Adsorption Mechanism and Characteristic Temperatures of the Monolayer Adsorption of CO2 on Graphite: The Role of Graphene Dimensions.

Although simulation results for gaseous adsorption on a surface of infinite extent, modeled with periodic conditions at the boundaries of the simulation box, agree with experimental data at high temperatures, simulated isotherms at temperatures below the triple point temperature show unphysical substeps because of the compromise of interactions within the box and interactions between the box and its mirror image boxes. This has been alleviated with surfaces of finite dimensions (Loi, Q. K.; Colloids Surf., A 2021, 622, 126690 and Castaño Plaza, O.; Langmuir 2023, 39 (21), 7456-7468) to account for free boundaries at the adsorbate patch on the surface, and the critical parameter of this model substrate is the size of the finite surface. If it is too small, the adsorbate patch does not model the physical reality; however, if it is too large, the computation time is excessive, making the simulation impractical. In this study, we used carbon dioxide/graphite as the model system to explore the effects of finite dimensions on the description of experimental data of Terlain, A.; Larher, Y. Surf. Sci. 1983, 125 (1), 304-311, especially for temperatures below the bulk triple point temperature. With the appropriate choice of graphene size, we derived the 2D triple point and 2D critical point temperatures of the monolayer, and most importantly, for temperatures below the 2D critical point temperature, the adsorption mechanism for the formation of the monolayer is due to the interplay between the boundary growth process and the vacancy filling. The extent of this interplay is found to depend on the fractional coverage of the surface.

Effects of a Free Adsorbate Boundary on the Description of an Argon Adsorbed Film on Graphite below the Bulk Triple Point.

Monte Carlo simulations have been carried out to study argon adsorption on graphite at temperatures below the bulk triple point temperature, Ttr(bulk) = 83.8 K. Two models for graphite have been used to investigate the effects of an adsorbate patch with a free boundary on the layering temperatures, the two-dimensional (2D)-triple point and the 2D-critical point for the three adsorbate layers on the surface. The first model (S-model) has a planar surface of infinite extent in the two directions parallel to the surface, and the second is a finite (2D-patch model). Although simulations using both models describe the characteristic temperatures, only the 2D-patch model can represent the experimental isotherms accurately, and the condensation pressures at which first-order transitions occur, while simulations with the S-model yield many unphysical substeps that are not observed experimentally in the first layer adsorbate, which leads to a poor description of higher adsorbate layers. These results support the interpretation that boundary growth of an adsorbate patch is the mechanism for argon adsorption at temperatures below the bulk triple point temperature. Combining the results derived from this simulation study for temperatures below the bulk triple point temperature, with results reported in the literature for temperatures above Ttr(bulk) and experimental data, we have constructed a generic pattern for the adsorption isotherms of simple gases on graphite at temperatures ranging from well below the bulk triple point temperature up to the bulk critical temperature, a comprehensive description not widely recognized in the literature.

Characterization of non-graphitized carbon blacks: a model with surface crevices.

Experimental isotherms for argon and nitrogen adsorption on two non-graphitized carbon substrates, Carbopack B and Cabot BP280, do not obey Henry's Law in the range of pressures accessible to the most sensitive MKS pressure transducers. At high pressures, close to the bulk coexistence pressure (P0), the isotherms at temperatures below the bulk triple point temperature cross the P0 axis at a finite loading, a behaviour which is interpreted as incomplete wetting. It was found that the adsorbed density at P0 for Cabot BP280 is lower than that for Carbopack B which is, in turn, only slightly lower than that for the highly graphitized Carbopack F, suggesting that there is a long-range effect of the surface structure in non-graphitized carbon blacks, in the accumulation of higher layers, especially for Cabot BP280. We have carried out extensive Monte Carlo simulations to compare experimental observations with a molecular model for substrate surfaces decorated with crevices of molecular dimensions. From the analysis of the experimental data, it was found that the typical width of crevices is of the order of 0.65-0.9 nm. In the high pressure region, the crossing of the P0 axis by isotherms at temperatures below the bulk triple point temperature can be explained by an adsorbate structure which is less dense and more disordered than the fcc structure of the bulk crystal, with a consequent raising of the coexistence pressure between the adsorbate and the gas phase above P0. Adsorbate loading at the point where the isotherm crosses the P0 axis for Cabot BP280 is lower than for Carbopack B which can be attributed to a higher concentration of crevices leading to a lower adsorbate density and an irregular arrangement of atoms at the interface separating the adsorbed phase and the gas phase. This results in weaker gas-adsorbate interactions which supresses the build-up of higher layers. We suggest that the use of the adsorbed density at the bulk coexistence pressure, at temperatures below the bulk triple point temperature, can be a useful tool for assessing the presence and concentration of surface crevices on non-graphitized carbon black.

On the canonical isotherms for bulk fluid, surface adsorption and adsorption in pores: A common thread.

Kinetic Monte Carlo simulated isotherms calculated in the canonical ensemble, at temperatures below the critical temperature, for bulk fluid, surface adsorption and adsorption in a confined space, show a van der Waals (vdW) loop with a vertical phase transition between the rarefied and dense spinodal points at the co-existence chemical potential, µco. Microscopic examination of the state points on this loop reveals features that are common to these systems. At state points with chemical potentials greater than µco the microscopic configurations show clusters, which coalesce to form two co-existing phases along the vertical section of the loop (the coexistence line). As more molecules are added, the dense region expands at the expense of the rarefied region, to the point where the rarefied region becomes spherical (cylindrical for 2D-systems) with a curvature greater than that of the coexisting phases. This results in a decrease of chemical potential from µco to the liquid spinodal point where the rarefied region disappears. With a further increase in loading, the chemical potential and the density increase. The existence of a vdW loop is the microscopic reason for the hysteresis observed in the grand canonical isotherm, where the adsorption and desorption boundaries of the hysteresis loop are first-order transitions, enclosing the vertical section of the vdW loop of the canonical isotherm. However, a first-order transition is rarely observed in experiments where transitions are usually steep, but not vertical. From our extensive simulations, we provide two possible reasons: (1) the finite extent of the system and (2) the existence of high energy sites that localize the clusters. In the first case, the desorption branch, and in the second case the adsorption branch, either comes close to, or collapses onto the coexistence line. When both occur, the hysteresis loop disappears and the isotherm is reversible, as often observed experimentally.

Nonwetting/Prewetting/Wetting Transition of Ammonia on Graphite.

Simulations of ammonia adsorption on graphite were carried out over a range of temperatures to investigate the transition from nonwetting to wetting. The process is governed by a subtle interplay between the various interactions in the system and the temperature. At temperatures below the bulk triple point, the system is nonwetting; above the triple point, we observed continuous wetting, preceded by a prewetting region in which the so-called thin-to-thick film transition occurs. This system serves as an excellent example of wetting/nonwetting behavior in an associating fluid as a function of temperature because the heat of sublimation (or condensation) is greater than the isosteric heat of adsorption at zero loading. The nonwetting-to-wetting transition (NW/W) is also strongly affected by the adsorbate-adsorbate interaction, which becomes important when this contribution to the isosteric heat is of a similar magnitude to the heat of condensation. An appropriate indicator of a NW/W transition at a given loading is therefore the difference between the isosteric heat and the heat of sublimation (or condensation). Our simulation results show the "thin-to-thick" film transition in the temperature range between 195 and 240 K, which has not been previously explained. Above 240 K, continuous wetting occurs. This study provides a basis for a better understanding of adsorption in a range of systems because ammonia is an intermediate between simple molecules, such as argon, and strongly associating fluids, such as water.

The role of adsorbate size on adsorption of Ne and Xe on graphite.

We have carried out an extensive grand canonical Monte Carlo simulation to investigate the adsorption of neon and xenon on graphite. The adsorbate collision diameters of neon and xenon are smaller and greater respectively, than the commensurate graphite lattice spacing λ=3×3R300 of 0.426 nm. Simulated isotherms and isosteric heats were obtained using a graphite model that has been shown to describe successfully the adsorbate transitions for krypton, methane and nitrogen by Prasetyo et al. (2017), which have collision diameters close to λ. Neon does not exhibit commensurate (C) packing because the gain in the intermolecular potential interactions in the incommensurate (IC) packing when molecules move away from carbon hexagon centres, does not compensate for the increase in the solid-fluid potential energy. Xenon, on the other hand, exhibits IC packing because its molecular size is greater than λ. Nevertheless, at a sufficiently high chemical potential, the first layer of xenon changes from the IC to C packing (in contrast to what is observed for krypton, nitrogen and methane). This transition occurs because the decrease in the xenon intermolecular interactions is sufficiently compensated by the increase in the solid-fluid interaction, and the increase in the fluid-fluid interactions between molecules in the first layer and those in the second layer. This finding is supported by the X-ray diffraction study by Mowforth et al. (1986) and Morishige et al. (1990).