MTASpec software for calculating the vibrational IR and Raman spectra of large molecules at ab initio level

Abstract

Fragmentation-based (FB) methods have been developed for enabling ab initio calculations on large molecules and clusters. However, a well-benchmarked FB-based utility, for calculating the vibrational spectra is not available. The present article reports a software package MTASpec, based on the FB-molecular tailoring approach (MTA), for computing the single point energy followed by vibrational IR and Raman spectra for spatially extended molecular systems. Accuracy and efficiency of MTASpec are assessed vis-à-vis their full calculation counterparts for some medium- to large-sized molecular systems using HF, DFT and MP2 theory employing large basis sets. The code is fully automated for use on linux platform, with the Gaussian suite of software at the back-end. It is envisaged that the MTASpec package would enable spectral studies of molecular systems containing ∼100 atoms and/or ∼10000 basis functions employing correlated theories with computational economy.

Program summary

Program title: MTASpec

CPC Library link to program files: https://doi.org/10.17632/m5b5zhxkfh.1

Licensing provisions: GPLv3

Programming language: Python, Fortran, C and Csh

External routines/libraries: Gaussian software [1]

Nature of problem: Single point energy, harmonic IR and Raman spectra computation for spatially extended molecular systems.

Solution method: The Fragment-based method, molecular tailoring approach (MTA) is employed for enabling economic and efficient ab initio computation of the vibrational spectra of large molecules and clusters.

References

• Gaussian 16, Revision A.03, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, G.A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A.V. Marenich, J. Bloino, B.G. Janesko, R. Gomperts, B. Mennucci, H.P. Hratchian, J.V. Ortiz, A.F. Izmaylov, J.L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V.G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J.A. Montgomery, Jr., J.E. Peralta, F. Ogliaro, M.J. Bearpark, J.J. Heyd, E.N. Brothers, K.N. Kudin, V.N. Staroverov, T.A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A.P. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, J.M. Millam, M. Klene, C. Adamo, R. Cammi, J.W. Ochterski, R.L. Martin, K. Morokuma, O. Farkas, J.B. Foresman, and D.J. Fox, Gaussian, Inc., Wallingford CT, 2016.

Introduction

Chemistry deals with the exploration of structures, properties and reactions of molecules. This involves the calculation and characterization of structures and relative energies not only of stable forms, but also of other stationary points on the potential energy surface, corresponding to reactive intermediates and transition structures. However, due to the required intricate experimental setup and/or rapid chemical phenomena, experimental spectroscopic scrutiny of molecules may be prohibitively difficult [1]. On the other hand, with the advent of powerful computational hardware and software, quantum chemical (QC) theories have turned out to be a major source of obtaining detailed information on molecular structure and reactivity. Apart from the calculation of energies, such high-level QC investigations offer valuable information on the vibrational frequencies and spectra of molecules.

Pursuing such ab initio-level QC investigations, however, requires large computational resources [2]. The computational cost of a QC method is determined by the level of theory and number of basis functions, N. The Hartree-Fock (HF) and Density Functional Theory (DFT) computations typically scale O($N^3$ or $N^4$) whereas correlated theories e.g. Møller-Plesset second- order perturbation theory (MP2) and coupled-cluster singles and doubles with perturbative triples viz. CCSD(T) scale as O($N^5$) and O($N^7$) respectively [3], [4]. With this in view, computing molecular properties employing correlated ab initio methods indeed proves to be a non-trivial task for large molecules. Furthermore, exploring spectral features of large molecular systems employing off-the-shelf hardware is generally not feasible [5].

To overcome the computational power requirements and to cut down the scaling complexities of the QC methods, several fragmentation-based [FB] methods have been proposed. These were inspired by the work of Christoffersen et al. in the early 1970s [6] followed by the studies by the groups of Yang and Gadre in the 1990s [7], [8]. These methods are summarized in References [9], [10], [11]. The central theme of all FB methods is to divide the parent system into subsystems such that the latter can be readily treated computationally, although the details of fragmentation differ. It is noteworthy to mention that all FB routes follow certain approximations due to which canonical full calculations (FC) which demand huge computational resources, can be circumvented effectively. However, most of the FB methods are generally restricted only to energetics and geometry optimization. The accurate and well-benchmarked theoretical treatment of vibrational IR and Raman spectra for large molecules require extensive computations (viz. the hessian matrix, derivatives of dipole moment and polarizability tensor etc.). We summarize below the attempts reported in the literature for the calculation of vibrational spectra employing FB methods.

In 2008, Li and co-workers demonstrated the Generalized Energy-Based Fragmentation (GEBF) [12] approach for geometry optimization and vibrational spectra of large molecules. In the same year, Gadre and coworkers employed the Molecular Tailoring Approach (MTA) for hessian and frequency calculation of large molecules [13]. Bieler et al. [14] proposed the Cartesian Tensor Transfer Method (CTTM) for calculating the vibrational spectra of polypeptides in 2011. The research groups of Kitaura and Fedorov [15] developed and applied fragmentation-based Fragment Molecular Orbital (FMO) for targeting large biomolecules. In 2014, the use of FMO was demonstrated [15] for the estimation of vibrational IR and Raman spectra for crambin, a protein, and similar large molecules. In 2015, Raghavachari and co-workers [16] reported similar development for computing vibrational IR and Raman spectra within Molecules-In-Molecules (MIM) procedure. Most of these methods are demonstrated for prototype molecules of moderate size and at the HF/DFT level of theory. In the case of MP2 theory, vibrational IR and Raman spectra of large molecules reported in the literature are restricted to either small molecules or are done with some approximations, e.g. resolution of identity (RI) method [17]. A clear-cut comparison with the conventional calculation of vibrational spectra and a discussion regarding the accuracy and efficiency of FB-based methods is also rather scantily found in the literature.

In general, FB methods are restricted to the developer's research group due to their intricate understanding and/or programming implementation. Thus, a well-benchmarked general utility within FB methods, is not available for uninitiated users. In this light, Collins and Bettens [10] have remarked that “the fragmentation methods should move ahead benchmarking phase and to create, as well as disseminate user-friendly programs”. In a similar spirit, Raghavachari and Saha [11] in their article on FB methods stated that “fragmentation schemes are user-specific, requiring substantial user-intervention, precluding their use as broad black box code”.

With this in view, the present work reports the development of the MTA-based black-box code, MTASpec, facilitating ab initio investigations on spatially extended large molecular structures. This package is aimed at the implementation of HF/DFT as well as correlated theories for the exploration of energetics and spectral features of large molecules/molecular clusters assisted by grafting procedure. The software is mostly written using python (version 2.7) as a programming language to be deployed on Linux-based machine/s with pre-installed Gaussian [18] software. In its present version, MTASpec can be utilized for an accurate estimation of energies/spectra (vibrational IR and Raman) of large, closed-shell molecular systems. We trust that systems containing up to 500 first- or second-row atoms and up to 2000 basis functions could be treated at HF/hybrid DFT theory by the present version. Further, molecules containing up to 100 first- or second-row atoms and up to 1000 basis functions could be treated at the MP2 level theory.