Model studies on phase separation at a capillary orifice 1

Abstract

Phase separation between red cells and plasma were studied using blood perfused model systems which consisted of glass capillaries (I.D. between 3.3 and 11.0 μm) branching from a large bore (I.D. 1.5 mm) feed channel. The orifice of the capillaries was either positioned in the centerline of the feed channel (model A) or flush with its wall (model B). Capillary discharge hematocrit (H_D) was measured at different flow conditions and related to the hematocrit in the feed channel (H_F). The ratio H_D/H_F was found to be correlated to the ratio between wall shear stress in the capillary and in the feed channel ( ${\mathit{\tau }}_{\mathit{T}}\mathrm{/}{\mathit{\tau }}_{\mathit{F}}$ ). While at low ${\mathit{\tau }}_{\mathit{T}}\mathrm{/}{\mathit{\tau }}_{\mathit{F}}$ almost no cells entered the capillary, H_D increased with increasing ${\mathit{\tau }}_{\mathit{T}}\mathrm{/}{\mathit{\tau }}_{\mathit{F}}$ and approached H_F in model A. In model B, H_D did not reach H_F within the experimental range of ${\mathit{\tau }}_{\mathit{T}}\mathrm{/}{\mathit{\tau }}_{\mathit{F}}$ ’ The flow fractionation in model B can be explained by “plasma skimming”. In model A cell entry into the capillary is determined by the relationship between forces acting on the red cell either in the direction of the feed channel or of the capillary. For this mechanism the term “screening effect” is proposed. This effect decreases with increasing H_F and increasing capillary diameter and vanishes when capillary diameter approaches 12 μm.