Filtration Workshop Baxter Bioscience 26 Sep 2013

Tan Ze Yu Application Specialist Filtration Technology Zeyu.Tan@Sartorius-Stedim.com

1

Workshop Agenda
Basics of Filtration Different Mechanisms and Applications of Filtration Production of Filter Cartridge Integrity Testing for Sterilizing Grade Filter Q & A Session

Basics of Filtration
The basic operation Filtration is a physical- mechanical procedure for the separation of substances Phase separations: Solid/liquid : eg. Particle removal out of water Solid/gaseous: eg. Collection of airborne microorganisms Liquid/gaseous: eg. Separation of water drops out of compressed air
3

Use of membrane filters

Residue Analysis If the retained particles or microorganism are of the interest

Filtrate If the visually clear, particle-free or sterile filtrate is needed: eg. Clarification, Particle Removal, Sterile Filtration
4

Classification of Filtrations

Filter Elements

MiniCartridges

MidiCaps

Standard Cartridges

MaxiCaps

Filter Classification

Depth filter Pre-filtration Particle filtration Clarification

Membrane filter Particle filtration Bio-burden reduction Sterile filtration Mycoplasma reduction Virus filtration / Nano filtration Microbiology / Particle analysis
8

Depth filters

Glass fiber - borosilicate glass - binder resin

Polypropylene - PP fibers

Cellulosic - cellulose fibres - diatomaceous earth - binder resin

Seite 28 9

Depth Filters
Filter materials: polypropylene glass fibre cotton activated carbon cellulose

nominal retention rate (0.5 - 100 m) particulate breakthrough at high differential pressures particle release possible integrity test is not possible
10

Depth Filters
Retention mechanism: - Sieve retention (mechanical) - Adsorptive retention (physical)

Particles are retained in the depth of the filter material

Thickness: 0.2 - 20 mm
11

Depth Filter Formats

1. Wounded cartridges on the basis of spun yarns materials: cotton, polypropylene, nylon 2. Wounded cartridges on the basis of several wound mat layer material: polypropylene, nylon 3. Melt blown filter cartridges: material: polypropylene 3. Pleated filter cartridges material: glass fiber layers with binder, polypropylene layers 4. Flat disc modules material: cellulose base depths filters (filter sheets, lenticular filters)
12

Formats: Wounded Depth Filter Cartridges

filter fleeces

direction of filtration fine coarse

13

Formats: Pleated Depth Filter Cartridges

direction of filtration

14

Determination of Particle Retention

15

Depth Filter: Profile of Particle Retention

16

Membrane Filters
Hydrophilic

Hydrophobic

17

Hydrophilic Membrane Filter Materials

Cellulose based

Polyamide PA

Modified hydrophobic polymers Polysulfone PS Polyethersulfone PES Polyvinylidendifluoride PVDF

18

Cellulose nitrate CN Cellulose acetate CA Regenerated cellulose RC

Nylon 6.6 modified Nylon 6.6 Nylon 6

Asymmetric PES Membranes

cross section direction of flow
Degree of asymmetry - can be influenced by casting conditions - can be varied for different applications such as clarification or sterile filtration
19

Heterogeneous Double Membrane

Application specific prefilter layer according to particle size and distribution:

Pre Filter Layer

Final Layer

Example: Sartopore 2 (PES) 0.8/0.2 m: big particles, eg. in protein solutions, complex cell culture media 0.45/0.2 m: standard applications 0.35/0.2 m: small particles, buffer solutions, chem. defined cell culture media

20

Hydrophilic Membrane Filters

absolute" retention rate, i.e. microbiological retention rate is specified pore size: 8.0 m 0.1 m Matrix with high mechanical stability (pressure peaks) small pore size distribution low dirt holding capacity thickness: 100 - 200 m integrity testing is possible
21

Definition of a Sterilizing Grade Filter

Bacterial retention: FDA Guidelines on sterile Drug Products produced by Aseptic Processing, 2004 The microorganism Brevundimonas diminuta (ATCC 19146) when properly grown, harvested and used, is a common challenge microorganism for 0.2 m rated filters because of its small size (0.3 m mean diameter). A challenge concentration of at least 107 organisms per cm2 of effective filtration area should generally be used, resulting in no passage of the challenge microorganism.
22

3. Integrittstestung von Membranfiltern

Regulatory Requirements for Sterile Filtrations

EMEA CPMP/QWP/486/95
http://www.emea.europa.eu/pdfs/human/qwp/048695en.pdf

Prefiltration may be needed - Bioburden reduction filter - Sterilizing grade filters

*NTU = not more than *CFU = colony forming unit 23

Hydrophobic Membrane Filters

Filter material: PTFE (Teflon or Polytetrafluorethylene) Polysulfone and Polyethersulfone PVDF (Polyvinylidendifluoride) Polypropylene Sterile filtration of gases Venting of tanks Aeration and exhaust filtration on fermenters Sterilizing filtration of non polar solvents Removal of small particles from gases
24

Applications:

For venting of non vacuum stable tanks PP or GF fiber filters should be used.

Hydrophobic PTFE Membrane Filters

PTFE / Teflon (6000times)

absolute retention rate high mechanical stability very robust construction high pore volume Small pore size distribution low dirt holding capacity thickness: 70 - 200 m integrity testable (e.g. WIT or IPA)
25

1. Ziel des Trainings

Pleated filter cartridge Sartopore 2

Outer Cage Upstream support layer Filter membrane (one or two layers) Downstream support layer Inner Core

26

Cross section of pleated filter cartridge

direction of filtration

Outer Cage Protecting fleece Pre-filtration membrane (optional) Filter membrane Drainage fleece Inner Core

27

Scale up <-> Scale down

- scalability most important for development products - same materials and construction should be used - other separation technologies such as centrifugation are scalable from lab to process scale

2012

Page 37

Scale up <-> Scale down

linear increase of flow and total throughput per filter element

same flow / cm2

Filtration area (cm2) Scalability for pleated filter elements. Flat filter membranes have different flow characteristics.
2012 Page 29 38

Membrane Production Evaporation Process (schematic presentation) Membranes based on cellulosic derivates
and big pore-size membranes
N2Tank

Granulate polymeres

exhaust

Solvent recovery

Mixing tank

Buffer tank

Casting machine

Tank system

30

Production of Filter Cartridges: Pleating

Control:

number of pleats hight width

Production of Filter Cartridges: Seaming

Production of Filter Cartridges: Assembly

33

Wetting and Autoclaving

Goal: flushing of the membranes preperation for autoclaving propper wetting for integrity testing
2012 Page 11 34

Production of Filter Cartridges: Lot Number Identification

Seite 22 35

Production of Filter Cartridges: Autoclaving

Validated sterilisation process for all capsules labeled sterile Thermal treatment of filter cartidges Validated sterilization for capsules

Shelf life: 5 years for unsterile filters 3 years for sterile filters

36

Overview of the filter production

37

Production of Filter Cartridges: Packaging and QA Release

38

Fundamentals of Filter Integrity Testing (FIT)

39

Why is Integrity Testing so Important?

Integrity testing of filter is necessary... To fulfill the regulatory requirement Determine the presence of oversized pores within the membrane to confirm the integrity of the filter prior to (out of box, after SIP/autoclaving) and after use (release requirement) to detect system leaks (housings, tubes, o-rings, etc.) To prevent damage to the product by the use of defective filters To prevent any risks to the environment (risk of contamination by harmful bacteria)

40

3. Integrittstestung von Membranfiltern

Basics and Regulatory Requirements Bacteria Challenge Test

HIMA Document No. 3, Vol. 4, 1982 / ASTM F 838-05, 2005 (Health Industry Manufacturers Association) Generally, a filter is considered to be a sterilizing filter when, after being challenged with 1.0 x 107 challenge organism per square centimeter (cm2) of effective filter area (EFA), the resulting filtrate is sterile...Pseudomonas diminuta (Brevundimonas diminuta), ATTC 19146 is commonly used.

ASTM F 838-05, 2005: Standard test method for determining bacterial retention of membrane filters utilized for liquid filtration.

41

Destructive Test Methods: Bacteria Challenge Tests

B.diminuta

dosing pump Mix

sterile water

> 107 CFU / cm and effective filtration area

Filter incubation on agar plates

analytical filter (0.45 m)

42

3. Integrittstestung von Membranfiltern

Non-destructive Integrity Testing

Non-destructive integrity Testing is performed by manufacturers for batch release and by operators for in.process control and batch release. Examples: Bubble-Point Test (BP) Diffusion test (Foreward Flow Test) Pressure drop test Water Intrusion Test (WIT) resp. Water-Flow Test (WFT) Correlation: To be able to use a non-destructive integrity test, physical tests were developed that correlate to the bacterial challenge test. A specification for the physical test correlates directly to the bacterial challenge test. Once this correlation is established, it is determined that a cartridge passing the physical test is an integral sterilizing filter.
43

Physical Background: Diffusion and Bubble Point

Ambient pressure Pressurized Air (P<BP) Pressurized Air (PBP)

wetted membrane

Against surface tension and capillary forces Air diffuses through the membrane

Direct flow through the largest pores of the membrane

44

3. Integrittstestung von Membranfiltern

Surface Tension
Surface tension...
...is a property of the surface of a liquid that allows it to resist an external force. ...decreases with rising temperature. ...can be influenced by wetting agents (detergents).

water strider

45

Physical Background: Bubble Point Test

Ambient pressure Pressurized Air (P<BP) Pressurized Air (PBP)

wetted membrane

Diffusion

Bubble Point

46

Physical Background: Bubble Point Test

Flow
Bulk Flow
(linear)

Bubble Point
(non-linearity)

Diffusive Flow
(linear)

PBP

Pressure

47

3. Integrittstestung von Membranfiltern

Bubble Point Test Facts

The Bubble Point is the pressure at which a given liquid in the wetted pores of a membrane is forcibly removed. (overproportional air flow via opened pores) At the Bubble Point the acting force (pressure) exceeds the capillary forces which holds the liquid inside the pores. The Bubble Point detects the largest pores of a fully wetted membrane. Recommendable for "small area" up to large filter systems (from filter discs to multiround systems). Appropriate integrity test for single layer or heterogeneous double layer membranes No correlation of Bubble Point and Bacteria Retention with homogenous double layer membranes !!! Test criterion: minimal Bubble Point (BPmin) A test is successful when the measured BP is BPmin
48

3. Integrittstestung von Membranfiltern

Bubble Point Test Manual Test Setup

low test pressure: single air bubbles

Bulk Flow
(linear)

Flow

Bubble Point

rel. high test pressure: continuous chain of bubbles

(non-linearity)

Diffusive Flow
(linear)

Pressure PBP
49

3. Integrittstestung von Membranfiltern

Diffusion Test - Background

Upstream
Test pressure: e.g. 2.5 bar: Good solubility of test gas in liquid. differential pressure (p)

Downstream
Atmospheric pressure: Bad solubility of test gas in liquid gas release

diffusion
Henry`s Law: The solubility of a gas in a liquid at a particular temperature is proportional to the pressure of that gas above the liquid.
50

3. Integrittstestung von Membranfiltern

Diffusion Test Facts

The diffusion is the gas flow through a wetted membrane, depending on the different solubility at atmospheric pressure and elevated test pressure. The diffusion gas flow is proportional to the differential pressure, the filter area and its thickness. Diffusion depends also from solubility and diffusivity of the test gas in the wetting liquid and from the temperature (of course, it should be kept constant!) Suitable for filters with 150 cm EFA. The test is performed at one defined test pressure (70-80% of BP). Test criterion: max. diffusion rate (Diff.max.)

A test is successful when the measured diffusion rate does not exceed the limit value.
51

3. Integrittstestung von Membranfiltern

Diffusion Test Manual Test

Direct measurement of diffusion (mL/min) using a glas burette

52

3. Integrittstestung von Membranfiltern

Diffusion Test Important Parameters

N = t N/t D L p F d : : : : : : d Gasflow per time [mol/s] Diffusion coefficient (liquid-gas-system) Solubility coefficient (liquid-gas-system) differential pressure boundary surface gas / liquid (~ membrane area) area Thickness of the layer (membrane) The extent of diffusion depends of the kind of test gas. e.g. correction factor for nitrogene as test gas: DiffN2 = DiffLuft x 0,82
53

D L p F

3. Integrittstestung von Membranfiltern

Automatic Filter Integrity Testing

In

Compressed Air Supply

Out

Ext.Ref. Tank

Atmospheric Pressure
54

Diffusion Test Automatic Detection with Sartocheck

Sartocheck measures the pressure decay within the constant volume of the filter housing. Using the pressure drop, net volume and test time, diffusion is calculated according to DIN 58356, part 2:

Diff = p1 VN t p0
p1 p0 VN t p

p1 [mL/min] ln p1 - p

= test pressure = atmospheric pressure (1000 mbar) = net volume = test time = pressure drop
55

Bubble Point Test Automatic BP Detection with Sartocheck

Sartocheck performs a stepwise increase of the test pressure and measures the flow for each step. Pressure / Flow
10 9 8 7 6 5 4 3 2 1

Dedicated algorithms are used to identify the point of overproportional increase

BP

Time / Pressure

The test time strongly depends on the required number of pressure steps until the BP is found

56

Bubble Point Test Automatic BP Detection with Sartocheck

Sartocheck uses special algorithms for detecting the BP. Detection parameters are adjusted to match the normal flow characteristics (small system, standard system, special system) This way of BP detection reliably detects an over-proportional increase of flow as indicator.
57

A2
Flow [mL/min]
continuous linearity Both curves fulfil the A2 criterion overproportional increase (A1fulfilled)

Pressure [mbar]

Diffusion vs. Bubble Point Always the Same Result? Diffusion Test
N = D L p F t d
N/t: D: L: p: F: d: Gasflow per time [mol/s] Diffusion coefficient Solubility coefficient Differential pressure Boundary surface gas/liquid (~membrane membrane area) area Thickness of the layer (membrane)

Bubble Point Test

PBP=
pBP: : : r: K:

2 cos . K r
Bubble Point Surface tension Wetting angle Radius of the pore correction factor

58

Diffusion vs. Bubble Point Always the Same Result?

Cross section of a pore (wetted)
Diffusion Test thickness, surface area e.g. 4 mL/min Bubble Point Test pore radius e.g. 3.7 bar

e.g. 8 mL/min

e.g. 3.7 bar

e.g. 15 mL/min

e.g. 3.7 bar

59

How to Test Hydrophobic Membranes?

Problem: One precondition for integrity testing is the wetting of the membrane. Hydrophobic membranes are not wettable with water. How to solve this? Two approaches:
Artificial Hydrophilisation e.g. 60% IPA/Water Standard Integrity Tests (Diff., BPT...)

Hydrophobic Membrane

Unmodified Membrane

Water Intrusion Test

60

Testing Hydrophobic Membranes: IPA Test

How to perform: Disadvantages:

Installation of cartridge in test housing Wetting with 60% IPA/water mixture

Test can only be performed before sterilization

IPA residues have to be reliably removed

Integrity test (BPT, Diff.) Removal of wetting liquid In-place installation In-line sterilisation Degree of hydrophobicity is not tested Correct installation of cartridge in the housing is not tested

61

Testing Hydrophobic Membranes: Water Intrusion Test (WIT)

The WIT measures the pressure drop which is generated by the intrusion of water molecules into the pore structure of a hydrophobic membrane.

0,3 bar

increasing pressure 2,5 bar

4,5 bar

0 bar

Water PTFE-Membrane 0,2 m

Water Penetration Point (WPP)

62

The Water Intrusion Test (WIT)

Integrity Tester
Compressed Air

The cartridge has to be completely covered with water

Test pressure below Water Penetration Point

Water

63

The Water Intrusion Test (WIT)

Integrity Tester V1 P1 = 2,5 bar (test pressure) P2 = ??? bar 10 min stab time 10 min test time Max. Intrusion Sartofluor GA 10", 0,2 m:13 ml/10min V2

64

3. Integrittstestung von Membranfiltern

The Water Intrusion Test (WIT)

80 70 60 50 40 30 20 10 0 Intrusion (mL/10 min)
During the first minutes ater pressurization the pleated membrane is mechanically compacted ( pressure drop). The test unit cannot distinguish between a pressure drop by compaction and water intrusion! The compaction of the cartridge is finalized within the first 6 min.

recommended stab time

6 8 10 Time (min)

12
Then the pressure drop is mainly generated by intrusion.

65

The Water Intrusion Test (WIT)

Prerequisites: Surface tension of test water >70 mN/m No temperature gradient between test water and air No temperature drift during the test Filter membrane has to be completely hydrophobic (no hydrophilic spots) Advantages of the WIT In-place tests are possible Test after steaming is possible Hydrophobicity of the membrane is tested The complete system is tested, not only the cartridge No manipulation of the downstream side of the filter required But... Intrusion rate is extremely low. Therefore, WIT only possible for EFA > 1000 cm.

66

Just keep in mind...

The integrity test just measures pressure/flow and the test result alone does not tell you anything about the sterile filtration capabilities! Therefore, a correlation of the physical tests with real microbial tests is needed.

67

3. Integrittstestung von Membranfiltern

Correlation Test Limits with Bacterial Challenge Tests

non sterile Safety margin sterile

68

Possible causes of FIT failure - Filter system leakage - Contaminant/product residual in the filter - insufficient wetting/flushing of filter cartridge/capsule - Temperature deviation - of course, it could be due to membrane damage. - etc
69

Questions?

70