Mechanical performance study of PVA fiber-reinforced seawater and sea sand cement-based composite materials

Abstract

Due to the swift progress in the construction sector, there is a global concern about the potential scarcity of river sand and freshwater resources. The development of new construction materials is considered an inevitable trend for industry growth. PVA fibers, known for their strong corrosion resistance, cost-effectiveness, and high toughness, have the potential to enhance the corrosion resistance and seismic performance of structures in marine environments. However, their mechanical properties and durability in the seawater and sea sand environment are not well understood. Therefore, the investigation of the impact of seawater and sea sand on the mechanical properties and durability of PVA fiber-reinforced cement composites is considered crucial. A mechanical performance analysis of PVA fiber-reinforced seawater and sea sand fiber cement composites was conducted in this study. PVA fiber volume fractions of 0%, 0.75%, and 1.5%, cement composite matrix strength grades of C30 and C50, and curing periods of 28 days, 90 days, and 180 days were examined, investigating their influence on the bending toughness of PVA fiber-reinforced seawater and sea sand cement composites. Specific conclusions include the addition of fibers increased the peak bending load, had a less corrosive effect in seawater, and improved the flexural toughness of the material. The most significant improvement was observed at 1.5% fiber content, where the load-deflection curve was fuller and the energy absorption capacity of the material increased by 33–109%, maintaining good bending toughness. Furthermore, higher fiber contents are required for high-strength cementitious composites to improve flexural toughness and durability. The formulation of calculation formulas for predicting bending strength and corresponding deflection, which fit well with the experimental results; and the development of a calculation model for the bending toughness index of PVA fiber-reinforced seawater and sea sand cement composites, providing an effective prediction of material bending toughness.

Introduction

Due to the swift progress of the economy and urbanization in China, the construction industry is thriving, resulting in a significant rise in the yearly utilization of construction raw materials and contributing to an alarming depletion of resources. Nationwide, the demand for sand in construction projects is estimated to be around 3 billion tons per year, and the usage of sand and gravel is projected to reach 250 billion tons by 2030^1,2. However, the annual extraction capacity of river sand in the Yangtze River Basin, a crucial source, is only 34 million tons, significantly less than the actual demand³. Moreover, the shortage of freshwater resources is a major constraint on the development of the construction industry⁴.

Although seawater and sea sand resources are abundant, their extraction poses challenges due to the presence of harmful substances. These ions can cause damage to reinforced concrete structures, ultimately impacting the durability of concrete^5,6. To meet the production needs of the construction industry, many countries opt to use desalinated sea sand as a substitute for river sand in construction. In China, the State Oceanic Administration issued a notice titled ‘Notice on the Comprehensive Implementation of Market-Oriented Transfer of Sea Sand Mining Rights’, outlining regulations for sea sand extraction and promoting its utilization⁷.Currently, methods for desalting sea sand, which are widely recognized and frequently employed globally, include natural stacking, freshwater washing, mechanical methods, hybrid methods, and the addition of rust inhibitors to reduce various harmful ions⁸. However, each of these methods has its drawbacks; natural stacking requires extensive land use, and both freshwater washing and mechanical methods consume significant freshwater resources, intensifying the global strain on freshwater resources. Additionally, various types of rust inhibitors are available on the market, and their effectiveness requires continuous in-depth research.Against the backdrop of conserving freshwater resources and reducing engineering construction costs, researchers have undertaken studies involving the direct use of untreated sea sand and seawater in concrete⁹.

Scholars such as Takahiro Nishida¹⁰, Keisaburo KATANO¹¹, Liu Wei¹², Qin Bin¹³, and Li Tianyu¹⁴ have utilized sea sand to produce concrete, incorporating special mixtures such as blast furnace slag, silica powder, fly ash, and calcium nitrate into reinforced concrete. They have explored the mechanical properties, including compressive strength, flexural strength, and splitting tensile strength, as well as the workability of newly mixed concrete. The results indicate that the use of sea sand as a building material is feasible in both non-corrosive reinforced concrete and unreinforced concrete structures. Studies by Xing Li¹⁵ have examined the influence of sea sand and shell materials on the slump, compressive strength, and carbonation resistance of concrete. Experimental results demonstrate that the slump of sea sand concrete increases with rising salt content and decreases with an increase in shell-like materials. In comparison to ordinary concrete, sea sand concrete exhibits lower early strength, but its carbonation resistance is superior. Yang Changhai¹⁶ investigated the addition of a composite rust inhibitor in seawater concrete, resulting in excellent corrosion inhibition and improved concrete flow.Presently, fiber-reinforced cement-based composite materials are employed in the field of construction engineering to strengthen structures^17,18. Polyvinyl alcohol (PVA) fibers, known for their high elastic modulus and excellent acid and alkali resistance¹⁹, are introduced into sea sand cement-based materials to create PVA fiber-reinforced sea sand cement-based composite materials. This approach aims to achieve a rational utilization of seawater and sea sand. Scholars worldwide continue to study fiber cement-based composite materials. However, fiber cement-based materials, particularly those represented by steel fibers, still struggle to control concrete crack widths to the required standards in practical engineering. Consequently, Victor C. Li²⁰ from the University of Michigan has developed Engineered Cementitious Composites (ECC) based on micromechanics and fracture mechanics, resulting in a material with exceptional deformability and ultimate tensile strains exceeding 3%.Boasting exceptional deformation resistance, Engineered Cementitious Composites (ECC) exhibit an ultimate tensile strain exceeding 3%. Scholars such as Wei Hu²¹, Bo Xu²², He Xixi²³, Zhu Changshu²⁴, and Xue Huiqing²⁵ have conducted research on the influence of Polyvinyl Alcohol (PVA) fiber at different volume fractions on various mechanical properties of concrete. Experimental results indicate that an increase in PVA fiber content reduces the slump of cement-based composite materials, requiring an increase in the unit mixing water volume. As the volume fraction of fibers increases, concrete’s compressive strength and elastic modulus also increase. Additionally, flexural strength, splitting tensile strength, ultimate tensile/flexural strength, impact resistance, and fracture toughness strain improve. Yang Yanhua²⁶ conducted research on the frost resistance of PVA fiber cement-based composite materials. The experimental results show that the frost resistance of cement-based composite materials enhances with an increase in fiber content up to 1.5%. The optimal frost resistance is achieved at a 1.5% volume fraction of PVA fibers, and the frost resistance decreases when the fiber volume fraction reaches 2%. Peng Zhang²⁷ investigated the mechanical properties of polyvinyl alcohol fiber-reinforced cementitious composites (PVA-FRCC) under the coupled effect of humid-thermal and chloride-salt environments. The results indicated that the mechanical properties of the cementitious composites decreased after being subjected to the coupling effect of the wetthermal and chloride salt environment. The incorporation of the PVA fibers improved the mechanical properties of the cementitious composites under the coupling effect of the wet-thermal and chloride salt environment. Ming-Gin Lee²⁸ examines the properties of ordinary and high-strength fiber-reinforced pervious concrete and conducts mechanical and abrasion resistance testing on pervious concrete specimens. The results show that optimized maintenance including vacuum cleaning and high-pressure rinsing restores water permeability to over 60%. The results highlight the significant impact of the ASTM C1747 impact abrasion test. Cumulative impact abrasion rates were as high as 60 percent for normal permeable concrete, compared to about 20 percent for other high-strength specimens. Hongwei Han²⁹ investigated the physicomechanical properties of seawater coral sand-engineered cementitious composites (SCECC) under different fiber volume fraction conditions. The experimental results indicated that the compressive strength, three-point bending strength, and split tensile strength of SCECC tended to increase with the rise in fiber volume fraction. The SCECC exhibits favorable strain-hardening and multi-crack cracking characteristics.

To drive the development of the construction industry, researchers worldwide have extensively studied the mechanical properties of seawater concrete, sea sand concrete, seawater sea sand concrete, and fiber cement-based composite materials. They have highlighted the advantages and disadvantages of seawater sea sand concrete in the future construction industry. While scholars have made significant contributions to the use of fiber cement-based materials, the harmful ions in seawater sea sand that damage reinforced concrete cannot be overlooked. Most studies by scholars, both domestically and internationally, focus on the separate investigation of seawater, sea sand, and fiber-reinforced cement-based composite materials. Few studies have integrated these elements. Hence, this research proposes the incorporation of corrosion-resistant PVA fibers into seawater sea sand to create PVA fiber-reinforced seawater sea sand cement-based composite materials. This study aims to delve deeper into the mechanical properties of PVA fiber-reinforced seawater sea sand cement-based composite materials, exploring the impact of PVA fiber content, seawater concentration, and test age on the mechanical properties of materials with different strength grades. The ultimate goal is to advance the safe utilization of seawater and sea sand in the construction industry.

Experimental design

Experimental variables

The specimen design took into account four parameters: strength grades, seawater concentration, PVA fiber content, and aging periods. For each set of experiments, three cuboid specimens measuring 100 mm × 100 mm × 400 mm were designed and produced. In total, 162 cuboid specimens were designed and fabricated for the experiment.

The experiment covered two matrix-strength grades of cement-based composite materials, C30 and C50. Under each strength grade, three aging periods (28 d, 90 d, and 180 d) were considered. For each aging period, three artificial simulated seawater concentrations (0 times, 1 time, and 3 times) were designed. Under each seawater concentration, three PVA fiber volume fractions (0%, 0.75%, and 1.5%) were considered. This design aimed to investigate the influence of matrix strength grades, seawater concentration, PVA fiber content, and aging periods on the mechanical properties of PVA fiber-reinforced seawater and sea sand cement-based composite materials. The specific specimen parameters are detailed in Table 1.

Table 1 Specimen parameters.

Mix ratio design

To ensure the uniform distribution of PVA fibers in the matrix and guarantee optimal specimen performance, extensive trial mixes were conducted by continuously adjusting the incorporation ratios of composite admixtures. The resulting mix proportions per cubic meter for the experiments are shown in Table 2. The concrete mix used in the experiments adhered to common formulations, including Portland cement as the binding agent, aggregates comprising sand and gravel to provide structural support, and the essential addition of water for the chemical reaction with cement during the hardening process. Moreover, admixtures were incorporated to modify specific properties of the concrete, such as durability.

Table 2 Mixing ratio design.

Raw materials and properties

PVA fiber-reinforced seawater and sea sand cement-based composite materials consist of materials such as cement, sand, water, and PVA fibers, as shown in Fig. 1. The specifications and basic properties of the main raw materials are as follows:

• (1) Cement The cement used in this experiment is ordinary Portland cement produced by Jiande Southern Cement Co., Ltd., with a strength grade of P·O 42.5. The residue on a 0.080 mm square sieve is 2.90%, and the density is 3.14 g/m³. The cement sample is shown in Fig. 1.

• (2) Sand Wenzhou local natural sea sand without desalination treatment was chosen for this experiment. The sand was dried before testing. The original sea sand, containing harmful ions, was tested by the 11th Geological Brigade of Zhejiang Province. The results indicated sodium ion content of 0.20%, magnesium ion content of 0.10%, calcium ion content of 2.27%, chloride ion content of 0.12%, and sulfate ion content of 0.13%. The fineness modulus μ_f of the sea sand is 1.762, classifying it as fine sand. The sea sand sample is shown in Fig. 1.

• (3) Fiber PVA fibers from Kuraray Co., Ltd., Japan, were selected to enhance the overall integrity and ductility of the cement-based composite material. The fiber has a length of 12 mm, diameter of 0.026 mm, density of 1300 kg/m³, tensile strength of 1500 MPa, elastic modulus of 36 GPa, and is categorized as bundle-shaped. The PVA fiber sample is shown in Fig. 1.

• (4) Composite Admixture The composite admixture used in the experiment includes a superplasticizer, defoamer, thickener, and plastic expansion agent. The superplasticizer is a polycarboxylate-based high-efficiency water reducer, appearing as a light-yellow powder, with a pH value (20% aqueous solution) of 8.5 ± 1.0, bulk density of 580 ± 50 kg/m³, cement mortar fluidity ≥ 240, and a water-reducing rate of ≥ 25%.

• (5) Preparation of simulated Seawater: In this experiment, artificially simulated seawater was used as mixing water for the tests. Three different concentrations of simulated seawater (0 times, 1 time, and 3 times) were designed. The preparation of simulated seawater with 1-time concentration was based on the analysis of salt content in Asian seawater as outlined in the “Standard Practice for the Preparation of Substitute Ocean Water”. The simulated seawater was prepared by adding four salts with higher content in seawater and a significant impact on the mechanical properties of cement-based materials. The salt content ratio for 1-time simulated seawater was 24.53 g/L NaCl, 5.2 g/L MgCl₂, 4.09 g/L Na₂SO₄, and 1.16 g/L CaCl₂. All mixing waters used in this experiment were prepared on-site. To ensure the uniformity of the test water for the 9 groups of experiments in each batch, tap water was used for 0-time simulated seawater. For 1-time and 3-time simulated seawater, containers with the same concentration of simulated seawater were prepared before the start of the experiment. Subsequently, the simulated seawater from these containers was sequentially transferred to small buckets for each experimental group, as shown in Fig. 2.

Figure 1

Test material. (a) cement, (b) sea sand, (c) the PVA fiber.

Figure 2

Artificial simulated seawater.

Test specimen fabrication and curing

Specimen fabrication

To ensure the uniformity of the experiment, it is ideal to complete the fabrication of all specimens at once^30,31,32. However, due to limitations such as the capacity of the mixer, site size, and the number of molds, it is not possible to pour all specimens at once. Therefore, the specimens were divided into four batches for fabrication. The first batch included specimens of strength grade C30 with two aging periods, 180 days and 90 days, fabricated simultaneously. The second batch included specimens of strength grade C50 with aging periods of 180 days and 90 days. The third batch consisted of C30 specimens with an aging period of 28 days, and the fourth batch consisted of C50 specimens with an aging period of 28 days. The pouring steps for each batch of specimens are as follows:

Preparation stage

Before pouring, calculate the quantities of various materials used in the experiment according to the mix ratio. Use an electronic scale to carefully measure the amounts of cement, sea sand, PVA fibers, composite admixtures, and simulated seawater. The measurements must be precise to ensure the accuracy of the experiment. When preparing the simulated seawater solution, first measure tap water in a large container, then add the solutes to the water. Stir with a handheld electric mixer until there are no solid substances in the water, and then measure the simulated seawater into small buckets for each experimental group, as shown in Fig. 3.

Figure 3

Test piece preparation.

After all materials are weighed, arrange them in the pouring order to prevent errors.

Prepare the required number of molds for specimens with dimensions of 100 mm × 100 mm × 400 mm. Use a brush to apply an appropriate amount of mold release agent to the inner sides of the molds. Prepare tools such as gray buckets, trowels, shovels, etc., needed during the experiment.

Pouring stage

Before pouring, moisten the mixer. Then, pour the materials into the mixer in the specified order. For each set of specimens, add the pre-measured sea sand, cement, composite admixtures to the mixer in sequence. After covering the mixer, dry mix for 3 min to ensure thorough mixing of the materials. Slowly and uniformly pour the pre-measured simulated seawater into the mixer, continue mixing for 3 min, and then evenly sieve the PVA fibers into the mixer. Mix for an additional 2 min. After mixing, pour the mixed material into a gray bucket, then into the prepared molds. For the 100 mm × 100 mm × 400 mm molds, pour in two steps, vibrating for 10 s on the vibrating table each time. Finally, place the fabricated specimens in sequence on a flat surface and use a trowel to smooth the surface.

The fabrication process is shown in Fig. 3.

Mold removal and maintenance

Remove the specimen after 3 days of pouring, and then place the prepared specimen in the standard curing room with 20 ± 2℃ and humidity above 95% to the 28d age, and then continue in the normal room temperature environment until the 28d age age, the specimen should be removed from the standard curing room 2 days in advance, and the surface should be dried for test as shown in Fig. 4.

Figure 4

Test piece after molding.

Test instruments and equipment

During the pouring process and testing process, the equipment used in this test are: 120 L forced concrete mixer for specimen production; FEI Nova NanoSEM 450 field emission scanning electron microscope for SEM test; electronic universal testing machine produced by MTS for bending toughness test, and the equipment are shown in Fig. 5.

Figure 5

Test instrument. (a) forced concrete mixer figure, (b) scanning electron microscope, (c) bending toughness test device.

Test methods and testing

This experiment employed the three-point bending method to conduct flexural toughness tests on specimens with dimensions of 100 mm × 100 mm × 400 mm. To measure the mid-span deflection of the specimen, displacement gauges with an accuracy of 0.001 mm were installed on both sides of the middle section. The mid-span deflection was determined by averaging the readings from the two displacement gauges. The test was conducted with displacement control at a rate of 0.05 mm/min and a data acquisition frequency of 1 Hz. The specific operational steps of the test are as follows:

• (1) To prevent specimen misalignment and accurately position the frame for fixing the displacement gauges, three equally spaced lines were marked on the specimen, with one each located 50 mm from both ends and another in the middle, dividing the specimen into three equal parts of 100 mm each.

• (2) The iron frame of the fixed displacement gauges was secured to the specimen along the marked lines. The frame was then accurately placed on the supports of the universal testing machine, with one support serving as a fixed hinge support and the other as a universal rolling hinge support. The rolling hinge support could roll and tilt freely forward and backward to minimize the impact on mid-span deflection during the loading process.

• (3) Displacement gauges were installed on both sides of the specimen, and the equipment was initiated using software control for data collection. The test continued until the test load dropped to approximately 15% of the peak load.

SEM test

Samples were taken from the tested rectangular specimens, glued onto conductive adhesive, coated with gold, and then photographed under an electron microscope. Specimens were selected at 180d age with 1.5% fiber dosing and seawater concentrations of 0×, 1×, and 3×. The single fiber and the part of fiber bonded with cementitious composite were sampled at both ends of the specimen where the PVA fiber was not stressed. The corrosion of the single fiber surface and the fiber-cementitious composite bond were observed at different seawater concentrations after 180d of age. The SEM results of the samples taken are shown in Fig. 6.

Figure 6

Scanning electron microscopy results. (a) C30, 0x, (b) C30, 1x, (c) C30, 3x, (d) C50, 0x, (e) C50, 1x, (f) C50, 3x, (g) C30, 0x, (h) C30, 1x, (i) C30, 3x, (j) C50, 0x, (k) C50, 1x, (l) C50, 3x.

By observing the fibers at the bond with the matrix, it was found that the specimens at both strength levels did not show any significant change in the bond between the PVA fibers and the matrix as the seawater concentration increased and that the interface at the junction of the matrix and the fibers was relatively smooth. By observing the electron microscope images of the single fiber, it was found that the single PVA fiber taken out after the test was wrapped by a layer of substance, which might be cementitious powder or salt substance in the mixing seawater solution, and this substance increased with the increase of seawater concentration, but the surface of the PVA fiber was relatively smooth, and at the age of 180d, with the increase of the seawater concentration, the PVA fiber was not found to have obvious corroded traces.

Evaluation of toughness for PVA fiber-reinforced seawater and seabeach sand cementitious composite materials

Toughness evaluation method

This study employed the American ASTM-C1018 standard, combined with the method of determining the initial crack point based on the tangent value, to analyze and evaluate the flexural toughness of the specimens.

Calculation results

According to CECS 13:2009 “Test Methods for Fiber Concrete”³³, the formula for the flexural cracking strength of PVA fiber-reinforced seawater and seabeach sand cementitious composite materials is:

$$ f_{cr} = \frac{{F_{cr} L}}{{bh^{2} }} $$

(1)

Including: f_cr—PVA fiber enhances the initial crack strength of seawater sea sand cement composite, MPa); F_cr—PVA fiber enhances the initial crack load of seawater sea sand cement composite, N; the span between the supports of L—test beam, mm; b—test beam width, mm; h—test beam height, mm.

The individual performance results for each group of tests based on the methodology in 2.4 and the calculations in 3.2 are shown in Fig. 7. The I₅, I₁₀, and I₂₀ refer to the flexural toughness indices of the test group. The bending toughness index is used to evaluate the bending toughness, and the fiber enhancement index is essentially the sum of the fiber characteristic values. The R5,10 and R10,20 refer to the residual strength factors of the test group. The American ASTM-C1018 standard evaluates the flexural toughness of concrete concerning the evaluation of ideal elastoplastic materials^34,35,36,37.

Figure 7

Individual performance results for each set of trials. (a) Initial deflection results for each group, (b) initial crack and initial crack load results for each group, (c) Bending toughness index and and residual strength coefficients for each group.

Based on Fig. 7, it can be observed that with the increase in fiber content, the initial cracking load of specimens in the C30 and C50 series of strength grades has increased to varying degrees. The initial cracking deflection of most specimens increases with the increase in fiber content, and the flexural toughness index of the specimens improves with the increase in fiber volume fraction.

• (1) With the continuous increase in fiber content, the peak bending load and the corresponding mid-span deflection of the specimens increase. The effect is more pronounced at 1.5% fiber content, with an approximate 2–4 times increase compared to the test group with 0% PVA fiber. For the C30 and C50 series, the increase in the mid-span deflection corresponding to the peak bending load for the 90d and 180d cure periods was limited for the 0.75% PVA fiber group compared to the 0% fiber group, and for the C50, 0.75% PVA fiber group with a 180d cure period, as there was no secondary peak load on the specimens. The increase in the mid-span deflection corresponding to the peak bending load was limited because the specimens did not experience a secondary peak load and entered the flexural-softening phase after cracking.

• (2) Except for specimens at 90 d curing period under C30 and 1 × seawater concentration, and specimens at 180 d curing period under C50 and 0 × seawater concentration, the initial cracking strength of the specimens increases with the increase in fiber content, demonstrating good crack resistance. The initial cracking strength of specimens under C30 and 0 × and 3 × seawater concentration, 0% fiber content, and under C50 and 3 × seawater concentration, 0.75% fiber content, showed less improvement as the curing period increased. The initial crack strength increased with age for all groups of specimens except for the C30, 0×, and 3×  seawater concentration, 0% fiber doping specimens, and the C50, 3 × seawater concentration, 0.75% fiber doping specimens.

• (3) For C30, 1 × seawater concentration, 0.75% fiber content specimens, the initial cracking strength stabilizes with the increase in curing period up to 180 d. The initial cracking strength of C30 specimens at 28 d curing period stabilizes after 28 d, while for C50 specimens, the initial cracking strength remains stable within 90 d but significantly increases at 180 d.

• (4) Under the same seawater concentration, as the PVA fiber content increases, the fle×ural toughness indices I₅, I₁₀, I₂₀, and residual strength factors R5,10 and R10,20 for all test groups increase, indicating that the effect of absorbing energy in specimens is more pronounced at high fiber content, resulting in better flexural toughness.

• (5) At the 28d curing period, when the strength grade of the specimen is lower, specimens with low fiber content exhibit the best flexural toughness at 1 × seawater concentration and the worst at 3 × seawater concentration. At high fiber content, specimens at 3 × seawater concentration show the best flexural toughness and the worst at 0 × seawater concentration. When the strength grade of the specimen is higher, specimens with low fiber content exhibit the best flexural toughness at 0 × seawater concentration, and specimens with high fiber content show little impact of seawater concentration on flexural toughness, with the worst performance observed at 1 × seawater concentration.

• (6) At the 90 d curing period, when the strength grade of the specimen is lower, specimens with low fiber content exhibit the best flexural toughness at 0 × seawater concentration, and at high fiber content, specimens at 3 × seawater concentration show the best flexural toughness. At 1 × seawater concentration, both fiber content levels exhibit poor flexural toughness. When the strength grade of the specimen is higher, specimens at both fiber content levels exhibit the best flexural toughness at 1 × seawater concentration, and at 3 × seawater concentration, flexural toughness is the worst.

• (7) At the 180 d curing period, when the strength grade of the specimen is lower, specimens at both fiber content levels show the best flexural toughness at 3 × seawater concentration. When the strength grade of the specimen is higher, specimens with low fiber content show the best flexural toughness at 1 × seawater concentration, while specimens with high fiber content exhibit the best flexural toughness at 0 × seawater concentration. Except for C30, 1.5% fiber content specimens, other specimens show a significant decrease in the residual strength factors R10,20 compared to R5,10, indicating a significant decrease in flexural toughness in the later stages of loading.

• (8) For C30, 0.75% fiber content test group, with the increase in curing period, the change in flexural toughness in the early loading stage is relatively small. However, in the later loading stage, flexural toughness decreases with the increase in curing period, indicating a continuous decline in the ability of specimens to absorb energy. For 1.5% fiber content, with the increase in curing period, the change in flexural toughness for all test groups is relatively small, and the residual strength factor can be maintained at a high value, indicating that the ability of specimens to absorb energy can remain stable during the 180 d curing period.

• (9) For C50, 0.75% fiber content test group, with the increase in curing period, the change in flexural toughness in the early loading stage is relatively small. However, in the later loading stage, flexural toughness decreases with the increase in curing period, indicating a continuous decline in the ability of specimens to absorb energy. For 1.5% fiber content, with the increase in curing period, the change in flexural toughness in the early loading stage is relatively small. Within the 90 d curing period, the ability of specimens to absorb energy remains relatively stable, and at 180 d curing period, the flexural toughness indices of all test groups show a certain decrease in the later loading stage, indicating a continuous decline in the ability of materials to absorb energy.

• (10) The flexural toughness of the specimens continued to deteriorate as the seawater concentration increased. The high seawater concentration has an enhancing effect on the toughness of the specimens in the later stages of the process. The toughness of the specimens at 180 d age was better in the pre-loading period. However, in the later stages, the bending toughness was better at the age of 28 d. The flexural toughness of specimens at higher seawater concentrations is better at longer ages. At low age, high seawater concentration has an early strengthening effect on the initial cracking strength of the specimens. At high age, high seawater concentration has a weakening effect on the initial crack strength of the specimens. As the age increases, there is not much difference in the ability of the specimens to absorb energy in the pre-loading period of the specimens. However, at the later stage, the capacity to absorb the energy of specimens at the age of 180d decreases. It decreases by around 24.4% compared to the lower age ones. The flexural toughness of the high-fiber doped specimens is similar to that of the low-fiber doped ones in the early stage after specimen cracking. However, in the later stages of loading, the performance of bending toughness will be very much better than that of the low-fiber doped specimens.

Beam strength and its deflection equations are established

The bending test shows that PVA fibers can increase the flexural strength of the material. Only a higher dosage of fibers can increase the deformation capacity of the material at a high age. Changes in seawater concentration and age have a small effect on the flexural strength of the material at the same fiber dosage. Therefore, according to the effects of age, fiber dosage, and seawater concentration on the material. The formulae for the flexural strength of PVA fiber-reinforced seawater sand cementitious composites and their corresponding deflections are established. The formula for calculating the flexural strength of a material is as follows:

$$ f_{m} = \alpha f_{m,0} + e^{{\frac{{\lambda_{a} }}{{\beta_{a} }} }} + e^{{\beta_{sc1} \lambda_{sc}^{2} + \beta_{sc2} \lambda_{sc} + \beta_{sc3} }} + e^{{\frac{{V_{f} }}{{\beta_{f} }}}} $$

(2)

Including: $f_{m}$,$f_{m,0}$—PVA fiber enhanced bending strength of seawater sea sand cement base composite and 28-d-old ordinary cement-based material, MPa; $\alpha$—bending strength of ordinary cement-based material; $\beta_{a}$, $\beta_{sc1}$, $\beta_{sc2}$, $\beta_{sc3}$, $\beta_{f}$—sample age, seawater concentration, PVA fiber mixing coefficient; $\lambda_{a}$, $\lambda_{sc}$,${ }V_{f}$—sample age, seawater concentration, and PVA fiber content.

The nonlinear curve fitting to the formula (2) from the test data indicates that:

$$ f_{m} = - 0.19753f_{m,0} + e^{{\frac{{\lambda_{a} }}{9.92362}}} + e^{{\left( {0.34883\lambda_{sc}^{2} - 1.17849\lambda_{sc} - 0.01467} \right)}} + e^{{\frac{{V_{f} }}{0.88284}}} $$

(3)

Calculate the bending strength of the material according to formula (3) and compare it with the test value, as shown in Fig. 8, The fitted correlation coefficient is given in R² = 0.92, with a good fit.

Figure 8

Comparison of bending strength test value and calculated values.

The deflection formula corresponding to the flexural strength is calculated as follows:

$$ \delta_{m} = \alpha f_{m,0} + \beta_{a1} \lambda_{a}^{2} + \beta_{a2} \lambda_{a} + \beta_{sc} \lambda_{sc} + \beta_{f1} V_{f}^{2} + \left( {\beta_{f2} + \beta_{f3} \lambda_{a} + \beta_{f4} \lambda_{sc} } \right)V_{f} + \gamma $$

(4)

Including: $\delta_{m}$—deflection corresponding to flexural strength, mm; fm, bending strength of 0–28d instar ordinary cement based material, MPa, flexural strength of $\alpha$ -28d instar ordinary cement based materials; $\beta_{a1}$, $\beta_{a1}$, $\beta_{sc}$, $\beta_{f1}$, $\beta_{f2}$, $\beta_{f3}$, $\beta_{f4}$, $\gamma$—test age, seawater concentration, PVA fiber content influence coefficient; $\lambda_{a}$, $\lambda_{sc}$, $V_{f}$—Test age, seawater concentration, PVA fiber content.

Linear regression of formula (4) from the test data yielded:

$$ \delta_{m} = - 0.043f_{m,0} + 0.017\lambda_{a}^{2} - 0.121\lambda_{a} + 0.004\lambda_{sc} + 0.254V_{f}^{2} + \left( {0.258 - 0.071\lambda_{a} - 0.04\lambda_{sc} } \right)V_{f} + 0.267 $$

(5)

The deflection corresponding to the material flexural strength is calculated according to formula (5) and compared with the test value, as shown in Fig. 9, and the fitted correlation coefficient is R² = 0.92, with a good fit.

Figure 9

Comparison of deflection test values for flexural strength.

Bending toughness index model establishment

In the four-point bending test of material, bending toughness is an important index to evaluate PVA fiber reinforced sea sand cement matrix composite, so it is necessary to conduct quantitative analysis and calculation, It is found that the tensile strength and bending toughness of PE fiber and PVA fiber cement matrix composite are linearly related with the characteristic fiber characteristic value^38,39;The mathematical model of bending toughness index and fiber enhancement index is proposed for the composite materials. Through test verification, it is generally applicable to the equivalent bending strength and bending toughness index, as shown in formula (6) and formula (7) ⁴⁰:

$$ T = T_{0} \left( {1 + \alpha S^{2} + \beta S} \right) $$

(6)

$$ S = \mathop \sum \limits_{i}^{n} \delta_{i} \rho_{i} \frac{{l_{i} }}{{d_{i} }}\left( {\frac{{f_{i} }}{{f_{s} }}} \right)^{m} $$

(7)

Including: bending toughness evaluation index of $T$ and $T_{0}$—fiber reinforced cement based composite and plain cement based material; S—sum of various fiber enhancement indexes; $\alpha$,$\beta$—Impact coefficient of fiber enhancement index; $\delta_{i}$ mechanical anchorage coefficient of—fiber and matrix, PVA fiber: 0.1; $\rho_{i}$_, $l_{i}$_, $d_{i}$—— Fiber content, length and diameter;$f_{i}$, $f_{s}$—tensile strength of fiber and steel fiber; m—— and fiber category, 0.5 for non-metallic fiber and 1 for steel fiber.

The bending toughness index is used to evaluate the bending toughness, and the fiber enhancement index is essentially the sum of the fiber characteristic values. This study is a single fiber reinforced seawater sea sand cement matrix composite material, so formula (6) and formula (7) are appropriately changed to obtain:

$$ I_{5} = I_{5,0} \left( {1 + \alpha S^{2} + \beta S} \right) $$

(8)

$$ I_{10} = I_{10,0} \left( {1 + \alpha S^{2} + \beta S} \right) $$

(9)

$$ I_{20} = I_{20,0} \left( {1 + \alpha S^{2} + \beta S} \right) $$

(10)

$$ S = \delta_{i} \rho_{i} \frac{{l_{i} }}{{d_{i} }} $$

(11)

Among them, the bending toughness evaluation index of cement base material when $I_{5}$, $I_{10}$ and $I_{20}$—fiber mixture is 0%, known by ASTM C1018, and the three values are 1.0.

Nonlinear curve fitting of formula (8), formula (9) and formula (10) by test data, and the results of bending toughness index of each group are shown in Table 3. The number of this table is different from the previous above, the number according to the strength grade-test age-seawater concentration format, such as number 3-1-0 means material strength grade C30, test age 28 d, 0 times seawater concentration test, number 5-1-0 means material strength grade C50, age 28 d, 0 times seawater concentration test, other test number and so on.

Table 3 Formula fitting of the bending toughness index.

As found in Table 3, the fitted correlation coefficient R² = 0.99, the fitting effect of the model is good, can accurately predict the PVA fiber enhanced sea sand cement composite bending toughness, observation parameters, can be found almost all negative, that the bending toughness index curve opening down, can be in the PVA fiber under the optimal fiber content, in order to achieve the optimal bending toughness. The parameters α and β are related to the test age, the seawater concentration and the material compressive strength in this test.

$$ \begin{aligned} \alpha = & 0.00{4} + 0.00{18}\lambda_{{\text{a}}} - 0.00{63}\lambda_{{{\text{sc}}}} - 0.00{12}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} + {6}.{43} \times {1}0^{ - 5} \lambda_{{\text{a}}}^{{2}} + {9}.{68} \times {1}0^{ - 5} \lambda_{{{\text{sc}}}}^{2} \\ & \; - {5}.{13} \times {1}0^{ - 5} \lambda_{{\text{a}}}^{{2}} \lambda_{{{\text{sc}}}}^{{2}} + \left( { - {2}.{94} \times {1}0^{ - 4} - {4}.{89} \times {1}0^{ - 5} \lambda_{{\text{a}}} + {3}.0{1} \times {1}0^{4} \lambda_{{{\text{sc}}}} + {2}.{79} \times {1}0^{ - 5} \lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}} \\ & \; + \left( {{3}.{87} \times {1}0^{ - 6} + {1}.{19} \times {1}0^{ - 7} \lambda_{{\text{a}}} - {3}.{64} \times {1}0^{ - 6} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}}^{{2}} \\ \end{aligned} $$

(12)

$$ \begin{aligned} \beta = & - 0.{18} - 0.{126}\lambda_{{\text{a}}} + 0.{4}0{8}\lambda_{{{\text{sc}}}} + 0.0{79}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} - 0.00{4}\lambda_{{\text{a}}}^{{2}} - 00{6}\lambda_{{{\text{sc}}}}^{{2}} \\ & \; + 0.00{3}\lambda_{{\text{a}}}^{{2}} \lambda_{{{\text{sc}}}} + \left( {0.0{17} + 0.00{4}\lambda_{{\text{a}}} - 0.0{19}\lambda_{{{\text{sc}}}} - 0.00{2}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{\text{c u}}} \\ & \; + \left( { - {2}.{24} \times {1}0^{{ - {4}}} - {1}.{11} \times {1}0^{ - 5} \lambda_{{\text{a}}} + {2}.{32} \times {1}0^{ - 4} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}}^{2} \\ \end{aligned} $$

(13)

$$ \begin{aligned} {\text{a}} = & 0.00{7} + 0.00{65}\lambda_{{\text{a}}} - 0.0{16}\lambda_{{{\text{sc}}}} - 0.00{3}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} + {8}.{48} \times {1}0^{ - 5} \lambda_{{\text{a}}}^{2} + {2}.{15} \times {1}0^{ - 4} \lambda_{{{\text{sc}}}}^{2} - {1}.{13} \times {1}0^{ - 4} \lambda_{{\text{a}}}^{2} \lambda_{{{\text{sc}}}}^{{2}} \\ & \; + \left( {{5}.{16} \times {1}0^{ - 4} - {1}.{82} \times {1}0^{ - 4} \lambda_{{\text{a}}} + {7}.{6} \times {1}0^{ - 4} \lambda_{{{\text{sc}}}} + {7}.0{4} \times {1}0^{ - 5} \lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}} \\ & \; + \left( {{6}.{45} \times {1}0^{ - 6} + {1}.0{1} \times {1}0^{{ - {7}}} \lambda_{{\text{a}}} - {8}.{97} \times {1}0^{ - 6} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}} \\ \end{aligned} $$

(14)

$$ \begin{aligned} \beta = & - 0.{21} - 0.{455}\lambda_{{\text{a}}} + {1}.0{37}\lambda_{{{\text{sc}}}} + 0.{219}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} - 0.00{5}\lambda_{{\text{a}}}^{2} - 0.0{14}\lambda_{{{\text{sc}}}}^{2} \\ & \; + 0.00{7}\lambda_{{\text{a}}}^{2} \lambda_{{\text{s c}}} + \left( {0.0{28} + 0.0{13}\lambda_{{\text{a}}} - 0.0{49}\lambda_{{{\text{sc}}}} - 0.00{5}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}} \\ & \; + \left( { - {3}.{44} \times {1}0^{{ - {4}}} - {7}.{46} \times {1}0^{ - 5} \lambda_{{\text{a}}} + {5}.{72} \times {1}0^{ - 4} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}}^{2} \\ \end{aligned} $$

(15)

$$ \begin{aligned} \alpha = & 0.00{1} + 0.0{14}\lambda_{{\text{a}}} - 0.0{31}\lambda_{{{\text{sc}}}} - 0.00{5}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} + {1}.{66} \times {1}0^{4} \lambda_{{\text{a}}}^{{2}} + {8}.{25} \times {1}0^{ - 4} \lambda_{{{\text{sc}}}}^{2} \\ & \; - {1}.{59} \times {1}0^{ - 4} \lambda_{{\text{a}}} {2}\lambda_{{{\text{sc}}}} + \left( { - {3}.{81} \times {1}0^{4} - {4}.{14} \times {1}0^{ - 4} \lambda_{{\text{a}}} + 0.00{1}\lambda_{{{\text{sc}}}} + {1}.{1} \times {1}0^{ - 4} \lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}} \\ & \; + \left( {{5}.{73} \times {1}0^{ - 6} + {2}.{73} \times {1}0^{ - 6} \lambda_{{\text{a}}} - {1}.{58} \times {1}0^{ - 5} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{{\text{cu}}}}^{2} \\ \end{aligned} $$

(16)

$$ \begin{aligned} \beta = & {1}.{17} - {1}.{254}\lambda_{{\text{a}}} + {1}.{789}\lambda_{{{\text{sc}}}} + 0.{3}0{5}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} - 0.00{8}\lambda_{{\text{a}}}^{{2}} - 0.0{52}\lambda_{{{\text{sc}}}}^{{2}} \\ & \; + 0.00{9}\lambda_{{\text{a}}}^{{2}} \lambda_{{{\text{sc}}}} + \left( { - 0.0{17} + 0.0{41}\lambda_{{\text{a}}} - 0.0{78}\lambda_{{{\text{sc}}}} - 0.00{6}\lambda_{{\text{a}}} \lambda_{{{\text{sc}}}} } \right){\text{f}}_{{\text{c u}}} \\ & \; + \left( {{1}.{32} \times {1}0^{ - 4} - {3}.{28} \times {1}0^{ - 4} \lambda_{{\text{a}}} + {8}.{62} \times {1}0^{ - 4} \lambda_{{\text{s c}}} } \right){\text{f}}_{{{\text{cu}}}}^{2} \\ \end{aligned} $$

(17)

According to formula (12) -formula (17), the value of the two parameters is calculated and compared with the fitting value. As shown in Fig. 10, the fitting correlation coefficient is high, and the fitting effect is good.

Figure 10

Comparison of fitted and calculated values for the parameters α and β. (a) I₅ The parameter α comparison, (b) I₅ of the parameter β for comparison, (c) I₁₀ The parameter α comparison, (d) I₁₀ of the parameter β for comparison, (e) I₂₀ The parameter α comparison, (f) I₂₀ of the parameter β for comparison.

Comparison between test value and calculated value of index of bending toughness

The bending toughness index I is calculated according to formula (12) -formula (17) I₅, I₁₀, I₂₀ And compared with the experimental values, as shown in Fig. 11.

Figure 11

bending toughness index test value and calculated value.

Therefore, Fig. 11 can find that the model established in the previous section can calculate the bending toughness index of this test, with small error, which shows that this model can better predict the bending toughness of PVA fiber-enhanced composite of sea sand cement matrix at different ages and seawater concentration.

Conclusion

In this study, the mechanical properties of PVA fiber-reinforced seawater-sea sand-fiber cementitious composites were investigated. The effects of PVA fiber content, seawater concentration and test age on the initial cracking properties and flexural toughness of the materials with strength grades C30 and C50 were investigated, and the flexural toughness indices of the beams were calculated. These specific concentrations were chosen to form a comprehensive spectrum of PVA fiber content, covering scenarios ranging from no added fibers (0%) to low concentration (0.75%) and a moderate increase (1.5%). The specific conclusions are as follows:

• 1. The addition of fibers increases the peak flexural load, with the most significant improvement at 1.5% fiber content. In the early age, due to the “bridging” effect of fibers, there is some strain hardening in the high fiber content specimens, and the lower the seawater concentration, the more pronounced the strain hardening effect. In low fiber content specimens, a bending-strengthening phenomenon is observed. With age, the strain-hardening effect of high-fiber-content specimens weakens, and low-fiber-content specimens transform into bending-softening. At the same age, with increasing seawater concentration, this characteristic persists, and the specimens exhibit faster deflection under load, indicating a significant influence of harmful ions in seawater and sea sand on the “bridging” effect of fibers.

• 2. The corrosion of single fiber surfaces and the bonded portion of fibers to the cement matrix at different seawater concentrations were observed by SEM and it was found that the fibers were less affected by seawater corrosion and that a substance, which may be matrix powder or salts, was adhered to the fiber surfaces. This substance increases with seawater concentration and it can be inferred that the bonding of the fibers to the matrix becomes stronger with increasing seawater concentration and age. When the specimen is cracked, most of the fibers at low seawater concentration are pulled out or slipped, and a few are pulled out, so there is a strain hardening or bending-strengthening phenomenon. When the age and seawater concentration are increasing, most of the PVA fibers in the matrix are tensile damaged and the initial cracking load of the specimen is higher. Therefore, the strain-strengthening phenomenon is weakened at high fiber dosage and the bending-strengthening phenomenon disappears at low fiber dosage.

• 3. By calculating and analyzing the bending toughness index and residual strength factors of the bending tests, it was found that fiber incorporation would improve the bending toughness of the material and that the bending toughness would increase more at a later stage under high fiber incorporation. At the same seawater concentration, the ability of the material to absorb energy increased by a minimum of 33% and a maximum of 109% compared to specimens at low fiber dosage.

• 4. When the material has lower strength grades and 0.75% fiber content, high seawater concentration reduces flexural toughness in the early stages, but with age, flexural toughness improves. For specimens with 1.5% fiber content, the material’s energy absorption capacity remains stable with increasing seawater concentration, maintaining good flexural toughness. When the material has higher strength grades, both 0.75% and 1.5% fiber content reduce flexural toughness in the later stages at high seawater concentrations, indicating a decrease in energy absorption capacity after material cracking. Under the same seawater concentration, at 180 days, specimens with low fiber content exhibit a decrease in flexural toughness in the later stages, but this decrease weakens with higher seawater concentration. Specimens with high fiber content in low-strength grade specimens show minimal changes in flexural toughness with age, while high-strength grade specimens exhibit a decrease in flexural toughness in the later stages.

• 5. In a seawater-sea sand environment, for lower strength grades, specimens with 1.5% fiber content exhibit better energy absorption capacity and durability, while for higher strength grades, specimens with 1.5% fiber content have poorer durability in the later stages after material cracking. High-strength cement-based composite materials require higher fiber content to improve flexural toughness and durability.

• 6. A formula was fitted to predict the compressive strength of cubic specimens of PVA fiber-reinforced seawater-sea sand-fiber cement-based composite materials under different ages, seawater concentrations, and PVA fiber content. And to predict the flexural strength and deflection formula of PVA fiber reinforced seawater sea sand cement matrix composites at different ages, different seawater concentration and different PVA fiber dosage:

  $$ f_{m} = - 0.19753f_{m,0} + e^{{\frac{{\lambda_{a} }}{9.92362}}} + e^{{\left( {0.34883\lambda_{sc}^{2} - 1.17849\lambda_{sc} - 0.01467} \right)}} + e^{{\frac{{V_{f} }}{0.88284}}} $$
  $$ \delta_{m} = - 0.043f_{m,0} + 0.017\lambda_{a}^{2} - 0.121\lambda_{a} + 0.004\lambda_{sc} + 0.254V_{f}^{2} + \left( {0.258 - 0.071\lambda_{a} - 0.04\lambda_{sc} } \right)V_{f} + 0.267 $$

The Complex Factors shaping the mechanical performance of PVA fiber-reinforced seawater-sea sand-fiber cement-based composites. The deliberate inclusion of varied PVA content 0%, 0.75%, and 1.5%unveiled nuanced patterns, showcasing heightened flexural load and improved toughness with increased fiber content. Early-age strain hardening transformed into bending-softening over time, influenced by seawater concentration. Fiber corrosion proved minor, and durability exhibited intricate dependencies on strength grades and fiber concentrations. Notably, 1.5% fiber content demonstrated superior energy absorption and durability in lower strength grades, while higher strength grades required elevated fiber content for enhanced toughness and durability. The derived predictive formulas contribute valuable tools for comprehensively assessing concrete behavior in diverse conditions.

Discussion

The study’s findings reveal valuable insights into the mechanical behavior of PVA fiber-reinforced seawater-sea sand-fiber cement-based composites. This paper is a conservative study of PVA fiber doping. Due to the experimental variables, only one type of fiber was investigated, and the effects of higher dosages and other types of fibers in the seawater sea sand environment are not yet known. The authors believe that the fiber dosage can be further increased to investigate the enhancement of flexural toughness and cracking improvement of seawater sand cementitious composites at higher fiber dosages. It is also possible to study the effect of mixing different types of corrosion-resistant fibers on the flexural toughness of specimens.