3
Hydrogen Fluoride1

Acute Exposure Guideline Levels

SUMMARY

Hydrogen fluoride (HF) is a colorless, highly irritating, corrosive gas. Reaction with water is rapid, producing heat and hydrofluoric acid. HF is used in the manufacture of artificial cryolite; in the production of aluminum, fluorocarbons, and uranium hexafluoride; as a catalyst in alkylation processes during petroleum refining; in the manufacture of fluoride salts; and in stainless-steel pickling operations. It is also used to etch glass and as a cleaner in metal finishing processes.

HF is a severe irritant to the eyes, skin, and nasal passages; high concentrations may penetrate to the lungs, resulting in edema and hemorrhage. Data on irritant effects in humans and lethal and sublethal effects in six species of mammal (monkey, dog, rat, mouse, guinea pig, and rabbit) were available for developing acute exposure guideline levels (AEGLs). The

1  

This document was prepared by the AEGL Development Team comprising Sylvia Talmage (Oak Ridge National Laboratory) and National Advisory Committee (NAC) on Acute Exposure Guideline Levels for Hazardous Substances member Larry Gephart (Chemical Reviewer). The NAC reviewed and revised the document and the AEGL values as deemed necessary. Both the document and the AEGL values were then reviewed by the National Research Council (NRC) Subcommittee on Acute Exposure Guideline Levels. The NRC subcommittee concluded that the AEGLs developed in this document are scientifically valid conclusions on the basis of the data reviewed by the NRC and are consistent with the NRC guidelines reports (NRC 1993, 2001).



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 3 Hydrogen Fluoride1 Acute Exposure Guideline Levels SUMMARY Hydrogen fluoride (HF) is a colorless, highly irritating, corrosive gas. Reaction with water is rapid, producing heat and hydrofluoric acid. HF is used in the manufacture of artificial cryolite; in the production of aluminum, fluorocarbons, and uranium hexafluoride; as a catalyst in alkylation processes during petroleum refining; in the manufacture of fluoride salts; and in stainless-steel pickling operations. It is also used to etch glass and as a cleaner in metal finishing processes. HF is a severe irritant to the eyes, skin, and nasal passages; high concentrations may penetrate to the lungs, resulting in edema and hemorrhage. Data on irritant effects in humans and lethal and sublethal effects in six species of mammal (monkey, dog, rat, mouse, guinea pig, and rabbit) were available for developing acute exposure guideline levels (AEGLs). The 1   This document was prepared by the AEGL Development Team comprising Sylvia Talmage (Oak Ridge National Laboratory) and National Advisory Committee (NAC) on Acute Exposure Guideline Levels for Hazardous Substances member Larry Gephart (Chemical Reviewer). The NAC reviewed and revised the document and the AEGL values as deemed necessary. Both the document and the AEGL values were then reviewed by the National Research Council (NRC) Subcommittee on Acute Exposure Guideline Levels. The NRC subcommittee concluded that the AEGLs developed in this document are scientifically valid conclusions on the basis of the data reviewed by the NRC and are consistent with the NRC guidelines reports (NRC 1993, 2001).

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 data were considered adequate for deriving the three AEGL classifications for the five exposure periods. Regression analyses of the reported concentration-exposure durations for lethality in the animal species determined that the relationship between concentration and time is C2×t=k (where C=concentration, t=time, and k is a constant). The AEGL-1 was based on an exposure at 3 parts per million (ppm) (range, 0.85–2.93 ppm) for 1 hour (h), which was the threshold for pulmonary inflammation, as evidenced by an increase in the percentage of several inflammatory parameters such as CD3 cells and myeloperoxidase in the bronchoalveolar lavage fluid of 20 healthy exercising adult subjects (Lund et al. 1999). There were no increases in neutrophils, eosinophils, protein, or methyl histamine at this or the next higher average exposure concentration of 4.7 ppm (range, 3.05–6.34 ppm). There were no changes in lung function and only minor symptoms of irritation at that concentration (Lund et al. 1997). Although healthy adults were tested, several individuals had increased immune factors, indicating atopy. The 3-ppm concentration was divided by an intraspecies uncertainty factor (UF) of 3 to protect susceptible individuals. Because there were no effects on respiratory parameters of healthy adults at concentrations up to 6.34 ppm in the Lund et al. (1997) study and at concentrations up to 8.1 ppm for 6 h/day (d) with repeated exposures in a supporting study (Largent 1960, 1961), the calculated AEGL-1 values will be protective of asthmatic individuals. Although the Lund et al. (1999) study duration was only 1 h, the longer exposures at higher concentrations in the supporting study (Largent 1960, 1961), and the fact that adaptation to mild sensory irritation occurs, support application of the 1-ppm concentration for up to 8 h. The 10-minute (min) AEGL-2 was based on an absence of serious pulmonary or other adverse effects in rats during direct delivery of HF to the trachea at 950 ppm for an exposure period of 10 min (Dalbey 1996; Dalbey et al. 1998a). The reported concentration-exposure value of 950 ppm for 10 min was adjusted by a combined UF of 10–3 for interspecies variation, because the rat was not the most sensitive species in other studies (but direct delivery to the trachea is a sensitive model), and an intraspecies UF of 3 to protect susceptible individuals. The resulting 10-min value clearly is below the serious injury categories of data from tests in monkeys, rats, dogs, mice, guinea pigs, and rabbits. The 30-min and 1-, 4- and 8-h AEGL-2 values were based on a study in which dogs exposed at 243 ppm for 1 h exhibited blinking, sneezing, and coughing (Rosenholtz et al. 1963). Rats exposed at a similar concentration (291 ppm) developed moderate eye and nasal irritation. The next higher

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 concentration (489 ppm for 1 h) resulted in respiratory distress and severe eye and nasal irritation in the rat, signs more severe than those ascribed to AEGL-2. The moderate eye and nasal irritation observed in dogs at 243 ppm was considered the threshold for impaired ability to escape. The 1-h value of 243 ppm was adjusted by a total UF of 10–3 for interspecies variation, because the dog is a sensitive species for sensory irritation, and 3 to protect susceptible individuals. The values were scaled across time using Cn×t=k, where n=2. The n value was derived using concentration-exposure duration relationships from animal lethality studies. It should be noted that the resulting 30-min AEGL-2 of 34 ppm is similar to the 32-ppm concentration that could be tolerated by human subjects for only minutes in the Machle et al. (1934) study. Using a larger total UF such as 30 would reduce the 1-h value to 8 ppm, a concentration that resulted in only slight irritation in healthy adults during repeated, intermittent exposures (Largent 1960, 1961). Because the time-scaled 8-h value of 8.6 ppm was inconsistent with the Largent (1960, 1961) study in which humans subjects inhaling 8.1 ppm intermittently suffered no effects other than slight irritation, the 8-h AEGL-2 was set equal to the 4-h AEGL-2. The 10-min AEGL-3 was based on the reported 10-min lethal threshold of 1,764 ppm reported in orally cannulated rats (Dalbey 1996; Dalbey et al. 1998). That value was rounded to 1,700 ppm and adjusted by UFs of 3 for interspecies differences (LC50 values [concentrations lethal to 50% of subjects] differ by a factor of approximately 2–4 between the mouse and rat) and 3 to protect susceptible individuals. The total UF for the 10-min AEGL-3 was 10. Application of a larger UF would reduce the 10-min AEGL-3 to a value below the 10-min AEGL-2. The 30-min and 1-, 4-, and 8-h AEGL-3 values were derived from a 1-h exposure that resulted in no deaths in mice (Wohlslagel et al. 1976). The data indicated that 263 ppm was the threshold for lethality. A comparison of LC50 values among species indicated that the mouse was the most sensitive species in the lethality studies. The 1-h value of 263 ppm was adjusted by an interspecies UF of 1, because the mouse was the most sensitive species, and an intraspecies UF of 3 to protect susceptible individuals. A modifying factor (MF) of 2 was applied to account for the fact that the highest nonlethal value was close to the LC50 of 342 ppm. The resulting value was scaled to the other AEGL-specified exposure periods using Cn×t=k, where n=2. A total factor of 6 is reasonable and sufficient, because application of a total factor of 20 (3 each for inter- and intraspecies uncertainties and 2 as a MF) would reduce the predicted 6-h AEGL-3 to 5.4 ppm, a concentration below the peak 8.1-ppm concentration that produced only irrita-

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 tion in humans (Largent 1960, 1961). Because HF is well scrubbed at low concentrations, and because the time-scaled 8-h AEGL-3 value of 15 ppm was inconsistent with data from repeated exposures in animal studies, the 8-h value was set equal to the 4-h value. The AEGLs for HF are summarized in Table 3–1. 1. INTRODUCTION HF is a colorless, highly irritating, corrosive gas with a molecular weight of 20.01 and a density of 1.27. It is extremely soluble in water; reaction with water produces heat and forms hydrofluoric acid. At atmospheric pressure, the gas is monomeric; at higher pressures, polymerization takes place, producing a gas of density greater than monomeric HF (Perry et al. 1994). Although HF is lighter than air and would disperse when released, a cloud of vapor and aerosol that is heavier than air may be formed under some release conditions (EPA 1993). Additional chemical and physical properties are listed in Table 3–2. Anhydrous HF is manufactured and used in the United States for the production of aluminum, fluorocarbons, cryolite, and uranium hexafluoride; in solutions used for glass etching, cleaning, stainless steel pickling, and chemical derivatives; as a catalyst for the production of gasoline; and for nuclear applications (EPA 1993; Perry et al. 1994). Recent production data were not located. In 1992, HF was manufactured in the United States by three companies at 10 sites with a total production capacity of 206,000 tons; U.S. production is approximately 90% of capacity. In addition, several aluminum producers make HF for on-site use. In 1991, users and/or producers of HF included 13 fluorocarbon production facilities and approximately 51 petroleum refineries that had HF alkylation units. Due to the phase-out of chlorofluorocarbon production, HF production was expected to fall slightly by 1996 (EPA 1993). Contact of liquid HF with the skin can produce severe burns; the gas is corrosive to the eyes and mucous membranes of the respiratory tract. The acute inhalation toxicity of HF has been studied in several laboratory animal species, and its irritant properties have been studied in human volunteers. Large differences in the concentrations causing the same effects in animal studies indicate that difficulties in measurement techniques were encountered by investigators in some of the early studies, thus limiting the value of their quantitative data. In addition, experimental details and descriptions of effects were inadequate in some of the studies.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 TABLE 3–1 Summary Table of AEGL Values (ppm [mg/m3]) Classification 10 min 30 min 1 h 4 h 8 h End Point (Reference) AEGL-1 (Nondisabling) 1.0 (0.8) 1.0 (0.8) 1.0 (0.8) 1.0 (0.8) 1.0 (0.8) Threshold, pulmonary inflammation in humans (Lund et al. 1997, 1999) AEGL-2 (Disabling) 95 (78) 34 (28) 24 (20) 12 (9.8) 12 (9.8) NOAEL for lung effects in cannulated rats (Dalbey 1996; Dalbey et al. 1998a);a sensory irritation in dogs (Rosenholtz et al. 1963)b AEGL-3 (Lethal) 170 (139) 62 (51) 44 (36) 22 (18) 22 (18) Lethality threshold in cannulated rats (Dalbey 1996; Dalbey et al. 1998a);c lethality threshold in mice (Wohlslagel et al. 1976)d a10-min AEGL-2 value. b30-min and 1-, 4-, and 8-h AEGL-2 values. c10-min AEGL-3 value. d30-min and 1-, 4-, and 8-h AEGL-3 values. Abbreviations: mg/m3, milligrams per cubic meter; ppm, parts per million. 2. HUMAN TOXICITY DATA 2.1. Acute Lethality No data were located regarding human deaths following inhalation-only exposure to HF. However, several studies indicate that humans have died from accidental exposure to hydrofluoric acid (Kleinfeld 1965; Tepperman 1980; Braun et al. 1984; Mayer and Gross 1985; Chan et al. 1987; Chela et al. 1989; ATSDR 1993). These accidents involved acute inhalation of HF in combination with dermal exposure involving severe dermal lesions. Deaths were attributed to pulmonary edema and cardiac arrhythmias, the latter a result of acidosis from pronounced hypocalcemia

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 TABLE 3–2 Chemical and Physical Data for Hydrogen Fluoride Parameter Value Reference Synonyms Hydrofluoric acid gas, anhydrous hydrofluoric acid Budavari et al. 1996 Molecular formula HF Budavari et al. 1996 Molecular weight 20.01 Budavari et al. 1996 CAS Registry Number 7664–39–3 Budavari et al. 1996 Physical state Gas Budavari et al. 1996 Color Colorless Budavari et al. 1996 Solubility in water Miscible in all proportions Perry et al. 1994 Vapor pressure 760 mm Hg at 20°C ACGIH 2002 Density (water=1) 1.27 at 34°C Perry et al. 1994 Melting point −87.7°C Perry et al. 1994 Flammability Not flammable Weiss 1980 Boiling point 19.5°C Perry et al. 1994 Conversion factors 1 ppm=0.82 mg/m3 1 mg/m3=1.22 ppm ACGIH 2002 and hypomagnesemia following dermal fluoride uptake. No doses or exposure levels could be determined. 2.2. Nonlethal Toxicity Ronzani (1909) and Machle et al. (1934) cite early reports in which a concentration of HF at 0.004% (40 ppm) was used in the treatment of tuberculosis. No exposure times were stated. The sharp, irritating odor of HF is noticeable at 0.02–0.13 ppm (Sadilova et al. 1965; Perry et al. 1994). Three groups of investigators studied the irritant effects of acute HF exposures in human volunteers. An additional study reported on exposures over a period of 10–50 d. Studies of industrial exposures and accidental releases were located, but exposure concentrations either were intermittent or were not measured; furthermore, those studies were confounded by the presence of other chemicals.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 2.2.1. Experimental Studies The studies using human volunteers are summarized in Table 3–3. Machle et al. (1934) exposed two male volunteers to concentrations of HF at 0.1, 0.05, and 0.026 mg/L (32, 61, and 122 ppm) for very short exposure periods. Inhalation of HF at 122 ppm produced marked conjunctival and respiratory irritation within 1 min and smarting of the exposed skin. At 61 ppm, eye and nasal irritation were marked, but smarting of the skin was not reported. Irritation of the eyes and nose was mild at 32 ppm, and that concentration was “tolerated” with discomfort. At all concentrations, irritation of the larger airways and a sour taste in the mouth were present. Repeated exposures (undefined) failed to produce adaptation. Collings et al. (1951) subjected two volunteers to an atmosphere containing HF and silicon tetrafluoride during an 8-h work shift; the subjects left the area for 15 min every 2 h and during a lunch break. The average concentration of fluoride during the exposure was 3.8 milligrams per cubic meter (mg/m3) (4.6 ppm); the concentration of HF alone was not measured, but would presumably have been ≤4.6 ppm. According to the authors, “both subjects experienced the anticipated irritant effect of thegases and the remarkably rapid acclimation which is so well known.” No further details on irritant effects were stated. Largent (1960, 1961) exposed five male volunteers (ages 17–46) to variable concentrations of HF for 6 h/d over a period of 10–50 d. Average individual concentrations over the exposure period ranged from 1.42 ppm to 4.74 ppm (average, 3.2 ppm; total range, 0.9–8.1 ppm). Effects were no more severe in two subjects who were exposed at concentrations up to 7.9 ppm and up to 8.1 ppm over a 25 d and 50 d period, respectively, than in the other subjects. Although it was stated that one subject tolerated 1.42 ppm for 15 d (6 h/d) without noticeable effects, exposure of the same subject at 3.39 ppm for 10 d at a later time resulted in redness of the face and, by day 11, some flaking of the skin. The subjects experienced very slight irritation of the eyes, nose, and skin at ≤2 ppm and noted a sour taste in the mouth during the exposures. It is not clear whether the subject exposed at 1.42 ppm for 15 d also experienced those effects. Application of a coating of face cream prior to exposure was found to prevent any discomfort or redness of the shaved facial skin. Any signs of discomfort disappeared after cessation of exposure. Systemic effects were not observed. Two subjects in this study displayed slightly different levels of sensitivity. One subject suffered from a cold for a few days during which there was heightened discomfort. Another subject did not use cosmetic cream.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 TABLE 3–3 Summary of Sensory and Irritant Effects in Humans Concentration (ppm) Exposure Time Effects Reference 0.02–0.13 NA Odor threshold Perry et al. 1994; Amoore and Hautala 1983; Sadilova et al. 1965 0.2–0.7 1 h No to low sensory and upper airway irritation; no change in FEV1, decrease in FVC; no change in components of BAL Lund et al. 1997, 1999 0.85–2.9 1 h No to low sensory and upper airway irritation; no change in FVC, FEV1; BAL showed increase in CD3 cells, lymphocytes, with no increase in neutrophils, eosinophils, protein Lund et al. 1997, 1999 3.0–6.3 1 h No eye irritation, but upper (3/14 subjects) and lower (1/14 subjects) respiratory airway irritation;a no change in FVC, FEV1; BAL showed increase in CD3 cells, lymphocytes, myeloperoxidase, cytokine, with no increase in neutrophils, eosinophils, protein Lund et al. 1997, 1999 1.42 6 h/d, 15 d No noticeable effect (single subject) Largent 1960, 1961 2.59–4.74 (average) 0.9–8.1 (range) 6 h/d, 10–50 d Slight irritation of the skin, nose, and eyes; sour taste in mouth Largent 1960, 1961 4.6 (average)b 3.5–7.1 (range) 7 h Irritant effect followed by adaptation Collings et al. 1951

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 32 3 min “Tolerated” with discomfort; mild irritation of eyes and nose Machle et al. 1934 61 Approx. 1 min Eye and nasal irritation Machle et al. 1934 122 Approx. 1 min Marked eye and respiratory irritation, skin irritation, highest concentration tolerated for >1 minute Machle et al. 1934 aUpper airways: symptoms of eye, nose and throat irritation; lower airways: symptoms of chest tightness, coughing, expectoration, wheezing. bExposure to gaseous HF and silicon tetrafluoride; value expressed as fluoride ion. Abbreviations: BAL, bronchoalveolar lavage fluid; FEV1, forced expiratory volume in one second; FVC, forced vital capacity; NA, not applicable.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 Lund et al. (1995) exposed 15 healthy male volunteers to concentrations at 1.5–6.4 mg/m3 (1.83–7.8 ppm) for 1 h in order to study sensory irritation, as indicated by inflammatory cells in bronchoalveolar lavage (BAL) fluid and changes in pulmonary parameters. HF induced a bronchial inflammatory reaction as indicated by an increase in the fraction of lymphocytes and neutrophils in the BAL fluid. The fraction of CD5 positive cells (cluster determinants, a subpopulation of T cells) increased from a pre-exposure value of 0.6% to 6.3% at 20–24 h after exposure. There were no changes in spirometry measurements. The data were reported in an abstract, and no further details were given. In a more recent publication that appears to be a continuation of the above study, 20 healthy, nonsmoking male volunteers, ages 21–44 y, were exposed to a constant concentration within the range of 0.24–6.34 ppm for 1 h in a 19.2-m3 chamber (Lund et al. 1997). Three subjects per exposure group were exposed twice with a 3-month (mo) interval between exposures. In order to analyze a dose-response relationship, the exposures were divided into ranges of 0.2–0.7 ppm (nine subjects); 0.85–2.9 ppm (seven subjects); and 3.0–6.3 ppm (seven subjects). Two of the subjects had hay fever; one of those and an additional subject had an increased total IgE immunoglobulin level. The exposure groups of these subjects were not identified. The authors stated that the rest of the subjects were not atopic or allergic. Exposure concentrations were monitored by an electrochemical sensor. Exact exposures were measured by collecting air samples on cellulose pads impregnated with sodium formate and analyzed with a fluoride selective electrode. Upper and lower airway and eye irritation were subjectively scored on a scale of 0 (no symptoms) to 5 (severe symptoms). In addition, FEV1 (forced expiratory volume in one second) and FVC (forced vital capacity) were measured before, during (every 15 min), and at the end of the exposures and again at 4 and 24 h post-exposure. Subjects rested during the first 45 min of exposure; during the last 15 min the subjects exercised on a stationary bicycle. Five subjects reported minor upper and lower respiratory symptoms (mild coughing or expectoration and itching of the nose) before entering the chamber. Symptoms increased after the 1-h exposure, but none of the subjects in the lower two exposure groups reported symptom scores of greater than 3. Specific scores for symptoms were not reported in the publication; however, a score of 1–3 was defined as low. The mean FVC was significantly decreased after exposure in the lowest exposure group, from 5.1 liters (L) to 4.8 L. The lack of significant changes in the higher exposure groups makes it unlikely that the change in FVC in the lowest group was a result

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 of chemical exposure. In the highest exposure group, no eye irritation was reported, but three subjects reported upper airway irritation (itching or soreness of the nose or throat) with scores of greater than 3, and one subject reported a lower airway irritation (chest tightness, soreness, coughing, expectoration, or wheezing) with a score of greater than 3. Specific symptoms and actual scores were not reported. The authors noted that lower airway symptoms were not reported to a significant degree in relation to exposure to HF, and none of the subjects had obvious signs of bronchial constriction. The authors note that the study was not blind and that the symptoms may have been overreported; however, the exposed subjects were unaware of the exposure concentration. In a second publication addressing the same study (Lund et al. 1999), the authors reported whether or not changes in BAL fluid components occurred 24 h after 1-h exposures at the above concentrations. In particular, they looked at an inflammatory response as indicated by changes in types of white blood cells and several noncellular components compared with measurements taken 3 weeks (wk) before the exposures. The aspirated BAL was divided into bronchial and bronchoalveolar portions, the latter reflecting the more distal air spaces of the lung. Results were provided in the form of cell differentials (%, median and interquartile ranges), making absolute comparisons difficult. The percentage of CD3-positive cells was significantly increased in the bronchial portions of the BAL in the two higher exposure groups and in the bronchoalveolar portions of the BAL in the highest exposure group (3.0–6.3 ppm). CD (cluster determinant) cells are a subpopulation of T cells (i.e., lymphocytes from the thymus) that are recognizable by a selective monoclonal antibody. Although neutrophils were not increased, myeloperoxidase and interleukin-6 (a cytokine) increased significantly in the bronchial portion in the highest exposure group. There were no dose-response related differences in percentages of lymphocytes, eosinophils, neutrophils, and macrophages among the groups for either portions of the BAL, although for the exposure groups combined, the percentage of lymphocytes increased slightly but significantly and the percentage of macrophages decreased slightly but significantly compared with pre-exposure values in both portions of the BAL. Methyl histamine and intercellular adhesion molecule-1 in the bronchial portion were unchanged and, surprisingly, several protein components, including albumin and total protein, were decreased in the bronchoalveolar portion. Although the authors refer to an inflammatory response, they considered the effects minor and could not identify a clear concentration-response relationship. Increases in neutrophils, eosinophils, mast cells, or serum protein in the

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 APPENDIX A Time-Scaling Calculations for Hydrogen Fluoride AEGLs Data (LC50 values) (Rosenholtz et al. 1963) Time (min) Concentration (ppm) Log time Log concentration 5 4,970 0.6990 3.6964 15 2,689 1.1761 3.4296 30 2,042 1.4771 3.3101 60 1,307 1.7782 3.1163 Regression Output: Intercept 4.0627 Slope −0.5260 R Squared 0.9948 Correlation −0.9974 Degrees of Freedom 2 Observations 4 n=1.9 k=5.3E+07

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 APPENDIX B Derivation of AEGL Values Derivation of AEGL-1 Key study: Lund et al. 1997, 1999 Toxicity end point: Biomarkers of exposure during 1 h exposure of exercising human subjects to several ranges of concentrations. Time-scaling: Not applied; adaptation occurs to the slight effects characterized by the AEGL-1. Uncertainty factors: 3 for differences in human susceptibility. The resulting concentration should be protective of asthmatic individuals because it is below the average (2 ppm) and ranges of concentrations (up to 8.1 ppm) (Largent 1960, 1961) that produced slight to mild irritation in healthy adult male subjects. Calculations: The 3 ppm concentration was divided by the intraspecies UF of 3. The resulting concentration, 1 ppm, was used for all AEGL-1 time points. Derivation of AEGL-2 Key studies: Dalbey 1996; Dalbey et al. 1998a; Rosenholtz et al. 1963 Toxicity end point: Lower respiratory tract effects (10-min value)—the 10-min NOAEL of 950 ppm in orally cannulated rats (Dalbey 1996) Irritation (30-min and 1-, 4-, and 8-h values)—signs of blinking, sneezing, and coughing in dogs exposed at 243 ppm for 1 h (Rosenholtz et al. 1963)

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 Time-scaling: C2×t=k, based on the data of Rosenholtz et al. (1963), where C=243 ppm, t=60 min, UF=10, and n=2   C2/10×t=k k=35429.4 ppm2·mm Uncertainty factors: 10-min AEGL-2 Combined uncertainty factor of 10 3 for interspecies (effects are unlikely to differ between species; LC50 values were similar among species; oral cannulation maximizes the dose to the lower respiratory tract) 3 for intraspecies (oral cannulation maximizes the dose to the lower respiratory tract and is a potentially realistic model for human response to corrosive gases)   30-min and 1-, 4-, and 8-h AEGL-2 Combined uncertainty factor of 10 3 for interspecies (the dog is a sensitive species for sensory irritation) 3 for intraspecies 10-min AEGL-2: 950 ppm/10=95 ppm 30-min AEGL-2: C2/10×30 min=35429.4 ppm2·min C=34.4 ppm 1-h AEGL-2: C=243 ppm/10=24.3 ppm 4-h AEGL-2: C2/10×240 min=35429.4 ppm2·min C=12.2 ppm 8-h AEGL-2: 12 ppm   Time-scaling to the 8-h exposure duration results in a value inconsistent with the human data. Because humans suffered no greater effect than slight irritation during intermittent exposures at 8.1 ppm on a repeated basis (Largent 1961, 1962), the calculated concentra-

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4   tion of 8.6 ppm was considered too low. Therefore, the 8-h AEGL-2 value was set equal to the 4-h value. Derivation of AEGL-3 Key studies: Dalbey 1996; Dalbey et al. 1998a; Wohlslagel et al. 1976 Toxicity end point: Lethality (10-min value)—LC05 in rats (1,764 ppm) (Dalbey 1996) Lethality (30-min and 1-, 4-, and 8-h values)—1-h no-death value in the mouse (263 ppm) (Wohlslagel et al. 1976) Time-scaling: C2×t=k, based on the data of Rosenholtz et al. (1963), where C=263 ppm, t=60 min, UF/MF=6, and n=2   (263 ppm/6)2×60 min=115,281.67 ppm2·min Uncertainty factors: 10-min AEGL-3 Combined uncertainty factor of 10 3 for interspecies (effects are unlikely to differ greatly among species; LC50 values were similar among species; oral cannulation maximizes the dose to the lower respiratory tract) 3 for intraspecies (oral cannulation maximizes the dose to the lower respiratory tract and is a potentially realistic model for human response to corrosive gases)   30-min and 1-, 4-, and 8-h AEGL-3 1 for interspecies (the mouse was the most sensitive species; LC50 values differed by approximately 3 between the rat and mouse and effects are unlikely to differ between species). 3 for intraspecies

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 Modifying factor: 2 to account for the fact that the highest nonlethal value was close to the LC50 (applied to the 30-min, and 1-, 4-, and 8-h values) 10-min AEGL-3: 1,700 ppm (rounded from 1,764)/10=170 ppm 30-min AEGL-3: C2/6×30 min=115,281.67 ppm2·min C=61.9 ppm 1-h AEGL-3: C=263 ppm/6=43.8 ppm 4-h AEGL-3: C2/6×240 min=115,281.67 ppm2·min C=21.9 ppm 8-h AEGL-3: C2/6×480 min=115,281.67 ppm2·min C=15.4 ppm (set equal to the 4-h value of 22 ppm)   The time-scaled 8-h AEGL-2 value of 15 ppm is considered inconsistent with the animal data. Rats survived for 8 d during a 14-d exposure at 5 ppm (Placke et al. 1990). No deaths occurred in groups of four male and female rhesus monkeys inhaling 690 ppm for 1 h (MacEwen and Vernot 1970). In a longer-term study, two rhesus monkeys survived a 50-d exposure at 18.5 ppm, 6–7 h/d for 50 d for a total of 309 exposure hours (Machle and Kitzmiller 1935). Therefore, the 8-h AEGL-3 was set equal to the 4-h AEGL-3.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 APPENDIX C ACUTE EXPOSURE GUIDELINE LEVELS FOR HYDROGEN FLUORIDE (CAS No. 7664–39–3) DERIVATION SUMMARY AEGL-1 10 min 30 min 1 h 4 h 8 h 95 ppm 34 ppm 24 ppm 12 ppm 12 ppm Key references: (1) Lund et al. 1997. Exposure to hydrogen fluoride: an experimental study in humans of concentrations of fluoride in plasma, symptoms, and lung function. Occup. Environ. Med. 54:32–37. (2) Lund et al., 1999. Increased CD3 positive cells in bronchoalveolar lavage fluid after hydrogen fluoride inhalation. Scand. J. Work. Environ. Health 25:326–334. Test species/strain/number: 20 healthy male volunteers Exposure route/concentrations/durations: Inhalation; average concentrations of 0.2–0.7 ppm (n=9), 0.85–2.9 ppm (n=7), and 3.0–9.3 ppm (n=7). Effects: At 0.2–0.7 ppm, no to low sensory irritation; no change in FVC, FEV1; no inflammatory response in bronchoalveolar lavage fluid (BAL). At 0.85–2.9 ppm, no to low sensory irritation; no change in FVC, FEV1; increase in the percentage of CD3 cells and myeloperoxidase in bronchial portion of BAL; no increases in neutrophils, eosinophils, protein, or methyl histamine in BAL. At 3.0–6.3 ppm, low sensory irritation; no change in FVC, FEV1; increase in the percentage of CD3 cells and myeloperoxidase in bronchial portion of BAL; no increases in neutrophils, eosinophils, protein, or methyl histamine in BAL. End point/concentration/rationale: Subthreshold concentration for inflammation of 3 ppm (0.85–2.9 ppm) for 1 h, which was without sensory irritation was chosen as the basis for the AEGL-1. Uncertainty factors/rationale: Total uncertainty factor: 3 Interspecies: Not applicable since human subjects were the test species. Intraspecies: 3. The subjects were healthy adult males. The resulting concentration is far below tested concentrations that did not cause symp toms of bronchial constriction in healthy adults (ranges up to 6.3 ppm [Lund et al. 1997] and 8.1 ppm [Largent 1960, 1961])

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 Modifying factor: Not applicable Animal to human dosimetric adjustment: Not applicable, human data used. Time-scaling: Not applied. AEGL-1 values were calculated by adjusting the 1-h concentration of 3 ppm by a UF of 3. Because the response to slight irritation would be similar at shorter exposure durations, the 10- and 30-min values were set equal to the 1-h concentration. Data adequacy: The values are supported by the earlier study of Largent (1969, 1961) in which five healthy human volunteers were exposed at 1.42–4.74 ppm for 10 to 50 d with no greater effects than slight irritation and reddened facial skin. Effects were no more severe in two individuals who were exposed at concentrations up to 7.9 ppm and 8.1 ppm during some of the exposure days.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 AEGL-2 10 min 30 min 1 h 4 h 8 h 95 ppm 34 ppm 24 ppm 12 ppm 12 ppm Key references: (1) Dalbey, W. 1996. Evaluation of the toxicity of hydrogen fluoride at short exposure times. Petroleum Environmental Research Forum Project 92–09. Performed at Stonybrook Laboratories, Inc., Pennington, NJ. (2) 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. Regulat. Toxicol. Pharmacol. 27:207–216. (3) 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. Amer. Ind. Hyg. Assoc. J. 24:253–261. Test species/strain/gender/number: female Sprague-Dawley rats, 20/exposure group (Dalbey 1996); mongrel dogs, 2/exposure group (Rosenholtz et al. 1963) Exposure route/concentrations/duration: 10-min inhalation exposures of orally cannulated rats to 135, 271, 950, or 1,764 ppm (Dalbey, 1996); 60-min inhalation exposures of mongrel dogs to 157 ppm or 243 ppm (Rosenholtz et al. 1963) Effects: In the 10-min inhalation exposures of orally cannulated rats (Dalbey 1996), at 135 ppm there was no effect; at 271 ppm there was no effect; 950 ppm was set as a no-observed-adverse-effect level (NOAEL) (increase in myeloperoxidase and polymorphonuclear leukocytes in BAL); and 1,764 ppm resulted in death of one of 20 animals. In the 60-min inhalation exposures of mongrel dogs (Rosenholtz et al. 1963), at 157 ppm there was mild eye irritation and sneezing and dry cough that persisted for 2 d; and at 243 ppm there was eye, nasal, and respiratory irritation (blinking, sneezing, and coughing during exposures; cough persisted for 2 d and during exercise for up to 10 d). End point/concentration/rationale: For the 10-min exposure, the NOAEL of 950 ppm in orally cannulated rats was chosen because it addresses the relevant exposure period and represents the highest concentration tested that did not result in death. In addition, direct delivery of hydrogen fluoride to the trachea via cannulation is a sensitive model and simulates 100% mouth-breathing in humans exposed to irritant gases. For longer-term exposures, the study by Rosenholtz et al. (1963) was chosen because it was well designed and used dogs which represent a sensitive model for irritants. The highest exposure

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 level of 243 ppm for 60 min, which resulted in symptoms/effects of great discomfort but is not expected to impair the ability to escape or result in irreversible or long-lasting effects, was chosen as the threshold for AEGL-2 effects. Uncertainty factors/rationale: 10-min AEGL-2 values Total uncertainty factor: 10 Interspecies: 3. A sensitive model was used (orally cannulated rats). Intraspecies: 3. Oral cannulation maximizes the dose to the lungs and is relevant to mouth-breathing humans. 30-min and 1-, 4-, and 8-h AEGL-2 values Total uncertainty factor: 10 Interspecies: 3. A sensitive species was used (other studies with irritant gases show an irritant response in the dog at concentrations that are nonir ritating to rodents). Intraspecies: 3. A greater factor would lower the value to concentrations that were non-irritating in human studies. Modifying factor: Not applicable Animal to human dosimetric adjustment: Insufficient data Time-scaling: Cn×t=k where n=2 was derived based on regression analysis of rat LC50 studies conducted at time periods of 5, 15, 30, and 60 min. A second study using rabbits and guinea pigs and conducted over time periods of 5 min to 6 h resulted in the same value for n (reported in a third study). End points for the second study were both irritation and death. Because the time-scaled 8-h value of 8.6 ppm was inconsistent with the data of Largent (1960, 1961), the 8-h AEGL-2 was set equal to the 4-h value. Data adequacy: Based on the following observations, there is considerable support for the scientific credibility of the AEGL-2 values. Oral cannulation bypasses nasal scrubbing and maximizes the dose to the lung. Two species (rat and dog) were tested but not at the same concentrations. A similar irritant response was observed in the rat at higher test concentrations. According to Alarie (Environ. Health Persp. 42:9–13), one-tenth of the mouse RD50 for irritant chemicals can be tolerated for “hours” by humans. The mouse RD50 is 151 ppm; deriving AEGL-2 values based on the RD50 results in a human 4- or 8-h exposure of 15 ppm, slightly higher than the AEGL-2 values based on irritant effects in the dog. The database for irritant effects of hydrogen fluoride is extensive. Five species were tested over a range of concentrations for time periods of 5 min to 6 h.

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 AEGL-3 10 min 30 min 1 h 4 h 8 h 170 ppm 62 ppm 44 ppm 22 ppm 22 ppm Key References: (1) Dalbey, W. 1996. Evaluation of the toxicity of hydrogen fluoride at short exposure times. Petroleum Environmental Research Forum Project 92–09. Performed at Stonybrook Laboratories, Inc., Pennington, NJ. (2) 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. Regulat. Toxicol. Pharmacol. 27:207–216. (3) 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. Test species/strain/gender/number: female Sprague-Dawley rats, 20/exposure group; female CF-1 mice, 10/exposure group Exposure route/concentrations/durations: 10-min inhalation exposures of orally cannulated rats at 135, 271, 950, or 1,764 ppm. 60-min inhalation exposures of female mice at 263, 278, 324, 381, or 458 ppm Effects: In the 10-min inhalation exposures of orally cannulated rats, at 135 ppm there were no effects; at 271 ppm there was no effect; 950 ppm was set as a no-observed-adverse-effect level (NOAEL) (increase in myeloperoxidase and polymorphonuclear leukocytes in BAL); and 1,764 ppm resulted in the death of one of 20 animals. In the 60-min inhalation exposures of female mice, at 263 ppm there were no deaths; at 278 ppm there were 1/10 deaths; at 324 ppm there were 7/10 deaths; at 381 ppm there were 6/10 deaths; at 458 ppm there were 9/10 deaths. End point/concentration/rationale: For the 10-min AEGL-3, the LC05 of 1,764 ppm was rounded down to 1,700 ppm. Although 1/20 deaths is higher than the usual threshold for the AEGL-3 (1/100 deaths), the oral cannulation model is conservative compared with normal nose breathing as it bypasses nasal scrubbing and maximizes the dose to the lung. No higher concentrations were tested at the 10-min exposure period. The concentration resulting in no deaths in the mouse, 263 ppm, was chosen for the longer exposure periods. This specific data set was selected because, based on LC50 values in several studies, the mouse was the most sensitive of three tested species (monkey, rat, and mouse). Uncertainty factors/rationale: 10-min AEGL-3 values Total uncertainty factor: 10

OCR for page 123
Acute Exposure Guideline Levels for Selected Airborne Chemicals, Volume 4 Interspecies: 3. Based on LC50 values in the same studies, the rat was approximately three times less sensitive than the mouse to the lethal effects of hydrogen fluoride; however, the delivery of hydrogen fluoride directly to the trachea via oral cannulation is a conservative model. Intraspecies: 3. Oral cannulation maximizes the dose to the lungs and is relevant to mouth breathing humans. 30-min and 1-, 4-, and 8-h AEGL-2 values Total uncertainty factor: 3 Interspecies: 1. Based on LC50 values, the mouse was the most sensitive of three tested species; of several studies involving the mouse, this study had the lowest lethal values. Intraspecies: 3. Application of a greater uncertainty factor would reduce concentrations to those found only slightly irritating in human studies. Modifying factor: For 30-min and 1-, 4-, and 8-h AEGL-2 values, 2. The highest non-lethal value was close to the LC50 value Animal to human dosimetric adjustment: Insufficient data Time-scaling: Cn×t=k where n=2 based on regression analysis of rat LC50 studies conducted at time periods of 5, 15, 30, and 60 min. A second study using rabbits and guinea pigs and conducted over time periods of 5 min to 6 h resulted in the same value for n (reported in a third study). End points for the second study were both irritation and death. Because the time-scaled 8-h AEGL-3 value of 15 ppm was inconsistent with results of longer-term studies with monkeys and rodents, the 8-h value was set equal to the 4-h value. Data adequacy: There is considerable support for the AEGL-3 values as the database for hydrogen fluoride is extensive with multiple studies of lethality conducted at several exposure durations and involving five species of mammals (monkey, rat, mouse, guinea pig, and rabbit). Studies with multiple dosing regimens generally showed a clear dose-response relationship. A few longer-term studies were also available and served as supporting data. Tissue and organ pathology indicated that the toxic mechanism was the same across species. Difficulties in maintaining/measuring exposure concentrations were encountered in some of the studies; studies in which these difficulties were described were not used to derive AEGL values.