Experimental Study on the Impermeability and Micromechanisms of Basalt Fiber-Reinforced Soil-Cement in Marine Environments

Abstract

In coastal areas, structures such as cement-soil dams are often eroded by seawater, so it is significant to study how to improve the impermeability of cement-soil. Basalt fiber with a strong tensile property, good stability and a high-performance price ratio was selected as the additive to study the influence of the basalt fiber content on the permeability of soil-cement. The permeability test and the chloride ion permeability test were used to evaluate the best mixing amount. The results of the permeability test showed that, although the permeability coefficient of soil-cement decreased with the increase in the basalt fiber content, the decreased rate of the permeability coefficient showed a slowing trend. The results of the chloride ion permeability test indicated that the chloride ion-related impermeability of soil-cement was enhanced with the increase in the basalt content, which was confirmed by the consistent findings of the contrast permeability test. The comprehensive analysis shows that the optimal content ratio of the basalt fiber was 1.5%. Furthermore, a SEM analysis established that the addition of the basalt fiber reduced the soil-cement porosity, improved the structural compactness and formed a more stable whole. This study could serve as a valuable reference for soil-cement used in projects with impermeability requirements.

1. Introduction

With the development of urbanization, land use is becoming increasingly tense and, with the development of rivers and oceans, structures often need to be built on soft soil. The soft soil has a low strength and large deformation, so it needs to be improved. Since adding cement to soft soil can improve its strength, reduce deformation, reduce permeability and obtain a good cost-effectiveness, soil-cement is widely used as a subgrade reinforcement for retaining wall waterproofing in dam and port constructions, etc. [1,2,3]. Cao et al. [4] added ultra-fine cement (UFC) to peat soil, which significantly improved the unconfined compressive strength of soil-cement, but the increase in the compressive strength was not obvious when high amounts of the UFC was added. Wu et al. [5] found that the elastic modulus had a linear relationship with the unconfined compressive strength. Wei et al. [6] found that soil-cement reinforcement can effectively improve the pile soil resistance, and thus, improve the horizontal bearing performance of the pile foundation. Wang et al. [7] studied the engineering characteristics of Xiamen marine silt cement-soil and found that, when the cement mixing ratio was less than 18%, the compressive strength and the cohesion of cement-soil increased approximately linearly, and when the cement mixing ratio was greater than 18%, the growth rate of the strength index decreased. Soil-cement structures are located below ground level and are exposed to the eroding effects of groundwater and seawater year-round. Therefore, in addition to the strength, research on the impermeability of cement-soil is also very important, which is an essential index to evaluate the engineering performance of cement-soil [8,9,10].

With the growth of related structural engineering requirements, many scholars began to study the permeability of soil-cement. The research on soil-cement permeability mainly focuses on the soil quality, cement content, curing age and environmental conditions. The research on the influence of the soil quality on the permeability of cement-soil includes Toshimitsu et al. [11], who examined the relationship between the soil-cement permeability and the soil structure compactness through microscopic testing, The results showed that the smaller the pore size and the better the compactness of the soil, the better the impermeability of cement-soil. Lu et al. [12] investigated the permeability of soil-cement containing silty sand and silty clay. The results showed that, with the increase in the cement content and curing age, the permeability coefficient of the silty sand decreased faster than that of the silty clay. The research on the influence of the cement content on the permeability of cement-soil includes Hong and Gong [13], who performed permeability tests to establish that the permeability coefficient of soil-cement decreased with the increase in the cement content and curing age. It was considered that the optimum cement mixing ratio for engineering application is 10%. Zhang et al. [14] determined the relationship between the soil-cement permeability coefficient and the cement content using a regression analysis. Zhu et al. [15] conducted an indoor permeability test on cement-soil with different cement contents and ages. The results showed that the permeability coefficient decreased with the increase in the cement content and age. Tao et al. [16] studied the influence of the cement content on the micropore distribution and the permeability of cement-soil. It was found that the permeability coefficient of cement-soil decreased with the increase in the cement content, which decreased sharply in the range of low cement content (4%~12%) and slowly in the range of high cement content (15%~25%). The research on the influence of age on the permeability of cement-soil includes Zhang et al. [17], who obtained the quantitative relationship between the permeability coefficient of cement-soil and the curing age using a regression analysis. Jiao et al. [18] studied the influence of age on the permeability of cement-soil. The results showed that the permeability coefficient of cement-soil gradually decreased with the increase in age. After the age was more than 28 days, the reduction rate of the permeability coefficient with age decreased. A prediction formula for the long-term permeability coefficient of cement-soil was proposed. The research on the influence of environmental conditions on the permeability of cement-soil includes Yang et al. [19], who considered the influence of a seawater environment on the permeability of cement-soil. The research showed that the permeability coefficient of cement-soil increased with the increase in age under the curing condition in seawater, which also showed that the cement-soil deteriorated in the seawater environment. Chen et al. [20] conducted a permeability test on cement-soil under a sewage environment. The test results showed that the permeability coefficient of cement-soil decreased gradually before the age of 60 days. After the age of 60 days, due to the erosion of cement-soil, the permeability coefficient increased gradually with the increase in age. The research on the influence of additives on the permeability of cement-soil includes Pang et al. [21], who studied the effect of adding fly ash and lime on the impermeability of cement-soil. The results showed that the effect of adding fly ash alone on the impermeability of cement-soil was small, but the effect was obvious after adding lime. In view of the soil properties studied, a reasonable ratio relationship was proposed. Chen et al. [22] conducted a permeability test on cement composite soil. The results showed that the permeability coefficient of cement composite soil decreased gradually with the increase in the cement content, bentonite content and fly ash content. The cement content had the greatest influence on the permeability coefficient of cement-soil, followed by the fly ash and bentonite. Mo et al. [23] studied the influence of the cement content, desulfurized gypsum content, fly ash content and water–cement ratio on the permeability of cement-soil. The results showed that the influence degree of the four factors on the permeability of the soil was the water–cement ratio > fly ash content > cement content > desulfurized gypsum content. Cui et al. [24] studied the influence of the fly ash content on the permeability of cement-soil. The results showed that the permeability coefficient of fly ash cement-soil with a cement content of 60% was higher than that of pure cement-soil before curing for 14 days. However, it became lower after curing for 28 days. Hu et al. [25] mixed different amounts of MgO into the fiber cement-soil and carried out a permeability test. The test results showed that there were some defects when the content of MgO was too high or too low, and the optimal content was 1%. Chen et al. [26] studied the influence of ferronickel slag powder on the permeability of cement-soil. The results showed that the addition of ferronickel slag powder can improve the impermeability of cement-soil in a clean water environment and a seawater environment, especially in the latter. When the content of ferronickel slag powder was more than 20%, the influence of the increase in the ferronickel slag powder content on the impermeability of cemented-soil gradually decreased.

It can be seen from the above studies that some scholars improved the impermeability of soil-cement by enhancing the compactness of soil-cement, increasing the cement content and adding various additives. From both an economic and engineering perspective, incorporating various admixtures has proven more cost-effective than increasing the cement content and has become a more obvious and effective strategy to improve the performance of soil-cement. The most important point arising from the above investigations is that appropriate admixtures can be selected to meet variable engineering conditions to improve the performance of specific features of the soil-cement materials in order to satisfy the design requirements of different projects [27,28,29]. Basalt fiber is widely used in engineering projects involving concrete due to its proven chemical stability, excellent temperature resilience, high-tensile properties, good dispersion capability and highly cost-effective performance. Despite these key attributes, its application in soil-cement to date has not been extensive.

At present, the research on basalt fiber applied to soil-cement mainly focuses on strength. The research on the compressive strength is as follows. Niu et al. [30] found that it was not suitable to use long fibers for reinforcement. On the one hand, it is easy to form cluster phenomenon, and on the other hand, it is easy to hang on the mixer. Ma et al. [31] found that the appropriate content of basalt fiber and sand can jointly promote the improvement of the soil-cement compressive strength. Chen [32] found that the addition of basalt fiber can significantly improve the compressive strength of the cement-soil samples at different ages, but with the increase in the fiber content, its reinforcement effect gradually weakened. The research on tensile strength includes Zhang et al. [33], who found that the order of influence of the fiber variables on the tensile properties was the length, content and diameter. A 9mm length of basalt fiber with the content of 1.5% are the optimal single mixing parameters of basalt fiber. Shen et al. [34] found that a fiber length of 9 mm + 12 mm mixed in a 3:1 ratio had the best tensile and residual strength. Chen [35] found that a too high fiber content could not produce an obvious strength increase, but the addition of basalt fiber enhanced the plasticity of cement-soil. The research on the shear strength includes Chen [36], who found that the incorporation of basalt fiber significantly improved the cohesion of cement-soil, significantly improved the strength of cement-soil and increased its plastic deformation. Chen et al. [37] found through a triaxial test that the stress–strain curve can be roughly divided into three stages: the linear elastic stage, plastic yield stage and peak softening stage. The research on the splitting strength includes Cao et al. [38], who found that excessive basalt fibers had a negative effect. When the content of basalt fiber was 1.5%, its impact splitting strength and absorption energy reach the maximum value. The research on fatigue performance includes Chen et al. [39], who found that basalt fiber can improve the fatigue properties of cement-soil mainly by limiting the formation and expansion of fatigue cracks, thus improving the fatigue life of cement-soil.

Through the above research, it was found that the diameter, length and content of basalt fiber affected the various strengths of cement-soil. When adding basalt fiber, it should not be too long to avoid agglomeration and twining. Although basalt fiber can enhance the strength and improve the plastic deformation ability, a too high basalt fiber content cannot significantly improve the strength and will increase the project cost.

With the development of marine engineering, it is urgent to solve the problem of the impermeability of cement-soil. However, there are few studies on the impermeability of basalt fiber cement-soil, and the impermeability of cement-soil in marine environment has not been given due attention. Therefore, it is necessary to study the impermeability of basalt fiber-enhanced soil-cement under the sea. In the study of the basalt fiber soil-cement strength, it was found that the strength increased continuously with the increase in the basalt fiber content, but an excessive content caused poor workability and the improvement effect was not obvious. Therefore, this paper mainly studies the influence of the basalt content on the permeability. Through the permeability and chloride ion permeability tests, the optimal basalt content was obtained. The influencing mechanism of the basalt content on the impermeability of soil-cement was microscopically scrutinized through an SEM examination of the microstructure of the compound.

The chloride ion permeability test in this test method is often used in permeability tests of high- and medium-strength concrete. However, few studies were identified that applied chloride ion permeability testing to establish the permeability of soil-cement. As soil-cement is generally located in environments rich in groundwater and seawater, it is essential to consider the eroding effects of chloride ions on this material. Therefore, this study carried out a chloride ion permeability test based on the permeability test and obtained the optimal content of basalt fiber cement-soil through a comprehensive analysis of the results of the two tests to improve the credibility. This study could, in the present authors’ view, serve as a valuable reference for soil-cement usage in projects with impermeability requirements.

2. Basalt Fiber Permeability Testing

2.1. Experimental Materials

The soil samples were taken from a riverbed under a bridge in Fuzhou, a coastal city in Fujian Province, China, at a depth of 8.4–13.2 m. The soil samples are shown in Figure 1 and the main indicators are shown in

Table 1. Main indicators of the soil samples.

Indicators	Natural Moisture Content	Unit Weight	Natural Void Ratio	Plasticity Index	Liquidity Index	Compressibility Factor	Compressive Modulus	Cohesion	Internal Friction Angle
	ω	γ	e	IP	IL	a0.1–0.2	Es0.1–0.2	C	ϕ
	(%)	(kN/m3)		(%)		(MPa−1)	(MPa)	(kPa)	(°)
The numerical	57.5	15.52	1.72	21.72	1.43	1.58	1.72	9.69	6.69

.

The soil samples were taken from depths reached by subway construction and, therefore, represented the characteristics of the coastal cities. The plasticity index Ip of the soil mass was 21.72, the natural water content was 57.5% and the porosity was 1.72. According to the Standard for Geotechnical Engineering Investigation (DBJ/T 13-84-2022) [40], the soil sample was judged as silt, and the fluidity index was 1.43. Therefore, the soil mass was judged as fluid plastic. The cohesiveness and internal friction angle of the soil mass were relatively small and the strength was naturally low. Therefore, the soil mass must be improved in roadbed reinforcement and retaining wall waterproof engineering.

Cement and basalt fibers were added to the soil in the test design. P.O 42.5 ordinary Portland cement was used. The main chemical element, CaO, accounted for 61.56%, SiO₂ accounted for 22.03% and SO₃ accounted for 2.63% of the overall material. Basalt fiber is a continuous fiber composed of molten basalt stone through high-speed drawing. It has stable chemical properties, a high elastic modulus and tensile strength, a high cost performance and good dispersion and compatibility. This test used chopped fiber, as shown in Figure 2, and its main physical and mechanical indexes are shown in

Table 2. Main indicators of basalt fibers.

Material	Monofilament Diameter	Length	Tensile Strength	Elastic Modulus	Fracture Elongation	Density
	(μm)	(mm)	(MPa)	(GPa)	(%)	(kg/m3)
Basalt fiber	17	24	4150–4800	93–110	3.1	2650

.

2.2. Test Plan

2.2.1. Permeability Test Plan

The permeability test is a common test method used to determine the permeability of cement-soil. The most important variable in this basalt fiber permeability experiment was the amount of basalt fiber, four content ratios of which were set on the basis of the available data, namely, 0%, 0.5%, 1.0% and 1.5%. The water content of cement-soil was set according to the natural water content of the undisturbed soil, i.e., 57.5%. According to the requirements of the Technical Standard for Building and Municipal Foundation (DBJT13-07-2021), the content of the cement for soil reinforcement should not be less than 15% [41]. The content of the cement used in this test was 15%, and 0.5% of the water–cement content. Four groups of specimens referenced as A, B, C and D were distinguished according to their different basalt content ratios. A second variable for this test was the curing age. Four curing periods of 7 days, 28 days, 60 days and 90 days were set. Three samples were produced for each mixing ratio and the results for each group were averaged.

Table 3. Mix design of the basalt fiber soil-cement permeability test.

Number	Moisture Content	Cement Mixing Ratio	Water/Cement Ratio	Basalt Fiber Blending Ratio	Curing Age/Number of Samples (unit)
	(%)	(%)		(%)	7d	28d	60d	90d
A	57.5	15	0.5	0	3	3	3	3
B				0.5	3	3	3	3
C				1.0	3	3	3	3
D				1.5	3	3	3	3

illustrates the specific test design.

2.2.2. Chloride Ion Permeability Test Plan

It was not sufficient to evaluate the optimal content of basalt fibers using only the results of the permeability tests because, when soil-cement is used as a waterproof and impermeable barrier in engineering, it is usually in groundwater or seawater environments. Long-term exposure to such environments inevitably leads to corrosion, which affects the impermeability of soil-cement. Among other corrosive elements, chloride ions have been shown to have the greatest impact on soil-cement impermeability as they can invade the interior of the material through osmosis, capillary action and electrochemical transfer, causing changes in its properties and seriously affecting its impermeability and reliable safety. Therefore, it was imperative to investigate the permeability of basalt fiber-reinforced soil-cement from chloride ions.

Since the hydration reaction of the cement needed to be completed in 90 days in order to study the influence of the basalt fiber content on the permeability of soil-cement, only the samples at 90 days were selected for this test. The variables set in this test were the same as those set in the permeability test.

2.2.3. Scanning Electron Microscope (SEM) Test Plan

In order to explain the mechanism of the influence of the basalt fiber content on the soil-cement permeability, scanning electron microscopy (SEM, FEI Company, Hillsboro, OR, USA) tests with a 0%, 0.5%, 1.0% and 1.5% basalt content were carried out. Using scanning electron microscopy to understand the microstructure of basalt fiber-reinforced soil-cement could be conducive to identifying reasonable explanations for some phenomena and laws displayed in the aforementioned permeability tests of basalt fiber-reinforced soil-cement.

The variables set in this test were the same as those set in the chloride ion permeability test.

2.3. Testing Procedure

2.3.1. Permeability Test Procedure

The permeability coefficient of cement-soil is relatively low, generally reaching the order of 10⁻⁸~10⁻⁹ cm/s. The permeability coefficient cannot be accurately measured using a conventional penetration test device. According to the requirements of “JGJT 233-2011 Design Regulations for Soil-cement Mix Ratio” [42], a TJSS-25 soil-cement permeability testing device was used due to its simple testing principle and operational convenience. In particular, its automatic pressurization can well control the test process and effectively improves the testing accuracy. The permeability test device is shown in Figure 3.

(1)

Sample Preparation and Curing

The test model was a round table with an upper inner diameter of 70 mm, a lower inner diameter of 80 mm and a height of 30 mm. The prepared samples were placed in a room for curing to the various corresponding durations. The samples during curing are shown in Figure 4.

(2)

Sealing of the Specimens

Strict requirements were applied to the sealing of the soil-cement samples in this test. The naturally air-dried cement-soil samples were placed in liquid paraffin and rolled evenly. After the paraffin on the surface was cooled, the samples were rolled again. This process was repeated three to four times. The steel penetration mold was placed into the drying oven and dried for at least half an hour. The mold was taken out of the oven and a layer of Vaseline was painted on its inner surface. The samples were placed in the center of the mold. The liquid paraffin was poured into the gaps between the sample and the mold. After the paraffin cooled, more liquid paraffin was poured to make the spaces between the sample and the mold denser.

(3)

Water Output Calculation

An infiltration device (TJSS-25, Zhongya Instrument Co., LTD., Cangzhou, Hebei, China) was used to measure the permeability coefficient of the samples as follows. Its pressure was adjusted to 0.2 MPa to ensure that the soil-cement penetration test was conducted under a constant water pressure while taking into account the low strength of the soil-cement specimens. Once their surface had begun to allow a relatively uniform water seepage, the water output Q within a specific time frame was recorded, together with the water temperature.

(4)

Data Processing

With the water temperature represented as T °C, the permeability coefficient of soil-cement could be calculated, as shown in Equation (1)

$$k_T=\frac{100Q{\gamma }_{w}h}{pAt}$$

(1)

where $k_T$ is the permeability coefficient of soil-cement at the water temperature T °C, cm/s; Q is the seepage volume in t period, cm³; A is the sectional area in the middle of the sample, cm²; T is the time interval, s; I is the hydraulic gradient with no measurement unit; P is the osmotic pressure, MPa; h is the height of the specimen, cm and ${\gamma }_w$ is the weight of the water, taken as 0.0098 N/cm³.

The standard temperature selected for the soil-cement penetration test was 20 °C, and the permeability coefficient at T °C was adjusted to 20 °C. The dynamic viscosity coefficient of the water complied with the relevant specifications of the Standard for Soil Test Methods (GB/T50123-2019) [43]. The permeability coefficient of soil-cement was calculated as shown in formula (2):

$$k_{20}=k_T\times \frac{{\eta }_T}{{\eta }_{20}}$$

(2)

where $k_{20}$ is the permeability coefficient of soil-cement at the standard temperature of 20 °C, cm/s; ${\eta }_T$ is the dynamic viscosity coefficient of the water at water temperature T °C, kPa·s and ${\eta }_{20}$ is the dynamic viscosity coefficient of the water at 20 °C, kPa·s.

2.3.2. Chloride Ion Permeability Test Procedure

The greatest difference between the chloride ion permeability test method and the general permeability test method was that the former focused on the seepage and migration of the chloride ions to the soil-cement pores to evaluate the permeability of the soil-cement to chloride ions. The chloride ion permeability testing methods can be subdivided into slow, fast and other methods, according to the duration of the tests. The rapid electric flux method was selected for the present study.

(1)

Preparation and Curing of the Samples

The test molds used in this test were cylinders of 100 mm in diameter and 300 mm in height. After the specimens had been cured for the corresponding period in a standard curing room at a temperature of 20 ± 3 °C and relative humidity of 90%, slices of about 70 mm were cut off from both ends of the samples. Their middle part was then further cut into three cylinders of 100 mm ± 1 mm in diameter and a height of 50 mm ± 2 mm. The samples after cutting are shown in Figure 5.

(2)

Sample Sealing with Wax and Soaking

The cut samples were placed in an oven at 80 °C for drying for 4 h before being laid on their side in melted liquid paraffin and rolled over evenly. Once the paraffin on the samples cooled, the rolling was repeated three to four times. Roughly, 0.3 mol/L of NaOH solution and an NaCl solution with a mass concentration of 3.0% were prepared. The wax-sealed samples were then placed vertically in an automatic salt-saturated water machine (BSJ type, Suzhi Yilong Instrument Co., LTD., Beijing, China) and soaked for 18 h, as shown in Figure 6.

(3)

Electric Flux Data Collection

A direct current (DC) of 60 V was connected to the samples using an RCM-DTL multifunctional chloride ion tester (Suzhi Yilong Instrument Co., LTD., Beijing, China) and maintained for 6 h, as shown in Figure 7. The device automatically collected and stored the latest values every minute.

(4)

Data Processing

The electric flux values of the samples could be calculated according to the following formula

$$Q={\displaystyle \underset{0}{\overset{T}{\int }}I\cdot dt}$$

(3)

where Q is the electric flux value, C; I is the current value, A; and T is the power of time, min.

The calculated value was the electric flux value of the 100 mm diameter samples, which then had to be converted into the electric flux values of the 95 mm diameter samples using to the following formula

$$Q={\displaystyle Q\times {\left(\frac{100}{95}\right)}^2=1.1080Q}$$

(4)

where Q is the electric flux value when the sample diameter was 100 mm, C; and Q₀ is the electric flux value when the sample diameter was 95 mm, C.

For each group of tests, the mathematical mean value of the electric flux recorded for the three samples was taken as the measured value for that group.

2.3.3. Scanning Electron Microscope Test Procedure

A QUANTA 250 multifunctional scanning electron microscope (FEI Company, Hillsboro, OR, USA), as shown in Figure 8, was selected for the test, which offers a high magnification and stereoscopic image. It can directly observe the fine structure of various uneven surfaces of samples and is very suitable for material science.

(1)

Sampling

After the chloride ion permeability test, the samples were crushed under pressure and small slices with flat upper and lower surfaces were selected from the crushed samples, as shown in Figure 9. The selected fragments were then immersed in pure ethanol for at least 24 h.

(2)

Gold Coating of the Surfaces

In order to avoid any charge accumulation affecting the mechanical stability and resulting in unclear images, the samples were coated before testing to provide conductivity. In this test, a layer of gold film was coated onto the surface of the samples with a gold spraying instrument.

(3)

Scanning Imaging

A QUANTA 250 multifunctional SEM tester was selected. The gold-plated samples were placed into the sample capsule and vacuumized to 10⁻³ Pa. The appropriate magnification and scanning speed were selected, and the images were saved after fine focusing.

3. Results

3.1. Penetration Test Results

According to the relevant specifications of the Standard for Soil Test Methods (GB/T50123-2019) [43], the permeability coefficient was modified, and the results are shown in

Table 4. Permeability coefficient of basalt fiber soil-cement (10⁻⁸ cm/s).

Number	Age (days)
	7	28	60	90
A	12.25	9.34	7.74	7.16
B	10.21	7.79	6.35	5.45
C	8.47	6.66	5.29	4.39
D	8.11	6.03	4.34	3.68

.

3.1.1. Influence of the Basalt Fiber Content on the Permeability Coefficient

Based on the test results in Table 4, the relationship curve between the soil-cement permeability coefficients and the basalt fiber content was drawn, as shown in Figure 10.

It can be seen intuitively in Figure 10 that the permeability coefficient of soil-cement at any age decreased as the basalt fiber content increased. The precise effects of the different basalt fiber content ratios on the permeability coefficient of soil-cement were analyzed using group A, which had a basalt fiber content of 0. This group remained the benchmark group for which the declining rates of the permeability coefficient of the remaining groups were calculated. The corresponding results are shown in

Table 5. Declining rates (%) of the permeability coefficient of basalt fiber soil–cement.

Number	Age (days)
	7	28	60	90
A	0	0	0	0
B	16.65	16.60	17.96	23.88
C	30.86	28.69	31.65	38.69
D	33.80	35.44	43.93	48.60

.

The following can be observed from Table 5 and Figure 10.

(1)

When the curing age was 7 days, the permeability coefficients of groups B, C and D were, respectively, 16.65%, 30.86% and 33.80% lower than those of group A, which contained no basalt fiber. This indicated that increasing the basalt fiber content improved the impermeability of soil-cement.

(2)

When the curing age was 28 days, the permeability coefficients of groups B, C and D decreased by 16.60%, 28.69% and 35.44%, respectively, compared to group A. However, when the basalt content was increased by 0.5%, the permeability coefficients decreased by 16.60%, 14.51% and 9.46%, respectively. This established that, although increasing basalt content improved the impermeability of soil-cement, the degree of enhancement gradually reduced.

(3)

At the curing ages of 60 days and 90 days, it can be seen from Figure 1 that the trend in the permeability coefficient changes was almost the same as that at 7 days and 28 days. With the increase in the fiber content (0 → 0.5% → 1.0% → 1.5%), the gradients of the broken lines also changed from steep to shallow.

While the addition of basalt fiber was furthermore found to reduce the early shrinkage of soil-cement, the later hardening dry shrinkage was also observed to have diminished due to the water retention properties of the fiber. The present authors, therefore, hypothesized that firstly, the internal cracks and porosity of soil-cement diminished with the addition of basalt fiber, whereas its compactness improved; secondly, that basalt fiber could mitigate the expansion of cracks, prevent the formation of connected cracks and reduce the formation of seepage channels; and thirdly, that the fiber itself could concurrently impede the flow of water. The authors, therefore, concluded that adding basalt fiber (0 → 0.5% → 1.0% → 1.5%) to soil-cement could enhance its impermeability.

Notwithstanding, the experiment also showed that, with the increase in the basalt content, the improvement of the soil-cement impermeability decreased gradually. This was because too much basalt fiber causes fiber aggregation, which in turn further expanded the soil-cement cracks, and thus, reduced the cohesion between the fiber and soil-cement. This was consistent with the experimental results that the basalt fiber content increased by the same amount (0.5%), while the permeability coefficient of soil-cement decreased by a decreasing amount.

In summary, although with the increase of basalt fiber content, the permeability of soil-cement decreases and the permeability increases, but the amount of basalt fiber is not the more the better. This was because, first, as we continued to increase the content of basalt fiber, the reduction in the permeability coefficient was not significant enough, so there was a problem of low cost performance. Secondly, the excessive increase in the basalt fiber caused fiber aggregation, which was unfavorable to the permeability of cement-soil. According to the research on the mechanical properties of basalt fiber cement-soil in the literature [32,35,36,37,39], the mechanical properties will not increase indefinitely, and the strengthening range will continue to decrease with the increase in the basalt fiber content. When the fiber content is greater than 1.0%, the strength growth range will be significantly reduced and the basalt fiber content with better mechanical properties will be between 1.0% and 1.5%. Therefore, the optimal content of basalt fiber in this study was 1.5%.

3.1.2. Relationship between the Permeability Coefficient and the Basalt Content

As previously indicated, the basalt fiber contents set in this paper were 0, 0.5%, 1.0% and 1.5%. The analysis of the above test results revealed that, although the impermeability of soil-cement improved as the basalt fiber content increased, in the course of the latter process, the rising trend in the soil-cement impermeability slowed and potentially even reversed. Due to the upper limit of the basalt fiber content set for this study, it was not possible to determine how the impermeability of soil-cement might have been affected when the fiber content exceeded 1.5%. It was, however, hypothesized that the permeability coefficient of soil-cement, when the basalt fiber content exceeded 1.5%, could reasonably be predicted by establishing the functional relationship between the basalt fiber content and the permeability coefficient of soil-cement, with the view of providing practical guidelines for the use of basalt fiber in soil-cement to achieve the optimal impermeability in engineering projects.

(1)

At 7 days, as shown in Figure 11a, the quadratic polynomial equation of the permeability coefficient changes according to the basalt fiber content variations was obtained as follows

$$k_7=1.68x^2-5.352x+12.304({\mathrm{R}}^2=0.9638)$$

(5)

(2)

At 28 days, as shown in Figure 11b, the quadratic polynomial equation of the permeability coefficient changes according to the basalt fiber content variations was obtained as follows

$$k_{28}=0.92x^2-3.592x+9.344({\mathrm{R}}^2=0.9908)$$

(6)

(3)

At 60 days, as shown in Figure 11c, the quadratic polynomial equation of the permeability coefficient changes according to the basalt fiber content variations was obtained as follows

$$k_{60}=0.44x^2-2.912x+7.729({\mathrm{R}}^2=0.9829)$$

(7)

(4)

At 90 days, as shown in Figure 11d, the quadratic polynomial equation of the permeability coefficient changes according to the basalt fiber content variations was obtained as follows

$$k_{90}=1.00x^2-3.801x+7.145({\mathrm{R}}^2=0.9850).$$

(8)

Figure 11. (a) Fitting curve of the permeability coefficient and the basalt fiber content at 7d; (b) fitting curve of the permeability coefficient and the basalt fiber content at 28d; (c) fitting curve of the permeability coefficient and the basalt fiber content at 60d; (d) fitting curve of the permeability coefficient and the basalt fiber content at 90d.

Figure 11. (a) Fitting curve of the permeability coefficient and the basalt fiber content at 7d; (b) fitting curve of the permeability coefficient and the basalt fiber content at 28d; (c) fitting curve of the permeability coefficient and the basalt fiber content at 60d; (d) fitting curve of the permeability coefficient and the basalt fiber content at 90d.

3.2. Chloride Ion Permeability Test Results

Since this test used a fully automated electric flux meter, the calculation and conversion of the sample electric flux values were automatically completed by the instrument. The specific results are shown in

Table 6. Electric flux value of basalt fiber soil-cement.

Number	Sample Height	Electric Flux Value	Average Value
	(mm)	(C)	(C)
A	50	580	542
	51	533
	50	512
B	49	465	487
	51	504
	50	492
C	52	523 *	446
	49	451
	50	440
D	49	398	417
	48	410
	51	444

Note: * marked as an abnormal value in the electric flux measurement; abnormal data are not required.

.

In order to reflect the influence of the basalt fiber content on the chloride ion permeability of soil-cement more intuitively, a broken line diagram of the altered electric flux values according to the basalt fiber content variations was drawn on the basis of the test results of Table 6, as shown in Figure 12.

It can be seen intuitively from Figure 12 that the flux value decreased with the increase in the basalt fiber content. In order to analyze the effects of the different basalt fiber content ratios on the flux value of soil-cement more specifically, group A with a basalt fiber content of 0 was taken as the reference group. The data from the different basalt fiber content groups were then compared with those of the reference group, after which the two groups with a rate of increase of 0.5% for the basalt fiber content were compared.

Table 7. Decreasing rate of the electric flux of basalt fiber soil-cement.

Control Group	B Contrast A	C Contrast A	D Contrast A	C Contrast B	D Contrast C
Rate of decline (%)	10.15	17.71	23.06	8.42	6.50

illustrates the results obtained.

As seen in Table 7, when the basalt fiber content was 0.5%, 1.0%, and 1.5%, the electric flux value of soil-cement decreased by 10.15%, 17.71% and 23.06%, respectively, compared to the electric flux value of the control group without basalt fiber. When the basalt fiber content increased from 0 to 0.5%, the electric flux decreased by 10.15%; when the basalt fiber content increased from 0.5% to 1.0%, the electric flux decreased by 8.42%; and when the basalt fiber content increased from 1.0% to 1.5%, the electric flux decreased by 6.50%. The data show that, as the fiber content increased, the electric flux value of soil-cement decreased. However, the rate of decrease weakened, indicating the strengthening effect of the soil-cement impermeability was slowing. It was, therefore, concluded that the permeability test results were essentially consistent with those of the hydraulic chloride ion permeability tests, and that the optimal dosage of basalt fiber was, indeed, 1.5%.

3.3. Scanning Electron Microscope (SEM) Test Results

The pore diameters, quantity, connectivity of the soil-cement particles and cementation between the fibers and soil-cement particles were observed. This SEM test qualitatively analyzed the permeability of soil-cement from the perspective of its microstructure, on the basis of which it derived and explained the macroscopic phenomena and laws that manifested in the previous tests. Figure 13a–d illustrates the SEM images of the samples with the different basalt content at 90 days.

It can be seen from the above SEM images that, at the age of 90 days, obvious flocculent structures had grown in the cement-soil of various basalt fibers, mainly CSH. In addition, a small amount of needle-shaped ettringite can be seen. However, hexagonal CH crystals are often covered by CSH gel due to their low strength and large volume, and their original shape is easily damaged during the sampling process, making them difficult to be directly observed. Since the curing time was long, the cement was constantly hydrated, and a large amount of gel was produced. On the one hand, it bonded the soil particles to make the soil particles more solid. On the other hand, the gel filled in the holes to improve the compactness of the soil. Therefore, it can be explained that adding cement to soil can improve the compactness and impermeability of the soil.

The length of the basalt fibers added in the test was 24 mm and the diameter was 17 μm. The length was obviously larger than the diameter, and the content of basalt fiber was not high. Therefore, if the small slice sampled was along the diameter direction of a basalt fiber, the morphology of the basalt fiber was not obvious. Basalt fibers cannot be clearly observed in (a), (b) and (c) in Figure 13, but the characteristics of the porosity and compactness can be seen. The Figure 13a shows the morphology of cement-soil without basalt fiber. It can be seen that there are small particles or needle-shaped units everywhere, and only a small part of large particle units exist. The overall skeleton was more dispersed than others. When the content of basalt fiber increased to 0.5%, as shown in Figure 13b, the pore diameter and the number of pores became significantly smaller and the particle units gradually cemented together, starting to form a more stable whole. When the content of basalt fiber increased to 1%, as shown in Figure 13c, the existence of the pores was almost invisible, the structure was significantly denser, small particles were significantly reduced and the unit body was further condensed, thus forming a more stable whole. With the increase in the basalt fiber content, the soil-cement had smaller pores, a higher degree of compactness and an enhanced impermeability. Basalt fiber filled the pores of soil-cement, making the structure denser and played the role of a seepage channel. Secondly, the hydration products of cement had a higher bonding force, which combined the fibers and soil particles well. The crystals on the surface of the fibers and the crystals in the pores were linked to each other, so that the basalt fibers, cement hydration products and soil particles formed a more stable whole.

Since the small slice sampled was along the length direction of the basalt fibers, the morphology of the basalt fibers can be clearly seen in Figure 13d. It can be seen that the basalt fiber was well wrapped by the cement-soil particles. The basalt fiber had a good toughness, and the cement crystallites wrapped the fiber surface, as if forming a hard shell outside the fiber, improve its stiffness, so that the fiber can play the role of a transfer force. When the basalt fiber cement is under tension, the basalt fiber can play the role of a tensile, reducing the formation of cracks and improving the impermeability of cement-soil.

To sum up, the addition of basalt fiber into soil can, on the one hand, improve the density because the pores in the soil-cement structure can be filled by basalt fiber. On the other hand, the cement crystallinity will wrap the fiber surface, improve the fiber stiffness and improve the bonding force between the reinforcement and soil contact surface, allowing the fiber to play the role of a transmission force, reducing the formation of cracks and improving the impermeability of soil-cement.

4. Discussion

The comparison between the research results of this paper and the previous results can provide support for the results of this study and guide the direction of future research.

Zhang et al. [4] carried out a uniaxial tensile test and an unconfined compressive strength test and found that the order of influence of the fiber variables on the tensile properties was the length, content and diameter. A 9mm length of basalt fiber and a content of 1.5% were the best single mixing parameters of basalt fiber. Ma et al. [44] obtained through the Hopkinson compression bar test that the appropriate amount of basalt fiber plays a positive role in the dynamic characteristics of cement-soil, and the appropriate amount of basalt fiber is 1.5%~2.0%. From the above analysis, it can be seen that the appropriate content of basalt fiber was also considered in the previous literature to be between 1.5% and 2.0%, which was consistent with the conclusions of this study.

Shen et al. [5] obtained the optimal tensile strength and residual strength when the fiber length of 9 mm + 12 mm was mixed at 3:1 through the tensile test of adding basalt fibers of different lengths. Therefore, the influence of the basalt fiber length and diameter on the permeability can be discussed in future studies, and the optimal mixing ratio can be studied.

Liu, et al. [45] thought that the UCS decreased with the increase in the silt content when the cement content was 30%. With the cement content ranging from 15%–25%, the UCS increased at first with the increase in the silt content but decreased once the silt content reached a ‘saturation’ point. It can be seen that the properties of the soil, such as the particle size and mineral composition, have a great impact on the performance of cement-soil. Therefore, in the future research, it is necessary to analyze the impermeability of basalt fiber cement-soil in combination with the local soil characteristics.

Ma et al. [2] found that the appropriate amount of basalt fiber and sand can jointly promote the improvement of the tensile and compressive strength of cement-soil. Liu, et al. [46] thought that both fly ash and Dura Crete could be used as partial replacements for cement. Therefore, the combination of basalt fiber and other additives can be considered in future research.

Chen et al. [47] tested the mechanical strength of basalt fiber soil-cement under the dry–wet cycle condition and concluded that the mechanical strength of soil-cement with a 0.3% fiber content reached the maximum. It can be seen that the optimal basalt fiber content was different under the various environmental conditions, so a comprehensive analysis should be carried out according to the local environmental conditions. Therefore, the influence of basalt fiber on the soil-cement impermeability under dry–wet cycle and freeze–thaw action can be considered in future studies.

5. Conclusions

Taking the soft soil commonly found in Fuzhou, Fujian Province, China as the research object, the influence of the basalt content on the impermeability of soil-cement was studied by adding cement and basalt fibers. In this study, the impermeability of basalt fiber-reinforced soil-cement was studied through permeability testing and chloride ion permeability testing, and the influence of the different basalt fiber content ratios on the permeability coefficient and electric flux value of soil-cement was determined. The microscopic mechanism of basalt fiber-reinforced soil-cement was studied using SEM testing, and the microstructure characteristics of basalt fiber-reinforced soil-cement were obtained. The phenomena and laws identified using macroscopic testing were explained by the analysis of the various microscopic mechanisms. According to the test results and the research scope of this study, the following was concluded.

(1)

With the increase in the basalt fiber content, the permeability coefficient of soil-cement decreased and the impermeability of soil-cement increased. The basalt fiber content increased from 1.0% to 1.5% and the permeability coefficient only decreased by 9.46%. In addition to the problem of low cost performance, a too high basalt fiber content is can form agglomeration and cause cracks. Therefore, the optimum content of basalt fiber is 1.5%.

(2)

According to the permeability test data, the relationship between the content of basalt fiber and the permeability coefficient of soil-cement was analyzed using regressive analysis, and the power index fitting curve between the permeability coefficient of soil-cement and the content of basalt fiber was obtained. Therefore, the permeability coefficient of basalt soil when the content of the basalt fiber exceeds 1.5% can be reasonably predicted, which provides practical guidance for using basalt fiber to achieve the best impermeability in engineering.

(3)

The findings of the chloride ion permeability test showed that, with the increase in the basalt fiber content, the total electric flux through soil-cement decreased within 6 h, indicating that the chloride ion permeability of soil-cement was enhanced, but the extent of the enhancement was weakened as the basalt fiber content increased. By comparing the two different permeability tests, it was found that both test outcomes were consistent. That is, the optimal content of basalt fiber was 1.5%, which further proved that basalt fiber had a strengthening effect on the impermeability of soil-cement.

(4)

It can be seen from the scanning electron microscope images that, with the increase in the basalt fiber content, the permeability resistance gradually increased mainly due to the following reasons. First, basalt can fill the pores of cement-soil, making the structure more compact. Secondly, the cement crystallites will wrap the fiber surface, improve the fiber stiffness and improve the bonding force between the reinforcement and soil contact surface so that the fiber can play the role of transmission force, reducing the formation of cracks and improving the impermeability of soil-cement.