DFT-computed formation energies at 0 K of Ti1-xMoxS2 configurations within prismatic (H) and octahedral (T) hosts.
The system shows a phase separating behaviour: all the configurations in the H host lie above the mixing line (black solid), while the energy gain coming from Jhan-Teller distortion arising the T host is not enough to stabilise the doped configuration with respect to the phase separating solution H-MoS2 + T-TiS2. The grey dot are the configuration predicted by a Cluster Expansion model that we will use to investigate the thermodynamics of the system at finite temperature.
I applied non-equilibrium statistical physics to the simplest low-order model in nanoscale friction: The Prandtl-Tomlinson model. We are able to calculate velocity dependent heat produced and heat dissipated for this model using expressions of entropy produced and entropy flow from non-equilibrium statistical mechanics.
Expected values of friction, expected work, difference between work and heat produced, and difference between heat produced and heat dissipated, for velocities 1 nm/s - 100 nm/s, and temperatures 100, 200, 300, 400 K, in the 1D thermal Prandtl-Tomlinson model with gamma = 0.06, C = 1.8 N/m, E0 = 0.3 eV, a = 0.25 nm, and f0 = 19.5 Hz. Quantities are given as time averages calculated in the steady-state and over a few stick-slip cycles. The relation friction-velocity is logarithmic. Work, on the other hand, is linear with velocity. Thermolubricity is observed as a reduction of friction of about 1 pN every 100 K, and a decrease of the slope of the work per unit of time versus velocity of about 0.5 eV/s every 100 K, in the range of temperatures, and velocities considered. It is expected that all work after a cycle is converted into heat produced, and all the heat produced into heat flow, if the energy of the system returns to the same state at the end of the stick-slip event. Discrepancies are at most of 0.5 %, which can be attributed to 1) numerical errors during the integration of the Master Equation, 2) small errors when finding the boundaries of a stick-slip event. Then for the Prandtl-Tomlinson model, the heat dissipated by the frictional system can be calculated as the integral of the force, however this is not the general case when the system has internal degrees of freedom and therefore, after each event, the expected heat dissipated may be lower than the expected heat produced and the expected work.
The figure shows comparison of frictional behaviour (friction force vs. load) of MoS2 monolayer and WS2 multilayer (3 layers) samples. The samples were deposited by chemical vapour deposition on Si/SiO2 substrates. Friction measurements were performed by friction force microscopy using the same Nanosensors™ PPP-CONT probe. The normal force constant was calibrated using thermal noise calibration method built-in the system and lateral forces were calibrated according to Wedge calibration method (Ogletree et al., 1996).
Both measurements have been fit to Hertz-plus-offset model (Macro-scale elastic contact model, which accounts for adhesion and accommodates for uncertainty during pull-off force measurement) (Schwarz et al., 1997). We get a very good fit for monolayer (R2 = 0.99), and bilayer (not shown on this figure, R2 = 0.98), the model fails to describe friction of multilayer sample, which means that the multilayer sample experiences different frictional mechanism. The possible causes for the deviation from the model could be: (i) due to higher thickness the layers may experience different elastic properties, which could mean that a different adhesion regime is present, (ii) tip shape effects (nano-scale wear may have altered the shape of the tip), (iii) plastic deformation of the contact (less likely as plastic deformation would show up on a topography scan), or (iv) puckering effect and sliding between the layers.
The X-ray Diffraction pattern shows the amorphous nature of the coating. The peaks found are of steel substrate and no peaks were seen based on the coating. The Mo-S-N coating was prepared using DC magnetron sputtering. As per the literature, The amorphous nature of TMD coatings were obtained only after 30 atomic % of nitrogen doping on a DC magnetron sputtering. Here, Due to usage of high ion energy and ion bombardment on Industrial scale machine at IREIS, It was made possible to obtain amorphous nature of Mo-S-N coating. The coating also reported 0.2 as coefficient of friction in humid air. The tribological study in vacuum is yet to be done. The coating obtained 2.9 GPa hardness. Still more works in terms of hardness improvement needs to be done. Literature study shows hardness reaching up to 7 GPa.
In the following picture, visualized with VMD, we investigate the complex interaction between a MoS2 layer and an octane molecule (used as a model for conventional lubricants) for different geometrical configurations, using DFT.
In the following picture, visualized with VMD, we use MD to expand our system to a larger scale and explore the physicochemical layer (MoS2)/liquid (octane) interactions, like adsorption.
Nanotoxicity Assay: Evaluation of ROS production in A549 cells with samples of MoS2 and WS2 obtained from ACS Material, LLC 959 E Walnut Street #100, Pasadena, CA 91106, www.acsmaterial.com).
Nanotoxicity Assay: Evaluation of Cell Viability using the Neutral Red Assay. Impact on A549 cells of exposure to micro and nano-MoS2 and WS2 samples (ACS Material, LLC 959 E Walnut Street #100, Pasadena, CA 91106, www.acsmaterial.com).
Discussion: The results of the variety of assays conducted indicate the excellent biocompatibility of MoS2 and WS2 toward mammalian tissue cells. The ROS generation tests showed that all the nanomaterials have far lower levels of ROS compared to H2O2, a known ROS generator. The Neutral Red viability assay showed a non-toxic effect for all the concentrations tested. In conclusion, we found that the commercial samples of graphene-like materials tested did not trigger the generation of high levels of ROS. In addition, no significate reduction on the number of viable cells were observed during an incubation time of 24 hours with the samples. These results of our investigation confirm the high biocompatible behaviour of 2D TMDs and enhance their potential for their use in biomedical applications.
Pure TMD films are porous but addition of carbon increases the compactness of the films. C doped MoSe2 films were deposited by DC magnetron sputtering of two individual targets. It can be seen in the figure, that the addition of carbon led to compactness and densification of the porous MoSe2 films (Fig. 1a). The deposited composite coatings are x-ray amorphous and did not show any contribution of TMD nanocrystals (Fig. 1b). A set of coatings having 50 at. % C were also deposited with substrate bias. Mo-Se-C films whether deposited with substrate bias or without, show similar patterns. Application of bias led to decrease in the Se/Mo ratio and increase in hardness (Fig. 1c). 90 V bias film showed maximum hardness of 4.9 GPa which is more than same coatings deposited without bias (2.7 GPa for 50 at. % C coating) and also the values reported in literature. Coatings show excellent tribological properties in ambient atmosphere with an average friction coefficient around 0.06 (Fig. 1d).
In the figure, Scanning Electron Micrographs from the thin films deposited by radio-frequency magnetrons sputtering are reported. a-d panels represent cross sections of thin films deposited by 3 different methods and the difference in the microstructure can be depicted ( a) and c) represent dense featureless microstructures while b represent more porous and columnar structure) d-f represent top views of the films presented at a-b respectively.
We investigated the atomic scale tribological properties of transition metal dichalcogenides (TMDs), using ab-initio techniques. Such compounds are formed by triatomic layers with MX2 stoichiometry (M: transition metal cation, X: chalcogen anion) held together by van der Waals forces.
We considered 6 prototypical MX2 TMDs (M=Mo, W; X=S, Se, Te) with hexagonal P63/mmc symmetry, focusing on how specific phonon modes contribute to their intrinsic friction.
Within the DFT framework, we described the exchange-correlation interaction energy by means of the PBE functional, including long range dispersion interactions in the Grimme formulation (DFT-D3 van der Waals).
We identified and disentangled the electro-structural features that determine the intra- and inter-layer motions affecting the intrinsic friction by means of electro-structural descriptors such as orbital polarization, bond covalency and cophonicity.
We show how the phonon modes affecting the intrinsic friction can be adjusted by means of an external electrostatic field. In this way, the electric field turns out to be a knob to control the intrinsic friction.
The presented outcomes are a step forward in the development of layer exfoliation and manipulation methods, which are fundamental for the production of TMD-based optoelectronical devices and nanoelectromechanical systems.
In this figure we present the results of sliding crystalline anti-parallel bilayer MoS2 in different sliding directions ranging from 0 to 60 degrees. In other words, we slide the top mobile layer over the static bottom layer while keeping the relative orientation between the two layers the same. The goal of the study was to find a possible sliding directional-dependent relation for the superlubricity of commensurate MoS2. In this figure we display the coefficient of friction as a function of sliding angle. We can distinguish two different friction regimes, one regime with higher coefficients of friction for when the sliding direction is oriented along the bonds accompanied with an observed stick-slip behavior during the sliding and a superlubricity regime for when we do not slide along the bonds and we find a smooth dynamical behavior. The main finding is that by using Molecular Dynamics I have shown that in contrast to the current understanding not only incommensurability can lead to the super low friction regime of MoS2 but also controlling the sliding direction of commensurate MoS2 can lead to the super low friction behavior. Some result.
The difference on performance by applying a solid lubricant coating can be easily seen in the following picture. It illustrates the profile of the wear track for two different couple of rings; silicon carbide substrate without coating and silicon carbide substrate with a diamond-like carbon coating. The test was conducted in a disc-on-disc rig under dry nitrogen environment during 4 hours. The wear volume of the uncoated pair was 2.08 x 109 µm3 while for the DLC case was two order of magnitude less, 3.63 x 107 µm3. According to the coefficient of friction it was 0.45 and 0.38 for the uncoated and coated pair, respectively.
On the Figures from left to right Ni‒P pure alloy coating and Ni‒P/WS2 composites plated from electrolytic baths containing 0.25, 0.5, 0.75 and 7 g L-1 of the WS2 particles, respectively are depicted. According to EDS analyses results, phosphorus content is around 4 wt.% and 8 wt.% for the two utilized baths containing 2 and 5 g L-1 of H3PO3 as a phosphorous source. Tungsten content plateaus around 6 wt.% for WS2 concentration of 0.75 g L-1 . The microhardness of Ni‒P deposits is around 800 HV from both baths. Composites are delamination prone and possess lower microhardness compared to the pure Ni‒P ones which is approximately 740 HV.
In the pictures above, Ni‒P/WS2 deposits’ microstructure is depicted, in the case of the addition of 0.075 g/particle g of surfactants SDS (up) and CTAB (down).
We have successfully grown few layer MoS2 on AFM tips for tribological studies. Optical images shown here represent an AFM tip covered with MoS2. The corresponding Raman spectra reveal the few-layer thickness of the films.