Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, EPHB 326-15a, 2008 Oct; :1-63
Researchers from the National Institute for Occupational Safety and Health (NIOSH) conducted a study of the recycling of electronic components at the Federal Prison Industries, Inc. (FPI) facilities (aka, UNICOR) in Marianna, Florida, in August, 2007 to assess worker exposures to metals and other occupational hazards, including heat, associated with these operations. The electronics recycling operations at Marianna can be organized into four production processes: a) receiving and sorting, b) disassembly, c) glass breaking operations, and d) packaging and shipping. A fifth operation, cleaning and maintenance, was also addressed but is not considered a production process per se. It is known that lead, cadmium, and other metals are used in the manufacturing of electronic components and pose a risk to workers involved in recycling of electronic components if the processes are not adequately controlled or the workers are not properly trained and provided appropriate personal protective clothing and equipment. Methods used to assess worker exposures to metals during this evaluation included: personal breathing zone sampling for airborne metals and particulate; surface wipe sampling to assess surface contamination; and bulk material samples to determine the composition of settled dust. Samples were analyzed for 31 metals with five selected elements (barium, beryllium, cadmium, lead and nickel) given emphasis. Heat exposures were determined using wet bulb globe temperature monitors. The results of air sampling conducted during the August visit indicated no overexposures of workers to metals above the most stringent occupational exposure limits during the routine and non-routine operations evaluated during that site visit. The highest exposures to metals (as determined by both arithmetic and geometric means) occurred to workers in the Federal Prison Camp (FPC) glass breaking operation while changing filters, while workers in the Federal Prison Camp (FPC) UNICOR factory had the highest exposure to airborne particulate during routine production operations. The results of two of those samples were affected by unanticipated events. In one instance, a worker touched the inlet of the cassette with her glove and some lint was sucked onto the filter. In the other, a worker unloading a truck reported that toner spilled onto her from surplus equipment she was unloading. When those two samples (which did not exceed allowable limits) are not considered, the particulate concentrations are well below levels of concern. When those two samples are not included in the analyses, the FPC glass breakers had the highest particulate exposures. These occurred during the filter change operation. Exposures to airborne metals during the filter change-out maintenance operation were higher than exposures during other operations in the FPC but were below the most stringent occupational exposure limits. Total airborne particulate levels were higher during this operation than elsewhere, when the two samples described above are disregarded. Total particulate concentrations during routine glass breaking operations ranged from <71 µg/m3 (140 minute sample) for a breaker to 891 µg/m3 (147 minute sample) for a feeder. During the filter change operation, they ranged from 4,912 µg/m3 (57 minute sample) for a worker working inside the glass breaking booth to 274 µg/m3 (45 minute sample) for a worker outside the booth. All airborne particulate measurements representing potential exposures during routine and non-routine operations were, however, below applicable occupational exposure limits (e.g., the OSHA PEL of 15 mg/m3 (15000 µg/m3), 8-hr TWA for total particulate). Although beryllium is used in consumer electronics and computer components, such as disk drive arms (beryllium-aluminum), electrical contacts, switches, and connector plugs (copper-beryllium) and printed wiring boards [Willis and Florig 2002, Schmidt 2002], beryllium in this study was not detected at levels above the detection limit of the analytical method. Most of the recycling activities at this facility resemble typical maintenance activities on consumer products (e.g., personal computers), such as opening cases and removing components. Willis and Florig  noted that most beryllium "in consumer products is used in ways that are not likely to create beryllium exposures during use and maintenance." This may account for the results seen at this facility. Other e-recycling activities that include further processing, such as shredding of the components, may produce higher exposures to beryllium but shredding does not occur at this facility. Samples collected during routine daily glass breaking operations showed that the highest exposure was less than 10% of the OSHA PEL for lead of 50 µg/m3 8hr TWA (4.5 µg/m3 8hr TWA for a 109 minute sample). The highest lead exposure measured during the filter change operation was 12.5 µg/m3 8hr TWA for a 57 minute sample. The highest cadmium result during routine glass breaking was 2.0 µg/m3 8hr TWA for a 143 minute sample, less than half the OSHA PEL of 5 µg/m3 8hr TWA. During the filter change operation, the highest cadmium concentration was 1.4 µg/m3 8hr TWA for a 57 minute sample. Samples collected on disassembly workers in the FCI factory area and on workers in the FPC factory area were well below levels of concern for cadmium, lead and nickel. Unless specified, the results of the samples presented are for the duration of sample and not calculated on an 8 hour time-weighted average basis. Lead, cadmium and other heavy metals were detected in the surface wipe and bulk dust samples. There are few established standards available for wipe samples with which to compare these data. Some of the surfaces tested for lead indicated levels exceeding the most stringent criteria. The wipe sample results can not be used to determine when the contamination occurred. They only represent the surface contamination present at the time the sample was collected. Environmental heat monitoring and estimates of work rate indicated that some workers in this facility were exposed to heat stress (e.g., above the ACGIH TLV or at risk of heat stress (e.g., exceeding the ACGIH Action Limit) during this survey period. The locations where heat stress was noted included the glass breaking operation (breakers, feeders, and outside workers) and the warehouse (truck crew), while a risk of heat stress was noted in the warehouse (other workers), FCI-disassembly and FCI-Refurbish. Recommendations resulting from this study include: 1. The implementation of a site specific health and safety program at Marianna that includes a heat stress program; 2. The respiratory protection program for this facility should be evaluated to ensure that it complies with OSHA regulations; 3. Attention should be focused on practices to prevent accidental ingestion of lead and other metals, such as housekeeping to reduce surface contamination and hand washing to prevent hand-to-mouth transfer of contaminants; 4. Management should evaluate the feasibility of providing and laundering work clothing for all workers in the recycling facility; 5. Change rooms should be equipped with separate storage facilities for work clothing and for street clothes to prevent cross-contamination; and 6. All UNICOR operations should be evaluated from the perspective of health, safety and the environment in the near future. A comprehensive program is needed within the Bureau of Prisons to assure both staff and inmates a safe and healthy workplace.
National Institute for Occupational Safety and Health, Division of Applied Research and Technology, Mail Stop R-5, 4676 Columbia Parkway, Cincinnati, OH 45226