Dry–Wet Cyclic Sulfate Attack Mechanism of High-Volume Fly Ash Self-Compacting Concrete

Abstract

High-volume fly ash replacing cement helps to improve the fluidity, volume stability, durability, and economy of self-compacting concrete (SCC). Sulfate attack is the most common form of the durability damage of hydraulic concrete; in particular, the performance degradation at the water level change position is more significant. Therefore, research on the influence effect and mechanism of fly ash on the durability is of great significance. In this paper, the change regularity of the SCC physical and mechanical properties with the fly ash replacement percentage and dry–wet cycles were studied by 60 dry–wet cycles of sulfate attack test. The 6 h electric flux, MIP, and SEM were used to study the performance degradation mechanism of SCC cured for 56 days, which had also been attacked by sulfate. The results show that the physical and mechanical properties of SCC increased first and then decreased with the dry–wet cycles of sulfate attack. After 10–15 cycles, the corresponding properties increased slightly, and then decreased gradually. When the fly ash content was 40%, the corrosion resistance coefficient, relative dynamic elastic modulus, and flexural strength retention were higher than those of the control specimen. However, when the fly ash content was 50%, they were close to the control and deteriorated obviously with the further addition of fly ash. For pore sizes in the range of 120–1000 nm, the porosity of SCC cured for 56 days was inversely proportional to the 6 h electric flux and the retention of mechanical properties, indicating that the porosity of the large pores is the decisive factor affecting the chloride ion permeability and corrosion resistance. The incorporation of fly ash in SCC can change the sulfate attack products and destruction mechanism. The sulfate attack damage of SCC with 40% of fly ash and the control specimen was dominated by ettringite crystallization and expansion, while those with a fly ash content of 50% and 60% had no obvious corrosion products, and the microstructures became looser. The appropriate fly ash replacement percentage could significantly improve the corrosion resistance of SCC.

1. Introduction

Self-compacting concrete (SCC) is favored by construction engineers for the excellent fluidity, homogeneity, and stability. Compared with ordinary concrete of the same strength, SCC has the characteristics of low water–binder ratio, high sand proportion, and large cement consumption [1,2,3], resulting in high hydration heat, which limits the engineering application. Cement is one of the most utilized construction materials due to its availability, but its production seriously affects environmental sustainability through the emission of high amounts of greenhouse gases such as CO₂ [4]. Replacing cement with a high-volume of fly ash can not only alleviate these problems, but also improve the environmental and economic effect of SCC [5,6,7].

As a high alkaline building structure material, SCC is highly vulnerable to sulfate attack in seawater, groundwater, saline soil, and polluted water, thus reducing the service life of building structures [8,9]. Particularly in acidic groundwater and high-clay soil environments, sulfates are contained in most cases, and the concrete itself may also contain sulfate. Therefore, under the above conditions, concrete will be eroded, causing concrete expansion, cracking, spalling, and other phenomena, causing the loss of strength and viscosity, leading to the destruction of the internal mechanism, which ultimately affects the durability of concrete [10].

Global scholars have systematically studied the microstructure and performance degradation laws, attack mechanisms, and improvement methods of concrete under sulfate attack [11,12]. The results showed that the mineral admixture improved the compactness and attack resistance of concrete through the active effect and microaggregate effect, among which the research and application of fly ash is the most extensive [13]. Fly ash helped to inhibit sulfate ions from entering concrete, and consumed the calcium hydroxide in the hydration products of concrete through secondary hydration, thereby reducing the production of ettringite during sulfate attack and further improving the corrosion resistance of concrete [14]. The results of Manthena et al. [15] illustrated that the strength and weight of SCC with 30% fly ash could not be reduced after curing in sulfate water for 90 days, and thus, had good sulfate resistance. Similar results were also obtained in the study of Ghafoori et al. [16]. The porosity of ordinary concrete or SCC and the generation and propagation of micro-pores and microcracks in a sulfate attack environment have an important impact on the deterioration of concrete performance [11].

The physical and mechanical properties of marine buildings and hydraulic concrete structures located in the area of tidal and groundwater level changing were more significantly degraded under the coupling effect of the dry–wet cycle and sulfate attack [17]. Repeated cycles of deliquescence–crystallization or hydration–dehydration cause the rapid decay of concrete compared to pure chemical attack from continuous immersion in a solution [18] Mineral admixtures such as fly ash and ground granulated blast slag could improve the resistance of concrete exposed to combined dry–wet cyclic sulfate attack, and long-term loading had a more serious influence on the penetration of chloride, regardless of being in a compressive region or tensile region [19]. Under the dry–wet cycle of sulfate attack, low-volume fly ash (10% and 20%) can effectively delay the deterioration of the mechanical properties of cellulose fiber concrete, whereas high-volume fly ash cannot improve the sulfate resistance. Macropores have a negative effect on the sulfate resistance durability of cellulose fiber concrete, while micropores and mesopores are favorable [14]. The dry–wet cycles significantly accelerated sulfate corrosion, and the concrete with fly ash performed better sulfate resistance than the concrete with ground granulated blast-furnace slag and silica fume [20,21]. In conclusion, the dry–wet cycle accelerated the performance degradation process of concrete under a sulfate environment, and the composition and replacement rate of mineral admixtures had complex effects on its corrosion resistance.

Although a lot of research has been conducted on the sulfate attack mechanism in fly ash-based concrete, and many scholars have focused on the attack mechanism of concrete under dry–wet cyclic sulfate attack and explored it experimentally, systematic studies on the sulfate attack mechanism of fly ash-based SCC, especially combined with dry–wet cycle, are seriously lacking. Based on this, the deterioration rule and mechanism of the physical and mechanical properties of high-volume fly ash SCC under dry–wet cyclic sulfate attack were studied in this article. Combined with the 6 h electric flux and mercury intrusion porosimetry (MIP) of SCC cured for 56 days as well as the scanning electron microscope (SEM) analysis before and after sulfate attack, the effect of fly ash on the mechanism of sulfate attack was illustrated. This study will provide the basic data for promoting the engineering application of SCC and high-volume fly ash SCC, and would be helpful for its safety evaluation under harsh environments.

2. Materials and Methods

2.1. Raw Materials

The cement used in the research was ordinary Portland cement with grade 42.5, the specific surface area was 382 m²/kg, and the water requirement for normal consistency was 28%. The properties and chemical composition were tested according to the Chinese standard GB/T17671-2021 ‘‘Test method of cement mortar strength (ISO)” [22], and the results are listed in

Table 1. The properties of ordinary Portland cement.

Setting Time (min)		Compressive Strength (MPa)		Flexural Strength (MPa)
Initial	Final	3 days	28 days	3 days	28 days
267	342	35.2	54.6	5.6	8.9

and

Table 2. The chemical composition of cement and fly ash.

Composition	SiO2	Al2O3	Fe2O3	CaO	MgO	SO3	K2O	Na2O	Loss
Cement (%)	17.43	5.23	3.54	60.87	3.15	3.43	1.04	0.35	4.96
Fly ash (%)	55.84	26.38	5.04	3.97	2.01	0.32	1.24	0.26	4.94

, respectively.

The fly ash had a specific surface area of 463 m²/kg, water content of 0.18 %, water demand ratio of 104.32 %, and the 28 day activity index was 79.34 %. The chemical composition of fly ash is shown in Table 2.

The slaked lime powder was industrial grade; its CaO content was no less than 70%.

The coarse aggregate was continuously graded crushed stone (5–19 mm) with a bulk density of 1.56 g/cm³ and apparent density of 2.61 g/cm³.

The fine aggregate was mixed sand and had a fineness modulus of 2.75, bulk density of 1.52 g/cm³, and apparent density of 2.65 g/cm³.

The experimental water was tap water.

The water reducer was 530 P polycarboxylate superplasticizer, which had a reducing rate of 30%.

2.2. Mixture Proportion

The absolute volume method was used to determine the mixture proportion (shown in

Table 3. The mixture proportion of SCC.

Code	Cementitious Materials (kg/m3)			W/B	Sand Proportion	Gravel (kg/m3)	Water Reducer (%)
	Cement	Fly Ash	Slaked Lime
SCC-0	479	-	-	0.37	0.50	883	0.4
SCC-40	287.5	175.0	16.5	0.37	0.50	852	0.4
SCC-50	237.5	208.0	33.5	0.37	0.50	844	0.4
SCC-60	191.6	241.6	45.8	0.37	0.50	836	0.4
SCC-70	141.5	275.0	62.5	0.37	0.50	828	0.4

) of SCC according to the “Technical specification for application of self-compacting concrete”(JGJ/T 283-2012) [23]. SCC-0 was partially replaced with fly ash and hydrated lime powder (referred as fly ash in this paper) in equal quantities. The fly ash replacement percent for cement was 40% (SCC-40), 50% (SCC-50), 60% (SCC-60), and 70% (SCC-70) by weight. The slaked lime powder was used to improve the alkalinity of the cementitious material system to ensure that the potential reactivity of fly ash was effectively stimulated. The mass composition of fly ash and slaked lime powder was determined by the mortar strength of cementitious materials.

2.3. Workability and Mechanical Properties

The workability and mechanical properties of SCC cured for 56 days were evaluated according to JGJ/T 283-2012 [23]. The test results are listed in

Table 4. The workability and mechanical properties of SCC.

Code	Slump Flow (mm)	T500 (s)	J-Ring Flow (mm)	Compressive Strength (MPa)	Flexural Strength (MPa)	Dynamic Elastic Modulus (GPa)
SCC-0	645	3.5	630	60.9	6.0	47.4
SCC-40	620	4.7	610	66.6	5.8	56.6
SCC-50	650	3.7	635	57.4	4.8	49.8
SCC-60	670	2.8	660	54.9	4.6	49.5
SCC-70	705	2.9	690	42.0	3.5	43.9

.

2.4. Experimental Methods

2.4.1. Dry–Wet Cyclic Sulfate Attack Test

The dry–wet cyclic sulfate attack test referred to the Chinese national GB/T50082-2009 standard for test methods of the long-term performance and durability of ordinary concrete [24]. Given that the content of fly ash was higher than 30%, the samples had to be cured for 56 days before the dry–wet cyclic sulfate attack test.

It took 24 h to complete a cycle. The experimental process was as follows:

(1)

The SCC samples were soaked in 5% Na₂SO₄ solution for 15 h, as shown in Figure 1a.

(2)

The samples were air-dried for 1 h, and then placed in an electric blast drying oven at 60 °C for 6 h, as shown in Figure 1b.

(3)

The samples were cooled for 2 h.

The dry–wet cyclic sulfate attack of SCC was carried out for 60 cycles in total. The mass and dynamic elastic modulus of samples were tested every five cycles. Compressive strength and flexural strength were tested every 15 cycles. Na₂SO₄ solution was replaced every 30 cycles.

2.4.2. Mass Loss Ratio

Specimens of SCC with the size of 100 mm × 100 mm × 400 mm were used and three replicates were made to verify the reproducibility of results in this experiment. The mass loss rate of the concrete specimen was calculated using Equation (1),

$$\mathrm{∆}{W}_n=\frac{W_0-W_n}{W_0}\times 100\%$$

(1)

where $\mathrm{∆}{W}_n$ is the mass loss ratio of the SCC specimens at the tested number of dry–wet cyclic sulfate attack, n, which is calculated as the average of the three specimens, %; $W_0$ is the mass of SCC specimens before corrosion, g; $W_n$ is the mass of SCC specimens after corrosion, g.

2.4.3. Relative Dynamic Elastic Modulus

The dynamic elastic modulus of SCC was tested according to the national standard GB/T50082-2009 [24]. In this study, specimens of SCC with the size of 100 mm × 100 mm × 400 mm were used to measure the dynamic elastic modulus at regular time intervals, and three samples were made as a group. The relative dynamic elastic modulus was calculated using Equation (2),

$$E_{rd}=\frac{E_{dn}}{E_{d0}}$$

(2)

where $E_{rd}$ is the relative dynamic elastic modulus; $E_{dn}$ is the dynamic elastic modulus of SCC specimens at the tested number of dry–wet cyclic sulfate attack, GPa; $E_{d0}$ is the dynamic elastic modulus of SCC specimens before corrosion, GPa.

2.4.4. Corrosion Resistance Coefficient and Flexural Strength Retention

The corrosion resistance coefficient is the ratio of the compressive strength of SCC with different dry–wet cycles of sulfate attack to the standardized specimens cured for 56 days, and the size of the samples was 100 mm × 100 mm × 100 mm. The flexural strength retention was the ratio of the flexural strength with the same calculation method as the corrosion resistance coefficient, and the size of the samples was 100 mm × 100 mm × 400 mm. Both were the average value of the test results of three specimens.

2.4.5. Chloride Ion Penetration Test

In order to clarify the relationship between the chloride ion permeability of SCC cured for 56 days and its corrosion resistance, the 6 h electric flux of SCC was tested according to the Chinese national standard GB/T50082-2009 [24]. The size of the samples was Φ100 mm × 50 mm. After sealing and vacuum water filling, the SCC samples were loaded into the battery box, the positive electrode was 0.3 mol/L NaOH solution, and the negative electrode was 3% NaCl solution, as shown in Figure 2. The DC voltage was 60 V and the power-on time was 6 h. The current was automatically collected every 5 min. The 6 h electric flux of SCC was obtained by integrating the current–time curve.

2.4.6. MIP Analysis

The samples used for MIP and the chloride ion penetration tests were prepared simultaneously. After standard curing for 56 days, hydration was then ceased in the samples by immersing in ethanol for 24 h. The samples were broken into pieces smaller than 1 cm³, and heated to 110 °C in a vacuum drying oven at 3 Pa (2.5 × 10⁻² Torr) for 4 h. The pore volume, pore size, and porosity of SCC were determined by an AutoPoreIV9500 automatic mercury porosimeter. The test pressure range was 0.10~61,000.00 psia, and the secondary contact angle was 130°.

2.4.7. SEM Analysis

The sample used for SEM measurement adopted the edge part of the sample after the compressive strength test, which was broken into regular pieces of 3–5 mm, and then hydration was ceased by immersing in ethanol for 24 h. The samples were dried to a constant weight in a vacuum drying oven at 60 °C, and then sprayed with gold. SCC specimens before and after sulfate attack were prepared in the same way. Then, the microscopic morphology of the hydration products and attack products of SCC were examined under SEM (Japanese JSM-7800F). The SEM was operated in high vacuum (<3 × 10⁻⁴ Pa) at a voltage of 3–20 kV with a spot size of 3.

3. Results and Discussion

3.1. Mass Loss Ratio

As can be seen from Figure 3, the mass of SCC decreased with the dry–wet cycles of sulfate attack at the initial stage, and increased with it gradually in the middle and later stages. The variation in the mass loss rate increased with the increase in the fly ash replacement percentage. The mass loss rate of SCC-40 was −0.35% and 0.85% at the 15th and 60th cyclic attack, while that of SCC-70 was −0.57% and 1.35% at the 10th and 60th cyclic attack, independently. During the 60 cycles of sulfate attack, the mass loss rate of SCC-40 was less than that of SCC-0, and the mass loss ratio of SCC-50 was close to that of SCC-0. Once the fly ash replacement percentage continued to increase, the mass loss rate of SCC-60 and SCC-70 was significantly higher than that of SCC-0. The results indicate that the appropriate replacement percentage of fly ash can improve the mass and volume stability of SCC effectively during the dry–wet cyclic sulfate attack.

3.2. Mechanical Properties

3.2.1. Corrosion Resistance Coefficient

The results in Figure 4 illustrate that the corrosion resistance coefficient of SCC increased first and then decreased with the dry–wet cycle sulfate attack. When there were 15 dry–wet cycles of sulfate attack, the corrosion resistance coefficient of SCC was greater than 1, then it decreased rapidly at 15–30 cycles and reduced slowly at 30–60 cycles. The corrosion resistance coefficient of SCC decreased with the increase in the fly ash content—SCC-40 was higher than that of SCC-0 during the whole dry–wet cycle sulfate attack test, SCC-50 was close to that of SCCC-0, and it was obviously lower than SCC-0 when the fly ash replacement percentage continued to increase. According to Table 4, the compressive strength of SCC-40 cured for 56 days was also higher than that of SCC-0. The results indicate that the mechanical properties and corrosion resistance of SCC-40 were both better than the control specimen.

3.2.2. Relative Dynamic Elastic Modulus

The dynamic elastic modulus describes the defects, pore characteristics, and properties of concrete with ultrasonic velocity [25,26]. The relative dynamic elastic modulus was used to characterize the effect of freeze–thaw cycles on the mechanical properties of concrete [27]. The damage mechanisms of dry–wet cyclic sulfate attack and the freeze–thaw cycles on the concrete hardened structures were different, but the damage process and results were quite similar. Therefore, the relative dynamic elastic modulus was used to describe the variation in the mechanical properties of SCC during dry–wet cyclic sulfate attack in this paper. According to Figure 5, SCC-40 had the highest relative dynamic elastic modulus during the whole sulfate attack process, which was 2% higher than that of SCC-0. Comparing Figure 4 with Figure 5, it can be seen that the variation tendency of the relative dynamic modulus with the fly ash content and the dry–wet cycles tended to be consistent with the corrosion resistance coefficient, and the former was slightly lower than the latter. This indicates that non-destructive testing can evaluate the change trend of the SCC mechanical properties in the process of dry–wet cyclic sulfate attack more conservatively.

3.2.3. Flexural Strength Retention

The flexural strength retention rate of SCC in Figure 6 increased significantly with the increase in the fly ash content in the early stage, and decreased in the intermediate and later stages. When the fly ash content was 70%, the flexural strength retention rate reached 1.23, which was 7% higher than SCC-0. SCC-40 was 0.97 and 0.90 at 30 and 60 cycles, respectively, which was 2% higher than that of SCC-0. At the initial stage of sulfate attack, the main corrosion product in concrete was acicular ettringite, which is beneficial to improve the compactness between the hydration products and optimize the flexural strength of concrete [28,29]. With the increase in the dry–wet cyclic sulfate attack, ettringite accumulates and expands continuously to produce microcracks. Bending load produces tensile stress at the bottom of the mid-span section of the specimen, and the continuous propagation and overlapping of microcracks leads to the instability and failure of the specimen. The research in [30] showed that the hydration products of tetracalcium aluminate (C₄AF) had good flexural tensile properties and corrosion resistance. The cement content in the SCC mixture decreased with the increase in the fly ash content (as shown in Table 3), and the relative volume of C₄AF decreased accordingly. Therefore, the flexural strength and its retention rate of SCC decreased with the increase in the fly ash content.

3.3. Dry–Wet Cyclic Sulfate Attack Mechanism

3.3.1. The 6 h Electric Flux of SCC Cured for 56 Days

The chloride ion permeability of SCC cured for 56 days can be negligible [31]. The results in Figure 7 indicate that the 6 h electric flux of high-volume fly ash SCC increased with the increase in the fly ash content, that of SCC-40 was 25% lower than SCC-0, and SCC-50 was close to SCC-0. The 6 h electric flux of SCC-70 reached 323 °C, which was 171% higher than that of SCC-0. Through comprehensive analysis of the test results in Figure 4, Figure 5 and Figure 6, it can be seen that the 6 h electric flux of SCC had a crucial effect on its resistance of dry–wet cycle sulfate attack. The mass loss rate of SCC was positively correlated with the electric flux, while the corrosion resistance coefficient, relative dynamic elastic modulus, and flexural strength retention rate were negatively correlated with the 6 h electric flux at 56 days.

3.3.2. Pore Structure of SCC Cured for 56 Days

As can be seen from Figure 8 and

Table 5. The pore structure parameter of SCC cured for 56 days.

Pore Structure Parameter		SCC-0	SCC-40	SCC-50	SCC-60	SCC-70
Cumulative porosity (v%)		17.65	20.06	23.30	23.76	26.28
Pore size distribution (v%)	<20 nm	9.11	10.45	10.05	11.18	9.41
	20–50 nm	2.97	3.10	4.18	3.34	4.28
	50–120 nm	3.17	3.93	4.31	4.17	6.38
	120–1000 nm	1.26	1.14	2.60	3.72	4.14
	>1000 nm	1.14	1.45	2.16	0.89	2.07

, the cumulative porosity of SCC increased with the increase in the fly ash content, as did the porosity of the pore size to above 20 nm. Because of the smaller fineness of fly ash and hydrated lime powder, the equal ball accumulation effect of fine particles forms small pores, which can effectively improve the fluidity of the mixture (Table 3). Although fly ash fills partial pores through secondary hydration in the process of setting and hardening, the incorporation of fly ash in cement still reduces the compactness of SCC in general. The pores below 50 nm accounted for more than 50% of the cumulative porosity, of which SCC-40 and SCC-50 accounted for 68% of the cumulative porosity, both higher than those of SCC-0. The results show that a high volume of fly ash cannot reduce the porosity of SCC, but can refine the capillary pores.

The findings of Mehta [32] proved that the impermeability of concrete mainly depends on the larger capillary pores, especially those with pore sizes exceeding 132 nm, while pores larger than 1000 nm had less correlation with the impermeability of concrete [32,33]. In light of the fact that the 132 nm aperture was not detected in MIP, 120 nm was taken as the reference in this paper. From the test data in Table 5 and Figure 6, it can be seen that the porosity of the capillary pores with a pore size between 120 and 1000 nm in SCC was linearly correlated with the 6 h electric flux. This indicates that the porosity of the larger capillary pores of SCC is the main factor that affects and even determines its chloride ion permeability and resistance to sulfate attack.

3.3.3. Microstructure and Morphology

From Figure 9a, it can be seen that the hydration products of SCC-0 were fibrous C–S–H gels, hexagonal calcium hydroxide (CH), and acicular ettringite crystals. As can be seen in Figure 9b, the micro morphology and hydration products of SCC-40 were close to SCC-0, the CH crystals were slightly reduced, and the fly ash beads and their surrounding hydration products were visible in the observation area. The C–S–H gel, a large number of hexagonal sheets CH, lotiform monosulfide calcium sulfoaluminate (AFm), and a few acicular ettringite crystals could be observed in Figure 9c. The SEM images of the SCC-60 section are also displayed in Figure 9d, where only the C–S–H gel and stacked CH crystals could be observed.

Considering that the physical and mechanical properties of SCC had been significantly degraded after 30 dry–wet cycles of sulfate attack, this study discussed the corrosion products through the SEM analysis of its section. It can be seen from Figure 10a,b that the main corrosion product of SCC-0 and SCC-40 was the radioactively distributed acicular ettringite crystalline beam. As shown in Figure 10c,d, the microstructure of SCC-60 and SCC-50 was loose and porous, pits were left after the flaking of fly ash beads caused by sulfate attack, and the C–S–H gel and other hydration products on the surface of the fly ash beads disappeared.

In the process of soaking in Na₂SO₄ solution, SO₄²⁻ migrated to the interior of the hardened concrete through capillary pores, reacted with CH in the pore solution to produce dihydrate gypsum, and then the dihydrate gypsum reacted with calcium aluminate hydrate (C–A–H) or AFm to produce ettringite. The generation and crystallization of dihydrate gypsum and ettringite both led to the volume expansion of the hardened structure [34,35]. In the initial stage of dry–wet cycle sulfate attack, the dihydrate gypsum and ettringite fill in the pores of SCC, improving the density, mass, and strength of SCC. In the intermediate and later stages, the crystallization and growth of the dihydrate gypsum and ettringite generate expansion stresses. This leads to the continuous growth of microcracks, which in turn causes the decrease in the quality and strength.

When the fly ash content was higher than 60%, the loss rate of the mechanical properties and mass slightly increased in the intermediate and later stages of dry–wet cyclic sulfate attack. This is due to the fact that the residual CH after secondary hydration increased with the fly ash content (Figure 9d). The residual CH can increase the local stress concentration caused by the crystallization and expansion of ettringite and dihydrate gypsum. At the same time, during the drying and air cooling of the dry–wet cycles, the concentration of CH in the pore solution increases, and then carbonation occurs to produce CaCO₃. During the subsequent process of sulfate attack, the C–S–H gel decomposes to produce non-cementitious thaumasite. The studies by [36,37] showed similar conclusions. In summary, the hydration products, pore structure, and impermeability of the hardened SCC affect and determine the resistance to dry–wet cycle sulfate attack. Incorporating the appropriate fly ash content in SCC can effectively improve its durability.

4. Conclusions

Experimental investigation was performed on the corrosion development of five groups of SCC under dry–wet cyclic sulfate attack. The degradation law and mechanism of the physical and mechanical properties of SCC with high-volume fly ash were studied. The participation of fly ash changed the performance degradation process and erosion products, and the appropriate replacement percentage of fly ash could effectively improve the corrosion resistance of SCC.

(1)

The mass of SCC increased at the initial stage of dry–wet cyclic sulfate attack and decreased in the intermediate and later stages. The mass loss ratio decreased with the increase in the fly ash content, that of SCC-40 and SCC-50 were lower than SCC-0, while SCC-60 and SCC-70 were higher than SCC-0. This indicates that the appropriate replacement percentage of fly ash can effectively improve the mass and volume stability of SCC during dry–wet cyclic sulfate attack.

(2)

The mechanical properties of SCC increased first and then decreased with the dry–wet cycles of sulfate attack and decreased with the increase in the fly ash content. After 60 dry–wet cycles of sulfate attack, the corrosion resistance coefficient, relative dynamic elastic modulus, and flexural strength retention rate of SCC-40 were higher than those of SCC-0 by 7%, 5%, and 2%, respectively, and the corrosion resistance of SCC-50 was also better than SCC-0. The relative dynamic elastic modulus was slightly lower than the corrosion resistance coefficient and the retention rate of flexural strength, indicating that nondestructive testing can evaluate the mechanical property of SCC more safely during the dry–wet cycle sulfate attack process.

(3)

The 6 h electric flux of SCC cured for 56 days can be negligible, and the lowest of SCC-40 was only 86 °C. The cumulative porosity of SCC increased with the fly ash content, and most of the pores were below 50 nm, which belong to less harmful pores. The porosity of the capillary pores with pore sizes between 120 and 1000 nm in SCC was linearly correlated with the 6 h electric flux and resistance to dry–wet cyclic sulfate attack. Therefore, when the content of fly ash was 40%, the mechanical properties and corrosion resistance of SCC were significantly improved.

(4)

The hydration products of SCC-0 and SCC-40 were fibrous C–S–H gels, CH, and acicular ettringite crystals. Those in SCC-50 were the C–S–H gel, CH, AFm, and a few ettringite crystals. Only the C–S–H gel and stacked CH crystals were observed on the section of SCC-60. The main attack product of SCC-0 and SCC-40 was the radioactively distributed acicular ettringite crystalline beam. Erosion pits were left after the flaking of fly ash beads caused by erosion. The erosion of SCC-0 and SCC-40 was dominated by ettringite expansion, while the microstructure of SCC-50 and SCC-60 became loose and porous.