Real-time time-dependent density functional theory (RT-TDDFT) has become a widely used tool for the simulation of optical response. Apart from its favorable scaling and its capacity to resolve the whole frequency dependent linear response functions in one go, RT-TDDFT provides a non-perturbative framework to apply finite electro-magnetic fields, mimicking actual experiments. In this contribution the versatility of RT-TDDFT is illustrated by showcasing its ability to simulate UV-VIS absorption, electric circular dichroism (ECD), (resonance) Raman and (resonance) Raman optical activity (ROA) spectra [1], the latter two within the short time approximation. For the formulation of the spectroscopic response tensors a unified formalism in terms of linear response propagators is applied, allowing insights into how the perturbation and response operators are distinguishable in the practical real-time linear response protocol. Special emphasis is on the choices of gauge, specifically length-, velocity- and symmetric gauges, and the coupling of the electro-magnetic fields to the non-local part of pseudo potentials, a proper handling of which proves to be vital for an adequate description of the chiral spectroscopic responses, ECD and ROA [2]. For UV-VIS absorption and Raman spectroscopy results for non-periodic and periodic simulation cells are presented, drawing on the velocity gauge and the modern theory of polarization. The results were obtained with a modified development version of the CP2K package. These developments allow applications beyond single molecules, e. g. the study of liquids and interfaces [3].  
  [1] J. Mattiat, S. Luber, J. Chem. Phys. 151, 234110, (2019) [2] J. Mattiat, S. Luber, J. Chem. Theory Comput. 18, 9, 5513–5526, (2022) [3] J. Mattiat, S. Luber, J. Chem. Theory Comput. 17, 1, 344–356 (2021)

Jahn-Teller active molecules demonstrate significant vibronic and spin-orbit couplings which has always posed a computational challenge in quantum chemistry. The presence of a conical intersection in the adiabatic potential energy surfaces makes getting a reliable quantum chemistry calculation of these terms challenging due to breakdown of Born-Oppenheimer (BO) approximation. Calculating experimentally observable parameters for rotationally resolved spectra of such molecules involves three basic steps, a) generating a potential energy surface in quasi-diabatic basis, b) transformation of the potential energy surface from Cartesian to cylindrical coordinates c) diagonalizing the spin-vibronic Hamiltonian. One of the major obstacles in these calculations is solving the eigenvalue problem in quasi-diabatic basis due to the size of matrices that need to be diagonalized. In this talk we talk about pushing the boundaries for such calculations with our program package (SOCJT3). We have implemented new algorithms that diagonalize huge matrices and include coupling terms up to quartic order. This is further utilized to calculate experimentally observable rotational parameters. The primary objective of our work is to generate a simulation to understand and characterize the previously unanalyzed rotationally resolved spectra of open-shell radicals.

As described in the previous talk, if one is analyzing a spectrum involving rotational and fine structure, the cylindrical representation of the potential energy matrix in the quasi-diabatic basis, V_cyl, is more convenient due to the simple form of rovibronic Hamiltonian. It also facilitates the identification of the symmetry of the vibronic basis functions.Bunker, P.R., and Jensen, P. 2006. Molecular symmetry and spectroscopy. 2nd ed. NRC Research Press, Ottawa, Canada, p. 290 ff.n the other hand, ab initio parameterization of the vibronic Hamiltonian is typically performed in the Cartesian representation, V_cart. The two matrix representations for a molecule with a simple, linear Jahn-Teller E ×e effect can be written as                 where Q_± = Q_a ±i Q_b are the normal modes. Here the corresponding electronic basis sets are related as Φ_± = \frac1√2 (Φ_a ±iΦ_b). Transforming V_cart into V_cyl through coordinate substitution together with the basis set transformation becomes tedious for higher-order expansions, especially considering multimode and multistate problems. It is therefore desirable to develop a general procedure that can be used for the transformation of operators from the Cartesian form to the cylindrical one. In this talk, we will show that the potential energy operator represented in a tensor form can be naturally transformed from Cartesian to cylindrical representation through a series of tensor-matrix multiplications, provided that the two matrices transforming the vibrational normal coordinates and the electronic basis set are known. In general, this method can be used to transform any quantum mechanical operator matrix. We demonstrate the effectiveness of this method with calculations for NO₃ and CH₃O.

---

Footnotes:

Bunker, P.R., and Jensen, P. 2006. Molecular symmetry and spectroscopy. 2nd ed. NRC Research Press, Ottawa, Canada, p. 290 ff.O

Absorption spectroscopy has long been known as a technique for making molecular concentration measurements and has received enhanced visibility in recent years with the advent of new techniques, like cavity ring-down spectroscopy, that have increased its sensitivity. To apply the method, it is necessary to have a known molecular absorption cross-section for the species of interest, which typically is obtained by measurements of a standard sample of known concentration. However, this method fails if the species is highly reactive, and indirect means for attaining the cross section must be employed. The HO₂ and alkyl peroxy radicals are examples of reactive species for which absorption cross-sections have been reported. This work explores and describes for these peroxy radicals the details of an alternative approach for obtaining these cross sections using quantum chemistry methods for the calculation of the transition dipole moment upon whose square the cross section depends. Likewise, details are given for obtaining the transition moment from the experimentally measured cross sections of individual rovibronic lines in the near-IR, Ã-~X electronic spectrum of HO₂ and the peaks of the rotational contours in the corresponding electronic transitions for the alkyl (methyl, ethyl, and acetyl) peroxy radicals. In the case of the alkyl peroxy radicals, good agreement for the transition moments, approximately 20%, is found between the two methods. However, rather surprisingly, the agreement is significantly poorer, approximately 40%, for the HO₂ radical. Possible reasons for this disagreement are discussed.

Assigning and interpreting the spectroscopy of large molecules and clusters often requires a manual trial-and-error approach. Here, we present our efforts to automate spectroscopic assignments through a data-driven approach. We will consider two cases: vibrational spectra of cold clusters and simulated electronic spectra of conjugated molecules. Our method first assumes a local mode form of the Hamiltonian and then performs an active search through the space of physically reasonable couplings that could be present in the system (e.g., anharmonicities for vibrations and electronic couplings for conjugated molecules). By finding the Hamiltonian(s) that can best fit the spectra, the assignment can be automatically performed. This approach employs a Bayesian-optimization-derived algorithm as its driver and does not require a large volume of initial training data. In this talk, we focus on applying the method to model problems, higher-level calculations of benchmark systems, and real experimental spectra found in the literature. Finally, we will present our efforts on modeling the framework’s robustness to noisy input data.

Luminescent organic molecules have been widely applied in optoelectronics and biological research. First-principles time-dependent density functional theory (TDDFT) and machine learning methods have demonstrated great success in the predictions of optical properties of large organic molecules. However, the systematic error, large time cost, and narrow range of predictable properties of TDDFT hinder its applications in high-throughput screening of real-life systems. On the other side, statistics-based methodologies have the advantages of high accuracy and low costs. While the generalizability of the models and synthesizability of the molecules still pose challenges. Herein, we developed a ML model that implemented semi-empirical quantum chemical properties to accurately predict the absorption frequencies, emission frequencies, and photoluminescence quantum yield (PLQY) of organic molecules. Based on the evaluation on chromophore families and chromophore-solvent pairs, we illustrated that our extension of the semi-empirical quantum chemical properties remarkably improved the accuracy and generalizability of the model with only a margin increase of computational costs. Meanwhile, tree-based algorithms outperformed neural networks and managed to reach mean absolute errors (MAEs) as low as 0.061 eV for absorption frequencies, 0.065 eV for emission frequencies, and 0.10 for PLQY. Tested on another database containing 96 million compounds with semi-empirical calculations, our model exhibited great success in the predictions of optical properties for all synthesizable molecules at very low computational costs, and thus substantially promote the discovery of potential optical materials at a large scale.

In the past few decades, metal nanoparticles have attracted a lot of interest due to their ability to confine light in nanometric volumes [1] resulting in interesting applications of plasmonic nanoparticles such as optical sensors [2]. Nanometer-sized metallic particles or structures can strongly absorb and scatter light due to their ability to support surface plasmon resonances - coherent oscillations of surface conduction electrons in response to the electric field of light [3]. In my talk, I will briefly present the theory from a classical electrodynamics perspective [4] and discuss my ongoing research on (a) plasmonic enhancement and circular dichroism response of 2D metamaterials (nanoslits arrays) with circularly-polarized light, and (b) optical response of 1D and 2D arrays of gold, silver and aluminium particles with linearly polarized light. Numerical modelling will be based on finite-difference time-domain [5] and coupled-dipole methods [6]. References: [1] E. Ozbay, Science 311, 189 (2006). [2] K. A. Willets and R. P. Van Duyne, Annu. Rev. Phys. Chem. 58, 267 (2007). [3] K. L. Kelly, E. Coronado, L. L. Zhao and G. C. Schatz, J. Phys. Chem. B 107, 668 (2003). [4] J. Zhao, A. O. Pinchuk, J. M. Mcmahon, S. Li, L. K. Ausman, A. L. Atkinson and G. C. Schatz, Acc. Chem. Res. 41, 1710 (2008). [5] K. S. Yee, IEEE Trans. Antennas Propag. 14, 302 (1966). [6] S. Zou, N. Janel and G. C. Schatz, J. Chem. Phys. 120, 10871 (2004).

Organic fluorescent molecules with desired optical properties attracted great attention, while the rational design was hindered by unclear structure properties relationship and the lack of rapid/affordable prediction methods. With the introduction of statistics-based methods in the prediction of photophysical properties for organic dyes, reverse design of fluorophores without traversing chemical space is still challenged by the features used for current methodologies. In this work, we construct a self-referencing embedded strings (SELFIES)-based variational autoencoder (VAE) and a prediction model, which uses the latent space as the input, for the organic fluorophores, in the absence of joint training. The VAE can reproduce the structure of midsize organic dyes with acceptable accuracy. A tree-based prediction model based on Gradient Boosted Regression Trees (GBRT) can estimate the optical properties of organic dyes with a MAE 0.134 eV for emission energy and an accuracy of 0.81 for photoluminescence quantum yield (PLQY), which is comparable with the state-of-the-art quantum-mechanical based approach, time-dependent density-functional theory (TD-DFT). The feasibility of our approach in reverse design is proved by preliminary attempts at skeleton optimization and validated by first-principles calculations. New experimental synthesized molecules demonstrated the accuracy of our prediction model. Meanwhile, due to the continuous values in the latent space, this VAE-based methodology makes gradient optimization become possible for large organic materials. Combined, our statistical learning methodology opens a new venue for the design of organic fluorophore, can also be extended to the field of organic solar cell (photo conversion efficiency, PCE) and organic field-effect transistor (conductivity).

Ammi Visnaga has been reported to possess various biological activities such as anti-inflammatory, antioxidant, antifungal, antidiabetic, cytotoxic, antibacterial effects etc. In the present study, nineteen selected constituents of Ammi Visnaga such as osthenol, visnadin, dihydrosamidin, samidin, apiumetin, celereoin, visnagin, khellin, visamminol, cimifugin, acacetin, quercetin, isoformonometin, visnaginone, khellinone etc. were docked to agaricus bisporus tyrosinase (PDB ID:2Y9X), priestia megaterium tyrosinase (PDB ID: 3NQ1) and homo sapiens tyrosinase (PDB ID: 5M8M) to investigate the potential anti-melanogenic activity. All compounds showed higher docking scores and binding free energy than cognate ligand tropolone for 2Y9X. However, A. visnaga constituents such as coumarin (osthenol), pyranocoumarins (visnadin, dihydrosamidin, samidin), furanocoumarins (apiumetin, celeroin), visamminol (furanochromones), flavonoids (cimifugin, quercetin) have higher binding energy than kojic acid. Kojic acid is a cognate ligand of human tyrosinase (PDB ID: 5M8M) and bacillus megaterium tyrosinase (PDB ID: 3NQ1). Apiumetin (L5) has the highest binding energy of all compounds in three tyrosinase enzymes than cognate ligands. Molecular dynamic analysis shows that Apiumetin (L5) is more stable than the cognate ligand in the binding pocket. ADME analyses calculated by the QikProp program show that all compounds obey Lipinski’s rule of five without violations. Schrödinger module was used for molecular docking (IFD) and molecular dynamic (Desmond) analyses. The binding free energies of the compounds were calculated by MM/GBSA approach.