Accepted Manuscript

Performance of passive methods in plastic shrinkage cracking mitigation

Sadegh Ghourchian, Mateusz Wyrzykowski, Luis Baquerizo, Pietro Lura

PII: S0958-9465(18)30122-7
DOI: 10.1016/j.cemconcomp.2018.05.008
Reference: CECO 3061

To appear in: Cement and Concrete Composites

Received Date: 2 February 2018

Revised Date: 1 May 2018
Accepted Date: 7 May 2018

Please cite this article as: S. Ghourchian, M. Wyrzykowski, L. Baquerizo, P. Lura, Performance of
passive methods in plastic shrinkage cracking mitigation, Cement and Concrete Composites (2018), doi:
10.1016/j.cemconcomp.2018.05.008.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

1 Performance of passive methods in plastic shrinkage

2 cracking mitigation
3 Sadegh Ghourchian1,2,*, Mateusz Wyrzykowski1, Luis Baquerizo3, Pietro Lura1,2

PT
1
4 Empa, Swiss Federal Laboratories for Materials Science and Technology, Dübendorf,
5 Switzerland
2
6 Institute for Building Materials (IfB), ETH Zurich, Switzerland

RI
3
7 LafargeHolcim Research Centre, Saint-Quentin Fallavier, France

SC
*
8 Corresponding author: sadegh.ghourchian@empa.ch

9 Abstract

U
10 The occurrence of plastic shrinkage cracking in fresh concrete may significantly compromise its
11 service life. Therefore, it is essential to reduce the crack width with appropriate, efficient and
AN
12 effective methods. This study is dedicated to comparing the ability of passive mitigation methods
13 including: Shrinkage Reducing Admixtures (SRA), accelerators based on Calcium Silicate
14 Hydrate seeding agents and curing compounds, in controlling the plastic shrinkage cracking in
M

15 the drying state. The performance of the different admixtures was evaluated according to the
16 ASTM C1579-13 standard. Furthermore, a series of complementary test methods were utilized to
17 study the mechanism of action of the different admixtures. Among the crack mitigation methods
D

18 that were studied in this paper, SRA were the most effective.
TE

19 Keywords:

20 Fresh Concrete; Plastic Shrinkage Cracking; Mitigation Methods; Admixtures; Additives

EP

21

22
C

23
AC

24

25

26

27

28
 ACCEPTED MANUSCRIPT

29 1. Introduction

30 Fresh concrete undergoes volumetric shrinkage, referred to as plastic shrinkage, due to the
31 effects of the gravity (in the bleeding state) [1,2] and pore water evaporation (in the drying state)
32 [3,4], from casting to the time of final set. Since structural concrete is always restrained (e.g., by

PT
33 reinforcement, differential depth or knobs and dents on the substrate base, and friction between
34 fresh concrete and substrate base), incompatibility of deformations occurs, which leads to the

RI
35 buildup of restraint stresses. During the above-mentioned period, concrete is plastic, possessing
36 insignificant strength, which makes it highly susceptible to cracking when the stresses increase

SC
37 beyond the strength limit [5]. Accordingly, it is mentioned that plastic shrinkage is the source of
38 roughly 80% of the early-age cracking of reinforced concrete structures [6]. The cracks that
39 appear on the surface may grow to widths larger than 1 mm and lengths of 50 to 1000 mm, and

U
40 may be spaced in an irregular pattern from 50 to 700 mm apart [7]. In shallow elements (e.g.
AN
41 concrete slabs) they can even grow throughout the whole element depth (see [8]). The cracks are
42 unsightly [8], and furthermore accelerate the ingress of harmful substances (e.g. chloride ions)
43 into the hardened concrete [9], leading to the reduction of serviceability. Therefore, they need to
M

44 be eliminated or at least their width needs to be reduced with effective methods [10–12].
D

45 The mitigation methods available in the engineering practice are divided into two groups: active
46 and passive solutions. The active solutions, e.g. the ACI method, are mainly based on the
TE

47 evaporation rate of the free bleeding water [13]. When the rate of water evaporation is higher
48 than the proposed critical evaporation rate (the latter derived based on a typical bleeding rate of
EP

49 concrete), necessary actions need to be taken. Examples of possible actions are external curing,
50 i.e. by wetting of the concrete surface, or covering the surface by plastic sheets [14]. Recently, an
C

51 improved method [15] was proposed to complement the ACI approach to mitigate plastic
52 shrinkage cracking in the drying state, which is based on ensuring that the total amount of bleed
AC

53 water is larger than the amount of evaporated water until the time of final set, instead of
54 evaluating the evaporation/bleeding rates only.

55 On the other hand, passive solutions to mitigate plastic shrinkage cracking rely on modifying the
56 material properties (e.g. by chemical admixtures), regardless of the environmental parameters,
57 i.e. the evaporation rate. It was observed that accelerators tend to reduce the cracking risk
58 [11,16,17], while on the contrary, retarders extend the time of set, extending the time window
 ACCEPTED MANUSCRIPT

59 when cracking may appear [18],and therefore generally tend to increase the cracking risk
60 [16,17,19]. It was also found that, with the reduction of mixing water (made possible by the use
61 of water reducing admixtures, i.e. plasticizers), the capillary pressure development starts almost
62 immediately after placing [17]. The fast pressure development can increase the cracking
63 tendency considerably, since some plasticizers are also known to retard cement hydration [16].

PT
64 Shrinkage reducing admixtures (SRA), i.e. surfactants, have been shown to be very effective in
65 controlling plastic shrinkage cracking in the evaporation state [10,11,16,20–22]. They reduce the

RI
66 surface tension of pore water and consequently, reduce the capillary pressure and plastic
67 shrinkage. However, precautions are needed in using the SRA, since they may reduce the

SC
68 compressive strength and retard the development of mechanical properties at high dosages
69 [22,23] and they are in general expensive. Therefore, the dosage of SRA needs to be optimized.

U
70 Viscosity modifying agents (VMA) [24], i.e. stabilizers [11], primarily affect the viscosity of the
71 pore fluid, while they also can reduce the surface tension of the pore fluid, but much less than
AN
72 SRA [11,24]. When they contact water, they swell first, and then dissolve, producing a viscous
73 solution [24]. Since VMA decrease the coefficient of permeability significantly, by increasing
M

74 the viscosity of pore fluid, they decrease the bleeding rate (i.e. the rate of vertical/volumetric
75 shrinkage) [24], and consequently, they can be very effective in controlling the cracks in the
D

76 bleeding state in deep elements. Reference is made to [25] for the relationship between bleeding
77 rate and coefficient of permeability. Similar to SRA, VMA also can retard the evolution of
TE

78 mechanical properties [11], thus their dosage needs to be optimized. Evaporation reducers
79 [14,26] (i.e. curing compounds [11]), e.g. paraffin dispersion, can form a film on the surface of
EP

80 fresh concrete and reduce the evaporation of pore fluid, and plastic shrinkage cracking. Leemann
81 et al. [11] confirmed that they are effective also as a concrete admixture, mixed in the volume of
C

82 concrete, and not only as a curing compound applied on the surface, as proposed in ACI 302.1R-
83 04 [14]. Moreover, if they are used as an admixture, it can be assured that the curing agent
AC

84 accumulates on the whole surface of the concrete by bleeding, and decreases the evaporation by
85 forming a curing film on the surface of the concrete.

86 It has been shown that pre-wetted lightweight aggregates (e.g. expanded shale, used in coarser
87 fraction that the river sand they substituted) can reduce plastic shrinkage cracking in the drying
88 state [27]. The mechanism of action is by providing the water required for evaporation [21] and
89 thus preventing the increase of the capillary pressure. However, utilizing very fine light-weight
 ACCEPTED MANUSCRIPT

90 aggregates (i.e. finer than the sand they substitute) might not mitigate cracking, since the
91 negative aspect of the increase of the stiffness in the dormant hydration period, caused by the
92 fine particles of the lightweight aggregates, may compensate the effect of the water provided by
93 internal curing. See [5,28] for the effect of the stiffness in the dormant hydration period on
94 plastic shrinkage cracking.

PT
95 Fibers (e.g. polypropylene fibers) have been utilized in fresh concrete to reduce the size and the

RI
96 area of the plastic shrinkage cracks in many studies [7,29–44] and they are also recommended by
97 ACI 302.1R-04 [14]. Their mechanism of action is by: 1- Reduction of bleeding by enhancement

SC
98 of the concrete stiffness caused by the interlocking network of fibers [29], therefore beneficial
99 against cracking in the bleeding state. However, the latter accelerates the capillary pressure (i.e.
100 the effective stress) increase in the evaporation state (see [28]). 2- Increase of the cohesion, and

U
101 therefore tensile strength [29], which can prevent or retard crack propagation in the evaporation
AN
102 state. Nevertheless, fibers are not able to prevent crack nucleation, as explained in [45]. Similar
103 conclusions can be made according to the simulations performed in [5], as well.
M

104 It has been shown that Calcium Silicate Hydrate (C-S-H) seeds act as cement hydration
105 accelerator [46–48], shortening the dormant period and increasing the rate of hydration.
D

106 According to Nicoleau [48], the efficiency of C-S-H seeding is less sensitive to the cement
107 composition than other conventional accelerators. While no results have been published so far, it
TE

108 is expected that C-S-H seeding may be able to mitigate plastic shrinkage cracking, similarly as
109 for the accelerators mentioned above [11,49].
EP

110 In order to choose the most efficient and robust mitigation method, the performance of the
111 above-mentioned passive solutions needs to be compared and discussed. Therefore, in this study
C

112 the performance of different passive solutions, including a curing compound, SRA and C-S-H
AC

113 seeding, were compared. The crack size distribution was obtained with the ASTM C1579-13
114 standard [50], while a DIC (Digital Image Correlation) system was used to measure the free and
115 restrained deformations in the plastic state. The temperature evolution was monitored by a semi-
116 adiabatic temperature measurement device, in the bleeding state on mortar samples
117 corresponding to the tested concretes. Furthermore, thermocouples and Vicat needle were
118 utilized to record the temperature change and setting times, respectively, in the drying state.
119 Finally, the mechanism of action of the above-mentioned additives was explained extensively,
 ACCEPTED MANUSCRIPT

120 according to the novel poromechanics model proposed in a previous study [28], utilizing the
121 above-mentioned complementary data and pore pressure measurements.

122

123 2. Materials and methods

PT
124 2.1. Materials and mix compositions
125 The used cement was a CEM II/A-LL 42.5 N with density and Blaine fineness, 3060 kg/m3 and

RI
126 401 m2/kg, respectively. The mix compositions are presented in Table 1. The names of the
127 concrete mixes indicate the type of utilized crack-reducing additive. CURE is the concrete with a

SC
128 paraffin-based curing agent (BASF Masterlife IC100), added in the mixer as an admixture.
129 CSRA is the concrete containing SRA (Sika Control 60). SEED is the concrete with the C-S-H

U
130 seeds (X-Seed 100 by BASF). The amounts of the additives were based on the values
AN
131 recommended by the producers. A polycarboxylate-based superplasticizer was added in an
132 amount allowing for obtaining similar consistency for all analyzed mixtures, see Table 1. The
133 amount of required superplasticizer increased considerably for the CURE concrete. In this case,
M

134 adjusting the needed amount in steps produced no satisfactory slump until a rapid increase in
135 slump (110 mm compared to 61 mm on average for other concretes), while the differences in
D

136 flow were much lower. The individual mixes had flow different by 28 mm at most from the
TE

137 average of 468 mm, see Table 1. The particle sizes of sand and gravel used as aggregates were 0-
138 4 mm and 4-16 mm, respectively. All concretes had the same cement content and water-to-
139 cement ratio (w/c). All the constituents were stored in a climate-controlled room at 20±0.3 °C for
EP

140 a minimum of 12 h prior to mixing. The experimental procedure was carried out within a 4-days’
141 time window. Slump and flow were measured according to EN 12350-2:2009 [51] and EN
C

142 12350-5:2009 [52], respectively. The density of the fresh concrete was also determined using a
143 13-l cylinder.
AC

144 Table 1. Mix compositions

Concrete REF SEED CURE CSRA
3
Cement (kg/m ) 348 348 348 348
3
Water (kg/m ) 174 163.8 171 176
3
Sand [0-4 mm] (kg/m ) 927 927 927 927
3
Gravel [4-16 mm] (kg/m ) 928 928 928 928
 ACCEPTED MANUSCRIPT

Superplasticizer (kg/m3) 1.88 1.66 3.132 1.89

Additive amount (% bwoc) 0 3.7 1.5 1.5
3
Additive amount (kg/m ) 0 12.87 5.22 5.22
w/c [-] 0.5 0.5 0.5 0.5
Slump (mm) 59 60 110 65

PT
Flow (mm) 460 440 490 480
3
Measured density (kg/m ) 2321 2280 2348 2300
145

RI
146 2.2. Plastic shrinkage

SC
147 2.2.1. Environmental conditions

148 A climate-controlled room was utilized to maintain stable temperature and humidity conditions

U
149 (33±0.5 °C and 40±5 %RH) during the test. An air blower was set to produce wind velocity of
AN
150 7.5 ± 0.5 m/s over the top surface of the specimens. A purposefully-designed channel was
151 utilized to obtain uniform wind speed on the samples (see [49]). The mass change and the
152 temperature of pure water in the above-mentioned environment, placed in a pan with 90 mm
M

153 depth with initial temperature of 21 °C, is shown in Fig. 1. The surface had higher initial
154 temperature because it was in direct contact with 33 °C air, while the temperature at the bottom
D

155 of the pan was lower. At first, the temperature of the water started to increase until a temperature
TE

156 of about 25 °C (which was lower than the room temperature but substantially higher than the
157 dew point temperature of 16 °C). After about 2 h, the temperature remained almost stable (with
158 only a slow decrease over time) and the rate of evaporation increased and then remained almost
EP

159 constant. The overall evaporation rate of pure water in the above-mentioned environment was
160 1.1 kg/m2/h. This fulfills the conditions of the necessary evaporation rate prescribed in the
C

161 ASTM standard (minimum 1.0 kg/m2/h). A pan with lower depth would likely reach constant
evaporation rate faster, because of faster temperature equilibration.
AC

162
 ACCEPTED MANUSCRIPT

Time (h)
0 1 2 3 4 5 6
26 0

25 -1

MASS CHANGE (kg/m2)

Temperature (°C)

PT
24 -2

23 -3

RI
22 -4

SC
21 Temperature at 20 mm below surface -5
Temperature at 90 mm below surface
Mass
20 -6

U
163
164 Fig. 1. Mass change and temperature evolution of pure water in the test environment, placed in a
AN
165 pan with 90 mm depth.
166
M

167 2.2.2. Measurements of plastic shrinkage and cracking

168 The susceptibility of concrete mixtures to plastic shrinkage cracking was studied according to
D

169 ASTM C1579-13 [50], as explained in [49]. Two samples with metal inserts (stress risers) and a
170 sample without any inserts, providing results on the restrained and free shrinkage, respectively,
TE

171 were tested. The assessment was performed by comparing the developed average crack widths
172 (resolving also their distribution along the length of the cracks). Furthermore, the free horizontal
EP

173 and vertical shrinkage, and the horizontal deformation across the crack was measured by DIC
174 (see also [49]). Two pairs of meshes were used as two virtual strain gauges for one of the
175 restrained samples and for the free sample to measure the across crack and horizontal plastic
C

176 shrinkage, respectively. The gauge lengths were equal to 60±1 mm and 150±1 mm for the
AC

177 restrained and unrestrained deformations, respectively. The molds were filled with concrete on a
178 vibrating table (12 s vibration time) and the top surface was finished using a trowel. At each test,
179 two restrained samples per concrete mixture were tested. Every experiment lasted for 6 h after
180 the air blower was started; the cracks were measured at the conclusion of each test. For
181 measuring the crack size, multiple high-resolution pictures (2413±50 ppi) of the crack
182 throughout its length were taken. The crack width was then quantified by an operator with image
 ACCEPTED MANUSCRIPT

183 processing software by measurements of a distance in pixels at resolution 0.01 mm at 10±1 mm

184 intervals throughout the length of the crack. To prevent possible wall-effects on the measured
185 crack widths, cracks widths within 25 mm of the sample edges were not taken into account.
186 The capillary pressure development in the free shrinkage specimen (additional specimen without
187 the stress risers) was monitored by pressure sensors connected to capillary pipes mounted in a

PT
188 mold. The temperature evolution was monitored by thermocouples of resolution ±0.1 °C: two
189 thermocouples at the bottom, two at the surface and two thermocouples at the middle of the free

RI
190 shrinkage samples. The evaporation rate was measured by recording the mass change of the free
191 shrinkage sample. The complete test setup was discussed in detail in [49].

SC
192
193 2.3. Bleeding

U
194 Bleeding tests were performed according to EN 480-4:2005 [53]. A cylindrical container with
195 diameter of 250±10 mm and height of 250±10 mm was filled with concrete, which was
AN
196 compacted by a needle vibrator. The container was kept in the climate-controlled chamber during
197 the parallel plastic shrinkage test and covered to avoid evaporation and desaturation. The water
M

198 accumulated on the samples was collected at 15 min, 30 min, 1 h, and then every hour until 6 h
199 from mixing. The mass of collected water was measured directly after collection. Next, the
D

200 containers with bleed water were dried at 105 °C and weighed again to account for the mass of
201 solid particles collected together with the bleeding water.
TE

202
203 2.4. Semi-adiabatic temperature evolution and Vicat setting times
EP

204 The measurements of the semi-adiabatic temperature evolution of mortars were performed in a
205 climate-controlled room at 20±0.3 °C, within a thermally isolated box, with sample openings
C

206 containing the cylindrical mortar samples with thermocouples placed underneath the samples
207 (HolcimHeat®). The measurements were performed on mortars obtained by sieving the
AC

208 corresponding concretes on a 3.2 mm sieve. This procedure was employed in order to obtain
209 higher content of the cement paste in the material and therefore higher signal, assuming that the
210 removed coarse aggregates had negligible effect on heat release during hydration. 1300 g of the
211 mortar were poured into the containers of the device and two samples were tested for each
212 mixture. The measurements started 80±10 min after mixing.
 ACCEPTED MANUSCRIPT

213 The Vicat test was performed on the mortars obtained in the same way as for the semi-adiabatic
214 calorimetry according to ASTM C807-13 [54] in the same room and conditions as in the plastic
215 shrinkage tests, in the drying state. The needle had 2 mm diameter and mass of 300 g. A
216 cylindrical mold with inner diameter of 76 mm was filled with the sieved mortar and placed in
217 the climate-controlled room under evaporative conditions similar to the samples for plastic

PT
218 shrinkage. Starting at 30 min from the start of mixing, the needle was released every 10 min and
219 its penetration depth was recorded. Before release, the needle was brought in contact with the

RI
220 surface of the mortar. The time when the needle failed to penetrate up to the bottom of the
221 sample was reported as the initial set and the time at which the needle did not penetrate into the

SC
222 surface was reported as the final set.
223

U
224 2.5. Compressive strength

The compressive strength test was performed on 150×150×150 mm3 samples, two samples for
AN
225
226 each test. They were stored in a climate-controlled room (20±0.3 °C, RH>95%) and tested after 7
227 days and 28 days of curing.
M

228 3. Results and discussions

3.1. Bleeding
D

229
230 As shown in Fig. 2, very small bleeding was measured on the REF and CURE concretes, while
TE

231 no bleeding could be measured on the SEED and CSRA concretes. The small bleeding is due to
232 the low coefficient of permeability, caused by the increase of viscosity of pore fluid due to the
EP

233 admixtures, and by the high initial stiffness [55].

C
AC
 ACCEPTED MANUSCRIPT

0.03
REF
CURE

BLEEDING (mm)
0.02

PT
0.01

RI
0

SC
0 1 2 3 4 5 6
234 TIME (h)

235 Fig. 2. Bleeding vs. time measured on the REF and CURE concretes. For the other concretes, no

U
236 bleeding was observed.
3.2. Temperature evolution
AN
237
238 The temperature evolution of the concretes at the bottom (temperatures of the bulk) and at the
239 surfaces are shown in Fig. 3. The presented results are the average from 2 thermocouples with
M

240 standard deviation of less than 0.5 °C. The bulk temperature of the samples became almost
241 identical within the first 4 h after mixing. Afterwards, SEED manifested accelerated temperature
D

242 increase, due to the faster hydration. A very interesting trend of temperature evolution could be
TE

243 observed at the surface of the CURE concrete (concrete with paraffin-based curing agent) at
244 around 4 h from mixing. Before the mentioned time, the surface temperature increased due to the
245 lack of surface water evaporation, while the temperature dropped later, due to a sudden increase
EP

246 of evaporation. The latter is discussed further in the discussion section.

247
C
AC
 ACCEPTED MANUSCRIPT

34 REF
SEED
CURE
32 CSRA

TEMPERATURE (˚C)
30

PT
28

26

RI
24
0 1 2 3 4 5 6

SC
248 TIME (h)

34
REF
SEED
CURE

U
32 CSRA
TEMPERATURE (˚C)

AN
30

28
M

26
D

24
0 1 2 3 4 5 6
TE

249 TIME (h)

250 Fig. 3. Average temperatures at the bottom of the samples (above) and at the surface of the
251 samples (below).
EP

252
253 3.3. Semi-adiabatic temperature evolution and Vicat setting times
C

254 Fig. 4 shows the semi-adiabatic temperature evolutions measured with HolcimHeat®. The
AC

255 influence of the additives on the hydration kinetics was evident. The seeding agent considerably
256 accelerated cement hydration. SRA addition only slightly delayed hydration (CSRA mixture),
257 while the curing agent (CURE mixture) delayed hydration significantly. The latter is concluded
258 by considering that the semi-adiabatic tests were started around 1 h after mixing. All ingredients
259 had identical initial temperature at mixing time, while CURE had the minimum temperature after
260 1 h, and the temperature difference (comparing to REF) started to increase around 4 h after
261 mixing.
 ACCEPTED MANUSCRIPT

262

40 REF
SEED
CURE
36 CSRA
TEMPERATURE (˚C)

PT
32

RI
28

24

SC
20

U
0 2 4 6 8 10 12
263 TIME (h)
AN
264 Fig. 4. Semi-adiabatic temperature evolution of concrete mixtures with and without
265 admixtures
266
267 The obtained setting times by Vicat test are shown in Table 3. The trend is similar to that
M

268 revealed in Fig. 4. However, the capillary pressure stiffening also accelerates setting in the
269 drying state [28]. The initial set measured by Vicat correlates well with the onset of horizontal
D

270 shrinkage as measured by DIC (see section 3.6), as also found before [28,49].
TE

271
272 Table 3. Setting times determined by Vicat test on samples open to drying.
Sample Initial Final
REF 1h 07 min 2h 00 min
EP

SEED 0h 55 min 1h 40 min

CURE 1h 15 min 2h 40 min
CSRA 1h 10 min 2h 20 min
C

273
AC

274
275 3.4. Pore pressure development
276 Fig. 5 shows the pore pressure development in the samples. The sudden jump and pressure
277 release visible in the curves were due to air penetration into the capillary tubes of the pressure
278 sensors. However, it is pointed out that the latter is a highly localized phenomenon and hence
279 might not necessarily correspond to the air-entry time at the concrete surface. In fact,
 ACCEPTED MANUSCRIPT

280 measurement made with several sensors placed within one sample at similar depths show that, as
281 long as the onset and initial rise of capillary pressure are fairly reproducible, the air penetration
282 of the sensor is very variable between different sensors. The capillary pressure developed in
283 SEED earlier than in other concretes. The other two concretes, CSRA and in particular CURE,
284 experienced significant delay in the onset of capillary pressure development.

PT
10 TIME (h)

RI
0 1 2 3 4 5 6
PORE PRESSURE (kPa)

-10

SC
-20

-30

U
-40
AN
-50 REF
SEED
-60 CURE
CSRA
M

-70
285
286 Fig. 5. Pore pressure development vs. time of concrete mixtures with and without admixtures.
287 The curves were utilized to monitor the rate of progress of capillary pressure, while the initial
D

288 values might not be calibrated.

289
TE

290 3.5. Mass change

291 The mass change vs. time is plotted in Fig. 6. All concretes had the same initial evaporation rate,
EP

292 due to the same initial volumetric water content (mixing water) and environmental conditions.
293 The different trends are discussed in the discussion section.
C
AC
 ACCEPTED MANUSCRIPT

TIME (h)
0 1 2 3 4 5 6
0
MASS CHANGE (kg/m2)

-0.4

PT
-0.8

RI
-1.2
REF

SC
SEED
-1.6 CURE
CSRA

-2

U
294
295 Fig. 6. Mass change vs. time of concrete mixtures with and without admixtures.
AN
296
297 3.6. Deformations
298 Strains measured with DIC are shown in Fig. 7 to Fig. 10. The distance between the wire meshes
M

299 across the crack was always 60±1 mm. The slight initial expansion observed in some samples is
300 likely due to thermal expansion. The order of the times of final set from earlier to later was: 1-
D

301 SEED 3h, 2- REF 3.5 h, CSRA 3.7 h, CURE 4.5 h. This confirms that hydration was
TE

302 significantly delayed and accelerated in the CURE and SEED concretes, respectively, (see also
303 Fig. 4). The black dots on the across-crack deformation curves show the crack initiation time.
304 Due to a technical problem (insufficient light that made the pattern on one of the wire meshes
EP

305 recorded by the DIC hardly visible) the results for only one virtual strain gauge were available
306 for the free shrinkage sample. Therefore, the standard deviations for horizontal deformations are
C

307 not available; however, previous studies by the same authors [28,49] showed that the standard
AC

308 deviation of the horizontal shrinkage measurements by the DIC method is below 250×10-6 [-].
 ACCEPTED MANUSCRIPT

10000 Free-horizontal
Free-vertical
7500 Across crack

5000
STRAIN (×10-6 [-])
2500
TIME (h)

PT
0
0 1 2 3 4 5 6
-2500

RI
-5000

-7500

SC
-10000
309
310 Fig. 7. Strain vs. time for REF concrete. The error bars in across crack and free-vertical curves
311 show the standard deviation between two and three measurements, respectively.
10000
U
AN
Free-horizontal
Free-vertical
7500 Across crack

5000
STRAIN (×10-6 [-])

2500
TIME (h)
0
D

0 1 2 3 4 5 6
-2500
TE

-5000

-7500

-10000
EP

312
313 Fig. 8. Strain vs. time for SEED concrete. The error bars in across crack and free-vertical curves
C

314 show the standard deviation between two and three measurements, respectively.
AC
 ACCEPTED MANUSCRIPT

10000
Free-horizontal
Free-vertical
7500 Across crack

5000
STRAIN (×10-6 [-])
2500
TIME (h)

PT
0
0 1 2 3 4 5 6
-2500

-5000

RI
-7500

SC
-10000
315
316 Fig. 9. Strain vs. time for the CURE concrete. The error bars in across crack and free-vertical

U
317 curves show the standard deviation between two and three measurements, respectively.
318
AN
10000
Free-horizontal
7500 Free-vertical
Across crack
M

5000
STRAIN (×10-6 [-])

2500
D

TIME (h)
0
0 1 2 3 4 5 6
TE

-2500

-5000

-7500
EP

-10000
319
C

320 Fig. 10. Strain vs. time for the CSRA concrete. The error bars in across crack and free-vertical
321 curves show the standard deviation between two and three measurements, respectively.
AC

322
323 3.7. Crack width
324 The cumulative crack width as a function of crack size and the average crack widths (presented
325 separately for two samples measured during one experiment) for the different mixtures are
326 shown in Fig. 11 and Table 4, respectively. The cumulative frequency was calculated by
 ACCEPTED MANUSCRIPT

327 measuring the crack widths every 10 mm along the crack’s length. All of the additives led to the
328 reduction of the crack width compared to the reference concrete REF.
329

100
CUMULATIVE FREQUENCY (%)

REF

PT
SEED
80 CURE
CSRA

RI
60

SC
40

20
U
AN
0
M

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

CRACK WIDTH (mm)
330
331 Fig. 11. Cumulative crack widths of concrete mixtures made with different crack-reducing additives. The
D

332 curves are the average and the error bars represent the range of the measured values from the two
333 samples tested during one experiment.
TE

334
335 Table 4. Average crack widths of concretes made with different crack-reducing additives. Results denoted
336 as 1 and 2 were measured on separate samples tested in parallel.
EP

Sample average crack width 1 average crack width 2 total average

(mm) (mm) (mm)
REF 0.389 0.466 0.428
C

SEED 0.263 0.279 0.271

CURE 0.318 0.159 0.239
AC

CSRA 0.182 0.108 0.145

337

338 3.8. Compressive strength

339 Compressive strength results are presented in Table 5. The reported values are the average of two
340 samples. No considerable differences could be observed between the REF, CURE and CSRA
341 concretes, while the SEED concrete had substantially lower 28-days compressive strength.
342
 ACCEPTED MANUSCRIPT

343 Table 5. Compressive strength results (average±standard deviation)

Sample 7-day (MPa) 28-day
(MPa)
REF 45.0±0.2 51.3±0.0
SEED 40.7±0.6 46.6±0.3
CURE 42.9±0.1 50.5±0.3
CSRA 42.6±0.5 51.0±1.1

PT
344

345

RI
346 4. Discussion

SC
347 The concretes did not bleed significantly, as shown in section 3.1. The latter was manifested also
348 by the onset of the pressure evolution and horizontal shrinkage, right after or very close to the
349 beginning of the plastic shrinkage test (see sections 3.4 and 3.6). See [55] and [28] for the effect

U
350 of the initial stiffness on bleeding and horizontal shrinkage, respectively. The latter is in line with
AN
351 the conclusions of [17], according to which in mixtures with high water demand requiring
352 superplasticizers, the onset of capillary pressure is accelerated and occurs almost immediately
353 after placing.
M

354 During the constant rate evaporation period, the curing agent accumulated on the surface pores
355 of CURE, forming a paraffin film that was apparently able to block the surface pores and reduce
D

356 the surface coefficient of permeability, which led to lower evaporation rate (see also eq. (2))
TE

357 [11,12,26]. However, at around 3.5 h to 4 h after mixing, the evaporation (see Fig. 6) and
358 consequently the pressure evolution (see Fig. 5) were accelerated. The latter can be related to the
359 evaporation of the layer of the water between the film and the surface of the concrete, when the
EP

360 evaporation rate decreased and less water was supplied from the bulk to keep the curing layer
361 wet. Consequently, when the layer of water finally disappeared from the surface, most likely the
C

362 film became discontinuous (i.e. pores opened again) and the rate of evaporation increased
AC

363 dramatically (at about 3.5 h from the start of mixing). After the opening, the evaporation rate was
364 even higher than in the CRP, since the surface temperature had increased more than for other
365 concretes beforehand (see Fig. 3), due to the lack of evaporation, leading to an increased gradient
366 of vapor pressure on the surface. The stability [50] and shrinkage and cracking [51] of the films
367 created by curing compounds are important factors which need to be investigated further under
368 different climatic conditions.
 ACCEPTED MANUSCRIPT

369 In CSRA, the evaporation was reduced, due to the effect of the SRA on the pore fluid’s surface
370 tension [10]. According to the Young-Laplace equation (eq. (1)), the characteristic air entry
371 pressure decreases in CSRA concretes, i.e. the pore drainage starts with lower capillary pressure
372 [28] (see Fig. 12). Therefore, as shown in Fig. 12, lower capillary pressure develops with the
373 same amount of lost water in CSRA concretes (see also [56] for soils).

PT
2γ cos θ
374 pc = − (1)
r

RI
375 In eq. (1), pc [Pa] is the capillary pressure, γ [Pa·m] is the surface tension of the pore fluid, θ
376 [rad] is the contact angle and r [m] is the radius of water-air menisci.

U SC
CAPILLARY PRESSURE
(absolute value-kPa)

AN

Sw<1
Reduction
of surface

Sw<1
tension
M

pae-REF
D
TE

Sw=1
pae-CSRA Sw=1
EP

VOLUMETRIC WATER CONTENT (m3/m3)

C

377
378 Fig. 12. Schematic conceptual diagram showing the effect of SRA on the Water Retention
AC

379 Characteristic Curve (WRCC) of reference concrete, right after mixing. pae [kPa] indicates the
380 characteristic air entry of concretes. Sw [-] is the saturation degree. It is assumed that after
381 mixing all water is present in the capillary pores.
382
383 Furthermore, CSRA concretes enter the Falling Evaporation Rate Period (FRP) earlier than the
384 reference concrete (see [10] for the effect of SRA dosage on the onset of FRP). The latter can be
385 explained by Fig. 12. When air enters the surface pores of CSRA, with a lower value of capillary
 ACCEPTED MANUSCRIPT

386 pressure compared to REF, the resulting desaturation decreases the relative hydraulic
387 conductivity (kr [-]) (see [57]), leading to a reduction of the effective coefficient of permeability
388 (krksat) in eq. (2). At the same time, with lower capillary pressure, the driving force for the
389 transport of water from the bulk towards the surface is reduced, see equation 2. This leads to
390 earlier onset of FRP (Falling Evaporation Rate Period), for CSRA compared to REF and SEED,

PT
391 according to the Darcy-Buckingham equation (eq. (2)) (see Fig. 6). In eq. (2), ksat [m/s] is the
392 saturated coefficient of permeability. The latter explanation is in line with the observations of a

RI
393 preceding study by neutron tomography [21]. In mortars possessing identical mixing water
394 content, the SRA-containing mortar had lower water content at the top surface, already few

SC
395 minutes after the beginning of the evaporation [21]. Also see [58] for the effect of surface
396 volumetric water content and hydraulic conductivity on the evaporation rate of soils. In addition,

U
397 the resulted slower progress of capillary pressure in the CSRA concrete also results in a smaller
398 contribution of the capillary pressure to the stiffness (see [5,28] for the effect of stiffness on the
AN
399 progress of capillary pressure in the plastic shrinkage problem).
r
r k r k sat r
400 q=− ⋅ (∇ ψ p + ρ f g ) (2)
M

ρf g
401 The evaporation decreased in SEED earlier than in REF, due to shorter dormant period and
D

402 earlier reduction of the pore sizes in SEED, and consequently earlier reduction of intrinsic
403 permeability as a consequence of faster hydration [28].
TE

404 Except for the CSRA mixture, the air entry occurred primarily due to the increased stiffness of
405 the bulk, which is due to the simultaneous reduction of pore size and evaporation rate in the
EP

406 hydration acceleration period, and slight progress of shrinkage, while the saturation degree most
407 likely remained very close to 1 [28]. However, as explained before, in CSRA the air entry was
C

408 due to the effect of the SRA on the surface tension during the dormant hydration period, when
409 pore sizes are constant [28]. Therefore, the shrinkage proceeded even after the air entry, when
AC

410 the concrete was not stiff enough, and eventually stopped at around the time of final set, when
411 the coupled effect of hydration and capillary pressure increased the stiffness of the bulk.
412 Accordingly, CSRA had lower horizontal shrinkage, due to the effect of SRA on the WRCC (see
413 Fig. 12), which also was manifested by later crack initiation time comparing to REF.
414 The capillary pressure started to develop in SEED at an earlier time, also manifested by faster
415 crack initiation (see Fig. 8), due to faster evolution of the bulk modulus as a consequence of
 ACCEPTED MANUSCRIPT

416 faster hydration (see Fig. 4 and Table 3). However, accelerating cement hydration decreases the
417 volumetric plastic shrinkage after the solid percolation. A previous study [49] showed that the
418 final crack size correlates well with the magnitude of the volumetric plastic shrinkage after solid
419 percolation. Therefore it can be concluded that, although acceleration of cement hydration
420 accelerated the crack initiation in this study, the reduction of shrinkage due to rapid increase of

PT
421 stiffness led to smaller final crack sizes. Furthermore, a faster evolution of mechanical properties
422 (i.e. stiffness and strength), can possibly arrest the crack propagation and decrease the final crack

RI
423 size [5,49]. It is worth here to mention that, opposite to the effect of cement hydration
424 accelerators, increasing the initial values of bulk modulus with inert materials (just by increasing

SC
425 the solid fraction or using finer particle size distribution) may on the contrary accelerate the
426 crack initiation [5] and also significantly increase the final crack size [5,49].

U
427 Although CURE had initially slower capillary pressure progress, the later sudden increase of the
428 evaporation rate (possibly due to the previously-mentioned film breakage) led to development of
AN
429 high capillary pressure, crack initiation and propagation. Since the intrinsic permeability should
430 be still relatively high at the time of film breakage, the shrinkage caused by severe evaporation
M

431 together with already high stiffness induced restraint stresses high enough to cause cracking.
432 Since cement hydration was delayed in the CURE concrete, one hour from the increase in
D

433 evaporation rate (corresponding to possible paraffin film breakage) to the time of final set was
434 enough for the crack to propagate to the surface (see Fig. 6 and Fig. 9). The short time between
TE

435 crack initiation and propagation is in line with the fast progress of plastic shrinkage cracking [8],
436 and the necessity of the curing up to the time of final set, as also mentioned in a previous paper
EP

437 [15]. Finally, the CSRA concrete showed the smallest overall crack width, since: 1- SRA delayed
438 the capillary pressure progress and therefore the effective stress progress [5], 2- SRA did not
C

439 retard the hydration (i.e. strength and stiffness) significantly.

440 Based on the above-mentioned discussion, it can be speculated that a combination of passive
AC

441 methods could be even more effective. For instance, the combination of SRA or evaporation
442 reducers with an accelerator (e.g. seeding agents) would be very effective, since the former
443 changes the WRCC and reduces the evaporation rate, and the latter accelerates the evolution of
444 mechanical properties. At the same time, the use of SRA would be even more effective than the
445 use of evaporation reducers, since SRA can modify the WRCC (i.e. reduce the characteristic air
446 entry pressure), while at the same time they also reduce the evaporation rate.
 ACCEPTED MANUSCRIPT

447 According to the compressive strength data, it can be concluded that, except for SEED, all
448 concretes had similar 7-day and 28-day strength. Although C-S-H seeding accelerates strength
449 enhancement up to 24 h after mixing [59], which is beneficial in controlling the plastic shrinkage
450 cracking, they might reduce to a certain extent the later strength, as also shown in [59]. This is
451 somewhat surprising, since contrary to other accelerators C-S-H seeds, when properly stabilized,

PT
452 should cause homogenous nucleation of C-S-H in the pore fluid, rather than heterogeneous
453 nucleation on the cement grains [60]. While this is only of marginal interest for the present study,

RI
454 the long-term effect of C-S-H seeding on the mechanical properties of concrete needs further
455 investigation.

SC
456 5. Conclusions

457 The performance of passive plastic shrinkage cracking mitigation methods was investigated
458
U
according to the ASTM standard, by comparing the final crack sizes. Furthermore,
AN
459 complementary tests were performed to explain the mechanisms of action of passive methods
460 according to an approach based on poromechanics.
M

461 Calcium Silicate Hydrate seeding accelerated both the pressure evolution and the evolution the
462 mechanical properties at very early ages. The fast hydration, accompanied by rapid stiffness and
D

463 strength increase, arrested the crack propagation and decreased the final crack size. However, it
TE

464 was also observed that Calcium Silicate Hydrate seeding slightly reduced the 7-day and 28-day
465 compressive strength.
EP

466 A curing compound (paraffin suspension) admixed into the concrete also was shown to be
467 initially effective in reducing plastic shrinkage. However, after a couple of hours the evaporation
468 rate of the concrete suddenly increased (possibly as a consequence of the breaking of the paraffin
C

469 film at the surface). In addition, the curing compound also retarded cement hydration.
AC

470 A shrinkage-reducing admixture delayed the increase in capillary pressure by reducing the
471 characteristic air entry value, and the evaporation rate, without substantially retarding the
472 evolution of mechanical properties (no effect on time of set). Therefore, the shrinkage-reducing
473 admixture was found to be the most effective admixture in this study.
 ACCEPTED MANUSCRIPT

474 Moreover, it was hypothesized that the combination of different passive methods, to target both
475 the water retention curve and the acceleration of evolution of mechanical properties, could
476 represent a robust solution for mitigation of plastic shrinkage cracks.

477 Acknowledgements

PT
478 We acknowledge Holcim Technology Ltd and Empa for funding the experiments. We thank the useful
479 discussions with Dr. Fabrizio Moro, and furthermore, the contribution of Mr. Daniel Käppeli in
480 performing the experiments.

RI
481

SC
482

483

U
484
AN
485 References

486 [1] T.C. Powers, Properties of Fresh Concrete, John Wiley and Sons, Inc N. Y. (1968) 301.
487 [2] T.S. Tan, T.H. Wee, S.A. Tan, C.T. Tam, S.L. Lee, A consolidation model for bleeding of cement
M

488 paste, Adv. Cem. Res. 1 (1987) 18–26.

489 [3] A. Radocea, A model of plastic shrinkage, Mag. Concr. Res. 46 (1994) 125–132.
490 [4] P.J. Uno, Plastic shrinkage cracking and evaporation formulas, ACI Mater. J. 95 (1998) 365–375.
D

491 [5] S. Ghourchian, M. Wyrzykowski, P. Lura, On the mechanism of plastic shrinkage cracking in fresh
492 cementitious materials, submitted to Cement and Concrete Research. (2018).
TE

493 [6] H.-G. Kwak, S. Ha, W.J. Weiss, Experimental and numerical quantification of plastic settlement in
494 fresh cementitious systems, J. Mater. Civ. Eng. 22 (2010) 951–966.
495 [7] S.H. Kosmatka, B. Kerkhoff, W.C. Panarese, Design and control of concrete mixtures, Portland
496 Cement Assoc., 2011.
EP

497 [8] W. Lerch, Plastic shrinkage, in: ACI 52nd Annu. Conv., Journal of the American Concrete Institute,
498 Philadelphia, 1957: pp. 797–802.
499 [9] D.P. Bentz, E.J. Garboczi, Y. Lu, N. Martys, A.R. Sakulich, W.J. Weiss, Modeling of the influence of
500 transverse cracking on chloride penetration into concrete, Cem. Concr. Compos. 38 (2013) 65–74.
C

501 [10] P. Lura, B. Pease, G.B. Mazzotta, F. Rajabipour, J. Weiss, Influence of shrinkage-reducing
502 admixtures on development of plastic shrinkage cracks, ACI Mater. J. 104 (2007).
AC

503 [11] A. Leemann, P. Nygaard, P. Lura, Impact of admixtures on the plastic shrinkage cracking of self-
504 compacting concrete, Cem. Concr. Compos. 46 (2014) 1–7.
505 [12] J. Plank, E. Sakai, C.W. Miao, C. Yu, J.X. Hong, Chemical admixtures—Chemistry, applications and
506 their impact on concrete microstructure and durability, Cem. Concr. Res. 78 (2015) 81–99.
507 [13] ACI Committee 305, ACI 305R-10: Guide to Hot Weather Concreting, Am. Concr. Inst. (2010).
508 [14] ACI 302.1R-04, Guide for Concrete Floor and Slab Construction, Am. Concr. Inst. (2004).
509 [15] S. Ghourchian, M. Wyrzykowski, P. Lura, A practical approach for reducing the risk of plastic
510 shrinkage cracking of concrete, submitted to RILEM Technical Letters. (2017).
 ACCEPTED MANUSCRIPT

511 [16] I. Löfgren, O. Esping, O.M. Jensen, P. Lura, K. Kovler, Early age cracking of self-compacting
512 concrete, in: Int. RILEM Conf. Vol. Chang. Hardening Concr. Test. Mitigation” Lyngby, 2006: pp.
513 251–260.
514 [17] A. Kronlöf, M. Leivo, P. Sipari, Experimental study on the basic phenomena of shrinkage and
515 cracking of fresh mortar, Cem. Concr. Res. 25 (1995) 1747–1754.
516 [18] C.A. Menzel, A tribute to “Causes and Prevention of Crack Development in Plastic Concrete,” ACI
517 Spec. Publ. 249 (2008) 161–172.

PT
518 [19] A.M. Alhozaimy, A.I. Al-Negheimish, Plastic Shrinkage in Hot and Arid Environments, Concr. Int. 31
519 (2009) 26–32.
520 [20] J. Mora-Ruacho, R. Gettu, A. Aguado, Influence of shrinkage-reducing admixtures on the reduction

RI
521 of plastic shrinkage cracking in concrete, Cem. Concr. Res. 39 (2009) 141–146.
522 [21] M. Wyrzykowski, P. Trtik, B. Münch, J. Weiss, P. Vontobel, P. Lura, Plastic shrinkage of mortars with
523 shrinkage reducing admixture and lightweight aggregates studied by neutron tomography, Cem.
524 Concr. Res. 73 (2015) 238–245.

SC
525 [22] J. Saliba, E. Rozière, F. Grondin, A. Loukili, Influence of shrinkage-reducing admixtures on plastic
526 and long-term shrinkage, Cem. Concr. Compos. 33 (2011) 209–217.
527 [23] F. Rajabipour, G. Sant, J. Weiss, Interactions between shrinkage reducing admixtures (SRA) and

U
528 cement paste’s pore solution, Cem. Concr. Res. 38 (2008) 606–615.
529 [24] S.-T. Lin, R. Huang, Effect of viscosity modifying agent on plastic shrinkage cracking of cementitious
AN
530 composites, Mater. Struct. 43 (2010) 651–664.
531 [25] S. Ghourchian, M. Wyrzykowski, P. Lura, The bleeding test: a simple method for obtaining the
532 permeability and bulk modulus of fresh concrete, Submitt. Cem. Concr. Res. (2015).
533 [26] J.P. Liu, L. Li, C.W. Miao, Q. Tian, Q.P. Ran, Y.J. Wang, Reduction of water evaporation and cracks
M

534 on plastic concrete surface by monolayers, Colloids Surf. Physicochem. Eng. Asp. 384 (2011) 496–
535 500.
536 [27] R. Henkensiefken, P. Briatka, D. Bentz, T. Nantung, J. Weiss, Plastic shrinkage cracking in internally
D

537 cured mixtures made with pre-wetted lightweight aggregate, Concr. Int. 32 (2010) 49–54.
538 [28] S. Ghourchian, M. Wyrzykowski, P. Lura, A poromechanics model for plastic shrinkage of fresh
539 cementitious materials, Cem. Concr. Res. 109 (2018) 120–132.
TE

540 [29] Y. Ma, M. Tan, K. Wu, Effect of different geometric polypropylene fibers on plastic shrinkage
541 cracking of cement mortars, Mater. Struct. 35 (2002) 165–169.
542 [30] C. Qi, J. Weiss, J. Olek, Characterization of plastic shrinkage cracking in fiber reinforced concrete
EP

543 using image analysis and a modified Weibull function, Mater. Struct. 36 (2003) 386–395.
544 [31] J. Kaufmann, F. Winnefeld, D. Hesselbarth, Effect of the addition of ultrafine cement and short
545 fiber reinforcement on shrinkage, rheological and mechanical properties of Portland cement
546 pastes, Cem. Concr. Compos. 26 (2004) 541–549.
C

547 [32] T. Rahmani, B. Kiani, M. Bakhshi, M. Shekarchizadeh, Application of different fibers to reduce
548 plastic shrinkage cracking of concrete, in: 7th RILEM Int. Conf. Crack. Pavements, Springer, 2012:
AC

549 pp. 635–642.

550 [33] P. Soroushian, F. Mirza, A. Alhozajiny, Plastic shrinkage cracking of polypropylene fiber reinforced
551 concrete, Mater. J. 92 (1993) 553–560.
552 [34] P. Soroushian, S. Ravanbakhsh, Control of plastic shrinkage cracking with specialty cellulose fibers,
553 Mater. J. 95 (1998) 429–435.
554 [35] H.R. Shen, Y.X. She, P.W. Gao, The influence of polypropylene fiber on the performance of
555 concrete pavement, in: Appl. Mech. Mater., Trans Tech Publ, 2012: pp. 1099–1103.
556 [36] Y. Shao, A. Mirmiran, Control of plastic shrinkage cracking of concrete with carbon fiber-reinforced
557 polymer grids, J. Mater. Civ. Eng. 19 (2007) 441–444.
 ACCEPTED MANUSCRIPT

558 [37] M. Nemoz-Gaillard, D. Nectoux, E. Dallies, D. Muller, E. des Mines d’Ales, Influence of AR glass
559 fibers on the cracking of concrete: Analysis at the very early age by digital image correlation, in: 23
560 Int. RILEM Conf. Early Age Crack. Cem. Syst.-EAC01, RILEM Publications, 2002: p. 237.
561 [38] A. Messan, P. Ienny, D. Nectoux, Free and restrained early-age shrinkage of mortar: Influence of
562 glass fiber, cellulose ether and EVA (ethylene-vinyl acetate), Cem. Concr. Compos. 33 (2011) 402–
563 410.
564 [39] M. Norfleet, A model to quantify the influence of fiber reinforcement on plastic shrinkage cracking,

PT
565 Purdue University, 2008.
566 [40] N.S. Berke, M.P. Dallaire, Effect of low addition rates of polypropylene fibers on plastic shrinkage
567 cracking and mechanical properties of concrete, Spec. Publ. 142 (1994) 19–42.

RI
568 [41] P.N. Balaguru, S.P. Shah, Fiber-reinforced cement composites, 1992.
569 [42] N. Banthia, R. Gupta, TEST METHOD FOR EVALUATION OF PLASTIC SHRINKAGE CRACKING IN FIBER-
570 REINFORCED CEMENTITIOUS MATERIALS, Exp. Tech. 31 (2007) 44–48.
571 [43] R. Malathy, K. Subramanian, M. Rameshkumar, Effect of glass fibers on restrained plastic shrinkage

SC
572 cracking of HPC with silica fume, (2007).
573 [44] A. Hosoda, PLASTIC SHRINKAGE CRACK CONTROLLING MECHANISM OF SYNTHETIC SHORT FIBER,
574 in: 30th Conference on Our World in Concrete and Structures (OWICs), Singapore, 2005.

U
575 http://cipremier.com/100030033.
576 [45] J. Branch, A. Rawling, D.J. Hannant, M. Mulheron, The effects of fibres on the plastic shrinkage
AN
577 cracking of high strength concrete, Mater. Struct. 35 (2002) 189–194.
578 [46] J.J. Thomas, H.M. Jennings, J.J. Chen, Influence of nucleation seeding on the hydration mechanisms
579 of tricalcium silicate and cement, J. Phys. Chem. C. 113 (2009) 4327–4334.
580 [47] M. Duriez, R. Lézy, Possibilités nouvelles dans le durcissement rapide des ciments, mortiers et
M

581 bétons, Ann. L’Institut Tech. BâTim. Des. 98 (1956) 482–483.

582 [48] L. Nicoleau, The acceleration of cement hydration by seeding: influence of the cement mineralogy,
583 ZKG Int. 1 (2013) 40–49.
D

584 [49] S. Ghourchian, M. Wyrzykowski, L. Baquerizo, P. Lura, Susceptibility of Portland cement and
585 blended cement concretes to plastic shrinkage cracking, Cem. Concr. Compos. 85 (2018) 44–55.
586 [50] ASTM C1579-13, Standard Test Method for Evaluating Plastic Shrinkage Cracking of Restrained
TE

587 Fiber Reinforced Concrete (Using a Steel Form Insert), ASTM International, West Conshohocken,
588 PA, 2013, www.astm.org, (n.d.).
589 [51] BS EN 12350-2:2009,Testing fresh concrete. Slump-test, (n.d.).
EP

590 [52] EN 12350-5:2009.Testing fresh concrete. Flow table test, (2009).

591 [53] BS EN 480-4:2005, Admixtures for concrete, mortar and grout. Test methods. Determination of
592 bleeding of concrete, (2005).
593 [54] ASTM C807-13,Standard Test Method for Time of Setting of Hydraulic Cement Mortar by Modified
C

594 Vicat Needle, (2013).

595 [55] S. Ghourchian, M. Wyrzykowski, P. Lura, The bleeding test: A simple method for obtaining the
AC

596 permeability and bulk modulus of fresh concrete, Cem. Concr. Res. 89 (2016) 249–256.
597 [56] A. Karagunduz, K.D. Pennell, M.H. Young, Influence of a nonionic surfactant on the water retention
598 properties of unsaturated soils, Soil Sci. Soc. Am. J. 65 (2001) 1392–1399.
599 [57] M.T. Van Genuchten, A closed-form equation for predicting the hydraulic conductivity of
600 unsaturated soils, Soil Sci. Soc. Am. J. 44 (1980) 892–898.
601 [58] E. Haghighi, E. Shahraeeni, P. Lehmann, D. Or, Evaporation rates across a convective air boundary
602 layer are dominated by diffusion, Water Resour. Res. 49 (2013) 1602–1610.
603 [59] K. Owens, M.I. Russell, G. Donnelly, A. Kirk, P.A.M. Basheer, Use of nanocrystal seeding chemical
604 admixture in improving Portland cement strength development: application for precast concrete
605 industry, Adv. Appl. Ceram. 113 (2014) 478–484.
 ACCEPTED MANUSCRIPT

606 [60] M. Bräu, L. Ma-Hock, C. Hesse, L. Nicoleau, V. Strauss, S. Treumann, K. Wiench, R. Landsiedel, W.
607 Wohlleben, Nanostructured calcium silicate hydrate seeds accelerate concrete hardening: a
608 combined assessment of benefits and risks, Arch. Toxicol. 86 (2012) 1077–1087.
609

PT
RI
U SC
AN
M
D
TE
C EP
AC