Acute Exposure Guideline Levels for Selected Airborne Chemicals: Volume 16 (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.

7 Tear Gas (CS)1 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 experience notable discomfort, irritation, or certain asymptomatic, nonsensory 1 This document was prepared by the AEGL Development Team composed of Cheryl Bast (Oak Ridge National Laboratory), Lisa Ingerman (SRC, Inc.), Chemical Manager Glenn Leach (National Advisory Committee [NAC] on Acute Exposure Guideline Levels for Hazardous Substances), and Ernest V. Falke (US 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 con- cluded that the AEGLs developed in this document are scientifically valid conclusions based on the data reviewed by the NRC and are consistent with the NRC guidelines re- ports (NRC 1993, 2001). 309 

310 Acute Exposure Guideline Levels 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. SUMMARY Tear gas is a white crystalline powder with a pepper-like odor. It was first synthesized by Corson and Stoughton in 1928 and is, thus, abbreviated as CS (Corson and Stoughton 1928; US Army Chemical School 2005). CS was devel- oped in the 1950s as a replacement for the chemical incapacitant, 1-chloro- acetophenone (CN), because CS was a much more potent irritant than CN, but was significantly less toxic (WHO 1970; Colgrave and Creasey 1975; Hu et al. 1989). It was adopted for use by the military, and was widely used in the Vi- etnam War (WHO 1970; Hu et al. 1989). It is currently used as an incapacitating agent both by military and law enforcement personnel (HSDB 2005). Upshall (1973) reported that an aerosol concentration of CS at 4 mg/m3 will disperse the majority of rioters within 1 min, and at 10 mg/m3 will deter trained troops. With the exception of more severe cutaneous reactions, recovery from exposure is generally rapid upon exposure to fresh air, generally within 30 min after expo- sure (Ballantyne 1977). CS may be manufactured through carbonyl condensa- tion by combining o-chlorobenzaldehyde and malononitrile (HSDB 2005). Re- cent production data on CS were not found. Human studies did not identify a no-effect level for CS or effects of CS that would be consistent with the definition of AEGL-1. The severity of the ef- fects observed at the lowest tested concentrations in humans (ocular stinging and watering, and nasal, throat, and mouth irritation) exceeded those defined by AEGL-1. Therefore, AEGL-1 values for CS are not recommended. AEGL-2 

Tear Gas (CS) 311 values were based on human exposure to CS at an average concentration of 0.75 mg/m3 for 60 min (Beswick et al. 1972). All five subjects tolerated the exposure, but reported ocular stinging and watering, increased salivation, cough, and face stinging. Some subjects also reported throat irritation (4 subjects), nasal stinging and running (3 subjects), mouth stinging (2 subjects), chest burning (2 subjects), nausea (2 subjects), and headache (2 subjects). An intraspecies uncertainty fac- tor of 3 was applied because contact irritation is a portal-of-entry effect and is not expected to vary widely among individuals. Furthermore, the responses of volunteers with jaundice, hepatitis, or peptic ulcer or who were 50-60 years old were similar to those of “normal” volunteers when exposed at a highly irritating concentration of CS for short durations. The ability to tolerate CS at 14-73 mg/m3 and the recovery time in volunteers with a history of drug allergies, sea- sonal allergies, asthma, or drug sensitivity were similar to normal volunteers; although more severe chest symptoms were reported in the people with pre- existing conditions (Gutentag et al. 1960; Punte et al. 1963). An interspecies uncertainty factor of 1 was applied because the study was conducted in humans. A modifying factor of 3 was also used because the effects observed at 0.75 mg/m3 were considered AEGL-2 effects. Time scaling was not performed be- cause irritation is a function of direct contact with CS and is unlikely to increase with duration of exposure at this level of severity (NRC 2001). AEGL-3 values were based on calculated lethality thresholds for CS at each exposure duration. Rat data from the studies by McNamara et al. (1969), Ballantyne and Callaway (1972), and Ballantyne and Swanston (1978) were used calculate LC01 (lethal concentrations, 1% lethality) values for CS. Calcula- tions were performed using the probit analysis-based dose-response program of ten Berge (2006). Time scaling was performed using the equation Cn × t = k, where the exponent n ranges from 0.8 to 3.5 (ten Berge et al. 1986). An empiri- cal value for n of 0.70 was determined on the basis of the rat data. The 4-h AEGL-3 value was adopted as the 8-h AEGL-3 value because time scaling yielded an 8-h value inconsistent with the AEGL-2 values, which were derived from robust human data. A total uncertainty factor of 10 was applied. A factor of 3 was used to account for interspecies differences, because clinical signs are likely caused by a direct chemical effect on the tissues and this type of portal-of- entry effect is unlikely to vary greatly between species. Furthermore, calculated LCt50 values for different species were all well within a factor of 2 of each other (88,480 mg-min/m3 for rats, 67,200 mg-min/m3 for guinea pigs, 54,090 mg- min/m3 for rabbits, and 50,010 mg-min/m3 for mice) (Ballantyne and Swanston 1978). An uncertainty factor of 3 was used to account for intraindividual varia- bility because contact irritation is a portal-of-entry effect and is not expected to vary widely among individuals. As noted above in support of the AEGL-2 val- ues, a factor of 3 is also supported by the results of studies by Punte et al. (1963) and Gutentag et al. (1960) in subjects with pre-existing conditions. AEGL values for CS are presented in Table 7-1. 

312 Acute Exposure Guideline Levels TABLE 7-1 AEGL Values for Tear Gas Classification 10 min 30 min 1h 4h 8h End Point (Reference) AEGL-1 NRa NRa NRa NRa NRa Insufficient data (nondisabling) AEGL-2 0.083 0.083 0.083 0.083 0.083 Ocular, nasal, and throat (disabling) mg/m3 mg/m3 mg/m3 mg/m3 mg/m3 irritation in humans (Beswick et al. 1972) AEGL-3 140 29 11 1.5 1.5 Threshold for lethality (lethal) mg/m3 mg/m3 mg/m3 mg/m3 mg/m3 (LC01) in rats (McNamara et al. 1969; Ballantyne and Callaway 1972; Ballantyne and Swanston 1978) a Not recommended. Absence of an AEGL-1 value does not imply that exposure below the AEGL-2 value is without adverse effects. The severity of effects observed at the low- est tested concentrations exceeded those defined by AEGL-1. 1. INTRODUCTION CS is a white crystalline powder with a pepper-like odor. It was first syn- thesized by Corson and Stoughton in 1928 (thus, the abbreviation CS) (Corson and Stoughton 1928; US Army Chemical School 2005). It was developed in the 1950s as a replacement for the chemical incapacitant, 1-chloroacetophenone (CN), because CS was a much more potent irritant than CN, but was significant- ly less toxic (WHO 1970; Colgrave and Creasey 1975; Hu et al. 1989). CS was adopted for use by the military, and was widely used during the Vietnam War (WHO 1970; Hu et al. 1989; Smith and Greaves 2002). It is currently used as an incapacitating agent by military and law enforcement personnel (HSDB 2005). It is reported that an aerosol concentration of 4 mg/m3 will disperse the majority of rioters within 1 min, and 10 mg/m3 will deter trained troops (Upshall 1973). With the exception of more severe cutaneous reactions, recovery from exposure is generally rapid upon exposure to fresh air, usually within 30 min after expo- sure (Ballantyne 1977). Because CS is stable when heated and has a low vapor pressure, it requires a means of dispersement (Blain 2003). Different forms of dispersement include the combination of CS with a pyrotechnic compound in a grenade or canister, generating a smoke or fog, and dispersement of a fine powder as an aerosol (WHO 1970; Smith and Greaves 2002). CS1 is a micronized powder formula- tion of CS containing 5% silica gel for dissemination by an explosive burst or dusting apparatus, and CS2 is the same as CS1 except that the CS1 is microen- capsulated with silicone to improve its weather resistance and flow properties (WHO 1970). In controlled studies investigating the toxicologic properties of CS aerosol, CS was disseminated as a 2-10% solution in methylene chloride or acetone by means of a pneumatic atomizing nozzle assembly (Gutentag et al. 1960; Owens and Punte 1963; Punte et al. 1963) or by thermal dispersion by spraying the mol- ten chemical (Gutentag et al. 1960; Punte et al. 1962, 1963). 

Tear Gas (CS) 313 CS may be manufactured through carbonyl condensation by combining o-chlorobenzaldehyde and malononitrile (HSDB 2005). Recent production data of CS were not available. Hydrolysis of CS produces malononitrile and o-chlorobenzaldehyde (NTP 1990). Hydrolysis of CS is relatively rapid, with a half-life of about 15 min at a pH 7, but CS reacts faster with an alkaline solution, having a half-life of about 1 min at a pH of 9 (Blain 2003). CS has a vapor pressure of 3.4 × 10-5 mm Hg; thus, at concentrations greater than 0.35 mg/m3, it will exist in vapor and aerosol forms. CS in the vapor phase will be degraded by reaction with photochemically produced hydroxyl radicals, with an estimated half-life of 110 h. CS in the particulate phase will be removed by wet and dry deposition. The chemical and physical properties of CS are presented in Table 7-2. TABLE 7-2 Chemical and Physical Properties of Tear Gas Parameter Value Reference Common name Tear gas Synonyms CS; o-chlorobenzylidenemalonitrile; HSDB 2005; O’Neil 2-chlorobenzalmalononitrile; et al. 2006 (2-chloro-phenyl)methylene) propanenitrile; 2-chlorobmn; beta, beta-dicyano-o-chlorostyrene CAS registry no. 2698-41-1 O’Neil et al. 2006 Chemical formula C10H5ClN2 O’Neil et al. 2006 Molecular weight 188.6 O’Neil et al. 2006 Physical state White crystalline solid O’Neil et al. 2006 Melting point 95-96°C ACGIH 1991 Boiling point 310-315°C US Army Chemical School 2005 Density (solid) Bulk: 0.24-0.26 g/mL; crystal: 1.04 g/mL US Army Chemical School 2005 Solubility in water (g/L) Sparingly soluble; 2.0 x 10-4 M ACGIH 1991; O’Neil et al. 2006 Vapor density (air =1) 6.5 US Army Chemical School 2005 Vapor pressure 3.4 × 10-5 mm Hg at 20°C US Army Chemical School 2005 Henry’s Law Constant 1.0 × 10-8 atm-m3/mol HSDB 2005 Volatility 0.71 mg/m3 at 25°C US Army Chemical School 2005 Stability/reactivity Combustible material; may burn but does US Army Chemical not ignite readily School 2005 Conversion factors 1 ppm = 7.71 mg/m3 Calculated 1 mg/m3 = 0.13 ppm 

314 Acute Exposure Guideline Levels 2. HUMAN TOXICITY DATA 2.1. Acute Lethality No human acute lethality data on CS were found. 2.2. Nonlethal Acute Toxicity 2.2.1. Experimental Studies In a review article, Blain (2003) reported a TC50 (concentration that caused a perceptible effect on 50% of the population exposed for 1 min) of 0.004 mg/m3 for ocular irritation and 0.023 mg/m3 for airway irritation. An ICT50 (concentration intolerable to 50% of the population exposed for 1 min) was also reported. No further details were presented. A group of male volunteers was exposed to CS aerosol with a mass medi- an diameter (MMD) of 0.9 microns (94 ± 15 mg/m3; 4% larger than 10 microns) or up to 60 microns (85 ± 16 mg/m3; 4% smaller than 20 microns) to assess dif- ferences in ocular and respiratory responses to different particle sizes of CS (Owens and Punte 1963). Six volunteers who had the best ability to tolerate CS were chosen from a group of approximately 50. Subjects wore tightly fitted goggles and a nose and mouth respirator designed to protect against particle sizes less than one micron, and were exposed individually in a wind tunnel with a constant air speed of 5 mph. The exposure protocol was designed to restrict exposure to either the small or large particles to the eyes, to the respiratory sys- tem, or to both the eyes and respiratory system. The wind tunnel was elevated to a height of 5 feet, and a rubber-lined port was installed in the bottom of the duct enabling the subject to insert his head into the airstream of the tunnel and re- move it quickly after the exposure. CS was disseminated from a 2% solution in methylene chloride by means of a pneumatic atomizing nozzle assembly. CS concentrations were determined from air samples collected using filter paper placed on air sampling probes located around the head area (one on top and one on each side at eye level), followed by extraction with ethanol and measurement with ultraviolet spectrophotometry. A modified cascade impactor was used to measure the CS aerosol containing the small particles, while the larger particles were sized microscopically, measuring and counting the various particles in the pre-ground material before dissemination. Tolerance time was defined as the time at which a subject could no longer remain in the atmosphere containing the compound and left the exposure chamber, and recovery time was defined as the time after the exposure when the subjects were able to sort and arrange a series of 24 playing cards from which the corner numbers were removed. Control val- ues were determined before each test. The results indicate that small particles are more effective in rapidly producing ocular irritation (see Table 7-3). The onset of ocular response is hypothesized to be faster with small particles because 

Tear Gas (CS) 315 TABLE 7-3 Tolerance and Recovery Time in Humans Exposed to Tear Gas Particles (1-60 microns) Subjects Tolerating 60-sec Recovery Time (sec) Exposure (%) Exposure Small Particlesa Large Particlesb Small Particlesa Large Particlesb Eyes 40 100 91 280 Respiratory system 0 67 51 9c Eyes and respiratory system 16 85 52 188 a Measured concentration of 94 ± 15 mg/m3. b Measured concentration of 85 ± 16 mg/m3. c 4/6 subjects were able to perform task immediately after exposure. Source: Adapted from Owens and Punte 1963. of they are more soluble than larger particles in ocular fluid. Once begun, how- ever, the irritation process would continue for a longer period with the large particles compared with the small particles. Respiratory effects were more se- vere with small particles (no volunteers could withstand exposure for more than 30 seconds [sec]) and required more time for recovery than the large particles. The difference in response is due to the ability of smaller-sized particles to pene- trate more deeply into the respiratory tract. When both the eyes and the respira- tory system were exposed to CS, the respiratory response predominated with exposure to the small particles, whereas the ocular response predominated with exposure to the large particles. A group of 4-6 volunteers was exposed to CS aerosol in a wind tunnel (8 × 8 × 8 feet; fixed wind speed of 5 mph) (Gutentag et al. 1960; Punte et al. 1963). Volunteers were both military and civilian personnel. Each volunteer’s medical history was recorded, and each was given pre-exposure and post-exposure phys- ical examinations. Volunteers were classified as “normal” or were placed in one of four special categories: those with hypertension (diastolic pressure of 80-110 mm Hg or normal blood pressure reading with a history of hypertension; pre- exposure tests included electrocardiogram, chest X-ray, liver function, and uri- nalysis); those with hay fever, drug sensitivity, or bronchial asthma (volunteers with asthma had normal chest X-ray before exposure); those with a history of jaundice, hepatitis, or peptic ulcers without gastrointestinal bleeding; and those that were 50-60 years of age. Subjects classified as normal were further catego- rized into untrained men with or without protective masks or trained men with or without protective masks. The trained men had previous exposure to CS, whereas the untrained men did not. CS was dispersed as a 10% solution in ace- tone or methylene dichloride with a spray nozzle (MMD 3.0 or 1.0 micron, re- spectively) or by thermal dispersion (spraying the molten chemical; MMD 0.5 micron). Airborne samples of the aerosol were collected at various points in the wind tunnel. Particle size was characterized using a 6-stage modified cascade impactor, and exposure concentrations were measured using ultraviolet spectro- photometry. The subjects did not report any noticeable difference in symptoms from the different dispersion methods. 

316 Acute Exposure Guideline Levels Groups of 3-6 untrained men without masks were exposed to CS in ace- tone, and tolerance times were recorded. Times ranged from 53 to >120 sec at 5 mg/m3, 19-43 sec at 12 mg/m3, and up to 5 sec at 442 mg/m3. When groups of 1- 7 trained men were exposed, tolerance times ranged from 37 to >120 sec at 4 mg/m3, 18-41 sec at 10 mg/m3, and up to 12-25 sec at 141 mg/m3. To compare the effects of hyperventilation on symptoms, untrained subjects ran for approx- imately 100 yards before exposure. Exercising subjects could not tolerate CS as well as normally breathing subjects; groups of three subjects exposed at 10, 13, or 39 mg/m3 could tolerate CS for up to 13, 13, and 9 sec, respectively. While ocular irritation was minimal, chest symptoms were more pronounced and re- covery time was slightly prolonged (by 1-2 min). The reactions of subjects with jaundice, hepatitis, or peptic ulcer or those that were 50-60 years old were simi- lar to those of normal subjects. Subjects with a history of drug allergies or sensi- tivities, hay fever, or asthma also tolerated exposure to CS at concentrations comparable to those tolerated by normal subjects, but the group with pre- existing conditions had a higher percentage of individuals with more severe chest symptoms, with many of them laying prostrate on the ground for several minutes. However, no wheezing or rhonchi were heard, and recovery was as rapid as that seen in other exposure groups. When subjects were exposed to CS at temperatures ranging from 0-95°F, tolerance to the chemical was slightly re- duced at the high temperature of 95°F. Whether the decrease in tolerance was an actual effect of the exposure, the uncomfortable climate, or a combination of both was unclear. An increase in skin-burning symptoms with increased temper- ature was attributed to an increase in perspiration. As part of the study described above, the potential for developing toler- ance to CS was investigated by exposing a group of four subjects to CS at 1.5 mg/m3 for 90 min in a 20,000-L chamber (Punte et al. 1963). No data were pro- vided about the monitoring of the CS aerosol. During exposure, subjects were allowed to smoke, read, and play cards. Only one subject reported nasal irrita- tion (after 2 min), three subjects reported headaches (after 45, 50, and 83 min), and all four subjects reported ocular irritation (after 20, 24, 70, and 75 min). In the second part of the experiment, the four subjects were exposed to CS at 1.5 mg/m3 for 40 min, and then additional CS aerosol was added to the chamber to achieve an airborne concentration of 11 mg/m3 in about 10 min. Although the subjects had not been told of the increase in concentration, they all left within 2 min due to respiratory irritation. The exposure concentration was estimated to be 4.3-6.7 mg/m3 when the subjects left the chamber. In the third part of the exper- iment, the subjects were exposed to CS at 6 mg/m3, which was attained over 10 min. Symptoms reported by the subjects included nasal and throat irritation, chest burning, sneezing, ocular irritation and lacrimation, headache, and dermal irritation. Three of the four subjects reported that the exposure was unbearable after 18, 20, and 29 min, with chest symptoms being the reason the subjects left the chamber. The remaining subject was able to tolerate the agent, and the expo- sure was terminated after 40 min. The investigators attempted to enter the chamber without the benefit of the gradual increase in exposure concentration, 

Tear Gas (CS) 317 and were unable to remain in the chamber. In the fourth experiment, a concen- tration of 6.6 mg/m3 was attained over 30 min. The usual signs and symptoms of CS exposure developed, but to a lesser degree. One of the subjects had to leave after 2 min because of a violent cough, but he returned to the exposure chamber after his cough had ceased upon exposure to fresh air. He remained in the expo- sure chamber for the duration of the 60-min exposure. To assess the potential effect of CS exposure on ventilation, cardiac fre- quency, and breathing pattern, a group of 11 healthy soldier volunteers was ex- posed to CS aerosol (particle diameter of 1 micron) at a concentration that was progressively increased from 0.2 mg/m3 to 1.3 mg/m3 (Cotes et al. 1972; Cole et al. 1977). The exposure duration was not specified, but appeared to be approxi- mately 80 min. CS aerosol was produced by saturating the exposure chamber the evening before the exposure, followed by flushing with air to remove all of the gas except that adsorbed onto the walls and equipment. During exposure, pyro- technic generators were ignited to progressively raise the concentration of CS throughout the exposure session. Subjects wore woolen or denim battle dress covered with cotton coveralls, boots, and gaiters. Electrocardiogram electrodes were applied to the chest, and subjects wore a full respirator into the chamber. For the commencement of exposure, each subject removed his own respirator. During each exposure, each subject completed two 8-min periods of exercise, which consisted of cycling at 20W up to 120W. During exercise, the subjects breathed through an oral-nasal mask and three-way valve box. Inspiration was from the chamber and expiration was through a 6-L capacity mixing bottle into a low resistance gas meter. Cardiac frequency was measured by electrocardio- graph, while a thermister in the valve box recorded respiratory frequency. A control exposure including exercise was conducted the day before and the day after exposure to CS. A major difference between the control and CS exposures was that ventilation was continued throughout the control session but not the CS-exposure session; therefore, the temperature was much higher during the CS-exposure sessions than the control session (~24° vs. 20.5°C for controls). All subjects experienced intense discomfort, including cough, lacrimation, and substernal pain, when first exposed to the CS aerosol. Discomfort was se- vere enough that two subjects withdrew (one before and one after the first period of exercise), and two additional subjects were unable to complete the first period of exercise due to coughing. Coughing coincided with ignition of the CS genera- tors. The discomfort disappeared with continuing exposure. Although cardiac frequency increased during exposure to CS compared with control air, the dif- ference was eliminated when the cardiac frequency was corrected for the in- creased ambient temperature (corrected to the arbitrary temperature of 20°C). The ventilation minute volume was reduced from exposure to CS compared with controls. The reduction appeared to be due to a decrease in respiratory frequen- cy. The exposure was repeated using 17 volunteers (Cole et al. 1975, 1977). Exposure conditions were the same with the following exceptions: the CS can- dles were ignited between and not during periods of exercise, CS concentrations were slightly higher (0.92-2.15 mg/m3), and the subjects were seen on five con- 

318 Acute Exposure Guideline Levels secutive half-day sessions (the first, third, and fifth sessions were for control observations and the other two sessions were allocated one each for exposure to ammonia and to CS [the order of exposure changed between the different weeks of the study]). Results were generally the same as those observed in the first study. The only difference was that the reduction in the ventilation minute vol- ume was the result of a diminution in tidal volume and occurred despite an in- crease in respiratory frequency. To investigate the potential for developing tolerance to CS, 35 healthy male volunteers were exposed for 60 min to increasing concentrations of CS aerosol (Beswick et al. 1972). Exposures were conducted in a 100-m3 chamber. The chamber was generally saturated an hour before the exposure, followed by air being blown through the chamber to remove the CS not absorbed on the walls and equipment. A number of parameters were assessed before and after exposure, including chest radiograph results, hematology and clinical-chemistry analysis, and respiratory-function tests to assess peak flow, tidal volume, and vital capacity. A total of 10 exposure trials were conducted, with no volunteers exposed more than once. Exposure concentrations were kept relatively constant in the first three trials: 0.53-0.86 mg/m3 in trial 1 (three subjects), 0.71-0.78 mg/m3 in trial 2 (five subjects), and 0.31-0.74 mg/m3 in trial 3 (six subjects). For the seven remaining trials, exposure concentrations were increased by a factor of 2, 3, or 4 during the exposure period: 0.8-1.4 mg/m3 in trial 4 (five subjects), 0.84-2.3 mg/m3 in trial 5 (four subjects), 0.7-2 mg/m3 in trial 6 (four subjects), 0.63-2.3 mg/m3 in trial 7 (two subjects), 0.57-2.1 mg/m3 in trial 8 (two subjects); 0.42-1.8 mg/m3 in trial 9 (two subjects), and 0.45-1.7 mg/m3 in trial 10 (two subjects). Chamber concentrations were measured at 10-min intervals. Volun- teers entered the chamber wearing full respirators and protective coveralls. CS was generated and allowed to mix for 3 min before removal of the respirator. Symptoms from all volunteers were reported during individual interviews after exposure. The results of the 10 trials were consolidated into five groups: group I included trial 2 (five subjects), group II included trials 1 and 3 (nine subjects), group III included trial 4 (five subjects), group IV included trials 5 and 6 (eight subjects), and group V included trials 7, 8, 9, and 10 (eight subjects). One of the six subjects in trial 3 left the exposure chamber after 8 min of exposure with complaints of severe stinging of the eyes, throat irritation, cough and dyspnea, salivation, and nausea, and one subject in group IV left after 55 min due to vom- iting. All other subjects remained in the chamber for the entire 60-min exposure period. A summary of the symptoms of the exposed individuals is presented in Table 7-4. The predominant symptoms included excess production of mucus and saliva (34/34 subjects), ocular irritation (stinging in 32/34 subjects and lacrima- tion in 32/34 subjects), runny nose (28/34 subjects), and face stinging (32/34 subjects). Symptoms generally resolved within 10 min of leaving the chamber. Nausea was reported by 11/34 subjects and two vomited, which appeared to follow swallowing of large amounts of saliva. The development of tolerance was assessed in two of the trials (group IV in Table 7-4); in these trials, half of the subjects removed the respirator at the start of the exposure (with the CS concen- 

Tear Gas (CS) 319 tration increasing with time), while the remaining subjects did not remove their respirators until the last 5 min of exposure. The subjects that were exposed to CS throughout the entire exposure period were able to withstand the entire 60- min exposure (concentrations increasing from 0.84 to 2.30 mg/m3 and 0.70 to 2.00 mg/m3) except for the one individual that had to leave the chamber at 55 min because of vomiting. Of the subjects that removed their respirators for the last 5 min of exposure, only one of eight subjects could remain in the chamber for more than 1 min; five left within 30 sec of removing their respirators. No exposure-related changes were observed in hematology or clinical chemistry parameters. Decreases in heart rate after exposure ceased were ascribed to the sense of relief each volunteer felt at the end of an uncomfortable experience, and the increase in systolic blood pressure observed in individuals when exposure commenced was due to the abrupt onset of discomfort; continued exposure re- sulted in normal blood pressure readings. No abnormalities were noted in meas- urements of respiratory function, but the investigator noted that the sample size was small and, thus, may not be representative. It was concluded that the main effects of CS are due to local irritation of exposed nerve endings, and systemic changes noted are due to stress. Three groups of volunteers were exposed to CS aerosol at various concen- trations to investigate potential effects on visual acuity (Rengstorff 1969). The first group was composed of 10 male volunteers exposed to CS2 aerosol (CS treated with Cab-o-Sil® 5 and hexamethyldisilaxane) at concentrations of 0.1 to 1.7 mg/m3. The exposure was conducted in a wind tunnel suspended 4.5 feet above the floor; the volunteer sat on a chair at the end of the wind tunnel and put his head through a rubber aperture in the tunnel until he could no longer tolerate the exposure or for a maximum of 10 min. A powder dispenser disseminated specific concentrations of CS2 (MMD of 0.8 microns) into the air at a wind speed of 4.5 mph. An Orthorater was used to measure the binocular far and near visual acuity of the subjects before and after exposure. The second and third groups were exposed to CS aerosol in a circular steel chamber. CS aerosol (MMD of 0.9 micron) in a methylene dichloride solution was disseminated us- ing a thermal generator, and introduced into the chamber as a uniform cloud. Subjects wore protective masks for the first 5 min in the chamber, and then re- moved their masks for the commencement of exposure. The second group was composed of 34 volunteers, and an Orthorater was again used to measure the binocular far and near visual acuity before and after exposure. A summary of the amount of time volunteers from this group could tolerate exposure to CS is pre- sented in Table 7-5. The third exposure involved 22 volunteers who had a base- line visual acuity of 20/20 and who could remain in the exposure chamber for 10 min. Binocular acuity was measured using a Snellen visual acuity projector be- fore, during, and a few minutes after exposure. The Snellen chart contained a row of 20/30, 20/25, and 20/20 letters. No exposure-related changes in visual acuity were noted except those due to the inability of some subjects to keep their eyes open because of intense ocular irritation. Visual acuity returned to normal in all subjects several minutes after exposure to CS ended. 

320 TABLE 7-4 Symptoms of Volunteers Exposed to Tear Gas for 60 Minutes Exposure Group (nominal-fold increase in concentration during exposure) Symptoms I (steady) II (× 2) III (× 2) IV (× 3) V (× 4) Total Observations Number exposed 5a 8b 5c 8d 8e 34 – Eyes: Stinging 5 8 4 7 8 32 No connection between severity and concentration, but duration of symptoms may be less with steady Watering 5 8 5 7 7 32 concentration. Nose: Stinging 3 4 1 6 4 18 Effects diminished in subjects exposed at steady concentrations or at concentrations that were Running 3 7 3 8 7 28 doubled during the exposure duration. Peppery feeling 2 4 2 4 5 17 Blocked 1 3 2 3 2 11 Mouth Irritation 2 4 – 6 3 15 Copious saliva production; subjects spitting appeared to be vomiting. Salivation 5 8 5 8 8 34 Throat: Irritation 4 5 3 6 5 23 – Dry – 1 – 6 1 8 Chest: Burning 2 2 – 3 1 8 More severe effects appeared to be due to deep breaths taken after holding breath; coughing was sporadic. Tight 1 3 2 2 3 11 Dyspnea – 2 2 2 3 9 Cough 5 2 4 3 4 18 

Nausea 2 3 1 3 2 11 – Vomiting – – – 1 1 2 Likely due to swallowing large quantity of saliva. (during exposure) Subject in Group V vomited within first 5 min and left chamber, but returned for duration of exposure. Subject in Group IV vomited after 55 min. Face stinging 5 7 5 7 8 32 Effects appeared to be of shorter duration in subjects exposed at steady concentrations. Stinging most unpleasant in shaved regions. Headache 2 1 2 1 – 6 Persisted in 3 subjects throughout exposure; three cases reported post-exposure. Appeared to be due to irritation of the frontal sinuses. a Five subjects exposed at 0.71-0.78 mg/m3. b Three subjects exposed at 0.56-0.86 mg/m3, and five subjects exposed at 0.31-0.74 mg/m3 (one subject excluded because he left the exposure chamber after 8 min). c Five subjects exposed at 0.8-1.4 mg/m3. d Four subjects exposed at 0.84-2.3 mg/m3, and four exposed at 0.7-2.0 mg/m3. To assess the development of tolerance, four subjects removed respirators at start of exposure, while the other four removed respirators 5-min before the end of exposure; the subjects that removed respirators at beginning of exposure were able to withstand the entire 60-min exposure, except for one individual that had to leave the chamber after 55 min because of vomiting. Of the subjects that removed their respirators during the last 5 min of exposure, only one of eight could remain in the chamber for more than 1 min and five left within 30 sec of removing their respirators. e Two subjects exposed at 0.63-2.3 mg/m3, two exposed at 0.57-2.1 mg/m3, two exposed at 0.42-1.8 mg/m3, and two exposed at 0.45-1.7 mg/m3. Source: Adapted from Beswick et al. 1972. 321 

322 Acute Exposure Guideline Levels TABLE 7-5 Tolerance of Humans Subjects Exposed to Tear Gas for Up to 10 Minutes in Study Evaluating Visual Acuity Concentration (mg/m3) Number of Subjects Exposure Duration (sec) 0.4 1 135 1 420 1 435 4 600 0.6 1 30 2 35 1 38 1 40 1 65 1 68 1 102 1 105 1 135 7 600 0.9 6 600 1.0 1 35 1 40 1 45 1 50 Source: Adapted from Rengstorff 1969. To assess the effect of CS exposure on respiration, a group of six volun- teers (four with previous exposure to CS) were exposed to various concentra- tions of CS (3 micron) in a wind tunnel while a portable breathing device moni- tored respiration (Craig et al. 1960). Subjects remained in the tunnel until the exposure became intolerable (see Table 7-6). Notable coughing was observed in subjects exposed at 15 mg/m3 for 61 sec or at 150 mg/m3 for 12 sec. On the ba- sis of the recordings made during exposure, it was concluded that although the breathing pattern of the volunteers was disrupted, adequate ventilation was maintained. Therefore, the incapacitation of CS is attributed to the unpleasant sensations of exposure rather than to any degree of respiratory failure. A group of 38 US Marines was exposed to a cloud of CS dispersed by a thermal canister as part of a training exercise to test the ability and speed of the trainees in donning their gas masks (Thomas et al. 2002). The exposure occurred after 6 days of strenuous training with minimal sleep and reduced food con- sumption, and was followed by a 1.5-mile run. Temperature and relative humidi- ty at the time of exposure were approximately 24°C and 91%, respectively. Clin- ical signs and symptoms began to develop 36-84 h post-exposure during and 

Tear Gas (CS) 323 TABLE 7-6 Tolerance of Human Subjects Exposed to Tear Gas in Study Evaluating Respiratory Effects Concentration (mg/m3) Exposure Duration (sec) 5 110 + 12 24 15 61 + 64 15 + 80 12 150 12 + + Previous exposure to CS. Source: Craig et al. 1960. after periods of strenuous exercise (one subject became symptomatic after a 1,000-meter pool swim at 36 h post-exposure; seven became symptomatic after a second swim consisting of a 1,000-meter open ocean swim 60 h post- exposure; and one became symptomatic after a third swimming event consisting of a 1,500-meter open ocean swim 84 h post-exposure). A total of nine subjects were affected, with four requiring admission into intensive care. Effects of ex- posure included dyspnea upon exertion, hemopytosis (ranging from frank blood to blood-tinged sputum), cough, rales, reduced arterial blood gas (range of 60- 68), and infiltrates visible on chest radiograph. Signs and symptoms resolved by 72 h, and lung function before and after exercise challenge returned to normal within 1 week post-exposure. When the exposure in this study was recreated (without test subjects) and air sampling was performed, CS concentrations were found to range from less than quantifiable to approximately 17 mg/m3. McDonald and Mahon (2002) proposed that the pulmonary symptoms in the Marines described in the study by Thomas et al. (2002) were not the result of CS exposure, but were rather the result of water aspiration or swimming-induced pulmonary edema (SIPE). Their conclusions were based on the observations that all subjects became symptomatic immediately after swimming, that there was a rapid resolution of symptoms, and that there was no evidence of airway dysfunc- tion. Delayed pulmonary effects of CS exposure are unusual, and there were no other reports of such symptoms even though approximately 200,000 Marines have been exposed to CS since 1996 under similar field conditions. 2.2.2. Case Reports The effects of exposure to CS are generally of an acute nature. However, re- active airways dysfunction syndrome was reported in two individuals exposed to CS. One case involved a healthy 21-year-old female exposed to CS smoke at a nightclub for 5-10 min (Hu and Christiani 1992). She exhibited the typical signs and symptoms of CS exposure, including tightness and burning in her chest and coughing. Results of a physical examination and chest radiography were normal, 

324 Acute Exposure Guideline Levels and she was released from the hospital. She continued to experience coughing and shortness of breath, and had reduced measurements of forced expiratory volume in 1 sec (68% of predicted) and forced vital capacity (78%) 4 weeks after exposure. Cough and shortness of breath were still present at the 2-year follow-up exam, and were made worse by exertion, cold air, and some environmental pollutants. The second case report involved exposure to a riot-control agent containing 1% CS and 1% oleo resin capsicum (Roth and Franzblau 1996). A healthy 53-year-old male was exposed for at least 30 sec, and immediately experienced symptoms of mu- cous membrane irritation, cough, and chest tightness. Wheezing and shortness of breath continued for months after exposure, and were severe enough to require hospitalization. Pulmonary function test results indicated reversible and fixed ob- structive pulmonary disease. Effects of exposure to the capsicum cannot be ex- cluded. A 4-month-old infant exposed to CS for 2-3 h developed pneumonitis and persistent leukocytosis (Park and Giammona 1972). The infant was exposed when a CS canister was fired into a house to subdue an adult. Upon hospitalization, the infant had copious nasal and oral secretions and was sneezing and coughing. A chest X-ray demonstrated that the lungs were clear, but laboratory testing revealed leukocytosis. The infant developed severe respiratory distress by the second day of hospitalization, with pulmonary infiltrates evident on X-ray by day 7. Pulmonary infiltration began to decrease on day 15, and the lungs were clear on day 17. White blood cell counts were elevated throughout hospitalization, and finally decreased when the infant was discharged from the hospital. CS is a common riot-control agent in Britain; consequently, typical symp- toms following exposure to CS have been described following its use in con- fined spaces, such as a night club (Breakell and Bodiwala 1998) or bus (Ka- ragama et al. 2003), use by police on individuals for self-defense (Euripidou et al. 2004), or under conditions of large-scale riot control (Himsworth 1969; An- derson et al. 1996). Symptoms of exposure included but were not limited to ocu- lar irritation, lacrimation, blurred vision, burning sensations sometimes accom- panied by first degree burns, cough, headache, shortness of breath, chest pain, sore throat, retching, vomiting, and salivation (Himsworth 1969; Anderson et al. 1996; Breakell and Bodiwala 1998; Karagama et al. 2003; Euripidou et al. 2004). In general, the symptoms resolved rapidly; however, there were reports of effects lasting longer that that predicted. The hand-held spray canisters used by police contain CS dissolved in methyl isobutyl ketone, an industrial solvent and denaturant (Gray 2000; Euripiou et al. 2004). It has, therefore, been pro- posed that the ketone combined with the CS may result in longer lasting adverse effects than CS preparations without the solvent. 2.3. Developmental and Reproductive Toxicity The National Teratology Information Service collected outcome data on 30 pregnant women who were exposed to CS gas: 12 women during the first 

Tear Gas (CS) 325 trimester, 11 during the second trimester, and seven during the third trimester (McElhatton et al. 2004). Acute maternal toxicity (transient symptoms of ear, nasal, and throat irritation) was reported by 50, 82, and 57% of the exposed women, respectively. Pregnancy outcome was not adversely affected by expo- sure to CS. Birth weight was within the normal range except for one female ba- by weighing less than 2,500 g. One infant had a congenital anomaly (hypospa- dia), and this anomaly has a background incidence of 1 in 1,000 live born male infants. No concentration or duration exposure parameters were described. 2.4. Genotoxicity No data on the genotoxicity of CS in humans were found. 2.5. Summary CS is a potent irritant, with symptoms of exposure including lacrimation, blepharospasm, erythema of the eyelids, chest tightness, coughing, nasal irrita- tion and discharge, salivation, throat irritation, nausea, vomiting (from swallow- ing excess saliva), and cutaneous irritation (ranging from stinging to contact irritation or allergic dermatitis). Upshall (1973) reported that an aerosol concen- tration of CS at 4 mg/m3 will disperse the majority of rioters within 1 min, and 10 mg/m3 will deter trained troops. With the exception of more severe cutaneous reactions, recovery from exposure is generally rapid upon exposure to fresh air, generally within 30 min after exposure (Ballantyne 1977). Data on human tolerance to CS are summarized in Table 7-7. Many stud- ies investigated the amount of time that elapsed before subjects could no longer remain in an atmosphere containing CS. Gutentag et al. (1960) and Punte et al. (1963) reported tolerances of 5 sec at 442 mg/m3, 12-25 sec at 141 mg/m3, 9 sec at 39 mg/m3, and more than 90 min at 1.5 mg/m3. Tolerance to low concentra- tions of CS could be increased when exposure concentration was increased over time (Punte et al. 1963; Beswick et al. 1972). A study investigating the differ- ences in respiratory and ocular responses to different particles sizes of CS found that small particles are more effective than larger particles in producing ocular and respiratory irritation. Recovery time for ocular irritation took longer for large particles, because the onset of irritation was delayed due to the lower solu- bility of large particles. Recovery time for respiratory irritation took longer for small particles, because the smaller sized particles penetrated further into the respiratory tract (Owens and Punte 1963). Pregnancy outcomes were not affected in a prospective case study of 30 pregnant women who were exposed to CS gas and experienced transient symp- toms of ear, nasal, and throat irritation (McElhatton et al. 2004). No other repro- ductive or developmental toxicity data of CS in humans were available. No hu- man data on the toxicity of repeated exposures to CS or on the genotoxicity or carcinogenicity of CS were found. 

326 TABLE 7-7 Summary of Selected Human Toxicity Data on Tear Gas Concentration (mg/m3) Tolerance Durationa Notes Reference 94b <60 sec 6 subjects chosen for ability to tolerate CS. Owens and Punte 1963 85c <60 sec 5 53 to ≥120 sec Untrained subjects; maximum exposure duration of 2 min. Punte et al. 1963; 12 19-43 sec Gutentag et al. 1960 442 5 sec 4 37 to ≥120 sec Trained subjects (previous exposure to CS); maximum exposure Punte et al. 1963; 10 18-41 sec duration of 2 min. Gutentag et al. 1960 141 12-25 sec 10 13 sec Untrained subjects; exercised before exposure. Punte et al. 1963; 13 13 sec Gutentag et al. 1960 39 9 sec 0.4 135-600 secd 4/7 tolerated 10 min. Rengstorff 1969 0.6 30-600 secd 7/17 tolerated 10 min. 0.9 600 secd 6/6 tolerated 10 min. 1 35-50 sec 6 (attained over 10 min) 18 min 1 subject Punte et al. 1963 20 min 1 subject 29 min 1 subject 0.75 (average)e 60 minf 5 subjects; all remained in chamber for duration of exposure; ocular, Beswick et al. 1972 nasal, mouth, and throat irritation, nausea, chest discomfort, headache, and stinging of the face. 0.56-0.86 8 min 9 subjects; concentration increased during exposure; 0.31-0.74 60 minf 1 subject in 0.31–0.74-mg/m3 group left after 8 min because of irritation; 8 subjects tolerated 60-min exposure with same signs as 0.75-mg/m3 group. 

0.8-1.4 60 minf 5 subjects; all tolerated exposure with same signs as 0.75-mg/m3 group. 0.84-2.3 60 minf 16 subjects; 1 subject vomited after 5 min, left chamber but returned for 0.7-2.0 duration of exposure; 1 subject vomited after 55 min; 14 subjects tolerated 0.63-2.3 60-min exposure with same signs as 0.75-mg/m3 group. 0.57-2.1 0.42-1.8 0.45-1.7 1.5 90 ming 4 subjects; 1 developed nasal irritation (at 2 min); 3 developed headache Punte et al. 1963 (at 45, 50, and 83 min); 4 had ocular irritation (at 20, 24, 70, and 75 min) a Time at which the subject could no longer tolerate exposure. b Mass median diameter of 0.9 microns. c Mass median diameter of 60 microns. d Exposure was for a maximum of 10 min. e Average concentration calculated using the six interval concentrations. f Exposure was for a maximum of 60 min. g Exposure was for a maximum of 90 min. 327 

328 Acute Exposure Guideline Levels 3. ANIMAL TOXICITY DATA 3.1. Acute Lethality 3.1.1. Monkeys Groups of eight immature male and female Macaca mulatta monkeys (3-4 kg) were exposed to a cloud of CS dispersed via an M7A3 CS grenade in a 20,000-L chamber at an average CS concentration of 900 mg/m3 for 3 min, 1,700 mg/m3 for 5 min, 2,850 mg/m3 for 10 min, or 2,500 mg/m3 for 32 min (Striker et al. 1967). The report stated that the cloud was sampled and measured at various times, but details were not provided. A group of eight monkeys served as controls; they were treated similarly to the exposed monkeys except they were not put into an exposure chamber. Monkeys were observed frequently for clinical signs during the first 72 h after exposure. Chest radiographs were taken before exposure and after 2, 6, or 12 h or 1, 3, 7, or 30 days post-exposure. Monkeys were killed after 12 h or after 3, 7, or 30 days. Clinical signs in mon- keys exposed to CS at 900 mg/m3 for 3 min or at 1,700 mg/m3 for 5 min includ- ed blinking and a “fright reaction” observed immediately after removal from the exposure chamber, which disappeared within a few minutes after the monkeys were moved to fresh air. Monkeys exposed at 2,850 mg/m3 for 10 min exhibited frequent blinking, labored respiration, coughing, oral and nasal discharge, occa- sional vomiting, and decreased activity and response to external stimuli. One monkey also had copious ocular discharge. Clinical signs were most severe at 12 h and were generally resolved within 72 h. Clinical signs in monkeys exposed to CS at 2,500 mg/m3 for 30 min were severe and included prostration, dyspnea, copious oral and nasal discharge, and scleral congestion after removal from the exposure chamber. Five monkeys died; four died 3-12 h after exposure and one died on day 4. Dyspnea was most severe at 12 h, while oral and nasal discharge and effects on the eyes were most severe at 24 h. Radiographic findings were present only in this group; infiltrates appeared 3 h after exposure, were most severe after 24 h, and cleared after 3 days. Pathologic examination of the monkeys 12 h after exposure to CS at 900 or 1,700 mg/m3 revealed mild pulmonary congestion, bronchorrhea, emphyse- ma, and atelectasis (Striker et al. 1967). These effects disappeared in the mon- keys examined on day 3, but recurred in monkeys examined on days 7 and 30. Pathologic lesions were more severe and developed earlier in monkeys exposed at 2,850 mg/m3 for 10 min. Pulmonary edema and congestion and bronchorrhea were found at 12 h, and progressed to purulent bronchitis and bronchopneumo- nia at day 3. After 1 week, acute pleuritis and interstitial pneumonitis were seen, and mucosal lesions and bronchopneumonia were resolving. Lesions were still present after 4 weeks, and included emphysema, atelectasis, and focal interstitial pneumonitis. Pathologic findings in monkeys that died after exposed to CS at 2,500 mg/m3 for 30 min included severe pulmonary edema and congestion. The three surviving monkeys were killed on days 3, 7, or 30. The monkey killed on 

Tear Gas (CS) 329 day 3 days had considerable edema, but congestion was less prominent. Exami- nation of the monkey killed on day 7 revealed emphysema involving all lobes and bronchiolitis, but most of the edema had cleared. The monkey killed on day 30 had small shrunken lungs, purulent mucoid material filling many small bron- chioles, and distinct bronchiolitis. McNamara et al. (1969) exposed groups of four monkeys (strain and sex not specified) to seven different CS concentration-duration combinations. No further experimental details were available. Mortality data from this study are summarized in Table 7-8. 3.1.2. Rats Groups of 10 rats were exposed to an aerosol of CS for 25-90 min (Punte et al. 1962). Animals were exposed in a dynamic inhalation chamber containing individual cages on racks. Aerosol was generated by passing dry nitrogen through an aspirator. Molten CS was maintained in a side-armed flask in an oil bath at 140-150°C. The aerosol was easily generated and liquid droplets recrys- talized before entering the exposure chamber. Chamber concentrations were measured by drawing chamber air through filter paper for subsequent analysis by spectrophotometry. Samples for particle-size determinations were collected by a Cascade impactor, and MMD was derived by use of stage calibrations based on the density of the compound; the particle size was about 1.5 microns (MMD). Observations for clinical signs were made during and after exposure. Surviving animals were maintained for 14 days, and then killed and examined histopathologically. Immediately after exposure began, the animals became ex- citable and hyperactive, and lacrimation and salivation occurred within 30 sec. Lethargy and dyspnea occurred after approximately 5-15 min. Dyspnea persisted for approximately an hour after exposure ceased, and all other signs subsided about 5 min after the rats were removed from the chamber. Histopathologic ex- aminations revealed an increase in the number of Goblet cells in the respiratory tract and conjunctiva, necrosis in the respiratory and gastrointestinal tracts only if particles had impacted the surface, and an occasional animal with pulmonary edema and hemorrhage in the adrenal glands. The calculated LCT50 was 32,500 mg-min/m3. An unpublished report by McNamara et al. (1969) appears to provide data additional to those that were published by Punte et al. (1962). Specific study details are not provided in the unpublished report, but one set of study results is consistent with those published by Punte et al. (1962). The report includes the mortality results from tests with additional animal species exposed by inhalation to CS, as well as mortality data for CS dispersed by different methods. As dis- cussed above, Punte et al. (1962) reported mortality data for rats, but the values were reported only in terms of mg-min/m3. Specific concentrations of CS (sprayed as molten agent) with corresponding exposure durations for these data are reported in the unpublished study by McNamara et al. (1969) and are pre- sented in Table 7-8. 

330 Acute Exposure Guideline Levels TABLE 7-8 Mortality Data from Studies by McNamara et al. (1969) in Different Species Exposed to Tear Gas Species Concentration (mg/m3) Duration (min) Mortality Time to Death (days)a Ratsb 560 25 1/10 1(1) 543 35 2/10 1(2) 489 45 3/10 2(1), 3(1), 4(1) 454 55 5/10 1(3), 3(2) 500 60 2/10 1(2) 500 80 6/10 1(1), 2(2), 6(3) 500 90 8/10 1(1), 3(2), 7(2), 11(3) Miceb 1,200 10 0/20 – 1,100 20 7/20 7(1), 8(3), 9(3) 900 30 2/20 7(2) 800 40 5/20 5(2), 9(3) 740 50 5/20 5(1), 6(3), 7(1) 683 60 14/20 5(4), 8(5), 9(4), 13(1) Guinea pigsb 400 5 1/10 7(1) 400 10 2/10 7(1), 8(1) 400 15 4/10 1(2), 6(2) 500 20 3/10 1(1), 6(1), 7(1) 400 25 7/10 2(5), 7(1), 8(1) 400 30 7/10 1(4), 5(1). 7(1), 9(1) 425 40 8/10 1(7), 3(1) Rabbitsb 500 30 1/4 6(1) 250 40 0/4 – 267 45 0/4 – 250 80 3/4 1(1), 2(1), 7(1) 333 90 4/4 1(1), 2(1), 3(1), 8(1) Dogsc 833 20 0/4 – 649 30 1/4 12(1) 508 36 2/4 5(1), 10(1) 899 40 2/4 1(1), 2(1) 520 45 2/4 1(1), 4(1) 612 45 2/4 1(2) 797 60 3/4 1(2), 3(1) 909 60 2/4 1(2) Monkeysc 469 24 1/4 5(1) 673 30 2/4 1(2) 381 45 2/4 1(2) 612 45 1/4 1(1) 699 60 1/4 1(1) 941 60 3/4 1(3) 1,057 60 2/4 1(2) a Number in parenthesis indicates number of deaths on that day. b Source of CS was the same for rats, mice, guinea pigs, and rabbits. c Source of CS was the same for dogs and monkeys, except CS used in the monkey studies had a mass-median diameter of 2.0-3.2 microns (ultraviolet analysis was conducted at 260 nanometers). Source: Adapted from McNamara et al. 1969. 

Tear Gas (CS) 331 Groups of 18 male albino SPF rats were exposed to pyrotechnically- generated CS smoke in a 10-m3 chamber (Colgrave and Creasey 1975). The rats were exposed to at 5,871 ± 476 mg/m3 for 15 min, at 6,030 ± 590 mg/m3 for 10 min, or at 6,800 ± 1,166 mg/m3 for 5 min (averages and standard deviations were calculated on the basis of the values reported by the investigators as 6,000, 6,000, and 6,400 mg/m3, respectively). CS was released from four CS cartridges, each containing CS (12.5 g), potassium chlorate (16 g), lactose (15 g), and kao- lin (7.5 g). The cloud of CS in the exposure chamber was sampled at approxi- mately 1-min intervals for the 10- and 15-min exposures and at 30-sec intervals during the 5-min exposure. The analytic method used to measure CS concentra- tions was not described. Survivors were killed at times ranging from 15 min to 2 days post-exposure. All animals were necropsied, and selected tissues were ana- lyzed by both light and electron microscopy. Nonexposed controls were used to establish the typical macroscopic and microscopic appearance of the tissues of the particular strain used. Mortality occurred in four rats exposed at 5,871 mg/m3 for 15 min (death occurred with 24 h), and in two rats exposed at 6,030 mg/m3 for 10 min (death occurred within 24 or 36 h). All animals exposed at 6,800 mg/m3 survived until the study terminated after 2 days. Animals that died after exposure to CS for 15 min developed marked pulmonary congestion with scattered alveolar hemorrhages and patchy edema. Survivors developed less marked pulmonary congestion and only occasional areas of edema and hemor- rhaging. Rats that died after exposure to CS for 10 min also developed pulmo- nary congestion, but the severity was much less than that seen with the 15-min exposures. Hemorrhages and edema were occasionally seen in the lungs of sur- vivors. Examination of rats exposed to CS for 5 min revealed mild pulmonary congestion with occasional hemorrhage up to 6 h post-exposure. Rats killed be- tween 12 h and 2 days post-exposure had no pulmonary findings except for one rat with moderate and extensive pulmonary congestion. Electron microscopic examination of the lungs from all exposed rats revealed changes in the epitheli- um and interstitium, with accumulation of fluid between the membrane layers and collagen-containing areas of the septum. Degenerative changes of the epi- thelium and endothelium led to rupture or dissolution of the capillary wall. The investigators stated that the changes were similar in all exposed rats, with the changes varying only in the degree of severity. Damage was evident as early as 15 min after exposure, and became more severe after 30 and 60 min. Ballantyne and Callaway (1972) exposed groups of male and female Wistar-derived SPF rats to pyrotechnically-generated CS smoke at concentra- tions of 750 mg/m3 for 30 min, 480 mg/m3 for 1 h, or 150 mg/m3 for 2 h in a 10-m3 exposure chamber. A group of control animals was also maintained, but no description of the treatment of the controls was provided to determine wheth- er they were exposed under similar conditions to clean air. The grenades used for the exposure contained CS (2 g), potassium chlorate (2.4 g), lactose (2.4 g), and kaolin (1.2 g). Although the report stated that the concentration of CS in the exposure chamber was sampled at the start of the exposure and at 6-min inter- vals up to and including 57 min, no information was provided about the analytic 

332 Acute Exposure Guideline Levels technique. Groups of animals were killed after 1, 10, and 28 or 29 days. Some of the animals exposed at 480 or 150 mg/m3 were retained for up to 32 months to evaluate potential lasting toxicity and pathology (Marrs et al. 1983a). Animals that died or were moribund after 1 month and those killed after 32 months were subjected to gross necropsy, and the heart, lungs, small intestine, liver, pancreas, spleen, kidneys, brain, gonads, and pituitary and adrenal glands were removed and processed for histologic examination. All animals exposed for 30 min at 750 mg/m3 survived to the scheduled necropsy, and histopathologic changes were found only on post-exposure day 1 (see Table 7-9). One rat had congestion of alveolar capillaries and a few scat- tered alveolar hemorrhages, while another rat had a few minute foci of renal tubular necrosis at the inner cortex. No pathologic changes were found in rats at post-exposure day 10 or 28. Exposure to CS at 480 mg/m3 for 1 h resulted in the mortality of some rats (see Table 7-9), with the majority of the deaths occurring on post-exposure days 1 and 2. Pathologic changes in animals surviving expo- sure at 480 mg/m3 were generally confined to post-exposure day 1; lesions were limited to minimal pulmonary congestion and hepatic congestion in one rat, minimal pulmonary hemorrhage and hepatic necrosis in another rat, and mild pulmonary congestion in a third rat. Two rats had mild pulmonary edema. A few rats killed after 10 days had healed lesions, as evidenced by binucleate liver cells around centrilobular veins and immature epithelium in some renal tubules. No abnormal pathologic changes were noted at day 29. Histopathologic findings in rats that died were much more severe and included renal changes (mild-to- moderate necrosis of the cortex, moderate-to-severe necrosis of the medulla, and some mild congestion), pulmonary changes (mild-to-severe congestion, mild hemorrhage, and some mild edema), and hepatic changes (glycogen depletion in all rats and a few cases of mild congestion and mild-to-moderate necrosis). Exposure to CS at 150 mg/m3 for 2 h resulted in no mortality. Pathologic examination of the animals revealed lesions only on day 1 post-exposure. Le- sions were confined to female animals; one rat had a few scattered alveolar hemorrhages, one had acute mucoid enteritis, and one had pneumonic consolida- tion of the upper right lung lobe. Exposure to CS at 480 mg/m3 for 1 h or at 150 mg/m3 for 2 h did not af- fect the lifespan of rats, and no statistically significant increases were found in non-neoplastic lesions in the exposed groups compared with controls (Marrs et al. 1983a). Common non-neoplastic lesions in male and female rats included changes in the lungs (engorgement, congestion, inflammatory changes, and pulmonary edema) and pyelonephritis of the kidneys. Liver congestion was also a common finding. No exposure-related neoplastic lesions were evident in male rats. Female rats in the 150-mg/m3 group exhibited an increased incidence of pituitary tumors; incidence was 26% in the control group, 29% in the group ex- posed to CS at 480 mg/m3 for 1 h, and 47% in the group exposed at 150 mg/m3 for 2 h. The increases were not statistically significant. 

TABLE 7-9 Summary of Acute Toxicity Data from Studies by Ballantyne and Callaway (1972) in Hamsters and Mice Exposed to Tear Gas Day 1 Day 10 Day 25 or 29 No. killed No. w/ lesions No. killed No. w/ lesions No. killed No. w/ lesions 750 30 Hamster Male 24 0 8 3 8 0 8 0 Rat Male 24 0 8 2 8 0 8 0 480 60 Hamster Male 47 16 (34%) 8 4 2 0 9 0 Female 59 15 (25%) 7 6 3 0 8 0 Rat Male 60 6 (10%) 6 2 2 0 8 0 Female 60 3 (3%) 7 1 2 0 8 0 150 120 Hamster Male 58 2 (3%) 8 0 6 0 6 0 Female 62 0 8 3 10 0 8 0 Rat Male 60 0 8 0 8 0 8 0 Female 60 0 8 2 8 1 8 0 Source: Adapted from Ballantyne and Callaway 1972. 333 

334 Acute Exposure Guideline Levels In another experiment, Ballantyne and Callaway (1972) exposed groups of 10 rats for 5 to 20 min to CS at an approximate concentration of 4,000 mg/m3, followed by a 14-day observation period. An anti-riot grenade containing ap- proximately 50 g of CS was ignited in a 10-m3 static chamber and allowed to burn to completion. All animals that died and the survivors killed at the end of the 14-day observation period were subjected to gross and histologic examina- tions. Clinical signs during exposure could not be recorded because the aerosol generated in the chamber resulted in a complete lack of visibility. Upon removal from the chamber, animals exhibited signs of increased buccal and nasal secre- tions and dyspnea, particularly at the longer exposure durations. Mortality data are summarized in Table 7-10. No animals died during exposure. Necropsy re- vealed pulmonary edema and congestion, often with multiple, variable sized areas of hemorrhage, and mucus in the trachea and major bronchi. Histopatho- logic examination of these animals revealed severe congestion of the alveolar capillaries and intrapulmonary veins and alveolar hemorrhage. Mucus was seen in some bronchi and bronchioles, and occasional areas of collapse and hemor- rhage were seen distal to a completely occluded bronchiole. Moderate-to- marked pulmonary edema was also observed in several animals. No evidence of acute inflammatory cell infiltrate was observed in any of the lungs examined, suggesting that the CS aerosol produced direct injury to the pulmonary capillary endothelium. Circulatory failure evidenced as congestion of the liver, kidneys, and spleen and dilation of the right ventricle was present in most of the animals that died. Animals that survived to day 14 days did not have any residual patho- logic changes at necropsy. Groups of 20 or 21 male Porton-Wistar rats were exposed by whole body inhalation to various concentrations of CS aerosol for 10-60 min (see Table 7- 11) (Ballantyne and Swanston 1978). Animals were exposed in a 1-m3 dynamic flow chamber. The aerosol was generated by filling a Collision spray with mol- ten CS (heated to 150°C) and passing pure nitrogen into the air stream. The re- sultant aerosol was fed into the diluting air stream. The chamber atmosphere was sampled for 1 min at 5-min intervals by aspirating air through glass fiber discs held in double cone filters. A bubbler containing hydrochloric acid in eth- anol was connected in line to the glass filter to act as an additional trap. The contents of the bubbler were used to elute CS from the filter discs, and the con- centration of CS in the resultant extract was measured by absorption spectropho- tometry and compared with a prepared standard. Signs of toxicity included in- creased nasal and buccal secretions and increased rates of respiration when removed from the chamber; effects disappeared within approximately 1 h post- exposure. No animals died during exposure; deaths generally occurred within the first 2 days following exposure. A summary of mortality data is presented in Table 7-11. Necropsy findings in animals dying within 48 h included pulmonary congestion and edema (with some animals also having multiple variable sized hemorrhages) and congestion of the trachea. Moderate amounts of mucus were also seen in the trachea. Histopathologic examination of the lungs from these animals revealed moderate-to-marked congestion, inter- and intra-alveolar hem- 

Tear Gas (CS) 335 orrhaging, and excess secretions in the bronchioles and intrapulmonary bronchi. Examination of animals dying after 48 h revealed similar findings, as well as evidence of early bronchopneumonia. Congestion of the liver, kidneys, spleen, and small intestines were also frequently seen in animals dying from exposure. No abnormal findings were found in animals surviving the 14-day observation period. 3.1.3. Mice Groups of 20 mice were exposed to an aerosol of CS for 10-60 min (Punte et al. 1962). Experimental procedures, clinical signs, and necropsy results are simi- lar to those described for the corresponding study in rats (see Section 3.1.2). The calculated LCT50 was 43,500 mg-min/m3. An unpublished report by McNamara et al. (1969) appears to provide data additional to those that were published in this study. Specific study details are not provided in the unpublished report, but one set of study results is consistent with those published by Punte et al. (1962). The re- port includes the mortality results from tests with additional animal species ex- posed by inhalation to CS, as well as mortality data for CS dispersed by different methods. Punte et al. (1962) reported mortality data for mice, but the values were reported only in terms of mg-min/m3. Specific concentrations of CS (sprayed as molten agent) with corresponding exposure durations for these data are reported in the unpublished study by McNamara et al. (1969) and are presented in Table 7-8. Ballantyne and Callaway (1972) exposed groups of 10 mice for 5-20 min to CS at an approximate concentration of 4,000 mg/m3, followed by a 14-day observation period. The experimental protocol, clinical signs, and necropsy find- ings were similar to those described in the corresponding study in rats (see Sec- tion 3.1.2). Mortality data are summarized in Table 7-10. TABLE 7-10 Summary of Mortality Data from Studies by Ballantyne and Callaway (1972) in Different Species Exposed to Tear Gas Exposure Mortality (No. died/No. exposed) Duration 3 Concentration (mg/m ) (min) Rat Mouse Guinea Pig Rabbit 3,950 5 0/10 1/10 1/5 0/5 4,760 5 0/10 0/10 0/5 0/5 4,250 10 1/10 0/10 5/5 0/5 4,330 10 1/10 4/10 3/5 2/5 4,150 15 0/10 3/10 3/5 2/5 5,167 15 7/10 3/10 5/5 2/5 4,000 20 9/10 8/10 5/5 4/5 4,300 20 8/10 6/10 5/5 5/5 Source: Adapted from Ballantyne and Callaway 1972. 

336 TABLE 7-11 Summary of Mortality Data from Studies by Ballantyne and Swanston (1978) in Different Species Exposed to Tear Gas Mortality Species Average Concentration (mg/m3) Duration (min) (No died/No. exposed) Mortality (%) LCT50 (mg-min/m3 ± 95% CI) Rat (male) 1,802 10 0/20 0 88,480 (77,370-98,520) 1,806 45 8/20 40 1,911 45 9/20 45 2,629 60 20/21 95 2,699 60 20/20 100 Mouse (male) 1,432 15 1/40 3 2,753 20 17/40 43 50,010 (42,750-60,220) 2,333 30 10/19 53 2,400 30 17/40 43 2,550 30 24/36 67 Guinea pig (female) 2,326 10 2/20 10 67,200 (59,200-78,420) 2,380 15 2/10 20 1,685 25 10/20 50 2,310 20 8/20 40 1,649 30 11/20 55 1,302 45 9/11 82 2,041 30 13/20 65 2,373 30 10/19 53 

Rabbit (female) 846 5 0/10 0 54,090 (42,630-70,400) 836 10 0/10 0 1,434 10 0/10 0 1,802 10 1/5 20 2,188 15 2/10 20 2,380 15 3/8 38 1,407 30 4/10 40 1,653 30 2/10 20 1,309 45 4/5 80 2,118 45 9/10 90 2,133 60 7/8 88 3,066 60 8/9 89 Source: Adapted from Ballantyne and Swanston 1978. 337 

338 Acute Exposure Guideline Levels Groups of 19-40 male albino mice were exposed by whole body inhalation to various concentrations of CS aerosol for 15-30 min (see Table 7-11) (Ballan- tyne and Swanston 1978). The experimental protocol, clinical signs, and necrop- sy findings were similar to those described in the corresponding study in rats (see Section 3.1.2). Mortality data are summarized in Table 7-11. 3.1.4. Guinea Pigs Groups of ten guinea pigs were exposed to an aerosol of CS for exposure durations of 5-40 min (Punte et al. 1962). Experimental procedures, clinical signs, and necropsy results are similar to those described for rats in Section 3.1. 2. The calculated LCT50 is 8,300 mg min/m3. An unpublished report by McNamara et al. (1969) appears to provide data additional to those that have been published. Spe- cific study details are not provided in this report, but one set of study results is consistent with those published by Punte et al. (1962). The report includes the mortality results of additional animal species exposed by inhalation to CS, as well as mortality data for CS dispersed by various methods. As described above, Punte et al. (1962) reported mortality data for guinea pigs, but the values were reported only in terms of mg min/m3. Specific concentrations of CS (sprayed as molten agent) with corresponding exposure durations for these data are reported in McNamara et al. (1969) and are presented in Table 7-8. Ballantyne and Callaway (1972) exposed groups of five guinea pigs for 5 to 20 min to an approximate CS concentration of 4,000 mg/m3, followed by a 14-day observation period. The experimental protocol, clinical signs, and nec- ropsy findings are similar to those described in the rat study (see Section 3.1.2). Mortality data are summarized in Table 7-10. Groups of ten to twenty female Dunkin Hartley guinea pigs were exposed by whole body inhalation to various concentrations of CS aerosol for durations of 10 to 45 min (see Table 7-11) (Ballantyne and Swanston 1978). Experimental protocol, clinical signs, and necropsy findings are as described for the rat study in Section 3.1.2. Mortality data are summarized in Table 7-11. 3.1.5. Rabbits Groups of four rabbits were exposed to an aerosol of CS for exposure du- rations of 30-90 min (Punte et al. 1962). Experimental procedures, clinical signs, and necropsy results are similar to those described for rats in Section 3.1.2, ex- cept that hyperactivity, salivation, and lachrymation were not reported. The cal- culated LCT50 is 17,000 mg min/m3. An unpublished report by McNamara et al. (1969) appears to provide data additional to those that have been published. Specific study details are not provided in this report, but one set of study results is consistent with those published by Punte et al. (1962). The report includes the mortality results of additional animal species exposed by inhalation to CS, as well as mortality data for CS dispersed by various methods. As described above, 

Tear Gas (CS) 339 Punte et al. (1962) reported mortality data for guinea pigs, but the values were reported only in terms of mg min/m3. Specific concentrations of CS (sprayed as molten agent) with corresponding exposure durations for these data are reported in McNamara et al. (1969) and are presented in Table 7-8. Ballantyne and Callaway (1972) exposed groups of five rabbits for 5 to 20 min to an approximate CS concentration of 4,000 mg/m3, followed by a 14-day observation period. The experimental protocol, clinical signs, and necropsy find- ings are similar to those described in the rat study (see Section 3.1.2). Mortality data are summarized in Table 7-10. Groups of five to ten female New Zealand white rabbits were exposed by whole body inhalation to various concentrations of CS aerosol for durations of 5 to 60 min (see Table 7-11) (Ballantyne and Swanston 1978). Experimental pro- tocol, clinical signs, and necropsy findings are as described for the rat study in Section 3.1.2. Mortality data are summarized in Table 7-11. 3.1.6. Hamsters Ballantyne and Callaway (1972) exposed groups of male and female gold- en hamsters to pyrotechnically-generated CS smoke at concentrations of 750 mg/m3 for 30 min, 480 mg/m3 for 1 h, or 150 mg/m3 for 2 h in a 10-m3 exposure chamber. The experimental protocol is the same as that for the corresponding study in rats (see Section 3.1.2). All animals exposed for 30 min at 750 mg/m3 survived to the scheduled necropsy, and histopathologic changes were observed only on post-exposure day 1 (see Table 7-11). Three hamsters had a few scat- tered alveolar hemorrhages, and one also had congestion of the alveolar capillar- ies. No pathologic changes were found at post-exposure day 10 or 28. Exposure at 480 mg/m3 for 1 h killed some hamsters (see Table 7-11); the majority of the death occurred after 1-2 days. Pathologic changes in animals surviving exposure at 480 mg/m3 were generally found only on post-exposure day 1. Eight hamsters had mild pulmonary congestion, and four of them also had mild pulmonary hemorrhage. One hamster with no lung lesions had mild renal congestion and necrosis in the medulla, and another had mild necrosis in the medulla. A few hamsters killed after 10 days had healed lesions, as evidenced by binucleate liv- er cells around centrilobular veins and immature epithelium in some renal tu- bules. No abnormal pathologic changes were found at day 29. Histopathologic findings in hamsters that died were generally similar to those in rats (see Section 3.1.2); however, the lesions were less severe in hamsters than in rats. Exposure to CS at 150 mg/m3 for 2 h resulted in the mortality of two male hamsters (after 12 or 16 days), and necropsy revealed bronchopneumonia. Path- ologic examination of surviving animals revealed lesions only on day 1. Lesions were found only in female hamsters; one had a few scattered alveolar hemor- rhages and two had a few scattered foci of acute renal tubular necrosis at the inner cortex. 

340 Acute Exposure Guideline Levels Exposure to CS at 480 mg/m3 for 1 h or at 150 mg/m3 for 2 h did not af- fect the lifespan of the hamsters, and no statistically significant increases in non- neoplastic lesions were found in the exposed groups compared with controls (Marrs et al. 1983a). Common non-neoplastic lesions in male and female ham- sters included changes in the lungs (engorgement, congestion, inflammatory changes, and pulmonary edema) and pyelonephritis of the kidneys. No expo- sure-related neoplastic lesions were evident in male or female hamsters. 3.1.7. Dogs McNamara et al. (1969) exposed groups of four dogs (strain and sex not specified) to eight different CS concentration-duration combinations. No further experimental details were available. Mortality data are summarized in Table 7-8. Following a 30-sec exposure to CS at 25 mg/m3, one dog exhibited in- creased blood pressure, an altered respiratory pattern, tachycardia, and increased femoral artery blood flow (Cucinell et al. 1971). In another test, two dogs were exposed for 23 min to CS at 2,600 mg/m3. One dog survived and the other dog died after 52 h. The investigators noted that dogs recover partially when exposed to a lethal dose of CS but then develop respiratory distress and die within 48-70 h. 3.2. Nonlethal Acute Toxicity 3.2.1. Mice An RD50 (concentration that reduces the respiratory rate by 50%) for CS of 4.0 mg/m3 (95% confidence interval: 3.3-5.2 mg/m3) was reported for male Swiss-Webster mice (Kane et al. 1979). 3.2.2. Rabbits To investigate whether CS exposure can cause diarrhea, four rabbits were exposed to thermally-generated pure CS in a 10-m3 chamber (Ballantyne and Beswick 1972). The exposures involved the following: one rabbit each was ex- posed at 58 mg/m3 for 30 min, 46 mg/m3 for 20 min, 54 mg/m3 for 12 min, or 17 mg/m3 for 17 min. Animals were placed singly in cages with removable trays lined with several layers of filter paper arranged to collect stool samples. The number of stool pellets passed, their total weight, and their water content were recorded for several day before and after exposure. Exposure to CS did not re- sult in an increased incidence of diarrhea. Two rabbits were exposed in a static chamber to the entire contents of a 3-ounce unit containing 71.5 g of CS (Gaskins et al. 1972). The unit required 20 sec to fully dispense. Both rabbits became unconscious after approximately 2 min of exposure and were moved to fresh air. The rabbits regained their righting 

Tear Gas (CS) 341 reflex approximately 10-20 min after exposure and were almost completely re- covered after 1 h (moderate ocular wetness was the only visible effect). Gross necropsy of the rabbits performed after 2 weeks did not reveal any abnormali- ties. Two other rabbits were exposed to 23.2 g of CS during dispersion of a CS unit requiring about 10 sec to completely discharge. The dispensed CS formed a cloud in the chamber. The rabbits tried to avoid the spray as it was dispensed, and then sat quietly with their eyes tightly closed for the remainder of the 5-min exposure. No abnormalities were observed in the eyes or skin of the rabbits. 3.3. Repeat-Dose Studies 3.3.1. Rats Groups of five male and five female F344/N rats were exposed to CS2 at concentrations of 0, 1, 3, 10, 30, or 100 mg/m3 for 6 h/day, 5 days/week for 2 weeks (NTP 1990). (CS2 contains 94% CS, 1% hexamethyldisilizane, and 5% Cab-o-Sil®). All rats exposed at 30 or 100 mg/m3 died before the end of the study. Rats from all exposure groups exhibited adverse clinical signs, ranging from erythema and blepharospasm at the lower concentrations to dacryorrhea, mouth breathing, listlessness, and mouth breathing at the higher concentrations. Rats in the 1-mg/m3 group gained more weight over the exposure period than controls, but at concentrations of 3 mg/m3 and higher body weight was generally decreased. Groups of 10 male and 10 female F344/N rats were exposed to CS2 at 0, 0.4, 0.75, 1.5, 3, or 6 mg/m3 for 6 h/day, 5 days/week for 13 weeks (NTP 1990). One male rat exposed at 6 mg/m3 died, and all others survived to study termina- tion. Clinical signs of ocular irritation (partial or complete eyelid closure) were noted in all exposure groups, and rats exposed at 6 mg/m3 developed erythema of the extremities that persisted overnight. Rats exposed to CS2 at 1.5 mg/m3 or higher gained significantly less weight over the study period than controls; final mean body weight was 17-44% lower than that of controls for males and 10- 24% lower for females. An approximate 46% reduction in thymus weight rela- tive to body weight was noted in male and female rats exposed at 6 mg/m3. Con- centration-related histopathologic changes included focal erosion with regenera- tive hyperplasia and squamous metaplasia of the respiratory epithelium. Acute inflammation and hyperplasia of the respiratory epithelium were also found. One group of 56 male rats was exposed to a mean CS concentration of 1,470 or 1,770 mg/m3 for 5 min/day for 5 days and another group of 49 male rats was exposed at a mean concentration of 12.5 or 14.8 mg/m3 for 80 min/day for 9 days to (Ballantyne and Callaway 1972). Exposures to the thermally- generated CS aerosol (MMD of 1-2 micrometers) were conducted in a 1-m3 chamber, with chamber air sampled continuously throughout exposure at a rate of 1 L/min using a double cone filter. The samples were analyzed for CS content (details not provided). Groups of three to five survivors were killed after 1, 6, 

342 Acute Exposure Guideline Levels and 24 h and 2, 3, 4, 5, 7, 10, 14, and 21 days, and gross and microscopic exam- inations were performed. All animals survived the 5-min exposures. Histopatho- logic examination revealed minimal congestion of the alveolar capillaries after 1 or 6 h in two of five rats and a few scattered alveolar hemorrhages after 2 days in one of four rats. Scattered patches of bronchopneumonia were found in one of five rats after 7 days, in one of three rats after 8 days, one of three rats after 10 days, and two of five rats after 18 days. Pathologic changes in control rats in- cluded scattered alveolar hemorrhages in two of 11 rats and subacute mucoid enteritis in one of 11 rats. Death occurred in five of the 49 rats exposed at 12.5 or 14.8 mg/m3 for 80 min/day; one died after the seventh exposure, two after the eight exposure, and two died 5 days after the final exposure. Necropsy revealed widespread acute bronchopneumonia. Histopathologic examination of the sur- viving animals revealed lesions for up to 5 days after exposure and not thereaf- ter. 3.3.2. Mice Groups of five male and five female B6C3F1 mice were exposed to CS2 at concentrations of 0, 1, 3, 10, 30, or 100 mg/m3 for 6 h/day, 5 days/week for 2 weeks (NTP 1990). All mice exposed at 10 mg/m3 and greater died before study termination. Mice from all exposure groups exhibited adverse clinical signs, ranging from erythema and blepharospasm at the lower concentrations to dacry- orrhea, mouth breathing, listlessness, and mouth breathing at the higher concen- trations. Mice exposed at 1 mg/m3 gained more weight over the exposure period than controls, but generally lost body weight at exposure concentrations of 3 mg/m3 and higher. Groups of 10 male and 10 female B6C3F1 mice were exposed to CS2 at 0, 0.4, 0.75, 1.5, 3, or 6 mg/m3 for 6 h/day, 5 days/week for 13 weeks (NTP 1990). All mice exposed at 6 mg/m3 died and one male and one female mouse from the 3-mg/m3 group died during the second week of exposure. Closed or partially- closed eyes during exposure were observed in mice from all exposure groups through week 6, and in mice exposed at 3 mg/m3 during weeks 12 and 13. Con- centration-related decreases in body weight compared with controls were found in all exposure groups; final mean body weights of mice in the 3-mg/m3 group were 13% lower for males and 9% lower for females. Exposure-related histo- pathologic changes were observed in mice exposed at 1.5 mg/m3 and higher, and included focal inflammation and squamous metaplasia (primarily in the nasal turbinates) and inflammation of the vomeronasal organ. 3.3.3. Rats, Mice, Guinea Pigs, and Rabbits Groups of five to 10 guinea pigs, five rabbits, 10 rats, and 10-20 mice were exposed to CS at approximate concentrations of 30-40 mg/m3 for 5 h/day for 1-7 successive days (Ballantyne and Callaway 1972). An anti-riot grenade 

Tear Gas (CS) 343 containing 0.5 to 0.75 g of CS was ignited every 30 min in a 10-m3 static cham- ber to maintain the nominal concentration. The investigators stated that concen- trations were determined by continuous sampling throughout the exposure, but no details were provided. Animals were removed to fresh air following each exposure, and were maintained for a 14-day post-exposure period. All animals that died and the survivors killed at the end of the study were given gross and histologic examinations. A summary of the mortality data is presented in Table 7-12. The description of clinical signs was limited to a statement that rabbits and rats exhibited more rhinorrhea and lacrimation than did mice, whereas guinea pigs showed few clinical signs apart from occasional sneezing during the first hour of exposure. Necropsy of animals that died revealed moderate-to-marked congestion of the alveolar capillaries and intrapulmonary veins and inter- and intra-alveolar areas of hemorrhage; many of the animals that died also had con- gestion of the liver, kidneys, and small intestine. Moderate pulmonary edema was noted in a “few of the animals.” No residual pathologic changes were found in animals that survived until the end of the study. 3.4. Developmental and Reproductive Toxicity Groups of 22-24 pregnant Porton strain rats or 12 pregnant New Zealand white rabbits were exposed to CS aerosol for 5 min/day on gestation days 6-15 or 6-18, respectively (Upshall 1973). CS The aerosol had a particle size of 1-2 micrometers and was generated by melting pure crystalline CS at 120°C using a Collison spray. A preliminary study investigated exposure to CS at 0 or 20 mg/m3, and was followed by a concentration-response study that evaluated CS at concentrations of 0, 6, 20, or 60 mg/m3. Control rats were recaged and moved out of their normal environment during the test-group exposure, and control rabbits were exposed to a siliconized silica aerosol at 60 mg/m3. Additional con- trol groups of pregnant rats were exposed to a particulate aerosol (60 mg/m3 of Neosil) or to water aerosol to evaluate the stress of aerosol exposure. Rats were killed on gestation day 21 and rabbits on gestation day 30. Cesarean sections were performed, and the fetuses were evaluated for skeletal or visceral abnor- malities. In addition, the lungs, liver, kidneys, and adrenal glands from the rabbit dams in the concentration-response study were evaluated histologically. No de- finitive effects of treatment were noted. In the preliminary rat study, exposed animals exhibited a decrease in maternal weight gain compared with controls (- 23%), but a clear concentration-response relationship was not observed in the main study (-23, -12, and 15% for the 6, 20, or 60 mg/m3 groups, respectively). Fetal weight appeared to decrease with increasing concentration in the main rat study (3.3, 3.2, and 3.1 g, respectively, vs. 3.5 for controls), but the fetal weights were comparable those in other studies. No other statistically significant effects were observed. No exposure-related effects were found in exposed rabbits or their offspring. Although the exposure concentrations were sufficient to cause extreme irritation, clinical signs in exposed rats and rabbits were not reported. 

344 Acute Exposure Guideline Levels TABLE 7-12 Summary of Mortality Data in Different Species Exposed to Tear Gas for 5 Hours per Day for Up to 7 Days Duration Mortality Species Hours/Day No. Days Concentration (mg/m3) (No. died/No. exposed) Guinea pig 5 1 44.7 0/5 3 36.0 2/5 4 34.2 3/10 6 35.2 2/5 7 43.7 10/10 Rabbit 5 3 36.0 1/5 5 34.2 2/5 Rat 5 1 37.0 1/10 3 36.0 9/10 5 34.2 7/10 Mouse 5 1 40.0 0/10 2 38.8 0/10 3 36.0 1/10 4 31.9 10/10 5 56.4 16/20 Source: Ballantyne and Callaway 1972. 3.5. Genotoxicity In general, CS was not mutagenic to Salmonella typhimurium. Mutations were not induced with or without the presence of S9 at CS concentrations of 12.5-800 μg/plate in strains TA97a, TA98, TA100, TA102, or TA104 (Meshram et al. 1992); of up to 1.5 mg/plate in strains TA98, TA100, TA1535, or TA1537 (Wild et al. 1983); ranging from 10 μg/plate to 2 mg/plate in strains TA98, TA1535, TA1537, or TA1538 (von Däniken et al. 1981); or at CS2 concentra- tions of 3.3-333 μg/plate in strains TA98, TA100, TA1535, or TA1537 (NTP 1990). Equivocal responses for CS and CS2 were reported in strain TA100 only without S9 (von Däniken et al. 1981; NTP 1990), and for CS2 in strain TA97 but only with 30% S9 (NTP 1990). Cytotoxicity from CS was observed starting at 200 μg/plate, but the presence of 30% S9 generally reduced the cytotoxicity. Other in vitro genotoxicity testing was generally positive. CS induced sis- ter chromatid exchanges and chromosomal aberrations in Chinese hamster ovary cells both with and without S9 at CS2 concentrations of 6 μg/mL and greater (NTP 1990). Trifluorothymidine resistance in mouse L5178Y lymphoma cells 

Tear Gas (CS) 345 was induced in the absence of S9 at a CS concentration of 2.5 μg/mL (McGregor et al. 1988; NTP 1990). V79 Chinese hamster cells exposed to CS in culture at 19, 38, or 75 μM for 3 h and evaluated 6 days later showed reduced survival (by about 20, 30, and 80%, respectively; estimated from a graph), and exhibited a concentration-related increase in the frequency of mutants resistant to 6- thioguanine (mutations induced approximately 4- to 5-fold above controls at the highest concentration) (Ziegler-Skylakakis et al. 1989). Exposure to CS also increased the frequency of micronuclei by approximately 2-fold at 19 μM and up to 18-fold at 75 μM (measured 24 h after exposure), but did not induce DNA- repair synthesis as assessed using the BrdUrd density-shift method. A concentra- tion-dependent increase was observed in spindle cell disturbances, particularly C-metaphases (chromosomes completely scattered in cytoplasm and often high- ly contracted), when cells were exposed to CS at 5, 9, 19, or 38 μM for 3 h (Schmid and Bauchinger 1991). The C-mitotic effect was also reflected in the appearance of a metaphase block and the disappearance of other mitotic figures (prophases and ana-telophases). When a differential staining technique was ap- plied to allow for visualization of the spindle apparatus and chromosomes, a concentration-dependent increase in the number of mitoses with abnormal spin- dles was again observed, particularly apolar mitoses (mitotic figures without any signs of polar spindle configurations) (Salassidis et al. 1991). Further investiga- tion into the mechanism of CS-induced c-mitotic spindle damage found that exposure of cells to CS at 38 μM for 20 h or 3 h followed by 20 h of recovery resulted in an increase in the number of aneuploid cells and in the polyploid index (Schmid and Bauchinger 1991). The number of aneuploid cells and the polyploid index were increased to a much greater extent by exposure to the me- tabolite o-chlorobenzaldehyde than to CS, suggesting that this metabolite may play a role in the induction of spindle damage. A comparison of the effective- ness of various exposure conditions revealed that cells exposed to CS at concen- trations of up to 38 μM for 20 h exhibited a concentration-dependent increase in the number of S-cells and the frequency of chromatid-type aberrations (single breaks, isolocus breaks and exchanges, and gaps). Exposure to CS for 3 h fol- lowed by a 20-h recovery period resulted in similar effects but was not as effec- tive. No effects were observed when cells were incubated with the supernatant from the 3 h exposure (Bauchinger and Schmid 1992). The cell cycle time of the V79 cell line is approximately 8-10 h; therefore, the cells had time to run through one or two S-phases. Genotoxicity testing in vivo was generally negative. CS did not bind to DNA in the liver or kidneys of rats injected intraperitoneally with radiolabeled CS at 13 mg/kg and evaluated 8 or 75 h after dosing, but did bind to nuclear proteins in these organs (von Däniken et al. 1981). CS did not cause an increase in sex-linked recessive mutations in germ cells of male Drosophila when admin- istered in the feed at concentrations ranging from 5 × 10-4 M to 2.6 × 10-3 M for 3 days (Wild et al. 1983), and did not increase micronucleated polychromatic erythrocytes in the bone marrow of NMRI mice administered CS by intraperito- 

346 Acute Exposure Guideline Levels neal injection at 19 or 38 mg/kg or by oral administration at 113 or 226 mg/kg (Wild et al. 1983). The oral dose of 226 mg/kg killed 10 of 13 exposed mice. 3.6. Chronic Toxicity and Carcinogenicity Groups of 50 male and 50 female B6C3F1 mice and 50 male and 50 female F344/N rats were exposed to CS2 at target concentrations of 0, 0.75, or 1.5 mg/m3 (mice) or 0, 0.075, 0.25, or 0.75 mg/m3 (rats) for 6 h/day, 5 days/week for 105 weeks (NTP 1990). Rats exposed at 0.75 mg/m3 developed histopathologic chang- es in the respiratory and olfactory epithelium of the nasal passage and inflamma- tion and proliferation of the periosteum of the turbinate bones. No neoplastic ef- fects were present. Lesions seen in the nasal cavity of exposed mice included inflammation in the anterior middle portions of the nasal passage and focal hyper- plasia and/or squamous metaplasia of the respiratory epithelium. No other adverse effects were noted. Female mice exhibited a statistically significant, exposure- related reduction in the incidences of hyperplasia and adenomas of the pituitary gland pars distalis (adenoma rates in the 0-, 0.75-, and 1.5-mg/m3 groups were 16/47, 5/46, and 1/46, respectively). Lymphomas in female mice also occurred with a significant negative trend (21/50, 12/50, and 8/50, respectively). Groups of 75 male SPF Porton strain mice, 50 male Porton Wistar-derived rats, and 50 Dunkin Hartley guinea pigs were exposed to CS at nominal concentra- tions of 0, 3, 30, or 300 mg/m3 (MMD of 3-4 micrometers) for 1 h/day, 5 days/week for up to 55 exposures (11 weeks) in mice and up to 120 exposures (24 weeks) in rats and guinea pigs (Marrs et al. 1983b). Exposure at the high concen- tration resulted in excessive mortality in mice and guinea pigs within days of ex- posure; therefore, tests at the high concentration was discontinued after three ex- posures in mice and after five exposures in rats and guinea pigs (the number of deaths were not provided) (Marrs et al. 1983b). During the first month of the ex- periment, 17% of the mice and 46% of the guinea pigs in the high-concentration groups died. A significant trend (p < 0.001) was found in the incidence of early death in mice with concentration. The investigators also reported a significant trend (p < 0.001) in the incidence of early death with concentration in guinea pigs; however, most of the mortality in guinea pigs occurred during the first month. Post-mortem examination of 10 guinea pigs that died during exposure revealed acute alveolitis in seven of the animals, with mild alveolitis present in the other three. The cause of death in mice that died during exposure to CS could not be determined. The investigators reported that toxic signs were not usually observed, and that death occurred suddenly and without warning. No cause of death could be ascribed to animals that died during the observation period. CS exposure did not affect the growth of rats or guinea pigs, but did result in a concentration-related decrease in the growth of mice. No definitive, exposure-related histologic findings were observed in mice, rats, or guinea pigs at the end of the study. No exposure- related neoplasms were found. 

Tear Gas (CS) 347 3.7. Summary Clinical signs in the acute and repeated-dose animal studies suggest that CS is highly irritating. The majority of the acute inhalation exposure data in animals focused primarily on lethality, and death was generally caused by pul- monary edema and congestion. Renal damage was also occasionally noted, but may have been secondary to anoxia. Results of genotoxicity tests were mixed. Results in gene mutations tests using S. typhimurium were generally negative, as were results of in vivo genotoxicity assays. CS induced trifluorothymidine re- sistance in mouse L5178/TK lymphoma cells in the absence of S9, and induced both sister chromatid exchanges and chromosomal aberrations in Chinese ham- ster ovary cells in the presence and absence of S9. No developmental toxicity was found in rats or rabbits, and there was no evidence of carcinogenicity in rats or mice. 4. SPECIAL CONSIDERATIONS 4.1. Metabolism and Disposition 4.1.1. Absorption One cat with a cannulated trachea was exposed to CS aerosol by an oral- nasal mask to assess absorption of CS by the upper respiratory tract (cannulation prevented access to the lower respiratory tract), while a second cat was exposed via a tracheal tube to assess absorption by the lower respiratory tract (Leadbeater 1973). Blood concentrations of CS and its metabolites after exposure to the up- per and lower respiratory tract were about 30 and 80%, respectively, of those measured in intact cats. 4.1.2. Toxicokinetics The half-lives of CS, 2-chlorobenzylmalonitrile, and 2-chlorobenzaldehyde were measured in cats and rabbits (Leadbeater 1973; Paradowski 1979). The chemicals were administered directly into the femoral artery via a cannula in cats and directly into the ear vein of rabbits. The half-lives of CS, 2- chlorobenzylmalonitrile, and 2-chlorobenzaldehyde in cats were 5.5, 9.5, and 4.5 sec, respectively, and in rabbits they ranged from 19-25, 38-55, and 38-41 sec, respectively. The in vitro half-lives of these chemicals in the blood of cats, humans and rats were also measured. In cat blood, the half-life was 5 sec for CS, 470 sec for 2-chlorobenzylmalonitrile, and 70 sec for 2-chlorobenzaldehyde. The respec- tive half lives in humans were 5, 660, and 15 sec and in rats were 7, 30, and 15 sec (Leadbeater 1973). The in vitro half-life of CS in the blood of rabbits was approx- imately 60 sec; the investigators postulated that this half-life might be longer than 

348 Acute Exposure Guideline Levels those of rats, cats, and humans because of the higher concentration of CS tested in rabbits (Paradowski 1979). CS incubated with rat liver homogenate for 5 min (ethanol-buffer; pH 7.4; 37°C) resulted in a 59% decrease in the initial amount of glutathione, with 26% of the depletion occurring spontaneously (non-enzymatically) (Rietveld et al. 1986). Binding to glutathione in vivo was confirmed by enhanced urinary thi- oether excretion in rats following intraperitoneal administration of CS (Rietveld et al. 1983, 1986). The thioether was identified as 2-chlorobenzylmercapturic acid. 4.1.3. Metabolism Metabolism of CS appears to be qualitatively similar in different species. In vivo, CS can be hydrolyzed to 2-chlorobenzaldehyde or malononitrile or can be reduced to 2-chlorobenzyl malononitrile (see Figure 7-1) (Leadbeater 1973; Paradowski 1979). 2-Chlorobenzaldehyde can then be either oxidized to 2- chlorobenzoic acid for subsequent glycine conjugation or reduced to 2- chlorobenzyl alcohol for ultimate excretion as 2-chlorobenzyl acetyl cysteine or 2-chlorobenzyl glucuronic acid. Malononitrile can break down to cyanide, and be excreted as thiocyanate. The reduction of CS to 2-chlorobenzyl malononitrile is a relatively minor pathway; 2-chlorobenzyl malononitrile can be conjugated with glycine or can be hydrolyzed to 2-chlorophenyl 2-cyanoproprionate. Radiolabeled CS was administered intravenously to rats at 0.08, 0.8, and 80 μmol/kg (3H-ring labeled), 0.8 and 80 μmol/kg (14C-cyanide labeled), or 0.8 and 80 μmol/kg of (14C=C side-chain labeled) (Brewster et al. 1987). Rats were also treated intragastrically with CS at 80, 106, and 159 μmol/kg (14C-cyanide labeled). The major urinary metabolites recovered in rats up to 96 h after intra- venous or intragastric administration of CS were 2-chlorohippuric acid (49% of dose), 2-chlorobenzyl glucuronic acid (10%), 2-chlorobenzyl cysteine (8%), and 2-chlorobenzoic acid (8%), and minor metabolites included 2-chlorophenyl ace- tyl glycine (3%), 2-chlorobenzyl alcohol (1.6%), and 2-chlorophenyl 2-cyano propionate (1.6) (see Figure 7-1). In another investigation, urinary concentrations of cyanide and thiocyanate were measured over a 24-h period in untreated rats, in rats administered CS intrave- nously, or in rats exposed to the CS hydrolysis product malononitrile intraperi- toneally or intragastrically (Brewster et al. 1987). Following CS and malono- nitrile administration, urinary cyanide concentrations remained at or below baseline levels, while thiocyanate concentrations generally increased with the CS or malononitrile dose. The percentage molar conversion from CS to thiocya- nate was 21.5% at an intraperitoneal dose of 212 μmol/kg and 30% at an intra- gastric dose of 212 μmol/kg. In tests with malononitrile, the percentage was 60% or more at an intraperitoneal dose of 80 μmol/kg or intragastric dose of 212 μmol/kg. 

2-chlorobenzoic acid 2-chlorohippuric acid 2-chlorobenzaldehyde sed roly hyd 2-chlorobenzyl acetyl cysteine CS hy 2-chlorobenzyl alcohol dro lys 2-chlorobenzyl glucuronic acid ed malonitrile (cyanide) thiocyanate re du ce d 1-chlorophenyl acetyl glycine 2-chlorobenzyl malononitrile 2-chloro phenyl-2-cayano propionamide 2-chlorophenyl 2-cyanopropionate FIGURE 7-1 Predominant metabolic pathways of tear gas in rats proposed by Leadbeater (1973), Paradowsk (1979), and Rietveld et al. (1983). 349 

350 Acute Exposure Guideline Levels Metabolism in rabbits is similar to that in rats. The predominant biotrans- formation pathway in the blood of rabbits administered high doses of CS by intravenous injection (0.5 LD50 to the LD50) was hydrolysis of CS to 2- chlorobenzyaldehyde and malononitrile (~30-40%) (Paradowski 1979). A minor pathway involved reduction to 2-chlorobenzyl malononitrile (10%) The investi- gators indicated that the remaining 50-60% of the administered CS disappeared from the blood by other means; no other explanation was provided. The liver is involved in the metabolism of CS, as demonstrated by an increase in the half- lives of CS and its metabolites in the blood of rabbits when the liver was ex- cluded from the circulation. More of the CS was accounted for after dosing, with approximately 75% of the CS hydrolyzed to 2-chlorobenzyaldehyde and 15% reduced to 2-chlorobenzyl malononitrile. When the kidneys were excluded from the circulation, no changes were observed in CS or metabolites in the blood. Maximum blood concentrations of CS and its derivatives in cats were attained 30 min after intragastric administration at 40 mg/kg (Leadbeater 1973). After 90 min, blood concentrations of 2-chlorobenzylmalonitrile and 2- chlorobenzaldehyde were still elevated, can CS concentrations had returned to zero. When anesthetized cats were exposed for 60 min via oral-nasal masks to CS aerosol (75 or 750 mg/m3 of pyrotechnically-generated CS aerosol, 750 mg/m3 of pure CS aerosol from molten CS, or 62.5 mg/m3 of CS aerosol generated from an aqueous suspension of micronized CS in acetic acid using a Collison sprayer), concentrations of CS and 2-chlorobenzylmalonitrile rapidly reached steady values, but concentrations of 2-chlorobenzaldehyde continued to increase. Comparison of the blood concentrations resulting from exposure to CS at 750 and 75 mg/m3 showed that there was not a 10-fold decrease in the concentration of CS and its metabolites 2-chlorobenzylmalonitrile and 2-chlorobenzaldehyde; the concentra- tions were reduced by 4.5, 7.7, and 5.9, respectively. Exposure of cats to CS at 100 mg/m3 for 5 min/day for 4 days, followed by exposure to CS at 75 or 750 mg/m3, resulted in reduced blood concentrations of CS and its derivatives. Rats receiving a single oral dose of CS at 50-500 mg/kg had lower blood concentrations of CS and its derivatives than cats (Leadbeater 1973). CS was only detected in high-dose group. Blood concentrations of 2-chlorobenzylmalonitrile and 2-chlorobenzaldehyde in rats and cats did not increase in a dose-related man- ner. Rats exposed by inhalation to CS aerosol at concentrations of 14-245 mg/m3 for 5 min had measurable amounts of CS and 2-chlorobenzylmalonitrile in their blood immediately after exposure, but 2-chlorobenzaldehyde was detected only in rats exposed at concentrations greater than 100 mg/m3. Animal data suggest that CS should be absorbed by the human respiratory tract following inhalation exposure, and that metabolism of CS should proceed via a pathway similar to those found in laboratory animals. However, humans are not able to tolerate concentrations of CS as great as those tolerated by ani- mals. Six healthy human males were exposed by inhalation to CS at 0.5-1.5 mg/m3 over 90 min, and blood was drawn before and after exposure to measure CS and its derivatives (Leadbeater 1973). Two men left the chamber within 20 

Tear Gas (CS) 351 min. CS and 2-chlorobenzaldehyde were not detected in the blood of any of the volunteers, and only a trace of 2-chlorobenzylmalonitrile was detected in the blood of one man who remained in the chamber for the entire exposure. 4.1.4. Distribution and Elimination To evaluate the fate of CS, radiolabeled CS was administered intravenous- ly (3H-ring labeled, 14C-cyanide labeled, or 14C=C side-chain labeled) or intra- gastrically (14C-cyanide labeled) to rats, and urine, feces, and CO2 were collect- ed for 96 h (Brewster et al. 1987). The majority of the administered dose was recovered in the urine (44.4 to 100%). Recovery in feces was 1.2-23.4%, and recovery in CO2 was minimal at 0-2.1%. Comparison of recovery data for the three different radiolabels after intravenous administration showed that more radioactivity was recovered in the feces of rats administered the 14C=C side- chain labeled CS (21-23%) than from rats administered the other two labels (4- 8%). Male mice were administered 14CN-CS by intravenous injection, and were killed at selected intervals to evaluate distribution by autoradiography (Brewster et al. 1987). A significant amount of radioactivity was present in the gastrointes- tinal tract after 5 min. After 1 h, significant amounts of radioactivity were pre- sent in the gastrointestinal tract, urinary bladder, mouth, and esophagus, with lesser amounts in the blood, liver, and salivary glands. At 24 h, most of the re- sidual radioactivity was present in the mouth, salivary glands, gastrointestinal tract, or urinary bladder. 4.2. Mechanism of Toxicity CS is an SN2 alkylating agent and, therefore, reacts directly with nucleo- philic compounds (Cucinell et al. 1971). Consequently, sulfhydryl-containing enzymes and other biologic compounds are prime targets. Most notably, CS reacts rapidly with the disulfhydryl form of lipoic acid, a coenzyme in the py- ruvate decarboxylase pathway. In in vitro studies, CS reacted readily with cyste- ine, N-acetyl L-cysteine, glutathione, dithiothreitol, and lipoic acid, and had first-order reaction constants of 0.33, 0.42, 0.85, 4.88, and 10.4, respectively. Incubation of rat liver homogenate with CS for 5 min (ethanol-buffer; pH 7.4; 37°C) resulted in a 59% decrease in the initial amount of glutathione, with 26% of the depletion occurring spontaneously (non-enzymatically) (Rietveld et al. 1986). Binding to glutathione in vivo was confirmed by enhanced urinary thi- oether excretion in rats after intraperitoneal administration of CS; the thioether was identified as 2-chlorobenzylmercapturic acid (Rietveld et al. 1983, 1986). In another study, rats administered CS intraperitoneally at a dose that was 120% of the LD50 became moribund approximately 30 min after injection (most likely 

352 Acute Exposure Guideline Levels due to the relatively slow generation of cyanide from the malononitrile metabo- lite). The role of cyanide in CS-induced lethality is supported by the observation that administration of thiosulfate intraveneously reduced mortality by 65% com- pared with control rats (21/32 exposed rats survived vs. 1/11 control rats). Intra- venous administration of CS at 8 mg/kg in dogs resulted in a rapid drop in the plasma sulfhydryl concentration, which returned to normal within approximately 3 h (Cucinell et al. 1971). 4.3. Other Relevant Information 4.3.1. Species Variability CS is a potent acute irritant. Ocular and pulmonary toxicity results from direct contact with CS and its associated alkylating properties; therefore, the mechanism of action is not expected to vary greatly between species. Ballantyne and Swantson (1978) calculated LCT50 values of 88,480 mg-min/m3 for rats, 67,200 mg-min/m3 for guinea pigs, 54,090 mg-min/m3 for rabbits, and 50,010 mg-min/m3 for mice. These values are well within a factor of two of each other. 4.3.2. Susceptible Populations CS is an irritant and the mechanism of toxicity is a direct contact effect; therefore, the mechanism of action is not expected to vary greatly between indi- viduals. The reactions in people with jaundice, hepatitis, or peptic ulcer or those that were 50-60 years old were similar to those of “normal” volunteers when exposed at highly irritating concentration of CS for short durations (Gutentag et al. 1960; Punte et al. 1963). Subjects with a history of drug allergies or sensitivi- ties, hay fever, or asthma also tolerated exposure to CS and were similar to nor- mal subjects, but the group with pre-existing conditions had a higher percentage of individuals with more severe chest symptoms (many of them laying prostrate on the ground for several minutes). However, no wheezing or rhonchi were heard, and recovery was as rapid as that seen in other exposure groups. 4.3.3. Concentration-Exposure Duration Relationship The concentration-exposure time relationship for many irritant and sys- temically-acting vapors and gases can be described by the equation Cn × t = k, where the exponent n ranges from 0.8 to 3.5 (ten Berge et al. 1986). The value of n for CS was determined on the basis of acute lethality data from studies of rats, mice, rabbits, guinea pigs, dogs, and monkeys (see Table 7-13). The dose- response software of ten Berge (2006) was used in the analyses (see Appendix A for details). 

Tear Gas (CS) 353 TABLE 7-13 Values of the Exponent n for Tear Gas Species n value 95% Confidence Limits Rat 0.704 0.543–0.865 Mouse 0.701 0.509–0.892 Rabbit 0.658 0.467–0.849 Guinea pig 0.559 0.018–1.099 Dog 0.356 -1.464–0.751 Monkey 0.187 -0.281–0.656 5. DATA ANALYSIS FOR AEGL-1 5.1. Human Data Relevant to AEGL-1 Several studies describe irritation in humans cause by CS (see Table 7-7); however, the severity of the effects exceeds those defined by AEGL-1. 5.2. Animal Data Relevant to AEGL-1 No animal studies of CS were available for deriving AEGL-1 values. 5.3. Derivation of AEGL-1 Values AEGL-1 values are not recommended because no studies were available in which toxicity was limited to AEGL-1 effects. Effects observed at the lowest tested concentrations exceeded the severity of those defined by AEGL-1. Ab- sence of AEGL-1 values does not imply that exposure below the AEGL-2 values are without adverse effects. 6. DATA ANALYSIS FOR AEGL-2 6.1. Human Data Relevant to AEGL-2 Five subjects exposed to CS at 0.71-0.78 mg/m3 (average concentration 0.75 mg/m3, calculated from the six interval measurements) tolerated a 60 min exposure (Beswick et al. 1972). All subjects reported ocular stinging and water- ing, increased salivation, cough, and face stinging. Other effects reported by some of the subjects included throat irritation (4 subjects), nasal stinging and running (3 subjects), mouth stinging (2 subjects), chest burning (2 subjects), nausea (2 subjects), and headache (2 subjects). Nausea was likely due to swal- lowing large amounts of saliva and the headaches were likely due to frontal si- nus irritation (Beswick et al. 1972). In another study, four subjects exposed to CS at 1.5 mg/m3 tolerated a 90 min exposure, but experienced clinical signs of irritation. One subject developed nasal irritation within 2 min, three subjects 

354 Acute Exposure Guideline Levels developed headache (at 45, 50, and 83 min), and all four experienced ocular irritation (at 20, 24, 70, and 75 min) (Punte et al. 1963). When the CS concentra- tion was gradually increased over the course of 60 min, most of the 30 subjects were able to tolerate exposure to concentrations ranging from 0.31 to 2.3 mg/m3; one subject left at 5 min because of vomiting but returned for the duration of the exposure, another vomited at 55 min of exposure, and one subject left after 8 min because of irritation (Beswick et al. 1972). The two cases of vomiting were attributed to swallowing large amounts or saliva. Clinical signs noted during the 60 min exposure included ocular, nasal, mouth, and throat irritation, nausea, chest discomfort, headache, and stinging of the face. 6.2. Animal Data Relevant to AEGL-2 Blinking, mild pulmonary congestion, and emphysema were observed in monkeys exposed to CS at 900 mg/m3 for 3 min or 1,700 mg/m3 for 5 min. Monkeys exposed at 2,500 mg/m3 for 32 min exhibited blinking, labored respi- ration, coughing, oral and nasal discharge, vomiting, decreased activity, pulmo- nary edema, and congestion (Striker et al. 1967). Mice exposed to CS at 40 mg/m3 for 5 h had rhinorrhea and lacrimation, and guinea pigs exposed at 45 mg/m3 for 5 h showed occasional sneezing during the first hour of exposure. 6.3. Derivation of AEGL-2 Value AEGL-2 values are based on human exposure to CS at 0.75 mg/m3 for 60 min (Beswick et al. 1972). All five subjects tolerated the exposure, but experi- enced ocular irritation, increased salivation, and coughing; some subjects also reported nasal, mouth, and throat irritation, nausea, and headache. An intraspe- cies uncertainty factor of 3 was applied because contact irritation is a portal-of- entry effect and is not expected to vary widely among individuals. This factor is also supported by the finding that responses of volunteers with jaundice, hepati- tis, or peptic ulcer or who were 50-60 years old were similar to those of “nor- mal” volunteers when exposed at a highly irritating concentration of CS for short durations. The ability to tolerate CS at 14-73 mg/m3 and the recovery time in volunteers with a history of drug allergies, seasonal allergies, asthma, or drug sensitivity were similar to “normal” volunteers; although more severe chest symptoms were reported in the volunteers with pre-existing conditions (Gutent- ag et al. 1960; Punte et al. 1963). An interspecies uncertainty factor of 1 was applied because the study was conducted in humans. A modifying factor of 3 was also used because the effects observed at 0.75 mg/m3 were considered AEGL-2 effects (watering eyes could impair escape). Time scaling was not per- formed because ocular irritation is a function of direct contact with the CS and is unlikely to increase with duration of exposure at this level of severity (NRC 2001). AEGL-2 values for CS are presented in Table 7-14, and the calculations are in Appendix B. 

Tear Gas (CS) 355 TABLE 7-14 AEGL-2 Values for Tear Gas 10 min 30 min 1h 4h 8h 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 These values are supported by the Punte et al. (1963) study in which four subjects tolerated a 90-min exposure at 1.5 mg/m3 and reported ocular and nasal irritation and headache. The values are also supported by other experiments conducted by Beswick et al. (1972). When an additional 30 subjects were ex- posed for 60 min to CS at 0.31-2.3 mg/m3, one subject left at 5 min because of vomiting, but returned for the duration of the exposure, and another vomited at 55 min of exposure. Both cases of vomiting were attributed to swallowing large amounts or saliva. One subject voluntarily left after 8 min because of irritation; this subject was exposed at 0.56-0.86 mg/m3 (AEGL-2 values are below this range). 7. DATA ANALYSIS FOR AEGL-3 7.1. Human Data Relevant to AEGL-3 No human studies of CS were available for deriving AEGL-3 values. 7.2. Animal Data Relevant to AEGL-3 Animal lethality data are available for rats, mice, rabbits, guinea pigs, dogs, and monkeys exposed to varying concentrations of CS for varying dura- tions (McNamara et al. 1969; Ballantyne and Callaway 1972; Ballantyne and Swanston 1978). Exposure durations ranged from 5 to 300 min and concentra- tions of CS ranged from 37 to 5,176 mg/m3. Mortality incidences ranged from 0 to 100%, depending on concentration-duration pairings. The experimental pa- rameters are summarized in Tables 7-8, 7-10, 7-11, and 7-12. 7.3. Derivation of AEGL-3 Values Using the rat, mouse, rabbit, guinea pig, dog, and monkey data sets of McNamara et al. (1969), Ballantyne and Callaway (1972), and Ballantyne and Swanston (1978), the lethality threshold for CS at each AEGL-3 exposure dura- tion was calculated using the probit analysis-based dose-response program of ten Berge (2006) (see Appendix A). The threshold for lethality was set at the LC01. The rat, mouse, and rabbit data all yielded similar time-scaling values and AEGL-3 values (see Appendix A). Large variances in the dog and monkey data precluded calculation of 95% confidence intervals. The rat data set was used to 

356 Acute Exposure Guideline Levels derive AEGL-3 values, because it yielded values that were the most consistent with the available human data. Time scaling was performed using the equation Cn × t = k, where the exponent n ranges from 0.8 to 3.5 (ten Berge et al. 1986). An empirical value for n of 0.70 was determined on the basis of the rat data. The 4-h AEGL-3 value was adopted as the 8-h AEGL-3 value because time scaling yielded an 8-h value inconsistent with the AEGL-2 values, which were derived from a rather robust human dataset. A total uncertainty factor of 10 was applied. A factor of 3 was used to account for interspecies differences, because clinical signs are likely caused by a direct chemical effect on the tissues and this type of portal-of-entry effect is unlikely to vary greatly between species. Furthermore, calculated LCt50 values for different species are all well within a factor of 2 of each other (88,480 mg-min/m3 for rats, 67,200 mg-min/m3 for guinea pigs, 54,090 mg-min/m3 for rabbits, and 50,010 mg-min/m3 for mice) (Ballantyne and Swanston 1978). An uncertainty factor of 3 was used to account for intraindi- vidual variability because contact irritation is a portal-of-entry effect and is not expected to vary widely among individuals. As noted above in support of the AEGL-2 values, a factor of 3 is also supported by the results of studies by Punte et al. (1963) and Gutentag et al. (1960) in subjects with pre-existing conditions. AEGL-3 values for CS are presented in Table 7-15, and calculations presented in Appendix B. The AEGL-3 values are considered protective. No mortality was noted in rats, rabbits, or mice exposed to CS at 1,802, 1,434, or 4,250 mg/m3 for 10 min, respectively (Ballantyne and Swanston 1978), suggesting that the 10-min AEGL-3 of 140 mg/m3 is appropriate. Similarly, no deaths were observed in 10 mice or five guinea pigs exposed at 40 or 44.7 mg/m3, respectively, for 5 h (Bal- lantyne and Callaway 1972). One of 10 rats died after exposure to CS at 37 mg/m3 for 5 h (Ballantyne and Callaway 1972). These data support the 4-h AEGL-3 of 1.5 mg/m3. 8. SUMMARY OF AEGLs 8.1. AEGL Values and Toxicity End Points AEGL-1 values were not recommended because effects exceeding the se- verity of AEGL-1 were detected at the lowest concentrations tested. AEGL-2 values are based on irritation in humans and AEGL-3 values are based on an estimated threshold for lethality in rats. AEGL values for tear gas are presented in Table 7-16. TABLE 7-15 AEGL-3 Values for Tear Gas 10 min 30 min 1h 4h 8h 140 mg/m3 29 mg/m3 11 mg/m3 1.5 mg/m3 1.5 mg/m3 

Tear Gas (CS) 357 8.2. Other Standards and Guidelines Standards and guidelines for CS are presented in Table 7-17. Differences between the emergency response planning guideline 3 (ERPG-3) and the 1-h AEGL-3 value and between the immediately dangerous to life or health and 30- min AEGL-2 values are due to differences in the point of departure and uncer- tainty factors. The ERPG-3 is based on a 1-h LC50 of 1,000 mg/m3 and undis- closed uncertainty factors (AIHA 2008). The IDLH is based on a Department of Army safety guide, which reported that a 2-min exposure to CS at 2-10 mg/m3 was considered intolerable by six of 15 people (NIOSH 1996). TABLE 7-16 AEGL Values for Tear Gas Exposure Duration Classification 10 min 30 min 1h 4h 8h AEGL-1 NRa NRa NRa NRa NRa (nondisabling) AEGL-2 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 (disabling) AEGL-3 (lethal) 140 mg/m3 29 mg/m3 11 mg/m3 1.5 mg/m3 1.5 mg/m3 a Not recommended. Absence of an AEGL-1 value does not imply that exposure below the AEGL-2 value is without adverse effects. The severity of effects observed at the low- est tested concentrations exceeded those defined by AEGL-1. TABLE 7-17 Standards and Guidelines for Tear Gas Exposure Duration Guideline 10 min 30 min 1h 4h 8h AEGL-1 NR NR NR NR NR AEGL-2 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 0.083 mg/m3 AEGL-3 140 mg/m3 29 mg/m3 11 mg/m3 1.5 mg/m3 1.5 mg/m3 a 3 ERPG-1 (AIHA) 0.005 mg/m a ERPG-2 (AIHA) 0.1 mg/m3 a ERPG-3 (AIHA) 25 mg/m3 b 3 IDLH (NIOSH) 2 mg/m PEL-TWA (OSHA)c 0.4 mg/m3 d REL-TWA (NIOSH) 0.4 mg/m3 e 3 TLV-STEL (ACGIH) 0.005 ppm (0.4 mg/m ) MAC (The Netherlands)f 0.4 mg/m3 a ERPG (emergency response planning guidelines, American Industrial Hygiene Associa- tion [AIHA 2008]). ERPG-1 is the maximum airborne concentration below which it is believed nearly all individuals could be exposed for up to 1 h without experiencing other than mild, transient adverse health effects or without perceiving a clearly defined objectionable odor. The 

358 Acute Exposure Guideline Levels ERPG-1 for CS is based on one of 10 individuals reporting a burning or itching sensation at 0.0004 mg/m3 and a calculated EC50 of 0.004 mg/m3. ERPG-2 is the maximum airborne concentration below which it is believed nearly all individuals could be exposed for up to 1 h without experiencing or developing irreversi- ble or other serious health effects or symptoms that could impair an individual’s ability to take protective action. The ERPG-2 for CS is based on human irritation data. ERPG-3 is the maximum airborne concentration below which it is believed nearly all individuals could be exposed for up to 1 h without experiencing or developing life- threatening health effects. The ERPG-3 for tear gas is based on animal lethality data (ab approximate 1-h LC50). b IDLH (immediately dangerous to life or health, National Institute for Occupational Safe- ty and Health) (NIOSH 1994) represents the maximum concentration from which one could escape within 30 min without any escape-impairing symptoms or any irreversible health effects. c PEL-TWA (permissible exposure limit – time-weighted average, Occupational Health and Safety Administration) (29CFR Part1910.1000 [1996]) is the time-weighted average concentration for a normal 8-h workday and a 40-h workweek, to which nearly all work- ers may be repeatedly exposed, day after day, without adverse effect. d REL-TWA (recommended exposure limit – time-weighted average, National Institute for Occupational Safety and Health) (NIOSH 2011) is defined analogous to the OSHA PEL-TWA. e TLV-STEL (threshold limit value – short-term exposure limit, American Conference of Governmental Industrial Hygienists) (ACGIH 2012) is defined as a 15-min TWA expo- sure which should not be exceeded at any time during the workday even if the 8-h TWA is within the TLV-TWA. Exposures above the TLV-TWA up to the STEL should not be longer than 15 min and should not occur more than four times per day. There should be at least 60 min between successive exposures in this range. f MAC (maximaal aanvaarde concentratie [maximal accepted concentration], Dutch Ex- pert Committee for Occupational Standards, The Netherlands) (MSZW 2004) is defined analogous to the OSHA PEL-TWA. 8.3. Data Adequacy and Research Needs Inadequate data are available to derive AEGL-1 values for CS. Adequate human data are available for deriving AEGL-2 values, and animal data are available for deriving AEGL-3 values. Additional data providing information at exposures that produce minimal irritation would be useful for deriving AEGL-1 values. 9. REFERENCES ACGIH (American Conference of Government and Industrial Hygienists). 1991. o-Chlorobenzylidene Malononitrile (CAS Reg. No. 2698-41-1). Pp. 275-278 in Doc- umentation of the Threshold Limit Values and Biological Exposure Indices, 6th Ed. American Conference of Government and Industrial Hygienists, Cincinnati, OH. ACGIH (American Conference of Government and Industrial Hygienists). 2012. TLVs and BEIs Based on the Documentation of the Threshold Limit Values for Chemical Sub- stances and Physical Agents and Biological Exposure Indices: o-Chlorobenzylidene 

Tear Gas (CS) 359 malononitrile (CAS Reg. No. 2698-41-1). American Conference of Government and Industrial Hygienists, Cincinnati, OH. AIHA (American Industrial Hygiene Association). 2008. Emergency Response Planning Guidelines: o-Chlorobenzylidene malononitrile (CAS Reg. No. 2698-41-1). Amer- ican Industrial Hygiene Association, Fairfax, VA. Anderson, P.J., G.S. Lau, W.R. Taylor, and J.A. Critchley. 1996. Acute effects of the potent lacrimator o-chlorobenzylidene malononitrile (CS) tear gas. Hum. Exp. Toxicol. 15(6):461-465. Ballantyne, B. 1977. Riot control agents: Biomedical and health aspects of the use of chemicals in civil disturbances. Pp. 7-41 in Medical Annual: The Year Book of Treatment, R.B. Scott, and J. Fraser, eds. Bristol: John Wright. Ballantyne, B., and F.W. Beswick. 1972. On the possible relationship between diarrhea and o-chlorobenzylidene malononitrile (CS). Med. Sci. Law 12(2):121-128. Ballantyne, B., and S. Callaway. 1972. Inhalation toxicology and pathology of animals exposed to o-chlorobenzylidene malononitrile. Med. Sci. Law 12(1):43-65. Ballantyne, B., and D.W. Swanston. 1978. The comparative acute mammalian toxicity of 1-chloroacetophenone (CN) and 2-chlorobenzylidene malononitrile (CS). Arch. Toxicol. 40(2):75-95. Bauchinger, M., and E. Schmid. 1992. Clastogenicity of 2-chlorobenzylidene malonitrile (CS) in V79 Chinese hamster cells. Mutat. Res. 282(4):231-234. Beswick, F.W., P. Holland, and K.H. Kemp. 1972. Acute effects of exposure to ortho- chlorobenzilidene malononitrile (CS) and the development of tolerance. Br. J. Ind. Med. 29(3):298-306. Blain, P.G. 2003. Tear gases and irritant incapacitants. 1-chloroacetophenone, 2-chlorobenzylidene malononitrile and dibenz[b,f]-1,4-oxazepine. Toxicol. Rev. 22(2):103-110. Breakell, A., and G.G. Bodiwala. 1998. CS gas exposure in a crowded night club: The consequences for an accident and emergency department. J. Accid. Emerg. Med. 15(1):56-57. Brewster, K., J.M. Harrison, L. Leadbeater, J. Newman, and D.G. Upshall. 1987. The fate of 2-chlorobenzylidene malononitrile (CS) in rats. Xenobiotica 17(8):911-924. Cole, T.J., J.E. Cotes, G.R. Johnson, H. de V. Martin, J.W. Reed, and M.J. Saunders. 1975. Comparison of effects of ammonia and CS aerosol upon exercise ventilation and cardiac frequency in healthy men. J. Physiol. 252(2):28P-29P. Cole, T.J., J.E. Cotes, G.R. Johnson, H. de V. Martin, J.W. Reed, and M.J. Saunders. 1977. Ventilation, cardiac frequency and pattern of breathing during exercise in men exposed to o-chlorobenzylidene malononitrile (CS) and ammonia gas in low concentrations. Q. J. Exp. Physiol. Cogn. Med. Sci. 62(4):341-351. Colgrave, H.F., and J.M. Creasey. 1975. Ultrastructure of rat lungs following exposure to o-chlorobenzylidene malononitrile (CS). Med. Sci. Law. 15(3):187-197. Corson, B.B., and R.W. Stoughton. 1928. Reaction of alpha, beta-unsaturated dinitriles. J. Am. Chem. Soc. 50(10):2825-2837. Cotes, J.E., L.R. Evans, G.R. Johnson, H. de V. Martin, and J.W. Reed. 1972. The effect of CS aerosol upon exercise ventilation and cardiac frequency in healthy men. J. Physiol. 222(1):77P-78P. Craig, F.N., W.V. Blevin, and E.G. Cummings. 1960. Breathing Patterns during Human Exposure to CS. Technical Report No. CWLR 2399. AD 318487. US Army Chem- ical Warfare Laboratories, Army Chemical Center, MD. June 1960. 

360 Acute Exposure Guideline Levels Cucinell, S.A., K.C. Swentzel, P. Biskup, H. Snodgrass, S. Lovre, W. Stark, L. Feinsil- ver, and R. Vocci. 1971. Biochemical interactions and metabolic fate of riot- control agents. Fed. Proc. 30(1):86-91. Euripidou, E., R. MacLehose, and A. Fletcher. 2004. An investigation into the short term and medium term health impacts of personal incapacitant sprays. A follow up of patients reported to the National Poisons Information Service (London). Emerg. Med. J. 21(5):548-552. Gaskins, J.R., R.M. Hehir, D.F. McCaulley, and E.W. Ligon. 1972. Lacrimating agents (CS and CN) in rats and rabbits: Acute effects on mouth, eyes, and skin. Arch. En- viron. Health 24(6):449-454. Gray, P.J. 2000. Is CS spray dangerous? Formulation affects toxicity. BMJ 321(7252):46- 47. Gutentag, P.J., J. Hart, E.J. Owens, and C.L. Punte. 1960. The Evaluation of CS Aerosol as a Riot Control Agent in Man. Technical Report CWLR 2365. AD 316 686. US Army Chemical Warfare Laboratories, Army Chemical Center, MD. April 1960. Haber, F.R. 1924. On the history of the gas war. Pp. 76-92 Five Lectures from the Years 1920-1923 [in German]. Berlin, Germany: Verlag von Julius Springer. Himsworth, H. 1969. Report of the Enquiry into the Medical and Toxicological Aspects of CS (Orthochlorobenzylidene Malononitrile). Part I. Enquiry into the Medical Situation Following the Use of CS in Londonderry on 13th and 14th August, 1969. Cmnd 4173. London: H.M.S.O. HSDB (Hazardous Substances Data Bank). 2005. 2-Chlorobenzalmalononitrile (CAS Reg. No. 2698-41-1). TOXNET Specialized Information Services, US National Library of Medicine: Bethesda, MD [online]. Available: http://toxnet.nlm.nih.gov/ [accessed Dec. 30, 2013]. Hu, H., and D. Christiani. 1992. Reactive airways dysfunction after exposure to tear gas. Lancet 339(8808):1535. Hu, H., J. Fine, P. Epstein, K. Kelsey, P. Reynolds, and B. Walker. 1989. Tear gas- harassing agent or toxic chemical weapon? J. Am. Med. Assoc. 262(5):660-663. Kane, L.E., C.S. Barrow, and Y. Alarie. 1979. A short-term test to predict acceptable levels of exposure to airborne sensory irritants. Am. Ind. Hyg. Assoc. J. 40(3):207- 229. Karagama, Y.G., J.R. Newton, and C.J. Newbegin. 2003. Short-term and long-term phys- ical effects of exposure to CS spray. J. R. Soc. Med. 96(4):172-174. Leadbeater L. 1973. The absorption of ortho-chlorobenzylidenemalononitrile (CS) by the respiratory tract. Toxicol. Appl. Pharmacol. 25(1):101-110. Marrs, T.C., E. Clifford, and H.F. Colgrave. 1983a. Late inhalation toxicology and pa- thology produced by exposure to a single dose of 2-chlorobenzylidene malono- nitrile (CS) in rats and hamsters. Med. Sci. Law 23(4):257-265. Marrs, T.C., H.F. Colgrave, N.L. Cross, M.F. Gazzard, and R.F. Brown. 1983b. A repeat- ed dose study of the toxicity of inhaled 2-chlorobenzylidene malononitrile (CS) aerosol in three species of laboratory animal. Arch. Toxicol. 52(3):183-198. McDonald, E.C., and R.T. Mahon. 2002. US Marine Corps Amphibious Reconnaissance (Recon) students requiring hospitalization with pulmonary edema after strenuous exercise following exposure to o-chloro-benzylidenemalonitrile. Mil. Med. 167(11):iii-iv. McElhatton, P.R., S. Sidhu, and S.H. Thomas. 2004. Exposure to CS gas in pregnancy. J. Toxicol. Clin. Toxicol. 42(4):547. 

Tear Gas (CS) 361 McGregor, D.B., A. Brown, P. Cattanach, I. Edwards, D. McBride, and W.J. Caspary. 1988. Responses of the L5178Y tk+/tk- mouse lymphoma cell forward mutation assay. II. 18 coded chemicals. Environ. Mol. Mutagen. 11(1):91-118. McNamara, B.P., E.J. Owens, J.T. Weimer, T.A. Ballard, and F.J. Vocci. 1969. Toxicol- ogy of Riot Control Chemicals CS, CN, and DM. Edgewood Arsenal Technical Report EATR-4309 AD 862 075. US Department of the Army, Edgewood Arsenal Medical Research Laboratory, Edgewood Arsenal, MD. November 1969. Meshram, G.P., R.P. Malini, and K.M. Rao. 1992. Mutagenicity evaluation of riot control agent o-chlorobenzylidene malononitrile (CS) in the Ames Salmonella/microsome test. J. Appl. Toxicol. 12(5):377-384. MSZW (Ministerie van Sociale Zaken en Werkgelegenheid). 2004. Nationale MAC-lijst 2004: o-Chloorbenzilydeenmalonitril. Den Haag: SDU Uitgevers [online]. Availa- ble: http://www.lasrook.net/lasrookNL/maclijst2004.htm [accessed Jan. 2, 2014]. NIOSH (National Institute for Occupational Safety and Health). 1994. Documentation for Immediately Dangerous to Life or Health Concentrations (IDLHs): o- Chlorobenzylidene malononitrile. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, Cincinnati, OH [online]: http://www.cdc.gov/niosh/idlh/2698 411.html [accessed Dec. 30, 2013] NIOSH (National Institute for Occupational Safety and Health). 2011. NIOSH Pocket Guide to Chemical Hazards: o-Chlorobenzylidene malononitrile. US Department of Health and Human Services, Centers for Disease Control and Prevention, Na- tional Institute for Occupational Safety and Health, Cincinnati, OH [online]. Available: http://www.cdc.gov/niosh/npg/npgd0122.html [accessed Dec. 30, 2013]. 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. NTP (National Toxicology Program). 1990. Toxicology and Carcinogenesis Studies of CS2 (94% o-Chlorobenzalmalononitrile Cas No. 2698-41-1) in F344/N Rats and B6C3F1 Mice (Inhalation Studies). Technical Report No. 377. NIH 90-2832. US Department of Health and Human Services, Public Health Service, National Insti- tute of Health, National Toxicology Program, Research Triangle Park, NC [online]. Available: http://ntp.niehs.nih.gov/ntp/htdocs/LT_rpts/tr377.pdf [accessed Dec. 30, 2013]. O’Neil, M.J., P.E. Heckelman, C.B. Koch, and K.J. Roman, eds. 2006. o- Chlorobenzylidenemalononitrile. P. 352 in The Merck Index: An Encyclopedia of Chemicals, Drugs and Biologicals, 14th Ed. Whitehouse Station, NJ: Merck. Owens, E.J., and C.L. Punte. 1963. Human respiratory and ocular irritation studies utiliz- ing o-chlorobenzilidene malononitrile aerosols. Am. Ind. Hyg. Assoc. J. 24:262- 264. Paradowski, M. 1979. Metabolism of toxic doses of o-chlorobenzylidene malononitrile in rabbits. Pol. J. Pharmacol. Pharm. 31:563-572. Park, S., and S.T. Giammona. 1972. Toxic effects of tear gas on an infant following pro- longed exposure. Am. J. Dis. Child. 123(3):245-246. Punte, C.L., J.T. Weimer, T.A. Ballard, and J.L. Wilding. 1962. Toxicologic studies on o-chlorobenzylidene malononitrile. Toxicol. Appl. Pharmacol. 4(5):656-662. 

362 Acute Exposure Guideline Levels Punte, C.L., E.J. Owens, and P.J. Gutentag. 1963. Exposures to ortho-chlorobenzylidene malononitrile: Controlled human exposures. Arch. Environ. Health 6(3):72-80. Rengstorff, R.H. 1969. The effects of the riot control agent CS on visual acuity. Milit. Med. 134(3):219-221. Rietveld, E.C., L.P. Delbressine, T.H. Waegemaekers, and F. Seutter-Berlage. 1983. 2-Chlorobenzylmercapturic acid, a metabolite of the riot control agent 2-chlorobenzylidene malononitrile (CS) in the rat. Arch. Toxicol. 54(2):139-144. Rietveld, E.C., M.M. Hendrikx, and F. Seutter-Berlage. 1986. Glutathione conjugation of chlorobenzylidene malononitriles in vitro and the biotransformation to mercapturic acids in rats. Arch. Toxicol. 59(4):228-234. Rinehart, W.E., and T. Hatch. 1964. Concentration-time product (CT) as an expression of dose in sublethal exposures to phosgene. Am. Ind. Hyg. Assoc. J. 25:545-553. Roth, V.S., and A. Franzblau. 1996. RADS after exposure to a riot-control agent: A case report. J. Occup. Environ. Med. 38(9):863-865. Salassidis, K., E. Schmid, and M. Bauchinger. 1991. Mitotic spindle damage induced by 2-chlorobenzylidene malonitrile (CS) in V79 Chinese hamster cells examined by differential staining of the spindle apparatus and chromosomes. Mutat. Res. 262(4):263-266. Schmid, E., and M. Bauchinger. 1991. Analysis of the aneuploidy inducing capacity of 2-chlorobenzylidene malonitrile (CS) and metabolites in V79 Chinese hamster cells. Mutagenesis 6(4):303-305. Smith, J., and I. Greaves. 2002. The use of chemical incapacitant sprays: A review. J. Trauma 52(3):595-600. Striker, G.E., C.S. Streett, D.F. Ford, L.H. Herman, and D.R. Helland. 1967. A Clinico- pathologic Study of the Effects of Riot Control Agents on Monkey: IV. o- Chlorobenzylidene Malononitrile (CS) Grenade. Technical Report EATR 4071. AD-808732. Edgewood Arsenal Aberdeen Proving Ground, MD. January 1967. ten Berge, W.F. 2006. Concentration-time Response in Acute Inhalation Toxicity. DoseRespBeta2006 [online]. Available: http://home.wxs.nl/~wtberge/doseresp.html [accessed July 5, 2013]. 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. Thomas, R.J., P.A. Smith, D.A. Rascona, J.D. Louthan, and B. Gumpert. 2002. Acute pulmonary effects from o-chlorobenzylidenemalonitrile “tear gas”: A unique expo- sure outcome unmasked by strenuous exercise after a military training event. Mil. Med. 167(2):136-139. Upshall, D.G. 1973. Effects of o-chlorobenzylidene malononitrile (CS) and the stress of aerosol inhalation upon rat and rabbit embryonic development. Toxicol. Appl. Pharmacol. 24(1):45-59. US Army Chemical School. 2005. Military compounds and their properties. Chapter 3 in Potential Military Chemical/Biological Agents and Compounds. FM 3-11.9, MCRP 3-37.1B, NTRP, 3-11.32, AFTTP(I) 3.255. January 2005 [online]. Availa- ble: http://www.fas.org/irp/doddir/army/fm3-11-9.pdf [accessed Dec. 30, 2013]. von Däniken, A., U. Friederich, W.K. Lutz, and C. Schlatter. 1981. Tests for mutagenici- ty in Salmonella and covalent binding to DNA and protein in the rat of the riot control agent o-chlorobenzylidene malononitrile (CS). Arch. Toxicol. 49(1):15-27. WHO (World Health Organization). 1970. Health Aspects of Chemical and Biological Weapons. Report of a World Health Organization Group of Consultants, WHO, 

Tear Gas (CS) 363 Geneva, Switzerland [online]. Available: http://whqlibdoc.who.int/publications/24 039.pdf [accessed Dec. 30, 2013]. Wild, D., K. Eckhardt, D. Harnasch, and M.T. King. 1983. Genotoxicity study of CS (ortho-chlorobenzylidenemalononitrile) in Salmonella, Drosophila, and mice. Failure to detect mutagenic effects. Arch. Toxicol. 54(2):167-170. Ziegler-Skylakakis, K., K.H. Summer, and U. Andrae. 1989. Mutagenicity and cytotoxi- city of 2-chlorobenzylidene malonitrile (CS) and metabolites in V79 Chinese ham- ster cells. Arch. Toxicol. 63(4):314-319. 

364 Acute Exposure Guideline Levels APPENDIX A TIME-SCALING CALCULATIONS FOR TEAR GAS The relationship between dose and time for any given chemical is a function of the physical and chemical properties of the substance and the unique toxicologic and pharmacologic properties of the individual substance. Historically, the relation- ship according to Haber (1924), commonly called Haber’s Law or Haber’s Rule (C × t = k, where C = exposure concentration, t = exposure duration, and k = a constant), has been used to relate exposure concentration and duration to effect (Rinehart and Hatch 1964). This concept states that exposure concentration and exposure duration may be reciprocally adjusted to maintain a cumulative exposure constant (k) and that this cumulative exposure constant will always reflect a specific quantitative and qualitative response. This inverse relationship of concentration and time may be valid when the toxic response to a chemical is equally dependent on the concentra- tion and the exposure duration. However, an assessment by ten Berge et al. (1986) of LC50 data for certain chemicals revealed chemical-specific relationships between exposure concentration and exposure duration that were often exponential. This rela- tionship can be expressed by the equation Cn × t = k, where n represents a chemical- specific (and even a toxic end point-specific) exponent. The relationship described by this equation is basically the form of a linear regression analysis of the log-log transformation of a plot of C vs. t. ten Berge et al. (1986) examined the airborne concentration (C) and short-term exposure duration (t) relationship relative to death for approximately 20 chemicals and found that the empirically derived value of n ranged from 0.8 to 3.5 among this group of chemicals. Hence, the value of the expo- nent n in the equation Cn × t = k quantifies the relationship between exposure con- centration and exposure duration. Haber's Rule is the special case where n = 1. As the value of n increases, the plot of concentration vs. time yields a progressive de- crease in the slope of the curve. An n of 0.70 mg/m3 for CS was obtained by analysis of lethality data in rats (McNamara et al. 1969; Ballantyne and Callaway 1972; Ballantyne and Swanston 1978) using the software of ten Berge (2006). This exposure-time relationship for lethality was considered appropriate for deriving AEGL-3 values. The 4-h AEGL-3 value was adopted as the 8-h AEGL-3 value because time scaling yielded an 8-h value inconsistent with the AEGL-2 values that were derived from robust human data. 

TABLE A-1 Results of ten Berge Program (1% Lethality) LC01 Point Estimate, mg/m3 (95% confidence limits) Species Exponent n 10 min 30 min 1h 4h 8h Reference(s) Rat 0.704 1,385 290 109 15 5.6 McNamara et al. 1969; Ballantyne (0.543–0.865) (477–2,500) (97–496) (32–196) (3.1–35) (-0.93–15) and Callaway 1972; Ballantyne and Swanston 1978 Mouse 0.701 998 208 77 11 4.0 McNamara et al. 1969; Ballantyne (0.509–0.892) (208–1,899) (36–404) (11–166) (-0.86–3.2) (-0.23–15) and Callaway 1972; Ballantyne and Swanston 1978 Rabbit 0.658 656 124 43 5.2 1.8 McNamara et al. 1969; Ballantyne (0.467–0.849) (227–1,136) (28–249) (7.0–103) (0.40–19) (0.094–8.6) and Callaway 1972; Ballantyne and Swanston 1978 Guinea pig 0.559 3.65 0.51 0.15 0.012 0.0036 McNamara et al. 1969; Ballantyne (0.018–1.099) (0–100) (0–25) (0–12) (0–3.3) (0–1.8) and Callaway 1972; Ballantyne and Swanston 1978 Dog 0.356 a a a a a McNamara et al. 1969 349 7,604 53,150 2,597,000 18,150,000 (-1.464–0.751) Monkey 0.187 a a a a a McNamara et al. 1969; Striker et al. 1967 26 0.075 0.0018 0.0000011 0.000000028 (-0.281–0.656) Monkey 2.123 (-21–25) a a a a a McNamara et al. 1969 11 6.6 4.7 2.5 1.8 a Large variances precluded estimating 95% confidence limits. 365 

366 Acute Exposure Guideline Levels TABLE A-2 Data in Rats Used in Log Probit Model Date: 02 October 2008 Time: 09:11:09 Sequence No. Concentration (mg/m3) Minutes Exposed Responded 1 560 25 10 1 2 543 35 10 2 3 489 45 10 3 4 454 55 10 5 5 500 60 10 2 6 500 80 10 6 7 500 90 10 8 8 750 30 8 0 9 150 120 8 0 10 3,950 5 10 0 11 4,760 5 10 0 12 4,250 10 10 1 13 4,330 10 10 1 14 4,150 15 10 0 15 5,176 15 10 7 16 4,000 20 10 9 17 4,300 20 10 8 18 1,802 10 20 0 19 1,806 45 20 8 20 1,911 45 20 9 21 2,629 60 21 20 22 2,699 60 20 20 23 37 300 10 1 Used Probit Equation Y = B0 + B1*X1 + B2*X2 X1 = conc mg/m3, ln-transformed X2 = minutes, ln-transformed ChiSquare = 50.11 Degrees of freedom = 20 Probability Model = 2.13E-04 Ln(Likelihood) = -47.54 B 0 = -9.6233E+00 Student t = -3.9484 B 1 = 1.1705E+00 Student t = 5.6748 B 2 = 1.6634E+00 Student t = 5.6382 

Tear Gas (CS) 367 Variance B 0 0 = 5.9402E+00 Covariance B 0 1 = -4.8485E-01 Covariance B 0 2 = -6.7032E-01 Variance B 1 1 = 4.2542E-02 Covariance B 1 2 = 4.9302E-02 Variance B 2 2 = 8.7045E-02 Estimation ratio between regression coefficients of ln(conc) and ln(minutes) Point estimate = 0.704 Lower limit (95% CL) = 0.543 Upper limit (95% CL) = 0.865 Estimation of conc mg/m3 at response of 1% Minutes = 10 Point estimate conc mg/m3 = 1.385E+03 for response of 1% Lower limit (95% CL) conc mg/m3 = 4.772E+02 for response of 1% Upper limit (95% CL) conc mg/m3 = 2.500E+03 for response of 1% Estimation of conc mg/m3 at response of 1% Minutes = 30 Point estimate conc mg/m3 = 2.906E+02 for response of 1% Lower limit (95% CL) conc mg/m3 = 9.659E+01 for response of 1% Upper limit (95% CL) conc mg/m3 = 4.963E+02 for response of 1% Estimation of conc mg/m3 at response of 1% Minutes = 60 Point estimate conc mg/m3 = 1.085E+02 for response of 1% Lower limit (95% CL) conc mg/m3 = 3.223E+01 for response of 1% Upper limit (95% CL) conc mg/m3 = 1.958E+02 for response of 1% Estimation of conc mg/m3 at response of 1% Minutes = 120 Point estimate conc mg/m3 = 4.052E+01 for response of 1% Lower limit (95% CL) conc mg/m3 = 1.021E+01 for response of 1% Upper limit (95% CL) conc mg/m3 = 8.137E+01 for response of 1% Estimation of conc mg/m3 at response of 1% Minutes = 240 Point estimate conc mg/m3 = 1.513E+01 for response of 1% Lower limit (95% CL) conc mg/m3 = 3.122E+00 for response of 1% Upper limit (95% CL) conc mg/m3 = 3.501E+01 for response of 1% Estimation of conc mg/m3 at response of 1% Minutes = 480 Point estimate conc mg/m3 = 5.649E+00 for response of 1% Lower limit (95% CL) conc mg/m3 = 9.345E-01 for response of 1% Upper limit (95% CL) conc mg/m3 = 1.540E+01 for response of 1% 

368 Acute Exposure Guideline Levels APPENDIX B DERIVATION OF AEGL VALUES FOR TEAR GAS Derivation of AEGL-1 Values AEGL-1 values are not recommended for CS because the effects observed at the lowest tested concentrations exceeded the severity of AEGL-1 effects. Derivation of AEGL-2 Values Key study: Beswick, F.W., P. Holland, and K.H. Kemp. 1972. Acute effects of exposure to orthochlorobenzilidene malononitrile (CS) and the development of tolerance. Br. J. Ind. Med. 29(3):298-306. Toxicity end point: Human exposure to CS at an average concentration of 0.75 mg/m3 for 60 min. All five subjects tolerated the exposure but reported ocular stinging and watering, increased salivation, coughing, and face stinging. Some subjects also reported throat irritation (4 subjects), nasal stinging and running (3 subjects), mouth stinging (2 subjects), chest burning (2 subjects), nausea (2 subjects), and headache (2 subjects). Time scaling: None. Irritation is a function of direct contact with CS and is unlikely to increase with duration of exposure at this level of severity (NRC 2001). Uncertainty factors: 1 for interspecies differences 3 for intraspecies variability; contact irritation is a portal-of-entry effect and is not expected to vary widely between individuals. Value of 3 is also supported by a study that showed that volunteers with a history of jaundice, hepatitis, or peptic ulcer or those that were 50-60 years old had responses similar to those of “normal” volunteers when exposed at a highly irritating concentration of CS for short durations. The ability to tolerate exposure to CS at 14-73 mg/m3 and the recovery time in people with a history of drug allergies, seasonal allergies, asthma, or drug sensitivity was similar to normal volunteers; although more severe chest symptoms were reported in the people with pre-existing conditions (Gutentag et al. 1960; Punte et al. 1963). Modifying factor: 3, because effects observed at 0.75 mg/m3 were AEGL-2 effects. 

Tear Gas (CS) 369 Calculations: 0.75 mg/m3 ÷ 9 = 0.083 mg/m3 (applied to all AEGL durations) Derivation of AEGL-3 Values Key studies: McNamara, B.P., E.J. Owens, J.T. Weimer, T.A. Ballard, and F.J. Vocci. 1969. Toxicology of Riot Control Chemicals CS, CN, and DM. Edgewood Arsenal Technical Report EATR-4309. US Department of the Army, Edgewood Arsenal Medical Research Laboratory, Edgewood Arsenal, MD. November 1969. Ballantyne, B., and S. Callaway. 1972. Inhalation toxicology and pathology of animals exposed to o chlorobenzylidene malononitrile. Med. Sci. Law 12(1):43-65. Ballantyne, B., and D.W. Swanston. 1978. The comparative acute mammalian toxicity of 1- chloroacetophenone (CN) and 2-chlorobenzylidene malononitrile (CS). Arch. Toxicol. 40(2):75-95. Toxicity end point: L01 for rats; calculated using probit analysis-based dose- response program of ten Berge (2006), see Appendix A. LC01 Point Estimate, mg/m3 (95% confidence limits) Exponent n 10 min 30 min 1h 4h 8h 0.704 1,385 290 109 15 5.6 (0.543-0.865) (477–2,500) (97–496) (32–196) (3.1–35) (-0.93–15) Time scaling: Cn × t = k; n = 0.70 (based on rat lethality data). No time scaling was performed for the 8-h AEGL value, because time scaling yielded a value that was inconsistent with the AEGL-2 values that were derived from robust human data. Uncertainty factors: 3 for interspecies differences. Effects from CS are likely caused by a direct chemical effect on the tissues. This type of portal-of-entry effect is unlikely to vary greatly between species. Value is also supported by calculated LCT50 values of 88,480 mg min/m3 for rats, 67,200 mg min/m3 for guinea pigs, 54,090 mg min/m3 for rabbits, and 50,010 mg min/m3 for mice (Ballantyne and Swanston 1978); values all well within a factor of two of each other. 

370 Acute Exposure Guideline Levels 3 for intraspecies variability. Effects from CS are likely caused by a direct chemical effect on the tissues. This type of portal-of-entry effect is not likely to vary greatly among individuals. Value is also supported by a study that showed that volunteers with a history of jaundice, hepatitis, or peptic ulcer or those that were 50-60 years old had responses similar to those of “normal” volunteers when exposed at a highly irritating concentration of CS for short durations. The ability to tolerate exposure to CS at 14-73 mg/m3 and the recovery time in people with a history of drug allergies, seasonal allergies, asthma, or drug sensitivity was similar to normal volunteers; although more severe chest symptoms were reported in the people with pre-existing conditions (Gutentag et al. 1960; Punte et al. 1963). Calculations: 10-min AEGL-3: 1,385 mg/m3 ÷ 10 = 140 mg/m3 30-min AEGL-3: 290 mg/m3 ÷ 10 = 29 mg/m3 1-h AEGL-3: 109 mg/m3 ÷ 10 = 11 mg/m3 4-h AEGL-3: 15 mg/m3 ÷ 10 =1.5 mg/m3 8-h AEGL-3: Set equal to the 4-h AEGL-3 of 1.5 mg/m3 

Tear Gas (CS) 371 APPENDIX C ACUTE EXPOSURE GUIDELINE LEVELS FOR TEAR GAS Derivation Summary AEGL-1 VALUES AEGL-1 values are not recommended for CS because the effects observed at the lowest tested concentrations exceeded the severity of AEGL-1 effects. AEGL-2 VALUES 10 min 30 min 1h 4h 8h 3 3 3 3 0.083 mg/m 0.083mg/m 0.083 mg/m 0.083 mg/m 0.083 mg/m3 Key reference: Beswick, F.W., P. Holland, and K.H. Kemp. 1972. Acute effects of exposure to orthochlorobenzilidene malononitrile (CS) and the development of tolerance. Br. J. Ind. Med. 29(3):298-306. Test species/Strain/Number: Humans, 5 Exposure route/Concentration/Duration: Inhalation, 0.71-0.78 mg/m3 (average: 0.75 mg/m3) for 60 min Effects: Clinical signs of irritation. Ocular stinging and watering, increased salivation, cough, and face stinging was reported in all subjects. Additional signs of irritation included throat irritation (4 subjects), nasal stinging and running (3 subjects), mouth stinging (2 subjects), chest burning (2 subjects), nausea (2 subjects), and headache (2 subjects). End point/Concentration/Rationale: Ocular, nasal, mouth, and throat irritation, coughing, nausea, and headache at 0.75 mg/m3 for 60 min. Uncertainty factors/Rationale: Total uncertainty factor: 3 Interspecies: 1, because data were from human volunteers Intraspecies: 3, contact irritation is a portal-of-entry effect and is not expected to vary widely between individuals. Value is also supported by a study that showed that volunteers with a history of jaundice, hepatitis, or peptic ulcer and those that were 50-60 years old had responses similar to those of “normal” volunteers when exposed at a highly irritating concentration of CS for short durations. The ability to tolerate the exposure to CS at 14-73 mg/m3 and the recovery time in people with a history of drug allergies, seasonal allergies, asthma, or drug sensitivity was similar to normal volunteers; although more severe chest symptoms were reported in the people with pre-existing conditions (Gutentag et al. 1960; Punte et al. 1963). Modifying factor: 3, because effects at 0.75 mg/m3 were considered AEGL-2 effects. Animal-to-human dosimetric adjustment: Not applicable Time scaling: Not applied. Irritation is a function of direct contact with the CS and is unlikely to increase with duration of exposure at this level of severity (NRC 2001). (Continued) 

372 Acute Exposure Guideline Levels AEGL-2 VALUES Continued Data adequacy: AEGL-2 values are supported by the data of Punte et al. (1963) and Beswick et al. (1972). Exposure of four subjects at 1.5 mg/m3 for 90 min resulted in ocular and nasal irritation in all subjects and headache in 3 subjects (Punte et al. 1963). When a total of 30 subjects were exposed for 60 min to gradually increasing CS concentrations ranging from 0.31-2.3 mg/m3, one subject left at 5 min because of vomiting but returned for the duration of the exposure, and another vomited at 55 min of exposure (vomiting in both cases attributed to swallowing large amounts or saliva). One subject voluntarily left the exposure after 8 min because of irritation; this subject was exposed in the range of 0.56-0.86 mg/m3, and the AEGL-2 values are below this range. AEGL-3 VALUES 10 min 30 min 1h 4h 8h 140 mg/m3 29 mg/m3 11 mg/m3 1.5 mg/m3 1.5 mg/m3 Key references: McNamara, B.P., E.J. Owens, J.T. Weimer, T.A. Ballard, and F.J. Vocci. 1969. Toxicology of Riot Control Chemicals CS, CN, and DM. Edgewood Arsenal Technical Report EATR-4309. US Department of the Army, Edgewood Arsenal Medical Research Laboratory, Edgewood Arsenal, MD. November. Ballantyne, B., and S. Callaway. 1972. Inhalation toxicology and pathology of animals exposed to o chlorobenzylidene malononitrile. Med. Sci. Law 12(1):43-65. Ballantyne, B., and D.W. Swanston. 1978. The comparative acute mammalian toxicity of 1- chloroacetophenone (CN) and 2-chlorobenzylidene malononitrile (CS). Arch. Toxicol. 40(2):75-95. Test species/Strain/Number: Rat, various strains; 8, 10, 20, or 21 per group Exposure route/Concentration/Duration: Inhalation, 37-5,175 mg/m3 for 5-300 min. Effects: Lethality End point/Concentration/Rationale: LC01 for rats calculated using probit-analysis dose- response program of ten Berge (2006). LC01 Point Estimate, mg/m3 Exponent n 10 min 30 min 1h 4h 8h 0.704 1,385 290 109 15 5.6 (0.543-0.865) (477-2,500) (97-496) (32-196) (3.1-35) (-0.93-15) Uncertainty factors/Rationale: Total uncertainty factor: 10 Interspecies: 3, effects from CS are likely caused by a direct chemical effect on the tissues. This type of portal-of-entry effect is unlikely to vary greatly between species. Value is also supported by calculated LCT50 values of 88,480 mg min/m3 for rats, 67,200 mg min/m3 for guinea pigs, 54,090 mg min/m3 for rabbits, and 50,010 mg min/m3 for mice (Ballantyne and Swanston 1978); values are all well within a factor of two of each other. Intraspecies: 3, effects from CS are likely caused by a direct chemical effect on the tissues. This type of portal-of-entry effect is not likely to vary greatly among individuals. Value is also supported by a study that showed that volunteers with a history of jaundice, hepatitis, or peptic ulcer or those that were 50-60 years old had responses similar to those 

Tear Gas (CS) 373 of “normal” volunteers when exposed at a highly irritating concentration of CS for short durations. The ability to tolerate the exposure to CS at 14-73 mg/m3 and the recovery time in people with a history of drug allergies, seasonal allergies, asthma, or drug sensitivity was similar to normal volunteers; although more severe chest symptoms were reported in the people with pre-existing conditions (Gutentag et al. 1960; Punte et al. 1963). Modifying factor: None applied Animal-to-human dosimetric adjustment: Not applicable Time scaling: Cn × t = k, where n = 0.70 based on rat lethality data. The 4-h AEGL-3 value was adopted as the 8-h AEGL-3 value because time scaling yielded an 8-h value inconsistent with the AEGL-2 values that were derived from robust human data. Data adequacy: The AEGL-3 values are considered protective. No mortality was noted in rats exposed to CS at 1,802 mg/m3 for 10 min (Ballantyne and Swanston 1978), in rabbits at 1,434 mg/m3 for 10 min (Ballantyne and Swanston 1978), or in mice and rabbits at 4,250 mg/m3 for 10 min (Ballantyne and Callaway 1972). Dividing these concentrations by a total uncertainty factor of 10, yields values ranging from 140-425 mg/m3, suggesting that the 10-min AEGL-3 is appropriate. No mortality was noted in guinea pigs exposed at 44.7 mg/m3 for 5 h or in mice exposed at 40 mg/m3 for 5 h (Ballantyne and Callaway 1972). Applying a total uncertainty factor of 10 to these concentrations yields a value of approximately 4.0 mg/m3 for 5 h. One of ten rats died when exposed at 37 mg/m3 for 5 h (Ballantyne and Callaway 1972). Dividing 37 mg/m3 by 2 to obtain an approximate threshold for lethality, yields 18.5 mg/m3; application of a total uncertainty factor of 10, yields a value of 1.9 mg/m3 for 5 h. The values derived from the 5-h data show that the AEGL-3 values are protective. 

374 Acute Exposure Guideline Levels APPENDIX D CATEGORY PLOT FOR TEAR GAS FIGURE D-1 Category plot of toxicity data and AEGL values for tear gas. 

TABLE D-1 Data Used in the Category Plot for Tear Gas Source Species Sex No. Exposures mg/m3 Minutes Category Comments AEGL-1 NR 10 AEGL AEGL-1 NR 30 AEGL AEGL-1 NR 60 AEGL AEGL-1 NR 240 AEGL AEGL-1 NR 480 AEGL AEGL-2 0.083 10 AEGL AEGL-2 0.083 30 AEGL AEGL-2 0.083 60 AEGL AEGL-2 0.083 240 AEGL AEGL-2 0.083 480 AEGL AEGL-3 140 10 AEGL AEGL-3 29 30 AEGL AEGL-3 11 60 AEGL AEGL-3 1.5 240 AEGL AEGL-3 1.5 480 AEGL Ballantyne and Calloway 1972 Guinea pig 1 45 300 2 Sneezing Ballantyne and Calloway 1972 Guinea pig 1 3,950 5 SL Mortality: 1/5 Ballantyne and Calloway 1972 Guinea pig 1 4,150 15 SL Mortality: 3/5 Ballantyne and Calloway 1972 Guinea pig 1 4,250 10 3 Mortality: 5/5 Ballantyne and Calloway 1972 Guinea pig 1 4,330 10 SL Mortality: 3/5 Ballantyne and Calloway 1972 Guinea pig 1 4,760 5 2 Mortality: 0/5 Ballantyne and Calloway 1972 Guinea pig 1 5,167 15 3 Mortality: 5/5 (Continued) 375 

376 TABLE D-1 Continued Source Species Sex No. Exposures mg/m3 Minutes Category Comments Ballantyne and Calloway 1972 Mouse 1 40 300 2 Rhinorrhea and lacrimation Ballantyne and Calloway 1972 Mouse 1 3,950 5 SL Mortality: 1/10 Ballantyne and Calloway 1972 Mouse 1 4,150 15 SL Mortality: 3/10 Ballantyne and Calloway 1972 Mouse 1 4,250 10 2 Mortality: 0/10 Ballantyne and Calloway 1972 Mouse 1 4,330 10 SL Mortality: 4/10 Ballantyne and Calloway 1972 Mouse 1 4,760 5 2 Mortality: 0/10 Ballantyne and Calloway 1972 Mouse 1 5,167 15 SL Mortality: 3/10 Ballantyne and Calloway 1972 Rabbit 1 3,950 5 2 Mortality: 0/5 Ballantyne and Calloway 1972 Rabbit 1 4,150 15 SL Mortality: 2/5 Ballantyne and Calloway 1972 Rabbit 1 4,250 10 2 Mortality: 0/5 Ballantyne and Calloway 1972 Rabbit 1 4,330 10 SL Mortality: 2/5 Ballantyne and Calloway 1972 Rabbit 1 4,760 5 2 Mortality: 0/5 Ballantyne and Calloway 1972 Rabbit 1 5,167 15 SL Mortality: 2/5 Ballantyne and Calloway 1972 Rat 1 37 300 SL Rhinorrhea, lacrimation, and mortality (1/10) Ballantyne and Calloway 1972 Rat 1 150 120 2 Mortality: 0/8 Ballantyne and Calloway 1972 Rat 1 3,950 5 2 Mortality: 0/10 Ballantyne and Calloway 1972 Rat 1 4,150 15 2 Mortality: 0/10 Ballantyne and Calloway 1972 Rat 1 4,250 10 SL Mortality: 1/10 Ballantyne and Calloway 1972 Rat 1 4,330 10 SL Mortality: 1/10 Ballantyne and Calloway 1972 Rat 1 4,760 5 2 Mortality: 0/10 Ballantyne and Calloway 1972 Rat 1 5,167 15 SL Mortality: 7/10 Ballantyne and Swanston 1978 Rabbit Female 1 846 5 2 Mortality: 0/10 

Beswick et al. 1972 Human 1 0.31 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.42 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.56 8 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.57 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.63 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.7 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.78 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.8 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 0.84 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 2 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 2.1 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 2.3 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches Beswick et al. 1972 Human 1 2.3 60 2 Ocular, nasal, throat irritation; nausea, chest discomfort, headaches McNamara et al. 1969 Dog 1 508 36 SL Mortality: 2/4 McNamara et al. 1969 Dog 1 520 45 SL Mortality: 2/4 (Continued) 377 

378 TABLE D-1 Continued Source Species Sex No. Exposures mg/m3 Minutes Category Comments McNamara et al. 1969 Dog 1 612 45 SL Mortality: 2/4 McNamara et al. 1969 Dog 1 649 30 SL Mortality: 1/4 McNamara et al. 1969 Dog 1 797 60 SL Mortality: 3/4 McNamara et al. 1969 Dog 1 833 20 2 Mortality: 0/4 McNamara et al. 1969 Dog 1 899 40 SL Mortality: 2/4 McNamara et al. 1969 Dog 1 909 60 SL Mortality: 2/4 McNamara et al. 1969 Guinea pig 1 400 5 SL Mortality: 1/10 McNamara et al. 1969 Guinea pig 1 400 10 SL Mortality: 2/10 McNamara et al. 1969 Guinea pig 1 400 15 SL Mortality: 4/10 McNamara et al. 1969 Guinea pig 1 400 25 SL Mortality: 7/10 McNamara et al. 1969 Guinea pig 1 400 30 SL Mortality: 7/10 McNamara et al. 1969 Guinea pig 1 500 20 SL Mortality: 3/10 McNamara et al. 1969 Monkey 1 381 45 SL Mortality: 2/4 McNamara et al. 1969 Monkey 1 469 24 SL Mortality: 1/4 McNamara et al. 1969 Monkey 1 612 45 SL Mortality: 1/4 McNamara et al. 1969 Monkey 1 673 30 SL Mortality: 2/4 McNamara et al. 1969 Monkey 1 699 60 SL Mortality: 1/4 McNamara et al. 1969 Monkey 1 941 60 SL Mortality: 3/4 McNamara et al. 1969 Monkey 1 1,057 60 SL Mortality: 2/4 McNamara et al. 1969 Mouse 1 683 60 SL Mortality: 14/20 McNamara et al. 1969 Mouse 1 740 50 SL Mortality: 5/20 McNamara et al. 1969 Mouse 1 800 40 SL Mortality: 5/20 McNamara et al. 1969 Mouse 1 900 30 SL Mortality: 2/20 

McNamara et al. 1969 Mouse 1 1,100 20 SL Mortality: 7/20 McNamara et al. 1969 Mouse 1 1,200 10 2 Mortality: 0/20 McNamara et al. 1969 Rabbit 1 250 40 2 Mortality: 0/4 McNamara et al. 1969 Rabbit 1 250 80 SL Mortality: 3/4 McNamara et al. 1969 Rabbit 1 267 45 2 Mortality: 0/4 McNamara et al. 1969 Rabbit 1 333 90 3 Mortality: 4/4 McNamara et al. 1969 Rabbit 1 500 30 SL Mortality: 1/4 McNamara et al. 1969 Rat 1 454 55 SL Mortality: 5/10 McNamara et al. 1969 Rat 1 500 60 SL Mortality: 2/10 McNamara et al. 1969 Rat 1 500 80 SL Mortality: 6/10 McNamara et al. 1969 Rat 1 500 90 SL Mortality: 8/10 McNamara et al. 1969 Rat 1 543 35 SL Mortality: 2/10 Owens and Punte 1963 Human 1 1.5 90 2 Nasal and ocular irritation, headache Owens and Punte, 1963 Human 1 6 18 2 Intolerable irritation; escape possible Owens and Punte 1963 Human 1 6.7 2 2 Intolerable irritation; escape possible Owens and Punte 1963 Human 1 85 1 2 Intolerable airway and ocular irritation Owens and Punte 1963 Human 1 94 1 2 Intolerable airway and ocular irritation Rengsdorf 1969 Human 1 1 1 2 Intense ocular irritation Rengsdorf 1969 Human 1 0.4 10 2 Intense ocular irritation Rengsdorf 1969 Human 1 0.6 10 2 Intense ocular irritation Rengsdorf 1969 Human 1 0.9 10 2 Intense ocular irritation Rengsdorf 1969 Human 1 1 1 2 Intense ocular irritation Striker et al. 1967 Monkey 1 900 3 2 Pulmonary congestion, emphysema (Continued) 379 

380 TABLE D-1 Continued Source Species Sex No. Exposures mg/m3 Minutes Category Comments Striker et al. 1967 Monkey 1 1,700 5 2 Pulmonary congestion, emphysema Striker et al. 1967 Monkey 1 2,500 32 SL Severe irritation, pulmonary edema, emphysema, mortality (5/8) Striker et al. 1967 Monkey 1 2,850 10 2 Pulmonary congestion, emphysema, ocular/respiratory irritation For category: 0 = no effect, 1 = discomfort, 2 = disabling, SL = some lethality, 3 = lethal