Poster Session 4

Poster Session 4

Author Attend Time: Tuesday 1:00 p.m. - 3:00 p.m.

*All posters are available for viewing in the expo hall from Monday 9:00 a.m. through Wednesday 1:00 p.m.

SR-404-01 Stu​dy on Generation Characteristics of Powder By-products for Novel Etching Process in the Semiconductor Manufacturing Industry

H. An, K. Choi, D. Kim, Samsung Electronics, Yong-in, Republic of Korea

Objective: Etching gases are the halogen compounds such as HBr, Cl2, CF4, BCl3, and NF3 mainly used for the chemical reaction; the b​​y-product also depends on the etching target substances. Since the trend in recent years of new target materials can be applied it is necessary to check by-product according to the change of the etching material. BSI (Backside Illuminated Image Sensor) is the new product with the backside of the wafer being used tungsten (W), titanium (Ti) and the like, is applied at least 6-10 times thicker than the conventional. Therefore, the focus of this study is to verify the change in the hazard of new process equipment of BSI to verify by-products generated in addition to etch gas.

Methods: The equipment studied are new process units of etching tungsten & titanium compounds and the material of the conventional silicon, the same model. By-product particles were collected from the inner part materials and air sampling during preventive maintenance of process equipment. The chemical composition of by-products was determined by field emission scanning electron microscopy (FE-SEM) equipped with energy dispersive spectroscopy (EDS) respectively.

Results: At the inner part materials, the particle compounds found were aluminum (Al), fluorine (F), titanium (Ti) and were expected, such as Titanium fluoride (TiF4), Aluminum chloride (AlCl3), aluminum oxide (Al2O3), titanium dioxide (TiO2). There was no big difference between the new and existing process equipment, in the case of area samples with personal air sampler except for the environmental component. In the new process equipment oxygen (O), silicon (Si), chromium (Cr), chlorine (Cl) were generated; additionally an aluminum (Al), iron (Fe), bromine (Br) were also generated in the conventional process equipment.

Conclusions: From the results of SEM-EDS analysis, other types of by-products occur in the new process equipment to produce BSI. For example, tungsten hexafluoride (WF6) is generated by the chemical reaction with tungsten and nitrogen trifluoride (NF3). This material is expected by adsorption on the chamber side wall after reaction with titanium (Ti). In summary, it would be demonstrated that by-products generated not from etch gases but target materials. Also, we figured out those titanium (Ti) by-products were not scattered in to the air. The present study elucidated the hazard characteristics such as chemical composition from by-products considering target materials.

CS-404-02 Bring Back the Impinger

D. Duffy, ESIS, Inc., Chicago, IL

Situation/Problem: Impingers were the gold standard devices for evaluating exposures and conducting area air monitoring for various inorganic acids, as well as compounds such as acetic acid, methanol, formaldehyde, hydrogen peroxide and others. We all agree that the use of impingers was cumbersome, difficult for the monitored employee and as CIH consultants, we were anxious to find a substitute for the impinger. Now, due to safety and security issues for travel, transporting impinger solutions and liquid samples has added to the difficulty. Since the days when impinger use was common, new media has been developed, including solid sorbent tubes, treated filters and passive monitoring badges, all with the purpose of making life easier for the CIH. At issue and as discussed in this poster, is whether we have sacrificed accuracy for convenience. In many instances, we were not finding significant contaminant concentrations with the use of solid sorbet media in relation to visual observations and employee concerns that suggested that higher exposure levels may be present. 

Resolution: To address this issue, we decided to conduct side-by-side exposure monitoring using impingers and the current and accepted “non-impinger” monitoring method. A portion of the side-by-side air samples were collected in the industrial environment(s) and a portion in the laboratory. We also decided to begin this study by conducting side-by-side monitoring for a variety of industrial chemicals as opposed to one group, such as inorganic acids.

Results: The use of impingers for many of our samples showed higher contaminant concentrations, and at least preliminarily, suggest that lower recovery and detection were observed on solid media that was used for comparison. Particle size of the contaminant and the presence of water vapor both can affect recovery on treated filters and on solid sorbent media, especially silica gel tubes. Side-by-side testing will be presented. 

Lessons Learned: Solid sorbent tubes, passive monitors and treated filters have replaced the impinging devices for exposure monitoring. Our data suggest that this replacement, while making the job of the CIH easier, may not be accurately assessing employee exposures and area air contaminant levels. It may be time to re-evaluate the need for impingers, and for some chemicals, to develop more user-friendly impinging devices. 

CS-404-03 Sampling for Peak/IDLH Exposures Using Field Portable GC/MS and Laboratory GC/MS to Investigate and Characterize Chemical Exposure

D. Pearce, J. Hill, P. Smith, OSHA, Sandy, UT

Situation/Problem: Occupational sampling for an organic chemical of interest is usually performed by taking samples on activated charcoal tubes. Samples that are collected for 15 minutes, when an exposure is only a few seconds to minutes, may produce low to not-detected results even when an exposure may have exceeded a peak or IDLH value.

Resolution: This project looked at headspace and personal sampling of tank gauging in the oil and gas industry. Nearly instantaneous sampling was performed at several locations by using a vacuum box to capture air into an aluminized gas bag. Samples were collected at tank openings and personal samples were collected from employee breathing zones. A field portable gas chromatograph mass spectrometer using needle trap with tri-bed sorbent and thermal desorption was used to qualitatively characterize these samples in the field. Then the hydrocarbons from the bag were stabilized on replicate charcoal tubes for quantitative analysis in the laboratory by GC/FID and GC/MS using carbon disulfide extraction. Samples of bulk crude oil were also collected for comparison.

Results: Several sites were characterized. Coconut shell charcoal tube sample results for most of the sites were essentially non-detects. However one sampled site (synthetic charcoal) had significant findings for the headspace sample and lower results for the personal air samples. All of the bulk samples gave similar results for C3-C20 hydrocarbons. Correlation of the field portable GC/MS results and the lab analyses showed good qualitative agreement. Quantitation was performed on 12 C5-C7 hydrocarbon solvent analytes. These analytes combined to produce bulk concentrations between 7–12%, headspace between 10,000-15,700 ppm, and personal between 0-1,500 ppm.

Lessons Learned: Exposure assessment for Peak and IDLH situations using a bag to collect instantaneous samples and either immediate analysis or stabilization to a sorbent tube can be a useful technique. Using field portable GC/MS to characterize a site can help determine what type of samples to collect showing that in this case light C3-C7 hydrocarbons should be characterized. Synthetic and coconut shell charcoal tubes do not retain propane and butane as desired. Even though the transfer from the bag to the tube passed 0.5 liter, significant breakthrough was observed on charcoal tubes when analytes were present in high concentrations.

CS-404-04 An Evaluation of a Quantitative X-ray Diffraction Method for Analysis of Asbestos-Containing Materials

L. Greene, J. Ennis, W. Winstead, RTI International, Research Triangle Park, NC

Situation/Problem: The use of appropriate test methods to identify and quantify asbestos in bulk samples is critical to ensure the best decisions may be made to protect human health. Although polarized light microscopy (PLM) is widely accepted in North America and Europe as the standard analysis method for bulk asbestos, powder X-ray diffraction (XRD) is also a useful tool for asbestos analysis. In the United States, XRD is included as a complement to PLM in the Environmental Protection Agency’s 1993 Method for the Determination of Asbestos in Bulk Building Materials. XRD is limited, however, by potential interferences from components commonly found in bulk building materials and may not be able to achieve a detection limit low enough to identify asbestos at or near 1% in all bulk samples.

Resolution: A quantitative XRD method was developed for the determination of asbestos in bulk materials. This standard additions method differs from the internal standard method outlined in the 1993 EPA test method. Powder mixtures of known composition were used for preliminary evaluation of the method. Both the method and the sample preparation techniques used to remove matrix interferences were further refined in routine use of XRD for the characterization of “real-world” materials used as proficiency testing materials.

Results: Analysis of formulated mixtures of known composition indicates that the XRD method has good weight composition accuracy and precision, with a detection limit approaching 1% or even lower. Matrix interferences challenge application of the method to complex real-world materials in which there is peak overlap. However, judicious application of gravimetric sample reduction techniques greatly enhances the method’s ability to quantify asbestos at low levels in many real-world materials.

Lessons Learned: The successful application of quantitative powder XRD to bulk asbestos analysis is sample-dependent. Method limitations include potential matrix interferences and the inability to differentiate between fibrous and nonfibrous analogs. When both fibrous and nonfibrous mineral forms occur in one sample, the presence of the nonfibrous component hampers quantitation of the fibrous component by XRD. XRD may be most effective as a complement and confirmation to microscopy techniques, such as visual estimation, that are subjectively dependent on analyst training and experience.

SR-404-05 Comparison of the Standard and Dark-Medium Objective Lens in Counting Asbestos Fibers by Phase-Contrast Microscopy

E. Lee, J. Nelson, M. Kashon, M. Harper, NIOSH, Morgantown, WV

Objective: A Japanese round-robin (RR) study revealed that the analysts who used a dark-medium (DM) objective lens reported higher fiber counts of Proficiency Analytical Testing (PAT) chrysotile samples than those with a standard objective lens but no causes of such differences were investigated. The purpose of this study was to determine the major sources of the differences of fiber counts between two objective lens types by performing two sets of RR studies.

Methods: For the first RR study, 15 proficiency test sample filters (five each of chrysotile and Amosite generated by water-suspended method and five chrysotile generated by aerosolization method) were purchased from the American Industrial Hygiene Association (AIHA) and slides were prepared with relocatable cover slips. Nine labs volunteered to participate in the exercise. A single microscope with both standard and DM objectives was circulated to each laboratory along with the prepared slides. A second RR study was then performed with six chrysotile field sample slides. Six out of nine labs who participated in the first RR study participated. Additionally, an 8-form diatom test slide was examined by eight analysts to compare resolutions between two objectives.

Results: For the PAT chrysotile reference slides, use of the DM objective resulted in consistently higher fiber counts (1.45 times for all data) than the standard objective (p-values: < 0.05), regardless of the filter generation (water-suspension or aerosol) method. For the PAT Amosite and chrysotile field sample slides, the fiber counts between the two objectives were not significantly different. There was no differences of resolution from the 8-form diatom test slide examination.

Conclusions: The findings support the contrast caused by the different phase plate absorption between the two objectives as the main factor affecting high number of PAT chrysotile fiber counts using the DM objective. The chrysotile fibers in the PAT samples are thinner than the airborne contaminant fibers collected from the field. If the thin PAT fibers are not generally representative of field samples, it is not necessary to recommend the DM objective be used for routine fiber-counting. However, the DM objective does allow more very fine fibers to be counted and may provide counts of those fine fibers, if present, similar to that achievable with an electron microscope.

SR-404-06 Development of Analytical Methods as a Tool for Checking Compliance with Newly Established OELs in Poland

J. Gromiec, S. Brzeznicki, M. Bonczarowska, A. Wziatek, Nofer Institute of Occupational Medicine, Lodz, Poland

Objective: In Poland Occupational Exposure Limits are called Maximum Admissible Concentrations (MAC) and are published by the Minister of Labour and Social Policy (based on documented proposals from the Interdepartmental Commission for MACs for Agents Harmful to Health in the Work Environment) and are legally binding. There is a practice that no MAC values are published unless the appropriate analytical method is available enabling determination of compliance. The objective of the project was to develop analytical methods for 2-ethylhexyl acrylate, methoxyaniline, cyclopentane and diphenylamine introduced to the MAC list in 2013–2014.

Methods: Parameters of air sample collection (sorbent selection, air volume, sampling rate, desorption efficiency) and analytical conditions were investigated. Validation parameters, required according to the European standard EN 482:2012 such as limit of detection, analytical range, precision, specificity and expanded uncertainty were also determined. Gas chromatography and high performance liquid chromatography were used in analysis of the compounds.

Results: Four analytical methods, meeting the criteria of EN 482:2012 were elaborated. The limits of determination are, in mg/m3, 3 for 2-ethylhexyl acrylate, 0.025 for isomers of methoxyaniline, 30 for cyclopentane and 0.4 for diphenylamine. The expanded uncertainty of all the methods is much better than ± 30%.

Conclusions: The developed methods were presented in the form analytical procedures meeting the requirements of the Polish Standardization Committee for the national standards. The methods, as specific and selective in presence of other substances expected in the same work environment enable evaluation of occupational exposure to the compounds newly introduced to the MAC list.

SR-404-07 The Second Report of Studies of Passive Sampler Equipped with FAN Motor for Indoor Sampling of Nicotine

Y. Suzuki, T. Enomoto, T. Enomoto, SIBATA Science Technology Ltd., Saitama, Japan; M. Noguchi, Seikei University, Musashino, Japan; Y. Yanagisawa, Tokyo Universtiy, Bunkyo ku, Japan

Objective: We examined whether or not, the diffusive sampler system for measurement of Nicotine in the air, will work efficiently. For passive sampling, Nicotine is one of objectives that substance measurements are difficult. This is because the Nicotine situation turns into gaseous, particulate, air oxidation, UV degradation, water absorption, dimer formation and so forth. It is also because this material causes adsorption (such as particle adhesion, adhesion to glass). Our study is to solve these problems.

Methods: In this study, we used the Semi-Active Sampler that we had reported in AIHce2014. At Nicotine trapping, we adopted a method of adsorbing to a PUF (Poly-Urethane Form). Herewith we carried out efficient collection by increasing the equilibrium velocity. The PUF can collect the Nicotine in a stable situation. Semi-Active Sampler is superior in serenity compared with the active collection. In recovery rate measurement of Nicotine, it was confirmed by the sampling performance. After adding Nicotine to the PUF directly, it is operated in clean air for a certain time, and the recovery rate superiority was confirmed by GC / MS (EI) with a solvent extraction adsorbent. The sampling was carried out in parallel with this method and with XAD-4 active sampling method in the smoking location.

Results: The spike recovery rates showed at about the percentages of 55. In 24-hour collection, the results were higher values even in comparison with the active method (XAD-4 collected). For the pump method, Semi-Active Sampler showed good correlation in Nicotine sampling results. However, a fluctuation was observed in the measured value in such state where smoke flux was drawn significantly. In addition, the results with this present measurement method showed a good correlation with those with a measurement of dust meter of light scattering method.

Conclusions: Nicotine Collection, using the PUF adsorbent, with Semi-Active Sampler turned out possible. However, we observed the fluctuation of a significantly higher value than the expectation in non-mixed state (such as near a smoking area) and we also observed the necessity of further consideration in the future. We would like to consider such substance as SVOC in the future.

CS-404-08 Using Gas Chromatography Mass Spectrometry in Single Ion Monitoring Mode for Analysis of Diacetyl and 2,3-Pentanedione

M. Simmons, USDOL/OSHA, Sandy, UT

Situation/Problem: OSHA is currently using OSHA Method 1013 for the sampling and analysis of diacetyl and OSHA Method 1016 for 2,3-pentanedione. Both methods use gas chromatography with flame ionization detection (GC/FID) for analysis. OSHA is also using OHSA Method 1012 for the sampling and analysis of diacetyl, with analysis by gas chromatography with electron capture detection (GC/ECD) after derivatization with O-pentafluorobenzyl hydroxylamine hydrochloride. All three methods use two 600 mg silica gel sorbent tubes in series, have recommended sampling times of 3 hours (9 L), and use ethanol for sample extraction. The advantage of methods 1013 and 1016 is sample preparation can be performed in one hour, but the reliable quantitation limit (RQL) is 41.0 µg/m3 for diacetyl and 38.0 µg/m3 for 2,3-pentanedione. The advantage of Method 1012 is that the RQL is 4.6 µg/m3 for diacetyl, but the derivatization procedure requires 36 hours with multiple derivatives of diacetyl (isomers) formed.

Resolution: Using OSHA “Validation Guidelines for Air Sampling Methods Utilizing Chromatographic Analysis”, the sample preparation procedure described in OSHA Method 1013 and 1016, and gas chromatography mass spectrometry in single ion monitoring mode (GC/MS-SIM), new RQLs were determined for diacetyl and 2,3-pentanedione. 

Results: A RQL of 11 µg/m3 was calculated for both diacetyl and 2,3-pentanedione using GC/MS-SIM. 

Lessons Learned: The use of GC/MS-SIM improved the RQL for both diacetyl and 2,3-pentanedione as compared to the GC/FID methods, and brought the diacetyl RQL closer to the reported value of the GC/ECD method while avoiding the time consuming derivatization step. 

SR-404-09 Validation of a Diffusive Sampler for Monitoring Siloxanes in Air

J. Elder, SKC, Inc., Eighty Four, PA

Objective: Siloxanes are a family of organic compounds that are used widely in the production of silicone polymers and other organosilicone substances, electronics, textiles, and personal care products. Siloxanes are found commonly in landfills where they can be volatilized easily. Some toxicology studies report effects on the liver, kidney, and female reproductive tract. Studies to date have not demonstrated if this effect occurs through a pathway that is relevant to humans. The objective of this study was to determine if the 575-001 Passive Sampler is suitable for monitoring octamethylcyclotetrasiloxane (D4) in air.

Methods: The method used to collect D4 was a 575-001 Passive Sampler filled with activated carbon and analysis was by GC-FID. The test concentration was 20 ppm for D4, at a relative humidity of 80% and a temperature of 30 C. Desorption efficiency studies were performed for D4 at 0.1 to 2.0 times the WEEL (Workplace Environmental Exposure Level) limit of 10 ppm. A storage study was conducted for 2 weeks at room temperature (22 C) and refrigerated temperature (< 4 C).

Results: The results showed that the mean sampling rate for D4 was 7.16 ml/min with an 8.38% Relative Standard Deviation (%RSD). The desorption efficiency (DE) was 87.2% with a 4.96% RSD. The storage study showed that the sample could be stored for up to 2 weeks at either ambient or freezer temperatures with less than a 10% loss in recovery.

Conclusions: In conclusion, the data indicate that diffusive sampling is a viable method for sampling D4 in workplace air.​