pressure drop across the cake and an average permeability of the cake.
Wakeman [4] derived an analysis of the formation and growth of
compressible filter cakes based on the principles of continuity, Darcy's
law and the knowledge of the porosity profile of the cake. However
Wakeman's method of treating the moving boundary of the cake
interface made it difficult to assess the validity of his analysis. An
approach describing cake formation in several solid–liquid separation
operations was given by Tiller et al. [5]. This method only demonstrated
its utility. The information of cake structure was not fully
understood. While useful theoretically and practically most of these
proposed modes were considered for ideal situations. For example,
when considering the retention of dust particulates on the surface of
the filter medium the effect of non-coverage of the cake in the cake
filtration is not considered. In the actual cake filtration process, cake
formation and growth are essential in the dust particulate packing
process. The packing is generally not ordered, instead, the dust
particulates generally form randomly or loosely packed structures.
There has been much anecdotal evidence in previous research that
suggest that dust cake formation is crucial to achieving high collection
efficiency [3–9]. The granular bed filter is one of the most promising
approaches for hot temperature gas cleanup and is used in advanced
coal conversion systems. The basic principle of the granular bed filter
is the removal of suspended particulates via the passage of flue gas
through the filter media. Unlike conventional filtration technologies, a
granular bed filter is especially attractive for hot gas clean-up because
it employs low-cost refractory filter media and can work in high
temperature environments. From our earlier series of cold tests of a
granular bed filter, we found that the existence of dust cakes would
have an influence on the collection efficiency of a filter system.
Consequently, in this study, we carry the investigation further. Here
we use a multipoint measurement method to obtain the average
Powder Technology 192 (2009) 217–224
⁎ Corresponding author. Tel.: +886 3 426 7341; fax: +886 3 425 4501.
E-mail address: sshsiau@cc.ncu.edu.tw (S.-S. Hsiau).
0032-5910/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.powtec.2008.12.014