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.

Highly Selective Electrochemical Nitrate to Ammonia Conversion by Dispersed Ru in a Multielement Alloy Catalyst.

Electrochemical reduction of nitrate to ammonia (NH3) converts an environmental pollutant to a critical nutrient. However, current electrochemical nitrate reduction operations based on monometallic and bimetallic catalysts are limited in NH3 selectivity and catalyst stability, especially in acidic environments. Meanwhile, catalysts with dispersed active sites generally exhibit a higher atomic utilization and distinct activity. Herein, we report a multielement alloy nanoparticle catalyst with dispersed Ru (Ru-MEA) with other synergistic components (Cu, Pd, Pt). Density functional theory elucidated the synergy effect of Ru-MEA than Ru, where a better reactivity (NH3 partial current density of -50.8 mA cm-2) and high NH3 faradaic efficiency (93.5%) is achieved in industrially relevant acidic wastewater. In addition, the Ru-MEA catalyst showed good stability (e.g., 19.0% decay in FENH3 in three hours). This work provides a potential systematic and efficient catalyst discovery process that integrates a data-guided catalyst design and novel catalyst synthesis for a range of applications.

Machine Learning in Environmental Research: Common Pitfalls and Best Practices.

Machine learning (ML) is increasingly used in environmental research to process large data sets and decipher complex relationships between system variables. However, due to the lack of familiarity and methodological rigor, inadequate ML studies may lead to spurious conclusions. In this study, we synthesized literature analysis with our own experience and provided a tutorial-like compilation of common pitfalls along with best practice guidelines for environmental ML research. We identified more than 30 key items and provided evidence-based data analysis based on 148 highly cited research articles to exhibit the misconceptions of terminologies, proper sample size and feature size, data enrichment and feature selection, randomness assessment, data leakage management, data splitting, method selection and comparison, model optimization and evaluation, and model explainability and causality. By analyzing good examples on supervised learning and reference modeling paradigms, we hope to help researchers adopt more rigorous data preprocessing and model development standards for more accurate, robust, and practicable model uses in environmental research and applications.

Ca-Based Layered Double Hydroxides for Environmentally Sustainable Carbon Capture.

The process of carbon dioxide capture typically requires a large amount of energy for the separation of carbon dioxide from other gases, which has been a major barrier to the widespread deployment of carbon capture technologies. Innovation of carbon dioxide adsorbents is herein vital for the attainment of a sustainable carbon capture process. In this study, we investigated the electrified synthesis and rejuvenation of calcium-based layered double hydroxides (Ca-based LDHs) as solid adsorbents for CO2. We discovered that the particle morphology and phase purity of the LDHs, along with the presence of secondary phases, can be controlled by tuning the current density during electrodeposition on a porous carbon substrate. The change in phase composition during carbonation and calcination was investigated to unveil the effect of different intercalated anions on the surface basicity and thermal stability of Ca-based LDHs. By decoupling the adsorption of water and CO2, we showed that the adsorbed water largely promoted CO2 adsorption, most likely through a sequential dissolution and reaction pathway. A carbon capture capacity of 4.3 ± 0.5 mmol/g was measured at 30 °C and relative humidity of 40% using 10 vol % CO2 in nitrogen as the feed stream. After CO2 capture occurred, the thermal regeneration step was carried out by directly passing an electric current through the conductive carbon substrate, known as the Joule-heating effect. CO2 was found to start desorbing from the Ca-based LDHs at a temperature as low as 220 °C as opposed to the temperature above 700 °C required for calcium carbonate that forms as part of the Ca-looping capture process. Finally, we evaluated the cumulative energy demand and environmental impact of the LDH-based capture process using a life cycle assessment. We identified the most environmentally concerning step in the process and concluded that the postcombustion CO2 capture using LDH could be advantageous compared with existing technologies.

Methane Emissions from Municipal Wastewater Collection and Treatment Systems.

Municipal wastewater collection and treatment systems are critical infrastructures, and they are also identified as major sources of anthropogenic CH4 emissions that contribute to climate change. The actual CH4 emissions at the plant- or regional level vary greatly due to site-specific conditions as well as high seasonal and diurnal variations. Here, we conducted the first quantitative analysis of CH4 emissions from different types of sewers and water resource recovery facilities (WRRFs). We examined variations in CH4 emissions associated with methods applied in different monitoring campaigns, and identified main CH4 sources and sinks to facilitate carbon emission reduction efforts in the wastewater sector. We found plant-wide CH4 emissions vary by orders of magnitude, from 0.01 to 110 g CH4/m3 with high emissions associated with plants equipped with anaerobic digestion or stabilization ponds. Rising mains show higher dissolved CH4 concentrations than gravity sewers when transporting similar raw sewage under similar environmental conditions, but the latter dominates most collection systems around the world. Using the updated data sets, we estimated annual CH4 emission from the U.S. centralized, municipal wastewater treatment to be approximately 10.9 ± 7.0 MMT CO2-eq/year, which is about twice as the IPCC (2019) Tier 2 estimates (4.3-6.1 MMT CO2-eq/year). Given CH4 emission control will play a crucial role in achieving net zero carbon goals by the midcentury, more studies are needed to profile and mitigate CH4 emissions from the wastewater sector.

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.

Membrane Distillation-Crystallization for Sustainable Carbon Utilization and Storage.

Anthropogenic greenhouse gas emissions from power plants can be limited using postcombustion carbon dioxide capture by amine-based solvents. However, sustainable strategies for the simultaneous utilization and storage of carbon dioxide are limited. In this study, membrane distillation-crystallization is used to facilitate the controllable production of carbonate minerals directly from carbon dioxide-loaded amine solutions and waste materials such as fly ash residues and waste brines from desalination. To identify the most suitable conditions for carbon mineralization, we vary the membrane type, operating conditions, and system configuration. Feed solutions with 30 wt % monoethanolamine are loaded with 5-15% CO2 and heated to 40-50 °C before being dosed with 0.18 M Ca2+ and Mg2+. Membranes with lower surface energy and greater roughness are found to more rapidly promote mineralization due to up to 20% greater vapor flux. Lower operating temperature improves membrane wetting tolerance by 96.2% but simultaneously reduces crystal growth rate by 48.3%. Sweeping gas membrane distillation demonstrates a 71.6% reduction in the mineralization rate and a marginal improvement (37.5%) on membrane wetting tolerance. Mineral identity and growth characteristics are presented, and the analysis is extended to explore the potential improvements for carbon mineralization as well as the feasibility of future implementation.

Underestimation of Sector-Wide Methane Emissions from United States Wastewater Treatment.

An increasing percentage of US waste methane (CH4) emissions come from wastewater treatment (10% in 1990 to 14% in 2019), although there are limited measurements across the sector, leading to large uncertainties in current inventories. We conducted the largest study of CH4 emissions from US wastewater treatment, measuring 63 plants with average daily flows ranging from 4.2 × 10-4 to 8.5 m3 s-1 (<0.1 to 193 MGD), totaling 2% of the 62.5 billion gallons treated, nationally. We employed Bayesian inference to quantify facility-integrated emission rates with a mobile laboratory approach (1165 cross-plume transects). The median plant-averaged emission rate was 1.1 g CH4 s-1 (0.1-21.6 g CH4 s-1; 10th/90th percentiles; mean 7.9 g CH4 s-1), and the median emission factor was 3.4 × 10-2 g CH4 (g influent 5 day biochemical oxygen demand; BOD5)-1 [0.6-9.9 × 10-2 g CH4 (g BOD5)-1; 10th/90th percentiles; mean 5.7 × 10-2 g CH4 (g BOD5)-1]. Using a Monte Carlo-based scaling of measured emission factors, emissions from US centrally treated domestic wastewater are 1.9 (95% CI: 1.5-2.4) times greater than the current US EPA inventory (bias of 5.4 MMT CO2-eq). With increasing urbanization and centralized treatment, efforts to identify and mitigate CH4 emissions are needed.

High-Efficiency Recovery of Acetic Acid from Water Using Electroactive Gas-Stripping Membranes.

Recovery of carbon-based resources from waste is a critical need for achieving carbon neutrality and reducing fossil carbon extraction. We demonstrate a new approach for extracting volatile fatty acids (VFAs) using a multifunctional direct heated and pH swing membrane contactor. The membrane is a multilayer laminate composed of a carbon fiber (CF) bound to a hydrophobic membrane and sealed with a layer of polydimethylsiloxane (PDMS); this CF is used as a resistive heater to provide a thermal driving force for PDMS that, while a highly hydrophobic material, is known for its ability to rapidly pass gases, including water vapor. The transport mechanism for gas transport involves the diffusion of molecules through the free volume of the polymer matrix. CF coated with polyaniline (PANI) is used as an anode to induce an acidic pH swing at the interface between the membrane and water, which can protonate the VFA molecule. The innovative multilayer membrane used in this study has successfully demonstrated a highly efficient recovery of VFAs by simultaneously combining pH swing and joule heating. This novel technique has revealed a new concept in the field of VFA recovery, offering promising prospects for further advancements in this area. The energy consumption was 3.37 kWh/kg for acetic acid (AA), and an excellent separation factor of AA/water of 51.55 ± 2.11 was obtained with high AA fluxes of 51.00 ± 0.82 g.m-2hr-1. The interfacial electrochemical reactions enable the extraction of VFAs without the need for bulk temperature and pH modification.

Simple and Low-Cost Electroactive Membranes for Ammonia Recovery.

Ammonia is considered a contaminant to be removed from wastewater. However, ammonia is a valuable commodity chemical used as the primary feedstock for fertilizer manufacturing. Here we describe a simple and low-cost ammonia gas stripping membrane capable of recovering ammonia from wastewater. The material is composed of an electrically conducting porous carbon cloth coupled to a porous hydrophobic polypropylene support, that together form an electrically conductive membrane (ECM). When a cathodic potential is applied to the ECM surface, hydroxide ions are produced at the water-ECM interface, which transforms ammonium ions into higher-volatility ammonia that is stripped across the hydrophobic membrane material using an acid-stripping solution. The simple structure, low cost, and easy fabrication process make the ECM an attractive material for ammonia recovery from dilute aqueous streams, such as wastewater. When paired with an anode and immersed into a reactor containing synthetic wastewater (with an acid-stripping solution providing the driving force for ammonia transport), the ECM achieved an ammonia flux of 141.3 ± 14.0 g.cm-2.day-1 at a current density of 6.25 mA.cm-2 (69.2 ± 5.3 kg(NH3-N)/kWh). It was found that the ammonia flux was sensitive to the current density and acid circulation rate.

Upscaling 3D Engineered Trees for Off-Grid Desalination.

More than 70% of the population without access to safe drinking water lives in remote and off-grid areas. Inspired by natural plant transpiration, we designed and tested in this study an array of scalable three-dimensional (3D) engineered trees made of natural wood for continuous water desalination to provide affordable and clean drinking water. The trees took advantage of capillary action in the wood xylems and lifted water more than 1 foot off the ground with or without solar irradiation. This process overcame some major challenges of popular solar-driven water evaporation and water harvesting, such as intermittent operation, low water production rate, and system scaling. The trade-off between energy transfer and system footprint was tackled by optimizing the interspacing between the trees. The scaled system has a ratio of surface area (vapor generation) to project area (water transport) up to 118, significantly higher than the prevailing flat-sheet design. The extensive surface area evaporated water at a temperature cooler than the surrounding air, drawing on multiple environmental energy sources including solar, wind, or ambient heat in the air and realized continuous operation. The total energy for evaporation reached over 300% of the one-sun irradiance, enabling a freshwater production rate of 4.8 L m-2 h-1 from an array of 16 trees in an enclosed room and 14 L m-2 h-1 under a 3 m/s airflow. Furthermore, we found that the ambient heat in the air contributed 60%-70% of the total latent heat of vaporization when energy sources were decoupled. During long-term desalination tests, the engineered trees demonstrated a self-cleaning mechanism with daily cycles of salt accumulation and dissolution. Combining the quantification from an evaporation model and meteorology data covering the globe, we also demonstrated that the 3D engineered trees can be of particular interest for sustainable desalination in the Middle East and North Africa (MENA) regions.

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.

Photochemical Behavior of Microbial Extracellular Polymeric Substances in the Aquatic Environment.

Microbially derived extracellular polymeric substances (EPSs) occupy a large portion of dissolved organic matter (DOM) in surface waters, but the understanding of the photochemical behaviors of EPS is still very limited. In this study, the photochemical characteristics of EPS from different microbial sources (Shewanella oneidensis, Escherichia coli, and sewage sludge flocs) were investigated in terms of the production of reactive species (RS), such as triplet intermediates (3EPS*), hydroxyl radicals (•OH), and singlet oxygen (1O2). The steady-state concentrations of •OH, 3EPS*, and 1O2 varied in the ranges of 2.55-8.73 × 10-17, 3.01-4.56 × 10-15, and 2.08-2.66 × 10-13 M, respectively, which were within the range reported for DOM from other sources. The steady-state concentrations of RS varied among different EPS isolates due to the diversity of their composition. A strong photochemical degradation of the protein-like components in EPS isolates was identified by excitation emission matrix fluorescence with parallel factor analysis, but relatively, humic-like components remained stable. Fourier-transform ion cyclotron resonance mass spectrometry further revealed that the aliphatic portion of EPS was resistant to irradiation, while other portions with lower H/C ratios and higher O/C ratios were more susceptible to photolysis, leading to the phototransformation of EPS to higher saturation and lower aromaticity. With the phototransformation of EPS, the RS derived from EPS could effectively promote the degradation of antibiotic tetracycline. The findings of this study provide new insights into the photoinduced self-evolution of EPS and the interrelated photochemical fate of contaminants in the aquatic environment.

Machine Learning: New Ideas and Tools in Environmental Science and Engineering.

The rapid increase in both the quantity and complexity of data that are being generated daily in the field of environmental science and engineering (ESE) demands accompanied advancement in data analytics. Advanced data analysis approaches, such as machine learning (ML), have become indispensable tools for revealing hidden patterns or deducing correlations for which conventional analytical methods face limitations or challenges. However, ML concepts and practices have not been widely utilized by researchers in ESE. This feature explores the potential of ML to revolutionize data analysis and modeling in the ESE field, and covers the essential knowledge needed for such applications. First, we use five examples to illustrate how ML addresses complex ESE problems. We then summarize four major types of applications of ML in ESE: making predictions; extracting feature importance; detecting anomalies; and discovering new materials or chemicals. Next, we introduce the essential knowledge required and current shortcomings in ML applications in ESE, with a focus on three important but often overlooked components when applying ML: correct model development, proper model interpretation, and sound applicability analysis. Finally, we discuss challenges and future opportunities in the application of ML tools in ESE to highlight the potential of ML in this field.

pH Dependence of Phosphorus Speciation and Transport in Flow-Electrode Capacitive Deionization.

Electrochemical processes such as capacitive deionization have shown great promise for salt removal and nutrient recovery, but their effectiveness on phosphate removal was lower than other charged ions. This study hypothesized that the speciation and transport behaviors of phosphate ions are highly influenced by electrolyte pH, and it used experimental and modeling approaches to elucidate such impacts in flow-electrode capacitive deionization (FCDI) cells. Phosphate removal was investigated in either constant current (CC) or constant voltage (CV) charging mode with pH ranged from 5 to 9 in the feed solution. Results showed that the average P removal rate increased from 20.8 (CC mode) and 16.8 mg/min (CV mode) at pH 9 to 38.3 (CC mode) and 34.3 mg/min (CV mode) at pH 5 (84-104% in improvement), respectively. Correspondingly, the energy consumption reduced from 1.04 kWh/kg P at pH 9 to 0.59 kWh/kg P at pH 5 (42.9-56.1% in saving). Such benefits were attributed to the shift in dominant P-species from HPO42- to H2PO4-. Conversely, high-electrolyte pH (pH = 11) for flow-electrode led to ∼74.8% higher phosphate recovery during discharge compared with pH 5, which was associated with the higher distribution of phosphate ions in the electrolyte versus on the flow-electrodes due to surface charge change. These results improved our understanding in ion distribution and migration and indicate that solution pH is critical for operating FCDI reactors. It shed lights on the best practices on electrochemical phosphate removal and recovery.

Molecular Transformation of Crude Oil Contaminated Soil after Bioelectrochemical Degradation Revealed by FT-ICR Mass Spectrometry.

Bioremediation is a low-cost approach for crude oil spill remediation, but it is often limited by electron acceptor availability. In addition, the biodegradation products of crude oil contaminants are complex, and transformation pathways are difficult to decipher. This study demonstrates that bioelectrochemical systems (BESs) can be effective in crude oil degradation by integrating biological and electrochemical pathways, and more importantly, it provides the first understanding on the daughter products of bioelectrochemical hydrocarbon degradation. Using electrospray ionization (ESI) Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) and two-dimensional gas chromatography (GC × GC), the results showed that the active BES reactor improved the total petroleum hydrocarbon (TPH) degradation by ∼70% than open circuit control reactors. After separating the daughter products into nine fractions (MA1-MA9) according to the molecular weight (m/z 200-1000) by modified aminopropyl silica (MAPS) fractionation, we found that active BES remediation resulted in 50% more polar, oxygen-containing naphthenic (NAP) acids. The MA4 fraction (centered at ∼550 Da) increased by 47%, and MA5 and MA7 fractions with higher molucular weight increased by a maximum of ∼7- and 9-fold, respectively. These results are in accordance with the variation of bulk elemental compositions in O2 species, where daughter transformation products doubled relative to parent oil extract. The contribution of newly generated NAP acids was mainly from higher-order oxygen species (O5-O6) with increased hydrophobicity in conjunction with a decreased abundance in lower-order oxygen species (O1). Overall, the study suggests that n-alkane degradation occurred via ß-oxidation to oxygenated transformation products with lower molecular weight, such as n-alcohols in O1 class and subsequently to n-fatty acids in O2 class.

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.

Autotrophic Vanadium(V) Bioreduction in Groundwater by Elemental Sulfur and Zerovalent Iron.

Vanadium (V) is an emerging contaminant in groundwater that can adversely affect human health. Although bioremediation has been shown effective, little is known on autotrophic V(V) bioreduction in the context of oligotrophic characteristics of groundwater. In this study, we demonstrate that efficient V(V) bioreductions can be coupled with bio-oxidation of elemental sulfur (S(0)) or zerovalent iron (Fe(0)), and the V(V) removal efficiencies reached 97.5 ± 1.2% and 86.6 ± 2.5% within 120 h using S(0) and Fe(0), respectively. V(IV) is the main reduction product and precipitates naturally in near-neutral conditions. Microbial community, functional gene, and metabolites analyses reveal that synthetic metabolisms among autotrophs and heterotrophs played major roles in V(V) reduction using S(0) and Fe(0). These results demonstrate a new approach for V(V) contaminated groundwater remediation.

Influences of Temperature and Metal on Subcritical Hydrothermal Liquefaction of Hyperaccumulator: Implications for the Recycling of Hazardous Hyperaccumulators.

Waste Sedum plumbizincicola, a zinc (Zn) hyperaccumulator during phytoremediation, was recycled via a subcritical hydrothermal liquefaction (HTL) reaction into multiple streams of products, including hydrochar, bio-oil, and carboxylic acids. Results show approximately 90% of Zn was released from the S. plumbizincicola biomass during HTL at an optimized temperature of 220 °C, and the release risk was mitigated via HTL reaction for hydrochar production. The low-Zn hydrochar (∼200 mg/kg compared to original plant of 1558 mg/kg) was further upgraded into porous carbon (PC) with high porosity (930 m2/g) and excellent capability of carbon dioxide (CO2) capture (3 mmol/g). The porosity, micropore structure, and graphitization degree of PCs were manipulated by the thermal recalcitrance of hydrochar. More importantly, results showed that the released Zn2+ could effectively promote the production of acetic acid via the oxidation of furfural (FF) and 5-(hydroxymethyl)-furfural (HMF). Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS) with negative electrospray ionization analysis confirmed the deoxygenation and depolymerization reactions and the production of long chain fatty acids during HTL reaction of S. plumbizincicola. This work provides a new path for the recycling of waste hyperaccumulator biomass into value-added products.