AIChE Journal最新文献

CO₂ solubility in deep eutectic solvent (DES) is a critical property to indicate the competence of using DESs for carbon capture. This work presents a systematic screening framework with computational and experimental procedures to identify the promising ammonium-based DESs for CO₂ absorption. First, 1556 CO₂ solubility data points are collected to regress UNIFAC-DES interaction parameters to extend the applicability of UNIFAC in DESs. Meanwhile, the data are utilized to calibrate the COSMO-RS model for enhancing accuracy. Then, 3192 hypothetical DESs are generated by rules. Their CO₂ solubilities are predicted using the two newly obtained models to produce two lists of top candidates. The melting points and viscosities of the candidates that appear in the two lists repeatedly are estimated and finally four ethylenecyanohydrin-based DESs are identified. With two DESs verified via experiment, their superiority over the best reference DES is proved. Their absorption mechanisms are analyzed through quantum chemistry calculations.

Nanofiltration (NF) membranes with sub-nano structure are promising for the extraction of lithium from brine resources, yet often face a trade-off between separation efficiency and permeation rate. Here, the ultrathin nanocomposite membranes (TFN_AM) with high Li⁺/Mg²⁺ selectivity and permeability are prepared by introducing the positive aminated-MXene interlayer in the process of interfacial polymerization (IP). This interlayer modulates the diffusion behavior of amine monomers and reduces the thickness of the polyamide (PA) layer by six times; meantime, it provides excellent positive charge on the membrane surface. The aminated-MXene interlayer endows the high positive charge density of TFN_AM enhancing the Donnan exclusion effect on the improvement of Li⁺/Mg²⁺ selectivity. As a result, the TFN_AM exhibits a high permeability of 12.5 L m⁻² h⁻¹ bar⁻¹ and a Li⁺/Mg²⁺ selectivity of 17.5, while maintaining performance over 100 h. This work proposes a facile and effective strategy to address the challenge of NF membrane limitations and advance lithium recovery technologies from brines.

Robust aluminum-based metal–organic framework (Al-MOF) nanoparticles with ordered nanochannels represent promising fillers for mixed-matrix membranes (MMMs) toward challenging liquid separations. Achieving both high separation factor and permeation flux, however, requires not only precise pore engineering but also reliable interfacial compatibility under complex feed conditions. Herein, we demonstrate that fluoroalkylsilane (FAS) coating concurrently enhances the processability of Al-MOF (CAU-23) nanoparticles in PIM-1 and tunes their pore apertures. Through this synergy, the FAS-anchored CAU-23 (FCAU-23) evolve from passive fillers into active nanochannel directors, establishing efficient, low-resistance pathways for linear n-hexane (n-C6) while selectively discriminating against branched isomers. The optimized 1-FCAU-23-10/PIM-1 MMM achieves a record n-C6 permeation flux of 1.45 kg m⁻² h⁻¹ with a separation factor (SF) of 5.1 in binary feeds. More importantly, it maintains robust performance under quinary feeds, yielding permeants enriched in linear n-C6 and mono-branched species (88 wt.%). This work highlights filler/interface engineering as a route to robust isomer separations.

Amine blend is a viable method to formulate superior absorbents for CO₂ capture from flue gas. However, the understanding of the degradation of blended amines, especially the interaction effect on the degradation pathway is limited, which restricts the solvent development. In this work, the oxidative degradation of 1-amino-2-propanol (1AP) and its blends with various tertiary amines is studied by analyzing amine loss, alkalinity loss, viscosity change, NH₃ emission, and heat-stable salt generation. Experimental results show that both triethanolamine and N-methyldiethanolamine have an excellent inhibition effect on 1AP oxidative degradation, reducing 1AP loss from 47.0% to 6.5% and 8.5%, respectively. The reaction pathways for the oxidative degradation of single and blended 1AP solutions are proposed and evaluated by density functional theory calculations. Computational results show that the reactions between 1AP peroxide radicals and peroxide radicals of tertiary amines reduce 1AP loss, which agrees with experimental results.

This paper proposes an offset-free model predictive control (MPC) framework for nonlinear systems modeled using neural network-based nonlinear autoregressive models with exogenous inputs (NARX). To address plant-model mismatch and ensure offset-free tracking, the NARX model is augmented with an integrating disturbance model, resulting in an extended state-space suitable for MPC. A nonlinear observer is developed to estimate both system and disturbance states in real time. The impact of training data quality on control performance is examined through two modeling scenarios: one with rich excitation data and another with limited excitation data, reflecting practical constraints. For both cases, offset-free MPC controllers are designed using the proposed framework. The approach is validated through simulations on a nonlinear chemical reactor and compared with a benchmark NARX-based offset-free MPC method employing bias correction from output prediction errors. Results show that the proposed method improves tracking performance, particularly when training data are limited.

Helical liquid-bridge flow, a novel semi-constrained flow, shows significant potential for enhancing gas-liquid mass transfer and reactions. Research confirms millimeter-scale spiral bridges offer high efficiency and low pressure drop in distillation, absorption, and reaction, while micrometer-scale flow enables controllable gas-liquid contact in microchemical engineering. However, behavioral differences across scales remain unclear, and quantitative models for spiral design and flow prediction are lacking. This study systematically investigated flow characteristics and morphology evolution in millimeter and micrometer liquid bridges. Then, a multi-parameter flow model was established based on the experiments. In addition, the consistency was improved from 0.825 to 0.996 by minimizing the surface energy optimization model. The model can accurately predict key parameters such as the liquid bridge profile, providing a reliable theoretical basis and effective methods for the selection and optimization of helical structures in gas-liquid mass transfer and reaction processes at different scales.

Carbon molecular sieve (CMS) membranes hold significant potential for advanced gas separation and purification, yet further improvements in molecular sieving properties remain a critical challenge. In this study, carboxylated-PIM-1 (PIM-COOH) polymers were synthesized by hydrolyzing PIM-1 polymers, serving as efficient precursors for CMS membrane fabrication. The pyrolysis process was optimized, with the temperature tuned to 800°C for 2 h, resulting in high-performance CMS membranes. These membranes demonstrated exceptional gas separation capabilities, achieving H₂ and O₂ permeabilities of 831.8 and 79.7 Barrer, respectively, with H₂/CH₄ and O₂/N₂ selectivities of 453.6 and 12.8, surpassing the latest 2015 upper bound for H₂/CH₄ and O₂/N₂ separation. Mixed gas tests further validated the single-gas results, revealing H₂/CH₄ and O₂/N₂ selectivities as high as 1402.1 and 16.8, respectively. This innovative strategy, leveraging PIM-COOH-derived CMS membranes, provides a promising pathway for next-generation hydrogen purification and air separation technologies.