Monte Carlo method time correlation function

The approach to the evaluation of vibrational spectra described above is based on classical simulations for which quantum corrections are possible. The incorporation of quantum effects directly in simulations of large molecular systems is one of the most challenging areas in theoretical chemistry today. The development of quantum simulation methods is particularly important in the area of molecular spectroscopy for which quantum effects can be important and where the goal is to use simulations to help understand the structural and dynamical origins of changes in spectral lineshapes with environmental variables such as the temperature. The direct evaluation of quantum time- correlation functions for anharmonic systems is extremely difficult. Our initial approach to the evaluation of finite temperature anharmonic effects on vibrational lineshapes is derived from the fact that the moments of the vibrational lineshape spectrum can be expressed as functions of expectation values of positional and momentum operators. These expectation values can be evaluated using extremely efficient quantum Monte-Carlo techniques. The main points are summarized below. 

Recent years have seen the extensive application of computer simulation techniques to the study of condensed phases of matter. The two techniques of major importance are the Monte Carlo method and the method of molecular dynamics. Monte Carlo methods are ways of evaluating the partition function of a many-particle system through sampling the multidimensional integral that defines it, and can be used only for the study of equilibrium quantities such as thermodynamic properties and average local structure. Molecular dynamics methods solve Newton s classical equations of motion for a system of particles placed in a box with periodic boundary conditions, and can be used to study both equilibrium and nonequilibrium properties such as time correlation functions. 

Dynamical quantities are harder to obtain, since the QMC representations only give access to imaginary-time correlation function. With the exception of measurements of spin gaps, which can be obtained from an exponential decay of the spin-spin correlation function in imaginary time, the measurement of real-time or real-frequency correlation functions requires an ill-posed analytical continuation of noisy Monte Carlo data, for example using the Maximum Entropy Method [46-48]. 

In ordinary applications of the Monte Carlo method the boundary conditions are carefully contrived to make the environment of the sample mimic that in the interior of a macroscopic system (cf. Chapter 4, Section 4). At the same time, one of the attractive features of such calculations is that one can extract from them, along with thermodynamic averages, microscopic information on the structure of the system. The most common such data concern the details of the pair correlation function, although more complicated information is also available. Along this line, for instance. Hoover have proposed a study... 

In the next four sections, we discuss the four principal types of application of molecular dynamics. Section 3 very briefly describes the problem of the approach to equilibrium. Section 4 deals with the evaluation of equilibrium thermodynamic functions through a discussion of the dynamical equation of state. In Section 5, we consider the evaluation of equilibrium time correlation functions, detailing the application of the combined Monte Carlo-molecular-dynamics method to the time correlation functions for self-diffusion. Section 6 deals with nonequilibrium molecular dynamics and in particular with a calculation for self-diffusion. 

This chapter is organized as follows. In section 1.1, we introduce our notation and present the details of the molecular and mesoscale simulations the expanded ensemble-density of states Monte Carlo method,and the evolution equation for the tensor order parameter [5]. The results of both approaches are presented and compared in section 1.2 for the cases of one or two nanoscopic colloids immersed in a confined liquid crystal. Here the emphasis is on the calculation of the effective interaction (i.e. potential of mean force) for the nanoparticles, and also in assessing the agreement between the defect structures found by the two approaches. In section 1.3 we apply the mesoscopic theory to a model LC-based sensor and analyze the domain coarsening process by monitoring the equal-time correlation function for the tensor order parameter, as a function of the concentration of adsorbed nanocolloids. We present our conclusions in Section 1.4. 

The Monte Carlo method, which is an alternative to molecular dynamics, consists of the generation of a sequence of molecular configurations in such a way that there is a Boltzmann distribution in the potential energies of the selected configurations. Then, the average of a molecular property over all configurations gives the appropriate value for comparisons with experiment. This method can only be u.sed to calculate equilibrium properties of a system, while equilibrium, transport (e.g., diffusion coefficients), time dependent (e.g., time correlation functions), and spectroscopic (e.g., infrared and Raman lineshapes) properties are accessible to calculation with the molecular dynamics method. 

Fortunately, progress can be made because the Heisenberg operator involved in the evolution of time correlation functions or the quantum mechanical density matrix involves not only the forward time evolution operator but also its inverse. It is well understood that a dramatic phase cancellation takes place between these two propagation steps upon integration, and this cancellation is entirely responsible for the failure of Monte Carlo methods. To remedy this situation, Mgtoi and Thompson proposed a forward-backward semiclassical approximation (36) in which the time evolution operator and its inverse are combined into a single semiclassical treatment. This procedure is equivalent to a sta-... 

The important difference between the microkinetics approach and the kinetic Monte Carlo simulation is that in the former diffusion is not explicitly included. Reaction probabilities are again based on the Eyring transition state rate expression. Its benefit is a substantial reduction in computational time length. Similar as in the kinetic Monte Carlo method, production rates as a function of reaction condition can be computed. These kinetic data can be correlated with changes in surface composition of the adsorbed reactant and intermediate overlayer. Also, rates of reaction intermediate production or removal can be deduced. 

For the equihbrium properties and for the kinetics under quasi-equilibrium conditions for the adsorbate, the transfer matrix technique is a convenient and accurate method to obtain not only the chemical potentials, as a function of coverage and temperature, but all other thermodynamic information, e.g., multiparticle correlators. We emphasize the economy of the computational effort required for the application of the technique. In particular, because it is based on an analytic method it does not suffer from the limitations of time and accuracy inherent in statistical methods such as Monte Carlo simulations. The task of variation of Hamiltonian parameters in the process of fitting a set of experimental data (thermodynamic and... 

Figure 7-1. Typical autocorrelation function of the energy. In this example it is calculated for the case of benzophenone in water simulated with Monte Carlo Metropolis method. The calculated auto-correlation function (circles) is fitted to the exponential decay (line) and the correlation time t is obtained using Eq. (7-7) applied in the fitted function shown in Eq. (7-8)...

At the same time, the LDA gave an a posteriori justification of the old Xa method by Slater, because the latter is a special LDA variant without correlation. The corresponding spin-dependent version of the LDA is called a local spin-density approximation (LSDA or LSD or just spin-polarized LDA), and even now when people talk of LDA functionals, they always refer to its generalized form for systems with (potentially) unpaired spins. Among the most influential LDA parametrizations, the one of von Barth and Hedin (BH) [154] and the one of Vosko, Wilk and Nusair (VWN) [155] are certainly worth mentioning. The latter is based on the very accurate Monte Carlo-type calculations of Ceperley and Alder [156] for the uniform electron gas, as indicated above. 

Buendia et al. calculated many states of the iron atom with VMC using orbitals obtained from the parametrized optimal effective potential method with all electrons included. Iron is a particularly difficult system, and the VMC results are only moderately accurate. The same authors also pubhshed VMC and Green s functions quantum Monte Carlo (GFMQ calculations on the first transition-row atoms with all electrons. GFMC is a variant of DMC where intermediate steps are used to remove the time step error. Cafiarel et al. presented a very careful study on the role of electron correlation and relativistic effects in the copper atom using all-electron DMC. Relativistic effects were calculated with the Dirac-Fock model. Several states of the atom were evaluated and an accuracy of about 0.15 eV was achieved with a single determinant. ... 

---