Study on the Effect of Fly Ash on Mechanical Properties and Seawater Freeze–Thaw Resistance of Seawater Sea Sand Concrete

Abstract

When using seawater and sea sand as mixes, the mechanical properties and durability of concrete are adversely affected because the raw materials themselves contain harmful ions. Fly ash is the tailings formed in the process of industrial production, the use of which does not require the burning of clinker, reducing CO₂ emissions. Moreover, it belongs to a new type of cementitious materials with low emissions and high environmental protection. Fly ash enhances the properties of concrete and reduces the effect of harmful ions on concrete. Based on the above considerations, the corresponding specimens were prepared and subjected to cubic compressive strength, flexural strength, and seawater freezing and thawing resistance tests by using fly ash admixture as the main variable. A combination of macro-analysis and micro-analysis was used to investigate the effect of fly ash on the performance of seawater sea sand concrete. The results showed that fly ash significantly enhanced the mechanical properties and resistance to seawater freezing and thawing of seawater sea sand concrete. The best improvement in compressive strength and resistance to seawater freezing and thawing was achieved at a substitution rate of 20%. The maximum increase in compressive strength was 13.22%. The maximum reduction in mass loss rate was 57.26% and the strength loss rate was 43.14% after the specimens were subjected to seawater freezing and thawing 75 times. The maximum enhancement in flexural strength was 17.06% for a substitution rate of 10%. Through microanalysis, it can be seen that the incorporation of coal ash can enhance the compactness of concrete through the microaggregate effect as well as the volcanic ash reaction to promote the secondary hydration reaction, so as to strengthen the seawater freeze–thaw resistance of seawater sea sand concrete. Finally, the damage prediction model established using the mean GM (1, 1) model of gray system theory meets the requirements of the first level of prediction accuracy and can accurately predict the damage of seawater sea sand concrete under seawater freezing and thawing.

1. Introduction

The demand for concrete as a major material for engineering and construction is increasing. China is a large country for concrete production and consumption, and the demand for sand, stone, and water as raw materials for concrete preparation is increasing. According to the 2015–2020 China Construction Industry Market Analysis and Investment Strategy Planning Report, China’s total sand and gravel production accounts for about one-third of the world’s total sand and gravel resources. The demand for sand and gravel in China reached 140 × 10¹⁰ t in 2014, and the demand is still increasing and is expected to reach 250 × 10¹⁰ t in 2030. The growing conflict between supply and demand for sand and gravel will lead to an increase in the price of the material. Uncontrolled exploitation leads to serious damage to ecological resources and causes many ecological problems [1,2].

As 70% of the earth’s area, the ocean is rich in resources, providing mankind with abundant fisheries, oil, natural gas, and other resources, of which seawater is considered an inexhaustible resource. With a vast sea area and a long and narrow coastline, China’s offshore sea sand is rich in resources, with total resources ranging from 67.96 × 10¹⁰ to 68.49 × 10¹⁰ m³ [3]. With the implementation of China’s ocean power and far-sea development strategy and the construction of ports, wharves, and islands, the demand for concrete is bound to increase. Moreover, there is a lack of river sand and freshwater resources in offshore areas. If seawater can be used instead of fresh water, sea sand instead of river sand can not only reduce the transportation cost and the overall construction cost [4,5] but also local materials and reduce the damage to the environment caused by mining river sand [6,7]. As a result, seawater sea sand concrete was created. In addition, some scholars have shown [8,9] that the basic mechanical properties of sea sand and river sand are similar, and the original sea sand is mainly medium-coarse sand compared with river sand, which has the advantages of lower mud content and moderate fineness; however, the content of shells and Cl⁻ is relatively high. The high content of shells can reduce the effective sand ratio of concrete, thus reducing the ease of concrete, strength, and the expansion and contraction of concrete in the later stage of the expansion and contraction, creep, and other adverse effects. The high content of chloride ions and other corrosive ions in seawater sand, once misused, will surely cause serious consequences, as evidenced by the emergence of the phenomenon of “seawater sand house”. A total of 166 commercial buildings were constructed in Zhoushan from 1994 to 1996, and the phenomenon of “seawater sand house” gradually appeared two years later. “In 2005, the Jinshan Mingzhu District in the suburb of Fenghua City caused the leakage of reinforcing steel due to the illegal use of sea sand [10]. Therefore, investigating the influence of corrosive ions in seawater sand on concrete structures is a prerequisite to ensure the quality and safety of the project.

Seawater sea sand concrete is a concrete material in which the mixing water is seawater and the fine aggregate is sea sand. Since seawater sea sand inherently contains chloride and sulfate salts [11], it is directly involved in the hydration reaction of cement when used as a mix. Among them, chloride and sulfate salts will accelerate cement hydration and play an early strength role, but the late strength growth is slow and, ultimately, comparable to the strength of ordinary concrete [12,13,14]. Tixier et al. [15], in their study of natural seawater sea sand concrete, found that despite the fact that seawater sea sand impeded the strength of the concrete, there was no significant difference in the overall mechanical properties compared to normal concrete. Islam et al. [16] compared seawater sea sand concrete with ordinary concrete. After the age of curing reached 7d, the strength of seawater sea sand concrete was superior to that of plain concrete regardless of the type of curing (seawater curing and freshwater curing). In view of the existence of corrosive ions in this seawater sea sand, many scholars at home and abroad have proposed methods to solve the problem. For example, the addition of mineral admixture, desalination of seawater sand, the use of corrosion-resistant reinforcement, etc. [17,18,19,20,21].

Fly ash is industrial waste residue, and considering China’s annual emissions of more than 400 million tons, the treatment of waste residue requires a large number of sites, as poor treatment will cause environmental pollution and even destroy the ecological environment [22]. Fly ash as a mineral admixture for concrete not only has the advantages of reducing the heat of hydration of concrete, optimizing the pore structure, anti-freezing, chemical corrosion resistance, etc., but also its own volcanic ash reaction can improve the late strength of concrete [23]. Scholars believe that, compared with ordinary silicate cement concrete, the addition of mineral admixtures makes the cement gel layer more compact and strengthens the interfacial transition zone (aggregate–cement interface) (ITZ), which, in turn, strengthens the concrete matrix [24,25]. Ying Tao Li et al. [26] conducted an experimental study on seawater, sea sand, and seawater coral sand concrete, and the results showed that the modulus of elasticity of seawater sea sand and seawater coral sand concrete is slightly lower compared with ordinary concrete, and the mechanical properties of seawater sea sand concrete can be improved by adding fly ash and slag. Karthikeyan et al. [27] formulated C30 grade concrete with sea sand mass replacements of 10%, 20%, 30%, and 40%, adding silica fume as an external admixture. It was shown that the highest strength of concrete was obtained when the replacement rate of sea sand was 30% and silica fume was added at 5%. Manjunath [28] et al. used granulated blast furnace slag as an admixture into concrete and found, using electron microscope scanning, that the internal structure of slag concrete after high temperatures was denser than that of normal concrete.

Considering that the majority of seawater sea sand concrete is used in marinas in offshore and coastal areas, the working environment is often a marine environment [29]. Concrete specimens are not only affected by physical factors such as tidal flushing or drying, wetting, freezing, and thawing but also by chemical reactions with salt in seawater. In the northern region of China, 22% of dams and 21% of small and medium-sized hydraulic buildings have freeze–thaw problems; moreover, concrete building damage caused by freeze–thawing is a common phenomenon in the northern region. The dual-factor action of freeze–thaw cycles and corrosive media exacerbates the damage to the durability of concrete, resulting in the premature failure of concrete specimens. It is thus necessary to study the process of seawater sea sand concrete change under seawater freeze–thaw conditions.

At present, to study the effect of seawater freeze–thawing on concrete, the mixing water and corrosion medium are mostly artificial seawater. Some scholars have studied the effect of single ions (CI⁻ and SO₄²⁻) on concrete [30,31]. This is somewhat different from the reality of the marine environment. Based on the above considerations, this paper investigates the mechanical properties and durability of seawater sea sand concrete using the replacement rate of fly ash as the main variable. The effects of freezing and thawing on the appearance, microstructure, and physical phase composition of seawater sea sand concrete were analyzed using SEM electron microscope scanning. Gray system theory was applied to predict the freeze–thaw damage of the seawater sea sand concrete to provide certain theoretical references for related engineering practice.

2. Experimental Design

2.1. Raw Material

(1) Cement: Bohai brand P-O 42.5 grade ordinary silicate cement is used in this test, originating from Huludao City, Liaoning Province (see Figure 1), and the related indexes are shown in

Table 1. Chemical composition and content of cement.

Chemical Composition	SiO2	Al2O3	Fe2O3	CaO	K2O	MgO
quantity contained (%)	19.1	6.2	3.31	64.6	5.91	2.93

and

Table 2. Basic mechanical properties of cement (MPa).

3d Compressive Strength	3d Flexural Strength	28d Compressive Strength	28d Flexural Strength
16.4	1720	2568	7.31

.

(2) Fine aggregate: the sea sand used in the test is the sea sand near Daling River in Linghai City, Jinzhou; the grade is medium sand with good grading; the sea sand is shown in Figure 2, and the related indexes are shown in

Table 3. Fine aggregate index.

Fineness Modulus	Apparent Density (kg/m3)	Packing Density (kg/m3)	Mud Content (%)
2.44	2660	1549	3.2

. The composition of the sea sand used in this test differs from that of other coastal areas in China, as shown in

Table 4. Chemical composition of sea sand and its content in different sea areas of China.

Coastal City	Sea Area	Cl− (%)	Shell (%)	Mud (%)
Zhujiangkou	South China sea	0.062	15	2.75
Fujian	East China sea	0.073	10.02	1.65
Qingdao	Yellow sea	0.149	1.10	1.78
Zhangzhou	Taiwan strait	0.073	9.80	2.64

.

(3) Coarse aggregate: Coarse aggregate uses natural gravel produced in Wendilou Township, Jinzhou City, with a particle size of 5~20 mm and continuous grading, see Figure 3, and relevant indexes are shown in

Table 5. Coarse aggregate index.

Particle Gradation (mm)	Mud Content (%)	Indicators of Crushing (%)	Apparent Density (kg/m3)	Packing Density (kg/m3)
5–20	0.3	9.4	2573	1460

.

(4) Seawater: The seawater used in the test was taken from the Bohai Sea waters of Jinzhou City, Liaoning Province, and its composition is shown in

Table 6. Chemical composition of seawater (mg/L⁻¹).

Ca2+	Mg2+	Na+	K+	SO42−	Cl−	HCO3−
255	975	7638	221	2755	15,611	331

.

(5) Fly ash: Fly ash adopts Huifeng brand first-class fly ash, the appearance of which is shown in Figure 4, and its composition is shown in

Table 7. Main components of fly ash (%).

Al2O3	SiO2	H2O	Cl−	SO3	CaO
24.2	45.1	0.85	0.015	2.1	5.6

.

2.2. Test Equipment

A vertical mortar mixer of type jw-400 produced by Beijing Road Engineering Group, China, was used in the preparation of the concrete tests; the rate of strength loss was tested using the YAW-5000J compression and shear tester produced by Changchun Science and Technology Company Limited, China. The freeze-thaw cycle test was carried out using a fully automatic low-temperature freeze-thaw tester produced by Beijing Dadi Instrument Company, China. The test equipment is shown in Figure 5.

2.3. Mixing Ratio Design

According to the “Standard for Test Methods of Long-term Properties and Durability of Ordinary Concrete” (GB/T50082-2009) and “Standard for Test Methods of Physical and Mechanical Properties of Concrete” (GB50081-2002), cubic specimens of 100 mm × 100 mm × 100 mm were subjected to cubic compressive strength, mass loss, and strength loss data collection [32,33]. Then, 100 mm × 100 mm × 400 mm prisms were subjected to flexural strength data collection.

According to the “ordinary concrete proportion design regulations” (JGJ55-2011) for the proportion design [34], the concrete strength design C40, a water–cement ratio of 0.47, and a sand rate of 35% was used. Fly ash admixture (0, 10%, 20%, 30%, and 40%) was taken as the variable. The basic mechanical properties (cubic compressive strength, flexural strength) of their 28d concrete were studied. Considering the working environment of seawater sea sand concrete, the effects of seawater freezing and thawing coupling effects of 20, 50, and 75 times on the rate of loss of quality and the rate of loss of strength of concrete are investigated. The specific mixing ratios and number of specimens are shown in

Table 8. Material table for concrete mix ratio (kg/m³).

Serial Number	Sea Water	Fine Aggregate	Coarse Aggregate	Cement	Fly Ash
W1S1F0C	195	626.5	1163.5	415	0
W1S1F0.1C	195	626.5	1163.5	373.5	41.5
W1S1F0.2C	195	626.5	1163.5	332	83
W1S1F0.3C	195	626.5	1163.5	290.5	124.5
W1S1F0.4C	195	626.5	1163.5	249	166

Note: W is seawater, S is sea sand, F is fly ash, C is cement, e.g., W₁S₁F_0.1C: concrete specimen with 10% fly ash substitution rate.

and

Table 9. Number and grouping of tests.

Serial Number	Mechanical Property		Resistant to Freezing and Thawing of Sea Water
	Compressive Strength	Flexural Strength	Mass				Strength
			25 Times	50 Times	75 Times	Contrast	25 Times	50 Times	75 Times	Contrast
W1S1F0C	3	3	3	3	3	9	3	3	3	9
W1S1F0.1C	3	3	3	3	3	9	3	3	3	9
W1S1F0.2C	3	3	3	3	3	9	3	3	3	9
W1S1F0.3C	3	3	3	3	3	9	3	3	3	9
W1S1F0.4C	3	3	3	3	3	9	3	3	3	9

.

2.4. Test Piece Fabrication

During the mixing process, first place the stone and sand into the mixer and dry mix for 60 s. Then, place in cement and fly ash and mix them dry for 60 s. Finally, add seawater and mix them wet for 90 s to make the concrete mix evenly.

Place the well-mixed concrete into the mold, then place the mold on the vibrating table using a plastering trowel along the mold around the insertion of the pounding. After compaction and smoothing, the specimen is completed, as shown in Figure 6.

2.5. STEPS of Seawater Freeze–Thaw Test

The specimens were placed in a pre-prepared stainless steel tank, and then seawater was poured into the tank and soaked with seawater for 4 days. In the immersion process, in order to ensure that the seawater has a sufficient level of liquid height, the specimen surface box side wall spacing is no less than 20 mm, and the seawater plane exceeds the uppermost surface of the specimen by 20 mm. After the end of the infiltration, the sink was placed into the freeze–thaw box together with the specimen, and the seawater freeze–thaw test is shown in Figure 7. Care should be taken to ensure that the concentration of seawater is maintained during the test and that the seawater used for immersion is changed every other week for this test. Weighing and strength tests were performed at 25 intervals.

2.6. Data Calculation Formula

Compressive strength and flexural strength data are calculated as Equations (1) and (2).

$$f_{cu}=\frac{F}{A}$$

(1)

where f_cu is the compressive strength of cubes (MPa); F is the load at the time of destruction of the specimen (N); A is the compressed area of the specimen (mm²).

$$f_f=\frac{Fl}{b{h}^2}$$

(2)

where f_f is the flexural strength of concrete (MPa); F is the load at the time of destruction of the specimen (N); l is the span between the supports (mm); h is the height of the section; and b is the width of the section.

The value of the mass loss rate was calculated according to Equation (3).

$$\Delta W_{ni}=\frac{W_{0i}-W_{ni}}{W_{0i}}\times 100$$

(3)

where $\Delta W_{ni}$ is the mass loss rate (%) of the ith concrete specimen after experiencing seawater erosion for N days, $W_{0i}$ is the mass (g) of the ith concrete specimen without experiencing seawater erosion, and $W_{ni}$ is the mass (g) of the ith concrete specimen after experiencing N freeze–thaw cycles.

The rate of strength loss is shown in Equation (4).

$$\Delta f_c=\frac{f_{c0}-f_{cn}}{f_{c0}}\times 100$$

(4)

where Δf_c is the strength loss rate (%) of the ith concrete specimen after experiencing seawater erosion for N days, f_c0 is the strength of the ith concrete specimen without experiencing seawater erosion (MPa), and f_cn is the strength of the ith concrete specimen after experiencing N freeze–thaw cycles (MPa).

3. Results and Analysis

The results of the mechanical property tests are shown in

Table 10. Mechanical testing test results.

Serial Number	Compressive Strength/Mpa	Flexural Strength/Mpa
W1S1F0C	42.67	5.45
W1S1F0.1C	45.56	6.38
W1S1F0.2C	48.31	6.09
W1S1F0.3C	43.29	5.73
W1S1F0.4C	41.18	5.51

and the data were plotted as bar graphs as shown in Figure 8.

3.1. Compressive Strength Analysis

From Figure 8, it can be found that the specimens show a tendency to first increase and then decrease in cubic compressive strength as the replacement rate of fly ash increases. The compressive strength of the specimens without fly ash was significantly lower than that of the fly ash incorporated series. This is because CI⁻ and SO₄²⁻ in seawater and sea sand react with hydrates in cement to form salt crystals (“Friedel” salts) and alumina crystals (AFt), which cause volume expansion of the concrete, generating internal stresses and resulting in cracks that encourage the diffusion of Mg²⁺ into the cementitious material. The conversion of Mg²⁺ with C-S-H into M-S-H reduces the alkalinity of the cement paste, destroys the conditions for the stable existence of C-S-H hydration products, and decomposes the hydration products such as C-S-H, resulting in a loss in concrete strength. With the enhancement of the fly ash admixture, the enhancement over the basic group W₁S₁F₀C was 6.77%, 13.22%, 1.45%, and −3.49%, respectively. Ming fly ash has the ability to enhance the compressive strength of seawater sea sand concrete. The reasons analyzed are as follows: (1) Adding fly ash reduces cement flocculation, releases some of the free water, and leads to an increase in the actual strength of the concrete. (2) The active components SiO₂ and Al₂O₃ in fly ash can only participate in the reaction when Ca(OH)₂ is produced by cement hydration, and with the curing age reaching 28d, the active components in fly ash react with the hydration products of cement to produce more hydrated calcium silicate and hydrated calcium aluminate, which reduces the internal pores of the concrete and makes the internal structure of the concrete more dense. (3) At the early stage of curing, the morphological effect and microaggregate effect of the fly ash itself are not stimulated. With the prolongation of the age of concrete curing, its effect plays a role, so that it promotes the formation of aggregate cementing material, thereby improving the particle gradation and ease of concrete while reducing porosity, improving concrete compactness, and, thus, improving the strength of concrete [35]. Analyzing the data, it can be seen that a fly ash dosage of 40% has a negative effect on compressive strength, indicating that the dosage of fly ash is not too large. The reason for this is the high CaO content within the cement and the large proportion of Ca(OH)₂ in the hydration products. Ca²⁺ has a significant promoting effect on the generation of C-S-H gels. The alkalinity provided in the liquid phase is sufficient to promote thorough hydration of SiO₂ and Al₂O₃. However, the low CaO content in fly ash leads to a low alkalinity coefficient of fly ash; moreover, if it replaces cement in a large admixture, the alkalinity in the liquid phase is not enough to promote the thorough hydration of the admixture, and the strength of the concrete decreases due to the insufficient cementitious material in the concrete [36].

3.2. Flexural Strength Analysis

As can be seen in Figure 8, the incorporation of fly ash enhances the flexural strength of the specimens significantly, and the strengths are all greater than those of the basic group W₁S₁F₀C. The flexural strength showed an increasing and then decreasing trend when the doping level was varied from 0% to 40%. The lowest flexural strength was obtained for W₁S₁F₀C. This is because seawater and sea sand carry chloride ions, sulfate ions, magnesium ions, and other components for the concrete material, causing a lot of uncertainty, and resulting in increased brittleness within the specimen, which adversely affects the flexural strength. The highest flexural strength of 6.38 MPa was obtained at a dosage of 10%, which is a 17.06% improvement over the basic group. Under the action of OH⁻, the active ingredient in fly ash undergoes three processes of dissolution, depolymerization, and condensation, generating a C-S-H gel with high adhesion, which strengthens the ITZ interfacial zone (aggregate–cement interfacial transition zone) of the matrix and improves the densification of the matrix, resulting in a certain increase in the strength of the matrix. As cement hydration proceeds, fly ash undergoes a secondary volcanic ash reaction with the hydration product Ca(OH)₂ and the C-S-H gel generated from clinker hydration, further generating a low alkalinity C-S-H gel, which significantly strengthens the bond of the cement paste [37].

The flexural strength curve shows a decreasing trend when the dosage of fly ash exceeds 10%. This is because as more cement is replaced by fly ash, the alkalinity in the cement paste is not enough to make the active ingredients in the fly ash completely transform into C-S-H gel; most of the fly ash only plays a filling role, in which the generation of cementitious substances in the matrix is reduced, resulting in a slight reduction in the flexural strength of concrete. In addition, the flexural strength of seawater sea sand concrete was significantly reduced due to the reduction in cement dosage by the replacement of cement with excess fly ash, the reduction in the production of calcium hydroxide, one of the hydration products, and the reduction in the heat generated by the hydration reaction, which resulted in the stagnation of the secondary hydration reaction of the fly ash.

4. Resistance to Seawater Freezing and Thawing

The test results are shown in

Table 11. Results of seawater freeze–thaw resistance test.

Serial Number	Mass Loss Rate/%			Rate of Loss of Strength%
	25 Times	50 Times	75 Times	25 Times	50 Times	75 Times
W1S1F0C	−0.477	0.468	0.985	7.300	11.966	18.122
W1S1F0.1C	−0.358	0.301	0.734	4.505	7.659	12.890
W1S1F0.2C	−0.334	0.153	0.421	3.673	6.626	10.305
W1S1F0.3C	−0.390	0.337	0.781	5.796	9.578	15.114
W1S1F0.4C	−0.435	0.391	0.930	6.430	10.374	16.091

.

4.1. Changes in Specimen Appearance and Morphology

Figure 9 shows the damage pattern of the outer surface of the specimen after 25, 50, and 75 times of coupling. It can be seen that, with the increase in the number of coupling actions of the specimen, surface honeycomb pitted defects gradually increased, even after 75 iterations of the coupling action of the lack of corners and the phenomenon of surface detachment. It can, therefore, be speculated that there is gradual damage received from the surface as well as inside. The reason for this is that, in the early stage of freezing and thawing, the salt crystals in the microcracks on the surface of the concrete are less constrained than those inside the concrete; therefore, the expansion force of seawater crystals on the surface weakened the bond between aggregate and salt crystals after the freezing and thawing erosion, which promotes the formation of cellular pockmarks on the surface. In addition, the specimen continued to hydrate at the beginning of the freeze–thaw period, the generation of “AFt” and “Friedel’s salt” inside the concrete increased gradually, and the expansion stress gradually became larger to promote the emergence and development of micro-cracks, coupled with the increase in the number of freeze–thaw cycles and amount of seawater penetrated into the matrix from the surface honeycomb and micro-cracks. In addition, with the increase in freeze–thaw cycles, seawater penetrates into the matrix from the surface honeycomb and microcracks, and the expansion stress generated by icing makes the microcracks continue to develop until cracking. On the other hand, because the freezing point of seawater is lower than ordinary tap water, supercooled water and iced water form when a part of the iced water freezes. The compression of supercooled water creates expansion stress and penetration pressure in the capillary pore. After undergoing multiple freezing and thawing cycles, they are subjected to a variety of stresses coupled with the role of the originally closed pore through the open pore to promote the continuous development of micro-cracks, resulting in an increase in the surface of the cellular pockmarked surface and surface spalling corner phenomenon.

4.2. Effect of Seawater Freezing and Thawing on the Rate of Mass Loss

As can be seen in Figure 10, the mass of the different specimens first increases and then decreases under the seawater freeze–thaw coupling. W₁S₁F₀C increased the most and the mass of W₁S₁F_0.2C increased the least. This is because seawater sea sand concrete is mixed with seawater and sea sand, and its own hydration products include Friedel’s salt and AFt, which have certain expansion properties, resulting in seawater sea sand concrete itself having certain pores. At the beginning of the freeze–thaw cycle, seawater enters the specimen along the cracks and its own pores produced by the freeze–thaw cycle, and the corrosive ions in the seawater and the cement hydrate continue to undergo a chemical reaction, generating AFt, gypsum, and “Friedel” salts and so on. The generated material is able to fill some of the pores to a certain extent, leading to an increase in the mass of the specimen. In addition, seawater enters the interior of the specimen and continues the hydration reaction with the cement and fly ash that have not completed hydration, resulting in a buildup of hydrides inside the specimen, leading to an increase in the mass of the specimen. As the number of freeze–thaw cycles increases, freeze–thaw damage and seawater erosion have a superimposed effect. When concrete undergoes freeze–thaw cycles, the internal pores and microcracks of concrete are further developed, making it easier for water solutions to intrude in seawater erosion, and the concrete suffers from increased stress caused by water absorption and expansion. The constant repetition of concrete in both forms of damage result in a faster rate of damage to the specimens in the seawater freeze–thaw environment.

Incorporation of fly ash, especially at a substitution rate of 20%, showed the most significant improvement in mass loss, which was reduced by 67.31% and 57.26% compared to the basic group after 50 and 75 seawater freeze–thaw cycles. This is due to the fact that the particle size of fly ash particles is smaller than that of cement, and the particles are intact and have a smooth surface. Mixed into concrete, it can play the role of water reduction and lubrication and has the effect of promoting the hydration of cement and changing the rheological properties of the mix [38]. The contribution to durability is due to its “active effect”, and SiO₂ and A1₂O₃ in the fly ash can be generated with the Ca(OH)₂ generated by cement hydration to occur during the second hydration reaction, generating C-S-H gel, filling the pores in the cement hydration products and playing a role in refining the pores, so that the cement paste is denser, which can improve the interface structure within the concrete, thus improving the resistance to seawater freezing and thawing.

4.3. Effect of Seawater Freezing and Thawing on the Rate of Strength Loss

Figure 11 shows the variation in the rate of strength loss for different groups of specimens. As the number of seawater freeze–thaw cycles continued to increase, the mass loss increased linearly, with the greatest strength loss occurring in the basic group W₁S₁F₀C, which reached 18.122% at 75 seawater freeze–thaw cycles. Seawater freeze–thaw damage is essentially a special form of freeze–thaw erosion damage. Salt ions have a high capacity for moisture absorption and water retention, which greatly increases the degree of water saturation of the concrete material and greatly reduces the time required for water saturation. The salt solution makes it more difficult to remove the water that enters the interior of the concrete, leaving it saturated for long periods of time. Therefore, the damage under the coupled effect of freeze–thaw cycle and seawater erosion is much more serious than the damage by freeze–thaw erosion alone. When the temperature at which the concrete material is located is below 0 °C, the water within the pores of the material is not frozen immediately because of the small pore size of the capillaries and the presence of salt ions in the solution in the pores. As the temperature continues to drop, the moisture within certain pores begins to freeze. At this point, the solution concentration inside the pores will increase due to the reduction in water. In order to reduce the concentration inside the pores where freezing occurs and to establish a new equilibrium in the salt solution concentration, water from the other unfrozen pores inside the concrete material will flow into the pores where freezing occurs. The new influx of moisture causes the volume of ice and solution in the pores to grow. At the same time, the osmotic pressure inside different pores due to the migration of water will continue to increase, and the continuously increasing osmotic pressure makes the cement paste begin to produce cracks and cracking phenomenon [39].

The incorporation of fly ash can effectively improve the damage to specimens caused by seawater freeze–thaw cycles. The rate of strength loss under successive freeze–thaw cycles was reduced by 49.68%, 44.63%, and 43.14% for a substitution rate of 20% compared to the basic group W₁S₁F₀C. The reason for this is that fly ash particles have a much smaller particle size, making their specific surface area larger. Therefore, the addition of fly ash improves the flow and workability of concrete when other aggregates remain unchanged. Small spherical fly ash particles can fill some of the space between the coarse particles of cement, making the particle size distribution of fine particles more reasonable. The addition of fly ash can increase the cohesion of concrete; moreover, because of its large specific surface area, many free waters can be constrained by fly ash, which can reduce the amount of water secretion and aggregate segregation. Fly ash phase has very strong “volcanic ash properties” [40], and although there is basically no hydration with the water, it can be used in the cement hydration product Ca (OH)₂ and some other compounds under the stimulation of the second hydration reaction, which occurs very quickly, generating a higher strength of the product of the gelling properties of the C-S-H gels. In addition to the microaggregate properties of fly ash, it can not only fill the harmful holes in the cement paste after freezing and thawing of seawater but also improve the microstructure of the cement paste through its secondary hydration products. The upliftment effect is less, at a replacement rate of 40%. This is because the excessive incorporation of fly ash affects the flowability and plasticity of concrete. Excessive incorporation of fly ash will deteriorate the flowability of concrete. In addition, the particle shape and size distribution of fly ash is different from that of conventional aggregates, which may affect the workability of concrete. The fineness and activity of fly ash are low, and excessive admixture may affect the cementation of concrete, which may lead to the lower performance of concrete.

5. Microanalysis

W₁S₁F₀C and W₁S₁F_0.2C were sampled, and the specimens were scanned using SEM to investigate the damage mechanism of seawater freezing and thawing on the specimens. Figure 12 shows the microstructure of W₁S₁F₀C and W₁S₁F_0.2C specimens before seawater freeze–thaw. As can be seen in Figure 11a, a large number of sheet-like Ca(OH)₂ and flocculent C-S-H gels are tightly articulated and have a relatively dense structure; however, there are slight pores in localized positions. This is due to the fact that AFm is a common substance in cement concrete called monosulfur-type calcium sulfoaluminate with the chemical formula 3CaO·Al₂O₃·CaSO₄·12H₂O, and AFm has less structural water compared to AFt. When sulfate ions in seawater and sea sand react with AFm and convert to AFt, the structural water increases and the volume expands, which can cause structural damage to the cement paste. CI⁻ in seawater and sea sand hydrates with cement to form “Friedel” salt. This substance is brittle and hinders the continuation of hydration when encapsulating the cement paste. In addition, SO₄²⁻ promotes the release of CI⁻ from the “Friedel” salt, which causes secondary damage to the concrete, as shown in Figure 12b. When mixed with fly ash, the hydration products are interspersed with each other, bonded to each other, and lapped to each other, and there are no obvious holes on the surface. This is because the fineness of fly ash is about a hundred times that of ordinary silicate cement, and such a high fineness is conducive to filling the capillary pores inside the concrete, enhancing the compactness of the concrete and impeding the infiltration of CI⁻ and SO₄²⁻. Thereby, the performance of the specimens after the incorporation of fly ash is better than the basic group, as shown in Figure 12c.

The internal condition of the specimen after the freezing and thawing of the W₁S₁F₀C specimen is shown in Figure 13a–c. It can be noticed that the crack width and number gradually increase with the amount of seawater freezing and thawing. With the increase in the number of seawater freezing and thawing, the capillary pores gradually increase, resulting in the acceleration of the propagation rate of corrosive ions, and the damage of the specimen is gradually aggravated. At the same time, AFt, “Friedel” salt, Mg(OH)₂, and other substances produced by the reaction between erosion ions in seawater and the hydration products of the slurry cover the surface of the unhydrated cementitious material particles to hinder their hydration, resulting in the cement slurry tending to be loose, with a large number of pores in the structure; moreover, there is almost no overlap between the hydration products’ looseness and porosity (Figure 13d).

Under coupling, a portion of the solution in the internal pores freezes as the temperature decreases. As the water becomes ice, the volume expands, forcing the unfrozen water in the hole to flow and thus generate hydrostatic pressure [41], as shown in Figure 14. When the hydrostatic pressure generated is greater than the tensile strength of the concrete pore generated for the cracks, the generation of micro-cracks accelerates the rate of seawater erosion so that the products generated by the reaction will be attached to the cracks. As the erosion time lengthens, the increase in reaction products will further widen the cracks, and freeze–thaw cycles occur simultaneously with seawater erosion. As the number of coupling increases, the cracks inside the specimen will gradually increase in size. When the coupling reaches a certain number of times, the pores will be penetrated by cracks between the pores, causing changes in the quality and strength of the specimen. Seawater freeze–thaw coupling is a typical damage problem of two-factor coupling between corrosion and freeze–thaw. Compared with seawater erosion and freeze–thaw cycle single factor, the coupling effect has a “1 + 1 > 2” damage effect. In essence, it is the superposition of chemical action and physical action, and the two influence each other to accelerate the damage of concrete specimens. Compared with ordinary concrete, seawater sea sand concrete contains chlorine salt and sulfate in its own mixture, and its hydration reaction product originally has “Friedel” salt and Aft; under the coupling effect, seawater erosion products will have a superposition effect, which will strengthen the damage degree of the coupling effect, which leads to a gradual increase in cracks in the specimen.

An EDS energy spectrometer was used to score the interior of the slurry of W₁S₁F₀C for elemental analysis, and the location of the score is shown in Figure 15a. The elemental composition and content of concrete after seawater freezing and thawing were obtained through an elemental capture of the samples after seawater freezing and thawing, and the spot-scan results are shown in Figure 15b–d. From Figure 15b–d, ions such as Ca, O, Al, and Cl can be seen. The hydrides can be introduced roughly as 3CaO·Al₂O₃·3CaSO₄·32H₂O (Aft), 3CaO·Al₂O₃·CaCl₂·10H₂O (“Friedel” salts), and CaSO₄·2H₂O (gypsum). The action of these substances with expansive properties led to the deterioration of the specimens after seawater freezing and thawing. Some support is provided for the interpretation of the internal generators as well as erosion products of seawater sea sand concrete.

As shown in Figure 16, after mixing fly ash and a freeze–thaw cycle, the hydration products are interspersed with each other and bonded to each other; moreover, the contact at the hydride lap point is better, the pores in the structure are fewer, and the densification of the structure is higher. The reasons for this are as follows: (1) the effect of fly ash on concrete can be divided into the volcanic ash reaction effect, microaggregate effect, and particle morphology effect. Among them, the volcanic ash effect is a chemical reaction and the latter two are physical effects. Studies have shown that fly ash plays a much greater physical than chemical role in the early stages of cement hydration; however, the role of the volcanic ash effect becomes progressively greater as age increases. Fly ash particles are smaller than cement particles, and more than 70% are smooth, intact glass particles that are mixed into the concrete to play the role of lubrication. The appropriate amount of mixing can change the working properties of cement, and the fine particles themselves can be filled into the pores produced by the hydration of cement so that the concrete itself is denser [42]. (2) Higher aluminum content in fly ash is conducive to reducing the penetration rate of chloride ions in concrete and improving the resistance of concrete to chloride ion penetration. (3) CI⁻ and cement hydrate can form sodium hydroxide, effectively promote fly ash vitreous network depolymerization and disintegration, and release its internal soluble active silica and alumina, affecting the cement hydration products of the calcium hydroxide in the volcanic ash reaction. This accelerates the speed of the fly ash cement hydration reaction so that the amount of cement hydration products continues to increase; moreover, the hardening of the cement paste porosity continues to decrease, thereby improving the performance of the concrete.

6. Damage Prediction Based on GM (1, 1) Modeling

6.1. Modeling Steps

Considering the large discrete nature of the experimental data, the establishment of a mean GM (1, 1) prediction model was used [43,44]. The raw data of the strength loss rate of seawater sea sand concrete under seawater freezing and thawing were weakened by the cumulative buffer operator method, using MATLAB R2020b software for the relevant weakening.

Step 1: Raw data of strength loss rate increases with the number of seawater freezing and thawing, and the strength damage increases, which belongs to a monotonically increasing trend. To weaken the original intensity loss data sequence X, according to Equation (5), the average weakening buffer operator can be calculated (see

Table 12. Weakened buffer treatment for strength loss rate.

Specimen Grouping		0 Times	25 Times	50 Times	75 Times
W1S1F0C	raw data	0	7.300	11.966	18.122
	Average weakening buffer operator	9.347	12.463	15.044	18.122
W1S1F0.1C	raw data	0	4.505	7.659	12.890
	Average weakening buffer operator	6.264	8.351	10.275	12.890
W1S1F0.2C	raw data	0	3.673	6.626	10.305
	Average weakening buffer operator	5.151	6.868	8.466	10.305
W1S1F0.3C	raw data	0	5.796	9.578	15.114
	Average weakening buffer operator	7.622	10.163	12.346	15.114
W1S1F0.4C	raw data	0	6.430	10.374	16.091
	Average weakening buffer operator	8.224	10.965	13.233	16.091

).

$$\mathrm{x}(\mathrm{k})\mathrm{d}=\frac{1}{\mathrm{n}-\mathrm{k}+1}\left[\mathrm{x}(\mathrm{k})+\mathrm{x}(\mathrm{k}+1)+\cdots +\mathrm{x}(\mathrm{n})\right]$$

(5)

Step 2: Let the average weakening buffer operator be the sequence X⁽⁰⁾.

$${\mathrm{X}}^{(0)}=[{\mathrm{X}}^{(0)}(1),{\mathrm{X}}^{(0)}(2),{\mathrm{X}}^{(0)}(3),\cdot \cdot \cdot ,{\mathrm{X}}^{(0)}(\mathrm{n})]$$

(6)

The 1-AGO sequence is obtained by performing one cumulative processing of Equation (6).

$${\mathrm{X}}^{(1)}=[{\mathrm{X}}^{(1)}(1),{\mathrm{X}}^{(1)}(2),{\mathrm{X}}^{(1)}(3),\cdot \cdot \cdot ,{\mathrm{X}}^{(1)}(\mathrm{n})]$$

(7)

The results of one cumulative treatment calculation are shown in

Table 13. 1-AGO Sequences Processed in One Accumulation.

Specimen Grouping		0 Times	25 Times	50 Times	75 Times
W1S1F0C	Average weakening buffer operator	9.347	12.463	15.044	18.122
	1-AGO	9.347	21.810	36.854	54.976
W1S1F0.1C	Average weakening buffer operator	6.264	8.351	10.275	12.890
	1-AGO	6.264	14.615	24.890	37.780
W1S1F0.2C	Average weakening buffer operator	5.151	6.868	8.466	10.305
	1-AGO	5.151	12.019	20.485	30.790
W1S1F0.3C	Average weakening buffer operator	7.622	10.163	12.346	15.114
	1-AGO	7.622	17.785	30.131	45.245
W1S1F0.4C	Average weakening buffer operator	8.224	10.965	13.233	16.091
	1-AGO	8.224	19.189	32.422	48.513

.

Step 3: Calculate the sequence of immediately neighboring mean equal weights of the buffer operator obtained from Equation (7) as in Equation (8).

$${\mathrm{Z}}^{(1)}=[{\mathrm{Z}}^{(1)}(1),{\mathrm{Z}}^{(1)}(2),{\mathrm{Z}}^{(1)}(3),\cdot \cdot \cdot ,{\mathrm{Z}}^{(1)}(\mathrm{n})]$$

(8)

Step 4: Construct the B and Y matrices based on the average weakening buffer operator for the sequence X⁽⁰⁾ and the immediately adjacent mean-equally weighted sequence Z⁽¹⁾, and compute the model parameter vector $\stackrel{\wedge }{a}={[a,b]}^T$ coefficients a and the gray role b.

$$\begin{gathered} \mathrm{B}=\left[\begin{array}{cc}-{\mathrm{z}}^{\left(1\right)}\left(2\right)& 1\\ -{\mathrm{z}}^{\left(1\right)}\left(3\right)& 1\\ ⋮& ⋮\\ -{\mathrm{z}}^{\left(1\right)}\left(\mathrm{n}\right)& 1\end{array}\right]\mathrm{Y}=\left[\begin{array}{c}-{\mathrm{x}}^{\left(0\right)}\left(2\right)\\ -{\mathrm{x}}^{\left(0\right)}\left(3\right)\\ ⋮\\ -{\mathrm{x}}^{\left(0\right)}\left(\mathrm{n}\right)\end{array}\right], \\ \stackrel{\wedge }{\mathrm{a}}={(\mathrm{a},\mathrm{b})}^{\mathrm{T}}={({\mathrm{B}}^{\mathrm{T}}\mathrm{B})}^{-1}{\mathrm{B}}^{\mathrm{T}}\mathrm{Y} \end{gathered}$$

Step 5: Construct the GM (1, 1) model time response using Equation (9) and add and subtract the reduction process to Equation (9) to obtain Equation (10). The prediction model is shown in

Table 14. Model for predicting the rate of loss of compressive strength of cubes.

Specimen Grouping	Predictive Modeling of Strength Loss Rate of Concrete in Freeze–Thaw Environment
W1S1F0C	${\widehat{\mathrm{x}}}^{(0)}(\mathrm{k})=9.973{\mathrm{e}}^{0.187(\mathrm{k}-1)}$
W1S1F0.1C	${\widehat{\mathrm{x}}}^{(0)}(\mathrm{k})=6.657{\mathrm{e}}^{0.218(\mathrm{k}-1)}$
W1S1F0.2C	${\widehat{\mathrm{x}}}^{(0)}(\mathrm{k})=5.612{\mathrm{e}}^{0.201(\mathrm{k}-1)}$
W1S1F0.3C	${\widehat{\mathrm{x}}}^{(0)}(\mathrm{k})=8.289{\mathrm{e}}^{0.198(\mathrm{k}-1)}$
W1S1F0.4C	${\widehat{\mathrm{x}}}^{(0)}(\mathrm{k})=8.21{\mathrm{e}}^{0.192(\mathrm{k}-1)}$

.

$${\widehat{\mathrm{X}}}^{(1)}(\mathrm{k})=[{\mathrm{X}}^{(0)}(1)-\frac{\mathrm{b}}{\mathrm{a}}]{\mathrm{e}}^{-\mathrm{a}(\mathrm{k}-1)}+\frac{\mathrm{b}}{\mathrm{a}}$$

(9)

$${\widehat{\mathrm{X}}}^{(0)}(\mathrm{k})=(1-{\mathrm{e}}^{\mathrm{a}})[{\mathrm{X}}^{(0)}(1)-\frac{\mathrm{b}}{\mathrm{a}}]{\mathrm{e}}^{-\mathrm{a}(\mathrm{k}-1)},\mathrm{k}=1,2,\cdots \mathrm{n}$$

(10)

According to the mean GM (1, 1) damage prediction model of different specimens derived from Table 14, the prediction model was assigned to obtain the predicted value, and the predicted value and the original data Origin 2018 plotting software was used to construct a graph of the strength loss pattern of seawater sea sand concrete under seawater freezing and thawing as shown in Figure 17. From the figure, it can be clearly seen that the original data and the predicted data are highly fitted, and the curve trend is basically the same; therefore, it can be preliminarily judged and concluded that the mean GM (1, 1) damage prediction model is suitable for the damage prediction of seawater sea sand concrete under the seawater freezing and thawing environment.

6.2. Validation of the Accuracy of the Mean GM (1, 1) Damage Prediction Model

An approximate judgment of the trend of the fitted image is not strong proof of the problem of the accuracy of the model for damage prediction; therefore, the relevant parameters of the mean GM (1, 1) damage prediction model are given in

Table 15. Model parameters and accuracy tests.

Specimen Grouping	a	b	Average Relative Error/%
W1S1F0C	−0.187	9.563	0.370
W1S1F0.1C	−0.218	6.043	0.614
W1S1F0.2C	−0.201	5.159	0.425%
W1S1F0.3C	−0.198	7.627	0.436%
W1S1F0.4C	−0.192	8.320	0.404%

. It can be seen that the average relative error of the predicted values of the five groups of specimens based on the mean GM (1, 1) damage prediction model is less than 5%, and the fitting accuracy of the prediction model is high, which meets the first-grade accuracy requirements [45]. On the other hand, the development coefficient of the mean GM (1, 1) damage prediction model is a∈[−0.22, −1.18]. According to the relevant inference of Prof. Liu Sifeng and Deng Julong, considering the development coefficient a ≥ −0.3, the damage model is applicable to the prediction in the medium and long term; when −0.5 ≤ a ≤ −0.3, the damage model is applicable to the short-term prediction [45,46]. Therefore, it can be confirmed that seawater sea sand concrete, based on the mean value of the damage prediction model GM (1, 1), can be used to predict the medium and long-term damage life.

7. Conclusions

(1) Fly ash enhances the cubic compressive strength, flexural strength, and seawater freeze–thaw resistance of seawater sea sand concrete to some extent. The maximum value of compressive strength is 48.31 Mpa, and the maximum improvement of compressive strength is 13.22%. The maximum value of flexural strength is 6.38 Mpa, and the maximum improvement of flexural strength is 17.06%. The minimum value of mass loss was 0.734% and the maximum reduction in mass loss rate was 57.26% after the specimens were subjected to seawater freezing and thawing 75 times. The minimum value of strength loss was 12.89% and the maximum reduction in strength loss rate was 43.14%. A comparison of the data shows that for compressive strength and resistance to seawater freeze–thaw tests, the optimum substitution rate of fly ash is 20%. For flexural strength, the optimum replacement rate of fly ash is 10%.

(2) Compared with single seawater erosion and freezing and thawing, the coupling effect has the damaging effect of “1 + 1 > 2”, and the fly ash can give full play to the “volcanic ash effect” because of its own nature in the more stable chemical properties. When subjected to seawater freezing and thawing, seawater will accelerate the volcanic ash reaction of fly ash.

(3) The cracks and holes inside the seawater sea sand concrete specimens under the microstructure provide natural conditions for seawater intrusion, and, in the process of seawater freezing and thawing, the salt inside the concrete crystallizes. The AFt accumulates continuously, which, ultimately, makes the inside of the concrete brittle and porous; moreover, the cracks dilate, which reduces the service life of the concrete. Regarding the addition of fly ash-generated C-S-H gel and Aft to enhance the degree of structural compactness, the specimen did not appear to have obvious holes, and the addition of fly ash C-S-H gel increased, which is conducive to enhancing the strength and compactness of the interfacial links and enhancing the resistance to seawater freezing and thawing performance.

(4) The basic idea of gray system theory is introduced into the study of seawater sea sand concrete resistance to seawater freezing and thawing, and the damage prediction model established by using the mean value of GM (1, 1) has a high accuracy, which can provide a more reliable damage prediction and evaluation of concrete under the action of seawater erosion.