Crack Propagation in Cement Mortar Exposed to External Sulfate Attack

- This study displays experimental investigation to find out the effect of the external sulfate attack on crack propagation in cement mortar cubes exposed to two sulfate solutions. For this purpose, five mixes of mortar are designed to cast 90 cubes (18 cubes of each mix) and two sulfate solutions (magnesium sulfate MgSO 4 and sodium sulfate Na 2 SO 4 ), each of 34000 p.p.m SO 4-2 concentration are prepared. The cubes are evenly distributed in each of the solutions and in tap water as well. The testing technique is carried out compressive strength and Ultrasonic Pulse Velocity (UPV) tests simultaneously to specify the stress/strength ratio at which cracks propagate. The tests are carried out after 56 and 150 days of exposing to sulfate solutions. The results show that crack propagation takes place in specimens exposed to sulfate solutions at stress/strength ratios higher than those which are kept in tap water. The main conclusion, according to this study results, is that sulfate particles delay the propagation of cracks in cement mortar cubes.


Introduction
Sulfate attack has many types; Ettringite formation, thomosite formation, magnesium sulfate attack and physical sulfate attack (PSA) -a subset of physical salt attack involving Sodium Sulfate [1]. The most common explanation to the deterioration caused by the sulfate attack is that sulfate attack causes expansion resulted from the formation of ettringite and gypsum. However, most, if not all reliable concrete references recognize the attack of magnesium sulfate as the most severe attack. Besides, some concrete references talk about the physical effect (not chemical). The initiation and propagation of cracking in concrete and other cementitious materials is a governing mechanism for many physical and mechanical material properties [2]. In concrete which is not subjected to any attack, the previous experimental results showed that microcracks initiation takes place at a stressstrength ratio ranging between 0.15 and 0.44 (rather than at a specific value) [3]. Microcracks already exist in concrete at the interface between aggregate and hydrated cement paste even prior to the application of load. These cracks remain stable and do not grow under stress up to 30 percent of the ultimate strength [4]. However, when concrete exposed to sulfate solution of high concentration, its physical and mechanical properties will be changed. This change may include crack initiation and propagation. Cracks in concrete are macro and micro, the former is noticeable by naked eyes, while the latter is not. Non-destructive test, specifically, the ultrasonic pulse velocity device is used (in this research) to specify the point at which the velocity starts to decrease continuously with loading increment till failure. This point is considered to be the initiation of crack propagation (this is the concept of this research). The UPV is changing with the load increment or in other words, with stressstrength ratio. The change in UPV gives a notion about the crack behavior.A comparison can be done between the UPV through specimens in tap water and specimens in sulfate solution(s) to find out the effect of sulfate compounds on crack behavior in hydraulic Portland cement mortar.

The Significance of Research
This research aims to find out the effect of the external sulfate attack on crack initiation and propagation in mortar. This aim needs answers to the following questions: Does external sulfate attack change the stressstrength ratio at which the crack starts to propagate? Does external sulfate attack change the behavior of crack propagation? Is the effect of magnesium sulfate similar to that of sodium sulfate on crack initiation and propagation? 214 1-Casting mortar cubes (50mm side size) of five mixes, total number of cubes are 90. The materials used are, Ordinary Portland Cement in one mix and Sulfate Resistant Cement in four mixes and river sand type M (the chemical analysis and properties of the materials are listed in the appendix). Mixing is done manually and compaction is done by tamping according to ASTM 109-87 [5]. Table 1 shows the cement type used in each of the five mixes, mix proportion and number of cubes. Mix 2 and 3 of the similar W/C ratio, which is 0.485, this ratio is specified in ASTM 1012 (the specification to test the external sulfate attack) [6]. Mix 1 is a 0.35 W/C ratio, which is low ratio and the mix needs more physical work than that of the standard compaction, so the mix is porous and allows to sulfate solution to penetrate easily into the cube body. Mixes 4 and 5 of W/C ratios 0.55 and 0.65 respectively, are to get different strengths. 2-Preparing MgSO 4 solution and Na 2 SO 4 solution, each of high SO 4 -2 concentration (34000 ppm). The salt compounds used are MgSO 4 .7H 2 O, its molecular weight is a 246gm / mole and Na 2 SO 4 , its molecular weight is 142gm/mole. A-The quantity of MgSO 4 . 7H 2 O dissolved in distilled water to produce a solution of (34000 ppm SO 4 -2 ) is: 87.125gm/litter (the mass number of the radical SO 4 -2 is 96, so it represents 0.39 of MgSO 4 .7H 2 O mass number, so 0.39*87.125 gm=34 gm, which gives a solution of 34000 ppm when dissolved in one liter of distilled water). B. The quantity of Na 2 SO 4 dissolved in distilled water to produce a solution of (34000 ppm SO 4 -2 ) is: 50.291 gm/liter. The volume of each solution is 3.5 times the total volume of the specimens (according to ASTM 1012, Note 2) [6]. {5 mixes*6 cubes of each mix* cube volume (5cm) 3 }*3.5= 13125cm 3 = 13.125 litter. Note that the concentration of SO 4 -2 is very severe according to ACI 318-11 classification [7], that it is much greater than 10000 p, p, m (10000 ppm is considered very severe) [ditto]. 3. Curing all the cubes for 28 days in tap water. 4. 6 cubes of each mix are put in MgSO 4 solution, the same is in Na 2 SO 4 solution and the rest is kept in tap water for another 150 days.

Test Method
Testing cubes compressive strength by using an electrical machine of 200KN capacity. Measurements are taken to reach the failure with no less than 20 and no more than 80 seconds, according to ASTM C 109. A PUNDIT of sensitivity 0.01 microsecond is used with transducers of frequency 200KH to suit the specimen dimension (50 mm) according to BS 4408: Part 5:1974 [8]. The two tests are carried out on each sample (cube) simultaneously; compressive test and ultrasonic pulse velocity test (as it is shown in the picture below). The ultrasonic pulse velocity test is one of the indirect methods that have been used to detect and observe microcracking in concrete [9]. This is done by recording the UPV with each increment of 1.2 MPa in compression loading. Then, when the ultrasonic velocity starts to decrease continuously with compression, that very point is considered (in this research) as the inflection point, which represents (according to this research) the initiation of crack propagation.

Results and Discussion
Generally, the initiation of crack propagation of all mixes in tap water after 74 days locates in the range (0.166-0.52) stress-strength ratio and in the range (0.16-0.52) stress/strength ratio after 150 days. The crack propagation of all the mixes in MgSO 4 solution after 56 days locates in the range (0.33-0.7) stress-strength ratio and (0.28-0.75) stress-strength ratio after 150 days. The crack propagation of all the mixes in MgSO 4 solution after 56 days locates in the range (0.33-0.7) stress-strength ratio and (0.28-0.75) stressstrength ratio after 150 days.
The crack propagation of all the mixes inNa 2 SO 4 solution after 56 days locates in the range (0.16-0.75) stress-strength ratio and (0.14-0.74) stress-strength ratio after 150 days. Table 2 shows the strength of each mix in each of the two sulfate solutions and in tap water, as well as it shows the stress-strength ratio at which the crack propagates. Mix1: the UPV through the specimens in tap water for (84) days has two stages, in the 1 st one, it is constant with loading, while in the 2 nd , it decreases with load increment. In sulfate solutions for (56) days, the UPV has three stages as it is shown in Table 3. The UPV through the specimens in tap water for (178) days as well as in sulfate solutions for (150)days, has two recognized stages, in the 1 st one, it is constant with loading, while in the 2 nd , it decreases with load increment, as it is shown in Table 4. Mix 2: the UPV through the specimens in tap water for both ages (84 and 178) days has two stages, in the 1 st one, it is constant with loading, while in the 2 nd , it decreases with load increment. In sulfate solutions for both (56 and 150) days, the UPV has three stages as it is shown in Table 5 and 6. Mix 3: the UPV through the specimens in tap water for both (84 and 178) days,as well as in sulfate solutions for both (56 and 150) days, has two stages, in the 1 st one, it is constant with loading, while in the 2 nd , it decreases with load increment, as it is shown in Table 7 and 8. Mix 4:the UPV through the specimens in tap water for both (84 and 178) days has two stages, in the 1 st one, it is steady with loading, while in the 2 nd , it decreases with load increment, in sulfate solutions for (56 and 150) days as well, as it is shown in Table  9) and 10. Mix 5: the UPV through the specimens in tap water for both (84 and 178) days has two stages, in the 1 st one, it is constant with loading, while in the 2 nd , it decreases with load increment, in sulfate solutions for both (56 and 150) days as well, as it is shown in Table  11) and 12.
The results clearly show that sulfate solutions affect the strength, the density (that the UPV is a function of the density) and the stress/strength ratio where the crack starts to propagate in all the specimens. The effect of sulfate solutions on crack propagation (the scope of this research) is that they (sulfate solutions) delay the initiation of crack propagation and change the behavior of cracking. The sulfate compounds ions and/or the compounds resulted from the reaction(s) between sulfate compounds ions and the other compounds or ions in the media (the body of the mortar) prevent the initiation of crack propagation. This is due to the additional surface area of the sulfate compounds or their products which leads to the need to an additional energy to cause crack propagation [10].
With regard to the behavior of cracking, the results show that the specimens in tap water have two phases; constant UPV with loading then the UPV starts and keeps decreasing till failure. Most of the results of the specimens in sulfate solutions show three phases; the 1 st phase is that the UPV decreases with the start of loading. This might be due to the internal stresses resulted from the sulfate attack. The values of these stresses are added to the mechanical stress. The 2 nd phase is that the UPV stays as it is with loading. The 3 rd and last phase is that the UPV decreases continuously with loading till failure.

Conclusions
1-The initiation of crack propagation in cement mortar specimens exposed to sulfate solutions takes place at a stressstrength ratio higher than that of specimens in tap water.