Filter cake compression, pressure filters

Theoretical studies (30) comparing the abihty to dewater compressible sohds by sedimenting and filtering centrifuges to pressure filters, have shown that at high G levels, scroU decanters produce drier cakes than pressure filtration. 

Cake Dewatering. Dewatering (qv), identified as a separate entity in filtration, is used to reduce the moisture content of filter cakes either by mechanical compression or by air displacement under vacuum pressure or drainage in a gravitational or centrifugal system. Dewatering of cakes is enhanced by addition of dewatering aids to the suspensions in the form of surfactants that reduce surface tension. 

The most important feature of the pressure filters which use hydrauHc pressure to drive the process is that they can generate a pressure drop across the medium of more than 1 x 10 Pa which is the theoretical limit of vacuum filters. While the use of a high pressure drop is often advantageous, lea ding to higher outputs, drier cakes, or greater clarity of the overflow, this is not necessarily the case. Eor compressible cakes, an increase in pressure drop leads to a decrease in permeabiUty of the cake and hence to a lower filtration rate relative to a given pressure drop. 

Since 1980, a number of new filters have appeared on the market, utilising some form of mechanical compression of the filter cake, either after a conventional pressure filtration process or as a substitute for it. In most designs the compression is achieved by inflating a diaphragm which presses the slurry or the freshly formed filter cake toward the medium, thus squee2ing an additional amount of Hquid out of the cake. 

A variation of the same principle is the DDS-vacuum pressure filter which has a number of small disks mounted on a shaft which rotates discontinuously. The cake is formed on both sides of the disks when they are at the bottom position, dipped into the slurry. When the disks come out of the slurry and reach the top position, hydrauhcaHy driven pistons squee2e the cake and the extra Hquid then drains from both sides of the cake. The cake is removed by blowback with compressed air. 

Belt Presses Belt presses were fiiUy described in the section on filtration. The description here is intended to cover only the parts and designs that apply expression pressure by a mechanism in adchtion to the normal compression obtained from tensioning the belts and pulling them over rollers of smaller and smaller diameters. The tension on the belt produces a squeezing pressure on the filter cake proportional to the diameter of the rollers. Normally, that static pressure is calculated as P = 2T/D, where P is the pressure (psi), T is the tension on the belts (Ib/hnear in), and D is the roller diameter. This calculation results in values about one-half as great as the measured values because it ignores pressure created by drive torque and some other forces [Laros, Advances in Filtration and Separation Technology, 7 (System Approach to Separation and Filtration Process Equipment), pp. 505-510 (1993)]. 

At n = 1 N-s/m, hj, = 1 m and u = 1 m/s, the value r = AP. Thus, the specific cake resistance equals the pressure difference required by the liquid phase (with a viscosity of 1 N-s/m ) to be filtered at a rate u = 1 m/s for a cake 1 m thick. This hypothetical pressure difference is, however, beyond a practical range. For highly compressible cakes, the value ro reaches 10 m or more. Assuming V = 0 (at the start of filtration) where there is no cake over the filter plate, the equation becomes ... 

The following example helps to illustrate the use of the equations presented up to this point. An aqueous slurry was filtered in a small laboratory filter press with a pressure drop of 0.5 atm and at a temperature of 20 C. After 10 minutes, 4.7 liters of filtrate were obtained after 20 minutes, 7.0 liters were collected. From experiments at other pressures, it was determined that the cake compression coefficient was s = 0.4. We wish to determine the volume of filtrate expected after 30 minutes from a filter press having a filtering area 10 times greater than the laboratory press if the filtration is to be performed at 1.5 atm pressure. The liquid temperature will be 55 °C. We also wish to determine the rate of filtration at the end of the process. 

The hydrostatic pressure varies from a maximum at the point where suspension enters the cake, to zero where liquid is expelled from the medium consequently, at any point in the cake the two are complementary. That is, the sum of the hydrostatic and compression pressures on the solids always equals the total hydrostatic pressure at the face of the cake. Thus, the compression pressure acting on the solids varies from zero at the face of the cake to a maximum at the filter medium. 

An increase in pressure influences not only coefficient rQ, but the cake s porosity as well. Since the cake on the filter plate is compressed, residual liquid is squeezed... 

The test is usually performed at a temperature of 300°F and a pressure of 600 psi over a 30-min period. When other temperatures, pressures, or times are used, their values should be reported together with test results. If the cake compressibility is desired, the test should be repeated with pressures of 200 psi on the filter cell and 100 psi back pressure on the collection cell. 

A 30 ml sample of broth from penicillin fermentation is filtered in the laboratory on a 3 cm2 filter at a pressure drop of 5 psi. The filtration time is 4.5 minutes. Previous studies have shown that the filter cake of Penicillium chrysogenum is significantly compressible with S = 0.5. If 500 litres of fermentation broth from a pilot plant have to be filtered in 1 hour, what size of filter is required for a pressure drop of 10 psi and 5 psi Neglect the resistance of the filter medium. 

Having a differential pressure in the above filtration process, and reducing the pressure drop from 10 to 5 psi, increases the filter area by 19%. The main reasons why an increase in the pressure drop results in less filter area are the compressed cake and the porosity of the filter cake. 

Filtration of 300 ml of fermentation broth was carried out in a laboratory-sized filter with a pressure drop of lOpsi. The filtration took 20 min. Based on previous studies, the filter cake obtained from Penicillium chrysogenum was compressible with the exponent 5 in the equation for calculation of filter area equal to 0.5. 

Measurements of filtration rates should be repeated at different pressures or different vacuum levels. This gives information on the influence of pressure on the specific cake resistance. The specific resistance of cakes that are difficult to filter is often pressure-dependent. Thus, use of excessive pressure can result in blocking of the cake, causing filtration to stop. In the case of compressible cakes, information is needed over the whole range of pressures being considered for industrial filters since extrapolation of compressibility beyond the experimentally covered region is always risky. The larger the scale of an experimental filter, the less risky predictions based on the experimental data. 

Filter cakes may be divided into two classes—incompressible cakes and compressible cakes. In the case of an incompressible cake, the resistance to flow of a given volume of cake is not appreciably affected either by the pressure difference across the cake or by the rate of deposition of material. On the other hand, with a compressible cake, increase of the pressure difference or of the rate of flow causes the formation of a denser cake with a higher resistance. For incompressible cakes e in equation 7.1 may be taken as constant and the quantity e3/[5(l — e)2S2] is then a property of the particles forming the cake and should be constant for a given material. 

---