Acute Exposure Guideline Levels for Selected Airborne Chemicals: Volume 18 (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

3 Selected Halogen Fluorides1 Acute Exposure Guideline Levels PREFACE Under the authority of the Federal Advisory Committee Act (FACA) P.L. 92-463 of 1972, the National Advisory Committee for Acute Exposure Guide- line Levels for Hazardous Substances (NAC/AEGL Committee) has been estab- lished to identify, review, and interpret relevant toxicologic and other scientific data and develop AEGLs for high-priority, acutely toxic chemicals. AEGLs represent threshold exposure limits for the general public and are applicable to emergency exposure periods ranging from 10 minutes (min) to 8 hours (h). Three levels—AEGL-1, AEGL-2, and AEGL-3—are developed for each of five exposure periods (10 and 30 min and 1, 4, and 8 h) and are distin- guished by varying degrees of severity of toxic effects. The three AEGLs are defined as follows: AEGL-1 is the airborne concentration (expressed as parts per million or milligrams per cubic meter [ppm or mg/m3]) of a substance above which it is predicted that the general population, including susceptible individuals, could 1 This document was prepared by the AEGL Development Team composed of Sylvia Talmage (Oak Ridge National Laboratory), Heather Carlson-Lynch (SRC, Inc.), Chemi- cal Manager William Bress (National Advisory Committee [NAC] on Acute Exposure Guideline Levels for Hazardous Substances), and Ernest V. Falke (U.S. Environmental Protection Agency). The NAC reviewed and revised the document and AEGLs as deemed necessary. Both the document and the AEGL values were then reviewed by the National Research Council (NRC) Committee on Acute Exposure Guideline Levels. The NRC committee has concluded that the AEGLs developed in this document are scientifi- cally valid conclusions based on the data reviewed by the NRC and are consistent with the NRC guidelines reports (NRC 1993, 2001). 66 

Selected Halogen Fluorides 67 experience notable discomfort, irritation, or certain asymptomatic, nonsensory effects. However, the effects are not disabling and are transient and reversible upon cessation of exposure. AEGL-2 is the airborne concentration (expressed as ppm or mg/m3) of a substance above which it is predicted that the general population, including sus- ceptible individuals, could experience irreversible or other serious, long-lasting adverse health effects or an impaired ability to escape. AEGL-3 is the airborne concentration (expressed as ppm or mg/m3) of a substance above which it is predicted that the general population, including sus- ceptible individuals, could experience life-threatening health effects or death. Airborne concentrations below the AEGL-1 represent exposure concentra- tions that could produce mild and progressively increasing but transient and nondisabling odor, taste, and sensory irritation or certain asymptomatic, nonsen- sory effects. With increasing airborne concentrations above each AEGL, there is a progressive increase in the likelihood of occurrence and the severity of effects described for each corresponding AEGL. Although the AEGL values represent threshold concentrations for the general public, including susceptible subpopula- tions, such as infants, children, the elderly, persons with asthma, and those with other illnesses, it is recognized that individuals, subject to idiosyncratic respons- es, could experience the effects described at concentrations below the corre- sponding AEGL. 1. GENERAL INFORMATION ON SELECTED HALOGEN FLUORIDES In this chapter, the bases of the AEGL values for the following three halo- gen fluorides are described: chlorine pentafluoride (ClF5), bromine pentafluoride (BrF5), and bromine trifluoride (BrF3). Information relevant to all three com- pounds is presented first, and is followed by separate sections on the individual chemicals. 1.1. Physical and Chemical Properties The physical and chemical properties for ClF5, BrF5, and BrF3 are present- ed in Table 3-1. ClF5 is an extremely reactive gas. The gas does not burn, but may ignite combustibles. It explodes on contact with organic materials, is vio- lently hydrolyzed by water, and reacts vigorously or explosively with metals and fuels. Explosions may result from reaction with other chemicals, including am- monia, carbon monoxide, hydrogen sulfide, sulfur dioxide, and hydrogen gas (Teitelbaum 2001). 

68 Acute Exposure Guideline Levels TABLE 3-1 Chemical and Physical Properties of Selected Halogen Fluoridesa Parameter Chlorine pentafluoride Bromine pentafluoride Bromine trifluoride Synonyms None Bromine fluoride None CAS registry no. 13637-63-3 7789-30-2 7787-71-5 Chemical formula ClF5 BrF5 BrF3 Molecular weight 130.45 174.89 136.91 Physical state Colorless gasb Liquid Colorless to pale yellow liquid Melting point -103ºCb -60.5ºC 8.77°C Boiling point -13.1ºCb 40.76ºC 125.75ºC Solubility in water Violently hydrolyzed Explodes on contact Reacts with water by waterc with water Density/specific gravity 1.79 mg/L at 20ºCd 2.4604 g/mL at 25ºC 2.803 g/mL at 25ºC (water = 1) Vapor density (air = 1) 5.3 g/Lb 6.05 4.7e Vapor pressure 3.4 bar at 20ºCf 328 mm Hg at 20ºCg 2.8030 g/cm3 c Flammability limits Nonflammable Nonflammable Nonflammable Conversion factors 1 ppm = 5.34 mg/m3 1 ppm = 7.15 mg/m3 1 ppm = 5.6 mg/m3 (calculated) 1 mg/m3 = 0.19 ppm 1 mg/m3 = 0.14 ppm 1 mg/m3 = 0.18 ppm a HSDB (2007a,b) except where noted. b Lide (1999). c Teitelbaum (2001). d Bailey and Woytek (2004). e Weiss (1980). f Air Liquide (2005). g NIOSH/OSHA (1992). ClF5 is proposed to react with water according to the following reaction (Darmer 1971): ClF5 + 2H2O → ClO2F + 4HF The reaction of ClF5 and other halogen fluorides with water is violent (Ai- gueperse et al. 2000; Bailey and Woytek 2004; Atwood 2006). Following hy- drolysis of ClF5, the reaction of the breakdown product hydrogen fluoride (HF) with OH is endothermic and, therefore, not a viable source of F atoms (Syage 1994). BrF5 is a colorless or light yellow liquid when below its boiling point of 40.8ºC. Above its boiling point, BrF5 is a colorless, pungent, and corrosive gas. It is stable to heat, shock, and electric sparks (ACGIH 2001). Although non- flammable, fire may result from contact of BrF5 with combustibles at room tem- perature. Reaction with water is violent, with potential release of bromine, fluo- rine, hydrogen bromide, and hydrogen fluoride (Dost et al. 1968; NIOSH/OSHA 1992; Teitelbaum 2001). 

Selected Halogen Fluorides 69 BrF3 is a highly toxic, colorless to yellow or red liquid. The commercial grade is yellow to red in color due to contamination with bromine. BrF3 has the highest boiling point of any of the halogen fluorides. It is extremely reactive; reaction with water releases bromine, oxygen, and bromic and hydrofluoric ac- ids. BrF3 etches glass, sets fire to paper and wood, and reacts violently with most organic compounds (Braker and Mossman 1980; Bailey and Woytek 2004; HSDB 2007a). Reaction of BrF3 with water is likewise violent, producing a complex mixture of products including bromine and bromic and hydrofluoric acids (Braker and Mossman 1980; Owen 2005). According to Weiss (1980), the primary product is HF. Information on the hydrolysis of ClF3 provides additional information on the hydrolysis of the halogen fluorides, as similar products are formed. ClF3 decomposes by hydrolysis to a variety of substances including ClOF (the initial product), ClF, ClO2F, ClO3F, ClO2, Cl2, and HF; higher humidity increases the rate of decomposition (reviewed by NRC 2007a). 1.2. Use and Production ClF5 was once considered for use as an oxidizing propellant for missile propulsion, along with hydrazine and monomethyl hydrazine (Darmer et al. 1972; Syage 1994). A typical missile-launch-propellant payload would involve 2,000 kg of ClF5 (Syage 1994). At present, ClF5 has no significant industrial use other than as a fluorinating and oxidizing agent (Teitelbaum 2001). The halogen fluorides are manufactured by the reaction of fluorine with the corresponding halogen (Bailey and Woytek 2004). No information on storage and transporta- tion of ClF5 was available. BrF5 is manufactured by the fluorination of bromine at 200ºC in a metal apparatus (O’Neil et al. 2001). It can also be prepared by heating a mixture of BrF3 and fluorine to 200ºC. It is shipped in compressed gas containers under its own vapor pressure (Braker and Mossman 1980; NIOSH/OSHA 1992). BrF5 is predominantly used as a fluorinating agent to produce fluorocarbons and as an oxidizer in rocket propellant systems (ACGIH 2001). Metal chlorides, bromides, and iodides are converted to fluorides by treatment with BrF5 (Braker and Mossman 1980). Uranium is converted to uranium hexafluoride by strong oxi- dizing agents, including BrF5 (Bailey and Woytek 2004). BrF3 is prepared by direct combination of one part bromine with three parts fluorine in a water-cooled copper reactor at temperatures between 15 and 50°C, or by the reaction of bromine fluoride with fluorine (Braker and Mossman 1980). No production data on BrF3 were available. BrF3 is used as a solvent for other fluorides and as a fluorinating agent. It is used as an oxidizing agent in cutting tools used in deep oil-well drilling. Uranium is converted by BrF3 to ura- nium hexafluoride. It has also been of interest as a propellant for rockets and missiles (Braker and Mossman 1980; Bailey and Woytek 2004). 

70 Acute Exposure Guideline Levels 1.3. Structure-Activity Relationships ClF5, ClF3, BrF5, BrF3, HF, and ClO2 are toxicologically related, and all produce the toxic effect at the point of absorption, which is primarily related to the agent’s physical form (vapor, mist, and aerosol). In the moist respiratory tract, ClF3 is predicted to hydrolyze to ClOF, which further degrades to ClO2F and ClF (Dost et al. 1974). ClO2F rapidly hydrolyzes to ClO2, HF, and ClOx anions; the first two products predominate and are thought to be responsible for ClF3 toxicity, as the ClOx anions are relatively nontoxic. Hydrolysis of ClF5 is predicted to follow a similar path. If a similar path exists for bromine to form BrO2, it is expected to be less toxic than ClO2, as BrO2 is less reactive than ClO2. Lethality data provide a means of comparing the relative toxicities of these compounds; 1-h lethality data for ClO2, CLF5, CLF3, BrF5, and HF are presented in Table 3-2 (no data were available for BrF3). On the basis of LC50 (lethal con- centration, 50% lethality) values, the relative toxicities of these agents are: ClO2 > ClF5 > ClF3 ≈ BrF5 > HF The toxicities could be expressed in terms of HF equivalents. ClF3 is ap- proximately seven times more toxic than HF, and ClF5 is approximately 10 times more toxic than HF. The relative toxicities indicate that ClO2, an interme- diate in the dissociation of ClFx, plays a role in the toxicity of these agents. Inha- lation studies with rats indicate that the toxicity of ClF3 is comparable to that of chlorine dioxide (ClO2) on the basis of chlorine equivalents and is comparable to that of HF on the basis of fluorine equivalents (Dost et al. 1974). The hydrolysis of ClF5 is exothermic. Therefore, the enhanced toxicity of ClF5 relative to HF on a fluorine atom equivalent basis may be due to the effect of heat from the reaction (Syage 1994). Few data are available for bromine fluorides. For BrF5, Dost et al. (1970) reported a 1-h 95% lethal concentration of 500 ppm in male Sprague-Dawley rats. When the exposure duration was reduced to 30 min, none of the rats died; however, the observation period after exposure was for only 20 h before the rats were killed. The 1-h LC50s for ClF5 and ClF3 in rats were 122 and 299 ppm, re- spectively. The data suggests that BrF5 is less toxic than ClF5, which is in ac- cordance with their chemical reactivity. By extension, BrF3 is predicted to be less toxic than ClF3. The order of relative toxicities closely follows the chemical reactivity of the halogenated fluorine compounds. According to Bailey and Woytek (2004), the chemical reactivity of these compounds is, in order of decreasing reactivity: ClF5 > ClF3 > BrF5 > iodine heptafluoride (IF7) > chlorine monofluoride (ClF) > BrF3 > iodine pentafluoride (IF5) > bromine monofluoride (BrF). 

Selected Halogen Fluorides 71 TABLE 3-2 One-hour LC50 Values for Halogen Fluorides and Related Compounds or Decomposition Productsa Species ClO2 (ppm) ClF5 (ppm) ClF3 (ppm) BrF5 (ppm) HF (ppm) Monkey – 173 230 – 1,774 Dog – 122 – – – Rat 10 < LC50 < 54b 122 299 <500 (LC95)c 1,276 Mouse – 57 178 – 501 a Darmer et al. (1972) except where noted. b NRC (2004). c Dost et al. (1968, 1970). 1.4. Absorption, Distribution, Metabolism, and Excretion As discussed above, the halogen fluorides are expected to hydrolyze rapid- ly in the moist respiratory tract. No information on the absorption, metabolism, or excretion of these compounds was found in the scientific literature. Few data are available to evaluate systemic fluoride deposition after expo- sure to the halogen fluorides. Bone fluoride concentrations in monkeys exposed by inhalation to ClF5 at 10 ppm for 60 min, 20 ppm for 30 min, or 30 ppm for 10 min were not substantially different from those of untreated controls; measure- ments were made 28 days after exposure (MacEwen and Vernot 1972). Dost et al. (1970) measured fluoride in rat tissues following exposure to BrF5 at 500 ppm for 30 min or ClF3 at 400 ppm for 15 min; selected results are presented in Table 3-3. The data show increased fluoride content in the blood and lungs im- mediately after exposure, which declined over time. Mean fluoride in bone in- creased from 300 µg/g immediately after exposure to BrF5 to 353 µg/g 20 h lat- er. After exposure to ClF3, mean fluoride in bone increased from 118 to 172 μg/g over a 24-h period. Although a small number of animals (four to six) are represented by the mean values and fluoride concentrations in untreated rats (see Table 3-3) varied widely, the data suggest that, at higher concentrations of halo- gen fluorides, fluoride may be absorbed into the bloodstream and distributed to bone and other tissues. Systemic distribution of fluoride is not believed to be a significant factor in the acute inhalation toxicity and lethality of the halogen fluorides, as the pri- mary symptoms and cause of death (see Section 1.5) are associated with corro- sion of the tissues at the site of contact. In contrast, acute systemic fluoride poi- soning leads to hypocalcemia and symptoms of convulsions, coma, hypotension, and acidosis (HSDB 2007a,b). 1.5. Mechanism of Toxicity The mechanism by which halogen fluorides exert their acute toxicity is by direct irritation and corrosion. In experimental animals, these strong oxidizing 

72 Acute Exposure Guideline Levels chemicals produce lacrimation, sneezing, and salivation, which progresses to nausea, difficult respiration, and unconsciousness with cyanosis. Lacrimation and rhinorrhea are responses to stimulation of the trigeminal nerve. The eyes and exposed skin suffer burns. At the high concentrations that cause lethality, the capacity of the upper respiratory tract to scrub the halogen fluorides from the inhaled air is exceeded, and the chemicals penetrate to the lungs, causing edema and destruction of the pulmonary tissue (Horn and Weir 1955; Dost et al. 1968, 1974; MacEwen and Vernot 1970, 1973; Darmer et al. 1972). These signs were observed in monkeys, dogs, rats, and mice exposed to ClF5 (Darmer et al. 1972) and in rats exposed to BrF5. The hydrolysis of ClF5 is exothermic. Acute intraperitoneal, intravenous, or intragastric administration of ClF5 (10, 25, or 50 μL) to rats, rabbits, guinea pigs, and cats resulted in rupture of the surrounding areas and destruction of the blood vessels leading to hemorrhage, and was followed by respiratory depres- sion, cardiac failure, and death (Weinberg and Goldhamer 1967). Hydrolysis of other halogenated fluorides may be similarly exothermic. Thus, at sufficiently high concentrations, the heat of hydrolysis may play a fundamental role in tissue destruction. 1.6. Species Variability As evidenced by the 15-, 30-, and 60-min LC50 values (see Table 3-4), the mouse is more sensitive to the lethal effects of ClF5 than the other species tested. The dog and rat are equally sensitive, and the monkey is the least sensitive to the toxic effects of ClF5. The 1- h LC50 values for ClF5 differ by a factor of 3 be- tween the most sensitive species, the mouse (57 ppm), and the least sensitive species, the monkey (173 ppm). TABLE 3-3 Fluoride Content in Bone, Blood Plasma, and Lungs of Rats Exposed to BrF5 (500 ppm for 30 min) or ClF3 (400 ppm for 15 min) and in Control Rats Exposure Bone (μg/g) Blood plasma (μg/g) Lungs (μg/g) Unexposed rats (group A) 308 (230-403) No data 2.0 (0.8-2.8) Unexposed rats (group B) 144 (112-174) 0.4 (0-0.6) 0.2 (0-0.6) Unexposed rats (group C) 145 (125-184) 1.6 (1.5-1.7) 1.2 (0.9-1.7) BrF5 (immediately after exposure) 300 (235-378) 7.8 (6.5-9.0) 5.9 (3.0-8.5) BrF5 (20 h after exposure) 353 (278-415) 3.7 (3.0-4.5) 2.0 (1.5-3.0) ClF3 (immediately after exposure) 118 (97-135) No data 2.6 (1.4-4.2) ClF3 (24 h after exposure) 172 (128-232) No data 1.3 (0.2-3.2) Source: Data from Dost et al. 1970. 

Selected Halogen Fluorides 73 TABLE 3-4 Species Variability in LC50 Values for ClF5 Species 15 min 30 min 1h Monkey 249 ppm - 173 ppm Dog 298 ppm 156 ppm 122 ppm Rat 257 ppm 194 ppm 122 ppm Mouse 144 ppm 105 ppm 57 ppm Source: Adapted from Darmer et al. 1972. No information was found on species variability in the lethal effects of BrF5 or BrF3. There is very little variation in species sensitivity to lethal concen- trations of the related compound ClF3. As Table 3-2 shows, the 1-h LC50 values for three species, the monkey, rat, and mouse, were remarkably similar, exhibit- ing a less than 2-fold difference. For HF, the order of relative sensitivity is gen- erally mouse > rat > monkey (Table 3-2). The nasal passages vary considerably in size and shape among species. The nasal passages of rodents and primates differ in gross anatomy, the amount and distribution of types of respiratory epithelium, and air-flow patterns. The noses of primates (humans and monkeys) show great similarity in these three factors (Schreider 1986), and the monkey is a more appropriate model for ex- trapolation of inhalation toxicity for irritants to humans than is the rodent. Fur- thermore, the respiratory rate of primates is lower than that of rodents. There- fore, the delivered dose to the respiratory tract in primates is lower than that of rodents exposed to the same concentration. Lethality data (Table 3-4) demon- strate this difference; the monkey is the least sensitive of the four species ex- posed to ClF5. On the basis of relative body size, the respiratory rate of humans is lower than that of monkeys, resulting in a lower dose to the target tissues in the respiratory tract in humans. 1.7. Susceptible Populations No information on subpopulations that are especially sensitive to the ef- fects of the halogen fluorides was found. Individuals with asthma may respond to exposure to respiratory irritants with increased bronchial responsiveness. The elderly and those who are ill may also have increased susceptibility to the effects of irritants. Individuals under stress, such as those involved in emergency situations and individuals engaged in physical activity, will experience greater deposition and pulmonary irritation than individuals at rest. Furthermore, individuals who breathe through their mouths would be at greater risk. 1.8. Concentration-Exposure Duration Relationship For ClF5, the dose-response curve for lethality is steep. Data on four spe- cies exposed to ClF5 (see Section 2.2.1.) show that less than a doubling of a con- 

74 Acute Exposure Guideline Levels centration is necessary to go from no deaths to the LC50 value. Dost et al. (1970) noted that the dose-response curve for fluoride oxidizing agents is generally steep. Using the mortality data for four species in the study of Darmer et al. (1972), ten Berge et al. (1986) calculated regression coefficients for the concen- tration-time-mortality response relationships of ClF5. The n value in the equa- tion Cn × t = k for the monkey, dog, rat, and mouse were 4.1, 1.4, 1.9, and 1.5, respectively. On the basis of lethality data, the n value for HF is 2 (NRC 2004) and for ClF3 is 1.3. A graph showing the regression analyses of 15-, 30-, and 60- min LC50 values for the rat is shown in Appendix A. The regression analysis confirmed the ten Berge et al. (1986) time-scaling value for n of 1.9 in the rat. The data from the MacEwen and Vernot (1971, 1972, 1973) studies of ClF5 indicate that, at least for the direct irritant responses to ClF5, concentration may be more important than exposure duration. However, for the other effects observed (e.g., lung pathology, pale liver and kidneys), the role of exposure du- ration versus concentration is difficult to interpret because the studies provided few qualitative and quantitative details of the pathology findings. Discordant findings could be due to the dissociation to other agents or to a metabolic path- way. The lethality data from the sparse data set on BrF5 (see Section 3.2.1.) indi- cate that the dose-response curve is steep. Using the two nonlethal data points, 500 ppm for 40 min and 1,000 ppm for 20 min, the value for the exponent n value in the relationship Cn × t = k might be approximated at 1. However, because lethality data for BrF5 is sparse, the more conservative default time-scaling values of n = 3 when extrapolating to shorter durations and n = 1 when extrapolating to longer durations were used when time-scaling the AEGL-3 values. No information on a concentration-exposure duration relationship for BrF3 was found. For comparison, an n value of 1.3 was used to time scale AEGL values for ClF3 (NRC 2007a) and an n value of 2.0 was used for hydrogen fluoride (NRC 2004). 1.9. Concurrent Exposure Issues The available data suggest that the halogen fluorides operate via direct contact irritation, and are highly reactive in a moist environment. As noted in Section 1.1, higher humidity is expected to increase the rate of decomposition of these compounds and, thus, may enhance their toxicity. Coexposure to more than one of the halogen fluorides may result in great- er exposure to the common decomposition products such as ClO2 and HF. In addition, use of BrF5 and BrF3 in the synthesis of uranium hexafluoride suggests the possibility of coexposure to bromine and uranium fluorides in the context of accidental releases. As HF is a hydrolysis product of both the bromine fluorides and uranium hexafluoride (NRC 2004), coexposure to these compounds might lead to greater toxicity due to higher levels of the reaction product HF. 

Selected Halogen Fluorides 75 1.10. Summary of AEGLs for Selected Halogen Fluorides AEGL-1 values are not recommended for ClF5, as the available data for AEGL-1 effects resulted in values that were very similar to the AEGL-2 values, indicating inadequate warning properties for this compound. AEGL-2 values were based on no-effect levels for irritation symptoms that might impair escape in four animal species exposed to ClF5 for 10 min or 1 h. AEGL-3 values were derived from the highest nonlethal concentration in rats. An uncertainty factor of 3 was used to account for interspecies differences and another factor of 3 was used to account for intraspecies variability. The factors are supported by data on ClF5 and related compounds that indicate limited interspecies variability and because the mechanism of action of ClF5 is direct irritation or corrosion, so met- abolic and physiologic differences are unlikely to play a major role. Time scal- ing was performed using the equation Cn × t = k. An empirical value for n of 1.9 was derived from lethality data on ClF5. No data pertinent to the AEGL-1 values for BrF5 were found, so AEGL-1 values are not recommended. Similarly, no data were available for AEGL-2 end points. In the absence of suitable data, the AEGL-2 values were set equal to those for the related compound ClF5. AEGL-3 values for BrF5 were based on the highest nonlethal concentration in rats. As with ClF5, uncertainty factor of 3 each were used to account for interspecies differences and intraspecies variabil- ity. Time scaling was performed using the equation Cn × t = k; default values of n = 1 when extrapolating to longer durations and n = 3 when extrapolating to shorter durations were used. No human or animal data on the toxicity of BrF3 were available. AEGL values for this compound were set equal to the AEGLs for the related compound ClF3 (NRC 2007a). Because BrF3 is expected to be less toxic than ClF3, a modi- fying factor to account for lack of data was not used. A summary of the AEGL values for the three halogen fluorides is present- ed in Table 3-5. For comparison, AEGL values for ClF3, HF, and ClO2 are pre- sented in Table 3-6. 1.11. Data Adequacy and Research Need Other than a brief case report on ClF5 exposure, there are no human data on the toxicity of ClF5, BrF5, or BrF3. Likewise, human data on the related compound ClF3 include a single case report of a brief (1-2 min) exposure (NRC 2007a). The acute toxicity of ClF5 in animals has been well-studied for durations up to 1 h, although some of the studies lack histopathology data. There are no data on the toxicity of ClF5 for exposures longer than 1 h. Data on the acute tox- icity of BrF5 include a single study (Dost et al. 1968) conducted in male rats exposed to one of two concentrations for durations of 20-60 min. The study pro- vided inadequate information on methods (in particular, duration of follow-up was not specified) and did not include microscopic examination of tissues. 

76 TABLE 3-5 AEGL Values for Selected Halogen Fluorides Classification 10 min 30 min 1h 4h 8h End Point (Reference) Chlorine pentafluoride AEGL-1 (nondisabling) NRa NRa NRa NRa NRa Insufficient warning properties. AEGL-2 (disabling) 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm No-effect level for impaired ability (3.7 mg/m3) (2.1 mg/m3) (0.91 mg/m3) (0.44 mg/m3) (0.30 mg/m3) to escape (MacEwen and Vernot 1972, 1973). AEGL-3 (lethal) 21 ppm 12 ppm 8.0 ppm 3.9 ppm 2.7 ppm Highest 1-h nonlethal concentration (110 mg/m3) (64 mg/m3) (43 mg/m3) (21 mg/m3) (14 mg/m3) in rats (Darmer et al. 1972). Bromine pentafluoride AEGL-1 NRa NRa NRa NRa NRa Insufficient data. (nondisabling) AEGL-2 (disabling) 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm Set equal to AEGL-2 values for (5.0 mg/m3) (2.8 mg/m3) (1.2 mg/m3) (0.57 mg/m3) (0.41 mg/m3) chlorine pentafluoride. AEGL-3 (lethal) 79 ppm 55 ppm 33 ppm 8.3 ppm 4.2 ppm Highest nonlethal concentration in (570 mg/m3) (390 mg/m3) (240 mg/m3) (59 mg/m3) (30 mg/m3) rats (Dost et al. 1970). Bromine trifluoride AEGL-1 (nondisabling) 0.12 ppm (0.67 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm Set equal to AEGL-1 values for mg/m3) (0.67 mg/m3) (0.67 mg/m3) (0.67 mg/m3) (0.67 mg/m3) chlorine trifluoride (NRC 2007a). AEGL-2 (disabling) 8.1 ppm 3.5 ppm 2.0 ppm 0.70 ppm 0.41 ppm Set equal to AEGL-2 values for (45 mg/m3) (20 mg/m3) (11 mg/m3) (3.9 mg/m3) (2.3 mg/m3) chlorine trifluoride (NRC 2007a). AEGL-3 (lethal) 84 ppm 36 ppm 21 ppm 7.3 ppm 7.3 ppm Set equal to AEGL-3 values for (470 mg/m3) (200 mg/m3) (120 mg/m3) (41 mg/m3) (41 mg/m3) chlorine trifluoride (NRC 2007a). a Not recommended. Absence of AEGL-1 values does not mean that exposures below the AEGL-2 values are without adverse effects. 

TABLE 3-6 AEGL Values for Hydrogen Fluoride, Chlorine Trifluoride, and Chlorine Dioxide Classification 10 min 30 min 1h 4h 8h End Point (Reference) AEGL-1 Hydrogen fluoride (NRC 2004) 1.0 ppm 1.0 ppm 1.0 ppm 1.0 ppm 1.0 ppm Threshold, pulmonary inflammation in humans (Lund et al. 1997, 1999). Chlorine trifluoride (NRC 2007a) 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm Slight irritation – dog (Horn and Weir 1956). Chlorine dioxide (NRC 2007b) 0.15 ppm 0.15 ppm 0.15 ppm 0.15 ppm 0.15 ppm Slight salivation, slight lacrimation, and slight chromodacryorrhea in rats exposed at 3 ppm for 6 h (DuPont 1955). AEGL-2 Hydrogen fluoride (NRC 2004) 95 ppm 34 ppm 24 ppm 12 ppm 12 ppm NOAEL for pulmonary effects in cannulated rats (Dalbey 1996, Dalbey et al. 1998)a, sensory irritation in dogs (Rosenholtz et al. 1963)b. Chlorine trifluoride (NRC 2007a) 8.1 ppm 3.5 ppm 2.0 ppm 0.70 ppm 0.41 ppm Threshold, impaired ability to escape – dog (Horn and Weir 1956). Chlorine dioxide (NRC 2007b) 1.4 ppm 1.4 ppm 1.1 ppm 0.69 ppm 0.45 ppm Lacrimation, salivation, dyspnea, weakness, and pallor in rats exposed at 12 ppm for 6 h (DuPont 1955). AEGL-3 Hydrogen fluoride (NRC 2004) 170 ppm 62 ppm 44 ppm 22 ppm 22 ppm Lethality threshold in cannulated rats (Dalbey 1996; Dalbey et al. 1998)c; lethality threshold in mice (Wohlslagel et al. 1976)d. Chlorine trifluoride (NRC 2007a) 84 ppm 36 ppm 21 ppm 7.3 ppm 7.3 ppm Threshold for lethality – monkey (MacEwen and Vernot 1970). Chlorine dioxide (NRC 2007b) 3.0 ppm 3.0 ppm 2.4 ppm 1.5 ppm 0.98 ppm No lethality in rats exposed at 26 ppm for 6 h (DuPont 1955). a 10-min AEGL-2 value. b 30-min and 1-, 4-, and 8-h AEGL-2 values. c 10-min AEGL-3 value. d 30-min and 1-, 4-, and 8-h AEGL-3 values. 77 

78 Acute Exposure Guideline Levels No human or animal data on the toxicity of BrF3 were available. Taken collectively with the toxicity data on the related halogenated compounds ClF3, HF, and ClO2, the available data provide a reasonable basis for deriving AEGL values for the selected halogen fluorides; however, additional studies would serve to refine the AEGL values. In particular, the following studies would be beneficial: studies of the acute lethality of BrF3 in laboratory animals, studies of ClF5 and BrF5 for exposure durations of 1-8 h, studies of the acute toxicity of BrF5 in species other than rats, and additional data on the concentration-time relationship for BrF5. 2. CHLORINE PENTAFLUORIDE 2.1. Human Toxicity Data ClF3 is a colorless gas with a pungent or suffocating odor (Teitelbaum 2001; Air Liquide 2005). No information on the odor threshold was found, nor was information available on lethal or sublethal concentrations, neurotoxicity, developmental toxicity, reproductive toxicity, genotoxicity, or carcinogenicity of ClF3 in humans. In the performance of animal studies of ClF5 at the Aerospace Medical Research Facility, Wright-Patterson AFB, Ohio, a research staff member had occasion to take a single breath of ClF3 at 30 ppm in the exposure chamber (MacEwen and Vernot 1973). The staff member experienced a mild “burning” of the lungs, mild nausea, an unpleasant taste in the mouth, and headache. The duration of these symptoms persisted was not reported. 2.2. Animal Toxicity Data 2.2.1. Acute Lethality Acute lethality data for ClF5 are summarized in Table 3-7 and discussed further below. 2.2.1.1. Nonhuman Primates Darmer et al. (1972; see also Darmer 1971; MacEwen and Vernot 1971) exposed groups of four male and female rhesus monkeys to various measured concentrations of ClF5 for 15, 30, or 60 min to determine LC50 values. The ob- servation period was 14 days. Chamber concentrations were measured with a fluoride ion specific electrode. Concentrations and mortality for the 15-, 30-, and 60-min exposures are presented in 3-8. Except for the death of one animal on the third day after exposure to ClF5 at 225 ppm for 15 min, all deaths occurred dur- ing or immediately following exposure. The investigators calculated 15-, 30-, 

Selected Halogen Fluorides 79 TABLE 3-7 Acute Lethality in Laboratory Animals Exposed to Chlorine Pentafluoride Species Concentration (ppm) Exposure Duration Effect Reference Monkey 249 15 min LC50 Darmer et al. 1972 165 15 min No deaths 218 30 min LC50 198 30 min No deaths 173 60 min LC50 116 60 min No deaths Dog 298 15 min LC50 Darmer et al. 1972 168 15 min No deaths 156 30 min LC50 102 30 min 25% mortality 122 60 min LC50 63 60 min No deaths Rat 400 10 min 100% mortality Weinberg and 200 10 min 10% mortality Goldhamer 1967 100 15 min/d for 6 d “Survival of most” Rat 257 15 min LC50 Darmer et al. 1972 175 15 min 10% mortality 194 30 min LC50 163 30 min No deaths 122 60 min LC50 80 60 min No deaths Mouse 144 15 min LC50 Darmer et al. 1972 100 15 min 25% mortality 105 30 min LC50 102 30 min 25% mortality 57 60 min LC50 35 60 min 10% mortality a The LC50 values are calculated values; the other mortality values are the empirical data. and 60-min LC50 values of 249 ppm (95% confidence limits [95% CI]: 191-326 ppm), approximately 218 ppm, and 173 ppm (95% CI: 148-204 ppm), respec- tively (see Table 3-8). Examination of the data indicates that the 30- and 60-min lethality data reported in this study exhibited a clear linear concentration- response relationship when plotted on a semi-log plot, lending confidence to the calculated LC50 values for these durations. In contrast, the 15-min data in mon- keys did not; thus, there is uncertainty associated with the 15-min LC50 value for monkeys. Almost immediately upon onset of exposure, monkeys had signs of irrita- tion that included salivation, lacrimation, sneezing, nausea, and labored breath- ing. Cyanosis was usually evident by the end of exposure. The signs disappeared in surviving animals within 30 min after exposure. Monkeys, dogs, rats, and mice (see following sections) that died during exposure exhibited similar pa- thology in the respiratory tract, the primary target of ClF5. Alveolar destruction 

80 Acute Exposure Guideline Levels TABLE 3-8 Acute Lethality in Monkeys Exposed to Chlorine Pentafluoride Concentration (ppm) Exposure Duration Mortality 165 15 min 0/4 193 15 min 1/4 225 15 min 3/4 335 15 min 3/4 395 15 min 3/4 198 30 min 0/4 218 30 min 2/4 236 30 min 4/4 116 60 min 0/4 122 60 min 1/4 140 60 min 1/4 189 60 min 2/4 215 60 min 2/4 223 60 min 4/4 Source: Adapted from Darmer et al. 1972. was indicated by fluid and blood in the lungs. Nasal passages generally con- tained large amounts of mucus and fluids; blood was also found in some cases. There were no apparent systemic effects. The authors suggested that chemical pneumonitis was the cause of death that occurred during or immediately after exposure. Animals that survived the 14-day observation exposure period had incomplete recovery of the pulmonary tissue. Residual damage involved scar- ring and consolidation of pulmonary tissue. Corneal opacity was observed in most species but was less pronounced in monkeys, possibly because they tended to keep their eyes closed during exposure. 2.2.1.2. Dogs Darmer et al. (1972) exposed groups of four male and female beagles to various measured concentrations of ClF5 for 15, 30, and 60 min to determine LC50 values. The observation period was 14 days. Most deaths occurred during the first 2 days after exposure. Concentrations and mortalities for the 15-, 30-, and 60-min exposures are presented in Table 3-9. Dogs exposed at 143 ppm for 60 min died 8-14 days after exposure, whereas all dogs exposed at 170 ppm for 60 min died during the first day after exposure. The 15-, 30-, and 60-min LC50 values were 298 ppm (95% CI: 238-374 ppm), 156 ppm (95% CI: 113-256 ppm), and 173 ppm (95% CI: 148-204 ppm), respectively (Table 3-7). Corneal opacity was a common occurrence. 2.2.1.3. Rats In preliminary experiments, Weinberg and Goldhamer (1967) exposed groups of rats (strain not specified) to ClF5 at 200 (10 rats) or 400 ppm (three groups of six rats) for various durations. In the first two groups of rats exposed 

Selected Halogen Fluorides 81 at 400 ppm for 10 or 60 min, all died either during the exposure or within 15 min of the start of exposure. Of another group of six rats exposed to ClF5 at 400 ppm for 10 min, three survived but were sacrificed immediately after exposure. Gross findings in the three survivors included marked pulmonary edema and hemorrhage, myocardial infarction, and congestion in the liver and brain. Among 10 rats exposed to ClF5 at 200 ppm for 10 min, nine survived the expo- sure and three survived for 24 h, when the animals were killed. Examination of the animals revealed protein leakage into the lungs in all animals and regenera- tion of the enzymes aspartate aminotransferase and alanine aminotransferase 16 h after exposure. No systemic effects were noted in these rats. In the same study (Weinberg and Goldhamer 1967), 40 rats were exposed to ClF5 at 100 ppm for 15 min/day for up to 5 consecutive days. Groups of three rats were killed either immediately after or 16 h after each daily exposure (six rats/day). The authors stated that, “almost all rats survived exposure to 100 ppm for 15 min daily up to 5 days.” The days on which deaths occurred were not clearly documented, but one death may have occurred on the first day of expo- sure. Six rats survived to day 5. Surviving rats began to lose weight by the third exposure day. Groups of 10 male Sprague-Dawley rats were exposed to ClF5 at various concentrations and durations to determine LC50 values (Darmer et al. 1972). LC50 values were 257 ppm for 15 min (95% CI: 210-314 ppm), 194 ppm for 30 min (95% CI: 135-278 ppm), and 122 ppm for 60 min (95% CI: 108-139 ppm) (Table 3-7). Concentrations and mortalities for this study are presented in Table 3-10. The rats were observed for 14 days following exposure. Most deaths oc- curred either on the day of exposure or the day after. A few delayed deaths oc- curred 8-14 days after exposure. Corneal opacity was a common occurrence. TABLE 3-9 Acute Lethality in Dogs Exposed to Chlorine Pentafluoride Concentration (ppm) Exposure Duration Mortality 168 15 min 0/4 202 15 min 1/4 300 15 min 2/4 360 15 min 2/4 443 15 min 4/4 102 30 min 1/4 150 30 min 2/4 190 30 min 2/4 223 30 min 3/4 252 30 min 3/4 274 30 min 4/4 63 60 min 0/4 110 60 min 1/4 128 60 min 2/4 143 60 min 4/4 170 60 min 4/4 Source: Adapted from Darmer et al. 1972. 

82 Acute Exposure Guideline Levels TABLE 3-10 Acute Lethality Data in Rats Exposed to Chlorine Pentafluoride Concentration (ppm) Exposure Duration Mortality 175 15 min 1/10 235 15 min 4/10 258 15 min 6/10 300 15 min 7/10 325 15 min 9/10 373 15 min 6/10 432 15 min 9/10 120 30 min 0/10 163 30 min 0/10 185 30 min 3/10 190 30 min 6/10 233 30 min 9/10 250 30 min 10/10 80 60 min 0/10 100 60 min 1/10 120 60 min 4/10 136 60 min 8/10 Source: Adapted from Darmer et al. 1972. 2.2.1.4. Mice Groups of 10 male ICR mice were exposed to ClF5 at various concentra- tions and durations to determine LC50 values (Darmer et al. 1972). LC50 values were 144 ppm for 15 min (95% CI: 112-186 ppm), 105 ppm for 30 min (95% CI: 93-119 ppm), and 57 ppm for 60 min (95% CI: 47-70 ppm) (Table 3-7). Concentrations and mortalities from the study are presented in Table 3-11. The mice were observed for 14 days. Most deaths occurred immediately after expo- sure or the next day. Delayed deaths occurred 8-14 days after exposure. Corneal opacity was a common occurrence. 2.2.2. Nonlethal Toxicity Nonlethal toxicity studies of ClF5 in the monkey, dog, rat, and mouse are summarized in Table 3-12. 2.2.2.1. Nonhuman Primates Groups of four male and two female rhesus monkeys were exposed to ClF5 at 10 ppm for 60 min, 20 ppm for 30 min, or 30 ppm for 10 min (MacEwen and Vernot 1972). The monkeys were killed after a 28-day observation period. Concentrations were measured with a fluoride ion specific electrode. Lacrima- tion and nausea were observed almost immediately after onset of exposure and 

Selected Halogen Fluorides 83 disappeared within 30 min after exposure. Signs were similar in the three treat- ment groups. One monkey in the group exposed at 20 ppm for 30 min was found dead six days after exposure. Necropsy of the animal revealed purulent material in the upper respiratory tract and focal bronchopneumonia in the lungs. Because no monkeys exposed to ClF5 at 198 ppm for 30 min died in an earlier experi- ment at the same facility (Darmer et al. 1972), the cause of death for this animal is uncertain. No gross lesions were observed in monkeys exposed at 30 ppm for 10 min. One monkey in the 30-ppm group had some multifocal, white caseous material in the lungs. Monkeys exposed at 10 ppm for 60 min exhibited congest- ed lungs. However, in summarizing this study in a report the following year, MacEwen and Vernot (1973) reported that “gross and histopathological exami- nation at 28 days postexposure failed to produce any evidence of permanent effects of exposure” in monkeys or any other species; thus, it is difficult to in- terpret the pathology findings in the various exposure groups. Although mean body weights were not significantly affected following a 28-day observation period, weight gains during this period were depressed. At necropsy, the bone fluoride content of the treated monkeys did not differ from that of the control group. TABLE 3-11 Acute Lethality in Mice Exposed to Chlorine Pentafluoride Concentration (ppm) Exposure Duration Mortality 100 15 min 2/10 130 15 min 4/10 166 15 min 7/10 174 15 min 7/10 195 15 min 6/10 212 15 min 9/10 231 15 min 8/10 305 15 min 9/10 360 15 min 15/15 70 30 min 2/10 90 30 min 3/10 117 30 min 6/10 120 30 min 5/10 140 30 min 8/10 145 30 min 8/10 166 30 min 9/10 175 30 min 10/10 35 60 min 1/10 47 60 min 2/10 62 60 min 5/10 75 60 min 9/10 Source: Adapted from Darmer et al. 1972. 

84 TABLE 3-12 Nonlethal Data from Studies of Laboratory Animals Exposed to Chlorine Pentafluoride Species Concentration (ppm) Exposure Duration Effects Reference Monkey 10 60 min Lacrimation, nausea, transient weight gain depression, and congested lungs. MacEwen and Vernot 1972 20 30 min One death with bronchopneumonia, lacrimation, nausea, transient weight gain depression, multifocal caseous material in lungs of one monkey, and no grossly observed pathology in the others. 30 10 min Lacrimation, nausea, transient weight gain depression, and no grossly observed pathology. Monkey 5 60 min Salivation, ocular irritation, lacrimation, and rhinorrhea in all groups. No MacEwen and Vernot 1973 10 30 min grossly observed pathology in any group. 30 10 min Dog 5 60 min Salivation, ocular irritation, lacrimation, and rhinorrhea in all groups; no MacEwen and Vernot 1973 10 30 min grossly observed pathology in any group. 30 10 min Rat 10 60 min Lacrimation, salivation, and pale livers and kidneys. MacEwen and Vernot 1972 20 30 min Lacrimation, salivation, and no grossly observed pathology. 30 10 min Lacrimation, salivation, and no grossly observed pathology Rat 5 60 min Salivation, ocular irritation, lacrimation, and rhinorrhea in all groups. No MacEwen and Vernot 1973 10 30 min grossly observed pathology. 30 10 min Rat 30 10 min Irritation and increase in lung wet weight when killed immediately after MacEwen and Vernot 1973 exposure. (Gross pathology not assessed.) Rat 3 10 min No irritation and no effect on body weight gain or lung wet weight. MacEwen and Vernot 1973 7 10 min Slight ocular irritation, and no effect on body weight gain or lung wet weight. Mouse 10 60 min Lacrimation and transient effect on body weight. MacEwen and Vernot 1972 20 30 min One death (cause not specified), lacrimation, and transient effect on body weight. 30 10 min Lacrimation, no effects on body weight, and no grossly observed pathology. 

Mouse 5 60 min Salivation, ocular irritation, lacrimation, and rhinorrhea in all groups. No MacEwen and Vernot 1973 10 30 min grossly observed pathology. 30 10 min Mouse 30 10 min Irritation, mild pulmonary congestion observed when killed immediately MacEwen and Vernot 1973 after exposure. 85 

86 Acute Exposure Guideline Levels In a follow-up study, groups of five rhesus monkeys were exposed to ClF5 at 30 ppm for 10 min, 10 ppm for 30 min, or 5 ppm for 60 min (MacEwen and Vernot 1973). A single concurrent control group was maintained. The protocol was the same as in the previously described study by MacEwen and Vernot (1972), with the exception that monkeys were observed for 6 weeks after expo- sure. Blood chemistry and body weight were monitored. Salivation, ocular irrita- tion, lacrimation, and rhinorrhea, described by the study authors as “typical ClF5 irritation and discomfort symptoms,” were seen during the exposures, with the most severe signs observed in the group exposed at 30 ppm for 10 min. All groups (including the control group) exhibited a slight weight loss 2 weeks after exposure due to a change in living quarters. All groups except the one exposed at 5 ppm for 60 min exhibited a weight gain 4 weeks after exposure. Both the 5- ppm (60 min) and the 30-ppm (10 min) groups exhibited slight weight loss 6 weeks after exposure. Clinical chemistry parameters failed to show a pattern consistent with exposure. Gross pathologic examinations revealed no lesions consistent with exposure. 2.2.2.2. Dogs Groups of eight beagles (gender not specified) were exposed to ClF5 at 30 ppm for 10 min, 10 ppm for 30 min, or 5 ppm for 60 min (MacEwen and Vernot 1973). A single concurrent control group was maintained. The protocol was the same as in the study by MacEwen and Vernot (1972) described earlier. Blood chemistry and body weight were monitored. Salivation, ocular irritation, lacri- mation, and rhinorrhea, described by the study authors as “typical ClF5 irritation and discomfort symptoms,” were seen during the exposures, with the most se- vere signs observed in the group exposed for 10 min at 30 ppm. Exposed groups exhibited weight gains similar to that of the control group. Clinical chemistry parameters failed to show a pattern consistent with exposure. Gross pathologic examinations revealed no lesions consistent with exposure. 2.2.2.3. Rats Groups of 30 male Sprague-Dawley rats were exposed to ClF5 at 30 ppm for 10 min, 20 ppm for 30 min, or 10 ppm for 60 min (MacEwen and Vernot 1972). Concentrations were measured with a fluoride ion specific electrode. Lacrimation and salivation were observed almost immediately after onset of exposure and disappeared within 30 min after exposure. Signs were similar in the three treatment groups. Mean body weights and body weight gains were un- affected during the 28-day observation period. Gross examination at necropsy revealed no effects in any of the control rats or rats exposed for 10 min. Rats exposed for 60 min had pale livers and kidneys. In a follow-up experiment, groups of 30 Sprague-Dawley rats were ex- posed to ClF5 at 30 ppm for 10 min, 10 ppm for 30 min, or 5 ppm for 60 min 

Selected Halogen Fluorides 87 (MacEwen and Vernot 1973). A single concurrent control group was main- tained. The protocol was the same the one used in the previously described study by MacEwen and Vernot (1972). Blood chemistry and body weight were moni- tored over a 4-week period. Salivation, ocular irritation, lacrimation, and rhinor- rhea, described by the study authors as “typical ClF5 irritation and discomfort symptoms,” were seen during the exposures, with the most severe signs ob- served in the group exposed for 10 min at 30 ppm. Exposed groups exhibited body weight gains similar to that of the control group. Gross pathologic exami- nations revealed no lesions consistent with exposure. An additional experiment was conducted to study the effects of ClF5 on the lungs of 10 Sprague-Dawley rats exposed at 30 ppm for 10 min (MacEwen and Vernot 1973). Rats were killed immediately after exposure and the lungs were weighed. The mean wet weight of the lungs was significantly increased over that of the controls (by 0.1 g), indicating the presence of edema (dry lung weights were identical for the two groups). The severity of the irritation was not described, but the study authors concluded that “the degree of discomfort expe- rienced by the experimental animals during exposure, and the fact that signifi- cant edema resulted in the lungs of rats exposed to this dose” indicated that 30 ppm would not be an acceptable emergency exposure limit. A fourth experiment was performed at lower concentrations (MacEwen and Vernot 1973). Groups of 20 male Sprague-Dawley rats were exposed to ClF5 at 3 ppm or 7 ppm for 10 min. Half of the rats in each group was killed immediately after exposure to determine lung weight and the remaining animals were followed for 28 days to evaluated weight gain patterns and pulmonary le- sions. Two groups of 10 rats each comprised the concurrent control groups. No signs of irritation were observed during the 10 min exposure at 3 ppm. Slight moistening of the eyes was observed following the 10-min exposure at 7 ppm. There was no effect on body weight gain or lung wet weight following the expo- sures. 2.2.2.4. Mice Groups of 30 male ICR mice were exposed to ClF5 at 30 ppm for 10 min, 20 ppm for 30 min, or 10 ppm for 60 min (MacEwen and Vernot 1972). Con- centrations were measured with a fluoride ion specific electrode. Lacrimation was observed almost immediately after onset of exposure and disappeared with- in 30 min after exposure. Signs were similar in the three treatment groups. One mouse exposed at 20 ppm for 30 min was found dead 4 days after exposure; the cause of death could not be ascertained. Mean body weight and body weight gain of all treatment groups were lower during the first week after exposure, but were unaffected during the remainder of the 28-day observation period. Gross examination at necropsy revealed no effects in any of the control mice or mice exposed for 10 min. “Significant pathology” in the group exposed for 60 min was reported, but no details were described other than pale liver and kidneys. 

88 Acute Exposure Guideline Levels In a follow-up experiment, groups of 30 ICR mice were exposed to ClF5 at 30 ppm for 10 min, 10 ppm for 30 min, or 5 ppm for 60 min (MacEwen and Vernot 1973). A single concurrent control group was maintained. The protocol was the same as that described for the study by MacEwen and Vernot (1972). Blood chemistry and body weight were monitored over a 4-week period. Saliva- tion, ocular irritation, lacrimation, and rhinorrhea, described by the study authors as “typical ClF5 irritation and discomfort symptoms,” were seen during the ex- posures, with the most severe signs observed in the group exposed for 10 min at 30 ppm. Exposed groups exhibited body weight gains during the first week after exposure. The group exposed for 30 min at 10 ppm exhibited slight weight loss 2-4 weeks after exposure. Gross pathologic examinations revealed no lesions consistent with exposure. A third experiment was conducted in which 10 mice were exposed to ClF5 at 30 ppm for 10 min, (MacEwen and Vernot 1973). The animals were killed immediately after exposure and their lungs were examined. “Typical” irritation symptoms were reported. Gross examination revealed mild pulmonary conges- tion in the exposed group but not in the control group. The severity of the irrita- tion was not described, but the study authors cited the “degree of discomfort experienced by the experimental animals during exposure” as a contributing rationale for concluding that 30 ppm would not be an acceptable emergency exposure limit. 2.2.3. Neurotoxicity No information on the neurotoxicity of ClF5 in animals was found. 2.2.4. Developmental and Reproductive Toxicity No information on the developmental or reproductive toxicity of ClF5 in animals was found. 2.2.5. Genotoxicity No information on the genotoxicity of ClF5 was found. 2.2.6. Chronic Toxicity and Carcinogenicity No information on the chronic toxicity or carcinogenicity of ClF5 was found. 2.2.7. Summary A series of studies with four species (monkey, dog, rat, and mouse) pro- vided information on the irritation and lethality of ClF5. One-hour LC50 values were 173 ppm for monkeys, 122 ppm for dogs, 122 ppm for rats, and 57 ppm for 

Selected Halogen Fluorides 89 mice (Darmer et al. 1972). The highest nonlethal concentrations for 1-h expo- sures for the monkey, dog, and rat were 116, 63, and 80 ppm, respectively. The four species were also exposed to ClF5 at 5 or 10 ppm for 60 min, 10 or 20 ppm for 30 min, and 30 ppm for 10 min. The animals had distinct signs of irritation, including lacrimation, salivation, nausea (monkey), and for some species, re- versible pulmonary congestion (MacEwen and Vernot 1972, 1973). Rats ex- posed at 3 ppm for 10 min had no signs of irritation, and rats exposed at 7 ppm for 10 min had slight ocular irritation (MacEwen and Vernot 1973). 2.3. Data Analysis for AEGL-1 2.3.1. Human Data Relevant to AEGL-1 No human data on ClF5 relevant to developing AEGL-1 values were found. 2.3.2. Animal Data Relevant to AEGL-1 Animal data relevant to developing AEGL-1 values for ClF5 are sparse. No signs of irritation were observed during a 10-min exposure of 20 Sprague- Dawley rats to ClF5 at 3 ppm (MacEwen and Vernot 1973). Ocular moisture was observed following a 10-min exposure at 7 ppm. Animals of both groups killed immediately after exposure had wet lung weights similar to those of the control groups. The absence of fluid in the lungs (edema) indicates that toxicologically significant quantities of ClF5 did not reach the lungs at those concentrations. Animals of both groups killed after a 28-day observation period had no alterna- tions in body weight gain or gross pathologic findings in the lungs. 2.3.3. Derivation of AEGL-1 Values An AEGL-1 value for chlorine pentafluoride could have been derived on the basis of a no-effect level for irritation in the rat of 3 ppm for 10 min (MacE- wen and Vernot 1973). A total uncertainty factor of 10 would have been applied; a factor of 3 for interspecies differences and a factor of 3 for intraspecies varia- bility. However, the resulting 10-min AEGL-1 value (0.30 ppm) exceeds the 4- and 8-h AEGL-2 values (0.24 and 0.17 ppm, respectively) and is close to the 1-h AEGL-2 value (0.50 ppm). Thus, AEGL-1 values are not recommended for ClF5 because of inadequate sensory warning properties. 2.4. Data Analysis for AEGL-2 2.4.1. Human Data Relevant to AEGL-2 There are no human data relevant to developing AEGL-2 values. An indi- vidual suffered mild “burning” of the lungs, unpleasant taste, nausea and head- 

90 Acute Exposure Guideline Levels ache following a single breath of ClF5 at 30 ppm (MacEwen and Vernot 1973). However, no meaningful understanding of the toxicity of ClF5 can be obtained from this report. 2.4.2. Animal Data Relevant to AEGL-2 Data relevant to deriving AEGL-2 values for ClF5 are available from a se- ries of experiments in monkeys, dogs, rats, and mice by MacEwen and Vernot (1972, 1973). In the first experiment, exposure of groups of six rhesus monkeys, 30 rats, and 30 mice to ClF5 at 10 ppm for 60 min, 20 ppm for 30 min, or 30 ppm for 10 min or groups of eight beagles at 30 ppm for 10 min resulted in irri- tation, as exhibited by salivation and lacrimation (MacEwen and Vernot 1972). All animals were monitored for several weeks. At gross necropsy, pulmonary congestion was observed in monkeys exposed at 10 ppm for 60 min, and rats had pale livers and kidneys. A second experiment at lower concentrations re- ported salivation, ocular irritation, lacrimation, and rhinorrhea, but no gross pathologic findings in groups of five rhesus monkeys, eight beagles, 30 rats, and 30 mice exposed to ClF5 at 5 ppm for 60 min, 10 ppm for 30 min, or 30 ppm for 10 min (MacEwen and Vernot 1973). Irritation was reportedly more severe in the groups exposed at 30 ppm for 10 min; however, no detail about the severity of the irritant symptoms was provided. No gross pathologic findings related to exposure were observed (MacEwen and Vernot 1973). In a third experiment, wet weights of the lungs were increased in a group of 10 rats killed immediately after exposure to ClF5 at 30 ppm for 10 min (the only concentration-exposure duration group examined in that experiment) (MacEwen and Vernot, 1973). As part of the same experiment, gross examination of 10 mice following exposure at 30 ppm for 10 min revealed mild pulmonary congestion; this finding was not observed in the control group (MacEwen and Vernot 1973). In the fourth exper- iment, 20 rats exposed at 7 ppm for 10 min exhibited only slight moistening of the eyes, but 20 rats exposed at 3 ppm for 10 min had no signs of irritation (MacEwen and Vernot 1973). Table 3-13 summarizes the data relevant to deriv- ing AEGL-2 values for ClF5. 2.4.3. Derivation of AEGL-2 Values The studies by MacEwen and Vernot (1972, 1973) provide little or no in- formation on the severity of irritant symptoms in the animals; as a consequence, it is difficult to identify no-effect levels for AEGL-2 end points. The data indicate that exposure to ClF5 at concentrations 30 ppm or higher for 10 min, 10 ppm or higher for 30 min, and 5 ppm or higher for 60 min results in irritant effects, includ- ing salivation, ocular irritation, lacrimation, and rhinorrhea in all species tested. Additionally, pathologic effects in the lungs were found in monkeys exposed at 10 ppm for 60 min, in one monkey exposed at 20 ppm for 30 min, and in rats and 

TABLE 3-13 Data Relevant to AEGL-2 Values for Chlorine Pentafluoride Duration (min) Concentration (ppm) Symptoms Gross pathology Other effects Species 10 3 None None No effect on body weight Rats 7 Slight moistening of eyes None No effect on body weight Rats 30 Salivation, ocular irritation, Reversible pulmonary Slight transient weight loss Monkeys, dogs, lacrimation, and rhinorrhea (all congestion and edema (monkeys). rats, and mice species); nausea (monkeys). (rat and mice). 30 10 Salivation, ocular irritation, None Transient weight gain loss Monkeys, dogs, lacrimation, and rhinorrhea. (monkey and mouse). rats, and mice 20 Lacrimation (all species); White caseous material in Transient weight gain Monkeys, rats, salivation (rats); nausea (monkeys). lung (one monkey). depression (monkeys and and mice mice); one monkey died with bronchopneumonia; one mouse died of undetermined cause. 60 5 Salivation, ocular irritation, None Slight transient weight gain Monkeys, dogs, lacrimation, and rhinorrhea depression (monkeys). rats, and mice (all species). 10 Lacrimation (all species); Congested lungs (monkeys); Transient weight gain depression Monkeys, rats, salivation (rats); nausea (monkeys). pale livers and kidneys (rats). (monkeys and mice) and mice Source: Data from MacEwen and Vernot 1972, 1973. 91 

92 Acute Exposure Guideline Levels mice exposed at 30 ppm for 10 min. Among the experiments conducted, minimal or no irritation was found only in the studies of rats exposed for 10 min at 3 or 7 ppm; no irritation occurred at 3 ppm and only slight moistening of the eyes was observed at 7 ppm. Without additional information on the severity and prevalence of the irritant symptoms, it is difficult to determine the concentration and exposure duration at which escape might be impaired. However, a conservative assumption is that the lacrimation and ocular irritation exhibited by all species exposed at 10 ppm for 30 min or at 5 ppm for 60 min were of sufficient severity to impair es- cape. Thus, those concentrations were considered effect levels for AEGL-2 end points. Under that assumption, the highest no-effect level is 7 ppm, which was associated with slight moistening of the eyes in rats exposed for 10 min. The 10-min no-effect level of 7 ppm was selected as the point-of- departure for the 10-min and 30-min AEGL-2 values. Because of the uncertainty associated with extrapolating a 10-min point-of-departure to exposure durations of 1 h and longer, the point-of-departure for the 1-, 4-, and 8-h AEGL-2 values was based on the effect level of 5 ppm for a 60-min exposure; that concentration was reduced by a modifying factor of 3 to account for extrapolation from an effect level to a no-effect level for AEGL-2 end points. Thus, the point-of- departure was 1.7 ppm. A total uncertainty factor of 10 was applied; a factor of 3 for interspecies difference and a factor of 3 for intraspecies variability. A fac- tor of 3 was selected for interspecies difference because LC50 values for ClF5 and related compounds were within a factor of 3 among different species (see Sections 1.3 and 1.6). Further, an interspecies uncertainty factor of 3 is appro- priate when the point-of-departure is obtained from data in the most appropriate species (NRC 2001); monkeys were included in the animals tested in the critical study, and monkeys are considered a more appropriate species to predict human toxicity than rodents. An intraspecies uncertainty factor of 3 was also applied; this uncertainty factor is appropriate when the mode of action involves a direct- acting mechanism in which metabolic or physiologic differences are unlikely to play a major role (NRC 2001). The values of the two uncertainty factors are also consistent with those used to derive AEGL values for the related compounds ClF3, HF, and ClO2 (NRC 2004, 2007a,b). Time scaling was performed using the equation Cn × t = k. The 30-min AEGL-2 value was extrapolated from the 10-min point-of-departure of 7 ppm, and the 4- and 8-h AEGL-2 values were extrapolated from the 1-h point of de- parture of 1.7 ppm. An empirical value for the exponent n of 1.9 was deriving from lethality data in the rat (see Section 1.8). The irritation observed in the study used as the basis for the AEGL-2 values are believed to exist on a contin- uum that leads to lung pathology and death at higher concentrations, supporting the use of a time scaling value (n) based on lethality. AEGL-2 values for ClF5 are presented in Table 3-14, the calculations are provided in Appendix B, and a category plot of toxicity data and AEGL values is presented in Appendix C. 

Selected Halogen Fluorides 93 TABLE 3-14 AEGL-2 Values for Chlorine Pentafluoride 10 min 30 min 1h 4h 8h 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm (3.7 mg/m3) (2.1 mg/m3) (0.91 mg/m3) (0.44 mg/m3) (0.30 mg/m3) 2.5. Data Analysis for AEGL-3 2.5.1. Human Data Relevant to AEGL-3 No human data relevant to developing AEGL-3 values for ClF5 were found. 2.5.2. Animal Data Relevant to AEGL-3 Lethality data on ClF5 were available for the monkey, dog, rat, and mouse (Darmer et al. 1972). The 15-, 30-, and 60-min LC50 values and the highest con- centrations that did not result in deaths are presented in Table 3-7. The highest 1-h nonlethal values were 116 ppm in the monkey, 63 ppm in the dog, and 80 ppm in the rat. 2.5.3. Derivation of AEGL-3 Values Analysis of the ClF5 data using benchmark-concentration analysis proved to be inappropriate for developing AEGL-3 values. Using the log-probit model, the 15-, 30-, and 60-min benchmark concentration-derived data failed to follow a concentration-response relationship for any of the four species. For example, the 15-, 30-, and 60-min BMCL05 values for the rat were 72, 146, and 81 ppm, respectively. The failure to follow a concentration-response relationship is most likely due to experimental error encountered with short exposure durations (15 min) as well as from the small data sets (groups of four monkeys and dogs or groups of 10 rats and mice). The LC50 values and the highest exposures to ClF5 resulting in no lethality for all four species follow a more realistic concentration- response relationship (see Table 3-2). The highest 60-min nonlethal concentration in rats of 80 ppm (Darmer et al. 1972) was used as the point-of-departure for the AEGL-3 values for ClF5. Although the rat is not the most sensitive species (mice exhibited the lowest LC50 values of the four species tested; see Table 3-2), the rat data were selected over the mouse data because they were more consistent with the lethality benchmarks observed in monkeys and, thus, would be expected to be a more appropriate species than the mouse to predict human response. The LC50 values in monkeys and rats were 249 and 257 ppm (15 min), 218 and 194 ppm (30 min), and 173 and 122 ppm (60 min), respectively. In contrast, the mouse LC50 values were one-half to one-third of the rat and monkey LC50 values. The rat data were selected over the monkey and dog data because they provided a better dose-response relationship over the 15- to 60-min periods and because a larger 

94 Acute Exposure Guideline Levels number of animals were tested. The longest exposure duration of 60 min was considered the most reliable value. An interspecies uncertainty factor of 3 was selected because differences in LC50 values for ClF5 and related compounds in various test species were within a factor of 3 of each other (see Sections 1.3 and 1.6). Further, an interspecies uncertainty factor of 3 is appropriate when the point-of-departure is obtained from data in the most appropriate species (NRC 2001). The monkey shares greater physiologic similarity with humans than rodents (see Section 1.6), and the monkey data for ClF5 were well-approximated by the rat data used for the point-of-departure. An intraspecies uncertainty factor of 3 was also applied, be- cause the mode of action involves a direct-acting mechanism in which metabolic or physiologic differences are unlikely to play a major role (NRC 2001). As discussed in Section 1.5, ClF5 and related compounds exert toxicity via direct irritation and corrosive action on the respiratory tissues. The values for the two uncertainty factors are also consistent with those used to derive AEGL values for the related compounds ClF3, HF, and ClO2 (NRC 2004; 2007a,b). In sum- mary, the 80 ppm point-of-departure was adjusted by a total uncertainty factor of 10. Time scaling was performed using the equation Cn × t = k. An empirical value for the exponent n of 1.9 was deriving from lethality data in the rat (see Section 1.8). AEGL-3 values for ClF5 are presented in Table 3-15. 2.6. Summary of AEGLs 2.6.1. AEGL Values and Toxicity End Points AEGL values for ClF5 are presented in Table 3-16. TABLE 3-15 AEGL-3 Values for Chlorine Pentafluoride 10 min 30 min 1h 4h 8h 21 ppm 12 ppm 8.0 ppm 3.9 ppm 2.7 ppm (110 mg/m3) (64 mg/m3) (43 mg/m3) (21mg/m3) (14 mg/m3) TABLE 3-16 AEGL Values for Chlorine Pentafluoride Exposure Duration Classification 10 min 30 min 1h 4h 8h AEGL-1 NRa NRa NRa NRa NRa (nondisabling) AEGL-2 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm (disabling) (3.7 mg/m3) (2.1 mg/m3) (0.91 mg/m3) (0.44 mg/m3) (0.30 mg/m3) AEGL-3 21 ppm 12 ppm 8.0 ppm 3.9 ppm 2.7 ppm (lethal) (110 mg/m3) (64 mg/m3) (43 mg/m3) (21 mg/m3) (14 mg/m3) a Not recommended. Absence of AEGL-1 values does not mean that exposures below the AEGL-2 values are without adverse effects. 

Selected Halogen Fluorides 95 2.6.2. Other Standards and Guidelines There are no other exposure standards or guidelines for ClF5. 3. BROMINE PENTAFLUORIDE 3.1. Human Toxicity Data No information on lethality, sublethal effects, neurotoxicity, developmen- tal toxicity, reproductive toxicity, genotoxicity, or carcinogenicity of BrF5 in humans was found. The odor threshold is unknown. According to Braker and Mossman (1980), BrF5 provides adequate warning of its presence by its sharp, penetrating odor. 3.2. Animal Toxicity Data 3.2.1. Acute Lethality Dost et al. (1968) exposed groups of 10-14 male Sprague-Dawley rats to BrF5 at 500 or 1,000 ppm for various durations (see Table 3-17). All 10 rats sur- vived a 40-min exposure at 500 ppm, but 11 of 14 rats exposed at 500 ppm for 50 min died. All 10 rats survived a 20-min exposure at 1,000 ppm, but all 12 rats exposed at 25 min died. Rats were observed for several days following expo- sure. All rats survived additional exposures to BrF5 at 500 ppm for durations shorter than 40 min, and all rats survived additional exposures to 1,000 ppm for durations shorter than 20 min (data not provided). Exposed rats exhibited corro- sive damage to the lungs, corneal and conjunctival damage, yellow and sticky fur, and necrotic damage to unprotected areas of the skin; however, the concen- trations and exposure durations associated with those effects were not reported. In citing their earlier, unpublished experiments on BrF5, Dost et al. (1970) reported a 1-h 95% lethal concentration of 500 ppm in rats. No deaths were re- ported when groups of 4-6 male Sprague-Dawley rats were exposed to BrF5 at 500 ppm for 30 min (half of the 95% lethal exposure duration) in a study evalu- ating systemic fluorine distribution. TABLE 3-17 Lethality Data from Studies of Rats Exposed to Bromine Trifluoride Concentration (ppm) Exposure Duration Effect 1,000 20 min No deaths (0/10) 25 min 100% mortality (12/12) 500 30 min No deaths (0/4-6) 40 min No deaths (0/10) 50 min 79% mortality (11/14) 60 min 95% mortality Source: Data from Dost et al. 1968, 1970. 

96 Acute Exposure Guideline Levels 3.2.2. Nonlethal Toxicity No information on the nonlethal toxicity, neurotoxicity, developmental toxicity, reproductive toxicity, genotoxicity, or carcinogenicity of BrF5 was found. 3.3. Data Analysis for AEGL-1 3.3.1. Human Data Relevant to AEGL-1 No human data relevant to developing AEGL-1 values for BrF5 were found. 3.3.2. Animal Data Relevant to AEGL-1 No animal data relevant to developing AEGL-1 values for BrF5 were found. 3.3.3. Derivation of AEGL-1 Values In the absence of chemical-specific data, no AEGL-1 values were devel- oped for BrF5. 3.4. Data Analysis for AEGL-2 3.4.1. Human Data Relevant to AEGL-2 No human data relevant to developing AEGL-2 values for BrF5 were found. 3.4.2. Animal Data Relevant to AEGL-2 No animal data relevant to developing AEGL-2 values for BrF5 were found. 3.4.3. Derivation of AEGL-2 Values In the absence of data relevant to deriving AEGL-2 values for BrF5, data for the structurally-related chemical, ClF5, were used. The database for ClF5 is more robust than the database for BrF5. ClF5 is considered more toxic than BrF5, on the basis that the highest 60-min nonlethal concentration of ClF5 in the rat is 80 ppm whereas the highest 40-min nonlethal concentration for BrF5 in the rat is 500 ppm. Thus, setting AEGL-2 values for BrF5 that are equal those for ClF5 should be protective. The AEGL-2 values for BrF5 are presented in Table 3-18. 

Selected Halogen Fluorides 97 TABLE 3-18 AEGL-2 Values for Bromine Pentafluoride 10 min 30 min 1h 4h 8h 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm (5.0 mg/m3) (2.8 mg/m3) (1.2 mg/m3) (0.57 mg/m3) (0.41 mg/m3) 3.5. Data Analysis for AEGL-3 3.5.1. Human Data Relevant to AEGL-3 No human data relevant to developing AEGL-3 values for BrF5 were found. 3.5.2. Animal Data Relevant to AEGL-3 A single study provided lethal and nonlethal concentration-exposure dura- tions for the rat. Dost et al. (1968) reported no deaths in rats from exposures to BrF5 at 500 ppm for 40 min or 1,000 ppm for 20 min (Table 3-17). The observa- tion periods were shorter (maximum of 20 h) than the usual 2-week observation period; however, for most strong oxidizing chemicals, death occurs during or shortly after exposure when concentrations approach lethal levels (MacEwen and Vernot 1970; Darmer et al. 1972; Dost et al. 1974). 3.5.3. Derivation of AEGL-3 Values Data were unavailable for calculating a benchmark concentration or an LC01 for BrF5. Therefore, the highest concentrations that resulted in no mortality in rats of 500 ppm for 40 min and 1,000 ppm for 20 min (Dost et al. 1968) were considered. The longer exposure duration of 40 min at 500 ppm was considered more a reliable point-of-departure. A total uncertainty factor of 10 was applied; a factor of 3 for interspecies differences and a factor of 3 for intraspecies varia- bility. Although the data on BrF5 are only from studies of rats, data on related compounds indicate little interspecies differences (within a factor of 3; see Sec- tion 1.6), supporting the selection of an interspecies uncertainty factor of 3. An intraspecies uncertainty factor of 3 was also applied; this uncertainty factor is appropriate when the mode of action involves a direct-acting mechanism in which metabolic or physiologic differences are unlikely to play a major role (NRC 2001). As discussed in Section 1.5, BrF5 and related compounds exert toxicity via direct irritation and corrosive action on the respiratory tissues. The values of the two uncertainty factors are also consistent with those used to de- rive the AEGL values for the related compounds ClF3, HF, and ClO2 (NRC 2004, 2007a,b). Application of a total uncertainty factor of 10 to the point-of- departure of 500 ppm results in a value of 50 ppm. Time scaling was performed using the equation Cn × t = k. Default values for n = 3 for extrapolating to short- er durations and n = 1 for extrapolating for shorter durations were used. The resulting AEGL-3 values are presented in Table 3-19, and the calculations are in Appendix B. 

98 Acute Exposure Guideline Levels Because of the sparse data base on BrF5, application of a modifying factor of 2 was considered when deriving the AEGL-3 values. A modifying factor was not applied because the AEGL-3 values reflect the toxicity of BrF5 relative to that of ClF5 and ClF3. The AEGL-3 values for the slightly more toxic ClF3 are 84 ppm for 10 min, 36 ppm for 30 min, 21 ppm for 1 h, 7.3 ppm for 4 h, and 7.3 ppm for 8 h (NRC 2007a). 3.6. Summary of AEGLs 3.6.1. AEGL Values and Toxicity End Points The AEGL values for BrF5 are presented in Table 3-20. 3.6.2. Other Standards and Guidelines Bromine pentafluoride has limited uses, and only a few exposure stand- ards and guidelines have been developed (see Table 3-21). The threshold limit value–time-weighted average established by the American Conference of Gov- ernmental Industrial Hygienists (ACGIH 2001, 2012) of 0.1 ppm was based on toxicologic analogy with ClF3, which at the time of the recommendation (1969) had a TLV-ceiling of 0.1 ppm. The recommended exposure limit–time-weighted average (REL-TWA) of the National Institute for Occupational Safety and Health (NIOSH 2011) for BrF5 is also 0.1 ppm. NIOSH has not established a concentration that is immediately dangerous to life or health, and the Occupa- tional Safety and Health Administration has not established a permissible expo- sure limit for BrF5. TABLE 3-19 AEGL-3 Values for Bromine Pentafluoride 10 min 30 min 1h 4h 8h 79 ppm 55 ppm 33 ppm 8.3 ppm 4.2 ppm (570 mg/m3) (390 mg/m3) (240 mg/m3) (59 mg/m3) (30 mg/m3) TABLE 3-20 AEGL Values for Bromine Pentafluoride Exposure Duration Classification 10 min 30 min 1h 4h 8h AEGL-1 NRa NRa NRa NRa NRa (nondisabling) AEGL-2 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm (disabling) (5.0 mg/m3) (2.8 mg/m3) (1.2 mg/m3) (0.57 mg/m3) (0.41 mg/m3) AEGL-3 79 ppm 55 ppm 33 ppm 8.3 ppm 4.2 ppm (lethal) (570 mg/m3) (390 mg/m3) (240 mg/m3) (59 mg/m3) (30 mg/m3) a Not recommended. Absence of AEGL-1 values does not mean that exposures below the AEGL-2 values are without adverse effects. 

Selected Halogen Fluorides 99 TABLE 3-21 Standards and Guidelines for Bromine Pentafluoride Exposure Duration Guideline 10 min 30 min 1h 4h 8h AEGL-1 NR NR NR NR NR AEGL-2 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm AEGL-3 79 ppm 55 ppm 33 ppm 8.3 ppm 4.2 ppm a TLV-TWA (ACGIH) – – – – 0.1 ppm REL-TWA (NIOSH)b – – – – 0.1 ppm c MAC (The Netherlands) – – – – 0.1 ppm a TLV-TWA (threshold limit value – time-weighted average, American Conference of Governmental Industrial Hygienists) (ACGIH 2012) is the time-weighted average con- centration for a normal 8-h workday and a 40-h workweek, to which nearly all workers may be repeatedly exposed, day after day, without adverse effect. b REL-TWA (recommended exposure limit – time-weighted average, National Institute for Occupational Safety and Health) (NIOSH 2011) is the time-weighted average concen- tration for up to a 10-h workday during a 40-h workweek. c MAC (maximaal aanvaarde concentratie [maximal accepted concentration], Dutch Ex- pert Committee for Occupational Standards, The Netherlands (MSZW 2004) is defined analogous to the ACGIH TLV-TWA. 4. BROMINE TRIFLUORIDE 4.1. Human Toxicity Data No reliable data on the toxicity of BrF3 in humans were found. According to Braker and Mossman (1980), concentrations of 50 ppm or more may be fatal in 30 min to 2 h. No reference was provided for this information, so it is of ques- tionable reliability. BrF3 is irritating and corrosive to the skin, eyes, mucous membranes, and respiratory tract (O’Neil et al. 2001). No information on suble- thal effects, neurotoxicity, developmental toxicity, reproductive toxicity, geno- toxicity, or carcinogenicity of BrF3 in humans was found. The odor of BrF3 is considered pungent and choking (Owen 2005), but the odor threshold is un- known. 4.2. Animal Toxicity Data No data on the toxicity of BrF3 in animals were found. According to Brak- er and Mossman (1980), the toxic effects of BrF3 are comparable to those of ClF3, which is considered the most toxic of the halogen fluorides. No reference was provided for this information. No information on sublethal effects, neuro- toxicity, developmental toxicity, reproductive toxicity, genotoxicity, or carcino- genicity of BrF3 in animals was found. 

100 Acute Exposure Guideline Levels 4.3. Data Analysis for AEGL Values 4.3.1. Human Data Relevant to AEGL Values No human data relevant to deriving AEGL-1, AEGL-2, or AEGL-3 values for BrF3 were found. 4.3.2. Animal Data Relevant to AEGL Values No animal data relevant to deriving AEGL-1, AEGL-2, or AEGL-3 values for BrF3 were found. 4.3.3. Derivation of AEGL Values In the absence of chemical-specific data, the AEGL-1, AEGL-2, and AEGL-3 values for BrF3 were based on its structure-activity relationship with other halogen fluorides and set equal to the AEGL values for the more toxic chemical analogue, ClF3. A modifying factor was not applied to the ClF3 AEGL values because BrF3 is expected to be less toxic than ClF3 (see Section 1.3). Based on chemical reactivity and relative toxicity, the chlorine fluorides are ex- pected to be more toxic than the bromine fluorides (see Section 1.3). Thus, bas- ing the BrF3 values on the more toxic ClF3 values was considered to provide reasonable protection. The AEGL values for BrF3 are presented in Table 3-22. 4.3.4. Other Standards and Guidelines There are no other exposure standards or guidelines for BrF3. TABLE 3-22 AEGL Values for Bromine Trifluoride Exposure Duration Classification 10 min 30 min 1h 4h 8h AEGL-1 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm (nondisabling) (0.67 mg/m3) (0.67 mg/m3) (0.67 mg/m3) (0.67 mg/m3) (0.67 mg/m3) AEGL-2 8.1 ppm 3.5 ppm 2.0 ppm 0.70 ppm 0.41 ppm (disabling) (45 mg/m3) (20 mg/m3) (11 mg/m3) (3.9 mg/m3) (2.3 mg/m3) AEGL-3 84 ppm 36 ppm 21 ppm 7.3 ppm 7.3 ppm (lethal) (470 mg/m3) (200 mg/m3) (120 mg/m3) (41 mg/m3) (41 mg/m3) 5. REFERENCES ACGIH (American Conference of Government and Industrial Hygienists). 2001. Docu- mentation of the Threshold Limit Values and Biological Exposure Indices: Bro- 

Selected Halogen Fluorides 101 mine Pentafluoride. American Conference of Government and Industrial Hygien- ists, Cincinnati, OH. ACGIH (American Conference of Governmental Industrial Hygienists). 2012. 2012 Threshold Limit Values and Biological Exposure Indices Based on the Documen- tation of the TLVs for Chemical Substances and Physical Agents and BEIs. ACGIH, Cincinnati, OH. Aigueperse, J., P. Mollard, D. Devilliers, M. Chemla, R. Faron, R. Romano, and J.P. Cuer. 2000. Fluorine Compounds, Inorganic. Ullman’s Encyclopedia of Industrial Chemistry, online edition. New York: John Wiley & Sons. Air Liquide. 2005. Safety Data Sheet: Chlorine Pentafluoride [online]. Available: http:// www.msdshazcom.com/MSDS/A/Air%20Liquide/023_AL_EN_Chlorine%20penta fluoride.pdf [accessed July 15, 2014]. Atwood, D. 2006. Fluorine: Inorganic Chemistry. Encyclopedia of Inorganic Chemistry, online edition. New York: John Wiley & Sons. Bailey, W.I., and A.J. Woytek. 2004. Halogen Fluorides. Kirk-Othmer Encyclopedia of Chemical Technology, online edition. New York: John Wiley & Sons. Braker, W., and A.L. Mossman. 1980. Bromine pentafluoride. In Matheson Gas Data Book, 6th Ed. Lyndhurst, NJ: Matheson. Dalbey, W. 1996. Evaluation of the Toxicity of Hydrogen Fluoride at Short Exposure Times. Petroleum Environmental Research Forum Project 92-09. Stonybrook La- boratories Inc., Pennington, NJ. Dalbey, W., B. Dunn, R. Bannister, W. Daughtrey, C. Kirwin, F. Reitman, A. Steiner, and J. Bruce. 1998. Acute effects of 10-minute exposure to hydrogen fluoride in rats and derivation of a short-term exposure limit for humans. Regul. Toxicol. Pharmacol. 27(3):207-216. Darmer, K.I. 1971. The acute toxicity of chlorine pentafluoride. Paper No. 21. Pp. 291- 300 in Proceedings of the 2nd Annual Conference on Environmental Toxicology, 31 August, 1 and 2 September 1971, AD-751452. AMRL-TR-71-120. Wright- Patterson Air Force Base, OH [online]. Available: http://www.dtic.mil/dtic/tr/full text/u2/751452.pdf [accessed July 15, 2014]. Darmer, K.I., C.C. Haun, and J.D. MacEwen. 1972. The acute inhalation toxicity of chlo- rine pentafluoride. Am. Ind. Hygiene Assoc. J. 33(10):661-668. Dost, F.N., D.J. Reed, A. Finch, and C.H. Wang. 1968. Metabolism and Pharmacology of Inorganic and Fluorine Containing Compounds. AMRL-TR-67-224. Aerospace Medical Research Laboratories, Wright-Patterson Air Force Base, OH [online]. Available: http://www.dtic.mil/dtic/tr/fulltext/u2/681161.pdf [accessed July 15, 2014]. Dost, F.N., D.J. Reed, T.D. Cooper, and C.H. Wang. 1970. Fluorine distribution in rats following acute intoxication with nitrogen and halogen fluorides and with sodium fluoride. Toxicol. Appl. Pharmacol. 17(3):573-584. Dost, F.N., D.J. Reed. V.N. Smith and C.H. Wang. 1974. Toxic properties of chlorine trifluoride. Toxicol. Appl. Pharmacol. 27:527-536. DuPont. 1955. Summary of Toxicological Evaluations of Chlorine Dioxide. Haskell Lab Report No. 80-55. Haskell Laboratory for Toxicology and Industrial Medicine, E.I. du Pont de Nemours and Company, Inc., Wilmington, DE. Horn, H.J., and R.J. Weir. 1955. Inhalation toxicology of chlorine trifluoride. I. Acute and subacute toxicity. A.M.A. Arch. Ind. Health 12(5):515-521. Horn, H.J., and R.J. Weir. 1956. Inhalation toxicology of chlorine trifluoride. II. Chronic toxicity. A.M.A. Arch. Ind. Health 13(4):340-345. 

102 Acute Exposure Guideline Levels HSDB (Hazardous Substances Data Bank). 2007a. Bromine Trifluoride (CAS Reg. No. 7787-71-5). TOXNET, Specialized Information Services, U.S. National Library of Medicine, Bethesda, MD [online]. Available: http://toxnet.nlm.nih.gov/cgi-bin/sis/ htmlgen?HSDB [accessed January 2013]. HSDB (Hazardous Substances Data Bank). 2007b. Bromine Pentafluoride (CAS Reg. No. 7789-30-2). TOXNET, Specialized Information Services, U.S. National Li- brary of Medicine, Bethesda, MD [online]. Available: http://toxnet.nlm.nih.gov/cg i-bin/sis/htmlgen?HSDB [accessed January 2013]. Lide, D.R. 1999. CRC Handbook of Chemistry and Physics, 80th Ed. Boca Raton: CRC Press. Lund, K., J. Ekstrand, J. Boe, P. Sostrand, and J. Kongerud. 1997. Exposure to hydrogen fluoride: An experimental study in humans of concentrations of fluoride in plasma, symptoms, and lung function. Occup. Environ. Med. 54(1):32-37. Lund, K., M. Refsnes, T. Sandstrom, P. Sostrand, P. Schwarze, J. Boe, and J. Kongerud. 1999. Increased CD3 positive cells in bronchoalveolar lavage fluid after hydrogen fluoride inhalation. Scand. J. Work Environ. Health 25(4):326-334. MacEwen, J.D., and E.H. Vernot. 1970. Toxic Hazards Research Unit Annual Technical Report: 1970. AMRL-TR-70-77. AD 714 594. Aerospace Medical Research La- boratory, Wright-Patterson Air Force Base, OH [online]. Available: http://www. dtic.mil/dtic/tr/fulltext/u2/714694.pdf [accessed July 15, 2014]. MacEwen, J.D., and E.H. Vernot. 1971. Toxic Hazards Research Unit Annual Technical Report: 1971. AMRL-TR-71-83. AD-734 543. Aerospace Medical Research La- boratory, Wright-Patterson Air Force Base, OH [online]. Available: http://ntrs.na sa.gov/archive/nasa/casi.ntrs.nasa.gov/19720005411.pdf [accessed July 15, 2014]. MacEwen, J.D., and E.H. Vernot. 1972. Toxic Hazards Research Unit Annual Technical Report: 1972. AMRL-TR-72-62. AD-755 538. Aerospace Medical Research La- boratory, Wright-Patterson Air Force Base, OH [online]. Available: http://www.dt ic.mil/dtic/tr/fulltext/u2/755358.pdf [accessed July 15, 2014]. MacEwen, J.D. and E.H. Vernot. 1973. Toxic Hazards Research Unit Annual Technical Report: 1973. AMRL-TR-73-83. AD-771 025. Aerospace Medical Research Labora- tory, Wright-Patterson Air Force Base, OH. MSZW (Ministerie van Sociale Zaken en Werkgelegenheid). 2004. Nationale MAC-lijst 2004: Broompentafluoride. Den Haag: SDU Uitgevers [online]. Available: http:// www.lasrook.net/lasrookNL/maclijst2004.htm [accessed July 16, 2014]. NIOSH (National Institute for Occupational Safety and Health). 2011. NIOSH Pocket Guide to Chemical Hazards: Bromine pentafluoride [online]. Available: http:// www.cdc.gov/niosh/npg/npgd0065.html[accessed July 16, 2014]. NIOSH/OSHA (National Institute for Occupational Safety and Health and Occupational Safety and Health Administration). 1992. Occupational Safety and Health Guideline for Bromine Pentafluoride [online]. Available: http://www.cdc.gov/niosh/docs/81-1 23/pdfs/0065.pdf [accessed July 15, 2014]. NRC (National Research Council). 1993. Guidelines for Developing Community Emer- gency Exposure Levels for Hazardous Substances. Washington, DC: National Academy Press. NRC (National Research Council). 2001. Standing Operating Procedures for Developing Acute Exposure Guideline Levels for Hazardous Chemicals. Washington, DC: Na- tional Academy Press. NRC (National Research Council). 2004. Hydrogen fluoride. Pp. 123-197 in Acute Expo- sure Guideline Levels for Selected Airborne Chemicals, Vol. 4. Washington, DC: The National Academies Press. 

Selected Halogen Fluorides 103 NRC (National Research Council). 2007a. Chlorine trifluoride. Pp. 53-91 in Acute Expo- sure Guideline Levels for Selected Airborne Chemicals, Vol. 5. Washington, DC: The National Academies Press. NRC (National Research Council). 2007b. Chlorine dioxide. Pp. 11-52 in Acute Expo- sure Guideline Levels for Selected Airborne Chemicals, Vol. 5. Washington, DC: The National Academies Press. O’Neil, M.J., A. Smith, and P.E. Heckelman, eds. 2001. Pp. 233 in The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals, 13th Ed. Whitehouse Station, NJ: Merck. Owen (Owen Compliance Services, Inc.). 2005. Material Safety Data Sheet: Bromine Tri- fluoride [online]. Available: http://www.ocsresponds.com/ref/msds/msds-che.pdf [accessed January 2013]. Rosenholtz, M.J., T.R. Carson, M.H. Weeks, F. Wilinski, D.F. Ford, and F.W. Oberst. 1963. A toxicopathologic study in animals after brief single exposures to hydrogen fluoride. Am. Ind. Hyg. Assoc. J. 24:253-261. Schreider, J.P. 1986. Comparative anatomy and function of the nasal passages. Pp. 1-25 in Toxicology of the Nasal Passages, C.S. Barrow, ed. New York: McGraw-Hill. Syage, J.A. 1994. Launch Safety, Toxicity, and Environmental Effects of the High Per- formance Oxidizer ClF5. ADA-A286 095. The Aerospace Corporation, El Segun- do, CA [online]. Available: http://oai.dtic.mil/oai/oai?verb=getRecord&metadata Prefix=html&identifier=ADA286095 [accessed July 15, 2014]. Teitelbaum, D.T. 2001. The halogens. Pp. 731-825 in Patty’s Toxicology, 5th Ed., Vol. 3, E. Bingham, B. Cohrssen, and C.H. Powell, eds. New York: John Wiley & Sons. ten Berge, W.F., A. Zwart, and L.M. Appelman. 1986. Concentration-time mortality response relationship of irritant and systemically acting vapours and gases. J. Haz- ard. Mater. 13(3):301-309. Weinberg, M.S., and R.E. Goldhamer. 1967. Pharmacology and Metabolism of Com- pound A. AMRL-TR-88-238. AD830412. Food and Drug Research Labs, Inc., Maspeth, NY. Weiss, G. 1980. Hazardous Chemicals Data Book. Park Ridge, NJ: Noyes Data Corpora- tion. Wohlslagel, J., L.C. DiPasquale and E.H. Vernot. 1976. Toxicity of solid rocket motor exhaust: Effects of HCl, HF, and alumina on rodents. J. Combust. Toxicol. 3:61- 69. 

104 Acute Exposure Guideline Levels APPENDIX A TIME-SCALING CALCULATION FOR CHLORINE PENTAFLUORIDE Best Fit Concentration x Time Curve 2.45 2.4 2.35 Log Concentration (ppm) 2.3 2.25 2.2 2.15 2.1 2.05 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Log Time (minutes) FIGURE A-1 LC50 values for chlorine pentafluoride in the rat. Source: Darmer et al. 1972. TABLE A-1 Oxygen Difluoride Lethality in Rats Time Conc. Log Time Log Conc. Regression Output: 15 257 1.1761 2.4099 Intercept 3.0552 30 194 1.4771 2.2878 Slope -0.5374 60 122 1.7782 2.0864 R Squared 0.9804 Correlation -0.9901 Degrees of Freedom 1 Observations 3 n = 1.86 1.27 k = 483897 245.78 Source: Lester and Adams 1965; Davis 1970. 

Selected Halogen Fluorides 105 APPENDIX B DERIVATION OF AEGL VALUES FOR SELECTED HALOGEN FLUORIDES Chlorine Pentafluoride Derivation of AEGL-1 Values AEGL-1 values are not recommended for ClF5 due to inadequate warning properties. Derivation of AEGL-2 Values Key study: MacEwen, J.D., and E.H. Vernot. 1973. Toxic Hazards Research Unit Annual Technical Report. AMRL-TR-73-83. AD-771 025. Aerospace Medical Research Laboratory, Wright-Patterson Air Force Base, OH. Toxicity end point: No-effect levels for escape-impairing irritation (7 ppm for 10 min in rats, and 1.7 ppm for 1 h in monkeys, dogs, rats and mice; latter value was extrapolated from an effect level of 5 ppm by dividing it by a modifying factor of 3). Time scaling: Cn × t = k; n = 1.9 (see Appendix A for calculation of n) For 30-min value: (7 ppm ÷ 10)1.9 × 10 min = 5.078 ppm-min For the 4- and 8-h values: (1.7 ppm ÷ 10)1.9 × 1 h = 0.17 ppm-h Uncertainty factors: 3 for interspecies differences 3 for intraspecies variability Modifying factor: 3 to extrapolate from an effect level to a no- effect level (4-h and 8-h AEGL-2 values) Calculations: 10-min AEGL-2: 7 ppm ÷ 10 = 0.70 ppm 30-min AEGL-2: (5.078 ppm-min ÷ 30 min)1/1.9 = 0.39 ppm 1-h AEGL-2: 1.7 ppm ÷ 10 = 0.17 ppm 

106 Acute Exposure Guideline Levels 4-h AEGL-2: (0.17 ppm-h ÷ 4 h)1/1.9 = 0.082 ppm 8-h AEGL-2: (0.17 ppm-h ÷ 8 h)1/1.9 = 0.057 ppm Derivation of AEGL-3 Values Key study: Darmer, K.I., C.C. Haun, and J.D. MacEwen. 1972. The acute inhalation toxicity of chlorine pentafluoride. Am. Ind. Hygiene Assoc. J. 33(10):661-668. Toxicity end point: Highest nonlethal 1-h concentration, 80 ppm in the rat. Time scaling: Cn × t = k; n = 1.9 (see Appendix A for calculation of n) (80 ppm ÷ 10)1.9 × 60 min = 3,119 ppm-min Uncertainty factors: 3 for interspecies differences 3 for intraspecies variability Calculations: 10-min AEGL-3: (3,119 ppm-min ÷ 10 min)1/1.9 = 21 ppm 30-min AEGL-3: (3,119 ppm-min ÷ 30 min)1/1.9 = 12 ppm 1-h AEGL-3: (3,119 ppm-min ÷ 60 min)1/1.9 = 8.0 ppm 4-h AEGL-3: (3119 ppm-min ÷ 240 min)1/1.9 = 3.9 ppm 8-h AEGL-3: (3,119 ppm-min ÷ 480 min)1/1.9 = 2.7 ppm Bromine Pentafluoride Derivation of AEGL-1 Values AEGL-1 values are not recommended for BrF5 because of insufficient data. Derivation of AEGL-2 Values No human or animal data relevant to deriving AEGL-2 values for BrF5 were available. AEGL-2 values were set equal to those for the related com- pound, ClF5. 

Selected Halogen Fluorides 107 10-min AEGL-2: 0.70 ppm 30-min AEGL-2: 0.39 ppm 1-h AEGL-2: 17 ppm 4-h AEGL-2: 0.082 ppm 8-h AEGL-2: 0.057 ppm Derivation of AEGL-3 Values Key study: Dost, F.N., D.J. Reed, T.D. Cooper, and C.H. Wang. 1970. Fluorine distribution in rats following acute intoxication with nitrogen and halogen fluorides and with sodium fluoride. Toxicol. Appl. Pharmacol. 17(3):573-584. Toxicity end point: Highest nonlethal 40-min concentration, 500 ppm in the rat. Uncertainty factors: 3 for interspecies differences 3 for intraspecies variability Time scaling: Cn × t = k; default values of n = 3 for extrapolating to shorter durations and n = 1 for extrapolating to longer durations (NRC 2001). (500 ppm ÷ 10)3 × 40 min = 5.0 × 106 ppm-min (500 ppm ÷ 10)1 × 40 min = 2.0 × 103 ppm-min Modifying factor: None applied Calculations: 10-min AEGL-3: C = ([5.0 × 106 ppm-min] ÷ 10 min)1/3 C = 79 ppm 30-min AEGL-3: C = ([5.0 × 106 ppm-min] ÷ 30 min)1/3 C = 55 ppm 1-h AEGL-3: C = (2.0 × 103 ppm-min) ÷ 60 min C = 33 ppm 4-h AEGL-3: C = (2.0 × 103 ppm-min) ÷ 240 min C = 8.3 ppm 8-h AEGL-3: C = (2.0 × 103 ppm-min) ÷ 480 min C = 4.2 ppm 

108 Acute Exposure Guideline Levels Bromine Trifluoride Derivation of AEGL-1 Values No human or animal data were available on BrF5. Therefore, AEGL-1 values were set equal to those for the related compound, ClF3 (see NRC 2007a). Appendix E provides a summary of how the AEGL-1 values for ClF3 were de- rived. 10-min AEGL-1: 0.12 ppm 30-min AEGL-1: 0.12 ppm 1-h AEGL-1: 0.12 ppm 4-h AEGL-1: 0.12 ppm 8-h AEGL-1: 0.12 ppm Derivation of AEGL-2 Values No human or animal data were available on BrF3. AEGL-2 values were set equal to those for the related compound, ClF3 (see NRC 2007a). Appendix E provides a summary of how the AEGL-2 values for ClF3 were derived. 10-min AEGL-2: 8.1 ppm 30-min AEGL-2: 3.5 ppm 1-h AEGL-2: 2.0 ppm 4-h AEGL-2: 0.70 ppm 8-h AEGL-2: 0.41 ppm Derivation of AEGL-3 Values No human or animal data were available on BrF3. AEGL-3 values were set equal to those for the related compound, ClF3 (see NRC 2007a). Appendix E provides a summary of how the AEGL-3 values for ClF3 were derived. 10-min AEGL-3: 84 ppm 30-min AEGL-3: 36 ppm 

Selected Halogen Fluorides 109 1-h AEGL-3: 21 ppm 4-h AEGL-3: 7.3 ppm 8-h AEGL-3: 7.3 ppm 

110 Accute Exposure Guideline Levels AP PPENDIX C CATEGORY C PLOTS P FOR SELECTED S H HALOGEN FLU UORIDES Chlorin ne Pentafluoriide FIGUR RE C-1 Category y plot of toxicity y data and AEGL L values for chloorine pentafluoridde. 

TABLE C-1 Data Used in Category Plot for Chlorine Pentafluoride Source Species ppm Time (min) Category Comments AEGL-2 0.70 10 AEGL AEGL-2 0.39 30 AEGL AEGL-2 0.17 60 AEGL AEGL-2 0.082 240 AEGL AEGL-2 0.057 480 AEGL AEGL-3 21 10 AEGL AEGL-3 12 30 AEGL AEGL-3 8.0 60 AEGL AEGL-3 3.9 240 AEGL AEGL-3 2.7 480 AEGL Darmer et al. 1972 Monkey 165 15 2 Severe signs of irritation. 249 15 SL LC50 198 30 2 Severe signs of irritation. 218 30 SL LC50 116 60 2 Severe signs of irritation. 173 60 SL LC50 Darmer et al. 1972 Dog 168 15 2 Severe signs of irritation. 298 15 SL LC50 102 30 SL 25% mortality. 63 60 2 Severe signs of irritation. 122 60 SL LC50 (Continued) 111 

112 TABLE C-1 Continued Source Species ppm Time (min) Category Comments Darmer et al. 1972 Rat 175 15 SL 10% mortality. 163 30 2 Severe signs of irritation. 194 30 SL LC50 80 60 2 Severe signs of irritation. 122 60 SL LC50 Darmer et al. 1972 Mouse 100 15 SL 25% mortality. 102 30 SL 25% mortality. 35 60 SL 10% mortality. MacEwen and Vernot 1972 Monkey, rat, mouse 10 60 2 Severe irritation, pulmonary pathology. 20 30 2 Severe irritation, pulmonary pathology. 30 10 2 Severe irritation, pulmonary pathology. MacEwen and Vernot 1972 Monkey, dog, rat, mouse 5 60 1 Irritation, discomfort. 10 30 1 Irritation, discomfort. 30 10 1 Irritation, discomfort. MacEwen and Vernot 1972 Rat 3 10 0 No obvious irritation. 7 10 1 Slight irritation. Weinberg and Goldhamer 1967 Rat 400 10 3 100% mortality. For category: 0 = no effect, 1 = discomfort, 2 = disabling, SL = some lethality, 3 = lethality. Severe signs of irritation might include salivation, lacrimation, rhinorrhea, and nausea. 

Selecteed Halogen Flu uorides 1113 FIGUR RE C-2 Category y plot of toxicity y data and AEGL L values for brom mine pentafluoride. TABLE E C-2 Data Useed in Category Plot P for Chlorinne Pentafluoridee Source Species ppm Time (min) Category Com mments AEGL-2 0.70 10 AEGL AEGL-2 0.39 30 AEGL AEGL-2 0.17 60 AEGL AEGL-2 0.082 240 AEGL AEGL-2 0.057 480 AEGL AEGL-3 79 10 AEGL AEGL-3 55 30 AEGL AEGL-3 33 60 AEGL AEGL-3 8.3 240 AEGL AEGL-3 4.2 480 AEGL Dost et al. a 1970 Rat 500 30 2 No mortality. 500 40 2 No mortality. 500 50 SL 79% %, mortality. 500 60 SL 95% % mortality. 1,000 20 2 No mortality. 1,000 25 3 1000% mortality. For cateegory: 0 = no efffect, 1 = discomffort, 2 = disablinng, SL = some leethality, 3 = lethhal- ity. 

114 Acute Exposure Guideline Levels APPENDIX D ACUTE EXPOSURE GUIDELINE LEVELS FOR SELECTED HALOGEN FLUORIDES Derivation Summary for Chlorine Pentafluoride AEGL-1 Values for Chlorine Pentafluoride Data relevant to AEGL-1 values for ClF5 involve only a 10-min exposure. Because the AEGL-1 values that would be derived from those data are similar to the 8-h AEGL-2 values, AEGL-1 values are not recommended. AEGL-2 Values for Chlorine Pentafluoride 10 min 30 min 1h 4h 8h 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm Key reference: MacEwen, J.D. and E.H. Vernot. 1973. Toxic Hazards Research Unit Annual Technical Report: 1973. AD-771 025; AMRL-TR-73-83; Aerospace Medical Research Laboratory, Wright-Patterson Air Force Base, OH. Test species/Strain/Sex/Number: Monkey, rhesus, males and females, 6 Dog, beagle, sex not specified, 8 Rat, Sprague-Dawley, male, 30 Mouse, ICR, male, 30 Exposure route/Concentration/Duration: Inhalation; 7 ppm for 10 min and 1.7 ppm for 60 min Effects: Slight moistening of eyes in rats exposed to ClF5 at 7 ppm for 10 min. Salivation, lacrimation, ocular irritation, and rhinorrhea (assumed to potentially impair escape) were observed in all species exposed at 30 ppm for 10 min, 10 ppm for 30 min, and 5 ppm for 60 min. The 1-h effect level of 5 ppm was reduced by a modifying factor of 3 (yielding a value of 1.7 ppm) to estimate a no-effect level for escape impairment. End point/Concentration/Rationale: No-effect level for irritation (7 ppm for 10 min and 1.7 ppm for 60 min) Uncertainty factors/Rationale: Total uncertainty factor: 10 Interspecies: 3, because interspecies variability in LC50 values ClF5 and related compounds was within a factor of 3 of each other. Further, an interspecies uncertainty factor of 3 is appropriate when the point-of-departure is obtained from data in the most appropriate species (NRC 2001); monkeys were included in the animals tested in the critical study, and monkeys are considered a more appropriate species to predict human toxicity than rodents. Intraspecies: 3, uncertainty factor is appropriate when the mode of toxic action involves a direct-acting mechanism in which metabolic or physiologic differences are unlikely to play a major role (NRC 2001). ClF5 and related compounds exert toxicity via direct irritation and corrosive action on the respiratory tissues. The values of the two uncertainty factors are consistent with those used to derive AEGL values for the related compounds ClF3, HF, and ClO2 (NRC 2004, 2007a,b). (Continued) 

Selected Halogen Fluorides 115 AEGL-2 Values for Chlorine Pentafluoride Continued Modifying factor: 3, to extrapolate from an effect level of 5 ppm for 60 min to a no-effect level of 1.7 ppm for escape-impairing irritation symptoms. Animal-to-human dosimetric adjustment: Insufficient data. Time scaling: Cn × t = k; n = 1.9, determined on the basis of the time-concentration relationship for LC50 values reported by Darmer et al. (1972) in rats exposed for 15, 30, and 60 min. Irritation symptoms observed in the study are believed to exist on a continuum that leads to pathologic effects in the lungs and death at higher concentrations, supporting the use lethality data to determine the value for n. Data adequacy: The acute toxicity of ClF5 has been well-studied in four species of animal for durations up to 1 h, although some of the studies lacked histopathologic data. There are no data on the toxicity of ClF5 for exposures longer than 1 h. Considered collectively with the toxicity data on the related halogenated compounds ClF3, HF, and ClO2, the data on ClF5 provide a reasonable basis for deriving AEGL-2 values; however, additional studies of ClF5 exposure for durations of 1-8 h would enhance the basis of the 4- and 8-h AEGL-2 values. AEGL-3 Values for Chlorine Pentafluoride 10 min 30 min 1h 4h 8h 21 ppm 12 ppm 8.0 ppm 3.9 ppm 2.7 ppm Key reference: Darmer, K.I., C.C. Haun, and J.D. MacEwen. 1972. The acute inhalation toxicity of chlorine pentafluoride. Am. Ind. Hyg. Assoc. J. 33(10):661-668. Test species/Strain/Sex/Number: Rat, Sprague-Dawley, male, 10 per group Exposure route/Concentration/Duration: Inhalation; 175, 235, 258, 300, 325, 373, 432 ppm for 15 min; 120, 163, 185, 190, 233, 250 ppm for 30 min; and 80, 100, 120, 136 ppm for 60 min. Effects: Highest nonlethal concentrations and calculated LC50 Duration Highest nonlethal concentration LC50 15 min Not identified 257 ppm 30 min 163 ppm 194 ppm 60 min 80 ppm 122 ppm 1-h calculated BMCL05 = 81 ppm End point/Concentration/Rationale: Highest nonlethal concentration in the rat was considered the threshold for lethality (80 ppm for 1 h) Total uncertainty factor: 10 Interspecies: 3, because interspecies variability in LC50 values ClF5 and related compounds was within a factor of 3 of each other. Further, an interspecies uncertainty factor of 3 is appropriate when the point-of-departure is obtained from data in the most appropriate species (NRC 2001); monkeys were included in the animals tested in the critical study, and monkeys are considered a more appropriate species to predict human toxicity than rodents. Intraspecies: 3, uncertainty factor is appropriate when the mode of toxic action involves a direct-acting mechanism in which metabolic or physiologic differences are unlikely to play a major role (NRC 2001). ClF5 and related compounds exert toxicity via direct irritation and corrosive action on the respiratory tissues. (Continued) 

116 Acute Exposure Guideline Levels AEGL-3 Values for Chlorine Pentafluoride Continued The values of the two uncertainty factors are consistent with those used to derive AEGL values for the related compounds ClF3, HF, and ClO2 (NRC 2004, 2007a,b). Modifying factor: Not applicable Animal-to-human dosimetric adjustment: Insufficient data. Time scaling: Cn × t = k; n = 1.9, determined on the basis of the time-concentration relationship for LC50 values reported in the same study. Irritation symptoms observed in the study are believed to exist on a continuum that leads to pathologic effects in the lungs and death at higher concentrations, supporting the use lethality data to determine the value for n. Data adequacy: The acute toxicity of ClF5 has been well-studied in four species of animal for durations up to 1 h, although some of the studies lacked histopathologic data. There are no data on the toxicity of ClF5 for exposures longer than 1 h. Considered collectively with the toxicity data on the related halogenated compounds ClF3, HF, and ClO2, the data on ClF5 provide a reasonable basis for deriving AEGL-3 values; however, additional studies of ClF5 exposure for durations of 1-8 h would enhance the basis of the 4- and 8-h AEGL-3 values. Derivation Summary for Bromine Pentafluoride AEGL-1 Values for Bromine Pentafluoride Data on BrF5 are insufficient for deriving AEGL-1 values, so no values are recommended. AEGL-2 Values for Bromine Pentafluoride 10 min 30 min 1h 4h 8h 0.70 ppm 0.39 ppm 0.17 ppm 0.082 ppm 0.057 ppm Data adequacy: No human or animal data relevant to deriving AEGL-2 values for BrF5 were available. Therefore, AEGL-2 values were set equal to those for the related com- pound ClF5. AEGL-3 Values for Bromine Pentafluoride 10 min 30 min 1h 4h 8h 79 ppm 55 ppm 33 ppm 8.3 ppm 4.2 ppm Key reference: Dost, F.N., D.J. Reed, A. Finch, and C.H. Wang. 1968. Metabolism and Pharmacology of Inorganic and Fluorine Containing Compounds. AMRL-TR-67-224, AD 681 161. Aerospace Medical Research Laboratories, Wright-Patterson Air Force Base, OH [online]. Available: http://www.dtic.mil/dtic/tr/fulltext/u2/681161.pdf [accessed July 15, 2014]. Test species/Strain/Sex/Number: Rat, Sprague-Dawley, male, 10-12 per group Exposure route/Concentration/Duration: Inhalation, 500 ppm for 30, 40, 50, or 60 min or 1,000 ppm for 20 or 25 min. (Continued) 

Selected Halogen Fluorides 117 AEGL-3 Values for Bromine Pentafluoride Continued Effects: Concentration Time Effect 500 ppm 30 min No deaths 500 ppm 40 min No deaths 500 ppm 50 min 79% mortality 500 ppm 60 min 95% mortality 1,000 ppm 20 min No deaths 1,000 ppm 25 min 100% mortality End point/Concentration/Rationale: The highest non-lethal concentration of 500 ppm for 40 min. Uncertainty factors/Rationale: Total uncertainty factor: 10 Interspecies: 3, although only rats have been used to study BrF5, data on related compounds indicate that there is little interspecies variability (within a factor of 3 of each other), supporting the selection of an interspecies uncertainty factor of 3. Intraspecies: 3, is appropriate when the mode of toxic action involves a direct-acting mechanism in which metabolic or physiologic differences are unlikely to play a major role (NRC 2001). BrF5 and related compounds exert toxicity via direct irritation and corrosive action on the respiratory tissues. The value of the two uncertainty factors is also consistent with those used to derive AEGL values s for the related compounds ClF3, HF, and ClO2 (NRC 2004, 2007a,b). Modifying factor: Not applicable Animal-to-human dosimetric adjustment: Insufficient data. Time scaling: Cn × t = k where n = 3 and 1 for shorter and longer exposure durations, respectively (NRC 2001). Data adequacy: Data on the acute toxicity of BrF5 include a single study (Dost et al. 1968) conducted in male rats exposed to one of two concentrations for durations of 20-60 min. The study provided inadequate information on methods (in particular, duration of follow-up was not specified) and did not include microscopic examination of tissues. Considered collectively with the toxicity data on the related halogenated compounds ClF3, HF, and ClO2, the data provide a reasonable basis for deriving AEGL values for BrF5. However, additional studies would serve to refine the AEGL-3 values, including studies of BrF5 exposure for durations of 1-8 h, studies of the acute toxicity of BrF5 in species other than the rat, and additional data on the concentration-time relationship for BrF5. Derivation Summary for Bromine Trifluoride AEGL-1 Values for Bromine Trifluoride 10 min 30 min 1h 4h 8h 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm Data adequacy: No human or animal data relevant for deriving AEGL-1 values for BrF3 were available. AEGL-1 values were set equal to those for the related compound ClF3 (see NRC 2007a). That approach is considered reasonable because qualitative and (Continued) 

118 Acute Exposure Guideline Levels AEGL-1 Values for Bromine Trifluoride Continued quantitative data on the halogenated fluorides suggest that the BrF3 is likely to act via the same mechanism of toxic action as ClF3, but is expected to be less toxic than ClF3. Thus, a modifying factor was not applied to account for the lack of data. Additional research on the chemical-specific toxicity of BrF3 would allow refinement of the AEGL-1 values. AEGL-2 Values for Bromine Trifluoride 10 min 30 min 1h 4h 8h 8.1 ppm 3.5 ppm 2.0 ppm 0.70 ppm 0.41 ppm Data adequacy: No human or animal data relevant for deriving AEGL-2 values for BrF3 were available. AEGL-2 values were set equal to those for the related compound ClF3 (see NRC 2007a). That approach is considered reasonable because qualitative and quantitative data on the halogenated fluorides suggest that the BrF3 is likely to act via the same mechanism of toxic action as ClF3, but is expected to be less toxic than ClF3. Thus, a modifying factor was not applied to account for the lack of data. Additional research on the chemical-specific toxicity of BrF3 would allow refinement of the AEGL-2 values. AEGL-3 Values for Bromine Trifluoride 10 min 30 min 1h 4h 8h 84 ppm 36 ppm 21 ppm 7.3 ppm 7.3 ppm Data adequacy: No human or animal data relevant for deriving AEGL-3 values for BrF3 were available. AEGL-3 values were set equal to those for the related compound ClF3 (see NRC 2007a). That approach is considered reasonable because qualitative and quantitative data on the halogenated fluorides suggest that the BrF3 is likely to act via the same mechanism of toxic action as ClF3, but is expected to be less toxic than ClF3. Thus, a modifying factor was not applied to account for the lack of data. Additional research on the chemical-specific toxicity of BrF3 would allow refinement of the AEGL-3 values. 

Selected Halogen Fluorides 119 APPENDIX E DERIVATION SUMMARY FOR CHLORINE TRIFLUORIDE (Excerpted from NRC 2007a) AEGL-1 Values for Chlorine Trifluoride 10 min 30 min 1h 4h 8h 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm 0.12 ppm Key reference: Horn, H.J., and R.J. Weir. 1956. Inhalation toxicology of chlorine trifluoride. II. Chronic toxicity. A.M.A. Arch. Ind. Health 13(4):340-345. Test species/Strain/Number: Two dogs and 20 rats, breed and strain not stated. Exposure route/Concentration/Duration: Inhalation: 1.17 ppm, 6 h/day, 5 days/week for 6 months. Effects during first day: Dogs: 1.17 ppm for 6 h - nasal discharge (began within 0 to 45 min) obvious lacrimation (after 3 h) Rats: 1.17 ppm for 6 h - no observed effects. End point/Concentration/Rationale: A concentration of 1.17 ppm for 3 h resulted in no signs of irritation in dogs other than nasal discharge. Nasal discharge is considered to be within the definition of the AEGL-1 (mild sensory irritation). Lacrimation after 3 h of exposure was considered the threshold for notable discomfort. Uncertainty factors/Rationale: Total uncertainty factor: 10 Interspecies: 3 – The dog is a sensitive species for nasal irritation and provides a good model for humans. Dogs exposed to 1.17 ppm showed obvious lacrimation after 3 h yet rats showed no effects at the same concentration for 6 h. Intraspecies: 3 – The concentration at which slight irritation is induced in the general population should not differ greatly. Modifying factor: Not applicable. Animal-to-human dosimetric adjustment: Insufficient data. Time scaling: Not applied; adaptation occurs to the slight sensory irritation that defines the AEGL-1. Data adequacy: Although only two dogs were tested in the key study, the concomitant exposure of 20 rats contributes to confidence in the data. The value was based on the dog, which appeared to be more sensitive to respiratory irritants than the rat. Although no histopathological examinations were performed until the termination of the experiment or death, exposure continued for 56 days (39 exposures) before a death occurred in the treated rats. The hydrolysis of ClF3 potentially produces three moles of hydrogen fluoride (HF). Confidence in the AEGL-1 values is boosted by the fact that the values for ClF3 are one- eighth of the AEGL-1 values for HF. The database for HF is extensive. 

120 Acute Exposure Guideline Levels AEGL-2 Values for Chlorine Trifluoride 10 min 30 min 1h 4h 8h 8.1 ppm 3.5 ppm 2.0 ppm 0.70 ppm 0.41 ppm Key reference: Horn, H.J., and R.J. Weir. 1955. Inhalation toxicology of chlorine trifluoride. I. Acute and subchronic toxicity. A.M.A. Arch. Ind. Health 12(5):515-521. Test species/Strain/Sex/Number: Two dogs and 20 rats, breed and strain not stated. Exposure route/Concentration/Duration: Inhalation: 5.15 ppm for 6 h/day, 5 days/week for 6 months. Effects (observed during the first day) for exposures to 5.15 ppm for 6 h: Dogs: strong irritation (salivation, lacrimation, rhinorrhea, coughing, sneezing) apparent recovery at end of day. Rats: no observed effects. End point/Concentration/Rationale: 5.15 ppm for 6 h resulted in strong signs of irritation (salivation, lacrimation, rhinorrhea, coughing, sneezing) in the dog. These signs and symptoms are consistent with the definition of the AEGL-2 (threshold for irreversible or other serious, long-lasting effects or impaired ability to escape). Following 2 days of exposure to 21 ppm, corneal ulcers were observed. Uncertainty factors/Rationale: Total uncertainty factor: 10 Interspecies: 3 – The dog is a sensitive species for nasal irritation and provides a good model for the human. Dogs exposed to 5.15 ppm showed signs of strong irritation (salivation, lacrimation, rhinorrhea, coughing, sneezing) during a 6 h exposure period yet rats showed no effects at the same concentration for 6 h. Intraspecies: 3 – The concentration that induces irritation among the general population should not vary greatly. Modifying factor: Not applicable Animal to Human Dosimetric Adjustment: Insufficient data. Time scaling: Cn × t = k where n = 1.3; based on the time-concentration relationship for LC50 values in monkeys, rats, and mice for exposure durations of 13.5-222 min (Horn and Weir 1955; MacEwen and Vernot 1970; Dost et al. 1974) . Data adequacy: Although only two dogs were tested in the key study, the concomitant exposure of 20 rats contributes to confidence in the data. The value was based on the dog which appeared to be more sensitive to respiratory irritants than the rat. No histopathological examinations were performed until termination of the experiment or death; exposures continued for 26 days before a death occurred in the treated dogs. AEGL-3 Values for Chlorine Trifluoride 10 min 30 min 1h 4h 8h 84 ppm 36 ppm 21 ppm 7.3 ppm 7.3 ppm Key reference: MacEwen, J.D. and E.H. Vernot. 1970. Toxic Hazards Research Unit Annual Technical Report: 1970. AMRL-TR-70-77. Aerospace Medical Research Laboratory, Wright-Patterson Air Force Base, OH [online]. Available: http://www.dtic.mil/dtic/tr/fulltext/u2/714694.pdf [accessed July 15, 2014]. Test species/Strain/Sex/Number: Male and female rhesus monkeys, 4/exposure group. (Continued) 

Selected Halogen Fluorides 121 AEGL-3 Values for Chlorine Trifluoride Continued Exposure route/Concentration/Duration: Inhalation: 127, 150, 200, 300, or 400 ppm for 1 h. Effects from 1 h exposure: Concentration Mortality 127 ppm: 0/4 150 ppm: 2/4 200 ppm: 1/4 300 ppm: 2/4 400 ppm 4/4 1-h LC50 is 230 ppm (provided in reference) 1-h LC01 could not be calculated End point/Concentration/Rationale: 127 ppm for 1 h, the highest non-lethal value in the monkey, was considered the threshold for lethality, the defined end point for the AEGL-3. Uncertainty Factors/Rationale: Total uncertainty factor: 6 Interspecies: 2 – Based on the similarity in respiratory parameters among primates. In addition, effects were similar among species and LC50 values varied by less than a factor of two for the monkey, rat, and mouse (indicating similar species sensitivity). Intraspecies: 3 – The concentration at which extreme irritation and pulmonary damage may lead to lethality should not differ by more than a factor of 3 among the general population. Modifying factor: Not applicable. Animal-to-human dosimetric adjustment: Insufficient data. Time scaling: Cn × t = k where n = 1.3; based on the time-concentration relationship for LC50 values in monkeys, rats, and mice for exposure durations of 13.5-222 min (Horn and Weir 1955; MacEwen and Vernot 1970; Dost et al. 1974). Data adequacy: The key study was well conducted and documented. LC50 values from several additional studies were within a factor of two for all tested species. Similar values can be derived using the rat data (MacEwen and Vernot 1970; Dost et al. 1974) and a total uncertainty factor of 10.