Iodotrifluoromethane: Toxicity Review (2004)

L. A. Chaney, Ph.D.,

Consultant

Toxicology Directorate

U.S. Army Center for Health Promotion and Preventive Medicine

Aberdeen Proving Ground, MD 21010-5403

Tel: (410) 436-3980 Fax: (410) 436-6710

leslie.chaney@apg.amedd.army.mil

PURPOSE: In 1987, 23 countries, including the United States, signed an agreement that would reduce the production of ozone depleting substances (ODS). Amendments to this agreement, called the “Montreal Protocol on Substances that Deplete the Ozone Layer,” placed controls on the production and consumption of ozone depleting materials, including the fire suppressants Halon 1211 and Halon 1301. These compounds are effective and have acceptable risk when used correctly, but have been identified as ozone depleting substances. The restrictions set forth in the Montreal Protocol forced a search for suitable replacements for Halon 1211 and Halon 1301, which are effective, safe, and environmentally acceptable. A number of candidate replacement agents for Halon 1301 that have been tested for efficacy and safety are currently in use. Iodotrifluoromethane (CF₃I) (a.k.a. iodotrifluoromethane, trifluoroiodomethane, trifluoromethyl iodide) is another candidate replacement agent currently being considered.

The USACHPPM has been supporting the search for a non-ozone depleting fire extinguishing agent to replace Halon 1301. A request for a Toxicology Profile for CF₃I was submitted by the Army Acquisition

Pollution Prevention Support Office of the Army Materiel Command in 1993. Since no toxicity data for CF₃I were available at that time, a battery of tests were recommended to characterize toxicity. In 1999, CHPPM published a Toxicity Review of CF₃I that presented a critical discussion of much of the new data. In this 2002 update, the current status of CF₃I is considered, particularly in regard to defining exposure levels that would be considered acceptable for military use of the agent.

CONCLUSIONS: Overall, the toxicity of CF₃I is relatively low. Available data indicate a potential health hazard exists in the area of cardiac sensitization following acute inhalation exposure to concentrations of CF₃I greater than 0.2%. The effect of CF₃I on mutagenicity and reproductive parameters is equivocal and may warrant further investigation. Human exposure to CF₃I could occur during the manufacturing, transportation, storage, or packaging processes. Accidental releases are also potential sources of exposure in the military setting.

USACHPPM will not endorse the NFPA Standard 2001 (2000) recommendations for “safe” exposure limits to CF₃I because these levels were determined using PBPK modeling data based on a LOAEL (0.4%) for cardiac sensitization in the dog that resulted in death of the animal.

RECOMMENDATIONS: Any proposed use of CF₃I in army systems at design concentrations greater than 0.2% must conform to EPA Significant New Alternatives Policy (SNAP) guidelines which accept CF₃I as a substitute for Halon 1301 in normally unoccupied areas only (Federal Register, 1995). Based on this ruling, any employee that could possibly be in the area must be able to escape within 30 seconds, and the employer must ensure that no unprotected employees enter the area during agent discharge.

In 1987, 23 countries, including the United States, signed an agreement that would reduce the production of ozone depleting substances (ODS). Amendments to this agreement, called the “Montreal Protocol on Substances that Deplete the Ozone Layer”, placed controls on the production and consumption of ozone depleting materials, including the fire suppressants Halon 1211 and Halon 1301. These compounds are effective and have acceptable risk when used correctly, but have been identified as

ozone depleting substances. The restrictions set forth in the Montreal Protocol forced a search for suitable replacements for Halon 1211 and Halon 1301, which are effective, safe, and environmentally acceptable. A number of candidate replacement agents for Halon 1301 that have been tested for efficacy and safety are currently in use. Iodotrifluoromethane (CF₃I) (a.k.a. iodotrifluoromethane, trifluoroiodomethane, trifluoromethyl iodide) is another candidate replacement agent currently being considered.

The USACHPPM has been supporting the search for a non-ozone depleting fire extinguishing agent to replace Halon 1301. A request for a Toxicology Profile for CF₃I was submitted by the Army Acquisition Pollution Prevention Support Office of the Army Materiel Command in 1993. Since no toxicity data for CF₃I were available at that time, a battery of tests were recommended to characterize toxicity. In 1999, CHPPM published a Toxicity Review of CF₃I that presented a critical discussion of much of the new data (McCain and Macko, 1999). In this 2002 update, the current status of CF₃I is considered, particularly in regard to defining exposure levels that would be considered acceptable for military use of the agent.

The physical properties of CF₃I are summarized in Table B-1. CF₃I is a gas at room temperature with a relatively high boiling point of 2.5°C, a melting point of 110°C, and a vapor pressure of 78.4 psia. CF3_I also has a C-I dissociation energy of 54 kcal/mol, indicative of a compound that can readily disassociate (Moore et al, 1994; NFPA, 1996). Exposure to CF₃I in the workplace is most likely to occur through inhalation.

There is evidence that CF₃I photolyzes in the presence of sunlight and common fluorescent lights, resulting in an atmospheric half-life of 1.15 days. Potential degradation products following release in a well-lighted area would include low concentrations of highly toxic carbonyl fluoride (COF₂), and hydrogen fluoride (HF) (Nyden, 1995). These compounds would be produced to a greater degree during fire suppression, but CF₃I would be expected to produce less HF than other candidate Halon replacement agents like HFC-125, HFC-227ea, or FC-218 (Gann, 1995).

Long-term stability testing indicates that CF₃I would degrade more rapidly in the presence of moisture, copper, and at temperatures above 100°C (Harris, 1995). Yamamoto et al. (1997) indicated that fluorinated compounds containing iodine or bromine atoms decomposed easier than perfluoridated compounds. It is unknown how product degradation will

affect toxicity. No attempt to identify or evaluate the toxicity of degradation products has been performed.

The proposed design concentration for CF₃I use in most systems exceeds the lowest observed adverse effect (LOAEL) for cardiac sensitization of 0.4%. The EPA published a final rule under its Significant New Alternatives Policy (SNAP) program to accept CF₃I as a substitute for Halon 1301 in normally unoccupied areas only (Federal Register, 1995). Based on this ruling, any employee that could possibly be in the area must be able to escape within 30 seconds, and the employer must ensure that no unprotected employees enter the area during agent discharge.

The EPA also published a final rule accepting CF₃I as a substitute for

Halon 1211 in nonresidential applications only (Federal Register, 1997). Because cardiac sensitization has been demonstrated at relatively low concentrations of CF₃I, the EPA prohibits use of this agent in consumer residential applications where the possibility exists of incorrect use by untrained individuals.

As of April 1, 2002, EPA removed restrictions previously imposed on the use of certain halon alternatives under the SNAP program. EPA rescinded use conditions imposed under SNAP that limit human exposure to halocarbon and inert gas agents used in the fire suppression and explosion protection industry. EPA considers these use conditions to be redundant with the safety standards outlined in the NFPA 2001 Standard. Currently, the EPA SNAP program recommends that use of CF₃I and several other halocarbon agents should be in accordance with the safety guidelines in the latest edition of NFPA Standard 2001 (Federal Register, 2002).

The NFPA 2001 Standard on Clean Agent Fire Extinguishing Systems (1996; 2000) is a guidance document prepared by the Technical Committee on Halon Alternative Protection Options to address the need for information regarding the design, installation, maintenance, and operation of systems using clean agent fire extinguishants. In the most recent edition (2000), NFPA endorses the use of physiologically-based pharmacokinetic (PBPK) modeling procedures to recommend “safe” exposure limits.

According to the NFPA 2001 Standard (2000), Section 1-6.1.2.1 (c) states, “In spaces that are not normally occupied and protected by a halocarbon system designed to concentrations above the LOAEL…, and where personnel could possibly be exposed, means shall be provided to limit exposure times using Tables 1-6.1.2.1(b) through 1-6.1.2.1(e).” The relevant table for CF₃I (see Table B-2 adapted from Table 1-6.1.2.1(e)) describes human exposure times that would be considered “safe” for exposure to increasing concentrations of CF₃I, based on estimates derived from PBPK modeling and LOAEL values established during cardiac sensitization testing in a dog model:

Based on the NFPA 2001 (2000) guidelines, it would be considered safe for a human to be exposed to levels of CF₃I above the NOAEL (0.2%) and up to 0.3% for as long as 5 minutes. At concentrations greater than 0.3%, the time for “safe” exposure decreases, but exposure is still allowed.

The Army does not have a separate policy regarding ozone-depleting substances. Health and safety issues are addressed in Army Regulation 40-5 (AR 40-5: Preventive Medicine) (1990). One of the Preventive Medicine functional areas of AR 40-5 is the Health Hazard Assessment Program (AR 40-10: Health Hazard Assessment Program in Support of the Army Materiel

Acquisition Process) (1991). The primary objective of this regulation is to identify and eliminate or control health hazards associated with the life cycle management of weapons, equipment, clothing, training devices, and materiel systems. One objective of this program is to preserve and protect the health of the individual soldier and other personnel. Another objective is to reduce the health hazards due to potential environmental contamination associated with the use of Army systems. This objective is protective of the stratospheric ozone and complies with all federal regulations and guidelines. The Army is in the process of revising both AR 40-5 and AR 40-10.

The USACHPPM considers all available standards and guidelines when evaluating agents proposed for use in an Army system. In the interest of its primary responsibility, to protect US Army personnel from exposure to potentially hazardous substances, USACHPPM has traditionally adopted a conservative approach when evaluating and approving such agents.

The minimum design concentration for a gaseous agent is determined by the ISO Cup Burner Test. The concentration of Halon 1301 necessary to extinguish a n-heptane fire by this test method is 3.3 vol%. The “best value” for CF₃I as determined by the National Fire Protection Association

(NFPA) Cup Burner Data Task Group is 3.2 vol% (Tapscott, 1999). Therefore, for n-heptane, the design concentration for CF₃I will be slightly lower than that for Halon 1301 regardless of the applied safety factor. The NFPA 2001 Standard (1996) requires a minimum 20% safety factor above the cup burner values with a minimum design concentration of 5.0% for Halon 1301. This safety margin was chosen as a requirement for extinguishment of Class A fires. According to Meyer (1997), the extinguishing concentration of CF₃I is almost half of the concentration needed by any other gaseous agent under consideration. In a turbulent spray burner test, CF₃I required the lowest mass fraction at extinction of any compound tested (Hamins, 1997).

The U.S. Army Environmental Hygiene Agency prepared a toxicity profile for CF₃I in 1993 (Haight and Macko, 1993). The profile indicated that no toxicity data were available for CF₃I and recommended that a number of toxicity tests be conducted in order to fully evaluate its safety. These suggestions included a skin and eye irritation test, acute and 14-day inhalation studies, genotoxicity testing, cardiac sensitization testing, and a full evaluation of combustion, pyrolysis and decomposition products. Comprehensive tests, such as reproductive and developmental toxicity and subchronic inhalation, were also suggested if projected use scenarios indicated a need. Many of these tests have since been conducted and data were incorporated into the comprehensive review of CF₃I released in 1999 by McCain and Macko. A summary of the toxicity studies performed for CF₃I is provided in Table B-3, and a brief overview of the results is provided below.

1. 15-Minute Acute Exposure. Several acute inhalation studies have been conducted using CF₃I. Ledbetter (1993) exposed 5 male and 5 female Sprague-Dawley rats to CF₃I vapor in a nose-only inhalation chamber for 15 minutes. The intended target concentration was 60,000 ppm (6.0%). Due to an error in the wavelength setting for the infrared monitoring system, the actual measured concentration during exposure was 127,289 ± 5,574, nearly

twice the targeted dose. No deaths were reported during the study. Immediately following exposure, severe salivation was noted for all exposed rats, and 2 males exhibited audible respiration (rales). All clinical signs resolved within 1 hour after exposure was discontinued. No other clinical signs or changes in body weight were reported during the 14-day post-exposure observation period. No gross abnormalities were reported at necropsy.

In a separate study, 5 male and 5 female rats were exposed to CF₃I vapor in a nose-only inhalation chamber at concentrations of 28.8% or 24% for 15 minutes (Ledbetter, 1994). Following exposure to 28.8%, 5 females and 2 males died. Necropsy findings indicated red lungs in 2 males and 1 female. One male rat died following exposure to 24%. Two male rats had hemorrhagic foci in the lungs and one had red lungs. All other organs were normal. The median lethal concentration (LC₅₀) following 15-minute exposure to CF₃I was determined using only two exposure levels and was estimated to be 27.4%.

2. 4-Hour Acute Exposure. A 4-hour whole-body exposure study was conducted using CF₃I at concentrations of 32%, 20%, 12.8%, and 10% (Ledbetter, 1994). Five males and five females were exposed at each concentration. All rats exposed to 32% became unconscious and died within 20 minutes of exposure. Necropsy findings indicated that the lungs in this exposure group were dark red and puffy. It was determined that the gas for the 32% group was contaminated with hydrogen fluoride (HF). A second group of animals was exposed at 20% CF₃I with a KOH scrubber system in place to remove HF prior to entry of test material into the chamber. Again, all exposed rats became unconscious and died after approximately 20 minutes of exposure. Upon necropsy, lungs from this group were puffy but much less red. No HF was detected in the test atmosphere. Remaining exposures were conducted using new CF₃I test material containing no detectable HF. No deaths were observed in the 12.8% or 10% exposure groups, although all rats became unconscious to semi-unconscious after approximately 30 minutes of exposure. All animals regained consciousness immediately (within 1-3 minutes) after exposure was discontinued. No other clinical signs were noted and rats exhibited normal weight gain during the 14-day observation period. Upon necropsy, the lungs of 2 male rats exposed to12.8% CF₃I were red. All other organs appeared normal.

In another study, male Fisher 344 rats (30/group) were exposed to CF₃I in a nose-only chamber for 4 hours at 0.0%, 0.5%, or 1.0% (Kinkead et al., 1994). No deaths occurred during exposure. Ten animals per group were sacrificed either immediately, at 3 days, or at 14 days after exposure. Body weights, clinical pathology, including thyroxine and thyroxine binding globulin assays, and histopathology evaluations were performed. No biologically significant findings were noted during the 4-hour exposure or during the 14-day post-exposure observation period.

3. 14-Day Repeated Exposure. A two-week range-finding study was performed using CF₃I at concentrations of 0%, 3%, 6%, and 12% (Kinkead et al., 1995; Dodd et al., 1997a). Five male Fischer 344 rats were exposed (nose-only) at each concentration for 2 hours per day, five days per week

(10 exposures). No deaths were reported. Lethargy and incoordination were observed in the 6% and 12% groups at the end of each daily exposure. A significant decrease in mean body weight gain was noted for rats in the 6% and 12% groups. There was also a 20% decrease in white blood cells of animals exposed in the two highest dosage groups (6% and 12%) and an 8% increase in serum albumin of animals exposed to 12% CF₃I. Elevated levels of serum thyroglobulin and reverse triiodothyronine (rT₃) were observed in all treatment groups. No changes in organ weights or gross lesions were observed. No histopathologic lesions were noted in the thyroid or parathyroid glands following examination of CF₃I exposed rats.

4. 13-Week Subchronic Exposure. A subchronic inhalation (90-day) study of CF₃I was performed in Fisher 344 rats (Kinkead et al., 1996; Dodd et al., 1997a). Fifteen males and females per group were exposed to 0%, 2%, 4%, or 8% CF₃I vapor for 2 hours/day, 5 days/week for 13 weeks in nose-only chambers. Clinical effects, body weights, hematology, bone marrow toxicity/mutagenicity (micronuclei induction), serum chemistry, organ weights, gross pathology and histopathology were evaluated. To investigate potential effects of CF₃I on thyroid function, morphometric image analysis and immunoradiometric assays for serum thyroid hormones were also performed. Five males and females per group were sacrificed after 30 days of exposure. Remaining animals were necropsied after 90 days. Six male rats in the 2% group died during the 9th exposure day, and one on the 13th exposure day. One male from the 8% group was also found dead following the 10th exposure. Remaining males from all study groups were placed into larger nose-only exposure tubes for the remainder of the study. All deaths were attributed to accidental death due to the restraint system. It is unknown if other measured parameters were affected by heat stress due to restraint. Mean body weights were significantly decreased for males and females in the 8% treatment group, and for males only in the 4% group. Hematological analysis showed a slight decrease in red blood cell count in male rats, and decreased total lymphocytes in both males and females in CF₃I treated groups. A statistically significant, dose-dependent increase was noted in micronucleated bone marrow polychromatic erythrocytes (PCE) in all rats exposed to CF₃I as well as a reduction in the PCE/NCE (normochromatic erythrocytes) ratio. Exposure to 8% CF₃I resulted in significant reductions in serum levels of calcium, alanine aminotransaminase (ALT), triglycerides (males only), and triiodothyronine (T₃), and increased levels of thyroglobulin, rT₃, thyroxine (T₄) and thyroid stimulating hormone (TSH). Similar changes in thyroglobulin, T₃, rT₃, T₄ and TSH were also found in males and females in the 4% and 2% groups. Organ to body weight ratios were significantly increased in the 8%

treatment group for brain, liver and thyroid, and significantly decreased for thymus and testes. The decrease in relative weight for thymus and increase for thyroid were also found in the 4% and 2% treatment groups. Biologically significant changes in histopathology included rhinitis, which was noted in all rats exposed to CF₃I concentrations of 4% and 8% after 30 days, but not after 90 days of exposure. A mild increase in thyroid follicular colloid content was observed in all treatment groups. Testicular atrophy with loss of spermatogonia and spermatids, including aspermia, of male rats was observed after 30 days of exposure to 4% and 8% CF₃I. These lesions were also present but less severe after 90 days of exposure. The finding of testicular degeneration was considered equivocal due to the potential heat stress associated with the method of restraint.

5. Cardiac Sensitization Testing. Cardiac sensitization studies for CF₃I (Kenny et al., 1995) were performed using experimental procedures developed by Reinhardt et al. (1971, 1973). Beagle dogs were initially challenged by injecting adrenaline (epinephrine, 0.1 mg/kg/sec) to establish the response of each individual dog to adrenaline alone. The appearance of multifocal ventricular ectopic activity (MVEA), or ventricular fibrillation following exposure indicated a positive response. Dogs were then exposed to CF₃I for 5 minutes and challenged again with adrenaline. For this study, selected CF₃I concentrations were 0.1%, 0.2%, 0.4% and 1.0%. A single dog exposed to CF₃I at a concentration of 1.0% displayed a severe positive response (fatal ventricular fibrillation) and died. A second dog also died following exposure to 0.4% CF₃I. No other animals were tested at these concentrations. Dogs exposed to CF₃I concentrations of 0.1% and 0.2% displayed no dysrhythmia following epinephrine challenge. The lowest observed adverse effect level (LOAEL) for this CF₃I was 0.4% and the no observed adverse effect level (NOAEL) was 0.2%.

6. Genotoxicity Testing. Genetic toxicity testing completed for CF₃I incudes the Salmonella typhimurium histidine reversion assay (Ames Assay), the in vivo mouse bone marrow erythrocyte micronucleus test, and the mouse lymphoma forward mutation assay using L5178Y cells (Mitchell, 1995a; b; c; Dodd et al., 1997b).

The Ames assay used five tester strains of Salmonella typhimurium (TA1535, TA1537, TA1538, TA98, and TA100) at 5 dilutions of CF₃I. Desired concentrations of CF₃I were determined following a range finding study. Actual concentrations of CF₃I achieved for exposures were 1060, 2775, 10586, 23230 and 85908 ppm (0.11%, 0.28%, 1.1%, 2.3% and 8.6%). Assays were conducted using three plates per dose level, in the presence and absence of S-9 metabolic activation. Tester strain TA1538 was not affected by CF₃I. Strains TA1537 and TA98 displayed a weak positive

response both with and without S-9 activation. Strong positive responses were displayed in strains TA100 and TA1535 with and without activation. The results indicate that CF₃I is mutagenic with and without activation, inducing frame-shift and base-pair mutations in Salmonella typhimurium.

For the in vivo mouse bone marrow micronucleus assay, Swiss Webster mice were exposed for 6 hours/day for 3 consecutive days to 2.5%, 5% or 7.5% concentrations of CF₃I. All animals survived and appeared normal during the study. Some treatment-related weight loss was observed in both male and female mice. Positive results were assessed according to criteria set forth by MacGregor, et al. (1988). The ratios of polychromatic erythrocytes (PCE) / 1000 erythrocytes of female mice were significantly decreased with increasing concentrations of CF₃I. This effect was also observed in male mice although one outlier prevented statistical significance. The ratio of micronucleated erythrocytes (MN)/1000 PCEs was significantly elevated in both genders for the 5.0% and 7.5% exposure groups. The results indicate that CF₃I can cause structural chromosomal aberrations in vivo. These data are supported by similar information obtained from Fisher 344 rats used in the 90-day inhalation study (Kinkead et al., 1996; Dodd et al., 1997a).

The forward mutation assay, using L5158Y/tk^+/- mouse lymphoma cells (clone 3.7.2C) was conducted using 5 concentrations of CF₃I (8.0%, 17.7%, 30.6%, 42.6%, 45.4%, 49.7%, and 51.8%). Tests were performed with and without metabolic activation by S-9. Results indicated no evidence of CF₃I-induced mutations of L5158Y/tk^+/- mouse lymphoma cells at any concentration tested.

Free radical modeling has indicated that CF₃I has the characteristics to be carcinogenic (Koski et al., 1997). The model was based on carbon tetrachloride, where it is thought that cellular damage is caused by free radicals produced when an electron is transferred from an enzyme to the carbon tetrachloride molecule. Vertical electron affinities were calculated for a series of halocarbons and suggested that CF₄ would be non-carcinogenic, CF₃Cl was equivocal, and CF₃Br and CF₃I were considered to be potent toxicants expected to be carcinogenic.

7. Reproductive Toxicity. Reproductive toxicity screening was performed in Sprague-Dawley rats by Dodd et al. (1998). These studies were designed to evaluate the effects of CF₃I on parental fertility, maternal pregnancy and lactation, and growth and development of offspring. Four groups of sixteen rats of each gender were exposed to concentrations of 0.0%, 0.2%, 0.7% and 2.0% CF₃I in a whole-body inhalation chamber. Animals were exposed for four weeks at 6 hours/day, 5 days/week prior to mating. During mating, gestation and lactation, rats were exposed for 6

hours/day, 7 days/week. Females were not exposed from gestation day 21 through lactation day 4 to allow for early parturition. Pups were not exposed to CF₃I, and were separated from the dams for 6 hrs/day, 5 days/wk during lactation days 5 through 21. Following the mating period, half of the male rats (8) from each group were sacrificed at 7 weeks. The remaining adult animals (males and females) were sacrificed after 14 weeks. Evaluated endpoints included measurement of body weights, hematology and clinical chemistry, thyroid hormone levels, bone marrow micronuclei, gross necropsy, organ weights and histopathology. Pups were examined at birth for viability and physical abnormalities and were sacrificed at weaning on postnatal day 21 with gross necropsies performed. The results of the study indicated no biologically significant differences in measured body weights, clinical pathology (except for thyroid hormone levels), relative or absolute organ weights, histopathology, bone marrow micronuclei, PCE/NCE ratios, or reproductive endpoints between animals exposed to CF₃I and control animals.

At both 7 and 14 weeks, T₃ levels were reduced and serum TSH, rT₃ and T₄ levels were increased. The observed changes in thyroid hormone levels are similar to those reported previously by Kinkead et al. (1996) in the 13-week subchronic exposure study.

An exposure assessment of CF₃I in handheld fire extinguishers was conducted to determine the exposure of fire fighters during simulated streaming scenarios. Three different room sizes were used in the study (912 ft³, 3822 ft³, and 5133 ft³). In each scenario, the firefighter stood 8 feet from a 1-foot target, and fully discharged the extinguisher. The firefighters discharged 2.5 lb., 5.0 lb., 9.0 lb., and 13 lb. fire extinguishers. Peak concentrations of CF₃I varied from approximately 10,000 ppm (1%) to 30,000 ppm (3%), depending on the height off of the floor, size of the room, and amount of CF₃I discharged. Average concentrations for the first 30 minutes varied from 1040 ppm (0.1%) to 4678 ppm (0.5%) (Skaggs, 1995).

Exposures from intentional release of CF₃I in an F-15 engine nacelle have been estimated (Vinegar et al., 1997; 1999). Portions of the data were obtained from air sampling conducted during a discharge test of an F-15A engine fire-suppression system at the Robbins Air Force Base, GA. The fire suppression bottle was filled with 6.6 pounds of CF₃I and charged with nitrogen at 600 psi. Air sampling for CF₃I concentrations was conducted using the Halonyzer, which provided accurate data for CF₃I concentrations

above 10,000 ppm (1%) and the Triodide analyzer, which accurately measured concentrations below 10,000 ppm. Two Fourier Transform Infrared Spectroscopy (FTIR) analyzers were used to sample extremely low concentrations of CF₃I. The samplers were strategically placed in various locations around the aircraft. Three crew locations appropriate for maintenance activities were identified: 1) kneeling or standing near engine bay, 2) working in or under the engine bay, and 3) prone near the engine bay. Paths of and time to egress were determined for each crew location. Blood concentrations of CF₃I were estimated using PBPK modeling. The estimated blood concentration resulting from a 5-minute exposure to 4000 ppm (0.4%) CF₃I, the LOAEL for cardiac sensitization, was 19 mg/L. Estimated blood concentrations for potentially exposed crewmembers ranged between 6 and 40 mg/L. The highest estimated blood concentration of CF₃I was predicted for individuals who were at head level inside the open engine nacelle. Concentrations of CF₃I in this area were in excess of 70,000 ppm (7%), which resulted in an estimated blood concentration of 40 mg/L. This estimated blood concentration for the “head-at-the-engine” scenario was obtained following the first breath, and remained above the level of cardiac sensitization for more than 30 seconds. Levels of CF₃I under the left wing remained above 4000 ppm for more than five minutes.

An event where two salesmen inhaled CF₃I from balloons as part of their sales demonstration has also been described (Vinegar et al., 1999). The salesmen reportedly inhaled deep breaths of CF₃I on 15 to 17 different occasions without reporting adverse effects. The average volume inhaled was estimated to be 1.25 L, resulting in a simulated peak blood concentration of 2000 mg/L and after five minutes, 71 mg/L. It is not known whether cardiac arrhythmia occurred since the salesmen were not monitored.

CF₃I and several other compounds have been screened as potential replacements for Halon 1301. The review of the available data indicates that adverse health effects could occur following exposure to CF₃I. Potential health hazards appear to exist in the area of cardiac sensitization and genotoxicity.

Since it is reasonable to expect that most exposures would be intermittent and of short duration, acute toxicity information is critical. The LC₅₀ for CF₃I following 15-minute nose-only inhalation has been approximated at 27.4%. This approximation was determined using two concentrations (24% and 28.8%) for 15-minute exposures. Normally, at least three

concentrations are used, and the animals are exposed for 4 to 6 hours. Due to the steep mortality curve, full determination of the LC₅₀ for CF₃I was not completed (Ledbetter, 1994). The LD₅₀ (or LC₅₀) is a somewhat imprecise value traditionally used to compare toxicity among chemicals. Lethality is only one of many parameters used to characterize acute toxicity. The slope of the dose-response curve, time to death, clinical signs, and pathological findings generally contribute more than the LC₅₀ in the evaluation of acute toxicity.

Abnormal cardiac activity, resulting in death, occurred when a single dog was exposed to CF₃I at 1.0% in the presence of epinephrine. Another dog died after exposure to CF₃I at 0.4% in the presence of epinephrine. The cardiac sensitization testing procedure is based on methodology developed by Reinhardt et al. (1971). Although developed for use as a screening test, it has traditionally been accepted by EPA for use as a conservative tool in setting regulatory exposure guidelines to halocarbon agents. The dose of epinephrine used in most cardiac sensitization testing procedures is more than 10 times the level produced by humans, and testing is performed using only one animal at each dose level. Other potential replacement compounds that have been tested using this methodology resulted in only mild to moderate MVEA, and not death (Reinhardt et al., 1971; 1973; Mullin et al., 1972; Trochimowicz et al., 1974; 1976). Cardiac sensitization data, therefore, is paramount in considering the risk associated with the use of CF₃I. A potential health hazard is believed to exist in the area of cardiac sensitization following acute exposure to concentrations of CF₃I greater than 0.2% (NOAEL).

Since the dog model used to measure cardiac sensitization is a conservative assessment of human risk, PBPK modeling can also be used to simulate concentrations of halocarbon agents in human blood following different exposure scenarios that may cause cardiac effects. PBPK modeling is a mathematical description of the uptake and disposition of chemicals based on quantitative interrelationships among critical determinates of these processes (Anderson, 1991). The PBPK model developed to evaluate blood levels of halocarbon agents may be used in some cases to provide extrapolations essential for dose-response assessment of this class of chemicals.

PBPK modeling was used to simulate a blood level of CF₃I for a salesman reportedly inhaling CF₃I from a balloon without adverse effects. This level, 2000 mg/L, was two orders of magnitude greater than that predicted for a response in humans (19 mg/L) based on cardiac sensitization testing in dogs. However, the available information regarding actual human exposure to high levels of CF₃I is anecdotal, at best. In the most recent edition of NFPA Standard 2001 (2000), NFPA endorses the use of PBPK

modeling procedures to recommend “safe” exposure limits to CF₃I. Given that the modeling data is based on the cardiac sensitization LOAEL (0.4%) in the dog that resulted in death of the animal, endorsement of this guideline and recommendation for use of this particular agent by the US military is unlikely.

Mutagenicity was demonstrated in two of the three screening techniques performed using CF₃I. The Ames Salmonella Reverse Mutation assay indicated that CF₃I was a potent mutagen. It induced both frameshift and base-pair mutations in Salmonella typhimurium tester strains with and without activation by mitochondrial S-9. Positive results were also obtained from the mouse bone marrow micronucleus assay, where elevated polychromatic erythrocyte (PCE) to erythrocyte ratios and micronuclei to PCE ratios were observed. These data are supported by similar information obtained from Fisher 344 rats used in the 90-day inhalation study (Kinkead et al., 1996; Dodd et al., 1997), but not by results from the 14-week reproductive toxicity studies performed in Sprague-Dawley rats (Dodd et al., 1998). Overall, the results indicate that CF₃I is capable of causing structural changes in the chromosomes in vivo. Positive results on these screens indicate that a potential for mutagenesis exists and that further testing is warranted. Furthermore, free radical modeling has indicated that CF₃I could potentially be carcinogenic (Koski et al., 1997). Examination of tissues taken from animals exposed to CF₃I in repeated-dose studies, however, has revealed no pre-neoplastic lesions. The ability of CF₃I to induce mutagenesis is considered to be equivocal and may warrant further investigation in additional developmental/reproductive toxicity and carcinogenicity testing.

Results of the 13-week subchronic nose-only inhalation study in Fischer-344 rats indicated a complete absence of sperm as well as a reduction in testicular weight and testicular atrophy in males from the two highest exposure groups (4% and 8%) (Kinkead et al., 1996; Dodd et al., 1997a). This finding was interpreted by the authors to be an effect of restraint resulting in heat stress, and not associated with CF₃I exposure. The fact that testicular changes were reduced at 90 days may support this hypothesis. However, the alterations were only seen in animals of the two highest dosage groups, not control animals. It is possible that CF₃I mediated reproductive effects observed in the Kinkead study may have been potentiated by heat stress. The pathology report indicated potential reproductive toxicity associated with exposure to CF₃I occurred at the high and medium dosage levels and recommended further investigation.

Reproductive toxicity was not demonstrated in a subsequent 14-week whole-body exposure study using male Sprague-Dawley rats (Dodd et al.,

1998). The highest dosage level used was 2%, but the exposure time (6 hours/day) was 3 times greater than that of the Kinkead study (2 hours/day). Strain differences, as well as the different inhalation exposure delivery systems may have contributed to the equivocal findings between these two studies. Although long-term inhalation is not an anticipated exposure scenario for CF₃I, further studies could be performed to clarify reproductive toxicology issues.

Subacute and subchronic exposures to CF₃I resulted in significant changes in thyroid hormone levels (Kinkead et al., 1995; 1996; Dodd et al., 1998). While these effects could be related to exposure to CF₃I, they could also be a result of species differences, with the rat being more susceptible to perturbations in the pituitary-thyroid axis (Capen, 2001; McClain et al., 1988; 1999).

Overall, the toxicity of CF₃I is low. Available data indicates a potential health hazard exists in the area of cardiac sensitization following acute inhalation exposure to concentrations of CF₃I greater than 0.2%. The effect of CF₃I on mutagenicity and reproductive parameters is equivocal and may warrant further investigation. Human exposure to CF₃I could occur during the manufacturing, transportation, storage, or packaging processes. Accidental releases are also potential sources of exposure in the military setting.

USACHPPM will not endorse the NFPA Standard 2001 (2000) recommendations for “safe” exposure limits to CF₃I because these levels were determined using PBPK modeling data based on a LOAEL (0.4%) for cardiac sensitization in the dog that resulted in death of the animal.

Any proposed use of CF₃I in army systems at design concentrations greater than 0.2% must conform to EPA Significant New Alternatives Policy (SNAP) guidelines which accept CF₃I as a substitute for Halon 1301 in normally unoccupied areas only (Federal Register, 1995). Based on this ruling, any employee that could possibly be in the area must be able to escape within 30 seconds, and the employer must ensure that no unprotected employees enter the area during agent discharge.

Andersen, M.E. 1991. Physiological modeling of organic chemicals. Ann. Occup. Med. 35:305-321.

Capen, C.C. 2001. Toxic responses of the endocrine system. In Casarett and Doull’s Toxicology: Basic Science of Poisons. Klaassen, C.D. (ed.). McGraw-Hill, New York, NY.

Dodd, D.E., E.R. Kinkead, R.E. Wolfe, H.F. Leahy, J.H. English, and A. Vinegar. 1997a. Acute and subchronic inhalation studies on trifluoroiodomethane vapor in Fischer 344 rats. Fundam. Appl. Toxicol. 35:64-77.

Dodd, D.E., A.D. Ledbetter, and A.D. Mitchell. 1997b Genotoxicity testing of the halon replacement candidates trifluroiodomethane (CF₃I) and 1,1,1,2,3,3,3-heptafluoropropane (HFC-227ea) using the Salmonella typhimurium and L5178Y mouse lymphoma mutation assays and the mouse micronucleus test. Inhal. Toxicol. 9:111-131.

Dodd, D.E., H.F. Lehay, M.L. Feldmann, A. Vinegar, and J.H. English. 1998. Reproductive toxicity screen of trifluoroiodomethane (CF₃I) in Sprague-Dawley rats. Army Medical Research Unit, Report # AFRL-HE-WP-TR-1998-0019.

Federal Register. 1995. Protection of the stratospheric ozone: Final Rule. Vol. 60, No. 113, pp. 31092-31107, 13 June 1995.

Federal Register. 1997. Protection of the stratospheric ozone: Listing of substitutes for ozone-depleting substances. Vol. 61, No 100, pp. 25585-25594, 22 May 1997.

Federal Register. 2002. Protection of the stratospheric ozone: Removal of restrictions on certain fire suppression substitutes for ozone-depleting substances; and listing of substitutes. Vol. 67, No 19, pp. 4185-4203, 29 January 2002.

Gann, R.G. 1995. Executive Summary. In Fire Suppression System Performance of Alternative Agents in Aircraft Engine and Dry Bay Laboratory Simulations. Gann, R.G. (ed.). National Institute of Standards and Technology SP 890, Vol. 1, iii-vi.

Haight, E.A. and J.A. Macko. 1993. Toxicity Profile: Iodotrifluoromethane. U.S. Army Environmental Hygiene Agency, Aberdeen Proving Ground, MD.

Hamins, A. 1997. Flame suppression by halon alternatives. In Thirteenth Meeting of the UJNR Panel on Fire Research and Safety, March 13-20, 1996. Beall, K.A. (ed.). Building and Fire Research Laboratory, National Institute of Standards and Technology, Technology Administration, U.S. Department of Commence. NISTR 6030.

Harris, R.H. 1995. Agent stability under storage. In Fire Suppression System Performance of Alternative Agents in Aircraft Engine and Dry Bay Laboratory Simulations. Gann, R.G. (ed.). Vol. 1, Section 7, National Institute of Standards and Technology SP 890, pp 249-398.

Kenny, T.J., C.K. Shepherd, and C.J. Handy. 1995. Iodotrifluoromethane and

iodoheptafluoropropane assessment of cardiac sensitization potential in dogs. Huntington Research Center, Ltd., Report # AL/OE-TR-1995-0031.

Kinkead, E.R., S.A. Salins, R.E. Wolf, H.F. Leahy, and J.H. English. 1994. Acute toxicity evaluation of halon replacement, trifluoroiodomethane (CF₃I). Report # AL/OE-TR-1994-0070.

Kinkead, E.R., R.E. Wolfe, M.L. Feldmann, H.F. Leahy, and J.H. English. 1995. Two-week range finding, nose-only inhalation toxicity study of trifluoroiodomethane (CF₃I). In Toxic Hazards Research Unit Annual Report, Dodd, D.E. (ed.), Report # AL/OE-TR-1995-0135.

Kinkead, E.R., R.E. Wolfe, M.L. Feldmann, H.F. Leahy, and C.D. Flemming. 1996. Ninety-day nose-only inhalation toxicity study of trifluoroidomethane (CF₃I) in male and female Fischer 344 rats. Report # AL/OE-TR-1996-0024.

Koski, W.S., S. Roszak, J.J. Kaufman, and K. Balasubramanian. 1997. Potential toxicity of CF₃X halocarbons. In Vitro Toxicol. 10:455-457.

Ledbetter, A. 1993. Acute inhalation toxicity of iodotrifluoromethane in rats: Study Number 1. Final Report. Project 6030-012, Mantech Environmental Technology, Research Triangle Park, NC.

Ledbetter, A. 1994. Acute inhalation toxicity of iodotrifluoromethane in rats. Final Report. Project 1530-001, Mantech Environmental Technology, Research Triangle Park, NC.

MacGregor, J.T., J.A. Heddle, M. Hite, B.H. Margolin, C. Ramel, M.F. Salamone, R.R. Tice, and D. Wild. 1988. Guidelines for the conduct of micronucleus assays in mammalian bone marrow erythrocytes. Mut. Res. 189:103-112.

McCain, W.C., and J. Macko. 1999. Toxicity review for iodotrifluoromethane (CF₃I). Project No. 85-8823-99. U.S. Army Center for Health Promotion and Preventive Medicine, Aberdeen Proving Ground, MD.

McClain, R.M., A.A. Levin, R. Posch, and J.C. Downing. 1989. The effects of phenobarbital on the metabolism and secretion of thyroxine in rats. Toxicol. Appl. Pharmacol. 99:216-228.

McClain, R.M., Posch, R.C., Bosakowski, T., and Armstrong, J.M. (1988). Studies on the mode of action of thyroid gland tumor production in rats by phenobarbital. Toxicol. Appl. Pharmacol. 94:254-265.

Meyer, D. 1997. Triodide–New gaseous extinguishing agent. Fire. May edition.

Mitchell, A.D. 1995a. Genetic toxicity evaluation of iodotrifluoromethane (CF₃I). Volume I of III: Results of the Salmonella typhimurium histadine reversion assay (Ames assay). Genesys Research, Inc., Report # AL/OE-TR-1995-0009.

Mitchell, A.D. 1995b. Genetic toxicity evaluation of iodotrifluoromethane (CF₃I). Volume II of III: Results of in vivo mouse bone marrow erythrocyte micronucleus testing. Genesys Research, Inc., Report # AL/OE-TR-1995-0009.

Mitchell, A.D. 1995c. Genetic toxicity evaluation of iodotrifluoromethane (CF₃I): Volume III of III: Results of the forward mutation assay using L5178Y mouse lymphoma cells. Genesys Research, Inc., Report # AL/OE-TR-1995-0009.

Moore, T.A., S.R. Scaggs, M.R. Corbitt, R.E. Tapscott, D.S. Dierdorf, and C.J. Kibert. 1994. The development of CF₃I as a Halon replacement. NMERI Report 1994/40/31790.

Mullin, L.S., A. Azar, C.F. Reinhardt, P.E. Smith, and E.F Fabryka. 1972. Halogenated hydrocarbon-induced cardiac arrhythmias associated with release of endogenous epinephrine. Am. Ind. Hygiene J. 33:389-396.

National Fire Protection Association (NFPA). 1996. NFPA 2001 Standard on Clean Agent Fire Extinguishing Systems. National Fire Protection Association, Quincy, MA.

National Fire Protection Association (NFPA). 2000. NFPA 2001 Standard on Clean Agent Fire Extinguishing Systems. National Fire Protection Association, Quincy, MA.

Nyden, M.R. 1995. Chapter 3: Photodegradation of CF₃I. In Fire Suppression System Performance of Alternative Agents in Aircraft Engine and Dry Bay Laboratory Simulations. Gann, R.G. (ed.). National Institute of Standards and Technology SP 890, Vol. 1, pp77-95.

Reinhardt, C.F., A. Azar, M.E. Maxfield, P.E. Smith, and L.S. Mullen. 1971. Cardiac arrhythmias and aerosol “sniffing”. Arch. Environ. Health. 22:265-279.

Reinhardt, C.F., L.S. Mullen, and M.E. Maxfield. 1973. Epinephrine induced cardiac arrhythmia potential of some common industrial solvents. J. Occup. Med. 15:953-955.

Skaggs, S.R., and Cecil, M.A. 1995. Pp. 561-569 in Exposure Assessment to Triodide in Streaming Scenarios. Proceedings of International CFC and Halon Alternatives Conference, Washington, DC.

Tapscott, R.E., 1999. Best values of cup burner extinguishing concentrations. Proceedings, Halon Options Technical Working Conference, Albuquerque, NM, 27-29 April, 1999.

Trochimowicz, H.J., A. Azar, J.B. Terrill, and L.S. Mullen. 1974. Blood levels of fluorocarbon related to cardiac sensitization: Part II. Am. Ind. Hyg. J. 34:632-639.

Trochimowicz, H.J., C.F. Reinhardt, L.S. Mullen, A. Azar, and B.W. Karrh. 1976. The effect of myocardial infarction on the cardiac sensitization potential of certain halocarbons. J. Occup. Med. 18:26-30.

U.S. Army. 1990. Army Regulation 40-5: Preventive Medicine. Headquarters, Department of the Army, Washington, DC.

U.S. Army. 1991. Army Regulation 40-10: Health Hazard Assessment program in support of the Army materiel acquisition decision process. Headquarters, Department of the Army, Washington, DC.

Vinegar, A., and G.W. Jepson. 1997. PBPK modeling of CF₃I releases from F15 engine nacelles. Proceedings of the Halon Options Technical Working Conference, pp 162-168.

Vinegar, A., G.W. Jepson, S.J. Hammann, G. Harper, D.S. Dierdorf, and J.H. Overton. 1999. Simulated blood levels of CF₃I in personnel exposed during its release from an F-15 jet engine nacelle and during intentional inhalation. Am. Ind. Hyg. Assoc. J. 60:403-408.

Yamamoto, T., A. Yasuhara, F. Shiraishi, K. Kaya, and T. Abe. 1997. Thermal decomposition of halon alternatives. Chemosphere. 35:643.