Two sets of naturally-occurring molybdenite samples were used in this work. The minerals, found in the mineral storage/collection at the Department of Geological Sciences at the University of Michigan, of unknown origin were employed for thermal oxidation experiments. Molybdenite samples obtained from Japan were employed for lithium-insertion measurements. Atomic force microscopy measurements were performed using a Digital Instruments AFM (NanoScope IV). The measurements were made in tapping mode. Scanning electron microscopy observations were made using two different instruments. A JEOL (model 6150) electron microscope was used to image the oxidation behavior of the molybdenite surfaces as a function of heating time. SEM imaging experiments on lithium-reacted molybdenite samples were made using a high-resolution electron microscope (Hitachi S-4700). Raman scattering (RS) spectra were measured using a Jobin-Yvon U1000 double-pass spectrometer equipped with a cooled, low-noise photomultiplier tube (ITT FW130). The incident light used for the experiments was the 515 nm Ar ion laser. TEM analysis was performed using a JEOL TEM 2010F at a 200 kV acceleration voltage. Phase and structure of the material were monitored using selected area electron diffraction (SAED). Specimens for TEM analysis were prepared by dispersing the MoS₂ sample on 3-mm Cu grid with a hole size of 1 × 2 mm. High resolution transmission electron microscopy (HRTEM) image processing including the fast Fourier transformation (FFT) was carried out using a Gatan Digital Micrograph 3.4.

Lithium intercalated Li_xMoS₂ samples ware prepared from natural 2H-MoS₂ treated with dilute butyllithium in hexane in a controlled water and oxygen-free environment. Prolonged treatment (~10 days) with very dilute solutions (~0.005 M) was used to produce the lithium-inserted samples for analytical experiments. On cleaving the surface was seen to be broken into mm sized regions of crystal, either shiny, or matt black (poor surface quality) indicating the intercalated areas.

Thermal oxidation experiments were performed in air. Molybdenite surfaces were heated to 400°C in a furnace. Various analytical measurements were performed as a function of heating time to understand the thermally-induced effects. The ion-beam experiments namely Rutherford backscattering spectrometry (RBS) and nuclear reaction analysis (NRA) measurements were carried out in the accelerator facility at the Environmental Molecular Sciences Laboratory (EMSL) of the Pacific Northwest National Laboratory (PNNL), Richland, Washington, USA. A helium (He⁺) ion beam of 2 MeV was incident on the sample surface at near normal and the scattered ions were detected at an angle of 170° from the sample normal for RBS measurements. A beam of 0.94 MeV deuterium (d⁺) ions was incident on the molybdenite sample and the reaction products were detected at a scattering angle of 170° from the sample normal for NRA measurements. A thin aluminized mylar film covered the detector to stop backscattered deuterium ions, allowing only the more energetic reaction products to enter the detector. The detected particles for these measurements were protons from the ¹⁶O(d,p)¹⁷O reaction.

The AFM image of a representative pristine natural MoS₂ sample surface is shown in Fig. 2. Molybdenite surfaces are often atomically flat over relatively large areas as seen in this work (Fig. 2) and other reports 141516. AFM images, in the area of scanning, show no specific surface features such as steps or kinks. The electron diffraction pattern of the molybdenite crystal is shown in (Fig. 3). The hexagonal lattice of the 2H polytype is evident from the image (Fig. 3).

The long-wavelength lattice vibrations of molybdenite are interesting and often provide a model system to understand the linear-chain model of the layered-type chalcogenide minerals. Group theory predicts two infrared (IR) and four Raman-active modes for 2H-MoS₂ 161718. The Raman scattering spectrum of the pristine molybdenite surface recorded at ambient temperature is shown in Fig. 4 along with the associated mode assignments. The A_1g mode at 407 cm^-1 is an intralayer mode involving the motion of S atoms along the c axis. The E_g mode at 383 cm^-1 is an intralayer vibrational mode involving motion of Mo+S atoms in the basal plane. The peak at 286 cm^-1 with weak intensity and E_1g symmetry is observed, which involves S atoms in the basal plane. The spectrum of the pure 2H-MoS₂ exhibits a rigid-layer (RL) mode E_g at 32 cm^-1. This mode is of interlayer type involving rigid motion of neighboring sandwiches in anti-phase. In layered crystals, the frequency of the RL mode provides direct information on the strength of the interlayer forces in these crystals, since for such RL motions the restoring forces are provided entirely by layer-layer interactions 19.

Figure 5 shows the SEM and electron diffraction micrographs of Li-inserted molybdenite produced by inserting the samples in n-butyl-lithium. It can be seen that some defects are created near the edges and in the steps of the specimen, after 0.2 hours of insertion (arrows in Fig. 5a). After 2 hours of insertion reaction, superlattice spots appear (label s in Fig. 5b). Notice also the splitting of the main spots denoted by the letter m. The micrograph taken from the same area reveals that the specimen is heavily disordered or contains a significant number of defects as a result of metal-insertion reaction in the molybdenite structure (Fig. 5c). The distribution of lithium vs. distance from the edges of the specimen has been determined by tracer studies using solid-state nuclear track detectors (SSNTD). A typical SSNTD image of Li_0.3MoS₂ is also shown in Fig. 5 (Fig. 5d). Continued insertion reaction for 20 hours results in the appearance of superlattice spots in areas far from the edges. Splitting of the (101¯0 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacPC6xNi=xH8viVGI8Gi=hEeeu0xXdbba9frFj0xb9qqpG0dXdb9aspeI8k8fiI+fsY=rqGqVepae9pg0db9vqaiVgFr0xfr=xfr=xc9adbaqaaeGaciGaaiaabeqaaeqabiWaaaGcbaGaeGymaeJaeGimaaZaa0aaaeaacqaIXaqmaaGaeGimaadaaa@2FA2@) spots still occurs, the inner spots being very intense while the outer ones are faint as shown in Fig. 5 (Fig. 5e). This is an indication that the intercalated system did not reach saturation, even locally, although the diffusion of lithium has progressed, implying large variations in the diffusivity of lithium in molybdenite.

The results of the lithium insertion reaction and associated structural modifications are discussed as follows. Insertion of foreign metal and/or ionic species into a host structure implies two main conditions: (1) the structural ability to accept ions in empty sites, and (2) the presence of acceptor levels in the electronic structure. From a crystal structure point of view, there are several possible locations to insert lithium into molybdenite (Fig. 6) as the unit cell of 2H_b-MX₂ possesses tetrahedral and octahedral inter-sandwich sites (this is equally applicable for inserting any other metal ions into the structure). Among the group-VI TMDs, MoS₂ is a typical example where insertion reactions induce a local structure modification. In this particular case, Mo presents a trigonal prismatic (TP) S coordination which changes to the octahedral (Oh) one, i.e. the TP → Oh transition 3202122. The structure modification is accompanied by an increase in the Mo-S bond ionicity in agreement with the respective stability of the new atomic arrangement, the Coulomb repulsion between partially charged ligands favoring the octahedral form. Also, comparison of the d-band density of states for 2H-MoS₂ and hypothetical 1T-LiMoS₂ shows that the occupied bands which contain six states are lower in the case of the Oh phase corresponding to the glide process between Mo and S atoms. This is a good example of destabilization through lithium reduction. The transformation from TP to Oh coordination is attributed to a process which is driven by a lowering of the electronic energy for the octahedral structure when the charge transfer occurs from Li ions to the MoS₂ layer (electrons are donated by Li) upon insertion reaction 310.

The NRA results are shown in Fig. 7. The curves are shown as a function of heating time of molybdenite surfaces in air. The molybdenite surface, introduced into the high-vacuum chamber immediately after cleavage, did not show any peak corresponding to oxygen. This observation indicates that there is no presence of oxygen on the pristine molybdenite surface before the thermal oxidation experiments. We claim that we start our thermal oxidation experiments with atomically clean molybdenite surfaces since the detection limit of NRA to determine the oxygen concentration (using the ¹⁶O(d,p)O¹⁷ reaction) is ~3 × 10¹⁵ atoms/cm². The evolution of the oxygen peak as a function of heating time is evident in the NRA spectra. The peak intensity increases with the increase in time of heating, which indicates progressive thermal oxidation of the molybdenite surfaces. However, the increase in oxygen peak intensity is somewhat low during the first hour of heating. This observation indicates that the oxidation during the first hour of heating is slow leading to oxidation only on the surface. A significant increase in peak intensity and change in shape (becoming more broad) with further heating after the first hour indicates that oxidation is progressively proceeding from the surface layers into the deeper layers.

SEM micrographs of the molybdenite surfaces are shown in Fig. 8 as a function of heating time. The thermally-induced oxidation completely changes the morphology of the molybdenite surface. The SEM images indicate the formation of small oxide crystals on the surface within the first hour of heating. More significant changes are observed after 2 to 3 hours of heating time. The crystals formed are light greenish indicating the oxide phases. The number of the oxide crystals formed on the surface increases significantly as a result of thermal oxidation. The crystals formed on the molybdenite surface as a result of thermal oxidation are assuming a regular shape with an elongated thin structure, which is in good agreement with the shapes reported for Mo oxides 23. Oxide crystals formed on the surface as a result thermally-induced oxidation during the first one hour of heating are, however, longer in size extending over 4–5 microns in length and about 2 microns in width. The SEM image shown for samples after 3 hours of heating clearly shows the density of such crystals formed on the surface increases significantly. The increase in the number of crystals at 3 hours of heating increases the surface roughness with random distribution of the crystals in addition to the irregular shapes. However, an interesting phenomenon is the size of the crystals. It can be seen that the crystals formed at 3 hours of heating are smaller in size when compared to the ones formed in the first hour of heating. It may be that the tendency of all crystals to grow within the limited surface area may inhibit the growth of larger crystals in the neighboring medium. The accumulation of several crystals contributing to the rough morphology is also observed in some areas, which supports the idea that the higher density of oxide crystals may inhibit the growth of large and uniform crystals as seen at the initial stages of oxidation. In other words, the formation of continuous oxide layers, initially formed on the top-most surface layers, may influence the lateral diffusion of the particles to join the already formed crystals and to grow into larger crystals.