Aqueous Aerosols in Atmospheric Chemistry and Airborne Diseases Citation Gu, Alan Yalun (2022) Aqueous Aerosols in Atmospheric Chemistry and Airborne Diseases. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/4spy-az88. https://resolver.caltech.edu/CaltechTHESIS:04142022-025410011 Abstract Aqueous atmospheric aerosols are small droplets (typically smaller than 5 μm) suspended in the air that are comprised of water and water-soluble components. These aerosols provide an air-water interfacial reaction environment on their surfaces, and act as a medium for airborne disease transmission. In this thesis, Chapters II and V explore atmospherically relevant reactions on the aqueous aerosol surface using an online mass spectrometry, while Chapter III investigates the SARS-CoV-2 airborne transmission considering suspended virus-laden aerosols as the transmission media. Spinning off this SARS-CoV-2 work, Chapter IV describes a newly developed quantitative RNA amplification test kit for COVID-19, with an emphasis on the amplification result photo recognition component. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: Aerosol Interface Atmosphere SARS-CoV-2 Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemical Engineering Awards: Demetriades-Tsafka-Kokkalis Prize in Entrepreneurship or Related Fields, 2022. Thesis Availability: Public (worldwide access) Research Advisor(s): Hoffmann, Michael R. Group: COVID-19, Resnick Sustainability Institute Thesis Committee: Wennberg, Paul O. (chair) Gray, Harry B. Goddard, William A., III Hoffmann, Michael R. Defense Date: 7 April 2022 Funders: Funding Agency Grant Number NSF AGS-1744353 Bill & Melinda Gates Foundation OPP1192379, INV-018569, and INV-030223 Resnick Sustainability Institute SRA Record Number: CaltechTHESIS:04142022-025410011 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:04142022-025410011 DOI: 10.7907/4spy-az88 Related URLs: URL URL Type Description https://pubs.acs.org/doi/10.1021/acs.est.1c01962 DOI Article adapted for ch. 2: Role of Ferryl Ion Intermediates in Fast Fenton Chemistry on Aqueous Microdroplets https://pubs.rsc.org/en/Content/ArticleLanding/2022/EA/D1EA00013F DOI Article adapted for ch. 3: Speech-Generated Aerosol Settling Times and Viral Viability Can Improve COVID-19 Transmission Prediction https://pubs.acs.org/doi/10.1021/acs.est.1c04623 DOI Article adapted for ch. 4: Membrane-Based In-Gel Loop-Mediated Isothermal Amplification (mgLAMP) System for SARS-CoV-2 Quantification in Environmental Waters ORCID: Author ORCID Gu, Alan Yalun 0000-0001-8095-3634 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 14550 Collection: CaltechTHESIS Deposited By: Alan Gu Deposited On: 20 Apr 2022 19:42 Last Modified: 04 Feb 2025 23:41 Thesis Files PDF - Final Version See Usage Policy. 5MB Repository Staff Only: item control page

Aqueous Aerosols in Atmospheric Chemistry and Airborne Diseases Thesis by Alan Yalun Gu In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CALIFORNIA INSTITUTE OF TECHNOLOGY Pasadena, California 2022 (Defended April 7, 2022) ii  2022 Alan Yalun Gu ORCID: 0000-0001-8095-3634 iii ACKNOWLEDGEMENTS Acknowledgement Prof. Michael Hoffmann has always been supportive and available during the past several years when I explored vastly different fields of research. When we first met in January 2018, I was a fledgling eager to join a group for rotation, but not sure what I could do. Mike took me into his group with no questions asked. I appreciated that very much because at that time, I would not have been able to answer any questions about research had he asked. Prior to COVID-19, I focused on experimental atmospheric chemistry studies using a custom mass spectrometer developed in house. When the pandemic hit, the laboratory was closed for months, and around the same time the funding for the atmospheric chemistry projects was coming to an end. Thanks to Mike inviting me to the biochemistry subgroup in our research group, I was able to continue my residency. In 2021, I switched to yet another research topic of carbon capture, with an entrepreneurial plan that may take off at any time. Mike accommodated this variability with utmost flexibility, ensuring a smooth transition for me and for the burgeoning company. My experience at Caltech would not have been nearly as educational and enjoyable without him. My thesis committee, Prof. Paul Wennberg, Prof. William Goddard III, and Prof. Harry Gray, took an extended period out of their busy schedule to offer advice on my work. My gratitude extends to them for this opportunity to learn from the experts in their respective fields. A. J. Colussi, Lijie (Sara) Li and Nathan Dalleska were my first mentors after my arrival in the Hoffmann Group. A.J. taught me how to do research, which served as the cornerstone of my scientific career. Thanks to his pedagogy, I was able to switch research topics over the past several years and continue to invent new technologies in my future career. Sara showed me how to perform the atmospheric chemistry experiments and analyze the data and bore with my unfamiliar experimental techniques when I broke the mass spectrometer microjet and interrupted her research. Nathan taught me from the fundamental principles to the maintenance procedures of analytical instruments, which were instrumental to my iv understanding of the chemistry involved in the experiments. I am so lucky to have the right mixture of advisor and mentors, and my scientific career will be a tribute to them. This thesis work is made possible by the generous support of the National Science Foundation for the atmospheric Fenton chemistry project, Bill & Melinda Gates Foundation for the SARS-CoV-2 transmission modelling and testing device projects, and Resnick Sustainability Institute for the CO2 projects. Thanks to them for having faith in me when I was new in the respective research fields and supported my exploratory work when success was by no means guaranteed. Thanks to Clément Cid, Stephanie Yanchinski, and Neil Fromer for their support in the business planning and commercialization efforts in the carbon capture project. Entrepreneurship is full of unknown unknowns for me, and their support turns these unknowns into knowns. The work presented in this thesis would not have been possible without my collaborators: Charles Musgrave, Prof. William Goddard III, Yi Zhang, and Lei Guo on the atmospheric chemistry projects; Yanzhe Zhu, Sean McBeath, and Xunyi Wu on the SARS-CoV-2 projects; as well as Heng Dong and Léopold Dobelle on the CO2 projects. Special thanks to Jing Li who was not only a collaborator but a mentor in biochemistry and microbiology, patiently teaching me the basics from my Grade 10 biology foundation and caring for my learning outcome. Prof. Lifeng Yin visited our laboratory prior to the pandemic and showed me how versatile a researcher can be. We shared laughter, solution-driven research goals, and an entrepreneurial spirit. Our collaboration is going to make both of our lives so much more interesting. Despite the short time Prof. Xuejing Yang visited our laboratory, her knowledge and thinking paradigm inspired the project in Chapter V, and she continued to offer help in our later correspondence. It was so fortunate for me to have met them this early in my career. v Prof. Michael Vicic made my teaching assistant duties an experience beyond just graduation requirements. His care for his teaching assistants and his students’ learning outcome is exemplary. John Nagara from the patent office was consistently dependable during our collaboration, and I could always rely on him to be on time for meetings and deadlines. Such professionalism is the pinnacle of what I strive to attain. Thanks to Prof. Paul Wennberg for inviting me to join his group meetings and activities. This small community meant a lot to me, especially at the start of my graduate studies when the only project I was working on was atmospheric chemistry and no one else in my own research group was working on anything remotely similar. Thanks to Yuanlong Huang, Prof. John Seinfeld, Prof. Richard Flagan, Prof. Jess Adkins, and Donna Wrublewski for thought-provoking discussions. The knowledge they offered in their respective fields transformed my unknown unknowns into known unknowns. The experimental efforts described in this thesis and beyond are reliant on the staff of the Linde Laboratory: Nora Oshima, Bronagh Glaser, and Leilani Rivera-Dotson. After the courses, the laboratory became my home. Thank you for making it a sweet one. The Barracuda Virtual Reactor simulation software team, especially Peter Glaser and the support team, were very helpful and accommodating our request as their new academic customer as well as the questions encountered during the learning process. Thanks to them, I can end my graduate studies on a high note. Graduate studies are more than projects, and I am fortunate to have met my lab mates and friends: Seungkyeum Kim, Axl LeVan, Prof. Kai Liu, Prof. Yang Yang, Cody Finke, Hugo Leandri, Eric Huang, Nissim Gore-Datar, Kevin Vu, Siwen Wang, and John-Alexander vi Preuß. Thank you for sharing a nice time with me, including, or perhaps especially, when not talking about research. Thanks to Sai Gu, for I owe it all to her. vii ABSTRACT Abstract Aqueous atmospheric aerosols are small droplets (typically smaller than 5 μm) suspended in the air that are comprised of water and water-soluble components. These aerosols provide an air-water interfacial reaction environment on their surfaces, and act as a medium for airborne disease transmission. In this thesis, Chapters II and V explore atmospherically relevant reactions on the aqueous aerosol surface using an online mass spectrometry, while Chapter III investigates the SARS-CoV-2 airborne transmission considering suspended virus-laden aerosols as the transmission media. Spinning off this SARS-CoV-2 work, Chapter IV describes a newly developed quantitative RNA amplification test kit for COVID-19, with an emphasis on the amplification result photo recognition component. viii PUBLISHED CONTENT AND CONTRIBUTIONS Published Content and Contributions Gu, A. Y., Musgrave, C., Goddard, W. A., Hoffmann, M. R., & Colussi, A. J. (2021). Role of Ferryl Ion Intermediates in Fast Fenton Chemistry on Aqueous Microdroplets. Environmental Science & Technology, 55(21), 14370-14377. doi:10.1021/acs.est.1c01962 G. A. Y. designed the experiments, performed the experiments, analyzed the data, and wrote the non-computational sections of the manuscript. Gu, A. Y.#, Zhu, Y.#, Li, J., & Hoffmann, M. R. (2022). Speech-Generated Aerosol Settling Times and Viral Viability Can Improve COVID-19 Transmission Prediction. Environmental Science: Atmospheres. doi:10.1039/d1ea00013f G. A. Y. conceived the project, designed the computational algorithm, collaborated in writing the code, analyzed the data, and wrote the manuscript other than the figures. Zhu, Y., Wu, X., Gu, A., Dobelle, L., Cid, C. A., Li, J., & Hoffmann, M. R. (2021). Membrane-Based In-Gel Loop-Mediated Isothermal Amplification (mgLAMP) System for SARS-CoV-2 Quantification in Environmental Waters. Environmental Science & Technology. doi:10.1021/acs.est.1c04623 G. A. collaborated in performing the experiments, analyzing the data, and writing the manuscript. ix TABLE OF CONTENTS Table of Contents Acknowledgement ....................................................................................................... iii Abstract......................................................................................................................... vii Published Content and Contributions..................................................................... viii Table of Contents......................................................................................................... ix List of Illustrations....................................................................................................... xi List of Tables .............................................................................................................. xvi Chapter I: Introduction................................................................................................. 1 Chapter II: Air-Water Interfacial Ozone Fenton Chemistry.................................. 4 1.1 Abstract ............................................................................................................. 4 1.2 Synopsis Statement ......................................................................................... 5 1.3 Introduction ...................................................................................................... 5 1.4 Methods............................................................................................................. 7 1.4.1 Experimental Section .................................................................................. 7 1.4.2 Quantum Mechanical Calculations ........................................................... 9 1.5 Results and Discussion ................................................................................. 10 1.5.1 Identification and Quantification of Products ....................................... 10 1.5.2 Rates and Product Yields at the Air-Water Interface vs Bulk Water 11 1.5.3 Ligand Effects ............................................................................................ 12 1.6 Summary......................................................................................................... 14 1.7 Supporting Information ................................................................................ 15 1.7.1 Online Electrospray Ionization Mass Spectrometry ............................ 15 1.7.2 Quantification of Ozone Concentration in Spray Chamber................ 15 1.7.3 Kinetic Analysis of Me2SO(aq) Ozonation in Bulk Water ................... 15 1.8 Figures ............................................................................................................. 18 1.9 Tables .............................................................................................................. 40 1.10 Acknowledgements .................................................................................... 42 Chapter III: SARS-CoV-2 Indirect Airborne Transmission Model ................... 43 2.1 Abstract ........................................................................................................... 43 2.2 Introduction .................................................................................................... 44 2.3 Methods........................................................................................................... 46 2.3.1 Data Mining ................................................................................................ 46 2.3.2 Aerosol Settling Behavior ........................................................................ 47 2.3.3 Viral Viability............................................................................................. 49 2.3.4 Transmission Model .................................................................................. 49 2.4 Results and Discussion ................................................................................. 50 2.5 Conclusions .................................................................................................... 56 2.6 Supplementary Note ..................................................................................... 57 2.7 Figures ............................................................................................................. 59 2.8 Tables .............................................................................................................. 71 2.9 Acknowledgements....................................................................................... 72 Chapter IV: Image Recognition for a SARS-CoV-2 Test Kit ............................ 73 3.1 Abstract ........................................................................................................... 73 3.2 Introduction .................................................................................................... 74 3.3 Image Analysis Methods.............................................................................. 76 3.4 Image Analysis Results: Proof of Portability ........................................... 77 3.5 Figures ............................................................................................................. 78 3.6 Acknowledgements....................................................................................... 79 Chapter V: CO2-Acivated H2O2 Oxidation ............................................................ 80 4.1 Abstract ........................................................................................................... 80 4.2 Introduction .................................................................................................... 80 4.3 Methods........................................................................................................... 81 4.4 Results and Discussion ................................................................................. 83 4.5 Conclusions .................................................................................................... 84 4.6 Figures ............................................................................................................. 85 4.7 Acknowledgements....................................................................................... 90 Bibliography ................................................................................................................ 91 Appendix A: Coordinates of DFT-optimized Structures ................................... 105 x xi LIST OF ILLUSTRATIONS List of Illustrations Number Page Figure 1. Positive (a) and negative (b) online electrospray ionization mass spectra of pH 4.5 (1mM Me2SO + 100µM Fe(ClO4)2 + 50µM Cs+ + 250µM ClO4-) aqueous microdroplets before (blue) and after (red) exposure to 283ppm O3(g) for < 1 ms. Cs+ and ClO4- non-complexing ions added as internal standards..................... 18 Figure 2. Mechanism of the Fenton oxidation of Me2SO into Me2SO2, MeSO3- and HSO4-. Me2SO2 is the only product of the ferryl ion pathway (bottom left). .......... 20 Figure 3. Structures of charge-neutral FeIVO and FeII species. Ligands in blue. Structures do not report actual coordination, stereochemistry, denticity, or hapticity. For DFT-optimized structures, see Figures 11-12.................................. 21 Figure 4. Ligand effects on the formation of dimethyl sulfone Me2SO2. [Me2SO2] vs E0(LFeIVO2+ + O2 → LFe2+ + O3) DFT formal reduction potentials calculated from ∆G(LFeIVO2+ → LFe2+) Gibbs free energies differences relative to ∆G(Me2SO → Me2SO2). The inset shows E0(LFeIVO2+ + O2 → LFe2+ + O3) values for the different ligands L. [Me2SO2] determined at the air-water interface of pH 4.5 [1mM Me2SO + 100 µM Fe(ClO4)2 + 200 µM ligand L + 150 µM NaClO4] aqueous microdroplets exposed to 283 ppmv O3(g) for < 1 ms (see text for more details). Because no Cs+ is used as the internal standard in the positive mode, the absolute concentration of [Me2SO2H+] measured has higher uncertainty than Figure 14, though the trend is unaffected. Me2SO2 concentration is measured as Me2SO2H+. Due to unavoidable complex formation between the negatively charged ligands and Me2SO2H+ cation, the yield measured is an underestimate of the actual yield. DMA = dimethylamine, TMA = trimethylamine, CAT = catechol, NAP = naphthoquinone, NLI = no ligand, PBQ = p-benzoquinone, MAL = malonate, CHL = chloride, OXA = oxalate, TAR = tartrate, EDT = ethylenediaminetetraacetate. ............................................ 22 Figure 5. Protonated dimethyl sulfone Me2SO2H+ (m/z+ = 95) mass signal intensities vs [EDTA]/[Fe(II)] at various pH values. Me2SO2H+ mass signal intensities are measured by online electrospray ionization mass spectrometry at the air-water interface of [1mM Me2SO + 100 µM Fe(ClO4)2] aqueous microdroplets exposed to 283 ppm O3(g) for < 1 ms. The mass signal intensities measure the net yield of Me2SO2 from Fenton chemistry and direct ozonolysis of Me2SO, after subtracting the Me2SO2H+ complexed by EDTA. No Fe(II) means experiments in the absence of EDTA or FeII............................................................................ 23 Figure 6. Experimental setup of online ES-MS. The aqueous reactants and gaseous reactants contact at the outlet tip of the microjet................................................... 24 Figure 7. Experimental setup of offline ES-MS. The aqueous reactants contact each other in the manifold tank reactor before entering the microjet. ............................ 25 Figure 8. Me2SO2 calibration curve using 1, 10 and 100 mM Me2SO2. Y-axis plots the mass signal intensity ratios of Me2SO2H+/Cs+ at pH = 4.5 and [Cs+] = 50 µM. ..... 26 xii - Figure 9. MeSO3 calibration curve using 1, 10 and 100 µM MeSO3Na. Y-axis plots the mass signal intensity ratios of MeSO3-/ClO4- at pH = 4.5 and [ClO4-] = 250 µM. .......................................................................................................................... 27 Figure 10. HSO4- calibration curve using 1, 10 and 100 µM NaHSO4. Y-axis plots the mass signal intensity ratios of HSO4-/ClO4- at pH = 4.5 and [ClO4-] = 250 µM. .... 28 Figure 11. Optimized structures of ligated Fe2+ and FeO2+ complexes from DFT calculation. ......................................................................................................... 31 Figure 12. Optimized structure of R3 transition state. ..................................................... 31 Figure 13. Relative product concentration of MeSO3- and HSO4- in interfacial ozoneFenton chemistry for 1mM Me2SO and 100 µM Fe(ClO4)2. ................................. 32 Figure 14. Product concentrations of (a) Me2SO2 (b) MeSO3- and (c) HSO4- at different initial FeII bulk concentration ([FeII]) and gaseous O3 concentration at pH 4.5 resulting from ozone Fenton oxidation of 1mM Me2SO. Concentrations on the y-axis are the corresponding bulk concentrations of Me2SO2, MeSO3- or HSO4in hypothetical equilibrium with the sampled surface concentrations. The product concentrations are used to calculate the ferryl:hydroxyl pathway product concentrations ratio, represented as (d) the cumulative ferryl:hydroxyl intermediate. CsClO4 and NaClO4 are added as the internal standards to a total concentration of 50 µM Cs+ and 200 µM ClO4-. ................................................... 33 Figure 15. Me2SO2 product concentration at the air-water interface for 1mM Me2SO and 100µM Fe(ClO4)2 including Me2SO2 produced from Me2SO direct ozonolysis. The y-intercepts show the product concentration from direct ozonation of Me2SO, and the slope shows the product concentration from ferryl O-atom transfer. ................................................................................................. 34 Figure 16. Me2SO2 product concentration in the bulk phase reaction from 10mM Me2SO, 5mM O3 and 100µM Fe(ClO4)2 reacting for 746 seconds, including Me2SO2 produced from Me2SO direct ozonolysis. The y-intercepts show the product concentration from direct ozonation of Me2SO, and the slope shows the product concentration from ozone-Fenton chemistry, including but not limited to O-atom transfer from ferryl species. The Me2SO2 product concentration originating from 100µM Fe2+ sets the upper bound for ferryl O-atom transfer kinetics. .............................................................................................................. 35 Figure 17. MeSO3- product concentration in bulk from 10mM Me2SO, 5mM O3, and 100µM Fe(ClO4)2 reacting for 746 seconds. The y-intercepts show the product concentration from direct ozonation of Me2SO, and the slope shows the product concentration from ozone-Fenton chemistry, including both ferryl and OH pathways. ........................................................................................................... 36 Figure 18. 5mM EDTA effect on the ozone Fenton chemistry between 10mM FeCl2(aq) and 283ppm O3(g). Blue denotes without EDTA and red with EDTA in the reaction mixture. The iron peaks are FeIICl3- most visible at m/z=161/163/165, FeIVOCl3- most visible at m/z=177/179/181, and FeIIICl4- most visible at m/z=196/198/200/202. The signal intensities are lowered for all three species with EDTA added, showing that ligands compete with Cl- for the coordination sphere of Fe2+ ions. ............................................................................................. 37 xiii Figure 19. Methanesulfonate formation [MeSO3 ] vs E (Fe + e ⟶ Fe ) reduction potentials. The inset shows calculated E0(Fe3+ + e- ⟶ Fe2+) potentials as functions of ligand. Standard reduction potentials of Fe3+ to Fe2+ in the presence of ligands shown on the x-axis of the inset are from Martell and Smith.10 Reactants are 1mM Me2SO, 100 µM Fe(ClO4)2, 200 µM ligand L, 283ppm O3 at the interface of gaseous O3 and aqueous microdroplets. Abbreviations: DMA = dimethylamine, TMA = trimethylamine, CAT = catechol, NAP = naphthoquinone, NLI = no ligand, PBQ = p-benzoquinone, MAL = malonate, CHL = chloride, OXA = oxalate, TAR = tartrate, EDT = ethylenediaminetetraacetate. ............................................................................... 38 Figure 20. Ligand effect on interfacial Me2SO2 product concentration in 25 weightpercent catechol/water and 50 volume-percent acetonitrile/water solvents. The concentrations of Me2SO and Fe(II) are fixed at 1mM and 100µM, respectively. O3(g) is at 283 ppm. ............................................................................................. 39 Figure 21. (a) illustration showing the model data flow in this work (b) typical COVID19 progression around the date of positive test result. The three periods are: the pre-symptomatic contagious period, the wait period to obtain the test result after taking the test, and the recovery period at the end of which the patient is modelled as either recovered and no longer contagious, or entering the intensive care unit (ICU) and isolated from the public. We assume that the patient takes the test at the onset of symptoms. Under this assumption, a positively tested patient is considered contagious in our model from 4 days before until 12 days after the positive test result. ................................................................................. 59 Figure 22. (a) Calculated settling velocities of aerosols of varying sizes using StokesCunningham Law. (b) The Reynolds number (Re) and Knudsen number (Kn) of droplets of varying sizes. At Kn < 10, the Stokes-Cunningham Law is the most applicable first-principle relationship to calculate settling velocity. ...................... 61 Figure 23. Evaporated aerosol sizes derived from the Köhler equation based on an initial size at different ambient (a) temperatures at 50% relative humidity and( b) relative humidity at 25 ℃. (c) Calculated settling times obtained from the empirical model using 6 µm as the initial droplet size and (d) viral viability at different ambient temperatures and relative humidity. .......................................... 62 Figure 24. Timeseries of daily case percentage increase in decimal format for April 2020 in counties studied. The predicted daily case increase of the last 4 days are shown as triangles with their associated one-sigma prediction intervals. Dashed lines show the model fitting from the 6th day to the 25th day of April. No fitting data obtained from the model for the first 5 days because they would require weather data from March (up to 5 days prior). LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long-/short-term memory recurrent neural network. The green filled areas represent 95% confidence intervals for LR predictions. The blue patterned areas represent 95% confidence intervals for VAR predictions. .................................... 64 Figure 25. Contour plots of the daily case percentage increase as a function of settling time and viral viability (represented by half-life) for different counties. Colour - 0 3+ - 2+ xiv shows the daily case percentage increase in decimal. LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long short term memory recurrent neural network.................................... 66 Figure 26. (a) Fitting and (b) predicted daily new case percentage increase for Maricopa County from May to August 2020. Interrupted data in (a) is due to interrupted weather history data from NOAA. Error bars show one-sigma prediction intervals. The training data in (a) are 75 days long and the testing data in (b) are 21 days long. LR: linear regression; Simple RNN: simple recurrent neural network; LSTM: long short term memory recurrent neural network. The green filled areas represent 95% confidence intervals for LR predictions. ...................... 67 Figure 27. Size distribution of speech-generated droplets before evaporation measured by Morawska et al.91 The peak number concentration is at 6 µm and is used as the input to the Köhler equation in this work........................................................ 68 Figure 28. Data points available to generate the contour plot in Fig. 5 for each county, with the daily case percentage increase as a function of settling time and viral viability (represented by half-life). Colour shows the daily case percentage increase in decimal. The daily cases percentage increase is shown as the colour of each data point. The dots represent data points for training, and the triangles represent data for prediction. LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long short-term memory recurrent neural network. .................................................................................... 70 Figure 29. Cloud-based image processing and portable device prototyping. (a) An example mgLAMP image taken by a smartphone. (b) In the same image, the squares highlight the recognized amplicon dots by our trained Google AutoML Vision model. (c) The accuracy of the trained ML model in amplicon dot counting. Sixteen smartphone images were tested, and the results were compared to manual counting of the same images. The dotted line represents the linear fitting of the observed ML counts versus manual counts, with the equation and R-square displayed. (d) Picture of our portable mgLAMP device prototype. The prototype holds 9 mgLAMP reactions, provides 65 °C heating, laser excitation, and fluorescence emission filter. (e) Examples picture of the output from the prototype. ........................................................................................................... 78 Figure 30. Oxidation of methanesulfinate MeSO2- by H2O2 or HCO4- to methanesulfonate MeSO3-. The oxidation byproducts are HCO3- (from HCO4-) and H2O (from H2O2).......................................................................................... 85 Figure 31. Experimental setup of the online ESI-MS. The aqueous reactants and gaseous reactants contact at the outlet tip of the microjet...................................... 86 Figure 32. Experimental setup of the offline ESI-MS. The aqueous reactants contact each other in the manifold tank reactor before entering the microjet. .................... 87 Figure 33. Percarbonate a.k.a. peroxymonocarbonate (HCO4-) formation kinetics (yaxis) measured at different pH, in the bulk aqueous phase and at the air-water interface. Rate constant is obtained assuming first order dependence on the concentration of H2O2, and CO2 or HCO3-. The ‘Bulk’ reaction forms peroxymonocarbonate from aqueous H2O2 and aqueous HCO3-; the ‘Interface’ reaction forms peroxymonocarbonate from aqueous H2O2 and gaseous CO2. ........ 88 xv - Figure 34. Kinetics of methanesulfinate (MeSO2 ) oxidation by H2O2 (orange bars) or HCO4- (green and purple bars) at different pH, in the bulk aqueous phase (orange and purple bars) or air-water interface (green bars). The rate constant is plotted on the y-axis. The rate constants are obtained assuming first order dependence on MeSO2-, H2O2, and CO2 or HCO3-. The ‘Bulk H2O2’ reactions involve oxidizing aqueous MeSO2- with aqueous H2O2 in a nitrogen atmosphere; the ‘Interfacial HCO4-’ reaction involve oxidizing aqueous MeSO2- with aqueous H2O2 in the presence of gaseous CO2; the ‘Bulk HCO4-’ reaction involve oxidizing aqueous MeSO2- with aqueous H2O2 in the presence of aqueous HCO3. ......................................................................................................................... 89 xvi LIST OF TABLES List of Tables Number Page Table 1. Ligand effect on the relative product concentrations of MeSO3- and HSO4-. ClO4- denotes no ligand as perchlorate does not ligate Fe2+ .................................. 40 Table 2. Important bond distances for the Fe systems investigated. All distances are in Å. Some structures contain multiple Fe-Ligand distances due to increased denticity. Structures are depicted in Figure 4 ....................................................... 41 Table 3. (a) R-squared of model fitting and (b) sum of squares of residuals of model prediction for each county and model during April 2020. (c) R-squared of model fitting and sum of squares of residuals of model prediction for Maricopa County during extended time frame from May to August 2020. LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long short term memory recurrent neural network.................................... 71 1 Chapter 1 INTRODUCTION Chapter I: Introduction Chapter II was completed prior to COVID-19, investigating the ozone Fenton chemistry at the interface between an aqueous aerosol containing ferrous ion and gas phase ozone. This reaction produces reactive ferryl ions and hydroxyl radicals, both capable of driving oxidation chemistries in the atmosphere. This chapter investigates the oxidation kinetics of ozone Fenton chemistry via both the ferryl and the hydroxyl oxidation intermediates simultaneously using an online electrospray ionization mass spectrometry. In the spray chamber of the mass spectrometer, an aqueous jet containing ferrous perchlorate and dimethyl sulfoxide reacts with an intersecting gas stream of ozone, where the reaction takes place at the air-water interface. Results suggest that the ferryl ions are the dominant intermediate over hydroxyl radicals by 2 orders of magnitude at acidic pH. The interfacial kinetics is measured to be 4 orders of magnitude faster than the bulk aqueous phase kinetics for this ozone Fenton reaction system. The ferryl ion intermediate also dominate over the hydroxyl radicals in the bulk aqueous phase. In the presence of a common atmospheric ligand, the oxidation of an atmospheric substrate by the ozone Fenton reaction via the ferryl intermediate may be accelerated or suppressed, and the kinetics may be predicted by the standard reduction potential of ligated ferryl ions. Chapter III was completed during the lockdown period of the COVID-19 pandemic and is a simulation investigation of the role of indirect airborne transmission of SARS-CoV-2. In the beginning of the pandemic, there was significant uncertainty towards the primary route of transmission, with contaminated surfaces and airborne aerosol transmission being the contenders. In contaminated-surface-based transmission, virus is deposited onto a surface by an infected individual, and is transmitted to another individual when that individual touches the contaminated surface. In airborne transmission, virus-laden aerosols are emitted by an infected individual, and another individual breathes in those suspended aerosols. Because the 2 settling time of those aerosols and the viability of the SARS-CoV-2 virus in those aerosols are direct functions of the ambient temperature and humidity, one expects to see correlation between weather and SARS-CoV-2 transmission when linked via the aerosol settling time and viral variability. Assuming a typical speech-generated aerosol size of 6 μm as the initial aerosol size and an asymptomatic transmission period of 5 days prior to testing, statistically significant correlation is observed in the reported data of 5 United States counties using linear regression and recurrent neural network. This result supports asymptomatic airborne transmission as a primary transmission route of SARS-CoV-2, and encourages epidemiologists to include meteorological variables in future disease transmission models. Chapter IV was completed after the initial lockdown period of the pandemic, where I developed an image recognition software to count RNA amplicons after a Membrane-Based In-Gel Loop-Mediated Isothermal Amplification (mgLAMP) reaction. The mgLAMP is an alternative to Reverse Transcription quantitative Polymerase Chain Reaction (RT-qPCR) in SARS-CoV-2 detection, but with fewer equipment requirements due to its isothermal nature. Taking advantage of this lower requirement, our group has developed a home test kit for SARS-CoV-2 capable from collecting an individual’s virus sample to obtaining a photo of the mgLAMP reaction glass slide. The Google AutoML Vision module presented in this chapter counts the amplicons in the photo, and returns a total count number. A non-zero count is a positive test result, and a zero count is a negative result. The computer count agrees with manual count with a slope of 0.9926 (R2 = 0.992) and may be used to reduce worker workload at testing centers or produce an automatic result for home test kit users. Chapter V is an ongoing experimental project investigating the importance of peroxymonocarbonate chemistry in the atmosphere. Peroxymonocarbonate is an oxidative intermediate formed from CO2 and H2O2, and has been found to oxidize sulfides and alkenes ~100 times faster than H2O2 alone. The importance of this reaction in the atmosphere, however, remains unexplored. This work quantifies the rate acceleration by introducing gaseous or dissolved CO2 into H2O2 oxidation reactions relevant in the atmosphere. The experiment is carried out in the online electrospray ionization mass spectrometer outlined in 3 Chapter II. Preliminary results suggest 2 orders of magnitude rate acceleration by introducing CO2, and another 4 orders of magnitude rate acceleration at the air-water interface vs bulk aqueous phase. CO2-activated H2O2 chemistry may play an important yet previously unrecognized role in atmospheric chemistry and may explain the observed higherthan-expected oxidation rates of various substrates in the atmosphere. 4 Chapter 2 AIR-WATER INTERFACIAL OZONE FENTON CHEMISTRY Chapter II: Air-Water Interfacial Ozone Fenton Chemistry Reproduced from the article below with permission from the American Chemical Society. Further permission related to this material should be directed to the American Chemical Society: https://pubs.acs.org/doi/10.1021/acs.est.1c01962 Gu, A. Y., Musgrave, C., Goddard, W. A., Hoffmann, M. R., & Colussi, A. J. (2021). Role of Ferryl Ion Intermediates in Fast Fenton Chemistry on Aqueous Microdroplets. Environmental Science & Technology, 55(21), 14370-14377. doi:10.1021/acs.est.1c01962 1.1 Abstract In the aqueous environment, FeII ions enhance the oxidative potential of ozone and hydrogen peroxide by generating the reactive oxoiron species (ferryl ion, FeIVO2+) and hydroxyl radical (.OH) via Fenton chemistry. Herein, we investigate factors that control the pathways of these reactive intermediates in the oxidation of dimethyl sulfoxide (Me2SO) in FeII solutions reacting with O3 in both bulk-phase water and on the surfaces of aqueous microdroplets. Electrospray ionization mass spectrometry is used to quantify the formation of dimethyl sulfone (Me2SO2, from FeIVO2+ + Me2SO) and methanesulfonate (MeSO3-, from .OH + Me2SO) over a wide range of FeII and O3 concentrations and pH. In addition, the role of environmentally relevant organic ligands on the reaction kinetics was also explored. The experimental results show that Fenton chemistry proceeds at a rate ∼104 times faster on microdroplets than in bulk-phase water. Since the production of MeSO3- is initiated by .OH radicals at diffusion-controlled rates, experimental ratios of Me2SO2/MeSO3- > 102 suggest that FeIVO2+ is the dominant intermediate under all conditions. Me2SO2 yields in the presence of ligands, L, vary as volcano-plot functions of 5 0 IV 2+ 2+ E (LFe O + O2/LFe + O3) reduction potentials calculated by DFT with a maximum achieved in the case of L ≡ oxalate. Our findings underscore the key role of ferryl FeIVO2+ intermediates in Fenton chemistry taking take place on aqueous microdroplets. 1.2 Synopsis Statement The ferryl ion intermediate in the reaction between dissolved iron and gaseous ozone can play a more important role than previously envisioned in the atmospheric system. 1.3 Introduction Atmospheric oxidants such as ozone (O3) and hydrogen peroxide (H2O2) react with the transition metal ions in natural aqueous media. In aerosol microdroplets, FeII/FeIII and CuI/CuII,1 catalyze the oxidation of the aerosol organic components by O3 and H2O2 via Fenton chemistry.2 This chemistry is ubiquitous in the atmosphere.3-13 The catalytic mechanism involves the production of more reactive intermediates, which in the case of FeII can be either high-valent state oxoiron (ferryl) FeIVO2+ ions14 or hydroxyl radicals .OH, reactions R1 and R2.15 FeII + O3 + H+⟶ FeIII + .OH + O2 FeII + O3 ⟶ FeIVO2+ + O2 [R1] [R2] Both, .OH radicals, E0(.OH + H+ + e- ⇌ H2O) = 2.7 V vs SHE, and FeIVO2+, E0(FeO2+ + H+ + e- ⇌ Fe(OH)2+) > 1.2V vs SHE, are strong one-electron oxidants, but their chemistries are quite different.16 FeIVO2+ can transfer O-atoms to low-valence atoms,17 such as divalent S, or selectively insert O-atoms into C-H bonds via Gif chemistry.18-20 Hydroxyl radicals (.OH), in contrast, are unselective and react with most substrates via fast H-atom abstractions or addition reactions.21 Both intermediates generate products that are generally different from those obtained in the uncatalyzed oxidations.22 These are significant issues in atmospheric chemistry. For example, one study estimated that up to ∼30% of atmospheric .OH would be accounted for by Fenton chemistry if reaction R1 were the exclusive channel.23 A FeIVO2+/.OH = 104 ratio would account for the oxidation of atmospheric SO2 in thick hazes.23 6 In the atmosphere, Fe O can oxidize dimethyl sulfoxide, epoxidize alkenes such as IV 2+ 24 25 isoprene and pinene,26 and hydroxylate short alkanes.27 It may also oxidize alcohols28 and organosulfides into aldehydes and sulfoxides. However, factors that control the relative importance of R1 vs R2 reactions are not well understood. Factors that could affect the outcome of Fenton reactions include the nature of the substrates, pH,16 solvents,29 and the presence and nature of complexing ligands.30 In this paper, we address the issues as they relate to the Fenton oxidation of dimethyl sulfoxide (Me2SO) 1) at the air-water interface of microdroplets by exposing aqueous (Me2SO + FeII) solutions to O3(g), and 2) in bulk water by mixing solutions of (Me2SO + FeII) with O3(aq). The main reaction products are dimethyl sulfone (Me2SO2) and methanesulfonate (MeSO3-) as shown in R3-R5.31 Me2SO + FeIVO2+ ⟶ Me2SO2 + Fe2+ Me2SO + .OH ⟶ .CH3 + MeSO2- + H+ MeSO2- + .OH + O2 ⟶ MeSO3- + HO2⋅ [R3] [R4] [R5] Our choice of Me2SO as the substrate is based, in part, on the fact that it is a common component of the marine aerosol resulting from the processing of biogenic dimethyl sulfide (Me2S) emissions.32 These emissions account for ∼15% of the global atmospheric sulfur budget. Moreover, in part, by the fact that Me2SO not only reacts with .OH radicals, but with FeIVO2+ ions via O-atom transfer to the S-atom.16 This feature lets us simultaneously track the competition between reactions R3 vs R4-R5 under identical conditions. The ozone Fenton reaction, FeII + O3, was studied because it is expected to be more important than the FeII + H2O2 reaction under atmospheric conditions. For example, the O3 and H2O2 tropospheric concentrations are comparable (typically 10 ppbv O3 vs 2 ppbv H2O2), but their Henry’s law constants are very different: H(H2O2) = 9 × 102 mol m-3 Pa-1 vs H(O3) = 1.0 × 10−4 mol m-3 Pa-1. Consequently, the concentrations of dissolved H2O2 are expected to be ∼ 10 larger than those of O3. Although O3 reacts only ∼ 10 times faster than H2O2: k(FeII + O3) 7 7 4 = 8.2×105 M-1 s-1 vs k(FeII + H2O2) = 76 M-1 s-1 in bulk water,33-34 our experiments show that O3(g) reacts with FeII(aq) ions at the air-water interface of aqueous microdroplets faster than H2O2,35 and ∼104 times faster than in bulk water. Therefore, O3 is expected to compete favorably with H2O2 at the air-water interface. Fenton chemistry as driven by O3(g) and H2O2(g) on aqueous microdroplets is similar.35 Based on these considerations and previous observations, we make the case that Fenton chemistry on aqueous aerosols mostly proceeds via reactions R1 and R2. 1.4 Methods 1.4.1 Experimental Section Positively and negatively charged reaction products were analyzed by electrospray ionization mass spectrometry (ESI-MS). In the gas-liquid experiments, aqueous solutions of Me2SO and Fe(ClO4)2 (in some experiments also containing one of the following ligands: sodium oxalate, sodium malonate, ethylenediaminetetraacetate, sodium dimethylamine, tartrate, trimethylamine, disodium-dihydrogen p-benzoquinone, 1,4- naphthaquinone or catechol) are injected (at 30 µL min-1 through a stainless steel Agilent G1946-68703 nozzle) into the spray chamber of the ESI mass spectrometer (Agilent G2445A Ion Trap MSD) (maintained at 1 atm with ultrapure N2(g), 293 ± 2 K) using a syringe pump (Cole-Parmer 74905-02 and 74905-50) . A coaxial, codirectional, fast (∼340m s-1) stream of N2(g) pneumatically nebulizes the outermost layers of the aqueous microjets into submicrometer-sized droplets (microdroplets) within 50 µs.36 Microjets and microdroplets were made to react with an orthogonal stream of O3(g) directed to the tip of the nozzle for ∼ 1 ms, which is the estimated residence time of the microdroplets in the spray chamber, before they are pumped into the low pressure section of the mass spectrometer.37-38 Microdroplets are naturally charged during nebulization by the segregation of anions from cations and deflected by an applied electric field towards the inlet to the low-pressure section of the mass spectrometer. Therein, microdroplets evaporate within a heated capillary tubing prior to undergoing Coulombic explosions that release the ions contained in the microdroplets to the 8 gas-phase for detection. A schematic of this experimental setup is shown in Figure 6. More details are provided in the Supporting Information (SI) section. Experiments carried out in bulk-phase water were performed by injecting (Me2SO + Fe(ClO4)2) and O3 aqueous solutions through separate ports into a manifold mixer (ColeParmer Manifold Mixing Valve EW-01356-21, Figure 7) modelled as a tank reactor. The reactions were then allowed to take place for 12 minutes before injection into the ESI mass spectrometer for analysis. Instrumental drift was eliminated by adding non-complexing Cs+ (50µM CsClO4) and ClO4- (200µM NaClO4) ions as internal standards. Me2SO2H+ (from the protonation of the Me2SO2 product) and MeSO3- product signals (see below) were quantified based on calibration curves obtained from Me2SO2 and MeSO3Na solutions (see Figures 810). Positive and negative ions were quantified in the same experiments by alternating positive/negative either in mass scanning or SIM (selected ion monitoring) detecting modes. The ESI mass spectrometer was operated as follows. Positive ions were detected by polarizing the capillary at -3.4 kV (voltages relative to ground), the end plate offset and the capillary exit at 500 V and 78 V, respectively. In the negative mode, the capillary voltage was at 3.5 kV, and the end plate offset and the capillary exit at 500 and -64.5 V, respectively. The nebulizer N2 gas source was maintained at 40 psi, and the drying gas (at 340 °C) flow rate was 12 L min-1. Ozone was generated by flowing ultrapure grade 99.9999 % O2(g) at 50 cm3 min-1 (via a mass flow controller (MKS 247) through an ozone generator (Ozone Solutions) and quantified via UV-visible absorption spectrophotometry (Shimazu UV-2101PC). The pH of aqueous solutions was adjusted before injection by adding HClO4/NaOH solutions and measured (within ± 0.5 pH units) with pH paper (Milipore MCholorpHast). Sodium oxalate, sodium malonate, sodium tartrate, disodium-dihydrogen ethylenediaminetetraacetate (Na2EDTA), dimethylamine (DMA), trimethylamine (TMA), p-benzoquinone, 1,4naphthaquinone, catechol, dimethylsulfoxide (Me2SO, DMSO), dimethylsulfone (Me2SO2, DMSO2), sodium methanesulfinate (MeSO2Na), sodium methanesulfonate (MeSO3Na), iron (II) perchlorate (Fe(ClO4)2), cesium perchlorate (CsClO4), sodium perchlorate (NaClO4), 9 sodium sulfate (Na2SO4), acetonitrile (ACN), perchloric acid (HClO4) and sodium hydroxide (NaOH) of purity 95% or higher were obtained from Sigma Aldrich and used as received. 1.4.2 Quantum Mechanical Calculations Quantum mechanics (QM) calculations were carried out using Density Functional Theory (DFT) with the B3LYP hybrid functional selected for its high accuracy and versatility.39 We used the Grimme-Becke-Johnson D3 empirical correction for describing London dispersion (van der Waals attraction) interactions.40 Initial geometry optimizations were performed in the vacuum where all atoms were described with a singly-contracted triple-zeta basis set (TZV). Following vacuum optimizations, solvated single-point energies were calculated with an extended triplet-zeta basis set containing additional polarization functions on nonhydrogen atoms (TZV*).41-42 To describe solvent effects, we used the Poisson Boltzmann Finite element (PBF) implicit solvent method with water solvent parameters of ε = 80.37 and a probe radius of 1.4 Å. The structure optimizations started with all structures having octahedral coordination of FeII/FeIV. To obtain accurate solvation energies for our clusters, we balanced the net charge to zero by replacing H2O ligands with OH- ions. For example, we performed calculations on the charge-neutral [FeIVO(H2O)3(OH)2], resulting from substituting two OH- ions for two waters, rather than on [FeIVO(H2O)5]2+ ions. We confirmed that charge neutralization barely affects the trend of reduction potential as a function of the ligand. After performing vacuum-optimizations at the B3LYP-D3/TZV level, all FeIVO2+ structures converged to 6-coordinate iron-centered structures, except for the cases of [FeIVO(H2O)2(OH)2(TMA)], [FeIVO(H2O)2(OH)2(DMA)] (and un-ligated [FeIVO(H2O)3(OH)2], where TMA=trimethylamine and DMA=dimethylamine). These amines were treated as fully protonated ligands, whereas all other ligands were either singly or doubly deprotonated depending on experimental pH and pKa values. The reaction free energies for the conversion of FeIVO2+ to FeII (calculated as:∆𝐺𝐺𝑟𝑟𝑟𝑟𝑟𝑟 = �𝐺𝐺𝐹𝐹𝐹𝐹 𝐼𝐼𝐼𝐼 + 𝐺𝐺𝑀𝑀𝑀𝑀2 𝑆𝑆𝑂𝑂2 � − �𝐺𝐺𝐹𝐹𝑒𝑒 𝐼𝐼𝐼𝐼𝑂𝑂2+ + 𝐺𝐺𝑀𝑀𝑀𝑀2 𝑆𝑆𝑆𝑆 �) are interpreted as FeIVO2+ 2-electron reduction potentials in which the reference oxidation is the conversion of Me2SO to Me2SO2. In all cases, unfilled coordination sites on the 6-coordinate Fe centers were filled with additional waters. Table 2, Figure 11-12 10 and Appendix A show the optimized structures of ligated Fe and Fe O complexes. The II IV 2+ structures presented in our work represent the lowest-energy conformations for each structure. All Fe(IV) structures were modeled as high-spin quintets (singly occupied dxy,dxz,dyz,dx2-x2) and all Fe(II) were modeled as high-spin quintets (doubly occupied dxy). Since DFT is a single-determinant method, the high-spin configurations are the most reliable states. 1.5 Results and Discussion 1.5.1 Identification and Quantification of Products Figure 1 shows the positive and negative mode ESI mass spectra obtained from (Me2SO + Fe(ClO4)2 + CsClO4) aqueous microdroplets at pH 4.5 before and after exposure to O3(g). In the absence of O3, signals appear at m/z+ = 133, 79, m/z- = 99, 101, which correspond to Cs+, Me2SOH+ and 35,37ClO4- , respectively. Upon O3 exposure, new signals appear at m/z+ = 95 (Me2SO2H+), m/z- = 95 (MeSO3-) and m/z- = 97 (HSO4-). The relative ratios of MeSO3and HSO4- signal intensities: I-95/I-97 > 1 (see Figure 13), are consistent with our previous study at pH 7.43 We verified that the Me2SO2 and MeSO3- products are inert in these experiments. Me2SO2 was also produced in the uncatalyzed ozonolysis of Me2SO in the absence of Fe(ClO4)2, as reported in the literature.44 Signal intensities were converted to concentrations using Me2SO2, MeSO3Na and HSO4Na calibration curves using CsClO4 and NaClO4 as the internal standards (Figure 8-10). We note that this procedure is approximate because the mass signals detected in our gas-liquid experiments correspond to species formed at the air-water interface, whereas calibration curves are based on mass signals from Me2SO2H+, MeSO3- and HSO4- ions present in bulk aqueous solutions. This procedure is correct to the extent that Me2SO2H+, MeSO3-, HSO4-, and internal reference Cs+ and ClO4- ions have similar surface affinities.45 The dependences of Me2SO2H+/Cs+, MeSO3-/ClO4- and HSO4-/ClO4- ratios as functions of experimental parameters are, of course, independent of such an assumption. The key result is that the main product of the Fenton oxidation of Me2SO on microdroplets is Me2SO2, which is produced in mM concentrations vs MeSO3- in µM concentrations. Figure 18a-c show Me2SO2H+, 11 - - II MeSO3 and HSO4 signal intensities as functions of initial Fe and O3 concentrations. Plotted Me2SO2H+ signal intensities, I95+ (Figure 14a), correspond to I95+ signals obtained in the presence of FeII minus those measured in the uncatalyzed oxidation of Me2SO by O3 (see Figure 15 for raw, uncorrected Me2SO2H+ signals). 1.5.2 Rates and Product Yields at the Air-Water Interface vs Bulk Water The formation of Me2SO2 (detected as Me2SO2H+) is consistent with one-step O-atom transfers from O3 or FeIVO2+ to Me2SO (reaction R3) (Figure 2). The formation of MeSO3and HSO4 from Me2SO, on the other hand, is a multi-step process involving the splitting of one and two Me groups via C-S bond fissions, reactions R4 and R5.43 Leaving methyl groups are formed by breaking the weak C-S bonds of the sulfur-centered free radicals produced from ⋅OH radical additions to reduced S atoms.43 ⋅OH radical addition reactions to S-atoms are very fast, with diffusion-controlled rate constants k4 ∼ k5 ∼ kDMSO+OH = 4.5 ×109 M-1 s-1 in bulk water.46 Therefore, if all of FeIVO2+, and most of the ⋅OH radicals produced in R1 react with Me2SO (see below), from: Me2SO2/(MeSO3- + HSO4-) < Me2SO2/MeSO3- = I95+/I95-= (k3/k4)[FeIVO2+]/[⋅OH], and the k3/k4 ≤ 1 condition, the observed I95+/I95- >> 1 ratios imply that [FeIVO2+] >> [⋅OH], that is: k2 >> k1. That both k4 ∼ k5 are diffusion-controlled is consistent with the rapid consumption via R5 of the undetected methyl sulfinate MeSO2- intermediate produced in R4. In other words, the rate of formation of MeSO3- is in fact controlled by R4. The Me2SO2H+/MeSO3- ratios, evaluated from I95+/I95- values measured in gas-liquid experiments, as functions of [Fe2+] at various O3(g) concentrations are shown in Figure 14. We found that ∼1.5 mM Me2SO2 is produced from pH 4.5 (1 mM Me2SO + 100 µM Fe(ClO4)2) solutions exposed to 113 ppm O3 (equivalent to ≤ 0.4 mM O3(aq) under equilibrium) for < 1 ms. The direct and inverse dependences of Me2SO2 and MeSO3- concentrations on [Fe2+], Figure 14a-b, suggest that a small fraction of the ⋅OH radicals produced in R1 reacts further with Fe2+ in competition with Me2SO, reaction R4. Note that the minor role assigned to ⋅OH radicals in these experiments is valid because the extrapolated MeSO3- concentrations at [Fe2+] = 0 still are in the µM scale (vs mM Me2SO2 concentrations). Me2SO2 and MeSO3- are produced in mM and µM 12 concentrations in experiments exposing pH 4.5 Fe(ClO4)2 in aqueous microdroplets to O3(g) for < 1 ms. The results imply that FeIVO2+ ions are produced in 300-1600 larger yields than ⋅OH radicals on the surface of aerosol microdroplets at typical cloud water pH and iron concentrations. To investigate the course of Fenton chemistry in bulk water, we rapidly mixed (Fe2++ Me2SO) solutions with O3(aq) solutions obtained by previously dissolving O3(g) in water. In contrast to the gas-liquid experiments, we found that only ≤ 0.1 mM Me2SO2 was produced by mixing pH 4.5 (10 mM Me2SO + 100 µM Fe(ClO4)2) with 10.6 mM O3(aq) solutions for 12 minutes (See Figure 16-1.17). This result is consistent with previously reported kinetic values in bulk water (Kinetic Analysis in the SI).44 Thus, under our experimental conditions, interfacial Fenton chemistry as catalyzed by Fe(H2O)62+ ions on microdroplets largely proceeds via the ferryl in, FeIVO2+, pathway at a rate that is 104 times faster than in bulk water. Figure 3 shows the structures of fully hydrated Fe2+ and FeO2+ ions with various ligands. It is apparent that the apical water of hexaquo Fe2+ ion needs to be replaced before (or during) an O-atom can be transferred from O3 to produce the FeO2+ ion via an outer-sphere process. This process would be facilitated if the hydration shell of Fe2+ at the air-water interface presented a vacancy, preferentially oriented to the air phase that could accept an O-atom without paying an enthalpy penalty of dehydration.35 1.5.3 Ligand Effects We tested the effects of environmentally relevant ligands (L) on the rates and products of interfacial ozone Fenton chemistry. Experiments involved (1 mM Me2SO + 100 µM Fe(ClO4)2 + 200 µM L) microdroplets at pH 4.5, except as indicated, exposed to O3(g) for < 1 ms. The mass spectrum recorded in a similar experiment is shown in Figure 18. Ligands were found to be inert under our experimental conditions. For example, we could not detect any new species from the oxidation of EDTA itself in these experiments. None of the ligands used in our experiments were expected to react with FeIVO2+.17, 47 Furthermore, none of the ligands at 200 µM is expected to significantly compete with 1 mM Me2SO for ⋅OH radicals 13 either, because rate constants for ⋅OH radical reactions with most of the ligands used in our experiments are at least 10 times slower than kDMSO+OH = 4.5 ×109 M-1 s-1.46 We found that the nature of the ligands has a significant effect on the dimethyl sulfone Me2SO2 concentrations (Figure 4), and a weaker effect on methanesulfonate MeSO3- yields (Figure 19). Figure 4 shows the dependence of Me2SO2 yields as a function of the calculated reduction potentials of the complexed iron couples: E0(LFeIVO2+ + O2 = LFe2+ + O3) . The volcano-type dependence of Me2SO2 yields as a function of E0 suggests the competition of opposite effects. A positive effect is most likely associated with faster rates of formation of the LFeIVO2+ intermediates with stronger Fe=O bonds expected at larger E0 values. Negative effects arise from the slower O-atom transfers expected from stronger Fe=O bonds to Me2SO. Me2SO2 yields are, therefore, expected to peak at an intermediate E0 value, which in our experiments corresponds to the ligand L = oxalate (Figure 4). This explanation is consistent with previous studies for which the reactivity of ferryl ion intermediates was linked to the electron affinity of the iron-bound oxygen derived from DFT calculations.48 The data of Figure 4 were obtained at pH 4.5, the pH of 100 µM FeII solutions in equilibrium with atmospheric CO2. Lower pH values are likely under atmospheric conditions.49 Figure 5 shows the effect of [EDTA] on Me2SO2 yields at pH 1, 2 and 4.5. The successive pKa’s for H4EDTA are 2.0, 2.7, 6.2 and 10.3. At pH 4.5, H2EDTA2- complexation has the net effect of increasing the reduction potential of Fe(IV) to Fe(II) as shown in Figure 4. This results in a decrease in the yield of the ferryl ion pathway as shown in Figure 5. Note that at pH 2.0, H3EDTA- complexation leads to an increase in the yield of the ferryl ion pathway. We ascribe this result to the effect of differently protonated ligands on Fe(IV) and Fe(II) reduction potentials. Under atmospheric conditions, aerosols typically are comprised of mixtures of water with organics. Figure 20 shows that EDTA has minor effects in catechol/water and acetonitrile/water mixtures at pH 4.5. The lack of significant effects in mixed media suggests that EDTA is preferentially solvated by the organic solvents while FeII and ferryl ion remain in the aqueous microphase.50 14 II The Fenton reaction between interfacial Fe ions and gas-phase ozone at the air-water interface of aqueous microdroplets mainly produces FeIVO2+ via reaction R2 at a rate ∼104 faster than the similar reaction in bulk water. The presence of complexing ligands affects both the relative and absolute rates of the ferryl ion and hydroxyl radical pathways. Dimethyl sulfoxide, Me2SO, is an important intermediate in the tropospheric oxidation of biogenic Scontaining gases. We have shown that the yields of dimethyl sulfone Me2SO2, the product of LFeIVO2+ + Me2SO ⟶ reactions, were found to depend on the nature of L according to a volcano plot function of the E0 (LFeIVO2+ + O2 ⇌ LFe2+ + O3) reduction potentials as calculated by DFT with a peak at L ≡ oxalate. The volcano dependence of Me2SO2 yields reflects competing effects on the formation of Me2SO2 resulting from the influence of L on the strength of the FeIV=O bond. The stronger FeIV=O bonds produced at faster rates in reaction R2 become weaker O-atom donors to Me2SO. Results show that ozone Fenton chemistry on aqueous microdroplets proceeds ∼104 faster than in bulk-phase water due to the production of reactive ferryl ion (FeIVO2+) intermediates that, in turn, depend on the specific nature of the available complexing ligands. 1.6 Summary The Fenton reaction between interfacial FeII ions and gas-phase ozone at the air-water interface of aqueous microdroplets mainly produces ferryl FeIVO2+ ions, reaction R2, ∼ 104 faster than the similar reaction in bulk water. Ligands L affect both the relative and absolute rates of the ferryl and hydroxyl pathways. In the oxidation of dimethyl sulfoxide Me2SO, an important intermediate in the tropospheric oxidation of biogenic S-containing gases, the yields of dimethyl sulfone Me2SO2, the product of LFeIVO2+ + Me2SO reactions, were found to markedly depend on the nature of L according to a volcano function of the E0(LFeIVO2+ + O2 = LFe2+ + O3) reduction potentials calculated by DFT, and peak at L ≡ oxalate. The volcano dependence of Me2SO2 yields reflects competing effects on the formation of Me2SO2 resulting from the influence of L on the strength of the FeIV=O bond. The stronger FeIV=O bonds produced at faster rates in reaction R2 become weaker O-atom donors to Me2SO. Present results show that ozone Fenton chemistry proceeds on aqueous microdroplets about 104 faster than in bulk water, mainly produces reactive ferryl FeIVO2+ ion intermediates, and 15 strongly depends on typical organic ligands. These are relevant issues in the heterogeneous chemistries of atmospheric aerosol and cloud microdroplets. 1.7 Supporting Information 1.7.1 Online Electrospray Ionization Mass Spectrometry In our gas-liquid experiments, the products generated in the microdroplets and the outermost layers of the microjets are analyzed without manipulation by online electrospray ionization mass spectrometry (ESI-MS).37, 51 The production of charged microdroplets requires the conversion of the kinetic energy of the fast sheathing nebulizer N2 stream into electrostatic energy associated with the separation of anions from cations and that associated with the creation of additional surface.36 Charge separation is a direct result of the nebulizer gas because ion signals (1) increase at higher nebulizer gas velocity and (2) extrapolate to zero as nebulizer gas velocity drops to zero.36, 51. The polarizing external electric field required to deflect the charged microdroplets does not affect the observed interfacial chemistry, as shown, e.g., in the interfacial α-tocopherol ozonation reaction in the positive and negative modes.52 This is consistent with the fact that the calculated electric field at the microjet tip is 2 orders of magnitude weaker than the electric field naturally present at the air-water interface.53-54 The surface specificity of online ESI-MS has been demonstrated in several reports from our laboratory.37, 55 1.7.2 Quantification of Ozone Concentration in Spray Chamber Quantification of interfacial ozone concentration assumes a well-mixed spray chamber with an O3/O2(g) inlet at 50 cm3/min and a jet of nebulizer gas stream through an annulus with an inner diameter of 100um and an outer diameter of 200um at sonic velocity. Ozone generator settings at 1200, 2700 and 3000 ppm in the O2/O3(g) stream correspond to 113, 254 and 283 ppm in the spray chamber, respectively. 1.7.3 Kinetic Analysis of Me2SO(aq) Ozonation in Bulk Water In this experiment, Me2SO(aq) and appropriate internal standards are dissolved in an aqueous solution in syringe A, while saturated O3(aq) aqueous solution in equilibrium with 3000ppm O3(g) is prepared in syringe B. Syringe A and B are each pushed at 15 μL/min into a manifold 16 tank reactor of volume 373 μL. The outlet of the manifold tank reactor connects to the ESI-MS microjet via PEEK capillary tubing of negligible internal volume. Levanov et al.56 reported the Henry’s Law constant for ozone HO3 as a function of temperature 𝑙𝑙𝑙𝑙𝐻𝐻𝑂𝑂3 = −(9.445) + 2406 𝑇𝑇 𝐻𝐻𝑂𝑂3 = 86.8 𝑎𝑎𝑎𝑎 25 ℃ Calculate the saturation concentration of O3(aq) in equilibrium with 3000ppm O3(g) �𝑂𝑂3 𝑎𝑎𝑎𝑎 � = 0.0106 𝑀𝑀 In the tank reactor, the concentration of O3(aq) is diluted by half, resulting in an initial reaction concentration of �𝑂𝑂3𝑎𝑎𝑎𝑎 � 2 = 0.00532𝑀𝑀 = 255 𝑚𝑚𝑚𝑚/𝐿𝐿 At this ozone concentration, the ozonation Me2SO follows zero-order kinetics with a rate constant k = 0.10 mM/min.44 We can calculate this rate constant from our bulk-phase experimental measurements. From Figure 8, we obtain the Me2SO2 concentration from direct ozonation to be 2.335 mM. We calculate the reaction time using a continuously stirred tank reactor model with a total volume of 373 µL and a total flow rate of 30 μL/min, and the residence time is 𝑡𝑡 = Hence the zero-order rate constant is 373 = 12.43 𝑚𝑚𝑚𝑚𝑚𝑚 30 2.335 𝑘𝑘0 = = 0.0188 𝑚𝑚𝑚𝑚/𝑚𝑚𝑚𝑚𝑚𝑚 12.43 17 Comparing with the literature reported value of 0.10 mM/min,44 this confirms the slow bulk phase kinetics compared to interfacial kinetics where 2 mM Me2SO2 are formed from Me2SO ozonolysis within 1 ms (Figure 16 y-intercept). 18 1.8 Figures Figure 1. Positive (a) and negative (b) online electrospray ionization mass spectra of pH 4.5 (1mM Me2SO + 100µM Fe(ClO4)2 + 50µM Cs+ + 250µM ClO4-) aqueous 19 + microdroplets before (blue) and after (red) exposure to 283ppm O3(g) for < 1 ms. Cs and ClO4- non-complexing ions added as internal standards. 20 Figure 2. Mechanism of the Fenton oxidation of Me2SO into Me2SO2, MeSO3- and HSO4-. Me2SO2 is the only product of the ferryl ion pathway (bottom left). 21 Figure 3. Structures of charge-neutral FeIVO and FeII species. Ligands in blue. Structures do not report actual coordination, stereochemistry, denticity, or hapticity. For DFToptimized structures, see Figures 11-12. 22 Figure 4. Ligand effects on the formation of dimethyl sulfone Me2SO2. [Me2SO2] vs E0(LFeIVO2+ + O2 → LFe2+ + O3) DFT formal reduction potentials calculated from ∆G(LFeIVO2+ → LFe2+) Gibbs free energies differences relative to ∆G(Me2SO → Me2SO2). The inset shows E0(LFeIVO2+ + O2 → LFe2+ + O3) values for the different ligands L. [Me2SO2] determined at the air-water interface of pH 4.5 [1mM Me2SO + 100 µM Fe(ClO4)2 + 200 µM ligand L + 150 µM NaClO4] aqueous microdroplets exposed to 283 ppmv O3(g) for < 1 ms (see text for more details). Because no Cs+ is used as the internal standard in the positive mode, the absolute concentration of [Me2SO2H+] measured has higher uncertainty than Figure 14, though the trend is unaffected. Me2SO2 concentration is measured as Me2SO2H+. Due to unavoidable complex formation between the negatively charged ligands and Me2SO2H+ cation, the yield measured is an underestimate of the actual yield. DMA = dimethylamine, TMA = trimethylamine, CAT = catechol, NAP = naphthoquinone, NLI = no ligand, PBQ = p-benzoquinone, MAL = malonate, CHL = chloride, OXA = oxalate, TAR = tartrate, EDT = ethylenediaminetetraacetate. 23 Figure 5. Protonated dimethyl sulfone Me2SO2H+ (m/z+ = 95) mass signal intensities vs [EDTA]/[Fe(II)] at various pH values. Me2SO2H+ mass signal intensities are measured by online electrospray ionization mass spectrometry at the air-water interface of [1mM Me2SO + 100 µM Fe(ClO4)2] aqueous microdroplets exposed to 283 ppm O3(g) for < 1 ms. The mass signal intensities measure the net yield of Me2SO2 from Fenton chemistry and direct ozonolysis of Me2SO, after subtracting the Me2SO2H+ complexed by EDTA. No Fe(II) means experiments in the absence of EDTA or FeII. 24 Figure 6. Experimental setup of online ES-MS. The aqueous reactants and gaseous reactants contact at the outlet tip of the microjet. 25 Figure 7. Experimental setup of offline ES-MS. The aqueous reactants contact each other in the manifold tank reactor before entering the microjet. 26 Figure 8. Me2SO2 calibration curve using 1, 10 and 100 mM Me2SO2. Y-axis plots the mass signal intensity ratios of Me2SO2H+/Cs+ at pH = 4.5 and [Cs+] = 50 µM. 27 Figure 9. MeSO3-calibration curve using 1, 10 and 100 µM MeSO3Na. Y-axis plots the mass signal intensity ratios of MeSO3-/ClO4- at pH = 4.5 and [ClO4-] = 250 µM. 28 Figure 10. HSO4- calibration curve using 1, 10 and 100 µM NaHSO4. Y-axis plots the mass signal intensity ratios of HSO4-/ClO4- at pH = 4.5 and [ClO4-] = 250 µM. 29 No Ligand: Dimethylamine: Trimethylamine: Chloride: Oxalate: 30 Malonate: Tartrate: Catechol: p-Benzoquinone: 1,4-naphthaquinone: 31 EDTA: Figure 11. Optimized structures of ligated Fe2+ and FeO2+ complexes from DFT calculation. Figure 12. Optimized structure of R3 transition state. 32 Figure 13. Relative product concentration of MeSO3- and HSO4- in interfacial ozoneFenton chemistry for 1mM Me2SO and 100 µM Fe(ClO4)2. 33 Figure 14. Product concentrations of (a) Me2SO2 (b) MeSO3- and (c) HSO4- at different initial FeII bulk concentration ([FeII]) and gaseous O3 concentration at pH 4.5 resulting from ozone Fenton oxidation of 1mM Me2SO. Concentrations on the y-axis are the corresponding bulk concentrations of Me2SO2, MeSO3- or HSO4- in hypothetical equilibrium with the sampled surface concentrations. The product concentrations are used to calculate the ferryl:hydroxyl pathway product concentrations ratio, represented as (d) the cumulative ferryl:hydroxyl intermediate. CsClO4 and NaClO4 are added as the internal standards to a total concentration of 50 µM Cs+ and 200 µM ClO4-. 34 Figure 15. Me2SO2 product concentration at the air-water interface for 1mM Me2SO and 100µM Fe(ClO4)2 including Me2SO2 produced from Me2SO direct ozonolysis. The yintercepts show the product concentration from direct ozonation of Me2SO, and the slope shows the product concentration from ferryl O-atom transfer. 35 Figure 16. Me2SO2 product concentration in the bulk phase reaction from 10mM Me2SO, 5mM O3 and 100µM Fe(ClO4)2 reacting for 746 seconds, including Me2SO2 produced from Me2SO direct ozonolysis. The y-intercepts show the product concentration from direct ozonation of Me2SO, and the slope shows the product concentration from ozone-Fenton chemistry, including but not limited to O-atom transfer from ferryl species. The Me2SO2 product concentration originating from 100µM Fe2+ sets the upper bound for ferryl O-atom transfer kinetics. 36 Figure 17. MeSO3- product concentration in bulk from 10mM Me2SO, 5mM O3, and 100µM Fe(ClO4)2 reacting for 746 seconds. The y-intercepts show the product concentration from direct ozonation of Me2SO, and the slope shows the product concentration from ozone-Fenton chemistry, including both ferryl and OH pathways. 37 Signal Intensity [a. u.] 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 150 160 170 180 190 200 210 m/z Figure 18. 5mM EDTA effect on the ozone Fenton chemistry between 10mM FeCl2(aq) and 283ppm O3(g). Blue denotes without EDTA and red with EDTA in the reaction mixture. The iron peaks are FeIICl3- most visible at m/z=161/163/165, FeIVOCl3- most visible at m/z=177/179/181, and FeIIICl4- most visible at m/z=196/198/200/202. The signal intensities are lowered for all three species with EDTA added, showing that ligands compete with Cl- for the coordination sphere of Fe2+ ions. 38 Figure 19. Methanesulfonate formation [MeSO3-] vs E0(Fe3+ + e- ⟶ Fe2+) reduction potentials. The inset shows calculated E0(Fe3+ + e- ⟶ Fe2+) potentials as functions of ligand. Standard reduction potentials of Fe3+ to Fe2+ in the presence of ligands shown on the x-axis of the inset are from Martell and Smith.10 Reactants are 1mM Me2SO, 100 µM Fe(ClO4)2, 200 µM ligand L, 283ppm O3 at the interface of gaseous O3 and aqueous microdroplets. Abbreviations: DMA = dimethylamine, TMA = trimethylamine, CAT = catechol, NAP = naphthoquinone, NLI = no ligand, PBQ = p-benzoquinone, MAL = malonate, CHL = chloride, OXA = oxalate, TAR = tartrate, EDT = ethylenediaminetetraacetate. 39 Figure 20. Ligand effect on interfacial Me2SO2 product concentration in 25 weightpercent catechol/water and 50 volume-percent acetonitrile/water solvents. The concentrations of Me2SO and Fe(II) are fixed at 1mM and 100µM, respectively. O3(g) is at 283 ppm. 40 1.9 Tables Ligand MeSO3-/HSO4- catechol 1.110045 EDTA 3.297675 malonate 1.774516 oxalate 1.867809 tartrate 1.966674 ClO4- 1.116289 Cl- 15.35224 p-benzoquinone 2.344543 1,4-naphthaquinone 0.86446 dimethylamine 2.680139 trimethylamine 0.758585 Table 1. Ligand effect on the relative product concentrations of MeSO3- and HSO4-. ClO4denotes no ligand as perchlorate does not ligate Fe2+. 41 Ligand Fe=O Fe(IV)-Ligand Fe(II)-Ligand distance (Å) distance(s) (Å) distance(s) (Å) No Ligand 1.65 - - Dimethylamine 1.65 2.05 2.18 Trimethylamine 1.64 2.17 2.17 Chloride 1.64 2.43 2.63 Oxalate 1.65 1.90, 1.93 1.98, 1.99 Malonate 1.64 1.92, 1.96 1.92, 1.96 Tartrate 1.64 1.92, 1.96 2.08, 2.23 Catechol 1.66 1.89, 1.91 1.98, 1.99 p-benzoquinone 1.65 2.16 2.01 1,4-naphthaquinone 1.65 2.12 2.03 EDTA Fe-N: 2.26, 2.33 Fe-N: 2.20, 2.22 Fe-O: 1.95, 1.96 Fe-O: 1.91, 1.90 1.63 Table 2. Important bond distances for the Fe systems investigated. All distances are in Å. Some structures contain multiple Fe-Ligand distances due to increased denticity. Structures are depicted in Figure 4. 42 1.10 Acknowledgements This project was supported by the U.S. National Science Foundation, Grant AGS-1744353. Special thanks are extended to Drs. Nathan Dalleska and Sara Lijie Li for their assistance. 43 Chapter 3 SARS-COV-2 INDIRECT AIRBORNE TRANSMISSION MODEL Chapter III: SARS-CoV-2 Indirect Airborne Transmission Model Reproduced from the article below with permission from the Royal Society of Chemistry. Further permission related to this material should be directed to the Royal Society of Chemistry: https://pubs.rsc.org/en/Content/ArticleLanding/2022/EA/D1EA00013F Gu, A. Y.#, Zhu, Y.#, Li, J., & Hoffmann, M. R. (2022). Speech-Generated Aerosol Settling Times and Viral Viability Can Improve COVID-19 Transmission Prediction. Environmental Science: Atmospheres. doi:10.1039/d1ea00013f. 2.1 Abstract Droplets during human speech are found to remain suspended in the air for minutes, while studies suggest that the SARS-CoV-2 virus is infectious in experimentally produced aerosols for more than one hour. However, the absence of a large-scale association between regional outbreaks and weather-influenced virus-laden speech-generated aerosol characteristics such as settling time and viral viability makes it challenging for policy making on appropriate infection control measures. Here we investigate the correlation between the timeseries of daily infections and of settling times of virus-containing particles produced by speaking. Characteristic droplet settling times determined by the Stokes-Cunningham equation as influenced by daily weather conditions were estimated based on local meteorological data. Daily infection data were calibrated from local reported cases based on established infection timeframes. Linear regression, vector autoregression, simple recurrent neural network, and long short-term memory models predict transmission rates within one-sigma intervals using the settling times and viral viability over 5 days before the day of prediction. Corroborating with previous health science studies, from the perspective of meteorology-modulated 44 transmission, our results strengthen that airborne aerosol transmission is an important pathway for the spread of SARS-CoV-2. Furthermore, historical weather data can improve the prediction accuracy of infection spreading rates. 2.2 Introduction The novel coronavirus (SARS-CoV-2) has caused more than 240 million infections and 4.8 million deaths globally from COVID-19 as of October 19, 2021.57 COVID-19 is known to cause considerable asymptomatic infections. Therefore, the ability to predict local COVID19 outbreaks is imperative for effective public health management.58 Faster flu transmission during winter months is often linked to lower temperatures and relative humidity than occur during the summer.59 Virus-laden aerosols from infected human hosts evaporate into smaller aerosol particles at lower humidity and as a result, they take longer to settle out of the atmosphere. In addition, viruses in aerosols survive longer at lower ambient temperatures, and thus, they remain contagious for longer periods of time while airborne.60 Speechgenerated aerosols may be suspended in air for 8 to 14 minutes61, while viruses encapsulated in aerosol droplets could remain viable for 49 hours62-63. Thus speech-generated aerosols are widely considered to have contributed to asymptomatic transmission of COVID-19.61, 64-65 The fate and transport of these virus-laden aerosol droplets could be used for predicting the spread of COVID-19. Airborne transmission of COVID-19 has been studied extensively over the past year.66-67 Previous studies on predicting COVID-19 transmission and similar airborne transmission diseases were focused on using an infected population (SIR model)68 or meteorological observation69 directly as the input variables when predicting COVID transmission. Considering the non-linear relationships connecting weather to settling time and viral viability63, 70, using weather-derived settling times and viability as input variables may improve the goodness of fit as well as elucidating additional factors affecting airborne transmission. Meteorological conditions such as temperature and humidity affect aerosol settling velocity by affecting the final size of aerosols after equilibration with ambient moisture through 45 evaporation or condensation. The settling velocity of the equilibrated aerosols in the atmosphere is often calculated using Stokes’ Law71, which has been traditionally used to estimate aerosol terminal velocity at ambient temperatures and pressures. Because it assumes no-slip boundary condition, it underestimates the terminal settling velocity for small particles of size < 1 μm. In air at 25 °C, the terminal velocity accounting for slip correction is 1.24 times faster than calculated from uncorrected Stokes’ Law for a 1 μm-diameter particle, and 2.2 times faster for a 200 nm-diameter particle. Stokes’ Law also assumes that aerodynamic stress is transferred primarily through viscous exchange, meaning it is valid for small Reynolds number Re < 1. Cunningham later introduces a correction factor to account for particle surface slippage and the resultant Stokes-Cunningham Law applies for aerosols sizes as small as 100 nm at ambient temperature and pressure.72 Other models, such as the one proposed by Epstein73 and Millikan74, are only applicable at Knudsen numbers Kn > 10, corresponding to nm-sized particles in the lower troposphere or micron-sized particles at millibar-level pressures75. In addition to settling time, weather also affects the viability of viruses in suspended aerosols.76 In the case of SARS-CoV-2, high temperature, relative humidity (within 20% 70% range) and Ultraviolet B (UVB) light produce higher decay rates,63 which is in agreement with previous studies on an enveloped virus77. In a study focused on the viability of SARS-CoV-2 on surfaces, investigators reported an extension of viability over longer times at low temperatures and humidities.78-79 Weather also affects influenza A virus viability, though the relationship depends on the specific solution medium.80 Given aerosol settling times and viral viability as the input variables, COVID-19 cases can be forecasted using regression analysis or machine learning models. Regression analyses such as linear regression and vector autoregression can identify key input variables among all the input variables but are limited to linear correlations only.81-82 Machine learning algorithms can find highly non-linear correlations, but they do not reveal any intuitive relationship between the input and response variables. Machine leaning has been introduced 46 as a promising alternative to existing forecasting models for influenza and SARS-CoV83 284 with temperature, humidity and sunlight intensity as input variables. Herein, we test the model fitting and prediction performance of the transmission rate of COVID-19 in the US using the settling times of speech-generated aerosols coupled with viral viability data. In order to achieve this goal, weather information, evaporated speech aerosol settling times, and viral viability are processed in regression and recurrent neural network (RNN) models to forecast SARS-CoV-2 daily transmission rates. We compared linear regression, vector autoregression (VAR), simple RNN and long-/short-term memory (LSTM) RNN in terms of prediction performance of COVID-19 transmission. We expect that inclusion of first principles such as the Köhler equation for vapor pressure reduction on aquated aerosol size and settling velocity calculation improvements should removes some of the non-linearity that models need to accommodate in order to achieve better fitting and forecasting performance. A good model fitting and prediction performance would indicate that speech-generated airborne aerosols are a significant transmission route for COVID-19 and that the weather-affected speech-generated aerosol properties may be incorporated to assist further predictive model development. 2.3 Methods Figure 21a shows the data flow of the model from weather data to predicted SARS-CoV-2 transmission in this work. Each section of the model is elaborated in this section. 2.3.1 Data Mining Five counties were selected for inclusion in our model development. They are Harris County, TX, King County, WA, Los Angeles County, CA, Maricopa County, AZ, and Santa Clara County, CA. The counties are representative of the top-20 most populated counties in the United States. Of the 5 counties selected none had zero-case days throughout April 2020. They also had moderately warm weather and no temperature below 0°C. When temperatures are below 0°C, additional data on water surface tension and sodium chloride solution partial molal volumes below normal melting point are needed. Constraining the predictive model to T > 0 oC avoids the complication of ice crystal formation within aquated aerosols85. The daily 47 local meteorological data, including daily average temperatures and relative humidities (RH) were obtained online from National Oceanic and Atmospheric Administration (NOAA) from 1 April to 29 August 2020. For counties with more than one station, the station with most data coverage for daily temperature and RH was chosen. The station numbers are 12960, 24233, 93134, 23183, and 23293 for Harris County, King County, Los Angeles County, Maricopa County, and Santa Clara County, respectively. The county-level COVID-19 confirmed case counts were obtained from USAFacts.org, who collected data from the Centers for Disease Control and Prevention (CDC), and the corresponding state- and local-level public health agencies. Data was acquired on 14 September 2020 and contained up-to-date daily confirmed cases. Given the extended asymptomatic period of COVID-19, the daily confirmed cases data was processed to reflect the daily active cases based on a disease progression timeline (Figure 21b) that summarizes information provided by the CDC86. The daily active cases of a certain day to study is therefore the sum of daily confirmed cases for the past 12 days and future 4 days. 2.3.2 Aerosol Settling Behavior Given the fast kinetics of water evaporation from micron-sized aerosols (seconds)87-88 compared to their settling time from a typical human height (minutes),89 the Köhler equation (Equation 1) is used to estimate the size of evaporated aerosols:70 ln ℎ = ξσ,0 𝑐𝑐 3 − 𝜉𝜉 3 𝜉𝜉 (1) 𝑀𝑀 𝜌𝜌 1 3 where ℎ is RH in decimal, 𝑐𝑐 = �𝜈𝜈 � 𝑀𝑀𝑤𝑤 � � 𝜌𝜌𝑑𝑑𝑑𝑑𝑑𝑑 �� = 1.10 for sodium chloride solution, 𝜉𝜉 = 𝑟𝑟 𝑟𝑟𝑑𝑑𝑑𝑑𝑑𝑑 𝑠𝑠 𝑤𝑤 𝑟𝑟 is the ratio of dry salt diameter to wet aerosol diameter, 𝜉𝜉𝜎𝜎,0 = 𝑟𝑟 𝜎𝜎,0 is the ratio of the 𝑑𝑑𝑑𝑑𝑑𝑑 characteristic length scale of Kelvin effect to dry salt diameter where the characteristic length scale is calculated as follows: 𝑟𝑟𝜎𝜎,0 = 2𝜈𝜈 ����𝜎𝜎 𝑤𝑤 𝑅𝑅𝑅𝑅 48 (2) in which 𝜈𝜈 ���𝑤𝑤� is the partial molal volume of water in the solution, 𝜎𝜎 is the surface tension of the solution-air interface, 𝑅𝑅 is the gas constant and 𝑇𝑇 the absolute temperature. The speech-generated aerosols are modelled as sodium chloride solutions at physiological concentration of 80 mM, which is a typical salivary sodium concentration90. The initial size of speech-generated aerosols before evaporation is taken as 6 µm, which is the most abundant size according to experimental measurements91. The partial molal volume of water in a sodium chloride solution,92 water vapor pressure,93 water surface tension94, and the binary diffusion constant of water vapor in air95 are taken from previous experimental data or semiempirical relationships. The settling velocity of the evaporated aerosol of a given size is calculated using the Stokes’ Law with the Cunningham correction factor shown in Equation 3: 𝑉𝑉𝑡𝑡 = 𝜌𝜌𝑝𝑝 𝐷𝐷𝑝𝑝2 𝑔𝑔 18𝜇𝜇 × 𝐶𝐶𝑐𝑐 (3) where 𝑉𝑉𝑡𝑡 is the terminal settling velocity, 𝜌𝜌𝑝𝑝 is the particle density, 𝐷𝐷𝑝𝑝 is the particle diameter, 𝑔𝑔 is the gravitational acceleration, mu is the viscosity of air, and 𝐶𝐶𝑐𝑐 is the Cunningham correction factor calculated as follows: 𝐶𝐶𝑐𝑐 = 1 + 2.52 𝐾𝐾𝐾𝐾 (4) where 2.52 is an empirical constant specific to air, and 𝐾𝐾𝐾𝐾 is the Knudsen number, which is the ratio of the mean free path of the gas molecules (𝜆𝜆) and the aerosol diameter (𝐷𝐷𝑝𝑝 ) as shown in Equation 5. 𝐾𝐾𝐾𝐾 = 𝜆𝜆 𝐷𝐷𝑝𝑝 Assuming ideal gas law, the mean-free path, λ, for a given gas is (5) 𝜆𝜆 = 1 𝑁𝑁 √2𝜋𝜋𝑑𝑑2 � 𝑉𝑉 � 49 (6) where d is the Van der Waals diameter of the gas molecule (3.10 × 10−10 𝑚𝑚 for N2), and is the molecular density of gas (2.46 × 10 for air at 25 °C. 25 𝑁𝑁 𝑉𝑉 at 25 °C and 1 atm total pressure). λ = 95 nm From the aerosol settling velocity, the settling time is calculated assuming aerosols attain their terminal settling velocity immediately after release at a height of 1.5 meters. Because the settling time is used as an intermediate variable in the model depicted in Figure 21 to check fitting and make predictions, the absolute height of release does not affect conclusions obtained. 2.3.3 Viral Viability Viral viability is calculated using empirical linear regression with interaction by Paul Dabisch63. Because the regression equation is obtained from a limited range of temperature (10 - 30 °C) and humidity (20 - 70 %), we focus on counties with moderate climate where the viability calculation is valid. 2.3.4 Transmission Model The variable describing SARS-CoV-2 transmission is the “new case percentage increase (NCP),” which is calculated as the number of new positive tests on a particular day divided by the “total number of active cases (TNAC)” on that day. The TNAC on a day is estimated by summing all positive tests from 12 days before until 4 days after the day of interest as stated above. The timeseries data for each county are separated into a training set and a test set, with the test data set containing the last 4 days of data and the training set containing the remaining data. VAR and RNN models are developed using the training data. Subsequently, the predictive accuracy of the trained models is tested using the test data. 50 Linear regression analysis uses the settling times and viral viability between the day of interest and 5 days before as the input variables (total of 10). VAR uses the settling times, viral viability, and “new case percentage increase” between 1 day and n days prior to the day of interest as the input variables, where n is the order of VAR and selected by Akaike's Information Criterion. As an autoregressive algorithm, predictions of more than one day in the future are calculated using the predictions of previous days, not the actual data as in the linear regression or RNN models. Simple RNN uses the same input variables as the linear regression model, one hidden layer of 70 nodes, a max epoch of 105 and a learning rate of 10-4. LSTM uses the same input variables as RNN, one LSTM layer of 120 units, a max epoch of 106 and a batch size of 72. All models use the new case percentage increase on the day of interest as the response variable, which represents the transmission rate. 2.4 Results and Discussion In order to investigate the gravity settling of the speech-generated droplets, the settling velocity and dimensionless numbers of the Stokes-Cunningham modification were estimated for droplets of 6 μm size (Figure 27), which is used as the peak initial size of speechgenerated droplets91. It should be noted that this size is comparable to the average diameter of cough-generated droplet size of 5 μm96. Thus, we use the size representing speechgenerated droplets considering asymptomatic transmission of SARS-CoV-297, which is at its most contagious before symptom onset98. Figure 22 shows the estimated terminal settling velocities of an evaporated aerosol as well as its associated Reynolds number and Knudsen number at that particular size and velocity in ambient air. The density of the aerosol is set to unity in this chart for illustration purposes; estimated sodium chloride solution density accounting for evaporation is used in producing all fitting and prediction results. Because the Stokes-Cunningham equation is only applicable to Re < 1 and particle size > 100 nm, the estimated terminal velocity is accurate up to approximately ~10 µm and down to 0.1 µm in terms of aerosol size. Thus, the size spectrum is broad enough to encompass the entire range of sizes produced by equilibrating speech-generated aerosols with ambient moisture (vide infra). For the range of sizes shown in Figure 22b, Kn << 10. Thus, the Epstein or Millikan equations73, 99 are not applicable in regard to the range covered (Fig. 2b). The decreasing 51 trend of Kn as droplet size increases also confirms the importance of surface slippage at small droplet sizes. Droplets of an initial size of 6 µm equilibrate with atmospheric moisture and evaporate into smaller aerosols or condense into larger droplets as shown in Figure 23 a-b. Figure 23a shows the temperature effect on the size of aerosol after evaporation or condensation, which is negligible within the temperature range seen in the counties investigated. Assuming that a few seconds are needed for droplets to evaporate to an equilibrium size,88 we further assume instantaneous kinetics, thus the temperature effects demonstrated in this work are expected to be smaller than in reality. Figure 23b shows the relative humidity effect on the size of aerosol after evaporation (below 90% relative humidity) or condensation (at 100% relative humidity). A higher relative humidity corresponds to a larger equilibrium size of droplet or aerosol as expected. An initial size of 6 µm yields a droplet of size 1 to 10 µm in equilibrium with moisture, and this final droplet size is used to calculate its settling time from the height of 1.5 m shown in Figure 23c. As expected from Figure 23a, the temperature effects on the settling velocity are minimal. The relative humidity effect on settling time is significant, yielding as short as 1 min at 100% relative humidity and >20 min at <10% relative humidity. The evaporation and settling calculations agree with the classic Wells model.89, 100 Similarly, the SARS-CoV-2 virus half-life is plotted as a function of ambient temperature and relative humidity in Figure 23d. Lower temperatures and humidities yield longer viral half-lives. However, the relationship is highly nonlinear. The non-linearity poses a challenge to previous models69, 101-102 using meteorological data directly as input variables. Current transmission models incorporating weather data as input variables have varying goodness of fits and correlation significances that may be due to how the meteorological variables were used85. For example, humidity has been factored into models as relative humidity,103 absolute humidity,104 or dew point.105 The correlation between humidity and transmission may be related to the hydrophilic interactions between water and the proteins on the outer surface of SARS-CoV-2 virus via hydrogen bonding.106 The range of virus half-lives varies from several minutes to over an 52 hour with typical ranges of temperature and humidity in April. These results underscore the potential effect of weather on airborne virus transmission. Results show that the weather affects the fate and transport of speech-generated, virus-laden droplets by changing the settling times and viral half-lives, and thus these intertwined effects may not be captured by a simple linear model. To establish an effective weather-based model for COVID-19 epidemic prediction, regression analyses (LR and VAR) and machine learning models (RNN and LSTM) were compared for 5 U.S. counties. Figure 24 shows the timeseries of daily case percentage increase in the different US counties. The model fittings follow the major trends of the actual data and capture most of their peaks and troughs; the actual data of the last 4 days also fall inside the one-sigma prediction intervals despite simplicity of the models used. The goodness of fit and the prediction accuracy generally rank as follows: LSTM > simple RNN > LR > VAR (see r2 for fitting and residual sum of squares (RSS) for prediction in Table 3a-b). Considering a key difference between VAR and the rest of the models are the use of autoregressively predicted settling times and viral viability data versus actual data starting from the second day of prediction, the lowest fitting and prediction accuracy of VAR suggests inaccurate aerosol settling times and viral viability predictions from past data as expected. It is clear that accurate weather-originated data input is required to predict transmission rates accurately. VAR also includes past transmission rate data as an input, which is not included in the other models explored. This suggests that past transmission is not a significant input variable for predicting future transmission compared to the two weather-originated variables as normalized into a percentage increase. Improved fitting for LSTM over simple RNN suggests that weather beyond 5 days prior affects current transmission. Better fitting and prediction performance of neural network models compared to LR suggests nonlinearity in the correlation between settling time, viral viability, and transmission rate, even though reasonable linear correlations are observed. For example, the r2 values for the counties considered vary from 0.36 to 0.80 with an average of 0.59, achieved using input variables capturing two types of weather influences on transmission. Variability in goodness of fit 53 among the counties may be explained by local residents, who have delay in time from the onset of symptom to getting a COVID test. To better understand how weather-originated aerosol settling times and viral viability affect transmission, the contours of model predictions are shown in Figure 25. The ranges of settling times and virus half-lives are determined in part by the local temperature and RH range during April, for each county of study. Note that the data points used to generate the contour plots are not uniformly distributed inside the contours, and the data to be predicted may not lie within the range of training data (see Figure 28). Although UV intensity is not a direct input variable in this model, it positively correlates with temperature107 and is, therefore, indirectly taken into account in this model. Counties have faster transmission at longer aerosol settling times or longer virus half-lives. These results indicate that active-virus-laden aerosols are a major pathway for COVID transmission. The only exception to this claim was seen for Santa Clara County for which there appeared to be faster transmission at low viral viability and settling times leading to a less accurate prediction compared to the other counties that were analysed in Figure 25. Harris, King and Maricopa counties show faster transmission with a longer virus half-life, while LA County had increased transmission rates at longer settling times. The LR, VAR and simple RNN predictions show clear trends, while the trend of LSTM predictions indicates hotspots for easy transmission in the 2D space of viral viability and aerosol settling times. This may be indicative of the small training data set used, considering the high accuracy of fitting and predictions by LSTM in Figure 24. The different trends between LA County vs. Harris, King and Maricopa counties may be a result of their different policies and human behaviours not captured by the input variables in this work. Future work in the training of an LSTM model with sufficient data over a wide range of weather conditions from all seasons may reveal a clear trend of correlation similar to the LR, VAR, and simple RNN models in this work. The performance of transmission rate prediction based on aerosol settling times and viral viability was also studied with an extended dataset of Maricopa County from May to August 54 2020, as shown in Figure 26. The r values are 0.172, 0.579, and 0.999956 for linear 2 regression, simple RNN, and LSTM, respectively. Like the April data, LSTM has the closest fitting, followed by the simple RNN, and a linear regression. All three models have similar prediction accuracies, with RSS values of 0.0110, 0.0156, 0.0160 for linear regression, simple RNN, and LSTM, respectively. The matching performance of these 3 models are also observed in April Maricopa County data. The observed increase in new cases line falls within the one-sigma prediction interval for the last 21 days of available data. The prediction from weather-driven settling times and viral viability to transmission rate in this work corroborates with previous findings that transmission is faster at low temperatures and humidites for COVID in major global cities from Nov 2019 – Feb 2020,108 in the US using state-level data over Jan – Apr 2020,109 and for Singapore using data from Jan – May 202069. Respiratory droplets travel can travel three times farther at lower temperatures and higher humidity compared to typical dry and hot environments.110 It should be noted that not all published work supports a link between weather and transmission. Linear machine learning models failed to establish the correlation between state-level (Italy and US) or country-level (rest of the world) transmission and meteorological data.103 This is most likely due to the non-linearity in linking temperature and humidity data to other variables that are important factors in transmission. For similar reasons, a recent multilinear regression model found no significant correlation between temperature, humidity and the basic reproductive number R0 of transmission.111 However, the lack of correlation between meteorological data and COVID transmission in China during early 2020 may be a result of strong policy changes overshadowing any weather effects.112 Other works have analysed the link between virus-laden aerosol settling and SARS-CoV-2 transmission from different perspectives.61, 65, 113 Smith et al. provided a useful model that assesses aerosol transmission of SARS-CoV-2 through respiratory droplet physics.113 TheirThe study calculated the number of virus particles inhaled via indirect airborne transmission by calculating the persistence (settling time) of cough-generated aerosols, and concluded that aerosol transmission is a possible but not efficient route of transmission of 55 113 SARS-CoV-2. 61 This conclusion as well as evidence suggested by Stadnytskyi et al. and Anfinrud et al.65 agree with the conclusion of the present work to the extent that indirect airborne transmission is a possible route of transmission of SARS-CoV-2. The WHO, in the most recent update (Apr 30, 2021), has also acknowledged aerosol transmission as one of the major routes of transmission for SARS-CoV-2.114 Homogeneity of the aerosols in the space studied is often assumed in these approaches to translate aerosol persistence to aerosol inhaled, which can be far from reality91. One advantage of this work is that by predicting transmission from aerosol persistence (and virus viability) via data analysis tools, homogeneity is not assumed. Because the infection risk assessment is embedded in the data analysis step connecting aerosol persistent and transmission, mathematical infection risk assessment models such as Wells-Riley and dose-response are also not required in this work. This approach reduces uncertainties introduced into the model as the infection threshold of SARS-CoV-2 is still unclear115. A key assumption in the models presented is that the timeframes of virus transmission, disease progression, test-to-results, and hospitalizations are consistent across a studied population, their location, and time span. However, timeframes could actually be fluctuating and thus undermine the accuracy of our model predictions. For example, since COVID case data that is reported may have inherent time delays due, for example, to the shortage of test kits. Delays are an important parameter in this study, and thus model fitting residuals associated with this input variable cannot be eliminated. Another underlying assumption of this study is that the fraction of asymptomatic infections of total infections is constant. However, this is still unknown to the best of our knowledge. Our models also have simplifications that may be additional sources of error. These simplifications include that a sodium chloride solution, which is used as a surrogate model of physiological fluids, is a good proxy for virus-laden aerosols and that the surface tension of an aerosol droplet is only a function of its temperature and solute concentration. The neural network models use a random set of parameters initially for each neuron, and the optimized result can be dependent on this initial set of parameters, if they are actually too different from the optimal set. 56 Although the models in this work use the outdoor weather input variables and transmission can occur indoors, the outdoor temperature correlates positively with the indoor environment116-117. The correlation coefficient (slope of linear regression), however, depends on the season and location. For example, Massachusetts has Toutdoor ~ 0.04Tindoor at T < ~10℃, and Toutdoor ~ 0.41Tindoor at T > ~10℃.118 South Korea has Toutdoor ~ 0.13Tindoor at T < ~15℃, and Toutdoor ~ 0.47Tindoor at T > ~15℃.119 The indoor absolute humidity also tracks the outdoor humidity across seasons and diverse locations.117, 120-121 As a result, the outdoor transmission risk predicted in this work tracks with, and can be used as a surrogate for the indoor transmission risk. Control measures such as mandatory mask-wearing and lockdowns are not accounted for by two input variables in this work. We limit our scope to April in Figure 24 when nationwide lockdown was still in effect to minimize this variable in terms of its influence on transmission. The extended-time analysis on Maricopa County for May-August in Figure 26 has lower fitting and prediction accuracy compared to the April results as shown in Table 3c. The lower accuracy for longer time periods of analysis may be the result of encompassing more non-weather-related events, such as a significant increase in mask-wearing and the mass public protests of 2020. Although it is possible that the models presented in this work are not capable of handling data over longer times, the RNN models typically benefit from additional training data to improve prediction accuracy. They are expected to have improved prediction performance for longer study times, if non-weather-related events would be represented in the model. 2.5 Conclusions Seasonality of airborne COVID transmission may be explained in part by weather-induced changes in the aerosol settling times and virus viability. We use Stokes’ sedimentation model with a Cunningham correction factor for surface slippage in order to estimate the settling times of speech-produced aerosols after evaporation for Re < 1 and Kn << 1. SARS-CoV-2 viral viability is estimated using an empirical relationship from local historical weather data. Linear regression, vector autoregression, and recurrent neural network models using the 57 settling time, viral viability and past transmission rate successfully predict future transmission rates within one-sigma prediction interval. Airborne speech-generated aerosol transmission is a significant transmission route of SARS-CoV-2. Including aerosol settling time and viral viability from historical weather data as input variables can improve the accuracy of transmission rate prediction. Corroborating with publications and public actions over the past year, the findings of this study support implementation of control measures including social distancing, enforcing mask wearing, and systematic preventive measures such as improved ventilation in both community and healthcare settings. Overall, the evidence on weather influence of transmission has been contradictory and inconclusive. We note that the present work does not aim to prove that aerosol settling time and virus viability are exclusively important on predicting transmission rate. The fitting and prediction performance of the models presented suggests that weather plays a considerable role in transmission. Thus, the incorporation of weather-derived, transmission-mechanismsbased input variables, including aerosol settling times and virus viability, into epidemiological prediction model may worth further investigation. Future work in model development should also include additional variables that play a role in airborne or surfacebased transmission such as wind speeds, turbulence (especially those created by speech which can lengthen the suspension time by 30-150 times122), and UVB intensity. Datasets should include more locations outside of the US where the weather system may be different. Furthermore, the study periods can be extended to allow for better machine learning algorithm training. 2.6 Supplementary Note Code for this work is available at: https://github.com/zhuyanzhe98/evaptransmission 58 59 2.7 Figures Figure 21. (a) illustration showing the model data flow in this work (b) typical COVID-19 progression around the date of positive test result. The three periods are: the presymptomatic contagious period, the wait period to obtain the test result after taking the test, and the recovery period at the end of which the patient is modelled as either recovered and no longer contagious, or entering the intensive care unit (ICU) and isolated from the public. We assume that the patient takes the test at the onset of symptoms. Under this assumption, 60 a positively tested patient is considered contagious in our model from 4 days before until 12 days after the positive test result. 61 Figure 22. (a) Calculated settling velocities of aerosols of varying sizes using StokesCunningham Law. (b) The Reynolds number (Re) and Knudsen number (Kn) of droplets of varying sizes. At Kn < 10, the Stokes-Cunningham Law is the most applicable firstprinciple relationship to calculate settling velocity. 62 Figure 23. Evaporated aerosol sizes derived from the Köhler equation based on an initial size at different ambient (a) temperatures at 50% relative humidity and( b) relative humidity at 25 ℃. (c) Calculated settling times obtained from the empirical model using 6 µm as the initial droplet size and (d) viral viability at different ambient temperatures and relative humidity. 63 64 Figure 24. Timeseries of daily case percentage increase in decimal format for April 2020 in counties studied. The predicted daily case increase of the last 4 days are shown as triangles with their associated one-sigma prediction intervals. Dashed lines show the model fitting from the 6th day to the 25th day of April. No fitting data obtained from the model for the first 5 days because they would require weather data from March (up to 5 days prior). LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long-/short-term memory recurrent neural network. The green filled areas represent 95% confidence intervals for LR predictions. The blue patterned areas represent 95% confidence intervals for VAR predictions. 65 66 Figure 25. Contour plots of the daily case percentage increase as a function of settling time and viral viability (represented by half-life) for different counties. Colour shows the daily case percentage increase in decimal. LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long short term memory recurrent neural network. 67 Figure 26. (a) Fitting and (b) predicted daily new case percentage increase for Maricopa County from May to August 2020. Interrupted data in (a) is due to interrupted weather history data from NOAA. Error bars show one-sigma prediction intervals. The training data in (a) are 75 days long and the testing data in (b) are 21 days long. LR: linear regression; Simple RNN: simple recurrent neural network; LSTM: long short term memory recurrent neural network. The green filled areas represent 95% confidence intervals for LR predictions. 68 Figure 27. Size distribution of speech-generated droplets before evaporation measured by Morawska et al.91 The peak number concentration is at 6 µm and is used as the input to the Köhler equation in this work. 69 70 Figure 28. Data points available to generate the contour plot in Fig. 5 for each county, with the daily case percentage increase as a function of settling time and viral viability (represented by half-life). Colour shows the daily case percentage increase in decimal. The daily cases percentage increase is shown as the colour of each data point. The dots represent data points for training, and the triangles represent data for prediction. LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long short-term memory recurrent neural network. 71 2.8 Tables Table 3. (a) R-squared of model fitting and (b) sum of squares of residuals of model prediction for each county and model during April 2020. (c) R-squared of model fitting and sum of squares of residuals of model prediction for Maricopa County during extended time frame from May to August 2020. LR: linear regression; VAR: vector autoregression; Simple RNN: simple recurrent neural network; LSTM: long short term memory recurrent neural network. 72 2.9 Acknowledgements This work was supported by the Bill & Melinda Gates Foundation Investment Grant INV018569. The authors thank Paul Dabisch at National Biodefense Analysis and Countermeasures Center for sharing his insights on SARS-CoV-2 viability. The authors would like to acknowledge Richard Flagan for helpful discussions. 73 Chapter 4 IMAGE RECOGNITION FOR A SARS-COV-2 TEST KIT Chapter IV: Image Recognition for a SARS-CoV-2 Test Kit Reproduced from the article below with permission from the American Chemical Society. Further permission related to this material should be directed to the American Chemical Society: https://pubs.acs.org/doi/10.1021/acs.est.1c04623. Zhu, Y., Wu, X., Gu, A., Dobelle, L., Cid, C. A., Li, J., & Hoffmann, M. R. (2021). Membrane-Based In-Gel Loop-Mediated Isothermal Amplification (mgLAMP) System for SARS-CoV-2 Quantification in Environmental Waters. Environmental Science & Technology. doi:10.1021/acs.est.1c04623. 3.1 Abstract Since the COVID-19 pandemic is expected to become endemic, quantification of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in environmental waters is critical for rapid risk assessment and for early detection of outbreaks. Herein, we report on the development of a membrane-based in-gel loop-mediated isothermal amplification (mgLAMP) system that is designed for the rapid point-of-use quantification of SARS-CoV2 particles in environmental waters. The mgLAMP system integrates viral concentration, inassay viral lysis, and an on-membrane hydrogel-based RT-LAMP quantification using an enhanced fluorescence detection with a target-specific probe. With a sample-to-result time of less than 1 hour, mgLAMP has a limit of detection (LOD) of 0.96 copies/mL in Milli-Q water and 93 copies/mL in surface water. The LOD of mgLAMP for SARS-CoV-2 detection in surface water was at least 10-fold lower than that of the reverse transcription quantitative polymerase chain reaction (RT-qPCR). A 3D-printed portable device is designed to integrate heated incubation and fluorescence illumination for the simultaneous analysis of 9 mgLAMP 74 assays. Smartphone-based imaging and machine learning-based image processing are used for interpretation of results. In this report, we demonstrate that mgLAMP is a promising method for large-scale environmental surveillance of SARS-CoV-2 without the need for specialized equipment, highly-trained personnel, and labor-intensive procedures. 3.2 Introduction The pandemic of coronavirus disease 2019 (COVID-19) is predicted to become endemic with seasonal fluctuations.123-124 The SARS-CoV-2 virions and their RNA are known be shed in feces of infected individuals at a concentration of 102 to 108 copies/g.125-127 Depending on sanitary practices the viral RNA is discharged in wastewater and surface waters.128 Detection of SARS-CoV-2 RNA in river water and wastewater has been widely reported globally such as in Italy,129 Japan,130 and Brazil131. In Spain, SARS-CoV-2 RNA was detected in wastewater before the first local case was confirmed.132 Virions may persist in environmental waters given that the time required for 99% viable virus load reduction (T99) of 19.5 days for wastewater and T99 at 18.7 days for river water at 4 °C.133 Thus large-scale environmental surveillance is necessary for quantitative risk assessment, for notice of potential outbreaks, population-wide infection prevalence monitoring, and for a better understanding of SARSCoV-2 contamination and potential spread via environmental waters.129, 134-135 Therefore, development of point-of-sampling testing technologies for large-scale environmental surveillance is needed to avoid the use of specialized equipment, highly-trained personnel, and labor-intensive laboratory procedures.136 Detection of SARS-CoV-2 is based on nucleic acid analysis (NAA) using primarily reverse transcription quantitative polymerase chain reaction (RT-qPCR).130, 132, 137-140 RT-qPCR is not readily adaptable for large-scale environmental surveillance especially in resourcelimited settings due to its requirement for a specialized thermocycling instrument and the need for highly-trained personnel. Furthermore, RT-qPCR does not produce absolute quantification. In addition, RT-qPCR targeting SARS-CoV-2 is sensitive to inhibitors that are present in wastewater leading to false negative results.141-142 To address these challenges, isothermal NAA methods, such as loop-mediated isothermal amplification (LAMP), are used for environmental quantification of other microbial pathogens including Zika virus,143 144 astrovirus, 145 MS2, 75 . RT-LAMP quantification has 146-147 E. coli, and Enterococcus spp, higher tolerances to inhibitors and shorter the amplification times (e.g., 30 min) compared to RT-qPCR. Huang et al. used a portable in-gel LAMP platform for the sensitive detection of MS2 coliphage in wastewater while RT-qPCR failed to produce a positive result.145 RTLAMP has been used in portable SARS-CoV-2 detection platforms for use on clinical samples.148-149 However, isothermal methods have yet to be developed for detecting SARSCoV-2 in environmental water samples. Point-of-use (POU) NAA for environmental surveillance normally requires the implementation of complex procedures required for sample preparation, including viral particle concentration, RNA extraction, and subsequent purification.140 SARS-CoV-2 loads observed in surface water and wastewater are typically lower than the detection limit of RTqPCR.128, 130, 132 Therefore, SARS-CoV-2 viral particles are commonly concentrated from 50-500 mL water sample for example by ultracentrifugation,150 ultrafiltration,137 polyethylene glycol (PEG) precipitation,151 or electronegative membrane filtration130, 137. The recovery rates using these methods for SARS-CoV-2 concentration of samples taken from wastewater and surface waters are poorly understood, while the recovery rates for similar-structured surrogates are generally in the range of 26.7-65.7%.137 Other existing infield virus concentration technologies, such as nanofiltration via superabsorbent polymer microspheres152-153 and bag-mediated filtration154, may be readily adaptable for SARS-CoV2. For RNA concentration and purification, specialized commercial kits for environmental samples are commonly employed but they involve a series of manual operations.130, 137, 140, 150-151 These sample preparation steps take at least 2 hours and require specialized instruments. Integration of sample preparation is needed for the development of rapid, sensitive, and POU quantification platforms targeting SARS-CoV-2 in environmental waters. Herein, we report on a membrane-based in-gel loop-mediated isothermal amplification (mgLAMP) system for absolute quantification of SARS-CoV-2 in environmental water samples within a 1-hr timeframe. The limit of detection (LOD) was 0.96 copy/mL in Milli- 76 Q water and 93 copies/mL in surface water. The mgLAMP assay and setup development is presented, and its performance was characterized in SARS-CoV-2 spiked Milli-Q water, river water samples, and wastewater samples. We further demonstrated the feasibility of POU applications by developing a portable device that integrates heat incubation, fluorescence illumination, and a cloud-based smartphone image analysis algorithm for quantitative results interpretation. We demonstrate that our integrated portable platform can be reliably used for the POU absolute quantification of SARS-CoV-2 in environmental water samples. Our mgLAMP, as an NAA-based platform, has been adapted for the detection of other microbial pathogens and can be modified easily for use in other sample matrices (e.g., clinical samples). 3.3 Image Analysis Methods For automated signal counting, a machine-learning model was developed using AutoML Vision module of the Google Cloud Platform (Google, Mountain View, CA). To establish the model, 10 fluorescence microscope images were uploaded to Google Cloud Storage and all signals in those images were labelled manually. Each image with more than 50 signals was split into a 3×3 grid (total of 9 smaller images), and each of the smaller images was uploaded and labelled separately to meet the Google Cloud Vision’s limit of 50 signals per image. All images uploaded were categorized randomly as training, testing and validation images with an 8:1:1 ratio. Training of the machine learning model used the high accuracy goal setting and 27 node hours. The trained model was deployed with 1 node for online access. To assess the accuracy of our trained model, 16 new smartphone images were analyzed and compared with manual counts. A portable device was designed to integrate heating, fluorescence illumination, and fluorescence emission filtering. The prototype was assembled with electrical components purchased from Digikey (Thief River Falls, MN) and Programmed Scientific Instruments (Arcadia, California) and plastic structures fabricated by 3D printing (Makerbot METHOD X, Makerbot, Brooklyn, NY) using PLA and ABS filaments (Makerbot, Brooklyn, NY). 77 3.4 Image Analysis Results: Proof of Portability To demonstrate the POU applicability of mgLAMP, we developed an image processing algorithm for mgLAMP pictures taken by a smartphone on the portable mgLAMP prototype. Quantitative mgLAMP results were generated by counting amplification dots. As an alternative to fluorescence microscopy, smartphone imaging of the fluorescence amplification dots was shown to be robust enough to deliver comparable image quality for dot counting.145 Paired with smartphone imaging, our image processing algorithm can be used for point-of-sample-collection quantitative result determination. A machine learning model was trained for 27 node hours using AutoML Vision module of the Google Cloud Platform by feeding 10 fluorescence microscope mgLAMP images with labelled amplicon dots. Models trained with up to 96 mgLAMP images (94 fluorescence microscope images and 2 smartphone images) and 20-54 node hours training time yielded lower recognition accuracy, likely due to overtraining.155-156 We then tested the trained model with 16 new mgLAMP images and compared the model-recognized signals with visual assessment. An example of this approach is shown in Figure 29a-b. In this image, 20 out of 24 signals were correctly recognized by AutoML vision with no false positive and 4 false negatives. For the 16 tested images, the results from the machine learning algorithm were close to those obtained by manual counting (y = 0.9926x, R2 = 0.992) (Figure 29c). The deviations from manual counting were mostly false negatives due to the low fluorescence intensity of the dots or vague boundaries between the adjacent dots with occasional false positives. Furthermore, our portable prototype (Figure 29d) was assembled with 3D-printed parts and commercially-available electrical components. The prototype is able to run 9 mgLAMP reactions simultaneously through 30 min of RT-LAMP heating at 65 °C, fluorescence excitation with high-powered LEDs, and fluorescence emission filtering (Figure 29e). The estimated cost of the prototype is approximately $500. This compact, low cost mgLAMP system can run assays without access to biological laboratories or the need for bulky labbased equipment. 78 3.5 Figures Figure 29. Cloud-based image processing and portable device prototyping. (a) An example mgLAMP image taken by a smartphone. (b) In the same image, the squares highlight the recognized amplicon dots by our trained Google AutoML Vision model. (c) The accuracy of the trained ML model in amplicon dot counting. Sixteen smartphone images were tested, and the results were compared to manual counting of the same images. The dotted line represents the linear fitting of the observed ML counts versus manual counts, with the equation and R-square displayed. (d) Picture of our portable mgLAMP device prototype. The prototype holds 9 mgLAMP reactions, provides 65 °C heating, laser excitation, and fluorescence emission filter. (e) Examples picture of the output from the prototype. 79 3.6 Acknowledgements General: The authors thank Caltech Biological Imaging Facility for use of the confocal microscope and Dr. Yuxing Yao for assistance with DLS measurements. Funding: The authors acknowledge the financial support provided by the Bill and Melinda Gates Foundation (grant nos. OPP1192379, INV-018569, and INV-030223). 80 Chapter 5 CO2-ACTIVATED H2O2 OXIDATION Chapter V: CO2-Acivated H2O2 Oxidation 4.1 Abstract Peroxymonocarbonate (HCO4-) is a reactive oxygen species previously used in chemical manufacturing and water detoxification. However, its role in atmospheric chemistry remains largely unexplored despite its high reduction potential for oxygen-atom transfer reactions (E(HCO4-/HCO3-) = 1.8V vs SHE) as well as the abundance of CO2 and H2O2 in the atmosphere. This work explores the formation and role of HCO4- at the interface of an aqueous H2O2 jet and an intersection stream of CO2 gas in an online electrospray ionization mass spectrometry (o-ESI-MS) reacting for ~50 µs. The interfacial kinetics of HCO4formation is measured to be >104 times faster than the bulk phase kinetics, possibly due to the incomplete hydrogen bonding network at the gas-liquid interface. The oxidation of methanesulfinate by H2O2 is 200 times faster in the presence of CO2 than in its absence. Peroxymonocarbonate may play an important yet previously unexplored role in the oxidation of atmospheric sulfides, (bi)sulfites and alkenes. 4.2 Introduction Peroxymonocarbonate (HCO4-) is an oxidant formed in the presence of H2O2 and CO2 via reaction R6157-158 H2O2 + CO2 ⟶ H+ + HCO4- [R6] Given its high estimated standard reduction potential (E(HCO4-/HCO3-) = 1.8V vs SHE)159 for oxygen atom transfer reaction, CO2 and HCO3- has been used as a green catalyst to activate H2O2 in alkene epoxidation160-161 and sulfide oxidation159, 162. In those applications, a rate acceleration of 2 orders of magnitude is observed for most substrates upon introduction 81 of CO2 or HCO3 . The role of peroxymonocarbonate is recognized as a possible approach - for soil remediation in which aqueous H2O2 is injected in the sub-surface where it reacts with environmental HCO3- to form HCO4-. The formed HCO4- may have similar importance as hydroxyl radicals (.OH) in oxidizing environmental contaminants under certain geological conditions.163 In the atmosphere, both H2O2 and CO2 also exist: H2O2 concentrations are measured within the range of 0.1-1 ppbv in the gas phase, and 1-100 µM in aerosols;164 CO2 is measured to be 412 ppm in the gas phase,165 and calculated to be 14 µM in the aqueous phase using Henry’s Law.166 Hence, peroxymonocarbonate is expected in the atmosphere, though its importance remains unexplored. Considering the abundance of H2O2 in the liquid aerosol phase and CO2 in the gas phase, gas-liquid interfacial chemistry may be significant in relation to the bulk aqueous or gas phase reactions. Previous studies (including the work presented in Chapter 2) found 3-4 orders of magnitude rate acceleration for interfacial reactions compared to the same reactions in the bulk liquid phase.35, 167-168 This work investigates the importance of peroxymonocarbonate in terms of its formation kinetics and oxidation of atmospherically relevant substrates, both at the air-water interface and in the bulk aqueous phase. The model substrate is chosen to be methanesulfinate (MeSO2-), which can accept an oxygen atom from either H2O2 or HCO4- to form methanesulfonate (MeSO3-) as shown in Figure 30. Methanesulfinate is an oxidation intermediate from biogenic dimethyl sulfide (Me2S) emissions.32 These emissions account for ∼15% of the global atmospheric sulfur budget.169 An accelerated oxidation kinetics by HCO4- compared to H2O2 may help explain higher than predicted oxidation products observed in the atmosphere. 4.3 Methods The experimental apparatus is largely the same as described in Chapter II of the thesis, with different reagents and mass spectrometer voltage parameters. Positively and negatively charged reaction products were analyzed by electrospray ionization mass spectrometry (ESI- 82 -1 MS). In the gas-liquid experiments, aqueous solutions of H2O2 are injected (30 µL min ) through a stainless-steel Agilent G1946-68703 nozzle) into the spray chamber of the ESI mass spectrometer (Agilent G2445A Ion Trap MSD) (maintained at 1 atm with ultrapure N2(g), 293 ± 2 K) using a syringe pump (Cole-Parmer 74905-02 and 74905-50). A coaxial, codirectional, fast (∼340m s-1) stream of N2(g) pneumatically nebulizes the outermost layers of the aqueous microjets into sub-micrometer-sized droplets (microdroplets) within 50 µs.36 Microjets and microdroplets were made to react with an orthogonal stream of O3(g) directed to the tip of the nozzle for ∼ 1 ms, which is the estimated residence time of the microdroplets in the spray chamber, before they are pumped into the low pressure section of the mass spectrometer.37-38 Microdroplets are naturally charged during nebulization by the segregation of anions from cations and deflected by an applied electric field towards the inlet to the lowpressure section of the mass spectrometer. Microdroplets evaporate within a heated capillary tubing prior to undergoing Coulombic explosions that release the ions contained in the microdroplets to the gas-phase for detection. A schematic of this experimental setup is shown in Figure 31. Experiments carried out in bulk-phase water were performed by injecting H2O2 and NaHCO3 aqueous solutions through separate ports into a manifold mixer (Cole-Parmer Manifold Mixing Valve EW-01356-21, Figure 32) modelled as a tank reactor. The reactions were then allowed to take place for 12 minutes before injection into the ESI mass spectrometer for analysis. In both interfacial and bulk-phase experimental setups, instrumental drift was eliminated by adding non-complexing Cs+ and ClO4- (50µM CsClO4) ions as internal standards. Positive and negative ions were quantified in the same experiments by alternating positive/negative either in mass scanning or SIM (selected ion monitoring) detecting modes. The ESI mass spectrometer was operated as follows. Positive ions were detected by polarizing the capillary at -3.434 kV (voltages relative to ground), the end plate offset and the capillary exit at 500 V and 60.7 V, respectively. In the negative mode, the capillary voltage was at 3.434 kV, and the end plate offset and the capillary exit at 500 and -60.7 V, 83 respectively. The nebulizer N2 gas source was maintained at 40 psi, and the drying gas (at 325 °C) flow rate was 10 L min-1. 99.99 % CO2(g) is supplied by Airgas and is introduced into the spray chamber of the ESIMS at 50 cm3 min-1 via a calibrated rotameter. The pH of aqueous solutions was adjusted before injection by adding HClO4/NaOH solutions and measured (within ± 0.5 pH units) with pH paper (Milipore MCholorpHast). Sodium methanesulfinate (MeSO2Na), sodium methanesulfonate (MeSO3Na), sodium bicarbonate (NaHCO3), cesium perchlorate (CsClO4), perchloric acid (HClO4) and sodium hydroxide (NaOH) of purity 95% or higher were obtained from Sigma Aldrich and used as received. 30% H2O2 is obtained from Sigma Aldrich and used as received. 4.4 Results and Discussion Figure 33 shows the formation kinetics of peroxymonocarbonate at moderately acidic and basic pH, in the bulk aqueous phase as well as at the air-water interface. In the bulk phase, a higher HCO4- product concentration is measured at pH = 8.5 compared to pH = 4.5, corroborating previous measurement of H2CO4 pKa ~ 4.159 The higher speciation fraction of HCO3- (pKa = 6.3 for H2O∙CO2/H2CO3 159) at pH = 8.5 may have also enhanced the formation of HCO4- from HCO3- and H2O2.158 At the air-water interface, however, both pH show similar HCO4- formation kinetics. This may be explained by previous observations of the superacidic weakly-hydrated hydronium ions at the air-water interface.167, 170 Despite having different bulk aqueous phase pH of 4.5 and 8.5, the superacidic hydronium ions at the interface protonates HCO3- and/or HCO4- to near completion in both scenarios, resulting in similar measured kinetics. In both scenarios, the interfacial reaction kinetics are measured to be 3-4 orders of magnitude faster compared the bulk liquid phase kinetics, similar to previous observations in other reaction systems.35, 167-168 The importance of peroxymonocarbonate in the atmosphere is showcased in Figure 34, where methanesulfinate is oxidized by H2O2 or peroxymonocarbonate. At pH = 4.5, no difference is measured in the oxidation rate of methanesulfinate upon introduction of gaseous 84 CO2 into the reaction system while aqueous H2O2 reacts with aqueous methane sulfinate. This choice of reactant phases simulates the atmospheric conditions. At pH = 8.5, the presence of aqueous HCO3- accelerates the H2O2 oxidation of methanesulfinate by 2 orders of magnitude in the aqueous phase (orange vs purple bar), which agrees with previous findings when oxidizing other sulfides with peroxymonocarbonate.159, 162 The presence of gaseous CO2 further accelerates the reaction by 4 orders of magnitude at the interface (purple vs green bar), agreeing with the interfacial rate acceleration observed for peroxymonocarbonate formation at pH = 8.5 in Figure 33. This indicate that peroxymonocarbonate is the reaction intermediate responsible for the observed rate acceleration. The interfacial rate acceleration may be attributed to the incomplete hydration sphere of methane sulfinate, where the unhydrated, exposed sulfur center may accept an oxygen atom transfer with less energy penalty than a hydrated sulfur center. 4.5 Conclusions Laboratory experiments indicate that gaseous and dissolved CO2 accelerate H2O2 oxidation of methanesulfinate, an atmospherically relevant substrate, by 2 orders of magnitude. The rate acceleration may be attributed to the formation of peroxymonocarbonate as the oxidation reaction intermediate. Formation of the peroxymonocarbonate intermediate and subsequent oxidation of the methanesulfinate substrate are accelerated by 3-4 orders of magnitude at the air-water interface compared to in the bulk aqueous phase. This chemistry may have been playing an important yet previously unrecognized role in atmospheric chemistry. The next step of this work may investigate other atmospherically relevant substrates that may accept an oxygen atom from peroxymonocarbonate such as dimethyl sulfoxide and dimethyl sulfide, as well as substrates whose oxidation may be initiated by accepting an oxygen atom, such as methyl vinyl ketone and methacrolein. 85 4.6 Figures Figure 30. Oxidation of methanesulfinate MeSO2- by H2O2 or HCO4- to methanesulfonate MeSO3-. The oxidation byproducts are HCO3- (from HCO4-) and H2O (from H2O2). 86 Figure 31. Experimental setup of the online ESI-MS. The aqueous reactants and gaseous reactants contact at the outlet tip of the microjet. 87 Figure 32. Experimental setup of the offline ESI-MS. The aqueous reactants contact each other in the manifold tank reactor before entering the microjet. 88 Figure 33. Percarbonate a.k.a. peroxymonocarbonate (HCO4-) formation kinetics (y-axis) measured at different pH, in the bulk aqueous phase and at the air-water interface. Rate constant is obtained assuming first order dependence on the concentration of H2O2, and CO2 or HCO3-. The ‘Bulk’ reaction forms peroxymonocarbonate from aqueous H2O2 and aqueous HCO3-; the ‘Interface’ reaction forms peroxymonocarbonate from aqueous H2O2 and gaseous CO2. 89 Figure 34. Kinetics of methanesulfinate (MeSO2-) oxidation by H2O2 (orange bars) or HCO4- (green and purple bars) at different pH, in the bulk aqueous phase (orange and purple bars) or air-water interface (green bars). The rate constant is plotted on the y-axis. The rate constants are obtained assuming first order dependence on MeSO2-, H2O2, and CO2 or HCO3-. The ‘Bulk H2O2’ reactions involve oxidizing aqueous MeSO2- with aqueous H2O2 in a nitrogen atmosphere; the ‘Interfacial HCO4-’ reaction involve oxidizing aqueous MeSO2- with aqueous H2O2 in the presence of gaseous CO2; the ‘Bulk HCO4-’ reaction involve oxidizing aqueous MeSO2- with aqueous H2O2 in the presence of aqueous HCO3-. 90 4.7 Acknowledgements This work is supported by the Resnick Sustainability Institute. Special thanks to Prof. Xuejing Yang for inspiration and helpful discussions. 91 BIBLIOGRAPHY Bibliography 1. Hoffmann, P.; Dedik, A. N.; Ensling, J.; Weinbruch, S.; Weber, S.; Sinner, T.; Gütlich, P.; Ortner, H. M., Speciation of iron in atmospheric aerosol samples. Journal of Aerosol Science 1996, 27 (2), 325-337. 2. Deguillaume, L.; Leriche, M.; Desboeufs, K.; Mailhot, G.; George, C.; Chaumerliac, N., Transition Metals in Atmospheric Liquid Phases: Sources, Reactivity, and Sensitive Parameters. Chemical Reviews 2005, 105 (9), 33883431. 3. Kuang, X. M.; Gonzalez, D. H.; Scott, J. A.; Vu, K.; Hasson, A.; Charbouillot, T.; Hawkins, L.; Paulson, S. E., Cloud Water Chemistry Associated with Urban Aerosols: Rapid Hydroxyl Radical Formation, Soluble Metals, Fe(II), Fe(III), and Quinones. ACS Earth Space Chem. 2019. 4. Paulson, S. E.; Gallimore, P. J.; Kuang, X. M.; Chen, J. R.; Kalberer, M.; Gonzalez, D. H., A light-driven burst of hydroxyl radicals dominates oxidation chemistry in newly activated cloud droplets. Science Advances 2019, 5 (5), eaav7689. 5. Ding, J.; Guan, Y.; Cong, Y.; Chen, L.; Li, Y.-F.; Zhang, L.; Zhang, L.; Wang, J.; Bai, R.; Zhao, Y.; Chen, C.; Wang, L., Single-Particle Analysis for Structure and Iron Chemistry of Atmospheric Particulate Matter. Analytical Chemistry 2020, 92 (1), 975-982. 6. Zhang, G.; Lin, Q.; Peng, L.; Yang, Y.; Jiang, F.; Liu, F.; Song, W.; Chen, D.; Cai, Z.; Bi, X.; Miller, M.; Tang, M.; Huang, W.; Wang, X.; Peng, P. a.; Sheng, G., Oxalate Formation Enhanced by Fe-Containing Particles and Environmental Implications. Environmental Science & Technology 2019, 53 (3), 1269-1277. 7. Chevallier, E.; Jolibois, R. D.; Meunier, N.; Carlier, P.; Monod, A., “Fentonlike” reactions of methylhydroperoxide and ethylhydroperoxide with Fe2+ in liquid aerosols under tropospheric conditions. Atmospheric Environment 2004, 38 (6), 921-933. 8. Chen, R.; Pignatello, J. J., Role of Quinone Intermediates as Electron Shuttles in Fenton and Photoassisted Fenton Oxidations of Aromatic Compounds. Environmental Science & Technology 1997, 31 (8), 2399-2406. 9. Gonzalez, D. H.; Cala, C. K.; Peng, Q.; Paulson, S. E., HULIS Enhancement of Hydroxyl Radical Formation from Fe(II): Kinetics of Fulvic Acid–Fe(II) Complexes in the Presence of Lung Antioxidants. Environmental Science & Technology 2017, 51 (13), 7676-7685. 10. Mao, J.; Fan, S.; Horowitz, L. W., Soluble Fe in Aerosols Sustained by Gaseous HO2 Uptake. Environmental Science & Technology Letters 2017, 4 (3), 98-104. 92 11. Daumit, K. E.; Carrasquillo, A. J.; Sugrue, R. A.; Kroll, J. H., Effects of Condensed-Phase Oxidants on Secondary Organic Aerosol Formation. The Journal of Physical Chemistry A 2016, 120 (9), 1386-1394. 12. Al-Abadleh, H. A., Review of the bulk and surface chemistry of iron in atmospherically relevant systems containing humic-like substances. RSC Advances 2015, 5 (57), 45785-45811. 13. Lim, Y. B.; Turpin, B. J., Laboratory evidence of organic peroxide and peroxyhemiacetal formation in the aqueous phase and implications for aqueous OH. Atmos. Chem. Phys. 2015, 15 (22), 12867-12877. 14. Pestovsky, O.; Bakac, A., Identification and Characterization of Aqueous Ferryl(IV) Ion. In Ferrates, American Chemical Society: 2008; Vol. 985, pp 167-176. 15. Bakac, A., Oxygen Activation with Transition-Metal Complexes in Aqueous Solution. Inorganic Chemistry 2010, 49 (8), 3584-3593. 16. Bataineh, H.; Pestovsky, O.; Bakac, A., pH-induced mechanistic changeover from hydroxyl radicals to iron(iv) in the Fenton reaction. Chemical Science 2012, 3 (5), 1594-1599. 17. Pestovsky, O.; Bakac, A., Aqueous Ferryl(IV) Ion: Kinetics of Oxygen Atom Transfer To Substrates and Oxo Exchange with Solvent Water. Inorganic Chemistry 2006, 45 (2), 814-820. 18. Barton, D. H. R.; Hu, B.; Taylor, D. K.; Rojas Wahl, R. U., The importance of carboxylate ligands in the differentiation of catalase reactivity from Gif ketonization systems. Tetrahedron Letters 1996, 37 (8), 1133-1136. 19. Barton, D. H. R.; Hu, B.; Taylor, D. K.; Wahl, R. U. R., The selective functionalization of saturated hydrocarbons. Part 32. Distinction between the Fe–Fe and Fe–Fe manifolds in Gif chemistry. The importance of carboxylic acids for alkane activation. Evidence for a dimeric iron species involved in Gif-type chemistry. Journal of the Chemical Society, Perkin Transactions 2 1996, (6), 1031-1041. 20. Stavropoulos, P.; Çelenligil-Çetin, R.; Tapper, A. E., The Gif Paradox. Accounts of Chemical Research 2001, 34 (9), 745-752. 21. Dorfman, L. M.; Adams, G. E., Reactivity of the hydroxyl radical in aqueous solutions. 1973. 22. Rush, J. D.; Maskos, Z.; Koppenol, W. H., [12] Distinction between hydroxyl radical and ferryl species. In Methods in Enzymology, Academic Press: 1990; Vol. 186, pp 148-156. 23. Deguillaume, L.; Leriche, M.; Chaumerliac, N., Impact of radical versus nonradical pathway in the Fenton chemistry on the iron redox cycle in clouds. Chemosphere 2005, 60 (5), 718-724. 24. Cantú Reinhard, F. G.; Faponle, A. S.; de Visser, S. P., Substrate Sulfoxidation by an Iron(IV)-Oxo Complex: Benchmarking Computationally Calculated 93 Barrier Heights to Experiment. The Journal of Physical Chemistry A 2016, 120 (49), 9805-9814. 25. Zhou, Z.; Dai, G.; Ru, S.; Yu, H.; Wei, Y., Highly selective and efficient olefin epoxidation with pure inorganic-ligand supported iron catalysts. Dalton Transactions 2019, 48 (37), 14201-14205. 26. Kameel, F. R.; Riboni, F.; Hoffmann, M. R.; Enami, S.; Colussi, A. J., Fenton Oxidation of Gaseous Isoprene on Aqueous Surfaces. Journal of Physical Chemistry C 2014, 118 (50), 29151-29158. 27. Huang, X. Y.; Groves, J. T., Beyond ferryl-mediated hydroxylation: 40 years of the rebound mechanism and C-H activation. Journal of Biological Inorganic Chemistry 2017, 22 (2-3), 185-207. 28. Keenan, C. R.; Sedlak, D. L., Factors Affecting the Yield of Oxidants from the Reaction of Nanoparticulate Zero-Valent Iron and Oxygen. Environmental Science & Technology 2008, 42 (4), 1262-1267. 29. Bataineh, H.; Pestovsky, O.; Bakac, A., Iron(II) Catalysis in Oxidation of Hydrocarbons with Ozone in Acetonitrile. ACS Catalysis 2015, 5 (3), 16291637. 30. Widger, L. R.; Davies, C. G.; Yang, T.; Siegler, M. A.; Troeppner, O.; Jameson, G. N.; Ivanovic-Burmazovic, I.; Goldberg, D. P., Dramatically accelerated selective oxygen-atom transfer by a nonheme iron(IV)-oxo complex: tuning of the first and second coordination spheres. J Am Chem Soc 2014, 136 (7), 2699702. 31. Baptista, L.; Clemente da Silva, E.; Arbilla, G., Oxidation mechanism of dimethyl sulfoxide (DMSO) by OH radical in liquid phase. Physical Chemistry Chemical Physics 2008, 10 (45), 6867-6879. 32. Watts, S. F.; Brimblecombe, P.; Watson, A. J., Methanesulphonic acid, dimethyl sulphoxide and dimethyl sulphone in aerosols. Atmospheric Environment. Part A. General Topics 1990, 24 (2), 353-359. 33. Walling, C., Fenton's reagent revisited. Accounts of Chemical Research 1975, 8 (4), 125-131. 34. Loegager, T.; Holcman, J.; Sehested, K.; Pedersen, T., Oxidation of ferrous ions by ozone in acidic solutions. Inorganic Chemistry 1992, 31 (17), 35233529. 35. Enami, S.; Sakamoto, Y.; Colussi, A. J., Fenton chemistry at aqueous interfaces. Proc Natl Acad Sci U S A 2014, 111 (2), 623-8. 36. Enami, S.; Colussi, A. J., Long-range specific ion-ion interactions in hydrogen-bonded liquid films. Journal of Chemical Physics 2013, 138 (18). 37. Colussi, A.; Enami, S., Detecting Intermediates and Products of Fast Heterogeneous Reactions on Liquid Surfaces via Online Mass Spectrometry. Atmosphere 2019, 10 (2). 94 38. Enami, S.; Mishra, H.; Hoffmann, M. R.; Colussi, A. J., Hofmeister effects in micromolar electrolyte solutions. Journal of Chemical Physics 2012, 136 (15). 39. Becke, A. D., Density‐functional thermochemistry. III. The role of exact exchange. The Journal of Chemical Physics 1993, 98 (7), 5648-5652. 40. Grimme, S.; Ehrlich, S.; Goerigk, L., Effect of the damping function in dispersion corrected density functional theory. Journal of Computational Chemistry 2011, 32 (7), 1456-1465. 41. Schäfer, A.; Horn, H.; Ahlrichs, R., Fully optimized contracted Gaussian basis sets for atoms Li to Kr. The Journal of Chemical Physics 1992, 97 (4), 25712577. 42. Schäfer, A.; Huber, C.; Ahlrichs, R., Fully optimized contracted Gaussian basis sets of triple zeta valence quality for atoms Li to Kr. The Journal of Chemical Physics 1994, 100 (8), 5829-5835. 43. Enami, S.; Sakamoto, Y.; Hara, K.; Osada, K.; Hoffmann, M. R.; Colussi, A. J., "Sizing" Heterogeneous Chemistry in the Conversion of Gaseous Dimethyl Sulfide to Atmospheric Particles. Environmental Science & Technology 2016, 50 (4), 1834-1843. 44. Chou, M.-S.; Chang, K.-L., Oxidation of Aqueous Dimethyl Sulfoxide (DMSO) Using UV, O3, and UV/O3. Ozone: Science & Engineering 2007, 29 (5), 391-397. 45. Cheng, J.; Vecitis, C. D.; Hoffmann, M. R.; Colussi, A. J., Experimental anion affinities for the air/water interface. J Phys Chem B 2006, 110 (51), 25598-602. 46. Bardouki, H.; da Rosa, M. B.; Mihalopoulos, N.; Palm, W. U.; Zetzsch, C., Kinetics and mechanism of the oxidation of dimethylsulfoxide (DMSO) and methanesulfinate (MSI−) by OH radicals in aqueous medium. Atmospheric Environment 2002, 36 (29), 4627-4634. 47. Jacobsen, F.; Holcman, J.; Sehested, K., Activation parameters of ferryl ion reactions in aqueous acid solutions. International Journal of Chemical Kinetics 1997, 29 (1), 17-24. 48. Mukherjee, G.; Lee, C. W. Z.; Nag, S. S.; Alili, A.; Cantu Reinhard, F. G.; Kumar, D.; Sastri, C. V.; de Visser, S. P., Dramatic rate-enhancement of oxygen atom transfer by an iron(iv)-oxo species by equatorial ligand field perturbations. Dalton Trans 2018, 47 (42), 14945-14957. 49. Pye, H. O. T.; Nenes, A.; Alexander, B.; Ault, A. P.; Barth, M. C.; Clegg, S. L.; Collett Jr, J. L.; Fahey, K. M.; Hennigan, C. J.; Herrmann, H.; Kanakidou, M.; Kelly, J. T.; Ku, I. T.; McNeill, V. F.; Riemer, N.; Schaefer, T.; Shi, G.; Tilgner, A.; Walker, J. T.; Wang, T.; Weber, R.; Xing, J.; Zaveri, R. A.; Zuend, A., The acidity of atmospheric particles and clouds. Atmos. Chem. Phys. 2020, 20 (8), 4809-4888. 50. Enami, S.; Ishizuka, S.; Colussi, A. J., Chemical signatures of surface microheterogeneity on liquid mixtures. The Journal of Chemical Physics 2019, 150 (2), 024702. 95 51. Enami, S.; Mishra, H.; Hoffmann, M. R.; Colussi, A. J., Hofmeister effects in micromolar electrolyte solutions. J Chem Phys 2012, 136 (15), 154707. 52. Enami, S.; Hoffmann, M. R.; Colussi, A. J., How phenol and alpha-tocopherol react with ambient ozone at gas/liquid interfaces. J Phys Chem A 2009, 113 (25), 7002-10. 53. Kebarle, P.; Verkerk, U. H., Electrospray: from ions in solution to ions in the gas phase, what we know now. Mass Spectrom Rev 2009, 28 (6), 898-917. 54. Xiong, H.; Lee, J. K.; Zare, R. N.; Min, W., Strong Electric Field Observed at the Interface of Aqueous Microdroplets. J Phys Chem Lett 2020, 11 (17), 74237428. 55. Enami, S.; Hoffmann, M. R.; Colussi, A. J., Proton Availability at the Air/Water Interface. The Journal of Physical Chemistry Letters 2010, 1 (10), 1599-1604. 56. Levanov, A. V.; Isaikina, O. Y.; Lunin, V. V., Thermodynamic and Kinetic Parameters of the Solubility of Ozone in Water. Russian Journal of Physical Chemistry A 2019, 93 (7), 1230-1234. 57. COVID-19 weekly epidemiological update - 19 Oct 2021; World Health Organization (WHO): 2021. 58. Qu, G.; Li, X.; Hu, L.; Jiang, G., An Imperative Need for Research on the Role of Environmental Factors in Transmission of Novel Coronavirus (COVID-19). Environ Sci Technol 2020, 54 (7), 3730-3732. 59. Ianevski, A.; Zusinaite, E.; Shtaida, N.; Kallio-Kokko, H.; Valkonen, M.; Kantele, A.; Telling, K.; Lutsar, I.; Letjuka, P.; Metelitsa, N.; Oksenych, V.; Dumpis, U.; Vitkauskiene, A.; Stasaitis, K.; Ohrmalm, C.; Bondeson, K.; Bergqvist, A.; Cox, R. J.; Tenson, T.; Merits, A.; Kainov, D. E., Low Temperature and Low UV Indexes Correlated with Peaks of Influenza Virus Activity in Northern Europe during 2010(-)2018. Viruses 2019, 11 (3). 60. Shaman, J.; Kohn, M., Absolute humidity modulates influenza survival, transmission, and seasonality. Proc Natl Acad Sci U S A 2009, 106 (9), 32438. 61. Stadnytskyi, V.; Bax, C. E.; Bax, A.; Anfinrud, P., The airborne lifetime of small speech droplets and their potential importance in SARS-CoV-2 transmission. Proc Natl Acad Sci U S A 2020, 117 (22), 11875-11877. 62. Dabisch, P.; Schuit, M.; Herzog, A.; Beck, K.; Wood, S.; Krause, M.; Miller, D.; Weaver, W.; Freeburger, D.; Hooper, I.; Green, B.; Williams, G.; Holland, B.; Bohannon, J.; Wahl, V.; Yolitz, J.; Hevey, M.; Ratnesar-Shumate, S., The Influence of Temperature, Humidity, and Simulated Sunlight on the Infectivity of SARS-CoV-2 in Aerosols. Aerosol Science and Technology 2020, 1-15. 63. Schuit, M.; Ratnesar-Shumate, S.; Yolitz, J.; Williams, G.; Weaver, W.; Green, B.; Miller, D.; Krause, M.; Beck, K.; Wood, S.; Holland, B.; Bohannon, J.; Freeburger, D.; Hooper, I.; Biryukov, J.; Altamura, L. A.; Wahl, V.; Hevey, 96 M.; Dabisch, P., Airborne SARS-CoV-2 Is Rapidly Inactivated by Simulated Sunlight. J Infect Dis 2020, 222 (4), 564-571. 64. Rodriguez-Palacios, A.; Cominelli, F.; Basson, A. R.; Pizarro, T. T.; Ilic, S., Textile Masks and Surface Covers-A Spray Simulation Method and a "Universal Droplet Reduction Model" Against Respiratory Pandemics. Front Med (Lausanne) 2020, 7, 260. 65. Anfinrud, P.; Stadnytskyi, V.; Bax, C. E.; Bax, A., Visualizing SpeechGenerated Oral Fluid Droplets with Laser Light Scattering. N Engl J Med 2020, 382 (21), 2061-2063. 66. Bazant, M. Z.; Bush, J. W. M., A guideline to limit indoor airborne transmission of COVID-19. Proc Natl Acad Sci U S A 2021, 118 (17). 67. Prather, K. A.; Marr, L. C.; Schooley, R. T.; McDiarmid, M. A.; Wilson, M. E.; Milton, D. K., Airborne transmission of SARS-CoV-2. Science 2020, 370 (6514), 303-304. 68. Kucharski, A. J.; Russell, T. W.; Diamond, C.; Liu, Y.; Edmunds, J.; Funk, S.; Eggo, R. M.; Sun, F.; Jit, M.; Munday, J. D.; Davies, N.; Gimma, A.; van Zandvoort, K.; Gibbs, H.; Hellewell, J.; Jarvis, C. I.; Clifford, S.; Quilty, B. J.; Bosse, N. I.; Abbott, S.; Klepac, P.; Flasche, S., Early dynamics of transmission and control of COVID-19: a mathematical modelling study. The Lancet Infectious Diseases 2020, 20 (5), 553-558. 69. Pani, S. K.; Lin, N. H.; RavindraBabu, S., Association of COVID-19 pandemic with meteorological parameters over Singapore. Sci Total Environ 2020, 740, 140112. 70. Lewis, E. R., An examination of Köhler theory resulting in an accurate expression for the equilibrium radius ratio of a hygroscopic aerosol particle valid up to and including relative humidity 100%. Journal of Geophysical Research 2008, 113 (D3). 71. Stokes, G. G., Mathematical and physical papers. Johnson reprint corp: 1850; Vol. 3. 72. On the velocity of steady fall of spherical particles through fluid medium. Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character 1997, 83 (563), 357-365. 73. Epstein, P. S., On the Resistance Experienced by Spheres in their Motion through Gases. Physical Review 1924, 23 (6), 710-733. 74. Millikan, R. A., The General Law of Fall of a Small Spherical Body through a Gas, and its Bearing upon the Nature of Molecular Reflection from Surfaces. Physical Review 1923, 22 (1), 1-23. 75. Jakobsen, A. B.; Merrison, J.; Iversen, J. J., Laboratory study of aerosol settling velocities using Laser Doppler velocimetry. Journal of Aerosol Science 2019, 135, 58-71. 97 76. Lin, K.; Marr, L. C., Humidity-Dependent Decay of Viruses, but Not Bacteria, in Aerosols and Droplets Follows Disinfection Kinetics. Environ Sci Technol 2020, 54 (2), 1024-1032. 77. Prussin, A. J., 2nd; Schwake, D. O.; Lin, K.; Gallagher, D. L.; Buttling, L.; Marr, L. C., Survival of the Enveloped Virus Phi6 in Droplets as a Function of Relative Humidity, Absolute Humidity, and Temperature. Appl Environ Microbiol 2018, 84 (12). 78. Chan, K. H.; Peiris, J. S.; Lam, S. Y.; Poon, L. L.; Yuen, K. Y.; Seto, W. H., The Effects of Temperature and Relative Humidity on the Viability of the SARS Coronavirus. Adv Virol 2011, 2011, 734690. 79. Shimasaki, N.; Morikawa, H., Prevention of COVID-19 Infection with Personal Protective Equipment. Journal of Disaster Research 2021, 16 (1), 6169. 80. Yang, W.; Elankumaran, S.; Marr, L. C., Relationship between humidity and influenza A viability in droplets and implications for influenza's seasonality. PLoS One 2012, 7 (10), e46789. 81. Dadar, M.; Fakhri, Y.; Bjorklund, G.; Shahali, Y., The association between the incidence of COVID-19 and the distance from the virus epicenter in Iran. Arch Virol 2020, 165 (11), 2555-2560. 82. Suhaimi, N. F.; Jalaludin, J.; Latif, M. T., Demystifying a Possible Relationship between COVID-19, Air Quality and Meteorological Factors: Evidence from Kuala Lumpur, Malaysia. Aerosol and Air Quality Research 2020, 20 (7), 1520-1529. 83. Venna, S. R.; Tavanaei, A.; Gottumukkala, R. N.; Raghavan, V. V.; Maida, A. S.; Nichols, S., A Novel Data-Driven Model for Real-Time Influenza Forecasting. IEEE Access 2019, 7, 7691-7701. 84. Tomar, A.; Gupta, N., Prediction for the spread of COVID-19 in India and effectiveness of preventive measures. Sci Total Environ 2020, 728, 138762. 85. Babin, S., Use of Weather Variables in SARS-CoV-2 Transmission Studies. Int J Infect Dis 2020, 100, 333-336. 86. Tanaka, K.; Shigemori, K.; Shionoya, M., Cyclic metallopeptides, cyclo[GlyL-Cys(terpyPt(II))](n)Cl(n). Chemical Communications 1999, (24), 24752476. 87. Maruyama, Y.; Hasegawa, K., Evaporation and drying kinetics of water-NaCl droplets via acoustic levitation. RSC Advances 2020, 10 (4), 1870-1877. 88. Gregson, F. K. A.; Robinson, J. F.; Miles, R. E. H.; Royall, C. P.; Reid, J. P., Drying Kinetics of Salt Solution Droplets: Water Evaporation Rates and Crystallization. J Phys Chem B 2019, 123 (1), 266-276. 89. Rezaei, M.; Netz, R. R., Airborne virus transmission via respiratory droplets: Effects of droplet evaporation and sedimentation. Curr Opin Colloid Interface Sci 2021, 55, 101471. 98 90. Kallapur, B.; Ramalingam, K.; Bastian; Mujib, A.; Sarkar, A.; Sethuraman, S., Quantitative estimation of sodium, potassium and total protein in saliva of diabetic smokers and nonsmokers: A novel study. J Nat Sci Biol Med 2013, 4 (2), 341-5. 91. Morawska, L.; Johnson, G. R.; Ristovski, Z. D.; Hargreaves, M.; Mengersen, K.; Corbett, S.; Chao, C. Y. H.; Li, Y.; Katoshevski, D., Size distribution and sites of origin of droplets expelled from the human respiratory tract during expiratory activities. Journal of Aerosol Science 2009, 40 (3), 256-269. 92. Millero, F. J., The apparent and partial molal volume of aqueous sodium chloride solutions at various temperatures. The Journal of Physical Chemistry 1970, 74 (2), 356-362. 93. Stull, D. R., Vapor Pressure of Pure Substances. Organic and Inorganic Compounds. Industrial & Engineering Chemistry 1947, 39 (4), 517-540. 94. Vargaftik, N. B.; Volkov, B. N.; Voljak, L. D., International Tables of the Surface Tension of Water. Journal of Physical and Chemical Reference Data 1983, 12 (3), 817-820. 95. Bolz, R. E., CRC Handbook of Tables for Applied Engineering Science. CRC press: 1973. 96. Somsen, G. A.; van Rijn, C.; Kooij, S.; Bem, R. A.; Bonn, D., Small droplet aerosols in poorly ventilated spaces and SARS-CoV-2 transmission. The Lancet Respiratory Medicine 2020, 8 (7), 658-659. 97. Zhao, H.; Lu, X.; Deng, Y.; Tang, Y.; Lu, J., COVID-19: asymptomatic carrier transmission is an underestimated problem. Epidemiol Infect 2020, 148, e116. 98. He, X.; Lau, E. H. Y.; Wu, P.; Deng, X.; Wang, J.; Hao, X.; Lau, Y. C.; Wong, J. Y.; Guan, Y.; Tan, X.; Mo, X.; Chen, Y.; Liao, B.; Chen, W.; Hu, F.; Zhang, Q.; Zhong, M.; Wu, Y.; Zhao, L.; Zhang, F.; Cowling, B. J.; Li, F.; Leung, G. M., Temporal dynamics in viral shedding and transmissibility of COVID-19. Nat Med 2020, 26 (5), 672-675. 99. Allen, M. D.; Raabe, O. G., Re-evaluation of millikan's oil drop data for the motion of small particles in air. Journal of Aerosol Science 1982, 13 (6), 537547. 100. Netz, R. R.; Eaton, W. A., Physics of virus transmission by speaking droplets. Proc Natl Acad Sci U S A 2020, 117 (41), 25209-25211. 101. Jamshidi, S.; Baniasad, M.; Niyogi, D., Global to USA County Scale Analysis of Weather, Urban Density, Mobility, Homestay, and Mask Use on COVID19. Int J Environ Res Public Health 2020, 17 (21). 102.Ogaugwu, C.; Mogaji, H.; Ogaugwu, E.; Nebo, U.; Okoh, H.; Agbo, S.; Agbon, A., Effect of Weather on COVID-19 Transmission and Mortality in Lagos, Nigeria. Scientifica (Cairo) 2020, 2020, 2562641. 103. Malki, Z.; Atlam, E. S.; Hassanien, A. E.; Dagnew, G.; Elhosseini, M. A.; Gad, I., Association between weather data and COVID-19 pandemic predicting 99 mortality rate: Machine learning approaches. Chaos Solitons Fractals 2020, 138, 110137. 104. Chanda, A., COVID-19 in India: transmission dynamics, epidemiological characteristics, testing, recovery and effect of weather. Epidemiol Infect 2020, 148, e182. 105. Hughes, V. C., The Effect of Temperature, Dewpoint, and Population Density on COVID-19 Transmission in the United States: A Comparative Study. American Journal of Public Health Research 2020, 8 (4), 112-117. 106. Joonaki, E.; Hassanpouryouzband, A.; Heldt, C. L.; Areo, O., Surface Chemistry Can Unlock Drivers of Surface Stability of SARS-CoV-2 in a Variety of Environmental Conditions. Chem 2020, 6 (9), 2135-2146. 107. Beckmann, M.; Václavík, T.; Manceur, A. M.; Šprtová, L.; von Wehrden, H.; Welk, E.; Cord, A. F.; Tatem, A., glUV: a global UV-B radiation data set for macroecological studies. Methods in Ecology and Evolution 2014, 5 (4), 372383. 108.Sajadi, M. M.; Habibzadeh, P.; Vintzileos, A.; Shokouhi, S.; Miralles-Wilhelm, F.; Amoroso, A., Temperature, Humidity and Latitude Analysis to Predict Potential Spread and Seasonality for COVID-19. SSRN 2020, 3550308. 109. Gupta, S.; Raghuwanshi, G. S.; Chanda, A., Effect of weather on COVID-19 spread in the US: A prediction model for India in 2020. Sci Total Environ 2020, 728, 138860. 110. Zhao, L.; Qi, Y.; Luzzatto-Fegiz, P.; Cui, Y.; Zhu, Y., COVID-19: Effects of Environmental Conditions on the Propagation of Respiratory Droplets. Nano Lett 2020, 20 (10), 7744-7750. 111. Pan, J.; Yao, Y.; Liu, Z.; Meng, X.; Ji, J. S.; Qiu, Y.; Wang, W.; Zhang, L.; Wang, W.; Kan, H., Warmer weather unlikely to reduce the COVID-19 transmission: An ecological study in 202 locations in 8 countries. Science of The Total Environment 2021, 753. 112. Yao, Y.; Pan, J.; Liu, Z.; Meng, X.; Wang, W.; Kan, H.; Wang, W., No association of COVID-19 transmission with temperature or UV radiation in Chinese cities. Eur Respir J 2020, 55 (5). 113. Smith, S. H.; Somsen, G. A.; van Rijn, C.; Kooij, S.; van der Hoek, L.; Bem, R. A.; Bonn, D., Aerosol persistence in relation to possible transmission of SARS-CoV-2. Phys Fluids (1994) 2020, 32 (10), 107108. 114. Organization, W. H., Coronavirus disease (COVID-19): How is it transmitted? 2021. 115. Evans, M. J., Avoiding COVID-19: Aerosol Guidelines. medRxiv 2020. 116. Pan, J.; Tang, J.; Caniza, M.; Heraud, J.-M.; Koay, E.; Lee, H. K.; Lee, C. K.; Li, Y.; Ruiz, A. N.; Santillan-Salas, C. F.; Marr, L., Correlating Indoor and Outdoor Temperature and Humidity in a Sample of Buildings in Tropical Climates. engrXiv 2021. 100 117. Nguyen, J. L.; Dockery, D. W., Daily indoor-to-outdoor temperature and humidity relationships: a sample across seasons and diverse climatic regions. Int J Biometeorol 2016, 60 (2), 221-9. 118. Nguyen, J. L.; Schwartz, J.; Dockery, D. W., The relationship between indoor and outdoor temperature, apparent temperature, relative humidity, and absolute humidity. Indoor Air 2014, 24 (1), 103-12. 119. Lee, K.; Lee, D., The Relationship Between Indoor and Outdoor Temperature in Two Types Of Residence. Energy Procedia 2015, 78, 2851-2856. 120. Habeebullah, T. M.; Abd El-Rahim, I. H. A.; Morsy, E. A., Impact of outdoor and indoor meteorological conditions on the COVID-19 transmission in the western region of Saudi Arabia. J Environ Manage 2021, 288, 112392. 121. Ahlawat, A.; Wiedensohler, A.; Mishra, S. K., An Overview on the Role of Relative Humidity in Airborne Transmission of SARS-CoV-2 in Indoor Environments. Aerosol and Air Quality Research 2020, 20 (9), 1856-1861. 122. Chong, K. L.; Ng, C. S.; Hori, N.; Yang, R.; Verzicco, R.; Lohse, D., Extended Lifetime of Respiratory Droplets in a Turbulent Vapor Puff and Its Implications on Airborne Disease Transmission. Phys Rev Lett 2021, 126 (3), 034502. 123. Phillips, N., The coronavirus is here to stay — here’s what that means. Nature Febuary 18, 2021, 2021, p 3. 124. Lavine, J. S.; Bjornstad, O. N.; Antia, R., Immunological characteristics govern the transition of COVID-19 to endemicity. Science 2021, 371 (6530), 741-745. 125. Xiao, F.; Sun, J.; Xu, Y.; Li, F.; Huang, X.; Li, H.; Zhao, J.; Huang, J.; Zhao, J., Infectious SARS-CoV-2 in Feces of Patient with Severe COVID-19. Emerg Infect Dis 2020, 26 (8), 1920-1922. 126. Zhang, N.; Gong, Y.; Meng, F.; Shi, Y.; Wang, J.; Mao, P.; Chuai, X.; Bi, Y.; Yang, P.; Wang, F., Comparative study on virus shedding patterns in nasopharyngeal and fecal specimens of COVID-19 patients. Sci China Life Sci 2020. 127. Wolfel, R.; Corman, V. M.; Guggemos, W.; Seilmaier, M.; Zange, S.; Muller, M. A.; Niemeyer, D.; Jones, T. C.; Vollmar, P.; Rothe, C.; Hoelscher, M.; Bleicker, T.; Brunink, S.; Schneider, J.; Ehmann, R.; Zwirglmaier, K.; Drosten, C.; Wendtner, C., Virological assessment of hospitalized patients with COVID-2019. Nature 2020, 581 (7809), 465-469. 128. Shutler, J. D.; Zaraska, K.; Holding, T.; Machnik, M.; Uppuluri, K.; Ashton, I. G. C.; Migdal, L.; Dahiya, R. S., Rapid Assessment of SARS-CoV-2 Transmission Risk for Fecally Contaminated River Water. ACS ES T Water 2021, 1 (4), 949-957. 129. Rimoldi, S. G.; Stefani, F.; Gigantiello, A.; Polesello, S.; Comandatore, F.; Mileto, D.; Maresca, M.; Longobardi, C.; Mancon, A.; Romeri, F.; Pagani, C.; Cappelli, F.; Roscioli, C.; Moja, L.; Gismondo, M. R.; Salerno, F., Presence 101 and infectivity of SARS-CoV-2 virus in wastewaters and rivers. Sci Total Environ 2020, 744, 140911. 130. Haramoto, E.; Malla, B.; Thakali, O.; Kitajima, M., First environmental surveillance for the presence of SARS-CoV-2 RNA in wastewater and river water in Japan. Sci Total Environ 2020, 737, 140405. 131. Prado, T.; Fumian, T. M.; Mannarino, C. F.; Resende, P. C.; Motta, F. C.; Eppinghaus, A. L. F.; Chagas do Vale, V. H.; Braz, R. M. S.; de Andrade, J.; Maranhao, A. G.; Miagostovich, M. P., Wastewater-based epidemiology as a useful tool to track SARS-CoV-2 and support public health policies at municipal level in Brazil. Water Res 2021, 191, 116810. 132. Randazzo, W.; Truchado, P.; Cuevas-Ferrando, E.; Simon, P.; Allende, A.; Sanchez, G., SARS-CoV-2 RNA in wastewater anticipated COVID-19 occurrence in a low prevalence area. Water Res 2020, 181, 115942. 133. de Oliveira, L. C.; Torres-Franco, A. F.; Lopes, B. C.; Santos, B.; Costa, E. A.; Costa, M. S.; Reis, M. T. P.; Melo, M. C.; Polizzi, R. B.; Teixeira, M. M.; Mota, C. R., Viability of SARS-CoV-2 in river water and wastewater at different temperatures and solids content. Water Res 2021, 195, 117002. 134. Tran, H. N.; Le, G. T.; Nguyen, D. T.; Juang, R. S.; Rinklebe, J.; Bhatnagar, A.; Lima, E. C.; Iqbal, H. M. N.; Sarmah, A. K.; Chao, H. P., SARS-CoV-2 coronavirus in water and wastewater: A critical review about presence and concern. Environ Res 2021, 193, 110265. 135. Shao, L.; Ge, S.; Jones, T.; Santosh, M.; Silva, L. F. O.; Cao, Y.; Oliveira, M. L. S.; Zhang, M.; BéruBé, K., The role of airborne particles and environmental considerations in the transmission of SARS-CoV-2. Geoscience Frontiers 2021, 12 (5). 136. Carducci, A.; Federigi, I.; Liu, D.; Thompson, J. R.; Verani, M., Making Waves: Coronavirus detection, presence and persistence in the water environment: State of the art and knowledge needs for public health. Water Res 2020, 179, 115907. 137. Ahmed, W.; Angel, N.; Edson, J.; Bibby, K.; Bivins, A.; O'Brien, J. W.; Choi, P. M.; Kitajima, M.; Simpson, S. L.; Li, J.; Tscharke, B.; Verhagen, R.; Smith, W. J. M.; Zaugg, J.; Dierens, L.; Hugenholtz, P.; Thomas, K. V.; Mueller, J. F., First confirmed detection of SARS-CoV-2 in untreated wastewater in Australia: A proof of concept for the wastewater surveillance of COVID-19 in the community. Sci Total Environ 2020, 728, 138764. 138. Cervantes-Avilés, P.; Moreno-Andrade, I.; Carrillo-Reyes, J., Approaches applied to detect SARS-CoV-2 in wastewater and perspectives post-COVID19. Journal of Water Process Engineering 2021, 40. 139. Pecson, B. M.; Darby, E.; Haas, C. N.; Amha, Y. M.; Bartolo, M.; Danielson, R.; Dearborn, Y.; Di Giovanni, G.; Ferguson, C.; Fevig, S.; Gaddis, E.; Gray, D.; Lukasik, G.; Mull, B.; Olivas, L.; Olivieri, A.; Qu, Y.; Consortium, S. A.C.-I., Reproducibility and sensitivity of 36 methods to quantify the SARS- 102 CoV-2 genetic signal in raw wastewater: findings from an interlaboratory methods evaluation in the U.S. Environmental Science: Water Research & Technology 2021, 7 (3), 504-520. 140. Alygizakis, N.; Markou, A. N.; Rousis, N. I.; Galani, A.; Avgeris, M.; Adamopoulos, P. G.; Scorilas, A.; Lianidou, E. S.; Paraskevis, D.; Tsiodras, S.; Tsakris, A.; Dimopoulos, M. A.; Thomaidis, N. S., Analytical methodologies for the detection of SARS-CoV-2 in wastewater: Protocols and future perspectives. Trends Analyt Chem 2021, 134, 116125. 141. Furuki, T.; Ito, T.; Asakawa, N.; Inoue, Y.; Sakurai, M., Effects of trehalose on the swelling behavior of hydrogel -visualization of the preferential hydration of disaeeharides. Chemistry Letters 2009, 38 (3), 264-265. 142. Kitajima, M.; Ahmed, W.; Bibby, K.; Carducci, A.; Gerba, C. P.; Hamilton, K. A.; Haramoto, E.; Rose, J. B., SARS-CoV-2 in wastewater: State of the knowledge and research needs. Sci Total Environ 2020, 739, 139076. 143. Chotiwan, N.; Brewster, C. D.; Magalhaes, T.; Weger-Lucarelli, J.; Duggal, N. K.; Ruckert, C.; Nguyen, C.; Garcia Luna, S. M.; Fauver, J. R.; Andre, B.; Gray, M.; Black, W. C. t.; Kading, R. C.; Ebel, G. D.; Kuan, G.; Balmaseda, A.; Jaenisch, T.; Marques, E. T. A.; Brault, A. C.; Harris, E.; Foy, B. D.; Quackenbush, S. L.; Perera, R.; Rovnak, J., Rapid and specific detection of Asian- and African-lineage Zika viruses. Sci Transl Med 2017, 9 (388). 144. Yang, B. Y.; Liu, X. L.; Wei, Y. M.; Wang, J. Q.; He, X. Q.; Jin, Y.; Wang, Z. J., Rapid and sensitive detection of human astrovirus in water samples by loopmediated isothermal amplification with hydroxynaphthol blue dye. BMC Microbiol 2014, 14, 38. 145. Huang, X.; Lin, X.; Urmann, K.; Li, L.; Xie, X.; Jiang, S.; Hoffmann, M. R., Smartphone-Based in-Gel Loop-Mediated Isothermal Amplification (gLAMP) System Enables Rapid Coliphage MS2 Quantification in Environmental Waters. Environ Sci Technol 2018, 52 (11), 6399-6407. 146. Li, J.; Zhu, Y.; Wu, X.; Hoffmann, M. R., Rapid Detection Methods for Bacterial Pathogens in Ambient Waters at the Point of Sample Collection: A Brief Review. Clin Infect Dis 2020, 71 (Suppl 2), S84-S90. 147. Fu, J.; Chiang, E. L. C.; Medriano, C. A. D.; Li, L.; Bae, S., Rapid quantification of fecal indicator bacteria in water using the most probable number - loop-mediated isothermal amplification (MPN-LAMP) approach on a polymethyl methacrylate (PMMA) microchip. Water Research 2021, 199. 148. Ning, B.; Yu, T.; Zhang, S.; Huang, Z.; Tian, D.; Lin, Z.; Niu, A.; Golden, N.; Hensley, K.; Threeton, B.; Lyon, C. J.; Yin, X. M.; Roy, C. J.; Saba, N. S.; Rappaport, J.; Wei, Q.; Hu, T. Y., A smartphone-read ultrasensitive and quantitative saliva test for COVID-19. Sci Adv 2021, 7 (2). 149. Ganguli, A.; Mostafa, A.; Berger, J.; Aydin, M. Y.; Sun, F.; Ramirez, S. A. S.; Valera, E.; Cunningham, B. T.; King, W. P.; Bashir, R., Rapid isothermal 103 amplification and portable detection system for SARS-CoV-2. Proc Natl Acad Sci U S A 2020, 117 (37), 22727-22735. 150. Prado, T.; Fumian, T. M.; Mannarino, C. F.; Maranhao, A. G.; Siqueira, M. M.; Miagostovich, M. P., Preliminary results of SARS-CoV-2 detection in sewerage system in Niteroi municipality, Rio de Janeiro, Brazil. Mem Inst Oswaldo Cruz 2020, 115, e200196. 151. Barril, P. A.; Pianciola, L. A.; Mazzeo, M.; Ousset, M. J.; Jaureguiberry, M. V.; Alessandrello, M.; Sanchez, G.; Oteiza, J. M., Evaluation of viral concentration methods for SARS-CoV-2 recovery from wastewaters. Sci Total Environ 2021, 756, 144105. 152. Xie, X.; Bahnemann, J.; Wang, S.; Yang, Y.; Hoffmann, M. R., "Nanofiltration" Enabled by Super-Absorbent Polymer Beads for Concentrating Microorganisms in Water Samples. Sci Rep 2016, 6, 20516. 153. Chen, W.; Wang, T.; Dou, Z.; Xie, X., Self-Driven "Microfiltration" Enabled by Porous Superabsorbent Polymer (PSAP) Beads for Biofluid Specimen Processing and Storage. ACS Mater Lett 2020, 2 (11), 1545-1554. 154. Fagnant, C. S.; Sanchez-Gonzalez, L. M.; Zhou, N. A.; Falman, J. C.; Eisenstein, M.; Guelig, D.; Ockerman, B.; Guan, Y.; Kossik, A. L.; Linden, Y. S.; Beck, N. K.; Wilmouth, R.; Komen, E.; Mwangi, B.; Nyangao, J.; Shirai, J. H.; Novosselov, I.; Borus, P.; Boyle, D. S.; Meschke, J. S., Improvement of the Bag-Mediated Filtration System for Sampling Wastewater and Wastewater-Impacted Waters. Food Environ Virol 2018, 10 (1), 72-82. 155. Tetko, I. V.; Livingstone, D. J.; Luik, A. I., Neural network studies. 1. Comparison of overfitting and overtraining. Journal of Chemical Information and Computer Sciences 1995, 35 (5), 826-833. 156. Sjöberg, J.; Ljung, L., Overtraining, Regularization, and Searching for Minimum in Neural Networks. IFAC Proceedings Volumes 1992, 25 (14), 7378. 157. Flangan, J.; Jones, D. P.; Griffith, W. P.; Skapski, A. C.; West, A. P., On the existence of peroxocarbonates in aqueous solution. Journal of the Chemical Society, Chemical Communications 1986, (1), 20-21. 158. Bakhmutova-Albert, E. V.; Yao, H.; Denevan, D. E.; Richardson, D. E., Kinetics and Mechanism of Peroxymonocarbonate Formation. Inorganic Chemistry 2010, 49 (24), 11287-11296. 159. Richardson, D. E.; Yao, H.; Frank, K. M.; Bennett, D. A., Equilibria, Kinetics, and Mechanism in the Bicarbonate Activation of Hydrogen Peroxide: Oxidation of Sulfides by Peroxymonocarbonate. Journal of the American Chemical Society 2000, 122 (8), 1729-1739. 160. Yao, H.; Richardson, D. E., Epoxidation of Alkenes with BicarbonateActivated Hydrogen Peroxide. Journal of the American Chemical Society 2000, 122 (13), 3220-3221. 104 161. Ho, K.-P.; Chan, T. H.; Wong, K.-Y., An environmentally benign catalytic system for alkene epoxidation with hydrogen peroxide electrogenerated in situ. Green Chemistry 2006, 8 (10), 900-905. 162. Bennett, D. A.; Yao, H.; Richardson, D. E., Mechanism of Sulfide Oxidations by Peroxymonocarbonate. Inorganic Chemistry 2001, 40 (13), 2996-3001. 163. Yang, X.; Duan, Y.; Wang, J.; Wang, H.; Liu, H.; Sedlak, D. L., Impact of Peroxymonocarbonate on the Transformation of Organic Contaminants during Hydrogen Peroxide in Situ Chemical Oxidation. Environmental Science & Technology Letters 2019, 6 (12), 781-786. 164. Vione, D.; Maurino, V.; Minero, C.; Pelizzetti, E., The atmospheric chemistry of hydrogen peroxide: a review. Ann Chim 2003, 93 (4), 477-488. 165. Goosse, H.; Barriat, P.-Y.; Brovkin, V.; Klein, F.; Meissner, K. J.; Menviel, L.; Mouchet, A., Changes in atmospheric CO2 concentration over the past two millennia: contribution of climate variability, land-use and Southern Ocean dynamics. Climate Dynamics 2022. 166. NIST.gov Carbon dioxide. https://webbook.nist.gov/cgi/cbook.cgi?ID=C124389&Mask=10#Solubility. 167. Enami, S.; Mishra, H.; Hoffmann, M. R.; Colussi, A. J., Protonation and Oligomerization of Gaseous Isoprene on Mildly Acidic Surfaces: Implications for Atmospheric Chemistry. Journal of Physical Chemistry A 2012, 116 (24), 6027-6032. 168. Gu, A. Y.; Musgrave, C.; Goddard, W. A.; Hoffmann, M. R.; Colussi, A. J., Role of Ferryl Ion Intermediates in Fast Fenton Chemistry on Aqueous Microdroplets. Environmental Science & Technology 2021, 55 (21), 1437014377. 169. Zhu, L.; Nicovich, J. M.; Wine, P. H., Temperature-dependent kinetics studies of aqueous phase reactions of hydroxyl radicals with dimethylsulfoxide, dimethylsulfone, and methanesulfonate. Aquatic Sciences 2003, 65 (4), 425435. 170. Mishra, H.; Enami, S.; Nielsen, R. J.; Stewart, L. A.; Hoffmann, M. R.; Goddard, W. A.; Colussi, A. J., Bronsted basicity of the air-water interface. Proceedings of the National Academy of Sciences of the United States of America 2012, 109 (46), 18679-18683. 105 Appendix A: Coordinates of DFT-optimized Structures no ligand O14 0.7680228217 -1.3052848462 -2.2919432579 H2 0.7584223414 -2.2501763863 -2.0661375355 Fe4 -0.7497366749 -0.0067991904 -2.0394507429 O5 -1.0603443397 0.3322663998 -4.0031504829 H19 -0.4036576990 0.0111010687 -4.6772658472 H20 -1.9243029173 0.7630532931 -4.2591255829 O9 -2.3928455969 0.6745027680 -1.3910942292 H23 -2.7752923342 0.7883218354 -0.5090241509 O10 -3.3577077614 1.4137914804 -3.6805626934 H11 -4.2761494874 1.2317566774 -3.9309555931 H12 -3.1649396027 1.2110512342 -2.6989924854 O25 0.8722547616 0.7285103625 -0.8588231959 H29 1.4614618184 -0.0136329128 -1.1579825200 H30 1.3011820145 1.5741223300 -0.6654673386 O15 1.0136826305 -0.8901396470 -4.8566046002 H16 1.8627748015 -0.8113791205 -5.3154377222 106 H17 1.0934667770 -1.1445303650 -3.8734238435 O14 0.7680228217 -1.3052848462 -2.2919432579 H2 0.7584223414 -2.2501763863 -2.0661375355 Fe4 -0.7497366749 -0.0067991904 -2.0394507429 O5 -1.0603443397 0.3322663998 -4.0031504829 H19 -0.4036576990 0.0111010687 -4.6772658472 H20 -1.9243029173 0.7630532931 -4.2591255829 O9 -2.3928455969 0.6745027680 -1.3910942292 H23 -2.7752923342 0.7883218354 -0.5090241509 O10 -3.3577077614 1.4137914804 -3.6805626934 H11 -4.2761494874 1.2317566774 -3.9309555931 H12 -3.1649396027 1.2110512342 -2.6989924854 O25 0.8722547616 0.7285103625 -0.8588231959 H29 1.4614618184 -0.0136329128 -1.1579825200 H30 1.3011820145 1.5741223300 -0.6654673386 O15 1.0136826305 -0.8901396470 -4.8566046002 H16 1.8627748015 -0.8113791205 -5.3154377222 dma-Fe(II) 107 H17 1.0934667770 -1.1445303650 -3.8734238435 O14 -0.1655340093 -1.7268063039 -2.6095353944 H2 -0.3944219122 -2.4108828075 -3.2632278177 Fe4 -0.7707577602 0.2290541593 -2.7723227762 O5 1.5230329615 0.0171089392 -3.2915139341 H19 2.2534865858 0.4255137552 -2.7999395253 H20 1.2970082143 -0.9133447146 -2.9833107091 O9 -1.0066552088 1.9487382218 -2.0424117983 H23 -0.4984929168 2.6898021231 -1.6832136443 O10 -2.7446668203 -0.2142757076 -1.8892807656 H11 -2.6311698838 -0.9745628100 -1.2442971945 H12 -2.9093194947 0.6370485030 -1.4418975663 C15 -0.3704657242 1.5432800780 -5.3973751225 N16 -1.3494873767 0.5781975649 -4.8293540290 C17 -1.3520134653 -0.6929262087 -5.5942865208 C18 -2.7126537661 1.1736537290 -4.7711098241 H17 -0.3698260792 2.4343964075 -4.7735754166 tma-Fe(II) 108 H18 0.6210145965 1.0993010209 -5.3613931753 H21 -0.6314231817 1.8119699517 -6.4290141523 H22 -3.3882196531 0.4632592938 -4.3008771379 H24 -2.6684113541 2.0624700457 -4.1458264821 H25 -3.0734805126 1.4336554283 -5.7743633641 H26 -2.0433699061 -1.3905771743 -5.1239071940 H27 -1.6590481908 -0.5333790764 -6.6357488419 H28 -0.3506032286 -1.1197469178 -5.5796786368 O25 -1.7216096955 -2.2138716360 -0.6271053384 H29 -0.9584423562 -2.1602443259 -1.3060684009 H30 -1.5271258971 -2.7088429317 0.1810021166 chloride-Fe(II) O14 0.3976702798 -1.4773996919 -2.7887423748 H2 0.3772259047 -2.3828298976 -2.4402323102 Fe4 -0.7744413237 0.0396878672 -2.3567048422 O5 -0.4405337329 0.9590007201 -4.1347828722 H19 0.2048408237 0.4563172448 -4.7563223266 H20 -0.7042765410 1.8602575824 -4.3663319178 109 Cl7 -2.0465718251 2.1906876266 -1.5487949815 O10 -2.7698035681 -0.5746316041 -2.2415785220 H11 -3.2845621037 -1.3024474873 -2.6210595601 H12 -3.2955943711 0.1981029804 -1.9355223327 O25 0.0677119898 0.4013039493 -0.4685131437 H29 0.9083819222 0.1111906487 -0.0839362415 H30 -0.3260204012 1.2087914841 -0.0767820033 O15 1.1076792753 -0.7458040964 -5.0658618888 H15 2.0233219486 -0.7675916562 -5.3795276228 H16 0.9471859164 -1.2229291780 -4.1439162864 oxolate-Fe(II) C1 -1.3518626782 1.1887077594 -4.9834823855 C2 -2.5532196878 1.7486002828 -4.1258181644 O3 -0.5120180158 0.4353111114 -4.2416854379 O4 -2.4781394429 1.3550388065 -2.8363637288 O5 -3.4346454728 2.4545424885 -4.6020014847 O6 -1.2233866228 1.4270608016 -6.1786955916 Fe12 -0.9645174245 0.2308612041 -2.3625262650 110 O13 0.7756771704 -0.8846087072 -2.4678183860 H14 1.4092866072 -1.4248819202 -1.9739703789 H15 0.9923498594 -0.7718231176 -3.4209800224 O19 -1.9321294166 0.3648460913 -0.5381685054 H20 -1.9192305335 0.1031588534 0.3936904919 H21 -2.6949722511 0.9318662328 -0.7925921136 malonate-Fe(II) C1 -1.4616918542 1.1304608514 -5.1661571065 C2 -1.2714329443 2.5023583365 -4.5030376946 O3 -1.0968863375 0.0667678130 -4.4235751927 C4 -1.8047285398 2.6919085695 -3.0755316379 H5 -0.1931553953 2.6889511511 -4.4657409888 O6 -1.8804805410 1.0237516653 -6.3236382582 O14 -0.8041295636 -2.0812953667 -3.1379597763 H2 -0.7242041158 -3.0082010414 -2.8728350970 H3 -0.8822713734 -1.9276317995 -4.1088390855 Fe4 -1.0391265617 -0.1163462224 -2.5090022568 O9 -1.2111909422 -0.0657317395 -0.4407086955 111 H23 -1.1927156691 -0.5752420184 0.3813569757 H11 -1.4663830043 0.8814262787 -0.3420113948 H21 -1.7252450582 3.2534853623 -5.1385035080 O21 -2.4638230784 3.6874660982 -2.7578290328 O23 -1.4646765589 1.7322976415 -2.1928647874 tartrate-Fe(II) O14 -1.6514421105 -1.8037446986 -3.2016132395 H2 -0.8782166179 -2.3375215406 -3.4554012886 H3 -2.0250784525 -1.2961771971 -3.9689215532 Fe4 -1.1622555927 -0.0680175937 -2.0574845815 O5 -0.4773773402 0.2721723939 -0.0245337343 H19 -0.0197998813 1.1195827942 -0.2373498981 H20 -0.9121196573 0.2754788917 0.8419678617 O9 -3.2069017891 0.3054045129 -1.6946044387 H23 -3.9243495893 0.0649971299 -1.0919904700 H11 -3.4851586737 0.5599383102 -2.6048014572 C1 0.4730157885 1.9902243702 -3.1855779800 C3 -0.0656901050 3.0697801899 -4.1047778962 112 O4 -0.1840267605 1.7659215797 -2.0719341901 O16 1.4376358294 1.2651186962 -3.5768053027 O6 0.2189023788 4.3719950158 -3.5274286290 H17 0.4206992235 2.9654748343 -5.0755107362 C8 -1.5954990426 2.9227511365 -4.2872196458 C9 -2.0158122284 1.5528200241 -4.8450934683 H21 -2.0597191075 3.0866891425 -3.3184752945 O11 -2.0231421899 3.9878224778 -5.1797119088 O12 -2.1214182042 0.5728996453 -3.9639221601 O13 -2.2477233008 1.4471339514 -6.0673969992 H24 -0.2777162296 5.0334969220 -4.0550501423 H25 -2.1656682464 3.5877255092 -6.0674881565 O25 0.8189499681 -0.9216095344 -2.3833207100 H26 1.2971945788 -0.1608816531 -2.8787536239 H27 1.3287089378 -1.2212996424 -1.6126850432 catechol-Fe(II) Fe12 -0.5956779278 -0.9401152716 0.2442911944 O19 -0.0641533657 -2.5860914580 -0.8414605665 113 H20 -0.4857873881 -3.4434153186 -0.9876671207 H21 0.7894117351 -2.3994176083 -1.4220580814 O3 0.9037483723 0.0994699490 -0.5350751272 O4 -1.1626559414 0.8549187687 0.8805537158 O2 -1.9006279704 -1.8627505236 1.5280222944 O9 1.9342327009 -1.6494963794 -1.9434301703 H14 -1.8529718429 -2.7250732998 1.9639493732 H15 -2.4626456572 -1.1440970974 2.0491894695 H16 1.7221785166 -0.7547518909 -1.4606774503 H17 2.2648531891 -1.5603602027 -2.8480258542 C18 1.7853262644 2.3837395314 -0.6464735707 C19 1.6252613215 3.7306074347 -0.2932518041 C20 0.5245932933 4.1333091482 0.4623212547 C21 -0.4308054519 3.1944058586 0.8748758531 C22 -0.2771226398 1.8582754859 0.5273418125 C23 0.8405206769 1.4491445309 -0.2410467840 H24 2.6380529391 2.0553087099 -1.2254254775 H25 2.3624061718 4.4568960240 -0.6084718613 114 H26 0.4040237131 5.1732759769 0.7345625678 H27 -1.2892951836 3.4918655486 1.4619238817 O24 -2.8870026391 0.1722230671 2.5095915311 H28 -2.2287624942 0.7235761957 1.9222535597 H29 -3.7774477684 0.5429349855 2.5857212916 benzoquinone-Fe(II) O14 -2.1186606245 -2.0330178699 -3.4766712638 H2 -1.7110437184 -2.8065899684 -3.8902478609 Fe4 -1.5826270926 -0.2725983695 -2.9442561269 O5 0.4667181523 -0.6707772548 -2.4140302042 H19 0.7853925743 -0.0436280888 -1.6736822804 H20 0.7894363751 -1.5789341100 -2.3176158604 O9 -1.7095024198 1.0633733797 -1.5067325193 H23 -2.5292589483 1.2951705630 -1.0419301855 C12 -0.9835245578 4.4261319780 -4.2637500302 C13 -0.7667867987 4.7969576318 -5.6762862064 C14 -0.7056628207 3.7041517454 -6.6690358545 C15 -0.8387391407 2.4191842623 -6.2960011504 115 C16 -1.0456754895 2.0722092741 -4.8896318749 C17 -1.1152442243 3.1409290226 -3.8902286842 H18 -1.0313787026 5.2453880592 -3.5602276380 O19 -0.6406081947 5.9946269726 -6.0237298167 H21 -0.5506789537 4.0018632805 -7.6963036701 H22 -0.8011329848 1.5959988243 -6.9950094579 O22 -1.1620190215 0.8568588095 -4.5495591471 H24 -1.2848007384 2.8159522541 -2.8701704676 O24 0.6212912162 1.0513770642 -0.5890245469 H25 1.1273764089 1.8511384761 -0.3856808558 H26 -0.3850810218 1.2049625036 -0.7995724634 O25 -3.8686768794 -0.3105215563 -3.3809351666 H27 -3.6100572652 -1.2646582916 -3.5814909317 H28 -4.6231026128 -0.2362409792 -2.7756407246 1,4-naphthaquinone-Fe(II) O14 -2.3330166693 -1.8529845874 -3.1567442412 H2 -2.1134553449 -2.2013546529 -4.0393784674 Fe4 -2.0369847501 0.1127559584 -2.7346519190 116 O5 -0.0259184080 -0.2820138265 -2.0054277126 H19 0.1336151055 0.3770723073 -1.3079220350 H20 0.0344234685 -1.2316951577 -1.6763840411 O9 -2.0594151145 1.6108230847 -1.5786382208 H23 -2.6538379048 1.9814170321 -0.9092494033 C12 -0.6673207976 4.4013859133 -4.4760857333 C13 -0.3348266812 4.6574404834 -5.8887552650 C14 -0.4403463468 3.5185485271 -6.8381550628 C15 -0.8513348222 2.2488021337 -6.3809740599 C16 -1.1765864991 2.0606680104 -4.9594305499 C17 -1.0613474043 3.1926932120 -4.0373574282 H18 -0.5728363762 5.2499105422 -3.8127930005 O19 0.0238390437 5.7978235123 -6.2625896270 O22 -1.5530482676 0.9283364746 -4.5281508752 H24 -1.3173729859 2.9720614262 -3.0025330623 O24 -0.6292734788 -2.6999963722 -1.4539537679 H25 -0.4185315064 -3.6185476096 -1.2356206943 H26 -1.4075003449 -2.5671155202 -2.1116253960 117 O25 -4.2599455684 -0.2228817903 -3.0519551784 H27 -4.0322538327 -1.1904662335 -3.1058509645 H28 -5.0274303788 -0.0117667436 -2.4984258317 C25 -0.1327384985 3.7016268505 -8.1878418825 C26 -0.2306508689 2.6345405221 -9.0778808383 C27 -0.6379922269 1.3753073624 -8.6246998064 C28 -0.9479634240 1.1824469747 -7.2819352209 H17 0.1799122487 4.6816649576 -8.5180291674 H21 0.0091713829 2.7807248456 -10.1218902618 H22 -0.7126979777 0.5497558889 -9.3186721519 H32 -1.2653547612 0.2183032713 -6.9129865198 Fe1 -0.0201860345 -0.0527985469 -1.5477579672 C2 0.2309594348 -0.5387212941 1.3284412618 H3 1.3159592907 -0.4577066522 1.3206807399 H4 -0.0443434899 -0.9914672865 2.2802826269 C5 -0.3938993231 0.8493336457 1.2016493685 H6 -1.4768699862 0.7673251401 1.1280123340 Edta-Fe(II) 118 H7 -0.1686644532 1.4402745669 2.0897425281 N8 0.0656556284 1.5531936413 -0.0457658184 N9 -0.1465031056 -1.4332481033 0.1815394047 C10 -0.9429043415 2.5550688064 -0.5162176409 H11 -0.4554144080 3.2367788674 -1.2128100505 H12 -1.3490559029 3.1464925086 0.3044915698 C13 -1.5210055282 -1.9736284817 0.2507162620 H14 -2.2347133009 -1.1525255293 0.2993244005 H15 -1.7405366469 -2.4810626367 -0.6890074449 C16 0.9102514866 -2.4570427149 -0.0983692907 H17 1.2861334043 -2.9106441524 0.8142741908 H18 0.4726075719 -3.2463783370 -0.7102906298 C19 1.4523813934 2.1037980291 0.0260191245 H20 2.1315944497 1.3111421669 0.3268475968 H21 1.7424284506 2.4260575356 -0.9683609208 C22 2.0740081352 -1.8402677262 -0.8998822036 O23 3.2157526157 -2.2986553661 -0.8094911302 O24 1.7188718275 -0.8098844195 -1.6745545121 119 C25 -1.8664813215 -2.9477064727 1.3671694415 O26 -2.9823940627 -3.4263989896 1.5258391652 O27 -0.8044216259 -3.2552904155 2.1845071381 C28 1.5660572782 3.2549419054 1.0034869647 O29 1.4894827336 3.1540821300 2.2273547232 O30 1.7158020658 4.4588388207 0.3676049058 C31 -2.0961154836 1.8544773266 -1.2661276086 O32 -3.2182232445 2.3628134285 -1.3113295574 O33 -1.7463647949 0.6997011022 -1.8401363165 H34 -1.0618301831 -3.9012358449 2.8762685144 H41 1.7789520038 5.2025571125 1.0052252337 No Ligand-Fe(IV) O14 -0.7676618425 -1.4982343393 -1.9747904571 H2 -1.4603922618 -2.0492513920 -2.3760190591 Fe4 -0.3771310312 0.2612086252 -2.4491642716 O5 -1.2255465979 -0.1137385897 -4.3173038879 H19 -0.9868767899 0.5176026774 -5.0152567168 H20 -2.2282784421 -0.2120524595 -4.1805713315 120 O9 -1.8757872386 1.1643433113 -1.7776676259 H23 -1.8222603195 2.1370839914 -1.7423046114 O10 -3.5410971322 -0.0349025390 -3.3144509048 H11 -4.3733520407 -0.4848852251 -3.1147829408 H12 -3.1219901715 0.4726787970 -2.5407150427 O25 1.0390217891 -0.2117036898 -1.0544518193 H29 0.8426441574 -1.1446232389 -0.8225357848 H30 1.9464288431 0.1041972450 -0.9403174720 O15 0.4919303872 1.5436287290 -3.0207902164 O14 0.2674558505 -1.3794106640 -2.7588283622 H2 0.1159597623 -2.3082724367 -3.0138471775 Fe4 -1.2238841767 -0.1876842959 -2.4887738774 O5 1.4084075656 0.2019226039 -4.5505656020 H19 2.1370135086 0.7888971378 -4.2909003747 H20 1.2014664353 -0.4672458846 -3.8292282519 O9 -0.5383389718 1.4469013938 -2.0067116754 H23 -0.1823221579 1.6285030809 -1.1181767825 dma-Fe(IV) 121 O15 -2.5597056719 -1.1505109371 -2.6038165346 O10 -0.9454140708 -0.7710294939 -0.3933960884 H11 -0.1440056731 -1.3249520780 -0.4926067915 H12 -1.7005156098 -1.2835516030 -0.0522736931 H13 -0.3820607680 0.8055453403 -4.5552807397 N16 -1.3718107850 0.5474139685 -4.4011932344 C17 -1.7473922921 -0.4967530355 -5.3897000424 C18 -2.2584289754 1.7393835961 -4.4411789683 H22 -3.2757345509 1.4299441311 -4.2068758849 H24 -1.9116677037 2.4471868986 -3.6941853312 H25 -2.2412562320 2.1978102885 -5.4333070991 H26 -2.7099995782 -0.9171316614 -5.1114728369 H27 -1.7920088278 -0.0719732354 -6.3955460585 H28 -0.9956903584 -1.2807216957 -5.3663394575 O14 -0.1219741393 -1.7468986180 -3.2158580786 H2 -0.6096047449 -2.5871383628 -3.2253702642 Fe4 -0.9465209825 -0.1360221249 -2.7506010041 tma-Fe(IV) 122 O5 1.1497702089 0.2124356110 -2.8163068011 H19 1.7273383080 0.6408188564 -2.1666872224 H20 1.3492389342 -0.7462703168 -2.9848872383 O9 -1.1592430766 1.4118689302 -1.7373747976 H23 -0.4261571934 1.9885561648 -1.4589095599 O15 -2.4539037517 -0.7254592138 -2.5025316566 O10 -3.8504478122 1.6793595182 -1.5420423367 H11 -4.0805065378 0.7329526296 -1.5881716589 H12 -2.8587845445 1.7255377424 -1.4955221646 C15 -0.1649511077 1.1252158610 -5.3840875548 N16 -1.4081500295 0.6494839274 -4.7180376158 C17 -2.0252329269 -0.4604072149 -5.5002952494 C18 -2.3812211988 1.7714299073 -4.5475343910 H17 0.2955054142 1.8971750752 -4.7744694571 H18 0.5333063299 0.2976498904 -5.4746687769 H21 -0.3922057191 1.5274899372 -6.3768099341 H22 -3.2336070476 1.4330645256 -3.9669507927 H24 -1.9073726560 2.5677048271 -3.9827798142 123 H25 -2.7028040285 2.1409068078 -5.5267923690 H26 -2.9145295579 -0.8000535808 -4.9786163153 H27 -2.2874137605 -0.1176659393 -6.5065740724 H28 -1.3182667238 -1.2837044734 -5.5605251724 chloride-Fe(IV) O14 -0.2567421246 -1.6753994051 -2.4544211939 H2 0.4234916029 -2.1870563155 -2.9360707400 Fe4 -0.2585410231 0.1073587212 -2.6339383757 O5 -1.4025090666 1.1067093371 -3.9558461383 H19 -1.0039047482 1.3825874610 -4.7942221047 H20 -2.3926097465 1.3177897201 -3.8353063178 Cl7 -2.0370420418 0.4222682937 -1.0038266588 O10 -3.7930373220 1.5064164494 -3.2401795179 H11 -4.6547677102 1.1984154573 -3.5607950357 H12 -3.6045148703 1.2226194088 -2.3106170341 O25 0.6728210513 1.3911790114 -1.3136725820 H29 1.5774761917 1.7226679869 -1.4245130099 H30 0.2156806574 1.6496971350 -0.4938924121 124 O15 1.0383441099 0.2068707632 -3.6384647994 oxolate-Fe(IV) C1 -0.8326044840 3.8464913545 -3.3168126623 C2 -1.5379645714 2.6302145144 -3.9087700508 O3 -1.6246800787 0.1953742130 -0.8615535043 O4 -2.3150763713 1.9695621752 -3.0876873066 O5 -1.2526116725 2.2325784777 -5.0773239562 O6 -1.1058372143 5.0040765779 -3.6018045093 O9 -1.0515866981 -1.9584668322 -2.3840111876 H10 -0.7824834896 -2.2788578886 -1.5055808062 H11 -0.5711698928 -2.3914120598 -3.1098370371 Fe12 -1.4848238163 0.1196158623 -2.4832132691 O13 0.2470919775 0.8173180211 -2.5862786070 H14 0.2594886869 2.5080330233 -2.3143586270 H15 0.7384598900 0.5992527177 -3.4044875255 O16 -1.4117585753 -0.2479259443 -4.5262738027 H17 -2.1532253612 -0.7773840329 -4.8686730954 H18 -1.3589046776 0.6716775381 -4.9974329570 125 O17 0.1370350613 3.5046483069 -2.4302358273 O18 -3.5400948776 -0.2696288763 -2.7589390177 H19 -4.0433764569 -0.6580468114 -2.0228138324 H20 -3.8384014829 0.6426104946 -2.9789504957 malonate-Fe(IV) C1 -1.0169153310 1.3080309354 -5.1643155328 C2 -1.3739699689 2.6769038444 -4.5472824056 O3 -1.1950598136 0.2366433093 -4.3711230525 C4 -2.0669554798 2.4361975529 -3.2358384413 H5 -0.4479801240 3.2330343770 -4.4064913840 O6 -0.5884048921 1.2179720765 -6.3186246903 O14 -1.3552295741 -1.9125338633 -3.1069856091 H2 -1.8545909338 -2.5931746663 -2.6296919226 H3 -1.5797833179 -1.8302019128 -4.0564140206 Fe4 -1.0948863774 0.0161832144 -2.4310777849 O5 0.9447216702 -0.1061917349 -2.8237499204 H19 1.6158969561 -0.2443484949 -2.1378884597 H20 1.2763759724 -0.1279179505 -3.7362423652 126 O9 -3.2696705630 -0.0069822216 -2.1251063657 H23 -3.4485606306 -0.1430629959 -1.1775795201 H11 -3.6324066402 0.8741827051 -2.4178876229 O15 -0.8661509385 -0.3880427825 -0.8551450215 H21 -2.0140813211 3.2147358942 -5.2398838509 O21 -3.3076421273 2.5377957425 -3.0990319894 O23 -1.2846358028 1.9169988091 -2.2742739389 tartrate-Fe(IV) O14 -1.8801956895 -1.7716265730 -2.8299685378 H2 -1.6315618328 -2.6364459058 -2.4678596123 H3 -1.8315928359 -1.7052421369 -3.8068413722 Fe4 -1.1220764282 0.0374152509 -2.1625962285 O5 0.7973379774 -0.7906793194 -2.6177268308 H19 1.3377312942 -0.0396253888 -3.0248419096 H20 1.2431597112 -1.1520996266 -1.8319777290 O9 -3.0737704682 0.7169229333 -1.8290321371 H23 -3.4017891893 0.7931860220 -0.9182440068 H11 -3.7719962682 0.7000856463 -2.5028347961 127 O15 -0.9959497283 -0.4214955489 -0.5944910721 C1 0.4705178640 2.1322604441 -3.1307995517 C3 -0.1703559743 3.1981177818 -3.9976274555 O4 -0.2382817293 1.7433499405 -2.0783573587 O16 1.5201339039 1.5510432504 -3.5017462528 O6 -0.1492938366 4.4762590210 -3.3194029374 H17 0.3773252730 3.2456446349 -4.9393791956 C8 -1.6385096893 2.7863906783 -4.2706634896 C9 -1.7650436030 1.3776720906 -4.8684928553 H21 -2.1606536056 2.8257563865 -3.3172677229 O11 -2.2210632654 3.7749793297 -5.1519207094 O12 -1.5140426455 0.3534024263 -4.0607964966 O13 -2.1067380684 1.2436720694 -6.0555170603 H24 -0.6681356974 5.1103850272 -3.8583643017 H25 -2.3413207967 3.3707520791 -6.0402548917 catechol-Fe(IV) Fe12 -0.8043398489 -0.7618731585 0.1663471937 O19 0.0945217808 -2.5595601633 -0.1652404955 128 H20 1.0451791380 -2.4446699329 -0.3454811913 H21 -0.3397245506 -3.2500927179 -0.6890240956 O14 -2.3542943879 -1.3319345919 0.0538900022 O3 1.0412266998 -0.1896954464 0.2247675633 O4 -1.2219017022 1.0715491300 0.3032417816 O2 -0.6671904426 -1.1013158427 2.2635509843 O9 -0.6909098452 -0.5498117682 -2.0460772650 H14 0.0638823690 -0.6694127106 2.7361109979 H15 -1.4989368126 -1.1237311354 2.7650912929 H16 -0.0561709325 0.1412880660 -2.3084304602 H17 -1.5924174104 -0.3469216281 -2.3567102223 C18 2.3398816848 1.8924428083 0.3375312608 C19 2.3003314316 3.2896680716 0.4187708728 C20 1.0773503832 3.9661067881 0.4637561504 C21 -0.1278488538 3.2560991196 0.4275607196 C22 -0.0919328082 1.8678954588 0.3472582401 C23 1.1437141201 1.1831560071 0.3024621724 H24 3.2763445030 1.3537643474 0.3035402664 129 H25 3.2262473013 3.8476346739 0.4471280626 H26 1.0593191539 5.0452967100 0.5269896708 H27 -1.0848267656 3.7566102302 0.4622408556 benzoquinone-Fe(IV) O14 -1.8416465905 -1.2460000322 -3.2146303901 H2 -1.9810727362 -2.0560253637 -2.6887940822 Fe4 -0.9229699265 0.1729729273 -2.3219502282 O5 0.9613788407 -0.7842530493 -2.8997914785 H19 1.3258565201 -1.3966132912 -2.2383433909 H20 0.8188109617 -1.2110197606 -3.7640459624 O9 -0.3660295322 1.8366937941 -1.7924975025 H23 -0.1464179029 1.9442715948 -0.8483748523 O15 -0.9578065529 -0.5086826880 -0.8242292217 C12 -1.8557968838 4.2035373878 -4.8983578826 C13 -1.7758582767 4.2651361516 -6.3701837301 C14 -1.2583132412 3.0730409030 -7.0841417507 C15 -0.8867841582 1.9755942623 -6.4108838101 C16 -0.9994344702 1.9222100466 -4.9426879916 130 C17 -1.4949388524 3.1035955349 -4.2190984117 H18 -2.2161213662 5.0945969042 -4.4057486248 O19 -2.1256395932 5.2847373652 -7.0034353276 H21 -1.2054740936 3.1479248719 -8.1606300652 H22 -0.5109850485 1.0851947104 -6.8931909917 O22 -0.6661102150 0.8622802563 -4.3498032786 H24 -1.5057872557 3.0148387771 -3.1443610814 O24 -3.0820280317 0.9174271800 -2.8399918594 H25 -3.1585001576 -0.0195911765 -3.1749001183 H26 -3.7120461012 1.1219291170 -2.1297580792 1,4-naphthaquinone-Fe(IV) O14 -1.6517175365 -1.3371660716 -3.2755191901 H2 -1.8598839606 -2.1025035525 -2.7051419368 Fe4 -0.9259663879 0.1781495427 -2.3361606991 O5 1.0049186327 -0.7798943905 -2.7955378076 H19 1.5138317566 -1.1475090487 -2.0534811069 H20 0.7215970585 -1.4655070393 -3.4316000790 O9 -0.4091663901 1.8443729204 -1.7683932977 131 H23 -0.3065956686 1.9551863096 -0.8044918061 O15 -1.1063799845 -0.4704895011 -0.8330219104 C12 -1.7561574411 4.2342106581 -4.9116753170 C13 -1.1953394137 3.1132756264 -8.5161699940 C14 -0.7897853233 1.9721995463 -9.2059928681 C15 -0.4487153311 0.8111273285 -8.5070879218 C16 -0.5084407571 0.7900109833 -7.1162844406 C17 -1.4252628860 3.1231272014 -4.2358846965 H18 -2.0758657459 5.1361103532 -4.4094766631 H17 -1.4675930302 4.0180463851 -9.0394092276 H21 -0.7425710890 1.9852406748 -10.2855435608 H22 -0.1407756992 -0.0730414349 -9.0465060458 O22 -0.6564111708 0.8721688031 -4.3185007719 H24 -1.4233839563 3.0446791991 -3.1589525881 O24 -3.0747216454 0.7796704919 -2.9512099589 H25 -3.0937048431 -0.1353383671 -3.3393985239 H26 -3.7485853563 0.9116276608 -2.2640455682 C25 -1.7019481536 4.3121292033 -6.3835626149 132 C26 -1.2581773985 3.1012789525 -7.1229018384 C27 -0.9077245369 1.9331094879 -6.4184464229 C28 -0.9784763185 1.9165609592 -4.9448195833 O19 -2.0206600320 5.3661449714 -6.9765392283 H30 -0.2580952964 -0.0979142411 -6.5552741201 Fe1 -0.1603003512 -0.1022444073 -1.3974872379 C2 0.4466970058 -0.4559928965 1.4506417070 H3 1.4800091851 -0.1825581253 1.2558330994 H4 0.4025673588 -0.9295879976 2.4303778493 C5 -0.4283897782 0.7940037394 1.4346118295 H6 -1.4803835430 0.5305103584 1.4991049860 H7 -0.1974096335 1.4138130166 2.2942967777 N8 -0.2573856866 1.5671524965 0.1502770960 N9 0.0365044162 -1.4045586258 0.3725860405 C10 -1.4434040114 2.3827221038 -0.2387364167 H11 -2.2943007975 1.7050620670 -0.3094218074 H12 -1.2766535251 2.7523600504 -1.2449683142 edta-Fe(IV) 133 C13 -1.2681303540 -2.0761317153 0.5873053527 H14 -2.0580306549 -1.3342159357 0.6494466263 H15 -1.5071850389 -2.6389700215 -0.3143490334 C16 1.0851496829 -2.3673177686 -0.0764890173 H17 2.0530581426 -1.8831258971 0.0191076008 H18 1.0937263964 -3.2745326999 0.5202130397 C19 1.0204509900 2.3519064989 0.0619310401 H20 1.8367166106 1.6415963744 -0.0325268122 H21 0.9909936577 2.9126895659 -0.8644892977 C22 0.8791892170 -2.7168118181 -1.5613762928 O23 1.3863765871 -3.7367931511 -2.0451615212 O24 0.1505319306 -1.8323229770 -2.2415436738 C25 -1.4156316478 -3.0070005110 1.7749270916 O26 -2.4759777929 -3.5369092388 2.0884745070 O27 -0.2572762715 -3.2162165207 2.4890097403 C28 1.2905829470 3.2790720496 1.2223116671 O29 1.6739108447 2.9352514084 2.3387956134 O30 1.0906318743 4.5983214088 0.8892396302 134 C31 -1.8414745162 3.5633776268 0.6192772333 O32 -2.4135527693 4.5605897196 0.1957896035 O33 -1.5588226995 3.3916397658 1.9554541418 H34 -0.4085316466 -3.8442180320 3.2269303750 O35 -0.1573087997 1.2691854094 -3.0059876856 H36 -0.8532358650 1.2267850911 -3.6840370991 H37 0.7475586686 1.3649587759 -3.3560842194 O38 -1.8196253529 -0.1454599430 -1.4163787988 H41 1.3057011566 5.1921796455 1.6400648647 H40 -1.8655126093 4.1581211449 2.4840582014 O41 1.7105668011 0.2230088512 -1.4632438641 H42 2.2790909087 -0.4051840608 -1.9457193572 135