Nonisobaric diffusion

Characterization of Nonisobaric Diffusion Due to Nonequimolar Fluxes in Catalyst Particles... 

It is in the non-pulse component (N ), which initially develops a peak flux which becomes negative Before finally approaching zero, that the effect of the pressure gradient is seen. A positive flux in Ng would not be predicted by the isobaric equations, since it must be equal and opposite to the positive He flux. The explanation is apparent by examining the total pressure gradient, where it can be seen that bulk motion of the fluid can result in a net positive flux of Ng. Similar results are obtained with longer pulses, and for these cases the differences between isobaric and nonisobaric diffusivity becomes larger. 

Eq. (8.8-5) or (8.8-6) is the basic equation for the case of diffusion under nonisobaric conditions. Alternative derivation of these two equations is given in Appendix 8.7. Knowing the individual fluxes obtained from eq.(8.8-6), the total flux Nt and the diffusive fluxes J are given by ... 

---