Mechanical Properties and Microstructure of Rubber Concrete under Coupling Action of Sulfate Attack and Dry–Wet Cycle

Abstract

In order to study the mechanical properties of rubber concrete (RC) with different rubber particle sizes after dry–wet cycles in a sulfate environment, apparent morphology analysis, mass loss analysis, relative dynamic elastic modulus analysis, compressive strength loss analysis, internal microscopic characteristics and deterioration degree analysis of ordinary concrete (NC) and rubber concrete after dry–wet cycles were compared and analyzed. The results show that with the increase in the number of dry–wet cycles, the surface caves of rubber concrete increase, the internal microcracks develop and penetrate, and the macroscopic strength increases first and then decreases significantly. The high elasticity of rubber effectively improves the expansion force caused by sulfate attack and the dry–wet cycle. The deterioration degree of RC in each dry–wet cycle stage is obviously better than that of NC. When the rubber particle size is 0.85 mm, the performance of the sample is the best. After 120 days of dry–wet cycle, the compressive strength is reduced by 37.4%, and the compressive strength of concrete with a rubber particle size of 0.85 mm is reduced by 11.2%. After cyclic loading, the deterioration degree of concrete is 5.1% lower than that of ordinary concrete.

1. Introduction

In recent years, with the rapid development of the automobile industry, there have been more and more waste rubber tires, and the utilization of waste rubber has also been widely studied [1,2,3]. It has been widely used in recent years to make rubber concrete by mixing waste rubber into rubber particles or rubber powder into concrete [4,5,6,7]. Compared with ordinary concrete, concrete mixed with a certain amount of rubber improves the frost resistance [8], toughness [9] and impact resistance [10] of concrete under certain conditions. Hanbing Liu et al. [11] used rubber to partially replace fine aggregate and mixture to produce rubber concrete. With an increase in the rubber content from 0% to 20%, the compressive strength of rubber concrete decreases from 34.76 MPa to 33.41 MPa, and the elastic modulus decreases from 31.75 GPa to 24.73 GPa. Osama Youssf et al. [12] prepared concrete mixtures with rubber contents of 0%, 10%, 20%, 30%, 40% and 50%. Different mechanical properties were measured. The results showed that the impact resistance of concrete increased by about 3.5 times when the rubber content of concrete increased from 0% to 50%. It has great potential in structural applications, especially in seismic zones. Jiaqing Wang et al. [13] prepared fiber rubber concrete by adding polypropylene fiber and rubber particles. Macro polypropylene fiber and rubber aggregate can improve the fracture energy of ordinary concrete. The ultrasonic pulse velocity shows that the quality of concrete samples is good. The fracture morphology and ESEM images show the positive effect of rubber polymer and polypropylene fiber on crack propagation. The incorporation of rubber particles can effectively alleviate the pollution problem of waste rubber tires and has good development prospects.

China contains a large amount of sulphate saline soil [14], which subtly erodes surrounding concrete houses, roads and tunnels [15]. Sulfate is a corrosive medium, which is extremely harmful and has a great influence on the durability of concrete [16,17,18]. The hydration products of ordinary Portland cement react with sulfate to form expansive crystals, such as ettringite, gypsum, etc., which produce volume expansion to induce the development of micro-cracks, and ultimately damage the concrete [19,20,21]. Dry–wet cycles caused by environmental conditions such as day and night temperature differences and rainy and snowy seasons will affect the strength of concrete. Dry–wet cycles accelerate the destruction process of concrete by sulfate erosion [22,23]. Jianming Gao et al. [24] outlined that, compared with the single sulfate attack damage process, the peak value of the dynamic elastic modulus under simple sulfate erosion is about 1.035. After 174 days of immersion, the dynamic elastic modulus decreased to 1.024. The dynamic elastic modulus value was 1.03 under the action of composite sulfate erosion and a 40% bending load, while the dynamic elastic modulus value decreased to 0.935 under the combined action of the sulfate and dry–wet cycle. Both the bending load and dry–wet cycle can accelerate the damage process of sulfate attacking concrete. Wu Ruidong et al. [25] carried out the sulfate dry–wet cycle test of concrete with different contents of metal tailings powder. The results show that the proper addition of metal tailings powder can improve the sulfate corrosion resistance of C50 concrete. Wei Tian et al. [26] revealed the damage evolution of concrete by scanning electron microscopy (SEM) and computerized tomography (CT); in the early stage of sulfate attack, the pore structure of concrete was destroyed, ettringite and gypsum were formed, and the quality and strength increased. As the deterioration progresses, the expansion force of the product and the salt crystallization pressure of the sulfate crystal act on the inner wall of the concrete and accelerate the deterioration. The quality and strength of concrete decreased sharply, indicating that the quality and strength of concrete increased first and then decreased with the increase in the number of dry and wet cycles of sulfate attack. Jin-Jun Guo et al. [27] studied the effect of different dry–wet time ratios on sulfate attack of concrete by designing different dry–wet time ratios. When the dry–wet ratio was 5:1, the compressive strength and splitting tensile strength of 252 days were reduced by 24% and 11%, respectively, compared with the initial value, and the concrete deterioration was the most serious. At present, the research on the dry–wet cycle of ordinary concrete by sulfate attack is relatively mature [28,29], but there is little research on the dry–wet cycle of rubber concrete by sulfate attack. Concrete roads and tunnels in the sulfate saline soil area are not only subjected to dry–wet cycles in a sulfate environment but are also damaged by cyclic loading [30,31,32]. Therefore, it is of great significance to ensure the durability and fatigue resistance of rubber concrete (RC) after a dry–wet cycle in a sulfate environment for the application of green concrete [33,34].

Therefore, in this paper, a dry–wet cycle test and cyclic loading–unloading test were carried out on rubber concrete with different rubber particle sizes in a sulfate environment. The damage and deterioration degree of RC after different dry–wet cycle stages and the fatigue performance of RC after cyclic loading–unloading were measured and analyzed. The internal damage of concrete was reflected by non-destructive ultrasonic testing technology [35] and scanning electron microscopy, and the microscopic mechanism was analyzed.

2. Test

2.1. Raw Materials

The test cement was P·C42.5-grade composite Portland cement produced by Huainan Bagongshan. The fly ash was grade I fly ash produced by Huainan Pingwei Power Plant. The main chemical composition of the cementitious material is shown in

Table 1. Chemical composition of cementin material.

Component		CaO	SiO2	Fe2O3	MgO	Al2O3	Na2O	SO3	Ignition Loss
content/%	cement	63.11	22.60	4.38	1.46	5.03	-	2.24	1.18
	fly ash	2.45	53.21	4.09	0.39	34.75	1.92	-	4.05

. The medium-sized sand of Huaihe River was used, with a fineness modulus of 2.67, bulk density of 1450 kg/m³ and apparent density of 2650 kg/m³. The stones used were gravel melon seeds with a particle size distribution of 5~15 mm. The rubber was produced by Dujiangyan Huayi Rubber Co., Ltd. (Deyang, China) The rubber particle size was 0.85 mm, 1–3 mm and 3–6 mm. The chemical composition of rubber particles is shown in

Table 2. Chemical composition of rubber particles.

Ultimate Composition	S	Si	Zn	C	O
content/%	2.86	0.11	1.63	92.19	3.23

, and the parameter index is shown in

Table 3. Rubber particle parameter index.

Rubber Particle Size/mm	Breaking Strength/MPa	Breaking Elongation/%	Iron Content/%	Fiber Content/%	Heating Loss/%	Ash Content/%	Sieve Residue/%
0.85	17.2	584	0.02	0	0.58	8.45	0.012
1–3
3–6

. The HPWR standard high-performance water-reducing agent produced by Shaanxi Qinfen Building Materials Factory (Weinan, China) in China was used as the water-reducing agent; the main component is polycarboxylic acid, the proportion is 1.0%, and the water reduction rate is 25.0%. The mixing water used was ordinary tap water.

2.2. Mix Proportion and Specimen Preparation

The test refers to the China ordinary concrete mix design specification to design the mix ratio, and the final mix ratio of ordinary concrete is determined as water:sand:stone:cement:fly ash:water reducer = 153:751:1121:330:43:3.5. The test rubber has 0.85 mm, 1–3 mm and 3–6 mm (three kinds of rubber particle size), with the rubber particles added according to a cementing material quality of 10% of the volume to an equal volume instead of sand. The rubber:concrete mix ratio is shown in

Table 4. Rubber concrete mixture ratio/(kg·m⁻³).

Concrete Number	Cementing Material		Fine Aggregate		Gravel	Water	Water Reducer
	Cement	Fly Ash	Sand	Rubber
RC-0.85	330	43	680.6	37.3	1121	153	3.5
RC-1-3	330	43	678.8	37.3	1121	153	3.5
RC-3-6	330	43	675.2	37.3	1121	153	3.5

.

The concrete specimens were prepared at room temperature. The concrete specimens in this test were cylindrical specimens of Φ50 mm × 100 mm. The specimens were placed in a standard curing room (relative humidity ≥ 95%, temperature 20 ± 2 °C) for 24 h after the concrete was molded. After the specimens were demoulded, they were placed in a saturated calcium hydroxide solution and maintained in a standard curing room for 90 days, and then the test was started. The concrete specimens were divided into four groups: NC, RC-0.85, RC-1-3 and RC-3-6. The uniaxial compressive strength and fatigue failure strength of each group of specimens after 0, 30, 60, 90, 120 cycles of dry and wet cycles were tested. Each group has 30 specimens, making up a total of 120 specimens.

2.3. Test Scheme Design

In this paper, the particle size of rubber particles is taken as the research parameter, and RC is taken as the research object. Among them, rubber particles are used as fine aggregate to replace sand in an equal volume, the dosage is designed to be 10%, and the particle sizes of rubber particles are 0.85 mm, 1–3 mm, and 3–6 mm, respectively. The high temperature of 105 ± 5 °C was set up in the electrothermal constant-temperature drying box to dry the test block, so as to avoid the error influence of the internal moisture of concrete on the dry–wet cycle damage, and the initial quality was weighed after drying and cooling to room temperature.

The dry–wet cycle test was carried out in an electrothermal constant-temperature drying oven. The sample was taken out after soaking in 5% Na₂SO₄ solution for 16 h, and then placed in an oven at 80 °C for 6 h. After drying, it was cooled for 2 h. One dry–wet cycle was 24 h, and the number of dry–wet cycles was 0, 30, 60, 90, and 120 times. To ensure the concentration of sodium chloride solution requirements, the solution was replaced every 30 days. After every 30 dry–wet cycles, the test block was placed in an electrothermal constant temperature drying oven and dried. After wiping the residual sulfate with clear water at both ends of the test block, the drying quality of RC and NC was measured. The NN-4B non-metallic ultrasonic detection analyzer was used to determine the ultrasonic parameters [36]. The sampling period was 0.4 µs and the emission voltage was 500 V.

The cyclic loading and unloading tests were carried out using an RDL-200 electronic creep relaxation tester. Two different test loading paths were designed for RC and NC with different dry–wet cycles, which were the uniaxial compression test path and cyclic loading–unloading test path. The uniaxial compression test adopts displacement control, the speed is set to 1 mm/min, until the test block is destroyed, and the uniaxial compressive failure load of the cylinder is measured. In the cyclic loading and unloading test, the test block is pre-loaded with a 500 N force to ensure the correct alignment and full contact between the test block and the test device. The test process adopts stress control, and the loading and unloading rate is 30 KN/min. The upper limit of the loading load is 60% of the uniaxial compressive failure load, and the lower limit of the unloading load is 100 N. A loading and unloading process is used as a cycle, and a total of 50 cycles are performed.

3. Experimental Results and Analysis

3.1. Appearance Analysis

In the early stage of dry–wet cycle, the appearance of RC specimens with three different rubber particle sizes did not change significantly. With the dry–wet cycle, the pit corrosion on the concrete surface increased significantly or small cracks appeared. The apparent morphology of RC specimens with different rubber particle sizes after 90 and 120 dry–wet cycles is shown in Figure 1.

As shown in Figure 1a,b, after 90 dry–wet cycles, there are many erosion holes on the surface of the concrete specimen, and the edges of the upper and lower ends are passivated. The cement mortar at the end of the RC specimen with a rubber particle size of 3–6 mm is slightly peeled off, and the damage is the most serious. After 120 dry–wet cycles, the concrete test block is seriously eroded, micro-cracks appear on the surface, the upper and lower ends are passivated, and even some test blocks appear to have fine aggregate shedding, exposing the internal coarse aggregate. This is due to the erosion of sulfate solution, and SO₄²⁻ establishes a three-dimensional transmission area at the end of the test piece to accelerate the erosion process. In the middle and late stages of the dry–wet cycle, the sulfate solution accelerates the deterioration rate of the concrete. With the continuous immersion and drying of water molecules in the RC, the internal pores increase, the cracks gradually penetrate, the internal bonding force and friction force of the concrete decrease, the structure gradually loosens, and the surface of the test block peels off significantly. Secondly, the rubber particles in RC do not participate in the cement hydration reaction, and there is a lack of strong chemical bonding force between rubber particles and other aggregates, resulting in a large gap and accelerating the erosion rate of the sulfate solution. With the increase in the particle size of the rubber particles, the gap around the rubber particles increases, and the range of crystal erosion damage caused by SO₄²⁻ to the concrete increases, which accelerates the invasion of SO₄²⁻. With the progress of the dry–wet cycle, the crystals accumulate in the pores of the concrete, forming a large crystal pressure, resulting in an increase in the pores and the penetration of the cracks, and the concrete surface forms more obvious cracks. The experimental phenomena show that with the increase in dry–wet cycles, the surface damage of RC specimens becomes more serious, and with the increase in rubber particles, the internal voids increase. At this time, the erosion damage of the sulfate solution to RC specimens becomes more serious, and the corrosion resistance of concrete specimens is worse. Among them, the damage of RC samples with a rubber particle size of 3–6 mm is the most serious, and the corrosion resistance of RC samples with a rubber particle size of 0.85 mm is the best.

3.2. Quality Loss Analysis

Mass loss is a macro damage index that can measure the concrete after dry–wet cycle to a certain extent. In the test of Gaowen Zhao et al. [37], the mass of concrete specimens increased first and then decreased with the increase in sulfate solution erosion time, and the higher the concentration of sulfate solution, the faster the mass growth and decline. Figure 2 plots the relative mass (K_m) of NC and RC at different cycles. With the increase in dry–wet cycles, the mass of samples increased slightly and then decreased obviously. At 60 dry–wet cycles, the mass of the RC specimen reached the peak, and then the mass of the specimen began to decrease gradually, and the mass after 120 dry–wet cycles was less than that before erosion. This is due to the early dry and wet cycle of SO₄²⁻ ions into the concrete, and cement hydration products to form ettringite and gypsum as well as other expansion, resulting in an increased sample quality. In the later stage of the dry–wet cycle, the high elasticity of rubber particles was unable to alleviate the damage caused by the expansion stress generated by the expansion, and the internal holes gradually penetrated into cracks, resulting in the spalling of cement mortar on the surface of the test block and a decrease in the quality of the test block. The calculation formula of the mass relative value is as follows:

$$Km=\frac{m_n}{m_0}\times 100\%$$

where K_m represents the relative value of mass (%); and m_n and m₀ represent the mass of the specimen at the nth and 0th dry–wet cycles (kg).

In the 60 dry–wet cycles, the relative value of the mass of ordinary concrete is less than that of rubber concrete, which is due to a large number of SO₄²⁻ ions entering the sample and reacting to generate expansion, and the mass is increased. The rubber particles in RC and the coarse and fine aggregates in RC optimize the gradation of concrete and slow down the rate of solution drying and immersion, and the mass loss is reduced compared with ordinary concrete. After 60 dry–wet cycles, the relative mass value of rubber concrete is significantly greater than that of ordinary concrete. This is because the high elasticity of rubber relieves the expansion stress generated by the expansion, and the crack generation rate is significantly slower than that of NC. The smaller the rubber particles, the greater the relative mass value of RC. The cement hydration products of concrete materials without sulfate attack are mainly composed of Ca(OH)₂ and Ca₄Al₂O₇·19H₂O. After 120 dry–wet cycles in sulfate solution, Ca(OH)₂ in cement hydrate gradually forms gypsum and ettringite under the chemical corrosion of sulfate solution and dry–wet cycles [38]. The damage of sulfate solution to concrete in the later stage is mainly due to the gradual accumulation of expansions caused by the main chemical reaction between Ca(OH)₂ and SO₄²⁻ ions in the solution.

$$\mathrm{C}\mathrm{a}(\mathrm{O}\mathrm{H}{)}_2+\mathrm{S}{\mathrm{O}}_4^{2-}+2{\mathrm{H}}_2\mathrm{O}\to \mathrm{C}\mathrm{a}\mathrm{S}{\mathrm{O}}_4\cdot 2{\mathrm{H}}_2\mathrm{O}+2\mathrm{O}{\mathrm{H}}^{-}$$

$$3{\mathrm{C}}_3\mathrm{A}+3\left(\mathrm{C}\mathrm{a}\mathrm{S}{\mathrm{O}}_4\cdot 2{\mathrm{H}}_2\mathrm{O}\right)+26{\mathrm{H}}_2\mathrm{O}\to 3\mathrm{C}\mathrm{a}\mathrm{O}\cdot \mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3\cdot 3\mathrm{C}\mathrm{a}\mathrm{S}{\mathrm{O}}_4\cdot 32{\mathrm{H}}_2\mathrm{O}$$

The K_m of samples with rubber particle sizes of 0.85 mm, 1–3 mm and 3–6 mm reached the maximum values, which were 0.66%, 1.24% and 2.06% different from that after 120 d dry–wet cycle. The larger the rubber particle size, the greater the change rate of K_m, the worse the impermeability and the higher the mass loss rate. It can be seen that the damage of 0.85 mm rubber mixed with concrete under the coupling of dry–wet cycle and sulfate attack is the smallest, and the mass loss is significantly reduced.

3.3. Relative Dynamic Elastic Modulus Analysis

The relative dynamic elastic modulus [39,40] can be used to reflect the internal damage of concrete specimens after dry–wet cycles in a sulfate environment and to evaluate the durability of concrete. Using an NN-4B non-metallic ultrasonic testing analyzer to measure the relative wave velocity of the sample, to calculate the relative dynamic elastic modulus of the specimen, the following formula [41] was used:

$$E_{\mathrm{dr}}=\frac{E_{\mathrm{dt}}}{E_{\mathrm{d}0}}\times 100\%=\frac{T_0^2}{T_{\mathrm{t}}^2}\times 100\%=\frac{V_{\mathrm{t}}^2}{V_0^2}\times 100\%$$

where E_dr represents the relative dynamic elastic modulus (%); E_dt and E_d0 represent the dynamic elastic modulus (MPa) of the specimens after different dry–wet cycles and before dry–wet cycles. T_t and T₀ represent the sound time (s) of the specimen after different dry–wet cycles and before dry–wet; V_t and V₀ represent the ultrasonic wave velocity (m/s) of the specimens after different dry–wet cycles and before dry–wet cycles.

The relationship between the number of dry–wet cycles and the relative dynamic elastic modulus of RC and NC specimens is shown in Figure 3. From the diagram, it can be seen that with the increase in the number of dry–wet cycles, the relative dynamic elastic modulus of concrete increases first and then decreases, and the relative dynamic elastic modulus of NC specimens is smaller than that of RC specimens. This is due to a large number of SO₄²⁻ ions entering the interior of the specimen to react to generate expansion, reducing the diffusion rate of sulfate, improving the compactness of concrete, and greatly reducing the speed of salt solution entering the void. The high elasticity of the rubber particles in the RC specimen alleviates the damage caused by the drying shrinkage force generated by the solution baking and immersion. In the 60 dry–wet cycles, the relative dynamic elastic modulus of RC specimens showed an increasing trend, and the relative dynamic elastic modulus of RC specimens with a rubber particle size of 1–3 mm was the highest. This is due to the rubber particles with a particle size of 3–6 mm. The interfacial transition zone between the rubber and the cement matrix in the RC specimen is obvious, the regional bonding force is weak, and it is easy to develop cracks. Therefore, the relative dynamic elastic modulus is lower than the test block with other rubber particle sizes.

The relative dynamic elastic modulus of concrete specimens reached the peak at 60 dry–wet cycles, and then decreased rapidly. The relative dynamic elastic modulus of RC and NC with rubber particle sizes of 0.85 mm, 1–3 mm and 3–6 mm decreased by 13.8%, 23.2%, 33.1% and 38.8%, respectively, at 120 dry–wet cycles. The four groups of concrete samples were damaged internally after dry–wet cycles. The internal damage to concrete mixed with 0.85 mm rubber was the smallest, and the internal damage to ordinary concrete was the largest. This is because after 60 dry–wet cycles, the expansion force generated by the expansion products inside the specimen is greater than the tensile strength of the internal structure, and the high elasticity of the rubber particles cannot alleviate the damage caused by the expansion force and dry–wet cycles. With the increase in the number of dry–wet cycles, the density of the test block decreases and the relative dynamic elastic modulus decreases. It can be seen that adding rubber can effectively improve the internal damage of concrete after the dry–wet cycle. The smaller the rubber particle size is, the smaller the damage is, and the 0.85 mm rubber particle size is the best.

3.4. Analysis of Compressive Strength Loss

In this paper, the compressive strength of the specimen was measured using an RDL electronic creep relaxation tester. The physical diagram and the damage morphology of the specimen are shown in Figure 4a,b.

The relative compressive strength K_f was used to measure the mechanical properties of concrete after dry–wet cycles of sulfate attack.

$$K_{\mathrm{f}}=\frac{{\mathrm{f}}_{\mathrm{cn}}}{{\mathrm{f}}_{\mathrm{c}0}}$$

where f_cn is the strength value of sulfate attack after dry–wet cycle, and f_c0 is the strength value of the same age under standard curing.

The relationship between dry–wet cycles and relative compressive strength of concrete specimens is shown in Figure 5. With the increase in dry–wet cycles, the compressive strength of RC and NC increased first and then decreased, which can be divided into the early growth stage and the later damage stage. The corrosion resistance coefficient of the compressive strength of the concrete specimens in the early stage of erosion is increasing. First, the concrete continues to undergo a cement hydration reaction in the sulfate solution and is still in the curing stage. At this time, the hydration reaction ability of the concrete is slightly greater than the sulfate erosion ability. Second, the SO₄²⁻ ions in the sulfate solution enter the concrete and react with the cement hydration products to form ettringite, gypsum and other expansion substances, which fill the pores inside the concrete, increase the compactness of the concrete and increase the compressive strength. Under the combined action of the two, the K_f value of concrete specimens increased in the early stage of erosion. With the further progress of erosion, the expansions such as ettringite generated by concrete hydration products and SO₄²⁻ ions increase, which promotes the rapid development of cracks. During the dry–wet cycle, the sulfate continuously eroded the weak interface connection between the coarse aggregate and the cement mortar, which had caused damage inside the concrete. Therefore, after 60 dry–wet cycles, the compressive strength of concrete samples continued to decline. This is consistent with its macroscopic quality loss and other changes.

In the process of the dry–wet cycle, the K_f value of rubber concrete is greater than that of ordinary concrete, and the smaller the rubber particle size, the stronger the sulfate resistance of the specimen and the better the durability. This is because the gypsum and ettringite formed by sulfate ions and cement hydration products fill the internal voids of concrete, and the air group at the interface between rubber particles in RC and cement mortar blocks the capillary pores, reduces the diffusion rate of sulfate ions and improves the compactness of concrete. The high elasticity of rubber particles alleviates the double damage caused by erosion and dry–wet cycle and alleviates the generation and development of cracks. The compactness of RC specimens is better than that of NC specimens, and the relative compressive strength is improved. The larger the particle size of rubber particles, the weaker the interfacial transition zone inside the RC specimen, the more prominent the defects, the easier the crack to develop and penetrate, the faster the erosion rate of sulfate solution, and the greater the compressive strength damage of the specimen. When the number of dry–wet cycles is 120, the K_f values of concrete with a rubber particle size of 0.85 mm, 1–3 mm and 3–6 mm is reduced by 11.2%, 25.1% and 26.3%, respectively, while the K_f value of ordinary concrete is reduced by 37.4%. The strength of concrete is greatly reduced, and the loss of compressive strength of RC with a particle size of 0.85 mm is the least. This shows that the rubber particles with an appropriate particle size enhance the corrosion resistance of concrete during the dry–wet cycle of sulfate solution. When the rubber particle size is large, the weak area of the RC interface is obvious, which makes the concrete specimen easy to damage.

3.5. Deterioration Analysis after Cyclic Loading

Constant amplitude cyclic loading–unloading tests were carried out on specimens subjected to different dry–wet cycles in sulfate solution, and the fatigue failure strength of the specimens was obtained. Compared with the initial uniaxial compressive strength of the specimens before cyclic loading–unloading, the degradation coefficient P was analyzed.

$$P=\frac{F_0-F_{\mathrm{i}}}{F_0}$$

P represents the deterioration coefficient; F₀ represents the uniaxial compressive strength of concrete before cyclic loading and unloading; F_i represents the fatigue failure strength of concrete after cyclic loading and unloading.

The relationship between the uniaxial compressive strength and fatigue failure strength of concrete specimens under different dry–wet cycles is shown in Figure 6. It can be seen from the figure that the uniaxial compressive strength of the four groups of concrete decreased under cyclic loading and unloading, and the deterioration coefficient increased continuously. With the increase in dry–wet cycles, the deterioration coefficient of RC and NC specimens increased, and the fatigue damage of concrete specimens caused by cyclic loading was more obvious. It can be concluded that the initial damage of the specimen is aggravated under the dual action of dry–wet cycle and sulfate attack, and the more the number of cycles, the greater the fatigue damage of the specimen after 50 cycles of loading and unloading.

After 60 dry–wet cycles, the deterioration degree of ordinary concrete is always greater than that of rubber concrete, and the RC with rubber particle size of 0.85 mm has the best fatigue resistance. After 120 dry–wet cycle, the deterioration coefficient of concrete with a rubber particle size of 0.85 mm, 1–3 mm and 3–6 mm is 10.8%, 14.2% and 15.5%, respectively, and the deterioration coefficient of ordinary concrete is 16.9%. Under the action of fatigue load, rubber particles, as a part of concrete, play a role of energy absorption and dissipation in weak parts such as initial cracks and crack development by compression and extension, inhibit the generation and further expansion of microcracks, and effectively alleviate the fatigue damage caused by the cyclic load to concrete [42,43]. The results show that rubber particles can optimize the fatigue performance of concrete to a certain extent. The smaller the rubber particle size, the smaller the deterioration coefficient and the smaller the fatigue damage.

3.6. Microscopic Analysis

Figure 7a–c show the micro-morphology of RC specimens with rubber particle sizes of 0.85 mm, 1–3 mm, and 3–6 mm after 120 dry–wet cycles in a sulfate solution environment. Figure 7d shows the micro-morphology of NC specimens after 120 dry–wet cycles. As shown in Figure 7a,b, there are a small number of cracks and local micro-pores in the sample, and there are a large number of clusters of ettringite and plate-column gypsum and other expansions to fill the gap cracks. The high elasticity of rubber particles alleviates the damage caused by dry–wet cycle and sulfate attack, and the interior is denser. As shown in Figure 7c, the larger holes and cracks, larger particle size rubber and cement matrix interface transition zone are weak, and cracks develop easily. As shown in Figure 7d, the cracks and voids in NC are more obvious, and the damage to the specimen is more serious. The expansion stress generated by cement hydration products (C-S-H) and expansion products such as ettringite and gypsum is far greater than the tensile strength inside NC. The cracks develop and penetrate, and the internal structure is further damaged, which can be reflected from the micro level to the macro level, that is, the compressive strength is significantly reduced.

Through microscopic mechanism analysis, it is found that the larger the rubber particle size, the more serious the concrete damage. In order to more intuitively reflect the internal damage of the samples at different cycles, the SEM images of the samples with a rubber particle size of 3–6 mm in different dry–wet cycle stages were selected. Figure 8a,b show SEM images of rubber particle size of 3–6 mm in 60 and 120 sulfate attack dry–wet cycles.

As shown in Figure 8a, after 60 dry–wet cycles, a large number of products such as ettringite generated by sodium sulfate erosion appear inside the sample, which are flocculent and plate-like distributed in microcracks, excessive weak areas of the interface, etc., filling internal voids and cracks, improving the compactness of the specimen, and effectively improving the corrosion resistance and durability of the specimen. Figure 8b shows the microstructure of RC after 120 dry–wet cycles. With the increase in dry–wet cycles, more expansions are produced. The high elasticity of rubber particles can no longer alleviate the expansion stress caused by expansions and the damage caused by dry–wet cycles. Due to the large particle size of rubber particles, the defects in the transition zone between rubber and cement are more prominent, which is easy to develop through cracks and accelerate the destruction of concrete samples.

4. Conclusions

Based on the comparative analysis of apparent morphology, mass loss, relative dynamic elastic modulus, compressive strength loss, deterioration degree after cyclic loading and internal microstructure of ordinary concrete (NC) and rubberized concrete (RC) under 120 sulfate attack and dry–wet cycle coupling, the main conclusions are as follows:

(1)

Adding appropriate rubber particles into concrete, the mechanical properties decreased compared with NC, but its corrosion resistance and durability were significantly improved. RC with a particle size of 0.85 mm was the best. After 120 dry–wet cycles, the relative compressive strength of RC with a particle size of 0.85 mm was 26.2% higher than that of ordinary concrete.

(2)

With the increase in dry–wet cycles, the surface damage and mass loss of the specimens were more serious, and the deterioration degree of the relative dynamic elastic modulus and compressive strength was significantly optimized compared with NC, indicating that the incorporation of rubber effectively improved the durability of concrete. The maximum mass relative value of the sample with a rubber particle size of 0.85 mm, 1–3 mm, and 3–6 mm is 0.66%, 1.24%, and 2.06% lower than that after 120 dry–wet cycles. The smaller the rubber particle size is, the better the impermeability is, and the smaller the mass loss rate is.

(3)

After cyclic loading, the deterioration degree of ordinary concrete is greater than that of rubber concrete. The high elasticity of rubber effectively alleviates the fatigue damage caused by cyclic loading and inhibits the generation of cracks. The smaller the rubber particle size, the smaller the deterioration coefficient and the smaller the fatigue damage. After cyclic loading, the deterioration degree of concrete with a 0.85 mm rubber particle size was 5.1% lower than that of ordinary concrete.

(4)

After 120 days of dry–wet cycle of sulfate attack, there are many pore cracks in the sample. The smaller the rubber particle size, the smaller the internal damage. In the early stage of erosion, the expansion of ettringite and gypsum filled the internal cracks, improved the compactness of the sample, and effectively increased the compressive strength of the sample. With the increase in erosion time, more expansion was produced, and the internal cracks of the sample gradually developed through the occurrence of damage.