Optimization of Microbial-Induced Carbonate Precipitation Parameters for Strength, Durability, and Environmental Safety of Phosphogypsum Road Base Materials

TL;DR

This study explores how to optimize a microbial process to strengthen and stabilize road materials made from phosphogypsum, while ensuring environmental safety.

Contribution

The study identifies optimal parameters for microbial-induced carbonate precipitation to enhance the strength and durability of phosphogypsum road base materials.

Findings

Optimal mechanical strength was achieved with 2.0 mol/L cementation solution and a 2:1 bacterial/cementation solution ratio.

MICP improves durability by coating phosphogypsum particles and filling voids with calcium carbonate.

Toxic leaching of F− and PO43− was significantly reduced in treated mixtures.

Abstract

This study investigates the mechanical properties, moisture stability, and environmental safety of microbial-induced carbonate precipitation (MICP)-treated phosphogypsum (PG)-based mixtures (MPGT) for road base utilization. Optimal cementation solution concentrations and bacterial-to-cementation solution ratios were determined via unconfined compressive strength (UCS), California bearing ratio (CBR), and splitting tensile strength tests. Durability was compared with untreated mixtures, and enhancement mechanisms were analyzed using XRD, SEM, and FTIR. Additionally, toxicity leaching tests evaluated environmental safety. Results indicated optimal parameters of 2.0 mol/L cementation solution and a 2:1 bacterial/cementation solution ratio for maximum mechanical strength. Under these conditions, MPGT durability significantly improved compared to untreated mixtures. Mechanism analysis revealed that MICP-generated calcium carbonate coats PG particles and fills voids, enhancing strength and durability. Furthermore, F− and PO43− leaching concentrations were significantly reduced. In summary, MICP improves the mechanical performance, durability, and environmental safety of PG-based mixtures, promoting PG recycling in road engineering.

Figures

Full text

1. Introduction

Phosphogypsum (PG) is the primary solid waste generated by the phosphate fertilizer industry as a byproduct of wet-process phosphoric acid production [1]. Global statistics indicate that the annual total amount of PG increases by approximately 300 million tons and reached 6 billion tons in 2024 [2]. This substantial accumulation of PG would occupy vast land resources and poses considerable challenges to society and environment [3,4]. Researchers have carried out a series of studies on recycling PG solid waste to mitigate its adverse effects. These applications include utilization in agriculture (e.g., soil amendment), building materials (e.g., gypsum board, plaster), and cement admixture (e.g., retarder, mineralizer) [5,6,7,8,9,10]. However, the current recycling rate of PG is relatively low, accounting for only 15% of the total amount. Therefore, it is imperative to explore alternative utilization methods for recycling or disposing of large quantities of PG.

It is widely recognized that pavement engineering utilizes a substantial quantity of natural mineral materials and binding materials, e.g., aggregate, cement, and lime. Researchers have demonstrated the feasibility of using PG in pavement construction to increase its recycling rate. Recent studies have highlighted the potential of PG as a sustainable alternative to natural mineral aggregates in road base applications. For instance, a cement-stabilized PG base material was successfully developed with enhanced mechanical performance, while the recycling of PG into high-performance cementitious materials was optimized through mechanistic analysis and preparing the PG whisker to enhance the pavement performance of asphalt binders [11,12,13]. Given that mineral particles account for over 90% of the weight of pavement construction material, the utilization of PG as an alternative material in the road subgrade layer and base course can significantly improve the recycling rate of PG waste [14]. However, untreated PG exhibits low strength, poor water stability and volume expansion, which is attributed to the formation of ettringite and secondary gypsum [15]. Given that the performance requirements for pavement subgrades are less stringent than those for base courses, the majority of existing research focuses on the application of PG particle as subgrade filling material.

To further facilitate the recycling of PG in road engineering, it is also needed to enhance the performance and durability of PG-based mixture for overcoming its technical challenges in order to meet the current performance requirements of road base course. Previous studies have demonstrated that stabilizers such as cement, lime, and industrial solid wastes, can be utilized to improve the unconfined compressive strength (UCS) and California bearing ratio (CBR) of PG-based mixture [16,17,18,19,20]. These findings confirmed that additives can be used to enhance the mechanical strength of PG-based mixture, thereby withstanding the actual traffic loads.

Nevertheless, it should be noted that there is a risk of PG-based mixture resulting in environmental pollution during its utilization in road engineering. Generally, the PG material contains various contaminants, including phosphates, fluorides, and trace heavy metals, which were originated from the phosphate rock processing [21]. Additionally, the acidic nature (pH 2–4) of PG material enhances the solubility of heavy metals under rainfall or groundwater infiltration [22]. Traditional enhancement techniques, such as cement or lime stabilization, primarily concentrate on mechanical strength but are unable to address the environmental problems related to the impurities of PG-based mixture [23,24]. Therefore, it is also essential to immobilize the inherent pollutants in PG-based mixture and mitigate the adverse effects on water bodies surrounding the road.

Microbial-induced carbonate precipitation (MICP) is an environmentally friendly bio-mineralization technique. Generally, ureolytic bacteria (e.g., Sporosarcina pasteurii) are used to hydrolyze urea to produce carbonate ions, which then react with calcium ions. Consequently, CaCO_3_ precipitate is generated, which can bond the particles and immobilize contaminants [25]. Due to its efficiency and adaptability, S. pasteurii has been extensively utilized in various engineering structures. For instance, in the field of building materials, it is widely applied in self-healing concrete to repair micro-cracks and enhance durability [26,27,28]. In geotechnical engineering, it has been successfully employed for soil reinforcement to improve the shear strength and prevent liquefaction [25,29].

In the context of roadway engineering, recent studies and systematic reviews have highlighted the extensive potential of ureolytic bacteria for enhancing the performance of sustainable road infrastructure [30]. For instance, Hu et al. [31] found that the mechanical strength of clayey sand-based subgrade mixture was increased by 198% using the pretreatment-mixing method of MICP. Meanwhile, the unconfined compressive strength of the graded stone base mixture was increased by 159% through surface permeation method of MICP [32]. More recently, bio-cementation has been successfully applied to stabilize aeolian sand as a green road base material, significantly improving its dynamic durability and wind erosion resistance [33].

Despite these achievements in natural aggregates, the application of MICP specifically for modifying phosphogypsum (PG)-based mixtures remains underexplored. From this point of view, it is prospective that MICP technique can be used to enhance the mechanical properties of PG-based mixture as well as mitigate the pollution problem related to PG material. Therefore, it is essential to obtain a comprehensive understanding of the effect of MICP on the performance and environmental safety of the PG-based mixture. Since the enhancement effect of the MICP technique is strongly influenced by the reaction conditions, it is also necessary to optimize the key preparation parameters—specifically the cementation solution concentration and volume ratio of bacterial solution to cementation solution—to ensure the effective stabilization of the PG-based mixture.

Despite the progress in PG utilization, traditional stabilization methods (e.g., cement, lime) primarily focus on meeting mechanical standards, often overlooking the long-term leaching risks of hazardous ions (e.g., fluoride and phosphate) inherent in PG. Furthermore, while MICP has been successfully applied to natural soils, its application to PG-based road base materials presents unique challenges due to the complex chemical environment (e.g., acidity and impurities) and the stringent durability requirements for road bases (e.g., moisture stability and erosion resistance). To bridge these gaps, this study distinguishes itself from general application studies by systematically optimizing the key MICP parameters specifically for PG mixtures. The novelty lies in establishing a rigorous link between parameter optimization and comprehensive performance evaluation—integrating mechanical strength, durability, and environmental safety—to validate the feasibility of MPGT as a sustainable green road base material.

Accordingly, the research methodology was structured into four systematic stages. Firstly, the optimal cementation solution concentration and the volume ratio of bacterial solution to cementation solution were determined based on the UCS, CBR value, and splitting tensile strength. On this basis, the durability of the MICP-treated PG-based mixture (hereinafter referred to as MPGT) was evaluated through water stability, freeze–thaw resistance, and erosion resistance tests. Subsequently, the enhancement mechanism of the MICP technique was investigated using XRD, SEM, and FTIR analyses. Finally, toxicity leaching tests were performed to strictly evaluate the environmental safety, specifically focusing on the effectiveness of immobilizing hazardous fluoride and phosphate impurities.

2. Materials and Methods

2.1. Materials

2.1.1. Phosphogypsum

The PG material, which appears as irregular blocks, is obtained from a stockpile in Dawu County, Hubei Province. The raw PG exhibits a grayish-white color with a natural moisture content of 19.5% and a pH value of 4.2. To ensure experimental consistency, the PG was dried to constant weight. The particle size distribution of the crushed PG particle was measured using a laser particle size analyzer, and the results are shown in Figure 1. The heavy compaction test was used to determine the optimum moisture content and maximum dry density of the PG mixture, and the results are shown in Figure 2. For pure PG, the maximum dry density is 1.75 g/cm^3^ and the optimum moisture content is 18.3%. It can be observed that the compaction curve displays a typical bell shape, which reflects the porous and hydrophilic characteristics of PG.

The chemical composition of PG was tested by XRF analysis, and the results are shown in Table 1. The main components of PG include sulfur (S), calcium (Ca), and silicon (Si). To clarify the mineralogical phase, the X-ray diffraction (XRD) analysis (detailed in Section XRD Phase Analysis) confirmed that the primary phase of the raw PG is gypsum dihydrate (CaSO_4_·2H_2_O) with traces of quartz. It is worth noting that although fluoride (F) is not explicitly quantified as a main oxide in XRF results, it constitutes a significant portion of the “Other” category and is a critical environmental indicator monitored in this study (Section 3.3). Additionally, the radioactivity of the raw PG was detected. The internal exposure index (I_Ra_) and external exposure index (I_γ_) were 0.18 and 0.25, respectively, which comply with the requirements for Class A materials specified in the standard GB 6566-2010 [34].

2.1.2. Bacterial Solution

The mineralization bacteria used in this study is S. pasteurii (ATCC 11859), which was obtained from the American Type Culture Collection (ATCC). The strain used was preserved in the laboratory after being activated from the freeze-dried powder. S. pasteurii is a Gram-positive, heterotrophic bacterium. Its cell surface carries a negative charge.

2.1.3. Preparation of Microbial Solutions and PG-Based Mixtures

The culture medium used for strain cultivation is a liquid medium. Based on the optimized parameters from our previous study [31], the medium consists of urea (serving as the nitrogen source), sodium chloride (maintaining cell osmotic pressure and acting as an inorganic salt), peptone (a nutrient substance), nickel chloride (a trace element), and sodium hydroxide (for adjusting the pH of the culture solution to 7–9).

The cementation solution used for the microbial induced carbonate precipitation (MICP) reaction comprises urea for supplying the carbonate ion (CO_3_^2−^) and calcium chloride for supplying the calcium ion (Ca^2+^) required for the mineralization reaction. Consistent with methodologies reported in relevant studies [35,36] and optimization results from previous research in our laboratory [37], the cementation solution was prepared with an equimolar ratio (1:1) of urea to calcium chloride. Therefore, the “concentration of cementation solution” refers to the molar concentration of these reactants (ranging from 1.0 to 3.0 mol/L). The specific compositions of the culture medium and cementation solution, along with the detailed experimental mixture design, are comprehensively summarized in Table 2.

The entire preparation process of MPGT is illustrated in Figure 3, which can be divided into four steps.

Step 1: The raw materials for the liquid culture medium were prepared based on the composition shown in Figure 3 and Table 2. Specifically, urea, NaCl, NiCl_2_, soy peptone, and pancreatic peptone were accurately weighed and dissolved. The pH was adjusted using NaOH. Simultaneously, the cementation solution was prepared by dissolving urea and calcium chloride in deionized water at an equimolar ratio (1:1). For instance, a 2.0 mol/L cementation solution contained 2.0 mol/L urea and 2.0 mol/L CaCl_2_.

Step 2: The liquid medium was first sterilized in an autoclave at 120 °C for 30 min. After cooling, the S. pasteurii strain was inoculated into the medium at a 5% ratio under sterile conditions. The culture was then incubated in a shaking incubator at 30 °C and 180 rpm for approximately 24 h. The bacterial concentration was determined by measuring the optical density at 600 nm (OD_600_) using a spectrophotometer (721G, Shanghai Yidian Analytical Instrument Co., Ltd., Shanghai, China). To ensure measurement accuracy, if the measured OD value exceeded 0.8, the bacterial suspension was diluted with the liquid medium to fall within the linear range of 0.2–0.8. The actual OD_600_ was then calculated by multiplying the measured value by the dilution factor. Crucially, prior to the preparation of MPGT specimens, the bacterial solution for all groups was strictly adjusted to a standardized OD_600_ of 0.8 and a urease activity of approximately 1.2 mS/cm/min (at 30 °C). This standardization ensures consistent catalytic capability across all experimental groups.

Step 3: The PG powder was thoroughly mixed with the prepared bacterial solution and cementation solution to ensure a homogeneous distribution. Crucially, the total volume of the liquid phase (bacterial solution + cementation solution) was strictly controlled to match the optimum moisture content (18.3%) of the PG. This strict control ensures that all specimens are compacted under the same moisture condition to achieve the maximum dry density.

Step 4: The mixture was placed into a cylindrical mold and subjected to static compaction to the targeted maximum dry density. The prepared MPGT specimens were then demolded and subjected to standard curing before testing.

For comparison, the untreated PG-based mixture (control group) was prepared by mixing PG with water at the optimum moisture content and compacted under the same conditions as the MPGT.

2.2. Mechanical Test

2.2.1. California Bearing Ratio Test (CBR)

The CBR test was conducted in accordance with method T 0134-2019 specified in the standard JTG 3430-2020 [38]. The specimens with dimensions of Φ 152 mm × h 120 mm were prepared using a heavy compaction apparatus. For each experimental group, three parallel specimens were prepared to ensure data reliability. The specimens were initially cured under standard conditions within a temperature of 20 ± 2 °C and a humidity exceeding 95% for 7 days and then immersed in water for 4 days to simulate the saturated condition. Finally, the penetration test was performed using a universal testing machine, with the loading rate strictly controlled at 1 mm/min, and the relationship between penetration depth and unit pressure was obtained.

2.2.2. Splitting Test

The splitting tensile test specimens with dimensions of Φ 50 mm × h 50 mm were prepared by static compaction. The specimens were cured for 7, 14, and 28 days. These curing ages were selected to verify compliance with the standard requirement and to evaluate the long-term crack resistance. The specimens were soaked in water for 24 h on the last day of the curing period. The splitting test was carried out using a universal testing machine in accordance with the method T 0806-1994 specified in the standard JTG 3441-2024 [39]. The loading rate was controlled at 1 mm/min. Six parallel specimens were prepared for each group. The maximum load at failure for each specimen was recorded. The splitting tensile strength is calculated by Equation (1).

$$[eqn]$$

where $[eqn]$ is the splitting tensile strength of the specimen, MPa; $[eqn]$ is the diameter of the specimen, mm; $[eqn]$ is the width of the loading strip, mm; $[eqn]$ is the central angle corresponding to half the strip width, °; $[eqn]$ is the maximum load at failure, N; $[eqn]$ is the height of the specimen after soaking, mm.

2.2.3. Unconfined Compressive Strength Test (UCS)

The specimens for unconfined compressive strength test were cylinders with diameters of Φ 50 mm × h 50 mm prepared by static compaction in accordance with the method T 0805-2024 specified in the standard JTG 3441-2024 [39]. To investigate the strength evolution, the specimens were cured for 1, 3, 7, 14, and 28 days. After the prescribed curing period, the unconfined compressive strength was tested using a universal testing machine at a constant loading rate of 1 mm/min. The reported UCS values represent the average of six parallel specimens for each group to ensure statistical reliability. The strength was calculated by Equation (2).

$$[eqn]$$

where $[eqn]$ is the unconfined compressive strength, MPa; $[eqn]$ is the maximum load at failure, N; $[eqn]$ is the cross-sectional area of the specimen, mm^2^.

2.2.4. Water Stability Coefficient Test

The water effect on the durability of the MPGT was evaluated by the water stability coefficient. The test was performed based on the unconfined compressive strength test method T 0805-2024 specified in the standard JTG 3441-2024 [39]. Specimens with diameters of Φ 50 mm × h 50 mm were cured for 7 d and 28 d to simulate the short-term and long-term immersion respectively. The specimens cured for 7 d specimens were maintained at 25 °C for 6 days and subsequently immersed in water at 20 °C for 24 h. The specimens cured for 28 days were kept at 25 °C for 27 days and then similarly immersed for 24 h. For each group, three parallel specimens were tested using a universal testing machine and the average value was obtained to evaluate the water stability of MPGT. The water stability coefficient Kw is determined by the ratio of UCS of soaked specimens to the unsoaked specimens under standard curing conditions, which is calculated by Equation (3).

$$[eqn]$$

where $[eqn]$ is the water stability coefficient; $[eqn]$ is the unconfined compressive strength of soaked specimens, MPa; $[eqn]$ is the unconfined compressive strength of unsoaked specimens, MPa.

2.2.5. Freeze–Thaw Test

The freeze–thaw test was conducted to evaluate the water stability of the MPGT in accordance with method T 0858-2009 specified in the standard JTG 3441-2024 [39]. For each group, eighteen cylindrical specimens with dimensions of Φ 50 mm × h 50 mm were prepared. After 28 days of curing, nine specimens served as the control group (without freeze–thaw treatment), while the other nine were subjected to five freeze–thaw cycles. Each cycle consisted of freezing at −18 °C for 16 h, followed by thawing in a water bath at 20 °C for 8 h. Before testing, the mass of all the specimens was measured, and then the compressive strength test was performed using a universal testing machine. The frost resistance index (FRI) and the mass loss rate were calculated according to Equation (4).

$$[eqn]$$

where $[eqn]$ is the frost resistance index (referred to as BDR in the standard JTG 3441), %; $[eqn]$ is the compressive strength after n freeze–thaw cycles, MPa; $[eqn]$ is the compressive strength of control specimens, MPa; $[eqn]$ is the mass before freeze–thaw, kg; $[eqn]$ is the mass after n freeze–thaw cycles, kg.

2.2.6. Erosion Test

Anti-erosion performance is a critical durability index for pavement base course materials, which is strongly related to the long-term stability under rainfall and surface water erosion. This test evaluates the hydraulic erosion resistance of the MPGT by simulating actual water flow conditions in accordance with the method T 0860-2009 specified in the standard JTG 3441-2024 [39]. For each group, three parallel specimens with dimensions of Φ 150 mm × h 150 mm were prepared using static compaction. After molding, specimens were initially cured under standard conditions at a temperature of 20 ± 2 °C and a humidity exceeding 95% for 28 days and then immersed in 20 °C water for 24 h. Before testing, the surface water on the specimens was wiped off, and the initial mass was recorded. The specimens were subjected to a standardized erosion test at 0.5 MPa pressure and 10 Hz frequency for 30 min. After the erosion testing, the turbid water and eroded material were collected and left undisturbed for 12 h. The sediment was dried at 105 °C to constant mass. The erosion mass loss rate was calculated as the ratio of eroded mass to the initial specimen mass according to Equation (5).

$$[eqn]$$

where $[eqn]$ is the erosion mass loss rate, %; $[eqn]$ is the mass of eroded material, g; $[eqn]$ is the mass of the specimen, g.

2.2.7. Determination of Calcium Carbonate Content

The calcium carbonate (CaCO_3_) content was determined using the acid-washing method. Initially, the crushed MPGT samples were rinsed with deionized water to eliminate the soluble salts (such as residual urea and sodium chloride) and then dried in an oven at 105 °C until a constant mass, namely $[eqn]$ . Subsequently, the dried samples were immersed and washed multiple times with a 1.0 mol/L hydrochloric acid (HCl) solution to ensure complete reaction with CaCO_3_. Finally, the residue was rinsed with deionized water and dried to a constant mass, which was recorded as $[eqn]$ . The CaCO_3_ content can be calculated according to Equation (6).

$$[eqn]$$

where $[eqn]$ is the dry mass of the crushed MPGT sample before acid washing, g; $[eqn]$ is the dry mass after acid washing, g.

2.3. Micro-Structural Analyses

The samples of PG and MPGT were dried in an oven at a temperature of 55 °C for 24 h. Subsequently, the samples were ground and sieved through a 200-mesh sieve. The resultant powders were employed for X-ray diffraction (XRD), Scanning Electron Microscopy (SEM), and Fourier Transform Infrared Spectroscopy (FTIR) analyses.

The XRD analysis was conducted using a Bruker D8 Advance X-ray diffractometer (Bruker, Karlsruhe, Germany), with a scanning from 2θ = 5° to 80° in steps of 0.04° and a dwell time of 1.5 s. The SEM analysis was performed on a JSM-6510LV scanning electron microscope (JEOL Ltd., Tokyo, Japan) at various magnifications. The FTIR analysis was carried out using a Fourier transform infrared spectrometer (Nicolet 6700, Thermo Fisher Scientific, Waltham, MA, USA) with a wavenumber range of 400 to 4000 cm^−1^.

2.4. Toxicity Leaching Test

In addition to mechanical strength and water stability, it is also essential to evaluate the environmental safety of the MPGT utilized in pavement base course [40]. In this study, deionized water was employed as the leaching agent in accordance with the Identification Standards for Hazardous Wastes—Identification of Leaching Toxicity GB 5085.3-2007 [41]. MPGT samples (50 g) and deionized water (500 g) were mixed in sealed bottles and oscillated at 110 ± 10 rpm for 8 h at 25 °C. Afterwards, the bottle containing the MPGT and deionized water was left undisturbed for 16 h. The leachate was filtered through a 0.22 μm membrane to eliminate potential impurities before analysis. The concentrations of F^−^ and PO_4_^3−^ in the leachate were measured using an ICS-1500 ion chromatograph (Thermo Fisher Scientific, Waltham, MA, USA).

3. Results

3.1. Determine the Preparation Parameters of Microbial Induced Carbonate Precipitation (MICP) for Stabilizing PG

Both the cementation solution concentration and the volume ratio of bacterial solution to cementation solution are critical parameters to preparing the MPGT. In this section, a comparative study was conducted to determine the optimal cementation solution concentration and the volume ratio of bacterial solution to cementation solution based on the test results of UCS, CBR, and splitting tensile strength of MPGT.

3.1.1. Concentration of Cementation Solution

Figure 4a presents the UCS test results of MPGT prepared within five cementation solution concentrations (1.0 mol/L, 1.5 mol/L, 2.0 mol/L, 2.5 mol/L, and 3.0 mol/L) and a fixed volume ratio of bacterial solution to cementation solution of 2:1. The UCS of MPGT initially increased with the cementation solution concentration and then started to decrease. The peak UCS value of 17.31 MPa was obtained when the cementation solution concentration was 2.0 mol/L. This phenomenon can be attributed to the urea hydrolysis effect on the MICP process.

When the cementation solution concentration was relatively lower than 2.0 mol/L, the generation of carbonate ions (CO_3_^2−^) increased with the cementation solution concentration. In addition, the enzyme activity test also indicated that the conductivity of solution increased from 1.2 mS/cm·min^−1^ to 4.1 mS/cm·min^−1^, which is also beneficial for urea hydrolysis. Consequently, a greater amount of carbonate ions (CO_3_^2−^) would result in generation of more CaCO_3_, which promotes the bonding of the PG particles and thereby enhances the UCS, which was in accordance with previous studies [42,43].

However, when the cementation solution concentration increased from 2.0 mol/L to 3.0 mol/L, the UCS of MPGT decreased by 13.2% from 17.31 MPa to 15.03 MPa. In this stage, the enzyme activity decreased by 18% correspondingly. It indicates that relatively larger amount of urea would inhibit the urease activity, thereby decreasing the generation of carbonate ions (CO_3_^2−^). Therefore, the UCS of MPGT decreased when the cementation solution concentration exceeded 2.0 mol/L.

Figure 4b illustrates the CBR and Splitting test results of PG mixtures within different cementation solution concentrations. Similar to the UCS test results in Figure 4a, the CBR value and splitting tensile strength also initially increased when the cementation solution concentration was below the 2.0 mol/L and then decreased when the cementation solution concentration increased from 2.0 mol/L to 3.0 mol/L. It can also be explained by the effect of cementation solution concentration on the urease activity and the generation of carbonate ions (CO_3_^2−^). In summary, the optimal cementation solution concentration for preparing the MPGT was determined as 2.0 mol/L according to test results of UCS, CBR, and splitting tensile strength.

3.1.2. Effect of Volume Ratio of Bacterial Solution to Cementation Solution on Mechanical Properties

The volume ratio of bacterial solution to cementation solution also significantly influences the mineralization reaction for MICP technique. Considering the optimal cementation solution concentration was 2.0 mol/L, five volume ratios (1:3, 1:2, 1:1, 2:1, and 3:1) were adopted to investigate the effect of the volume ratios of bacterial solution to cementation solution on the MPGT. Figure 5a illustrates the UCS of different PG mixtures. The UCS initially increased with the volume ratios of bacterial solution to cementation solution and then began to decrease. A lower ratio implies insufficient bacterial biomass relative to the cementation solution, whereas a higher ratio indicates an excess of bacteria with limited calcium availability. When the volume ratio increased from 1:3 to 2:1, the urease concentration increased and more CaCO_3_ was generated to bond the PG particles, enhancing the UCS of MPGT. However, when the volume ratio was greater than 2:1, the less available calcium was insufficient to support the excess bacteria, resulting in the decrease in the UCS of MPGT.

Figure 5b presents the CBR and Splitting test results of MPGT prepared with different volume ratios of bacterial solution to cementation solution. The CBR value and splitting tensile strength show a similar change trend with the volume ratio of bacterial solution to cementation solution as the UCS test results in Figure 5a. Therefore, the volume ratio of bacterial solution to cementation solution was determined as 2:1, which represents the balanced proportion of bacterial solution to cementation solution.

3.1.3. Validation of Durability of Treated Phosphogypsum Mixtures

According to Chinese specification JTG/T 3610–2019 [44], the durability of MPGT was investigated to validate the optimum parameters for preparing the MPGT determined in Section 3.1.1 and Section 3.1.2. Specifically, the MPGT specimens were prepared using the optimal cementation solution concentration of 2.0 mol/L and a bacterial-to-cementation solution volume ratio of 2:1. A comparative study was conducted on the water stability, frost resistance and erosion resistance of the untreated PG and MPGT. For the frost resistance test, the specimens were subjected to five freeze–thaw cycles, with each cycle consisting of freezing at −18 °C for 16 h followed by thawing at 20 °C for 8 h. The results are shown in Table 3.

The microbial-induced carbonate precipitation (MICP) treatment significantly enhances the durability of MPGT compared to the untreated PG. In terms of frost resistance, the untreated PG exhibited a low FRI of 25.8%, whereas the FRI value of MPGT reached 83.6%, satisfying the Chinese requirement of ≥70%. Similarly, the erosion mass loss rate of MPGT was only 8.04%, significantly lower than the 50% for the untreated PG.

However, regarding water stability, although the coefficient increased from 18.60% to 55.05%, it must be acknowledged that this value still falls short of the stringent requirement (≥80%) for high-grade road base materials specified in JTG 3441. This limitation is primarily attributed to the inherent water sensitivity of the phosphogypsum matrix. Although microbial calcite precipitation (CaCO_3_) provides effective bonding bridges between particles, it cannot fully prevent the softening and partial dissolution of gypsum dihydrate crystals (CaSO_4_·2H_2_O) upon prolonged immersion.

Therefore, solely relying on MICP is insufficient to completely resolve the water stability issue. For practical engineering applications requiring strict water stability, a composite stabilization strategy (e.g., incorporating a small amount of inorganic binders like cement or lime into the MICP-treated PG) is recommended as a necessary measure to further waterproof the matrix and meet the qualification standard.

3.2. Enhancement Mechanism of Microbial Induced Carbonate Precipitation on Phosphogypsum-Based Mixture

3.2.1. Correlation Between CaCO3 Production and Strength Improvement

As described in Section 2.2.7, the acid-washing method was used to determine the quantity of CaCO_3_ produced in the MPGT prepared with different cementation solution concentrations and the volume ratios of bacterial solution to cementation solution. The results are summarized in Table 4.

It is evident that there is a strong positive correlation between the CaCO_3_ production and the UCS of the PG mixture. As listed in Table 4, the maximum CaCO_3_ production was 12.8% when the cementation solution concentration was 2.0 mol/L and the volume ratio was 2:1. This trend aligns perfectly with the UCS results presented in Figure 4.

Specifically, the mechanism of strength enhancement can be attributed to the “pore-filling” and “particle-binding” effects of the biomineralization products. As the urea concentration increased to the optimal level (2.0 mol/L), the promoted urease activity generated sufficient carbonate ions to react with calcium ions. The resulting CaCO_3_ precipitates (up to 12.8%) effectively filled the inter-particle voids and cemented the loose phosphogypsum particles into a coherent matrix. This densification process, driven by the accumulation of CaCO_3_, directly reduced the macro-porosity of the mixture, thereby significantly enhancing the macroscopic mechanical strength (UCS). However, excessive urea or unbalanced solution ratios inhibited the reaction, resulting in reduced CaCO_3_ content and a consequently looser microstructure with lower strength.

3.2.2. Strength Evolution of Treated Phosphogypsum Mixtures

Standard specimens were also prepared within different curing period levels (1 day, 3 days, 7 days, 14 days, and 28 days) and then performed by the UCS test to investigate the strength evolution of the MPGT. The test results are presented in Figure 6.

A nonlinear regression model was developed to analyze the strength evolution characteristics. The growth trend with respect to curing age can be divided into the following two stages: (1) a rapid increase during the initial phase (1–7 days); (2) slower growth in the intermediate phase (7–28 days). When the curing age increased from 1 day to 7 days, the UCS increased by 109%, from 7.46 MPa to 15.61 MPa, while the corresponding results for the curing age increasing from 7 days to 28 days was only 10.9%. Additionally, as shown in Table 4, the UCS exhibits a strong correlation with the CaCO_3_ generated by the mineralization reaction. Therefore, it can be concluded that the majority of the mineralization reaction was completed within the initial 14 days to achieve nearly maximum strength.

3.2.3. Micro-Structural Studies

XRD Phase Analysis

Figure 7 presents the XRD analysis results of the untreated PG-base mixture and the MICP stabilized mixture. Generally, the chemical composition can be inferred from the position of diffraction peaks. As shown in Figure 7, the chemical composition untreated PG-base mixture (blue line) consists of CaSO_4_·2H_2_O (2θ ≈ 11.6°) and SiO_2_ (2θ ≈ 26.6°), which are sourced from the PG material. For the MPGT sample (red line), new diffraction peaks appeared and the characteristic peak was located at the position of 2θ ≈ 29.4°, indicating the existence of calcite-type CaCO_3_. The results testified the formation of calcite crystal precipitation catalyzed by the microbial urease, which enhances the mechanical strength of the PG-based mixture.

SEM Analysis

Figure 8 and Figure 9 show the SEM images of the untreated PG-base mixture and the MICP stabilized mixture, respectively. For the untreated PG-based mixture, the PG particles exhibit a loose cauliflower-like accumulation structure with irregular shapes and serrated edges. The surface of PG particles was rough and highly porosity, which explains its high-water absorption capability. As shown in Figure 9, the structure of MPGT becomes denser due to the effect of calcite crystal precipitation. It can be observed that the calcite crystals coat the PG particle surface and fill the voids between the PG particles, resulting in a smoother texture. This implies that the MICP effect enhances the degree of compaction and the bonding characteristics of PG-based mixture, contributing to greater mechanical strength and better durability, in accordance with results in Figure 4, Figure 5, Figure 6 and Figure 7 and Table 3.

FTIR Spectra Analysis

Figure 10 presents the FTIR spectra for the untreated PG and the MPGT samples. The spectrum of untreated PG exhibits distinct bands characteristic of gypsum (CaSO_4_·2H_2_O), including broad O–H stretching vibrations at 3549 cm^−1^ and 3408 cm^−1^, alongside pronounced SO_4_^2−^ bending vibrations near 1141 cm^−1^. Notably, impurity peaks within the fingerprint region (836 cm^−1^ and 794 cm^−1^) are associated with P–O bonds originating from soluble phosphates or lattice impurities [45].

Moreover, the spectrum of the MPGT specimen reveals significant alterations that elucidate the immobilization mechanism of hazardous ions. The sharp impurity peaks observed in untreated PG were significantly reduced and masked, suggesting a dual-mechanism for pollutant immobilization. Specifically, the biomineralized calcite forms a dense coating layer on the PG particle surfaces (corroborated by SEM results), physically blocking the dissolution channels of soluble fluoride and phosphate (physical encapsulation). Simultaneously, the high concentration of Ca^2+^ in the cementation solution promotes the reaction with soluble F^−^ and PO_4_^3−^ to form insoluble phases (e.g., fluorite CaF_2_ or Calcium Phosphates), which are then co-precipitated and firmly embedded within the calcite crystal lattice (Chemical Stabilization).

Furthermore, specific organic–inorganic functional groups presented in the MPGT sample highlighted its strengthening mechanism. Although the spectral region around 1600–1700 cm^−1^ is typically dominated by crystal water bending vibrations in gypsum, variations in peak intensity around 1687 cm^−1^ indicate superposition with bacterial organic matrix components (C=O stretching of Amide I). These organic constituents serve as nucleation sites for calcite precipitation, thereby forming a dense “organic–inorganic” hybrid cementing network. It can not only enhance the strength of PG-based mixture but also reduces the release of hazardous ion into the environment.

3.3. Toxicity Leaching Test and Environmental Safety Evaluation

3.3.1. Validation of the Harmless Effect of Microbial Induced Carbonate Precipitation on Treated Phosphogypsum Mixtures

In Section 3.1, the cementation solution concentration was determined as 2.0 mol/L and the volume ratio of bacterial solution to cementation solution determined as 2:1 from the mechanical strength and durability aspects. Here, the toxicity leaching test were performed on the MPGT specimens to demonstrate the harmless effect of the MICP technology on the PG-based mixture. Generally, the F^−^ and PO_4_^3−^, as well as the pH value, serve as the critical indicators for evaluating the environmental safety of PG-based material.

In this paper, a comparative study was conducted on the immobilization efficiency of F^−^ and PO_4_^3−^ for the untreated PG-based and MPGT samples, and the results are shown in Figure 11. Compared to the untreated PG-based mixture, the leaching concentration of F^−^ for the MPGT was decreased by 67.5%, meanwhile the leaching the PO_4_^3−^ ions were almost completely immobilized. Additionally, the acidity of the leaching solution was significantly neutralized. The untreated PG exhibited a low pH of 4.2, indicating strong acidity. After MICP treatment, the pH value of the MPGT increased to 7.0, which satisfies the standard requirement (pH 6−9). It confirmed that the MICP technology can enhance the immobilization efficiency and environmental compatibility of the PG−based mixture.

3.3.2. Effect of Cementation Solution Concentration on Leaching Behavior

The effect of cementation solution concentration on the immobilization efficiency of F^−^ and PO_4_^3−^ is summarized in Figure 11a. Since the MICP was used to stabilize the PG particles, the leaching concentration of F^−^ and PO_4_^3−^ showed a significant decrease in comparison to the untreated PG−based mixture. The increase in pH value can be attributed to the urea hydrolysis process, which generates ammonium (NH_4_^+^) and hydroxide ions (OH^−^), thereby neutralizing the acidic impurities in PG.

Nevertheless, when the cementation solution concentration was as low as 1.0 mol/L, the amount of cementation solution was limited, which constrained the generation of carbonate ions and alkalinity. As a result, the insufficient calcite precipitate was unable to form a continuous protective layer around PG particles, resulting in the relatively high leaching concentration of F^−^. When the cementation solution concentration was increased to 2.0 mol/L, urease activity reached a high level and facilitated the generation of sufficient CaCO_3_ precipitation. In this case, the leaching concentration of F^−^ decreases to the minimum value of 5.15 mg/L, while that of PO_4_^3−^ was 0.05 mg/L, indicating that abundant mineral crystals effectively encapsulate the toxic ions component, making them difficult to leach.

However, the leaching concentration of F^−^ and PO_4_^3−^ increased to 7.24 mg/L and 1.85 mg/L, respectively, when the cementation solution concentration was increased from 2.0 mol/L to 3.0 mol/L. A high urea concentration inhibits bacterial metabolism, and meanwhile, the rapid generation of CaCO_3_ precipitation impeded the substrate and Ca^2+^ diffusion into the porous structure of the PG−based mixture. Consequently, F^−^ and PO_4_^3−^ can still be released at a noticeable level and their leaching concentrations started to increase. It can be concluded that the cementation solution concentration has a considerable effect on the leaching behavior of F^−^ and PO_4_^3−^ for the PG-based mixture.

3.3.3. Effect of Volume Ratio on Leaching Behavior

Figure 11b illustrates the leaching concentrations of fluoride (F^−^) and phosphate (PO_4_^3−^) for MPGT specimens prepared with different volume ratios of bacterial solution to cementation solution (V_b_:V_c_). Compared to the untreated PG mixture, all MPGT groups showed a significant reduction in the leaching of F^−^ and PO_4_^3−^, confirming that MICP technology effectively immobilizes these pollutants. This trend is consistent with the optimization results of the cementation solution concentration.

However, the immobilization efficiency was strongly influenced by the V_b_:V_c_ ratio. When the ratio was optimized at 2:1, the supply of bacterial cells (nucleation sites) and cementation solution (reactants) reached an optimal balance. This facilitated sufficient mineralization to interconnect adjacent PG particles, forming a dense skeleton–cementation structure that maximized the physical encapsulation of F^−^ and PO_4_^3−^.

In contrast, at lower ratios (where cementation solution volume exceeded bacterial volume), the bacterial concentration was inadequate. This led to reduced total urease activity and slower precipitation rates, resulting in unsatisfactory immobilization. Conversely, when the ratio increased from 2:1 to 3:1, the relative amount of cementation solution decreased. In this scenario, insufficient calcium availability limited the generation of calcium carbonate crystals, causing the leaching concentrations of F^−^ and PO_4_^3−^ to rebound.

In summary, both the cementation solution concentration and the V_b_:V_c_ ratio critically affect the immobilization efficiency. Based on the leaching data in Figure 11, the maximum immobilization efficiency was achieved with a 2.0 mol/L cementation solution and a 2:1 volume ratio, which aligns with the optimal parameters for mechanical strength and durability.

Furthermore, it is important to address the scope of the environmental risk assessment in this study. While the leaching tests focused on fluoride (F^−^) and phosphate (PO_4_^3−^)—identified as the primary mobile pollutants in phosphogypsum posing eutrophication risks—the potential impacts of heavy metals and radioactivity were also considered. As detailed in Section 2.1.1, the radioactivity indices of the raw PG satisfy the Class A limit (GB 6566-2010), and trace heavy metals (e.g., Pb, Cr) are known to be effectively immobilized via co-precipitation within the calcite crystal lattice generated by MICP.

Regarding the leaching conditions, the use of deionized water in this study follows the standard method (GB 5085.3-2007) to simulate a baseline scenario of groundwater infiltration. Although realistic engineering environments may involve acidic rainwater, the abundant calcium carbonate (CaCO_3_) produced by bio-mineralization provides significant acid-buffering capacity. This alkalinity helps neutralize acidic pore fluids, thereby maintaining the chemical stability of the solidified matrix and preventing the release of encapsulated pollutants.

3.4. Cost Analysis and Economic Feasibility

A preliminary cost analysis was performed to evaluate the economic feasibility of MPGT as a road base material. It is acknowledged that the use of analytical-grade reagents in this laboratory study—specifically to ensure experimental precision and avoid impurity interference during toxicity testing—results in a theoretically high cost (~1.26 million CNY per lane-kilometer). However, this does not represent the actual cost in practical engineering applications.

In large-scale engineering projects, “Industrial Grade” raw materials (purchased in bulk/wholesale) are utilized. Based on current industrial market prices in China (e.g., industrial urea ~2.8 CNY/kg, industrial calcium chloride ~1.5 CNY/kg, and industrial-grade peptone ~40.0 CNY/kg), the material cost can be significantly reduced. As presented in Table 5, the re-evaluated comprehensive cost of MPGT is approximately 460 CNY/m^3^ (including estimated bio-fermentation and processing costs). Although this cost is approximately 1.8–2 times that of traditional cement-stabilized macadam (~200–250 CNY/m^3^), it is considered acceptable for demonstration projects given the superior environmental benefits of phosphogypsum recycling and the enhanced durability of the road base.

Furthermore, recent parallel studies by our research group have demonstrated the feasibility of using low-cost agricultural/industrial by-products (e.g., corn steep liquor powder) as a substitute for peptone [46]. Preliminary results indicate that this substitution could further reduce the nutrient cost by approximately 60–80%. Therefore, the cost presented in Table 5 represents a conservative estimate, and the actual engineering cost has significant potential to be further lowered in future applications to be competitive with traditional binders.

4. Conclusions

Based on the experimental findings on the mechanical properties, durability, and environmental safety of microbial induced carbonate precipitation (MICP)-treated phosphogypsum (MPGT) road base material, the following conclusions can be drawn:

1. The cementation solution concentration and the volume ratio of bacterial solution to cementation solution critically influence the performance of the PG-based mixture. The optimal preparation parameters were determined to be a cementation solution concentration of 2.0 mol/L and a bacterial-to-cementation solution volume ratio of 2:1. Under these conditions, the mixture achieved a maximum unconfined compressive strength of 17.31 MPa.
2. The MICP treatment significantly enhanced the durability of the PG-based mixture. Under the optimal conditions, the frost resistance index (FRI) improved to 83.6% and the erosion mass loss rate decreased to 8.04%, satisfying the relevant standards. Although the water stability coefficient increased significantly to 55.05%, it indicates that for high-grade road base applications, composite stabilization is recommended to fully meet the stringent water stability requirements.
3. Microstructural analyses (XRD, SEM, and FTIR) confirmed that strength enhancement is attributed to the biomineralization of calcium carbonate. These calcite precipitates coat the PG particles and fill the inter-particle voids, creating a dense and stable microstructure that improves both mechanical strength and resistance to water/frost erosion.
4. MICP treatment effectively immobilized hazardous impurities in the phosphogypsum. The leaching concentration of fluoride (F^−^) decreased by 67.5%, and the phosphate (PO_4_^3−^) concentration was reduced from 60.18 mg/L to 0.05 mg/L. Additionally, the leachate pH was neutralized, demonstrating that the MPGT meets environmental safety standards for road engineering applications.

Future research should consider the evolution of MPGT as a road base under the coupled effects of actual load and environmental conditions, such as moisture and temperature variation. Moreover, the pollutant characteristics of the PG-based mixture should be comprehensively investigated, and more pollutant indexes should be taken into consideration to strictly evaluate environmental safety. Long-term field studies are also needed to validate the laboratory findings and assess the durability of the MICP-stabilized PG-based mixture in real service conditions. In the context of such large-scale engineering applications, the central plant mixing method is recommended based on the current findings. This approach, which involves spraying premixed biological solutions onto phosphogypsum aggregates in standard stabilized soil mixing stations, ensures precise dosage and uniform distribution, thereby guaranteeing the performance consistency of the road base.