SUPER ABSORBENT POLYMERS TO STIMULATE SELF HEALING IN ECC

Transcription

1 SUPER ABSORBENT POLYMERS TO STIMULATE SELF HEALING IN ECC J.S. Kim (1) and E. Schlangen (1) (1) Delft University of Technology, Faculty of Civil Engineering and Geoscience, Micromechanics Laboratory (MICROLAB), Delft, The Netherlands Abstract The load-bearing capacity of structure can be violated by the degradation of concrete and reinforcement. Defects in materials may also lead to poor serviceability or inconvenience in the use of a structure. For instance, cracks in concrete structures may cause carbonation of concrete and corrosion of reinforcement. Under certain conditions, the phenomenon of self-healing in concrete structures is well established. To find these conditions of self-healing is important to prolonging service life of concrete structure. Even for buildings and bridges over land, many concrete structures always serve under frequently changing weather conditions, in particular, the humidity. For new materials like Engineered Cementitious Composites (ECC) investigation into the performance of specimen under frequent changing of humidity is necessary and of great importance for evaluating the durability. In the current paper, the experimental studies for the development of an ECC using Super Absorbent Polymers (SAP) are presented. The experimental results showed that ECC specimen included SAP s performed comparably with reference ECC specimen. When specimens were under the action of wet dry cycling, the self healing properties of ECC specimen included SAP s were much inferior to that of reference ECC specimen. It is therefore suggested that including SAP s could contribute to self healing where wet dry cycling weather occurs. 1. Introduction The crack development in concrete structures leads to large permeability and consequently to durability problems. Controlling crack development and crack width in ordinary concrete has been a great issue for the structural community, while proposed measures seem either insufficient or impractical. For the last fifteen years, the development of ECC [,7] has received serious attention by structural engineers. ECC are a unique type of high performance fibre reinforced cementitious composites (HPFRCC), which feature high tensile ductility with moderate fibre fraction, typically % polyvinyl alcohol (PVA) fibres by volume. Special interest has been given to the capability of 9

2 ECC materials to deform up to high tensile strains, regularly over 3%, while maintaining very tight crack widths, shown to be on the order of to μm on average. The design of ECC, makes self-healing possible, because the material is capable to bend and crack in narrow hairlines rather than break and split in wide gaps, as the traditional cement behaves. Self-healing is a phenomenon that can repair the cracked material via chemical reactions which take place, connecting the two crack surfaces. Usually a wet environment is needed to start self healing. In the present study self healing is promoted in by the use of SAP in the mix (Schlangen et al. 1). The SAP s are filled with water during the mixing process and form in such a way water pockets in the concrete that can be used for hydration of the cement and thus self healing in a later stage. SAP s are known as additive to mitigate autogenous shrinkage in concrete (Jensen & Hansen 1). The self healing capacity of the strain hardening cementitious composites(shcc) is already improved by adding these SAP s in specimens that are cracked and subsequently stored in water (Antonopoulou 9, Tziviloglou 9). Storing specimens under water is not a realistic case. However, these SAP s can also work for specimens stored in air. The water pockets are emptied during or shortly after the first hydration. When the material cracks at a later stage no water is left anymore. But after some rain on the structure the SAP s located in the cracked zone are again filled and then slowly release the water for the self healing mechanism. This seems to be a realistic and practical scenario which is currently under investigation. First results are presented in this paper.. Experimental program.1 SAP SAP s are a group of polymeric materials that can absorb and retain extremely large amounts of a liquid relative to their own mass. Water absorbing polymers, classified as hydrogels, absorb aqueous solutions through hydrogen bonding with the water molecule. So SAP's ability to absorb water is a factor of the ionic concentration of an aqueous solution. SAP s can absorb water up to few hundred times of its own weight (Figure 1). Figure 1: A dry and a swollen SAP (Jensen & Hansen 1). 5

3 Important properties of SAPs include swelling capacity, strength and elastic modulus of the swollen gel. All these properties depend on the cross-link density of the network: the elastic modulus increases and the swelling capacity decreases with increasing cross-link density. Generally, The gel strength is firmer and can maintain particle shape even under modest pressure. Their high swelling degree has been mainly supported by macromolecular polymer network and poly-electrolyte constituents. Due to the water absorbing mechanism of polymers, when adding a sufficient quantity in a cement paste the porosity and permeability are noticeably reduced, together with the good dispersion of cement particles a more homogeneous material is formed. Therefore, polymer powder/cement composites are known to have higher flexural strength, deformability, adhesion and durability than standard cementitious materials.. materials Table 1 and show that three different mix proportion of the ECC PVA %, PVA % + SAP.5% and PVA % + SAP 1 % were investigated in this study. All three groups had the same composition as the ECC mixture, except the amount of water and SAP was different. The ECC mixture is comprised of Portland cement CEM I.5 N, fly ash, limestone powder, water, super plasticizer and % of the total volume of PVA fibers. These PVA fibers had length of mm and diameter of µm. Moreover, the SAP s were put in a proportion of.5% of the cement weight for the second group mixture and the third had 1%. Table 1: Proportion of fibres and SAP s Group Proportion 1 PVA % PVA % and.5% of SAP s 3 PVA % and 1% of SAP s Table : Mix proportion of the ECC (weight %) Group Cement Limestone... Fly Ash Water..9.9 Super Plasticizer PVA fiber (by volume) SAP

4 .3 Specimen configuration The specimen configuration in this study was taken similar as in the work of Antonopoulou and Tziviloglou [, ]. They conducted an experimental study on specimens of 1 mm-long, 3 mm-thick and 1 mm-height. In these studies, the fresh ECC was cast into moulds with the dimension of mm mm 1mm for the four-point bending test. These coupon specimens were moisture cured for hours and then demoulded. After demoulding, the coupon specimens were evenly sawn into four pieces with the dimensions of 1 mm 3mm 1mm. These specimens were used in the four-point bending test.. Environmental conditioning As described in the previous section, the structures always serve under frequently changing weather condition. So in the investigation of this study, an environmental conditioning regime has been adopted. This includes cyclic wetting and drying. Specifics of the conditioning regime are given below. The pre-cracked ECC specimens are subjected to wet and dry cycles. The specimens are submersed in water of ºC for 1 hour and subsequently dried in laboratory air for 3 days. This regime is used to simulate cyclic outdoor environments such as rainy days and unclouded days. Actually the number of raining days per year in Netherlands is 13, so it is reasonable to assume that this raining cycle occurs one time per three days. The reference groups were stored in the laboratory (approx. 5% RH, C) until final testing..5 Four point bending test The four point bending test was the main tool of this research. The overall program for the four-point bending test included four different loading and curing scheme is depicted in Figure 3. The test was used in order to characterize the strength and the deformation capacity of the ECC specimens for the three different mixtures. The support span of the four point bending test set up was 11 mm, and the load span was 3 mm as shown in Figure. The load was applied at the centre of the middle span, under displacement control at a constant speed of. mm/s. The vertical deformation at the mid cross-section of the samples was measured with two LVDTs. Five specimens were tested for each variable in the testing program. The flexural strength and deflection were calculated based on the average results of five tests. 11 mm 3 mm 1 mm P/ P/ Figure : Four point bending test 5

5 Figure 3: Loading and curing scheme 3. Experimental results 3.1 Flexural performance Four point bending test were conducted to measure the flexural mechanical properties of specimen after self healing. Under four point bending load, all the mixtures exhibit multiplecracking behaviour (Table 3). Table 3: Crack width and number of cracks on pre-cracked specimens Age of preloading Group Number of cracks Average crack width(μm) 7 days old days old As is stated above, the main method in this research was the four point bending test on specimens with the dimensions of 1 mm 3mm 1mm and this was applied to three 53

6 mixtures in the study. Each mixture has two main categories, those tested 7 days after casting and days. Each category contains three basic schemes, first was loading to final failure, second and third were preloaded up to mm and then cured in either 9 wet-dry cycle or in laboratory air for days. Figure and 5 show the flexural stress-deformation curves of the three mixtures. Only a single curve (out of the five specimens tested) curve is presented in the graphs, showing the average behaviour Group 1 Group 1 1 Figure : Loading to final failure flexural stress-deformation curves of 7 days cured ECC specimen Figure 5: Loading to final failure flexural stress-deformation curves of days cured ECC specimen ECC specimens measuring 1 mm 3mm 1mm were prepared and preloaded to mm vertical deformation at 7days and days after casting (Figure and 7) Group Group Figure : Preloading flexural stress-strain relations of 7 days cured ECC specimen Figure 7: Preloading flexural stress-strain relations of days cured ECC specimen After straining and unloading, half of the cracked specimens were exposed to 9 wet-dry cycles and the rest was stored in laboratory air condition for days. Four point bending test were conducted again on the healed specimens. In the stress-strain curve of the reloading stage, the permanent residual strain introduced in the preloading stage is not accounted for. 5

7 Figure, 9, 1 and 11 show the reloading flexural stress-strain curves of ECC specimens after conditioning Group Group 1 Figure : Reloading after days in laboratory air condition flexural stressstrain relation of 7 days cured ECC specimen Figure 9: Reloading after 9 wet-dry cycles flexural stress-strain relation of 7 days cured ECC specimen Group Group 1 Figure 1: Reloading after days in laboratory air condition flexural stressstrain relation of days cured ECC specimen Figure 11: Reloading after 9 wet-dry cycles flexural stress-strain relation of days cured ECC specimen As stated in the introduction, the purpose of the present research is to investigate it the SAP s located in the cracked zone can be filled again and subsequently release the water for the self-healing mechanism. Above figures show that all specimens that are air cured have a flexural stress-strain curve which has a smooth and even shape. This can be explained by the fact that when the preloaded specimens are reloaded, there were no new crack formed but the pre-existing cracks were re-opened and extended. It shows that the self-healing did not occur or occurred only partial. 55

8 In contrast, the stress-strain curves obtained from the specimens that went through the wetdry cycles of group and 3 had a significant roughness and unevenness. The principle of multiple cracking in ECC-like materials lies on the assumption that, after the first crack has formed, the energy needed to let this crack grow is lager than the energy needed to form a new crack. So the experimental results obtained in these tests could be explained by the fact that new cracks occurred because the strength of the self healed part was higher than the strength of existing matrix. The specimens of group 1 that followed also the wet-dry cycles (figure 9) have again a smooth and even shape. The specimens of group 1 were made of ECC without SAP s. In that case the material can not hold enough water during the wetting cycle to create any significant self healing in the sample. 3. Flexural strength and recovery In order to evaluate the performance in strength of cured samples (both wet dry cycles and air cured either preloaded 7 or days after casting) and to be able to compare it with the correspondent value of bending strength of virgin specimens, the strength ratio was calculated. The ratio was computed from equation (1). Cured Virgin Strength Recovery Ratio = Cured flexural stress (1) Table summarizes the ultimate bending stress of each mixture for the virgin and the water and air cured specimens. As it can be seen from Table 5, the strength of cyclic wet-dry cured specimens exceeds the strength of the virgins in all mixtures for both cases of 35 and 5 days old. Particularly, strength recovery ratio of group and 3 is higher than the specimens of group 1. Apparently, this is also evidence that SAP s kept the water in the cracked zone and helped the self healing. In Table the strength recovery for air cured specimen is given. It can be seen that the influence of SAP is limited. In case of 35 days curing the SAP still contribute, because they probably help to store a small amount of water that can be used for self healing. However at 5 days air curing the contribution of SAP s has a negative effect. Table : Ultimate flexural stress (MPa) Group 7 days days_water cured days_air cured days days_water cured days_air cured

9 Table 5: Strength recovery ratio for the cyclic wet-dry cured specimens (%) 5 days 5 days Table : Strength recovery ratio for the air cured specimens (%) 5 days 5 days CONCLUSIONS In the present experimental research, the self-healing performance of preloaded ECC was investigated. A group of ECC were mixed with Portland cement CEM I.5 N, fly ash, limestone powder, water, super plasticizer, % of the total volume of PVA fibers and SAP s. The mix proportion was designed experimentally by adjusting the amount of SAP s, and four point bending tests were conducted to preloaded samples of three different mixtures up to a deformation of mm, both on the 7 th and th day after casting. The pre-cracked specimens were cured in cyclic wetting and drying condition or in laboratory air condition for days. The following conclusions can be derived from this experimental study: 1. Under four point bending test, multiple-cracking behaviour is commonly observed in all groups of ECC specimen. But the crack width and spacing of each group varies considerably.. The air cured ECC specimens and the wet-dry cycling cured ECC specimens of group 1 show none or limited self-healing of cracks. The wet-dry cycling cured ECC specimens of group and 3 (containing SAP s), by contrast, show considerable amount of self-healing. 3. The results of the present paper shows that the group which contains.5% of SAP s achieves better strength recovery ratio values compared with the group 3 which contains 1%. Future research will focus on obtaining optimum mix designs for ECC contating SAP s with respect to self healing properties. REFERENCES [1] Antonopoulou, S., Self healing in ECC materials with high con-tent of different microfibers and micro particles, MSc Thesis, Delft University of Technology, 9. [] Jensen, O.M., and Hansen, P.F., Water-entrained cement-based materials: I. Principles and theoretical background, Cem. Concr. Res. 31() 1, 7 5. [3] Jensen, O.M., and Hansen, P.F., Water-entrained cement-based materials: II. Experimental observations, Cem. Concr. Res. 3, [] Schlangen, E., Jonkers, H., Qian, S., and Garcia, A., Recent developments on self healing of concrete, Proc. FraMCoS7, Jeju, Korea, May. 3-7, 1. 57

10 [5] Tziviloglou, E., Self-healing in ECC materials with low content of different microfibers and micro-particles, MSc Thesis, Delft University of Technology, 9. [] Li, V.C., and Yang, E.H., Self healing in concrete materials, in: Self Healing Materials. An Alternative Approach to Centuries of Materials Science, , 7. [7] Li, V.C., Engineered cementitious composites Tailored composites through micromechanical modeling, in: Fiber Reinforced Concrete: Present and the Future, Canadian Society for Civil Engineering, Montreal, -97,