Document 65yx6nwaerEnjMZZVxp0VNkq1

FILE NAME: Brakes (BRK) DATE: DOC#: BRK2 DOCUMENT DESCRIPTION :Z&Z ZhZZ rnVIRonmfnTAL {-SCIence kTeClino1ogg Determining the Frequency of Asbestos Use in Automotive Brakes from a Fleet of On-Road California Vehicles Joseph De Vita/ Stephen Wall,:j: Jeff Wagner,:j: Zhong-Min Wang,:j: and Leela E. Rao-I-,* tCalifornia Air Resources Board, Mobile Source Control Division, El Monte, California 91731, United States 'California Department of Public Health, Environmental Health Laboratory, Richmond, California 94804, United States ABSTRACT: Asbestos is a known human carcinogen, and recent regulation in California limits asbestiform fibers in brakes to trace levels beginning in 2014, although there is no corresponding federal requirement. In order to gauge the current prevalence of asbestos use in automotive brake applications, the California Air Resources Board tested brake linings from 137 light- and medium-duty vehicles and 54 heavy-duty vehicles. Only about 3% of the light- and mediumduty vehicle brake linings contained chrysotile asbestos. All of those brake linings were drum-type shoes, which are generally being phased out. No asbestos was found in low mileage vehicles presumed to have their original stock linings from the vehicle manufacturer. Additionally, no asbestos was found in the heavy-duty vehicle brake shoe linings sampled. Given the small percentage of vehicle brake linings with asbestos observed, it appears that the prior federal ban that was subsequently overturned, in combination with a threat of litigation, has reduced asbestos use in brake linings. However, our study was limited in scope and without a national ban, the current and future prevalence of asbestos in brakes is uncertain, suggesting the need for continued monitoring of materials released as toxic air contaminants in normal braking operations. INTRODUCTION Asbestos has been used in many products, including automotive brake linings, cements, millboard, roofing materials, floorings, heat-resistant gaskets, and electrical wiring insulation. I For decades, asbestos was the primary material in automotive brake linings due to its thermal stability, low wear rate, low cost, and other useful characteristics. 2 It is estimated that in 2003, 10% of the estimated world sales of chrysotile asbestos was used for brake linings and clutch faclngs. 3 Chrysotile asbestos was eventually shown to be a human carcinogen and has become the source of much debate and an aggressive litigation industry over the past few decades. Central to the litigatIOn are risks to auto mechanics from handling brake linings containing asbestos, as identified by Castleman et al.4 and reviewed subsequently by Lemens and Welch.6 However, asbestos particles are also released into the general environment as a result of routine driving and braking, with airborne concentratIOns of asbestos close to major highways and roadways up to 3-6 times higher than at locations away from roadways?,8 As of 2011, 5S countries, including all member states of the European Union, have banned the use of all forms of asbestos In manufactured products.9 In 1989, the United States Environmental Protection Agency (USEPA) issued a rule that would have phased out the manufacture, import, and processing of onginal equipment manufacturer (OEM) asbestos-containing automotive brakes by 1996,10 but that rule was subsequently overturned in 1991 by the fifth Circuit Court of Appeals.1! In reaction to the impending U.S. asbestos ban, rising concern in other Industrialized countries about asbestos health effects, and ongoing litigation, manufacturer; of brake linings began searching for asbestos alternatives. A 1998 report indicated 25% of new vehicles built In the V.S. With OEM drum brakes contained asbestos,12 but by 2004 a VSEPA survey of the nine major OEMs found that none of the major manufacturers sold asbestos brakes In new vehicles or as replacement parts in the V.S. 13 Instead, OEMs were Optlllg to use alternative formulatIOns such as metalhc, semlmetalhc, synthetic, and ceramic fibers. However, no legal Impediment to uSing asbestos exists, and a 2007 OccupatIOnal Safety and Health Administration (OSHA) survey found that some vehicles of two OEMs were still equipped With asbestos containing friction products. 14 Furthermore, aftermarket brake linings containing asbestos are still available as demonstrated by Senator Patty Murray (D-WA), who brought an asbestos brake lining purchased by her staffer to a 2007 V.S. Senate Subcommittee heanng in an effort to dispel the myth that asbestos products are no longer available in the V.S. IS Current litigation regarding asbestos-related cancer may lead to further reductions in asbestos use, and even greater natIOnWide reductions may occur due to a California ban on asbestos 111 frictIOn materials that goes into effect in 2014. 16 KCCCJ\'cd: Kcvised: Accepted PublIshed: August 1, 2011 December 14, 2011 December 14, 2011 December 14, 2011 V ACS Publications 2011 Amencan Chemical Society 1344 dx dOl erg/la 1021/es202667d I EnVlron So Technol 2012,46, 1344-13'> 1 Asbestos IS known to cause lung cancer, pleural fibrosis (scarnng of the lung tissue), and mesothehoma, a cancer of the membrane that lines the body cavities and some internal organs.'7 The onset of mesothelIoma can mvolve a latency of 10-50 years from the time of exposure, thereby accounting for current litigatlOn based on exposures from past decades. The etJology of mesothelIoma and other asbestos-related diseases mvolves vanous environmental sources of asbestos, most of which are anthropogenic. Historically, asbestos ftbers of tOJ{lcologlcal concern have been those that can penetrate deep mto the lungs, which are typically between 5 and 10 /-lm in length. As such, OSHA has permissible exposure limits on airborne asbestos fibers, which are defined as those 5 /-lm or longer, with an aspect ratlO (l.e., length-to-width ratio) of 3: 1 or longer. IS However, the definitIOn of health-relevant fibers is somewhat variable and a recent review by Lemens found that there was significant evidence for toxicity of asbestos fibers less than 5 pm m length, which outnumber fibers greater than 5 Jim m brake wear debns by a ratio of 300: 1. Relevant motor vehicle sources of asbestos are brake linings, clutch facings, gaskets, and automatic transmission components. Of Importance m this context IS the mhalation exposure to brake wear dust emitted from the vehicle and entrained in ambient air. Due to the high heat generated during braking, much of the surface asbestos is converted to forsterite, a nonftbrous material. Thus, both forstente and short-fiber length chrysotlle asbestos are present in brake wear debris. While the tOXlClty of chrysotlle asbestos is well documented, the health nsk aSSOCiated With forsterite exposure IS not well studied, although evidence suggests that both forsterite and brake lining dust have a cytotoxic effect.s The percentage of brake wear mass that becomes airborne has been reported as anywhere between 7 and 70%?,19-23 Mechanics performing brake replacements are especially susceptible to high exposures.4 In 2010, brake wear accounted for 19% of Callforma's onroad motor vehicle emiSSIOns inventory of PM lO (particulate matter :510 pm), as based on the California Air Resources Board's (CARB's) model for estimatmg California's motor vehicle emlsslOnsH For PM2s (partlculate matter :52.5 pm), brake wear accounted for 11% of the 2010 on-road motor vehicle emissions inventory.24 Human exposure to brakerelated asbestos emiSSIOns IS likely to be the highest in urban and suburban enVIronments compared to rural environments.7,g Currently, the proportion of ambient asbestos and the aSSOCiated level of human exposure due to brake wear emissions m CalIforma are unknown. Given that human exposure may be occurnng, and the fact that asbestos IS a known human carc111ogen, an assessment was made to gauge the prevalence of asbestos use 111 automotive brake applications. The purpose of thiS study was to estimate the current level of asbestos use in the Cahforma on-road fleet of light-duty, medium-duty, and heavy-duty motor vehicles, and evaluate the regulatory ImplIcatIOns of the findings. III MATE:HIALS AND METHODS Light-Duty and Medium-Duty Vehicles. Vehicles were procured under CARB's 18th Vehicle Surveillance Program (VSPI8 program), a test program routinely conducted for the purpose of assessmg emiSSIOn levels of hght-duty and mediumduty vehicles representative of the on-road California fleet. Under thiS program vehicles are randomly selected from a registered-vehicles database maintained by the California Department of Motor Vehicles. SelectIOn of approximately .,@. 300 vehicles was based on stratified samphng by model year (i.e., weighted by the number of vehicles still on the road). The model year of the vehicles sampled ranged from 1989 to 2006 and sampling was termmated in July of 2008. The sampling area encompassed an approximate 50 mlle radius from the CARB, El Monte office, located roughly ten miles east of downtown Los Angeles. This area includes all of Los Angeles County, Orange County, and portions of nelghboring counties, which are likely to be representative of urban/suburban populations in southern California. Vehicle selection IS also presumed to include approximate representation across all mcome levels, all manufacturers, all models, and targeted vehicle classes. However, consideration of potential biases m this study is given in a later section. Sampling of these vehicles was halted after 137 vehicles, when this subsample of vehicles was not statistically different (at p :5 0.05) from the target fleet of 300 with respect to model = = year (p = 0.069, t = 2.04, df = 32) and manufacturer (p = 0.403, t 2.03, df 7) based on a paired sample, two-tailed t-test (using the arcsine-square root transformation). Although not statistically significantly different, the mean model year age for the target fleet was slIghtly lower than our sample fleet (1996 versus 1999), with our sample fleet distnbutlOn skewed slIghtly more toward newer model vehicles. Replacement of brake linings for the light- and medium-duty vehicles was performed by certified CARB auto mechanics. Brake lining samples were bagged and labeled with bar-coded vehicle identifiers. The following samples of brake lining were collected: (a) the two inner disk linings from the front axle and (b) the two inner disk linmgs from the rear axle, or in the case of drum brakes, the two primary drum linings from the rear axle. One front and one rear lining from each vehicle was sent the California Department of Pubhc Health EnVIronmental Health Laboratory (EHL) for asbestos analysis, and the other set was retained as "back-up." As part of the vehicle procurement protocol conducted under the VSP18 program, the owner was given a questionnaire to assess use, condition, and integrity of the vehicle. Questions on household income and where vehicle repairs were typICally conducted were also asked. Vehicle owners who agreed to participate in the program were asked if they wanted their brakes replaced at no cost. Regarding the replacement of brake linings in particular, the vehicle owner was also asked If the front and/or rear brake linings on their vehicle were ever replaced and, if so, how long ago. The answer to this question was intended to assist in the determmatlOn of whether the linings on the vehicle were the ongmal hnings installed on the vehicle at the time of purchase, or if they were aftermarket linings. This distinction was considered important because communications with automotive trade groups and manufacturers strongly suggested that both U.S. and foreign auto manufacturers were no longer selling new vehIcles eqUIpped with brake linings containing asbestos. A typical vehicle IS likely to have its brake linings replaced anywhere from 4 to 8 times during its hfetime, and in many cases With the use of aftermarket brake linings (i.e., non OEM). Heavy-Duty Vehicles. The sampling of heavy-duty vehicles was initiated in November 2008 and terminated in October 2009, The combined fleet of 54 heavy-duty vehide~ consisted of line-haul tractors, line-haul trailers, city-delivery vehicles, motor homes, dump trucks, and buses. Due to the large size and mass of heavy-duty vehicle brake linings, two- to four-m ch square pieces were removed for analysis. The sampled pieces 1345 dx dOl org/10 l021/es202667d Ifnvlron So Technol 2012, 46, 1344-1351 Environmental Science & Technology Figure 1. Images for subs ample from brake hnmgs showing (a) sampled drum brake hnmg, (b) sampled disk brake lmmg, (c) 6.3x stereozoom microscope image of subsample from drum brake hnmg, and (d) 32X stereozoom microscope Image of drum brake subsample showmg fiblom asbestos <50 flm from the outer surface. were bagged for transport and labeled with appropriate identifiers. Thirty-sIX heavy-duty vehicles were obtained from one repair shop in the Los Angeles area alongside a major California highway. In all but two vehicles, both the steer and drive axle brake linings were sampled. Twelve samples were obtained from a traIler repair facility located in the Los Angeles suburb of Compton. Samples collected from this facility consisted of stockpiled linings with no associated vehicle information other than the fact they were removed from linehaul trailers. Linings from this stockpile were arbitrarily selected, with the only criterion being a composition visually different from all others, for example, based on presence or absence of copper particles or visually different fibers. Five samples were obtained from a repair shop at French Camp, a community located near the city of Stockton, and one additional sample was from a bus owned by a school district in Anaheim, a city in Orange County. Asbestos Analysis. Asbestos analysis was conducted by EHL, which has state-of-the-art facilities for identification of the following species of asbestos: chrysotile, actinolite, tremolite, crocidolite, anthophyllite, and amosite. A screening procedure was developed based on established National Institute for Occupational Safety and Health (NIOSH) methodology, which allowed for the qualitative identification (presence or absence) of chrysotile asbestos, the most common asbestos mineral used in automotive brake linings. All brake linings were qualitattvely screened for asbestos using techniques specified in NIOSH Method 9002,25 which uses low-power stereozoom microscopy and polarized light microscopy (PLM) and has a detection limit of <1% asbestos content. For light- and medium-duty vehicles, samples of each brake lining were prepared by breakmg off pieces of the brake lining in a ventilatIOn hood with the use of a hammer and awl (Figure la and b). For heavy-duty vehicles, this was not necessary smce these samples were already removed from the backing of the brake lmmg. All brake hnmg samples were then mspected under the stereozoom microscope Using a scalpel and forceps, brake matenal was removed from the brake lining. For the light- and medIUm-duty vehicle brake linings containing asbestos, brake material was removed from three locations in the following manner: (a) by scrapmg from top to bottom along the cross-sectional edge, referred to as "cross section," (b) by scraping the outer fnctlOnal surface, referred to as "outer surface," and (c) by scrapmg the inside surface of a piece broken off from the bottom of the brake lining, referred to as the"subsurface sample." AnalYSIS of these brake lining sybsamples was performed m order to account for any heterogeneity in the spatial distribution of asbestos, the presence of wear dust, and/or the thermochemlcal transformation of asbestos. Brake lining samples were placed in 1.550 refractive index Cargille oil on a slide WIth a covershp. PLM was used to characterize the fibers in the brake material with respect to the follOWIng key opttcal properties: flber morphology, color under plane polarized hght, extmctlon angle under crossed polars, sign of elongation usmg a full wave plate, and birefringence as measured by disperSIOn stammg (Figure 2). If any fibers were "positive" for chrysotlle, the sample was ashed in a muffle furnace at 500 DC for 24 h to remove the interfering organic matrix, thereby facihtatmg improved asbestos Identtfication and quantttation Without denatunng the asbestos. The asbestos content was determmed by 1346 dx dOl org/10 l021/es202667d IEnVlfon So Technol 2012, 46, 1344-1351 ~!l':lronmental SCIence & Technology +1$8 (a) (b) Figure 2. lOOX PLM analysIs of chrysottle asbestos subsampled from a 1997 Oldsmobue Silhouette drum brake lInIng. (a) Posittve sign of elongatlOn under crossed polanzers WIth full wave plate. (b) Characteristic dispersion stainmg colors. Table 1. Summary Data for Light-Duty, Medium-Duty, and Heavy-Duty Vehicles Sampled vehicle type no. vehicles average model year avera~e odometer miles) avera~e GVWR lbs)" no. axles sampled no. samples analyzed no. vehicles WIth asbestos b hght-duty + medlUm- 137 1999 100390 4856 274 274 4 duty heavy-duty Ime haul 13 2004 386070 49881 27 33 0 heavy-duty city dehvery 10 1999 165571 22629 21 36 0 heavy-duty motor 7 2001 58235 16936 14 24 0 home heavy-duty bus 3 1998 heavy-duty trailer" 12" -" 313 852 - -38,667 4 12" 8 12' 0 0 heavy-duty other 9 1999 132205 38289 23 35 0 heavy-duty total 54 2001 219 373 32554 101 148 0 "GVWR = gross vehicle weight rating, defined as the maximum allowable total weight of a road vehicle or trauer carrying a full load, inclUSive of vehicle weight, fuel, passengers, cargo, and traIler weIght. bAll brake lmmgs With asbestos were rear drum brakes. 'Trauer brake lInmgs taken from a stockpile at a traIler repaIr faCUlty and With each lackmg associated vehicle Information. These assumed to originate from different axles based on differences m Vlsual appearance, I.e., ViSUal compositIOn. - ----_.._--_._-- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - measuring the weight percent asbestos in the ash, correcting for only four vehicles had brake linings that contained asbestos, the amount of matrIX removed, and then calculatmg the Implied percentage m the ongmal brake. whIch m all cases was identified as chrysotile asbestos. The heavy-duty fleet consisted of SIX vehicle categories representmg various specialties of use. Of the 42 vehicles for ill HESULTS AND DISCUSSION Fleet Characteristics. Charactenstlcs of the vehicle fleet sampled are given m Table 1. In total, 422 brake lining samples from light-, medium-, and heavy-duty vehicles were analyzed for asbestos. Of the 137 light- and medium-duty vehicles sampled, the average model year was 1999 With an average odometer of 100390 miles, and an average weIght of 4856 lbs gross vehicle rate welghtmg (GVWR), which is defined as the actual vehicle which vehicle information was available, the average model year was 2001, the average odometer reading was 219373 miles, and the average weight 32 554 lbs GVWR. The stockpiled brake linings from the heavy-duty trailers were esttmated by the owner to have been removed from vehicles over the 20052008 time frame, but no vehicle-related information was available for these samples. No asbestos was found in any of the heavy duty brakes that were sampled. Asbestos is used primarily in drum brakes, though web-based weight plus Its maxImum allowable load. Of thiS sample fleet, searches have indicated that asbestos-containing disk brakes are 1347 dx dOl orgl10 1021/es202667d IfnVlfon. SC/ Technol 2012,46, 1344-1351 Environmental Science & Technology Table 2. Asbestos Content and Vehicle Description in Brake Lining Samples with Asbestos" vehicle no. 7Rb model year 1997 vehicle informatIOn mfr model Oldsmoblle Silhouette sample location on lining" cross sectIOn outer surface subsurface mean :::: sample mformatlOn wt % asbestos collection date 26 1/23/07 23 32 27 analysis date 6/7/07 9/9/08 9/9/08 edge coded FASA-RX-Ff 76 1989 Cadlllac Coupe DeVille cross section outer surface subsurface mean = 30 II/S/07 12/24/07 no edge code 20 9/9/08 26 9/9/08 25 116 1990 Ford Thunderbird cross section outer surface subsurface mean :::: 32 3/28/08 4/23/08 FASA-RX-FF 27 9/9/08 30 9/9/08 30 140 1998 Toyota 4 Runner cross sectIOn 33 7/11/08 7/28/09 FASA-OBX-H outer surface 2S 9/9/08 subsurface 29 9/9/08 mean ::;: 29 Grand Mean = 27.5 aVehicles procured under Surveillance Program No. 18, Project No. 2S06CI and Project No. 2R0609. bAn 'R" after the vehicle no. mdlcates that the vehicle was rejected from the study due to excessive mechanical problems. eSee text for details regarding the methods and locatIOn tram which samples were taken. dThe edge code is an identification markmg stamped on the edge of the brake linmg. available on the market. The four vehicles found with asbestos in this study were all equipped with front disk and rear drum brakes, with asbestos present only in the drum brakes. Currently, the majority of light- and medium-duty vehicles have disk brakes on all four wheels. Of the light- and mediumduty vehicles we sampled, 60% employed only disk brakes, with the remainder utilizmg front disk brakes and rear drum brakes. The current study should be viewed as a "snapshot" in time, conducted over a relatively short period of only 18 months for the light- and medium-duty fleet (beginning]anuary 2007) and 12 months for the heavy-duty fleet (beginning]une 2008). The heavy-duty trailer brake linings were estimated to have been removed from vehicles over the 2005-2008 time frame, but with no associated vehicle information available. Thus, cautIOn must be exercised in the interpretation of these data relative to past or future prevalence of asbestos in the on-road vehicle population. Sample Bias. Five possible sources of bias in the light- and medium-duty vehicle populations sampled include: (a) overrepresentation of low-cost vehicles resulting from the offer of a free brake lining replacement and a free "loaner" vehicle, (b) over-representation of high mileage vehicles also resulting from the offer of a free brake lining replacement and a free "loaner" vehicle, (c) under-representation of four-wheel drive vehicles resulting from limitations in dynamometer emissions testing, (d) under-representation of high-cost vehicles from possible owner concern of vehicle damage, and (e) geographic variation, resulting in either an over-representation or under-representation in the supply and thus use of brand-specific aftermarket brake linings. With particular respect to geographical representation, the nearness of a brake repair facility to particular manufacturers and distributors of brake linings could bias the brands of brakes sold mto that local market and sold to a particular repair shop. Investigation into the brake lining "edge codes" indicate that three of the four brake linings containing asbestos originated from the same Canadian manufacturer, With Canada bemg one of the largest producers of asbestos in the world; the manufacturer of the fourth hnmg I, unknown due to an illegible edge code. Smce the study described herem was conducted in southern Califorma, there exists the possiblhty that the mCldence of asbestos limngs may be higher in northern California and the more northern u.s. states since, to our knowledge, no brake manufacturer m the U.S. produces asbestos-contammg brake Immgs. Similar to the light- and medium-duty populatIOn of vehlCb sampled, there was a geographic bias in the heavy-duty vehicles sampled, namely, a bias toward southern Califorma. In additIOn, the heavy-duty vehicles were serviced on site at their respective facility, thus reqUiring the selection of facilities that were wtllmg to coordinate with CARB staff. Besides resulting m additional logistlcal complexities, this also resulted m the inablhty to collect a random sample. A sampling Issue and likely bias specific to the stockpile of heavy-duty trailer brake Immgs was selection of friction material samples based solely on visual differences, resulting in sampling that was not random. Asbestos Content. AnalYSIS of the three subsamples from each of the four light-duty and medIUm-duty vehicles with asbestos brake linings are reported in Table 2, and mclude mean values of the relative asbestos contents for each vehicle These means ranged from 25 to 30 wt %, With a mean of 27.5 wt % (n = 4). A two-way ANOVA (analysis of vanance) was conducted to assess the presence of any statlstlcally significant differences (p ~ 0.05) between asbestos brake hl1lng content of the four vehicles, as well as any differences between the three locations where the brake subsamples were taken (n = 12). No statistically significant differences (F3,6 = 1.76, P = 0252) were found among the four vehicles With respect to asbestos content in the brake linmg samples. However, there was a statistically slgmficant difference (F2,6 = 7.47, P = 0.023) found among the three locations from which subsamples were taken. A posthoc 1348 dx dOl org/1 0 1021Jes202667d I Envlron So Te( hnoJ 2012. 46, 1344~ 13S 1 ~!1Vironmental Science & Technology Table 3. Characteristics of the Four Vehicles with Asbestos Brake Liningsa vehicle model odometer GVWR' no. year (mlles) (lbs) comments 7R" 1997 101947 5357 damage: stains on door, 4 broken Items outside, 13 paint scratches (one deep metal), 2 dents/rungs, one clupped pamt, approXlmately 40 Items mdlcated on damage diagram/report, one deep-metal scratch mechamcal. leak10g radiator, approximately $1300 m reparrs conducted by dealership, nOlse when turnmg sharp left or nght household mcome $35-50K other: vehicle fruled Smog Check on the frrst test, drrty ou, mamtenance typically conducted by garage/semce statIOn 76 1989 167482 4507 damage. carpet stams and tear;, nps on pas;enger door, broken light on dnver's side, steermg wheel damage, some dent, on roof, cracked wmdshield, apprOJumately 20 Items total on damage diagram/report mechamcal: dnver's side wmdow has operatlOnal problems, reparred 2 vacuum hnes and 02 sensor household mcome: $50-70K other' faued two emiSSIOn tests for mtrogen oxides, vehicle mamtenance typically conducted by garage/semce statIOn, owner rarely follows recommended mamtenance schedule 116 ]990 18054] 4885 damage rearvtew mirror fell off, driver's seat leans to left, weak AC, brakes squeak, stams on seat and carpet, approXimately 41 Items total on damage diagram/report mechantcal two drums replaced, rear spring kit, tow wheel cylinder, muffler was repaired, had 011 leak With puddle on ground household mcome: >$90K other' failed one eml~SlOn test for mtrogen oxides and hydrocarbons, low coolant level in reservOlf, mamtenance typICally conducted by garage/service statlOn 140 1998 142465 5250 damage: front wmdshleld pitted, approXlmately 29 Items total on damage diagram/report ~. mechantcal: master cylmder rear right axle seal replaced household mcome: dechned other: mamtenance typically conducted by dealerslup or owner "VehIcles procured under CARE's LIght-Duty VehIcle SurveIllance Program No. 18. b'R" after the vehlde no. mdlcates that the velude was rejected from the study due to the need for excessive mechanical repairs. 'GVWR = gross vehIcle weight ratmg, defined as the maxunum allowable total weIght of a road vehIcle or traIler carrymg a full load, inclusive of vehicle weight, fuel, passengers, cargo, and trauer weight. Tukey HSD (honestly sigmficant difference) test indicated that the cross-sectIOn samples contamed SIgnificantly more asbestos than the outer surface samples, with the subsurface sample asbestos content lying between these extremes. The outer surface was expected to have the least amount of asbestos due to the conversIOn of the asbestos to forsterite through fnctional heatmg, whIch creates a relatively thin, asbestos-poor layer (Figure lc 1d). However, we should note that the p-value of 0.023 mdlcates only a weak trend due, in large part, 1'0 the hmlted saI)1ple sIze. These results mdicate that care should be taken m sampling brake linings for asbestos content by avoiding sample contamination from the brake dust and also by avoiding sampling any subsurface forsterite resulting from frictional heat. The average approximate asbestos content of 27.5 wt % found (n this study can be compared to previous estimates of 20-60 wt %,z2,26-29 However, only a very small fraction of the ongmal asbestos survives the thermal temperatures experienced dunng a brakmg event due to the conversion to forsterite.5 The percentages by weIght of airborne asbestos found in brake wear dust worn from asbestos brake lmings are typIcally less than 1.0 wt %, rangmg from 0.02 to 0.50 wt %,z1,22,30 Variability in these estlmates IS likely due to dIfferences in the asbestos contents of the brake lmmgs, as well as dIfferences m the driving cycles or the number and mtenslty of stoppmg events, and the sampling methodology used to collect and size the asbestos fibers. Such vanablhty exacerbates any attempt to estimate the contribution of motor vehIcle brake wear emISSIOns to the ambIent air concentration of asbestos. Characteristics of Vehicles with Asbestos. In our study, four out of the 137 vehicles sampled (Le" 2.9%) of the hghtand medmm-duty fleet had brake linings with asbestos. Table 3 presents charactenstlcs of the four vehicles found to have asbestos brakes. Two clearly distmguishing factors among the four vehIcles determmed to have asbestos-containing brake friction material were high mileages and earlier model years. In particular, the average model year of the four vehicles was 1994 \ compared to the sample fleet average of 1999. Odometer mileages of the four vehicles averaged 148000 miles, compared to 101000 miles for the total sample fleet average. From the perspective of mechanical condition, one of the four vehicles required so many repairs (totaling approximately $1300), that It was ultimately rejected from the program and never tested for tailpipe emissions. Another metric of assessing the mechamcal condition of the vehicle is the number of VSP18 program emission tests that it fails. Of the three vehicles tested, one failed two tests and two of the vehicles faIled one test. A test engineer for the program qualitatively estimated that 60% or more of the vehicles do not fail any of the emission tests. Regardmg vehicle maintenance, three of the four owners had their maintenance performed at a service station or independent garage rather than a dealership. For the fourth vehicle, maintenance was conducted by the owner or a dealership. Maintenance and repair costs at dealerships tend to be significantly higher than service stations or independent repair facilities. In addition, asbestos linings are less expensive than nonasbestos linings3l and so the owner may have requested less-expensIve brakes. In summary, these four vehicles can be charactenzed as early model, hIgh-mIleage, and mechamcally compromised, with their owners tending to minimize the cost of maintenance and repairs. Based on hIgh mileage and/ or confirmation by vehicle owners, all four vehIcles with asbestos-containing brakes were assumed to have had theIr brakes replaced at least once, thereby indicating a hIgh probability of an aftermarket source of the brake hnings. Although no single characteristic deSCrIbed above would likely serve to distinguish a vehicle with asbestos brake Imings compared to one that does not, the ensemble of characteristics nonetheless serves to typify those light- and medmm-duty 1349 dx dOl orgll0 l021/es202667d 1Envlron So Technol 2012,46, 1344-1351 Environmental Science & Technology vehicles that do have asbestos brake linings. However, this small sample size of four vehicles does warrant caution in extrapolating these results to a larger fleet. These characteristics are not likely to be useful in predicting if a given vehicle will have asbestos brake linings, since many vehicles may have the above traits, but only a very few of those are likely to have asbestos-containing brake linings. The lack of asbestos-containing brake linings in the heavy- duty vehicles sampled may be due to the low sample SIZe of 54 and/or the result of heavy-duty brake specifications that preclude the use of asbestos. There is also a possibility of some bias in the heavy-duty brake samples collected, since front tractor brake linings may be of identical composItion to the rear tractor brake linings because they typically have the same specifications and are often replaced at the same time with the same brake linings. The same is often true for trailer brakes; when trailer brake linings are replaced, all are typically changed at the same time. There is also a sparse representation regarding the variety of use in the total fleet sampled, with only three buses sampled and many other specialty vehicles not included. Regulatory Implications. In the mid to late 1980s, brake linmg manufacturers began to develop alternatives to asbestos, motivated by litigation and the aforementioned asbestos bans in European countries and an imminent USEPA ban on asbestos use in friction products. However, following the ban enacted by the USEPA in 1989 and the court's subsequent reversal of that ban in 1991,10,11 auto manufacturers and the aftermarket brake lining industry continued to use asbestos. In 1998, it was estimated that approximately _~~.-S'f~e l~ht-dl!~and __ j ~!ll-dq!c.'ehlcles ~old_}~~~ U'~2 were equil'y~d _-~;t~~ OEM drum brake~C()ntaining asbestos. At the same time, the attermarket -We industry in the U.S. was reported as being approximately 10 times the size of the OEM market. 12 The consensus among automotive trade associations is that no U.S. automotive manufacturers currently use asbestos in OEM brakes, and that no foreign vehicle manufacturers are selling brakes with asbestos in the U.S. market. However, without a federal ban aftermarket providers of brake linings throughout the U.S. are likely to continue to include asbestos- containing linings within the mix of their product offerings. The future trend nationally in the use of asbestos is likely to depend on availability, cost advantage, and the likelihood of a ban on importation and sales. Although our study identified only four vehicles with asbestos-containing brake linings, all four were hIgh mileage vehicles with a high likelihood of aftermarket brake linmgs, with three of four having linings from the same Canadian aftermarket manufacturer. In late 2010, after completIOn of our study, _Cal~ signed mto law Senate Bill (SB) 346,16 which specifie! a de mimimis level o~ by ~eight of toxics in motor vehicle bra!<e-f'ffction materials, including aSbeStIf'Omlfibers, as well as compounds of cadmium, chromium(VI), and mercury. This regulation (takes effect on )anual}'l~ and essentially bans asbestos and these other toxic compounds in brake lining formulations for vehicles sold in the state. At this time, California appears to be the only state to ban asbestos use in automotive brake lmmgs. Our sampling program found that only 3% of light- and medium-duty vehicles and 0% of heavy-duty vehIcles had brake linings WIth asbestos, supporting the contention that the U.S. is currently experiencmg a downward trend in the use of asbestos in brake linings. Although there have been repeated calls by national and international organizations to implement a worldWIde ban on asbestos products, most recently by the Collegium Ramazzini (a nonpartIsan internatIOnal academIC society that examines public health issues), no U.S. or worldwide ban appears forthcommg 32;33 Thus, the momtonng of asbestos use and other tOXlC materials m brake hnmgs IS a prudent course of action. In this context, ongoing samplmg programs will be required to assess the prevalence of asbestos and other toxic substances present m the brake dust released from the operatIOn of on-road vehicles, and adherence to state regulation. AUTHOR INFORMATION Corresponding Author -~ *Phone: 626-350-6469; fax: 626-459-4480; e-maIl: lrao@arb.ca. /)-. gov. /" ACKNOWLEDGMENTS John Urkov, Raphael SusnoWltz, Tom ValenCIa, Don Cherlllch, Brian Weitz, Jerry Ho, and Mike O'Connor were very supportive of this project and assisted WIth fundmg, sample collection, and logistics. Ron Haste served as the backup engineer for this program. Marco Dlaz and Carl Law were very helpful in performing the brake lining replacement. The following staff provided excellent field mpport for the collectIOn of heavy-duty vehicle brake samples: Bob Torres, Mark Burnitzki, Gene Sato, Manny Arevalos, Javler Chavez, Esther Gonzales, William Intenano, John O'Cam, Gary Okimoto, and Luis Ordonez. Daimon Pon from the CaiIforllla Department of Public Health proVIded custom bar-coded sample labels. REFERENCES (1) Vlrta, R. 1. WorldWide Asbestos Supply and ConsumptIOn Trends from 1900 through 2003, US. Geological Survey Circular 1298, UOlted States GeologICal Survey: Reston, VA, 2006; http';/pubsusgs.gov/ circ/2006/1298/c1298.pdf (accessed Apnl 6, 2011). (2) Blau, P. ]. CompOSitIOns, FunctIOns, And Testing of FrictIOn Brake Materials and Their Add,tives, ORNL/TM-2001/64; Oak RIdge NatIOnal Laboratory: Oak Ridge, TN, 2001; http://www.ornl gOY / ~webworks/cppr/y200l/rpt/1l2956.pdf (accesset! Apnl 6, 2011). (3) Moore, P. Chrysotile in CriSIS. Industrial Mine'als MagaZine, Apnl 2004, 56-61. (4) Castleman, B.; Camarota, L. A.; Fritsch., A ].; Mazzoechl, S., Crawley, R. G. The hazards of asbestos for brake mechaOlcs. Pubilc Health Rep. 1975,90 (3), 254-256. (5) Lemen, R. A. Asbestos 10 brakes' exposure and nsk of dIsease Am. J. Ind. Med. 2004, 45 (3),229-237. (6) Welch, L. S. Asbestos exposure causes mesothehoma, but not thIS asbestos exposure: An amICus bnef to the MichIgan Supreme Court. Int. J. Occup. Environ. Health. 2007, 13 (3),318-327 (7) Llm, H.; Kim,]. Y.; SakaJ, K.; Hlsanaga, N. AIrborne asbestos and non-asbestos fiber concentratIOns 10 non-occupatIOnal envIronment> 10 Korea. Ind. Health 2004, 41, 171-178. (8) Kohyama, N. Airborne asbestos levels III non-occupatlOnal environments in Japan. In Non-OccupatIOnal Exposure to Mineral Flbers (lARC SCientifiC PublIcatIOns No.90); Blgnon, ]., Peto, J., SaraCCl, R., Eds.; International Agency for Research on Cancer' Lyon, 1989; pp 262-276. (9) Kazan-A1len, L Current asbestos bans anc restnctlOns, 2011 InternatIOnal Ban Asbestos Seeratanat. http.; /Ibasecretanat org/ alpha_ban_hst.php (accessed October 6, 2011). (10) Asbestos: Manufacture, Importahon, process1Og, and dlstnbuhon in commerce prohIbitIOn. Code of Federal RegulatIOns, Part 763, Title 40, Subpart I, SectlOns 762.160 - 763179, 1989 1350 dx dOl orgl1 0 , 021/es202667d I Envlron So Telhno/ 2012, 46, 1344~ 1351 [:!~.v.2!.()nm'~ntal S( lence & Technology (ll) CorrosIOn Proof Fittings, et al., PetitIOners, v. the Environmental ProtectIOn Agency and Wdllam K. Rei/ly, AdminIstrator, Respondents, No. 89-4S96 (Sth Clr. Oct. 18, 1991). (12) The Friction Product and Materials Market; GB-212; Busmess Communications Company, Ine Wellesley, MA, 1998. (13) V.S. Environmental ProtectlOn Agency. Release of draft brochure on current best practlces to prevent asbestos exposure among brake and clutch repair workers: NotICe of avallablllty. Fed. Reglst 2006, 71 (164), S0060 - S0062. (l4) Asbestos. Stdl Lethal/Stdl Legal - The Need to Better Protect the Health ofAmerican Workers and Their Families; Subcommittee Hearing Statement of Barry Castleman, Environmental Consultant; Senate Health, Educatlon, Labor, and PenSlOns Committee, Subcommittee on Employment and Workplace Safety: Washmgton, DC, 2007; http:// help.senate.gov/lmo/medla/ doc/ Castleman.pdf (accessed October 10, 20ll) (15) Mlchaels, D.; Monforton, C. How htlgatlOn shapes the scientific hterature: asbestos and disease among automobile mechanics. J. Law Policy 2007, 5 (3), ll37-1169. (16) Hazardous matenals: motor vehicle brake friction matenals. Senate Bill 346, Kehoe Califorma Health and Safety Code, Chapter 6.5, DlVlSlOn 20, Article 13.5, SectlOns 25250.50 - 25250.65, September 27, 2010; http//e-lobbYlst.com/galts/vlew/62361 (accessed May 24, 20ll). (17) Osmubl, O. Y., Gochfeld, M.; Kipen, H. M. Health effects of asbestos and nonasbestos fibers. Envzron. Health Perspect. 2000, 108 (Suppl 4), 665-674. (18) Detalled procedure for asbestos sampling and analYSIS - nonmandatory. Code of Federal Regulations, Part 1910, Title 29, Subpart Z, SectIOn 1910.1001 Appendix B., 1996; http://www.osha.gov/pls/ oshaweb/owadisp.show_ document l p_ table=standards&p_ld=9997 (accessed Apnl 6, 2011). (19) Sanders, P. G.; Xu, N.; Dalka, T. M.; Maricq, M. M. Airborne brake wear debns: SIZe dlstnbutlOns, composition, and a companson of dynamometer and vehICle tests. Envlron. SCI. Technol. 2003, 37 (18), 4060-4069. (20) Garg, B. D.; Cadle, S. H.; Mulawa, P. A; Grobhcki, P. J; Laroo, C; Parr, G A Brake wear partlculate matter emiSSIOns. Envzron. Sc!. Technol. 2000, 34 (21), 4463-4469. (21) Cha, S; Carter, P.; Bradow, R L. SImulation of automobIle brake wear dynamiCs and estimatlOn of emlSSlOns. SAE Tech. Pap. Ser. 1983, SAE 831036, 1-21. (22) WlIhams, R L.; Muhlbaler, ]. L. Asbestos brake emlSSlOns. Envlron. Res. 1980,29 (1),70-82. (23) Sanders, PG.; Dalka, T. M.; Xu, N.; Mancq, M. M; Basch, R H. Blake dynatnometer measurement of airborne brake wear debns. SAE Tech. Pap. Ser. 2002, SAE 2002-01-1280, 1-8. (24) EMFAC2007, verSlOn 2.3; Callforma Alr Resources Board: Sacramento, CA, 2006, http./ /www.arb.ca.gov/msei/mseLhtm (accessed Apnl 7, 2011) (25) Asbestos (Bulk) by PLM 9002. NIOSH Manual of Analytical Methods, NIOSH Pubhcahon Number 2003-154 (Third Supplement); Center for Disease Control and Prevenhon: Atlanta, GA, 2003. (26) Morshed, M. M., Haseeb, A S. Physical and chemical charactensttcs of commercially avallable brake shoe hnmgs materials: A comparative study J. Mater. Process. Tech. 2004, 155-156, 1422- 1427. (27) Salto, T. Indirect estimatIOn of asbestos concentration due to vehIcles 111 partlculate matter m the atmosphere and on roads by pyrolYSIS - gas chromatography. ] Anal. Appl. Pyrol. 1995, 32, 177- 178 (28) Rowson, D. M. The chrysotlle content of the wear debns of brake hmngs. Wear 1978,47 (2), 315-321. (29) Wall, S. DetermznatlOn of Asbestos Content of Current Automotzve Dry FrictIOn Materials and the Potenhal ContributIOn of Asbestos to Partlculate Matter Denvedfrom Brake Wear, Contract No. 01-333 Fmal Report; Cahforma Atr Resources Board Research Division: Sacramenta, CA, 2007. .fiWW (30) Jacko, M. G.; DuCharme, R. T.; Somers, J. H. Brake and clutch emissions generated dunng vehicle operation. SAE Tech. Pap Ser. 1973, SAE 730548, 1-20. (31) Asbestos Materials In the AutomotIVe Maintenance and RepaIr Industry, GUidance Note; Chatnber of Commerce and Industry Risk Solutions: East Perth, Australia, 1999. (32) Collegium Ramazzini. Asbestos IS sttll With us: Repeat call for a universal ban. Arch. Envlron. Occup. Health 20lO, 65 (3), 121-126. (33) LaDou, J.; Castleman, B.; Frank, A; Gochfeld, M.; Greenberg, M.; Huff, J; Joshi, T. K.; Landngan, P. J.; Lemen, R; Myers, ]., Soffrittl, M.; Soskolne, C. L.; Takahashi, K.; Teltelbaum, D., TerraClm, B.; Watterson, A The case for a global ban on asbestos. Envlron. Health Perspect. 2010, 118 (7), 897-901. f)M~!ir7/ -1 /U[) k b~ . /~., " . /j/6: d) ,. .); /7/),' ; .v ' ~~).{.n::;lft'/?~'?c.cl~/' y; t r:;t ? d Un1t,ifVj;/ YJ/~ ,vtf'7 I f A v J {.- G {lVl /' / /I:9j/t(Y/"'~'/ tv'A '/ ~(Y//,t i f"/~i<' It, , / L.-t J t..-tA.:..J (,' '" ;-Y11 /// 1:t' /' ()/tv-. / ? / 4/ J / /,/V I /l,./~~ /t/~l ~ //1'fC' ,(,/ .' L/ t) I d<-f';jlI1VfI./;yI',;/ ,,>1 ~~/ITyJ',1/-t?/1~J/?L j;;; ~t,~~jv"" ~I ~,~/~ i~'~ .,/V'~. v'Jv~~ttf~,1 ptV"';'/1;iv 1351 dx dOl orgl10 l021/es202667d I EnVlfon Se/ Techno/ 2012, 46, 1344-1351