Abstract The use of industrial wastes as oxygen carriers (OCs) in chemical looping combustion (CLC) has gained recent attention as these materials demonstrate the potential to be a cost-effective alternative to synthetic OCs. A key limitation in the development of CLC with industrial waste OCs is the lack of modeling efforts for CLC systems with these materials. This work presents a dynamic multiscale packed-bed reactor CLC model of the oxidation and reduction stages to investigate the performance of red mud, an industrial waste from the alumina refining industry, as an OC. Kinetics describing the oxidation reaction of red mud with oxygen and the reduction reactions of red mud with CH4, CO, and H2 fuels were fitted and assessed using lab-scale experimental data. To further advance CLC with red mud, the packed-bed model is used to design a pilot-scale CLC system, following scaling constraints and design heuristics. The proposed pilot-scale design exhibits moderate utilization of the red mud OC, an average of 89% fuel conversion during the dynamic reduction stage, and is estimated to produce a total thermal energy output of 235.6 kWh. Evaluation of the red mud performance under various reactor conditions at pilot scale provides valuable insights into the potential for this industrial waste to advance CLC toward industrial relevance and adoption.
Abstract We present an ion-specific parameterization of the electrolyte Cubic-Plus-Association (eCPA) equation of state, which we refer to as ms-eCPA, for phase equilibria of CO2 and CH4 in multisalt aqueous brines containing Na+, Cl–, K+, Ca2+, Mg2+, and SO42–. The model includes a Soave–Redlich–Kwong physical term with a Huron–Vidal mixing rule, Wertheim 4C association of water with CO2 cross-association, and Debye–Hückel plus Born electrostatic contributions on an ionic basis; the static permittivity is closed self-consistently through the Onsager–Kirkwood relation. The parameterization is hierarchical─each level of added physical description (nonpolar fluids, the associating solvent, ions in the solvent, gases over brine) introduces only parameters fitted to that level’s own data─with the top-level gas–ion interaction energies and ion volume translations calibrated against 1,931 experimental CO2–brine VLE points in five salts (T = 283–473 K, P = 1.3–1200 bar, ms = 0.09–6 mol kg–1), 581 CH4–brine points in six salts (T = 298–598 K, P = 4–1379 bar, ms = 0.25–6 mol kg–1), and 905 high-accuracy brine densities (T = 283–473 K, P = 9–686 bar). A colligative verification of the inherited ion parameters against the Pitzer osmotic-coefficient standard uncovers, and corrects, two defects in the published ion set (the Ca2+ parameters and the Mg2+ polarizability). Across the parametrized range, at salt molalities to 6 mol kg–1, the model predicts CO2 solubility to within 4.3–11.1% AARE per salt and CH4 solubility within 1.6–10.2% AARE per salt, reproduces brine density across the full salinity and temperature range through a temperature-dependent ion-specific Péneloux volume translation with a single ion-additive coefficient set, predicts CO2 solubility in mixed-salt brines with no mixed-brine tuning, and predicts measured calorimetric enthalpies of CO2 dissolution in water and NaCl brines to 0.6 kJ mol–1 with no adjusted parameters. The parameterization provides the thermodynamic basis for subsurface CO2 sequestration and CH4 production in realistic formation brines. In a separate study, the model is paired with robust two- and three-phase stability and flash computations and extended to N2, H2, C2H6, C3H8, and CH4–CO2 gas mixtures.
Abstract The catalytic hydrodeoxygenation (HDO) of biomass-derived oxygenates requires bifunctional catalysts that combine acid and metal sites, yet the reaction networks of polyfunctional substrates remain incompletely understood. In this paper, two bifunctional catalysts based on sulfonated porous aromatic frameworks with different platinum loadings (PAF-30-SO3H–Pt(1) and PAF-30-SO3H–Pt(3)) were prepared and studied in the HDO of two model substrates: 1,1-bis(5-methylfuran-2-yl)cyclohexane (C6-(MF)2) and the condensation products of cyclohexanone. These two substrates represent two classes of biomass-derived SAF precursors─the furanic C6-(MF)2, obtained by alkylation of lignocellulose-derived 2-methylfuran with cyclohexanone, and the self-condensation products of cyclohexanone, a common model compound of lignin bio-oils accessible by partial hydrodeoxygenation of lignin-derived phenolics. Their hydrodeoxygenation yields fuel-range naphthenic hydrocarbons, suitable as SAF components. The catalysts were characterized by CHNS and ICP-AES analysis, FTIR, XPS, N2 adsorption, TEM, and SEM, and the reaction products were identified by GC-MS. The transformations proceeded through the combined action of the acidic SO3H groups and platinum. Additional reaction routes were identified for both substrates, including the formation of methyldecalins and an acid-catalyzed cyclization competing with hydrogenation, which was markedly more favorable for the cyclohexanone trimers, as supported by DFT calculations. Higher temperature, H2 pressure, and platinum loading increased the yield of hydrocarbons, which reached 91% for the cyclohexanone condensation products and 51% for 1,1-bis(5-methylfuran-2-yl)cyclohexane after 4 h of reaction at 250 °C and H2 pressure of 5 MPa. Upon reuse, the catalyst was substantially deactivated, mainly through the accumulation of carbonaceous deposits that blocked the pores, while the framework and sulfonic groups remained intact.
Abstract Determining the concentrations of rare earth elements and yttrium (REY) in coal is essential for evaluating coal-hosted critical metal resources and understanding their geochemical controls. However, conventional laboratory analyses are costly, time-consuming, and spatially sparse, limiting regional-scale assessment. In this study, we compiled 527 coal samples from 30 coalfields across China’s five major coal-bearing regions and developed an explainable machine learning framework for REY prediction. Unlike previous site-specific models, this framework integrates feature selection, Bayesian optimization, and SHAP interpretation for REY prediction at a broad regional scale. After data preprocessing, multicollinearity diagnosis, and correlation-based feature selection, 16 predictors were retained for model construction. Four ensemble regression models, including Random Forest (RF), Extra Trees (ET), CatBoost (CAT), and Gradient Boosting Decision Tree (GBDT), were optimized using Bayesian hyperparameter search and 5-fold cross-validation. The ET model achieved the best test-set performance among the four ensemble models, with RMSE = 39.962, MAE = 29.669, and R2 = 0.716, although predictions for highly enriched samples remained uncertain because of the long-tailed REY distribution. SHAP analysis identified Ga as the most influential predictor, followed by geological age and P2O5, suggesting that clay-related terrigenous input, stratigraphic context, and minor phosphate-associated phases may jointly influence model-predicted REY enrichment. These results demonstrate that the proposed framework can serve as an effective prescreening tool to prioritize coal samples for detailed REY analysis, while independent validation is recommended for high-value target areas.
Abstract SO2 removal in marine exhaust gas cleaning systems is evaluated together with the CO2 concentration. Absorbent design must therefore consider selective SO2 removal in the presence of CO2. This study compared the SO2 absorption performance, pH stability, and SO2/CO2 selectivity of NaOH, NaHCO3, and Na2CO3 under SO2-only conditions and in the presence of 8% CO2 using simulated marine exhaust gas. At an equal compound molarity of 0.25 M under SO2-only conditions, Na2CO3 showed the highest conversion retention and pH stability. Its average SO2 conversion was 99.6%, the duration at pH ≥ 7 was 98.6 min, and t90 exceeded 120 min. Na2CO3 also exhibited the highest overall gas-phase mass transfer coefficient, KG, of 8.932 × 10–3 mol·m–2 · s–1 · kPa–1 in wetted-wall column (WWC) tests. In the presence of 8% CO2, all three absorbents maintained approximately 99% SO2 conversion. NaHCO3 showed the highest uptake-based apparent SO2/CO2 selectivity of 0.213. These results guided the design of NaHCO3–Na2CO3–NaOH buffer-type mixed absorbents. The 1%-A and 5%-B formulations were selected from the 1 and 5 wt % groups, respectively, as candidates for subsequent wet scrubber evaluation. Both maintained pH ≥ 7 for 180 min in the bubble reactor. Their apparent CO2 uptakes were 36.6 and 22.9 mmol·L-solution–1, respectively. In the wet scrubber, 1%-A and 5%-B achieved SO2 conversions of 96.5% and 94.6%, respectively, while the outlet CO2 concentration remained close to the inlet level. The study identifies bicarbonate-rich, low-NaOH buffer-type formulations that maintain SO2 removal while showing a small decrease in outlet CO2 concentration.
