A program for the direct calculation of excitation energies of atoms and molecules in strong magnetic fields is presented. The implementation includes the equation-of-motion coupled-cluster singles-doubles (EOM-CCSD) method for electronically excited states as well as its spin-flip variant. Differences to regular EOM-CCSD implementations are due to the appearance of the canonical angular-momentum operator in the Hamiltonian causing the wave function to become complex. The gauge-origin problem is treated by the use of gauge-including atomic orbitals. Therefore, a modified Davidson method for diagonalizing complex non-Hermitian matrices is used. Excitation energies for selected atoms and molecules that are of importance in the astrochemical context are presented and their dependence on the magnetic field is discussed.
Finite-field EOM-CCSDT: a highly accurate method for the theoretical prediction of excitation energies and electronic spectra in strong magnetic fields.
Quasiparticle energies of the atoms H-Ne have been computed in the GW approximation in the presence of strong magnetic fields with field strengths varying from 0 to 0.25 atomic units (0.25 B0 = 0.25 e −1 a −2 0 ≈ 58763 T). The GW quasiparticle energies are compared with equationof-motion ionization-potential (EOM-IP) coupled-cluster singles-and-doubles (CCSD) calculations of the first ionization energies. The best results are obtained with the evGW @PBE0 method, which agrees with the EOM-IP-CCSD model to within about 0.20 eV. Ionization potentials have been calculated for all atoms in the series, representing the first systematic study of ionization potentials for the first-row atoms at field strengths characteristic of magnetic white dwarf stars. Under these conditions, the ionization potentials increase in a near-linear fashion with the field strength, reflecting the linear field dependence of the Landau energy of the ionized electron. The calculated ionization potentials agree well with the best available literature data for He, Li, and Be.
An implementation of transition-dipole moments at the equation-of-motion coupled-cluster singles-doubles (EOM-CCSD) and CCSD linear response (LR) levels of theory for the treatment of atoms and molecules in strong magnetic fields is presented. The presence of a finite magnetic field leads, in general, to a complex wave function and a gauge-origin dependence, necessitating a complex computer code together with the use of gauge-including atomic orbitals. As in the field-free case, for EOM-CC, the evaluation of transition-dipole moments consists of setting up the one-electron transition-density matrix (TDM) which is then contracted with dipole-moment integrals. In the case of CC-LR, the evaluation proceeds with a modified TDM but additionally requires a second contribution accounting for the amplitude response which is missing in EOM-CC theory for properties. We present a selected set of transitions for the sodium atom and investigate the LiH molecule in both a parallel as well as a perpendicular magnetic field. The dependence of excited-state energies and transition moments on the magnetic-field strength is discussed with a focus on magnetic-field-induced avoided crossings. Additionally, the differences between field-dependent EOM-CCSD and CCSD-LR transition moments are investigated.
Magnetic white dwarfs with field strengths below 10 MG are easy to recognise since the Zeeman splitting of spectral lines appears proportional to the magnetic field strength. For fields ≳ 100 MG, however, transition wavelengths become chaotic, requiring quantum-chemical predictions of wavelengths and oscillator strengths with a non-perturbative treatment of the magnetic field. While highly accurate calculations have previously been performed for hydrogen and helium, the variational techniques employed become computationally intractable for systems with more than three to four electrons. Modern computational techniques, such as finite-field coupled-cluster theory, allow the calculation of many-electron systems in arbitrarily strong magnetic fields. Because around 25 percent of white dwarfs have metal lines in their spectra, and some of those are also magnetic, the possibility arises for some metals to be observed in very strong magnetic fields, resulting in unrecognisable spectra. We have identified SDSS J114333.48+661531.83 as a magnetic DZ white dwarf, with a spectrum exhibiting many unusually shaped lines at unknown wavelengths. Using atomic data calculated from computational finite-field coupled-cluster methods, we have identified some of these lines arising from Na, Mg, and Ca. Surprisingly, we find a relatively low field strength of 30 MG, where the large number of overlapping lines from different elements make the spectrum challenging to interpret at a much lower field strength than for DAs and DBs. Finally we model the field structure of SDSS J1143+6615 finding the data are consistent with an offset dipole.
Magnetic white dwarfs with field strengths below 10 MG are easy to recognise since the Zeeman splitting of spectral lines appears proportional to the magnetic field strength. For fields 100 MG, however, transition wavelengths become chaotic, requiring quantum-chemical predictions of wavelengths and oscillator strengths with a non-perturbative treatment of the magnetic field. While highly accurate calculations have previously been performed for hydrogen and helium, the variational techniques employed become computationally intractable for systems with more than three to four electrons. Modern computational techniques, such as finite-field coupled-cluster theory, allow the calculation of many-electron systems in arbitrarily strong magnetic fields. Because around 25 percent of white dwarfs have metal lines in their spectra, and some of those are also magnetic, the possibility arises for some metals to be observed in very strong magnetic fields, resulting in unrecognisable spectra. We have identified SDSS J114333.48+661531.83 as a magnetic DZ white dwarf, with a spectrum exhibiting many unusually shaped lines at unknown wavelengths. Using atomic data calculated from computational finite-field coupled-cluster methods, we have identified some of these lines arising from Na, Mg, and Ca. Surprisingly, we find a relatively low field strength of 30 MG, where the large number of overlapping lines from different elements make the spectrum challenging to interpret at a much lower field strength than for DAs and DBs. Finally we model the field structure of SDSS J1143+6615 finding the data are consistent with an offset dipole.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.