TY - JOUR
T1 - Novel spacers for mass transfer enhancement in membrane separations
AU - Li, F.
AU - Meindersma, G.W.
AU - de Haan, A.B.
AU - Reith, T.
PY - 2005
Y1 - 2005
N2 - The optimal flow pattern for mass transfer enhancement in spacer-filled channels is characterized by the coexistence of transversal and longitudinal vortices in the flow close to the channel walls and minimal cross-flow power consumption in the middle of the channel. The mass transfer enhancement of spacers with modified filaments, twisted tapes and multi-layer structures, which were expected to generate these flow patterns, was investigated experimentally. The results indicate that the performance of spacers with modified filaments and twisted tapes is generally worse while the performance of spacers with multi-layer structure is generally better than that of the optimal non-woven net spacer. An optimal multi-layer spacer was designed with optimal non-woven nets in the outer layers and twisted tapes in the middle-layer. Its average Sherwood number is about 30% higher than the Sherwood number of the optimal non-woven spacer at the same cross-flow power consumption whereas the cross-flow power consumption is only about 40% of the consumption of the optimal non-woven spacer at the same Sherwood number. The Reynolds number based on the height of a spacer-filled channel varies from 40 to 500 in present study.
AB - The optimal flow pattern for mass transfer enhancement in spacer-filled channels is characterized by the coexistence of transversal and longitudinal vortices in the flow close to the channel walls and minimal cross-flow power consumption in the middle of the channel. The mass transfer enhancement of spacers with modified filaments, twisted tapes and multi-layer structures, which were expected to generate these flow patterns, was investigated experimentally. The results indicate that the performance of spacers with modified filaments and twisted tapes is generally worse while the performance of spacers with multi-layer structure is generally better than that of the optimal non-woven net spacer. An optimal multi-layer spacer was designed with optimal non-woven nets in the outer layers and twisted tapes in the middle-layer. Its average Sherwood number is about 30% higher than the Sherwood number of the optimal non-woven spacer at the same cross-flow power consumption whereas the cross-flow power consumption is only about 40% of the consumption of the optimal non-woven spacer at the same Sherwood number. The Reynolds number based on the height of a spacer-filled channel varies from 40 to 500 in present study.
U2 - 10.1016/j.memsci.2004.12.019
DO - 10.1016/j.memsci.2004.12.019
M3 - Article
SN - 0376-7388
VL - 253
SP - 1
EP - 12
JO - Journal of membrane science
JF - Journal of membrane science
IS - 1-2
ER -