An on-the-fly deep neural network for simulating time-resolved spectroscopy: predicting the ultrafast ring opening dynamics of 1,2-dithiane.

Revolutionary developments in ultrafast light source technology are enabling experimental spectroscopists to probe the structural dynamics of molecules and materials on the femtosecond timescale. The capacity to investigate ultrafast processes afforded by these resources accordingly inspires theoreticians to carry out high-level simulations which facilitate the interpretation of the underlying dynamics probed during these ultrafast experiments. In this Article, we implement a deep neural network (DNN) to convert excited-state molecular dynamics simulations into time-resolved spectroscopic signals. Our DNN is trained on-the-fly from first-principles theoretical data obtained from a set of time-evolving molecular dynamics. The train-test process iterates for each time-step of the dynamics data until the network can predict spectra with sufficient accuracy to replace the computationally intensive quantum chemistry calculations required to produce them, at which point it simulates the time-resolved spectra for longer timescales. The potential of this approach is demonstrated by probing dynamics of the ring opening of 1,2-dithiane using sulphur K-edge X-ray absorption spectroscopy. The benefits of this strategy will be more markedly apparent for simulations of larger systems which will exhibit a more notable computational burden, making this approach applicable to the study of a diverse range of complex chemical dynamics.

Photostability of the deprotonated forms of the UV filters homosalate and octyl salicylate: molecular dissociation versus electron detachment following UV excitation.

While common molecular anions show a strong propensity to undergo electron detachment upon UV excitation, this process often occurs in competition with molecular ion dissociation. The factors that affect the balance between these two major possible decay pathways have not been well understood to date. Laser photodissociation spectroscopy of the deprotonated forms of the UV filter molecules, Homosalate (HS) and Octyl Salicylate (OS), i.e. [HS - H]- and [OS - H]-, was used to acquire gas-phase UV absorption spectra for [HS - H]- and [OS - H]-via photodepletion from 3.0-5.8 eV. No photofragmentation (i.e. dissociation of the ionic molecular framework) was observed for either [HS - H]- and [OS - H]- following photoexcitation, revealing that electron loss entirely dominates the electronic decay pathways for these systems. High-level quantum chemical calculations were used to map out the excited states associated with [HS - H]- and [OS - H]-, revealing that the minimum-energy crossing points (MECPs) between the S1 and S0 states are located in elevated regions of the potential energy surface, making internal conversion unlikely. These results are consistent with our experimental observation that electron detachment out-competes hot ground state molecular fragmentation. More generally, our results reveal that the competition between molecular dissociation and electron detachment following anion photoexcitation can be determined by the magnitude of the energy gap between the excitation energy and the MECPs, rather than being a simple function of whether the excitation energy lies above the anion's vertical detachment energy.

Beyond structural insight: a deep neural network for the prediction of Pt L2/3-edge X-ray absorption spectra.

X-ray absorption spectroscopy at the L2/3 edge can be used to obtain detailed information about the local electronic and geometric structure of transition metal complexes. By virtue of the dipole selection rules, the transition metal L2/3 edge usually exhibits two distinct spectral regions: (i) the "white line", which is dominated by bound electronic transitions from metal-centred 2p orbitals into unoccupied orbitals with d character; the intensity and shape of this band consequently reflects the d density of states (d-DOS), which is strongly modulated by mixing with ligand orbitals involved in chemical bonding, and (ii) the post-edge, where oscillations encode the local geometric structure around the X-ray absorption site. In this Article, we extend our recently-developed XANESNET deep neural network (DNN) beyond the K-edge to predict X-ray absorption spectra at the Pt L2/3 edge. We demonstrate that XANESNET is able to predict Pt L2/3 -edge X-ray absorption spectra, including both the parts containing electronic and geometric structural information. The performance of our DNN in practical situations is demonstrated by application to two Pt complexes, and by simulating the transient spectrum of a photoexcited dimeric Pt complex. Our discussion includes an analysis of the feature importance in our DNN which demonstrates the role of key features and assists with interpreting the performance of the network.

Measurement of the Population of Electrosprayed Deprotomers of Coumaric Acids Using UV-Vis Laser Photodissociation Spectroscopy.

The measurement of deprotonation sites in multifunctional molecules following electrospray ionization is important to better inform a wide range of spectroscopic and photophysical studies that use electrospray to prepare molecular species for study in the gas phase. We demonstrate that low-resolution UV-vis laser photodissociation spectroscopy can be applied in situ to identify the deprotomers of three coumaric acids, trans-para-coumaric acid (CMA), trans-caffeic acid (CA), and trans-ferulic acid (FA), formed via electrospray. Electronic absorption spectra of the deprotonated coumaric acids are recorded via photodepletion and photofragmentation following electrospray from solutions of ethanol and acetonitrile. By comparing the experimental spectra to wave function theory calculations, we are able to confirm the presence of phenoxide and carboxylate deprotomers upon electrospray for all three coumaric acids, when sprayed from both protic and aprotic solvents. Ratios of the phenoxide:carboxylate deprotomers are obtained by generating summed theoretical absorption spectra that reproduce the experimental spectra. We find that choice of electrospray solvent has little effect on the ratio of deprotomers obtained for deprotonated CMA and CA but has a greater impact for FA. Our results are in excellent agreement with previous work conducted on deprotonated CMA using IR spectroscopy and demonstrate that UV photodissociation spectroscopy of electrosprayed ions has potential as a diagnostic tool for identifying deprotomeric species.

Enhancing the analysis of disorder in X-ray absorption spectra: application of deep neural networks to T-jump-X-ray probe experiments.

Many chemical and biological reactions, including ligand exchange processes, require thermal energy for the reactants to overcome a transition barrier and reach the product state. Temperature-jump (T-jump) spectroscopy uses a near-infrared (NIR) pulse to rapidly heat a sample, offering an approach for triggering these processes and directly accessing thermally-activated pathways. However, thermal activation inherently increases the disorder of the system under study and, as a consequence, can make quantitative interpretations of structural changes challenging. In this Article, we optimise a deep neural network (DNN) for the instantaneous prediction of Co K-edge X-ray absorption near-edge structure (XANES) spectra. We apply our DNN to analyse T-jump pump/X-ray probe data pertaining to the ligand exchange processes and solvation dynamics of Co2+ in chlorinated aqueous solution. Our analysis is greatly facilitated by machine learning, as our DNN is able to predict quickly and cost-effectively the XANES spectra of thousands of geometric configurations sampled from ab initio molecular dynamics (MD) using nothing more than the local geometric environment around the X-ray absorption site. We identify directly the structural changes following the T-jump, which are dominated by sample heating and a commensurate increase in the Debye-Waller factor.

Linking Electronic Relaxation Dynamics and Ionic Photofragmentation Patterns for the Deprotonated UV Filter Benzophenone-4.

Understanding how deprotonation impacts the photophysics of UV filters is critical to better characterize how they behave in key alkaline environments including surface waters and coral reefs. Using anion photodissociation spectroscopy, we have measured the intrinsic absorption electronic spectroscopy (400-214 nm) and numerous accompanying ionic photofragmentation pathways of the benzophenone-4 anion ([BP4-H]-). Relative ion yield plots reveal the locations of the bright S1 and S3 excited states. For the first time for an ionic UV filter, ab initio potential energy surfaces are presented to provide new insight into how the photofragment identity maps the relaxation pathways. These calculations reveal that [BP4-H]- undergoes excited-state decay consistent with a statistical fragmentation process where the anion breaks down on the ground state after nonradiative relaxation. The broader relevance of the results in providing a basis for interpreting the relaxation dynamics of a wide range of gas-phase ionic systems is discussed.

Ultrafast excited-state dynamics of promising nucleobase ancestor 2,4,6-triaminopyrimidine.

The ultrafast excited-state dynamics of 2,4,6-triaminopyrimidine - thought to be a promising candidate for a proto-RNA nucleobase - have been investigated via static multireference quantum-chemical calculations and mixed-quantum-classical/trajectory surface-hopping dynamics with a focus on the lowest-lying electronic states of the singlet manifold and with a view towards understanding the UV(C)/UV(B) photostability of the molecule. Ultrafast internal conversion channels have been identified that connect the lowest-lying ππ* electronically-excited state of 2,4,6-triaminopyrimidine with the ground electronic state, and non-radiative decay has been observed to take place on the picosecond timescale via a ππ* out-of-plane NH2 ("oop-NH2") minimum-energy crossing point. The short excited-state lifetime is competitive with the excited-state lifetimes of the canonical pyrimidine nucleobases, affirming the promise of 2,4,6-triaminopyrimidine as an ancestor. Evidence for energy-dependent excited-state dynamics is presented, and the open question of intersystem crossing is discussed speculatively.

Structure retrieval in liquid-phase electron scattering.

Electron scattering on liquid samples has been enabled recently by the development of ultrathin liquid sheet technologies. The data treatment of liquid-phase electron scattering has been mostly reliant on methodologies developed for gas electron diffraction, in which theoretical inputs and empirical fittings are often needed to account for the atomic form factor and remove the inelastic scattering background. In this work, we present an alternative data treatment method that is able to retrieve the radial distribution of all the charged particle pairs without the need of either theoretical inputs or empirical fittings. The merits of this new method are illustrated through the retrieval of real-space molecular structure from experimental electron scattering patterns of liquid water, carbon tetrachloride, chloroform, and dichloromethane.

The Role of Structural Representation in the Performance of a Deep Neural Network for X-Ray Spectroscopy.

An important consideration when developing a deep neural network (DNN) for the prediction of molecular properties is the representation of the chemical space. Herein we explore the effect of the representation on the performance of our DNN engineered to predict Fe K-edge X-ray absorption near-edge structure (XANES) spectra, and address the question: How important is the choice of representation for the local environment around an arbitrary Fe absorption site? Using two popular representations of chemical space-the Coulomb matrix (CM) and pair-distribution/radial distribution curve (RDC)-we investigate the effect that the choice of representation has on the performance of our DNN. While CM and RDC featurisation are demonstrably robust descriptors, it is possible to obtain a smaller mean squared error (MSE) between the target and estimated XANES spectra when using RDC featurisation, and converge to this state a) faster and b) using fewer data samples. This is advantageous for future extension of our DNN to other X-ray absorption edges, and for reoptimisation of our DNN to reproduce results from higher levels of theory. In the latter case, dataset sizes will be limited more strongly by the resource-intensive nature of the underlying theoretical calculations.

Volatile and Thermally Stable Polymeric Tin Trifluoroacetates.

Tin trifluoroacetates are effective vapor phase single-source precursors for F-doped SnO2, but their structures have been poorly understood for decades. Here we undertook a comprehensive structural analysis of these compounds in both the solid and gas phases through a combined single-crystal X-ray crystallography, gas phase electron diffraction, and density functional theory investigation. Tin(II) bis(trifluoroacetate) (1) thermally decomposes into a 1:1 mixture of 1 and ditin(II) µ-oxybis(µ-trifluoroacetate) (2) during sublimation, which then polymerize into hexatin(II)-di-µ3-oxyoctakis(µ-trifluoroacetate) (3) upon solidification. Reversible depolymerization occurred readily upon heating, making 3 a useful vapor phase precursor itself. Tin(IV) tetrakis(trifluoroacetate) (5) was also found to be polymeric in the solid state, but it evaporated as a monomer over 130 °C lower than 3. This counterintuitive improvement in volatility by polymerization was possibly due to the large entropy change during sublimation, which offers a strategic new design feature for vapor phase deposition precursors.

Direct Experimental Observation of in situ Dehydrogenation of an Amine-Borane System Using Gas Electron Diffraction.

In situ dehydrogenation of azetidine-BH3, which is a candidate for hydrogen storage, was observed with the parent and dehydrogenated analogue subjected to rigorous structural and thermochemical investigations. The structural analyses utilized gas electron diffraction supported by high-level quantum calculations, while the pathway for the unimolecular hydrogen release reaction in the absence and presence of BH3 as a bifunctional catalyst was predicted at the CBS-QB3 level. The catalyzed dehydrogenation pathway has a barrier lower than the predicted B-N bond dissociation energy, hence favoring the dehydrogenation process over the dissociation of the complex. The predicted enthalpy of dehydrogenation at the CCSD(T)/CBS level indicates that mild reaction conditions would be required for hydrogen release and that the compound is closer to thermoneutral than linear amine boranes. The entropy and free energy change for the dehydrogenation process show that the reaction is exergonic, energetically feasible, and will proceed spontaneously toward hydrogen release, all of which are important factors for hydrogen storage.

Structure of 4-(Dimethylamino)benzonitrile Using Gas Electron Diffraction: A New Lease of Life for the Only Gas Electron Diffractometer in the U.K. .

The continued demand for gas-phase molecular structures has led to the recommissioning of a gas electron diffractometer, formerly housed at the University of Reading. The gas electron diffractometer, now the only one of its kind in the U.K., is currently housed at the University of York, where it is now used routinely to determine directly structures of isolated molecules in the gas phase. The instrument has been fitted with an air-heated nozzle assembly to increase the range of molecules accessible to study in the gas phase; the efficacy of this assembly is demonstrated in this article via the determination of the gas-phase structure of 4-(dimethylamino)benzonitrile (DMABN) at high temperature. A series of complementary theoretical calculations using the B2PLYP DFT functional of Grimme et al. with correlation-consistent basis sets of double, triple, and quadruple-ζ quality are also presented. The agreement between the experimental and theoretical structural parameters attests to the accuracy of the applied theoretical calculations and of our gas-phase structural solution.