No abstract

No abstract

CF 3 OH, an important and controversial by-product of atmospheric decomposition of CF 3 CFH 2 ͑HFC-134a͒ and other hydrofluorocarbons, has been examined by photoionization mass spectrometry. The ionization onset is characterized by a broad Franck-Condon distribution, arising primarily from a substantial elongation of the CO bond upon ionization. An upper limit to the adiabatic ionization potential ͑IP͒ of р 13.08 Ϯ 0.05 eV has been established. The appearance potentials ͑APs͒ of the first two fragments have been accurately determined by fitting with appropriate model functions as AP 0 ͑CF 2 OH ϩ /CF 3 OH͒р13.830Ϯ0.005 eV and AP 0 ͑CF 3 ϩ /CF 3 OH͒р13.996Ϯ0.005 eV. While the exact nature of the lowest-energy fragment ͑nominally CF 2 OH ϩ ͒ is not clear, the CF 3 ϩ fragment threshold leads unambiguously to ⌬H f 298 ‫ؠ‬ ͑CF 3 OH͒уϪ217.2Ϯ0.9 kcal/mol and D 298 ͑CF 3-OH͒р115.2Ϯ0.3 kcal/mol. With previously derived ⌬H f 298 ‫ؠ‬ ͑CF 3 O͒ϭϪ151.8 Ϫ1.1 ϩ1.7 kcal/mol, this yields D 298 ͑CF 3 O-H͒ ϭ117.5 Ϫ1.4 ϩ1.9 kcal/mol, very close to, or only slightly weaker than the O-H bond energy in water: D 298 ͑CF 3 O-H͒-D 298 ͑HO-H͒ϭϪ1.8 Ϫ1.4 ϩ1.9 kcal/molϷ0 kcal/mol. Similarly, with the recently redetermined value for ⌬H f ‫ؠ‬ ͑CF 2 O͒, this implies a 298 K reaction enthalpy for the 1,2-elimination of HF from CF 3 OH of 2.8 Ϫ1.1 ϩ1.7 kcal/mol. CF 3 OF and CF 3 OCl have also been examined by photoionization. CF 3 OF produces a very weak parent, with an apparent adiabatic IP͑CF 3 OF͒ р12.710Ϯ0.007 eV. An analysis of the CF 3 ϩ and CF 2 O ϩ fragments from CF 3 OF, when combined with literature data, suggests ⌬H f 298 ‫ؠ‬ ͑CF 3 OF͒ϭϪ176.9 Ϫ1.3 ϩ1.8 kcal/mol. The fitted value for the appearance potential of CF 3 ϩ from CF 3 OCl, AP 0 ͑CF 3 ϩ /CF 3 OCl͒р12.85Ϯ0.01 eV, leads to ⌬H f 298 ‫ؠ‬ ͑CF 3 OCl͒уϪ175.6Ϯ1.0 kcal/mol, D 298 ͑CF 3-OCl͒р88.4Ϯ0.3 kcal/mol, and D 298 ͑CF 3 O-Cl͒р52.8 Ϫ1.5 ϩ2.0 kcal/mol.

Ab initio quantum mechanical methods have been used to examine clusters formed of molecules of methane and water. The clusters contained one molecule of one component (methane or water) and several (10, 8, 6, 4, and 1) molecules of the other component. The Møller−Plesset perturbation theory (MP2 method) was used in the calculations. The cluster geometries were obtained via optimization and the interaction energies between the nearest neighbors were calculated for the geometries obtained in the first step. It is shown that the interaction energies and intermolecular distances between the molecules of methane and water are quite different in the clusters CH4···(H2O)10 and H2O···(CH4)10. They are also different from those in the water/methane dimer. The structure of the cluster CH4···(H2O)10 is highly affected by the hydrogen bonding among the water molecules, and the methane molecule is located inside a cage formed of water molecules. In contrast, the molecules of methane and water are randomly distributed in the cluster H2O···(CH4)10. The average methane/water intermolecular distance in the cluster CH4···(H2O)10 provided by the quantum mechanical calculations is in agreement with the experimental and simulation results regarding the position of the first maximum in the radial distribution function g oc = g oc(r oc) in dilute mixtures of methane in water, where r oc is the distance between the C atom of methane and the O atom of water. It is shown that the water molecules in the vicinity of a central methane molecule can be subdivided into two groups, A and B. Molecules of type A are touching nearest neighbors of the central methane molecule. They are located on a sphere with a radius corresponding to the first maximum in the radial distribution function g oc = g oc(r oc) and are tangentially oriented toward the central methane molecule. The layer of A water molecules is somewhat denser than bulk water. The molecules of type B are also located in the first hydration layer of a central methane molecule (up to a distance given by the position of the first minimum of the radial distribution function g oc = g oc(r oc)), but are not touching nearest neighbors. They are distributed more randomly than the molecules of type A, because they are less affected by the hydrophobic core of the solute.

We discuss issues in developing scalable parallel algorithms and focus on the distribution, as opposed to the replication, of key data structures. Replication of large data structures limits the maximum calculation size by imposing a low ratio of processors to memory. Only applications which distribute both data and computation across processors are truly scalable. The use of shared data structures that may be independently accessed by each process even in a distributed memory environment greatly simplifies development and provides a significant performance enhancement. We describe tools we have developed to support this programming paradigm. These tools are used to develop a highly efficient and scalable algorithm to perform self‐consistent field calculations on molecular systems. A simple and classical strip‐mining algorithm suffices to achieve an efficient and scalable Fock matrix construction in which all matrices are fully distributed. By strip mining over atoms, we also exploit all available sparsity and pave the way to adopting more sophisticated methods for summation of the Coulomb and exchange interactions. © 1996 by John Wiley & Sons, Inc.

Several parallel algorithms for Fock matrix construction are described. The algorithms calculate only the unique integrals, distribute the Fock and density matrices over the processors of a massively parallel computer, use blocking techniques to construct the distributed data structures, and use clustering techniques on each processor to maximize data reuse. Algorithms based on both square and row‐blocked distributions of the Fock and density matrices are described and evaluated. Variants of the algorithms are discussed that use either triple‐sort or canonical ordering of integrals, and dynamic or static task clustering schemes. The algorithms are shown to adapt to screening, with communication volume scaling down with computation costs. Modeling techniques are used to characterize algorithm performance. Given the characteristics of existing massively parallel computers, all the algorithms are shown to be highly efficient for problems of moderate size. The algorithms using the row‐blocked data distribution are the most efficient. © 1996 by John Wiley & Sons, Inc.

Nonoverlapping UPSIT items were identified that were individually associated with age and disease. Despite a modest predictive value of the AD-specific items for conversion to AD, the AD-specific items may be useful in enriching samples to better identify those at risk for AD. Further studies are needed with monomolecular and unilateral stimulation and orthogonal biomarker validation to further refine disease- and age-associated signals.

Nicotine dependence (ND) has a reported heritability of 40–70%. Low-coverage whole-genome sequencing was conducted in 1,889 samples from the UCSF Family study. Linear mixed models were used to conduct genome-wide association tests of ND in this and five cohorts obtained from the database of Genotypes and Phenotypes. Fixed-effect meta-analysis was carried out separately for European (n = 14,713) and African (n = 3,369) participants, and then in a combined analysis of both ancestral groups. The meta-analysis of African participants identified a significant and novel susceptibility signal (rs56247223; P = 4.11 × 10−8). Data from the Genotype-Tissue Expression (GTEx) study suggested the protective allele is associated with reduced mRNA expression of CACNA2D3 in three human brain tissues (P < 4.94 × 10−2). Sequence data from the UCSF Family study suggested that a rare nonsynonymous variant in this gene conferred increased risk for ND (P = 0.01) providing further support for CACNA2D3 involvement in ND. Suggestive associations were observed in six additional regions in both European and merged populations (P < 5.00 × 10−6). The top variants were found to regulate mRNA expression levels of genes in human brains using GTEx data (P < 0.05): HAX1 and CHRNB2 (rs1760803), ADAMTSL1 (rs17198023), PEX2 (rs12680810), GLIS3 (rs12348139), non-coding RNA for LINC00476 (rs10759883), and GABBR1 (rs56020557 and rs62392942). A gene-based association test further supported the relation between GABBR1 and ND (P = 6.36 × 10−7). These findings will inform the biological mechanisms and development of therapeutic targets for ND.