Analytical gradients of the state-average complete active space self-consistent field method with density fitting

Analytical SA-HCISCF Nuclear Gradients from Spin-Adapted Heat-Bath Configuration Interaction

This work reports an implementation of the analytical nuclear gradients and nonadiabatic couplings with state-averaged SCF wave functions from a spin-pure selected configuration interaction (SCI) method. At the core of the implementation lies the evaluation of the Lagrange multipliers required for the variational calculation of the nuclear gradient. Using the same code infrastructure, we developed a fully CI-coupled second-order orbital optimization method. Both the calculation of the nuclear gradient and the second-order orbital optimization make use of density fitting in order to accelerate the calculation of the two-electron integrals. We demonstrate the use of analytical nuclear gradients in excited-state geometry optimizations for conjugated molecules. In addition, the first triplet excited-state geometry of a transition-metal catalyst, Fe(PDI), was optimized with up to 30 orbitals in the active space. Our results outline the capabilities of the implemented methods as well as directions for future work.

Charge-induced isomerization in alkyl imine molecular motors: a reduced energy barrier approach

Molecular motors offer promising applications in the fields of nanodevices and biological systems, as the accurate control of directional rotation at the molecular scale holds great potential. In this context, it is highly relevant to study a new class of molecular motors that can undergo isomerization. Since the first report of the chiral N-alkyl imine-based motors, most investigations have focused on the unidirectional rotation process induced by light and heat. However, this work explores an alternative mechanism - the electron-induced stimulating mechanism of the molecular motor. We theoretically investigate how charge injection and extraction can influence molecular rotation. The rotation occurs around the central axle, which is measured as the torsion angle between the rotor and the stator fragments of the molecule against the C[double bond, length as m-dash]N double bond. Our computational study reveals that the introduction of charge reduces the energy barrier, facilitating more favourable molecular rotation than in the neutral singlet state. The charged molecule in a quartet spin state can rotate internally, while that in the doublet state cannot. Our findings provide a molecular scale understanding of the reaction pathways and highlight the significant role of charge in promoting the isomerization and rotational behaviour of the molecular motor.

Finite-field Cholesky decomposed coupled-cluster techniques (ff-CD-CC): theory and application to pressure broadening of Mg by a He atmosphere and a strong magnetic field

For the interpretation of spectra of magnetic stellar objects such as magnetic white dwarfs (WDs), highly accurate quantum chemical predictions for atoms and molecules in finite magnetic field are required. Especially the accurate description of electronically excited states and their properties requires established methods such as those from coupled-cluster (CC) theory. However, respective calculations are computationally challenging even for medium-sized systems. Cholesky decomposition (CD) techniques may be used to alleviate memory bottlenecks. In finite magnetic field computations, the latter are increased due to the reduction of permutational symmetry within the electron-repulsion-integrals (ERIs) as well as the need for complex-valued data types. CD enables a memory-efficient, approximate description of the ERIs with rigorous error control and thus the treatment of larger systems at the CC level becomes feasible. In order to treat molecules in a finite magnetic field, we present in this work the working equations of the left and right-hand side equations for finite field (ff)-EOM-CD-CCSD for various EOM flavours as well as for the approximate ff-EOM-CD-CC2 method. The methods are applied to the study of the modification of the spectral lines of a magnesium transition by a helium atmosphere that can be found on magnetic WD stars.

Tensor hypercontraction for fully self-consistent imaginary-time GF2 and GWSOX methods: Theory, implementation, and role of the Green's function second-order exchange for intermolecular interactions

We present an efficient MPI-parallel algorithm and its implementation for evaluating the self-consistent correlated second-order exchange term (SOX), which is employed as a correction to the fully self-consistent GW scheme called scGWSOX (GW plus the SOX term iterated to achieve full Green's function self-consistency). Due to the application of the tensor hypercontraction (THC) in our computational procedure, the scaling of the evaluation of scGWSOX is reduced from O(nτnAO5) to O(nτN2nAO2). This fully MPI-parallel and THC-adapted approach enabled us to conduct the largest fully self-consistent scGWSOX calculations with over 1100 atomic orbitals with only negligible errors attributed to THC fitting. Utilizing our THC implementation for scGW, scGF2, and scGWSOX, we evaluated energies of intermolecular interactions. This approach allowed us to circumvent issues related to reference dependence and ambiguity in energy evaluation, which are common challenges in non-self-consistent calculations. We demonstrate that scGW exhibits a slight overbinding tendency for large systems, contrary to the underbinding observed with non-self-consistent RPA. Conversely, scGWSOX exhibits a slight underbinding tendency for such systems. This behavior is both physical and systematic and is caused by exclusion-principle violating diagrams or corresponding corrections. Our analysis elucidates the role played by these different diagrams, which is crucial for the construction of rigorous, accurate, and systematic methods. Finally, we explicitly show that all perturbative fully self-consistent Green's function methods are size-extensive and size-consistent.

Galván I, Rivalta I, Aquilante F, Garavelli M, Lindh R, Segarra-Martí J. Characterizing Conical Intersections in DNA/RNA Nucleobases with Multiconfigurational Wave Functions of Varying Active Space Size

We characterize the photochemically relevant conical intersections between the lowest-lying accessible electronic excited states of the different DNA/RNA nucleobases using Cholesky decomposition-based complete active space self-consistent field (CASSCF) algorithms. We benchmark two different basis set contractions and several active spaces for each nucleobase and conical intersection type, measuring for the first time how active space size affects conical intersection topographies in these systems and the potential implications these may have toward their description of photoinduced phenomena. Our results show that conical intersection topographies are highly sensitive to the electron correlation included in the model: by changing the amount (and type) of correlated orbitals, conical intersection topographies vastly change, and the changes observed do not follow any converging pattern toward the topographies obtained with the largest and most correlated active spaces. Comparison across systems shows analogous topographies for almost all intersections mediating population transfer to the dark 1nO/Nπ* states, while no similarities are observed for the "ethylene-like" conical intersection ascribed to mediate the ultrafast decay component to the ground state in all DNA/RNA nucleobases. Basis set size seems to have a minor effect, appearing to be relevant only for purine-based derivatives. We rule out structural changes as a key factor in classifying the different conical intersections, which display almost identical geometries across active space and basis set change, and we highlight instead the importance of correctly describing the electronic states involved at these crossing points. Our work shows that careful active space selection is essential to accurately describe conical intersection topographies and therefore to adequately account for their active role in molecular photochemistry.

The OpenMolcas Web: A Community-Driven Approach to Advancing Computational Chemistry

The developments of the open-source OpenMolcas chemistry software environment since spring 2020 are described, with a focus on novel functionalities accessible in the stable branch of the package or via interfaces with other packages. These developments span a wide range of topics in computational chemistry and are presented in thematic sections: electronic structure theory, electronic spectroscopy simulations, analytic gradients and molecular structure optimizations, ab initio molecular dynamics, and other new features. This report offers an overview of the chemical phenomena and processes OpenMolcas can address, while showing that OpenMolcas is an attractive platform for state-of-the-art atomistic computer simulations.

Interpreting the Cu-O2 Antibonding Nature in Two Cu-O2 Complexes from Cu L-Edge X-ray Absorption Spectra

Cu-O2 structures play important roles in bioinorganic chemistry and enzyme catalysis, where the bonding between the Cu and O2 parts serves as a fundamental research concern. Here, we performed a multiconfigurational study on the copper L2,3-edge X-ray absorption spectra (XAS) of two copper enzyme model complexes to gain a better understanding of the antibonding nature from the clearly interpreted structure-spectroscopy relation. We obtained spectra in good agreement with the experiments by using the restricted active space second-order perturbation theory (RASPT2) method, which facilitated reliable chemical analysis. Spectral feature interpretations were supported by computing the spin-orbit natural transition orbitals. All major features were assigned to be mainly from Cu 2p to antibonding orbitals between Cu 3d and O2 π*, Cu 3d-πO-O* (type A), and a few also to mixed antibonding/bonding orbitals between Cu 3d and O2 π, Cu 3d ± πO-O (type M). Our calculations provided a clear illustration of the interactions between Cu 3d and O2 π*/π orbitals that are carried in the metal L-edge XAS.

Efficient quantum analytic nuclear gradients with double factorization

Efficient representations of the Hamiltonian, such as double factorization, drastically reduce the circuit depth or the number of repetitions in error corrected and noisy intermediate-scale quantum (NISQ) algorithms for chemistry. We report a Lagrangian-based approach for evaluating relaxed one- and two-particle reduced density matrices from double factorized Hamiltonians, unlocking efficiency improvements in computing the nuclear gradient and related derivative properties. We demonstrate the accuracy and feasibility of our Lagrangian-based approach to recover all off-diagonal density matrix elements in classically simulated examples with up to 327 quantum and 18 470 total atoms in QM/MM simulations with modest-sized quantum active spaces. We show this in the context of the variational quantum eigensolver in case studies, such as transition state optimization, ab initio molecular dynamics simulation, and energy minimization of large molecular systems.

Modelling Photoionisation in Isocytosine: Potential Formation of Longer-Lived Excited State Cations in its Keto Form

Studying the effects of UV and VUV radiation on non-canonical DNA/RNA nucleobases allows us to compare how they release excess energy following absorption with respect to their canonical counterparts. This has attracted much research attention in recent years because of its likely influence on the origin of our genetic lexicon in prebiotic times. Here we present a CASSCF and XMS-CASPT2 theoretical study of the photoionisation of non-canonical pyrimidine nucleobase isocytosine in both its keto and enol tautomeric forms. We analyse their lowest energy cationic excited states including 2 π + , 2 n O + and 2 n N + and compare these to the corresponding electronic states in cytosine. Investigating lower-energy decay pathways we find - unexpectedly - that keto-isocytosine+ presents a sizeable energy barrier potentially inhibiting decay to its cationic ground state, whereas enol-isocytosine+ features a barrierless and consequently ultrafast pathway analogous to the one previously found for the canonical (keto) form of cytosine+ . Dynamic electron correlation reduces the energy barrier in the keto form substantially (by ∼1 eV) but it is nevertheless still present. We additionally compute the UV/Vis absorption signals of the structures encountered along these decay channels to provide spectroscopic fingerprints to assist future experiments in monitoring these intricate photo-processes.

Analytic gradients for multiconfiguration pair-density functional theory with density fitting: Development and application to geometry optimization in the ground and excited states

Density fitting reduces the computational cost of both energy and gradient calculations by avoiding the computation and manipulation of four-index electron repulsion integrals. With this algorithm, one can efficiently optimize the geometries of large systems with an accurate multireference treatment. Here, we present the derivation of multiconfiguration pair-density functional theory for energies and analytic gradients with density fitting. Six systems are studied, and the results are compared to those obtained with no approximation to the electron repulsion integrals and to the results obtained by complete active space second-order perturbation theory. With the new approach, there is an increase in the speed of computation with a negligible loss in accuracy. Smaller grid sizes have also been used to reduce the computational cost of multiconfiguration pair-density functional theory with little effect on the optimized geometries and gradient values.

Optimization of atomic density-fitting basis functions for molecular two-electron integral approximations

A general procedure for the optimization of atomic density-fitting basis functions is designed with the balance between accuracy and numerical stability in mind. Given one-electron wavefunctions and energies, weights are assigned to the product densities, modeling their contribution to the exchange and second-order correlation energy, and a simple weighted error measure is minimized. Generally contracted Gaussian auxiliary basis sets are optimized to match the wavefunction basis sets [D. N. Laikov, Theor. Chem. Acc. 138, 40 (2019)] for all 102 elements in a scalar-relativistic approximation [D. N. Laikov, J. Chem. Phys. 150, 061103 (2019)].

Analytic gradients for state-averaged multiconfiguration pair-density functional theory

Analytic gradients are important for efficient calculations of stationary points on potential energy surfaces, for interpreting spectroscopic observations, and for efficient direct dynamics simulations. For excited electronic states, as are involved in UV-Vis spectroscopy and photochemistry, analytic gradients are readily available and often affordable for calculations using a state-averaged complete active space self-consistent-field (SA-CASSCF) wave function. However, in most cases, a post-SA-CASSCF step is necessary for quantitative accuracy, and such calculations are often too expensive if carried out by perturbation theory or configuration interaction. In this work, we present the analytic gradients for multiconfiguration pair-density functional theory based on SA-CASSCF wave functions, which is a more affordable alternative. A test set of molecules has been studied with this method, and the stationary geometries and energetics are compared to values in the literature as obtained by other methods. Excited-state geometries computed with state-averaged pair-density functional theory have similar accuracy to those from complete active space perturbation theory at the second-order.

Modern quantum chemistry with [Open]Molcas

MOLCAS/OpenMolcas is an ab initio electronic structure program providing a large set of computational methods from Hartree-Fock and density functional theory to various implementations of multiconfigurational theory. This article provides a comprehensive overview of the main features of the code, specifically reviewing the use of the code in previously reported chemical applications as well as more recent applications including the calculation of magnetic properties from optimized density matrix renormalization group wave functions.

Exploring the Mechanism of a Chiral N-Alkyl Imine-Based Light-Driven Molecular Rotary Motor at MS-CASPT2//CASSCF and MS-CASPT2//(TD) DFT Levels

The working mechanism including the photoisomerization and thermal isomerization steps of a chiral N-alkyl imine-based motor synthesized by Lehn et al. are revealed by MS-CASPT2//CASSCF and MS-CASPT2//(TD-)DFT methods. For the photoisomerization process of the imine-based motor, it involves both the bright (π,π*) state and the dark (n,π*) state. In addition, the MECI has similar geometry and energy to the minimum of the S₁ state, which shows that the process is barrierless and keeps the unidirectionality of rotation well; the result confirms the imine-based motor is a good candidate for a light-driven molecular rotary motor. For the thermal isomerization process of the imine-based motor, there are two even isomerization paths: one with the mechanism of the in-plane N inversion, the energy barriers of which are 29.6 kcal mol^-1 at MS3-CASPT2//CAM-B3LYP level and 29.2 kcal mol^-1 at MS3-CASPT2//CASSCF level; the other with the mechanism of ring inversion of the cycloheptatriene moiety, with energy barriers of 28.1 kcal mol^-1 at MS3-CASPT2//CAM-B3LYP level and 18.1 kcal mol^-1 at MS3-CASPT2//CASSCF level. According to the structural feature of the stator moiety, the imine molecule can be used as a two-step or a four-step light-driven rotary motor.

Ultrafast and radiationless electronic excited state decay of uracil and thymine cations: computing the effects of dynamic electron correlation

In this article we characterise the radiationless decay of the first few electronic excited states of the cations of DNA/RNA nucleobases uracil and thymine, including the effects of dynamic electron correlation on energies and geometries (optimised with XMS-CASPT2). In both systems, we find that one state of ²n and another two of ²π⁺ character can be populated following photoionisation, and their different minima and interstate crossings are located. We find strong similarities between uracil and thymine cations: with accessible conical intersections suggesting that depopulation of their electronic excited states takes place on ultrafast timescales in both systems, suggesting that they are photostable in agreement with previous theoretical (uracil⁺) evidence. We find that dynamic electron correlation separates the energy levels of the "3-state" conical intersection (D₂/D₁/D₀)_CI previously located with CASSCF for uracil⁺, which will therefore have a different geometry and higher energy. Simulating the electronic and vibrational absorptions allows us to characterise spectral fingerprints that could be used to monitor these cation photo-processes experimentally.

A direct-compatible formulation of the coupled perturbed complete active space self-consistent field equations on graphical processing units

We recently developed an algorithm to compute response properties for the state-averaged complete active space self-consistent field method (SA-CASSCF) that capitalized on sparsity in the atomic orbital basis. Our original algorithm was limited to treating small to moderate sized active spaces, but the recent development of graphical processing unit (GPU) based direct-configuration interaction algorithms provides an opportunity to extend this to large active spaces. We present here a direct-compatible version of the coupled perturbed equations, enabling us to compute response properties for systems treated with arbitrary active spaces (subject to available memory and computation time). This work demonstrates that the computationally demanding portions of the SA-CASSCF method can be formulated in terms of seven fundamental operations, including Coulomb and exchange matrix builds and their derivatives, as well as, generalized one- and two-particle density matrix and σ vector constructions. As in our previous work, this algorithm exhibits low computational scaling and is accelerated by the use of GPUs, making possible optimizations and nonadiabatic dynamics on systems with O(1000) basis functions and O(100) atoms, respectively.

Assessment of the Potential Energy Hypersurfaces in Thymine within Multiconfigurational Theory: CASSCF vs. CASPT2

The present study provides new insights into the topography of the potential energy hypersurfaces (PEHs) of the thymine nucleobase in order to rationalize its main ultrafast photochemical decay paths by employing two methodologies based on the complete active space self-consistent field (CASSCF) and the complete active space second-order perturbation theory (CASPT2) methods: (i) CASSCF optimized structures and energies corrected with the CASPT2 method at the CASSCF geometries and (ii) CASPT2 optimized geometries and energies. A direct comparison between these strategies is drawn, yielding qualitatively similar results within a static framework. A number of analyses are performed to assess the accuracy of these different computational strategies under study based on a variety of numerical thresholds and optimization methods. Several basis sets and active spaces have also been calibrated to understand to what extent they can influence the resulting geometries and subsequent interpretation of the photochemical decay channels. The study shows small discrepancies between CASSCF and CASPT2 PEHs, displaying a shallow planar or twisted ¹(ππ*) minimum, respectively, and thus featuring a qualitatively similar scenario for supporting the ultrafast bi-exponential deactivation registered in thymine upon UV-light exposure. A deeper knowledge of the PEHs at different levels of theory provides useful insight into its correct characterization and subsequent interpretation of the experimental observations. The discrepancies displayed by the different methods studied here are then discussed and framed within their potential consequences in on-the-fly non-adiabatic molecular dynamics simulations, where qualitatively diverse outcomes are expected.

Complete-active-space second-order perturbation theory (CASPT2//CASSCF) study of the dissociative electron attachment in canonical DNA nucleobases caused by low-energy electrons (0-3 eV)

Low-energy (0-3 eV) ballistic electrons originated during the irradiation of biological material can interact with DNA/RNA nucleobases yielding transient-anion species which undergo decompositions. Since the discovery that these reactions can eventually lead to strand breaking of the DNA chains, great efforts have been dedicated to their study. The main fragmentation at the 0-3 eV energy range is the ejection of a hydrogen atom from the specific nitrogen positions. In the present study, the methodological approach introduced in a previous work on uracil [I. González-Ramírez et al., J. Chem. Theory Comput. 8, 2769-2776 (2012)] is employed to study the DNA canonical nucleobases fragmentations of N-H bonds induced by low-energy electrons. The approach is based on minimum energy path and linear interpolation of internal coordinates computations along the N-H dissociation channels carried out at the complete-active-space self-consistent field//complete-active-space second-order perturbation theory level. On the basis of the calculated theoretical quantities, new assignations for the adenine and cytosine anion yield curves are provided. In addition, the π1 (-) and π2 (-) states of the pyrimidine nucleobases are expected to produce the temporary anions at electron energies close to 1 and 2 eV, respectively. Finally, the present theoretical results do not allow to discard neither the dipole-bound nor the valence-bound mechanisms in the range of energies explored, suggesting that both possibilities may coexist in the experiments carried out with the isolated nucleobases.

Molcas 8: New capabilities for multiconfigurational quantum chemical calculations across the periodic table

In this report, we summarize and describe the recent unique updates and additions to the Molcas quantum chemistry program suite as contained in release version 8. These updates include natural and spin orbitals for studies of magnetic properties, local and linear scaling methods for the Douglas-Kroll-Hess transformation, the generalized active space concept in MCSCF methods, a combination of multiconfigurational wave functions with density functional theory in the MC-PDFT method, additional methods for computation of magnetic properties, methods for diabatization, analytical gradients of state average complete active space SCF in association with density fitting, methods for constrained fragment optimization, large-scale parallel multireference configuration interaction including analytic gradients via the interface to the Columbus package, and approximations of the CASPT2 method to be used for computations of large systems. In addition, the report includes the description of a computational machinery for nonlinear optical spectroscopy through an interface to the QM/MM package Cobramm. Further, a module to run molecular dynamics simulations is added, two surface hopping algorithms are included to enable nonadiabatic calculations, and the DQ method for diabatization is added. Finally, we report on the subject of improvements with respects to alternative file options and parallelization.