Đang tải... (xem toàn văn)
phương pháp thử nghiệm độ toàn vẹn lọc khí
Water Intrusion Test Integrity Testing Test description Conventional integrity testing of membrane filters using the bub- ble point, diffusion or pressure hold tests requires that the filter membrane be completely wetted with a suitable wetting agent. Hydrophobic filter elements can be tested by this method only if the surface tension of the wetting agent is less than the critical surface tension of the filter material being used. Other- wise, complete wetting does not take place. This is why different organic solvents like ethanol or solvent| water mixtures are used as prac- tical wetting media. This aspect is especially detrimental when the system is used as a “sterile air filter" for venting tanks, bio- reactors and fermenters . Integrity testing of filters for liq- uids is usually performed in situ. However, the in situ integrity testing of hydrophobic gas filters is only possible to a limited degree due to safety reasons (e.g., explosion protection when test- ing with organic solvents) and to keep the product from solvent contamination. Besides these complicated individual tests must be followed by a drying phase before any filter elements can be installed. Additionally the filter sealing in the housing still has to be tested in situ. Since the filters should be retested immediately aher in-line steam sterilization, tests with a solvent|water mixture are impractical for the user. Moreover, the drying procedure must be validated since solvent residues can contaminate the product if drying has been insufficient. The Water Intrusion Test (WIT) was developed to overcome all of these disadvantages. The WIT can be used to run routine integrity tests simply, easily and reliably. Definition Water Intrusion Test (WIT). The Water Intrusion Test (WIT) is an in-situ integrity test for hydrophobic filters. The WIT measures the decay rate of a pressure level imposed upon a hydrophobic membrane enveloped in water. Theoretical principles. Among other things, a pressure gradient that is dependent on pore size is necessary to over- come negative capillary forces (cohesive forces). This pressure is generally called the "Water Penetration Point" (WPP), the pressure at which water is pressed through a hydrophobic membrane. The WPP is depend- ent on the hydrophobicity of the filter material and the pore size and is comparable to the bubble point. The relationship can be illustrated as follows: D max = Diameter of the largest pore σ = Surface tension of the liquid in dynes/cm (water 72) θ = Angle of contact (greater than 90° in hydrophobic filters) ∆ p = Available upstream differential pressure (bar|psi) k = Correction factor (required since membrane filter pores are not cylindrical capillaries) Introduction 4 · σ· cosθ D max =·k ∆ p Intrusion (ml/10 min) Test Pressure (bar) 350 300 250 200 150 100 50 0 0123456 Water Intrusion Water Penetration Water Intrusion Characteristics at Different Test Pressures Pore Size: 0.2 µm, Height: 10’’, Temperature: 20°C Principle of Water Intrusion Test The Water Intrusion Test (WIT) was developed as an integrity testing method to test hydropho- bic sterilizing grade filters. These filters are often used as inlet air and off-gas filters for fermenters bioreactors and for venting sterilizers, freeze dryers auto- claves and tanks. The WIT is employed in both the qualification of filter elements and as an inplace test procedure. The WIT - like the mercury intrusion test - is based on the capillary depression of non wetting liquids on the outer surface of the membrane. To withstand these capillary forces, a certain pressure gradient, which is dependent on the pore size among other factors, is necessary. The test run is in a similar manner to the Diffusion Test, although there is a major difference. In the Diffusion Test, the diffusive gas flow through a wetted membrane is measured. However, during the WIT a hydrophobic filter installed in a filter housing is flooded with water on the upstream side. The pressure drop detected by an automatic integrity tester indi- rectly measures the volume of water intruding into the mem- brane matrix. At first the test system has to be flooded with water, pressurised at the defined test pressure and then stabilised for a defined time. After the stabilisation time is complete, the water intrusion will be measured. Origin of the pressure drop. The pressure drop has two main reasons: 1. The upstream volume is in fact increasing due to remaining compaction processes. Since this compaction is an asymp- totic process, it is never absolutely completed. There- fore, a certain time has to be determined after which most of the compaction is already done. That is the main reason for the stabilization time of 10 Minutes. 2. Water molecules actually get transported through the mem- brane. In case of an integer membrane, this process is mainly due to evaporation of water molecules through the pore structure. If the mem- brane is punctured, water is mainly flowing through the punctured hole. Water vanishes from the up- stream side. The effect is, that the upstream volume is increasing (keep in mind that water is not subject to compaction). This amount of water (∆V) can be determined using the fundamen- tal law of Boyle and Mariott (p 1 · V 1 = p 2 · V 2 ) Principle of WIT testing compressed air|gas p 1 p 2 V 1 · ∆ p (p 1 - ∆ p) = ∆V ∆ t V n · ∆ p (p 1 – ∆ p) · ∆ t ∆V = V 2 - V 1 = V 1 · - 1 = with: p 1 : Absolute pressure at the beginning of the pressure drop measurement [mbar] p 2 : Absolute pressure at the end of the pressure drop measurement [mbar] ∆p: = p 1 - p 2 V 1 : Upstream Volume at the beginning of the measurement [ml] (= V n ) V 2 : Upstream Volume at the end of the measurement [ml] The increase in Volume per time is defined as: This value is defined – as one easily understands – as the “Water Flow Value”. This value is basically linked to the Water Intrusion Value by a simple mathematical correlation: The Water Intrusion is a value that has been defined in analogy to the diffusion. For small pressure drops (< 5% of the absolute test pressure), the following equation holds: with: V n : Upstream Volume [ml] ∆p: Pressure Drop [mbar] 1000 mbar: Reference Pressure ∆t: measurement duration [min] Physically, air molecules pass through the wetted membrane. If one collects all these molecules, they will fill a certain volume at 1000 mbar reference pressure at the ambient temperature as reference temperature. This volume pre time is the Diffusion value. The Water Intrusion is calculated using the same formula: If one imagines the volume ∆V that has been added to the upstream Volume due to the Water Flow to be filled with gas, it would use a different volume at an ambient pressure of 1000 mbar. Again, the required volume at 1000 mbar (∆V10OO) can be calculated with Boyle Mariott's law to be: ∆V ∆ t V n · ∆ p (p 1 - ∆ p) · ∆ t WIT = · = · = p 1 - ∆ p 1000 mbar (p 1 - ∆ p) 1000 mbar V n · ∆ p 1000 mbar · ∆ t Diff. = V n · ∆ p 1000 mbar · ∆ t WIT = V n · ∆ p 1000 mbar · ∆ t ∆V 1000 = ∆V· (p 1 - ∆ p) 1000 mbar Test parameters Practical aspects Water Intrusion Test. • Allows inplace testing after sterilization • Avoids using wetting agents, e.g. IPA/Water - Therefore no contamination - No removal problems No alcohol residue or down- stream contamination • The hydrophobicity of the filter stays hydrophobic during the test • High sensitivity of the test • The hydrophobicity of the filter is evaluated by the test • The time is reduced, especially at multiple systems, which reduces the shut down periods • Directly correlated to the ASTM Bacteria Challenge Test. System set-up. The WIT can be used in a variety of air filtration applications that require minimal engineering changes, however, some times systems may have to be reconfig- ured. This is where the Sartorius Technical Support team comes to your assistance. The diagram to the right shows an example of a manual WIT sys- tem. The basic construction does not differ significantly from that used for the diffusion test. In principle, WIT systems can be designed as manual, semi-auto- matic and fully automatic units to suit the applications, e.g., sterile venting systems for auto- claves, Iyophilizers, fermenters and tanks. Basic requirements for the test are: 1. The surface tensions of the water used must be > 72 dynes/cm. 2. Minimized temperature differ- ence between the water and the inlet air ± 1k. 3. Temperature fluctuations must be avoided during the test. 4. Filter element must be com- pletely hydrophobic. 5. Adequate effective filter area, i.e., > 0.1 m 2 . 6. The integrity tester must have sufficient accuracy, e.g., Sartocheck ® unit. Operating sequence: 1. Completely fill the upstream side of the filter housing with water. 2. Close all upstream valves. 3. Connect the integrity tester, e.g., Sartocheck 3 unit. 4. Start the test: The unit per- forms the WIT automatically. 5. The test is completed and the results are printed out. 6. Empty the filter housing through an appropriate drain or condensate valve. 7. Briefly vent dry using pressu- rized air with the inlet and drain valves open. 8. Start operating the system. Maximum permissible intrusion rates False failures and trouble- shooting. If the filter cartridge does not pass the test, it might be that partial hydrophilization of the membrane has occurred. In this case, the filter must be steam- sterilized for at least 30 min. at 121°C or 134°C and then repeat- edly blown dry with hot air. Other false failures might be attributable to water surface tension and temperature fluctua- tions. All of these factors will invariably cause elevated water intrusion values (false failures). Experience has shown that these increased values indicate “marginal“ failures and do not lead to high water flow rates or extreme pressure drops. An isopropanol|water test will deter- mine whether one of these causes is the problem or whether the filter cartridge is out of specification. Fault free filters will pass the solvent test because this test cannot detect any hydrophilization or reduced sur- face tension of the water used for the initial WIT. After the WIT, the filter cartridges need to reach their original air flow rate as quickly as possible. In other words, they must not become blocked with water, e.g., residual water in the fleeces. The time that the filter cartridge needs to achieve 100% of the original flow rate is called the “blow-down" time. Test pressure. The high air pressures needed to force water penetration of the PTFE membranes used in the Sartofluor filter cartridges result in very high test pressures used for the various pore size ratings. This increases the tests reliability, accuracy and precision. Pore Size Water WIT Test Penetration Pressure Pressure bar psi bar psi 0.20 µm 4.5 65 2.5 36 0.45 µm 2.8 41 1.5 22 These variables are dependent on the introduction of water with a surface tension of > 72 dynes/cm. Stabilization and test time The compressible upstream air volume above the water column and the actual water intrusion level are both low. This illustrates the importance of maintaining sufficient stabilization and test times. Enough time must be available for the water column to become saturated with air at the test pressure. Similarly, the water must be distributed over the entire membrane area without the presence of air bubbles and the filter pleating must be fully compacted. The following, values have been established for the stabilization and test times: Stabilization time 1 13 min Stabilization time 2 10 min Test time 10 min If these times are shortened the intrusion values may be too high due to insufficient distribution of water across the membrane surface or faulty compacting of the membrane pleats. As in the diffusion test (and in all integrity tests), the maximum permissible water intrusion rates given by the filter manufacturer for the WIT must follow the ASTM bacteria challenge test methodology (HIMA Document No. 3, Vol. 4, April 1982). Factors influencing the Water Intrusion Test Hydrophilizing active agents. As already mentioned, the use of WIT requires that the filter elements be com pletely dry and hydrophobic. Surface active agents like solvent vapors, detergents or oils can lead to partial membrane hydrophiliza- tion and thus to false negative results. Moreover, these agents might also reduce the air flow rate if parts of the membrane matrix are blocked by these agents. During normal operations, strict atten- tion must be paid to these factors and appropriate counter measu- res taken when necessary. If partial hydrophilization occurs the filter cartridge must be hand- led as described on previously page. Difference in temperature. False failures may occur if the temperature difference between the test air and the water is too great. If the water temperature is much lower than the air tem- perature, great drops in pressure and high intrusion rates can result. If the water temperature is much higher, the pressure drops will be too low. This situation can be remedied by storing the water in a storage tank in the room where the test will be performed or by filling the housing with water adjusted to the right temperature. Low surface tension of the water. This situation mostly arises when the tanks being used are not cleaned and rinsed thoroughly, old ion exchange resins are used for preparing the water, or the water temperature is too high (> 32°C|90° F). These parameters can be avoided by taking preven- tive measures or adapting the water temperature. Fluctuations in ambient temperature. Such temperature fluctuations affect the pressure drop measured during the test. The same applies to the diffusion and pressure drop test. The possi- bilities for the test being affected are minimized because the area of the housing that comes into contact with the test air is very small compared to that in the diffusion test. Moreover, the great heat capacity of water compensates for any fluctuations. Integrity tests are normally performed in airconditioned rooms. This source of error is therefore encountered very rarely. Inlet volume. Generally with large-volume systems, the precise volume can be easily determined, whereas the measured pressure drop may merely amount to a few mbar. The results may then border on the accuracy limits of the integrity tester and false results could be obtained. Similarly, when dealing with very low net volumes, high pressure drops can occur, leading to inaccurate test results. The net inlet volume must there- fore be selected so that both parameters relevant for intrusion measurement – namely, volume and pressure drop – can be deter- mined as accurately as possible. Sartorius AG Weender Landstrasse 94–108 37075 Goettingen, Germany Phone +49.551.308.0 Fax +49.551.308.3289 www.sartorius.com Sartorius Corporation 131 Heartland Boulevard, Edgewood, New York 11717, USA Phone +1.631.2544249 Fax +1.631.2544253 Toll-Free +1.800.3687178 Sartorius Limited Longmead Business Centre Blenheim Road, Epsom, Surrey KT 199 QN , Great Britain Phone +44.1372.737159 Fax +44.1372.726171 Sartorius S.A. 4, rue Emile Baudot 91127 Palaiseau, France Phone +33.1.69192100 Fax +33.1.69200922 Sartorius S.p.A Via dell’Antella, 76/ A 50011 Antella (FI), Italy Phone +39.055.634041 Fax +39.055.6340526 Sartorius K.K. No. 3 Hoya Building 8–17 Kamitakaido 1-chome Suginami-ku Tokyo 168-0074, Japan Phone +81.3.33295533 Fax +81.3.33295543 Sartorius S.A. C/Isabel Colbrand 10-12 Edificio Alfa III planta 4, of. 121 28050 Madrid, Spain Phone +34.91.3586100 Fax +34.91.3588804 Specifications subject to change without notice. Printed in Germany on paper that has been bleached without any use of chlorine. W/sart-153a · G Publication No.: SP-4504-e04014 Order No.: 85030-507-71 . Definition Water Intrusion Test (WIT) . The Water Intrusion Test (WIT) is an in-situ integrity test for hydrophobic filters. The WIT measures the decay rate. to change without notice. Printed in Germany on paper that has been bleached without any use of chlorine. W/sart-153a · G Publication No.: SP-4504-e0 4014