Machine Learning for Polymer Design to Enhance Pervaporation-Based Organic Recovery.

Pervaporation (PV) is an effective membrane separation process for organic dehydration, recovery, and upgrading. However, it is crucial to improve membrane materials beyond the current permeability-selectivity trade-off. In this research, we introduce machine learning (ML) models to identify high-potential polymers, greatly improving the efficiency and reducing cost compared to conventional trial-and-error approach. We utilized the largest PV data set to date and incorporated polymer fingerprints and features, including membrane structure, operating conditions, and solute properties. Dimensionality reduction, missing data treatment, seed randomness, and data leakage management were employed to ensure model robustness. The optimized LightGBM models achieved RMSE of 0.447 and 0.360 for separation factor and total flux, respectively (logarithmic scale). Screening approximately 1 million hypothetical polymers with ML models resulted in identifying polymers with a predicted permeation separation index >30 and synthetic accessibility score <3.7 for acetic acid extraction. This study demonstrates the promise of ML to accelerate tailored membrane designs.

Predicting Extraction Selectivity of Acetic Acid in Pervaporation by Machine Learning Models with Data Leakage Management.

The extraction of acetic acid and other carboxylic acids from water is an emerging separation need as they are increasingly produced from waste organics and CO2 during carbon valorization. However, the traditional experimental approach can be slow and expensive, and machine learning (ML) may provide new insights and guidance in membrane development for organic acid extraction. In this study, we collected extensive literature data and developed the first ML models for predicting separation factors between acetic acid and water in pervaporation with polymers' properties, membrane morphology, fabrication parameters, and operating conditions. Importantly, we assessed seed randomness and data leakage problems during model development, which have been overlooked in ML studies but will result in over-optimistic results and misinterpreted variable importance. With proper data leakage management, we established a robust model and achieved a root-mean-square error of 0.515 using the CatBoost regression model. In addition, the prediction model was interpreted to elucidate the variables' importance, where the mass ratio was the topmost significant variable in predicting separation factors. In addition, polymers' concentration and membranes' effective area contributed to information leakage. These results demonstrate ML models' advances in membrane design and fabrication and the importance of vigorous model validation.

ES&T in the 21st Century: A Data-Driven Analysis of Research Topics, Interconnections, And Trends in the Past 20 Years.

Environmental Science & Technology (ES&T) has served a leadership role in reporting advanced and significant research findings for decades and accumulated tremendous amount of high-quality literature. In this study, we developed tailored text mining methods and analyzed 29 188 papers published in ES&T from 2000 to 2019, and we performed data-driven analyses to reveal some critical information and guidance on what has been published, what topical changes have evolved, and what are the areas that deserve additional attention. While top research keywords remained stable (water, sorption, soil, emiss, oxid, exposur), the trending up and emerging keywords showed clear shift over the years. Keywords related to nanobased materials (nanoparticl, nanomateri, carbon nanotub), climate and energy (climat, ch4, greenhouse gas emiss, mitig, energi), and health (exposur, health, ingest) demonstrated the strongest uptrend in the past 10 years, while plastics and PFAS were among clear emerging topics in the past 5 years. Co-occurrence analysis showed distinct associations between media (water, soil, air, sediment), chemicals (pcb, humic subst, particulate matt), processes (sorption, remov, degrad), and properties (kinet, mechan, speciat). Furthermore, a rule-based classification deciphered trends, distributions, and interconnections of articles based on either monodomains (air, soil, solid waste, water, and wastewater) or multidomains. It found water and wastewater cross-discipline articles tended to have higher citation values, while air domain tended to stand alone. Water and air monodomains consistently increased their shares in publications (together 56.3% in 2019), while shares of soil studies gradually declined. This study provides new data-driven methods on literature mining and offers unique insights on environmental research landscape and opportunities.

Hydrophobic Gas Transfer Membranes for Wastewater Treatment and Resource Recovery.

Gaseous compounds, such as CH4, H2, and O2, are commonly produced or consumed during wastewater treatment. Traditionally, these gases need to be removed or delivered using gas sparging or liquid heating, which can be energy intensive with low efficiency. Hydrophobic membranes are being increasingly investigated in wastewater treatment and resource recovery. This is because these semipermeable barriers repel water and create a three-phase interface that enhances mass transfer and chemical conversions. This Critical Review provides a first comprehensive analysis of different hydrophobic membranes and processes, and identifies the challenges and potential for future system development. The discussions and analyses were grouped based on mechanisms and applications, including membrane gas extraction, membrane gas delivery, and hybrid processes. Major challenges, such as membrane fouling, wetting, and limited selectivity and functionality, are identified, and potential solutions articulated. New opportunities, such as electrochemical coating, integrated membrane electrodes, and membrane functionalization, are also discussed to provide insights for further development of more efficient and low-cost membranes and processes.

Membrane electrolysis distillation for volatile fatty acids extraction from pH-neutral fermented wastewater.

Volatile fatty acids (VFAs) serve as building blocks for a wide range of chemicals, but it is difficult to extract VFAs from pH-neutral wastewater using evaporation methods because of the ionized form. This study presents a new membrane electrolysis distillation (MED) process that extracts VFAs from such fermentation solutions. MED uniquely integrates pH regulation and joule heating to facilitate the efficient evaporation of VFAs. This integration occurs alongside a hydrophobic membrane that ensures effective gas-liquid phase separation. Operating solely on electricity, MED achieved an acid flux rate of 12.03 g/m2/h at 6V. In contrast, the control results without the joule heating or pH swing only obtained a 0.23 g/m2/h and 0.32 g/m2/h flux, respectively. In addition, a physicochemical model was developed to assess the impacts of temperature on membrane surface pH. This system enhances resource recovery from waste streams and helps achieve a circular carbon economy.

Microbial electrochemical nutrient recovery in anaerobic osmotic membrane bioreactors.

This study demonstrates that by incorporating a microbial electrochemical unit into an anaerobic osmotic membrane bioreactor (AnOMBR), the system addressed several challenges faced by traditional anaerobic membrane bioreactors and recovered biogas, nitrogen, and phosphorus while maintaining high effluent quality with low dissolved methane. The microbial recovery cell (MRC)-AnOMBR system showed excellent organic (>93%) and phosphorus removal (>99%) and maintained effluent COD below 20 mg/L. Furthermore, the reactor effectively recovered up to 65% PO43- and 45% NH4+ from the influent, which can be further improved if membranes with higher selectivity are used. Nutrients removal from bulk solution mitigated NH4+ penetration to the draw solution and reduced scaling potential caused by PO43-. The maximum methane yield was 0.19 L CH4/g COD, and low methane (<2.5 mL CH4/L) was detected in the effluent. Further improvement can be made by increasing charge efficiency for better nutrient and energy recovery.

Microbial fuel cells and osmotic membrane bioreactors have mutual benefits for wastewater treatment and energy production.

This study demonstrates that microbial fuel cells (MFCs) and osmotic membrane bioreactors (OMBRs) can be mutually beneficial when integrated together for wastewater treatment. When connecting MFCs with OMBRs, the solute buildup increased conductivity and buffer capacity, which greatly increased MFC power density from 3 W/m(3) up to 11.5 W/m(3). In turn, the MFCs conditioned and reduced sludge production and therefore reduced forward osmosis (FO) membrane fouling. The MFC-OMBR equipped with new thin-film composite (TFC) membrane showed excellent organic (>95%) and phosphorus removal (>99%) and therefore maintained effluent sCOD below 20 mg/L. However, the nitrogen removal was limited due to the negative surface charge of the thin-film composite membrane and solution chemistry, which led to higher flux of ammonium toward the OMBR draw solution. Further studies are needed to improve nitrogen removal, reduce fouling, and optimize system integration.