Approximate Analytical Gradients and Nonadiabatic Couplings for the State-Average Density Matrix Renormalization Group Self-Consistent-Field Method

Freitag, Leon; Ma, Yingjin; Baiardi, Alberto; Knecht, Stefan; Reiher, Markus

doi:10.1021/acs.jctc.9b00969

Cited by 22 publications

(25 citation statements)

References 149 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the method relies on conventional 1- and 2-RDMs, it can be easily applied to both MCSCF and DMRGSCF optimizations. Second, a promising approach for calculation of state-averaged DMRGSCF analytical gradients was proposed and implemented on the basis of OpenMolcas. The method is closely related to the formalism based on Lagrange multipliers, which is used in state-averaged CASSCF.…”

Section: Discussionmentioning

confidence: 99%

Ab Initio Study of Low-Lying Excited States of Carotenoid-Derived Polyenes

Khokhlov

Belov

2020

J. Phys. Chem. A

View full text Add to dashboard Cite

Knowledge about excited states of carotenoids is essential for understanding photophysical processes underlying photosynthesis. However, due to the presence of a large number of optically dark states, experimental study of the excited-state manifold is limited to a significant extent. In this paper, we apply high-level ab initio quantum chemical methods to study the low-lying excited states of polyenes containing from 8 to 13 conjugated double bonds, which serve as a model for natural carotenoids. Vertical and adiabatic excitation energies from the ground 1Ag – state to the excited 2Ag –, 1Bu +, and 1Bu – states were evaluated by means of density matrix renormalization group (DMRG) with NEVPT2 perturbative correction. The energies of all excited states are highly sensitive to nuclear geometry, especially the 2Ag – state. Thus, the 2Ag – and 1Bu + states interchange their relative positions upon geometry relaxation, while the vertical excitation energy to the 2Ag – state is rather high. At the same time, the 1Bu – state energy is shown to be higher than other studied excited states at any geometry. With relaxed geometries of the excited states, absorption and transient absorption spectra were calculated within the Franck–Condon approximation bridging the gap between experimental spectroscopic data and computational results.

show abstract

Section: Discussionmentioning

confidence: 99%

Ab Initio Study of Low-Lying Excited States of Carotenoid-Derived Polyenes

Khokhlov

Belov

2020

J. Phys. Chem. A

View full text Add to dashboard Cite

show abstract

“…As an alternative to deal with large active spaces, the density matrix renormalization group (DMRG) can be used . A DMRG-SCF calculation is similar to a CASSCF calculation, but instead of a FCI solution of the active space, an approximated solution with DMRG is obtained to avoid the exponential scaling of the computational costs with the number of active orbitals. − Very recently, transcorrelated DMRG (tcDMRG) was introduced for strongly correlated systems…”

Section: Quantum Chemical Theory and Methodsmentioning

confidence: 99%

Machine Learning for Electronically Excited States of Molecules

2020

View full text Add to dashboard Cite

Electronically excited states of molecules are at the heart of photochemistry, photophysics, as well as photobiology and also play a role in material science. Their theoretical description requires highly accurate quantum chemical calculations, which are computationally expensive. In this review, we focus on not only how machine learning is employed to speed up such excited-state simulations but also how this branch of artificial intelligence can be used to advance this exciting research field in all its aspects. Discussed applications of machine learning for excited states include excited-state dynamics simulations, static calculations of absorption spectra, as well as many others. In order to put these studies into context, we discuss the promises and pitfalls of the involved machine learning techniques. Since the latter are mostly based on quantum chemistry calculations, we also provide a short introduction into excited-state electronic structure methods and approaches for nonadiabatic dynamics simulations and describe tricks and problems when using them in machine learning for excited states of molecules.

show abstract

“…A DMRG-SCF calculation is similar to a CASSCF calculation, but instead of a FCI solution of the active space, an approximated solution with DMRG is obtained to avoid the exponential scaling of the computational costs with the number of active orbitals. [296][297][298][299][300][301] As an alternative, ML can be used to determine an active space. 70…”

Section: Configuration Interactionmentioning

confidence: 99%

Machine learning for electronically excited states of molecules

Westermayr,

Marquetand

2020

Preprint

View full text Add to dashboard Cite

Electronically excited states of molecules are at the heart of photochemistry, photophysics, as well as photobiology and also play a role in material science. Their theoretical description requires highly accurate quantum chemical calculations, which are computationally expensive. In this review, we focus on how machine learning is employed not only to speed up such excited-state simulations but also how this branch of artificial intelligence can be used to advance this exciting research field in all its aspects. Discussed applications of machine learning for excited states include excited-state dynamics simulations, static calculations of absorption spectra, as well as many others. In order to put these studies into context, we discuss the promises and pitfalls of the involved machine learning techniques. Since the latter are mostly based on quantum chemistry calculations, we also provide a short introduction into excited-state electronic structure methods, approaches for nonadiabatic dynamics simulations and describe tricks and problems when using them in machine learning for excited states of molecules.

show abstract

Approximate Analytical Gradients and Nonadiabatic Couplings for the State-Average Density Matrix Renormalization Group Self-Consistent-Field Method

Cited by 22 publications

References 149 publications

Ab Initio Study of Low-Lying Excited States of Carotenoid-Derived Polyenes

Ab Initio Study of Low-Lying Excited States of Carotenoid-Derived Polyenes

Machine Learning for Electronically Excited States of Molecules

Machine learning for electronically excited states of molecules

Contact Info

Product

Resources

About