Pub Date : 2026-08-01Epub Date: 2026-02-04DOI: 10.1016/j.fuel.2026.138575
Xiaoyuan Jiang , Lin Lan , Shengqiang Yang , Buzhuang Zhou , Jiawen Cai
Differences in gas environments lead to distinct active group reactions and gas products during coal oxidation versus pyrolysis. However, as a fundamental underlying reaction, coal pyrolysis invariably coexists with the oxidation process, inevitably affects the precise analysis of the correlation between gas product release and active group oxidation reactions. To address this, the study first employed Electron Spin Resonance (ESR), Fourier Transform Infrared Spectroscopy (FTIR), and Gas Chromatography (GC) experiments to measure radical parameters, functional group contents, and CO/CO2 release amounts under oxidation and pyrolysis conditions, respectively. The differences in active groups under these two conditions were compared and their causes analyzed. Subsequently, correlation analysis was used to calculate the correlation degrees between gas products and active groups under oxidation, pyrolysis, and differential conditions (the difference between oxidation and pyrolysis). The results show that under oxidation conditions, the active groups most strongly correlated with CO and CO2 generation are ether bonds and carbonyl groups, respectively; under pyrolysis conditions, aliphatic hydrocarbons are the dominant groups for both CO and CO2 generation. This indicates that gas products under oxidation primarily originate from reactions of oxygen-containing functional groups, whereas pyrolysis gases mainly result from aliphatic chain cracking. Differential correlation analysis further highlights the strong correlations between CO generation and ether bonds, and CO2 generation and carbonyl groups under oxidation. In contrast, the strong correlations between CO/CO2 generation and aliphatic hydrocarbons under pyrolysis are weakened. Ultimately, the micro-generation pathways of coal spontaneous combustion gases revealed by this differential study provide the core theoretical foundation for precise early-warning and efficient inhibitor development.
{"title":"Differential correlation analysis of gas products and active groups reactions in coal oxidation and pyrolysis","authors":"Xiaoyuan Jiang , Lin Lan , Shengqiang Yang , Buzhuang Zhou , Jiawen Cai","doi":"10.1016/j.fuel.2026.138575","DOIUrl":"10.1016/j.fuel.2026.138575","url":null,"abstract":"<div><div>Differences in gas environments lead to distinct active group reactions and gas products during coal oxidation versus pyrolysis. However, as a fundamental underlying reaction, coal pyrolysis invariably coexists with the oxidation process, inevitably affects the precise analysis of the correlation between gas product release and active group oxidation reactions. To address this, the study first employed Electron Spin Resonance (ESR), Fourier Transform Infrared Spectroscopy (FTIR), and Gas Chromatography (GC) experiments to measure radical parameters, functional group contents, and CO/CO<sub>2</sub> release amounts under oxidation and pyrolysis conditions, respectively. The differences in active groups under these two conditions were compared and their causes analyzed. Subsequently, correlation analysis was used to calculate the correlation degrees between gas products and active groups under oxidation, pyrolysis, and differential conditions (the difference between oxidation and pyrolysis). The results show that under oxidation conditions, the active groups most strongly correlated with CO and CO<sub>2</sub> generation are ether bonds and carbonyl groups, respectively; under pyrolysis conditions, aliphatic hydrocarbons are the dominant groups for both CO and CO<sub>2</sub> generation. This indicates that gas products under oxidation primarily originate from reactions of oxygen-containing functional groups, whereas pyrolysis gases mainly result from aliphatic chain cracking. Differential correlation analysis further highlights the strong correlations between CO generation and ether bonds, and CO<sub>2</sub> generation and carbonyl groups under oxidation. In contrast, the strong correlations between CO/CO<sub>2</sub> generation and aliphatic hydrocarbons under pyrolysis are weakened. Ultimately, the micro-generation pathways of coal spontaneous combustion gases revealed by this differential study provide the core theoretical foundation for precise early-warning and efficient inhibitor development.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138575"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186134","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-06DOI: 10.1016/j.fuel.2026.138665
Taotao Cao , Jiacheng Zeng , Ye Chen , Anyang Pan , Hu Liu , Jing Liao , Gaofei Ning
<div><div>Adsorbed gas makes a great contribution to shale gas reserve and shale gas development, and therefore it is very important to clarify the dynamic evolution mechanism of adsorbed gas. However, the co-evolution modes of pore structure and methane adsorption capacity are still unclear. Accordingly, a series of experiments, including total organic carbon (TOC), XRD, low-pressure N<sub>2</sub> adsorption (N<sub>2</sub>GA), scanning electron microscopy (SEM), field-emission scanning electron microscopy (FE-SEM) and methane adsorption capacity, were performed on a series of artificially simulated samples from a low-mature Dalong shale in the Sichuan Basin. The results demonstrate that the evolution of shale pores particularly OM-related pores, coupled with a decrease in OM content and the alternation of clay minerals, is significantly controlled by thermal maturity. At mature and early high-mature stages, kerogen evolves into liquid hydrocarbons which subsequently fill shale pores, and with further increased thermal maturity, liquid hydrocarbons are cracked and OM pores reappear. As thermal maturity increases, OM pores are developed extensively in solid bitumen and pyrobitumen. Methane adsorption capacities of the simulated samples range from 2.29 to 3.80 cm<sup>3</sup>/g and from 1.85 to 3.18 cm<sup>3</sup>/g under dry and moisture-equilibrated conditions, respectively. Methane adsorption capacity is predominately contributed by OMs, and the contribution of clay minerals accounts only for 15.15-22.85% across the simulated samples. Clay mineral content exerts a positive influence on methane adsorption capacity at the mature stage, while it shifts to negligible effect at the overmature stage. The rise in methane adsorption capacity can be attributed to: the decomposition of kerogen, the progression of thermal maturity, the cracking of liquid hydrocarbon, and the expansion of micro- to mesopore networks. In terms of dynamic evolution of shale pores and methane adsorption behavior, four distinct stages of methane adsorption can be identified. At the early mature stage (eqv<em>R</em>o < 1.30%), methane is adsorbed on the surface of OMs and clay minerals. At the high mature stage (1.30%<eqv<em>R</em>o < 2.0%), methane is adsorbed in single-layer form within pores related to original OM, solid bitumen and clay minerals. At the over-mature stage (2.0%<eqv<em>R</em>o < 3.0%), methane exists in pores related to original OM in a single-layer form, and in pyrobitumen pores and clay mineral pores in a multi-layer form, and 4) at extremely over-mature stage (eqv<em>R</em>o > 3.0%), methane is predominantly adsorbed firstly in a multi-layer form in OM pores of pyrobitumen and secondly in pores related to original OM, and the adsorption capacity of clay minerals is enhanced due to the discharge of water. These findings revealed the mechanism of adsorbed gas in different thermal mature stages, and also provided guidance for shale gas evaluat
{"title":"Evolution of pore structure and methane adsorption in thermally matured shales: A case study of low-mature Dalong shale in Sichuan Basin, China","authors":"Taotao Cao , Jiacheng Zeng , Ye Chen , Anyang Pan , Hu Liu , Jing Liao , Gaofei Ning","doi":"10.1016/j.fuel.2026.138665","DOIUrl":"10.1016/j.fuel.2026.138665","url":null,"abstract":"<div><div>Adsorbed gas makes a great contribution to shale gas reserve and shale gas development, and therefore it is very important to clarify the dynamic evolution mechanism of adsorbed gas. However, the co-evolution modes of pore structure and methane adsorption capacity are still unclear. Accordingly, a series of experiments, including total organic carbon (TOC), XRD, low-pressure N<sub>2</sub> adsorption (N<sub>2</sub>GA), scanning electron microscopy (SEM), field-emission scanning electron microscopy (FE-SEM) and methane adsorption capacity, were performed on a series of artificially simulated samples from a low-mature Dalong shale in the Sichuan Basin. The results demonstrate that the evolution of shale pores particularly OM-related pores, coupled with a decrease in OM content and the alternation of clay minerals, is significantly controlled by thermal maturity. At mature and early high-mature stages, kerogen evolves into liquid hydrocarbons which subsequently fill shale pores, and with further increased thermal maturity, liquid hydrocarbons are cracked and OM pores reappear. As thermal maturity increases, OM pores are developed extensively in solid bitumen and pyrobitumen. Methane adsorption capacities of the simulated samples range from 2.29 to 3.80 cm<sup>3</sup>/g and from 1.85 to 3.18 cm<sup>3</sup>/g under dry and moisture-equilibrated conditions, respectively. Methane adsorption capacity is predominately contributed by OMs, and the contribution of clay minerals accounts only for 15.15-22.85% across the simulated samples. Clay mineral content exerts a positive influence on methane adsorption capacity at the mature stage, while it shifts to negligible effect at the overmature stage. The rise in methane adsorption capacity can be attributed to: the decomposition of kerogen, the progression of thermal maturity, the cracking of liquid hydrocarbon, and the expansion of micro- to mesopore networks. In terms of dynamic evolution of shale pores and methane adsorption behavior, four distinct stages of methane adsorption can be identified. At the early mature stage (eqv<em>R</em>o < 1.30%), methane is adsorbed on the surface of OMs and clay minerals. At the high mature stage (1.30%<eqv<em>R</em>o < 2.0%), methane is adsorbed in single-layer form within pores related to original OM, solid bitumen and clay minerals. At the over-mature stage (2.0%<eqv<em>R</em>o < 3.0%), methane exists in pores related to original OM in a single-layer form, and in pyrobitumen pores and clay mineral pores in a multi-layer form, and 4) at extremely over-mature stage (eqv<em>R</em>o > 3.0%), methane is predominantly adsorbed firstly in a multi-layer form in OM pores of pyrobitumen and secondly in pores related to original OM, and the adsorption capacity of clay minerals is enhanced due to the discharge of water. These findings revealed the mechanism of adsorbed gas in different thermal mature stages, and also provided guidance for shale gas evaluat","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138665"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186136","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-04DOI: 10.1016/j.fuel.2026.138627
Chaomin Ding , Ying Kan , Anni Liu , Xiaoyan Guo , Zhengdong Zhang , Junting Tong , Ke Li , Fan Liu , Xiaoping Song , Yong Yan , Qi Li , Tao Hong
Density is an important physical and chemical property of diesel, closely linked to its combustion performance in engines. Ensuring diesel quality in the refined oil market necessitates rapid and accurate density determination. While the density meter method, as a standard technique, is highly accurate, its complexity and lengthy analysis time limit its suitability for rapid on-site detection. Mid-infrared (MIR) and near-infrared spectroscopy (NIR) are commonly used in rapid oil testing. The spectral fusion of MIR and NIR simultaneously reflects fundamental, overtone, and combination frequency absorption of hydrogen-containing group vibrations, effectively addressing the limitations of single-spectrum methods in representing molecular structural information. Combined with chemometric models, this approach forms a novel method for rapid diesel density determination by addressing the limitations of conventional data integration strategies. PLS models were constructed using MIR and NIR spectra under four preprocessing conditions (no preprocessing, Savitzky-Golay first derivative, Savitzky-Golay second derivative, and vector normalization) for single spectra, and low-level, mid-level, high-level, and optimized high-level (Opt-HLDF) spectral data fusion. Under different spectral preprocessing conditions, the root mean squared error of prediction (RMSEP) for the best single-spectrum, low-level, and mid-level data fusion models were 1.10, 0.84, and 0.38 kg/m3, respectively, with the prediction coefficient of determination () of 0.98, 0.98, and 0.99, demonstrating the effectiveness of low- and mid-level data fusion in improving prediction performance. However, the high-level data fusion model performed poorly in external validation, with RMSEP and values of 1.75 kg/m3 and 0.95. To address this, an Opt-HLDF-PLS model was developed by integrating sub-models screened using a performance benchmark from optimal single-spectrum results. The Opt-HLDF-PLS model achieved the highest external validation performance, with RMSEP and values of 0.35 kg/m3 and 0.99, respectively, improving external validation accuracy by 7.89%–82.93% compared to other models and effectively mitigating the predictive instability frequently observed in conventional high-level fusion. F-tests and T-tests confirmed that the precision and accuracy of the optimized high-level data fusion method are highly comparable to the standard method. In conclusion, this study developed an optimized high-level spectral data fusion model based on NIR and MIR, enabling accurate and rapid diesel density determination.
{"title":"Rapid determination of diesel density using near-infrared and mid-infrared spectroscopy data fusion modelling","authors":"Chaomin Ding , Ying Kan , Anni Liu , Xiaoyan Guo , Zhengdong Zhang , Junting Tong , Ke Li , Fan Liu , Xiaoping Song , Yong Yan , Qi Li , Tao Hong","doi":"10.1016/j.fuel.2026.138627","DOIUrl":"10.1016/j.fuel.2026.138627","url":null,"abstract":"<div><div>Density is an important physical and chemical property of diesel, closely linked to its combustion performance in engines. Ensuring diesel quality in the refined oil market necessitates rapid and accurate density determination. While the density meter method, as a standard technique, is highly accurate, its complexity and lengthy analysis time limit its suitability for rapid on-site detection. Mid-infrared (MIR) and near-infrared spectroscopy (NIR) are commonly used in rapid oil testing. The spectral fusion of MIR and NIR simultaneously reflects fundamental, overtone, and combination frequency absorption of hydrogen-containing group vibrations, effectively addressing the limitations of single-spectrum methods in representing molecular structural information. Combined with chemometric models, this approach forms a novel method for rapid diesel density determination by addressing the limitations of conventional data integration strategies. PLS models were constructed using MIR and NIR spectra under four preprocessing conditions (no preprocessing, Savitzky-Golay first derivative, Savitzky-Golay second derivative, and vector normalization) for single spectra, and low-level, mid-level, high-level, and optimized high-level (Opt-HLDF) spectral data fusion. Under different spectral preprocessing conditions, the root mean squared error of prediction (RMSEP) for the best single-spectrum, low-level, and mid-level data fusion models were 1.10, 0.84, and 0.38 kg/m<sup>3</sup>, respectively, with the prediction coefficient of determination (<span><math><msubsup><mi>R</mi><mrow><mi>P</mi></mrow><mn>2</mn></msubsup></math></span>) of 0.98, 0.98, and 0.99, demonstrating the effectiveness of low- and mid-level data fusion in improving prediction performance. However, the high-level data fusion model performed poorly in external validation, with RMSEP and <span><math><msubsup><mi>R</mi><mrow><mi>P</mi></mrow><mn>2</mn></msubsup></math></span> values of 1.75 kg/m<sup>3</sup> and 0.95. To address this, an Opt-HLDF-PLS model was developed by integrating sub-models screened using a performance benchmark from optimal single-spectrum results. The Opt-HLDF-PLS model achieved the highest external validation performance, with RMSEP and <span><math><msubsup><mi>R</mi><mrow><mi>P</mi></mrow><mn>2</mn></msubsup></math></span> values of 0.35 kg/m<sup>3</sup> and 0.99, respectively, improving external validation accuracy by 7.89%–82.93% compared to other models and effectively mitigating the predictive instability frequently observed in conventional high-level fusion. F-tests and T-tests confirmed that the precision and accuracy of the optimized high-level data fusion method are highly comparable to the standard method. In conclusion, this study developed an optimized high-level spectral data fusion model based on NIR and MIR, enabling accurate and rapid diesel density determination.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138627"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186743","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-10DOI: 10.1016/j.fuel.2026.138681
Yunyun Liu , Heng Zhang , Rui Cao , Haojie Gao , Qiang Yu , Meng An
Predicting the specific capacitance of biomass-derived supercapacitors is challenging owing to the compositional variability of biomass and nonlinear interactions among synthesis parameters. Traditional methods struggle to accurately model these complex relationships. To address this limitation, this study employed six interpretable machine learning (ML) algorithms, including Gradient Boosting Decision Tree (GBDT), Random Forest (RF), Categorical Boosting (CATBoost), Gaussian Process Regression (GPR), Gradient Boosting Regression (GBR), and Extra Trees Regression (ETR), to predict capacitance based on 100 experimental datasets. After hyperparameter optimization, CATBoost achieved superior prediction accuracy with an R2 of 0.9558, whereas GPR and GBR showed higher overfitting tendencies. Shapley Additive exPlanations (SHAP) analysis indicated that specific surface area (SSA), total pore volume (TPV), and activation temperature had positive impacts on specific capacitance, whereas the ID/IG ratio and activation time exhibited negative effects. Validation using Partial Dependence Plots (PDPs) confirmed that the maximum specific capacitance reached 370F/g under optimal conditions: SSA exceeding 1250 m2/g, TPV above 1.1 cm3/g, activation temperature near 600 °C, and ID/IG within 0.8–0.85. These results demonstrate that the interpretable ML–based CATBoost model can effectively guide the synthesis of high–performance biomass–derived supercapacitors and provide a data–driven strategy for sustainable energy material design.
{"title":"Interpretable machine learning for optimizing the specific capacitance of biomass-derived supercapacitors","authors":"Yunyun Liu , Heng Zhang , Rui Cao , Haojie Gao , Qiang Yu , Meng An","doi":"10.1016/j.fuel.2026.138681","DOIUrl":"10.1016/j.fuel.2026.138681","url":null,"abstract":"<div><div>Predicting the specific capacitance of biomass-derived supercapacitors is challenging owing to the compositional variability of biomass and nonlinear interactions among synthesis parameters. Traditional methods struggle to accurately model these complex relationships. To address this limitation, this study employed six interpretable machine learning (ML) algorithms, including Gradient Boosting Decision Tree (GBDT), Random Forest (RF), Categorical Boosting (CATBoost), Gaussian Process Regression (GPR), Gradient Boosting Regression (GBR), and Extra Trees Regression (ETR), to predict capacitance based on 100 experimental datasets. After hyperparameter optimization, CATBoost achieved superior prediction accuracy with an R<sup>2</sup> of 0.9558, whereas GPR and GBR showed higher overfitting tendencies. Shapley Additive exPlanations (SHAP) analysis indicated that specific surface area (SSA), total pore volume (TPV), and activation temperature had positive impacts on specific capacitance, whereas the ID/IG ratio and activation time exhibited negative effects. Validation using Partial Dependence Plots (PDPs) confirmed that the maximum specific capacitance reached 370F/g under optimal conditions: SSA exceeding 1250 m<sup>2</sup>/g, TPV above 1.1 cm<sup>3</sup>/g, activation temperature near 600 °C, and ID/IG within 0.8–0.85. These results demonstrate that the interpretable ML–based CATBoost model can effectively guide the synthesis of high–performance biomass–derived supercapacitors and provide a data–driven strategy for sustainable energy material design.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138681"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-10DOI: 10.1016/j.fuel.2026.138690
Huitian Peng , Yuhang Li , Wen Nie , Lidian Guo , Fei Liu , Heng Liu , Mingyuan Lu , Jinkun Zhang , Yunshuo Shi , Meiyang Liao , Xingchao Cha
To address the challenges of low filtration efficiency and non-uniform velocity distribution in coal-mine dust-control fan cartridges, this study employed a porous media model to represent the filter medium and an orthogonal experimental design to systematically examine the effects of permeability α, pressure-jump coefficient C2, and cartridge thickness Δm on velocity distribution and filtration performance. An L9(34) orthogonal scheme was established, and numerical simulations were performed to analyze flow-field characteristics and dust-filtration behavior under different parameter combinations. The results show that when the permeability α is set to 3 × 10⁻11 m2, the pressure-jump coefficient C2 to 4 × 104 m⁻1, and cartridge thickness Δm to 0.0008 m, the standard deviation of the flow-field gradient decreases to 0.09, suggesting improved uniformity in the velocity distribution within the cartridge. Particle-transport simulations further demonstrate that this parameter combination effectively captures particles smaller than 10 μm, with dust predominantly depositing on the outer surface of the cartridge and at the bottom of the housing, thereby reducing the outlet dust concentration to approximately 9 mg/m3. Experimental validation confirms that, under the optimized conditions, the cartridge achieves a filtration efficiency exceeding 98% for dust particles in the size range of 0.2–10 μm. Overall, the optimization of key parameters results in a synergistic improvement in both flow-field uniformity and filtration performance, providing valuable insights for the advancement of mine dry dust-removal technology.
{"title":"Optimization of dust collector cartridge performance parameters using a porous media model","authors":"Huitian Peng , Yuhang Li , Wen Nie , Lidian Guo , Fei Liu , Heng Liu , Mingyuan Lu , Jinkun Zhang , Yunshuo Shi , Meiyang Liao , Xingchao Cha","doi":"10.1016/j.fuel.2026.138690","DOIUrl":"10.1016/j.fuel.2026.138690","url":null,"abstract":"<div><div>To address the challenges of low filtration efficiency and non-uniform velocity distribution in coal-mine dust-control fan cartridges, this study employed a porous media model to represent the filter medium and an orthogonal experimental design to systematically examine the effects of permeability <em>α</em>, pressure-jump coefficient <em>C<sub>2</sub></em>, and cartridge thickness <strong>Δ</strong><em>m</em> on velocity distribution and filtration performance. An L9(3<sup>4</sup>) orthogonal scheme was established, and numerical simulations were performed to analyze flow-field characteristics and dust-filtration behavior under different parameter combinations. The results show that when the permeability <em>α</em> is set to 3 × 10⁻<sup>11</sup> m<sup>2</sup>, the pressure-jump coefficient <em>C<sub>2</sub></em> to 4 × 10<sup>4</sup> m⁻<sup>1</sup>, and cartridge thickness <strong>Δ</strong><em>m</em> to 0.0008 m, the standard deviation of the flow-field gradient decreases to 0.09, suggesting improved uniformity in the velocity distribution within the cartridge. Particle-transport simulations further demonstrate that this parameter combination effectively captures particles smaller than 10 μm, with dust predominantly depositing on the outer surface of the cartridge and at the bottom of the housing, thereby reducing the outlet dust concentration to approximately 9 mg/m<sup>3</sup>. Experimental validation confirms that, under the optimized conditions, the cartridge achieves a filtration efficiency exceeding 98% for dust particles in the size range of 0.2–10 μm. Overall, the optimization of key parameters results in a synergistic improvement in both flow-field uniformity and filtration performance, providing valuable insights for the advancement of mine dry dust-removal technology.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138690"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-09DOI: 10.1016/j.fuel.2026.138616
Yong Tang , Yu Sun , Youwei He , Zhuxin Zhang , Jiazheng Qin
CO2-enriched hydrocarbon gas mixtures take advantage of the low miscibility pressure of CO2 and the compositional similarity of hydrocarbons, enabling enhanced oil recovery (EOR), reducing CO2 emissions, and mitigating carbon source limitation. However, the potential principles of their multi-component mass transfer in oil remain unclear. In this paper, the EOR performance and mechanisms of CO2-hydrocarbon gas co-injection in unconventional reservoirs are clarified by integrating phase behavior with compositional simulation. The physical properties of CO2-hydrocarbon gas mixtures and their phase behaviors with oil are studied in a fluid model. The impacts of different gases on oil–gas miscibility, gas front migration, and mixing mass transfer are analyzed using a slim-tube simulation study. The EOR mechanisms of CO2-hydrocarbon gas co-injection and its feasibility in reservoirs with carbon source limitation are explored based on compositional simulations. Results show that regulating the CO2 ratio in the gas mixtures improves both oil–gas miscibility and carbon utilization efficiency. The comparison between gas front migration and oil component distribution under miscible and immiscible status indicates that CO2-hydrocarbon gas miscible flooding enhances CO2 dissolution, promotes mass transfer, and delays gas breakthrough. Gas dissolution, oil expansion, component extraction, oil–gas miscibility and energy replenishment are essential mechanisms that boost the synergistic CO2-hydrocarbon gas-EOR process. Nearly 52.53% CO2 is stored availably, while the rest is produced after the end of 20-year EOR. This impure CO2 injection technology combines abundant associated gas with limited CO2 for oil production, which provides engineers with valuable insights into methane management and carbon reduction in unconventional reservoirs.
{"title":"Feasibility and mechanism of miscible oil recovery in unconventional reservoirs using CO2-enriched hydrocarbon gas mixtures: Insights from phase behavior and compositional simulation","authors":"Yong Tang , Yu Sun , Youwei He , Zhuxin Zhang , Jiazheng Qin","doi":"10.1016/j.fuel.2026.138616","DOIUrl":"10.1016/j.fuel.2026.138616","url":null,"abstract":"<div><div>CO<sub>2</sub>-enriched hydrocarbon gas mixtures take advantage of the low miscibility pressure of CO<sub>2</sub> and the compositional similarity of hydrocarbons, enabling enhanced oil recovery (EOR), reducing CO<sub>2</sub> emissions, and mitigating carbon source limitation. However, the potential principles of their multi-component mass transfer in oil remain unclear. In this paper, the EOR performance and mechanisms of CO<sub>2</sub>-hydrocarbon gas co-injection in unconventional reservoirs are clarified by integrating phase behavior with compositional simulation. The physical properties of CO<sub>2</sub>-hydrocarbon gas mixtures and their phase behaviors with oil are studied in a fluid model. The impacts of different gases on oil–gas miscibility, gas front migration, and mixing mass transfer are analyzed using a slim-tube simulation study. The EOR mechanisms of CO<sub>2</sub>-hydrocarbon gas co-injection and its feasibility in reservoirs with carbon source limitation are explored based on compositional simulations. Results show that regulating the CO<sub>2</sub> ratio in the gas mixtures improves both oil–gas miscibility and carbon utilization efficiency. The comparison between gas front migration and oil component distribution under miscible and immiscible status indicates that CO<sub>2</sub>-hydrocarbon gas miscible flooding enhances CO<sub>2</sub> dissolution, promotes mass transfer, and delays gas breakthrough. Gas dissolution, oil expansion, component extraction, oil–gas miscibility and energy replenishment are essential mechanisms that boost the synergistic CO<sub>2</sub>-hydrocarbon gas-EOR process. Nearly 52.53% CO<sub>2</sub> is stored availably, while the rest is produced after the end of 20-year EOR. This impure CO<sub>2</sub> injection technology combines abundant associated gas with limited CO<sub>2</sub> for oil production, which provides engineers with valuable insights into methane management and carbon reduction in unconventional reservoirs.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138616"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-09DOI: 10.1016/j.fuel.2026.138577
Jieyuan Yang , Yang Yang , Kesheng Liang , Xiangfei Geng , Bin Ding , Ming Gao , Shuangxiang Ye , Zhiming Wu , Dalong Sun
Gas channeling during CO2 geological storage reduces sweep efficiency and increases leakage risk, while conventional anti-channeling agents suffer from limited stability and retention. In this study, a novel CO2-responsive in-situ mineralization plugging (RIMP) system, composed of aqueous ammonia (NH4OH) and calcium chloride (CaCl2), simultaneously achieves CO2 in-situ mineralization and mitigates channeling. The optimized RIMP formulation, determined through compatibility and mineralization capacity tests, demonstrated a CO2 mineralization capacity of 6.13 mg CO2/g RIMP with 1.6 wt% CaCl2 and 1.0 wt% NH4OH. Real-time monitoring of pH, mass change, Ca2+ concentration, and in-situ infrared spectra revealed a four-stage mineralization process governed by seven elementary reactions. XPS, XRD, SEM, and FTIR analyses showed that the mineralization product was composed of 84.4% calcite and 15.6% vaterite. Microscopic visualization and numerical simulations of in-situ CO2 mineralization in capillaries showed that CaCO3 particles form at the gas–liquid interface, diffuse into the liquid phase due to concentration gradients, and adhere to the capillary wall when the adsorption force dominates, resulting in localized accumulation. A reaction–diffusion model, established based on Fick’s law and the Stokes-Einstein equation, showed that the distribution of mineralization products is mainly influenced by shut-in periods and pore size. Coreflooding tests demonstrated that the RIMP system increased plugging efficiency from 65.25% to 88.16% after 40 h of shut-in, and CO2 storage and mineralization efficiencies improved by 26.3% and 23.5%, respectively, consistent with the reaction–diffusion model predictions. The CO2-responsive in-situ mineralization characteristic of RIMP offers a promising dual-function solution that mitigates gas channeling and enhances CO2 storage security through permanent mineral trapping.
{"title":"CO2-responsive in-situ mineralization system for mitigating channeling and enhancing storage efficiency of CO2 during CCUS","authors":"Jieyuan Yang , Yang Yang , Kesheng Liang , Xiangfei Geng , Bin Ding , Ming Gao , Shuangxiang Ye , Zhiming Wu , Dalong Sun","doi":"10.1016/j.fuel.2026.138577","DOIUrl":"10.1016/j.fuel.2026.138577","url":null,"abstract":"<div><div>Gas channeling during CO<sub>2</sub> geological storage reduces sweep efficiency and increases leakage risk, while conventional anti-channeling agents suffer from limited stability and retention. In this study, a novel CO<sub>2</sub>-responsive in-situ mineralization plugging (RIMP) system, composed of aqueous ammonia (NH<sub>4</sub>OH) and calcium chloride (CaCl<sub>2</sub>), simultaneously achieves CO<sub>2</sub> in-situ mineralization and mitigates channeling. The optimized RIMP formulation, determined through compatibility and mineralization capacity tests, demonstrated a CO<sub>2</sub> mineralization capacity of 6.13 mg CO<sub>2</sub>/g RIMP with 1.6 wt% CaCl<sub>2</sub> and 1.0 wt% NH<sub>4</sub>OH. Real-time monitoring of pH, mass change, Ca<sup>2+</sup> concentration, and in-situ infrared spectra revealed a four-stage mineralization process governed by seven elementary reactions. XPS, XRD, SEM, and FTIR analyses showed that the mineralization product was composed of 84.4% calcite and 15.6% vaterite. Microscopic visualization and numerical simulations of in-situ CO<sub>2</sub> mineralization in capillaries showed that CaCO<sub>3</sub> particles form at the gas–liquid interface, diffuse into the liquid phase due to concentration gradients, and adhere to the capillary wall when the adsorption force dominates, resulting in localized accumulation. A reaction–diffusion model, established based on Fick’s law and the Stokes-Einstein equation, showed that the distribution of mineralization products is mainly influenced by shut-in periods and pore size. Coreflooding tests demonstrated that the RIMP system increased plugging efficiency from 65.25% to 88.16% after 40 h of shut-in, and CO<sub>2</sub> storage and mineralization efficiencies improved by 26.3% and 23.5%, respectively, consistent with the reaction–diffusion model predictions. The CO<sub>2</sub>-responsive in-situ mineralization characteristic of RIMP offers a promising dual-function solution that mitigates gas channeling and enhances CO<sub>2</sub> storage security through permanent mineral trapping.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138577"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146186852","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-03DOI: 10.1016/j.fuel.2026.138644
Dian Jiao , Gaige Liu , Zhisheng Liu , Tianzhen Feng , Xiangkun Li , Xiangjun Li
Polystyrene (PS) is a common microplastic can have toxic effects on the anaerobic digestion system of sludge by affecting the secretion of organic matter and volatile fatty acid, the activity of related enzymes, and the structure of microbial communities. Magnetite is a common conductive magnetic material can enhance anaerobic digestion of sludge. For the purpose to investigate the improvement of magnetite to an anaerobic digestion system inhibited by high quantity of PS microplastics, for anaerobic digestion of sludge were built and run with different conditions. The results demonstrated that high concentration microplastic stress of 300 particles/gTS initially inhibit VFAs synthesis, followed by further inhibition of protease and coenzyme F420 functions, ultimately leading to the accumulation of propionic acid/butyric acid and inhibiting anaerobic digestion for methane production, and an appropriate amount of magnetite (1 g/L) can effectively alleviate the methane production inhibition induced by microplastics (from 6.41 mlCH4/gVS∙d to 9.24 mlCH4/gVS∙d) through enhanced butyrate kinase activity and coupled with selective enrichment of Methanospirillum to optimize VFAs utilization. The microbial community reorganization supplemented by magnetite increased bacterial abundance and promoted the surface oxidation and fragmentation degree of PS. The findings reveal the addition of magnetite can alleviate the toxic effects of PS particles on the system and enhance its running performance.
{"title":"Improving anaerobic sludge digestion system combined with high quantity of polystyrene microplastics through adding magnetite","authors":"Dian Jiao , Gaige Liu , Zhisheng Liu , Tianzhen Feng , Xiangkun Li , Xiangjun Li","doi":"10.1016/j.fuel.2026.138644","DOIUrl":"10.1016/j.fuel.2026.138644","url":null,"abstract":"<div><div>Polystyrene (PS) is a common microplastic can have toxic effects on the anaerobic digestion system of sludge by affecting the secretion of organic matter and volatile fatty acid, the activity of related enzymes, and the structure of microbial communities. Magnetite is a common conductive magnetic material can enhance anaerobic digestion of sludge. For the purpose to investigate the improvement of magnetite to an anaerobic digestion system inhibited by high quantity of PS microplastics, for anaerobic digestion of sludge were built and run with different conditions. The results demonstrated that high concentration microplastic stress of 300 particles/gTS initially inhibit VFAs synthesis, followed by further inhibition of protease and coenzyme F420 functions, ultimately leading to the accumulation of propionic acid/butyric acid and inhibiting anaerobic digestion for methane production, and an appropriate amount of magnetite (1 g/L) can effectively alleviate the methane production inhibition induced by microplastics (from 6.41 mlCH<sub>4</sub>/gVS∙d to 9.24 mlCH<sub>4</sub>/gVS∙d) through enhanced butyrate kinase activity and coupled with selective enrichment of Methanospirillum to optimize VFAs utilization. The microbial community reorganization supplemented by magnetite increased bacterial abundance and promoted the surface oxidation and fragmentation degree of PS. The findings reveal the addition of magnetite can alleviate the toxic effects of PS particles on the system and enhance its running performance.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138644"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146102739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-03DOI: 10.1016/j.fuel.2026.138542
Haihong Zhang, Chongchong Wu, Huiling Zhang, Ning Wang, Yu Song, Longnian Han, Siran Zhang, Mengjun Su, Jian Liu, Zhanggui Hou
The tightening of global environmental regulations and the growing emphasis on carbon neutrality have driven the petroleum industry to pursue ultra-clean fuels. Diesel, as a dominant transportation fuel, faces particularly stringent sulfur limits due to the adverse environmental and operational impacts of sulfur oxides. Hydrodesulfurization (HDS) remains the most mature and efficient industrial technology for producing low-sulfur diesel; however, the escalating demands for ultra-deep desulfurization and low-carbon operation have placed unprecedented challenges on catalyst performance. Despite significant advances in catalyst formulation, systematic reviews dedicated specifically to HDS catalyst development remain scarce. This review provides a comprehensive overview of recent progress in HDS catalysis, emphasizing strategies for modulating active phases, supports, and novel unsupported catalysts. The fundamental mechanisms governing desulfurization pathways are briefly summarized to establish the theoretical foundation for subsequent discussions. Recent advances in tuning metal–support interactions, tailoring acidity, and employing heteroatom or promoter modifications to enhance catalytic activity are critically analyzed. Support regulation strategies—including compositional modification of γ-Al2O3 and zeolite-based systems—are examined with attention to structure–performance correlations and industrial applicability. The review further explores the emerging class of unsupported catalysts, highlighting template-assisted synthesis and three-dimensional framework designs as promising directions. Finally, key scientific challenges and future prospects are outlined, aiming to provide guidance for the rational design of next-generation HDS catalysts that enable sustainable production of ultra-clean diesel.
{"title":"Hydrodesulfurization catalysts for ultra-clean diesel: Recent progress, modulation strategies, and emerging alternatives","authors":"Haihong Zhang, Chongchong Wu, Huiling Zhang, Ning Wang, Yu Song, Longnian Han, Siran Zhang, Mengjun Su, Jian Liu, Zhanggui Hou","doi":"10.1016/j.fuel.2026.138542","DOIUrl":"10.1016/j.fuel.2026.138542","url":null,"abstract":"<div><div>The tightening of global environmental regulations and the growing emphasis on carbon neutrality have driven the petroleum industry to pursue ultra-clean fuels. Diesel, as a dominant transportation fuel, faces particularly stringent sulfur limits due to the adverse environmental and operational impacts of sulfur oxides. Hydrodesulfurization (HDS) remains the most mature and efficient industrial technology for producing low-sulfur diesel; however, the escalating demands for ultra-deep desulfurization and low-carbon operation have placed unprecedented challenges on catalyst performance. Despite significant advances in catalyst formulation, systematic reviews dedicated specifically to HDS catalyst development remain scarce. This review provides a comprehensive overview of recent progress in HDS catalysis, emphasizing strategies for modulating active phases, supports, and novel unsupported catalysts. The fundamental mechanisms governing desulfurization pathways are briefly summarized to establish the theoretical foundation for subsequent discussions. Recent advances in tuning metal–support interactions, tailoring acidity, and employing heteroatom or promoter modifications to enhance catalytic activity are critically analyzed. Support regulation strategies—including compositional modification of γ-Al<sub>2</sub>O<sub>3</sub> and zeolite-based systems—are examined with attention to structure–performance correlations and industrial applicability. The review further explores the emerging class of unsupported catalysts, highlighting template-assisted synthesis and three-dimensional framework designs as promising directions. Finally, key scientific challenges and future prospects are outlined, aiming to provide guidance for the rational design of next-generation HDS catalysts that enable sustainable production of ultra-clean diesel.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138542"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146102829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-08-01Epub Date: 2026-02-03DOI: 10.1016/j.fuel.2026.138586
Gen-Wang Ma , Xiao-Han Guo , Wei-Zhuo Gai , Yang Yang , Jie Zhang , Zheng-hui Pan , Zhen-Yan Deng
Aluminum (Al) is an active metal with a redox potential as low as of −1.66 V, but its surface passive oxide film prevents inner Al from contacting outside. In the past twenty years, Al surface modification by covering nanometer oxides was successfully used to activate Al so that it can rapidly generate hydrogen with water. However, the underlying mechanism is controversial so far. In this work, a model experiment was done by soaking and heat-treating Al sheet and powder so that a layer of porous nanometer γ-Al2O3 or Al(OH)3 covers on Al surfaces. It was found that there is a sharp wettability transition from hydrophobic on pristine Al surfaces to hydrophilic on modified Al. The contact angle of water droplet decreases from ∼ 90° to 30-40° and underwater hydrogen bubble contact angle increases from ∼ 125° to ∼ 180° after Al surface modification. Nanometer porous structures on modified Al surfaces push water towards inner Al such that the hydration process of Al surface passive oxide film is speeded up, and hydrogen bubble adhesion force decreases from > 100 μN on pristine Al surfaces to almost zero on modified Al, leading to a shorter induction time for the beginning of Al-water reaction to generate hydrogen. Cyclic voltammetry tests and electrochemical impedance spectroscopy showed that Al surface modification enhances the current density and reduces the charge transfer resistance, which are beneficial to Al reduction reaction. The present study provides a new mechanism and route for Al activation in hydrogen generation.
{"title":"Improvement of surface hydrophilicity by nanometer oxides to promote aluminum-water reaction to generate hydrogen","authors":"Gen-Wang Ma , Xiao-Han Guo , Wei-Zhuo Gai , Yang Yang , Jie Zhang , Zheng-hui Pan , Zhen-Yan Deng","doi":"10.1016/j.fuel.2026.138586","DOIUrl":"10.1016/j.fuel.2026.138586","url":null,"abstract":"<div><div>Aluminum (Al) is an active metal with a redox potential as low as of −1.66 V, but its surface passive oxide film prevents inner Al from contacting outside. In the past twenty years, Al surface modification by covering nanometer oxides was successfully used to activate Al so that it can rapidly generate hydrogen with water. However, the underlying mechanism is controversial so far. In this work, a model experiment was done by soaking and heat-treating Al sheet and powder so that a layer of porous nanometer γ-Al<sub>2</sub>O<sub>3</sub> or Al(OH)<sub>3</sub> covers on Al surfaces. It was found that there is a sharp wettability transition from hydrophobic on pristine Al surfaces to hydrophilic on modified Al. The contact angle of water droplet decreases from ∼ 90° to 30-40° and underwater hydrogen bubble contact angle increases from ∼ 125° to ∼ 180° after Al surface modification. Nanometer porous structures on modified Al surfaces push water towards inner Al such that the hydration process of Al surface passive oxide film is speeded up, and hydrogen bubble adhesion force decreases from > 100 μN on pristine Al surfaces to almost zero on modified Al, leading to a shorter induction time for the beginning of Al-water reaction to generate hydrogen. Cyclic voltammetry tests and electrochemical impedance spectroscopy showed that Al surface modification enhances the current density and reduces the charge transfer resistance, which are beneficial to Al reduction reaction. The present study provides a new mechanism and route for Al activation in hydrogen generation.</div></div>","PeriodicalId":325,"journal":{"name":"Fuel","volume":"417 ","pages":"Article 138586"},"PeriodicalIF":7.5,"publicationDate":"2026-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146102749","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}