Document km1q3B3m1Bv2D1xxnvKwzRv8q

.'jurnal: American Water Works Association, v.70, no 2, (February) 1978, pp.97-102 Is Chrysotile Asbestos Released From Asbestos-Cement Pipe Into Drinking Water? William H. Hallenbeck, Edwin H. Chen, Carolyn S. Hesse, Kusum Patel-Mandlik, and Arthur H. Wolff To investigate the possibility that asbestos-cement pipe does, as has been suggested, release asbestos as a contaminant in drinking water, a study was made of water samples from fifteen public water-supply systems before and after their flow tiirough asbestos-cement pipes of various ages, lengths, and diameters. More than 200 000 miles of asbestoscement water distribution pipe have been placed into service in the United States since its introduction in 1929. Onethird of all water distribution pipe currently being sold in the US is made of asbestos-cement. Its principal attributes are resistance to corrosion and scaling, light weight, and low cost.1 A question has arisen concerning the possible release of asbestos fibers from asbestos-cement pipe into drinking wa ter. Occupational epidemiology studies indicate that there may be a human health hazard associated with the inges tion of asbestos because death rates due to cancer of the digestive system are elevated in asbestos workers.1 This find ing may be related to the swallowing of asbestos that was inhaled and cleared from the respiratory system via the respi ratory clearance mechanism.1 Asbestos is a generic term for a number of fibrous silicate minerals and represents 16-18 per cent (by weight) of the composition of asbestos-cement po table-water pipe.1 Chrysotile refers to the serpentine [M&ShO.ofOH),] variety of asbestos* and comprises 80 per cent or more of the asbestos used in asbestoscement pipe. The remaining asbestos is of the crocidolite variety.1 The Johns-Manville Corp. has studied two asbestos-cement pipes in two differ ent municipal water systems in order to determine if chrysotile was released. These pipes of unspecified ages were studied over a period of about one year. Paired water samples were taken two to three times a month from water before and after it flowed through the pipes. It was concluded that chrysotile release from both pipes was on the order of nanograms per liter at the 10 per cent level of significance.1 Since the effect of pipe age on the release of chrysotile was unknown, the FEBRUARY 1978 0003-150X / 78 /0200-0Q97S01.00 c 1978 American Water Work.A**ocntion . LLENBECK ET AL 97 CAPCO JEN 0021188 1 TABLE 1 Genera) Characteristics of Selected Asbestos-Cement Field Pipes Water Systems Groundwater svstems Westmont lisle Hoffman Estates Rolling Meadows York Center Lake Michigan systems Sannockburn Bradlev Road Zion-Benton Waukegan Zion Midlothian Blue Island Brookfield Glenview Highland Park Pipe Age veers Length ft Oiameter in OS 200 1 3000 18 3700 20 4400 27 1100 1 1300 14 1000 16 10000 19 600 26 1150 27 1500 30 3244 35 40000 37 1500 40 275 12 6 6 12 6 10 6 8 6 6 6 8 20 6 6 TABLE 2 Saturation Index and Data Pertinent to its Calculation' Water Systems PGruundwater systems Westmont Ltsle Hoffman Estates Roiling Meadows York Center Lake Michigan systems Bannockburn Bradley Road Zion-Benton Waukegan Zion Midlothian Blue Island 8rookfield Glenview Highland Park 'Abbreviations; H total hardness (CaCO,) A - total alkalinity iCaCO,) Ca" calctum I - pH - pH,, Data not available H k 50 * -o/es.f 0 75 456 4 45 306 3 26 I 42 1 42 135 180 135 t 23 123 123 t.28 1.30 A x 10 * Ca-- x 10 ' T *<? * moles '1 c log K pH pH, 712 0 35 10 234 7.1 79 5 76 2.40 10 2.34 77 74 300 2 23 10 2.34 82 7 7 568 195 10 2 34 7 5 74 7 28 3 13 10 2.34 7-7 72 2 16 0 93 5 2 47 84 6.3 2.16 093 s 2.47 6.4 &3 2.18 0.90 10 2 34 6.1 8.2 2.80 0 95 5 2 47 t 62 2.16 090 10 234 8.1 6.2 2.16 0.63 IS 2.21 62 6.! 2.16 063 IS 2.21 6.2 8.1 2.16 063 15 2 21 82 6.1 1.96 0.83 10 2.34 7.6 82 200 ass 5 2.47 7.9 8.3 ^Saturation "Index (f) -08 03 05 01 05 0.1 0.1 -0.1 -0.1 01 0.1 (11 -0.6 -04 K - (Ca--) (HCO,)/H T - temperature pH - - log H' pH,, - log K 1 * 53 2S (4H-A) * pCa" * pA. for pH < 5 5 (4H-AJ TABLE 3 Aggressiveness index and Data Pertinent to its Calculation Ftg. 1. Chrysotile asbestos fibril in a field sample (length * 1.8 pm. diameter - 0.04 pm) Water Systems Total Alkalinity pH mg 1CoCO, Calctum Hardness mg/I CaCO, Groundwater svstems Westmont Lisle Hoffman Estates Roiling Meadows York Center Lake Michigan systems Bannockburn Bradley Road Zion-Benton Waukegan Zion Midlothian Blue Island Brookfield Glenview Highland Park 71 7.7 82 75 77 84 84 8.1 t 8.1 82 82 82 7.6 79 356 268 150 284 364 108 108 108 140 106 108 106 108 96 100 35 240 223 195 313 93 93 90 95 90 83 63 83 83 65 *x - total alkalinity in mg`I CaCO.. y - calcium hardness in mg'l CaCO, tData not available Aggressiveness Index pH + log (xy)` 11 2 12 5 12.7 122 128 12.4 12.4 12.1 t 12.1 12.2 12.2 122 11.5 118 Aggressiveness moderate none none none none none none none t none Qone Qone none moderate moderate TABLE 4 Number of Grid Squares Scanned. Average Number of Fibers Per Grid Square, and Standard Deviation' Fig. 2. Selected area electron diffraction pattern (SAED) of a chrysotile asbestos fibril Before Pipe Water Systems Number of Grid Squares Scanned Average Number of Fiberst Per Grid Square Groundwater systems Weatmonl Lisle Hoffman Estates Rolling Meadows York Center Lake Michigan systems Bannockburn Bradley Road Zion-Benton Waukegan Zion Midlothian Blue Island Brookfield Glenview Highland Park 20 015 (0 49) 10 010 (0 32) 20 0 25 (0.72) 20 0.40 (0 60) 20 005(0 22) 20 0.10(0 31) 20 0.10 (0 31) 10 0.10 (0 32) 20 0 20 (0.70) 10 010(032) 20 0.40 (1.3S) 20 1.15 (1.57) 10 0 70 (0 62> 20 0.05 (0.2 20 0.15 (o.r *!n parentheses (fiber designates either a fibril or bundle of fibrils. (Standard deviation not applicable when count u zero. After Pipe Number of Grid Squares Scanned Average Number of fibers Per Cnd Square 20 0.10 (0 31) 10 060(1.90} 20 0 t 20 o X 20 0 25 (0.72) 20 0.30(0 57) 20 065(0 88) 10 0.30 (0 67) 20 0.10(0 31) 10 0.10 (032) 20 045 (0 94) 20 095 (1.23) 10 030(0.48) 0 25 (0 64) 0 20 (0 52) 98 WATER TECHNOLOGY/QUALITY JOURNAL AWWA CAPCO JEN 0021189 may be broken down during transit and thus create the appearance of chrysotile release, in terms of fiber count, from asbestos-cement pipe. Some evidence TABLE 7 Distribution of Chrysotile Fiber Lengths was found to support this hypothesis from the following analysis of Lake Michigan systems. All the fiber lengths from the before samples were compared Length of Fiber ufn Groundwater Systems Absolute Frequency number Cumulative Frequency per cent Lake Michigan Systems Absolute Frequency number Cumulative Frequency per cent to all the fiber lengths in the after samples by using the Mann-VVhitney test. The lengths of the after fibers were found to be somewhat shorter (P = 0.063) than those of the before samples. The same analysis of groundwater systems did not reveal a signifi cant difference between before and after fiber lengths (P = 0.382). Because of the possibility that fibers may be broken down during transit through the pipe, a before-after increase was considered significant only if a statistically signifi 0l 02 OJ 04 OS 06 or 08 10 1-1-2.5 2 6-S 0 5.1-100 10 1-150 15 1-200 20 1-25 0 Tola! 2 14 6 2 1 2 l 0 0 0 0 1 0 0 0 31 6 51 b "1 0 *7 4 6? 93 5 968 968 968 968 968 100 0 IS 15 9 0 6 2 1 2 a 3 4 l 2 2 77 26 22.1 416 53 2 59? 07 5 70 I 71 4 74 0 844 883 93 5 945 97 4 1000 cant increase was demonstrated by both fiber count and mass data. Referring to Table 6. it can be seen that these two criteria were not satisfied at the 5 per TABLE 8 Fiber and Mass Concenlrations of Chrysotile* cent level of significance. Table 7 contains the distribution of chrysotile fiber lengths. Eighty-seven per Water Systems Before Pipe 6bers*l PS'I After Pipe fibers/f n't cent of the chrysotile fibers were in the form of single fibrils and 13 per cent were composed of bundles of fibrils. The length of the longest fibril in a bundle was taken as the length of the bundle. The mean and median lengths of Lake Michigan fibers (2.2 nm and 0.4 (im. respectively) were somewhat greater than those encountered in groundwater systems (0.5 jim and 0.2 pm, respective ly)The diameters of chrysotile fibrils in groundwater and Lake Michigan systems were distributed over a narrow range from 0.02 pm to 0.1 pm: mean and median diameters were both 0.05 pm for both Groundwater sv stems Westmont Lisle Hoffman Estates Rolling Meadows York Center Lake Michigan systems Bannockburn 8radle> Road Zion-Benton Waukegan Zion Midlothian Blue Island Brookfield Glenview Highland Park 1 2 x 10* 8 0 x 10* 2.0 x 10' 3 2 X 101 40 X 10* 5 8 X 10* 5 8 x 10* 5 8 x 10* 1.2 x 10* 5 8 x 10* 23 x 10* 6.7 x 10> 4 1 x 10* 2 9 x 10* 88 x 10* These are preliminary 'Jala; see text for details. +Below detection V______________________ 120 100 220 9SO 24 31 3500 240 10 100 160 740 17500 540 230 140 8 0 x 10* 4 8 x 10` t 2 0 X 10' 1 8 x 10* 38 x 101.8 x 10 5 8 X 10' 5.8 X 10* 26 X ;0' 55 X 10* 1.8 X 10V 1.5 X 10' 1.2 X 10* 740 350 + t 220 2200 240 730 2800 200 600 3900 160 480 38 ______________________ / types of water. Bundles of fibrils were not included in the analysis of diameters because, in general, they do not have circular cross-sections nor uniform widths. The central question addressed in the present study was whether an increase of chrysotile could be detected in water In conclusion, fifteen asbestos-cement pipes of various ages, lengths, and diam eters were studied for possible release of chrysotile asbestos. Although several water systems contained moderately aggressive water, no statistically signifi cant release of chrvsotile was observed. References 1. Personal communication. Mr. J.C. Jackson. A/C Pipe Producers Association. Wash ington. D.C. (1977) 2. Hallenbeck, W.H. & Hesse, C.S. A Review of the Health Effects of Ingested Asbestos. Rev. Envir. Health, 2:3:157 passing through asbestos-cement pipe and. if so, how much of an increase. Because the before sample served as the reference or control for the after sample, there was no need to analyze samples specifically for the purpose of deter mining laboratory contamination. The problem of laboratory contamination is widely recognized by asbestos re searchers and may result from the presence of chrysotile in the air. mem brane filters, or grid coatings. When dat: are presented in an absolute sense as ir Table 8 (Fiber and Mass Concentration: of Chrysotile). it is important to contro lor contamination. Since appropriatf control samples were not analyzed foi T kut0ry contam'nat'on- the data ir Table 8 must be considered preliminan la nature. It may be possible, however, that chrysotile is released from pipes at levels below the detection limits used in this study or at levels that were obscured by labora tory contamination or by chrysotile already present in the water. Acknowledgements Support was provided by the Univ. of Illinois Water Resources Ctr. under a grant from the US Dept, of the Interior's Ofce. of Water Res. & Tech, and the Campus Research Board of the Univer sity of Illinois at the Medical Center. Chicago. The information and coopera tion provided by the Illinois Environ mental Protection Agency was greatly appreciated. We thank Ms. Lucille Vaughn for typing the manuscript. (1977). 3. Evans, J.C., et al. Studies on the Deposi tion of Inhaled Fibrous Material in the Respiratory Tract of the Rat and Its Subsequent Clearance Using Radioactive Tracer Technique. Environ. Res.. 6:2:180 (Jun. 1973). 4. Berry. L.G. & Mason, B. Mineralogy. Free man, San Francisco (1959). 5. A Study of the Problem of Asbestos in Water. AWWA Research Foundation. Prepared for the A/C Pipe Producers Association. Denver (Jul. 1974). 6. McCrone, W.C. & Delly. J.C. The Particle Atlas Vol 1. Ann Arbor Science Pub., Ann Arbor (2nd ed,, 1973). 7. Beaman, D.R. & File, D.M. Quantitative Determination of Asbestos Fiber Con centrations. Anai. Chem.. 48:1:101 (Jan. 1976). 8. Rickards, A.L. Estimation of Submicron ! fOWUARY 1978 W.%H. HALLENBECK ET AL 101 CAPCO JEN 0021190 Quantities of Chryson.V -\sr`-"sfos by Electron Microscopy A,-.u< Chem. 45 4 609 .Apr 19731. 9. Kay G. Ontario Intensifies Search for Asbestos in Drinking Water Wtr. Pollut. Control. 111.9 33 (Sept. 1973). 10. Sms.se. M I.: Talbot. IH. i Re.voall, R.EG. Electron Diffraction Patterns of U I.C.C. Asbestos Samples. Envir. Res.. 4 3:141 (Apr. 1971). 11 Ch.aTfield. E [. A Pi.-llas. H Measuring Asbestos in thp Environment Cun Res. Dev.. 7 6 33 (N'ov-Dec. 1974). 12. Cunningham. H M. A Pontefract. R. Asbestos Fibers in Beverages and Drink ing Water. Nature. 332 5309 332 (|ul. 1971). 13. McFarren. E.F: MlLLETTE. [.R. & LlSHKA. R.|. Asbestos Analysis by Electron Microscopy. Proc. AVV'WA Water Qual. Con/.. Xiv'-l-.i (Dec. 1974). 14. Gaze. R. The Physical and Molecular Structure of Asbestos Ann NY Arad Sci,. 132.1.23 iDec. 19651 15, Birks. L.S. et al. Quantitative Analysis of Airborne Asbestos by X-Ray Diffrac tion: Feasibility Study. Naval Research Laboratory (Washington). Prepared for USEPA. Washington D C. (Feb. 28. 1975). 16. Fair, G.M.: Geyer. |.C. & Okun. D.A. Water and Wastewater Engineering Vol 2. Wiley & Sons. New York (1968). 17 Personal communication. Dr. James Letneweber. Research and Development. Johns-Manville Corporation. Denver. Colo. (1977). 18. Conover. W.|. Practical S'onparametric Statistics. Wiley. New York (1971). 19. Halle.vbeck, W.H.. et al. Precision of Analysis for Waterborne Chrysotile As bestos by Transmission Electron Micros copy. Bull. Envir. Contom. and Toxicol.. 17:5:551 (May 1977). 20 Biometriko Tables tor Statisticians W| I ES Pearson A H.O. Hartley. ed ors). Cambridge University Press. London [2nd ed.. 19621. 21. Chaxravarti. I.M. A Rao. C.R. Tables for Some Small Sample Tests of Significance for Poisson Distributions and 2x3 Contingency Tables. Sankhya: Indian Jour. Statistics. 21:315 (1959). An article contributed to and selected by the JOURNAL, authored by William H. Hallen- beck, asst, prof., envir.. health sci.; Edwin H. Chen, assoc, prof., biometry; Carolyn S. Hesse, res. assoc., envir. health sci.; Kusum Patel-Mandlik, visiting asst, prof.; and Arthur H. Wolff, prof, and dir. envir. health sci.. all of the School of Public Health, Univ. of Illinois at the Medical Ctr., Chicago, III. 53465 4300, 5535 Reprinted and copyrighted as a part of JOURNAL AMERICAN WATER WORKS ASSN. VoL 70 No. 3 February 1178 Printed in VS*A. V CAPCO JEN 0021191