Fast exciton annihilation by capture of electrons or holes by defects via Auger scattering in monolayer metal dichalcogenides

Abstract

The strong Coulomb interactions and the small exciton radii in two-dimensional metal dichalcogenides can result in very fast capture of electrons and holes of excitons by mid-gap defects from Auger processes. In the Auger processes considered here, an exciton is annihilated at a defect site with the capture of the electron (or the hole) by the defect and the hole (or the electron) is scattered to a high energy. In the case of excitons, the probability of finding an electron and a hole near each other is enhanced many folds compared to the case of free uncorrelated electrons and holes. Consequently, the rate of carrier capture by defects from Auger scattering for excitons in metal dichalcogenides can be 100-1000 times larger than for uncorrelated electrons and holes for carrier densities in the 10 11 -10 12 cm −2 range. We calculate the capture times of electrons and holes by defects and show that the capture times can be in the sub-picosecond to a few picoseconds range. The capture rates exhibit linear as well as quadratic dependence on the exciton density. These fast time scales agree well with the recent experimental observations and point to the importance of controlling defects in metal dichalcogenides for optoelectronic applications.

Surface Recombination Limited Lifetimes of Photoexcited Carriers in Few-Layer Transition Metal Dichalcogenide MoS2

“…The refractive index response is given by the changes in the imaginary part of the optical conductivity which can affect the probe transmission through the term containing the product processes in which carrier capture by defects occurs via Auger scattering have been presented by the authors in previous works 12,18,25 , and used successfully to model the carrier recombination dynamics in monolayer and bulk MoS 2 samples 12,18 . A prominent feature of Auger scattering is recombination times that are independent of the temperature but depend on the carrier density (or the pump fluence).…”

“…We assume a fast recombination time τ s (as given by the products of the Auger rate constants and the defect density) for the two surface layers in a few-layer sample, and a slow recombination time τ i for all the inner layers, and then estimate the actual recombination time for a few-layer sample by weighing the inverse lifetime with the probability of electron (or hole) occupation of each layer as given by the electron (or hole) wavefunction in a few-layer sample. This procedure is equivalent to assuming a spatially varying defect structure in a more formal calculation technique such as the one presented by Wang et al 25 . This procedure also assumes that the carriers are mobile in the direction normal to the plane of the layers, an assumption that seems to be justified by the large splittings of the energy subbands observed in the conduction band minima and the valence band maxima in few layer MoS 2 in calculations 47 .…”

“…Recently, defect-assisted electron-hole recombination was proposed as the dominant non-radiative recombination mechanism in monolayer and bulk MoS 2 12,18,25,35 . The carrier density and the temperature dependence of the observed recombination dynamics suggested that photoexcited carriers are captured by defects via Auger processes 12,18,25 . It is reasonable to expect that surface layers of few-layer TMDs have far more defects and impurities than the inner layers and, consequently, if defect-assisted processes are indeed responsible for electron-hole recombination in TMDs, then the recombination time would scale in some meaningful way with the number of layers in few-layer TMDs.…”

“…The ultrafast dynamics of photoexcited carriers and the non-radiative recombination mechanisms in monolayer TMDs, and in MoS 2 in particular, have been the subject of several recent experimental 13,[18][19][20][21][22][23][24] and theoretical studies 25 . Non-radiative electron-hole recombination in MoS 2 monolayers results in photoexcited carrier lifetimes in the few tens of picosecond range.…”

Exciton trapping is responsible for the long apparent lifetime in acid-treated MoS2

“…5 Calculations predict a room temperature lifetime in the range of hundreds of picoseconds, 14,15 in agreement with some experiments. 16,17 However, many processes can obscure the actual radiative lifetime, such as defect-mediated non-radiative decay, 18 exciton-exciton interactions, 19 or equilibration with dark states. 20 Here we show that the long apparent lifetime in acid-treated MoS 2 is due to the presence of long-lived (4 µs) dark states that extend hundreds of meV into the bandgap.…”