Bronek Z. Drozdowicz, Ph.D.
Air Products and Chemicals, Inc.
7201 Hamilton Boulevard, Allentown, PA 18195-1501

BIOGRAPHY
Bronek Z. Drozdowicz is the Manager of Toxicology and Risk Control at Air Products and Chemicals, Inc. He is a graduate of the Warsaw University (Warsaw, Poland) and received his Ph.D. in Genetics from Cornell University (Ithaca, NY). He has conducted post-doctoral research in mutagenesis, somatic cell genetics, DNA repair, and carcinogenesis at the Smithsonian Institution, the Johns Hopkins University School of Hygiene and Public Health, and the Case Western Reserve University. In his current position, he is responsible for the toxicological evaluations and risk assessments of Air Products' raw materials, processes, and products. He serves on his company's Product Stewardship committee and provides regulatory guidance in the areas of TSCA and FDA compliance.

Dr. Drozdowicz is a member of the Genetic Toxicology Association, the Environmental Mutagen Society, the American College of Toxicology, and the New York Academy of Sciences. He also serves on the technical and regulatory committees of the Chemical Manufacturers Association, the Society of Plastics Industries and the American Industrial Health Council.

ABSTRACT
Nitrogen trifluoride (NF₃) is used in the electronic industry as a cleaning and etching agent. It is relatively inert at ambient temperatures but will dissociate at temperatures greater than 300°C into fluorine species that react with most materials. It is a nonflammable, colorless, and odorless gas. Its LC50 in rats is 6700 ppm/hr. Its ACGIH TLV is set at 10 ppm. The immediate effect of acute overexposure is methemoglobin formation. Unpublished results of in vitro genotoxicity testing of NF₃ samples of various origins in the USA (Air Products) and Japan (Japanese Bioassay Center) indicated a possibility of NF₃ having mutagenic properties. We have investigated the genotoxic potential of NF₃ using the current high-purity commercial product and current testing protocols. NF₃ was examined for mutagenic activity in the Ames' Salmonella typhimurium and Escherichia coli assays. The tests were performed with Salmonella strains TA1535, TA98, TA100, and TA102 and E. coli strain WP2 (uvrA) both in the presence and absence of an Aroclor 1254-induced rat-liver metabolic activation system. NF₃ was found not to be mutagenic under the test conditions used. NF₃ was examined for clastogenic activity in the in vivo bone marrow micronucleus test in Swiss-Webster mice and NF₃ was found not to be clastogenic. In conclusion, NF₃ was found not to be genotoxic using in vitro and in vivo test systems.

DATA
High-purity Air Products commercial NF₃ (Nitrogen Trifluoride NG Grade (see Table 1 for analytical details) was tested for mutagenic potential in the modified and expanded microbial Ames Test and for chromosomal aberrations induction (clastogenic) potential in the Mouse Micronucleus Assay. A preliminary cytotoxicity assay was performed in preparation for the mouse lymphoma mutation assay. No cytotoxicity was observed and no further testing was performed.

I. MICROBIAL MUTAGENICITY
Since the standard plate test is not suitable for the testing of gaseous materials, a modified procedure was used with special exposure chamber to deliver the gas to the bacterial tester strains [Salmonella TA1535, TA98, TA100, TA102 and E. coli WP2 (uvrA)]. NF₃ was tested (with and without Aroclor-1254-induced rat liver metabolic activation system) at 0.1, 0.25, 0.5, 1.0, 2.0, and 5.0 molar percent dose levels. Details of the experimental results are shown in Table 2. No statistically significant increases in the number of revertants and no cytotoxicity was observed. Positive and negative controls were within acceptable historical limits. The high-purity NF₃ was not mutagenic under the conditions of these bioassays.

II. MOUSE INHALATION MICRONUCLEUS TEST
The animals were exposed to the test material in nose-only, "flow-past" exposure units. The test doses for the definitive assay were chosen from the results of a range-finding assay in which animals were exposed to 217, 394, 845, or 1850 ppm of NF₃ for a single 4-hr exposure and sacrificed 72 hr after exposure. In the definitive assay five mice per sex per treatment level were exposed to test article concentrations of 840, 1274, or 2469 ppm. A negative control group was exposed similarly and concurrently to cylinder air. A group of 15 male mice was treated by gavage with 300 mg/kg of urethane in water (positive control) on the same day as the inhalation exposure. Five mice of each sex were sacrificed 24, 48, or 72 hours after exposure.

Bone marrow smears were then prepared from each mouse. The number of micronucleated polychromatic erythrocytes (PCEs) in at least 1000 PCEs per animal, which provides an index of chromosomal damage, and the number of PCEs in at least 200 erythrocytes per animal, which provides an index of cytotoxicity, were analyzed. The details of experimental results are shown in Tables 3 and 4. A statistically significant increase in micronucleated PCE was observed in male mice at 72 hr after exposure to 840 ppm of nitrogen trifluoride. However, this effect was not dose-related and was not reproducible (i.e., no increase in micronucleated PCE was observed in slides from the range-finding assay). No significant increase in micronucleated PCE was found in any other dose groups. Therefore the high-purity nitrogen trifluoride was determined to be non-clastogenic in the mouse bone marrow micronucleus assay based on the criteria established in the protocol.