Search

Machine learning (ML) has shown to advance the research field of quantum chemistry in almost any possible direction and has also recently been applied to investigate the multifaceted photochemistry of molecules. In this paper, we pursue two goals: (i) We show how ML can be used to model permanent dipole moments for excited states and transition dipole moments by adapting the charge model of Gastegger et al. [Chem. Sci. 8, 6924–6935 (2017)], which was originally proposed for the permanent dipole moment vector of the electronic ground state. (ii) We investigate the transferability of our excited-state ML models in chemical space, i.e., whether an ML model can predict the properties of molecules that it has never been trained on and whether it can learn the different excited states of two molecules simultaneously. To this aim, we employ and extend our previously reported SchNarc approach for excited-state ML. We calculate UV absorption spectra from excited-state energies and transition dipole moments as well as electrostatic potentials from latent charges inferred by the ML model of the permanent dipole moment vectors. We train our ML models on C H 2 N H 2 + and C2H4, while predictions are carried out for these molecules and additionally for CHNH2, CH2NH, and C 2 H 5 + . The results indicate that transferability is possible for the excited states. [ABSTRACT FROM AUTHOR]

We introduce weighted atom-centered symmetry functions (wACSFs) as descriptors of a chemical system’s geometry for use in the prediction of chemical properties such as enthalpies or potential energies via machine learning. The wACSFs are based on conventional atom-centered symmetry functions (ACSFs) but overcome the undesirable scaling of the latter with an increasing number of different elements in a chemical system. The performance of these two descriptors is compared using them as inputs in high-dimensional neural network potentials (HDNNPs), employing the molecular structures and associated enthalpies of the 133 855 molecules containing up to five different elements reported in the QM9 database as reference data. A substantially smaller number of wACSFs than ACSFs is needed to obtain a comparable spatial resolution of the molecular structures. At the same time, this smaller set of wACSFs leads to a significantly better generalization performance in the machine learning potential than the large set of conventional ACSFs. Furthermore, we show that the intrinsic parameters of the descriptors can in principle be optimized with a genetic algorithm in a highly automated manner. For the wACSFs employed here, we find however that using a simple empirical parametrization scheme is sufficient in order to obtain HDNNPs with high accuracy. [ABSTRACT FROM AUTHOR]

We regard the rovibrational wave packet dynamics of NaI in a static electric field after femtosecond excitation to its first electronically excited state. The following quasibound nuclear wave packet motion is accompanied by a bonding situation changing from covalent to ionic. At times when the charge separation is present, i.e., when the bond-length is large, a strong dipole moment exists and rotational excitation takes place. Upon bond contraction, the then covalently bound molecule does not experience the external field. This scenario repeats itself periodically. Thus, the vibrational dynamics causes a situation which is comparable to the interaction of the molecule with a train of pulses where the pulse separation is determined by the vibrational period. © 2004 American Institute of Physics. [ABSTRACT FROM AUTHOR]

We investigate the combined electronic and nuclear motion in a simple model system as proposed by Shin and Metiu [J. Chem. Phys. 102, 9285 (1995)]. A variation of the electron-nuclei interaction energy allows us to study the transition from an adiabatically uncoupled situation to the case where the adiabatic approximation breaks down. Wave-packet calculations illustrate the dynamical changes of electronic and nuclear probability densities for the case of an adiabatic motion and during a nonadiabatic transition. The influence of the coupling on the transport of electron density between different nuclei is characterized with the help of time-dependent Laplacians. [ABSTRACT FROM AUTHOR]

We study alchemical atomic energy partitioning as a method to estimate atomization energies from atomic contributions, which are defined in physically rigorous and general ways through the use of the uniform electron gas as a joint reference. We analyze quantitatively the relation between atomic energies and their local environment using a dataset of 1325 organic molecules. The atomic energies are transferable across various molecules, enabling the prediction of atomization energies with a mean absolute error of 23 kcal/mol, comparable to simple statistical estimates but potentially more robust given their grounding in the physics-based decomposition scheme. A comparative analysis with other decomposition methods highlights its sensitivity to electrostatic variations, underlining its potential as a representation of the environment as well as in studying processes like diffusion in solids characterized by significant electrostatic shifts., (© 2024 Author(s). Published under an exclusive license by AIP Publishing.)

We perform time-resolved ionization spectroscopy measurements of the excited state dynamics of CH 2 I 2 and CH 2 IBr following photoexcitation in the deep UV. The fragment ions produced by ionization with a vacuum-ultraviolet probe pulse are measured with velocity map imaging, and the momentum resolved yields are compared with trajectory surface hopping calculations of the measurement observable. Together with recent time-resolved photoelectron spectroscopy measurements of the same dynamics, these results provide a detailed picture of the coupled electronic and nuclear dynamics involved. Our measurements highlight the non-adiabatic coupling between electronic states, which leads to notable differences in the dissociation dynamics for the two molecules.

We compare the excited state dynamics of diiodomethane (CH 2 I 2 ) and bromoiodomethane (CH 2 BrI) using time resolved photoelectron spectroscopy. A 4.65 eV UV pump pulse launches a dissociative wave packet on excited states of both molecules and the ensuing dynamics are probed via photoionization using a 7.75 eV probe pulse. The resulting photoelectrons are measured with the velocity map imaging technique for each pump-probe delay. Our measurements highlight differences in the dynamics for the two molecules, which are interpreted with high-level ab initio molecular dynamics (trajectory surface hopping) calculations. Our analysis allows us to associate features in the photoelectron spectrum with different portions of the excited state wave packet represented by different trajectories. The excited state dynamics in bromoiodomethane are simple and can be described in terms of direct dissociation along the C-I coordinate, whereas the dynamics in diiodomethane involve internal conversion and motion along multiple dimensions.

Many approaches, which have been developed to express the potential energy of large systems, exploit the locality of the atomic interactions. A prominent example is the fragmentation methods in which the quantum chemical calculations are carried out for overlapping small fragments of a given molecule that are then combined in a second step to yield the system's total energy. Here we compare the accuracy of the systematic molecular fragmentation approach with the performance of high-dimensional neural network (HDNN) potentials introduced by Behler and Parrinello. HDNN potentials are similar in spirit to the fragmentation approach in that the total energy is constructed as a sum of environment-dependent atomic energies, which are derived indirectly from electronic structure calculations. As a benchmark set, we use all-trans alkanes containing up to eleven carbon atoms at the coupled cluster level of theory. These molecules have been chosen because they allow to extrapolate reliable reference energies for very long chains, enabling an assessment of the energies obtained by both methods for alkanes including up to 10 000 carbon atoms. We find that both methods predict high-quality energies with the HDNN potentials yielding smaller errors with respect to the coupled cluster reference.

Full multiple spawning is a formally exact method to describe the excited-state dynamics of molecular systems beyond the Born-Oppenheimer approximation. However, it has been limited until now to the description of radiationless transitions taking place between electronic states with the same spin multiplicity. This Communication presents a generalization of the full and ab initio multiple spawning methods to both internal conversion (mediated by nonadiabatic coupling terms) and intersystem crossing events (triggered by spin-orbit coupling matrix elements) based on a spin-diabatic representation. The results of two numerical applications, a model system and the deactivation of thioformaldehyde, validate the presented formalism and its implementation.

Ground- and excited-state UV photoelectron spectra of thiouracils (2-thiouracil, 4-thiouracil, and 2,4-dithiouracil) have been simulated using multireference configuration interaction calculations and Dyson norms as a measure for the photoionization intensity. Except for a constant shift, the calculated spectrum of 2-thiouracil agrees very well with experiment, while no experimental spectra are available for the two other compounds. For all three molecules, the photoelectron spectra show distinct bands due to ionization of the sulphur and oxygen lone pairs and the pyrimidine π system. The excited-state photoelectron spectra of 2-thiouracil show bands at much lower energies than in the ground state spectrum, allowing to monitor the excited-state population in time-resolved UV photoelectron spectroscopy experiments. However, the results also reveal that single-photon ionization probe schemes alone will not allow monitoring all photodynamic processes existing in 2-thiouracil. Especially, due to overlapping bands of singlet and triplet states the clear observation of intersystem crossing will be hampered.

An efficient perturbational treatment of spin-orbit coupling within the framework of high-level multi-reference techniques has been implemented in the most recent version of the Columbus quantum chemistry package, extending the existing fully variational two-component (2c) multi-reference configuration interaction singles and doubles (MRCISD) method. The proposed scheme follows related implementations of quasi-degenerate perturbation theory (QDPT) model space techniques. Our model space is built either from uncontracted, large-scale scalar relativistic MRCISD wavefunctions or based on the scalar-relativistic solutions of the linear-response-theory-based multi-configurational averaged quadratic coupled cluster method (LRT-MRAQCC). The latter approach allows for a consistent, approximatively size-consistent and size-extensive treatment of spin-orbit coupling. The approach is described in detail and compared to a number of related techniques. The inherent accuracy of the QDPT approach is validated by comparing cuts of the potential energy surfaces of acrolein and its S, Se, and Te analoga with the corresponding data obtained from matching fully variational spin-orbit MRCISD calculations. The conceptual availability of approximate analytic gradients with respect to geometrical displacements is an attractive feature of the 2c-QDPT-MRCISD and 2c-QDPT-LRT-MRAQCC methods for structure optimization and ab inito molecular dynamics simulations.

The importance of triplet states in the photorelaxation dynamics of SO2 is studied by mixed quantum-classical dynamics simulations. Using the SHARC method, standing for Surface Hopping including ARbitrary Couplings, intersystem crossing (ISC) processes caused by spin-orbit coupling are found occurring on an ultrafast time scale (few 100 fs) and thus competing with internal conversion. While in the singlet-only dynamics only oscillatory population transfer between the (1)B1 and (1)A2 states is observed, in the dynamics including singlet and triplet states we find additionally continuous ISC to the (3)B2 state and to a smaller extent to the (3)B1/(3)A2 coupled states. The populations obtained from the dynamics are discussed with respect to the overall nuclear motion and in the light of recent TRPEPICO studies [I. Wilkinson, A. E. Boguslavskiy, J. Mikosch, D. M. Villeneuve, H.-J. Wörner, M. Spanner, S. Patchkovskii, and A. Stolow, "Excited state dynamics in SO2. I. Bound state relaxation studied by time-resolved photoelectron-photoion coincidence spectroscopy," J. Chem. Phys. 140, 204301 (2014)].

A new approach for the calculation of resonance Raman spectra is presented. The new method is based on dynamic polarizabilities from real-time time-dependent density functional theory, and its estimations are compared to two established techniques for the prediction of resonance Raman spectra. These established methods either use dynamic polarizabilities from linear-response time-dependent density functional theory or employ excited-state gradients. The three different ways to calculate resonance Raman spectra are investigated using the example of ortho-nitrophenol. The three methods give very similar results, respectively, for the four different exchange-correlation functionals applied. Thus, the new approach is validated for the calculation of resonance Raman intensities and advantages as well as disadvantages are discussed.

We focus on the applications of shaped-dump laser pulses in the femtosecond regime. Calculations on a model system show that a wealth of information on molecular properties can be obtained from spectroscopy with such pulses. Systematic parameter scans are presented for linearly chirped pulses and for colored double pulses, revealing the dynamical properties of the system. For characterization of the system, quantum control fitness landscapes prove to be a powerful tool.

Local control theory (LCT) is employed to achieve molecular photoassociation with shaped laser pulses. Within LCT, the control fields are constructed from the response of the system to the perturbation which makes them accessible to a straightforward interpretation. This is shown regarding the ground-state collision of H+F and H+I atoms. Different objectives are defined, which aim at the formation of vibrational cold or hot associated molecules, respectively. Results are presented for s-wave scattering, where the rotational degree of freedom is ignored and also for full scale calculations including rotations, in order to describe more realistic conditions.

The driven wave-packet dynamics in potentials exhibiting several potential wells is investigated. Therefore, local-control strategies are employed where the control field is constructed from the system's dynamics at any instant of time. It is shown that particles can be moved successively between various potential minima. Furthermore, results presented indicate that the intuitive local-control scheme allows for the initiation of a clockwise or counterclockwise rotational motion of a model molecular motor.

Local control theory, where the instantaneous response of a system to an external field determines the control field, is employed for the purpose of inducing molecular fragmentation processes via infrared excitation. In particular, the effects of the orientational motion are investigated and compared with the idealized case of a frozen rotation. It is shown that the rotational degree of freedom is crucial for the applicability of the employed local control algorithm. The addition of an additional static electric field which induces a molecular preorientation offers an efficient way for the local control. In particular, with increasing static field strength, the fragmentation yield approaches unity so that the idealized rotationless case is recovered. Numerical results are presented for the NaI molecule.