Gel Foam Loaded with Oxygen-Consuming Bacterial Communities for the Coupled Inhibition of Lignite Spontaneous Combustion

Abstract

To improve the durability of microbial inhibition of coal spontaneous combustion, oxygen-consuming bacterial communities from mining environments were immobilized in gel foam to construct a bioactive carrier for lignite protection. Two previously screened aerobic communities, D₂ and D₃, were loaded into four candidate gel-foam systems and evaluated in terms of carrier stability, water retention, bacterial compatibility, and functional retention. The sodium alginate-based Y₁ formulation exhibited the best overall carrier performance and maintained sustained microbial oxygen consumption. Importantly, after foam collapse and air replenishment, the residual Y₁ gel phase remained metabolically active and decreased the O₂ volume fraction from approximately 21% to approximately 10%, demonstrating that microbial function could persist beyond the lifetime of the foam structure. Scanning electron microscopy (SEM) and X-ray diffraction (XRD) analyses showed that the bacterial-loaded gel foam coated the coal surface and filled pores and fissures, while no major reconstruction of the overall XRD diffraction profile was observed after treatment. Thermogravimetric/derivative thermogravimetric (TG-DTG) and kinetic analyses further showed that Y₁-D₂ and Y₁-D₃ shifted the T₂/T₃ temperatures from 153.20/309.91 °C to 194.81/333.07 °C and 192.35/327.13 °C, respectively, and increased the apparent apparent activation energy of the oxygen-absorption stage by 93.59% and 109.20%. The inhibition results from the coupled effects of physical sealing, water retention and bacterial immobilization, sustained biological oxygen consumption, and local gas-environment modification. This work highlights gel foam as a functional microbial carrier rather than solely as a passive fire-retardant material.

1. Introduction

Coal spontaneous combustion remains one of the major hazards threatening coal-mine safety. In goafs, residual coal is readily exposed to air leakage and undergoes low-temperature oxidation, which is a major cause of spontaneous combustion incidents. , Spontaneous combustion in goafs is characterized by a wide distribution range, concealed ignition sources, difficulty in early detection, continuous oxidation under oxygen-deficient conditions, and a high risk of reignition. It not only affects the safe and efficient extraction of coal but may also induce secondary disasters, including methane explosions, smoke poisoning, and equipment damage. − Therefore, the development of long-lasting and environmentally compatible technologies for preventing and controlling coal spontaneous combustion in complex goaf environments remains an urgent issue in mine fire management.

At present, the prevention and control of coal spontaneous combustion in mines mainly rely on physicochemical methods, such as grouting, inert gas injection, and the spraying of chemical inhibitors. − Although these techniques can reduce coal temperature, isolate oxygen, or delay oxidation reactions to some extent, they still have limitations in fractured coal bodies and goaf environments. The migration and diffusion of grouting materials are strongly affected by pore and fracture structures, making it difficult to achieve uniform coverage in deep or highly permeable regions and potentially causing material loss and secondary contamination. Inert gases usually have a short residence time and require continuous replenishment to maintain an effective inerting concentration, while their distribution in goafs is easily affected by air leakage pathways and spatial structures. In addition, some chemical inhibitors suffer from limited environmental compatibility and short effective duration. − In recent years, microbial approaches for fire prevention and suppression have attracted attention because of their environmental compatibility and potential for sustained action. , Oxygen-consuming microbial communities can reduce local oxygen concentrations through metabolic activity and, under suitable conditions, decrease the contact probability between coal and oxygen, thereby providing a biological route for inhibiting coal spontaneous combustion. ,

In our previous study, the mine-derived D₂ and D₃ bacterial communities exhibited strong oxygen-consuming activity and suppressed lignite oxidation by delaying characteristic oxidation behavior, reducing CO release, and modifying reactive functional groups. Nevertheless, these results were obtained mainly with free bacterial suspensions under controlled laboratory conditions. For application in complex goaf environments, free microbial communities remain vulnerable to variations in temperature, pH, moisture, oxygen supply, and nutrient conditions. In addition, insufficient adhesion and retention within fractured coal may cause cell loss from the target region and a progressive decline in sustained oxygen-consuming capacity. Therefore, the simple screening of efficient oxygen-consuming microorganisms does not by itself ensure persistent fire-prevention performance; a carrier capable of immobilizing cells, maintaining a moist microenvironment, improving spatial retention, and preserving metabolic activity is required. Gel foam is attractive for this purpose because it combines the high mobility and fissure-filling ability of foam with the water-retention, structural stability, and sealing functions of a gel network, thereby providing both a physical oxygen barrier and a protective microenvironment for oxygen-consuming bacterial communities.

Gel foam is a promising carrier material because it can protect microbial communities while adapting to complex goaf conditions. It combines the high fluidity, diffusivity, and fissure-filling ability of foam with the water-retention, foam-stabilization, and sealing properties of gel systems, and has therefore been regarded as a composite material for the control of residual coal and hidden fire sources in goafs. , Compared with ordinary water-based foam, gel foam is less prone to rapid drainage and collapse and can form a relatively stable covering layer on the coal surface. Compared with conventional gels, it has better spreading and penetration abilities, making it more suitable for fragmented coal masses and complex fissure spaces. , Previous studies have shown that the incorporation of polysaccharides, polymers, clay minerals, or natural polyphenols can further improve the bubble-film strength, water-retention capacity, and thermal stability of gel foam. Lu et al. developed a novel gel foam with antioxidant properties. Fan et al. found that a xanthan gum/guar gum/polyacrylamide gel foam could inhibit low-temperature coal oxidation by improving foam-film stability, increasing the reaction energy barrier, and suppressing the reactivity of active functional groups. Shi et al. demonstrated, from the perspective of foam-film evolution and the fire-extinguishing process, that gel-stabilized foam can slow the diffusion of bubbles and gas. Han et al. developed a sodium alginate/calcium lactate dual-cross-linked biomass gel foam, which significantly improved long-term stability and green fire-retardant performance. However, most existing studies on gel foams have focused on their intrinsic sealing, water-retention, and thermal fire-retardant properties, whereas their use as specialized carriers for oxygen-consuming microorganisms remains insufficiently explored. In particular, it remains unclear how different gel-foam formulations affect microbial activity, sustained oxygen consumption, microbial retention after foam collapse, and the low-temperature oxidation behavior of coal.

In this study, aerobic bacterial communities from mining environments were used as functional components to construct a gel foam loaded with oxygen-consuming bacterial communities, and its coupled inhibitory effects on lignite spontaneous combustion were investigated. First, aerobic bacterial communities were enriched and screened from deep mine wastewater, settled sludge, and soil from abandoned mine shafts. D₂, derived from settled sludge, and D₃, derived from abandoned-mine soil, were selected as the functional bacterial communities. Subsequently, four candidate gel-foam carriers were prepared, and their foaming performance, gelation behavior, foam stability, water retention, and bacterial-loading suitability were systematically compared to identify an appropriate sodium alginate-based bacterial-loaded gel-foam system. On this basis, closed-chamber oxygen-consumption tests, oxygen-consumption retention tests after foam collapse, SEM, XRD, and TG-DTG analyses were performed to evaluate the sustained oxygen-consuming capacity under sealed laboratory conditions of the bacterial-loaded gel foam, its coal-surface sealing characteristics, and its influence on the low-temperature oxidation process of lignite. This study aims to clarify the coupled inhibition mechanism of gel foam loaded with oxygen-consuming bacterial communities through physical sealing, water retention and bacterial immobilization, sustained oxygen consumption, and local gas-environment modification.

2. Experiments and Methods

2.1. Materials

Lignite samples were collected from Dongtan Coal Mine in Jining, Shandong Province, China. After collection, lump coal samples were vacuum-packed and transported to the laboratory. The surface oxidation layer was removed, and the coal samples were then crushed, sieved, and stored in sealed containers. Coal samples used for bacterial contact tests and surface morphology observations were sieved to a particle size of 0.25–0.50 mm. After drying at room temperature for 24 h, these samples were sterilized at 121 °C for 30 min, cooled to room temperature, and stored for subsequent experiments. Coal samples used for TG-DTG analysis were sieved to 80–100 mesh and stored in sealed bags after drying to minimize the influence of particle-size differences on thermal analysis. These preparation procedures were adopted to reduce uncontrolled variability associated with indigenous microorganisms, moisture condition, and particle size and were not intended to reproduce the heterogeneous conditions of an actual goaf.

Microbial samples were collected from the environment surrounding Dongtan Coal Mine, including deep mine wastewater, settled sludge, and soil from abandoned mine shafts, which were designated as D₁, D₂, and D₃, respectively. For settled sludge and abandoned-mine soil samples, 2.0 g of sample was added to 20 mL of sterile water. The mixture was thoroughly shaken, allowed to settle, and filtered to remove coarse particles. For the deep mine wastewater sample, 2.0 mL of wastewater was added to 20 mL of sterile water and mixed thoroughly. Subsequently, 10 mL of the supernatant from each sample was inoculated into 90 mL of LB medium for enrichment culture. After five successive passages, the changes in O₂ and CO₂ volume fractions in a sealed system were monitored using gas chromatography, and oxygen-consumption capacity was used as the criterion for selecting dominant microbial communities. The enriched communities were subsequently activated in nitrate mineral salts (NMS) medium before the oxygen-consumption tests. The composition of the NMS medium is listed in Table . The selected D₂ and D₃ communities were used for the subsequent preparation and evaluation of bacterial-loaded gel foams, and their microbial community compositions were analyzed by Shanghai Majorbio Bio-pharm Technology Co., Ltd.

1. Composition of the NMS Culture Medium.

Number	Reagents	Proportions(g/L)
1	NH4Cl	0.50000
2	KH2PO4·7H2O	0.40000
3	K2HPO4	0.40000
4	MgSO4·7H2O	0.30000
5	CaCl2·2H2O	0.02000
6	KNO3	1.60000
7	NaCl	0.30000
8	EDTA	0.02000
9	FeSO4·7H2O	0.00400
10	ZnSO4·7H2O	0.00034
11	CuSO4·5H2O	0.00130
12	MnSO4·H2O	0.00040
13	NaMoO4·2H2O	0.00024

2.2. Preparation of Bacterial-Loaded Gel Foams

To screen a gel-foam carrier suitable for oxygen-consuming bacterial communities, four candidate systems (Y₁–Y₄) were selected from previously reported gel-foam formulations. The selection was based on multiple criteria: foaming and spreading ability, controllable gelation without premature agglomeration or excessive drainage, sufficient structural stability and water retention, and compatibility with bacterial suspensions. Y₁–Y₄ were further chosen to represent different foaming, gelling, stabilizer, and crosslinking chemistries, allowing their bacterial-loading performance to be compared under a unified framework. −

For the Y₁ system, alkyl polyglucoside (APG, 0.2 g) and tea saponin (TS, 0.1 g) were used as composite foaming agents, sodium alginate (SA, 0.4 g) was used as the gelling agent, and calcium lactate (CL, 0.1 g) was used as the cross-linking agent. For the Y₂ system, lauramidopropyl betaine and sodium α-olefin sulfonate were used as composite foaming agents, with a total dosage of 0.3 g; SA (0.5 g) was used as the gelling agent; and polyaluminum chloride (PAC, 0.8 g), disodium ethylenediaminetetraacetate (EDTA, 0.5 g), and glucono-δ-lactone (GDL, 2.0 g) were used as the cross-linking and curing system. In the Y₃ system, SDS and AES were mixed at a mass ratio of 1:1 as the foaming agent, with a total mass fraction of 0.7%. Sodium carboxymethyl cellulose (CMC) was used as the gelling agent; guar gum and xanthan gum were mixed at a mass ratio of 1:1 as the foam stabilizer; and an aluminum citrate system was used as the cross-linking and curing component. The mass fractions of CMC, cross-linking agent, foam stabilizer, and composite foaming agent were 1.5–1.6%, 4–5%, 0.35–0.4%, and 0.7%, respectively. The Y₄ system used a composite foaming agent consisting of SDS and APG, with a total mass fraction of 0.92%. Xanthan gum and guar gum were mixed at a mass ratio of 1:1 as the foam stabilizer, with a total mass fraction of 0.49%; the mass fractions of polyacrylamide (HPAM) and aluminum citrate (AlCit) were 0.34% and 5%, respectively.

The bacterial-loaded gel foam was prepared using a three-component method. The foaming component was prepared as Solution A, the gelling component as Solution B, and the cross-linking and curing component as Solution C. During preparation of the bacterial-loaded system, the aqueous phase in Solution A was replaced with an activated bacterial suspension, allowing bacterial cells to disperse in the foaming system. Solution B was then added and fully foamed. After foam formation, Solution C was added to trigger cross-linking and curing, thereby immobilizing the bacteria within the gel-foam network. Sterile gel foam, in which the bacterial suspension was replaced with sterile culture medium or sterile water, was used as the control. The D₂ and D₃ bacterial suspensions were activated before use, and their concentration and inoculation volume were kept consistent across all groups.

2.3. Evaluation of Gel Foam Properties

Foaming performance was evaluated by measuring the foam volume. Immediately after foaming, the foam was transferred to a graduated cylinder, and the total foam volume was recorded. The differences between the sterile and bacterial-loaded groups were then compared. Foam stability was evaluated by drainage behavior and half-drainage time. Freshly prepared samples were allowed to stand under identical environmental conditions. The volume of liquid drained from the bottom was recorded at different time points, and the time required for the drained liquid to reach 50% of the final drainage volume was defined as the half-drainage time. ,

Gelation behavior was determined using the inversion method. Gelation was considered complete when the system showed no macroscopic flow and remained stationary for 30 s after inversion. Timing started immediately after the addition of the cross-linking and curing component. The sample was placed in a transparent container and allowed to stand, and it was slowly tilted or inverted at predetermined intervals. When the system no longer flowed as a whole and remained stable for 30 s without visible flow, gelation was considered complete, and the corresponding time was recorded as the gelation time. If localized agglomeration occurred before a stable foam structure formed, the system was considered to exhibit excessive gelation. If gelation was too slow and accompanied by obvious liquid separation and foam collapse, the system was considered to exhibit insufficient gelation.

Water retention was evaluated by the water loss rate. A gel sample with a specified mass was placed in a weighing dish, allowed to stand under identical environmental conditions, and weighed at predetermined time points. The water loss rate was calculated using eq :

$$W_t=\frac{m_0-mt}{m_0}\times 100\%$$ 1

where W_t is the water loss rate (%), m ₀ is the initial mass of the sample (g), and m_t is the mass at time t (g).

2.4. Microbial Loading Compatibility and Oxygen Consumption Tests

To evaluate the ability of the gel foam to maintain microbial activity, bacterial-loaded gel foams from each group were placed in sealed containers fitted with rubber stoppers, and the headspace volume was kept consistent among all groups. The samples were incubated under identical environmental conditions. At predetermined time points, headspace gas was collected using an airtight syringe, and the O₂ and CO₂ volume fractions were measured by gas chromatography. The reported O₂ and CO₂ values represent measured headspace volume fractions and were not converted into absolute molar gas inventories. The sample volume and sampling interval were kept consistent among all groups to minimize the influence of sampling disturbance on gas composition.

The foam collapse process was recorded by monitoring the stratification pattern and foam volume over time. During static standing, the foam layer, bottom liquid phase, and residual gel phase were observed at regular intervals. Collapse was defined as the point at which the foam layer had markedly decreased, the system had stabilized into distinct layers, and the continuity of the foam structure had significantly weakened. The corresponding time was recorded as the collapse time. After foam collapse, sterile air was injected into the sealed container to restore the headspace oxygen concentration to a level close to that of ambient air. The container was then sealed again for further incubation, and the changes in O₂ and CO₂ volume fractions were measured to evaluate the survival of the microbial community in the residual gel phase and its capacity for sustained oxygen consumption.

2.5. Preliminary Phytocompatibility Test

A wheat seedling growth experiment was conducted to preliminarily evaluate the phytocompatibility of the bacterial-loaded gel foam. Wheat seeds from the same batch were selected and uniformly sown in seedling trays containing the same growth medium. The seedlings were cultivated under consistent temperature, light, and watering conditions. After emergence and stabilization of seedling growth, three groups were established: a clean water control group, a bacterial suspension treatment group, and a bacterial-loaded gel foam treatment group. Each group was sprayed with an equal volume of the corresponding treatment solution at regular intervals, and obvious runoff was avoided during spraying. Seedling emergence, growth status, leaf color, and wilting were observed, and the final emergence rate was calculated.

2.6. Characterization of Coal Treated with Bacterial-Loaded Gel Foam

To analyze the coating, filling, and retention characteristics of the bacterial-loaded gel foam on the coal surface, the optimized Y1 formulation was selected for interaction with coal samples. For SEM analysis, the treated samples were freeze-dried, and regions with relatively intact coating layers were selected for gold sputtering. The film-forming and coating behavior of the gel foam on the coal surface, the filling of pores and fissures, and the retention of the residual gel layer after foam collapse were observed.

For XRD analysis, both raw and treated coal samples were dried, ground using an agate mortar, and sieved through standard mesh screens to 80–100 mesh (0.15–0.18 mm). Diffraction patterns were collected using Cu Kα radiation (λ = 1.5406 Å) at 40 kV and 40 mA, with a scanning range of 5°–80° (2θ) and a scanning speed of 2°/min, under identical conditions for all samples. The positions and intensities of the main diffraction peaks were compared to analyze the effects of bacterial-loaded gel foam treatment on the mineral phases and surface structure of the coal samples.

2.7. Thermogravimetric Analysis and Kinetic Calculations

To evaluate the effect of bacterial-loaded gel foam on the thermal behavior of lignite during oxidation, coal samples were uniformly mixed with gel foam at a mass ratio of 5:2. This ratio was used as a standardized laboratory mixing condition to ensure sufficient coal–gel contact during thermal analysis and does not represent an optimized field dosage. Four samples were prepared: raw coal, Y₁-treated coal, Y₁-D₂-treated coal, and Y₁-D₃-treated coal. The treated coal samples were freeze-dried before thermogravimetric analysis. Freeze-drying was used to place the treated samples under a relatively comparable initial moisture condition before TG-DTG testing and was not intended to reproduce the wet-state operating condition of the gel foam in a goaf. Accordingly, the TG-DTG analysis focuses on comparing the thermal-oxidation response of the treated solid-phase coal samples under standardized test conditions. Thermogravimetric analysis was performed using a TA Discovery TGA 550 (TA Instruments, USA). Approximately 10 mg of each sample was placed in an open alumina (Al₂O₃) crucible (70 μL, no lid) and heated from 30 to 800 °C at 10 °C/min under an air flow of 50 mL/min. Temperature and sensitivity calibration of the instrument was performed before sample measurements using Curie point standards (ASTM E1582) and built-in calibration weights, respectively. TG and DTG curves were recorded as functions of temperature.

The characteristic temperature points for each treated coal sample were extracted from the TG-DTG curves, including the initial moisture loss temperature T₁, the characteristic temperatures T₂/T₃ of the oxygen absorption weight gain stage, the ignition temperature T₄, the temperature corresponding to the maximum mass-loss rate T₅ and the combustion completion temperature T₆. To further evaluate the combustion behavior of different coal samples, the combustion stability index C and the comprehensive combustion index S were calculated: ,

$$C=\frac{DT{G}_{\max }}{T_i^2}$$ 2

$$S=\frac{DT{G}_{\mathrm{mean}}DT{G}_{\max }}{T_i^2{T}_b}$$ 3

In eqs and , DTG _max and DTG _mean represent the maximum and average mass loss rates, respectively; T _i and T _b are the ignition temperature and burnout temperature, respectively.

The kinetic parameters of coal oxidation were calculated using the Coats–Redfern method. The conversion rate α of the coal sample was calculated using eq :

$$\alpha =\frac{m_0-m_t}{m_0-m_{\infty }}$$ 4

Where m ₀, m _t and m _∞ represent the initial mass, the mass at temperature T, and the final mass of the sample, respectively. For a first-order reaction, the mechanism function is expressed as eq :

$$g(\alpha )=-\ln (1-\alpha )$$ 5

The Coats–Redfern integral is expressed as

$$\ln \left[\frac{g(\alpha )}{T^2}\right]=\ln \left[\frac{AR}{\beta E}(1-\frac{2RT}{E})\right]-\frac{E}{RT}$$ 6

In eq , A is the pre-exponential factor, min^–1; R is the gas constant, 8.314 J/(mol·K); E is the apparent activation energy of the reaction, kJ/mol; β is the heating rate, K/min; and T is the absolute temperature. Based on the linear relationship between ln­[g(α)/T ²] and 1/T, the apparent apparent activation energy of each oxidation stage was calculated from the slope of the fitted line. The first-order Coats–Redfern treatment was used here as a simplified stage-wise comparative model and does not imply that the complex oxidation of lignite follows a unique elementary first-order mechanism. Accordingly, the calculated kinetic parameters are interpreted as apparent activation energies for comparison among samples under identical experimental and model assumptions.

3. Results and Discussion

3.1. Screening of Oxygen-Consuming Bacterial Communities and Analysis of Community Composition

Aerobic bacterial communities form the functional basis for the construction of bacterial-loaded gel foam. In our previous work, three aerobic bacterial communities, D₁, D₂, and D₃, were enriched and isolated from deep mine wastewater, settled sludge, and soil from disused mine shafts, respectively, and their inhibitory effects on lignite oxidation and free-radical activity were systematically investigated. Based on the previous screening results, D₂, derived from settled sludge, and D₃, derived from abandoned-mine soil, were selected as the functional bacterial communities for constructing the oxygen-consuming bacterial-loaded gel foam. Preliminary results showed that both D₂ and D₃ had strong oxygen-consuming capacities and could decrease the O₂ volume fraction in a sealed system from approximately 20% to below 5% within a short period. This rapid oxygen depletion provides a functional basis for the coupled inhibition process involving biological oxygen consumption and gel sealing.

The microbial community compositions are shown in Figure . The dominant genera in D₂ mainly included Citrobacter, Macellibacteroides, Clostridium sensu stricto, Acinetobacter, and Lachnoclostridium, among which Citrobacter and Macellibacteroides accounted for 40.89% and 33.95%, respectively. The dominant taxa in D₃ mainly included Pseudomonas aeruginosa, Bacillus, and Bacteroidota. Notably, D₃ contained a high relative abundance of Pseudomonas aeruginosa, an opportunistic pathogen. Therefore, D₃ is used in the present study only as a laboratory proof-of-concept consortium and should not be interpreted as suitable for direct field release. Although P. aeruginosa was not detected among the dominant taxa of D₂, this does not establish the biosafety of D₂ for environmental release. Strain-level identification, virulence-factor and antimicrobial-resistance screening, together with broader ecological and occupational-health risk assessments, are required before any field application of either consortium or related microbial formulations. Because the oxygen-consuming capacity of free bacterial communities is strongly affected by carrier protection, moisture retention, and spatial retention, the compatibility of D₂ and D₃ with different gel-foam systems and their coupled inhibitory effects on low-temperature lignite oxidation were further evaluated.

1.

Results of bacterial community identification: (a) D₂ from settled sludge; (b) D₃ from abandoned-mine soil.

3.2. Performance of Gel-Foam Carriers and Their Suitability for Bacterial Colonization

When used as carriers for oxygen-consuming bacterial communities, gel foams should possess good foaming capacity, controllable gelation behavior, structural stability, and water-retention performance. Foaming capacity determines the spreading and coverage ability of the material in coal pores and fissures. Gelation behavior affects foam solidification and bacterial immobilization. Foam stability and foam half-life reflect the ability of the system to maintain its structure after formation, whereas water retention is essential for the survival of bacterial communities and their sustained oxygen-consuming activity. Therefore, the foaming capacity, drainage stability, gelation time, and foam half-life of four candidate gel foams, Y₁–Y₄, were first evaluated, and the water-retention performance of the selected systems was further compared.

As shown in Figure a, all four gel foams could form a certain foam volume, but their responses to bacterial loading were different. Compared with the sterile control, the foam volume of Y₁ decreased by only 7.78% and 5.56% after loading with D₂ and D₃, respectively, indicating that the bacterial suspension had little effect on its foaming capacity. The foam volumes of Y₂ decreased by 17.27% and 12.00%, respectively, whereas those of Y₃ decreased by 14.71% and 16.18%, indicating moderate inhibition of foaming. By contrast, Y₄ showed the largest decreases, 30.00% and 36.67%, respectively, suggesting that its foaming performance was the most sensitive to bacterial-suspension disturbance.

2.

Coupled material performance tests: (a) foaming performance; (b) drainage stability; (c) gelation time; (d) foam half-life.

The drainage stability results further showed clear differences among the systems after bacterial loading. In Y₄, water separation was accelerated after bacterial loading, and the foam structure was prone to destabilization. Although Y₂ exhibited a relatively large initial foam volume, its stability after loading with D₂ and D₃ differed markedly, indicating that the system was strongly affected by the properties of the bacterial suspension. In comparison, Y₁ showed a relatively stable drainage process after bacterial loading, indicating good structural tolerance. For microbial carriers, high foaming volume alone is insufficient; if the foam rapidly drains or collapses after formation, it cannot provide a stable support environment for bacterial retention and metabolism.

The gelation time results are shown in Figure c. All measurements were performed in triplicate, and results are expressed as mean ± standard deviation. The gelation time of the sterile Y₁ system was 10.0 ± 0.7 min and changed only slightly after loading with D₂ or D₃, indicating that its cross-linking and curing process remained stable. The sterile Y₃ system had a gelation time of 11.2 ± 0.7 min, which also remained within a controllable range after bacterial loading. In contrast, the gelation time of the sterile Y₂ system was 9.8 ± 0.6 min but decreased to 5.2 ± 0.6 and 6.5 ± 0.9 min after loading with D₂ and D₃, respectively. This marked acceleration in gelation may cause local curing before a uniform foam structure is fully formed, resulting in uneven pore distribution and local agglomeration, which is unfavorable for practical delivery and uniform spreading.

The foam half-life results are shown in Figure d. The foam half-life of the sterile Y₄ system was 15.0 d, but it decreased to 8.0 and 6.0 d after loading with D₂ and D₃, respectively, indicating poor structural retention after bacterial loading. The foam half-life of the sterile Y₂ system was 12.0 d; after loading with D₂, it increased to 15.0 d, whereas after loading with D₃, it decreased to 9.0 d, showing obvious differences between the two bacterial communities. The sterile Y₃ system had a foam half-life of 30.0 d, which decreased to 21.0 and 26.0 d after loading with D₂ and D₃, respectively, but still remained relatively long. The foam half-life of the sterile Y₁ system was 18.0 d and remained 16.0 and 15.0 d after loading with D₂ and D₃, respectively. The relatively small decrease indicates that Y₁ maintained good foam stability under bacterial-loaded conditions.

Based on the comprehensive evaluation of foaming capacity, drainage stability, gelation time, and foam half-life, Y₄ showed a significant decrease in foam volume and insufficient stability after bacterial loading, while Y₂ exhibited excessively rapid gelation and large differences in bacterial compatibility. Therefore, neither Y₂ nor Y₄ met the requirements for stable bacterial loading. In contrast, Y₁ and Y₃ maintained good foaming, gelation, and structural retention properties after bacterial loading and were selected for subsequent water-retention tests.

As shown in Figure , the water loss rates of the Y₁ and Y₃ bacterial-loaded gel foams increased with standing time at room temperature, showing a rapid increase in the early stage and a slower increase in the later stage. This trend indicates that water release gradually shifted from rapid free-water loss to slower bound-water release. Overall, the water loss rates of Y₁ were lower than those of Y₃ at all tested time points, indicating better water-retention performance. Good water retention is favorable for the formation of a moist residual gel phase after foam dissipation, thereby providing a microenvironment for bacterial retention and metabolic activity. Therefore, both Y₁ and Y₃ were further used for oxygen-consumption validation, with Y₁ showing a more pronounced advantage in water retention.

3.

Water loss rate.

3.3. Oxygen Consumption Properties of Bacterial-Loaded Gel Foam and Its Retention Capacity after Collapse

3.3.1. Sustained Oxygen-Consumption Capacity

To further evaluate the ability of gel foam to maintain bacterial metabolic activity, Y₁ and Y₃ were selected for oxygen-consumption validation. As shown in Figure , panels (a) and (b) correspond to the Y₁ and Y₃ bacterial-loaded gel-foam systems, respectively. Both systems exhibited O₂ consumption and CO₂ accumulation, indicating that the bacterial communities retained metabolic activity within the gel-foam matrices. However, the two carriers differed significantly in their ability to sustain this function during the later stage.

4.

Changes in O₂ and CO₂ volume fractions over time: (a) Y₁ system; (b) Y₃ system.

In the Y₁ system, the O₂ volume fraction continuously decreased, and the O₂ and CO₂ curves intersected at approximately 120 h. By 150 h, the CO₂ volume fraction had reached approximately 30%, and it further increased to more than 60% at 250 h. These results show that the headspace gas composition of the Y₁ system continued to change throughout the 250 h laboratory observation period, consistent with sustained metabolic activity of the immobilized bacterial communities. In contrast, the O₂ concentration in the Y₃ system decreased more slowly and remained at approximately 14–15% at 150 h, while the CO₂ concentration was approximately 33%. After 150 h, the changes in gas composition gradually slowed, suggesting a decline in bacterial metabolic activity during the later stage. These results demonstrate that Y₁ not only had better material stability but was also more suitable for maintaining the sustained oxygen-consuming function of bacterial communities.

Because system pressure, dissolved CO₂, and absolute gas inventories were not simultaneously quantified, the O₂ and CO₂ volume-fraction data in Figure should be interpreted as changes in headspace gas composition rather than as a quantitative O₂–CO₂ mass balance. In addition, the oxygen-consumption experiment did not include a complete set of sterile-gel, free-bacteria, and inactivated-bacteria controls; therefore, the individual contributions of physical sealing and biological O₂ depletion cannot be quantitatively separated, and the present results do not establish a quantitatively demonstrated supra-additive interaction.

3.3.2. Foam-Collapse Process

Y₁ was selected for the detailed foam-collapse and post-collapse evaluation because the preceding carrier screening and oxygen-consumption tests showed that it provided the best overall combination of structural tolerance, water retention, and sustained microbial oxygen-consumption performance. Figure shows the stratification and collapse evolution of the Y₁ bacterial-loaded gel foam during static standing. With increasing standing time, gravitational drainage caused the system to gradually form an upper foam layer and a lower liquid layer. The foam layer became thinner and developed local cracks, but the residual gel phase retained a certain degree of continuity within the system. This residual gel phase is important for the bacterial-loaded system. On the one hand, it provides a moist microenvironment for the bacterial communities and reduces bacterial inactivation caused by water loss. On the other hand, the gel network helps retain bacterial cells and prevents their loss with the supernatant.

5.

Hierarchical structure changes of the Y₁ bacterial-loaded gel foam from 0 to 480 h.

3.3.3. Survival of Bacterial Communities and Oxygen Consumption after Collapse

The oxygen-consumption results after foam collapse are shown in Figure .

6.

Oxygen consumption after collapse and air replenishment of the Y₁ bacterial-loaded gel foam.

After the foam collapsed and air was replenished, the system continued to show obvious gas-conversion ability. The O₂ volume fraction decreased from approximately 21% to approximately 10%, while the CO₂ volume fraction increased from nearly 0% to approximately 15%. This indicates that the collapse of the foam structure did not completely inactivate the bacterial communities. Instead, the residual gel phase continued to support bacterial metabolism through entrapment and water retention. Therefore, the advantage of Y₁ lies not only in its initial foam coverage and structural stability but also in its ability to retain a bioactive residual gel phase after foam collapse, thereby prolonging the duration of oxygen consumption. Based on the combined evaluation of material performance and oxygen-consumption function, Y₁ was identified as the optimal bacterial-loaded gel foam.

3.4. Preliminary Phytocompatibility Assessment

A wheat seedling growth experiment was conducted to preliminarily evaluate the environmental compatibility of the Y₁ bacterial-loaded gel foam.

As shown in Figure , seedlings in the clean water control group, bacterial suspension group, and Y₁ bacterial-loaded gel foam group all emerged normally after 2 d of cultivation, with little difference in early growth. After 10 and 20 d of cultivation, seedlings in all groups continued to grow, and their leaf color and expansion remained generally normal. No obvious toxic symptoms, such as yellowing, wilting, growth arrest, or local death, were observed. Quantitative results showed that the final seedling emergence rates of the clean water control group, bacterial suspension group, and bacterial-loaded gel foam group were 93%, 92%, and 90%, respectively, with only slight differences among the groups. These results indicate that the Y₁ bacterial-loaded gel foam did not show an obvious inhibitory effect wheat seedling growth at the tested stage and showed no obvious phytotoxicity under the tested conditions. Because this assessment was limited to a 20 d plant-growth test using a single species, it should be regarded only as a preliminary phytocompatibility evaluation rather than a comprehensive environmental-safety assessment. Longer-term and broader ecological and biosafety evaluations are still required before field application.

7.

Growth of wheat seedlings in different periods.

3.5. Characteristics of the Interaction between Bacterial-Loaded Gel Foam and the Coal Surface

3.5.1. SEM Analysis

The SEM results are shown in Figure . The raw coal surface exhibited well-developed pores and fissures, which can provide pathways for oxygen diffusion and oxygen–coal contact. After treatment with Y₁ gel foam, a continuous coating layer appeared on the coal surface, and some pores and fissures were covered or filled by the gel film. This indicates that the Y₁ gel foam had good interfacial spreading and sealing abilities.

8.

SEM images of coal samples treated with bacterial-loaded gel foam.

After bacterial loading, bacterial cells were observed not only on the surface of the gel coating but also within the gel network, suggesting that the gel structure provided a habitat for bacterial retention. After foam collapse, residual gel still remained in coal fissures and surface depressions, and bacterial cells could still be observed. This further confirms that the Y₁ system retained interfacial adhesion and bacterial-retention capacity even after structural degradation.

3.5.2. XRD Analysis

The XRD results are shown in Figure . Raw coal, Y₁-treated coal, Y₁-D₂-treated coal, and Y₁-D₃-treated coal exhibited broadly similar overall diffraction profiles, including broad diffuse features in the range of 2θ ≈ 20–30°. Compared with raw coal, the treated samples showed changes in the positions and intensities of several weak diffraction features, particularly near 2θ ≈ 12°. These variations may be associated with crystalline constituents in the coal and/or deposits introduced by the gel treatment. Overall, no major reconstruction of the XRD diffraction profile was observed after treatment. However, because XRD provides limited information on disordered organic structures and surface chemical functionalities, the present results cannot determine whether specific oxygen-containing functional groups or the amorphous carbon structure were chemically altered. Such changes require further characterization by complementary techniques such as FTIR, Raman spectroscopy, or XPS.

9.

XRD patterns of raw and treated coal samples.

3.6. Analysis of TG-DTG Characteristic Temperatures and Coal Oxidation Kinetics

3.6.1. TG-DTG Analysis

To evaluate the effect of bacterial-loaded gel foam on the thermal behavior of lignite during oxidation, TG-DTG analyses were conducted on raw coal, Y₁-treated coal, Y₁-D₂-treated coal, and Y₁-D₃-treated coal. As shown in Figure , all coal samples exhibited typical stages, including moisture loss, oxygen-absorption weight gain, pyrolysis, and combustion-induced weight loss. Compared with raw coal, the TG-DTG curves of the Y₁-treated and bacterial-loaded groups shifted toward higher temperatures, indicating enhanced thermal stability and suppression of the low-temperature oxidation process.

10.

Thermogravimetric analysis diagram of each sample: (a) Raw coal; (b) Coal-Y₁; (c) Coal-Y₁-D₂; (d) Coal-Y₁-D₃.

In terms of characteristic temperatures, the T₁ value of raw coal was 41.84 °C. After treatment with Y₁, Y₁-D₂, and Y₁-D₃, T₁ increased to 44.37, 50.46, and 54.94 °C, respectively, with Y₁-D₃ showing the most pronounced delay during the initial moisture-loss stage. During the oxygen-absorption weight-gain stage, the T₂ and T₃ values of raw coal were 153.20 and 309.91 °C, respectively. After Y₁-D₂ treatment, they increased to 194.81 and 333.07 °C, while after Y₁-D₃ treatment, they increased to 192.35 and 327.13 °C. These results indicate that both bacterial-loaded gel foams significantly delayed the low-temperature oxygen-absorption weight-gain process.

During the combustion stage, the T₅ and T₆ values of raw coal were 512.68 and 556.75 °C, respectively. These values increased to 528.21 and 566.75 °C for Y₁-D₂ and to 525.76 and 565.17 °C for Y₁-D₃, indicating that the combustion process of the treated coal samples was delayed. Overall, Y₁ alone enhanced the thermal stability of lignite, and the inhibitory effect was further strengthened after loading with D₂ or D₃. Y₁-D₂ showed more pronounced increases in T₂, T₃, T₅, and T₆, indicating stronger inhibition during the oxygen-absorption and combustion stages, whereas Y₁-D₃ showed higher T₁ and T₄ values, suggesting a stronger delaying effect during the initial moisture-loss and ignition-transition stages. The characteristic temperature points of the coal samples are shown in Table .

2. Characteristic Temperatures of Coal Samples.

Coal sample	T1 (°C)	T2 (°C)	T3 (°C)	T4 (°C)	T5 (°C)	T6 (°C)
Raw coal	41.84	153.20	309.91	471.94	512.68	556.75
Coal-Y1	44.37	182.36	323.07	478.36	525.10	563.79
Coal-Y1-D2	50.46	194.81	333.07	478.78	528.21	566.75
Coal-Y1-D3	54.94	192.35	327.13	479.96	525.76	565.17

3.6.2. Analysis of Combustion Characteristic Parameters

The combustion characteristic parameters further confirmed the inhibitory effect of the bacterial-loaded gel foam on the oxidative combustion process of lignite. Compared with raw coal, both the ignition temperature and burnout temperature of Y₁-D₂ and Y₁-D₃ increased, indicating that the treated coal samples required higher temperatures to enter and complete the combustion stage.

Compared with raw coal, Y₁ treatment alone increased the ignition temperature from 471.94 to 478.36 °C and the burnout temperature from 556.75 to 563.79 °C, indicating that the gel matrix itself contributed to delaying the combustion process. The comprehensive combustion index S changed only slightly from 5.27450 × 10^–9 for raw coal to 5.26766 × 10^–9 for Coal-Y₁. After bacterial loading, S further decreased to 5.04486 × 10^–9 and 5.00320 × 10^–9 for Coal-Y₁-D₂ and Coal-Y₁-D₃, respectively. These results indicate that Y₁ itself affects the thermal behavior of lignite, whereas bacterial loading is associated with additional changes in the combustion characteristics. The comparison of combustion characteristic parameters is shown in Table .

3. Combustion Characteristic Parameters of Coal Samples.

Coal sample	Ignition temperature (°C)	Burnout temperature (°C)	Maximum rate of change in mass (%/°C)	Average rate of change in mass (%/°C)	Combustion Stability Index C	Combustion Index S
Raw coal	471.94	556.75	–0.8937	–0.73185	–4.01253E–06	5.27450E–09
Coal-Y1	478.36	563.79	–0.9224	–0.73678	–4.03084E–06	5.26766E–09
Coal-Y1-D2	478.78	566.75	–0.9115	–0.71904	–3.97635E–06	5.04486E–09
Coal-Y1-D3	479.96	565.17	–0.9087	–0.71683	–3.94467E–06	5.00320E–09

The apparent apparent activation energy during the oxygen-absorption stage and the comprehensive combustion index S characterize different portions of the oxidation–combustion process and therefore should not be expected to change to the same extent. The pronounced kinetic changes observed during the oxygen-absorption stage indicate that the principal effect of the treatment occurs during low-temperature oxidation, whereas the relatively small decreases in S suggest a comparatively limited additional effect on combustion intensity after ignition.

3.6.3. Kinetic Analysis of Coal Oxidation

Figure shows the kinetic fitting curves of different coal samples at different oxidation stages, including the moisture-loss stage, oxygen-absorption weight-gain stage, pyrolysis stage, and combustion stage. The stage-wise kinetic fittings yielded R² values of approximately 0.89–0.96, indicating acceptable linear consistency within the selected temperature intervals. However, these fitting coefficients do not uniquely validate a first-order reaction mechanism, and the calculated kinetic parameters are therefore interpreted as apparent activation energies for comparative purposes. The activation energies calculated from the slopes of the fitted lines showed that the treated coal samples generally had higher activation energies than raw coal at each stage.

11.

Kinetic fitting curves of different coal samples at different stages.

The activation energies of raw coal during the moisture-loss, oxygen-absorption weight-gain, pyrolysis, and combustion stages were 29.606, 68.083, 146.717, and 257.177 kJ/mol, respectively. After Y₁-D₂ treatment, the activation energies increased to 34.827, 131.802, 163.412, and 290.574 kJ/mol, corresponding to increases of 17.63%, 93.59%, 11.38%, and 12.99%, respectively. After Y₁-D₃ treatment, the activation energies increased to 33.580, 142.427, 167.361, and 294.407 kJ/mol, corresponding to increases of 13.42%, 109.20%, 14.07%, and 14.48%, respectively.

The increase in apparent activation energy was most pronounced during the oxygen-absorption weight-gain stage, indicating that the bacterial-loaded gel foam had a significant inhibitory effect on coal–oxygen complexation in the early stage of low-temperature oxidation. Y₁-D₃ showed higher activation energies than Y₁-D₂ during the oxygen-absorption, pyrolysis, and combustion stages, suggesting a more pronounced increase in the reaction energy barrier for subsequent oxidation and combustion. In contrast, Y₁-D₂ showed greater increases in characteristic temperatures such as T₂, T₃, T₅, and T₆, suggesting that the two bacterial communities exhibited different dominant inhibition stages when combined with the gel foam. Overall, both Y₁-D₂ and Y₁-D₃ increased the reaction energy barrier of lignite oxidation, thereby reducing its low-temperature oxidation activity and spontaneous combustion tendency. The apparent activation energy data are summarized in Table .

4. Apparent Activation Energies and Fitting Coefficients of Selected Coal Samples.

Coal sample	Parameters	Dehydration and weight loss phase	Oxygen-absorption weight-gain stage	Pyrolysis stage	Combustion stage
Raw coal	E(kJ/mol)	29.606	68.083	146.717	257.177
	R2	0.985	0.898	0.949	0.930
Coal-Y1-D2	E(kJ/mol)	34.827	131.802	163.412	290.574
	R2	0.947	0.925	0.957	0.909
Coal-Y1-D3	E(kJ/mol)	33.580	142.427	167.361	294.407
	R2	0.937	0.891	0.955	0.916

It should be noted that Coal-Y₁ was not included in the present Coats–Redfern comparison. Therefore, the kinetic analysis cannot quantitatively distinguish the contribution of the Y₁, gel matrix from that of bacterial loading, and the apparent activation energies of Coal-Y₁-D₂ and Coal-Y₁-D₃ represent the combined response of the composite treatments.

Because the treated samples were freeze-dried before TG-DTG testing, the results mainly reflect the comparative thermal-oxidation behavior of the treated solid-phase coal samples under standardized dry conditions and do not account for the evaporative-cooling effect of retained water or possible structural changes induced by wet–dry cycling. Further wet-state oxidation and wet–dry cycling experiments are therefore required to evaluate these effects under conditions closer to those in a goaf. Because the present TG-DTG experiments were conducted at a single heating rate of 10 °C/min, multi-heating-rate isoconversional analyses such as KAS and FWO could not be performed. Future studies using multiple heating rates are required to independently validate the present kinetic trends.

3.7. Coupled Inhibition Mechanism of Bacterial-Loaded Gel Foam

Considering the carrier properties, oxygen-consuming function, coal-surface characterization, and thermogravimetric results, the inhibition of lignite spontaneous combustion by the Y₁ bacterial-loaded gel foam is not caused by a single factor. Instead, it results from the coupled action of the gel-foam carrier and the oxygen-consuming bacterial communities. The coupled inhibition mechanism can be summarized in four aspects. ,

First, the gel foam has good flowability, spreading ability, and gelation stability. It can form a continuous coating layer on the coal surface and fill some pores and fissures, thereby reducing oxygen diffusion into the coal matrix and limiting oxygen contact with active sites. Second, the sodium alginate-based gel network has good water-retention and encapsulation capacities, providing a relatively moist and stable microenvironment for D₂ and D₃. This allows the bacterial communities to remain in the residual gel phase and maintain metabolic activity even after foam degradation or collapse. Third, the bacterial communities continuously consume O₂ around the coal mass, thereby weakening the coal–oxygen complexation reaction. Meanwhile, the accompanying increase in headspace CO₂ under sealed conditions may contribute to local inerting; however, the present measurements do not establish a quantitative O₂–CO₂ stoichiometric relationship. Finally, the combined effects of physical sealing, water retention and bacterial immobilization, sustained oxygen consumption, and CO₂ inerting increase the characteristic temperatures of the oxygen-absorption and combustion stages, raise the apparent apparent activation energy, and reduce the low-temperature oxidation activity of lignite.

Consequently, under the tested laboratory conditions, the Y₁ bacterial-loaded gel foam inhibits lignite oxidation through the combined actions of physical sealing, water retention and bacterial immobilization, sustained oxygen consumption, and thermal stabilization. These findings indicate that the gel foam serves not only as a fire-prevention material but also as a functional carrier for oxygen-consuming bacterial communities, providing a composite material strategy for further laboratory and engineering evaluation of residual lignite spontaneous combustion control. No techno-economic analysis was conducted in the present study; therefore, whether the additional performance associated with bacterial loading justifies its cultivation, preparation, transportation, and application costs remains to be evaluated before engineering scale-up.

4. Conclusions

This study developed a gel foam loaded with oxygen-consuming bacterial communities and evaluated its coupled inhibition of lignite spontaneous combustion in terms of carrier performance, bacterial compatibility, sustained oxygen consumption, and thermal-oxidation behavior. The main conclusions are as follows:

• (1)The D₂ and D₃ bacterial communities, obtained from settled sludge and abandoned-mine soil, respectively, showed strong oxygen-consuming capacity and rapidly reduced the O₂ volume fraction from approximately 20% to below 5% in a sealed system. Microbial community analysis revealed that D₂ was dominated by Citrobacter (40.89%) and Macellibacteroides (33.95%), while D₃ mainly consisted of Pseudomonas aeruginosa, Bacillus, and Bacteroidota.
• (2)The sodium alginate-based Y₁ gel foam exhibited good bacterial-loading compatibility and functional retention. After loading with D₂ and D₃, the foaming volume decreased by only 7.78% and 5.56%, respectively, with foam half-lives of 16.0 and 15.0 d. During 250 h of sealed incubation, the Y₁ bacterial-loaded systems continuously consumed O₂ and increased the CO₂ volume fraction to over 60%. Importantly, after foam collapse and air replenishment, the residual gel phase continued to support bacterial metabolism, decreasing O₂ from approximately 21% to 10%, demonstrating that Y₁ can retain bioactive bacteria beyond the lifetime of the foam structure.
• (3)The Y₁ bacterial-loaded gel foam reduced the spontaneous combustion tendency of lignite through coupled surface sealing and thermal-oxidation inhibition. SEM observations showed effective surface coating and pore/fissure filling, while the XRD profiles showed no major reconstruction of the overall diffraction pattern after treatment. TG-DTG and kinetic analyses revealed delayed oxidation behavior and increased apparent activation energy (by up to 109.20%) during the oxygen-absorption stage, indicating a significantly raised reaction barrier for low-temperature oxidation.

Overall, the gel foam loaded with oxygen-consuming bacterial communities provides a promising laboratory-scale strategy for coupled inhibition of residual-coal oxidation. Further dynamic-flow, long-duration, biosafety, and field-scale evaluations are required before its engineering applicability can be established.

Data will be made available upon request.

J.Y.: Investigation, Methodology, Writingoriginal draft. Y.W.: Data curation, Writing review & editing. Z.Y.: Conceptualization, Supervision. L.Z.: Investigation, Methodology. H.X.: Formal analysis, Visualization.

The authors declare no competing financial interest.