Species Analysis of Inorganic Compounds in Workroom Air by Atomic Spectroscopy
ANALYTICAL SCIENCES VOL. 7 SUPPLEMENT 1991 1029
SPECIES ANALYSIS OF INORGANIC COMPOUNDS IN WORKROOM AIR BY ATOMIC SPECTROSCOPY
SIRI HETLAND', IVAR MARTINSEN', BERNARD RADZIUK2 and YNGVAR THOMASSEN'
1 National Institute of Occupational Health , P.O.Box 8149 DEP, N-0033 Oslo 1, Norway
2 Department of Applied Research , Bodenseewerk Perkin-Elmer GmbH, P.O.Box 10 11 64, D-7770 Uberlingen, Germany
Abstract - Sampling methods for the determination of volatile inorganic compounds in workroom air are described. The main factor affecting the adsorption capacity of activated charcoal for iron pentacarbonyl is air flow rate. Nickel tetracarbonyl is collected quantitatively at flow rates of up to 11/min if the volume of air sampled does not exceed 1001. Arsine, phosphine and stibine are collected quantitatively from air by use of a double filter air cassette Silver nitrate impregnated backing pads are shown to trap the hydrides with better than 99.9% efficiency. Several atomic spectroscopic methods have been applied to the determination of these volatile species, with electrothermal atomic absorption and inductively coupled plasma atomic emission spectroscopy being preferred. Concentrations of the hydrides in various workroom atmospheres were found to be well below the Threshold Limit Values.
Key words arsine, stibine, phosphine, iron pentacarbonyl, nickel tetracarbonyl, species analysis
The term chemical species refers to a specific form (monatomic or molecular) or configuration in which an element can occur, or to a distinct groups of atoms consistently present in different compounds or matrices . The individual species may be gaseous, solid, crystalline or in solution depending upon the nature of the sample. A parent species is an unchanging component (e.g. CH3Hg+), a matrix species is a combined form (e.g. CH3HgL, with L-OH, protein etc.) while an analyte species is the form (e.g. CH3HgC1) detected by the analytical method used. Species analysis( sPeciation) is the identification and quantification of these three types of species. Species distribution describes the abundance or fractional occurrence, species reactivity is related to characteristic reactions, responses or effects while species transformation is the conversion of one species to another. The above definitions have recently been suggested for use in "speciation'' studies {1J.
Exposure of workers in occupational settings is most frequently to inorganic compounds present mainly in particulate or volatile forms (gases). The recently increasing interest in the occurrence of organometallie compounds in the work environment has also led to a demand for new sampling and analytical methods with adequate selectivity and sensitivity. An excellent review of this topic has been published in Arbete och Halsa [2].
The Threshold Limit Values( TLV's) which have been established in most countries refer to airborne concentrations of substances and represent conditions under which it is believed that nearly all workers may be repeatedly exposed day after day without adverse health effects. These limits are neither fine lines between safe and dangerous concentrations nor a relative index of toxicity. In spite of the fact that serious injury is not believed likely as a result of exposure to the threshold limit concentrations, the best preventive practice is to maintain concentrations of all atmospheric contaminants as low as possible. Moreover, the potential health hazard posed by chemical substances in inhaled air depends mainly on the parent species. While most guidelines are set for the total amounts of elements, some threshold limit values for parent species have been recommended as well. The most important inorganic substances for which speciation is needed are listed in Table 1. Of the species listed in Table 1, hydrogen fluoride occurs the most frequently in workroom air. This compound is emitted during production of primary aluminium metal and is one of the important components in the evaluation and monitoring of the exposure of workers in the aluminium industry. The preferred method for determination of HF in air is trapping on sodium hydroxide soaked filters followed by analysis using an ion selective electrode. The carbonyls of iron and nickel are highly toxic compared with other compounds of these metals which are found in workroom air. These volatile compounds are, however, only generated under particular conditions, e.g. during 1030
production of nickel metal using the carbonyl process. There has arisen some concern that these compounds may also be produced in environments with elevated temperatures and high concentrations of c arbon monoxide in the presence of iron and nickel metal particles.
TABLE 1. TLV's for some volatile inorganic species
Volatile hydrides of arsenic, antimony and phosphorus are formed during certain industrial processes. During charging processes in the manufacture of lead batteries nascent hydrogen is generated which reacts with traces of antimony and arsenic in the sulphuric acid solutions to form stibine and arsine. The reaction of phosphides , arsenides and antimonides with atmospheric humidity forming the elemental hydrides may occur both in the metallurgical and metal working industries. Sampling of workroom air in which these parent species are present in combination with particulate forms of the same elements requires methods for the separation and quantitative collection of the different species. The air sampling is accordingly performed by means of various types of filters for particulates or involatile aerosols, while the volatile compounds are collected on adsorbents [3]. Various analytical techniques may be used to detect the analyte species collected on the adsorbent or filter media. After sample preparation with inorganic acids these species will be present in aqueous solutions which are well suited for quantitative measurements by, for example, inductively coupled plasma atomic emission spectrometry or atomic absorption spectrometry either with flame or graphite tube atomization.
The presently available sampling procedures do not make possible the separate measurement of the above mentioned carbonyls and hydrides in general occupational settings. In the present study, sampling procedures based on the combination of a membrane filter with a tube containing activated charcoal adsorbent and on a two-filter air sampling cassette have been investigated. Both atomic absorption and atomic emission spectroscopic procedures have been developed for quantification of these species in a variety of occupational environments.
EXPERIMENTAL
Apparatus A Perkin-Elmer Zeeman 5100 atomic absorption spectrometer equipped with an HGA 600 graphite atomizer, an AS-40 automatic sampler and hollow cathode and electrodeless discharge lamps was used . Pyrolytical graphite coated tubes were applied throughout the study. Atomic emission measurements were performed using a Perkin-Elmer 5500 Inductively Coupled Plasma Atomic emission spectrometer. A peristaltic pump (0.75 mIJmin) introduced the sample solution into a cross-flow nebulizer operated at an argon flow of 1.0 llmin which was controlled by a mass-flow meter . Additional argon was supplied to the quartz torch (131/min). The plasma was maintained at 1.2 kW applied with a frequency of 27.2 MHz. For the determination of the adsorption capacity and breakthrough time for air sampling , a manifold was used in which up to 10 units could be exposed simultaneously to gas mixtures or real workroom atmospheres . Air was sampled using Personal Sampling Pumps with fixed flow-rates of 0.1-2.0 1/mm. Reea ents Commercial adsorbent tubes (SKC , Pittsburg, PA) packed with activated carbon were used. The tubes were composed of two sections , the first containing 100 mg and the second 50 mg. Measurement on adsorption capacity could be made by analysing the material in the two sections separately . Activated Carbon (Fluka) was ground and the 0.6-1.2 mm fraction was boiled with 65% nitric acid for 1 h, five times in succession . Th e carbon was then washed with demineralized water , air-dried and reactivated in air at 450 °C for 30 min. This procedure reduced the concentrations of iron and nickel (see Table 3). Cellulose acetate membrane filters with 0.8-pm pore-size and 37mm diameter were used for the collection of particulate matter from air. The cellulose backing pad which supports the membrane filter in the standard air sampling cassette (Nucleopore) was dipped in a solution containing 2.5% AgN03 by weight and oven dried at 75 C overnight. This double filter system was used to collect particulates and hydrides simultaneously. Procedures for dissolution extraction The cellulose membrane filters were dissolved and the charcoal fractions extracted with nitric acid [4J. The silver nitrate impregnated backing pads were extracted with 2.5 ml 65% HNp3 ANALYTICAL SCIENCES VOL. 7 SUPPLEMENT 1991 1031 at 100° C for 1 h, and the final solution was diluted to 25 ml with demineralized water . The results were corrected for the volume of undissolved material (activated charcoal and cellulose fibers) by preparing standard solutions containing the same amount of charcoal and a backing pad, respectively.
RESULTS AND DISCUSSION
Collection efficiencies-carbonyls Charcoal-filled collection tubes were exposed to various volumes of gas mixture containing the two metal carbonyls at TLV's concentrations at various flow rates in order to assess adsorption efficiency. Table 2 gives a summary of the adsorption capacity data.
TABLE 2. Adsorption capacity for activated charcoal tubes
As can be seen in the table, breakthrough losses for iron pentacarbonyl are significant even at low flow rates. The recommended maximum air flow is 0.5 llmin for sampling periods less than 100 min. No detectable breakthrough was detected for nickel tetracarbonyl under the experimental conditions. Recovery studies of the adsorbed carbonyls based on nitric acid extraction was between 95 to 105 % for all exposed tubes. No detectable amounts (<1%) were found in ethanol extracts of the charcoal indicating a rapid species transformation catalyzed by the adsorbent. The transformed species, presumably the elemental forms, are quantitatively extractable with concentrated nitric acid. Detection of carbonyls Table 3 shows that the blank values of the charcoal for iron and nickel seriously influence the detection limit of the method.
TABLE 3. Iron and nickel extractable from two types of carbon adsorbents and the detection limits of the method
The amount of iron remaining in the adsorbent even after careful acid washing and reactivating is so considerable that the detection limit is limited to one tenth of the TLV for iron pentacarbonyl. Lower concentrations could be detected if uncontaminated charcoal were available. The detection limit for nickel tetracarbonyl is adequate for measurements under workroom conditions for which the concentration is considerably lower than the TLV level. Collection of hydrides The collection capacity of silver nitrate impregnated backing pads for arsine and stibine was tested in real workroom atmospheres during charging of grids for lead batteries. A manifold containing parallell double filter cassettes was exposed for 8 hours with a flow rate of 1.8-2.0 llmin. The cassettes were 1032
packed with two impregnated pads i n series in order to measure any bre akthrough of the hydrides. No detectableof eitherantimonyor amounts arsenicw ere found in the second pad even at atmospheric concena'ations whichtheTLV exceededb y a factor of five. Under all sampling conditi . This effeciency was also c ons, more than 99.9 % of the hydrides collectedfoundin the werefirst pad onfirmed far phosphine under laboratory condiuons.It is presumedthatthe parent volatile hydride species are tr ansformed by solid silver nitrate into non-volatile silver phosphides/arsenides/antimonid es. These transformed species are e asily extracted into nitric acid and recoveries99±2%havebeen ofestimated by c omparison against pads which had b een spiked with standard solutions. Thelimits fordetectionthe hydrides are giv en in Table 4 far the different at omic spectroscopic techniques applied.
TABLE 4. Typical detection limits` in mg/m3 for ato mic spectroscopic techni ques applied in the measurement of airborne hydrides .
The methods of choice are either ICPJAES for detection of the three elements at concentrations of interest ar ETAAS with a considerably higher sensitivity but the limitation of single element determination. The silver extracted with the analyte is beneficial in ETAAS as it both thermally stabilizes arsenic , antimony and phosphorus up to 1100° C and renders the matrix of the sample uniform . Calibrations of the two methods can be done using standard solutions matrix-matched with regard to silver and nitric acid . The high concentrations of silver in the solutions interfere in hydride generation atomic absorption spectrometry and this method is not generally applicable without further development. Measurements of hydrides in working environments The present method has been used in the species analysis of hydrides in workroom air. Phosphine was determined during the turning of cast iron and various other metallurgical processes.The concentrationsranged from slightlyabove the 'Tq,Vto less than 0.003 mg/m3.'3'he majority of the results were lower than 0.01 mg/m3. Concentrations higher than the TI..V were found in only 3 of 150 filters exposed using personal samplers. Even for those workers complaining of the characteristic smell of phosphine there are very few examples from our studies which indicate that workers may be exposed to excessive amounts of phosphine. The majority of measurements made for arsine and sribine during production of lead batteries gi ve results which a re also well below the UV's . The average arsine and sribine concentrations measured over the period 1985-1990 in several battery plants in Norway were 0 .002 (n=90) and 0.04 (n=150) mg/m3, respectively. Thus, based on thethe workingTLV's conditi ons with respect to these compounds are acceptable. However, biological monitoring of antimony in whole blood and urine of workers has shown considerably elevated concentrations compared with non-exposed groups indicating a high absorption rate in the lungs. This is under further investigation.
REFERENCES
1. LE. Nieboer,XXVII-CSI Post-Symposium on Speciationof Elements in Environmentaland BiologicalSciences. oen, Norway,June lb-1$, 1991. 2. O. Nygren,Arbeta och Halsa, 27, (1987). 3. HP.M.Eller(Ed). NIOSH Manual of AnalyticalMethods , 2,3rd ed., NationalInstitute for OccupationalSafety and ealth, Cincinnati,Ohio, USA,(1984). 4. O. R~ysetand Y. T'homassen, Anal.Chim. Acta,188,247 (1986)