Sodium and Magnesium Sulfate Resistance of Mortars with

Interground Limestone and Limestone Powder Replacing

Cements
Ittiporn Sirisawat , Lalita Baingam, Warangka Saengsoy, Pitisan Krammart , Somnuk TangtermsirikXO

Journal of Advanced Concrete Technology, volume 12 ( 2014 ), pp. 403-412

XRD/Rietveld analysis of the hydration and strength development of slag and limestone blended cement

Seiichi Hoshino , Kazuo Yamada, Hiroshi Hirao

Journal of Advanced Concrete Technology, volume 4

( 2006 ), pp. 357-367

Effects of Shape and Packing Density of Powder Particles on the Fluidity of Cement Pastes with
Limestone Powder

Etsuo Sakai , Keisuke Masuda , Yasuo Kakinuma , Yutaka Aikawa

Journal of Advanced Concrete Technology, volume 7 ( 2009 ), pp. 347-354

403

Journal of Advanced Concrete Technology Vol. 12, 403-412, October 2014 / Copyright 2014 Japan Concrete Institute

Technical report

Sodium and Magnesium Sulfate Resistance of Mortars with Interground

Limestone and Limestone Powder Replacing Cements
Ittiporn Sirisawat1, Lalita Baingam2, Warangkana Saengsoy3, Pitisan Krammart4* and
Somnuk Tangtermsirikul5
Received 9 March 2014, accepted 3 October 2014

doi:10.3151/jact.12.403

Abstract
Sulfate resistance of mortars containing limestone powder was investigated in a laboratory in which mortar specimens
were continuously immersed in 33,800 ppm of sodium and magnesium sulfate solutions for 1,700 days. Mortars made
from interground limestone and limestone powder, replacing cements with different proportions (10%, and 20% limestone by weight of the blended cement), were used to compare with the control Portland Type 5 cement mortar. Expansion and weight loss were measured. Microstructural analyses such as SEM, MIP, TGA and XRD techniques were also
performed on the paste samples. It was observed that interground limestone cement specimens had higher expansion
than the limestone powder replacing cement specimens due to smaller average pore size and lower total porosity, providing less spaces for depositing products of expansion. Contrary to the expansion, the specimens made from interground limestone cement lost less weight than those made from the limestone powder replacing cement because of
lower average pore size and total porosity, making the specimens denser. For mortar containing limestone powder
specimens in MS solution, the MS decreased the systems pH. There occurred higher dissolution of CaCO3 from the
limestone powder, which contributed to the formation of gypsum, magnesite and dolomite. This reduced the conversion
of C-S-H to M-S-H, which resulted in less weight loss and less severe surface etching than the specimens made from
OPC.

1. General
Sulfate attack is a type of deterioration resulting from
chemical reactions which occur when concrete is exposed to solutions containing a sufficiently high concentration of dissolved sulfates (SO42-). Sulfate attack has
been reported in many parts of the world (Miller and
Manson 1951; Mehta 1975; Al-Amoudi 1998; Tangtermsirikul 2003; Baingam et al. 2012). This is particularly prevalent in sulfate containing regions such as in
water or in soils. Necessary conditions for sulfate attack

PhD Student, School of Civil Engineering and

Technology, Sirindhorn International Institute of
Technology, Thammasat University, Pathum Thani,
Thailand.
2
Graduate Student, School of Civil Engineering and
Technology, Sirindhorn International Institute of Technology, Thammasat University, Pathum Thani, Thailand.
3
Researcher, Construction and Maintenance Technology
Research Center (CONTEC), Sirindhorn International
Institute of Technology, Thammasat University, Pathum
Thani, Thailand.
4
Assistant Professor, Department of Civil Engineering,
Rajamangala University of Technology Thanyaburi,
Pathum Thani, Thailand.
*Corresponding author, E-mail: pitisan.k@en.rmutt.ac.th
5
Professor, School of Civil Engineering and Technology,
Sirindhorn International Institute of Technology,
Thammasat University, Pathum Thani, Thailand.

should be investigated and preventive measures should

be taken to eliminate or minimize the damage due to
sulfate attack. Therefore, there is a need to clarify the
mechanisms of sulfate attack for proper practices.
Blended cements are defined as hydraulic cements
consisting of an intimate and uniform blend of a number
of different constituent materials (ACI 116R-00 2000).
They are produced by intergrinding Portland cement
clinker with other materials or by blending Portland
cement with other materials or a combination of intergrinding and blending. Recently, pozzolans and byproducts such as fly ash, silica fume, blast furnace slag
and filler material such as limestone powder are typically recommended for improving the properties of concrete. When used as partial replacement of cement, they
modify both the physical and the chemical properties of
the hydrated products and improve the impermeability
(Berry and Malhotra 1980; Rachel et al. 1996; Heikal et
al. 2004)
Limestone powder (LP) has been increasingly used in
the concrete industry as a filler in concrete mixture or as
a component in blended cement, for many years. It is
applied in high performance concrete as well as in normal concrete. Studies related to the mechanical properties of concrete incorporating LP were carried out to
achieve the reliability needed for increasing its usage
(Kaewmanee et al. 2006). Moreover, there have been
numerous studies in Thailand on the durability of LP
cement concrete when exposed to aggressive conditions
(Thasanakosol et al. 1999; Sukayanudist et al. 2009;
Ponpo et al. 2010; Wattanakul et al. 2010; Krammart et

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

al. 2011). In recent years, there have been some interests

in research of interground limestone cement (Tsivilis et
al. 1999; Tsivilis et al. 2002). However, there are few
studies to compare the effect of intergrinding and replacing on the properties of limestone cement, especially on sulfate resistance.
This study aimed to investigate sulfate resistance
properties of the interground limestone cement and
compare its properties with that of the limestone powder
replacing cement. The sulfate resistance performances
of mortar specimen exposed in NS and MS solutions
were evaluated by measuring expansion and loss in
weight. In order to describe the mechanisms of the difference, studies on microstructure and product composition were also conducted.

complying with TIS 15 P1-2004. Two interground limestone cements, with 10% and 20% limestone by mass
(ILC10 and ILC20), were used in this study. All of the
Portland cements and the interground cements were
manufactured by a cement manufacturer in Thailand.
Chemical composition, physical properties and particle
size distributions of all tested powder materials are
shown in Table 1 and Fig. 1. In addition, Figs. 2(a)2(b) show the scanning electron microscopy images
(SEM) of the interground limestone cements while Fig.
2(c) is for limestone powder used in the LP replacing
cement. The figures show that the sizes of LP in the
interground limestone cement are relatively smaller than
those of the limestone powder used in the LP replacing
cement. The LP used for LP replacing cement in this
study was a fine grain LP, with a mean particle size of
about 3.2 microns, and was manufactured by a limestone powder supplier in Thailand. The chemical composition and physical properties of the LP are shown in
Table 1. The particle size distributions of the LP are
also presented in Fig. 1.
Twelve different mixture conditions of the tested

2. Experimental tests
2.1 Materials and mix proportions
The mortar mixtures used were prepared with OPC 1
(Ordinary Portland cement) and OPC 5 (Sulfateresisting cement) Portland cements of commercial grade
OPC 1

OPC 5

ILC10

ILC20

LP

100

C u mulative Passing %

80

60

40

20

0
0.001

0.01

0.1

1
Particle Diameter (m)

10

100

1000

Fig. 1 Particle size distribution of binders.

Table 1 Chemical composition, physical properties and mineralogical composition of binders.
Chemical
Composition (%)
SiO2
Al2O3
Fe2O3
CaO
MgO
Na2O
K2O
SO3
Free Lime
Loss on Ignition
Specific Gravity
Blaine fineness (cm2/g)
Mean diameter (m)

404

OPC 1

OPC 5

ILC10

ILC20

LP

19.51
4.97
3.78
65.38
1.08
<0.01
0.47
2.16
1.00
2.27
3.12
3,550
15.41

21.87
3.87
4.34
64.56
1.11
<0.01
0.24
2.08
0.76
1.59
3.18
3,830
14.43

17.48
4.49
3.36
63.29
1.02
<0.01
0.44
2.35
0.78
7.20
3.17
4,500
12.34

16.20
4.13
3.01
62.67
0.98
<0.01
0.43
2.37
0.61
9.93
3.09
4,600
11.34

0.46
0.06
0.03
55.25
0.37
<0.01
0.01
<0.01
43.79
2.79
8,840
3.23

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

405

Table 2 Mix proportions of the tested mortar specimens.

Mix

Mix Designation

Cement Type

1
M1-0.40
2
M1 90 LP10-0.40
3
M1 80 LP20-0.40
4
M5-0.40
5
ML10 -0.40
6
ML20 -0.40
7
M1-0.55
8
M1 90 LP10-0.55
9
M1 80 LP20-0.55
10
M5-0.55
11
ML10-0.55
12
ML20-0.55
* interground limestone cement

OPC1
OPC1
OPC1
OPC5
ILC10
ILC20
OPC1
OPC1
OPC1
OPC5
ILC10
ILC20

Mix proportion (ratio by weight)

LP
Water
0.0
0.40
0.1
0.40
0.2
0.40
0.0
0.40
0.0
0.40
0.0
0.40
0.0
0.55
0.1
0.55
0.2
0.55
0.0
0.55
0.0
0.55
0.0
0.55

Cement
1.0
0.9
0.8
1.0
1.0*
1.0*
1.0
0.9
0.8
1.0
1.0*
1.0*

Cement

Sand
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.75

Cement
LP

Cement

LP
Cement

a) 10% Interground limestone cement

b) 20% Interground limestone cement

c) Limestone powder used in limestone powder replacing cement

Fig. 2 SEM images of binders used in this study (x2,500).

mortars were designed for the tests (see Table 2). The
mortar mixtures were made from various cements with a
ratio by volume of sand to cement of 2.75. Water to
binder ratio was controlled at 0.40 and 0.55. LP was
used at 10%, and 20% of the total powder materials. It
should be noted that interground limestone cements
(ILC10 and ILC20) were used to prepare the ILC mortar
specimens while the specimens made from limestone
replacing cement were prepared by mixing the limestone powder with cement and then other ingredients in
the mortar mixer during the mortar mixing. The methods for preparing mortar specimens and testing the sulfate expansion are according to ASTM C 157. To study
the microstructural characteristics, X-ray Diffraction
(XRD), Mercury Instrusion Porosimetry (MIP), Thermogravimetric Analysis (TGA) and Scanning Electron

Microscopy with Back-Scatter Electron Mode

(SEM/BSE) with EDX analysis techniques were conducted on pastes with water to binder ratio of 0.40 for
all mix proportions.
2.2 Sulfate resistance test
Mortar bar specimens (25x25x285 mm) and mortar
cube specimens (50x50x50 mm) were prepared in accordance with ASTM C 1012 and ASTM C109, respectively. Immediately, after casting, the molds were covered with plastic sheets and the specimens were demolded at one day of age. After demolding, all bars and
cube specimens were stored in a plastic tank of saturated limewater for 28 days. After that the specimens
were immersed in plastic tanks containing sulfate solutions. The sodium sulfate (NS) solution used in this

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

study contained 50 g of NS dissolved in 1.0 liter of water (SO42- of 33,800 ppm or 5% by weight of solution)
whereas 42.36 g of magnesium sulfate (MS) was used to
prepare the magnesium sulfate solution in order to obtain the same concentration of SO42- as that of the NS
solution. The solutions were mixed 24 hours before use,
and stored at a constant temperature of 302 oC. The
volume ratio of the sulfate solution to specimens in each
storage container was approximately 4 to 1. The solutions were replaced every two months of exposure.
Expansion was measured on mortar bar specimens
according to ASTM C1012. After 28 days of curing in
saturated lime water, the initial length of the mortar bar
specimens was measured by using a length comparator.
Subsequently, they were placed in the sulfate solutions
and the length change was measured at various ages of
exposure. A value of expansion was obtained from the
average of three specimens. In addition, the mortar cube
specimens immersed in MS solution were subjected to
weight loss testing by measuring the difference between
the initial weight of the specimen after 28-day immersion in water and the weight of the specimen after immersion in sulfate solutions, indicated in % of the initial
weight.

Amount of Ca(OH) 2 , %

20.0
15.0

10.9
10.0
5.0

OPC1

ILC10

ILC20

Fig. 4 Amount of Ca(OH)2 at the age of 91 days by TGA

analysis of OPC 1 paste and pastes made from ILC10
and ILC20 (Sirisawat et al. 2014).

3. Results and discussion

3.1 Expansion
3.1.1 Expansion in NS of interground limestone
and limestone powder replacing cement mortars
Relationships between the expansion and period of immersion in NS solution of OPC and blended cement
mortar bar specimens with w/b of 0.40 and 0.55 are
shown in Fig. 3 (Sirisawat et al. 2014). These data indicated that, for specimens placed in NS solution, the expansion of OPC 1 cement mortar was, as generally
known, higher than that of an OPC 5 cement specimen.
A smaller quantity of C3A of OPC 5 cement (1.7%) than
that of OPC 1 (5.5%) leads to less ettringite. At higher
w/b (0.55), the effect of composition of cement becomes
more significant. For the specimens made from interground limestone cement (ILC), the expansion of ILC10
mortar bar specimen was significantly larger than that of
ILC20 mortar bar specimen but still lower than that of
the OPC 1 mortar. This is partly because of lower
Ca(OH)2 when comparing between the amount of
Ca(OH)2 in an ILC specimen (ILC10, ILC20) and that
of the OPC 1 specimen, as shown in Fig. 4 (Sirisawat et
al. 2014). The incorporation of limestone also reduces
the amount of C3A in the ILC.
For LP replacing cement mortars (with 10% and 20%

M 5 -0.40
M L20 -0.40

3.00

M1 -0.55

M5 -0.55

ML10 -0.55

ML20 -0.55

3.00
NS

NS

2.50
Expansion, %

2.50
Expansion, %

18.0

0.0

2.3 Microstructure observations and product

characterization
Before initial exposure and after a certain period of exposure, the paste specimens were collected from the
containers and washed in acetone to stop further hydration. In order to determine the mechanisms of sulfate
attack of mixtures with interground and replacing cements, microstructure observations and product characterization were conducted. X-ray Diffraction (XRD)
with Rietveld analysis was used to identify phases and
amount of the products formed by sulfate attack. Mercury Intrusions Porosimetry (MIP) was used to measure
total porosity and average pore size, and Thermogravimetric Analysis (TGA) was used to determinate
Ca(OH)2 content in the specimens. Scanning Electron
Microscopy with Back-Scatter Electron Mode (SEM/
BSE) with EDX analysis was utilized to identify formation of products such as gypsum, M-S-H, and CaCO3.
M1 -0.40
ML10 -0.40

18.1

406

2.00
1.50
1.00
0.50

2.00
1.50
1.00
0.50

0.00

0.00

-0.50

-0.50
0

200

400

600

800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

200

400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

Fig. 3 Relationship between expansion and period of immersion in NS solutions of OPC and interground limestone cement mortars with w/b of 0.40 and 0.55.

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014
M 1 -0.40
M 1 90 LP10-0.40

M1 -0.55
M1 90 LP10-0.55

M 5 -0.40
M 1 80 LP20-0.40

M5 -0.55
M1 80 LP20-0.55

3.00

3.00
2.50

2.50

NS

NS

2.00

2.00

Expansion, %

Expansion, %

407

1.50

1.50

1.00

1.00

0.50

0.50

0.00

0.00

-0.50

-0.50
0

200

400

600

800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

200

400

600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

(a) w/b = 0.40

(b) w/b = 0.55
Fig. 5 Relationship between expansion and period of immersion in NS solutions of OPC and limestone replacing cement
mortars with w/b of 0.40 and 0.55.
ML10 -0.40
ML20 -0.40

3.00

ML10 -0.55
ML20 -0.55

3.00

NS

2.50
2.00
1.50
1.00
0.50

2.00
1.50
1.00
0.50

0.00

0.00

-0.50

-0.50
0

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

M1 90 LP10-0.55
M1 80 LP20-0.55

NS

2.50
Expansion, %

Expansion, %

M1 90 LP10-0.40
M1 80 LP20-0.40

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

(a) w/b = 0.40

(b) w/b = 0.55
Fig. 6 Relationship between expansion and period of immersion in NS solutions of interground limestone cement and
limestone replacing cement mortars with w/b of 0.40 and 0.55

limestone powder by weight), the expansion of 10% and

20% LP replacing cement mortars immersed in NS solution was significantly smaller than that of the OPC 1
mortars and was close to or even smaller than that of the
OPC 5 mortars, as shown in Figs. 5 (Sirisawat et al.
2013). This is similar to the case of interground limestone cement, because OPC 1 cement replaced with
limestone powder has a smaller quantity of C3A and
produces less Ca(OH)2 than OPC 1 due to the partial
replacement of limestone powder.
3.1.2 Expansion comparison between interground limestone and limestone powder replacing cement mortars in NS
For mortar bar specimens made from interground limestone cement (ILC) and LP replacing cement (with 10%
and 20% limestone by weight), the relationships between expansion and period of exposure in NS solution
are shown in Fig. 6. The expansion of interground limestone cement was higher than that of the LP replacing
cement specimens. This is due to smaller average pore
size and lower total porosity of the ILC paste, as can be
seen in Figs. 7(a) and (b), causing smaller spaces for
deposition of the expansion products in the paste matrix
of the ILC mortars.

3.1.3 Expansion comparison between mortar

specimens immersed in NS and MS
Figure 8 shows the comparison of expansion in NS and
MS solutions of limestone replacing cement mortars
(Sirisawat et al. 2013). The figures show that the expansion in MS solution of these specimens was higher than
that in NS solutions when compared at the same immersion period. This is in contrast to the behaviors of other
types of binder system that usually produce larger expansion in NS solution than in MS solution. This is because MS solution decreases the pH of the system. For
the system with LP, there occurred dissolution of CaCO3
from the limestone powder that contributed to gypsum
formation (Irassar et al. 2003; Bonen et al. 1992), which
is confirmed by SEM in Figs. 9(a) and (b). This increases the expansion of the specimens.
3.2 Weight loss
3.2.1 Weight loss in NS of interground limestone and limestone powder replacing cement
mortars
Figure 10 presents the effect of blending method and
content of binders on weight loss of mortar cube specimens with w/b of 0.55 immersed in NS solutions for
1,700 days. It is obvious from this figure that no significant weight loss in each mixture was found. Some

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

Intergrinding

Replacing

Intergrinding

Replacing

0.04

25

27.31

30

25.96

25.42

Total Porosity, %

0.064

0.056

0.055

0.06

0.060

35

0.08

27.76

40

0.10
Average pore size (m)

408

20
15
10

0.02

0.00

0
10%LP

10%LP

20%LP

20%LP

(a) Average pore size

(b) Total porosity
Fig. 7 Average pore size and total porosity at the age of 91 days by MIP analysis of interground limestone and limestone
powder replacing cement pastes.
NS-0.55

NS-0.55

M S-0.55

M1 80 LP20

M1 90 LP10

1.50
Expansion, %

1.50
Expansion, %

MS-0.55

2.00

2.00

1.00
0.50

1.00
0.50
0.00

0.00

-0.50

-0.50
0

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

(a) 10% replacement

(b) 20% replacement
Fig. 8 Effect of type of sulfate on expansion of LP replacing cement mortars with w/b of 0.40 and 0.55.

Gypsum

Gypsum

M-S-H

M-S-H

a) ILC cement paste

(b) LP replacing cement paste
Fig. 9 SEM photographs of interground limestone and limestone powder replacing cement paste showing the formation
of M-S-H and gypsum near the exposed surface.

specimens in NS solution still gained weight or had very

low weight loss although the test period had reached
1,700 days. These weight loss results do not provide any
useful information for evaluation of NS resistance as the
mechanism of NS attack is mainly the ettringite formation which results in expansion rather than weight loss
(Al-Amoudi et al. 1995).
3.2.2 Weight loss in MS of interground limestone and limestone powder replacing cement
mortars
For MS solutions, relationships between the weight loss
and period of immersion of interground limestone and

limestone powder replacing cement mortar cube specimens are shown in Figs. 11 and 12, respectively. These
data indicated that, for specimens placed in MS solution,
ILC cement specimens showed smaller weight loss than
the limestone replacing cement specimens did with their
weight loss equivalent to that of the OPC 5 specimens.
In addition, the ILC cement with 20% LP gave smaller
weight loss than the ILC cement with 10% LP.
In MS solution, the ingress and diffusion of magnesium ions (Mg2+) into a specimen are limited because of
their low diffusivity, causing deteriorated products such
as gypsum and M-S-H near the surface of specimen
(Irassar et al. 2003; Bonen et al. 1992; El-Hachem et al.

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

2012). MgSO4 reacts with Ca(OH)2, forming gypsum

(CaSO4.2H2O) and brucite (Mg(OH)2) near the surface.
As conversion of Ca(OH)2 to Mg(OH)2 takes place, pH
in the cement matrix is gradually reduced. When pH
decreases, C-S-H is unstable and easy to decalcify into

M-S-H, causing a reduction in binding properties (Lee

2007). Eqs. (1) to (3) present the mechanism of MS attack (Cohen and Bentur 1988).
Ca(OH) 2 + MgSO 4 + 2H 2 O CaSO 4 2H 2 O + Mg(OH) 2 (1)

2.0
At 1,700 days of immersion
Weight Loss (%)

1.0

0.60

0.42

0.30

-1.0

-0.45

-0.87

-2.0

Fig. 10 Effect of blending method and content of limestone powder on weight loss of mortars in NS solutions
(w/b of 0.55).
M 1 -0.40
M L10 -0.40

7.0

(2)

4Mg(OH) 2 + SiO 2 (H 2 O)11

(MgO) 4 SiO 2 (H 2 O)8.5 + (n 4.5)H 2 O

(3)

M 1 -0.55
M L10 -0.55

7.0

3.0
1.0
-1.0

M 5 -0.55
M L20 -0.55

MS

5.0
Weight loss, %

Weight loss, %

M 5 -0.40
M L20 -0.40

MS

5.0

xCaO ySiO 2 zH 2 O + xMgSO 4 + (3 x + 0.5 y z )H 2 O

xCaSO 4 2H 2 O + xMg(OH) 2 + 0.5 y (SiO 2 ) 2 H 2 O

For a system with limestone, the solubility of CaCO3

is a function of the pH value. The MS attack was found
to decrease the pH causing higher dissolution of CaCO3,
as shown in Eq. (4), which contributes to gypsum formation, as in Eq. (5) and Figs. 9(a) and (b). Magnesite
in Eq. (6) and dolomite in Eq. (7) (Sirisawat et al. 2014;
Sirisawat et al. 2013) were also formed in the limestone
powder specimens which were exposed to MS solution.
Formation of gypsum, magnesite, and dolomite resulted
in less weight loss and less severe surface etching of
ILC specimens since the formation mitigated the conversion of C-S-H to M-S-H in MS attacking mechanisms. Due to magnesite and dolomite formation, the
Mg2+ ions in pore solution were reduced (Eqs. (1) and

0.0
-0.29

409

3.0
1.0

-1.0
0

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

(a) w/b = 0.40

(b) w/b = 0.55
Fig. 11 Relationship between the weight loss and period of immersion in MS solutions of OPC and interground limestone
cement mortars.
M1 -0.40
M1 90 LP10-0.40

M1 -0.55
M1 90 LP10-0.55

M5 -0.40
M1 80 LP20-0.40

7.0

7.0

MS

MS

5.0
Weight loss, %

5.0
Weight loss, %

M 5 -0.55
M 1 80 LP20-0.55

3.0
1.0

3.0
1.0

-1.0

-1.0
0

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, days

(a) w/b = 0.40

(b) w/b = 0.55
Fig. 12 Relationship between the weight loss and period of immersion in MS solutions of OPC and limestone replacing
cement mortars.

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

Dolomite

Magnesite

3.2.3 Weight loss comparison between interground limestone and limestone powder replacing cement mortars in MS
Figure 15 shows that the ILC cement mortars exhibit
lower weight loss than the LP replacing cement specimens do. This is because ILC cement specimens had
lower average pore size and total porosity, so they are
denser than the LP replacing cement specimens as mentioned before (see Fig. 7). Specimens made from ILC
cement lost less weight than those made from the LP
replacing cement.

Amount of products, %

0.50
0.41
0.40

0.34
0.28

0.30
0.20
0.10

0.07

0.05

0.00

0.00

OPC1

ILC10

4. Conclusions

ILC20

Fig. 13 Amount of magnesite and dolomite by XRD

analysis of OPC1 paste and interground limestone cement pastes (ILC10 and ILC20) (Sirisawat et al. 2014).

(2)), causing a reduction of M-S-H (Eq. (3)).

Dissolving of CaCO3
CaCO3 (s )
Ca 2+ + CO 3

(4)

+ 2H 2 O
CaSO 4 2H 2 O

(5)

Gypsum formation
Ca 2+ + SO 4

Magnesite formation
Mg 2+ + CO 3

MgCO 3

(6)

Dolomite formation
CaCO 3 + Mg 2+
Ca [Mg(CO 3 )2 ]

(7)

The formation of magnesite and dolomite of systems

with LP in MS solution was confirmed by the results of
XRD analysis as shown in Fig. 13 (Sirisawat et al.
2014). It was found that much more magnesite and
dolomite were detected in LP cement paste specimens
than in the OPC specimen.
Another reason supporting the lower weight loss of
limestone powder specimens than that of the OPC
specimen was that the precipitation layer of calcite was
found on the exposed surface of the ILC and LP replacing cement paste, as shown by SEM images in Figs.
14(a) and (b). This layer helped delay the ingress of
magnesium ions into the paste.

Ca(CO)3

410

The study on sulfate resistance of mortars made from

interground limestone and limestone powder replacing
cements indicated that the limestone replacing cement
had advantages over the interground limestone cement
in NS solution since it is less susceptible to the expansion due to higher porosity, which increased available
spaces for ettringite deposition. On the other hand, the
specimens made from interground limestone cement lost
less weight than those made from the limestone replacing cement because of lower average pore size and total
porosity, making the specimens denser. For mortars containing limestone powder in MS solution, the MS decreased the systems pH. There occurred higher dissolution of CaCO3 from the limestone powder, which contributed to more gypsum formation as well as resulted in
formation of magnesite and dolomite. This mitigated the
conversion of C-S-H to M-S-H, and then resulted in less
weight loss and less severe surface etching than the
specimens made from OPC.
Acknowledgement
This research was supported by research funds from
Srinakharinwirot University, Siam Research & Innovation Co Ltd., National Science and Technology Development Agency (NSTDA), and Center of Excellence in
Material Science, Construction and Maintenance Technology. The research is also partially supported by the
Higher Education Research Promotion and National
Research University Project of Thailand, Office of the
Higher Education Commission and the Thailand Research Fund (grant No. MRG5080440).

Ca(CO)3

(a) ILC cement paste

(b) LP replacing cement paste
Fig. 14 SEM photographs of interground limestone and limestone powder replacing cement paste showing that the external layer consisted of calcium carbonate.

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014
M L10 -0.40

M L10 -0.55

M 1 90 LP10-0.40

5M

5M
5.0
Weight loss, %

5.0
Weight loss, %

M 1 90 LP10-0.55

7.0

7.0

3.0

1.0

3.0

1.0

-1.0

-1.0
0

200 400 600 800 1,000 1,200 1,400 1,600 1,800

Immersing time, day

200

(a) 10% replacement, w/b = 0.40

M L20 -0.40

400

600

800 1,000 1,200 1,400 1,600 1,800

Immersing time, day

(b) 10% replacement, w/b = 0.55

M 1 80 LP20-0.40

ML20 -0.55

7.0

M 1 80 LP20-0.55

7.0
5M

5M
5.0

5.0
Weight loss, %

Weight loss, %

411

3.0
1.0
-1.0

3.0

1.0

-1.0
0

200 400 600

800 1,000 1,200 1,400 1,600 1,800

Immersing time, day

(c) 20% replacement, w/b = 0.40

200

400 600

800 1,000 1,200 1,400 1,600 1,800

Immersing time, day

(d) 20% replacement, w/b = 0.55

Fig. 15 Relationship between the weight loss and period of immersion in MS solution of interground limestone and limestone powder replacing cement mortars (10% and 20% replacement).

References
ACI 116R-00, (2000). Cement and concrete
terminology. Farmington Hills, MI, USA: American
Concrete Institute.
Al-Amoudi, O. S. B., (1998). Sulfate attack and
reinforcement corrosion in plain and blended cements
exposed to sulfate environments. Building and
Environment Journal, 33(1), 53-61.
Al-Amoudi, O. S. B., Maslehuddin, M. and Saadi, M.
M., (1995). Effect of magnesium sulfate and sodium
sulfate on the durability performance of plain and
blended cements. ACI Materials Journal, 92(1), 1523.
ASTM C 1012. Test method for length-change of
hydraulic cement mortars exposed to a sulfate
solution. Annual Book of ASTM Standards.
Baingam, L., Lawtrakul, L., Saengsoy, W., Sirisawat, I.
and Tangtermsirikul, S., (2012). Study on
mechanism of magnesium sulfate attack on hardened
cementitious system by using computation methods.
In: The Pure and Applied Chemistry International
Conference (PACCON 2012), Chiang Mai, Thailand.
Berry, E. E. and Malhotra, V. M., (1980). Fly ash for
use in concrete. ACI Journal, 77(8), 59-73.
Bonen, D. and Cohen, M. D., (1992). Magnesium
sulfate attack on Portland cement paste II. Chemical
and mineralogical analyses. Cement and Concrete
Research, 22, 707-718.
Bonen, D. and Cohen, M. D., (1992). Magnesium
sulfate attack on Portland cement paste. I. Micro-

structural analysis. Cement and Concrete Research,

22, 169-180.
Cohen, M. D. and Bentur, A., (1988). Durability of
Portland cement-silica fume pastes in magnesium
sulfate and sodium sulfate solutions. ACI Materials
Journal, 85(3), 148-157.
El-Hachem, R., Rozire, E., Grondin, F. and Loukili, A.,
(2012). Multi-criteria analysis of the mechanism of
degradation of Portland cement based mortars
exposed to external sulphate attack. Cement and
Concrete Research, 42(10), 1327-1335.
Heikal, M., Helmy, I., El-Didamony, H. and El-Raoof,
F.A., (2004). Electrical properties, physico-chemical
and mechanical characteristics of fly ash-limestonefilled pozzolanic. Cement Ceramics-Silikty, 48(2),
49-58.
Irassar, E. F., Bonavetti, V. L. and Gonzalez, M., (2003).
Microstructural study of sulfate attack on ordinary
and limestone Portland cements at ambient temperature. Cement and Concrete Research, 33, 31-41.
Kaewmanee, K., Tangtermsirikul, S. and Ratchayut, K.,
(2006). Effect of particle size of limestone powder
on properties of normal concrete and special
concrete. In: Proceedings of the 11th National
Convention on Civil Engineering, Engineering Institute of Thailand, Phuket, Thailand, CD Rom.
Krammart, P., Wattanakul, Y., Sirisawat, I. and
Tangtermsirikul, S., (2011), Sulfate resistance of
concrete with fly ash and limestone powder. In:
Proceedings of Annual Concrete Conference 7,

I. Sirisawat, L. Baingam, W. Saengsoy, P. Krammart and S. Tangtermsirikul / Journal of Advanced Concrete Technology Vol. 12, 403-412, 2014

Rayong, Thailand, MAT 226-235.

Lee, S.-T., (2007). Performance deterioration of
Portland cement matrix due to magnesium sulfate
attack. KSCE Journal of Civil Engineering, 11(3),
157-163.
Mehta, P. K. (1975). Evaluation of sulfate-resisting
cements by a new test method. ACI Journal, ACI
Digest paper 72-40, 573-575.
Miller, D. G. and Manson, P. W., (1951). Long-time
tests of concretes and mortars exposed to sulfate
waters. University of Minnesota, Agricultural
Experiment Station, USA, Technical Bulletin 194.
Ponpo, N., Krammart, P., Sirisawat, I. and
Tangtermsirikul, S., (2010). Expansion of mortar
with fly ash and limestone powder in sodium sulfate
solution. In: Proceedings of Annual Concrete
Conference 6, Phetcha Buri, Thailand, MAT 399-405.
Rachel, J. D., Javed, L. B. and Sankar, B., (1996).
Supplementary cementing materials for use in
blended cements. Research and development bulletin
RD111T.
Sirisawat, I., Saengsoy, W., Baingam, L., Krammart, P.
and Tangtermsirikul, S., (2014). Durability and
testing of mortar with interground fly ash and
limestone cements in sulfate solutions. Construction
and Building Materials, 64, 39-46.
Sirisawat, I., Saengsoy, W., Baingam, L., Krammart, P.
and Tangtermsirikul, S., (2013). Test methods for
evaluation of sulfate resistance of limestone powder
replacing cement mortars. In: International
Symposium on New Technologies for Urban Safety of

412

Mega Cities in Asia (USMCA 2013), Hanoi, Vietnam,

12 p.
Sukayanudist, S., Krammart, P., Sirisawat, I. and
Tangtermsirikul, S., (2009). Effect of limestone
powder on expansion of mortar in sodium sulfate
solution. In: Proceedings of Annual Concrete
Conference 5, Nakorn Ratchasima, Thailand, MAT
241-248.
Tangtermsirikul, S., (2003). Durability and mix design
of concrete. 1st edn. Printing House of Thammasat
University, Pathum Thani, Thailand.
Thasanakosol, S., Chatveera, B. and Tangtermsirikul, S.,
(1999). Durability of limestone powder concrete.
In: Proceedings of the 5th National Convention on
Civil Engineering, Engineering Institute of Thailand,
Patthaya, Thailand, MAT 122-127.
TIS15 P1-2004, (2005). Standard for Portland
cement. Thai Industrial Standard Institute, Thailand.
Tsivilis, S., Chaniotakis, E., Kakali, G. and Batis, G.,
(2002). An analysis of the properties of Portland
limestone cements and concrete. Cement and
Concrete Composites, 24, 371-378.
Tsivilis, S., Voglis, N. and Photou, J., (1999). A study
on the intergrinding of clinker and limestone: Technical note. Minerals Engineering, 12(7), 837-840.
Wattanakul, Y., Krammart, P., Sirisawat, I. and
Tangtermsirikul, S., (2010). Cementitious properties
and expansion in sodium sulfate of concrete with fly
ash and limestone powder. Journal of Engineering,
RMUTT, vol.2, 1-9.