Pub Date : 2026-01-22DOI: 10.1016/j.cherd.2026.01.044
Juan Ramiro Lezama , Sofía Micaela Guerrero Soler , Laura Emilia Giménez , Eleonora Erdmann
The objective of this study is to conduct a comparative analysis of the two principal technologies for obtaining lithium carbonate currently: evaporative processes versus Direct Lithium Extraction (DLE) based on selective adsorption with resins, to determine the production potential for each of them and the convenience of their implementation in salars. The mass and energy balances of each process are resolved, considering a common calculation base. Aspen Plus is used as a simulation tool to access the thermodynamic data of the chemical species involved. Key Performance Indicators (KPIs) are determined, including brine feed, water consumption, reagents used, and energy consumption per ton of lithium carbonate produced. In turn, the interaction between virgin brine and depleted brine is evaluated. Each method has advantages and disadvantages in terms of these indicators, which must be evaluated for each project. In addition, electrical energy consumption was included to compare the peak power demand between both processes.
The selection of these processes necessitates an evaluation of environmental, economic, and social factors to guarantee the sustainable advancement of the lithium sector in these areas. A better approach, from the production point of view, optimizing the factors mentioned, is to propose a combined scheme of both technologies.
{"title":"Evaluation of current processes for lithium carbonate production: Determination of key performance indicators","authors":"Juan Ramiro Lezama , Sofía Micaela Guerrero Soler , Laura Emilia Giménez , Eleonora Erdmann","doi":"10.1016/j.cherd.2026.01.044","DOIUrl":"10.1016/j.cherd.2026.01.044","url":null,"abstract":"<div><div>The objective of this study is to conduct a comparative analysis of the two principal technologies for obtaining lithium carbonate currently: evaporative processes versus Direct Lithium Extraction (DLE) based on selective adsorption with resins, to determine the production potential for each of them and the convenience of their implementation in salars. The mass and energy balances of each process are resolved, considering a common calculation base. Aspen Plus is used as a simulation tool to access the thermodynamic data of the chemical species involved. Key Performance Indicators (KPIs) are determined, including brine feed, water consumption, reagents used, and energy consumption per ton of lithium carbonate produced. In turn, the interaction between virgin brine and depleted brine is evaluated. Each method has advantages and disadvantages in terms of these indicators, which must be evaluated for each project. In addition, electrical energy consumption was included to compare the peak power demand between both processes.</div><div>The selection of these processes necessitates an evaluation of environmental, economic, and social factors to guarantee the sustainable advancement of the lithium sector in these areas. A better approach, from the production point of view, optimizing the factors mentioned, is to propose a combined scheme of both technologies.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"227 ","pages":"Pages 106-119"},"PeriodicalIF":3.9,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146076144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-21DOI: 10.1016/j.cherd.2026.01.043
Yellam Naidu Kottavalasa , Andrea Battaglia , Giovanni Bevilacqua , Gianni Marchetti , Andrea Salfinger , Lauro Snidaro
Fouling, a phenomenon in which materials originating from the process fluid settle onto heat-exchange surfaces, significantly reduces thermal efficiency, increases energy consumption, and raises maintenance costs, particularly in high-pressure tubular reactors used for Ethylene-Vinyl Acetate (EVA) polymerization. Accurate, real-time prediction of fouling factor is therefore essential for optimizing operational efficiency, maintaining product quality, and preventing unplanned downtime. This paper proposes a novel attention-based neural network that integrates parallel Bidirectional Gated Recurrent Unit branches with a Multi-Head Attention mechanism to enhance temporal feature extraction and focus on the most informative time steps. In addition to the neural architecture, the framework incorporates Mutual Information-based feature selection stage to retain highly relevant process variables, derived from temperature, pressure, and flow rate measurements collected through sensors across the reactor system. The model was trained on six years of industrial EVA reactor data from Versalis. Experimental results demonstrate that the proposed model consistently outperforms baseline architectures, achieving the lowest test MSE (), RMSE (), and highest (0.82) on normalized data. These improvements highlight the model ability to capture complex temporal dependencies and generalize under varying operational conditions. The proposed approach offers a scalable and effective solution for predictive fouling monitoring in polymerization heat exchangers, with potential applicability across other energy-intensive chemical manufacturing processes.
{"title":"Attention-based neural network fusion for fouling prediction in Ethylene-Vinyl Acetate heat exchangers","authors":"Yellam Naidu Kottavalasa , Andrea Battaglia , Giovanni Bevilacqua , Gianni Marchetti , Andrea Salfinger , Lauro Snidaro","doi":"10.1016/j.cherd.2026.01.043","DOIUrl":"10.1016/j.cherd.2026.01.043","url":null,"abstract":"<div><div>Fouling, a phenomenon in which materials originating from the process fluid settle onto heat-exchange surfaces, significantly reduces thermal efficiency, increases energy consumption, and raises maintenance costs, particularly in high-pressure tubular reactors used for Ethylene-Vinyl Acetate (EVA) polymerization. Accurate, real-time prediction of fouling factor is therefore essential for optimizing operational efficiency, maintaining product quality, and preventing unplanned downtime. This paper proposes a novel attention-based neural network that integrates parallel Bidirectional Gated Recurrent Unit branches with a Multi-Head Attention mechanism to enhance temporal feature extraction and focus on the most informative time steps. In addition to the neural architecture, the framework incorporates Mutual Information-based feature selection stage to retain highly relevant process variables, derived from temperature, pressure, and flow rate measurements collected through sensors across the reactor system. The model was trained on six years of industrial EVA reactor data from Versalis. Experimental results demonstrate that the proposed model consistently outperforms baseline architectures, achieving the lowest test MSE (<span><math><mrow><mn>3</mn><mo>.</mo><mn>48</mn><mo>×</mo><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>3</mn></mrow></msup></mrow></math></span>), RMSE (<span><math><mrow><mn>4</mn><mo>.</mo><mn>17</mn><mo>×</mo><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>2</mn></mrow></msup></mrow></math></span>), and highest <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> (0.82) on normalized data. These improvements highlight the model ability to capture complex temporal dependencies and generalize under varying operational conditions. The proposed approach offers a scalable and effective solution for predictive fouling monitoring in polymerization heat exchangers, with potential applicability across other energy-intensive chemical manufacturing processes.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"227 ","pages":"Pages 23-34"},"PeriodicalIF":3.9,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-20DOI: 10.1016/j.cherd.2026.01.042
Kang Zhao , Shasha Zhao , Zhen Liu
Addressing the core challenge of insufficient selectivity of traditional solvents in the separation of long-chain α-olefins/alkanes (such as 1-octene/n-octane), this study innovatively constructs a silver trifluoromethanesulfonate (AgOTf)/N,N-dimethylformamide (DMF) deep eutectic solvent (DES) system based on the “Ag⁺-DMF” synergistic coordination and hydrogen bonding network mechanism for the extraction and separation of 1-octene/n-octane. Through systematic optimization of process parameters, under conditions of 273 K, an alkane/olefin mass ratio of 2:1, water content of 5 %, HBA:HBD molar ratio of 1:2, and a rotational speed of 200 rpm, the system achieves a remarkable separation selectivity of 27.23 for 1-octene/n-octane. Quantum chemical calculations based on Density Functional Theory (DFT) indicate that 1-octene is specifically captured by AgOTf/DMF through strong Ag⁺-π bonding (binding energy: −45.0 kJ/mol) and OTf⁻∙∙∙H-C hydrogen bonding. In contrast, n-octane exhibits only weak non-specific interactions with AgOTf/DMF (binding energy: −0.2 kJ/mol), resulting in a binding energy difference of 44.8 kJ/mol, which overcomes the separation bottleneck caused by the similar physicochemical properties of olefins and alkanes. Additionally, the DES demonstrates excellent low-temperature regenerability and cyclic stability, offering a promising new solvent solution for industrial olefin/alkane separation.
{"title":"Silver-based deep eutectic solvent for the extraction and separation of 1-octene/n-octane: Insights into the underlying molecular mechanism","authors":"Kang Zhao , Shasha Zhao , Zhen Liu","doi":"10.1016/j.cherd.2026.01.042","DOIUrl":"10.1016/j.cherd.2026.01.042","url":null,"abstract":"<div><div>Addressing the core challenge of insufficient selectivity of traditional solvents in the separation of long-chain α-olefins/alkanes (such as 1-octene/n-octane), this study innovatively constructs a silver trifluoromethanesulfonate (AgOTf)/N,N-dimethylformamide (DMF) deep eutectic solvent (DES) system based on the “Ag⁺-DMF” synergistic coordination and hydrogen bonding network mechanism for the extraction and separation of 1-octene/n-octane. Through systematic optimization of process parameters, under conditions of 273 K, an alkane/olefin mass ratio of 2:1, water content of 5 %, HBA:HBD molar ratio of 1:2, and a rotational speed of 200 rpm, the system achieves a remarkable separation selectivity of 27.23 for 1-octene/n-octane. Quantum chemical calculations based on Density Functional Theory (DFT) indicate that 1-octene is specifically captured by AgOTf/DMF through strong Ag⁺-π bonding (binding energy: −45.0 kJ/mol) and OTf<sup>⁻</sup>∙∙∙H-C hydrogen bonding. In contrast, n-octane exhibits only weak non-specific interactions with AgOTf/DMF (binding energy: −0.2 kJ/mol), resulting in a binding energy difference of 44.8 kJ/mol, which overcomes the separation bottleneck caused by the similar physicochemical properties of olefins and alkanes. Additionally, the DES demonstrates excellent low-temperature regenerability and cyclic stability, offering a promising new solvent solution for industrial olefin/alkane separation.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"226 ","pages":"Pages 656-666"},"PeriodicalIF":3.9,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurately forecasting the fast transients that govern catalytic reactors remains difficult because first-principles ordinary differential equation (ODE) models neglect unmodelled heat and mass-transfer effects and therefore perform poorly (baseline CO-oxidation rate = –0.231). For the above reason, this study presents a systematic hybrid mechanistic machine-learning (ML) framework that couples a physically rigorous CSTR model with data-driven residual learning to close these physics gaps. A six-factor design of experiments generated 500 operating scenarios, and after simulation, quality screening, derivative estimation, and residual/outlier filtering, the residual-learning dataset comprised approximately 33,096 usable samples. Five regressors (XGBoost, LightGBM, SVR, MLP and sparse Gaussian-process regression) were hyperparameter-tuned with Optuna and blended through weight optimisation. Uncertainty was propagated with GP posterior bands and inter-model disagreement. The optimised ensemble lifted test-set accuracy to = 0.755, RMSE = 0.006 and MdAPE = 93 % a dramatic recovery over the mechanistic baseline. ±2σ GP bands captured 94 % of unseen points, providing actionable epistemic bounds. Performance deteriorated by only ∼21 % when 5 % Gaussian sensor noise was injected, confirming robustness for on-line use. By modularising experiment design, physics-guided feature engineering, automated model selection, and calibrated uncertainty quantification, this workflow delivers interpretable, real-time-capable surrogate models within the modelled operating envelope, outperforming pure ODE and single-model ML baselines. The protocol is transferable to other catalytic systems and establishes a reproducible path toward uncertainty-aware reactor optimisation and control.
{"title":"A systematic hybrid mechanistic–machine learning framework for catalytic reactor modelling and computational validation using CO oxidation","authors":"Ebenezer Aquisman Asare , Dickson Abdul-Wahab , Elsie Effah Kaufmann , Rafeah Wahi , Zainab Ngaini , Abigail Ampadu","doi":"10.1016/j.cherd.2026.01.039","DOIUrl":"10.1016/j.cherd.2026.01.039","url":null,"abstract":"<div><div>Accurately forecasting the fast transients that govern catalytic reactors remains difficult because first-principles ordinary differential equation (ODE) models neglect unmodelled heat and mass-transfer effects and therefore perform poorly (baseline CO-oxidation rate <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> = –0.231). For the above reason, this study presents a systematic hybrid mechanistic machine-learning (ML) framework that couples a physically rigorous CSTR model with data-driven residual learning to close these physics gaps. A six-factor design of experiments generated 500 operating scenarios, and after simulation, quality screening, derivative estimation, and residual/outlier filtering, the residual-learning dataset comprised approximately 33,096 usable samples. Five regressors (XGBoost, LightGBM, SVR, MLP and sparse Gaussian-process regression) were hyperparameter-tuned with Optuna and blended through weight optimisation. Uncertainty was propagated with GP posterior bands and inter-model disagreement. The optimised ensemble lifted test-set accuracy to <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> = 0.755, RMSE = 0.006 <span><math><mrow><mi>mol·</mi><msup><mrow><mi>m</mi></mrow><mrow><mo>−</mo><mn>3</mn></mrow></msup><mi>·</mi><msup><mrow><mi>s</mi></mrow><mrow><mo>−</mo><mn>1</mn></mrow></msup></mrow></math></span> and MdAPE = 93 % a dramatic recovery over the mechanistic baseline. ±2σ GP bands captured 94 % of unseen points, providing actionable epistemic bounds. Performance deteriorated by only ∼21 % when 5 % Gaussian sensor noise was injected, confirming robustness for on-line use. By modularising experiment design, physics-guided feature engineering, automated model selection, and calibrated uncertainty quantification, this workflow delivers interpretable, real-time-capable surrogate models within the modelled operating envelope, outperforming pure ODE and single-model ML baselines. The protocol is transferable to other catalytic systems and establishes a reproducible path toward uncertainty-aware reactor optimisation and control.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"227 ","pages":"Pages 1-22"},"PeriodicalIF":3.9,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146015801","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-20DOI: 10.1016/j.cherd.2026.01.040
Harwin Sandhu, Sangeeta Garg, Shashikant Yadav
Mineral carbonation using slurry-phase wollastonite (CaSiO3) composed of suspended micron-scale particles represents an effective strategy for permanent CO2 sequestration, yet the interactions among ionic strength, dissolution–precipitation kinetics, and surface passivation remain poorly constrained. This study presents a mechanistic thermodynamic–kinetic model integrating CO2 solubility, wollastonite dissolution, and silica and calcium carbonate precipitation in multi-ionic brines (NaCl, MgCl2, CaCl2, Na2SO4) relevant to deep saline aquifers. CO2 solubility decreases across all brines due to salting-out, most strongly in Mg2+ - and SO42--rich systems, while wollastonite buffering enhances uptake with solubility ratio coefficients up to 1.95. Dissolution rates peak near pH 3–4 but drop by more than 30 % in Mg2+ -rich solutions because of competitive adsorption of Mg2+ with Ca2+ at wollastonite surface sites, which suppresses Ca2+ release. In sulfate-rich brines, SO42- primarily regulates Ca2+ activity through sulfate complexation and potential gypsum buffering, thereby delaying the onset of CaCO3 supersaturation and precipitation. Silica precipitation evolves from reaction-controlled polymerization to diffusion-limited growth as passivating layers develop. CaCO3 precipitation is triggered only at high supersaturation, limited by CO32- transport and diffusional barriers at elevated pH. Ion flux declines by over 95 % once silica layers approach ∼500 nm, corresponding to a sharp reduction in effective diffusivity and permeability due to pore-space occlusion by secondary mineral phases. Interactions between silica layers and co-precipitating minerals, such as carbonates, modulate layer porosity and diffusivity, suggesting that insights from shale reservoirs can refine predictions of passivation and reaction–diffusion transitions in engineered brine systems. The model predicts a progressive decrease in porosity and transport capacity as silica and CaCO3 layers thicken, providing a quantitative mechanistic link between mineral reprecipitation, evolving transport properties, and the observed transition from reaction-controlled to diffusion-limited carbonation.
{"title":"Modeling kinetics of wollastonite dissolution and carbonate precipitation in multi-ionic brine systems","authors":"Harwin Sandhu, Sangeeta Garg, Shashikant Yadav","doi":"10.1016/j.cherd.2026.01.040","DOIUrl":"10.1016/j.cherd.2026.01.040","url":null,"abstract":"<div><div>Mineral carbonation using slurry-phase wollastonite (CaSiO<sub>3</sub>) composed of suspended micron-scale particles represents an effective strategy for permanent CO<sub>2</sub> sequestration, yet the interactions among ionic strength, dissolution–precipitation kinetics, and surface passivation remain poorly constrained. This study presents a mechanistic thermodynamic–kinetic model integrating CO<sub>2</sub> solubility, wollastonite dissolution, and silica and calcium carbonate precipitation in multi-ionic brines (NaCl, MgCl<sub>2</sub>, CaCl<sub>2</sub>, Na<sub>2</sub>SO<sub>4</sub>) relevant to deep saline aquifers. CO<sub>2</sub> solubility decreases across all brines due to salting-out, most strongly in Mg<sup>2+</sup> - and SO<sub>4</sub><sup>2-</sup>-rich systems, while wollastonite buffering enhances uptake with solubility ratio coefficients up to 1.95. Dissolution rates peak near pH 3–4 but drop by more than 30 % in Mg<sup>2+</sup> -rich solutions because of competitive adsorption of Mg<sup>2+</sup> with Ca<sup>2+</sup> at wollastonite surface sites, which suppresses Ca<sup>2+</sup> release. In sulfate-rich brines, SO<sub>4</sub><sup>2-</sup> primarily regulates Ca<sup>2+</sup> activity through sulfate complexation and potential gypsum buffering, thereby delaying the onset of CaCO<sub>3</sub> supersaturation and precipitation. Silica precipitation evolves from reaction-controlled polymerization to diffusion-limited growth as passivating layers develop. CaCO<sub>3</sub> precipitation is triggered only at high supersaturation, limited by CO<sub>3</sub><sup>2-</sup> transport and diffusional barriers at elevated pH. Ion flux declines by over 95 % once silica layers approach ∼500 nm, corresponding to a sharp reduction in effective diffusivity and permeability due to pore-space occlusion by secondary mineral phases. Interactions between silica layers and co-precipitating minerals, such as carbonates, modulate layer porosity and diffusivity, suggesting that insights from shale reservoirs can refine predictions of passivation and reaction–diffusion transitions in engineered brine systems. The model predicts a progressive decrease in porosity and transport capacity as silica and CaCO<sub>3</sub> layers thicken, providing a quantitative mechanistic link between mineral reprecipitation, evolving transport properties, and the observed transition from reaction-controlled to diffusion-limited carbonation.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"227 ","pages":"Pages 54-76"},"PeriodicalIF":3.9,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-19DOI: 10.1016/j.cherd.2026.01.038
Yuxiu Zhang , Xiangang Liu , Lei Wang , Bingzheng Song , Haibo Gao , Jiaqi Zong , Yongqi Liu , Xiaofei Pan , Yanxia Wang
The motion behavior of boiler bottom slag in a heat exchanger with radial pipes was experimentally investigated to enhance waste heat recovery. The results indicate that increasing rotational speed from 1.5 to 5.5 rpm leads to a steady rise in the repose angle, accompanied by an increase in active zone thickness from 10 mm to 30 mm and a reduction in passive zone thickness from 55 mm to 35 mm. As the filling ratio increases from 8 % to 24 %, both upper and lower repose angles show a moderate upward trend, while the thicknesses of the active and passive zones increase simultaneously. In contrast, raising the number of radial pipes from 0 to 8 enhances boiler bottom slag agitation, resulting in larger repose angles, a thicker active zone from 10 mm to 23 mm, and a thinner passive zone from 58 mm to 46 mm. Pearson correlation analysis reveals that, for the repose angle, rotational speed (r ≈ 0.95) and the number of radial pipes (r ≈ 0.94) are the dominant factors, both far exceeding the influence of filling ratio (r < 0.9). Regarding the zone thicknesses, the number of radial pipes shows the strongest positive correlation with the active zone (r ≈ 0.99) and a corresponding strong negative correlation with the passive zone (r ≈ -0.99), followed by the influence of rotational speed, while the filling ratio again exhibits a secondary role. These findings establish radial pipes as dual-functional elements that simultaneously enhance heat transfer surface and actively modify granular flow, providing quantitative guidelines for optimizing exchanger design.
{"title":"Experimental investigation on the motion behavior of boiler bottom slag in heat exchanger with radial pipes","authors":"Yuxiu Zhang , Xiangang Liu , Lei Wang , Bingzheng Song , Haibo Gao , Jiaqi Zong , Yongqi Liu , Xiaofei Pan , Yanxia Wang","doi":"10.1016/j.cherd.2026.01.038","DOIUrl":"10.1016/j.cherd.2026.01.038","url":null,"abstract":"<div><div>The motion behavior of boiler bottom slag in a heat exchanger with radial pipes was experimentally investigated to enhance waste heat recovery. The results indicate that increasing rotational speed from 1.5 to 5.5 rpm leads to a steady rise in the repose angle, accompanied by an increase in active zone thickness from 10 mm to 30 mm and a reduction in passive zone thickness from 55 mm to 35 mm. As the filling ratio increases from 8 % to 24 %, both upper and lower repose angles show a moderate upward trend, while the thicknesses of the active and passive zones increase simultaneously. In contrast, raising the number of radial pipes from 0 to 8 enhances boiler bottom slag agitation, resulting in larger repose angles, a thicker active zone from 10 mm to 23 mm, and a thinner passive zone from 58 mm to 46 mm. Pearson correlation analysis reveals that, for the repose angle, rotational speed (<em>r</em> ≈ 0.95) and the number of radial pipes (<em>r</em> ≈ 0.94) are the dominant factors, both far exceeding the influence of filling ratio (<em>r</em> < 0.9). Regarding the zone thicknesses, the number of radial pipes shows the strongest positive correlation with the active zone (<em>r</em> ≈ 0.99) and a corresponding strong negative correlation with the passive zone (<em>r</em> ≈ -0.99), followed by the influence of rotational speed, while the filling ratio again exhibits a secondary role. These findings establish radial pipes as dual-functional elements that simultaneously enhance heat transfer surface and actively modify granular flow, providing quantitative guidelines for optimizing exchanger design.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"226 ","pages":"Pages 561-568"},"PeriodicalIF":3.9,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-16DOI: 10.1016/j.cherd.2026.01.036
Nikolai A. Jessen , Alexander Findeisen , Krist V. Gernaey , Ulrich Krühne
Spray dryers are highly dynamic systems in which complex airflow patterns, heat and mass transfer, and particle motion interact across multiple length and time scales. Accurately resolving these interactions with fully transient CFD-DPM simulations is computationally expensive, limiting their use for process optimization. This study addresses this challenge by developing and validating a frozen time-averaged airflow CFD-DPM approach for a co-current spray dryer and embedding it within a surrogate-based inverse optimization framework. Compared to fully transient simulations, the frozen airflow approach reproduces key product outputs with comparable accuracy while reducing computational time from approximately 72 h to 15 min, corresponding to a speedup close to 300 for a pilot-scale co-current spray dryer. The workflow is used to train a polynomial surrogate model embedded in an inverse optimization framework. The differences between surrogate and CFD-DPM results remain small, with mean particle temperature deviations below 2 °C, particle diameter differences under 1.2 µm, moisture discrepancies below 0.024 kg kg⁻¹ , and product yield differences below 0.01. The proposed framework can be used as a practical tool for examining the operating input space and for rapid spray dryer optimization without the need for expensive computational hardware.
喷雾干燥机是高度动态的系统，其中复杂的气流模式，传热和传质，以及粒子运动在多个长度和时间尺度上相互作用。通过完全瞬态CFD-DPM模拟准确地解决这些相互作用在计算上是昂贵的，限制了它们在工艺优化中的使用。本研究通过开发和验证用于共流喷雾干燥机的冻结时间平均气流CFD-DPM方法，并将其嵌入基于代理的逆优化框架，解决了这一挑战。与完全瞬态模拟相比，冻结气流方法以相当的精度再现了关键产品的输出，同时将计算时间从大约72 h减少到15 min，相当于中试规模的共流喷雾干燥机的加速接近300。该工作流用于训练嵌入在逆优化框架中的多项式代理模型。代孕法和CFD-DPM法的结果差异很小，平均颗粒温度偏差小于2°C，颗粒直径偏差小于1.2 µm，水分偏差小于0.024 kg kg⁻¹ ，产物得率差异小于0.01。所提出的框架可以用作检查操作输入空间和快速喷雾干燥机优化的实用工具，而不需要昂贵的计算硬件。
{"title":"Inverse optimization of spray drying by surrogate models trained on frozen-field CFD simulations","authors":"Nikolai A. Jessen , Alexander Findeisen , Krist V. Gernaey , Ulrich Krühne","doi":"10.1016/j.cherd.2026.01.036","DOIUrl":"10.1016/j.cherd.2026.01.036","url":null,"abstract":"<div><div>Spray dryers are highly dynamic systems in which complex airflow patterns, heat and mass transfer, and particle motion interact across multiple length and time scales. Accurately resolving these interactions with fully transient CFD-DPM simulations is computationally expensive, limiting their use for process optimization. This study addresses this challenge by developing and validating a frozen time-averaged airflow CFD-DPM approach for a co-current spray dryer and embedding it within a surrogate-based inverse optimization framework. Compared to fully transient simulations, the frozen airflow approach reproduces key product outputs with comparable accuracy while reducing computational time from approximately 72 h to 15 min, corresponding to a speedup close to 300 for a pilot-scale co-current spray dryer. The workflow is used to train a polynomial surrogate model embedded in an inverse optimization framework. The differences between surrogate and CFD-DPM results remain small, with mean particle temperature deviations below 2 °C, particle diameter differences under 1.2 µm, moisture discrepancies below 0.024 kg kg⁻¹ , and product yield differences below 0.01. The proposed framework can be used as a practical tool for examining the operating input space and for rapid spray dryer optimization without the need for expensive computational hardware.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"226 ","pages":"Pages 611-625"},"PeriodicalIF":3.9,"publicationDate":"2026-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-15DOI: 10.1016/j.cherd.2026.01.025
Richel Annan Dadzie , Massimiliano Zanin , William Skinner , Richmond Asamoah , Jonas Addai-Mensah , George Blankson Abaka-Wood
Reprocessing copper flotation tailings is limited by particle-size effects and mineralogical challenges, especially poor liberation. Preliminary studies aimed at recovering copper minerals from complex, low-grade rougher flotation tailings have shown limited success, with values within the intermediate particle size range (-150 +53 µm) failing in conventional mechanical flotation cells. To improve copper recovery, the current study employs a flowsheet that uses Hydrofloat ™ fluidised-bed flotation on the deslimed + 53 µm fraction, considers the REFLUX ™ Flotation Cell (RFC) for the (-53µm) slimes, and compares performance against a Denver mechanical cell. Flotation performance (recovery, grade, size-by-size) was combined with data from Quantitative Evaluation of Minerals by Scanning Electron Microscopy (QEMSCAN), including liberation and locking statistics, to explain the flotation response observed. At similar mass pull (29.3–33 %), Hydrofloat ™ achieved 61.8 % Cu recovery at 0.74 % Cu (upgrade ratio 3.44), outperforming the mechanical flotation cell (44. 6 % Cu at 0. 10 % Cu grade; upgrade ratio 0.6) when processing the deslimed feed, rather than the whole (unsplit) tailings feed. Preliminary RFC tests on the - 53 µm stream maintained high recoveries of fine particles (often >80 %) with improved Cu concentrate grades, aligning with the presence of well-liberated chalcopyrite in slimes. Overall, the results support a split-flotation process for complex copper low grade ores. Although liberation ultimately limits recovery this integrated method significantly enhances copper recovery and upgrade relative to conventional mechanical cells, offering a practical route to unlock value from low-grade sulphide tailings.
铜浮选尾矿的再处理受到粒度效应和矿物学挑战的限制，尤其是解离性差。从复杂的、低品位的粗细浮选尾矿中回收铜矿物的初步研究表明，在常规机械浮选池中，在中等粒度范围内（-150 +53 µm）的数值是失败的。为了提高铜的回收率，目前的研究采用了一个流程，在脱泥+ 53 µm部分使用Hydrofloat™流化床浮选，考虑了（-53µm）泥的REFLUX™浮选池（RFC），并将其性能与丹佛机械池进行了比较。浮选性能（回收率、品位、粒度）与QEMSCAN矿物定量评价（Quantitative Evaluation of Minerals by Scanning Electron Microscopy， QEMSCAN）的数据（包括解离和锁定统计）相结合，来解释观察到的浮选反应。在相同的质量拉力（29.3-33 %）下，Hydrofloat™的铜回收率为61.8 %，铜回收率为0.74 %（升级比3.44），优于机械浮选池（44）。6 % Cu at 0。10 % Cu品位；在处理脱泥料时升级比0.6)，而不是处理整个（未分裂）尾矿料。在- 53 µm流上进行的初步RFC测试保持了高细颗粒回收率（通常为>； 80% %），铜精矿品位提高，与泥中黄铜矿的充分释放一致。总体而言，研究结果支持对复杂低品位铜矿石进行分选浮选。尽管释放最终限制了铜的回收，但与传统的机械电池相比，这种集成方法显著提高了铜的回收和升级，为从低品位硫化物尾矿中释放价值提供了实用的途径。
{"title":"Enhanced size-split flotation of sulphide tailings: Mechanical, HydroFloat™, REFLUX ™ flotation benchmarking","authors":"Richel Annan Dadzie , Massimiliano Zanin , William Skinner , Richmond Asamoah , Jonas Addai-Mensah , George Blankson Abaka-Wood","doi":"10.1016/j.cherd.2026.01.025","DOIUrl":"10.1016/j.cherd.2026.01.025","url":null,"abstract":"<div><div>Reprocessing copper flotation tailings is limited by particle-size effects and mineralogical challenges, especially poor liberation. Preliminary studies aimed at recovering copper minerals from complex, low-grade rougher flotation tailings have shown limited success, with values within the intermediate particle size range (-150 +53 µm) failing in conventional mechanical flotation cells. To improve copper recovery, the current study employs a flowsheet that uses Hydrofloat ™ fluidised-bed flotation on the deslimed + 53 µm fraction, considers the REFLUX ™ Flotation Cell (RFC) for the (-53µm) slimes, and compares performance against a Denver mechanical cell. Flotation performance (recovery, grade, size-by-size) was combined with data from Quantitative Evaluation of Minerals by Scanning Electron Microscopy (QEMSCAN), including liberation and locking statistics, to explain the flotation response observed. At similar mass pull (29.3–33 %), Hydrofloat ™ achieved 61.8 % Cu recovery at 0.74 % Cu (upgrade ratio 3.44), outperforming the mechanical flotation cell (44. 6 % Cu at 0. 10 % Cu grade; upgrade ratio 0.6) when processing the deslimed feed, rather than the whole (unsplit) tailings feed. Preliminary RFC tests on the - 53 µm stream maintained high recoveries of fine particles (often >80 %) with improved Cu concentrate grades, aligning with the presence of well-liberated chalcopyrite in slimes. Overall, the results support a split-flotation process for complex copper low grade ores. Although liberation ultimately limits recovery this integrated method significantly enhances copper recovery and upgrade relative to conventional mechanical cells, offering a practical route to unlock value from low-grade sulphide tailings.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"227 ","pages":"Pages 35-45"},"PeriodicalIF":3.9,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-15DOI: 10.1016/j.cherd.2026.01.035
Shida Gao , Cuimei Bo , Guo Yu , Quanling Zhang , Furong Gao , Genke Yang , Jian Chu
Ethylene glycol (EG) serves as a primary raw material in the polyester industry, with syngas-to-dimethyl oxalate (DMO) conversion representing an advanced EG production method. However, this process encounters conflicting objectives between maximization of economic benefits and minimization of carbon emissions, particularly exacerbated by constraints and market prices. To address this challenge, we developed a multi-objective optimization framework for various working conditions: First, we establish a steady-state simulation system incorporating reaction kinetics and mechanisms to model the DMO synthesis process. Then, an innovative economy-carbon emission multi-objective optimization problem is formulated, where the ranges of pivotal operating parameters are determined by sensitivity analysis, and the response surface method is used to obtain the reference points under different conditions. Finally, the optimization problem is solved by the Pareto frontier (PF) estimation algorithm to solve the irregular PF problem, which arises from the complex nonlinear interactions between process variables under various working and price conditions. Under regular working conditions, we compare the knee point among the obtained Pareto solution set with the reference point, and the framework reduces carbon emissions by 19.63% (129.5 kmol/h) while increasing economic benefits by 1.38% (1253.1 yuan/h). Considering three typical conditions of sharp increase of DMC prices, limited production capacity and short-term negative profits, our framework identifies solutions that dominate the reference points and the original turning points in the obtained PF. The results have verified that this study is able to support the decision-making in providing solutions with a good balance between economy and carbon emissions under various working and price conditions.
{"title":"Multi-objective optimization for sustainable dimethyl oxalate synthesis: A plant-wide framework balancing economic benefits and carbon emissions","authors":"Shida Gao , Cuimei Bo , Guo Yu , Quanling Zhang , Furong Gao , Genke Yang , Jian Chu","doi":"10.1016/j.cherd.2026.01.035","DOIUrl":"10.1016/j.cherd.2026.01.035","url":null,"abstract":"<div><div>Ethylene glycol (EG) serves as a primary raw material in the polyester industry, with syngas-to-dimethyl oxalate (DMO) conversion representing an advanced EG production method. However, this process encounters conflicting objectives between maximization of economic benefits and minimization of carbon emissions, particularly exacerbated by constraints and market prices. To address this challenge, we developed a multi-objective optimization framework for various working conditions: First, we establish a steady-state simulation system incorporating reaction kinetics and mechanisms to model the DMO synthesis process. Then, an innovative economy-carbon emission multi-objective optimization problem is formulated, where the ranges of pivotal operating parameters are determined by sensitivity analysis, and the response surface method is used to obtain the reference points under different conditions. Finally, the optimization problem is solved by the Pareto frontier (PF) estimation algorithm to solve the irregular PF problem, which arises from the complex nonlinear interactions between process variables under various working and price conditions. Under regular working conditions, we compare the knee point among the obtained Pareto solution set with the reference point, and the framework reduces carbon emissions by 19.63% (129.5 kmol/h) while increasing economic benefits by 1.38% (1253.1 yuan/h). Considering three typical conditions of sharp increase of DMC prices, limited production capacity and short-term negative profits, our framework identifies solutions that dominate the reference points and the original turning points in the obtained PF. The results have verified that this study is able to support the decision-making in providing solutions with a good balance between economy and carbon emissions under various working and price conditions.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"226 ","pages":"Pages 536-549"},"PeriodicalIF":3.9,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2026-01-15DOI: 10.1016/j.cherd.2026.01.022
Ping Gu , Yongmin Zhang , Hui Du
Insufficient filtration performance had long been a critical issue for industrial slurry bed reactors, which hindered plant capacity improvement and increased production costs. To address this, this study experimentally investigated in-vessel filtration characteristics with gas leakage considered, aiming to develop solutions. Gas leakage flux (Jg,l) was originally defined to characterize gas leakage effects, and pressure variation-based method was proposed to measure liquid filtration flux (J) and its average (Javg) for quantifying filtration performance. Results showed that when superficial gas velocity (ug) increased from 0.027 to 0.063 m/s, Jg,l increased by 684.4 %, impairing filtrate outflow and reducing Javg by 44 %. When particle concentration (Cw) increased from 5 % to 15 %, Javg and Jg,l decreased by 74.6 % and 69 %, with Javg dropping more sharply. Javg for 30 μm filter tubes was 325.93 L/(m2·h), which was much lower than for 50 μm (496.98 L/(m2·h)) and 80 μm (568.65 L/(m2·h)), indicating excessively high ug, high Cw, and smaller filter tube pore size (df) as key causes. Increasing df from 30 to 80 μm boosted Javg by 74.5 %, while increasing Jg,l by 178.8 %. Filtrate analysis showed low particle concentrations and no substantial long-term particle loss, confirming reasonably increasing df as a feasible solution for in-vessel filtration performance enhancement. This study identified the root causes of insufficient filtration performance and proposed a feasible solution, which could serve as a viable reference for addressing analogous issues and thus held considerable academic and engineering significance.
{"title":"Experimental study on in-vessel filtration characteristics of Fischer-Tropsch slurry bed reactors considering influence of gas leakage","authors":"Ping Gu , Yongmin Zhang , Hui Du","doi":"10.1016/j.cherd.2026.01.022","DOIUrl":"10.1016/j.cherd.2026.01.022","url":null,"abstract":"<div><div>Insufficient filtration performance had long been a critical issue for industrial slurry bed reactors, which hindered plant capacity improvement and increased production costs. To address this, this study experimentally investigated in-vessel filtration characteristics with gas leakage considered, aiming to develop solutions. Gas leakage flux (<em>J</em><sub><em>g</em>,<em>l</em></sub>) was originally defined to characterize gas leakage effects, and pressure variation-based method was proposed to measure liquid filtration flux (<em>J</em>) and its average (<em>J</em><sub><em>avg</em></sub>) for quantifying filtration performance. Results showed that when superficial gas velocity (<em>u</em><sub><em>g</em></sub>) increased from 0.027 to 0.063 m/s, <em>J</em><sub><em>g</em>,<em>l</em></sub> increased by 684.4 %, impairing filtrate outflow and reducing <em>J</em><sub><em>avg</em></sub> by 44 %. When particle concentration (<em>C</em><sub><em>w</em></sub>) increased from 5 % to 15 %, <em>J</em><sub><em>avg</em></sub> and <em>J</em><sub><em>g</em>,<em>l</em></sub> decreased by 74.6 % and 69 %, with <em>J</em><sub><em>avg</em></sub> dropping more sharply. <em>J</em><sub><em>avg</em></sub> for 30 μm filter tubes was 325.93 L/(m<sup>2</sup>·h), which was much lower than for 50 μm (496.98 L/(m<sup>2</sup>·h)) and 80 μm (568.65 L/(m<sup>2</sup>·h)), indicating excessively high <em>u</em><sub><em>g</em></sub>, high <em>C</em><sub><em>w</em></sub>, and smaller filter tube pore size (<em>d</em><sub><em>f</em></sub>) as key causes. Increasing <em>d</em><sub><em>f</em></sub> from 30 to 80 μm boosted <em>J</em><sub><em>avg</em></sub> by 74.5 %, while increasing <em>J</em><sub><em>g</em>,<em>l</em></sub> by 178.8 %. Filtrate analysis showed low particle concentrations and no substantial long-term particle loss, confirming reasonably increasing <em>d</em><sub><em>f</em></sub> as a feasible solution for in-vessel filtration performance enhancement. This study identified the root causes of insufficient filtration performance and proposed a feasible solution, which could serve as a viable reference for addressing analogous issues and thus held considerable academic and engineering significance.</div></div>","PeriodicalId":10019,"journal":{"name":"Chemical Engineering Research & Design","volume":"226 ","pages":"Pages 589-596"},"PeriodicalIF":3.9,"publicationDate":"2026-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146034284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号