Next Article in Journal
Life Cycle Sustainability Assessment of Agriproducts in Latin America: Overview Based on Latent Dirichlet Allocation
Previous Article in Journal
One-Step Synthesis of In Situ Sulfur-Doped Porous Carbons for Efficient CO2 Adsorption
Previous Article in Special Issue
Management of Tanning Waste from Leather Processing by Anaerobic Digestion Using a Dynamic Method on a Semi-Technical Scale
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 17
Issue 11
10.3390/su17114953
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Substrate Composition Effects on the Microbial Enhancement of Biogenic Methane Production from Coal

by
Liu Zhu
,
Wangjie Diao
*,
Chenyao Gong
,
Haihan Wang
,
Peilin Zhu
and
Yi Liu
School of Mines, China University of Mining & Technology, Xuzhou 221116, China
*
Author to whom correspondence should be addressed.
Sustainability 2025, 17(11), 4953; https://doi.org/10.3390/su17114953
Submission received: 11 April 2025 / Revised: 14 May 2025 / Accepted: 19 May 2025 / Published: 28 May 2025
(This article belongs to the Special Issue Sustainable Development of Biomass Energy under the Trend of Carbon Neutrality)
Browse Figures
Versions Notes

Abstract

:
The conversion of coal to biomethane is an environmentally friendly and sustainable method of coal utilization, and algae is a nutrient additive that enhances the economic sustainability of coal-to-biomethane production. The key regulatory factors and interaction mechanism of methane production were studied by carrying out anaerobic fermentation experiments on coal and microorganisms. Spearman correlation analysis, multiple linear regression, random forest and principal component analysis (PCA) were used to evaluate the effects of 14 coal-quality and microorganism composition parameters on methane production. The results showed that the hemicellulose content of microorganisms was significantly positively correlated with methane production, while total sugar and total fat significantly reduced the gas production. The protein content of microorganisms in a reasonable range could promote methane production. Among the coal-quality parameters, the C/H ratio (β = 0.43) and dry volatile matter (β = 0.17) had a weak positive contribution to methane production, while a high carbonization degree (C% > 80%; vitrinite reflectance > 1.2%) significantly inhibited the fermentation activity. The higher the maturity of the coal, the lower the methane production. The optimal methanogenic performance was concentrated in the combination of a low degree of coalification in coal (PC1 < −1.5) and high hemicellulose in microorganisms (PC2 > 1.8). In this study, a process optimization strategy was put forward, and the combination of low-rank coal with vitrinite reflectance < 0.5%, volatile matter > 35%, microorganisms with hemicellulose > 4.5%, and total sugar < 20% was optimized in an anaerobic fermentation experiment of coal and microorganisms. The results provide theoretical support for the directional control of anaerobic digestion of coal enhanced by microorganisms.

Graphical Abstract

1. Introduction

Coal, as the fossil fuel with the longest history of human use, provides power support for industrial civilization while facing development challenges due to its non-renewable nature and high pollution. Currently, the proven coal reserves worldwide can only meet approximately 150 years of demand, and a large amount of pollutants such as carbon dioxide and sulfur oxides are released during traditional combustion and utilization processes. With breakthroughs in frontier technologies such as synthetic biology and microbiomics, a new path for the green development of coal resources using biotransformation is emerging, opening up revolutionary prospects for the sustainable utilization of this energy source.
Recent studies have demonstrated that indigenous microbial communities or artificially cultured and directionally acclimated microbial consortia can progressively degrade coal into methane via biochemical processes [1,2]. This discovery suggests the feasibility of utilizing microbial conversion to transform coal into coalbed methane (CBM) for in situ extraction. Microbial coal-gasification technology involves the injection of nutrient solutions and microbial agents into coal seams, where anaerobic fermentation converts solid coal into extractable CBM [3,4,5,6]. This innovative approach not only enhances natural gas production but also contributes to CO2 emission reduction through optimized carbon conversion pathways [7,8,9,10]. Microbial in situ coal-gasification technology demonstrates substantial potential for application in unmined residual coal deposits and in situ coal seams, offering a sustainable alternative for unconventional coal resource development [11,12].
However, current microbial coal-gasification efficiency remains suboptimal for commercial-scale applications [13]. Current enhancement strategies focus on three primary approaches—coal pretreatment [14,15,16,17], microbial augmentation [18,19], and nutrient stimulation [20,21,22]—which can be individually implemented or in combination. Coal pretreatment involves physical comminution techniques to increase surface area accessibility or chemical modification methods to enhance coal solubility [23]. Microbial augmentation refers to the strategic introduction of exogenous microbial consortia to activate/enhance biogenic methane production. These inoculants are generally composed of syntrophic microbial consortia (bacteria and archaea) with laboratory-optimized community structures that feature high phylogenetic diversity [24] and comprehensive metabolic pathways for organic substrate utilization. Nutrient stimulation entails the injection of macronutrients (e.g., nitrogen and phosphorus) or micronutrients (e.g., vitamins and trace metals) into coal seams to activate indigenous methanogenic microbial communities [25,26,27].
In nutrient stimulation research, Penner et al. [28] demonstrated that tryptone supplementation as a methanogenic stimulant achieved a 55-fold enhancement in methane yield compared with coal-only controls. A comparative analysis by Barnhart et al. [29] systematically evaluated the effects of yeast extract and its constituent components (peptone, glutamate, and vitamins) on coal-to-methane bioconversion, and revealed that yeast extract was the most effective enhancement agent. Subsequent optimization studies by Zhang et al. [30] identified optimal concentration thresholds for yeast extract and tryptic peptone, and demonstrated the efficacy of complex nutrient formulations (tryptic soy broth and corn steep liquor) in boosting biomethane production. However, the prohibitively high cost of these composite nutrient formulations substantially limits their large-scale field implementation. In contrast, Barnhart et al. [29] investigated lipid-extracted algal biomass as an alternative methanogenic stimulant and revealed a comparable methane yield enhancement to yeast extract supplementation. This finding was corroborated by Davis et al. [31], who confirmed that algal biomass is a viable enhancer of in situ CBM production. Recent advances have suggested that phototrophic microalgae and cyanobacteria can be cultivated in surface reservoirs adjacent to CBM production sites [32], enabling dual-function systems. Furthermore, algal biomass demonstrates multifunctional potential, including lipid production for biofuel refinement and the synthesis of high-value chemicals as well as applications as biofertilizers and feedstock for aquaculture [33]. These coproduct streams create economic synergies that offset the operational costs associated with CBM enhancement technologies.
Algal biomass has significant potential as a nutritional supplement to enhance coal-derived biomethane production. However, the correlation between coal–microorganism physicochemical characteristics and methane yield remains unclear, particularly in terms of the effective coal-rank range that is amenable to algal-enhanced bioconversion. This study employed defined microbial consortia to conduct systematic batch anaerobic co-fermentation experiments using coal–cyanobacterial mixtures. Multivariate statistical analyses were performed to elucidate the critical coal–microorganism property correlations that govern methane production. The study findings establish a predictive framework to optimize algal-based stimulant selections while providing fundamental insights into accelerating the technological transition from laboratory-scale microbial CBM enhancement to field-scale implementation.

2. Materials and Methods

2.1. Coal Samples and Microorganisms

Coal samples were collected from actively operating underground Chinese mines, including the Baiyinhua Coal Mine (BYH), Xiezhuang Coal Mine (XZ), Madiliang Coal Mine (MDL), Zhulinshan Coal Mine (ZLS), and Lingxin Coal Mine (LZ). Fresh samples were obtained from the excavation working faces, immediately wrapped in an airtight plastic film to prevent oxidation, and transported to the laboratory under controlled conditions. The coal specimens exhibited varying maturity levels based on their geological origins, which were subsequently quantified using standardized petrographic and proximate analyses. The microorganism species were bought from biotech companies on Alibaba (https://www.1688.com/, accessed on 21 October 2024) for the co-fermentation experiments (Table 1). These microorganisms were cultivated in the laboratory of a biotech company. They are generally used to make food. To enhance the biochemical reactivity of the raw coal, microorganisms underwent mechanical pulverization. A cryogenic grinding protocol was implemented to mitigate oxidative degradation, and the coal samples were flash-frozen in liquid nitrogen prior to comminution using a high-precision pulverizer. The resultant coal powder was fractionated using an electric vibratory sieve shaker, with particles ≤ 200 mesh (74 μm) retained for experimental use.

2.2. Culture Media

An oxygen-stable mineral solution (pH 6.0) was formulated using the following components per liter: NaCl 0.8 g, NH4Cl 1.1 g, KH2PO4 0.1 g, KCl 0.1 g, MgCl2·6H2O 0.165 g, and CaCl2·2H2O 0.04 g.
The trace metal solution (per liter) contained 2 g nitrilotriacetic acid as a chelating agent, supplemented with MnCl2·H2O 0.88 g, FeCl2·6H2O 0.41 g, CoCl2·6H2O 0.2 g, ZnCl2 0.08 g, CuCl2·2H2O 0.02 g, NiCl2·6H2O 0.02 g, Na2MoO4·2H2O 0.02 g, Na2SeO3 0.02 g, and NaWO4·6H2O 0.02 g. The pH was adjusted to 6.0 ± 0.1 using a KOH solution added dropwise under continuous stirring.
The vitamin solution contained (per liter) pyridoxine hydrochloride 10 mg, thiamine hydrochloride 5 mg, riboflavin 5 mg, calcium pantothenate 5 mg, lipoic acid 5 mg, p-aminobenzoic acid 5 mg, nicotinic acid 5 mg, vitamin B12 5 mg, mercaptoethanesulfonic acid (MESA) 5 mg, biotin 2 mg, and folic acid 2 mg.

2.3. Experimental Procedure

Anaerobic fermentation was conducted in 500 mL serum bottles (working volume of 200 mL liquid + 300 mL headspace). Twenty-five experimental groups were established by combining five coal types with five algal species (10 g coal + 10 g microorganisms per bottle) in triplicate (n = 3). Each bottle contained 97 mL of the mineral solution and 1 mL of the trace metal solution. Bottles were loosely capped with aluminum foil and sterilized via an autoclave (126 °C; 15 min). Post-sterilization, 1 mL of a vitamin solution was aseptically injected through a 0.22 μm syringe filter at 56 °C. After cooling to mesophilic conditions (35 °C), 100 mL of a standardized inoculum (with a microbial composition as per the literature) [34] was anaerobically added.

2.4. Test Methods for Coal Samples and Microorganisms

A proximate analysis was performed using a 5E-MAG6700 (Changsha, China) fully automated industrial analyzer following ASTM D7582 standards [35]. An ultimate analysis (C, H, N, and S content) was conducted using an Elementar UNICUBE elemental analyzer (Elementar Analysensysteme GmbH, Langenselbold, Germany) equipped with a thermal conductivity detector. All measurements were performed in triplicate with strict quality controls, and the relative standard deviations (RSDs) were maintained below 2% using periodic calibration with NIST-certified reference materials (SRM 1632d bituminous coal [36]). Mean values were calculated after eliminating outliers via Grubbs’ test (α = 0.05).
The total sugar content (%) was determined using the phenol–sulfuric acid method, where samples were defatted with 80% ethanol, reacted with concentrated sulfuric acid to dehydrate polysaccharides into furan derivatives, and spectrophotometrically quantified at 490 nm using a glucose standard curve.
Total protein (%) was analyzed using the Kjeldahl method, where samples (0.5 g) were digested with concentrated sulfuric acid (420 °C; 4 h); distilled ammonia was absorbed in boric acid, titrated with 0.01 M HCl, and the protein content was calculated using a nitrogen-to-protein conversion factor of 6.25.
Total fat (%) was measured using Soxhlet extraction, where dried samples (105 °C) were refluxed with petroleum ether (boiling range: 30–60 °C) for 6 h and the fat content was gravimetrically determined.
Hemicellulose (%) and cellulose (%) were analyzed using the Van Soest detergent fiber analysis. Here, neutral detergent fiber (NDF) was obtained by boiling samples in a neutral detergent solution (with α-amylase) at 100 °C for 1 h; acid detergent fiber (ADF) was derived by further treating NDF residues with acidic detergent (containing sulfuric acid). The hemicellulose content was calculated as NDF% − ADF%, while cellulose% was determined by treating ADF residues with 72% sulfuric acid and subtracting the ash content after ignition.

2.5. Statistical Analysis Methods

In this study, four statistical analyses were employed to fully reveal the relationship between the substrate composition and biomethane production. First, monotonic correlations between variables were assessed using Spearman’s rank correlation analysis. This method processes non-normally distributed data by rank transformation and is particularly suitable for datasets with outliers or nonlinear but monotonous correlations. Second, a multiple linear regression analysis was constructed to explore the independent effects of the independent variables of the substrate composition and their joint explanatory power for biomethane production. The relative contributions of the substrate composition variables were compared using standardized regression coefficients (β-values). To capture possible nonlinear interaction effects between the substrate composition and biomethane yield, a random forest analysis was then implemented. This machine learning algorithm, based on decision tree integration, effectively reduces the risk of overfitting using bootstrap aggregation and random feature selection, and identifies key predictors using feature importance scores (Mean Decrease Gini). Finally, a principal component analysis was used for data dimensionality reduction and structure exploration. The original variables were transformed into uncorrelated principal components by orthogonal transformation, the effective components were retained according to Kaiser’s criterion (eigenvalue > 1), and the effect of dimensionality reduction was evaluated by combining the variance explained.

3. Results

3.1. Composition of Coal Samples and Microorganisms

The results of the proximate analysis for the coal samples are presented in Table 2, and their ultimate analysis data are summarized in Table 3. A comparative study of the elemental composition and industrial parameters of five samples (BYH, XZ, MDL, LX, and ZLS) revealed significant variations in the degree of carbonization and thermal evolution characteristics. In terms of the elemental composition, the carbon content (C%) demonstrated a progressive increase from 46.74% to 73.81%, with ZLS exhibiting the highest carbon enrichment (73.81%). Conversely, the hydrogen content (H%) displayed an inverse trend, ranging from 4.44% to 2.88%, where the lowest H% in ZLS (2.88%) suggested a reduced proportion of aliphatic structures. The sulfur content (S%) peaked in BYH (0.99%), whereas MDL and LX showed markedly lower sulfur levels (≤0.29%), reflecting distinct coal-forming environments or sedimentary conditions. The C/H ratio escalated from 11.89 (BYH) to 25.63 (ZLS), whereas the C/N ratio spanned 42.88–97.25.
The vitrinite reflectance (Ro%) results of the coal samples are listed in Table 4. According to the ISO 11760 classification standard [37], the Ro% is a key indicator of maturity. Samples BYH and XZ were classified as low-rank coal, whereas MDL and LX represented medium-rank coal. Notably, the ZLS sample attained a high-rank status, with a Ro% of 2.671%, significantly exceeding the values from other samples (0.25–0.762%) and confirming its advanced maturation stage. The proximate analysis further corroborated these trends: volatile matter on a dry basis (VM_d) sharply decreased with increasing maturity, whereas fixed carbon (FC_d) increased to 80.35%. This aligned with the fundamental principle of coalification, where volatile components dissipate and carbon frameworks become enriched.
Additionally, the algal nutritional properties exhibited substantial variations (Table 5). Dunaliella predominantly accumulates saccharides, Nannochloropsis and Aphanizomenon were protein-rich, whereas Porphyra demonstrated superior cellulose accumulation.

3.2. Methane Production

Figure 1 illustrates the methane production profiles from the co-fermentation of different coal samples with algal species. Figure 1f shows methane generation using algal mono-fermentation (without the addition of coal). The cumulative methane production volumes are summarized in Table 2. BYH coal coupled to Nannochloropsis exhibited the highest methane yield (26.43 mL), significantly surpassing the other combinations. In contrast, XZ coal paired with Dunaliella showed the lowest methane production (0.03 mL). Methane yields across all groups ranged from 0.01 mL (XZ + Dunaliella) to 26.43 mL (BYH + Nannochloropsis), with an extreme difference of 26.40 mL, demonstrating the critical role of combinatorial selections in methanogenic efficiency.
The BYH coal consistently outperformed the other coal types across all algal combinations (Table 6), achieving a mean methane yield of 13.18 mL (range: 0.03–26.43 mL). Conversely, ZLS coal displayed the lowest methane production (mean: 6.46 mL; range: 0.03–19.05 mL), primarily attributed to its high vitrinite reflectance (2.671%) and low volatile matter content (7.06%). Intermediate methane yields were observed for MDL and LX coals, with means of 10.12 mL and 11.99 mL, respectively, corresponding with their moderate vitrinite reflectance values (0.504% and 0.762%).
Nannochloropsis emerged as the most effective algal partner for coal-derived methane enhancement, generating a mean yield of 21.45 mL across all coal combinations. Its high protein content (37.4%) and moderate hemicellulose content (4.51%) likely facilitated methanogen activity. In contrast, Dunaliella demonstrated the lowest co-fermentation efficiency (mean: 0.02 mL; range: 0.01–0.03 mL), potentially due to its extremely low protein (0.4%) and bioavailable carbon sources (69.5% total sugars but only 0.04% cellulose), which could induce metabolic inhibition. Notably, Aphanizomenon exhibited application potential under specific conditions, achieving a 6.53 mL methane yield when combined with high-volatile coal (e.g., LX, with 29.26% volatile matter).

3.3. Statistical Analysis

3.3.1. Spearman Correlation Analysis

The Spearman correlation analysis revealed a monotonic correlation between the parameters and methanogenic volume in the co-fermentation system of coal and microorganisms. The results showed that the hemicellulose content in the algal components was significantly and positively correlated with the methanogenic volume (r = 0.803; p < 0.001), whereas the total sugar (r = −0.691; p < 0.001) and total fat (r = −0.401; p = 0.047) contents were significantly and negatively correlated (Figure 2). Hemicellulose in algal cell walls could be the main carbon source for methanogenesis. Hemicellulose mainly comprises easily degradable polysaccharides, such as xylan and mannan, and its hydrolysates (such as xylose and mannose) can be rapidly transformed into acetic acid and hydrogen by acidogenic bacteria, directly providing substrates for methanogens [38]. There was a significantly negative correlation between total sugar and methanogenic volume (r = −0.691; p = 0.0001), which was contrary to conventional understanding. Soluble sugars are generally considered to be good substrates for anaerobic fermentation. However, the total sugar content of microorganisms can also contain refractory components. A high sugar content can also inhibit methanogenic activity by inducing acidification [39]. In addition, the collinearity of total sugars and hemicellulose can interfere with a correlation analysis, which needs to be verified by a partial correlation analysis.
The vitrinite reflectance (r = −0.232; p = 0.265) and dry-basis volatile matter (r = 0.232; p = 0.265) in the coal-quality parameters showed a certain correlation but did not reach a significant level, indicating that the direct impact of coal maturity on the co-fermentation system was limited. In terms of the elemental analysis, the negative correlation between the C/H ratio (r = −0.255; p = 0.218) and C% (r = −0.255; p = 0.218) could have been related to the aromaticity of the coal molecular structure, as a highly aromatic structure can reduce bioavailability. Cellulose (r = 0.290; p = 0.159) and total protein (r = 0.224; p = 0.282) showed no significant effect, indicating that their decomposition could have been limited by the complexity of the lignin–cellulose complex structure or the nitrogen metabolism pathway. In summary, the Spearman correlation analysis effectively screened out the key variables; however, owing to its ability to recognize nonlinear relationships, it needed to be utilized in combination with other methods.

3.3.2. Multiple Linear Regression Analysis

The results of the multiple linear regression analysis showed that hemicellulose (β = 7.76) and cellulose (β = 3.96) had the most significantly positive contribution to methane production, and each 1% increase in hemicellulose content could increase methane production by 7.76 mL/GVS (Figure 3). This verified the core role of hemicellulose as a biodegradable matrix, and its β-value was significantly higher than that of the other parameters (p < 0.001). Total fat (β = −3.96) showed an inhibitory effect, which was consistent with the results of the Spearman correlation analysis, and could have been due to the toxic effect of long-chain fatty acids on methanogens. If the lipid content of microorganisms increased from 5% to 10%, the methane production was expected to decrease by 19.8 units. The weak positive coefficient of total sugar (β = 0.27) exhibited a negative correlation with the Spearman’s rank correlation coefficient, potentially due to the model’s inability to fully eliminate collinearity. For instance, hemicellulose, a subclass of total sugars, could have significantly contributed to this effect, thereby obscuring the true impact of the total sugar content.
The C/H ratio (β = 0.43) and volatile matter (β = 0.17) showed a weak positive effect on the coal-quality parameters. This contribution was weak and did not pass the significance test (p > 0.05). This could have been because the organic matter of coal undergoes multistep degradation (such as lignin depolymerization) to release small molecular substrates, whereas the aromatic structure of coal with high vitrinite reflectance (>0.5%) has strong resistance to degradation. The negative coefficients of fixed carbon (β = −0.04) and C% (β = −0.26) indicated that the bioavailability of highly carbonized coal was reduced. The model-adjusted R2 was 0.852, indicating that the selected variables explained 85.2% of the variation in methane production.

3.3.3. Random Forest Analysis

The random forest model revealed the nonlinear effects of the variables on methanogenesis based on the characteristic importance score (IS). The IS of hemicellulose was as high as 0.585, accounting for 58.5% of the total contribution, and was much higher than that of the other variables, confirming its dominant role in the complex fermentation system (Figure 4). This finding aligned with the results of the Spearman’s correlation and regression analyses.
The IS for total fat was only 0.029, which was significantly lower than the degree of contribution in the linear model. It could show “all or none” of the characteristics of the nonlinear relationship owing to its inhibitory effect. When the fat content was less than 3%, it did not significantly affect methanogenesis. However, when the fat content exceeded 5%, the inhibitory effect sharply increased. A random forest can capture such threshold effects; however, the importance score is sensitive to data distribution, and the specific impact mode must be visualized using partial dependence plots (PDPs).
The importance of the coal-quality parameters was less than 0.05. The importance of microorganism total protein (IS = 0.036) was higher than the coefficient (β = −0.16) in the linear model, which could have been due to the fact that protein degradation products promote microbial growth at low concentrations and inhibit methanogens at high concentrations. The proteins in microorganisms can enhance methane production by optimizing the carbon-to-nitrogen ratio in the reaction system. Such nonlinear relationships can only be identified using machine learning models.

3.3.4. PCA Analysis

The PCA results (Figure 5) showed that the cumulative interpretation rate of the first two principal components was 59.8% (PC1 = 40.5%; PC2 = 19.3%). The variance contribution of PC1 mainly originated from coal-quality parameters, in which the C/H ratio (loading = 0.416), vitrinite reflectance (loading = 0.416), and C% (loading = 0.358) showed significant positive loads, whereas volatile matter (loading = −0.415) and fixed carbon (loading = 0.404) on a dry basis showed negative loads. This showed that PC1 could be used to characterize the coal-maturity gradient. A high PC1 score reflects highly carbonized coal, whereas a low PC1 score corresponds with low-rank coal with high volatile matter and low fixed-carbon contents. The lower the maturity of the coal, the higher the methane production. The PC1 > 2 area had a high C/H ratio and high vitrinite reflectance but a low volatile matter content, which represents the characteristics of high-rank coal. Samples from this area were negatively correlated with methane production. The average volatile content of the samples in the PC1 < −1 area was 37.2%, and the C/H ratio was as low as 12.1. This area corresponded with high methane production.
The C/N ratio (loading = 0.591) and C% (loading = 0.294) showed a strong positive load, whereas N% (loading = −0.498) and S% (loading = −0.547) showed a significantly negative load. The positive score of PC2 reflected an algal substrate with high cellulose and a high C/N ratio, whereas the negative score was associated with high total sugar and sulfur contents. The sample score (Figure 5) showed that the samples with high methanogenic yields were concentrated on the negative axis of PC1 (low degree of coalification) and the positive axis of PC2 (high hemicellulose), and the optimal methanogenic area corresponded with the quadrants of PC1 < −1.5 and PC2 > 1.8.

4. Discussion

4.1. Critical Components Affecting Methane Production

The three modelling approaches—Spearman, regression, and random forest—collectively underscored the pivotal role of hemicellulose, although they exhibited variations in the quantification of its significance. The Spearman coefficient showed a strong monotonic relationship between hemicellulose and methanogenesis (r = 0.803), the regression model assigned the highest coefficient (β = 7.76), and the importance score of the random forest (0.585) highlighted its predictive power. This difference in the magnitude of the correlation stemmed from the methodological characteristics, with Spearman’s correlation reflecting monotonic associations but ignoring multivariate interactions, the regression model emphasizing linear additive and summative effects, and the random forest capturing nonlinear relationships and feature interactions. These three methods consistently confirmed that hemicellulose is a key driver of methanogenesis, possibly because of its ease of degradation, which provides sufficient substrates for acidic and methanogenic bacteria. In other anaerobic reaction systems, the co-fermentation of cellulose and hemicellulose promotes efficient and stable biogas production [40]. Two-stage anaerobic digestion of sugarcane straw hemicellulose hydrolysate has been shown to energetically favor hydrogen and methane production compared with single-stage digestion [41]. The inhibitory effect of total fat was significant in both the Spearman (r = −0.401) and regression models (β = −3.96) but was less significant in the random forest analysis (0.029), probably due to a threshold effect of inhibition, which was significant only in a specific concentration range. In other studies, the anaerobic digestion of fat-rich waste, such as fats, oils, and greases, has shown high potential for methane production but is challenged by long-chain fatty acid inhibition [42].

4.2. Components Weakly Correlated with Methane Production

The Spearman correlation analysis showed a significantly negative correlation between total sugar and methane production (r = −0.691; p = 0.0001). The results of the multiple linear regression analysis showed that total sugar was a weak positive contributor (0.27), which contradicted the Spearman analysis results. The random forest results showed that total sugars were moderately important (IS = 0.056) and did not clearly indicate inhibition or promotion. Algal polysaccharides contain hemicellulose fractions and total sugars could be covariates of hemicellulose, causing the regression model to misclassify their true effects. In other studies, the optimization of TS content and incorporation of sugar-rich additives increased methane production from various agricultural and industrial waste through anaerobic digestion [43]. Spearman’s correlation analysis revealed no significant correlation between cellulose and methane production (r = 0.290; p = 0.159). The results of the multiple linear regression analysis showed that cellulose had a strong positive contribution (β = 3.96) and worked synergistically with hemicellulose. The random forest results showed that cellulose was less important (IS = 0.033), probably because of degradation rate differences that were not emphasized by the nonlinear model. Algal polysaccharides contain hemicellulose fractions and total sugars could be covariates with hemicellulose, causing the regression model to misclassify their true effects. Cellulose must be converted to fermentable sugars by complex hydrolysis and its degradation efficiency is limited by the fermentation conditions, which leads to a weak, short-term correlation with methane production.
The mechanism underlying the effects of coal-quality parameters on the methane yield is complex. Mirror mass group reflectance was not significant in the univariate analysis (p = 0.265) and showed weak positive effects in the regression models (β = 0.06) and random forests (IS = 0.025). Coal with a low metamorphic degree was positively correlated with methane in the PCA loading analysis. Other findings suggest that the organic-matter structure of coal can determine its anaerobic degradation potential and that high-maturity coals are often difficult for microorganisms to directly utilize [44]. Low-maturity coal samples are suggested to be selected for biomethane production. In another study, algal amendments increased methane production and shortened the production time for different coal grades [45]. The positive enhancement of methane production by microorganisms could mask the effects of the coal-quality parameters. In addition, most previous studies have used pure coal fermentation, whereas the addition of microorganisms alters the microbial community composition. The contradictory performance of the C/H ratio (Spearman r = −0.255; regression β = 0.43) suggested that its relationship with methanogenicity was moderated by other coal-quality parameters in this paper. The PCA loading analysis showed that the C/H ratio was strongly correlated with PC1 (loading = 0.416), suggesting that the effect could be indirectly mediated through the main coal-quality component.

4.3. Limitations and Research Implications

Multiple linear regression models ignore interactions between variables, such as the synergistic degradation of hemicellulose and cellulose. Multiple linear regression models capture only linear effects and cannot explain thresholding or the saturation of variables. The importance score of the random forest model relies on the distribution of features, which could overestimate the contribution of hemicellulose if its content is concentrated, and the model is unable to quantify the interactions between variables. Nonetheless, random forests provide new perspectives for mechanism exploration in complex biological systems, and it is recommended that follow-up studies incorporate metabolomics to validate the critical pathways. The PCA failed to effectively isolate the interaction effects of coal and algal parameters (cumulative explanation rate of 59.8%), and it is recommended that follow-up studies employ a partial least squares discriminant analysis. In predicting the performance of a co-fermentation system, nonlinear models such as random forests are recommended to be used in conjunction with a PCA to identify the key component interaction patterns and avoid the limitations of a single method.
This study provides theoretical guidance for microorganism-enhanced biomethane production from coal. The established random forest model predicted the methane yield of diverse feedstock combinations. The results of this paper showed that low-rank coal with specular group reflectance <0.5% and volatile matter >35% was preferably combined with microorganisms comprising hemicellulose >4.5% and total sugars <20% for fermentation experiments. Higher-order coals with specular group reflectance >2.5%, volatile matter <10%, and high-sugar microorganisms with total sugar >60% are not recommended. Microorganisms can be broken down by pretreatment with hot gases for sugar–high-fat microorganisms. Future studies should incorporate macrogenomics to reveal the functional gene expression profiles for hemicellulose degradation and develop indicators to evaluate coal bioactivity based on the C/H ratio and volatile fraction.

5. Conclusions

In this study, the key regulatory factors of a coal and microorganism co-fermentation system were analyzed using various statistical methods. The main conclusions were as follows:
(1)
The highest methane-producing combination was that of BYH coal and Nannochloropsis, with a methane production of 26.43 mL, which was significantly higher than the other combinations tested. The lowest methane-producing combination was XZ coal and Dunaliella, with a methane production of only 0.03 mL. Coal performed the best among all algal combinations, and the microorganism that efficiently promoted methane production from coal was Nannochloropsis.
(2)
The hemicellulose content was a determinant of methane production, with each 1% increase in hemicellulose content increasing methane production by 7.76 mL/g coal. Total sugar and total fat decreased the efficiency of methane production.
(3)
The fermentation experiment preferred a combination of low-rank coal with a specular group reflectance <0.5%, volatile matter > content >35%, hemicellulose content >4.5%, and total sugar content <20%. Higher-order coal with a specular group reflectance >2.5%, volatile fraction <10%, and high-sugar microorganisms with total sugar > content >60% are not recommended.
(4)
The results of this study provide a basis for the application design of microorganism-enhanced methane production from coal; however, future studies need to be combined with microbial communities to improve the accuracy of the model predictions and further reveal the key regulators of the coal–microorganism co-fermentation system.

Author Contributions

Conceptualization, writing—original draft, and funding acquisition: L.Z.; conceptualization: W.D.; investigation: C.G.; methodology: H.W.; data curation, P.Z. and Y.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Yunlong Lake Laboratory of Deep Underground Science and Engineering Project, grant number 104024003; the National Natural Science Foundation of China, grant number 52404149; the Natural Science Foundation of Jiangsu Provincial Basic Research Program, grant number BK20220024; the Jiangsu Funding Program for Excellent Postdoctoral Talent, grant number 2024ZB177; and the Postdoctoral Fellowship Program of CPSF, grant number GZC20241927.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data is contained within the article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Bao, Y.; Ju, Y.; Huang, H.; Yun, J.; Guo, C. Potential and constraints of biogenic methane generation from coals and mudstones from huaibei coalfield, eastern China. Energy Fuels 2019, 33, 287–295. [Google Scholar] [CrossRef]
  2. Davis, K.J.; Gerlach, R. Transition of biogenic coal-to-methane conversion from the laboratory to the field: A review of important parameters and studies. Int. J. Coal Geol. 2018, 185, 33–43. [Google Scholar] [CrossRef]
  3. Su, X.; Zhao, W.; Xia, D.; Hou, S.; Fu, H.; Zhou, Y. Experimental study of advantages of coalbed gas bioengineering. J. Nat. Gas Sci. Eng. 2022, 102, 104585. [Google Scholar] [CrossRef]
  4. Xia, D.; Zhang, H.; Hong, J.; Su, X.; Liu, X. Variation in endogenous trace elements during methane generation from different coal ranks. Energy Fuels 2017, 31, 12168–12173. [Google Scholar] [CrossRef]
  5. Huang, Z.; Sednek, C.; Urynowicz, M.A.; Guo, H.; Wang, Q.; Fallgren, P.; Jin, S.; Jin, Y.; Igwe, U.; Li, S. Low carbon renewable natural gas production from coalbeds and implications for carbon capture and storage. Nat. Commun. 2017, 8, 568. [Google Scholar] [CrossRef]
  6. Bao, Y.; Hao, Y.; Guo, Z.; Hu, Y.; Li, D. Experimental research on biogenic methane generation pathway and microflora constraints based on in-situ gas-water-microbe. Fuel 2024, 377, 132828. [Google Scholar] [CrossRef]
  7. Ponnudurai, V.; Kumar, P.S.; Muthuvelu, K.S.; Velmurugan, S.; Subhani, S.; Arumugam, L.; Rajarathinam, R. Investigation on future perspectives of ex-situ biogenic methane generation from solid waste coal and coal washery rejects. Fuel 2022, 318, 123497. [Google Scholar] [CrossRef]
  8. Fallgren, P.H.; Jin, S.; Zeng, C.; Ren, Z.; Lu, A.; Colberg, P.J.S. Comparison of coal rank for enhanced biogenic natural gas production. Int. J. Coal Geol. 2013, 115, 92–96. [Google Scholar] [CrossRef]
  9. Colosimo, F.; Thomas, R.; Lloyd, J.R.; Taylor, K.G.; Boothman, C.; Smith, A.D.; Lord, R.; Kalin, R.M. Biogenic methane in shale gas and coal bed methane: A review of current knowledge and gaps. Int. J. Coal Geol. 2016, 165, 106–120. [Google Scholar] [CrossRef]
  10. Fuertez, J.; Boakye, R.; Mclennan, J.; Adams, D.J.; Sparks, T.D.; Gottschalk, A. Developing methanogenic microbial consortia from diverse coal sources and environments. J. Nat. Gas Sci. Eng. 2017, 46, 637–650. [Google Scholar] [CrossRef]
  11. Welte, C.U. A microbial route from coal to gas. Science 2016, 354, 184. [Google Scholar] [CrossRef] [PubMed]
  12. Aramaki, N.; Tamamura, S.; Ueno, A.; Badrul, A.A.K.M.; Murakami, T.; Tamazawa, S.; Yamaguchi, S.; Aoyama, H.; Kaneko, K. Experimental investigation on the feasibility of industrial methane production in the subsurface environment via microbial activities in northern Hokkaido, Japan—A process involving subsurface cultivation and gasification. Energy Conv. Manag. 2017, 153, 566–575. [Google Scholar] [CrossRef]
  13. Zhu, L.; Yao, Q.; Diao, W.; Huang, Z.; Li, X. Bioelectrochemically enhanced biomethane production from low-rank coal using multiple microbial strains. J. Clean. Prod. 2024, 442, 141028. [Google Scholar] [CrossRef]
  14. Feng, X.; Zhang, Z.; Guo, H.; Li, Z.; Huang, Z.; Urynowicz, M. Enhancement of biogenic methane production from coal using supercritical CO2. J. Supercrit. Fluids 2023, 201, 106023. [Google Scholar] [CrossRef]
  15. Huang, Z.; Urynowicz, M.A.; Colberg, P.J.S. Stimulation of biogenic methane generation in coal samples following chemical treatment with potassium permanganate. Fuel 2013, 111, 813–819. [Google Scholar] [CrossRef]
  16. Webster, J.; Lee, M.; Gurba, L.W.; Manefield, M.; Thomas, T. The effect of oxidative treatment on soluble compounds from australian coal. Fuel 2019, 257, 116071. [Google Scholar] [CrossRef]
  17. Haq, S.R.; Tamamura, S.; Ueno, A.; Tamazawa, S.; Aramaki, N.; Murakami, T.; Alam, A.K.M.B.; Igarashi, T.; Kaneko, K. Biogenic methane generation using solutions from column reactions of lignite with hydrogen peroxide. Int. J. Coal Geol. 2018, 197, 66–73. [Google Scholar] [CrossRef]
  18. Piao, D.; Song, Y.; Kim, D. Bioelectrochemical enhancement of biogenic methane conversion of coal. Energies 2018, 11, 2577. [Google Scholar] [CrossRef]
  19. Gao, L.; Feng, X.; Zhang, Y.; Guo, H.; Mu, X.; Huang, Z.; Urynowicz, M. Methane production from the biodegradation of lignite with different sizes by mixed fungi-methanogen microflora. FEMS Microbiol. Lett. 2024, 371, fnae037. [Google Scholar] [CrossRef]
  20. Li, Z.; Cheng, J.; Duan, K.; Cheng, Y.; Guo, H.; Huang, Z.; Urynowicz, M.; Ali, M.I. The succession of microorganisms and organics in the process of methane generation by co-degradation of anthracite and rice straw. Int. J. Energy Res. 2022, 46, 15116–15126. [Google Scholar] [CrossRef]
  21. Xiao, D.; Peng, S.; Wang, E. Fermentation enhancement of methanogenic archaea consortia from an illinois basin coalbed via dol emulsion nutrition. PLoS ONE 2015, 10, e124386. [Google Scholar] [CrossRef] [PubMed]
  22. Khan, S.; Deng, Z.; Wang, B.; Yu, Z. Coal-straw co-digestion-induced biogenic methane production: Perspectives on microbial communities and associated metabolic pathways. Sci. Rep. 2024, 14, 26554. [Google Scholar] [CrossRef] [PubMed]
  23. Zhang, W.; Huang, Z.; Zhang, D.; Medunic, G.; Urynowicz, M.; Liu, F.; Guo, H.; Haider, R.; Ali, M.I.; Jamal, A.; et al. Enhancement of biomethane production by huminite-enriched lignite pretreated with hydrogen peroxide. Int. J. Coal Geol. 2023, 274, 104284. [Google Scholar] [CrossRef]
  24. Yang, L.; Zhang, Y.; Hao, Z.; Ding, D.; Liu, Z. Clean utilization of lignite to produce biomethane by optimizing the microbial community. Energy 2023, 262, 125533. [Google Scholar] [CrossRef]
  25. Fallgren, P.H.; Zeng, C.; Ren, Z.; Lu, A.; Ren, S.; Jin, S. Feasibility of microbial production of new natural gas from non-gas-producing lignite. Int. J. Coal Geol. 2013, 115, 79–84. [Google Scholar] [CrossRef]
  26. Mcleish, A.G.; Vick, S.H.W.; Grigore, M.; Pinetown, K.L.; Midgley, D.J.; Paulsen, I.T. Adherent microbes in coal seam environments prefer mineral-rich and crack-associated microhabitats. Int. J. Coal Geol. 2021, 234, 103652. [Google Scholar] [CrossRef]
  27. Li, B.; Guo, H.; Chen, Z.; Xu, Q.; Xia, D.; Lv, J.; Yu, H. Metabolism mechanisms of biogenic methane production by synergistic biodegradation of lignite and guar gum. Sci. Total Environ. 2024, 946, 174085. [Google Scholar] [CrossRef]
  28. Penner, T.J.; Foght, J.M.; Budwill, K. Microbial diversity of western canadian subsurface coal beds and methanogenic coal enrichment cultures. Int. J. Coal Geol. 2010, 82, 81–93. [Google Scholar] [CrossRef]
  29. Barnhart, E.P.; Davis, K.J.; Varonka, M.; Orem, W.; Cunningham, A.B.; Ramsay, B.D.; Fields, M.W. Enhanced coal-dependent methanogenesis coupled with algal biofuels: Potential water recycle and carbon capture. Int. J. Coal Geol. 2017, 171, 69–75. [Google Scholar] [CrossRef]
  30. Zhang, J.; Park, S.Y.; Liang, Y.; Harpalani, S. Finding cost-effective nutrient solutions and evaluating environmental conditions for biogasifying bituminous coal to methane ex situ. Appl. Energy 2016, 165, 559–568. [Google Scholar] [CrossRef]
  31. Davis, K.J.; Lu, S.P.; Barnhart, E.P.; Parker, A.E.; Fields, M.W.; Gerlach, R. Type and amount of organic amendments affect enhanced biogenic methane production from coal and microbial community structure. Fuel 2018, 211, 600–608. [Google Scholar] [CrossRef]
  32. Smith, H.J.; Schweitzer, H.D.; Barnhart, E.P.; Orem, W.; Gerlach, R.; Fields, M.W. Effect of an algal amendment on the microbial conversion of coal to methane at different sulfate concentrations from the powder river basin, USA. Int. J. Coal Geol. 2021, 248, 103860. [Google Scholar] [CrossRef]
  33. Priyadarshani, I.; Rath, B. Commercial and industrial applications of micro algae—A review. J. Algal Biomass Util. 2012, 3, 89–100. [Google Scholar]
  34. Zhu, L.; Yao, Q.; Huang, Z.; Li, X.; Ma, Z. Biogenic methane generation from lignite coal at different temperatures. Gas Sci. Eng. 2023, 117, 205016. [Google Scholar] [CrossRef]
  35. ASTM D7582; Standard Test Methods for Proximate Analysis of Coal and Coke by Macro Thermogravimetric Analysis. ASTM: West Conshohocken, PA, USA, 2015.
  36. SRM 1632d; Trace Elements in Coal (Bituminous). NIST: Gaithersburg, MD, USA, 2014.
  37. ISO 11760; Classification of Coals. ISO: Geneva, Switzerland, 2018.
  38. Liu, Z.; Ma, C.Q.; Gao, C.; Xu, P. Efficient utilization of hemicellulose hydrolysate for propionic acid production using propionibacterium acidipropionici. Bioresour. Technol. 2012, 114, 711–714. [Google Scholar] [CrossRef]
  39. Xie, D.D.; Sun, Y.Q.; Lei, Y.A. Effect of glucose levels on carbon flow rate, antioxidant status, and enzyme activity of yeast during fermentation. J. Sci. Food. Agric. 2022, 102, 5333–5347. [Google Scholar] [CrossRef]
  40. Ma, S.S.; Wang, H.L.; Li, J.X.; Fu, Y.; Zhu, W.B. Methane production performances of different compositions in lignocellulosic biomass through anaerobic digestion. Energy 2019, 189, 116190. [Google Scholar] [CrossRef]
  41. Tomasini, M.; Faber, M.D.; Ferreira-Leitao, V.S. Sequential production of hydrogen and methane using hemicellulose hydrolysate from diluted acid pretreatment of sugarcane straw. Int. J. Hydrogen Energy 2023, 48, 9971–9987. [Google Scholar] [CrossRef]
  42. Hao, J.H.; de Los Reyes, F.L.; He, X. Fat; oil, and grease (fog) deposits yield higher methane than fog in anaerobic co-digestion with waste activated sludge. J. Environ. Manag. 2020, 268, 110708. [Google Scholar] [CrossRef]
  43. Aboudi, K.; Alvarez-Gallego, C.J.; Romero-García, L.I. Influence of total solids concentration on the anaerobic co-digestion of sugar beet by-products and livestock manures. Sci. Total Environ. 2017, 586, 438–445. [Google Scholar] [CrossRef]
  44. Strapoc, D.; Mastalerz, M.; Dawson, K.; Macalady, J.; Callaghan, A.V.; Wawrik, B.; Turich, C.; Ashby, M. Biogeochemistry of microbial coal-bed methane. Annu. Rev. Earth Planet. Sci. 2011, 39, 617–656. [Google Scholar] [CrossRef]
  45. Platt, G.A.; Davis, K.J.; Schweitzer, H.D.; Smith, H.J.; Fields, M.W.; Barnhart, E.P.; Gerlach, R. Algal amendment enhances biogenic methane production from coals of different thermal maturity. Front. Microbiol. 2023, 14, 1097500. [Google Scholar] [CrossRef] [PubMed]
Sustainability 17 04953 g001
Figure 1. Characterization of methane production. (a) BYH and microorganisms; (b) LX and microorganisms; (c) XZ and microorganisms; (d) MDL and microorganisms; (e) ZLS and microorganisms; (f) Only microorganisms.
Figure 1. Characterization of methane production. (a) BYH and microorganisms; (b) LX and microorganisms; (c) XZ and microorganisms; (d) MDL and microorganisms; (e) ZLS and microorganisms; (f) Only microorganisms.
Sustainability 17 04953 g001
Sustainability 17 04953 g002
Figure 2. Spearman correlation coefficient.
Figure 2. Spearman correlation coefficient.
Sustainability 17 04953 g002
Sustainability 17 04953 g003
Figure 3. Linear regression coefficients.
Figure 3. Linear regression coefficients.
Sustainability 17 04953 g003
Sustainability 17 04953 g004
Figure 4. Random forest feature importance.
Figure 4. Random forest feature importance.
Sustainability 17 04953 g004
Sustainability 17 04953 g005
Figure 5. PCA analysis.
Figure 5. PCA analysis.
Sustainability 17 04953 g005
Table 1. Taxonomic classification of samples.
Table 1. Taxonomic classification of samples.
DomainPhylumClassOrderFamilyGenusMicroorganism Sample for Experiment
PlantaeChlorophytaChlorophyceaeChlamydomonadalesDunaliellaceaeDunaliellaDunaliella
ChromistaHeterokontophytaEustigmatophyceaeEustigmatalesMonodopsidaceaeNannochloropsisNannochloropsis
BacteriaCyanobacteriaCyanophyceaeNostocalesAphanizomenonaceaeAphanizomenonAphanizomenon
PlantaeRhodophytaBangiophyceaeBangialesBangiaceaePorphyraPorphyra
EukaryoteBigyraLabyrinthuleaThraustochytridaThraustochytriidaeSchizochytriumSchizochytrium
Table 2. Proximate analysis of coal samples.
Table 2. Proximate analysis of coal samples.
SampleMoisture, % (ar)Ash, % (d)Volatile Matter, % (d)Fixed Carbon, % (d)Dry Basis, % (d)Volatile Matter, % (ad)Volatile Matter, % (daf)Fixed Carbon, % (ad)Fixed Carbon, % (daf)
BYH16.439.5935.5938.3911.4742.5948.1145.9451.89
XZ7.414.5030.1457.954.8632.5534.2262.5965.78
MDL5.2021.6529.5343.6222.8431.1540.3746.0159.63
LX11.909.8729.2651.7611.2030.0533.8458.7566.16
ZLS2.679.927.0680.3510.197.258.0882.5691.92
Table 3. Ultimate analysis of coal samples.
Table 3. Ultimate analysis of coal samples.
SampleC, %N, %H, %S, %C/HC/N
BYH46.741.093.930.9911.8942.88
XZ62.590.844.110.6715.2374.51
MDL60.420.934.440.2913.6164.97
LX65.160.673.670.2117.7497.25
ZLS73.811.262.880.6125.6358.58
Table 4. Vitrinite reflectance test of coal samples.
Table 4. Vitrinite reflectance test of coal samples.
SampleAverage Reflectivity, %Reflectivity, %
BYH0.2500.25 to <0.30
XZ0.4590.45 to <0.50
MDL0.5040.50 to <0.55
LX0.7620.75 to <0.80
ZLS2.6712.65 to <2.70
Table 5. Nutrients of microorganism samples.
Table 5. Nutrients of microorganism samples.
MicroorganismTotal Sugar, %Total Protein, %Total Fat, %Hemicellulose, %Cellulose, %
Dunaliella69.50.440.080.04
Nannochloropsis1237.43.24.510.18
Aphanizomenon17.250.52.860.084.85
Porphyra45.321.82.240.675.03
Schizochytrium16.117.22.861.523.43
Table 6. Accumulated biomethane.
Table 6. Accumulated biomethane.
SampleDunaliella/mLNannochloropsis/mLAphanizomenon/mLPorphyra/mLSchizochytrium/mL
BYH0.0326.430.3323.4315.7
XZ0.0114.440.1819.9817.05
MDL0.0312.160.6523.4914.29
LX0.0325.086.539.059.27
ZLS0.026.080.2921.284.62
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Zhu, L.; Diao, W.; Gong, C.; Wang, H.; Zhu, P.; Liu, Y. Substrate Composition Effects on the Microbial Enhancement of Biogenic Methane Production from Coal. Sustainability 2025, 17, 4953. https://doi.org/10.3390/su17114953

AMA Style

Zhu L, Diao W, Gong C, Wang H, Zhu P, Liu Y. Substrate Composition Effects on the Microbial Enhancement of Biogenic Methane Production from Coal. Sustainability. 2025; 17(11):4953. https://doi.org/10.3390/su17114953

Chicago/Turabian Style

Zhu, Liu, Wangjie Diao, Chenyao Gong, Haihan Wang, Peilin Zhu, and Yi Liu. 2025. "Substrate Composition Effects on the Microbial Enhancement of Biogenic Methane Production from Coal" Sustainability 17, no. 11: 4953. https://doi.org/10.3390/su17114953

APA Style

Zhu, L., Diao, W., Gong, C., Wang, H., Zhu, P., & Liu, Y. (2025). Substrate Composition Effects on the Microbial Enhancement of Biogenic Methane Production from Coal. Sustainability, 17(11), 4953. https://doi.org/10.3390/su17114953

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop