Document dY3VV9mxe6DGDqg23QvMZ7DJB

0641688 TASK ORGANIZATION REPORT EPA Contract No. 03-01-5915 October 14, 1980 (Revised October 20, 1980) TASK 32 - STUDY PROTOCOL FOR THE COLLECTION AND ANALYSIS OF VERMICULITE AND RELATED SAMPLES FOR THE EVALUATION OF FIBER CONTEXT WITH EMPHASIS ON ASBESTIFORM FIBERS I. Background In December 1978, the venniculite industry submitted information to the EPA regarding health problems experienced by employees who were pro cessing asbestos-contaminated vermiculite. The original submission indi cated that bloody pleural effusions had been detected in 4 of 350 employees; symptomatology and clinical findings in the employees were similar to those found in individuals with asbestos-related diseases. Subsequent follow-up studies by the Occupational Safety and Health Administration (OSHA) revealed an even higher prevalence of health problems among the employees. Venniculite is a hydrated magnesium-iron-aluminum silicate which has been mined in the U.S. since 1929. W. R. Grace & Company ("Grace") pro duces most of the 300 thousand tons consumed domestically. After mining, vermiculite is usually processed to remove impurities; however, all contam inants are not removed. Such is the case with the vermiculite from the Grace Libby Montana mine which historically has had at least 11 tremolite asbestos remaining as a contaminant in their vermiculite. Recent technical advances have reportedly reduced the levels of tremolite. Information has been received which suggests that the three major domestic deposits in Montana, South Carolina, and Virginia contain asbestiform minerals. Authorities in geology have expressed the opinion that all vermiculite mines may be contaminated with asbestiform minerals, including the South African eine which claims to be free of asbestiform minerals. Although vermiculite may contain fibrous materials, the health effects from vermiculite itself are unknown at this time. A priority review of asbestos-contaminated vermiculite, completed by the Office of Testing and Evaluation in June 1980, suggested that the asbestos in vermiculite may be responsible for the reported adverse health effects and it concluded that certain information gaps need to be filled before an indepth risk assess ment on vermiculite could be initiated. Several projects have been initiated to fulfill the information gaps and complete the preregulatory analysis on vermiculite. A control op tions analysis has been initiated to determine regulatory strategy to con trol asbestiform mineral-contaminated vermiculite and a substitutes analysis is in preparation to evaluate replacements for vermiculite products. Work has also been initiated on a materials balance to show the mass flow of vermic ulite along with the release of any associated asbestiform mineral and the development of a mineralogy profile with a sampling and analysis protocol of vermiculite is underway to characterize the fibrous materials within vermiculite. 1 0641689 From the available information on the composition of vermiculite, it seems there is the possibility that contamination of vermiculite does occur. For example, in their comments and response to the Consumer Product Safety Commission and the Environmental Protection Agency's joint-effort Announced Notice of Proposed Rulemaking on asbestos, Grace explains that the inadvertent mineral contaminant form (tremolite), which is generally associated with vermiculite appears in a fibrous and nonfibrous form. This type of information suggests that it may be difficult to assess the magnitude of the contamination of the vermiculite. Therefore, the objective of this protocol is to specify the sam pling and analysis procedure to determine the composition of vermiculite, particularly the amount of asbestiform minerals present in the vermiculite. This will provide the needed information on the risk to the population ex posed to asbestiform minerals from vermiculite at each of the various stages of its commercial distribution. Included in the protocol will be an indepth analysis of the asbestiform fibers present in the vermiculite. The protocol will address: 1. The selection, collection, and analysis of bulk samples of: (a) raw vermiculite ore samples from the four major vermiculite mines in the U.S and from the ports of entry; (b) beneficiated vermiculite ore samples v from the four major vermiculite mines in the U.S. and from the ports of entry; and (c) exfoliated vermiculitp ore samples from a representative number of exfoliation plants in the U.S. The exfoliation plants where- sampling will occur will be statistically chosen by Exposure Evaluation Division (EED) to include all the major sources of vermiculite. 2. The selection, collection, and analysis of air samples taken at sites associated with (a) the handling of raw vermiculite ore at the four major vermiculite mines in the U.S.; (b) the beneficiation of vermiculite ere at the four major vermiculite mines in the U.S.; and (c) the exfoliation of vermiculite ore from a representative number of exfoliation plants in the U.S. The exfoliation plants where sampling will occur will be statis tically chosen by EED. II. Preparation for Sampling A. Inventory and Preparation of Supplies 1. All necessary equipment for sampling bulk vermiculite and air will be gathered and inventoried. 2. Filters to be used for the collection of airborne particles will be assembled and labeled before samples are obtained. 3. Calibration of the pumps will be performed prior to their shipment to the sampling site. 4. All supplies will be packed and shipped to the vermiculite sampling site at least 2 days in advance of the arrival of the crew. 2 - Jim a m mmw ra 0441690 S. Replicate prenuobered sample identification labels will be prepared for field application to sample containers and the field notebook. B. Site Investigation 1. Survey site - Upon arrival at the site, the crew chief and other designated persons will survey the site to determine the location of the facility, its boundaries, and the locations of various operations within the facility. 2. Select sampling points - The crew chief and other designated persons will select appropriate points for the collection of ore samples and airborne particulates. Officials of the host plant will be invited to participate and assist in the survey and selection of sampling points. III. Sampling A. Bulk Material 1. Basis for selection of protocol - No American Society for Test ing Materials (ASTM) method-was found that is directly applicable to this situation. The following related methods are used for guidance. a. American National Standards Institute (ANSI)/ASTM D 75-71 (1978) Standard Methods of Sampling Aggregates. b. ASTM Designation: D 2234-72 Standard Methods for Col lection of a Gross Sample of Coal. c. ANSI/ASTM E 105-58 (1975) Standard Recommended Practice for Probability Sampling of Materials. d. ASTM Designation: C 702-72 Standard Methods for Reduc ing Field Samples of Aggregate to Testing Size. e. ASTM Designation: C 516-75 Standard Specifications for Vermiculite Loose Fill Insulation. f. BS 812 (British Standards Institution) Methods for Sam pling and Testing of Mineral Aggregates, Sands and Fillers. g. Other Considerations - The minimum quantity of any sam ple should be 5 to 10 times the anticipated analytical needs. The analyti cal needs will vary with particle size and range from approximately 40 g for fine material to 1,000 g for 25-mm particle size. Therefore, sample size minimums should range from 400 g to 10 kg. Each sample may consist of a composite of individual sampling increments representing different times and/or locations. Increments will be sampled and stored separately with a composite made under laboratory con ditions by combining representative fractions of each increment. /3 2. Samples to be collected * The objective of sampling is to ob tain samples that are representative of the operations or sites. It is an ticipated that properties of the materials of similar types will vary with time, operation and specific mine site origin. Therefore, to obtain repre sentative samples, it is necessary that composite samples be prepared of a given sample type from individual sample increments, each representing a specific sample time or site. It is likely that a historical sample collec tion is maintained (by the mine company) from several operation points within the facility. If these historical samples are available, it would be helpful to obtain selected increment samples for both the preparation of a time averaged composite and a comparison of present to past conditions. The number of increment samples to be collected must depend oo the variability of the sample and availability of increment sources, with a decision made by an experienced sampler depending upon increment availability and proper sampling procedures. All decisions will be documented with copies sent to the EPA task manager. The following sample types will be collected, except where specific operational conditions dictate variations from these sample categories. Any variations from these sample categories will be documented and copies will be sent to the EPA task manager. a. Raw ore (1) A bulk sample of the raw vermiculite ore before crushing representing different parts of the mine. (2) A bulk sample of the raw vermiculite ore after crushing for beneficiation. (3) A bulk sample of dust from the dust collection equipment where such equipment exists. operations. (4) A water sample from washings and dust control b. Beneficiated ore (1) A bulk sample of each of five grades of benefici ated vermiculite ore. (2) A bulk sample of ore from one to three intermedi ate beneficiation processing steps. ment. (3) A bulk sample of dust from dust collection equip erations . (4) A water sample from washing and dust control op 4 0641692 c. Exfoliated material vermiculite ore. (1) A bulk sample of each of five grades of exfoliated {2) A bulk dust sample from dust collection equipment and other appropriate related material. B. Air Samples 1. General considerations - Airborne particulate samples will be collected at designated points inside and outside the plant boundaries. The sampling will generally follow the EPA method described in Electron Microscope Measurement of Airborne Asbestos Concentrations - A Provisional Methodology Manual. EPA-600/2-77-178, Revised June 1978. This method uses polycarbonate 0.4 pm Nucleopore filters when possible, but allows for the use of cellulose acetate (Millipore) filters. There are advantages and disadvantages to the use of either Nuclespore or Millipore filters. The sample collected on the smooth polycarbonate (Nucleopore) surface has poor retention efficiency and the sample may be lost or redistributed during transport. The cellulose acetate (Millipore) filters requires an ashing procedure and reconstitution on a polycarbonate filter for TEM analysis. Ashing and reconstitution is an extra step in the procedure but has the potential advantage of eliminating interfering organic matter and provides the opportunity to adjust the filter loading for optimum particle analysis. This reduces the need of multiple time sampling to obtain a range of filter loadings. Asbestos contamination has been reported in some lots cf both Nucleopore and Millipore filters. The current consensus of leaders in the field is to favor the use cf Millipore filters for the type of sampling and transport that will be required on this program. The potential presence of asbestos fibers in the filters them** selves will require careful attention to the selection and analysis of filter blanks and field blanks. 2. Control blanks a. Filter blanks - Four filters from each package of 100 filters, one from each box of 25, will be selected by random numbers as filter blanks. The four filters will be quartered and one-fourth of each combined as a composite sample. b. Field blanks - One of every 10 filters will be a field blank, subjected to all processing conducted with an actual air sample ex cept for the sampling itself. 3. Sampling procedure - All sampling, fixed and personal, will b taken using 37 mm 0.45 pm Millipore filters backed with 5 pm Millipore filters and a Millipore support pad. The sampling rate will be approximately /5 06*1693 2 liters/min with the exact rate determined periodically throughout the sampling period by the use of calibrated flow meters. Sampling will be scheduled for 8 hr (or longer for ambient and background samples). However, if the flow is found to reduce during sampling, indicating that the filter is loaded, sampling will be stopped and the time and flow rate recorded. 4. Air samples to be collected - Personal air zone samples and fixed samples located at targeted areas will be taken. The use of personal samples will depend on the individual work patterns and on the cooperation of the host company. When possible, individuals at each work station will be equipped with personal samplers. When possible, air samples will be taken which correspond to the bulk samples (Section III-A-2). The following air samples will be taken. Any variation from these sampling locations will be documented. a. Air samples taken from the raw ore operations. (1) An air sample at the mine during mining. (2) An air sample at the crushing operation. (3) Air samples around the mining facility. (4) Air samples downwind of dusty operations. (5) Air samples along shipping lanes. (6) An air sample as a background control within the geographic region. b. Air samples taken from the ore beneficiation operations. (Where mining and beneficiation are at the same general location, several of the samples may be the same.) (1) An air sample at each of selected work stations in the beneficiation operation. (2) Air samples around the beneficiation plant. (3) Air samples downwind from dusty operations. (4) Air samples along shipping lanes. (5) An air sample as a background control within the geographic region. c. Air samples taken from the exfoliation operation. 6 0641694 (1) An air sample from each of selected work stations in the exfoliation operations. (2) Air samples around the exfoliation plant. (3) Air samples downwind from dusty operations. d. A meteorological station will be installed on-site to collect air speed and direction data throughout the sampling period. IV. Sample Handling A. Bulk Samples The increment samples will be shipped to a central laboratory. Each increment sample will be divided by appropriate procedures (riffle di vided or cone and quarter). Part of each increment will be retained and the remainder combined with other appropriate increments to form a composite sample. The composite will be mixed and split (riffle or cone and quartered) to provide appropriate split analytical samples. The composite samples will be properly designated and submitted for analyses. B. Air Samples 1. Special handling - When sampling-is completed, the filter car tridge will be turned to a position with the filter horizontal and the col lection surface up, the cartridge disconnected from the pump and the inlet and exit holes plugged. This horizontal filter position will be maintained during transport and storage. The cartridge will be placed in a special container for transport to MRI. 2. Each filter will be divided and band carried to different laboratories to provide for replicate analysis. V. Sample Analysis A. Bulk Sample Analysis The analysis protocol for the bulk vermiculite samples will in clude parallel approaches which, to a degree, support one another. However, because of the great differences in the detection limits of the different methods, the justification of some approaches is their simplicity as pre liminary screening procedures rather than their sensitivity. X-ray diffrac tion and optical microscopy of the unfractionated samples are examples of simple procedures with limited sensitivity. These methods may serve to iden tify some samples with gross quantities of asbestiform minerals and eliminate the need for continued analysis. 7 0641695 1. Unexfoliated raw and beneficiated ore a. Examine the sample as received with low power (30X) stereomicroscopy for quantities of visible fibers. b. If no fibers are found, proceed to step f. c. If fibers are visible, determine if they are free and not attached to the veroiculite particles and are in a weigbable quantity. If not, proceed to step e; otherwise continue. d. Hake a density separation, remove (hand pick) the non* asbestos material from the fraction, and weigh the asbestiforo fibers. e. Identify the asbestiform mineral by appropriate means (PLM, TEH, SEM, SAED, XRD, or EDXA). Proceed to step f if additional analy sis is needed. f. Exfoliate the sample by heating for 3 to 6 sec at 80CC and proceed to 2. 2. Exfoliated vermiculite - Steps a and b below are conducted in parallel. Examination of the unfractionated sample will have a low sensi tivity for asbestos but will complement the overall procedure. a. Examination of unfractionated sample (1) Grind the sample using a mortar and pestle. Sieve the sample with a 140 U.S. series equivalent sieve. (2) Regrind the fraction that does not pass the screen. Any material that still does not pass through the 140 sieve is primarily vermiculite and mica and contains a greatly reduced asbestos content. amphiboles (3) Examine the ground material by XRD for total fibers. (4) Examine the ground material by PLM for asbestos b. Examination of fractionated (asbestos-enriched) samples Am important feature of the analytical procedure to achieve high microfiber detection sensitivity is the fractionation of the sample to remove much of the interfering vermiculite, thereby greatly enriching whatever asbestiform fibers that may be present. The basis of fractionation is the floatation en water of the exfoliated vermiculite and the wetting and sinking of the asbestos and other fibers. This assumes that a proportionally high frac tion of the fibers are not physically attached to the vermiculite particles. This is a reasonable assumption but one that will be verified by the examina tion of representative samples of the fraction that floats. 8 0641496 Start with a sample quantity depending on particle size, Particle Size Sample Size 1 mm 5 am 10 mo 25 mm 10 g 50 g 100 g 250 g and proceed as follows: (1) Float separation - Place sample in water; vermiculite will float; released asbestos will sink. Skim off vermiclite and dis card.* (2) Disperse heavy material with ultrasonic treatment. Remove an aliquot during treatment for a preliminary PLM examination and for TEM analysis. Double dilution may be necessary ts obtain proper grid loading. examination.** (3) Examine preparation by PLM as a preliminary parallel (4) Prepare TEH grid by EPA carbca-coated Nucleopore filter technique. (5) Fiber count - Determine chrysatile and total amphiboles. Count 100 fibers of chrysotile or amphiboles or 10 grids of 200mesh screen. Determine limit of detection and count mare grids if necessary. (6) Identify specific amphiboles using SAED or zone axis selected X-ray diffration. 3. Miscellaneous bulk samples a. Dust samples (1) A preliminary examination of the dust sample will be made by optical microscopy including PLM for the identifieation_of gross quantities of asbestiform fibers. If gross quantities of fibers are identi fied, the quantities will be estimated and the analysis terminated. (2) If the sample is not adequately characterized by PLM, a portion of the sample will be dispersed in water and filtered for EM analysis. * Selected samples will be examined to verify the absence of asbestos in this fraction. ** If PLM examination reveals a gross quantity of identifiable asbestiform fibers, the quantity should be estimated and the analysis terminated. 9 0641697 b. Wash water samples - The solids present will be dispersed in the water and aliquots filtered for appropriate optical and EM analysis. B. Air Samples The analysis of air samples will basically follow that specified in the EPA document, "EPA-600/2-77-178, Revised June 1978, Electron Micros scope Measurement of Airborne Asbestos Concentrations - A Provisional Methodology Manual." 10 Butte, Montana 406/723-8075 0641698 Pocatello. Idaho 208/237-2763 SAMPLING POINTS I TRANSFER POINT 1) OPERATOR (4:30) 2) ROCK GATE CONTROL AREA OR OUTSIDE CONTROL ROOM OVER ORE BELTS 3) SIDE OF SCREEN 4) OTHER SIDE OF SCREEN BY THE STAIRS 5) 30TT0M FLOOR ON SWITCH BOX OR CONCRETE WALL 3X/0.S.&E. 1) EO GASTON- BELT KAN (3:30) 2) -S AIR SYSTEM 3) =: AIR SYSTEM (NOTE WHETHER STACKER OR RECLAIMER SIDE) ttvmill curators (all leave at 3:30) r 1) --r It! P-0PERAT6R',r" 2) 30TT0M FLOOR OPERATOR^UTILITY) 3) MIDDLE FLOOR OPERATOR (WO.I> 4) UPPER FLOORS OPERATOR(5-10) NOTE WHETHER HE'S CHANGED SCREENS 5) CONTROL ROOM OPERATOR ) ' TT\ MILL TEST LAB 1) RICHARD FIELDS (LAB ASSAYER) 2) .MINE TESTER 3) KILL TESTER X*MILL FLOORS (1-10) NOTE WHETHER ORE SPILLED ON ANY FLOOR, AIR PURIFIERS RUNNING, AIR PURIFIERS CLEAN, WHETHER OVERHEAD DOORS ARE OPEN. PLACE 2 SAMPLES ON 4th FLOOR: ON FIRE STATION, OR SCREEN AREA OR OUTSIDE LAE AREA H> DRYER 5AGH0USE (1-4) nr X SCREEN:NG PLANT OPERATORS 1) OPERATOR (3:00) 2) SHUTTLE TRUCK DRIVER (3:00) y-- 3) HAUL TRUCK- KENWORTH (2:00) 4) TEST LAB- JOHN CRAVER (4:30) 5) PAY CARLSON- LOADOUT Butte, Montana 406/723-8075 "IT/SCREENING PLANT 1) FLOORS (1-6) 2) BELT HOUSE 3) CONTROL ROOM TT EXPORT 1) KENNY FREDRICKS- DRIVES FOR EXPORT OR SCREENING PLANT (V%0) 2) TALE VINSON- FOREMAN 3) BAG STACKER 4) BAGGINE ROOM 2* MINE PERSONNEL 1) :5.ILLS (DO BOTH IF RUNNING) 2) ::zer 3) LOADER 4) GFADER 5) '*:SILE TEST LAB (4:30) 6) FIT HAUL TRUCKS 0641699 Pocatello. Idaho 208/237-2763 /  -C / ># r> /h^tt/JTT -`-'T yrz i -- 0641700 / crscfret/ oV^ifs/rs As -s y l) FO n't nd }cJ<?r O/o r filter * / d) Siw HU d^t \/i r r z y) ~L/p>e ra i~Orm----- z? / triks/z*:- jfjL *-*.y /'i OfiCrA' n<2 v/-j, t) i hst* / 7dr'it*`*' t / 0 $ Un/fHj + -si o ----- bidy--i&tA+r. -tf-2 frtt'L fffrr'tci* / ) /v> S. 'fa.r 2f? COr n ,f I /3o dOfr !cor '"it / - Sf*+;Z 4 r-y St?*]}f/ors j' Ac r*-* *11 ^ n 6'A l Ok C p *'"' i k/ 'l // c/oMU Ut) hJ 2* J " z<}--iVj;;.,J C?m;ri*tr7/sv--f<n' (~t,J.-U/J -- * '' ^ 'ft,* ituclt /.>Z. sZk. i/^v'> z' " S n J&/ dcf) /m<Q* .OO WO -r-*V'' "" 2- Oh rO* */:ctcSSi Cl io *4 tOo/ ( c r*sz * 2- * Car/wy Cr*<L roc,i Sw a-f M'<) (<*pt~i.r\d*) ~x" O* Ui- hJ dtr<cii oh ^ o y* **"> 2. 0641701 Ci L to/. c(l, /< / ) tiSif*k*---^) v/ if MB \ 'jT ' ".'. ) (f) (Lotrfro/ 4(Jr t~) t'n d cr~f$ ^ropt.-^y z_ l) $cr*ss r'tVt*- Jicfjx/' c-rr /o**//*t^j Z'Me %- cfie*~/ Ull ?J K i i, t----- ~~*f ^~'i!-W ' J-ti. Ti'/T Mf*.- } 2~ Me 'ic.cr c /sj; c ( S'tc? /1 tu-S 3 3c. r# .* fl', o |,v r * *f \ o * <irr/// *< /* fr b*t*` J -- <'/Pc~ff g. / 7 <r jj/. ,v ^^VAC*_ <- ' Cmc'^. , ^ C 1, A*VW~w<u^ . .At- &. ^uA.^u^fjur (jia^no# ) ?. jV*vvt C . ts^Xjii'L. icv^i i'^ 'K--( n> p. (. ^VA^I<AU^ A. ~ft'- JLIn | tfltd JULc-k <yi*-wCG^ C. i Jt-Mfi c'jU^LtTz- Z. ^-uct, ^ ^6cX_ *'^ >6^C'*.aX1<'-*V r> p. Xe* / /Vt^- Q. utu.-v+~<fl C. tv I .., i'v<^ </. Cc-CCu,\ 0641702