Document 2NbedZJvboym4zxEB0RwNy407

FILE NAME: Asbestos in Hair Dryers (HD) DATE: 1979 Apr 27 DOC#: HD024 DOCUMENT DESCRIPTION: Unpublished Report of NBS to CPSC with Cover Letter - Report on Hair Dryers, Test Method Development / V,\ \ / V'' UNITED STATES DEPARTM ENT OF COMMERCE National Bureau of Standards Wasrimgion. D.C. eQ22l May 2, 1979 MEMORANDUM FOR-Dale Scott, CPSC * Attention: Gail Wyer Through: William West, C . ->7" Deputy AED (Engineering Walter G. Leight, Chie. Product Safety Technology From : Sid Greenwald, --* Product Safety Technology Division Subject: Examination of Hair Dryers from CPSC (#7166 and #7167) for Asbestos On March 26, 1979, the Consumer Product Safety Commission delivered two hand-held hair dryers to NBS and requested an examination to answer the following questions: Ql. Does the insulating liner surrounding the heater element contain asbestos fibers? Al. Scrapings from each dryer were immediately examined by electron microscopy and were found to be primarily chrysotile asbestos. Q2. Does the effluent air from the operating dryer contain chrysotile asbestos fiber? (The switch setting giving the highest air volume rate would be used, since this would maximize any possible cr erosion from the surface of the liner material.) . , Quantitative results were not required, but it w ^ -- ~ ^ requested that the size range of any detected asbestos fibrils, bundles, or clumps be noted. i*: . ~ ' A2. Small amounts of chrysotile asbestos were detected ' JL in the effluent from each of the hair dryers, some rr- in the respirable size range. However, similar -- amounts and size distributions were found on : control (blank) filters used to collect samples without hair dryers. GI00072 2 IC should be noted that the results summarized in this memorandum and its attachments are:based on limited experiments vith but two hair dryers, both of which were brand new at the initiation of this task. Obviously, results cannot be generalized for other devices, nor is it possible to state at this juncture whether degradation might occur in the course of time. The results of the investigation are given in the attached two part memo report. The first part deals primarily with the methods used to sample the air from the two hair dryers. In addition, some brief experiments were run, noting the effects .of temperature on a typical insulating liner. " In the first sampling method used, only a very small percentage of the air coming from the hair dryer was captured by a polycarbonate membrane filter. The remainder of the output escaped to the open air. This method was employed because of back pressure and overheating difficulties originally encountered in trying to develop a total air capture system. Four air samples were taken in this manner, and sent to the Gas and Particulate Science Division (GPSD) for analysis. In the second method, all of the hair dryer output was captured by a 20 X 25 cm (8 X 10 inch) polystyrene filter. Four trials were run as follows: ` Trial__________Dryer_________ Time 1 Sample A (#7167) 4.5 hrs. 2 Sample B (#7166) 4.5 hrs. 3 Blank 16.0 hrs. 4 Blank 4.0 hrs. The two blanks were run, each in a slightly different manner, in order to determine the asbestos background level. These would serve as the control. ' All four samples were sent to the GPS Division for analysis. 'Throughout the air sampling procedure, every effort was made to keep the filters as free as possible from any outside contamination. GI00073 3 The second pa t t o f this report deals with the results of the analysis made by scientists In the Gas and Particulate Science Division. The Instrumentation Involved both Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM). The TEM Is considered superior for this investigation because of its ability to detect smaller fibrils due to its higher resolution capability. Attachments cc: W. Porter - CPSC S. Warshaw - NBS J. Small - NBS S. Toner - NBS H. Rook - NBS J. Hodgeson - NBS GI00074 REPORT ON HAIR DRYERS TEST METHOD DEVELOPMENT Prepared by . Samuel D. Toner, Div. 763 April 27, 1979 On March 26, 1979, two hand-held hair dryers were delivered by the Consumer Product Safety Commission. These dryers were said to contain a thermal Insulating material composed, at least in part, of asbestos. The insulation in question was a cylindrical sleeve of paper-like material inserted in the section of the dryer barrel surrounding the heating element. NBS was requested to determine, if possible, whether asbestos fibers and/or fiber bundles are released from the insulating material and emitted into the air stream of the dryer during operation. In addition, estimates of the range of size of fibrils and bundles was desired: For the purposes of this report, the two commercially available hair dryers are identified as Sample A (CPSC #7167) rated at 1200 watts, and Sample B (CPSC #7166) rated at 1000 watts. A third dryer, Sample C, produced by the manufacturer of Sample A, was obtained from the NBS Product Performance Engineering Division. These three dryers were sent to the NBS Fluid Engineering Division for measurements of the air flow capacity. These measurements were made using a Thermo-Systems Inc., Ionflo Meter, S/N 074, Range 0 - 150 cubic feet per minute i). The results obtained at maximum air flow capacity were as follows: Sample A, 35.5 CFM; Sample B, 28.0 CFM, and Sample C, 25.0 CFM. Insulating Material ' ' There is some concern'that, during the life of a dryer, any binder present in the insulation might be decomposed by the heat generated during normal use, thus possibly leading to an increase in the number of fibers that might be released into the air stream. Consequently', the Insulating liner was removed from a new dryer identical to Sample C. When a piece of this material was subjected to a bunsen burner flame, charring occured immediately, indicating the presence of a binder. A few minutes of additional heating burned off the char, leaving the paper brittle and friable. A section of the insulation was dried to constant weight in a circulating air oven at 110*C. After one hour the material exhibited a weight loss of 0.37%and after an additional 0.5 hour at 110*C, exhibited a total weight loss of 0.38Z. The specimen was then placed in a circulating air oven for 4 hours at 300*C. On removal, the material exhibited some slight discoloration, but no evidence of charring, and an additional weight loss of 2.78%. The specimen was reexposed for an additional 6 hours at 300*C, after which time the total weight loss was 2.88%, based on the constant weight after conditioning at 110C. There was no apparent additional discoloration of the material. There was no apparent evidence that the insulation had become friable, a condition which might result in release of larger quantities of fibers with continued usage. Thermal measurements were then conducted on Sample C, by placing a v -rmocouple against the inner surface of the insulating material, and measuring the temperature rise during normal operation of the dryer, with the controls set GI00075 2 . ... ,,,,j m f l o w rate. The temperature of the insulation for maximum temperature and highest iiow rat , nf did ,,ocP.,ced 38-C ^ i e v ^ d u r i o s the r - t . of^ rise to 50* - 55*C over the next 20 - 30 seconds, and then cooled to ambient temoerature in approximately 10 minutes. Temperatures in the air stream of this dryer, measured at random points just inside the protective grill. w " e . observed to be in the range of about 65* - 80'C. It was concluded that, if the insulation used in this dryer were typical to the dryer industry, then therms, decompositon of the binder due to dry heat alone was considered unlikely. Test Method Development . - Although several potential methods for collecting emitted fiber, if ay were considered, it appeared desirable to attempt to develop a system in which all of the air emitted by the dryer passed through a filter medium. Such closed-system technique vas expected to minimize potential loss of respira 1 fibers and fiber bundles, and to provide a better assessment of the amount and range of sizes of the fibers. The NBS Gas and Particulate Science Division (GPSD) provided a polystyrenefiber filter medium suitable for collecting respirable asbestos fiber and also amenable to further sample preparation procedure required prior to analysis by means of the electron microscope. A new dryer, identical to Sample C, was used in developing the test method. The dryer was attached to one end of a section of rigid, 2-inch inside diameter tube. A water-filled manometer was attached to the tube as a means of monitoring any changes in the air flow capacity of the dryer that might result from the air impinging on the filter medium. Since the techniqes required to prepare a sample for the electron microscope are simplified by use of *Z " 11 fJ1 . * the first attempt simply involved covering the open end of the tube with a Piac* . of the filter. The available surface area of the filter was appraataately 15 2 cm^ (3.9 in2). In this case, back pressure in the tube, created by the low air flow rate through the filter, caused rapid overheating of the dryer with subsequent activation of the thermal overload switch. This event occurred in about 7 - 8 seconds. The test equipment was then modified to accept a larger il" , * 73.6 cm2 (11.4 in2) in area. Although the back pressure problem was alleviated to some extent, the operating time prior to activation of the thermal overload switch was increased bv only 6 - 8 seconds. Further increase in the size of the filter to about 171 cm2 (26.5 in2) in area did not result in a sipiificant improvement in the procedure. Additional modification of the equipment invo ve the attachment of a section of 2-inch tube on the downstream side of the filter, a DC motor and fan assembly, removed from a dryer identical to Sample , was sealed into the open end of the tube in an effort to induce a negative pressure on the downstream side of the filter as a potential means of balancing the pressure on the upstream side of the filter. This did not result in a significant improvement over the preceding methods. Substitution of a vacuum pump for the ^-yer motor/fan assembly also failed to solve the back pressure problem. Parc GI00076 3 the problem seemed to be due to the difficulty in obtaining air-tight seal, on the downstream side of the filter in this temporary set-up, and pafc 1 appeared to be due to the difficulty in achieving a d L n s t r e ^ air flow canacitv of a sufficient nature to off-set the pressure dfop across the filter P ^ teaoorariw* P01?*' ?rk thft closed-system collection technique was fiandned ac the request of CPSC staff members, in favor of a b p m: ` " da t h a t b * * > . 4 as =; * j- ibLh: Ib. hair 4*7 was placed ia a hoo4 capable of providing a positive flow o " l L t e V s " f r o V ^ * rblni*k* T " - Th' ^ P . S S * 2 i i S d ,f ITTM' r L: - "iaHd^e *- of air fin er ahaf the filter surface was approximately normal to the direction to the fllter"h o L e r Cp r o v i d i n p'res'ur.lrop'.cS'I t h ^ U t e r V<,T ; T e f iecci" " 11" 1 Cryirs^"^ ' ! ' 01 th' i Ile" *urfac* o< the order of 5.0 on* (0.78 in*) * ' fnHrswith l r s s s s nr hr s ;iio,i!- "s i* - A staff of the CPS Division Indicated "f sb" s f* p-- si2 eofro" dHi!r * contlnuous 8-hour period using a membrane filter with a rn_ evelPental work was then resumed on a closed-system collection technics, r % T \ s Z ; iia l ^ : r r i Po iL r s 1hi aiefa ^ iL i rc%d^ci u i : sfofter holde; S ^ i o ^ c h piece of^he0 ?** Th# a;cooPanyia8 filter holder required a nominal collecting area of 0 " i'a TM o r ^ " *VaUabl* !"r>" s:es iin. c o : ^ a ^ : " "o1c!dr7h:eu" du`n'Suchdr ted' a i c s -s a n s ; -- - - technigue. When the fan va, operating s r - i n c . "" rate^u'wLlapabL aooroximacpiv k i j n8ativ * pressure on the downstream side of the filter pp ximately balanced the pressure caused by the air flow rate of rh v the upstream side of the h h , r___ W air n o w rate of the dryer on GI00077 4 pressure on the downstream side of the filter was only of the order of 1 - 2 mm of water pressure greater that the positive pressure produced by the air flowing from the dryer. Subsequently, tests were conducted on the dryers Samples A and B were;individually subjected to 4.5 hours of continuous opiration. Control tests to determine the presence of any background asbestos were also t c Ut: i r v i ' f thT " ** . period ol 16 h o u r " in" h i . test the dryer was replaced with an extension tube and a linear air flow meter so that the air flow impinging on the filter surface was intermediate to that ' of Samples A and B. The second test was conducted for four hours. In this test the cover plate was not used so that the entire available surface of the filter was exposed to the flow of filtered air emitted by the hood. All filters were asbestos^* ^ DiVisiQ for 8ubae*ut analyses for the possible presence of M GI00078 "800010 (!!*) US. DEPARTMENT OF COMMERCE NATIONAL. BURKAU OP TANDAROS WABMINOTON. O.C. WLM ;.. REPORT OF ANALYSIS Of Electron Microscopy on Lining Materials from Hair Dryers and Filter Samples Submitted by Sid Greenwald, Division 763 Laboratory Nos. Project No. Requisition NO. Analysts 553 39024 - 39029 763-1411 763-0936 . . John A. Small and Eric B. Steel Request: The laboratory was requested to analyze the lining material from two hair dryers and determine if the material contained asbestos. In addition, the laboratory was also requested to determine if eight filter samples, four poly carbonate and four polystyrene, contain asbestos and to qualitatively estimate fiber loading and fiber size for each sample. Sample Preparation: Lining material. Each dryer was disassembled, and a piece approximately 5 mm2 was removed from the insulating sheet. These pieces were placed in separate vials. Approximately 25 ml of ethyl alcohol was added to each vial. The vials were sealed and shaken both by hand and ultrasonically for a period of about 10 minutes, in order to free some of the fibers from the piece of insulating sheet. Next, 200 mesh Cu TEM grids coated with 10 nm carbon foils were dipped into the solutions. After drying, one grid from each sample was investigated with a transmission electron microscope (TEM). The TEM was operated at 125 kV accelerating potential. ' . - Polycarbonate filters. The polycarbonate filters were coated with approximately 20 nm of carbon. After coating, two 5 m m 2 portions were cut from each filter and placed in a Jaffewick washer. One of the 5 mm2 sections from each filter was placed in the washer with the carbon side up; the other was placed with the carbon side down. Once the filter material was cleared, the grids were investigated with the TEM. GI00081 Polystyrene filters. One-fourth of each 8" x 10" poly styrene filter was dissolved in approximately 500 ml of xylene. When the filter had completely dissolved, the solution was * filtered through a 0.2 um polycarbonate filter, which was backed with a teflon filter. After filtration, the filter was washed with an additional 25 ml of xylene and allowed to air dry. Next, the filters were carbon coated, and a 5 mm2 portion of each filter was removed for scanning electron microscopy (SEM) analysis. The remainder of the filter was prepared on a TEM grid with a condensation washer. The grids were then analyzed at 125 kV with the TEM. There was no evidence of particle loss duetto the condensation washing. The 5 mm2 portions removed for SEM analysis were mounted on aluminum stubs with double-sided tape and carbon coated with approximately 10 nm of carbon. The SEM was. operated at 7.5 kV accelerating potential. Results : The samples used in the TEM were scanned at about .. 5,000 to 20,000 times magnification for fibers. Those used in the SEM were scanned at about 1,000 to 10,000 times magni fication for fibers. Fiber identification was based on morphology and selected area electron diffraction in the TEM, and morphology and energy-dispersive x-ray analysis in the SEM. Detailed analyses were not done on all fibers. Representative fibers from each sample were chosen at random for detailed analysis, and the remaining fibers were identified by morphology only. Lining material. The lining material was analyzed by TEM only. The results are shown in Figures 1 - 4 . Figures 1 and 2 show fibers with the characteristic morphology associated with chrysotile, very thin fibers (less than 50 nm wide) with hollow tubes running down the center. Figures 3 and 4 are selected area electron diffraction images of representative fibers. These images show the slurring of the diffraction spots which is also characteristic of chrysotile. Nuclepore filters. Sample 1 2 3 4 CPSC CPSC CPSC CPSC 7167, 7167, 7166, 7167, Label 8 hr. test, 2 hr. test, 2 hr. test, 1 hr. test, 0.8 um filter 0.1 um filter 0.1 um filter 0.8 um filter The TEM and SEM results from these filter samples were in conclusive. .A few fibers were seen on the TEM and identified as chrysotile. However, there were not enough fibers prp^snf on any of the samples to conclude that the loading was above blank values. Delbag filters. Sample 5 6 7 8 Label CPSC 7166, 4.5 hr. test CPSC 7167, 4.5 hr. test Blank for Background Asbestos, Blank for Background Asbestos, 16 hr. 4 hr. test test GI00082 All the samples and blank;! showed the presence of small amounts of chrysotile, the fiber loading on Samples 5 and 6 was not significantly higher than on Samples 7 and 8. The results from the SEM analyses are shown in Figures 5 - 8 . The x-ray spectra show that the fibers contain Mg and Si.^ These two elements are the major constituents of chrysotile. In several of the fibers, however, the ratio of Mg to Si is, to a first approximation, higher than expected for chrysotile. This can be attributed to several factors such as the possibility of a binder material mixed with the chrysotile or residue from the filter dissolution. As before, the fiber loading on Samples 5 and 6 was not significantly higher than the loading on Samples 7 and 8. Conclusions: From the results of the microscopic analyses, the following conclusions can be drawn. ^ The lining material in both the dryers is primarily chrysotile. 2. The polycarbonate and polystyrene filters, both samples and blanks, contain small amounts of chrysotile. 3. TEM and SEM analyses indicate that the polycarbonate and polystyrene filters from CPSC 7166 and 7167 do not' - contain a significantly higher loading of chrysotile than the blanks (polystyrene filter 7 and 8). 4. Based on the qualitative analysis, there was no apparent difference in the fiber,size distribution between sample and blank. / Research Chemist Acting Chief Gas & Particulate Science Division "Philip p\ LaFleur Directomj Center for Analytical Chemistry Robert W. Burke Service Analysis Coordinator Center for Analytical Chemistry 79-92 TCI analysis of hair dryer insulating material ^l5 cp'OiMi & 'Vi'd? f^Ssi.% ySi; 'JZ--r ^ z Z ^ ^ C - -'' '^.VV^- - 1.0pm Figure one: TC! inagc of insulating aaterial froa CPSC 3 7166 SSSS^K 24^933 -rtoTSi is is XXZB. &SQgfc& A - 0.2 p m *^r`-WK *v: pVCVf J > B g ^ ^ L a i i ! ^ ^ s a i e p ^ j a i B Pg Figure 2: TCI iaage of insulating aaterial fro- C?3r. ?7167 it^4 GI00085 i