THERMOPHORETICALLY AUGMENTED MASS-, MOMENTUM- AND ENERGY- TRANSFER RATES IN HIGH PARTICLE MASS-LOADED LAMINAR FORCED CONVECTION SYSTEMS

In all previous treatments of thermophoretically-modified aerosol particle transport, even those which attempt to allow for variable host gas properties, it has been explicitly (or implicitly) assumed that the particle mass fraction is small enough to neglect the influence of the suspended particles on the host gas momentum density- and energy density-fields. However, in high intensity material processing applications (eg. optical waveguide (OWG) preform deposition from silica mist-laden combustion products) particle mass loadings often exceed 1/3, and the thermophoretically enhanced particle mass deposition flux itself modifies the local mixture velocity and temperature fields in the vicinity of the deposition surface. A self-consistent pseudo- single-phase mixture ("diffusion") approximation which exploits the fact that the volume fraction of suspended particles is negligible even when the particle mass fraction is quite near unity is introduced here to calculate the fully coupled problem of mass-, energy- and momentum diffusion for laminar boundary (LBL) flow of a combustion gas mixture containing submicron particles of appreciable thermophoretic diffusivity but negligible Brownian diffusion. It is shown that high particle mass loading systematically increase the wall fluxes of momentum (shear stress), heat and particle mass, much like those effects associated with 'massive suction' in single-phase LBL-theory. Indeed, a simple rational correlation approach is introduced to capture the essential interactions leading to higher transport coefficients (C{dollar}sb{lcub}rm f{rcub}{dollar},St{dollar}sb{lcub}rm h{rcub}{dollar}, St{dollar}sb{lcub}rm m{rcub}{dollar}) in such high mass-loaded aerosol systems. The transfer coefficients we obtain from our 'exact' numerical integrations of the self-similar LBL equations can be used to estimate LBL-development and associated particle mass fluxes to surfaces of 'arbitrary' shape. Moreover, this approach to high particle mass-loaded materials processing applications is extended to rationally predict the partitioning of dopants (eg. Ge,B,P dots) between the vapor and particle phases and, hence, the rather complex connection between deposit (OWG preform) and process fed compositions.