The chapter 2, we deal with the challenge b). It focuses on the variational Monte Carlo (VMC) and the wavefunction optimization methods based on VMC. The performance of different methods are displayed through the op- timization of the Jastrow factor in our test case Beryllium dimer and the efficiency is improving surprisingly during the evolution of these methods. In chapter 3, we focus on the challenge a). It describes the wavefunc- tion ansatz used by our simulation. In this thesis, we introduce the atomic hybrid orbitals which significantly increase the compactness of our wavefunc- tion without hurting accuracy. This chapter also explain how to optimize the determinant in a way that the number of variational parameters scales only linearly with the system size. This further helps the efficiency of the wavefunction optimization. In chapters 4 and 5, the issue c) is explained in detail. In chapter 4, a second order Langevin dynamics (SLD) scheme is devised particularly for QMC and this thesis improves this scheme by developing a better integration method. Here, we also highlight the remarkable power of the force covari- ance matrix which can be defined only in QMC and is capable of accelerating the slow modes of a dynamics. In chapter 5, this SLD for QMC is validated through intensive benchmarking on the calculation of the vibrational frequen- cies of water and other small molecules. It is shown that many systematic biases in our MD scheme and QMC evaluation can be controlled so that we are confident to push forward this ab initio molecular dynamics for applica- tions on large systems. Finally in chapter 6, we perform the simulation of liquid water with all the preparation done in the previous chapters. The results are encouraging since we’ve closed the discrepancy of the peak positions of RDFs between experi- ments and ab initio simulations. The power of QMC is also demonstrated by the fact that the shapes of our RDFs are much less structured than previous DFT-based ab initio simulations even if the two water molecule interaction is dealt with the same level of accuracy as the DFT/BLYP calculation. In this chapter, we have also studied the features of hydrogen bonds in our simulation of liquid water. All our results indicate that it is important to consider the quantum nature of the ions for a faithful description of liquid water. This will be left for future studies, possible in principle even within the QMC approach.
The chapter 2, we deal with the challenge b). It focuses on the variational Monte Carlo (VMC) and the wavefunction optimization methods based on VMC. The performance of different methods are displayed through the op- timization of the Jastrow factor in our test case Beryllium dimer and the efficiency is improving surprisingly during the evolution of these methods. In chapter 3, we focus on the challenge a). It describes the wavefunc- tion ansatz used by our simulation. In this thesis, we introduce the atomic hybrid orbitals which significantly increase the compactness of our wavefunc- tion without hurting accuracy. This chapter also explain how to optimize the determinant in a way that the number of variational parameters scales only linearly with the system size. This further helps the efficiency of the wavefunction optimization. In chapters 4 and 5, the issue c) is explained in detail. In chapter 4, a second order Langevin dynamics (SLD) scheme is devised particularly for QMC and this thesis improves this scheme by developing a better integration method. Here, we also highlight the remarkable power of the force covari- ance matrix which can be defined only in QMC and is capable of accelerating the slow modes of a dynamics. In chapter 5, this SLD for QMC is validated through intensive benchmarking on the calculation of the vibrational frequen- cies of water and other small molecules. It is shown that many systematic biases in our MD scheme and QMC evaluation can be controlled so that we are confident to push forward this ab initio molecular dynamics for applica- tions on large systems. Finally in chapter 6, we perform the simulation of liquid water with all the preparation done in the previous chapters. The results are encouraging since we’ve closed the discrepancy of the peak positions of RDFs between experi- ments and ab initio simulations. The power of QMC is also demonstrated by the fact that the shapes of our RDFs are much less structured than previous DFT-based ab initio simulations even if the two water molecule interaction is dealt with the same level of accuracy as the DFT/BLYP calculation. In this chapter, we have also studied the features of hydrogen bonds in our simulation of liquid water. All our results indicate that it is important to consider the quantum nature of the ions for a faithful description of liquid water. This will be left for future studies, possible in principle even within the QMC approach.
Quantum Monte Carlo methods are accurate and promising many body techniques for electronic structure calculations which, in the last years, are encountering a growing interest thanks to their favorable scaling with the system size and their efficient parallelization, particularly suited for the modern high performance computing facilities. The ansatz of the wave function and its variational flexibility are crucial points for both the accurate description of molecular properties and the capabilities of the method to tackle large systems. In this paper, we extensively analyze, using different variational ansatzes, several properties of the water molecule, namely, the total energy, the dipole and quadrupole momenta, the ionization and atomization energies, the equilibrium configuration, and the harmonic and fundamental frequencies of vibration. The investigation mainly focuses on variational Monte Carlo calculations, although several lattice regularized diffusion Monte Carlo calculations are also reported. Through a systematic study, we provide a useful guide to the choice of the wave function, the pseudopotential, and the basis set for QMC calculations. We also introduce a new method for the computation of forces with finite variance on open systems and a new strategy for the definition of the atomic orbitals involved in the Jastrow-Antisymmetrised Geminal power wave function, in order to drastically reduce the number of variational parameters. This scheme significantly improves the efficiency of QMC energy minimization in case of large basis sets.
Quantum Monte Carlo methods are accurate and promising many body techniques for electronic structure calculations which, in the last years, are encountering a growing interest thanks to their favorable scaling with the system size and their efficient parallelization, particularly suited for the modern high performance computing facilities. The ansatz of the wave function and its variational flexibility are crucial points for both the accurate description of molecular properties and the capabilities of the method to tackle large systems. In this paper, we extensively analyze, using different variational ansatzes, several properties of the water molecule, namely: the total energy, the dipole and quadrupole momenta, the ionization and atomization energies, the equilibrium configuration, and the harmonic and fundamental frequencies of vibration. The investigation mainly focuses on variational Monte Carlo calculations, although several lattice regularized diffusion Monte Carlo calculations are also reported. Through a systematic study, we provide a useful guide to the choice of the wave function, the pseudo potential, and the basis set for QMC calculations. We also introduce a new strategy for the definition of the atomic orbitals involved in the Jastrow - Antisymmetrised Geminal power wave function, in order to drastically reduce the number of variational parameters. This scheme significantly improves the efficiency of QMC energy minimization in case of large basis sets.
Since the form of PN is the same as that occurring in the statistical mechanics of the classical gas (replace f 2 ( r ij ) by exp[- V ( r ij /kT ]), we can use the same integration techniques that are used in the classical problem. W. L. McMillan, 1965 Summary Building on the random walk methods developed in the previous chapter, we show how to compute properties of many-body trial wavefunctions using a random walk. This method, called variational Monte Carlo, is the simplest stochastic quantum many-body technique. Whereas mean-field methods are usually limited to single determinants, variational Monte Carlo can treat any correlated trial function, as long as its values are computable. We discuss how to optimize such trial wavefunctions, how to compute their momentum distribution, how to use non-local pseudopotentials, how to compute excited states, and how to correct for the finite size of the simulation cell. Deterministic quantum methods have difficulties. For example, the Hartree–Fock method assumes the wavefunction is a single Slater determinant, neglecting correlation. If one expands as a sum of determinants, it is very difficult to have the results size-consistent since the number of determinants needed will grow exponentially with the system size. As we have seen, the DMFT method introduced in Ch. 16 assumes locality. In Ch. 6 we discussed general properties of many-body wavefunctions. Using Monte Carlo methods, we can directly incorporate correlation into a wavefunction, without having to make any further approximations other than the form of the correlation factors. In many cases the energy and other properties are very close to the exact results. Some of the usual restrictions on the form of the many-body wavefunction are not an issue in variational Monte Carlo. The most important generalization of the HF wavefunction is to put correlation directly into the wavefunction via the “Jastrow” factor. At next order, one can use the “backflow” wavefunction, in which correlation is also built into the determinant. The variational Monte Carlo method (VMC) was first used by McMillan [44] to calculate the ground-state properties of superfluid 4 He. One of the key problems at that time was whether the observed superfluid properties were a consequence of Bose condensation.
TurboRVB is a computational package for ab initio Quantum Monte Carlo (QMC) simulations of both molecular and bulk electronic systems. The code implements two types of well established QMC algorithms: Variational Monte Carlo (VMC) and diffusion Monte Carlo in its robust and efficient lattice regularized variant. A key feature of the code is the possibility of using strongly correlated many-body wave functions (WFs), capable of describing several materials with very high accuracy, even when standard mean-field approaches [e.g., density functional theory (DFT)] fail. The electronic WF is obtained by applying a Jastrow factor, which takes into account dynamical correlations, to the most general mean-field ground state, written either as an antisymmetrized geminal power with spin-singlet pairing or as a Pfaffian, including both singlet and triplet correlations. This WF can be viewed as an efficient implementation of the so-called resonating valence bond (RVB) Ansatz, first proposed by Pauling and Anderson in quantum chemistry [L. Pauling, The Nature of the Chemical Bond (Cornell University Press, 1960)] and condensed matter physics [P.W. Anderson, Mat. Res. Bull 8, 153 (1973)], respectively. The RVB Ansatz implemented in TurboRVB has a large variational freedom, including the Jastrow correlated Slater determinant as its simplest, but nontrivial case. Moreover, it has the remarkable advantage of remaining with an affordable computational cost, proportional to the one spent for the evaluation of a single Slater determinant. Therefore, its application to large systems is computationally feasible. The WF is expanded in a localized basis set. Several basis set functions are implemented, such as Gaussian, Slater, and mixed types, with no restriction on the choice of their contraction. The code implements the adjoint algorithmic differentiation that enables a very efficient evaluation of energy derivatives, comprising the ionic forces. Thus, one can perform structural optimizations and molecular dynamics in the canonical NVT ensemble at the VMC level. For the electronic part, a full WF optimization (Jastrow and antisymmetric parts together) is made possible, thanks to state-of-the-art stochastic algorithms for energy minimization. In the optimization procedure, the first guess can be obtained at the mean-field level by a built-in DFT driver. The code has been efficiently parallelized by using a hybrid MPI-OpenMP protocol, which is also an ideal environment for exploiting the computational power of modern Graphics Processing Unit accelerators.
Dans cette thèse nous présentons des progrès algorithmiques ainsi que plusieurs applications des méthodes de Monte Carlo quantique (QMC) pour simulations à partir des premiers principes. Les améliorations que nous proposons permettent d'étudier par QMC des systèmes de plus grosse taille voire périodiques, avec l'ambition de faire du QMC une alternative valable à la théorie de la fonctionnelle de la densité (DFT). Tous les résultats ont été obtenus par le logiciel TurboRVB. D'abord, nous présentons une implémentation du QMC basée sur la fonction d'onde Jastrow-Geminale qui combine une grande flexibilité avec un traitement précis des corrélations électroniques. On a appliqué une technique originale de plongement pour réduire la taille de la base atomique à la molécule d'eau ainsi qu'à un modèle simplifié du transfert de protons (TP) dans l'eau. Nos résultats ouvrent la voie à l'étude des phénomènes microscopiques tels que le TP directement par QMC. Ensuite, on a amélioré notre méthode afin de simuler les solides cristallins. Grâce à une nouvelle procédure pour choisir de manière appropriée les conditions aux limites, nous avons pu réduire les erreurs de taille finie qui affectent les simulations QMC des solides. Sur la base des techniques développées, nous étudions enfin le supraconducteur FeSe. Le QMC fournit le meilleur résultat concernant sa structure cristalline; via une étude systématique du paysage énergétique à différentes configurations magnétiques, nous montrons un lien fort entre la structure, le magnétisme et les mouvements de charge dans ce matériau, prélude à une compréhension quantitative de la supraconductivité à haute température des premiers principes.
Simulations de Monte Carlo quantique améliorées : de systèmes ouverts aux solides cristallins Dans cette thèse nous présentons des progrès algorithmiques ainsi que plusieurs applications des méthodes de Monte Carlo quantique (QMC) pour simulations à partir des premiers principes. Les améliorations que nous proposons permettent d'étudier par QMC des systèmes de plus grosse taille voire périodiques, avec l'ambition de faire du QMC une alternative valable à la théorie de la fonctionnelle de la densité (DFT). Tous les résultats ont été obtenus par le logiciel TurboRVB. D'abord, nous présentons une implémentation du QMC basée sur la fonction d'onde Jastrow-Geminale qui combine une grande flexibilité avec un traitement précis des corrélations électroniques. On a appliqué une technique originale de plongement pour réduire la taille de la base atomique à la molécule d'eau ainsi qu'à un modèle simplifié du transfert de protons (TP) dans l'eau. Nos résultats ouvrent la voie à l'étude des phénomènes microscopiques tels que le TP directement par QMC. Ensuite, on a amélioré notre méthode afin de simuler les solides cristallins. Grâce à une nouvelle procédure pour choisir de manière appropriée les conditions aux limites, nous avons pu réduire les erreurs de taille finie qui affectent les simulations QMC des solides. Sur la base des techniques développées, nous étudions enfin le supraconducteur FeSe. Le QMC fournit le meilleur résultat concernant sa structure cristalline; via une étude systématique du paysage énergétique à différentes configurations magnétiques, nous montrons un lien fort entre la structure, le magnétisme et les mouvements de charge dans ce matériau, prélude à une compréhension quantitative de la supraconductivité à haute température des premiers principes.