Determination of trace elements in hair for monitoring environmental and occupational exposure

Hair has a number of advantages over more conventional sample types, such as serum, whole blood and urine, as an indicator of the intake and exposure to trace elements. While the analysis of these conventional sample types can only be used as a measure of recent exposure, hair, in contrast, can provide an historical record of exposure episodes. Furthermore the levels of trace elements in hair are higher than in blood and as a result small changes in intake are magnified. In addition, the collection of hair samples is non-invasive and the samples are easy to store. However, the difficulty in differentiating surface bound contamination from trace elements incorporated into the hair matrix complicates the interpretation of the hair analysis results. A review of the literature showed that a variety of washing procedures to remove external contamination have been proposed but as yet no standardised procedures are available. In this study, methods for the pretreatment and determination of antimony, arsenic, cadmium, chromium, lead, mercury and selenium in human hair by ICP-MS were developed. Chemical speciation of mercury and methyl mercury in hair was also achieved using LC-ICP-MS. For the digestion of hair samples the optimum method for these elements was found to be a digestion mixture of nitric acid and hydrogen peroxide. Certified reference materials CRM 397 and CRM 07601 were digested using this method and the values obtained were seen to be in good agreement with the certified values. Investigations of various washing procedures to remove external contaminants showed that in unexposed hair samples cadmium, lead and mercury were significantly removed from hair using a 0.1M HCl wash, with 87%, 73% and 5% respectively being washed off. Whilst the removal of antimony, arsenic and chromium from unexposed hair was more efficient with 1% v/v sodium lauryl sulphate, with 43%, 40% and 13% of each element respectively being washed off. Selenium could not be removed from the hair by any of the washing methods studied. Experiments with simulated sweat spiked with each of the elements showed that exogenously bound chromium, cadmium and lead could be removed after washing with 0.1M HC1. In contrast, antimony, arsenic, selenium and mercury were irreversibly bound and could not be removed with any of the washing solutions investigated. Using a LC-ICP-MS system it was possible to separate inorganic mercury and methyl mercury in hair, without any modifications to the existing instrumentation. The results showed that in order to determine methyl mercury the sample had to be cold digested in 2: 1 HNO3: H202 and that a minimum of 0.1g of hair is required. Using this method of analysis it was seen that with spiked simulated sweat solutions exogenously bound methyl mercury could be removed from hair after washing with 0.1M HCI, whereas the inorganic mercury was irreversibly bound to the hair. It was also found that inorganic mercury was the major mercury species in hair from dental subjects. The results of a six week selenium supplementation study of 29 subjects showed an increase in hair selenium levels of 19%, confirming that hair concentrations of selenium can give a good reflection of selenium status in the body. A control group of 40 unexposed samples were analysed to determine normal levels of antimony, arsenic, cadmium, chromium, lead, mercury and selenium in human hair. Occupational studies were then carried out and the data obtained from these studies showed a significant increase in the elements of exposure. Hair samples from dental workers (n = 43) showed a mean mercury level of 813 ng g-1 compared to a control value of 287 ng g-1; hair samples from semiconductor workers (n = 24) showed a mean arsenic level of 208 ng g'1 compared to a control value of 34 ng g'1; hair samples from chromium platers (n = 7? showed a mean chromium level of 2201 ng g-1 compared to a control value of 985 ng g-1 ; and hair samples from lead foundry workers (n = 13) showed a mean lead level of 124337 ng g-1 compared to a control value of 1030 ng g-1. The lead levels in the foundry workers' hair samples were reduced by 94% after washing with 0. IM HCl from 124 000 to 7440 ng g-1, which was still above the normal group mean hair lead level of 1030 ng g-1.