Successful PhD Defense of Daniela Trogolo

In this thesis, quantum chemical methods have been applied to elucidate the thermodynamics and the kinetics of reactions involving reactive species in water. Due to their high reactivity in water, many transient species are difficult to study by experimental means only. Here, quantum chemical models are used to provide a deeper insight into the chemical nature and aqueous behaviors of such species.

In the first chapter, I investigate the gas phase electronic structure and the thermodynamics of inorganic chloramines, bromamines, and bromochloramines, collectively termed halamines. The halamines are halogen oxidants that arise from reactions between ammonia and hypohalous acids during water disinfection processes, and these reactive species are implicated in the formation of disinfection byproducts that are harmful to human health. Despite their relevance in both drinking water chemistry and in biochemistry, the stabilities and speciation of these molecules are difficult to investigate by experimental means. To accurately predict the electronic structures and gas phase thermodynamic properties of halamines, I design a computational protocol, TA14, based on the high-quality Weizmann and Feller-Peterson-Dixon composite methods. TA14 combines a systematic sequence of wave function theory calculations, including the evaluations of dynamical and static electron correlation (CCSDTQ), core/valence electron correlation contributions, scalar and spin-orbit relativistic contributions, and VPT2 anharmonic vibrations. Using TA14, I successfully assess the gas phase total atomization energies, free enthalpies of formation, Δ_fH^o_gas, and Gibbs free energies of formation, Δ_fG^o_gas, of halamines within uncertainty bounds of 1-3 kJ mol^-1.

Analysis of the energy components contributing to total atomization energies of halamines reveals that N-Cl and N-Br bonds are held together mostly or entirely by electron correlation forces, with small or even negative Hartree Fock contributions. For example, the Hartree Fock component of the total atomization energy, TAE_HF, is negative for both NBr₃ and NBr₂Cl, implying that these molecules would be predicted as unstable without accounting for dynamical electron correlation. Reported thermochemical data enable the determination of equilibrium constants for reactions involving halamines, opening possibilities for more quantitative studies of the chemistry of these poorly understood compounds.

In the second chapter, I evaluate the aqueous equilibria and speciation of halamines.

I combine theoretical benchmark-quality gas phase Gibbs free energies of formation (chapter 2) with the computed Gibbs free energies of solvation, ΔG*_solv, thereby obtaining aqueous phase Gibbs free energies of formation, Δ_fG^o_aq, for halamines. The 'half-and-half' solvation approach, based on averaging the estimates of SMD implicit solvent model and the cluster-continuum solvent model, produces an average error of 3.3 kJ mol^-1 in ΔG*_solv for a set of structurally related molecules containing H, N, O, and Cl. Taking into consideration the combined uncertainties of the computed Δ_fG^o_gas values and the computed ΔG*_solv values, we assign an uncertainty of 6-7 kJ mol^-1 to the theoretical standard Gibbs free energies of formation in aqueous phase, Δ_fG^o_aq, of halamines.

Δ_fG*_aq values are key thermodynamic properties for investigating chemical processes involving halamines during drinking water treatment. The newly reported thermodynamic data can be used to determine the stabilities (reaction equilibria) of halamines in water. Based on our estimated uncertainties of 6-7 kJ mol^-1 in the computed Δ_fG*_aq values of halamines, we expect roughly 1 order-of-magnitude uncertainty in the aqueous equilibrium constants for the reactions leading to the production of halamines in water. I also estimate acid dissociation constant values, pK_a, of both the protonated and neutral halamines and methylated amines, based on quantum chemical computations. These results bring new insights into the electronic nature and stabilities of these reactive species in water, enabled by quantitative computational investigations of properties that are not possible to determine experimentally.

In the third chapter, I evaluate a previously proposed molecular pathway that leads from N,N-dimethylsulfamide to N,N-nitrosodimethylamine (NDMA), a US Environmental Protection Agency regulated carcinogen, during ozonation of bromide-containing natural water.

The molecular modeling involves several challenges, including electronic structures having multireference character, intermediates and transition state structures that are aqueously solvated, and intermolecular migration reactions. To describe this transformation, I take advantage the M05 and B2PLYPD density functional theory methods that were previously developed and tested for investigating multireference electronic structures and reaction barrier heights. The multi-step molecular mechanism involves the reactions of N,N-dimethylsulfamide with hypobromous acid, produced by the ozonation of naturally-occuring bromide, and with ozone. This leads to an anionic intermediate that resembles a fragile complex between an electrophilic nitrosyl bromide (BrNO) molecule and an electron-rich dimethylaminosulfinate (SO₂)N(CH₃)₂^- fragment. This loosely bound intermediate decomposes by two branches: an exothermic channel that produces NDMA, and an entropy-driven channel giving non-NDMA products.

In the last chapter, I computationally investigate the molecular pathways that lead from biogenic sulfur-containing compounds (thiols, polysulfides, and thioethers) to carbonyl sulfide (OCS) in the ocean. OCS ventilates from the oceans to the atmosphere, where it is directly implicated in the cloud condensation nuclei and in the formation of the stratospheric sulfate layer. However, the predominant molecular sources of OCS in the oceans are unknown, and previously reported experimental results suggested that thiyl/sulfhydryl radicals could play a key role in both the photochemical and non-photochemical production pathways of OCS.

Quantum chemical modeling and reaction kinetics simulations are applied to study the complex molecular reactions leading to OCS in water.

Hypothesized reaction mechanisms involving thiyl/sulfhydryl radicals and CO are assessed with density functional theory electronic structure methods. These results show that thiyl radicals, which can be produced from the oxidation of thiolates or possible from the bond cleavage of dimethylsulfide, are key precursors of OCS in seawater. Hydrogen sulfide and polysulfide species such as HS^., S^.-, S^2-, and S₂^2- are likely involved in the formation of OCS in ocean surface as well. APEX photochemical models reveal that the transient species, Br₂^.-, plays a main role in the phototransformation of dimethylsulfide, methionine, cysteine, and glutathione, whereas CO₃^.- could play a significant role in the photodegradation of cysteine and glutathione at low concentrations of DOC. The results of the study have implications for our understanding of the role of OCS in ocean biogeochemistry and climate.