Physics and chemistry of MoS₂ intercalation compounds

Molybdenum disulfide is a layered diamagnetic semiconductor. Trigonal prismatic coordination exists between the Mo and the S atoms, which are primarily covalently bonded. The layers can stack to form either a hexagonal (2H), or a rhombohedral (3R) polymorph. The reaction of MoS₂ with alkali and alkaline-earth metals dissolved in liquid ammonia results in a series of metallic superconducting intercalation compounds. The alkali and alkalline-earth intercalates, A_xMoS₂ (A = Na, K, Rb, Cs, Ca, and Sr) can be separated into two groups according to their stoichiometry, structure and superconducting properties. The first group (I) consists of the Na, Ca, and Sr intercalates, which have nonhexagonal structures, less well-defined stoichiometries, small intercalate ion diameters, and superconduct at temperatures of 3.6, 4.0, and 5.6 K, respectively. The uncertainties in the stoichiometries in this group are Na_xMoS₂ (0.3 ≤ x ≤ 0.6), Ca_xMoS₂ (0.05 ≤ x ≤ 0.07), and Sr_xMoS₂ (0.06 ≤ x ≤ 0.1). The second group (II) consists of the K, Rb, and Cs intercalates which form compounds of definite stoichiometry (x {all equal to} 0.3), have a well-ordered hexagonal crystal structure, larger intercalate ion diameters, and superconduct at 6.9 K. The compound Li_xMoS₂ differs from the above groups in that NH₃ or NH₂ intercalates along with Li, and the structure exibits considerable disorder. It superconducts at 3.7 K. The electronic energy band structure of MoS₂ results from strong intralayer hybridization between metal d_z² and d_xy, d_x²-y² bands resulting in at least a 1 eV gap within the d band manifold. The lowest d sub-band is filled in MoS₂, and upon intercalation the Fermi level is raised into the upper d sub-band. The high density of states at the Fermi energy in the intercalates is believed to be the major reason for superconductivity up to 7 K. The superconducting critical magnetic fields are highly anisotropic, and are very large (greater than 20 Tesla for group II) for fields parallel to the layer planes. The critical field angular dependencies can be fitted to either a model for thin film superconductivity (group I), or to an anisotropic effective mass model (group II). An unusual positive curvature in the critical field vs. temperature plots is found in all compounds.