ALBERT

Notes: Hydrogen bond strength depends on both temperature and pressure. The gradient for hydrogen bond strength with temperature, or pressure, depends upon the hydrogen bonded structure. These features create an intimate connection between quantum mechanics and thermodynamics in the structure of liquid water. The equilibrium structural model of liquid water developed from analysis of the heat capacity at constant pressure is complex. The model is based on the assumptions that: (i) the hydrogen bond length and molecular packing density of water both vary with temperature; (ii) the number of different geometries for hydrogen bonding is limited to a small set; (iii) water molecules that possess these hydrogen bonding geometries are in equilibrium with each other under static conditions; (iv) significant changes in the slope of the heat capacity, Cp, and to a lesser extent other properties of the liquid, reflect the onset of significant changes in the chemical structure of the liquid; (v) the partial molal enthalpies and entropies of the different water arrays generated from these building blocks differ from each other in their dependence upon temperature; and (vi) the structure of the liquid is a random structural network of the structural components. The equilibrium structural model for liquid water uses four structural components and the assumptions listed above. At the extrapolated-homogeneous nucleation temperature, 221 K, a disordered hexagonal-diamond lattice (tetrahedrally hydrogen bonded water clusters) is the structure of liquid water. At the homogeneous nucleation temperature, ∼238 K, liquid water is a mixture of disordered tetrahedral water arrays and pentagonal water arrays. The abundance of tetrahedral water structures at this temperature causes the system to self-nucleate. As the temperature increases to 266 K the proportion of disordered pentagonal water clusters in the equilibrium mixture increases. At 256 K, the temperature of the previously unrecognized maximum in the heat of fusion of water, "planar"-hexagonal water arrays appear in the liquid. At 273 K the concentration of tetrahedral hydrogen bonded water approaches zero. At the temperature of maximum density, 277 K, the liquid consists of a disordered dodecahedral-water lattice. The equivalence point between pentagonal and "planar"-hexagonal water arrays occurs near 291 K, the approximate temperature of minimum solubility of large hydrocarbons in water. At temperatures above 307.6 K, the minimum in Cp, square water arrays first appear in significant concentrations. Pentagonal water arrays become insignificant in the liquid at the temperature of minimum isothermal compressibility, ∼319 K. The equilibrium point between "planar"-hexagonal and square water arrays occurs near 337 K. As the temperature increases the liquid structure becomes dominated by disordered cubic arrays of water molecules. Structures with fewer than four hydrogen bonds per water molecule appear in the liquid near 433 K. "Planar"-hexagonal clusters are no longer present in the liquid at the temperature of the maximum dissociation constant for water, 513 K. These views are certainly oversimplified. Simple models for density are introduced. A model for viscoscosity based on the variation of hydrogen bond strength with temperature is introduced. Attempts to model density, heat capacity, or other thermodynamic properties of liquid water, using only two functions will not capture the subtle complexity of the equilibrium process. The equilibrium structural model of water has the potential to provide a basis for quantitative descriptions of the liquid's seeming anomalies. © 1998 American Institute of Physics.

Notes: Perturbation molecular orbital (PMO) theory is used to approximate the electronic matrix element in the semiclassical expression for the rate of nonadiabatic electron transfer (ET). The resulting expression gives a satisfactory account of the intramolecular ET rate data reported by Closs, Miller, and co-workers. We develop the idea of electron-transfer efficiency for the contribution of electron transfer to the observed kinetics of ion–molecule collisions followed by electron transfer. Electron-transfer efficiency comes from the calculated ET rate divided by the maximum calculated ET rate. Electron-transfer efficiency values are also obtained by dividing the observed reaction rate by the collision rate, calculated by the PMO treatment of ion–molecule collision rates. We applied this approach to data on electron transfer from sulfurhexafluoride or perfluoromethylcyclohexane anions to aromatic acceptors. The structural reorganization energies, λs, for these reactions were 0.016 and 0.046 eV, respectively. The vibrational reorganization energies, λv, for the reactions were 1.01 and 1.00 eV, respectively. Electron transfer from either of the donor anions to fluoranil occurs in the inverted region. © 1997 American Institute of Physics.

Notes: Lengths and strengths of hydrogen bonds are exquisitely sensitive to temperature and pressure. Temperature and pressure sensitivity is the result of the fact that hydrogen bonds are so weak that the internal energy of the bond is important to bond strength, and the equilibrium bond distance is controlled by a combination of thermodynamics and quantum mechanics, rather than quantum mechanics alone. The importance of thermodynamics in the bond length, and strength, of hydrogen bonds is the result of a breakdown in the Born-Oppenheimer approximation that occurs when the energy of the first vibrational excitation of a bond is of the order of kT. Variation of water–water hydrogen bond length and strength with temperature and pressure is discussed in light of the data for the specific volume of ice Ih, the enthalpy of vaporization of liquid water, and the internal energy of the liquid. In most chemical contexts, correction of covalent bond strength for internal energy is not necessary. For hydrogen bonds this is not the case. In hydrogen bonded systems, like liquid water, the internal energy associated with hydrogen bonding is a significant fraction of the internal energy of the system. The variation of hydrogen bond length with temperature is approximately quadratic. Bond strength should also be quadratic with temperature because bond strength depends linearly on bond distance in second order. The internal energy correction is empirically quadratic in temperature. The net result is a linear dependence of apparent hydrogen bond strength on temperature. This can be seen directly in the variation of ΔHvaporization0/T for water with the reciprocal of temperature. The known variations in hydrogen bond equilibria with temperature in liquid formamide are discussed. Variation of the density of ice Ih with pressure, at constant temperature, demonstrates the nonlinear pressure dependence of hydrogen bond length. Because hydrogen bond strengths depend upon temperature and pressure, equilibria that involve hydrogen bonds explicitly depend upon temperature and pressure in addition to the universally appreciated dependence of the equilibrium constant on temperature. The temperature and pressure dependence of hydrogen bond length needs to be explicitly considered when one is modeling the properties of hydrogen bonded networks such as liquid water. Temperature dependence can be easily introduced by utilization of the hydrogen bond length, temperature relationship that is known for ice Ih and using a perturbation molecular orbital (PMO) treatment for bond formation. Our PMO treatment of hydrogen bonding involves second order perturbations between the donor and acceptor molecules. A random structural network model for liquid water based on this approach should be relatively easy to construct. The PMO model gives the relationship between hydrogen bond strength and hydrogen bond length as linear. This quantum mechanical result is quite distinct from the bond strength–bond length relationships obtained in classical models. © 1998 American Institute of Physics.

Notes: Methane chemical ionization (CI) mass spectra for a series of ten polycyclic chlorinated insecticides and metabolites have been examined. In all cases except heptachlor epoxide the base peak corresponded to elimination of Cl, or OH from the molecule ion. In the spectrum of heptachlor epoxide the [M + H]+ and [M - Cl]+ clusters were of approximately equal intensity. The CI spectra were remarkably simple, invariably less complex than the corresponding electron-impact (EI) mass spectra and the intensity of the ions with high information content, e.g. [M - CI]+ was uniformly high. All of these features are important to the analytical potential of these studies. Retro Diels-Alder (RDA) fragments were observed for the chlordanes, aldrin, isodrin, nonachlor and heptachlor epoxide. The reported preliminary data suggest that the relative intensity of RDA ions in CI mass spectra may be useful in establishing molecular configurations.

Notes: The alkylhalide-halide association ions, [RX2]- that are observed in the negative chemical ionization mass spectra of alkyl halides appear to be directly related to the corresponding SN2 transition states in solution. ‘Frontside’ association of halide ions with bridgehead alkyl halides does not occur in our system. The Change in heats and entropies of association for the chloromethane series is consistent with delocalization in the [RX]2- ions.

Notes: The heats and entropies of gas phase association of chloride, bromide and iodide with the respective methyl halides are reported. Comparison of these results with results published for SN2 reactions in solution suggests that the solvent is the dominant factor in determining relative halide nucleophilicities for the reactions in solution.

Notes: The gas phase heats and entropies of SN2-like association of bromide ions with a series of alkyl bromides are reported. Comparison of this data with published data for the same reactions in solution suggests that the alkyl group structural effects on SN2 reactivity in solution are controlled entirely by the solvent.

Notes: 2-Acetamido-1,3,4,6-tetra-O-acetyl-2-deoxy-α-D-glucopyranose (I), and its analogs specifically mono (trideuterioacetylated) at O-1 (III), at N-2 (II), at O-4 (IV) and at O-6 (V), have been examined by high-resolution mass spectrometry. From the elemental compositions of the fragment ions, the mass-number shifts resulting from deuterium incorporation and analysis of metastable transitions, it has been possible to specify in detail the fragmentation pathways undergone by this molecule. The principal degradations of I proceed by initial rapid decomposition of the molecular ion (whose intensity is insignificant) by three routes: (i) by loss of the C-1 acetoxyl group as a radical to give the glycosyl cation (a), (ii) by loss of the 1-acetyl group as a radical to give an acyclic ion m/e 346 (b) and (iii) by loss of a C-6 fragment and acetic acid derived from the 3-acetate group to give m/e 241 (c).

Notes: The mass spectra of both nonhindered and sterically hindered methyl phenanthrene derivatives follow the patterns established for other alkyl aromatics. The ‘unimolecular’ thermochemistry of these systems parallels the mass spectral fragmentation in a qualitatively predictable way. Generally the major thermochemical products correspond to one of the five most intense fragment ions, and the importance of products which correspond to ions with high appearance potentials increases with increasing temperature of the thermolysis system. On the basis of this empiricism we have outlined a tentative procedure for anticipating the results of thermolysis experiments based on mass spectral data. The use of appearance potential differences to estimate steric strain energies in the compounds studied did not give satisfactory results.

Notes: Negative ion mass spectra obtained under chemical ionization conditions (NCI) employing methane, isobutane or methylene chloride as the enhancement gas are presented for a series of chlorinated polycyclic insecticides. All of the compounds examined except 1-hydroxychlordene yielded molecular anions of substantial relative abundance (6 to 39%). The most significant features of the spectra are the prominent peaks at masses greater than that of the molecule ion formed via ionmolecule association reactions. Peaks representing association of the parent molecule with ionic species such as H-, O-, OH-, Cl-, H2OCl-, HCl2-, ClO- and Cl3- were observed in some cases. The base peak in all spectra was associated with the isotopic group of the [M + Cl]- on if contributions from other negative, even electron ions of low mass values present in high concentrations (Cl-, H2OCl- Cl2- and HCl2-) are neglected. Fragmentation processes were limited to elimination reactions involving loss of combinations of the even electron neutral species H, Cl and HCl. In addition, fragmentation resulting from a nucleophilic radical displacement of Cl by O- from the parent molecule was observed in all cases except 1-hydroxychlordene when the source was modestly wet (methane as reactant gas). NCI mass spectra of polycyclic chlorinated pesticides are reproducible, intense, interpretable in terms of classical carbanion chemistry and thus may have important analytical utility, particularly when used in conjunction with positive electron-impact and chemical ionization mass spectral methods and selective use of different enhancement gases.