The AQUA-MER databases and aqueous speciation server: A web resource for multiscale modeling of mercury speciation.

To assess the chemical reactivity, toxicity, and mobility of pollutants in the environment, knowledge of their species distributions is critical. Because their direct measurement is often infeasible, speciation modeling is widely adopted. Mercury (Hg) is a representative pollutant for which study of its speciation benefits from modeling. However, Hg speciation modeling is often hindered by a lack of reliable thermodynamic constants. Although computational chemistry (e.g., density functional theory [DFT]) can generate these constants, methods for directly coupling DFT and speciation modeling are not available. Here, we combine computational chemistry and continuum-scale modeling with curated online databases to ameliorate the problem of unreliable inputs to Hg speciation modeling. Our AQUA-MER databases and web server (https://aquamer.ornl.gov) provides direct speciation results by combining web-based interfaces to a speciation calculator, databases of thermodynamic constants, and a computational chemistry toolkit to estimate missing constants. Although Hg is presented as a concrete use case, AQUA-MER can also be readily applied to other elements. © 2019 Wiley Periodicals, Inc.

Molecular Dynamics Simulation of the Structures, Dynamics, and Aggregation of Dissolved Organic Matter.

Dissolved organic matter (DOM) plays a significant role in the transport and transformation of pollutants in the aquatic environment. However, the experimental characterization of DOM has been limited mainly to bulk properties, and the molecular-level interactions among various components of DOM remain to be fully characterized. Here, we use molecular dynamics (MD) simulations to probe the structural properties of model DOM systems at atomic detail. The 200 ns simulations, validated by available experimental data, reveal processes and mechanisms by which chemical species (cations, peptides, lipids, lignin, carbohydrates, and some low-molecular-weight aliphatic and aromatic compounds) aggregate to form complex DOM. The DOM aggregates are dynamic, consisting of a hydrophobic core and amphiphilic exterior. The lipid tails and other hydrophobic fragments form the core, with hydrophilic and amphiphilic groups exposed to water, making DOM accessible to both polar and nonpolar species. Thus, the lipid component acts as a nucleator, whereas cations (especially Ca2+) connect the molecular fragments on the surface by coordinating with the O-containing functional groups of DOM. The structural details revealed here provide new insights including surface accessible atoms, overall assemblage, and interactions among the molecules of DOM for understanding the kinetics and mechanisms through which DOM interacts with metal and other contaminants.

Hydrodynamic Radii of Intrinsically Disordered Proteins Determined from Experimental Polyproline II Propensities.

The properties of disordered proteins are thought to depend on intrinsic conformational propensities for polyproline II (PPII) structure. While intrinsic PPII propensities have been measured for the common biological amino acids in short peptides, the ability of these experimentally determined propensities to quantitatively reproduce structural behavior in intrinsically disordered proteins (IDPs) has not been established. Presented here are results from molecular simulations of disordered proteins showing that the hydrodynamic radius (Rh) can be predicted from experimental PPII propensities with good agreement, even when charge-based considerations are omitted. The simulations demonstrate that Rh and chain propensity for PPII structure are linked via a simple power-law scaling relationship, which was tested using the experimental Rh of 22 IDPs covering a wide range of peptide lengths, net charge, and sequence composition. Charge effects on Rh were found to be generally weak when compared to PPII effects on Rh. Results from this study indicate that the hydrodynamic dimensions of IDPs are evidence of considerable sequence-dependent backbone propensities for PPII structure that qualitatively, if not quantitatively, match conformational propensities measured in peptides.

Activation-strain analysis reveals unexpected origin of fast reactivity in heteroaromatic azadiene inverse-electron-demand diels-alder cycloadditions.

Heteroaromatic azadienes, especially 1,2,4,5-tetrazines, are extremely reactive partners with alkenes in inverse-electron-demand Diels-Alder reactions. Azadiene cycloaddition reactions are used to construct heterocycles in synthesis and are popular as bioorthogonal reactions. The origin of fast azadiene cycloaddition reactivity is classically attributed to the inverse frontier molecular orbital (FMO) interaction between the azadiene LUMO and alkene HOMO. Here, we use a combination of ab initio, density functional theory, and activation-strain model calculations to analyze physical interactions in heteroaromatic azadiene-alkene cycloaddition transition states. We find that FMO interactions do not control reactivity because, while the inverse FMO interaction becomes more stabilizing, there is a decrease in the forward FMO interaction that is offsetting. Rather, fast cycloadditions are due to a decrease in closed-shell Pauli repulsion between cycloaddition partners. The kinetic-thermodynamic relationship found for these inverse-electron-demand cycloadditions is also due to the trend in closed-shell repulsion in the cycloadducts. Cycloaddition regioselectivity, however, is the result of differences in occupied-unoccupied orbital interactions due to orbital overlap. These results provide a new predictive model and correct physical basis for heteroaromatic azadiene reactivity and regioselectivity with alkene dieneophiles.

Intermolecular N-H oxidative addition of ammonia, alkylamines, and arylamines to a planar σ(3)-phosphorus compound via an entropy-controlled electrophilic mechanism.

Ammonia, alkyl amines, and aryl amines are found to undergo rapid intermolecular N-H oxidative addition to a planar mononuclear σ(3)-phosphorus compound (1). The pentacoordinate phosphorane products (1·[H][NHR]) are structurally robust, permitting full characterization by multinuclear NMR spectroscopy and single-crystal X-ray diffraction. Isothermal titration calorimetry was employed to quantify the enthalpy of the N-H oxidative addition of n-propylamine to 1 ((n)PrNH2 + 1 → 1·[H][NH(n)Pr], ΔHrxn(298) = -10.6 kcal/mol). The kinetics of n-propylamine N-H oxidative addition were monitored by in situ UV absorption spectroscopy and determination of the rate law showed an unusually large molecularity (ν = k[1][(n)PrNH2](3)). Kinetic experiments conducted over the temperature range of 10-70 °C revealed that the reaction rate decreased with increasing temperature. Activation parameters extracted from an Eyring analysis (ΔH(⧧) = -0.8 ± 0.4 kcal/mol, ΔS(⧧) = -72 ± 2 cal/(mol·K)) indicate that the cleavage of strong N-H bonds by 1 is entropy controlled due to a highly ordered, high molecularity transition state. Density functional calculations indicate that a concerted oxidative addition via a classical three-center transition structure is energetically inaccessible. Rather, a stepwise heterolytic pathway is preferred, proceeding by initial amine-assisted N-H heterolysis upon complexation to the electrophilic phosphorus center followed by rate-controlling N → P proton transfer.

1,2-Addition of dihydrogen across rhodium(III)-OMe bonds.

The Rh(III) complexes [((t)bpy)2Rh(OMe)(L)][X]n ((t)bpy = 4,4'-di-tert-butyl-2,2'-bipyridyl; L = MeOH, n = 2, X = OTf (OTf = trifluoromethanesulfonate), TFA (TFA = trifluoroacetate); L = TFA, n = 1, X = OTf) have been shown to activate dihydrogen via net 1,2-addition of the H-H bond across the Rh(III)-OMe bond. The bis(methoxide) complex [((t)bpy)2Rh(OMe)2][OTf] was synthesized by addition of CsOH·H2O in methanol to [((t)bpy)2Rh(OTf)2][OTf] in CH3CN. The addition of HTFA to [((t)bpy)2Rh(OMe)2][OTf] leads to the formation of [((t)bpy)2Rh(OMe)(MeOH)][OTf][TFA], which exists in equilibrium with [((t)bpy)2Rh(OMe)(TFA)][OTf]. The mixture of [((t)bpy)2Rh(OMe)(MeOH)][OTf][TFA] and [((t)bpy)2Rh(OMe)(TFA)][OTf] activates dihydrogen at 68 °C to give methanol and [((t)bpy)2Rh(H)(TFA)][OTf]. Studies indicate that the activation of dihydrogen has a first-order dependence on the Rh(III) methoxide complex and a dependence on hydrogen that is between zero and first order. Combined experimental and computational studies have led to a proposed mechanism for hydrogen activation by [((t)bpy)2Rh(OMe)(MeOH)][OTf][TFA] that involves dissociation of MeOH, coordination of hydrogen, and 1,2-addition of hydrogen across the Rh-OMe bond. DFT calculations indicate that there is a substantial energy penalty for MeOH dissociation and a relatively flat energy surface for subsequent hydrogen coordination and activation.

Selective CH functionalization of methane, ethane, and propane by a perfluoroarene iodine(III) complex.

Direct partial oxidation of methane, ethane, and propane to their respective trifluoroacetate esters is achieved by a homogeneous hypervalent iodine(III) complex in non-superacidic (trifluoroacetic acid) solvent. The reaction is highly selective for ester formation (>99%). In the case of ethane, greater than 0.5 M EtTFA can be achieved. Preliminary kinetic analysis and density functional calculations support a nonradical electrophilic CH activation and iodine alkyl functionalization mechanism.

Theory of divalent main group H2 activation: electronics and quasiclassical trajectories.

Density functional theory (DFT), absolutely localized molecular orbital (ALMO) analysis, and quasiclassical trajectories (QCTs) were used to study the structure, barrier heights, thermodynamics, electronic properties, and dynamics of dihydrogen (H2) activation by singlet divalent main group compounds (ER2; E = C, Si, Ge). ALMO energy and charge decomposition calculations reveal that in the transition state CR2 acts as an ambiphile toward H2 because of equal forward-bonding and back-bonding orbital stabilization while SiR2 and GeR2 act as nucleophiles with dominant orbital energy stabilization arising from ER2 to H2 donation. Frontier molecular orbital (FMO) energy gaps do not provide a reasonable estimate of energy stabilization gained between the ER2 and H2 in the transition state or an accurate description of the nucleophilic versus electrophilic character because of electron repulsion and orbital overlap influences that are neglected. In CR2 transition states, forward-bonding and back-bonding are maximized in the nonleast motion geometry. In contrast, SiR2/GeR2 transition states have side-on geometries to avoid electron-electron repulsion. Electron repulsion, rather than orbital interactions, also determines the relative barrier heights of CR2 versus SiR2/GeR2 reactions. Examination of barrier heights and reaction energies shows a clear kinetic-thermodynamic relationship for ER2 activation of H2. A computational survey of R groups on ER2 divalent atom centers was performed to explore the possibility for H2 activation to occur with a low barrier and thermodynamically reversible. QCTs show that dihydrogen approach and reaction with CR2 may involve geometries significantly different than the static transition-state structure. In contrast, trajectories for dihydrogen addition to SiR2 involve geometries close to the side-on approach suggested by the static transition-state structure. QCTs also demonstrate that addition of H2 to CR2 and SiR2 is dynamically concerted with the average time gap of bond formation between E-H bonds of approximately 11 and 21 fs, respectively.

Metal-mediated dihydrogen activation. What determines the transition-state geometry?

Density functional theory and absolutely localized molecular orbital energy decomposition analysis calculations were used to calculate and analyze dihydrogen activation transition states and reaction pathways. Analysis of a variety of transition-metal complexes with d(0), d(6), d(8), and d(10) orbital occupation with a diverse range of metal ligands reveals that for transition states, akin to dihydrogen σ complexes, there is a continuum of activated H-H bond lengths that can be classified as "dihydrogen" (0.8-1.0 Å), "stretched or elongated" (1.0-1.2 Å), and "compressed dihydride" (1.2-1.6 Å). These calculations also quantitatively for the first time reveal that the extent to which H(2) is activated in the transition-structure geometry depends on back-bonding orbital interactions and not forward-bonding orbital interactions. This is true regardless of the mechanism or whether the metal ligand complex acts as an electrophile, ambiphile, or nucleophile toward dihydrogen.

Theory of late-transition-metal alkyl and heteroatom bonding: analysis of Pt, Ru, Ir, and Rh complexes.

Density functional and correlated ab initio methods were used to calculate, compare, and analyze bonding interactions in late-transition-metal alkyl and heteroatom complexes (M-X). The complexes studied include: (DMPE)Pt(CH(3))(X) (DMPE = 1,2-bis(dimethylphosphino)ethane), Cp*Ru(PMe(3))(2)(X) (Cp* = pentamethylcyclopentadienyl), (DMPE)(2)Ru(H)(X), (Tp)(CO)Ru(Py)(X) (Tp = trispyrazolylborate), (PMe(3))(2)Rh(C(2)H(4))(X), and cis-(acac)(2)Ir(Py)(X) (acac = acetylacetonate). Seventeen X ligands were analyzed that include alkyl (CR(3)), amido (NR(2)), alkoxo (OR), and fluoride. Energy decomposition analysis of these M-X bonds revealed that orbital charge transfer stabilization provides a straightforward model for trends in bonding along the alkyl to heteroatom ligand series (X = CH(3), NH(2), OH, F). Pauli repulsion (exchange repulsion), which includes contributions from closed-shell d(π)-p(π) repulsion, generally decreases along the alkyl to heteroatom ligand series but depends on the exact M-X complexes. It was also revealed that stabilizing electrostatic interactions generally decrease along this ligand series. Correlation between M-X and H-X bond dissociation energies is good with R(2) values between 0.7 and 0.9. This correlation exists because for both M-X and H-X bonds the orbital stabilization energies are a function of the orbital electronegativity of the X group. The greater than 1 slope when correlating M-X and H-X bond dissociation energies was traced back to differences in Pauli repulsion and electrostatic stabilization.

Variational fractional-spin density-functional theory for diradicals.

Accurate computation of singlet-triplet energy gaps of diradicals remains a challenging problem in density-functional theory (DFT). In this work, we propose a variational extension of our previous work [D. H. Ess, E. R. Johnson, X. Q. Hu, and W. T. Yang, J. Phys. Chem. A 115, 76 (2011)], which applied fractional-spin density-functional theory (FS-DFT) to diradicals. The original FS-DFT approach assumed equal spin-orbital occupancies of 0.5 α-spin and 0.5 ß-spin for the two degenerate, or nearly degenerate, frontier orbitals. In contrast, the variational approach (VFS-DFT) optimizes the total energy of a singlet diradical with respect to the frontier-orbital occupation numbers, based on a full configuration-interaction picture. It is found that the optimal occupation numbers are exactly 0.5 α-spin and 0.5 ß-spin for diradicals such as O(2), where the frontier orbitals belong to the same multidimensional irreducible representation, and VFS-DFT reduces to FS-DFT for these cases. However, for diradicals where the frontier orbitals do not belong to the same irreducible representation, the optimal occupation numbers can vary between 0 and 1. Furthermore, analysis of CH(2) by VFS-DFT and FS-DFT captures the (1)A(1) and (1)B(1) states, respectively. Finally, because of the static correlation error in commonly used density functional approximations, both VFS-DFT and FS-DFT calculations significantly overestimate the singlet-triplet energy gaps for disjoint diradicals, such as cyclobutadiene, in which the frontier orbitals are confined to separate atomic centers.

Chiral salicyloxazolines as auxiliaries for the asymmetric synthesis of ruthenium polypyridyl complexes.

Chiral auxiliaries are promising emerging tools for the asymmetric synthesis of octahedral metal complexes. We recently introduced chiral salicyloxazolines as coordinating bidentate chiral ligands which provide excellent control over the metal-centered configuration in the course of ligand substitution reactions and can be removed afterward in an acid-induced fashion under complete retention of configuration (J. Am. Chem. Soc. 2009, 131, 9602-9603). Here reported is our detailed investigation of this sequence of reactions, affording virtually enantiopure ruthenium polypyridyl complexes. The control of the metal-centered chirality by the coordinated chiral salicyloxazolinate ligand was evaluated as a function of reaction conditions, the employed bidentate 2,2'-bipyridine and 1,10-phenanthroline ligands, and the substituent at the asymmetric 5-position of the oxazoline heterocycle. Most striking was the strong influence of the reaction solvent, with aprotic solvents of lower polarity providing the most favorable diastereoselectivities. Through a combination of computational and experimental results, it was revealed that the observed stereoselectivities are under thermodynamic control. The removal of the chiral salicyloxazoline auxiliary under retention of the configuration requires acidic conditions and a coordinating solvent such as MeCN or THF in order to prevent partial racemization. This method represents the first general strategy for the asymmetric synthesis of enantiopure heteroleptic ruthenium polypyridyl complexes.

Antimony(v) cations for the selective catalytic transformation of aldehydes into symmetric ethers, α,ß-unsaturated aldehydes, and 1,3,5-trioxanes.

1-Diphenylphosphinonaphthyl-8-triphenylstibonium triflate ([][OTf]) was prepared in excellent yield by treating 1-lithio-8-diphenylphosphinonaphthalene with dibromotriphenylstiborane followed by halide abstraction with AgOTf. This antimony(v) cation was found to be stable toward oxygen and water, and exhibited exceptional Lewis acidity. The Lewis acidity of [][OTf] was exploited in the catalytic reductive coupling of a variety of aldehydes into symmetric ethers of type in good to excellent yields under mild conditions using Et3SiH as the reductant. Additionally, [][OTf] was found to selectively catalyze the Aldol condensation reaction to afford α-ß unsaturated aldehydes () when aldehydes with 2 α-hydrogen atoms were used. Finally, [][OTf] catalyzed the cyclotrimerization of aliphatic and aromatic aldehydes to afford the industrially-useful 1,3,5 trioxanes () in good yields, and with great selectivity. This phosphine-stibonium motif represents one of the first catalytic systems of its kind that is able to catalyze these reactions with aldehydes in a controlled, efficient manner. The mechanism of these processes has been explored both experimentally and theoretically. In all cases the Lewis acidic nature of the antimony(v) cation was found to promote these reactions.

Main-group compounds selectively oxidize mixtures of methane, ethane, and propane to alcohol esters.

Much of the recent research on homogeneous alkane oxidation has focused on the use of transition metal catalysts. Here, we report that the electrophilic main-group cations thallium(III) and lead(IV) stoichiometrically oxidize methane, ethane, and propane, separately or as a one-pot mixture, to corresponding alcohol esters in trifluoroacetic acid solvent. Esters of methanol, ethanol, ethylene glycol, isopropanol, and propylene glycol are obtained with greater than 95% selectivity in concentrations up to 1.48 molar within 3 hours at 180°C. Experiment and theory support a mechanism involving electrophilic carbon-hydrogen bond activation to generate metal alkyl intermediates. We posit that the comparatively high reactivity of these d(10) main-group cations relative to transition metals stems from facile alkane coordination at vacant sites, enabled by the overall lability of the ligand sphere and the absence of ligand field stabilization energies in systems with filled d-orbitals.

Direct stereospecific synthesis of unprotected N-H and N-Me aziridines from olefins.

Despite the prevalence of the N-H aziridine motif in bioactive natural products and the clear advantages of this unprotected parent structure over N-protected derivatives as a synthetic building block, no practical methods have emerged for direct synthesis of this compound class from unfunctionalized olefins. Here, we present a mild, versatile method for the direct stereospecific conversion of structurally diverse mono-, di-, tri-, and tetrasubstituted olefins to N-H aziridines using O-(2,4-dinitrophenyl)hydroxylamine (DPH) via homogeneous rhodium catalysis with no external oxidants. This method is operationally simple (i.e., one-pot), scalable, and fast at ambient temperature, furnishing N-H aziridines in good-to-excellent yields. Likewise, N-alkyl aziridines are prepared from N-alkylated DPH derivatives. Quantum-mechanical calculations suggest a plausible Rh-nitrene pathway.

Are they linear, bent, or cyclic? Quantum chemical investigation of the heavier group 14 and group 15 homologues of HCN and HNC.

The singlet potential-energy surface (PES) of the system involving the atoms H, X, and E (the (H, X, E) system) in which X=N-Bi and E=C-Pb has been explored at the CCSD(T)/TZVPP and BP86/TZ2P+ levels of theory. The nature of the X-E bonding has been analyzed with charge- and energy-partitioning methods. The calculations show that the linear isomers of the nitrogen systems lin-HEN and lin-HNE are minima on the singlet PES. The carbon compound lin-HCN (HCN=hydrogen cyanide) is 14.9 kcal mol(-1) lower in energy than lin-HNC but the heavier group 14 homologues lin-HEN (E=Si-Pb) are between 64.8 and 71.5 kcal mol(-1) less stable than the lin-HNE isomers. The phosphorous system (H, P, E) exhibits significant differences concerning the geometry and stability of the equilibrium structures compared with the nitrogen system. The linear form lin-HEP of the former system is much more stable than lin-HPE. The molecule lin-HCP is the only minimum on the singlet PES. It is 78.5 kcal mol(-1) lower in energy than lin-HPC, which is a second-order saddle point. The heavier homologues lin-HPE, in which E=Si-Pb, are also second-order saddle points, whereas the bent-HPE structures are the global minima on the PES. They are between 10.3 (E=Si) and 36.5 kcal mol(-1) (E=Pb) lower in energy than lin-HEP. The bent-HPE structures possess rather acute bending angles H-P-E between 60.1 (E=Si) and 79.7° (E=Pb). The energy differences between the heavier group 15 isomers lin-HEX (X=P-Bi) and the bent structures bent-HXE become continuously smaller. The silicon species lin-HSiBi is even 3.1 kcal mol(-1) lower in energy than bent-HBiSi. The bending angle H-X-E becomes more acute when X becomes heavier. The drastic energy differences between the isomers of the system (H, X, E) are explained with three factors that determine the relative stabilities of the energy minima: 1) The different bond strength between the hydrogen bonds H-X and H-E. 2) The electronic excitation energy of the fragment HE from the X (2)Π ground state to the (4)Σ(-) excited state, which is required to establish a E≡X triple bond in the molecules lin-HEX. 3) The strength of the intrinsic X-E interactions in the molecules. The trends of the geometries and relative energies of the linear, bent, and cyclic isomers are explained with an energy-decomposition analysis that provides deep insight into the nature of the bonding situation.