The Influence of Salt Erosion on the Mechanical Performances of Ultra-High-Performance Concrete with Secondary Aluminum Dross

Abstract

Secondary aluminum dross containing a large amount of active substance can be used to prepare concrete. The mechanical strengths, the mass loss rate (MR) and the relative dynamic modulus of elasticity (RME) of ultra-high-performance concrete with secondary aluminum dross are researched. The NaCl freeze–thaw cycles (F-Cs) and dry–wet alternation (D-A) effects with NaCl and Na₂SO₄ are considered. The corresponding permeability of chloride ions and the carbonation depth (D_c) are obtained. The scanning electron microscope (SEM) photos are researched to reveal the variation of the mechanical mechanism. Results show that after specimens’ suffering from the action of 20 NaCl D-As, the MR of ultra-high-performance concrete is the highest. Specimens exposed to 200 NaCl F-Cs show the lowest MR and CMC. The RME of UHPC under salt actions increase in the order of 20 NaCl D-As < 20 Na₂SO₄ D-As < 200 NaCl F-Cs. After suffering 200 NaCl F-Cs, 20 Na₂SO₄ D-As and 20 NaCl D-As, the corresponding D_c values are 1.86 mm to 2.31 mm, 1.79 mm to 2.23 mm and 2.11 mm to 2.76 mm. The flexural strength decreases at the rates of 0.99%–25%, 3.92%–27.84% and 1.47%–21.59% respectively. The MR increases and the RME decreases as the cubic function changes with the amount of salt erosion. After the secondary aluminum dross is added, the CMC decreases at the rates of 0% to 11.53%, 0% to 33.17% and 0% to 8.41% during the process of the salt action. The SAD can reduce the D_c with the decreasing rates of 19.48%, 23.55% and 19.73%. The SAD can increase the compactness of ultra-high-performance concrete. Ultra-high-performance concrete suffering from 20 NaCl D-As shows the largest number and the highest width of cracks. However, when the specimens are exposed to 20 Na₂SO₄ D-As, the number of cracks is the lowest and the width is the narrowest.

1. Introduction

The generation and accumulation of solid waste in today’s society are relatively complex environmental problems that trouble human beings [1]. Considering that the landfill deposits of this material will cause serious pollution to the environment, the solidification technology of solid waste appears to be very necessary [2,3]. The solidification technology of solid waste includes their use in concrete and the solidification effect of CO₂ applied for the disposal of the solid waste [4,5].

Secondary aluminum dross (SAD) is a by-product of aluminum production, causing environmental pollution if its treatment is not reasonable. The SAD possesses a large content of active oxide, which is beneficial to cement hydration. The use of SAD in cement-based materials can solidify the SAD to achieve the purpose of reducing its environmental pollution. Several studies have centered on the usage of SAD in cement concrete [6,7,8,9]. Satish found that SAD can increase the flexural and compressive strengths by 28.6% and 21.5%, respectively [10]. Moreover, SAD can decrease the chloride ion migration coefficient by 13.2% to 33.1% [11,12,13,14]. However, the concrete structure is usually exposed to various erosive environments. In a coastal environment, the erosion of NaCl and Na₂SO₄ accelerates the deterioration of concrete performance. Therefore, cement concrete materials in coastal environments require a high durability. The ultra-high-performance concrete (UHPC) showing excellent strength and durability can be used in salt erosion environments.

The UHPC is composed of continuous graded quartz sand, cement and a large amount of mineral admixtures [15,16]. This material is manufactured through the maximum density theory. Compared with other cement concrete, UHPC has the least pores and excellent mechanical and durable properties. Several researchers have found the positive effect of solid waste ash in UHPC’s strength and durability. The solid waste ash and stove ash can increase the corresponding flexural strengths by the maximum rates of 23.4% and 17.1%, while the compressive strengths are increased by the maximum rates of 14.1% and 16.7% [17,18,19]. Additionally, the corrosion resistance of a reinforced UHPC matrix subjected to NaCl action is also improved by adding SAD [20,21,22]. In coastal environments, the mechanical performance can deteriorate when subjected to salt erosion. However, the SAD–UHPC’s mechanical performance under the environment of salt erosion has not been investigated.

In this study, the flexural and compressive strengths (f_t and f_cu) of SAD–UHPC are measured. The NaCl F-C, NaCl D-A and Na₂SO₄ D-A actions on the specimens’ mass loss rates (MRs), relative dynamic moduli of elasticity (RDMEs), carbonation depths and chloride ion permeabilities are studied. The scanning electron microscope (SEM) photos, energy dispersive spectrometer (EDS), thermogravimetric analysis (TG), X-ray diffraction spectrum (XRD) and mercury intrusion porosimetry (MIP) are utilized for the analysis of the mechanical mechanisms after NaCl and Na₂SO₄ action. This paper will offer theoretical supports for the application of SAD–UHPC in salt erosion environments.

2. Experimental Section

2.1. Raw Materials

Ordinary Portland cement (OPC) with the initial and final setting times of 103.1 min and 413.2 min purchased from the Ningguo cement plant of Anhui Conch Cement Co., Ltd., Ningguo city, China is used. The density of the OPC is 3.03 g/cm³. Zhejiang Yongji Metal Materials Technology Co., Ltd., Jinhua, China offers the SAD for this research. The density of the SAD is 1.21 g/cm³. The contents of AlN, Al₄C₃, AlP, Al₂S₃ and Al₅O₆N of the SAD are 0.03%, 0.01%, 0.02%, 0.025% and 0.013% of the total mass of the SAD. Other chemical compositions are shown in

Table 1. Chemical composition (%).

Types	SiO2	Al2O3	FexOy	MgO	CaO	SO3	K2O	Na2O	Ti2O	Loss on Ignition
OPC	20.3	5.8	3.9	1.9	62.2	2.8	-	-	-	3.1
FA	90.40	0.33	0.62	0.35	0.42	0.21	7.49	-	-	-
QS	98.7	-	1.3	-	-	-	-	-	-	-
SAD	4.7	79.2	4.3	5.2	1.6	-	-	0.9	-	3.8

. Taiyue Mining (Shandong) Co., Ltd., Jinan, China, offers the ultra-fine fly ash (FA) used as the mineral admixture. The FA has a density of 2.45 g/cm³ and particle diameters of 0.005 mm to 0.05 mm. Henan Tengze Environmental Technology Co., Ltd., Zhengzhou, China produces the quartz sand (QS) as aggregate for the UHPC. Efficient polycarboxylic acid water-reducing agent bought from Jilin Zhongxin New Building Materials Co., Ltd., Changchun, China, is used for the adjustment of the fresh UHPC’s fluidity. The aggregates’ particle sizes range from 3.21 to 1.57 mm, 0.83 to 0.33 mm, and 0.31 to 0.21 mm, respectively, with the corresponding mass ratio of 1:1:5:1 applied. Table 1 and

Table 2. The accumulated pass rate (%).

Types	Particle Size/μm
	0.3	0.6	1	4	8	64	360
OPC	0.13	0.36	3.1	14.8	29.2	92.8	100
FA	32.11	59.12	83.16	100	100	100	100
QS	0	0	0	0	0.035	24	100
SAD	0.04	0.31	0.55	1.11	3.95	25.4	88.13

show the accumulated pass ratios and the chemical compositions.

2.2. The Preparing of Specimens

The binder materials (OPC, BFS, SF and SAD) are mixed uniformly in the NJ-160A mixer (Shanghai Meiyu Instrument Technology Co., Ltd., Shanghai, China) with a stirring speed of 60 rpm. After this mixing, the water and water-reducing agent are added to the mixer, stirring at the speed of 80 rpm for another 3 min. Then, the specimens with sizes of 40 × 40 × 160 mm³, Φ100 × 50 mm³ and 100 × 100 × 100 mm³ are manufactured. After curing in the environment (relative humidity of 40% and temperature of 20 ± 2 °C) for 1 day, the specimens are moved to an environment of 20.3 °C and 96.2% relative humidity for 23 days. The mixing proportions are shown in

Table 3. The mixing proportions of UHPC (kg/m³).

Samples	Water	OPC	SAD	FA	QS	WR
SAD-0%	241.2	843.3	0	365.5	965.4	16.1
SAD-25%	241.2	843.3	91.4	274.1	965.4	16.1
SAD-50%	241.2	843.3	182.8	182.8	965.4	16.1
SAD-75%	241.2	843.3	274.1	91.4	965.4	16.1
SAD-100%	241.2	843.3	365.5	0	965.4	16.1

, which are used for manufacturing UHPC specimens. In this paper, six samples are prepared for each experiment. The average of the measured values is used as the testing result. The coefficient of variation of all experimental test data in this article is less than 0.1, ensuring the accuracy of the study.

2.3. The NaCl F-C Action

A 3% NaCl solution is used for immersing specimens for 4 days. After the immersion, specimens are moved to the TDS-300 fully automatic rapid freeze–thaw test box provided by Nanjing Annai Testing Equipment Co., Ltd., Nanjing, China, for the experiment. Specimens are moved to the sealed stainless-steel freeze–thaw boxes filled with the NaCl solution at the same concentration. The temperature in the F-C box ranges from −15 °C to 8 °C. The freezing time of each F-C is 2–3 h, while the thawing time of each F-C is 1–2 h. All parameters are determined for each of the 50 NaCl F-Cs. Figure 1 shows the process of the NaCl F-C.

2.4. The NaCl and Na₂SO₄ D-A Actions

The immersion process of specimens in NaCl is the same as the NaCl F-C action. After the immersion, the specimen are moved to the concrete corrosion-resistant dry–wet cycle testing machine bought from Tianjin Huida Experimental Instrument Co., Ltd., Tianjin, China for the experiment of NaCl D-A. During each D-A, 8 hours of solution immersion in NaCl or Na₂SO₄ are applied to the specimens; after that, the surfaces of the specimens are dried using a rag. Then, the specimens are dried at a temperature of 80 °C for 36 h. Ultimately, the samples are cooled for 2 h. Figure 2 shows the process of Na₂SO₄ D-A.

2.5. The Measurement of MR

Specimens’ masses are determined after the specimens are immersed in NaCl for 4 days. After that, the specimens’ surfaces are wiped. Then, the masses of the specimens are determined using the DH-8905 electronic scale manufactured by Dongguan Dingheng Electronics Co., Ltd., Dongguan, China, with the measurement range of 0–3 kg and the minimum accuracy of 0.1 g. The MR can be calculated with Equation (1),

$$MR=\frac{m_t-m_1}{m_1}$$

(1)

where m₁ is the mass of the specimen after the 4-day immersion in 3% NaCl solution and m_t is the mass of the specimen after 50 F-Cs or 10 D-As.

2.6. The Measurement of the Relative Dynamic Modulus of Elasticity

A HS600 digital ultrasonic flaw detector manufactured by Shanxi Gangyan Testing Technology Co., Ltd., Taiyuan, China, is used for the determination of the relative dynamic modulus of elasticity (RDEM). The probes of the receiving and transmitting ends of the ultrasound are tightly adhered to the center positions of the specimens. The baseline is used for the match between the probes and the specimens. The RDEM is calculated by Equation (2),

$$RDEM=(\frac{v_t}{v_1}{)}^2$$

(2)

where v₁ and v_t are the ultrasonic velocities of the specimens after 4 days of NaCl or Na₂SO₄ immersion and each 50 NaCl F-Cs, 10 NaCl D-As or 10 Na₂SO₄ D-As. Specimens with sizes of 100 × 100 × 100 mm³ are used for the measurements of MR and RDEM. The flowchart of the experimental process is shown in Figure 3.

2.7. The Measurement of Mechanical Strengths

The flexural strength (f_t) and compressive strength (f_cu) are determined with the fully automatic bending and integrated testing machine (Shanghai Qunhong Instrument Equipment Co., Ltd., Shanghai, China) with the loading rates of 0.05 kN/s and 2.4 kN/s, respectively. The mechanical strength measurements are carried out with specimens of the size 40 × 40 × 160 mm³.

2.8. The Chloride Ion Migration Coefficient

The CMC of the UHPC is measured with the RCM-6 concrete chloride ion diffusion coefficient tester manufactured by Shanghai Shengshi Huike Testing Equipment Co., Ltd., Shanghai, China. All specimens are firstly processed with the vacuum saturation treatment and then moved to the measurement of the CMC. The vacuum saturation machine (Shanghai Shengshi Huike Testing Equipment Co., Ltd., Shanghai, China) with a vacuum volume of 18 L and accuracy of 0.1% is used for the vacuum saturation. A concrete chloride ion diffusion coefficient tester (Cangzhou Taiding Hengye Testing Instrument Co., Ltd., Cangzhou, China) with the measuring voltages of 10 V to 60 V and a voltage accuracy of 0.1 V is used for measuring the CMC.

2.9. The Carbonation Depth

The carbonation depth (D_c) of UHPC is determined after treatment in the TH-2 concrete carbonization test box (Hebei Zhongyi Weichuang Test Instrument Co., Ltd., Cangzhou, China). The detailed testing process can be described as follows:

The UHPC specimens are placed into the carbonization box. The concentration of the CO₂ in the carbonization box is 20 ± 3%. The temperature and the humidity in the carbonization box are 20 ± 5 °C and 70 ± 5%. After the specimens are cured in the carbonization box for 28 days, the specimens are taken out and broken to determine their carbonization depth. The specimens are cut off by half of their width. The remaining powder on the section of the cut specimen is scraped off, and then a 1% concentration of phenolphthalein alcohol solution (containing 20% evaporated water) is sprayed. After 30 s, the carbonization depth of each point on both sides at every 10 mm marked previously is measured in a carbonization depth test box. In this paper, the variation coefficients of all macroscopic performance experimental results are lower than 0.1.

2.10. The SEM Experiment

The samples’ inner UHPC with the diameter of 0.5 mm to 1 mm is used for the experiment of SEM. All samples are dried in the vacuum oven at 105 °C for 2 days. The samples are coated with gold until dry. Finally, the samples coated with gold are moved to the SEM to obtain the photos magnified 10,000 times.

3. Results and Discussions

3.1. The MR of UHPC

In this paper, the N of all fitting equations presents the amount of salt action. a, b, c and d represent the MR of specimens with the increasing numbers of NaCl F-Cs, NaCl D-As and Na₂SO₄. Figure 4 displays the MR of the UHPC exposed to erosion action. The MR of specimens increases with the increasing number of NaCl F-Cs, NaCl D-As and Na₂SO₄ D-As. The NaCl F-Cs can increase the frost-heaving stress and crystallization stress inside the UHPC [23,24], thus, accelerating the surface peeling on UHPC. Therefore, the MR increases with the NaCl F-C action. The NaCl D-A can increase the crystalline stress of the UHPC; hence, the surface peeling on the UHPC is exacerbated, resulting in the increased MR of UHPC [25]. Additionally, the chloride accelerates the alkali aggregate reaction of UHPC, which increases the mass loss of the UHPC. Meanwhile, the dry–wet cycling of sulfate acts on the UHPC, the internal cementitious material undergoes a chemical reaction with sulfate and decomposes to form an erosion product, ettringite, which appears inside the UHPC. While the quantity of SAD increases, the MR of UHPC decreases. This can be ascribed to the reason that the SAD can form an aluminum hydroxide gel with cement, thus making UHPC more compact; therefore, the resistance to salt erosion is enhanced [26,27,28,29]. The MR of UHPC is the highest after suffering NaCl D-As, while the UHPC after suffering the NaCl F-Cs shows the lowest MR. The relationships between the number of NaCl F-C, NaCl D-A and Na₂SO₄ actions and the MR fit with the function of Equation (3).

Table 4. The fitting equations of the relationship between the amount of erosion action and the MR.

Equation Types	SAD Content	a	b	c	d	R2
NaCl F-C	0%	6.33 × 10−7	−1.96 × 10−4	2.17 × 10−2	1.42857 × 10−4	0.99
	25%	5.58 × 10−7	−1.75 × 10−4	1.75 × 10−2	−1 × 10−3	0.99
	50%	5.12 × 10−7	−1.51 × 10−4	1.42 × 10−2	−1.99 × 10−16	1.00
	75%	4.44 × 10−7	−1.28 × 10−4	1.27 × 10−2	−5.96 × 10−16	1.00
	100%	3.33 × 10−7	−9.51 × 10−5	9.5 × 10−3	2.29 × 10−3	0.98
NaCl D-A	0%	4.40 × 10−4	−1.54 × 10−2	2.38 × 10−1	−8.57 × 10−3	0.99
	25%	2.60 × 10−4	−1.04 × 10−2	1.97 × 10−1	−9.57 × 10−3	0.98
	50%	2.13 × 10−4	−9.03 × 10−3	1.77 × 10−1	−7.43 × 10−3	0.99
	75%	1.87 × 10−4	−7.71 × 10−3	1.58 × 10−1	−7.71 × 10−3	0.99
	100%	6.67 × 10−5	−4.11 × 10−3	1.27 × 10−1	−9.71 × 10−3	0.98
Na2SO4 D-A	0%	3.67 × 10−4	−1.27 × 10−3	0.21	−1.01 × 10−2	0.99
	25%	1.67 × 10−4	−7.51 × 10−3	0.17	−6.71 × 10−3	0.99
	50%	6.00 × 10−5	−4.06 × 10−3	0.13	−3.86 × 10−3	0.99
	75%	4.67 × 10−5	−3.46 × 10−3	0.12	−1.86 × 10−3	0.99
	100%	6.67 × 10−5	−3.40 × 10−3	0.10	−6 × 10−3	0.98

shows the fitting equations of the relationships between the amount of erosion action and the MR. The R-squared is higher than 0.98, which ensures the rationality of fitting equations. After 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the MR of UHPC are 0.76% to 1.47%, 1.24% to 1.97% and 1.42% to 2.14%, while the MR are decreased at the decreasing rates of 0% to 48.2%, 0% to 33.6% and 0% to 37.1% by adding the SAD.

$$MR=a{N}^3+b{N}^2+cN+d$$

(3)

3.2. The RDME of UHPC

Figure 5 shows the RDME of UHPC after suffering different numbers of NaCl F-C, NaCl D-A and Na₂SO₄ D-A actions. The RME of UHPC decreases with increased NaCl F-C, NaCl D-A and Na₂SO₄ D-A numbers. The frost heaving-stress stemming from NaCl F-Cs and the crystalline stress generated by NaCl F-Cs and NaCl D-As can facilitate the development and number of cracks inside the UHPC. The cracks obstruct the propagation of ultrasounds, resulting in a decrease in the RDEM [30,31]. When the UHPC undergoes the Na₂SO₄ erosion, the internal UHPC is exposed to a chemical reaction with sulfate and decomposes to form an erosion product, ettringite, leading to more cracks in the specimens. The erosion product expands, and the voids are irregularly distributed in the mortar. The presence of these voids will gradually deteriorate the RDEM of the mortar. The amount of salt erosion shows the cubic function with the RME, which is shown in Equation (4).

Table 5. The fitting equations of the relationships between the amount of erosion action and the RDEM.

Equation Types	SAD Content	a	b	c	d	R2
NaCl F-C	0%	1.00 × 10−6	−1.14 × 10−4	−6.86 × 10−2	99.98	0.99
	25%	1.93 × 10−6	−5.49 × 10−4	−1.21 × 10−2	99.97	0.99
	50%	8.67 × 10−7	−3.03 × 10−4	−1.26 × 10−2	99.98	0.99
	75%	−8.67 × 10−7	1.34 × 10−4	−2.47 × 10−2	100.02	0.99
	100%	−2 × 10−7	−1.43 × 10−5	−1.26 × 10−2	100.02	0.99
NaCl D-A	0%	−4.67 × 10−4	−2.69 × 10−2	−1.04	99.98	0.99
	25%	1.73 × 10−3	−4.43 × 10−2	−0.33	99.97	0.99
	50%	6.00 × 10−4	−1.77 × 10−2	−0.33	100.02	0.99
	75%	−1.13 × 10−3	2.49 × 10−3	−0.41	100.03	0.99
	100%	1.33 × 10−4	−8.29 × 10−3	−0.15	100.01	0.99
Na2SO4 D-A	0%	1.20 × 10−3	−0.02	−0.61	99.94	0.99
	25%	2 × 10−3	−0.06	−0.09	99.94	0.99
	50%	1 × 10−3	−0.03	−0.14	100.02	0.99
	75%	−6 × 10−4	0.01	−0.24	100.04	0.98
	100%	2 × 10−4	−0.01	−0.06	100.01	0.99

shows the fitting equations of the relationships between the amount of erosion action and the RDEM. The R-squared of the fitting curves are higher than 0.98. However, the RME is increased by adding SAD. Due to the improved compactness and decreased inner cracks, the ultrasound’s speed is increased [32,33]. Therefore, the addition of SAD increases the RME. Meanwhile, the RME of UHPC is the highest after 200 NaCl F-Cs. The UHPC shows the lowest RDEM after suffering 20 NaCl D-As, due to the most serious erosion action of NaCl D-As on the UHPC. After 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the RMEs of UHPC are 89.7% to 95.3%, 89.4% to 95.7% and 86.3% to 94.8%. Additionally, the SAD can increase the RDEM with the increasing rates of 5.88%, 8.97% and 6.58%.

$$RDEM=a{N}^3+b{N}^2+cN+d$$

(4)

3.3. The Mechanical Strengths of UHPC

The f_t of UHPC after suffering from different types of salt erosion are shown in Figure 6. The f_t of UHPC decreases with the number of NaCl F-Cs, NaCl D-As and Na₂SO₄ D-As, due to the reason that the salt erosion can increase the number and widen the size of cracks in UHPC, decreasing the f_t. As found in Figure 6, the UHPC after NaCl D-A action shows the lowest f_t. However, when the mixed content of SAD is increased, the f_t of the UHPC is increased. The SAD can increase the amount of aluminum hydroxide, which makes the hydration products more compact [34,35]. Hence, the f_t is improved by adding SAD. The f_t of UHPC after NaCl F-C action is the highest, while the f_t of UHPC is the lowest after suffering from NaCl D-A erosion. After 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the f_t of UHPC decreases with the decreasing rates of 0.99%–25%, 3.92%–27.84% and 1.47%–21.59%. The SAD can increase the f_t by 0%–23.4%, 0%–25.1% and 0%–13.4%.

Figure 7 shows the f_cu of UHPC after suffering from different types of salt erosion. It can seen in Figure 7 that similar varying levels of f_cu and f_t occur. The f_cu shows an increasing trend with the increasing quantity of SAD and a decreasing trend with the number of NaCl F-Cs, NaCl D-As and Na₂SO₄ D-As. The reasons for these results are the same with the f_t’s results [36,37]. The relationship between the number of NaCl F-Cs, NaCl D-As and Na₂SO₄ D-As and the f_cu coincides with the cubic function. The UHPC demonstrates the highest compressive strength when exposed to the NaCl F-C environment. However, the f_cu of UHPC is the lowest after suffering 20 NaCl D-As. After 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the f_cu of UHPC decreases with the decreasing rates of 1.59%–10.31%, 3.09%–11.45% and 1.63%–9.62%. After the SAD is added, the f_cu of UHPC increases to 4.13%–21.16%, 5.41%–22.76% and 5.53%–21.41%.

3.4. The CMC of UHPC

The CMC curves of the UHPC after suffering from different NaCl F-Cs, Na₂SO₄ and NaCl D-As are shown in Figure 8. The CMC of UHPC increases with the increased numbers of NaCl F-Cs and Na₂SO₄ and NaCl D-As, which is attributed to the cracking number and width [38,39]. The chloride ions migrate along the cracks, thereby increasing the CMC. The increasing quantity of SAD decreases the CMC of the UHPC, due to the improved compactness with the addition of SAD, which prevents the chloride ions from migrating along the cracks, thereby reducing the CMC. The relationships between the amount of salt erosion activity and the CMC fit the model of the cubic function shown in Equation (5).

Table 6. The fitting equations of the relationship between the amount of erosion action and the CMC.

Equation Types	SAD Content	a	b	c	d	R2
NaCl F-C	0%	1.4 × 10−7	−3.49 × 10−5	1.21 × 10−2	1.99	0.96
	25%	2.67 × 10−8	3.06 × 10−5	3.12 × 10−3	1.94	0.98
	50%	2.53 × 10−7	−2.63 × 10−5	5.62 × 10−3	1.84	0.98
	75%	3.73 × 10−7	−5.8 × 10−5	7.17 × 10−3	1.70	0.97
	100%	3.2 × 10−7	−4.34 × 10−5	6.09 × 10−3	1.62	0.94
NaCl D-A	0%	1.63 × 10−3	−0.06	0.80	2.06	0.94
	25%	1.55 × 10−3	−0.06	0.74	1.99	0.97
	50%	1.03 × 10−3	−0.04	0.58	1.89	0.93
	75%	1.29 × 10−3	−0.05	0.61	1.75	0.95
	100%	8.73 × 10−4	−0.03	0.35	1.66	0.97
Na2SO4 D-A	0%	−6.00 × 10−5	2.6 × 10−3	0.07	−1.98	0.91
	25%	−6.67 × 10−5	6.11 × 10−3	−2.62 × 10−3	1.95	0.98
	50%	2.27 × 10−4	−1.63 × 10−3	3.59 × 10−2	1.84	0.98
	75%	3.87 × 10−4	−5.51 × 10−3	0.05	1.71	0.98
	100%	3.8 × 10−4	−4.46 × 10−3	0.03	1.63	0.99

shows the fitting equations of the relationships between the amount of erosion activity and the CMC. The R-squared of the fitting equations is higher than 0.91. After suffering 200 NaCl F-Cs and 20 Na₂SO₄ and 20 NaCl D-As, the CMCs are 3.68 × 10⁻¹² m²/s to 4.16 × 10⁻¹² m²/s, 3.59 × 10⁻¹² m²/s to 3.92 × 10⁻¹² m²/s and 4.03 × 10⁻¹² m²/s to 6.03 × 10⁻¹² m²/s. As established by the results, the CMC is the highest after treatment with 20 NaCl D-As. However, the CMC is the lowest after being exposed to 20 Na₂SO₄ D-As. The NaCl D-As can cause UHPC cracking the most seriously, simultaneously increasing the concentration of chloride ions in the UHPC pore solution. Therefore, the CMC of UHPC is the highest after the action of 20 NaCl D-As. However, the Na₂SO₄ D-A action induces the lowest number and width of cracks and cannot introduce chloride ions. Hence, the CMC of UHPC with 20 Na₂SO₄ D-As is the lowest. After the salt action, the CMC are decreased, with rates of 0% to 11.53%, 0% to 33.17% and 0% to 8.41% by adding SAD. As observed in Section 3.1, Section 3.2, Section 3.3 and Section 3.4, when the addition of MS is 100% (30% of the total binder materials), the resistance of the NaCl and Na₂SO₄ dry–wet alternations are the highest, which is ascribed to the pozzolanic effect.

$$CMC=a{N}^3+b{N}^2+cN+d$$

(5)

3.5. The Carbonation Depth of UHPC

Figure 9 exhibits the carbonation depth (D_c) of UHPC. The D_c shows the increasing trend with more occurrences of NaCl F-Cs and Na₂SO₄ and NaCl D-As, due to the increased cracks by the salt action [40]. The CO₂ can infiltrate into the interior of the UHPC along the crack, leading to an accelerated carbonation reaction, thereby increasing the depth of carbonation. The relationships between the amount of salt erosion action and D_c fit the model of the cubic function shown in Equation (6).

Table 7. The fitting equations of the relationships between the amount of erosion action and the D_c.

Equation Types	SAD Content	a	b	c	d	R2
NaCl F-C	0%	1.8 × 10−7	−5.89 × 10−5	9.42 × 10−3	1.34	0.99
	25%	8 × 10−8	−2.86 × 10−5	7.31 × 10−3	1.21	0.99
	50%	−9.33 × 10−8	2.49 × 10−5	3.16 × 10−3	1.15	0.95
	75%	−6.67 × 10−9	8.57 × 10−7	4.45 × 10−3	1.07	0.97
	100%	−8.67 × 10−8	2.91 × 10−5	1.89 × 10−3	1.00	0.98
NaCl D-A	0%	1.2 × 10−4	−5.94 × 10−3	0.14	1.35	0.98
	25%	−6.67 × 10−5	4 × 10−4	0.08	1.24	0.96
	50%	−1.8 × 10−4	4.77 × 10−3	0.04	1.18	0.91
	75%	−1 × 10−4	−2 × 10−3	0.06	1.10	0.93
	100%	−1.87 × 10−4	4.89 × 10−3	0.03	1.02	0.95
Na2SO4 D-A	0%	6.67 × 10−6	−2.86 × 10−5	0.04	1.33	0.96
	25%	−2 × 10−5	2.86 × 10−5	0.05	1.20	0.97
	50%	−6 × 10−5	2.11 × 10−3	0.02	1.15	0.97
	75%	−6 × 10−4	5.37 × 10−3	−6.6 × 10−3	1.08	0.99
	100%	−1.47 × 10−4	6.17 × 10−3	−0.03	1.01	0.99

shows the fitting equations of the relationships between the amount of erosion action and the CMC. The R-squared of the fitting equations is higher than 0.91. After suffering from 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the D_c values are 1.86 mm to 2.31 mm, 1.79 mm to 2.23 mm and 2.11 mm to 2.76 mm. The D_c is the highest after being treated with 20 NaCl D-As. This is ascribed to the most serious cracking by the NaCl D-As, which increases the D_c. However, the crack width and number induced by the Na₂SO₄ D-A action is the lowest, resulting in the lowest D_c. The addition of SAD can decrease the D_c, with the decreasing rates of 19.48%, 23.55% and 19.73%. It can be determined from Figure 9, that a UHPC with 100% MS (30% of the total binder materials) shows the optimum anti-carbonation and penetration performance.

$$D_c=a{N}^3+b{N}^2+cN+d$$

(6)

3.6. The SEM Photos of UHPC

The SEM pictures of the UHPC with 0%, 50% and 100% SAD are provided in Figure 10a–c. Moreover, the UHPC with 50% SAD exposed to 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As are shown in Figure 10d–f. It can be observed in the photos that the rhombic hydration products decrease with the addition of SAD. With the increasing quantity of SAD, the hydration products become more compact. Some cracks are found in the SEM photos. The UHPC shows a higher number and wider cracks after 20 NaCl D-As. Meanwhile, the cracks are the lowest when the UHPC suffers from 20 Na₂SO₄ D-As.

4. Conclusions

The paper aims to investigate the attenuation of mechanical properties after suffering salt erosion. The conclusions are as described below.

The MR of UHPC increases in the cubic function with NaCl F-C, NaCl D-A and Na₂SO₄ D-A numbers. After 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the MR are 0.76% to 1.47%, 1.24% to 1.97% and 1.42% to 2.14%.

The UHPC’s RME decreases in the cubic function with more rounds of NaCl F-Cs, NaCl D-As and Na₂SO₄ D-As. After suffering from 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the RMEs of UHPC are 89.7% to 95.3%, 89.4% to 95.7% and 86.3% to 94.8%.

The relationship between the CMC and D_c and the amount of salt action shows the cubic function. When 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As are finished, the CMCs are 3.68 × 10⁻¹² m²/s to 4.16 × 10⁻¹² m²/s, 3.59 × 10⁻¹² m²/s to 3.92 × 10⁻¹² m²/s and 4.03 × 10⁻¹² m²/s to 6.03 × 10⁻¹² m²/s, while the corresponding D_c values are 1.86 mm to 2.31 mm, 1.79 mm to 2.23 mm and 2.11 mm to 2.76 mm.

The f_t and f_cu demonstrate decreasing trends with the NaCl F-C, NaCl D-A and Na₂SO₄ D-A actions. After 200 NaCl F-Cs, 20 Na₂SO₄ and 20 NaCl D-As, the flexural strengths of UHPC decrease with the decreasing rates of 0.99%–25%, 3.92%–27.84% and 1.47%–21.59%, while the f_cu of UHPC decreases with the decreasing rates of 1.59%–10.31%, 3.09%–11.45% and 1.63%–9.62%. The CMCs are decreased with the rates of 0% to 11.53%, 0% to 33.17% and 0% to 8.41% by adding SAD, after the salt action. The SAD can reduce the D_c with the decreasing rates of 19.48%, 23.55% and 19.73%.

UHPC with SAD shows an improved compact structure. After exposure to salt action, the UHPC that suffered from 20 NaCl D-As shows the largest number and highest width of cracks. However, when specimens are exposed to 20 Na₂SO₄ D-As, the number of cracks is the lowest and the width is the narrowest. When the mass ratio of SAD is 100% (30% of the total binder materials), the resistances of the UHPC’s CO₂ penetration and NaCl and Na₂SO₄ dry–wet alternations are the best.