Factors Affecting Soybean Crude Urease Extraction and Biocementation via Enzyme-Induced Carbonate Precipitation (EICP) for Soil Improvement

Abstract

Enzyme-induced carbonate precipitation (EICP) is a new biogeotechnical ground improvement technique that uses calcium carbonate (CaCO₃) formed by biochemical processes to increase soil strength and stiffness. In this paper, crude urease extracted from soybeans was employed to catalyze the precipitation of CaCO₃ in sand. To optimize the urease extraction efficiency, factors affecting the soybean crude urease extraction, including the powdered soybean particle size, concentration, soaking time, and soaking temperature, were addressed. This paper also provided further insight regarding the impact of the urease activity of soybean crude extract on the chemical conversion efficiency and the biocementation performance in EICP. The findings revealed that the powdered soybean concentration and the particle size were the two most important factors affecting the urease activity of the soybean crude extract. The enzyme activity utilized in the EICP process might further lead to different reactant efficiencies of urea-CaCl₂ solution, and consequently, the improvement in the physical and mechanical properties of biocemented sand. Considering the chemical conversion efficiency and the biocementation performance, 60 g/L of powdered soybean was concluded as the preferred quantity for extracting the crude urease, with an enzyme activity of 6.62 mM urea min⁻¹. Under this condition, a chemical conversion efficiency of approximately 95% for 0.5 M urea-0.5 M CaCl₂ could be obtained in merely 12 h, and the unconfined compressive strength (UCS) of the EICP-treated sand exceeded 4 MPa with a CaCO₃ content of ~8%. As a high-efficient cost-effective alternative to the purified enzyme for carbonate precipitation, the soybean crude urease showed great potential for ground improvement.

1. Introduction

Biocementation is an innovative soil-improvement method for geotechnical and environmental engineering applications [1,2,3,4,5,6,7]. The most effective forms of biocementation, including enzyme-induced carbonate precipitation (EICP) and microbially induced carbonate precipitation (MICP), are achieved via urease-catalyzed urea hydrolysis, whereby soluble calcium (Ca²⁺) is converted into solid calcium carbonate (CaCO₃) crystals that cement individual soil particles together, leading to increased soil strength and stiffness [8,9,10]. The biochemical process can be described by the following equations:

$$\mathrm{CO}{({\mathrm{NH}}_2)}_2+2{\mathrm{H}}_2\mathrm{O}\stackrel{\mathrm{Urease}}{\to }2{\mathrm{NH}}_4{}^{+}+{\mathrm{CO}}_3{}^{2-}$$

(1)

$${\mathrm{Ca}}^{2+}+{\mathrm{CO}}_3{}^{2-}\stackrel{}{\to }{\mathrm{CaCO}}_3\left(\mathrm{s}\right)$$

(2)

Urease catalyzes the hydrolysis of urea into carbonate ions and ammonium (Equation (1)). The production of ammonium results in an increase in the pH and creates an alkaline environment favorable for carbonate formation. The precipitation of CaCO₃ crystals occurs when soluble calcium sources are provided (Equation (2)). The end-product CaCO₃ crystal, a natural and non-toxic mineral, can be an excellent alternative to the traditional Portland cement, which is energy-intensive and environmentally unfriendly during production [11].

EICP is distinguished from MICP due to its use of free urease to catalyze the hydrolysis of urea (Equation (1)) rather than the intercellular urease in live bacteria. The free urease enzyme has a size of 10 nm, 2–3 orders-of-magnitude smaller than the bacterial cells, facilitating its transport in soil pores [12,13]. In addition, using free urease (EICP) instead of live bacteria (MICP) eliminates the challenges associated with controlling the microbial growth (activity) in soils [14].

Though the advantages of EICP are visible, some problems are yet to be fully addressed before EICP’s further application can be implemented. The cost of purified urease enzyme, comprising more than half of the total cost of the EICP solution, is one of the major challenges in large-scale applications [13]. The EICP process catalyzed by the purified urease enzyme lacks the nucleation sites for CaCO₃ formation, which may cause disorder in the precipitation morphology [15,16]. Additional stabilizers, such as powdered non-fat milk or foreign CaCO₃ crystals, are needed for calcium carbonate precipitation in the biochemical process as nucleation points. These further increase the costs of EICP [13,17]. Furthermore, the purified urease is relatively unstable, especially under a high temperature and extreme pH conditions, leading to the catalyst deactivation and lower reaction efficiency of EICP compared with MICP [9,15,18]. Soybean crude extract, containing an appreciable amount of urease and protein, may be an appropriate alternative to the expensive and unstable purified urease enzyme to catalyze the precipitation of CaCO₃.

This study exploited the key factors affecting the urease activity of soybean crude extract and the performance of soybean crude urease in EICP. An optimized extraction process of the soybean crude urease was proposed, including soybean smashing and soaking, the salting-out of the excess protein, and liquid-solid separation. The impact of several possible factors, including the powdered soybean particle size and concentration, soaking time, and temperature, on the enzyme activity of soybean crude urease was investigated with the aim of achieving the efficient use of soybeans for crude urease extraction. The influence of the dosage of soybean crude urease on the efficiency of carbonate precipitation and the physical-mechanical properties of Ottawa standard sand were evaluated for the optimization process. Analytical methods, including X-ray diffraction (XRD) and scanning electron microscopy (SEM), were employed to confirm CaCO₃ precipitation and characterize the microstructure development in the biocemented sand.

2. Materials and Methods

2.1. Soybean Crude Urease Preparation and Urease Activity Measurement

Soybeans were purchased from a nearby market and smashed with a grinder. The crude urease was extracted by mixing powdered soybean and tap water (pH = 7.5). The extraction process was as follows: (a) the powdered soybean was soaked in tap water and fully dispersed by stirring; (b) the mixture was preserved at different storage times and temperatures; (c) food-degraded CaSO₄·2H₂O (CaSO₄·2H₂O content ≥ 98%), as an additive to salt out excess protein, was added into the powdered soybean suspension with a dosage of 0.1 g CaSO₄·2H₂O/g powdered soybean, and the mixture was stirred for five minutes; (d) the mixed solution was centrifuged (4000 rpm, 15 min, and 4 °C) and the supernatant was collected; and (e) the insoluble substances in the supernatant were removed by filtration and the soybean crude urease was obtained without further purification. Before other experiments, the soybean crude urease was stored in a refrigerator at 4 °C.

Urease activity reflecting the urea hydrolysis rate was measured based on the change of electric conductivity of 1 M of urea at 25 °C. The increased rate of conductivity (mS/min/cm) was related to the urease activity (mM urea min⁻¹) with a ratio of 1: 11.1, according to Whiffin et al. [19]. Note that the change in electric conductivity tended to be stable after mixing the soybean crude urease and urea solution for 30 min, as shown in Figure 1, and the electric conductivity variation value after mixing for 30 min was used for calculating the urease activity. The batch tests were conducted in triplicates, and the average value (±standard deviation) was calculated and will be presented herein.

2.2. Parameters Affecting Soybean Crude Urease Extraction

To optimally use the soybeans to extract crude urease for biocalcification, a study of several factors that affect the enzymatic activity of prepared soybean crude urease was carried out, including the degree to which the soybeans are crushed (related to the powdered soybean size), the powdered soybean concentration, the soaking time of powdered soybean, and the soaking temperature. Five sets of powdered soybean samples with particle size ranges of 0–0.075 mm, 0.075–0.25 mm, 0.25–0.5 mm, 0.5–1 mm, and 1–2 mm were obtained by filtration through different sieves. The sieved powdered soybean was soaked in tap water, which diluted it to the desired powdered soybean concentrations (i.e., 20, 40, 60, 80, and 100 g/L). The soaking time was controlled from 0.5 to 24 h, and the soaking temperature varied from 5 to 50 °C.

2.3. Sand Samples’ Preparation and Testing Scheme

To investigate the effect of soybean crude urease activity on the performance of EICP in Ottawa sand, four kinds of powdered soybean concentrations were employed to extract the soybean crude urease, i.e., 20 g/L (2.15 mM urea min⁻¹), 40 g/L (4.86 mM urea min⁻¹), 60 g/L (6.62 mM urea min⁻¹), and 80 g/L (8.17 mM urea min⁻¹), as shown in Figure 2. During the extraction, the powdered soybean size was 0–0.25 mm, the soaking time of powdered soybean was 0.5 h, and the soaking temperature was around 25 °C. The cementation solution (CS) contained 1 mol/L urea and 1 mol/L calcium chloride.

Poorly-graded Ottawa sand [20] with a mean size of 0.36 mm was used throughout this study.

Table 1. Properties of sand used in this study.

Name	D10 (mm)	D50 (mm)	Gs (g/cm3)	emax	emin	Dr%
Ottawa sand	0.26	0.36	2.65	0.735	0.467	45

summarizes the physical properties of the tested sand. Polyvinyl chloride (PVC) cylindrical molds were employed to make the sand specimens (diameter = 50 mm, length = 100 mm). Dry sand was placed into each mold at a relative density of ~45% by pluviation. Each column was positioned vertically with top and bottom open. A low-temperature premixed percolation method was used in this study to bio-stabilize the sand. Biological reagents (soybean crude urease and cementation solution) with a low temperature (4 °C) were firstly premixed and then immediately introduced to the sand by surface percolation (from the top). The infiltration of liquid was due to the gravity and capillary forces. The low-temperature premixed method’s key feature was that the urease activity could be temporarily suppressed (see Figure 3) during the infiltration phase, thus inhibiting the fast accumulation of CaCO₃ precipitation in the surficial soil. Under surface percolation (fully drained) conditions, the water retention capacity of Ottawa sand in the columns was ~72 mL. The retained water was characterized as capillary water and adsorbed water [21]. The low-temperature premixed percolation process consisted of the following four steps:

• A total of 72 mL tap water (~25 °C) was introduced into each column to provide moist conditions for the follow-up EICP treatment;
• A total of 36 mL soybean crude urease (4 °C) and 36 mL cementation solution (4 °C) were mixed adequately and then poured immediately onto the sand surface;
• The sand samples were incubated at room temperature (~25 °C) for the biochemical process to occur for 12 h;
• Steps (2) and (3) were performed several times (2–8) to obtain different cementation levels of the soil samples.

After the last EICP treatment, 500 mL of tap water was flushed through each sand column from the top to remove the residual unreacted CaCl₂, urea and the by-product NH₄Cl in the sand samples. During the experiments (including the washing stage), the effluents collected from the outlet were used for calcium content measurement.

2.4. Chemical Conversion Efficiency, Total Calcium Carbonate Content, and Calcium Carbonate Distribution in Sand

The chemical conversion efficiency of cementation solution during the EICP process was calculated as the ratio of the chemical reactant’s (CaCl₂) mass that took part in the biochemical reaction (Equation (2)) to the total amount of CaCl₂ inputs at the endpoint of each test. The quantity of calcium involved was calculated as the difference between the amount of calcium introduced into sand samples and the calcium content in the effluent container. Excessive HCl solution (2 M) was added to each effluent container to dissolve the possible precipitated carbonates. Then, the calcium content was determined using the EDTA titrimetric method [22].

The total content of CaCO₃ in the sand samples was determined by the quantity of Ca²⁺ participating in the biochemical reaction (as mentioned previously). After the sand specimens were broken during the unconfined compressive test, sand fragments (~10 g each) from the top, middle, and bottom of each sample were collected for calcium carbonate content measurement using the acid-washing method [11].

2.5. Water Permeability Tests and Unconfined Compressive Tests

Water permeability tests were carried out using the constant head permeability method following Cheng et al. [23]. The permeability tests were conducted inside the PVC column, where the sand specimens were kept undisturbed. After the permeability tests, the biocemented samples were oven-dried for 48 h at 50 °C. Then, the unconfined compressive tests were conducted. The elastic modulus, E50, was determined as the slope of secant drawn from the origin of the stress–strain curve through the point at 50% of peak stress [24].

2.6. Microscale Identification Analysis and Data Analysis

X-ray diffraction (XRD) analysis was employed to identify the crystalline phases of the carbonate precipitates. The oven-dried sand samples (after the unconfined compressive tests) were crushed for XRD analysis using a high-resolution X-ray diffractometer (PANalytical XPert PRO, The Netherlands). Scanning Electron Microscopy (SEM) analysis was performed to examine microstructure development and the morphology of precipitates in the biocemented sand. Several small pieces of sand samples collected after the unconfined compressive tests were oven-dried, sputter-coated with gold, and examined using a Hitachi S4800 Field Emission Scanning Electron Microscope (FE-SEM), Hitachi Ltd., Tokyo, Japan.

The one-way analysis of variation (ANOVA) (P = 0.05) method was used to compare the mean values of experiment result. P is the probability of the outcome in a statistical experiment.

3. Results and Discussion

3.1. Factors Affecting Soybean Crude Urease Extraction

3.1.1. Powdered Soybean Size and Concentration

To investigate the effect of the powdered soybean size and concentration on soybean crude urease enzyme activity, the soaking time and the temperature were set as 0.5 h and 25 °C, respectively. These measures were taken to make efficient use of the soybeans while extracting crude urease. It is evident from Figure 4 that both the size and the concentration of the powdered soybean had a significant impact on the soybean crude urease activity. For all the powdered soybean sizes considered in this study, the soybean crude urease activity increased with the increase in the powdered soybean concentration. Therefore, the amount of soybean crude urease used in the EICP could be quantified according to the enzyme activity and the powdered soybean. In addition, the urease activity in the soybean crude extract increased with increasing the degree of crushings of the soybean for a given powdered soybean concentration. For instance, when the powdered soybean concentration was 40 g/L, the urease activity in the soybean crude extract prepared using powdered soybean with a particle size range of 0.075–0.25 mm was 4.93 mM urea min⁻¹, about 1.67 times higher than that of 1–2 mm. A decreased powdered soybean size resulted in a higher increasing rate of urease activity. However, the two ranges of the powdered soybean size (i.e., 0.075–0.25 mm and 0–0.075 mm) yielded statistically similar results (p > 0.05) in terms of soybean crude urease activity. This similarity might be attributed to the rough extraction method adopted in this study. The soybean urease was adequately extracted with the powdered soybean particle size of 0.075–0.25 mm; therefore, crushing the particle size to the range of 0–0.075 mm could not further improve the urease activity.

It is commonly accepted that the higher the urease activity, the faster the rate of the biochemical reactions become and the more the CaCO₃ precipitates [13,25]. During the EICP process, the formerly-precipitated CaCO₃ might act as nucleation seeds for calcite precipitation, facilitating the subsequently-precipitated CaCO₃ to agglomerate and form large calcite clusters at the contact areas between the sand particles [26]. The large calcite clusters (as it is known for effective CaCO₃) that accumulated within and progressively filled the interstitial gaps of the inter-particles bind separate sand particles together, thereby improving the already strong bonding force to demonstrate an even higher strength [1,23,25,27]. Consequently, more CaCO₃ precipitation in porous materials significantly contributed to improving the mechanical properties such as shear strength and erosion resistance. Considering the urease activity and the cost among all the degrees of crushing of the soybeans, the optimal particle size of the powdered soybean was 0–0.25 mm. Unless particularly stated, the particle size of the powdered soybean used hereafter was 0–0.25 mm. In addition to the degree of crushing, the urease activity in the soybean crude extract could be improved by substantially increasing the dosage of the powdered soybean, which would dramatically enhance the total cost of biocalcification.

3.1.2. Soaking Time and Temperature

For the application of EICP technology in large-scale fields, achieving the economical production of urease in natural environments is of great significance. To achieve this goal, a simple and efficient extraction strategy (i.e., under an such as appropriate soaking time and temperature) is needed. Figure 5 plots the effect of the soaking time and temperature on the urease activity of soybean crude extract. The powdered soybean size was within the range of 0–0.25 and the powdered soybean concentration was set to be 40 g/L. It is indicated from Figure 5 that the enzyme activity of the soybean crude urease extracted at the temperatures of 5, 15, and 25 °C appeared broadly similar (p > 0.05). As the soaking temperature increased to 40 and 50 °C, the urease activity in the soybean crude extract kept nearly constant in the first 5 h and then sharply declined. The reduction in the urease activity should be due to the deterioration of protein and the inactivation of enzyme at relatively high temperatures. The findings revealed a strong correlation between urease activity and temperature, and high temperatures might impede urease activity. Dilrukshi et al. [28] measured the urease activity of crude extracts from watermelon seeds at various temperatures (from 25 to 70 °C). They claimed that the activity of the crude extract decreased with time and that the decay rate was more pronounced at high temperatures. Similar observations regarding the effect of environmental temperature on the biocementation process can also be found in Cheng et al. [23] and Sun et al. [29].

The soaking time had no impact on the urease activity in the soybean crude extract at room temperature, implying the feasibility of the soybean crude urease obtained in this work for potential engineering applications. Compared with the urease-producing bacteria (UPB) cultivation, the production of urease with soybean crude urease was more time-saving (less than 1 h in this study) and less sensitive to seasonal (temperature) variations for the biomineralization of CaCO₃. Based on the results in Figure 5, the soaking temperature and the soaking time of 25 °C and 0.5 h were suggested and adopted in the remainder of this study.

3.2. Effect of Powdered Soybean Concentration (Urease Activity) on Chemical Conversion Efficiency and Calcium Carbonate Content of EICP-Treated Sand

The bio-catalyzed precipitate of CaCO₃ accumulates as more EICP solutions (soybean crude urease and cementation solution) are introduced to the soil. To efficiently use the cementation solutions for CaCO₃ precipitation, a high chemical conversion efficiency is required in the soybean crude urease-induced biochemical process. Figure 6a displays that an increase in the powdered soybean concentration (urease activity) increased the efficiency of the biological reaction. Since a larger amount of urea was hydrolyzed, a greater mass of calcium was bio-transformed into CaCO₃. In fact, during the first two treatments (12 h for one treatment), a powdered soybean concentration of 60 g/L (6.62 mM urea min⁻¹ before premixing) enabled the production of almost 95% of the theoretical mass of calcium carbonate for 0.5 M urea-0.5 M CaCl₂ (after premixing) within 12 h. However, for the same cementation solution, the use of 40 g/L (4.86 mM urea min⁻¹ before premixing) and 20 g/L (2.15 mM urea min⁻¹ before premixing) powdered soybean concentration reduced the average chemical conversion efficiency to 41.2% and 70.9% (in the first two treatments), respectively. The chemical conversion efficiency declined with the increase in the number of EICP treatment cycles when the powdered soybean concentrations were 60 and 80 g/L. The EICP solution retention capacity (initially 72 mL) of Ottawa sand in the columns might decrease due to the accumulation of CaCO₃ crystals in the pores and the formation of preferential flow paths, which in turn, resulted in the leakage of excess biological reagents and thereby the decrease in the total chemical conversion efficiency. The increased chemical conversion efficiency with the augmentation of the EICP treatment for lower powdered soybean concentrations (40 and 20 g/L) might be due to the accumulation of urease residues in the soil from the former treatments.

The powdered soybean concentration affected the chemical conversion efficiency of the cementation solution and the content of CaCO₃ in the EICP-treated soil samples (Figure 6b). To improve the chemical conversion efficiency and reduce the waste of CaCl₂ and urea, a lower urea-CaCl₂ solution concentration or a longer retention time should be adopted for low powdered soybean concentrations (40 and 20 g/L). However, more EICP solution was required to obtain the target level of cementation at a lower concentration of cementation solution, further increasing the total cost of EICP. Moreover, a longer retention time could further increase the time cost of the EICP. The statistical analysis revealed that the overall biochemical reactants’ efficiency and the total content of CaCO₃ in the EICP-treated soil samples were similar for the powdered soybean concentration of 60 and 80 g/L (p > 0.05). From the aspects of the reactants’ efficiency and application cost, 60 g/L of the powdered soybean was preferred for extracting the crude urease in this study.

It has been reported that below a specific urea-CaCl₂ input rate (0.042 mol/L/h) and at a bacterial optical density (OD₆₀₀) between 0.8 and 1.2, the reaction efficiency of MICP is higher than 90% (Al Qabany et al., 2012). In this study, with a urea-CaCl₂ input rate of ~0.042 mol/L/h (retention time: 12 h, urea-CaCl₂ concentration after premixing: ~0.5 M) and a urease activity of ~3.31 mM urea min⁻¹ (after premixing), the reactant efficiency of the soybean crude urease-induced CaCO₃ precipitation was comparable to those obtained from MICP treatment [30]. A review of the literature shows that the chemical conversion efficiency of EICP varied among different research groups. Yasuhara et al. [31] premixed urease powder with sand before injecting cementation solution and found that the chemical conversion efficiency ranged from 30 to 60%. Almajed et al. [13] reported that when a treatment solution consisting of 1 M urea, 0.67 M calcium chloride, and 3 g/L enzyme (3500 U/g) was adopted to biocement sand, the chemical conversion efficiency of EICP (after curing for 7 days) in the soil fell in the range of 70–95%. Carmona et al. [32] conducted a series of test-tube (aqueous environments) experiments. They found that with a curing time of 24 h and urease activity of 4000 U/L, the reactant efficiency was less than 50% for 0.5 M urea and 0.5 M CaCl₂ solution. The chemical conversion efficiency of soybean crude urease-induced CaCO₃ precipitation in this study (60 g/L powdered soybean with urease activity of ~3.31 mM urea min⁻¹ after premixing) was higher than those obtained from purified enzyme-induced CaCO₃ precipitation within a short retention time (12 h). Two factors accounted for this phenomenon: firstly, bacteria act as the nucleation sites for the formation of CaCO₃ crystals through their adsorption of Ca²⁺ and by creating localized supersaturation, whereas using purified enzymes may lead to a lack of nucleation points for the CaCO₃ precipitation; secondly, the purified urease is unstable and can easily be affected by high temperatures, extreme pH, and other environmental factors, limiting their applications in field conditions [15,18].

Soybean crude urease (without further purification) contains an appreciable amount of urease and soybean protein. The soybean protein had an isoelectric point of around pH 4.5, as measured by an isoelectric focusing analysis. The isoelectric point is the pH value at which the zeta potential value is zero, implying that there is no electric charge on the surface of the soybean protein. The net charge on the protein molecule is affected by the pH of its surrounding environment. It becomes negatively charged when the surrounding environment’s pH exceeds the isoelectric point [33]. During the soybean crude urease-based biochemical process, the pH of the EICP solution was much higher than the isoelectric point of soybean protein (~pH 4.5). The negatively charged soybean proteins (similar to the urease-producing bacteria) may also act as nucleation sites through the adsorption of Ca²⁺ to their surfaces, creating a localized supersaturation that favored the precipitation of CaCO₃. Furthermore, the macromolecular soybean proteins acted as an organic stabilizer to protect the urease against environmental change, which was similar to the commonly used dried non-fat milk [17].

3.3. Effect of Powdered Soybean Concentration (Urease Activity) on the Properties of Biocemented Sand

3.3.1. UCS and Elastic Modulus

The influence of the treatment cycle and the powdered soybean concentration (urease activity) on the UCS of the biocemented soil samples is provided in Figure 7. In general, the value of the UCS increased with the reduplicative percolation of the EICP solution, as illustrated in Figure 7a. In addition to the treatment times, the powdered soybean concentration (urease activity) was another crucial factor impacting the mechanical properties of the biocemented sand specimens (Figure 7a). For the same treatment cycles, the UCS of sand increased significantly with the increase in the powdered soybean concentration from 20 to 60 g/L and then remained roughly constant even when a higher powdered soybean concentration (80 g/L) was employed. A statistical analysis of the results indicated that the UCS was similar (p > 0.05) when the powdered soybean concentration was 60 and 80 g/L. The phenomena mentioned above could be related to the precipitated CaCO₃ (Figure 7b). For all the powdered soybean concentrations, the UCS increased with the increase in the CaCO₃ amount, i.e., a more considerable amount of bio-mediated CaCO₃ increased the bonding strength between soil particles, leading to an improvement in the mechanical properties (UCS) of the biocemented sand samples.

Figure 8 compares the present and previously reported relationship between the UCS and the biocatalyzed CaCO₃ content, including the present EICP results obtained using soybean crude urease and the previously reported EICP (using purified urease) and MICP (using urease producing bacteria) results. The CaCO₃ content refers to the mass of generated CaCO₃ to the mass of sand. The trend of the UCS in relation to the CaCO₃ content recorded in this work was comparable to that obtained from MICP treatment [23,34] and EICP treatment [13,35]. However, the UCS value obtained in the present work was at a noticeably higher range than that of EICP- and MICP-treated data in terms of comparable CaCO₃ precipitation levels. It is worth noting that the overall amount of CaCO₃ precipitation could not be considered as the sole factor for the biocementation effects [36,37,38]. Wang et al. [27] pointed out that the precipitated CaCO₃ could be categorized into effective CaCO₃ and non-effective CaCO₃ in terms of the different nucleation sites. The effective CaCO₃ may act as solid bridges to bind separate sand particles together, thereby improving strength and stiffness [23]. In addition, the influence of other factors (e.g., crystal polymorphs) may either influence the biocementation effect even at an identical CaCO₃ content. Dyer and Viganotti [39] suggested that the crystal morphology dramatically affected the failure mechanism of MICP-treated samples due to the different geometrical interferences. Consequently, in consideration of the abovementioned factors, a microscale and microstructure analysis should be conducted for further investigation.

To investigate the possible explanation for the abovementioned difference, the crystalline phase and microfeatures of the precipitate within the soil matrix were examined using an X-ray diffractometer (XRD) and scanning electron microscopy (SEM), respectively. The XRD analysis (Figure 9) shows that quartz and calcite existed in the soybean crude urease-based biocemented sand, while no calcite formation was observed in the original sand. Calcite may be the most desirable crystal of biocatalyzed CaCO₃ for geotechnical applications due to its thermodynamic stability [15]. A similar CaCO₃ crystal type was observed in other enzyme-induced calcium carbonate precipitation studies [13,32], suggesting that soybean crude urease can be a good alternative to the purified urease for catalyzing the precipitation of calcite for ground improvement. The small precipitated crystals (typical crystal size of 1–3 μm) were agglomerated to form large clusters with an approximate size range of 20–100 μm in the soybean crude urease (60 g/L powdered soybean) based biocemented sand. One notable feature of this deposition pattern was that although some of the large crystal clusters precipitated on the grain surface, the gaps between the sand grains were entirely filled with clustered crystals (Figure 10), which could bridge the adjacent sand grains and provide a strong enough bonding force to bear the high shear strength [23,40,41].

After eight cycles of treatment, the typical stress–strain curves of the biocemented soil samples treated with different powdered soybean concentrations were obtained and are presented in Figure 11a. All the EICP-treated sand samples exhibited a brittle nature comparable to those observed in the MICP [11,34] and EICP [35,42] treatments. During the unconfined compressive tests, failures tended to occur within the whole core (Figure 11b). This result was in line with the relatively uniform distribution of CaCO₃ precipitation in the sand columns (Figure 12). A low-temperature premixed percolation strategy was employed in this work. Due to the low urease activity of the soybean crude extract at 4 °C, a period of lag phase existed in the percolation phase of the EICP solution, which ensured the relatively uniform distribution of the premixed biological agents within the sand column (10 cm) and consequently the uniformity of the CaCO₃ precipitation.

The elastic modulus (E50), defined as the ratio of 0.5 UCS and the corresponding axial strain, was used to quantify the stiffness of the biocemented soil [24,40,43]. The E50 of the EICP-treated sand samples ranged from 18.47 to 200.96 MPa, according to the powdered soybean concentration and treatment cycle. For example, after eight cycles of biocementation treatment, the E50 increased from 87.27 to 195.12 MPa (i.e., 2.24 times) while the powdered soybean concentration rose from 20 to 60 g/L (Figure 13a). As evident in Figure 13b, the E50 increased with the total CaCO₃ content in the EICP-treated samples, which was comparable to that observed in the previous research [24,34]. As mentioned earlier, the results indicated that the strength and stiffness of the EICP-treated sand exhibited a similar trend regarding the treatment cycle and powdered soybean concentration.

Biomineralization is a highly sophisticated process because of the involvement of nucleation sites and the precipitation of crystals. For example, biomineral-associated macromolecules, such as proteins and proteoglycans, may initiate and stabilize non-equilibrium crystal polymorphs and morphologies through interactions between anionic moieties and cations in solution or at the minerals’ surfaces [44]. Therefore, a further investigation of this complex biocatalyzed process to find the possible reasons for forming the peculiar deposition pattern of calcite is worth carrying out in the future based on the current study. It should be emphasized that the UCS and E50 values from the ureolysis-based biocementation-treated soil can be highly variable, depending on diverse experimental factors. The results of experiments in this area can be highly variable due to the various properties of biocementation-treated soil such as relative density [45], particle shape [46], and particle size [34]. In Zhao et al. [47], the UCS value of MICP-treated sand with a greater d50 was much higher compared with those with a smaller d50 under an identical treatment strategy. Regarding the gradation, Mahawish et al. [48] demonstrated that the biocementation-treated column samples with the lower percentage of fine materials (25%) showed a more uniform distribution of CaCO₃ precipitation along its length compared to the samples, which contained more fines and a gap-graded distribution led to higher UCS values of the biocementation-treated soil. In addition, biological agent compositions such as sources of urease [49], sources of calcium [50], and urea-CaCl₂ concentration [45] had a significant impact on the biocementation process. Mujah et al. [51] pointed out that MICP-treated sand samples with lower cementitious solution concentrations possessed higher UCS values at a given CaCO₃ content. Nevertheless, Zhao et al. [47] stated an opposite observation between the UCS value and the cementitious solution concentration: the higher the cementation concentration, the higher the UCS value. The influence of environmental factors (e.g., pH and temperature) on the performance of urelysis-based biocemented soil cannot be ignored either [29,52]. In conclusion, much more effort should be made to provide an insight into the optimization of EICP based on soybeans’ crude urease for various possible applications.

3.3.2. Permeability

The measured permeability values of the sand samples after EICP treatment were firstly normalized by the permeability value of the original sand. As expected, permeability reduction appeared with the advancement of the EICP treatments (Figure 14a). The permeability was reduced from 2.4 × 10⁻⁴ m/s for untreated sand to 1.3 × 10⁻⁵ m/s for biocemented sand (80 g/L) after eight cycles of treatment. Similar outcomes were reported by [23], wherein the permeability of the biocementation treated sand decreased with the increase in the content of CaCO₃ precipitation. Figure 14b indicated that although the reduction in permeability of the EICP-treated sand samples was mainly controlled by the CaCO₃ precipitation content, the concentration of the powdered soybean (urease activity) used in the EICP treatment had a non-negligible impact. For the same amount of CaCO₃ precipitate, the permeability reduction of biostabilized sand treated with higher powdered soybean concentrations (for example, 60 and 80 g/L) was more than that of the sand treated with lower powdered soybean concentrations (for example, 20 and 40 g/L). The higher urease activity (60 and 80 g/L) may cause the rapid precipitation of CaCO₃ in the top area of the sand samples, which would block the soil pores and lead to clogging during the EICP treatment [53].

3.4. Implications of Soybeans Crude Urease for Ground-Improvement Applications

The application of biocementation for ground stabilization requires a great degree of cementing of the calcium carbonate between the soil grains to produce an adequate improvement in strength. The cost of urease-producing bacteria or urease enzymes is one of the main obstacles in adopting the bio-mediated CaCO₃ technique in large-scale practice. For instance, nutrients in the culture medium (urease-producing bacteria) are expensive and can account for as high as 60% of the total operation cost of MICP [54]. In contrast, the cost of commercially available urease enzyme comprises 57–98% of the total cost of the EICP treatment [13]. Thus, low-cost urease sources are highly desirable for the large-scale application of biocementation. Soybeans can be a cost-effective alternative source of urease to catalyze the precipitation of CaCO₃ crystals. It is also crucial to produce greater amounts of crude urease from soybeans and to make efficient use of the enzyme for cost minimization. The findings obtained from the present work show that powdered soybean concentration and particle size are the two crucial factors affecting the urease activity of soybean crude extract. The enzyme activity used in EICP process can further result in different reactant efficiencies of urea-CaCl₂ solution, and consequently, the improvement in the physical and mechanical property of biocemented sand. For ground improvement purposes, 60 g/L of the powdered soybean was preferred for extracting the crude urease with an enzyme activity of 6.62 mM urea min⁻¹, leading to a chemical conversation efficiency of approximately 95% for 0.5 M urea-0.5 M CaCl₂ over 12 h. In this study, considering the current price of soybeans, the material cost of the crude urease enzyme (powdered soybean concentration: 60 g/L) was about USD 0.036/L⁻¹, which was much lower than that of the commonly used Sporosarcina pasteurii and purified urease enzyme. Furthermore, the cost can be further reduced if the remaining soybean dregs after extracting the crude urease are used to produce soya meal, which is rich in protein and can be used to feed animals or culture microorganisms. This work uses the low temperature premixed percolation method, which can be used wherever the surficial soil needs to be treated, such as strengthening the surface of slopes, preventing shallow ground liquefaction, and controlling wind or water induced erosion. Although the study demonstrates the feasibility of using soybean crude urease for sand consolidation in short columns, its meter-scale performance is still unknown. In addition, the soil conditions, such as the grain size and mineral composition, vary considerably in different regions, which may have a non-negligible impact on the ground reinforcement of EICP. Furthermore, the effects of the residual unreacted CaCl₂ and urea as well as the by-product of NH₄Cl on the soil microenvironment remain to be seen. Many more efforts must be made to address the practical problems mentioned above, which are critical for the large-scale application of EICP.

4. Conclusions

This study placed emphasis on the key factors affecting the urease activity of soybean crude extract and the performance of soybean crude urease in EICP. Soybean crude urease was extracted simply by mixing powdered soybean and tap water, and a modified method was proposed to optimize the urease extraction efficiency. The powdered soybean concentration and particle size were the two critical factors affecting the urease activity of soybean crude extract. To make the most of soybeans in terms of crude urease extraction, the powdered soybean particle size should be less than 0.25 mm. Under this condition, soybean crude urease can be fully extracted at room temperature in less than 1 h.

The soybean power concentration can be used as an indicator of the extracted soybean crude urease activity due to the positive correlation. A series of sand column tests were performed to investigate the effects of soybean power concentration on the chemical conversion efficiency and mechanical properties of the biocemented ASTM-graded Ottawa sand. When the soybean power concentration increased to 60 g/L (6.62 mM urea min⁻¹), a chemical conversion efficiency of approximately 95% for 0.5 M urea-0.5 CaCl₂ could be obtained in merely 12 h. Furthermore, with a CaCO₃ content of ~8% (after eight cycles of treatment), the UCS and E50 of EICP-treated sand exceeded 4 and 190 MPa, respectively.

The formation of large, agglomerated calcite crystal clusters filled the gaps between the sand grains, showing the high efficiency of soybean crude urease in improving the strength and stiffness of EICP-treated sands. This study indicates that soybean crude urease, as a cost-effective and more efficient alternative to the purified enzyme and urease-producing bacteria, can be used to catalyze the precipitation of calcite in sands for ground improvement.