Toxicity of Military Smokes and Obscurants: Volume 3 (1999)

THE MAJOR dye components of the smoke formulation are benzanthrone (BZA) (54% of the dye components) and dibenzochrysenedione (DBC) (38%) (Lundy and Eaton 1994). DBC is also referred to as vat yellow 4. In the old M18 grenades, the colored dyes are mixed with a pyrotechnic mixture containing sulfur, potassium chlorate, and sodium bicarbonate. Optional amounts of a mixture of pure, refined kerosene and tricalcium phosphate are added for control of dusting and caking, respectively.

Chin et al. (1984) (as cited in Lundy and Eaton 1994) studied the combustion products of the old yellow smoke formulation and found that combustion produced no major chemical changes in the dyes. Additional information on the combustion products of the old yellow-smoke formulation is presented in Chapter 1.

The only toxicokinetics data available for the smoke formulation is the finding that yellow particles were present in the airways of exposed mice, rats, and guinea pigs many days (number of days not specified) after a 1-hr inhalation exposure to the combusted smoke formulation (Weeks and Yevich 1963). See Appendixes A and B for information on the toxicokinetics of the component dyes.

This section reviews the toxicity of the old yellow smoke formulation and its combustion products. The toxicity of the component dyes is evaluated in Appendixes A and B.

No epidemiological studies have been conducted on military workers or others exposed to the smoke formulation or its combustion products. A dermal lethal dose for 50% of the test animals (LD50) in rats exposed to the smoke formulation is more than 4.6 grams per kilogram (g/kg) of body weight (Lundy and Eaton 1994). An intraperitoneal LD₅₀ is 1.5 g/kg in the rat and 0.29 g/kg in the mouse.

In a study by Weeks and Yevich (1963), grenades containing an old yellow-smoke formulation were fired inside a static exposure chamber to achieve initial concentrations of particles that were high (13.4 grams per cubic meter (g/m³)), medium (4.8 g/m³), or low (0.9 g/m³). The dye composition of the smoke formulation was reported to be BZA and indanthrene (a blue powder) rather than BZA and vat yellow 4. Rats (male), mice (female), and guinea pigs (both sexes) were exposed in the static chamber for 1 hr at 13.4, 4.8, or 0.9 g/m³. Carbon monoxide (CO) concentrations reached 1% for the high-concentration group and might account for some of the toxic effects observed, including eye irritation, nasal discharge, gasping, and lethargy. Surviving animals were sacrificed 24 hr after the exposure or at later time points for up to 4 weeks after the exposure and then examined histologically.

All animals died during or soon after the exposure to the high concentration. After exposure to the medium concentration, 3 of 5 male guinea pigs and 8 of 10 male rats died. No animals died after exposure to the low concentration. All surviving animals had a retarded growth rate after the exposure; however, those exposed to the medium and low concentrations recovered and eventually gained weight at a normal rate.

Histopathological examination on animals sacrificed 24 hr after exposure indicated that the yellow-dye particles clogged the nasal passages of the animals and sometimes obliterated bronchi and bronchioles. Lungs exhibited necrosis, sloughing of the mucosa, and edema in the alveolar space. Necrosis of the epithelium of the tracheobronchial tree extended to the basement membrane and the submucosa. Gastric changes were observed with isolated areas of necrosis and hemorrhage. Those changes might be a result of ingestion of smoke components from grooming by the animals; it is unlikely that humans would be exposed by that route. At the later sacrifice times, there were signs of chronic inflammation in the nose, and the lung contained "foreign-body giant cells" that obliterated the alveoli.

Two female rhesus monkeys were exposed to the medium concentration (4.8 g/m³). One monkey collapsed and went into convulsions 22 min into the exposure and was removed from the exposure chamber. The second monkey was exposed for 30 min. Both monkeys developed a hacking cough resulting in a yellow mucoid exudate from the nostrils. The first monkey was sacrificed 5 days after exposure and the second, 27 days after exposure. Histopathological changes consisted of pulmonary mucosal necrosis and sloughing, serocellular plugs, necrosis extending to the adventitia in some areas, submucosal polymorphonuclear infiltration, and dilation of the mucoid glands.

One of the dyes used in the smoke formulation is BZA, a known photosensitizer and a compound that causes dermal toxicity in workers (Dacre et al. 1979). The other component, vat yellow 4, is relatively nontoxic for both noncancer (by oral and dermal route) and cancer end points. However, no inhalation toxicity studies have been completed on this chemical. For a complete review on the toxicity of the component dyes, see Appendixes A and B.

The major dye components of the smoke formulation are BZA (24% of the dye components), 1,4-di-p-toluidino-9,10-anthraquinone (PTA)

(62%), and vat yellow 4 (13%) (Lundy and Eaton 1994). PTA is also called solvent green 3. In the old M18 grenades, the colored dyes are mixed with a pyrotechnic mixture containing sulfur, potassium chlorate, and sodium bicarbonate. Optional amounts of a mixture of pure, refined kerosene and tricalcium phosphate are added for control of dusting and caking, respectively.

Chin et al. (1984) (as cited in Lundy and Eaton 1994) studied the combustion products of the old green smoke formulation and found that combustion produced no major chemical changes in the dyes. Additional information on the combustion products of the old green-smoke formulation is presented in Chapter 1.

A toxicokinetics study was conducted on a mixture of solvent green 3 and solvent yellow 33 dyes (Medinsky et al. 1986). The study demonstrated that solvent green 3 cleared slowly from rat lungs after inhalation. See Chapter 3 for details of this study.

The only information on the toxicokinetics of the combustion products of the old green smoke formulation is from the studies of Weeks and Yevich (1963). The composition of the smoke formulation was not exactly that of the Army's smoke formulation, but the major dye component of the smoke formulation was solvent green 3. Animals exposed to the combustion products had green particulate matter in their nasal passages and lungs for several days after a 1-hr exposure. See Appendixes A, B, and D for information on the toxicokinetics of the component dyes.

This section reviews the toxicity of the old green-smoke formulation and its combustion products. The toxicity of the component dyes is evaluated in Appendixes A, B, and D.

No epidemiological studies have been conducted on military workers

or others exposed to the smoke formulation or to its combustion products. An acute oral LD₅₀ of the smoke formulation in rats at a concentration of 3.1 g/kg of body weight was reported by Lundy and Eaton (1994). In a study by Weeks and Yevich (1963), grenades containing a green-smoke formulation were fired inside a static exposure chamber to achieve initial concentrations of particles that were high (12.1 g/m³), medium (4.8 g/m³) or low (0.6 g/m³). The dye composition consisted of solvent green 3 (75% of the smoke formulation) and auramine hydrochloride (25%) rather than solvent green 3, BZA, and vat yellow 4, as in the Army's smoke formulation. Rats (male), mice (female), and guinea pigs (both sexes) were exposed in the static chambers for 1 hr. CO concentrations reached 1% at the end of the high-concentration exposure and might account for some of the toxic effects observed in the animals, including eye irritation, nasal discharge, gasping, and lethargy. Surviving animals were sacrificed 24 hr after exposure or at later time points for up to 4 weeks after the exposure and then examined histopathologically.

Death occurred during or soon after the exposure to the high concentration in all rats, 7 of 10 mice, and 3 of 5 female guinea pigs. No animals died after exposure to the medium concentration, and 1 of 10 rats died after exposure to the low concentration. Surviving animals in all groups had at least a slightly retarded growth rate after the exposure. Animals in the high-concentration exposure group did not recover a normal rate of weight gain; however, those in the medium-and low-concentration groups recovered and gained weight at a normal rate. Histopathological analysis on animals sacrificed 24 hr after exposure indicated that the green-dye particles clogged the nasal passages and sometime obliterated bronchi and bronchioles. Lungs exhibited necrosis, sloughing of the mucosa, and edema in the alveolar space. Gastric changes were observed with isolated areas of necrosis and hemorrhage, and few areas of ulceration were seen in the duodenum. Those gastric changes might be a result of ingestion of smoke components from grooming by the animals; it is unlikely that humans would be exposed by that route. At later sacrifice times, signs of chronic inflammation were observed in the nose, and the lung contained "foreign-body giant cells" that obliterated the alveoli.

A repeated-exposure inhalation study of the toxicity of the combustion products of a smoke formulation containing a mixture of solvent yellow 33 (13%), disperse red 9 (16%), and solvent green 3 (19%) was performed in female rats, mice, and guinea pigs (Marrs et al. 1984). The animals were exposed at concentrations of 0.1, 0.3 and 1.0 g/m³ for 1 hr

per day, 5 days per week until they received up to 100 exposures. The animals were sacrificed 1 year after the exposure period with the exception of a group of mice killed 40 weeks after exposure. All animals had evidence of retention of dye in the lungs. In the guinea pigs, the high-dose exposure was stopped after 16 exposures because 14 of 50 animals had died from severe pulmonary congestion and alveolitis. Histologically, there were no significant effects in guinea pigs exposed at the two lower-doses except for alveolitis. There was a dose-related trend of alveolitis and chronic pneumonia in mice sacrificed 40 weeks after exposure. Mice in the high-dose group sacrificed 1 year after exposure had an increased number of macrophages and dye particles in the lungs compared with other exposure groups and controls. In rats, there was a significant dose-related trend in the occurrence of lesions related to inflammatory processes in the lung. It is important to note that the results of the Marrs et al. (1984) study could not be ascribed to solvent green 3 alone, and therefore, the relevance in quantitatively describing the toxicity of the old green-smoke formulation is questionable.

The primary dye component of the old green-smoke formulation is solvent green 3. Although relatively nontoxic when delivered orally or dermally to animals, solvent green 3 is insoluble in the lung and accumulates there when inhaled (Sun et al. 1987). The accumulation results in an inflammatory response in the lung. One of the other dyes used in the old green-smoke formulation is BZA, a known photosensitizer and a compound that causes dermal toxicity in workers (Dacre et al. 1979). Another component, vat yellow 4, is relatively nontoxic for both noncancer end points (by oral and dermal routes) and cancer end points. However, no inhalation toxicity studies have been completed on this compound. For a complete review on the toxicity of the dye components, see Appendixes A, B, and D.

The primary dye component of old red-smoke formulation is 1-methyl-

aminoanthraquinone (MAA) (40% of the total components) (Rubin and Buchanan 1983). MAA is also called disperse red 9. Other components are anthraquinone (2-3%) and trace materials (less than 1%). A dark, high-molecular-weight material that is chloroform insoluble and nonvolatile is also in the mixture (Rubin and Buchanan 1983). The authors stated that the material could not be fully characterized. In the old M18 grenades, the colored dyes are mixed with a pyrotechnic mixture containing sulfur, potassium chlorate, and sodium bicarbonate. Optional amounts of a mixture of pure, refined kerosene and tricalcium phosphate are added for control of dusting and caking, respectively.

Rubin et al. (1982) characterized the combustion products produced by old M18 grenades containing red-smoke formulation. The grenades were detonated inside sealed canvas tents. Samples of particulate matter and vapor generated by the smoke formulation were collected. Disperse red 9 accounted for 86% of the particulate matter in the combustion products compared with 98% in the red-smoke formulation. Approximately 10% of the disperse red 9 was converted to aminoanthraquinones (1-AA and 2-AA). The concentration of 1-AA and 2-AA increased more than 10 fold over their concentration in the old red-smoke formulation. Additional information on the combustion products of the old red-smoke formulation is presented in Chapter 1.

No studies have been conducted on the toxicokinetics of the old red-smoke formulation or its combustion products.

This section reviews the toxicity of the old red-smoke formulation and its combustion products. The toxicity of the component dye is evaluated in Appendix F.

No epidemiological studies have been conducted in military workers or others exposed to the smoke formulation or its combustion products. An acute intraperitoneal LD₅₀ in rats for the smoke formulation at a concentration of 1.5 g/kg of body weight was reported by Lundy and Eaton (1994).

The toxicity of the combustion products were tested in the monkey, dog, goat, swine, rabbit, rat, and guinea pig (Owens and Ward 1974). Grenades containing a red-smoke formulation were fired in test chambers exposing the animals to concentrations ranging from 1.5 to 18.0 g/m ³. Exposure duration ranged from 10 to 240 min. The animals were observed over a period of 30 days. The results were presented as a Bliss analysis of the combined mortality of total number of animals of all species exposed to the combustion products. For the combustion products, the lethal concentration to 50% of the test animals multiplied by exposure time (LCT₅₀) ranged from 0.4 to 0.6 g•min/m³. All animals showed signs of upper-respiratory irritation and salivation immediately after exposure. In the dog, swine, goat, and monkey, gagging and regurgitation of a thick red mucus were seen, and the urine was a dark red color for 24 hr immediately after exposure. Labored breathing was seen in all species for 7 days after exposure. The swine and the goat were the most resistant of all species tested. Most deaths occurred within 24 hr, and 97% of deaths occurred by day 14. Only general information could be gleaned from this study because the data were reported in such a manner that it is not possible to evaluate the results of the old red-smoke formulation on any single species.

Slaga et al. (1985) examined the initiation and promotion properties of the old red-smoke formulation in the Sencar mouse. The smoke formulation was studied for its complete carcinogenicity (administered as an initiator and a promoter to the same animals) or as an initiator only (administered as an initiator followed by 12-O-tetradecanoyl phorbol 13 acetate administration). There was no tumor response when the smoke formulation was tested as a complete carcinogen or as an initiator.

A component of the old red-smoke formulation, disperse red 9, was reported to cause skin irritation in humans (Dacre et al. 1979) (reviewed in Appendix F). The dose and length of exposure were not reported;

however, a one-time dermal application of disperse red 9 to the abraded skin of rabbits at a concentration of 2 g/kg of body weight resulted in negligible dermal irritation, as did ocular exposure at 0.05 g/kg of body weight (Martin et al. 1983). Oral administration of disperse red 9 to dogs at up to 8 g/kg of body weight resulted in no significant differences between exposed and control animals (Sendelbach 1989). One study testing disperse red 9 for mutagenicity reported positive results (Lundy and Eaton 1994). Two other studies reported negative results (Dacre et al. 1979; Sigman et al. 1985). There is no evidence for carcinogenicity of disperse red 9 (Griswold 1968; Sigman et al. 1985). The combustion product 2-AA was found to be carcinogenic in bioassays using rats and mice (NCI 1978).

The major dye components of old violet-smoke formulation are 1,4-diamino-2,3-dihydroanthraquinone (DDA), significant amounts of disperse red 9, as well as an insoluble residue and a number of organic materials in trace amounts (Rubin and Buchanan 1983). The mixture is formulated to contain 80% DDA and 20% disperse red 9. DDA is easily converted to 1,4-diaminoanthraquinone (DAA) in air. In the old M18 grenades, the colored dyes are mixed with a pyrotechnic mixture containing sulfur, potassium chlorate, and sodium bicarbonate. Optional amounts of a mixture of pure, refined kerosene and tricalcium phosphate are added for control of dusting and caking, respectively.

Rubin et al. (1982) characterized the combustion products produced by old M18 grenades containing violet-smoke formulation. The grenades were detonated inside canvas tents, and vapor and particle samples were collected. Upon combustion, DDA is known to convert to DAA. Disperse red 9 and AA were also detected in the combustion products along with small amounts of other organic materials, including cyanoanthraquinone, hydrocarbons, aliphatic amides, and phthalates. The inorganic materials

detected were potassium and sodium chlorides, sulfur, and trace metals. Additional information on the combustion products of old violet-smoke formulation is presented in Chapter 1.

No studies have been conducted on the toxicokinetics of old violet-smoke formulation or its combustion products.

This section reviews the toxicity of the old violet-smoke formulation and its combustion products. The toxicity of the component dyes is evaluated in Appendixes H and I.

The combustion products are more significant toxicologically than the noncombusted material (the smoke formulation), because the potential mutagenicity of DAA is greater than that of DDA (Rubin et al. 1982) (see Appendix I).

No epidemiological studies have been conducted on military workers or others exposed to either the smoke formulation or its combustion products. There are no studies on the acute inhalation toxicity of the uncombusted smoke formulation. Acute inhalation studies of the combustion products disseminated from M18 munitions were conducted in the monkey, dog, goat, swine, rabbit, rat, and guinea pig (Owens and Ward 1974). The animals were exposed to concentrations ranging from 1.3 to 7.8 g/m³ for 8 to 142 min. Exposure was followed by a 30-day observation period. The results were presented as a Bliss analysis of the combined mortality of the total number of animals of all species exposed to the combustion products. The combined LCT ₅₀ for the combustion products ranged from 0.21 to 0.20 g•min/m³. Immediately after exposure, all animals showed upper-respiratory irritation and salivation. Gagging was evident in the dog, swine, goat, and monkey. Prostration was noted in all species for 1 to 4 hr after exposure. Most deaths occurred within the first week after exposure.

Slaga et al. (1985) conducted studies of the ability of the old violet-smoke formulation to exhibit complete carcinogenicity as well as its

ability to be an initiator in the Sencar mouse. There was no tumor response to the formulation when tested as a complete carcinogen or as an initiator.

Lundy and Eaton (1994) tested the old violet-smoke formulation for mutagenicity and the formulation was positive in the Ames assay.

No data are available on the toxic effects of DDA or DAA in humans. In animals, no data are available on DDA effects; however, positive and ''marginally adequate'' effects have been cited in the Ames assay (Dacre et al. 1979; Lundy and Eaton 1994). With respect to DAA, moderate eye irritation in rabbits was found at a dose of 0.5 g for a 24-hr period. DAA has a reported LD₅₀ at 4.9 g/kg of body weight (RTECS 1981-82), and positive effects have been shown in the Ames assay (Lundy and Eaton 1994). For complete reviews on the toxicity of DDA and DAA, see Appendixes H and I, respectively.

There are no well-controlled inhalation toxicity studies on combustion products of old yellow-, green-, red-, and violet-smoke formulations that could provide a basis for the subcommittee to assess the potential health effects of exposure to these combustion products by military personnel or to recommend guidance levels.

The subcommittee is concerned about the toxicity of several of the component dyes. BZA, a major component of old yellow-and green-smoke formulations, is a dermal toxicant. Disperse red 9, a major component of old red-and violet-smoke formulations, also might be a dermal toxicant. Thus, even masking will not protect against the toxicity of the combustion products of these smoke formulations. Solvent green 3, a component of the old green-smoke formulation, was shown to accumulate in the lung with repeated exposures, resulting in an inflammatory response. Although DDA, a component of the old violet-smoke formulation, and DAA, a combustion product, produced positive results in the Ames assay, there is insufficient information to assess the toxicity of these compounds.

There are no previous recommended exposure limits for the combustion products of the old yellow-, green-, red-, and violet-smoke formulations. Army Health Hazard Assessment Reports (AEHA 1992, 1993a,b) on the M18 grenade and the 40 mm-cartridge components recommend masking when a smoke haze exceeds 4 hr or when passing through or operating in a smoke fog.

The data base for assessing the potential toxicity of the combustion products of the old yellow-, green-, red-, and violet-smoke formulations is inadequate to make recommendations for exposure guidance levels. The BZA component of the old yellow-and green-smoke formulations is not suitable for use as a component dye because of the dermal toxicity of the dye and its photosensitization properties. The subcommittee recommends that additional research be conducted on the toxicity of the combustion products of the old red-and violet-smoke formulations if the Army is to continue to use them. At a minimum, the Army should conduct acute inhalation studies in experimental animals to test the potential toxicity of the combustion products. Acute toxicity studies would be most appropriate for recommending the emergency exposure guidance level and the short-term public emergency guidance level. Such exposures might occur during training exercises in which military personnel might be exposed for several minutes, twice per day, two to four times per year. Some military personnel, particularly instructors involved in training exercises, might be repeatedly exposed to the smokes over several years, as might a community living near a military-training facility. For those exposure scenarios, the repeated exposure guidance level and the repeated public exposure guidance level are the most appropriate exposure guidance levels, and studies assessing the potential toxicity of the smokes following repeated exposure would provide the most appropriate data for setting those guidance levels. Thus, the Army should also consider conducting subchronic inhalation studies in experimental animals to test the toxicity of the smokes under conditions of repeated exposure. Both acute and repeated inhalation studies should be carried out using combusted smokes, and the particle

size and combustion products should be representative of the smokes used by the Army. Toxicity testing of other smoke formulations or of individual dyes might not provide results similar to those obtained with the combusted smokes.

Additionally, concerns for potential sensitization resulting from exposure to disperse red 9, a component of both the old red-and violet-smoke formulations, and BZA, a component of both the old yellow-and green-smoke formulations, suggest that studies to assess contact allergic dermatitis and respiratory-tract hypersensitivity be conducted in animal models appropriate for testing hypersensitivity. Those studies should also be conducted using the combusted smokes. To ensure that such studies are designed correctly, the Army should consult with an expert panel before conducting them.

At this time, the subcommittee recommends that Army policy be followed with regard to respiratory and dermal protection from the combustion products of the old yellow-, green-, red-, and violet-smoke formulations. It also recommends that the old colored smokes grenades be used for signaling and not for obscuring. The subcommittee recommends that the Army avoid exposing the general public to the combustion products of the old yellow-, green-, red-, and violet-smoke formulations.

AEHA (U.S. Army Environmental Hygiene Agency). 1992. Initial Health Hazard Assessment Report (RCS MED 388) on the M18 Colored (Red, Green and Yellow) Smoke Grenade Components. 69-37-XN46-92. U.S. Army Environmental Hygiene Agency, Aberdeen Proving Ground, Edgewood, MD.

AEHA (U.S. Army Environmental Hygiene Agency). 1993a. Health Hazard Assessment Report (RCS MED 388) on the 40 MM Colored Smoke Ground Marker. Report No. 69-37-XY57-93. U.S. Army Environmental Hygiene Agency, Aberdeen Proving Ground, Edgewood, MD.

AEHA (U.S. Army Environmental Hygiene Agency). 1993b. Health Hazard Assessment Report (RCS MED) on the M18 Colored (Red and Violet) Smoke Grenade Components. Report No. 69-37-XN46-93. U.S. Army Environmental Hygiene Agency, Aberdeen Proving Ground, Edgewood, MD.

Chin, A., J.D. Plummer, and L. Borer. 1984. Studies of the Effluents from Burning Navy Green, Yellow, Red and Orange Colored Smoke Devices, NSSW/CR/RDTR-232, Applied Sciences Department, Naval Weapons Support Center, Crane, IN.

Dacre, J.C., W.D. Burrows, C.W.R. Wade, A.F. Hegyeli, T.A. Miller, D.R. Cogley. 1979. Problem Definition Studies on Potential Environmental Pollutants. V. Physical, Chemical, Toxicological, and Biological Properties of Seven Chemicals Used in Pyrotechnic Compositions. Technical Report. No. 7704. AD A090631 U.S. Army Medical Bioengineering Research and Development Laboratory, Fort Detrick, Frederick, MD.

Griswold Jr., D.P., A.E. Casey, E.K. Weisburger, and J.H. Weisburger. 1968. The carcinogenicity of multiple intragastric doses of aromatic and heterocyclic nitro or amino derivatives in young female Sprague-Dawley rats. Cancer Res. 28:924-933.

Lundy, D., and J. Eaton. 1994. Occupational Health Hazards Posed by Inventory U.S. Army Smoke/Obscurant Munitions (Review Update). WRAIR/RT-94-0001. AD-A276-774. Prepared by U.S. Army Medical Research Detachment, Wright-Patterson Air Force Base, OH, for Walter Reed Army Institute of Research, Washington, DC.

Marrs, T.C., H.F. Colgrave, M. Gazzard, and R.F.R. Brown. 1984. Inhalation toxicity of a smoke containing solvent yellow 33, disperse red 9 and solvent green 3 in laboratory animals. Human Toxicol. 3(4):289-308.

Martin, B.W., G.W. Ivie, E.M. Bailey. 1983. The acute toxicity of 1-methyl-aminoanthraquinone in dogs and rabbits and its metabolism in sheep. Arch Environ. Contam. Toxicol. 12(4):499-507.

Medinsky, M.A., Y.S. Cheng, S.J. Kampcik, R.F. Henderson, and J.S. Dutcher. 1986. Disposition and metabolism of ¹⁴C-solvent yellow and solvent green aerosols after inhalation. Fundam. Appl. Toxicol. 7(1):170-178.

NCI (National Cancer Institute). 1978. Bioassay of 2-Aminoanthraquinone for Possible Carcinogenicity. Carcinogenesis Tech. Rep. Ser. 144. DHEW Publ. No. (NIH) 781399. National Institutes of Health, National Cancer Institute, Bethesda, MD.

Owens, E.J., and D.M. Ward 1974. A Review of the Toxicology Of Colored Chemical Smokes and Colored Smoke Dyes. AD/A 003-827. Edgewood Arsenal, Aberdeen Proving Ground, Edgewood, MD.

RTECS (Registry of Toxic Effects of Chemical Substances). 1981-82. 1,4-Di-aminoanthraquinone. National Institute of Occupational Health and Safety, Cincinnati, OH.

Rubin, I.B. and M.C. Buchanan. 1983. Chemical Characterization and Toxicologic Evaluation of Airborne Mixtures. Chemical Characterization of Army Colored Smokes: Inventory Smoke Mixes (Red, Violet, Yellow and Green). Final Report. ORNL/TM -8956. AD A134777. Oak Ridge National Laboratory, Oak Ridge, TN.

Rubin, I.B. M.C. Buchanan, and J.H. Moneyhun. 1982. Chemical Characterization and Toxicologic Evaluation of Airborne Mixtures. Chemical Characterization of Combusted Inventory Red and Violet Smoke Mixes. Final Report. ORNL/TM-8810. AD A131527. Oak Ridge National Laboratory, Oak Ridge, TN.

Sendelbach, L.E. 1989. A review of the toxicity and carcinogenicity of anthraquinone derivatives. Toxicology 57(30):227-240.

Sigman, C.C., P.A. Papa, M.K. Doeltz, L.R. Perry, A.M. Twhigg and C.T. Helmes. 1985. A study of anthraquinone dyes for the selection of candidates for carcinogen bioassay. J. Environ. Sci. Health A20(4):427-484.

Slaga, T.J., L.L. Triplett, and R.J.M. Fry. 1985. Chemical Characterization and Toxicologic Evaluation of Airborne Mixtures. Tumorigenicity Studies of Diesel Fuel-2, Red Smoke Dye and Violet Smoke Dyes in the SENCAR Mouse Skin Tumorigenesis Bioassay System . Final Report. ORNL/TM-9752. AD A159728. Oak Ridge National Laboratory, Oak Ridge, TN.

Sun, J.D., R.F. Henderson, T.C. Marshall, Y.-S. Cheng, J.S. Dutcher, J.A. Pickrell, J.L. Mauderly, F.F. Hahn, D.A. Banas, F.A. Seiler, and C.H. Hobbs. 1987. The inhalation toxicity of two commercial dyes: Solvent yellow 33 and solvent green 3. Fundam. Appl. Toxicol. 8:358-371.

Weeks, M.H., and P. Yevich 1963. The toxicity of combustion products of pyrotechnics. Ind. Hyg. J. 24:622-629.