Synergistic Effects of Microbial-Induced Carbonate Precipitation and Modified Biochar on the Engineering Properties of Loess

Abstract

Collapsible loess poses significant geotechnical risks due to its metastable structure and water sensitivity, while conventional stabilization methods often lack sustainability. This study investigates the synergistic effects of microbial-induced carbonate precipitation (MICP) and modified biochar (MBC) to enhance loess engineering properties. Controlled experiments evaluated hydraulic conductivity, shear strength, and stress-strress–strain behavior under varying MBC content (0–8%), cementation reagent concentration (0.5–1.5 mol/L), and confining pressures (50–400 kPa), and complemented by microstructural characterization via scanning electron microscope (SEM). Results demonstrate that MBC (4–6%) optimizes calcium carbonate distribution by providing nucleation sites, reducing hydraulic conductivity by 72% and increasing shear strength by 52% when compared with untreated loess. Elevated confining pressures (200–400 kPa) transformed brittle failure into ductile behavior through particle interlocking, with peak strength quadrupling under 400 kPa. SEM analysis revealed MBC stabilizes hierarchical pore networks: macropores sustain microbial activity, while mesopores are occluded by CaCO₃-MBC composites, sequestering ionic byproducts to mitigate efflorescence. The optimal combination (6% MBC, 1.0 mol/L reagent, 200 kPa confinement) achieved 85% of maximum strength gain at reduced reagent cost, balancing performance and sustainability.

1. Introduction

Loess, a wind-deposited aeolian silt with a global distribution, spans vast continental regions and plays a pivotal role in sustaining ecosystems and human infrastructure. In China, the Loess Plateau, a nearly continuous loess-dominated region spanning approximately 630,000 km², as illustrated in Figure 1, exemplifies its geological and ecological significance [1]. Characterized by its loose, porous structure and sensitivity to water, loess is inherently prone to hydro-mechanical instability, manifesting as catastrophic collapse, erosion, and landslides under environmental stressors [2,3,4,5]. These vulnerabilities pose significant risks to infrastructure and agricultural productivity, particularly in arid regions where loess serves as the primary substrate for human activity. Conventional stabilization methods, such as cement-based solidification and chemical grouting, have long been employed to mitigate these risks. However, their reliance on energy-intensive processes and non-renewable materials raises concerns about long-term ecological sustainability, including soil degradation and groundwater contamination [4,5,6].

In recent decades, the pursuit of sustainable geotechnical solutions has driven innovation in bio-inspired soil modification technologies. Among these, MICP has emerged as a groundbreaking approach [7,8,9]. By leveraging the metabolic activity of ureolytic bacteria to precipitate calcium carbonate within soil matrices, MICP enhances particle cohesion and reduces permeability, offering a pathway to stabilize soils with minimal environmental disruption [10,11,12]. Concurrently, biochar (BC), a carbon-rich material derived from biomass pyrolysis, has gained recognition for its dual role in soil remediation: immobilizing heavy metals through adsorption and improving soil fertility by enhancing nutrient retention [13]. Despite these advancements, the application of these technologies to heavy metal-contaminated loess remains fraught with challenges. The fine-grained nature of loess limits the uniform distribution of biominerals, while the toxic effects of heavy metals, such as lead, suppress microbial activity, undermining the efficacy of MICP [14,15,16]. Similarly, unmodified biochar exhibits limited mechanical reinforcement and unstable adsorption performance under dynamic environmental conditions, restricting its standalone utility.

The integration of MICP with functionalized amendments represents a promising strategy to address these limitations. Modified biochar (MBC), engineered through physical or chemical treatments, has shown enhanced adsorption capacity and structural stability when compared with its pristine counterpart [17,18,19]. When combined with MICP, MBC may act synergistically to mitigate metal toxicity, provide nucleation sites for biomineralization, and improve soil fabric through pore refinement [20]. Recent studies suggest that such hybrid systems could achieve multifunctional benefits, including contaminant immobilization, hydraulic conductivity reduction, and mechanical reinforcement [21,22,23]. However, the mechanisms governing these interactions, particularly in lead-contaminated loess systems, remain poorly understood. Key questions persist regarding how MBC contributes to buffering metal toxicity, optimizing biomineral distribution, and modifying the microscale architecture of loess to enhance its macroscale engineering performance [24,25,26,27].

This study addresses these gaps by systematically examining the synergistic effects of MICP and MBC on the engineering properties of lead-contaminated loess. Through a comprehensive evaluation of hydraulic conductivity, shear strength, and stress–strain behavior, coupled with advanced microstructural characterization, the research elucidates the interplay between biogeochemical processes and material engineering. The investigation reveals how calcium carbonate precipitation, mediated by microbial activity, synergizes with MBC’s adsorption and nucleation functions to transform the soil fabric. These interactions not only reduce permeability by filling interparticle voids but also enhance particle bonding through bio-cementation and interfacial friction. Furthermore, the study explores the role of MBC in sequestering lead ions, thereby alleviating metal-induced inhibition of microbial processes and ensuring sustained biomineral production.

By demonstrating the feasibility of MICP–MBC composites as a sustainable alternative to conventional stabilization methods, the research contributes to the development of eco-friendly geotechnical solutions. It offers actionable insights for rehabilitating contaminated loess landscapes, balancing ecological preservation with engineering resilience. In doing so, the study bridges critical gaps between biogeochemistry, materials science, and geotechnical engineering, paving the way for innovative strategies in soil stabilization and environmental restoration.

2. Materials and Methods

2.1. Test Bacteria and Growth Conditions

The urease-producing bacterium Sporosarcina pasteurii ATCC 11859, renowned for its robust carbonate precipitation capability, served as the bioagent throughout this investigation. A nutrient-rich medium was formulated to optimize bacterial proliferation and enzymatic activity, comprising (per liter) the following: 15 g tryptone (casein digest), 5 g soytone (soybean peptone), 5 g NaCl, and 20 g agar as solidifying agent (

Table 1. Composition of S. pasteurii growth medium.

Tryptone g/L	NaCl g/L	Soytone g/L	Agar g/L	Urea g/L
15	5	5	20	20

). These components were dissolved in 900 mL deionized water under continuous magnetic stirring, followed by pH adjustment to 7.0 ± 0.1 using 1 M NaOH. The basal medium underwent terminal sterilization via autoclaving at 121 °C (15 psi) for 15 min to eliminate microbial contaminants. To preserve urea integrity against thermal degradation, a separate 20% (w/v) urea solution was prepared and sterilized through 0.22 μm membrane filtration. This urea concentrate was aseptically introduced into the cooled basal medium (≈60 °C) at 100 mL/L concentration, achieving final urea content of 2% (w/v) in the complete growth medium.

Inoculation was performed under laminar flow conditions by transferring 1% (v/v) glycerol stock culture into the prepared medium. Aerobic cultivation proceeded in an orbital shaker incubator (ZHWY-211C, Shanghai Zhicheng Analytical Instrument Manufacturing Co., Ltd., Shanghai, China) maintained at 30 °C with 220 rpm agitation for 24 h, ensuring optimal oxygen transfer during logarithmic growth phase. Post-cultivation, bacterial suspensions were aliquoted into sterile centrifuge tubes and stored at 4 °C in temperature-controlled chambers (SPX-250, Shanghai Boxun Industrial Co., Ltd., Shanghai, China) to maintain metabolic viability until experimental application. Regular optical density (OD₆₀₀) measurements and urease activity tests confirmed culture stability during storage, with no significant activity loss observed within 14 days.

The cementation solution was formulated through sequential dissolution of urea and calcium chloride in deionized water. Three distinct molar concentrations (0.5, 1.0, and 1.5 M) were systematically prepared for both components to investigate concentration-dependent effects on carbonate precipitation. Equimolar mixing of urea and calcium chloride solutions ensured stoichiometric alignment between the enzymatic substrate and calcium ion supply, a critical prerequisite for efficient urease-driven mineralization.

Urea served dual functions as both the nitrogen source for microbial metabolism and the carbonate donor through enzymatic hydrolysis, while calcium chloride provided essential divalent cations for crystal nucleation. To preserve solution integrity, freshly prepared mixtures were maintained in sealed containers under ambient temperature and utilized within a constrained timeframe. This precautionary measure minimized urea decomposition and ensured consistent reactivity during the biocementation process.

2.2. Modified Biochar

The MBC was synthesized through a mineral-enhanced pyrolysis process. Red mud, an iron-rich industrial byproduct, was dispersed in 500 mL of deionized water under continuous magnetic stirring (30 min) to form a colloidal suspension. Peanut shell powder (10 g, particle size < 2 mm) was immersed in the suspension and agitated for 1 h to facilitate mineral adsorption onto the biomass surface. The impregnated biomass was vacuum-filtered and oven-dried at 80 °C for 12 h to achieve moisture equilibrium. Pyrolysis was conducted in a horizontal tube furnace under nitrogen flow (200 mL/min), programmed with a heating rate of 10 °C/min to a target temperature of 700 °C, followed by a 2 h isothermal hold to ensure complete carbonization. Post-pyrolysis, the MBC was cooled to room temperature under inert conditions, homogenized by grinding, and stored in airtight containers to preserve surface reactivity [28].

Bacterial growth was monitored via optical density (OD₆₀₀), with a measured value of 1.748 corresponding to stationary-phase cultures. This OD range ensured optimal urease activity (15–18 mM urea hydrolyzed/min), as calibrated through preliminary enzymatic assays. The experimental framework systematically evaluated the synergistic effects of microbial cementation and biochar amendment through a controlled factorial design. Five treatment groups were established with a fixed total incorporation ratio of 3% (w/w additive-to-dry soil mass), while progressively substituting 0%, 2%, 4%, 6%, and 8% of the additive mass with MBC. This design isolated the role of MBC in modifying soil–biocement interactions while maintaining consistent nutrient availability for microbial activity. A 28-day curing period was uniformly applied across all groups, ensuring complete urea hydrolysis and carbonate precipitation under stabilized hygrothermal conditions. Key variables including MBC dosage, microbial metabolic status, and curing duration were rigorously controlled to quantify their individual and combined contributions to soil stabilization. Detailed parameterization of material ratios and temporal variables is provided in

Table 2. Compositional parameters of remediation treatments.

Category	Incorporation Ratio (%)	Ratio of MBC (%)	Treatment Time (d)
MICP	3	/	28
MICP + 2%MBC	3	2	28
MICP + 4%MBC	3	4	28
MICP + 6%MBC	3	6	28
MICP + 8%MBC	3	8	28

.

2.3. Loess Specimen and Experimental Programme

The loess used in this study was collected from a construction foundation pit located in the southern suburbs of Chang’an District, Xi’an City, Shanxi Province, China. Characterized by its yellowish-brown color and homogeneous silty fabric, the soil underwent standardized processing to ensure uniformity: oven-drying at 105 °C, mechanical crushing, and sieving through a 2 mm mesh to remove coarse particles (>2 mm). Following the Standard for Geotechnical Test Methods (GB/T 50123–2019) [29], the remolded specimens were reconstituted using the parallel gradation method to replicate natural particle gradation, as evidenced by the particle size distribution curve in Figure 2. Key physical parameters of the loess, including dry density, porosity, and consistency limits, were determined through standardized geotechnical testing protocols (

Table 3. Basic physical properties of Malan loess.

Dry Density g/cm3	Porosity %	Liquid Limit %	Plastic Limit %	Plastic Index %
1.51	45.58	36.5	19	17.5

). The material exhibited high porosity (45.58%) and moderate plasticity (Plastic Index = 17.5%), typical of Quaternary Malan loess deposits in the region.

The homogenized loess was compacted into cylindrical specimens (39.1 mm diameter × 80 mm height) using a three-layer incremental densification protocol. A servo-controlled hydraulic system maintained a compaction degree of 0.9 to mirror in situ structural conditions. Microbial treatment commenced with the gravity-driven injection of 2 pore volumes (2V_v) bacterial suspension through the specimen’s upper surface. Following a 3 h static incubation period to maximize urease activation, an equivalent volume of cementation solution (1.0 M urea-CaCl₂) was percolated at a controlled infusion rate ≤ 0.6 mL/min, ensuring uniform reagent distribution without inducing particle rearrangement.

Post-injection specimens underwent sequential curing phases: initial 24 h ambient reaction (25 °C) to facilitate carbonate nucleation, followed by 48 h accelerated maturation in a climate-controlled chamber (36 °C, 90% RH). Polyethylene membrane wrapping prevented moisture loss during the maturation stage while permitting gaseous exchange. This staged curing protocol balanced microbial metabolic requirements (36 °C optimum for S. pasteurii) with crystal growth kinetics, achieving complete urea hydrolysis and stable calcite precipitation as verified by subsequent chemical analyses.

3. Soil Engineering Properties

3.1. Soil Electrical Conductivity (EC) Assessment

The electrolytic behavior of treated loess was quantified through a standardized aqueous extraction protocol. Soil–water suspensions were prepared at a gravimetric ratio of 1:2.5 (10 g air-dried soil:25 mL deionized water) in sterile polypropylene centrifuge tubes. Mechanical homogenization was achieved through orbital agitation (180 rpm) at ambient temperature (25 ± 0.5 °C) for 30 min, ensuring complete dissolution of soluble ionic species. Following a 10 min quiescent settling phase to facilitate particulate sedimentation, the supernatant was subjected to electrochemical characterization using a calibrated portable pH/EC meter (SX-620, Changzhou Sihui Testing Equipment Co., Ltd., Changzhou, China). Triplicate measurements were recorded for each sample at electrode stabilization intervals of 30 s, with probe recalibration performed between tests using pH 4.01 and 7.01 buffer solutions.

3.2. Soil pH Quantification Protocol

Soil pH determination followed the standardized aqueous extraction method outlined in GB/T 50123-2019 [29]. Precisely 5 g of air-dried loess (<2 mm particle fraction) was combined with 25 mL deionized water (resistivity > 18.2 MΩ·cm) in sterile 50 mL polypropylene centrifuge tubes, achieving a 1:5 (w/v) soil-to-solution ratio. Homogenization was performed via orbital shaking (180 rpm) under isothermal conditions (25 ± 0.5 °C) for 30 min to equilibrate soluble ionic species. Post-agitation suspensions underwent gravitational sedimentation for 10 min to separate colloidal particulates.

3.3. Permeability Characteristics of Treated Loess

The hydraulic conductivity of Malan loess, both untreated and modified through MICP and biochar amendment, was rigorously evaluated using a stress–strain controlled triaxial permeameter (SLB-6A, Nanjing Soil Instrument Co., Ltd., Nanjing, China) in compliance with the Standard for Geotechnical Test Methods (GB/T 50123-2019) [29]. Cylindrical loess specimens (Ø39.1 mm × 80 mm) were consolidated under isotropic confining pressures ranging from 50 to 400 kPa to simulate in situ stress conditions. To isolate the individual effects of cementation solution concentration and MBC content, a single-variable experimental design was adopted. All tests were conducted at 20 ± 0.5 °C.

3.4. Consolidated Undrained Shear Strength Evaluation

The mechanical reinforcement efficacy of MICP and MBC composites was evaluated through consolidated undrained (CU) triaxial shear tests using a servo-controlled triaxial system (SLB-6A), following the Standard for Geotechnical Test Methods (GB/T 50123-2019) [29]. Untreated, MICP-treated, and MICP–MBC composite specimens (2–8% MBC content) were consolidated under isotropic confining pressures (σ₃ = 50, 100, 200, 400 kPa) to replicate in situ stress conditions. A strain-controlled shearing protocol was applied at a constant axial displacement rate of 0.12 mm/min until specimen failure or 15% axial strain. The experimental design focused on isolating the effects of MBC incorporation (0–8% w/w) and confining pressure on shear strength evolution. Untreated and treated specimens were subjected to identical saturation protocols and drainage conditions to ensure comparability. Real-time pore pressure and axial load measurements were synchronized at a 10 Hz sampling frequency, enabling precise reconstruction of stress–strain relationships under undrained shearing.

3.5. SEM Tests

Post-treatment specimens underwent a sequential stabilization protocol prior to microstructural analysis. MICP–MBC composites were initially cured under controlled hydrating conditions (30 °C, 90% RH) for 7 days to facilitate complete carbonate precipitation, followed by ambient air-drying (25 ± 2 °C, 50% RH) for an equivalent duration to achieve moisture equilibrium. Fresh fracture surfaces perpendicular to bedding planes were generated through cryogenic fracturing in liquid nitrogen to preserve native fabric architecture. Microstructural evaluation was conducted using a field-emission scanning electron microscope (JSM-5600LV, JEOL Ltd., Tokyo, Japan) operated at 15 kV accelerating voltage with a working distance of 10 mm [30].

4. Results and Discussions

4.1. Soil pH

Soil pH critically governs microbial activity and carbonate mineralization during MICP. As shown in Figure 3, MBC amendment induced a dose-dependent pH increase from 8.2 (untreated loess) to 8.53 (8% MBC). This alkalization arose from three synergistic mechanisms. First, alkaline minerals such as calcium oxide and potassium oxide within MBC released hydroxyl ions, neutralizing soil acidity. Second, microbial urea hydrolysis by S. pasteurii ATCC 11859 generated carbonate and ammonium ions, amplifying alkalinity. The calcium carbonate precipitation mechanism observed in this study aligns with previous reports using the analogous strain S. pasteurii DSM 33 under similar biocementation conditions [31]. Notably, our integration of modified biochar appears to enhance crystal nucleation patterns when compared with S. pasteurii DSM 33-based approaches [31], suggesting potential synergistic effects between microbial activity and carbonaceous substrates in loess modification. Third, MBC’s cation exchange capacity immobilized acidic ions, including aluminum and hydrogen, suppressing acidification. The pH range of 8.2 to 8.53 optimized calcite precipitation while sustaining microbial viability, avoiding both undersaturated conditions (pH < 8) and inhibitory alkalinity (pH > 9). Enhanced pore filling by calcium carbonate, facilitated by MBC’s nucleation sites, correlated with pH elevation, directly improving shear strength (52% increase) and carbonate content (38% increase). These results demonstrate that MBC integration refines the biogeochemical niche for MICP, enabling efficient soil stabilization without compromising microbial functionality.

4.2. Soil EC

Soil EC, a critical indicator of soluble ion mobility, reflects ionic byproduct accumulation during MICP. As shown in Figure 4, EC decreased progressively from 288 μs/cm (MICP-only) to 143 μs/cm (MICP + 8%MBC), demonstrating MBC efficacy in ionic regulation. This reduction arises from three mechanisms: (1) MBC’s porous structure (–COOH/–OH functionalized) adsorbs NH₄⁺ and Ca²⁺ ions from urea hydrolysis; (2) enhanced CaCO₃ precipitation on MBC surfaces immobilizes free Ca²⁺; and (3) alkaline pH (8.2–8.53) promotes ion speciation toward less soluble forms. The inverse correlation between MBC content and EC (R² = 0.94) highlights biochar’s dose-dependent ionic buffering. EC below 200 μs/cm (achieved at ≥4% MBC) minimizes osmotic stress on S. pasteurii ATCC 11859, sustaining 85% urease activity versus 62% in MICP-only groups. Reduced ionic mobility also improves hydraulic conductivity by stabilizing double-layer structures around soil particles. These results validate MBC’s dual role as a biomineralization catalyst and salinity moderator, resolving a key limitation of conventional MICP while supporting eco-friendly soil stabilization.

4.3. Permeability

4.3.1. Influence of Cementation Reagent Concentration on Permeability

The permeability (k) of MICP-treated loess demonstrates a nonlinear dependency on the concentration of cementation reagent under varying confining pressures (σ₃), as illustrated in Figure 5. At a fixed σ₃, k decreases sharply with initial increases in reagent concentration (0→0.5 M), followed by progressively attenuated reductions at higher concentrations (0.5→1.5 M). For instance, under σ₃ = 100kPa, k declines by 40.7% (from 30.2 × 10⁻⁵ cm/s to 17.9 × 10⁻⁵ cm/s) as the concentration rises from 0 to 0.5 M, but only 20.7% (from 17.9 × 10⁻⁵ cm/s to 14.2 × 10⁻⁵ cm/s) from 0.5 to 1.0 M. This nonlinear trend reflects pore-filling thresholds: lower reagent concentrations preferentially occlude macropores (>10 μm) via calcium carbonate bridging, while higher concentrations target smaller mesopores (1–10 μm) with diminishing efficiency.

Confining pressure synergistically enhances the permeability reduction driven by reagent concentration. At 0.5 M, increasing σ₃ from 50 to 400 kPa reduces k by 34.3% (from 20.1 × 10⁻⁵ cm/s to 13.2 × 10⁻⁵ cm/s), compared with a 61.9% reduction (from 39.1 × 10⁻⁵ cm/s to 14.9 × 10⁻⁵ cm/s) for untreated loess. This amplification arises from dual mechanisms: (1) Pore structure compression under elevated σ₃ narrows pore throats, improving CaCO₃ precipitation efficiency and (2) stress-aligned particle contacts promote continuous cementation chains rather than isolated nodules. The optimal reagent concentration for permeability control lies at 1.0 M, beyond which further increases yield marginal improvements (e.g., only a 9.3% kk reduction at 400 kPa when increasing from 1.0 to 1.5 M). This aligns with the pore-clogging saturation observed microscopically, where ≥1.0 M concentrations achieve 75–82% pore throat coverage. Field applications should prioritize 1.0 M cementation reagent under moderate confining pressures (100–200 kPa) to balance permeability targets (achieving k ≤ 12 × 10⁻⁵ cm/s) and material costs.

4.3.2. Influence of MBC Content on Permeability

The permeability (k) of MICP-treated loess demonstrates a nonlinear reduction with increasing MBC content under varying confining pressures (σ₃), as shown in Figure 6. At fixed σ₃, k decreases sharply with MBC addition up to 6% (w/w), followed by diminishing improvements at higher dosages (6–8%). For instance, under σ₃ = 200 kPa, k declines by 49.2% (from 25.8 × 10⁻⁵ cm/s to 13.9 × 10⁻⁵ cm/s) at 4% MBC and further reduces to 6.1 ×10⁻⁵ cm/s (76.4% reduction) at 8% MBC. This trend reflects two distinct phases: (1) rapid pore occlusion (0–6% MBC), where biochar’s high surface area and hydroxyl/carboxyl functional groups enhance CaCO₃ nucleation and macropore (>10 μm) filling and (2) saturation-limited refinement (6–8% MBC), where excessive MBC agglomeration restricts reagent diffusion into micropores, limiting additional permeability reduction.

Elevated confining pressures synergistically enhance MBC’s efficacy. At 6% MBC, increasing σ₃ from 50 to 400 kPa reduces k by 38.2% (from 11.0 × 10⁻⁵ cm/s to 6.8 × 10⁻⁵ cm/s), compared with a 49.1% reduction (from 37.1 × 10⁻⁵ cm/s to 18.9 × 10⁻⁵ cm/s) for MICP-only specimens. This amplification arises from stress-induced pore throat contraction (average radius reduction: 45–68%) and improved CaCO₃ -MBC interfacial bonding under compression.

4.4. Consolidated Undrained Triaxial Shear Testing

4.4.1. Stress–Strain Behavior of Untreated Loess

The stress–strain response of untreated Malan loess under varying confining pressures reveals a fundamental transition in mechanical behavior, as shown in Figure 7. At low confining pressures, the material exhibits strain-softening characteristics, where the deviatoric stress reaches a peak followed by a marked decline due to localized shear band formation and structural collapse. This post-peak degradation is driven by rapid dilatancy and the propagation of failure planes, which destabilize the soil fabric. In contrast, under high confining pressures, the loess demonstrates strain-hardening behavior, with deviatoric stress stabilizing after reaching its peak value. Elevated confinement suppresses dilatancy and compresses the soil matrix, promoting particle interlocking and distributed microcracking rather than macroscopic rupture.

The transition from softening to hardening reflects the interplay between confinement and soil fabric evolution. At low pressures, limited lateral constraints allow shear bands to propagate freely, leading to brittle failure. As confining pressure increases, particle rearrangement and enhanced frictional resistance stabilize the post-peak response, aligning the material with a critical state behavior where volumetric changes are minimized. This confinement-dependent strength amplification underscores the critical role of geostatic stress in governing loess stability, with deeper layers inherently more resistant to collapse due to stress-induced densification.

4.4.2. Stress–Strain Behavior of MICP–MBC Modified Loess

The stress–strain relationships of MICP–MBC modified loess under varying MBC content and cementation reagent concentrations are illustrated in Figure 8, comprising four subplots that compare deviatoric stress (σ₁ − σ₃) evolution across confining pressures. At a constant cementation reagent concentration (0.5 mol/L), increasing MBC content (Figure 8a,b) enhances deviatoric stress across all confining pressures. This improvement arises from biochar’s dual role as a nucleation template and structural reinforcer. MBC particles provide abundant sites for calcium carbonate precipitation, forming continuous cementation bonds between soil particles. Additionally, the rigid carbon skeleton of MBC interlocks with the loess matrix, redistributing stress more uniformly and suppressing localized shear failure. Under higher confining pressures, lateral compression forces MBC particles into closer contact with the soil fabric, amplifying interparticle friction and stabilizing post-peak behavior.

When MBC content is held constant (2%), increasing cementation reagent concentration (Figure 8a,c) significantly strengthens the deviatoric stress response. Higher reagent concentration enhances calcium ion availability, accelerating urea hydrolysis and generating denser calcium carbonate precipitation. This pore-filling effect occludes critical flow paths and reinforces particle bonding, particularly under low confining pressures where reagent diffusion is less restricted. At elevated confining pressures, compressed pore throats limit reagent mobility, but increased concentration compensates by ensuring sufficient reactant supply to sustain mineralization in constrained voids. The combined increase in MBC content (6%) and cementation reagent concentration (1.0 mol/L) (Figure 8b,d) maximizes deviatoric stress across all confinement levels. MBC preferentially nucleates calcium carbonate in macropores, while high-concentration reagents fill mesopores, creating a hierarchical reinforcement network. This multiscale occlusion reduces porosity and enhances load transfer efficiency. Under high confining pressures, the compressed MBC–soil interface optimizes carbonate adhesion, transforming brittle loess into a ductile composite resistant to strain localization.

4.5. Scanning Electron Microscope Microstructural Analysis

The scanning electron microscopy (SEM) analysis of untreated and modified loess (Figure 9) reveals critical insights into the microstructural mechanisms governing the efficacy of MICP and MBC integration. Untreated loess (Figure 9a) exhibits a metastable, porous fabric dominated by loosely packed silt particles and interconnected macropores, rendering it prone to hydro-collapse. MICP treatment alone (Figure 9b) introduces calcium carbonate (CaCO₃) precipitates, primarily as rhombohedral calcite crystals bridging adjacent particles. However, uneven crystal distribution leaves residual macropores and localized cementation clusters, explaining the limited hydraulic and mechanical improvements observed in MICP-only groups.

The incorporation of MBC fundamentally transforms this microstructure. At 2% MBC (Figure 9c), biochar particles act as nucleation templates, promoting smaller and more uniformly distributed CaCO₃ crystals that fill mesopores and coat particle surfaces. This dual-phase reinforcement—where MBC provides structural scaffolding and CaCO₃ enhances particle bonding—significantly reduces pore connectivity while maintaining microbial activity in preserved macropores. At 6% MBC (Figure 9d), a hierarchical reinforcement network emerges: biochar aggregates interlock via CaCO₃ bridges, forming continuous load-bearing chains, while residual NaCl byproducts are sequestered within MBC’s microporous structure, mitigating salt-induced degradation.

The stabilization mechanism of loess hinges on the controlled crystallization of calcium carbonate (CaCO₃), a process inherently aligned with crystalline materials science. SEM analysis revealed that MBC incorporation directs CaCO₃ polymorphism, favoring rhombohedral calcite formation at higher biochar content (6%) due to epitaxial growth on MBC’s oxygen-functionalized surfaces. This phase selectivity, transitioning from metastable spherical vaterite (dominant at 2% MBC) to thermodynamically stable calcite, exemplifies Ostwald’s step rule, where substrate chemistry governs nucleation kinetics and polymorphic transitions. Calcite’s crystallographic coherence and interfacial alignment with loess particles enhanced mechanical interlocking, directly correlating with the observed 52% strength gain. These findings demonstrate how biochar-mediated crystallization tailors crystal architecture (phase, morphology, orientation) to optimize geotechnical functionality, bridging microbial mineralization with crystalline materials engineering.

By elucidating CaCO₃’s crystallization pathways, including nucleation templating, phase selection, and stress-aligned growth, this work advances the design of bio-inspired crystalline composites. The interplay between microbial activity, biochar surface chemistry, and crystal–matrix interactions provides a novel framework for engineering crystalline materials in porous media, resonating with the emphasis on structure–property relationships in functional crystalline systems.

5. Conclusions

This study systematically investigates the synergistic effects of microbial-induced carbonate precipitation (MICP) and modified biochar (MBC) on the engineering properties of collapsible loess. The integration of MBC with MICP not only enhances mechanical strength and hydraulic stability but also addresses critical limitations of standalone MICP, such as uneven biomineral distribution and ionic byproduct accumulation. Through multi-scale characterization and controlled experimentation, the research establishes a framework for optimizing biochar-enhanced biomineralization in geotechnical stabilization. Key conclusions are drawn as follows:

(1)

Combining MBC (4–6% w/w) with MICP (1.0 mol/L cementation reagent) increases shear strength by 52% and reduces hydraulic conductivity by 72% compared with untreated loess. The biochar’s nucleation sites optimize calcium carbonate distribution, while its porous structure balances pore occlusion and microbial activity.

(2)

Elevated confining pressures (200–400 kPa) transform loess from a brittle (strain-softening) to a ductile (strain-hardening) behavior by enhancing particle interlocking and suppressing dilatancy. Peak strength quadruples under 400 kPa confinement, emphasizing stress-dependent design for deep foundations.

(3)

SEM analysis revealed that MBC stabilizes dual-scale pore networks, in which macropores sustain microbial colonization, while mesopores are occluded by CaCO₃-MBC composites. Residual salts are sequestered within biochar pores, mitigating efflorescence risks.

(4)

Optimal treatment parameters (6% MBC, 1.0 mol/L reagent, 200 kPa confinement) achieve 85% of maximum strength gain at 50% lower reagent cost, offering a scalable solution for eco-friendly infrastructure in loess regions.