Developement of DFT XC functional

Computational modeling based on DFT calcualtions is advantageous over other theoretical methods, such as those are based on wave function based theory and on many-body perturbation theory. This is because of relatively small and tractable computational cost for medium to large system sizes by the DFT method. However, predictions from DFT simulations largely depend on the choice of exchange and correlation functional (XC) employed in the calculations. Standard LDA and gradient corrected semi-local GGA functionals are known to produce erroneous data for the transport gap for molecules (band gaps for solids) and severely underestimated excitonic energies for the charge transfer (CT) excited states. The former one, i.e., gap underestimation is related with the lack of derivative discontinuity present in the commonly use XC functionals (such as LDA and GGA). While the wrong electron-electron and electron-hole interactions potentials provided by these XC functionals within time-dependent-DFT result in incorrect description of the CT excitons. This eventually calls for a smart XC functional which can simultaneously produce quantitatively accurate gap values and energies for the CT excitonic states.

Optimally-tuned Range-separated hybrid (OT-RSH) functionals have shown great promise along this direction only very recently, where electron-electron interaction potential is splited into a short-range (SR) and long-range (LR) parts via a system dependent range-separation parameter and in general numerically solved by using error functions. The SR part is treated with stardard GGA type exchange, whereas a full HF-exchange is used for the LR portion. This OT-RSH functional produces quantitatively accurate transport gap and physically meaningful orbital energies. Importantly, it also provides correct asymptotic description of electron-electron Coulomb potential at large distances and thereby resulting in quantitatively accurate exciton binding energy. These OT-RSH XC functionals mitigate the problems, such as lack of derivative discontinuity and electrons self-interactions error and inaccurate electron-electron interactions potential imposed by commonly used XC functionals (such as LDA, GGA and even with hybrid). Furthermore, this ingenious OT-RSH functional has been extended to include the electrostatic screening effects due to electronic polarization by the surrounding environment in molecular crystals and extended solids.

We design smart and modern XC functionals in order to capture quantitatively accurate and for providing correct physical and chemical picture of the problems in a range of molecules and diverse materials that are of our extensive research interest.