Optically Active Chalcogen Vacancies in Monolayer Semiconductors

Abstract

atomically thin 2D layered materials, such as monolayer transition metal dichalcogenides (TMDs), [5][6][7][8] hexagonal boron nitride (hBN). [9,10] and gallium selenide, [11] are attractive alternative hosts to overcome such fundamental limitations of bulk counterparts.Following the initial reports on single photon emitters observed in naturally occurring defects in as-grown and as-exfoliated 2D TMDs, [5][6][7][8] various strain and crystal defect engineering approaches have been developed to deterministically generate these quantum emitters. While local strain introduced by nano-pillars/ holes or nano-indents commonly results in single photon emitters in a variety of 2D materials including hBN, WS 2 , WSe 2 , and MoSe 2 , [12][13][14][15][16][17] creation of point defects by ion [10,18] and electron [9,19] beam irradiation has proved to be a viable route to inducing similar quantum emitters. Further, direct writing of quantum emitter arrays on monolayer MoS 2 with precisely controlled positions has been demonstrated using a focused helium ion beam. [20][21][22] Chalcogen vacancies (V X , X = S/Se), which are commonly present in TMDs, are known to introduce in-gap states. [23][24][25][26] Studies on helium-ion treated MoS 2 (Refs. [18,(20)(21)(22)27,28]) found that transitions involving such in-gap states can be optically bright and yield anti-bunched photons at sufficiently low-density defects. However, emission peaks commonly attributed to chalcogen vacancies in other TMDs are often broad (FWHM > 100 meV), lacking typical features of quantum emitters. [26,29,30] Defect engineering of atomically thin semiconducting crystals is an attractive route to developing single-photon sources and valleytronic devices. For these applications, defects with well-defined optical characteristics need to be generated in a precisely controlled manner. However, defect-induced optical features are often complicated by the presence of multiple defect species, hindering the identification of their structural origin. Here, we report systematic generation of optically active atomic defects in monolayer MoS 2 , WS 2 , MoSe 2 , and WSe 2 via proton-beam irradiation. Defect-induced emissions are found to occur ≈100 to 200 meV below the neutral exciton peak, showing typical characteristics of localized excitons such as saturation at high-excitation rates and long lifetime. Using scanning transmission electron microscopy, it is shown that freshly created chalcogen vacancies are responsible for the localized exciton emission. Density functional theory and ab initio GW plus Bethe-Salpeter-equation calculations reveal that the observed emission can be attributed to transitions involving defect levels of chalcogen vacancy and the valence band edge state.

Room-Temperature Photoluminescence Mediated by Sulfur Vacancies in 2D Molybdenum Disulfide

“…One way to investigate changes in the electronic structure of 2D TMD semiconductors due to vacancy defects is to characterize the absorption and PL from defect-induced in-gap states. Absorption from such defect states has been observed at room temperature (RT ∼300 K), whereas light emission is typically observed at low temperatures at which excitons are bound to defect states to facilitate radiative recombination. , As a result of this strong localization, the defect-mediated emission generally exhibits long lifetimes . In this work, we find that it is possible to observe emission from trapped excitons at RT when defect density ( n d ) is in the range of 10 13 cm –2 < n d < 10 15 cm –2 .…”

“…The emission energy of LX D is consistent with that of sulfur-vacancy-induced emission generated by in vacuo annealing, 13 focused He + irradiation, 13 and proton irradiation. 30 As the samples were cooled from 300 to 10 K, the dominant recombination pathway shows a transition from A − /A to LX N emission in the case of pristine MoS 2 . For the annealed sample, the transition from A − /A to In-gap states trap excitons and subsequently increase their PL decay lifetime.…”

“…For the annealed sample, the transition from A − /A to In-gap states trap excitons and subsequently increase their PL decay lifetime. 29,30 We measured the PL decay of the LX D emission at 300 and 8 K by time-resolved PL spectroscopy (Figure 3b). At 300 K (Figure 3b, top), pristine MoS 2 shows an A − /A exciton lifetime of 70 ps, and the annealed MoS 2 shows a slightly longer A − /A exciton lifetime of 163 ps due to the increased electron doping, which results in a larger proportion of trions 46 (see the Supporting Information for details of the time-resolved PL decay fitting procedure).…”

Gate-Tunable Bound Exciton Manifolds in Monolayer MoSe2

“…We employ proton beam irradiation to generate uniform V Se defects in monolayer MoSe 2 and study their optical properties after hBN encapsulation similar to our previous study 26 (Figure 1a). Figure 1b shows the low-temperature photoluminescence (PL) spectrum of the pristine and irradiated samples.…”

“…The first method involves substitutional doping during the growth process, while the second approach uses postgrowth treatment, such as ion or electron irradiation, which is known to preferentially generate chalcogen vacancies (V X ) in TMDs. Recent studies − consistently reported observation of V X -related subgap emissions in irradiated monolayer TMDs. There is, however, little consensus on the physical origin of BXs arising from V X defects.…”