Energy & Fuels最新文献

Biomass chemical looping gasification (BCLG) is an advanced technology that enables the efficient transformation of biomass into hydrogen-rich syngas. However, its practical implementation requires multiple complex steps, including oxygen carrier screening, reactor configuration, and process parameter optimization, which are often time-intensive and costly. This study proposes a machine learning framework to predict BCLG performance, providing a data-driven alternative to traditional empirical methods. A database comprising 525 samples was constructed, and among six machine learning models evaluated, the Extra Tree model exhibited the highest predictive accuracy, achieving an R² value of 0.86 after hyperparameter optimization. Employing the SHAP algorithm, we identified key factors influencing gasification performance, including the concentrations of Ni, Al, and Fe in oxygen carriers and the C and H contents of biomass, which were shown to significantly impact syngas yield. The optimized Extra Tree model identified ideal conditions with a steam-to-biomass ratio of 0.8 and a gasification temperature of 860 °C, accurately predicting syngas yields for four new-input oxygen carriers, with the best-performing carrier maintaining a yield of 79% after 10 cycles. These findings underscore the potential of machine learning methods to replace trial-and-error approaches, offering an efficient pathway for accelerating the development and optimization of BCLG systems.

The axial mixing/segregation behavior of single plastic particles in a bubbling fluidized bed reactor has been investigated by noninvasive X-ray imaging techniques in the temperature range of 500–650 °C and under pyrolysis conditions. Experimental results showed that the extent of mixing between the plastic particle and the fluidized bed increases as both the temperature and fluidization velocity increase. Three modeling approaches were proposed to describe the axial mixing/segregation behavior of the plastic particle, i.e., a purely mechanistic model, a physics-informed neural network (PINN), and an augmented PINN (augPINN). The former model is based on the second law of motion. The second model is a standard PINN, built by simply embedding the second law of motion in the loss function. The third approach involves the introduction of a new interphase distribution parameter, P, into the model. This parameter represents the relative importance of the effects of the emulsion and bubble phases on the plastic particle. This parameter was obtained by training the neural network using the X-ray axial displacement data. The augPINN has been shown to outperform both the mechanistic and the standard PINN models in describing the axial mixing/segregation of polypropylene particles. Moreover, the obtained parameter P was found to be physically interpretable. The main novelty of this work is to show how different frameworks based on the concept of physics-informed machine learning can be successfully applied to complex and real-world hydrodynamic data sets.

A Ni–Ru-based electrocatalyst combined with polyoxovanadate (POV) was found to show superior activity in the methanol oxidation reaction (MOR), suppressing the oxygen evolution reaction (OER). The role of POV was evaluated by comparing the activity with two other catalysts without POV. The NiO-Ru/RuO₂ catalyst decorated with POV exhibited both OER and MOR activity in an alkaline medium with an onset potential of 1.531 ± 0.002 vs RHE and 1.370 ± 0.001 V vs RHE (at 50 mV s^–1), respectively. The other two synthesized Ni–Ru-based catalysts, although they showed OER and MOR activity, were found to be inferior in the MOR in terms of the onset potential and current density compared to the similar catalyst modified with POV. However, it was observed that the Ni(OH)₂-RuO₂ exhibited better OER activity without POV. The low Tafel slope value of 112 mV dec^–1 in the MOR for NiO-Ru/RuO₂@POV suggested higher electrocatalytic activity with a low kinetic barrier. Under the electrocatalytic conditions, the activity of NiO-Ru/RuO₂@POV remained stable for 20,000 s. The linear correlation between the current density and the square root of the scan rate signified a diffusion-controlled MOR process. The mechanism of the electrocatalytic oxidation of methanol (CH₃OH) with NiO-Ru/RuO₂@POV was studied through FTIR and Raman analysis. From the FTIR analysis, it was evident that the oxidation process proceeded through the formation of intermediates like HCHO and HCOOH, while from the Raman study, it was concluded that the Ni and Ru centers participated more strongly in the MOR process.

As the mining depth increases, the pressure and temperature within the coal seams rise correspondingly. This change has a non-negligible impact on the adsorption characteristics of gas within the coal seams. There exists a research gap regarding the chemical interaction between methane molecules and the active functional groups of coal under the influence of temperature and pressure. This work examined the alterations of functional groups before and after methane adsorption under different temperature and pressure conditions by Fourier transform infrared spectroscopy. Moreover, based on the experimental results of elemental analysis, carbon-13 nuclear magnetic resonance, and X-ray photoelectron spectroscopy experiments, a coal molecular model was constructed for simulating saturated adsorption and identifying dominant adsorption sites of methane. Finally, based on density functional theory, a transition state search was conducted for the reaction between the functional groups and methane at the dominant adsorption positions. The results show the following: (i) the structural parameters of functional groups during the methane adsorption process by coal under different temperature and pressure conditions were calculated by means of infrared spectral peaks. A potential chemical interaction between the oxygen-containing functional groups within coal and methane transpired at 4 MPa and 140 °C. (ii) The methane saturation adsorption energy under different conditions demonstrates that neither the valence energy nor the nonbonding energy is zero during the process of coal adsorbing CH₄, implying that both physical adsorption and chemical adsorption occur during the process of coal adsorbing CH₄. (iii) The simulation results of reaction pathways disclose that methane molecules may chemically interact with −OCH₃, −COOH, and −COH in coal molecules at specific temperatures. The research is intended to augment the theory of methane adsorption in coal and establish a theoretical basis for the safe mining of deep coal and the prevention and control of gas disasters.

Mixed gas hydrates have been an area of active research for natural gas storage and transportation. While previous studies in the literature have focused on the structure, morphology, phase equilibria, and growth kinetics of these hydrates, limited data are available to describe their nucleation kinetics. This work reports the first measurement of mixed (methane–THF) gas hydrate nucleation rates and formation probability using a high-pressure stirred automated lag-time apparatus (HPS-ALTA). Ramped temperature experiments at three ramp rates were analyzed with the hazard function to extract nucleation rates, which were comparable with those previously measured for pure methane hydrate. Overall, these results suggest that THF is an effective gas hydrate promoter as it can yield hydrate formation at more favorable pressure/temperature conditions without compromising the associated nucleation rate.

Cyclanes serve as crucial platform compounds for the production of polymers, chemicals, and biofuels. Utilizing renewable biomass to synthesize cyclohexane is imperative not only to mitigate dependence on fossil fuels but also to advance green development strategies. The use of metal catalysts in the hydrodeoxygenation (HDO) reaction is essential for obtaining biomass-derived cyclanes. During the HDO process, the size of the metal particles and the adsorption capacity of the substrate significantly impact the reaction. This study reports the preparation of a fine and stable Ni nanoparticle material for an efficient HDO process. A postconfinement strategy was employed to synthesize a three-dimensional confined Ni/USY catalyst with finely dispersed particles. Initially, the Ni precursor was introduced into the ultrastable Y (USY) chemical microenvironment to facilitate nucleation and growth. Subsequently, H₂ reduction of Ni was within the three-dimensional confined scaffold at high temperatures. Compared with conventional Ni-based catalysts, these confined spaces restrict Ni aggregation, resulting in finer Ni particle sizes, increased exposure of active sites, and enhanced catalytic activity. To understand the mechanism for the HDO of soluble portion (SP), the catalytic activity of Ni/USY was investigated using 2,2-oxydinaphthalene (ODN) as lignin-related model compounds. As a result, ODN was completely converted to decalin at 160 °C under an initial hydrogen pressure of 2 MPa for 2.5 h, while the lignin was converted to soluble portion (SP) dominant by cyclanes and alkanes at 250 °C for 8 h over Ni/USY, and the yield of SP was about 83.3%. The characterization of Ni/USY and the time profiles of the product proved that Ni/USY can activate H₂ to H···H, which subsequently cleaves into mobile H⁺ and immobile H^–. The addition of H···H and H⁺ promoted the hydrogenation of benzene rings in the ODN and the removal of oxygen atoms, respectively. In addition, Ni/USY exhibits excellent reusability after 3 runs. In addition, the HDO properties and physicochemical properties of Ni loaded onto different kinds of zeolite supports were compared and used to reveal the reason for the good HDO performance of Ni/USY.

This study investigates the use of shape-modified silica nanoparticles functionalized with sodium (C14–16) olefin sulfonate (SOS) for enhancing oil recovery in oil-wet sandstone reservoirs. Characterization techniques, including scanning electron microscopy (SEM), Brunauer–Emmett–Teller (BET), Thermogravimetric analysis (TGA), and Fourier transform infrared spectroscopy (FTIR), verified successful surface modification. Functionalization reduced the mean particle size from 188 ± 15 to 98 ± 14 nm and enhanced stability, with zeta potential increasing from −11 to −46 mV. Nanoemulsion tests showed that SOS-functionalized nanoparticles achieved the lowest creaming degree and produced smaller oil droplets. The interfacial tension between crude oil and SOS-functionalized nanoparticles decreased from 24 to 1 mN/m, with further reductions observed upon the addition of alkali. Wettability alteration was also achieved, with contact angles shifting from 20° (oil-wet) to 173° (strongly water-wet) in the presence of SOS-functionalized nanoparticles. Spontaneous imbibition tests demonstrated oil recoveries of 77% with SOS-functionalized nanoparticles, outperforming SOS alone (42%) and unmodified nanoparticles (35%). Micro-CT scanning of the samples after imbibition test showed lower pore connectivity reduction with SOS-functionalized nanoparticles (31%) compared to unmodified nanoparticles (59%). Micromodel flooding tests confirmed enhanced oil recovery, with SOS-functionalized nanoparticles achieving 86% recovery compared to SOS (38%) and unmodified nanoparticles (18%). This study highlights the potential of SOS-functionalized silica nanoparticles to improve oil recovery in oil-wet sandstone reservoirs through wettability alteration, interfacial tension reduction, and stabilized emulsions.

CO₂ storage in the closed goaf is an essential negative carbon technical reserve to solve the carbon emissions in the coal industry, and ensuring the safety of carbon storage is a prerequisite for realizing this technology. The CO₂ breakthrough pressure of rocks and penetration time of strata in the bending subsidence belt are critical reference indices for evaluating the safety of CO₂ storage in a closed goaf. This study conducted CO₂ permeability and breakthrough pressure tests of rocks from a bending subsidence belt using the coal/rock solid–gas coupled comprehensive test system with the hydrogeological conditions of a real closed goaf. The effects of the axial pressure and water saturation on the rock CO₂ permeability and breakthrough pressure were analyzed. Based on the theory of gas–liquid two-phase flow in porous media, the CO₂–H₂O two-phase flow model of the fissure-matrix dual porous media was established, and a numerical simulation of CO₂ penetration in sand-mudstone combination caprocks at the site scale was performed innovatively. The effects of water saturation and fissures on the CO₂ penetration time and damage were analyzed. Finally, we provided an outlook on the CO₂ sealing enhancement techniques for bending subsidence belts. The results indicated that CO₂ breakthrough pressures in water-bearing sandstone range from 3.1 to 4.5 MPa in the mechanical environment of the bending subsidence belt. Internal fissures weakened the sealing ability of the combination caprocks to CO₂, but the increase in water saturation enhanced the sealing ability of the combination caprocks to CO₂. The penetration time of combination caprocks containing fissures decreased by 5.98–13.61% compared to those without fissures as the water saturation of the sandstone stratum increased from 25% to 100%. The higher the water saturation of the sandstone stratum, the greater the damage to the combination caprocks after CO₂ penetration, and the average damage of the mudstone is larger than that of the sandstone stratum. The internal fissures in the combination caprocks weaken this damaging effect. The study provides a foundation for the safety evaluation of CO₂ storage in a closed goaf.

CO₂ storage in geological formations, particularly deep saline aquifers, is a critical component of carbon capture and storage technology, offering significant potential for mitigating greenhouse gas emissions. However, high salinity of these aquifers poses the risk of salt precipitation, leading to pressurization and injectivity reduction. Developing a method to prevent salt precipitation remains a challenge, and this is an area that this study is focused on. Dissolved-water CO₂ injection (dwCO₂ injection) is proposed here as a novel method to prevent salt precipitation where water is dissolved in CO₂ before injection into an aquifer. Presence of water in the CO₂ stream prevents more dissolution of water into CO₂ (evaporation) and, hence, prevents salt precipitation. Before presenting this method and in order to provide a good mechanistic understanding of the interactions involved in a CO₂ storage process, six different scenarios are examined using the CMG-GEM simulator within a carbonate aquifer. The results showed that saturating CO₂ with water reduced the precipitation nearly to zero, and dissolving 2000 ppmv water decreased the salt precipitation to one-third. It should be noted that injection of humid CO₂ requires special methods to tackle the potential challenges, including corrosion and hydrate formation risks, and the paper also discusses them.

Aqueous zinc ion batteries (AZIBs) have gained unprecedented attention based on their high volumetric energy density, low price, and excellent eco-friendness. However, the persistent challenges posed by zinc dendrite formation and severe side reactions hinder their application in industry. Therefore, we introduce the in situ self-assembly of dodecanoic acid (DA) into an ultrathin amphiphilic layer on the zinc surface, which can suppress the troublesome side reactions and promote uniform deposition on the Zn anode. The carboxyl group of the DA molecular head is bonded onto the zinc anode surface, and the hydrocarbon tail is arranged vertically on the zinc foil surface to form a hydrophobic layer that effectively isolates the active zinc from the aqueous electrolyte. Accordingly, the symmetric cell of the DA-modified Zn (DA@Zn) achieved stable long cycle performance over 3000 h at an areal current density of 2 mA cm^–2 and a specific areal capacity of 1 mAh cm^–2. Furthermore, DA@Zn anodes also exhibited enhanced rate performance and cycling stability in full cells compared with bare zinc anodes. The introduction of the amphipathic DA layer onto the zinc anode provides new insights into the realization of stable zinc metal anodes.