Gas permeation through graphene membranes has received considerable attention for water purification and molecular sieving applications. However, characterization of the permeation has been limited to long timescales of minutes. This thesis shows a method for measuring gas permeation through porous graphene membranes at the microsecond timescale. Suspended porous graphene membranes, with an average pore size of 14 nm and a single 400 nm pore, are brought into sinusoidal motion by optothermal actuation. By monitoring the frequency dependent phase delay between actuation signal and mechanical motion, the gas dependent permeation time of the porous membrane is determined. The permeation time constant is demonstrated to be proportional to the square root of the molecular mass, indicating an effusion dominated permeation mechanism. The determination of permeation at timescales below 1 ms using a femtoliter gas cavity opens up opportunities for novel nanoscale porous graphene based gas sensors, with very fast response times.

Diffusivities and lithium distributions in Li₆PS₅X (X = Cl, Br, I) and Li₇PX₆ (X = S, Se) are investigated by means of molecular dynamics simulations. The best lithium conduction is found in Li₆PS₅Cl with diffusivity of over 1 S/cm at 600 K. The halide rich Li₅PS₄Cl₂ shows promising results with diffusivity of over 2 S/cm at 600 K, performing better in simulations than the existing compounds.

Simulations show a beneficent effect of chlorine and bromine disorder on anion sites, opening lithium pathways in the neighborhood and enabling higher conductivities. No significant influence of vacancies on the diffusivity of lithium in these materials can be reported. Lithium jump rates in both material families are in the order of 10¹⁰ s^-1, while Li₆PS₅Cl and Li₅PS₄Cl₂ show the highest jump rates of respectively 2.10·10¹¹ s^-1 and 2.14·10¹¹ s^-1 at 600 K.