The use of superabsorbent polymers to reduce cracking of bonded mortar overlays

Cement & Concrete Composites 52 (2014) 1–8

Contents lists available at ScienceDirect

Cement & Concrete Composites journal homepage: www.elsevier.com/locate/cemconcomp

The use of superabsorbent polymers to reduce cracking of bonded mortar overlays H. Beushausen ⇑, M. Gillmer University of Cape Town, Department of Civil Engineering, Concrete Materials and Structural Integrity Research Unit, South Africa

a r t i c l e

i n f o

Article history: Received 19 April 2013 Received in revised form 10 March 2014 Accepted 28 March 2014 Available online 4 April 2014 Keywords: Blended cement mortars Superabsorbent polymers Restrained shrinkage Cracking Tensile relaxation Elastic modulus Concrete repair

a b s t r a c t The use of superabsorbent polymers (SAP) was found to significantly improve cracking characteristics of normal strength bonded mortar overlays containing silica fume. Mortars were designed with two water/ binder ratios and three different SAP contents and tested for relevant material properties such as tensile strength, tensile relaxation, elastic modulus and drying shrinkage. With respect to overlay cracking, SAP addition generally resulted in an improvement in all tested properties, especially with respect to increasing relaxation and reduced elastic modulus. To observe cracking in real bonded overlays, mortars were cast on a rigid substrate and their cracking behavior observed for a period of approximately 7 weeks. The results indicate that SAP additions of 0.4% and 0.6% are very effective in reducing cracking in bonded mortar overlays with water/binder ratios of 0.45 and 0.55, respectively. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Bonded mortar overlays are common in concrete repair projects and for the lining and leveling of concrete members. The performance requirements for bonded overlays include crack resistance and sufficient bond strength to the concrete substrate. Further, for repair patches, sufficient resistance against penetration of aggressive agents such as carbon dioxide and chlorides is usually required. A common shortfall of bonded overlays is cracking due to restrained shrinkage, which is unsightly, can initiate debonding, and locally increases penetrability (see, for example [1,2]). With respect to cracking, the most important material parameters for mortar overlays are tensile strength, elastic modulus, relaxation, and shrinkage [3,4]. When subjected to restrained shrinkage, bonded mortars can be expected to remain crack-free if the following (simplified) condition is fulfilled:

ac  e  Et  wt ft

61

ð1Þ

where ar is the factor describing the degree of restraint [3]; e the free shrinkage strain; Et the elastic modulus in tension; wt the factor accounting for tensile relaxation and ft is the tensile strength. ⇑ Corresponding author. Address: University of Cape Town, Department of Civil Engineering, Private Bag, Rondebosch 7701, South Africa. Tel.: +27 21 650 5181; fax: +27 21 689 7471. E-mail address: [email protected] (H. Beushausen). http://dx.doi.org/10.1016/j.cemconcomp.2014.03.009 0958-9465/Ó 2014 Elsevier Ltd. All rights reserved.

In terms of overlay material properties, increasing tensile strength results in increased crack resistance. However, the common method to increase strength is to decrease the w/b ratio, which also results in a higher elastic modulus and reduced creep and relaxation, and therefore in higher stresses. Consequently, the overall effect of changing the mix design to increase strength may actually result in increased cracking. What is therefore needed from a material technology perspective is a method to effectively change some of the above material properties without negatively affecting the others. The aim of the research discussed in this paper was to use superabsorbent polymers (SAP) to alter mortar material properties to obtain an overall positive effect on cracking. For a number of years, SAP has been used to improve the performance of high strength concrete and reduce cracking due to autogeneous shrinkage [5,6]. The principle of internal curing with SAP relates to the provision of water-filled cavities in the hardened cement paste, which gradually release the water during cement hydration and thus facilitate the development of a denser microstructure [7]. The addition of SAP can result in a higher number of voids in the concrete and thus in reduced elastic modulus and increased creep [8]. These effects are undesirable for most structural applications but very helpful reducing stresses due to restrained deformation (compare Eq. (1)). The addition of SAP was therefore expected to have a positive effect on material properties and crack development of bonded mortar overlays. This hypothesis was tested by investigating the influence of SAP

2

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

addition on all material properties included in Eq. (1), and by observing the cracking behavior of bonded mortars containing SAP. Application of Eq. (1) for the analytical modeling of stress development in concrete subjected to restrained shrinkage would require knowledge on the time-dependent development of all material properties included in the equation. However, in the scope of this research, some of these properties were measured at individual test ages only, which does not permit time-development stress modeling. Eq. (1) is therefore simply shown to summarize how various material properties affect cracking performance of bonded overlays. In the experimental work, cracking was analyzed based on the cracking performance of real bonded overlays under restrained deformation. In a preliminary experimental study, the effect of SAP addition on durability-related mortar properties was investigated, which generally yielded very positive results, especially for mortars containing silica fume. For the investigation of crack development, mortars containing silica fume and various amounts of SAP were therefore selected. 2. Experimental details and test results 2.1. Mix design and concrete manufacture The aim of this research was to identify if the addition of SAP to bonded mortar overlays results in decreased overlay cracking and to explain the observed cracking behavior with the measurement of relevant mortar material properties. Various contents of SAP were used in the mix design to identify if an optimum content of SAP exists at which cracking can be minimized. Mortars were produced with 2 different w/b ratios of 0.45 and 0.55. A total of 8 mixes were designed, as presented in Table 1. The oxide composition of the binders used is presented in Table 2. The water content was kept constant at 250 l/m3 and superplasticizer (SP) was added when necessary. The SP consisted of a sodium salt of polymerized naphthalene sulfonic acid, supplied in dry powder form by BASF. The sand used was silicious pit sand with a fineness modulus of 2.7 and relative density of 2.65. The target slump was set at 50 ± 20 mm. SP was added to the mix after an initial slump test had been performed, the amount being adjusted to achieve the desired slump value. The SAP content was set at 0% (reference mortar), 0.2%, 0.4% and 0.6%, in order to identify if an optimum replacement level can be found for the cracking performance of repair mortars. In order to compare the direct influence of SAP addition on mortar properties it was deemed practical to use the same water content in all mortars, such that the same basic mix was used. Since the water demand was not adjusted to the SAP content, the higher absorption of water in mortars with higher SAP contents can be expected to have had an influence on the microstructure of the cement paste. This in turn will have an effect on the mechanical material properties that were measured in this research. It was

Table 2 % Oxide composition of binders. Oxide

CEM I42.5N

SF

CaO SiO2 Al2O3 Fe2O3 SO3 MgO K2O TiO2 Mn2O3 Na2O P2O5

64.12 20.75 4.17 3.21 2.30 0.74 0.73 0.28 0.05 0.04 0.08

0.0 91.9 0.7 0.8 0.46 0.26 1.33 0.11 0.09 0.42 Not available

therefore assumed that the effects of changes in microstructure due to SAP addition were sufficiently identified in the experimental program. Also from a practical point of view, it was considered relevant to assess how the addition of SAP to a given mortar mix of constant composition would affect its cracking performance. SAP was added as a dry powder the mix prior to water addition. The workability of the mortars was assessed using a common slump test. As expected from literature (e.g. [6]), a decrease in workability with increasing SAP content was observed, resulting from the water uptake of the polymer particles. Similarly, increasing SAP contents resulted in more efforts for casting, compaction and finishing operations, as well as reduced bleed water accumulation on the concrete surface. The SAP material used in this research consisted of covalently cross-linked acrylamide/acrylic acid copolymers. The suspension polymerized spherical particles had an average particle size of approximately 200 lm. The properties of the SAP material, including absorption characteristics, were previously published by Jensen and Hansen [7]. All samples were kept in their moulds under plastic sheeting for 1 day and then demoulded and cured for a further 2 days under plastic sheeting to prevent moisture loss and replicate common site curing conditions for repair work. Subsequently, samples were exposed to various environmental conditions, depending on the test parameter, as discussed in the following sections. A shortcoming of the data analysis in the following sections is that the measured individual material properties were not statistically evaluated, partly due to insufficient numbers of test specimens. Consequently, a statistically significant comparison of mortars with various SAP contents could not be performed. However, such a comparison was not the focus of this research, which aimed at identifying the influence of SAP on actual overlay cracking performance. The individual material properties were obtained to only provide an indicative explanation for the effect that increasing SAP addition has on mortar cracking performance. A detailed investigation into time-dependent mortar properties was therefore not considered relevant, but should be considered in future work.

Table 1 Mix design and workability (slump). Mix

a

w/b

Total binder (kg)

CEM I42.5 (kg)

FA (kg)

GGBS (kg)

SF (kg)

Water (kg)

Sand 002 (kg)

SAP (kg)

Slump (mm)

SP (kg)

7 8 9 16 17 18

0.45 0.45 0.45 0.55 0.55 0.55

556 556 556 455 455 455

500 500 500 409 409 409

0 0 0 0 0 0

0 0 0 0 0 0

56 56 56 45 45 45

250 250 250 250 250 250

1490 1490 1490 1530 1530 1530

1.11 2.22 3.33 0.91 1.82 2.73

40 50 40 70 50 50

1.00 1.76 2.94 0.25 0.59 2.35

Controls 21 24

0.45 0.55

556 455

500 409

0 0

0 0

56 45

250 250

1490 1530

0 0

80 50

2.35 0.59

a

The mix number was contained from the complete research project, which included testing of other mixes for durability characteristics.

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

2.2. Compressive and tensile strength Compressive strength was tested after 3, 7 and 28 days on cubes with 100 mm side length. A total of 3 cubes were tested for each parameter at each age. Tensile strength was measured on ‘‘dog bone’’ samples (Fig. 1) using a Zwick Roell Universal Testing Machine. Test ages for tensile strength were 3 and 14 days. Two specimens were tested for each parameter and age. Strength samples were cured as discussed earlier and subsequently stored at 23 ± 2 °C and 50 ± 5% RH until the age of testing. Compressive strength was tested at 3 different ages to identify and, if applicable, contrast the influence of SAP on early-age strength and strength development at later ages (up to 28 days design strength). Tensile strength was tested at ages that were deemed significant for tensile cracking, i.e. 3-day strength at the onset of stress development on completion of curing, and 14 day strength, which was the age when most of the overlays showed a noteworthy degree of first cracking. Test results for compressive and tensile strength are presented in Figs. 2 and 3, respectively. Compressive strength, which was just tested as a control parameter but which does not really have a significant influence on the performance of common patch repairs, was not much affected by SAP addition. For mortars with w/b of 0.45, the addition of SAP generally had a slightly positive effect on the tensile strength, especially at later ages. The mortars with w/b = 0.55 showed decreasing tensile strength with increasing SAP addition at an age of 3 days. However, at an age of 14 days, mixes containing SAP generally outperformed the reference mixes. Hence, if early cracking is avoided, SAP addition can be considered to increase tensile strength and hence have a positive effect on crack resistance of mortar containing silica fume. 2.3. Drying shrinkage Drying shrinkage was measured on prismatic specimens with dimensions of 50 mm  50 mm  300 mm, using a BAM extensometer. Demec targets were placed over a gauge length of 100 mm on 2 opposite faces of the prism. Three prisms were made per mix, resulting in a total of 6 measurement locations. Test results were recorded as the mean value from the 6 locations after exclusion of outlying values where applicable.

3

Drying shrinkage samples were cured as discussed earlier and subsequently exposed to a controlled environment at 23 ± 2 °C and 50 ± 5% RH. Strain measurements were taken in regular intervals for a duration of approximately 8 weeks. Autogeneous shrinkage was not measured as it was considered negligible at the relatively high w/b ratios used. Test results for mortars with w/b ratios of 0.45 and 0.55 are shown in Figs. 4 and 5, respectively. In general, shrinkage was not significantly affected by the addition of SAP. However, at both w/b ratios a slight decrease in shrinkage strain was observed for SAP contents of 0.4% and 0.6%, compared to the reference mix, indicating a generally positive effect of SAP addition. The results correspond to observations made by Jensen [6] who states that SAP may potentially be used to reduce drying shrinkage but that the effect can be expected to be small. Other publications [5,10,11] reported slightly increased drying shrinkage as a result of SAP addition to high strength concrete. However, the effects of SAP on drying shrinkage of high strength concrete may be different to normal strength concrete, which can be explained by the different microstructure characteristics. Further research is needed to fully explain this. 2.4. Elastic modulus For the estimation of stresses from restrained shrinkage in bonded overlays, information the elastic modulus (E-mod) in tension is required. However, testing E-mod in tension is generally difficult and in this research, E-mod was measured in compression at an age of 28 days. Since the test results were analyzed on a comparative basis, the single testing age and the fact that compressive testing was used instead of tensile testing were considered sufficient to give an indication on the influence of SAP addition on the elastic properties of the mortars. The E-mod was measured on prismatic specimens with dimensions of 100 mm  100 mm  200 mm. Test samples were cured as discussed earlier and subsequently stored at 23 ± 2 °C and 50 ± 5% RH until the age of testing. The E-mod was recorded as the mean values obtained from 3 different samples per specimen type; test results are shown in Fig. 6. The addition of SAP resulted in a noteworthy reduction in elastic modulus for mortars with both w/b ratios, increasing SAP contents resulting in decreasing E-mod. The results are consistent with research carried out by van der Ham et al. [9], who found between 5% and 10% reduction in E-mod for concretes with w/b ratios between 0.32 and 0.39 and SAP contents between 0.3% and 0.4%. In this research, SAP contents of 0.2%, 0.4% and 0.6% resulted in a reduction in E-mod values by approximately 5%, 10%, and 15%, respectively, compared to the reference mixes. It should however be noted that the experimental results obtained by Ham et al. [9] cannot be directly compared against the results obtained in this study. In their mix design, Ham et al. adjusted the water content to account for water absorption by the SAP particles and hence kept the effective w/b ratio the same in all mixes. In contrast, in this research, the water content was kept constant across all mixes, which probably resulted in a lower effective w/b ratio, and hence denser microstructure, of the mixes with higher SAP content. The results obtained in this study, i.e. reduced elastic modulus with increasing SAP content and decreasing effective w/b ratio highlight the dominant effect of increasing porosity, and hence loss in stiffness, with increasing SAP contents. 2.5. Tensile relaxation

Fig. 1. Geometry of tensile strength and tensile relaxation specimens (measurements in mm).

Tensile relaxation was measured on the same type ‘‘dog bone’’ specimens as tensile strength (Fig. 1). Tensile relaxation was measured as the time-dependent reduction in stress under sustained

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

Compressive strength (MPa)

4

60

28 d 7d 3d

50 40 30 20 10 0

w/b 0.45 w/b 0.45 w/b 0.45 w/b 0.45 w/b 0.55 w/b 0.55 w/b 0.55 w/b 0.55 0% 0.2% 0.4% 0.6% 0% 0.2% 0.4% 0.6%

Binder type and SAP content Fig. 2. Compressive strength test results.

Tensile strength (MPa)

4

14 d 3d

3

2

1

0 w/b 0.45 w/b 0.45 w/b 0.45 w/b 0.45 w/b 0.55 w/b 0.55 w/b 0.55 w/b 0.55 0% 0.2% 0.4% 0.6% 0% 0.2% 0.4% 0.6%

Binder type and SAP content Fig. 3. Tensile strength test results.

Time after completion of curing (days)

Time after completion of curing (days) 0

-100 -200

0

10

20

30

40

50

w/b = 0.45

-300 0%

-400

0.2%

-500

0.4%

-600

0

60

0.6%

-700 -800 -900

Drying shrinkage strain (10-6)

Drying shrinkage strain (10-6)

0

10

20

30

40

50

60

-100

w/b = 0.55 -200 -300 0%

-400 0.2%

-500 0.4%

-600 0.6%

-700 -800

Fig. 4. Shrinkage strain development, w/b = 0.45.

imposed strain, using a Zwick Roell Z020 Testing Machine (UTM) with a maximum capacity of 20 kN. The test specimens were loaded to a stress equivalent to approximately 70% of the tensile strength at the age of testing. The resulting imposed strain was kept constant and the stress decay in the specimens automatically recorded. The remaining stress after a period of 24 h was used in the analysis of test results. Subsequent to curing and prior to test commencement, specimens were coated with paraffin wax to ensure that no moisture loss occurred during testing, as this would have resulted in increasing stress from restrained drying shrinkage.

Fig. 5. Shrinkage strain development, w/b = 0.55.

Due to equipment and time constraints, only one specimen could be tested for each different mortar, which was considered sufficient for the comparative analysis done in this study. Subsequent to curing as discussed earlier, specimens were tested at an age of 3 days. Test results are expressed as percentage stress-decay over the 24-h testing period, as presented in Fig. 7. The 24-h relaxation values measured in this study ranged between 10% and 18%, which was relatively low, compared to previous studies on relaxation carried out by the authors [12].

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

Elastic modulus (GPa)

35 w/b 0.45

30 w/b 0.55

25

20

15 0%

0.20%

0.40%

0.60%

SAP content Fig. 6. Elastic modulus test results.

However, the results allow a comparative analysis with respect to SAP content, with mortars of both w/b ratios showing increasing stress relaxation with increasing SAP addition. Similar results were presented in [9], who reported on increasing creep factors resulting from SAP addition to concrete. 2.6. Bonded overlay cracking characteristics The assessment of the cracking behavior of real bonded mortar overlays formed the main part of the experimental investigation. For the manufacture of composite concrete substrate-to-mortar specimens, the authors made use of the circumstance that, at the time of the experimental research, their existing laboratory was about to be demolished in connection with the construction of new facilities. Consequently, permanent damage could be inflicted to the heavy-duty laboratory concrete floor, which was therefore used as substrate for overlay application. To provide sufficient roughness for maximum bond strength, the top approximately 5 mm of the concrete floor were removed with a jackhammer. Subsequently, the rough surface was pressure-air-cleaned to remove any lose debris and dust. Prior to application of the mortars, the substrate was water-saturated for 24 h and then left to dry for about 60 min. The industrial substrate concrete floor was assumed to be infinitely rigid and hence provide maximum possible restraint to shrinkage deformations of the bonded mortars. Overlays were made with a thickness of 40 mm and dimensions of 1500 mm  300 mm

5

(Fig. 8). Mortar overlays were applied and compacted with a trowel. Temporary wooden formwork was used to shape and support the overlays during manufacture and hardening. Curing was done with plastic sheeting for 3 days and the specimens subsequently exposed to the laboratory environment at approximately 50–60% RH and 22–25 °C. The cracking behavior of the overlays was assessed visually on a week-daily basis, for a total duration of approximately 7 weeks. All overlays were cast on the same day and were positioned next to one another over a total test area of approximately 3000  2000 mm. The test area was situated on the laboratory floor in a corner of the room and remote from windows or doors, with limited air draft exposure. The climatic conditions of the individual overlays were therefore assumed to be identical and the observed differences in cracking behavior considered a result of the different overlay material properties. The crack development of the bonded mortar overlays is presented for selected ages in Figs. 9 and 10 for mortars with w/b ratios of 0.45 and 0.55, respectively. Cracks were highlighted with a marker. The analysis of test results was based on a simple visual comparison of cracking patterns and crack development. For both w/b ratios, the addition of SAP prolonged the time until first cracking and reduced the total crack area, the effect depending on the w/b ratio and the amount of SAP addition. The reference mortar (0% SAP) with a w/b ratio of 0.45 showed extensive cracking already after 2 weeks of exposure. Mortars with a SAP content of 0.2% showed a significantly lesser number of cracks and in mortars with SAP contents of 0.4% and 0.6%, cracking was completely prevented at this age. Considering long-term crack development, the SAP content of 0.2% was ineffective in improving cracking behavior, whereas SAP contents of 0.4% and 0.6% resulted in significantly lower number of cracks and reduced crack lengths, compared to the reference mortar. For mortars with a w/b ratio of 0.55, the lower SAP content of 0.2% proved to be ineffective both at early and later ages. Cracking behavior was marginally improved with a SAP content of 0.4% and the improvement was observed mainly at an early age (14 days). The best performance was observed with a SAP content of 0.6%, which resulted in significantly less cracking, compared to the reference mortar. For the overall performance of the mortars it needs to be noted that the specimens were designed to crack in order to enable a comparative analysis. No effort was made to prevent cracking through mix design or curing procedures. Therefore, even though specimens containing SAP showed extensive cracking, the superior performance achieved as a result of SAP addition indicates the materials’ promising potential in reducing cracking in bonded mortar overlays.

20

Relaxation (%-stress decay)

w/b 0.55

15

w/b 0.45

10

5

0 0%

0.20%

0.40%

0.60%

SAP content Fig. 7. Tensile relaxation results.

Fig. 8. Bonded mortar overlays cast on the roughened concrete floor of the laboratory.

6

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

Fig. 9. Photographs of bonded mortar specimens: crack-development in mortars with w/b ratio of 0.45 (30 mm ruler as a scale).

As a general observation it is interesting to note that mortars at the higher w/b ratio of 0.55 showed a higher degree of cracking, compared to specimens with w/b = 0.45, with cracks developing both perpendicular and longitudinally.

3. Summary and discussion of results As expected, the addition of SAP to bonded mortar overlays significantly reduced overlay cracking. However, this effect depends on mix design parameters such as w/b ratio and SAP content. For mortars with a lower w/b ratio of 0.45, all investigated SAP contents prolonged the onset of cracking. However for long-term cracking, the lowest SAP content of 0.2% was ineffective, while both SAP contents of 0.4% and 0.6% resulted in a similar improvement in cracking, compared to the reference mortar. For mortars with a higher w/b ratio of 0.55, only the highest SAP content of 0.6% was found to effectively reduce cracking due to restrained shrinkage. The most important material parameters for cracking characteristics in bonded concrete and mortar overlays are shrinkage, elastic modulus, relaxation, and tensile strength. Testing these individual

material properties indicates an improvement in all these properties with SAP addition to cementitious mortar containing silica fume, which was observed for both w/b ratios that were used. The results obtained for elastic modulus and tensile relaxation suggest increasingly improving cracking behavior with increasing SAP content. This is probably a consequence of the larger amount of voids resulting from higher SAP contents, which results in larger deformability and increasing creep and relaxation. Tensile strength results suggest a general improvement with SAP addition, which is a result of improved concrete microstructure despite a presumably higher amount of larger voids. For drying shrinkage, SAP contents of about 0.4% yielded the best results, i.e. the lowest shrinkage values. This trend needs to be confirmed in further research for various binder types. However, the test results can be taken as a general indication that drying shrinkage of normal strength mortars is probably not increased by the addition of SAP. Considering the test results of the individual material properties, the positive effect of SAP addition on cracking behavior of bonded overlays can be assigned mainly to reduced elastic modulus and increased relaxation, as this is where the largest effect of SAP addition was observed. In future work, the combined influence of SAP particle properties and mass content should be investigated in view of changes

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

7

Fig. 10. Photographs of bonded mortar specimens: crack-development in mortars with w/b ratio of 0.55 (30 mm ruler as scale).

in microstructure and performance of repair mortars. The literature contains information on how to account for the influence of SAP addition on free water availability in cementitious mixes [13,14]. For example, Esteves [13] proposes a model based on Fick’s second law of diffusion to model the water absorption kinetics of SAP particles in relation to their size, concluding that larger SAP particles have higher absorption capacity. For ASR-related durability properties of cement-based materials, Esteves [13] therefore proposes that SAP particles with larger diameter are preferable since they result in larger air voids and also in a larger overall porosity. A similar deduction can probably made with respect to the cracking properties of mortars subjected to restrained deformations. Larger air voids and larger void contents, resulting from larger SAP particles, would likely result in lower values for elastic modulus and higher creep and relaxation, both of which should in turn, improve cracking characteristics. The influence of SAP particle size on mortar cracking was not part of this research project, but should be considered for future research on the optimization of repair mortar mixes containing SAP. In another research project, Esteves [15] identified that the effect of SAP on mechanical properties of cement-based mortars containing silica fume is dependent not only on physical properties

of the SAP particles, but also on mix design parameters such as aggregate addition. Further studies into these aspects will help to optimize the mix design and constituent material selection for of repair mortars containing SAP. 4. Conclusions – The addition of SAP to normal strength mortars containing silica fume significantly reduces elastic modulus and increases tensile relaxation, with increasing SAP contents being increasingly effective. Increased creep and decreased elastic modulus is a common shortfall when SAP is used in structural applications, but has a very positive effect on cracking behavior of bonded overlays. – The effect of SAP additions of 0.4% and 0.6% were found to slightly decrease drying shrinkage of normal strength mortars. – Despite a slight retardation in strength development, the longterm tensile strength of silica fume mortars can be expected to increase as a result of SAP addition. – Due to the above influences on mortar material properties, the cracking behavior of bonded overlays is significantly improved by addition of SAP. However, this effect depends on mix design

8

H. Beushausen, M. Gillmer / Cement & Concrete Composites 52 (2014) 1–8

parameters such as w/b ratio and SAP content. For lower w/b ratios of 0.45, a SAP content of 0.4% seems to be the threshold beyond which a higher addition of SAP does not improve cracking behavior any further. For mortars with a higher w/b ratio of 0.55, a SAP content of 0.6% was needed to really improve overlay performance. – The experimental results obtained in this research indicate that SAP can be used as a very effective addition to bonded mortar overlays when cracking is of concern.

Acknowledgements The authors wish to acknowledge with gratitude the Concrete Institute (TCI) and the National Research Foundation (NRF) for financial support of this work. In addition, support was given by Afrisam, PPC, and Sika (SA) Pty Ltd. The authors further thank Ole Jensen, Technical University of Denmark, for contributing a suitable SAP material for this research. References [1] Banthia N, Gupta R. Plastic shrinkage cracking in cementitious repairs and overlays. Mater Struct 2009;42:567. [2] Bentur A, Kovler K. Evaluation of early age cracking characteristics in cementitious systems. Mater Struct 2003;36:183–90. [3] Beushausen H, Alexander MG. Failure mechanisms and tensile relaxation of bonded concrete overlays subjected to differential shrinkage. Cem Concr Res 2006;36(2006):1908–14.

[4] Beushausen H, Alexander MG. Localised strain and stress in bonded concrete overlays subjected to differential shrinkage. Mater Struct 2007;40(2):189–99. [5] Jensen OM, Hansen PF. Water-entrained cement-based materials. I. Principles and theoretical background. Cem Concr Res 2001;31(4):647–54. [6] Jensen OM. Use of superabsorbent polymers in concrete. Concr Int 2013:48–52. [7] Jensen OM, Hansen PF. Water-entrained cement-based materials – II. Experimental observations. Cem Concr Res 2002;32(6):973–8. [8] Kolver K, Jensen OM., editors. RILEM TC 196-ICC: Internal Curing of Concrete, State-of-the-Art Report. France. p. 161. [9] van der Ham HWM, Koenders EAB, van Breugel K. Visco-elastic properties of concrete mixtures including Super Absorbent Polymers, creep, shrinkage & durability mechanics of concrete and concrete structures. In: Proceedings of CONCREEP 8 conference, Volume 2, Ise-Shima, Japan; 2008. p. 799–804. [10] Pierard J, Pollet V, Cauberg N. Mitigating autogenous shrinkage in HPC by internal curing using superabsorbent polymers. In: International RILEM conference on volume changes of hardening concrete: testing and mitigation; 2006. Lyngby, Denmark. [11] Mechtcherine V, Dudziak L, Schulze J, Staehr H. Internal curing by Super Absorbent Polymers (SAP) – effects on material properties of self-compacting fibre-reinforced high performance concrete. In: Jensen OM, Lura P, Kovler K, editors. RILEM Proceedings of PRO 52, volume changes of hardening concrete: testing and mitigation. RILEM Publications S.A.R.L., Copenhagen, Denmark; 2006. p. 87–98. [12] Beushausen H, Masuku C, Moyo P. Relaxation characteristics of cement mortar subjected to tensile strain. Mater Struct 2012;45(8):1181–8. [13] Esteves LP. Superabsorbent polymers: on their interaction with water and pore fluid. Cem Concr Compos 2011;33:717–24. [14] Jensen OM. Water absorption of superabsorbent polymers in a cementitious environment. In: Leung C, Wan KT, editors. Advances in construction materials through science and engineering, September 5–7 (Hong Kong, China), RILEM PRO 79; 2011. p. 22–35. [15] Esteves LP. An ongoing investigation on modeling the strength properties of water-entrained cement-based materials. In: Concrete repair, rehabilitation and retrofitting III, proceedings (ICCRRR2012). CRC Press, Taylor & Francis Group; 2012. p. 521–4.