Achintya Kumar Dutta

Towards a pair natural orbital coupled cluster method for excited states

The use of back-transformed pair natural orbitals in the calculation of excited state energies, ionization potentials, and electron affinities is investigated within the framework of equation of motion coupled cluster theory and its similarity transformed variant. Possible approaches to a more optimal use of pair natural orbitals in these methods are indicated.

Performance of the EOMIP-CCSD(2) Method for Determining the Structure and Properties of Doublet Radicals: A Benchmark Investigation.

We present a benchmark study on the performance of the EOMIP-CCSD(2) method for computation of structure and properties of doublet radicals. The EOMIP-CCSD(2) method is a second-order approximation to the standard EOMIP-CCSD method. By retaining the black box nature of the standard EOMIP-CCSD method and adding favorable N(5) scaling, the EOMIP-CCSD(2) method can become the method of choice for predicting the structure and spectroscopic properties of large doublet radicals. The EOMIP-CCSD(2) method overcomes the typical problems associated with the standard single reference ab initio treatment of doublet radicals. We compare our results for geometries and harmonic vibrational frequencies with those obtained using the standard EOMIP-CCSD method, as well as unrestricted Hartree-Fock (UHF)- and restricted open-shell Hartree-Fock (ROHF)-based single-reference coupled-cluster and second order many-body perturbation theory (MBPT(2)) methods. The effect of the basis set on the quality of the results has been studied using a hierarchy of Dunnings correlation-consistent aug-cc-pVXZ (X = D, T, Q) basis sets. Numerical results show that the EOMIP-CCSD(2) method, despite its N(5) scaling, gives better agreement with experimental results, compared to the UHF- and ROHF-based MBPT(2), as well as the single-reference coupled-cluster methods.

Perturbative approximations to single and double spin flip equation of motion coupled cluster singles doubles methods

Spin flip equation of motion coupled cluster (EOM-SF-CC) can correctly treat situations involving electronic degeneracies or near degeneracies, e.g., bond breaking, di- and tri-radicals, etc. However, for large systems EOM-SF-CC (even in single and double excitations) is computationally prohibitively expensive. Therefore, earlier approximations to EOM-SF-CC methods such as spin flip configuration interaction singles with perturbative doubles (SF-CIS(D)) have been proposed. In this work, we present a new perturbative approximation to EOM-SF-CC, which has been found to be more accurate than SF-CIS(D). The capabilities, advantages, and timings of the new approach have been demonstrated considering the singlet-triplet gaps in di- and triradicals as well as bond breaking examples. The method is extended to double spin flip EOM-CC and its capabilities have been tested.

Speeding up equation of motion coupled cluster theory with the chain of spheres approximation

In the present paper, the chain of spheres exchange (COSX) approximation is applied to the highest scaling terms in the equation of motion (EOM) coupled cluster equations with single and double excitations, in particular, the terms involving integrals with four virtual labels. It is found that even the acceleration of this single term yields significant computational gains without compromising the desired accuracy of the method. For an excitation energy calculation on a cluster of five water molecules using 585 basis functions, the four virtual term is 9.4 times faster using COSX with a loose grid than using the canonical implementation, which yields a 2.6 fold acceleration for the whole of the EOM calculation. For electron attachment calculations, the four virtual term is 15 times and the total EOM calculation is 10 times faster than the canonical calculation for the same system. The accuracy of the new method was tested using Thiels test set for excited states using the same settings and the maximum absolute deviation over the whole test set was found to be 12.945 cm(-1) (59 μHartree) for excitation energies and 6.799 cm(-1) (31 μHartree) for electron attachments. Using MP2 amplitudes for the ground state in combination with the parallel evaluation of the full EOM equations in the manner discussed in this paper enabled us to perform calculations for large systems. Electron affinity values for the two lowest states of a Zn protoporphyrine model compound (224 correlated electrons and 1120 basis functions) were obtained in 3 days 19 h using 4 cores of a Xeon E5-2670 processor allocating 10 GB memory per core. Calculating the lowest two excitation energies for trans-retinal (114 correlated electrons and 539 basis functions) took 1 day 21 h using eight cores of the same processor and identical memory allocation per core.

Partitioned EOMEA-MBPT(2): An Efficient N(5) Scaling Method for Calculation of Electron Affinities.

We present an N(5) scaling modification to the standard EOMEA-CCSD method, based on the matrix partitioning technique and perturbative approximations. The method has lower computational scaling and smaller storage requirements than the standard EOMEA-CCSD method and, therefore, can be used to calculate electron affinities of large molecules and clusters. The performance and capabilities of the new method have been benchmarked with the standard EOMEA-CCSD method, for a test set of 20 small molecules, and the average absolute deviation is only 0.03 eV. The method is further used to investigate electron affinities of DNA and RNA nucleobases, and the results are in excellent agreement with the experimental values.

EOMIP-CCSD(2)*: An Efficient Method for the Calculation of Ionization Potentials

A new approximation within the domain of EOMIP-CC method is proposed. The proposed scheme is based on the perturbative truncation of the similarity transformed effective Hamiltonian matrix. We call it the EOMIP-CCSD(2)* method, which scales as noniterative N(6) and its storage requirement is very less, compared to the conventional EOMIP-CCSD method. The existing EOMIP-CCSD(2) method has a tendency to overestimate the ionization potential (IP) values. On the other hand, our new strategy corrects for the problem of such an overestimation, which is evident from the excellent agreement achieved with the experimental values. Furthermore, not only the ionization potential but also geometry and IR frequencies of problematic double radicals are estimated correctly, and the results are comparable to the CCSD(T) method, obviously at lesser computational cost. The EOMIP-CCSD(2)* method works even for the core ionization and satellite IP, where the earlier EOMIP-CCSD(2) approximation dramatically fails.

Intermediate Hamiltonian Fock Space Multireference Coupled Cluster Approach to Core Excitation Spectra

The Fock space multireference coupled cluster (FSMRCC) method provides an efficient approach for the direct calculation of excitation energies. In intermediate Hamiltonian (IH-FSMRCC) formulation, the method is free from intruder state problems and associated convergence difficulties, even with a large model space. In this paper, we demonstrate that the IH-FSMRCC method with suitably chosen model space can be used for the accurate description of core excitation spectra of molecules, and our results are in excellent agreement with the experimental values. We have investigated the effect of choice of model space on the computed results. Unlike the equation-of-motion (EOM)-based method, the IH-FSMRCC does not require any special technique for convergence and in singles and doubles approximation gives a performance comparable to that of the standard EOMEE-CCSD method, even better in some of the cases.

Automatic active space selection for the similarity transformed equations of motion coupled cluster method

An efficient scheme for the automatic selection of an active space for similarity transformed equations of motion (STEOM) coupled cluster method is proposed. It relies on state averaged configuration interaction singles (CIS) natural orbitals and makes it possible to use STEOM as a black box method. The performance of the new scheme is tested for singlet and triplet valence, charge transfer, and Rydberg excited states.

Exploring the accuracy of a low scaling similarity transformed equation of motion method for vertical excitation energies

The newly developed back transformed pair natural orbital based similarity transformed equation of motion (bt-STEOM) method at the coupled cluster singles and doubles level (CCSD) is combined with an appropriate modification of our earlier active space selection scheme for STEOM. The resulting method is benchmarked for valence, Rydberg, and charge transfer excited states of Thiels test set and other test systems. The bt-PNO-STEOM-CCSD method gives very similar results to canonical STEOM-CCSD for both singlet and triplet excited states. It performs in a balanced manner for all these types of excited states, while the EOM-CCSD method performs especially well for Rydberg excited states and the CC2 method excels at obtaining accurate results for valence excited states. Both EOM-CCSD and CC2 perform worse than bt-PNO-STEOM-CCSD for charge transfer states for the test cases studied.

Structure, stability, and properties of the trans peroxo nitrate radical: the importance of nondynamic correlation.

We report a comparative single-reference and multireference coupled-cluster investigation on the structure, potential energy surface, and IR spectroscopic properties of the trans peroxo nitrate radical, one of the key intermediates in stratospheric NOX chemistry. The previous single-reference ab initio studies predicted an unbound structure for the trans peroxo nitrate radical. However, our Fock space multireference coupled-cluster calculation confirms a bound structure for the trans peroxo nitrate radical, in accordance with the experimental results reported earlier. Further, the analysis of the potential energy surface in FSMRCC method indicates a well-behaved minima, contrary to the shallow minima predicted by the single-reference coupled-cluster method. The harmonic force field analysis, of various possible isomers of peroxo nitrate also reveals that only the trans structure leads to the experimentally observed IR peak at 1840 cm(-1). The present study highlights the critical importance of nondynamic correlation in predicting the structure and properties of high-energy stratospheric NOx radicals.