Study on the Effects and Mechanism of the Reinforcement of Soft Clay via Microbially Induced Carbonate Precipitation

Abstract

A series of microbial-induced carbonate precipitation (MICP) experiments were conducted using Sporosarcina pasteurii to reinforce coastal soft clay in Zhejiang. By analyzing the physical and mechanical parameters of samples of varying ages, specifically focusing on each sample’s unconfined compressive strength, triaxial shear strength, and permeability coefficient, it was revealed that MICP technology can be used effectively to reinforce coastal clay. The unconfined compressive strength of treated soil increased by 23% compared to untreated soil, while the permeability coefficient decreased by 75%. The internal friction angle of the clay remained almost constant, whereas cohesion significantly increased by approximately 53%. One-dimensional compression experiments were also performed, yielding consolidation parameters such as the compression coefficient, compression index, and consolidation coefficient. The results indicated a notable decrease in the soil compression index. Furthermore, microscopic analysis revealed that clay particles were cemented by calcium carbonate, whose precipitation was induced by the bacteria. Our XRD results also indicated that the bacteria facilitated the conversion of Ca²⁺ present in the soil into calcium carbonate.

1. Introduction

With advancements in foundation reinforcement technology, microbially induced carbonate precipitation (MICP) [1,2] has garnered significant attention due to its environmental benefits, its low energy consumption, and the fact that it generates minimal pollution. Traditional methods, such as the use of cement and chemicals, are costly, especially for large-scale applications. Unlike traditional methods such as vacuum preloading and chemical grouting, MICP technology utilizes naturally occurring soil microorganisms. These microorganisms adsorb Ca²⁺ from environmental solutions onto their cell surfaces, decompose urea, and subsequently produce CO₃²⁻ and NH₄⁺ through urease activity. The Ca²⁺ then reacts with CO₃²⁻ to form substantial amounts of calcium carbonate crystals on the cell surfaces. Moreover, bacteria present in geotechnical materials produce large quantities of calcium carbonate, which bonds with microparticles in the soil, filling internal pores and fissures [3]. This process enhances soil strength, reduces soil permeability, and improves soil compressibility [4,5]. Construction methods that involve MICP generate minimal chemical pollution in surrounding soil and water environments and produce almost no toxic or harmful substances. These methods are not only simple and cost-effective but also offer substantial advantages over traditional soil-reinforcement techniques in terms of cost efficiency, operational efficiency, and environmental impact across various soil types [6,7].

MICP technology was initially applied to treat sandy soil [8,9,10]. Researchers such as Whiffin et al. [9] and Soon et al. [11] utilized the MICP method to treat sandy soil, resulting in significant improvements in cementing properties and anti-permeability. Rather than using traditional grouting, Oliveira et al. [12] mixed urease and cementation liquid with silty soil, followed by compaction and storage, which yielded effective reinforcement outcomes. Extensive research has been conducted on sandy and silty soils [13], but there are relatively few studies on the use of the MICP method for the reinforcement of soft clay. Sun et al. [14] employed MICP to improve sand–clay mixtures, and they found that the addition of kaolin clay inhibits the urease activity of bacteria. Peng et al. [15] applied the slip casting method to consolidate organic clay, demonstrating significant improvements in unconfined strength and permeability coefficients. However, their study only focused on the relationship between calcium ion or ammonia nitrogen concentrations and the number of grouting applications, without considering the effects on strength, the permeability coefficient, or other characteristics over time. Canakci et al. [16] investigated Bacillus-pasteurii-induced carbonate precipitation in organic soft soil, revealing that bacterial treatment effectively enhances the compression performance and shear strength of the soil. Compared to sandy soil, soft soil has a low permeability coefficient and small pore diameters, which significantly affect the migration of microorganisms within the soil [17].

MICP excels in sand reinforcement, enhancing soil strength, stiffness, erosion resistance, and permeability by binding loose sand particles into a cohesive whole with bio-cement. For different soil types, such as those incorporating the use of rubber particles for reinforcing calcareous sand [18] and the addition of activated carbon in red clay [19], the reinforcement effect is improved. In recent years, the incorporation of glass fibers, basalt fibers, and carbon fibers, combined with MICP, has led to sand samples having unconfined compressive strength (UCS) values that are 4.5 times, 7 times, and 11 times that of samples without fibers, respectively [20]. Zhao et al. [21] also compared the effects of several fibers on MICP-treated samples, finding that carbon fibers performed the best, while basalt fibers were the most cost-effective. The use of single MICP reinforcement is generally considered to be more cost-effective than the use of traditional reinforcement materials such as cement, but the cost of reinforcement is influenced by various factors, including the use of additives, solution concentration, bacterial strain, and the type of soil being reinforced. Due to the particles of clay being very fine, MICP treatment may necessitate greater quantities of bacterial strains and solutions, along with the associated costs of additives such as activated carbon. Additionally, the penetration and uniform distribution of MICP in clay can be more challenging, potentially resulting in higher construction costs compared to sand.

Previous studies in the literature have found that methods such as the fiber-based reinforcement of soft clay and sand–clay mixtures are effective for MICP reinforcement. Incorporating fibers not only enhances the shear strength of clay but also increases the colonization space for microorganisms, thereby improving the deposition efficiency and yield of calcium carbonate. The cementing effect of calcium carbonate further promotes fiber reinforcement [22]. Using controlled amounts of clay in clean sands can enhance the efficiency of MICP treatment. Pore clogging, caused by low percentages of clay, may not be a significant issue, and clay minerals can provide additional surface area for bacteria to attach to, leading to the precipitation of larger quantities of bio-cement [23]. The above methods can optimize the migration and distribution of microorganisms within the soft clay. Future research will consider incorporating these methods into further experimental studies on MICP reinforcement in soft clay to enhance the reinforcement performance.

The research described in this paper employed a method consisting of mixing bacterial fluid, cementation liquid, and soft clay to simulate the process of soft clay ground reinforcement via MICP. The mineral composition of the soil following MICP reinforcement was analyzed by measuring the physical and mechanical indices and conducting microanalyses of soil cultivated over a specific period. This study investigated this treatment’s effects on soft clay and examined the reinforcement mechanisms of MICP technology.

2. MICP-Based Clay-Reinforcement Experiments

2.1. Bacteria and Cementation Liquid

This experiment utilized Sporosarcina pasteurii (ATCC No. 11859), a chemoheterotrophic bacterium characterized by rod-shaped or spherical cells (2–3 μm in length) and Gram-positive staining. The culture medium comprised 20.0 g yeast extract, 10.0 g (NH₄)₂SO₄, and 24 mg NiCl₂ per liter, with the pH being adjusted to approximately 9.0 using a 1 mol/L sodium hydroxide solution.

The cementation liquid used in this experiment was prepared with CaCl₂ and urea. CaCl₂ provided Ca²⁺ for the MICP biochemical reaction, while urea served as a nitrogen source to sustain Sporosarcina pasteurii. This study utilized the urea and CaCl₂ mixing ratio described by Peng et al. [15], and thus concentrations of 2 mol/L urea and 1 mol/L CaCl₂ were used to prepare the cementation liquid.

2.2. Soil Samples and Sampling Method

For this experiment, soil samples were collected from Taizhou, Zhejiang Province, using thin-wall samplers by drilling to a depth of approximately 4–10 m. The natural water content of the soil was 45%, and the natural gravity was 18.2 kN/m³. The bacterial fluid and remolded soil were mixed, maintaining a proportion of 1:1 between the bacterial fluid and the cementation liquid.

The molded soil was stored in moisturized vessels and maintained for 7, 14, and 28 days. The temperature in the maintenance room was kept at 25–28 °C, with a humidity level of 95%. After the maintenance period, a distinct white bacterial colony group was observed on the soil surface (Figure 1).

3. Experimental Results and Analysis

3.1. Unconfined Compressive Strength Tests

After the sampling preparation procedure described in Section 2.2, we prepared multiple cylindrical specimens from the cured soil sample, each with a diameter of 50 mm and a height of 100 mm, for unconfined compressive strength (UCS) tests. We saturated the specimens using a saturator for a duration of 24 h.

This study utilizes the Geotechnical Digital Systems (GDS) apparatus (GDS Instruments, Hook, UK), which primarily consists of a confining pressure chamber, seepage tubes, and a pressure–volume controller. These components enable real-time readings of seepage rates, confining pressure, and head changes at the top and bottom of the sample during the seepage process. The GDS reaction frame loading system uses a motor for uniform loading and can measure displacement and reaction force values through the displacement and force sensors located at the top of the reaction frame. During the UCS test, the loading rate is set to a strain rate of 1% per minute, corresponding to an actual loading rate of 1 mm/min. Loading is stopped when the strain reaches 20%. Both treated and untreated soils exhibit strain-hardening properties without a strength peak; therefore, the stress value at 15% strain was selected as the unconfined compressive strength of the soil. The experimental results are shown in Figure 2.

As the treatment duration increased, the unconfined compressive strength of the soil notably improved. The optimal curing age required for the MICP-treated soft clay to achieve maximum strength and durability was 14 days, the time within which the peak value is reached. After this point, the strength increase plateaued. The unconfined compressive strength increased by 17 kPa, with a growth rate of approximately 23%, suggesting that further curing does not substantially enhance the strength.

The strength development of MICP-treated soft clay can be categorized into three stages. In the early stages (the initial days), the strength increase is minimal because the generation of calcium carbonate is not yet complete, resulting in minimal cementation and a slow strength increase. During the intermediate period (up to 14 days), the strength increase is rapid, as microbial activity leads to the complete generation of calcium carbonate, forming comprehensive carbonate cementation around the microbes, significantly enhancing the strength. In the later stages (beyond 14 days, as shown in Figure 2), the strength increase plateaus as microbial activity gradually declines and the cementation process stabilizes, meaning additional curing does not substantially improve the mechanical properties. The results in [24] indicate that traditional long-term reinforcement methods, such as cement-based reinforcement, exhibit relatively uniform and continuous strength growth across different ages. In contrast, the strength increase induced by the MICP reinforcement method stops relatively early. This is why some studies suggest that to ensure the effects of MICP are long term, periodic treatments are necessary [25].

3.2. Triaxial Shearing Strength Tests

For the triaxial shearing strength test, a GDS triaxial testing apparatus is used. This instrument utilizes three controllers to independently control and monitor the confining pressure, back pressure, and base pressure (pore water pressure) in real time. Triaxial shearing was carried out using the GDS reaction frame loading system. The sample dimensions and saturation preparations are the same as those for the unconfined compressive strength test. Conventional triaxial tests are conducted under confining pressures of 100 kPa, 200 kPa, and 300 kPa, with the loading rate controlled to a strain rate of 0.1% per minute, corresponding to an actual loading rate of 0.1 mm/min.

Animesh and Ramkrishnan (2016) [26] investigated the effects of varying bacterial concentrations on intermediately compressible clay and highly compressible clay. Their results indicated that the unconfined compressive strength (UCS) of both soil types increased with higher bacterial cell concentrations. In contrast, for MICP-treated soft clay, the cohesion increases notably, while the internal friction angle remains essentially unchanged. Therefore, the carbonate cementation generated after MICP treatment significantly enhances the cohesion of the soil. This cementation effect is primarily reflected in the increased cementation of soil particles rather than in the friction and occlusion resulting from the interlocking effect among the particles. According to previous research, for MICP-treated sandy soil, both the internal friction angle and cohesion increase significantly [17,27,28,29]. Their results indicate that as the casting time increases, the volume of generated calcium carbonate also increases. Consequently, the sandy soil cohesion can increase from 40 kPa to 160 kPa, and the internal friction angle can increase linearly from 32° to 35.5°.

The results are depicted in Figure 3. The left axis of the figure represents the cohesive force, while the right axis indicates the internal friction angle. According to the experimental data, as the cultivation period increases, the soil cohesion markedly improves. After 7 days, soil cohesion increased by 30.5% compared to the control group. After 14 days, soil cohesion increased by 53% and then stabilized. The internal friction angle, however, remained essentially unchanged at approximately 23°.

3.3. Laboratory Permeability Tests

To further analyze the physical and mechanical properties of the MICP-treated soil, we conducted constant-water rate permeability tests on the MICP-treated soil over three different periods. For constant head permeability tests, soil samples cured to a certain age are used, and the tests are conducted with a GDS consolidation-permeability apparatus. The sample preparation procedures are the same as those for the unconfined compressive strength test. In this test, the k_v-back and k_v-base values represent the permeability coefficients at the top and bottom of the sample, respectively. These values are calculated by the back and base pressure volume controllers of the GDS permeability system based on changes in the volume of the fluid that exudes or infiltrates within a certain unit of time. The permeability coefficient is then calculated using Darcy’s law and the relationship between the drainage volume and flow rate of the soil. The results are shown in Figure 4. As the cultivation period increases, the permeability coefficient of the soil notably declines, stabilizing after approximately 14 days, with a final reduction rate of approximately 76%.

According to several previous studies, for silty and sandy soils, the permeability coefficient of MICP-treated soil declines sharply [11,30]. MICP technology can be utilized to close fissures and enhance the anti-permeability performance of soil. Similar to sandy soil, microbially induced carbonate cementation in soft clay results in soil pore filling and a decrease in the permeability coefficient, also enhancing the soil strength. This behavior aligns with the development patterns observed in soil strength. The permeability coefficient stabilizes after approximately 14 days.

3.4. One-Dimensional Compression Tests

For the one-dimensional compression characteristics test, a GDS one-dimensional consolidation apparatus is used. Axial pressure is applied by injecting water into the axial pressure bladder via an axial pressure volume controller, which also monitors the axial pressure in real time. The back pressure controller and base pressure controller simultaneously control and measure the water pressure at the top and bottom of the sample. Using a top displacement sensor and full computer control, the system can achieve constant stress loading and high-precision monitoring.

Table 1. Loading scheme for one-dimensional compression test.

No.	Period (Day)	Axial Stress (kPa)
1	0	25→50→100→200→100→50→100→200→400→800→1600
2	14	25→50→100→200→100→50→100→200→400→800→1600

summarizes the loading scheme used in the compression test.

One-dimensional compression tests were performed on the MICP-treated soil. The loading scheme is shown in Table 1. As the load gradually increases, the deformation of the untreated soil samples is more pronounced compared to the MICP-treated soil. Therefore, MICP treatment significantly improves the compression properties of soil under large external loads.

As shown in Figure 5, the compression coefficient of the MICP-treated soil decreased notably. We selected a linear section of the e-lg p curve to calculate the compression index of the soil before treatment (C_c = 0.225) and after 14 days of treatment (C_c = 0.172). These results are similar to the findings of Hanifi Canakci’s research on Sporosarcina-pasteurii-induced carbonate precipitation in organic soil [16]. Both studies indicate that bacterial treatment effectively improves the compression performance of the soil.

We also performed unloading–reloading tests and obtained the relationship curve between the pore–space ratio and stress under springback–recompression conditions. The results indicate that the recompression index before treatment is Ce = 0.0226, and the recompression index after treatment is Ce = 0.0188. Similar to the compression index, the compression and recompression indexes of the MICP-treated soil are lower than those of the soil before treatment. Consequently, MICP treatment effectively decreases the compression index of soft clay and improves its overall compression performance.

According to the strain–time relationship of the soil under loading, we can calculate the consolidation coefficient of the soil under each loading condition. Zeng et al. [31] proposed that the time square root method is the best approach when the loading ratio is larger than 5 and the loading time is limited. As shown in Figure 6, as the load increases, the consolidation coefficient of the soil consistently increases. However, the consolidation coefficient of the soil after treatment is lower than that of the soil before treatment. Consequently, the soil after MICP treatment takes a longer time to complete consolidation.

3.5. Microanalysis Results for the MICP-Treated Soil

3.5.1. SEM Morphology Analysis

We selected MICP-treated soil and control group samples with ages of 7 days, 14 days, and 28 days to conduct SEM (Scanning Electron Microscopy) experiments. Using a knife, we made two parallel scratches on each soil sample; the scratches were approximately 0.5 mm deep and 2 mm apart. We broke each sample along the scratches to obtain an undisturbed SEM sample. After drying a sample, we coated it with gold to make it conductive. We then used the large-chamber tungsten filament scanning electron microscope (Zeiss EVO18, Oberkochen, Germany) at the Ningbo Institute of Materials Technology and Engineering, Chinese Academy of Sciences, to capture a sufficient number of SEM images at an appropriate magnification. This allowed us to examine the internal soil microstructure and analyze the MICP reinforcement mechanism at the microlevel. The surface micromorphology of the soil samples at different periods was observed and analyzed.

We specifically identified clearly generated calcium carbonate crystals in the spaces between the soil particles after 7 days (Figure 7). These calcium carbonate crystals exhibit obvious regular shapes and are mostly bulk crystals. According to this figure, calcium carbonate crystals serve as a filling material between the soil particles, which results in increased soil cohesion.

In contrast to treated sandy soil, where calcium carbonate occurs as separate granules adjacent to two or more sand particles, in soft clay, the calcium carbonate forms as cement mineral granules among clay particles. These observations indicate that in soft clay, microbes can induce calcium carbonate precipitation around soil particles, thereby enhancing soil cohesion.

Upon examining the soil sample after 28 days, we found locally concentrated calcium carbonate generation in the soil (Figure 7). The microbes in soft clay are relatively activated and effectively induce calcium carbonate precipitation around soil particles, thereby realizing soil reinforcement.

3.5.2. XRD Analysis of the Soil Composition

We conducted XRD experiments on MICP-treated soil and control group samples with ages of 7, 14, and 28 days using the Advance X-ray diffractometer (D8 advance; Bruker AXS, Madison, WI, USA). Figure 8 and

Table 2. Mass fractions for the composition of the MICP-treated soil.

Category	Chemical	0 Days (Control Group)	7 Days	14 Days	28 Days
Quartz	SiO2	17.3	15.7	17.2	17.2
Anorthite	Ca (Al2Si2O8)	0.4	0.5	0.7	0.6
Microcline	K (AlSi3O8)	29.6	33.5	29	29.1
Pegmatite	KAl2 (AlSi3O10) (OH)2	47.3	43.8	45.6	46.1
Dolomite	CaMg (CO3)2	1.6	1.9	2.3	1.8
Calcite	Ca (CO3)	1.7	2.8	2.9	3.3
Others	/	2.1	1.7	2.2	1.9

summarize the results. Dolomite and calcite are present in the soil in the form of calcium carbonate, and the sum of their contents is considered the total soil calcium carbonate content. The results obtained from XRD should be regarded as semi-quantitative references due to the inherent limitations in precision. Future research will incorporate chemical analysis to achieve more accurate and reliable assessments.

The amount of precipitated calcium carbonate can reflect the effectiveness of MICP reinforcement. This paper uses the acid washing method to measure the amount of precipitated calcium carbonate. The specific steps are as follows:

• Take the sample that failed in the unconfined compressive strength test and dry it to a constant weight (recorded as M1);
• Immerse and dissolve the sample using an excess of hydrochloric acid with a concentration of 1 mol/L. Stir until no bubbles escape from the soil. Then, rinse multiple times with an excess of deionized water and dry it in an oven;
• Repeat the immersion step until no bubbles are produced, and dry it to a constant weight (recorded as M2).

The calcium carbonate content in the sample is calculated using the following formula:

$$C_{{\mathrm{CaCO}}_3}={\displaystyle \frac{M_1-M_2}{M_2}}\times 100\%$$

(1)

Based on the CaCl₂ concentration (1 mol/L), the Ca²⁺ mass in the cementation liquid, corresponding to a per soil mass unit, is 5 g. In comparing the soil masses after different periods (Figure 9), after 14 days, the fixed Ca²⁺ content in the soil was 6.2 g/kg, which was higher than the content in the cementation liquid. Thus, microbes continuously induced calcium ions to generate calcium carbonate or magnesium calcium carbonate, utilizing both Ca²⁺ from the added cementation liquid and the original dissociative Ca²⁺ in the soil. Consequently, the final generation amount was notably larger than the amount of Ca²⁺ mixed in the cementation liquid.

The reinforcement effect is influenced not only by the concentration of CaCl₂ but also by the concentration of bacteria. Cheng et al. (2007) [32] found that there is a relationship between the concentration of bacteria and the morphology of calcium carbonate crystals. They found that at high bacterial concentrations, most calcium carbonate crystals were spherical, whereas at low bacterial concentrations, the crystals were predominantly regular rectangular parallelepipeds or agglomerates. Due to the limitations regarding this paper’s scope, future research could explore this factor in greater detail.

As shown in Figure 9, the experimental results indicate that the total calcium carbonate content in soil changes with the cultivation period. Calcium carbonate is generated in the soil primarily within the first 14 days, which is consistent with the changes observed in the physical and mechanical soil parameters such as the unconfined compressive strength. Additionally, microbially induced carbonate generation is the mechanism underlying the use of MICP for soft soil reinforcement.

4. Discussion

At present, MICP is widely applied for the reinforcement of sandy soil, and many scholars have studied its reinforcement effect on sandy soil, as well as its reinforcement mechanism. In this paper, we have primarily examined the reinforcement effect and mechanism of MICP technology for soft clay. According to the experimental results, our analysis of the physical, mechanical, and consolidation properties, and our microanalysis of the MICP-treated soil, the Sporosarcina pasteurii in the cementation liquid effectively induces Ca²⁺ ions in the soil to precipitate as carbonate, thereby reinforcing soft clay.

According to the experimental results for the MICP-treated soil, over the three different periods, we found that reinforcement effects are notable within the first 14 days. The XRD results indicate that the fixed Ca²⁺ content in the soil does not increase after 14 days. Additionally, as the Ca²⁺ content increases, the unconfined compressive strength of the soil also changes. Therefore, Sporosarcina-pasteurii-induced carbonate precipitation is a significant driving factor behind the strength increase in soft clay, and the soil-reinforcement effect stabilizes after 14 days, as little additional calcium carbonate is generated.

5. Conclusions

In this study, a series of experiments were conducted to study the use of MICP technology for reinforcing coastal clay. The physical and mechanical indexes of the MICP-treated soil at different ages were obtained. Additionally, microscopic analyses of the treated clay were performed using SEM and XRD experiments. The main conclusions obtained from the analysis are as follows:

(1)

The unconfined compressive strength of the soil strengthened using MICP increased by about 25%, and the permeability coefficient decreased by 76%. The shear strength of the soil treated using MICP increased, with its internal friction angle remaining unchanged at 22°, and the cohesive force increased from 40 kPa to 60 kPa. The soil treated using MICP was less compressible, with its compression index decreasing from 0.225 to 0.172;

(2)

The results of the mechanical and XRD experiments show that the process of microbial-induced calcium carbonate formation, which reinforces the soil, mainly occurred in the first 14 days;

(3)

The calcium carbonate crystals between reinforced clay aggregates were relatively small, primarily serving a bonding role and having little effect on the internal friction angle of the soil. The XRD experiment results also show that microorganisms not only induce the Ca²⁺ provided in the consolidating fluid to produce calcium carbonate but also promote the production of calcium carbonate from the Ca²⁺ originally contained in the soil.