Page 354 - Chemical Process Equipment - Selection and Design

11.5. LABORATORY TESTING AND SCALE-UP 317

3. Cake drying is critical. to test washing rates and efficiencies and rates of moisture removal
4, Cake removal may be a problem. with air blowing. Typical data of these kinds are shown in Figure
5, Precoating may be needed. 11.3. Detailed laboratory procedures are explained by Bosley
(1977) and Dahlstrom and Silverblatt (1977). Test and scale-up
'll.5. LABORAUOR'S TESTING AND SCALE-UP procedures for all kinds of SLS equipment are treated in the book
edited by Purchas (1977).
Laboratory filtration investigations are of three main kinds: Before any SLS equipment of substantial size is finally selected,
it is essential to use the results Qf pilot plant tests for guidance.
1. observation of sedimentation rates; Although many vendors are in a position to do such work, pilot
2, with small vacmm or pressure leaf filters; equipment should be used at the plant site where the slurry is made.
3. with pilot plant equipment of the types expected to be suitable Because slurries often are unstable, tests on sKpments of slurry to
for the plant. the vendors pilot plant may give misleading results. It may be
possible to condition a test slurry to have a maximum possible
Sedimentation tests are of value particularly for rapid evaluation of resistivity, but a plant design based on such data will have an
the effects of aging, flacculants, vibration, and any other variables unknown safety factor and may prove uneconomical.
that conceivably could affect a rate of filtration. The results may
suggest what kinds of equipment to exclude from further con- COMPRESSION-PERMEABILITY CELL
sideration and what kind is likely to be worth investigating. For
instance, if sedimentation is very rapid, vertical leaves are excluded, Such equipment consists of a hollow cylinder fitted with a
and top feed drums or horizontal belts are indicated; or it may be permeable bottom and a permeable piston under controlled
indicated that the slurry should be preconcentrated in a thickener pressure. Slurry is charged to the slurry, cake is formed with gentle
before going to filtration. If the settling is very slow, the use of filter suction, and the piston is lowered to the cake level. The rate of flaw
aids may be required, etc. Figure 11.1 illustrates typical of filtrate at low head through the compressed cake is measured at a
sedimentation behavior. Figure 11.2 summarizes an experimental series of pressures on the piston. From the results the resistivity of
routine. the cake becomes known as a function of pressure. The data of
Vacuum arid pressure laboratory filtration assemblies are Figures 11.4(b) and (c) were obtained this way; those of Figure
shown in Figure. 11.7. Mild agitation with air sometimes may be 11.4(a) by filtration tests.
preferable to the mechanical stirrer shown, but it is important that There is much evidence, however, that the resistivity behavior
any agglomerates of particles be kept merely in suspension and not of a cake under filtration conditions may be different from that
broken up. The test record sheet of Figure 11.8 shows the kind of measured in a CP cell. The literature is reviewed by Wakeman
data that normally are of interest. Besides measurements of filtrate (1978). CP cell data are easily obtained and may be of value in a
and cake amounts as functions of time and pressure, it is desirable qualitative sense as an indication of the sensitivity of resistivity to
pressure, but apparently are not of acceptable engineering accuracy
for the design of filtration equipment. The deduction of resistivities
from filtration tests is illustrated in Example 11.1.
TABLE 1'1.7. Specific Resistances of Some Filter
Cakes THE SCFT CONCEPT
Filtration Resistance No serious attempt has yet been made to standardize filtration tests
Pressure SI Units, and to categorize filtration behavior in generally accepted terms. A
Material psi m/kg possibly useful measure of filterability, however, has been proposed
10'
High grade kieselguhr - 1.64~ by Purchas (1977; 1981). The time in minutes required to form a
Ordinary kieselguhr 25 1.15X 10" cake lcm thick when the cell is operated with a differential of
100 1.31 X IO" 500Torr (0.67bar) is called the Standard Cake Formation Time
Carboraffin charcoal 1.4 3.14X IO'' (SCFT), tp The pressure of 500Torr is selected because it is
10 5.84 x 1o'O obtained easily with common laboratory equipment. The procedure
Calcium carbonate 25 2.21 x 10" suggested is to make a series of tests at several cake thicknesses and
(precipitated) 100 2.68 X 10" to obtain the SCET by interpolation, rather than to interrupt a
Ferric oxide (pigment) 25 8.04X IO" single test to make observations of cake thickness. A direct relation
100 14.12 X 10" exists, of course, between the SCFT and resistivity x; some
Mica clay 25 4.81 X IO"
100 8.63 X 10" examples are
Colloidal clay 25 5.10 X IO'*
100 6.47 x 10" Material a (m/kg) SCFT tF (rnin)
10''
Magnesium hydroxide 25 3.24~ Filter aid 1.64(E9) 0.26
(gelatinous) 100 6.97 X IO" CaCO, 2.21(E11) 34.6
Aluminium hydroxide 25 2.16~ 10'~ Colloidal clay 5.10(E12) 798
(gelatintous) 100 4.02 x 1013
Ferric hydlroxide 25 1.47 x 1013 Full scale filtration equipment requirements can be estimated
(gelatinsous) 100 4.51 x ioT3 quickly in terms of tp For instance, when the resistance of the filter
Thixotropic mud 80 6.77 X 1 0l4
Theoretical figures for medium is neglected, the constant pressure Eq. (11.3) may be
rigid spheres: written as
d=lQpm 6.37 x io9
d=l pm 6.37 X IOlq (11.27)
d=0.1 pm 6.37 x IO',
(Carmian, 1938). where L is the thickness of the cake in meters. Upon rationing in