J. Cross, ChemCounsel Corporation, Houston, TX.
Diffusion dosimeters are well-accepted for their ease of use and the promise of accuracy implied by their adherence to Fick’s First Law. In practice, however, performance differs from promise. Dosimeters with short diffusion paths have been found to deviate from predicted performance. This deviation has meant that dosimeter performance could not be predicted and that validation schemes had to be empirical. More importantly, the parameters controlling sampling rates could not be clearly characterized. Both of these conditions have, in the opinion of the author, prevented fuller use of dosimeters. Recently, the development of Gradient Analysis has demonstrated a better way to apply Fick’s law to dosimeters. This paper extends Gradient Analysis to describe a validation scheme that significantly reduces the validation effort and at the same time increases understanding of the mechanisms controlling the sampling process.
The validation scheme begins with separating the sampling process from the adsorption process. This is possible, because the analytical method associated with Gradient Analysis does not depend on the properties of the adsorbent. Separating the sampling device from the adsorbent divides the validation process into two more-easily-managed efforts. Two critical adsorbent properties, capacity and rate of adsorption, will be discussed as they relate to sampling conditions encountered in the field. The parameters of the sampling process that will be discussed include predicting the sampling rate (the calibration factor), determining the effect of inadequate airflow, correcting for airflow effects encountered in a specific application, checking the validation at periodic intervals, and validating a dosimeter for multiple compounds.
W. Hendricks, G. Schultz, U.S. DOL/OSHA, Salt Lake City, UT.
OSHA researchers have developed a new sampling and analytical method to measure low-level occupational exposures to moderately volatile organic chemical substances. The new method features diffusive sampling with analysis by thermal desorption and GC/mass spectrometry. The analytical technique permits simultaneous identification and accurate quantitation of many chemical substances that are often encountered in sampling previously uncharacterized workplace atmospheres. The method is intended for use to monitor exposures of one part-per-million or less with sampling times as long as eight hours. The new method represents a significant departure from traditional methods that are typically validated for only one substance at a high level. Low levels of many substances are difficult to monitor with traditional methods and exposure to these low levels often cause adverse health effects and employee complaints. Diffusive sampling rates were determined for the 16 solvents that were tested, and literature or surrogate sampling rates could be used to estimate air concentrations of untested substances. The new method is also sufficient for use in environmental and indoor air quality investigations.
S. Tsai, H. Chen, China Medical University, Taichung, Taiwan, Republic of China.
Exposure to BTEX has been associated with skin and sensory irritation, central nervous system depression, and effects on the respiratory system. Sources of exposure to BTEX vary from industrial processes, automobile emissions, as well as indoor air pollutions. A rapid, cost-effective, and accurate sampling system for exposure assessment to low-concentration BTEX therefore would be of extreme benefit. The aim of this research was to develop a new diffusive sampler for BTEX in air based on the technique of solid-phase microextraction. The Carboxen/Polydimethylsiloxane fiber was used, and the fiber assembly was put into a Teflon tubing to construct the sampler. The sampling constant was determined by exposures at known face velocities and vapor concentrations which were prepared in gas bags and/or by a dynamic standard gas generation system. After sampling, the sampler was inserted into the injection port of gas chromatography/mass spectrometry for thermal desorption and Selected Ion Recording mode was used for sample analysis. The desorption efficiency was 100% when the time for thermal desorption was 5 min. The experimental sampling rates were (7.46 ± 0.63)×10-3, (6.61 ± 0.87)×10-3, (7.22 ± 0.71)×10-3, and (7.39 ± 0.84)×10-3 cm3/min, for benzene, toluene, ethylene, and o-xylene, respectively. The experimental sampling rates of BTEX were not affected by different temperatures (4°C ~ 25°C) and wind velocities(0 ~ 50 fpm). Relative humidity at 80% also had no effects on the sampling rates for benzene and ethylbenzene (compared to RH = 10%) while statistical differences were observed for toluene and o-xylene. Besides, the designed method was found to be able to collect BTEX effectively at instantaneous low-concentration (e.g., 5 min at 0.1 ppm) and sample stability test showed that the recovery was around 100 ± 20% after 48 hours of storage at 4°C. Compared with other methods, the current designed sampler provides a convenient and sensitive tool for the exposure assessments of indoor low-level VOCs .
W. Fowler, D. Segers, C. Lagrone, S. Keeton, CMS Field Products–OI Analytical, Pelham, AL.
Due to the very high toxicities of the chemical warfare (CW) agents, they must be detected at exceedingly low concentrations if their adverse health effects are to be avoided. Consequently, today’s CW monitoring and analysis methodologies tend to rely heavily on the use of solid sorbents for preconcentrating CW vapors prior to thermally desorbing them into a suitable chemical-analysis instrument. This talk provides an overview of solid-sorbent sampling and thermal desorption as applied to CW detection and identification, with an emphasis on recent developments and experience in this field. Specific topics to be discussed include sorbent selection, sorbent conditioning, sorbent tube construction, mechanical problems with sorbent tubes, useful sorbent/detector combinations for CW detection, CW breakthrough volumes, optimum sorbent particle sizes, and optimum conditions for sampling and thermal desorption. Laboratory data are presented to illustrate a number of issues that can arise during solid-sorbent-based CW monitoring and analysis.
C. Lungu, J. Park, University of Minnesota, Minneapolis, MN.
Granular activated carbon is commonly used as an adsorbent in air purifying respirator cartridges. The performance of these cartridges is usually assessed by conducting breakthrough tests using a single challenging compound. However, these performance tests have little relevance when the cartridges are used in the field against a compound having different properties compared with the test compound. There are very few published data available to compare the adsorption characteristics of a certain type of granular activated carbon challenged with compounds having different chemical characteristics. In this study, the dynamic adsorption of an aromatic non-polar compound (Toluene) was compared with the adsorption of a polar compound (Ethanol) on an 80-CTC coconut shell activated carbon. The adsorption isotherms of the two compounds were obtained separately from the breakthrough curves generated by continuous injection of the contaminants at controlled rates into the airflow passing through a fixed activated carbon bed. The carbon sample (2 g) was placed into a copper cylinder immersed in a temperature controlled water bath. A gas chromatograph and an infrared gas analyzer monitored the effluent concentration. The challenge concentration was 50, 100, 150, 200, and 250 ppm for toluene and 50, 100, 200, 400, and 600 ppm for ethanol. By fitting each breakthrough curve with the known Wheeler equation the adsorption capacity, We and the adsorption rate constant kv were determined. As expected the adsorption capacity for ethanol was much reduced compared with toluene. The adsorption capacity ranges from 341 to 480 mg/g for the toluene challenge concentration interval, and from 24 to 141 mg/g for ethanol. The corresponding breakthrough time for 50, 100, and 200 ppm is higher for toluene with 232, 156, and 87%, respectively. The influence of humidity on the adsorption of the two compounds was also assessed.
S. Paik, Abbott Laboratories, North Chicago, IL; M. Puskar, Abbott Laboratories, Abbott Park, IL.
Two important factors to consider when measuring potentially explosive atmospheres are instrument accuracy and response time. These factors are especially critical when there are rapid rises in concentration, since they determine how confidently one can set alarm levels to ensure that such atmospheres are not formed. Initial measurements obtained from the exhaust duct of a fluid bed dryer (used for drying solvent-laden pharmaceuticals) using a MIRAN 1A® Gas Analyzer showed that a rapid rise in vapor concentration occurred during initial startup. The actual concentration attained was difficult to predict due to the delayed response of the instrument. For the optimal protection of equipment and personnel, an instrument was sought that could (1) accurately measure multiple vapors simultaneously, (2) quickly respond to the concentrations, and (3) conveniently be installed as a continuous real-time monitor. Towards these ends, a Fourier Transform Infrared (FTIR) spectrometer with a custom 5-cm extractive cell was evaluated using acetone, ethanol, isopropyl alcohol, and isopropyl acetate as the test vapors. The tests revealed that the accuracy for single vapor measurements was better than ± 5% (with slightly worse accuracies for mixtures) and the T90 response time, defined as the time it takes for an instrument to respond to 90% of its stabilized concentration upon exposure, was approximately 0.7 seconds for all the vapors tested. The FTIR was also shown to have many practical benefits in the field. An embedded temperature and pressure probe enabled measurements to be automatically corrected for air density differences between sample and reference conditions. The embedded computer’s hard drive enabled measurements to be obtained at 0.7-second intervals over the course of several days, and potentially several weeks. These features, in addition to data output capabilities for process control, make the FTIR an ideal instrument for real-time and continuous monitoring of potentially explosive atmospheres.
M. Harper, M. Andrew, NIOSH, Morgantown, WV; T. Hallmark, Mississippi Space Services, John C. Stennis Space Center, MS.
Personal and area air samples were taken at a scrap lead smelter operation at a bullet manufacturer using the 37-mm styrene/acrylonitrile filter cassette, the 37-mm GSP or “cone” sampler, the 25-mm Institute of Occupational Medicine (IOM) inhalable sampler, and the 25-mm Button sampler (developed by the University of Cincinnati). Pure, homopolymer, polyvinylchloride filters were used to capture lead particulate. The filters were pre- and post-weighed, and analyzed for lead content using a portable x-ray fluorescence (XRF) analyzer. The filters were then extracted with dilute nitric acid in a sonic bath and the solutions analyzed by inductively coupled plasma optical emission spectroscopy (ICP-OES). The 25-mm filters were analyzed using a single XRF reading, while three readings on different parts of the filter were taken from the 37-mm filters in accordance with current NIOSH/OSHA methodology. The single reading from the 25-mm filters was adjusted for the nominal area of the filter to obtain the mass loading, while the three readings from the 37-mm filters were inserted into two different algorithms for calculating the mass loadings, one from NIOSH and one from OSHA, and the algorithms were compared. The IOM is the only sampler where material collected in the sampler, but not caught on the filter, is intended to be part of the sampler. Therefore, the cassettes were rinsed separately to determine how this might bias the on-filter analysis, but ICP analysis found only insignificant amounts of lead, in line with other studies. All four samplers gave very good correlations between the two analytical methods above the limit of quantitation of the XRF procedure, although the limit was lower for the 25-mm filters (3 µg) than for the 37-mm filters (10 µg). For both types of 37-mm filter, the OSHA algorithm gave results closer to the ICP values than the NIOSH algorithm.
C. Glowacki, J. Stone, Technikon LLC, McClellan, CA.
Workplace exposure to hazardous materials in the metal casting industry has been the topic of many publications and presentations over the last 30 years. Most have focused on physical hazards such as heat and physical stress, inorganic chemicals such as silica, and a limited number of organic compounds. The Casting Emission Reduction Program (CERP), a public/private cooperative research and development program, has identified numerous compounds emitted from a variety of metal casting processes that greatly increase the potential for excessive workplace exposures. Many of these compounds and the method(s) of their formation and release to the environment have not been reported in the literature. The CERP target analyte list contains over 100 specific compounds that are routinely measured to concentrations of 10 to 100 parts per billion. This presentation will detail those organic compounds emitted from major metal casting processes such as core making, mold making, and pouring, cooling, and shakeout activities. The information will be subdivided by the type of metal being cast, the chemistry of the binder system used to prepare the core or mold, and the type of molding or core making process. The specific compounds found in each scenario will be ranked in order of the amount of compound emitted compared to the weight of metal cast, sand used, and/or binder used.
M. Reed, Trident Technical College, Charleston, SC; D. Underhill, C. Feigley, University of South Carolina, Columbia, SC.
The diffusive sampler is a valid air sampling device which can be used to measure airborne contaminants in the workplace. The primary limitation to the diffusive sampler is the uncertainty in the determination of the diffusivity coefficient as the diffusivity coefficient varies with environmental factors that affect complete sample uptake and retention. Accurate determination of the airflow rate cannot always be achieved due to the uncertainty of the diffusivity coefficient affected by these factors. Our objective was to improve the accuracy and assess abuse or misuse of the diffusive sampler by developing a self-validating diffusive sampler that will significantly improve the quality of air sampling data. A volatile pre-impregnate was investigated with adsorptive properties that will result in reverse diffusion under ambient conditions and that has a chemical half-life. The properties of the volatile pre-impregnate include: (1) a sufficiently high vapor pressure such that significant desorption will occur if the sampler is allowed to desorb after the sampling period; (2) easy detection so that only trace levels of the volatile pre-impregnate need to be placed onto the adsorbent; (3) not commonly present in the workplace so that uptake of the volatile pre-impregnate from the atmosphere will not be a confounding factor; (4) chemically stable; (5) readily available; and (6) low in toxicity. Diffusive samplers were exposed to the volatile pre-impregnate in a chamber, allowed to reach equilibrium, exposed to the ambient atmosphere, and desorbed with chlorobenzene. The volatile pre-impregnate was analyzed by gas chromatography with a flame ionization detector and confirmed by gas chromatography with mass spectrometry. There are very few chemical compounds that have the physical and chemical charcteristics suitable for use as a volatile pre-impregnate in a self-validating diffusive sampler. Chloromethane meets the criteria and appears to be the most promising volatile pre-impregnate.
L. Saarinen, Finnish Institute of Occupational Health, Helsinki, Finland.
Methylene bisphenyl diisocyanate (MDI) is atomised and evaporated during the exothermic polymerisation process. The data on the relationship between exposure to MDI and the effects it induces are scarce. This may be due to the complex etiology of the effects and/or in the shortcomings of exposure measurements. Sampling methods for MDI were compared in pairs to identify problems.
In the test chamber, the MDI aerosol was produced with a Liu-Lee aerosol generator without a polyol component, and a vapor-aerosol mixture was generated in the polymerisation process. The aerosols were characterized with a scanning electron microscope and a particle size counter. Solvent absorption and variable coated filters were used in the tests. 1-(2-methoxyphenyl)piperatzine (MPP) was the derivative reagent. The filters were coated with 20 or 2 µmol MPP in toluene. The samples were collected under variable airflows and analysed with a high-performance liquid chromatograph.
The collection efficiency for solvent absorption methods was mainly a function of flowrate. The sampling efficiency was poor for particles under 1 µm in size. The loss depended also on the slowness of the reaction rate both in the glassware and on the coated filters. The collection efficiency in the solid-state system of coated filters depended on sampling rate, amount of applied reagent, diffusion, and porosity.
All the tested sampling methods were incomplete, and underestimated the actual concentrations of fast curing MDI present as a mixture of aerosol and vapor. To minimize the underestimation, complementary methods for determining the exposure to MDI are recommended to be used simultaneously. The detection limit of analysis restricts the measurement of very short but high exposure peaks. It was therefore difficult to assess the actual exposure and dosage. Alternating efficiencies under variable sampling conditions impede the establishing of the dose-response relationship.
M. Harper, M. Andrew, NIOSH, Morgantown, WV; Z. Akbar, University of Alabama at Birmingham, Birmingham, AL; B. Muller, Honda Manufacturing of Alabama LLC, Lincoln, AL.
Particle size distributions in the inhalable size range collected by different personal samplers for wood dust were compared. Samples were collected over a short sampling period (<2 hours) on PVC filters, removed, suspended, and re-deposited evenly on a MCE filter, which was then cleared, and portions examined by optical microscopy. The aerodynamic equivalent diameter (AED) of particles was calculated from their dimensions, shape factor, and density. This method is particularly appropriate to wood-dust particles which are generally large and close to rectangular prisms in shape. Personal samples were collected using the traditional 37-mm closed-face polystyrene/acrylonitrile cassette (CFC), the Institute of Occupational Medicine (IOM) inhalable sampler, and the Button sampler developed by the University of Cincinnati. Total mass concentration results from the method described above were in approximately the same ratios as those from traditional long-term gravimetric samples, but about an order of magnitude higher. The particulate mass appears to be concentrated in the range 10–70 µm AED, but with contributions from particles larger than 100 µm AED in 65% of the IOM samples, 42% of the CFC samples, and 32% of the Button samples. Where present, these “ultra-large” particles dominated the mass collected, contributing an average 53% (range 10–95%), but significant differences were still found even after accounting for their effect. The IOM and CFC samplers appeared to operate in accordance with previous laboratory studies, as they both collected similar quantities of particles at the smaller size-ranges, up to about 30–40 µm AED, and for larger size-ranges the CFC collection was reduced compared to the IOM. The Button sampler collected significantly less than the IOM at most particle sizes. The Button sampler appears to sample less than the CFC at smaller particle diameters, consistent with some, but not all, relevant laboratory studies.
S. Erdal, L. Brown, S. Freels, L. Conroy, N. Esmen, University of Illinois at Chicago, Chicago, IL.
Exposure to wood dust has been associated with a number of health outcomes including respiratory irritation and nasopharyngeal cancer. It is important to measure wood dust concentrations that are encountered in a workplace to determine compliance and to provide necessary input to health effects studies. The customary exposure assessment tools used for aerosol monitoring involve samplers, which provide time-integrated measures of exposure. However, samplers with the near real-time capability have the ability to significantly improve our understanding of temporal distribution of workplace exposures. They also allow assessment of specific activities in the workplace, which result in high exposures. This information is very useful for exposure reduction and risk management purposes. We conducted a wood dust exposure assessment study in a mid-size wood working facility using a number of exposure assessment tools at a static sampling location in the middle of the facility. For time-integrated measures of exposure, 37-mm closed-cassette and a Marple Cascade Impactor were used. For near real-time measurements, MIE’s Personal DataRam (pDR-1200) was used. Sampling for total dust was performed on 10 different days in which different activities by the workers were performed (e.g., sawing, sanding, cutting). Sampling duration was four hours in each sampling event. While the MCI total dust measurements varied from 0.5 to 2.8 mg/m3, the 37-mm measurements at the same location were in the range of 0.2 to 1.3 mg/m3. The comparison of the time-integrated data against the near real-time data for total wood dust revealed the importance of obtaining information on temporal distribution of exposure, thus the need for development of reliable real-time monitoring tools and instruments for aerosol monitoring in the workplace.
Posted May 30, 2004