2.1. Properties of Hardened Concrete
The water absorptions of R40-CON, R40-NS5, R40-NS7.5, and R40-NS10 were 3.38%, 4.17%, 4.18%, and 4.20%, respectively. And the total porosities of corresponding admixtures were 7.74%, 9.48%, 9.67%, and 10.06%, severally. It was confirmed, from test results, that the MSCA increased both absorption and porosity of concrete, but the modified sulfur content at the coated aggregates slightly affected them. This result can be explained as followings: (1) the coated aggregate with modified sulfur causes its absorption to be reduced; (2) the space occupied by surplus water forms the high porosity in the MSCA concrete. The bulk densities of R40-CON, R40-NS5, R40-NS7.5, and R40-NS10 were 2.29, 2.26, 2.26, and 2.25 Mg/m3, respectively. Those were not much different regardless of amount of modified sulfur.
2.2. Compressive Strength
Figure 1 shows the compressive strength changes in the concrete cylinders with MSCAs, which were coated with various amounts of sulfur, in relation to W/C and curing time. The compressive strengths shown in
Figure 1 are the average values of three specimens for each mixture. For admixtures with W/C ratio of 40%, the compressive strengths of R40-NS5, R40-NS7.5 and R40-NS10 at a 7-day age had decreased by 19%, 23% and 28% relative to R40-CON, respectively (see
Figure 1a). At a 28-day age, the compressive strength of corresponding admixtures decreased by 6%, 20%, and 32%. At 56 days, these values had decreased to 5%, 18%, and 33%. The reduction rate of compressive strength of R40-NS5 decreased with passing curing age, whereas those of R40-N7.5 and R40-N10 were almost constant regardless of curing age. It was obvious from test results that the high dosage of modified sulfur may cause the cement setting problem and thus could delay the hydration process in concrete.
Figure 1.
Compressive strength of cylinders with (a) water-cement ratio (W/C) of 40%; (b) W/C of 45%; and (c) W/C of 50% with respect to W/C and curing time.
Figure 1.
Compressive strength of cylinders with (a) water-cement ratio (W/C) of 40%; (b) W/C of 45%; and (c) W/C of 50% with respect to W/C and curing time.
The compressive strength of the MSCA concrete decreased further when a higher amount of modified sulfur was used to coat the aggregate. There could be three reasons: (1) the remains of coated sulfur can act as impurities or light-weight aggregate in the concrete; (2) the additional quantity of superplasticizer needed to obtain an identical slump in concrete with more sulfur can lead to the formation of more pores in the cement paste; (3) the more amounts of modified sulfur owing to its low stiffness and strength result in the weaker microstructure of ITZ, leading to a decreased compressive strength.
Of the specimens with 5% MSCA, the compressive strengths at 56 days in R40-NS5, R45-NS5, and R50-NS5 cylinder samples (see
Figure 1b,c) had decreased by 5%, 9%, and 11% from that of each control sample with normal aggregate. In other words, the compressive strength tended to decrease with an increased W/C. A possible explanation is that the surplus of modified sulfur may have a greater effect on the strength of concrete with a higher W/C because the amount of cement decreased with the higher W/C.
2.3. Flexural Strength
The flexural strengths of the prism beams are plotted in
Figure 2. The strengths shown in
Figure 2 are the averaged values of three specimens for each mixture. The flexural strengths at 28 and 56 days were almost the same for each W/C. Of the specimens with a W/C of 40% (see
Figure 2a), the strength at 28 days of R40-NS5 had increased by about 0.7% as compared to the control specimen of R40-CON. On the contrary, the strengths at 28 days of the others had decreased up to 15%–16%. The results of this experiment are in agreement with the experimental result obtained in a study by Morin
et al. [
21], which used latex-coated aggregate. The flexural strength loss in the concrete with aggregate coated by more than 7.5% of modified sulfur may have the same explanation as the loss of compressive strength discussed above.
Figure 2.
Flexural strength of prism beams with (a) W/C of 40%; (b) W/C of 45%; and (c) W/C of 50% in relation to W/C and curing time.
Figure 2.
Flexural strength of prism beams with (a) W/C of 40%; (b) W/C of 45%; and (c) W/C of 50% in relation to W/C and curing time.
Additionally, the flexural strengths at 56 days of specimens with different W/C and 5% MSCA had increased when compared with each control sample, but the variation rate was slight (e.g., 0.4% for W/C = 40%, 2.6% for W/C = 45% and 0.5% for W/C = 50%). In other words, the 5% MSCA did not significantly affect the flexural strength loss regardless of W/C.
2.4. Length Change
The length change of concrete with passing time is presented in
Figure 3. The averaged length changes of R40-CON, R40-N45, R40-N7.5, and R40-N10 admixtures at 7-day age were −2.50 × 10
−4, −2.10 × 10
−4, −2.03 × 10
−4, and −2.17 × 10
−4 mm/mm, repetitively (see
Figure 3a). The MSCA led to a slight decrease of length change, but the amount of modified sulfur at the MSCA did not significantly affect the length change. At the age of 56 days, the length changes of corresponding admixtures had increased by −6.57 × 10
−4, −6.47 × 10
−4, −6.50 × 10
−4, and −6.53 × 10
−4 mm/mm, respectively (see
Figure 3a). The length change of MSCA concrete increased with the higher content of modified sulfur, but it was determined that the modified sulfur had little effect on the length changes because the change rates of MACA concrete were almost similar to that of R40-CON specimen. The length changes at 7–56 days in specimens having a W/C of 45% (see
Figure 3b) were greater in each control sample than in the specimen with the MSCA. On the other hand, the length changes of two specimens with a 50% W/C (see
Figure 3c) went up and down until 28 days. After 28 days of curing, the length changes of two specimens were almost identical. Based on the length change observations, it was found that the length change due to the MSCA was slightly smaller than or equal to that of the control specimens.
Figure 3.
Length change of prism beams with (a) W/C of 40%; (b) W/C of 45%; and (c) W/C of 50% in relation to W/C and curing time (unit: mm/mm).
Figure 3.
Length change of prism beams with (a) W/C of 40%; (b) W/C of 45%; and (c) W/C of 50% in relation to W/C and curing time (unit: mm/mm).
2.5. Freezing and Thawing
The relative dynamic modulus of elasticity and weight change in the freezing and thawing test are presented in
Figure 4 and
Figure 5, respectively. The freezing and thawing test was carried out using the prism beam specimens with only W/C of 40%. The freezing and thawing test was performed for only admixtures with a W/C of 40% because of the capacity of equipment. After 30 cycles of lowering and raising the temperature, the relative dynamic modulus of elasticity became 91.0% in beam R40-CON, 91.9% in R40-NS5, 92.1% in R40-NS7.5, and 92.8% in R40-NS10, respectively. The relative dynamic modulus of elasticity changed to 88.1%, 92.7%, 91.8%, and 90.9% at 150 cycles. That in beam R40-CON with normal aggregate rapidly decreased after 150 cycles, and then it reached 61.4% at 300 cycles. However, the relative dynamic modulus of elasticity of the specimens with the MSCA barely changed while maintaining the range of 90%–95%.
The weight losses for each sample (e.g., R40-CON, R40-NS5, R40-NS7.5 and R40-NS10 prism beams) were 0.17%, 0.16%, 0.0% and 0.14% at 30 cycles, respectively. When 300 cycles passed, the weight losses in each sample were 2.8%, 1.4%, 1.23% and 1.2%. The loss of the control specimen was almost twice as large as that of the MSCA concrete. The weight losses of the MSCA concrete with different amounts of sulfur were not significantly different.
Figure 6 shows the corrosion changes of four samples (e.g., R40-CON, R40-NS5, R40-NS7.5 and R40-NS10 prism beams) with passing cycles.
The surface of the R40-CON sample started popping out considerably from 90 cycles, and exposure of the aggregate in the concrete was seen plainly after 180 cycles had passed. On half of the surface, aggregate exposure developed after 300 cycles, as illustrated in
Figure 6a. In the MSCA concrete, the initial porosity faded after 300 cycles, but aggregate exposure due to popping out never occurred, as shown in
Figure 6b–d. Based on the surface porosity observation, the MSCA significantly improved the freezing and thawing resistance. It was observed that the corrosion progress in R40-CON had been visually much more rapid than that in samples with MSCA. When water freezes, a volume expansion of 9.1% occurs, and the expansion moves to opening gap in cement paste. If there are no pores, mainly caused by entrained air, to relax the expansion, high pressure builds up in the concrete, and this is one cause of deterioration. Therefore, it was obvious that the use of MSCA results in the high porosity in the concrete and thus develops the resistance against the freezing and thawing of concrete.
Figure 4.
Relative dynamic modulus of elasticity by freezing-thawing cycles.
Figure 4.
Relative dynamic modulus of elasticity by freezing-thawing cycles.
Figure 5.
Weight change by freezing-thawing cycles.
Figure 5.
Weight change by freezing-thawing cycles.
Figure 6.
Surface status of (a) R40-CON; (b) R40-NS5; (c) R40-NS7.5; and (d) R40-NS10 specimens according to number of cycles.
Figure 6.
Surface status of (a) R40-CON; (b) R40-NS5; (c) R40-NS7.5; and (d) R40-NS10 specimens according to number of cycles.
2.6. Sulfate Resistance
The resistance of the MSCA concrete to sulfuric acid was investigated every 7 days for 4 weeks. The effect of MSCA on the sulfate attack resistance of concrete is shown in
Figure 7. Of the samples with W/C = 40% (see
Figure 7a), the weight in only RS40-CON with normal aggregate decreased by 1.5% after 1 week of immersion; however, that of the other samples with MSCA increased up to about 2% after 1 week. The weight increase may have been caused by absorption of the liquid phase due to concrete surface porosity. After that, the weight in all samples decreased gradually until 4 weeks. The final weight loss of the sample with normal aggregate was around 9% but those of the samples with MSCA were 1.2%–1.6%. There was little mass change, and it was insignificant for the samples with MSCA. It is known that the dilute hydrochloric and sulfuric acids, in general, cannot affect the sulfur. The sand and aggregate used in the concrete mixture consisted of mineral oxides of alkalinity. Acid damage on concrete is started from the chemical reactions of basic oxides, acid oxides, and amphoteric oxides. Because the coating modified sulfur prevents reaction between acid and basic oxides, the resistance to sulfate improves.
As compared to samples having MSCA with 5% modified sulfur, the weight losses were 1.2% in the W/C = 40%, 1.0% in the W/C = 45%, and 0.3% in the W/C = 50% samples after 4 weeks of immersion. Even though the loss of weight in two samples (R45-CON andR50-CON) with uncoated aggregate was not significantly different, the weight loss tended to be reduced with increased W/C. As test results, a higher W/C enhanced the acid resistance of the modified sulfur concrete.
Corrosion probably developed on the sample surfaces in open pores that were not coated or protected by sulfur and lasted until the test ended (see
Figure 8). The specimen of R40-CON immersed in sulfuric acid solutions for 4 weeks showed high corrosion. By contrast, the specimens with MSCA showed few signs of corrosion. This indicates that MSCA contributes to better acid resistance in comparison with the control sample with normal aggregate.
Figure 7.
Weight change of cylinders with (a) W/C of 40%; (b) W/C of 45%; and (c) W/C of 50% in relation to W/C and immersion time.
Figure 7.
Weight change of cylinders with (a) W/C of 40%; (b) W/C of 45%; and (c) W/C of 50% in relation to W/C and immersion time.
Figure 8.
Surfaces of (a) R40-CON; (b) R40-NS5; (c) R40-NS7.5; and (d) R40-NS10 samples after 4 weeks of immersion in sulfuric acid.
Figure 8.
Surfaces of (a) R40-CON; (b) R40-NS5; (c) R40-NS7.5; and (d) R40-NS10 samples after 4 weeks of immersion in sulfuric acid.