Development of generalized potential-energy surfaces using many-body expansions, neural networks, and moiety energy approximations.

A general method for the development of potential-energy hypersurfaces is presented. The method combines a many-body expansion to represent the potential-energy surface with two-layer neural networks (NN) for each M-body term in the summations. The total number of NNs required is significantly reduced by employing a moiety energy approximation. An algorithm is presented that efficiently adjusts all the coupled NN parameters to the database for the surface. Application of the method to four different systems of increasing complexity shows that the fitting accuracy of the method is good to excellent. For some cases, it exceeds that available by other methods currently in literature. The method is illustrated by fitting large databases of ab initio energies for Si(n) (n=3,4,...,7) clusters obtained from density functional theory calculations and for vinyl bromide (C(2)H(3)Br) and all products for dissociation into six open reaction channels (12 if the reverse reactions are counted as separate open channels) that include C-H and C-Br bond scissions, three-center HBr dissociation, and three-center H(2) dissociation. The vinyl bromide database comprises the ab initio energies of 71 969 configurations computed at MP4(SDQ) level with a 6-31G(d,p) basis set for the carbon and hydrogen atoms and Huzinaga's (4333/433/4) basis set augmented with split outer s and p orbitals (43321/4321/4) and a polarization f orbital with an exponent of 0.5 for the bromine atom. It is found that an expansion truncated after the three-body terms is sufficient to fit the Si(5) system with a mean absolute testing set error of 5.693x10(-4) eV. Expansions truncated after the four-body terms for Si(n) (n=3,4,5) and Si(n) (n=3,4,...,7) provide fits whose mean absolute testing set errors are 0.0056 and 0.0212 eV, respectively. For vinyl bromide, a many-body expansion truncated after the four-body terms provides fitting accuracy with mean absolute testing set errors that range between 0.0782 and 0.0808 eV. These errors correspond to mean percent errors that fall in the range 0.98%-1.01%. Our best result using the present method truncated after the four-body summation with 16 NNs yields a testing set error that is 20.3% higher than that obtained using a 15-dimensional (15-140-1) NN to fit the vinyl bromide database. This appears to be the price of the added simplicity of the many-body expansion procedure.

[1]  Janet E. Jones.  On the determination of molecular fields. —II. From the equation of state of a gas , 1924 .

[2]  P. Morse.  Diatomic Molecules According to the Wave Mechanics. II. Vibrational Levels , 1929 .

[3]  Michel Dupuis,et al.  Theoretical three-dimensional potential-energy surface for the reaction of Be with HF , 1983 .

[4]  S. Carter,et al.  Approximate single-valued representations of multivalued potential energy surfaces , 1984 .

[5]  M. Baskes,et al.  Embedded-atom method: Derivation and application to impurities, surfaces, and other defects in metals , 1984 .

[6]  J. Murrell,et al.  Molecular Potential Energy Functions , 1985 .

[7]  Lawrence B. Harding,et al.  Ab initio calculations of electronic and vibrational energies of HCO and HOC , 1986 .

[8]  J. Tersoff,et al.  New empirical model for the structural properties of silicon. , 1986, Physical review letters.

[9]  J. Tersoff,et al.  Empirical interatomic potential for silicon with improved elastic properties. , 1988, Physical review. B, Condensed matter.

[10]  J. Tersoff,et al.  Modeling solid-state chemistry: Interatomic potentials for multicomponent systems. , 1989, Physical review. B, Condensed matter.

[11]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[12]  H. C. Andersen,et al.  Interatomic potential for silicon clusters, crystals, and surfaces. , 1990, Physical review. B, Condensed matter.

[13]  D. Brenner,et al.  Empirical potential for hydrocarbons for use in simulating the chemical vapor deposition of diamond films. , 1990, Physical review. B, Condensed matter.

[14]  D. W. Noid,et al.  Potential energy surfaces for macromolecules. A neural network technique , 1992 .

[15]  M. Baskes,et al.  Modified embedded-atom potentials for cubic materials and impurities. , 1992, Physical review. B, Condensed matter.

[16]  M. A. Collins,et al.  Molecular potential energy surfaces by interpolation , 1994 .

[17]  Martin T. Hagan,et al.  Neural network design , 1995 .

[18]  Steven D. Brown,et al.  Neural network models of potential energy surfaces , 1995 .

[19]  Donald J. Kouri,et al.  Distributed approximating functional approach to fitting multi-dimensional surfaces , 1996 .

[20]  Mark N. Gibbs,et al.  Combining ab initio computations, neural networks, and diffusion Monte Carlo: An efficient method to treat weakly bound molecules , 1996 .

[21]  Donald J. Kouri,et al.  Distributed approximating functional approach to fitting and predicting potential surfaces. 1. Atom-atom potentials , 1996 .

[22]  H. Rabitz,et al.  A global H2O potential energy surface for the reaction O(1D)+H2→OH+H , 1996 .

[23]  Donald J. Kouri,et al.  Distributed approximating functional fit of the H3 ab initio potential-energy data of Liu and Siegbahn , 1997 .

[24]  Joel M. Bowman,et al.  Vibrational self-consistent field method for many-mode systems: A new approach and application to the vibrations of CO adsorbed on Cu(100) , 1997 .

[25]  J. J. Soares Neto,et al.  The fitting of potential energy surfaces using neural networks: Application to the study of vibrational levels of H3+ , 1998 .

[26]  Martin Quack,et al.  Global analytical potential hypersurfaces for large amplitude nuclear motion and reactions in methane. I. Formulation of the potentials and adjustment of parameters to ab initio data and experimental constraints , 1998 .

[27]  Kersti Hermansson,et al.  Representation of Intermolecular Potential Functions by Neural Networks , 1998 .

[28]  J. J. Soares Neto,et al.  The fitting of potential energy surfaces using neural networks. Application to the study of the photodissociation processes , 1998 .

[29]  H. Rabitz,et al.  General foundations of high‐dimensional model representations , 1999 .

[30]  T Hollebeek,et al.  Constructing multidimensional molecular potential energy surfaces from ab initio data. , 2003, Annual review of physical chemistry.

[31]  H. Rabitz,et al.  Efficient input-output model representations , 1999 .

[32]  Joel M. Bowman,et al.  Variational Calculations of Rotational−Vibrational Energies of CH4 and Isotopomers Using an Adjusted ab Initio Potential , 2000 .

[33]  H. Rabitz,et al.  High Dimensional Model Representations , 2001 .

[34]  Herschel Rabitz,et al.  The Ar–HCl potential energy surface from a global map-facilitated inversion of state-to-state rotationally resolved differential scattering cross sections and rovibrational spectral data , 2001 .

[35]  B. Braams,et al.  A local interpolation method for direct classical dynamics calculations , 2001 .

[36]  R. Komanduri,et al.  Molecular dynamics simulation of the nanometric cutting of silicon , 2001 .

[37]  Herschel Rabitz,et al.  Efficient Implementation of High Dimensional Model Representations , 2001 .

[38]  Herschel Rabitz,et al.  Global, nonlinear algorithm for inverting quantum-mechanical observations , 2001 .

[39]  Herschel Rabitz,et al.  Constructing global functional maps between molecular potentials and quantum observables , 2001 .

[40]  H. Rabitz,et al.  High Dimensional Model Representations Generated from Low Dimensional Data Samples. I. mp-Cut-HDMR , 2001 .

[41]  Emily Weiss,et al.  Achieving the laboratory control of quantum dynamics phenomena using nonlinear functional maps , 2001 .

[42]  H. Rabitz,et al.  Practical Approaches To Construct RS-HDMR Component Functions , 2002 .

[43]  Michael A. Collins,et al.  Molecular potential-energy surfaces for chemical reaction dynamics , 2002 .

[44]  Nicholas C. Handy,et al.  On the representation of potential energy surfaces of polyatomic molecules in normal coordinates , 2002 .

[45]  H. Rabitz,et al.  Random Sampling-High Dimensional Model Representation (RS-HDMR) with Nonuniformly Distributed Variables: Application to an Integrated Multimedia/ Multipathway Exposure and Dose Model for Trichloroethylene , 2003 .

[46]  H. Rabitz,et al.  Reproducing kernel Hilbert space interpolation methods as a paradigm of high dimensional model representations: Application to multidimensional potential energy surface construction , 2003 .

[47]  Donald L. Thompson,et al.  Interpolating moving least-squares methods for fitting potential energy surfaces: Detailed analysis of one-dimensional applications , 2003 .

[48]  Herschel Rabitz,et al.  High‐dimensional model representations generated from low order terms—lp‐RS‐HDMR , 2003, J. Comput. Chem..

[49]  Akio Kawano,et al.  Improving the accuracy of interpolated potential energy surfaces by using an analytical zeroth-order potential function. , 2004, The Journal of chemical physics.

[50]  Akio Kawano,et al.  Interpolating moving least-squares methods for fitting potential energy surfaces: applications to classical dynamics calculations. , 2004, The Journal of chemical physics.

[51]  Herschel Rabitz,et al.  Multicut‐HDMR with an application to an ionospheric model , 2004, J. Comput. Chem..

[52]  A. Gross,et al.  Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks , 2004 .

[53]  Neil Shenvi,et al.  Efficient chemical kinetic modeling through neural network maps. , 2004, The Journal of chemical physics.

[54]  R Komanduri,et al.  Ab initio potential-energy surfaces for complex, multichannel systems using modified novelty sampling and feedforward neural networks. , 2005, The Journal of chemical physics.

[55]  Joel M Bowman,et al.  Ab initio potential energy and dipole moment surfaces for H5O2 +. , 2005, The Journal of chemical physics.

[56]  Ranga Komanduri,et al.  Molecular dynamics investigations of the dissociation of SiO2 on an ab initio potential energy surface obtained using neural network methods. , 2006, The Journal of chemical physics.

[57]  Sergei Manzhos,et al.  Using neural networks to represent potential surfaces as sums of products. , 2006, The Journal of chemical physics.

[58]  H. Rabitz,et al.  Random sampling-high dimensional model representation (RS-HDMR) and orthogonality of its different order component functions. , 2006, The journal of physical chemistry. A.

[59]  Sergei Manzhos,et al.  A random-sampling high dimensional model representation neural network for building potential energy surfaces. , 2006, The Journal of chemical physics.

[60]  T. Carrington,et al.  A nested molecule-independent neural network approach for high-quality potential fits. , 2006, The journal of physical chemistry. A.

[61]  Satish T. S. Bukkapatnam,et al.  Parametrization of interatomic potential functions using a genetic algorithm accelerated with a neural network , 2006 .

[62]  Herschel Rabitz,et al.  Estimation of molecular properties by high-dimensional model representation. , 2006, The journal of physical chemistry. A.

[63]  Richard Dawes,et al.  Interpolating moving least-squares methods for fitting potential energy surfaces: computing high-density potential energy surface data from low-density ab initio data points. , 2007, The Journal of chemical physics.

[64]  Sergei Manzhos,et al.  Using redundant coordinates to represent potential energy surfaces with lower-dimensional functions. , 2007, The Journal of chemical physics.

[65]  Donald L Thompson,et al.  Interpolating moving least-squares methods for fitting potential energy surfaces: Improving efficiency via local approximants. , 2007, The Journal of chemical physics.

[66]  M. Malshe,et al.  Theoretical investigation of the dissociation dynamics of vibrationally excited vinyl bromide on an ab initio potential-energy surface obtained using modified novelty sampling and feedforward neural networks. II. Numerical application of the method. , 2007, The Journal of chemical physics.

[67]  Donald L Thompson,et al.  Interpolating moving least-squares methods for fitting potential energy surfaces: an application to the H2CN unimolecular reaction. , 2007, The Journal of chemical physics.

[68]  Sergei Manzhos,et al.  Using neural networks, optimized coordinates, and high-dimensional model representations to obtain a vinyl bromide potential surface. , 2008, The Journal of chemical physics.

[69]  Joel M Bowman,et al.  Full-dimensional quantum calculations of ground-state tunneling splitting of malonaldehyde using an accurate ab initio potential energy surface. , 2008, The Journal of chemical physics.

[70]  Richard Dawes,et al.  Interpolating moving least-squares methods for fitting potential energy surfaces: a strategy for efficient automatic data point placement in high dimensions. , 2008, The Journal of chemical physics.

[71]  L. Raff,et al.  Cis-->trans, trans-->cis isomerizations and N-O bond dissociation of nitrous acid (HONO) on an ab initio potential surface obtained by novelty sampling and feed-forward neural network fitting. , 2008, The Journal of chemical physics.

[72]  R Komanduri,et al.  A self-starting method for obtaining analytic potential-energy surfaces from ab initio electronic structure calculations. , 2009, The journal of physical chemistry. A.