The Forward Flow Test is designed to test the integrity of sterilizing and virus grade hydrophilic and hydrophobic membrane filters.  This qualitative test is based on measuring the gas flow across a completely wetted membrane at a defined constant test pressure on the upstream side.  When the downstream side of the membrane is at atmospheric pressure a diffusion flow of gas is established due to the pressure differential.  The Forward Flow test is Pall’s recommended integrity test for capsule and cartridge filters.

Measured water flow that exceeds the test limit during Water Intrusion testing can be caused by a range of root causes, some of which are listed below. It can be seen that there are several factors which can lead to false test failures.

Possible root causes for false failures are:

• System leaks (filter housing, fittings, tubes, etc.)
• Insufficient test time
• Insufficient stabilization time
• Temperature influence
• Reversible partial wetting of the filter membrane (due to condensation of moisture within the pores or excessive pressure events)

More unlikely for unused filters (pre use test):

• Foreign substances / contaminations deposited on the filter

Root causes for true failures are:

• Filter defects
  
• Compromised O-ring seal with the housing

The WIT is performed on a non-wetted (“dry”) hydrophobic filter. The upstream side of the filter assembly is completely filled with water, covering the entire filter.  An air test pressure lower than the actual water intrusion (or water breakthrough) pressure of the largest membranes pores is applied to the system. The membrane pores remain “dry” during the test. The test pressure drives a transport of water vapors from the water phase across the filter membrane following the pressure differential. Transport of liquid water through all wetted pores and wetted flow pathways or defects will also occur. The WIT quantitatively measures the sum of water vapor (evaporation) and liquid water flow through the hydrophobic filter.

For a sterilizing grade filter, the maximum allowed flow (integrity test limit) is derived from the generic filter validation and correlated with the bacteria retention capability of the filter.

For more information, please see Pall Publication, USD 3033 Application Note: Best Practices for Successful Filter Integrity Testing Using the Water Intrusion Test (WIT) Method 

Pall recommends the use of water to lubricate the O-rings of a filter in order to ease installation into the housing.

Pall does not recommend the use of alcohol (or an alcohol/water mixture) as it could come into contact with the filter membrane, causing a hydrophilic spot that will allow water passage, resulting in a false failure test result, especially in a water intrusion test.

Additionally, if any alcohol gets trapped between the two O-rings grooves, there will be a localized area where the alcohol will expand when the filter is sterilized (autoclave or steam-in-place).  This can potentially damage the filter adapter, or impact the filter to housing seal.

For more information, please see Pall Publication, USD 3033 Application Note: Best Practices for Successful Filter Integrity Testing Using the Water Intrusion Test (WIT) Method 

In general, Pall does not recommend performing an integrity test at temperatures above 50°C when using water as the wetting fluid.

Forward Flow measurements at temperatures above 50 °C are considerably higher than at ambient temperature, and are more difficult to keep constant to the ±1 °C test specification.  This may introduce inaccuracies in the measurement and in calculating appropriate test limits.  When integrity testing is performed at elevated temperatures, the use of jacketed housing or a heating jacket is recommended to keep the temperature stable during the test.

While Pall can provide calculated test limits for elevated temperatures when water is used as the wetting fluid, additional testing should be performed to confirm these limits.  In addition, monitoring actual test results by the end user can also show that these limits are appropriate.

Please contact a Pall representative if you would like more information on developing Forward Flow and Bubble Point limits for all wetting fluids at elevated temperatures.

The Bubble Point Test is designed to detect the largest pores of hydrophilic and hydrophobic membrane filters.  The Bubble Point test is based on measuring the gas flow across a completely wetted membrane at increasing gas test pressure, until the point at which the wetting fluid is expelled from the pores, and bulk flow is measured. 

The Bubble Point test is considered a subjective test, and the results can vary depending on the algorithm of the test instrument.  The Bubble Point method is the preferred integrity test method for testing filter discs, as the Forward Flow is often too low to be accurately measured. 

The Water Intrusion Test measures water flow through a submerged filter when pressure is applied to the upstream side of the filter housing. Because this test can only be performed on a hydrophobic filter, the WIT measurement for an integral filter is primarily evaporative flow of water through the pores of the membrane.

When a hydrophobic filter becomes partially wetted with a low surface tension liquid such as an alcohol water mix, or condensate (from autoclave or Steam-In-Place (SIP)), then the WIT may result in a false failure.  This is due to a water channel forming through the membrane in areas where it has become wet, resulting in the free flow of water.

If the filter has become partially wet, it must be restored to a fully dry state before a WIT can be performed successfully.  Flowing compressed air through the filter for several hours is often required.  Alternatively, oven drying can be performed. Please contact your local Pall representative for the appropriate filter drying conditions.

Pall recommends the following to prevent a hydrophobic filter from becoming wet:

• Keep the filter away from potential sources of low surface tension liquids such as alcohol mixtures.
• If the filter is autoclaved, use a slow exhaust cycle and a vacuum drying cycle.
• If the filter is subjected to SIP, use a cooling gas such as air or nitrogen following SIP.

In certain applications, post-use testing using WIT is impractical due to product contamination on the filter, i.e. bioreactor exhaust filter.  In these cases, Pall recommends performing a post-use Forward Flow integrity test.

For more information, please see Pall Publications:

USD 3033 Application Note: Best Practices for Successful Filter Integrity Testing Using the Water Intrusion Test (WIT) Method

USTR2821 Application Note: Scientific and Technical Report – In Situ Drying of Pall Hydrophobic Air Filters Prior to Moist Heat Sterilization

The Water Intrusion Test is a practical and validated test which can be used for in-situ integrity testing of hydrophobic gas filters. This test is conducted with deionized or higher quality water, without the need for low-surface-tension flammable solvents (such as Isopropyl Alcohol or Ethanol). Due to occupational risks, environmental regulations, safety guidelines and cost associated with handling these low surface tension solvents, water intrusion is becoming the method of choice for integrity testing hydrophobic microbial rated filters for air or gas applications.

Water Intrusion testing is the preferred test where the hydrophobic microbial rated gas filter:

• Is integrity tested in-situ
• A pre-use test is performed (especially after sterilization)
• Alcohol use is restricted or not allowed in the production area

The Forward Flow or Bubble Point test, is the preferred method to integrity test filters in applications where:

• A small area filter is used
• An off-line filter integrity test is performed
• To confirm filter integrity following a water intrusion test failure evaluation

For more information, please see Pall Publication, USD 3033 Application Note: Best Practices for Successful Filter Integrity Testing Using the Water Intrusion Test (WIT) Method

To successfully perform a Water Intrusion Test (WIT), the entire length of the filter has to be covered by (submerged in) water for the duration of the test.

During pressurization of the filter housing assembly to the water intrusion test pressure, the water level drops upstream of the filter. The reason is due to compression of the membrane pleats and elimination of gas bubbles during initial pressurization.

This can leave a portion of the filter exposed to pressurized air, which will freely flow through the exposed area and result in a false test failure.

To troubleshoot the cause of a test failure, Pall recommends:

·       To refill the filter housing with water and repeat the test, or

·       Before the test, increase the upstream volume, to have more water above the filter, to compensate for the reduced water levels resulting from the compression.

For more information, please see Pall Publication, USD 3033 Application Note: Best Practices for Successful Filter Integrity Testing Using the Water Intrusion Test (WIT) Method

In order to consider the influence of the temperature, each filter integrity test method needs to be treated independently, as they are based on different physical principles.

Forward Flow

Forward Flow integrity test limits issued for Pall filters, when wet with standard wetting fluids such as water, 60/40 IPA or similar solutions, apply to a test temperature of 20 °C ± 5 °C. 

Any variation in the temperature of any gas volume in the filter test assembly during the measurement phase has an effect on the flow measurement. Most integrity test instruments measure Forward Flow on the upstream side of filter, as a function of gas pressure change¹. Variations in temperature during the test lead to expansion or compression of the gas in the test assembly (tubing and housing upstream of the filter).  Such variations of the gas volume may lead to inaccurate flow measurements.  Therefore, it is recommended to keep the temperature of the filter assembly constant during the test period.

Pall recommends that the temperature of the filter assembly during the test should not vary more than ± 1 °C. Some simple ways to accomplish this are listed below.

Bubble Point

Changes of surface tension are of direct relevance for bubble point testing as the surface tension influences the capillary forces holding the wetting liquid in the membrane pores. The measured bubble point of a given filter which is fully wetted will increase and decrease in direct proportion with the surface tension of the wetting liquid.

This means that lower bubble point values will be measured at higher temperature, and higher bubble point values will be measured at lower temperature.

Water Intrusion Test (WIT)

For Pall filters, the water intrusion limit values apply to ambient temperature (20 °C) with a specified range of ± 2 °C.  During the test period, the temperature of the filter assembly should not vary more than ± 1 °C.

For the WIT, the temperature of the gas in the filter assembly will have the same effects on the gas volume as indicated above for the Forward Flow test.  Water temperature will also have an effect on the measurements.

As the WIT measures evaporative flow, which is typically much lower than diffusion measurements (as measured by the Forward Flow test), any temperature changes will have a greater effect on the WIT measurements compared to the Forward Flow test and may not be identified by environmental or assembly temperature monitoring.

Maintaining Constant Temperature During Integrity Testing

The following is a list of approaches to maintaining a constant temperature during integrity testing:

• Acclimate the filter and fluids to room temperature before starting the test.  This is especially important for the WIT:  Our guidance is to dispense the water into a container and acclimate for >4 hours.
  
• Avoid placing the filter assembly under heating or cooling registers.
• Avoid handling the filter during the test.
• In cases where room temperature fluctuates, it may be necessary to insulate tubing and filter assembly.

¹ The Palltronic Flowstar line of integrity test instruments measure Forward Flow by direct measurement.  The impact of a change in temperature during the measurement (“Test”) phase will result in an unstable flow measurement, which will extend the test time.

Pall Corporation’s approach for testing multi-modular, sterilizing-grade filter assemblies is to provide assembly-specific Forward Flow limit values below the sum of the maximum allowable FF limit values for the individual filter modules (i.e. a multiplying or reducing factor).  This reduction in allowable Forward Flow is designed to reduce the risk that a marginal filter test failure (in the unlikely event that one is present) cannot be detected in a multi-element assembly.  It results in a tighter, more conservative test limit when compared to a linear multiplier.

While the use of multiplying factors is not a regulatory requirement, it is Pall Corporations’ philosophy on integrity testing to use practices that provide the maximum safety for large filter area installations.  This needs to be balanced with the risk that a set of integral modules may fail the test due to the application of a multiplying factor that is too stringent (“false fail”).  This approach is also described in the 2008 revision of PDA Technical Report No. 26 on Sterilizing Filtration of Liquids 1.

Pall Corporation’s basis for defining the appropriate multiplying factor for a specific multi-modular installation is based on several parameters including:

• Type of membrane and number of elements

• Statistical distribution of observed Forward Flow (FF) values for the specific filter

• Mean deviation for Forward Flow distribution

• Standard deviation of Forward Flow distribution

• Point of first failure during microbial challenge (if observed)

Forward Flow limits obtained from Pall Corporation for multi-element filter assemblies will incorporate the appropriate multiplier.

PDA Journal of Pharmaceutical Science and Technology, Technical Report No. 26 Sterilizing Filtration of Liquids, Rev. 2008. Supplement Volume 62, No. S-5

Prior to performing an integrity test, it is imperative that the pore structure of the filter membrane be filled with the wetting fluid.  To assure complete wetting, Pall recommends applying downstream flow restriction (sometimes referred to as back pressure, not to be confused with reverse pressure) during the filter flush.

The use of a downstream flow restriction helps to:

1. Ensure uniform flow distribution through the entire length of the filter
2. Overcome the tendency for fluid to flow through the path of least resistance. 
3. Remove air entrapped in the membrane pleats, by further solubilizing the air (due to increased pressure in the system) and by compressing air bubbles to a size where they can freely pass through the membrane.

When testing filters larger than 254 mm (10 in.) or multi-round systems, a downstream flow restriction will ensure the fluid reaches the top of the housing during venting.  Otherwise, the hydrostatic pressure of the liquid will cause it to flow out of the downstream side of the housing before reaching the vent at the highest point in the system.

To apply a downstream flow restriction, Pall recommends installing a pressure gauge downstream of the filter housing, followed by a flow control valve such as a diaphragm valve.

For more information, please see Pall Publication, USD3297_User Guide: Wetting and Flushing of Pall Microbially_Rated_Filter Cartridges and Capsules

Pall manufactures different membrane filters for gas filtration. We use either Polytetrafluoroethylene (PTFE) filter membrane or Polyvinylidene Fluoride (PVDF) modified, for such a purpose.

In our Pall Allegro™ single-use systems, the method to sterilize these systems is by gamma irradiation.

The Polytetrafluoroethylene (PTFE) filter membrane, is incompatible with gamma irradiation, so it could not be used as a gas filter in our Single-Use Systems.

The Polyvinylidene Fluoride (PVDF) modified is compatible with gamma irradiation and for this reason is the device to be used with our Single-Use-Systems.

Pall Emflon II V002 sterilizing grade filters uses a filter membrane made of polyvinylidene difluoride (PVDF).

Integrity test failures can be caused by a range of root causes, some of which are listed below:

• System leaks (filter housing, fittings, tubes, etc.)
• Insufficient test time
• Incorrect temperature of wetting fluid
• Insufficient stabilization time
• Temperature influence
• Incomplete wetting of the filter
• Incorrect test limits
• Incorrect pressure source
• Filter defects or damage
• Compromised O-ring seal
• Incorrect filter selection
  

When an integrity test failure occurs, the first step is to verify the system setup and test parameters.  After all of these conditions have been verified, the filter should be re-wet and tested again (using the Forward Flow test, even if the initial failure was recorded using a bubble point test, except for small area filters (<200cm²). 

If the result is a pass, then the filter is integral.  If it fails, the Forward Flow test should be repeated after a more vigorous wetting step.  This can include a larger flush volume, application of back pressure (downstream flow restriction), or increased differential pressure.  If the filter fails again, a flush and Forward Flow integrity test with a low surface tension wetting fluid, such as 60:40 IPA/water, should be performed.  If the filter fails the integrity test again, the filter (still installed in the housing if possible) should be returned to Pall Corporation for further analysis.

When investigating an integrity test failure, one of the first steps is to determine if any leaks are present in the system under test. Integrity test instruments perform the measurements on the upstream side of the filter, thus preserving the downstream sterility of the system.  The test instrument is not able to differentiate gas flow through the filter from gas flow through a leak in the hardware.

Prior to performing the leak test, a visual inspection of the system is recommended to identify any potential leak sources.  Damage to the housing or to the o-ring seals are common sources of leaks.  Such signs of damage include cuts on the O-rings, or misshapen or dented housings.

Pall recommends performing a leak test using the Palltronic^® Flowstar^® IV,  as follows:

1. The leak test can be performed on a filtration system with or without a filter installed.  If a filter is installed, as is the case with capsule filters, the filter must be dry.  Diffusive or bulk gas flow through a wet filter cannot be differentiated from leaks in the assembly.
2. The outlet of the housing should be sealed with either a blank endcap, valve, or other suitable termination method.
3. All drain and vent valves must be in the closed position, except in cases where the connection to the instrument is made through a vent.
4. A length of pneumatic tubing with the Palltronic Flowstar IV external vent valve installed should be used to connect the OUT port of the Palltronic^® Flowstar IV integrity test instrument to the housing vent port.
5. From the Palltronic Flowstar IV main menu, select the “Leak Test” function.
6. The Forward Flow test pressure that is used to test the filters should be used as the pressure for the leak test. 
7. The maximum system size for the leak test is 50 liters.

If the Palltronic Flowstar IV integrity test instrument is not able to detect a pressure loss in the volume or the leak rate is too small to be detected, it will report “no leak detectable” or “flow within limits’ as the result.  These results confirm that the filter system under test is free of significant upstream leaks.

If the test reports “flow outside limits” (> 1 mL/min)

• Ensure that the test is being carried out under stable temperature conditions.
• Re-check the connections and housing enclosure are tightened fully.
• If sanitary fittings are an integral part of the filter housing, then the elastomeric seals should be examined and changed if necessary.
• For smaller housings, submerging the pressurized system in a water bath and looking for bubbles will indicate the location of the leak.
• A soap-based leak detecting product can be used around the fittings and connections to locate a leak.

For more information, please contact Pall’s NEW Equipment Support Hotline.

The Bubble Point test measures the pressure region at which diffusive flow transitions into bulk flow, by looking for a deviation from the stable background flow versus flow through open pores.  However, a marginal leak in the upstream side of the system under test, or a minor damage to the filter, can add a small amount of gas flow.  This could be interpreted by the test instrument as part of the background flow, and therefore, the leak would not be detected.  The module factor has been designed into the software of the Palltronic® Flowstar IV software to detect these types of flow contributions.

As stated in USD 2594, Instructions for use for the Palltronic® Flowstar IV integrity test instrument, the Module Factor (MF) defines the sensitivity for the Leak Test phase which is executed at the beginning of the Bubble Point test sequence.

The Bubble Point integrity test using the Flowstar IV integrity test instrument features this stabilization/leak test before proceeding to the actual step-wise increase of pressure for the “Bubble Point” test.  This initial leak test seeks to establish that filter system under test shows an expected and typical background flow. 

For this test, the upstream side of the filter is pressurized to 80% of the minimum programmed BP value, followed by a measurement of the gas flow occurring at this gas pressure, which is compared against a limit value represented by the programmed filter type related ‘Module Factor’.  If the gas flow measured is below the limit value, the BP test proceeds.

The lower the “Module factor” is set, the more sensitive it will be to detect a deviation from the typical background flow for the filter under test during this step in the test sequence.

The expected, typical and thus acceptable background flow caused by gas diffusion is dependent on the filter area.  Pall recommends that the “Module Factor” should be set to the number of 10” modules being tested. It should be set to 1.0 for 10” cartridges, 0.5 for 5” cartridges (e.g. AB05 or SLK7002), 0.2 for smaller (e.g. AB02 or SLK7001) filter and 0.1 for flat membranes (e.g. 142 mm discs) or mini capsules (KA02 or similar).  The default value is 1 if nothing is entered.

This will ensure that the test sequence is aborted when a filter clearly displays a gas flow above the expected typical background value.

For software versions > 2.0b, the user can set the Module Factor to as low as 0.01 for small area filters such as the Acro 25/50 and KM5 devices.  See Table I below for the recommended input values for the module factor.

For more information, please contact Pall’s NEW Equipment Support Hotline.