1
Interpretability Meets Accuracy in Computational Spectroscopy: The Virtual Multifrequency Spectrometer

Vincenzo Barone1 and Cristina Puzzarini2

1 Scuola Normale Superiore, Piazza dei Cavalieri 7, 56126 Pisa, Italy

2 Università di Bologna, Dipartimento di Chimica "Giacomo Ciamician", Via Selmi 2, 40126 Bologna, Italy

The virtual multifrequency spectrometer (VMS), under active development in our laboratories over the last few years, is shortly described in this chapter by means of selected spectroscopic techniques and a few representative case studies. The VMS project aims to offer an answer to the following question: is it possible to turn strongly specialized research in the field of computational spectroscopy into robust and user-friendly aids to experiments and industrial applications? VMS contains a number of tools devised to increase the interaction between researchers with different background and to push toward new frontiers in computational chemistry. As a matter of fact, the terrific advancements in computational spectroscopy and the wide availability of computational and analytic tools are paving the route toward the study of problems that were previously too difficult or impossible to be solved and let imagine even more ambitious targets for fundamental and applied research. Under such circumstances, a robust, flexible, and user-friendly tool can allow for moving data analysis toward a proactive process of strategic decisions and actions. This chapter starts from these premises, and it proposes a perspective for a new virtual platform aimed at integrating past developments in theory, algorithms, and software with new workflow management and visualization tools. After a short review of the underlying theoretical framework, the features of the principal tools available in the current version of VMS for a selection of spectroscopic techniques are addressed in some details. Next, four case studies are presented, thus aiming to illustrate possible applications of VMS to systems of current interest for both fundamental and applied research. These applications convincingly show that even if several extensions of the software are planned or already under development, VMS represents a powerful and user-friendly tool for both computational and experimentally oriented spectroscopists.

1.1 Introduction

Spectroscopic techniques provide a wealth of qualitative and quantitative information on the chemical and physical-chemical properties of molecular systems in a variety of environments. Nowadays, sophisticated experimental techniques, mainly based on vibrational, electronic, and resonance spectroscopies, allow studies under various environmental conditions and in a noninvasive fashion [1,2]. Particularly effective strategies are obtained when different spectroscopic techniques are combined together and further supported and/or integrated by computational approaches. Indeed, not only the spectral analysis is seldom straightforward, but also molecular spectra do not provide direct information on molecular structures, properties, and dynamics [3,4]. The challenges can be posed by the intrinsic properties and complexity of the system and/or caused by thermal or environmental effects, whose specific roles are not easy to separate and evaluate. In such a context, computational spectroscopy is undoubtedly a powerful and reliable tool to unravel the different contributions to the spectroscopic signal and understand the underlying physical phenomena [5,6]. However, direct vis-à-vis comparisons between experimental and computed spectroscopic data are still far from being standard. To fill this lack, a virtual multifrequency spectrometer () (http://dreamslab.sns.it/vms/) has been implemented with the aim of providing a user-friendly access to the latest developments of computational spectroscopy, also to nonspecialists [7-11]. As it will be better explained in the following section, VMS integrates state-of-the-art computational implementations of different spectroscopies with a powerful graphical user interface () [12], which offers an invaluable aid in preorganizing and displaying the computed spectroscopic information. For the sake of clarity, it should be noted that several codes incorporate implementation of spectroscopic properties at different levels of theory together with graphic engines. However, none of these tools offer the characteristics that should be considered mandatory for state-of-the-art computational spectroscopy (e.g. rigorous treatment of anharmonicity, vibronic contributions, etc.) and/or for flexible user-friendly graphical tools. In particular, it should emphasize the uniqueness of VMS in incorporating both general utilities needed by experimentally oriented scientists (e.g. conversion of theoretical quantities to experimental observables, manipulation of several spectra at the same time, etc.) and advanced tools for theoreticians and developers (e.g. resonance Raman [] spectra).

The aim of the present chapter is to provide an overview of the VMS software, thus focusing on its peculiarities and unique features. The chapter is organized as follows. In the following section, a brief summary of the general machinery of the VMS program and of the main technical aspects will be provided. This will be followed by a short introduction of the theoretical background for the selected spectroscopies (e.g. rotational, vibrational, vibronic, and magnetic) and of the corresponding quantum chemical () requirements. Then, the current status of VMS will be presented in some detail with specific reference to rotational, vibrational, vibronic, and magnetic spectroscopy. Finally, applications will be illustrated with the help of four case studies, which will allow the capabilities of VMS to be demonstrated. Some general considerations will conclude the chapter.

1.2 The Virtual Multifrequency Spectrometer

VMS is a tool that integrates a wide range of computational and experimental spectroscopic techniques and aims at predicting and analyzing different types of molecular spectra as well as disclosing the static and dynamic physical-chemical information they contain [7]. VMS is mainly composed of two parts, namely, VMS-Comp, which provides access to the latest developments in the field of computational spectroscopy, and VMS-Draw, which provides a powerful GUI for an intuitive interpretation of theoretical outcomes and a direct prediction or comparison to experiment (http://dreamslab.sns.it/vms/) [7].

The spectroscopies supported by VMS are electron spin resonance (), nuclear magnetic resonance (), rotational (microwave []), infrared (), vibrational circular dichroism (), nonresonant Raman (), resonance Raman, Raman optical activity (), resonance Raman optical activity (), electronic one-photon absorption () (i.e. UV-vis) and one-photon emission () (i.e. fluorescence), electronic circular dichroism (), and circularly polarized luminescence ().

1.2.1 The VMS Framework

The framework of the VMS program is graphically shown in Figure 1.1 [7]. The key feature of VMS is to provide a user-friendly access to computational spectroscopy tools also to nonspecialists. VMS integrates a powerful GUI, VMS-Draw, which offers an invaluable aid in the pre- and post-processing stages [12]. This permits a direct way to present the information produced by in vitro and in silico experiments, thus allowing the user to focus the attention on the underlying physical-chemical features without being concerned with technical details. VMS-Draw is interfaced with VMS-Comp [8,9,13], which takes care of QC computations of the required spectroscopic parameters and all high-performance computing () aspects [7,12]. Both VMS-Draw and VMS-Comp modules are either fully embedded with the Gaussian package [] or loosely bound to other suites of QC programs, such as CFOUR [15]. In the last case, general input-output facilities as well as ad hoc scripts that permit effective interactions with other electronic structure codes than Gaussian have been developed or are still under development (see, for example, Ref. [10]). Overall, VMS has access to almost all computational models and to properties that are not yet available in the reference QC Gaussian suite. In addition to the large availability of QC methods and properties, VMS has the unique feature of allowing state-of-the-art computational spectroscopy studies driven by a flexible user-friendly graphical tool that furthermore includes those general utilities needed by experimentally oriented scientists (e.g. manipulations of several spectra at the same time, spectral normalization, etc.) and advanced tools for theoreticians and developers (e.g. resonance Raman spectroscopy). In the following sections, the theoretical background and the QC requirements for quantitative...