In this paper we apply STD to calculations of the first order photocurrent in a 19.6 nm silicon p-n junction from DFT-NEGF. We recently adopted this Special Thermal Displacement (STD) approach to study electron transport in silicon systems with over 1000 atoms including electron-phonon coupling within the DFT-NEGF formalism Gunst et al. Recently, Zacharias and Guistino Zacharias and Giustino ( 2016) introduced a very efficient method for including phonon induced absorption processes using a single super cell calculation in which the atoms are displaced away from their equilibrium positions. Therefore state-of-the-art DFT calculations of phonon-assisted absorption has so far been limited to bulk crystals where the supercell contains only a few atoms Noffsinger et al. The study of phonon-assisted photon absorption from first-principles is notoriously difficult as it involves a double sum over fine grids of k points and complex two excitation processes.
A number of recent studies also show that EPC plays a key role in the outstanding performance of perovskite PV cells Yang et al. Silicon has an indirect bandgap and as such absorption of a photon around the bandgap energy must be accompanied by the absorption/emission of a phonon to conserve momentum. In spite of the influx of new thin film based PV cells, silicon remains the market leader and about 90 % of PV cells are still based on silicon where large modules with high efficiency and stability can be produced Philipps and Warmuth ( 2017). previously used to improve a continuum model study of transport through the interface between CZTS and the buffer material CdS important for the CZTS solar cell efficiency Crovetto et al. DFT combined with the non equilibrium Green’s function (NEGF) formalism was f.ex. These effects can, however, be captured using atomistic models based on density functional theory (DFT). It is difficult to include important effects such as confinement of electrons and phonons, surface- and strain in the continuum models. Here it is stressed how the abundance of candidate materials together with the lack of efficient devices highlight the need for efficient predictive device calculations.Ĭontinuum models are used extensively in the field of PV to extract benchmark parameters from measurements on devices and to predict the performance of new device geometries Burgelman et al. Recently a review was published on the design of new materials using first-principles calculations Butler et al. The field of computational material science has seen massive progression and as a result the difference between system size and complexity attainable in simulations and experiments is becoming smaller every day. ( 2016).Ĭlearly there is still room for discovery of new materials to improve on the cost/efficiency relationship. In the last couple of decades many promising thin film absorber materials have been discovered, all of them with unique strengths and weaknesses.ĬdTe and CIGS (CuInGaSe 2) can produce high efficiencies, but include rare and toxic elements, CZTS includes only nontoxic earth abundant elements, but suffers from low efficiency and open circuit voltage ( V o c) Philipps and Warmuth ( 2017) Fthenakis ( 2004) Woodhouse et al. Photovoltaics (PV) represents a promising technology as a replacement for burning fossil fuels. This work represents a recipe for computational characterization of future photovoltaic devices including the combined effects of light-matter interaction, phonon-assisted tunneling and the device potential at finite bias from the level of first-principles simulations. Our calculations illustrate the pivotal role played by EPC in photocurrent modelling to avoid underestimation of the open-circuit voltage, short-circuit current and maximum power. The first-principles results are successfully compared to experimental measurements of the temperature and light intensity dependence of the open-circuit voltage of a silicon photovoltaic module. We apply the method to a silicon solar cell device and demonstrate the impact of including EPC in order to properly describe the current due to the indirect band-to-band transitions.
The photocurrent is calculated using nonequilibrium Green’s function with light-matter interaction from the first-order Born approximation while electron-phonon coupling (EPC) is included through special thermal displacements (STD). We present a straightforward and computationally cheap method to obtain the phonon-assisted photocurrent in large-scale devices from first-principles transport calculations.