ERT 313 : BIOSEPARATION ENGINEERING Mechanical - Physical Separation Process “1. FILTRATION” By; Mrs Haf iza Bint i Shu kor ERT 313/4 BIOSEPARATION ENGINEERING.

ERT 313 : BIOSEPARATION ENGINEERING

Mechanical - Physical Separation Process

“1. FILTRATION”By; Mrs Hafiza Binti Shukor

ERT 313/4 BIOSEPARATION ENGINEERINGSEM 2 (2010/2011)


Students should be able Students should be able to; to;

APPLY and CALCULATE based on filtration principles; ANALYZE cake filtration, Constant Pressure Filtration, Continuous Filtration and Constant Rate Filtration.


IntroductionIntroduction

• Filtration is a solid-liquid separation where the liquid passes through a porous medium to remove fine suspended solids according to the size by flowing under a pressure differential.

• The main objective of filtration is to produce high-quality drinking water (surface water) or high-quality effluent (wastewater)


2 categories of filtration, which differ according to the direction of the fluid feed in relation to the filter medium.

Results in a cake of solids depositing on the filter medium

Minimize buildup of solids on the filter medium


Application of Filtration in Application of Filtration in Bio-industryBio-industry

Recovery of crystalline solidsRecovery of cells from fermentation

mediumClarification of liquid and gassesSterilization of liquid for heat sensitive

compound


Filtration EquipmentFiltration EquipmentFiltration for biological materials is generally completed using batch

filtration, rotary drum filtration, or ultrafiltration methods.

1. Batch Filtration• Usually performed under constant pressure with a pump that

moves the broth or liquor through the filter• Filter cake will build-up as filtration proceeds and resistance

to broth flow will increase• The filter press is the typical industrial version of a batch

vacuum filter, using a plate and frame arrangement• Can be used to remove cells, but does not work particularly

well for animal cell debris or plant seed debris


Cont….Filtration EquipmentCont….Filtration Equipment

2. Rotary Drum Filtration• Rotary vacuum filters can be used to efficiently remove

mycelia, cells, proteins, and enzymes, though a filter aid or precoat of the septum may be necessary

3. Ultrafiltration• Utilizes a membrane to separate particles that are much larger

than the solvent used • Successful removal occurs in the partical size range of 10

solvent molecular diameters to 0.5 μ


Filter MediaFilter Media• To act as an impermeable barrier for particulate matter.• Filtration media for cross-flow filtration are generally referred as

“MEMBRANE”• First and foremost, it must remove the solids to be filtered from the

slurry and give a clear filtrate• Also, the pores should not become plugged so that the rate of

filtration becomes too slow• The filter medium must allow the filter cake to be removed easily

and cleanly• Some widely used filter media (for conventional filtration) like filter

paper, ceramics, synthetic membrane, sinterd & perforated glass, woven materials (woven polymer fiber).


Filter AidsFilter Aids• Substance (solid powder)that are mixed with the feed for

creating very porous cakes ( increase filtration rate very significantly)

• Can be added to the cake during filtration to increases the porosity of the cake and reduces resistance of the cake during filtration

• Can also be added directly to the feed to: i) maintain the pores in the filter cake open

ii) Make the cake less compressibleiii)Provide faster filtration

• Common types of filter aid is diatomite (types of algae) and perlite.

The structure of diatomite particles gives them a high intrinsic permeability


Filtration PrinciplesFiltration Principles• When a slurry containing suspended solids flow against a filter medium by the application of a

pressure gradient across the medium, solids begin to build up on the filter medium• The buildup of solids on the filter medium is called a cake• This type of filtration is sometimes referred to as “dead-end” filtration• Darcy’s law describes the flow of liquid through a porous bed of solids and can be written as follows:

• where V is the volume of filtrate, t is time, A is the cross-sectional area of exposed lilter medium, Δp is the pressure drop through the bed of solids (medium plus cake), µ0 is the viscosity of the filtrate, and R is the resistance of the porous bed. In this case, R is a combination of the resistance Rm of the filter medium and the resistance Rc of the cake solids:

• It is convenient to write the cake resistance Rc in terms of specific cake resistance α as follows:

• where ρc is the mass of dry cake solids per volume of filtrate.

• Thus, the resistance increases with the volume filtered• Combining Eq. (1), (2) and (3), we obtain

(1)

(2)

(3)

(4)


• For the case of zero filtrate at time zero (before start an exp), integration of this equation yields

• where and

(can determine specific cake resistance,α and medium resistance, Rm by plotting the graph)

In a cake filtration process where a significant amount of cake is allowed to accumulate, the medium resistance, Rm become neglegible compare witn the cake resistance. (Rm=0). So,

Incompressible CakeIncompressible Cake

(5)

BA

VK

V

At

PK o

2

P

RB m

CmXY

2

2

A

V

Pt o


Example 1Example 1Batch Filtration A Buchner funnel 8 cm in diameter is available for testing the filtration of a cellculture suspension, which has a viscosity of 3.0 cp. The data in Table E1 wereobtained with a vacuum pressure of 600 mm Hg applied to the Buchner funnel.The cell solids on the filter at the end of filtration were dried and found to weigh14.0 g. Determine the specific cake resistance α and the medium resistanceRm. Then estimate how long it would take to obtain 10,000 liters of filtrate from

this cell broth on a filter with a surface area of 10 m2 and vacuum pressure of500 mm Hg.

TABLE E1


SolutionAccording to Equation (5), we can plot t/(V/A) versus V/A and obtain α from the slopeand Rm from the intercept. We see that the data are reasonably close to a straight line.

A linear regression of the data in this plot gives the following results (Figure E1):

Example 1Example 1

Figure E1 Plot of batch filtration data for the determination of α and Rm.

(5)

BA

VK

V

At

CmXY


From these values, we can calculate α and Rm:

This is a typical value of Rm for a large-pore (micrometer-sized) filter.To determine the time required to obtain 10,000 liters of filtrate using a filterwith an area of 10 m2, we must make the assumption that α does not change atthe new pressure drop of 500 mm Hg. We use Equation (5) and solve for time:

Example 1Example 1

(5)


We calculate the two components of this equation as follows:

and finally

Thus, this filter is probably undersized for the volume to be filtered. In addition, from thiscalculation we see that at the end of the filtration,

Therefore, the filter medium is contributing very little of the resistance tofiltration, a typical situation in a lengthy dead-end filtration.

Example 1Example 1(5)


• Almost all cakes formed for biological material are compressible.• As these cake compressed, filtration rate drop (flow become relatively more

difficult as pressure increase)• The pressure drop is influence by α, the specific cake resistance• α can be increased if the cake is compressed• The specific resistance of the cake is directly affected by Δpc, the pressure drops

across the cake• Studies have shown that the relationship between specific resistance and pressure

drop commonly takes the form:

• where α’ and s are empirical constants.• The power s has been called the “cake compressibility factor”. (for incompressible

cake, s=0 and for compressible cake, s=0.1-0.8)

Compressible CakeCompressible Cake

(6)


Cake WashingCake Washing• After filtration, the cake contains a significant amount of

solute-rich liquid broth that usually removed by washing the cake

• 2 function of washing:A) displaces the solute-rich broth trapped in

pores in the cakeB) allows diffusion of solute out of the biomass in

the cake(enhance recovery if the desired product is in the biomass)


• It is often necessary to wash the filter cake with water or a salt solution to maximize the removal of dissolved product from the cake.

• Frequently, the wash must be done with more than the volume of the liquid in the cake because some of the product is in stagnant zones of the cake, and transfer into the wash liquid from these zones occurs by diffusion, which takes place at a slower rate than the convective flow of wash through the cake

• Data for the washing of the filter cakes has been correlated by Choudhary and Dahlstrom using the following equation:

• where R’ is the weight fraction of solute remaining in the cake after washing (on the basis that R’ = 1.0 prior to washing), E is the percentage wash efficiency, and n is the volume of wash liquid per volume of liquid in the unwashed cake.

• Assuming that the liquid viscosity and the pressure drop through the bed solids are the same during the filtration of the solids, the washing rate per cross-sectional area can be found from the filtrate flow rate per unit area given in Equation (4) at the end of the filtration

• Thus, for negligible filter medium resistance for filtrate volume Vf at the end of time tf to form the cake, this yields

(7)

(8)


• If Vw is the volume of wash liquid applied in time tw, then

• Using the definition of (dv/dt)V=Vf from Eq. (8), we obtain

• At the end of filtration, the integrated form of the filtration equation (Eq. 5), with Rm neglected, can be written

• Substituting this expression for Vf/A in Eq. (10) and simplifying gives

Filtration PrinciplesFiltration Principles

(9)

(10)

(11)

(12)


• From Eq. (11) and (12), the ration of tw to tf is

• It is helpful to write tw/tf in terms of n, the ration of the volume Vw of wash liquid to the volume Vr of residual liquid in the cake:

• where f is the ratio of Vr to the volume Vf of filtrate at the end of filtration.

• The ratio f can be determined by a material balance• Thus, for a given cake formation time tf, a plot of wash time tw versus wash ratio n will be a

straight line

Filtration PrinciplesFiltration Principles

(13)

(14)


Example 2Example 2Rotary Vacuum Filtration It is desired to filter a cell broth at a rate of 2000 liters/h on a rotary vacuum filter at avacuum pressure of 70 kPa. The cycle time for the drum will be 60 s, and the cakeformation time (filtering time) will be 15 s. The broth to be filtered has a viscosity of 2.0cp and a cake solids (dry basis) per volume of filtrate of 10 g/liter. From laboratory tests,the specific cake resistance has been determined to be 9 x 10 cm/g. Determine the area of the filter that is required. The resistance of the filter medium can be

neglected.

Solution:We can use the integrated form of the filtration equation, Equation (5), with Rm = 0:

We solve for A2 to obtain

In applying this equation, it is helpful to focus on the area of the drum, which is where the cake is being formed and where filtrate is being obtained.


We use this volume of filtrate with t = 15 s in the equation for A2 to obtain

The area A’ of the entire rotary vacuum filter can be calculated from the cake formation time (15s) and the total cycle time (60s) as

This is a medium-sized rotary vacuum filter, with possible dimensions of 1.0 m diameter by 1.0 m long.

Thus, A is the area of that part of the drum. We can calculate the volume of filtrate that needs to be collected during the cake formation time of 15 s:


Example 3Example 3Washing of a Rotary Vacuum Filter Cake For the filtration in Example 2, it is desired to wash a product antibiotic out of the cake sothat only 5% of the antibiotic in the cake is left after washing. We expect the washingefficiency to be 50%. Estimate the washing time per cycle that would be required.

Solution;From Equation (7) for the washing efficiency of a filter cake

where R’ is the weight fraction of solute remaining in the cake after washing (on the basis of R’ = 1.0 before washing), E is the percentage wash efficiency, and n is the volume of wash liquid per volume of liquid in the unwashed cake. Substituting R’ = 0.05 and E = 50% into this equation gives

From Equation (14), the relationship between the washing time tw, and the cake

formation time tf is given by


where f is the ratio of volume Vr of residual liquid in the cake to the volume of filtrate Vf

after time tf. Thus, we need to estimate the volume of residual liquid in the filter cake to determine tw. At the end of the 15 s cake formation time,

Assuming the cake is 70 wt% water, which is typical for filter cakes, we find

Thus,

Cake solids per volume of filtrate

Volume of filtrate need to be collected during the cake formation time of 15s

llg

gmV

soV

m

194.0/1000

194

,


The End

ERT 313 : BIOSEPARATION ENGINEERING Mechanical - Physical Separation Process “1. FILTRATION” By; Mrs Haf iza Bint i Shu kor ERT 313/4 BIOSEPARATION ENGINEERING.

Documents

cake filtration

rate of filtration

filtration rate

filtration media

continuous filtration

filtration proceeds

conventional filtration

filtration principles