On the accuracy of the general, state-specific polarizable-continuum model for the description of correlated ground- and excited states in solution

Abstract

Equilibrium and non-equilibrium formulations of the state-specific polarizable-continuum model (SS-PCM) are evaluated in combination with correlated ground- and excited-state densities provided by the algebraic-diagrammatic construction method (ADC) for the polarization propagator via the computationally efficient intermediate-state representation (ISR) formalism. Since the influence of the SS-PCM onto quantum-chemical method is naturally limited to the presence of the apparent surface charges in the one-electron Hamiltonian and hence fully contained in the polarized MOs, the herein presented solvent model can be combined with all implemented orders and variants of ADC. Employing ADC/SS-PCM, the symmetric, ionized dimers of neon, ethene and nitromethane are investigated. Their broken-symmetry wavefunctions exhibit a low-lying charge-transfer state that is symmetry-equivalent to the ground state. This curious though ultimately artificial feature is convenient as it allows for a direct comparison of ADC/SS-PCM for the CT state to the Møller-Plesset/PCM description of the ground state. The agreement down to 0.02 eV for a wide range of dielectric constants validates the ADC/SS-PCM approach. Eventually, the relaxed potential-energy surfaces of the ground and lowest excited states of 4-(N,N)-dimethylaminobenzonitrile in cyclohexane and acetonitrile are computed, and it is demonstrated that the ADC(2)/SS-PCM approach affords excellent agreement with experimental fluorescence data. Only at the ADC(3) level of theory, however, the experimentally observed solvent-dependent dual fluorescence can be explained.

CCSD‐PCM Excited State Energy Gradients with the Linear Response Singles Approximation to Study the Photochemistry of Molecules in Solution

“…This phenomenon is due to a twisted intramolecular charge transfer state (TICT, called CT in the following) that is stabilized in polar solvents, indeed becoming more stable than the locally excited state (LE) from where the fluorescence occurs in gas phase and low-polar solvents. This compound has been extensively studied with a variety of single and multi-reference methods, [15][16][17][18][19][20][21][22] but this is the first time that excited state geometry optimizations were performed at CCSD level. Some studies have investigated various local minima for each electronic state, but we focus on the lowest energy structures for the CT and LE states, also shown in Figure 1 from the gas phase optimizations.…”

“…Thus, emission occurs from less twisted configurations due to vibrational motion that move the position of the observed band to higher energies. [22]…”

“…We report the equations for the implementation of the energy gradients in Section 2, details of the calculations in Section 3, and an application of the method to 4-(N,N)-dimethyl-aminobenzonitrile (DMABN) in a polar and a low-polar solvent in Section 4. This molecule has been extensively studied experimentally and computationally, [15][16][17][18][19][20][21][22] and it allows us to discuss some limitations of the LR and SS formalisms. Final considerations and concluding remarks are reported in Section 5.…”

How Boron Doping Shapes the Optoelectronic Properties of Canonical and Phenylene‐Containing Oligoacenes: A Combined Experimental and Theoretical Investigation

“…Alternatively, one could employ long‐range corrected/range‐separated functionals such as CAM‐B3LYP. To estimate the impact of a typical nonpolar and polar solvent onto relaxed excited‐state energies, we employed the difference‐density based, non‐equilibrium implementation of COSMO as implemented in ORCA with the dielectric constant ( ϵ ) and optical dielectric constant (squared refractive index, n 2 ) set to 2 and 10, respectively (see the caption of Table and the Supporting Information for details on this approach) …”

“…Alternatively, one coulde mploy long-range corrected/range-separated functionals such as CAM-B3LYP.T oe stimate the impact of at ypical nonpolarand polar solvent onto relaxed excited-state energies, we employed the difference-density based, non-equilibrium implementation of COSMO [53] as implementedi nO RCAw ith the dielectric constant (e)a nd optical dielectric constant (squared refractive index, n 2 )s et to 2a nd 10, respectively (see the captiono fT able2 and the Supporting Information for details on this approach). [54,55] In iDBP and DBP,S 1 is al ocally excited (LE) state, whiche xhibits al arge oscillator strength( f Osc )o f0 .34 and 0.41, Figure 6. Energetics of MeCN-adduct formationso fDBI, DBP,and DBA at the highest CEPA/CBS//SCS-MP2/def2-TZVPleveloft heory (dE)w ith zeropoint contributions (ZPE) and thermal corrections( RT) computed at the B3LYP-D3BJ/SVP level of theory.…”