Document LMbN9VbDBgL6neeyGen1LL7X

AB800-1-72 QUALITYASSURANCE PRACTICE and STANDARDS FOR ULTIPOR BACTERIA REMOVAL FILTERS PALL *a9< Prepared by Erwin A Kirnbauer, Corporate Consultant, Contamina tion, Pall Corporation, January 1972. Pall Corporation is a leading international producer of filters and other fluid clarification devices for such diverse customers as phar maceutical, electronic, fluid power, chemical, aerospace, biomedical and other industries. COPYRIGHT PALL CORPORATION. GLEN COVE, N Y . U S A 1972 INTRODUCTION This illustrated report contains data supporting the ability of Pall Ultipor bacteria removal cartridges to provide completely sterile, superclean effluents. The tests employed to obtain this data are indicative of pro cedures established by Pall Corporation to assure high quality standards demanded by industry. Pall Corporation welcomes the opportunity to provide techni cal assistance in the application of Pall filters. This assistance is available by mail, phone or by contacting your local Pall representative. 3 QUALITY ASSURANCE PRACTICE AND STANDARDS FOR ULTIPOR BACTERIA REMOYAL FILTERS 1. Pall Ultipor bacteria removal filters, when properly installed and operated, will meet the following standards: They provide sterile effluents They exhibit no media migration With respect to "built-in" dirt, they provide effluent essentially free of particulate solids 2. Standards applied: These bacteria removal filters are subject to the following tests: When ordered with the "A" option, 100 c/o of the elements are sub jected to bacteria removal test, using 109/100 ml incident bac teria. Downstream bacteria count must be zero. Statistical examination to assure very low particle content in effluent filtered fluid. Pall Ultipor filter elements are manufactured under very clean con ditions with great care to avoid "built-in" dirt. Despite the most careful precautions, some particles and fibers are inevitably picked up on the downstream side of the element. The elements are, however, far better in this respect than any competitive filter, including membranes; and after a short period of use they provide an effluent with an essentially zero count of detectable particles and fibers. Table 1 shows the very low average effluent p article count fo r U ltipor AR, AX, UR and UX elem ents. The in itial 20 liter counts show some v ariatio n ; an occasional elem ent w ill show tw ice th e co u n t noted. T he " su b seq u en t 120 lite r " counts a re v ery consistent, and the typical effluent count during service is zero, when the average blank count is subtracted from th e average effluent count. TABLE 1 Average effluent particle count of Pall Corporation production elements Effluent count ', per lite r per sq. ft. of area P articles larger than 5 pm 2 Initial 20 liters Subsequent 120 liters Typical 120 liter effluent during service Blank1 (Average, computed to 120 liters) 130 5 1.0 1.0 1subtract blank count to get actual particle release numbers. 2um = micrometers = 10-6meter; formerly referred to as " microns" . Fibers 5 A .2 .2 4 3. The detailed test procedure for determining particle release counts is outlined on page 10 entitled "Quality Control Test Methods for Ultipor Bacteria Removal Filters." 4. What is the "Blank"? The blank count is that obtained when the test is performed without the test element in place. It represents particles dislodged from the test jig, from the piping downstream of the test jig, and the upstream parts of the jig which clamp the analysis membrane test disc, as well as par ticles originally present on the test disc when it was installed. The greater part of the blank count is represented by particles present on the analysis membrane disc when installed. Typical standards for a 47 mm membrane disc are : No. of particles greater than 5pm: <80 No. of fibers: 2 5. How clean is "clean"? It is interesting to compare our effluent counts with the following stan dards for cleanliness: D E S C R IP T IO N Ultraclean surface (Grumman Spec. LSP140011A) One liter I.V. bottle Pharmaceutical ampule (30 cc) STANDARD Max. allowable number of particles/ sq. ft. above 10 pm is 111 150 particles larger than 5 pm 30 particles larger than 5 pm 6. How do our bacteria removal elements compare with competition? Data for our production elements vs competitive elements purchased on the open market are presented in Table 2. The data shows a substantial variation in the effluent contamination levels of the filters tested. Those filters which show high counts after the initial sample are suspected of exhibiting media migration. It is also interesting to note that the membrane tubes tested showed particle and fiber counts higher than our bacteria filters. TABI FILTER SAMPLE #1** #2** #3** #4** #5** #1** #2** #3** #4** # lt #2t MANUFACTURER Pall MICRON RATING (BY MANUFACTURER) .2 pm abs. Pall .35 pm abs. Q Submicron #1* R .45 pm # lt #2t # lt #2t #1* # 2* # lt #2t # lt #2t #3t #4t #5t #6t #7t #8t #9t S T U V w Submicron Submicron ,45 pm .2 pm abs. .22 pm abs. Note *Failed Serratia marcescens test in our fab SAMPLE DESCRIPTION MCY1001ARA MCY1001URA Filter cartridge made of 9 fibrous filter medium discs, bound to a center core Fibrous filter medium in cylinder configuration Fibrous discs, edge sealed around a rigid separator Flat sheet fibrous filter medium Fibrous filter medium in corrugated filter element Corrugated membrane tube Non-corrugated membrane tube 'Passed Serratia marcescens test in our lab 7. Contaminants other than particulates: Glycerine: Many type membranes contain glycerine as a plas ticizer; the glycerine contaminates the effluent, which is fre quently very undesirable. Further, the plastic is even more brittle after the glycerine has been washed out. Surfactants: Certain type membranes contain Triton X100, a wetting agent. Figure 1 on page 8 shows the drop in surface tension of the initial effluent. The presence of the surfactant in the effluent is frequently very undesirable (for example tissue culture work). 6 2jE EFFLUENTCOUNT/LITER/SQ. FT. OFAREA INITIAL 20 LITER EFFLUENT PARTICLES LARGER THAN 5 Am FIBERS SUBSEQUENT 120 LITER EFFLUENT PARTICLES URGER THAN 5 Am FIBERS 35 161 172 184 190 1 15177 156 1 117 67,000 H57.000 0.2 10 11 22 9.8 0.8 10 46 2.5 33,000 100,000 3.8 11.8 2.6 6.8 3.3 12 15 4.4 1.6 2,200 2,800 0.02 0.4 .9 1.0 .7 0 05 0 05 0.4 .10 5,300 3,300 5,500,000 5,500,000 175,000 175,000 1,700,000 1,700,000 3,300 13,000 165,000 630,000 11,000 5,000 180 335 6,000 1,100 7 100 30 5 329 495 178,000 164 184 170 750 590 512 13 26 18 19 9 13 16 7 6 tN ot tested fo r bacteria removal 62,000 78,000 9,400 11,000 1.5 4.9 5.2 44 30 36 13 6.9 108 4.3 7.7 5.5 5.7 Blanks not subtracted. .15 1.9 0.66 0.32 2 4.6 2.4 .4 2.4 1.0 .4 .5 .4 : Our new bacteria removal cartridges contain no wetting agents and no glycerine. ! I 8. How long can the P all b a cte ria rem oval c a rtrid g e s be kept on J stream before sterilization is necessary? Many customers keep our bacterial removal cartridges on their high volume water systems for a four week period, which is for many systems the routine time interval for sterilization of the system. Others find it necessary to resterilize their systems more frequently. The interval between successive sterilizations is a function of a number of factors, as system contamination can originate from a number of sources. In 7 FIGURE I r !NOV 22, 1971 ____ 4 s U O I/ Z O oc =u>J : : m VN UFA C T U R ER W 2 2 m \N U F A C TU R ER V 2 2 , m 1 F L U ID : V\/A T E R FLO W RA TE 4 5 L /M N /S Q . FT _ _ _ _ _ _ _ _ _ I_ _ _ _ _ _ _ _ _ _ 1 i i D O W N S FREAAc C O N F A M I N A T IO r IN M E M E R A N E FILTE R M E 31A i VOLUME OF WATER PASSED, LITERS/SQ. FT. OF AREA our experience, where Pall filters have been used, we have not ob served any instance of system contamination due to "grow thru". Thus, the time intervals between system sterilization is predominantly a function of the customer's system maintenance and use procedures, as well as the construction of the system. In this connection it should be noted that our bacteria removal filters have a very important advantage (over the membrane tubes) in that our filters can be repeatedly "in situ" stream sterilized and, therefore, do not need to be removed from the system while system stream sterilization is accomplished. 8 QUALITY CONTROL TEST METHODS for ULTIPOR BACTERIA REMOVAL FILTERS BACTERIA TEST PROCEDURE. The following test procedure is employed on a routine production basis for Ultipor bacteria removal filters: 1. Test Organism--Serratia marcescens, Strain (8UK) (SM) The test organism is grown in a nutrient medium at 30 C for 12 hours and is harvested when the pH reaches 7.5. The culture is centrifuged and the cells are washed twice with saline to remove cell debris and culture medium. The cells are resuspended in spent supernatant 1:1 by weight, resulting in a slurry having approximately 12.5 per cent solids and a viable count of 6.5 x 1011 per gram. This suspension is stored at --40 to --70 C. When needed for test purposes, a portion of the frozen SM is removed and resuspended in gel-phosphate buffer (0.2 per cent gelatin, 0.4 per cent anhydrous NA2HPO4, in deionized or distilled water adjusted to a pH of 7.0). No loss in viability has been detected in the frozen SM after IV2 years of storage. 2. Description of Test. The filter element, fitted with suitable seals, is immersed into a suspension of SM (having a viable count of at least 109 per 100 ml*) so that the filter material and its end cap element seals are immersed. The suspension is drawn through the filter element with vacuum, and 5 to 10 ml of this effluent is filtered through a membrane disc % ins. in diameter. This disc is then placed on trypticase soya agar in a petri dish and in cubated at 30 C to 32 C for 24 to 36 hours. The membranes are then examined for Serratia marcescens colonies. After the filter element has been tested, it is sterilized in an autoclave at 121 C at 15 to 20 psi for 20 minutes. The filter is then flushed with hot water. 3. Filter Classification. If no colonies appear on the membrane discs, the filter elements, all of which have serial numbers, are given `A' classification. * F o r U R A an d A R A elem en ts a c o n c e n tra tio n of 109/100 m l is used. * F o r U X A an d A X A elem ents, a c o n c e n tra tio n o f 105/ 100 m l is used. 9 EFFLUENT CLEANLINESS TEST PROCEDURE. A test set up per Figure 2 is prepared. Then the test set up without the test filter element and analysis membrane installed is flushed for a minimum of 30 minutes. Without the filter element installed in the filter housing but with the analysis membrane (A) installed, a blank test is performed by running 20 liters of water through the setup at a flow rate of 4 liters per minute. The analysis membrane is then removed and the contaminant collected on its surface is microscopically counted. A new analysis membrane and the filter element to be tested are then installed and 20 liters of water at a flow rate of 4 liters/per minute are run through the filter. The analysis membrane is removed for particle count analysis and a new analysis membrane is installed. The test is then continued by running an additional 120 liters of water through the test filter and counting the particles collected on the analy sis membrane. All particle counts are performed per SAE-ARP-598. Particles larger than 5 microns and fibers are recorded separately. WATER FIGURE 2 M EM BR A N E FILTER FIXTURE W ITH M EM BR A N E INSTALLED 0 8 /m i! & ANALYSIS M EM BRANE FILTER (A ) 0 .8 /*m TEST FILTER 10 PALL ULTIPOR Bacteria Removal Filters Available from : Pall T rinity Micro C orporation C o rtlan d , N.Y. 13045 Pall Europe, Ltd. Walton Road, F arlington Portsm outh, H ants, England Pall (Canada) Ltd. 4880 Hickm ore S treet St. L au ren t, M ontreal 9, C anada Pall S. A. A p a rta d o P o sta l No. 199 Cd. N aucalpan de Ju arez Edo. deMexico, Mexico Pall GMBH 6072 D reieichenhain bei F ra n k fu rt S im em sstrasse 6, P o stfa c h West Germany Pall Trincor Corporation 459 C hestnut S treet Union, New Jersey 07083 Pall W estern Corporation 3951 E a st H untington D rive P asadena, Calif. 91107 P.T.M . D istric t S ales Office 24634 Five Mile Road D etroit, M ichigan 48239 Biomedical Division Pall Corporation 30 Sea Cliff Avenue Glen Cove, N.Y. 11542 M arusho Industrial Ltd. 11-2 Ginza 4-Chome Chuo-ku, Tokyo 104, Ja p a n 11