透過您的圖書館登入
IP:3.129.13.201
  • 學位論文

二氯甲烷之量子化學勢能計算與分子動力學模擬

Quantum Chemistry Calculated Intermolecular Interaction and Molecular Dynamics Simulation of Dichloromethane

指導教授 : 趙聖德

摘要


量子計算部分我們採用自洽理論(Hartree-Fock,HF)、微擾理論(Møller-Plesset perturbation theory,MP2)、密度泛函理論(Density Functional Theory,DFT)、以及耦合簇理論(Coupled Cluster method,CC)等四種理論來計算二氯甲烷分子之間的作用力並且皆搭配BSSE(Basis Set Superposition Errors)修正,其中HF、DFT及MP2等方法搭配Dunning’s correlation consistent及Pople’s medium size基底來計算單體及雙體的最佳化構型,並且使用MP2計算結果與CCSD(T)搭配aug-cc-pVXZ(X=D、T、Q)基底計算出的結果來做比較,接著使用四種DFT方法搭配11種基底來與CCSD(T)的結果來做比較,以找出節省計算資源的方法。再來我們針對其12種構型配合PSI4軟體中的SAPT(Symmetry-Adapted Perturbation Theory)分析分解出不同構型的能量組合,其中包含靜電能、誘導能、交換能及色散能,以便我們分析其吸引力及排斥力對二氯甲烷分子二聚體的影響。 分子動力學模擬部分我們使用MP2搭配aug-cc-pVQZ進行單體最佳化,並且從其可能出現的38種構型裡選擇12種較具代表性的構型來進行擬合,擬合部份我們選擇5site Lennard-Jones potential model並加上庫倫項來擬合量子化學計算得到的二聚體能量曲線,接著代入牛頓方程式來進行分子動力學模擬,得到二氯甲烷的平衡性質和動態性質。二氯甲烷的分子動力學模擬我們模擬了徑向分佈函數(Radial Distribution Function,RDF)、速度自相關係數(Velocity Autocorrelation Function,VAF)、擴散係數(Diffusion Constant)以及黏滯係數(Viscosity),並與現有的文獻做比較,皆有不錯的準確度,這表示以量子化學計算所建構出的力場來進行分子動力學模擬有一定的可靠度。

並列摘要


In the quantum chemical calculation part, we use Hartree-Fock self-consistent theory(HF), second-order Møller-Plesset perturbation theory(MP2), Density Functional Theory (DFT), and Coupled Cluster method (CC). We use these four theories to calculate the intermolecular interaction of methylene chloride and the correction of the basis set superposition errors(BSSE) has been included. In the optimized structure of dimer and binding energy calculations of methylene chloride, we use HF, DFT and MP2 method with Dunning’s correlation consistent basis sets(cc-pVDZ up to aug-cc-pVQZ) and Pople’s medium size basis sets(6-31(G) up to 6-311++G(3df,3pd)), and the results are compare with the calculated by single-point coupled cluster with single and double and perturbative triple excitation(CCSD(T)). Furthermore, we analyzed the energy composition of the 12 configurations with the Symmetry-Adapted Perturbation Theory(SAPT) in the PSI4 software, including electrostatic energy, induction energy, exchange energy and dispersion energy. In the molecular dynamics simulation part, we used MP2 with aug-cc-pVQZ for monomer optimization, and selected 12 representative configurations from 38 possible configurations to fit. In the fitting part, we chose the 5site Lennard-Jones potential model including the Coulomb term to fit the ab initio data, and then substituted into the Newton equation for molecular dynamics simulation to obtain the equilibrium and dynamic properties of dichloromethane. We compared the radial distribution function (RDF), velocity autocorrelation function (VAF), diffusion constant, and viscosity with the scientific literature, the simulation results are consistent with experiment data. This indicates that molecular dynamics simulations using force fields constructed by quantum chemical calculations can accurately reproduce thermodynamic properties.

參考文獻


[1] Li, A. H. T., Huang, S. C., & Chao, S. D. (2010). Molecular dynamics simulation of liquid carbon tetrachloride using ab initio force field. The Journal of chemical physics, 132(2), 024506.
[2] Yin, C. C., Li, A. H. T., & Chao, S. D. (2013). Liquid chloroform structure from computer simulation with a full ab initio intermolecular interaction potential. The Journal of chemical physics, 139(19), 194501.
[3] Chao, S. W., Li, A. H. T., & Chao, S. D. (2009). Molecular dynamics simulations of fluid methane properties using ab initio intermolecular interaction potentials. Journal of computational chemistry, 30(12), 1839-1849.
[4] Saenger, W., & Jeffrey, G. A. (1991). Hydrogen bonding in biological structures. Springer-Verlag, Berlin.
[5] Stone, A. J. (1996). The theory of intermolecular forces. International Series of Monographs on Chemistry, vol. 32.

延伸閱讀