| Since the 1970 s,with the boom in computer science and technology,the use of electronic computers for scientific computation has become the dominant paradigm in theoretical research.The subdiscipline of chemistry,“theoretical chemistry”,has grown to be “theoretical and computational chemistry”.The main platform nowadays of chemical theory have changed from the formulae and graphs in the past to algorithms and programs.The development of theoretical models of chemistry and corresponding computer programs has become one of the best ways to develop chemical theories.Vibrational spectroscopy is one of the mainstream methods to study the dynamics of chemical systems.High-accuracy molecular vibrational spectra reflect the transition between vibrational energy levels and thus the Hamiltonian of a chemical system.Probe spectra of complex systems reflect dynamical information about the chemical environment in which the probe molecule is located,and reflect the way energy flows through the system.Simulating vibrational spectrum from first principles requires the potential energy surface and dipole moment function of the system.Transport properties in non-equilibrium statistical mechanics(including viscosity,diffusion,thermal conductivity,thermal diffusion)are important properties for gases.The theoretical analysis and prediction of these properties require solving the thermodynamic equations of the system.For the special case of dilute atomic gases,the Boltzmann equations of the system are formulated and solved by the Chapman–Enskog method.The transport properties of the gas can be thus obtained.However,to our knowledge,there is no stable and powerful computer software publicly available that can perform the whole set of operations from the intermolecular potential to the transport properties of the gas.During the period of doctoral study,the author has completed a series of innovative works to address the scientific and technical requirements above.The author developed new models and new software to solve such problems,in which high-precision potential energy surface plays the central role.The structure of the thesis is briefly described as follows.In Chapter 1,we reviews the history of the development of theoretical and computational chemistry,summarizes that one of the core tasks of the current stage of theoretical and computational chemistry is to develop new models and new software,and introduces the main programming languages and programming paradigms of current scientific computing software.In Chapter 2,the basic theoretical knowledge involved in the later chapters is briefly introduced,including(1)the fundamentals of molecular spectroscopy,introducing the central role of energy level gaps and transition dipole moment in the basic theory of spectroscopy;(2)the fundamentals of molecular vibrational spectroscopy,introducing the vibrational dynamics of molecules,leading to the concepts of potential energy surface and dipole moment surface;(3)the fundamentals of statistical mechanics,deriving statistical mechanics theories such as the Liouville equation,and derived the Boltzmann equation for dilute gases.In Chapter 3,multi-dimension Morse/Long-Range(md MLR)potential energy model is introduced.Spectroscopically accurate Potential Energy Surfaces(PES’s)are fundamental for explaining and making predictions of the Infrared and Microwave spectra of van der Waals(vd W)complexes,and the model used for the potential energy function is critically important for providing accurate,robust and portable analytical PES’s.The Morse/Long-Range(MLR)model has proved to be one of the most general,flexible and accurate 1-D model potentials,as it has physically meaningful parameters,is flexible,smooth and differentiable everywhere,to all orders,and extrapolates sensibly at both long and short range.The md MLR potential energy model described herein is based on that 1-D MLR model,and has proved to be effective and accurate in the potentiology of various types of vd W complexes.In this chapter,we introduce the current status of development of the md MLR model and its application to vd W complexes.The future of the md MLR model is also discussed.This chapter can serve as a tutorial of construction of an md MLR PES.In Chapter 4,the accurate computation of dipole moment function is discussed.Recently,more attention have been paid on the construction of dipole moment functions(DMF)using theoretical methods.However,the computational methods to construct DMFs are not validated as much as those for potential energy surfaces do.In this chapter,using Ar···He as an example,we tested how spectroscopy-accuracy DMFs can be constructed using ab initio methods.We especially focused on the basis set dependency in this scenario,i.e.,the convergence of DMF with the sizes of basis sets,basis set superposition error,and mid-bond functions.We also tested the explicitly correlated method,which converges with smaller basis sets than the conventional methods do.This work can serve as a pictorial sample of all these computational technologies behaving in the context of constructing DMFs.In Chapter 5,the Local Quantum Vibration Embedding(LQVE)theoretical framework is introduced.The vibrational spectroscopy of probe molecules in complex system is a common way to study the dynamics.Vibration of the probe has to be described by quantum mechanics,whilst the dynamics of the complex system influence of spectra of the probes greatly,and thus dynamics information is included.We developed the classical dynamics/quantum mechanics hybrid spectroscopic theory,where in every step of molecular dynamics simulation,the quantum description of the probe is embedded,and the instantaneous variables such as vibrational frequency and transition dipole moment are computed on-thefly.Using the separation of degrees of freedom,we can complete the simulation of the probe spectra with low computational costs.In Chapter 6,the computer program PENG computing the transport properties of dilute binary gas mixture is introduced.The fundamental properties of molecules bridge experiment and theory.Transport properties(diffusion,thermal diffusion,thermal conductivity and viscosity)of binary mixtures are measurable in experiments,and well-defined in theory,but difficult to compute with high accuracy.In addition to high-accuracy inter-molecular potential energy curves(PECs),a reliable and high-order solution program that compute the properties based on the PECs is required.In this work,we present a computer program called PENG that performs the collision integration numerically,and solves the Boltzmann equation in Chapman–Enskog fashion.The program has been devised to perform both parts of the solution procedure to arbitrary order,so that no hard-coded limitation will prevent a user from computing at higher precision,except the amount of RAM and the required computational time.PENG is welldesigned in an Object-Oriented Programming(OOP)fashion,which make the program clear and easy to modify.In addition to the end-user oriented program,PENG is also compiled as a dynamic shared library that may readily be extended and embedded in users’ programs.Summary and outlook are given in Chapter 7. |
PDF Full Text Request