New Dimensions in Theoretical Multiphoton Spectroscopy

Modern lasers can produce light with such a high intensity that more than one photon can be absorbed simultaneously, bringing the molecule into an excited state. Such processes have higher focality (that is, that excitations occurs in a smaller volume) than regular absorption of only one photon. These multiphoton processes give rise to exciting applications such as 3-dimensional data storage as well as imaging of biological tissue or other materials. Multiphoton absorption allows low-energy light to induce electronic excitations, which in turn can lead to the breaking of chemical bonds, providing interesting opportunities for use in photodynamic therapy. When we use infrared photons in such processes, we can unravel molecular structure and dynamics of both molecules and complex systems. Infrared photons will lead to excitations in the molecular vibrations (the relative motion of the atoms), and when several of these are excited simultaneously, we will have a unique opportunity to see couplings between different parts of a molecule, for instance whether a molecule is elongated or curled up into a ball (secondary structure). Such methods can also be used to study the structure and dynamics of molecules at the interface of two liquids or between a liquid and a gas. In the project periode, we have in particular been working on the following four aspects of the project: 1) We have completed the implementation of a polarizable embedding (PE) methodology for describing solvent effects on vibrational infrared (IR) and Raman spectra. The polarizable embedding approach provides an atomistic, multiscale and dynamic approach to solvent effects on vibrational spectra. The dynamics allows to sample the conformational space, including solvent-solute configurations arising for instance due to hydrogen bonding. This will lead to a more realistic and detailed simulation of the vibrational spectra, allowing for a more detailed comparison with experiment, and also allow theory to provide an analysis of features in the experimentally observed spectra. We have applied the methodology developed to acetone in different solvents. The results have illustrated that this atomistic PE approach is capable of describing effects that do not follow from the polarity of the solvent effects arising from hydrogen bonding effects in solvents such as water. A manuscript has been submitted for publication. 2) Our polarizable embedding model for the calculation of vibrational spectra has also been extended to include anharmonic effects in vibrational spectra. These effects are important in order to obtain quantitive agreement with experiment (assuming a sufficiently high level of electronic structure theory) and for the prediction of overtones and combination bands, both of which are purely anharmonic effects. The largest (future) impact of these anharmonic effects is in the calculation of multidimensional vibrational spectra, where the most interesting bands are of purely anharmonic origin, while at the same time providing detailed insight into interactions between the solute and the solvent (such as hydrogen bonds). We have applied the methodology developed to a study of acetonitrile in different solvents, and also to endohedral water in C60 (a water molecule within the cavity of the C60 "football" molecule). The latter examples illustrates the applicability of the PE model to more general molecular complexes, and which goes beyond the functionality of the closest alternative approaches for polarizable, atomistic modeling of solvent effects. A manuscript is being finalised and will be submitted for publication. 3) In our work on relativistic studies of X-ray spectroscopy, the main focus in the reporting period has been on applications on the methodology that has been developed. One application has been studies of the chiroptical absorption spectrum of the L-edge (excitations from the spin-orbit split 2p orbitals) for a bitiophene complex, that is, X-ray circular dichroism spectra. Such complexes hold great promise in a wide range of applications involving nonlinear optical processes. This has turned out to be a challenging problem, as the electronic states of the two sulphur atoms are nearly degenerate but of opposite polarization, and thus even minor errors may lead to qualitative changes in the calculated spectra. Through systematic exploration of computational parameters, a computational protocol that ensure reliable results have been established. In parallel to this work, we have worked on the L- and M-edges of a series of organometallic complexes, where good agreement with experiment has been obtained. Manuscripts are being prepared for both these applications. 4) In the project period, we have also continued with the extension of our relativistic code to the study of open-shell molecules (molecules with unpaired electrons). This work is still in progress.

Prosjektet har utviklet metoder (dataprogrammer) som gjør det mulig å modellere en lang rekke flerdimensjonale vibrasjonelle spektroskopier for molekyler i løsning. Dette er nye eksperimentelle teknikker med høy informasjonsverdi, og de metoder som er utviklet i dette prosjektet vil kunne gjøre det mulig å øke anvendelsesområdet og å hente ut mer informasjon fra eksperimentelle spektre. Prosjektet har også utviklet nye dataprogrammer som vil muliggjøre ny innsikt i forholdet mellom molekyler struktur og molekylære egenskaper, og hvordan disse avhenger av omgivelsene (løsningsmidler). Dette vil kunne danne grunnlaget for ny forståelse av struktur-egenskapsforhold. Prosjektet har også utviklet analyseverktøy for å forstå sammenhengen mellom molekylær struktur og ikke-lineære optiske egenskaper. Dette vil kunne gi verdifulle bidrag i utviklingen av nye funksjonelle materialer med spesifikke ikke-lineære optiske egenskaper.

Multiphoton processes are becoming increasingly more widespread due to its potential for a wide range of novel applications, including 3D data storage, multiphoton microscopy, photodynamic cancer therapy and drug delivery. The focality that can be achieved through the use of multiple lasers, and the fact that nonlinear, multidimensional vibrational spectroscopies probes anharmonicites in the vibrations in the molecule, allows for a detailed understanding of coupling of different vibrational modes in complex systems, such as biomolecular systems or interfacial systems. For such complex systems, multiscale models are required in order to get a faithful computational model that accounts for the complexity of the systems being studied. Computationally, these spectroscopies are difficult to calculate due to the fact that they probe the anharmonic parts of the molecular vibrations. This requires that efficient computational tools for calculating high-order geometric derivatives of energies and (hyper)-polarizabilities are available. The goal of this project is to develop analytic methods for evaluating a wide range of multidimensional vibrational spectroscopies based on recursive techniques applied to the vibrational sum-over-states expression. This is a highly non-trivial problem for the general case. The necessary geometrical derivatives of the energies and (hyper)polarizabilities will be obtained by extending our recently developed open-ended quasi-energy derivative theory to the inclusion of medium effects from polarizable embedding and polarizable density embedding models. We will use this methodology to study existing and new, novel multidimensional vibrational spectroscopies, with a particular focus on the potentials these methods hold for understanding interactions in biomolecular systems (hydrogen bonding, secondary structure), as well as surface-specific spectroscopies. Recent novel experimental developments such as 2D-Raman will be studied computationally.