Obviously, to enable full exploitation of the unique properties of atomically-thin MoS2 films, similar to traditional semiconductor technology, stable n and p-type conduction in the TMDs is a key requirement for all intended applications, such as heterojunction bipolar transistors, light emitting diodes, and photodetectors . Yet, the bipolar doping strategy of the TMDs needs to be refined. Customarily, controlling majority charge carriers via doping is realized in conventional semiconductors by covalently bonded impurities. However, as to MoS2, the insight on the ability of robust and tunable ‘conventional’ doping appears limited, in particular regarding p-type dopants.
In this respect, the current work deals with multifrequency electron spin resonance (ESR) study of three types of covalently bonded impurity acceptors in p-type 2H MoS2 crystals, where ESR is demonstrated as an exclusive technique for defect identification, quantification, and, outstandingly, first inference of the corresponding electrical activation energy Ea.
A first part will deal with p-type geological (geo) 2H MoS2 crystals, where ESR reveals p-type doping predominantly originating from As atoms substituting for S sites in densities (spin S = ½) of ~2.4 ´ 1017 cm-3. Observation of a ‘half field’(g ~ 3.883) signal firmly correlating with the central Zeeman As accepter signal indicates the presence of spin S > ½ As agglomerates, which together with the distinct multicomponent makeup of the Zeeman signal points to manifest non-uniform As doping; only ~13% of the total As response originates from individual decoupled As dopants. From ESR monitoring of the latter vs.T, an activation energy Ea = (0.7 ± 0.2) meV is obtained. This unveils As as a noticeable shallow acceptor, appropriate for realization of efficient and stable p-type doping.
Next, in studying p-type synthetic 2H MoS2, we will present results on the first ESR observation of the N impurity dopant. It is emerging as a signal of axial-symmetry characteristic for a hole-type center in MoS2 . The signal, corresponding to a defect density (S = ½) of ~2.3 x 1017 cm-3, is identified as originating from N acceptor dopants, the N atoms substituting for S sites. For the applied magnetic field B//c-axis, the signal is mainly comprised of a 14N hyperfine triplet with, on top, a center line accounting for ~23% of the total signal intensity. The additional observation of a weak half-field signal (g = 3.19) correlating with the main full-field Zeeman response points to the presence of spin S³ 1 N agglomerates. The overall signal properties indicate that only ~ 77% of the N acceptors occur as isolated decoupled dopants. Careful monitoring of the total ESR signal intensity over a broad T range unveils the N dopant as a shallow acceptor, making it a promising candidate for stable substitutional p-type doping in MoS2-based novel nanoelectronic devices.
Finally, dealing with one more observation on p-type geo 2H MoS2 crystals, we address a previously unreported ESR signal characterized by g// = 2.102, g2 ~1.998 and g3 ~1.973, and corresponding to a defect density (S = ½) of ~5 ×1014 cm-3. For the applied magnetic field B//c-axis, the response is comprised of a single central asymmetric Zeeman peak at zero-crossing g = 2.102(1), amid a symmetrically positioned hyperfine doublet of 6.6-G splitting. The g pattern points to a defect of lower than axial symmetry, likely orthorhombic (C2v). Based on signal composition analysis, it is ascribed to Pb impurity atoms substituting for Mo, the defect operating as acceptor dopant with estimated thermal activation energy >10 meV.
The adduced insight from ESR experiments on covalently-bonded substitutional acceptor dopants may contribute to developing appropriate and tunable p-type doping strategies for few-layer MoS2 application in future high-efficiency electronic nano-devices. The key experimental information provided, i.e., regarding advanced atomic models and a pioneering platform of experimental activation energies, may stimulate theoretical verification through carrying out first-principles calculations.
 B. Radisavljevic, A. Radenovic, J. Brivio, V. Giacometti, and A. Kis, Nat. Nanotechnol. 6, 147 (2011)
 Q.H. Wang, K. Kalantar-Zadeh, A. Kis, J.N. Coleman, and M.S. Strano, Nat. Nanotechnol. 7, 699
 R. S. Title and M. W. Shafer, Phys. Rev. B8, 615 (1973).