The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent of the COVID-19 pandemic. Computer simulations of complete viral particles can provide theoretical insights into large-scale viral processes including assembly, budding, egress, entry, and fusion. Detailed atomistic simulations, however, are constrained to shorter timescales and require nearly billion-atom simulations for these processes. In my talk I will present the current status and on-going development of a largely “bottom-up” coarse-grained (CG) model of the SARS-CoV-2 virion. Structural data from a combination of cryo-electron microscopy (cryo-EM), x-ray crystallography, and computational predictions have been used to build molecular models of the structural SARS-CoV-2 proteins, which were then assembled into a complete CG virion model. I will describe how CG molecular interactions were derived from all-atom simulations, how viral behavior difficult to capture in atomistic simulations was incorporated into the CG models, and how the CG models are iteratively improved as new data becomes publicly available. The initial CG model and the detailed methods behind it are intended to serve as a resource for researchers working on COVID-19 who are interested in performing multiscale simulations of the SARS-CoV-2 virion. Recent updates to this multiscale model will also be a focus of my talk.
Enveloped viruses, such as SARS-CoV-2, infect cells via fusion of their envelope with the host membrane. By employing molecular simulations to characterize viral envelopes, researchers can gain insights into key determinants of infection. Here, petascale supercomputers are leveraged for large-scale modeling and analysis of authentic viral envelopes, whose lipid compositions are complex and realistic. Visual Molecular Dynamics (VMD) with support for MPI is employed, overcoming previous computational limitations and enabling investigation into virus biology at an unprecedented scale. The techniques applied here to an authentic SARS-CoV-2 envelope at two levels of spatial resolution (29 million particles and 280 million atoms) are broadly applicable to the study of other viruses. A general framework for carrying out scalable analysis of simulation trajectories on petascale supercomputers is presented, expanding the utility of the machine in humanity's ongoing fight against infectious diseases.
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an enveloped, positive-sense, single-stranded RNA virus that has emerged in Wuhan, China, at the end of 2019, causing the ongoing COVID-19 pandemic. Protruding from the SARS-CoV-2 spherical membrane bilayer is the spike (S), a homotrimeric fusion glycoprotein that recognizes and latches onto the angiotensin-converting enzyme 2 (ACE2) to promote host cell infection. Similar to other viral proteins, the SARS-CoV-2 S protein is cloaked by a sugary shield made of 22 N-glycans and two O-glycans per protomer. In the context of viral pathogenesis, SARS-CoV-2 S protein’s extensive glycosylation plays a critical role as it is used to thwart and evade the host immune response. Here, building on the available cryo-EM structures of the spike in the open and closed states, we used in-silico approaches to construct a full-length model of the glycosylated SARS-CoV-2 S protein, allowing for an atomic-level understanding and annotation of the roles of glycans. All-atom molecular dynamics simulations revealed that, despite covering a vast amount of the S protein surface area, the glycan shield changes depending on the conformational state (open or closed) of the protein. Moreover, beyond shielding, two N-linked glycans at positions N165 and N234 modulate the conformational dynamics of the spike’s receptor binding domain (RBD), which is responsible for ACE2 recognition. As corroborated by biolayer interferometry experiments, these two glycans are found to "load-and-lock" the RBD in the up conformation, priming the spike for infection. Overall, our results highlight the importance of glycans in the evasion of the host immune response and in the modulation of the RBD dynamics. Our work sheds light on the full structure of this critical target and points out opportunities, strategies and challenges for vaccine design and drug development.
The SARS-CoV-2 virus enters human cells after interacting with two proteins on the host's cell surface, ACE2 and the serine protease TMPRSS2. Here we exploit and develop machine learning (ML) and molecular simulation methods to better understand and improve TMPRSS2 inhibitors that may inhibit SARS-CoV-2 cell entry and therefore contribute to early-stage Covid-19 therapeutics. We exploit machine-learning methods for enhanced sampling and Markov modeling in order to generate an equilibrium ensemble of the TMPRSS2 protease, and use this ensemble to understand the molecular mechanisms of the already confirmed inhibitors Nafamostat and Camostat. We also develop an ML-driven drug search and optimization framework in collaboration with our experimental partners.
Infection of human cells by the SARS-CoV2 relies on its binding to a specific receptor and subsequent fusion of the viral and host-cell membranes. The fusion peptide (FP), a short peptide segment in the spike protein, plays a central role in this process by inserting into the host cell membrane, thereby initiating the fusion of the two membranes. Here, we use a large array of all-atom molecular dynamics (MD) simulations using our accelerated membrane model, the Highly Mobile Membrane Mimetic (HMMM), to investigate the interaction of the SARS-CoV2 FP with a lipid bilayer representing mammalian cellular membranes, and to characterize the membrane-inserted form of the peptide. Several independent initial systems with diverse initial positioning and orientation of the FP with respect to the membrane and randomized lipid mixings were generated. Each system was simulated as multiple independent replicas. The majority of the simulations capture stable membrane binding of the FP. Clustering of the results reveals three major membrane binding modes, from which one is the most representative binding mode. The resulting membrane-bound mode shows a deep penetration of the FP into the membrane accompanied by extension of the helical structure of the peptide. This mode substantiates the results of biochemical studies on amino acids involved in the interaction of the FP with cellular membranes. Taken together, the results shed light on a key step involved in SARS-CoV2 infection with potential implications for designing novel inhibitors.
Membrane-bound form of SARS-CoV2 fusion peptide to human cellular membrane
Neuropilin-1 (Nrp1) was recently identified as a host factor for SARS-CoV-2 entry. As the Spike protein (S protein) of SARS-Cov-2 is cleaved into S1 and S2 domain by furin protease, Nrp1 binds to the newly created C-terminal RRAR segment of the S1 domain (Science 370(6518):856; Science 370(6518):861). Beyond the very S1 C-terminal peptide, it is at present unknown at the atomistic detail how Nrp1 binds to the Spike protein and more importantly what the mechanistic consequences of this interaction are. Here, we study the association of a Nrp1 (a2-b1-b2) protein with the Spike protein via molecular modeling and dynamics simulation (bioRxiv doi: https://doi.org/10.1101/2021.01.06.425627). We predict the possible binding modes between Nrp1 and Spike protein, and investigate the co-binding models of ACE2/Nrp1 with Spike protein. Importantly, by using the pulling molecular dynamics simulation, we explore the exit mechanism of S2 from the S1 domain with the assistance of Nrp1. We find that the Nrp1 could stimulate faster separation of Spike S1 and S2 domain. This would lead to enhanced membrane fusion via the separated S2 domain. The study is insightful for understanding the molecular mechanism of virus infection, especially for a better appreciation of the role of the newly discovered virus co-receptor—Nrp1, in facilitating cell infection.
PAMAM dendrimer is the most studied and well exploited class of commercially available dendrimers. PAMAM is described as a carrier in the biomedical field either in conjugation with or encapsulation of drugs, biomolecules or ions. In this work, we discuss on the design, virtual screening and synthesis of a guanylthiourea-PAMAM library and its evaluation as furin inhibitors. Zero generation ethylene diamine cored PAMAM dendrimer (G0 PAMAM) has a positive surface and act as a source of multi-basic functionality due to the presence of amine and amide units. Furin is a proprotein convertase enzyme responsible for the activation of spike (S) protein of SARS-CoV-2 which is attributing to the high spreading rate of SARS-CoV-2 among human. Presence of a greater number of Asp and Glu residues at the catalytic site of furin demands multi-basic ligands as substrates. Hence, we designed a library of guanyl thiourea conjugates of G0 PAMAM which could be prepared easily from commercially available reagents. Molecular docking studies revealed that similar to m-guanidinomethyl-Phac-RVR-Amba, strongest furin inhibitor reported till date, all the derivatives along with G0 PAMAM could mask the substrate binding site of furin and the distributions of different arms of dendrimer conjugates into the furin pockets were influenced by the substituent at the guanylthiourea unit. Molecular dynamics studies (100ns) revealed that some of the conjugates bind to the catalytic triad (Asp153, His194 and Ser368) essential for the furin catalytic activity, thereby making it unavailable for the S protein cleavage. All the selected candidates showed higher binding free energy (MM/GBSA method) towards furin catalytic site. Synthesis of the designed guanylthiourea-G0 PAMAM conjugates was achieved in a single step reaction in our laboratory and the details will be presented.
3D depiction of PAMAM conjugates at the catalytic site of furin
We present a computational study of the structure and dynamics of the SARS-CoV-2 envelope protein (protein E, a viroporin) in the monomeric form. The protein consists of three parts: α-helical transmembrane domain (TMD) and amphiphilic α-helices H2 and H3, which are connected by flexible linkers. We show that TMD is inclined in the membrane, with phenylalanines Phe20, Phe23 and Phe26 facing the lumen. H2 and H3 reside at the membrane border. Orientation of H2 is not affected by glycosylation, but strongly influenced by palmitoylation pattern of cysteines Cys40, Cys43 and Cys44. The protein induces curvature in the membrane, which can be ascribed to the cytoplasmic part (H2 and H3). The protein is also curvature-sensitive, preferably localizing with the cytoplasmic part at the convex regions of the membrane. Such localization may be favorable for assembly of the protein E oligomers, whereas induction of curvature may facilitate budding of the viral particles. We hope that the presented results will be helpful for understanding the function of coronaviral protein E and viroporins in general, and for overcoming the ongoing SARS-CoV-2 pandemic.
Structure of the SARS-CoV-2 envelope (E) protein monomer. Schematic model showing a fully palmitoylated and glycosylated protein E and conformations of the protein E observed in coarse-grained simulations.
Track: [CATL] Division of Catalysis Science & Technology
Division/Committee: [CATL] Division of Catalysis Science & Technology
The Chemical Catalysis for Bioenergy (ChemCatBio) consortium is an R&D consortium dedicated to identifying and overcoming catalysis challenges for biomass conversion processes. Led by U.S. Dept. of Energy national laboratories, ChemCatBio seeks to accelerate the development of catalysts and related technologies for the commercialization of biomass-derived and waste-derived fuels and chemicals, leading to enhanced energy security and national leadership in the global bioeconomy. This symposium will foster engagement with industry and academia to tackle challenges spanning from foundational science to applied engineering, leverage our unique capabilities, and collaboratively generate new advanced technologies.
Track: [CATL] Division of Catalysis Science & Technology
Improvements to current catalyst technologies are required to produce biofuels at a cost that is competitive with fossil fuels. Understanding the physical and chemical properties that control catalyst performance (i.e. activity, selectivity, and durability) is critical information needed to develop synthetic methods for producing better-performing catalysts. X-ray absorption spectroscopy (XAS) is a catalyst characterization technique that can provide insight into physical and chemical properties such as geometric and/or electronic structure, respectively, of a catalyst that influence its performance. As an enabling technology of the Chemical Catalysis for Bioenergy Consortium (ChemCatBio), XAS is being employed to accelerate catalyst development for a number of process technologies including catalytic fast pyrolysis and catalytic upgrading of C1 and C2 alcohols to produce gasoline and diesel/jet fuels. Combining the results from XAS with results from other characterization techniques, such as electron microscopy and infrared/Raman spectroscopy, coupled with advanced synthesis methods is helping to facilitate the development of next generation catalysts. In this presentation, we will present results from XAS studies on beta zeolite-supported metal catalysts for the conversion of methanol to high-octane gasoline (collaboration with NREL) and ethanol to C3-C6 olefins (collaboration with ORNL) and discuss how these results are enabling the development of better catalysts. A brief perspective will also be given on potential new applications of advanced XAS techniques for accelerating catalyst development.
Track: [CATL] Division of Catalysis Science & Technology
Conversion of methanol and dimethyl ether to high-octane gasoline catalyzed by beta zeolite (BEA) provides an opportunity for the production of high-quality fuels from renewable carbon sources (e.g., gasified biomass). Recent research demonstrated that a Cu-modified BEA zeolite catalyst (Cu/BEA) offered advantages over the unmodified BEA catalyst due to multifunctional Cu species that enabled incorporation of co-fed H2, reactivation of light alkanes, and reduction of products from the aromatic hydrocarbon pool. The shift in hydrocarbon pool chemistry has the potential to influence the identity and relative composition of surface carbon species that are often linked to deactivation. A detailed understanding of these carbon species is important to develop an effective and efficient regeneration procedure that can enable the transition from fundamental catalyst development to commercial application. Here, we applied complementary ex situ and in situ characterization techniques to compare the structures of surface carbon species on post-reaction Cu/BEA and unmodified BEA catalysts. Both catalysts contained acyclic and aromatic hydrocarbons along with graphitic carbon species. However, the post-reaction Cu/BEA catalyst had a lower polycyclic aromatic content, and further, the graphitic species were more hydrogenated and defective. It was also found that the presence of Cu promoted carbon removal at lower temperatures than for unmodified BEA through activation of O2 by Cu during thermal oxidation. The fundamental insight into the composition of surface carbon species enabled the design of an effective and efficient regeneration strategy for the DME homologation reaction over Cu/BEA, resulting in full recovery of the catalyst activity.
Track: [CATL] Division of Catalysis Science & Technology
Typical petrochemical processes in the production of conventional jet fuel lead to a product rich in aromatics and lights. Improvements to the jet fuel properties such as the sooting index and seal swelling properties can be achieved by replacing this aromatic fraction with cycloalkane compounds. Insights on how to cost-effectively produce cycloalkane rich fuel from waste streams will influence the future of the sustainable aviation fuel. We have developed a selective pathway to produce sustainable, cycloalkane-rich jet fuels through renewable ethanol derived from process waste streams. Ethanol can be effectively transformed into the longer chain ketones through a ketonization reaction at high selectivities and conversion over a Pd-promoted ZnO-ZrO2catalyst. This ketone mix is subsequently cyclized to produce branched cyclohexenones ranging from C9 to C15, with the specific composition being controlled by changes in the feed composition and reaction conditions. Effective hydrodeoxygenation and hydrogenation yields a jet-fuel range product rich in branched cycloalkanes appropriate for fuel blending. Catalyst and the process development, characterization of fuel properties based on ASTM standards as well as technoeconomic analysis of this process will be presented.
Track: [CATL] Division of Catalysis Science & Technology
Ethanol is a promising platform molecule for the production of a variety of fuels and chemicals. The ethanol “blend wall,” coupled with advancements in production efficiency and feedstock diversification will potentially lead to excess ethanol, at competitive prices, make it an attractive feedstock. This creates an opportunity for ethanol producers to diversify their product offerings, but processes for producing fuels and chemicals from ethanol are currently lacking. We have developed a metal-supported ZrO2/SiO2 catalyst system with specially tailored metal and Lewis acid sites useful for producing either 1,3-butadiene (BD) or n-butene directly from ethanol, and with excellent activity, selectivity, and stability. BD is a valuable building block currently produced in the petrochemical industry with an annual global market size of 12 MMT. While demand for BD is steadily increasing its supply is expected to decrease as ethylene manufactures shift to lighter feedstocks and produce less BD byproduct. Butene-rich olefins are precursors used for producing renewable jet and diesel fuel blendstock. Producing butene-rich olefins directly from ethanol enables an improvement relative to the state-of-the-art alcohol-to-jet process, as it eliminates one unit operation while simultaneously offering the potential for energy savings. In our lab we have demonstrated the direct production of BD from ethanol with 99% conversion and 71% selectivity, when ethanol is co-fed with N2. Whereas with H2 co-feed an 88% selectivity to butene-rich olefins was achieved and with 99% conversion. Markedly stable production of n-butene was also demonstrated for 100+ hours’ time-on-stream. Additionally, successful demonstration with real fermentation-derived ethanol feedstock was performed, thus illustrating the potential for commercial adaption. Catalyst characterizations were performed in order to discern structure-function relationships. XPS indicated that the product composition varies depending on metal oxidation state. Reactivity measurements coupled with operando NMR experiments and computational modeling provided insight into the reaction mechanism and structure sensitivity of the catalyst. Ultimately, the catalyst system reported here shows promise in enabling renewable production of BD or n-butene-rich olefins from ethanol.
Track: [CATL] Division of Catalysis Science & Technology
The identification of catalytic pathways for generating renewable fuels and fuel additives from biomass-derived feedstocks is of critical importance for increasingly carbon-neutral operations and the improved economic viability of processes for renewable feedstock valorization. Here, we present our latest work on designing catalyst combinations consisting of multifunctional bimetallic Zn-Y/Beta zeolites and “single-atom alloy” (SAA) Pt-Cu supported metal catalysts for the thermocatalytic upgrading of ethanol into butene-rich olefins as precursors to jet-range (C8-16) hydrocarbons. Ethanol conversion to 1,3-butadiene and subsequent hydrogenation to butene isomers occurs through a complex reaction network consisting of dehydrogenation, aldol condensation, dehydration, and selective hydrogenation reactions where each reaction step requires unique active site requirements for efficient conversion. The development of multifunctional catalysts, accomplished here through the incorporation of multiple transition metals (e.g. Zn, Y, Cu, Pt) onto a single support (e.g. Beta, Al2O3), is investigated using a suite of in situ and ex situ characterizations (e.g. EXAFS, HAADF-STEM, pyridine DRIFTS) to identify Lewis and Brønsted acid sites. These site configurations and mixed metal center identities are correlated to specific reaction steps at 588 K and are evaluated at both high (>95%) and low (<10%) ethanol conversions. Reaction testing at high conversions indicates improved ethanol valorization into butene-rich olefin streams (65% butenes, 78% C3+ olefin selectivity at 94% conversion) while avoiding further hydrogenation to form butanes or other saturated hydrocarbons. This is ethanol-to-olefins (ETO) reaction network is accomplished without cofed hydrogen over a SAA Pt-Cu catalyst capable of catalyzing butadiene hydrogenation at stoichiometric hydrogen and butadiene partial pressures.
Track: [CATL] Division of Catalysis Science & Technology
Abundant and low-cost shale gas has replaced naphtha as the feedstock of choice for C2-C4 olefin production. This change has led to a shortage of 1,3-butadiene (BD), a critical intermediate for the manufacture of synthetic rubber. The constrained BD supply has triggered price fluctuations and interest in on-purpose BD production. Cellulosic ethanol is a sustainable feedstock quickly becoming mainstream and its conversion into BD significantly reduces lifecycle greenhouse gas emissions when compared to petroleum-derived BD. Supported ZnO-ZrO2/SiO2 catalysts show particular promise for the one-pot conversion of ethanol to BD, yet little is known about the atomic structures giving rise to catalytic activity. We report how insights about catalytic activity and selectivity, as probed by temperature-programmed surface reaction spectroscopy, are made possible via in situ X-ray absorption, Raman, and UV-vis spectroscopies. These new fundamental insights point towards structure-activity/selectivity relationships to guide the rational design of improved catalyst performance.
Urban stormwater runoff often contains potentially harmful contaminants, including metals, which can create public health and environmental concerns. Bioretention systems are increasingly being utilized to treat contaminated urban stormwater; however, bioretention media should be carefully selected. In this study, batch experiments were conducted to investigate five low-cost sorbent materials (biochar (BC), iron amended biochar (FeBC), scrap tire (ST), coir coconut (CC) and blast furnace slag (BFS)) potential to treat six common toxic metal contaminants (Pb, Cr, Cd, Cu, Ni and Zn). The effects of pH, ionic strength and dissolved organic carbon (DOC) of solutions on the sorption of metals were also investigated. CC and BFS had great removal efficiencies for all evaluated metals (>90%), while removal efficiencies of ST and BC were high (>90%) for some of metals (Cr, Pb and Cu). FeBC showed the worst performance removing only Pb and Cr (30-50%). Kinetic experiments showed that metals removal occurred rapidly and reached equilibrium < 5 h and experimental data were well described by a pseudo-second order model. Sorption of metals on the surface of sorbents was strongly dependent on solution pH, as sorption was highest at pH=9 and lowest at pH=3, which is likely due to a combination of H+ competition and more charged metal species. In addition, sorption was negatively affected by DOC, potentially due to increased metal solubility or site blocking. Ionic strength did not show significant effect on sorption performance. The sorption mechanism of metals on the surface of sorbents was mainly due to electrostatic attraction and precipitation. Vegetated intermittent flow column studies are currently underway to evaluate metals removal in stormwater infiltration relevant systems. Overall, this study demonstrates strengths and limitations of each sorbent, appropriate selection of sorbent as filter media for removing metals from runoff, and on-going testing.
Mining-impacted waters contain unsafe levels of toxic elements like arsenic, zinc, cadmium, and lead. Such metals and metalloids can be incorporated in the solid phase via inclusion of foreign ions in the crystalline structure (coprecipitation). In this work, we focus on coprecipitation for simultaneous heavy metal(loid) and CO2 removal, motivated by naturally observed and engineered process of carbonation in mine tailings for the purpose of water treatment. This approach can be considered effective if it can be applicable to commonly found heavy metals, is permanent, and mitigates risk of rerelease into the environment. Therefore, quantifying the degree of uptake, and mapping the path of elements is critical since mineral precipitates of different crystalline structure or composition exhibit varying degrees of reactivity and stability. Carbonates, along with sulfates and oxides were precipitated in a series of synthetic mine water-based experiments to explore elemental distributions and uptake mechanisms. ICP-MS data of the treated solution indicated As and Zn to be most effectively taken up in the solid phase, but Cd and Pb concentrations in the treated solution still reached levels below U.S. EPA national primary drinking water standards. Experiments conducted at acidic conditions showed abundance of calcium sulfate (gypsum) precipitates, and S/TEM/EDS analyses showed homogeneous distributions of the four elements, strongly suggesting incorporation of the cations and/or anion in the crystalline structure. Experiments conducted at near neutral to basic conditions revealed metal(loid)-rich calcites and Fe-oxides, and stabilization of vaterites, a phase that is commonly overlooked due to its metastability. Results from this work are especially useful as it provides evidence for coprecipitated metal(loid)s in the form of stable carbonate and/or commonly found secondary minerals. We show that with pH adjustments and CO2 addition, carbonate precipitation can be promoted utilizing existing cations in mine waters for effective water treatment.
Approximately 33% of roadways in the United States are unpaved, and these surfaces account for more than 35% of all particulate matter (<10 µm, PM10) emissions in the United States. The U.S. Environmental Protection Agency regulates PM10 and the smaller PM2.5 (<2.5 µm) due to the risks posed to the human respiratory system as well as vegetative and aquatic health. In order to mitigate fugitive dust emissions, dust suppressants are applied to unpaved roads, either causing fine particles to aggregate together or creating a physical barrier between the roadbed surface and vehicle impact. A variety of dust suppressants are commonly used, including water, non-petroleum organic material (vegetable oils), petroleum products (solvents, tars), synthetic polymers, or most commonly, salts and brines. Despite the availability of commercial products, their cost, which can range from $4,000 to $10,000 per mile per application, can be prohibitive in areas operating with small road maintenance budgets. To reduce costs while limiting the release of fine particulate matter, some states allow oil and gas wastewater from conventional operations (COGWW) to be spread on unpaved roads. These brines are often characterized by high TDS, primarily sodium, chloride, calcium, and magnesium ions, the latter two being particularly important for dust suppression. However, other contaminants in these wastes, such as radium and toxic heavy metals, have been shown to be mobile in roadbed aggregate and could therefore pose risks to adjacent water resources. These risks have motivated some states to place moratoriums on using COGWW as a dust suppressant. The goal of this study is to compare runoff chemistry from a model road treated with six dust suppressants during a simulated rainfall event. The chosen dust suppressants include: synthetic rainwater, treated and untreated COGWW, treated COGWW dosed with radium, a calcium chloride brine, and soybean oil. Runoff samples were periodically collected over a 24 hour period and evaluated for specific conductance, salt anions, major cations, trace metals, organic composition, and radioisotopes.
Wildfire occurrence and intensity are increasing worldwide due to climate change. With increasing burning, destruction of wildland-urban interface communities may cause contamination of surrounding waterways by ash and debris from burned structures, cars, and buildings. However, the effects of burned urban residues in surface water are not well understood in terms of contaminants of emerging concerns. In this study, we collected stormwater samples for one year following the November 2018 Camp Fire, the most destructive fire in California history with near 18,000 structures and thousands of vehicles burnt. We comprehensively characterized organic contaminant profiles and dynamics via targeted quantification of 35 stormwater-derived chemicals and complementary HRMS suspect screening. For quantified analytes, pentachlorophenol, a wood preservative pesticide and a likely human carcinogen, dominated the chemical profiles with concentrations up to 2400 ng/L and detection frequency of 89%. In one sampling site close to a burnt pharmacy, 6400 ng/L of ibuprofen and 1500 ng/L of caffeine was detected in November 2018, and suspect screening revealed acetaminophen and lidocaine presence in the same sample with rare detections of these pharmaceuticals in other locations. Suspect screening also revealed high levels of perfluorinated compounds in samples collected in November 2018 compared with those collected in the following year. These results highlighted the impact of urban burning on the release of contaminants of emerging concerns into surrounding watersheds, and identified potential threats to human and ecological health.
The natural aquatic environment plays an important role in the emergence and spread of antibiotic resistance. Recent work has highlighted that water environments receive inputs of antibiotic compounds from wastewater treatment plants (WWTPs), livestock operations, aquaculture, and industry, which may contribute to selection pressure for elevation of resistance levels in native bacteria, and in turn, they could also serve as sinks, reservoirs, and sources of antibiotic resistant bacteria and resistance genes of clinical concern. In this context, this project is undertaken to investigate occurrences, concentrations, and geospatial distribution of antibiotics across surface water environment in Minnesota State (US). Water samples were collected during July-October 2020 at a total of 39 surface water sites from lakes and creeks in or near the Minneapolis/St. Paul (MSP) metropolitan area. They include 30 lakes in (sub)urban areas variably impacted by anthropogenic activities, and one lake with limited impact (acting as a reference to assess the background levels of antibiotics), and 4 creeks (for some, multiple samples were taken along the creek to evaluate small-scale geospatial distributions). Concentrations of 25 antibiotics in different classes (including sulfonamides, macrolides, tetracyclines, fluoroquinolones, beta-lactams, and other unclassified compounds) were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Preliminary results showed that 23 out of the 25 antibiotics were detected in the 39 lake/creek samples, with detection frequencies of the classified antibiotics declining in the order of sulfonamides (95%) > macrolides (79%) > beta-lactams (67%) > tetracyclines (28%) > fluoroquinolones (26%). In general, total antibiotic concentrations were found higher at sites in more urban regions, with several hotspots of certain antibiotics identified in the MSP area. Ongoing work is being undertaken to (1) investigate partitioning of the target antibiotics between the water columns and sediments in the selected lakes, and (2) explore relations of antibiotic concentrations with land use and hydrogeologic condition and identify primary sources of antibiotic input due to human activities. Outcomes of this work will be applied to develop an “antibiotic footprint” map that enables predictions of antibiotic geospatial distribution in a statewide scale and aid in mitigation of antibiotic resistance in the natural water environment.
Surface runoff from wildfire-affected areas carries ash, black carbon particulates, and pollutants to downgradient surface waters, potentially negatively affecting ecosystems and water quality. Yet, the factors that affect the surface water quality or the extent to which water quality deteriorates after a wildfire is unclear. We analyzed 43 studies from worldwide to evaluate how the burned area, surface water flow rates, and sampling time affect the peak concentration of different types of pollutants. Results reveal that contaminant concentration in surface waters after a wildfire can increase by two orders of magnitude, but the peak concentration and its arrival time depend on pollutant type. Increases in wildfire burned area disproportionally increase total suspended solids (TSS) concentration in downstream surface waters. Conversely, increases in surface water flow up to 10 m3 s-1 increase TSS concentration and further increases in flow decrease in TSS concentration potentially due to dilution or exhaustion of TSS source. The peak concentration for suspended solids and nutrients appears immediately after a wildfire, whereas the peak for heavy metals such as Cu, Pb, and Zn delayed by a year or longer, indicating the slow dissolution of heavy metals from fire-affected areas or debris. Polyaromatic hydrocarbon peaked within a year after wildfire but the peaks did not exceed EPA limits. Collectively, this analysis reveals that post-wildfire risk to downstream aquatic systems is influenced by wildfire burn area, flow rate, and pollutant type. The results could inform management efforts to protect downstream areas and their ecosystem functions.
For decades, targeted gas chromatography-mass spectrometry (GC-MS) methods have been used to carefully acquire measurement data in support of risk-based assessments of volatile and semi-volatile chemicals. While targeted measurements of known chemicals have supported most exposure studies, recent discoveries of new chemicals in diverse media samples have driven a shift towards broader-scope non-targeted analysis (NTA) methods. To date, NTA methods have largely produced qualitative chemical screening results with little focus on quantitative interpretations. For NTA results to be most useful in a risk-based context, multi-step methods must be developed to estimate chemical concentrations in prepared solution (i.e., sample extracts) and ultimately in the original sampled media. Furthermore, needs exist for statistically defensible error-bounding methods if NTA data are to be considered in risk-based decisions. Here, we illustrate a proof-of-concept risk-based prioritization using a mixture of 66 volatile and semi-volatile chemicals commonly found in drinking water, extracted from spiked activated carbon filters, and analyzed using GC high-resolution mass spectrometry (HRMS). Semi-quantitative concentration estimates of all 66 chemicals were determined using a regression-based model and surrogate response factors. Statistically defensible error bounds taken from a normalized response factor distribution were applied to prepared solution concentration estimates. Media concentrations were derived from the solution estimates using percent recovery from carbon filter extract data, and compared to existing regulatory levels as a means of preliminary prioritization. This research serves as a model to focus larger NTA datasets on priority chemical lists for further targeted analysis and risk assessment. The views expressed are those of the author(s) and do not necessarily reflect the views or policies of the US EPA.
In this study, we used molecular images as a representation for organic compounds and combined them with a convolutional neural network (CNN) to develop quantitative structure-activity relationships (QSARs) for predicting compound rate constants toward OH radicals. We applied transfer learning and data augmentation to train molecular image-CNN models and the Gradient-weighted Class Activation Mapping (Grad-CAM) method to interpret them. Results showed that data augmentation and transfer learning can effectively enhance the robustness and predictive performance of the models, with the root-mean-square-error (RMSE) values on the test dataset (RMSEtest) decreasing from (0.395–0.45) to (0.284–0.339) after applying data augmentation, and the RMSE on the training dataset (RMSEtrain) decreasing from (0.452–0.592) to (0.123–0.151) after applying transfer learning. The obtained molecular image-CNN models showed comparative predictive performance (RMSEtest 0.284–0.339) with the molecular fingerprint-based models (RMSEtest 0.30–0.35). Grad-CAM interpretation showed that the molecular image-CNN models correctly chose the molecular features in the images and identified key functional groups that influenced the reactivity. The applicability domain analysis showed that the molecular image-CNN models have a broader applicability domain than molecular fingerprints-based models and the reactivity of any new compounds with a maximum similarity of over 0.85 to the compounds in the training dataset can be reliably predicted. This study demonstrated that molecular image-CNN is a new tool to develop QSARs for environmental applications and can be used to build trustful models that make meaningful predictions.
A useful reference for understanding molecular processes occurring at mineral-water interfaces is the bulk liquid, away from the interface. Halide ions pose a challenge to molecular investigations due to competing hydrogen bonds between neighboring water molecules versus water-anion interactions. This situation leads to competing structures and anharmonic vibrations for simple clusters of two or more waters interacting directly with an anion (H2O)nX-. Specific properties of the anions, such as their Hofmeister ranking and selective adsorption to mineral-water interfaces, depends in part on subtle differences in cluster structures and vibrational modes of motion. Here we apply density functional theory molecular dynamics and quasi-chemical free energy theory to compare hydrogen-bond structures, energies, and vibrational modes of clusters of the first four halide ions with water. Our analysis provides new insights about which structural, dynamic, and thermodynamic properties can be used to differentiate these anions, which may be useful for designing materials for selective anion absorption.
Ion adsorption at mineral-water interfaces plays a significant role in geochemical processes such as mineral growth and dissolution and particle aggregation. Here, we use barite as a model mineral to understand divalent cation interactions with ionic crystals. The coverage and location of either lead (Pb) or strontium (Sr) sorbed to the (001) and (210) surfaces were studied as a function of [Pb]aq or [Sr]aq using in situ specular x-ray reflectivity (XR) and resonant anomalous x-ray reflectivity (RAXR). In the presence of Pb, the XR signals become progressively more distorted as [Pb]aq increases, implying that the presence of Pb alters the structure of the two surfaces. In contrast, the presence of Sr has smaller effects on the XR. The interfacial structural changes imply Pb and Sr sorption and that the extent of Pb sorption may be greater than that of Sr sorption.
The specific Sr or Pb sorption behavior at the (001) and (210) surfaces was measured using RAXR. At both surfaces, Sr and Pb incorporate into the topmost barite layer and adsorb as an inner-sphere complex with a small outer-sphere fraction. Roughly half the sorbed Sr is incorporated and the other half is adsorbed, regardless of concentration. However, a larger amount of Sr sorbs to the (001) as compared with the (210). Pb is similarly partitioned half incorporated and half adsorbed at both surfaces at [Pb]aq ≤ 75 μM. At [Pb]aq ≥ 200-225 μM, less than a third of the Pb sorbed to the (001) is incorporated, but up to three-quarters of Pb sorbed to the (210) is incorporated. As with Sr, Pb coverage is greater at the (001) as compared with the (210). These results suggest that the (001) has a higher sorption capacity than the (210), despite sorption occurring through simultaneous incorporation and adsorption at both surfaces.
Ion sorption processes are important in understanding and predicting the fate of toxic metal ions in the environment. Here, we studied desorption of lead (Pb) ions at the barite (001) -water interface using in situ x-ray reflectivity (XR) and resonant anomalous x-ray reflectivity (RAXR) measurements. Previous results on barite (001) surfaces reveal complex Pb sorption behavior. These results indicate that Pb ions adsorb to and incorporate into the barite (001) surface at ~0-5 Å above the surface and ~0-2.5 Å into the crystal. Increased sorption correlates with increasing [Pb]. While the sorption behavior of Pb at barite is known, it is essential to study desorption mechanisms at the surface, which will enable prediction of the potential release of this toxic metal.
In situ XR and RAXR measurements were conducted on barite samples pre-reacted with Pb using a transmission cell. After introducing a Pb-free barite saturated solution into the cell, the amplitude of the RAXR signal decreases, which corresponds to a decrease in the Pb coverage. After 0.5 hours of reaction time, ~50% of the sorbed Pb species desorbs. Pb continues to desorb over time, but even after reacting for 12 hours, 31% of the originally sorbed Pb still remains. The phase of the RAXR signals, which is proportional to the height of sorbed Pb, decreases with desorption. This indicates the adsorbed species are preferentially removed and the incorporated species are resistant to desorption. Together, these results suggest that while incorporation of Pb is a relatively fast process, exchange of Ba for incorporated Pb is slow. The incorporation of Pb within the barite surface leads to additional stabilization of Pb, making barite an potential host mineral for Pb sequestration.
Uranium and lead are two of the most important contaminants of concern in natural waters in Earth’s critical zone. In order to create detailed experimental and theoretical models of molecular interactions between these cations and sediment-water interfaces, it is crucial to understand important modalities and key characteristics of complex natural solid-solution interfaces. Historically, model systems have employed well-defined iron or aluminum oxide minerals, in the presence or absence of simplified organic molecules or humic or fulvic acids. Yet, reactive solid surfaces in aquifers and reservoirs may be decorated with other types of organic matter such as microbial biomass, or may be non-mineral particulate organic matter, may exhibit nano-scale heterogeneity with inorganic and organic domains, or may be modified through reaction with water and solutes. It is helpful to consider studies of natural systems to gain a better understanding of aqueous-metal ion interactions at these complex natural surfaces and to better represent this complexity them in model systems. In this talk, we will address the question: What types and compositions of solid substrates (inorganic vs organic and functional groups present) control U(IV) and Pb(II) behavior in natural sediments and groundwater? Recent studies of redox-variable riparian soils have demonstrated a greater and more complex role for organics in governing U(IV) and Pb(II) sorption at solid surfaces than previously appreciated. For example, in spite of the low solubility of U(IV) mineral phases, U(IV) surface complexes on particulate organic matter and clay-organic assemblages dominate uranium inventories in contaminated anoxic sediments and influence groundwater quality regionally across the upper Colorado River Basin. Similarly, Pb(II) is widely considered to partition strongly to iron/manganese oxides, sulfides, clays, and dissolved organic matter. Yet, recent studies show that solid-associated organic matter plays a major role in controlling Pb(II) behavior in riparian sediments under variable redox conditions. These studies broadly emphasize the need to better understand molecular processes occurring at complex organic-dominated solid-water interfaces.
Green rust (GR) minerals are mixed-valent iron (Fe) hydroxide minerals that are believed to be wide-spread at redox boundaries in natural and engineered environments. These Fe phases consist of brucitic FeII(OH)2 sheets in which a portion of the ferrous (divalent) Fe cations have been replaced with ferric (trivalent) Fe. Because of their layered structure, nano-particulate size, and high Fe(II) content, GR minerals are well known to be effective reductants of a broad range of organic and inorganic compounds, and their redox reactivity has been studied in considerable detail. In contrast, little is known about the reactivity of GR minerals towards redox-stable trace metals such as Ni(II) and Zn(II). The aim of the research presented here was to assess the mechanisms of Ni(II) and Zn(II) sorption onto GR in circumneutral anoxic solutes. The work involved a combination of macroscopic batch kinetic experiments and synchrotron-based X-ray absorption spectroscopic analyses to determine prevailing sorption mechanisms as a function of time. The results demonstrate that GR behaves quite differently as a sorbent of trace metal Ni(II) and Zn(II) than ferric-oxides such as goethite and hematite, as it induces the formation of secondary metal precipitate phases rather than surface complexes as the main mechanism of metal retention. This implies that GR acts as a dynamic sorbent of trace metals in suboxic systems, and suggests that precipitation rather than surface complexation reactions control the solubility of trace metals in Fe redox transitions zones.
Over the next century, sea level is predicted to rise by up to 1 m, leading to the salinization of coastal environments. Salinity affects aqueous geochemistry (i.e., ionic strength, pH), which in turn influences iron (Fe) mineral reactivity, and adsorption capacity. Although Fe (oxyhydr)oxide phases are well-established adsorbents of organic matter (OM), mineral complexation of OM under different salinity regimes has not been investigated. Here, we examined the temporal adsorption and molecular fractionation of coastal dissolved OM to ferrihydrite (Fh) in deionized water (DI), freshwater (FW) and seawater (SW) matrices, using a suite of aqueous geochemical techniques.
We found that salinity matrices substantially impacted the extent, rate, and composition of OM adsorption to Fh surfaces. The extent of dissolved organic carbon removal was similar in DI (46.7 ± 4.5%) and FW (40.5 ± 7.1%); however, removal decreased in SW, with only 26.5 ± 0.36%. Overall, this decreased removal in SW corresponds to a slower rate of removal (-0.004 mg/L min) over 24 hours compared to DI (-0.009 mg/L min) and FW (-0.01 mg/L min). Higher concentrations of divalent cation adsorbed to Fh (2-9 mM) in the SW system, compared to the FW system (0.04-0.1 mM), likely limited OM removal rate and extent. Fourier Transform ion cyclotron resonance mass spectrometry (FT-ICR-MS) revealed DI and FW systems exhibited sequential adsorption of aromatic, lignin-like unsaturated, and aliphatic compounds, previously observed in goethite systems, while the SW system demonstrated rapid, unselective adsorption.
These results suggest a salinity threshold effect on the adsorption of OM to Fh, in which freshwater minorly slows but otherwise does not alter OM adsorption dynamics, while seawater results in decreased, indiscriminate adsorption. This threshold is likely due to occupation of specific surface area by divalent cations, reduced ligand-exchange reactions with organic carboxyl functional groups and subsequent lack of OM-OM scaffolding. Thus, environments subject to salinization could potentially experience higher DOC concentrations, leading to elevated microbial metabolism and CO2 production than previously observed.
Production of struvite (MgNH4PO4●6H2O) from dairy lagoon wastewater (DW) improves phosphorus (P) resource sustainability and reduces the deposition of P into surface waters. Struvite formation can be hindered by dissolved organic matter (DOM) and calcium (Ca), but the prevalence of the phenolic functional group (R-OH) in DW DOM may counteract any limiting effects. A constant composition reactor was used to determine the effect of R-OH on struvite precipitation from simulated DW in the presence of Ca, with phenol (Ph) as a model R-OH. The presence of Ph increased struvite precipitation rate relative to Ca+Ph, Blank (no Ca or Ph) and Ca treatments (51.3, 30.5, 5.8, 2.7 µmol-struvite min-1 respectively). Solids analysis showed that Ph limited the struvite morphology deformation caused by Ca and Ca mineral coprecipitation. Coprecipitation with Ca-P minerals with or without Ph present, did not significantly change the thermal properties of struvite, with the thermogravimetric (TG) mass loss associated with ammonia gas and water vapor emissions occurring at 80-160 ○C. The observed effects of Ph on struvite formation suggest a beneficial contribution of dissolved low molecular weight organics such as R-OH to improved struvite recovery from DW.
Phenol increases struvite precipitation rate from simulated dairy wastewater for improved phosphorus nutrient recovery in the presence of dissolved organics
Track: [I&EC] Division of Industrial and Engineering Chemistry
In developing its stranded hydrocarbon gas fields that reside in high CO2 environments, PETRONAS is committed to zero net carbon emission. For this reason, as part of its development strategy, PETRONAS envisages pipeline transportation of dense-phase CO2 separated during the process to sequestration sites. However, this strategy will be faced with two main challenges: dense phase CO2 is highly corrosive in the presence of free water, and the pipeline may be subjected to running ductile fracture due to the decompression of dense-phase CO2 when ruptured. Realizing the challenges, PETRONAS has embarked on developing the technologies that will address the challenges. For highly corrosive environments, firstly, PETRONAS in collaboration with Ohio University developed a CO2 corrosion model that would yield more accurate corrosion rates of carbon steel in high CO2 environments; that was because existing corrosion models seemed to overpredict CO2 corrosion rates which would consequently disqualify the use of carbon steel. Secondly, PETRONAS developed corrosion inhibitors that would reduce corrosion rates to target values. Thirdly, PETRONAS identified an in-line corrosion monitoring tool suitable for dense-phase CO2 pipelines. As for running ductile fracture, PETRONAS in collaboration with RINA-CSM enhanced existing Battelle Two-Curve Model to enable adequate fracture toughness in the design of the pipelines. The technologies are basically available for the development of the high CO2-containing hydrocarbon gas fields.
Track: [I&EC] Division of Industrial and Engineering Chemistry
Solid particles entrained in fluids can impact pipelines and equipment causing wear and material removal. In addition to interaction of particles with the carrier fluids, particles interaction with solid materials makes this process highly complex and produces effects that are interesting and yet important to predict for practical engineering applications. Solid particle erosion of pipeline and equipment depends on many parameters that can be categorized into three interacting components or elements: One element is solid particles themselves that vary in shape, size, hardness and density. Another important element is the materials characteristics such as hardness of materials, ductility, density, and wear characteristics. The last and certainty not the least is the carrier fluid properties and complexity of the flow regimes. Interaction between these three elements are the most difficult to understand and model especially in multiphase gas-liquid-particle flows. Erosion wear occur due to repeated impacts of solid particles and even liquid droplets with materials surfaces. The hardness of particles relative to the harness of the material target is especially important in mining, process and petroleum industries as many types of particles with various shapes and hardness and size hit various pipe materials and coatings. One of the most interesting variable that occur in industry practice is the size variation of the particles themselves. The size variation of particles entrained in fluids cause impact speed and angle variations with the target materials. The size and shape variability also causes various crater sizes within the target material and these are most difficult to model. Materials characteristics are another important, complex and fascinating part in modeling solid particle erosion. Particles impacting solid particles cause craters, fractures and/or cracks to develop that eventually causes small pieces of materials to be removed from the target materials. Both ductility and hardness of materials play an important role in solid particle erosion. Fluids and their interactions with solid particles adds another order of magnitude in complexity for the prediction of solid particle erosion. Erosion prediction in multiphase pipelines involving gas-liquid and particles is a complex problem due to the lack of understanding of solid particle velocity distributions in the liquid and gas phases.
Track: [I&EC] Division of Industrial and Engineering Chemistry
Anthropogenic CO2 transportation by carbon steel pipelines is one of the most discussed topics in the last decades to move the CO2 from the extraction/separation site to a storage reservoir. It is normally aimed at Enhanced Hydrocarbon Recover when injected in oil fields to increase the site production, at exploiting gas fields on which large amount of CO2 is present in natural gas mixture and, more in general, at reducing the CO2 anthropogenic emissions in atmosphere using the Carbon Capture Transportation and Sequestration solution. The latter is based in a first phase consisting in the separation of CO2 from the flue gases on the production plants and following transportation to specific sites where it can be stored in a safe way. Limited literature is available for CO2 transportation pipelines and only a few full-scale propagation tests have been performed worldwide. This is the reason why the approach to fracture propagation control for CO2 pipelines is also based on the experience gained for natural gas projects. However, some differences exist between natural gas and CO2 pipelines: - Higher operating pressure for CO2 pipeline to guarantee dense/liquid phase. - Different decompression behavior for the CO2 mixture due to change of phase from liquid to gas. - Larger influence on the decompression behavior of possible impurities inside the CO2 mixture with respect to the natural gas mixtures. One relevant issue is related to control the ductile fracture propagation which may be originated from a third-party damage, as well as corrosion flaws. In such cases, ensuring a crack arrest in a short length is fundamental to minimize any possible effects on outer environment and human beings. The fluid decompression occurring during fracture propagation represents the main driving force and, for dense-phase CO2, the plateau in the same decompression and its saturation pressure plays a key role in fracture control. In this paper, it is introduced the CO2 pipelines fracture control and is shown how the presence of impurities in CO2 produced by human activities strongly affect the saturation pressure and, consequently, the material fracture arrest ability. The effect of initial operating conditions is also discussed. Specific numerical tools are used to evaluate the CO2 decompression, able to calculate the thermodynamic and transport properties of industrially important fluids and their mixtures.