Intercalation of alkali atoms within the lamellar transition metal dichalcogenides is a possible route toward a new generation of batteries. It is also a way to induce structural phase transitions authorizing the realization of optical and electrical switches in this class of materials. The process of intercalation has been mostly studied in three-dimensional dichalcogenide films. Here, we address the case of a single-layer of molybdenum disulfide (MoS 2 ), deposited on a gold substrate, and intercalated with cesium (Cs) in ultra-clean conditions (ultrahigh vacuum). We show that intercalation decouples MoS 2 from its substrate. We reveal electron transfer from Cs to MoS 2 , relative changes in the energy of the valence band maxima, and electronic disorder induced by structural disorder in the intercalated Cs layer. Besides, we find an abnormal lattice expansion of MoS 2 , which we relate to immediate vicinity of Cs. Intercalation is thermally activated, and so is the reverse process of de-intercalation. Our work opens the route to a microscopic understanding of a process of relevance in several possible future technologies, and shows a way to manipulate the properties of two-dimensional dichalcogenides by "under-cover" functionalization.tion is thermally activated, being completed after few tens of minutes at a temperature of 550 K. Above 850 K, deintercalation is efficient and completed within a few tens of minutes.Intercalated cesium forms a a Cs monolayer with an ill-ordered structure compatible with a ( √ 3 × √ 3)R30 • reconstruction with respect to Au(111). We reveal electron transfer from Cs to MoS 2 , modifications of the relative positions of the valence band maxima in MoS 2 , and electronic disorder induced by structural disorder in the intercalated layer. Upon intercalation, MoS 2 is lifted, and adopts an unusually large lattice parameter. Our analysis combines scanning tunneling microscopy (STM), reflection high-energy electron diffraction (RHEED), grazing incidence X-ray diffraction (GIXRD), reflectivity (XRR), X-ray photoelectron spectroscopy (XPS), and angle-resolved photoemission spectroscopy (ARPES) all performed under ultrahigh vacuum, in some cases in operando during intercalation. Further insights are brought by density functional theory (DFT) calculations.
MethodsThree ultrahigh vacuum systems were used for our experiments. A first one is coupled to the X-ray synchrotron beam delivered at the BM32 beamline of the ESRF. It has a base pressure of 3×10 −10 mbar and is equipped with a quartz micro-balance and a RHEED apparatus. The second one, at Institut Néel (Grenoble), with a base pressure of 2×10 −10 mbar, is part of a larger ultrahigh vacuum system comprising a STM, a RHEED apparatus, and a quartz microbalance. The samples were prepared in each system before being investigated by RHEED, STM, GIXRD, and XRR. Temperatures were measured with a pyrometer in both systems. Note that the pyrometers and the chamber configurations are different in the two systems, which implies a plausible variability...