Cora Went1 Joseph DuChene1 Yi-Rung Lin1 Wei-Hsiang Lin1 Michelle Sherrott1 Zakaria Al Balushi1 Deep Jariwala1 Joeson Wong1 Harry Atwater1

1, California Institute of Technology, Pasadena, California, United States

The two-dimensional transition metal dichalcogenides (2D TMDs) are excellent candidates for ultrathin, high-efficiency optoelectronic devices, such as solar cells, due to their direct bandgaps, high luminescence radiative efficiencies upon passivation, and high absorption efficiencies per unit thickness (one to two orders of magnitude higher than conventional semiconductors)[i]. A challenge for large-area optoelectronic or photovoltaic applications of TMDs is to demonstrate high optoelectronic quality in films grown by large-area synthesis methods, such as metalorganic chemical vapor deposition (MOCVD). In this work, we observe improvements in photoluminescence quantum yield in exfoliated molybdenum disulfide (MoS2) samples and in MOCVD-grown samples by a solution-based passivation with the superacid bis(trifluoromethane)sulfonimide, similar to previous reports[ii]. Our MOCVD-grown MoS2 films were grown in submonolayer, monolayer, and multilayer form on 1 cm2 SiO2/Si substrates using molybdenum hexacarbonyl and diethylsulfide precursors at T = 600 C. The MoS2 samples were passivated by immersion in solutions of bis(trifluoromethane)sulfonimide in acetonitrile under atmosphere, as previously reported[iii], and were characterized using photoluminescence mapping, photoluminescence spectroscopy and Raman spectroscopy. We demonstrate a higher photoluminescence quantum yield in MOCVD-grown MoS2 than in exfoliated MoS2 layers pre-treatment, and propose mechanisms to explain this surprising observation. Comparative luminescence measurements for exfoliated and MOCVD-grown MoS2 samples indicated an integrated band-edge intensity for MOCVD-grown layers that was at least 3 times higher than for exfoliated samples. We find that the vapors of the superacid solution can also passivate, and develop a cleaner, residue-free passivation process based on this observation. Since the superacid treatment on its own is unstable under exposure to water and other conditions, we explore the use of hexagonal boron nitride films of varying thicknesses for stable encapsulation of TMDs, as an alternative to encapsulating with a fluoropolymer film[iv]. Finally, we will discuss the role of TMD absorber layer passivation and encapsulation in solving some of the challenges that remain for 2D TMD solar cells including 1) increasing absorption through light-trapping, 2) limiting defects to enable a high radiative efficiency and high open circuit voltage, and 3) enhancing compability with carrier-collecting heterostructures such as p-n homojunctions, TMD heterojunctions or carrier-selective contacts to extract photogenerated charge carriers[v].

[i] Cao, L., MRS Bulletin 40, 592-599 (2015).

[ii] Amani, M. et al, Science 350, 1065-1068 (2015).

[iii] Amani, M. et al, ACS Nano 10, 6535-6541 (2016).

[iv] Kim, H.; Lien, D.; Amani, M.; Ager, J. W.; Javey, A., ACS Nano 11, 5179-5185 (2017).

[v] Jariwala, D.; Davoyan, A. R.; Wong, J.; Atwater, H. A., ACS Photonics (2017).