Superplasticizers & Other Admixtures For Concrete

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
COPYRIGHT 2003; ACI International (American Concrete Institute) Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
00:29:06 MDT Questions or comments about this message: please call the Document
Policy Management Group at 1-800-451-1584.
A C 1 SP-148 94 W 0662947 0519770 187 W
Fourth
CANMET/ACI International
Conference on Superplasticizers and
Other Chemical Ädmktures
in Concrete
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
V.M. Malhotra
Editor SP-148
A C 1 SP-148 9 4 W Obb29Y9 0519771 013 W
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
DISCUSSION of individual papers in this symposium may be submitted in

accordance with general requirements of the AC1 Publication Policy to AC1
headquarters at the address given below. Closing date for submission of
discussion is April 1, 1995. All discussion approved by the Technical
Activities Committee along with closing remarks by the authors will be
published in the July/August 1995 issue of either AC1 Structural Journal or
AC1 Materiais Journal depending on the subject emphasis of the individual
paper.
The Institute is not responsible for the statements or opinions expressed in

its publications. Institute publications are not able to, nor intended to,
supplant individual training, responsibility, or judgment of the user, or the
supplier, of the information presented.
The papers in this volume have been reviewed under Institute publication
procedures by individuals expert in the subject areas of the papers.
Copyright O 1994
AMERICAN CONCRETE INSTITUTE
P.O. Box 19150, Redford Station
Detroit, Michigan 48219
All rights reserved including rights of reproduction and use in any form or
by any means, including the making of copies by any photo pracess, or by any
electronic or mechanical device, printed or written or oral, or recording for
sound or visual reproduction or for use in any knowledge or retrieval system
or device, unless permission in writing is obtained from the copyright
proprietors.
Printed in the United States of America
Editorial production Victoria Wieczorek
Library of Congress catalog card number 94-72367
A C 1 SP-148 94 Obb2949 0537772 T 5 T
PREFACE
In May 1978, The Canada Centre for Mineral and Energy

Technology (CANMET), in association with the American Concrete
Institute, the AC1 Eastern Ontario and Quebec Chapter, and the Structural
Division of the Canadian Society for Civil Engineering, sponsored a three-
day conference in Ottawa on the use of superplasticizers in concrete.
Selected papers from the symposium were published as American Concrete
Institute special publication SP-62.
In 1981, CANMET in association with the above organizations, spon-
sored a second three-day international conference in Ottawa on the use of
superplasticizers in concrete. The proceedings of the conference were
published as American Concrete Institute special publication SP-68.
In October 1989, CANMET in association with the American Con-
crete Institute and the AC1 Eastern Ontario and Quebec Chapter sponsored
the third three-day international conference in Ottawa. The purpose of the
third conference was to review the progress made since the meetings in
1978 and 1981; and to bring together representatives of the chemical
admixtures, cement, and concrete industries to exchange information and
to delineate the new areas of needed research. The scope of this conference
was expanded to include chemical admixtures other than superplasticizers.
The proceedings of the conference were published as American Concrete
Institute special publication SP-119.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
In October 1994, CANMET in association with the American Con-

crete Institute sponsored the fourth conference on the subject in Montreal,
Canada. The objective of this conference was to bring to the attention of
the concrete community the new developments in chemical admixtures since
the last conference in 1989. A total of 44 papers from more than 12 coun-
tries were received and reviewed by an expert panel of AC1 reviewers, and
25 were accepted for publication in the proceedings of the conference. The
proceedings were published as American Concrete Institute special publica-
tion SP-148. The co-operation of the authors in accepting the reviewers’
suggestions is greatly appreciated.
Particular thanks are extended to the members of the review panel
for their contributions towards the publication of this volume,
The help and assistance of A. Bilodeau (Chairperson), B. Fournier,
R. Chevrier (Members) of the Slide Review Panel are gratefully acknow-
ledged. Thanks are also extended to Harry S. Wilson, Mana Venturino, and
iii
Gordon D. Brearly for their help in the processing of the manuscripts, both
for the AC1 proceedings and the supplementary volume. Acknowledgement
is also made to Helayne Beavers, Managing Editor, Journals and Special
Publications, for her help in getting the proceedings ready on time.
V.M. Malhotra, P. Eng.

Editor
Chairman, Fourth CANMET/ACI International Conference
on Superplasticizers and Other Chemical Admixtures in Concrete
October 1994
Organizing Committee
Fourth CANMET/ACI Intemational Conference

on
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Superplasticizers and Other Chemical
Admixtures in Concrete
Montreal, Canada
October 1994
V.M. Malhotra
Chairman
Theodore W. Bremner (Member) Wilbert (Wib) S. Langley (Member)

Henri L. Isabelle (Member) Vangi S. Ramachandran (Member)
Harry S. Wilson
Secretary-Treasurer
COPYRIGHT 2003; ACI International (American Concrete Institute)

iv
Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-348 9 4 D 0662947 0539774 8 2 2 W
CONTENTS
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
PREFACE ....................................................
1.
.
III
ORGANIZING COMMITTEE .......................................... iv
ROLE OF SUPERPLASTICIZERS AND SLAG FOR PRODUCING HIGH

PERFORMANCE CONCRETE
by R.N. Swarny, M. Sakai, and N. Nakarnura ............................. 1
APPLICATION OF A NEW SUPERPLASTICIZER FOR ULTRA HIGH-STRENGTH CONCRETE

by K. Mitsui, T. Yonezawa, M. Kinoshita, and T. Shimono ...................... 27
CEMENT/SUPERPLASTICIZER/AlR-ENTRAINING AGENT COYPATIBILIM

by M. Baalbaki and P.C. Aïîcin ...................................... 47
PROGRESS IN UNDERSTANDING THE FUNCTIONAL PROPERTIES OF SUPERPLASTICIZERS

IN FRESH CONCRETE
by C Jolicoeur, P.C. Nkinamubanzi, M.A. Simard, and M. Piotîe .................. 63
INFLUENCE OF ACRYLIC POLYMERS ON THE RHEOLOGY OF MORTARS

by S. Maximilien, J. Ambroise, and J. Péra ............................... 89
SLUMP LOSS AND RHEOLOGY OF SUPERPLASTICIZED MORTAR AND CONCRETE WITH

DIFFERENT POLYSACCHARIDE SYRUPS
by G. Mantegazza and E. Alberti .................................... 105
WORKABILITY CHARACTERISTICS OF HIGH-STRENGTH CONCRETE INCORPORATING

AN AIR-ENTRAINING, HIGH-RANGE WATER-REDUCING ADMIXTURE
by M. Kagaya, H. Tokuda, M. Kawakami, and T. Kaneko ..................... 121
EFFECT OF A PHOSPHONATE-BASED COMPOUND ON THE HYDRATION OF CEMENT AND

CEMENT COMPONENTS
by V.S. Ramachandran and M.S. Lowery ............................... 131
THE INFLUENCE OF CROSS-LINKED AND NSF SUPERPLASTICIZER ON THE FLOW

PROPERTIES OF BLENDED CEMENTS
by T. Sone, S.L Sarker, and H. Uchikowa .............................. 153
V
A C 1 SP-148 ï Y m Obb2949 0539775 769 M
A STUDY ON THE FLOW OF HIGHLY SUPERPLASTICIZED CONCRETE

by S. Nishibayashi, S. h u e , A. Yoshino, T. Kuroda, and T. Kume ................ i77
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
APPLICATIONS OF RHEOLOGICAL MODIFIERS AND SUPERPLASTICIZERS IN

CEMENTITIOUS SYSTEMS
by C.B. Skaggs, W.G. Rokihky, and S.F. Whitaker .......................... 189
MIXTURE PROPORTION, FLOWABILITY, STRENGTH AND DURABILITY OF FLOWING CONCRETE

by Y. Kasoi, S. Hiraishi, K. Tobinai, and K. Osada ......................... 209
AN EXPERIMENTAL CONSTRUCTION OF MODEL STRUCTURE WITH FLOWING CONCRETE

by N. Nirhiyama, Y. Kasai, N. Yuam, and Y. Nakata ........................ îî7
SHRINKAGE AND CRACKING OF HIGH-STRENGTH CONCRETE AND FLOWING CONCRETE

AT EARLY AGES
by K. Yokoyama, S. Hiraishi, Y. Kasai, and K. Kishitani ...................... 243
THE INFLUENCE OF SUPERPLASTICIZER ON MIXTURE PROPORTIONING AND THE STRENGTH

AND DURABILITY OF SILICA FUME CONCRETE
by SA. Austin and PJ. Robins ..................................... 259
PROPERTIES OF AN ACRYLIC GRAA COPOLYMER-BASED NEW SUPERPLASTICIZER FOR

ULTRA HIGH-STRENGTH CONCRETE
by M. Kinoshita, T. Suzuki, T. Yonezawa, and K. Mitsui ...................... 281
SPUT ADDITION OF A HRWRA AND ITS EFFECT ON HIGH-PERFORMANCE CONCRETE

by JJ. Schemmel, V. Arora, and J. Williams ............................. 301
EFFECT OF MOIST CURING AND MODIFIER CONTENT ON PERFORMANCE IN FREEZING

AND THAWING OF TWO LATEX-MODIFIED CONCRETES
by C.M. Newîson and DJ. Janssen ................................... 317
EFFECT OF CHEMICAL CHARACTERISTICS OF ALCELL@ LIGNIN-BASED

METHYLSULPHONATES ON THEIR PERFORMANCE AS WATER-REDUCING ADMIXTURES
by J. Zhor, T.W. Bremner, and J.H. Lora ............................... 333
PRACTICAL IMPLICATIONS OF LITHIUM-BASED CHEMICALS AND ADMIXTURES IN

CONTROLLING ALKALI-AGGREGATE REACTIONS
by H. Wang, S. Tysl, and J.E. Gilloîî .................................. 353
PLASTICIZING CHARACTERISTICS OF SULFONIC ACID POLYMER CONTAINING

METHACRYLIC ACID DERIVATIVES
by E.I. TOZOWQ,A. Yonekura, M. Takahashi, S. Miyazawa, and K. Kawai ............ 367
PERFORMANCE EVALUATION OF SET ACCELERATORS FOR CONCRETE

by AA. Jeknavorian, N.S. Berke, and D.F. Shen ........................... 385

vi
A C 1 SP-148 94 0bb2949 0.539776 bT5
ROLE OF SUPERPLASTICIZERS IN GYPSUMLESS PORTLAND CEMENTS

by J. Hrazdira ............................................... 407
MODIFICATION OF CEMENT MORTARS BY POLYMERIC LATEXES AND THEIR USE

FOR REPAIRING CONCRETE
by G. Mantegona, A.M. Penna, and S. Taiîoni ........................... 415
PARAMETERS INFLUENCING THE OXYGEN AND CHLORIDE PERMEABILITY OF

SUPERPLASTICIZED NORMAL PORTLAND CEMENT AND FLY ASH-NORMAL
PORTLAND CEMENT MORTAR AND CONCRETE
by CJ. Lynsdale and J.G. Cabrera ................................... 435
SI (Metric) TABLES ............................................. 451
INDEX .................................................... 453

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 94 = 0662949 0519777 531 =
SP 148-1
Role of Superplasticizers and Slag for

Producing High Performance Concrete
by R.N. Swamy, M. Sakai, and N. Nakamura
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
-Da: This paper presents a detailed investigation into the role and
effectiveness of ground granulated blast-hace slag and a high range water
reducer (HRWR) on the quality of concrete in terms of bleeding, setting times,
heat evolution, strength development and pore structure. The tests were carried
out in two parts. (a) A slag of normal fineness was used, and both the
replacement level and water-binder ratio were varied. It was found that both
the slag and the HRWR acted as set retarders in terms of setting times and heat
evolution. The water-binder ratio was the predominant factor affecting the rate
of bleeding. The presence of slag, on the other hand, caused low early strength
and slow strength development but had significant beneficial influence on the
totai pore volume and pore size distribution. (b) T i e fineness of siag was varied
from 453 to 1 160 m2kg and the replacement level was kept constant at 50%. It
was then possible to obtain compressive strength in excess of 30 MPa at 3 days
and 100 MPa at 28 days with very substantial reductions in total porosity and
water permeability. The bleeding rate was also reduced and the setting times
also improved. The overall conclusion of this study is that a judicious
combination of HRWR and slag fineness can lead to a very effective synergic
interaction to produce concretes of high strength, high modulus and very low
porosity.
Kewords: Blast furnace slag; bleeding (concrete); durability; hiah performance concretes;
permeability; porosity; setting (hardening); h; strength; superplastitizen; water reducing
agents
COPYRIGHT 2003; ACI International (American Concrete Institute) 1Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-148 9 4 m Obb2949 0539778 478 m
2 Swamy, Sakai, and Nakamura
AC1 Fellow Prof. R. Narayan Swamy is at the Structural Integrity Research

Institute and the Centre for Cement and Concrete, University of Sheffield,
England. His research interests are in concrete materials and concrete
structures. He received the George Stephenson gold Medal of the Institution of
Civil Engineers and the Henry Adams Diploma from the Institution of
Structural Engineers. He is Editor of Cement and Concrete Composites and the
Book Series on Concrete Technology and Design. He recently received the
CANMET/ACI Award for contributions to Durability of Concrete.
Masami Sakai is a general manager at Building and Construction Centre, NKK

Corporation, Tokyo, Japan. He received his Doctor of Engineering degree in
1985 from the University of Tokyo. His main research interests are in
composite structures and utilization of steels and slag.
Nobuyuki Nakamura is a senior engineer at Building and Construction Centre,

NKK Corporation, Tokyo. He received his MS degree in 1976 from Tokyo
Institute of Technology. His main research interests are in high strength
concrete with slag and steel-concrete composites.
INTRODUCTION
Neither the use of superplasticizers nor the incorporation of ground,

granulated b l a s t - h a c e slag (GGBFS) in concrete are new to concrete
technology and to the construction industry. Water-reducing agents and
plasticizers and, more recently, high range water-reducers (HRWR) are,
however, often largely seen as workability agents, with the emphasis on
possible savings in cement and increases in compressive strength. The
incorporation of GGBFS as a cement-replacement material, on the other hand,
is often viewed as a means of reducing the temperature rise due to hydration, in
addition to its chemical resistance, with a penalty on early age and subsequent
development of strength. Both these perspectives are narrow and limited, and if
one takes the broader and global view of the need for concrete structures to
perform safely, satisfactorily and durably for a long time, these concepts may
be argued to be not only misplaced, but unrepresentative of the potential of
these concrete constituents to contribute to long-term durability, and to the
conservation of resources and energy.
Whilst good workability is accepted as an essential requirement for proper

placing and adequate compaction of concrete, the role of HRWRs should not be
seen to be solely related to water reduction nor are HRWRs to be treated as
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14B 94 M Ob62949 0519779 304
Superplasticizers and Chemical Admixtures 3
agents of workability. GGBFS, on the other hand, has the inherent ability to
contribute to strength, stiffness and durability, but this hidden potential is
chemically bound and locked within the material itself, and needs to be
extracted and mobilised for utilisation. A combination of slag and
superplasticizer can help to bring out the unique properties of each of these
components, and show that the synergic interaction between slag and
superplasticizers can produce a more durable and stronger concrete than when
either of these materials is used alone with portland cement. The aim of this
paper is to show how this synergistic reaction between superplasticizers and
portland cement-slag combination can be designed for and achieved, and
thereby influence beneficially bleeding, setting time, heat of hydration,
mechanical properties, and more importantly, the factors that control the
durability properties of the resulting composite concrete.
SCOPE OF INVESTIGATION
This investigation was carried out in two parts. The aim of the first part
was to identi@ the roles and effectiveness of slag and the HRWR in terms of
bleeding, heat evolution, strength development and pore structure. To achieve
this, tests on twelve concrete mixtures were carried out. In these tests, the
fineness of the slag and the water content were kept constant; the water-binder
ratio, the amount of HRWR and the cement replacement level were varied to
achieve concretes of consistent workability. The second part was designed to
substantially enhance the qualities of the slag concrete in terms of strength
development, porosity and water permeability. This was studied by increasing
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
the fineness of the slag but keeping the replacement level, water-binder ratio
and water content constant, but proportioning the concrete mixtures to have the
same high workability as in the first part.
The concrete mixtures used in this study were proportioned to have low
water-binder ratios and high workability with slumps in the range of 150 to 200
mm. In the first part of the study, for mixtures with the same water-binder
ratio, the portland cement was replaced, mass for mass, by slag of the same
fineness of 453 m2kg by amounts of 30%, 50% and 70% . The properties of
the fresh concrete such as temperature, density, bleeding rate and setting times
were determined. Heat evolution profiles from hydrating cement and slag were
measured by a conduction calorimeter using cement paste. The pore structure,
water permeability and the strength characteristics of the concrete were then
determined. The results of these tests are compared with those of concrete
without slag. In the second part of the study, slags of three different fineness,
namely 453, 786 and 1 1 60 m2kg, were used at a replacement level of 50%.
These concrete mixtures were proportioned similar to those in the first part, and
their strength development was related to porosity and water permeability. The
two sets of data show that the combined use of a HRWR and slag can lead to
A C 1 SP-148 9 4 0bb2949 0 5 3 9 7 8 0 O26 W
effective synergic interaction to produce concrete of high strength, high

modulus and very low porosity.
EXPERIMENTAL DETAILS
Concrete Materiais
Normal portland cement, ASTM Type 1, was used for all the concretes.
The physical properties and chemical analysis of the cement are shown in Table
1. The GGBFS used in the first part of the study had similar average particle
size to that of portland cement but had about 40% higher surface area at 453
m2kg compared to 323 m2kg of the portland cement. Table 1 also contains the
physical properties and chemical analysis of the slag. The concrete aggregates
consisted of crushed sandstone with a maximum size of 20 mm with a specific
gravity of 2.65, fineness modulus of 6.61 and water absorption of 0.69%. The
fine aggregate was a river sand with 5 mm maximum size, 2.60 specific gravity,
2.93 fineness modulus and a water absorption of 1.20%.
The only variable in the second part of the study was the fineness of the
slag. The physical properties and chemical analysis of these slags are also
shown in Table I.
Concrete Mixture Details
In the first part of the study, twelve concrete mixtures were tested with
water-binder ratios of 0.45, 0.40 and 0.35. The cement replacement levels,
mass for mass, were O, 30,50 and 70%. The aggregate contents were slightly
adjusted for each mixture, with sandítotal aggregate ratios varying from 0.470
to 0.426, to give dense mixtures. The mixture proportions used in the first part
of the tests are shown in Table 2.
All the mixtures were proportioned to have high workability with

cohesive and flowable matrices, and slumps in the range of 150 to 200 mm.
The total water content for each mixture was kept at 160 ks/m3. The high
workability and control of slump was achieved by using a high range water
reducer (HRWR), polyether carboxylic acid. The amount of HRWR was
adjusted for the required slump, and the actual quantity used in the mixtures
varied from about 3.8 to 5.8 kg/m3, Le. from about 1.12 to 1.22% by weight of
the binder.
Test Programme
Immediately after the concrete mixing, the slump, air content, concrete
temperature and density were measured. In addition, the bleeding rate and the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0 6 6 2 9 4 9 05l19781 T62 W
initiai and final setting times were determined according to ASTM C 232-87
and ASTM C 403-88 respectively.
The heat evolved during hydration of the portland cement and slag paste
under constant temperature conditions was measured. These heat evolution
profiles were determined by a conduction calorimeter using neat cement paste.
The test was carried out at 20°C. Compressive strength and elastic modulus
were determined from 100 x 200 mm cylinders which were cured at 20°C and
100% RH for 24 hr, demoulded, and then cured in water at 20°C. The pore
structure of ali the mixtures was determined from mortar samples through
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
mercury porosimetry. The test samples were cured at 2OoC in a curing room for
24 hr, demoulded and then cured in water until testing. The water permeability
test was camed out on 150 x 300 mm size cylinders, which were cured in water
at 20°C for 28 days followed by air curing at 20°C and 65% RH for 7 days. By
applying a water pressure of 1.5 MPa for 48 hrs, the average depth of
penetration and the coefficient of water diffusion were determined.
TEST RESULTS AND DISCUSSION
Properties of Fresh Concrete
The properties of the fresh concrete immediately after casting are given in
Table 3. The slump results show that the designed workability was obtained for
all the mixtures. The actual slump varied from 165 to 185 mm giving concretes
of excellent flow characteristics. The effect of the incorporation of GGBFS in
concrete in enhancing their flow properties can be assessed from the amount of
HRWR added to these mixtures. For similar workabilities, it was possible to
reduce the amount of HRWR with increase in the substitution rate of the slag
for cement. In general, with the highest replacement level of 70%, about 10%
less HRWR was used compared to that for normal portland concrete.
The air content of the fresh concrete vaned from 0.7% to 1.4%. In
general, the entrapped air decreased with the amount of cement replacement.
The temperature of the fresh concrete immediately after mixing remained
relatively constant, at 21-22"C, with the slag concrete mixtures registering more
or less the same temperature as the concrete without slag. The density of the
fresh concrete also varied little between the different mixtures, ranging from
2390 to 2430 kgím3. The incorporation of slag had thus little effect either on
the temperature (immediately after casting) or the density of the fresh concrete
compared to concrete without slag.
A C 1 SP-148 9 4 m Ob62949 051’3782 9Tï m
Rate of Bleeding
The rate of bleeding of all the mixtures was determined according to

ASTM C 232-87, The bleeding rate (YO) and the volume of bleeding water are
shown in Table 4, and typical variations of bleeding with time are shown in Fig.
1 and 2. In general, the bleeding rate is seen to increase with an increase in the
substitution rate of slag for cement, although there appears to be no consistent
pattern between the amount of bleeding and the level of cement replacement.
However, the results taken altogether show that there is a progressive reduction
in bleeding as the water-binder ratio is decreased. The major reductions in
bleeding occur when the water-binder ratio is decreased from 0.45 to 0.40, and
still substantial reduced bleeding rates can be obtained when the water-binder
ratio is decreased still fùrther to 0.35.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
The bleeding test shows that the water-binder ratio is the principal factor
affecting bleeding, but that the replacement level has also an important
influence on the rate of bleeding. Tests reported elsewhere (1) also show that
the slag fineness is also a parameter affecting this property, so that a judicious
combination of these three factors can ensure a very highly workable mixture
with the minimum of bleeding.
Settinp Times
The initial and final setting times of all the mixtures were determined
according to ASTM C 403-88. The results of these tests are included in Table 4
and typical variation of the penetration resistance with time is shown in Fig. 3
for cement pastes with water-binder ratio of 0.40.
These results show that although the water-binder ratio has some minor
influence on the setting times, the major factor influencing this property is the
amount of cement replacement. Both the initial and final setting times increase
as the level of cement replacement increases. In general terms, at a replacement
level of 70%, both the initial and final setting times are increased by about 1 hr
to 1% hr. In the setting time tests, the mortar temperature was almost equal to
the concrete temperature taken soon after mixing. The temperature in all cases
was almost equai to 20°C.
The increased setting times with the incorporation of slag confirm that the
cement-slag hydration is a two-stage process (2-5) and that the presence of slag
acts in some respects as a set retarder. The combination of the HRWR with the
slag is also responsible to some extent to the increased setting times. The
carboxylic acid ether HRWR used in these tests is designed to obtain high
slumps easily, particularly with low water-binder ratios, and it is also designed
to control the slump loss. The setting time with the J3RWR therefore tends to
lag behind that without the admixture as confirmed by the heat of hydration
tests shown later in this paper.
A C 1 SP-148 7Y W 0bb2949 0519783 835 W
The use of high slag replacements may thus appear to prolong the time
during which the concrete is vulnerable to plastic shrinkage. This may
particularly be undesirable in deep section concrete members where the slag
concrete may bleed longer than comparable cement concrete. In practice, this
does not appear to be the case because of the cohesive and dense nature of the
concrete mixtures, and the relatively short increases in the setting times
achieved by the controlled use of the HRWR. In situations where the
environment is such as to likely enhance the adverse effects of bleeding and
increased setting times, other precautions need to be taken irrespective of
whether the concrete is made without or with slag.
Heat of Hvdration
The heat evolved during the hydration of the cement-slag combination

was measured by conduction calorimetry. The tests were carried out at 20°C on
neat cement pastes without and with the HRWR and with cement-slag mixtures.
Fig. 4 shows typical heat evolution profiles for mixtures with water-binder ratio
of 0.40 and cement replacement of 50%.
The results of the conduction calorimetry shown in Fig. 4 emphasize the

role of both the HRWR and slag on the heat evolved during hydration. The
presence of a HRWR, capable of reducing slump loss with time, has extended
the peak time of heat evolution by a factor of almost two, from about 9.2 hr to
about 17.2 hr. Replacing portland cement by 50% of slag at a water-binder
ratio of 0.40 is seen, on the other hand, to reduce the peak heat evolution to
about 60% of that of portland cement alone and this peak occurs at about the
same time as that for portland cement. The incorporation of a HRWR, in
addition to slag, on the other hand, again delays, as for portland cement, the
occurrence of the peak heat evolution by about the same period. It is thus clear
that both slag and the HRWR act as retarders so far as both setting times and
heat evolution are concerned.
The heat of hydration is a major factor influencing the long-term

durability of concrete. Early-age thermal cracking, arising from undesirable
thermal gradients, is not easy to heal, and takes a long time to disappear. The
results of the study reported here show that incorporating slag and a HRWR can
not only reduce the heat evolved but can also extend the time of its evolution
which will allow proper placement and compaction of the concrete in
reinforcement congested sections and in hot, dry environments to be carried out.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 O b b 2 9 4 9 0519784 771
8 Swami, Sakai, and Nakamura
PROPERTIES OF HARDENED CONCRETE
Compressive Strength and Elastic Modulus
Many studies have shown that the hydration of slag in combination with
portland cement is a two-stage process (2-5), and because slag hydration tends
to lag behind that of the portland cement component's hydration, there will
always be a penalty on early age and subsequent development of strength and
elastic modulus of slag concrete (5-9). The data presented in Table 4 and Fig. 4
provide basic confirmation of this phenomenon in terms of both setting times
and time of peak heat evolution. Conventional mixture proportioning methods
of incorporating slag in concrete are therefore likely to show a slower rate of
strength development than the corresponding portland cement concrete.
However, modifying the mixture proportions to compensate this can achieve
high early strength as shown elsewhere (5). The test data reported here are,
however, based on direct replacement of cement by slag, mass for mass, with no
mixture modifications to counteract the slower development of early strength.
The early development of compressive strength and Young's modulus

measured on 100 x 200-mm cylinders for the mixture proportions given in
Table 2 is shown in Fig. 5 and 6. The compressive strength results of Fig. 5.
show that at 3 and 7 days the slag concrete had lower compressive strength than
the portland cement concrete for all substitution levels and water-binder ratios.
The reduction in strength is almost inversely proportional to the amount of slag
substitution. At 28 days all the mixtures with a substitution rate of slag up to
50% were able to achieve strength equal to that of the portland cement concrete,
and this compressive strength varied from 50 to 70 MPa. At 70% replacement
level, however, it took about 91 days for the slag concrete to achieve about the
same compressive strength as portland cement concrete. At this age, all the slag
concretes showed higher strength than the portland cement concrete, and this
varied from 60 to 85 MPa.
Fig. 6 shows the variation of Young's modulus with compressive strength.

At 28 days, this value varied from 27.0 to 34.0 GPa, and only two concretes
with 70% slag at water-binder ratios of 0.45 and 0.40 showed lower modulus
than the corresponding portland cement concrete. At 91 days, all the slag
concretes showed higher elastic modulus than the portland cement concrete, and
this ranged from 35 to 38 GPa.
Pore Structure
The influence of slag on the pore structure of the mixtures with water-
binder ratios of 0.40 was determined by mercury intrusion porosimetry.
Typical data on the cumulative pore volume and pore size distribution on
portland cement mixtures and portland cement with SO% slag mixtures are
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 I0 6 6 2 9 4 7 051i7785 608 I
shown in Fig. 7 to 9. These figures clearly show that the presence of slag has a
significantly beneficial influence on the porosity of the resulting concrete, both
in terms of total pore volume and the distribution of pore sizes. These are
substantial advantages which positively contribute to enhanced durability and
superior serviceability life compared to normal portland cement concrete:
however, the fact remains that this superior pore structure is obtained at the
expense of lower and slower early-age strength up to about 28 days at ali
cement replacement levels.
CONCRETES WITH SLAG OF VARYING FINENESS
It is obvious that the maximum technical, practical and economic

advantages of utilizing slag concrete can be realised if one could combine the
superior pore structure with high-early strength and rapid strength development
comparable at least to that of portland cement concrete. Because of the inherent
nature of the portland Cement and slag hydration processes, it is also clear thn
low and slow development of strength will always be part of slag concretes,
unless some slag activation techniques are used. On the other hand, the results
presented above show that the combination of HRWR and slag possesses
intrinsic capabilities to produce high-early strength, high strength at later ages
and excellent durability. One method of overcoming some of the limitations
recognized in the first part of this study then appears to be to grind the slag
finer, and use the finer slag.
It is recognised that higher costs will be incurred in grinding the slag

finer, but these cost disadvantages can be greatly offset by the substantial
improvements in the quality of the resulting concrete and particularly in its pore
structure and water tightness, as shown later. In the tests reported here the slag
S4 with 453 m2kg slag fineness was obtained by the ordinary grinding process.
The slags S8 and S12 of higher fineness of 786 and 1160 m2kg respectively
were obtained by classifying the S4 slag with an air classifier.
In the following sections, the strength development, pore structure and

water permeability of concretes containing these finer slags and the same
HRWR used previously are presented. In these tests three degrees of slag
fineness, namely, 453, 786 and 1160 d k g were used; the slag replacement
level was kept at 50% by mass, and the water-binder ratio at 0.40 and 0.30.
Only typical data are shown here to emphasize how the use of fine slag and a
HRWR can lead to concrete of high strength and excellent durability.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 D Obb29LI9 0 5 L 9 7 8 b 5 4 4 M
DEVELOPMENT OF STRENGTH AND ELASTIC MODULUS
Table 5 presents data on cylinder (1O0 x 200-mm) compressive strength

development with age for portland cement concretes without and with 50% slag
replacement. These concretes were proportioned in a way similar to that shown
in Table 2; the water content was kept at 160 kglm3 and sufficient HRWR was
used to give slumps of 170 to 190 mni. Although not reported here to save
space, both the bleeding rate and the setting times also improved as the fineness
of slag was increased. The strength results show that slags of 786 and 1160
m2kg fineness gave strengths comparable to portland cement concrete at 3 days
and that these concretes gave strengths at 91 days in excess of about 60% to
70% compared to that of portland cement concrete. The strength data show that
with slag fineness of 786 and 1160 m2/kg, cylinder compressive strengths in
excess of 30 MPa at 3 days and 100 MPa at 28 days can be obtained.
Fig. 10 shows the variation of Young's modulus with cylinder

compressive strength for all slag concretes tested in this study with slag
finenesses of 453 to 1160 m2kg. The elastic moduli of concretes with
compressive strength in excess of 85 MPa obtained with slags of 786 and 1160
m2kg fineness are identified separately in Fig. 10. These results show that slag
concretes can develop elastic moduli comparable to that of portland cement
concrete. For the very high cylinder strengths of 110 to 120 MPa, the elastic
modulus was in excess of 40 GPa.
Porositv
The total pore volume of portland cement concrete and of slag concretes
of varying slag finenesses is shown in ?'able 5. Fig. 11 shows a comparison of
the cumulative pore volume and pore size distribution of portland cement paste
with that of cement-slag combination with the highest slag fineness. These data
show dramatic improvements in the refinement of pores and reduction in pore
volume of slag concretes compared to that of portland cement concrete. Two
significant factors stand out. Whilst the period of wet curing has a strong
influence on the resulting pore structure, the data in Table 5 show that while the
total pore volume for portland cement concrete is reduced by about 50% after
91 days curing, compared to that at 3 days, that of concrete even with the
coarsest slag fineness of 453 m/kg at the same age of 91 days is only about 115
of its total pore volume at 3 days. As the slag fineness is increased, further
significant reductions in total pore volume compared to that of portland cement
concrete are also obtained. These are major improvements in the quality of slag
concrete which will have a direct impact on its long-term durability.
Secondly, typical data as shown in Fig. 11, portray how the pore structure
becomes finer as the slag fineness becomes higher. With ali slag finenesses, the
pore size is almost uniformly small, whilst the portland cement concrete shows
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 m 0662949 0519787 480
a large size pore distribution at all ages compared with even the coarsest slag
(Fig. 7 to 9). These results show that slag concrete has an inherently finer pore
structure, while portland cement concrete will always have a coarser pore
structure than the slag concrete.
Water Permeabiìitv
The water permeability tests reported here were canied out in accordance
with what is called the “input method” reported elsewhere (1 i). The tests were
conducted on 150 x 300-mm size cylinders which were initially cured in water
at 2OoC for 28 days followed by air curing at 2OoC and 60% RH for 7 days.
The cylinders were then subjected to a water pressure of 1.5 MPa for 48 hr.
The depth of water penetration is then measured, and the coefficient of
diffusion calculated (12).
The results of these water permeability tests in terms of the average depth
of penetration and coefficient of diffusion are shown in Table 5. Fig. 12 shows
typical cylindrical specimens of portland cement concrete and slag concrete of
1 160 m2kg fineness broken after the water permeability test. The depth of
water penetration in the two concretes is clearly seen in this Figure.
The results of the water permeability test in Table 5 and Fig. 11 show that
portland cement concrete has the highest water permeability compared to all the
slag concretes. it will be seen that even the coarsest slag imparts significant
improvements to water impermeability, compared to portland cement concrete.
The depth of water penetration is then reduced by about 35%, whilst the
diffusion coefficient is only about 45% of that of portland cement concrete.
Still further improvements to the quality of concrete can be achieved when the
slag fineness is increased from 453 to 786 m2kg at the same water-binder ratio
of 0.4; however, very major benefits are obtained when the slag fineness is
increased further to 1160 m2kg. At this slag fineness, the average depth of
water penetration is less than 25% of that of portland cement concrete. The
difíùsion coefficient, on the other hand, is still further reduced. It is less than
10% of that of portland cement concrete. Quantitatively these are dramatic
improvements, but when considering that water is the major transport medium
for deleterious agents, the superior quality of concrete arising from the
incorporation of slag in concrete can be appreciated.
There is one still further implication to the data shown in Table 5 .

Comparing mixtures C, D and E, it can be seen that mixture D with the 786
m2’kg slag can give almost the same microstructural quality as concrete with the
1160 m2kg slag, if instead of grinding the slag finer, a reduction in the water-
binder ratio is effected. Thus concrete with slag of 786 m2kg fineness and
water-binder ratio of 0.30 can impart durability properties similar to that of
concrete with 1160 m2kg slag fineness at a higher water-binder ratio of 0.40.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 0662749 05L9788 317 W
A judicious combination of slag fineness and water-binder ratio can thus

achieve very substantial improvements in strength, porosity and water-tightness
of concrete. The data in Table 5 thus confirm that a moderate increase in slag
fineness can achieve important and large improvements in durability properties.
The moderate increase in costs involved can thus be fully justified by longer
and durable service life and less repair costs. Indeed, it might even be possible
to achieve similar outstanding durability properties with slag fineness of 453
m2kg and a water-binder ratio of 0.30.
One important deduction from the porosity and water permeability data is
the good correspondence between the results of these tests. The total pore
volume is thus a good indicator of the denseness of the concrete and its
resistance to water permeability.
CONCLUSIONS
The major conclusions derived from this study are as follows.
The HRWR enabled concrete of consistent and high workability in excess

of 150 mm slump to be produced, and for similar workabilities, it was possible
to reduce the amount of HRWR with increase in the substitution rate of slag.
The heat evolution profiies, on the other hand, showed that the HRWR also
acted as a set retarder by almost doubling the time of occurrence of the peak
heat evolution, for all concretes without and with slag. The bleeding rate was
found to be primarily dependent on the water-binder ratio, although it did
increase with increasing amounts of slag content. The major factor influencing
setting times, on the other hand, was the slag content, although the water-binder
ratio also had some minor influence on this property. Replacing portland
cement by 50% of slag at a water-binder ratio of 0.40 reduced the peak heat
evolution to about 60% of that of portland cement alone, but a combination of
slag and HRWR not only maintained this reduction in peak heat evolution, but
also extended the occurrence of the peak heat evolution by about the same time.
The development of compressive strength of slag concrete was almost

directly dependent on the water-binder ratio and the amount of cement
replacement. In general, the reduction in strength was almost inversely
proportional to the amount of slag substitution, but neither the setting times nor
the heat of hydration appeared to have a direct influence on or relationship to
strength development. At 28 days all the concretes with slag substitution up to
50% achieved strength equal to that of portland cement concrete. At 70%
replacement level, it took about 91 days for the slag concrete to achieve about
the same strength as portland cement concrete. Beyond this age, all the slag
concretes possessed higher elastic modulus than portland cement concrete. The
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0539789 253 =
slag had also a significantly beneficial infíuence on the porosity of the concrete,
both in terms of total pore volume and pore size distribution.
A combination of the HRWR and a finer slag than normal produced

significant improvements in strength development, total pore volume and water
permeability. Cylinder compressive strengths in excess of 30 MPa at 3 days
and 100 MPa at 28 days were obtained with Young's elastic modulus values in
excess of 40 GPa. There were also very substantial reductions in total pore
volume and impermeability to water. The use of a very fine slag with a HRWR
can thus produce high strength, high modulus and very low porosity.
Grinding the slag finer will involve higher costs, but the Substantial
improvements in the quality of the resulting concrete, and particularly in its
pore structure and water tightness, can greatly offset the higher initial costs
through longer and durable service life and less repair costs. However, the data
also show that a judicious combination of slag fineness and wata-binder ratio
can achieve similar outstanding durability properties with only minimal increase
in costs.
REFERENCES
1. Niamura, N., Sakai, Ni., and Swamy, K.N.,"EfÏect of Slag Fineness on

the Engineering Properties of High Strength Concrete", Blending Cements
in Construction, Editor R.N. Swamy, Elsevier Applied Science Publishers,
London, 1991, pp.302-316.
2. Roy, D.M., and Idorn, G.M., "Hydration, Structure and Properties of Blast
Furnace Slag Cements, Mortars and Concrete", AC1 Journal Proc. Vol. 79,
No. 6 , Nov-Dec. 1982, pp.444-457
3. Mather, Bryant, "Laboratory Tests of Portland Blast-Furnace Slag

Cements", AC1 Journal, Proc. Vol. 54, NO.3, Sept 1957, pp.205-232.
4. Fulton, F.S., "The Properties of Portland Cement Containing Milled

Granulated Blast-Furnace Slag", Monograph, Portland Cement Institute,
Johannesburg, 1974, pp.4-46.
5. Swamy, R.N., and Bouikni, A., "Some Engineering Properties of Slag
Concrete as Influenced by Mix Proportioning and Curing", AC1 Materials
Journal, Vol. 87, No. 3, May-June 1990, pp.210-220.
6. Douglas, E., Wilson, H., and Malhotra, V.M., "Production and Evaluation
of a New Source of Granulated Blast-Furnace Slag", Proc. Int. Workshop
on Granulated Blast-Furnace Slag, (Toronto 1987), CANMET, Ottawa,
pp.79-112.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
14 Swamy, Sakai, and Nakamuro
7. Malhotra, V.M., Carette, G.G., and Bremer, T.W., "Durability of

Granulated Blast-Furnace Slag in Manne Environment", Ibid, pp. 171-201.
8. Wainwright, P.J., "The Influence of Slag Cement on Some of the

Properties of Concrete Related to Thermal Cracking", Ibid, pp.203-227.
9. Wainwright, P.J., and Tolloczko, J.J.A., "Early and Later Age Properties
of Temperature Cycled Slag-OPC Concrete", Second Int. Conf. on Fly
Ash, Slag, Silica Fume and Natural Pozzolans in Concrete, AC1 publn.
SP-91, V.M. Malhotra, Editor, 1986, pp.1293-1321.
1o. Malhotra, V.M., "Properties of Concrete Incorporating Low Quantity of

Cement and High Volumes of Ground Granulated Slag", AC1 Materials
Journal, Vol. 89, Nov-Dec 1992, pp.554-566.
II. Murata, J., "Studies on the Permeability of Concrete", Trans. of Japan

Society of Civil Engineers, No. 77, Nov. 1961, pp.69-103, (In Japanese).
12. Nakamura, N., Sakai, M., and Swamy, R.N., "Effect of Slag Fineness on
the Development of Concrete Strength and Microstructure", Int. Conf. on
Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, AC1 Publ.
SP-132, Vol. 2, Editor V.M. Malhotra, 1992, pp.1343-1366.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0 b b 2 9 4 9 0519791 901
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 1 - PHYSICAL PROPERTIES AND CHEMICAL ANALYSES OF PORTLAND CEMENT

AND BLAST FURNACE SLAG
Portland- Slag
Description of Test
cement - s4 S8 s12
Physical properties
Fineness
88 pm (passing) % 99.5 100.0 -
15 pm (passing) % 54.0 63.0 97.0 100.0
Average particle 14.0 11.7 5.1 3.1
Surface area,
Air permeability 323 453 786 1160
(m2k3)
Specific gravity 3.16 2.93 2.93 2.93
Mortar strength @Pa) *

3 days 16.0 -
7 days 23.8
28 days 42.3 - -
Chemical analvsis
YO
Ca0 63.0 41.6 41.2 41.0
Si02 21.6 33.5 33.8 33.9
A1203 5.3 15.0 14.5 14.6
Mg0 2.6 6.8 6.9 6.9
Fe203 3.1 Tr Tr Tr
so3 2.2 Tr Tr Tr
Na20 0.34 0.18 0.20 0.21
K2O 0.44 0.18 0.20 0.21
Loss on ignition 1.2 0.0 o. 1 o. 1
insoluble 0.2 o. 1 o. 1 o. 1
Bogue Compound composition
%
c3s 46.0
c2s 27.4
C3A 8.8
c4AF 9.4
* Based on 40-mm cubes
A C 1 SP-148 94 06629q9 0539792 8413 M
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
B
m
141
O 1 % O
2
m m m
vi
in
d 3 d
*?
8 in
m
O
O
z
A C 1 SP-148 74 Ob62747 0517773 7 8 4 W
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14894 m Obb2949 0519794 bL0 m
i n w o o V I V I 0 0
m r n m u u m u o
I l I I I l I I
w w r - r - w r - r - 0 0
I A V I I A 0
c . 4 b - m
I l I I
m o m o
O W b O
Q t - Q t -
O 0 0 0
0 0 0 0 O 0 0 0
V I W W r -
- u q m
O 0 0 0
VI IA
x O
U
O
m
O
m w r - 0 0
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 M 0662749 0539795 5 5 7 M
U
O
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Y
5a
æ
4
A C 1 SP-148 94 Ob62949 0539796 493
"t
ELAPSED TIME, BRS
Fig. 1-Bleeding rate for mixtures with water-binder ratio of 0.45
WIC = 035
Bleed VoLof
Mix Rate Bleed Water
% cm3/cm2
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
ELAPSED TIME, HR!3
Fig. 2-Effed of slag replacement level on bleeding rate for water-binder ratio of 0.35
A C 1 SP-14ô 94 0662949 0519797 32T
Superplasticizers and Chemical Admixtures
= 21
FINAL SETTING
mo---- -- - - - - - - _ -
WIC+S -= 0.40 I
CURING TEMP. 2O'C
Mix 7
-- OX -- 50.h
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
MU 8 70%
1000 -
ELAPSED TIME, HRS

Fig. L l n i t i a l and final setting times for slag concretes with water-binder ratio of 0.40
OPC + HRWR
15-
OPC f .'. SLAG BLAINE 453 d / k g
/ \, w¡C+S = 0.40, S/C+S = 50%
10-
5-
O0 10 20 30 LO
ELAPSED TIME, HRS
Fig. &Effect of HRWR and slag on heat evolution
A C 1 SP-I148 94 0 6 6 2 9 4 9 0539798 2bb
SLAG BLAINE 453 m2kg
rn
o- 0%
o- 30%
O - 50%
O - m%
37 28 91 3 7 28 91 37 28 91
CURING TIME, DAYS
Fig. %Cylinder compressive strength development of slog concrete
25
LO’I
30 LO 50
I 60 70 ‘
80
COMPRESSIVE STRENGTH, MPa
Fig. &Variation of static Young’s modulus with cylinder compressive strength

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14A Ob62949 O5I19799 LTE'
94
Superplastitizen and Chemical Admixtures 23
EFFECT OF SLAG FINFNESS ON POROSITY
NO SLAG
wtc+s = 0.40
s/c+s = o
PORE RADIUS 'i
10-
N- ,,," _
.' _ _ _ _ _ _
L-
' .-.-.. .. . .._
'
:-i
o- , , , , ,,,, , I , ,,;, , ,,
z [P 1v
PORE RADIUS ,i
Fig. 7-Pore structure of portland cement and slag concretes: 7 days
EFFECT OF SLAG F"ENESS ON POROSITY
W I -1
i.... " ,,,, ,

I
PORE RADIUS A
Fig. W o r e volume of portland cement and slag concretes: 28 days
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 0bb2949 0519800 7 4 4 W=
EFFECT OF SLAG FINENESS ON POROSITY

60
QawGTDb6 mAYS
NO SLAG
50- mALpo116vcuQ(E
-smE"685MR
-
'
/
t w/c+s= 0.40
SIC+S = o
10-
...............
"
I
Y
10
Oal, , ,-,,i.,
10
__/-.;,
'... . ........
DSIRÆüílON
.-______
, , , ,,,
1Q
.....-.......
-.-._
- _ _----
., , , ,
poRE RADIUS A
lo<
................
1GJ 1GJ
SLAG 453 MVKG

CmmBSmsnmNlll-
SIC+ S=O.50
10
........ ............... ...........

10
........
10
PORE RADIUS A
Fig. O-Effect of slag pore volume: 91 days
4/
30
I
I
LO
1
50 60 70 80 90 100
I I I I l I
110
I
120
COMPRESSIVE STRENGTH. MPa
Fig. lû-Variation of Young's modulus with cylinder compressive strength for all slag concretes
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 Ob62949 0519801 680
0
60 -' \
S12 - 1160 m2/ kg
2E 50- \\ W/(C+S)=O.L
60
ol 1 . A C
20 DAYS
',OPC
\
,5
AGE91 DAYS
w ----.\IL)PC
iI
\
\
9 20-1 \ CU MU LAT IVE
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
PORE RADIUS, Å
Fig. 11-Pore structure of portland cement concrete and slog concrete with very fine slog
A C 1 SP-148 94 Obb2949 0539802 5 3 7
W/C + S = 0.40
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 12-Typical broken cyiindricol specimens showing effectiveness of slag in reducing

water permeability
A C 1 SP-148 94 = Ob62947 0519803 4 5 3
SP 148-2
Application of a New Superplasticizer

for Ultra High-Strength Concrete
by K. Mitsui, T. Yonezawa,
M. Kinoshita, and T. Shimono
Synopsis. The icey to casting high-strength concrete with compressive strength of

more than 100 MPa into complicated reinforced shuctures are to give the concrete
high fluidity as well as to improve strength. The authors developed an acrylic
copolymer-based new superplasticizer that can improve fluidity of concrete with
water to binder ratio of around 0.20.
This paper gives the results of a senes of studies conducted to d e t e m e the
properties of fresh and hardened high-strength concrete using the newly developed
superplasticizer. The effect of the new superplasticizer was e x m e d with varying
water to binder ratio, type of cement, and temperature compared with conventional
superplasticizers The new superpiasticizer needed much less dosage than
conventional superplasticizersto attam d certain slump (250 mm) for a water to binder
ratio of around 0.20 and it significantly reduced the viscosity of concrete. Sufficient
workability w&. kept for 2 hours without much delay III settmg time while
conventional superplasticizers showed large slump loss and excessive delay in settmg
time. The results of strength development, drying shnnkage and freezing and thawing
resistance did not show any harmful effect.
Field studies were conducted on application of the high-strength concrete to a
prestressed concrete bridge with design strength of 100 MPa usmg the new
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
superplasticizer Workability and strength development of concrete were tested and

resulted in sufficient quality
Keywords: Bridaes [structures]; coarse aggregates; compressive strength; drying shrinkage;

flowability; fly ash; freeze thaw durability; hiah strencith concretes; mix proportioning; setting
(hardening); silica fume; slump; suDerDlasticizers; temperature; workability

27
zu
A C 1 SP-148
MltSUI et al
74 = 0662949 05L9804 3 ï T W
Kenro Mitsui is AC1 member, Deputy Chief Researcher, Research and Development
institute, Takenaka Corporation, Chiba, JAPAN
Toshio Yonezawa, Ph.,D. is Chief researcher, Research and Development institute,

Takenaka Corporation, Chiba ,JAPAN
Mitsuo Kmoshita is Chef research Engineer, Chemical Admixture Division,

Takemoto Oil&l:at Co.,Ltd., Aichi, JAPAN
Toshhde Shmono is Chef research engmeer, Chemical Admixture Division,

Takernoto Oil&I;at Co.,Ltd., Aich, JAPAN
INTRODUCTION
The development of concrete technology in recent years has led to considerable

increases in the strength of concrete. As a result, hgh-nse buildings and long-span
bndges have been constructed with concrete, and high-strength concrete has become
essential for such huge structures. Further increases in the strength of concrete are
now required 111 order to construct structures of even greater scales economically.
Commercializationof ultra high-strength concrete exceeding 100 Mpa is anticipated.
It is necessay to limit the water-to-binder ratio to a level as low as 0.20, m order

to produce ultra high-strength concrete with a design strength exceeding 100 MPa
This produces v.scous concrete and a loss of fluidity hampenng the in-situ placement
of such concrete
The authors developed a new acrylic superplasticizer that greatly improves the
fluidity of concrete with a water-to-binder ratio 111 an extremely low range, and retams
sufficient fluidity for a longer time, when compared with conventional
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
superplasticizers
Ths paper rcports on the results of expenments to compare the properhes of ultra
high-strength concrete in fresh and hardened states contanmg the newly developed
superplasticizer with those of ultra high-strength concrete contauung conventional
superplasticizers. It also reports on the expenmental study and actuai application of
this superplasticizer to a prestressed concrete bndge with a design strength of 100
m a
A C 1 SP-148 94 Ob62949 0537805 226
OUTLINE OF LABORATORY EXPERIMENT
Compansons were made between the new superplasticizer and conventional

superplasticizers, with regard to dosages, slump and viscosity losses over time, semng
I
l time, and comprzssive strength development.
Materiais used
New superplasticizer. The newly developed superplasticizer (herein referred to as

SSP) cornpnses an acrylic graft copolymer with graft chams and polyethylene glycol
chains, with an adequate mole ratio of carboxylic groups and sulfonic groups (-
c00w-s03 = SODO) III the molecule
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Other chemical admixtures: A conventional polycarboxylic superplasticizer (SP)
and a conventional naphthalene superplasticizer (NSF) were used for companson with
the SSP.
Cements: Two types of Portland cement were used, i.e. normal Portland cement
(N) and Portlanc fly ash cement (FB).
Aggregates: River sand with a specific gravity of 2.6 I, water absorption of 1.16%,
and fineness modulus of 2.81 was used as the fine aggregate. Crushed limestone with
a maximum size of 20 mm, a specific gravity of 2.70, water absorption of 0.38%, and
fineness modulus of 6.42 was used as the coarse aggregate.
Silica fume: Undensified silica fume as presented u1 Table 1 was used to replace
10% of the cement by weight.
Mixture propoi-
The mixture proportions of the concrete are presented in Table 2. The concrete
were proportioned to have the same volume of cementinous paste content (cernent +
silica fume + wkter). The flow, which was d e t e m e d as the average diameter of the
concrete after spreading at slump test, was measured to d e t e m e the workability of
concrete. The dosages of the superplasticizers were established to provide a flow of
5 5 0 k 5 0 mm. The ax content was 2.0 k I 0%.
Experimental procedure
Mix mortar iòr 1 minute KI a pan-type mixer Add coarse aggregate and mix for 2
minutes Measure the slump, flow, a r content, and viscosity of the concrete at 30, 60,
90, and 120 minutes after mixing. Measure semng time in accordance with ASTM
C403 Test the compressive strengths using cylindr~calspecimens 100 rnm in
diameter and 200 mm in length. Measure the drying shnnkage using beam specimens
100 x 100 x 400 mm in size at a temperature of 20 OC and relative humidity of 60%.
Test the freezing and thawing resistance in accordance with ASTM C666 procedure
A
A C 1 SP-148 9 4 Ob62949 0519806 L b 2
30 Mitsui et al
Measurement of viscosity of concrete
Because high-strength concrete with a very low water-to-binder ratio has

considerably high viscosity, the workability of the concrete should be evaluated not
only by slump tests, but also III terms of the viscosity of the concrete (i).
The viscosity of concrete was measured by an L-shaped flow tester, a handy

viscometer proposed by the authors (2) The shape of this tester is shown in Fig. 1.
Concrete is filled III the vertical part of this meter. When the gate is opened, the
concrete IS allowed to flow mto the honzontai part. Sensors are attached to the
horizontal part at distances of 50 and 100 mm from the gate. The time the concrete
takes to pass through this distance of 50 mm (T) is recorded. The average flow
velocity of the concrete over the distance, Lv = 50R (mmísec), is referred to as L-
flow velocity. It represents the plastic viscosity of the fresh concrete. When the
concrete has stopped, the lateral flow (Lf) and vertical slump (Ls) are measured. Ls
and Lf correspond to slump and flow values III conventional slump testing.
It is known that the flow properties of high-slump concrete are approximated to

those of Bingham liquid (3), for which the constitutive equation is
where 7:shear stress, 2’. shear yield stress, p:plastic viscosity, Yshear strain rate
Concrete wh i the same slump is expected to have the same shear yield stress.
When such concrete flow through the L-flow meter, shear stress histories dunng
deformahon will not differ very much The L-flow velocity is a parameter that
represents shear stram rates. Thus, it is expected, for concrete with the same slump,
that the inverse of the L-flow velocity is a parameter expressing the plastic viscosity of
the concrete
RESULTS AND DISCUSSIONS
Properties of fresh concrete
Dosages of superpiashcizers
Fig. 2 shows the relationship between the dosages of the superplasticizers and the
water-to-binder -atios with the types of cement, when the target flow is 550 mm. The
dosage of each superplasticizeris converted and expressed as the solid content.
The superphisticizers demand to obtain the required flow increased as the water-
to-binder ratio ûecreased in ail cases, but the tendencies vaned with the type of
superplasticizers. When the water-to-binder ratio was 0.30, the demand for SSP was
similar to that fbr SP, and slightly smaller than that of NSF. As the water-to-binder
ratio decreased down to 0.20, however, the demand for NSF increased, up to more
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 0662747 0517807 OT7
than 3 times the demand for SSP, which was less than 1%. As for SP, a dosage of
more than 5% did not lead to the required flow. Thus the superplasticizers showed
marked differences in their performance. In other words, when compared with
conventional superplasticizers, the effects of SSP on the fluidity of concrete becomes
more evident the water-to-binder ratio decreases, and SSP imparts sufficient
workability to thr concrete with a smaller dosage.
Viscosity of concm
Fig. 3 shows the relationship between the L-flow velocity and the water-to-binder
ratio. As explauied above, concrete with a higher L-flow velocity has lower viscosity
Both in the case of N and FB cement, lower L-flow velocity, and thus higher viscosity
were observed s the water-to-binder ratio decreased and the strength of concrete
increased. Concrete with FB cement showed higher L-flow velocity than with N
cement. When the water-to-binder ratio was 0.30, the L-flow velocity of concrete
with either one of the superplasticizers was practically the same. However, the
differences in L-flow velocity by the types of superplasticizersincreased as the water-
to-binder ratio decreased. When the water-to-binder ratio was 0.20, the L-flow
velocity of concrete containing NSF was approximately half and concrete contauiing
SP was less than 1/10, of that of concrete containing SSP. It appears that SSP
substantially reduces the viscosity, and improves the placeability, when compared
with conventiond superpiasticizers. The effect of conventional superplasticizers on
fluidizing concrete results from the electronic repulsion between cement parûcles to
the molecules of the superplasticizer adsorbed on the surface of the cement (4). in
spite of low dosage, the reason for the excellent fluidmng effect of SSP with a low
water to binder ratio is considered to be due to the synergistic effect of the
electrostatic repalsion and stabilization of the adsorbed protective layer due to the
steric hindrance of the graft chains (5).
Fig. 4 shows the relationship between the dosage, the L-flow velocity, and the
mixing temperature. in the range of a concrete temperature between 10 and 30 OC,
the demand of SSP and NSF to obtain the required flow does not vary with
temperature However, SSP showed higher L-flow-velocity as temperature rose,
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
while NSF showed no difference with temperature. in other words, as the

temperature rises, the NSF's effect of improwig the fluidity remains unchanged,
whereas SSPs effect increases
Loss of fluidity over time after mimg

Fig 5 shows the relationship between the loss of flow and L-flow velocity and the
temperature hi the case of 10 and 2OoC, the flow of concrete contauung SSP
remained practically the same for 2 hours after m m g . The L-flow velocity also
showed little loss, indicating good fluidity retention effects even with a veq low
water-to-binder ratio. Concrete containing NSF showed slightly larger loss in flow
than in the case of SSP, but the loss of L-flow velocity was much greater than in the
case of SSP. li indicates that, III high-strength concrete , the viscosity of concrete
would increase over time after mixing while the flow would not show sipficant loss,
and it varies with type of superplasticizer The slump loss is considered to be caused
by the physical agglomeration of particles (6). The stenc hdrance of SSP's graft
A C 1 SP-I148 9 4 m 0662999 0539808 T35 m
32 Mitsui et al
chains is considered to be effective in preventmg this physical agglomeration

especially with kery low water to bindei raho.
Setting time
The results of the sei-hng time tests are tabulated in Table 3. Because concrete
contauung a IoNer water-to-binder ratio requires a higher dosage of superplasticizer,
the semng time iends to be delayed. With a water-to-binder ratio of 0.25, the dosages
of the superplasticizers, especially SP, were so high that proper semng was
considerably delayed. Concrete contaming SP did not set even at 7 days after mimg.
Semng of concrete contauiing NSF also tended to be delayed substantially. On the
other hand, the sethng time of concrete containing SSP with a water-to-binder ratio of
0.20 was faster than that of concrete contauiing NSF with a water-to-binder ratio of
0.30. Thus the concrete mcorporatmg SSP did not show such excessive set-retarding
effects as found m conventional superplasticizers, even with a very low water-to-
binder ratio.
Properties of hiirdened concrete
Compressive strzngth
Fig. 6 shows the results of compression tests at 7, 28, and 91 days. With water-
to-binder ratios of 0.30 and 0.20, no appreciable difference was observed between the
effects of the superplasticizers on the strength development. With a water-to-binder
ratio of 0.20, however, the concrete contauiuig SP showed unsatisfactory setting
charactenstics due to an overdose. It had not set at 7 days, and the strengths at later
ages were substantially low Though the concrete contauiuig NSF attained a similar
strength at 91 days as that contauung SSP, the strengths at 7 and 28 days were lower
than that contauiing SSP strengths This may be due to the marked delay caused by
the high dosage of NSF, similar to the case of the concrete mcorporahng SP.
Drying shnnkagc
Fig. 7 shows the results of drying shnnkage tests. The drying shnnkage of the
concrete containing SSP tended to be slightly lower than that of concrete containmg
NSF The concrete contauiing either SSP or NSF showed smaller drymg shnnkage
with water-to-biiider ratios of 0.25 and 0.20 than with water-to-bmder ratio of 0.30.
Freezing and thawinn resistance

Fig. 8 shows the results of freezing and thawmg resistance tests. Because the a r
content was set to as low as 2.0% m this investigation, great losses in the relative
dynamic modulus of elashcity were observed in spite of the low water-to-binder ratio
of 0.25 and 0.20. However, the concrete contauiing the SSP indicated higher
durability than that contaumg NSF. The concrete containmg the SSP with a water-to-
binder ratio of O 20 showed no loss in the relative dynamic modulus of elasticity up to
300 cycles of fieezmg and thawmg, in spite of the ar content bemg as low as 2.0%. It
IS possible to improve freezing and thawing resistance without entraining au m such
high strength coiicrete.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-LYB 94 0662749 05L9809 971
APPLICATION OF SSP FOR A 100 MPa HIGH-STRENGTH CONCRETE

BRIDGE
Based on the results of the laboratory tests, the newly developed superplasticizer
was applied to il cast-ui-place ultra high-strength prestressed concrete bndge with a
design strength of 1O0 MPa. The concrete was of the highest strength ever cast III situ
for structures m Japan. The structure is a bndge connectmg two buildings. As shown
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
in Fig. 9, the ratio of the beam depth to the span is as small as 1/40. Therefore the
design strength was set at 1O0 ma, so as to ensure the strength and ngidity required
for the structure. The concrete was proportioned as presented m Table 4, so as to
satisfy the requirements established from the results of the laboratory tests, i.e
W/(C+SF) = 0.20, slump = 250 mm, flow = 550 mm and a r content = 2.0%.
The concrete was produced III batches of 1.5 m3 with a pan-type mxer at a ready-
mixed concrete plant. Silica fume was charged into the mixer by a special device,
which was developed by the authors, (as shown in Fig. 10). This device is designed
to automatically store, convey, batch, and charge undensified silica fume, which has
stronger effects of increasing the fluidity and strength of concrete than densified silica
fume (7). The mxmg time was set at 70 sec, which is 1.5 times the time for normal
concrete with WIC around 0.55
MOCK-UP TES2
Before the actual application, a construction test was conducted using a füll-scale
model, in order to study the quality and placeability of the fresh concrete, as well as
the surface texture and strength development of the hardened concrete. The model
was made to have identical cross-sections and bar arrangement as the bndge, but was
1/10 in length. The filling capability of the concrete was investigated by using
transparent forms. The concrete was deposited with a bucket, and was consolidated
with an internal vibrator.

Table 5 presents the results of the tests on fresh concrete. Though the concrete
amved at the site 1O0 m u t e s after mixing, changes m the qualities of fresh concrete
were marginal The required qualities such as slump = 250 mm, flow = 550 mm and
ar content = 2.03/0,were retained at the time of placing.
Compressive strength development

Fig. 11 shows the results of compressive strength tests on the control specimens
and on cores dnlled from the model. The controì specimens were cured in water at 20
OC, seal-cured on site, or seal-cured to be synchromed with the thermal history
monitored at the center of the model Both water-cured and field-cured specimens
indicated 11O MPa at 28 day, exceeding the design strength. At 9 1 days, the water-
cured and field-cured specimens had compressive strength of 140 MPa and 121 MPa,
respectively The dnlled cores and the specimens synchromzed with the thermal
A C 1 SP-248 94 0662949 0.519810 6 9 3
34 Mitsui et al
history of the model also reached strength of 115 MPa at 28 days. Thus it was
confirmed that concrete with sufficient strength of the structure was obtmable.
Thermal history of the model

Fig. 12 shows the thermal history of the model monitored with thermocouples
embedded in thc center and near the surface of the model. The temperature of the
concrete at the time of placing was 18.4 OC, and no delay III setting was observed.
The temperature began to nse 11 hours after placmg, due to the heat of hydration, and
the temperature 111 the center reached a maximum of 55 OC 3 1 hours after mimg. It
has been pointed out that, in high-strength concrete, high volume cement content
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
causes high temperature rises wthm structures, resulting m lower strength

development 111 dnlled cores than field-cured specimens. Here the temperature rise
was curbed, because the new superplasúcizer enabled the unit water content to be
limited to 135 kg/m3, and the use of fly ash and silica fume enabled the cement
content to be limited to 470 kg/m3.
Filling of concrete
Fig. 13 shows the vemcal distnbunon of compressive strength and coarse
aggregate content tested by cores dnlled verhcally across the center of the model. The
vertical vanations of the strength and aggregate content were small, and no
segregation or defective filling was observed. Adequate filling of the concrete was
confirmed by uispection of the transparent forms and by the appearance of the
concrete after removal of the forms.
ACTUAL CONSTRUCTION
As in the mock-up expenment, concrete m the actual construction was deposited

with buckets. Because of no accumiilation of bleeding water, the surface of the
concrete was kept moist by being sprayed with water from immediately after placing
through the final settmg, so as to avoid plastic shrinkage cracks due to rapid diying.

Fig. 14 shows the results of the tests on the fresh concrete immediately after
mixing and upon arnval at the site. Though it took 2.5 hours to haul the concrete,
changes in the quality over time were marginal, and all testing lots satisfied the quality
requirements ïlus is because the atmospheric temperature was 14 O C , at which the
new superplasticizer's fluidity-retainingeffects are Mly exerted.
Comuressive strength development

Fig. 15 shows the results of the compressive strength tests. The strength of the
water-cured specimens was 127 MPa at 28 days and 134 MPa at 91 days. The
strength of the specimens seal-cured on site was 62 MPa at 5 days, when prestressing
was applied, and 110 MPa at 91 days. The synchronized specimens had a strength of
112 MPa. Thus the concrete y attamed required the design strength.
A C 1 SP-14ô 9 4 0662949 05L98LL 5 2 T
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 16 shows the appearance of the bndge after the completion. The concrete
filled every comer of the forms without any cold jomt or honeycomb, and was III
accordance with the quality control requirements.
CONCLUSIONS
The newly developed acrylic graft copolymer-based superplasticizer greatly

improves the fluidity of ultra high-strength concrete with an extremely low water-to-
binder ratio. This enabled an ultra high-strength concrete with a design strength of
100 MPa to be placed on site.
I The newly developed acrylic graft copolymer-based superpiashcizer reduces the

viscosity of concrete with a lower dosage than conventionai carboxylate
superplasticizers and naphthalene superplasticizers. The viscosity-reducing effect of
the new superpiashcizer, when col spared with conventionai superplasticizers,
becomes more evident as the water-to-binder ratio decreases.
2 The new superpiashcizer mamtams fluidity of even an ultra high-strength concrete
with a water-to-binder rah0 of 0.20 for 2 hours without appreciable losses in the flow
and viscosity. Also, concrete contauung this superplashcizer develops sufficient
strength without excessive retardahon m semng, and has the same or better durability
than concrete contauung conventional superplasticizers.
3 An ultra high-strength concrete made usmg the new superplasticizerwith a design
strength of 1O0 MPa was applied to a prestressed concrete bndge, and was confirmed
to have good placeability on site and adequate strength development.
REFERENCES
( 1 )Tmgawa,Y. and Mon H., SuperplasticizedSteel Fiber Reinforced High-Strength
Concrete, Transiiction of Japan Concret? Institute, Vo1.6,1984, pp.89-94
(2)Yonezawa,Y ,bumi I.,Mitsui, K.and Okuno, T.,A Study on the Flowability of

High-Strength Concrete using the L-flow testproceedings of Japan Concrete
Inshtute,Vol.l1,1989, pp.171-176
(3)Tattersall, G.H. and Banfdl, P.F.G, The Rheology of Fresh Concrete, Pitman
Advanced Publishing Program, 1983
(4)Damon, M. and Roy, D.M., Rheological Properties of Mixes: Zeta Potential and
Prelimary Viswsity Studies, Cement and Concrete Research, Vo1.9, 1979, pp. 103-
I10
(S)Kmoshita, M , Yonezawa, T. and Yuki, Y., Chemical Structure and Performance

of New Type €úgh-Range Water Reducmg Agent, Proceedings of Japan Cement
Association, No.44,1990, pp.222-227
(6)Hatton, K., Mechanism of Slump Loss and Its Control, J o m d of the Society of
Materials Science,Japan, Vo1.29, 1980, pp.240-246
36 Mitsui et al
(7)Yonezawq T , Izm, I., Okuno T. and Asakura, E.; Reducing Viscosity of High-
Strength Concrete Using Silica Fume, Supplemenkuy Papers, 4th International
Conference on Fly Ash, Silica Fume, Slag and Natural Pozzolans UI Concrete,
Istanbul, May, 1992, pp.765-780
TABLE 1 - PHYSICAL PROPERTIES AND CHEMICAL COMPOSITION OF THE SILICA FUME
Specific Bulk Surface Si02 Carbon Moisture Losson

gravity densir ,2,
area Ignition
(kg/m ) ( g) (YO) (YO) (YO) pó)
2.35 325 14.1 93 8 O. 78 0.40 2.76
TABLE 2 - MIXTURE PROPORTIONS USING FLY ASH CEMENT

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 3 - TIME OF SETTING
I Super- I SettingTime(h0ur)
lasticizer
W/(C+SF)
0.30 7.17 8.50

0.25 SSP 7.75 9.17
0.20 8.58 10.33
0.30 7.75 9.50
0.25 SP 9.00 10.17
o.20 * *
0.30 9.42 10.58
0.25 NSF 9.83 11.17
0.20 11.08 12.17
A C 1 SP-148 7 4 Obb2747 0517813 3 T 2 =
TABLE 4 - MIXTURE PROPORTION OF 100 MPo HIGH STRENGTH CONCRETE

W/(C+SF) Unit Weight (kg/m3)
Water Cement Silica Fine Coarse Super-
Fume Aggregate Aggregate plastxcizer
020 135 574 101 748 836 13.5
Testing Time after slump flow Air Concrete L-flow

Point rnimg content Temperature Velocity
(inin.) (mm1 (mm) (%) (“c) (&sec)
plant O 260 605 2.1 18.0 96
site 1 O0 250 540 2.3 18.4 56
200/
400
c-i Lf
Fig. 1-L-flow meter
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662947 0519814 2 3 9
38 Mitsui et al
-e- SSP
-e- SP 5.0 -e- SP
-E-.NSF
\
4.0 \
m
Co
\
\
Co \
8 1.0 y----
V
--
c.20 0.25 0.30 0.20 0.25 0.30
water to binder ratio water to binder ratio
(a) Fly ash cement (b) Normal Portland cement
Fig. 2-Relationship between W/(C + SF) and dosage of superplasticizer
JU.U‘
JU.UT
-
A
a,
O
20.0 4
1
g- 5.0- /
2 3.0- Cr’ / I! 3.0-
3
c
1
2.0-
/
’ +SSP
-e-SP
B 2.0-
F
F 2.0-
J
1.0-
o8
/
/
+SSP
-e- SP
1.0-
4s / *-.
NSF 45 / -E-.NSF
c,
n- I O
0.20 0.25 0.30 0.20 0.25 0.30
waterto binder ratio water to binder ratio
(a) Fly ash cement (b) Normal Portland cement
Fig. LRelationship between W/(C + SF) and 1-flow velocity
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
I A C 1 SP-14ô 9 4 E 0662947 05L9B15 175 E
I
+NSF:: W/(C+SF)=O.20
-E- .NSF::W/(CtSF)=00.30
-
N +SSP: W/(CtSF)=0.20
+. SSP W/(CtSF)=O.30
t;;
.z5.0-
-
%
f 3.0-
B NSF W/(C+s F)=o.îc
*----u 9
"7
2.0-
-1
1.0-
-U
0- - -- -Q- - -- -0
O I I
10 20 30 1'0 20 30
temperature ( O C ) temperature ("C )
(a dosage of superplasticizer (b) L-flow velocity
Fig. +Effect of temperature on dosage of superplasticizer and 1-flow velocity (fly ash cement)
80
50
40
30
20
10
~~ 20% +-c-
~+---~--
30°C +-L.
o0 30 60 90 120
5
O 30 60 90 120
time aiter mixing (min.) time after mixing (min.)
(a) loss of flow (b) loss fo L-flow velocity
Fig. !LLoss of fluidity of concrete with time after mixing (fly ash cement, W/(C + SF) = 0.20)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 m Obb2949 051981b O O L m
40 Mitsui et al
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
SSP SP NSF CSP SP NSF SSP SP NSF

W/(C+SF)=O.30 W/(C+SF)s).25 W/(CtSF)s).20
Fig. ó-Compressive strength development (fly ash cement)
duration of drying (weeks)

5 10 15 20 25 :O
O'
I 1
SSP NSF
9
- 100 ' w/c=0.20-+ -t
4
~
X
v
~ w/c=o.25- -t ,
g 200
22
.E
r 300
m
al
'g 400
U
500
Fig. 7-Relationship between drying shrinkage and age (fly ash cement)
A C 1 SP-148 94 0 6 6 2 9 4 9 0519817 T 4 8 =
freezing and thawing cycles

h
0 50 100 150 200 250 300
òe 120 i-
v
a
J
3 100
U
E 80
.-
o
60
c
6 40
+
-
al
.-
>
<u 20
w/c=O.20 -- -c-
?
O
Fig. &Freezing and thawing resistance (fly ash cement)
Modd on Mock-upTest
S~an:41.216m /
u b-b Cross Section

a-a Cross Section
Fig. W i r d e n and cross-sections of the prestressed concrete bridge
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 05L98Lô 984 W
42 Mitsui et al
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 1û-A special device for using silica fume and ready-mixed concrete plant
A C 1 SP-148 74 I0662747 0517819 810
-
IW
m
g
r
140-
‘Ei>
= 120-
L
tj
curing conditions
0 20°C in water
0 seal-cured on site
A simulatedtemperature of the model
o core
I I
28 91
age (days)
Fig. 1 I-Compressive strength development on mock-up test
60
50
10
O
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 12-Temperature of the concrete inside the structure aíter placement
A C 1 SP-L48 9 4 M 0662947 0537820 532 =
44 Mitsui et al
" 80 90 100 110 120 20 40 60

compressive strength (MPa) volume ratio of coarse agg. (%)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 13-Verticol distribution of the core strength and amount of coarse

aggregate of the model structure
-1
300
--___
3% -
UY
03
plant siîe plant siîe
(a) slump (b) slump flow spread
8 30
plant site plant site

(c) air content (d) L-flow veioaíy
Fig. 14-Properties of fresh concrete on actual construction
A C 1 SP-148 9 4 W 0662947 05L9821 479 W
I UV
-
140
---o
a ___----o
5 - ___---
r 100
F"
al
p
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
-1 60 curing conditions
g 0 20°C inwater
U seal-cured an site
A under simulated temperature c. the model
37 28 91
age (day)
Fig. ILRelationship between compressive strength and age on actual construction
Fig. Ió-Appearance of the bridge using 100 MPa ultra high strength concrete after completion
A C 1 SP-148 94 0 6 6 2 9 4 9 0539822 305 H
SP 148-3
Cement/Su perplasticizer/Air-Entraining
Agent Compatibility
by M. Baalbaki and P.C. Akin
S v n o D s i s : It is well-known that in order to pass successfully

ASTM C666 (Procedure A) for rapid freezing and thawing resistance,
normal-strength concrete must contain an adequate amount of
entrained air composed of minute air bubbles with the right spacing
factor.
As concrete slump is increasingly restored at the jobsite using

superplasticizer instead of retempering with water, it is essential that
slump increase does not alter the total air content and the air-void
system if the concrete is to be frost resistant.
Since mixed results have been reported when superplasticizer

is added to air-entrained concrete at the jobsite, a research program
was undertaken to study the compatibility between three air-
entraining agents, four water reducers, and one polynaphthalene
sulfonate superplasticizer currently used in Eastern Canada.
Experimental r e s u l t s conducted on t w e l v e d i f f e r e n t
combinations of admixtures with a Type 10 (ASTM Type I) portland
cement show that the addition of superplasticizer nearly always
increased the air content without changing the bubble spacing. The
only case in which the air bubble spacing was significantly altered
was when the air content of the concrete was lower than 4.5 percent
70 minutes after batching. In this case, the total air content
decreased after the introduction of the superplasticizer, while the
spacing factor increased significantly.
A second Type 10 cement was used to duplicate these results.

No significant difference was found between the results of the two
sets of experiments.
Keywords: Air entrained concretes; air entrainina oaents; cements; cammibilitv methods; freeze
thaw durability; plasticizers; scaling; slump; suDerdasticizen
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
COPYRIGHT 2003; ACI International (American Concrete Institute) 47 Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-148 94 O b 6 2 9 4 9 0539823 243 m
48 Baalbaki and Akin
Pierre-Claude Aitcin is a Professor of Civil Engineering at the

Université de Sherbrooke. He is a member of AC1 Committees 234.
Silica Fume in Concrete, and 363, High-Strength Concrete.
Moussa Baalbaki is a research engineer in the Civil Engineering

Department at the Université de Sherbrooke, Québec, Canada.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
INTRODUCTION
It is well-known that in order to pass successfully ASTM C666
(Procedure A) for rapid freezing and thawing resistance, normal-
strength concrete must contain an adequate amount of entrained air
composed of minute air bubbles with the right spacing factor (1-4).
As concrete slump is increasingly restored at the jobsite using

superplasticizer instead of retempering with water, it is essential that
slump increase with superplasticizer does not alter the total air
content and the bubble spacing factor if the placed concrete is to
have freeze-thaw resistance ( 5 ) .
Few documented field data have been published despite

conflicting results in the cases when superplasticizers are used to
restore concrete workability (6-9). In some cases, the slump increase
is reported to be accompanied by a loss of the total air content, while,
in others an increase in the air content is observed.
An extensive research program involving different

commercial air-entraining agents, water reducers, and one
polynaphthalene superplasticizer was undertaken with two Type 10
cements (ASTM Type I) to investigate the effect of superplasticizer on
the air-void stability of air-entrained concrete.
EXPERIMENTAL STUDY
Admixtures currently used in Eastern Canada were selected for
the experimental study and are listed in Table 1. The twelve possible
combinations of air-entraining agents and water reducers were
studied even though some were commercially incompatible.
Moreover, five combinations were tested twice in order to check the
reproducibility of the results.
The concrete mixture selected for this experimental study

corresponds to class C1 concrete according to CSA-A23.1 M90. This
class of concrete is usually specified in Canada for bridge decks,
floors, sidewalks or parking garage and inclined ramps. For
freeze/thaw and scaling resistance, such concrete is required a
water/cement ratio not higher than 0.40 and a total air content of 5 to
8 percent.
A C 1 SP-148 94 = 0662949 0519824 LBB
Each batch of concrete was prepared in 70 litres (170 kg). It

was made in a laboratory tilt mixer with a nominal capacity of i00
litres. Slump was measured 10 minutes after mixing, then 40 minutes
later. and finally 70 minutes later, before and after the addition of
0.7 L/m3 of superplasticizer. Concrete compositions and properties
are given in Table 2.
In order to cross-check the measured values of the air content,

the unit weight of the fresh concrete was measured for all mixes, the
results are presented in Fig. 3. The linear relationship between the
unit weight and the total air content is quite good with a high
correlation index showing that both measurements were done
properly. The unit weight of a zero percent air concrete can be
extrapolated to 2453 kg/m3 and each percent of air resulted in a
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
2 5 k g / m 3 decrease of the unit weight. Finally, four more sets of

concrete mixtures were done using a different brand of Type 10
cement.
RESULTS AND DISCUSSION

Presentation of the resulb
As seen in Fig. 1 (a, b, c , and d), the initial slump varied

between 110 and 150 mm, while the initial air content varied from 6
to 8 percent. Seventy minutes after batching, the slump dropped to
70-90 mm, while the air content ranged between 5.5 to 7 percent, as
shown in Fig. 2 (a, b, c, and d) except in one case (Fig. 2a). where it
was 4 percent (AE#3 - WR#A).
The addition of superplasticizer 70 minutes after batching

resulted in a slump increase, and the final slump was equal to 100 -
150 mm, as seen in Fig. 1 (a, b, c, and d). In ali cases except mix AE#3
- WR#A, the a i r content increased after the addition of the
superplasticizer from 5.4 to 7.8 percent, as shown in Fig. 2(a, b, c, and
d).
Bubble spacing factors and specific surfaces before and after

the addition of the superplasticizer are presented in Fig. 4. Figure 5
gives the slump and air content variations observed with the second
brand of Type 10 cement and the four different admixture
combinations, 10 minutes after mixing, 70 minutes later before the
addition of superplasticizer, and 10 minutes after its addition.
A C 1 SP-148 74 = Ob62949 0519825 014 W
50 Baalbaki and Aïtcin
The relation between compressive strength and air content in

fresh concrete i s presented in Fig. 6. Moreover, the scaling
resistance values of the mixtures made with Cement A are presented
in Fig. 7.
i3luudm
During t h e first 70 minutes, all t h e tested concretes
experienced a similar slump loss, averaging 70-90 mm. In order to
avoid any segregation o r bleeding, so that the top surface of concrete
can exhibit good deicing-salt scaling resistance, it was decided to
adjust the superplasticizer dosage so that the final slump of the
concrete would be around 150 mm. It was found that, for the
particular mix under study, a dosage of 0.7 L/m3 would increase the
slump from 70-90 mm to 120-150 mm.
Figure 8 shows that all concrete mixtures experienced an air

loss before superplasticizer addition. However, the use of the A.E.#3
air-entraining agent resulted in a greater (26%) air loss than the
other two (17%). After the superplasticizer addition, the air content
increased by around 12-16% with A.E.#l and 2-7% with A.E.#3. In
almost all cases, after the introduction of the superplasticizer, the air
content went back up to nearly the initial value except when the
combination of W.R.#A and A.E.#3 was used (see Figure 2a). It should
be pointed out that this was the only case for which the amount of
entrained air dropped to less than 4.5 percent. In all the other cases
the amount of air before the introduction of the superplasticizer was
higher than 5 percent.
The last WR#A, AE#3 mix was reproduced but this time with a
higher initial a i r content of 7.6 percent so that before the
introduction of the superplasticizer the air content was 5.3 percent.
After the introduction of the superplasticizer the air content was 5.4
percent.
T h e spacing factors that have been obtained after the

introduction of the superplasticizer were well under the 230 pm
value specified by CSA A23.1 for freeze-thaw resistance, except for
the concrete made with the combination WR#A, AE#3. It should be
mentioned that the spacing factor for this mix was nearly at the
maximum limit of CSA specification before the introduction of the
superplasticizer. This is confirmed in Figure 4b, where it can be seen
that this mix had the lowest specific surface area, meaning that the
air bubbles were coarser than in the other mixes.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = Obb2949 0539826 T50 m
f the results another brand of cement
Figure 5a shows that Cement B yielded slightly higher slumps

than Cement A when used with air-entraining agent AE#3. The air
loss was always greater when this particular air-entraining agent
was used. However, after the introduction of the superplasticizer, the
air content increased as when Cement A was used.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Scaling resistance
Scaling resistance was tested according to ASTM C672 using a 3

percent sodium chloride solution. The results, presented in Fig. 7,
show that the mass loss after 50 cycles falls between 100 and 200
g/m2, which is excellent. Sweedish standard SS137244 suggests that a
1 kg/m2 mass loss is the maximum value accepted for good scaling
resistance after 56 cycles.
In Fig. 7b, it is seen that the mix made with the water reducer
WR#3 shows a higher initial mass loss, but that after 20 cycles, the
mass loss continues at the same rate as for other concretes. This can
be attributed to the weakness of the concrete skin in this particular
case. This water reducer is a strong retarder and tends to promote
bleeding, so that the top surface of the concrete has a weaker
microstructure. Once this thin layer of weak concrete is gone, the
remaining concrete present a good scaling resistance.
Cornmessive s t renpth
Since all the tested concretes did not contain the same amount
of entrained air, some differences were observed in the compressive
strength in spite of the fact that the concrete had the same water-
cement ratio. Figure. 6 illustrates the relationship between air
content and compressive strength. It is seen that, for these 35-MPa
concretes (on average), a one percent air content increase results in
a loss of 2 MPa less compressive strength, which in terms of
percentage of the initial compressive strength, represents a loss of
approximately 6 percent.
CONCLUSION
Under the selected experimental conditions, it was found that,

in almost all cases, the addition of superplasticizer 70 minutes after
batching resulted in a slight air increase with no significant
modification of the bubble spacing, except in one case. In that case,
the air content at 70 minutes was lower than 4.5 percent and dropped
to 3.8 percent after the introduction of the superplasticizer. A
similar concrete with a slightly higher initial air content of 7.6
percent (instead of 5.2 percent) had a 85-mm slump after 70 minutes.
A C 1 SP-14B 94 0662949 0519827 997 =
In this case, the addition of the superplasticizer resulted in an

increase in the air content.
Some of the mixes were duplicated using another brand of
Type 10 cement with almost the same experimental results. The
addition of the superplasticizer did not significantly modify the
bubble spacing factor, with the one exception cited above.
Therefore, it seems that the mixed results reported in the field could
be explained in terms of slump value when the superplasticizer is
added rather than as an incompatibility problem. However, as is
always the case with concrete, it could be dangerous to generalise the
results obtained in these particular experimental conditions and to
apply them to all kinds of admixtures and cements.
This research was sponsored and financed by the members of

the Industrial Chair on Concrete of the University of Sherbrooke, and
the Natural Science and Engineering Research Council of Canada.
REFERENCES
PIGEON, M., 1989, "La durabilité au gel du béton", Matériaux et
construction, Vol. 22, No.127, pp. 3-14.
PIGEON, M., 1992, "The Frost Durability of Concrete", Cement
and Concrete Science and Technology, S.N.Ghost Ed., ABI
Books PvT, New Delhi, India, Vol. 1, Part II, pp. 417-448.
PIGEON, M., PLEAU, R., and AITCIN, P.C.. 1986, "Freeze-Thaw
Durability of Concrete With and Without Silica Fume on ASTM
C666 (Procedure A) Test Method : Internal Cracking Versus
Scaling", Cement, Concrete, and Aggregate, Vol. 8, No. 2, pp. 76-
85.
PIGEON, M., PERRATON, D., and PLEAU, R., 1987, "Scaling Test of
Silica Fume Concrete and the Critical Spacing Factor Concept",
Katherine and Bryant Mather International Conference on
Concrete Durability, AC1 Special Publication SP-100, John M.
Scanlon Ed., pp. 1155-1182.
PIGEON, M., SAUCIER, F., and PLANTES., 1990, "Air Void Stability
Part IV : Retempering", AC1 Materials Journal. Vol. 87, No. 3,
pp. 252-259.
PLANTE, P., SAUCIER, F., and PIGEON, M., 1989, "1989. "Air Void
Stability Part II : Influence of Superplasticizer and Cement",
AC1 Materials Journal. Vol. 86, No. 6, pp. 581-589.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 0662947 0519828 8 2 3 m
(7) SAUCIER, F.. PIGEON, M., and PLANTE. P., 1990, "Air Void
Stability Part III : Field Tests of Superplasticized
Concretes", AC1 Materials Journal, Vol. 87, No. 1 , pp. 3-11.
(8) PIGEON, M., and LANGLOIS, M., 1991, "Étude de la résistance au

gel de bétons contenant un fluidifiant", Canadian Journal of
Civil Engineering, Vol. 18, No. 4, pp. 581-589.
(9) PIGEON, M., PLEAU, R., PLANTE, P.,and BANTHIA, N., 1992, "The
Influence of Soluble Alkalis on the Production and Stability
of the Air-Void System in Superplasticized and Non-
Superplasticized Concrete", AC1 Materials Journal, Vol. 89, No.
1, pp. 24-31.
TABLE 1 - ADMIXTURES
Chemical nature
A.E.#l Fatty acid salt 11
& I
entraining
agent
A.E.#2 I Fatty acid salt I 20 I
A.E.#3 Synthetic agent 5
I W.R.#A I Lignosulfonate I 41 I
Water W.R. #B Hydroxycarboxylic acid
reducer
W.R.#C Carbohydrate
W.R. #D Lignosulfonate 40
Superplasticizer Polynaphthalene sulfonate 42
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 94 0 b b 2 9 4 9 0539829 7bT U
TABLE 2 - CONCRETE COMPOSITION AND PROPERTIES
W/C I 0.38' I
w- I 170** I
cement I 450
Fme
620
aggregate
com 1M5
0.7
W.R. #A W.R. #B W.R. #î W.R. #D
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
WaQTredUcer
(mU100kg
of m e n t ) 400 230 300 300
** 165 with cement B for validation test
A C 1 SP-I148 9 4 = 0662949 0519830 481 =
200
180 180
5 160 g 160
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
-g
v
140 2 140
0
8 lu)
loo
$
5;
lm
100
80 80
60 60
o 20 40 60 80 1M O 20 40 60 80 O
Elapsed time (min) Elapsed t i e (min)
200
180 180
5 160 g 160
-
v v
3 140 2 140
e
g
a
*
IM
100
5 l2O
100
80 80
60 60
O M 40 60 80 100 O 20 40 60 80 LM
Elapsed time (min) Elapsed t i e (min)
Fig. 1-Slump loss obtained with Cement A and the different air entraining agents combined
with lignosulfonate [(a), (d)], hydroxycarboxylic (b), and carbohydrate, (c) water reducers
A C 1 SP-148 94 0662949 0519831 318
6-
5.
Ö
-
AH.#I
4---0-
-C
2 i
A.E.#2
A.E#3
4 i o s o s o i o o o
-
a
A.E.#2
o 4 0 s o 8 0 1 a o
Elapsed h e (min) Elapsed time (min)
9-
(C)
* v5
dditim
8-
5s
7-
2
6-
-P
.!i
5.
- o
5-
--
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
4- A.EI1 .!a 4- A.E.# 1
. --O-
--C AE.#2
A.E.#3
4 ,
--O-
A.E.#2
A.E.#3
Fig. 2 4 i r content variation in fresh concrete obtained with Cement A and the different air
entraining agents combined with lignosulfonate [(a), (d)], hydroxycarboxyiic (b), and
carbohydrate (c) water reducers
A C 1 SP-148 9 4 0 6 6 2 9 4 9 0539832 254 W
Y = 2453 - 25 X
2340 R2 = 0.9
N=51
2320j
22ml
2240
2220
3
!
4
I
5
I
6
I
7
I .
"\
I
Air content in fresh concrete
Fig. 3-Relation between air content and unit weight in fresh concrete
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 = 0 6 6 2 9 4 7 0539833 190
Before superplasticizer After superplasticizer

addition addition
60
e W.R.#D- A.E#1
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
50 WR.#B - AE.#1
(b) W .#C - AE#.2
40
W.R.#D - A.E#2
30
20 WR.#A - AE.#3
10
C I I
Before superplasticizer After superplasticizex

addition addition
Fig. &Variation of the spacing factor (a) and specific surface (b) before and after
superplasticizer addition
A C 1 SP-148 9 4 M Ob62949 0519834 027 D
After
superplasticizer
aàdition
1801
160
140 -
120 -
100 -
80 - W.R
1
O 20 40 60 80 100
ElapEd lime (min)
After
WR.#B - A.E.#3 superplastic izer
additi
10 -
- WR.#A - A.E#3
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
8-
. W.R.#A - A.E.#l
WR.#B - A.E.#l
J I I - 1 I
O 20 40 60 80 100
Elapsed time (min)
Fig. L S l u m p (a) and air content variation in fresh concrete (b) for concretes made
with Cement B
A C 1 SP-148 94 0 6 6 2 9 4 9 0519835 Tb3 H
45
Y = 49-2X
RA2= 0.7
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
B
99
N
25
Air content in fresh concrete (%)
Fig. blelotionship between compressive strength and air content in fresh concrete
A C 1 SP-14B 94 0662949 0 5 3 9 8 3 6 9 T T m
AX# 1
AEJ 2
ABI3
-lml O
O 1 0 2 0 M 4 0 5 0
Il-( . , . , . , . , . .
0 1 0 u ) ~ 4 0 5 0 6 0
Number of freeze-thaw cycles Number of freeze-thaw cycles
2 800
100
O
0 1 0 a ) M 4 0 M 6 0 0 1 0 ~ 3 0 4 0 s 0 6 0
Number of freeze-thaw cycles Number of freeze-thaw cycles
Fig. 7-Scaled material versus number of cycles obtained with Cement A and the different air
entraining agents combined with lignosulfonate [(a), (d)], hydroxycarboxylic (b), and carbohydrate
(c) water reducers
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-AY8 7Y m Obb29q9 0539837 836 m
8-
A.E.#2
1-
A.E.#l
6-
A.E.#3
5- .Y
4-
3. I I I
10 min 70 min 80 min
Elapsed time
Fig. Wir lor;

* The air lors is expressed M a percentage
\F(aj - ar, a$ x 100
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Where ai = initial air content

af = final air content
A C 1 SP-148 9 4 M 0662949 0519838 7 7 2
SP 148-4
Progress in Understanding the

Functional Properties of
Superplasticizers in Fresh Concrete
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
by C. Jolicoeur, P.C. Nkinamubanzi,

M.A. Shard, and M. Piotte
Synopsis: A variety of new and literature data on the properties of cement

pastes and concentrated slurries of various types of mineral particles are
examined in order to elucidate the origin of the fluidification of cement
pastes by superplasticizers. The influence of sodium poly-0-naphthalene
sulfonate superplasticizers (NaPNS) of different molecular weights on the
rheological properties of pastes and on the early heat of hydration of
cement, together with results from other physico-chemical measurements
(adsorption, zeta potential), suggest that the unique fluidification effect of
these admixtures depends on at least four distinct phenomena.
With reference to fluidification of slurries of "inert" minerals, the

superplasticizer effect in concrete can be understood in terms of non-specific
(physical) and specific (chemical) effects. The "physical" effects comprise:
1) Adsorption of the superplasticizer molecules by van der Waals and
electrostatic forces (direct or assisted by cations); 2) surface charging which
induces long-range interparticle repulsive forces; 3) steric hinderance
between adsorbed polymer molecules on neighbouring particles, leading to
added short-range repulsive forces. The "chemical" effect involves a
reaction of the PNS superplasticizer molecules with the most reactive sites
of cement particles (particularly C,A), substantially reducing the initial
surface hydration rate. This description is largely based on data relevant to
PNS-type superplasticizers but, with proper allowance for specific chemical
effects, it should also be valid for other types of superplasticizers.
Kewords: Adsorption; calorimeters; cement pastes; cements; heat of hydration; mineral

admixtures; rheologicol properties; slurries; superplasticizen
COPYRIGHT 2003; ACI International (American Concrete Institute) 63

C. Jolicoeur is professor of physical-chemistry at the Université de
Sherbrooke; he heads a colloidal chemistry laboratory which is part of the
Network of Centres of Excellence on High-Performance Concrete.
P.-C. Nkinamubanzi is a Ph.D. student in the colloidal chemistry

laboratory.
M.-A. Simard is a senior research associate in the colloidal chemistry group.
M. Piotte is a project manager in the colloidal chemistry laboratory.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
INTRODUCTION
Polymeric dispersants are increasingly used as water reducers and

fluidifiers (superplasticizers) in cementitious systems, not only in high
performance concrete, but in normal concretes as well. The functional
properties and benefits of these admixtures are well documented (1-3) and
their mode of action is usually described in terms of improved
deflocculation and dispersion of the cement particles. Although this simple
explanation is appealing, it does not provide the level of detail required to
account for all of the observed influences of superplasticizers in fresh
concretes, or provide a basis for the development of improved superplasti-
cizer systems.
Given the complexity of the cement mineral phases and the chemical
processes occurring in a fresh concrete, it should be expected that a variety
of physicochemical phenomena will be required to explain the behavior of
superplasticized concretes. Important variations in the performance of
superplasticizers in the presence of cements of different type or origin are
strong reminders of the chemical complexity of fresh cement pastes. A
realistic description of the mode of action of superplasticizers should
therefore take into account, both the physical effects (common to many
types of particles), and the chemical effects (specific to cement particles).
To achieve a detailed understanding of the rheological properties of

superplasticized concrete, it would be most useful to decouple the physical
and chemical parts; the former could be viewed as the phenomena
occurring at very short times (t = O) in the system, while the latter would
account for reaction processes as function of time. While a true discrimina-
tion of the phenomena at t = O , and t > O , is hardly possible, alternate
strategies can be developed to achieve the same purpose. Such a strategy is
A C 1 SP-348 94 0662949 0539840 320
being attempted in ongoing work aimed at i) elucidating the physical and

chemical aspects of the interactions between the superplasticizer molecules
and the cement particles, 2) understanding the influence of the superplas-
ticizer molecules on the initial surface hydration reactions and 3) identifying
the influence of superplasticizers on the development of the hydration
products and the microstructural evolution in the cement matrix.
STRATEGY
The rheology of slurries of particulate materials depends on

numerous factors related both to the particles and to the suspending media.
The shape, size and size distribution, chemical nature, surface area, porosity
and surface reactivity are among the main factors related to the particles.
O n the other hand, the ionizing properties of the solvent, its viscosity and
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
its content in dissolved species are important properties of the suspending
media which affect the suspension rheology.
For slurries of cement particles, the hydration reactions of the

mineral phases will also exert a major influence on the rheological
properties. The slump loss of a cement paste as function of time is a clear
manifestation of these hydration reactions. Conceptually, a cement paste
could thus be described as follows:
at t =O: a concentrated slurry of "non-reactive'' mineral particles dis-
persed in an electrolyte solution;
at t > O: a heterogeneous reaction system in which the dispersed mineral
phases react with components of the solution phase (water, ions)
to yield the hydration products.
At t = O, the rheological properties of the slurry are governed by physical
effects which depend on the properties of the particle and liquid phase as
listed above; the particle surface charge (zeta potential) and dissolved
electrolytes exert important influences. At t > O, the rheological properties
of the slurry will progressively reflect the consequences of the ongoing
hydration reactions: changes in particle size, shape, charge and surface
texture due to the formation of hydrates; nucleation and growth of new
particles; interparticle bridging by hydrate products; progressive reduction
in the quantity of available water.
To understand the nature of the superplasticizer-cement interactions

and the consequences of these interactions on the properties of fresh
concrete, an investigation has been initiated involving extensive compari-
sons between cement pastes and concentrated slurries of various types of
mineral particles. This investigation comprises systematic studies on:
A C 1 SP-148 9 4 W 0662949 0539843 267 W
66 Jolicoeur et al
- the rheological and other properties of concentrated slurries of various

types of mineral particles in aqueous electrolytes solutions, with and
without superplasticizer;
- the heat of hydration (t=O-30 minutes) of cement particles upon
immersion in aqueous solution of varying compositions, again with and
without superplasticizer;
- the rheology (t = 0-2 hrs) and heat of hydration (t = 0.5-24 hrs) of cement
pastes in the presence of superplasticizer.
Evidently, the consequences of superplasticizer-cement interactions

must also depend on the physico-chemical parameters of the superplasticizer
itself: chemical composition, molecular weight (and weight distribution),
degree of ionization, etc. Controlled variations of these parameters can
thus be used to provide important mechanistic information.
In the following pages, selected experimental data derived from these

studies are illustrated. Through these and other literature results, a compre-
hensive, though largely qualitative, description of the different contributions
to the "superplasticizer effect" can be formulated.
EXPERIMENTAL
Materials
A naphthalene-based superplasticizer (NaPNS) provided by a

Canadian producer was fractionated to obtain polymer fractions of different
molecular weights. The polymers used for measurements of the initial heat
of hydration were fractionated by selective precipitations from organic
solvents while those used in rheology, adsorption and zeta potential
measurements were fractionated by ultrafiltration (4). In all cases, the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
sulfate content of samples was low ( c 0.4 dry wt %), except for the high
molecular weight fraction obtained by precipitation which contained 3.0%.
For concentrated slurries (rheology, calorimetry), the concentrations of
superplasticizer, or of other additives, are expressed relative to the weight
of mineral or cement; for diluted suspensions (adsorption, zeta potential,
initial heat of hydration), they are relative to the water content.
A common Type-10 CSA (ASTM Type I) portland cement was used

in most experiments, except for the 0-24 heat of hydration studies where a
type-30 was used. The non-reactive minerals used are listed in Table I,
along with some of their physico-chemical properties; further details on
these minerals can be found elsewhere (5).
A C 1 SP-148 94 m 0 b b 2 9 4 9 0519842 I T 3 m
Methods
Rheoloa - Apparent viscosities were measured with a Brookfield

viscometer equipped with an "Helipath stand" which displaces the probe
vertically (an inverted T-shaped rod rotating at 12 rpm) during measure-
ment in order to avoid cavity formation. Minislump measurements were
carried out as described in reference 6. Slurries were prepared, first by
dissolving additives (electrolytes, superplasticizers) in water, then by adding
the solid to the water while mixing manually. The procedure was
completed by mixing with a high speed mechanical mixer for two minutes.
Adsomtion and zeta potential - Adsorption isotherms were

measured at 25OC. Forty milligrams of solid were suspended in 20 mL of
aqueous solution containing NaPNS of varying concentrations with or
without other additives. The NaPNS equilibrium concentrations were
measured by UV absorption. Zeta potential measurements were performed
on a Malvern Zetasizer on the same suspension as that used for the
adsorption experiments.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Initial heat of hvdration - The temperature increase was measured
when 15 g of cement were immersed in 250 mL of water containing
NaPNS of varying concentrations, or other additives in a high sensitivity
adiabatic calorimeter (7). The temperature of the slowly-stirred slurry was
recorded for approximately 45 minutes.
OBSERVATIONS ON SLURRIES OF NON-REACTIVE MINERALS
In the following paragraphs, various results derived from studies on

aqueous slurries (or pastes) of several types of mineral particles are
examined. Variations in the composition of the liquid phase (pH, electro-
lytes, superplasticizer) of these slurries provide broad illustrations of the
phenomenology of mineral pastes; they also provide a simple basis for
understanding the "non-specific"effects in a cement paste and the influence
of superplasticizer on the latter.
Fluidity of mineral slurries
Influence of dissolved electrolytes - The amount of dissolved

electrolytes (ionic strength) exerts an important influence on the fluidity of
most slurries of mineral particles. Fig. 1 illustrates the effect of NaC1 and
of a superplasticizer on the fluidity of an aqueous slurry of quartz particles.
Fig. l a shows the minislump spread area of the quartz paste in water;
addition of NaC1 to this paste leads to an important loss of fluidity (Fig.
ib), while subsequent addition of PNS re-fluidifies the paste (Fig. ic).
68 Jolicoeur et al
The paste "stiffening" effect observed upon addition of electrolytes

can be attributed largely to electrostatic forces. Due t o surface ionization,
or ion adsorption, the particles in contact with water initially develop a
surface charge; the latter creates a long-range electrostatic potential which
causes repulsive interactions between particles, minimizing their self-
aggregation and the viscosity of the slurry. The addition of significant
quantities (e.g., 0.1%) of soluble electrolytes leads to a screening of the
repulsive potentials allowing the particles to come closer and interact
through attractive van der Waals forces (i,e., as in the basic concept of
coagulation) (8). Other electrostatic effects, such as the electroviscous
effects, also contribute to the stiffening of the slurry upon the addition of
electrolytes (8). The fluidifying effect of the superplasticizer may then be
viewed as a reversal of the electrolyte effect: the charged superplasticizer
molecules adsorb onto the mineral particles, increasing their surface charge,
thus re-establishing interparticle repulsive forces.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
These views are consistent with the basic principles of colloid

stability (8), and readily explain why superplasticizers and similar com-
pounds are broadly used as dispersants in suspensions and slurries.
However, before these concepts can be generalized to concentrated mineral
slurries, a note of caution is in order: the long-range electrostatic forces
invoked to explain the dispersion of colloidal particles in dilute suspensions
cannot be entirely adequate to described interactions between closely
packed particles in a paste, i.e., when the particles are only surrounded by
a film of water. Localized electrostatic effects and short-range forces may
play an important role; manifestations of such specific effects can be
observed, for example, when varying the added electrolyte.
The type and charge of the added electrolytes can strongly influence
the rheological behavior of mineral slurries. To illustrate this point, several
results are reported in Fig. 2 for aqueous quartz slurries to which one of
the following electrolytes was added (NaCl, CaCl, or Na,S04). The data
show that CaCl, is much more effective than NaCl in reducing the fluidity
of the slurry; the weak opposite effect of Na,S04 contrasts sharply with
that of CaCl,, although both electrolytes are 2:1 salts (containing a divalent
ion).
These observations are partly predicted from electrostatic theory of

colloids, i.e., the influence of electrolytes on the coagulation of colloidal
suspensions is found to be proportional to 2, z being the valence of the
ions of charge opposite to that of the colloid (8). Furthermore, since Ca2+
is known to adsorb on silica particles (9, it is highly probable that Ca2+
ions act as "ionic bridges" between two negatively charged particles,
promoting further particle aggregation and loss of fluidity. On the other
hand, the fluidification effect observed with Na,S04 is not readily
A C 1 SP-148 74 = 0662747 0517844 T7b
interpreted from simple colloid stability principles, although it seems

plausible that the SO:- ions can contribute additional repulsive forces
between the negatively charged quartz particles.
The foregoing observations suggest that localized electrostatic effects

will depend on the nature of both, the electrolyte and mineral surface. The
influence of a given electrolyte may, however, be generalized to many types
of mineral particle slurries. This is readily seen in Fig. 3 showing rheology
data (apparent viscosity) of various mineral slurries with an without added
CaC1,. A marked loss of fluidity is induced in all of these systems by the
electrolyte, and the effect is most certainly to be encountered in cemen-
titious systems as well.
Influence of p H -- The addition of acids or alkalies to vary the p H

will contribute a change in the ionic strength and thus influence the
rheology of slurries as noted above. However, with some minerals,
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
particularly those containing oxides and hydroxides, the rheological
properties are strongly dependent on pH; in these cases, the surface
dissociation or adsorption of H+ or OH-alters the surface charge and zeta
potential of the particles and their related properties.
As an example, we illustrate in Fig. 4 the apparent viscosity of a

Tio, slurry as function of pH. The viscosity peaks at a p H value of = 5.3
which is close to the pH at zero charge (pzc) occurring at 5.5 for Tio,. O n
either side of the pzc, the surface exhibits a net surface charge (positive at
low pH, negative at high pH); the interpaiticle repulsive forces induced by
the surface charges increase the particle dispersion and the fluidity of the
slurries.
The fluidification effect of superplasticizers - Much as the

"stiffening" effect of electrolytes, the fluidifying influence of superplasti-
cizers on mineral slurries also appears general as shown by the data in
Fig. 5. Although the magnitude of the effects depends on the physico-
chemical properties of the mineral, the addition of a superplasticizer greatly
increases the fluidity of the slurries in all cases.
The amplitude of the electrolyte stiffening, or superplasticizer

fluidification, effects on the slurries depends of course on the concentration
of these additives. This is illustrated in Fig. 6 which shows the change in
slurry viscosity, upon addition of salts, for a concentrated suspension of
quartz containing varying amounts of superplasticizer (the superplasticizer
concentrations in these experiments are similar to those commonly used to
fluidify cement pastes). As noted above, the presence of the superplasticizer
counteracts the effect of the electrolyte; for a fixed concentration of
superplasticizer, the higher the salt concentration, the higher the viscosity.
A C 1 SP-148 94 0662949 0539845 902
70 Jolicoeur et al
This behavior is similar to that observed for cement systems containing

various amounts of alkalies (10).
The superplasticizer fluidification was also found to depend

markedly on the molecular parameters of the polymers, in particular, their
size or molecular weight. Low molecular weight polymers are much less
effective than those of intermediate or high molecular weights for
fluidification of Tio, suspensions as shown in Fig. 7a. A commercial
superplasticizer, containing a broad range of molecular weights, shows a
fluidifying effect between that of the low and high molecular weight
fractions extracted from the same commercial product; a very similar
behavior is observed in cement pastes as also illustrated from mini-slump
data in Fig. 7b.
The relationship between the molecular weight of the superplasti-

cizer and its fluidification power may originate in electrostatic effects i.e.,
the larger molecules conveying a higher surface charge to the particles,
increasing interparticle repulsive forces. However, it could also originate
in shorter range repulsive forces due to the steric hinderance between
polymer molecules adsorbed on neighbouring particles, i.e., the adsorbed
molecules preventing particle-particle contact (11). In the systems studied
here, the steric repulsion effect is not readily demonstrated in a direct
fashion. However, the above observations and other results discussed below
strongly suggest that steric repulsive forces are operative in superplasticized
cement pastes, contributing significantly to fluidification.
Superplasticizer adsorption on mineral particles
Much of the preceding discussion of superplasticizer effects on the

rheology of aqueous mineral slurries implies that the superplasticizer
molecules adsorb onto the various mineral surfaces. To document this
assumption in a quantitative way, the adsorption of NaPNS molecules on
several different minerals has been extensively investigated. Some of the
results are illustrated in Fig. 8 (for the purpose of comparison, the data
were normalized relative to the nitrogen surface area of the solids).
The quantity of superplasticizer adsorbed per square meter of surface

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
area is found to be highly dependent on the nature of the solid. It ranges

from nearly O for quartz and silica up to 1.2 pmol/mz for alumina and
titanium dioxide. The surface charge (zeta potential) of the particles
appears t o be a dominant factor for NaPNS adsorption: solids with a zero
or positive value of zeta potential (Alzo3,Al(OH), and Ti03 exhibit a high
affinity towards NaPNS, while those with more negative values (silica,
quartz) show very low NaPNS adsorption.
A C 1 SP-148 94 = 0662747 0539846 849 =
Experiments carried out with NaPNS fractions of different

molecular weight further showed that, for a given solid, the NaPNS
adsorption is dependent on both the molecular weight of the polymer and
the ionic composition of the suspending solution (Fig. 9). The affinity of
the Tio, surface is found similar for the intermediate and high molecular
weight polymers, but is much lower for polymers of MW < 3 kD
(1 kilodalton = io3 g/mol); this has also been observed for other solids.
These adsorption data parallel the fluidity results which were also
maximum for high molecular weight polymers. The adsorption of high
molecular weight NaPNS on Tio, is strongly enhanced by the presence of
Ca" ions as also shown in Fig. 9; the influence of Ca" on the adsorption
of low molecular weight PNS is comparatively very weak. These
adsorption results on Tio, are consistent with other published results (12).
Relation to cementitious systems
The various effects discussed above for slurries of "inert" mineral

particles must also be operative in slurries of cement particles, although
their magnitude and importance will evolve as the hydration reactions
proceed. Based upon the observations described above, the non-specific
(electrostatic) effects occurring in a cement paste at very early ages (t = O)
may be pictured as follows.
In contrast to homogeneous particulate minerals examined previous-

ly, the cement particles comprise different mineral phases (as micro-
domains, or grains) which, upon immersion in water will acquire electro-
static charges of different magnitude and/or sign. Hence, in a cement paste,
like-charged particles will repel each other, while neighbouring particles
which exhibit opposite surface charges (average or local) will experience
strong attractive forces. As a net result of the attractive and repulsive
forces (and also due to the low amount of water available), the particles can
aggregate to form stable floc structures; the latter can be further stabilized
by soluble electrolytes (e.g., alkalies in cement which dissolve very rapidly),
through the "screening" and "ion-bridging'' mechanisms discussed above.
By analogy with observations in slurries of non-reactive minerals,

the fluidification effect observed upon addition of a superplasticizer in
cement pastes may thus initially originate from: adsorption of the
superplasticizer molecules onto the cement particles; screening of attractive
forces between oppositely charged particles (deflocculation) and, inter-
particle repulsive forces due to the high negative charge conveyed to the
particles by the adsorbed superplasticizer (dispersion); and possibly the
steric repulsive forces inferred above.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 Sp-148 94 0662949 0519847 7 8 5
72 Jolicoeur et al
There is ample evidence in the literature that the superplasticizer

adsorbs onto the cement particles, (i, 13-18) and studies on pure phases (i,
19-20) have shown large variations in the affinities of the different phases
for the negatively charged polymer molecules. Since the superplasticizer
adsorption experiments cannot be done at t =O, the data describe the
adsorption of superplasticizer on the surface-hydrated particles. In this
respect, it is important to recall that the fluidification effect in cement
pastes is observed wether the superplasticizer is introduced in the mixing
water, or at later times after the mixing (i, 3), i.e., the action of the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
superplasticizer does not appear strongly dependent on the extent of surface

hydration of the cement particles. In fact, the quantity of superplasticizer
required to achieve a given fluidity is lower when the superplasticizer is
added after all the other components have been mixed together. These
observations provide clues to the importance of chemical-type interactions
between cement particles and superplasticizer molecules. Further under-
standing of the latter may be gained from studies of the influence of
superplasticizers on the early cement hydration rate.
INFLUENCE OF SUPERPLASTICIZER O N CEMENT HYDRATION
It is broadly acknowledged that the presence of a superplasticizer in

a fresh concrete delays the cement hydration and frequently leads to a
retardation of the paste setting (1). The effect appears quite general,
although it is dependent on the nature of the superplasticizer and markedly
influenced by the cement composition; usually, retardation is inversely
related to C,A content (3, 21). This observation raises several important
questions in relation to the mode of action of superplasticizer, typically:
Which hydration reactions are affected primarily? Are the hydration
reactions paths and products altered in the presence of superplasticizers?
What are the key molecular parameters of the superplasticizer polymers
which determine their effect on hydration reactions?
In this section, some of these issues are addressed from new and
published results on the influence of superplasticizer on the hydration
processes in various cementitious systems.
Influence of suDemlasticizers on initial cement hvdration
To investigate the nature and magnitude of superplasticizer effects

on the cement hydration reactions, the heat liberated as function of time
was measured when cement particles are immersed in water, or in aqueous
solutions containing admixtures of interest (7). Some of the results are
illustrated in Fig. 10, showing selected thermograms obtained when a
sample of cement powder is immersed in water and in NaPNS solutions.
A C 1 SP-L4ô 74 0662947 0517ô48 b L L
The thermograms exhibit a sharp temperature rise upon cement-water

contact, followed by a weak near-linear increase. The rapid rise is due to
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
a combination of phenomena namely, wetting, dissolution of electrolytes,

adsorption of solution species and surface hydration reactions. The weak
continuous temperature increase thereafter mainly reflects the hydration
reactions occurring during the lag phase (in the present experiments, the
W/C ratio is high, i.e., = 17, hence, the formation of an initial surface
hydrate layer can be very rapid).
The presence of NaPNS leads to a decrease in the heat evolved

(curves b, c) and the magnitude of the effect is found qualitatively related
to the NaPNS concentration, although there appears to be a saturation
condition. A close examination of the thermograms shows that, in the
presence of excess NaPNS, the initial hydration rate is most strongly
affected (first few minutes), i.e., the steep temperature rise curve is strongly
altered both in shape and magnitude. Similar experiments with the pure
mineral phases of portland cement show that it is mainly the C,A and
C,AF hydration rate that is decreased in the presence of NaPNS (7).
Ramachandran (19) also observed a lower heat flux for the hydration of
C,A in presence of a melamine-based superplasticizer compared to values
obtained in pure water. However, in cement pastes, Vernet (22) (C,A) and
Uchikawa (18) only observed minor changes of the heat of hydration in the
presence of PNS superplasticizer.
The molecular features of the superplasticizer were also found to

have a marked influence on the cement hydration kinetics. The hydration
thermograms obtained in the presence of NaPNS of different molecular
weights reveal unexpected trends (curves d, e). Indeed, the low molecular
weight NaPNS appears more effective in reducing surface hydration than
the high molecular weight species. With polymers of molecular weight
lower than 3 kD, the initial heat is decreased by a factor of two, while with
polymers of molecular weight higher than 100 kD, the initial heat of
reaction is of magnitude comparable to that measured in pure water. This
contrasts sharply with other observations noted above, namely: in
adsorption on non-reactive particles, and in the fluidification of mineral or
cement pastes, the high molecular weight species are most effective. It
should also be noted that numerous other additives (e.g., sulfates, some
retarding admixtures) can reduce the initial heat of hydration without being
effective fluidifiers (data not shown).
Implication for the mode of action of superplasticizer in early hydration
With respect to the early cement hydration reaction rates, the

observations reported above strongly suggest that the larger polymers,
although strongly adsorbed on the cement particles, do not oppose an
A C 1 SP-148 94 0662747 0519849 5 5 8 =
74 Jolicoeur et al
effective barrier t o water (and ions) transport at the mineral/solution

interface. Instead, the hydration kinetics appear controlled by specific
surface site-polymer reactions, i.e., as opposed to non-specific (electrostatic)
polymer-particle interaction. Such reactions would be analogous to those
involving the sulfate, and it is highly probable that PNS (bearing SO,
groups) and SO, species compete for the same reactive sites, particularly on
C,A (16). In such surface reactions, a larger number of small molecules can
yield a more complete coverage of individual surface sites, and a more
compact surface packing, than a small number of large polymers. Hence,
for a given amount of PNS, low molecular weight polymers can be more
effective at blocking hydration reactions at specific sites.
The information derived from calorimetric studies of initial cement

hydration are consistent with recent results (23) showing that the PNS
molecules not only act to disperse cement particles, but are involved in the
hydration reactions, leading to changes in the morphology of the hydrates.
Specifically, in the presence of PNS, ettringite crystals exhibit a globular
morphology rather than their usual needle shape crystal form. This
alteration of morphology should have a marked influence on the rheologi-
cal properties of cementitious systems at very early ages.
The results examined above, both on cementitious systems and inert

mineral pastes, force the conclusion that superplasticizer molecules act both
as dispersants (physical effect) and surface binding reagents for the more
reactive sites of cement particles (chemical effect). As seen from the
changes in the initial heat of hydration, the chemical effect is important and
virtually instantaneous. Since, as pointed out before, the amount of
superplasticizer required to achieve a given concrete fluidity is lower when
the admixture is introduced as the last component, the early cement-
superplasticizer surface reactions apparently result in a high initial
"consumption" of the superplasticizer molecules. Finally, recalling that
other chemical additives, e.g., sodium gluconate, can also reduce the initial
heat of hydration, it can be safely concluded that the superplasticizer-surface
reaction invoked will primarily influence the' paste fluidity as function of
time (i.e., slump retention).
Influence of suuerulasticizers on the setting of cement uastes
As noted earlier, it is frequently observed that superplasticizers delay

the setting of cement pastes, sometimes by as much as several hours. While
this retardation effect must originate in the cement-superplasticizer
interaction discussed above, it is interesting to determine the duration of
this effect, and to what extent the presence of a superplasticizer can
influence the hydration reactions during and after the acceleration phase.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 M Ob62947 0517850 2 ï T M
To investigate the influence of superplasticizers on the kinetics of

cement hydration during the setting period, the heat evolved in cement
pastes was measured as function of time (0-24 hrs) in the presence and
absence of superplasticizers (21). Selected results are reproduced in Fig. il.
Upon the first addition of a NaPNS superplasticizer, the shape of the
thermogram changes drastically, and the heat flux peak is shifted substan-
tially to shorter times (not shown); these phenomena are most certainly
due to an improved water-cement contact upon deflocculation and a more
homogeneous dispersion of the cement particles. Upon further addition of
NaPNS, a pronounced retardation is observed for both, the beginning of
the accelerated phase, and the maximum of the thermal peaks (21).
The superplasticizer effects depicted in Fig. i1 were also found with

many other cementitious systems, the magnitude of the changes depending
on the type of cement and the nature of the superplasticizers. Beyond the
effect of the first addition of superplasticizer, the results obtained to date
exhibit several common features: first, the shapes of the hydration curves
are not drastically altered as the concentration of superplasticizer is
increased; second, the start of the acceleration phase (setting time) and the
time elapsed to the maximum heat flux are shifted to longer times in a
parallel way (4,21). These observations, combined with the heat of
hydration data near t =O, are consistent with the notion that the superplas-
ticizer molecules bind irreversibly (chemically) to the surface of the
hydrating particles. The molecules are incorporated progressively into
some hydration products, and when most of the superplasticizer is depleted,
further hydration can proceed virtually unperturbed. Calorimetric data
obtained with NaPNS of different molecular weights have shown that the
low molecular weight polymers induce more pronounced retardation effects
than the high molecular weight species (4, i.e., in agreement with the
molecular weight dependence on initial hydration kinetics (in different
conditions, Kondo et al. (24) had reported a greater retardation with high
molecular weight polymers).
TOWARDS A DESCRIPTION OF SUPERPLASTICIZER ACTION
In the foregoing overview, a variety of phenomena occurring in

concentrated slurries of cement and of various types of mineral particles
were examined. Data on superplasticizer adsorption, on paste fluidification
and on cement hydration rates provide a number of clues from which a
more detailed description of the "superplasticizer effect" may be attempted.
Based on the latter, and keeping in mind that the unique property of
superplasticizers is their ability to fluidify cement pastes at very low W/C;
the following "physical" and "chemical" roles may be assigned.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = Obb2949 0519851 Lob
76 Jolicoeur et al
Interfacial effects - Superplasticizer molecules usually do not

exhibit high surface activity (surfactancy), but they adsorb onto most
hydrated mineral surfaces including cement particles by virtue of non
specific van der Waals forces and electrostatic forces due to particle surface
charges (direct binding to positively charged surfaces, or binding assisted by
positive ions (e.g., Ca") onto negatively charged surfaces.
Surface charging and related effects -- The adsorption of superplas-

ticizer molecules will convey a net negative surface charge on all particles,
irrespective of their initial surface charge. This induces interparticle
repulsive forces which promote the dispersion of the particles and the
breaking-up of any floc structure, thereby improving fluidity. Because
electrolytes markedly affect the electrostatic phenomena at interfaces, the
fluidification effect depends on the ionic composition of the solution phase;
everything else being equal, the quantity of superplasticizer required will
increase as the concentration of dissolved electrolytes increases. (The
solution phase of fresh cement paste contains between 0.1 and 1.0mol/L
of electrolytes.)
Binding to specific surface sites -- Superplasticizer molecules can bind

onto highly reactive surface sites (e.g., of C,A or C,AF) and reduce the rate
of subsequent hydration reactions. The early hydration reaction behavior
(i.e., t - O ) shows that the superplasticizer effect is not merely that of a
physical barrier to water and ions diffusion at the interface; it seems best
described as a partial blocking of specific surface sites which play a key role
in hydration reactions. As noted earlier (í!2), this chemical effect leads to
the inclusion of superplasticizer molecules in new hydrate products.
Steric hinderance -- The various effects described above, i.e.,

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
adsorption, dispersion, retardation of cement hydration, are not exclusive

features of superplasticizers. Other chemical compounds, for example,
sodium gluconate (a dispersant and retarding admixture), exhibit similar
features, but without the important fluidification effect at very low W/C
observed with superplasticizers. The latter requires very short-range
repulsive forces which will prevent particles from "sticking" together
although they are closely packed (in order to achieve low porosity, high
density, concretes). These additional forces are most likely provided by
steric repulsion (ii), i.e., polymer molecules adsorbed onto neighbouring
particles oppose their adhesion and aggregation by steric hinderance. This
short-range repulsive effect was invoked in a preceding section as a possible
contribution to the fluidification effect of superplasticizers in slurries of
mineral pastes. Again, the steric repulsion effect cannot be demonstrated
unequivocally, but there is sufficient basis for inferring this phenomena in
explaining the unique functional properties of superplasticizers in cement
pastes and concretes.
A C 1 SP-148 9 4 W Obb2949 0 5 3 9 8 5 2 042 W
Of the various phenomena discussed in this paper, the extent of

adsorption and the fluidification effect were both found to increase
markedly with the molecular weight of a NaPNS superplasticizer. The
surface charge (as measured by the zeta potential in dilute slurries) appeared
relatively insensitive to the molecular weight of the superplasticizer.
Finally, the importance of specific site surface binding and early hydration
control was found inversely related to the superplasticizer molecular
weight. The different trends observed as function of molecular weight are
consistent with general principles and observations in colloid chemistry:
Polymer adsorption at interfaces becomes increasingly irreversible

(entropic stabilization) with increasing molecular weight.
The extent of surface charging depends on the quantity of polymer
adsorbed and on the intrinsic charges of the polymers; hence, surface
charging can vary significantly with polymer molecular weight, and it
can only be predicted if the charge (number of ionic groups and degree
of ionization) is known as function of molecular weight.
The polymer steric hinderance effect, which provides short-range inter-
particle repulsive forces is also known to increase with increasing
molecular weight of the polymer (11).
Finally, the variation in the surface chemical effect with molecular
weight is readily understood from statistical considerations; a large
number of small molecules can be more effective than a small number
of large molecules (although the surface binding is weaker for the small
molecules).
CONCLUDING REMARKS
The foregoing overview on the functional properties of superplasti-

cizers in aqueous slurries of non-reactive minerals and cement pastes
provides a basis for discriminating those effects which are non-specific (thus
common to any slurry), from those who are characteristic of the cement-
superplasticizer systems. The data examined yield direct confirmation of
the following non-specific (physical) effects:
i) interfacial adsorption through electrostatic and van der Waals forces;

2) surface charging and electrostatic screening of interparticle attractive
forces and induction of repulsive forces.
The data further showed direct evidence of a specific (chemical)

effect:
3) surface binding to reactive sites, inhibiting early hydration reactions.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 94 = Obb2949 0539853 T 8 9
78 Jolicoeur et al
Finally, the combined information available from mineral and

cement slurries enables one to infer another nonspecific effect:
4) steric hinderance leading to short-range interparticle repulsion.
In order to achieve the dispersion of cement particles and fluidifi-

cation of cement pastes, a superplasticizer must perform each of these four
functions. The effectiveness of a superplasticizer in these functions, for a
given cementitious system, depends on its chemical composition and
molecular parameters. Amongst the latter, the molecular weight was seen
to play an important role.
In particular, it was found that increasing the molecular weight of

the (PNS) superplasticizer molecules enhances the adsorption of these
molecules on mineral particles (an effect which may be further promoted
by Ca2+ions) and increases their fluidification performance. On the other
hand, it appeared that the early surface hydration reactions are best
controlled (inhibited) with low molecular weight PNS. Such effects can be
qualitatively understood in terms of general polymer and colloid chemistry
concepts, taking into account the specific reactivity of cement particles.
This should help to develop more effective uses of superplasticizers in high
performance concretes and also suggest more focused investigations in the
development of new superplasticizers, for example, with regards to the
optimal distribution of polymer molecular weights.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the financial support of the

National Sciences and Engineering Research Council, the Network of
Centers of Excellence on High Performance Concrete and Handy
Chemicals Ltd. The contributions of Pierre Gagné towards some of the
data reported here and numerous discussions with Prof. P.-C. Aïtcin and
many other academic and industrial colleagues are also gratefully acknowl-
edged.
REFERENCES
1 Ramachandran, V.S. and Malhotra, V.M., "Superplaticizers"in Concrete

Admixtures Handbook, V.S. Ramachandran ed., Noyes Publ., 1984, p.
21 1-268.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
2 Rkom, M.R. and Mailvaganam, N.P., "Chemical Admixtures for

Concrete", E and F.N. Spon Ltd., 2nd ed., 1986.
3 Dodson, V., "Concrete Admixtures", Van Nostrand Reinhold, 1990.
A C 1 SP-148 94 0662949 0519854 715
Piotte, M., Ph.D. thesis, Université de Sherbrooke, 1993.
Nkinamubanzi, P.-C., Ph.D. thesis, Université de Sherbrooke, 1994.
Kantro, D.L., "Influence of Water Reducing Admixtures on Properties

of Cement Pastes -A Miniature Slump Test", Cem. Concr. Aggregates,
-
2, 1980, 95-102.
Gagné, P., M.Sc. thesis, Université de Sherbrooke, 1993.
Shaw, D.J., "Introduction to Colloid anti Surface Chemistry", 3rd ed.,

Butterworths, 1980.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Zerrouk, R., Foissy, A., Mercier, R., Chevallier, Y and Morawski, J.C.,
"Study of Ca*+-InducedSilica Coagulation by Small Angle Scattering",
J. Colloid Interface Sci., 139, 190, 20-29.
10 Ref. 3, p. 53.
11 Napper, D.H., "Polymeric Stabilization of Colloidal Dispersions",

Academic Press, 1983.
12 Pierre, A., Carquille, C., Lamarche, J.M.., Foissy, A. and Mercier, R.,
"Adsorption d'un polycondensat d'acide naphtalène sulfonique (PNS) et
de formaldéhyde sur le dioxyde de titane, Cem. Concr. Res, 18, 1988,
18-28.
13 Blank, B., Rossington, D.R. and Weinland, L.A., "Adsorption of

Admixtures on Portland Cement", J. Am. Ceram. Soc., 46, 1963, 395-
399.
14 Burk Jr., A.A., Gaidis, J.M. and Rosenberg, A.M., Adsorption of

Naphthalene-Based Superplasticizerson Different Cements, presented at
the 2nd Int. Conf. on Superplasticizers in Concrete, Ottawa, 1981.
15 Collepardi, M., Corradi, M., "Influence of Polymerization of Sulfonated

Naphthalene Condensate and its Interaction with Cement", in Proceed-
ings of the 2nd Int. Conf. on Superplasticizers in Concrete, V.M.
Malhotra ed., AC1 SP-68, 1981, 485-498.
16 Andersen, P.J., Kumar, A., Roy, D.M. and Wolfe-Confer, D., "The
Effect of Calcium Sulphate Concentration on the Adsorption of a
Superplasticizer on a Cement: Methods, Zeta Potential and Adsorption
Studies" Cem. Concr. Res., 0,1986, 255-259.
A C 1 SP-148 94 0 6 6 2 9 4 9 0539855 8 5 1 W
80 Jolicoeur et al
17 Andersen, P.J. and Roy, D.M., "The Effect of Superplasticizer Molecu-

lar Weight on its Adsorption on, and Dispersion of, Cement", Cem.
Concr. Res., 18, 1988, 980-986.
18 Uchikawa, H., Hanehara, S., Shirasaka, T. and Sawaki, D., "Effect of

Admixture on Hydration of Cement, Adsorptive Behavior of Admix-
ture and Fluidity and Setting of Fresh Cement Paste", Cem, Concr.
Res., 22, 1992, 1115-29.
19 Rarnachandran, V.S., "Adsorption and Hydration Behavior of Tri-

calcium Aluminate-Water and Tricalcium Aluminate-Gypsum-Water
Systems in the Presence of Superplasticizers", AC1 Journal, May-June
1983, 235-241.
20 Fukaya, Y. and Kato, K., "Adsorption of Superplasticizers on CSHO

and Ettringite", 8th Int. Congr. Chem. Cem., Rio de Janeiro, Vol. 2,
1986, 142-147.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
21 Simard, M.-A., Nkinamubanzi, P.-C., Jolicoeur, C., Perraton, D. and
Aïtcin, P.-C., "Calorimetry, Rheology and Compressive Strength of
Superplasticized Cement Pastes", Cem. Concr. Res., 2,1993, 939-950.
22 Vernet, C., and Noworita, G., "Interaction des adjuvants avec l'hydra-
tation du C,A: Point de vue chimique et rhéologique", Technodes S.A.,
1992.
23 Sarkar, S.L. and Aimin, X., "Preliminary Study of Very Early

Hydration of Superplasticized C,A + Gypsum by Environmental
SEM", Cem. Concr. Res., 2,1992, 605-608.
24 Kondo, R., Daimon, M. and Sakai, E., "Interaction Between Cement

and Organic Polyelectrolytes" 11 Cemento, 3/1978, 225-229.
A C 1 SP-148 9 4 W Obb2947 0537856 798
TABLE 1 - PHYSICAL CHARACTERISTICS OF MINERALS
Mineral Chemical Mean Surface

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
formula diameter area

dum) m2/g
Quartz 15 1.6
Silica* 20 2.0
Alumina 15 0.9
Hydrated alumina 20 1.0
Titanum dioxide 26 3.5
* Refers to amorphous silica

** Slurries for rheological experiments
A C 1 SP-148 74 = 0bb2747 0517857 624 =
82 Jolicoeur et al
Fig. 1-Mini-slump of amorphous silica powder suspensions prepared in a) pure water;

b) NaCI solution; c) NaCI solution containing a small quantity of superplasticizer
Control NaC1 CaC1, N+SO,
Fig. 2-Apparent viscosity of aqueous slurries of quartz powder containing 0.1 percent of
different electrolytes (the control is without added electrolyte)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-34ô 9 4 I0 6 6 2 9 4 9 0 5 3 9 8 5 8 5 b 0
60
o
O
C 50
*e
õ 40
.Z
* 30
Y
= 20
8
m
g 10
4 o
Quartz Silica Alzo, AI(OH), Tio,
Fig. 3-Apparent viscosity of various mineral slurries in water: or in CaCI, O1 percent solution:
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O
u
h
-o
o
O
rc
.iI
v>
40 -
O
O O
O 0
O
.-E>
n
B
Y 20 -0
3
9
2 O0 o
4 o I l I l n
Fig. A n f l u e n t e of pH on apparent viscosity of a Tio, slurry in water
A C 1 SP-348 94 0662949 0539859 4 T 7
84 Jolicoeur et al
-
Quartz Silica AZO3 A(OH), Tio,
Fig. U p p a r e n t viscosiîy of mineral slurries prepared in water: or in O1 percent NaPNS

aqueous solution: B
NaC1 concentration (W’t %)

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. ó-Influence of NoCI concentration on the apparent viscosity of quam slurries prepared in pure
water: O, in 0.1 percent NaPNS: 0,or in 05 percent NaPNS solutions: A
A C 1 SP-148 94 W 0662949 0519860 119 H
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
MW<3kD 3-1OOkD MW>lWkD Miaum

Molar mass fraction
Fig. ia-Effect of NaPNS molecular weight on the apparent viscosity of Tio, slurries
(1 kilodalton = i@ g/mol)
loo I
O 30 60 90
Time (min)
Fig. 7b-Mini-slump area, as function of time, of cement pastes containing 0.3 percent of NaPNS.
O: MW > 100 kD, a: MW < 3 kD
A C 1 SP-148 94 Ob62949 0519861 0 5 5
86 Jolicoeur et al
O io 20 30 40
Equilibrium conc. (pmol/l)
Fig. 8-Adsorption isotherms of NaPNS, MW > 100 kD, on Tio,: v, AI,$: AIIOH),: O, silica:
A and quartz: 4 (adsorption expressed as micromoles of polymer repeat unit per square meter
of N, surface area)
8
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
6
bc)
$ 4
E
a
2
o
O 3 6 9 12 15
Equiiibrium conc. (pmol/l)
Fig. 9-Adsorption isotherms of NaPNS on Ti$ suspended in water or in 0.1 percent CaCI2.
MW > 100 kD in water: 0;in 0.1 percent CaC12: m;
3kD e MW e 100 kD; in water: O;
MW < 3 kD: in water: A; in 0.1 percent CaC12: A
A C 1 SP-L4ô 94 Ob62949 0.539862 T91
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
- - - - - - --
I l I
O 10 20 30
Time (min)
Fig. 10-Heat evolved when a cement powder sample is immersed in water or in NaPNS solutions.
a) water; b) 0.08 percent NaPNS; c) 0.4 percent NaPNS; d): 0.4 percent of a low molecular weight
fraction of NaPNS MW c 3 kD; e) 0.4 percent of a high molecular weight fraction of NaPNS MW
> 100 kD
88
A C 1 SP-/qô 9 4
Jolicoeur et al
= 0662749 05L98b3 728
a
b n
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 4 8 12 16 20
Time (h)
Fig. 11-Hydration thermograms of cement pastes (W/C = 0.35) in water (a) or in presence
of NaPNS 05 wt percent (b); 1 5 w i percent (c)
A C 1 SP-148 9 4 H 0bb2949 0519864 8 6 4
SP 148-5
Influence of Acrylic Polymers on

the Rheology of Mortars
by S. Maximilien, J. Ambroise, and J. Péra
: The influence of an acrylic polymer on the rheology of mortars was

investigated using a mixer in which the torque on the impeller shaft was
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
continuously measured. The polymer was added to specific mortars either alone
or in combination with aqueous solutions of sulfonated naphthalene
formaldehyde condensate, or sulfonated melamine formaldehyde condensate.
Two cements were ground from two different clinkers to specific surfaces of
270 and 400 m2/kg, respectively.
The flow properties of these fresh mortars closely approximate the

Bingham model, whatever the time after initial mixing may be.
When used alone, the polymer decreases the plastic viscosity of the
mortars. When used in combination with sulfonated melamine or naphthalene
formaldehyde condensates, it decreases the yield value.
Kewords: Acrylic polymers; clinker; flowability; mortars fmoteriol); rheoloaical properties;

viscosity
A C 1 SP-II48 7Y = Obb%7i.17 0 5 1 7 8 6 5 7T0 =
90 Maximilien, Ambroise, and Péra
S. Maximilien is a Civil Engineer. She is a Ph. D. student involved in the

development of new chemical admixtures for concrete.
J. Ambroise, Ph. D. in Physical Science, is "Maître de Conférences" at the
Institut National des Sciences Appliquées de Lyon, France. He is engaged in the
development of technical concretes : self-levelling concretes, porous concretes,
fibre-reinforced concretes.
J. Péra, AC1 and MRS member, is Professor of Civil Engineering and Head of
the Inorganic Materials Laboratory at the Institut National des Sciences
Appliquées de Lyon.
INTRODUCTION
The great potential of superplasticizers as high range water-reducing
agents has brought some disadvantages to job sites where concrete is placed at
flowing consistency. The presence of a superplasticizer does not prevent the
loss of workability with time. It can be ameliorated by delaying the addition :
one part of the superplasticizer is added with the mixing water, the other part is
introduced just before casting the concrete. In this way, the workability can be
maintained for 135 minutes (i, 2). But this technical solution is not always well
managed. Problems of bleeding can appear and durability requirements are not
always met. The difficulties which might arise from bleeding and segregation in
concrete mixtures with high slump can be avoided by using combinations of
stabilizing agents and sulfonated melamine or naphthalene formaldehyde
condensates.
The aim of the present study is to investigate the behaviour of new
admixtures. In these admixtures, an acrylic polymer is mixed with sulfonated
melamine or sulfonated naphthalene formaldehyde condensates.
A specific mortar was chosen such that its rheological behaviour
correlates to that of concrete (3). The cement content in the test mortar was
reduced comparatively to the normal mortar in order to get a cementísand ratio
close to that of the cementíaggregate ratio in concrete.
Ail the measurements on mortars were carried out using a mixer in which
the torque on the impeller shaft was continuously recorded. The d u e n c e of the
polymer addition was studied on two cements which were round from two
different clinkers to specific surfaces of 270 and 400 m2gk g , respectively.
Measurements of bleeding were also done.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662749 O539866 637
MATERIALS
The Bogue potential compositions and specific surfaces of the cements

are shown in Table 1. Cements Ci and C2 came from clinker K1, C3 and C4
from K2.
The sand was a French Norm sand (2 mm maximum). The admixtures
were sulfonated melamine (MF) and naphthalene (NF) formaldehyde
condensates and an acrylic polymer (Pi), with about 50 % solids. The molecular
weight of this acrylic solution was about 13 000. This polymer is patented.
MIXTURE PROPORTIONS
In order to approximate the behaviour of concrete, specific mortars were

designed. Normal mortar cast according to French standards NFP 15-403
contains a large amount of cement in comparison with the actual cement content
of concrete. The mixture proportions of mortars are based on one prepacked
bag of sand Containing 1350 g Normsand. The normal mortar cement content is
450 g (cementhand = 0.33 by weight). Our specific mortar only contained 350
g cement cementfsand = 35011350 = 0.26). In an average concrete containing
i
350 kg/m cement, 750 kg/m3 sand, and 1100 kglm3 coarse aggregate, the
cement/(sand + aggregate) ratio is 350/1850= 0.19.
In the specific mortar the water content was 245 g, giving a
waterkement ratio of 0.70. This quantity of water allowed the measurement of
reliable rheological values, even with a control mortar without any admixture
mixed for 90 minutes.
The individual admixture contents were as follows :
Melamine (MF) : 0.2 %
Naphthalene (NF) : 0.2 %
Polymer (Pi) : 0.1 %
The contents are expressed as dry solidsicement.
RHEOLOGICAL MEASUREMENTS
ADDaratus
A special mortar mixer was used, which has a blade shaped like an open
shield which rotates in planetary motion at different speeds. The bowl containing
the mortar is supported on a suitable arm that can be raised or lowered. The
blade is immersed in the mortar and is rotated about its own axis. A flow curve
was obtained from measurements of the torque required to rotate the blade.
Previous work carried out by Banfll (4, 5), on the rheology of fresh
mortar has shown it to undergo structural breakdown, and that the equilibrium
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 m Obb2949 0519867 573 m
flow conforms to the Bingham model. The relation between torque and speed is
therefore :
T=g+hN
where T is the torque at a speed N, g the intercept on the torque axis, and h the
slope of the line. The quantity g is a measure of the yield value of the mortar
and the quantity h is a measure of its plastic viscosity, both of which are
fundamental properties of the material (6).
In the case of the slump test, the measurement is made on a stationary
cone of concrete and it is reasonable to suppose that the rate of shear associated
with the test is zero, or, in other words, that the slump test assesses the yield
value of the concrete and takes no account at all of the plastic viscosity.
In our study, the flow curve in the form of a graph of torque T against
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
speed N was constructed from values of torque obtained at each speed : in cases
where structural breakdown occured during the test, these were the final torques
reached at the end of each period at constant speed.
It is clear that the actual numerical values of g and h depend on
characteristics of the apparatus, specifically on the design and dimensions of the
impeller and bowl. Tattersall (6) shows that g and h may be converted to
fundamental units if desired. For practical use, this is not necessary, and the
effort required to carry out the calibration would not be justified.
ocedura
The procedure for preparing the mortars before testing was as follows :
the cement was placed in the dry mixing bowl and sand was added over a period
of 30 s. with the blade rotating at 140 rev/min. Then, the water and the
admixture were added and the mortar mixed at 285 rev/min. for 2 minutes. For
the control mortar, only water was added.
Then, the following procedures were carried out :
i) Speed was held constant for 30 s. at each of 50, 100, 150, 200, 250,
300 revímin. ;
ii) Speed was kept equal to zero for 1 minute ;
i¡¡) Speed was held constant for 30 s. at each of 300, 250, 200, 150, 100,
and 50 rev/min.
This procedure gave a total test cycle time of 7 minutes. It was applied
for different times &er the preparation of the mortar : at t = O min., 30 min.? 60
min. and 90 min. Between each test cycle, the mortar was continuously mixed
at 50 rev/min.
The speed regimes and resulting torque-time traces, from which the
individual values or torque at each speed can be read, are shown schematically in
Fig. 1.
A C 1 SP-I148 7 4 0662949 05198b8 4 0 T
An example of flow curve is given in Fig. 2. It is shown that the

downcurve is at lower torques than the upcurve, and the points on the
downcurve approximate to a straight l i e . According to Banfill (3), this suggests
that structural breakdown is complete after the upcurve. The hysteresis loops
for mortar do not denote thixotropic behaviour, but rather result from
irreversible breakdown of structure due to shearing in the test. There is no
evidence that structure rebuilds &er shearing. In the present study, we just
investigated the downcurve, and computed the values of g and h for each time
(O, 30, 60, and 90 minutes) after the preparation of mortars.

Fig. 3 to 18 show the evolution of plastic viscosity h and yield value g
versus mixing time for the four cements from the downcurves of mortars. Table
2 summarizes the main trends of the mixture variables investigated in the present
study as compared with their effects on the rheology of fresh mortar. The
plastic viscosity and yield value of the admixture treated mortars generally
decreases. The acrylic polymer used alone has a slight effect on g.
-h
Effect of admixtures on the plastic viscos'itv
From Fig. 3 to 18, it can be seen that :

i) The plastic viscosity of non-admixture treated mortars increases with
mixing time, especially in the period 60 to 90 minutes, while that of admixture
treated mortars remains quite constant ;
ii) The efficiency of admixtures is higher when mixing time increases ;
iii) The efficiency of admixtures depends upon the characteristics of
cement
- with cement C l , polymer P i used alone has the higher effect :
the combination polymer-melamine gives better results than melamine alone ;the
combination polymer-naphthalene was not as efficient as plain naphthalene ;
- with cement C2, the admixtures leading to higher reductions in
h value are combinations polymer-melamine and polymer-naphthalene ;
- with cements C3 and C4, the more efficient admixture is the
combination polymer-naphthalene.
iv) For a given admixture, the fineness of cement can play a different role
- for clinker K1, the lowest values of plastic viscosity are
obtained either with the coarser cement ( C i ) when polymer P 1 and naphthalene
NF are used alone or with the finer cement (C2) for plain melamine and the
combinations polymer-melamine and polymer-naphthalene ;
- for clinker K2, whatever the admixture may be, the lowest
values of h are obtained with the finer cement (C4).
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0539867 346
tof-
The effect of admixtures on g value can be summarized as follows :
i) g increases gradually with mixing time whatever the admixture may

be ; g is higher for non-admixture treated mortars than for admixture treated
mortars ;
ii)The polymer Pi has the least effect on g values while the lowest g
values are obtained with combinations of polymer-melamine and polymer-
naphthalene ;
iii) The efficiency of admixtures depends upon the type of clinker and the
cement fineness :
- for clinker K i , the lowest g values are obtained with the coarser
cement (Ci);
- for clinker K2, the lowest g values are obtained with the finer
cement (C4).
The bleeding tests are done using the same specific mortar, and the same
W/C (= 0.70) as the rheological measurements.
The results of the bleeding tests are presented in Fig. 19 to 26. The
bleeding rate depends upon the type of clinker, the fineness of cement, and the
type of admixture.
For cements issued from clinker K i , the following trends emerge :
i) Polymer Piused alone decreases bleeding whatever the fineness of
cement may be ;
ii) The use of naphthalene (NF) or melamine (MF ) alone or in

association with polymer P i leads to a drastic increase of bleeding which can be
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
explained by the high watedcement ratio chosen for this study. In this case,
naphthalene and melamine are not used as high-range water reducers.
iii) Bleeding is lower with the finer cement (C2) and in this case, the
effect of admixtures is very slight.
For cements issued from clinker K2, the conclusions are as follows :
i) Polymer P i used alone has a slight effect on bleeding, especially in the
case of the finer cement (C4) ;
ii) Naphthalene (NF) used alone or in association with polymer P i leads
to the lowest bleeding values, whatever the fineness of cement may be ;
A C 1 SP-148 94 0 6 6 2 9 4 9 0 5 L 9 8 7 0 Ob8 w
iii) The use of plain melamine (MF) increases bleeding of mortars

containing the coarser cement (C3) while the combination melamine-polymer
considerably reduces this bleeding. The influence of melamine-based admixtures
on bleeding of mortars containing the finer cement (C4) is not so well-marked as
with cement (C3).
The general conclusions emerging from the bleeding data performed on
mortars cast with a high water content (WK = 0.70) are as follows :
i) The finer is the cement the lower is the bleeding ;
ii) The use of polymer P i alone allows the reduction of bleeding of
mortars cast with the coarser cements (Ci) and (C3) ;
iii) There is no general tendency for combinations naphthalene-polymer
and melamine-polymer.
CONCLUSION
The general conclusions about the rheology of fresh mortars containing
different types of admixtures (acrylic polymer, melamine, and naphthalene) used
alone or in combination, are as follows :
i) For mortars, the use of admixtures leads to a constant plastic viscosity
with time while that of non-admixture treated mortar increases with time. The
efficiency of admixtures is greater with increasing mixing time. The use of
admixtures also decreases the yield value, especially for combinations
naphthalene-polymer and melamine-polymer.
ii) The efficiency of admixtures depends upon both the nature of clinker
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
(C3S content) and the fineness of cement. For example, for clinker Ki, the
mortars cast with the coarser cement(C i) present the best rheological behaviour,
particularly towards the yield value. For clinker K2, the use of the finer
cement(C4) leads to the lowest values of plastic viscosity and yield value.
iii) Acrylic Polymer Pi considerably reduces plastic viscosity but has
slight effect on the yield value when used alone. It acts as a cohesion agent.
This study has to be renewed on cements presenting other C3A contents.
This parameter was kept constant in the present approach.
REFERENCES
1. CHORTNSKY, E.G.F. "Influence of plasticizers on the bleeding
characteristics of flowing concrete". Admixtures for concrete -
Proceedings of the International RILEM Symposium. Barcelona 1990.
Chapman and Hall. pp. 429-432.
96
A C 1 SP-148 7 4 - 0662947 0519871 T T 4
Maximilien, Ambroise, and Péra
2. PAILLERE, A.M., SERRANO, J.J., and GRIMALDI, M. "Influence du

dosage et du mode d'introduction des superplastifiants sur le maintien de
maniabilité optimum des bétons à hautes performances avec et sans
fumées de silice". Admixtures for Concrete. Proceedings of the
International RILEM Symposium. Barcelona 1990. Chapman and
Hall. pp. 63.79.
3. BANFILL, P.F.G. "The rheology of fresh mortar". Magazine of
Concrete Research. 199 1. 43. no 154. pp. 13-21.
4. BANFILL, P.F.G. "Feasibility study of a coaxial cylinders viscometer
for mortar. Cement and concrete Research, 1987. Vol. 17, pp. 329-339.
5. BANFILL, P.F.G. "Acoaxial cylinders viscometer for mortar : design
and experimental validation". Proceedings of the International
Conference on rheology of cement and concrete. Liverpool, 1990. Ed :
& F.N. SPON. London, 1990, pp. 217-226.
6. TATTERSALL, G.H. "Workability and Quality Control of Concrete".

Ed : E & F.N. SPON, London, 1991,262 p.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9i.r 0 6 6 2 9 4 7 0 5 1 9 8 7 2 930
TABLE ’ - BOGUE POTENTIAL COMPOSITION AND SPECIFIC SURFACE OF CEMENTS
Cements
..
c1 c2 C3 c4
B o g e comr>osition
c 3 s (%) 62.0 61.8 57.6 57.5
c 2 s (%) 11.0 11.1 16.7 16.7
C3A (%) 7.4 7.6 7.1 7.1
c4AF (“h) 10.5 10.7 9.2 9.2
Specific surface
rn%g 271 407 275 403
Quantity of gypsum
(%) 6.5 6.5 5.5 5.5
TALE 2 - EFFECTS OF MIXTURE COMPOSITION ON THE RHEOLOGY OF FRESH MORTAR

AFTER 90 MINUTES
Clinker Change (other parameters Effect on mortar

constant)
g h
Cement fineness No change Decrease
Admixture
K1 -polymer Slight decrease Decrease
- naphthalene + polymer Decrease Decrease
- melamine + polymer Decrease Decrease
Cement fineness Slight decrease Decrease
Admixture
K2 -polymer Slight decrease Decrease
- naphthalene + polymer Decrease Decrease
- melamine + polymer Decrease Decrease
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 7Y = 0662949 05L9873 8 7 7 m
a structura4
- 1
Fig. 1-Speed-time and torque-time traces from which flow curves are constructed
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 5) lm rn IOD m m
SPEED (revimin)
Fig. 2-Up and down curves of a same mortar
A C 1 SP-148 94 m Ob62949 051987Y 703 9
. . . . . . . . . . . . . .
o o a e O B Q ¶I
=(m =(m
Fig. M e m e n t C1 Fig. d e m e n t C1
melamine and acrylic polymer naphthalene and acrylic polymer
h6
o o a 9
=(m
Fig. M e m e n t C2 Fig. X e m e n t C2
Fig. 3-lû-Influence of cement and admixture on plastic viscosity h (lû' N.m.min)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 m 0662949 0539875 64T
8. . . . . . . .
h6 _ _ _
4 L 1;
2 ' 2- . . .
O 3 a b O 9 a 9)
=<Mi) =(Mi)
Fig. 7-Cement C3 Fig. P e m e n t C3

O 9 a 9) O 9 a a
!
=(w nww
+rd +Fi +lrp +
Fig. O m e n t C4 Fig. Iû-Cement C4

Fig. 3-1û-Influence of cement and admixture on plastic viscosity h (i@

Nmmin)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 D 0662949 05L9876 5Bb =
Superplasticizers and Chemical Admixtures 1O1
c
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O o 8 e
mm (W
Fig. 11-Cement C1 Fig. 1 2 4 e m e n t C1

m . . .
O
O o 6l e O o m e
(W (W
+& +R +M? +Fi
Fig. 13-Cement C2 Fig. 1 A e m e n t C2

Fig. 11-18-Influence of cement and admixture on yield value g (Nm)
A C 1 SP-14ô 9Y m Obb2949 0517877 412 m
. . . . . . . . . . . . . . . . .
. . . aL . . . . . .
g
m -
al
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 3 œ ¶ O B 4 ¶
= (W = (W
+d +r¶ +1w +
Fig. 15-Cement C3 Fig. Ió-Cement C3

n O B 4 ¶
= (W
+id +Fl +W +
Fig. 17-Cement C4 Fig. 18-Cement C4

Fig. ll-1Pnfluence of cement and admixture on yield value g (N.m)
A C 1 SP-148 94 0 b b 2 9 4 9 0519878 3 5 9 =
% -
[i
5üi
$ . .
Fig. 19-Cement C1 Fig. 2 M e m e n t C1

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
[
?'
d
. . .
4
O E S Q O Z T J
rn(MY)
+wuII +w +m
Fig. 2 1 4 m e n t C2 Fig. 22-Cement C2

melamine and acryiic polymer naphthalene and acrylic polymer
Fig. 19-21lnfluence of cement and admixture on bleeding
A C 1 SP-148 9 4 m O b 6 2 9 4 9 0539879 295 m
:1
. . . . .
Fig. 23-Cement C3 Fig. 2 4 A e m e n t C3

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 21-Cement c4 Fig. 2 L m e n t C4

Fig. 19-2ó-influerice of cement and admuture on bleeding
A C 1 SP-148 9 4 m 0662949 05L9880 T O ? m
Slump Loss and Rheology of

S uperpIasticized Mortar a nd Concrete
with Different Polysaccharide Syrups
by G. Montegozzo and E. Alberti
S v n o p s i s : The rheological properties of fresh mortar and

concrete were studied with 5 types of polysaccharide syrups added to
either naphthalene or melamine sulphonate based
superplasticizers.
The effects of the binary admixtures system (superplasticizer
a n d s y r u p ) on the physic0 mechanical properties of the
cementitious system, based on two types of cement (1 Portland and
1 Pozzolan) were studied.
The properties, which were studied at 25°C and 4OoC, were :
slump, loss of workability, setting time, compressive and flexural
strength. For one type of syrup, added to the superplasicizer,
synergistic effects were observed. In particular, it was demonstrated
that, at 4OoC, it is possible to increase the slump and reduce the loss
of workability using a n admixture based on naphthalene
sulphonate, plus syrup G62 and melamine sulphonate, plus syrup
G62, compared to normal superpiasticizer retarding admixtures.
Keywords: Admixtures; compressive strength; concretes; flexural strength; morion fmateriol!;

polvsaccharides; rheoloaicai orooerties; setting (hardening); slump; suoerdasticizen; workability
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---

105Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-L4ô =
74 Ob62949 05L988L 743
106 Mantegazza and Alberti
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Giovanni Mantegazza- graduated in Chemistry. He is working
as Tecnicai Manager for Ruredil- Milano- Italy where is responsible
for the development of new materials and admixture for concrete.
Elena Alberti- She is Assistant of Tecnical Manager -Ruredil-

Milano- Italy
INTRODUCTION
Superplasticizers are admixtures that permit a high reduction

of the water content in a given concrete mix without affecting the
workability, or considerably increase the workability without
affecting the water content, or produce both effects simultaneously.
These admixtures are used in the precast industry to obtain

high strength at early age (8-18 hours), reducing the water to cement
ratio without reducing the slump. They are also used for premix
concrete to produce flowing concrete to reduce vibrations as well as
the problem of cutting and adapting the formwork in very
congested reinforcement structures.
The superplasticizers are also advantageously used in the

production of high, durable concrete since they can reduce the
water content by approx. 30 %. Nevertheless there are some
problems associated with their use : one of these is the high rate of
loss of slump for superplasticizing concrete (1).
The ease with which plastic concrete can be deformed by an

applied stress is known as workability of the concrete and it is
measured by standard tests such as Ve Be, etc., under arbitrarily
chosen conditions of sample preparation and magnitude of the
applied stress. Slump is an indirected measure of workability.
A C 1 SP-148 94 = 0 6 6 2 9 4 9 0519882 88T
The workability of the concrete is progressively reduced as the

hydration of the cement paste proceeds. This process starts with the
mixing of the components and continues until the concrete or
mortar has been placed.
The problem is usually solved by using retarding admixtures,

water-soluble organic compounds such as IignosuIphonates (21,
hydroxycarboxylic acids (3),or carbohydrates with low molecular
weight (4). However, these molecules increase the setting time of
the cement paste, and is possible that these admixtures, under
particular conditions, also accelerate the slump loss (5-6).
In this paper different types of polysaccharide syrups at

different temperatures (25-4OOC) were studied, which can modify
the rheology and loss of workability of mortars and concretes, if
added to the naphthalene and/or melamine based superplasticizers.
MATERIALS
Polvsaccharides SvruP - The polysaccharides syrup is produced

by enzymatic hydrolysis of starch (7). Starch is stored in the roots,
seeds, and fruits of plants (corn, potatoes, wheat, tapioca, and rice).
All forms of starches contain dextrose and other compounds
constituted by two or more chemically linked dextrose molecules.
These compounds are called polysaccharide syrups and have
various dextroge equivalent values.
D.E. (equivalent dextrose) is defined as the value in grams of

reducing sugars (named dextrose) contained in 100 grams of dry
material.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 E 0 6 6 2 9 4 9 0519883 716
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Pure dextrose has D.E. = 100. On the contrary the large

molecules (polysaccharides) have a very low D.E. value. (5-10 D.E.)
The more the starch is hydrolyzed, the higher is the D.E. value.
In this paper 5 different types of polysaccharide syrups were

studied, called G6, G21, G33, G47 and G62, respectively where the
number following the letter G indicates the D.E. value.
Superplasticizers - Admixtures composed of salts from

naphthalene formaldehyde sulphonic acids (FL), or melamine
formaldehyde sulphonic acids (M) blended with polysaccharide
syrups, were studied, respectively called FLG-X and MG-X, where X
indicates the D.E. value and G the polysaccharides syrup.
There are two superplasticizers:
i) FL + polysaccharide (FLG-X)
2) M + polysaccharide (MG - X)
Cements - Two cements were studied : Portland cement

Type II (according to ASTM-250) and portland-pozzolanic cement
(PPC) Type IP (according to ASTM C595).
Mortars : mixture proportions The mortars were prepared

according to the Italian Cement Standards (D.M. 3-6-1968).
The admixtures used were :
a) polysaccharides syrup G6, G21, G33, G47, G62 were added
at 0.1, 0.2, 0.4, 0.8 % on cement weight. The mortars were
made using Portland cement.
b) naphthalene based superplasticizer (FL) with different
syrup contents
c) melamine based superplasticizer (M) with different syrup
contents.
A C 1 SP-L48 94 0662949 0529884 652 =
Each mortar composition consisted of several samples. Their

following properties were determined :
- flexural and compressive strength (tables 1-41
- workability (fig. 1-2)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
- loss of workability (fig. 1-2)

- setting time (tables 5-6)
Concrete : mix proportions - The mix design of concrete made

with Portland and pozzolanic cement at 25OC and 40°C is reported
in table 7. Three types of admixtures were used : FL G62, M G62, and
a reference superplasticizer retarding admixture according to ASTM
C494 - type G, named CL. Tables 8 and 9 show the compressive
strength, and tables 10 and 11 the loss of workability of concrete
made at 25°C and 40°C using the admixtures FL G62 and M G62.
DISCUSSION
The tests on plastic mortar shown in tables 1-4 indicate that

there is a minimum dosage (0.4 %) below which the syrups d o not
influence the mechanical properties of the mortars.
At the dosage of 0.1 % no significant flow and mechanical
strength variations (compared to the reference mortar) occurs as
can be seen in table 1.
At the dosage of 0.2 - 0.4 % an increase of flow and a slight
increase of mechanical strength at all age can be observed.
At a dosage of 0.8%, the different syrups delay the development

of compressive strength d u e to the presence of different percentages
of glucose, disaccharide and trisaccharide (table 4) as pointed out in
the literature (8) but the late-age strength can be higher.
A C 1 SP-148 94 M Obb2749 0519885 599 m
These substances delay the starting and the end of the setting
compared to the reference (tables 5-6); they however permit a better
hydration of the cement particles at later ages and therefore an
increase of the compressive strengths results after 28 days of age
(tables 3-4).
Figures 1 and 2 show the workability obtained with the syrups

at 0.1 % and 0.8 %. Brief descriptions of the data are as follows:
- At the dosage of 0.1 % no improvement of

workability for any
type of polysaccharides can be noticed, while at the dosage of 0.8 %
all syrups improve workability. Syrups G47 and G62 show the best
effects.
- The best results can be observed with G47 and G62.

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Several theories have been proposed to explain the retarding

effects of hydroxylated polymers such as polysaccharides (9-11).
According to Young (12) the retarding effects of polyhydroxylated
organic compounds is correlated to the total number of hydroxyl,
carboxylic acid, and carbonyl groups in the molecules.
From the analysis of tables 5 and 6 it can be seen that all syrups
used show values of initial set which increases as the D.E.
increases (reduction of the degree of polymerization). This confirms
that these molecules can be used as additives to maintain the
workability by delaying set.
The increase in the degree of polymerization (low D.E.) has a

low impact on the mechanical strength and on the setting, possibly
because the oxydriles group present in the large polysaccharide
molecules results in being "hidden" a n d therefore not easily
available to react with the hydration products of the cement.
A C 1 SP-L4ô 9 4 m 0 6 6 2 9 4 9 05L9ôBb 425 m
O n the bases of the results obtained with mortars, two

superplasticizing additives to be used with concrete have been
formulated :
a) FL G62, constituted by 80 parts of naphthalene sulphonate
and 20 parts of G62 syrup.
b) M G62, constituted by 80 parts of melamine sulphonate and
20 parts of G62 syrup.
A classical retarding superplasticizer, based o n calcium

gluconate and naphthalene sulphonate has been used as reference
(named CL).
The results shown in table 7, show that for both cements the
slump increases respectively by 40% and 30% with the FL G62 and
M G62, at both 25OC and 40°C) when compared with the reference
admixture.
Tables 8 and 9 show that these new additives, compared to the

classical retarding superplasticizer (CL), increase the compressive
strength at 28 days by approx. 20 % for both the Portland and the
pozzolanic cements.
This can be explained by the fact that polysaccharide syrups)

used delay the setting of the cement, permit better hydration, with a
consequent improvement of the compressive strengths at 28 days.
The results are particularly significant for the pozzolanic

cement ,
The loss of workability, results at 40°C are shown in tables 10

a n d 11. The traditional superplasticizer CL shows a loss of
workability of 75% in 1 hour, compared to the loss of 44% for FLG
62 and 53% for MG 62 in Cement II.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 m 0bb2949 0539887 3 6 1
112 Mantegana and Alberti
In cement type 'lI the traditional superplasticizer CL shows a

loss of workability of 69% in 1 hour, if compared to the loss of 55%
for FLG 62 and 59% MG 62.
CONCLUDING REMARKS
A new chemical admixture based on polysaccharides syrup

derived from enzymatic hydrolysis of starch is used as a retarding
and workability control of mortars and concrete.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
The best results are obtained with polysaccharide syrup G62
(DE=62 and a low degree of polymerization) added to melamine and
naphthalene sulphonate, compared to a normal retarding
superplasticizer admixtures of ASTM C494 - type G.
REFERENCES
1. Rixom, M.R. Chemical Admitures for Concrete, London:

E. & F.N. Spon LTD (1978)
2. Ramachandran, V.S. and Feldman, R.F., Adsorption of
calcium lignosulphonate on tricalcium aluminate and its
hydrates in a non-aqueous medium, Cement Tecnology 2
(4): 121-29 (1971)
3. Danielsson, S. Studies of the effects of some simple
organic admixtures on the properties of cement paste,
International Symposium on Admixtures for Mortar and
Concrete, Vol. 2,57-68,Brussels 1967
4. Massazza, F. and Testolin, M., Latest Developments in the
Use of Admixtures for Cement and Concrete , I1 Cemento
77: 73 - 146, (1980)
A C 1 SP-I48 94 066294'7 0 5 L 9 B B ô 2T8
5. Ravina, D., Retempering of Prolonged-Mixed Concrete

with Admixtures in Hot Weather, American Concrete
Institute Journal , 72:291-95, (197.5).
6. Previte, R.W., Concrete Slump Loss, American Concrete
Institute Journal, 74: 361-67, (1977).
7. S.B. Hendrickson, D.J. Cram, G.C. Hammond "Organic
Chemistry" The chemistry of natural products - pag. 1059
Mc Graw-Hill Series.
8. ANOM- Admixture for Concrete, Concrete Society 1967
Report 1
9. Steinour, H. M., Discussion of "Actions of calcium sulfate
on the hydration and the microstructure of hardened
mortar of C3S", by W.C. Hansen, ASTM Techincal
Publication 266,25-37 (1960)
10. Taplin , J.H., Discussion of "Some chemical additions and
admixtures in cement paste and concrete" by H.E. Vivian,
Proceedings of the Fourth International Symposium on
the Chemistry of Cements, Washington, D.C., Vol.
2,924-25, (1962).
11. Daugherty, K.E., and Kowalesky, M.J. Jr., Effect of Organic
Compounds on the Hydration Reactions of of Tricalcium
Aluminate, Proceedings of the 5th International
Symposioum Chemistry of Cements, Tokyo Vol. 4,42-52,
(1968).
12. Young, J.F., A Review of the Mechanism of set Retardation
in Portland Cement. Pastes Containing organic
Admixtures, Cement and Concrete Resaerch, 2: 415 -33,
(1972).
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 05398159 3 3 4
TABLE 1 - MORTARS, 0.1 PERCENT ADMIXTURE DOSAGE. W/C 05
Compressive strength IFlexural strength

I
w
IAdmixtures
IReference 'îî,5 (24,l (34,s 144,7
12,4 124,9 133,B 144 3'3 4,7 5,6 7,6
IG21 12,3 124,9 133,7 143,s 3,3 14,9 159 17,ó 1
I G33
IG47 I ' I
3,5 153 163 18,l
TABLE 2 - MORTARS, 0.2 PERCENT ADMIXTURE DOSAGE. W/C = 05

Compressive strength Flexural strength I
íMPa)
Admixtures
Reference 12,5 24,l 34,5 44,7
12,s 25,2 34,l 44,8
12,7 25.1 34,5 44,s
G33 13,l 25,9 34,l 43,5
G47 12,7 25,2 33'8 45,3
G62 13,9 26.2 35

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Compressive strength IFlexural strength

(MPa) I (MPa)
IAdmixtures Ild
24,9 135,5 143,B 13,3 1.5 16,4
G47 13,7
G62 14,6
A C 1 SP-148 94 IOb62949 0519890 95b w
(Compressive strength ]Flexural strength I

(MPa) (MPa)
Admixhires Id
-
12,s
IG6
I
12,l 36,2 '5 (6,5 (7,7
IG21 11.9 36.6 44,6 3,l
11 46.8 3.2
10,6 25,4 ~4,8
IG62 10,5 24,9 139,5 4,8 6,3
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 5 - INITIAL AND FINAL SET OF MORTARS,
0.4 PERCENT ADMIXTURE DOSAGE
]Initial IFinal I
G21 215 275
I F
IG62
210
1250 1 2 9 0 1
TABLE 6 - INITIAL AND FINAL SET OF MORTARS,

0.8 PERCENT ADMIXTURE DOSAGE
Initial I Final I
set (min.) I set(min.1 I
Reference 185 235
G6 240 285
G21 325 380
G33 320 380
G47 330 380
G62 350 400
A C 1 SP-148 94 0662947 05L989L ô 9 2 =
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 M 0662747 0539892 727
TABLE 8 - MECHANICAL PROPERTIES OF CONCRETE MADE WITH CEMENT TYPE II
Admixture W/C Initial Air

Type IDOS.% Slump content mixt.
(crn) (%I ("Cl Id 36 7d 28d
0,5 8 291 25 16 30,2 36,3 45,5
CL 2 0,45 14 2,8 25 20 38,l 45,3 54,3
FLG62 2 0,45 20 2,2 25 20,4 39,7 46,8 57,2
M G62 2 0.45 18 1,8 25 22.5 40.8 47.3 56.8
I I
CL 2 0,45 12 2,9 40 22,5 40,2 45,8 53,s
FLG62 2 0,45 18 2,l 40 23,l 40,8 46,5 56,3
M G62 2 0,45 17 1,8 40 24,8 42,l 47,8 55,9
TABLE 9 - MECHANICAL PROPERTIES OF CONCRETE MADE WITH CEMENT TYPE IP
CL 2 0,46 13 40 22,l 29,6 36 48

FLG62 2 0,46 18 40 21,5 30,8 38,5 51
M G62 2 0,46 17 40 23,6 33,3 37,8 50,6
TABLE 10 - WORKABILITY LOSS OF CONCRETE MADE WITH CEMENT TYPE II AT 40 C
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-248 74 Ob62747 05l17873 665
TABLE 11 - WORKABILITY LOSS OF CONCRETE MADE WITH CEMENT TYPE IP AT 40 C
1 O0 0 Ref.
G6
80
E
-E
B 60
G21
O G33
ü
40
G47
20 0G62
O
O 15 30 4 5 60 90
tirne(rninutes
Fig. 1-Workability loss of polysaccharide syrups at 25 C Dosage 0.1 percent on cement
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 M 0662749 05L7874 5 T l =
140
120
1 O0 0Ref.
Y
E 80 G6
I G21
2 60
LI.
G33
40
1G47
20
0G62
O
O 15 30 45 60 90
tirne(minutes
Fig. 2-Workability loss of polysaccharide syrups at 25 C. Dosage 0,8 percent on cement
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662747 0 5 3 9 8 9 5 4 3 8 =
SP 148-7
Workability Characteristics of
High-Strength Concrete Incorporating
an Air-Entraining, High-Range
Water-Reducing Admixture
by M. Kagaya, H. Tokuda,
M. Kawakami, and T. Kaneko
Synopsis: Righ s t r e n g t h c o n c r e t e w i t h a i r - e n t r a i n i n g h i g h r a n g e
Water-reducing a d m i x t u r e (ABW) ,0.35 water-to-cement r a t i o , 148kg
/ n i 3 u n i t w a t e r c o n t e n t and a b o u t 6OMPa c o m p r e s s i v e s t r e n g t h was
p r o d u c e d , a n d t h e a d m i x t u r e c o n t e n t a n d s a n d p e r c e n t a g e were
v a r i e d . The c o n s i s t e n c y p r o p e r t y was measured by t h e slump t e s t ,
a n d t h e c o m p a c t i o n p r o p e r t y a n d s e g r e g a t i o n r e s i s t a n c e was
d e t e r m i n e d by t h e f l o w time t h r o u g h au i n v e r t e d slump cone. An
a p p r o p r i a t e m i x t u r e which b a l a n c e d t h e s e p r o p e r t i e s r e l a t i v e t o
t h e admixture c o n t e n t and t h e sand p e r c e n t a g e v a s determined
from t h e t e s t . The a p p r o p r i a t e m i x t u r e c a n b e c o n s o l i d a t e d by
u s i n g a s h o r t e r v i b r a t i n g time and a lower f r e q u e n c y t h a n i n t h e
c a s e of o r d i n a r y c o n c r e t e .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Keywords: Admixtures; air entrainina aaents; compaction; compressive strength; consolidation;

hhh strenpth concretes; segregation; slump; vibrators (machinery); water redvcina aaents;
workability

A C 1 SP-148 94 m Obb2949 0 5 1 9 8 9 6 374 m
122 Kagaya et al
AC1 member H a k o t o K A G A Y A i s a a s s o c i a t e p r o f e s s o r a t t h e
Department o f C i v i l Engineering, Akita U n i v e r s i t y . Recent
r e s e a r c h themes i n c l u d e s l a g c o n c r e t e , r e c y c l e d - a g g r e g a t e
c o n c r e t e and r o l l e r c o m p a c t e d c o n c r e t e .
AC1 member H i r o s h i Tokuda i s t h e Dean o f t h e M i n i n g C o l l e g e ,
A k j t a U n i v e r s i t y . He h a s p u b l i s h e d numerous p a p e r s on t h e r m a l
and o t h e r c o n c r e t e p r o p e r t i e s .
A C 1 member Makoto KAWAKAHI i s a p r o f e s s o r a t t h e D e p a r t m e n t of
C i v i l E n g n e e r i n g , Akita U n i v e r s i t y . He h a s p u b l i s h e d p a p e r s on
concrete s t r u c t u r a l analysis.
T a d a h i k o K A N E K O i s c h i e f e n g n e e r a t t h e Kanto c o n c r e t e p r o d u c t
i n d u s t r y , D P S B r i d g e Works C O . L T D . One o f h i s m o s t r e c e n t
p r o j e c t s i s t h e p r o d u c t i o n of a h i g h s t r e n g t h c o n c r e t e p r e c a s t
elemen t .
INTRODUCTION
The u s a g e o f a new c l a s s o f c h e m i c a l a d m i x t u r e s h a s
i n c r e a s e d s u b s t a n t i a l l y i n t h e ready-mixed and p r e c a s t c o n c r e t e
i n d u s t r i e s . The a i r - e n t r a i n i n g , h i g h - r a n g e w a t e r - r e d u c i n g
a d m i x t a u r e ( A H W ) , c a n be u s e d t o i n c r e a s e s i g n i f i c a n t l y s l u m p
w i t h o u t a d d i n g more w a t e r , or t o g r e a t l y r e d u c e w a t e r c o n t e n t
w i t h o u t l o s s i n s l u m p . S u b s e q u e n t l y , a c a r e f u l c h o i c e of t h e
a d m i x t u r e c o n t e n t and t h e s a n d p e r c e n t a g e i s r e q u i r e d t o o b t a i n
a h i g h c o m p a c t i o n p r o p e r t y and s e g r e g a t i o n r e s i s t a n c e o f t h e
concrete(1).
I n t h i s p a p e r , slump, compaction and d e g r e e of s e g r e g a t i o n
o f h i g h - s t r e n g t h c o n c r e t e i n c o r p o r a t i n g ARW i s d i s c u s s e d i n
r e l a t i o n t o t h e a d m i x t u r e c o n t e n t and t h e sand p e r c e n t a g e , and
an a p p r o p r i a t e m i x t u r e p r o p o r t i o n i s d e t e r m i n e d from t h e r e s u l t
of t h e i n v e r t e d slump cone t e s t i n g . Furthermore, the
c o n s o l i d a t i o n c h a r a c t e r i s t i c s of t h e c o n c r e t e a r e c o i p a r e d t o
t h o s e of t h e o r d i n a r y c o n c r e t e .
EXPERIMENTAL OUTLINE
m e r i a l s and M i x t u r e P r o p o r t i o n s
Normal p o r t l a n d cement. r i v e r sand ( s p e c i f i c g r a v i t y : 2 . 5 6 ,
w a t e r a b s o r p t i o n : 2 . 9 1 % and f i n e n e s s m o d u l u s : 2 . 7 3 ) and c r u s h e d
s t o n e ( s p e c i f i c g r a v i t y : 2 . 6 0 , water absorption:2.17% f i n e n e s s
a o d u l u s : 6 . 8 0 , maximum size:ZOmm)were u s e d . An a i r - e n t r a i n i n g ,
high range water-reducing admixture (AHW) c o n s i s t i n g mainly of
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L48 94 0 6 6 2 9 4 9 0519897 200 =
p o l y a l k y l a r y l s u l f o n a t e and r e a c t i v e h i g h polymer was u s e d . The

a d m i x t u r e c o n t e n t r a n g e was 1 . 0 t o 3.0% by weight of cement.
A s u p p l e m e n t a r y a i r - e n t r a i n i n g a g e n t was u s e d t o c o n t r o l t h e a i r
c o n t e n t . The water-to-cement r a t i o arid w a t e r c o n t e n t were 0 . 3 5
a n d 148kg/m3 r e s p e c t i v e l y , a n d t h e s a n d p e r c e n t a g e was v a r i e d
from 35.0 t o 50.0%. O r d i n a r y c o n c r e t e was a l s o made i n o r d e r t o
compare i t s c o n s o l i d a t i o n c h a r a c t e r i s t i c s w i t h t h o s e o f t h e
a b o v e . T h e s e m i x t u r e p r o p o r t i o n s a n d c h e m i c a l c o m p o s i t i o n of
normal p o r t l a n d cement a r e shown i n T a b l e 1 and 2.
Measurement o f Compaction P r o p e r t y and Degree o f S e g r e g a t i o n
The c o m p a c t i o n p r o p e r t y was d e t e r m i n e d b y u s i n g t h e
compressive s t r e n g t h r a t i o of t h e c y l i n d r i c a l t e s t specimen.100x
200nm i n s i z e , c a s t w i t h o u t a n y c o n s o l i d a t i o n a n d by t h e
s t a n d a r d method f o l l o w i n g J a p a n e s e I n d u s t r i a l S t a n d a r d s A 1132.
T o m e a s u r e t h e d e g r e e o f s e g r e g a t i o n by v i b r a t i o n , two
c y l i n d r i c a l s p e c i m e n s , 150X300nm i n s i z e , were made. C o n c r e t e
was p l a c e d i n t o t h e mold i n one l a y e r ; o n e was c o n s o l i d a t e d by
v i b r a t i o n for 10 s e c o n d s and t h e o t h e r f o r 30 s e c o n d s .
C o n c r e t e samples were t a k e n from b o t h t h e upper and lower h a l v e s
of e a c h specimen. The r a t i o o f t h e c o a r s e a g g r e g a t e c o n t e n t i n
t h e u p p e r a n d l o w e r p a r t s a m p l e s was d e t e r m i n e d by m i x t u r e
a n a l y s i s . T h e d e g r e e o f s e g r e g a t i o n was d e t e r m i n e d b y
s u b t r a c t i n g t h e r a t i o a t 30 s e c o n d s v i b r a t i n g t i m e from t h e o n e
a t 10 s e c o n d s . The lower t h e v a l u e , the h i g h e r t h e s e g r e g a t i o n
r e s i s t a n c e . The v i b r a t o r u s e d 210Hz f r e q u e n c y a n d had a 2 5 m m
diameter.
I n v e r t e d Slump Cone T e s t
I n o r d e r t o measure t h e w o r k a b i l i t y o f a s p e c i f i c t y p e of
c o n c r e t e a s f o r e x a m p l e t h e c o m p a c t i o n p r o p e r t y and d e g r e e o f
s e g r e g a t i o n d e g r e e , t h e flow time t h r o u g h an i n v e r t e d slump cone
was measured a s shown i n F i g . l ( 2 ) .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Consolidation C h a r a c t e r l s t i c s B g
The c o n c r e t e was p l a c e d i n t o t h e c y l i n d r i c a l mold, 15OX
3 0 0 m m i n s i z e , and c o n s o l i d a t e d a t d i f f e r e n t f r e q u e n c i e s (150
and 210Hz) and v i b r a t i n g t i m e ( O t o 3 0 s e c . I . A f l e x i b l e
v i b r a t o r (30mm d i a m e t e r ) was u s e d . The p e r c e n t c o n s o l i d a t i o n
was d e t e r m i n e d f r o m t h e c o m p r e s s i v e s t r e n g t h r a t i o o f t h e
s p e c i m e n c o n s o l i d a t e d d u r i n g e a c h v i b r a t i o n p e r i o d and by t h e
s t a n d a r d method f o r b o t h t h e h i g h - s t r e n g t h a n d t h e o r d i n a r y
concrete.
Compressive S t r e n g t h T e s t
.________
The t e s t was c a r r i e d o u t a t t h e a g e of 28 d a y s , a f t e r t h e
specimens had been c u r e d i n a w a t e r b a t h a t 21T.
A C 1 SP-148 94 Obb2949 0519898 247
124 Kagayo et al
RESULTS A N D DISCUSSION
slupip
Fig. 2 shows t h e r e l a t i o n s h i p between a d m i x t u r e c o n t e n t (Ad!
and s l u m p . C o n c r e t e slump i n c r e a s e s w i t h a n i n c r e a s e i n t h e
a d m i x t u r e c o n t e n t and g r a d u a l l y a p p r o a c h e s 200mm beyond t h e 2%
admixture content. This i n d i c a t e s t h a t f u r t h e r consistency
improvement c a n n o t be g a i n e d by a d d i n g much more a d m i x t u r e i n
these mixtures.
F i g . 3 shows t h e r e l a t i o n s h i p b e t w e e n s a n d p e r c e n t a g e
( s / a ) and slump. I n t h e c a s e of a 2% a d m i x t u r e c o n t e n t , slump
d e c r e a s e s w i t h t h e i n c r e a s e i n sand p e r c e n t a g e , and t h e
d e c r e m e n t r a t e i n c r e a s e s beyond t h e 42.5% s a n d p e r c e n t a g e . I n
t h e c a s e of 3 % a d m i x t u r e c o n t e n t , t h e s i g n i f i c a n t v z r i a t i o n of
slump c a n n o t be o b s e r v e d and t h e v a r i a t i o n r a n g e i s from 180 t o
210mm r e g a r d l e s s of t h e i n c r e a s e i n s a n d p e r c e n t a g e . T h e s e
r e s u l t s d e m o n s t r a t e t h a t t h e d e c r e a s i n g t e n d e n c y of slump w i t h
t h e i n c r e a s e i n sand percentage d i s a p p e a r s g r a d u a l l y by an
i n c r e a s e i n admixture content.
Compaction P r o p e r t y and D e g r e e o f S e g r e g a t i o n
Fig.4 shows t h e r e l a t i o n s h i p between a d m i x t u r e c o n t e n t (Ad)
a n d , s a n d p e r c e n t a g e ( s / a ) , a n d c o m p a c t i o n p r o p e r t y . The
c o m p r e s s i v e s t r e n g t h r a n g e of t h e specimen made b y t h e s t a n d a r d
method was from 56.9 t o 58.9MPa a t t h e a g e of 28 d a y s . In t h e
l e f t s i d e f i g u r e , compaction p r o p e r t y i n c r e a s e s w i t h an i n c r e a s e
i n t h e a d m i x t u r e c o n t e n t and g r a d u a l l y a p p r o a c h e s 0.6 beyond t h e
2% a d m i x t u r e c o n t e n t . I n t h e r i g h t s i d e f i g u r e , t h e c o n p a c t i o n
p r o p e r t y f o r 2% a d m i x t u r e c o n t e n t d e c r e a s e s from 0.55 t o 0 . 2
when s a n d p e r c e n t a g e i n c r e a s e s from 35 t o 50% and t h e p r o p e r t y
f o r 3 % a d m i x t u r e c o n t e n t becomes a l m o s t c o n s t a n t a t 0 . 6 w i t h
r e s p e c t t o sand p e r c e n t a g e .
These r e s u l t s i n d i c a t e t h a t some compaction w i l l always be
required because consolidation is n o t needed f o r a 1.0
c o m p a c t i o n p r o p e r t y . Compaction e a s e depends on t h e a d m i x t u r e
c o n t e n t a n d t h e s a n d p e r c e n t a g e , b u t a d d i n g more a d m i x t u r e
r e d u c e s t h e s e mutual d e p e n d e n c i e s . T h e s e c o m p a c t i o n p r o p e r t y
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
t e n d e n c i e s r e l a t e d t o a d m i x t u r e c o n t e n t and sand p e r c e n t a g e a r e
s i m i l a r t o t h o s e of t h e s l u m p m e n t i o n e d a b o v e . I t a l s o shows
t h a t s l u m p d e t e r m i n e s compaction p r o p e r t y i n t h e s e m i x t u r e s .
F i g . 5 shows t h e r e l a t i o n s h i p between t h e a d m i x t u r e c o n t e n t
(Ad) and t h e sand p e r c e n t a g e ( s / a ) , and d e g r e e of s e g r e g a t i o n .
I n t h e l e f t s i d e f i g u r e , t h e d e g r e e of s e g r e g a t i o n i n c r e a s e s
(more s e g r e g a t i o n ) w i t h an i n c r e a s e i n a d m i x t u r e c o n t e n t . I n
t h e r i g h t s i d e f i g u r e , t h e d e g r e e f o r 2 . 0 and 3 . 0 % a d m i x t u r e
c o n t e n t d e c r e a s e s ( l e s s s e g r e g a t i o n ) w i t h an i n c r e a s e i n s a n d
p e r c e n t a g e . These d e g r e e of s e g r e g a t i o n t e n d e n c i e s r e l a t e d t o
a d m i x t u r e c o n t e n t and s a n d p e r c e n t a g e a r e somewhat d i f f e r e n t
A C 1 SP-148 9 4 0662747 0537899 083
from t h o s e of t h e slomp.
From t h e r e s u l t s o f F i g . 4 a n d 5, i t i s e v i d e n t a t f i r s t
t h a t a n improvement of t h e compaction p r o p e r t y by a n i n c r e a s e i n
a d m i x t u r e c o n t e n t makes t h e d e g r e e o f s e g r e g a t i o n g r o w , a n d
s e c o n d l y t h a t o n e way t o r e d u c e i t i s t o i n c r e a s e t h e s a n d
p e r c e n t a g e . I t i s n e c e s s a r y t o choose p r o p e r p r o p o r t i o n s h a v i n g
a n a d e q u a t e a d n i x t u r e c o n t e n t and s a n d p e r c e n t a g e from t h e view
p o i n t s of b o t h compaction p r o p e r t y and s e g r e g a t i o n r e s i s t a n c e t o
vibration.
A
_p_
p r o p r i a t e M i x t u r e E v a l u a t e d by I n v e r t e d Slump Cone T _
e s t.
Fig.6 shows t h e r e l a t i o n s h i p between a d m i x t u r e c o n t e n t (Ad)
and sand p e r c e n t a g e ( s / a ) , and i n v e r t e d cone t i m e . I n v e r t e d
cone t i m e d e c r e a s e s by an i n c r e a s e i n t h e a d m i x t u r e c o n t e n t a n d
i n c r e a s e s by an i n c r e a s e i n t h e s a n d p e r c e n t a g e w i t h and s i t h o u t
v i b r a t i o n . These t e n d e n c i e s i n d i c a t e t h a t t h e r e i s improvement
i n t h e m o b i l i t y or t h e f l u i d i t y o f t h e c o n c r e t e when a d m i x t u r e
c o n t e n t i n c r e a s e s f o r a g i v e n s a n d p e r c e n t a g e , and t h e sand
p e r c e n t a g e d e c r e a s e s f o r a g i v e n a d m i x t u r e c o n t e n t . Aowever,
non-flow t h r o u g h an i n v e r t e d slump cone o c c u r s w i t h o u t v i b r a t i o n
a t 1 . 0 % a d m i x t u r e c o n t e n t f o r 40% s a n d p e r c e n t a g e and a t 50% o f
s a n d p e r c e n t a g e f o r 2.9% a d m i x t u r e c o n t e n t . S i n c e a n i n v e r t e d
cone time with v i b r a t i o n responds well t o a l l a i x t u r e s i n a
s h o r t time c o m p a r e d w i t h t h e time w i t h o u t v i b r a t i o n , i t i s
e v i d e n t t h a t t h i s v a l u e e f f e c t i v e l y m e a s u r e s t h e w o r k a b i l i t y of
t h e concrete under i n t e r n a l v i b r a t i o n i n t h e s e mixtures.
F i g . 7 shows t h e r e l a t i o n s h i p b e t w e e n t h e i i i v e r t e d c o n e
t i m e w i t h v i b r a t i o n a n d c o m p a c t i o n p r o p e r t y . T h i s shows t h a t
t h e compaction p r o p e r t y i s almost, c o n s t a n t f o r l e s s t h a n 5
s e c o n d s i n v e r t e d c o n e t i m e b u t d e c r e a s e s by a n i n c r e a s e i n t h e
t i m e beyond 5 s e c o n d s . T h i s d e c r e a s i n g t e n d e n c y d e m o n s t r a t e s
t h a t t h e g r e a t e r t h e inverted cone time, the lower the
compaction p r o p e r t y .
F i g . 8 shows t h e r e l a t i o n s h i p b e t w e e n t h e j n v e r t e d c o n e
time and t h e d e g r e e of s e g r e g a t i o n . The d e g r e e of s e g r e g a t i o n
f a l l s s h a r p l y w i t h a 6 second i n v e r t e d cone time. A ï t e r t h i s
t i m e , t h e d e c r e m e n t r a t e d e c r e a s e s , a n d t h e d e g r e e s a r e much
l e s s t h a n t h o s e p r e c e d i n g , so t h e s e g r e g a t i o n r e s i s t a n c e o f
t h e s e m i x t u r e s t o v i b r a t i o n i s much b e t t e r .
Fig. 7 and 8 r e s u l t s i n d i c a t e t h a t the degree of
c o m p a c t i o n a n d s e g r e g a t i o n c a n b e e v a l u a t e d by i n v e r t e d c o n e
time r e g a r d l e s s of t h e d i P f e r e n c e i n p r o p o r t i o n , such as t h e
admixture c o n t e n t and sand p e r c e n t a g e . These r e s u l t s s u g g e s t
t h a t a balanced mixture between compaction p r o p e r t y and
s e g r e g a t i o n r e s i s t a n c e c a n be o b t a i n e d a t 5 t o 6 s e c o n d s
i n v e r t e d c o n e t i m e . I n t h i s c a s e , a 2% a d m i x t u r e c o n t e n t and a
40 t o 4 2 . 5 % s a n d p e r c e n t a g e make a g o o d c o m b i n a t i o n f o r t h e
m i x t u r e w i t h a slump of a b o u t 1 7 0 m .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 H 0662949 0539900 625 =
126 Kagoya et al
Consol i d a t i_______
on C h a r a c t e r i s t i c s
F i g . 9 shows t h e r e l a t i o n s h i p b e t w e e n t h e v i b r a t i n g t i m e
and t h e p e r c e n t c o n s o l i d a t i o n . For t h e r e s u l t s of h i g h s t r e n g t h
c o n c r e t e , t h e p r e v i o u s l y m e n t i o n e d b a l a n c e d m i x t u r e was u s e d .
The p e r c e n t c o n s o l i d a t i o n r a p i d l y i n c r e a s e d , a n d i t b e c a m e
a l m o s t c o n s t a n t w i t h r e s p e c t t o v i b r a t i n g t i m e . I n t h e c a s e of
210Hz v i b r a t o r f r e q u e n c y , 1 0 0 % c o n s o l i d a t i o n i n d i c a t i n g f u l l
c o m p a c t i o n s t a t e c a n be o b t a i n e d a t 6 and 2 0 s e c o n d s v i b r a t i n g
time f o r h i g h - s t r e n g t h c o n c r e t e and o r d i n a r y c o n c r e t e
r e s p e c t i v e l y . I n t h e c a s e o f 150Hz v i b r a t o r f r e q u e n c y , however
t h e p e r c e n t c o n s o l i d a t i o n s a r e no more t h a n 96 and 85% even a t
f i n a l v i b r a t i n g time.
These r e s u l t s s u g g e s t t h a t an a p p r o p r i a t e p r o p o r t i o n i n g
m i x t u r e i n c o r p o r a t i n g AHH c a n p r o v i d e a r e d u c t i o n of c o m p a c t i o n
e f f o r t s , n o i s e of p l a c e m e n t and s e g r e g a t i o n bu s i g n i f i c a n t l y
s h o r t e n i n g t h e v i b r a t i n g t i m e requirement. and l o w e r i n g t h e
v i b r a t o r f r e q u e n c y compared w i t h t h a t o f o r d i n a r y c o n c r e t e .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
C O N C L U S IONS
In o r d e r t o i n v e s t i g a t e t h e workability c h a r a c t e r i s t i c s of
h i g h - s t r e n g t h c o n c r e t e w i t h AHN and w i t h a water-to-cement r a t i o
o f 0 . 3 5 , s l u m p , c o m p a c t i o n p r o p e r t y and d e g r e e o f s e g r e g a t i o n
r e l a t e d t o t h e admixture c o n t e n t and sand p e r c e n t a g e were
m e a s u r e d and t h e c o n s o l i d a t i o n c h a r a c t e r i s t i c s were compared
w i t h o r d i n a r y c o n c r e t e . The f o l l o w i n g c o n c l u s i o n s may he drawn
from t h e i n v e s t i g a t i o n :
( 1 ) C o n c r e t e s l u m p i n c r e a s e s w i t h an i n c r e a s e i n a d m i x t u r e
c o n t e n t and d e c r e a s e s w i t h ai! i n c r e a s e i n s a n d p e r c e n t a g e .
T h e s e t e n d e n c i e s g r a d u a l l y d i s a p p e a r by an i n c r e a s e i n t h e
a d m i x t u r e content..
( 2 ) T h e improvement o f compaction by an i n c r e a s e i n t h e a d m i x t u r e
c o n t e n t makes t h e d e g r e e of s e g r e g a t i o n i n c r e a s e , and one way
t o r e d u c e t h e d e g r e e of s e g r e g a t i o n i s t o i n c r e a s e t h e s a n d
percentage.
( 3 ) T h e compaction and d e g r e e of s e g r e g a t i o n c a n be e v a l u a t e d by
i n v e r t e d c o n e t i m e r e g a r d l e s s of t h e d i f f e r e n c e i n t h e
m i x t u r e s , such a s t h e a d m i x t u r e c o n t e n t and sand p e r c e n t a g e .
( 4 ) A b a l a n c e d m i x t u r e f o r c o m p a c t i o n p r o p e r t y and s e g r e g a t i o n
r e s i s t a n c e can be o b t a i n e d a t 5 t o 6 seconds i n v e r t e d cone
t i m e . I n t h i s c a s e , a 2% a d m i x t u r e c o n t e n t and a 40 t o 4 2 . 5 %
sand p e r c e n t a g e make a good coinbination w i t h a s l u m p of a b o u t
170nim.
(5)A balanced mixture can provide a r e d u c t i o n i n compaction
e f f o r t , n o i s e o f p l a c e m e n t and s e g r e g a t i o n by s i g n i f i c a n t l y
s h o r t e n i n g t h e v i b r a t i n g t i m e r e q u i r e m e n t and by l o w e r i n g t h e
A C 1 SP-148 9 4 0662949 O539903 561
v i b r a t o r f r e q u e n c y r e q u i r e m e n t as compared w i t h t h a t r e q u i r e d
for ordinary concrete.
REFERENCES
1 . Committee Report A C 1 212.4R: Guide f o r t h e Use o f High-Range

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Water-Reducing Addmixtures ( S u p e r p l a s t i c i z e r s ) i n C o n c r e t e ,
Concrete I n t e r n a t i o n a l , April 1993 pp.40-47.
2 . Committee R e p o r t AC1 514.ZR-89: Measurement OP P r o p e r t i e s of
F i b e r R e i n f o r c e d C o n c r e t e , pp.1-10.
TABLE 1 - MIXTURE PROPORTION OF CONCRETE
i
(I)iU
14 420 597
95*0 14ü / MO
C /597
S
j E /Ad.
1130 8.40
1130 12.60
1 AE
7.6
-
155
TABLE 2 -
CHEMICAL COMPOSITION
OF NORMAL PORTLAND CEMENT
Blaine: 320 m 2 h g
128
A C 1 SP-198
Kagaya et al
94 = 0662949 0539902 4 T 8 =
Slump cone
Fig. 1-Inverted slump cone test
p6Rl
1.0 2.0 3.0 35 40 45 50
Ad(X) sla(X)
Fig 2-Admixture content Fig. 3-Sand percentage
Ad versus slump sla versus slump
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O
k0.4
i3
.r(
w
o 0.2
O
O
O
l.U 2.0 3.0 35 40 45 50
AdíX) sla(X)
Fig. 4-Admixture content Ad and sand percentage sla versus compaction property
o. 20
O. 16
I
*
.e
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
[o. 12
u
VI
*
;o.oe
k
O. 04
O
1.0 2.0 3.0 35 40 45 5ü
4d(X) s/aW
Fig. M d m i x t u r e content Ad and sond percentage sla versus degree of segregation
40
n
o
W
OI
32
W
I
.4
CI
;24
O
O
F
c,
16
L1
W
æ
2 8
o
1.0 2.0 3.U 35 40 45 50
Ad(%) da(%)
Fig. X d m i x t u r e content Ad and sand percentage slo versus inverted cone time
A C 1 SP-348 94 0662949 0539904 270 m
130 Kagaya et al
o. 20
0.8 O. 16
h
c>
2 0.6
O
' O . 12
Li
a
g 0.4 ' O O. o8
4
O. 04
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
n O
"0 4 8 12 16 o 4 8 12 16
Inverted cone t i i e (rec) Inverted cone tiae bec)
Fig. 7-Inverted cone time Fig. &Inverted tone time

versus compaction property versus degree of segregation
Ord inaru concrete High strength concrete
O 10 20 30 O 5 10 15
Vibrating tiae (sec)
Fig. 9-Vibrating time versus percent consolidation
A C 1 SP-148 9 4 m 0662947 0 5 3 9 9 0 5 107 m
SP 148-8
Effect of a Phosphonate-Based
Compound on the Hydration of
Cement and Cement Components
by V.S. Ramachandran and M.S. lowery
Svnopsis: The influence of a phosphonate-based admixture called

aminotri(methy1ene phosphonic acid), abbreviated to ATMP, on the
hydration of C3A, C3A and 25% gypsum, C3S and Type I cements
(normal, high alkali and low alkali types) was evaluated. Isothermal
conduction calorimetric investigations of various mixtures containing
0.03-0.05% of phosphonic acid, at a water:cement ratio of 0.35 and
1.0, were carried out up to 72 hrs. The ATMP acted as a super-
retarder for all the mixtures studied. At a dosage of only 0.05%, the
exothermal effect for one of the high alkali Type I cements was
delayed by about 16 hrs. The hydration of C3A and C3S was also
retarded by about 40-45 mins and > 50 hrs respectively at a dosage
of 0.05%. In the C3A and gypsum system the third peak
corresponding to the reaction of C3A with ettringite was extended by
about 8-9 hrs.
Keywords: Admixtures; calcium aluminates; calcium silicates; calorimeters; cements; gypsum;

hydration; phosphonate; portland cements
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---

A C 1 SP-148 94 Obb29Y7 0539906 043
132 Ramachandran and Lowery
V.S. Ramachandran is a Distinguished Research Officer/Principal

Research Officer, Materials Laboratory, Institute for Research in
Construction, National Research Council of Canada. He is the author
of numerous research papers and chapters and several books. He
has won several awards for his research in cement and concrete
science.
M.S. Lowery is a technical officer at the Materials Laboratory, Institute

for Research in Construction, National Research Council of Canada,
and has been involved in research on admixtures in concrete. He
has published four research papers on novel admixture systems.
INTRODUCTION
An admixture is an ingredient that is added in small amounts to

concrete to obtain several beneficial effects such as controlled setting
and hardening, improved workability and strength, and increased
frost and sulfate resistance. Most concrete placed in North America
contains one or more admixtures. A patent literature survey on
admixtures indicates that in the years 1990-1992, on an average,
more than 80 patents were taken annually, suggesting that there has
been a sustained interest in the development of novel admixtures.
Retarding admixtures in small amounts are used in concrete

practice to delay the setting times of cement paste, mortar and
concrete. In hot weather concreting, delays in transport and handling
between mixing and placing may result in early setting and loss of
workability and in such instances incorporation of retarders becomes
necessary. Retarders may be used in steam curing of concrete, to
offset the long term lower strengths developed by the silicate phase
and in the construction of large structural units, dams and the
fabrication of exposed aggregate panels. Another important
application of retarders is to maintain returned concrete from ready-
mixed trucks in a workable condition overnight by completely freezing
the hydration.
Many organic and inorganic compounds, including those

derived as by-products from industries are capable of acting as
retarders in concrete. The retarders are based on unrefined Na, Ca
or NH4 salts of lignosulfonic acids, hydroxycarboxylic acids and their
salts, phosphates, fluorates, carbohydrates, lead and zinc oxides,
borates and magnesium salts (1-5).
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 94 Obb2ïLl7 0 5 3 9 9 0 7 T B T
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
The action of many of the admixtures is related to their ability to
form complexes with the hydrating compounds in the cement-water
system (6-1O). Phosphonic acid-based chemicals are known to form
complexes with the inorganic species. Thus if present in small
amounts in the cement-water systems, they should be able to form
complexes and influence the hydration reactions. Phosphonatec are
suggested for use in high temperature oil and gas plugging
operations where a temperature in the order of 90°C or more is
encountered (1 1). Phosphonates may also be used for soil-cement
mixtures, gypsum plasters and as set time extenders for cements.
Most work related to phosphonates is to be found only in patents and
their role in the hydration of portland cement at normal temperatures
has recently been documented (12). The acids were found to be
much better retarders than their salts.
In order to use phosphonates in concrete for different

purposes, it is essential to understand their effect on the individual
compounds of cement. Hence the effect of a phosphonic acid based
compound (aminotri(methy1ene phosphonic acid)) or ATMP on the
hydration behavior of portland cements of different compositions,
tricalcium silicate, tricalcium aluminate, and tricalcium aluminate-
gypsum was investigated.
EXPERIMENTAL
Materials
Cements -- Four Normal Type I portland cements were used

in this study. They included an Exshaw brand from Alberta, two types
from Lafarge Canada Inc., and a fourth supplied by St. Laurent
Independent. Alkali levels, based on Na20 equivalence, among the
four cements ranged from 0.44 to 0.85. The tricalcium silicate and
aluminate components were obtained from the Construction
Technology Laboratory of Skokie, Illinois. A reagent grade hydrated
gypsum (CaS04.2H20) was employed. The physical properties and
chemical analysis of the cements are given in Table 1.
Phosphonic acids -- The phosphonic acid compound used

was manufactured and supplied by Monsanto Chemical Company,
St. Louis, MO., U.S.A. The phosphonate compound studied was
Aminotri(methylenephosphonic acid) (ATMP). Figure 1 shows the
chemical nomenclature, abbreviation and molecular structure of this
compound.
A C 1 SP-148 94 m Ob62949 0519908 9Lb m
Tech niaues
- -- Isothermal conduction calorimetry

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Conduction Calo rimetry

was used to measure the heat evolved during hydration as a function
of time. Most of the heat was released during the first 72 hrs of
hydration. Calorimetry work investigating the C3A reaction phase
was confined to 1 hr in duration because of the rapid heat liberation
characteristic of this compound. Studies involving the C3A and
gypsum systems usually produced heat up to 18 hrs after which the
reactions were very slow.
The calorimeter contained 6 cells mounted on a metal base

plate surrounded by foam insulation. A thermopile located under
each cell measured the heat production. Each cell contained a
Teflon-coated aluminum specimen holder into which a polyurethane
insert was placed holding the sample. An aluminum cover,
surrounded by a rubber O-ring, was securely fastened to the base
plate to keep the vessel watertight. The calorimetry unit was
manufactured by the Institute of Applied Physics, TNO-TU Delft,
Netherlands.
The calorimeter used for the C3A investigations was a single

cell apparatus. The design of this unit was similar to the description
above except for a syringe attachment which permitted injection of
the reaction fluids intp the sealed sample chamber. The reaction
vessel was also attachkd to an eight step vibrating mixer.
The calorimeters were placed in an isothermal water bath. The

voltage signals from each thermopile were recorded by a Datataker
DTl O0 Data Logger manufactured by Data Electronics Ltd., Australia.
The voltages were converted to calories using a formula that
considered sample weights and cell sensitivities. Data were
calculated as rate of hydration (cal/g/hr) and integral heat (cal/g).
Differential Scannina Calorimetry (DSC) -- DSC quantitatively

measures the heat evolved by a sample with respect to mass and
time as a function of temperature. It was used to determine the
hydration products formed after specified periods of cement paste
hydration. Samples of cement pastes hydrated for different periods
were subjected to heating at a constant rate. This caused the
hydrates present to decompose which produced thermal peaks. The
magnitude and temperature at which these peaks appeared provided
information on the identity of the hydrates.
The differential scanning calorimeter employed was a Dupont

Instruments, Model DSC 2910. It consisted of a furnace which
precisely heated an aluminum pan containing the sample. The DSC
A C 1 SP-148 9 4 M Obb294q 0 5 3 ~ 9 0 98 5 2 W
2910 controlled the heating of the sample and recorded data every
second.
Procedu re
Conduction Calorimetrv -- A 3.5 ml aqueous solution

containing the required amount of phosphonic acid was added to
10.000 g of cement at a water-cement ratio of 0.35. A 2.0 ml aqueous
solution containing the required amourit of phosphonic acid was
added to 2.000 g of C3A at a water-cement ratio of 1.0. The same
procedure was used for the C3A and 25% gypsum except the 2.000 g
sample contained 1.500 g (75%) of C3A and 0.500 g (25%) of
CaS04.2H20. Concentrations (percentage phosphonic acid by
weight of cement) ranging from 0.03% to 0.05% were used for ATMP.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Samples were mixed by hand for approximately 1 min and
placed in the calorimeter. The apparatus was sealed and then
submerged in a water bath maintained at 25.0f0.5°C. The C3A
calorimeter containing an unwetted sample was placed in a bath
along with the syringe of aqueous solution for an half hour to attain
temperature equilibrium. The reaction commenced by injecting the
solution and vibrating the chamber for 15 secs.
Averaged millivolt signals for each thermopile were collected

every 10 mins for up to 72 hrs by the Datataker. The Datataker
collected averaged millivolt signals from the thermopile between 2 to
30 secs for 2 to 18 hrs during the C3A tests.
Differential Scannina Calorimetrv -- The paste samples were

removed from the conduction calorimeter after specific elapsed time
intervals. Hydration of the samples was halted by vacuum-drying at
room temperature for a minimum of ten hours. The samples were
then ground to a fine powder and approximately 25.00 mg of a
sample was placed in the DSC. The samples were heated from room
temperature to 600°C at 15"C/min in a continuous flow of nitrogen
(100 ml/min). Data were recorded as the heat flow (W/g) of the
sample as a function of temperature (OC).
RESULTS AND DISCIJSSION
Fig. 2 to 5 represent the conduction calorimetric curves for the

hydration of four different normal portland Type I cements both in the
absence and presence of the phosphonic acid additive. The rate of
hydration (A) as well as the total amount of heat (B) developed up to
A C 1 SP-148 9i1 W 0 6 6 2 ï Y 9 05L99LO 57i1 m
136 Ramathandran and Lowery
72 hrs are shown. Generally, the reference samples (without

phosphonic acid) for the four cements exhibited a rate of hydration,
typical for normal portland Type I cement.
The following attributes were common to the rate curves for all
four cements (Fig. 2A to 5A). All exhibited an initial exotherm within
the first 10 mins of hydration and this is attributed to a combination of
reactions such as the hydration of free lime, heat of wetting and the
formation of ettringite of formula 3CaO(A1203,Fe203).3CaS04.31-32
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
H20. Only a small inflection was registered in the figures because

the cement was placed in the calorimeter a few minutes after it had
been mixed with a solution outside the calorimeter. This effect was
followed by a period of relatively low chemical reactivity from about 30
mins to 2.0 hrs, known as the 'induction' or 'dormant' period. At the
end of this period an increase in heat evolution occurred with a peak
at about 8.5-10.5 hrs and this is due to the hydration of tricalcium
silicate component of cement that yields calcium silicate hydrate and
calcium hydroxide. After this peak, there was a gradual diminution in
the evolution of heat, with the attainment of a steady state after about
50 hrs.
The exothermal peak resulting from the hydration of the C3S

component was observed between 8.5 and 10.5 hrs and had a
magnitude ranging between 3.33 to 4.05 cal/g/hr for the four
reference cements. The maximum elapsed time before the
occurrence of the peak of 10.5 hrs and the minimum magnitude of
3.33 cal/g/hr for the C3S peak, stated above, was recorded for the
Exshaw reference cement which contained the lowest amount of
C3S (Table 1).
The hydration of all the Type I cements was retarded with the
addition of the phosphonic acid. The duration of the retardation was
increased with higher dosages of the additive. Approximate
increases in the induction period ranging from 10.5 hrs for the low-
alkali Lafarge cement to 15.7 hrs for the high-alkali St. Laurent
cement were noted with the addition of 0.05% phosphonic acid.
The greater effect of the retarder on the St. Laurent cement

could be attributable to lower tricalcium aluminate levels. It has been
observed that the C3A phase consumes larger amounts of retarder
than the C3S phase during the hydration of Portland cement (13). In
cements with low C3A contents, lower amounts of retarder are
adsorbed and hence more retarder is available in the aqueous phase
to retard the hydration of the silicate phase. The St. Laurent cement
contained the lowest amount of C3A of the four cements. In addition,
the St. Laurent cement had the lowest Blaine surface area (Table 1)
of the four cements; therefore, it would be expected to exhibit a lower
degree of reactivity, at earlier times.
A C 1 SP-148 94 m Ob62949 0.539933 400 m
The second longest extension of the induction period occurred

with the Lafarge high-alkali cement. This cement contained higher
amounts of C3A than the St. Laurent cement, and a lower amount of
C3S and that would result in higher amounts of adsorption of the
retardant.
The Exshaw cement exhibited an extension of the induction

period of only 10.8 hrs with the addition of 0.05% phosphonic acid.
Conversely to the St. Laurent cement, the Exshaw brand contained
the highest amount of C3A compound and one with the highest
Blaine surface area values (Table 1).
The Lafarge low alkali cement was the least affected by the
retarder. This is contrary to some observations which have found that
the alkali in cement is capable of interacting with the retarder and
destroying its capability to retard hydration, implying the lower the
alkali, the better the retardation (4). Possibly, the moderately high
C3A and SO3 levels and high Blaine value diminished the retarder's
effectiveness enough to minimize the enhancement of retardation
normally associated with lower alkali levels.
Fig. 2B to 58 were obtained by integrating, at different times,

the curves in Fig. 2A to 5A. The total amount of heat developed at
different times may be used to estimate the degree of hydration of the
silicate phases in cement. The total heat generated by the hydration
of the Type I reference cements after 72 hrs ranged from 70.9 to 81.5
cal/g. Even after 3 days of hydration, the effects of the addition of
0.05% phosphonic acid was apparent. The 72 hr total heat values for
the four cements in the presence of 0.05% phosphonic acid were
72.7, 77.1, 65.4 and 69.0 cal/g for the St. Laurent high-alkali, Lafarge
high-alkali, Exshaw moderate-alkali and Lafarge low-alkali cements
respectively. These values correspond to 7.0, 5.4, 7.8 and 5.6%
reductions in the heat of hydration with respect to their reference
values.
The two largest reductions in the total heat of hydration

between the reference cement and samples treated with 0.05%
phosphonic acid were the St. Laurent cement (7.0%) and the Exshaw
cement (7.8%). It seems that Blaine fineness values and C3S, C3A
and alkali contents play a role.
Phosphonic acid affects the hydration of the C3S component of

cement (Fig. 6 ) . The reference C3S peak produced a maximum
exothermal heat rate of 5.32 cal/g/hr at 8.5 hrs. Addition of
phosphonic acid considerably reduced the maximum peak value,
delayed the peak occurrence and modified the peaks appearance. A
maximum peak value of only 1.79 cal/g/hr occurred at 11.5 hrs with
the addition of 0.03% phosphonic acid and the C3S exotherm was
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 74 IObb2949 05L99L2 347
transformed from a distinctive peak to a hump. A similar hydration

curve in appearance was produced with the addition of 0.04%
phosphonic acid and the exothermic maximum was 1.41 cal/g/hr and
occurred at 31.O hr. Addition of 0.05% phosphonate appears to totally
dampen the C3S peak for a duration greater than 72 hrs. It is
apparent that the retardation of phosphonic acid is related to a
chelating effect on the C-S-H and CH phases.
Treatment of C3A with phosphonic acid had a retarding effect

(Fig. 7). The reference C3A peak produced a maximum exothermal
value of 113.7 cal/g/hr at 3.6 mins. Addition of phosphonic acid
reduced the maximum peak value, delayed the peak occurrence and
modified the appearance of the peaks. A maximum peak value of
only 76.0 cal/g/hr occurred at 19.2 mins with the addition of O.O3O/o
phosphonic acid and the C3A exotherm was transformed from a
distinctive peak to a more rounded and broader based peak. A
similar hydration curve in appearance was produced with the addition
of 0.04% phosphonic acid, but the exothermic maximum was not
depressed. A value of 109.2 cal/g/hr occurred at 28.2 mins. The
hydration of C3A forms the hexagonal phases initially, which
subsequently are converted to the cubic phase. It has been observed
that some retarders enter the hexagonal phases and retard the
conversion effects (14). In the presence of phosphonic acid, in the
initial stages below 0.5 hrs, the heat values were lower for lower
concentrations of phosphonic acid. However, at longer times 0.03%
of acid was somewhat more efficient than the 0.04% concentration in
the retardation effects. No immediate explanation is obvious for this
anomalous effect.
Fig. 8 shows the effect of phosphonic acid on the calorimetric

curves of the C3A and gypsum system. The reference curve contains
three separate exotherms. The initial peak occurred almost
instantaneously at approximately 2 mins and generated 153 cal/g/hr.
Heat of wetting and hydration of some free lime present in C3A were
likely responsible for this initial heat evolution. A second peak of
approximately 75 calígíhr was observed at about 14 mins which was
produced by the formation of ettringite (high sulfoaluminate hydrate)
from the C3A phase reacting with gypsum. The third peak occurred
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
after about 1 hr and 15 mins and generated around 62 cal/g/hr (Fig.

8b). After all the gypsum converts to ettringite, the excess C3A
reacted with the ettringite to form a low sulfoaluminate hydrate and
this reaction could generate this peak. DSC was conducted on the
C3A and gypsum hydrates with and without phosphonic acid at
various times in order to determine the possible origin of these peaks.
A few DSC curves are shown in Fig. 9. The unhydrated C3A and
gypsum mixture exhibits an intense endothermal peak in the range
100-150OC caused by the dehydration of CaS04.2H20. Depending
on the conditions of heating, two stepwise hydration peaks are also
A C 1 SP-148 9 4 W 0662749 0519913 2 8 3 =
observed. Both phosphonic acid-treated and untreated samples

hydrated for 1 hr exhibit a second peak at about 13OOC that is caused
by the dehydration effect of ettringite. At 14 hrs of hydration two broad
endothermal peaks appear in all samples. The first peak at about
18OOC is caused by the low sulfoaluminate and the second peak at
about 300% by some cubic hydration product of C3A.
Addition of a phosphonic acid had an affect on all three

exothermic peaks. The initial peak was decreased in magnitude with
increasing concentrations of phosphonic acid (Fig 8a). The second
(ettringite) peak was also reduced in magnitude and the phosphonate
treated peaks were transformed to more of a shoulder or plateau
appearance. There was no significant shift in the time of occurrence
of the first or second peaks with the addition of phosphonic acid. The
third (low sulfoaluminate) peak for the ATMP treated samples
decreased in magnitude and delayed in time of occurrence with
respect to the reference. ATMP concentrations of 0.03 to 0.05%
reduced the reference peak from a magnitude of 62 cal/g/hr to
between 25 and 12 cal/g/hr respectively and shifted its occurrence
from 1 hr and 15 mins to between 5 to 10 hrs respectively. The
sequence of hydration reactions that occurs in the formation of
ettringite and low sulfoaluminate by the hydration of C3A and C3A
and gypsum systems is influenced by the adsorption of phosphonic
acid. The mechanism of retardation of the hydration in the C3A and
gypsum system is similar to that suggested for the same reactions in
the presence of superplasticizers (1).
CONCLUSIONS
Addition of aminotri(methy1ene phosphonic acid) (ATMP) results

in a substantial retardation of the hydration of cement, tricalcium
silicate, tricalcium aluminate and tricalcium aluminate-gypsum.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
ATMP appears to be a very effective retarder as evidenced by a

low dosage of 0.05% ATMP that increases the induction period
of Portland Type I cement by about 16 hrs.
The retardation effect was increased in cements with low C3A

contents suggesting that more phosphonic acid was available for
the retardation of the C3S component.
A high alkali cement was retarded more than the low alkali
cement in the presence of ATMP possibly because of chelating
effects of the phosphonic acid with the alkali.
A C 1 SP-148 9Li M O b b 2 9 4 9 0 5 1 9 9 1 4 L L T
5) The retardation effect of phosphonic acid on C3S was more

efficient and the peaks were less intense than in the cement. In
cement a substantial amount of the phosphonic acid is adsorbed
by the C3A component and lower amounts are available for the
effective retardation of C3S.
6) The hydration of C3A is retarded presumably by a large

adsorption effect. In the C3A and gypsum systems the rate of
hydration and interconversional effects are also retarded.
REFERENCES
1. Ramachandran, V. S., "Concrete Admixtures Handbook" (Noyes

Publications, New Jersey, 1984) pp. 626.
2. Ramachandran, V. S., Feldman, R. F. and Beaudoin, J. J.,

"Concrete Science", (Heyden, London, 1981) pp. 427.
3. Dodson, V., "Concrete Admixtures" (Van Norstrand Reinhold,

New York, 1990) pp. 21 1.
4. Rixom, M. R. and Mailvaganam, N. P., "Chemical Admixtures for

Concrete" (E. & F.N. Spon, London, 1986) pp. 306.
5. Brown, P. W. (Ed)., Cements Research Progress (Annual

Publication)' American Ceramic Society, Westerville, USA.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
6. Ramachandran, V. S., "Possible States of Chloride in the

Hydration of Tricalcium Silicate in the Presence of Calcium
Chloride", Materials and Construction,4 (1971) pp. 3-12.
7. Ramachandran, V. S.,"Admixture and Addition Interactions in

the Cement-Water System", Il Cemento, 83 (1986) pp. 13-38.
8. Young, J. F., "The Influence of Sugars on the Hydration of

Tricalcium Aluminate", Proceedings of V International Symp.
Chem. Cements, 1968, Tokyo, pp. 256-267.
9. Diamond, S., "Interactions Between Cement Minerals and

Hydroxycarboxylic Acid Retarders III, Infrared Spectral
Identification of the Aluminosilicate Complex", J. Amer. Ceram.
SOC.,5 5 , (1972) pp. 405-408.
10 Lieber, W., "Influence of Zinc Oxide on the Setting and

Hardening of Portland Cement", Zem. Kalk. Gips., 20, (1967) pp.
91-95.
A C 1 SP-348 9 4 W 0662949 0539935 056 W
1 1 . Nielson, E. B., "Well Cementing" (Elsevier, Amsterdam, 1990) pp.

496.
12. Ramachandran, V. S., Lowery, M. S., Wise, T. and Polomark,

G.M., "The Role of Phosphonates on the Hydration of Portland
Cement", Materials and Structures, 26 (1 993) pp. 425-432.
13. Ramachandran, V. S.,"Influence of Superplasticizers on the

Hydration of Cement", Third International Conference on
Polymers in Concrete, 1981, Koriyama, Japan, pp. 1071-1 081.
14. Ramachandran, V. S., "Adsorption and Hydration Behavior of

Tricalcium Aluminate-Water and Tricalcium Aluminate-Gypsum-
Water Systems in the Presence of Superplasticizers", J. Am.
C O ~ CInsf.,
~ . 80 (1983) pp. 235-241.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0 6 6 2 9 4 9 05L99Lb T92 =
TABLE 1 - CHEMICAL ANALYSIS AND PHYSICAL PROPERTIES OF TYPE I

CEMENTS AND CEMENT COMPONENTS
Cement Oxide Composition (%ì \ I
St. Laurent Lafarge Exshaw Lafarge

Hiah Alkali Hiah Alkali Moderate Alkali Low Alkali
si02 20.25 20.05 22.00 21.28

A1203 4.21 4.60 4.88 4.30
Ti02 0.21 0.20 0.20 0.22
p205 0.09 0.19 0.1 1 0.04
Fe203 3.33 2.47 1.93 2.97
Ca0 63.48 61.88 63.29 63.19
SrO 0.00 0.14 0.05 0.00
MgO 2.44 2.37 4.24 3.81
Na20 0.26 0.19 0.17 0.17
K20 0.90 0.94 0.78 0.41
SO3 3.1 1 3.79 1.95 2.64
LOI -
2.45 2.44 0.70 1.43
TOTAL 100.73 99.26 100.30 100.46
Phase Composition (%):

St. Laurent Lafarge Exshaw Lafarge
High High Moderate Low
Alkali Alkali Alkali Alkali
Tricalcium Silicate 62.61 54.29 49.35 54.85

G3S)
Dicalcium Silicate 10.82 16.53 25.85 19.63
(CZS)
Tricalciu m Aluminate 5.52 8.01 9.67 6.37
(C34
Tetracalci um Al urnino- 10.13 7.52 5.87 9.04
ferrite (C4AF)
TOTAL 89.08 86.35 90.74 89.89
Na20 equivalence 0.85 0.81 0.68 0.44

Blaine fineness 332 348 351 357
(m2/kg)
Tricalcium silicate - 326 m2/kg Tricalcium aluminate - 301 m2/kg.
* Cement nomenclature: C = Cao; S = Si02; A = Al2O3; F = Fe203
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 94 Ob62949 0539937 929 =
ATMP
Aminoîri(methy1enephosphonicacid)
Fig. 1-The molecular struciure, chemical nomenclature and

abbreviation of the phosphonic acid compound
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 I0 6 6 2 9 4 9 05L9918 ô65
Rate of Hydration
St. Laurent High Alkali
Portland Type I Cement
o 12 24 36 48 60 72
Time, hours
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 2a-Calorimetric curves showing the effect of a phosphonic acid on the rate of
hydration of high alkali portland Type I cement (WK = 0.35)
80
B .Total Heat of Hydration
70 Ft. Laurent High Alkali
o 12 24 36 48 60
Time, hours
Fig. 2b-Calorimetric curves showing the effect of a phosphonic acid on the total heat
of hydration of high alkali portland Type I cement (WK = 0.35)
A C 1 SP-LqA '34 m 0662949 0519919 7TL
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A 5
Rate of Hydration
, Reference
Lafarge High Alkali
O 12 24 36 48 60 72
Time, hours
hydration of high alkali portland Type I cement (WIC = 0.35)
90
Lafarge High Alkali

60
.
50
-
P .
o 40 .
*
30 *
20 -
10 '
n,
O 12 24 36 48 60 72
Time, hours
of hydration of high alkali portland Type I cement (W/C= 0.35)
A C 1 SP-148 94 m 0662749 0519920 413 m
146 Rarnachandran and Lowery
A 5
Rate of Hydration
4
Exshaw Moderate Alkali
L
E 3
-P
3 2
"O 12 24 36 48 60 72
Time, hours
hydration of moderate alkali portland Type I cement (W/C = 0.35)
80 -
' Total Heat of Hydration
70 *
, Exshaw Moderate Alkali
60 , Portland Type I Cement
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 12 24 36 48 60 !
Time, hours
of hydration of moderate alkali portland Type I cement (W/C= 0.35)
A C 1 SP-148 94 = 0662949 0 5 3 9 9 2 3 35T
Rate of Hydration
Lafarge Low Alkali
O 12 24 36 48 60 72
Time, hours
hydration of low alkali portland Type I cement (W/C = 0.35)
O 12 24 36 48 60 72
Time, hours
of hydration of low alkali portland Type I cement (WlC = 0.35)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 m Ob62949 0 5 3 9 9 2 2 296 W
148 Rarnachandran and Lowery
Rate of Hydration
Tricalcium Silicate
0.041 ATMP
1 0.05% ATMP
O
c / //--===--
W
O 12 24 36 48 60 72
Time, hours
Fig. 6a-Calorimetric curves showing the effect of a phosphonic acid on the rate of hydration
of the tricalcium silicate component of cement (W/C = 0.35)
80
B .
70 . Total Heat of Hydration
60
60 . Tricalcium
Tricalcium Silicate
Silicate
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 12 24 36 48 60 72
Time, hours
Fig. 6b-Calorimetric curves showing the efíed of a phosphonic acid on the total heat of hydration
of the tricalcium silicate component of cement (W/C = 0.35)
A C 1 SP-148 94 = Obb2947 0517723 1 2 2
130 - Rate of Hydration - Tricalcium Aluminate

A 120 :
. - . . . - - - - . - * * - - . - - .
o
O 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Time, hours
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 7a-Calorimetric curves showing the effect of a phosphonic acid on the rate of hydration
of the tricalcium aluminate component of cement (WK = 1.0)
Rate of Hydration 0.04% ATMP

Tricalcium Aluminate /
j:* 20
10
"O
-
4 ...
0.1 0.2
Fig. m-lalorimetric curves showing the effect of a phosphonic acid on the total heat of hydration
0.3 0.4 0.5
Time, hours
0.6 0.7 0.8 0.9
. .
1
of the tricalcium aluminate component of cement (W/C = 1.0)
A
140
> Rate of Hydration
120 Tricalcium Aluminate & Gypsum Paste
* k 10(I
P
3 E(I
o
6íI 0.03% ATMP
0.04% ATMP
20
o l * * - .
O 0.1 0.2
-
0.3
. * .
0.4
* . -
0.5
.
0.6
- -
0.7
* .
0.8
* -
0.9
* '
1
Time, hours
Fig. &-Calorimetric turves showing the effect of a phosphonic acid on the rate of hydration of the
tricalcium aluminate (75 percent) and hydrous calcium sulfate (25 percent) mixture (W/C = 1.0)
B 1601
1401 Rate of Hydration

Tricalcium Aluminate & Gypsum Paste
120 t
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
.-u I , ".".
~~
I 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Time, hours
Fig. Ob-Colorimetric turves showing the effect of a phosphonic acid on the rate of hydration of the
tricalcium aluminate (75 percent) and hydrous calcium sulfate (25 percent) mixture (W/C = 1.0)
A C 1 SP-14A 9 4 m Ob62949 0539925 T T 5 m
140
C
120
1O0
e
- 80
m
60
40
20 Tricalcium Aluminate & Gypsum Paste
O
O 1 2 3 4 5 6 7 8 9 1011 1 2 1 3 1 4 1 5 1 6 1 7 3
Time, hours
Fig. &-Calorimetric curves showing the effect of a phosphonic acid on the total heat of hydration
of a tricalcium aluminate (75 percent) and hydrous calcium sulfate (Upercent) mixture (W/C = 1.0)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
t
E
3
I
IF;
o 100 200
-
ñbimnncs 1 Hi. HyünaOn
300
Temperature, 'C
400 500
Y
I
s
L
8
t
o 100 200 300

Temperature, 'C
400 500
+
Fig. 9-D.S.C thermograms for tricalcium aluminate (Upercent) gypsum mixtures that were
a) unhydrated and hydrated for 1 hr; and b) hydrated for 14 hours
A C 1 SP-LYô 94 m O b 6 2 9 4 9 0519926 931 m
SP 148-9
The Influence of Cross-Linked and

NSF Superplasticizer on the Flow
Properties of Blended Cements
by T. Sone, S.L. Sarkar, and H. Uchikawa
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
SvnoDsis: The corporate use of mineral admixtures, such as slag, silica fume,
and fly ash, and superplasticizer in concrete is steadily rising for reasons of
economy, enhanced strength, low heat generation, increased durability and better
rheological control.
This contribution reports the results of the influence of a cross-linked and
a NSF type of superplasticizer on the flow properties of blended cements. The
cross-linked superplasticizer was constituted of polycarboxylic ether and cross-
linked polymer, whereas the NSF type was a modified lignin, alkylsulfonate and a
polymer. In view of the difference in their molecular structure, their effect was
studied on two types of cement, a normal portland cement, and the other a
moderate heat portland cement (belite rich, low in C3A) to which different
proportions of slag, silica fume, Class C and F fly ash were added to simulate
binary and ternary blended cement compositions.
Following a detailed chemical and mineralogical characterization of these
blending components, the slump flow of twenty five mortar blends were tested at a
sand:binder ratio of 1.5 with the superplasticizer dosage varying from 2.5 to 3%
by weight of cement. The waterhinder (W/B) of these mixtures ranged from 0.31
to 0.35.
Marked differences in flow characteristics (determined by different
methods) were recorded as a function of the cement type, blending component,
and superplasticizer composition. Viscometric measurements made on the
corresponding cement paste mixtures using a rheometer, also exhibit pronounced
differences in terms of their apparent viscosity. The possible superplasticizer
interactions that occur in these blended cementitious systems are discussed. This
study reiterates the cemenîhperplasticizer compatibility factor which is currently
under intense discussion among researchers.
Keywords: Admixtures; blended cements; flowability; slump; superplastitizers; viscosity

153 Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-148 9 4 E 0 6 6 2 9 4 9 0539927 878
154 Sone, Sarkar, and Uchikawa
T. Sone is associate general manager of cement and concrete laboratory, Onoda

Cement Co., Ltd. He received his master of engineering degree from Hosei
University in 1975. His recent research has been on the development of low-heat
type cement.
Shondeep L. Sarkar has been a faculty member in the Department of Civil
Engineering, Université de Sherbrooke, Canada, since 1986. He is a
microstructural scientist specializing in HPC, concrete durability/deterioration,
high early strength and blended cement, and waste utilization. He is the author of
many papers, and has edited a number of books and proceedings.
Hiroshi Uchikawa is the Executive Director in charge of Research and
Development of Onoda Cement Co., Ltd., Tokyo, Japan. He graduated from the
University of Tokyo in 1956 and earned his Ph.D. from the same university in
1967. He has devoted more than thirty five years for research on cement and
concrete from the standpoint of materials science. He has published more than 200
papers and several books, and won awards from the Ceramic Society of Japan and
Cement Association of Japan for his work in cement science. He has been also a
guest professor at the Nagaoka University of Technology since 1987. He is a
member of several scientific committees of the government.
INTRODUCTION
The use of high-range water reducing admixtures, or more commonly,
superplasticizers, in concrete is steadily rising. Not only does it provide better
rheological control of the concrete, given proper mixture proportioning and
materiais selection [ 1,2], its use can drastically reduce the waterlcement (W/C); this
yields a much denser paste matrix, stronger paste-aggregate bonding and lower
amount of Ca(OH)2 at a microstructural level. At the macro-level, its use is linked
to higher strength and impermeability, greater resistance to chemical and
environmental attack, which effectively reflects on its higher durability [3].
It is not surprising therefore, the constant interest shown by --```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
superplasticizer manufacturers and researchers alike to modify the composition of

existing superplasticizers, and introduce new types in order to achieve better flow
properties, improved stabilization and dispersability retention, and further
reduction in W/C.
Over the years, the incorporation of mineral admixtures, such as slag,
silica fume, or fly ash, either as a blending component in cement or additives in
concrete has become equally common. Different proportions of these components,
ranging from low (5% by weight) to as high as 70% by weight are often added in
conjunction with a superplasticizer for economic reasons, enhanced strength,
increased durability, and better rheology [4]. Numerous studies relating to the
advantageous use of admixtures in concrete have been reported to date [5,6]. As a
matter of interest, ternary-component blended cement is now used in field
applications for mass marine concreting in Japan [7].
This contribution reports the influence of a cross-linked and a NSF
superplasticizer on the flow properties of blended cements. Since the chemical
composition of these two superplasticizers is different, specially at the chain
structure level (Figure l), difference in the flow behaviour of cement was
A C 1 Sp-144 94 0662949 0519928 704 M
anticipated. This was investigated with reference to two cement types, a normal
portland cement (NPC), and a moderate heat portland cement (MPC) to which
different proportions of slag, silica fume, Class C and F fly ash were added.
It is well established that when a superplasticizer is added to a cement,
there is interaction with selective mineralogical phases present in the cement [8].
Therefore, two cement types with wide differences in mineralogical composition
were selected for amplifying this effect.
EXPERIMENTAL
Materials
The chemical composition of the two cements, slag, fly ashes, and silica
fume are given in Table 1. The Class C fly ash apart from being high (13.2%) in
Cao, is rich (6.5%) in Na20. The Fe203 content in the Class F fly ash, on the
other hand, is relatively high (23.8%). The slag is high (14.8%) in MgO, and its
Al203 content is lower than that of the fly ashes. The silica fume is composed
mostly of Sioz. The MPC represents a special kind of cement, normally used for
hydraulic constructions that require lower adiabatic temperature in concrete [9].
From the calculated Bogue composition given in this table, it is evident that MPC
is a belite-rich, low C3A cement.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
The Blaine fineness was 323 and 328 m%g, and density 3.16 and 3.20
for NPC and MPC respectively. The slag, silica fume, Class C and F fly ashes
originated from Canada, whereas the cements were Japanese. The specific surface
areas of the fly ashes were 348 m2/kg for the Class C and 3 18 m2/kg for the Class
F.
The cross-linked superplasticizer was a polycarboxylic ether and cross-
linked polymer (PCE) [lo], whereas the NSF type was a modified lignin
alkylsulfonate and a continuous activation polymer (LAS).
The LAS structure consists of alkylsulfonate of the Type RSO3- where R

is the organic complex group of high molecular weight, while the PCE comprised
an acrylic polymer of molecular weight 3000 as the main chain with polyethylene
glycol ether of varying molecular weights as grafted side chains. The PCE
superplasticizer was a 16.3% aqueous solution, whereas the LAS was 36.4%
aqueous. Other relevant physical and chemical properties of the two
superplasticizers are given in Table 2.
For mortars a river sand of fineness modulus 2.7 was chosen for its
particle size gradation.
Methods
A mini slump test cone measuring 10 cm base diameter, 5 cm apex
diameter, and 15 cm height, meeting the JIS-A1173 specification was used to
determine the flow characteristics of mortars made with a sand:binder ratio of 1.5.
Since slump test of high-flowability mixtures becomes questionable, the "flow
A C 1 SP-148 94 =
Ob62949 0539929 640 H
table" method was employed. The standard mixing procedure was carried out
(JIS-R5201). All the flow tests were performed at 28"C, RH 84%.
A rheometer was used to determine the rheological coefficient of
corresponding fresh cement paste mixtures. The mixing procedure was as
fol10ws:
Mixing by hand
small mixer -3
5 min
-+ Mixing in a small -+
measurement
Characterization of blending components
Besides particle size and shape, the mineralogical phase composition of

cement and blending components are known to affect the rheological behaviour of
grout, mortar, and concrete in the superplasticized state [ i i]. Therefore, before
performing the rheological tests, a careful characterization of the blending
components was carried out in order to predict their behaviour in blended cement.
The electron micrographs of the three blending agents are shown in
Figure 2. Compared to fly ash and silica fume, the slag particles are much more
angular; their morphology and particle size resembles that of coarse cement
particles. The TEM micrograph of silica fume confirms it to be ultra-fine. Both
the fly ashes comprised mostly circular to semicircular particles, some of which are
hollow. As compositional variations can not be accurately determined from
backscattered electron imaging, electron microprobe analysis (EMPA) at different
spots confirmed their chemical heterogeneity (Table 3), wherein the slag
composition was found to be fairly constant, and comparable to that of the XRF-
derived chemical analysis (Table 1).
From X-ray diffraction (XRD) analysis both the fly ashes were found
to contain quartz and mullite as common crystalline phases (Figure 3). Detectable
amount of free lime is present in Class C fly ash, whereas Class F fly ash contains
anhydrite, hematite and magnetite. All these mineralogical phases correspond well
with their chemical analysis (Table i). The slag contains a prominent hump
between 25 and 35"28, representing the glassy phase. The only crystalline phase
identified in it was melilite (same figure). Considering the hump between 20 and
30'28, Class C fly ash appears to be more glassy.
The condensation degree of silicate anion in slag and silica fume
determined by nuclear magnetic resonance with Fourier Transform (FT-NMR),
showed distinct differences, such as the characteristic QI dimer for slag, whereas
for silica fume it was the 4 4 structure (Figure 4). The condensation degree of
silicate anion in silica fume is unusually high because strictly speaking it is not
glass, but its amorphous phase is a product of gas-gas reaction.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0 6 6 2 9 4 9 0519930 362 =
Slump flow characteristics of mortars

In order to evaluate the cementlsuperplasticizer compatibility factor, the
slump flow of plain cement mortars was tested at WIC =0.31 (Table 4). In the
absence of blending agent(s), the influence of cement phase mineralogy on the
rheology becomes more conspicuous with PCE superplasticizer, especially when
one considers that the cement fineness is the same for both, and thus it plays no
effective role. Relatively high slump flow is obtained with NPC irrespective of the
type of superplasticizer used, although compared to PCE, the LAS superplasticizer
yields better slump flow. What is also interesting is that when the cement type is
changed to MPC, there is total loss of slump flow with PCE superplasticizer even
when its dosage is increased by 0.5% to 3.0%, whereas in the case of LAS
superplasticizer a small increase in slump flow was recorded, which can be related
to the lower C3A content in MPC, yielding higher flowability. On the other hand,
it is possible that a strong interaction between PCE and C4AF phase which is
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
significantly higher in MPC, impairs the dispersibility. Alternatively, assuming

that part of the C3A is in the orthorhombic form or alkali stabilized, this cement
does not provide the requisite amount of alkalis for the cross-links to cleave, which
is the mechanism responsible for its dispersive property.
This slump loss with PCE superplasticizer led to some additional tests
to study the effect of mixing time on the flow characteristics. The WIC was
increased to 0.35, and the mixtures were dosed with 3.0% PCE to a obtain better
4
flow. Since silica fume is known to affect the workability, a fixed amount (2.5%
by weight) was intentionally added to each of these mixtures. In general, the
flowability improves due to higher WIC. The flow increased by extending the
mixing time from the usual 60 to 120 sec. This is illustrated in Figure 5. One
notes that this effect is more prominent in NPC. Increasing the shear action no
doubt makes the defloculation process more effective, but associated with it are the
early chemical reactions of the cement brought about by its mineral constituents.
Although all the six mixtures contain the same amount of silica fume, cement phase
mineralogy appears to play a more dominant role as far as superplasticizer
compatibility is concerned.
To study the effect of silica fume replacement on the interaction of the
two superplasticizers, NPC was selected (to eliminate the influence of cement
phases which appears with MPC in the presence of PCE superplasticizer). The
results in Figure 6 clearly indicate that the slump flow is a direct function of the
silica fume replacement percentage in cement; as the silica fume content increases,
the slump flow decreases. This reduction, however, is not as much for LAS
superplasticizer, especially when one considers that a much lower W/C (0.32 vs
0.35) was used for LAS. The downward trend in slump flow is nearly parallel for
both the superplasticizers. Nevertheless, the lower workability with PCE even at a
higher W/C in mortars containing silica fume is suggestive of the fact that the
defloculating action of PCE is distinctly reduced in the presence of silica fume.
This may be due to the electrical charge property of the silica fume particles or their
ultrafine size which may hinder particulate dispersion.
The effect of the two superplasticizers was next studied on ternary-
component blended cements (Table 5 ). It is evident that when MPC is used in
conjunction with 27% Class F fly ash and 50% slag, the slump flow is high, in the
range of 338 to 347 mm at W/C =0.3 1. Changing the fly ash type to Class C does
not alter the flow properties much, except that they improve slightly. The same
A C 1 SP-146 94 0662747 0 5 3 9 9 3 3 2 T 9
applies for the mixtures containing Class F fly ash and NPC. However, when
Class C fly ash is used in combination with NPC, the flow decreases considerably
with the same dosage of LAS superplasticizer, but increases significantly with
PCE. This signifies that some of the constituents of NPC and Class C fly ash act
to the rheological advantage when PCE is used. NPC is high in C3A, whereas
Class C fly ash is high in alkalies (Table 1). Assuming that most of the alkalis
enter into a soluble state including that from the C3A, they facilitate the cleaving
action of PCE superplasticizer cross-linkage to improve the flowability. This is
further proved by the fact that this interaction does not appear when Class F fly ash
or MPC is used, when the slump flow in the presence of LAS is better.
The role of slag in all these mixes appears to be neutral as far as
rheology is concerned. Here it needs to be mentioned in passing that the rationale
for these replacement values of slag and fly ash in cement selected for this series
has been elaborated by Nagataki et al. [9].
Another series of ternary-component cements containing fly ash and
silica fume yield interesting slump flow results (Table 6). These mixtures contain
40% MPC or NPC, 50% Class C or F fly ash, and 10% silica fume; the W/C was
kept at 0.31 to 0.32, and the superplasticizer dosage ranged from 2.5 to 3.0%.
The results indicate that in the case of PCE superplasticizer, the slump flow is
severely affected when silica fume is introduced in the system, irrespective of the
type of cement used. As a matter fact, no slump flow occurred, only the slump '
loss could be recorded. This corroborates with the previous results presented in
Figure 6. The corresponding mixtures with LAS superplasticizer, however,
continue to demonstrate flow properties in the range of 312 to 368 mm. This is
illustrated in Figure 7.
By increasing the superplasticizer dosage by another 0.5%, thus
effectively increasing the W/C ratio to 0.32, and by changing the fly ash to Class
F, improvement in flow properties does occur, but the corresponding mixture with
LAS superplasticizer becomes still more flowable. Undoubtedly, increase in W/C
ratio is partly responsible for this higher flowability. These slump flow results
reiterate the influence of fly ash composition on the flow characteristics when PCE
superplasticizer is used.
Thus, it implies that in these ternary systems containing fly ash and
silica fume, the effect of the latter is much stronger in terms of reduced flowability
when PCE superplasticizer is used. It also demonstrates that compared to plain
cement mortar, certain ternary cement compositions yield much higher flow in the
presence of PCE superplasticizer, and this may be an effective way of improving
the rheology when PCE is to be used.
Viscosity measurements
In the rheological tests performed on neat and blended cement pastes,
the low yield values obtained (Table 7) can be attributed to the fact that a mortar
requires higher amount of water to obtain the same flowability as that of a cement
paste. In other words, the W/C of the pastes tested proved to be too high, with the
exception of a few test results.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 9Y Ob62949 0539932 3 3 5 D
In Table 7, the apparent viscosity of NPC is much higher compared to

MPC, but the type of superplasticizer does not appear to affect the apparent:
viscosity. The initial chemical reaction due to the higher C3A content of NPC may
be responsible for the increase in its apparent viscosity. Although no slump flow
was obtained with MPC when PCE superplasticizer was added (Table 4), high
shear rate results in reducing its apparent viscosity. Alternatively, the steric
hindrance property of PCE comes into play, resulting in particulate dispersion after
a certain period of time. The high yield value of NPC cement with LAS
superplasticizer is difficult to explain within this framework of experiments.
The yield value and apparent viscosity increase as a function of silica
fume replacement in NPC, irrespective of the type of superplasticizer (Figure S),
although in general the rheological coefficient is higher in pastes with PCE.
Similar trend in slump flow reduction of corresponding mortars was also observed
(Figure 6). Figure 8 shows that the apparent viscosity is minimum at 2.5% (by
weight) replacement of silica fume dosed with 2.5% LAS. This can be due to a
better gradation in the particle size distribution at this replacement value, thus
yielding higher flowability.
The apparent viscosity of ternary component pastes containing Class F
fly is higher, irrespective of the cement type (Figure 9). This effect is more
conspicuous in the case of LAS superplasticizer. The same figure shows that for
MPC containing Class C or F fly ash, lower apparent viscosity can be obtained by
using PCE superplasticizer.
The relationship between mortar slump flow and apparent viscosity of
ternary cements is illustrated in Figure 10. The pastes with LAS superplasticizer
demonstrate higher apparent viscosity, and therefore greater resistance to material
segregation. Also, the apparent viscosity of cement containing Class F fly ash is
higher than for Class C fly ash. The two distinct clusters in this figure show the
influence of cement and superplasticizer type on the rheological behaviour. No
direct correlation, however, can be drawn between the apparent viscosity and
slump flow.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Function of superplasticizers
The distinct rheological differences observed in the different blended
cements also depend on the functional mode of the two superplasticizers. One
needs to consider electrical repulsion versus steric repulsion for a better
understanding of this phenomenon.
In LAS, electrical repulsion enlarges the zeta potential of the cement
surface, whereas PCE increases the repulsive force by steric expansion of the
adsorbed layer. LAS is adsorbed in the shape of layered rigid rods in which the
cement particles are dispersed due to the strong electrical repulsion of the
negatively charged ions in the sulfonate group. In the cross-linked PCE polymer,
electrical repulsion of the negatively charged carboxyl group is accompanied with
steric repulsion of the main and graft chains [12]. That steric hindrance promotes
additional fluidity was also advocated by Kreijger [ 131. Polycarboxylic ether
resembles comb or side chains consisting of ether bonds. The oxygen in these
bonds and water molecules form strong hydrogen bonds. This results in a thick
A C 1 SP-14ô 94 Obb2949 0539933 071
hydrophylic protective layer. It is this layer that is considered to contribute to the

dispersion stability [3,10].
Gel permeation chromatography (GPC) and I3C-NMR results [ 101

demonstrate that the ester bond (-CO-O-) at the cross-linkage point is hydrolyzed or
cleaved by the available alkalis in the cement, resulting in an additional water
reducing agent being supplied to the hydrating cement mixture; this cleaving action
leads to the formation of polycarboxylic acid. This would explain the superior
flow properties of NPC in the presence of Class C fly ash (both richer in alkalis).
The adsorption mechanism of LAS and PCE superplasticizers is shown
schematically in Figure 11.
Since LAS is adsorbed in layers on cement particles, this layer is soon
covered with cement hydrates which reduces the effect of electrical repulsion,
whereas due to steric hindrance the dispersibility is retained longer with PCE.
That some mixtures despite low slump flow demonstrate low apparent viscosity
can be related to this phenomenon.
CONCLUSIONS
This study focused on the rheological behaviour of blended cements
using two types of cement (normal and moderate heat portland cement), containing
slag, silica fume, and Class F and C fly ashes as blending components in different
proportions, and two superplasticizers with entirely different types of functional
organo molecular structures.
Mortar slump flow test results and rheological coefficients of pastes at
relatively low W/C (0.31 to 0.35) and high superplasticizer dosage (2.5 to 3%)
address the issue of the complexity of establishing the cementitious/superplasticizer
compatibility factor. While one superplasticizer may demonstrate superior
rheological properties in a particular neat, binary or ternary component cement, it
may not perform equally well if one of the components is substituted by another,
for example, Class F fly ash by Class C fly ash, or normal portland cement by a
belite rich, low C3A cement, such as MPC.
The overall chemical composition, and phase mineralogy of the
cementitious components play determinative roles as far as flow properties are
concerned. This is compounded by the difference in the dispersive mechanism of
the superplasticizers.
ACKNOWLEDGEMENTS
The authors would like to thank Dr. S. Hanehara, Messers T. Okamura,
Y. Okada, H. Hirao, D. Sawaki and T. Shirasaka of Onoda Cement Co. for their
assistance. Dr. Shondeep L. Sarkar gratefully acknowledges the financial support
from Japan Science and Technology Fund for this collaborative work. Thanks are
also due to Dr. Mohan V. Malhotra, CANMET for providing the blending agents.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 94 W 0662949 0519934 T O B =
REFERENCES
1. Sarkar, S.L. and Baalbaki, M. "The Influence of the Type of Cement on
the Properties and Microstructure of High-performance Concrete",
Proceedings of the 9th International Congress on the Chemistry of Cement,
New Delhi, 1992, Vol. IV, pp. 89-94.
2. Uchikawa, H., Hanehara, S., Shirasaka, T. and Sawaki, D. "Effect on
Admixture on Hydration of Cement, Adsorptive Behavior of Admixture
and Fluidity and Setting of Fresh Cement Paste", Cement and Concrete
Research, Vol. 22, 1992, pp. 1115-1129.
3. Uchikawa, H. "Durability of High Strength Concrete with Superior
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Workability Estimated from the Composition and Structure", P.K. Mehta
Symposium on Durability of Concrete, Nice, K.H. Khayat, ed., 1994, to
be published.
4. Mehta, P.K. "Pozzolanic and Cementitious By-products in Concrete",
Proceedings of the Third International Conference on Fly Ash, Silica
Fume, Slag and Natural Pozzolans in Concrete, Trondheim, V.M.
Malhotra, ed., SP- 114, American Concrete Institute, Detroit, 1989,
pp. 1-44.
5. Larbi, J.A. and Bijen, J.M. "Effect of Mineral Admixtures on the Cement
Paste-Aggregate Interface", Proceedings of the Fourth International
Conference on Fly Ash, Silica Fume, Slag and Natural Pozzolans in
Concrete, Istanbul, V.M. Malhotra, ed., SP-142, American Concrete
Institute, Detroit, 1992, pp. 655-670.
6. Roper, H., Kirby, G.G. and Baweja, D. "Long-term Durability of
Blended Cements Concretes in Structures", Proceedings of the Second
International Conference on Fly Ash, Silica Fume, Slag and Natural
Pozzolans in Concrete, Madrid, V.M. Malhotra, ed., SP-991, American
Concrete Institute, Detroit, 1986, pp. 463-482.
7. Uchikawa, H. and Okamura, T. "Binary and Ternary Components
Blended Cements", in Mineral Admixtures in Cement and Concrete, S.L.
Sarkar, ed., ABI Books Ltd., N. Delhi, 1993, in press.
8. Ramachandran, V.S. "Use of Superplasticizers in Concrete", I1 Cemento,
Vol. 84(3), 1987, pp. 273-298.
9. Nagataki, S . , Sone, T. and Matsui, A. "Properties of Concrete Using
Newly Developed Low-Heat Cements and Experiments with Mass
Concrete Models", Proceedings of the Fourth International Conference on
Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete, Istanbul,
V.M. Malhotra, ed., SP-132, American Concrete Institute, Detroit, 1992,
pp. 1413-1432.
10. Tanaka, Y.O. and Okazawa, S . "A New Polycarboxylate Based Polymer:
Chemistry and Dispersing Performance", Proceedings of the International
Conference: Concrete 2000, Dundee, September 1993.
A C 1 SP-148 94 0662949 0539935 9 4 4
1 1. Uchikawa, H. "Similarities and Discrepancies of Hardened Cement Paste,

Mortar and Concrete from the Standpoints of Composition and Structure",
in Advances in Cement Manufacture and Use, E. Gartner, ed., Engineering
Foundation, New York, 1989, pp. 271-194.
12. Tanaka, Y. and Ohta, A. "Chemical Structure and Action Mechanism of
Superplasticizers", NMB Transaction, No. 9, Japan, 1992, pp. 5-1 1.
13. Kreijger, P.C. "Plasticizers and Dispersing Admixtures", in Admixtures,
The Construction Press, Lancaster, 1980, pp. 1-16.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 W 0662949 0539936 ô80 W
I I I I
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L48 94 m 0662949 0539937 717 m
7 7
9
O
9
O
7 c?
O O
Ln
9
w #
o a
a 1
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 74 M 0hh2947 0 5 3 9 9 3 8 6.53
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
166
A C 1 SP-148 94
Sone, Sarkar, and Uchikawa
Obb2447 05L9737 5 9 T =
1; ...... ..
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
c
H
O
z
A C 1 SP-L4ô 94 0 b b 2 9 4 9 0537940 2 0 1
.. 1 ..
g g
1
Li LI ILI
O 0 0
T
ö
z
-IN
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 W 0 6 6 2 7 4 7 0519941 148 W
I
ö
z
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0539942 084
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 O b 6 2 9 4 9 0539943 T L 0 D
R R
SO& SO$ R = H, CH3 C2H5
@)
NSF Cross-linked PCE
Fig. 1Uolecular structure of (a) NSF and (b) cross-linked polycarboxylate superplastitizen
Fig. 2-showing (a) slag particles under the SEM (b) TEM micrograph of silica fume particles,
(c) BEI of Class F and (d) Class C fly ash
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 =Superplasticizers
06b2949 0539944 957
and Chemical Admixtures 171
I 28 31
Fig. L X R D traces of (a) Class C, (b) Class F fly ash, and (c) slag, where 1 = quartz,
2 = mullite, 3 = free lime, 4 = anhydrite, 5 = hematite, 6 = magnetite, and 7 = melilite
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. d - N M R of (a) slag showing Q, dimer and (b) silica fume Q, structure
A C 1 SP-148 94 Ob62949 0 5 3 9 9 4 5 893
172 Sone, Sarkar, and Uthikawa
Mixing time (sec.)
Fig. &Effect of mixing time on the slump flow of NPC and MPC, using PCE superplasticizer
Fig. &Slump flow venus silica fume replacement in NPC
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662749 0.537946 7 2 T
300 (2.5)
Fig. 7-Slump flow of ternary-component cemenh using W superplasticizer
240
i
k 200
E
Y
<d
p.
o-
.e
160
i
.-
E 120
5n
å
80
O 2.5 5.0 7.5 10.0
Replacement of silica fume (96)
Fig. Wnfluence of silica fume replacement on apparent viscosity of pastes
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 0662749 0537747 6 6 6
Class F
Fly ash
Fig. M p p a r e n t viscosity of ternary-component cements as a function of fly ash

Class when PCE and LAS superplasticizers are used
? C : FA' : Slan ='23 : 27 :'50 1 I
d
p
Y
E
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
100 250 300 350 400 450 500

Slump flow (mm)
Fig. lû-Relotionship between slump flow and apparent viscosity
A C 1 SP-348 94 Ob62949 0539748 5T2 W
Coagulated-
o cement
particles
J.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
NSF eg PCE
Fig. 11-The adsorption mechanism of W and cross-linked PCE superplasticizen, aher [lo]
A C 1 SP-L48 94 0662947 0539749 439
SP 148-10
A Study on the Flow of Highly

Superplasticized Concrete
by S. Nishibayashi, S. houe, A. Yoshino,
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
T. Kuroda, and T. Kume
S y n o p s i s : The aim of t h i s s t u d y i s t o estimate t h e w o r k a b i l i t y of

h i g h l y s u p e r p l a s t i c i z e d c o n c r e t e , which c a n s a t i s f a c t o r i l y f i l l
up a l l t h e c o r n e r s o f t h e f o r m s i n a r e i n f o r c e d c o n c r e t e w i t h o u t
using a mechanical compactor. I n order t o place c o n c r e t e i n t o
r e i n f o r c e d c o n c r e t e members, t h e c o n c r e t e must h a v e s e g r e g a t i o n
r e s i s t a n c e a n d h i g h f l u i d i t y . T h e s e p r o p e r t i e s a r e o b t a i n e d by
u s i n g v e r y f i n e powders w i t h a n AE t y p e h i g h r a n g e water r e d u c i n g
admixture.
The h i g h l y s u p e r p l a s t i c i z e d c o n c r e t e w a s p r e p a r e d by m i x i n g
b l a s t - f u r n a c e s l a g , l i m e s t o n e powder o r s i l i c a fume w i t h o r d i n a r y
p o r t l a n d cement. The f l u i d i t y w a s measured by t h e box f l o w t e s t
( w i t h and w i t h o u t t h e r e i n f o r c e m e n t ) and by r h e o l o g i c a l tests o f
w e t - s c r e e n i n g m o r t a r . The e f f e c t s of m i x t u r e p r o p o r t i o n s and t h e
s p a c i n g o f r e i n f o r c e m e n t s on t h e f l u i d i t y of h i g h l y s u p e r -
p l a s t i c i z e d c o n c r e t e h a v e been d e t e r m i n e d .
Kewords: Aggregates; blast furnace slag; flowability; limestone; reinforced concrete; rheological
properties; segregation; silica fume; superplasticizers; tests; water reducing agents; workability
A C 1 SP-148 9 4 W Obb29r19 0539950 150 =
178 Nishibayashi et (il
S. N i s h i b a y a s h i i s a p r o f e s s o r o f C i v i l E n g i n e e r i n g a t T o t t o r i
U n i v e r s i t y , T o t t o r i , J a p a n , a l s o a member o f AC1
S. I n o u e i s a n a s s o c i a t e p r o f e s s o r o f C i v i l E n g i n e e r i n g a t
T o t t o r i U n i v e r s i t y , T o t t o r i , Japan
A. Y o s h i n o a n d T. K u r o d a a r e r e s e a r c h a s s i s t a n t s o f C i v i l
Engineering a t T o t t o r i University, T o t t o r i , Japan
T. Kume is a member o f t h e t e c h n i c a l d e p a r t m e n t s t a f f of Muramoto

C o n s t r u c t i o n Co.,Ltd., Osaka, J a p a n
INTRODUCTION
F o r t h e p u r p o s e o f i m p r o v i n g t h e w o r k a b i l i t y of c o n c r e t e ,
h i g h l y s u p e r p l a s t i c i z e d c o n c r e t e which can s a t i s f a c t o r i l y f i l l
f o r m s w i t h o u t u s i n g a m e c h a n i c a l c o m p a c t o r h a s been s t u d i e d . To
p l a c e c o n c r e t e i n r e i n f o r c e d c o n c r e t e members, t h e h i g h
f l o w a b i l i t y c o n c r e t e must h a v e s e g r e g a t i o n r e s i s t a n c e a n d h i g h
f l u i d i t y . C o n s e q u e n t l y t h e c o n c r e t e m i x t u r e may r e q u i r e a c e m e n t
r e p l a c e m e n t and a n a d d i t i o n of v i s c o s i t y - i n c r e a s i n g a d m i x t u r e t o
e l e v a t e h i g h f l u i d i t y and p r e v e n t s e g r e g a t i o n . However, t h e f l o w
b e h a v i o r c a n n o t b e e s t i m a t e d o n l y f r o m s l u m p and s l u m p f l o w
v a l u e , b e c a u s e it which t h e h i g h l y s u p e r p l a s t i c i z e d c o n c r e t e
p o s s e s s e s of is d i f f e r e n t from t h a t o f n o r m a l c o n c r e t e .
T h i s i n v e s t i g a t i o n w a s c a r r i e d o u t to study t h e e f f e c t of
materials and of m i x t u r e p r o p o r t i o n s on t h e f l o w b e h a v i o r of
h i g h l y s u p e r p l a s t i c i z e d c o n c r e t e . F i r s t , c o n c r e t e s were p r e p a r e d
by u s i n g b l a s t - f u r n a c e s l a g , t h r e e k i n d s o f l i m e s t o n e p o w d e r s and
t w o k i n d s of s i l i c a f u m e s , and a l s o t h e i r e f f e c t s on f l u i d i t y of
c o n c r e t e were s t u d i e d . S e c o n d l y , t h e r h e o l o g i c a l c o n s t a n t s o f
m o r t a r , t h e amount of c o a r s e a g g r e g a t e and t h e s p a c e between
r e i n f o r c e m e n t s were s e l e c t e d a s t e s t v a r i a b l e s . T h e e f f e c t o f
t h e s e f a c t o r s on t h e f l o w b e h a v i o r was s t u d i e d by t h e box f l o w
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
tests.
EXPERIMENTAL
MATERIALS
Ordinary p o r t l a n d cement, blast-furnace s l a g , t h r e e k i n d s of

l i m e s t o n e powders and t w o k i n d s of s i l i c a fumes were u s e d i n t h i s
series of e x p e r i m e n t s . The p h y s i c a l p r o p e r t i e s and t h e c h e m i c a l
c o n s t i t u e n t s o f t h e s e materials are shown i n T a b l e s 1 and 2. The
A C 1 SP-144 94 = Ob62949 0539953 0 9 7 =
c o a r s e a g g r e g a t e was c r u s h e d s t o n e . The f i n e a g g r e g a t e was a

m i x t u r e o f c r u s h e d and n a t u r a l sand. The p h y s i c a l p r o p e r t i e s of
t h e a g g r e g a t e s u s e d a r e g i v e n i n T a b l e s 3 a n d 4 . Two k i n d s o f AE
t y p e h i g h r a n g e water r e d u c i n g a g e n t s were u s e d . T h e m a i n com-
p o n e n t s are g i v e n i n T a b l e 5.
STUDY OF EFFECT OF TIPE OF POWDER
The water c o n t e n t , 170 kg/m3, t h e s l u m p f l o w , 6 0 0 t 5 0 m m and

a s / a p e r c e n t a g e , 49% were c o n s t a n t l y k e p t . The l i m e s t o n e powder
c o n t e n t was O, 30 a n d 50%, a n d t h e r a t i o o f water t o b i n d i n g
m a t e r i a l r a t i o (W/P) was 0.34. The s i l i c a f u m e was m i x e d w i t h
' o r d i n a r y p o r t l a n d c e m e n t and b l a s t f u r n a c e s l a g , and t h e s i l i c a
f u m e c o n t e n t w a s O , 8 a n d 15%. T a b l e 6 s h o w s t h e m i x t u r e p r o -
p o r t i o n . The f l u i d i t y o f t h e c o n c r e t e w a s m e a s u r e d by m e a n s o f
u s i n g t h e L-type f l o w test (Fig.1). I n t h i s t e s t , t h e L-type
f l o w v a l u e (LF) a n d t h e f l o w v e l o c i t y (VL) i n t h e r a n g e o f O t o
200mm from t h e e x i t of t h e a p p a r a t u s were measured.
The r h e o l o g i c a l c o n s t a n t s ( y i e l d v a l u e a n d p l a s t i c
v i s c o s i t y ) of t h e w e t - s c r e e n i n g m o r t a r were measured by u s i n g a
pull-up v i s c o m e t e r (1).
STUDY OF FLOWABILITY OF CONCRETE BETWEEN REINFORCEMENTS
The t e s t i n g f a c t o r s w h i c h a f f e c t t h e f l o w a b i l i t y o f t h e
h i g h l y s u p e r p l a s t i c i z e d c o n c r e t e between t h e r e i n f o r c e m e n t s are
as follows:
1) r h e o l o g i c a l c o n s t a n t s of m o r t a r
2) coarse aggregate content
3) s p a c i n g between r e i n f o r c e m e n t s
The e f f e c t s of t h e s e f a c t o r s on t h e f l o w a b i l i t y of c o n c r e t e
were a l s o d e t e r m i n e d by t h e L - t y p e f l o w t e s t a n d t h e t e s t of
f l o w a b i l i t y b e t w e e n r e i n f o r c e m e n t s (Fig.2). The r e i n f o r c e m e n t s i n
Fig.2 were deformed s t e e l b a r s (D13mm), and t h e s p a c i n g b e t w e e n
r e i n f o r c e m e n t s w a s v a r i e d as f o l l o w s :
1) 27 mm ( 4 / 3 times maximum s i z e of a g g r e g a t e )
2) 30 mm ( 3 / 2 times maximum s i z e o f a g g r e g a t e )
3) 37 mm ( 9 / 5 times maximum s i z e o f a g g r e g a t e )
The method on t e s t i n g t h e f l o w a b i l i t y b e t w e e n r e i n f o r c e m e n t s
was t h e same a s t h e L - t y p e f l o w t e s t . A d d i t i o n a l l y , a l i t e r o f
c o n c r e t e i n f r o n t of and b e h i n d r e i n f o r c e m e n t was sampled a f t e r
t h e test. These s a m p l e s were t h e n used f o r t h e washing a n a l y s i s
t e s t and t h e u n i t w e i g h t o f c o a r s e a g g r e g a t e w a s c a l c u l a t e d .
The s l a g c o n t e n t was 50%) a n d t h e c o n c r e t e m i x t u r e p r o -
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0539952 T23
180 Nishibayashi et al
p o r t i o n s were as shown i n t h e f o l l o w i n g t w o series.

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
1) I n t h e f i r s t series, t h e water c o n t y t was 1 7 kg/m3 and

t h e c o a r s e a g g r e g a t e c o n t e n t w a s 850 kg/m (316 L/m ). W/P was 9
0.32, 0.34 a n d 0.36. T h e d o s a g e o f AE t y p e h i g h r a n g e water
r e d u c i n g a g e n t was s e l e c t e d so t h a t t h e s l u m p f l o w o f c o n c r e t e
w a s 600t50 mm.
2 ) I n a n o t h e r s e r i e s , t h e w a t e r c o n t e t w a s 1 7 0 kg/m3 a n d
t h e b i n d i n g m a t e r i a l c o n t e n t was 500 kg/m (W/P = 0.34). The 9
s a n d l a g g r e g a t e p e r c e n t a g e ( s / a ) was 4 6 , 4 9 , 5 2 a n d 5 5 % ,
r e s p e c t i v e l y . The d o s a g e o f AE t y p e h i g h r a n g e w a t e r r e d u c i n g
a g e n t w a s s e l e c t e d so t h a t t h e r h e o l o g i c a l c o n s t a n t s o f w e t -
s c r e e n i n g m o r t a r was t h e same as t h o s e o f t h e c o n c r e t e w i t h slump
f l o w of 600+50 m m and s/a p e r c e n t a g e o f 49%.
T a b l e 7 s h o w s t h e m i x t u r e p r o p o r t i o n s w h i c h were d e c i d e d
f r o m t h e above c o n d i t i o n s .
TEST RESULTS AND DISCUSSIONS
EFFECT OF TYPE OF POWDER
Fig.3 shows t h e e f f e c t o f l i m e s t o n e powder c o n t e n t on t h e L-

t y p e f l o w v e l o c i t y ( V L ) and t h e p l a s t i c viscosity(P.V.) of w e t -
s c r e e n i n g m o r t a r when W/P was 0.34. I n t h e case o f r e p l a c e m e n t of
ce e n t by b l a s t - f u r n a c e s l a g ( w i t h B l a i n e f i n e n e s s v a l u e of 7000
Y
cm /g), t h e f l u i d i t y o f c o n c r e t e d e c r e a s e d , though t h e e f f e c t o f
s l a g on f l u i d i t p o f c o n c r e t e was c o n s t a n t w i t h i n t h e r a n g e o f
4000 t o 6000 c m /g(2). I n t h e case of r e p l a c e m e n t w i t h l i m e s t o n e
p o w d e r , t h e f l u i d i t y o f c o n c r e t e u s i n g LS3 w i t h t h e h i g h e s t
B l a i n e f i n e n e s s v a l u e was t h e h i g h e s t among t h e t h r e e k i n d s o f
l i m e s t o n e powders used. Though t h e L-type f l o w v e l o c i t y and t h e
p l a s t i c v i s c o s i t y o f m o r t a r were c o n s t a n t i n s p i t e o f t h e
r e p l a c e m e n t p e r c e n t a g e o f LS1 a n d LS2, f o r t h e c o n c r e t e u s i n g
LS3, t h e L-type f l o w v e l o c i t y i n c r e a s e d and t h e p l a s t i c v i s c o s i t y
of t h e mortar decreased w i t h i n c r e a s e i n t h e replacement
p e r c e n t a g e o f LS3.
To u n d e r s t a n d why t h e f l u i d i t y o f c o n c r e t e c o n t a i n i n g LS3
was t h e h i g h e s t , t h e p a r t i c l e s i z e d i s t r i b u t i o n o f t h e powder was
d e t e r m i n e d . The r e s u l t s are shown i n Fig.4.
From F i g . 4 , t h e r a n g e of s i z e o f LS3 p a r t i c l e s was w i d e r

t h a n t h a t o f LS1 and LS2. T h e r e f o r e , i t i s t h o u g h t t h a t t h e s o l i d
v o l u m e p e r c e n t a g e o f t h e c e m e n t p o w d e r w h i c h c o n t a i n e d LS3
l i m e s t o n e powder was l a r g e s t . The p l a s t i c v i s c o s i t y of t h e p a s t e
d e c r e a s e d d u e t o t h e e f f e c t of s o l i d v o l u m e p e r c e n t a g e o f t h e
powder and t h e f l u i d i t y of c o n c r e t e i n c r e a s e d .
Fig.5 shows t h e e f f e c t o f s i l i c a fume c o n t e n t on t h e L-type
A C 1 SP-148 74 m Ohb2947 0517953 9bT
flow v e l o c i t y and t h e p l a s t i c v i s c o s i t y of w e t - s c r e e n i n g m o r t a r
when W/P was 0.34. T h e L - t y p e f l o w v e l o c i t y i n c r e a s e d a n d t h e
p l a s t i c v i s c o s i t y of t h e m o r t a r d e c r e a s e d w i t h a n i n c r e a s e i n t h e
r e p l a c e m e n t p e r c e n t a g e of s i l i c a fume. T h e r e f o r e , i t i s s u g g e s t e d
t h a t t h e f l u i d i t y of highly superplasticized concrete increased
by t h e b a l l b e a r i n g e f f e c t of s i l i c a fume(3).
FLOWABILITY OF CONCRETE BETWEEN REINFORCEMENTS
Fig.6 shows r e l a t i o n s h i p s between W/P and t h e r e s u l t s o f L-

t y p e f l o w t e s t , as w e l l as between W/P and t h e r h e o l o g i c a l con-
s t a n t s f o r t h e wet-screening mortar. I n t h i s experiment, t h e
s l u m p f l o w was 600+50 m m , and t h e L-type f l o w v a l u e (LF) and t h e
y i e l d v a l u e (Y.V.) o f t h e m o r t a r were i n d e p e n d e n t o n W/P. T h e
f l o w v e l o c i t y ( V ) a n d t h e p l a s t i c v i s c o s i t y (P.V.) i n c r e a s e d
with decreasing Wb.
Therefore, t h e v i s c o s i t y of high f l o w a b i l i t y
c o n c r e t e a p p e a r s t o b e c o n s i d e r a b l y a f f e c t e d by t h e m i x t u r e pro-
p o r t i o n s , though t h e slump flow i s constant.
Fig.7 g i v e s t h e r e s u l t s o f t h e f l o w a b i l i t y t e s t between
r e i n f o r c e m e n t s . LFs/LF means t h e r a t i o of t h e f l o w v a l u e from t h e
f l o w a b i l i t y t e s t t o t h e L-type f l o w v a l u e . I n t h i s t e s t , t h e f l o w
v e l o c i t y o b t a i n e d w a s l o w e r t h a n L-type f l o w v e l o c i t y b e c a u s e of
t h e p r e s e n c e of t h e r e i n f o r c e m e n t and t h e l o w e r i n e r t i a f o r c e .
The f l o w v a l u e was a l s o smaller t h a n t h e L-type f l o w v a l u e . The
b r o k e n l i n e i n t h i s f i g u r e (LFs/LF=0.75) i n d i c a t e s t h e b l o c k i n g
c r i t e r i o n w h i c h was j u d g e d v i s u a l l y . T h e c o n c r e t e s a f f e c t e d by
t h e b l o c k i n g due t o t h e c o a r s e a g g r e g a t e a p p e a r below t h i s broken
l i n e . G s i n F i g . 8 w a s g i v e n by
GS = (Gl-GZ)/(Gl+GZ) x 100 (1)
where G 1 i s t h e u n i t weight of coarse aggregate i n f r o n t of t h e

r e i n f o r c e m e n t a n d G2 i s t h e v a l u e behind t h e r e i n f o r c e m e n t .
I n Fig.7, when t h e s p a c i n g between r e i n f o r c e m e n t s was 37 mrn,

t h e c o n c r e t e f l o w e d f r e e l y w i t h o u t b e i n g b l o c k e d by t h e c o a r s e
a g g r e g a t e . However, when t h e s p a c i n g between r e i n f o r c e m e n t s was
r e d u c e d t o 27 o r 30 m m , t h e c o n c r e t e was b l o c k e d by t h e o a r s e
a g g r e g a t e . T h e r e f o r e , when t h e mass of a g g r e g a t e i s 300 L/m , t h e 5
m a i n c a u s e o f b l o c k i n g is t h e s p a c i n g b e t w e e n r e i n f o r c e m e n t s
r a t h e r t h a n t h e p l a s t i c v i s c o s i t y of t h e m o r t a r . From t h e r e s u l t s
of G s when t h e s p a c i n g between r e i n f o r c e m e n t s was 27 m m , c o n c r e t e
s e g r e g a t i o n o c c u r r e d , and t h e c o n c r e t e b e h i n d t h e r e i n f o r c e m e n t
had g r e a t e r m o r t a r c o n t e n t .
Fig.8 g i v e s t h e e f f e c t of t h e amount of c o a r s e a g g r e g a t e on
f l o w a b i l i t y between t h e r e i n f o r c e m e n t s w i t h 30 mm s p a c i n g . When
t h e r h e o l o g i c a l c o n s t a n t s of t h e m o r t a r s were t h e same, t h e f l o w -
a b i l i t y between r e i n f o r c e m e n t s i n c r e a s e d as t h e amount coarse 05
a g g r e g a t e d e c r e a s e d , and t h e c o n c r e t e c o n t a i n i n g 276 L/m c o a r s e
a g g r e g a t e c o u l d f l o w w i t h o u t b e i n g blocked.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14B 94 m 0662747 0519954 8 T b m
From t h e a b o v e r e s u l t s , i t was f o u n d t h a t t h e f l o w a b i l i t y
b e t w e e n r e i n f o r c e m e n t s i s a f f e c t e d m a i n l y by t h e s p a c i n g b e t w e e n
r e i n f o r c e m e n t s and t h e amount o f c o a r s e a g g r e g a t e i n t h e con-
crete. The r e q u i r e d s p a c i n g between r e i n f o r c e m e n t s w i t h o u t
b l o c k i n g a p p e a r i n g i n c r e a s e d w i t h i n c r e a s i n g mass of c o a r s e
a g g r e g a t e . T h e r e f o r e , t h e r e l a t i o n s h i p b e t w e e n LFs/LF and i n d e x M
was p l o t t e d ( F i g . 9 ) . T h e i n d e x M i s a r a t i o o f t h e s p a c i n g i n d e x
( t h e s p a c e b e t w e e n r e i n f o r c e m e n t s d i v i d e d by t h e maximum s i z e o f
c o a r s e a g g r e g a t e ) t o mass o f c o a r s e a g g r e g a t e . F i g . 9 s h o w s t h a t
t h e highly superplasticized c o n c r e t e s c o u l d f l o w w i t h o u t b e i n g
b l o c k e d , when t h e i n d e x M w a s a b o v e 5.4.
CONCLUSIONS
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
I n t h e case of r e p l a c e m e n t o f cement by l i m e s t o n e powders,

t h e f l u i d i t y o f c o n c r e t e w i t h t h e f i n e s t l i m e s t o n e powder i s t h e
h i g h e s t . The p l a s t i c v i s c o s i t y o f p a s t e d e c r e a s e s d u e t o t h e
e f f e c t of t h e s o l i d v o l u m e p e r c e n t a g e o f t h e b i n d i n g m a t e r i a l
w h i c h was r e p l a c e d by t h i s l i m e s t o n e p o w d e r . I n t h e case of
r e p l a c e m e n t o f c e m e n t by s i l i c a f u m e , t h e f l u i d i t y o f c o n c r e t e
i n c r e a s e s w i t h a n i n c r e a s e i n r e p l a c e m e n t percentage of s i l i c a
f ume .
The e f f e c t o f t h e s p a c i n g b e t w e e n r e i n f o r c e m e n t s and o f t h e
amount of c o a r s e a g g r e g a t e on t h e f l o w a b i l i t y b e t w e e n r e i n f o r c e -
m e n t s is l a r g e r t h a n t h a t of p l a s t i c v i s c o s i t y of w e t - s c r e e n i n g
m o r t a r . When t h e s p a c i n g b e t w e e n r e i n f o r c e m e n t s was b e l o w 3 / 2
t i m e s t h e maximum s i z e o f c o a r s e a g g r e g a t e , h ' g h f l o w a b i l i t y
3
c o n c r e t e w i t h c o a r s e a g g r e g a t e c o n t e n t of 316 L/m c o u l d n o t f l o w
w i t h o u t b e i n g blocked. However, t h e c o n c r e t e c o u l d f l o w w i t h o u t
b l o c k i n g when t h e s p a c i n g b e t w e e n r e i n f o r c e m e n t s was 9 / 5 times
t h e maximum s i z e of c o a r s e a g g r e g a t e . Whether t h e h i g h f l o w i n g
c o n c r e t e c a n f l o w between r e i n f o r c e m e n t is d e t e r m i n e d by a n i n d e x
g i v e n by t h e s p a c i n g b e t w e e n r e i n f o r c e m e n t s d i v i d e d by maximum
size of c o a r s e aggregate.
REFERENCES
1. N i s h i b a y a s h i , S . , Yarnura,K. a n d I n o u e , S . , " R h e o l o g i c a l P r o -
p e r t i e s of S u p e r p l a s t i c i z e d C o n c r e t e " ; SP-68, A C I , 1 9 8 1 ,
pp.441-466.
2. C h i k a d a , Y . , M a t s u s h i t a , H . a n d Maeda,Y., "Study on t h e
Rheology of Cement P a s t e U s i n g Ground B l a s t F u r n a c e Slag";
P r o c . o f J C I , Vo1.13, No.1, 1 9 9 1 , pp.113-118.
3 . Yonezawa,T., Y a n a g i b a s h i , K . , Kouno,H., a n d A s a k u r a , E .
" D i s p e r s i o n o f a S i l i c a Fume i n H i g h S t r e n g t h C o n c r e t e " ;
P r o c . o f JCI, Vo1.15, No.1, 1 9 9 3 , pp.69-74.
A C 1 Sp-148 94 0662949 0 5 3 9 9 5 5 7 3 2
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 1 - PHYSICAL PROPERTIES OF POWDER CONSTITUENTS
Constituents Specific Blaine fineness

gravity (mZM
Portland cenent I 3.15 II 3220
I
I
I
1 2:; I
Blast-furnaceslag 2.89 6020
Limestone powder 1 I 2.73 I 3560
Limestone powder 2
Limestone powder 3 ",s:
Silica,fripe 1 2.20 200000
Silica,fume 2 2.20 I 180000 I
TABLE 2 - CHEMICAL CONSTITUENTS OF POWDERS
TABLE 3 - PHYSICAL PROPERTIES OF COARSE AGGREGATE

Max. Sieve test (Cumuiative Specific Unit
size percentage retainded) F.M.
(m) 2Onm 15 10 5
20 2 16 57 100 6.78 2.69 O. 87 1.59
o I l~ievitestl
(Cumulative percentage retainded)
5mn 2.5
14
1.2
36
0.6
51
I
0.3 0.15
76 95 2.72
I
TABLE 4 - PHYSICAL PROPERTIES OF FINE AGGREGATE
Specific
F.M. gravity Absorption
2.67 1.40
A C 1 SP-14ô 74 0662949 051775b b79
TABLE 5 - ADMIXTURES USED
Mark Kind Main components

Complex of ûenatured Ligin.
Ml AE type high *e Alkyarylsulfonic Acid and
reducingagent Continuous Activation Polynier
AE type high range water Cornplex of Polycarboxylic Ether
reducing agent and Cross-linked Polymer
TABLE 6 - MIXTURE PROPORTIONS USED TO STUDY THE EFFECT OF POWDER
LSP : Limestone powder, SF : Silica fume,

S : Sand, G : Crushud gravel
TABLE 7 - MIXTURE PROPORTIONS FOR FLOWABILITY TESTS
W/P d a (Wms) Type of

(%) W I C ISlagI S I G Admixture
0.32 I
I
48 I170 I266
I I
I266 II769 I1850 I
I
0.34 I I
46 170 250 I
250 753 I893 I
0.34 I 49 I170 I250 I250 I796 I850 I
0.34 52 i70 250 250 852 794
0.34 55 170 250 250 902 743 I
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 9 4 = Ob62947 0539957 505
I I
1000
Fig. 1-1-type flow test Fig. 2-Flowability test between reinforcements
O '
150
100-
m
0
> 50-
a
O '
O 30 5 0
Replace ment
percentage (% 1
Fig. 3-Effect of replacement percentage of limestone powder
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 94 W 0662949 051995ô 4 4 1
O1 1 10 100
Particle s i z e (urn)
Fig. 4-Distribution of limestone powder particle size
40
-p 3 0
Co
Q
Y
2 0
>:
1 0
O
O 8 1 5
Replacement
(%1
percentage
Fig. 5-Effect of replacement percentage of silica fume
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 94 Obb2949 0539959 388
E
o
LL Slag : 50%
4i
61 1
F’ov
-I
o .
E sE 2 O
2 . 5
1 .o
O O
h
LL A
--.
-1
08
A
2
i ;i:.\,__
0.5 o
1
Lm
v)
0.0
> 50 O 5 0 100150200
a: - 0 P. V. (Pas)
0.32 0.34 0.36
WIP
Fig. &Effect of W/P on FL, Fig. 7-Result of flowability
V, and rheological constants test (effect of plastic viscosity)
of mortar
1.o 1.o
LL
--.
-I
0.5
-1
--.
v) 0.5
LL LL
-I -I
0.0 0.0
250 300 350 4 5 6
A m o u n t of (L,m3)
M
coarse agg.
Fig. 9-Relation between LFs/LF

Fig. L R e s u I t of flowability test
and index M
(effect of amount of coarse aggregate)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 Ob62949 0539760 O T T =
SP 148-11
Applications of Rheo IogicaI

Modifiers and Superplasticizers
in Cementitious Systems
by C.B. Skaggs, W.G. Rakitsky, and S.F. Whitaker
Svnopsis: The impact of superplasticizers and water soluble-polymers, i.e.,

rheological modifiers, on the rheology and performance of cement-based
systems has been investigated. Combinations of water soluble-polymers and
superplasticizers can be used to formulate grouts, mortars, and concretes with
properties tailored for specific applications. Cement-based systems studied
ranged from highly fluid injection grouts to cohesive, flowable, concretes
suitable for underwater construction and repair applications.
This paper demonstrates how the rheology and performance characteristics of
cement-based systems can be manipulated using superplasticizers and
rheological modifiers. Specifically, the performance properties of a high-
molecular weight polysaccharide produced by fermentation are compared and
contrasted with cellulose derivatives. Combinations of water-soluble polymers
and superplasticizers can be formulated to produce a continuum of properties
ranging from highly fluid, non-separating grouts to low-slump concretes with
enhanced workability and water retention. Choice of the proper combination
of superplasticizer and water-soluble polymer is determined by the functional
demands of each application.
Kevwords: Cellulose fibers; cement-based svstems; compressive strength; grout; polysaccharides;

rheoloqical properties; superplasticizers; underwater construction; water retentivity
189Document
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
COPYRIGHT 2003; ACI International (American Concrete Institute) provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-348 94 0662949 0539963 T36 =
190 Skaggs, Rakitsky, and Whitaker
C. B. Skaggs is a Project Manager for Welan Gum Applications at Kelco

Division of Merck & Co., San Diego, CA
S . F. Whitaker is a Staff Chemist at Kelco Division of Merck & Co., Inc. San
Diego, CA
Dr. W. G. Rakitsky is Manager, Applied Research & Development Kelco
Division of Merck & Co., San Diego, CA
INTRODUCTION
Superplasticizers and water-soluble polymers function by different
mechanisms. Superplasticizers adsorb onto the surface of cement particles,
neutralize the inter particle attractive forces, and create a repulsive surface
charge. This action breaks up the flocculated cement particles and improves
flowability. However, the addition of superplasticizers to cement paste can
cause excessive bleeding and particle segregation.
The type of water-soluble polymers referred to in this study increase the
viscosity, or thicken, the interstitial water. Extremely small concentrations of
water-soluble polymers can dramatically alter the rheological, or flow,
properties of a solution. A properly proportioned cement-based system
containing both superplasticizers and water-soluble polymers eliminates the
undesirable features caused by the superplasticizer and enhances the flow
properties. For example, highly fluid grouts containing an efficient
combination of superplasticizer and water-soluble polymers have uniform
suspension, are highly flowable, and develop no excess bleed water. The
water-soluble polymers used in this study include welan gum, hydroxyethyl
cellulose (HEC), and hydroxypropyl methylcellulose (HPMC).
SUPERPLASTICIZERS: Superplasticizers provide a number of advantages
in cement-based systems including enhanced workability and reduced water
content. Sulfonated naphthalene-formaldehyde condensate (SNF) and
sulfonated melamine-formaldehydecondensate (SMF) are the most commonly
used superplasticizers in the construction industry. Modified lignosulfonates
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
(MLS) have found acceptance in oil field cements and are also used in
construction applications. Cellulose ethers are compatible with SMF, while
welan gum functions equally well with SNF, SMF, or MLS superplasticizers.
The chemical structure of these superplasticizers is described elsewhere.(i)
Studies show that up to a point, the efficiency of SNF increases with
increasing molecular weight and decreases with increased monomer
concentration.(2.3,4) The molecular weight of SMF superplasticizers can range
as high as twenty to thirty thousand.(i) Lignin is produced by most higher
plants and is composed of phenylpropane repeat units. It is separated from
cellulose during pulp and paper production and processed for cement-based
systems and various other applications. The molecular weight of MLS
superplasticizers can range up to one hundred thousand.(i) The performance
of a given superplasticizer is dependent on molecular weight, concentration,
A C 1 SP-148 94 9 0662949 05199b2 972
type of cement, w/c, presence of other admixtures, and order or time of

addition.
WATER-SOLUBLE POLYMERS: The water-soluble polymers used in this
study include cellulose ethers, HEC and HPMC, and welan gum. They are
classified as polysaccharides, or chains of saccharide molecules. Cellulose is
the most abundant of all natural molecules. It exists as a linear chain of
glucose molecules that exhibits strong inter and intra molecular hydrogen
bonding that makes them insoluble in water. Water-soluble cellulose
derivatives are manufactured by treating cotton or wood fibers with a strong
base that causes the molecule to swell. This swelling permits penetration and
reaction with the desired chemical compound to produce the water-soluble
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
ether. Commercial production of cellulose derivatives was established in
1938.(5) The molecular weight can range from 10,000 to over one miiìion.(b)
Cellulose derivatives improve workability, working time, and water retention
in cement-based environments.
Welan gum is a naturally occurring, high-molecular weight polysaccharide
that is produced by a carefully controlled aerobic fermentation using an
nlcaligenes strain ATCC 3 1555. Process steps include fermentation,
pasteurization, alcohol precipitation, drying, and milling to the desired particle
size. The welan gum used in cement-based systems is processed under
. specific conditions designed to assure optimum performance. The backbone
of welan gum is composed of D-glucose, D-glucuronic acid, D-glucose, and L-
rhamnose with a single side substituent consisting of either L-mannose or L-
rhamnose.(7) The molecular weight is estimated to be around two million.(8)
It appears to exist in solution as a double helix that builds viscosity through
direct helical interaction via hydrogen bonding or through ion mediated
association of helices. Intermolecular hydrogen bonding between the main
chain and the side chains contributes to the stiff conformation of the molecule.
This leads to properties such as high viscosity at low rates of shear, salt
tolerance, stability to changes in pH, and temperature tolerance. Because of its
lack of significant hydrophobic substituents, welan gum has little activity at
the air-water interface and does not cause foaming problems.
POLYMER SOLUTION RHEOLOGY: Polymer solutions can be

classified according to their rheological behavior. The polymer solutions used
in this study are classified as pseudoplastic, which means that the viscosity
decreases with increased shear rate. The degree of pseudoplasticity varies for
different types of polymer solutions. Random coil polymer solutions tend to
be less pseudoplastic than those of rigid rod polymers. Cellulose ethers are
random coil polymers that have found wide acceptance in cement-based
applications. Welan gum is a rigid rod biopolymer that is used in oil-field
cementing applications and is gaining acceptance in the construction industry.
Fig. 1 compares the efficiency of weIan gum to hydroxypropyl methylcellulose
(HPMC), at a shear rate of 1 S-I. The solution containing 0.25 percent welan
gum exhibits similar viscosity to one containing 0.80 percent HPMC. For a
given application, the amount of water-soluble polymer required is directly
related to the w/c. Polymer concentration requirement for a given cement
based system decreases as the w/c decreases and increases as the water content
increases. Typical use levels of HPMC in cement-based applications range
from 0.02 to over 2% by mass of cement. Welan gum use levels range from
0.02 to 0.2% by mass of cement.
A C 1 SP-148 94 m 0662749 0519963 8 0 9
Fig. 2 shows that at one third of the concentration of HPMC, welan gum has
more than 100 times the viscosity of the cellulose derivative at 10-5 s-1 while
exhibiting approximately one tenth the viscosity of HPMC at 1000 s-1. This is
indicative of welan gum’s excellent shear thinning properties. Both polymers
solutions have sigmoidally shaped flow profiles with regions at low and high
shear rates where they display Newtonian behavior (viscosity independent of
shear rate). The lower Newtonian region is not well defined for the weIan gum
solution because rheological measurements at lower rates (< 10-5 s-1) of shear
are not possible with existing instrumentation. Although the efficiency of
polymers can be adjusted by changing molecular weights, their flow profiles
remain distinct. Note that both solutions exhibit the same apparent viscosity at
approximately 10-o,8 s-l. At lower shear rates, the welan solution is more
viscous. However, as the shear rate is increased from 10-0.8 s-1, the HPMC
solution is more viscous. These data indicate that welan is the preferred
rheological modifier for low shear rate applications such as bleed and/or sag
control and particle suspension. HPMC is more efficient in higher shear rate
applications such as particle transport.
PROPERTIES IN CEMENT-BASED SYSTEMS

CEMENT PASTE RHEOLOGY: The influence of SMF, hydroxyethyl
cellulose (HEC), and welan gum on grout rheology is presented in Table 1.
All samples were prepared using type 1/11 cement at w/c = 1. A 400 mL
unsanded grout sample was mixed for 10 minutes at 600 rpm using a 50 mm
diameter turbine blade impeller. Rheology was measured immediately after
mixing using a concentric cylinder viscometer fitted with a 17.245 mm radius
bob and a 18.4 I5 radius mm rotor.
The first grout contains no additives, a second grout contains 1% SMF by

mass of cement, the third grout contains 1% SMF and O. 12% HEC by mass of
cement and the fourth grout contains 1% SMF and 0.06% welan gum by mass
of cement. Addition of the 1% SMF by mass of cement reduces the viscosity
measured at 510 s-I from I I to 8 m P a s At 5 s-I, the SMF reduces the
viscosity from 260 to 180 mPa.s. These data also demonstrate the significance
of shear rate for the two different water-soluble polymers. The addition of
0.12% HEC increases the viscosity at 51G s - I from 8 to 21 mPa.s while
increasing the viscosity at 5 s-1 from 180 to 210 m P a s The addition of 0.06%
weIan increases the viscosity measured at 510 s-I from 8 to 13 m P a s
However, at 5 s-1, it increases the viscosity from 180 to 280 mPa.s.
Fig. 3 is a plot of the apparent viscosity verses shear rate of two 0.6 w/c grouts
containing 1.14% SMF by mass of cement. Both grouts have the same
viscosity at about 75 s-‘. At shear rates less than 75 S I , the grout containing
weIan gum becomes more viscous. The grout containing HPMC becomes
more viscous than the welan grout as the shear rate is increased beyond 75 S I .
WATER RETENTION: Many applications place tough water retention

demands on the fresh cement paste. The term “water retention” is
synonymous with filtrate and fluid loss control. The factors that influence
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
fluid flow through a porous media are described by Darcy’s Law as illustrated
in Fig. 4.p) Any treatment that reduces the migration of water through the
interstitial spaces within the fresh paste will result in a corresponding
improvement in water retention. For example, water retentivity is improved as
the w/c is reduced. Superplasticizers can improve particle packing and thus
reduce the permeability (K) of the fresh cement paste. Water-soluble polymers
increase the viscosity (p)of the matrix water and improves water retention.
SEDIMENTATION: The combined use of superplasticizers and water-

soluble polymers influence :wo of the key variables required to reduce the
particle settling rate as described in Fig. 5.( 10) Note that the settling velocity
(V,) increases with the square of the particle diameter therefore, particle
dispersion is critical in order to eliminate or reduce sedimentation. Viscosity
(p) has a direct impact on particle settling. The settling velocity is reduced as
the viscosity is increased.
The sedimentation properties of 0.6 w/c grouts containing 1.14% SMF by

mass of cement are illustrated in Fig. 6. The fresh grouts were poured into a
38 mm diameter plastic tube, allowed to set, and cut into 50 mm segments.
Slurry sedimentation is determined by measuring the density of the 50 mm
segments. Fig. 6 shows the density of the hardened cement paste for a control
(SMF only), a second grout containing 0.10% HPMC by mass of cement, and
a third grout containing 0.05% welan gum by mass of cement. The sample
containing welan gum did not bleed or settle and produced a uniform column
of hardened cement paste. The grout containing HPMC controlled bleeding,
but exhibited sedimentation. The control exhibited both bleeding and
sedimentation. At low shear rates, ¡.e., those expected of settling cement
particles, the welan is highly viscous and minimizes particle settling and
bleeding.
COMPRESSIVE STRENGTH: Water soluble polymers influence primarily

the fluid properties of cementitious systems and have minimal impact on the
cement hydration process. However, characteristics imparted to the fluid
system by the use of these polymers in combination with superplasticizers, can
influence the hardened properties. As an example, HPMC exhibits surface
active properties at the air-water interface, facilitating the formation of foam.
This results in an increased hardened air content and corresponding decrease in
compressive strength as illustrated in Fig. 7. Anti-foaming agents are often
used in mixtures containing cellulosics to prevent excessive air.
APPLICATIONS
POST-TENSIONING GROUTS: Cement grout is used in post-tensioning
applications to protect the tendons from corrosion. Potential problems include
bleeding and sedimentation. The use of strand tendons can magnify these
problems when differential pressure causes the water to migrate from the grout
into the voids between the strands. A substantial vertical rise can compound
these effects and high points could be left entirely ungrouted. For these
reasons, it is extremely important that these grout systems provide suspension,
resist bleeding, and control fluid loss through steel strands without sacrificing
workability.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 94 O b b 2 9 4 9 0.519965 b ô L m
Water-soluble polymers, used in conjunction with a superplasticizer, can

significantly increase the water retentive properties of a cement grout. Fig. 8
illustrates the fluid loss properties of 0.40 wlc, neat cement grouts containing
welan gum and a high molecular weight HEC with 1.14% by mass of cement
of a SNF/SMF blend. The fluid loss is measured by applying a 0.55 MPa
differential pressure to the grout and measuring the loss of water through a fine
mesh screen.(l i ) Use of these polymers reduced the fluid loss to less than 2
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
mL (1.5% of the total water). A grout containing no superplasticizer or water-
soluble polymer exhibited 55 mL (54% of the total water) fluid loss.
Neat cement grouts containing welan gum and a high molecular weight HPMC
were formulated for the same fluid loss control and their plastic and hardened
properties evaluated. The results are presented in Table 2. The 0.40 w/c
grouts were prepared with portland type 1/11 cement and 0.28% by mass of
cement SNFISMF. The welan gum or HPMC was slurried into the liquid
superplasticizer then post-added to the wetted cement. Fluid loss, consistency
(modified flow cone), set time (ASTM C - 19 1-82), strength development
(ASTM C-109-86), and hardened air content (ASTM C-457-71) were
measured. Test methods were modified as necessary for evaluation of cement
grouts.
Highly fluid grouts, comparable to the control in consistency, were produced

using 0.05% by mass of cement welan gum and 0.08% by mass of cement
HPMC. The water retentive properties, however, were greatly improved,
reducing the fluid loss from 49 to 8-9.5 mL. Even greater fluid loss control
was achieved by increasing the polymer concentration to 0.15% by mass of
cement welan gum and 0.20% by mass of cement HPMC, resulting in only 1-3
mL of fluid loss. None of the grouts containing polymer displayed static
bleeding compared to significant bleeding observed for the control. At the
higher use levels of polymer, flow cone times were increased but the systems
were still fluid. By manipulating the concentrations of water-soluble polymer
and superplasticizer, grouts with differing consistencies can be produced.
Set times for the grouts containing polymer were equivalent. Water-soluble
polymers have minimal impact on the set time of cement-based systems,
instead it is the type and concentration of superplasticizer which has the
overriding effect. Addition of welan gum to the grout slightly increases the
hardened air content. As already indicated, HPMC exhibits surface active
properties at the aidwater interface resulting in an increase in the hardened air
content. This explains the somewhat lower strengths observed for the
hardened systems. The use of deair agents in mixes containing cellulosics can
help prevent excess air generation. Optimized formulations provide the
opportunity to produce cement grouts of differing consistencies with strength
development similar to a standard cement grout but with far superior fluid loss
control.
INJECTION GROUTS: Rheology and fluid loss control form the basis of
many applications for water-soluble polymers in grouts and concretes. An
example of the impact combinations of admixtures have on the performance of
injection grouts is shown in Table 3. An ideal injection grout must exhibit a
low viscosity at injection shear rates to facilitate penetration, minimize water
loss as it passes through an absorptive surface, and keep the cement particles
A C 1 SP-148 7 4 0bb2949 0519966 518
suspended once injection ceases. The grout must remain homogeneous

through initial set in order to produce desirable mechanical and barrier
properties in the grouted soil or structure. Combinations of welan gum and
superplasticizer produce an injectable grout with excellent stability, and
significantly reduced fluid loss.
OIL FIELD CEMENT: Oil wells are completed by lowering a steel pipe, or
casing, into the drilled hole and filling the formation and casing annulus with a
cement slurry. The slurry is pumped down the casing and up the annulus. A
plug is displaced behind it to prevent the fresh cement slurry from re-entering
the casing. Once placed, the hydrostatic pressure exerted by the cement
column serves as the sole means of preventing formation fluids such as natural
gas from entering the wellbore. Particle sedimentation reduces the slurry
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
density and causes a corresponding loss of hydrostatic pressure. Therefore, it
is critical to prevent sedimentation of the plastic cement slurry. The grout
must be highly fluid to prevent excessive friction pressure during placement,
must not lose excessive water to the formation, and must provide ample low
shear viscosity to prevent creating free water and particle sedimentation. The
w/c typically ranges from 0.38 to over 0.6 and superplasticizers are added to
reduce friction pressure and improve fluid loss control. Grouts containing a
superplasticizer have a lower viscosity and are easier to mix and pump. The
dispersion of particles caused by the superplasticizer improves particle size
distribution and allows the cement particles to deposit a tight, low permeability
filter cake. Hydroxyethyl cellulose is used to control fluid loss in high salt
cements. Carboxymethyl hydroxyethyl cellulose is used to retard cement
hydration and control fluid loss. WeIan gum improves fluid loss and is used to
eliminate free water and prevent sedimentation.( I 2 )
The American Petroleum Institute (API) Committee on Cementing has

developed a set of procedures that govern the laboratory procedures used to
test oil well cement slurries. It covers slurry preparation and conditioning in
addition to measuring the rheology, thickening time, fluid loss, and
compressive strength.( I 3) Rheological profiles of four class H oil well cement
slurries are shown in Fig. 9. Ail slurries were mixed at 0.45 w/c in a Waring
blender using API procedures and conditioned in an atmospheric
consistometer at 66 "C for twenty minutes prior to testing. At 5 s-1, the slurry
containing O. 1% weIan is substantially more viscous than the slurry containing
0.75% HEC. These data demonstrate the efficiency, even at high temperature,
of the welan. At about seventy-five s-1, both slurries have the same viscosity.
At shear rates above 75 s-1, the slurry containing 0.1% w e h is less viscous
than the slurry containing 0.75% HEC. The density of the set, or hardened
cement paste is shown in Fig. 10. Note that 0.03% welan eliminated
sedimentation in this cement slurry.
UNDERWATER CONCRETE: Concretes designed to resist dilution when

poured through water usually incorporate an antiwashout admixture (AWA) to
prevent loss of cement fines. Concrete formulations prepared with welan gum
or cellulose derivatives resist washout but will not self consolidate.
Superplasticizers are added to allow the concrete to flow into place.
Performance of concretes containing combinations of water-soluble polymers

and superplasticizers cannot be predicted from simple slump tests. Fig. 11
196
A C 1 SP-148 94
Skaggs, Rakitsky, and Whitaker
-
0bb29i.19 0519967 Y54 W
illustrates the differences in washout resistance of medium strength concretes

with equal slumps. The washout mass loss test consisted of filling a perforated
steel basket with a sample of fresh concrete, and then allowing the basket to
free fall through a test column 1.68 m in height filled with water.( 14) The loss
in mass of the concrete is measured at the conclusion of each of three drops.
The AWA concretes in Fig. 1 1 contained O. 15% welan gum or 0.50% HPMC
by mass of cement. All concretes were formulated with a melamine based
superplasticizer and used the same mix proportions with the exception of the
HPMC concrete. To minimize air entrapment in the HPMC concrete, 0.20%
by mass of cement deair agent was added. The concentration of SMF in the
control concrete was 6.5 mL/kg of cement compared to 13 mL/kg of cement
for the AWA concrete in order to generate similar slump values. No washout
was observed for either AWA concrete at one drop. At three drops the AWA
concretes exhibit approximately 0.50% mass loss while the control concrete
mass loss was 9.0%. Concrete slump can be increased by raising the level of
superplasticizer at the expense of washout resistance for a given concentration
of AWA.
Concretes containing antiwashout admixtures can resist segregation and

bleeding. Using ASTM test method C 232-87, bleeding characteristics were
evaluated for control concrete and underwater concretes containing welan gum
and HPMC. Results of the tests shown in Fig. 12 confirm that AWA concretes x\
exhibit minimal bleeding tendencies despite having slumps of 180 mm.

Manipulation of the level of AWA in the concrete will result in the production
of bleed water should it be required to enhance finishing in above water
applications. Even in the absence of bleeding, sedimentation can occur. In
recent work at the University of Sherbrooke, bleeding elimination has been
found to be a necessary but insufficient condition to prevent segregation.
The setting times of control and AWA concretes containing welan gum and
HPMC, measured in accordance with ASTM C403, are shown in Fig. 13. The
control concrete reached initial set in approximately 5 hours while the AWA
concretes containing 2 times more superplasticizer than the control achieved
initial sets in approximately 6.5 hours. Impact of combinations of AWA and
superplasticizers on setting times cannot be predicted based on individual
results. Tests must be performed with representative samples of cement and
the admixtures being considered. However, concretes can be formulated with
a range of setting times to match job requirements using existing admixture
technology.
Strength development for concretes containing superplasticizer and either

weIan gum or HPMC formulated for underwater applications are compared to
control concrete in Fig. 14. Ali specimens were cast above water.
Compressive strength values are the average of five specimens. At 7 days,
concrete with welan gum demonstrated 87% of control strength while the
HPMC concrete was 70% of control strength. At 84 days, the welan gum and
HPMC concretes were 95% and 84% of control strength respectively.
Hardened air content for the control was 1.0% while the hardened air content
of the AWA concretes was 2.4%. Considering the air content, both polymers
have minimal impact on strength. Despite being slightly higher in air content,
the concrete with the w e h gum exhibited compressive strengths comparable
to control concrete.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14B 94 0662949 0529968 390
The concentration of AWA required to produce washout resistant concrete is

dependent on the mix proportions, w/c ratio, polymer type, and the desired
flow and leveling properties. Concrete for underwater construction and repair
should be formulated with a combination of rheology modifier and
superplasticizer. Concrete lacking either component will not function
acceptably in underwater applications.
SUMMARY
The rheological, or flow properties of fresh cement paste are efficiently
controlled with combinations of superplasticizers and water-soluble polymers.
Superplasticizers adsorb onto the cement and disperse the particles whereas
water-soluble polymers viscosify the water. The result is a synergistic
improvement in the performance of fresh cement paste. The dispersion of
cement particles reduces rates for both particle settling and flow of the
interstitial water. The rheological profile of a cement based system varies for
different water-soluble polymers. WeIan provides excellent suspension
viscosity and efficiently controls bleeding and settling in high w/c grouts.
Cellulose ethers are more viscous at higher shear rates. Combinations of
superplasticizers and water-soluble polymers lead to enhanced systems
ranging from highly fluid injection grouts to low slump concretes.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 0662949 0.539969 2 2 7 W
REFERENCES:
1. Ramachandfran, V. S., “Concrete Admixtures Handbook, Properties,
Science, and Technology,” Noyes Publications, Park Ridge, New Jersey, USA,
1984, p 212.
2. Basile, F., et. el., “Influence of Different Sulfonated Polymers On The

Fluidity of Cement Paste,” Paper SP 1 19-11 , Presented at The Canada Centre
for Mineral and Energy Technology (CANMET), May, 1988.
3. Collepardi, M., Corradi, M. and M. Valenti, “Influence of Polymerization
of Sulfonated Naphthalene Condensate and Its Interaction With Cement,”
Paper SP 68-27, Presented at The Canada Centre for Mineral and Energy
Technology (CANMET), May, 1988.
4. ibid.. 2.
5. Whistler, Roy L., and B. E. Miller, James N., “Industrial Gums

Polysaccharides and Their Derivatives ,” Second Edition, Academic Press,
1973.
6. Whistler, Roy L., and B. E. Miller, James N., “Industrial Gums
Polysaccharides and Their Derivatives ,” Third Edition, Academic Press,
1993.
7. Clare, K., “Applications of Novel Biogums,” CHEMSPEC USA
Symposium, 1988.
8. Rakitsky, W. G., “Suspending Agent Applications in Cementitious
Systems,” Presented at the ConChem International Exhibition & Conference,
Karlsruhe, Germany, November, 1993.
9. Mayer-Gurr, Alfred, “Petroleum Engineering,” Volume 3, Ferdinand Enke
Publishers, Stuttgart, 1976, p 19.
10. Gover, G. W. and Aziz, K., “The Flow of Complex Mixtures in Pipes,”
Robert E. Krieger Publishing Company, Malabar, FL, 1987, p 6.
1 I . Schupack, Morris, “Admixtures for Controlling Bleed in Cement Grout
Used in Post-Tensioning,’’ PCI Journal, November-December, 1974, pp 28-
39.
12. Ganguli, K. K., “Biopolymers as Free Water and Settling Control Agent,”
Paper SPE 25437, Presented at the Production Operations Symposium held in
Oklahoma City, Oklahoma, March 21-23, 1993.
13. API Specifications for Materials and Testing for Well Cements, API
Specification 10, 2nd ed., American Petroleum Institute, Washington, D.C.,
June 15, 1984.
14. CRD C61 “Test Method for Determining the Resistance of Freshly-Mixed
Concrete to Washing Out in Water.” Handbook for Concrete and Cement, U.
S. Corps of Engineers, Waterways Experiment Station, Vicksburg, MS.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-34ô 94 I0 6 6 2 9 4 9 0539970 T49 =
TABLE 1 - INFLUENCE OF ADDITIVES ON GROUT VISCOSITY
Grout Viscositv ímPa s)

Shear Rate (s-1) #1 #2 #3 #4
510 11 8 21 13
340 13 10 22 14
170 19 13 23 18
102 27 18 27 25
51 42 26 34 40
10 130 90 125 150

5 260 180 210 280
#1 = Control grout (no admixtures)

#2 = 1.0% SMF (by mass of cement)
#3 = 1.0% SMF + 0.12% HEC (by mass of cement)
#4 = 1.0% SMF + 0.06% Welan (by mass of cement)
Type I/ii cement, w/c = 1
TABLE 2 - NEAT CEMENT GROUT CHARACTERISTICS
- Control Welan HPMC Welan HPMC

WIC 0.40 0.40 0.40 0.40 0.40
0.08
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Polymer (W cwt) 0.05 0.15 0.20

SNFBMF (% cwt) o 28 0.28 0.28 o 28
Fluid Loss (mL) 49.0 9.5 8.0 3.0 1.0
Condition Very Fluid Very Fluid Very Fluid Fluid Fluid
Flow Time (sec) 9.2 12.7 10.4 25.3 27 o
Set Time (hours) 6.0 8.3 8.0 8.5 8.0
Ave. 7 Day f c (MPa) 48.2 (2.8)' 49.0 (1.7) 46.3 (2.9) 49.4 (1.3) 50 1 (0.2)
Ave. 28 Day f e (ma) 64.0 (1.1) 62.6 (1.7) 57.0 (1.4) 63.7 (1.4) 58.1 (4.9)
Ave. 56 Day f c (ma) 72.6 (2.8) 66.6 (1.4) 62.8 (3.6) 71.6 (1.0) 63.1 (2.1)
Hardened Air Content % 0.88 1.07 2.33 1.33 4.24
*7& 28 day strengths, n = 3; 56 day strengths. n = 5; standard deviations are in parenthesis
A C 1 SP-148 9 4 Ob62949 051997L 985
m
I
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14B 94 m 066294'7 0539972 8 3 3 m
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
10000
/*
I ow
o
1O0
z
B
1 10
-t CPMC
1 . . ~ I ~ " I . ' . " . ' " " I ~ ' ' I

0.0 0.20 0.a 0.60 0.80 1.0 1.2
Polymer Concentration (wi X )
Fig. 1-Polymer efficiency measured at 1 s-'

i
#'
Fig. 2-Shear rheology of w e h gum and HPMC
A C 1 SP-148 94 IOb62949 0539973 758
=
loo0 -
....*... O 10 H P X C
-I
b
'1
i * .................. ..... .....
~
x)
Fig. &Shear viscosity profiles of cement grouts (0.6 w/c, 1.14 percent SMF)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. &Flow through porous media
A C 1 SP-I148 94 0 6 6 2 9 4 9 0 5 1 9 9 7 4 694
Fig. L l n f l u e n t e of particle size and viscosity on settling rate
2.2
2.0
P
a 1.8
1.6
1.4
1 2 3 4 5 6 7 8 9 l O l l l 2 B
BOITOM Column Segements (50 mm segments) TOP
Fig. &Effect of water-soluble polymers on grout uniformity (0.6 wkc, 1.14 percent SMF)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4B 74 Obb2949 0539975 520 H
O 4 wlc Cement Grouts
Control
i
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
70-
L
m 20% HPMC
id
.05% WeIan (.28% SNFBMF)
(.28% SNF/SMF)
60- .Oô% HPMC
(.2û% SNF/SMF)
2 ,254 HPMC
i1 14%CMF,
8 501
O 1 2 3 4 5
nIIardenedALIc0lltent
Fig. 7-Effect of hardened air content on compressive strength
m
ao
45
3 s
1:
i:
u)
5
O
0.060 0.075 0.100 0.125 0.150 0.175 0.200 0.225 O250
poiymerConœn~fion(%cwt)
Fig. &Fluid loss of welan gum and cellulosic (0.4 w/c, 1.15 percent SNF/SMF, 55 MPA for 10 min)
A C 1 SP-148 9 4 0662949 0539976 467
Superplasticizers and Chemical Admixtures
= 205
- 0.5%HEî
h
r -O--. O 754. HEC
.-o ._..
0.... 03%Welam
i -&--- 1% Welm
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
I 10 100
Shew Rate (s-1)
Fig. 9-Rheology of cement slurry measured at 66 C
No Polyrncr
O 5% HEC
0.15% HEC
c
0.03%W A N
0.1%WELAN
I I I I
16 1.8 2 2.2 14
Specifc Gravity
Fig. 1û-Specific gravity of hardened cement paste cured at 66 C
A C 1 SP-148 94 0662747 0537977 3T3 W
a
A S wlc I
W Welin Gum
HPMC
d
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
C
Drop 1 Drop 2 Drop 3
Number o( b o p s Through Water
Fig. 11-Washout resistance of 180 mm slump concretes
- AWAConcrein
slump = 180 mm
O
O 100 an m
Age (minutes)
Fig. 12-Bleeding characteristics of AWA and control concretes
A C 1 Sp-148 94 = 0 6 6 2 9 4 9 05l19978 23T W
w -
cz
O
&
* HPMC Concreta, super=l3 mVKg cm
E
O 33-
2O
k a-
t
œ m-
C
O
-
h 15
2 Y)-
2
5-
01
o 1DJ an m a m n
Seiiing Time (minutes)
Fig. 13-Setting behaviour of washout resistant concrete (.45 wkc)
N.
ControlConcre(e 8 WelanGumConcrete H HPMCConmte
O
&
Fig. 14-Strength gain profiles of concretes formulated at .45 w/c
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 9 4 I0 6 6 2 9 4 9 0539979 176
SP 148-12
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Mixture Proportion, Flowability,

Strength and Durability of
. Flow ing Concrete
by Y. Kasai, S. Hiraishi, K. Tobinai, and K. Osada
SvnoDsis: Flowing concrete with high flowability prepared with a river gravel and
a crushed stone was investigated for mixture proportions, flowability, strength,
shrinkage, carbonation and freezing and thawing resistance. This concrete has
proven to be highly feasible both in terms of cost and in performance . The main
findings can be summarized as follows.
(1) The slump of the flowing concrete capable of filling sufficiently with slight
compaction ranges from 24 to 26 cm, which corresponds to a flow from 50 to
60 cm and a differential height less than 8 cm in the box test .
(2) Flowing concrete with a water-cement ratio from 30% to 60% can be made by
using a new admixture and by a simple correction of the standard table of the
mixture proportion.
(3) Flowing concrete can be produced with specified concrete strengths ranging
from 18 to 60 MPa.
(4) The strength and durability of flowing concrete showed no significant
difference from that of AE concrete without any special admixtures.
Kevwords: Admixtures; carbonation; compaction; durabilitv; flowability; freeze thaw durability;

mix proportioning; shrinkage; slump; strength
A C 1 SP-I148 74 Ob62749 0539980 978 =
210 Kasai et al
Y.Kasai is a professor of building materials at Nihon university, College of

Industrial Technology. He has been in charge of various field of building
technology including durability of concrete, non-destnictive testing of concrete and
recycling of construction by-products.
S. Hiraishi is a research manager of technical development division at Yamaso
Chemicals Co. He has been in charge of research and development in chemical
admixtures for concrete.
K. Tobinai is a chief research engineer of building materials development
division at Research Institute of Mitubishi Materials Co. He has been in charge of
technical development of inorganic building materials. r
K. Osada is an engineer of Higashi-matsuyamaconcrete technical laboratory at
Nippon Pulp Co. Ltd.
INTRODUCTION
The reinforced concrete structures in Japan have been designed to have large
amounts of reinforcement to resist earthquakes. The wet consistency concrete has a
long history since the first table of standard mixture was specified in 1933. P
concrete with a slump of 24 cm was documented in the standard specification for -
concrete construction of The Architectural Institute of Japan (JASS 5, Reinforced
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Concrete) in 1935. This concrete is highly flowable, easily compacted, and

introduced in a complicatedjoint zone, and can be free of the honeycomb problem,
although it tends to form cracks due to drying shrinkage.
Considering the shortage of good aggregate resources and durability
problems, the current JASS 5 , revised in 1986 has recommended the slump of
fresh concrete to be as low as 18 cm and the unit water content as low as 185
kg/m3, which has resulted in substantial labor work in concreting on site. A special
concrete without compaction has been developed and introduced (I) to meet this
requirement. It contains considerable amount of fine binder such as blast furnace
slag or fly ash and superplasticizerto improve the filling capability and flowability,
however its long term stability is still unknown.
The present paper deals with experimental results on the mixture design,
flowability, strength, shrinkage, carbonation and freezing and thawing resistance
of the flowing concrete designed to save labor work during concreting by means
of a special admixture and slight compaction, and which further meets the quality
requirements of JASS 5 based on the normal strength (Fc =18 - 60 MPa) of wet
consistency concrete.
EXPERIMENTAL PROCEDURE
Materials Used
Cement and admixtures used for the exDeriments are listed in Table-1. and
aggregates in Table-2. Admixtures used are t h o types of water reducing agents ;
naphthalene sulfonate and poly-carboxylate both of which were mixed with water
soluble acrylic polymers which can give high segregation resistance as shown in
Table 1.
A C 1 Sp-L4& 9 4 m Ob62949 0519981 8 2 4
Superplasticizers and Chemical Admixtures 21 1
Testing. Method
Mixing of 60 liters per batch of concrete was made in a 100 liters capacity
mixer. The mixing time was 30 seconds after the cement, fine aggregate, water and
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
admixtures were introduced and 90 seconds after the coarse aggregates were
introduced. Tests on the fresh concrete included the flowability test, air content,
temperature just after mixing and bleeding, while compressive strength, change in
length, carbonation and rapid freezing and thawing were performed on the
hardened concrete.
The following three tests were carried out for the flowability evaluation (2).
>y [ 11 Slump test: Based on JIS A 1101
[2] Flow test: Based on the manual of under water, segregation free
concrete (3).
[3] Box test: Fresh concrete was filled in the half part compartment of a
15 x 15 x 40 cm acrylic box shown in Fig. 1. Then the partition was pulled up to
allow the concrete to flow into the other part , and the differential height between
the two parts was measured.
For the hardened concrete, compressive strength and change in length were
evaluated according to JIS A 1108 and 1129 respectively. The drying was
maintained for 26 weeks and the accelerated carbonation test was done
subsequently at a temperature of 20 60 % R.H. and 5% of CO2 concentration.
O C ,
The freezing and thawing test was conducted up to 300 cycle based on ASTM C
! 666A.
Details of the Experiments

1 --Preliminary test
The mixture proportion of the test concrete was determined with reference to
the standard mixture proportion of concrete in JASS 5 (1957) (4). The special
admixture used for this experiment was chosen based on preliminary test results. A
concrete with this special admixture shows no increase in the differential height in
the box test even though the flow increases as shown in Fig.2. The unit amount of
water was kept constant at 185 kg/m3 , and the cantent of fine aggregate and
admixture were varied to keep the specified flowability as shown in Fig.3.
As a result of the preliminary mixing, a slump to provide sufficient filling
capability with slight compaction was determined to be 24 to 26 cm, a flow from
50 to 60 cm and differential height in the box test less than 8 cm.
2--Main experiment
The experiment comprised five series. For the series 1 experiment, river sand
and river gravel were used, and a comparison between wet consistency concrete
with AE agent and with AE-water reducing agent was made. Effects of blast
furnace slag and expansive cement admixture were also investigated.
For series 2, river sand and crushed coarse aggregate were used, and effects
of the maximum size of coarse aggregate and lime stone powder were examined.
For series 3, river sand and crushed stone were used, and effects of the
water-to-cement ratio was examined.
In series 4 and 5, river sand - river gravel and river sand - crushed sand
were used and water-to-cement ratios ranging from 60, 50, 40 and 30 % were
investigated. The mixture proportions of these concretes and their test results are
shown in Table-3 and 4 respectively.
A C 1 SP-148 94 O b b 2 9 4 9 0 5 3 9 9 8 2 760
212 Kasai et al
Mixture Proaortions of Flowinp Concrete

A method of correction for mixture proportions by the standard table is
shown in Table-5. It has been pointed out that the higher the unit amount of water
is, the lower the apparent yield value becomes, and the resulting flowability
increases (5). The unit amount of water can be reduced with the help of admixtures
whose dosage however, still has a limit due to the set-retarding effect and slow
development in the early strength .
The unit amount of water used in these experiments was 185 kg/m3 which
was within the permissible range recommended in JASS 5. With respect to river
sand-river gravel system, the amount of fine aggregate at 60% water-cement ratio
was increased by 4.9% for river sand - crushed aggregate and 6.4% for 50%
crushed sand substitution so as to keep the designed flowability . The dosage of
special admixture varied according to the aggregate type; the river sand-river gravel
system required the least amount while the 50% crushed sand substitution system
required 15 % larger dosage. Table 4 shows that the fine aggregate ratio (da) of
concrete with a water-cement ratio ranging from 50% to 30% was decreased by
3.5% to 9.1 % compared to concrete with a water-cement ratio of 60%.
i
Flowabilitv
The relationship between the slump and the differential height is shown in
Fig.4 (a). The differential height decreases proportionally as the slump increases,
and it holds for all aggregate combinations except for the crushed sand-crushed
stone system. For example, a slump which corresponds to a differential height
below 8 cm ranges from 24 to 26 cm. The same trend can be seen in the
relationship between flow and the differential height shown in Fig.4 (b). For
example, a flow which corresponds to a differential height below 8 cm ranges from
50 to 60 cm.
These results imply that larger are the slump and flow, the smaller the
differential height becomes, and the fluidity of flowing concrete can be evaluated,
to some extent, by measuring either the slump, flow and the differential height.
Within the range of these experiments, a desirable fluidity for the flowing concrete
can be considered to have a slump of 24 to 26 cm, a flow of 50 to 60 cm and a
differential height less than 8 cm. When the maximum size of the coarse aggregate
is reduced to as low as 13 mm, the fluidity evaluated by the box test was not so
significant. However, it may contribute to better casting and compaction at
complicated joint zones. The desirable fluidity could not be obtained when either
AE or AE-water reducing agent was solely used as shown in Table-4.
Bleeding
A relationshir, between slumr, and the differential height with resuect to the
amount of bleeding is shown in Fig.5. The bleeding test h a s executed using a
container with a diameter of 15 cm and a height of 30 cm. The measured amount of
bleeding for the flowing concrete was 0.2 mYcm2 or less. When the water-cement
ratio was less than 40 %, it was and O. 1 mVcm2 or less as shown in Fig. 6. These
values fully met the value of 0.5 mVcm2 or less recommended by JASS 5.
Compressive Strength
A relationship between the normalized compressive strength and the type of
concrete is shown in Table-6. The development of compressive strength of the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-3Y8 74 0662949 0 5 3 9 9 8 3 bT7
concrete with river sand - river gravel substituted by 30 % fine blast furnace slag
was not so satisfactory at early stages and showed a normalized strength of 0.96 at
the age of 28 days, while that for river sand - crushed stone substituted by 10 %
lime stone powder was more noticeable with a substantial increase at a constant unit
amount of water, and with a normalized strength of 1.18 at the same age. The
compressive strength of the flowing concrete is nearly equal to that of concrete
using an AE admixture. No effect of the maximum aggregate size on the
compressive strength could be seen when the maximum size was reduced to 13
mm. A relationship between the water-cement ratio and the compressive strength is
shown in Fig.7, where linear relationships can be seen at any given age.
Shrin kaPe
l--Effects of water-cement ratio
The effect of water-cement ratio on the drying shrinkage behavior of the
flowing concrete is shown in Fig.8(a). An increase in drying shrinkage was
observed at an early stage of drying with a decrease in water-cement ratio, while
after 4 weeks, drying shrinkage was almost equal to each other except in the case
of water-cement ratio 30 %, and ranged from 6.5 to 7 x at the age of 26
weeks. The drying shrinkage in the case of 30 % water-cement ratio was 6 x 10-4
at the age of 26 weeks. These results may be attributed, at an early drying stage, to
the difference in the amount of cement since the water content was kept constant at
185 kg/m3. The drying shrinkage after the early stage may be controlled in
combination with the development of strength and shrinkage stress; This is a
general trend similarly observed in normal concrete (6).
2--Effects of chemical admixture
The effects of the type of chemical admixture on the drying shrinkage
behavior of the flowing concrete are shown in Fig.8 (b). A flowing concrete using
the special admixture A showed greater drying shrinkage than that of a control
normal concrete using AE agent. No significant difference in drying shrinkage
between normal and flowing concrete each using AE water reducing agent was
observed. The drying shrinkage of each concrete ranged from 6 to 7.5 x 10-4 at the
age of 26 weeks, and it is evident that flowing concrete shows satisfactory drying
shrinkage in comparison to a standard value of 8 x specified in JASS 5
commentary.
3--Effects of inorganic admixture
The effects of the type of inorganic admixture upon drying shrinkage are
shown in Fig.S(c). No significant difference could be found between them, and the
shrinkage was approximately 7.5 x The mixture in which cement was
substituted by 30% of blast furnace slag showed relatively greater shrinkage than
others at the early stage of drying. For the mixture in which fine aggregate was
substituted by 10% of lime stone powder, an improvement of shrinkage was
expected, however, no significant effect could be observed in the long term.
4--Effects of the aggregate type
The effects of the type of aggregate upon drying shrinkage are shown in
Fig.S(d). Drying shrinkage was relatively small when river sand and river gravel
were used, and was 7 x after 26 weeks of drying. However, drying shrinkage
increased when crushed stone was used as a coarse aggregate. This tendency was
particularly observed when the maximum size of the coarse aggregate became
smaller, which may be attributed to the difference in curing condition (the concrete
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
I
A C 1 SP-148 94 Obh2949 0539984 5 3 3 =
~
214 Kasai et al
with river sand and river gravel was subjected to standard curing up to one week of
age), difference in the intrinsic properties of river gravel and crushed stone, and in
the amount of the special admixture.
5--Loss of mass and drying shrinkage
The relationships between loss of mass and drying shrinkage are shown in
Fig.8 (e) and (f). The drying shrinkage showed a high correlation with loss of
mass irrespective of the water-cement ratio, and their gradients became steeper with
smaller water-cement ratio. This means that large shrinkage may be produced with ,
a small amount of water loss. The condition of curing at an early age may also
affect significantly the relationship between loss of mass and drying shrinkage.
Carbonation
The relationships between duration of accelerated carbonation curing and depth of
carbonation are shown in Fig.9. The carbonation depth of flowing concrete ranged
from 10.6 to 13.5 mm after 13 weeks of the curing duration , and showed no
significant difference with that of AE concrete without any special admixture.
Freezing and Thawing Resistance
A typical result of the freezing and thawing test is shown in Fig. 10. The relative
dynamic modulus of elasticity of flowing concrete at 300 cycles ranged from 93 to
102 % and the loss of mass of flowing concrete ranged from 0.09 to 2.54%. These
results are almost similar to that of AE concrete without any special admixture.
The relationship between water-cement ratio, the loss of mass and the durability
factor is shown in Fig.1 1. With a decrease in water-cement ratio, the durability
factor increased while the loss of mass showed a decreasing tendency.
The relationship between spacing factor and durability factor is shown in Fig. 12.
The spacing factor of flowing concrete was 229 to 262 micrometer, and showed no
significant difference to that of AE concrete without any special admixture.
CONCLUSIONS
This experimental study was aimed to redesign the wet consistency concrete
used in constructions as Flowing Concrete by means of special admixtures.
Mixture proportion, flowability, strength, shrinkage, carbonation and freezing and
thawing resistance were evaluated in the laboratory. The resulting characteristics
of the flowing concrete were as follows.
(1) The slump of the flowing concrete that could be placed with slight compaction
ranges from 24 to 26 cm, which corresponds to a flow from 50 to 60 cm and a
differential height in the box test less than 8 cm.
(2) Mixture proportions of the flowing concrete with a water-cement ratio from
30% to 60% can be made by using a new admixture and by a simple correction
of the standard table of the mixture proportion for the conventional wet
consistency concrete.
(3) Flowing concrete can be produced with a specified concrete strength ranging
from 18 to 60 MPa.
(4)The strength, shrinkage, carbonation and the freezing and thawing resistance of
flowing concrete showed no significant difference from that of AE concrete
without any special admixtures.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 m 0662747 0519985 4 7 T m
REFERENCES
1 Ozawa, K. Maekawa, K. and Okamura,Y. “Development of high performance

concrete”; Proc. Ann. Meeting JCI, 11(1), pp. 699-704, 1989. (in Japanese)
2 Sahara, Y., Shoji, Y. and Takeshita, H. A study for the workability test and
“
compaction of the highly flowable concrete”; Summary of Prep. Ann. Meeting

A U (A), pp. 941-942, 1991. (in Japanese)
3 Manual of the under water, non-segregation concrete (design and

construction), The Technical Center for the Seaside Development and The
Technical Laboratory for Harbor and Fisherman Community. (in Japanese)
4 JASS 5 (Japanese Architectural Standard Specification, 5: Reinforced concrete
work ), pp. 112-119, 1957. (in Japanese)
5 Wami, H., Yoda, K., Yanagida, K. and Abe, T. “Experimental study on the
flow properties of high strength concrete”; Summary of Prep. Ann. Meeting
i AIJ (A), pp. 927-928, 1991. (in Japanese)
-1
6 Hisaka, M. et. al. “Effect of water-cement ratio on the drying shrinkage
cracking of concrete”; Trans. Cem. Concr., 44, pp. 678-683, 1990. (in
Japanese)
7 Kasai, Y. et al. “Experimental study on the flowing concrete (Part 1 - 7)”;
Summary of Prep. Ann. Meeting AIJ, pp. 937-950, 1992. (in Japanese)
8 Kasai, Y. et al. “Experimental study on the flowing concrete (Part 8 - 13)”;
Summary of Prep. Ann. Meeting AIJ , pp. 1161-1172, 1993. (in Japanese)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662749 0 5 L 9 9 8 b 30b
216 Kasai et al
TABLE 1 - CEMENT AND ADMIXTURES USED FOR THE TEST
Mater ials Sort and nature

Normal portland cement
Cement
(s. a. 3270cm2/g, s. g. 3.16)
Blast-furnace slag (s. a. 4050cm2/g, s. g. 2.92)
Mineral Fly ash (s. a. 3000cm2/g, s. g. 2.20)
admixtures Lime stone powder (s. a. 3140cmz/g, s. g. 2.72)
Expansive admixture (s. a. 2550cm2/g, s. g. 2.92, CSA)
AE agent (Natural resin acid)
Chemical AE water reducing agent (Lignin sulfonate)
admixtures Special admixture A (Naphthalene sulfonate and
specific water-soluble polymer)
Special admixture B (Poly-Carboxylate and
sDecif ic water-soluble Dolvmer)
I
\-
TABLE 2 - AGGREGATES USED FOR THE TEST
Absorption
Type of aggregate
%
Fine River sand 2. 5 2.58 1.70 2. 76 67. ‘I
I 2:6:
2. 5 2. 56 i.48 2. 85 63. 6
Coar se I R i v e r gravel
1. 10
o. 59
sand stone 2. 62 O. 66
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 Ob62949 0519987 242
l
v
*
f
I
I I
o o o o o o o o o m m o o o o o o o o o o o o o o o o ~
N N N N N N N N N - - N N N N N N N N N N N N N N N N
I IP
I I I I .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 O b b 2 9 4 9 0519988 I489
218 Kasai et al
TABLE 4 - TEST RESULTS
Sarias EXP. slunip

-Flow
P.
., ert ies of-fresh conci --
e
Air
Propertias of
I concrete
NO. No. fe strength
M I
1-1
-
cm -
cm
22. o 42 x 41
-%
4.2
-
nelcm'
o. 20
Iday W a y iday
- - 17. 9 26. 8
1-2 24. 5 49 x 49 4. 6 o. 23 - - 17. 8 28.8
1 1-3 25. o 51 X 51 3.6 O. 06 - - 17. 8 29.8
1-4 25. o 54 x 53 3. 7 O. 15 - - 18. 4 30.a
1-5 24. 5 57 x 54 6. O 4. o o. 10 - - 13.4 28. 7
1-6 25. o 57 x 55
- 3. o -
3. 8 o. 12 - - 12.7 26. 9
2- 120. o 29 x 29 18. o X 5. 3 O. 15 3.4 11.4 17. 1 28. o
2-2 23.5 50 X 46 10. o X 5.2 O. 13 2.8 14.3 20.6 31. O
2-3 25. 5 60 X 57 8. o 5.5 O. 13 2.8 12.7 18. 1 26.9
2 2-4 21. o 45 x 44 16. 0 X 5.5 o. 20 1.9 13.2 19. 6 30. O
2-5 24. o 56 53 7. o Q 4. 5 o. 27 32. 8
j i
2.4 13.8 22. 3
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
23. o 44 x 42 5. O o. o9 17.4 25.5 35. 1
2-6
2-7
3- 1
-
25.5 48 x 55
25.5 60 X 59
'!:oo.0 o -
4.5
3. 1
o. 11
O. 17
3.5
3.4
1.4
16.9
10.4 19.2
33.3
28. 7
3 3-2
3-3
3-4 -
25. o 55 x 54
23. 5 53 x 49
23. o 46 X 46
1.0
43: o I
I 0 3.2
4. 8
-
4. 7
O. 17
o. 12
o. 18
2.3
1.4
2.5
10.8 18.6
10. 5 18. 6
11.0 -19.3
28.5
28. 6
27.5
4
4- 1 24.5 E i 5 5
4-2
4-3
4-4
25. o 63 X 60
24. 5 54 x 53
26.5 68 x 67
27. o 66 X 64
4.
0::o.5
I O
Q
A
4. o
4. 1
4.7
4.9
o. 22
O. 17
O. 03
o. o1
o. o1
3. 7
6. 1
IO. 8
!2. 9
12.8 18. 1
20.5 29.7
28. 5 37. 9
43. 8 56. 6
29. 7
37.7
48. O
66. 7
4-5 - -
4. 4 9. 1 32. 9 -
44. 7 -
64. 2
5- 1 24. 5 53 x 50 4. 1 o. 20 3. o 12. 5 19.7 30.9
5-2 24. 5 52 x 51 4. o o. 11 5.2 18.3 29.7 41.8
5 5-3 24. o 49 x 49 1. o 4. o o. 02 o. 1 19.3 44.2 51. 1
1) Temperature
5-4
15-5 -
25.5 55 x 55
26. O 64 X 63 ::2
f resi concretc was 18-20%.
I 4. 7
-
3. 9 -
o. o1
o. o2
!O. 8
9. 1
46. 7 63. 4
- 67.8
29.9 50. 1
78.4
2) Evaluation of f l o w a b i l i t y was made form slump 24-26cn, flow 50-60cm and

Box d i f f e r e n t i a l height of concrete less than 8cm.
8 :S a t i s f i e d 3 Subjects. 0 : S a t i s f i e d 2 SUbJeCtS out of 3 .
A : S a t i s f i e d 1 subject out of 3. X : No subjects were s a t i s f ied.
A C 1 SP-148 9 4 I0 6 6 2 9 4 9 0 5 3 9 9 8 9 015
TABLE 5 - THE CORRECTION MITHOD OF CONCRETE MIXTURE ACCORDING TO THE TYPE

OF AGGREGATE (CORRECTION FROM THE STANDARD TABLE)
Type o f aggregate I C o r r e c t i o n from r i v e r sand and g r a v e l

Fine
aggregates I Coarse
aggregates I U n i t water
content I I Spec i f i c
adm i x t u r e
R i v e r sand Crushed stone *O + 4.9% +6.7%
Crushed sand (30%) Crushed stone f0 +5.8% +8.3%
Crushed sand (50%) Crushed stone I I *O I +6.4% I +15.0%
TABLE 6 - NORMALIZED COMPRESSIVE STRENGTH OF DIFFERENT CONCRETES
Type o f c o n c r e t e
I R a t i o o f compressive
s t r eng t h (W/C=60%)
Aggregates Admixtures lday 3days 7days 28days
R i v e r sand * River gravel - - - 0.60 1.00
R i v e r sand - River gravel Slag 21 - - O. 45 0.96
R i v e r sand - Crushed stone - O. O9 O. 48 O. 69 1. 04
River sand * Crushed stone L.s.P.~’ 0.12 0.58 0.86 1.18
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
R i v e r sand * Crushed stone(l3mm) - O. 06 O. 44 O. 66 1. O0

R i v e r sand Crushed stone AE agent O. 12 O. 39 O. 58 O. 96
Crushed sand * Crushed stone - O. 05 O. 35 O. 62 O. 96
Fig. 1-Box test apparatus (made of acrylic resin)
26
- W/C=60%
* 24
a
s River sand.
- Crushed stone admix ture
s
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
,+
7
High range
AE water-reducing agent
,-& 20 I I
5 24 I
( )Amount of admix trires (kg/m”
I I
c>
c
.?i
+ (3. 7)
$ 8 -
- Spec ia1
Li admix ture
0
+4
+4
.?i
I
(11. i)
“ O
m 30 40 50 60 70
Flow (cm)
Fig. 2-Preliminary test results (relations between flow and slump, box differential height)
A C 1 SP-148 74 m Obb27Li7 051999L 773 m
30 I 1
A
e
o
W/C=6O%, Admixtures=2. Okg/ni3
v River sand-Crushed stone
20
W = 175 k g / d
\
10 -
-,i
P
x
O
m
O I
30 40 50 GO
Flow (cm)
Fig. LRelationship between flow and box differential height
( a > Slump [ b ) Flow

24 I I 1 I I 1
DM River s a n d s c r u s h e d s t o n e
M
AE
'G
c
16
0 AE O AE
DAF,
O ion n
20 22 21 6 30 40 50 60
Slump (cm) Flow (cm)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. LRelotionship between slump, flow, and the box differential height
A C 1 S P - I 1 4 8 74 0662949 0529992 bOT
222 Kasai et al
O.
-
2 -
IIAE
O AE
a
OAE -
o
AA A
- -
I-
O
0 AE
O AE
AEW -
-
O D
ci o. 1 - U
O
u
-
u
- o - - O -
I
O I I I i
Fig. SRelationship between slump, box differential height, and amount of bleeding
n
N
8
--.
O
9)
0
. River sandscrushed stone
R
v
+I
O
30 40 50 60
Fig. &Relationship between water-cement ratio and amount of bleeding
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0 h b 2 9 4 9 0539993 5 4 6
80
2 60
æ
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
5M
c
c> 40
rn
al
b-
.r(
v)
v)
8e 20
O
O
O
C/W 1.5 2.o 2.5 3. o 3. 5
W/C(%) 60 50 40 30
Fig. 7-Relationship between VW and compressive strength
(a) WIC
10
River
.~~... sandsgiver gravel
a9 .. . . . . ..
h
c
.. .. .
O . ~ ~~ ~~ ~~ ~~ ~ ~ u. >
- 4 8 rb
X
- 7
o>
as
a 6
-!d
.z
Li
5
c
* 4
M
.c;h 3
& 2
1
O
Fig. Ea-Effects of W/C on drying shrinkage y L a s

2.5U ,zc
0.4
o>E
jg
0.d
aie
W f ad am
<-o m m m m
Fig. Eb-Effedr of the type of

chemical admixture on drying shrinkage
A C 1 SP-148 94 0 6 6 2 9 4 9 0519994 4 8 2
224 Kasai et al
(c) Inorganic admixture

I1..W/C=60%
...................... ..................
I
River sand. River sand.
ed stone Age
......
Q) $+
s al
2 a
e o
alB
o a
w
w
al
R
.rl
-1
Fig. &-Effects of the type of

inorganic admixture on drying shrinkage
Fig. Ed-Effects of the type of
aggregate on drying shrinkage
(e) Effect of W/C

101 ' I * I ' 1 . I
A
-f
O
?I
x
v
M
C
.ri
2
n
O 1 2 3 4 O 1 2 3 4
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Loss of mass (%I Loss of mass ( 9 6 )
Fig.ponhsiitaR&le?- between loss Fig. If-Relationship between loss

of moss and drying shrinkage of mass and drying shrinkage
(effect of W/C) (effect of curing)
A C 1 SP-148 94 Ob62949 O539995 3 3 9 D
O 4 8 13
Accelerated duration (week)
Fig. 9-Relotionship between the accelerated durotion of CO, gos and the depth of corbonation
r I I I l I I I
3- .......*.
95
90 I l
1/C
I
-
60%
I
60% 50% 10% 30%
+ ...A... <+ ...o...
I I I I
30?4
+
l
-r
O 60 120 180 240 300

Number of freezing and thawing (cycle)
Fig. lû-Results of freezing ond thawing test
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 I0 6 6 2 9 4 7 0519996 255 =
226 Kasai et al
n
x
v
v 1 1 -
v1
2
y-i
O
v)
2 - River sand-Crushed stone
O
v) Special admixture B
a
30 40 50 60
w/c (%I
Fig. 114elotionship between water-cement ratio, loss of mass, and the durability factor
I I I I I I 1 I
- O
100 -
o (Replaced 30% Fig ash -
n
-
x
W
LI
- River gravel
O
c>
- 0
0. River sand-
U
- €!I
Q
CH AE
95 -
c.>
.d
,-l
- rn AE -
.r(
n - u7 e A E
m
LI
-
1
n
- O
I I I I I I I I I
90
Fig. li-Relationship between spacing factor and durability factor
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0539997 191 M
SP 148-13
An Experimental Construction of Model

Structure with Flowing Concrete
by N. Nishiyama, Y. Kasai, N. Yuasa, and Y. Nakata
Synopsis: Flowing concrete is characterized by high flowability, requiring

only slight consolidation by vibrating and easy control in a plant. An experimental
model structure using flowing concrete was constructed. This paper deals with
placing capability on site and properties of hardened flowing concrete in the
structure.
Pumpability, flowability, and capability to fill forms were investigated for
the fresh flowing concrete and distribution of compressive strength, carbonation
depth, cement content, air permeability, and water absorption were measured for
the hardened flowing concrete.
Compressive strength of core samples taken from the model structure and
standard cylinder specimens from the plant mixture were approximately 24 MPa.
The average estimated cement content was 333 kg/m3 and the standard deviation
was 15 Kg/m3 within a wall of 3 x 4.2 m. Measured carbonation depth was
smaller in the freely flowed parts than those in the upper portion of vibrated parts.
Keywords: Absorption; carbonation; compressive strength; flowabilitv; models; nondestructive tests;

permeability; tests
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---

A C 1 SP-148 74 = 0662949 0539998 028
228 Nishiyama et al
N Nishiyama is a chief research engineer of technical research laboratory of

Nishimatsu Construction Co., Yamato-si, Japan. He has been in charge of
research into high performance concrete as well as plant management.
Y. Kasai is a professor of building materials at Nihon university, College of
Industrial Technology. He has been in charge of various field of building
technology including durability of concrete, non-destructive testing of concrete,
and recycling of construction by-products.
N. Yuasa is an assistant of Nihon university, College of Industrial Technology.
He has been in charge of NDT study as well as the surface and cover zone
analysis of concrete for the durability of concrete
Y. Nakata is a research engineer of technical research laboratory of Nishimatsu
Construction Co. He has been in charge of the study in the estimation of cement
content of hardened concrete by means of using gluconic acid.
INTRODUCTION
Reflecting demands for higher productivity, mechanization of construction

works for reinforced concrete structures has been increasing during the past 25
years. As was seen in the change from a bucket and a buggy to a pump truck,
labor saving have been mainly realized by mechanization, and that the concrete
itself shows no advance except for those of superplasticizers. Development of a
new concrete to meet the requirement for simplification and higher quality without
consolidation has been studied in the flowing concrete technical committee.
Flowing concrete is characterized by high flowability and sufficient durability
without increasing the unit amount of water, which was made possible by a newly
developed superplasticizer.
The present paper deals with an experimental construction of full-scale
model structure with flowing concrete, the results of test for properties of
hardened concrete, and discussion of the applicability to practical constructions.
Quality control techniques, pumping capability, rapid measurement in the
construction of a full-scale structure are described, and compressive strength
evaluation using core samples, estimation of cement content, rapid measurement
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
techniques for carbonation depth, air permeability, and water absorption are
applied to confirm the quality of hardened concrete in the model structure.
METHOD OF EXPERIMENTS
Construction of full-scale model structure with flowinp concrete

1-- The full-scale model structure
The full-scale reinforced concrete structure comprises two columns with
dimensions of 700 x 700 x 3000 mm and walls with thickness of 150 mm as
shown in Fig.1 and 2. Reinforcing steel bars for the column were 4-D22 for the
main reinforcement and D22 with a pitch of 100 mm for the hoop reinforcement,
A C 1 SP-14ô 94 = OSuperplasticizers
6 6 2 9 4 9 0519799 T 6 4
and Chemical Admixtures 229
and D13 for vertical and D10 for horizontal reinforcement each with a pitch of 200
mm were arranged for the wall reinforcement keeping 30 mm for a cover
thickness. An internal partition was set at the center of the opening to test two
different mixtures, and the form was made of semi-transparent resin to be able to
observe the flow behavior of concrete.
2--Mixture proportions of flowins concrete
The mixture proportions of flowing concrete are shown in Table-1. A
preliminary batch using two admixtures was executed to determine the mixture
proportions for the full-scale construction.
3--Manufacture and transportation of flowing concrete
The flowing concrete was mixed by a tandem forced mixer with a capacity
of 3 m3 in a concrete plant, and the admixture was manually batched.
Transportation was made by agitating trucks with 6 m3 capacity, and took 30
minutes from mixing to placing.
4--Test of quality for flowing concrete
Slump, flow, air content, temperature of concrete, and box test described
below were executed to confirm the quality of fresh flowing concrete on arrival of
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
the agitating truck. Sampling was made both at discharge and at the discharge of a
pump truck. A concrete specimen of approximately 10 litres was set in a half of
150 x 150 x 400 mm of box test apparatus shown in Fig.3, then the shutter was
opened up and the differential height between two parts was measured. Control of
compressive strength was made for specimens both under the standard curing and
on-site curing at the age of 28 days.
5--Placing of flowing concrete
Flowing concrete was placed, as shown in Photo.1, by the squeeze-type
concrete pump truck at points indicatcd in Fig.1. Pumping capability, angles of
repose and lateral pressure against formwork were examined. The lateral pressure
against formwork was measured primarily using strain gauges attached to the
separators shown in Fig. 1 , and then converted into pressure.
Properties of hardened concrete taken from the structure

1--Compressive strength of cores and estimation of cement content
Cores were taken from the points in the model structure as indicated in
Fig.4, and compressive strength was measured according to JIS A 1107 at the age
of 28 days. These specimens were also subjected to the estimation of cement
content by the method using sodium gluconic acid.
The test procedure is, as shown in Fig.5, that a 2.000 g of powdered
concrete (So) is dissolved in a 15 %, 300 mL of sodium gluconic acid for 30
minutes under constant temperature of 60 "C and the amount of residue (R) is
measured. An estimated ceniei;t content is then calculated by a following formula,
A C 1 SP-L4ô 9 4 H 0 6 6 2 9 4 9 0 5 2 0 0 0 0 L 7 T
230 Nishiyama et al
and
where Co is a dissolution rate of powdered concrete in %, Wc is a loss of

powdered concrete at 500 "C, So is a mass of powdered concrete after heating to
500 OC, R is a residue of powdered concrete, Sd is an absolute dry mass of
concrete, V is a volume of specimen and Cm is an estimated cement content.
2--Rapid tests for carbonation depth. air permeability. and water absorption
Rapid tests for carbonation depth, air permeability and water absorption
were executed at 5 months after placing to confirm the quality of hardened
concrete. Sampling points were determined to have sufficiently small distance for
both column and wall as shown in Fig.1, and drilled to have a hole, which was 10
mm in diameter and shared ror every tests.
Rapid test for carbonation depth was made by using concrete powder
produced by drilling. A filter paper with phenolphthalein solution was put on a
plastic substrate, and slid so as to accept new concrete powder. When the powder
first showed red, the drilling was stopped to measure the depth of the hole by a
slide calipers as shown in Fig.6.
Rapid test for air permeability was executed with the same hole as used in
the carbonation test as shown in Fig.7. The hole with a diameter of 10 mm and a
depth of 80 mm was sealed by a cap of silicon rubber, and evacuated by a vacuum
pump down to X I Pa through a needle for injection set in the rubber cap.
Subsequently , a time T (sec.) for X I (Pa) to reach an equilibrium pressure x2 (Pa)
by incoming air through the wall o i the hole was measured. A rapid air
permeability K (Paísec.) could be calculated by the following formula ( 3 ) ,
representing the permeability of concrete to air.
where X I and x2 were typically 2.1 x104 and 2.5 x lo4 Pa respectively while 1.3
and 3.3 x lo4 Pa were used when T was as short as 10 second.
Rapid test for water absorption was executed immediately after the air
permeability test making use of the rubber cap as shown in Fig.8. The second
injection needle was put into the cap and connected to a measuring pipette. Water
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
was introduced to the drilled hole through the first injection needle used for the air
permeability test, and a time required for the subsequent decrease of the water for
0.01 mL in the measuring pipette due to absorption from the concrete wall was
measured as T in second. The amount of absorption W (mL) can be set for 0.03
ml when T was as short as 10 second. The rapid water absorption rate Q (musec)
can be calculated by the following formula (4),and taken for representing the
permeability of concrete to water.
A C 1 SP-14ô 9 4 I0bb2949 0 5 2 0 0 0 1 O06
Construction of full-scale model structure with flowinp concrete

1--Test of quality for flowing concrete
Quality of fresh flowing concrete tested at the plant, at discharge on site, and
at the discharge of pump is shown in Table-2. Regardless of an effect of
transportation, slump, air content, and box differential height remained within the
acceptable range of specified quality of flowing concrete with slump from 24 to
26 cm, flow from 50 to 60 cm, air content of 4k1%, and box differential height as
low as 8 cm, though a slight increase in slump-flow at the discharge was
observed. Difference in compressive strength at the age of 28 days was not
observed regardless of tests executed in plant, at discharge on site and at the
discharge of pump. These results show that flowing concrete can be produced in
the conventional concrete plant.
2--Placing of flowing concrete
The angles of repose of flowing concrete exhibited gently sloped shape in
general and became more gentle as they approached to an end point as shown in
Fig.4 and Photo.2, though a slight difference between mixtures of A and B was
observed. No vibrator was needed in the placing for a height up to 2 m owing to
an excellent flowability of admixtures A and B. The time taken for placing
concrete of one step (375 mm in the height of wall) without vibration was 4 - 6
minutes. The distribution of lateral pressure on the formwork as shown in Fig.9
was changing linearly along with the vertical direction, and exhibited sufficient
flowability of flowing concrete. Pumping was also successfully done without an
excessive charge in the same way as that of conventional concrete with a slump of
180 mm.
Properties of hardened flowinp concrete taken from the structure

1--Cornmessive strength of cores and estimation of cement content
Results of compressive strength of cores and estimation of cement content
are shown in Table-3. Average compressive strength of A and B wall was 26.0
MPa and 22.3 MPa respectively, which was nearly equal to that of standard cured
specimens. The difference in compressive strength between A and B may be
attributed to an effect of solar radiation. The estimated cement content showcd
good agreement with the designed value (312 kg/m3), and no substantial
difference along with the height. Flowing concrete in the test walls was proven to
flow without segregation.
2--Rapid test for carbonation depth, air permeability, and water absorption
Results of rapid test for carbonation depth, air permeability, and water
absorption are shown in Fig.10, 11, and 12, and the numerical data are shown in
Table.4. Evolution of carbonation in A wall was slower in a part where the
concrete flowed by itself than the other part where the vibrator was applied. Air
permeability and water-absorption rate showed higher value at the central part of
wall where the concrete flowed by itself.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 7 4 S 0662947 0520002 T 4 2 S
232 Nishiyama et al
CONCLUSIONS
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A full-scale model structure with the flowing concrete was experimentally

constructed. The followings conclusions can be drawn.
(1) Quality of flowing concrete can be controlled sufficiently in the conventional

concrete plants.
(2) Capability of flowing concrete was proven to be so satisfactory as to fill a
double reinforcement wall with a thickness of 150 mm.
(3) Surface finishing was easier than normal concrete with a slump of 180 mm
owing to its low viscosity and high flowability.
(4) Flowing concrete can be placed solely with a slight vibration.
(5) Flowing concrete can be conveyed by the conventional pumps with less
pumped load than for normal concrete with a slump of 180 mm.
(6) Quality of hardened flowing concrete showed little variance and was stable.
An experimental construction of full-scale model structure with the flowing
concrete has proven its promising future and feasibility using existing plant
facilities, site facilities, and quality control techniques.
REFERENCES
1 Kasai, Y. et al. “Experimental Study of Flowing Concrete (Part 1 - 7)”;

Summary of Technical Papers of Annual Meeting, Architectural Institute of
Japan, pp. 937-950, 1993. (in Japanese)
2 Kasai, Y. et al. “Experimental Study of Flowing Concrete (Part 8 - 13)”;

Summary of Technical Papers of Annual Meeting, Architectural Institute of
3 Yuasa, N. et. al. “Testing Method for Permeability and Cement Content of
Structural Concrete”; Proc. Int. Symp. Non-destructive Testing and Stress-
Strain Measurement, FENDT92, Tokyo, pp. 485-492, 1992
4 Kasai, Y. et al. “A Proposal for the Evaluation Method of Unit Amount of
Cement in the Hardencd Concrete by Means of Sodium Gluconate Acid”;
Proceedings of the Japan Concrete Institute, Vol. 12 , pp. 264-269, 1990 (in
Japanese)
5 Nakata, Y. et al. “Distribution of Unit Amount of Cement in the Experimental

Full-scale Model Structure”; Prep. Ann. Meeting CAJ, pp. 270-273, 1993 (in
Japanese)
A C 1 SP-148 9 4 0 6 6 2 9 4 9 0520003 9 8 9
m
d
9
o>
o
g
c
8
iLi
c
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
<cr
O
* -
O 6
.-eM
i
- o
u -
e>
P
L
u /I P.
,/81
- Lo
.-8I.-3x
e>=
us
c-u
z
ô
A C 1 SP-148 9 4 I 0 6 6 2 9 Y 9 0520004 815
234 Nisniyama et ai
TABLE 3 - RESULTS OF COMPRESSIVE STRENGTH OF CORES AND CEMENT CONTENT TESTS
Core Height Compressive Estimated cement

Wal 1 sample strength content
No. ( mm 1 ( MPa ) ( kdm3 )
2830 23.3 320

1500 27.1 329
700 23.3 307
2750 28.5 348
A 1470 25.6 315
720 26.6 323
2690 24.2 352
1550 27.1 304
470 28.4 349
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
2630 20.2 307
1450 21.9 324
660 20.6 356
2610 19.7 320
B 1440 24.7 319
650 23.5 339
2550 22.6 306
1450 24.9 329
640 22.6 332
A C 1 SP-148 9 4 I0662949 0520005 7 5 1 - =
TABLE 4 - RESULTS OF CARBONATION DEPTH, AIR-PERMEABILITY SPEED AND

WATER-ABSORPTION SPEED
- - - ubairticadeph ir-penneabiliiy rpx¿ 'mer-atsorptionspec<
No. X Y
1
-15
-
289
(mm)
2.2
(mmHg I sec)
0.516
X i U ' (ml/rec)
9.10
2 15 27 5 3.0 9.677 ~1.93
3 15 2a5 3.1 1.422 13.85
4 15 17 5 1.5 6.383 14.27
.. 5 .....
15
15
.....
130
75
.... a:>......
1.3
- ___ ..Q~906-..
0.816
___ .......MI? I....
11.55
7 15 25 0.6 0.667 11.95
8 15 10 0.4 0.365 (.ia
9 75 289 2.0 O. 564 9.11
..XP .7.5 - 175. ... 2.>6 ...... ....Q. ,? 3.0 ...... .......&J...1 7 . 1?I5 .....
11 75 ais 1.6 O. 531
ia 80 17 5 1.1 1.395 3.16
13 80 130 0.8 0.744 16.61
14 80 75 1.2 1.875 18.38
15
16
80
.....
80
25
.....
10
..............
1.7
1.5
.................. ..................
0.567
0.704
7.75
9.57
17 135 289 2.0 0.691 10.03
18 135 27 O 2.5 0.459 13.42
19 125 225 2.2 1.538 19.16
.....
20
21
??>. -
125
......
17 5
13 O
...............0.4
0.5
...........2.....0...9 8
1.038
........1.0...2..
i7.9a
6.....
22 125 75 1.6 1.282 14.99

23 125 25 1.0 0.530 10.83
24 175 10 0.0 0.831 9.09
.____
25
26
175
17 5
...289
...
27 O
...............1 . 5
2.2
.......0....3.7..4.....
O. 682
..................
13.97
15.67
27 17 O 22s 1.0 2.239 ~3.09
28 170 175 0.0 - 13.61
29 170 130 0.1 0.574 6.71
.....
30 ......
17 O ......
70 ..............
0.0 ..................
0.485 7.83
..................
13.61
31 110 30 1.3 O . 806
31 170 10 1.1 o, a44 4.56
33 235 289 0.6 0.530 10.62
34 135 270 0.6 1.224 11.04
35
.....
36
......
235
12s
215
.....
170
.....0 . 8
0.4
..... ..................
1.478
O . 501
..................
3.53
11.98
37 225 125 0.3 1.195 13.89
38 225 75 2.0 1.060 8.83
39 225 15 0.7 0.651 10.22
..?O ......
a25 .....
10 ..........
0.6 ..........
0.4 9
0.1 8
....... .................
4.96
3.43
41 27 5 289 0.7 . .
42 27 5 270 0.5 0.4 2 7.15
43 275 225 1.0 0.4 3 4.72
44 280 170 0.9 - 11.04
....45
. 280
..... .A?! ..............
1.1 ..................
1.579 .................. 0.00
46 mo 75 1.9 1.351 19.96
41 a80 25 0.6 0.693 9.40
48 280 10 0.8 0.450 3.44
49 335 289 0.2 0.920 8.58
--.. .E?.
5c
51 335
...170
...
225
..............
0.3
0.6
O. 458
..................
1.230
.................. 9.02
7.32
sa 340 170 0.9 1.685 14.73
53 340 115 0.6 0.815 8.86
54 340 15 1.1 0.785 8.47
.... 55 340
..... 25
..... ..............
0.0 &.... :O+? ..... .................. 2 . 3 6
1
5( 340 10 0.5 0.339 0.00
51 375 289 0.5
. . 0.501 7.11
5f 375 27 O 0.4 - . 2~
1 40 . . 8.U
51 375 225 1.0 1.456 11.49
.....6 i 380
..... 17 O
..... ......0... .1..... .......0 : w ..... ................
8.79
61 380 125 0.0 1.563 15.48
62 370 15 0.6 1.786 11.48
6: 360 75 0.0 O . 680 9.19
64 380 25 0.8 0.806 12.36
....
6! 380
..... 10
..... 1.1
.............. 0.502
..................
1.199
..................
4.14
9.27
61 425 289 0.1
61 425 270 0.6 0.562 4.47
61 415 235 1.4 1.261 7.86
61 425 17 O 0.0 0.759 7.07
__..
71
71
425
425
i a 5 ......0..
._.._
75
..
9.....
1.2
..................
0.785
1.463
..........1..8.47
......
6.98
7: 425 25 0.7 1.714 13.40
7: - -
41s 10 0.9 0.609 10.22
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9Y 0662749 O520006 698 W
236 Nishiyama et al
-
z
mncrete discharge
Place of measured
I
lateral pressure
1 Ii
I Surface
Wall
Vertical-bar DlWZGü
Horizontal-bar DlC@ZOO
A
t
r
Separator
A wall
I@*
I
I
I
Fig. 1-Plan of model structure
150
n
\ /-
Separator Partition Separator
r-l opening O
9
m
I I Base mncrete
2.500 2.500
i
Fig. 2 4 - A section of model structure
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 I0662749 0520007 5 2 4 m
t Sliding shutter
height
Fig. L B o x test apparatus (made of acrylic resin)

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. d b s e r v e d angles of repose and locations of coring
A C 1 SP-148 94 m 0bb2949 0520008 460
238 Nishiyama et al
- Y--
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 94 Ob62949 0520007 3T7
Filter paper soaked

with phenolphtnalein
alcohol liquid
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. L a r b o n a t i o n depth test by drilling method
........
. .. n , . . . . . I
..........
Vacuum pump
80 (mm) manometer
Fig. 7-Rapid air-permeability test apparatus
A C 1 SP-148 94 0662747 0 5 2 0 0 1 0 O19 9
240 Nishiyama et al
silicone cap Injection needle
... .r,. .. . ..
Pipette
I Pipe
B 80 (mm)
Fig. &Rapid water-absorption test apparatus
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O
( 4900 9800 14700 19600 24
internai pressure (Pa)
Fig. 9 4 n t e r n a l pressure of concrete in form
A C 1 SP-148 94 0662947 052001L T55
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Photo 1-Condition of concrete placement
Photo 2-ûbserved angles of repose
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 m 0662949 0520033 828 m
SP 148-14
Shrinkage and Cracking of High-

Strength Concrete and Flowing
Concrete at Early Ages
by K. Yokoyama, S. Hiraishi,
Y. Kasai, and K. Kishitani
Svnopsis: The present paper deals with shrinkage and cracking at early ages of
highly flowable flowing concrete with a special admixture (unit amount of water
of 185 kg/m3) and of normal high strength, viscous concrete with high-range AE
water reducing agents (unit amount of water of 170 kg/m3) under constant air
flow.
The results are as follows.
(1) Shrinkage at early ages of flowing concrete and normal high strength concrete
became greater with a decrease in water - cement ratio. Shrinkage of normal
high strength concrete varied according to the composition of admixture.
(2) The cracking of both flowing concrete and normal high-strength concrete
within a restricting steel frame initiated at an earlier age and became remarkable
with a decrease in water - cement ratio. The time and amount of crack initiation
of normal high strength concrete varied according to the composition of
admixtures.
Keywords: Admixtures; air entraining agents; crackina (fracturinq); crack propagation; flowability;
high strength concretes; shrinkaae; water reducing agents
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---

243 Document provided by IHS Licensee=Aramco HQ/9980755100, User=, 07/23/2003
A C 1 SP-146 94 = 0662949 0520014 764
244 Yokoyama et al
K. Yokoyama is a professor of Junior College of Technology, Nihon

University. He has been in charge of study in the cracking of concrete at early
ages.
S. Hiraishi is a research manager of technical development division at Yamaso
Chemicals Co. He has been in charge of research and development in chemical
admixtures for concrete.
Y. Kasai is a professor of building technology at Nihon university, College of
Industrial Technology. He has been in charge of study in various field of building
technology including durability of concrete, non-destructive testing of concrete
and recycling of construction by-products.
K. Kishitani is a professor of building materials at College of Science and
Technology, Nihon University. He has been in charge of study in the durability of
building materials.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
INTRODUCTION
The rationalization of on-site labor works in reinforced concrete

construction has been widely studied for years in terms of labor-saving and quality
improvement by means of compaction-free concrete ( 1-2). Experimental study on
flowing concrete using a special admixture for high flowability has been made on
the basis of an experience of wet consistency concrete with a slump of 24 cm once
used in the building construction in Japan.
The properties and the full-scale test construction of flowing concrete with a
designed strength from 18 to 24 MPa (3-4) and 18 to 60 MPa (5) have been
reported elsewhere. The control parameter for flowing concrete to have a sufficient
fill-up capability with a slight compaction is known as a slump from 24 to 26 cm
which corresponds to a flow from 50 to 60 cm and a box differential height as low
as 8 cm. However, the shrinkage behavior of flowing concrete at early ages has
not been studied so far.
On the other hand, high-strength concrete with a compressive strength from
30 to 48 MPa has been used for high-rise reinforced concrete structures for
collective housing and others, and a super high-strength concrete with a
compressive strength greater than 60 MPa is being developed(6). Normally, high-
range AE water reducing agents are employed for the production of these high-
strength concrete since the lowering of water - cement ratio by decreasing the unit
amount of water is proven to be effective. Concrete with this type of admixture
has lower amount of free water for bleeding, and cracks at early ages due to a
rapid drying of concrete surface immediately after placing may be initiated.
The present study deals with the assessment of shrinkage and cracking
behavior of high-strength flowing concrete with a unit amount of water of 185
kg/m3 and water - cement ratio from 0.6 to 0.27, as well as normal high-strength,
low workability concrete with a unit amount of water of 170 kg/m3.
A C 1 SP-148 94 0662949 0520015 bTO 9
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
METHOD OF EXPERIMENTS
Materials Used
1--Cement:
An ordinary Portland cement with compressive strength
- at 28 days of 41.5 to 42
MPa was Üsed.
2--A~regates
Coarse aggregate of crushed hard sandstone (p = 2.66, F.M. = 6.68) and fine
aggregate of river sand (p = 2.62, F.M. = 2.70) were used.
3--Chemical admixtures
A polycarbonic acid type with a special water soluble polymer was used for
making flowing concrete, while three types of high-range AE water reducing
agents ( naphthalene type, polycarbonic acid type A and B) were used for normal
high-strength concrete.
Mixture Proportions
The water-cement ratio of flowing concrete was 0.60,0.45,0.35 ,0.30 and
0.27, while that of normal high-strength concrete was 0.45, 0.35, 0.30 and 0.27.
The unit amount of water was constant at 185 kg/m3 for flowing concrete and at
170 kg/m3 for normal high-strength concrete, and the sand - aggregate ratio and
the amount of admixture were adjusted to obtain the designed flowability. The
mixture proportions and properties of fresh concrete used in the experiments are
shown in Table 1.
Test Methods
1--Test of crack propagation at early apes
The apparatus used for the evaluation of cracks is shown in Fig. 1. A 60 x
60 x 5 cm of steel form with stud bolts mounted regularly at each of the four sides
of the frame restricts possible shrinkage of concrete due to drying so as to initiate
cracks in the concrete (7). Special care was taken to prevent possible friction
between the concrete and the bottom plate of the restriction frame by applying a
Teflon sheet on the bottom plate side and a paraffin paper for the concrete side.
Concrete was placed in the restriction steel frame and cured for 2 hours under a
constant temperature of 30 "C.Subsequently, it was exposed to an air flow with a
velocity of 8 &sec under a constant temperature of 30 "C and relative humidity of
60 %. Then the age of crack initiation, crack length and crack width were
measured until the age of 24 hours.
2--Free shrinkage test
In conjunction with the crack test, free shrinkage with the lapse of time was
measured for specimens of 10 x 10 x 40 cm demolded at an age of 2 hours.
Shrinkage within the age of 24 hours was measured, as shown in Fig.2, by two
dial gauges with a precision of 1/1000 m in contact with two glass plates (with a
30 mm long anchor pin ) which was set prior to molding and then attached to the
A C 1 SP-148 94 0662949 0520036 537 m
246 Yokoyama et al
center of the longitudinal end side of the concrete specimen. Weight loss of the
specimen for the shrinkage test was also measured by a platform scale with a
precision of 1 g and maximum capacity of 20 kg.
Free Shrinkage with Time

Shrinkage and loss of weight for flowing concrete with the lapse of time are
shown in Fig. 3(a) to (e). In each case, a remarkable evolution of shrinkage was
observed within 2 hours after the air flow was applied ( at the age of 4 hours)
Shrinkage increased with a decrease in water - cement ratio at less than 0.45,
though the loss of weight at these ages shows no significant difference with each
other.
Shrinkage of a specimen with water - cement ratio of 0.60 was greater than
that of 0.45. After an age of 4 hours, the loss of weight became smaller for a
specimen with lower water - cement ratio, since it had a larger amount of cement
which resulted in consumption of more water for hydration, while shrinkage
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
became moderate owing to the formation of the matrix. Shrinkage and the loss of
weight with time for a specimen with water - cement ratio of 0.30 exposed to
variable air flow rates of 5, 7 and 8 d s e c are shown in Fig. 3(f). The loss of
weight as well as shrinkage increased with an increase in air flow rate, while after
an age of 14 hours, the loss of weight decreased with an increase in air flow rate.
Shrinkage and loss of weight for normal high-strength, viscous concrete
with the lapse of time are shown in Fig. 4 in conjunction with the results for
flowing concrete. The amount of shrinkage of normal high-strength concrete,
within 1.5 hours after the air flow was applied, was remarkable and it increased
with a decrease in water - cement ratio. The subsequent values depend upon the
composition of high-range AE water reducing agent. The evolution of shrinkage
of normal high-strength concrete with a naphthalene type admixture increased
with a decrease in water - cement ratio, and this may be attributed to a relative
increase in unit amount of cement with a decrease in water - cement ratio, and to a
retardation effect in the presence of larger amount of the admixture. Progress of
shrinkage of normal high-strength concrete with a polycarbonic acid type
admixture was significantly smaller than that with the naphthalene type admixture,
and this may be attributed to a relatively low retardation effect of this type of
admixture. The shrinkage behavior of normal high-strength concrete thus varies
according to the composition of the admixture even though the mixture proportion
is equal to each other.
Shrinkage of flowing concrete with a larger unit amount of water and its
progress after the air flow was applied was low in comparison with that of
normal high-strength concrete using naphthalene type admixture with a smaller
amount of water. Shrinkage of normal high-strength concrete using polycarbonic
acid type admixture and its evolution after the air flow was applied was low in
comparison with that using a naphthalene type admixture, and even lower than
that of flowing concrete after one hour of exposure to the air flow.
Crack Initiation
A C 1 SP-148 9 4 0 0662949 0520017 473 0
Change in length of cracks on the surface of flowing concrete restricted by

the steel frame is shown in Fig.5, and Fig.6 illustrates the cracks observed at the
age of 24 hours (end of the test). Initiation of cracks was observed at 1.5 to 2
hours (at the age of 3.5 to 4 hours) after an air flow 'was applied, and it appeared
earlier with a decrease in water - cement ratio. Subsequent crack evolution became
very slow at the age of 5 to 7 hours and actually showed no growth even at the age
of 24 hours. Under variable air flow rate, cracks could not be observed with an air
flow rate of 5 mísec, while cracks initiated with that of 7 or 8 mísec. With an
increase in air flow rate, the age of crack initiation became earlier and the
subsequent crack propagation was accelerated.
Change in length of cracks on the surface of normal high-strength concrete
under similar restrictive conditions is shown in Fig.7, and Fig.8 illustrates the
cracks observed at the end of this test. Initiation of the crack was observed at a
very early stage after an air flow was applied, and crack initiation was at the age
of 2.5 to 3 hours when naphthalene type admixture was used, and it started earlier
with a decrease in water - cement ratio. The age of crack initiation of normal high-
strength concrete with polycarbonic acid type admixture was at the age of 2.3
hours, which is earlier than those using naphthalene type admixture with the same
water - cement ratio. Evolution of crack length showed remarkable growth in a
short while after the initiation and stopped at the age of 4 to 8 hours for those
using naphthalene type admixture, and at the age of 4 to 4.5 hours for those using
polycarbonic acid type B admixture. Subsequent growth was not observed as in
the case of flowing concrete. The cracks of those containing polycarbonic acid
type B admixture exhibited larger number and finer characteristics, greater total
length and narrower mean width than those using naphthalene type admixture.
No cracks were observed for concrete with water - cement ratio of 0.27
when the restriction frame was covered by a transparent acrylic plate to prevent
drying immediately after placing. A preventive measure against cracking for low
water - cement ratio concrete is to prevent water from drying immediately after
placing, and to supply water when the concrete has set.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Crack Evaluation
Parameters representing crack propagation characteristics at the age of 24
hours ( at the end of test) may be the average crack area, number of cracks per unit
area and total crack area. A quantitative analysis was made as follows:
1--Average crack area
Taking into account the crack width and length, and a half of their product,
the average crack area [a] can be calculated as follows:
where W is the maximum width of ith crack, L is the length of ith crack and N is
the total number of cracks in the test area.
2--Number of cracks per unit area
The degree of crack distribution ( the number of cracks) may be influenced
by the degree of restriction, rate of shrinkage and other factors. The number of
cracks per unit area [b] can be calculated as follows:
A C 1 SP-148 94 Ob62949 0520038 30T m
248 Yokoyama et al
b =- N (number/m’) ....................................................... (2)

A
where A is the area of the restriction frame (0.6 x 0.6m = 0.36 m2)
3--Totd crack area
Total crack area [c] can be calculated as follows:
c=a.N (mm2), ....................................................... (3)

where a is the average crack area.
The calculated results for flowing concrete on the basis of equation (1), (2)
and (3) are shown in Fig.9. No remarkable difference in these parameters between
those with water - cement ratio of 0.6 and that of 0.45 could be observed, while
those with water - cement ratio below 0.45 showed greater value with a decrease
in water-cement ratio reflecting significant crack propagation.
Results of the normal high-strength concrete are shown in Fig.9. These
three parameters showed nearly equal value for those using naphthalene type
admixture with water - cement ratio of 0.45 and 0.35, while it is greater for those
with water - cement ratio of 0.30 in comparison with that of 0.45 and 0.35.
For those using polycarbonic acid type admixture A with water - cement
ratio of 0.27, all parameters were nearly equal to those with naphthalene type
admixture with water - cement ratio of 0.45 and 0.35. Those using polycarbonic
acid type admixture B showed equal or slightly larger number of cracks per unit
area than those using naphthalene type admixture, while the other two parameters
showed substantially smaller values. The cracking behavior of normal high-
strength concrete varied, in this way, according to the composition of admixtures
even though water - cement ratio was equal, and those using naphthalene type
admixture showed significantly larger number of cracks than those using
polycarbonic acid type admixtures.
Taking into account these results, a comparison of cracking behavior of
flowing concrete with that of normal high-strength concrete can be made. The
crack initiation of flowing concrete was slight compared to a normal high-strength
concrete using naphthalene type admixture, and was notable compared to those
using polycarbonic acid type admixtures.
CONCLUSIONS
Flowing concrete and normal high-strength concrete exposed to air flow

were comparatively evaluated in terms of cracking and shrinkage at early ages.
The following results were obtained.
(1) Evolution of shrinkage of flowing concrete with the lapse of time was
remarkable in the first 2 hours (at the age of 4 hours) after air flow was
applied, and it became greater with a decrease in water - cement ratio.
Evolution of shrinkage of normal high-strength concrete was also notable in
the first 1.5 hours (at the age of 3.5 hours) after air flow was applied, and it
became greater with a decrease in water - cement ratio. Subsequent shrinkage
depended upon the composition of the high-range AE water reducing agents.
Those using naphthalene type admixture showed more notable shrinkage
than that of those using polycarbonic acid type admixtures.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 IObb29Ll9 0520019 246
Cracks of flowing concrete in the restriction frame initiated at the age of 3.5
to 4 hours, and they appeared earlier and were greater in growth with a
decrease in water - cement ratio when they stopped at the age of 5 to 7 hours.
Crack initiation of normal high-strength concrete in the restriction frame
varied according to the type of admixture, and was at the age of 2.5 to 3
hours and 2.3 hours for those using naphthalene type and polycarbonic acid
type admixture B respectively. It was earlier than that of flowing concrete.
The change in crack length with the lapse of time depended upon the
composition of admixtures, and the age when the change came to an end
varied from each other.
The final crack morphology could be evaluated quantitatively in terms of
average crack area, number of cracks per unit area and total crack area. In
general, crack initiation of flowing concrete and normal high-strength
concrete became notable with a decrease in water - cement ratio . Crack
initiation of normal high-strength concrete using naphthalene type admixture
was remarkable compared to those using polycarbonic acid type admixtures.
Crack initiation of flowing concrete was slight compared to those using a
naphthalene type admixture, but was notable compared to those using
polycarbonic acid type admixtures.
REFERENCES
Ozawa, K., Maekawa, K. and Okamura, H. “Development of High

Performance Concrete”; Proceedings of the Japan Concrete Institute, Vol. 11,
No.1, pp. 699-704, 1989. (in Japanese)
Takeshita, H., Sawara, H. and Yokota, S. “Fundamental Study on Super
Flowing Concrete without Compaction”; Concrete Research and Technology,
Japan Concrete Institute, Vol.1, No.1, pp.143-153, 1990. (in Japanese)
Kasai, Y., Hiraishi, S., Tobinai, K. and Osada, K. “Experimental Study on
Mix Proportion, Consistency, Strength and Shrinkage of Flowing Concrete”;
Proceedings of the Japan Concrete institute, Vol. 14, No. 1, pp. 79-84, 1992.
(in Japanese)
Kasai, Y. et al. “Experimental Study of Flowing Concrete (Part 1 - 7)”;
Summary of Technical Papers of Annual Meeting A, Architectural Institute of
Kasai, Y. et al. “Experimental Study of Flowing Concrete (Part 8 - 13)”;
Summary of Technical Papers of Annual Meeting A, Architectural Institute of
Japan, pp. 1161-1 172, 1993. (in Japanese)
The Architectural Institute of Japan, “State-of-the-Art Report on High-
Strength Concrete”, 1991. (in Japanese)
Kasai, Y., Matsui, I. and Yokoyama, K. “Shrinkage and Cracking of
Concrete at Early Ages”; Proceedings of the International Conference of
Concrete at Early Ages, Ecole Nationanle des Ponts et Chaussees, Paris,
V01.1, pp. 45-50, 1982.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0520020 Tbô
250 Yokoyama et al
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 1 - MIKTURE PROPORTIONS AND PROPERTIES OF FRESH CONCRETE

Unit weight
s/a (kg/m3) Admix- Sluil Plow Air Concrete
-
W/C - ture
(!
a
--- W - C c (CXX) (ci)
- (XI
íc3 R3.
8 O. 60 55. o 185 308 977 808 1.8 23. 1 50x54 4.8 23. I
8 o. 45 49. 1 185 411 830 E69 1.4 24.2 50x47 5. O 21. O
8 o. 35 47. o 185 528 149 853 1.8 26. O 65x63 5.0 23.0
Plowing 5 O. 30 47.4 185 617 720 808 2.0 24.5 54x53 4.7 21.5
concrete
7 O. 30 47.4 185 617 720 808 2.0 26.9 68 x 65 5. o 20.2
O. 30 47.4 185 617 720 808 2.0 26.5 68 x 68 4.0 22.0
O. 27 45.3 185 -
685 663 808 2.0 60 X 57 5. O
25.4 - 22. O
- -- -
o. 45 44. o 170 378 770 998 1.3' 17.2 28 x 25 4.8 23.8
8 o. 45 44. o 170 378 770 998 0.4" 17. O 30 X 28 4.2 20. O
Normal
high- 8 o. 35 39. o 170 486 647 1032 1.6' 17. O 32 X 31 4.0 24.5
strength
concrete 8 O. 30 37. o 170 567 590 1021 2.3' 18.7 32x31 4.7 25.5
8 O. 30 37. o 170 567 592 1020 0.5" 18. 1 32 X 30 4. O 20.8
O. 27 3
-8.O - 630 595
170 - 992 1.9" 24.5 53x52 3. 1 22.5
- _. -
' Naphthalene type ** Polycarbonic acid type A *** Polycarbonic acid type B
A C 1 SP-I148 9 4 m 0662949 0520023 7 T 4 m
600
1
E 609
,c-9 75
A sheet of paraffin
heet of teflon
Fig. 1-Mold for cracking test

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
pin : 3 hOfm
LenRth I Specimen
/ I Dial gauge
baraffin
teflon
Fig. 2-Method of measurement for shrinkage of concrete at early ages
A C 1 SP-148 94 Obb2949 0520022 ô30 U
252 Yokoyama et al
Age (hours)
2 5 10 2024 2 5 10 2024L
Fig. U g e versus shrinkage and loss of weight of specimens (flowing concrete)

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 0662947 0520023 7 7 7
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. î - A g e versus shrinkage and loss of weight of specimens (normal high strength concrete)
A C 1 SP-148 9 4 0662949 0 5 2 0 0 2 4 b03 =
254 Yokoyama et al
U
M
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662749 0520025 54T
-.-.
zm
m
o
0 -
w n m
O G ;
\os
u
3 0
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 Ob62949 0520026 486
256 Yokoyama et al
300
200
1O 0
10 20242 5 10 2024
Age (hours)
Fig. 7 4 g e venus total crack length (normal high strength concrete)

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 S P - l 1 4 ô 9 4 I0 6 6 2 9 4 9 0520027 312
W/C:O.45(N type) W/C:O.45(P type B) W/C:O. 35(N type)

1)3-04 2112 3)110 1)2-20 2)iO 31181 1)2-45 2113 3)106
4)O. 08-0.90(0.25) 4)0.04-0.20(0.07) 4)O. 15-1.3(0.57)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
W/C:O. 30(N type) W/C:O. 30w type 6) W/C:O.27(P type A)
112-30 2)7 3197 02-20 3 1 1 31276 113-00 2)11 31131
4)O. 15-2.0(0.49) 4)O. 04-0.20(0.07) 4)0.20-0.90(0.48)
1)Cracking time(h-m) 2)Number of cracks a t 24 hours 3)Totai length Df
cracks at 24 hours(cm) 4)Width of cracks a t 24 hours(Average)(m)
Fig. "Sketches of cracks at 24 hours (normal high-strength concrete)
A C 1 SP-/48 94 = O b 6 2 9 4 9 0520028 259
258 Yokoyama et al
W/C:0.30 1 Naphthaieiie
Poiy!arbonic
acid type A
Pol ycarlioii i c
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
3 800
7
t
_1
? 600
n 1
U
-r
0 400
-
0
200
5 7 8
o 0.60 0.45 0.35 0.30 0.27
w/c Wiiid velociîy ( d s ) W/C
Fig. M v e r a g e crack area, number of crocks per unit area, and total crack area
A C 1 SP-148 74 Ob62947 0520029 I175
SP 148-15
The Influence of Superplasticizer

on Mixture Proportioning and the
Strength and Durability of
Silica Fume Concrete
by S.A. Austin and PJ. Robins
Svnopsis: This paper reports on part of a substantial research programme on the

properties of condensed silica fume (CSF) concretes cured in temperate and hot
climates, carried out in the Depument of Civil Engineering at Loughborough.
The research approach was to investigate mixtures proportioned to have equal
workability and 28 day strength (when water cured at 20°C). This paper examines
the effect of superplastizer, curing method (water and polythene) and curing
environment (temperate and hot) on the compressive strength, permeability and
pore structure of 40 MPa concretes. More specifically, the paper contrasts the
performance of two 15%CSF mixtures (replacement by weight of cement) where
workabilities were controlled by the addition of extra water or superplasticizer.
The development of the concretes' strength and subsurface permeability index (air
and water) with age (from 7 to 180 days) is described, together with the intrinsic
permeability (air and water) and pore structure of their equivalent mortar fraction.
The use of superplasticizer to control workability increased the compressive
strength of CSF concrete mixtures by around 18% and 10% in the temperate and
hot environments respectively. The superplasticized concrete had lower air and
water permeabilities which is attributed to an improved pore structure as confirmed
by mercury intrusion porosimetq date. The improvements were more marked in
the CSF concretes cured in a hot environment.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Keywords: Climate; Compressive strenlith; curing; durobilitv; mix proportioning; permeability;

porosity; silico fume; superplosticizers; workobility

A C 1 SP-I148 94 W 0bb2949 0520030 907 W
260 Austin and Robins
Simon Austin is a Senior Lecturer in Structural Engineering at Loughborough

University of Technology, with 15 years research experience in concrete materials
and technology and the authodco-author of over 30 papers on the subject. Dr
Austin is a chartered engineer, Member of the Institution of Civil Engineers and
Fellow of the Concrete Society in the U.K.
Peter Robins is a Senior Lecturer in the Department of Civil Engineering at
Loughborough University. Concrete has been his main research interest over the
last 25 years, including work on deep beams, NDT, fibre reinforcement cement
replacements and concrete repair. Dr Robins is the authodco-author of over 50
papers, a chartered Member of the Institution of Civil Engineers and
CANMET/ACI reviewer.
INTRODUCTION
As a well-established supplementary cementitious material, silica fume has
received wide coverage at the four CANMET/ACI international conferences (1-4)?
and has been the subject of numerous literature reviews (5-8). Silica fume is now
no longer regarded as a waste by-product from the silicon and ferrosilicon alloy
industries but rather as a pozzolanic material that can be used to enhance the
properties of portland cement products, both in their fresh and hardened states.
The beneficial effect of CSF is generally attributed to the improvement in the
microstructure of the hydrated cement paste, which results in a reduction in
volume of larger pores and an increase in strength.
Silica fume has a very fine particle size and hence a high surface area. Its use as
an additive or as a partial replacement for cement in concrete mixtures can therefore
significantly increase the amount of mixing water necessary to achieve a particular
level of workability, which counters the beneficial effects of the pozzolan. This
increase in water demand can be overcome by the use of water reducing or
superplasticizingadmixtures. It is natural to be concerned about the effects of an
admixture on the mechanical properties, particularly those that affect long-term
durability. The use of silica fume as a cement replacement for concrete in hot
climates has been investigated at Loughborough, concentrating on curing methods,
replacement levels and durability related properties as well as strength (9-11).
This paper investigates the use of a superplasticizer to control the increased water
demand of a 15% CSF mixture, subjected to both hot and temperature
conditioning. This study compares normal portland cement concrete and modified
mixtures designed to have equal workability and 28 day strength when cured
under water at 2OoC; this is in contrast to the practice of simple weight for weight
replacement which makes direct comparison of performance less meaningful.
Duplicate samples were also conditioned in the temperate climate of the laboratory
to identify the relative efficiencies of the replacement materials in the different
climates. The research has studied the effects of both curing method and duration
on the development of strength and permeability of the concretes.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 94 = 0662949 0520033 843 =
SUPERPLASTICIZERS AND CSF

Because of the small size. and spherical shape of CSF particles they are able to fill
the void space between the much larger cement grains, which would otherwise be
occupied by water. Such water is ’trapped’and not available to act as free water
and contribute to the fluidity and workability of fresh concrete. Thus the resulting
improvement in grain size distribution of a cement and CSF blend can reduce the
water demand needed to attain a desired level of workability. However this
reduction in water demand is generally only reported (12) for lean mixtures with
additions of CSF at 10% or less. In more normal mixtures, experience in the UK
and abroad (13,14) has shown that a well proportioned mixture containing up to
300 kglm3 of normal portland cement and less than 10% of silica fume will not
exhibit a significantly increased water demand.
In richer concrete mixtures and those with higher silica fume dosages the higher
specific surface area of the CSF particles results in an increase in cohesiveness of
the concrete, making it more difficult to place and consolidate. This increase in
cohesiveness is due to the reduction in internal bleeding of the fresh concrete
because of the high surface area of CSF particles and also due to the increase in
contact areas between grains of cement. Carette and Malhotra (15) found that the
increase in water demand is almost directly proportional to the amount of silica
fume added. This increase in water demand can be overcome with the use of
water reducing or superplasticizing admixtures and to a lesser degree by reducing
the fine aggregate content of a mixture. The effect of silica fume, cement content
and plasticizer dosage on the water demand of concrete has been studied by Loland
and Hustad (16), who reported that in all mixtures containing plasticizer and silica
fume, water demand decreased at all dosages of silica fume because of the
dispersing action of the plasticizer. Similar conclusions can be drawn from
Sellevold and Radjy’s work (17).
High range water reducing admixtures (HRWRA) or superplasticizers are now
commonly used to offset the additional water demand caused by the use of silica
fume in concrete mixtures. Such admixtures are of paramount importance in the
production of ultra high-strength concretes which tend to contain excessive
volumes of cementitious materials compared to normal concrete (18,19).
TEST PROGRAMME
Materials. mixture proportions and sDecimen preparation
The ordinaq portland cement conformed to B12:1989 (20). Table 1 gives the
average percentages of the main oxide components and the fineness ranges for the
portland cement and silica fume. The fiie aggregate was a river sand of zone M as
specified by BS882:1983 (21), and the coarse aggregate was river gravel with a
maximum size of 20mm. The superplasticizerwas melamine-formaldehyde based
in a fluid form, with no significant retarding or &-entraining effects.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
To provide a sound basis for comparison of performance, the normal portland

cement and normal portland cement/CSF concretes (15% replacement by weight)
were proportioned to have equal workability (3-6s Vebe time, equivalent to 30-
6Omm slump) and 28-day compressive strength (40MPa) when cured under water
at 20°C in accordance with BS 1881 : Part 111 : 1983 (22). The mixture
A C 1 SP-148 94 = Obb2949 0520032 7 8 T
proportions are given in Table 2. The dosage of superplasticizer was in the range
13 to 14% by weight of the silica fume.
The proportions of the mortar mixtures (Table 2), were derived from the concrete
mixtures by omitting the coarse aggregate fraction and adjusting the water content
to ailow for that absorbed by the coarse aggregate in 30 minutes.
To ensure thorough mixing and dispersion of the materials the following
procedure was used. Coarse aggregate, CSF, half the water plus superplasticizer
were mixed (1 minute), then the cement added and mixed (1 minute) and lastly the
sand and rest of water added and mixed (1 minute). Following mixing, the
concrete was cast in 1OOmm cube moulds, compacted by external vibration and
covered with a polythene sheet. The specimens were then cured as described
below.
CunnP
- andenvironmental conditioning
One third of the specimens were transferred to be conditioned in a climatic room

which simulated a hot arid climate. Further details of the conditioning room and
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
regime selected are presented elsewhere (9-11). The temperature and humidity
were cycled during the first 24 hours between 4OoC, 15% RH (day) and 2OoC,
35% RH (night), 10 hours being spent at each condition with two 2-hour
changeovers. After demoulding at 1 day these specimens were wrapped in two
layers of polythene sheet and transferred to a 1 m3 Fisons climatic cabinet for
further conditioning up to the test age in the same temperaturehumidity cycle,
Specimens were cured in the polythene for a further 7 days, after which the
polythene was removed and the samples left exposed to the cyclic hot environment
until testing.
The other two-thirds of the specimens were left in their cube moulds in the
temperate laboratory conditions for 24 hours. Here the temperature and humidity
over 24 hours typically varied between 16-2OOC and 4040% RH. After
demoulding, a set of specimens were cured in water at a temperature of 20oC for 7
days, whilst the rest were wrapped with two layers of polythene sheet for 7 days;
followingcuring, all these specimens were left exposed to the temperate laboratory
environment until testing.
Concrete test methods
Compressive strength tests were carried out in accordance with BS 1881 : Part 116
: 1983 (23). using 1OOmm cubes. The results given are the average of five
specimens.
The sub-surface permeability data reported here were measured using the Figg
water and Figg air test methods (24,25), in which either water is absorbed or air
evacuated from a drilled and sealed hole, 1Omm in diameter and 40mm deep. In
each case a permeability index is calculated and the value is dependent on the pore
structure in the sub-surface cover zone of the concrete. Permeability
measurements were made on 1OOmmconcrete cubes, the holes for each test being
progressively drilled out from 3mm to 7mm to 1Omm to minimise cracking. The
cube specimens were conditioned by being dried in an oven at 105oCk5oC to
constant weight (this being taken as a change in weight of no more than O. 1% over
twenty-four hours) then removed from the oven, placed in polythene bags and left
A C 1 SP-148 94 0662949 0520033 616
overnight to cool down. For the air permeability tests a vacuum pressure of 55 kPa
was applied and the time taken for a pressure drop of 5 kPa gives the air
permeability index (in s). The water permeability index is the time (in s) taken for
the meniscus to travel 50mm along the standard glass capillary tube at a head of
1OOmm above the hole. High index values equate to low permeability. The Figg
air and water indices given m the average of three readings.
Mortar test methoh
For the air permeability tests, sample preparation entailed drilling cores (21mm
diameter) from 1OOmm diameter mortar cylinders. The cores were washed and
dried in a ventilated oven at 105f50C for two days, after which they were
transferred to a desiccator and allowed to cool down. The 21mm diameter by
50mm long specimens were loaded into a Hassler type pressure cell and a sealing
pressure of 3.45 MPa was applied. An inflow air pressure of 0.21 MPa was
applied on the face of the specimens, and the specimens were then allowed to
equilibrate for 15 minutes before recording the air flow from the water bubble flow
manometer. The calculation of air permeability was based on Darcy's law for
isothermai steady state flow of gas through a porous media.
The 1OOmm diameter by 25mm thick mortar discs for water permeability testing
were cut from the centre of 100mm diameter by 220mm long cylinders. The discs
were washed and then dried in a ventilated oven at 1051t5OC for two days, after
which they were taken out and stored in a well sealed polythene bag for
conditioning. The mortar discs were sealed with epoxy resin around the edges.
Water flow through the single celi water permeability apparatus was recorded each
24 hours by measuring the fall in water height taking place in a perspex tube
attached to the cell. Specimens were loaded into the cell and tested for 24 days.
The calculations of water permeability from the test data was based on Darcy's
law.
For the mercury intrusion porosimetry cylindrical cores 25mm in diameter and
70mm in length were taken from mortar cylinders 1OOmm in diameter and 1OOmm
long. Samples for the pore sizer were small discs 25mm in diameter by about 10-
12mm in thickness cut from the 25 x 70mm cores. The samples were dried in a
ventilated oven at 105°C for 24 hours. The instrument used in this investigation
was a Micromeritics Autopore 9310. This instrument is capable of generating 207
MPa pressure and calculates the volume of pores and their distribution in the range
300 to 0.006pm.
MIXTURE PROPORTIONING
The most common method of proportioning silica fume in concrete is to simply
substitute silica fume for portland cement on a one-for-one basis by weight. Silica
fume concrete proportioned by this method will nearly always have a comparable
strength with a control normal portland cement mixture at early ages, (1 to 3 days),
and higher strengths at 28 days and onwards (6). It is important to have a mixture
proportioning rationale which is able to proportion silica fume concrete for
workability, strength and economy. Our approach was to proportion a blend of
silica fume and normal portland cement to produce concrete of comparable
workability and 28 days compressive strength to a control plain normal portland
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 D O b 6 2 9 4 9 0520034 552 M
cement concrete mixture. The method was based on the BRE Design of Normal
Concrete Mixes (26).
Initially the effect of simple substitution of CSF for normal portland cement on a
one-for-one basis by weight on the workability and compressive strength at 3, 7
and 28 days was measured. Theoretical normal portland cement-CSF blends were
examined experimentaily to verify blend proportions. The experimental data were
used to establish the relationship between compressive strength and
water/cementitious materials ratio both with and without the use of a
superplasticizer.
Three concrete mixtures were investigated in the full research programme: C25,
C40 and C55. Silica fume concrete mixtures were established on the basis of
simple (one-for-one) substitution of normal portland cement in the plain mixture
by 5,10,15 and 20% by weight of n o d portland cement.
In this research work two approaches were used for controlling workability. In
the first, mixing water was increased and, in the second, a superplasticizer was
used. The admixture was used in a sufficient quantity to secure the workability
level without changing the initial mixing water content. A number of observations
were made on the effect of CSF on concrete workability:
1. In the lean and normal concrete mixtures containing 230 and 300 kg/m3 of
cement, silica fume can replace up to 10% by weight of normal portland
cement without requiring extra water. However, in the mixtures
containing 400 kg/m3 of normal portland cement, silica fume can only
replace 5% of normal portland cement without reducing the workability.
2. The increase in W/C+CSF to maintain a medium workability was found to
be almost linearly related to the amount of CSF, Le. it appears that the
water demand increase is independent of the amount of cement in the
mixture. This agrees with the findings of Carette & Malhotra (15).
However the test results disagreed with those reported by Loland and
Hustad (16).
3. The increase in water content required to produce a medium workability is
dependent on the W/C in the control plain mixture and CSF quantity. Fig.
l(a) shows the percentage increase in mixing water necessary to secure
medium workability (measured by Vebe) in the CSF mixtures.
4. Test results showed that a superplasticizer dosage of 12.5 to 16% by
weight of condensed silica fume (around 2.7% of total binder) was
necessary, depending on the CSF dosage (see Fig. l(a)), and that these
dosages are the same for all three concrete mixtures.
When maintaining workability by increasing the water content, the replacement of
normal portland cement with up to 15% CSF increased the 7 day Compressive
strength of the C40 and C55 mixtures (Fig. l(b)). At 20% replacement there was
a marked reduction in compressive strength. In the C25 mixtures normal portland
cement can be replaced with up to 20% CSF without impairing strength.
The use of the superplasticizer to maintain medium workability improved both
early and late age compressive strength of CSF concrete mixtures compared with
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 D 0bb2747 OC20035 479 D
those in which workability was maintained by adding extra water (see Fig. l(b)).
This improvement may in part be due to maintaining the water/(cement + CSF)
ratio and also may result from the more uniform dispersion of cement and CSF
particles.
CONCRETE
ComDressive strenmh
The development of strength with age of the C40 mixtures under three different
curing regimes is shown in Fig. 2. When comparisons of the different mixtures
are made it must be remembered that the mixtures have been proportioned to have
equal workability and 28-day strength (40 MPa) when temperate water cured.
Moreover, when considering the effects of the superplasticizer it is important to
note that the two CSF mixtures have the same water/(cement + CSF) ratio. Thus
any differences in strengths recorded will have resulted from the effects of
different curing regimes and/or the different rates of strength gain of the different
mixtures.
Fig. 2 enables comparison to be made between the CSF and plain C40 mixtures
and between the 15% CSF mixtures with and without the superplasticizer. Under
temperate conditions (water or polythene cure) the strengths of the C40/0 and
C40/15 mixtures were very similar. In contrast, the superplasticized C40/15S
mixture produced greater strengths at later ages through a higher degree of
hydration, possibly caused by better dispersion of the cementitious particles
brought about by use of the superplasticizer.
The contrasting detrimental and beneficial effects of hot climate curing on the
normal portland cement and CSF mixtures, respectively, is clearly evident in the
strengtwage curves for the hot polythene cure. At early ages (3 and 7 days) hot
curing increased the strengh of the normal portland cement mixture compared with
the temperate cure, but at later ages (28 to 180 days) the situation was reversed.
This trend has been observed by many researchers (27) and is usually attributed to
the increased early age hydration producing a coarse crystalline structure which is
detrimental to later age hydration. The CSF mixtures exhibited a different trend to
the normal portland cement one. Hot curing was beneficial to both early and late-
age strength. It may be surmised that the high curing temperature which initially
accelerates the hydration of the normal portland cement, has consequently
accelerated the pozzolanic reaction between the Ca(OH)2 and the silica fume. The
structure of the normal portland cement/CSF paste appears to allow continuation of
normal portland cement hydration and of the pozzolanic reaction at later ages as
well.
Under hot curing conditions the superplasticized CSF mixture outperformed the
one without superplasticizer, at all ages, again possibly due to the improved
dispersion of the cementitiousparticles.
Air and water Dermeabiiitv
The sub-surface air and water permeabilities, as determined by the Figg air and
water tests, are presented in Fig. 3 and 4. In general, permeability decreases with
age at a decreasing rate. Compared with the normal portland cement mixture, the
permeabilities of the CSF mixtures decrease at a higher rate. This higher rate is
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 m 0bb2949 O520036 325 m
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
most likely a combination of both the physical densifying effect of the additional
finer CSF particles and the chemical effects of additional hydrates produced by the
pozzolanic reaction. Temperate water and polythene curing produced quite similar
trends with, as expected, the water cured concrete permeabilitiesbeing slightly less
than those of the polythene cured specimens.
Generally the hot curing increased the late-age (28 days onwards) a u and water
permeabilities of the normal portland cement concrete and significantly decreased
the permeabilities of the CSF concrete mixtures at ail ages. The effects of different
curing regimes are discussed in more detail elsewhere (10).
The permeabilities of the superplasticized CSF mixture were less than the non-
superplasticized ones at all ages and under all curing regimes, These differences
are again (see previous section) most likely due to the improved dispersion of the
cementitious particles in the superplasticizedmixture, resulting in a more even and
refined pore structure.
MORTAR
and water mmeability
The true air and water permeabilities for the mortar mixtures are presented in Fig.
5 and 6. The similarity of the general trends to the Figg air and water
permeabilities of the concrete mixtures (Fig. 3 and 4) is apparent. Permeability
reduces with increased length of hydration period. Permeabilities of the CSF
mortars are less than those of the normal portland cement mortars (due to the finer
pore stnicture) and they decrease at a higher rate with time.
Hot curing is detrimental to the later-age permeabilities of the plain mortar mixtures
and beneficial to the CSF mortar mixtures, and again the CSF mortar with a
superplasticizerconsistently outperformsthe one without.
The water permeability values in Fig. 6 compare favourably with published data.
The Concrete Society (28) reported a number of researchers' results with a
permeability of m/s being typical for normal portland cement concretes and
m/s being a lower bound at watedcement of 0.4. Khyatt and Aïtcin (8)
quote values of 10-11 m/s for normal portland cement control concrete compared
with 6 x 4 s with 5% CSF and less than 4 s with 10% CSF and
above,
The intrinsic mortar permeabilities are compared with the Figg concrete
permeability indices in Fig. 7 and 8. There appears to be a good exponential fit
between the two water permeability measurements (Fig. 7) and a somewhat more
scattered exponential fit between the air permeability values. Recalling that the
mortars were proportioned to be representative of the concrete mixtures' mortar
fraction, the data suggest that the relatively non-destructive in-situ Figg tests give a
reasonable indication of intrinsic permeability.
Pore structure
The mercury intrusion porosimetry test produces a range of information on the
pore size distribution which can be used to describe the pore structure of the
mortars. The three parameters selected for discussion in this paper are pore
A C 1 SP-146 74 0662747 0520037 261
surface area (PSA), median pore diameter by volume (MPD), and the volume of
pores greater than O.1pm (LPV). Pore surface area is directly related to the
number of pores in the matrix and hence at a constant porosity, the greater the pore
surface area the higher the number of small pores. The median pore diameter by
volume relates to the relative volume of the smailer pores, the smaller this value the
greater the volume of smaller pores.
The effect of age on the pore structure of plairi and CSF mortars can be illustrated
by its effects on PSA and MPD which are shown in Fig. 9 and 10. The pore
structure becomes finer with increasing age (over the range investigated). The rate
of increase in PSA and reduction in MPD is higher at early ages and gradually
decreases at later ages. Results also reveal that the combined effects of cement
hydration and pozzolanic reaction produces a finer pore structure in the CSF
mortars (compared with the OPC ones) at later ages under both curing
environments. The general refinement of pores in CSF mortars has been observed
by other researchers (e.g, Sarkar and Aiitcin (29), Zhang and Gjorv (30)). This
f i e r pore structurecan be explained by a combination of the physical properties of
CSF particles (i.e. their extreme fineness and large surface area) and the presence
of highly reactive CSF altering the pore struc:tureof the paste matrix by reacting
with the liberated Ca(0H)Z to form dense secondary calcium silicate hydrate gels
in the capillary pores. However, Feldman and Cheng (31) and Feldman (32) have
suggested that at least some of the apparent increase in small pores is actually a
result of the intrusion at high pressures of relatively large unconnected pores, as
shown by mercury re-intrusion.
Figures 9 and 10 reveal that at 7 days under temperate curing conditions the PSA
(and MPD) of all mortar mixtures are similar. It is only at later ages that the
marked beneficial effects of the CSF on a mortar's pore structure can be observed.
Furthermore, at 7 days the hot curing regime is seen to have had very little effect
on either the PSA or MPD of the plain normal portland cement moms. However,
hot curing has clearly accelerated the beneficial effects of the inclusion of silica
fume into the mortar mixtures and by 7 days their PSA and MPD are vastly
superior to those of the plain mortars. Clearly the higher curing temperature has
accelerated the normal hydration of OPC grains and caused the pozzolanic reaction
to commence earlier. This will have resulted in a higher rate of deposition of
hydration products in the capillary water-filled spaces, reducing their sizes and
volumes. This behaviour is revealed by the higher PSA and lower MPD-V at early
ages.
Figures 9 and 10 reveal that most of the divergence between the CSF and plain
mortars occurs between 7 and 28 days. This is in agreement with others who have
noted that the pozzolanic reaction is largely confined to this period (29,33). The
effect on the pore structure of securing the workability in the mortar mixtures by
using a superplasticizer compared to that in which extra water was used is also
shown in Fig. 9 and 10. Results indicate that using a superplasticizer to control
the workability of mixture improved the pore structure of the paste which had
higher PSA and lower MPD than the non-superplasticized mortars. This better
pore structure is presumably caused by the effect of the superplasticizer on the
dispersion of the cementitious particles, resulting in a more uniform distribution of
the hydration products. Interestingly, the greatest improvement brought about by
the use of the superplasticizer are to the pore surface area under the hot curing
regime.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662947 0520038 L T ô
A comparison was made between the volume of the coarse pores (those having a
diameter greater than 0.1 pm, as a percentage of the specimen volume), in mortars
made with normal portland cement and combinations of normally portland cement
and CSF. The reason for the selection of this value is that the variables of curing
environment, CSF content and the use of superplasticizer were found to influence
mainly the volume of pores with a diameter greater than O. 1 pm. Moreover Mehta
and Manmohan (34) found that permeability and durabiiity were mainly influenced
by the volume of these large pores. The relationships between age and the relative
volume of larger pores of normal portland cement and normal portland
cement/CSF mortars cured in temperate and hot environments is shown in Fig. 11,
which shows that the volume of large pores decreases rapidly at early ages, but at
a slower rate at the later ages up to 180 days. Sarkar and Kitcin (29) have also
observed reduction in LPV with age, their very high strength superplasticized
concrete (113 MPa) had virtually no pores greater than 0.1 pm by 91 days.
The large pore volume of plain normal portland cement mortars cured in hot
environment is higher at later ages (from 28 days and onwards) compared to that
of those cured in temperate environment. However, normal portland cement/CSF
mortars show a completely different trend. The volume of coarse pores of CSF
mortars cured in a hot environment is lower than those cured in a temperate
environment, again reflecting the beneficial effect of the hot environment on the
pozzolanic reaction.
The superplasticized mortar mixture had lower large pore volume compared to the
non- superplasticized one, under both temperate and hot curing environments.
CONCLUSIONS RELATING TO THE EFFECTS OF THE
SUPERPLASTICIZER IN CSF CONCRETES AND MORTARS
1. in the concrete mixtures where the superplasticizer was used to secure the
workability, the required dosage was found to be independent of the
normal portland cement content and only dependent on the CSF dosage.
2. The use of the superplasticizer to control workability increased the
compressive strength of CSF concrete mixtures by around 18% and 10%
in temperate and hot environmentsrespectively. This effect is attributed to
the efficient dispersion of the hydration products.
3. Superplasticized CSF concrete and mortar mixtures had lower water and
air permeabilitiescompared to the equivaient non-superplasticizedmixtures
in both temperate and hot environments.
4. The superplasticized mortars had a better pore structure compared to
equivalent non-superplasticized ones. The use of the superplasticizer
produced a paste with higher pore surface area and lower mean pore
diameter. Moreover, the superplasticizedmortar had a lower proportion of
larger pores under both temperate and hot curing environments.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 Ob62747 0520037 034 M
REFERENCES
1. Malhotra, V.M., Editor. Proceedings, First International Conference on
the Use of Fly Ash, Silica Fume, Slag, and Other Mineral By-products in
Concrete, Montebello, Canada, AC1 SP-79, Vol. 2, July-Aug. 1983, 1181
PP.
2. Maihotra, V.M., Editor, Proceedings, Second International Conference on
the Use of Fly Ash, Silica Fume, Slag, and Natural Pozzolans in Concrete,
Madrid, Spain, AC1 SP-91, April 1986.
3. Maihotra, V.M., Editor. Proceedings, Third International Conference on
Trondheim, Norway, AC1 SP-114, June 1989, 1714 pp.
4. Malhotra, V.M., Editor. Proceedings, Fourth International Conference on
Istanbul, Turkey, AC1 SP-132, May 1992, 1675 pp.
5. AC1 Committee 226, 'Silica Fume in Concrete', AC1 Materials Journal,
V01.84, No.2, 1987, pp. 158-166.
6. Sellevold, E.J. and Nilsen, T. 'Condensed Silica Fume in Concrete: A
World Review', Supplementary Cementing Materialsfor Concrete, SP86-
8E, CANMET, Ottowa, Editor: Malhotra, V.M., 1987, pp. 167-243.
7. Swamy, R.N. (Editor), Cement Replacement Materials, Surrey University
Press, 1986, 259 pp.
8. Khayat, K.H. and Aitcin, P.C. 'Silica Fume in Concrete - An Overview'.
Proceedings, Fourth International Conference on the Use of Fly Ash,
Silica Fume, Slag, and Natural Pozzolans in Concrete, Istanbul, Turkey,
AC1 SP-132, May 1992, pp. 835-872.
9. Al-Eesa, A.S.S., 'Silica Fume Concrete in Hot and Temperate
Environments', PhD Thesis, Loughborough University of Technology,
1990.
10. Austin, S.A., Robins, P.J. and Al-Eesa, A.S.S. 'Influence of Early
Curing on the Sub-surface Permeability and Strength of Silica Fume
Concrete', paper submitted to Materials Journal of ACI, December 1993.
11. Austin, S.A., Robins, P.J. and Al-Eesa, A.S.S. 'Influence of Early
Curing on the Surface Permeability and Absorption of Silica Fume
Concrete', Procs. of 3rd International Conference on Durability of
Concrete, Nice, France, May 1994.
12. Grutzeck, M., Roy, D.M., and Wolfe-Confer, D. 'Mechanism of
Hydration of Portland Cement Composites Containing Ferrosilicon Dust',
Proceedings, Fourth International Conference on Cement Microscopy, Las
Vegas, Nevada, 1982, pp.193-202.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 m 0662749 0520040 856 m
13. Parker, D.G. 'Microsilica Concrete: Part 1, The Material', Concrete

Society current practice sheet No. 104, Concrete, London, Oct. 1985,
p.21-22.
14. Regourd, M., Mortureux, B, Aitcin, P.C. and Pinsonneault, P.
'Microstructure of field concrete containing silica fume'. Proc. 4th
International Conference on Cement Microscopy, March/Apnl 1982, Las
Vegas, Nevada, (15A), p.249-260.
15. Carette, G.G. and Malhotra, V.M. 'Mechanical properties, durability and
drying shrinkage of Portland cement concrete incorporating silica fume'.
ASTM J . Cement, Concrete, Aggregate, Vo1.5, No.1, 1983.-
16. Loland, K.E. and Hustad, T. 'Fresh Concrete and Method of Data
Analysis'. Report No. STFGS A81031, Cement and Concrete Research
Institute, The Norwegian Institute of Technology,Trondheim, June 1981.
17. Sellevold, E.J. and Radjy, F.F. 'Condensed silica fume (Microsilica) in
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
concrete: water demand and strength development'. Proceedings, First
International Conference on the Use of Fly Ash, Silica Fume, Slag, and
Other Mineral By-products in Concrete, 1983, pp.677-694.
18. Baalbaki, M., Sarker, S.L., Kitcin, P.C. and Isabelle, H. 'Properties and
Microstructure of High-Performance Concretes containing Silica Fume,
Slag, and Fly Ash', Proceedings, Fourth International Conference on the
Use of Fly Ash, Silica Fume, Slag, and Natural Pozzolans in Concrete,
Istanbul, Turkey, AC1 SP-132, May 1992, pp.921-942.
19. Kakizaki, M., Edhairo, H., Tochigi, T. and Niki, T. 'Effect of Mixing
Method on Mechanical Properties and Pore Structure of Ultra High-
Strength Concrete', Proceedings, Fourth International Conference on the
Use of Fly Ash, Silica Fume, Slag, and Natural Pozzolans in Concrete,
Istanbul, Turkey, AC1 SP-132, May 1992, pp. 997-1015.
20. British Standards Institution, 'Specification for Portland Cements', BS
12:1989, British Standards Institution, London, 1989.
21. British Standards Institution, 'Aggregates from Natural Sources for
Concrete', BS 882:1983, British Standards Institution, London, 1983.
22. British Standards Institution, 'Method of Normal Curing of Test
Specimens (2PC method)', BS 1881:Part I I I : 1983, British Standards
institution, London, 1983.
23. British Standards Institution, 'Method of Determination of Compressive
Strength of Concrete Cubes', BS 1881:Part 116:1983, British Standards
Institution, London, 1983.
24. Figg, J.W., 'Methods of Measuring the AU and Water Permeability of
Concrete', Magazine of Concrete Research, Vol. 25, No. 85, December
1973, pp.2 13-219.
A C 1 SP-14ô 74 0 6 6 2 7 4 9 0520043 792
25. Cather, R., Figg, J.W., Marsden, A.F. and O'Brien, T.P.,
'Improvements to the Figg Method for Determining the Air Permeability of
Concrete', Magaxzine of Concrete Research, Vol. 35, No. 129, December
1984, pp.241-245.
26. Teychenné, D.C. et al. 'Design of Normal Concrete Mixes', Building
Research EstablishmentReport, BR106, 1988,42 pp.
27. AC1 Committee 305, 'Hot Weather Concreting', ACI Materials Journal,
Vol.88, No.4, 1991, pp.417-436.
28. Concrete Society, 'Permeability Testing of Site Concrete - A Review of
Methods and Experience', Technical Report No.3 1, Concrete Society,
London, 1987,95 pp.
29. Sarkar, S.L. and Aitcin, P.-C., 'Dissolution rates of silica fume in very
high strength concrete', Cement and Concrete Research, Vol. 17, No.4,
1987, pp.591-601.
30. Zhang, M.-H. and GjØrv,O.E., 'Effect of silica fume on pore structure
and chloride diffusivity of low porosity cement pastes', Cement and
Concrete Research, V01.21, No.6, 1981, pp.1006-1114.
31. Feldman, R.F. and Huang Cheng-yi, 'Properties of portland cement-silica
fume pastes I-porosity and surface properties', Cement and Concrete
Research, Vo1.15, N0.5, 1985, pp.765-774.
32. Feldman, R.F., 'The effect of sandkement ratio and silica fume on the
microstructure of mortars', Cement and Concrete Research, Vol. 16, No.1,
1986, pp.31-39.
33. Huang Cheng-yi and Feldman, R.F., 'Influence of silica fume on the
microstructural development in cement mortars', Cement and Concrete
Research, Vo1.15, No.2, 1985, pp.285-294.
34. Mehta, P.K. and Manmohan, M. 'Pore size distribution and permeability
of hardened cement paste'. Proc. 7th Int. Congress on Chemistry of
Cement, V3, 1980, pp.VlI 1-5.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662947 0520042 629
TABLE 1 - CHEMICAL AND PHYSICAL PROPERTIES OF CEMENTITIOUS MATERIALS
oxide Nonnai portland cement Condensed silica fume

(76) (76)
Si02 21 90
A1203 4.6 1.1
Fez03 2.7 1.4
Cao 65 O .4
SOS 2.8
MgO 1.3 2.0
K20 0.6 1.8
Na0 o. 1 0.8
I
I I
c3s 55.3
c2s 18.9
C3A 7.6
C4AF 8.3
Fineness 360 - 380 mg2/kg 15,000 - 20,000 m*/kg
- I
l
Mixture Cement
SilicaFume Water Sand Coarse Aggregates W/(C+SF)
Reference &g/m3) &drn3) (kdm3) &g/m3) &idm3) ratio
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
1Omm 20mm
C40/0 300 185 660 390 780 0.62

c40/15 210 45 15 210 660 390 780 0.82
C40/15S 175 45 15 185 660 390 780 0.84
-
M40/0 300 175 660 0.58
M40/15 210 45 15 200 660 0.78
M40/15S 175 45 15 175 660 0.79
-
A C 1 SP-14B 94 0662947 0520043 5b5 =
40
Workability by adding water
20
.-
I IO
E a5
I
10 15 20
E I0
Workability by adding superplasticizer
vi
u
10 15 20
CSF (I
wt of OPC)
Fig. la-Influence of silica fume content on water demand

and superplasticizer dosage to maintain workability
-c 80 -?
Workability by adding water
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 10 20 30
-
~
f
Workability by adding superplasticizer
O 10 20 30
Age (days)
Fig. 1b-Development of compressive strength of CSF

mixes with extra water or superplasticizer
A C 1 SP-148 94 0662947 0520044 4TL
60-
Temperaie water atre
O 1
O 50 100 150 200
60,
-z
Temperate polythene cure
40-
t
L:
t
.- 20 -
Bo
u O 1
O 50 100 150 200
60
Hot polythene aut A
t o
üì 20
uE
o coo/o
A c(o/15
4 uo/iss
O 7
O 50 100 150 200

Age (days)
Fig. 2 - E f f e d of curing environment on

concrete compressive strength development
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 0662747 0520045 338
250
Temperate water cure
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O4
O 50 100 150 200
250
Temperate polythene cure
200
O 50 100 150 200

250
Hot polythene c m I
!i
50
A C40/15S
O
O 50 100 150 200
Age (days)
Fig. 3-Effect of curing environment on

concrete sub-surface air permeability
A C 1 SP-14B 94 m 0662949 0520046 274 m
150
Temperate water cure
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 50 100 150 200
150 1 Hot polythenerim
Fig. &Effect of curing environment on

concrete sub-suríace water permeability
A C 1 SP-248 94 = 0 6 6 2 9 4 9 0520047 L O O
+;I O
O
50
Wl15
M4û/ES
50
100
-&----
100
A -
Age (days)
150
-_______
150
---4
200
200
Fig. SEffed of curing environment on

mortar intrinsic air permeability
;>^ 1E&
Q Temperate polythene cure
c
x 1E+E
v>
\
o
1E+03
E
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
1E+02 -
I I \.
O 50 100 150 200

Age (days)
Fig. ó-Effed of curing environment on

mortar intrinsic water permeability
A C 1 SP-148 74 H Obb2749 0520048 047
O 50 100 150
Figg water index (5)
Fig. 74omparison of intrinsic and

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
sub-surface water permeabilities
\ A O 0
O
X
.r: 100
--E
x
.-
A
B
1
E
10
I
1
\AA
I
)4 *o 4.015x I
I
O 50 100 150 200
Figg air index (s)
Fig. U m p o r i s o n of intrinsic and

sub-surface air permeabilities
A C 1 SP-148 94 0662747 0520047 T ô 3 m
6
Temperate polytbene cure
O 50 100 150 200
6T----7=7 Hot polythene cure
O 50 100 150 200

Age (days)
Fig. %Effect of curing environment

on mortar pore surface a n a
O.?
Temperate polythen-1
O 1 1
O 50 100 150 200
E,- I \...* 'o-..... .. -.* .......
O 50 100 150 200

Age (days)
Fig. 1LEffect of curing environment

on mortar median pore diameter
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 94 Obb2949 0520050 7 T 5 M
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
20
Temperate polyihene cure
04 I
O 50 100 150 200
1 Hot polyihene cure
O 50 100 150 200

Age (days)
Fig. 11-Effect of curing environment on mortar

large pore volume (> 0.1 pm diameter)
A C 1 SP-348 94 0bb2949 0 5 2 0 0 5 3 6 3 3
SP 148-16
Properties of an Acrylic Graft

Copolymer-Based New Superplasticizer
for Ultra High-Strength Concrete
by M. Kinoshita, 1. Suzuki,
T. Yonezawa, and K. Mitsui
Synopsis: In order to place ultra high strength concrete with a

compressive strength exceeding 100 MPa on site. a technology is
required to impart high fluidity to the concrete, with a water-
to-cementitious material ratio in an extremely low range of 0.25
or less.
For this purpose, the authers synthesized a new superplasticizer
comprising water-solubleacrylic graft copolymer, which has an
excellent cement dispersing capability.
This paper reports the investigation of the surface chemical
properties of this new superplasticizer and the properties of
the cement paste and mortar containing it.
It was confirmed that the new superplasticizer imparts a higher
fluidity to cement paste and mortar with an extremely low range
water-to-cementitious material ratio than conventional superpla-
sticizers. It nas also found that the surface tension of the
solution of the new superplasticizer is similar to that of conv-
entional polycarboxylate superplasticizers, whereas the adsorpt-
ion by cement and zeta potential of the new superplasticizer are
between those of the p-naphthalene superplasticizers and the
polycarboxylate superplasticizers. The high fluidity of the
cement paste and mortar containing the new superplasticizer with
a very low range water-to-cementitious material ratio may be
particularly attributable to the above-mentioned properties with
respect to surface tension as well as the molecular weight and
chemical structure of the graft copolymer.
Keywords: &;adsorption; cernent pastes; compressive strength; hioh strenoth concretes;

mortars (material); plastics, uolvrners, and resins; silica fume; suuerplasticizen; tension; viscosity;
water cernent ratio; workability
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Mitsuo Kinoshita is Chief Research Engineer, Chemical
Admixture Division, Takemoto Oil & Fat.Co.,Ltd.,Aichi, JAPAN.
Takaji Suzuki is Director, Head of Chemical Admixture
Division, Takemoto Oil & Fat.Co.,Ltd..Aichi, JAPAN.
Toshio Yonezawa,.Ph.D.,is Chief Researcher, Research and
Development Institute, Takenaka Corporation, Chiba, JAPAN.
Kenro Mitsui is Researcher, Research and Development
Institute, Takenaka Corporation, Chiba, JAPAN.
In recent years an extensive number of studies have been

conducted regarding ultra high strength concrete with a
compressive strength of 100 MPa o r higher for further increasing
the height of concrete highrise buildings.
In order to enable an ultra high strength concrete with a compr-
essive strength exceeding 100 MPa to be placed on site, it is
important to ensure fluidity, as well as strength. of concrete
with a water-to-cement (or-cementitious material) ratio in an
extremely low range of approximately 0.20.
The viscosity in concrete in this range of water-to-cement ratio
is generally very high, resulting in a substantial loss in
placeability, such as filling capability and pumpability.
It is therefore important to develop a new superplasticizer
which has a higher cement dispersing capability than convention-
al superplasticizers (1-4) for ensuring good placeability and
which improves the consistency of concrete. The authers synthe-
sized a novel superplasticizer which comprises an acrylic graft
copolymer, using the molecular design technique (5).
This paper reports the results of the investigation into the
surface chemical properties attributable to the chemical struct-
ure of the superplasticizer, as well as into the properties of
cement paste and mortar containing it.
Synthesis of the nem superplasticizer

The new superplasticizer (herein referred to as SSP) used in
this experiment was synthesized by radical copolymerization in
an aqueous system, using the macromonomer method ( 6 ) . Macromon-
omer is a monomer of a high molecular weight containing at one
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14B 74 0662747 0520053 404
end a polymerizable functional group. In this experiment, copc-

lymerization nas conducted employing methoxypolyethylene glycol
methacrylate and polyetylene glycol mono allyl ether as the mac-
romonomers and methacrylates, acrylic: esters, p-methallyl oxybe-
nzenesulfonates, and methallylsulfonates as the comonomers. In
order to obtain adequate molecular weight as a dispersing agent,
the molecular weight was controlled tiy utilizing the properties
of methallylsulfonate which is effective in chain transfer (7).
Chemical structure of the new superplasticizer
The chemical structure of the SSP is shown in Fig.1. The
structure is characterized by comprising an acrylic graft copol-
ymer with polyethylene glycol chains as the graft chains, and
containing in the molecule carboxyl groups and sulfonic groups
as anion groups of an adequate mole ratio (-COOH/-S03H=80/20).
A number average molecular weight of 3900 by Pullulan conversior
was adopted as the optimum molecular weight of SSP, which
strongly affects the dispersing capability. Fig.2 shows the
molecular weight distributions measured by a gel permeation
chromatograph (Type 244 from Waters Co.).
In these experiments, conventional superplasticizers (poly-
carboxylate type (4) and naphthalene type) were also tested for
comparison.
Materials used
(a) Admixtures ;
SSP : Acrylic graft copolymer based new graft copolymer

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
SP : Polycarboxylic graft copolymer based superplasticizer
NSF : fi-Naphthalene sufonated formaldehyde condensate
The superplasticizers (SSP, SP. and NSF) were all used in the
form of aqueous solutions, the solid concentration of which was
adjusted to 30%. A polyether defoaming agent was added in
advance to the SSP and SP solutions at 0.2% of the solution.
(b) Cements ; Five types were used.
N : Ordinary portland cement
H : High early-strength cement
M : Moderate heat portland cement
BB: Portland blast-furnance slag cement (slag 30-60%)
FB: Portland fly ash cement (fly ash 10-20%)
(c) Microsilica (SF) ; Type 940-U from Elkem Materials
(d) Sand ; River sand (FM=2.73, specific gravity-2.63)
Table 1 presents the chemical composition of the cementitious
materials.
A C 1 SP-148 94 W 0662749 0520054 3 Y O D
284 Kinoshita et al
Measurement and method

Measurement of surface tension
The surface tension of the aqueous solution of each superpl-
asticizer adjusted to a certain concentration was measured by a
Du Nouy tensiometer.
Measurement of adsorption
The adsorption of SSP and SP were measured by the pyrolysis
gas chromatogaphy. That of NSF was measured by mans of the
ultraviolet absorption spectrum (228 nm). Cement pastes, conta-
ining SSP, SP, or NSF, with a water-to-cement ratio of 0.5 were
prepared. The adsorptions were calculated from the differences
of the amounts of the superplasticizers dissolved in the liquid
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
phase before and after the adsorption.
Measurement of zeta potential
The zeta potential of colloidal particles can be measured by
determining the rate of transfer of particles in a given electr-
ic field. 100 mg of cement, 100 g of water and the specified
amounts of the superplasticizers were mixed to measure the zeta
potential by a laser zee meter model 501 manufactured by Pen Kem
Inc.
Evaluation of fliiidity of cement paste
Different dosages of the superplasticizers were added to the
various with or without silica fume, and nixed for 3 minutes at
20 deg C. Then flow tests were conducted in accordance with JIS
R 5201, to evaluate the fluidity of the cement paste.
Evaluation of fluidity of mortar
Cement containing silica fume, sand, and water containing
SSP proportioned as shown in Table 2 were added to a mortar
mixer in that order, and were mixed for 3 minutes in a room with
a temperature of 20 deg C and a relative humidity of 80%.
The evaluation of fluidity was made as per cement pastes, i.e.
measuring the spread of the mortar after lifting the cone, in
accordance with JIS R 5201.
The changes in mortar fluidity over time were tested by taking
saaïples from the mortar left to stand for specific periods at 20
deg C.
Heat evolution history
Six hmdred grams of mortar were charged in a plastic conta-
iner 6 cm in diameter and 10 cm in height, covered with an
insulating material, which was then encased in a vacuum bottle
and sealed. The heat evolution due to heat of hydration was
A C 1 SP-148 74 = Obb2949 0520055 2 8 7 M
recorded with a thermocouple inserted in the mortar
RESULTS Aw DISCUSSION
Interfacial properties
Surface tension of superplasticizer solutions
The results of the surface tension measurements are shown in
Fig.3. Marginal additions of SSP and SP reduce the surface
tension of water, because both superplasticizers have polyethyl-
ene glycol chains in the molecules. This property on one hand
imparts wettability to the cement slurry and increases the disp-
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
ersing capability, while on the other is considered to be a

cause of unstable air entrainment during mixing. In order to
apply SSP to ultra high strength concrete, in which excessive
air entrainment is not favored, a defoaming agent was added, so
as to restrict the air entrainment.
Adsorption by cement particles
Fig.4 shows the isothermal adsorption of SSP by ordinary
portland cement particles. The curves indicated Langmuir-type
monomolecular adsorption. Being graft copolymer superplasticiz-
ers, SSP and SP show smaller apparent saturated adsorptions when
compared with NSF. This may be due to the different degree of
adsorption on the cement particles. The conformation of adsorp-
tion of graft copolymers is considered to be more bulky and
three-dimensional than that of NSF, which is a linear polymer.
The reason why the adsorption of SSP is greater than that of SP
may be that the molecular weight is less, the proportjon of
anion groups is higher, and the graft ratio is lower in the che-
mical structure of SSP than in that of SP.
Zeta potential of cenent suspension
Fig.5 shows the result of measuring zeta potential.
The absolute values of zeta potential of SSP are higher than
those of SP by 10 mV when the solid concentration is 0.596,
indicating stronger electrostatic repulsion than SP. Yet the
values are smaller than those of NSF.
Dispersing capability of cement paste
Fig.6 shows the results of measuring the flow values of
cement pastes of ordinary portland cement with different dosages
of SSP, SP, and NSF. When W/C is 0 . 3 9 , SSP and SP shows similar
flow values, but as W/C decreases, the difference in the capabi-
lity becomes more evident, and SSP shows a higher flow value
than SP and NSF. In other words, the excellent cement dispersi-
A C 1 SP-148 94 0662749 052005b I113
286 Kinoshita et al
ng capability of SSP becomes evident as the water-to-cement

ratio becomes as low as 0.20-0.25.
Fig.7 shows the results of the similar tests conducted using
different types of cements. The dispersing capability of SSP
decreased in the different cements in the descending order of
FB, BB. M, N, and H. This suggests that the phase mineralogy of
cementitious materials influences the rheology.
Moderate heat cement and blended cement both showed excellent
dispersion with a small addition of SSP.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig.8 shows the flow values and plastic viscosity measured

by a Brookfield type viscometer of a cement paste of ordinary
portland cement (N) with a water-to-cement ratio of 0.20, with
and without silica fume of 10% by inner percentage of the
cement. Good dispersing capability was indicated even with a
smaller dosage of superplasticizers in cement pastes containing
silica fume than those without it.
The viscosity of the pastes was aïso lower with silica fume.
Cement containing silica fume are effective in obtaining slurri-
es o f low viscosity. In the case of NSF, the effect of viscosi-
ty reduction was smaller than that of SSP, even when the dosage
was increased.
Discussion on cement dispersing action
When compared with NSF. SSP has a greater surface-tension
-reducing effect, a smaller adsorption by cement, and a lower
zeta potential. When compared with SP, SSP has a similar sidace
-tension-reducing effect, a greater adsorption by cement, and a
greater absolute value zeta potential. These results concrete
well with the characteristics of the chemical structure of SSP,
e.q. having a smaller molecular weight than SP.
In other words, SSP has surface chemical properties midway
between those of NSF and SP.
It has been said that the fluidity of cement induced by the
addition of a superplasticizer results from the electrostatic
repulsion between cement particles to the molecules of the supe-
rplasticizer adsorbed on the surface of the cement (8).
However, the excellent fluidity of cement imparted by SSP in
spite of the substantially smaller adsorption and lower zeta po-
tential of SSP than those of NSF suggest that factors other than
electrostatic repulsion are also important.
The reason why the adsorptions of SSP and SP are smaller
than that of NSF may be that the form of adsorption of graft
copolymers by cement particles is different from that of linear
polymer such as h5F. In other words, graft copolymers are
spatially more extensive due to steric hindrance of the graft
chains, and thus they have greater coverage over the surface of
cement particles when they are adsorbed.
Consequently, the reason for the excellent dispersing capability
of SSP with a low water-to-cement ratio is considered to be due
to the synergistic effects of such factors as the wetting effect,
ellectrostatic repulsion, and stabilization of the adsorbed
A C 1 SP-lt48 94 0662947 0520057 05T
protective layer due to the steric hindrance of the graft chains

Fig.9 shows a dispersion model of mixture of cement and
silica fume by SSP. Conventionally, it has been maintained by
Bache at al (9) that the increase in the fluidity of cement
paste containing silica fume with an extremely low range water
-to-cement ratio is brought about by pore filling action of
silica fume, which is fully dispersed by the superplasticizer,
in the space between cement particles.
fonezawa et al (10) recently suggested that the increase in
fluidity by silica fume is mainly attributable to the ball
bearing effect of silica fume which coagulates and covers the
surface of the cement particles, as well as to electrostatic
repulsion. When SSF is used, in addition t o these factors, the
dispersing effects may be enhanced by the stabilization of the
adsorbed protective layer due to the steric hindrance of the
graft chains and by the moistness of the surface of the
particles effectively imparted by SSP.
It has been pointed out that the fluidity of concrete
containing naphthalene or melamine superplasticizers drastically
decreases at a certain period after mixing. The cause of this
slump l o s s , which has long been debated, is now considered to be
the physical agglomeration of particles (11). The stabilization
of the adsorbed protective layer hy the steric hindrance SSP
graft chains is effective in inhibiting this physical
agglomeration of the cement particles. In other words, SSP,
which comprises graft copolymer, is a superplasticizer having
advantages both in fluidization and fluidity retention.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Properties of mortar
Fluidity of mortar
The results of the measurement of fresh mortar proportioned
as presented in Table 2 are shown in Table 3 and Fig.10. With
a fluidity of a water-to-cementitious material ratio 0.17, the
flow values of the ordinary and high eary strength cements were
low, whereas those of the moderate heat and blended cements were
high. The measurement of the changes in flow values of mortar
over time with a water-to-cementitious material ratio of 0.20
indicated excellent stability of the flow values for at least 2
hours, excepting high early-strength cement, as shown in Fig.11.
Setting time
A s seen from Table 3 , the setting time of mortar containing
SSP showed no apprecible retarding properties. The results are
practically in the range of no effect.
A C 1 SP-148 94 Obb2949 O520058 T9b
288 Kinoshita et al
Heat evolution history

Fig.12 shows the adiabatic heat evolution of mortars
containing various cement. The ordinary and high early strength
cements led to a great temperature rise due to the heat of
hydration of the cement, whereas the moderate heat and blended
cements proved to be advantageous in reducing the heat of
hydration.
Cumpressive str- of mortar
Fig.13 shows the result of compression test. The specimens
cured in water of 20 deg C showed strength of 120-140 K a al: 28
days and 140-180 MF'a at 91 days.
(1) The new superplasticizer shows excellent cement dispersing

capability and imparts high flowability to cement at a lower
water-to-cementitious material ratio than conventional naphthal-
ene superplasticizers.
(2) The cement dispersing capability of the new superplasaticiz-
er is particularly evident with moderate heat and blended
cements. This implies that the cemetitious phase composition is
an important consideration for cement/superplasticizer compatib-
ility
(3) The new superplasticizer reduces the surface tension of
water more than conventional naphthalene superplasticizers. The
adsorption of the new superplasticizer by cement is smaller than
those of naphthalene superplasticizers, but greater than those
of polycarboxylate superplasticizers. The zeta potential is
lower than those of naphthalene superplasticizers, but higher
than those of polycarboxylate superplasticizers. Thus the new
plasticizer has the properties midway between those conventional
naphthalene type and polycarboxylate types.
(4) The properties of the new superplasticizers are closely
related with the chemical structure of the graft copolymer cons-
tituting the.superplasticizer. The factors affecting the prop-
rties are considered to include molecular weight, the percentage
of sulfonic group, carboxyl group and anion group, and such
balances as the percentage and length of the graft chains.
(5) ultra high strength silica fume mortar containing the new
superplasticizer with a water-to-cementitious material ratio of
0.20 provides fluidity sufficient for placing a high compressive
strength concrete exceeding 100 MPa.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 W 06629LI7 0520057 922 =
REFERENCES
1. Hattori,K., “Experiences with Mighty Superplasticizers in

Japan”; AC1 Special Publication SP-62, 1979, pp.37-66.
2. Ramachandran, V.S. and Malhotra,V.M. (1984)
“Superplasticizers”, in Concrete Admixtures Handbook, Editor
V.S.Ramachandran, Noyes Publication, Park Ridge, New Jersy,
USA, pp.211-268.
3 . K.Mitsui, H.Kasami, Y.Yoshioka, and M.Kinoshita, “Properties
of High-Strength Concrete with Silica Fume using High-Range
Water Reducer of Slump Retaining Type”; American Concrete
Institute, SP-119, 1989, pp.79-97.
4. Kinoshita,M.,Yamaguchi,S.,Yammoto,T. and Tomosawa,F.,
“Chemical Structure and Performance of new type High-Range
Water-Reducing Agent”; CAJ Proceedings of Cement & Concrete,
No.44, 1990, pp.222-227.
5. Kinoshita,M.,Yonezawa,T. and Yuki,Y. “Chemical Structure
and Performance of a new type High-Range Water-Reducing
Agent for Ultra High Strength Concrete”;The 47 Annual
Meeting of JCA, 1993, pp.206-211.
6. Ito,K., “Synthesis and Application of Macromonomers”;
Journal of the Adhesion Society of Japan ,Vo1.24,No.9,
1988, pp.29-39.
7. U.S.Patent No. 4,962,173 , NO. 5,087,648
8 . Daimon,M. and Roy, D.M., “Rheological Properties of Mixes :
li Zeta Potential and Preliminary Viscosity Studies”;
Cem. Con. Res. vo1.9, 1979, pp.103-110.
9. Bache H. H., “Densified Cement / Ultra-fine particle-Based
Materials, Second International Conference on Superplastici-
zers in Concrete, June, 1981 Ottawa, Canada.
10. Yonezawa,T.,Yanagibashi.K., Ikeo,Y. and Asakura.E.,
“Dispersion of Silica Fume in High Strength Concrete”;
Proceedings of the Japan Concrete Institute, ~01.15,No.1,
1993, pp.69-74.
11. Hattori,K.. “Mechanism of Slump Loss and Its Control“;
Journal of The Society of Materials Science, Japan, ~01.29,
1980, pp.240-246.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0 6 6 2 7 4 9 05200b0 b 4 4 =
290 Kinoshita et al
TABLE 1 - CHEMICAL COMPOSITION AND PHYSICAL PROPERTIES OF CEMENTITIOUS MATERIALS
Chemical compaiitlon 06) Blanc Svccl-

-
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Binde - t i c
KI O :ofai (d//a) gravlty
-
o. 2 91. 8 8280 3. 1 6
o. 5 98.4 4240 3.13
o. 4 98. 6 8880 3. 2 0
o. 4 98. o 3790 3. o 5
F B o. 4 81. 8 9480 2. 9 1
............. ....... ............ .....,,,,....... ...................
-
SF 1. 1 96. 7 141000 2. 3 6
TABLE 2 - MIX PROPORTIONS OF MORTAR
N.H,M 17 10 56.5 8.6 51.5 5.8 32.4 1.7

BB. F B
A C 1 SP-348 9 4 0662947 0520063 5 8 0
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 3 - PROPERTIES OF FRESH MORTAR
* Setting time
Dosage FI ow Air (hou r 6 -min)
Of SSP content
Cement (CIS0 X 96 (u) (96) initial final
20 3. 3 290 1. 5
H 20 10 3.3 275 1. 8 7-20 8-55
M 20 10 2 . 1 295 1. 2 11-05 12-50
BB 20 10 2. 2 310 1. 3 14-10 16-20
FB 20 10 2.2 303 1. o 10-15 12-00
* S S P = 3 O%a 9.
A C 1 SP-148 94 m 0662949 05200b2 417 m
292 Kinoshita et al
CHa
I
C-CHI
I
X
I
SOS N ô
HCH
X=CH, and - C H p O - O HEH
R = H or
Fig. 1-Chemical strudure of acrylic graft copolymer based new superplasticizer (SSP)
I I l I I
S S P ( Yn=3900, lhr=16000 1
S P ( Yn=4200. b21500 1
1
1o2 1 O' 104 105 1 O6
Molecular leight
Fig. Z-Molcular weight distribution of graft copolymer (SSP, SP)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 W 0662949 05200b3 353
( in aq. a t 209: 1 LNSF

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
o o. o 1 o. 1 1. o
C o n c e n t r a t i o n ( solid % )
Fig. 3-Surface tension of aqueous of superplasticizers (SSP, SP, NSF)
A C 1 SP-148 94 0662949 05200b4 29T W
294 Kinoshita et al
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 7 4 Obb27Ltï 05200b.5 1 2 b
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
400
W/C=O, 30 W/C=O, 2 5 w/c=o. 2 0 -
300
200 NSF-
1O 0
I I
-- I I
o 0.51.00 0.51.0 o 0.51.01.5
Dosage (CX%)
Fig. b f l u i d i t y of ordinary poitland cement paste
e
e
Y 400
w/c=o*20
3 FB
O
Li 300
o
c>
OI
(d
a 200
4
CI
I
o
e 100
pl
u o o. 5 I. o 1. 5 2. o
Dosage (Cx%) of SSP
Fig. 7-Fluidity of each type of cement paste
A C 1 SP-148 94 = 0662949 0520066 Ob2
296 Kinoshita et al
E!
e
Y
400
-
3
O
Lr,
300
dJ
3
rA
d
a 200
c>
CI
al K4
Y
edJ 100 O
U
O 1 2 3 4 5 6 7
Dosage (Cx%)
Fig. LViscosity and flow of cement paste
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0bb2949 O520067 T T 9
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. W d s o r p t i o n model of graft copolymer (SSP) onto the cement particles mixed with silica fume
A C 1 SP-148 9 4 M Obb29Y9 0520068 935
298 Kinoshita et al
300
200
100
N H M BB FB
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Type of Cement
Fig. 10-Mortar flow for each type of cement
350 I .................... ! I I
..................................................................
250
ay
3 0 0 - ...................................
W/ ( C + S F ) =20%
SF/(C+SF) =lo%
at20.c
200 ...................................................................
o o. 5 1. o 1. 5 2. o 2. 5
Time a f t e r mixing (hrs.)
Fig. 1 1 4 o r t a r flow versus time after mixing
A C 1 SP-148 94 = Ob62949 0520069 ô71
v j sF/(CtsF)=lO x
O 10 2 CI 30 40
Time (hrs)
Fig. 12-Heat evolution curves ot fresh mortar using each type of cement
-
F:
M
c
200
W/ (C+SF)=20%;
al
L
* n1 5 0
a
cd
Qia
>z
.- ? 1 0 0
In
In
Qi
ti
a 50
E
O
o
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
N H M B B F B
Type of Cement
Fig. 13-Compressive strength development of hardened mortar using each type of cement
A C 1 SP-148 9 4 IObb2949 0 5 2 0 0 7 0 593
SP 148-17
Split Addition of a HRWRA and its

Effeci on High-Petformance Concrete
by JJ. Schemmel, Y. Arora, and J. Williams
Svnopsis; High-performance concretes (HPC's)typically have low w/c's in order to

achieve the desired levels of strength and durability. As a result, H K ' s have a
tendency to be stiff and to lose their workability rather quickly. Often high-range
water-reducing admixtures (HRWRA) are used to improve the workability of
HPC. Care must be exercised when using any admixture, or combination of
admixtures, to insure that there are no detrimental side effects which might shorten
the life of the concrete.
Research has shown that although retempering concrete with a HRWRA will
generally improve the workability and maintain the strength of low w/c concretes,
it may also reduce the resistance to freezing and thawing. Thus, an experimental
study was undertaken to evaluate the use of a naphthalene based HRWRA in the
production of a high-early strength (HES) HPC. in a laboratory investigation,
dosing of the HRWRA was split to mimic both batch plant and job site addition.
This research stems from work conducted at the University of Arkansas as part of
Strategic Highway Research Program (SHRP) contract C-205, "Mechanical
Behavior of High Performance Concretes .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Keywords: Admixtures; coarse aggregates; concretes; freeze thaw durability; hiah performance
concretes; strength; water reducinp wents; workability
A C 1 SP-348 94 Obb2949 0520073 4 2 T
302 Schemmel, Arora, and Williams
John J. Schemmel is an Associate Professor of Civil Engineering at the University

of Arkansas. He received his Ph.D. from North Carolina State University. His
research interests include structurai materiais and design. He has taken a
sabbatical to work at the Corps of Engineers' Waterways Experiment Station. He
is a member of AC1 Committees 343 and C 610.
Vikas Arora received a Masters of Science degree in Civil Engineering jn 1993

from the University of Arkansas, Mr. Arora is now employed by private industry.
Jennifer Williams received a Bachelor of Science degree in Civil Engineering from

the University of Arkansas in 1993. Currently, Ms. Williams is pursing a Masters
at the University of Arkansas. Upon completion of her post-graduate studies, Ms.
Williams will be taking a position with the Arkansas Highway and Transportation
Department.
INTRODUCTION
A recently completed study examined the production and behavior of a number of

high-performance concrete (HPC) mixtures. The four-year study, conducted for
the Strategic Highway Research Program (SHRP), included determining the
proportions for five HPC mixtures, an extensive laboratory investigation of the
mechanical properties of these mixtures, as well as the construction and monitoring
of five field installations. Details of aü aspects of this research program can be
found elsewhere (1,2,3,4).
The HPC mixtures developed for the SHRP study are intended for use in highway
pavement and bridge applications. Three classes of HPC were developed, i )
Very-early strength (VES) for use in rapid pavement repairs, 2) High-early
strength (HES) for use in patching, bridge decks, and overlays, and 3) Very-high
strength (VHS) for use in bridge girders and piers. Ali three classes of concrete
were proportioned with w/c's of 0.35 or less and are to have durabiiity factors of
80% or more after 300 freezing and thawing cycles. The concretes have
compressive strength requirements of at least 14 MPa in 6 hours for the two VES
mixtures, at least 34 MPa in 24 hours for the HES mixture, and at least 69 Mpa in
28 days for the two VHS mixtures. Each class of concrete was produced using
four different coarse aggregates and two High-range water-reducing admixtures
(HRWRA). The engineering properties of ail five concrete mixtures were studied
in the laboratory. Field installations were consmcted using only the HES mixture.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662949 0520072 3 6 6 =
For ail three classes of concrete, it was important that the laboratory developed
mixtures be easily reproduced in the field. This, along with the performance
criteria stated above, influenced the selection of the type and quantity of materiais
used, the recommended batching and mixing procedures, and the curing methods
employed. A review of relevant literature highlighted potential problems with the
batching, mixing, and placing of low w/c concretes.
Research has shown that low w/c concretes typically require the use of a HRWRA
in order to achieve the desired level of workability at the job site. Typically, the
admixture is either added in full or retempered at the job site to overcome the
problem of rapid slump loss. AC1 212 (5) presents a two-step method for adding a
HRWRA "...when the concrete is placed in slabs that must be fmished by
troweling". A portion of the total quantity of admixture is added at the batch plant
"...to assure control of water..." and then again in the field "...for placing
purposes". Studies indicate that delayed addition and retempering with a HRWRA
can lead to increased workability and higher compressive strengths. However, as
summarized below, some studies suggest that job site addition and retempering
may have a deleterious effect on long-term resistance to freezing and thawing.
Smutzer and Zander (6)) examined the effect of retempering concrete mixtures
with a HRWRA. One of the mixtures considered in their research consisted of
Type III portland cement, an air-entraining admixture (AEA), and a HRWRA.
They found that when this concrete was retempered with a HRWRA, in order to
improve workability at the job site, the relative durability factor consistently fell
within a range of 50 to 70 percent. Smutzer and Zander recommended that
concretes containing a HRWRA not be retempered in situations requiring good
freezing and thawing durability. This recommendation was adopted by the Indiana
Department of Highways who sponsored the research.
Sprinkel (7) reported on the performance of a number of field installations

constructed for the Virginia Department of Highways and Transportation
(VDHT). Some of these installations involved the use of a naphthalene-based
HRWRA which was added entirely at the job site and subsequently retempered in
some cases. Sprinkel found that specimens cast at the job site and later tested for
their freezing and thawing resistance had low durability factors as compared to
concrete which did not contain the HRWRA.
These, and other studies, have raised questions about the durability of concretes
which have a HRWRA added or retempered at the job site. As a result, a decision
was made that job site addition of a HRWRA would not be permitted for the
concretes developed as part of the SHRP C-205 research program. Until further
durability studies could be conducted, any HRWRA required for a mixture would
be added only at the batch plant. Also, batch plant addition of the HRWRA was
felt important due to the low w/c and water content of these concretes.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 = 0662949 0520073 2 T 2 =
This paper reports on the results of a laboratory study which examined the effect
of splitting the addition of HRWRA between the batch plant and job site on the
performance of the concrete. The dosage sequence ranged from 100%batch plant
addition to 100% job site addition. In no case was the concrete retempered with
HRWRA. The investigation looked at the effect of the dosage sequence on the
workability, compressive strength, and resistance to freezing and thawing of the
concrete.
PROPORTIONS, MATERIALS, AND PRODUCTION
The HES mixture was the only type of HPC considered in this research.
Proportions for this concrete were originally developed by researchers at North
Carolina State University (NCSU) (i). The mixture is based, in part, on the
premise that the number of constituents, and the amount of each, should be kept to
a minimum without sacrificing performance. Modifications were made to NCSUs
basic proportions to account for the characteristics of the constituents used in this
study. The resulting proportions, along with typical fresh properties, are given in
Table 1. These proportions are based on a 19 mm nominal maximum sized
limestone coarse aggregate and natural river sand as the fine aggregate. Type ïïi
portland cement was used for its early strength development characteristics. A
HRWRA was used to achieve a low w/c. A calcium nittite accelerator was used to
help promote early strength gain. This non-chloride admixture was also chosen for
its reported corrosion retarding qualities as the HES mixture has potential for
bridge-deck applications.
A conventional water-reducing admixture was not considered for the HES

concrete due to the potential for significant slump loss and set retardation when
used at high dosages. In addition, a combination of conventional and HRWRA
was not used as this would added to the number of constituents needed to produce
the concrete. Fiially. so called "extended set" water reducers were not used due
to the fact that the applicationsthe HES concrete had been developed for required
rapid setting and rapid strength development.
Water present in the non-chloride accelerator was accounted for in computing the
batch quantities for this concrete. The particular accelerator used in this research
is a 30% solution of calcium nitrite. Thus, each gallon of accelerator contributes a
substantial amount of water to the mixture. This water, along with any free
aggregate moisture, must be subtracted from the 180 kdm3 total mixing water to
insure that the proper amount of water is added to the mixture. Typically, about
160 kg of water are batched for a cubic metre of concrete. Water present in the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 œ 0662949 05200'74 L39 œ
HRWRA was not accounted for in determining the batch quantities for simplicity.
Relative to the non-chloride accelerator, the HRWRA contributed little additional
water.
Cement
The cement used in this study was a Type III portland cement furnished by Blue
Circle Cement from their plant in Tulsa, Oklahoma. The chemical composition and
physical characteristics of the cement, as provided by the manufacturer, are given
in Table 2.
@re pates
Locally available aggregates were used in this research. The coarse aggregate was
a 19 mm nominal maximum size dense limestone. This aggregate came from the
McCiinton-Anchor quarry in West Fork, Arkansas. The fine aggregate used was
Arkansas River sand and came from Van Buren, Arkansas. The grading and
physical characteristics of both aggregates, as determined by the authors, are
presented in Tables 3 and 4. Experience has shown these aggregates to be
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
durable.
Admix tures
Three chemical admixtures are used in the production of the HES concrete
considered. They include a Vinsol-resin air-entraining admixture, a naphthalene-
based HRWRA, and a calcium-nitrite set acceleratorkorrosionretarder. DaravairTM
and DCITM were used as the AEA and set accelerator, respectively. Both are W R
Grace products. PSI SuperTM,supplied by Cormix, was used for the HRWRA.
The dosage rates for all three admixtures, shown in Table 1, were establish based
on manufacturer's recommendations and extensive trial batching. Although the
dosage rates are higher than is typical for conventional concrete, the least amount
of each to achieve the desired performance was used.
Batchinp. Mixinc. and Curin

. Procedures
The primary variable in this research was the dosage sequence for the HRWRA.
The quantity of HRWRA added to the concrete remained constant throughout the
A C 1 SP-148 9 4 m 0bb2949 0520075 075 m
investigation. The control condition was defined as the case where 100% of the
HRWRA was added at the batch plant and none was added at the job site. The
addition of the HRWRA was then split between the batch plant and job site. The
amount of HRWR4 added at the batch plant was reduced in increments of 25% of
the total dosage until 100% of the HRWRA was being added at the job site and
none was added at the batch plant. As a result, five dosage sequences were
considered 1) 100-0, 2) 75-25,3) 50-50,4) 25-75,and 5) 0-100,where the first
values is the percent of HRWRA added at the batch plant and the second value is
the percent added at the job site.
Three series of tests, A, B, and C, were conducted for replication. Ali batches
were mixed in a 0.17 m3 revolving-drum mixer. Batch sizes were 0.045 m3 for
series A and B and 0.057 m3 for series C. Corrections to the water content were
made for each batch of concrete to account for free aggregate moisture and water
in the calcium niîrite only.
Batching and mixing of the control concrete occurred as follows. First, the mixer
was charged with about 20 litres of water and allowed to mix for a few minutes at
which time the water was discharged. Next, roughly 112 of the coarse aggregate,
2/3 of the fine aggregate, U 3 of the water, and ali of the AEA were added to the
mixer. The AEA was pre-mixed with the mixing water. The mixer was run for
one minute at mixing speed and then stopped. Au of the cement followed by the
remaining coarse and fine aggregate were added. The HRWR4 was pre-mixed
with the remaining water and also added. For the record, the start of the mixing
process was as taken as the time at which the cement and final portion of water
were combined in the mixer. The concrete was then mixed at mixing speed for 5
minutes. This part of the batching and mixing process was intended to simulate
production at the ready-mixed concrete batch plant.
After the initial mixing of all materials was complete, the mixing action of the drum
was slowed to simulate agitation of the concrete during transport. This slow roll
mixing action was continued for 20 minutes. At the end of the 20 minute agitation
period, the calcium nitrite solution was added. Adding the calcium nitrite at this
time was to simulate job site addition. The concrete was then mixed at mixing
speed for an additional 5 minutes.
In order to accommodate the split addition of the HRWRA, the batching and
mixing process had to be slightly modified. Following the addition of the cement,
and remaining aggregate portions, a pre-determined amount of HRWRA was
mixed with the water and added to the concrete. The remaining portion HRWRA
was added just prior to the calcium niîrite solution. The concrete was thoroughly
mixed after the HRWRA was added and before the calcium nitrite was added.
*Inail cases, once batching and mixing was complete the concrete was discharged
in to a large non-absorptive pan. Fresh concrete properties were measured and
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 94 Obb2949 0520076 T O 1
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
specimens cast for future testing. All specimens were cured in their molds for 24
hours at 23" C. After 24 hours the specimens were removed from their molds and
placed in sealed plastic bags and maintained at 23' C until tested. These same
batching, mixing, and curing procedure were used throughout the SHRP C-205
investigation.
TESTING
Fresh Concrete ProDerties
in ail cases, the properties of the fresh concre1.e were evaluated immediately after
the concrete was discharged from the mixer. The elapsed time from the start of
mixing to discharge was the same for all dosage sequences. Slump, air content,
unit weight, and temperature were determined according to the appropriate ASTM
test methods by AC1 certified field testing technicians. The air content was
determined using a pressure meter. Changes in the fresh concrete properties over
time were not monitored.
Commessive StrenPth
Nine 102 by 203 mm cylinders were cast for compressive strength testing. Plastic
molds with lids were used in this process. Three cylinders were tested at each of
the foilowing ages, 1, 14, and 28 days. The cylinders were tested for their
compressive strength using a Soiltest 1,780-kN hydraulic machine. Unbonded
neoprene pads confined in steel rings were used as the end preparation.
Freezing and Thawing
Four 76 mm by 76 mm by 406 mm prisms were cast for freezing and thawing tests.
Three prisms were subjected to between 5 and 8 freezing and thawing cycles per
day. Testing was conducted as per ASTM C 666 Procedure A, rapid freezing and
thawing in water. The fourth prism was placed in a refrigerator, covered with
water, and maintained at 4.4 f 1.7 "C throughout the testing program. Specimens
were placed in the testing chamber or refrigerator at 14 days of age. Testing
continued until the specimens had been subjected to 300 freezing and thawing
cycles or the durability factor fell below 80%, whichever occurred first.
A C 1 SP-14B 9 4 O b 6 2 9 4 9 0520077 948 =
At roughly 30-cycle intervals, prisms in the freezing and thawing chamber were
stabilized at 4.4 f 1.4 OC. One by one, ail prisms were removed from the testing
chamber or refrigerator, their mass determined in air and in water, and they were
tested for transverse resonant frequency and dynamic modulus was calculated as
per ASTM C 215. Each specimen was tested for its resonant frequency at least 5
times to ensure replication of the data. It was often difficult to obtain replicate
frequency measurements in cases were extensive freezing and thawing damage had
occurred. In such cases a visual inspection of the prism confirmed the
deterioration of the concrete. The location of each specimen was shifted one
position upon its return to the freezing-and-thawingchamber.
Results from tests on the fresh and hardened concrete for all three series are
presented in Tables 5 and 6. Due to a batching error, there is no data for the 75-
25 dosage sequence in Series A. Also in Series A, no freezing and thawing control
specimen was prepared for the 0- 100 dosage sequence.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fresh Concrete Pro-perties
The fresh concrete properties for ail @ree test series are given in Table 5. The
slump ranged from 50 mm to 220 mm and the air content ranged from 3.2% to
11.6% for the entire test program. in general, the measured slump was higher
when a greater proportion of the HRWRA was added at the job site. Thus,
relative to the control condition (all the HRWRA added at the batch plant) the
total slump loss decreases as the proportion of HRWRA added later increases.
Aithough a few air content measurements were found to be on the high side, the
air content of the fresh concrete appears to be relatively unaffected by the split
addition of the HRWRA. This is particularly visible with the batches from Series
C. There may even be a very slight increase in the air content when all of the
HRWRA is added at the job site. in two of the three cases, the air content
increased for the 0-100 batch as compared to the 25-75 batch.
Although not reported here, the unit weight measurements were consistent with
the measured air contents. The temperature of the batches in Series A and B
ranged from 25' to 29' C. For Series C the concrete temperature was 18' C for
all five batches.
A C 1 SP-148 94 0662949 0520078 884
ComDressive Wen&h
The results for the compression tests are given in Table 5. in ail cases, the
compressive strength of the concrete increased with age as expected. For alì
batches, very little increase in the compressive strength occurred after 14 days.
only two batches had 1-day compressive strengths below the desired level of 34
MPa. These strengths are italicized in Table 5. The high air content, 11.6%, of
Series A batch 50-50 is likely the cause of its low strength. Series A batch 0-100
had a l-day strength only about 1% below the desired level. Thus, except for
Series A batch 50-50, ail batches produced can be considered to have satisfied the
strength requirement for HES concrete.
Overall, the compressive strength of the concrete does not appear to be affected
directly by the HRWRA dosage sequence through 28 days of age. Rather, changes
in the compressive strength can be attributed to differences in the air content of the
concrete. For ail three test series, increases in compressive strength are matched
with decreases in air content and vice versa.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Freezing and Thawing
Table 6 presents results from the freezing-and-thawing tests. For each batch of
concrete, the durability factor (DF) was determined for each individual test
specimen. An average DF, for the three test specimens, was also computed. The
DF's were determined based on equations found in ASTM C 666 and C 215. The
measured fundamental transverse frequencies were used to compute values of
relative dynamic modulus (RDM) which were in turn used to calculate the DF's.
The relative increase in the dynamic modulus for the control specimen was also
computed and is presented in Table 6.
The results from the freezing-and-thawing tests are mixed. In most cases, the
computed DF's are weil above the minimum level specified for HES concrete (80%
after 300 cycles of freezing and thawing). Excluding those specimens which did
not satisfy this performance specification, the DF's ranged from 98% to 119% with
most values falling in the 100% to 110% range. Specimens whose DF feii below
the 80% minimum level are given in italics in Table 6.
In Series A testing, ail three specimens from the 0-100 batch had DF's below 50%.
The average DF for the entire batch was 40%. In Series C, two specimens from
the 25-75 batch had DFs below about 30%. However, in this instance the
remaining specimen had a DF of 100%. The average DF for the entire batch was
50%. Smutzer and Zander (6) found the average DF for low w/c concrete
retempered with a HRWRA to range from 49% to 67%. Their retempering was
A C 1 SP-148 74
0662547 0520079 710
comparable to the 75-25 dosage sequence used in this research. in the field work
described by Sprinkel (7), which corresponds to a 0-100 dosage sequence, the
average DF ranged from 8% to 70%. Thus, the data would appear to suggest that
there may be a problem with frost resistance in low w/c concretes in which the
addition of a HRWRA is split or delayed.
At this time, an analysis of the air-void system in the hardened concrete specimens
tested has not been conducted but is scheduled for the future. However, the air
content of the fresh concrete may provide an indication of why some of the
concretes investigated in this work had poor frost resistance. in a summary of
research on the freezing and thawing resistance of concretes containing HRWRA's
(8), it is noted that concretes containing a HRWRA can be frost resistant so long
as they are properly air entrained. From the tabulated data within the published
report, one can see that frost resistance is low for concretes which contain a
HRWRA and have an air content less than about 3.8%. Siebe1 (9) notes that it
may be necessary to increase the minimum air content of concretes containing a
HRWRA to insure that they are frost resistant. He suggests the specified air
content be about 1% higher for mixtures which employ a HRWRA as compared to
those that do not contain the admixture. Another option might be to maintain
currently specified ranges for air content based on maximum aggregate size but
require the air content of HRWRA concrete to be in the upper half of the range.
Looking again at the data in Table 6, the two batches of concrete which had low
DF's (Series A 0-100 and Series C 25-75) also had low air contents. The 0-100
batches from Series B and C and the 25-75 batch from Series A, aii of which had
higher air contents, were found to be frost durable. Other batches with low air
contents were found to be durable. However, these batches had an equal or
greater portion of the HRWRA added at the batch plant. Thus, it may be that
lower air contents are not as serious when the majority of the H R m is added at
the batch plant. As more of the admixture is added at the job site, the importance
of small changes in the air content of the fresh concrete and air void spacing of the
hardened concrete may be magnified. An air content between 5% and 8% seems
to be adequate for strength and durability. Further study of this issue is needed.
CONCLUSIONS
The main benefit of splitting the addition of a HRWRA between the batch plant
and job site is an improvement in the workability of the concrete. SigniFcantiy
higher slumps were found when the proportion of HRWRA added after 20 minutes
of mixing was greater than that initiaüy added. The air content of the fresh
concrete was relatively unaffected by the split addition. Compressive strength
appears to be influenced more by the air content of the concrete than by the dosage
sequence of HRWRA. Resistance to freezing and thawing is good when an equal
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662747 0520080 4 3 2
or greater proportion of the HRWRA is added at the batch plant. There remains
some question as to the frost resistance of a concrete which has more of its
HRWRA added at the job site. A properly &-entrained concrete having an
adequate air-void system is the key to frost resistant concrete, regardless of the
dosage sequence. An air content of between 5% and 8% will probably provide
such a system. Field data indicates that this range in air content will not adversely
effect the strength developing characteristicsof a high-early strength concrete.
RECOMMENDATIONS
Based on the findings from this research it is recommended that no more than 50%
of the total amount of a HRWRA be added at the job site. At this time, there is
some degree of uncertainty regarding the durability of low w/c concretes which
have more than this proportion of admixture added at the job site. For low w/c
concretes with split addition of a HRWRA, the air content of the fresh concrete
should be specified between 5% and 8%. With this air content it should be
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
possible to produce a durable concrete with the desired level of strength.
Further study of the effects of a split HRWRA dosage sequence on the properties
of HPC is recommended. In particular, how the dosage sequence effects air
content, what impact a longer agitation period would have on performance, and
whether or not supplemental field dosages would adversely effect performance are
just a few of the issues that should be addressed.
REFERENCES
1. Zia, P., Leming, M.L., Ahmad, S.H., Schemmel, J.J., and Elliott, R.P.,
"Mechanical Behavior of High Performance Concretes, Volume 2: Production
of High Performance Concrete," SHRP-C-362, Strategic Highway Research
Program, National Research Council, Washington, DC, 1993.
2. Zia, P., Ahmad, S.H., Leming, M.L., Schemmel, J.J., and Elliott, R.P.,
"Mechanical Behavior of High Performance Concretes, Volume 3: Very Early
Strength Concrete," SHRP-C-363, Strategic Highway Research Program,
National Research Council, Washington, DC,1993.
"Mechanical Behavior of High Performance Concretes, Volume 4: High Early
National Research Council, Washington, DC,1993.
A C 1 SP-148 94 D Obb2949 0520081 379 D
"Mechanical Behavior of High Performance Concretes, Volume 5 : Very High
National Research Council, Washington,DC,1993.
5. AC1 Committee 212, "Guide for the Use of High-Range Water-Reducing

Admixtures (Superplasticizers) in Concrete," AC1 212.413-93, AC1 Manual of
Concrete Practice, Part 1, 1993.
6. Smutzer, R.K., and Zander, A.R., "A Laboratory Evaiuation of the Effects of
Retempering Portland Cement Concrete With Water and a High-Range, Water
Reducing Admixture," Transportation Research Record, No. 1040, TRB,
National Research Council, Washington, D.C., 1985, pp. 34-39.
7. Sprinkel, M.M., "Super-Water-ReducedConcrete Pavements and Bridge Deck

Overlays," Transportation Research Record, No. 720, TRB, National Research
Council. Washington, D.C., 1979, pp. 19-28.
8. Attiogbe, E.K., Nmai, C.K., and Gay, F.T.,"Air-Void Parameters and Freeze-
Thaw Durability of Concrete Containing Superplasticizers,"Durable Concrete,
Compilation 24, American Concrete Institute, Deboit, 1993, pp, 41-45.
9. Siebel, E., "Air-Void Characteristics And Freezing and Thawing Resistance of

Superplasticized &-Entrained Concrete With High Workability,"
Superplasticizers and Other Chemical Admixtures in Concrete, AC1 SP-119,
V.M. Malhotra, editor, Detroit, 1989, pp. 297-319.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 0 6 6 2 9 4 9 0520082 205
TABLE 1 - HES MIX PROPORTIONS

Matenai Dry Weight Proportions
Wm3)
Cement 515 kg
Water (1) 180 kg
Coarse Aggregate 1,000 kg
Fine Aggregate 610 kg
HRWRA (Naphthalene) 5.2 L
AEA (Vinsol Resin) 1.4 L
Calcium Nitrite 20 L
(I) Adjust for free aggregate moisture and water in
Calcium Nitrite.
TABLE 2 - PHYSICAL CHARACTERISTICS OF TYPE 111 CEMENT
Chemical Analvsis %
Loss on Ignition 1.08
insoluble Residue 0.21
Calcium Oxide (Cao) 63.44
Silica (Si02) 21.05
Alumina (AI203) 4.95
Sulfur Trioxide (SO?) 3.98
Iron Oxide (Fe2O3) 2.37
Magnesia (MgO) 2.16
Potassium Oxide (K20) 0.73
Sodium Oxide (Na20) 0.52
Na20 Equivalent 0.90
Phases %
Tricalcium Silicate (C3S) 50.15
Dicalciaum Silicate (C2S) 22.52
Tricalcium Aluminate (C3A) 10.06
Tetracalcium Aluminoferrite (C4AF) 7.2 1
Compressive Strength (MPa)

1 day 23
3 days 32
7 davs 38
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0520083 LYL
TABLE 3 - AGGREGATE GRADING

Percent Passing
Sieve Size Coarse Aggregate Fine Aggregate
19 mm 100.0
12.5 mm 82.0
9.5 mm 48.0
4.75mm 5.6 96.0
2.36mm 2.8 88.0
1.18 mm 2.4 75.0
600 pm 55.0
300 pm 12.5
150 pm 0.4
TABLE 4 - PHYSICAL CHARACTERISTICS OF THE AGGREGATE
Property Coarse Aggregate Fine Aggregate

Specific Gravity (SSD) 2.69 2.61
Absorption(%> 0.85 0.65
Unit Weight (kg/m3) 1,590
Fineness Modulus 2.72
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 W 0662947 0520084 O ô 8 U
TABLE 5 - FRESH PROPERTIES AND AVERAGE COMPRESSIVE STRENGTH

Dosage Slump Air Compressive Strength
Sequence Content (MPa)
(mm) mo) Id 14d 28d
Series A
100-0 65 3.7 40.8 52.3 52.7
50-50 220 11.6 26.3 37.4 40.1
25-75 220 7.4 36.8 53.0 54.8
0-1O0 200 3.5 33.7 48.0 52.8
Series B
100-0 120 6.5 34.1 46.9 49.2
75-25 115 4.8 44.9 56.8 58.0
50-50 120 3.5 50.5 62.3 63.1
25-75 150 3.2 51.6 62.7 64.5
0- 100 200 5.3 49.0 60.7 61.9
Series C
100-0 50 3.6 39.7 52.7 58.9
75-25 85 3.5 41.9 58.5 60.0
50-50 110 3.5 41.8 60.5 61.5
25-75 115 3.4 42.0 61.7 64.9
0- 100 200 5.7 37.0 59.8 62.2
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 m Obb2949 0520085 T L 4 m
TABLE 6 - FREEZING AND THAWING TEST RESULTS

Dosage Air Durability Factor
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Sequence Content Control 1 2 3 Average
Series A
100-0 3.7 117 119 116 117 117
50-50 11.6 113 106 101 102 103
25-75 7.4 106 110 101 101 104
o- 100 3.5 NA 24 48 48 40
Series B
100-0 6.5 108 110 107 105 107
75-25 4.8 113 106 104 106 105
50-50 3.5 110 106 106 101 104
25-75 3.2 112 109 103 98 103
0-100 5.3 109 104 108 100 104
Series C
100-0 3.6 114 109 105 107 107
75-25 3.5 110 103 104 104 104
50-50 3.5 109 104 104 102 103
25-75 3.4 107 31 100 19 50
o- 100 5.7 110 105 104 100 103
A C 1 SP-148 74 0662949 0520086 950
SP 148-18
Effect of Moist Curing and

Modifier Content on Performance
in Freezing and Thawing of Two
Latex-Modified Concretes
by C.M. Newtson and DJ. Janssen
synopsis:
Two different types of latex modifiers were used to determine how curing
conditions and latex-modifier content influenced compressive strength and
durability. Resistance to fkeezing and thawing in the presence of a 2.5 percent
NaCI solution was evaluated by measuring both relative dynamic modulus and
mass of scaled material. W e all specimens were 14 days old at the start of
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
testing, compressive strength increased as the period of initial wet curhg

increased, durability factor values (ASTM C 666) were insensitive to the period
of initial wet curing, and scaling resistance was improved by increasing the wet
curing time. The a u void system, described by the spacing factor, was found to
have a greater innuence on durability and scaling than either latex-modifier
content or duration of wet cure. A control mixture made using a high-range
water-reducing admixture (HRWRA) was used as a basis of comparison.
Kewords: Air entrainment; compressive strength; freeze thaw durability; (atex; moist curing;
scaling; superplasticizers
A C 1 SP-148 94 = Obb29i19 0520087 8 9 7 =
318 Newtson and Janssen
Craig Newtson received BSc and MSc degrees fiom the University of Wyoming
and is currently a PhD student in the Department of Civil Engineering at the
University of Washington. His research interests include nondestructive
evaluation of portland cement concrete materiais and structures.
AC1 Member Donald J. Janssen is an Associate Professor of Civil Engineering at

the University of Washington, Seattle, Washington. Dr. Janssen is a member of
AC1 Committee 201, Durability; and a member of RiLEM TC 117, Freeze-Thaw
and Deicing Resistance of Concrete.
INTRODUCTION
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Latex-modified concretes were originally developed in the 1920's as an
alternative to air-entrained concrete with lower permeability[l]. These latex-
modified concretes have been used in applications such as bridge decks[2] where
the impermeability of the concrete enhances the protection of reinforcmg steel
against corrosion. Other properties that have been attributed to latex-modified
concrete include increased tende and flexural strength[l] and increased
resistance to scaling in the presence of deicer salts[3].
It has also been shown that strengths of latex-modified concretes may be

optimized by wet curing the concrete for some time and then aiiowing it to &y
for several days[l,4]. In i d e n t w g this characteristic of latex-modified
concretes, only three merent wet curing periods were studied. Previous studies
investigating the effects of this wet curing period followed by a drying period on
properties other than strength have been quite qualitative in nature.
RESEARCH SIGNIFICANCE
This study was conducted to determine how physical properties of latex-

modified concretes, such as strength, dynamic modulus, and resistance to
deterioration and scaling during fieezing and thawing, vary with latex content
and length of wet cure. Quantitative analyses in this area are needed to identi@
optimum wet curing periods and latex contents to maximize durability.
A C 1 SP-148 74 0662949 0520088 723 m
MATERIALS
Two latex-modifiersproduced by a single manufacturer were used in this

study. The conventional latex-modiñer consisted of a 48.1 percent solution of
styrene-butadienepolymer latex solids suspended in water. This latex-modifier is
readily available at construction supply outlets. The developmental latex was
obtained directly fiom the manufacturer and consisted of a 44.6 percent solution
of methylmethacrylatepolymer latex solids suspended m water.
Type I-II cement, coarse aggregates, and h e aggregates were used in all
concrete mixtures. Coarse aggregate was well rounded and fi-om an alluvial
gravel deposit with a nominal maximum size of 9.5 mm. The ñne aggregate was
also well rounded and fi-om a different alluvial deposit. Chemical analysis of the
cement, provided by the manufacturer, is summarized m Table 1. A control mix,
containing no latex, was produced using a naphthalene-based high-range water
reducer and an air-entraining admixture based on organic acid salts.
EXPERIMENTAL METHODS
Concrete Mixtures
Seven different concrete mixtures with varying latex-modiñer contents

and latex-modiñer s p e s were tested for compressive strength according to
ASTM C 39 and fieezing and thawing resistance according to modified ASTM C
666 practices. A water-cement ratio of 0.35 was used for ail mixtures.
Two mixtures were prepared with conventional latex. These mixtures

were proportioned with latex solids contents of 10 percent and 15 percent by
mass of cement and were proportioned to have the same paste content by
volume, where paste components were water, cement, and latex-modifier
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
solution. Saturated surface-dry mixture quantities and paste contents, for all
mixtures, are presented in Table 2. Workability was good for all mixtures
containing the conventionallatex-modifier.
Four mixtures were prepared with the developmental latex-modifier.

Three of these mixtures were proportioned for latex solids contents of 5 , 7.5, and
10 percent, all with equivalent paste contents. This paste content, different íkom
that used for the conventional latex-modified mixtures, was selected by assigning
the cement content of the 5 percent developmental latex-modiñed mixture the
cement content used for the 10 percent conventional latex-modified mixture.
This adjustment was necessary to ensure adequate mixing of the materials.
Because segregation occurred in the 7.5 and 10 percent mixtures, a fourth
A C 1 SP-148 9 4 Ob62949 0520089 bbT
mixture was prepared with 10 percent latex-modifier content but a lower overall
paste content. This mixture had the same quantity of latex-modiñer as the 7.5
percent mixture, but less cemqt. Hereafter, the 10 percent developmental latex-
modified mixture with the low paste content will be referred to as the "10percent
lean mixture" while the 10 percent developmentallatex-modiñed mixture with the
normalized paste content will be refmed to as the "10percent rich mixture".
One control mixture was also prepared. This mixture contained no latex
but did contain a HRWRA and an &-entraining admixture. Paste content for the
control mixture was equivalent to the paste content used for the conventional
latex-modified mixtures.
Mixing and Curing
Mixtures were prepared with t h e aggregate in a cooled, oven-dry

condition and coarse aggregate wet ûom a 24-hour soaking period. Mixing time
was ñve minutes, with a two minute rest before discharging ûom the mixer and
measuring slump.
Each batch of concrete was mixed with enough materials to produce

0.0369 m30f concrete. This was enough to produce six prisms measuring
75x102~406mm and six 102 by 203 mm cylinders. Cylinders were compacted in
three lifts by rodding. Prisms were compacted in two lifts on a vibrating table.
For all of the concrete mixtures containing latex, three prisms and three cylinders
were wet cured under burlap for one day. The remaining three prisms and three
cylinders were wet cured under burlap for three days. For the concrete batch
containing no latex, three prisms and three cylinders were wet cured under burlap
for three days and the remaining specimens were wet cured under burlap
overnight and then in a fog room, the prisms for 13 days and the cylinders for 14
days. All specimenswith a wet cure period less than 13 days were ailowed to dry
cure in the laboratory (humidity 50-60 percent) untii they reached an age of 13
days. After 13 days, prisms were placed m a 5OC bath m order to prepare them
for placement in the ûeezing and thawing apparatus. At 14 days, the prisms were
placed in the ûeezing and thawing environment and ail six cylinders were tested
for compressive strength according to ASTM C 39.
Testing Program
Prism specimens were tested according to modified ASTM C 666

practices. In accordance with this test method, they were subjected to 300 cycles
of ûeeziug and thawing or less if their relative dynamic modulus dropped below
60 percent. One set of specimens made with 15 percent conventional latex was
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-/4ô 94 W O662949 0 5 2 0 0 9 0 381
tested in fieshwater while all other fieezing and thawing specimens were tested in
a 2.5 percent solution of NaC1 and deionized water. Specimens tested in
fieshwater were each wrapped with cloth to keep the d a c e moist during
freezing and thawing, in accordance with the proposed Procedure C for
AASHTO T 161[5]. Specimens tested in the salt solution were each held in a
plastic container, in accordance with Procedure A, ASTM C 666. These plastic
containers served the dual purpose of maintaining a constant salt solution
environment and retaining all material scaled fiom the specimens[ó]. Scaled
material retained in the plastic containers was, captured in a metal tray each time
the specimen was removed to be tested for fundamental fiequency. The material
scaled fiom each beam was then weighed in a saturated-drained state after the
specimen had been subjected to 50, 100, 200, and 300 cycles of fieezing and
thawing. A saturated-drained state was achieved by wet sieving the material
through 425pm and 75pm sieves.
Once ûeezing and thawing cycles were completed for a batch of

specimens, a sample was taken fiom them for microscopical determination of
parameters of the air-void system according, to ASTM C 457. Hardened air
content, spacing factor, and specific d a c e were determined for all mixtures.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
RESULTS
Compressive strengths and initial dynamic modulus values are provided in

Table 3. The compressive strengths as a function of latex-modifier content and
curing are shown in Fig. 1. These results show that both compressive strength
and initial dynamic modulus increased as the length of wet curing increased. The
exceptions to this trend are that the control mixture had higher compressive
strengths for specimens with 3 days of wet curing than for specimens with 14
days of wet curing and the 10 percent developmental rich mixture had higher
dynamic modulus values for specimens wet cured for one day than for specimens
wet cured for 3 days. It appears that in the control mixture specimens wet cured
for three days surface tension, induced by drying, increased the compressive
strength beyond the compressive strength oïthe specimens wet cured 14 days.
However, this surface tension appears to have been alleviated in the prism
specimens, at least in part, during the 24-hour soaking period prior to the
measurement of the initial dynamic modulus. Compressive strength also appears
to vary with latex-modifier content. For both latex-modifier types, the
compressive strength increased with decreasing latex content, although only
modestly for the conventional latex-modified mixtures. Initial dynamic modulus
values increased with decreasing latex content for the developmental latex-
modiñer but decreased with decreasing latex content for the conventional latex-
modiñer.
A C 1 SP-L4ô 94 O b 6 2 9 4 9 0520091 2 L B
Fig. 2 shows the relative dynamic modulus versus cycles of fieezing and
thawing for the 7.5 percent developmental mixture. This mixture showed the
greatest variation related to Curing time for all of the mixtures containing latex.
Considering the variabiiity of the individual samples, there is no clear effect of
curing time on the relative dynamic modulus. Table 4 summarizes the average
durability factor values for all of the mixtures, calculated in accordance with
ASTM C 666. The greatest effect of Curing was on the control mixture. Drymg
prior to fieezing and thawing appeared to improve the resistance to fieezing and
thawing of this mixture.
Mass of the material scaled fiom the specimens, retained as described

above during fieeziug and thawing cycles in saltwater, was measured in a
saturated-drained condition aRer 50, 100, 200, and 300 cycles. Results for the
control mixture are shown in Table 5 . These data show that for the mixture with
no latex, the ratio between mass loss, determined fiom mass measurements of the
pdsm when removed for fundamental frequency testing, and mass of the scaled
material remained fairly constant. In general this ratio varied drastically,
decreasing as the number of fieeziug and thawing cycles and amount of scaled
material increased. Mass of the scaled matenal for specimens cured one and
three days is shown graphically in Fig. 3 and Fig. 4, respectively. These plots
show that specimens fiom the 10 percent conventional, control, 5 percent
developmental, and 10 percent developmental lean mixtures with shorter wet
cure periods scaled less than did specimens with longer wet curing periods.
Specimens fkom the other mixtures, which had much greater mass losses, resisted
scaling better when the wet curing period was longer.
Figures 3 and 4 suggest that at 100 cycles of fieezing and thawing in 2.5
percent NaCl solution, a clear distinction can be made between those mixtures
showing a lot of scaling and those showing relatively little. For comparison
purposes, scaling to a depth of 1 mm over the entire surface is approximately
equivalent to 2.3 kg/m2. Durabfity factor values are plotted versus scaled mass
in Fig. 5. There does not appear to be a meaningfid relation between scaling and
durabiiity factor. Some specimens with what would be considered acceptable
durability factor values showed considerable scaling.
Only the control mixture containing no latex used an &-entraining

admixture, but microscopic examination indicated that the latex-modified
mixtures contained some voids small enough to be considered entrained. The
results of linear traverse analyses (ASTM C 457) are provided in Table 6. For
the conventionallatex mixtures, reducing the latex content improved the air-void
system (lower spacing factor and higher speciñc mfkce). This was not true for
the developmental latex mixtures. These mixtures appear to be much more
sensitive to other parameters, water and cement contents, in terms of the amount
of air that is entrained.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662949 0520092 1 5 4
The infiuence of spacing fhctor on durability factor for the mixtures is

shown in Fig. 6. In this graph, ali of the mixtures are shown. Spacing factor
appears to completely explain the variation in durability factor for the latex-
modified mixtures. As mentioned previously, the length of the moist cure does
not appear to significantlyinfiuence durability factor. The inûuence of variations
in latex type or amount appear to have no effect on durability factor other than
the effect that these variations have on the air void system. To maintain a
durability factor value of at least 80 the maximum spacing factor for latex-
modded concrete appears to be around 0.40 mm. This is higher than the
normally accepted value of 0.20 mm for conventional concrete[7]. The control
mixture shown in Fig. 6 also appears to have acceptable resistance to freezing
and thawing (durability factor value greater than 80) at a spacing factor greater
than 0.20 mm. The literature suggests that acceptable freezing and thawing
resistance has been obtained usiig mixtures with HRWRA that have higher
spacing factor@]. It is not clear if this simply means that such concrete takes
longer to become critically saturated and consequently, vulnerable to damage
when subjected to freezing.
Scaling at 100 cycles of fieezing and thawing as a function of spacing

factor is presented in Fig. 7. A spacing factor of approximately 0.33 m appears
to separate the scaling and non-scaling latex-modified mixtures. This is a stricter
requirement than for durability factor, above. Also, latex-mower appears to be
more effective than high-range water reducer in reducing scaling.
DISCUSSION
Compression tests demonstrated that reducing the latex content of latex-

modified concretes and maximizing the wet curing period increased the strength
of the mixtures. The length of the wet curing period did not appear to have a
significant effect on durability factor, and only appeared to have a major effect on
scaling for those mixtures showing considerable scaling. For those mixtures,
increased wet curing reduced scaling.
Comparison of two mixtures with the same proportions tested for

fieezing and thawing resistance, one in freshwater and one in saltwater, suggest
that deterioration of the concrete mass will occur in the presence of either
freshwater or saltwater. Apparently, saltwater wiil act as a catalyst for both
deterioration and scaling, while freshwater primarily facilitates deterioration.
Linear-traverse results show that even latex-modiñed concretes with high

latex contents are subject to deterioration ifinsdñcient entrained air is provided
or if the spacing factor is excessive. This demonstrates that the air-void system
has more influence on concrete durabiiity than latex-modifier content or the
duration of wet curing. These obsemations are consistent with those of other
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 W Obb2545 0520093 O90 W
researchers[3] who showed that spacing factor had a much greater influence on
scaling resistance in saltwater than did latex-modifier content.
CONCLUSIONS
Compressive strengths of latex-modified concretes tend to increase as the

amount of latex in the mixture is reduced. These strengths also tend to increase
as the amount of wet curing time is increased.
The quaiity of the air-void system of the concrete is much more essential
to durable concrete than latex-modiñer content or duration of wet cure.
Development of a proper air-void system should be the primary concern when
proportioning a concrete mixture for resistance to fieezing and thawing.
To develop both strength and durability in latex-modified concrete, effort

should be made to allow the latex-modified concrete to moist cure for a
reasonable amount of time m order to develop its compressive strength.
The standard practice of using 15 percent latex should be re-examined to

determine the optimum latex content for a speciñc application. This study has
shown that latex content can be reduced without harming the resistance to
Sreeziug and thawing in saltwater as long as a reasonable air-void system is
maintained. The air-void system requirements appear to not be as strict for latex-
modified concrete as for conventional concrete.
Finally, it is important to note that the range of curing and dryhg periods
that have been studied for latex-modified concretes is very limited. A much
greater database of information on this topic is needed before the use of iatex-
modified concretes can be optimized.
ACKNOWLEDGMENTS
The authors would like to thank Dow Chemicai for supplying the
developmental latex-modifier and W.Elkey for performing the linear traverse
analyses.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 Ob62949 0520094 T 2 7
REFERENCES
1. Ohama, Y., "Principle of Latex Modification and Some Typical Properties of

Latex-Modified Mortars and Concretes," AC1 Materials Journal, V. 84. No. 6,
Nov.-Dec. 1987, pp. 511-518.
2. Bishara, A. G., "Latex Modified Concrete Bridge Deck Overlays- Field

Performance Analysis," Federal Highway Administration, Report No.
FHWA/OW79/004, Oct.1979.
3. Bordeleau, D., Pigeon, M., and Banthia, N.,"Comparative Study of Latex-

Modified Concretes and N o m l Concretes Subjected to Freezing and Thawing in
the Presence of a Deicer Salt Solution," AC1 Materials Journal, V. 89, No. 6,
Nov.-Dec. 1992, pp. 547-553.
4. Hashimoto, H., and Ohama, Y., "Effects of Curing Methods on Strengths of

Polymer-Modified Concretes," Journal, College of Engineering, Nihon
University, SeriesA., V. 19, Mar. 1978,pp. 113-119.
5. Janssen, D. J. and Snyder, M. B., "Resistance of Concrete to Freezing and

Thawing," Final Report, SHRP Contract (2-203, Strategic Highway Research
Program, Washington, D.C., 1994.
6. Janssen, D. J. and Snyder, M. B., "Mass IASSExperience With ASTM C666:

With and Without Deicing Salt," International Workshop on the Resistance of
Concrete to Scaling Due to Freezing in the Presence of Deicmg Salts, Centre de
recherche interuniversitaire sur le beton (CRIB), Universite de Sherbrooke-
Universite Laval, August 1993, pp. 137-152.
7. AC1 Committee 201, "Guide to Durable Concrete", (AC1 201.2R-92), AC1

Manual of Concrete Practice, Volume 1, 1992.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
8. Attiogbe, E. K., Nmai, C. IC, and Gay, I . T., "Air-Void System Parameters
and Freeze-Thaw Durability of Concrete Containing Superplasticizers", Concrete
International, July 1992, pp. 57-61.
A C 1 SP-148 94 0662949 0520095 9 b 3
TABLE 1 - CHEMICAL COMPOSITION OF TYPE 1-11 CEMENT
Na20
Chemical Si02 A703 Feio3 Ca0 MgO SO, (Equiv.)
%(mass) 22.5 4.2 3.0 65.0 1.1 2.4 0.56
TABLE 2 - COMPOSITION OF CONCRETE MIXTURES (kg/m3)

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Au quantities are for saturated surface dry conditions.

* Paste content is expressed as percent by volume.
** Control mixture contained 673 mL of HRWRA and 110 mL of air-entraining
admixture per 100 kg of cement.
A C 1 SP-148 9Y 0bb2949 0520096 B T T
TABLE 3 - AVERAGE COMPRESSIVE STRENGTHS AND INITIAL

DYNAMIC MODULUS VALUES FOR ALL MIXTURES
10% Developmental 1 38.7 37.3

(Rich) 3 41.3 35.8
10% Developmental 1 35.0 I 34.3
(Lean) 3 40.5 36.4
TABLE 4 - AVERAGE DURABILITY FACTORS FOR ALL MIXTURES
Mixture Wet Cure Period Durability Factor

15% Conventional (Freshwater) 1 51
3 52
15% Conventional(Saltwater) 1 34
3 34
10% Conventional 1 109
3 103
Control (No Latex) 3 99
14 82
5% Developmental 1 104
3 101
7.5% Developmental 1 74
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
3 79
1 90
I 10% Developmental (Lean)

3
1
3
90
107
102
A C 1 SP-148 94 = 0662947 0520097 736
TABLE 5 - MASS LOSS AND SATURATED-DRAINED MASS OF SCALED

MATERIAL FROM THE CONTROL MIXTURE
Cure Freezing and Scaled

Period Thawing MassLoss Material Scaled-
Mixture (days) Cycles (grams) (grams) LostRatio
Control 3 50 110 227 2.06
100 183 358 1.96
200 287 542 1.89
300 358 666 1.86
14 50 198 332 1.68
1O0 402 655 1.63
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
200 5 82 977 1.68
300 730 I 1209 I 1.66
TABLE 6 - LINEAR TRAVERSE RESULTS
Hardened Air (%) Spacing Factor Specific Surface

Mixture (mm) (mm2/mm3)
15 % Conventional
(1) 2.0 0.49 17
15% Conventional
(2) 2.0 0.50 16
10% Conventional 6.3 0.26 18
Control
(No Latex) 2.1 0.32 25
5% Developmental 3.8 0.32 18
7.5% Developmental
2.6 0.41 17
10% Developmental
(Rich) 4.1 0.34 16
10% Developmental
(Lean) 5.7 0.27 17
A C 1 SP-148 94 = Obb2949 0520098 6 7 2
60
50
40
30
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
20
10
O
control 15% Conv. 10% Canv. 5% Dev. 7.5% Dev. 10%Dev.(R)li 6 Dev.(L)
Mixture
Fig. l-lompressive strength varying with duration of wet cure and latex-modifier content
. . . . . . ...............................................
..................
1-daywet cure
3-day wet cure
O
O 50 100 150 200 250 300
fieezing and Thawing Cycles
Fig. 2 D e l a t i v e dynamic modulus versus cycles of freezing and thawing for the 75 percent
developmental latex-modified concrete mixture
A C 1 SP-14ô 94 Obb29V9 0520099 509
50 100 150 ao0 a60 500

M i n g and Thawing Cycles
Fig. ?Mass of scaled material, normalized with respect to surface area, versus cycles of freezing
and thawing for control specimens and all specimens moist cured for one day and tested in saltwater
- 15nConv.
..............................................................
Control (14)
5% Dev.
7.5% Dev.
............. I
.............. .............. j . . . . . . . . . . . . . .j .....
10%Dev. (L)
50 100 150 200 !250 500

Freeaing and Thawing Cyciea
Fig. &Mass of scaled material, normalized with respect to surface area, versus cycles of freezing
and thawing for control specimens and specimens moist cured for three days and tested in saltwater
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0520LOO 050
100 %+ o
t
1-Day Cure
20 3-DayCure
0 Control (3)
x Control (14)
O
O 5 10 15 20 25
Scaled Material (kg/sq. m.)
Fig. S D u r a b i l i t y factor versus scaling for all mixtures and wet cure periods
100 ** a
O
'
80 t
b
ci
60
P
3
%
n$ 4 0 1-Day Cure
20 3-Day Cure
Control (3)
Control (14)
0.25 0.3 0.35 0.4 0.45 0.5 0.55
Spacing Factor (mm)
Fig. ó-Durability factor versus spacing factor for all mixtures and wet cure periods
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 94 Obb2949 0520LOL T 9 7
332 Newîson and Janssen
:.
i-Das C ~ r e
* 3-DayCure ............j ............. i ......... ...:.~ ...........
0 Control (3)
x Control (14)
............................................................
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O
0.2 0.25 0.3 0.35 0.4 0.4 0.5 0.55
Spacing M o r (mm)
Fig. 7-Scaling at 100 cycles of freezing and thawing versus spacing factor for all mixtures
and wet cure periods
A C 1 SP-148 94 Obb2949 0520102 9 2 3
SP 148-19
Effect of Chemical Characteristics

of AlceIl@ Lignin-Based Methyl-
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
sulphonates on Their Performance
as Water-Reducing Admixtures
by J. Zhor, T.W. Bremer, and J.H. Lora
Synopsis: High purity ALCELL" lignin based methylsulphonates were prepared by

sulphomethylolation of different lignin fractions obtained in an organosolv pulp mill.
These methylsulphonateswere characterizedand their performance in cement slurries and
mortars was studied. The chemical analysis focused on determining the degree of
sulphonation and molecular weight distribution. Then the twelve samples were tested as
possible water-reducing admixtures for concrete.
The fluidifying influence of ALCELIQ lignin based methylsulphonates on
cement slurries was evaluated by measuring the torque resistance in a specially
developed mixing chamber. Each test involved adding the methylsulphonate in 4 steps
of 0.2% by weight of cement to a portland cement slurry of waterhment ratio = 0.45.
The decrease of torque resistance indicated the relative effectiveness of the various lignin
samples. Also this test provided a simple indication of the change in time of set caused
by a particular admixture.
The sulphomethylolatedALCELL@lignin samples were tested in cement mortars
as well. The mortars were prepared with graded silica sand and normal portland cement.
The plasticizing effect was determined using the flow table. The increase of flow caused
by the addition of different samples in the range of 1.0% and 2.0% by weight of cement
was tested. The air-entraining effect of these samples was determined from the unit
weight measurement of fresh cement mortar.
The results of this testing indicated the importance of the chemical characteristics
of sulphomethylolated ALCELL" lignin based water-reducing admixtures on their
effectiveness in cement mortar and concrete.
Kewords: Admixtures; air entrainment; cements; chemical analysis; chromatography; emission

spectroscopy; lianin and derivotives; mortars (material); plasticizers; rheoloaical properties; setting
(hardening); sulphonates; sulfur; tests; water reducing agents
A C 1 SP-348 94 = Ob62949 0520303 BbT m
334 Zhor, Bremner, and Lora
AC1 member Jiri Zhor is a doctoral candidate in the Department of Civil Engineering at
the University of New Brunswick in Fredericton, N.B., Canada. After 10 years of
experience in areas of concrete sîructures, materials, repairs and testing, he has been
engaged in research on concrete admixtures since 1986. His current research interests are
on the chemistry of superplasticizers for concrete and properties of fresh concrete.
AC1 fellow Theodore W.Bremner is a professor of civil engineering at the University

of New Brunswick in Fredericton. N.B., Canada. He received his PN)degree from the
University of London in 1981. He is an active member of ACI's Concrete Research
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Council and several technical committees. His broad research interests include süuctural
lightweight concrete, durability of concrete, and concrete in the marine environment.
Jair0 H. Lora is the Director of ALCELL@Technology, Repap Technologies Inc., Valley

Forge, PA. He received his PhD from University of Toronto in 1981. After working as
a process engineer for a subsidiary of Container Corporation (now Jefferson Smurfit) in
Cali, Colombia he joined Repap in 1982. He has been involved in the development of
the ALCELL' process including applications for lignin and other Co-products.
INTRODUCTION
Lignosulphonate based concrete admixhires are usual1y prepared from the waste
liquor formed by the production of sulphite pulp. By neutralization, precipitation and
fermentation of this liquor a range of lignosulphonates of varying purity, composition
and molecular weight distribution is produced (I). Unlike the traditional sulphite pulping,
the new ALCELL~organosoiv pulping technology allows for the recovery of a pure,
unsulphonated form of lignin (2). "bis ALCELL@lignin is a suitable raw material for
the preparation of water-reducing admixtures for concrete (3).
Water-reducing admixtures are generally surface-activechemicals or surfactants.

These are amphipathic long-chain organic molecules bearing a hydrophobic tail and a
hydrophilic head. Lignosulphonates are classified as anionic surfactants since the
hydrophilic groups associated with the organic polymers are sulphonates (-SO,?. When
adsorbed onto cement particles, these surfactants impart a strong negative charge which
lowers the surface tension of the surrounding water, This greatly enhances the fluidity
of the system (4). The chemical structure and composition of water-reducing admixtures
influences their surfactant properties which determine their effectivenessin cement-water
mixtures (5-18).
The influence of molecular weight of different water soluble polymers on their

performance in cement-water systems has been tested (9-12) using methods such as gel
-
permeation chromatography GPC (9,10,13) or dynamic viscosity measurements (11,12)
for molecular weight determination. Water-reducing admixtures and superplasticizers
have also been analyzed for their chemical composition including the content of sulphur,
sulphate, chloride. sodium and calcium (14.15).
A C 1 SP-148 94 0662949 0520304 7Tb M
By studying the chemistry of lignosulphonates it was found that increasing the

molecular weight of lignosulphonatesimproves their fluidifying effect on cement mortars
(17). However, not all investigators have the same opinion as to what determines the
retardation and air-entraining effect (16,17,18).
The objective of the work described in this paper was a preliminary

investigation of the influence of molecular weight and sulphonation level of ALCELL@
lignin based methylsulphonates on their behaviour as water-reducing admixtures for
concrete. The methylsulphonates were prepared by sulphomethylolation of different
ALCELL@lignin fractions obtained in an organosolv pulp mill.
A group of 12 different samples of ALCELL@ lignin methylsulphonates

composed of 4 different molecular weight groups with 3 sulphonation levels (see Table
5) and 1 sample of commercial naphthalene sulphonate were incorporated into the
program. After determining the molecular weight distribution by gel permeation
chromatography and total sulphur content by atomic emission spectroscopy the following
testing of the effect of methylsulphonateson cement-water systems was conducted:
1. Cement Slurries : - fluidifying effect

- retardation of setting
2. Cement Mortars : - plasticizing effect
- air-entraining effect
TESTING PROGRAM
Materials
The following materials were used in the laboratory tests:

Cement: Normal portland cement - CSA-A5 - Type 10 (ASTM C150, Type I)
Aaareaate: Silica sand graded as standard sand (ASTM C778)
Water: Potable, municipal water supply
Admixtures: Water-reducing:
a) 12 different samples of sulphomethylolated ALCELL@lignin
b) 1 sample of commercial naphthalene sulphonate superplasticizer
Some of the physical and chemical properties of these products are reported in
Tables 1, 2, 3 and 4.
Mixture ~ r o ~ o r t i o n s
Cement Slurry: Portland cement, Type 10 :5OOog

Water : 2250 g
WJC : 0.45
Admixture - lignin methylsulphonate : 40 mi (4 times 10 mi)
- naphthalene sulphonate : 30 mi (3 times 10 mi)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0520305 6 3 2
Cement Mortar: Portland cement, Type 10 : 500g

Graded silica sand : 1375 g
Water : variable
WIC : 0.610.S510.46
Admixture :5/10 ml (17% 2%vol.by
weight of cement)
Equipment and Procedures
Gel permeation chromatomaphv -- The molecular size distribution of the admixtures was
defined by gel permeation chromatography (GPC). Low pressure GPC was performed
using a 1.5 x 30 cm glass column packed with Sephadex G- 100 (Pharmacia), a Buchler
rotary sample collector and Perkin-Elmer Lambda 3 UVNIS spectrophotometer
(A=280nm). A 0.1 mM NaOH eluant at an operating pressure of 60 cm H,O and a flow
rate of 0.13 ml.min-' was used for the analysis. The column had Vp8.7 ml (determined
using blue dextran as the test substance) and V,=27.6 mi. Protein standards were used
for column calibration.
Atomic emission suectrometrv -- The total sulphur content in the tested admixtures was
determined by atomic emission spectrometry (AES). Inductivelyaupled plasma emission
by aii argon plasma (ICAP) was used for AES analysis. The tests were run on a Thermo
Jarre11 Ash ICAP 9OOO machine at a wave length of 1820.4 nm and an operating power
of 1 kW.The applied samples were diluted 1:1O,OOO in water.
Torque test -- The fluidifying infiuence of the admixtures and their effect on the initial
set retardation was tested on cement slurries using the torque test. This test measured the
changes of the torque resistance by rotating a spindle in the cement slurry placed in a
specially developed mixing chamber. The spindle was operated by a GH Heller --```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
laboratory stirrer and a 3497A Hewlett Packard acquisition unit. A rotational speed of
150 rpm was used. The data were evaluated using a SM-200Samtron computer. 'Ehe
cement-water mixtures were premixed by hand for 10 minutes and then mixed in the
mixing chamber for another 10 minutes. After the indicated torque resistance reached the
value of 8 N.m, the first 10 mi (i.e. 0.2%by weight of cement) of an admixme was
applied. Additional doses of equal volume were added at 5 minute intervals until the
saturation point was reached.
Cement mortar testinE -- The plasticizing effect of the admixtures on cement mortars was
tested by flow determination. The air-entraining effect was tested on cement mortars as
well. The fresh mortars were prepared by mechanical mixing in a 3 litre Hobart mixer
using the procedure given in Section 6 of ASTM C 305. nie admixtures were added to
the @Xing water in the bowl before adding the cement. The Flow of the mortar was
determined using the flow table, flow mould and other tools conforming to the
requirements of ASTM C 230. The procedure used in previous works (19,20,21) was
modified in accordance with ASTM C 109. The air content of the cement mortar was
determined by weighing a 400 ml measure filled with fresh mortar. The measurement
and consequent calculation were in accordance with ASTM C185.
A C 1 SP-148 94 œ 0bb2949 052030b 579 W
Chemical Characteristics
Molecular weight disîribution -- The results of gel permeation chromatography are

presented in Fig.1 to 4. Molecular weights were interpolated from a K, vs log molecular
weight plot. As only protein standards were available for column calibration, the plot
used for interpolating molecular weights was obtained by shifting the protein standard
plot to a dextran equivalent in order to provide a better approximation for
lignosulphonates. Thus the plotted molecular size distribution does not necessarily
express absolute values of molecular weight. However, since all samples were run under
the same conditions, a comparison of their molecular size distributions can be used to
describe their differences. It should also be noted that the effective separation, obtained
over a molecular weight range of 1 to 100 kD, describes only 75 to 95% of these
samples, with 5 to 25% exceeding the effective limits. The samples were formulated in
four groups of increasing molecular weight, each group being sulphonated at three
different levels. By comparing Fig.l,2 and 3 it is noticeable that the degree of
sulphonation influences the molecular weight of the samples. The higher the
sulphonation, the greater the low molecular weight portion of the molecules. This could
mean that mainly low molecular weight sulphonates are present in the samples.
Nevertheless, the differences between the four molecular weights in each sulphonation
level.are quite evident. It is also clear that the commercial naphthalene sulphonate based
superplasticizer (Fig. 4) shows a remarkable difference in molecular size disîribution,
probably due to its different chemical origin.
Total sulphur content -- The determinationof total sulphur content by the ICAP analysis
is presented in Table 6. The results given in g.1.' were recalculated into % by weight of
solids according to the solids content and density of each sample. It was generally found
that the sulphur content of low sulphonation levei ALCELL" Lignin samples was 1.5 to
1.8% by weight of sample solids, whereas the sulphur content of medium and high
sulphonation level samples was 1.8 to 2.5% and 3.0 to 4.0%, respectively. The sulphur
content in the commercial naphthalene sulphonate based superplasticizer (CNS) was
10.81%. These results are in agreement with those of other investigators (14). It is
important to realize that total sulphur content gives only a preliminary idea of the degree
of sulphonation since it does not describe the amount and position of the sulphonate
group incorporated into the sample molecules.
Testing on Cement Slurries
The dependence of the torque resistance on time tested in cement slurries is

shown in Fig.5 to 8. The data obtained from these measurements were used for the
determination of fluidifying and set retarding effects of tested admixtures.
Fìuidifiing effect -- The fluidifying effect was expressed as the decrease in torque
resistance due to the addition of 40 rnl of the lignin admixture or 30 mi of commercial
superplasticizer into the cement slurry. The results are shown in Table 7 and 10. The
minimum torque resistance was chosen as the value of the torque resistance at time t=0.7
hour for the ALCELL" lignin samples and at time t=OS hour for the commercial
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-348 94 0662949 0520307 405
superplasticizer. From the results in Table 10 it is obvious that the fluidifying effect of
the admixtures increases with an increase in the sulphonation level. A not so obvious
correlation was found between molecular weight and fluidifying effect. Nevertheless, the
fluidifying effect seems to increase with a decrease in molecular weight. Other
investigators found either no correlation (9,17) or noted an increase in fluidifying effect
with increasing molecular weight. As can be seen in Table 10 the commercial
naphthalene sulphonate superplasticizer had significantly greater fluidifying effect than
all of the lignin samples.
Senha time retardation -- In this work, the time of initial set of the cement slurries was
arbitrary chosen as that when the torque resistance of the sluny reached the starting
value of 8 N.m. The time of initial set of normal portland cement Type 10 tested in
accordance with ASTM C 150 was used in the calculation of the setting time retardation
of the sluny. The results are shown in Table 7 and 8. It is apparent that the retardation
decreases when the molecular weight increases while the influence of sulphonation is not
obvious.
Testing on Cement Mortars
Plasticizing effect -- The plasticizing effect was evaluated in terms of the percentage
increase in the average base diameter of the mortar obtained in the flow test. 'Ihe results
are presented in Table 9 and 11. The plasticizing effect increases with an increase in
suiphonation level and with a decrease in molecular weight. The influence of the
admixtures on cement mortars seems to be in agreement with tests done on cement
slurries.
Air-entrainina effect -- The air content was calculated from the fresh mortar unit weight
measurement and the results are summarized in Table 9 and 12. The air-entraining effect
of the samples decreases significantly with an increase in molecular weight. It is
anticipated that the decrease of the air content also contributes to the decrease of the
plasticizing effect of tested admixtures. In some cases the &-entraining effect also
increases with an increase in sulphonation level, but this correlation seems insignificant.
CONCLUSIONS
The influence of chemical characteristicsof sulphomethylolatedALCELL' lignin

samples on their performance as water-reducing admixtures was proved. An increase in
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
the sulphonation level of the samples caused a noticeable increase in plasticizing effect
as well as an increase in time of set, but did not increase the air content significantly.
The increase in molecular weight of the lignin samples caused a significant decrease in
the time of set retardation and also a decrease in the air content and the plasticizing
effect. The close correlation between the air content and the plasticizing effect was
anticipated. The optimization of the degree of sulphonation and molecular weight of
ALCEU@ lignin methylsulphonates requires further study in order to develop an
admixture with effectivenesscomparableto the commercialnaphthalene sulphonatebased
superplasticizers.
A C 1 SP-148 9 4 = 0662949 0520308 3 4 3 =
ACKNOWLEDGEMENTS
The authors would like to thank Dr.P.Cashion and K.Gordon of University of

New Brunswick for their interest, cooperation and valuable advice during the GPC
analysis. Special thanks are due to Dr.I.Delaney and S.Venugopa1 of Research and
Productivity Council for carrying out the ICAP analysis. The authors also wish to thank
Dr.E.K.Pye, President of Repap TechnologiesInc., for granting permission to publish the
paper.
REFERENCES
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Rixom, M.R. and Mailvaganam, N.P. "Chemical Admixtures for Concrete"; 2nd
ed., E.8zF.N. Spon, New York, 1986.
Pye, E.K. and Lora, J.H. "The ALCELL@Process - A Proven Alternative to

Kraft Pulping"; Tappi Journal, Vo1.74, 1991, No.3, pp.113-118.
Bremner, T.W. and Zhor, J. "High Purity &CELL@ Lignin as a Concrete

Admixture";Final report for NSERC, Universityof New Brunswick, Fredericton,
October 1993.
Mehta, P.K. and Monteiro, P.J.M. "Concrete. Structure, Properties, and

Materials"; 2nd ed. Prentice Hall, Englewood Cliffs, NJ, 1993.
Myers, D. "Surfactant Science and Technology"; VCH Publishers, New York,

1988.
Jolicoeur, C."The Chemistry of Superplasticizers.Concept, Characterization and

Functional properties"; In Proceedings, Superplasticizers, Sherbrooke, 1990,
pp.13-25.
Ramachandran, V.S. and Mailvaganam, N.P. "New Developments in Chemical

Admixtures"; In Proceedings, Advances in Concrete Technology, Athens, 1992.
Edited by V.M.Malhotra. CANMET/ACI, pp.859-898.
Collepardi, M. and Ramachandran, V.S. "Effect of Admixtures"; In Proceedings,

9th International Congress on the Chemistry of Cement, New Delhi, 1992. Vol.1,
pp.529-568.
9. Basile, F., Biagini, S., Ferrari, G. and Collepardi, M. "Influence of Different

Sulphonated Naphthalene Polymers on the Fluidity of Cement Paste"; In
Proceedings, Superplasticizers and Other Chemical Admixtures in Concrete,
Ottawa, 1989. Edited by V.M.Maihotra. AC1 SP-119, pp.209-220.
10. Sebök, T."The Effects of the Molar Parameters of Sulphonated Melamine and
Formaldehyde Polycondensates on Mortar Properties"; In Proceedings, 8th
A C 1 SP-148 9 4 M 0662949 0520309 288 M
International Congress on the Chemistry of Cement, Rio de Janiero, 1986.

v01.1Ir. pp.34-39.
11. Collepardi, M., Corradi, M. and Valente, M. "Influence of Polymerization of

Sulphonated Naphthalene Condensate and its Interaction with Cement"; In
Developments in the Use of Superplasticizers.Edited by V.M.Maihob-a. AC1 SP-
68, Ottawa, 1981. pp.485-498.
12. Lahalih, S.M., Absi-Halabi, M. and Ali, A.M. "Effect of Polymerization

Conditions of Sulphonated-Melamine Formaldehyde Superplastizers on
Concrete"; Cement and Concrete Research, Vo1.18, 1988, N0.4, pp.513-531.
13. Roy, D.M., Varadi, G., Tamas, F.D., Paiyi. G. and Barîha, B. "Application of
GPC for the Analysis of the Oligomer Disîribution of Naphthalene-Based
Superplasticizers"; Cement and Concrete Research, Vol. 14, 1984, N0.3, pp.439-
442.
14. Ista, E. and Verhasselt, A. "Chemical Characterization of Plasticizers and

Superplasticizers"; In Proceedings, Superplasticizers and Other Chemical
Admixtures in Concrete, Ottawa, 1989. Edited by V.M.Malhotra. AC1 SP-119,
pp.99-116.
15. Jolicoeur, C., Pmeault, F., Simard, M.-A. and Nuyt, A. "The Chemistry of
Concrete Superplasticizers.Rheology and Hydration Kinetics of Portland Cement
Pastes Containing Mixtures of Naphthalene- and Melamine-Based
Superplasticizers";In Proceedings,International Seminar on Concrete Durability,
Monterrey, N.L., 1993, pp.185-199.
16. Rixom, M.R. and Waddicor, J. "Role of Lignosulphonates as Superplasticizers";

In Developments in the Use of Superplasticizers.Edited by V.M.Maïhoh-a. AC1
SP-68, Ottawa, 1981, pp.485498.
17. Bialski, A.M. and Nayak, KV. "Superplasticizers from Lignosulphonates"; In

Proceedings, Fourth International Symposium on Wood and Pulping Chemistry,
P ~ s 1987.
, V01.1, pp.83-90.
18. Miyake, N., Ando, T. and Sakai, E. "Superplasticized Concrete Using Refined
Lignosulphonate and Its Action Mechanism"; Cement and Concrete Research,
Vo1.15, N0.2, 1985, pp.295-302.
19. Zhor, J. "Testing of Concrete Admixtures in Cement Mortars"; Stavivo (Prague),

V01.66, 1988, N0.12, pp.503-507.
20. Zhor, J. "Attestation of Superplasticizers"; Zement und Beton (Vienna), Vo1.34,

1989, N0.2, pp.61-62.
21. Zhor, J. "Effect of Newly Developed Naphthalene-Based Superplasticizer on

Some Properties of Cement Pastes and Mortars. In Proceedings, 8th Silichem
Conference, Zilina, 1990, pp.125-126.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14B 74 m 0662947 0520LLO TTT m
TABLE i - CHEMICAL ANALYSIS AND PHYSICAL PROPERTIES OF CEMENT'
CSA-AS Type 10 Cement

Description of Test
Sample 1 Sample 2
a Chemical Analvsis, %
Silicon dioxide (SiO,) 20.2 20.1

Aluminum oxide (40,) 5.3 5.3
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Ferric oxide (?&O,) 2.0 2.0
Calcium oxide, total (TCaO) 64.1 64.3
Magnesium oxide (MgO) 1.4 1.4
Sulphur tnoxide (SO,) 4.0 4.1
Equivalent alkali (as N%O) 0.82 0.92
Calcium oxide, free (FCaO) 0.6 0.5
Loss on ignition 2.3 2.5
Potential Comuound Composition, %
Tricalcium silicate (C,S) 57.8 57.0

Dicalcium silicate (C,S) 14.2 14.5
Tricalcium aluminate (CpA) 10.6 10.7
Tetracalcium aluminofemite (C,AF) 6.1 6.2
Physical Tests
Fineness, passing 45pm sieve, % 86 86

Surface area (Blaine), m2.kg-' 375 380
Initial setting time (Vicat), h:min 2:20 2:lO
Autoclave expansion, % O O
Compressive strength, MPa 3 days 28.0 29.9
7 days 33.0 34.9
28 days 38.7 40.3
Supplied by the manufacturer
A C 1 SP-I148 9 4 m 0 6 6 2 7 4 9 0520LLL 93b m
342 Zhor, Bremer, and Lora
TABLE 2 - CHEMICAL ANALYSIS AND PHYSICAL PROPERTIES OF SILICA SAND'
Description of test I Nova Scotia Silica Sand

Mical Chemical Analysis. %
Silicon dioxide (SiO,) 98.70

Numinum oxide (Ai,O,) 0.18
Ferric oxide @%O,) 0.20
Titanium oxide (Tio2) 0.52
Calcium oxide (Cao) 0.05
Magnesium oxide (MgO) 0.01
Loss on ignition 0.06
Physical Promties
Specific gravity 2.60

Uniformity coefficient 2.00
Mohs hardness number 7
Organic impurities, % o. 1
Acid solubility, % c1
Absorption, % 0.6
Dry unit weight, kg.m-3 1600
'Supplied by the manufacturer
TABLE 3 - SIEVE ANALYSIS OF SILICA SAND
Percentage by mass
Sieve size
Individual Cumulative
No mm Retained Retained Passing
16 1.180 O O 0
30 0.600 0.6 O. 6 99.4
40 0.425 34.1 34.7 64.7
50 0.300 35.5 70.2 29.2
100 0.150 28.0 98.2 1.8
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 W 0662947 05201L2 872
TABLE 4 - BASIC PROPERTIES OF WATER-REDUCING ADMIXTURES
Admixture
Commercial Experimental
Chemical basis Naphthaierie
Physical state Aqueous soliition Aqueous solution
Appearance Brown liquid with Dark brown liquid

and odour slight moth-bail with woody odour
naphthalene odour
Specific gravity - 1.2 1.1 - 1.2
Solids concentration, -40 30 - 40

% by weight
Solubility in water Complete High
Volatiles 60% water 60 - 70% water
7.5 - 11.5 10- 12
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0520LL3 709
TABLE 5 - CLASSIFICATION OF ALCELL* LIGNIN SAMPLES
Molecular
Weight
LOW Medium High
Lowest ALO1 AL05 a109
Lower AL02 AL06 a110

Higher AL03 AL07 AL11
Highest ALO4 ALO8 a112
TABLE 6 - SULPHONATION LEVEL OF WATER-REDUCING ADMIXTURES
Density Solids Content Sulphur Content

Code
g.cm” g.1-’ % by weight g.1-1 % by weight
of solids
ALO1 1.112 359.0 32.28 5.6 1.56

AL02 1.125 410.2 36.46 7.2 1.76
AL03 1.120 376.1 33.58 4.7 1.25
ALO4 1.132 411.0 36.31 6.9 1.68
ALO5 1.115 398.2 35.71 9.7 2.44

ALO6 1.165 429.5 36.87 9.8 2.28
AL07 1.115 394.0 35.34 7.7 1.95
ALO8 1.143 457.8 40.05 7.9 1.73
ALO9 1.129 407.7 36.11 12.5 3.07

AL10 1.176 439.4 37.36 17.5 3.98
AL11 1.120 404.2 36.09 1.2 1.78
AL12 1.156 430.5 37.24 12.9 3.00
CNS I 1.197 I 506.6 I 42.32 I 54.74 I 10.81
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 94 E 0 6 6 2 9 4 9 0520114 6 4 5 E
TABLE 7 - TORQUE RESISTANCE DECREASE AND SETTING TIME RETARDATION

ON CEMENT SLURRIES
Admixture Torque Resistance, N.m Initiai Set, min

Code I Dosage,ml I Minimum I Decrease I Timeof Set I Retardation
1-
--- None --- -_- 140 ---
ALO1 40 4.42 3.58 340 200
AL06
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
AL12 3.94 4.06 260 120
CNS 30 2.26 5.74 160 20
Molecular Sulphonation Level

Weight
Low Medium High
Lowest 200 380 380

Lower 240 320 400
Higher 320 60 20
Highest O 120 120
A C 1 SP-348 94 Obb2749 0520335 5 8 3 W
TABLE 9 - PLASTICIZING AND AIR ENTRAINING EFFECT ON CEMENT MORTARS

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 0662749 0520LLb 418 U
TABLE 10 - FLUIDIFYING EFFECT ON CEMENT SLURRIES AS DETERMINED BY TORQUE

RESISTANCE DECREASE (N.m)
Weight I
b Medium High
Lowest I 3.58 4.18 4.28
Lower I 2.95 4.04 4.26
Higher I 3.74 3.60 3.51
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Highest I 3.32 3.36 4.06
TABLE 11 - FLOW OF CEMENT MORTARS (PERCENT)
Weight
LOW Medium High
59 60 62
42 51 60
Higher _- 49 --
41 45 52
TABLE 12 - AIR CONTENT OF CEMENT MORTARS (PERCENT)
Molecular Sulphonation Level

Weight
Low Medium High
37 37 37
26 30 31
Higher -- 32 --
Highest 25 27 30
A C 1 SP-148 94 0bb2749 0520117 354 W
MOLECULARWEIGHT (kû)
Fig. 2-U”lecular size distribution - medium sulphonotion level
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 0 6 6 2 9 4 9 052OLLB 290
1 10 100
MOLECULAR WEIGHT (kû)
Fig. 3-Molecular size distribution - high sulphonation level
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
1 10 100
MOLECULAR WEIGHT (kû)
Fig. 4-Molecular size distribution - commercial superplasticizer
A C 1 SP-148 74 m 0662947 0520117 127 m
350 Zhor, Bremner, and lora
Fig. 5-Torque resistance versus time - low sulphonation level
ELAPSEDTIME (hours)
Fig. &Torque resistance versus time - medium sulphonation level
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 94 H 0662949 0520320 ïY9 H
ELAPSED TIME (hais)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 7-Torque resistance versus time - high sulphonation level
O 1 2 3 4 5
ELAPSEDTIME@us)
Fig. &Torque resistance versus time - commercial superplasticizer
A C 1 SP-148 9 4 m 0662’749 0520121 885 m
SP 148-20
Practical Implications of Lithium-

Based Chemicals and Admixtures in
Controlling Alkali-Aggregate Reactions
by H. Wang, S. Tysl, and J.E. Gillott
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Synopsis: Two well understood and defined reactions in concrete are known as
alkali-silica reaction (ASR) and alkali-carbonate reaction (ACR). These two
reactions are extremely detrimental to the integrity of concrete structures. It has
been known that lithium-based chemical compounds can control and/or eliminate
the expansion caused by ASR. Based on this technology, a commercial chemical
admixture has been developed. The admixture is a ready to use liquid product
formulated to meet ASTM C494 requirements as a Type A water-reducing
admixture. This product is not intended to control ACR-induced expansion.
Therefore, the identification of the types of alkali-aggregate reactions is necessasi
in order to effectively control concrete expansion.
Keywords: Admixtures; air entrainment; alkali aqqreqate reactions; compressive strength;

expansion; flexural strength; freeze thaw durability; lithium compounds; water cement ratio
A C 1 SP-148 74 0662747 0 5 2 0 3 2 2 7 3 3 9
354 Wang, Tysl, and Gillott
AC1 member Hugh H.Wang is a Senior Project Engineer at Master Builders, Inc.,
Cleveland, Ohio. U. S. A. He obtained his BS and MS in Cement Technology from Wuhan
University of Technology, Hubei, China, and received PhD in Materials Science from the
University of Calgary, Alberta, Canada. He is currently involved in the development of
admixtures for enhancing concrete durability.
AC1 member Stephen R. Tysl is a Project Engineer at Master Builders, Inc., Cleveland,
Ohio, U. S. A. He has a ES in Civil Engineering from the University of Akron and is a
registered State of Ohio Engineer in Training. He is a member of AC1 Committees 20 1,
.Durability; 207, Mass Concrete; and 221, Aggregates.
AC1 member Jack E. Gillott is a professor in the Department of Civil Engineering at the
University of Calgary, Alberta, Canada. He received his PhD from the University of
Liverpool and his DSc in Engineering from the University of London, United Kingdom.
His research interests include concrete durability and the influence of microstructure and
composition on the behavior of materials.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
INTRODUCTION
Failure of concrete due to the expansion caused by the chemical reaction

between alkalis in concrete pore solution and the reactive silica in aggregate is a
world-wide problem. It was first identified in the late 1930's and early 1940's.
Since its discovery, various efforts have been made to solve this problem including
using low alkali cement, selecting non-reactive aggregate, and blending mineral
admixtures with cement. As early as the 1950's (i), it was found that lithium salts
were effective chemicals to controlleliminate the expansion caused by alkali silica
reaction (ASR). In recent years, widespread problems of ASR have renewed the
interest of lithium compounds as a mean of combating ASR-induced concrete
destruction (2, 3). The development of a commercial chemical ASR inhibitor
admixture is based on this technology.
This paper presents the results from test programs dealing with the effect
of lithium compounds on expansion caused by both ASR and ACR. A lithium-
based commercial chemical admixture was thoroughly tested to investigate its ASR
inhibition characteristics and its effect on the concrete properties.
SIGNIFICANCE OF ASR AND ACR
Both chemically and mineralogically, aggregates used in concrete differ

significantly from source to source. No aggregate is completely inert in a concrete
environment. Chemical reactions constantly occur between cement, (through the
pore solution), and aggregates. ASR and ACR are the two well understood and
A C 1 SP-148 9 4 W 0662949 0520123 658 m
defined reactions that have been found to be extremely detrimental to the integrity
of concrete structures. In order to effectively control ASR-induced expansion with
lithium compounds and lithium-based chemical admixtures, a clear understanding
of ASR and ACR reaction and expansion mechanisms is necessary.
ASR is the reaction between alkalis in cement pore solution and a reactive
silica component in certain aggregates. The reaction develops a product called
alkali-silica gel The gel exists in an unstable energy state and has a strong
tendency to absorb moisture to reduce its free energy. When it absorbs moisture,
the gel expands. The phenomenon is typically described by a two part reaction and
expansion process summarized as.
Chemical reaction:
xSiOz + yNa(K)OH -+ Na(K)ySlxOz.aq
Reactive silica Alkali in concrete Alkali-silica gel
in aggregate pore solution
Expansion:
Na(K)ySixOz.aq + Moisture 4 Na(K)ySixO,~wH~O
Alkali-silica gel Expanded alkali silica
gel
The Na(K):Si ratio of alkali-silica gel in concrete varies. The volume

change is reversible depending on the moisture conditions. From the reaction, it is
shown that alkalis are critical to the formation of the alkali-silica gel.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
ACR is the reaction between alkalis in cement pore solution and a reactive
dolomite mineral in certain aggregates. It is understood that the concrete
expansion is a result of a chemical dedolomitization reaction which results from the
presence of alkalis in the pore solution. The dedolomitization reaction is expressed
as follows:
CaMg(C03)2 + 2NaOH + C a c o 3 + Mg(OH), + 2Na+ + C03-2

Dolomite Calcite Brucite
There are two principal explanations of the dedolomitization-induced

expansion. A recent explanation claims that direct expansion occurs due to the
crystallization pressure of brucite in a restricted area (4). An earlier explanation
involves indirect expansion due to unwetted clay impurities in the dolomite being
exposed to moisture. This moisture is allowed into the dolomitic limestone matrix
A C 1 SP-148 74 m Ob62747 0520124 594 m
356 Wang, Tysl, and Gilloft
because of the increase of permeability resulting from the dedolomitization

reaction (5).
EXPERIMENTAL PROGRAMS AND RESULTS
Three independent experimental programs were conducted. The first was

to test the expansiveness of lithium-silica complex, the second to investigate the
effect of lithium compounds on ACR, and the third to examine the effect of a
lithium-based commercial chemical admixture on ASR and other engineering
properties.
Expansiveness of Lithium-SilicaComplex
Materials and Sample Preparation: Highly alkali-reactiveopal was obtained

from a source in Nevada, U.S.A. It was crushed to pass No. 100 mesh sieve (150 p
m). The chemical composition and physical properties of the opal are given in
Table 1. Reagent grade Ca(OH)2 , NaOH and LiOH.H20 were used. Three pastes
were prepared by mixing powdered opal with (a) NaOH solution; (b) LiOH.H,O
and distilled water and (c) LiOH.H,O and NaOH solution. The composition and
designation of the pastes are given in Table 2. Each of the three pastes had a
watedopal ratio of 0.48. Thoroughly mixed pastes were sealed in plastic containers
and stored in an oven conditioned at a temperature of 38f2"C.After three weeks,
the pastes were taken out, oven dried and ground to pass a No. 200 mesh sieve
(75pm) to measure free swelling.
Free Swelling ExDerimental Test: The detailed free swelling technique was
developed by Struble and Diamond to determine the expansive properties of
synthetic alkali-silica gel (6). A slightly modified apparatus was explained and
utilized in a previous paper (7).
Results: Free swelling results are shown in Fig. 1. It is shown that the
alkali-opal complex developed a large amount of fiee swelling, while lithium-opal
complex developed little free swelling. The paste with the molar ratios of
Li20/SiO2=0.1 and Na20/Si02=0. I also showed little free swelling.
Effect of Lithium Compounds on ACR
Materials: Several lithium compounds were selected to investigate their

effect on ACR. The compounds included in the ACR investigation were lithium
hydroxide, lithium carbonate, an organic lithium salt and a lithium-containing high-
range water-reducing polymer. ASTM Type I portland cement and an alkali
carbonate reactive aggregate were used in this program. The carbonate reactive
aggregate was from the Pittsburg quarry, Ontario, Canada.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 m 0662949 0520325 420 m
Experimental Technique The method adopted for expansion tests was a

short-term accelerated test method Concrete prisms of 3x3x14-inch (7 6x7 6x35 5
cm) size were cast A cement factor of 304 kg/m3 was used in all mixtures The
prisms were demolded after 1-day pre-curing, and moist cured for an additional 6
days at a controlled moisture room temperature of about 23°C (73'F) Then, the
prisms were saw-cut to produce prismatic test specimens with a 3/4-inch square
cross-section and 2-inch length (1 9x1 9x5 1 crn) The test specimens were
conditioned in one molar sodium hydroxide solution at an oven temperature of 80"
C (176°F) for 24 hours prior to initial length measurements After the initial length
measurements, the specimens remained in the solution until the time of testing The
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
length measurements reported were taken at 35 days
Results The effect of lithium compounds on ACR is shown in Fig. 2. All of

the lithium compounds tested increased linear expansion with the exception of an
organic lithium salt. It can also be shown that in special cases, expansion will
increase proportionally to the lithium hydroxide content.
The Performance of Commercial Aikali-Silica Reaction Inhibitor (ASRI)
This section presents the data collected from the investigation of the effect
of ASRI on ASR and the properties of concrete.
Effect of ASRI on ASR
Materials: The ASRI admixture used is a lithium-based product which meets the
requirements of ASTM C-494 as a type A admixture. One silica reactive aggregate
used to evaluate the effectiveness of ASRI in controlling ASR was from West
Central Pennsylvania, U.S.A ASTM Type I portland cement (ASTM C-150) was
used. The chemical composition and mineral component of the cement is also
given in Table 1. A Pyrex glass was also used as a silica reactive aggregate to
evaluate the effectiveness of ASRI in controlling ASR.
Expansion Test: CSA A23.2-14A was used to test the effect of ASRI on reactive
aggregate. The alkali content in the cement was increased to 1.25% sodium oxide
equivalent by adding sodium hydroxide to the mixing water. The concrete mixture
proportions followed the requirements in CSA A23.2-14A. Concrete prisms of
3x3xll-inch (7.6x7.6x27.9cm) were cast, demolded after 1-day pre-curing, and
initial length measurements were taken. All prisms were stored above water in
plastic containers. All containers were put in a hot room conditioned at a
temperature of 35*2OC (95*30F).
ASTM C-441 was also used to test the effectiveness of ASRI in controlling
expansion caused by Pyrex aggregate. 50% Pyrex was used in all mortar bar
mixtures. The alkali content was increased to 1.0% sodium oxide equivalent.
Mortar bars of lxlxll-inch (2.54~2.54~27.9 cm) were cast, demolded after 1-day
A C 1 SP-148 94 0662949 0520126 367
pre-curing, and initial length measurements were taken. All bars were stored above
water in plastic containers. Ail containers were put in a hot room conditioned at a
temperature of 35*2OC (95*30F).
Results: Fig. 3 shows the expansion results of concrete prisms made with silica
reactive aggregate up to one year. It can be seen that expansion was reduced by
the addition of ASRI and the higher the dosage, the more effective was the
admixture. Fig. 4 shows the effect of ASRI on expansion of mortar bars made with
Pyrex aggregate.
Effect of ASRI on the Properties of Concrete Made with Reactive Agqegates
Material: The effect of ASRI on the plastic properties and strength development of
concrete made with two silica reactive aggregates was also investigated. One
aggregate was from the Spratt quarry, Ontario, Canada and the other one was
from New Mexico U S.A A Type 10 high-alkali portland cement (CSA-A5-M88)
was used in this study. The alkali content of the cement was 0.85% sodium oxide
equivalent. The total alkali content in concrete was increased to 4.0 kg/m3 (6.77
lb/yd3) by addition of NaOH into mixing water in order to accelerate the ASR.
Experiment and Results: The mixture proportions, plastic properties, strength

development, air void parameters and durability factor of freezing and thawing of
concrete made with the Spratt reactive aggregate are shown in Table 3. The plastic
properties of the ASRI treated concretes did not show significant differencesfrom
the reference mixture. The W/C was reduced by about 7% at 980 and 1630
mL/I 00-kg cement doses. To obtain the same level of air content in concrete, the
dosage of air-entraining admixture decreased. Both initial and final setting times
were delayed about one hour at 980 mL/IOO-kg cement dose and about one and
half hours at 1630 mL/lOO-kg cement dose. The compressive strength of ASRI
treated concrete increased at 980 and 1630 mL/iOO-kg cement doses due to the
W/C reduction. The flexural strength was also increased at 1630 mL/lOO-kg
cement doses. The spacing factor of entrained air system was not significantly
changed by the admixture. Durability factor of freezing and thawing was also not
significantlyaffected by the admixture.
The mixture proportions, plastic properties, strength development and

durability factor of freezing and thawing of concrete made with New Mexico
reactive aggregate are shown in Table 4. Again the plastic properties of the ASRI
treated concretes did not show significant differences compared to the reference
mixture. The W/C was reduced by about 7% at 980 mL/iOO-kg cement and 1630
mL/lOO-kg cement doses. To obtain the same level of air content in concrete, the
dose of air-entraining admixture was decreased. The initial setting time was
delayed about one hour and the final setting time was delayed about forty minutes
at both doses. Both compressive and flexural strengths increased at both dosages
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 I0662749 0520327 2 T 3
of ASRI due to the WIC reduction. Durability factor of freezing and thawing was
not significantly affected.
DISCUSSION
ASR and ACR are very different reactions and have equally different
expansion mechanisms. As a result, lithium coinpounds were found to have
different effects on the expansion caused by these two types of reactions.
After careful examination of the inhibiting action of lithium compounds on

ASR, Sakaguchi et. al. (2) found that the alkali concentration of the mortar
containing no lithium decreased dramatically due to the consumption of alkalis in
the alkali-silica reaction. Conversely, the alkali concentration in the pore solution
of the mortar containing lithium compounds remained consistently high because of
the reaction of lithium ions with reactive silica Evidently, the presence of lithium
ions in pore solution inhibits the formation of soluble alkali-silica gel. This was
further confirmed by analyzing the reaction interface of cement paste and reactive
Pyrex aggregate with Energy Dispersive X-Ray Spectrometer techniques.
Sakaguchi et. a1.(2) observed that a distinct 300 pm reaction zone existed between
the paste and Pyrex glass after a 4 month reaction at 40OC. No such zone was
found in the mortar containing lithium compounds.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
The lithium-silicate complex is insoluble in water. As a result, lithium
silicate does not have the capacity to absorb water and swell as demonstrated by
the free swelling tests (Fig. 1). It is interesting 'to note that when half of the lithium
was replaced by sodium, the reaction product still remained non-expansive. This
implies that in concrete, conditions occur in which alkalis and lithium ions Co-exist
and the inhibition action of the lithium may be unaffected.
Since the reaction and expansion mechanisms of ASR and ACR are
different, it is not surprising that lithium compounds do not inhibit expansion
caused by ACR. In fact, the addition of lithium hydroxide may aggravate the ACR
induced expansion.
A lithium-based commercial chemical admixture has been formulated to

controiíeliminate ASR. This admixture does not adversely affect the properties of
either fresh or hardened concrete. A systematic study indicates that the admixture
generally reduced water-cement ratio without decreasing workability of concrete.
Setting time is not significantly changed and meets the requirements of ASTM C
494 as a Type A admixture. Concrete strength is increased because of the WIC
reduction. Suppression of expansion caused by ASR is one of the main functions
of this product. However, this product is not intended to control ACR-induced
expansion. Therefore, in engineering practice, a clear identification of the reactive
aggregate type is necessary prior to using this product.
A C 1 SP-148 94 œ 0662949 0 5 2 0 1 2 8 1 3 T œ
CONCLUSIONS
1). Lithium silicate is a water insoluble and non-expansive product. Lithium-

incorporated reaction with silica can reduce expansion potential due to
sodium(potassium)-silicareaction. Significant level of lithium is needed in order to
maintain the effective control on ASR-induced expansion.
2). Lithium chemicals and lithium-based admixtures only work on ASR. All the
lithium compounds tested showed no reduction of ACR-induced expansion. In one
specific case, lithium hydroxide aggravated ACR induced expansion. This implies
that the identification of alkali aggregate reaction type in aggregates is necessary in
order to ensure effective control of alkali aggregate reaction by using lithium
technology.
3). A lithium-based commercial admixture can be used to delay the onset of ASR
and reduce concrete expansion. The mechanical properties of the concrete treated
with this admixture are enhanced due to the WIC reduction. The other physical
properties of fresh and hardened concrete are not adversely affected.
4). Up to one year expansion data of concrete prism made with silica reactive
aggregate are presented in this paper. The lithium-based admixture can reduce
expansion by about 60% at one year age with a dosage of 1630 mL1100-kg
cement. It is still in process to monitor longer term expansion and other
engineering properties.
ACKNOWLEDGMENT
This paper is published with the permission of Master Builders, Inc. The
first two authors wish to thank their colleagues in Master Builders Technique
Center, Cleveland, Ohio for their contributions in the development of ASRI.
Special thanks to Mr. Richard Buehner, Senior Technical Specialist, for his
assistance in carrying out concrete mixing and tests.
Part of the work presented in this paper was carried out by the first author
at the University of Calgary where he was employed as Research Associate.
REFERENCES
(1) McCoy, W. J. and Caldwell. A. G., "New Approach to Inhibiting Alkaii-

Aggregate Expansion", Journal of the American Concrete Institute, Volume 47,
1951, pp. 693-706.
(2) Sakagushi, Y., et al., "The Inhibitive Effect of Lithium Compounds on Alkali-
Silica Reaction", Proceedings of the 8th International Conference on Aikaii-
Aggregate Reaction in Concrete, Kyoto, Japan, 1989, pp.228-234.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-LYB 9Y m Obb29Y9 0 5 2 0 3 2 9 076 m
Superplastitizers and Chemical Admixtures 361
"Lithium Salt Admixtures - An Alternative Method to Prevent

( 3 ) Stark, D.,
Expansive Alkali-Silica Reactivity." Proceedings of the 9th International
Conference on Alkali-Aggregate Reaction in Concrete, London, U. K., 1992,
Volume 2, pp.1017-1025.
(4) Tang, M. S., Liu, Z. and Han, S. F., "Mechanism of Alkali-Carbonate

Reaction", Proceedings of the 7th International Conference on Alkali-Aggregate
Reaction in Concrete, Ottawa, Canada, 1986, pp.275-279.
( 5 ) Gillott, J. E. and Swenson, E. G., "Mechanism of Alkali-Carbonate Rock

Reaction", Quarterly Journal of Engineering Geology, Volume 2, 1969, pp. 7-23
(6) Struble, L. and Diamond, S., Swelling Properties of Synthetic Alkali Silica
I'
Gels", Journal of American Ceramic Society, Volume 64, I98 1, pp. 6 1 1-6 17.
(7) Wang. H.and Gillott, J. E., "Effect of Ca(OH)2on Alkali-Silica Reaction".

Magazine of Concrete Research, Volume 43. 1991, pp. 2 15-2 18.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 9Y W Obb2949 0520330 8 9 8 m
TABLE 1 - CHEMICAL COMPOSITION AND PHYSICAL PROPERTIES OF OPAL AND CEMENT
opal Type I Cement

Chemical Anaivsis
Calcium oxide (Cao) % 0.41 63.85
Silicon dioxide (Si02)'Y0 94.19 2 1.22
Aluminum oxide (Ai2O3)% 0.90 4.69
Femc oxide (Fe203)Yo 1.44 3.17
Magnesium oxide (MgO)% O. 14 2.87
Sulfur trioxide (SO3) 2.48
Sodium oxide (Na20)% 0.0 0.15
Potassium oxide (K20)% 0.03 0.64
Loss on ignition % 2.73 1.00
Insoluble residue % 0.10

Total alkali as Na20 0.02 0.57
Surface area m2/g 18.39
Specific gravity 1.942
Bouge's Potential Compounds
c3s 56.0
c2s 19.0
C3A 71
10.0
TABLE 2 - PROPORTION AND DESIGNATION OF OPAL PASTE
I Proportion: mole ratio

I Designation
I
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Na20:Si02=0.2 OPAL-Na2
Li20 Si02=0.2 OPAL-Li2
Na20:Si02=0.1+Li20:Si02=0.1 OPAL-NalLi1
A C 1 SP-148 9 4 ûbb2949 0520333 7 2 4
TABLE 3 - MIXTURE PROPORTIONS, PLASTIC PROPERTIES, STRENGTH DEVELOPMENT, AIR VOID

PARAMETERS, AND DURABILITY FACTOR OF CONCRETE MADE WITH SPRAll REACTIVE AGGREGATE
Reference ASRI ASRI

Mixture 980mL/100-kg 1630mL/lOO-k~
cement cement
Type 10 High Alkali Cement 383 387 388
(0.85% Na2O equivalent)
(Wm3)
Spratt Aggregate (kg/m3) 1061 1073 1076
Non -reactive Fine Aggregate 673 680 682
(kg/m3)
Water (kg/m3) 167 158 157
w/c 0.44 O 41 0.41
Vins01 Resin (mL/l OOkg 55.3 45.6 29 3
cement)
ASRI (mLí100kg cement) O 980 1630
Slump (mm) 152 133 133
Air Content (%) 5.9 6.0 5.8
Unit Weight (kg/m3) 2273 2299 2299
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Setting Time (hr:min)

Initiai (3.45MF'a) 4:25 5:30 5155
Final (27.6MPa)' _5.45
. ._ 6:50 7.05
Compressive Strength ( m a ) [.Percent as Reference]
7d 28.8 I
31.0[108] 31.6[110] I I
28d I 34.0 I 36.8[108] I 37.4[ 1101
Flexural Strength ( m a )
Air Content by Petrographic 6.15 5 76 4.85

Method (%)
Spacing Factor (mm) 0.22 0.21 0.20
A C 1 SP-14ô 94 W 0662949 0520332 660 W
TABLE 4 - MIXTURE PROPORTIONS, PLASTIC PROPERTIES, STRENGTH DEVELOPMENT AND

DURABILITY FACTOR OF CONCRETE MADE WITH NEW MEXICO REACTIVE AGGREGATE
Reference ASRI ASRI
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Mixture BOmL/lOO-kl 1630mL/100-
cement kg cement
Type 10 high alkali Cement 384 385 387
(0.85% Na20 equivalent)
(kg/m3)
New Mexico Aggregate 1066 1070 1072
(kg/m3)
Non-reactive Fine Aggregate 656 682 683
(kdm3)
Water (kg/m3) 165 155 154
w/c o 43 0.40 0.40
Vinsol Resin (mL/100kg 46.9 35.2 23.4
cement)
ASRI (mL/100kg cement) O 980 1630
Slump (mm) 152 140 140
Air Content (%) 5.8 5.6 5.5
Unit Weight (kg/m3) 2266 2286 2289
Setting Time (hr:min)
Initial (3.45MPa) 4.15 5:lO 5:15
Final (27.6MPa) 5:30 6:15 6: 10
3d 22.9 26.7[116] 30.6[133]

7d 28.4 33.3[117] 35.1[123]
28d 35.0 39.0[ 11 i] 40.4[1151
A C 1 SP-148 9 4 m Obb2949 0 5 2 0 3 3 3 5 T 7 m
CIE
'O sol 4
~~~
4
4 Opal-Na2
4
4 -
Opal-Li2
'
A Opal-NalLil
L-
0
4 L L , L
2
L4 O L * * I
O 10 20 30
Time in Days
~ _ ~ _ ~
Fig. 1-Free swelling of sodium-opal and lithium-opal pastes
- 20
10
U Lithium hydroxide
@0.25%
c Lithium hydroxide
.2 0 @1.1%
cn
IT Lithium carbonate
m
g -10 @0.25%
w Lithium carbonate
.5 -20 @I .O%
c Organic lithium sal
O
c -30 Lithium-containing
2
T3 high-range water-
-40 reducing polymer
rY
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
-50
Fig. '>-Effect of lithium compounds on the reduction in expamion due to alkoli carbonate reaction
A C 1 SP-148 94 D O b 6 2 9 4 9 0520334 LI33 =
0.2
0.18
0.16
[
0.14
O Reference
W
BASRI @ 980 m U
c 0.12
.-O 1O0 kg cement
u) 0.1
i 0.08
MASRI @ 1630
mL/1O0 kn cement
J 0.06
0.04
0.02
O
6 IO 12
Time in months
Fig. &Effect of ASRI on expansion of silica reactive aggregate
-5 0.5
0.6
0.7
I - Reference
+ASRI@1630
cement
ml/lOOkg
A
w
.-8 0.4 I
rn
E
I
0.3
e 0.2
m
0.1
O 4
O 0.5 1 2 3 4 5 6 7
Time in Months
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. &Effect of ASRI on the expansion of mortar bars made with pyrex aggregate
A C 1 SP-148 94 Ob62949 0520135 3 7 T
SP 148-21
Plasticizing Characteristics of
Sulfonic Acid Polymer Containing
Methacrylic Acid Derivatives
by E.I. Tazawa, A. Yonekura, M. Takahashi,
S. Miyazawa, and K. Kawai
Synopsis : Various types of superplasticizers which can maintain the slump of

concrete for longer period, have been extensively investigated. A new type of
superplasticizer of high water reducing and slump maintaining capacities, that
is composed of sulfonic acid polymer containing methacrylic acid derivatives,
has recently been developed.
In this study, influences of cement type, cohcrete temperature, and
pozzolans on the properties of fresh and hardened concrete with this superplas-
ticizer were investigated. Reference superplasticizers were a naphtalene-based
superplasticizer(A) and an amino sulfonic acid(B) which are widely used all
over the world.
A significant increase in water reducing capacity to get the same consist-
ency was observed at a much lower dosage. Absolute value of zeta potential of
cement particle with the superplasticizer increased with elapsed time until 90
minutes after mixing, which can explain the high slump retention capacity of
the concrete.
Plasticizing effects of superplasticizers were more pronounced for the
concretes with fly ash or blast furnace slag used as blending agents. Bleeding
of concrete decreased slightly. Properties of hardened concrete such as com-
pressive strength and drying shrinkage were nearly of the same level as those
of concrete with naphtalene-based superplasticizer.
Kewords: Bleeding (concrete); concretes; fly ash; fresh concretes; plasticizers; elastics. Dolvrners,
and resins; setting (hardening); silica fume; slags; slump; suuemlasticizen
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9ir 0 b b 2 9 4 9 052033b 206 W
368 Tazawa et al
Ei-ichi Tazawa, ACI-member, is a professor of civil engineering, Hiroshima

University, Higashi-hiroshima city, Japan. He received his B.Sc, from the
University of Tokyo in 1960, M.Sc. from MIT in 1968 and Ph.D. from the
University of Tokyo in 1978. He had been a chief research engineer of Taisei
Corporation in Tokyo.
Asuo Yonkura is a professor of civil engineering, Hiroshima University, Hi-
gashi-hiroshima city, Japan. He received his M.Sc. from Hiroshima University
in 1967, and his Ph.D. from Tokyo Institute of Technology in 1981. He has
published papers on high-strength concrete, flowing concrete. etc.
Masatoshi Takahashi is a research manager, chemical research laboratories,
Lion corporation, Tokyo Japan. His recent study is on superplasticizers of
concrete.
Shingo Miyazawa is a research associate of civil engineering , Hiroshima
University, Higashi-hiroshima city, Japan. He received his B.Sc from Tokyo
Institute of Technology in 1982 and his Ph.D from Hiroshima University in
1992.
Kawai Kenji is a research associate of civil engineering, Hiroshima University
Higashi-hiroshima city, Japan. He received his M.Sc. in 1987 from University
of Tokyo, his Ph.D. in 1990 from the University of Tokyo. His recent study is
on migration of substances in concrete.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
INTRODUCTION
Three types of new superplasticizers composed of sulfonic acid polymer
containing methacrylic acid derivatives have recently been developed. In this
study, effects of these superplasticizers on fluidity, slump loss, setting time,
hydration characteristics and strength of concrete were experimentally investi-
gated.
Mixture proportions were decided by varying the water:cement and
dosage of superplasticizers at different temperatures of the experimental envi-
ronment.
In this study, blast furnace slag, fly ash and silica fume were used to
replace cement by .SO%, 20% and 10% respectively. Besides replacement of
cement, silica fume was used also as an anti-segregation admixture. To
determine the capacity of dispersion of cement particles due to superplasticiz-
ers, zeta potential tests were carried out. The results were compared with those
of a reference superplasticizer.
OUTLINE OF EXPERIMENTS
Materials
Superplasticizers--Eight types of superplasticizers were used (Table 1). The

new superplasticizers, C, E, CE, C/E, CE/E and C/CE/E are of sulfonic acid
polymer containing methacrylic acid derivatives. The reference superplasticiz-
ers, are named A and B, where A is a naphtalene-based superplasticizer and B
A C 1 SP-14B 94 0662949 0520337 1 4 2
is amino sulfonic acid.

Superplasticizers C and E were not used separately but were mixed with
each other forming the four types of superplasticizers.
Cement-- Cements used were normal and moderate heat portland cements.
The specific gravities of these cements were 3.13,3.13 and 3.18, respectively.
Aggregate and admixtures-- Coarse aggregate was quartzite trachyte crushed
rock of 20mm maximum size and specific gravity of 2.67, absorbed water
0.31%, and fineness modulus of 6.55. Fine aggregate was weathered granite
sand of specific gravity of 2.57, absorbed water 1.99% and fineness modulus of
2.88.
In order to adjust air content, Vinsol resin air entraining agent was used.
Silica fume, slug and fly ash of specific gravities of 2.2, 2.9 and 2.27 and
specific surface areas of 160000, 6130 and 4200 (cm2/g) respectively were
used.
Tests for Concrete Properties

The dosages of superplasticizers were decided such that the slump of
concrete would be between 18cm and 22cm. The mixing was carried out by a
pan type mixer. Slump, rate of slump flow and final slump flow were meas-
ured. The mixture was kept on a hand mixing pan and covered by vinyl sheet to
avoid the influence of the surrounding air.
Slump, slump flow and rate of slump flow tests were repeated after 30,
60 and 90 minutes with the tested concrete bt:ing isolated from the initial mix-
ture in the pan. Tests for bleeding and setting of concrete were also carried out.
Development of strength of concrete aí 3, 7 and 28 days was studied by
compressive strength of cylindrical specimens of 100mm diameter and 200mm
height after standard curing.
Tests
_ _ _ for_Cement
_ _ _Paste
~ Properties
Zeta potential--As an indicator of dispersion tendency of cement particles,

zeqotential of cement particles with superplasticizers was determined using
an electrophoresis type apparatus. The samples for zeta potential measurement
were hand mixed in a beaker by using distilled water.
The cement paste sample for the test was diluted 1500 times of its volume
in distilled water just before the zeta potential measurement.
The zeta potential was measured after 30, 60 and 90 minutes. During that
interval the specimen was covered to avoid the influence of surrounding air.
-__--
Test for the rate of hydration --The rate of hydration was determined by
the measurement of chemical shrinkage caused by hydration of cement. The
method was established in this laboratory and the absorbed amount of water by
the created pores was observed (Fig.l)[l]. Cement paste used for this test had
water cement of 0.3 and 0.5, while the dosage of superplasticizers was half of
those for concrete.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 9 9 D 0 6 6 2 9 4 9 0520338 089
370 Tazawa et al
TEST RESULTS AND DISCUSSION
Properties of Fresh Concrete with Superplasticizers
The
_ _ -effect superplasticizers on the fluidity of concrete-- Fig. 2 shows
the behaviour of slump and slump flow due to the dosage of superplasticizer.
In this test the superplasticizer was C and normal portland cement with W/C=
0.3 was used, while water content was 170 kg/m3..
From this figure, it can be seen that slump and slump flow increased
significantly with the change of dosage from 0.15% to 0.25%.
For dosages of more than 0.25%, segregation occurred, which indicates
that fluidity of concrete is fairly sensitive to the dosage of superplasticizer.
Change of slump and slump flow with time --Figures 3 and 4 shows
the change of slump and slump flow with time in concrete with normal port-
land cement, which had W/C of 0.3, water content of 160kg/m3and W/C of 0.5
with water content of 180kg/m3.
From the figures, it can be noted that for methacrylic acid polymer (new
type of superplasticizer), the minimum segregation was obtained at the opti-
mum dosage of superplasticizer at which slump loss was low, while for a
reference superplasticizer (TYPE B) slump loss was low but the slump flow
loss was large.
The superplasticizer C/E-2 (high mixing ratio of E) produced concrete
with slump loss which was nearly equal to the concrete with C/E-1 (low
mixing ratio of E) but its slump flow loss was lower at the same dosage.
In all the cases shown in Fig. 3, initial slump was from 22cm to 25cm.
The optimum dosage at which segregation and slump loss did not occur was
obtained for concrete with the slump of from 20cm to 25cm for all the super-
plasticizers. In the concretes with W/C = 0.5, the slump losses were very
small, but the flow losses were different for the different types of superplasti-
cizer. Particularly, it is noted that high fluidity was obtained for the concrete
with the superplasticizer CE/E at a half the dosage of reference superplasticizer
A.
Bleeding of concrete-- For all the types of superplasticizers tested, it was
observed that in the case of concrete with W/C of 0.3 bleeding did not occur,
while for W/C of 0.5 there was bleeding. As shown in Fig. 5 , for concrete
with the superplasticizer CE/E of dosage 0.25% and 0.3%bleeding was low.
But for CEE with a dosage of 0.4%, bleeding was 2.5 to 3 times of those with
dosages from 0.25% to 0.3%.
For the reference superplasticizer A, the bleeding of concrete as well as
its slump loss with the dosage of 0.6% was low. Its bleeding at the dosage of
0.7% was about 3 times larger than that at 0.6%.In some cases, when the
dosage of superplasticizer was increased in order to prevent the slump loss for
one to two hours, bleeding increased considerably.
Setting time of concrete-- From Fig. 6, it can be seen that by using the
superplasticizers C/E-1 and C/CE/E, the setting time of concrete did not
change with dosages varying from 0.5 to 0.6 percent, but for CE-2, the setting
time was larger. Setting time of concrete with the reference superplasticizer B
of dosage of 0.5% was nearly equal to that of plain concrete without superplas-
ticizer. But at a dosage of 0.5596, the setting time increased considerably.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 0662949 0520139 TL5
From Fig. 7, the setting time of concrete with C E E at W/C of 0.5 was
from 1.5 to 2 times that of plain concrete. For W/C of 0.5, slump losses were
low, but slump flow loss was observed for concrete with various superplasti-
cizcrs.
From Fig. 4 it is noted that high fluidity was obtained for concrete with
superplasticizers CE/E-l and 2 at about half of dosage of the reference super-
plasticizer A. While the setting time of concrete with superplasticizer A was 2
to 2.5 times longer than that of plain concrete.
The Effects of Cement Type, Concrete Emperature and Pozzolans on the

Properties of Fresh and Hardened Concrete with Superplasticizers.
From Figs. X and I), it can be seen that, slump losses of concrete with
the new superplasticizers( CEE) at low temperatures (5°C) were lower than
those of concrete at high temperatures (30°C). In Fig. 10, the slump losses of
concrete were slightly decreased by using blast furnace slag or fly ash in both
cases ( new superplasticizer C E E or reference superplasticizer A).
From Table 3, the compressive strengths of concrete with C E or C were
same as those of a reference superplasticizer A (MT 2000 WH) or A(MT 150)
respectively.
As shown in Table 4, the drying shrinkages of concrete with methacrylic
acid polymer (CEE, C E , C) were same as those of reference naphtalene based
superplasticizer concretes A(MT 2000S, MT IOOOWH, MT 150).
The Effects of Type of Superplasticizers and their Dosages on the Zeta Poten-
tial of Fresh Cement Paste with Superp1asticii:er
From Figs. 11 and 12, it can be seen that absolute values of zeta potential
of cement particles mixed with C E E (dosage 0.15%) at W/C=O.6 were higher
compared to those mixed with reference superplasticizer (dosage from 0.25%
to 0.3%). The absolute value of zeta potential increased with time for CEIE,
while, the zeta potential of plain cement particles (without superplasticizer)
changed from negative to positive ut about 12 minutes after mixing. In the case
of reference superplasticizers A and B the absolute values of zeta potential
decreased. For plain cement paste (without superplasticizer), it is noted that
immediately after mixing, the surfaces of cement particles were negatively
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
charged but with the progress of hydration, the calcium cations were adsorbed
at the surfaces of the cement particles after being dissolved out of cement parti-
cles, to crcatc a positive charge[2].
In the cements with superplasticizers C E , C E E and C/CE/E, the abso-
lute value of zeta potential increased with time until 90 minutes after mixing,
quite different to the plain pastes and pastes with reference admixtures. The
differences observed in the three cases may be responsible for the high slump
retention capacities shown by the new types of superplasticizers. Figures 11
and 12 show that the absolute value of zeta potential for cement particles with
the superplasticizer CE/E was larger than that of cement particles with the
reference superplasticizer A and its value tended to increase with time. This
behaviour explains the high increases in slump and low slump loss.
A C 1 SP-248 94 0662947 0520340 737
372 Tazawa et al
Effect of Superplasticizers on the Rate of Hydration of Cement Paste

--
Figs. 13 and 14 show the relationship between the chemical shrinkage of

cement paste with time for cement pastes with WIC of 0.3 and 0.5 respectively.
From these figures, the rate of hardening of cement pastes can be determined
from the assumption that the chemical shrinkage occurs due to the hydration
process. It is noted that the rate of hydration changed with WIC.
For W/C of 0.3, the rate of hydration of cement paste with superplasti-
cizer was lower than that of plain cement paste (without superplasticizer),
while for W/C of 0.5 the rate of hydration of cement paste was a little smaller
than that of plain cement paste.
Flow Characteristics & Concrete Superplasticizers Containing Silica
It was noted that fluidity of concrete with new superplasticizers was

sensitive to small changes in the dosages of the superplasticizers. Fig. 15,
shows that with silica fume as a replacement by 10% of cement, neither slump
loss nor significant loss of slump flow was observed although in this case the
dosage of the superplasticizers was double that of the one without silica fume.
So with the use of silica fume the sensitivity of fluidity to changes in the dosage
of superplasticizer is reduced, and thus its practical applicability is improved.
Properties of Hardened Concrete
From Fig. 16, it is evident that the development of strength for all super-
plasticizers except for C/CE/E was about the same. In the case of concrete with
silica fume, the strength at early ages (3 and 7 days) was slightly lower but the
28 days strength was approximately same for all.
For the W/C of 0.5 concrete with a reference superplasticizer A and
dosage of 0.796, showed lower strength probably due to the defects caused by
bleeding water[3].
CONCLUDING REMARKS
From this study, the following conclusions were obtained.
1. For WIC of 0.5, the superplasticizer dosage needed to achieve a given
fluidity can be reduced by half of reference superplasticizer A by using new
superplasticizer CE/E (methacrylic acid derivatives).
2. For W/C of 0.3, fluidity of concrete containing the new superplasticizers
was sensitive to slight changes in dosages, so dosages must be strictly con-
trolled in order to get the optimum condition (no slump loss and segregation),
as was the case with the existing superplasticizers. This sensitivity was over-
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-LYB 74 D Obb27i.17 0520343 673 D
come by the use of silica fume. The replacement of cement by silica fume has
an advantage of widening the margin of dosages without a significant change in
the consistency of concrete and thus the superplasticizers can be easily used in
practice
3. Zeta potential of cement particles of cement paste with superplasticizers
C E , C E E and C/CEE determined respectively, increased with time up to 90
minutes after mixing. This explains the high slump retention capacity exhibited
by concrete with these superplasticizers.
4. With the use of the new superplasticizers, setting is delayed from 6 to 8
hours compared to that of plain concrete, and also the rate of hydration with
these superplasticizers was slightly lower at an early age.
References
1. Ei-ichi TAZAWA, Shingo MIYAZAWA and Tetsurou KASAI :

Shrinkage due to chemical reaction of cement and intrinsic voids in hardened
cement paste, CAJ REVIEW, 40, pp.74-77, 1986.
2. Ei-ichi TAZAWA, Tetsurou KASAI and Tsutomu KASHIWAGI :
Double mixing effect and rheology of cement paste, CAJ REVIEW, 42,
pp.128- 131, 1988.
3. Ei-ichi TAZAWA : Effects of local defects on compressive strength of
concrete, International symposium on admixtures for concrete, Barcelona,
1990.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-14ô 94 0662949 0 5 2 0 3 4 2 5 0 T
374 Tazawa et al
TABLE 1 - SUPERPLASTICIZERS USED
Classification of Component Part (Role)

Superplasticizer
. -
C Sulfonic acid polymer Methamylic acid Improving fluidity
E Sulfonic acid polymer Methacrylic acid ester Maitaining slump
CE Sulfonic acid polymer Methaqlic acid Improving fluidity
Methacrvlic acid ester
W Sulfonic acid polymer Mixture of C and E Improving fluidity
and maintaining
slump
cE/E Sulfonic acid polymer Mixture of CE and E hPmVing fluidity
and maintaining
slump
C/cE/E Sulfonic acid polymer Mixture of C, CE and E Improving fluidity
and maintaining
slump
A Commercial NaDhthalene sulfonate ImProving fluidity
superplasticizer and maintaining
siumn
r .
B Commercial Amino sulfonic acid Improving fluidity
superplasticizer and maintaining
slump
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 2 - MIXTURE PROPORTION OF CONCRETES
Dosage Water-cement Sand-ratio Unit Weight (kghP)

(%wt.of Cement) ratio W/C(%) (%vol.) W C S G
C X 0.30
CX0.35
50 43 180 360 736 1013
CX 0.6
A
C k 0.7
c c >: 0.48
W-1 c X 0.5
c/E-2 C X 0.6 30 40 160 533 648 1011
c/cE/E C X 0.6
B CX 0.5
A C 1 SP-148 7 4 W Ob62749 0 5 2 0 3 4 3 44b W
TABLE 3 - COMPRESSIVE STRENGTH OF CONCRETE WITH SUPERPLASTICIZERS

(SPECIMEN: CYLINDER @ 10 x 20 cm, W/C = 30 PERCENT)
Compressive strength (k@anz)

Age NO.1 N0.2 N0.3 AVERAGE
CE 7 302 338 326 322.0
28 409 461 461 443.7
A(MT2000wH) 7 483 522 522 509.0
28 673 650 639 654.0
C 7 586 522 590 566.0
28 544 678 683 635.0
A(MT150) 7 544 540 492 525.3
28 625 601 596 607.3
c/ysF) 7 488 498 520 502.0
28 646 670 670 662.0
A(Mï2000WH(SF)) 7 561 487 571 539.7
28 775 708 766 749.7
TABLE 4 - DRYING SHRINKAGE OF CONCRETES AT TWO W/G (x lo6)
W/C Chemical Admixture 1W 4W 2M 3M 6M

2
CW
A (MT2000S) 130 277 266 438 524
0.50 CE/E (FA) 182 314 439 481 552
CWE (BS) 223 380 450 483 537
3A (MT2OOOS(BS)) 261
CE 161 307 389 413 501
A (MT200OWH) 151 279 350 383 470
C E (SF) 153 236 258 305 364
0.30
A(MT2000WH(SF)) 168 269 313 344 393
C 146 267 349 378 437
MT150 139 268 347 384 457
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 99 m Obb2999 0520L44 382 m
376 Tazawa et al
of water
(1)Placing
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
cement paste
cement paste
Fig. 1-The method for chemical shrinkage of cement paste
I Water content W=170kg/m3

W/C=30%, sJa=40%
20
E I
Wo
I
Dosage of superplasticizer
Fig. Z-ReIationship between dosage and slump, slump flow of concrete
A C 1 SP-148 94 m Obb2949 0 5 2 0 3 4 5 219 m
Water content W=MOkgmi

I W/C=30%, s/a=40% I
Dosage Slump Slumpflow ir 1 I-- n

2 o L l m
aE
)
L
I, --L-._____-
30 W
60 15
Time after mixing (minutes)
-Relationship between slump, slump flow, and time after mixing of concrete
3o 75
Dosage Slump Slumpflow Air
A Cx07%
cm CkO.25% v 3.8%
C E E Ck0.3 % 4.5%
03 W
t
G4
5 6
O
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 4-Relationship between slump, slump flow, and time after mixing of concrete
A C 1 S P - 1 4 8 94 W Obb2949 0520346 155 W
378 Tazawa et al
Water oontent W=180kg/m3

W/C=50%, s/a=43%
Time (hours)
Fig. %Bleeding of concrete as a function of superplasticizer type
CR-1 Cx0.6% O
CKEE M . 6 % a
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Time (minutes)
Fig. &Effect of superplasticizer type and dosage on setting time
A C 1 SP-34ô 94 9 0662747 0520347 O93
5000rWaler content
I
b
u
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
e% 3000
.-vi
vi
e t I
3 2000
.C
+
g 1 I sil i '
L1 ,I J
O.
& 1000
Initial set
Time (minutes)
Fig. 7-Effect of superplasticizer type and dosage on setting time
W/C=0.5, Ordinary portland cement
25m
:A(MT2000S,Dosage 0.4%),5 "c
Fig. "The effect of temperature on the fluidity of concrete with superplasticizers
A C 1 SP-LYB 99 = 0662949 0520348 T 2 B
380 Tazawa et al
25 I 1 I I I I I I l
W/C=0.5, Moderate heat portland cement
.. *
O CEE (Dosage 0.21%)30°C
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O cE/E (Dosage 0.21%) 5°C
w A(MT2OOOS, Dosage 0.4%)30°C
O A(MT200C6, Dosage 0.4%) 5°C
Time (minutes)
Fig. %The effect of moderate heat cement on the fluidity of concrete with the superplasticizer
A A(MT2000S.Dosage 0.4%).OPC+BS
0PC:Ordinary Portland Cement

BS :Blastfurnace Slag
O
Time (minutes)
Fig. 10-The effect of blending components on the fluidity of concrete with the superplasticizer
A C 1 SP-148 94 = 0 6 6 2 9 4 9 0520349 9b4
O 90
Time (minutes)
o'l
c.
.i(
Y
8
O
a
' -2
I
Plain L I
Fig. 11-Zeta potential of cement particles with and without superplasticizer
90 --```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Time (minutes)
?i
w
E -10
Fig. 12-Zeta potential of cement particles with and without superplasticizer
A C 1 SP-148 94 Obb2ïi.iï 0520350 b ô b =
382 Tazawa et al
Time of hydration
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
WlC=50%
Mark
V':CE/E,
0:A Cx0.35%
0:Plain
Fig. 1 A h e m i t a l shrinkage of cement paste
A C 1 SP-148 9 4 m 0662949 0520151 512
Superplastidzen and Chemical Admixtures 383
s/a=40%Si/(C+Si)=O.l
Fig. l!i-ERect of silica fume on slump and slump flow of concrete
Watercontent 160kg
WIC=30% Silica hime
da*% Watercontent 160kg
Dosage Mad< WlC=30%
1,000 C l C m CxO.6% a sJa=4û%
C CxO.48% b . ,
Si/~C+Sl)=O.i Watercontent ISOkg
CE-1 CxOS% c Dosage Mat WIC=SO%
B Cx0.58 d OE-1 C x l . 0 8 i sJa=43%
U 5 2 Cxû.6% e h g e Mad<
A C&.6% f
A CxO.78
i
I n CEE Cx0.88
,Compmsive stmgth
500 at 28day's
ÍÍ Compressivestmgth
/ at 7day's
'
i/ Compressive sttngth
at 3day's
O
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
a b c d e i f g h
Fig. ló-compressive strength of concrete
A C 1 SP-148 94 O h b 2 9 g 9 0520152 4 5 9
SP 148-22
Performance Evaluation of Set

Accelerators for ConCrete
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
by AA. Jeknavorian, N.S. Berke, and D.F. Shen
Svnopsis; Among the requirements for Type C and E accelerators in the ASTM C494 Specification
on Admixtures, limits for set and strength performance are stipulated for concrete mixed and cured
at 23°C (73OF). However, set accelerating admixtures are predominantly used for cold weather
concreting operations where they can significantly increase the rate of early strength development
at low temperatures, thus reducing the curing and protection periods necessaty to achieve
specified strengths.
This paper discusses a laboratory program designed to evaluate the effectiveness of various set
accelerating admixture fomulations. The scope of the program includes measuring set, strength,
and air entraining performance over a wide range of mix and curing conditions, and selection of
cementitious materials. Furthermore, considering that a wide range of soluble inorganic salts,
used over a relatively large dosage range, have been shown to accelerate the setting and early
hardening of portland cement, a protocol for evaluating the corrosive potential of accelerator
formulations containingthese salts is also discussed.
Kewords: Acceleratina agents; admixtures; air entraining agents; cold weather construction;
corrosion; inorganic salts; setting (hardening); strength

A C 1 SP-148 94 =
066294'7 0520153 395
386 Jeknavorian, Berke, and Shen
AC1 member, Ara A. Jeknavorian is a Research Manager, W.R. Grace & Co.-Conn., Cambridge, MA.
AC1 member, Neil Berke is a Research Manager, W.R. Grace & Co:Conn., Cambridge, MA.
Din Feng Shen is a Research Chemist, W.R. Grace & Co.-Conn., Cambridge, MA.
INTRODUCTION
The use of accelerating admixtures in concrete provides the opportunity for decreasing
construction costs, especially in cold weather by reducing requirementsfor protection and curing,
shortening the time for the reuse of formwork, and allowing an eariier start for surface finishing.
These benefits result from the admixture's ability to accelerate the hydration reactons associated
with portland cement, which in turn, shorten the time of setting and increase the rate of strength
gain of the treated concrete.
According to AC1 212.3R1 compounds known to accelerate the set and strength performance of
portland cement concrete can include the following soluble inorganic salts and organic
compounds: chlorides, bromides, fluorides, carbonates, nitrates, thiosulfates, silicates, nitrites,
aluminates, formates, acetates, propionates, butyrates, and triethanolamine. Among the chemicals
listed, calcium chloride is considered the most common accelerator on the bases of its economic
use in concrete, and its effectiveness on an equivalent weight basis versus other accelerating
compounds. However, because of the tendency for chlorides to promote corrosion of reinforcing
steel embedded in concrete, the American Concrete Institute (ACI) has recommended that either
no or very low levels of calcium chloride be allowed in reinforced concrete that will be exposed to
moisture and chlorides.1g2
The tendency of chloride ions in concrete to increase the potential for corrosion of embedded
metals has necessitated the development of non-chloride, nonarrosive accelerating admixtures.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Several chloride-free accelerators, whose primary component consists of either calcium nitrite,
calcium nitrate, or calcium formate are commercially available. Even though certain formulations
containing alkali nitrate, nitrite, and thiosulfate have also been recently introduced, wide spread
use is doubtful because of the adverse effects of added alkali with regards to the potential for
expansive alkali-aggregate reactions. Performance specifications for two types of accelerating
admixtures are described in the ASTM C494 'Standard Specificationsfor Chemical Admixtures for
C~ncrete.'~The requirements for each type are listed in Table 1, and are based on the
performance of reference and test concrete mixes prepared and cured at 23.01 1.7"C (731 3°F).
However, the effect of accelerators on set and strength performance is far more pronounced at
lower concrete temperatures, at which accelerating admixtures are primarily used. For example,
concrete admixed with 2.0% calcium formate by weight of cement was reportedto reduce the initial
set time of concrete cured at 4.4% ( 4 O O F ) by approximately 3.4 hours versus 1.7 hours for the
same concrete cured at 21OC QOOF).~
A wide range of test protocols have been used to evaluate the effectiveness of set accelerators.
The selection of concrete (mixing) and ambient (curing) temperaturesbelow
23" C (73°F) have included the following mixturdcure temperature combinations: 22/22"C (72172'
F), 20/5"C (W41°F),20/-5"C (68i23'F), lO/lO"C (5O/5O0F), 41-7"C (39/19"F), and 201-20"C (681-
4 O F ) 5f'~7~*~9.
Furthermore, some of the studies have also included tests to assess the corrosive
effect of the accelerator formulation8~
A C 1 SP-148 94 m 0662949 0520154 2 2 1
TEST PROGRAM
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A portion of the laboratory test program used in our lab to evaluate the performance of various
non-chloride set accelerating formulations is outlined in Table 2. In addition to the requirements
described in ASTM C494 (mixturehre temperatures of 23/23"C (73/7'3"F)), testing is also
conducted at three other mixturekuring temperature conditions: 221-6.7OC (72/2OoF),15,6/15,6"C
(60/60"F), and 15.614.4"C (60140°F). The 15.6"C (60°F) and 4.4"C (40°F) mixturekure
temperatures were chosen as they CONeSpond to the general temperature limits reported in Table
3.1 of AC1 306R-88on Cold Weather Concretinglo. A minimum of three cements and two fly
ashes (class C and class F) were used to examine response over a range of compositions.
An important performance criteria not included in the ASTM C494 specification involves
determining the effect, if any, that non-chloride accelerators could have on the corrosion of steel
reinforcement. One study on various compounds used in accelerator formulations found that
corrosion may be promoted by thiocyanates at dosages above 0.75% by mass of cementt1, while
nitrites have been establishedas corrosion i n h i b i t 0 r s . l ~ ~Considering
~ 3 ~ ~ ~ that little information is
available on the effect of composition and dosage of mixtures formulated with two or more
compounds, especially non-chloridesalts, accelerated corrosiontests such as cyclic polarization,a
potentiostatic polarization method using mortar cylinders with embedded steel, and a steel coupon
immersion procedure were included in our accelerator evaluation program. Each test was
performedwith and without admixed chlorides to better understand how corrosive properties could
be affected by the ingress of chloride ion. The tests without chloride would represent exposures to
on-chlorideenvironments, whereas the admixed chloride results would be applicable to deicing salt
and marine exposures. If an accelerator dosage significantly increased corrosion activity relative to
the control concrete in the presence of chloride, it would suggest that this dosage should not be
used in environments where the possibility of chlorideexposure exists.
Cyclic polarization testing indicates the pitting tendency of steel in a particular environment and
has been used by several r e s e a r c h e r s . l 5 1 ~ ~ ,In~ this
~ , ~test
~ a potentiostat is used to vary the
potential in the noble direction (more positive) of the steel relative to a reference electrode and
measuring the current generated. At a breakdown potential, Eb, there will be a significant
increase in current. When the current exceeds 200 uAlcm2 the scan direction is reversed and the
protection potential, Epl is determined at the point where the reverse current curve intersects the
passive current in the increasing potential region of the cume. In chloride free calcium hydroxide
solutions, the Eb and Ep values are essentially equal and above 500 mV vs a saturated calomel
electrode (SCE). In general, if Ep is more positive than -280 mV vs SCE and Eb is more positive
than 200 mV vs SCE pitting is unlikely in concrete. In this program, a scan rate of 5 mVls was
utilized in a calcium hydroxide saturated solution containingthe admixtures with and without added
sodium chloride.
The constant potential polarizations, at +260 mV vs. SCE for 24 hours, were carried out in a mortar
produced at a 0.5 w/c using an ASTM Type 5 cement, C$c5%. This test is based u on a method
used in Europe and is used to determine the harmlessness of chemical admixtures. It has also pi
been shown that this method provides informationas to whether or not admixtures are beneficial in
the presenceof chloride by admixing chloride into the mortar.lg In general, an admixture is
consideredto be harmless if the average current over 24 hours is within a factor of 2 of the control
specimen or under 1 uAlcm*.
A C 1 SP-148 9 4 m 0662949 0520355 L b 8
Coupon mass loss tests were also conducted using saturated calcium hydroxide solutions and
various addition rates of the admixtures withhrithout sodium chloride. These tests give an
indication of corrosivity without applying a current. In general, the admixtures are considered to be
harmless in the environment if the mass loss and pitting are similar (not more than 2 x the control
values) to the control specimens.
Performanceof Non-Chloride Accelerators
Using the test program outlined in Table 2, the performance of two commercially available non-
chloride accelerators was investigated. One accelerator, coded CNNOP, is a blend of several
compounds of which the major inorganic salts are calcium nitrite and calcium nitrate. The second
accelerator, coded CNSCN, is also a multicomponent blend, whose primary inorganic salts are
calcium nitrate and sodium thiocyanate.
The chemical composition of the three cements and two fly ashes used in this study are shown in
Table 3 and Table 4, respectively. The gradations for the fine and coarse aggregates conformed
to the requirements of ASTM C a . The coarse aggregate was a crushed granite with a nominal
size of 9.5 to 19mm (316 to 314 inch), and the fine aggregate was a natural washed sand with a
fineness modulus of 2.76.
The concrete for all the batches was mixed and sampled according to ASTM C192, except for the
temperatures at which the materials were conditioned, mixed, and cured. The mix designs used
for the portland cement and cementlfly ash concrete mixes are presentedin Table 5.
RESULTS
Concrete Performance
Results of the 22OC (72°F) tests which correspond to the temperature requirement in ASTM C494
indicate that accelerator CNN02 met the setting time and strength specification for two out of the
three cements (Table 6). Accelerator CNSCN exhibited satisfactory set performance and 26-day
compressive strengths; however, 3-day strengths were below the 125% of control with all three
cements. Among the three cements, the generally poorer performance with cement #133 can not
be readily attributed to any particular chemical or physical property.
The performace of lhe two accelerators at a mixturelcure temperature of 15.4/4.4"C (6014OOF) is

shown in Table 7. Though the addition rates are higher than the dosages used for the 22°C (72"
F) cured concrete, both accelerators provided a significantly greater reduction in set time and
improvement in the 3-day strength. Accelerarator CNNOP generally exhibited a better dosage
responsewith regard to set performance. Interestingly, cement #i%, which exhibited little
response to the accelerators at 22°C (72"F), performed well at the higher dosages and lower
temperatures.
Results of the 22/-6.7"C (72/20"F) mixturelcure temperature tests are reported in Table 8. This
experiment was designed to evaluate each admixture's capability of promoting adequate set
performance and strength development prior to the onset of freezing. The function of so-called
antifreeze admixtures is to depress the freezing water and acceleratethe cement hydratio process,
which, in turn, allows the cement to continue hydrating at a low temperature, and subsequently
reduce the amount of freezable water2O The results indicate that when the acceleratorsare used
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4B 7 4 Ob62949 0520156 O T 4
at sufficient dosage, the initial set time of the -6.7% (20°F) cured concrete can actually be less
than concrete cured at 22°C (72°F). The 3 and 7-day strength results demonstrate that some
cement hydration and associated strength development is occurring despite the sub-freezing
curing. When the concrete was allowed to cure at 22°C (72°C) after 7-days, strengths for the
accelerated concrete were comparable to the control concrete cured at 22°C (72°F). The overall
performance of the two admixtures indicated that accelerator CNN02 provided set times about 1.3
hours shorter than CNSCN. 28-day strengths were also more favorable with CNN02 by about 4
MPa (600 psi).
The effect of the two accelerators in fly ash concrete was evaluated because of the tendency of
certain concretes in which the cement has been partially replaced by fly ash to exhibit slower set
times and lower early strengths versus concrete made with the cement alone. Accelerator CCSCN
and CNNO2 were each tested with cements #O51 and #143, which were each blended with a class
F and class C fly ash at a 75/25 cementíash ratio. The complete mix design is shown i l Table 5.
The concrete batches were mixed at 22°C (72"C), and cured at 155°C (60°F) in the case of an
accelerator dosage of 1960 mVlOO kg (30 odcwt) dosage, and 4.4% (40°F) for the 3910 mü100
kg (60 odcwt) dosage.
The results reported in Tables 9 and 10 indicate that the 1960 mWOO kg (30 odcwt) dosage is
more than sufficient to reduce the set time of the 155°C (60°F) cured concrete below that of the
22°C (72°F) control mix. However, in the case of the concrete mixes cured at 4.4% (40°F) and
dosed with 3910 (60 odcwt) accelerator, the set times lagged behind those of the control.
Accelerator CNNOL exhibited slightly better set performance with the class C ash mixes, while
results with the class F ash concrete was mixed. With regard to strength development, accelerator
CNNO2 at a 1960 m1/100 kg (30 odcwt) dosage, nearly provided comparable strength gain to that
of the control mix cured at 22°C (72°F). This was also the case for the 28-day strengths produced
by the concretes batched with the class F ash, which was cured at 4.4"C (40°F) and dosed with
3910 ml1100 kg (60 odcwt) of accelerator. These results are consistent with earlier findings
regarding the strength enhancing performance of nitrites5 Strengths with accelerator CNSCN
consistently lagged behind those of the control mix despite the use of sodium thiocyanate, an
effective strength enhancer for portland cement concrete.
The effect of Accelerator CNN02 dosage on the quality of air-void parameters in air entrained
concrete was evaluated as an indication of freeze-thaw durability. Four concrete mixes were
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
prepared using a Vinsol Resin based air entraining agent. Dosages oí CNNO2 were O, 980 (15),
3260 (50),and 6520 (100) mlkg (odcwt). Three additional air entrained mixes were also prepared
with a NSFC-based high range water reducing admixture. Mix designs for the normal and
superplasticized concrete mixes are shown in Table 11.
The plastic air contents of each mix along with the results of an ASTM C457 hardened air
analysis2' is shown in Table 12. For the normal mixes (1 through 4), a constant dosage of 50
mU100 kg (0.75 odcwt) was used. Other than a slight decrease in air content at a dosage of 6520
mill00 kg (100 odcwt), all the air void parameters were within the recommendations of ASTM
C457-90 (Table 12) to achieve adequate protection against freeze-thaw damage. For the
superplasticized mixes, a slight increase in the air entraining agent addition rate was used with
increasing dosage of CNNOZ, but was most probably unnecessary. Again, all the air-void
parameters met the specifications put forth in ASTM C457. Finally, no significant relationship
between the plastic and hardened air contents was evident with increasing dosage of CNNO2.
A C 1 SP-148 94 =
Obb2949 0520157 T30 =
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Corrosion Studies
Cyclic polarization data are given in Table 13. All of the admixtures did not induce pitting in the
absence of chloride throughout their suggested dosage rate ran e as determined by this test
d
method. In the presence of 0.25 M chloride (equivalent to 3 kg/m or 5 Ibs/yd3), CNSCN did not
appreciably change the protection potential indicating that it has no effect on pitting induced
corrosion, The CNNO2 admixture showed a marked improvement in resistance to pitting in the
presence of choride, even at the lower dosage rate. This is not surprising since one of the
components is calcium nitrite which is an excellent corrosion i n h i b i t ~ r . ~ ~ ~ ~ ~ ~ ~ ~
The potentiostatic data at 260 mV vs SCE in mortars is given in Table 14. In the absence of
admixed chlorides, the CNNO2 admixture was noncorrosive at all dosage rates. The CNSCN
admixture was noncorrosive at 980 mVlOO kg (15 ozlcwt), but corrosive at a 5220 mV100 kg (80
odcwt) addition rate. Thus, if used under severe conditions at high dosage rates, there is a
possibility that CNSCN might be corrosive.
The addition of only 3 kg/m3 (5 ibs/yd3) of sodium chloride, or only 1.8 kglm3 (3 ibs/yd3) of
chloride was corrosive as the control specimens had an average current density of 264 uNcm2. In
this case, both CNNO2 additions 980 and 5220 mll100 kg (15 and 80 odcwt) significantly lowered
the currents to values almost equivalentto those without chloride. However, these currents are not
as low as those obtained with calcium nitrite, which in another study produced currents equivalent
to and under that of a control concrete without chloride at a 3 kglm3 chloride level at approximately
4500 mV 100 kg (70 o d ~ w t ) . ~ ~
The CNSCN admixture treated mortars with chloride showed severe corrosion at both of the
dosages tested. At 980 mV100 kg (15 odcwt), CNSCN showed some minor improvement;
however, the current density of 71 uNcm2 was two orders of magnitude over a control in the
absence of chloride. The higher dosage of 5220 mU100 kg (80 odcwt) was detrimental and had a
current over three orders of magnitude beyond a control without chloride. This is not surprising
considering that the admixture at this dosage rate without chloride exhibited corrosive behavior.
The coupon test results are given in Table 15 and show that both admixtures were noncorrosive in
the absen'ce of chloride. As in the potentiostatic tests, the CNN02 specimens showed a benefit in
the presence of chloride. The CNSCN specimens were corrosive at both 980 and 5220 mIl100 kg
(15 and 80 oz/cwt). These tests are much less severe than the cyclic polarization tests and agree
with the result that CNSCN admixture does not prevent corrosion in the presence of chloride.
Since the chloride control in this test had not yet shown severe corrosion behavior these tests
support the potentiostatic mortar tests in indicating that the CNSCN admixture might be corrosive
in the presenceof chloride.
The corrosion tests indicate that high dosage rates of CNSCN 5220 mV100 kg (80 odcwt) or
higher could be detrimental even in the absence of chloride. This result might be due to the
increased NaSCN content at the higher dosage rates.l
A C 1 SP-148 94 = 0bb2949 0520158 977
Conclusion
The test program described in this study demonstrated the advantages that non-chloride
admixtures can provide over a wide range of mixicure temperatures. The comparative evaluation
of two commercially available set accelerators in concrete mixes containing different cements and
fly ashes indicates the need to test with job materials in order to achieve the most cost effective
performance.
This paper also illustrated the importance of conducting corrosion testing which can be used to
predict long term behavior of steel in concrete. Significantly different conclusions concerning how
non-chloride accelerators can promote corrosion are possible when testing is performed in the
presence of chlorides. In particular, it was shown that the CNNOP accelerating admixture was
non-corrosiveat all usable addition rates, and offered some corrosion inhibition in the presence of
chloride, whereas, the CNSCN accelerating admixture might induce corrosion at high addition
rates, even in the absence of chloride. In the presence of chloride, low addition rates of CNSCN
were shown to have only a minor effect on the corrosion, while high addition rates actually
acceleratedcorrosion.
Based upon the above work, future evaluations of accelerating admixtures should include testing
with different cements and pozzolanic additions as well as at lower temperatures. If the admixture
is to be used with reinforcing steel, then corrosion tests similar to those performed here should be
conducted.
Acknowledgement
The authors would like to express their appreciationto the concrete crew who was responsible for
generating the concrete test data necessary for this paper.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-146 94 m 0 6 6 2 9 4 9 0520159 803 =
REFERENCES:
1. AC1 Cornmittee 212, 'Chemical Admixtures for Concrete," (AC1212,3R3-93)American

Concrete Institute, Detroit, 1993.
2. AC1 Committee 318, 'Building Code Requirements for Reinforced Concrete", (AC1
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
318, R-35) American Concrete Institute, Detroit, 1993.
3. ASTM C494, 'Standard Specificationfor Chemical Admixtures for Concrete,' Annual

Book of ASTM Standards, Vol. 04.02.
4. Dodson, M . "Concrete Admixtures', Van Nostrand Reinhold, New York, p. 89, 1990.
5. Rear, K.B. and Chin, D. "Non-ChlorideAcceleratingAdmixtures for Early

Compressive Strength Gain', Concrete lntemational, Vol. 12, No. 10, 1990,
6. Koronen, C. J., Cortez, E. R., and Charest, B. A., "Strength Development of Concrete
Cured at Low Temperature", Concrete International,Vol. 14, No. 12, 1992, pp. 34-39.
7. Sakai, K., Watanabe, H., Nomachi, H., and Hamabe, K, "AntifreezeAdmixture

Development in Japan', Concrete International,Vol. 13, No. 3, 1991, pp. 26-30.
8. Brook, J. W., Factor, D. F., Kinney, F. D., and Sarkar, A. K., 'Cold Weather
Admixture," Concrefe international, Vol., IO, No. 10,1988, pp. 44-49.
9. McCurnch, L. H., Hardman, M. P. and Lammiman, S.A., "Chloridefree Accelerators",

Concrete,Vol. 13, No. 3, March 1979, pp. 29-32.
1o. AC1 Committee 306, 'Cold Weather Concreting", (AC1 306R-88),ACI, Detroit,
1993, p. 306R-3.
11. Manns, W., and Eickler, W. R., 'The Corrosion-PromotingAction of Concrete

Admixtures Containing Thiocyanate," BetonWerk & Fertigteil-Technik (Wiesbaden),
V. 48, NO.3,1982, pp. 154-162.
12. Berke, N. S.,"The Effects of Calcium Nitrite and Mix Design on the Corrosion
Resistanceof Steel in Concrete (Part 2, Long-TermResults)," Proceedings,
Corrosioní87 Symposium on Corrosion of Metals in Concrete, National Association
of Corrosion Engineers, Houston, 1987, pp, 134-144.
13. Berke, N. S., Pfeifer, D. W., and Weil, T. G., 'Protection Against Chloride-Induced
Corrosion,' Concrete International: Design and Construction,V. IO,No. I2, Dec. 1988,
pp. 45-55.
14. Berke, N. S., and Roberts, L. R., 'Use of Concrete Admixtures to Provide Long- Term
Durability from Steel Corrosion," Supephsficizersand Ofher Chemical
Admixtore in Concrete,SP-119, American Concrete Institute, Detroit, 1989,
pp.383-403.
A C 1 SP-/4ô 94 =
0662949 0520360 525
15. Page, C. L. and Treadaway, K.W.T., 'Aspects of the Electrochemistryof Steel in

Concrete,' Nature, Vol. 297, pp. 109-115, 1982.
16. Berke, N. S.,"The Use of Anodic Polarization to Determine the Effectiveness of

Calcium Nitrite as an Anodic Inhibitor," Corrosion Effects of Stray Currents and the
Techniques for Evaluating Corrosion of Rebars in Concrete, ASTM STP 906, V.
Chaker, Ed., American Society for Testing and Materials, Philadelphia, 1988,
pp. 78-91.
17. Rechberger, P., "Elektrochemische Modellversuche Zur Frage der

Chloridinduzierten Betonstahl Korrosion,' Cement-Kalk-Gips,Vol. 36,
pp. 582-590,1983.
18. Berke, N. S.and Hicks, M. C., "Electrochemical Methods of Determining the

Corrosivity of Steel in Concrete,' ASTM 25th Annivercary Symposium, STP 1000,
R. Baboian and S.Dean Eds., ASTM, Philadelphia, pp. 425440,1990.
19. Berke, N. S.,Hicks, M. C.,Hoopes, R. J., and Tourney, P. G., 'Evaluation of

Laboratory Techniques Used to Evaluate Long-Term Durability of Steel Reinforced
Concrete Exposed To Chloride Ingres,' To be presented at the third CANMETIACI
International Conference on Concrete Durability, May 1994,
Nice, France, M. Malhotra, ed.,
20. Korhonen, C. J. and Cortez, E. R., 'Antifreeze Admixtures for Cold Weather
Concreting", Concrete International, Vol. 13, No. 3,1991, pp. 38-41.
21. ASTM C457 'Standard Test Method for Microscopical Determination of Parameters of
the Air-Void System in Hardened Concrete", Annual Book of ASTM Standards,
Vol. 04.02.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-II48 9 4 9 0662949 052OIIbII 4 6 1
TABLE 1 - ASTM C 494 PERFORMANCE REQUIREMENTS FOR SET ACCELERATING ADMIXTURE
TYPE E
TYPE C WATER REDUCING
PARAMETER ACCELERATOR AND ACCELERATING
WATER CONTENT
(Maximum %of Control) 95
TIME OF SEllING (HR:MIN)

INITIAL AT LEAST 1:OO EARLIER AND
NOT MORE THAN 3 3 0 EARLIER
FINAL AT LEAST l:W EARLIER
COMPRESSIVE STRENGTH,
(Minimum % of control)
3-DAY 125 I 25
7-DAY 100 110
28-DAY I O0 100
6-MONTH 90 1O0
I-YEAR 90 1O0
FLEXURAL STRENGTH,
(Minimum % of control)
3-DAY 110 110

7-DAY 1O0 100
28-DAY 90 100
LENGTHCHANGE
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
(% Increase over control) 0.010 MAX. 0.010 MAX.
RELATIVE DURABILIN FACTOR

(Minimum) ao ao
A C 1 SP-248 94 O662747 0520362 3 T ô W
TABLE 2 - GENERAL LABORATORY PROGRAM FOR EVALUATING SET ACCELERATING ADMIXTURES
CONFORMANCETO ASTM C494 REQUIREMENTS
EFFECTIVENESSAT CONCRETE MIXING AND CURING TEMPERATURES

OF 22/-6.7"C (72/20°F), 15.6./15.6"C (6O/6O0F), AND 15.6/4.4"C (60/40"F).
EFFECTIVENESSWITH CONCRETE CONTAINING FLY ASH
EFFECT ON THE PERFORMANCEOF AIR ENTRAININGAGENTS AND HIGH

RANGE WATER REDUCERS
AIR VOID PARAMETERS (ASTM C457)
SUSCEPTABILITY TO CORROSION
TABLE 3 - CHEMIIAL COMPOSITION OF PORTLAND CEMENTS (PERCENT BY MASS)

CEMENT DESIGNATION
#O51 #413 #133

T.vII -T-l -T-1
Si02 20.84 20.74 21.ô6

A1203 4.45 4.63 4.22
Fe203 3.08 2.48 3.85
Ca0 62.88 63.89 64.80
MgO 3.46 2.87 1.o5
so3 2.91 2.55 2.16
Alkali as Na20 0.58 0.13 0.63
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
LOI 1.54 1.42 0.91
BOGUE COMPOUNDS
c3s 53 57 56
c2s 20 17 20
C3* 7 9 5
C4AF 9 8 12
c3mo3 2.4 3.5 2.3
FINENESS(BLAINE) 3940 3600 3500

CM2/G
A C 1 SP-148 4 4 I0662949 0520163 234
TABLE 4 - CHEMICAL COMPOSITION OF FLY ASHES (PERCENT BY MASS)

CLASS c CLASS F
si02 33.4 56.9

A1203 19.2 23.4
Fe203 5.9 5.6
Ca0 27.0 8.2
M9O 5.5 0.7
so3 1.9 2.1
LOI 0.2 2.5
TABLE 5 - MIX PROPORTIONS

PORTLAND CEMENT PORTLAND CEMENT
CONCRETE FLY ASH CONCRETE
PORTLAND CEMENT 306 K a 3 (517 LEWD3) 246 K a 3 (415 LEWD3)
FLY ASH 80 KW3(134 LBSIyD3)
COARSE AGGREGATE 1041 K a 3 (1750 LEWD3) 1041 K a 3 (1750 LBS/YD3)
FINE AGGREGATE 888 K a 3 (1500 LBS/YD3) 840 K a 3 (1410 LBSND3)
WIC 0.55 --```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
WiCtF 0.55
SLUMP 76-127MM (3-5 IN.) 76-127MM (3-5 IN.)
A C 1 SP-148 94 W 0662749 0.520364 170
TABLE 6 - CONCRETE PERFORMANCE AT 22/22 C (72/72 F) MIXTURE/CURE

TEMPERATURE ADDITION RATE = 1960 MU100 kg (30 ozlwt)
SElliNG TIME, HR.MIN
CEMENT 051 CEMENT 143 CEMENT 133 AVERAGE

INITIAL FINAL INFTIAL FINAL INITIAL FINAL INITIAL FINAL
CONTROL 5.80 8.05 5.45 7.50 4.10 5.75 5.10 7.10
CNSCN 4.45 6.35 3.30 5.10 3.80 5.20 3.85 5.55
CNN02 4.45 6.15 3.35 4.85 3.50 5.30 3.75 5.40
3-DAY COMPRESSIVE STRENGTH. MPA (PClh % OF REFERENCE
CONTROL 17.7 (2570), 100 15.0 (2170), 100 18.1 (2620), 100 16.9 (2450), 100
CNSCN 17.6 (2550), 99 18.2 (2640), 122 19.1 (2770), 106 15.0 (2650), 100
CNNO2 24.9 (3610), 140 19.0 (2750), 127 18.6 (2700), 103 20.8 (3020), 123
28-DAY COMPRESSIVE STRENGTH. MPA (PSI). % OF REFERENCE
CONTROL 36.8 (5340), 100 35.9 (5210), 100 33.3 (4830), 100 35.3 (5130), 100
CNSCN 36.9 (5350), 100 32.5 (4710), 90 34.4 (4990), 103 34.6 (5020), 98
CNNO2 37.0 (5360), 100 34.1 (4940), 95 32.4 (4700), 97 34.5 (SOOO), 98
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 W 0662949 0520165 D O 7
TABLE 7 - CONCRETE PERFORMANCE AT 155/4.4 C (60/40 F) MIXTURE/CURE TEMPERATURE
DOSAGE
MUIOOKG CEMENT 051 CEMENT 143 CEMENT 133 AVERAGE
ADMIXTURE ( O U C W INITIAL FINAL INITIAL FINAL INITIAL FINAL INITIAL FINAL
CONTROL - 13.50 24.75 11.64 18.08 10.05 15.10 11.70 19.30
CNCCN 2930(45) 8.40 13.70 7.10 11.65 5.30 10.10 6.95 11.82
CNN02 2930(45) 8.80 14.05 7.05 10.90 4.75 10.00 6.85 11.65
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
CNSCN 3910(60) 7.95 13.10 6.65 11.80 5.10 9.55 6.55 11.50
CNNO2 3910(60) 6.90 11.65 5.05 9.60 3.75 8.05 5.20 9.75
Q-DAYCOMPRESSIVESTRENGTH. MPA (PSI!. % OF REFERENCE

CONTROL - 6.3 (910), 100 6.5 (940), 100 10.1 (1470), 100 7.6 (lllO), 100
CNSCN 2930 (45) 8.3 (1200), 132 10.0 (1450),154 10.8 (1560), 106 9.7 (1400), 126
CNNO2 2930 (45) 9.05 (1310), 145 11.6 (1680), 179 11.3 (1640), 112 10.6 (1540), 139
CNCCN 3910(60) 8.6 (1250), 138 10.9 (1580), 168 10.8 (1560),106 10.1 (1460), 132
CNN02 3910 (60) 11.5 (1670), 184 12.2 (1770),188 13.7 (1980), 135 12.5 (1810), 163
SB-DAY COMPRESSIVESTRENGTH. MPA !PSI). % OF REFERENCE

CONTROL - 29.6 (4300), 100 31.4 (4550), 100 33.0 (4790), 100 31.3 (4550), 100
CNCCN 2930 (45) 35.2 (5100), 116 32.5 (4720), 101 34.4 (5000), 95 34.1 (4940), 104
CNN02 2930 (45) 34.5 (SOOO), 116 31.6 (4580), 101 31.4 (4550), 95 32.5 (4710), 104
CNCCN 3910 (60) 31.7 (4600),107 33.8 (4900), 108 33.6 (4870), 102 33.0 (4790), 105
CNNOS 3910 (60) 34.8 (5040), 117 31.7 (4600), I01 34.9 (5060), 106 33.8 (4900), 108
A C 1 SP-148 94 D 0662947 0520166 T 4 3
TABLE 8 - CONCRBE PERFORMANCE AT 22/-6.7 C (72/20 F) MIXTURE/CURE TEMPERATURE
DOSAGE
MU100KG CEMENT 051 CEMENT 143 CEMENT 133 AVERAGE
ADMIXTURE- I M U l A L E l M A L ~ E l M A L ~ E l M A L ~ E l M A L
CONTROL(’) 5.70 7.95 5.25 7.75 4.40 625 5.12 7.32
CNSCN 4890(75) 5.95 9.80 7.10 11.10 5.00 9.00 6.02 9.97
CNNOZ 4890 (75) 4.85 8.80 5.05 10.05 4.30 9.90 4.73 9.92
ADMIXTURE !xblwfm-- AVERAGE

CONTROL(^) - 19.9 (2880) 16.1 (2330) 19.2 (2790) 18.4(2670)
CNSCN 4890 3.7 (540) 2.9 (420) 3.8 (550) 3.5 (soo)
(75)
CNN02 4890 3.9 (560) 4.4 (640) 3.9(560) 4.1 (590)

(75)
1-DAY COMPRESSIVE STRENGTH. MPA (PSI)
ADMIXTURE CEMENT 051 CEMENT 143 CEMENT 1% AVERAGE
CONTROL(^) - 26.0 (3TIO) 21.4 (3110) 28.3 (4100) 24.5 (3550)
CNSCN 4890 (75) 8.9(1290) lO.ô(l570) 7.6 (1100) 9.1 (1320)
CNNO2 4890 (75) 7.6 (1100) 11.3 (1640) 8.3 (1210) 9.1 (1320)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 O b b i 9 4 9 05203b7 98T W
TABLE 8 cont. - CONCRETE PERFORMANCE AT 2 2 / 4 7 C (72/20 FI MIXTURE/CURE TEMPERATURE
MIXTUROCURE TEMPERATURE
ACCELERATOR
-
TABLE 9 - ACCELERATOR PERFORMANCE WITH CLASS C FLY ASH CONCRETE
AVERAGE RESULTS mM Wlff ASH
22115.5'C
CNN@
t211.4'C
CNSCN
W4.4"C
C W
DOCAGE, MIA Do KO I p I I p z p 1 p z p 1 p
SETTING ilME RiR:MIM

INITUL 6.13 4.38 525 726 7.10
FINAL 8.13 7.84 8.66 11.18 11.82
P
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
MAY 14.4 (2093) 128(11)60) 13.9(2010) 76(113û) 8.0(11W)

70AY 20.1 (020) 19.11(2870) 21.1 (3060) 14.9(2160) 15.1(2Po)
2BDAY 29.4 (4260) 28.1 (raso) 28.5(4130) 26.8(W) 28.8 (4180)
A C 1 SP-148 94 m 0662949 0520Lbô 8Lb =
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 10 - ACCELERATOR PERFORMANCE WITH CLASS F FLY ASH CONCRETE
NWCURE TENPERANRE
ACCELERATOR
DOUCE. Yu1W KG
15.2(21w) 121 (1750) 17.1 (2480) 9.5(1360) lOd(157û

23.7(wo) al(3210) 25.6(37710) 163(2334) 175(254û
33.4 (rs00) Zû.1(433) 33.0 (4W) 294 (42áO) 31.3 (4540
A C 1 SP-348 9 4 W 0662949 0 5 2 0 3 6 9 7 5 2
TABLE 11 - MIX PROPORTIONS FOR AIR ENTRAINED CONCRETE
PORTLAND CEMENT
PORTLAND CEMENT CONCRETE WITH HIGH
PORTLAND CEMENT 308 KCM’ (520 LBSIYD‘ 308 KG/M3(520 LBS/YD-‘)

COARSE AGGREGATE 1065 KGM’(1800 m) 1065 KGM3 (1800 L B W )
FINE AGGREGATE 735 KGM’ (1240 L w ) 780 KCIM’ (1320 LBWD-’)
WIC O54 OJO
AIR RANGE 5-7 % 57%
SLUMP 127-1155MM (5.0 - 63”) 127-165 MM (5.0 6.5”)-
TABLE 12 - AIR VOID PARAMETERS BY LINEAR TRAVERSE METHOD, ASTM C 457
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
NORMAL SUPERPLASTICIZED
CONCRETE CONCRETE
m M l X 2 kux3Mlx4MIX.s w MIX‘I
DARAVAIR-M, MU100 KG 50 50 m 50 78 91 II7
(OZ/CWT) (0.75) (0.75) (0.75) (0.75) (1.2) (1.4) (1.8)
CNN02, ,MU100 KG - 980 3260 6520 980 3260 6520
(O~CWT) - (15) (50) (100) (15) (SO) (100)
AIR CONTENT (%I, PLASTIC 6.7 6.8 6.6 5.3 6.6 5.3 7.5
AIR CONTENT (%), C457 6.9 7.7 8.0 5.9 7.5 6.2 9.2
AVC.CHORDLENGTH,MM 0.122 0.135 0.140 0.132 0.193 0.140 0.119
SPECIFIC SURGFACE, VMM 325 30.0 28.5 30.2 24.6 1.6 33.3
SPACING FACTOR, MM 0.119 0.117 0.117 0.147 0.165 0.145 0.081
*CO3CRETE COYIAMIhC l9MM (Y4 M.1MAXIMLM SIZED ACGRECAlE
A C 1 SP-348 9 4 Ob62949 0520370 4 7 4
TABLE 13 - CYCLIC POLARIZATION TEST RESULTS (SIAN RATE = 5 d V S )
A. SATURATED Ca!OHIZ
DOSAGE RATE
CONCRETE EQUIPMENT
SAMPLE a mül00 kg OUCWT EP'. mV
CONTROL NO PillING
CNNO2 34 980 15 NO PITTING

I82 5220 80 NO PillING
213 1820 120 NO PITTING
CNSCN 34 980 15 NO PITTING

I82 5220 80 NO PITTING
213 1820 120 NO PITTING
B. 0.25M NaCI & SATURATED Ca !OH12
DOSAGE RATE
CONCRETE
EQUIPMENT
SAMPLE g!L miil O0 kg OUCWT EP'. mV
CONTROL -548
CNNO2 34 980 15 -280

182 5220 80 -105
CNSCN 34 980 15 -527

182 5220 80 -541
* AVERAGE PROTECTIONPOTENTIALS
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 I0662949 0520171 300
TABLE 14 - POTENTIOSTATIC POLARIZATION TEST RESULTS
A. NO NaCI
AVERAGE
ADMIXTURE DOSAGE RATE CURRENT DENSITY
SAMPLE mu1Ol kq QmJYI uA/cmZ
CONTROL 0.54
CNNO2 980 15 0.56

5220 80 0.36
7820 120 0.57
CNSCN 980 15 0.56

5220 80 8.58
B. WITH KGlh43 f5 LBSIYO 3) ....CI
AVERAGE
ADMIXTURE DOSAGE RATE CURRENT DENSITY
SAMPLE mU100 kg OZICWT yAicm2
CONTROL 264
CNNO2 980 15 2
5220 80 3
CNSCN 980 15 71
5220 80 552
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 Ob62949 0520172 2 4 7
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 15 - COUPON TEST RESULTS (AFTER 45 DAYS OF EXPOSURE)
A. SATURATEDCafOHP
DOSAGE RATE
EUIPMENTCONCRETE THICKNESS LOSS'
SAMPLE a mV100 KG OUCWT umlYR
- LOSS.MPY
CONTROL 3.6 0.14
CNN02 34 980 15 3.6 0.14

I82 5220 80 4.1 0.16
CNSCN 34 980 15 3.8 0.15

182 5520 80 4.6 0.18
B.NaC1 ti SATURATED CafOHl2
DOSAGE RATE
EUIPMENT CONCRETE THICKNESS LOSS'
SAMPLE a mU1M)KÇ OUCWT umlYR LOSS.MPY
CONTROL 3.9 0.15
CNN02 34 980 15 2.6 0.10

I82 5220 80 3.3 0.13
CNSCN 34 980 15 28 1.I

182 5220 80 28 1.1
A C 1 SP-148 7 4 W 0bb2747 0520173 183
SP 148-23
Role of Superplasticizers in
Gypsumless Portland Cements
by J. Hrazdira
Synopsis: Gypsumless portland cements iGPC) are

itiorgariic binders, which may be described as system
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
of: qround portland clinker having a specific surface
of 4ú0-500 mZ/kq {Blaine), a superplasticizer with
hydroxyl groups and a hydrolyzable alkali metal salt.
The major feature of GPC is the use of a low
water-to-cement ratio (under O . 301, which is based on
the etrong liquefying action of the superplasticizer
and alkali salt i n the suspension of ground clinker in
the absence of gypsum.
The effect of the dosage of sodium
lignosulfonate in combination with sodium carbonate on
compressive strengths, the setting time and
consistency of cement pastes was studied.
It was found, that properties of pastes strongly
deperids on the dosage and mutual ratio of the
admixtures. The optimum composition of the setting and
liquefying regulator. causes high compressive strengths
past 3 to ‘7 hours from mixing of binders (15-45 MPa).
Kewords: Clinker; compressive strength; consistency tests; pastes; portland cements; setting
(hardening); sodium carbonate; superplosticizers

407
A C 1 SP-148 94 Obb29Y9 0520174 OLT
408 Hrazdira
J. Hrazdira deals with the epecial cements having high

strengths. He also solves some environmental problems
with the help of blended cements.
INTRODUCTION
New cements, developed in recent years, are able

to reach both higher strengths and fracture toughness
than normal portland cement 11, 2, 3 ) . New
developments in the making of very strong cements have
resulted from modifying cement compositions and
manipulating the microstructures (4).
Brunauer and Co-workers developed a cement
system without gypsum (5). Substituting a suitable
setting regulator for gypsum is the way how to improve
properties of binders based on portland clinker.
Gypsumless portland cements (GPC) differ from normal
portland cement by having better rheolcgical
properties oí pastes and mortars under low
water-to-cement ratios, b y having high short and
long-term strengths (compressive strength 15-45 MPa
after 3-7 hours, 40-65 MPa. at 2 4 hours) and low
porosity of hydration products.
This paper describes the results of a study
dealing with the aim to control the course of
hardening and the development of strengths in GPC.
EXPER IMENT AL
Materials
Portland clinker used was obtained from a plant

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
in Bohemia. Its chemical and mineralogical composition

i s given in Table 1. This composition is similar to
ASTM type III. The clinker was crushed, homogenized
and its 0-2 mm fraction was ground in a ball mill in
the presence of a salt of dodecylbenzen- sulfonic
acid, which is known as the efficient grinding aid for
GPC ( 6 ) . The specific surface of the product was 450
mzikg (Blaine).
The superplasticizer used was a sodium
lignosulfonate (BORRESPERSE-NA) of Norwegian origin.
?'he alkali carbonate, N a z c a , was made in Czech
Hepubl i c.
Pastes
All pastes were prepared by mixing the qround

clinker with water solutions containing various
combinations of sodium carbonate (1-3 96 of the weight
of the qround clinker) and lignosulfonate (0.8-1.6 %
oí the weight of the ground clinker). The
A C 1 SP-/L)ô 74 0662949 0520175 T5b
water-to-cement ratio was kept at O. 26.
Methods
The setting time and the time course of

hydration reactions of the pastes were determined by
a conduction calorimeter (7).
Cement paste consistency was measured with
a mini-slump test technique in which fluidity is
related to the area attained by a pat of cement paste
on its release from a slump cone ( 8 ) .
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
The mixing oí pastes was performed by hand for
5 minutes. After mixing, the pastes were trarisfered
into steel molds having cube dimensions of 2 x 2 ~ 2cm.
The molds were then placed in containers at 100 %
humidity for one day. At the desired age ( 3 , 5 , 7
hours, 1, 7, 28 days) the samples were tested for
compressive strengths. After one day samples were
placed in water at 2 0 0 .
The variations of the mini-slump of pastes are

given in Fig.1. 'The results indicate that the higher
concentrations of sodium carbonate in solutions
produce lower fluidity oí pastes. BY increasing the
dosage of lignosulfonate from O. 8 to 1.6 wt. Y& the
fluidity of pastes is increased so that the
niirii-slump goes up.
The variation oí setting times of cement pastes
iri the presence of varying concentration of sodium
carbonate and sodium lignosulfonate is given in
Fig.2. It was found that the higher dosages oí sodium
carbonate in pastes, the shorter were the setting
times obtained, except the composition of 1.2 wt. % of
lignosulfonate and 2.0 wt. % oí sodium carbonate. The
higher concentration of lignosulfonate caused longer
setting times.
The time course oí hydration reactions of GPC
with various amounts of sodium carbonate and
lignosulfonate is shown in Fig. 3 . 4, 5. The
heat-evolution profiles of GPC consists of two
exothermic peaks separated by two induction periods
( 7 ) . It is clear that the area of peaks i s strongly
affected by the dosage of lignosulfonate. The higher
cincentration of lignosulfonate retards hydration
reactions of GPC so that the area of the first
exothermic peak falls.
The compressive strengths of pastes are given in
Table 2. It was observed that the higher dosage of
sodium carbonate improves short-term compressive
I strengths of GPC ( 4 8 . 5 MPa after 7 hours of hydration
for 3 wt. % of sodium carbonate and 0.8 wt. % of
lignosulfonate). The trend is strong especially within
A C 1 SP-148 94 0662947 0520376 992
410 Hrazdira
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
24 h o u r s of h y d r a t i o n . The d i f f e r e n c e s a r e minimal
a f t e r 28 d a y s oi h y d r a t i o n a n d b i n d e r s are e q u i v a l e n t
as r e g a r d s t h e i r l o n g - t e r m s t r e n g t h s .
T o e x p l a i n .the p r e s e n t d a t a t h e p o s s i b i l i t y o f
ttie e x i s t e n c e of a h i g h l y a n i o n i c complex b e t w e e n
lignosulfonate a n d c a r b o n a t e w a s a c c e p t e d (Y). The
complex is i o r m e d a n d a d s o r b e d o n t h e cement p a r t i c l e
surface, preierentially on t h e h y d r a t e d C j A surface.
'The c o m b i n a t i o n of a d m i x t u r s e mentioned above a f f e c t e
FH a n d a b u i f e r c a p a c i t y o f l i q u i d p h a s e . T h e i r m u t u a l
r a t i o probably iniluences the s t a b i l i t y and ttie
rfficiency of the c o m p l e x arid h e n c e t h e h y d r a t i o n
i.eactiuns. The h i g h e r c o n c e n t r a t i o n s of sodium
c a r b o n a t e probably d e c r e a s e t h e eff i c i e n c v o f ttie
complex d e s c r i b e d as art e f i e c t - i v e d i s p e r s a n t . Hence i t
f o l l o w s , that, t h e i l u i c i i t y o f pastes is d e c r e a s e d . I n
a d d i t i o i n , t h e complex is n o t s t a b l e for a l o n g t i m e s o
that a 1-etardinq effect is w e a k e r a n d s h o r t - t e r m
s t r e n g t h s are h i g h e r compared t o l o w e r c o r i c e r i t r a t i o n s
of earlotiate,
CONCLUSIONS
F'roni t h e r e s u l t s d e s c r i b e d i n t h i s paper it w a s
found t h a t :
1 . R e t a r d i n g and l i q u e f y i n g e i f e c t of coiiibiiiations uí
sodium c a r b o n a t e and l i g n o s u l f o n a t e i n GPC s t r o n q l y
d e p e n d s on the dosage and mutual r a t i u o f tlie
adrni x t u t - e s .
2 . IL is p o s s i b l e to control the p r o p e r t i e s of. CPC

cements t y optiiniziny the cornpusition of
requlator.
3 . H i s h c o n c e n t r a t i o n s of sodium c a r b o n a t e decrease
the i l u i d i t y and improve s h o r t - t e r m c o m p r e s s i v e
strenqLlis oí GPC p a s t e s . ' ï h e s a t t i i i y times Lecoine
shorter. Hiyh d o s a g e s of 1 i r i i o s u l f undLe retskrd
l i y d r a t i o n r e a c t i o n s and d e c r e a s e t h e h e a t e v o l u t i u n
oi GPC p a s t e s .
4. I t is p o s s i b l e t o u s e t h e s i r 3 n i í i c a n t i n c r e a s e i r i
compressive s t r e n g t h s (15-45 MPa p a s t 3-7 h o u r s
from mixing) ior repairs, siiotcretinq, the
- aiictiorage of b o l t s and o t h e r s p e c i a l works.
A C 1 SP-148 94 0662949 0520377 ô 2 9
REFE HE NCES
1. . C a i . ? ' . ~ s c i i i i i1.0, I < . L.

" P r o d u c t i o n oí High S t r e r i y t h
p d 4._t ,ES,
.. . Mortais and C o n c r e t e s " , Materiais R e s e a r c h
Society S_;vinposia Proceedings, Vol. 42, pp.
1 5 1 - -ltö, I4oveiiiber- 1'984
L Aitcin, P.C., Bedard. C., Plumat. M.. Haddad, G.

" V e t - y tliql-i S t i e n s t h C e m e r i t for V e r y H i g h S t r e i i g t f i
C o n c rei,e", Materials H e s e a r c l i Society S y m p o s i a
k ' r o c e e d i i i q s , V a i . 42, p p . L i U 1 - 2 1 0 , N o v e m b e r 1984
3 . DairrmrI, 8. "Very H i g h Str.erigtti Cement-Based

Materials-- a Prospective", Materials R e s e a l - < t i
Society Syniposia Proceedir~gs, Vol. 42, PF'.
233-243, Nove;.rriGei* i 9 8 4
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
4. B e a u J o i i i , J . J . , E ' e l d m a r i , k.F. " H i y l i - - S t r e n g t h C e m e r i t
Pastes- a C r i t i c a l A p p r a i s a l " , C e m e n t and C o n c r e t e
H e s e a r c h . V o l . 15, P P . 105-116, 1985
' I .bi u r l a u e r , S . , Yudenireund, M . , O d l e r , I. , Skairiy,

J. "kiai-dened Port larid cement i"astes of Low
L'or o s i t y , i . Materials a n d E n p e r i m e n t a l Methods",
C e m e t i l - aiid C : r > i i c r e t e k s e a r c h , Vol. L , P P . 313-3313,
1-472
b. H r a z c l i r a . J. " l h e E f t e c t of G r i n d i n q A i d s on t h e
t , r operC lis of G y p s u m l e s s Lenieiits" , C e r a m i c s , V o l .
3 - 3 , r+. 3U.7- , ì 5 , l 9 w
8. Perericiiio, V.F. , W f i i t ì r i q , b.A . , K a n t r w , D. L. "Water

Iirductiori, Slump L u s s a r i d Enti-airied A i r - V o i d
S y s t eins as l i i i ll_ieiiced by S u p e r p l a s t i c i z e r s " , A C i
S P - G ¿ , S u p e r p l a s t i c i z e r s i n C o n c r e t e , pp. 137-156,
1j ï Y
9. Jawed, I., K l e m m , W . A . , S k a l n y , J . " H y d r a t i o n of

Cement.- Licjtiosul f o r i a t e - A l k a l i Carbonate System",
J o u i i i a l uf Lhe Ame? icari C e r a m i c S o c i e t y , v o l . 6 2 ,
pi,. h 1 -4 4 , î ~ ' / 9
412 Hrazdira
A C 1 SP-148 94 m 0 6 6 2 9 4 9 0520L78 7b5 =
TABLE 1 - CHEMICAL AND MINERALOGICAL COMPOSITION OF PORTLAND CLINKER
Chemical Composition ut. %
Si& 20.44
Ca0 63.41
A l 2 o3 6.40
Fez % 3.02
MsO 3.53
S% O. 63
KZO o. 99
Na? 0 O. 18
Ti& O. 33
MnO o. o9
Free Ca0 O. 52
Loss on Ignition O. 65
Mineralogical wt. %
Composition
Ca s 71.9
cz s 6.2
C4 AF 8.1
Co A 13.5
Free Ca0 O. 3
TABLE 2 - COMPRESSIVE STRENGTH OF PASTES
o. a 5.4 23.3 35.5 55.0 65.5 100.o

0.8 6.3 28.o 42.5 67.5 72.5 104.5
O. 8 9.5 40.O 48.5 67.0 81.0 96.3
1.2 2.5 6.9 32.4 57.3 93.5 100.5
1.2 3. O 12.8 33.5 59.0 97.8 99.o
1.2 4.1 1 19.5 44.8 68.5 90.5 105.O
1.6 <1.0 1 5.5 15. O 50.5 55.5 102.5
1.6 <1.0 a 1 8.8 18.5 37.0 56.5 100.5
1.6 (1.0 ' 11.5 21.0 39.0 52.5 98.8
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662949 0520379 b T 1
300
wt. %
250 -
-0.8 E-NA
E 200 -
U + 1.2 E-NA
5 150-
3
VI
-I -E- 1.6 8-NA
I 100-
E
50 -
OL
Fig. 1-Influence of Na,CO, and lignosulfonate on GPC paste mini-slump
100,
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
E
E
W-
I
I-
80
60
'
- 0.8 B-NA
1.2 B-NA
'
$ 40
20 '
I 1 I I I I
O
O 0.5 1 1.5 2 2.5 3 3.5
No2C03. wt. 9.
Fig. 2-Influence of NalCO, and lignosulfonate on GPC paste setting time
A C 1 SP-148 7 4 0662749 0 5 2 0 3 8 0 3 3 3 =
414 Hrazdira
12
10 wt. %
O 50 1 O0 150 200 250 300

TIME, min.
Fig. &Time course of hydration reactions of GPC pastes with 1 percent of Na2C0,
and various amounts of lignosulfonate
12
+ 1.2 E-NA
1.6 B-NA
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
O 50 1O 0 150 200 250 300
TIME, min.
Fig. "Time course of hydration reactions of GPC pastes with 2 percent of Na,CO,
O wf. %
- 0.8 E-NA
+ 1.2 E-NA
-s- 1.6 8-NA
O 50 1O 0 1 50 200 250 300

TIME, mln.
Fig. &Time course of hydration reactions of GPC pastes with 3 percent of Na,CO,
A C 1 SP-148 94 0662949 0520383 2 5 T =
SP 148-24
Modification of Cement Mortars by

Polymeric Latexes and Their Use
for Repairing Concrete
by G. Mantegana, A.M. Penna, and S. Taîîoni
Svnousis: This paper reports the results of an experimental research

program on polypropylene fiber-reinforced cement mortars. The purpose of the
work was to define the effects produced on a basic mortar by the addition to
the mixture of various synthetic polymers in the form of latexes. The significant
features considered, that characterize a cement mortar formulated for repairing
concrete structures, include workability, low permeability, dimensional
stability, mechanical resistance (compressive and flexural strength), Young's
modulus of elasticity, ductility, toughness and durability. The test results
suggest that optimal characteristics have been obtained (given a minimum
percentage of fibers) with a content of 10%of silica fume by weight of cement
and by using acrylic acid copolymer water dispersion at the rate of 9% of the
solid polymer to cement.For protective use, it is possible to modify the basic
mortar by a styrene-butadiene copolymer water dispersion; however, the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
modification of the basic mortar by using styrene-acrylic polymer latex,
generates a product suitable either for reparing or coating of reinforced concrete
stnictures. The reliability of this last modified mortar as a protective coating
has been also investigated.The most significant results relate to the penetration
of C1- ed S04-- ions, the variation of bond strength between mortar and
concrete, the effectiveness of a thin layer of such a cement composite,
correlated to rate of steel corrosion.Equivalence criteria have been described
among different mortar covers in comparison with the minimum cover
thickness of concrete as prescribed by European codes, by using an
electrochemical test procedure.
Kemrds: Compressive strength; corrosion; m;ductility; durability; flexural strength; latex;

modulus of elasticity; mortars fmoterialb permeability; rePoin; workability
A C 1 SP-148 94 Obb2949 0520182 1 9 6
416 Mantegazza, Penna, and Tattoni
INTRODUCTION
This work synthesizes the results of a large research program (1-3) with
the purpose of defining a repair mortar for normal or prestressed reinforced
concrete structures. The major features of this material are ease of
application, even in difficult conditions, the resistance to mechanical strain,
the durability, the compatibility with the support, the control of the shrinkage
and good aesthetical appearance.
The achievement of one of the above listed features is sometimes
obtained at the detriment of the others. For instance, silica fume gives the
cement mortars good properties in terms of durability, strength and stiffness,
but a low ductility; the adding of polypropylene fibers limits the shrinkage of
the mortar and increases its ductility, but it reduces its workability at the same
water-to-cement ratio as the control mortar; the polymeric latexes improve the
workability and the tenacity of the mortar, but the mechanical resistance is
reduced (4, 5).Based upon the above premises, research has been started to
find out the most effective formulation of a cement mortar for repair
purpose.Also the compatibility of the mortar with the existing support and its
durability, or, better, the capacity of the repair layer to give the structural
element on which it is applied a suitable protection, has been investigated.
SELECTION OF THE TYPE OF REPAIR MORTAR
Experimented Materials
The composites investigated consist of a base mortar, prepared with
ASTM portland cement Type II (1 part), additive fillers and polypropylene
fibers of 10 mm length (0.25 parts, of which 0.008 fibers), sand with a max
size of 3 mm (1.25 parts) and silica fume in varying proportions, replacing
same quantities of the filler.To improve the workability, a superplasticizing
additive based on naftalenesulfonate, was used, which gave the mortar a
base consistency of 50 - 70 mm (measured following italian standard UNI
7044/72).The dosage and the characteristics of the base mortar are shown on
Table 1.
After having established in mortar type C (Table 1) the optimum dosage of
the silica fume to obtain a low permeability and a high mechanical resistance,
this was further modified with different types and dosages of the polymers
listed below (5, 6 ) .
STI.ACR1 copolymer of acrylic acid in anionic aqueous dispersion
ACRI acrylic polymer in aqueous dispersion
STI.BUT styrene-butadienecopolymer in aqueous dispersion
TER.VIN1 vinylic terpolymers obtained by means of copolymeration of various
types of vinylic monomers.
In some mixtures, the W/C was reduced in order to correct the reduction
of resistance given by the polymer.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 M 0 6 6 2 9 4 9 0 5 2 0 1 8 3 O22
Workabilitv and uermeabilitv

The optimum dosage for the polymer has been investigated; a low
dosage keeps the permeability high; on the other hand an excessive
percentage of the polymer reduces the workability of the mortar due to the
beginning of the setting of the polymer itself, which, in turn confers the
mixture a gluing consistency and makes its application difficult. It was
experimentally established that the optimum dosage is approximately 9
% (1). Fig. 1 gives the workability (measured by means of the consistency
test) of the mortar modified with different types of polymers.Fig. 2 shows the
permeability results obtained for the optimum mixtures.
Mechanical strength. Younv's modulus of elasticitv and shrinkage

These parameters are especially significant for the correct application
and identification of a repair mortar, besides being a significant index for the
compatibility with the support on which the mortar will have to be applied.
Table 2 and Fig. 3 , 4 show the most significant data, referred to mortars with
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
optimum dosage of polymers (9% of dry polymer by mass of cement) and

optimum WlC for the mechanical resistance.
With the exception of styrene-acrylic latex, the adding of polymers
reduces the compressive strength and the Young's modulus of elasticity,
although within reasonable limits, flexural strength is generally maintained as
compared to the base mortar. This characteristic, together with the reduction
of the Young's modulus of elasticity, is an index for the increase of
"ductility", by which the capacity of the hardened mortar to follow the
deformations of the support concrete without cracking is intended.
The test results for shrinkage for the various mixtures, is shown in Fig. 5.
The important expansion caused by the curing, has however been observed
during the first part of the setting (1 to 3 days); this helps the action of the
expansive agent, present in the base mortar. After this, the compensation of
the hydraulic shrinkage depends in a significant way on the type of the
polymer used; the highest stability is obtained with the styrene-acrylic and the
ter-vinylic polymers.
Tenacity index
The presence of fibers in the base mortar C confers it a high tenacity;
instead the adding of the polymer latex reduces it (Fig. 6 ) . The anticipated
increased adhesion between the polymer and the matrix seems therefore not
to take place. This could be explained by the fact that the organic materials
generally have a low surface energy, and therefore the adhesion between the
polymer and the polypropylene fibers does not develop completely, as it
happens for steel fibers.
A C 1 SP-146 94 W Obb2949 0.520384 Tb9 W
418 Montegazza, Penno, and Tattoni
Selection of the orximum mortars

The adding of polymer latexes, the percentage of which turned out to be
9%of the dry polymer by mass of cement, induces modifications to the base
mortar,some of which are common to all used polymers, though in different
degrees: a) less permeability, at same W/C; b) better workability, and therefore
possibility to reduce W/C, increase the mechanical resistance and reduce the
permeability (styrene-acrylic and styrene-butadiene); c) better adhesion and
deformability, and good volume stability (styrene-acrylic and ter-vinylic); d)
poor or limited adhesion of the concrete matrix to the poly-propylene
fiber, and reduction of the tenacity index (this negative effect is still more
important for the acrylic polymer); the tenacity index is anyway higher to the
one of the non-modified base mortar, and can be improved reducing the
W/C.From the above investigations it could be established that there is no
ideal polymer for all applications, but it is necessary to select it carefully
according to the characteristics required.
For the applications for structural repair mortars, the most suitable
modification agents appeared to be the acrylic type polymers.
DURABILITY OF THE ADHESION TO THE BASE CONCRETE
Mode of investigation
To evaluate the durability of the mortadbase concrete system, the
penetration capacity of the following substances has been investigated (Fig.
7), because these influence in a negative way the adhesion to the base
concrete, the maintenance of the reinforcing steel and the integrity of the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
compound.water; C1- ions (CaC12 i n aqueous solution 30 g/l, pH=5, following
the italian standard UNI 7928);S04-- ions (Na2S04 in aqueous solution
71.03 gA, pH = 5, following the italian standard UNI 8019). To accelerate
the damaging action, the specimen were submitted to soaking and drying
cycles (Fig. 8).
Materials
Concrete plates 50x40~10cm were made to support the mortars (2).
The surface on which the mortar was to be applied, was treated in order to
obtain an average dimension of surface roughness of 1 mm.The applied
mortars (Table 1, 2) were A (reference), C2 and C5. Based upon the
laboratory results, these found out to be the most appropriate for the repair
and the protection of the concrete in structures to be repaired.
Measuring of the adhesion
The adhesion of the mortar to the base concrete is checked measuring
the pull-out strength of a 50 mm diameter core from the test plate, by means
of an adherence meastiring device (CONTROL 1000 daN).
A C 1 SP-148 94 Obb29Y9 0 5 2 0 1 8 5 9T5
Not only the specific strength value f was considered to evaluate the
adherence to the base concrete, but also the mode of breaking and the
formation of crack (Fig. 9);on the contact surface between mortar and the
base concrete (adhesion fad, Fig. 9b);inside the mortar or inside the support
(tensile strength of the mortar fmt or of the support fct (Fig. 9c, 9d);inside
both materials in different percentages (fmt = fct, Fig. 9e).
The results of the adhesion tests, in terms of mode of breaking, are
shown in Fig. 10, 11, 12, whereas Fig. 13, 14, 15 show the adhesion
strength, when the breaking occurs according to Fig. 9b.
Results
Base mortar A. When curing i n water, the failure occurs mainly at the
mortar level, and secondarily at the surface (adhesion strength). The pull-
out strength does not vary with time (Fig. 13).
When curing i n presence of chloride ions, the mortar-adhesion-
cement system does not change in a significant way, and the failure can
occur also in the concrete. The dispersion of results is due to the variability
of strength of the single components, rather than to the degradation of one of
them. Also when curing in presence of sulfates, the failure can occur in the
three components, at least in the initial stage. The passing of time points out
a higher degradation of the adhesion.
C5 mortar. Under all three conditions of curing (Fig. 14) no significant
alterations can be observed, and the failure occurs more frequently in the
concrete, which is the weakest link in the system. The anomalous tendency in
case of curing in CaC12 must be ascribed to the dispersion of the measures,
more than to a precise physical fact. The failure due to the adhesion stength
occurs in a sporadic and negligible way.
C2 mortar. The failure occurred always in the concrete.
The analysis of the variation i n the pull-out strength in case of curing in
presence of sulfates shows that this phenomenon does not depend on an
alteration of the cement of the support due to a chemical aggression, since
the pull-out strength follows its normal increase with age (Fig. 15).
PENETRATION OF AGGRESSIVE AGENTS
The penetration of the aggressive agents in a mortar, which has been

modified with a polymer of acrylic type, has been measured using analytical
stratigraphy, and the results has been compared with these obtained for a
refence mortar (7).
The results of the penetration of the ions are shown in Fig. 16 and 17; the
reduction in the penetration of the ions can be noticed on the specimen
protected with the modified mortar, when compared to the traditional mortar
due to the differences in the permeability of the two. This effect is more
significant for the C1- ion than for the the S04-- ion.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-I148 9 4 m 0662949 0520386 8 3 3 =
PROTECTION OF THE REINFORCEMENT

After having venfieù the adhesion as a function of time and the resistance
to the penetration by aggressive agents in a modified mortar, the research
was extended to evaluate the efficiency of a thin layer of coating on steel
reinforcement, using the C2 mortar, which is specifically suitable for repairs of
reinforced concrete and prestressed concrete.
Furthermore, the comparison with a traditional concrete coating was
carried out, the thickness and composition of which is in conformity with the
prescriptions of the present technical recommendations (8).
Test method
The experiment was performed on concrete specimens in which steel
bars were placed at different depths (Fig. 18, 19).
These specimens were then exposed to the action of a 3% NaCI
solution according to the following cycle: one week of soaking, followed by
two weeks of drying in air.
The ionic solution, penetrating into the specimen, forms a macro cell in
between the steel bars activating a electrolytic corrosion process, witnessed by
the current flow between the upper bar and the two lower bars. The current
density was measured after the soaking phase with an AMEL mod. 668/RM
electrometer.
As there is a correlation between the speed of corrosion and the current
density which is located in the macro cell with the electrolytic solution (9), the
evaluation of the efficiency of the different protecting coatings (different
materials and thicknesses) has been based on the measure of this last value.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Test elements and materials

The preparation of the specimen was performed in conformity with
the pending standard CEN (10). To simulate the effect of the cover
protection, the specimen were manufactured with two zones (Fig. 18) with
different compounds and different thickness of zone 2 (Table 3).
For zone 1, a poor quality conglomerate was used, similar to that
present in old, degraded structures.
BC (Base concrete):
portland cement 425 ASTM type II 300 kg/m3
aggregate (max. size 10 mm) 1194 kgm3
silica fume 597 kg/m3
water (W/C=0.7) 210 Um3
A C 1 SP-148 94 0662749 0520187 778 M
For zone 2, acting as coating, a concrete compatible with exposition class 4a

(manne environment without frost) or 5a (prestressed concretes in chemically
aggressive environment) /lo/.
DC (Durable concrete):
portland cement 425 ASTM type II 300 k d m 3
aggregate (max. size 10 mm) 1194 kg/m3
silica fume 597 kg/m3
superplasticizing additive 4.5 Vm3
water (W/C=0.55) 165 Vm3
and the C2 mortar (Table 1,2)
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Results
After a transitory period which elapses within the first 20 weeks of
treatment, the value of the current density tends to stabilize, although with
wide oscillations; it is true the specimens under test (Fig. 20, 21, 22).The
current density in the base concrete protected specimens is much higher
than that measured for the same time on the specimens protected with
durable concrete and C2 mortar (series D, E, F and G, H, I) for which the
current density is stabilized at extremely low values, with the passage of time.
Only exception is the bi-component mortar C2 applied with 5 mm
thickness (Fig. 22, senes G ) . This anomalous behavior can be explained
keeping in mind that it is extremely difficult to create a thin and regular
coating mortar layer, and that the importance of the flaw can significantly
reduce the thickness of the layer.
Based upon the correlation between the current density and the rate of
the metal corrosion, expressed in mm/year, the maximum corrosion rate,
deduced from the measurements, related to the thickness, have been
compared for the different compounds, perhaps i n a more significant way
(Fig. 23).
Besides the case of the 5 mm thickness, it can be observed that with a
limited layer of bi-component mortar (10-15 mm) a protective function for
the bars can be obtained which, maintaining the same other conditions, is
comparable to the one offered by a durable concrete applied in thicknesses
ranging from 20 to 40 mm. An equivalence criteria can therefore be
established for the thickness of the different materials to coat the
reinforcements as a function of the actual protection which is offered.
CONCLUDING REMARKS
Based upon the intrinsic characteristics (workability, permeability,
volume stability, strength, deformability and tenacity) it has been made
sure that the optimum dosage of a polypropylene, fiber-reinforced concrete
mortar is of 10% by mass of cement for the silica fume and 9% (by mass of
the dry polymer to the cement) for the polymer latexes.
A C 1 SP-148 94 W O b 6 2 9 4 9 0520188 604 W
For the purpose of repair and protection of concrete and prestressed

concrete structures, the polymers which confer the most valuable
modifications are: acrylic styrene (STLACRI), which increases the strength and
the Young's modulus of elasticity of the base mortar and reduces its
permeability; the result is a material which is suitable for structural repairs,
even in aggressive environments;acrylic (ACRI), which on the other hand
increases the deformability of the base mortar and confers thyxotropic
properties to it. Considering the limited mechanical characteristics, this
product is suitable for leveling or finishing jobs and for the application in
masonry; it is not suitable for the use in aggressive environments.
Compared to traditional mortars, the bi-component mortars showed a
significant maintenance of the adhesive bond, and of the intrinsic strength,
even in the aggressive environment.
This behavior is due to the extremely low permeability of the
composites which interferes significantly with the penetration of the aggressive
agents.
Expressly, for the Ci- ions, the composites can offer an effective
protection to the steel bars which can be present in a structure to be prepared.
The protection effect is better for the styrene-acrylic polymer modified mortar,
because of the extremely low permeability and for the uniformity of the
obtained results.
The capacity of the C2 mortar to protect the steel bars, even if applied at
low thickness (10 to 15 mm), is equal to that of a concrete of very good
quality with a coating thickness of 30 to 40 mm. The use of this type of
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
modified mortar is extremely well suitable to rebuild the cover of

degraded structures, wherever the thickness and the weight of the added
materials is preferably maintained low.
ACKNOWLEDGMENTS
This research program has been partially supported by the financial
contribution of CNR (National Council of Research) and MURST (Ministery
of the University and Scientific and Technological Research).
A C 1 SP-348 9 4 Ob62947 0520389 5 4 0
REFERENCES
1. Mantegazza G., Penna A. M., Tattoni S . , "Malte da npristino fibrorin

forzate: modificazione delle carattenstiche mediante polimeri sintetici",
Studi e Ricerche Vol. 11/89, Politecnico di Milano, 1990.
2. G. Mantegazza, A. M. Penna, S . Tattoni, "Compatibilit e durabilit di
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
compositi cementizi modificati con polimen e fibrorinforzatiper il

ripristino di strutture in c.a.", Studi e Ricerche Vol. 12/90,
Politecnico di Milano, 1991.
3. G . Mantegazza, A. M. Penna, S . Tattoni, "Sull'efficacia dei compositi
cementizi modificati con polimeri impiegati per il ripristino del
copriferro di strutture in c.a.", Proc. Giomate AICAP, Pisa 3-5
giugno, 1993.
4. Mantegazza G., Puccio M., Finzi F., Bolzoni A., "Role of Fly-Ash,
Silica Fume and Superplasticizerin the Modification of the Mechanical
Characteristics and of Durability of Cement Mortars", Proc. Third Int.
Conf. on Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete,,
Trondheim Norway, 1989.
5. Atzeni C., Mantegazza G., Massidda L., Sanna U., "Effect of Polymeric
Latexes on Cement Systems Containing Silica Fume", Proc. AC1 Int. Conf.
on Fly-Ash, Silica Fume Slag & Natural Pozzolans in Concrete, Ottawa
Canada 1989.
6. Ohama Y., Endo M., Miyara M., "Properties of Hybrid Fibre Reinforced
Polymer-Modified Concrete", Proc. of RILEM Sym. on Developments
in Fiber Reinforced Cement abd Concrete , Vol. I, july 1986.
7. Baronio G., Carminati G., Mantegazza G., "Malte da ripristino
stnitturale caratterizzateda elevata durabilit e sistemi di controllo del
l'affidabilit dei coprifeni", Proc. II Convegno CNR - Progetto
Finalïzzato Edilizia, Sottoprogetto 3, Venezia 1990.
8. ENV 206, "Calcestruzzo - Prestazioni, produzione, getto e criteri di
conformit", febbraio 1989.
9. G. Bianchi, F. Mazza, "Corrosione e protezione dei metalli", ed.
Masson, Milano 1989.
10. Draft Reccomendation CEN/TC 104 WG 3, "Efficacy testing of
corrosion inhibiting admixtures for concrete exposed to chlorides".
A C 1 SP-148 94 0662949 0520390 262 M
TABLE 1 - CHARACTERISTICS OF THE BASE MORTAR

Characteristic mortar A mortar B mortar C mortar D
Silica Fume % 0,o 5,O 10,o 15,0
w/c 0,375 0,375 0,375 0,375
Consistency mm 60 50 28 25
Specific weight g/cm3 2,Ol 2,11 2,07 2,06
Permeability cm 6,O 4,o 3,O 3,0
Unrestrained expansion % -0,l -0,2 -0,5 -0,5
Compressive strength MPa
1 day 22,6 24,l 26,7 19,2
3 days 44,5 44,9 45,7 40,7
7 days 51,O 52’5 56,4 54,0
28 days 60,2 62,O 68,5 61,2
Flexural strength MPa
1 day 4,5 4,5 4,6 4,0
3 days 6,O 5,5 65 5,5
7 days 7,8 7,o 83 7,0
28 days 9.0 8,O 10,6 8,5
TABLE 2 - COMPRESSIVE AND FLEXURAL STRENGTH AND YOUNG‘S MODULUS

Mortar denomination
-
A c2 c5 c9 c12
Type of polymer STI.ACRI ACRI STI.BUT TER.VIN1
- 9% 9% 9% 9%
w/c 0,375 0,375 0,300 0,250 0,375
Consistency mm 60 80 40 55 40
Compr. strength MPa
1 day 22,6 22,8 21,7 15,9 10,9
7 days 51,O 54,O 40,4 43,6 38,0
28 days 60,2 66,5 51,5 55,4 51,9
Flexural strength MPa
1 day 4,5 5,O 3,4 3,8 2,9
7 days 7,s 82 66 72 7,8
28 days 9,O 12,o 92 10,l 10,0
Young’sModulusMPa
1 day 21490 18250 13780 15190 10810
7 days 24440 24460 19230 20890 19360
28 days 25140 26820
- 20030 22820 20620
Specimen Type of Mixtures Cover

Zone 1 Zonc 2 Inn
A BC BC 20
B BC BC 30
C BC BC 40
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
D BC DC 20
E BC DC 30
F BC DC 40
G BC c2 5
H BC c2 10
I BC c2 15
A C 1 SP-248 9 4 I0662949 0520292 L T 9 I
100
1
A c c2 c5 CO c12
Type of mortar
Fig. 1-Workability of the modified compounds
A C C2 c5 CO ci2
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Type of mortar
Fig. 2-Permeability of the modified compounds
Fig. LCompressive strength of the modified compounds
A C 1 S P - I 1 4 8 74 0662949 0520392 035 W
426 Mantegazza, Penna, and Taîîoni
"1
10
O 10 2 0 JO
Age (days)
Fig. &Young's modulus of the modified compounds
0,o -
Y%
c2
c12
4.2 c9
c5
4.3
C
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. %Restrained expansion of the modified compounds
c cz c5 c9 c12
Type of Mortar
Fig. &Toughness index of the modified compounds
A C 1 SP-348 94 W 0662747 0520373 T7L
Concrole
Fig. 7-Specimen for durability tests
Temp. (" c)
I
TIME (days)
Fig. &Soaking and drying cycles
Fig. 9-Types of failure: a) specimen, b) adhesion failure, c) failure in mortar,

d) failure in base concrete, e) mixed failure
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4ô 74 Obb2747 0520394 908
.
.
100
ea
Matar
8
Y
-a
e!"o
.-
40
Adheam
Cwiuete
e
O
x
F
20
O
O 30 60 90 120 lea
Age (days)
Fig. 1û-Type of failure in H,O
c5 cs e5 cs cs cs
100 A C2 A c2 A C2 A C2 A C2 A C2
-
A
ae
ea
o
Adheron
e *
-m Concie1e
40
O
c"g 2 0
O
O 30 60 90 I20 180
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Age (days)
Fig. 11-Type of failure in CaCI?
Fig. 12-Type of failure in Na2S04
A C 1 SP-148 94 Obb2949 0520395 A44
-a 4.0 1 o o
3.0
2
v
Z
0 2.0
-
ln
Y AGEING IN H20
cxxxx~mortar A
MW moriar C5
A- mortar C2
0
-
AGE (days)
Fig. 13-Adhesion strength with time in H,O

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
CB O
w - O AGEING IN CaCI,
I
2 1.0 -
m223 mortar A
MUMmortar C5 0
A ~ M mortar
A C2
0.0 4
O 30 120 150
,
180
%E ?days)
Fig. l d d h e s i o n strength with time in CaCIl

4.0 1
AGEING IN Na2S04
QXXXImoriar A
maw mortar C5
A L U A mortar C2
0.0 I
O 30 60 90 120 150 180
AGE (days)
Fig. 15-Adhesion strength with time in No,SO,
A C 1 SP-148 94 0662947 0520196 780 =
~~
Reference mortar A
Depth:
0,07
-1 0.06
-g
L
E
C
P
o
p
-
O
æ
0 0.01
0.05
0.04
0.03
0.02
o
7ed1.5. 2.7 cm
0 3 . 2 . 4.4 cm
H4.9 . 6.1 cm
1 3 6 12 24
Age (month)
Mortar C5
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Depth:
0.01 5
D
-
.-E
L
0.01 H 1.5 - 2.7 cm
al 3.2 - 4.4 cm
a
l
P
al 0,005 4.9 - 6.1 cm
P-
-
O
æ
o
O
1 3 6 12 24
Age (monih)
~~~~
Fig. l b c h l o r i d e penetration
A C 1 SP-148 9 4 IObb2949 0520397 bL7
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Reference moriar A
0.6 DeDth:
c
q 0.4
2
c 0.3 0 3 . 2 - 4.4 cm
n
o 0.2 H4.9 - 6.1 cm
83
c
i 0.1
O -I
1 3 6 12 24
Age (monlh)
Mortar C5
Depth:
H 1.5 - 2.7 cm
0 3.2 - 4.4 cm
4.9 - 6.1 cm
1 3 6 12 24
Age (rnonlh)
Fig. 17-Sulfate penetration
A C 1 SP-148 9 4 W 0662949 0520198 5 5 3 =
wer
25
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. lû-corrosion test specimen
I NaCI solution 3%
\ Reinforcing bar 0 = 12 mrn
Fig. 19-Corrosion test method
A C 1 SP-148 9 4 06629g9 0520199 4 9 T
I I I
I I I
! I I
I
, - - -.-,--- - - - - _,. ...- - - r - - - - -

I 1 I
I
8
- covbr 20 rnp
OPCLBP cover 30 rnm I
A+ covpr 40 m p
I I
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig, 2 U u r r e n t density I for covers in base concrete
Fig. 2 1 4 u r r e n t density I for covers in durable concrete
A C 1 SP-148 94 =0bb2949 0520200 T 3 L =
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 224urrent density I for covers in mortar C2
O
o
I
E
O
0.001
=0.0001
:
-
-
oewo base concrete BC
durable concrete DC
M r e p a l r mortar C2
Fig. 23-Norrnalized rate of corrosion for the tested covers
A C 1 SP-L4ô 74 m 0 6 6 2 7 4 7 052020L 9 7 8 m
SP 148-25
Parameters Influencing the Oxygen

and Chloride Permeability of
Superplasticized Normal Portland
Cement and Fly Ash-Normal Portland
Cement Mortar and Concrete
by CJ. Lynsdale and J.G. Cabrera
Svnopsis; inadequate performance of concrete structures is often caused by

deficient construction practices, poor supervision, and lack of appropriate
specifications for controlling the properties of concrete that are related to
adequate performance during the expected life of a structure. Work carried out for
many years at the Civil Engineering Materials Unit (CEMU) of the University of
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Leeds has shown that the durability of concrete can be assessed effectively by
measuring its permeability to gases, liquids, and ions.
This paper presents the findings of a laborato9 study of the properties of normal
portland cement and fly ash-normal portland cement mortar and concrete mixtures
that influence their oxygen and chloride-ion permeability. The study involves
twenty eight mixtures incorporating the use of five chemically different
superplasticisers and three water-cementitious materials ratios.
Statistical models which relate compressive strength, porosity, pore size-

distribution and water-cementitious materials ratio to oxygen and chloride
permeability are presented.
Kewords: Chlorides; compressive strength; durobiliiy; models; mortars frnaterialj;

oxygen; performance; permeobilitu; porosity; portland cement; superplasticizen; water cement
ratio

A C 1 SP-148 9 4 Ob62749 0520202 804 =
436 Lynsdale and Cabrera
Dr C J Lynsdaie is a lecturer at Sheffield University in the Department of Civil

and Structural Engineering. He obtained his PhD at Leeds University. His main
interests are concrete materials. He is currently engaged in the study of factors
which influence the durability of concrete.
Professor J G Cabrera is Professor of Civil Engineering Materials in the

Department of Civil Engineering at the University of Leeds. He has carried out
extensive research on the performance, properties and durability of concrete. He is
also responsible for the development of a number of test methods to assess
concrete performance and has writen extensively on the subject of performance of
civil engineering materials.
INTRODUCTION
Laboratory concrete mixture proportioning in the United Kingdom is largely bawd

on the attainment of a target compressive strength achieved by limiting the
maximum water-cement ratio (W/C) and the minimum cement content. Although
these criteria are adequate for compliance with the structural requirements they fall
short of the requirements for adequate performance and long term durability,
especially if the concrete is exposed to aggressive environments, Many researchers
agree that while a relationship exist$ between permeability and compressive
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
strength for a particular mixture, such a relationship cannot be applied when
dealing with concrete mixtures of different compositions ( I , 2, 3, 4). This is
attributed to the fact that different mixtures incorporating mineral or chemical
admixtures or both exhibit different pore structure characteristics. Permeability
measurements are therefore needed to predict performance, a need which is
recognised by the proposed draft of the European Standard through specifying a
meawrement related to permeability and the new CEB-FIP Code which is
supported by a separate Design Guide for Durable Concrete Structures (5,6).
The Civil Engineering Materials Unit (CEMU) of the Department of Civil

Engineering, University of Leeds has for many years been working in the
development of concrete durability models which may be used for mixture
proportioning, field control, and assessment of built concrete structures. The data
presented in this paper is part of this work. It wac obtained by the senior author
while working for his PhD at Leeds. The paper deals with the mixture
proportioning parameters that influence the oxygen and chloride-ion permeability.
Statistical models are presented based on a study of 28 concrete and mortar
mixtures incorporating a normal portland cement (npc) and a pozzolanic cement
consisting of 70% npc and 30% fly ash (fa) made with addition of five chemically
different superplasticisersand three different W/C.
A C 1 SP-148 94 0662949 0520203 740
MIXTURE PROPORTIONING AND MATERIALS
The mixture proportioning used in this investigation was based on the concept of
minimum porosity (7) which involves the selection of concrete components to
achieve maximum packing.
The resultant mixture proportions were as follow: 1 part of cement, 2.33 parts of
fine aggregate and 3.5 parts of coarse aggregate. The npc or fa-npc content of the
mixture was 325 kg!m3. This proportioning was used to prepare the three control
concretes by varying the W/C from 0.4 to 0.55 to 0.65. Table I gives the
composition and fresh concrete properties of the mixtures prepared with and
without superplasticisers. Five superplasticisers were added to the control
concretes at different dosages to obtain concretes with a wide range of workability
up to flowing concrete (see Table I).
The codes and superplasticiser types are given in Table 2.
The chemical composition of the normal portland cement and fly ash used are
given in Table 3. The fine and coarse aggregates were sand and gravel that were
obtained from natural deposits in Nottinghamshire, England and were both of
quartzitic nature and complied with the appropriate British Standards.
METHODS OF TESTING
Air content, slump, flow table (see Table I), and compressive strength were
measured according to the relevant parts of BS 1881 (8).
Total porosity was obtained by helium pycnometry using a Micromeritics

Autopycnometer and by the use of a Micromeritics Mercury intrusion Porosimeter
(MIP) which at the same time provided data on the pore-size disuibution.
The oxygen permeability was measured using the Leeds permeameter (9) and the
chloride-ion permeability by the electrical method developed by the FHWA and
adopted by AASHTO as method T277 using a computer controlled apparatus
which allows the testing of six samples simultaneously and records the data
automatically. This has been fully described in reference (2).
PRESENTATION OF RESULTS AND DISCUSSION
n h
Since compressive strength is the main parameter used for the proportioning of
concrete mixtures there is in theory an advantage if this could also be used to
assess the durability of concrete. in this work an attempt was made to relate
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0 6 6 2 9 4 9 O520204 687 W
438 Lynsdale and €abrera
oxygen permeability of concrete (i?. to)compressive strength (f,) of the mixtures

indicated in Table 1 tested at different ages, from three days up to one year. This
resulted in a cloud diagram as shown in Fig. 1. Careful observation of the results
showed that a good statistical relationship can be obtained for any "one" concrete
mixture, but this cannot be generalised for application to different types of
mixtures. Lawrence (IO) reporting results of oxygen permeability and strength of
concretes made with different types of pozzolanic cements indicated that there is a
unique rekttionship between these two parameters; however, examination of his
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
resultq showed that for a particular value of K, concrete strength varied from 30
to 60 MPa. The results presented in this paper confirm the findings of Lawrence
and other researchers (10, 11) who have indicated that strength alone cannot be
used to ensure adequate performance and long term durability of concrete.
Chioride-ion permeability (Ci$. was measured using mortars and the results
obtained were also plotted against the compressive strength (fm) of the same
mortars. The results are shown in Fig. 2. These closely resemble the trend shown
in Fig. 1, i.e. for any value of compressive strength there is a wide range of values
of Cip. When the results are separated into npc concretes, fa-npc concretes, and
discarding the mixtures containing CLS2 admixture which entrains large quantities
of air and therefore exhibit low strength without impairment of its chloride-ion
permeability, the relationship between Cip and fm can be expressed by two valid
statistical equations:
a) For mixtures made with npc:
fm = 130.79 - 9.42(Ci ) 0.25

R2 = 0.82, CV = 113
.!%
b) For mixtures made with fa-npc:
fm = 64.97 -0.36(Cip) 0.5

R2 = 0.81, CV = 9.8%
Proposing valid equations which are limited to mixture type and composition are
not very meful except to highlight the fact that there is no possibility of ensuring
high quality and durability of concrete if other important variables which
characterise any particular solid are not considered and assessed quantitatively.
The main variables which are normally considered are those that influence pore
structure characteristics. For this study the data considered waq the pore-size
distribution and the total porosity; they were used directly to assess their infiuence
on the permeability of the superplasticised and control concretes, but other
parameters that indirectiy influence the pore structure characteristics were also
evaluated these were: W/C, age of the concrete (t), air content of concrete (A), air
A C 1 SP-148 94 m 0 6 6 2 9 4 9 0520205 513 m
content of mortar (Am), and total porosity of the paste fraction of concrete or
mortar (Pp>.
Relation between Dermeabiiitv and 90re structure
The most common parameters used to quantify pore structure are the total pore
volume and the distribution of pore sizes. Many researchers have explored the
relation between these properties of concrete, mortar or hydrated cement paste,
and permeability. Comparisons are often difficult due to the differences in
procedures for the measurement of permeability and pore structure. Coto and Roy
(12) for example found that the volume of pores greater than O. 15 pm in hydrated
cement was the controlling parameter for permeability. Mehta and Manmohan (13)
reported good correlation between coefficient of permeability and volume of pores
larger than 0.132 pm and concluded that when hydrated cement pate did not
contain pores larger than O. 132 pm the permeability decreases very rapidly. Nyame
and Illston (14) found a good correlation between permeability and the maximum
continuous pore ratio and found that this relationship holds regardless of the
water-cement ratio or age of a hydrated paste. Careful analysis of the data
presented in the paper of Nyame and nlston for example, shows that although the
statistical correlation is good (due to the wide range of values of the dependent
variable, i.e. to 10-14 mísec) the actual measured values differ by one to two
orders of magnitude from the best fitted line.
Because the range of permeability values measured in this investigation was

comparatively narrow (0.15 to 16 x 10-'6 m2) and the measurement$ were made
using concrete and mortar samples the variability of the measurements was, as
expected, relatively high. An extensive statistical analysis wa$ carried out
exploring the relationship between permeability and many of the parameters which
describe the pore-size distribution i.e. average pore diameter, median pore
diameter on volume or surface area basis, and the volume of different ranges of
sizes of pores.
The best correlation was found between the oxygen permeability and the volume
of pores larger than 0.01 pm (Vo,oi). The correlation explained 68% of the
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
variability on permeability. Fig. 3 shows this relationship and confirm other

researchers findings that the V,,, correlates reasonably weil with permeability
(15, 16).
The same type of analysis was carried out using C. as the dependent variable, the
'P
results confirmed that the best statistical relationship was when Cip was correlated
with V,,,,. in this case however, the correlation coefficient was lower than for the
oxygen permeability - Vo.oi relationship.
A C 1 SP-148 74 m Ob62749 O520206 4 5 T m
Relation between aermeabüitv and Other concrete mixture naramete-
From the statistical analysis carried out for evaluating permeability against most
mixture parameters which were determined in this study, it was found that
permeability is affected by W/C, age, air content, and the porosity of the cement
paste. From these studies, two models are proposed: one for concrete oxygen
permeability and one for mortar oxygen permeability. The numerical expression of
the models are:
Log K,= 0.079A(W/C) - 0.012(t)0.66+O.O37Pp(W/C) -16.86

(3)
R2 = 0.80
Log K, = 0.027(A,) + 1.473(W/C) (t)-0.66+ O.O12PP(W/C) - 16.72

(4)
Rz = 0.8 1
Where:
K, = oxygen permeability of concrete (m2)
K, = oxygen permeability of mortar (m2)
W/C = water-cement ratio
t = age of specimen (days)
A = air content of concrete (S)
Am = air content adjusted for mortar (%)
Pp = porosity of paste fraction (%)
The calculated results using the models given were plotted against the measured
values; these are shown in Fig. 43. The models and variables included were highly
significant on the 95% confidence limit.
Relation between oxwen and chloride ion permeabilities
The results of mortar oxygen permeability were plotted against the values of
chioride-ion permeability and, although there was a trend relaring these values, the
scatter of the data was wide. However, from this relationship it is apparent that
concrete mixtures containing CLS2 and CPI superplasticiser and mixtures
containing fa (see Table 1) exhibit high oxygen permeability when compared to
other mixtures made at the same W/C of 0.4; this is attributed to the high air
content and porosity which CLS2 CPI and fa cause in these mixtures. The
chloride-ion permeability for these mixtures is similar to that of other mixtures
made with the same W/C. This is an indication that the movement of solvated
chloride ion, induced by the applied voltage is governed by the open pore
structure of the matrix and the presence of air voids does not necessarily increase
the flow rate, contrary to their effect on the flow of oxygen.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 S P - I J ~ B9 4 0bb2747 0520207 376
A regression equation of statistical validity is found between oxygen and chloride-

ion permeability when excluding mixtures with high air content (those mixtures
mentioned before). The equation is given below and a graphical representation is
presented in Fig. 6.
Log Km = 5.07 (Ci$ x - 16.6

R2 = 0.63
Where:
Km = oxygen permeability of mortar (m2)
Measurement of compressive strength and porosity is done using test methods that
do not require any sophistication and that are available in most concrete field
laboratories unlike for example, pore size distribution apparatus. For this reason,
and because it has been shown that strength is inversely related to permeability and
porosity directly related to it, the data of this investigation was used to ascertain if
models could be found to predict permeability from these two simple parameters.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
When analysing the data of permeability is was found that the statistical relations
could be vastly improved by separating the data for npc and fa-npc mixtures. The
equation which describe the models are given below.
For npc concrete and mortar:
Log K,= 1.1 14 Log (P/fc) - 15.54

R2 = 0.7 1
For fa-npc concrete and mortar:
Log KC = 0.97 Log (P/fc) - 15.97

R2 = 0.87
Log Cip = 30783 (P/fc) O.5 - 16593 (9)

R2 = 0.86
Fig. 7 is the graphical representation of equations 6 and 8 and Fig. 8 is the

graphical representation of equations 7 and 9. in these figures the parametric space
limited by porosity and strength has been divided with qui-permeability lines. The
A C 1 SP-146 94 0 6 6 2 9 4 9 0520208 222
data presented in this form allows for very rapid predictions of permeability for
any strength and associated porosity value.
The models offered here can be used as an aid during the mixture proportioning
stage for the proportioning of durable concrete. These models can also be used to
investigate why concrete smctures have deteriorated when the strength of the
concrete is apparently adequate and complies with the appropriate standards. For
this it will be necessary to determine iimiting acceptable values of permeability
which obviously wiil be related to the aggressiveness of the environment where a
structure has been built.
CONCLUSIONS
1. The analysis of the data shows that, with the exception of concretes made
with superplasticisers which entrain excessive amounts of air,
superplasticised and control concretes exhibit similar permeability and
other properties.
2. The study c o n f i i s that strength alone cannot be used to predict

permeability of concrete mixtures of different pore structure
characteristics.
3. The volume of pores larger than an equivalent diameter of 0.01 pm

correlates reasonably well with permeability.
4. Useful statistical models relating W/C, age, volume of air entrained, and
porosity of the paste fraction have been presented for concretes made with
npc or fa-npc.
5. The models relating compressive strength and porosity of concrete to its

permeability are offered as the models that could be used to aid mixture
proportioning for performance and durability and for assessment of
concrete structures where signs of poor performance have been detected
even if the structural properties of the concrete are adequate.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 I0bb2749 0520209 Lb9
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
REFERENCES
I. Hilsdorf H K. Durability of Concrete - a Measurable Quantity? IABSE

Symposium on Durability of Structures No. 57, Vol. 1, pp 111 - 123.
Lisbon, 1989.
2. Cabrera J G and Lynsdale C J. Measurement of Chloride Permeability in

Superplasticised Ordinary Portland Cement and Pozzolanic Cement
Mortars. Proc. Int Conf. on Measurements and Testing in Civil
Engineering. Vol. 1, pp 179 - 191, Lyon, 1988.
3. Ramachandran V S, Feldman R F and Beaudoin J J. Concrete Science p

31-32, Hyden, 1981.
4. Cabrera J G and Nwaubani S O. Strength and Chloride Permeability of

Concrete Containing Red Tropical Soils. Magazine of Concrete Research
Vol. 45, No. 164, pp 169-178, 1993.
5. CEB. Guide to Durable Concrete Structures. Durable Concrete Structures

Bulletin D'information No. 183, Thomas Telford, 1992.
6. CEB-FIP Model Code 1990, First pre-draft 1988, CEB-Bulletin

d'information No. 190.
7. Cabrera J G . Concrete Mixture Design Based on the Concept of Minimum

Porosity. Concrete Research Seminar, Leeds 1985.
8. British Standard institution. BS881:1983 Parts 102, 105, 106 and 116.
BSI, London, 1983.
9. Cabrera J G and Lynsdale C J. A New Gas Permemeter for Measuring the

Permeability of Mortar and Concrete. Magazine of Concrete Research Vol.
40. no. 144, pp 77 - 182, 1988.
10. Lawrence C D. Measurements of Permeability. Proc. 8th Int. Congress on

the Chemistry of Cement. Vol. 5, pp 29 - 34, Rio de Janeiro 1986.
Il. Rostm S . Influence of Material Properties on the Durability of Structures.

IABSE Symposium No. 57, Vol. 1, pp 125-136, Lisbon 1989.
12. Goto S and Roy D M. The Effect of W/C Ratio and Curing Temperature
on the Permeability of Hardened Cement Paste. Cement and Concrete
Research, Vol. I l , pp 575-579, 1981.
A C 1 SP-148 7 4 W 0662749 0520210 980 =
13. Mehta P K and Mammohan D. Pore Size Distribution and Permeability of

Hardened Cement Pa5tes. Proc. 7th inkmationai Congress on the
Chemistry of Cement. Vol. III, Section VII, pp i - 5, Paris, 1980.
14. Nyame B K and nlston J M. Capillary Pore Structure and Permeability of

Hardened Cement Paste. Proc. 7th international Congress on the
Chemistry of Cement, Vol III, Section VI, pp 18 1 - 185, Paris, 1980.
15. Collepardi M and Massidda L. The Influence of Water Reducing

Admixtures on Cement Paste and Concrete Properties. Conf. on Hydraulic
Cement Pastes, their Structure and Properties. pp 256-267. Sheffield 1976.
1h. Cabrera J G . The Use of Pulverised Fuel Ash to Produce Durable Concrete
in improvements in Concrete Durability. Thomas Telford, pp 19 - 35,
London, 1985.
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 94 = 0662749 0520233 B37
TABLE 1 - MIXTURES USED IN THE INVESTIGATION
* Dosage is based on solid/soiid weightlweight of cement
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-L4& 94 0 6 6 2 7 4 9 0520232 753 =
TABLE 2 - SUPERPLASTICIZER USED IN THE INVESTIGATION

Code
MLS
SNFC 1
I Superplasticiser Description
I Modified lknosulohonate
Sulphonated naphthaiene formaldehyde condensate (powder)
I
SNFC2 Sulphonated naphthalene formaldehyde condensate (liquid)
SMFC Sulphonated melamine formaldehyde condensate
CP Co-polymer based on acrylic acid and hydroxy-propyl-methacrylate
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
TABLE 3 - PROPERTIES OF NORMAL PORTLAND CEMENT AND FLY ASH
Specific surface mz/g I 0.390 0.230
63.92
10.35
9.85
6.60
A C 1 SP-148 94 0 6 6 2 9 4 9 0520213 6 î T =
r Key: A A N ~Ls3
B ENZ o np3
c c Q PFN
D PfH R BUZ
__
E a
F EN3
S ESFZ
T CP1-
0 CU2 U P E
H PFL V BU3
J ELSZ W BSF3
K õP2 X Sf1
L CU2
U PFM
10 I
0.1
I
i.o
1
10.0
Oxygen permeability x lo-'' (m2)
Fig. 1 4 l o u d diagram of the relation beiween compressive strength and permeability
Key: A A N BLS3
B EN2 O BP3
c c Q PFN
O PFH R EM2
E B 5 ESFZ
F EN3 T CP1
C G CW2 U PFS
H PFL V BM3
x L c p J EL52 W BSFJ
T K 6PZ X CSFl
L CM2 Y PFAl
T x M PFY Z PFAZ
Q C P
A W
y C ZN
C
a
H Z ME"
Y
20 I I I I I l I I
O 2000 4000 6000 8000 10000 12000 14000 16000
Charge passed (coulomb)

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Fig. 24ompressive strength versus chloride permeability for the various mixtures investigated
A C 1 SP-148 9 4 W 0662947 0520214 526
-15.2 -
N -15.4
v
E
X X
. -15.6
--1
c
x x X
X
-15.8 lr x x
X
m X x x X
E
m -16.0 X X y x
x
a u
c
-16.2
X
x>K< x X
eo -16.4 - x q x x2a x x
C
X
x x x xx x Xx x x X >k
u -16.6 X
o
1 -16.8
-17.0 1 1 1 , , , , , ~ , , , , ,
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Percent volume of pores > 0.01 micron
Fig. L R e l a t i o n between Log concrete permeability and the volume of pores greater thon 0.01 pm
-15.6 ,.
-15.7
N
v
E -15.8
2”
- -15.9
-16.0
8
E -16.1
m
a -16.2
o -16.6
O
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
pradictrd values vi. axp*rimantal value3

Ilnr = equality Ilna
Log concrete permeability (m2)
Fig. 4-Measured concrete permeability versus predicted permeability obtained using the
model proposed in equation (3)
A C 1 SP-1413 9 4 0 6 6 2 9 4 9 0520235 4 6 2
N
-15.2 -
& -15.4 -
.-2 -15.6 -
o
E -15.8
o
n
-
2 -16.4
predicted valuos V.I experimental values

Ilne = squailiy line
-17.0 -16.8 -16.6 -16.4 -16.2 -16.0 -15.8 -15.6 -15.4 -15.2 -15.0
Log mortar permeability (m2)
Fig. &Measured mortar permeability versus predicted permeability obtained from the
model proposed in equation (4)
N--15.4
v
E
-15.6
-
& x x x
0 -15.8
o
-O
1
-16.8 x
-17.0 I I I I l I I
Fig. blelationship between oxygen permeability and chloride permeability for the various
mixtures investigated with the exception of mixtures CLS2, CP1, PFL, PFH, PFN, AND FPM
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 9 4 O662949 0.520236 3 T 9
Compressive strength ( W o )
Fig. 7-The oxygen permeability-strength-porosity model
Compressive strength (MPa)
Fig. L T h e chloride ion permeability-strength-porositr model
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
A C 1 SP-148 74 I0662347 0520237 235
SI (Metric) Tables 451
CONVERSION FACTOR-INCH-POUND TO SI (METRIC)*
To convert from to multiply by
Length
inch ............................... millimeter(mm) . . . . . . . . . . . . . . . . . . . . . . . .4Et

foot ................................. meter(m) ......................... 0.3048E
yard ................................. meter (m) ......................... 0.9144E
mile (statute) ........................ kilometer (km) ....................... 1.609
Area
square inch ..................... square millimeter (mm’) . . . . . . . . . . . . . . . . .645.1

square foot ........................ square meter (ni2) ...................... 0.0929
square yard ........................ square meter (m’) ...................... 0.8361
Volume (capacity)
ounce ............................. milliliters (mL) ...................... 29.57

gallon ............................ cubic meter (m3)$ ...................... OM3785
cubic inch ....................... cubic millimeter (mm3) . . . . . . . . . . . . . . . .16390
cubic foot ......................... cubic meter (m3) ...................... 0.02832
cubic yard ......................... cubic meter (m3)t: ...................... 0.7646
Force
kilogram-force ........................ . . ........................

newton(N) 9.807
kip-force ..........................
kilonewton(kN) ................ . . . 4.448
pound-force .......................... newton (N) . . . . . . . . . . . . . . . . . . . . f4.448
Pressure or stress (force per area)
kilogram-force/square meter . . . . . . . . . . . . . . . pascal (Pa) .................. . . . 9.807

kip-forcehquare inch (ksi) . . . . . . . . . . . . . megapascal ( m a ) ................ . . . 6.895
newton/square meter (N/m’) . . . . . . . . . . . . . . pascal (Pa) .................. . . . IMME
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
pound-forcehquare foot . . . . . . . . . . . . . . . . . . pascal (Pa) .................. . . 47.88

pound-forcehquare inch (psi) ............ kilopascal ( H a ) . . . . . . . . . . . . . . . . . . . . 6.895
Bending moment or torque
inch-pound-force . . . . . . . . . . . . . . . . . . . newton-meter (Nm) . . . . . . . . . . . . . . . ... 0.1130

foot-pound-force . . . . . . . . . . . . . . . . . . . newton-meter (Nm) ............... . . . 1.356
meter-kilogram-force . . . . . . . . . . . . . . . .newton-meter (Nm). . . . . . . . . . . . . . . ... 9.801
A C 1 SP-148 94 m 0662949 0520238 3 7 1 m
452 SI (Metric) Tables
To convert from to multiply by
Mass
ounce-mass (avoirdupois) .................. gram (9) .................... . 28.34

pound-mass (avoirdupois) . . . . . . . . . . . . . . .kilogram (kg) ................... . ~ 0.4536
ton (metric) ......................... megagram (mg) .................. . . LOWE
ton (short, uxw) Ibm) ................... kilogram (kg). .................. 907.2
Mass per volume
pound-masshbic foot . . . . . . . . . . . .kiiogramicubic meter (kg/m3) ............. . 16.02

pound-madcubic yard ............ kiiogram/cubicmeter (kg/m3) ............. . . 0.5933
pound-mass/gaüon . . . . . . . . . . . . . . . kilogram/cubicmeter (kg/m3) ............. 119.8
Temperatures
degrees Fahrenheit (F) . . . . . . . . . . . . . . degrees Celsius (C) ........ = (tp 32)/1.8 -

degrees Celsius (C) . . . . . . . . . . . . . . . .degrees Fahrenheit (F) . . . . . . . . tp = 1.8t, + 32
'X i s selpted list gives practical conversion factorsof unitsfound in concrek technology.nie refwenee soum for information
on SI units and more exact conversion factors are ASTM E 380 and E 621. Symbols of metric uniis axe given in puenthesis.
t E Iodicstts that tbc factor given is exact.
Ooc liter (cubic decimeter) equais 0.001 m3 or loo0 cm3.
0 7ñe.w equations convert one temperalure reading lo another and include the necessary mrrstiono. To amvert a
difference in lemperaturc from Fahrenheit degrees lo Celsius degrees, divide by 1.8 o+, e.¡, a change from 70 to 88 F represents
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
a change of 18 Por lS/l.â = 10 C deg.
A C 1 SP-348 9 4 = 0662949 0520237 008
Index 453
INDEX
Absorption, 227 Chromatography, 333

Accelerating agents, 385 Climate, 259
Acrylic polymers, 89 Clinker, 89, 407
Acrylics, 281 Coarse aggregates, 27, 301
Admixtures, 105, 121, 131, 153, 209, 243, 301, Cold weather construction, 385
333, 353, 385 Compaction, 121, 209
Adsorption, 63, 281 Compatibility methods, 47
Aggregates, 177 Compressive strength, 27, 105, 121, 189, 227,
Air entrained concretes, 47 259, 281, 317, 353, 407, 415, 435
Air entraining agents, 47, 121, 243, 385 Concretes, 105, 301, 367
Air entrainment, 317, 333, 353 Consistency tesis, 407
Alttin, P., 47 Consolidation, 121
Alberti, E., 105 Corrosion, 385, 415
Alkali aggregate reactions, 353 Cover, 415
Ambroise, J., 89 Cracking (fracturing), 243
Arora, V., 301 Crack propagation, 243
Austin, S., 259 Curing, 259
Baalbaki, M., 47 Drying shrinkage, 27

Berke, N., 385 Ductility, 415
Blast furnace slag, 1, 177 Durability, 1, 209, 259, 415, 435
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
Bleeding (concrete), 1, 367

Blended cements, 153 Emission spectroscopy, 333
Bremner, T., 333 Expansion, 353
Bridges (structures), 27
Flexural strength, 105, 353, 415
Cabrera, J., 435 Flowability, 27, 89, 153, 177, 209, 227, 243
Calcium aluminates, 131 Fly ash, 27, 367, 435
Calcium silicates, 131 Freeze thaw durability, 27, 47, 209, 301, 317,
Calorimeters, 63, 131 353
Carbonation, 209, 227 Fresh concretes, 367
Cellulose fibers, 189
Cement-based systems, 189 Gillott, J., 353
Cement pastes, 63, 281 Grout, 189
Cements, 47, 63, 131, 333 Gypsum, 131
Chemical analysis, 333
Chlorides, 435 Heat of hydration, 63
A C 1 SP-I48 94 0662749 0520220 8 2 T =
454 Index
High performance concretes, 1, 301 Nishiyoma, N., 227

High strength concretes, 27, 121, 243, 281 Nkinomubanzi, P., 63
Hiraishi, S., 209, 243 Nondestructive tests, 227
Hrazdira, J,. 407
Hydration, 131 Osada, K., 209
Oxygen, 435
Inorganic salts, 385
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---
houe, S., 177 Postes, 407
Penno, A., 415
Janssen, D., 317 Péro, J., 89
Jeknovorian, A., 385 Performante, 435
Jolicoeur, Cl 63 Permeability, 1, 227, 259, 415, 435
Phosphonote, 131
Kagoya, M., 121 Piotte, M., 63
Koneko, T., 121 Plasticizers, 47, 333, 367
Kasai, Y., 209, 227, 243 Plastics, polymers, and resins, 281, 367
Kawa¡, K., 367 Polysaccharides, 105, 189
Kawakami, M., 121 Porosity, 1, 259, 435
Kinoshita, M., 27, 281 Portland cernent, 131, 407, 435
Kishitani, K., 243
Kume, T., 177 Rakitsky, W., 189
Kuroda, T., 177 Ramachandran, V., 131
Reinforced concrete, 177
Latex, 317, 415 Repairs, 415
Lignin and derivatives, 333 Rheological properties, 63, 89, 105, 177, 189,
Limestone, 177 333
Lithium compounds, 353 Robins, P., 259
Lora, J., 333
Lowery, M., 131 Sakai, M., 1
Lynsdole, C., 435 Sarkar, S., 153
Scaling, 47, 317
Montegano, G., 105, 415 Schemmel, J., 301
Maximilien, S., 89 Segregation, 121, 177
Minerol admixtures, 63 Setting (hardening), 1, 27, 105, 333, 367, 38,
Mitsui, K., 27, 281 407
Mix proportioning, 27, 209, 259 Shen, D., 385
Miyazawa, S., 367 Shimono, T., 27
Models, 227, 435 Shrinkage, 209, 243
Modulus of elasticity, 415 Silica fume, 27, 177, 259, 281, 367
Moist curing, 317 Simord, M., 63
Morton (material), 89, 105, 281, 333, 415, 435 Skoggs, C., 189
Slogs, 1, 367
Nokamuta, N., 1 Slump, 27, 47, Xi,121, 153, 209, 367
Nokata, Y., 227 Slurries, 63
Newtson, C., 317 Sodium carbonate, 407
Nishibayashi, S., 177 Sone, T., 153
A C 1 SP-148 7 4 0662749 0520223 766
index 455
Strength, 1, 209, 301, 385 Vibrators (machinery), 121

Sulfur, 333 Viscosity, 153, 281
Sulphonates, 333
Superplasticizers, 1, 27, 47, 63, 105, 153, 177, Wang, H., 353
189, 259, 281, 317, 367, 407, 435 Water cement ratio, 281, 353, 435
Suzuki, T., 281 Water reducing agents, 1, 121, 177, 243, 301,
Swomy, R., 1 333
Water retentivity, 189
Tattoni, S., 415 Whitaker, S., 189
Jazawa, E., 367 Williams, J., 301
Temperature, 27 Workability, 27, 105, 121, 177, 259, 281, 301,
Tension, 281 415
Tests, 177, 221, 333
Tobinai, K., 209 Yokoyama, K., 243
Takahashi, M., 367 Yonekura, A., 367
Tokuda, H., 121 Yonezawa, T., 27, 281
Tpl, S., 353 Yoshino, A., 177
Yuasa, N., 227
Uchikawa, H., 153
Underwater construction, 189 Zhor, J., 333
--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---

Superplasticizers & Other Admixtures For Concrete

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Superplasticizers & Other Admixtures For Concrete

Hochgeladen von

Copyright:

Verfügbare Formate

--```,``,,`,````,``,,,`,````,,-`-`,,`,,`,`,,`---

DISCUSSION of individual papers in this symposium may be submitted in

The Institute is not responsible for the statements or opinions expressed in

Printed in the United States of America

Editorial production Victoria Wieczorek

Library of Congress catalog card number 94-72367

In May 1978, The Canada Centre for Mineral and Energy

In October 1994, CANMET in association with the American Con-

V.M. Malhotra, P. Eng.

Fourth CANMET/ACI Intemational Conference

Theodore W. Bremner (Member) Wilbert (Wib) S. Langley (Member)

COPYRIGHT 2003; ACI International (American Concrete Institute)

ORGANIZING COMMITTEE .......................................... iv

ROLE OF SUPERPLASTICIZERS AND SLAG FOR PRODUCING HIGH

APPLICATION OF A NEW SUPERPLASTICIZER FOR ULTRA HIGH-STRENGTH CONCRETE

CEMENT/SUPERPLASTICIZER/AlR-ENTRAINING AGENT COYPATIBILIM

PROGRESS IN UNDERSTANDING THE FUNCTIONAL PROPERTIES OF SUPERPLASTICIZERS

INFLUENCE OF ACRYLIC POLYMERS ON THE RHEOLOGY OF MORTARS

SLUMP LOSS AND RHEOLOGY OF SUPERPLASTICIZED MORTAR AND CONCRETE WITH

WORKABILITY CHARACTERISTICS OF HIGH-STRENGTH CONCRETE INCORPORATING

EFFECT OF A PHOSPHONATE-BASED COMPOUND ON THE HYDRATION OF CEMENT AND

THE INFLUENCE OF CROSS-LINKED AND NSF SUPERPLASTICIZER ON THE FLOW

A STUDY ON THE FLOW OF HIGHLY SUPERPLASTICIZED CONCRETE

APPLICATIONS OF RHEOLOGICAL MODIFIERS AND SUPERPLASTICIZERS IN

MIXTURE PROPORTION, FLOWABILITY, STRENGTH AND DURABILITY OF FLOWING CONCRETE

AN EXPERIMENTAL CONSTRUCTION OF MODEL STRUCTURE WITH FLOWING CONCRETE

SHRINKAGE AND CRACKING OF HIGH-STRENGTH CONCRETE AND FLOWING CONCRETE

THE INFLUENCE OF SUPERPLASTICIZER ON MIXTURE PROPORTIONING AND THE STRENGTH

PROPERTIES OF AN ACRYLIC GRAA COPOLYMER-BASED NEW SUPERPLASTICIZER FOR

SPUT ADDITION OF A HRWRA AND ITS EFFECT ON HIGH-PERFORMANCE CONCRETE

EFFECT OF MOIST CURING AND MODIFIER CONTENT ON PERFORMANCE IN FREEZING

EFFECT OF CHEMICAL CHARACTERISTICS OF ALCELL@ LIGNIN-BASED

PRACTICAL IMPLICATIONS OF LITHIUM-BASED CHEMICALS AND ADMIXTURES IN

PLASTICIZING CHARACTERISTICS OF SULFONIC ACID POLYMER CONTAINING

PERFORMANCE EVALUATION OF SET ACCELERATORS FOR CONCRETE

COPYRIGHT 2003; ACI International (American Concrete Institute)

ROLE OF SUPERPLASTICIZERS IN GYPSUMLESS PORTLAND CEMENTS

MODIFICATION OF CEMENT MORTARS BY POLYMERIC LATEXES AND THEIR USE

PARAMETERS INFLUENCING THE OXYGEN AND CHLORIDE PERMEABILITY OF

SI (Metric) TABLES ............................................. 451

INDEX .................................................... 453

Role of Superplasticizers and Slag for

AC1 Fellow Prof. R. Narayan Swamy is at the Structural Integrity Research

Masami Sakai is a general manager at Building and Construction Centre, NKK

Nobuyuki Nakamura is a senior engineer at Building and Construction Centre,

Neither the use of superplasticizers nor the incorporation of ground,

Whilst good workability is accepted as an essential requirement for proper

effective synergic interaction to produce concrete of high strength, high

Concrete Mixture Details

All the mixtures were proportioned to have high workability with

TEST RESULTS AND DISCUSSION

Properties of Fresh Concrete

The rate of bleeding of all the mixtures was determined according to

The heat evolved during the hydration of the cement-slag combination

The results of the conduction calorimetry shown in Fig. 4 emphasize the

The heat of hydration is a major factor influencing the long-term

PROPERTIES OF HARDENED CONCRETE

Compressive Strength and Elastic Modulus

The early development of compressive strength and Young's modulus

Fig. 6 shows the variation of Young's modulus with compressive strength.

CONCRETES WITH SLAG OF VARYING FINENESS

It is obvious that the maximum technical, practical and economic

It is recognised that higher costs will be incurred in grinding the slag