Evaluation of a standardized method for determining soluble silver in workplace air samples

Abstract

Several occupational exposure limits and guidelines exist for silver, but the values for each depend on the chemical form of the silver compound in question. In the past, it generally was not possible, without prior knowledge of the work process, to distinguish soluble silver from insoluble silver compounds collected in workplace air samples. Therefore, analytical results were historically reported as total silver. In this study, work was conducted to evaluate a method to differentiate between the quantities of water-soluble silver compounds and total silver collected on filters. The investigation entailed an evaluation of an International Organization for Standardization method to determine soluble silver in airborne particulate matter. The study design incorporated laboratory experiments to evaluate analytical figures of merit, such as selection of appropriate filter media and extraction solution, analytical recovery, and sample stability during storage. Polytetrafluoroethylene (PTFE) filters (2 μm, 37 mm) in opaque cassettes were either spiked with known amounts of silver nitrate or contained a known mass of solid silver nitrate. Results showed that over 90% of the silver was recovered from PTFE filters. Also, field studies were conducted in which workplace air samples were collected in two silver refineries. Some of these samples were analyzed only for soluble silver while others were sequentially extracted and analyzed, first, for soluble silver, then for total silver. The mass fractions of soluble silver, as compared to total silver, were approximately 2% or less. This investigation served to validate an international standard procedure for the determination of soluble silver in workplace air samples.

---

Introduction

In the United States, the Occupational Safety and Health Act of 1970¹ established a permissible exposure limit (PEL) of 0.01 mg m⁻³ for all forms of airborne silver. In 1977, when the Mine Safety and Health Administration (MSHA) was formed as a result of the Federal Mine Safety and Health Act,² MSHA also adopted an exposure limit for silver of 0.01 mg m⁻³. In 1981, the American Conference of Governmental Industrial Hygienists (ACGIH) separated the Threshold Limit Values® (TLVs) for silver into soluble (0.01 mg m⁻³) and metallic (0.1 mg m⁻³) forms.³ ACGIH based its separate TLVs for silver on data that demonstrated that “silver salts have a greater propensity to cause argyria than does the dust or fume of metallic silver.”³ In 1994, the European Commission, assisted by the scientific expert group on occupational exposure limits, recommended an 8 hour time-weighted average (TWA) exposure limit of 0.1 mg m⁻³ for metallic silver⁴ while retaining the lower exposure limit of 0.01 mg m⁻³ for soluble silver salts. The Health and Safety Executive of the United Kingdom established the same recommendation in 1996.⁵ Examples of other countries that have separate occupational exposure limits for soluble and insoluble silver compounds include Australia, Belgium, Finland, France, and Sweden.⁶ A current review of the literature⁷ found widespread support for establishing separate occupational exposure limits for soluble and insoluble silver compounds.

Although the limit values for soluble silver compounds have been in existence for over two decades, a method for the determination of soluble silver in workplace air was unavailable until 2001 when the International Organization for Standardization (ISO) published a relevant international standard.⁸ A portion of this standard describes the extraction of soluble metals and metalloids for subsequent analysis by inductively coupled plasma–atomic emission spectrometry (ICP-AES).⁹ The ISO method describes the use of deionized water for extraction of soluble metal compounds from unreactive filter materials used to collect airborne particulate samples. Despite the existence of the ISO procedure, there remains a dearth of performance data regarding the measurement of soluble silver in workplace air samples.¹⁰ In an effort to fill this void, this study evaluated and validated the ISO method for the determination of water-soluble silver compounds in workplace air filter samples.

Experimental

Solution preparation

Preliminary experiments showed that the presence of acid in the spiking solution (i.e., commercially available silver standards) enhanced recoveries and thus introduced a positive bias; therefore, recovery experiments were conducted using a silver nitrate solution, with no acid added, for spiking onto filters. A 1000 μg mL⁻¹ solution of silver was prepared by dissolving 0.15764 g of Premion, 99.995% silver nitrate (Alfa Aesar, Johnson Matthey Company, Ward Hill, MA, USA) in 100 mL of deionized water (18 MΩ, Millipore Milli-Q Plus, Molsheim, France). This solution was then diluted to prepare a 100 μg mL⁻¹ silver stock solution. Both solutions were analyzed by ICP-AES (Applied Research Laboratories 34 000, Sunland, CA, USA) and inductively coupled plasma mass spectrometry (ICP-MS) (Varian UltraMass, St. Helens, Australia) to verify the silver concentrations. Measured silver concentrations of the 100 and 1000 μg mL⁻¹ solutions, respectively, were 98.2 μg mL⁻¹ and 981.7 μg mL⁻¹ by ICP-AES and 100.7 μg mL⁻¹ and 1002 μg mL⁻¹ by ICP-MS. The solutions were stored in the dark when not in use to minimize photoreduction of silver ions.

All standard and calibration check solutions were prepared from a 1000 μg mL⁻¹ silver standard (Inorganic Ventures Plasma Standard solutions in 3.5% nitric acid, Lakewood, NJ, USA). The ICP-AES standard solutions were prepared at 0 and 10 μg mL⁻¹ in 0.1% nitric acid (OPTIMA, Seastar Chemicals, Inc., Pittsburgh, PA, USA), and a calibration check solution was independently prepared at 5 μg mL⁻¹ in 0.1% nitric acid. A solution of 10 μg mL⁻¹ of silver in 0.1% nitric acid was used to prepare all ICP-MS standard solutions and calibration check solutions. The ICP-MS solutions were prepared daily at 50 and 100 ng mL⁻¹ in 0.5% nitric acid in 50 mL centrifuge tubes (Falcon, Becton-Dickinson, Franklin Lakes, NJ, USA) with an indium internal standard concentration of 100 ng mL⁻¹. The stock indium solutions of 15 and 100 μg mL⁻¹ were prepared from a 1000 μg mL⁻¹ indium standard solution (Inorganic Ventures, Inc., Custom Grade Standard in 1.4% nitric acid, Lakewood, NJ, USA). Calibration check solutions were prepared independently from standard solutions at concentrations of 10 and 75 ng mL⁻¹ of silver.

Method performance samples

The prescribed method for the determination of metals in workplace air samples¹¹ calls for the use of mixed cellulose ester (MCE) filters; consequently, the method for soluble silver was first attempted using MCE filters. The filters (37 mm, 0.8 μm, SKC, Inc., Eighty Four, PA, USA) were placed into 50 mL beakers, spiked with a known volume of silver nitrate solution, and extracted with 5 mL of deionized water for 60 min in a 37 °C shaker bath in accordance with the ISO procedure.⁸ The average recovery for silver from these filters was consistently less than 60%, and the results were imprecise. These same filters were then digested in nitric acid on a hotplate to extract any remaining silver. Total recoveries after nitric acid digestion ranged from 90% to 100%, indicating that silver was retained on the MCE filters after treatment with the soluble extraction procedure. The water extraction experiment was repeated with significantly reduced levels of laboratory lighting to minimize photoreduction of silver ions. Recovery of silver from the MCE filters improved (70% to 85%), but was still low. Because both the filter material and the amount of light were causing poor and erratic recoveries, polytetrafluoroethylene (PTFE) filters in opaque sampling cassettes were used thereafter for experiments in which soluble silver compounds were extracted.

Opaque plastic sampling cassettes pre-loaded with PTFE filters (37 mm, 2 μm pore size, with a spacer ring and no backup pad; SKC, Inc., Eighty Four, PA, USA) were either spiked with an aliquot of a known concentration of silver solution or had a known mass of solid silver nitrate added. For the liquid spikes, an aliquot of 10 to 50 μL of either 100 or 1000 μg mL⁻¹ silver solution was added directly into the cassette using a micropipette (Rainin EDP-Plus, Woburn, MA, USA). For the cassettes containing solid silver nitrate, approximately 0.025 g of pulverized material was added to each cassette through a small funnel.

Air sample collection

Workplace area and personal air samples were collected in two refineries where silver was a major constituent of the material being processed. Personal air samples were collected on PTFE filters in opaque plastic sampling cassettes. The samples were collected at a flow rate of 2.0 (±0.1) L min⁻¹ with personal sampling pumps (MSA Escort Elf, Pittsburgh, PA, USA). Sample collection times varied from 158 min for task-based samples to 534 min for individual samples. Area air samples were collected using a custom-made eight-port sampler. The multi-port sampler was connected to a high-volume (15 L min⁻¹) air sampling pump (Gilian AirCon2, West Caldwell, NJ, USA). The measured flow rate for each individual port was 1.8 (±0.1) L min⁻¹. All eight samples were simultaneously collected. This multi-port sampler has been shown in previous studies to give uniform loadings among the individual samplers, as evidenced by relative standard deviations (RSDs) of 4% or less for gravimetric determinations.¹²

Analysis

High-concentration silver measurements (0.1 to 10 mg L⁻¹) were made at 328 nm using a simultaneous ICP-AES side-view instrument, with a pneumatic nebulizer, operated in single-element mode. The ICP-AES method detection limit (MDL) for silver is 10 μg L⁻¹ for the conditions and instrument used for this study. Because lower levels of detection are needed for measuring soluble silver, an ICP-MS was used for the low-level analysis of silver (5 to 100 μg L⁻¹). Silver measurements by ICP-MS were made at Ag-107 and Ag-109, with indium (In-115) as an internal standard, using a pneumatic nebulizer. For silver, the MDL by ICP-MS in this study was 0.02 μg L⁻¹.¹³

Standard solutions were prepared daily in 0.1% nitric acid at 50 and 100 ng mL⁻¹ of silver. For all measurements, the calibration check solutions were within ±5% of the standard value. For extraction of soluble silver, 5 mL of deionized water were added to each sampling cassette using a mechanical pipettor (Finnpipette, Thermo Labsystems, Helsinki, Finland). The inlet plug was securely replaced, and the cassettes were placed in a 37 °C reciprocating water bath (Precision Reciprocal Shaking Bath, Winchester, VA, USA) for 60 min at 60 rpm. After the cassettes were removed from the shaker bath and blotted dry, the extracting fluid was emptied into plastic 30 mL sample cups (VWR International, West Chester, PA, USA). First the inlet plug was removed, and the cassette was inverted over the sample cup, then the outlet plug was removed. The solution was withdrawn from the sample cup with a 10 mL syringe (Henke Sass Wolf GMBH, Norm-Ject, Tuttlingen, Germany) and filtered through a 0.45 μm, 25 mm PTFE syringe filter (Nalgene Syringe Filters with PTFE Membranes, Rochester, NY, USA) into another clean sample cup. Aliquots of 0.5 to 4 mL were transferred to 15 mL centrifuge tubes (Falcon, Becton-Dickinson, Franklin Lakes, NJ, USA) containing 50 μL of concentrated nitric acid and 100 μL of 15 μg mL⁻¹ indium and diluted to 15.0 mL with deionized water. After mixing, the samples were analyzed by ICP-MS.

For the samples analyzed for total silver, the cassettes were opened, and the PTFE filters were transferred to 50 mL beakers (VWR International, West Chester, PA, USA). Three milliliters of concentrated nitric acid were added to each beaker. The beaker was covered with a watch glass (VWR International, West Chester, PA, USA) and heated on a hotplate to 140 °C. The samples were heated until approximately 0.5 mL of liquid remained. The beakers were cooled, and an additional 2 mL of concentrated nitric acid were added and heated. When the volume of the liquid was reduced to approximately 0.5 mL, the beakers were cooled, and 3 mL of deionized water was added to each beaker. The liquid was quantitatively transferred to 15 mL centrifuge tubes, diluted to 10.0 mL with deionized water, and analyzed by ICP-AES.

Some of the samples were sequentially extracted to determine the amounts of both soluble and total silver in the same sample. Following the extraction for soluble silver, as previously described, the cassettes were opened, and the PTFE filters were transferred to 50 mL beakers. The PTFE filters from the syringe filters were removed using a mechanical press and transferred to the 50 mL beakers. Total silver was then determined as previously described.

Results and discussion

Extraction and measurement of soluble silver

NIOSH guidelines for the evaluation of workplace air sampling and analysis methods¹⁴ recommend a suite of experiments be conducted in order to validate a potential method under investigation. For a method to meet the NIOSH accuracy criteria,¹⁴ the recovery of the analyte from the filter media should be greater than or equal to 75% of the true concentration at least 95% of the time. To determine if the recovery of soluble silver from the sample media was quantitative and precise, replicates of each loading level of interest were prepared at 0.1, 0.5, 1.0, and 2.0 times the TLV for soluble silver, based on an 8-hour TWA, and sampling at a flow rate of 2 L min⁻¹. The PTFE filters in opaque cassettes were spiked with unacidified silver nitrate solutions to yield the desired concentrations of silver (i.e., 10 μL of 100 μg mL⁻¹, 50 μL of 100 μg mL⁻¹, 10 μL of 1000 μg mL⁻¹, and 20 μL of 1000 μg mL⁻¹, giving 1.0, 5.0, 10.0, and 20.0 μg Ag filter⁻¹, respectively).

The spiked filters were prepared and analyzed as previously described for aqueous extraction and ICP-MS measurement of soluble silver (Table 1). The average recovery for all concentration levels was 99.6% (range 98.0%–101%) with the RSD less than or equal to 2.25%. The bias (B), as defined by NIOSH,¹⁴ was calculated as B = [(μ/T) − 1], where μ is the measured mean and T is the true concentration of the spike. Acceptable methods must have an absolute bias less than 10%. The bias values for these experimental results were within ±2.0%, indicating homogeneity over the range of concentrations.

Table 1 Analysis of silver nitrate spikes on PTFE filters in opaque cassettes at 0.1, 0.5, 1.0, and 2.0 times the TLV of 10 μg m⁻³ for soluble silver^a

Spike concentration/μg filter^−1	Measured mean/μg filter^−1 (±SD)	%Recovery	%RSD	%Bias
a n = 6 for all concentration levels. SD = standard deviation. RSD = relative standard deviation.
1.0	1.01 (±0.023)	101	2.25	0.91
5.0	4.96 (±0.084)	99.2	1.69	−0.83
10.1	10.1 (±0.144)	100	1.43	0.15
20.2	19.8 (±0.354)	98.0	1.79	−1.97

---

To evaluate the ability to recover soluble silver from the filter media over time, spiked samples were generated, stored under ambient laboratory conditions, and analyzed at specified time periods.¹⁴ A set of 42 filters was spiked with 50 μL of 100 μg mL⁻¹ silver solution (equivalent to 5 μg filter⁻¹). These samples were divided into six groups: one group of 12 samples and five groups of six samples each. The group of 12 samples was analyzed immediately following preparation (day 0). The remaining five groups of six samples each were analyzed after 7, 10, 14, 21, and 30 days. At each day of analysis, a set of freshly prepared spike samples (n = 6) were analyzed along with the stored samples to check analytical variability. The NIOSH acceptance criterion for the sample storage study states the set of samples analyzed on day 7 should not differ by more than 10% from the set analyzed on day 0. Samples should be stable for a minimum of 7 days under ambient conditions to allow samples to be shipped to a laboratory for analysis. Mean recoveries of silver from the filters were unchanged until day 21, when a 5% reduction was observed (Table 2). Even after 30 days of storage, only an 8.8% reduction was measured. Although this indicated the effectiveness of sample storage in opaque sampling cassettes, samples should be analyzed within 21 days of collection to maximize analytical accuracy. The differences between freshly prepared and stored samples were within ±1% of the result obtained on day 0, indicating that sample preparation and instrumental analysis did not contribute to the variability seen in the stored samples.

Table 2 Stability of silver nitrate spikes on PTFE filters in opaque cassettes^a

Day	n	Mean/μg filter^−1 (±SD)	%Mean recovery	%RSD	%Bias
a SD = standard deviation. RSD = relative standard deviation.
0	12	4.90 (±0.076)	97.3	1.56	−2.7
7	6	4.88 (±0.035)	96.9	0.718	−3.1
7 (fresh)	6	4.83 (±0.033)	96.0	0.684	−4.0
10	6	4.93 (±0.059)	97.8	1.21	−2.2
10 (fresh)	6	4.86 (±0.028)	96.6	0.569	−3.4
14	6	4.85 (±0.068)	96.3	1.40	−3.7
14 (fresh)	6	4.90 (±0.032)	97.4	0.646	−2.6
21	6	4.66 (±0.074)	92.5	1.59	−7.5
21 (fresh)	6	4.91 (±0.023)	97.6	0.478	−2.4
30	6	4.45 (±0.048)	88.5	1.08	−11.5
30 (fresh)	6	4.89 (±0.026)	97.1	0.529	−2.9

---

Measurement by ICP-AES and ICP-MS

Solid silver nitrate was weighed directly into opaque cassettes containing PTFE filters (n = 10; range 0.0214 to 0.0279 g). After the samples were extracted using the soluble extraction procedure, aliquots of 150 μL were transferred to 50 mL centrifuge tubes containing 250 μL of nitric acid and diluted to 50.0 mL with deionized water (first dilution). These samples were analyzed by ICP-AES. The mean recovery was 101% ± 0.65%. A second dilution was prepared by transferring 250 μL of the first dilution to 50 mL centrifuge tubes containing 250 μL of nitric acid and internal standard and diluted to 50.0 mL with deionized water. These samples were analyzed by ICP-MS. The mean recovery was 102% ± 1.03%. The 1% difference in the mean indicates that the ISO extraction method can be used to accurately and precisely measure solid silver nitrate on PTFE filters by both instrumental techniques.

Preparation and analysis of generated samples

A set of eight air filter samples was collected during the heating of silver solder inside an enclosed chamber and was extracted according to the method for soluble silver as previously described. When vaporized metal comes in contact with oxygen in the air, metal oxides are produced.¹⁵ Thus, the fumes generated while soldering with silver brazing alloy were expected to produce a measurable amount of silver oxide in the fume (Table 3). Under these experimental conditions, the silver solder fumes were uniformly deposited between the eight filters as demonstrated by the good precision (RSD = 5.3%) among the filters.

Table 3 Results from air samples collected in a multi-port sampler during silver soldering^a

Port #	Soluble Ag/μg filter^−1
a SD = standard deviation. RSD = relative standard deviation.
1	11.8
2	11.8
3	12.0
4	12.2
5	11.4
6	11.6
7	11.2
8	13.3
Avg. (±SD)	11.9 (±0.63)
%RSD	5.3

---

Samples were collected at two silver refineries to evaluate the method performance on complex samples and to obtain information about the ratio of soluble silver compounds to total silver present on air filter samples. During the mining process, silver ore is extracted, crushed, leached, concentrated, and refined. The refining process entails mixing the silver concentrates with flux material, melting the mixture in a furnace, and pouring the molten metal into bars or buttons. To collect a range of loading on the filters, the eight-port sampler was positioned in various locations throughout both refineries. The sampler was situated in areas where the concentrates were being mixed with flux material (Table 4). These samples were collected for 196 min. Another set of samples was collected outside, directly beside the ventilation dust-collection system (Table 5). When the refinery workers were in this area, a plume of very fine material was observed. These samples were collected for 275 min.

Table 4 Results from workplace air samples collected while concentrates were being mixed with flux material. Samples were analyzed for soluble and/or total silver

Sample number	Soluble Ag/μg m^−3	Total Ag/μg m^−3	%Soluble Ag
C 1	0.144	28.8	0.5
C 2	0.133	31.6	0.42
C 3	0.419	33.5	1.25
C 4	0.135	29.6	0.46
C 5	—	45.4	—
C 6	—	49.1	—
C 7	—	54.6	—
C 8	—	60.4	—
%Avg. soluble Ag			0.66

---

Table 5 Results from workplace air samples collected near the ventilation dust-collection system. Samples were analyzed for soluble and/or total silver

Sample number	Soluble Ag/μg m^−3	Total Ag/μg m^−3	%Soluble Ag
D 1	0.039	179	0.02
D 2	0.127	194	0.07
D 3	0.058	267	0.02
D 4	0.038	291	0.01
D 5	—	402	—
D 6	—	631	—
D 7	—	186	—
D 8	—	381	—
%Avg. soluble Ag			0.03

---

For these refinery samples, it was desired to determine the fraction of soluble silver relative to total silver. To do this, four of the eight samples were analyzed for soluble silver and then sequentially extracted with nitric acid for analysis of total silver. The remaining four samples were analyzed for total silver only, as previously described. The sequential extraction technique used was semi-quantitative. The mechanical press used to punch out the syringe filter caused the loss of some of the particulate matter contained in the syringe filter. However, it is important to note that the objective was not to perfect the sequential extraction technique, but rather to determine the mass fraction of soluble silver compared to total silver. The results from the sequential extractions presented in Tables 4, 5, and 6 underestimate the amount of total silver in these samples and, therefore, overestimate the percentage of soluble silver.

Table 6 Results from personal air samples collected at silver refineries. Some samples were analyzed for both soluble and total silver

Sample number	Soluble Ag/μg m^−3	Total Ag/μg m^−3	%Soluble Ag
1	0.05	167	0.03
2	0.114	49.6	0.23
3	0.097	37.5	0.26
4	0.293	53.7	0.55
5	0.119	46.3	0.26
6	0.053	15.0	0.35
7	1.79	273	0.66
8	1.41
9	0.26
10	2.22
11	0.789

---

In the set of samples collected outside by the ventilation system (Table 5), the average total silver concentration for all eight filters was quite high at about 315 μg m⁻³; however, only about 0.03% of the total silver was present as soluble silver. In the set of samples collected during concentrate mixing (Table 4), about 0.7% of the total concentration of silver on the filters was present as soluble silver. Given that the sequential extraction technique is semi-quantitative, the results still show that in both locations the fractions of soluble silver in the generated aerosols were very small.

Presumably, the highest concentrations of soluble silver would be found in the fumes emitted from the furnace, because the flux materials (sodium nitrate, borax, calcium fluoride, and soda ash), combined with the high temperatures and oxygen, could enhance the oxidation of silver metal during the melting process.¹⁵ To sample aerosols generated from the furnace, the eight-port sampler was set up in two locations around the furnace. It was first positioned adjacent to the furnace in proximity to the furnace operators’ breathing zone. Samples were collected for 515 min (Table 7). The average soluble silver concentration was 0.057 μg m⁻³. Then, the eight-port sampler was positioned slightly above the furnace, directly in the path of the fumes being captured by the ventilation system. These samples were collected for 531 min. The average soluble silver concentration was 4.33 μg m⁻³. The lack of precision in the samples collected adjacent to the furnace (%RSD = 57.4) may be due to the wide range of particle sizes in the refinery air. However, if sample 2-3 (a statistical outlier) is excluded from the analysis, precision is greatly improved (%RSD = 12.4). Better precision (%RSD = 4.17) was found among the samples collected above the furnace. This finding was probably an effect of the particle sizes of the fumes being much smaller.¹⁶ Even in this ‘worst case scenario’, the measured soluble silver concentrations were less than half the TLV of 10 μg m⁻³.

Table 7 Results from workplace air samples collected adjacent to and directly above the furnace to capture a range of loading on the filters. The samples were analyzed for soluble silver only^a

Port number	Soluble Ag/μg m^−3
	Adjacent to furnace	Above furnace
a SD = standard deviation. RSD = relative standard deviation. b Outlier. c Results excluding outlier.
1–3	0.043	4.30
2–3	0.136^b	4.02
3–3	0.045	4.55
4–3	0.040	4.58
5–3	0.038	4.36
6–3	0.044	4.31
7–3	0.053	4.18
8–3	0.053	4.30
Avg. (±SD)	0.057 (±0.032)	4.33 (±0.180)
%RSD	57.4	4.17
Avg. (±SD)	0.045 (±0.006)^c
%RSD	12.7^c

---

Personal samples were also collected at both refineries from workers who performed a wide variety of tasks throughout the day. Seven of the 11 samples were sequentially extracted for soluble silver compounds, and then the same filters were digested and analyzed for total silver. The results in Table 6 show a wide range in measured silver concentrations. All seven samples analyzed for total silver exceeded the current OSHA PEL of 10 μg m⁻³, but only two of the seven samples exceeded the ACGIH TLV of 100 μg m⁻³ for total silver. All 11 samples were less than the ACGIH TLV of 10 μg m⁻³ for soluble silver. Even the worker with the highest measured total silver exposure (273 μg m⁻³) only had a soluble silver concentration of 1.79 μg m⁻³. The average mass fraction of soluble silver to total silver in these samples was 0.33%, indicating that in these silver refineries, exposure to the more toxic soluble silver compounds was quite low.

In conclusion, this investigation validated an international standard procedure for the determination of soluble silver in workplace air samples. By using opaque samplers housing PTFE filters, it was possible to quantitatively recover soluble silver compounds from spiked samples. No problems arose with sample storage stability within the specified timeframe. Sampling and analysis of aerosols in silver refineries showed that the mass fractions of soluble silver compounds were less than 1.3% of total silver. It would be of interest to validate methods for other soluble metal compounds of concern in occupational health and safety.¹⁷

Acknowledgements

The authors gratefully acknowledge Kennecott Rawhide Mining Company for permitting multiple onsite studies to be conducted and Paul Pierce for the laboratory generated samples.

References

1. Occupational Safety and Health Act of 1970 (Public Law 91-596), 29 CFR 1910.1000, Table Z-1: Limits for Air Contaminants, U.S. Government Printing Office, Washington, DC, 1970.
2. Mine Safety and Health Administration, 30 CFR 57.5001: Exposure Limits for Airborne Contaminants, U.S. Government Printing Office, Washington, DC, 1978.
3. American Conference of Governmental Industrial Hygienists, Threshold Limit Values and Biological Exposure Indices, ACGIH, Cincinnati, OH, 6th edn, 1991.
4. European Commission, Occupational exposure limits: Recommendations of the scientific expert group 1991–92 (EUR 15091), European Commission, Luxembourg, 1994.
5. Health and Safety Executive, Metallic Silver—HSE review 1996 (Report no. D97), HSE, London, UK, 1998.
6. National Institute for Occupational Safety and Health, Registry of Toxic Effects of Chemical Substances (RTECS): Silver (VW3500000), NIOSH, Cincinnati, OH, 2003, http://www.cdc.gov/niosh/rtecs/vw3567e0.html, last update August 2003, accessed May 2005.
7. P. L. Drake and K. J. Hazelwood, Ann. Occup. Hyg., 2005, 49, 575–585.
8. International Organization for Standardization, ISO 15202-2—2001, Workplace Air – Determination of Metals and Metalloids in Airborne Particulate Matter by Inductively Coupled Plasma Atomic Emission Spectrometry—Part 2: Sample Preparation, ISO, Geneva, Switzerland, 2001.
9. ISO 15202-2, Annex B, Sample Dissolution Method for Soluble Metal and Metalloid Compounds, ISO, Geneva, Switzerland, 2001.
10. K. Ashley, Appl. Occup. Environ. Hyg., 2001, 16, 850–853.
11. NIOSH, NIOSH Manual of Analytical Methods, Method No. 7300: Elements by ICP-AES, NIOSH, Cincinnati, OH, 4th edn, 1994.
12. K. J. Hazelwood, P. L. Drake, K. Ashley and D. Marcy, J. Occup. Environ. Hyg., 2004, 1, 613–619.
13. Environmental Protection Agency, Inductively Coupled Plasma-Mass Spectrometry (Method No. 6020), EPA, Washington, DC, 1994.
14. E. R. Kennedy, T. J. Fischbach, R. Song, P. M. Eller and S. A. Shulman, Guidelines for Air Sampling and Analytical Method Development and Evaluation (DHHS [NIOSH] Publ. No. 95-117), NIOSH, Cincinnati, OH, 1995.
15. J. M. Antonini, Crit. Rev. Toxicol., 2003, 33, 61–103.
16. P. A. Baron, in NIOSH Manual of Analytical Methods, NIOSH, Cincinnati, OH, 4th edn, 2003, 3rd suppl., ch. O, Factors Affecting Aerosol Sampling.
17. K. Ashley and R. Fairfax, in NIOSH Manual of Analytical Methods, NIOSH, Cincinnati, OH, 4th edn, 2004, 3rd suppl., ch. M: Sampling and Analysis of Soluble Metal Compounds.

---

Footnotes

† Disclaimer: The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health (NIOSH). Mention of specific products and manufacturers does not imply endorsement by NIOSH.

‡ Presented at the Fifth International Symposium on Modern Principles of Air Monitoring & Biomonitoring, June 12–16 2005, Norway.

---