EPA 0553
Health advisories for
lOCs and SOCs
RX000027511

-------
EPA  0553

  RX000027511
                                                                                      April, 1992
                                                ALDRIN

                                      Drinking Water Health Advisory
                                             Office of Water
                                   U.S. Environmental Protection Agency
         I.  INTRODUCTION
               The Health Advisory Program, sponsored by the Office of Water (OW), provides
         information on the health effects, analytical methodology, and treatment technology that
         would be useful in dealing with the contamination of drinking water. Health Advisories
         (HAs) describe nonregulatory concentrations of drinking water contaminants-at which adverse
         health effects would not be anticipated to occur over specific exposure durations.  Health
         Advisories contain a margin of safety to protect sensitive members of the population.

               Health Advisories serve  as informal technical guidance to assist Federal, State, and
         local officials responsible for protecting public health when emergency spills or contamination
         situations occur. They are not to be construed as legally enforceable Federal standards.  The
         HAs are subject to change as new information becomes available.

               Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
         years, or 10% of an individual's lifetime) and Lifetime exposures based on data describing
         noncarcinogenic end points of  toxicity. For those substances that are known or probable
         human carcinogens, according  to the Agency classification scheme (Group A or B), Lifetime
         HAs are not recommended. For substances with a carcinogenic potential, chemical
         concentration  values are correlated with carcinogenic  risk estimates by employing  a cancer
         potency (unit risk) value together with assumptions for lifelong exposure and the ingestion of
         water. The cancer unit risk is  usually derived from a linearized multistage model with 95%
         upper confidence limits providing a low-dose estimate of cancer risk. The cancer risk is
         characterized  as being an upper limit estimate, that is, the true risk to humans, while not
         identifiable, is not likely to exceed the upper limit estimate and in fact may be lower. While
         alternative risk modeling approaches may be presented, for example One-hit, Weibull, Logit,
         or Probit, the range of risks described by using any of these models has little biological
         significance unless data can  be used to support the selection of one model over another.  In
         the interest of consistency of approach and in providing an upper-bound on the potential
         carcinogenic risk, the Agency recommends using the linearized multistage model.

-------
Aldrin
                                     April, 1
II.  GENERAL INFORMATION'AND PROPERTIES

CAS No.  309-00-2

Structural Formula
                                                     H
Synonyms
      Compound 118; ENT 15,949; exo-dimethanonaphthalene; HHDN; hexachloro-
      hexahydroendo, l,2,3,4,10,10-hexachloro-l,4,4a,5,8,8a-hexahydro-l,4,5,8-
      dimethanonaphthalene; octalene (Sax, 1975; Sax and Lewis, 1987; Sittig, 1981).
Uses
  • .  Aldrin is an organochlorine insecticide that is highly effective against many soil-dwelling
      pests. Aldrin is principally used in agriculture at concentrations ranging from 0.5 to 5,
      kg/hectare.  Other uses include control of termite and ant infestation. Manufacture
      and use of aldrin have been discontinued in the United States since 1974 (Clayton and"
      Clayton, 1981; Windholz et al., 1983).

Properties  (Sittig,  1981; Windholz et al., 1983; Worthing and Walker, 1983; U.S. EPA, 1987).
      Chemical Formula
      ".v/olecular Weight
      Physical State
      Boiling Point
      Meltirg Point
      Density (20ฐQ
      Vapor Pressure; (20ฐC)
      Specific Gravity (20ฐC)
      Water Solubility (27ฐC)
      Log Octanol/Water
         Partition Coefficient
      Taste Threshold
      .Odor Threshold
      Conversion Factor
 C12H8C16
 364.93
 Technical aldrin—a tan to dark-brown
 crystalline solid with a mild chemical odor;
 nonflammable; stable between pH 4
: and 8.  Pure aldrin - a white powder.
 145ฐC at 2 mm Hg
 Pure:  104ฐC; technical: 49 to 60ฐC
 1.54 g/raL
 7.5 x 10'J mm Hg
 1.70
 0.027 m
 3.01
 0.017 mg/L water
 1 ppm = 14.96 mg/mj at 25ฐC and 760 mm Hg

-------
Aldrin                                                                       April, 1992
Occurrence

      •  Aldrin and dieldrin residues are more likely to be found in soil than in water.  The
         highest dieldrin concentration found in sediment collected from bays and rivers of
         North  Carolina was 18 ppb, and a study of groundwater in South Carolina showed
         aldrin  levels of 0.007 ppb (U.S. Army Medical Research  and Developmental
         Laboratory, 1975).

      •  The present occurrence of aldrin in the environment is not known, although it is
         expected to be much lower now than in the past because the manufacture and use
         of aldrin were discontinued in 1974.

Environmental Fate

      •  In the  soil, aldrin is easily oxidized to dieldrin (Elgar,  1975). The investigator
         reported the results of single applications of aldrin to various soils from 12
         locations. The proportion of dieldrin in the total soil residue ranged from 60% 1
         year after application to 90% after 3 years.

      •  Aldrin is firmly bound to soil, is extremely resistant to leaching and does not
         migrate from the point of application (Shell Chemical  Company,  1986).

      •  Aldrin and dieldrin have low solubility in water and are strongly sorbed to soil,
         particularly to organic matter.  Eye (1968) concluded from a study on leachability
         that dieldrin residues are not significantly  transported-through soil into subsurface
         waters. These results were confirmed in studies by ThompJaAi et al. (1970).

      •  Aldrin (technical grade, purity unspecified), applied to subirri-gated soil at 4 pounds
         active  ingredient per acre (Ib a.i./A), was found to be immobile in a Hager silty clay
         loam and in a Lakeland sandy loam soil (Harris, 1969).
                                                          ,    - ~\ *
      •  In a soil-column (52 mm diameter by 60 cm length) leaching study, Foschi et al.
         (1970) found that aldrin (purity not specified) did not  leach into the 20- to 40-cm
         section (or below) after 15 days of soaking ."vith.'approximately 35.4 inches (89.8 cm)
         of water in a greenhouse at 26ฐC.  Residues of aldrin were detected by gas
         chromatography at 3.68 ppm in the top 5-cm  section and at 3.07 ppm in the next
         15-crn segment.                           :
                                                  .     "   i')' '
      •  In a laboratory study using labeled material, 2 to; 26% of the applied aldrin was
         found to volatilize from various substrates (Lichtenstein'and^Schulz, 1970).
         Approximately 19 to 26% of the aldrin volatilized from tapwater, a soil-water
         mixture, a pH 7 buffer solution, or glass beads, and 2  to 3% of the pesticide

-------
Aldrin                                                                       April,
         volatilized from silt loam soil or silt loam soil plus a detergent (0.1% linear
         alkylbenzenesulfonate). In another test, fruit flies died when exposed to aldrin
         vapors above an aldrin-water mixture containing 25 /j.g aldrin. Approximately 6 /xg
         aldrin volatilized in a 24-hour period.

      •  Aldrin residues in samples of rotational crops ranged from 0 to 2.45 ppm; the
         highest levels were found in potatoes.  The relative concentration of dieldrin in the
         recovered residues increased with time; only dieldrin was detected 5 years post-
         treatment.  Other aldrin degradates were not assayed. Aldrin, at 25 Ib a.i./A,
         dissipated in plots of silt loam soil planted to crops each year, declining from 9.77
         to 0.05 ppm in 5 years after soil incorporation to a depth of 4 to 5 inches.
         Approximately 11% of the total applied dose was present after 5 years;  the
         percentage of dieldrin found in recovered residues was 97%. Aldrin residues in a
         subplot planted to alfalfa as a cover crop dissipated  more slowly, declining from
         4.04 ppm in 1959 to 0.96 ppra in 1963.  During the test period, the-average annual
         rainfall was 30.16 inches (76.61 cm), the average annual temperature was 45.3ฐF
         (7.4ฐC) and the average temperature for the period  from May to October was
         62.4ฐF (16.9ฐC) (Lichtenstein and Schulz, 1965).
III. PHARMACOKINETICS

Absorption

      •  Quantitative data on the gastrointestinal (GI) absorption of aldrin by laboratory
         animals are not available.  However, the results of metabolism studies indicate that
         aldrin is slowly absorbed via the portal venous system following oral administration
         (U.S. EPA, 1987). In a time-course distribution/metabolism study by Farb et al.
         (1973),  aldrin was detected in the stomach and small intestine (levels not specified)
         of neonatal rats up to 6 days after administration of single oral doses of 10 mg/kg.
         In the same group of animals, the concentration of aldrin in the liver increased to
         about 13% of the administered dose by the third day.

      •  Ludwig et al. (1964) estimated that approximately 10% of the radio label
         administered to two male  rats (strain not reported) as wC-labeled aldrin dissolved
         in corn oil (43 /ig/rat/day  for 90 days) was absorbed by the GI tract.

      •  Results of inhalation studies on human volunteers suggest that up to 50% of
         inhaled aldrin vapor (exposure levels not specified) may be retained (Beyermann
         and Eckrich, 1973).

   _....•  Feldmann and Maibach (1974) reported that approximately 8% of the dermally
         applied aldrin (in acetone vehicle) was absorbed in humans.

-------
Aldrin                                                                      April, 199Z
Distribution
      •  Aldrin is rapidly metabolized to dieldrin.  Bioaccumulation of this metabolite in
         adipose tissue has been reported (U.S. EPA, 1987).  As discussed below, oral
         administration of dieldrin at concentrations as high as 160 ppm in rats (Deichmann
         et al., 1975) and 70 ppm in dogs (Deichmann et al., 1969), were found to
         accumulate in fat. No significant differences were noted between the amount of
         insecticide found in lipids and the amount found in the total carcass or abdominal
         fat.

      •  Farb et al. (1973) detected aldrin in the kidneys of neonatal rats for up to 72 hours
         after administration  of single oral doses of 10 mg/kg.  In the liver, aldrin
         concentrations increased for the first 6 hours following dosing to a maximum of
         13% of the administered dose. After 72 hours, aldrin concentration declined to less
         than 0.1%.  The only metabolite detected in the liver was dieldrin, which was
         identified as early as 2 hours post-treatment. A maximum of 31% of the
         administered dose was detected as dieldrin in liver tissue 24 hours following dosing.

      •  Ludwig et al. (1964) reported that the ratio of dieldrin to aldrin was  15:1 in the
         carcass and 18:1 in the abdominal fat of two male rats given daily oral doses of 4.3
         Mg uC-aldrin for 3 months. Approximately 3.60,  1.77 and 1.83% of the total
         administered dose of "C was found in the carcass, fat and other tissues,
         respectively.

      •  Increasing amounts of dieldrin were noted in the abdominal fat and total
         extractable lipids of mice from the F,, F2 and F} generations fed diets containing 5
         or 10 pprn aldrin (purity 95%) (doses equivalent to 0.75 and 1.5 mg/kg, respectively,
         based on Lehman, 1959).  Dieldrin residues in the parents of each generation were
         measured after about 260  days on the test diets; dieldrin concentrations  in total
         lipids were between 115 and 121 ppm in males and between 149 and 159 in females
         (Deichmann et al., 1975).

      •  In a study in which six male beagles were given 0.6 mg/kg of aldrin in corn oil daily
         for 10 months, dieldrin concentrations in the fat and liver increased  during months
         1 through 10, reaching levels of 70 and 20 ppm,  respectively (Deichmann et al.,
         1969). Dieldrin concentrations gradually decreased, and 12 months after aldrin
         administration was halted, dieldrin  concentrations had dropped to 25;and 6 ppm in
         the fat and liver, respectively.
Metabolism
      •  Following absorption, aldrin is readily oxidized to dieldrin in mammals (U.S. EPA.
         1987).

-------
Aldrin                                                                      April,
      •  Formation of the epoxide (dieldrin) is a major step in the biotransformation of
         aldrin. This transformation is probably mediated by  mixed-function mono-
         oxygenases (aldrin epoxidases), which are found in many organisms  including plants
         (Mehendale et al., 1972), insects (Krieger and Wilkinson, 1969), fish (Bums, 1976)
         and various mammals (Levi and Hodgson, 1985; Davies and Keysell, 1983; Wolff et
         al., 1980; Brooks and Harrison, 1966) and humans (McMannus et al., 1984).

      •  Soto and Deichmann (1967) reported that approximately 30% of the aldrin
         administered intravenously to dogs was converted to  dieldrin within  24 hours of
         dosing.

      •  Klein et al. (1968) identified one of the polar metabolites of aldrin in rat urine as
         l,l,2,3a,7a-pentachloro-5,6-epoxydecahydro-2,4,7-metheno-3H-cyclopenta[a]-
         pentalen-3-one (pentachloroketone).

Excretion

      •  In a  study in which male rats were given  daily oral doses of 4.3 /ig l4C-aldrin for 3
         months, the elimination of radioactivity, expressed as percent of dose administered
         weekly per rat, increased from 31% on the second day of dosing to about 100%
         during weeks 10 to 12 (Ludwig et al., 1964). Most of the radioactivity was excreted
         in the feces. During weeks 10 through 12, fecal excretion accounted  for 93 to 94%
         of the weekly dose.  Urinary excretion was approximately 9%.  Biliary excretion was
         not measured. The authors reported that after 8 weeks, the amount of radioactivity
         being excreted was equivalent to the amount administered, suggesting that an
         equilibrium between the intake and storage of aldrin had been established.  After
         dosing was discontinued, excretion of radioactivity  decreased rapidly. Analysis of
         urine and feces indicated that predominantly hydrophilic metabolites were excreted.
IV. HEALTH EFFECTS

Humans

  Short-term Exposure

      •  Acute intoxication following aldrin exposure in humans is characterized by a brief
         period of excitation or drowsiness, followed by convulsions, muscle twitching and
         coma.  Hypothermia generally accompanies death. The majority of individuals
         intoxicated with aldrin, however, usually regain consciousness and recover (Hayes,
         1982; Jager, 1970).

-------
Aldrin                                                              '         April, 1992.
      •  Jager (1970) reported the acute oral lethal dose of aldrin in an adult male to be 5.0
         g (approximately 70 mg/kg,  assuming a body weight of 70 kg).

      •  The ingestion of approximately 120 mg (8:2 mg/kg)  of aldrin by a 3-year-old female
         resulted in collapse and convulsions within 5 minutes and death within 12 hours
         (Hayes, 1982).

      •  A 23-year-old male experienced convulsions about 20 minutes after ingesting 25.6
         mg/kg aldrin (Spiotta, 1951). After treatment with pentobarbital, his convulsions
         ceased.  However, he exhibited restlessness, hypothermia,  tachycardia and
         hypertension for up to 5 days and electroencephalogram (EEG) abnormalities for
         up to 6 months after ingestion of aldrin.

  Long-term Exposure
                                                                      *
      •  One male, employed 21 years at a chemical plant and reassigned to the handling of
         aldrin concentrate (period and levels of exposure  not specified), experienced
         involuntary jerking (rapid flexor movement) of his hands and forearms, vomiting
         and chronic irritability and insomnia (Hodge et al.,  1967).  His EEG showed alpha-
         wave  irregularities, with discharges of slow and sharp waves.  After exposure to
         aldrin was discontinued, his condition rapidly improved.

      •  Dieldrin (mean 13 ng/g whole milk) was found in the breast milk of women whose
         homes were treated annually (or more frequently) with organochlorine pesticides
         (Stacey and Tatum, 1985).  A correlation between dieldrin  levels in the milk and
         aldrin treatment of homes was  observed.  Dieldrin levels in breast milk rose until
         the seventh or eighth month after treatment of homes was  discontinued.  No data
         were  provided on the health effects of children exposed to dieldrin-contaminated
         breast milk.

Animals

  Short-term Exposure

      •  The oral median lethal dose (LDj,,) values for aldrin in laboratory animals are as
         follows: mice, 44 mg/kg (purity not reported) (Borgmann et al., 1952); rats, 39 to 60
         mg/kg (purity not reported) (Gaines, 1969);. guinea  pigs, 33 mg/kg (purity not
         reported) (Borgmann et al., 1952); female rabbits, 50 to 80 mg/kg (purity, 95%)
         (Treon and Cleveland, 1955); and dogs, 65 to 95 rag/kg (purity not reported)
         (Borgmann et al., 1952).

      •  Acute toxicity in animals is  characterized by increased irritability, salivation,
         hyperexcitability, tremors followed by clonic/tonic convulsions, anorexia and loss of

-------
Aldrin                                                                     April,  19|
         body weight, depression, prostration and death (Borgmann et al., 1952; Hodge et
         al., 1967).

      •  Decreased body weight gain and increased mortality were observed in male and
         female Osbome-Mendel rats fed 320 ppm (16 mg/kg/day, based on a food
         consumption factor of 0.05 from Lehman, 1959) of aldrin (technical grade; 95%
         pure) in the diet for 42 days and observed for an additional 14 days (NCI, 1978).
         Groups of five rats of each sex were fed diets containing 0, 40, 80, 160 or 320 ppm
         aldrin (0, 2, 4, 8 or 16 mg/kg/day, respectively).  The No-Observed-Adverse-Effect
         Level (NOAEL) for this study was 160 ppm (8 mg/kg/day).  This study was a range-
         finding study for a long-term carcinogenicity study; therefore, a complete toxicology
         profile was not obtained (e.g., biochemical and hematology measurements).

      •  In groups of B6C3Fl mice (five/sex/group) fed aldrin (technical grade; 95% pure)  at
         concentrations of 0, 2.5, 5, 10, 20, 40 or 80 ppm (0, 0375, 0.75, 1.5? 3, 6 or 12
         mg/kg/day,  respectively, based on a food consumption factor of 0.15 from Lehman,
         1959) in the diet for 42 days, 100% mortality was observed .in the 40- and 80-ppm
         (6- and 12-mg/kg/day) groups. One male and one female died in the 20-ppm (3-
         mg/kg/day) group; 10 and 20 ppra (1.5 and 3 rag/kg/day) were therefore considered
         the NOAEL and Lowest-Observed-Adverse-Effect Level (LOAEL)  for this study
         (NCI, 1978).  This study was a range-finding study for a long-term carcinogenicity
         study; therefore, a complete toxicology profile was not obtained.

      •  In a  subchronic toxicity study (see the discussion in "Dermal/Ocular Effects" and
         "Long-term Exposure" below), Treon and Cleveland (1955) observed 100%
         mortality within 2 weeks in groups of male and female Carworth rats (total number
         and number/sex not reported) fed 300 ppm aldrin (purity 95%) (15  mg/kg/day,
         based on Lehman, 1959). No mortality was noted at lower doses.  Administration
         of a  diet containing 25 ppm aldrin (purity 95%) (0.625 mg/kg/day) to two male and
         three female beagle dogs induced fatalities after periods of feeding  ranging from 9
         to 15 days.

  Dermal/Ocular Effects

      •  Treon and Cleveland (1955) reported dermal LD,,, values of 600 to  1,250 mg/kg for
         powdered aldrin (purity 95%) in rabbits. The investigators also demonstrated that
         the application of powdered aldrin is less toxic to rabbits than aldrin in a vehicle.
         The  ranges of minimum lethal dosages obtained for aldrin in the powdered form,  in
         vegetable oil or in Ultraseneฎ (solvent) were 3.5 to 123, 10 to 26, and < 4.8
         mg/kg/day, respectively.  The rabbits were exposed for 2 hours/day,  5 days/week for
         10 weeks.

-------
Aldrin                                                                      April, 199Z
  Long-term Exposure

      •  Treon and Cleveland (1955) fed male and female Carworth rats (number/dose level
         not reported) 2.5, 5, 25, 75 or 300 ppm aldrin (purity 95%) (0.125, 0.25, 1.25, 3.75
         or 15 mg/kg/day, respectively,  based on Lehman, 1959) in the diet for 6 months;
         100% mortality was observed  at 15 mg/kg/day within 2 weeks.  No differences in
         mortality rate were noted among other test groups and the controls.  The authors
         did not report data on other effects. A NOAEL for the subchronic effects of aldrin
         cannot be determined from these data.

      •  Fitzhugh et al. (1964) fed groups of 12 male  and 12 female Osborne-Mendel rats
         aldrin (purity 99%) in the  diet at levels of 0.5, 2, 10, 50, 100 or 150 ppm (0.025, 0.1,
         0.5, 2.5, 5.0 and 7.5 mg/kg/day, respectively, based on Lehman, 1959) for 2 years.  A
         dose-related increase in mortality was  observed at a dietary level of 50 ppm or
         greater.  In addition, significant (p <. 0.05) dose-related increases in  relative liver
         weights were observed.  Histopathological changes observed in the livers of aldrin-
         treated rats were chiefly the characteristic "chlorinated insecticide" lesions that
         occur only in rodents. These  lesions consist of enlarged centrilobular hepatic cells,
         with increased cytoplasmic oxyphilia, and peripheral migration of basophilic
         granules. The incidence and severity of this nonneoplastic histologic change
         increased with increasing dietary level. In  rats ingesting amounts of aldrin at 50
         ppm or higher, distended and hernorrhagic urinary bladders, enlarged livers and
         increased incidences  of nephritis were observed. The apparent LOAEL for this
         study was 0.5 ppm (0.025 mg/kg/day).  A NOAEL was not established.

      •  Deichmann et al. (1970) fed groups of 50 male and 50 female  Osbome-Mendel rats
         20, 30 or 50 ppm (1,  1.5 or 25 mg/kg/day, respectively, based on Lehman, 1959)
         aldrin (technical grade, 95% pure) for 31 months.  Groups of 100 rats of each sex
         served as controls. Survival and body  weight gains were comparable  between the
         treated and the control groups. Treated animals exhibited tremors and clonic
         convulsions. Liver-to-body weight ratios were increased in males fed 30 or 50 ppm.
         A moderate increase (not  dose-related) in  the incidence of hepatic centrilobular
         cloudy swelling and necrosis was observed  in all aldrin-treated male and female rats,
         but not in controls.  The LOAEL of 20 ppm (1 mg/kg/day) was established for this
         study.  A NOAEL was not determined.

      •  Aldrin (technical grade, 95%  pure) was administered in the diet at 4 or 8 ppm (0.6
         or 12 mg/kg/day, respectively, based on Lehman, 1959) to groups of 50 male mice
         and at 3 or 6 ppm (0.45 or 0.90 mg/kg/day, respectively, based on Lehman, 1959) to
         groups of 50 female mice  for  80 weeks.  In a trend  test, a significant  (p = 0.037),
         dose-dependent increase in mortality was observed  in females; a similar effect was
         not observed in males.  A NOAEL was not established because of toxicity at 3 ppm
         (0.45 mg/kg/day), the lowest dose tested  (NCL 1978).

-------
Aldrin                                                                      April, 1
      •  Kitselman and Borgmann (1952) fed groups of seven mongrel dogs of both sexes
         (number/sex not specified) 0.2, 0.6 or 2 mg/kg/day of aldrin in medicated meatballs
         for up to 228 days. The test material was 99% pure.  Dogs administered 2
         mg/kg/day exhibited marked body weight loss, and all dogs died between days 60
         and 90.  No treatment-related effects were observed in dogs receiving 0.2 mg/kg/day
         for 190 days or in those administered 0.6 mg/kg/day for 228 days. Based on body
         weight loss, 0.6 mg/kg/day and 2 mg/kg/day were considered to be the NOAEL and
         LOAEL, respectively, for this study.

      •  In a long-term feeding study by Treon and Cleveland (1955), beagles (two/sex/dose)
         fed diets containing 1 or 3 ppm (0.043 to 0.091 or 0.12 to 0.25 mg/kg/day,
         respectively, as reported by the authors) aldrin (purity 95%) for 15.6 months gained
         weight at rates similar to control dogs.  However, at 3 ppm, significant (p < 0.05)
         increases in absolute and relative liver weights were noted. Histopathologic
         changes, such as fatty degeneration of the liver and vacuolation of renal tubular
         cells, were observed in both sexes at the 3-ppm level. At the 1-ppm level, females
         exhibited vacuolation of the distal renal tubules.  The LOAEL for this study was 1
         ppm (0.043 to 0.091 mg/kg/day).

      •  Fitzhugh et al. (1964) administered 0.2, 0.5, 1, 2  or 5 mg/kg/day aldrin (purity 99%)
         to 12  mongrel dogs (sexes combined), 6 days/week for periods of up to 25 months. I
         Each  group consisted of one dog/sex except for the 0.5-mg/kg/day group, which had
         one male and three female  dogs. All dogs receiving 1, 2 or 5 mg/kg/day died within
         49 weeks; the first death occurred on day 22 in a female administered 5  mg/kg/day.
         Prior to death, the animals  exhibited  body weight loss, dehydration and convulsions.
         Slight to moderate fatty degeneration was noted in hepatic and  renal tubular cells,
         and reduced numbers of mature erythroid cells were found in the bone marrow.  In
         animals receiving 0.5 mg/kg/day, clinical signs of toxicity were limited to  convulsions
         in one male dog during the 24th month.  Dogs in the 0.2-mg/kg/day group exhibited
         no adverse effects. The NOAEL in this study appears to be 0.2 mg/kg/day based on
         the absence of clinical signs of toxicity, body weight loss and histopathological
         changes. However, the adequacy of this study for establishing a NOAEL is
         questionable owing to the small number of dogs  used.

  Reproductive Effects

      •  A dietary level of 12.5 ppm (0.625 mg/kg/day, based on Lehman, 1959) fed over
         three generations was reported to cause a reduction in the pregnancy rate of
         Carworth rats (Treon and Cleveland, 1955). Groups of rats (number not specified)
         were  fed aldrin (purity 95%) at concentrations of 0, 2.5, 12.5 or 25.0 ppm (0.125,
         0.625 or 125 mg/kg/day, based on Lehman, 1959). No reduction in the  number of
         live pups/litter was evident  in dams fed 0.125, 0.625 or 1.25 mg/kg/day in the diet.
                                           10

-------
Aldrin                                                                       April, 1992.
         However, the viability of the offspring during lactation was decreased at levels of
         0.625 and 1.25 mg/kg/day.

      •  No apparent effects on the fertility or pregnancy rates were evident in mongrel dogs
         (two females/group) receiving 0.2, 0.6 or 2.0 mg/kg/day of aldrin (purity 99%) in
         medicated meatballs for 1 year (Kitselman, 1953). However, the majority of
         apparently healthy pups that were delivered at all levels died within 3 days
         postpartum.

      •  In a reproduction study reported by Deichmann et al.  (1971), groups of beagles
         were administered 0.15 (four females) or 0.3 (four males, three females) mg/kg/day
         of aldrin (purity 95%) by capsule 5 days/week for 14 months. Two of the four
         females administered 0.15 mg/kg/day failed to achieve  estrus during the following 8-
         month period in which aldrin feeding had been halted.  Similar problems were not
         evident in dogs given 0.3 mg/kg/day.  During lactation, the viability, of pups from
         dams receiving  0.15 or 0.3 mg/kg/day was decreased; 84, 75 and 44% of pups from
         dams ingesting  0, 0.15 and 0.3 mg/kg/day, respectively, survived until weaning. The
         reduced pup survival may have been due  to a prenatal effect or to toxicity
         associated with dieldrin in the mothers' milk. Mammary development and milk
         production also appeared to be depressed.  Some males reportedly exhibited a
         depressed sexual drive.

  Developmental Effects

      •  Administration of 50 mg/kg of aldrin (purity > 99%) via oral intubation to pregnant
         Syrian Golden  hamsters on day 7, 8 or 9 of gestation resulted in significant (p <
         0.05) increases  in fetal death, resorption,  anomalies (open eye, webbed foot, cleft
         palate and/or lip and fused ribs) and growth retardation (Ottolenghi et al., 1974).
         In pregnant CD-I mice given 25 mg/kg of aldrin on day 9 of gestation, significant (p
         < 0.01) increases in total anomalies,  as well as incidences of open eye, webbed foot
         and cleft palate, were also observed.  No  increases in fetal death or decreases in
         fetal body weight were noted in mice, however.  No information on maternal
         toxicity was presented.

  Mutagem'city

      •  In bacterial reverse mutation assays conducted by several investigators, aldrin was
         not mutagenic  to Salmonella typhimurium (Simmon and  Kauhanen, 1978; Cotruvo
         et al., 1977; Simmon et al., 1977).

      •  Simmon and Kauhanen (1978) reported that aldrin, at concentrations of 10 to 5,000
         jig/plate, did not cause gene conversion in Saccharomyces cerevisiae D3 with or
         without Aroclor-induced rat liver microsomes.

                                           11

-------
Aldrin                                                                      April, 19,
      •  Georgian (1975) reported that aldrin induced chromosome aberrations in human
         lymphocytes in vitro and in rat and mouse bone marrow cells in vivo.  The evidence
         for a clastogenic response was, however, inconclusive because increases in the
         frequency of chromosomal aberrations in the in vivo assays occurred only at
         cytotoxic levels. Additionally, chromosome and chromatid gaps, which are unreliable
         indicators of damage to genetic material, were included as aberrant figures. The
         extent of true chromosomal damage cannot, therefore,  be ascertained.

      •  Exposure of primary rat hepatocytes to aldrin at concentrations ranging from 0.5 to
         1,000 nmol/mL for 5 to 20 hours did not induce unscheduled DNA synthesis (Probst
         et al., 1981).

  Carcinogenic! tv

      •  NCI (1978) reported increased combined incidences of follicular ceil adenoma and
         carcinoma of the thyroid in both male  and female Osbome-Mendel rats (50
         animals/sex) fed aldrin (technical grade, 95% pure) at concentrations of 30 or 60
         ppm (1.5 and 3 mg/kg/day, respectively, based on Lehman, 1959) for 74 to 80 weeks
         compared to controls. The incidences were 4/48,14/38 and 8/38 for males and 3/52,
         10/39 and 7/46 for females from the pooled control, low-dose and high-dose groups^
         respectively. Differences were significant (p = 0.001) in the low-dose group but nci
         in the high-dose group.  Significant (p  = 0.001) increases in the incidence of
         cortical adenoma of the adrenal gland  were also observed in the low-dose females,
         but this was not considered to be compound-related.  Aldrin produced no
         significant effect on the mortality of rats of either sex.  The authors concluded that
         the observed tumors were not associated with treatment.  However, the U.S. EPA
         (1987) concluded that the occurrence of thyroid tumors should be considered
         equivocal evidence of carcinogenicity.

      •  In a carcinogenicity bioassay,  aldrin (technical grade, 95% pure) was  administered
         in the diet at 4 or 8 ppm (0.6  or 1.2 mg/kg/day, based on Lehman, 1959) to 50 male
         B6C3Ft mice and at 3 or 6 ppm (0.45 or 0.90 mg/kg/day, based on Lehman, 1959)
         to 50 female B6C3F! mice,  for 80 weeks.  The animals were observed for an
         additional 10 to 13 weeks.  A  significant (p ฃ 0.001) dose-related increase in the
         incidence of hepatocellular carcinomas was observed in male but not  female mice
         fed diets containing 4 or 8 ppm (0.6 or 12 mg/kg/day, respectively) when compared
         to matched or pooled controls. Incidences were 3/20 17/92, 16/49 and 25/45 for the
         matched control,  pooled control, low-dose male and high-dose  male groups,
         respectively (NCI, 1978).

      •  Davis and Fitzhugh (1962) and Davis (1965) fed 10 ppm (approximately 1.5
         mg/kg/day, based on Lehman, 1959) of aldrin (purity not reported) in the diet to a
         group of 215 C3HeB/Fe mice ("approximately  equally divided by sex") for 2 years:

                                          12

-------
 Aldrin
                                                     April, 1992.
          the control group consisted of 217 mice. It was reported that the incidence of
          hepatic cell adenomas was significantly (p < 0.001) higher in treated mice (35 of
          215) than in nine controls (9 of 217).  Upon revaluation of the liver lesions, most
          of the liver tumors were found to be hepatic carcinomas (Epstein, 1975).
 V. QUANTIFICATION OF TOXICOLOGICAL EFFECTS

      Health Advisories (HAs) are based upon the identification of adverse health effects
 associated with the most sensitive and meaningful noncarcinogenic end point of toricity. The
 induction of this effect is related to a particular exposure dose over a specified period of
 time, most often determined from the results of an experimental animal study. Traditional
 risk characterization methodology for threshold toxicants is applied in HA development. The
 general formula is as follows:

                (NOAEL or LOAEL)  x (BW)          ..
                                             = — mg/L
(UF)
                                Uday)
 where:

NOAEL or LOAEL

              BW

               UF


          _L/day
=   No- or Lowest-Observed-Adverse-Effect Level (in rag/kg bw/day).

=   assumed body weight of a child (10 kg) or an adult (70 kg).

=   uncertainty factor, (10, 100,1,000 or 10,000), in accordance with
    EPA or NAS/OW guidelines.

=   assumed daily water consumption of a child (1 L/day) or an adult
    (2 L/day).
 One-day Health Advisory

     No suitable information was found in the available literature for determining the One-
 day HA for aldrin.  The modified Drinking Water Equivalent Level (DWEL) for a 10-kg
 child of 0.0003 mg/L (03 fig/L, (calculated below) is recommended for use as a conservative
 estimate of the one-day HA for aldrin.
                                          13

-------
Aldrin                                                                     April, 199!
    For a 10-kg child, the adjusted DWEL is calculated as follows:

       DWฃL m  (0.00003 mg/kg/day) (10  kg)  = Q ^          ^ to QJ
                         (1 L/day)

where:

 0.00003 mg/kg/day     =   RfD (see Lifetime Health Advisory Section).

             10 kg     =   assumed body  weight of a child.

           1  L/day     =   assumed daily  water consumption of a child.

Ten-day Health Advisory
                                                                       ซ

    No suitable information was found in the available literature for determining the Ten-
day HA for aldrin. The modified DWEL for a 10-kg child of 0.0003 mg/L (0.3 ng/L) is
recommended for use as a conservative estimate of the  ten-day HA for aldrin.

Longer-term  Health Advisory

    No suitable information was found in the available literature for determining the
Longer-term  HA for aldrin.  Several subchronic studies were available. However, either
inadequate study design (e.g., a small number of animals exposed or an inadequate length of
exposure) or  inadequate reporting resulted  in the disqualification of these studies as the basis
for derivation of the Longer-term HA. The modified DWEL for a 10-kg child of 0.0003 mg/L
(0.3 Mg/L) is recommended for use as a conservative estimate of the longer-term HA for
aldrin.

Lifetime Health Advisory

    The Lifetime HA represents that portion of an individual's total exposure that is
attributed to  drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure.  The Lifetime HA is derived in a three-step process.  Step 1
determines the Reference Dose (RfD), formerly called the Acceptable Daily Intake (ADI).
The RfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily
exposure to the human population (including sensitive subgroups) that is likely to be without
appreciable risk of deleterious health effects during a lifetime, and is derived from the
NOAEL (or  LOAEL),  identified from a chronic (or subchronic) study, divided by an
uncertainty factor(s). From the RfD, a Drinking Water Equivalent Level (DWEL) can be
determined (Step 2). A DWEL is a medium-specific (i.e., drinking water) lifetime exposure
level, assuming 100% exposure from that medium, at which adverse, noncarcinogenic health
effects would not be expected to occur. The DWEL is derived from the multiplication of the

                                          14

-------
Aldrin                                                                       April, 1992-
RfD by the assumed body weight of an adult and divided by the assumed daily water
consumption of an adult. The Lifetime HA in drinking water alone is determined in Step 3
by factoring in other sources of exposure, the relative source contribution (RSC).  The RSC
from drinking water is based on actual exposure data or, if data are not available, a value of
20% is assumed.  If the contaminant is classified as a known, probable, or possible human
carcinogen, according to the Agency's classification scheme of carcinogenic potential (U.S.
EPA, 1986), then caution must be exercised in making a decision on how to deal .with
possible lifetime exposure to this substance. For human (A) or probable  (B) human
carcinogens, a Lifetime HA is not recommended.  For possible (C)  human carcinogens, an
additional 10-fold safety factor is used in the calculation of the Lifetime HA. The risk
manager must balance this assessment of carcinogenic potential and the quality of the data
against the likelihood of occurrence and significance of health effects related to
noncarcinogenic end points of toxicity. To assist the risk manager in this process, drinking
water concentrations associated with estimated excess lifetime cancer risks over the range of 1
in 10,000 to 1 in 1,000,000 for the 70-kg adult drinking 2 L of water/day are pfovided in the
Evaluation of Carcinogenic Potential section.

     The study of Fitzhugh  et al. (1964) in which groups of Osborne-Mendel rats were fed
diets containing aldrin at concentrations of 0.5, 2, 10, 50, 100 or 150 ppm (approximately
equivalent to 0.025, 0.1, 0.5, 2.5, 5 or 7.5 mg/kg/day, based on Lehman, 1959) for 2 years has
been selected to serve as the basis for the Reference Dose because  it was the most
appropriate chronic study found that established a LOAEL for aldrin in this species. In this
study, a LOAEL of 0.5 ppm (0.025 mg/kg/day) was identified, based on an increased
incidence of liver lesions characteristic of exposure to organochlorine pesticides and increased
relative liver weights in animals from all aldrin-treated groups. The liver lesions consisted of
enlarged centrilobular hepatic cells, with increased cytoplasmic oxyphilia and peripheral
migration of basophilic granules. A NOAEL was not established Furthermore, the results of
several other  long-term feeding studies in rats using aldrin at higher concentrations (2.5 to 60
ppm; equivalent to 0.13 and 3 mg/kg/day,  based on Lehman, 1959) support the  results of the
Fitzhugh et al. (1964) study (NCI, 1978; Deichmann et al.,  1970; Treon and Cleveland, 1955).

     Using the Fitzhugh et  al. (1964) study, the Lifetime HA is derived as follows:

Step 1:  Determination of the RfD

     RfD m  (0.025 mg/kg/day)  = Q QQQ^ mg/kg/day (rounded to 0.00003 mg/kg/day)
                  (1,000)

where:

   0.025 mg/kg/day     =   LOAEL, based on an increased incidence in liver lesions and
                           increased relative liver weights in rats fed aldrin for 2 years.
                                           15

-------
Aldrin                                                                      April, 1991
             1,000     =   uncertainty factor, chosen in accordance with EPA or NAS/OW
                          guidelines in which a LOAEL from an animal study is employed.

Step 2:  Determination of the DWEL

            DWEL n (0.00003 mg/kg/day) (70 kg) = Qm mg/L    ^
                               (2  L/day)

where:

 0.00003 mg/kg/day     =   RfD.

             70 kg     =   assumed body weight of an adult.

           2 L/day     =   assumed water consumption of a 70-kg adult.

Step 3:  Determination of the Lifetime HA

     Aldrin has been classified in Group B2:  probable human carcinogen (U.S. EPA, 1986a);
thus, a Lifetime HA is not recommended. The estimated excess cancer risk associated with
lifetime exposure to drinking water containing aldrin at 0.88 /xg/L is 4.25 x 10"*. This estimate
represents the upper 95% confidence limit from extrapolations prepared by EPA's
Carcinogen Assessment Group (U.S. EPA, 1987), using the linearized multistage model. The
actual risk is unlikely to exceed this value.

Evaluation of Carcinogenic Potential

     •  Three adequately conducted long-term carcinogenicity bioassays of aldrin have been
        conducted with B6C3Ft, C3HeB/Fe and C3H mice. Based on these studies, there is
        sufficient evidence of carcinogenicity for aldrin.  Dietary administration of aldrin
        induced statistically significant (p < 0.001) increases in hepatocellular carcinomas in
        male B6C3F, mice (NCI, 1978), hepatomas in male and female C3HeB/Fe mice
        (Davis and Fitzhugh, 1962) and hepatomas  in both sexes of C3H mice (Davis, 1965,
        as cited in Epstein, 1975). Reevaluation of the hepatomas observed in the latter two
        studies showed the hepatomas actually to be hepatocellular carcinomas (Epstein,
        1975).

     •  Dietary administration of aldrin increased the combined incidences of follicular cell
        adenomas and carcinomas of the  thyroid in both male and female Osborne-Mendel
        rats; however, the increase was not dose-related'and was significant (p = 0.001) only
        at the low dose. This increase was not  considered to be treatment-related. It was
        concluded that aldrin was not carcinogenic to rats (NCI, 1978).
                                          16

-------
AJdrin                                                              '        April, 1992-
    •  Based on the available data, IARC (1987) concluded that there was limited evidence
        for the carcinogenicity of aldrin in animals and inadequate evidence in humans.
        lARC's conclusion was based on the occurrence of malignant liver neoplasms in
        mice, since one study in rats could not clearly associate the occurrence of thyroid
        tumors with  aldrin treatment, three additional studies in rats gave negative results
        and another  rat study was judged to be inadequate.  Consequently, IARC classified
        aldrin as a Group 3 chemical, a possible human carcinogen.

    •  Applying the criteria described in EPA's guidelines for assessment of carcinogenic
        risk (U.S. EPA, I986a), aldrin and its metabolite dieldrin may be classified in Group
        B2:  probable human carcinogen. This category is for agents for which there  is
        inadequate evidence of carcinogenicity in human studies and sufficient evidence  of.
        carcinogenicity in animal studies.

    •  Several carcinogenicity studies have provided evidence that aldrin is carcinogenic in
        mice.  Three data sets from these studies are adequate for quantitative risk
        estimation.  Utilizing the linearized multistage model, the U.S. EPA (1987)
        performed potency estimates for each of these data sets.  The geometric mean of the
        potency estimates, (qt*) = 17 (mg/kg/day)'1,  was estimated as the cancer potency of
        aldrin for the general population.

    •  Using this cancer potency estimate and assuming that a 70-kg human adult consumes
        2 liters of water a day over a 70-year lifespan, the linearized multistage model can be
        used to estimate that concentrations of 0.2, 0.02 and 0.002 fig liter of aldrin may
        result in an excess cancer risk of 10"*,  10"5 and 10"*, respectively.

    •  The linearized multistage model is only one  method of estimating carcinogenic risk.
        From the data contained in the U.S. EPA (1987) report, it was determined that one
        of the three  data sets were suitable for determining slope estimates for the probit,
        logit, Weibull and gamma-multihit models. The cancer risk estimate (at  the upper
        95% confidence limit) that can cause one excess cancer per 1,000,000 (10"6) is
        associated with exposure to aldrin levels in drinking water of 0.00206 /ug/L for the
        multistage model, 0.00356 /ig/L for the probit model, 0.00376 /Ag/L for the logit
        model, 0.00356 /xg/L for the Weibull model and 0.00310 ngfL for the multihit model.
        Each model is based on different assumptions.  Based on the current understanding
        of the biological mechanisms of carcinogenesis, the relative accuracy of these models
        cannot be predicted
VI.  OTHER CRITERIA. GUIDANCE AND STANDARDS

     •  The U.S. EPA (1980) has established a National Ambient Water Quality Criterion
        for aldrin of 0.074 ng/L for human health based on a predicted cancer risk of

                                          17

-------
Aldrin                                                                   April, 19J
        1:1,000,000.  This estimate is based on the ingestion of contaminated drinking water
        and aquatic organisms.  A criterion of 0.079 ng/L is recommended for ingestion of
        contaminated aquatic organisms only.

    •   The Occupational Safety and Health Administration (OSHA) has established an 8-
        hour Time-Weighted Average (TWA) atmospheric Permissible Exposure Limit
        (PEL) of 0.25 mg/mj with a  recommendation to reduce skin absorption of aldrin by
        using protective measures (OSHA, 1989).

    •   Aldrin is regulated  as a hazardous substance, with a reportable quantity of 1 Ib (0.454
        kg) under section 102 of the Comprehensive Environmental Response,
        Compensation and  Liability  Act (CERCLA) for release from vessels and facilities
        (U.S. EPA, 1986b).

    •   The American Conference of Governmental and Industrial Hygienists (ACGIH)
        established a TWA Threshold Limit Value (TLV) of 0.25 mg/mj for exposure  to
        aldrin (ACGIH, 1986).

    •   The National Institute for Occupational Safety and Health (NIOSH) recommended a
        PEL of 0.25  mg/mj for aldrin (NIOSH, 1988).
VII.  ANALYTICAL METHODS

    •  Aldrin has been included in numerous EPA methods for analyzing chlorinated
       hydrocarbon pesticides.  In earlier procedures aldrin was used as an internal
       standard, since it had been rarely detected in the environment, degrading fairly
       quickly to dieldrin.

    •  Aldrin can be analyzed by EPA Methods 505 (U.S. EPA, 1988a) and 508 (U.S. EPA,
       1988b). Both methods use electron capture gas chromatography for analysis, but
       Method 505 is a micro-extraction procedure and Method 508 is one 1-L liquid-liquid
       extraction. The detection limit for aldrin is 0.075 mg/L.
VIII.  TREATMENT TECHNOLOGIES

    •  Granular-activated carbon (GAC) adsorption and reverse osmosis (RO) can reduce
        the levels of aldrin in drinking water supplies.

    •  Lafomara (1978) described case histories of the use of a trailer-mounted spills-
        treatment process to remove a number of toxic organic compounds from water,
        including aldrin. The treatment trailer, which had a capacity of approximately 0.3

                                        18

-------
Aldrin                                                                      April, 1994.
        million gal/day (1 million L/day), consisted of three mixed-media filters, each 3.5 feet
        (107 cm) in diameter and containing a 2-foot-deep (61-cm) bed of powdered
        anthracite on top of a 1.5-foot (46-cm) layer of sand plus three GAG columns, each 7
        feet (213 cm) in diameter and containing 5,940 Ib (2,690 kg) of GAG.  The
        contaminated water (containing 8.5 mg/L of aldrin) was passed through one filter and
        one GAG column with a contact time of 17 minutes.  After 100,000 gal (380,000 L)
        were treated, the effluent contained 0.19 mg/L of aldrin.

    •   Van Dyke et al. (1986) evaluated the ability of a home-use water filter to remove a
        number of organic chemicals, including aldrin.  The filtering system consisted of a
        nonwoven prefilter, a pressed carbon block and  a porous polyethylene-fritted core.
        The water was supplied at a constant pressure of 50 pounds per square inch (3.5
        kg/cm2) gauge. Each run  consisted of passing a volume of water equal to 150% of
        the  filter rated life of 500 gal (1,900 L) through  the filter,  and analyzing for various
        contaminants.  Aldrin was present in the influent at a concentration of 68 mg/L, and
        the  filter removed aldrin below its detection limit of 0.1 mg/L.

    •   Abron and Osburn (1973) investigated polyaraide hollow fiber RO membranes for
        use in the removal of aldrin from aqueous solutions. This  system was operated at 600
        psi, a flow rate of 0.48 gph and a flux rate of 0.077 to 0.082 gpd/ft2. Complete
        removal of aldrin with the RO membranes was attained at influent concentrations of
        both 4 mg/L and 28.1 mg/L. Aldrin is classified as a membrane-interacting solute.
        The aldrin removal may not have been achieved by the reverse osmotic pressure, but
        by membrane-solute interaction.

    •   Lykins  et al. (1986) tested the effect of disinfection, sand filtration and GAG
        adsorption on several organic compounds at a pilot plant in Jefferson Parish, LA.
        Four disinfectants were used:  chlorine, monochloramine,  chlorine dioxide and ozone.
        Aldrin was present in the non-disinfected influent at an average concentration of 0.53
        ng/L. The disinfection system was designed for a contact time  of 30 minutes. Sand
        filtration did not reduce the concentration of chlorinated hydrocarbon insecticides
        (CHI).  About 90 to 93% of the CHI was removed by GAC over the 1-year
        operational period.  No analytical data are presented on aldrin concentration in
        treated effluent.

    •  Data were not found for the removal of aldrin from drinking water by aeration.
        However, because of its low vapor pressure, aldrin is not likely to be amenable  to
        removal by aeration.
                                          19

-------
Aldrin                                                                     April, l!
IX. REFERENCES

Abron, L.A. and J.O. Osbum. 1973. A transport mechanism in hollow nylon fiber reverse
    osmosis membranes for the removal of DDT and aldrin from water.  Water Research
    7:461-477.

ACGIH.  1986.  American Conference of Governmental and Industrial Hygienists. TLV.
    Threshold-Limit  Values and biological exposure indices for 1986-1987.  Cincinnati, OH:
    ACGIH, pp. 5, 9, 16.

Beyermann,  K. and W. Eckrich.  1973.  Gas-chromatographische bestimmung von insecticid-
    spuren in Luft. Z. Anal. Chem. 265:4-7 (in German; translation.).

Borgmann, A.R., C.H. Kitselman, P.A. Dahm, I.E. Pankaskie and F.R. Dutra.  Kansas State
    College. 1952. Toxicological studies of dieldrin on small laboratory animals.
    Unpublished report.  July.

Brooks, G.T. and A. Harrison.  1966. Metabolism of aldrin and dihydroaldrin by houseflies
    (M. domestica  L.) in vivo and by housefly and pig liver microsomes. Life Sci. 5:2315-232Q

Burns, K.A.  1976.  Microsomal mixed-function oxidases in an estaurine fish Fundulus
    heteroclitus, and their induction as a result of environmental contamination. Comp.
    Biochem. Physiol. 538:443-446.

Clayton, C.D. and F.E. Clayton. 1981.  Patty's industrial hygiene and toxicology, Vol. 2. New
    York, NY:  John Wiley and Sons, pp. 3056-3057.

Cotruvo, J.A., V.F. Simmon and RJ. Spanggord. 1977.  Investigation of rautagenic effects of
    products of ozonation reactions in water.  Ann.  N.Y. Acad. Sci. 298:124-140.

Davies, M.W. and G.R. Keysell. 1983.  Sex differences in the metabolism of aldrin by rat
    liver preparation. J. Pharm. Pharmacol. 35(Suppl.):82.

Davis, HJ.  1965.  Pathology report on mice for aldrin, dieldrin, heptachlor or heptachlor
    epoxide for two years. Memorandum to A.J. Lehman. International FDA. July 19.

Davis, KJ. and  O.G.  Fitzhugh. 1962.  Tumorigenic potential of aldrin and dieldrin for mice.
    Toxicol. Appl. Pharmacol. 4:187-189.

Deichmann, W.B.,  M. Keplinger, I. Dressier and F. Sala.  1969. Retention of dieldrin and
    DDT in tissues of dogs fed aldrin and DDT individually and as a mixture.  Toxicol. Appl.
    Pharmacol. 14:205-213.
                                         20

-------
Aldrin                                                                      April, 1992*
Deichraann, W.B., W.E. MacDonald, A.G. Beasley and D. Cubit.  1971.  Subnormal
    reproduction in beagle dogs induced by DDT and aldrin. Ind. Med. Surg. 40:10-20.

Deichmann, W.B., W.E. MacDonald, E. Blum, M. Bevilacqua, J. Radomsld, M.  Keplinger
    and M. Balkus.  1970. Tumorigenicity of aldrin, dieldrin and endrin in the albino rat.
    Ind. Med. Surg.  39:426-434.

Deichmann, W.B., W.E. MacDonald and D.A. Cubit.  1975.  Dieldrin and DDT in the tissues
    of mice fed aldrin and DDT for seven generations.  Arch. Toxicol. 34:173-182.

Elgar, K.E.  1975. The dissipation and accumulation of aldrin and dieldrin residues in soil.
    Environ. Qual. Safety 3(Suppl.):250-257.

Epstein, S.S.  1975.  The carcinogenicity of dieldrin.  Part I.  Sci. Total Environ. 4:1-52.
                                                                      ซ
Eye, J.D.  1968.  Aqueous transport of dieldrin residues in soil. J. Water Pollut. Control.
    Fed. 40:R316-R332.
Farb, .R.M^ T. Sanderson, E.G. Moore and A.W. Hayes.  1973.  Interaction:  the effect of
     selected mycotorins on the tissue distribution and retention of aldrin and dieldrin in the
     neonatal rat. In: Deichmann, W.B., ed. Pesticides in the environment.  New York, NY:
     International Medical Book Corporation, pp. 179-187.

Feldmann, R J. and H.I. Maibach.  1974. Percutaneous penetration of some pesticides and
     herbicides in man. Toxicol. Appl. Pharmacol. 28:126-132.

Fitzhugh,  O.G., A.A. Nelson and M.L. Quaife.  1964.  Chronic oral toxicity of aldrin and
     dieldrin in rats and dogs.  Food Cosmet. Toxicol. 2:551-562.

Foschi, S., A. Cesari and I. Ponti. 1970.  Investigation into degradation and vertical
     movement of agricultural chemicals in soil.  Notiz. Mai Piante 82:37. (In  Italian;
     translation.)

Gaines, T.B.  1969. Acute toxicity of pesticides.  Toxicol. Appl. Pharmacol. 14:515-534.

Georgian, L. 1975. The comparative cytogeru'c effects of aldrin and phosphamidon. Mutat.
     Res.  31:103-108.

Harris, C. 1969.  Movement of pesticides in soil. J. Agric. Food Chem. 17:80-82.

Hayes, WJ. 1982.  Pesticides studied in man.  Baltimore, MD:  The Williams and Wilkins
     Company, pp. 234-247.
                                          21

-------
Aldrin                                                                       April,
Hodge, H.C., A.M. Boyce, W.B. Deichmann and H.F. Kraybill.  1967.  Toxicology and no-
    . effect levels of aldrin and dieldrin. Toxicol. Appl. Pharmacol. 10:613-675.

IARC. 1987.  International Agency for Research on Cancer. Evaluation of the carcinogenic
    risk of chemicals to humans.  Overall evaluations of carcinogenicity. Suppl. 7:88-89.

Jager, K.W.  1970. Aldrin, dieldrin, endrin, and telodrin:  An epidemiological and
    lexicological study of long-term occupational Exposure. New York, NY:  Elsevier
    Publishing Company, pp. 121-131.

Kitselman, C.H. 1953.  Long-term studies on dogs fed aldrin and dieldrin in sublethal
    dosages with reference to the histopathological findings and reproduction.  J. Am. Vet.
    Med. Assoc. 123:28-30.

Kitselman, C.H. and A.R. Borgrnann.  1952. A comparative study of the reaction of dogs as
    a susceptible species to sublethal doses of aldrin and dieldrin. J. Am. Vet.  Med. Assoc.
    121:383-385.

Klein, A.K., J.D. Link and N.F. Ives. 1968. Isolation and purification of metabolites found in
    the urine  of male rats fed aldrin and dieldrin. J. Assoc. Anal. Chem. 51:895-898.

Krieger, R.I. and C.F. Wilkinson.  1969. Microsomal mixed-function oxidases in insects. I.
    Localization and properties of southern armyworm (Prodenia eridania). Biochem.
    Pharmacol. 18:1403-1415.

Lafomara, J.P. 1978. Cleanup  after spills of toxic substances. J. Water Pol. Con. Fed.
    50(4):617-627.

Lehman, A. 1959. Appraisal of the safety of chemicals in foods, drugs and cosmetics.
    Association of Food and Drug Officials of the United States.

Levi, P.E. and E. Hodgson.  1985.  Oxidation of pesticides by purified cytochrome P-450
    isozymes from mouse liver. Toxicol. Lett.  24:221-228.

Lichtenstein, E. and K. Schulz.  1970.  Volatilization of insecticides from various substrates.
    J. Agric. Food Chem. 18:814-818.

Lichtenstein, E. and K. Schulz.  1965.  Residues of aldrin and heptachlor in soils and their
    translocation into various crops. J. Agric.  Food Chem. 13:57-63.

Ludwig, G., J. Weis and F. Korte.  1964. Excretion and distribution of aldrin-MC and its
    metabolites after oral administration for a long period of time. Life Sci.  3:123-130.
                                          22

-------
Aldrin
Lykins, B.W., W.E. Koffskey and R.G. Miller.  1986.  Chemical products and toricologic
  .  effects of disinfection. J. AWWA. 78(ll):66-75.

McMannus, M.E., A.R. Boobis, R.A. Michin, D.M. Schwartz, S. Murray, D.S. Davies and S.S.
    Thorgeirsson.  1984.  Relationship between oxidative metabolism of 2-
    acetylaminofluorine, dibrisoquine and aldrin in human liver microsomes. Cancer Res
    44:5692-5697.

Mehendale, H.M., R.F. Skrentny and H.W. Dorough.  1972. Oxidative metabolism of aldrin
    by subcellular root fractions of several plant species. J. Agri. Food Chem. 20:398-402.

NCI.  1978.  National Cancer Institute. Bioassay of aldrin and dieldrin for possible
    carcinogenicity. NCI-CG-TR-21.  Publication No.  (NIH) 78-821. Washington, DC:
    Department of Health, Education and Welfare.

NIOSH.  1988.  National Institute for Occupational Safety and Health.  Pocket guide to
    chemical hazards.  Publication no. 85-114.  Washington, DC: U.S. Department of Health,
    Education and Welfare, pp. 46-47.

OSHA.  1989. Occupational Safety and Health Administration. OSHA safety and health
    standards for general industry.  29 CFR 1910.1000. OSHA3112. Washington, DC: U.S.
    Government Printing Office.

Ottolenghi, A.D., J.K.  Haseman and F. Suggs.  1974.  Teratogenic effects of aldrin, dieldrin
    and endrin in hamsters and mice.  Teratology 9:11-16.

Probst, G.S., R.E. McMahon, L.W. Hill, D.Z. Thompson, J.K Epp and S.B. Neal.  1981.
    Chemically-induced unscheduled DNA synthesis in primary rat hepatocyte cultures: a
    comparison with bacterial rautagenicity using 218 chemicals. Environ. Mutagen. 3:11-32.

Sax, N.L 1975.   Dangerous properties of industrial materials, 4th ed. New York, NY: Van
    Nostrand Reinhold Company, p. 1101.

Sax, N.L and RJ. Lewis Sr. 1987. Hawley's condensed chemical dictionary.  New York, NY:
    Van Nostrand Reinhold Company, pp. 1050-1051.

Shell Chemical Company.  1986.  Shell Chemical Company.  Aldrin termiticide, health and
    environmental aspects. The Hague, The Netherlands:  Shell International Petroleum
    Maatschappi B.V.

Simmon, V.F. and K. Kauhanen. 1978.  Stanford research report. In vitro microbiological
    mutagenicity assays of aldrin. Stanford, CA:  Stanford Research International, p. 15.
                                          23

-------
Aldrin                                                                      April, 1
Simmon, V.F., K. Kauhanen and R.G. Tardiff.  1977.  Mutagenic activity of chemicals
     identified in drinking water. Develop. Toxicol. Environ. Sci. 2:249-258.

Sittig, M  1985.  Handbook of toxic and hazardous chemicals and carcinogens, 3rd ed.  Park
     Ridge, NJ: Noyes Publications.

Sittig, M.  1981.  Handbook of toxic and hazardous chemicals and carcinogens, 2nd ed.  Park
     Ridge, NJ: Noyes Publications, pp. 792-793.

Soto, A.R. and W.B. Deichmann.  1967. Major metabolism and acute toxicity of aldrin,
     dieldrin and endrin.  Environ. Res. 1:307-322.

Spiotta, EJ.  1951.  Aldrin poisoning in man. Arch. Ind. Hyg. Occup. Med. 4:560-566.

Stacey, C.I. and T. Tatum.  1985. House treatment with organochlorine pesticides and  their
     level in milk.  Path, Western Australia. Bull. Environ. Contam. Toxicol.  35:202-208.

Thompson, A.R., C.A. Edwards, MJ. Edwards and K.I. Beyon.  1970. Movement of dieldrin
     through soils.  II.  In sloping troughs  and soil columns. Pestic. Sci. 1:174-178.

Treon, J.F. and P.P. Cleveland.  1955.  Toxicity of certain chlorinated hydrogen insecticides
     for laboratory animals with special  reference to aldrin and dieldrin. Agric. Food Chem.
     3:402-408.

U.S. Army Medical Research and Developmental Laboratory. 1975.  Problem definition
     studies on potential environmental  pollutants.  Technical Report 7509.

U.S. EPA.  1980.  U.S. Environmental Protection Agency.  Ambient water quality criteria for
     aldrin/dieldrin. Report PB81-11730/OWRS. Washington, DC:  U.S. EPA, Criteria and
     Standards Division.

U.S. EPA.  1986a.  U.S. Environmental Protection Agency.  Guidelines for carcinogen risk
     assessment. Fed. Reg. 51(185):  33992-34003.

U.S. EPA. 1986b. U.S. Environmental Protection Agency.  Effluent guidelines and
     standards. 40 CFR, Subchapter N. Section 401.15. Part 401. General provisions,  pp. 5,
     8.

U.S. EPA. 1987. U.S. Environmental Protection Agency. Carcinogenicity assessment of
     aldrin and dieldrin. EPA/600/6-87-006. Washington, DC:  U.S. EPA, Office of Health
     and Environmental Assessment, Carcinogenesis Assessment Group.
                                          24

-------
Aldrin                                                                    April,
U.S. EPA.  1988a.  U.S. Environmental Protection Agency.  Method 505-Analysis of
    organochlorine pesticides in water by micro-extraction and gas chromatography, revision
    2.1.  Cincinnati, OH: Environmental Monitoring and Support Laboratory.  December.

U.S. EPA.  1988b.  U.S. Environmental Protection Agency.  Method 508-Determination of
    chlorinated pesticides in water by gas chromatography with an electron-capture detector,
    revision 2.1. Cincinnati, OH: Environmental Monitoring and Support Laboratory.
    December.

Van Dyke, K., R. Kuennen, J. Stiles, J. Wezeman and J. O'Neal. 1986. Test stand design
    and testing for a pressed carbon block water filter.  Am. Lab. 18(9):118-132.

Windholz M., S. Budavari, R.F. Bluemetti and E.X. Otterbein, ed  1983.  The Merck index--
    An encyclopedia of chemicals and drugs, 10th ed. Rahway, NJ: Merck and Company,
    Inc.

Wolff, T., H. Green, M-T. Huang, G.T. Miwa and Y.H.  Lu.  1980.  Aldrin epoxidation
    catalyzed by purified rat-liver cytochromes P-450 and P-448. Eur. J. Biochera. 11:545-
    551.

Worthing, C.R. and S.B. Walker, eds.  1983.  The pesticide manual.  In:  A world
    compendium,  7th ed.  Lavenhara, Suffolk, Great Britain:  The Lavenham Press Limited,
     pp. 6, 191.
                                         25

-------
EPA  0553

KX000027511
                                                                               April 1992
                                       AMMONIA
                              Drinking Water Health Advisory
                                      Office of Water
                           U.S. Environmental Protection Agency
 I.      INTRODUCTION
       The Health Advisory (HA) Program, sponsored by the Office of Water (OW), provides
 information on the health effects, analytical methodology and treatment technology that would.
 be useful in dealing with the contamination of drinking water.  Health Advisories describe
 nonregulatory concentrations of drinking water contaminants at which adverse health effects
 would not be anticipated to occur over specific exposure durations.  Health Advisories contain a
 margin of safety to protect sensitive members of the population.

        Health Advisories serve as informal technical guidance to assist Federal, State and local
 officials responsible for protecting public health when emergency spills or contamination
 situations occur.  They are not to be construed as legally enforceable Federal standards.  The
 HAs are subject to change as new information becomes available.

        Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
 years, or 10% of an individual's lifetime) and Lifetime exposures based on data describing
 noncarcinogenic end points of toxicity.  For those substances that are known or probable human
 carcinogens, according to the Agency classification scheme (Group A or B), Lifetime Health
 Advisories are not recommended.  The chemical concentration values for Group A or B
 carcinogens are correlated with carcinogenic risk estimates by  employing a cancer potency (unit
 risk) value together with assumptions for lifelong exposure and the ingestion of water.  The
 cancer unit risk is usually derived from a linearized multistage model with 95% upper
 confidence limits providing a low-dose estimate of cancer risk. This provides a low-dose
 estimate of cancer risk to humans that is considered unlikely to pose a carcinogenic risk in
 excess of the stated values. Excess cancer  risk estimates may also be calculated using the
 one-hit, Weibull, logit or probit models. There is no current understanding of the biological
 mechanisms involved in cancer to suggest that any one of these models is able to predict risk
 more accurately than another. Because each model is based on differing  assumptions, the
 estimates that are derived can differ by several orders of magnitude.
 II.     GENERAL INFORMATION AND PROPERTIES

        Ammonia is a ubiquitous naturally occurring inorganic chemical. In aqueous solution, it
 exists in a number of forms that include the following: NH,, NH4OH, NHj.HjO and NH4*.  For
 purposes of this document, ammonia is used as a general term encompassing NHj and NH4*.

        CAS  No. 7664-41-7

-------
Ammonia                                                                      April 1992,
       Structural Formula
                                       H—N—H
                                           I
                                           H

                                       Ammonia

       Synonyms

       •     None.

       Uses

       •     Ammonia is used in fertilizers, refrigeration systems and manufacturing processes
             (Windholz et al.t 1983).

       •     Used in conjunction with chlorine to form chloramine which is used as drinking
             water disinfectant.

       Properties  (Campbell et al., 1958; Verschueren, 1977; Windholz et al., 1983)

       Chemical Formula                        NH3
       Molecular  Weight                        17.03
       Physical  State (25 ฐC)                     Liquid
       Boiling Point                             33.4ฐC
       Melting  Point                            -77.7 ฐC
       Density  (-33 ฐC, 1 atm)                    0.6818
       Vapor Pressure (20 ฐC)                    8.5 atm
       Specific  Gravity                           —
       Water Solubility (20ฐC)                   531 g/L
       Log Octanol/Water Partition               —
        Coefficient (log K^,)
       Taste Threshold                          34 mg/L
       Odor threshold (air)                     0.04 g/ra3
       Conversion Factor                        —

       Occurrence

       •     Survey of total ammonia (NH, + NH*4) concentrations in surface waters indicate
             an  average of 0.18 mg/L in most areas, and 0.5 mg/L in waters near large
             metropolitan  areas (U.S. EPA, 1979).  Levels in ground water are usually low,

-------
Ammonia                                                                      April 1992
              since ammonia is generally immobile in soil (Feth, 1966).  Ammonia is effectively
              eliminated when drinking water is chlorinated.

              Ammonia is a negligible natural constituent of food, but ammonium compounds
              are added in small amounts (0.001 to 2%) to various foods as stabilizers,
              leavening agents, flavorings or for other purposes (FASEB, 1974).
       Environmental Fate
              Ammonia is introduced into the environment through sewage effluents, fertilizer
              application, agricultural runoff, drainage from feedlots and through industrial
              discharge.  Once present  in an environment, ammonia participates in the
              constant fluxing of nitrogen levels and states of the nitrogen cycle.  Processes
              related to this cycle are nitrogen fixation, nitrification,  denitrification and
              ammonification. Ammonia is subject to assimilation by chemotrophic and
              phototropic organisms, or may be biologically oxidized (nitrification).
              Nonbiological means of moving ammonia within an environment include
              diffusion, dilution (in waters), volatilization and sorption to particles.
III.    PHARMACOKINETICS

       Absorption
             Ammonia is produced in humans in the stomach, duodenum, ileum, colon and
             feces at an estimated 4,200 mg/kg/day, with the colon and fecal content
             contributing about 73%. Of the total amount produced, 4,150 mg/kg/day is
             absorbed and 50 rag/kg/day is excreted (Summerskill and Wolpert, 1970).

             Conn (1972) administered  9  rag NH4Cl/kg as uncoated tablets to 20 normal
             human subjects and 50 cirrhotic patients.  Blood ammonia concentration peaked
             (mean, 140 /ig NHj/100 mL)  at 15 minutes and returned to fasting levels (mean,
             105 /ig NHj/100 mL) by 30 minutes in normal human subjects. In 50 patients
             with cirrhosis of the liver, however, blood ammonia levels increased from
             elevated fasting levels (mean, 155 /ig NHj/100 mL) to higher peak concentrations
             (mean, 370 /tg NHj/100 raL)  at 15 minutes.  This was followed by a slow decrease
             in ammonia levels, reflecting impaired hepatic urea synthesis.

             Castell and Moore (1971) studied ammonia absorption from the  human gut by
             direct perfusion in four patients in whom surgical colon bypass procedures had
             been performed previously for chronic hepatic encephalopathy.  Eight solutions
             at pH 5 (acetate buffer) and eight solutions at pH 9 (trihydroxyaminomethane
             buffer) with varying NH3-N concentrations between 10 and 150 /tg/mL were

-------
Ammonia                                                                       April 1992
              infused at 15 mL/rain into a proximal colonic stoma for 20 minutes. In
              two patients, 12 infusions were also given at varying pH with both phosphate (pH
              5, 6.6 and 8.2)  and trihydroxyaminomethane buffer (pH 7, 8, 9).  Ammonia
              absorption in the human colon showed a positive linear (or slightly curvilinear)
              relationship with intraluminal concentrations of ammonia up to 150 /ig/mL.
              There was greater absorption of ammonia occurred consistently at  all
              concentrations, as the pH of the infused fluid was changed from 5 to 9.
       Distribution
              In healthy individuals, ammonia that is absorbed following oral administration is
              mainly converted in the liver to urea; therefore, relatively small amounts reach
              systemic circulation (Summerskill and Wolpert, 1970).  No other details were
              provided.
       Metabolism
              Much of the ammonia absorbed in the gut is transformed to urea in the liver,
              while some is incorporated into tissue proteins (Richards et al., 1968, 1975;
              Summerskill and Wolpert, 1970). The transformation of ammonia into tissue
              protein varies inversely to the amount of protein consumed in the diet (Kies and
              Fox, 1978).
       Excretion
              Excretion of ammonia following oral administration to humans is modified by
              protein intake. Richards et al. (1968, 1975) administered "NJ^Cl orally to
              healthy male volunteers fed a normal (70 g of protein in 24 hours) or
              protein-restricted, diet (20 g of protein in 24 hours). The total dose of 9.2 to
              17.2 mg NH4Cl/kg was administered as five divided doses at 4-hour intervals.
              Within 7 days, approximately 70% of the ingested isotope was excreted in the
              urine and feces of the test group on the normal diet. In the protein-restricted
              diet group, the excretion value was approximately 35% of the ingested isotope.
              In uremic patients, excretion of the isotope was comparable to that of healthy,
              protein-restricted individuals.

-------
Ammonia                                                                     April 1992
IV.    HEALTH EFFECTS

       Humans

             Short-term Exposure
                    Ingestion of an ammonium hydroxide solution containing 2.4% ammonia
                    resulted in the death of an adult male.  Autopsy revealed a hemorrhagic
                    esophago-gastro-duodeno-enteritis, with ammonia odor in the stomach
                    contents (Klendshoj and Rejent, 1966).
             Long-term Exposure

             •     No data were found in the available literature on the chronic toricity to
                    humans of ammonia following oral administration.

       Animals

             Short-term Exposure

             •     An acute oral LDj,, of 350 mg/kg was reported in rats administered
                    ammonia (Smyth et al., 1941).

             Dermal/Ocular Effects
             •     No studies were found in the available literature on the dermal or ocular
                    effects of ammonia by oral administration.

             Long-term Exposure

             •     No studies were found in the available literature on the long-term effects
                    of ammonia by oral administration.

             Reproductive Effects

             •     No studies were found in the available literature on the reproductive
                    effects of ammonia by oral administration.

             Developmental  Effects

             •     No studies were found in the available literature on the developmental
                    effects of ammonia by oral administration.

-------
 Ammonia                                                                      April 1992
               Mutagenicity

               •     No studies were found in the available literature on the mutagenic effects
                     of ammonia by oral administration.

               Carcinogenic! ty

               •     No studies were found in the available literature on the carcinogenicity of
                     ammonia by oral administration.

               Organoleptic Consideration

               •     Campbell  et al. (1958) determined the threshold concentration for
                     ammonia in redistilled water based on the responses of 21 to 22 judges
                     participating in "difference tests of the triangle type.11  At ammonia
                     concentrations of 26, 52 and 105 mg/L, the percentages of correct
                     identification by the judges were 61.9*. 71.4 and 85.7, respectively.
                     Defining the threshold concentration as the level at which correct
                     identification is 50% greater than that expected by chance, the taste
                     threshold of ammonia was determined to be 34 mg/L.  Based on assumed
                     water consumption of 2 L/day and average  body weight of 70 kg, this dose
                     corresponds to 0.97 mg/kg/day.
 V.     QUANTIFICATION OF TOXICOLOGICAL EFFECTS

        Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
 (up to 7 years) and Lifetime exposures if adequate data are available that identify a sensitive
 noncarcinogenic end point of toxicity. The HAs for noncarcinogenic toxicants are derived using
 the following formula:

               (NOAEL or LOAEL^ x (BW   =	mg/L (	^tg/L)
                 (UF) (_ L/day)

 where:

NOAEL or LOAEL    =     No- or Lowest-Observed-Adverse-Effect Level (in mg/kg bw/day).

               BW    =     assumed body  weight of a child (10 kg) or an adult (70 kg).

               UF    =     uncertainty factor. (10, 100, 1,000 or 10,000), in accordance with
                            EPA or NAS/OW guidelines.

-------
Ammonia                                                                      April 1992
         	L/day    =     assumed daily water consumption of a child (1 L/day) or an adult
                           (2 L/day).

       One-day Health Advisory

       No data were found in the available  literature that were suitable to use in the
determination of the One-day Health Advisory (HA).

       Ten-day Health Advisory

       No data were found in the available  literature that were suitable to use in the
determination of the Ten-day HA.

       Longer-term Health Advisory

       No data were found in the available  literature that were suitable to use in the
determination of the Longer-term HA.

       Lifetime Health Advisory

       The Lifetime HA represents that portion of an individual's total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure.  The Lifetime HA is derived in a three-step process. Step 1
determines  the Reference Dose (RfD), formerly called the Acceptable Daily Intake (ADI).  The
RfD is an estimate (with  uncertainty spanning perhaps an order of magnitude) of a daily
exposure to the human population (including sensitive subgroups) that is likely to be without
appreciable risk of deleterious health effects during a lifetime,  and is derived from the NOAEL
(or LOAEL), identified from a chronic (or subchronic) study, divided by an uncertainty
factor(s). From the RfD, a Drinking Water Equivalent Level (DWEL) can be determined
(Step 2). A DWEL is a medium-specific (i.e., drinking water)  lifetime exposure level, assuming
100% exposure from that medium,  at which adverse, noncarcinogenic health effects would not
be expected to occur. The DWEL  is derived from the multiplication of the RfD by the assumed
body weight of an adult and divided by the assumed daily water consumption of an adult.  The
Lifetime HA in drinking water alone is determined in Step 3 by factoring in other sources of
exposure, the relative source contribution (RSC). The RSC from drinking water  is based on
actual exposure data or, if data are not available, a value of 20% is assumed.

       If the contaminant is classified as a known, probable, or possible human carcinogen,
according to the Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986), then
caution must  be exercised in making a decision on how to deal with  possible lifetime exposure
to this substance.  For human (A) or probable (B) human carcinogens, a Lifetime HA is not
recommended.  For possible (C) human carcinogens, an additional  10-fold safety factor is used
in the calculation of the Lifetime HA.  The  risk manager  must balance this assessment of

-------
Ammonia                                                                     April 1992
carcinogenic potential and the quality of the data against the likelihood of occurrence and
significance of health effects related to noncarcinogenic endpoints of toxicity. To assist the risk
manager in this process, drinking water concentrations associated with estimated excess lifetime
cancer risks over the  range of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult drinking 2  L of
water/day are provided  in the Evaluation of Carcinogenic Potential section.

       There are no suitable studies available in the literature for the derivation of a Lifetime
HA for ammonia. Ammonia is produced in humans in the stomach, duodenum, ileum, colon
and feces at an estimated  4,200 mg/kg/day, with the colon and fecal content contributing about
73%.  However, ingestion of 2.4% ammonium hydroxide  solution (24 g/L) has resulted in the
death of an adult male.  Autopsy revealed a hemorrhagic esophago-gastro-duodeno-enteritis,
with ammonia odor in the stomach contents.  It appears that the observed toxic effects were due
to the local effects rather than the systemic effects and ammonia at low concentration, per se, is
not very toxic. Therefore, it is recommended that the taste and odor level of 34 mg/L be used
as a guide for the Lifetime HA.

       Evaluation of Carcinogenic Potential

       •      The International Agency for Research on Cancer (IARC) has not evaluated the
              carcinogenic potential of ammonia.

       •      The weight of evidence that ammonia is a carcinogen has not yet been evaluated
              by the EPA.  Applying the criteria described in EPA's guidelines for assessment
              of carcinogenic risk (U.S. EPA, 1986), ammonia may be placed in Group D:  not
              classifiable.  This  category is for agents with inadequate animal evidence of
              carcinogenicity.
VI.    OTHER CRITERIA. GUIDANCE AND STANDARDS

       •     The American Conference of Governmental Industrial Hygienists (ACGIH)
             suggests a Threshold Limit Value (TLV) of 25 ppm (18 mg/m3) as a
             Time-Weighted Average (TWA) for an 8-hour work day (ACGIH, 1987-88). A
             Short-Term Exposure Limit (STEL) of 35 ppm (27 mg/mj) has been suggested.
VII.   ANALYTICAL METHODS

       •     Ammonia is one of the classical water quality monitoring parameters, the
             concentration of which is important in eutrophication problems.  It has been
             monitored at low levels in ambient waters for a long time.  EPA Methods 350.1,
             350.2, 350.3 are available for the determination of ammonia. Method 350.3 is a
             specific ion electrode method, the others are colorimetric/titrimetric/

-------
Ammonia                                                                     April 1992
             potentiometric. Method 350.1 is an automated procedure. The type of
             interferences from other nitrogen compounds such as amines is well documented.
             The detection limits available by these methods ranges from 10 to 50 /xgm/L
             (U.S. EPA, 1979).
VIII.   TREATMENT TECHNOLOGIES
             Available data indicate that aeration, reverse osmosis (RO), combination lime
             softening and RO, and adsorption by natural zeolite significantly reduce
             ammonia concentrations in the drinking water supply.

             Powers et al. (1987) developed a mathematical model to examine the operating
             parameters that might affect the ammonia removal rate by a batch
             aerated-bubble stripping process.  The model requires the liquid tank volume,
             lime dosing for pH adjustment, steam flow rate dosing and air flow rate to
             produce an ammonia concentration profile at a given initial ammonia
             concentration. Using this model, the optimum operating conditions for this
             stripper are:  air flow rate  of 110 cfm and a  maximum steam flow rate of
             700 lb/hr., with a required  stripping time of 14 hours.  This model predicted a
             reduction of the ammonia  concentration by 96% from a concentration of
             4,000 mg/L as ammonia-nitrogen.

             Shpirt (1981) using a bench-scale apparatus  and a mathematical model
             demonstrated that diffused aeration is a feasible technology for ammonia
             reduction.  The apparatus consisted of two different types, one giving coarse
             bubbles (a section of glass pipe), the other giving fine bubbles (diffuser stone).
             The initial concentration of ammonium chloride was in the range of 50 to
             100 mg N/L, and the pH was adjusted to 11.5 with a 40% by weight solution of
             sodium hydroxide. A series  of experiments were performed at different column
             heights and air flow rates.  The results showed that the overall coefficient  of mass
             transfer increases with  increasing air flow rate and decreases with increasing
             depth of diffuser submersion. Furthermore, fine bubble aeration had twice the
             efficiency as coarse bubble aeration in removing ammonia.

             Houel et al. (1979) studied and reported the removal of ammonia by aeration in
             a counter current stripping tower with a cross section of 23 inches by 18 inches
             and a total packing depth of 13.83 feet. Air flow rate was maintained at
             1,250 cfm and the water flow rate  was  maintained at 5 gpm.  Two types of
             packing materials were tested.  Type A packing was a 0 J inches thick "egg crate"
             type polystyrene sheet, and Type B packing  was made from polyvinyl chloride
             (PVC) sheets.. Ammonia was present at an  influent concentration of 80 mg/L.

-------
Ammonia                                                                      April 1992
             Type A packing removed an average of 88% of the ammonia, while Type B
             packing removed an average of 53%.

             Benneworth and Morris (1972) studied the factors affecting ammonia removal by
             aeration.  They concluded that the rate of ammonia desorption increases rapidly
             as the pH rises from 7 to 10.5 and above pH 10.5 no significant improvement can
             be obtained.  The air requirements are governed by the value of the solubility
             coefficient. A theoretical minimum air to liquid ratio was calculated to be
             1,400 for 90% ammonia removal.

             O'Farrell and Bishop (1976) described a pilot plant which used aeration to
             remove ammonia from a second-stage clarifier effluent at pH 10.5. The tower
             was operated at a loading  rate of a 2 gpm/sq. ft. at an air to water ratio of 2,600.
             In general, this system, operated at 45ฐF, removed 56% of the ammonia from an
             initial concentration of 9.5 mg/L measured  as ammonia-nitrogen.

             Terril and Neufeld (1983)  reported data from a reverse osmosis unit used to treat
             blast-furnace  scrubber effluent.  The ammonia concentration in the influent was
             128 mg/L.  The RO unit contained a cellulose acetate membrane (CA)  and was
             operated at pressures  of 350 to 450 psig and a water recovery rate of 70 to 80%.
             This system achieved 93% reduction in ammonia levels.

             Argo (1984) reported  the performance of a lime softening/RO plant for water
             reclamation.  Potable water was reclaimed  from the unchlorinated effluent of an
             activated sludge wastewater treatment plant by lime softening, reacidification and
             RO.  Two 5,000 gpd RO pilot plants consisting of tubular aromatic polyamide
             membranes in a 2-1 array  configuration were operated in parallel at a flux rate of
             7.14 gpd/sq ft and at an applied pressure of 250 psi.  This system reduced
             ammonia concentration by more than 95% from an average influent
             concentration of 15 mg/L.

             Blanchard et al. (1984) studied removal of ammonia in a  pilot plant using natural
             zeolite clinoptilolite, as an ion exchanger.  The pilot plant consisted of two
             columns operated in series, each 8 inches in diameter and packed with 40 inches
             of zeolite, and a flow rate  of 12 bed volumes (BV) per hour, for a total BV =
             1.16 cu. ft. Ammonia breakthrough occurred after 480 BV at an influent
             concentration of 2.63 mg/L.  Breakthrough concentration was set at 50 jig/L, after
             which the zeolite was regenerated with NaCl at a  flow rate of 10 BV/hr.

             No data were found for the removal of ammonia from drinking water by
             activated carbon adsorption. However, ammonia  may not be amenable to
             removal by activated carbon adsorption due to its very  high solubility and low
             molecular weight.

                                           10

-------
Ammonia                                       .                              April 1992
IX.    REFERENCES

ACGIH.  1987-88.  American Conference of Governmental Industrial.Hygienists. Threshold
       limit values and biological exposure indices. Cincinnati, OH:  American Conference of
       Governmental Hygienists, p. 12.

Argo, O.R. 1984.  Use of lime clarification and reverse osmosis in water reclamation.  Jour.
       Water Pollut. Cont. Fed.  56(12): 1,238-1,246.

Benneworth, N.E. and N.G. Morris.  1972. Removal of ammonia by air stripping.  Jour. Water
       Pollut. Cont. Fed. 71(5):485-492.

Blanchard, G., M. Maunaye and G. Martin. 1984. Removal of heavy metals from waters by
       means of natural zeolites. Water Resources  18(12): 1,501-1,507.

Campbell, C.L., R.K. Dawes, S. Deolakkar and M.C. Merritt.  1958. Effects of certain
       chemicals in water on the flavor of brewed coffee.  Food Res. 23:575-579.

Castell, D.O. and E.W. Moore.  1971. Ammonia absorption from the human colon.
       Gastroenterology 60(l):33-42.

Conn, H.O.  1972.  Studies of the source and significance of blood ammonia. IV. Early
       ammonia peaks after ingestion of ammonium salts.  Yale J. Biol. Med. 45:543:549.

FASEB.  1974.  Federation of American Societies for Experimental Biology. Evaluation of the
       health aspects of certain ammonium salts as food ingredients.  FDA contract no. FDA
       72-85, NTIS PB-254-532.

Feth, J.H. 1966. Nitrogen compounds in natural water-A review.  Water Resour. Res. 2:41-58.

Houel, N., F.H. Pearson and R.E. Selleck. 1979.  Air stripping of chloroform from water.  Jour.
       of Environ. Engr. 105:777-781.

Kies, C. and H.M.  Fox.  1978. Urea as a dietary  supplement for humans. Adv. Exp. Med. Biol.
       105:103-118.

Klendshoj, N.C. and T.A. Rejent. 1966. Tissue levels of some poisoning agents less frequently
       encountered. J. Forensic Sci. 11:75-80.

O'Farrell, T.P. and D.F. Bishop.  1976. Conventional tertiary treatment. U.S. Environmental
       Protection Agency.  EPA-600/2-76-251.
                                           11

-------
Ammonia                   '                                                  April 1992
Perry, R.H. and C.H. Chilton.  1973.  Chemical engineers handbook, 5th ed. New York, NY:
       McGraw Hill Book Co.

Powers, S.E., A.G. Collins, J.K. Edzwald and J.M. Dietrich.  1987.  Modeling an aerated bubble
       ammonia stripping process.  Jour. Water Pollut. Cont. Fed. 59(2):92-100.

Richards, P., B.J. Houghton, A. Metcalfe-Gibson, E.E. Ward and O. Wrong. 1968.
       Incorporation of orally administered ammonia into tissue proteins in man: The
       influence of diet and uraemia. In:  Life Systems,  Inc., ed., Nutrition and renal disease.
       Edinburgh,  Scotland:  E&S Livingstone, pp. 93-98.

Richards, P., C.L. Brown, BJ. Houghton and D.M. Wrong. 1975. The incorporation of
       ammonia nitrogen into albumin in man: The effects of diet, uremia and growth
       hormone.  Clin. Nephrol. 3(5):172-179.

Shpirt, E.  1981.  Diffused-air stripping of ammonia in advanced wastewater treatment.
       Chemistry in Water Reuse 2:497-508.

Smyth, H.F., Jr., J. Seaton and L. Fischer.  1941. The single dose toxicity of some glycols and
       derivatives.  J. Ind. Hyg. Toricol. 23(6):259-268.

Summerskill, W.H.J. and E. Wolpert. 1970.  Ammonia metabolism  in the gut.  Am. J. Clin.
       Nutr. 23(5):633-639.

Terril, M.E. and R.D. Neufeld. 1983. Reverse osmosis of blast-furnace  scrubber water.
       Environ. Progress  2(2):121-127.

U.S. EPA. 1979.  U.S. Environmental Protection Agency. Methods for  the chemical analysis of
       water and wastes.  EPA-600/4-79-020. Environmental Monitoring and Support
       Laboratory, ORD.  Cincinnati, OH: U.S. EPA.

U.S. EPA. 1986.  U.S. Environmental Protection Agency. Guidelines for carcinogenic risk
       assessment.  Fed. Reg. 51(185):33992-34003.

U.S. EPA. 1979.  U.S. Environmental Protection Agency. Development document for effluent
       limits, guidelines and standards for the  petroleum refinery point-source category. EPA
       440/1-79-0146.

Verschueren, K.  1977.  Handbook of environmental data on organic chemicals. New York,
       NY: Van Nostrand Reinhold Co., p. 96.
                                           12

-------
Ammonia                                                                     April 1992
Windholz, M., S. Budavari, R.F. Blumetti and E.S. Otterbein, eds.  1983.  The Merck index—
       An encyclopedia of chemicals, drugs and biologicals, 10th ed.  Rahway, NJ: Merck and
       Co., Inc., p. 498.
                                           13

-------
EPA 0553

RX000027511                                                                          Apnl
                                            ANTIMONY

                                   Drinking Water Health Advisory
                                           Office of Water
                                U.S. Environmental Protection Agency
   I.      INTRODUCTION
          The Health Advisory Program, sponsored by the Office of Water (OW), provides
   information on the health effects, analytical methodology, and treatment technology that would be
   useful in dealing with the contamination of drinking water.  Health Advisories (HAs) describe
   nonregulatory concentrations of drinking water contaminants at which adverse health effects would
   not be anticipated to occur  over specific exposure durations.  Health Advisories contain a margin of
   safety to protect sensitive members of the population.

          Health Advisories serve as informal technical guidance to assist Federal, State, and local
   officials responsible for protecting public health when emergency spills  or contamination situations
   occur. They are not to be  construed as legally enforceable  Federal standards. The HAs are subject
   to change as new information becomes available.

          HAs are developed for One-day, Ten-day, Longer-term (approximately 7 years, or 10% of
   an individual's lifetime), and Lifetime exposures based on data describing noncarcinogenic endpoints
   of toxicity.  For those  substances that are known or probable  human carcinogens, according to the
   Agency classification scheme (Group A or B), Lifetime Health Advisories are not recommended.
   For substances with a carcinogenic potential, chemical concentration values are correlated with
   carcinogenic risk estimates  by employing a cancer potency (unit risk) value together with
   assumptions for lifelong exposure and the ingestion of water.  The cancer unit risk is  usually
   derived from a linearized multistage model with 95% upper confidence  limits providing a low-dose
   estimate of cancer risk.  The cancer risk is characterized as being  an upper limit estimate, that is,
   the true risk to humans, while not identifiable, is not likely  to exceed the upper limit estimate and in
   fact may be lower.  While  alternative risk modeling approaches may be presented, for example One-
   hit, Weibull, Logit, or Probit, the range of risks described by using any of these models has little
   biological significance  unless data can be used to support the selection of one model over another.
   In the interest of consistency of approach and in providing an upper-bound on the potential
   carcinogenic risk, the Agency recommends using the linearized multistage model.

-------
Antimony Health Advisory
                                       April  1992.
II.     GENERAL INFORMATION AND PROPERTIES

       CAS No.

       •      Antimony — 7440-36-0
              Potassium antimony tartrate — 2800-74-5
              Sodium antimony tartrate — none
              Sodium antimony bis(pyrocatechol) 2,4-disulfate — none

       Synonyms

       •      Antimony black; CI77050;  Regulus of antimony; Stibium (HSDB,  1988)

       Uses

       •      Alloy in semiconductor technology, batteries, antifriction compounds, ammunition,
              cable sheathing, flameproofing compounds, ceramics, glass and pottery (Weast et
              al., 1986).  Also used  in type castings for commercial printing.

       Properties (ACGIH,  1980; HSDB, 1988; Weast et al., 1986; Windholz et al., 1983)
Antimony
                                         Potassium
                                         antimony
                                         tartrate
              Sodium
              antimony
              tartrate
Sodium antimony
bis(pyrocatechol)
2.4-disulfate
Chemical Formula
Molecular Weight
Physical State (25 ฐC)
Boiling Point
Melting Point
Density (20ฐC)
Vapor Pressure
Water Solubility
Specific Gravity
Log Octanol/Water
  Partition Coefficient
Taste Threshold

Odor Threshold
Sb             KSbOCAOj
121.75         324.92       308.83
Silver-white,    Colorless     Transparent
hard, brittle     crystals or    or whitish
metal          white powder scale or
                            powder
1,635ฐC        -           -
630ฐC          1008C        -
6.691          -           -
1 mmHg <3> 886ฐC
Insoluble       8.3 mg/L
6.691          2.6
Sweetish
metallic
Odorless
                             895.21
                             Fine crystals
              66.7 mg/L      Soluble
       Occurrence

       •      Antimony (Sb) is a naturally occurring element found as various salts in seawater,
              surface water, soils and sediments. Its terrestrial abundance is of the order of 0.7
              Mg/g (Brannon and Patrick, 1985). Antimony concentration in the earth's crust is

-------
Antimony Health Advisory                                                          April  199,2.
              about 0.2 to 1.0 mg/kg (Rompp, 1979).  More than half of the naturally occurring
              ,Sb in sediments is bound to extractable iron and aluminum (Crecelius et al.,  1975).
              Sb(III) and Sb(V) forms and methyl antimony compounds have been shown to exist
              in natural waters (Andreae et al., 1981; Byrd and Andreae, 1982).  Antimony occurs
              in seawater at about 0.2 ^g/L (Venugopal and Luckey, 1978).  Schroeder et  ai.
              (1970) found Sb in the air only in a few cities (4 of 58 had 0.42 to 0.85 /zg Sb/m3)
              and in a  few non-urban areas (3 of 29 with 0.001  to 0.002 /ig Sb/m3) of the United
              States. Smoke condensate of cigarettes contains 35 to 60 mg Sb/kg (Gerhardsson
              1983).

         •    Ragaini et al. (1977) reported that lead smelting operations in the Kellog Valley of
              Idaho resulted in soil Sb concentrations ranging from 5 to 260 jig/g. Crecelius et al.
              (1975) reported total Sb concentrations in Puget Sound sediments within 8 to 15 km
              from a copper smelter were 2 to 3 times  higher than background values. Elevated
              concentrations of Sb in sediment have also been noted near the outfall of sewage and
              fertilizer facilities (Papakkostides et al., 1975). Sludges used for manuring soils in
              Indiana (United States) and collected near Vienna (Austria) from the Danube  River
              contained between 4 and 22 mg/kg of Sb.

       Environmental Fate

       •      Various forms of Sb found in the environment from natural and anthropogenic
              sources undergo a complex cycle of chemical interconversion and transfer between
              media. Antimony in water may undergo  either oxidation or reduction, depending on
              pH  and other ions  present.  Soluble forms of Sb (e.g., antimony potassium oxalate
              and antimony potassium tartrate) tend to be quite mobile in water, while less  soluble
              species adsorb to clay or soil  particles (Callahan et al., 1979).

       •      Antimony in gaseous, vapor and paniculate forms enters the atmosphere and  is
              transported via air until it undergoes atmospheric fallout or washout and is deposited
              in oceans, estuaries, lakes, rivers, sediments and terrestrial systems.  Antimony may
              enter the food chain via root uptake by terrestrial plants and via bioaccumulation in
              fish- and plant-eating mammals.  Antimony deposited in sediment can also be
              released to the atmosphere through microbial activity under anaerobic conditions.
              Antimony may leach from municipal landfills, sewage sludge,  oil-fired plant
              incinerator ash and fertilizers  to contaminate ground water, surface water and
              sediment (Callahan et al.,  1979).

       •      Brannon  and Patrick (1985) determined that So-amended sediment releases volatile
              Sb compounds during anaerobic incubation. Antimony evolution rates of 8 ^g/
              mVweek were detected.  Release of volatile Sb compounds, presumably stibine, was
              most pronounced in the first week of incubation.  Two sediment samples released
              additional Sb when their overlying water  was aerobic. These observations indicate
              that release of volatile Sb compounds from anaerobic sediments containing recent
              deposits of soluble Sb can occur regardless of the oxygen status of the overlying
              water. However; no releases of volatile Sb compounds were noted from sediments
              containing no added Sb.  These results also suggest that So-amended sediments will
              potentially release  greater amounts of Sb  during water-sediment interactions than
              native  So-containing sediments (Brannon  and Patrick, 1985).

-------
Antimony Health Advisory                                                          April 1992.
        ป      In a study on leaching of contaminants from a municipal landfill site, Cyr et al.
               (1987) found that Sb was present at a level of 0.01 mg/L in the landfill leachate but
               was not detected in the background monitoring well.  The river sediment near the
               landfill site contained 23.9 mg/kg of Sb; however, there was no detectable Sb in the
               sediment at the monitoring station further upstream.  These results suggest that some
               Sb  is adsorbed by  sediments.

         •     Antimony  is only slightly bioaccumulated and has been little studied in aquatic
               organisms.  Leatherland  et al. (1973) found low levels of Sb in various fish and
               invertebrates collected off the northwest coast of Africa; Sb was generally present in
               higher concentrations in  invertebrates than in fish.  Aquatic organisms from the
               Danube River and Danube  Canal in Vienna, Austria, were found to contain only
               background levels  of antimony (Rehwoldt et al., 1975).  Bertine and Goldberg
               (1972) obtained similar results in clams, mussels and shrimp.

         •     Antimony  may be  found  as a gas in the form of stibine or its methylated derivatives.
               Stibine can be formed  by reduction of Sb in the sediments.  The elements Sn, Pb,
               As, Se and Te, which  surround Sb in the periodic table, are subject to
               biomethylation, suggesting  the possibility of similar biomethylation pathways for Sb
               (Parris and Brinkman,  1975,  1976).  Stibine is rapidly oxidized in air or oxygenated
               waters to  form Sb2O3.  It is likely then, that most of the stibine formed in the
               sediments reacts in the water column to produce the oxide, resulting in
               remobilization of Sb.

         •     The methylated forms  of Sb are also subject to oxidation. Parris and Brinkman
               (1976) estimate the rate of  oxidation of trimethylstibine as greater than 10"2/M/sec.
               The product of this reaction,  (CHj)3SbO, is much more soluble than trimethylstibine,
               and, therefore, this oxidation would probably tend to reduce volatility. The rapid
               rate of oxidation implies  that, if trimethylstibine is formed in natural systems, much
               of it would be oxidized before it volatilized and only  a small amount of the volatile
               antimony compounds formed by either abiotic or biotic  mechanisms would be
               liberated to the atmosphere.

         •     It has been reported that  a species of bacteria, Stibiobaaer senarmontii, utilizes the
               energy released by metabolically induced oxidation of Sb (Lyalikova,  1974), but the
               distribution and ecological importance of this organism is unknown.
III.    PHARMACOKINETICS

       Absorption

       •      Antimony is not readily absorbed from the gastrointestinal (GI) tract. Approximately
              15% of the administered dose of radioactive Sb (4.4 mg potassium antimony tartrate/
              kg plus 7 jig of mSb) was estimated to be absorbed from the intestine of rats.  The
              animals (30 white rats) were dosed via gastric gavage or intravenous injection
              (Moskalev, 1959).

-------
Antimony Health Advisory                                                          April
       •      Gerber et al.  (1982) reported that the total-body radioactivity (1.7% of the daily
              intake) reached an equilibrium  within 4 days in pregnant BALB/c mice following
              exposures via diet and intraperitoneal (ip) injection. Assuming a half-life of 6 hours,
              7% of the ingested SbCl3 (given as '"SbClj in food) was absorbed in pregnant
              BALB/c mice following repeated dosing.

       •      Combined elimination data from cows administered single  oral or intravenous (iv)
              doses of 124SbCl3 indicated that very little (< 5%) of the orally administered dose
              was absorbed via  the GI tract of ruminants.  About 82% of the orally administered
              radiolabel was excreted in the feces (Van Bruwaene et al., 1982).

       •      Very  little (9 to 15%) trivalent or pentavalent Sb was absorbed when administered
              via oral gavage to 20 to 34 Syrian hamsters. Animals received oral doses  containing
              1 or 2 /iCi/mL of 124Sb-tartrate. For both valence states, Sb was retained in the
              body, with  a half-life of less than 1 day (Felicetti et al., 1974).

       Distribution

       •      Gerber et al.  (1982) indicated that concentrations of mSb (expressed as percent daily
              dose per gram tissue) in lung, bone, ovaries and uterus ranged from 0.085 to 0.2%
              when given in food to pregnant BALB/c mice.  The diet containing mSbC!3 was
              started on the day the vaginal plug was observed.  The animals were sacrificed 6
              days later, and tissue levels of  125Sb were measured.  The results were judged by the
              authors to be somewhat unreliable due to low levels of radioactivity  observed.

       •      Westrick (1953) fed diets containing 0 or 2%  Sb,Oj for 7 weeks to five male
              Sprague-Dawley rats.  Using a mean body weight of 0.18 kg (the mean of reported
              initial and final weights) and assuming average food consumption of 12 g/day
              (Arlington, 1972), this corresponds to an average daily dose of about 1,100 mg Sb/
              kg/day. After 7 weeks, average concentration of Sb in the liver, kidney, heart,
              spleen, lung,  adrenal and thyroid was 8.9, 6.7, 7.6, 18.9,  14, 67.8 and 88.9 /xg Sb/g
              tissue, respectively.  A wide variation was observed among replicates of some
              tissues.

       •      Gerber et al.  (1982) also measured tissue distribution in pregnant BALB/c mice
              following ip injection of l25SbCl3  on day 12 of pregnancy.  Peak concentrations  in
              tissues (percent dose per gram  tissue) were observed 2 to 6 hours after injection.
              Highest tissue levels (approximately 50%) were seen in the intestine and bone
              surfaces.  Levels in other tissues  observed at 2 hours were 1 to 5% in the uterus and
              ovary, and 0.01 to 1.0%  in the kidneys, liver, spleen, lung, thyroid, blood, muscle,
              skin and brain.  Low levels (about 0.1%) were measured in the placenta and fetus.

       •      Casals (1972) injected two groups of 10 female mice intramuscularly (im)  either with
              antimony dextran  glucoside (RL-712, 52 mg Sb/kg) or with N-methyl-glucamine-
              antimonate  (glucantime, 50.3 mg Sb/kg). Animals were sacrificed at various
              intervals between  6 hours and 6 weeks after dosing. High levels (100 to 150  Mg) of
              Sb were found mainly in the liver and spleen throughout that period; the levels
              decreased gradually with time.

-------
Antimony Health Advisory                                                          April 199Z
       •      Rowland (1971) presented a detailed mathematical model of the distribution of Sb in
              humans following a single iv injection of 124Sb-labeled potassium antimony tartrate
              (PAT) .  Four main compartments (blood, liver, skeletal tissue and urine) were
              considered by determining radioactivity in the blood and the urine, and via surface
              scanning over the liver and over the thigh (as representative of skeletal tissue).  The
              results from studies on a single patient revealed that 12 minutes after an injection of
              48 mg labeled PAT, 12% of the radioactive dose was present in the blood, which
              declined to 8 and 6%  at  1 and 2 hours post-injection. Another experiment revealed
              that 40 to 50%  of a 24 mg dose of labeled PAT was taken up by the liver within a
              few minutes to 48 hours  post-injection.  On average, this was followed by a slow
              decline with about 50%  of the peak value present 16 hours post-treatment. The
              content of skeletal tissue reached its peak in 1 hour with a slow decline with 30% of
              the dose remaining after  10 days post-treatment.  In the urine, 7% of a 48 mg
              labeled PAT dose appeared in the first hour followed by 15, 22, 25 and 45%  of the
              dose at 6 hours and day  1, 2 and 3, respectively.

       •      Leffler and Nordstroem (1984) demonstrated the transfer of Sb from maternal to
              fetal blood in three Syrian Golden  hamsters intratracheally exposed t;o Sb  on days 13
              and 15 after fecundation.  Concentration gradients of Sb determined in maternal and
              fetal blood suggested a possible carrier function of the placenta.

       •      Molokhia  and Smith (1969) incubated Sb (trivalent or pentavalent) compounds with
              equine whole blood in vitro and found that the erythrocyte membrane was permeable
              to trivalent antimony and impermeable to pentavalent antimony.  Trivalent Sb  bound
              to plasma proteins but not to erythrocytes.

       Metabolism

       •      Otto and Maren  (1950) found large amounts of Sb in erythrocytes following im
              injection of stibanose (6 mg Sb(V)/kg) in 11 dogs. Other results (Molokhia and
              Smith, 1969; Otto et al., 1947) had shown that Sb(V) does not enter erythrocytes but
              that Sb(IH) does, suggesting  that the Sb(V) may have been reduced to Sb(III) in
              vivo.  In contrast, at a lower im dose of 0.5 mg Sb(V)/kg, Otto and Maren (1950)
              did not detect Sb accumulation in erythrocytes of dogs.  Similarly, iv injection of
              either 0.5 or 5 mg Sb(V)/kg did not reveal Sb entry into red cells.  The authors
              stated that the available data were not sufficient to support a conclusion regarding
              possible reduction of Sb(V) to Sb(III).

       •      Goodwin and Page (1943) used polarography to analyze the valence state of
              antimony in blood and urine of seven humans injected iv with Sb(V).  During the
              first  12 hours after the administration of Sb(V) (sodium antimony gluconate
              equivalent to 50 Mg Sb),  83.5% (average of three subjects) of the administered dose
              was excreted in the urine as  Sb(V). Only 2.5% of the administered dose was
              excreted as Sb(III) during the same period, indicating that reduction of Sb(V) to
              Sb(III) was slight. Otto and Maren (1950) pointed out that some of the Sb(III) found
              in urine may have been formed during sample preparation in hydrochloric acid for
              polarographic examination.

-------
Antimony Health Advisory                                                         April 1992.
              Otto et al. (1947) administered two Sb(m) compounds (lithium antimony thiomalate
              and monosodium antimony thioglycollate) im and two Sb(V) compounds (stibanose
              and neostibosan) iv to 14 adult male filariasis patients.  Antimony concentrations in
              red blood cells and plasma were measured colorimetrically.  For both Sb(III) and
              Sb(V), plasma concentrations were sustained for only a short time (well under 24
              hours).  For both trivalent compounds, Sb was found largely inside the red blood
              cells, with very little in plasma, and the converse was observed for both pentavalent
              compounds.  The authors concluded that Sb(III) readily enters red blood cells, but
              Sb(V) does not.
       Excretion
              Otto and Maren (1950) reviewed the routes of excretion of parenterally administered
              antimony in the mouse, white rat, hamster, guinea pig, rabbit, dog and human.
              Trivalent antimony was excreted via the feces and urine.  With the exception of the
              mouse, pentavalent antimony was excreted primarily in the urine. While the percent
              of the dose excreted in the feces was less than 5% for all  species tested, the percent
              excreted in the urine was approximately 80, 60, 65, 70, 10 and 43% in the white
              rat, hamster, guinea pig, rabbit, dog and human, respectively.

              Casals (1972) injected (im) female mice either with antimony dextran glucoside (RL-
              712, 52  mg Sb/kg) or with N-methyl-glucamine-antimonate (glucantime, 50.3 mg
              Sb/kg), and female albino rats with RL-712 (50 mg Sb/kg).  In 48 hours, only 12
              and 10% of the doses  administered were excreted in the urine of mice and rats,
              respectively.

              Van Bruwaene et al. (1982) administered single oral doses of 124SoCl3 (2.84, 2.72 or
              2.00 mCi) to three lactating cows.  Since the compound had a specific activity of 3.5
              x 10"2  mCi/mmol, the average dose corresponds to 21.1 mg Sb/kg.  Total excretion
              of Sb  in feces was triphasic and totaled about 82% of the dose.   Most of the
              radioactivity in the feces appeared shortly after dosing (t,/, = 0.91 day).  Excretion
              in urine  and milk was  biphasic and totaled 1.1 and 0.008% of the dose in urine and
              milk,  respectively.  Most of the urinary radioactivity appeared in the initial phase (t,/
              2  = 0.97 day).  Radioactivity in tissues, at 102 days after dosing, totaled 0.024% of
              the oral dose. The highest radioactivity was found in the spleen, liver, bone and
              skin.

              Lippincott et al. (1947) administered potassium antimony tartrate (18 mg Sb/mL to
              77 patients over 29 days, for a  total of 576 mg/patient) or ruadin (8.7 mg Sb/mL to
              33 patients over 25 days, for a  total of 566 mg/patient) parenterally  to humans as
              treatment for infection with Schistosoma japonicum.  The average 24-hour excretion
              of Sb  in urine of subjects given potassium antimony tartrate ranged from 12 to 25%,
              whereas excretion in the group  given ruadin ranged from 17 to 42%. The combined
              excretion of Sb in urine and feces within 48 hours was approximately 55 % of the
              administered dose.

              In the Otto et al. (1947) study in adult males, most (21.6 to 70% of the daily dose)
              of the Sb administered daily was excreted in urine in 24 hours with only low levels
              (0.8 to 8.4% of the daily dose) present in feces.  Pentavalent antimony was excreted

-------
Antimony Health Advisory                                                          April  199,2,
              in urine more rapidly than trivalent antimony (up to 52% in 24 hours as opposed to
              15% in 24 hours for trivalent).
IV.    HEALTH EFFECTS

       Humans

         Short-term Exposure

       •      Kaplan and Korff (1937) briefly reviewed several reports of "food poisoning"  that
              were traced to Sb extracted from enamel-coated vessels by acid contents (lemonade).
              Symptoms were not detailed, but acute attacks of vomiting occurred in at least one
              case.  The amount of Sb ingested was not reported. Tests performed by the authors
              indicated that 0.5 to 2.6 mg of Sb,  an amount equal to about one-fourth of an emetic
              dose, could be extracted from 200 g of sauerkraut.

       •      Miller (1982) reported a fatality due to Sb poisoning.  The patient was administered
              two or three oral doses of James powder (each dose containing 66 mg of Sb) as
              treatment for headache, kidney trouble and fever.  The total dose was 132 to 198 mg
              of Sb (1 to 1.5 mg/kg of body weight). The treatment resulted in severe vomiting
              and diarrhea lasting for 18 hours and, finally, death.

       •      Jolliffe (1985) reported that sodium stibogluconate  ("Pentostam"), given iv in a
              standard daily dose of 600 mg Sb(V) for 10 days to 16 British soldiers with
              cutaneous leishmaniasis, did not adversely affect either glomerular or renal function.

       •      Schroeder et al. (1946) reported the effect of trivalent and pentavalent antimony
              compounds on the electrocardiogram of human patients being treated for
              schistosomiasis.  Sodium antimony bis(pyrocatechol-2,4-disulfonate) (Stibophen NF
              or fuadin) was given im and potassium antimony tartrate was given iv daily or on
              alternate days for about 1 month. Assuming an average body weight of 70 kg,
              average daily doses ranged from 0.24 to 0.89 mg Sb/kg/day. Examination of 315
              electrocardiograms (EKGs)  from 100 patients revealed that the EKGs were not
              indicative of cardiac damage or serious impairment of cardiac function.

       •      Rugemalila (1980) reported two deaths due to parenteral antimony (astiban)
              intoxication. The first case involved a 4-year-old girl with a history of periodic
              fevers. She was administered (route not specified) 100 mg astiban (stibocaptate)  for
              active schistosomiasis (about 2 mg Sb(UI)/kg), followed by a second dose 2 weeks
              later. The second case involved a 70-year-old woman who was  put on weekly
              injections of 320 mg astiban (about 2 mg Sb(III)/kg) to control  hookworm. Both
              patients died shortly after the second dose.

         Long-term Exposure

       •      Oliver (1933) examined six adult males who had worked in  a Sb smelter for 2 to 13
              years.  The workers had received considerable exposure to Sb as evidenced by the
              presence of Sb in the feces  (an average of 47.5 mg total/day) but not in  the urine.

-------
Antimony Health Advisory                                                         April  L99JJ.
              No signs of adverse effects were identified, including cardiac, kidney or bladder
              effects, general health and hematology.

       •      Brieger et al.  (1954) examined workmen in a plant where antimony trisulfide was
              used in the manufacture of grinding wheels.  Antimony levels ranged from 0.58 to
              5.5  mg/m3 (equal to approximately 0.4 mg/kg antimony trisulfide). In the workers
              studied,  14 of 113 had blood pressures that were > 150/90 mmHg, and 37 of 75
              showed significant changes in their EKGs, mostly in the T-wave.  Ulcers were
              detected in 7 of 111 exposed persons (63/1,000) as compared with 15/1,000 in  the
              total plant population.  No other disorders suspected of being related to Sb exposure
              were observed.

       •      Chulay et al. (1985) observed EKG changes in 59 Kenyan patients dosed orally with
              10, 20 or 40 to 60  mg/kg/day Sb(V) (sodium stibogluconate) for leishmaniasis.
              Dose-related increases in EKG abnormalities were found  following 65 courses of Sb
              treatment which lasted for 4 months. The incidences of EKG abnormalities were
              22% (2/9 patients) at  10 mg Sb/kg/day; 52% (25/48 patients) at 20 to 30 mg Sb/kg/
              day;  and 100% (8/8 patients) at 40 to 60 mg Sb/kg/day.  Furthermore, the frequency
              of EKG abnormalities increased with the duration of treatment.

       •      Belyaeva (1967) presented evidence suggestive of possible adverse effects of Sb in
              female workers employed in an Sb plant. The female workers in the Sb plant
              showed increased incidence of spontaneous late abortions  (12.5%) when compared to
              female workers working under similar conditions but not  exposed to Sb dust (4.1 %).

       •      Doll (1985) compared mortality due to lung cancer with mortality  due to other
              causes in a British factory manufacturing antimony oxide. According to the authors,
              the compound was manufactured before World War II, but no records were available
              before 1961 either for dust measurements or for males who terminated employment.
              An increase in lung cancer (Standardized Mortality Ratio  =  186) was observed  for
              men  first employed prior to  1961.  Antimony oxide  dust concentrations have been
              greatly reduced since the 1960s.

       •      Potkonjak and Pavlovich (1983) reported clinical findings for 51 male workers (ages
              31 to 54) exposed for 9 to 31 years to dust containing a mixture of antimony trioxide
              (ranging from 39 to 89%) and antimony pentoxide (ranging from 2 to 8%) in an
              antimony smelting plant. Characteristic changes observed  in the smelters were
              described as a form of pneumoconiosis simplex or antimoniosis. The symptoms
              observed included chronic coughing, conjunctivitis, orange-colored staining of
              frontal tooth surface, chronic bronchitis, chronic emphysema, inactive tuberculosis
              and pleural adhesions.  "Antimony dermatitis" characterized by vesicular or pustular
              lesions was seen in more than half the exposed workers.

       Animals

         Short-term Exposure

       •      The acute oral LDM values for potassium antimony tartrate (tartar emetic) in mice
              and rats range from 115 to 600 mg Sb/kg (Bradley and Fredrick, 1941;  HSDB,

-------
Antimony. Health Advisory                                                          April  1992,
              1988), whereas an oral LDM of 15 mg Sb/kg has been reported for rabbits (HSDB,
              1988). The iv and ip LDM values for Sb and its various salts in mice, rats, guinea
              pigs and rabbits are generally somewhat lower, ranging from 11 to 329 mg Sb/kg
              (Bradley and Fredrick, 1941;  Ercoli, 1968; Ghaleb et al.,  1979; Girgis et al., 1965;
              HSDB, 1988).

       •      Flury  (1927) determined emetic doses of six Sb compounds (antimony trioxide,
              antimony pentoxide, sodium antimonate, potassium antimonate,  sodium meta-
              antimonate and potassium antimony tartrate) in dogs.  Potassium antimony tartrate
              was the most effective compound, causing emesis at 33 mg/kg (about 12 mg Sb/kg).
              A dose of 16 mg/kg (about 6  mg Sb/kg) produced no  apparent effect.

       •      Cats appear to be more sensitive than dogs to the emetic effect of potassium
              antimony tartrate. Flury (1927) observed emesis in three cats given 11.5  or 14.3
              mg/kg (4.3 or 5.4 mg Sb/kg)  in 50 mL of water via stomach tube.  One cat dosed
              with 6.9 mg/kg (2.6 mg Sb/kg) showed no apparent response.

       •      Flury  (1927) fed high doses of five Sb  compounds to rats (one per chemical) for 9
              days.   Each rat received daily doses increasing from 100 mg to  2 or 3 g.  Doses up
              to 2 g/day of antimony trioxide or antimony pentoxide or up to  3 g/day of sodium
              meta-antimonate caused no adverse effects.  Potassium antimony tartrate was found
              to be toxic, however, causing death after the daily dose was increased to 500 mg
              (about 1,000 mg/kg  Sb/kg) on day 7.  Potassium antimonate produced adverse
              effects at dose levels of 2 g/day, but recovery was rapid when dosing ceased.

       •      Pribyl  (1927) investigated the toxicity and effect on nitrogen metabolism in four
              rabbits given 15 mg potassium antimony tartrate/kg/day (given in a milk plus sugar
              solution) over a 7- to 22-day period.  This corresponds to a dose of 5.6 mg Sb/kg/
              day. Nonprotein nitrogen, urea nitrogen, and ammonia nitrogen  were measured in
              the blood and urine of each animal before and after exposure. A small rise (10 to
              13%)  in nonprotein nitrogen in blood and urine was observed (no p value given);
              this was partly due to an increase in urea nitrogen.  Mean urine  ammonia nitrogen
              was also slightly increased (7%, no p value given).  The author  interpreted these
              increased nitrogen levels in blood and urine as evidence of increased protein
              catabolism. Gross and microscopic examination  showed hemorrhagic lesions in the
              stomach and small intestine, liver atrophy with fat accumulation and congestion, and
              hemorrhage in the kidney cortex, with some tubular necrosis. This study suggested
              a Lowest-Observed-Adverse-Effect Level (LOAEL) of 5.6 mg/kg/day based on
              minimal histological injury in  tissues.

         Dermal/Ocular Effects

       •      No information was  found in the available literature on the dermal/ ocular effects of
              Sb in experimental animals. However, an accumulated dose of 30 g of Sb (iv
              injections of potassium antimony tartarate, 87.5  g in 86 days) produced leukoderma
              and rough,  dull, bumpy, granular skin in an 18-year-old male patient
              (Christopherson, 1921).
                                             10

-------
Antimony Health Advisory                                                           April 199ฃ,
       •      Fifty-one male workers, exposed for 9 to 31 years to dust containing a mixture of
              antimony trioxide and antimony pentoxide in an antimony smelting plant, were
              examined for clinical effects. Thirty-two of the 51 had "antimony dermatitis,"
              characterized by vascular or pustular lesions (Potkonjak and Pavlovich,  1983).

         Long-term Exposure

       •      Flury (1927) exposed rats (two per test group) to potassium antimony tartrate,
              potassium antimonate, antimony trioxide, or antimony pentoxide in food.  Doses
              began at 0.1 mg/day and were steadily increased over the course of 107 days to a
              final level of 4 mg/day.  No toxic effects were observed,  and growth was unaffected
              except for a stimulation  of growth at low doses.

       •      In another study, Flury (1927) tested higher doses of potassium antimony tartrate,
              antimony trioxide,  and sodium meta-antimonate in food for 131 days.  Groups of
              two rats were exposed to doses of the first two compounds beginning at 1  mg/day
              for 45 days, and then increasing over the course of 86 days to 200 mg/day, with the
              dose being  increased steadily over the course of exposure.  The third compound was
              given in doses from 3 to 1,000 mg/day in a similar pattern.  No effects were seen,
              even at the highest doses, for antimony trioxide and sodium meta-antimonate, but
              potassium antimony tartrate caused a systemic deterioration and death at 200 mg/
              day.  This corresponds to about 485 mg Sb/kg/day, based on a mean body weight of
              155 g reported by the author..

       •      In a 91-day oral toxicity study in male/female Wistar rats, Bombard et al.  (1982)
              reported that two Sb-containing pigments produced no effects on behavior, food
              consumption, growth, mortality, hematological and clinical data, or organ  weights.
              The pigments were nickel rutile yellow [(Tio ggSb0.0jNioa7J)OJ and chrome rutile
              yellow [(Tio.ปซSbo.
-------
Antimony Health Advisory                                                          April 199Z
              females during the first year, but did result in weight loss in males after 18 months
              (p < 0.025) and decreased weight gain in females measured at 12 and 18 months (p
              <0.005).  In females, Sb shortened the median and 75% life spans by 49 to 86 days
              whereas males were only minimally affected. Upon necropsy, histologic
              examination of the liver revealed no significant differences in the incidence or
              degree  of fatty degeneration between controls (22.2%) and animals fed Sb (16.4%).
              This study suggests a LOAEL of 0.5 mg Sb/kg/day based on minimal liver fatty
              degeneration and decreases in weight and longevity (an average mouse weight of 40
              g, and 4 mL water consumption/day containing 5 ppm Sb is assumed).

       •      Schroeder et ai. (1970) administered potassium antimony tartrate (0 or 5 mg Sb/L)
              in drinking water to groups of at least 50 male and 50 female Long-Evans rats from
              the time of weaning until death.  This corresponds to an average daily dose of 0.35
              mg Sb/kg/day, based on authors' calculations.  Mean longevity in days ฑ  standard
              error (SE) was 1,160ฑ27.8 for control males, 1,304ฑ36.0 for control females,
              999ฑ7.8 for treated males, and  1,092ฑ30.0 for treated females.  Antimony had a
              negligible effect on body weight. Serum cholesterol levels were increased in  male
              rats (97.6ฑ4.9 mg/100 mL in dosed  males versus 77.5ฑ2.1 mg/100 mL in controls)
              and decreased in female rats (97.0ฑ5.6 mg/100 mL in dosed females versus
              116.0ฑ6.0  in controls) when compared to control animals.  Fasting blood glucose
              levels were not significantly different in either males or females, but nonfasting
              blood glucose levels were lower in both males (94.5 ฑ6.2 mg/100 mL in dosed
              males versus 134.4ฑ5.1 mg/100 mL in controls) and females (82.5 ฑ7.0 mg/100
              mL in dosed females versus 114.2ฑ5.4 mg/100 mL in controls).  No significant
              effects of Sb on glucosuria, proteinuria, heart weight or heart/body weight ratio
              were observed.  There was no evidence that Sb induced carcinogenesis.  Deposition
              of Sb in kidney, liver, heart, lung and spleen also was observed (mean range  10.14
              to 17.67 /xg/g, value variation range 1.7 to 60.1 ng/g; no Sb was detected  in control
              samples). Antimony accumulated in the soft tissues with age (from 279 to 1,070
              days); pooled samples  showed a  tendency to increase in concentration (p < 0.05),
              with a correlation coefficient of 0.525.  This study identified a LOAEL of 0.35 mg
              Sb/kg/day based on decreased longevity, and altered blood glucose and serum
              cholesterol levels.

         Reproductive Effects

       •      Belyaeva (1967) investigated the reproductive effects of antimony trioxide in rats
              following repeated inhalation exposures to 250 mg/m1 of SbO, dust over a 2-month
              period. Sterility and fewer offspring were noted in dosed rats when compared to the
              control group.

       •      Hodgson et al. (1927)  injected female rabbits with 7 to 17 10-mg doses  of sodium
              antimony tartrate (2.2  mg Sb/kg) or 9 to 16 50-mg doses of an unknown organic
              antimony compound over 16 to 38 days, and English white mice (male and female)
              with 30 to 39 doses of 10 mg of another unknown organic antimony salt over 60 to
              77 days.  In general, in the female rabbits and mice, contraception, abortion and
              fetal damage (details not specified) occurred; in male mice, the antimony salt did not
              cause sterility.
                                            12

-------
Antimony  Health Advisory                                                         April 1992,
         Developmental Effects

              James et al. (1966) fed antimony potassium taitrate to four yearling ewes at a dose
              level of 2 mg/kg of body weight for 45 days or throughout gestation.  All ewes fed
              Sb gave birth to normal, full-term lambs.  No adverse effects were noted in ewes at
              necropsy.

              Casals (1972) reported the absence of any abnormalities in Wistar rat fetuses whose
              mothers were exposed to the pentavalent antimonial drug RL-712 (antimony dextran
              glycoside) during gestation.  Five im injections of 125 or 250 mg Sb/kg were
              administered between days 8 and 14 of gestation.

         Mutagenicitv

              Kanematsu and Kada (1978) and Kanematsu et al.  (1980) determined that antimony
              trichloride, antimony pentachloride and antimony trioxide were mutagenic in the
              Bacillus subtilis (H17 and M45) rec-assay. An improved rec-assay procedure was
              employed to reveal the DNA damaging capacity of the three  Sb compounds.

              Potassium antimony tartrate and sodium antimony tartrate induced chromosomal
              aberrations in cultured human leukocytes (Paton and Allison, 1972; Hashem and
              Shaw Id, 1976).  Piperazine antimony tartrate and potassium antimony tartrate
              induced chromosomal aberrations in bone marrow cells of rats injected ip (El Nahas
              etal., 1982).

         Carcinogenicitv

              Schroeder et al. (1968), as described previously in the Longer-term Exposure
              section, studied the effect of lifetime exposure to Sb on tumor frequency in mice.
              Tumors (benign and malignant) were found in 34.8% of control animals (no
              explanation was given for the high tumor incidence in controls) and 18.8%  of the
              So-treated animals.  The authors concluded that Sb exposure  had no effect on the
              incidence or type of spontaneous benign or malignant tumors.

              Schroeder et al. (1970), as described previously (in the Longer-term Exposure
              section) studied the effect of lifetime exposure to antimony on tumor frequency in
              rats. No significant effects of Sb exposure on tumor frequency were observed in
              either male or female rats.

              Watt (1983) reported that antimony trioxide induces fibrosis and neoplasms in female
              rats when inhaled at levels close to the Threshold Limit Value (TLV).  Female CDF
              rats and S-l miniature swine were exposed to antimony trioxide dust at 1.6 ฑ1.5 mg/
              mj (as Sb) or 4.2ฑ3.2 mg/m3 (as Sb) for 6 hours/day, 5 days/week, for
              approximately 1 year.  The lungs of exposed animals (rats and swine) were mottled
              and heavier than the lungs of unexposed animals. Primary lung neoplasms were seen
              in rats but not in swine.  Most of the neoplasms were seen in the higher dose group
              and were either scirrhous carcinomas, squamous cell carcinomas or bronchoalveolar
              adenomas. The incidence and/or severity of the  response was related to the
              exposure time and the exposure level.
                                            13

-------
Antimony Health Advisory
                                                                          April 199*.
              Groth et al. (1986) exposed three groups (90 males and 90 females/ group) of albino
              rats via inhalation to SbjO, (mean Time-Weighted Average (TWA)  = 45.0 and 46.0
              mg Sb2Oj/mJ in two chambers), to Sb ore concentrate (mean TWA  = 36.0 and 40.1
              mg Sb ore/m3 in two chambers) or to filtered air.  Histopathological examinations
              revealed the presence of lung neoplasms (squamous cell carcinomas,
              bronchioloalveolar adenomas, bronchioloalveolar carcinomas or scirrhous
              carcinomas) in 27% of the females in the SbjO, group and 25%  of the females in the
              Sb ore concentrate group.  No lung tumors were found in the male rats or control
              females.
V.
QUANTIFICATION OF TOXICQLOGICAL EFFECTS
       Health Advisories (HAs) are based upon the identification of adverse health effects
associated with the most sensitive and meaningful noncarcinogenic end point of toxicity. The
induction of this effect is related to a particular exposure dose over a specified period of time, most
often determined from the results of an experimental animal study.  Traditional risk characterization
methodology for threshold toxicants is applied in HA development.  The general formula is as
follows:
where:
      NOAEL

            or

      LOAEL


          BW


        UF(s)
                 No-Observed-Adverse-Effect Level (the exposure dose in mg/kg/ bw/
                 day)
                 Lowest-Observed-Adverse-Effect Level (the exposure dose in mg/kg bw/
                 day)

                 Assumed body weight of protected individual (10 kg for child or 70 kg
                 for adult)

                 Uncertainty factors, based upon quality and nature of data (10, 100,
                 1,000, or 10,000 in accordance with EPA or NAS/OW guidelines)
     	L/day      =   Assumed water consumption (1 L/day for child or 2 L/day for adult)

       One-day Health Advisory

       No information was found in the available literature that was suitable for determination of a
One-day Health Advisory (HA) for Sb.  Accordingly, it is recommended that the Drinking Water
Equivalent Level (DWEL) (10 /tg/L, calculated below) for a 10-kg child be used at this time as a
conservative estimate of the One-day HA.
                                            14

-------
Antimony Health Advisory                                                          April 1992.
       Ten-dav Health Advisory

       The study by Pribyl (1927) was evaluated as the basis for calculation of the Ten-day HA for
a 10- kg child. However, this study appeared dated and offered very little study details upon which
a confident analysis could be made. In the absence of appropriate data, it is recommended that the
DWEL of 10  jig/L be used as a conservative estimate for a 10-day exposure in children.
       Longer-term Health Advisory

       No information was found in the available literature that was suitable for determination of
the Longer-term HA value for Sb.  It is, therefore, recommended that the DWEL (0.01 mg/L,
calculated below) for a 70-kg adult be used as a conservative estimate for a longer-term exposure.
Since the Reference Dose (RfD) and DWEL were based on a lifetime study in rodents and a large
safety factor (1,000) was incorporated into their derivation, it can be assumed that the DWEL will
more than adequately protect both adult and children over longer-term exposure.

       Lifetime Health Advisory

       The Lifetime HA represents that portion  of an individual's total exposure that is attributed to
drinking water and is considered protective of noncarcinogenic adverse health effects over a lifetime
exposure.  The Lifetime HA is derived in a three-step process.  Step 1  determines the RfD,
formerly called the Acceptable Daily Intake (ADI). The RfD is an estimate (with uncertainty
spanning perhaps an order of magnitude) of a daily exposure to the  human population (including
sensitive subgroups) that is likely to be without appreciable risk of deleterious health effects during
a lifetime, and is derived from the NOAEL (or LOAEL),  identified from a chronic (or subchronic)
study, divided by an uncertainty factor(s).  From the RfD, a Drinking Water Equivalent Level
(DWEL) can be determined (Step 2).  A DWEL is a medium-specific (i.e., drinking water) lifetime
exposure level, assuming 100% exposure from that medium, at which adverse, noncarcinogenic
health effects would not be expected to occur.

       The DWEL  is derived from the multiplication of the RfD by the assumed body weight of an
adult and divided by the assumed daily water consumption of an  adult.  The Lifetime HA in
drinking water alone is determined in Step  3 by factoring  in other sources of exposure, the relative
source contribution (RSC).  The RSC from drinking water is based on actual exposure data or, if
data are not available, a value of 20%  is assumed.

       If the contaminant is  classified  as a known, probable, or possible carcinogen, according to
the Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986), then caution must
be exercised in making a decision on -how to deal with possible lifetime exposure to this substance.
For human (A) or probable (B) human carcinogens, a Lifetime HA is not recommended. For
possible (C) human  carcinogens, an additional 10-fold safety factor is used in the calculation of the
Lifetime HA.  The risk manager must balance this assessment of carcinogenic potential and the
quality of the data against the likelihood of occurrence and significance of health effects related to
noncarcinogenic endpoints of toxic ity.  To  assist the risk manager in this process, drinking  water
concentrations associated with estimated excess lifetime cancer risks over the range of 1 in  10,000
to 1 in 1 ,000,000 for the 70-kg adult drinking 2  L of water/day are  provided in the Evaluation of
Carcinogenic Potential section.
                                             15

-------
(ฐ'35        kg/day) = ฐ-00035  mg S^S/day  (rounded to 0.0004  mg Sb/kg/day)
Antimony Health Advisory                                                         April  1992.
       The study by Schroeder et al. (1970) has been selected to serve as the basis for calculation
of the Lifetime HA for Sb because it involved lifetime exposure of rats to potassium antimony
tartrate (the most toxic of the common antimony compounds) given in drinking water.  This study
identified a LOAEL of 0.35 mg/kg/day on the basis of decreased longevity and altered blood levels
of glucose and cholesterol.

       Using a LOAEL of 0.35 mg Sb/kg/day, the Lifetime HA is calculated as follows:

       Step  1: Determination of RfD
   RfD
where:

    0.35 mg Sb/kg/day     =   LOAEL, based on decreased longevity and altered blood glucose
                               and cholesterol in rats exposed to potassium antimony tartrate in
                               drinking water for a lifetime (Schroeder et al., 1970).

                1 ,000     =   uncertainty factor. This uncertainty factor was chosen in
                               accordance with EPA or NAS/OW guidelines for use with a
                               LOAEL from  an animal study.

Step 2:  Determination of Drinking Water Equivalent Level (DWEL)

     DWEL =  (0.0004  mg Sb/kg/day) (70 kg)  = Q Q14 fflg $b/L  (rQunded ^ {Q ^
  a                       (2  L/day)

where:

  0.0004 mg Sb/kg/day     =   RfD.

              2 L/day     =   assumed water consumption of 70-kg adult.

Step 3:  Determination of the Lifetime Health Advisory

           Lifetime HA - (0.014  mg Sb/L) (20%) • 0.0028 mg Sb/L  (3  /xg Sb/L)

where:

        0.014 mg Sb/L     =   DWEL

                 20%     =   assumed relative source contribution from water.
                                            16

-------
Antimony Health Advisory                                                         April 1992,
Evaluation of Carcinogenic Potential

•      No evidence of carcinogenicity was found in two lifetime studies ia which mice and rats
       were supplied with drinking water containing 0 or 5 ppm Sb (as potassium antimony tartrate)
       from the time of weaning until death (Schroeder et al., 1968, 1970).

•      Primary lung neoplasia was observed in rats exposed via inhalation to antimony trioxide dust
       at  1.6ฑ1.5 mg/m3  (as Sb) or 4.2ฑ3.2  mg/m3 (as Sb) for 6 hours/day, 5 days/week for
       approximately 1 year (Watt, 1983). Since no systemic neoplasia was evident and no
       absorption data were presented, these data cannot be utilized in assessing the potential
       carcinogenicity of a soluble form of Sb in drinking water.

•      Groth et al. (1986) exposed rats to antimony ore and antimony trioxide via inhalation.
       Histopathological examinations revealed the presence of lung neoplasms (squamous cell
       carcinomas, bronchioloalveolar adenomas, bronchioloalveolar carcinomas or scirrous
       carcinomas) in 27% of females dosed with SbjOs and 25% of the females in the Sb ore
       category.  No lung rumors were found  in male rats or control females.

•      Applying the criteria described in EPA's guidelines for assessment of carcinogenic risk
       (U.S. EPA, 1986), Sb may be classified in Group D:  not classifiable.  This group is for
       substances with inadequate human  and animal evidence of carcinogenicity.
VI.    OTHER CRITERIA. GUIDANCE AND STANDARDS

       •      The American Conference of Governmental Industrial Hygienists recommends a
              TWA of 0.5 mg Sb/m3 (HSDB, 1988).
VII.    ANALYTICAL METHODS

       Methods for metal analysis involve spectroscopy, either emission or absorption.  In all of the
       methods the metal is dissolved and thermally excited. All elements when excited emit or
       absorb light frequencies characteristic of that element.  Most metal spectroscopy is done in
       the ultra-violet and x-ray regions.

       •     Direct Aspiration Atomic Absorption Spectroscopy (AA).  In this technique, the
             dissolved metals are aspirated into a flame source, and excited to the point that the
             metals are dispersed to a mono-atomic state, a light source whose cathode is the
             metal of interest passes through the flame, the resulting absorption of light by  the
             element of interest is directly proportional to concentration. The disadvantages of
             this technique is the one at a time determination  of the metals and the insensitivity of
             the technique.  This is EPA Method 204. 1 with a detection limit of 50
             Graphite Furnace Atomic Absorption (GFAA). In this technique a specific amount
             of liquid is dried on the thermal source, effecting a concentration step.  The sample
             is electro-thermally excited.  This technique has great sensitivity. This is EPA
             Method 204.2 with a detection limit of 0.8 /*g/L.
                                             17

-------
Antimony Health Advisory                                                         April 199iL
              Inductively Coupled Plasma Atomic Emission Spectroscopy.  This method uses
              aspiration of a liquid sample, but the flame is actually a plasma torch of Argon
              excited to super hot levels by radio-frequency (RF) radiation.  Here the metals are
              excited to the levels where they emit radiation.  By using classical dispersion grating
              optics a large number of metals can be determined simultaneously.  This is EPA
              Method 200.7 with a detection limit of 8 /ig/L.

              Inductively Coupled Plasma Mass Spectrometry  (TCP/MS).  In this  case the
              excitation is again by an RF plasma, but the excited atoms are then interfaced into a
              mass spectrometer. Quantitation is achieved by computerized software programs
              similar to those in other EPA MS organic methods. This is EPA Method 6020 with
              detection limit of 0.02 /xg/L.
VIII.   TREATMENT TECHNOLOGIES

       •     The available literature indicates that conventional coagulation/ filtration will possibly
             remove Sb from contaminated drinking water. If granular activated.carbon (GAC)
             adsorption is added as a post-treatment, then the removal efficiency will be
             improved.

       •     Hannah et al. (1977) tested the effectiveness of a pilot plant utilizing coagulation/
             filtration or excess lime treatment in removing Sb.  The plant consisted of a rapid
             mix designed for a capacity of 4 ppm, a flocculator, a sedimentation basin, and dual-
             media filtration. Antimony was present in the influent at a concentration of 0.6 mg/
             L.  Hydrated lime was added at a dose of 415 mg/L and a pH of 11.5; ferric
             chloride was added at a dose of 40 mg/L and a pH of 6.2; and alum was added at a
             dose of 220 mg/L and a pH of 6.4. Excess lime treatment produced an Sb reduction
             of 28 percent. The ferric chloride coagulation produced an antimony reduction of 65
             percent, and the alum coagulation produced  an Sb reduction of 62 percent.

       •     Hannah et al. (1977) also reported the results of using GAC adsorption as a post-
             treatment to the conventional coagulation/filtration mentioned above, following dual-
             media filtration. Using the above So-containing  effluents from the coagulation/
             filtration process, two GAC columns, operated in parallel and designated as  "old"
             and "new," were tested.  The "old" GAC had been in use for several months before
             this evaluation was made.  When the lime coagulation effluent was processed,
             through the "old" GAC column  and additional 36% of the antimony was removed for
             a total of 64%, while the "new" GAC column removed an additional 24% for a total
             of 52%. When the ferric chloride coagulation effluent was processed through the
             "old" GAC column an  additional 7% of the  antimony was removed for a total of
             72%, and the "new" column removed no additional antimony. When the alum
             coagulation effluent was processed, through  the "old" GAC column an additional
             13% of the antimony was removed for a total of 75%, while the "new" GAC column
             removed an additional'9%, for a total of 71%.
                                            18

-------
Antimony Health Advisory                                                         April 1992.
IX.  REFERENCES

ACGIH.  1980.  American Conference of Governmental and Industrial Hygienists.  Documentation
    of the threshold limit values and biological exposure indices, 4th ed.  and Suppl., p. 20.

Andreae, M.O., J. Asmode and L. Van't Dack.  1981.  Determination of antimony (III), antimony
    (V)  and methylantimony species in natural waters by atomic absorption spectrometry with
    hydride generation. Anal. Chem. 53:1766-71.

Arrington, L.R.  1972. Introductory laboratory animal science.  The breeding, care and
    management of experimental animals.  Danville, IL: Interstate Printers and Publishers, Inc., pp.
    9-11.

Belyaeva, A. P. 1967.  The effect of antimony on reproductive function.  Gig. Tr. Prof.  Zabol.
Bertine, K.K. and E.D. Goldberg.  1972.  Trace elements in clams, mussels, and shrimp.   Limnol.
   Oceanogr.  17:877-884.

Bombard, E., E. Loser, A. Dornemann and B. Schilde.  1982.  Subchronic oral toxicity and
   analytical studies on nickel rutile yellow and chrome rutile yellow with rats.  Toxicol. Lett.
   14:189-194.

Bradley,  W.R. and W.G.  Fredrick.  1941.  The toxicity of antimony.  Ind. Med., Ind. Hyg. Sect.
   2:15-22.

Brannon, J.M. and W.H.  Patrick, Jr.  1985. Fixation and mobilization of antimony in sediments.
   Environ. Pollut. (series B) 9:107-126.

Brieger H., C.W. Semisch, J. Stasney and D.A. Piatnek.  1954.  Industrial antimony poisoning.
   Ind. Med. Surg. 23:521-523.    •

Byrd, J.T. and M.O. Andreae.  1982.  Distribution of arsenic and antimony in species in the Baltic
   Sea.  Abstract. Trans. Am. Geophys.  63:71.

Callahan, M.A., M.W. Slimak and N.W. Gable.  1979.  Water-related environmental fate of 129
   priority pollutants, Vol. 1. Final report. EPA contract nos. 68-01-3852 and 68-01-3867, pp. 5-1
   to 5-8. EPA-440/14-79-029a & b.

Casals, J.B. 1972. Pharmacokinetic and lexicological studies of antimony dextran glycoside (RL-
   712).  Br. J.  Phannacol.  46:281-288.

Chulay, J.D., H.C. Spencer and M. Mugambi.  1985. Electrocardiographic changes during
   treatment of leishmaniasis with pentavalent antimony (sodium stibogluconate). Am. J. Trop.
   Med. Hyg.  34:702-709.

Christopherson, J.B.  1921.  Further notes on the intravenous injection of antimony tartrate.  Lancet
   i: 522-525.
                                            19

-------
Antimony Health Advisory                                                         April
Crecelius, E.A., M.H. Bothner and R. Carpenter.  1975.  Geochemistries of arsenic, antimony,
    mercury, and related elements in sediments of Puget Sound.  Environ. Sci. Technol.  9:325-
    333.

Cyr, F., M.C. Mehra and V.N. Mallet.  1987.  Leaching of chemical contaminants from a
    municipal land  fill site.  Bull. Environ. Contam. Toxicol.  38:775-782.

Doll, R.  1985.  Occupational cancer:  a hazard for epidemiologists.  Int. J. Epidemiol.  14:22-31.

El Nahas, S., S.A. Temtamy and H.A. deHondt.  1982. Cytogenetic effects of two antimonial
    antibilharzial drugs:  tartar emetic and bilharcid. Environ. Mutagen.  4:83-91.

Ercoli, N.  1968. Chemotherapeutic and lexicological properties of antimonyl tartrate-
    dimethylcysteine chelates (33304).  Proc. Soc. Exp. Biol. Med. 129:284-290.

Felicetti, S.A., R.G.  Thomas and R.O. McClellan.  1974.  Metabolism of two valence states of
    inhaled antimony  in hamsters.  Am. Ind. Hyg. Assoc. J.  35:292-300.

Flury, F.  1927.  Zur toxikologie des antimons.  Arch. Exp. Path. Pharmakol.  126:87-103.

Gerber, G.B., J.  Maes and B. Eykens.  1982.  Transfer of antimony  and arsenic to the developing
    organism.  Arch.  Toxicol. 49:159-168.

Gerhardsson, D.  1983.  Antimony in lung, liver, and kidney tissue from deceased smelter workers.
    Scand. J. Work Environ. Health.  8:201-208.

Ghaleb, H.A., H.A. Shoeb, N. El-Gawhary, A.W. El-Borolossy,  S.A. El-Halawany and M.K.
    Madkour.  1979.   Acute toxicity studies of some new organic trivalent antimonials. J. Egypt.
    Med. Assoc.  62:45-62.

Girgis, G. R., P. Scon, A.R. Schulert and H.G. Browne.  1965.'  Acute tolerance of mice to tartar
    emetic.  Toxicol.  Appl. Pharmacol. 7:727-731.

Goodwin, L.G. and I.E. Page. 1943.  A study of the excretion of organic antimonials using a
    polargraphic  procedure.  Biochem. J. 37:198-209.

Groth, D.H., L.E.  Stettler, J.R. Burg, W.M. Busey,  G.C. Grant and L. Wong.  1986.
    Carcinogenic effects of antimony trioxide and antimony ore concentrate in rats.  J. Toxicol.
    Environ. Health  18:607-626.

Hannah, S.A., M. Jelus and J.M. Cohen.  1977.  Removal of uncommon trace metals by physical
    and chemical treatment processes.  Journal WPCF.  49(11):2297-2309.

Hashem, N. and  R. Shawki.  1976. Cultured peripheral lymphocytes: one biologic indicator of
    potential drug hazard. Afr. J.  Med. Sci. 5:155-163.

Hodgson, E.G., A.C.  Vardon and Z. Singh.  1927.  Studies of the effects of antimony salts on
    conception and  pregnancy in animals. Indian J. Med.  Res.  15:491-495.
                                            20

-------
Antimony Health Advisory                                                        April  199.2,
HSDB.  1988.  Hazardous Substances Databank.

James, L.F., V.A. Lazar and W. Binns.  1966.  Effects of sublethal doses of certain minerals on
    pregnant ewes and fetal development.  Am. J. Vet. Res.  27:132-135.

Jolliffe, D.S.  1985.  Nephrotoxicity of pentavalent antimonials.  Lancet i:584.

Kanematsu, K. and T. Kada.  1978.  Mutagenicity of metal compounds.  Mutat. Res.  53:207-208.

Kanematsu, K., M. Hara and T. Kada.  1980.  Rec assay and mutagenicity studies on metal
    compounds.  Mutat. Res.  77:109-116.

Kaplan, E. and F.A. Korff. 1937.  Antimony in food poisoning.  Food Res.  1:529-536.

Leatherland, T.M., J.D.  Burton, F. Guilds, M.J. McCartney and R.J. Morris. 1973.
    Concentrations of some trace metals in pelagic organisms and of mercury in northeast Atlantic
    Ocean water.  Deep-Sea Res. 20:679-785.

Leffler, P. and S. Nordstroem. 1984.  Metals in maternal and fetal blood. Investigation of possible
    variations of the placental barrier function.  Statens Vattenfallsverk,  Vaellingby (Sweden).
    Projekt KOL-HAELSA-MILJOE.  KHM-TR-71.  DE83751427.  Government reports,
    announcements, and indexes, issue 9.

Lippincott, S.W., L.D. Ellerbrook, R*. Rhees and P. Mason.  1947.  A study  of the distribution and
    fate of antimony when used as tartar emetic and fuadin in the treatment of American soldiers
    with schistosomiasis japonica. J. Clin. Invest.  26:370-378.

Lyalikova, N.N.  1974.  Stibiobacter senarmontii, a new microorganism oxidizing antimony.
    Abstract.  Mikrobiologiya 43:941-948.

Miller, J.M.  1982.  Poisoning by antimony: a case report. South. Med. J. 75:592.

Molokhia, M.M. and H. Smith.  1969. The behaviour of antimony in blood. J. Trap. Med. Hyg.
    72:222-225.

Moskalev, Y.I.  1959. Materials on the distribution of radioactive antimony. Med. Radiol. 4(3):6-
    13.

Oliver, T.  1933.  The health of antimony oxide workers. Br. Med. J. 1,094-1,095.

Otto, G.F. and T.H. Maren.  1950.  VI.  Studies on the excretion and concentration of antimony in
    blood and other tissues following the injection of trivalent and pentavalent  antimonials into
    experimental animals. Am. J. Hyg. 51:370-385.

Otto.G.F., T.H. Maren and H.W. Brown.  1947. Blood levels and excretion  rates of antimony in
    persons receiving trivalent and pentavalent antimonials.  Am. J. Hyg. 46:193-211.

Papakkostides, G., A.P.  Grimmanis, D. Zarifiropoulus, G.B. Griggs and T.S. Hopkins.  1975.
    Heavy metals in sediments from Athens sewage outfall area. Mar. Pollut. Bull.  6:136-9.


                                            21

-------
Antimony Health Advisory                                                         April
Parris, G.E. and F.E. Brinkman.  1975.  Reactions which relate to the environmental mobility of
    arsenic and antimony.  I.  Quarternization of trimethylarsine and trimethylstibine. J. Org.
    Chem.  40:3,801-3,803.

Parris, G.E. and F.E. Brinkman.  1976.  Reactions which relate to the environmental mobility of
    arsenic and antimony.  II. Oxidation of trimethylarsine and trimethylstibine.  Environ  Sci
    Technol. 10:1,128-1,134.

Paton, G.R. and A.C. Allison.  1972.  Chromosome damage in human cell cultures  induced by
    metal salts. Mutat. Res.   16:332-336.

Potkonjak,  V.  and M. Pavlovich.  1983.  Antimoniosis:  a particular form of pneumoconiosis. I.
    Etiology, clinical and x-ray  findings.  Int.  Arch. Occup.  Environ. Health.  51:199-207.

Pribyl, E.  1927.  On the nitrogen metabolism  in experimental subacute arsenic and antimony
    poisoning.  J. Biol. Chem. 74:775-781.

Ragaini, R.C., H.R. Ralston and N. Roberts.   1977.  Environmental trace metal contamination in
    Kellog, Idaho near a lead smelting complex.  Environ. Sci.  Technol.  11:773-81.

Rehwoldt, R.,  D. Karimian-Tehenani and H. Altmann.  1975. Measurement and distribution of
    various heavy metals in the Danube River and Danube Canal aquatic communities in the vicinitj
    of Vienna, Austria. Sci. Total Environ. 3:341-348.

Rompp, 1979.  Chemie-Lexikon, Vol. 1.  Stuttgart: Franckh 'sche Verlag Shandlung, p. 237 FF.

Rowland, H.A.K.  1971.  Mathematical methods in research in tropical medicine. Int. Rev. Trop.
    Med.  4:175-238.

Rugemalila, J.B. 1980.  Fatal stibocaptate toxicity.  East Afr. Med. J. 57:720-722.

Schroeder,  E.F., F.A. Rose and H. Mast.  1946.  Effect of antimony on the electrocardiogram.
    Am. J.  Med. Sci.  212:697-706.

Schroeder,  H.A., M. Mitchener, J.J. Balassa, M. Kanisawa and A.P. Nason.  1968. Zirconium,
    niobium, antimony and fluorine in mice: effects on growth, survival and tissue levels.  J. Nutr.
    95:95-101.

Schroeder,  H.A., M. Mitchener and A.P. Nason.  1970.  Zirconium, niobium,  antimony vanadium
    and lead in rats:  life term studies.  J. Nutr. 100:59-68.

U.S. EPA.   1986.  U.S. Environmental Protection Agency. Guidelines for carcinogenic risk
    assessment. Fed. Reg. 51(185):33992-34002. September 24.

Van Bruwaene, R.E., G.B. Gerber, R.  Kirchmann and J. Colard. 1982.  Metabolism of antimony-
    124 in lactating cows.  Health Phys. 43:733-738.

Venugopal, B.  and T.D. Luckey, eds.  1978.  Metal toxicity in mammals,  vol.  2. Chemical
    toxicity of  metals and metalloids. New York, NY:  Plenum Press, p. 213.


                                            22

-------
Antimony Health Advisory                                                         April
Watt, W.D.  1983.  Chronic inhalation toxicity of antimony trioxide: validation of the threshold
    limit value. Environ. Sci. Diss. Abstr. Int. 44:739.B.

Weast, R.C., M.J. Astle and W.H. Beyer, eds.  1986.  CRC handbook of chemistry and physics,
    67th ed.  Boca Raton, FL:  CRC Press, Inc.

Westrick, M.L.  1953.  Physiologic responses attending administration of antimony, alone or with
    simultaneous injections of thyroxin.  Proc. Soc. Exp. Biol. Med.  82:56-60.

Windholz, M., S. Budavari, R.F. Blumetti and E.S. Otterbein, eds.  1983. The Merck index—An
    encyclopedia of chemicals, drugs, and biologicals, 10th ed.  Rahway, NJ:  Merck and
    Company, Inc., p. 715.
                                            23

-------
EPA  0553

HX000027511                                                                     APnl-



                                        BERYLLIUM

                                Drinking Water Health Advisory
                                        Office of Water
                             U.S. Environmental Protection Agency


   I.  INTRODUCTION

          The Health Advisory Program, sponsored by the Office of Water (OW), provides
   information on the health effects, analytical methodology and treatment technology that would
   be useful in dealing with the contamination of drinking water. Health Advisories describe
   nonregulatory concentrations of drinking water contaminants at which adverse health effects
   would not be anticipated to occur over specific exposure durations.  Health Advisories contain a
   margin of safety to protect sensitive members of the population.

          Health Advisories serve as informal technical guidance to assist Federal, State and local
   officials responsible for protecting public health when emergency spills or contamination
   situations occur.  They are not  to be construed as legally enforceable Federal standards. The
   HAs are subject to change as new information becomes available.

          Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
   years,  or 10% of an individual's lifetime) and Lifetime exposures based on data describing
   noncarcinogenic end points of toxicity. For those substances that are known or probable human
   carcinogens, according to the Agency classification scheme (Group A or B), Lifetime HAs are
   not  recommended. The chemical concentration values for Group A or B carcinogens are
   correlated with carcinogenic risk estimates by employing a cancer potency (unit risk) value
   together with  assumptions for lifetime exposure and the consumption of drinking water.  The
   cancer unit risk is usually derived from the linear multistage model with 95% upper confidence
   limits.  This provides a low-dose estimate of cancer risk to humans that is considered unlikely to
   pose a carcinogenic risk in excess of the stated values. Excess cancer risk estimates may also be
   calculated using the One-hit, Weibull, Logit or Probit models. There is no current understand-
   ing  of the biological mechanisms involved in cancer to suggest that any one of these models is
   able to predict risk more accurately than another. Because each model is based on differing
   assumptions, the estimates that are  derived can  differ by several orders of magnitude.

-------
Beryllium                                                                     April. 1992.
II.  GENERAL INFORMATION AND PROPERTIES

        Information is available on a variety of beryllium compounds that have a broad range of
solubilities in water.  In the development of Health Advisories (HAs) for beryllium, data on the
more soluble beryllium salts are considered to be the most relevant for two reasons:  first,
insoluble beryllium salts and ores are not considered likely to be present in public drinking
water systems; second, there is currently no evidence to suggest that insoluble beryllium salts or
ores can be absorbed by the gastrointestinal (GI) systems of animals or humans.
CAS Nos.
       •      Beryllium - 7440-41-7
       •      Beryllium chloride - 7787-47-5
       •      Beryllium sulfate - 13510-49-1
       •      Beryllium oxide - 1304-56-9

Structural  Formulas

        Beryllium carbonate:             BeCO3
        Beryllium ortho-phosphate:      Bej(PO4)2
        Beryllium chloride:              BeClj
        Beryllium oxide:                 BeO
        Beryllium sulfate:                BeSO4

Synonyms

        •  Glucinium. Precious forms of beryl:  emerald, aquamarine.

Uses

        •  Beryllium is used in high-performance products in metallurgical, aerospace and
           nuclear technologies because of its unique combination of properties, such as an
           unusually high melting  point, high modulus of elasticity, extreme hardness, low
           coefficient of thermal expansion and a high stifmess-to-weight ratio. Also, because
           beryllium has a low atomic weight, it  is highly permeable to X-rays, and thin sheets
           are commonly used as windows for X-ray tubes (U.S. EPA, 1988).

-------
Beryllium
                               April. 199Z
Properties (Windholz et al., 1976; Weast et al., 1986)

        •  The properties of inorganic beryllium compounds vary with the specific compound;
           some  examples  follow:
                                   Beryllium
                                   Sulfate
Beryllium
Chloride
                                                            Beryllium
                                                            Oxide
BeSO4
105.07
Crystals
550-600ฐC
2.443
BeCl2
79.93
Whitish
crystals
482.3 ฐC
399.2ฐC
1.90
BeO
25.01
Light
amorphous
powder
2,530ฐC
Beryllium
Carbonate

BeCO,
69
White
powder
                                               Very
                                               soluble
            0.2 mg/L (30ฐC)
  Chemical Formula
  Molecular Weight
  Physical State (at 25ฐC)
  Boiling Point (at 25 mm Hg)
  Melting Point
  Density
  Vapor Pressure
  Water Solubility

  Octanol/Water Partition
    Coefficient (log K^)
  Taste Threshold (water)
  Odor Threshold (water)
  Odor Threshold (air)
  Conversion Factor

Occurrence
        •  Beryllium is a naturally occurring element found in the earth's crust at an average
           concentration of 2.5 pprn.  Beryllium is found chiefly as the minerals beryl
           (Be,Al2Si6Ol8), bromellite (BeO), chrysoberyl (BeAljO4) and beryilonite (NaBePO4)
           (Weast et al., 1986).

        •  Many common beryllium compounds (for example, the chloride and nitrate) are
           readily soluble in water. Others, such as the sulfate complex, are only moderately
           soluble, and the carbonate and hydroxide compounds are almost insoluble in cold
           water.  Ionic beryllium is not likely to be found in natural waters except in trace
           amounts, because, in the normal pH range of these waters, the oxides  and
           hydroxides formed at pH 5-8 are relatively insoluble (U.S. EPA, 1988).

        •  In general, beryllium concentrations are well below 1 jtg/L in surface,  ground, and
           rain waters (Callahan  et al., 1979),
* Described as insoluble in cold water and decomposing in hot water (Weast et al., 1986).

-------
Beryllium                                                                       April, 1992.



Environmental Fate

        •   There are currently no data available on the environmental fate of beryllium.

        •   Callahan et al. (1979) have predicted that because beryllium salts commonly found
            in natural waters have very low aqueous solubilities, they are probably precipitated
            or adsorbed onto solids soon after introduction to the aqueous environment.

            Complexing agents may solubilize beryllium into its ionic form, but ambient water
            quality data suggest that concentrations of this element in heavily polluted waters
            are quite low.  Thus, beryllium in natural water systems is found predominantly in
            paniculate rather than dissolved form.


III. PHARMACOKINETICS

Absorption

        •   The available absorption data indicate  that the uptake of orally administered
            beryllium sulfate is extremely poor. Water solubility data on this compound
            indicate that it is relatively insoluble at lower water  temperatures, but decomposes
            to BeSO4.4HjO at higher water temperatures. Thus, unless stomach  acids enhance
            dissolution of beryllium sulfate, it is unlikely that a significant concentration of
            beryllium ion would reach the GI tract.

        •   Single oral radiolabeled doses of beryllium  were administered as a single dose to
            mice, rats, monkeys and dogs (Furchner et al., 1973).  Based on a weighted average
            of the 2-day cumulative urinary excretion data, absorption of beryllium was
            estimated to be 0.6%.

        •   In studies of guinea pigs fed 10 or 30 rag/day of beryllium sulfate (approximately
            13.3 or 40 mg/kg/day, based on the assumptions of Lehman,  1959), the amount of
            beryllium absorbed was 0.006% of that ingested (approximately 0.08 or 0.24
            mg/kg/day) (Hyslop et al., 1943).

        •   Reeves (1965) added 6.6 or 66.6 /ig/day of beryllium as beryllium sulfate to the
            drinking water of male Sprague-Dawley rats for 24 weeks. Of the total amount of
            beryllium administered during the study, more than 99%  of that recovered was
            found in the feces. Total body uptake  was  estimated to be less than 1% for both
            groups, with most of this found in the skeleton at levels that were independent of
            dose.  The author speculated that most of the beryllium in intestinal Quid was
            probably precipitated as a phosphate, which would account for the observed low
            levels of absorption and  uptake observed.

        •   Exposure of 160- to 180-gram Fischer 344 rats to  447 /*g/raj of beryllium oxide at a
            concentration of 447 /ig/raj for 1 hour resulted in  the incorporation of 0.2 /xg of

-------
Beryllium                                                                       April, 199/L
            beryllium into the lung tissue (Hart et al., 1984).  Assuming a respiration rate of
            0.0057 m3/hour, the percent uptake by inhalation  equals 0.27(0.0057 m3 x 447 /ig/ra3),
            or 7.8%.
Distribution
        •   In studies of orally administered beryllium compounds, the organs that retained
            significant amounts of beryllium were the skeleton, liver and kidney.  When Reeves
            (1965) administered beryllium sulfate in the drinking water of rats for 24 weeks,
            beryllium levels were highest in the GI tract and the skeleton, with somewhat lower
            levels in the blood and the liver.

        •   In vitro studies have indicated that when beryllium sulfate solutions are incubated
            with artificial or natural human serum, beryllium orthophosphate and hydroxide
            precipitates are found in these  fluids (Reeves and  Vorwald, 1961).
Metabolism
Excretion
            No information was found in the available literature on the metaboh'sm of
            beryllium.
            In rats, the route of administration influences the route of excretion of beryllium.
            Fecal excretion is the major route if beryllium is administered orally or via
            inhalation.  Urinary excretion is the major route following intravenous
            administration.

            Approximately 80% and 76%, respectively, of a 6.6 or 66.6 /xg/day oral dose of
            beryllium (administered via drinking water as beryllium sulfate) was recovered in
            the feces of rats during a 24-week study,  this was more than 99% of the total
            recovered dose.  Less than 1% was recovered from the urine at either dose
            (Reeves, 1965).

            Rhoads and Sanders (1985) showed that nearly all of the beryllium  cleared from the
            lungs of rats administered beryllium oxide via inhalation was excreted in the feces.
            Urinary excretion was the major route following parenteral administration of
            beryllium as beryllium chloride (very water soluble compound) in rats (Furchner et
            al., 1973).

-------
Beryllium                                                                      April, 1992.



IV.     HEALTH EFFECTS

Humans

  Short-term Exposure

        •  No information was found in the available literature on the human health effects of
           short-terra oral exposure to beryllium compounds.

  Long-term Exposure

        •  No information was found in the available literature on the human health effects of
           longer-term  oral exposure  to beryllium compounds.

  Carcinoeenicitv

        •  Evidence of the carcinogenic'potential of beryllium and its compounds in humans is
           considered to be limited, whereas, in animals there is sufficient evidence of
           carcinogenicity (IARC, 1987).

Animals

  Short-term Exposure

        •  In general, beryllium compounds administered orally produce less acute toxicity in
           animals than those administered by other routes,  probably because beryllium salts
           are poorly absorbed from the GI tract (Schroeder and Mitchener, 1975b).

        •  One of the earliest observed effects of short-term exposure to beryllium was the
           development of rickets in young rats fed diets containing beryllium carbonate
           (Guyatt et al., 1933; Businco, 1940).

        •  Guyatt et al. (1933) reported that 21- to 24-day-old rats fed diets containing
           beryllium carbonate at five dosage levels from 0.125% to 2.0% developed rickets
           after 3 weeks.  The effects were dose dependent,  with the lowest dose (1.25 g/kg of
           BeCO, diet, 163 rag/kg of Be diet or approximately 163 rag/kg/day of Be) resulting
           in a mild case of rickets, while higher doses (2.0% or approximately 261 rng/kg/day
           of Be) resulted in almost a complete lack of calcification of the long bones (the
           femur and the tibia).

        •  Businco (1940) conducted  a series of experiments in which young rats (strain not
           specified) were fed beryllium carbonate mixed in  milk at doses of 0.06 g/day on days
           0 to 14; 0.16 g/day on days 15 to 34 and 0.24 g/day on days 35 to 83.  A
           Time-Weighted Average (TWA) dose of 0.19 g/day (about 700 mg/kg/day of Be)
           was estimated by the U.S.  EPA (1988) based on the data provided by the author.
           No effects on body weight or general appearance  were  observed when animals were

-------
Beryllium                                                                       Aprili
            fed 0.06 g/day of beryllium carbonate (about 260 rag/kg/day of Be).  After 40 days
            of exposure, a 25% weight loss was observed in treated rats compared with controls.
            A reduction in weight of greater than 50% was seen in treated animals at the end
            of the study (day 83).  In addition, histologjcal and radiographic examination of the
            long bones (femur, tibia, and fibula) and vertebrae revealed typical rachitic lesions.

  Dermal/Ocular  Effects

        •   No data have been located on the dermal or ocular  effects of exposure to beryllium
            compounds.

  Long-term Exposure

        •   In the limited information available on the toxic effects of long-term oral exposure
            to beryllium sulfate, there is no clear evidence of adverse effects other than slight
            depressions in body weight.

        •   Schroeder and Mitchener (1975a) administered 5 mg/L of beryllium as beryllium
            sulfate in the drinking water of Long-Evans rats (52 per sex) until natural death.
            Based on data provided in the study, this level corresponds to approximately 0.538
            mg/kg/day of Be. Gross and microscopic pathological  changes were evaluated and
            clinical chemistry and urine analyses were performed.  No treatment-related effects
            were  observed in any parameter tested.  There was a slight depression in growth of
            male  rats from 2 to 6 months of age.  In a similar study by Schroeder and
            Mitchener (1975b), 54  male and 54 female Charles River (CD) mice were
            administered about 1 mg/kg/day of beryllium as beryllium sulfate in their drinking
            water. A slight decrease in body weight in females 6 to 8 months of age (p <
            0.025, p  < 0.01,  respectively)  and a slight general increase in male body weight were
            noted. No other treatment-related effects were found.

        •   Cox et al. (1975) fed rats 0, 5, 50 or 500 ppm of beryllium sulfate in the diet for 2
            years. Based on the assumptions of Lehman (1959), these dietary levels are
            equivalent to dosage levels of approximately 0, 0.25, 2.5 and 25 mg/kg/day of Be.  A
            slight decrease in body weight in the high-dose group was reported.

  Reproductive Effects

        •   No information was found in the available literature about the reproductive effects
            of oral, dermal or inhalation exposure to beryllium.

  Developmental  Effects

        •   No information was found in the available literature on the developmental effects
            of oral exposure to beryllium.

-------
Beryllium                                                                       April, 1992,
        •  Tsujii and Hoshishima (1979) observed mild neurotoric effects in the offspring of
           pregnant CFW mice administered 140 rag/day of beryllium sulfate (approximately
           0.0047 rng/kg/day of Be) by intraperitoneal injection. The mice were given this dose
           for 3 consecutive days and eight times every other day for a total of 11 days during
           pregnancy.

  Mutagcnicity

        •  Simmon (1979a) reported beryllium sulfate to be negative in rautagenic response in
           Salmonella typhimurium strains  TA1535, TA1536, TA1537, TA98 and TA100, with
           and without S-9 metabolic activation.  Rosenkranz and Poirier (1979) also reported
           negative results in this assay (the Ames test).

        •  Beryllium sulfate was also negative when tested for mitotic recombination activity in
           Saccharomyces  cereviseae D3 (Simmon, 1979b).

        •  Simmon et al. (1979) reported that beryllium sulfate was not rautagenic in the
           host-mediated assay with S. typhimurium strains TA1530, TA1535 and TA1538 and
           Saccharomyces  cerevisiae D3.  Mice were injected intramuscularly with 25  rag/kg
           beryllium sulfate or administered an oral dose of 1,200 rag of beryllium sulfate.
           Four hours after the treatment, microorganisms were recovered from the peritoneal
           cavity and plated for mutant colonies. Mutation frequencies  were not significantly
           increased for either strain.

        •  Larramendy et al. (1981) demonstrated that beryllium sulfate was clastogenic in
           cultured human lymphocytes. This study showed a six-fold increase in aberration
           frequency, which was primarily  due to an increase in DNA breaks.  This study also
           demonstrated the potential of beryllium  sulfate to induce  sister-chromatid
           exchanges (SCEs) in both cultured human lymphocytes and Syrian hamster embryo
           cells.

        •  Kanematsu et al. (1980) found beryllium sulfate to be weakly  rautagenic in the
           recombination  assay.  Inhibition of growth because of DNA damage was observed
           in Bacillus subtilis strains H17 (rec*) and M45 (rec). Similar results were also
           obtained by Kada et al. (1980).

        •  Rosenkranz and Poirier (1979)  reported negative results in the Pol assay, which
           measured the ability of cells deficient in their ability to repair DNA damage to
           grow after exposure to beryllium sulfate (250 /tg). Escherichia coli strains pol A*
           zndpol A* were used in this study, both in the presence and absence of an S-9
           activation system.

        •  Beryllium sulfate was reported  to produce negative results in the DNA repair test
           using rat primary hepatocyte cultures (Williams et al.,  1982) and in the mitotic
           recombination  assay using the yeast Saccharomyces cerevisiae D, (Simmon, 1979b).

-------
Beryllium                                                                       April, 199JL
        •   Kubinski et al. (1981) reported that ionic beryllium induced the formation of DNA
            protein complexes (adducts) when E. coli cells and Ehrlich ascites cells were treated
            with radioactive DNA in the presence of 30 rnM beryllium.

        •   Skilleter et al. (1983), using synchronized rat  liver-derived epithelial cell cultures
            (BL9L), found that beryllium sulfate blocked the cell cycle at the Gv phase and
            caused inhibition of cell division.

        •   Beryllium ion (2+) has been reported to increase the raisincorporation of
            nucleotides during DNA polymerization (Luke et al., 1975). Beryllium chloride
            increased the error frequency in the incorporation of nucleotide bases into DNA
            (Sirover and Loeb, 1976).  Nucleotide base substitutions induced by beryllium
            chloride were found to be two- to three-fold greater than those found
            spontaneously in the lacl system (Zakour and Glickman, 1984).

  Carcinogenicitv

        •   Beryllium compounds administered by injection or inhalation can induce malignant
            tumors in laboratory animals.  The two types  of cancers observed using these routes
            were lung cancer and osteosarcoraa.

        •   Gardner and Heslington (1946) reported induction of osteosarcoma following
            intravenous injections of zinc beryllium oxide into rabbits.  (Zinc oxide, zinc silicate
            and silicic acid were all 'noncarcinogenic by this route.) Since then, numerous other
            studies (e.g., Cloudman et al., 1949; Nash, 1950; Dutra and Largent, 1950; Barnes et
            al., 1950; Hoagland et al., 1950) have demonstrated that many different beryllium
            compounds,  including beryllium metal, are tumorigenic when administered
            intravenously.

        •   Barnes et al. (1950) injected beryllium into the ear veins of 4-kg rabbits  twice
            weekly for 5 weeks. The beryllium was administered as an aqueous suspension of
            particles of 5 fi or less in diameter of zinc beryllium silicate (an ore of questionable
            solubility); the suspension contained a total of 72 mg of beryllium.  The TWA daily
            dose over the 120-week period of the experiment was 0.0021 mg/kg/day of Be.  No
            adjustment was made for a less than lifetime  observation period because other
            studies have indicated that osteosarcomas almost always develop within 2 years of
            exposure (the lifespan of the test animals was 6 years). Four of nine animals
            surviving 32  weeks or longer developed bone  tumors.  Osteosarcomas did not
            develop in animals injected with zinc silicate.

        •   The only study reporting the development of osteosarcoma following inhalation of
            beryllium was that of Dutra et al. (1951).  In  this study, one of six rabbits exposed
            to an aerosol of beryllium oxide at 6 mg/mj for 25 hours/week for 13 months
            developed malignancies,

-------
Beryllium                                                                      April, 1992,
        •  Pulmonary tumors were produced in 18 of 19 rats that survived a 17-month
           exposure to 15 rag/mj of beryl ore (Wagner et ah, 1969).

        •  Reeves et al.  (1967) exposed 150 rats to beryllium sulfate in an aerosol at a
           beryllium concentration of 3425 /ig/m3 for 35 hours/week for a 72-week period.
           The initial response appeared 4 weeks after the first exposure and consisted of
           hyperplasia of the pulmonary epithelium, which progressed to metaplasia, anaplasia
           and lung cancer. The first tumors were discovered after 9 months of exposure, and
           the incidence of tumors was  100% at 13 months. All of the tumors were alveolar
           adenocarcinomas.

        •  Schroeder and Mitchener (1975a,b) reported a slight,  statistically nonsignificant
           increase in the incidence of lymphoma leukernias in female mice and a slightly
           higher but still nonsignificant increase in the incidence of grossly observed tumors
           in male rats given beryllium sulfate at a concentration of 5  mg/L (approximately
           0.538  mg/kg/day) in drinking water over a lifetime.

        •  In an unpublished 2-year study which has  not been peer reviewed, Cox et al. (1975)
           administered  beryllium sulfate at Be dietary levels of 0, 5, 50 and 5dO ppm
           beryllium (as  beryllium sulfate) in the diet to Wistar albino rats (50/sex/group).
           (These levels  are equivalent to approximately 0, 025, 2.5 and 25 rag/kg bw/day
           based on the  dietary assumptions of Lehman, 1959). Mild body-weight depression
          . was observed at the highest dosage level.  Reticulum cell sarcomas in the lung were
           seen in all dose groups and in controls, and similar lesions were seen in lymph
           nodes, bone marrow and abdominal organs.  The incidence of lung reticulum cell
           sarcoma was  higher in males than females, and was statistically significant in males
           at the lowest  two doses but not at the highest dose (tumor incidences were 10, 17,
           16 and 12 at 0, 5, 50 and 500 ppra, respectively). This study is considered to be
           suggestive of a carcinogenic response to ingested beryllium, but the lack of a
           statistically significant response at the highest dose level severely limits its
           interpretation as a positive study.  In addition, a more recent abstract (Morgareidge
           et al., 1977) based on the same study concludes that this dietary regimen had no
           effect on tumorigenesis.
V. QUANTIFICATION OF TOXICOLOGICAL EFFECTS

        Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
(up to 7 years) and Lifetime exposures if adequate data are available that identify a sensitive
noncarcinogenic end point of toxicity. The HAs for noncarcinogenic toxicants are derived using
the following formula:


          (NOAEL or LOAEL^ x (EW) =	rag/L (	
              (UF)(_Uday)
                                           10

-------
 Beryllium                                                                      April, 199Z
where:
        NOAEL or LOAEL      =    No- or Lowest-Observed-Aciverse-Effect Level (in
                                      mg/kg bw/day).

                       BW      =    assumed body weight of a child (10 kg) or an adult
                                      (70 kg).

                       UF      =    uncertainty factor, (10, 100, 1,000 or 10,000, in
                                      accordance with EPA or NAS/OW guidelines.

                  	L/day      =    assumed daily water consumption of a child (1 L/day)
                                      or an adult (2 L/day).
One-day Health Advisory

       Studies containing exposure duration data appropriate for the calculation of a One-day
HA could not be located in the available literature. It is recommended, therefore,  that the
Ten-day HA of 30 mg/L be used as an estimate of exposure for a One-day HA for  a 10-kg child.

Ten-day Health Advisory

       The study by Businco (1940) has been  selected to serve as the basis of the Ten-day HA
for the 10-kg child because  it was conducted for a more appropriate exposure duration than
other available studies using the oral route. In this study, 0.06 g of beryllium carbonate (0.26
g/kg of Be) was fed to 30- to 40-g young rats for 14 days.  This resulted in no adverse effects on
body weight or general appearance.  No other toxicity end points were measured for this time
period or dose; however, animals fed higher doses showed a dose-related decrease  in the rate of
body-weight gain and a decrease in calcification and development of long bones.  A No-
Observed-Adverse-Effect-Level (NOAEL)  for beryllium of 260 mg/kg/day was established based
on the absence of adverse effects on body weight and gross toxicological effects in this study.

       The Ten-day HA for the 10-kg child is calculated as follows:
       Ten-day HA = (260 mg/kg/dav> (10 k$ = 26 nig/L (rounded to 30,000 /tg/L)
                          (100) (1 L/day)
where:
              260 mg/kg/day     =     NOAEL, based on the absence of adverse effects on
                                      body weight gain and bone development in rats fed
                                      beryllium carbonate for 2 weeks (Businco, 1940).

                      10 kg     =     assumed body weight of a child.


                                           11

-------
Beryllium                                                                      April, 1992.
                        100      =    uncertainty factor chosen in accordance with EPA and
                                     NAS/OW guidelines in which a NOAEL from an
                                     animal study is employed.

                    1 L/day      =    assumed daily water consumption of a child.
Longer-term Health Advisory

       The study by Businco (1940) was selected to serve as the basis for the Longer-term HA
because of its exposure duration and because it provides information on an important toxicity
end point for beryllium exposure (i.e., the development of rickets).  In this study, young rats (30
to 40 g) were fed increasing doses of beryllium carbonate for 83 days. Rats received daily doses
of 60 mg of beryllium carbonate (7.8 rag Be) on days 0 to 14, 160 mg (20.8 rag Be) on days 15
to 34 and 240 mg (31.2 rag Be) on days 35 to 83. The estimated average body weight of the
treated animals was 55.5 g.  A TWA dose of 0.191 g/rat/day (3.4 g/kg) of beryllium carbonate
can be estimated based on the information provided by the authors.  The 3.4 g/kg-dose (443 mg
Be) resulted in a greater than 50% reduction in body weight gain and decreased development
and calcification of the  long bones.  Therefore, 443 mg/kg/day of beryllium is considered to be
the  Lowest-Observed-Adverse-Effect Level (LOAEL).  No other adverse effects were reported
in this study.

       No other available study was found to be more appropriate for the calculation of the
Longer-term HA. Therefore, the LOAEL of 443 mg/kg/day from the Businco (1940) study is
recommended for the calculation of the Longer-term  HA.

       The Longer-term HA for a 10-kg  child is calculated as follows:


    Longer-terra HA = C443 me/ke/dav') (10 kg) = 4.4 rag/L (rounded to 4,000 /tg/L)
                         (1,000) (1 L/day)
where:
            443 rag/kg/day      =    LOAEL, based on suppressed body weight gain and
                                    development of rickets in rats fed BeCO, for 83 days
                                    (Businco, 1940).

                    10 kg      =    assumed body weight of a child.

                    1,000      =    uncertainty factor chosen in accordance with EPA and
                                    NAS/OW guidelines in which a LOAEL from an animal
                                    study is employed.

                  1 L/day      =    assumed daily water consumption of a child.
                                           12

-------
 Beryllium                                                                      April, 1992.
       The Longer-term HA for a 70-kg adult is calculated as follows:
        Longer-term HA = C443 mg/kg/day") (70 kg) = 15-5 mg/L (rounded to 20,000
                             (1,000) (2 L/day)
 where:
             443 mg/kg/day      =    LOAEL, based on suppressed body weight gain and
                                     development of rickets in rats fed BeCOj for 83 days
                                     (Businco, 1940).

                     70 kg      =    assumed body weight of an adult.

                     1,000      =    uncertainty factor chosen  in accordance with EPA and
                                     NAS/OW guidelines in which a LOAEL from an animal
                                     study is  employed.

                   2 L/day      =    assumed daily water consumption of an adult.
 Lifetime Health Advisory

       The Lifetime HA represents that portion of an individual's total exposure that is
 attributed to drinking water and is considered protective of noncarcinogenic adverse health
 effects over a lifetime exposure.  The Lifetime  HA is derived in a three-step process. Step 1
 determines  the Reference Dose (RfD), formerly called the Acceptable  Daily Intake (ADI).  The
 RfD is an estimate of a daily exposure level to  the human population that is likely to be without
 appreciable risk of deleterious effects over a lifetime, and is derived from the NOAEL (or
 LOAEL), identified from a chronic (or subchronic) study, divided by an uncertainty  factor(s).
 From the RfD, a Drinking Water Equivalent Level (DWEL) can be determined (Step 2). A
 DWEL is a medium-specific (i.e., drinking water) lifetime exposure level, assuming 100%
 exposure from that medium, at which adverse,  noncarcinogenic health effects would not be
 expected to occur.  The DWEL is derived from the multiplication of the RfD by the assumed
. body weight of an adult and divided by the assumed daily water consumption of an adult. The
 Lifetime HA is determined in Step 3 by factoring in other sources of exposure, the relative
 source contribution (RSC).  The RSC from drinking water is based on  actual exposure data or,
 if data are not available, a value of 20% is assumed.

       If the contaminant is classified as a known, probable or possible human carcinogen,
 according to the Agency's classification scheme of carcinogenic potential  (U.S. EPA, 1986), then
 caution must be exercised in making a decision on how to deal with possible lifetime exposure
 to this substance.  For human (A) or probable  human (B) carcinogens, a Lifetime HA is not
 recommended.  For possible human carcinogens (C), an additional 10-fold safety factor is used
 in the calculation of the Lifetime HA. The risk manager must balance this assessment of
 carcinogenic potential and the quality of the data against the likelihood of occurrence and
                                            13

-------
Beryllium                                                                      April, 199&
significance of health effects related-to noncarcinogenic end points of toxicity. To assist the risk
manager in this process, drinking water concentrations associated with estimated excess cancer
risks over the range of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult drinking  2 L of water per
day are provided in the Evaluation of Carcinogenic Potential section.

       The study by Schroeder and Mitchener (1975a) has been selected to serve as the basis
for the Lifetime HA because currently it is the only available published study using an
appropriate route and exposure duration. In this study, male and female rats were administered
5 ppm of beryllium in their drinking water for a  lifetime.  The only significant effect was a slight
reduction  in body weight in males from 2 to 6 months of age. A NOAEL of 0.538 rng/kg/day
for beryllium was calculated by multiplying the 5-ppra dose (5 rag/L) by an average water
consumption rate of 0.035  L/day for rats and dividing the product by the average rat body
weight (0.325 g) given in this study.  It should be noted, however, that several weaknesses have
been identified in this study, such as the presence of other trace elements and minerals
(including chromium)  in the drinking water, the  use of nonrandornized animals and the
administration of only one dose.

       The unpublished study of Cox et al. (1975) (also reported in an abstract by Morgareidge
et al., 1977) could also be used to calculate the reference dose (RfD) for beryllium.  In this
study, rats were exposed to the 0, 5, 50 or 500 ppm (approximately 0, 025, 2.5 or 25 rag/kg/day)
of beryllium in the diet for 2 years.  The only toxic effect confirmed was a slight decrease in
body weight in the highest dose group.  A NOAEL of 25 mg/kg/day for beryllium can be
identified  based on the results of this study.  Since this study is unpublished and presumably has
not been peer reviewed, it cannot be used in the calculation of a Drinking Water Equivalent
Level (DWEL) for beryllium. In addition, there is some question about the finding of
pulmonary reticulum cell sarcoma at the two lowest doses.  However, until a more detailed
review of the Cox et al. (1975) study is conducted, the Agency-verified RfD  (U.S. EPA, 1988)
based on the Schroeder and Mitchener (1975a) study is recommended for use in calculating the
DWEL.

Step 1: Determination of the RfD
       RfD = (0.538 mg/kg/dav) = 0.005 mg/kg/day
                    (100)
where:
    0.538 mg/kg/day    =   adjusted NOAEL, based on the absence of effects in rats given
                          BeSO4 in drinking water over a lifetime (Schroeder and Mitchener,
                          1975a).

               100    =   uncertainty factor chosen in accordance with EPA and NAS/OW
                          guidelines in which a NOAEL from an animal study is employed.
                                           14

-------
Beryllium                                                                      April, 199jL
Step 2:  Determination of the DWEL


       DWEL = C0.005 me/kg/dav) (70 kg) = 0.175 rng/L (rounded to 0.2 rag/L)*
                      (2 Uday)

where:

   0.005 rag/kg/day    =  RฃD.

             70 kg    =  assumed body weight of an adult.

           2 L/day    =  assumed daily water consumption of an adult.


Step 3:  Determination of the Lifetime Health Advisory

       Beryllium is classified by the U.S. EPA as Group B2: probable human carcinogen.
Therefore, the determination of a Lifetime HA is not  recommended.

       The estimated excess cancer risk associated with lifetime exposure to drinking water
containing 175 /ig/L beryllium is approximately 2.5 x 10"2.  This estimate represents the upper
95% confidence limit from extrapolations prepared by U.S. EPA's Human Health Assessment
Group using the linearized multistage model. The actual  risk is unlikely to exceed this value.

Evaluation of Carcinogenic Potential

        •   IARC (1987) has classified beryllium and beryllium compounds in Group 2A:
            probable human  carcinogen.  This group  includes chemicals for which there is
            limited evidence  of carcinogenicity in humans and  sufficient evidence in animals.

        •   Using the U.S. EPA (1986) Guidelines for Carcinogen Risk Assessment, beryllium
            is classified as a Group B2 carcinogen. This classification is used for compounds in
            which there  is sufficient evidence of carcinogenicity in animal studies and
            inadequate evidence in human studies.

        •   The U.S. EPA (1987) has calculated an inhalation carcinogenic potency factor for
            beryllium of 2.4 x 10"J (/ig/m1)"1 based on  the epidemiological study by Wagoner et
            al. (1980) and the industrial hygiene  reviews  by NIOSH (1972) and Eisenbud and
            Lisson (1983).  These data have been combined arithmetically to estimate a
            plausible upper-bound incremental lifetime cancer risk of 2 x 10"J for exposure to
            air containing 1 pg/m} of beryllium.

        •   Because beryllium administered via inhalation and injection is carcinogenic in
            animals, it may be carcinogenic when administered orally as well.  Thus, oral
    ?This is the DWEL based on the verified RfD; however, as discussed previously, it may be
appropriate to the RfD.

                                           15

-------
Beryllium                                                                       April, 199Z
            exposure to beryllium in drinking water may represent a carcinogenic risk to
            humans.

        •  The U.S. EPA (1988) has derived an oral carcinogenic potency factor for beryllium .
           based on the results of intravenous infusion studies in which osteosarcomas were
           induced in rabbits. Based on these data, a human carcinogenic potency factor of
           1,843 (mg/kg/day)'1 for intravenous infusion of beryllium was identified.  After
           adjusting for an oral  absorption efficiency of 0.6% (Furchner et al. 1973), an oral
           potency factor of 11 (mg/kg/day)'1 is obtained (U.S. EPA, 1988).

        •  An oral carcinogenic potency estimate can be derived by extrapolation or the
           inhalation slope factor based on human data, after adjusting for relative absorption
           efficiency (McGinnis, 1988). Absorption from the GI tract has been reported to be
           approximately 0.6% (Furchner et al. 1973), and absorption  from the lungs was
           approximately 7.8%  (Hart et al. 1984). The relative difference in absorption
           efficiency is 0.6/7.8, or 0.077. Therefore, the extrapolation-based oral carcinogenic
           potency estimate equals 2.4 (rag/ra3)'1 x 70 kg/20rn3/day x 0.077, or 0.65
           (mg/kg/day)'1.

        •  If the geometric mean of the slope factors based on animal inhalation studies (U.S.
           EPA, 1987) is adjusted without surface-area correction for a 7Q-kg person inhaling
           20 raVday, and a relative absorption factor between the inhalation and oral routes
           of 0.77 is applied, an oral carcinogenic potency factor of 12.7 rag/kg/day'1 is derived
           (McGinnis, 1988).

        •  A carcinogenic potency estimate of 43 (rag/kg/day)'1, based on extrapolations from
           the Schroeder and Mitchener (197Sa) drinking water study in rats, has been verified
           by the Agency-wide CRAVE workgroup (U.S. EPA, 1991), despite the lack of a
           significant  tumorigenic  response.  Since no significant response was detected, this
           estimate is an upper-bound value; that is, the risk is not expected to be greater, but
           may be  less, than the derived value. The limited evidence of carcinogenicity  in the
           Cox et al. (1975) study provides further support for this conclusion.  When a
           linearized multistage model is applied to the Schroeder and Mitchener (1975a)
           animal data, the resulting criteria  are 0.8, 0.08 and 0.008 /tg/L, with corresponding
           carcinogenic risk levels of 10"*, 10"5 and 10"*, respectively, for a 70-kg man
           consuming 2 L/day of drinking water.
VI. OTHER CRITERIA. GUIDANCE AND STANDARDS

        •  The World Health Organization (WHO) has not established a guideline for
           drinking water quality for beryllium (IRPTC, 1987).

        •  National regulation by OSHA (1985) established a Permissible Exposure Limit
           (PEL) 8-hour TWA of 2 /xg/rn1, an acceptable ceiling limit of 5 pg/m1 and an
           acceptable maximum peak above ceiling of 25 jtg/raj for 30 minutes.


                                            IS

-------
Beryllium                                                                      April, 1993,
           Advisories issued by various agencies for beryllium in the air are as follows:
           NIOSH (1972) recommended an occupational exposure limit of 0.5 /zg/ra3; ACGIH
           (1987) recommended a TWA-Threshold Limit Value (TWA-TLV) of 0.002 mg/m3.

           The U.S. EPA (1980) established ambient water quality criteria of 68 ng/L for the
           consumption of 2 L of ambient water and fish, and 1,170 ng/L for the consumption
           of aquatic organisms, to correspond with a risk level of 10"s.
VII.  ANALYTICAL METHODS

        •  Methods available to analyze beryllium are summarized below. The bulk of the
           methods available for metal analysis involve spectroscopy, either emission or
           absorption.  In all of these  methods, the metal is dissolved and thermally excited.
           When excited, all elements  emit or absorb light frequencies characteristic of that
           element. Most metal spectroscopy is done in the ultraviolet and x-ray regions.

        •  Direct Aspiration Atomic Absorption Spectroscopv CAA).  In this technique (EPA
           Method 210.1), dissolved metals are aspirated into a flame source and excited to
           the point of dispersion into a mono-atomic state; a light source whose cathode is
           the metal of interest passes through the flame, and the resulting absorption of light
           by the element of interest is directly proportional to its  concentration.
           Disadvantages of this technique include the inability to analyze more than one
           metal at a time and the insensitiviry of the technique. The detection limit is 0.25
           Graphite Furnace Atomic Absorption (GFAA). This technique differs from AA, in
           that a specific amount of liquid is dried on the thermal source, effecting a
           concentration step.  The sample is then  electrothermally excited. This technique is
           very sensitive, but is still a tedious one-raetal-at-a-time determination.

           Inductively Coupled Plasma Atomic Emission Spectroscopv. This method utilizes
           aspiration of a liquid sample, but  the flame is actually a plasma torch of Argon
           excited to super hot levels by radio-frequency radiation. In this technique, the
           metals are excited sufficiently to emit radiation. By using classical dispersion-
           grating optics, a large number of metals can be analyzed simultaneously.  Currently,
           the utility of this technique is limited by lack of sensitivity for drinking water
           Maximum Contaminant Levels (MCLs), but sample preparation techniques under
           development that will overcome this problem should be available after March of
           1989.

           Inductively Coupled Plasma Mass Spectrometrv (TCP/MS). This technique (EPA
           Method 602.0) is the most expensive of those listed in this section, but it may be
           the most effective and cost-effective if used for all U.S. EPA programs seeking
           lower detection levels and/or MCLs.  In this technique excitation is effected using a
           radio-frequency  plasma, but the excited atoms are interfaced into a mass
           spectrometer. Quantitation is achieved by computerized software programs similar

                                           17

-------
Beryllium                                                                    April, 199.2.
           to those used in other U.S. EPA MS organic methods. The main problem with
           TCP/MS is metal deposition on the interface, resulting in giving the possibility of
           "memory" when an extremely low-concentration sample is run after a high-
           concentration sample. However, normal analytical steps can be used to resolve this
           problem.  The detection limit of TCP/MS is 0.1 /xg/L.


VIII.  TREATMENT TECHNOLOGIES

        •  The available information indicates that reverse osmosis (RO) systems and
           conventional coagulation/filtration will remove beryllium in drinking water.

        •  Fox and Sorg (1987) described the effectiveness of using an RO unit as a
           point-of-use device for removing beryllium. A laboratory-size RO unit containing a
           spiral-wound polyamide membrane was operated at a pressure of 42 ฑ 2 psi.  This
           system removed 97.7% of the beryllium from a 0.043-rag/L influent.

        •  Hannah et al. (1977)  tested the effectiveness of a pilot plant utilizing
           coagulation/filtration or excess lime treatment in removing beryllium. The plant
           consisted of a rapid mix- designed for a capacity of 4 gpm, a flocculater, a
           sedimentation basin and dual-media filtration. Beryllium was present in the
           influent at a concentration of 0.1 rng/L. Hydrated lime was added at a dose of 415
           rag/L and a pH of 11.5; ferric chloride was added at a dose of 40 rag/L and a pH of
           62', and alum was added at a dose  of 220 rag/L and a pH of 6.4.  Excess lime
           treatment reduced the concentration of beryllium by 99.4%. Ferric chloride
           coagulation reduced the beryllium concentration by 94%, while alum coagulation
           reduced the beryllium concentration by 98.1%.

        •  Hannah et al. (1977)  also reported the results of using granular-activated carbon
           (GAC) adsorption as a post-treatment to the conventional coagulation/filtration
           mentioned above. Using the beryllium-containing.effluents from the
           coagulation/filtration  process above, two GAC columns, operated in parallel and
           designated as "old" and "new," were tested. The "old" GAC was in use for several
           months before an evaluation of its  performance was made.  When the lime
           coagulation effluent was processed through the "new" or the "old" GAC columns, an
           additional 0.1% of the beryllium was removed, for a total of 99.5%.  When the
           ferric chloride coagulation effluent was processed through the "old" GAC column,
           an additional 1.4% of the beryllium was removed, for a total of 98.7%; the "new"
           GAC column removed an additional 4.9%  for a total of 98.9%.  When the  alum
           coagulation effluent was processed through the "old" or the "new" GAC column an
           additional 0.8% of the beryllium was removed, for a total of 98.9%.
                                           18

-------
Beryllium                                                                     April, 1992,
IX.  REFERENCES

ACGIH. 1987. American Conference of Governmental and Industrial Hygienists.  Threshold
        limit values and biological exposure indices for 1987-1988.  Cincinnati, OH: ACHIH,
        p. 13.

AJdrich Chemical Co., Inc.  1986. Aldrich catalog handbook of fine chemicals. Aldrich
        Chemical Co., Inc., Milwaukee, WI.

Barnes, J.M., F.A. Denz and H.A. Sissons.  1950.  Beryllium bone sarcomata in rabbits.  Br. J.
        Cancer 4:212-222.

Branion, H.D., B.L. Guyatt and H.D. Kay.  1931.  Beryllium rickets.  J. Biol. Chera. (Science
        Proceedings XXV) 92:11.

Businco, L.  1940. The rickets-producing effect of beryllium carbonate.  Ind. Med.  Rev.
        11:417-442. (In Italian; translation.)

Callahan, M.A., M.W. Slimak, N.W. Gabel, I.P. May, C.F. Fowler, J.R. Freed, P'. Jennings, R.L
        Durfee, F.C. Whitmore, B. Maestri, W.R. Mabey, B.R. Holt, and C. Gould. 1979.
        Water related environmental fate of 129 priority pollutants. Publication no. EPA
        440/4-79-029.   Washington, DC: U.S. Environmental Protection Agency.

Cloudraan, A.M., D. Vining,  S. Barkulis, and JJ. Nickson.  1949.  Bone changes observed
        following intravenous injections of beryllium.  Am. J. Pathol. 25:810-811.

Cox, G.E., D.E. Bailey and K. Morgareidge.  Food and Drug Research Laboratories, Inc. 1975.
        Chronic feeding studies with beryllium sulfate in rats.  Draft final report.  Pittsburgh,
        PA.  Aluminum Company of America.

Dutra, F.R. and EJ. Largent, 1950. Osteosarcoma induced  by beryllium oxide. Am. J. Pathol.
        26:197-209.

Dutra, F.R., EJ. Largent and J.L, Roth.  1951. Osteogenic  sarcoma after inhalation of
        beryllium oxide.  Arch. Pathol. 51:473-479.

Eisenbud, M. and J. Lisson.  1983. Epidemiological aspects  of beryllium-induced nonmalignant
        lung disease:  A 30-year update. J. Occup. Med. 25:196-202.

Fox, K.R. and TJ. Sorg.  1987. Controlling arsenic, fluoride and uranium by point-of-use
        treatment  J. AWWA. 79(10):81-84.

Furchner, I.E., C.R. Richmond and J.E. London.  1973. Comparativeraetabolism of
        radionuclides in mammals. VIE. Retention of beryllium in the mouse, rat,  monkey and
        dog.  Health Phys. 24:293-300.
                                           19

-------
Beryllium                                                                     April, 1992*
Gardner, L.U. and HJF. Heslington.  1946.  Osteosarcorna from intravenous beryllium
       compounds in rabbits. Fed. Proc. 5:221.

Guyatt, G.L., H.D. Kay and H.D. Branion.  1933.  Beryllium rickets.  J. Nutr. 6:313-324.

Hannah, S.A., M. Jelus and J.M. Cohen.  1977.  Removal of uncommon trace metals by physical
       and chemical treatment processes. J. Water Pollut. Cont. Fed.  49:2297-2309.

Hart, B.A., A.G. Harmsen, R.B. Low and R. Emerson.  1984.  Biochemical, cytological and
       histological alterations in rat lung following acute beryllium aerosol exposure. Toxicol.
       Appl. Pharmacol.  75:454-465.

Hawley, G.G.  1981.  The condensed chemical dictionary, 10th ed.  New York, NY: Van
       Nostrand Reinhold Co.

Hoagland, M.B., R.S. Grier, and M.B. Hood 1950. Beryllium and growth.  I.
       Beryllium-induced osteogenic sarcomata. Cancer Res. 10:629-635.

Hyslop, F., E.D. Palmes, W.C. Alford, A.R. Monaco and L.T. Fairhall.  1943. The toxicity of
       beryllium.  NIH Bull. no. 181. Washington DC: National Institutes of Health, p. 56.

I ARC. 1987. International Agency for Research on Cancer.  IARC monographs on the
       evaluation of carcinogenic risks to humans. Suppl. 7. Beryllium and beryllium
       compounds. Lyon, France: IARC, pp. 127-128.

IRPTC. 1987.  International Register of Potentially Toxic Chemicals.  IRPTC data profile on
       beryllium.  Geneva, Switzerland: United Nations Environment Programme.

Kada, T., K. Hirano and Y. Shirasu.  1980.  Environmental chemical mutagens  by the Rec-assav
       system with Bacillus subtilis.  Chem. Mutagens.  6:149-173.

Kanematsu, N., M. Hara and T. Kada.  1980. Rec assay and mutageniciry studies on metal
       compounds. Mutat. Res. 77:109-116.

Kay, H.D. and D.I. Skill  1934.  CLXDC Beryllium rickets IL The prevention and cure of
       beryllium rickets.  Biochem. J. 28:1222-1227.

Kubinski, H., G.E. Gutzke and Z.O. Kubinski.  1981. DNA-cell-binding (DCB) assay for
       suspected carcinogens and mutagens.  Mutat. Res. 89:95-136.

Larraraendy,  ML., N.C Popescu and J.A. diPaolo. 1981.  Induction by inorganic metal salts of
       sister chromatid exchanges and chromosome aberrations  in human and Syrian hamster
       cell strains.  Environ. Mutagen. 3:597-606.

Lehman, A.  1959.  Appraisal of the safety of chemicals in foods, drugs and cosmetics.
       Association of Food and Drug Officials of the United States.


                                          20

-------
Beryllium                        .                                             April, 1992.
Luke, M.Z., L Hamilton and T.C. Hollocher.  1975.  Beryllium-induced raisincorporation by a
        DNA polymerase. Biochem. Biophys.  Res. Comm. 62:497-501.

McGinnis, P. 1988. Review of beryllium file for CRAVE Summary, Task #29. Memorandum
        to L. Papa, Environmental Criteria and Assessment Office, U.S. Environmental
        Protection Agency.  Cincinnati, Ohio.  Syracuse Research Corporation, December 28
        1988.

Morgareidge, K., G.E. Cox, D.E. Bailey and MA. Gallo.   1977. Chronic oral toxicity of
        beryllium  in the rat.  Toxicol. Appl. Pharmacol. 41:204-205.

Nash, P. 1950.  Experimental production of malignant tumors by beryllium.  Lancet. 1:519.

NIOSH. 1972.  National Institute for Occupational Safety and Health. Criteria for a
        recommended standard occupational exposure to  beryllium.  Washington, DC:
        Department of Health, Education and Welfare, p. 72-10269.

OSHA.  1985.  Occupational Safety and Health Administration. Occupational standards and
        permissible exposure limits.  Code of Federal Regulations 29:1910.1000.'

Perry, R.H. and C.H. Chilton.  1973.  Chemical engineers handbook,  5th ed. New York, NY:
        McGraw Hill Book Co.

Reeves, A.L. 1965. The absorption of beryllium from the gastrointestinal tract. Arch. Environ.
        Health 11:209-214.

Reeves, A.L., D. Deitch and AJ. Vorwald. 1967. Beryllium carcinogenesis.  I. Inhalation
        exposure of rats to beryllium sulfate aerosol.  Cancer Res. 27:439-445.

Reeves, A.L. and AJ. Vorwald 1961. The humoral transport  of beryllium. J.  Occup. Med.
        3:567-571.

Rhoads, K. and C.L. Sanders.  1985.  Lung clearance, translocation and acute toxicity of arsenic,
        beryllium, cadmium, cobalt, lead, selenium, vanadium and ytterbium oxides following
        deposition in rat lung.  Environ. Res. 36:359-378.

Rosenkranz, H.S. and LA. Poirier.   1979.  Evaluation of  the rautagenicity and DNA-modifying
        activity of carcinogens and noncarcinogens in microbial systems.  J. Natl. Cancer Inst.
        62:873-892.                '

Schroeder, H.A. and M. Mitchener.  1975a. Life-term studies in rats: Effects of aluminum,
        barium, beryllium and tungsten. J. Nutr. 105:421-427.

Schroeder, H.A. and M Mitchener.  1975b. Life-term effects of mercury, methyl mercury and
        nine other trace metals on mice. J. Nutr. 105:452-458.
                                           21

-------
Beryllium                                                                    April, 199,2.
Scott, J.K., W.F. Neuraan and R. Allen.  1950.  The effect of added carrier on the distribution
       and excretion of soluble 7Be. J. Biol. Chera. 182:291-298.

Simmon, V.F.  1979a.  In  vitro mutagenicity assays of chemical carcinogens and related
       compounds with Salmonella typhimurium. J. Natl. Cancer Inst. 62:893-899.

Simmon, V.F.  1979b.  In  vitro assays for recornbinogenic activity of chemical carcinogens and
       related compounds with Saccharomyces cerevisiae D3. J. Natl. Cancer Inst. 62:901-909.

Simmon, V.F., H.S. Rosenkranz, E. Zeiger and LA. Poirier.  1979. Mutagenic activity of
       chemical carcinogens and related compounds in the intraperitoneal host-mediated assay.
       J. Natl.  Cancer Inst. 62:911-918.

Sirover, MA. and LA. Loeb.  1976.  Metal-induced infidelity during DNA synthesis. Proc.
       Natl. Acad. Sci. 73:2331-2335.

Skilleter, D.N., RJ. Price  and R.F. Legg.  1983. Specific GrS phase cell block by beryllium as
       demonstrated by cytofluorometric analysis.  Biochern. J. 216:773-776.

Tsujii, H. and K. Hoshishiraa. 1979. The effect of the administration of trace amounts of
       metals to pregnant mice upon the behavior and learning of their offspring. Shinshu
       Daigaku Nogakuba Kiyo.  16:13-27.

U.S. EPA.  1991.  U.S. Environmental Protection Agency. Integrated Risk Information System
       (IRIS) online. Cincinnati, Ohio:  U.S. EPA Office of Health and Environmental
       Assessment, Environmental Criteria and Assessment Office.

U.S. EPA.  1988.  U.S. Environmental Protection Agency. Drinking water criteria document for
       beryllium. Draft.  ECAO-CIN-0003. Cincinnati, OH; U.S. EPA Environmental Criteria
       and Assessment Office.

U.S. EPA.  1987.  U.S. Environmental Protection Agency. Health assessment document for
       beryllium. Document no. EPA 600/8-84/-026F. Research Triangle Park, NC: U.S. EPA
       Office of Health and Environmental Assessment, Environmental Criteria and
       Assessment Office.

U.S. EPA.  1986.  U.S. Environmental Protection Agency. Guidelines for carcinogen risk
       assessment  Fed.  Reg. 51(185):33992-34003. September 24.
                                          22

-------
Beryllium                                                                     April, 1992.
U.S. EPA.  1980.  U.S. Environmental Protection Agency.  Ambient water quality criteria
       document for beryllium.  Prepared by the Office of Health and Environmental
       Assessment, Environmental Criteria and Assessment Office. Document no. EPA
       440/5-80-024. Washington, DC: Office of Water Regulations and Standards.

U.S. EPA.  1977.  U.S. Environmental Protection Agency.  Multimedia environmental goals for
       environmental assessment.  Vol. II. MEG charts and background information.
       Document no. EPA 600/17-77-136b.  Washington, DC: U.S. EPA.

Wagner, W.D., D.H. Groth, J.L. Holtz, G.E.  Madden and HJE. Stokinger.  1969.  Comparative
       inhalation toxicity of beryllium ores, bertrandite and beryl, with production of
       pulmonary tumors by beryl.  Toxicol. Appl. Pharmacol. 15:10-29.

Wagoner, J.K., P.F. Infante and D.L. Bayliss.  1980.  Beryllium: An etiologic agent in the
       induction of lung cancer, nonneoplastic respiratory disease among industrially exposed
       workers.  Environ. Res. 21:15-34.

Weast, R.C. 1982. CRC handbook of chemistry and physics, 62nd ed. Boca Raton, FL:  CRC
       Press, Inc.

Weast, R.C., MJ. Astle and W.H. Beyer.  1986. CRC handbook of chemistry and physics.
       1985-1986, 66th ed. Boca Raton, FI:  CRC Press, Inc.

Williams, G.M., M.F. Laspia and V.C. Dunkel. 1982. Reliability of hepatocyte primary culture.
       Mutat. Res. 97:359-370.

Windholz M., S. Budavari, L.Y. Strqumtsos and MJ*. Fertig, eds. 1976. The Merck index. An
       encyclopedia of chemicals and drugs, 9* ed.  Rahway, NJ:  Merck and Co., Inc.

Zakour, R.A. and B.W. Glickman.  1984.  Metal-induced mutagenesis in the lacl gene of
       Escherichia coli.  Mutat.  Res. 126:9-18.
                                          23

-------
  EPA 0553

  EX000027511
                                                                              April 1992
                                        BORON
                            Drinking Water Health Advisory
                                    Office of Water
                          U.S. Environmental Protection Agency
I.       INTRODUCTION

        The Health Advisory Program, sponsored by the Office of Water (OW), provides
information on the health effects, analytical methodology, and treatment technology that
would be useful in dealing with the contamination of drinking water. Health Advisories (HAs)
describe nonregulatory concentrations of drinking water contaminants at which adverse health
effects would not  be anticipated to occur over specific exposure durations.  Health Advisories
contain a margin of safety to protect sensitive members of the population.

        Health Advisories serve as informal technical guidance to assist Federal, State, and
local officials responsible for protecting public health when emergency spills or contamination
situations occur.  They are not to be construed as legally enforceable Federal standards. The
HAs are subject to change as new information becomes available.

        HAs are developed for One-day, Ten-day, Longer-term (approximately  7 years, or
10% of an individual's lifetime), and Lifetime exposures based on data describing
noncarcinogenic endpoints of toxicity. For those substances that are known or probable
human carcinogens, according to the  Agency classification scheme (Group A or B), Lifetime
Health Advisories are not recommended.  For substances with a carcinogenic potential,
chemical concentration values are correlated with carcinogenic risk estimates by employing a
cancer potency (unit risk) value together with assumptions for lifelong exposure and the
ingestion of water. The cancer unit risk is usually derived from a linearized multistage model
with 95% upper confidence limits providing a low-dose estimate of cancer risk.  The cancer
risk is characterized as being an upper limit estimate, that is, the true risk to humans, while
not identifiable, isjiot likely to exceed the upper limit estimate and in fact may  be lower.
While alternative  risk modeling approaches  may be presented, for example one-hit, Weibull,
logit, or probit, the range of risks described by using any of these models has little biological
significance unless data can be used to support the  selection of one model over another.  In
the interest of consistency of approach and  in providing an upper-bound  on the potential
carcinogenic risk, the Agency recommends using the linearized multistage model.
II.      GENERAL INFORMATION AND PROPERTIES

        CAS Nos.

        •     Boron — 7440-42-8
        •     Boric acid — 10043-35-3
        •     Sodium tetraborate — 1303-96-4

                                            1

-------
Boron
                                                                        April 1992
        Structural Formula

        •     Not applicable.

        Synonyms

        •     Boron:  none
        •     Boric acid: orthoboric acid
        •     Sodium tetraborate (as the decahydrate): borax
        Uses
              Elemental boron (B) and its carbides are used in composite structural
              materials, in high-temperature abrasives, in special  purpose alloys and in steel-
              making (Hawley, 1981). Boric acid and borates are used in glass manufacture,
              cleaners, wood and leather preservatives,  flame retardants, cosmetic products
              and neutron absorbers for nuclear installations (Sittig, 1981).  Borax is also
              used as an insecticide for cockroaches and black carpet beetles (Windholz et
              al., 1983). Boron halides are used as catalysts in the  manufacture of
              magnesium alloy products, metal refining, magnesium solder fluxes, rocket
              fuels, and detoxification of nitrosamine-contaminated wastes (U.S. EPA, 1975).
              Boron hydrides are used as reductants to control heavy metal discharges in
              wastewater, as catalysts and in jet and rocket  fuels  (U.S. EPA, 1975).  They are
              also used as household insecticides.
        Properties  (Windholz et al., 1983)

                            Boron       Boric acid
                                                        Borax
Chemical Formula
Molecular Weight
Physical State
Boiling Point
Melting Point
Density
Vapor Pressure
Water Solubility
Specific Gravity
Log Octanol/Water
  Partition
  Coefficient (log
Taste Threshold
Odor Threshold
Conversion Factor
                            B
                            10.81
                            Solid
                            —
                            2,300ฐC
                            2.35
                            —
                            Insoluble
                            2.45
H3BO3
61.84
Solid

171ฐC
1 g/18 mL (cold)
1.43
3813
Solid
320ฐ
75ฐ (rapid heating)
1.73

1 g/16 mL (cold)
1.73
                            NA
NA
35.27

-------
Boron                                                                       April 1992
      Occurrence

      •     The most common forms of boron in nature are boric acid and sodium
            tetraborate.

      •     Boron, a naturally occurring element, is found in soil at an average concentration
            of 10  mg/kg (Weast, 1988), in  ocean  waters at a concentration of 4.6 mg/L
            (Weast, 1988) and in freshwater at a concentration of 0.01 /ig/g (U.S. EPA,
            1980).

      •     The total daily boron intake in normal human diets has been reported to range
            from  2.1 to 4.3 mg/day (Zook  and Lehman, 1965) and from 1.3 to 4.4 mg/day
            (Hamilton and Minski, 1972).

      •     Boron found in coal, oil shale and geothermal fluids contributes to
            environmental pollution  (White, 1980).  Excluding  borate production, an
            estimated 1,000 to 4,000  tons of boron are released to the environment each year
            (U.S.  EPA, 1975).

      •     Municipal sludge  and industrial wastewater contribute to the release of boron to
            soil, ground water, and estuaries (Hemphill et al., 1981).

      Environmental Fate

      •     No information regarding the  environmental fate of boron was located in the
            literature.
III.   PHARMACOKINETICS

      Absorption

      •     Jansen et al. (1984b) administered a single oral dose of 750 mg of boric acid to
            each of six human volunteers (approximately 1.9 mg/kg of boron; average body
            weight 70.8 kg).  At least 93.9% of the dose was absorbed from the
            gastrointestinal (GI)  tract, as measured by the recovery of boric acid in urine
            after 96 hours.

      •     Kent and McCance (1941) orally administered a total of 352 mg of boron (as
            boric acid) to each of two women over a 3-day period (approximately 2.35 mg/kg/
            day, body weight 50 kg).  At least 93 to 94% of the boric acid was absorbed from
            the GI tract, as measured by recovery of boric acid in urine within a week.

-------
Boron                                                                         April  1992
      •     Draize and Kelley (1959) estimated that urinary excretion of boron accounted
            for 50 to 66% of the quantities of boric acid (17.1 to 119.9 mg/kg/day of boron)
            administered orally to rabbits on four consecutive days.

      •     Draize and Kelley (1959) reported no increase in the urinary excretion of boron
            by a human volunteer following a 4-hour exposure of the forearm to 15 g of
            boric acid (approximately 37.5 mg/kg of boron, body weight 70 kg) wetted with
            physiological saline. Results indicated that little or no dermal absorption of
            boron occurred through the intact skin.

      •     Mulinos et al. (1953) applied a commercial talcum powder containing 5% boric
            acid to the skin of six infants with  no (one infant) or varying degrees (mild, one
            infant; moderate, two infants; marked, one infant) of diaper rash. Boric acid was
            not detected in the urine of any of the infants before powder application.
            Twenty-four hours after powder application, boric acid was detected in the urine
            of the infants with moderate to marked rashes, and  it persisted in the urine for
            at least 48 hours. No boric acid was detected in the urine of infants with no rash
            or mild diaper rash, indicating that little or no absorption occurred through the
            intact skin.  No quantitative data on absorption were presented in this study.

      •     Vignec  and Ellis (1954) applied a powder containing 5% boric acid 7 to 10 times
            daily for at least 1 month to infants varying in age from  1.25 to 10 months.  The
            infants were exposed to a calculated dose of approximately 2.33 g boric acid/
            infant/day.  Boron concentrations in the blood and urine were determined
            colorimetrically.  In the test group containing 12 infants, blood boron levels
            varied from 0.01 mg to 0.14 mg/100 mL (0.04 ฑ 0.05 mg/100 mL, mean ฑ SD),
            and the urine levels varied  from 0.02 mg  to 0.44 mg/100  mL (0.16 ฑ 0.14 mg/100
            mL, mean +. SD).  Corresponding values in the 12 control infants were 0 to 0.19
            mg/100 mL (0.10 ฑ 0.05 mg/100 mL, mean ฑ SD) for blood and 0 to 0.3 mg/100
            mL (0.08 ฑ 0.07 mg/100 mL, mean SD) for urine.  Twelve other infants tested
            developed diaper rashes varying in severity from mild erythema to moderately
            severe diaper rash with blood levels of boron ranging from 0 to 0.15 mg/100 mL
            (0.03 ฑ 0.04 mg/100 mL, mean ฑ SD).  Similar boron levels in the serum and
            urine of control  and test infants indicate.that no remarkable topical  absorption
            of.boric acid occurred in this study.

      •     Draize and Kelley (1959) applied 5% aqueous boric acid (35 mg/kg of boron) to
            the intact or damaged skin of rabbits for 1.5 hours a day for 4 days. Net urinary
            excretion of boric acid amounted to 0.25, 1.3 or 3.7 mg/kg of boron, depending
            on whether the application had been made to intact, abraded or burnt and
            partially denuded skin, respectively. Application of a wet boric acid powder (700
            mg/kg of boron) resulted in net urinary excretion of boric  acid amounting to

-------
Boron                                                                         April 1992
            0.14, 0.98 or 22.2 mg/kg of boron, depending on whether the application had
            been made to intact, abraded or burnt and partially denuded skin, respectively.

      Distribution

      •     Locksley and Sweet (1954) reported that borate (administered in borax)
            concentrations in the tissues of normal mice were directly proportional to the
            dosage of borax administered (1.8 to 71.0 mg/kg of boron) for at least 2 hours
            after intraperitoneal  injection.  After equilibrium was achieved in
            nephrectomized  mice, about 90% of the borate was distributed uniformly
            throughout the body water, and about 10% was bound in the intracellular
            compartment.

      •     Grella et al. (1976) described transplacental distribution  of boric acid in humans.
            A 34-week pregnant  female accidentally swallowed 70 g of boric acid
            (approximately 245 mg/kg of boron assuming a 50-kg body weight).  A fetus
            delivered 2 hours later by Cesarean section died shortly  afterward from
            cardiovascular failure.

      Metabolism

      •     No information on the metabolism of boron was found in the literature.

      Excretion

      •     Jansen et al.  (1984b) administered single oral doses of boric acid (approximately
            1.9 mg /kg of boron) to six male human volunteers aged 30 to 58 years (mean
            47.3 years), over 93% of the dose was excreted in the urine within 96 hours of
            dosing.

      •     Jansen et al.  (1984a) administered via a 20-minute intravenous infusion a total
            dose of boric acid of 570 to 620 mg (corresponding to approximately 1.4 to 1.5
            mg/kg of boron for a 70-kg male) to each of eight male adults (22 to 28 years
            old).  Within 120 hours of administration,  98.7% of the dose was excreted in the
            urine.

-------
Boron                                                                         April 1992



IV.   HEALTH EFFECTS

      Humans

            Short-term Exposure

            •    The lowest oral lethal dose (LD^) of boric acid in humans is reported to
                  be 640 mg/kg (112 mg/kg of boron), and the dermal LD^, is 8,600 mg/kg
                  (1,505 mg/kg of boron); the intravenous LD^ for boric acid is reported to
                  be 29 mg/kg (5.1 mg/kg of boron) (Stokinger, 1981).

            •    Six fatalities occurred within 3.5 days after eight infants ingested formula
                  prepared with a 2.5% aqueous solution of boric acid.  In one case (3.49 kg
                  body weight), it was estimated that the amount present in the whole body
                  was between 1 and 2 g of boric acid at death, 45 hours after ingestion
                  (50.1 mg/kg of boron, for ingestion of 1 g boric acid).  The other infants
                  also may have ingested less than 3 g of boric acid (Young et al.,  1949).

            •    Five fatalities were reported among 11  infants who drank formula
                  prepared with 2.5% aqueous boric acid. The total amount of boric acid
                  ingested by each of the five infants who died was estimated to range from
                  4.5 to 14.1 g; ingestion by the six survivors was estimated to range from
                  2.0 to 4.5  g. It was not specified whether ingestion was limited to one
                  dose in all cases.  For one surviving female infant, who apparently
                  consumed only one dose, the estimated ingestion was 3 g of boric acid
                  (165 mg/kg of boron in an infant weighing 3.2 kg) (Wong et al., 1964).

            •    A 5-month-old child (6.1 kg body weight) ingested 1.83 g of a boric acid
                  solution (53 mg/kg of boron).  Except for vomiting 1 hour later and a
                  slight lethargy at about 12 hours after ingestion, the child was alert and
                  well 36 hours after ingestion. Based on the serum level of boric acid, a
                  total body boric acid level of 491 mg (i.e., 85.9 mg boron, or 14 mg/kg)
                  was estimated at 7 hours after ingestion (i.e., 14 mg/kg of boron) (Martin,
                  1971).

            •    A 109-kg 44-year-old female ingested 14 g of boric acid (i.e., 22.5 mg/kg
                  of boron). The patient exhibited erythema, desquamation, elevated liver
                  function tests and central nervous system (CNS) involvement. The patient
                  was discharged after 2 weeks in the hospital (Schillinger et al., 1982).

            •    Linden et al. (1986) reported four cases of  nonfatal ingestion of boric
                  acid.  Two adult females ingested 298 g of a 99% boric acid-containing
                  insecticide and 80  g of boric acid in a fungicide, respectively (presumably

-------
Boron                                                                         April  1992
                   1.0 and 0.28 g/kg of boron, respectively, assuming a 50-kg body weight).
                   The doses ingested by the two other subjects were not clearly specified.
                   All four subjects recovered, and the two adults presented no systemic
                   signs of toxicity following release from the hospital.

             •      Six male volunteers aged 30-58 years (mean  47.3) received single oral
                   doses of boric acid (Jansen et al. 1984b). Three volunteers ingested  750
                   mg boric acid dissolved in 100 mL of water (1.9 mg/kg of boron based on
                   an assumed 70-kg body weight). Three other volunteers swallowed 24.95
                   g  to 49.6 g of a commercial water-emulsifying ointment containing 2.97%
                   (w/w) boric acid (total amount of boric acid consumed by the individuals
                   was 0.740 g to  1.473 g; 1.8 mg to 3.6 mg/kg of boron). No adverse health
                   effects were reported in any of the individuals  following a single ingestion
                   of 1.8 to 3.6 mg/kg boron during the 96-hour observation period.

             •      In another study (Jansen et al., 1984a), eight 22- to 28-year-old male
                   volunteers were given 20-minute intravenous infusions of 21 mg/mL boric
                   acid in  sterile water; the boric acid was infused at a rate of 28.52  to 31.0
                   mg/min, resulting  in a total dose of 570 to 620 mg (1.4 to 1.5 mg/kg) of
                   boron,  for a 70-kg adult. Within 120 hours of treatment, 98.7% of the
                   boron dose was excreted in the urine.  This  indicated that excretion of
                   boron was nearly complete, and that boron exhibited no tendency to
                   accumulate in the tissues.  None of the volunteers reported any
                   discomfort following the infusion.  The No-Observed-Adverse-Effect Level
                   (NOAEL) in this  study was 1.4 to 1.5 mg/kg  of boron.

             Long-term Exposure

             •      Following ingestion of borax (present in pacifiers coated with a borax-
                   honey mixture) over of period of several weeks, two infants suffered
                   seizures and exhibited neurological  symptoms. One infant (4.5 months
                   old) also had erythema of the scalp, trunk and limbs;  scanty hair  and
                   increased cellularity in bone marrow aspirates.  Other clinical findings  for
               ,    the second child (9 months old) were unremarkable.  The following
                   ingestions of borax were estimated:  an accumulated total of 125  mg over
             =v     a  12-week period  for  the first child and 9 g over 5 weeks for the second
                   child (approximately 66 and 13 mg/kg/day of boron, respectively)  (Gordon
                   et al., 1973).

             •      In a human nutrition study using basal diets supplying 0.25 mg/day of
                   boron,  Nielsen et al. (1987) reported that supplementing the basal diets of
                   12 postmenopausal women with 3 mg/day of boron for 119 days (an
                   approximate total of 0.065  mg/kg/day of boron for a 50-kg woman)

-------
Boron                                                                         April 1992
                   reduced calcium and magnesium excretion in the urine and elevated
                   serum steroid levels. The effect appeared to be enhanced by low
                   magnesium levels in the diet.

      Animals

            Short-term Exposure

            •      Acute oral LDJO values for boric acid in mice and rats range from 466 to
                   900 mg/kg of boron, and the acute oral LDJO values for borax in rats range
                   from 396 to 689 mg/kg of boron (Pfeiffer  et al, 1945; Wang et al., 1984;
                   Weir and Fisher, 1972).

            •      Three of six dogs dosed orally with boric  acid at 350 mg/kg of boron (as
                   boric acid) died within 48 hours of exposure (Pfeiffer et al., 1945).

            •      Immature rats (five groups of 20 to 24) of unspecified weight were
                   administered boric acid  in their drinking water at levels of 1 g/L, 2.5 g/L
                   or higher unspecified concentrations (approximately 22.8, 57 or more mg/
                   kg/day of boron in animals weighing 100 g and consuming  12 to 14 mI7
                   day of water) for at least 30  days. Growth was not inhibited in animals
                   receiving the lowest dose, but growth was inhibited in all other animals
                   after 20 or 30 days. There were no significant findings in hematology or
                   in gross and microscopic pathology (Pfeiffer et al., 1945).

            •      Seal and Weeth (1980) administered 0, 150 or 300 mg of boron (as borax)
                   per liter of drinking water to 45 male Long-Evans rats for 70 days. The
                   basal diet contained approximately 54 ^g  boron/g of feed.  The total
                   intake of boron of the treated rats was 23.7 and 47.4 mg/kg/day,
                   respectively (assuming a body weight of 0.35 kg, a fluid intake of 0.049 L/
                   day and a daily food consumption equal to  5% of the body weight). Both
                   doses of borax produced significant (p  <  0.05) decreases in body weight;
                   in the weights of the testes, seminal vesicles, spleen and right femur and
                   in the levels of plasma triglycerides. In addition, spermatogenesis was
                •   impaired in animals receiving the highest  dose.  The Lowest-Observed-
                   Adverse-Effect Level (LOAEL) in this study was 23.7 mg/kg/day of boron.

            •      Forbes and Mitchell (1957) administered  19 (unsupplemented control
                   diet), 73,104 or 198 mg/kg/day of a boric acid-supplemented diet to 48
                   male Sprague-Dawley weanling rats; based on the assumptions of Lehman
                   (1959), these doses correspond to approximately 3.8,  14.6, 20.8 or 39.6 mg/
                   kg/day of boron.  At the end of 8 weeks,  animals in the two higher-dose
                   groups had significantly (p = 0.05) lower body weights than the untreated

                                           8

-------
Boron                                                                        April 1992
                  controls; no effect was observed in animals receiving 73-mg/kg of boron.
                  However, the percentage of the boron dose retained in soft tissue and
                  skeleton was inversely proportional to the dietary boron level, ranging
                  from 45% in the 19-mg/kg group to 16%  in the 198-mg/kg group. Based
                  on the results of this study, the suggested NOAEL is 14.6  mg/kg/day.

            •     Green et al. (1973) observed no adverse effects in six 30-day-old rats
                  administered 75 ppm borax (0.2 mg/kg/day of boron).

            Dermal/Ocular Effects

            •     Roudabush  et al. (1965) studied the dermal irritation potential of borax
                  and boric acid. Ten mL of 5% aqueous borate (5.7 g/L of boron) and 5
                  mL of 10%  aqueous boric acid (17.5 g B/L) were applied  under occlusion
                  to the clipped,  intact and abraded skin of rabbits and guinea pigs. Sites
                  were scored for irritation at 24 and 72 hours. Both borax  and boric acid
                  were found  to be mild-to-moderate irritants.

            •     In an unpublished study sponsored by the Cosmetics, Toiletries  and
                  Fragrances Association (CTFA) and reviewed by an American College of
                  Toxicology expert  panel (ACT, 1983), a bath preparation  containing 0.4%
                  boric acid (0.7  g/L of boron) was tested for dermal irritation in albino
                  rabbits.  The test solution was applied undiluted and under occlusion to
                  the shaved intact skin of each  animal for 24 hours. Of the nine animals
                  tested, eight experienced mild-to-moderate erythema.

            •     In an unpublished CTFA-sponsored study reviewed by ACT (1983), a bath
                  preparation containing 0.4% boric acid (0.7 g/L of boron) was tested for
                  ocular irritation in albino rabbits.  The eyes of the rabbits were  rinsed
                  with warm water 4 seconds after instillation of the test material.  The
                  product was found to be moderately irritating.

            Long-term Exposure

            •     In a subchronic study conducted by the National Toxicology Program
                  (NTP, 1987), groups of 10 male and 10 female B6C3F, mice were fed
                  diets containing 0, 1,200, 2,500, 5,000,10,000 or 20,000 ppm of boric acid
                  for 13 weeks (approximately 0, 34, 68,136, 272 or 544  mg/kg/day of boron
                  for males and 0, 47, 94, 188, 376 or 752 mg/kg/day of boron for  females,
                  respectively, based on reported average values for feed consumption by
                  controls on week 4 of the experiment).  Over 60% mortality was observed
                  at the highest  dose level, and  10% mortality was observed among males
                  dosed with  10,000 ppm boric acid. At doses of 5,000 ppm or higher,

-------
Boron                                                                        April  1992
                  degeneration or atrophy of the seminiferous tubules was observed in
                  males, and weight gain was suppressed in animals of both sexes.
                  Extramedullary hematopoiesis of the spleen of minimal to mild severity
                  was observed in all dosed groups. The lowest dose tested (1,200 ppm)
                  was determined to be the LOAEL for this study.

            •     Weir and Fisher  (1972) administered borax or boric acid in the diet to
                  Sprague-Dawley  rats (10/sex/dose) for 90 days at doses of 0, 52.5, 175, 525,
                  1,750 or 5,250 ppm as boron  equivalents; these doses corresponded to
                  approximately 0,  2.6, 8.8, 26,  88 or 260 mg/kg/day of boron, respectively,
                  assuming a food  consumption equivalent to 5% of the body weight as
                  recommended by Lehman  (1959). Both borax and boric acid produced
                  100% mortality in the highest-dose group and complete atrophy of the
                  testes in all males fed diets containing 1,750 ppm.  At 1,750, both
                  compounds produced significant  (p < 0.05) decreases in body weight and
                  in the mean weights of the liver,  kidney,  spleen and testes.

                  At lower doses, changes in organ weights were inconsistent. At 52.5 ppm,
                  borax produced increases in the mean weights of the spleen, kidneys and
                  ovaries in females, and boric acid produced an increase in the mean
                  weight of the liver in both  males and females. Male rats fed 175  ppm
                  exhibited increased kidney weights. These changes, however, were not
                  observed in animals fed 525 ppm of either compound.  Microscopic
                  examination revealed partial  testicular atrophy in four males fed  borax at
                  525 ppm and in one male fed boric acid  at 525 ppm. Although 52.5 ppm
                  would appear to  be  a LOAEL in this study, changes in organ weights were
                  inconsistent  at this level of dietary boron and no histopathological data
                  were provided.

            •     Weir and Fisher  (1972) administered a diet containing 0, 17.5, 175 or
                  1,750 ppm of boron  in the form of borax or boric acid to groups of five
                  young male and five young female beagle dogs for 90 days (corresponding
                  to approximately 0, 0.44, 4.4 or 44 mg/kg body weight/day, assuming a
                  daily food intake of 2.5% of  the body weight as recommended by Lehman
                  (1959).  Except for one death in  a male dog at the 1,750-ppm boron level
                  (as borax), dogs fed both boron compounds were normal in appearance,
                  behavior, elimination, body weights and food consumption.  At 1,750 ppm
                  of boron, both compounds produced significant (p < 0.05) decreases in
                  thyroid- and testes-to-body weight ratios and severe testicular atrophy with
                  degeneration of the spermatogenic epithelium in all male dogs.  At  175
                  ppm of boron (as boric acid), a decrease in the testes-to-body weight ratio
                  was observed. This  effect, however, was not  accompanied by histologica!
                  changes. At 17.5 ppm of boron (as borax), the LOAEL for this  study, the

                                          10

-------
Boron                                                             '           April 1992
                  spleen-to-body weight ratio in males was significantly (p  < 0.05) elevated.
                  No histological alterations were observed in dogs fed 175 ppm (or less) of
                  boron (as boric acid); histological data for borax at lower doses were not
                  reported.

            •     Settimi et al. (1982) administered 3 g/L of borax in the drinking water to
                  male Wistar rats for 14 weeks (approximately 20.8 mg/kg/day of boron,
                  assuming a body weight of 0.350 kg and an average water consumption of
                  21.4 mL as reported in the study data.  Significant (p < 0.05) elevations in
                  RNA concentration and in succinate dehydrogenase and acid proteinase
                  activities were reported in the brains of the treated rats.  Significant
                  decreases (p  < 0.05) in reduced nicotinamide adenine dinucleotide
                  phosphate (NADPH) cytochrome reductase activity and in the content of
                  cytochromes b, and P-450 were observed in the liver. No effect on body
                  weights or on liver, kidney or testes weights was observed after exposure
                  to boron.  Histopathology was not performed.  Because only one dose
                  level was studied, a NOAEL or LOAEL cannot be identified.

            •     Wang et al. (1984) administered 0, 1, 5, 50 or 500  mg/L of boron (as
                  borax) in the drinking water to rats of an unspecified strain for up to 198
                  days. The treatment groups consisted of 5 rats/sex/group, and the control
                  group consisted of 10 rats/sex. The doses given corresponded to
                  approximately 0, 0.055, 0.280, 2.8 or 28 mg/kg/day  of boron (assuming an
                  average body weight of 350 g and a water consumption rate of 19i5 mL/
                  day  (water intakes of 22 mL for males and 17 mL  for females were
                  reported at week 26 of the study).  Although changes were reported in the
                  pancreas-to-body weight ratios of all treated animals, detailed results were
                  not  provided.  Changes included a significant decrease in the pancreas-to-
                  body weight ratio in female rats on day 98, and a significant  increase in
                  the pancreas-to-body ratio in males on day 198. However, no
                  histopathologic changes were noted.

            •     In a lifetime study, Schroeder and Mitchener (1975) administered 5 ppm
                  of boron (as sodium metabo rate) in the drinking water to 162 male and
                  female Charles River CD Swiss mice; this dose corresponded to
                  approximately 8.1 mg/kg/day of boron, assuming a water intake of 5.5 mL/
                  day  and an average body weight of 34 g (based on the reported weights at
                  90 days).  No effects were observed in body weights or longevity, the only
                  parameters studied. Consequently, the NOAEL in this study was 8.1 mg/
                  kg/day.

            •     Weir and Fisher (1972) fed groups of  four young male and four young
                  female dogs diets containing 0, 58, 117 or 350 ppm of boron (as borax or

                                           11

-------
Boron                                                                         April 1992
                   boric acid) for 2 years; these doses corresponded to approximately 0, 1.5,
                   2.9 or 8.8 mg/kg/day of boron, assuming a daily food intake of 2.5%  of the
                   body weight as recommended by Lehman (1958).  An additional group of
                   dogs was subsequently fed a diet containing 1,170 ppm of boron as borax
                   (approximately 29 mg/kg/day) for 38 weeks. At this dose, severe testicular
                   atrophy and spermatogenic arrest were evident within 26 weeks of
                   treatment. At the lower doses, there were no apparent effects on body
                   weight, food consumption, organ weights, organ-to-body weight ratios or
                   clinical parameters.  Gross and microscopic findings were comparable to
                   those found in the controls.  This study defines a NOAEL of 8.8 mg/kg/
                   day of boron.

             •     Weir and Fisher (1972)  fed groups of 35 male and 35 female
                   Sprague—Dawley  rats 0, 117, 350 or 1,170 ppm of boron as borax or boric
                   acid in the diet for 2-years; these doses corresponded to approximately 0,
                   5.9, 17 J or 58.5 mg/kg/day of boron, assuming a daily food consumption
                   of 5% of the body weight  as recommended by Lehman (1959). At 1,170
                   ppm, both animals fed borax and animals fed boric acid exhibited
                   decreased food consumption during the first 13 weeks of study and
                   suppressed growth throughout the study. The weights of the testes and
                   the testes-to-body weight ratio were significantly (p  < 0.05) decreased,
                   and the brain- and thyroid-to-body weight ratios were also significantly (p
                   < 0.05) decreased.  In addition, the seminiferous epithelium was
                   atrophied and the size of the tubules in the testes was decreased.  No
                   treatment-related effects were observed in rats treated with 350 or 117
                   ppm boron as borax or boric acid.  Therefore the LOAEL in this study
                   was 1,170 ppm (58.5 mg/kg/day of boron) based on the findings noted
                   above.  A NOAEL of 350 ppm (17.5 mg/kg/day of boron) was identified in
                   this study.

             Reproductive Effects

             •     In  two acute exposure studies, adult male rats were dosed  orally with 0 or
                   200 mg/kg of boric acid  in a single administration, or dosed with 0, 250,
                   500, 1000, or 2000 mg/kg in a single administration.  Equivalent doses of
                   boron were 0 or 41 mg/kg or 0, 41, 82,164 or 328 mg/kg, respectively. No
                   definite histologic  changes were detected in animals given  250 or 500
                   mg/kg of boric acid.   In animals dosed with 1,000 or 2,000 mg/kg of boric
                   acid, adverse  effects to the testes and sperm were noted (Linder et al.,
                   1990).

             •     Male rats administered 9,000 ppm of boric acid (equivalent to 74
                   mg/kg/day of boron based on the assumed weight of an adult rat) in their

                                           12

-------
Boron                                                                         April 1992
                   diet for 28 days exhibited testicular lesions.  These lesions were expressed
                   as an inhibition of spermiation followed by exfoliation of germ cells and
                   pachytene-cell  death. After 28 days, extreme epithelial disorganization
                   and germ cell loss were noted. It was also determined that boron
                   exposure reduced basal testosterone levels after 4 days of dosing and
                   levels remained low during dosing (Treinen and Chapin, 1991).

            •      Dixon et al. (1976) administered 0, 0.3, 1.0 or 6.0 mg/L of boron (as
                   borax) in the drinking water to male Sprague-Dawley rats (10/group) for
                   up to 90 days (corresponding  to approximately  0.042, 0.14 or 0.840 mg/kg/
                   day of boron assuming a water consumption rate of 35 mL/day and body
                   weight of 0.250 kg). There were no observable reproductive effects, no
                   changes in serum chemistry or weight of the body, testes, prostate, or
                   seminal vesicles.  Fructose, zinc and acid phosphatase levels in the
                   prostate were also unchanged. No effects on male fertility were observed,
                   suggesting  a NOAEL of 0.840 mg/kg/day (the highest dose tested).

            •      Silaev et al. (1977)  administered a single daily oral dose of 1 g/kg of boric
                   acid (175 mg/kg/day of boron) to 12 experimental and six control sexually
                   mature male albino  rats for 2  weeks.  Vacuolation and granulation of the
                   cytoplasm as well as an almost total absence of nuclear chromatin were
                   observed in the spermatids of most seminiferous tubules.  In some
                   seminiferous tubules, an appreciable reduction  in tubular diameter and a
                   complete absence of germinative cells were observed

            •      Krasovskii  et al. (1976) administered 0, 0.3, 1.0 or 6.0 mg/L of boron (as
                   boric acid) in drinking water to random-bred white male rats (number not
                   specified) for 6 months approximately 0.015, 0.05 or 0.3 mg/kg/day of
                   boron, based on  the authors' calculations.  No  significant toxic effects
                   were observed at 0.3 mg/L of  boron.  A significant decrease in mobility
                   time (p < 0.01) and number of spermatozoa (p < 0.01) was observed in
                   animals in the two  higher-dose groups, the intensity of these effects
                   increased in dose-related fashion.  In addition,  at the highest dose level,
                   there was a significant (p < 0.01).decrease in the organ-to-body weight
                   ratio for gonads and a significant (p < 0.01) increase in serum levels of
                   blood aldolase. This study defines a NOAEL and a LOAEL of 0.02 and
                   0.05 mg/kg/day of boron,  respectively.  The value of this data is
                   questionable,  however, since these results have not been repeated
                   elsewhere.

            •      Dixon et al. (1976, 1979)  administered, 0, 500,  1,000 or 2,000 mg/kg of
                   boron (as borax) in the diet to male Sprague-Dawley rats (18/group) for
                   60 days (approximately  25, 50 or 100 mg/kg/day of boron assuming a food

                                           13

-------
Boron                                                                         April 1992
                   consumption of 5% of the body weight, Lehman, 1959).  Testicular and
                   plasma levels of boron were elevated in a dose-dependent fashion at 30
                   and 60 days of treatment. Significant (p < 0.05) decreases in the weights
                   of the liver, testes and epididymis were observed at the 1,000- and 2,000-
                   mg/kg dietary levels.  Seminiferous-tubule diameter was significantly (p <
                   0.05) decreased in a dose-dependent manner in all treatment groups;
                   however, loss of germinal cell elements was observable only at the 1,000-
                   or 2,000-mg/kg dietary level.  Aplasia was complete at the highest dose.
                   The activities of hyaluronidase (H) and sorbitol dehydrogenase  (SDH),
                   testicular enzymes associated with postmeiotic spermatogenic cells, were
                   significantly (p < 0.05) decreased in apparent dose-related fashion for all
                   treatment levels. The activity of a related testicular enzyme, lactic acid
                   dehydrogenase (isozyme-X), was decreased at the 500-mg/kg dietary level
                   but was significantly (p < 0.05) decreased only at the two higher-dose
                   levels.  Plasma levels of follicle-stimulating  hormone (FSH) were
                   significantly (p < 0.05) elevated in a dose-dependent fashion at all dose
                   levels.  Serial mating studies revealed reduced fertility without change in
                   copulatory behavior  only at the two higher-dose  levels. Based upon the
                   finding of a dose-dependent tubular germinal aplasia which was reversible
                   at low doses, this study defined a NOAEL  of 25 mg/kg/day of boron.

             •     Weir and Fisher (1972), in a  multigeneration reproduction study,
                   administered 0, 117, 350 or 1,170 ppm boron (as borax or boric acid) in
                   the diet to groups of 8 male and 16 female Sprague-Dawley rats; these
                   doses corresponded  to approximately 0, 5.9, 17.5 or 585 mg/kg/day of
                   boron,  assuming a food consumption of 5% of the body weigh as
                   recommended by Lehman (1959).  No adverse effects on  reproduction  or
                   gross pathology were observed in the rats dosed with 5.9  or 17.5 mg/kg/
                   day of boron as borax or boric acid. Animals in test groups fed 58.5 mg/
                   kg/day of boron1 as borax or boric acid were found to be sterile.  In
                   addition, males exhibited a lack of spermatozoa in their atrophied testes,
                   and females exhibited decreased ovulation  in the majority of the ovaries
                   examined. This study identified a NOAEL of 175 mg/kg/day of boron.

             Developmental Effects

             •  .   In the multigenerational  study conducted by  Weir and Fisher (1972), in
                   which female rats were administered 117 to 1,170 ppm of boron, no
                   reduction in  the number of living offspring and no physical defects were
                   found.
                                           14

-------
Boron                                                                         April 1992
            Mutagenicity

            •      Boric acid either was not mutagenic (Iyer and Szybalski, 1958; Szybalski,
                   1958) or produced equivocal results (Demerec et al., 1951) in Escherichia
                   coli Sd-4.

            •      Sodium borate and boric acid did not cause gene mutations in the
                   Salmonella typhimurium preincubation assay with and without rat liver S9
                   (Benson et al., 1984).

            •      Similarly, boric acid, both with and without rat and Syrian hamster liver
                   S9 preparations, was negative in an interlaboratory evaluation of the S.
                   typhimurium  preincubation assay (Haworth et al., 1983).

            •      Boric acid did not induce forward mutations in L5178Y mouse lymphoma
                   cells under nonactivated or S9-activated conditions.  Similarly, boric acid
                   did not induce chromosome aberrations or  increase the frequency of sister
                   chromatid exchanges in Chinese hamster ovary cells under nonactivated or
                   S9-activated conditions (NTP, 1987).

            •      Boron was reported to induce mitotic suppression and  chromosomal
                   abnormalities  in Papaver somniferum; no details were provided (Sopova et
                   al., 1981).

            Carcinogenicity

            •      Groups of 50 male and 50 female B6C3F, mice were fed diets containing
                   0, 2,500 or 5,000 ppm boric acid for 103 weeks (approximately 0, 21.9 or
                   43.8 mg/kg/day of boron, assuming a daily food intake of 5% of the body
                   weight). Mortality in the  dosed animals was significantly higher in the
                   treated males  (40 and 56% at 2,500 and 5,000 ppm,  respectively) than  in
                   controls (18%). Body weight gain and food consumption were lower in
                   both sexes. No treatment-related clinical signs of carcinogenicity were
                   observed.  At  5,000 ppm, boric  acid caused an increased incidence  of
                   testicular atrophy and interstitial hyperplasia in males.  Under the
                   conditions of this study, there was no evidence of carcinogenicity for boric
                .   acid in male or female mice (NTP, 1987).
V.    QUANTIFICATION OF TOXICOLOGICAL EFFECTS

      Health Advisories (HAs), are based upon the identification of adverse health effects
associated with the most sensitive and meaningful noncarcinogenic endpoint of toxicity.  The

                                           15

-------
Boron
                                                   April  1992
induction  of this effect is related to a particular exposure dose over a specified period of time,
most often determined from the results of an experimental animal study. Traditional risk
characterization methodology for threshold toxicants is applied in HA development.  The
general formula is as follows:

          UA    (NOAEL or LOAEL)  (BW)           _ .
          HA =  -:	;	'.—:	L  =     mg/L (rounded to  	ug/L)
                       (UF) (    L/day)         	   *  V            	^  '
where:
          NOAEL  =
No-Observed-Adverse-Effect Level (the exposure dose in mg/kg
bw/day).
or
          LOAEL  =     Lowest-Observed-Adverse-Effect Level (the exposure dose in mg/
                           kg bw/day).

              BW  =     assumed body weight of protected individual (10 kg for child or
                           70 kg for adult).

            UF(s)  =     uncertainty factors, based upon  quality and nature of data (10,
                           100, 1,000, or 10,000 in accordance with EPA or NAS/OW
                           guidelines).
            L/day   =     assumed water consumption  (1 L/day for child or 2 L/day for
                           adult).
       One-day Health Advisory

       Although no ideal acute toxicity study was found for calculating the One-day Health
Advisory (HA) for boron, the study of Jansen et al. (1984b) has been selected as providing the
best available data.  In this study, no adverse effects were reported following oral ingestion of
3.6 mg/kg of boron, as boric  acid, by each of six human, males.  Other studies on accidental
ingestion of boric acid solution reported deaths in six out of eight infants (Young et al., 1949)
ingesting 50 to 100 mg/kg of boron; or in five of eleven infants  ingesting 200 to 700 mg/kg of
boron (Wong et al., 1964). In one  study (Martin,  1971), a five-month-old child (6.1 kg)
ingested approximately 1.83 g of boric acid (14 mg/kg) of boron; the child vomited 1 hour later
and exhibited slight lethargy  for about 12 hours. The child was  alert  and well 36 hours after
ingestion.
                                           16

-------
Boron                                                             '          April 1992
       The NOAEL from the Jansen et al., (1984b) study (3.6 mg/kg/day) was used to
calculate the One-day HA for boron. The One-day HA for the 10-kg child is calculated as
follows:


                  One-day HA -  (3-6 mg/kg/day) (10 kg)  =
                         7            (10) (1 L/day)             &

where:

   3.6 mg/kg 1 day   =      NOAEL based on absence of adverse effects in human adults
                           ingesting boric acid.

            10 kg   =      assumed weight of child.

               10   =      Uncertainty factor; this uncertainty factor was chosen in
                           accordance with EPA or NAS/OW guidelines in which  a
                           NOAEL from a human study is employed.

           1 L/day   =      assumed water consumption by a 10-kg child.

       Ten-day Health Advisory

       The available data are insufficient  to develop a Ten-day HA for boron.  Therefore, it
is recommended that the Longer-term HA for a 10-kg child, 0.9 mg/L, be used as a
conservative estimate of the Ten-day HA.

       Longer-term Health Advisory

       The study conducted by Weir and  Fisher (1972)  involving dietary ingestion of boric
acid and borax for 90 days in dogs was selected as the basis for calculation of the Longer-term
HA for boron. In this study, dogs were administered 0,  1.5, 2.9, 8.8 or 29 mg/kg/day of boron
as borax  or boric  acid. This study defined  a NOAEL of  8.8 mg/kg/day, based on the  absence of
reproductive effects, gross pathology, and  testicular histopathology. Higher doses of boron
(23.7 and 29 mg/kg/day) have been shown  to produce testicular atrophy in Long-Evans rats
(Seal and Weeth,  1980) and  in beagle dogs (Weir and Fisher, 1972), respectively. The results
of subchronic studies in rats and dogs (Weir and Fisher, 1972), although well-conducted, were
not considered because histopathology was not completely reported at the lower doses. In
addition, the organ-weight changes observed  in the rats  in lower-dose groups were
inconsistent.  The NTP (1987) study reported extramedullary hematopoiesis at all doses (34 to
752 mg/kg/day of boron) in mice administered boric acid for 13 weeks.  No NOAEL was
defined in this study and, at  the LOAEL of 34 mg/kg/day, mild spleen pathology was evident.
                                          17

-------
Boron
                                                   April 1992
Therefore, the NOAEL identified in the Weir and Fisher (1972) study, 8.8 mg/kg/day, was
used to calculate the Longer-term HA for boron.

       The Longer-term HA for a 10-kg child is calculated as follows:
Longer-term HA
    6
                         <8'8
kg) =
                             (100) (1 L/day)
                     = 08g
                                                             (rounded to 0.9 mg/L)
                                                             v               '&  '
where:
       8.8 mg/kg/day



               100


             10kg

           1 L/day
      NOAEL, based on the absence of reproductive effects, or
      gross pathology and testicular histopathology,  in a chronic
      2-year study in dogs exposed to borax or boric acid.

      uncertainty factor, chosen in accordance with  EPA or
      NAS/OW guidelines for use with a chronic animal study.

      assumed weight of a child.

      assumed water consumption of a 10-kg child.
       The Longer-term HA for a 70-kg adult is calculated as follows:

      Longer-term HA =  (8-8 mg/kg/day)  (70  kg)  =                  ed
         6                   (100) (2 L/day)              &    v               &  '
where:

     8.8 mg/kg/day   =



              100   =


             70kg   =

           2 L/day   =
NOAEL, based on the absence of reproductive effects, or gross
pathology in a chronic 2-year study, in dogs exposed to borax or
boric acid.

uncertainty factor, chosen in accordance with EPA or NAS/OW
guidelines for use with a chronic animal study.

assumed weight of an adult.

assumed water consumption  of an adult.
                                           18

-------
Boron                                                                        April 1992
       Lifetime Health Advisory

       The Lifetime HA represents that portion of an individual's total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure. The Lifetime HA is derived in a three-step process.  Step 1
determines the  Reference Dose (RED), formerly called the Acceptable Daily Intake (ADI).
The RfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily
exposure to the human population (including sensitive subgroups) that is likely to be without
appreciable risk of deleterious health effects during a lifetime, and is derived from the
NOAEL (or LOAEL), identified from a chronic (or subchronic) study, divided by an
uncertainty factor(s). From the RfD, a Drinking Water Equivalent Level (DWEL) can be
determined (Step 2). A DWEL is a medium-specific (i.e., drinking water) lifetime exposure
level,  assuming  100% exposure from that medium, at which adverse, noncarcinogenic health
effects would not be expected to occur. The DWEL is derived from the multiplication of the
RfD by the assumed body weight of an adult and divided by the assumed daily water
consumption of an adult. The Lifetime HA in drinking water alone is determined in Step 3 by
factoring in other sources of exposure, the relative source contribution (RSC). The RSC from
drinking water is based on actual exposure data or, if data are not available, a value of 20% is
assumed.

       If the contaminant is classified as a known, probable, or possible carcinogen, according
to the Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986), then caution
must be exercised in making a decision on how to deal with possible lifetime exposure to this
substance.

       For human (A) or probable (B) human carcinogens, a Lifetime HA is not
recommended.  For possible (C) human carcinogens, an additional 10-fold safety factor is used
in the calculation of the Lifetime HA. The risk  manager must balance this assessment of
carcinogenic potential and the quality of the data against the likelihood of occurrence and
significance of health effects related to noncarcinogenic endpoints of toxicity.  To assist the
risk manager in this process, drinking water concentrations associated with estimated excess
lifetime cancer  risks over the range of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult
drinking 2 L of water/day are provided  in the Evaluation of Carcinogenic Potential section.

       The chronic toxicity study  in dogs by Weir and Fisher (1972) was selected to serve as
the basis for calculation  of the Lifetime HA for boron.  Other chronic studies  were
considered, including a chronic study with rats (Weir and Fisher, 1972) and a lifetime study
with mice (Schroeder and Mitchener, 1975).  However, limitations or  deficiencies in these
studies precluded their use in the  development of the Lifetime HA for boron.

       Weir and Fisher (1972) reported a NOAEL for boron of 8.8 mg/kg/day in dogs fed
boric  acid in the diet for 2 years.  This NOAEL was selected because there is  clear evidence
that the dog is more sensitive than the rat to boron.

                                           19

-------
Boron                                                                       April 1992
       Schroeder and Mitchener (1975) reported no effects on longevity or body weight in
mice administered 8.1 mg/kg/day of boron (as metaborate) for a lifetime.  This study was not
selected because a limited number of parameters were tested and because only one dose level
was tested.

       In the study selected as a basis for development of the Lifetime HA for boron (Weir
and Fisher, 1972), dogs were administered 0, 1.5, 2.9, 8.8 or 29 mg/kg/day of boron as borax or
boric acid for 2 years. This study defined a NOAEL for boron of 8.8 mg/kg/day based on the
absence of either testicular atrophy or spermatogenic arrest at this level of exposure.

       Using the NOAEL from this study, identified in the chronic dog study by Weir and
Fisher  the Lifetime HA is derived as follows:

Step 1:  Determination of the RfD

         RfD  = (8.8 mg/kg/day)  = Q 08g mg/kg/day  (rounded to 0.09  mg/kg/day)
                     (100)

where:

     8.8 mg/kg/day   =      NOAEL, based on the absence of gross on light-microscopic
                           pathology in dogs exposed to borax or boric acid in the diet  for 2
                           years.

               100   =      uncertainty  factor, chosen in accordance with EPA or NAS/OW
                           guidelines for use with a NOAEL from a chronic animal study.

Step 2:  Determination of the DWEL

          DWEL . (0.09 mg/kg/day) (70 kg) =             ^
                           (2 L/day)                 &   ^               & '

where:

    0.09 mg/kg/day   =      RfD.

            70 kg   =      assumed weight of an adult.

          2 L/day   =      assumed water consumption of a 70-kg adult.
                                          20

-------
Boron                                                                        April 1992



Step 3:  Determination of the Lifetime HA

               Lifetime HA = (3.15 mg/L)  (20%) = 0.63 mg/L  (0.6 mg/L)

where:

        3.15 mg/L   =     DWEL.

              0.20   =     assumed relative source contribution.

       Evaluation of Carcinogenic Potential

       •     According to the U.S. EPA classification scheme for carcinogenic potential
             (U.S. EPA, 1986), boron is classified as Group D: not classifiable.

       •     No quantitative assessment of excess cancer risk attributable to boron has been
             reported.

       •     The NTP (1987) conducted a bioassay with B6C3Ft mice fed diets containing
             boric acid at levels of 0, 2,500 or 5,000 ppm for 103 weeks. No evidence of
             carcinogenicity was found in male or female mice.

       •     Boric acid has not been found to be mutagenic in bacterial  and mammalian cell
             assays.


VI.    OTHER CRITERIA. GUIDANCE AND  STANDARDS

       •     The U.S. EPA has established tolerances for total boron of 30 ppm in or on
             cottonseed and of 8 ppm in or on citrus fruits (U.S. EPA, 1987b).


VII.   ANALYTICAL METHODS

       •     Methods for metal analysis involve spectroscopy, either emission or absorption.
             In all of these methods, the metal is dissolved and then thermally excited.  All
             elements when excited emit or absorb light frequencies characteristic of that
             element. Most metal spectroscopy is done in the ultraviolet and x-ray regions.

       •     Direct Aspiration Atomic Absorption Spectroscopy (AA).  In this  technique,
             dissolved metals are aspirated into a flame source and excited to the point of
             the dispersion into a mono-atomic state; a light source whose cathode is the
             metal of interest passes through the flame, and the resulting absorption of light

                                          21

-------
Boron                                                                        April 1992
             by the element of interest is directly proportional to its concentration.
             Disadvantages of this technique include the inability to analyze more than one
             metal at a time and the insensitivity of the technique.

             Graphite Furnace Atomic Absorption (GFAA). In this technique, a specific
             amount of liquid is dried on the thermal source, effecting a concentration step.
             The sample is then electrothermally excited.  This technique is very sensitive.

             Inductively Coupled Plasma Atomic Emission Spectroscopv. This method (EPA
             Method 200.7) utilizes aspiration of a liquid sample, but the flame is actually a
             plasma torch  of Argon excited to super hot levels by radio- frequency radiation.
             In this technique,  the metals are excited sufficiently to emit radiation.  By using
             classical dispersion-grating  optics, a large number of metals can be analyzed
             simultaneously.  This technique has a detection limit of 1.25
             Inductively Coupled Plasma Mass Spcctrometrv (ICP/MSV In this technique,
             the excitation is effected using a radio- frequency plasma, but the excited atoms
             are interfaced into a mass spectrometer. Quantitation is achieved by
             computerized software programs similar to those used in other U.S. EPA MS
             organic methods.
VIII.   TREATMENT TECHNOLOGIES

       •     Available data indicate that reverse osmosis (RO), electrodiaiysis (ED/EDR),
             lime softening, ion exchange (IX) and granular activated carbon (GAC)
             adsorption can significantly reduce boron levels in contaminated drinking water
             supplies.

       •     Jarrett (1978) reported that an RO treatment designed for bakery use reduced
             boron levels by 60% (from an initial concentration of 0.3 mg/L).  This system
             was designed to treat 20,000 gpd of water at an operating pressure of 600 psi
             and a water recovery rate of 75%.  The source water was well water with a total
             hardness of 1,830 mg/L.

       •     Folster et al. (1980)  reported boron removal by RO  and ED/EDR treatment
             plants in San Jon and Alamogordo, New Mexico.  The RO systems in both
             locations consisted of hollow-fiber  (HF) and spiral-wound  (SW) configuration
             membranes. The HF RO was operated  at 515 psi and a water recovery rate of
             78%, while the SW RO was operated at 430 psi and a water recovery rate of
             79%.  The ED/EDR plants were operated at a water recovery rate of 85%.  At
             the San Jon plant, boron was present in the influent at an  initial concentration
             of 0.8 mg/L. ED/EDR treatment removed 20% of the boron, while HF RO

                                          22

-------
Boron                                                                        April 1992
             treatment removed 15% of the boron. The SW RO system did not remove any
             boron.  At the Alamogordo site, ED/EDR treatment removed 50% of the boron
             (from an initial concentration of 0.01 mg/L), and SW RO treatment removed at
             least 94% of the boron (from an initial concentration of 0.09 mg/L). No
             removal was achieved by HF RO.

       •   .  Potable water was reclaimed from unchlorinated effluent by an activated sludge
             wastewater treatment plant using lime softening, reacidification and RO at
             Water Factory-21 (WF 21) in Orange County, California (Argo, 1984). The
             influent was softened at pH 11. The softened water was carbonated to pH 8.0,
             filtered and passed through an RO system.  Two 5,000 gpd RO units with
             tubular aromatic polyamide membranes arranged in a 2-1 array configuration
             were used.  The average flux rate was 7.14 gpd/sq ft at 250 psi applied pressure.
             The lime/RO process removed 80% of the boron (from an initial average
             influent concentration of 0.55 mg/L).

       •     Wong (1984) studied the use of two different IX resins to reduce boron levels in
             water from a utility plant, thereby reclaiming it for  potable use.  A boron-
             specific anion exchange resin  (IRA-743) and a strong base anion exchange resin
             were tested.  Both resins tested removed boron to below 0.1 mg/L from an
             initial concentration of 10 mg/L. The IX columns contained 50 ft, of resin and
             were operated at a loading rate of 2 gpm/ft3. Both types of resins were
             regenerated with 4 Ib/ft3 of a sodium hydroxide solution.  IRA-743 provided an
             exchange capacity of 0.125 Ib  boron/ft3 resin, while  the strong base anion resin
             provided an  exchange capacity of 0.36 Ib boron/ft3 resin.

       •     Choi and Chen (1979) developed batch adsorption  isotherms for a variety of
             commercially available GAC systems.  Regardless of the characteristics of the
             background solution, a boron removal efficiency of approximately 90% can be
             achieved with Filtrasorb* carbon at a dose of 25 g/L if the initial boron
             concentration does not exceed 5 mg/L.
                                          23

-------
Boron                                                                        April 1992
IX.    REFERENCES

ACT. 1983.  American College of Toxicology, expert panel.  Final report on the safety
       assessment of sodium borate and boric acid.  J. Am. Coll. Toxicol.  2:87-125.

Aldrich Chemical Co., Inc. 1986. Aldrich catalog handbook of fine chemicals.  Milwaukee,
       WI: Aldrich Chemical Co., Inc.

Argo, D.R.  1984.  Use of lime clarification and reverse osmosis in water reclamation.  Water
       Pol. Con. Fed.  56(12): 12381246.

Benson, W.H., WJ. Birge and H.W. Dorough.  1984. Absence of mutagenic activity of
       sodium borate (borax) and boric acid in the Salmonella preincubation test.  Environ.
       Toxicol. Chem. 3:209-214.

Choi, W. and K.Y. Chen.  1979.  Evaluation of boron removal by adsorption on solids.
       Environ. Sc. Tech. 13(2):189-196.

Demerec, M., G. Bentani and J. Flint.  1951. A survey of chemicals for mutagenic action on
       E. coli.  Am. Nat. 85:119-136.

Dixon, R.L., I.P. Lee and RJ. Sherins.  1976.  Methods to assess reproductive effects of
       environmental chemicals. Studies of cadmium and boron administered orally.
       Environ. Health Perspect.  13:59-67.

Dixon, R.L., RJ. Sherins and I.P. Lee.  1979.  Assessment of environmental  factors affecting
       male fertility. Environ. Health Perspect. 30:53-68.

Draize, J.H. and E.A. Kelly.  1959.  The urinary excretion of boric acid preparations following
       oral administration and topical applications to intact and damaged skin of rabbits.
       Toxicol. Appl. Pharmacol. 1:267-276.

Folster, H.G., D.B. Wilson, S. Hanson and R. Duran.  New Mexico Water Resources
       Research Institute. 1980. Water treatment for small public supplies. WRRI Report
       no.
Forbes, R.M. and.H.H. Mitchell.  1957. Accumulation of dietary boron and strontium in
       young and albino rats.  A.M.A. Arch. Ind. Health 16:489-492.

Gordon, A.S., J.S. Prichard and M.H. Freedman. 1973. Seizure disorders and anemia
       associated with chronic borax intoxication.  Can. Med. Assoc. J. 108:719-724.
                                          24

-------
Boron                                                                        April  1992
Green, G.H., M.D. Lott and H.J. Weeth. 1973. Effects of boron-water on rats.  Proc. West.
       Sec. Am. Soc. Anim. Sci. 24:254.

Grella, P., B. Tambuscio and V. Suma.  1976.  Boric acid poisoning during pregnancy.
       Description of one case.  Acta Anaesthesiol. Italica 27:745-748. (In Italian;
       translation.)

Hamilton, E.I. and M.J. Minski.  1972. Abundance of the chemical elements in man's diet
       and possible relations with environmental factors. Sci. Tot. Environ.  1:375-394.

Hawley, G.G.  1981.  Condensed chemical dictionary, 10th ed.  New York: Van Nostrand
       Rheinhold Company,  pp. 143-145.

Haworth, S., T. Lawlor, K. Mortelmans, W. Speck and E. Zeiger.  1983.  Salmonella
       mutagenicity test results for 250 chemicals.  Environ. Mutagen. (Suppl.) 1:3-142.

K.mphill, D.D., K.K. Roberts and T.E. Clevenger. 1981. 3rd International conference on .
       heavy metals  in the environment,  pp. 391-394.

Iyer, V.N. and W. Szybalski. 1958. Two simple methods for the detection of chemical
       mutagens.  Appl. Microbiol.  6:23-29.

Jansen, J.A., J. Andersen and J.S. Schou. 1984a.  Boric acid single dose pharmacokinetics
       after intravenous administration to man. Arch. Toxicol.  55:64-67..

Jansen, J.A., J.S. Schou and B. Aggerbeck.  1984b. Gastrointestinal absorption and in vitro
       release of boric acid from water-emulsifying ointments.  Food  Chem. Toxicol.  22:49-
       53.

Jarrett, R.  1978. Reverse osmosis in  modern water treatment system for bakers.  The Bakers
       Digest, Feb.,  pp. 44-50.

Kent, N.L. and R.A.  McCance.  1941. The  absorption and excretion of minor elements by
       man.  I.  Silver, gold, lithium, boron and vanadium. Biochem.  J.  35:837-844.

Krasovskii, G.N., SJ*. Vashavskaya and A.I. Borisov.  1976.  Toxic and gonadotropic  effects of
       cadmium and boron relative to standards for these substances in drinking water.
       Environ. Health Perspect. 13:69-75.

Lehman, AJ.  1959.  Appraisal  of the safety of chemicals in foods, drugs and cosmetics.
       Association of Food and Drug Officials of the United States.
                                           25

-------
Boron                                                                        April 1992
Linden, C.H., A.H. Hall, K.W. Kulig and B.H. Rumack.  1986. Acute ingestions of boric acid.
       Clin. Toxicol.  24:269-279.

Linder, R.E., L.F.  Strader and G.L. Rehnberg.  1990.  Effect of acute exposure to boric acid
       on the male reproductive system of the rat. J.  Tox. and Environ.  Health 31:133-146.

Locksley, H.B. and W.H. Sweet.  1954. Tissue distribution of boron compounds in relation to
       neutron-capture  therapy of cancer. Proc. Soc.  Expt. Biol. Med. 86:56-63.

Martin, G.I.  1971.  Asymptomatic  boric acid intoxication.  N.Y. State J. Med.  71:1,842-1,844.

Mulinos, M., C. Connant and E. Hauser. 1953. The toxicity of boric acid and the clinical
       implications of the use of borated baby powders.  Bull. N.Y. Med. Coll.  16:92-101.

Nielsen, F.H., CD. Hunt, L.M. Mullen and J.R. Hunt.  1987.  Effect of dietary boron on
       minerals, estrogen and testosterone metabolism in postmenopausal women.  FASEB J.
       1:394-397.

NTP. 1987.  National Toxicology Program. Toxicology and carcinogenesis studies of boric acid
       (CAS No. 10043-35-3) in B6C3FL mice (feed studies). NTP technical report series no.
       324.  Research Triangle Park, NC: National Toxicology Program.

Perry, R.H. and C.H. Chilton.  1973. Chemical engineers handbook, 5th ed. New York, NY:
       McGraw Hill Book Co.

Pfeiffer, C.C., L.F. Hallman and  I.  Gersh.  1945.  Boric acid ointment.  A study of possible
       intoxication in the treatment of burns.  J. Am.  Med. Assoc.  128:266-274.

Roudabush,  R.L., DJ. Terhaar, D.W. Fasset and S.P. Dziuba.  1965. Comparative acute
       effects of some chemicals on the skin of rabbits and guinea pigs.  Toxicol. Appl.
       Pharmacol.  7:559-565.

Schillinger, B.M., M. Berstein, L.A. Goldberg and A.R. Shalita. 1982. Boric acid poisoning.
       Am.  Acad. Dermatol.   7:667-673.

Schroeder, H.A. and M. Mitchener.  1975. Life-term effects of mercury, methyl mercury and
       nine other trace  metals on mice. J. Nutrit. 105:452-458.

Seal, B.S. and H.J. Weeth.  1980. Effect of boron in drinking water on  the male laboratory
       rat.  Bull. Environ. Contam. Toxicol.  25:782-789.

Settimir L., E. Elovaara  and H. Savolainen.  1982. Effects of extended peroral borate
       ingestion on rat  liver and brain.  Toxicol. Lett.  10:219-223.

                                           26

-------
Boron                                                                        April 1992
Silaev, A.A., A.A. Kasparov, V.V. Korolev and V.V. Nebstrueva.  1977. Electron-microscopic
       investigation of the effect of boric acid on the seminiferous tubules of albino rats.
       Bull. Exp. Biol. Med. (U.S.S.R.) 83:588-591.

Sittig, M.  1981.  Handbook of toxic and hazardous chemicals. Park Ridge, NJ: Noyes
       Publications, pp. 94-96.

Sopova, M., D. Petrovska, A. Musalevski, C. Najcevska and Z. Sekovski. 1981. Cytogenetical
       and morphological effect of general toxicity caused with different concentrations of
       boron. Mutat. Res. 85:229.

Stokinger, H.E.  1981.  The halogens and the nonmetals boron and silicon. In: Clayton, G.D.
       and F.E.  Clayton, eds.  Patty's industrial hygiene and  toxicology, 3rd  ed.  New York,
       NY: John Wiley and Sons, pp. 2,978-3,005.

Szybalski,  W.  1958. Special microbiological system. II.  Observations on chemical
       mutagenesis in microorganisms. Ann. N.Y. Acad. Sci. 76:475-489.

Treinen, K.A. and R.E.  Chapin.  1991. Development of testicular lesions in  Fischer-344 rats
       after treatment with boric acid. Tox. Appl. Pharm. 107:325-335.

U.S. EPA. 1987a.  U.S. Environmental Protection Agency. Health effects assessment for
       boron and compounds.  Report no. EPA/600/888/021. Washington,  DC:  U.S.
       Environmental Protection Agency.

U.S. EPA. 1987b.  U.S. Environmental Protection Agency. Boron: Tolerances for residues.
       Code of Federal Regulations.  40 CFR  180.271.  July  1.

U.S. EPA. 1986. U.S. Environmental Protection Agency.  Guidelines for carcinogen risk
       assessment.  Fed. Reg. 51(185):33992-34003.

U.S. EPA. 1980. U.S. Environmental Protection Agency.  Biological  monitoring of toxic trace
       elements. Report no.  EPA-600/380-090. Washington, DC: U.S. Environmental
       Protection Agency.

U.S. EPA. 1975. U.S. Environmental Protection Agency.  Preliminary investigation of effects
       on the. environment of boron, indium, nickel, selenium, tin, vanadium and their
       compounds.  Vol. 1:  Boron.  Report no. EPA-560/2-75-005A. Washington, DC: U.S.
       Environmental Protection Agency,  Office of Toxic Substances.

Vignec, AJ. and R. Ellis.  1954. Inabsorbability of boric acid in infant powder.  Am. J. Dis.
       Child. 88:72-80.
                                           27

-------
Boron                                                                      April 1992
Wang, E., X. Xun, H. Wang, Y. Chen, X. Meng, S. Zhang, J. Huang, S. Cong and G. Tao.
       1984. Toxicological effect of boron  in laboratory rats.  Zhonghua Yufangyixue Zazhi
       18(l):20-22.  (In Chinese; translation.)

Weast, R.C.  1988. CRC handbook of chemistry and physics, 69th ed.  Boca Raton, FL: CRC
       Press, Inc.  pp. F-143-146.

Weast, R.C.  1981-1982.  CRC handbook of chemistry and physics, 62nd ed. New York, NY:
       McGraw Hill Book Co.

Weir, RJ. Jr. and R.S. Fisher.  1972.  Toxicologic studies on borax and boric acid.  Toxicol.
       Appl. Pharmacoi. 23:351-364.

White, K.L. 1980. Health implications of new energy technology.  In: Rom, W.N. and V.E.,
       Archer,  eds. Ann Arbor, Michigan:  Ann Arbor Science, pp 553-564.

Windholz, M., S. Budavari, R.F. Bluminetti and E.X. Ottenbein, eds. 1983. The Merck index
       — An encyclopedia of chemicals, drugs and biologicals, 10th ed. Rahway, NJ: Merck
       and Company, Inc.,  pp.  1,324-1,325.

Wong, J.M.  1984. Boron control in power plant reclaimed water for potable reuse. Environ.
       Prog. 3(1):5-11.

Wong, L.C., M.D. Heimbach, D.R. Truscott and B.D. Duncan. 1964.  Boric acid poisoning:
       Report of 11 cases. Can. Med. Assoc. J. 90:1,018-1,023.

Young, E.G., R.P. Smith and O.C. Macintosh.  1949. Boric acid as a poison. Report of six
       accidental  deaths in infants.  Can. Med.  Assoc. J. 61:447-450.

Zook, E.G. and J. Lehman.  1965.  Total diet study:  Content of ten minerals—-aluminum,
       calcium, phosphorus, sodium, potassium, boron,  cooper, iron, manganese and
       magnesium.  J. Assoc. Off. Agric. Chem. 48:850-855.
                                          28

-------
SPA  0553

RX000027511

                                                                               April  1992
                                    CHLORPYRIFOS

                              Drinking Water Health Advisory
                                     Office of Water
                           U.S. Environmental Protection Agency
 I.     INTRODUCTION

       The Health Advisory (HA) Program, sponsored by the Office of Water (OW), provides
 information on the health effects, analytical methodology, and treatment technology that would
 be useful in dealing with the contamination of drinking water. Health Advisories describe
 nonregulatory concentrations of drinking water contaminants at which adverse health effects
 would not be anticipated to occur over specific exposure durations. Health Advisories contain a
 margin of safety to protect sensitive  members  of the population.

       Health Advisories serve as informal technical guidance to assist Federal,  State, and local
 officials responsible for protecting public health when emergency spills or contamination
 situations occur. They are not to be construed as legally enforceable  Federal standards. The
 HAs are subject to change as new information becomes available.

       Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
 years,  or 10% of an individual's lifetime),  and Lifetime exposures based on data describing
 noncarcinogenic endpoints of toxicity. For those substances that are known or probable human
 carcinogens, according to the Agency classification scheme (Group A or B), Lifetime HAs are
 not recommended.  The chemical concentration values for Group A or B carcinogens are
 correlated with carcinogenic risk estimates by employing a cancer potency (unit risk) value
 together with assumptions for lifelong exposure and the ingestion of water. The cancer unit risk
 is usually derived from a linearized multistage model with 95% upper confidence limits.  This
 provides a low-dose estimate of cancer risk to humans that is  considered unlikely to pose a
 carcinogenic risk in excess of the stated values. Excess cancer risk estimates may also be
 calculated using the one-hit, Weibull, ibgjt, or probit models.  There is no current understanding
 of the biological mechanisms involved in cancer to suggest that any one of these models is able
 to predict risk more accurately than another.   Because each model is  based on differing
 assumptions, the estimates that are derived can differ by several orders of magnitude.
 II.    GENERAL INFORMATION AND PROPERTIES

       CAS No.      2921-88-2

-------
Chlorpyrifos                                                                   April 1992
       Structural Formula
                                                 3   J~^

                                      Chlorpyrifos

       Synonyms

       •     Brodan; Chlorpyrifos; Chlorpyrifos-ethyl; Detmol U.A.; Dowco 179; Dursban;
             Dursban F; Ent 27311; Eradex; Ethion, dry; Lorsban; NA 2783 (Dot); OMS-0971;
             Phosphorothioic acid, O,O-diethyl O-(3,5,6-trichloro-2-pyridyl) ester; Pyrinex.

       Uses

       •     Chlorpyrifos is a broad-spectrum insecticide with many uses. An estimated 7 to
             11 million pounds of Chlorpyrifos are produced each year in the United States for
             domestic use.  Of the total domestic Chlorpyrifos usage, 57% is applied to corn
             and 5 to 6% to cotton. Commercial pest control and lawn and garden services
             consume 20 to 22% of the annual Chlorpyrifos usage, followed by domestic
             household and lawn and garden application (9 to 13%).

       Properties (Kenaga, 1980; Windholz et al., 1983; Worthing, 1987)
       Chemical Formula
       Molecular Weight                        350.57
       Physical State (25ฐC)                     White,  granular crystals
       Boiling Point                             —
       Melting Point                            41 to 43.5ฐC
       Density                                 —
       Vapor Pressure (258C)                    1.87 x 10's mm Hg
       Water Solubility (25ฐC)                   2 ppm
       Specific Gravity                           —
       Log Octanol/Water Partition               4:99
        Coefficient (log K^)
       Taste Threshold (water)                   —
       Odor Threshold (water)                   —
       Conversion  Factor
        (ppm  air as mg/m3)                      —

-------
Chlorpyrifos                                                                     April 1992
       Occurrence

       •      Based on a 3-year, 20-city nationwide study conducted by the Food and Drug
              Administration, Gartrell et al. (1985) estimated that the average daily intake of
              chlorpyrifos from food and beverages (including water) is approximately 0.001 to
              0.005 /zg/kg.  Contaminated grain and cereal products and garden  fruits were the
              food groups through which exposure occurred.

       Environmental Fate

       •      Chlorpyrifos hydrolyzes readily in water, its rate of hydrolysis increases with
              temperature (Worthing,  1987). When mixed with distilled water (pH 6.5) or
              pasture water (pH 7 to 9), chlorpyrifos levels dropped an average  of 25% at 24ฐC
              and 48% at 38ฐC within 8 hours (Schaefer and Dupras, Jr., 1969).

       •      Chlorpyrifos administered at a rate of 3.4 kg chlorpyrifos/hectare (ha) dissipated
              fairly rapidly in sand and organic muck soils with respective half-lives of 2 and 8
              weeks in the top 15 cm of soil (Chapman and Harris, 1980). Low levels of
              chlorpyrifos (2 to 3% of the amount applied) remained in both soils for up  to 2
              years.  3,5,6-Trichloro-2-pyridinol was the primary degradation product, reaching
              maximum concentrations of 13 and 39% of the chlorpyrifos applied to the sand
              and muck soils, respectively.  The concentrations of the oxygen analog of
              chloropyrifos were ฃ 0.004 ppm in all samples.  Chlorpyrifos (EC, eraulsifiable
              concentrate), applied at 4 kg chlorpyrifos/ha to turf grass, dissipated rapidly with
              a half-life of <14 days in the soil and turf cover (Sears and Chapman, 1979).
              Movement of chlorpyrifos from the turf into the soil was minimal  (<18% of the
              recovered chlorpyrifos at any time during the study).

       •      Breakdown of chlorpyrifos in soil primarily results from microbial metabolism
              (Miles et al., 1979).  Chlorpyrifos (10 ppm) is degraded more rapidly in sandy
              loam soil (half-life, <1 week)  than in organic soil (half-life, 25 weeks).  In
              sterilized soils, the half-life for chlorpyrifos is >17 weeks.  Half-lives of 11 to 141
              days were reported in another study in soils ranging in texture from loamy sand
              to clay (Bidlack, 1979).  3,5,6-Trichloro-2-methoxypyridine and two unidentified
              minor metabolites of chlorpyrifos were recovered after a 1-year incubation
              period; most of the radiocarbon, however, was  recovered as 14CO2 with small
              amounts incorporated into soil organic matter.

       •      After  30 days of aerobic aging of soil, uC-chlorpyrifos degraded with half-lives of
              15 days  in loam and 58 days in clay soils. The half-lives in treated and
              anaerobically incubated loam  and clay were 39 and 51 days, respectively. The
              major degradation product formed was 3,5,6-trichloro-2-pyridinol. Degradation
              of this compound was very slow.  Evolution of 14CO2 was insignificant, and
              incorporation of UC into the soil organic matter was slow. Relatively low levels

-------
Chlorpyrifos                                                                    April 1992
              (<3%) of 3,5,6-trichloro-2-methoxypyridine and two unidentified metabolites
              were present in small amounts of the samples (Bidlack, 1979).

              Chlorpyrifos was  immobile in loamy sand and sandy loam soil; only 2 to 13% of
              the applied radioactivity leached out of the zone of application with 6 to 10
              inches of water.  Mobility in beach sand was low.  After leaching with 10 inches
              of water, 78.9% of the surface-applied Chlorpyrifos remained in the top inch of
              sand (Dow Chemical Company, 1972).

              Chlorpyrifos was  not persistent in pond water treated at 0,05 Ib active ingredient
              per acre (a.i./a) or in a polluted aquatic  environment treated at 0.23 to 0.27 Ib
              a.i./a (Schaefer and Dupras, 1970; Madder, 1977). The rates of decline were  not
              determined, and losses to underlying segments were not investigated.
              Chlorpyrifos applied to pond or rice floodwater as a slow-release formulation
              (chlorinated polyethylene  pellets) exhibited no patterns of decline in 22 weeks
              (Nelson and Evans, 1973). The concentration of Chlorpyrifos was extremely
              variable in the top 1 inch  of pond sediment and rice plot soil; however, there was
              a clear trend toward the partitioning of Chlorpyrifos  from water onto soil and
              sediments.

              14C-Chlorpyrifos residues found in wheat, soybeans and beets planted 119  days
              after treatment of loamy sand soil with 14C-chlorpyrifos at 2 Ib a.i./acre amounted
              to 0.31, 0.31 and 0.03 ppm Chlorpyrifos equivalents, respectively.  Chlorpyrifos
              was  largely degraded in the soil before the crops were  planted, however, and  the
              plant residues consisted primarily of unidentified UC residues. Residues in wheat
              and soybeans concentrated in the vegetative portions of the plants (Bauriedel et
              al., 1976).
III.    PHARMACOKINETTCS

       Absorption

       •      Chlorpyrifos (unlabeled, 99.8% pure) was readily absorbed from the
              gastrointestinal (GI) tract in six men given a single oral dose at 0.5 mg/kg (Nolan
              et al., 1984). Absorption was estimated to be approximately 70% over a 5-day
              period. Blood Chlorpyrifos levels remained low (<30 ng/mL) throughout the
              study. Mean blood concentrations of the principal metabolite of Chlorpyrifos,
              3,5,6-trichloro-2pyridinol  (3,5,6-TCP), peaked at 0.93 /ig/mL 6 hours after
              ingestion. There was a 1- to 2-hour delay in the absorption of the oral dose.

       •      Approximately 90% of a single oral dose of 50 mg ^Cl-chlorpyrifos/kg (in corn
              oil) was absorbed from the GI tract of male Wistar rats within 2 to 3 days after
              dosing (Smith et  al., 1967).

-------
Chlorpyrifos                                                                     April 1992
       •      A single oral dose of uC-chlorpyrifos (2,6-ring-labeled) (19.1  mg/kg, 23.5 /xCi/mg;
              or 6.9 mg/kg, 48.4 /iCi/mg) was absorbed rapidly by male Sprague-Dawley rats
              (Dow Chemical Company, 1972).  At least 73% of the l4C-chlorpyrifos was
              absorbed by the rats within 3 days after administration. Only 1.6 to 2.5% of the
              administered radioactivity remained in the tissues and carcass at 72 hours
              postdosing; blood I4C levels peaked 1 to 3 hours after dosing, accounting for 3 to
              6% of the ingested 14C-chlorpyrifos.  About 68 to 76%, 5 to 15%, and 0.15 to
              0.63% of the administered 14C was eliminated in the urine, feces and expired air,
              respectively, within 72 hours. The authors reported that absorption may have
              been slightly reduced in some animals as a result of predose starvation and
              frequent bleeding at 2-hour intervals.

       •      Less  than 3% of single doses of analytical grade (unlabeled, 99.8% pure)
              chlorpyrifos (5.0 mg/kg, dissolved in dipropylene glycol methyl ether or methylene
              chloride) was absorbed 7 days after dermal application to six  men (Nolan et al.,
              1984).  Blood levels of 3,5,6-TCP, which were used to determine absorption  and
              clearance rates of chlorpyrifos,  peaked at 0.063 /ig/mL 24  hours post-dosing. The
              average half-life for the appearance of 3,5-6-TCP in the blood was 22.5 hours.

       Distribution

       •      Because of the rapid  elimination of chlorpyrifos and its metabolites following
              administration  of a single oral dose of 0.5 mg/kg to six men (Nolan et al., 1984),
              they  are not expected to accumulate to any appreciable extent in  humans.

       •      The highest levels of radioactivity in male Wistar rats given a single oral dosage
              of MCl-chlorpyrifos (50 mg/kg) were recovered at 4 hours  post-dosing in the
              kidneys, liver, lung and fat (0.0924, 0.0690, 0.406 and 0317 mmol radioactive
              equivalents/kg tissue, respectively) (Smith et al., 1967). Radioactivity was
              eliminated  rapidly from the liver (tw, 10 hours), kidney (tw, 12 hours) and
              muscle (tw, 16 hours) but was retained for a longer period of time by fat tissue
              (tw, 62 hours).

       •      Tissue 14C residue levels were low (<1 ppm) 72 hours after male Sprague-Dawley
              rats were given a single oral dosage of [2,6-l4Cpyridyl]chlorpyrifos (19.1  mg/kg;
              23.5 /xCi/rag) (Dow Chemical Company, 1972).  Fat and intestines contained the
              highest levels of radioactivity (approximately 0.757 and 0.363  ppra, respectively);
              brain 14C residue concentrations were <0.010 ppm.

       Metabolism

       •      Very low levels (<30 mg/mL) of unchanged chlorpyrifos were found in the blood
              and no parent  compound was recovered in the urine during the 5 days after six
              men  were given a single oral dose (0.5 mg/kg) of the pesticide (Nolan et al.,

-------
Chlorpyrifos                                                                     April 1992
              1984).  Most of the chlorpyrifos was converted to 3,5,6-TCP; however, other
              metabolites were not identified.

              In studies ot" workers exposed occupationally to chlorpyrifos, several urinary
              metabolites of the insecticide were identified by gas chromatography.  O,O-
              Diethyl phosphate was found in 96% of the urine specimens, and O,O-diethyl
              phosphorothionate was recovered from 28% of the samples (Hayes et al., 1980;
              Lores and Bradway,  1977).  Hayes et al. (1980) reported that 8-hour exposure
              levels were <27.6 /zg/m3 for workers exposed to spray emulsions of Dursban E-2
              containing 23.5% chlorpyrifos.

              Two major metabolites, 36Cl-3,5,6-trichloro-2-pyridyl phosphate and MCl-3,5,6-
              TCP, were recovered from the urine and feces of male Wistar rats administered a
              single oral dose of 50 mg 36Cl-chlorpyrifos/kg (in corn oil) (Smith et al., 1967).

              Male Sprague-Dawley rats given a single  oral dose of 14C-chlorpyrifos (19.1
              mg/kg, 23.5 /zCi/mg)  excreted 3,5,6-TCP as the major metabolite and  another
              unidentified compound in the urine (Dow Chemical Company, 1972).  A total of
              1% of the  14C recovered in expired air, almost all of which was UC-CO2,
              suggesting that some cleavage of the pyridyl ring had occurred.

              In an in vitro study using rat hepatic microsomes, 14C-chlorpyrifos (10 mg/mL,
              10.6 mCi/mmol) was readily metabolized  to 3,5,6-TCP (Dow Chemical Company,
              1972).  No other  metabolites were found. The reaction was NADPH-dependent,
              and binding of chlorpyrifos to microsoraes occurred prior to catabolism.  These
              findings were also noted in studies by Sultatos et al. (1981, 1982, 1985) and
              Sultatos and Murphy (1983). They indicated that chlorpyrifos may  be
              metabolized by a glutathione-mediated process. Male Charles River Swiss mice
              injected with chlorpyrifos (70 mg/kg) displayed a "moderate but transient"
              depletion of hepatic  glutathione (Sultatos et al., 1982).

              Mostafa et al. (1983) reported that the in vivo  alkylating activities of l-14C-ethyl-
              labeled chlorpyrifos were high following intraperitoneal injection of 5- or 15-
              rng/kg doses in male mice (strain  not given). Labeled 7-ethylguanine found in
              hepatic RNA hydrolysates measured approximately 5.5 x 10 3% of the
              administered radioactivity.  The two major unidentified radioactive peaks
              associated with hepatic DNA hydrolysates corresponded to 3 x 10"%  and 2.3 x
              10"3% of the applied 14C dose. The authors reported that the total incorporation
              of 14C into mouse liver nucleic acids was greater for RNA than for DNA. In
              addition, the degree  of 14C-incorporation  appeared to be dose related.

-------
Chlorpyrifos                                                                    April 1992
       Excretion

       •      Within 5 days after ingesting a single oral dose of chlorpyrifos (0.5 mg/kg), six
              men eliminated an average of 70% of the administered insecticide via the urine,
              with a urinary elimination half-life of 27 hours (Nolan et al., 1984). Fecal
              elimination of chlorpyrifos and/or its metabolites was not measured.

       •      Smith et al. (1967) reported that approximately 90% of a radioactive dose of MC1-
              chlorpyrifos (50 mg/kg) administered orally to male Wistar rats was excreted in
              the urine within 2 to 3 days. The remaining 10% was eliminated in the feces.

       •      Approximately 68 to 70%, 14 to 15%, and 0.15 to 0.39% of a single oral dose of
              14C-chlorpyrifos (19.1 mg/kg, 23.5  ^Ci/mg) administered to two male Sprague-
              Dawley rats were eliminated in the urine, feces and exhaled air, respectively,
              within 72 hours after dosing (Dow Chemical Company, 1972).  Thus, the urine
              provided the primary route of elimination for the insecticide and/or its
              metabolites.

       •      Essentially all of the 3% of a dermal dose (5.0 mg/kg) of chlorpyrifos absorbed
              by male volunteers was eliminated in the urine within 7 days post-administration
              (Nolan et al., 1984). An elimination half-life of 27 hours was reported.


IV.    HEALTH EFFECTS

       Humans

              Short-term Exposure

              •     Plasma cholinesterase (ChE) activity was  depressed to about 15% of
                    predose levels following administration of a single oral dose of 0.5 mg
                    chlorpyrifos/kg to six men  (Nolan et al., 1984). Enzyme activity  returned
                    to near-normal (i.e.,  80 to 90% of predose levels) within 4 weeks.  No
                    other signs or symptoms of toxicity were observed during the 30-day post-
                    treatment period.

              •     In a study by Dow Chemical Company (1972), 16 human  male volunteers
                    (four/dose) received  0, 0.014,  0.03 or 0.10 rag chlorpyrifos/kg/day (in
                    capsule form) for 28, 28. 21  or 9 days, respectively. The high-dose
                    treatment (0.10 mg/kg/day) was discontinued after 9 days due to a runny
                    nose  and blurred vision in one individual.  The authors did not state why
                    administration of the 0.03  mg/lcg dose was terminated on day 21. Mean
                    plasma ChE  activity  in the high-dose (0.10 mg/kg) group was inhibited by
                    about 30% when compared to the mean control value (p  < 0.05) and by

-------
Chlorpyrifos                                                                    April 1992
                    about 65% when compared to baseline (i.e., pretreatment) levels.  In the
                    group receiving 0.03 mg/kg/day doses, plasma ChE activity averaged about
                    70% of pretreatment levels and 87% of concurrent control values;
                    however, these differences were not statistically significant.  Plasma ChE
                    activity was comparable for low-dose and control individuals. Plasma
                    ChE activities of all affected persons returned to pretreatment levels
                    within 4 weeks after administration of test material was terminated. No
                    effect on erythrocyte ChE activity was observed at any dose.  This  study
                    identified a No-Observed-Adverse-Effect Level (NOAEL) of 0.03 mg/kg/
                    day and a Lowest-Observed-Adverse-Effect Level (LOAEL) of 0.1
                    mg/kg/day based on the absence or presence of decreased plasma ChE
                    activity.

              •     Five office workers exposed to chlorpyrifos in the air (levels not  reported)
                    for 5 to 21 hours over a 3-day period had significantly (p <0.01) reduced
                    eythrocyte  ChE levels 1 month after exposure, when compared to values
                    obtained 4 months post-exposure (Hodgson et al., 1986).  Erythrocyte
                    ChE activity measured on the first day after exposure was estimated to be
                    approximately 33% of the 4-month value. Physical examinations, nerve
                    conduction studies, and routine blood and urine tests were normal for all
                    but one worker, who developed numbness and tingling in the fingertips of
                    both hands 3 weeks after exposure. Most of the individuals complained
                    of fatigue,  weakness and anxiety and experienced diarrhea, abdominal
                    pain and nausea within hours and also during the first 3 weeks after
                    exposure to chlorpyrifos. Symptoms were resolved by 4 weeks, and no
                    chlorpyrifos was detected in the office air 2 weeks after the initial
                    exposure period.

              •     A 42-year-old man who ingested approximately 300 mg chlorpyrifos/kg
                    was comatose and showed acute signs of cholinergic toxicity through day
                    17. Longer term neurological effects (leg weakness, reduced or abolished
                    tendon reflex, reduced or lost vibration sense, and muscle denervation)
                    were present from day 40  and became progressively worse with time
                    (Lotti et al., 1986).  Blood concentration of chlorpyrifos dropped in an
                    exponential manner from 680 nmol/L on day 3 to 49 nraol/L on day 10;
                    none was detected  13 days after ingestion.  Blood ChE, plasma
                    butylcholinesterase and lymphocyte neuropathy target esterase (Nit)
                    activity levels were markedly depressed on day 30 but began to increase
                    thereafter, through day 90. Inhibition of NTE preceded the development
                    of polyneuropathy.

              •     An 11-day-old boy, exposed to  chlorpyrifos in the home, became lethargic
                    and cyanotic  prior to respiratory arrest (Dunphy et al., 1980). The infant
                    was resuscitated but remained  limp and relatively unresponsive to  stimuli.

-------
Chlorpyrifos                                                                     April 1992
                     His red blood cell ChE activity level was about 50% below normal. After
                     8 days, the infant appeared well but ChE activity was not measured.
                     Exposure was probably via both the oral and cutaneous routes, since
                     chlorpyrifos residues were found on dish towels, food preparation  surfaces
                     and the infant's clothing. Direct inhalation exposure may also have
                     occurred since the house reportedly smelled strongly of insecticides when
                     the baby was taken to the hospital.

                     Insecticide-related signs or symptoms of toxicity were not observed in any
                     of six men who received a single dermal application of 5.0 mg
                     chlorpyrifos/kg (Nolan et al., 1984). Mean plasma ChE activity was
                     depressed slightly (to about 13% of predose levels) but did not exhibit a
                     consistent pattern among the individual volunteers.

                     Seven human adults (sex not reported) were exposed dermally, by patch
                     tests, to 1.0, 1.5, 3.0, 5.0 or 7.5 mg chlorpyrifos/kg; the total exposure
                     areas ranged from 2.25 to 13 JO in2, and the length of exposure was 12
                     hours (Dow Chemical Company, 1972).  No skin irritation was observed
                     in any of the subjects, and both erythrocyte and plasma ChE levels
                     remained unchanged throughout the experimental period. In addition,  no
                     morphological alterations were observed in lymphocytes obtained  from
                     exposed sites. The data indicate that low levels of chlorpyrifos do not
                     present a significant toxicity hazard from acute skin exposure.

                     Plasma ChE activities in a group of seven adult humans (sex not
                     reported) decreased by about 30% following multiple 12-hour dermal
                     exposures to chlorpyrifos (Dow Chemical Company, 1972). During the
                     first test period, individuals received three applications  of 25 mg
                     chlorpyrifos/kg, and in the second experiment, each  subject received
                     applications of 5 mg chlorpyrifos/kg.  No other effects, including dermal
                     irritation, were observed.  ChE activity levels returned to normal within 7
                     to 9 days after the final exposure.

                     Spray workers exposed to 0.5% chlorpyrifos emulsion in field trials for
                     malaria control snowed decreased plasma and erythrocyte ChE activity
                     levels (Eliason et al., 1969).  In this study, five of seven sprayers showed
                     more than a 50% reduction in ChE within 2 weeks after spraying  began.

                     In a study by Ludwig et al. (1970), groups of two to three human
                     volunteers were exposed to one of several thermal aerosols containing
                     chlorpyrifos.  Exposures of 3 to 8 minutes at concentrations of about 0.8
                     /im/m3 produced no significant changes in ChE levels. This concentration
                     is similar to the application rate recommended in thermal fogging.

-------
Chlorpyrifos                                                                    April 1992
              Long-term Exposure

              •     Plasma ChE was significantly (p <0.001) inhibited in a group of 17
                    workers exposed occupationally to a Time-Weighted Average (TWA) of
                    7.54 mg chlorpyrifos/m3 for 8 hours/day, 5 days/week, for 2 years, when
                    compared to age- and sex-matched controls (Hayes et al., 1980). Most
                    workers experienced headaches and complained of aggravated nasal or
                    respiratory problems.  General physical examinations were normal,
                    however, as were erythrocyte ChE activity levels.

              •     Two groups of machine-operating  farm workers (numbers not reported)
                    exposed daily to a granular insecticide containing 5% chlorpyrifos  were
                    examined over a 2-year period (Majczakowa et al., 1985). These tractor
                    drivers and feeder operators were  in contact with insecticide
                    concentrations not exceeding 0.015 and 0.040 mg/m3, respectively,  based
                    on samples periodically analyzed from breathing areas of workers. The
                    authors reported that up to 2 mg chlorpyrifos were recovered from the
                    workers'  hands  at various sampling intervals.  Average potential exposure
                    at work to chlorpyrifos was estimated to be 0.373 mg/hour for the first
                    year of the study and 0.034 mg/hour for the second year. No signs of
                    toxicity or changes in blood ChE activity were observed.

       Animals

              Short-term Exposure

              •     An oral LDJO of 152 mg/kg was reported for female mice and 169  mg/kg
                    for female rats given chlorpyrifos by intubation in soy bean oil (details of
                    the chlorpyrifos formulation were not provided) (Berteau and Deen,
                    1978). Oral LD50 values for male and female rats ranged from  118 to 245
                    mg/kg; no significant sex-related differences were observed  (Gaines, 1969;
                    McCollister et al., 1974).  The acute oral LD^ for male guinea pigs was
                    504 mg/kg, and no deaths were noted in male and female rabbits dosed
                    with 1,000 mg/kg (McCollister et al., 1974).

              •     In a study conducted by Dow Chemical Company (1972), each of  three
                    rhesus monkeys (sex not specified) was given a single oral dose of 3.5 mg
                    chlorpyrifos/kg.  Erythrocyte ChE levels were 60% below pretreatment
                    levels at 4 hours post-dosing but increased to 66, SO and 82% of baseline
                    values at 8, 24 and 48  hours, respectively.  Plasma ChE levels were more
                    severely affected and were only 6,  8, 14 and 30% of baseline values at the
                    respective sampling times tested above.
                                           10

-------
Chlorpyritbs                                                                    April 1992
             •     Two rhesus monkeys (sex not reported) given a single oral dose of 2
                    mg/kg/day chlorpyrifos for 3 consecutive days showed no clinical signs of
                    toxicity (Dow Chemical Company, 1972).  A sharp decrease (15 to 25% of
                    control values) in plasma ChE activity was observed 24 hours after the
                    initial dosing. An additional 5% reduction was observed after
                    administration of the second and third doses. Erythrocyte ChE activity
                    levels dropped only slightly during the first day; greater reductions (to 60
                    to 65% of control levels) were observed on the second and third days of
                    the study.

             •     In a range-finding study conducted by Dow Chemical Company (1972),
                    pairs of beagle dogs consuming a diet containing 0.6 ppm (0.015
                    mg/kg/day, based on Lehman, 1959) chlorpyrifos for 12 days  showed no
                    changes in either plasma or erythrocyte ChE activity.  When the chemical
                    was administered for 28 days at a dietary concentration of 2  ppm (0.1
                    mg/kg/day), the  plasma ChE activity in one female was reduced by 50%
                    within 7 days after the study began. In another study, dogs fed 6, 20 or
                    60 ppm (0.15, 0.5 and 1.5 mg/kg/day)..chlorpyrifos for 35 days showed
                    reduced plasma ChE activity to 42%, 25% and 17% of pretreatment
                    values, respectively; however, erythrocyte and brain ChE activities did not
                    change. From these two studies, it was concluded that the NOAEL was
                    0.015 mg/kg/day  for dogs exposed orally to chlorpyrifos.

             •     Symptoms of severe ChE inhibition developed in beagle dogs (two/sex/
                    group) fed 2,000 (50 mg/kg/day) or 600 (15 mg/kg/day) ppm chlorpyrifos
                    in the diet for 5  and 16 days, respectively  (Dow Chemical Company, 1972;
                    conversions based on Lehman, 1959).  These dogs were taken off their
                    respective diets  and placed on a 200-ppm  diet. Additional groups of dogs
                    consumed a 200-ppm (5-mg/kg/day) diet for up to 45 days or a 20- or 60-
                    pprn (0.5 or 1.5 mg/kg/day) diet for up to  88 days.  Slowed growth was
                    observed in all males and in females consuming 200 ppm chlorpyrifos.
                    Plasma and erythrocyte ChEs were depressed in all groups of animals.
                    Brain ChE activity was decreased in both  sexes receiving 200 ppm but
                    only in females consuming the 60-ppra diet. Gross and histological
                    examination of tissues was normal in all dogs.  This study indentified a
                    brain ChE-depression NOAEL of 0.5  mg/kg/day and a LOAEL of 1.5
                    rag/kg/day for beagles of both sexes.

             •     Acute dermal and inhalation exposures to chlorpyrifos (in 65% xylene)
                    were as toxic to  mice and rats as were oral exposures. A dermal LD50
                    value of 202 mg/kg for rats was reported by Gaines (1969), and inhalation
                    LCjo values of 152 and 169 mg/kg were reported for female mice and rats,
                    respectively, by  Berteau and Deen (1978).
                                           11

-------
Chlorpyritbs                                                                    April 1992
              •     In a study by Berteau and Dean (1978), groups of 16 mature female
                    NAMRU mice (30 to 35 g) inhaled a 65% xylene aerosol cloud containing
                    the equivalent of 0.1 to 50 mg chlorpyrifos/kg for 27 to 50 minutes. Dose-
                    related decreases in plasma acetylcholinesterase were observed; enzyme
                    activity was 55% of predosing activity following exposure to 0.2 mg/kg and
                    less than 10% after exposure to 50 mg/kg.

              •     Groups of 10 male and 10 female  Sprague-Dawley rats that inhaled an
                    aerosol cloud containing 5 mg chlorpyrifos/L for an unspecified amount of
                    time exhibited lachrymation, slight nasal discharge and gasping during
                    exposure (Dow Chemical Company,  1972).  Animals appeared normal
                    during the 14-day post-inhalation period, and postmortem examination of
                    tissues revealed no gross pathological changes.

              Dermal/Ocular Effects

              •     Chlorpyrifos (0.5 mL of a 24% solution) was applied to the intact and
                    abraded skin of six New Zealand albino rabbits (sex and age not
                    reported) (Dow Chemical Company, 1972).  Animals were exposed to the
                    test material for 24 hours. Moderate to severe erythema developed on all
                    exposed areas; slight necrosis was  observed on four of the intact areas and
                    five of the abraded areas.  All exposed skin  areas had some degree of
                    edema. Reactions of intact and abraded skin of three additional rabbits
                    exposed to chlorpyrifos (24% in solution)  for 6 hours included slight
                    erythema, slight edema and slight  necrosis within 10, 30 to 60, and 90 to
                    210 minutes, respectively.

              •     Instillation of chlorpyrifos (0.1 mL of a 24% solution) into the
                    conjunctival sac of the right eye of six New Zealand albino rabbits  (sex
                    and age not  reported) produced conjunctival redness, iritis and corneal
                    injury in  all treated eyes (Dow Chemical Company, 1972).

              •     No skin or eye irritation developed in any of the 40 adult male and
                    female mongrel dogs (number/sex not reported) or 85 puppies dipped
                    repeatedly in 0.0125, 0.025, 0.05 or 0.10% chlorpyrifos solutions (Dow
                    Chemical Company, 1972).  Adults were dipped three to six times at 15-
                    or 30-day intervals; puppies (6 to 8 weeks old) were dipped up to three
                    times in the  0.025% solution but only once in the 0.05% solution.

              •     Percutaneous injections of 1.0, 2.0 or 3.98 g chlorpyrifos (as a  25%
                    solution) into groups of four albino rabbits (sex not reported)  induced
                    slight to moderate  erythema, swelling, and necrosis (Dow Chemical
                    Company, 1972).  One mid-dose rabbit died 3 days after exposure, and
                  .  three high-dose animals died within 6 to 9 days post-dosing.

                                           12

-------
Chlorpyrifos                                                                    April 1992
              Long-term Exposure

              •     Groups of 20 albino rats (10/sex/dose) were maintained on diets
                    containing 10, 30, 100 or 300 ppm chlorpyrifos (approximately 0.5, 1.5, 5
                    or 15 mg/kg, respectively, based on Lehman, 1959) for 90 days (Dow
                    Chemical Company, 1972). Another group of rats that received 1,000
                    ppm (50 mg/kg) chlorpyrifos in the feed was included in this study, but
                    due to high  mortality, this group was terminated after 28 days. Plasma
                    and erythrocyte ChE activity levels were depressed in a dose-related
                    manner, at 1,000 ppm the average plasma ChE activity  for both sexes was
                    less than 1% of the control value. Brain ChE activity also was reduced to
                    about 30%,  20% and 10% of control values in animals consuming 100,
                    300 and 1;000 ppm chlorpyrifos, respectively.  Exposure to 0.5 mg/kg/day
                    for 90 days caused a 3 to 7% reduction in brain ChE activity, and 1.5-
                    mg/kg/day doses reduced brain ChE activity by 19 to 22% after 90 days
                    (neither was significantly different from control values at p = 0.05).  The
                    animals dosed at 1,000 ppm exhibited signs of severe ChE depression
                    (e.g.,  tremors, bloody noses, circling and backing, ulceration of the cornea
                    and nostrils), decreased food consumption, significant weight loss and
                    increased mortality.  Rats consuming the 300-ppm feed experienced
                    tremors, slight diuresis and slight growth  retardation. Consumption of the
                    three lowest doses produced no signs of toxicity. The NOAEL based on
                    reduced brain ChE activity was 0.5 mg/kg/day.

              •     In a 91-day  study conducted by Dow Chemical Company (1972), groups
                    of 20 albino rats (10/sex/dose) fed 3.0 or 10.0 mg chlorpyrifos/kg/day
                    exhibited reduced plasma and erythrocyte ChE activities (35 to 58% and
                    14 to 26% of control values, respectively) and showed slight to severe
                    signs of ChE inhibition and toxicity (e.g., hunched appearance, tremors,
                    weight loss). Rats consuming a 03- or 1.0-mg/kg/day diet had depressed
                    plasma  and  erythrocyte ChE levels.  Male rats given 03 mg
                    chlorpyrifos/kg/day had reduced body weight gains.  No adverse effects
                    were observed at the 0.03- or 0.1-mg/kg/day dose levels. Survival was not
                    affected at any exposure level, and ChE activities returned to normal
                    within 1 to 2 weeks after withdrawal of the test compound from the diet.
                    This study identified a NOAEL of 0.1 mg/kg/day  and a LOAEL of 03
                    mg/kg/day (for male rats).

              •     Albino  rats  (20/sex/group) consuming dietary levels of 0.03, 0.15 or 0.75
                    mg chlorpyrifos/kg/day for 6 months showed no significant clinical or
                    histological  signs or symptoms of organophosphate poisoning (Dow
                    Chemical Company,  1972). Animals ingesting the high-dose feed
                    exhibited reduced plasma  and erythrocyte ChE activities (i.e.,  35 to 60%
                    and 50% of control values, respectively).  Brain ChE activity was not

                                            13

-------
Chlorpyrifos                                                                    April 1992
                     affected at any treatment level. This study identified a NOAEL of 0.15
                     mg chlorpyrifos/kg/day.

              •      A group of 20 Sprague-Dawley rats (4 weeks old. sexes combined) were
                     fed diets containing 100 ppm chlorpyrifos (5 mg/kg/day) for 1 year.
                     Seventeen  animals also consumed a diet supplemented with both
                     chlorpyrifos (100 ppm) and corn oil (final concentration,  20%) (Buchet et
                     al., 1977; conversions based on Lehman, 1959).  Control  animals
                     consumed nonsupplemented or corn oil-only supplemented diets (19 and
                     18, respectively).  All rats grew normally and had normal levels of
                     hormone sensitive  lipase, lipoprotein  lipase, serum fatty acids (free and
                     total), serum glycerol and serum cholesterol (total).  Total glycerol and
                     cholesterol content of the aorta in test animals were also comparable to
                     controls, but total aortic fatty acids were increased in animals consuming
                     both chlorpyrifos and 20% oil in the diet. Total blood ChE activity was
                     reduced by 40% in the chlorpyrifos-exposed group on the normal feed
                     and by 60% in those animals on the fat-enriched regimen.  A LOAEL of
                     5 mg/kg/day, based on reduced ChE activity and increased total  aortic
                     fatty acids, was identified in this study.

              •      In a study conducted by McCollister et al. (1974), groups of 7-week-old
                     Sherman rats (25/sex/dose)  fed a diet containing 1.0 or 3.0 mg
                     chlorpyrifos/kg/day for 2 years  exhibited significantly (p < 0.05)  depressed
                     plasma ChE activity levels.  Erythrocyte ChE activities were depressed (p
                     < 0.05) by approximately 67% and 85% of control values in rats fed the
                     1.0- and 3.0-mg/kg/day diets, respectively; brain ChE activity was
                     significantly (p < 0.05) reduced (to about 57% of controls) in the  high-
                     dose animals only. Effects on  ChE activity were reversible when
                     consumption of a chlorpyrifos-free diet was resumed. ChE activity in
                     animals fed 0.1 mg/kg/day was comparable to control values.  No clinical
                     signs of toxicity were observed at any dose.  A NOAEL of 0.1 rag
                     chlorpyrifos/kg/day based on plasma ChE activity, and a NOAEL of 3.0
                     mg/kg/day based on systemic effects were established in this study.

              •      Groups of three or. four rhesus monkeys (males and females combined)
                     that received chlorpyrifos by gavage at  doses of 0.08, 0.4 or 2.0 mg/kg/day
                     for 6 months showed no significant compound-related clinical effects
                     compared to control animals (Dow Chemical Company, 1972).  The only
                     evidence of exposure to chlorpyrifos was reduced plasma and erythrocyte
                     ChE activities in the mid- and high-dose monkeys (significance levels not
                     reported).  Midbrain and cerebrum ChE values were not affected  in any
                     group.  Histological examination revealed that the liver and kidney
                     showed no abnormalities in any of the animals. A NOAEL of 0.08
                                            14

-------
Chlorpyrifos                                                                      April 1992
                     mg/kg/day was identified, based on the absence of inhibition of ChE at
                     this dose.

              •      Plasma ChE activity was depressed by 40 to 75% of pretest or control
                     values in groups of three or four male and female beagle dogs (aged  10 to
                     11 months) given 0.1, 1.0 or 3.0 mg chlorpyrifos/kg/day in the diet for 1 to
                     2 years (McCollister et al., 1974). Brain ChE activity was slightly
                     depressed (by 8 to 21% of the control value) at the highest dose level;
                     however, these decreases were not statistically significant (p < 0.05).
                     Both plasma and brain ChE activities returned to normal when dosed
                     animals were placed on the control diet.  No significant health effects
                     were observed at any dose (0.01 and 0.03  mg/kg/day) using the following
                     criteria: mortality, body weight,  food intake, hematological and clinical
                     chemistry parameters, organ weight, tumor incidence,  and gross and
                     histopathologicar examination of tissues.  The only notable difference was
                     a statistically significant (p <  0.05) increase in the mean liver-to-body
                     weight  ratio of high-dose males administered chlorpyrifos for 2 years.
                     The NOAEL for dogs identified  in this study was 0.03 mg chlorpyrifos/kg/
                     day based on plasma ChE activity levels and was 3.0 mg/kg/day based on
                     systemic effects.

              Reproductive Effects

              •      In a three-generation reproduction study, groups of 15 male and 15
                     female Sprague-Dawley albino rats that received up to 1.0 mg
                     chlorpyrifos/kg/day in the feed showed no adverse reproductive or
                     postnatal effects, as judged by fertility, gestation, viability and lactation
                     indices (Dow Chemical Company, 1972). Litter size, pup weight and sex
                     ratios of offspring from treated rats also were unaffected by exposure to
                     the test compound.  In addition, ingestion of chlorpyrifos (0.03, 0.1 or 0.3
                   .  mg/kg/day by the first-generation rats and 0.1, 03 or 1.0 mg/kg/day by the
                     second- and third-generation  rats) had no adverse effects on survival,
                     body weight gains and  food consumption of either male or female
                     parents.  Third-generation rats (both sexes) consuming the 1.0-mg/kg/day
                     diet had depressed plasma and erythrocyte ChE activities, as did females
                     given feed containing 0.3 mg  chlorpyrifos/kg/day.  It was concluded that
                     the reproductive NOAEL from this study is 0.1 mg/kg/day.

              •      Dow Chemical Company (1972)  reported that multiple exposures to
                     chlorpyrifos (0.025, 0.05 or 0.10% solutions) via dipping produced no
                     maternal toxicity in mongrel dogs and had no effect on gestation or
                     parturition.  Twelve dogs were dipped one to four times at 15- or 30-day
                     intervals.  Animals were cither not pregnant or up to 58 days pregnant at
                     the time of the first dip (average gestation period,  63  ฑ 7 days).

                                             15

-------
Chlorpyrifos                                                                     April 1992
              •      Everett (1982) studied the effects of dermal applications of a test material
                     consisting of 43.2% chlorpyrifos on 185 Holstein bulls. No other dosing
                     information was provided.  In the 172 bulls that did not exhibit severe
                     toxic effects, semen production initially was depressed but returned to
                     normal within 6 months post-treatment.  Normal semen production was
                     measured for at least 12 months post-treatment. Six bulls became very
                     sick following exposure to chlorpyrifos, and their semen production did
                     not return to normal within  12 months.  The remaining seven bulls died
                     after being treated with the  insecticide.

              Developmental Effects

              •      In a developmental study conducted by Deacon et al. (1980), 40 to
                     47 pregnant CF-1 mice were dosed, by gavage, with 1, 10 or 25 mg
                     chlorpyrifos/kg/day on gestation days 6 through 15.  Fifty-one animals
                     were used in the control group.  Severe maternal toxicity, including
                     mortality, clinical signs of ChE inhibition, and significant  (p < 0.05)
                     decreases in maternal body weight gains and food and water consumption
                     were reported at 25 mg/kg/day. Plasma and erythrocyte ChE levels were
                     significantly (p < 0.05) reduced at all dose levels tested when compared
                     with controls. Developmental toxic effects were reported at 25 mg/kg/day.
                     The findings included significant reductions in fetal body weight and
                     crown-rump lengths.  Exencephaly was noted in four fetuses from three
                     litters of mice dosed with 25 mg/kg/day (nonsignificant) and in five fetuses
                     from five litters  in the 1-mg/kg/day dose group (significant at p < 0.05).
                     Significant increases (p < 0.05) in the incidences of delayed ossification  of
                     the skull and sternebrae were  also reported at the highest dose level. The
                     incidence of sternebrae abnormalities was high (p < 0.05) among fetuses
                     bom to dams  in the lowest dose group (1 mg/kg/dayj.  These results were
                     not repeatable, however, when additional groups of 35 to 41 CF-1 mice
                     were given 0, 0.1, 1 or 10 mg chlorpyrifos/kg by gavage on gestation days
                     6 through 15 (Deacon et al., 1980). Thus, although the first study
                     suggests a fetal LOAEL of 1 mg/kg/day based on reduced ChE activity
                     and adverse developmental effects, the data are equivocal due to the lack
                     of any significant response in the second group of test animals. From
                     both studies combined, the NOAEL appears to be 0.1 mg/kg/day.

              •  .    Chlorpyrifos was not teratogenic in rats, as judged by  external, skeletal
                     and visceral examination of  second-litter fetuses from  third-generation
                     Sprague-Dawley female rats administered the insecticide by gavage at 1.0
                     mg/kg/day on  gestation days 6 through 15 (Dow Chemical Company,
                     1972).  Parental females received chlorpyrifos in the diet at levels of 0.1,
                     0.3 or  1.0 mg/kg/day for the  rest of their lives.  Maternal weight gains and

-------
Chlorpyrifos                                                                     April 1992
                     food consumption, corpora lutea, resorptions, fetus viability, pup weights
                     and sex ratios also appeared to be unaffected.

              •      In the dog reproduction study by Dow Chemical Company (1972), 58 of
                     85 (68%) pups born to mongrel dogs that had been dipped repeatedly in
                     chlorpyrifos (0.025, 0.05 or 0.10% solutions) died before 8 weeks of age.
                     Only one puppy death was attributed to chlorpyrifos (due to symptoms
                     suggestive of organophosphate toxicity). No other details were given.

              Mutagenicity

              •      Chlorpyrifos was not mutagenic in Salmonella typhimurium or Escherichia
                     coli either in the absence or presence of both mammalian and plant
                     microsomes (Gentile et al., 1982; Moriya et al., 1983; Shirasu  et al., 1976,
                     Waters et al.,  1982).  Similarly, chlorpyrifos  did not induce reverse
                     mutations in Zea mays (Gentile et al., 1982; Seehy et al., 1984) or cause
                     sex-linked recessive lethal mutations in Drosophila melanogaster (Waters
                     et al.,  1982).

              •      Evidence of in vivo induction of micronucleated polychromatic
                     erythrocytes in mouse bone marrow cells following intraperitoneal or oral
                     administration of chlorpyrifos have been reported (Amer and Fahmy,
                     1982).  Chlorpyrifos induced mitotic suppression and increased the
                     frequency of chromosome aberrations in Vtcia faba (Abdou and Abdei-
                     Wahab, 1985). It produced clastogenic effects in barley (Kaur and
                     Grover, 1985).

              •      Chlorpyrifos, without exogeneous metabolic activation,  was genotoxic in
                     DNA polymerase I-deficient E. coli and recombination-deficient S.
                     typhimurium (Waters et al., 1982).  However, inconsistent results have
                     been seen in the Bacillus subtilis rec-assay.  Waters et al. (1982) reported
                     a positive response, but Shirasu et al. (1976) and Kada et al. (1980) found
                     no genotoxic activity.  Chlorpyrifos was not recombinogenic in
                     Saccharomyces cerevisiae D3 either with or without rat liver microsomes
                     (Waters et al., 1982) or in S. cerevisiae D4 with or without mammalian and
                     plant microsomes (Gentile et al., 1982). Unscheduled DNA synthesis  was
                     not increased in chlorpyrifos-treated human lung fibroblast (Waters et al.,
                     1982).  Chlorpyrifos was judged  negative for the induction of  sister
                     chromatid exchanges in Chinese hamster ovary cells, chick embryos
                     (Muscarella et al., 1984). and the LAZ-007 human lymphoid cell line
                     (Sobti et al., 1982).

-------
    Chlorpyrifos                                                                    April 1992
                 Carcinogenic! tv

                 •      Chlorpyrifos was not tumorigenic in a chronic feeding study in which 7-
                        week-old Sherman rats were administered up to 3.0 mg chlorpyrifos/kg/
                        day for 2 years (McCollister et al., 1974).  Thirty rats/sex/dose were used;
                        an additional five to seven rats/sex/group killed  for interim gross and
                        microscopic pathological examinations were normal throughout the
                        experimental period. In this study too few animals were included to fully
                        assess the carcinogenicity of Chlorpyrifos in rats. Since there was no
                        evidence of toxicity at  tested doses, a minimal toxic dose (MTD) may not
                        have been used.

                 •      No excess tumors developed in groups of 10-  to 11-month-old beagle dogs
                        (three or four/sex/group) fed 0.01, 0.03, 0.1, 1.0 or 3.0 mg chlorpyrifos/kg
                        in  the diet for 1 to 2 years (McCollister et al., 1974). In addition, gross
                        and microscopic examination of tissues was normal for all dose levels
                        throughout the experimental period. However, this study is considered of
                        limited usefulness in providing  data to assess  the carcinogenic potential of
                        Chlorpyrifos: the study was relatively short (i.e.,  less-than-lifetime for
                        dogs), and the animals may not have been tested at MTD.
   V.     QUANTIFICATION OF TOXICOLOGICAL EFFECTS

          Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term,
   and Lifetime exposures if adequate data are available that identify a sensitive noncarcinogenic
   end point of toxicity.  The HAs for noncarcinogenic toxicants are derived using the following
   formula:
   where:

NOAEL or LOAEL =   No- or Lowest-Observed-Adverse-Effect Level in rag/kg bw/day.

               BW =   assumed body weight of a child (10 kg) or an adult (70 kg).

               UF =   uncertainty factors (10. 100, 1.000, or 10,000), in accordance with EPA or
                        NAS/OW guidelines.

           _ L/day =   assumed daily water consumption of a child (1 L/day) or an adult 2 L/
                        day).

-------
Chlorpyrifos                                                                    April 1992
       One-day Health Advisory

       No suitable information was found in the available literature for determining the One-
day Health Advisory (HA) for chlorpyrifos. The Ten-day HA for a child, 30 /xg/L, calculated
below, is recommended for use as a conservative estimate for a 1-day exposure to chlorpyrifos.

       Ten-day Health Advisory

       The human oral-exposure study by Dow Chemical Company (1972) has been selected to
serve  as the basis  for the Ten-day HA because it contains adequate human data of appropriate
length.  In this study, groups of four healthy adult men were administered chlorpyrifos in
capsule form at doses of 0, 0.014, 0.03 or 0.10 mg/kg for 28, 28, 21 or 9 days, respectively.
Adverse health effects were observed  only in the high^dose (0.10 mg/kg) group; at this exposure
level,  plasma ChE activity was reduced by approximately 65% when compared to control values.
One individual in  this group also  experienced a runny nose and blurred vision. A NOAEL of
0.03 mg/kg/day was identified from this study.

       Another study by Dow Chemical Company (1972), in which a NOAEL of 0.015 mg/kg
was reported for beagle dogs consuming chlorpyrifos in the feed for 12 days, was also
considered for the calculation of the Ten-day HA. However, this value was not used because
the available human data were within  one order of magnitude of the animal data. Therefore,
the results of this dog study support the human data used to calculate the Ten-day HA but are
not the most appropriate for the derivation of this HA.

       The Ten-day HA for the 10-kg child is calculated as follows:

       Ten-day HA -  (ฐ^ (?^yff  ^  - 0-03 mg/L - (rounded to 30 Mg/L)


where:

 0;03 mg/kg/day =   NOAEL, based on the absence of decreased plasma ChE activity in male
                    human subjects exposed to chlorpyrifos via  the oral route for 21 days
                    (Dow Chemical Company, 1972).

          10 kg =   assumed weight of child.

             10 =   uncertainty factor,  chosen in accordance with EPA or NAS/OW guidelines
                    in which a NOAEL from a human study is employed.

        1 L/day =   assumed water consumption of 10-kg child.
                                           19

-------
Chlorpyrifos                                                                   April 1992
       Longer-term Health Advisory

       The study by Dow Chemical Company (1972) in which humans received daily oral doses
or" chlorpyrifos tor up to 28 consecutive days has been selected to serve as the basis for the
Longer-term HA because it contains human data that identify a NOAEL and a LOAEL.
Although several adequate subchronic animal studies were available, data from these studies
indicate that humans are more sensitive to chlorpyrifos following oral administration than other
species. Using human data would, therefore, allow for a more conservative estimate of a
Longer-term HA for a 10 kg child.  The following studies  used plasma and erythrocyte ChE
activity levels as the basis for NOAEL and LOAEL values. Two 3-month feeding studies with
rats identified NOAELs of 0.1 and 1.5 mg/kg/day; respective LOAELs were 0.3 and 5 mg/kg/day
(Dow Chemical Company, 1972). A 6-month study in which rats received chlorpyrifos in the
diet identified a NOAEL of 0.15 mg/kg/day and a LOAEL of 0.75 mg/kg/day (Dow Chemical
Company, 1972). Finally, monkeys exposed via oral gavage to chlorpyrifos for 6 months showed
no adverse effects at 0.08 mg/kg/day doses but had reduced ChE activities at 0.4 mg/kg/day
(Dow Chemical Company, 1972). The human study, also conducted by Dow Chemical Company
(1972), identified a NOAEL of 0.03 mg/kg/day and a LOAEL of 0.1 mg/kg/day based on the
absence or presence of reduced plasma ChE activity following administration of chlorpyrifos (in
capsules) to groups of four healthy male adults for 9 and 21 days, respectively.

       The Longer-term HA for the 10-kg child is calculated as follows:

      Longer-term  HA = (0-03 nig/kg/day) (10 kg) = QM  mg/L  (rounded t(j 3Q
                              (10) (1 L/day)

where:

 0.03 mg/kg/day =   NOAEL, based on the absence of decreased plasma ChE activity in male
                    human subjects exposed to chlorpyrifos via the oral route for 21 days
                    (Dow Chemical Company, 1972).
          10 kg =   assumed body weight of a child.

             10 =   uncertainty factor, chosen in accordance with EPA or NAS/OW guidelines
                    in which a NOAEL from a human study is employed.

        1 L/day =   assumed daily water consumption of a child.

       The Longer-term HA for the 70-kg adult is calculated as follows:
    Longer-term HA =  (ft03 ™f ^ <70  kg)  . 0.105 mg/L   (rounded to  100 Mg/L)
                             (10) (2 L/day)
                                           20

-------
Chlorpyrifos                                                                    April 1992
where:

  0.03 mg/kg/day =   NOAEL, based on the absence of decreased plasma ChE activity in male
                    human subjects exposed to chlorpyrifos via the oral route for 21 days
                    (Dow Chemical Company, 1972).

          70 kg =   assumed body weight of an adult.

             10 =   uncertainty factor, chosen in accordance with EPA or NAS/OW guidelines
                    in which a NOAEL from a human study is employed.

        2 L/day =   assumed daily water consumption of an adult.

       Lifetime Health Advisory

       The Lifetime HA represents that portion of an individual's  total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure. The Lifetime HA is derived in a three-step process. Step 1
determines the Reference Dose (RfD),  formerly called the Acceptable Daily Intake (ADI).  The
RfD is an estimate of a daily exposure to the human population that is likely to be without
appreciable risk of deleterious health effects during a lifetime, and is derived from the NOAEL
(or LOAEL), identified from a chronic  (or subchronic) study, divided by an uncertainty
factor(s).  From the RfD, a Drinking Water Equivalent Level (DWEL) can be determined (Step
2).  A DWEL is a medium-specific  (i.e., drinking water) lifetime exposure  level, assuming 100%
exposure from that medium, at which adverse, noncarcinogenic health effects would not be
expected to occur.  The DWEL is derived from the multiplication of the RfD by the assumed
body weight of an adult and divided by  the assumed  daily water consumption of an adult. The
Lifetime HA in drinking water alone is  determined in Step 3 by factoring in other sources of
exposure, the relative source contribution (RSC). The RSC from drinking water is based on
actual exposure data or, if data are not  available, a value of 20% is assumed.

       If the contaminant is classified as a known, probable, or possible human carcinogen,
according to the Agency's classification  scheme of carcinogenic potential (U.S. EPA,  1986), then
caution must be exercised in making a decision on how to deal with possible lifetime  exposure
to this substance. For human (A) or probable human (B) carcinogens, a Lifetime HA is not
recommended.  For possible human carcinogens (C), an additional 10-fold safety factor is used
to calculate the Lifetime  HA. The risk manager must balance this assessment  of carcinogenic
potential and the quality  of the data against the likelihood of occurrence and significance of
health effects related to noncarcinogenic endpoints of toxicity. To  assist the risk manager in
this process, drinking water concentrations associated with estimated excess lifetime cancer risks
over the range of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult drinking 2 L of water/day  are
provided in the Evaluation of Carcinogenic Potential section.

-------
Chlorpyrifos                                                                   April 1992
       The oral-exposure study with humans by Dow Chemical Company (1972) has been
selected to serve as the basis for the Lifetime HA because it contains adequate human exposure
data that are supported by animal data.  A NOAEL of 0.03 mg/kg/day for humans was
identified in this study. Three animal studies also were considered for the calculation of the
Reference Dose (RfD), Drinking Water Equivalent Level (DWEL), and Lifetime HA. The
chronic (2-year) feeding study  in rats (McCollister et al., 1974) was judged unacceptable due to
insufficient  numbers of animals (25/sex/dose with  an additional 5 to 7/sex/dose for interim
sacrifices).  The other animal studies, which included the reproduction study with rats by Dow
Chemical Company (1972) and the chronic feeding study with dogs by McCollister et al. (1974),
were considered adequate. However, the human  oral-exposure study  by Dow Chemical
Company (1972) was judged most appropriate for the calculation of the Lifetime HA for
chlorpyrifos. Although this human study was coreclassified as supplementary because only four
males/dose were used, when considered with the available experimental data in animals, the
NOAEL for ChE inhibition in humans (0.03  mg/kg/day) appeared  comparable to that in rats
(0.1 mg/kg/day) and dogs (0.03 mg/kg/day).

       Using the human study (Dow Chemical Company, 1972), the Lifetime HA is derived as
follows:

Step 1: Determination of the RfD

                            m  (0.03 mg/kg/day) =
                                     (10)                * &   '

where:

    0.03 mg/kg/day   =     NOAEL, based on the absence of decreased plasma ChE activity
                          in male human subjects exposed to chlorpyrifos via the oral  route
                          for 21 days (Dow Chemical Company, 1972).

               10   =     uncertainty factor, chosen in accordance with EPA or  NAS/OW
                          guidelines in which  a NOEL from a  human study is employed.

Step 2: Determination of the DWEL

       ™T/TrT    (0.003 mg/kg/day) x (70 kg)    n ,nc    „   ,     .  , t  inn   n.
       DWEL = .1	?—ฐL—LL	i	ฐL = 0.105 mg/L  (rounded to 100 ug/L)
                          (2 L/day)

where:

   0.003 mg/kg/day   =     RfD.

            70 kg   =     assumed body weight of an adult.

-------
Chlorpyritbs                                                                   April 1992



           2 L/day   =     assumed daily water consumption of an adult.

Step 3:  Determination of the Lifetime HA

                  Lifetime HA = (0.1 mg/L) (20%) = 0.02 mg/L (20 /zg/L)

where:

       0.105 mg/L   =     DWEL.

             20%   =     assumed relative source contribution from water.

       Evaluation of Carcinogenic Potential

       •     No evidence of carcinogenicity was found in rats or dogs fed up to 3.0 mg
             chlorpyrifos/kg/day for 1 to 2 years (McColIister et al., 1974).

       •     The International Agency for Research on Cancer (IARC) has not evaluated the
             carcinogenic potential of chlorpyrifos.

       •     Applying the criteria in EPA's guidelines for assessment of carcinogenic risk
             (U.S. EPA, 1986), chlorpyrifos may be classified in Group D: not classifiable.
             This category is for agents with inadequate animal evidence for carcinogenicity.


VI.    OTHER CRITERIA. GUIDANCE. AND STANDARDS

       •     The American Conference of Governmental Industrial Hygienists recommends a
             Threshold Limit  Value-Time-Weighted Average of 0.2 rag/m3 and a Short-term
             Exposure Limit of 0.6 mg/m3 for dermal exposures (ACGEH, 1988).

       •     The Food and Agriculture Organization/World Health Organization (FAO/
             WHO, 1984) Acceptable Daily Intake for chlorpyrifos is 0.01 mg/kg/day (Gartrell
             et al., 1985).


VII.   ANALYTICAL METHODS

       •     Chlorpyrifos is one of the phosphorus-containing pesticides. The relative ease
             with which this pesticide can be monitored by element-specific detectors has
             usually led to its inclusion in pesticide monitoring studies.  Chlorpyrifos can be
             analyzed by Methods 622  (U.S. EPA, 1982) and 507 (U.S. EPA, 1988). In both
             methods a liter of sample is extracted with methylene chloride, the solvent is then
             exchanged for hexane or methyl tertiary butyl ether. Analysis  is by an element-

                                          23

-------
Chlorpyrifos                                                                    April 1992
              specific thermionic nitrogen-phosphorus detector, which allows even relatively
              "dirty" samples to be analyzed with no clean-up. The estimated detection limit
              for this residue is 0.3 /ig/L.
VIII.   TREATMENT TECHNOLOGIES

       •      There is presently no information available describing treatment technologies
              capable of removing chlorpyrifos from contaminated drinking water supplies.

       •      Chlorpyrifos may be amenable to removal by activated carbon adsorption due to
              its solubility.

       •      Chlorpyrifos is probably not amenable to removal by aeration on the basis of its
              Henry's Coefficient value.
                                           24

-------
Chlorpyrifos                                                                    April 1992
IX.    REFERENCES

Abdou, RiF. and M.A. Abdel-Wahab.  1985. Cytological and developmental effects of certain
       insecticides in Vicia faba.  Int. Pest Control. 27:123-125.

ACGIH.  1988. American Conference of Governmental Industrial Hygienists. Threshold limit
       values and biological exposure indices for 1987-1988. 2nd printing.  Cincinnati, OH:
       ACGHI, p. 16.

Amer, S.M. and M.A. Fahmy. 1982.  Cytogenetic effects of pesticides.  I. Induction of
       micronuclei in mouse bone marrow by the insecticide Dursban. Mutat. Res.  101:247-
       255.

Bauriedel, W.R., R.L. McKellar and J.H. Miller. 1976. A rotational crop study using UC-
       iabeled chlorpyrifos: GHrC 876. Dow Chemical U.S.A., Midland,  MI."  Unpublished
       study. Contract No. 464-448. Publication No. CDL:224341-A.

Berteau, P.E. and W.A. Deen.  1978. A comparison of oral and inhalation toxicities of four
       insecticides to mice and rats. Bull. Environ. Contam. Toxicol. 19:113-120.

Bidlack, H.D.  1979.  Degradation of chlorpyrifos in soil under aerobic, aerobic/anaerobic and
       anaerobic conditions: method GH-C 1258.  Dow Chemical U.S.A., Midland, MI."
       Unpublished study. Contract No. 464448. Publication  No. CDL:241547-A.

Buchet, J.P., R. Lauwerys and R. Roels.  1977.  Long term exposure to organophosphorous
       pesticides and lipid metabolism in the rat. Bull. Environ. Contam. Toxicol.  17:75-183.

Chapman, R.A. and C.R. Harris.  1980. Persistence of chlorpyrifos in a mineral and  an organic
       soil.  J. Environ. Sci. Health B15:39-46.

Deacon, M.M., J.S. Murray, M.K. Pilny, K.S. Rao, D.A. Dittenber, T.R. Hanley, Jr. and J.A.
       John. 1980. Embryotoxicity  and fetotoxicity of orally administered chlorpyrifos in mice.
       Toxicol. Appl. Pharmacol. 54:31-40.

Dow Chemical Company.' 1972.  Product literature.  Dowco 179.  No. 112118.  Washington,
       DC:  U.S. Environmental  Protection  Agency.

Dunphy, J., M. Kesselbrenner, A. Stevens, B. Vlec and R.J. Jackson.  1980.  Pesticide poisoning
       in an infant - California.  MMWR. 29:254-255.
    "Confidential Business Information.  Submitted to the EPA Office of Pesticide Programs.

                                           25

-------
Chlorpyritbs                                                                     April 1992
Eliason, D.A., M.F. Cranmer, D.L. von Windeguth, J.W. Kilpatrick, I.E. Suggs and H.F. Schoof.
       1969.  Dursban premises applications and their effect on the cholinesterase levels of
       spraymen.  Mosq. News 29:591-595.

Everett, R.W. 1982. Effect of Dorsban 44 on semen output of Holstein bulls. J.  Dairy Sci.
       65:1781-1794.

FAO/WHO. 1984.  Food and Agriculture Organization of the United Nations/World Health
       Organization.  Guide to codex recommendations concerning pesticide  residues.  Part 2:
       Maximum limits for pesticide residues.  Rome:  FAO/WHO.

Gaines, T.B.  1969. Acute toxicity of pesticides. Toxicol. Appl. Pharmacol. 14:515-5344.

Gartrell, M.J., J.C. Craun, D.S. Podrebarac and E.L. Gunderson.  1985.  Pesticides, selected
       elements, and other chemicals in adult total diet samples, October 1979 - September
       1980.  J. Assoc. Off. Anal. Chem. 68:1,184/1,197.

Gentile, J.M., GJ. Gentile, J. Bultman, R. Sechriest, E.D. Wagner, and MJ. Plewa.  1982. An
       evaluation of the genotoxic properties of insecticides following plant and animal
       activation.  Mutat. Res. 101:19-29.

Hayes, A.L., R.A. Wise and F.W. Weir.  1980.  Assessment of occupational exposure to
       organophosphates in pest control operators. Am. Ind. Hyg. Assoc. J. 41:568-575.

Hodgson, MJ., G.D. Block and D.K. Parkinson.  1986.  Organophosphate poisoning in office
       workers.  J. Occup. Med. 28:434-437.

Kada, T., K. Hirano and Y. Shirasu. 1980. Screening of environmental chemical mutagens by
       the rec-assay system with Bacillus subtilis.  In: de Serres, F.T. and A. Hollaender, eds.
       Chemical mutagens: Principles and methods  for their detection.  Vol. VI.  New York:
       Plenum, pp. 149-174.

Kaur, P. and I.S. Grover.  1985. Cytologica! effects of some organo-phosphorus pesticides.  I.
       Mitotic effects.  Cytologia 50:187-197.

Kenaga, E.E.  1980. Correlation of bioconcentration factors of chemicals in aquatic  and
       terrestrial organisms with their physical and chemical properties.  Environ. Sci. Technol.
       14:553-556.-

Lehman, A. 1959.  Appraisal of the safety of chemicals in foods, drugs, and cosmetics.
       Association of Food and Drug Officials of the United  States.

Lores,  E.M. and D.E.  Bradway.  1977. Extraction and recovery of organophosphorus
       metabolites from urine using an anion exchange resin. J. Agric. Food Chem.  25:75-79.

-------
Chlorpyritbs                                                                     April  1992
Lotti. M., A. Moretto, R. Zoppellari, R. Dainese, N. Rizzuto and G. Barusco.  1986. Inhibition
       of lymphocytic neuropathy target esterase predicts the development of organophosphate-
       induced delayed polyneuropathy. Arch. Toxicol. 59:176-179.

Ludwig, P.D., DJ. Kilian, HJ. Dishburger and H.N. Edwards.  1970.  Results  of human
       exposure to thermal aerosols containing Dursban" insecticide. Mosq. News 30:346-354.

Madder,  DJ.  1977.  The disappearance from efficacy in and effect on nontarget organisms of
       diflubenzuron, methoprene and chlorpyrifos in a lentic ecosystem.  Master's thesis.
       University f Manitoba.

Majczakowa, W., H. Badach, Z. Soczewinska-Klepacka and A. Molocznik.  1985.  Evaluation of
       the conditions of work while using pesticide in the granular form - Dursban 5G.
       Medycyna Wiejska. 20(4):269-278. (In Polish; summary in English.)

McCollister, S.B., R.J. Kociba, C.G. Humiston, D.D. McCollister and PJ. Gehring. 1974.
       Studies of the acute and long-term oral  toxicity of chlorpyrifos (0,0-diethyl-0(3,5,6-
       trichloro-2-pyridyl)phosphorothioate).  Food Cosmet. Toxicol. 12:46-61.

Miles, J.R.W., C.M. Tu and C.R. Harris.  1979. Persistence of eight organo phosphorus
       insecticides in sterile and nonsterile mineral and organic soils. Bull. Environ. Contam.
       Toxicol. 22:312-318.

Moriya, M., T. Ohta, K. Watanabe, T. Miyazawa, K. Kato and Y. Shirasu.  1983.  Further
       mutagenicity studies on pesticides in bacterial reversion assay systems.  Mutat. Res.
       116:185-216.

Mostafa, I.Y., Y.M. Adam and S.M.A.D. Zayed.  1983. Bioalkylation of nucleic acids in mice
       by insecticides.  I.  Alkylation of liver RNA and DNA by chlorpyrifos. Z. Naturforsch.
       38c:461-464.

Muscarella, D.E., J.F. Keown and S.E.  Bloom.  1984. Evaluation of the  genotoxic and
       embryotoxic potential of chlorpyrifos  and its metabolites in vivo and in vitro. Environ.
       Mutagen. 6:13-23.

Nelson, J.H. and E.S. Evans, Jr.  1973.  Field evaluation of the  larvicidal  effectiveness, effects on
       nontarget species and environmental  residues of a slow-release polymer formulation of
       chlorpyrifos:  March-October 1973. Study No. 44-022-73/75.  U.S. Army Environmental
       Hygiene  Agency.

Nolan, R.J., D.L. Rick, N.L. Freshour and J.H. Saunders. 1984.  Chlorpyrifos:
       Pharmacokinetics in human volunteers.  Toxicol. Appl. Pharmacol. 73:8-15.

-------
Chlorpyrifos                                                                     April 1992
Schaefer, C.H. and E. Dupras, Jr.  1969.  The effects of water quality,  temperature and light on
       the stability of organophosphorus larvicides used for mosquito control. Proc. Papers
       37th Ann. Conf. Calif. Mosq. Control Assoc., pp. 67-75.

Schaefer, C.H. and E.F. Dupras, Jr. 1970. Factors affecting the stability of Dursban in polluted
       waters. J. Econ. Entomol. 63:701-705.

Sears, M.K and R.A. Chapman.  1979.  Persistence and movement of four insecticides applied
       to turfgrass. J. Econ. Entomol. 72:272-274.

Seehy, M.A., M. Moussa  and E. Badr.  1984.  Induction of reverse mutation in waxy locus of
       Zea mays pollen grain by pesticides. Egypt. J. Genet. Cytol.  13:137-142.

Shirasu, Y., M. Moriya, K. Kato, A. Furuhashi and T. Kada. 1976.  Mutagenicity screening of
       pesticides in the microbial system.  Mutat.  Res. 40:19-30.

Smith, G.N., B.S.. Watson and F.S. Fischer.  1967.  Investigations on Dursban insecticide.
       Metabolism of [J6Cl]0,0-diethyl 0-3,5,6-trichloro-2-pyridyl phosphorothioate in rats.  J.
       Agric. Food Chem. 15:132-138.

Sobti, R.C., A. Krishan and C.D. Pfaffenberger. 1982. Cytokinetic and cytogenetic effects of
       some agricultural  chemicals on human lymphoid cells in vitro:  organophosphates.
       Mutat. Res. 102:89-102.

Sultatos, L.G., L.G. Costa and S.D. Murphy.   1981. The role of glutathione in the
       detoxification of chlorpyrifos and methyl chlorpyrifos in mice. Pharmacologist 23:214.

Sultatos, L.G., L.G. Costa and S.D. Murphy.   1982. Factors involved in the differential acute
       toxicity of the insecticides chlorphyrifos and methyl chlorpyrifos in mice.  Toxicol. Appl.
       Pharmacol. 65:144-152. '

Sultatos, L.G., L.D. Minor and S.D. Murphy.  1985. Metabolic activation of phosphorothioate
       pesticides:  Role of the liver. J.  Pharmacol. Exp. Then 232:624-628.

Sultatos, L.G. and S.D. Murphy.  1983.  Hepatic microsoraal detoxification  of the
       organophosphates paraoxon and chlorpyrifos in the mouse.  Drug Metab. Dispos. 11:232-
       238.

U.S. EPA.  1982.  U.S. Environmental Protection Agency.  Method 622 - The determination of
       organophosphorus pesticides in industrial and municipal wastewater. Cincinnati, OH:
       Environmental Monitoring and Support Laboratory. January.

U.S. EPA.  1986.  U.S. Environmental Protection Agency.  Guidelines for carcinogen risk
       assessment.  Fed.  Reg. 51(85):33992-34003   September 24.

-------
Chlorpyrifos                                                                    April 1992
U.S. EPA.  1988.  U.S. Environmental Protection Agency.  Method 507--The determination of
       nitrogen- and phosphorus-containing pesticides in water by gas chromatography with a
       nitrogen-phosphorus detector. Cincinnati, OH:  Environmental Monitoring and Support
       Laboratory.  December.

Waters, M.D., S.S. Sandhu, V.F. Simmon, K.E. Mortelmans, A.D. Mitchell, T.A. Jorgenson,
       D.C.L. Jones, R.  Valencia and N.E. Garrett.  1982. Study of pesticide genotoxicity.
       Basic Life Sci. 21:275-326.

Windholz, M., S. Budavari, R.F. Blumetti and E.S.  Otterbein, eds.  1983. The Merck index—An
       encyclopedia of chemicals, drugs, and biologicals, 10th ed. Rahway, NJ:  Merck and
       Company, Inc., pp. 309-310.

Worthing, C.R. and S.B. Walker, eds. 1987. The pesticide manual, 8th ed. Lavenham, Suffolk:
       The British Crop Protection Council, pp. 179-180.
                                           29

-------
EPA 0553

RX000027511
                                                                                     April 199ฃ
                                           ISOPHORONE

                                    Drinking Water Health Advisory
                                           Office of Water
                                 U.S. Environmental Protection Agency
       I.  INTRODUCTION
             The Health Advisory (HA) Program, sponsored by the Office of Water (OW), provides
       information on the health effects, analytical methodology and treatment technology that would
       be useful in dealing with the contamination of drinking water.  Health Advisories describe
       nonregulatory concentrations of drinking water contaminants at which adverse health effects
       would not be anticipated to occur over specific exposure durations.  Health Advisories contain
       a margin of safety to protect sensitive members of the population.

             Health Advisories serve as informal, technical guidance to assist Federal, State and local
       officials responsible for protecting public health when emergency spills or contamination
       situations occur.  They are not to be construed as legally enforceable Federal standards. The
       HAs are subject to change as new information becomes available.

             Health Advisories are developed for One-day, Ten-day, Longer-term  (approximately 7
       years, or 10% of an individual's  lifetime) and Lifetime exposures based on data describing
       noncarcinogenic end points of toxicity. For those substances that are  known or probable
       human carcinogens, according to the Agency classification scheme (Group A or B), Lifetime
       HAs are not recommended.  The chemical concentration values for Group A or B carcinogens
       are correlated with carcinogenic risk estimates by employing a cancer  potency (unit risk) value
       together with assumptions  for lifetime exposure and the consumption of drinking water.  The
       cancer unit risk is usually derived from the linear multistage model with 95% upper confidence
       limits. This provides a low-dose estimate of cancer risk to humans that is considered unlikely
       to pose a carcinogenic risk in excess of the stated values.  Excess cancer risk estimates may
       also be calculated using the one-hit, Weibull, logit or probit models. There is no current
       understanding of the biological mechanisms involved in cancer to suggest that any one of these
       models is able to predict risk more accurately than another. Because  each model is based on
       differing assumptions, the estimates that are derived can differ by several orders of magnitude.

-------
ISOPHORONE                                                               April 1997.

  II.  GENERAL INFORMATION AND PROPERTIES

 . CAS No.  78-59-1

  Structural Formula
  Svnonvms (CAS, 1988)
        •   3,5,5-Trimethyl-2-cyclohexen-l-one; l,l,3-Trimethyl-3-cyclohexen-5-one;
           Trimethylcyclohexenone, Isoacetophorone

  Uses (NIOSH, 1978)

        •   Isophorone is used primarily as a solvent for many types of lacquers, including vinylic
           coating resins. It is also used as a solvent/cosolvent for polyvinyl and nitrocellulose
           resins, pesticides, herbicides, fats, oils and gums.  Isophorone is used as a chemical
           intermediate in the manufacture of other solvents and plant growth retardants. It
           has also been used as a repellent to stop woodpeckers from damaging utility poles.

  Properties (Amoore and Hautala, 1983; Union Carbide,  1968; Veith et al., 1980; Hawley, 1981)
           Chemical Formula
           Molecular Weight                   138.21
           Physical State                       Liquid
           Boiling Point                        215.28C
           Melting Point                       -8.1ฐC
           Density (20ฐC)                      0.923 g/mL
           Vapor Pressure (25ฐC)               0.44 mm Hg
           Specific Gravity                     0.923
           Water Solubility (20ฐC)               12 g/L
           Log Octanol/Water Partition          1.67
           Coefficient (log K,,)
           Taste Threshold
           Odor Threshold (air)                0.2 ppm (peppermint-like)
           Conversion Factors (25ฐC)            ppm = 5.65 mg/m3
                                              mg/m3 = 0.18 ppm

-------
ISOPHORONE                                                                 April

  Occurrence

        •  Production figures for isophorone were not available. The greatest potential for
           exposure is probably in the workplace.  NIOSH (1978) estimated that over
           1.5 million persons were occupationally exposed to isophorone, primarily via
           inhalation and dermal contact.

        •  Isophorone has been detected in drinking water. The highest concentration of
           isophorone reported in finished drinking water was 9.5 /ig/L (U.S. EPA, 1975).
           Lesser concentrations of isophorone have been found in drinking water in Cincinnati
           (0.02 /ig/L) and New Orleans (1.5 to 2.9 /ig/L) (U.S. EPA, 1974).  Trace quantities
           (<0.01 ppb) of isophorone have been reported in the Delaware River near an
           industrial area (Sheldon and Hites 1978), in the wastewater from a tire
           manufacturing plant (Jungclaus et al., 1976) and in effluents from  latex and chemical
           plants (Shackelford and Keith, 1976).

  Environmental Fate

        •  Little has been published with regard to the environmental fate of isophorone.  In
           aqueous solutions, isophorone is converted by sunlight into three different tricyclic
           diketodimers (Jennings, 1965). The significance of this reaction in reducing the
           concentration of isophorone  in surface water is unknown. Isophorone is degraded'by
           microorganisms in both domestic wastewater and in synthetic saltwater (Price, et al.,
           1974).


  III. PHARMACOKINETICS

  Absorption

        •  No data were found on how much isophorone is absorbed by any route of
           administration in animals or  humans. However, evidence of systemic toxicity
           following oral administration indicates that absorption does occur.
  Distribution
           No data were found on the distribution of isophorone to specific tissues by any route
           of exposure in humans or animals.
  Metabolism
           Dutertre-Catella et al. (1978) identified metabolites in Wistar rats and New Zealand
           rabbits that received a single oral dose of isophorone at 1 g/kg bw in olive oil by
           gavage. Metabolites included:

-------
ISOPHORONE                                                                April 199Z

              5,5-dimethyl-2-cyclohexen-l-one-3-carboxylic acid, thought to be formed from
              methyl oxidation.

              isophorol (3,5,5-trimethyl-2-cyclohexen-l-ol), found as the glucuronide conjugate
              and formed by reduction of the ketone.

           -  dihydroisophorone (3,5,5-trimethylcyclohexanone) resulting from the
              hydrogenation of the cyclohexene double bond.
  Excretion
           Following oral administration of isophorone (1 g/kg bw), rabbits partially eliminated
           isophorone unchanged in expired air and in the urine.  The remainder was oxidized
           to a carboxylic acid (oxidation of methyl group) or reduced to isophorol (ketone
           reduction), which was eliminated in the urine as a glucuronide conjugate (Patty,
           1982).

           No other information on excretion of isophorone and its metabolites'in humans or
           animals was found in the literature.
  IV. HEALTH EFFECTS

  Humans

    Short-term Exposure
           Isophorone vapors produce irritation to the eyes, nose and throat of unacclimatized
           subjects after 15-minute exposures to 25 ppm (125 mg/m3) (Silverman et al., 1946).
           The majority of the 12 volunteers were not uncomfortable at 10 ppm isophorone (50
           mg/m3).

           Union Carbide (1963) reported that volunteers exposed to isophorone (via
           inhalation) found levels of 200 ppm (1,000 mg/m3) to be intolerable even for
           exposure durations of 1 minute, as were exposures of 40 ppm (200 mg/m3) for 4
           minutes.  Subjects reported eye, nose and throat irritation, headaches, dizziness and
           nausea at concentrations of 40 and 85 ppm (200 and 425 mg/m3) and faintness,
           drunkenness and a feeling of suffocation at levels of 200 and 400 ppm (1,000 mg/m3
           and 2,000 mg/m3) isophorone.

           Silverman et al. (1946) reported that  unacclimatized persons (12 males and 12
           females) found 25 ppm of isophorone (15 minutes) caused irritation of eyes, nose
           and throat.

           N1OSH (1978) reported that occupational workers complained of fatigue and malaise
           (not specified) after 1 month of working at levels of 5 to 8 ppm (25 to 40 mg/m3)

-------
ISOPHORONE                                                                April 1992

           isophorone.  After workroom levels were reduced to 1 to 4 ppm (5 to 20 mg/m3)
           through improvements in ventilation, no further complaints were received.

    Long-term Exposure

        •  Workers in a screen printing plant were routinely exposed to several organic solvents
           in air, including isophorone (Samimi, 1982).  Workers were reported to have
           time-weighted average (TWA) exposures to isophorone of 8.3 to 23 ppm. No reports
           of worker complaints were mentioned.

  Animals

    Short-term Exposure

        •  The oral LD^ of isophorone has been reported to be 2.0 and 2.37 g/kg bw for mice
           and rats, respectively (Patty, 1982).

        •  A skin penetration LDj<, of 1.39 g/kg (in rabbits) was reported in a technical data
           booklet  (Union Carbide, 1968).  This result was obtained from a 24-hour covered
           skin contact with isophorone. No details on experimental protocol such as the
           number  of animals exposed were presented.

        •  When rats were exposed for 8 hours in an isophorone-saturated atmosphere
           (approximately 580 ppm), 1/6 animals died (Union Carbide, 1963).

        •  Groups  of 10 male Wistar rats were exposed to 5-17.8 mg/L (8,853, 150 ppm) of
           isophorone for 4 hours.  An LCX of 7.0 mg/L (1,240 ppm) was calculated (Hazleton
           Laboratories, 1965).

        •  At low concentrations, isophorone  caused respiratory irritation which was measured
           by the reflex decrease in respiratory rate. Six male  Swiss OF, mice were exposed to
           4 to 100 ppm (23 to 565 mg/m3) of isophorone for 5 minutes (De Ceaurriz et al.,
           1981). The concentration of isophorone causing a 50% decrease in respiratory rate
           was 28 ppm (157 mg/m3).

        •  The neurobehavioral effects of isophorone were quantified by measuring the
           duration of immobility in a "behavioral despair" swimming test (De Ceaurriz et al.,
           1984). Mice (10 male Swiss OF, mice) were  exposed to 89 to 137 ppm (500 to 770
           mg/m3) isophorone for 4 hours.  The concentration  that produced a 50% decrease in
           immobility (IDjo) was 110 ppm (620 mg/m3).

        •  Male and female CD rats (10/sex) were exposed via inhalation to isophorone for 4
           weeks.  The average measured concentration of isophorone was 0.208 mg/L (37 ppm)
           5 days/week for 6 hours/day. These animals were compared with an unexposed
           control group (Hazleton Laboratories, 1968). Mild  transient nasal bleeding was
           reported in treated rats. No other abnormal clinical signs were found.  Body weight

-------
ISOPHORONE                                                                April 1992.

           gain was reduced significantly in male rats. Absolute and relative liver weights were
           also reduced in males.  No exposure-related effects were found by gross necropsy.
           Hematologic parameters were not affected by treatment.

        •  Ten  male Wistar rats and male and female guinea pigs (10/sex/group) were exposed
           via inhalation to isophorone for 6 weeks. The concentrations ranged from 25,500
           ppm (1,402,830 mg/m3) 8 hours/day, 5 days/week.  No effects were seen in animals
           exposed to 25 ppm (140 mg/m3) (Smyth et al., 1942). No distinction was made
           between  rats and guinea pigs, since the toxic effects found in both species were
           similar. Histopathologic changes were found in animals that survived exposure to
           >50 ppm (280 mg/m3). These included effects on the lung (congestion), kidney
           (dilation of Bowman's capsule, cloudy swelling), liver (congestion) and spleen
           (congestion). Deaths were reported at concentrations of >100 ppm (565 mg/m3).
           Histologic examination of  animals that died from isophorone exposure revealed
           severe lung and kidney damage. Reduced body growth was noted in animals exposed
           to ~100 ppm (565 mg/m3). Other effects, reported only at 500 ppm (2,830 mg/m3),
           included excretion of albumin in the urine, hematologic changes, conjunctivitis and
           nasal irritation. Thus, the NOAEL and LOAEL are identified as 25-and 50 ppm,
           respectively.

  Dermal/Ocular Effects

        •  Isophorone had a weak irritant action on rabbit  and guinea pig skin. The dermal
           LDM was reported to be 1,390 mg/kg bw (Union Carbide, 1968) and 1,500 mg/kg
           (Patty, 1982) for rabbits. Moderate skin irritation was observed when isophorone
           was held in contact with guinea pig skin for 24 hours.

        •  Isophorone produced corneal opacity, inflammation of eyelids, discharge and
           conjunctiva when administered to the eyes of rabbits (Truhaut et al., 1972).
           Carpenter and Smyth (1946) reported moderate corneal injury and ocular burns
           when undiluted isophorone was administered to  the eyes of rabbits.

    Long-term Exposure

        •  Nor-Am Agricultural Products (1972a) conducted a 90-day feeding study with
           isophorone in rats.  CFE rats (20/sex/group) were fed isophorone in their daily diet
           for 90 days at levels of 0, 750, 1,500 or 3,000 ppm.  Based on the actual feed intake,
           the doses are equivalent to 0, 57, 102.5 or 233.8  mg/kg/day in the males and 0, 78.9,
           163.8 or 311.8 mg/kg/day in the females respectively. No compound-related effects
           were observed in the female animals.  However, at 3,000 ppm, the male rats suffered
           a significant decrease (8 to 11%) in body weight gain from week 6 through 11 (p
           <0,01). Thus the study identifies a LOAEL of 3,000 ppm and a NOAEL of 1,500
           ppra (234 and 103 mg/kg/day, respectively).

        •  Beagle dogs (four/sex/group) were  administered isophorone for 90 days at doses of 0,
           35, 75 or 150 mg/kg bw/day in gelatin capsules (Nor-Am Agricultural Products,

-------
ISOPHORONE                                                                April 199,2.

           1972b). No compound-related effects were observed either clinically or
           histopathologically. The NOAEL for this study is identified as 150 mg/kg/day.

        •  In a 13-week study (NTP,  1986), F344/N rats (13/sex/group) were administered
           isophorone by gavage at levels of 0, 62.5, 125, 250, 500 or 1,000 mg/kg bw per day 5
           days/week, (equivalent to average daily doses of 0, 45, 89, 179, 357 or 714 mg/kg/day).
           One female rat that received the highest dose died. No compound-related gross or
           microscopic effects were observed in either sex.  Examination of the kidneys of
           high-dose and control animals confirmed a lack of nephrotoxicity. Rats receiving
           1,000 mg/kg were reported to be sluggish and lethargic after dosing, suggesting CNS
           effects. These effects indicate that the LOAEL is less than 1,000 mg/kg/day
           (equivalent to an average daily dose of 714 mg/kg/day), and the NOAEL  is  less than
           500 mg/kg (equivalent to an  average daily dose of 357 mg/kg/day).

        •  In a 13-week study (NTP,  1986) of B6C3F, mice (10/sex/group) administered
           isophorone by gavage (5 days/week) at doses of 0, 62, 125, 250, 500 or 1,000 mg/kg
           bw.  Three of 10 females administered 1,000 mg/kg isophorone died before  the end
           of the study.  Investigators considered the deaths to be compound-related. No
           compound-related gross or microscopic effects were observed in either sex, and
           examination of the kidneys of high dose  and control animals confirmed a  lack of
           nephrotoxicity.  Significant mortality at the high dose indicates that the LOAEL is
           below 1,000 mg/kg (equivalent to an average daily dose of 714 mg/kg/day) and the
           NOAEL is below 500 mg/kg (equivalent to an average daily dose of 357 mg/kg/day).

        •  A 2-year bioassay was conducted by NTP (1986) on F344/N rats (50/sex/group).
           Isophorone was administered by gavage at doses of 0, 250, or 500 mg/kg bw 5
           days/week for 2 years.  These doses are equivalent to average daily doses  of 0, 179,
           or 357 mg/kg/day.  The overall incidence of nephropathy was similar between dosed
           and vehicle control male rats. (Dosed male rats exhibited increased mineralization
           of the kidney tubules [control, 1/50; low dose,  31/50; high dose, 20/50], and epithelial
           hyperplasia of the renal pelvis [control, 0/50; low dose, 5/50; high dose 5/50]).
           Nephropathy incidence in female rats was somewhat increased (low dose,  39/50; high
           dose, 32/50) compared to female vehicle controls (21/50). Based on the conditions of
           this study, the LOAEL was 179 rag/kg/day (the lowest dose tested).

        •  B6C3F, mice (50/sex/group)  were administered 0, 250 or 500 mg/kg bw isophorone by
           gavage 5 days/week for 2 years (NTP, 1986).  Dosed male mice showed an increased
           incidence  of hepatic coagulative necrosis (control 3/48; low dose,  10/50; high dose,
           10/50) and hepatomagaly (control, 23/48; low dose, 39/50; high dose, 37/50) when
           compared to controls.  These effects in female mice were comparable in treated and
           control animals.  Chronic focal inflammation was also observed at increased
           incidences in dosed male mice.  Based on the  conditions of this study, the LOAEL
           was 250 mg/kg (equivalent to an average daily dose of 179 mg/kg/day).

-------
ISOPHORONE                                                                April 1991

    Reproductive Effects

        •  No reports of studies on reproductive effects associated with exposure to isophorone
           were found in the literature.

    Developmental Effects

        •  An inhalation teratology study was conducted with pregnant F344/N rats and CD-I
           mice (22/group) at isophorone levels of 0, 25, 50 or 115 ppra for 6 hours/day on days
           6 to 15 of gestation (Bio/dynamics, 1984). No statistically significant differences
           among control and treatment groups were found for any of the fetal external, visceral
           or skeletal parameters in either rats or mice.  However, maternal toxicity was noted
           at 115 ppm in both rats and mice as evidenced by statistically significant differences
           in mean body weight and food consumption of the dams.  Rats also had increased
           alopecia and genital staining. The NOAEL for this study in rats and mice appears to
           be 50 ppm (2,500 mg/m3).

    Mutagenicitv

        •  Isophorone showed a weak mutagenic response in the L5178Y tk+/tk- Mouse
           Lymphoma Forward Mutation Assay in the absence of S9 mix (McGregor et
           al., 1988). Isophorone was not mutagenic in Salmonella strains TA100, TA1535 or
           TA98, with or without activation (NTP, 1986;  Cheh, 1986). However, when
           isophorone was chlorinated under conditions similar to those used in wastewater
           chlorination, mutagenic activity by the Ames/Salmonella assay using strain TA100 was
           increased  (Cheh, 1986).

        •  There are conflicting results as to whether isophorone is mutagenic in mammalian
           cells in vitro. When tested at concentrations of 4,001,200 /ig/mL, positive results
           were obtained in the mouse lymphoma L5178Y/TK ฑ assay in the absence of
           metabolic activation (Nil', 1986).  Two subsequent replications of this study by
           McGregor et al. (1988) verified these results.  However, when this assay was
           performed in another laboratory, using similar concentrations of isophorone
           (1,301^300 Mg/mL), the results were negative, with or without activation (O'Donoghue
           et al., 1988).

        •  Isophorone induced sister-chromatid exchanges  in Chinese hamster ovary cells in the
           absence (NTP, 1986), but not the presence of, Arochlor 1254-induced rat liver S9 mix
           (O'Donoghue et al., 1988).  In addition, isophorone did not induce chromosomal
           aberrations in Chinese hamster ovary cells,  with or without S9 mix (NTP, 1986), or
           mouse bone marrow assay in vivo (O'Donoghue et al, 1988).

     Carcinogenicitv

        •  F344/N rats (50/sex/group) were administered isophorone by gavage at doses of 0,
           250 or 500 mg/kg bw 5 days/week for 2 years (NTP, 1986). These doses were

                                            8

-------
ISOPHORONE                                                                April 199JL

           equivalent to an average daily dose of 179 and 357 mg/kg/day.  Dosed males showed
           a slight increased incidence of renal tubular cell adenomas and adenocarcinomas
           over the vehicle controls [controls, 0/50; low-dose, 3/50 (6%); high-dose, 3/50 (6%)].
           The authors considered that these tumors were significant, since the historical
           incidence for such tumors at this laboratory was 4/1,091 (0.4%). Dosed male rats
           also had an increased incidence of epithelial hyperplasia of the renal pelvis, and
           among the high-dose males, 4/50 animals also exhibited tubular cell hyperplasia.
           High-dose males had a significantly higher  incidence of carcinomas of the preputial
           gland than did the low-dose or control males  (controls and low-dose, 0/50; high-dose,
           5/50).  However, the prepuce is among those  tissues examined microscopically only
           when a neoplasm is visible to the prosector. Therefore, the actual incidence of all
           types  of proliferative lesions of the prepuce is not known, since only five high-dose
           males and two low-dose females were sampled for histological examination, and the
           diagnosis or actual occurrence of preputial  gland tumors has been  sporadic in vehicle
           controls in previous National Toxicology Program (NTP) studies.  (The NTP
           historical control values for preputial gland tumors ranged from 0  to 7, and 5 were
           observed in corn oil vehicle controls in one previous comparable NTP study in the
           same  laboratory.) Therefore, the NTP concluded that isophorone  exposure produced
           some evidence of carcinogicity  in male rats but not in females.

        •  Isophorone was administered by gavage to  B6C3F, mice (50/sex/group) at doses of 0,
           250 or 500 mg/kg bw 5 days/week for 2 years  (NTP, 1986). These  doses were
           equivalent to an average daily dose of 179 and 357 mg/kg/day.  In the high-dose male
           mice, isophorone exposure appears to be associated with a statistically significant
           increase in hepatocellular adenomas or carcinomas compared with low-dose and
           control male mice (control, 18/48 [38%]; low-dose, 18/50 (36%); high-dose, 29/50
           [58%]). High-dose male mice also showed  an increase in mesenchymal tumors
           (fibroma, fibrosarcoma, neurofibrosarcoma or sarcoma) of the integumentary system
           (control, 6/48; low-dose, 8/50; high-dose, 14/50). There was an increase in
           lymphomas or leukemias in the  low-dose male mice but not in the  high-dose group.
           However, the survival of male mice was low (final rates: control, 16/50; low-dose,
           16/50; high dose, 19/50). The NTP, owing to  the reduced survival,  analyzed the data
           with Life Table Test, and found no significant elevation or trends for these sites.
           Although the unadjusted tests showed significant elevations, the survival was so low,
           including early deaths and high control losses, the results in male mice are
           considered by NTP and the U.S. Environmental Protection Agency (EPA) as
           "equivocal."   No treatment-related neoplasms were observed in female mice.


  V.  QUANTIFICATION OF TOXICOLOGICAL EFFECTS

        Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
  (approximately  7 years) and Lifetime exposures if adequate data are available that identify a
  sensitive noncarcinogenic end point of toxicity.  The HAs for noncarcinogenic toxicants are
  derived using the following formula:

-------
ISOPHORONE                                                               April 1992.

            (NQAEL or LOAEL) x
                  (UF) (	L/day)

  where:

       NOAEL or LOAEL     =                No- or Lowest-Observed-Adverse-Effect
                                               Level (in mg/kg bw/day).

                      BW     =                Assumed body weight of a child (10 kg) or
                                               an adult (70 kg).

                      UF     =                Uncertainty factor, (10, 100, 1,000 or
                                               10,000, in accordance with EPA or
                                               NAS/OW guidelines.

                 	L/day     =                Assumed daily water consumption of a
                                               child (1 L/day) or an adult (2 IVday).

  One-day Health Advisory

        No data were found in the available literature that were suitable for the determination
  of the One-day HA value for isophorone. Therefore, it is recommended that the
  Longer-term HA value for a 10-kg child (15 mg/L, calculated below) be used as a
  conservative estimate for a One-day exposure.

  Ten-day Health Advisory

        No data were found in the available literature that were suitable for the determination
  of the Ten-day HA value for isophorone. Therefore, it is recommended that the Longer-term
  HA value for a 10-kg child (15 mg/L, calculated below) be used as a conservative estimate of
  a Ten-day exposure.

  Longer-term Health Advisory

        The 90-day subchronic  study in dogs (Nor-Am Agricultural Products, 1972b) was
  selected as the basis for the Longer-term HA for isophorone.  In the study, a NOAEL of
  150 mg/kg/day was identified. The dietary study in rats (Nor-Am Agricultural Products,
  1972a) identified a lower NOAEL of 103 mg/kg/day, had been considered as  the basis for the
  Longer-term HA. However, in the subchronic study in the dog,  isophorone was administered
  via capsule which provided for a better control of the dose.  The actual amount of isophorone
  ingested by rats in the dietary study might be less than the amount assumed (partly due to
  evaporation), making the NOAEL uncertain and the dog study identified a higher  NOAEL,
  making it the most appropriate study.
                                           10

-------
ISOPHORONE                                                                April 199Z

        The Longer-term HA for a 10-kg child is calculated as follows:


                              (150 mg/kg/day) (10 ke)
     Longer-term HA =           (100) (1 L/day)	      = 15 mg/L O5-000 ^S71-)


              150 mg/kg/day       =     NOAEL, based on the 90-day dog study (Nor-Am
                                        Agricultural Products, 1972b).

                     10 kg       =     Assumed body weight of a child.

                       100       =     Uncertainty factor, chosen in accordance with EPA
                                        or NAS/OW guidelines for use with a NOAEL  from
                                        a study in animals.

                    1 L/day       =     Assumed daily, water consumption of a child.


        Since isophorone has been classified by EPA as a Group C carcinogen based on limited
  evidence of carcinogenicity in animals, the chemical is considered to have potential to cause
  cancer in humans.  Therefore, the Longer-term HA is kept at 15 mg/L and  not be rounded
  up to 20 mg/L.  For prudent purposes, the Longer-term HA for a child (15  mg/kg/day,
  calculated above) is also recommended as the Longer-term HA for an adult.

  Lifetime Health Advisory

        The Lifetime HA represents that portion of an individual's total exposure that is
  attributed to drinking water and is considered protective of noncarcinogenic adverse health
  effects over a lifetime exposure.  The  Lifetime HA is derived in a three-step process.  Step  1
  determines the Reference Dose (RfD), formerly called the Acceptable Daily Intake (ADI).
  The RfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily
  exposure to the human population (including sensitive subgroups) that is likely to be without
  appreciable risk of deleterious health  effects during a lifetime,  and is derived from the
  NOAEL (or LOAEL), identified from a chronic (or subchronic) study, divided by  an
  uncertainty factor(s).  From the RfD,  a Drinking Water Equivalent Level (DWEL) can be
  determined (Step 2).  A DWEL  is a medium-specific (i.e., drinking water) lifetime exposure
  level, assuming 100% exposure from that medium, at which adverse, noncarcinogenic health
  effects would not be expected to occur. The DWEL is derived from the multiplication of the
  RfD by the assumed body weight of an adult and divided by the assumed daily water
  consumption of an adult.  The Lifetime HA in drinking water alone is determined in Step 3
  by factoring in other sources of exposure, the relative source contribution (RSC).   The RSC
  from drinking water is based on  actual exposure data or, if data are not available, a value of
  20% is assured.

        If the contaminant is classified as a known, probable, or possible human carcinogen,
  according to the Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986),

                                           11

-------
ISOPHORONE                                                                April 1992.

  then caution must be exercised in making a decision on how to deal with possible lifetime
  exposure to this substance.  For human (A) or probable (B) human carcinogens, a Lifetime
  HA is not recommended. For possible (C) human carcinogens, an additional 10-fold safety
  factor is used in the calculation of the Lifetime HA.  The risk manager must balance this
  assessment of carcinogenic potential and the quality  of the data against the likelihood of
  occurrence and significance  of health effects related  to noncarcinogenic endpoints of toxicity.
  To assist the risk manager in this process, drinking water concentrations associated with
  estimated excess lifetime cancer risks over the range of 1 in 10,000 to 1 in  1,000,000 for the
  70-kg adult drinking 2 L of water/day are provided in the Evaluation of Carcinogenic
  Potential section.

       The subchronic study in dogs (Nor-Am Agricultural Products, 1972b) has been selected
  as the basis for the RfD for isophorone. In this study, male and female dogs were fed
  isophorone in capsules at  0, 35, 75 or 150 mg/kg/day  for 90 days. All animals survived the
  study with no signs of adverse effect. A NOAEL of  150 mg/kg/day was identified.  The NTP
  (1986) 2-year bioassay in rats and mice demonstrating systemic effects at an average daily
  dose of 179 mg/kg/day supports the  selection.
                                                                         •
       As described above, the RfD  was calculated using the NOAEL of 150 mg/kg/day from
  the subchronic dog study (Nor-Am Agricultural Products, 1972b).

  Step 1:  Determination of the Reference Dose (RfD)


       RfD =     *—(l"ooof  ^^      =     ฐ'15 MSfc&tey (rounded to 0.2 mg/kg/day)

  where:

             150 mg/kg/day     =   NOAEL, based on the subchronic study in dogs.

                      1,000     =   Uncertainty factor, chosen in accordance with EPA or
                                   NAS/OW guidelines for  use with a NOAEL from a
                                   subchronic study in animals.

  Step 2:  Determination of the Drinking Water Equivalent Level (DWEL)


       ™I/CT        (0.2  mg/kg/dav) (70 kg)         _   n  ^/w^   „ ^
       DWEL  =     	(2L/day)	      =    mg/^( '    /*8^

  where:

              0.2 rng/kg/day     =   RfD.

                     70 kg     =   Assumed body  weight of an adult.

                   2 L/day     =   Assumed daily water consumption of an adult.

                                            12

-------
ISOPHORONE                                                                April

  Step 3:  Determination of the Lifetime Health Advisory
  where:

               7 mg/kg/day     =    DWEL.

                      20%     =    Assumed relative source contribution from water.

                        10     =    Additional safety factor based on OW policy to account
                                    for possible carcinogenicity of Group C carcinogens.

  Evaluation of Carcinogenic Potential

        •  The International Agency for Research on Cancer (IARC) has not evaluated the
           carcinogenic potential of isophorone.

        •  Applying the criteria described in EPA's guidelines for assessment of carcinogenic
           risk (U.S. EPA, 1986), isophorone has been classified by EPA in Group C:  possible
           human carcinogen.  This  category applies to agents for which there is limited
           evidence of carcinogenicity  from  animal studies in the  absence of human data.

        •  In the NTP 2-year gavage study,  the report concluded  that there is some evidence of
           carcinogenicity in male rats due to an increased incidence of relatively rare  renal
           tubular cell tumors at 250 and 500 mg/kg/day and rare  preputial gland carcinomas at
           500 mg/kg/day. Based on the combined incidences of preputial gland and renal
           tubular cell tumors in male  rats, the  carcinogenic potency (ql*) was estimated by
           EPA to be 4 x 10'J per (mg/kg/day) using the multistage model:

        •  The Office of Water used the following mathematical models for comparison of the
           oncogenic lifetime risk for a 70-kg adult. The cancer risk estimates (95% upper
           limit) that may cause one excess cancer in 1,000,000 (10~*) population is associated
           with exposure to isophorone levels in drinking water of 9 /ig/L (multistage), 3 /ig/L
           (one hit), 0.3 /ig/L (logit), 0.1  /ig/L (Weibull), 0.05 /ig/L (multihit) and 220 /tg/L
           (probit).


  VI. OTHER CRITERIA. GUIDANCE AND STANDARDS

        •  Based on reports of eye,  nose and throat irritation, fatigue and malaise at levels
           below 10 ppm (NIOSH, 1978; Union Carbide, 1963), the American Conference of
           Governmental Industrial  Hygienists (ACGIH) has set a Threshold Limit Value
           (TLV) of 5  ppm with the stipulation that 5 ppm is also a ceiling concentration to
           prevent eye and throat irritation.  OSHA (1989) has lowered the isophorone

                                            13

-------
ISOPHORONE                                                               April 199Z.

           standard from 25 ppm to 4 ppm for an 8-hour Time-Weighted Average (TWA).
           NIOSH currently recommends a permissible exposure limit of 4 ppm (23 mg/m}) as
           a concentration for an 8-hour TWA.

  VII. ANALYTICAL METHODS

        •  Isophorone can be analyzed by either EPA Method 609 (U.S. EPA, 1984a) or 625
           (U.S. EPA, 1984b), depending on the detection level sought.  In the Methods, 1L of
           sample is extracted with methylene chloride and concentrated to ImL. For Method
           609, the solvent must be exchanged to hexane, with analysis by Flame-Ionization
           Gas Chromatography. Method 625 is the Semi-volatile Gas Chromatography Mass
           Spectrometry procedure.  A detection limit of 5.7 ^.g/L can be  reached with Method
           609.

  VIII.  TREATMENT TECHNOLOGIES

        •  Available data indicate that lime softening, granular activated carbon (GAC)
           adsorption and chemical oxidation will significantly reduce levels of isophorone in
           drinking water supplies.

        •  McCarty et al. (1982) reported the removal of isophorone at Water Factory 21, a
           wastewater treatment plant in Orange County, CA, using an excess-lime softening
           treatment process, followed by GAC.  The lime dose was 350 to 400 mg/L as CaO
           to raise the pH to 11.  The softened water was recarbonated to a pH of 8.0, filtered,
           and chlorinated.  Isophorone was present at an initial concentration of 0.3 Mg/L.
           The GAC columns contained Filtrasorb 300 carbon and were each designed for a
           capacity of 1 million gallons per day, a hydraulic flow rate of 4.9 gallons/min/sq ft
           and an empty bed contact time (EBCT) of 30 minutes.  Lime softening alone
           reduced isphorone concentration to 0.05 /ig/L and GAC further reduced isophorone
           to below 0.05 jig/L.

        •  Borup and Middlebrooks (1987) conducted tests to determine the feasibility of
           treating contaminated water by a UV light-catalyzed oxidation process with
           hydrogen peroxide (H2O2) as an oxidant.  Oxidation with  250 jig/L of hydrogen
           peroxide and UV irradiation with an intensity of 1,210 microwatts/cm2 for 60
           minutes reduced initial isophorone concentrations of 62 mg/L to below 0.05 mg/L.

        •  Data were not found for the removal of isophorone from  drinking water by
           aeration. However, isophorone is probably only slightly amenable to removal by
           aeration due to its moderate Henry's Coefficient value, high solubility and low vapor
           pressure.
                                           14

-------
ISOPHORONE                                                                April 199Z.

  IX. REFERENCES

  Amoore, I.E. and E. Hautala. 1983. Odor as an aid to chemical safety: Odor threshold limit
        values and volatilities for 214 industrial chemicals in air and water dilution. J. Appl.
        Tpncol. 3:272-290.

  Bio/dynamics.  1984. Inhalation teratology study in rats and mice.  Final Report 3223772.
        Unpublished study performed by Bio/dynamics Inc., East Millstone, NJ, for Exxon
        Biomedical Science, East Millstone, NJ.  OTS Section 4 submission Doc ID 40-855049.
        Microfiche No. OTS 0507224.

  Borup, M.B. and EJ. Middlebrooks. 1987. Photocatalyzed oxidation of toxic organics.
        Water Science and Technology 19:381-390.

  Carpenter, C.P. and H.F. Smyth Jr. 1946. Chemical burns of the rabbit cornea. Am. J.
        Ophthalmol. 29:1,363-72.

  CAS.  1988.  Chemical Abstract  Service Online Registry File. August 3.

  Cheh, A.M.  1986. Mutagen production by chlorination of methylated alpha, beta-
        unsaturated ketones.  Mutat.  Res. 169 No. 1-2:1-9.

  De Ceaurriz, J.C., J.C. Micillion, P. Bonnet and J.P. Guenier.  1981. Sensory irritation caused
        by various industrial airborne chemicals.  Toxicol. Lett. (AMST)  9(2):137-143.

  De Ceaurriz, J.C., J.C. Micillino, B. Barignac, P. Bonnet, J. Muller and J.P. Guenier.  1984.
        Quantitative evaluation of sensory irritating and neurobehavioral  properties of aliphatic
        ketones in mice.  Fd. Chem. Tox. 22(7):545-549.

  Dutertre-Catella, H., P. Nguyen, Q. Dang and R. Truhaut.  1978.  Metabolic  transformation
        of the 3,5,5-trimethyl-2-cyclohexene-l-one (isophorone).  Toxicol. Eur. Res. 1:209-216.

  Hawley, G.G., 1981. The condensed chemical dictionary, 10th ed. New  York, NY: Van
        Nostrand Reinhold Co. p.  581.

  Hazleton Laboratories.  1968. Assessment and comparison of subacute inhalation toxicities of
        three ketones.  Final Report. Prepared by Hazleton Laboratories, Inc.  Falls Church,
        VA for Exxon Chem Amers.  Houston, TX.  OTS 8d submission Doc ID. 878210935,
        Microfiche No. 206267.

  Hazleton Laboratories.  1965. UCX determination,  acute inhalation exposure - rats. Final
        Report.  Prepared by Hazleton Laboratories,  Inc., Falls Church, VA for Exxon Chem.
        Amers., Houston, TX. OTS 8d submission Doc.  ID. 878210933, Microfiche No. 206267.

  Jennings, P.  1965. Photochemistry of isophorone.  Diss. Abstr. 26:698.
                                           15

-------
ISOPHORONE                                                               April 1992,

  Jungclaus, G., L Games and R. Kites.  1976.  Identification of trace organic compounds in
        tire manufacturing plant wastes. Anal. Chera. 48:1,894-1,896.

  McCarty, D.L., D. Argo, M. Reinhard, J. Graydon, N. Goodman, and M. Aieta.  1982.
        Performance of Water Factory 21 in removing priority pollutants. Proc. Water Reuse
        Symposium: 2,325-2,349.

  McGregor, D.B., A. Brown, P. Cattanach, I. Edwards, D. McBride, C. Riach and
        WJ. Caspary. 1988. Responses of the L5178Y tk+/tk- mouse lymphoma cell forward
        mutation assay:  III. 72 coded chemicals. Environ. Molec. Mut.  12:85-154.

  NIOSH.  1978.  National Institute for Occupational Safety and Health. Criteria for a
        recommended standard, occupational exposure to ketones.  DHEW (NIOSH)
        Publication No. 78-173.  pp. 44, 75, 176.

  Nor-Am Agricultural Products. 1972a. 90-Day subchronic toxicity of isophorone in the rat.
        MRID No. 00123976.  Available from EPA.  Write to FOI, EPA, Washington, D.C.
        20460.

  Nor-Am Agricultural Products. 1972b. 90-Day subchronic toxicity of isophorone in the dog.
        MRID No. 00123977.  Available from EPA.  Write to FOI, EPA, Washington, D.C.
        20460.

  NTP.  1986. National Toxicology Program.  Toxicology and carcinogenesis studies of
        isophorone (CAS No. 78-59-1)  in F344/N rats and  B6C3F, mice (gavage studies).  NTP
        TR-291.  NIH Publication No. 86-2547.

  O'Donoghue, J.L, S.R. Haworth, R.D. Curren, P.E. Kirby, T. Lawlor, EJ. Moran, R.D.
        Phillips, D.L Putnam, A.M. Rogers,  R.S. Slesinski. 1988. Mutagenicity studies on
        ketone solvents: methyl ethyl ketone, methyl isobutyl ketone and isophorone.  Mutat.
        Res. 206(2):149-161.

  OSHA.  1989. Air contaminants. Final Rule. U.S. Department of Labor. Occupational
        Safety and Health Administration.  Code of Federal Regulations. 29 CFR 1910. Fed.
        Reg. 54(12):2941.

  Patty, F.A.  1982.  Patty's industrial hygiene and toxicology, 3rd rev. ed. Vol. 2C:4,786-4,787.

  Price, K., G. Waggy and R. Conway.  1974.  Brine shrimp bioassay and BOD of
        petrochemicals. J. Water Pollut. Control Fed. 46:63-77.

  Samimi, B.  1982.  Exposure to isophorone and other organic solvents in a screen  plant.  Am.
        Ind. Hyg. Assoc. J. 43:43-48.

  Shackelford, W. and L. Keith.  1976.  Frequency of organic compounds identified  in water.
        U.S. EPA 600/4-76-062.  Athens, GA: U.S. Environmental Protection Agency.

                                           16

-------
ISOPHORONE                                                                April 1992,

  Sheldon, L. and R. Hites.  1978.  Organic compounds in the Delaware River.  Environ. Sci.
        Technol. 12:1,188-1,194.

  Silverman, L.,  H. Schulte and M. First.  1946. Further studies on sensory response to certain
        industrial solvent vapors.  J. Ind. Hyg. Toxicol.  28:262-266.

  Smyth, H., J. Seaton and L Fischer.  1942.  Response of guinea pigs and rats  to repeated
        inhalation of vapor of mesityl oxide and isophorone.  J. Ind. Hyg. Toxicol. 24:46-50.

  Truhaut, R., H. Dutertre-Catella and P. Nguyen.  1972.  Study of the toricity of an industrial
        solvent,  isophorone. Irritating capacity with regard to the skin and the mucosae. J.
        Eur. Toxicol.  5:31-37.

  Union Carbide Corporation.  1968. Ketones booklet F-419771.  New York, NY: Union
        Carbide  Corporation.  21 p.

  Union Carbide Corporation.  1963. Toxicological studies-isophorone summary data sheet.
        New York, N.Y.:  Ind. Med. Toxicol. Dept, Union Carbide Corporation:

  U.S. EPA. 1974.  U.S. Environmental Protection Agency.  Draft analytical report: New
        Orleans  area water supply study.  Region IV, Dallas, TX.  Surveillance and Analysis
        Division, Lower Mississippi River facility, Slidell, LA.

  U.S. EPA. 1975.  U.S. Environmental Protection Agency.  Preliminary assessment of
        suspected carcinogens in drinking water.  Report to Congress. Washington, DC: U.S.
        Environmental Protection Agency.

  U.S. EPA. 1984a.  U.S. Environmental Protection Agency. U.S. EPA Method 609 -
        Nitroaromatics and isophorone. 40 CFR part 136, Oct 26.

  U.S. EPA. 1984b.  U.S. Environmental Protection Agency. U.S. EPA Method 625 -
        Base/neutrals and acids.  40 CFR part 136, Oct 26.

  U.S. EPA. 1986.  U.S. Environmental Protection Agency.  Guidelines  for carcinogen risk
        assessment. Fed. Reg.  51(185):33,992-34,003.

  Veith, G.D., KJ. Macek, S.R. Petrocelli and J. Carroll. 1980. An evaluation of using
        partition coefficients and water solubility to estimate bioconcentration factors for
        organic chemicals in fish.  ASTM STP 707. Aquatic Toxicology.  In: Easton J.G. et al.,
        eds. Amer. Soc. Test Mater, pp. 116-129.
                                            17

-------
EPA 0553                                                                         April.  199.2

RX000027511
                                           MALATfflON

                                   Drinking Water Health Advisory
                                           Office of Water
                                U.S. Environmental Protection Agency
      I.    INTRODUCTION
            The Health Advisory (HA) Program, sponsored by the Office of Water (OW), provides
      information on the health effects, analytical methodology and treatment technology that
      would be useful in dealing with the contamination of drinking water. Health Advisories
      describe nonregulatory concentrations of drinking water contaminants at which adverse health
      effects would not be anticipated to occur over specific exposure durations. Health Advisories
      contain a margin of safety to protect sensitive members of the population.

            Health Advisories serve  as informal, technical guidance to assist Federal, State and
      local officials responsible for protecting public health when emergency spills or contamination
      situations occur. They are not to be construed as legally enforceable Federal standards. The
      HAs are subject to change as new information becomes available.

            Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
      years, or 10% of an individual's lifetime) and Lifetime exposures based on data describing
      noncarcinogenic end points of toxicity.  For those substances that are known or probable
      human carcinogens, according  to the Agency classification scheme (Group A or B), Lifetime
      HAs are not recommended.  The chemical concentration values for Group A or B
      carcinogens are correlated with carcinogenic risk estimates by employing a cancer potency
      (unit risk) value together with  assumptions for lifetime exposure and the consumption of
      drinking water.  The cancer  unit risk is usually derived from the linear multistage model with
      95% upper confidence limits.  This provides a low-dose estimate of cancer risk to humans
      that is considered unlikely to pose a carcinogenic risk in excess of the stated values.  Excess
      cancer risk estimates may also be calculated using the one-hit, Weibull, logit or probit models.
      There is no current understanding of the biological mechanisms involved in cancer to suggest
      that any one of these models is able to predict risk more accurately than another.  Because
      each model is based on differing assumptions, the estimates that are derived can differ by
      several orders of magnitude.

-------
 Malathion                                                                  April. 1992
II.  GENERAL INFORMATION AND PROPERTIES

CAS No. 121-75-5

Structural Formula
                               S     CH2-C-0-C2H5
                         CHj-0-P-S-CH -C  - "
                O.O-Dimethvl S-d.Z-dicarfaethoxvethvDphosphorodithioate
Synonyms
         Calmathion, Carbetox, Carbofos, Celthion, Chemathion, Cimexan, Cythion,
         Dorthion, Emmatox, Ethiolacar, Etiol, Extermathion, Forthion, Fosfothion,
         Fosfotion, Fyfanon, Hilthion, Karbofos, Kypfos, Lucathion, Malacide, Malafor,
         Malamar, Malaphele, Malathiazol, Malathon, Malathyl, Malatol, Malatox, Maldison,
         Maltox, Mercaptothion, Phosphothion, Prioderrn, Sadofos, Sumitox, Vetiol, Zithion,
         Compound 4049 (U.S. EPA, 1988a; Meister, 1987; IARC,  1983).
Uses
      •  Malathion is an organophosphorous insecticide used extensively to control a wide
         variety of insects and mites. It is used particularly where a high degree of safety to
         mammals is desired (Meister, 1987).

      •  An estimated 15 to 20 million pounds of the active ingredient are used annually,
         based on 1985 and 1986 data (U.S. EPA, 1988a). Industrial, commercial and
         government applications constitute 40% of the annual use in the United States, and
         approximately 33% is used in and around the home. Direct application to
         agricultural crops accounts for 12% of use, and agricultural noncrop use is 15% of
         the total.

Properties (Windholz et al., 1983; Worthing and Walker,  1987; U.S. EPA, 1988a)

      Chemical Formula                   C10H,9O6PS2
      Molecular Weight                   330.36
      Physical  State (at 25ฐC)              liquid
      Boiling Point (at 0.7 mm Hg)         156-157ฐC
      Melting Point (ฐC)                   2.9
      Density (at 20ฐC)                    1.20 g/mL
      Vapor Pressure (at  30ฐC)             0.00004 mm Hg

-------
Malathion                                                                   April, 1992.
      Specific Gravity                      —
      Water Solubility (at 20ฐC)             145 mg/L
      Log Octanol/Water Partition          2.89
         Coefficient (log K^)
      Taste Threshold                     —
      Odor Threshold                     —
      Conversion Factor                   1 ppm = 1 mg/m3
Occurrence

      •  In a national surface water monitoring program (1976 to 1980), malathion was
         found in 0.3% of the samples, with a maximum value of 0.18 ppm (Carey and Kutz,
         1985).  Malathion was not detected in sediments.

      •  Ambient air concentrations of malathion as a result of crop treatment were 0.01 to
         0.60 mg/m3 in work areas, and 0.001 mg/m3 in an agricultural community (Hayes,
         1971).

      •  In the National Soils Monitoring Program, malathion was detected at levels of 0.08
         to 0.19 mg/kg in cropland soil (Carey et al., 1978, 1979).

      •  In FDA total diet studies conducted from 1982 to 1986,  no tolerance-exceeding
         residues of malathion were observed  (U.S. EPA, 1988a).

Environmental Fate

      •  Malathion at 10 mg/L was added to raw river water in closed glass containers (at
         room temperature  and exposed  to natural and artificial light) (Eichelberger and
         Lichtenberg,  1971). Within 4 weeks,  malathion residues were not detectable.

      •  Malathion is very mobile in loamy sand and sandy loam soils (LaFleur, 1979).
         Adsorption ratios determined in Norfolk loamy sand and Cecil sandy loam soils
         treated with malathion at 10, 20, 40 or 80 mmol/kg were 0.73 to 0.95, and
         desorption ratios were 0.75 to 0.95.

      •  It is not possible to fully assess the environmental fate of malathion because
         acceptable data are lacking.  Based on theoretical calculations, however, application
         of malathion  to land could result in aquatic concentrations of 0.3 mg/L owing to
         surface runoff and aerial drift (U.S. EPA, 1988a).

-------
Malathion                                                                   April. 1992_
III.  PHARMACOKINETICS

Absorption

      •  Malathion is absorbed through the gastrointestinal tract, the respiratory tract and
         the skin, as shown by the rapid onset of postexposure symptoms of poisoning (NAS,
         1977; WHO, 1982).

      •  Wester et al. (1983) determined the  percutaneous absorption of 5 mg/cm2 malathion
         repeatedly administered to the ventral  forearm skin of male volunteers for 14 days.
         Monitoring of urine indicated that dermal absorption of malathion on day 1 (4.5%)
         was not significantly different from dermal absorption on day 8 (3.5%).  Feldman
         and Maibach (1970, 1974) reported absorption of 7.8% and 8.2% after application
         of a single dose of malathion at 4 jig/cm2 to the ventral forearm of human
         volunteers.

Distribution

      •  Distribution of malathion is general, and low concentrations  are found in many
         tissues (NAS, 1977; WHO, 1982).

      •  Morgade and Barquet (1982) reported the analyses of tissues obtained at  the
         autopsy of a suicide victim (times of ingestion, death and autopsy were unavailable).
         Malathion was detected primarily in  adipose tissue (78 ppra), kidney (17 ppm),
         brain (5 ppm), spleen (1 ppm) and bile (1 ppm).  Malaoxon, the main active
         metabolite, was detected primarily in adipose tissue (8 ppm), brain (1 ppm) and
         kidney (1 ppm).  Malathion monocarboxylic acid was detected primarily in bile (221
         ppm), kidney (106 ppm) and spleen (59 ppm).

Metabolism

      •  Malathion is a phosphorodithioate. Its oxidized analogue, the phosphate malaoxon,
         is an anticholinesterase agent.  The conversion of malathion  to malaoxon is carried
         out by the liver microsomal monooxygenase system. Competing with the activation
         of malathion are phosphatase and carboxylesterase enzymes which degrade
         malathion to less toxic metabolites such as malathion monocarboxylic acid and
         dicarboxylic acid, various  phosphoric acids and the O-demethylated product. The
         degradation rate of malaoxon exceeds its activation rate, so there is generally little
         accumulation of malaoxon in mammalian systems (NAS, 1977).

Excretion

      •  Malathion is rapidly metabolized in mammals, and its metabolites are excreted
         mostly in the urine within 24 hours (IARC, 1983; WHO, 1982).

-------
Malathion                                                                   April. 199,2-
      •  Urinary excretion of l4C-malathion after intravenous administration was
         approximately 90%, with a half-life of 3 hours (Feldman and Maibach, 1974).

      •  Dermal administration of uC-malathion resulted in peak urinary excretion of
         radioactivity between 12 and 24 hours of the first day. The excretion half-life of
         dermally applied malathion was 30 hours (Wester et al., 1983).

      •  In the Health and Nutrition Examination Survey II (1976-1980) conducted by the
         National Center for Health Statistics, the malathion metabolites, malathion
         dicarboxylic acid and monocarboxylic acid, were detected in 1.1% and 0.5%,
         respectively, of human urine samples collected from individuals 12 to 74 years of
         age (Carey and Kutz, 1985).
IV. HEALTH EFFECTS

Humans

  Short-term Exposure

      •  The acute toxicity of malathion is due to inhibition of acetylcholinesterase at nerve
         endings, leading to an accumulation of endogenous acetylcholine.  The effects are
         manifested by muscarinic, nicotinic and central nervous system symptoms.  The
         cause of death is primarily respiratory failure (IARC, 1983).

      •  Technical  malathion may contain organophosphorous impurities that diminish the
         organism's ability to detoxify malathion.  Isomalathion, trimethyl phosphothioates
         and other organophosphate impurities have been shown to potentiate the toxicity of
         malathion (WHO, 1982).

      •  Moeller and Rider (1962) administered  malathion (of unspecified  purity) in
         capsules to human male volunteers at 8 mg/day for 32 days, 16 mg/day for 47 days
         and 24 mg/day for 56 days (five subjects/treatment group).  Assuming a body weight
         of 70 kg, the doses were 0.11, 0.23 and 034 mg/kg/day. There was no significant
         depression of plasma or erythrocyte cholinesterase  activity in the 0.11- and
         0.23-mg/kg/day  treatment groups (maximum cholinesterase inhibition of
         approximately 10%). At the highest dose, plasma and erythrocyte cholinesterase
         were depressed a maximum of 25% by about 3 weeks after the end of treatment.
         There were no clinical signs of poisoning. Thus, the NOAEL in humans for this
         study was  0.23 mg/kg/day and the LOAEL was 0.34 mg/kg/day.

      •  Between 1981 and 1985, malathion was  the third most common cause of pesticide
         illness in California. However, it ranked only eleventh as a cause  of hospitalized
         pesticide illnesses reported between 1982 and 1986 (U.S. EPA, 1988a; Brown et al..
         1989).

-------
 Maiathion                                                                   April. 199JL
       •  Thirty-five cases of poisoning by ingestion of unspecified amounts of malathion in
          India were examined (Chhabra et al., 1970). Pinpoint pupils, cyanosis, excessive
          salivation and pulmonary edema were observed.  Electrocardiographic abnormalities
          were observed in 37% of the cases.  Histopathological examination at autopsy
          indicated damage to the myocardium.

       •  A number of studies of workers spraying malathion have been conducted in Haiti
          (Miller and Shah, 1982; Warren et al., 1985), India (Gupta et al., 1980; Siddiqui et
          al., 1980) and Pakistan (Baker et al., 1978; Miller and Shah, 1982).  Weakness,
          headache, nausea and mild eye irritation were reported.  Miller and Shah (1982)
          measured plasma and erythrocyte cholinesterase  levels in 68 malathion sprayers. In
          12 workers,  plasma cholinesterase levels ranged from 31.8 to 48.6% of unexposed
          controls, and erythrocyte cholinesterase  levels ranged from 51.2 to 65.4% of
          controls.  Lower levels of cholinesterase inhibition were reported in workers using
          better  safety procedures.

      •   In two cases of attempted suicide by raalathion ingestion (one victim ingested
          approximately one cup; the amount  ingested by the second victim was unknown),
          serum  cholinesterase of both patients was completely inhibited for 2 or 3 days
          following admission to the hospital;  recovery to normal levels occurred in 15 to 20
          days following intensive atropine and 2-PAM therapy.  No followup was reported
          after release from the hospital (Hanna and Choo-Kang, 1983).

  Long-term  Exposure

      •   Although raalathion has been used as an insecticide for many years, no reports of
          chronic or delayed toxicity to humans resulting from long-term exposure were
          located.  There is no documented case of peripheral neuropathy due to malathion
          exposure, even though other organophosphates have been shown to have this effect
          (IARC, 1983).

Animals

  Short-term  Exposure

      •   In mammals, acute oral LDjg values have been reported as 1,000 to 1,375 mg/kg in
          Sherman rats (Gaines, 1969) and 1,680 mg/kg in mice (Berteau and Deen, 1978).
      •  Oral LDjg values in Wistar rats were age dependent.  Values were 209 mg/kg at 1 to
         2 days, 707 mg/kg at 6 to 7 days, 1,085 mg/kg at 12 to 13 days and 1,806 mg/kg at 17
         to 18 days of age (Mendoza and Shields, 1977).

      •  Male rats (five/dose) fed diets  containing 0, 100 or 500 ppm of 98% pure technical
         malathion (approximately 0, 0.5 and 25 mg/kg/day, based on the dietary assumptions
         of Lehman, 1959) for 8 weeks  had no significant depression of whole-blood

-------
Malathion                                                                    April, 1993.
         cholinesterase levels (Frawley et al., 1957).  The dose of 25 mg/kg/day, the highest
         dose tested, may be considered a NOAEL for rats in this study.

      •  In a 33-day study, rats (10/dose) were fed diets containing malathion at levels
         resulting in doses of 0, 10, 90 or 470 mg/kg/day (Golz and Shaffer, 1955).  There
         were no gross signs of toxicity related to treatment.  Cholinesterase activity was
         determined in plasma, erythrocytes, brain and liver.  Significant inhibition of
         erythrocyte Cholinesterase was observed at 90 and 470 mg/kg/day, and significant
         inhibition of plasma Cholinesterase was observed at 470 mg/kg/day.  The NOAEL
         for rats in this study was  10 mg/kg/day,  and the LOAEL was 90 mg/kg/day.

      •  Other organophosphorus insecticides can potentiate the toxicity of malathion when
         appropriate  levels are fed in the diet. The toxicity of malathion was increased as
         much as 10-fold by feeding o-ethyl-o-p-nitrophenyl phenylphosphothioate (EPN),
         parathion, azinphosmethyl, Folex and triorthotolyl phosphate (TOTP). The
         potential toxicity was correlated with  the inhibition of liver aliesterases (Su et al.,
         1971).

  Dermal/Ocular Effects

      •  A single dermal application of 4,444 mg/kg of malathion in an emulsifiable
         concentrate killed 4 out of 10 Sherman  rats (Gaines, 1969).

      •  Technical malathion is nonsensitizing and only mildly irritating to the eyes and skin
         (U.S. EPA, 1988a).

  Long-term  Exposure

      •  Mixed-breed dogs (one of each sex/group) were administered diets containing 0, 25,
         100 or  250 ppm of 98% pure technical malathion (approximately 0, 0.6, 2.5 and 6.3
         mg/kg/day based on the dietary assumptions of Lehman, 1959)  for 12 weeks
         (Frawley et al., 1957).  No significant depression of plasma Cholinesterase  was
         observed.  The highest dose  tested, 6.3  mg/kg/day, caused significant erythrocyte
         Cholinesterase inhibition  (maximum of 25%).  The level of statistical significance
         was not provided by the authors.  Thus, 2-5 mg/kg/day (100 ppm in feed) may be
         considered a NOAEL, and 6.3 mg/kg/day (250 ppm in feed) may be considered a
         LOAEL for dogs in this study.  The small number of animals in each treatment
         group reduces the degree of confidence in the results of the study.

      •  Albino rats (20 to 30 rats/group) were fed 90% pure technical malathion in the diet
         for 2 years at 0, 6, 70 or 350 mg/kg/day (Hazelton and Holland, 1953; Golz and
         Shaffer, 1955). At 6 mg/kg/day, erythrocyte, plasma and brain  Cholinesterase were
         inhibited by 10 to 30% of control  levels. At 70 mg/kg/day, 60 to 95% inhibition of
         erythrocyte Cholinesterase was observed, and at 350 mg/kg/day, 60 to 95% inhibition

-------
Malathion                                                                   April. 1992.
         of erythrocyte, plasma and brain cholinesterase was reported.  A NOAEL for
         cholinesterase inhibition was not established, but the LOAEL was 6 mg/kg/day
         based on the conditions of this study. The methods and results of this study were
         not completely reported.

      •  Pure-bred beagle dogs were administered malathion (95% purity) by capsule at
         doses of 0, 62.5, 125 or 250 mg/kg/day (six dogs/sex/group) for 12 months (Tegeris
         Laboratories, 1987).  Creatinine, serum alanine aminotransferase (SGPT) and blood
         urea nitrogen (BUN) were decreased in the high-dose animals. Elevated liver and
         kidney weights were observed at all doses, generally in a dose-related manner.
         Erythrocyte and plasma cholinesterase were inhibited at all dose levels.  Erythrocyte
         cholinesterase was inhibited at about 75% of the control activity at all doses.  A
         NOAEL for cholinesterase inhibition was not established, and the LOAEL was 62.5
         mg/kg/day.

  Reproductive Effects

      •  In a two-generation continuous feeding study (Kalow and Marton, 1961), Wistar
         rats were fed 240 mg/kg/day of 95%  pure technical grade malathion.  Males and
         females  were bred after 10 weeks of treatment. Survival of the pups was
         significantly reduced at 7 and 21 days after birth,  and growth retardation was
         observed 9 weeks after birth.  Because adverse effects were observed at the only
         dose tested, a NOAEL cannot  be established for this study.

  Developmental Effects

      •  Kimbrough and Gaines (1968)  administered malathion intraperitoneally at 600 and
         900 mg/kg to pregnant Sherman rats on day 11 of gestation.  There were no
         significant  differences between the malathion-treated females and the controls
         relative to dead fetuses per litter, resorptions, average weight of fetuses, average
         weight of placenta or malformations  of fetuses.

      •  Khera et al. (1978) administered technical malathion (purity not specified) at 50,
         100, 200 and 300 mg/kg to female Wistar rats by gastric intubation on days 6
         through  15 of gestation. There  was no evidence of fetal or maternal toxicity.

  Mutagenidty

      •  Malathion produced negative results in the Ames assay, with and without S9
         activation, in Salmonella typhimurium tester strains TA97a, TA98 and TA100, at
         concentrations up to 5 x 10'J M (Pednekar et al., 1987).  Tester strains TA1536,
         TA1537, TA1538, TA98 and TA100 were negative in another test, but malathion
         was positive at the highest assay concentration (300 mg/plate) for TA1535 and for
         Bacillus subtilis TKJ6321 strain (Shiau et al., 1980).

-------
Malathion                                                                  April. 1992-
      •  In Drosophila melanogaster, malathion did not induce point mutations when
         administered at 50 ppm in culture media. It was also ineffective in producing .total
         or partial sex-chromosome losses and nondisjunction (Velazquez et al.,  1987).

      •  In a test using human hematopoietic cell cultures, malathion inhibited cell growth in
         proportion to the doses of 50, 100, 200 and 400 mg/mL. In studies of chromosomal
         aberrations, however, the incidence of metaphases with aberrations did not increase
         (Huang, 1973).  However, Wiszkowska et al. (1986) reported chromosomal
         aberrations in lymphocyte cultures exposed to 10, 40 or 70 /xg/mL of malathion.
         Malathion induced increases in the number  of aberrant cells in the bone marrow of
         Syrian hamsters treated in vivo at 240 and 2,400 rng/kg, but the increases  were not
         dose related (Dzwonkowska and Huebner, 1986).

      •  Malathion was positive in the mouse micronucleus test after cutaneous  exposure at
         120, 240 and 480 mg/kg, but the effect was not dose  related (Dulout et  al., 1982).
         In mice treated with raalathion intraperitoneally,  subchromatid and chromatid-type
         aberrations in bone marrow cells were induced at 230 mg/kg, but net at 115 mg/kg
         (Dulout et al.,  1983). In male mice receiving an intraperitoneal injection of
         malathion at 300 mg/kg, there was no increase in chromosome aberrations in bone
         marrow cells and spermatogonia (Degraeve  and Moutschen, 1984).

      •  Malathion did not induce dominant lethal effects in mice when administered in the
         diet for 7 weeks at unspecified doses (Jorgenson et ah, 1976) or when injected
         intraperitoneally (single dose) at 300 mg/kg (Degraeve and Moutschen, 1984).

  Carcinogenicitv

      •  In National Cancer Institute bioassays (NCI, 1978, 1979a),  technical grade
         malathion was  administered to Osbome-Mendel rats, Fischer 344 rats and B6C3F,
         mice.  The Osborne-Mendel rats (50/sex/group) were administered Time-Weighted
         Average (TWA) doses of 0, 4,700 or 8,150 ppm malathion in the diet for  80 weeks,
         and subsequently observed for 29  to 33 weeks. Feed consumption rates were not
         provided, but using the dietary assumptions of Lehman (1959), the doses  could be
         estimated at 0, 235 and 408 mg/kg/day. There was no significant evidence of
         carcinogenicity (NCI, 1978; Huff et al., 1985; U.S. EPA, 1988a). The IARC
         Working Group (IARC, 1983) noted that the duration of treatment was 80 weeks,
         which  is not a rodent lifetime.

      •  Fischer 344 rats (50/sex/group) were fed 95% pure malathion in the diet at
         concentrations of 0, 2,000 and 4,000 ppm for 103 weeks (NCI,  1979a).  They were
         then observed for an additional 2  to 3 weeks. Feed consumption rates  were  not
         provided, but using the dietary assumptions of Lehman (1959), the doses could be
         estimated at 0, 100 and 200 mg/kg/day. No  treatment-related evidence  of
         carcinogenicity was observed (NCI, 1979a; Huff et al.,  1985; U.S. EPA, 1988a).

-------
Malathion                                                                   April, 199Z.
      •  Charles River B6C3F, mice (50/sex/group) were administered 95% pure raalathion
         in the diet at concentrations of 0, 8,000 and 16,000 ppra for 80 weeks and observed
         14 to 15 weeks (NCI, 1978).  Feed consumption rates were not provided, but using
         the dietary assumptions of Lehman (1959), the doses could be estimated at 0, 1,200
         or 2,400 mg/kg/day.  A significant increase in hepatocellular tumors (combined
         carcinomas plus neoplastic nodules) was observed in the high-dose male group; the
         increase was not statistically significant under a time-adjusted analysis.  NCI (1978)
         noted that historically the incidence of hepatocellular carcinoma in controls in this
         strain of mouse is often higher than that observed in the high-dose male group in
         this study. NCI (1978) and IARC (1983) concluded that no significant  evidence of
         carcinogenicity was found. IARC (1983)  and U.S. EPA (1988a) noted  that duration
         of treatment was only 80 weeks.  U.S. EPA (1988a) noted a dose-related trend in
         the incidence of hepatocellular carcinoma (p = 0.019) and increased incidence of
         hepatocellular carcinoma in the high-dose males (p = 0.031); however,  these
         findings were considered questionable.

     •  NCI conducted oncogenicity studies on the raalathion metabolite, roalaoxon, in the
         Fischer 344 rat and the B6C3F! mouse (NCI, 1979b). Fischer 344 rats
         (50/sex/group) were administered 0, 500 or 1,000 ppm of malaoxon in the diet for
         103 weeks and were observed for 1 to 2 additional weeks. Feed consumption rates
         were  not provided, but using the dietary assumptions of Lehman (1959), the doses
         could be estimated at 0, 25 or 50 mg/kg/day. No significant increase in tumor
         incidence was observed in male rats (NCL 1979b; Huff et al., 1985; U.S. EPA,
         1988a). A significant increase in combined thyroid C-cell adenomas and carcinomas
         in female rats was observed; however, this was  considered to be of questionable
         biologic relevance since historical data showed  that the incidence of these tumors
         was higher than that of the controls in this study (NCL 1979b). In a re-examination
         of the study,  Huff et al. (1985) concluded there was equivocal evidence  of
         carcinogenicity in rats with regard to thyroid C-celi adenomas and carcinomas.

     •   Charles River B6C3Ft mice were administered  malaoxon via dietary doses of 0, 500
         or 1,000 ppra for 103 weeks and  were observed for 1 to 2 additional weeks. Feed
         consumption rates were not provided, but using the dietary assumptions of Lehman
         (1959), the doses could be estimated at 0, 75 or 150 rag/kg/day. No evidence  of
         carcinogenicity of malaoxon in the mice was observed (NCI, 1979b; U.S. EPA,
         1988a).

     •   The above data in mice and rats were also evaluated by the Office of Pesticide
         Programs (U.S. EPA, 1990) and found to be flawed for the purpose of  adequately
         assessing the carcinogenic potential of this chemical. This was determined because,
         among other things,  the maximum tolerated dose may not have been reached in
         these studies.
                                          10

-------
 Malathion                                                                  April, 199^
 V.  QUANTIFICATION OF TOXICOLOGICAL EFFECTS

      Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
 (up to 7 years) and Lifetime exposures if adequate data are available that identify a sensitive
 noncarcinogenic end point of toricity. The HAs for noncarcinogenic toxicants are derived
 using the following formula:

                (NOAEL or LOAEL)  x (BW)          ...
               *	(UF) (_ Uday)        = — mg/L (rฐUnded tO
where:
    NOAEL or LOAEL      =     No- or Lowest-Observed-Adverse-Effect  Level (in
                                   rag/kg bw/day).

                   BW      =     assumed body weight of a child (10 kg)'or an adult
                                   (70 kg).

                    UF      =     uncertainty factor, (10, 100, 1,000 or 10,000), in
                                   accordance with EPA or NAS/OW guidelines.

              	L/day      =     assumed daily water consumption of a child (1 L/day)
                                   or an adult (2 L/day).
One-day Health Advisory

     No studies were located that are suitable for calculating the One-day HA. It is
recommended that the adult DWEL adjusted for a 10-kg child, 200 jig/L, calculated as
described in the Longer-term HA for the 10-kg child, be used as a conservative estimate of
the One-day HA.

Ten-day Health Advisory

     No studies were located that are suitable for calculating the Ten-day HA.  Khera et al.
(1978) administered technical malathion to pregnant Wistar rats at levels of 50,100, 200 or
300 mg/kg/day during days 6 to 15 of gestation.  Although Khera et al. (1978) did not find any
evidence of fetal or maternal toricity, this study is not suitable for determining a NOAEL
value because of lack of testing for acetylcholinesterase inhibition, a relevant endpoint for
malathion.  As reported in Golz and Schaffer (1955), significant inhibition of erythrocyte
cholinesterase was observed in rats administered malathion for 33 days at doses of 90 or 470
mg/kg/day.  Thus, some inhibition of acetylcholinesterase also may have occurred, and gone
unobserved, in the Khera et al. (1978) study, making it unsuitable for risk assessment.  It is
therefore recommended that the adult DWEL adjusted for a 10-kg child, 200 /ig/L, calculated

                                          11

-------
 Malathion                                                                  April, 199Z,
 as described in the Longer-term HA for the 10-kg child, be used as a conservative estimate of
 the Ten-day HA.

 Longer-term Health Advisory

     The study in humans by Moeller and Rider (1962) used for determining the DWEL has
 also been selected as the basis for the Longer-term HA. As described in the next section,  a
 NOAEL of 0.23 mg/kg/day and a LOAEL of 0.34 rag/kg/day for erythrocyte cholinesterase
 inhibition in humans were identified in this study.

     The Longer-term HA for a 10-kg child is calculated as follows:

     Longer-term HA  = ^"f^ (*ฐ kg)  = 0.23 mg/L  (rounded to 200 MgflL)
                             (iU) (1 iv day)

where:

     0.23 mg/kg/day   =      NOAEL, based on the absence of cholinesterase depression in
                             humans exposed orally to malathion for 32 to 56 days.

     10 kg           =      assumed weight of a child.

     10              =      uncertainty factor, this uncertainty factor was chosen in
                             accordance with EPA or NAS/OW guidelines in which a
                             NOAEL from a human study was employed.

     1 L/day         =      assumed water consumption of a 10-kg child.

     The DWEL of 0.7 mg/L (700 /ig/L) derived for the Lifetime HA (see next section) is
used for the Longer-term HA for the 70-kg adult.

Lifetime Health Advisory

     The Lifetime HA represents that portion of an individual's total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure. The Lifetime HA is derived in a three-step process. Step 1
determines the Reference Dose (RfD), formerly called the Acceptable Daily Intake  (ADI).
The RfD is an estimate of a daily exposure level to the human population that is likely to be
without appreciable risk of deleterious effects over a lifetime, and is derived from the
NOAEL (or LOAEL), identified from a chronic (or subchronic) study, divided by an
uncertainty factor(s). From the RfD, a Drinking Water Equivalent Level (DWEL) can be
determined (Step 2).  A DWEL is a medium-specific (i.e.,  drinking water) lifetime exposure
level, assuming 100% exposure from that medium, at which adverse, noncarcinogenic health
effects would not be expected to occur.  The DWEL is derived from the multiplication of the
RfD by the assumed body weight of an adult and divided by the assumed daily water

                                          12

-------
 Malathion                      •                                            April, 199i.
consumption of an adult. The Lifetime HA is determined in Step 3 by factoring in other
sources of exposure, the relative source contribution (RSC). The RSC from drinking water is
based on actual exposure data or, if data are not available, a value of 20% is assumed. If the
contaminant is classified as a known, probable or possible human carcinogen, according to the
Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986), then caution must
be exercised in making a decision on how to deal with possible lifetime exposure to this
substance.  For human (A) or probable human (B) carcinogens, a Lifetime HA is  not
recommended. For possible human carcinogens (C), an additional 10-fold safety factor is
used in the calculation of the Lifetime HA The risk manager must balance this assessment
of carcinogenic potential and the quality of the data against the likelihood of occurrence and
significance of health effects related to noncarcinogenic end points of toxicity.  To assist the
risk manager in this process, drinking water concentrations associated  with estimated excess
cancer risks over the range of 1 in 10,000 to  1 in 1,000,000 for the 70-kg adult drinking 2 L of
water per day are provided in the Evaluation of Carcinogenic Potential section.

     The study by Moeller and Rider (1962)  has been selected as the basis for  the Lifetime
HA. Human males were treated orally with malathion in capsules at 0.11, 0.23 and
0.34 mg/kg/day for 32 to 56 days.  There were no clinical signs of poisoning, and at the two
lowest  doses, 0.11 and 0.23 mg/kg/day, there was no inhibition of plasma or erythrocyte
cholinesterase. Administration of 034 mg/kg/day resulted in a maximum inhibition of plasma
and erythrocyte cholinesterase of 30% of pretest values. Thus, the NOAEL was
0.23  mg/kg/day and the LOAEL was 034 mg/kg/day, under the conditions of this study.

     Chronic animal studies were  evaluated for calculating the Lifetime HA, but were not
judged to be suitable. Golz and Shaffer (1955) identified a LOAEL of 6 mg/kg/day for
cholinesterase inhibition in rats administered malathion for 2 years. The methods and results
were not adequately  reported, and a NOAEL was  not identified. Tegeris Laboratories (1987)
identified a LOAEL of 62J mg/kg/day for cholinesterase inhibition and systemic effects in
dogs administered  malathion  for 1 year.  A NOAEL was not identified.  In contrast to the
above animal studies, the Moeller and Rider (1962) 56-day study in humans identified a
NOAEL and LOAEL of 023 and 034 mg/kg/day,  respectively.  In addition, because it was
performed in humans, the Moeller and Rider (1962) study provides a more sensitive indicator
of the potential for cholinesterase inhibition in humans following oral exposure to  malathion.

     Using this study, the Lifetime HA is calculated as follows:

Step 1:  Determination of the Reference Dose (RfD)

         RfD , (0.23 mg/kg/day)  = Q Q23 mg/kg/day   (rounded to 0.02 rag/kg/day)


where:
                                          13

-------
 Malathion                                                                 April, 199JL
          0.23 mg/kg/day      =     NOAEL, based on the absence of cholinesterase
                                   depression in humans exposed orally to raalathion for
                                   32 to 56 days.

                     10      =     uncertainty factor, this uncertainty factor was chosen in
                                   accordance with EPA or NAS/OW guidelines in which
                                   a NOAEL from a human study was employed.

Step 2:  Determination of the Drinking Water Equivalent Level (DWEL)

          DWEL m  (0.02 mg/kg/day) (70 kg)  = Q ?  mg/L  (rounded ^ ^ ^
                           (2 L/day)

where:

        0.023 mg/kg/day      =     RfD (before  rounding).

                  70 kg      =     assumed body weight of an adult.

                2 L/day      =     assumed daily water consumption of an adult.


Step 3:  Determination of the Lifetime HA

           Lifetime HA = (0.7 mg/L) (20%) = 0.14  rag/L (rounded to 100 Mg/L)

where:

         0.81 rag/kg/day     =     DWEL.

                 (20%)     = .     assumed relative source contribution for a drinking
                                   water disinfectant.


Evaluation of Carcinogenic Potential

     •  Oncogenicity studies of raalathion were negative in rats (NCI, 1978, 1979a; U.S.
        EPA, 1988).  A study in mice was judged negative by NCI (NCI, 1978; Huff et al.,
        1985), but the data were considered equivocal by U.S. EPA (1988a).  Another study
        was required

     •  An oncogenicity study of malaoxon was equivocal in rats (NCI, 1979b; Huff et al.,
        1985; U.S. EPA, 1988a) and negative in mice (NCI,  1979b; Huff et al., 1985; U.S.
        EPA, 1988a).
                                         14

-------
 Malathion                                                                  April, 199Z,
     •   The Office of Pesticide Programs reevaluated the available oncogenicity studies in
         rats and mice (U.S. EPA, 1990) and concluded that the oncogenic potential of this
         chemical remains undetermined from the available data (Group D).

     •   IARC (1983) has classified malathion and raalaoxon in Group 3: chemicals that
         cannot be classified as to their carcinogenicity for humans.

     •   The carcinogenic potential of malathion has been evaluated by the U.S. EPA.
         Applying the criteria described in EPA's guidelines for assessment of carcinogenic
         risk (U.S. EPA, 1986), malathion may be classified in Group D: not classifiable.
         This category is for agents with inadequate animal or no evidence of carcinogenicity.
VI. OTHER CRITERIA. GUIDANCE AND STANDARDS

     •  The Safe Drinking Water Committee of the National Academy of Sciences
        calculated a Suggested-No-Adverse-Effect Level (SNARL) in drinking water for
        malathion of 0.16 rag/L, based on the Moeller and Rider (1962) study (NAS, 1977).

     •  The current exposure criteria for airborne malathion in the United States are
        15 mg/mj, 10-hour TWA (NIOSH, 1985) and 10 mg/m3 (ACGffl, 1986).

     •  Tolerances for malathion  in a variety of raw agricultural commodities have been
        established. For most fruits and vegetables, the level is 8 ppra (U.S. EPA, 1988a).
VII.  ANALYTICAL METHODS

     •  Malathion is one of the class of phosphorodithioate. pesticides. The relative ease
        with which this pesticide can be monitored by element-specific detectors has led to
        its inclusion in many pesticide-monitoring studies.  Malathion can be analyzed by
        EPA Methods, 622 (U.S. EPA, 1982) and 507 (U.S. EPA, 1988b). In both methods,
        a solvent is exchanged for hexane or methyl tertiary butyl ether (MTBE).  Analysis
        is by an element-specific thermionic nitrogen-phosphorus detector, which allows
        even relatively "dirty" samples to be analyzed with no cleanup. The estimated
        detection limit for malathion is Q25 jig/L.
VIII.  TREATMENT TECHNOLOGIES

     •  Available data indicate that ozone oxidation and activated carbon adsorption can
        significantly reduce malathion levels in drinking water.

     •  Sharma et al. (1987) used activated charcoal to remove malathion from saline water
        with an initial concentration of 1.6 g of malathion/L. The activated charcoal
        adsorption  capacity for malathion was found to be 117 mg malathion/g charcoal.
                                         15

-------
Malathion                                                                   April, 199.2,
     •  Zeff and Harris (1984) evaluated the applicability of the UV-ozonation process for
        water reuse from industrial wastewater at a model P-602 (Ultrox*) pilot plant in
        Lathrop, California.  Preliminary bench tests were conducted to provide operating
        guidelines for the initial pilot plant tests.  Malathion was present in the influent at a
        concentration of 0.5 mg/L.  The bench results indicated that the pilot plant should
        operate with a retention time of 2 to 5 hours, and an ozone dosage of 2 to 5 g/L at
        an ozone concentration of 2 to 3 weight percent.  Under these conditions, the
        concentration of raalathion in the effluent reached nondetectable levels in 2 hours.
        Using the operating conditions derived from the bench-scale work, the pilot plant
        was put into operation.  The pilot plant removed 93% of the influent 24 mg/L
        raalathion when operated at a 12 L/min wastewater flow and an ozone dosage of
        3.1 g/L. Lime pretreatment increased the removal to 995%.

     •  Data were not found for the removal of raalathion from drinking water by aeration.
        However, because of its low vapor pressure, malathion probably is not amenable to
        removal by aeration.
                                          16

-------
Maiathion                                                                  April, 1995.
IX.  REFERENCES

ACGIH.  1986. American Conference of Government Industrial Hygienists.  Documentation
     of the threshold limit values and biological exposure indices, 5th ed. Cincinnati, OH:
     ACGIH.

Baker, EX., Jr., M. Warren, M. Zack, R.D. Dobbin, J.W. Miles, S. Miller, L. Alderman and
     W.R. Teeters.  1978.  Epidemic rnalathion poisoning in Pakistan malaria workers.
     Lancet 1:31-34.

Berteau,  P.E. and W.A. Deen. 1978. A comparison of oral and inhalation toricities of four
     insecticides to mice and rats. Bull. Environ. Contam. Toricol. 19:13-120.

Brown, S.K., R.G. Ames, and D.C. Mengle. 1989.  Occupational illnesses from
     cholinesterase-inhibiting  pesticides among agricultural applicators in California,
     1982-1985. Arch. Environ. Health. 44:34-39.

Carey, A.E. and F.W. Kutz.  1985. Trends in ambient concentrations of agrochemicals  in
     humans and the environment of the United States.  Environ. Monit. Assess.  5:155-163.

Carey, A.E., J.A. Gowen, H. Tai, W.G. Mitchell and G.B. Wiersma. 1979.  Pesticide residue
     levels in soils and crops from 37 states, 1972.  National Soils Monitoring Program (IV).
     Pestic. Monit. J. 12:209-229.

Carey, A.E., J.A. Gowen, H. Tai, W.G. Mitchell and G.B. Wiersma. 1978.  Pesticide residue
     levels in soils and crops,  1971. National Soils Monitoring Program (III).  Pestic. Monit.
     J. 12:117-136.

Chhabra, M.L., G.C. Sepha, S.R. Jain, R.R. Bhagwat and J.D. Khandekar.  1970.  E.C.G. and
     necropsy changes in organophosphorus compound (rnalathion) poisoning. Indian J.
     Med. Sci.  24:424-429.

Degraeve, N. and J. Moutschen.  1984.  Genetic and cytogenic effects induced in the mouse
     by an organophosphorus insecticide:  rnalathion.  Environ. Res. 34:170-174.

Dulout, F.N, O.A. Olivero, H. von Guradze and M.C. Pastori.   1982. Cytogenic effect of
     rnalathion assessed by the micronucleus test.  Mut. Res. 105:413-416.

Dulout, F.N., M.C. Pastori and O.A. Olivero.  1983. Malathion-induced chromosomal
     aberrations in bone-marrow  cells of mice:  dose-response relationships.  Mut. Res.
     122:163-167.

Dzwonkowska, A. and H. Huebner.  1986.  Induction of chromosomal aberrations in the
     Syrian hamster by insecticides tested in vivo.  Arch. Toxicol. 58:152-156.
                                          17

-------
 Malathion                                                                  April, 1992,
 Eichelberger, J.W. and JJ. Lichtenberg.  1971.  Persistence of pesticides in river water.
     Environ. Sci. Technol. 5:541-544.

 Feldraan, R J. and H.I. Maibach.  1974.  Percutaneous penetration of some pesticides and
     herbicides in man.  Toxicol. Appl. Pharmacol. 28:126-132.

 Feldraan, RJ. and H.I. Maibach.  1970.  Absorption of some organic compounds through the
     skin in man. J. Invest. Dermatol. 54:399-404.

 Frawley, J.P., H.N. Fuyat, E.G. Hagan, J.R. Blake and O.G. Fitzhugh.  1957.  Marked
     potentiation in mammalian toxicity from simultaneous administration of two
     anticholinesterase compounds.  J. Pharmacol. Exp. Ther. 121:96-106.

 Gaines, T.B.  1969.  Acute toxicity of pesticides.  Toxicol. Appl. Pharmacol.  14:515-534.

 Golz, H.H. and C.B. Shaffer.  1955.  Malathion:  summary of pharmacology and toxicology.
     Central Medical Department, American  Cyanamid Co.

 Gupta, S.K., M.K. Pandga, J.P. Jani  and S.K  Kashyup.  1980. Health risks in ultra-low
     volume (ULV) aerial spray of malathion for mosquito control.  J. Environ. Sci. Health.
     B15:287-294.

 Hanna,  WJ. and E. Choo-Kang.  1983.  Malathion poisoning: A report of 2 cases.  West
     Indian Med. J. 32:109-111.

 Hayes, W.J., Jr.  1971.  Studies on exposure during the use of anti-cholinesterase pesticides.
     Bull. World Health Organ. 44:277-288.

Hazelton, L.W. and E.G. Holland. 1953. Toxicity of malathion:  Summary of mammalian
     investigations.  AMA Arch. Indus. Hyg. Occup. Med. 8:399-405.

Huang,  C.C.  1973.  Effect on growth but not on chromosomes of the mammalian cells after
     treatment with three organophosphorus insecticides.  Proc. Soc. Exp.  Biol. Med.
     142:36-40.

Huff, J.E., R. Bates, S.L. Eustis, J.K. Haseman and E.E. McConnell. 1985. Malathion and
     malaoxon:   histopathology reexamination of the National Cancer Institute's
     carcinogenesis studies. Environ. Res. 37:154-173.

IARC.  1983. International Agency for Research on Cancer.  IARC monographs on the
     evaluation of carcinogenic risk of chemicals to humans. Vol. 30. Lyon, France:  IARC.
     pp. 103-129.
                                          18

-------
Malathion                                                                   April,  199JL
Jorgenson, TA^ CJ. Rushbrook and G.W. Newell.  1976. In vivo mutagenesis investigations
     of ten commercial pesticides.  Toxicol. Appl. Pharmacol. 37:109 (abstract).

Kalow, W. and A. Marton.  1961. Second-generation toxicity of raalathion in rats.  Nature
     192(4801):464-465.

Khera, K.S., C. Whalen and G. Trivett.  1978. Teratogenicity studies on linuron, malathion
     and methoxychlor in rats. Toxicol. Appl. Pharmacol. 45:435-444.

Kimbrough, R.D. and T.B.  Gaines.  1968. Effect of organic phosphorus compounds and
     alkylating agents on the rat  fetus. Arch. Environ. Health.  16:805-808.

LaFleur, K. 1979. Sorption of pesticides by  model soils and agronomic soils: Rates and
     equilibria. Soil Science 127:94-101.

Lehman, AJ.  1959.  Appraisal of the safety  of chemicals in foods, drugs, and cosmetics.
     Association of Food and Drug Officials  of the United States, Quarterly Bulletin.

Meister, R.T., ed.  1987.  Farm Chemicals Handbook. Willoughby, OH:  Meister Publishing
     Co.

Mendoza, C.E. and J.B. Shields.  1977. Effects on esterases and comparison of IM and LDSO
     values of malathion in suckling rats. Bull. Environ. Contain. Toxicol. 17:9-15.

Miller, S. and MA. Shah.   1982.  Cholinesterase activities of workers exposed to
     organophosphorus insecticides in Pakistan and Haiti and an evaluation of the tintometric
     method. J. Environ. Sci. Health.  B17(2):125-142.

Moeller, H.C. and J.A. Rider.  1962. Plasma and red blood cell cholinesterase activity as
     indications of the threshold  of incipient  toxicity of ethyl-p-nitrophenyl
     thionobenzenephosphonate  (EPN) and malathion in human beings.  Toxicol. Appl.
     Pharmacol.  4:123-130.

Morgade, C. and A. Barquet.  1982.  Body distribution of malathion and its metabolites in a
     fatal poisoning by ingestion.  J. Toxicol. Environ. Health 10:321-325.

NAS. 1977. National Academy of Sciences.  Drinking Water and Health. Vol. 1.
     Washington, DO National  Academy Press,  p. 622.

NCI. 1979a.  National Cancer Institute. Bioassay of malathion for possible carcinogenicity.
     U.S. Department of Health, Education & Welfare.  (Tech. Rep. Ser. No. 192; DHEW
     Publ. No (Nffl) 79-1748.) Washington, DC.
                                           19

-------
 Malathion                      •                                            April,
 NCI. 19795. National Cancer Institute. Bioassay of raalathion for possible carcinogenicity.
     U.S. Department of Health, Education & Welfare. (Tech. Rep. Ser. No. 135; DHEW
     Publ. No. (NIH) 79-1390.) Washington, DC

 NCI. 1978. National Cancer Institute. Bioassay of raalathion for possible carcinogenicity.
     U.S. Department of Health, Education & Welfare. (Tech. Rep. Ser. No. 24; DHEW
     Publ. No. (NIH) 78-824.) Washington, DC.

 NIOSH.  1985.  National Institute for Occupational Safety and Health.  NIOSH Pocket
     Guide to Chemical Hazards.  U.S. Department of Health and Human Services.  Public
     Health Service.  Printing (DHEW [NIOSH] Pub. No. 78-120).  Washington, DC

 Pednekar, M.D., S.R. Gandhi and MS. Netrawali. 1987. Evaluation of mutagenic activities
     of endosulfan, phosalone, malathion and permethrin, before and after metabolic
     activation,  in the Ames Salmonella test. Bull. Environ. Contam. Toxicol. 38:925-933.

 Sharma, S.R., H.S. Rathore and S.R. Ahmed.  1987.  Studies on removal of raalathion from
     water by means of activated charcoal.  Ecotoxicology and Environmental Safety 14:22-29.

Shiau, S.Y., R.A. Huff, B.C. Wells and I.C. Felkner.  1980.  Mutagenicity and DNA damaging
     activity for several pesticides tested with Bacillus subtilis mutants.  Mut. Res. 71:169-179.

Siddiqui, M.KJ., T.D. Seth and MC. Saxena.  1980.  Acetyicholinesterase activity in red
     blood cells of healthy, diseased and exposed persons.  Indian J. Med. Sci.  34:289-292.

Su, M.Q., F.K. Kinoshita, J.P. Frawley and KJP.  DuBois. 1971. Comparative inhibition of
     aliesterases and cholinesterase in  rats fed eighteen organophosphorus insecticides.
     Toxicol. Appl. Pharmacol. 20:241-249.

Tegeris Laboratories, Inc.1  1987.  One-year oral toxicity study in purebred beagle dogs with
     AC6.601. Laboratory study number 85010.  MRTD 401885-01.  February 10,1987.

U.S. EPA. 1990.  U.S. Environmental Protection Agency.  1990. Carcinogenicity peer review
     of malathion.  Office of Pesticide  Programs. April 12.

U.S. EPA. 1988a.  U.S. Environmental Protection Agency.  Guidance for the reregistration
     of pesticide products containing malathion as the active ingredient.   Washington, DC:
     Office of Pesticides and Toxic Substances.  January.

U.S. EPA. 1988b.  U.S. Environmental Protection Agency.  Method 507, the determination
     of nitrogen and phosphorus containing pesticides in water by gas chromatography with a
 'Confidential Business Information submitted to the Office of Pesticide Programs.


                                          20

-------
Malathion                                                                  April, 1992-
     nitrogen-phosphorus detector.  Cincinnati, OH:  Environmental Monitoring and Support
     Laboratory.  December.

U.S. EPA.  1986.  U.S. Environmental Protection Agency.  Guidelines for carcinogen risk
     assessment.  Fed. Reg. 51:33992-34003.

U.S. EPA.  1982.  U.S. Environmental Protection Agency.  Method 622, the determination of
     organophosphorus pesticides in industrial and municipal wastewater. Cincinnati, OH:
     Environmental Monitoring and Support Laboratory.

Velazquez,  A., A. Creus, N. Xamena and R. Marcos. 1987. Lack of rautagenicity of the
     organophosphorus insecticide malathion in Drosophila melanogaster.  Environ. Mutagen.
     9:343-348.

Warren, M., H.C. Spencer, F.C. Churchill, VJ. Francois, R. Hippolyte and MA. Staiger.
     1985.  Assessment of exposure to organophosphate insecticides during spraying in Haiti:
     monitoring of urinary metabolites and blood cholinesterase levels. Bull. World Health
     Org. 63:353-360.

Wester, R.C., H.I. Mailbach, DA.W. Bucks and R.H. Guy.  1983.  Malathion percutaneous
     absorption after repeated administration to man. Toricol. Appl. Pharmacol.  68:116-119.

Windholz, M., S. Budavari, R.F. Blumetti and E.S. Otterbein, eds.  1983. The Merck
     index—An encyclopedia of chemicals, drugs, and biologicals, 10th ed. Rahway, NJ:
     Merck and Co., Inc.  p. 813.

Wiszkowska, H., I. Kulamowicz, A.  Malinowska and Z. Walter. 1986.  The effect of
     malathion on RNA polymerase activity of cell nuclei and transcription products in
     lymphocyte culture.  Environ. Res. 41:372-377.

WHO. 1982. World Health Organization.  Recommended health-based limits in
     occupational  exposure to pesticides:  Malathion. WHO Technical Report Series No.
     677. pp. 13-37.

Worthing, CR. and S.B. Walker, eds. 1987.  The  pesticide manual — A world compendium,
     8th ed. The  British Crop Protection Council.  United Kingdom:  Thornton Health.

Zeff, J.D. and JA. Harris. 1984. Chemistry and application of ozone and ultraviolet light for
     water reuse-pilot plant demonstration. Proceedings, Industrial Waste Conference.
     38:105-116.
                                          21

-------
EPA  0553
RX000027511
                                                                               April,
                                     p-NITROPHENOL

                               Drinking Water Health Advisory
                                       Office of Water
                             U.S. Environmental Protection Agency
   I.     INTRODUCTION
         The Health Advisory (HA) Program, sponsored by the Office of Water (OW), provides
   information on the health effects, analytical methodology and treatment technology that
   would be useful in dealing with the contamination of drinking water.  Health Advisories
   describe nonregulatory concentrations of drinking water contaminants at which adverse health
   effects would not be anticipated to occur over specific exposure durations.  Health Advisories
   contain a margin of safety to protect sensitive members of the population.

         Health Advisories serve as informal technical guidance to assist Federal,  State and local
   officials responsible  for protecting public health when emergency spills or contamination
   situations occur.  They are not to be construed as legally enforceable Federal standards.  The
   HAs are subject to change as new information becomes available.

         Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
   years, or 10% of an individual's lifetime) and Lifetime exposures based on data describing
   noncarcinogenic end points of toxicity. For those substances that are known or probable
   human carcinogens,  according to the Agency  classification scheme (Group A or B), Lifetime
   HAs are not recommended. The chemical concentration values for Group  A or B
   carcinogens are correlated with carcinogenic risk estimates by employing a  cancer potency
   (unit risk) value together with assumptions for lifetime exposure and the consumption of
   drinking water. The cancer unit risk is usually derived from the linear multistage model with
   95% upper confidence limits.  This provides a low-dose estimate of cancer risk to humans
   that is considered unlikely to pose a carcinogenic risk in excess of the stated values.  Excess
   cancer risk estimates may also be calculated using the one-hit,  Weibull, logit or probit models.
   There is no current  understanding of the biological mechanisms involved in cancer to suggest
   that any one of these models is able to predict risk more accurately than another.  Because
   each model is based on differing assumptions, the estimates that are derived can differ by
   several orders of magnitude.

-------
P-Nitrophenol                                                               April, 1992.



II.    GENERAL INFORMATION AND PROPERTIES

CAS No.  100-02-7

Structural Formula

                                 OH
Synonyms
      •  4-Nitrophenol; 4-hydroxynitrobenzene;  phenol,4-nitro;
         p-hydroxynitrobenzene; para-nitrophenol (Windholz et al., 1989).
Uses
      •  /7-Nitrophenol is used in the manufacture of the pesticides, ethyl and methyl
         parathions and n-acetyl-p-arainophenol.  Other uses include production of dyes,
         fungicides and leather preservatives.  It is also used as an analytical indicator
         (Windholz et al., 1989; Hawley, 1987; NIOSH, 1981).

Properties  (NIOSH, 1981)

      Chemical Formula                   CjHsNOj
      Molecular Weight                    139.11
      Physical State                       colorless to slightly
                                          yellow crystals
      Boiling Point                        279ฐC (decomposes)
      Melting Point                       113-114ฐC
      Density (20ฐC)                      1.479 g/cmj
      Vapor Pressure (146ฐC)              2.2 mm Hg
      Specific Gravity (20ฐC)               1.479
      Water Solubility (25ฐC)              16 g/L
      Log Octanol/Water Partition
       Coefficient                        1.91
      Taste Threshold                     43.4  rag/L
      Odor Threshold (water)              2.5 rag/L
      Conversion  Factor                   1 rng/rn3 = 0.18 ppm
       (in air at 25ฐC)                      (rounded from
                                         0.176 ppm)

-------
P-Nitrophenol                                                                April,  1992.
Occurrence

      •  Alber et al. (1989) detected /7-nitrophenol in rain and snow at concentrations
         ranging from 0.49 jig/L to 17.1 /xg/L. p-Nitrophenol has also been detected in the
         secondary effluent from publicly owned treatment works and in industrial effluents
         in surface waters (Ellis et al., 1982).  Alber et al. (1989) attributed the high
         concentrations of /7-nitrophenol and other phenols in rain and snow to
         photochemical reactions of aromatic hydrocarbons, NOX and OH' radicals.

      •  Lokke (1985) attributed the presence ofp-nitrophenol in Danish soils to leaching
         from chemical waste disposal sites.

Environmental Fate

      •  p-Nitrophenol is soluble in water and is not removed by aeration (Batchelder,
         1975).  The bioconcentration of p-nitrophenol in fish and other aquatic life is
         considered insignificant based on its  low water solubility and low octanol/water
         partition coefficient (< 100) (Leo et al., 1971).

      •  The biodegradation ofp-nitrophenol in lake and river waters is faster at higher p-
         nitrophenol concentrations than  at lower concentrations (Zaidi et al., 1989; Siragusa
         and DeLaune, 1986).  Zaidi et al. (1989) conducted a study to assess the factors
         limiting biodegradation of chemicals present at low concentrations. p-Nitrophenol
         at 50, 75 and 100 /ig/L was extensively decomposed in lake water inoculated with
         Corynebacterium sp. but not in the  uninoculated water.  However, the presence of
         this bacterium resulted in the mineralization of only 35% of p-nitrophenol present
         at a concentration of 26 /tg/L.  Addition of 250 mg of cyclohexiraide, an inhibitor of
         microbial growth, increased biodegradation to 75% within 70 hours.  This effect of
         cyclohexirnide was absent  at higher levels of p-nitrophenol and in uninoculated
         waters.  Contaminants may be of limited impact if the compound is found at low
         concentrations or if bacterial inhibitors are present in the polluted environment. In
         other studies (Spain et al., 1980; Spain  and Van Veld, 1983), pre-exposure of
         coastal sediment to various compounds affected their biodegradation rates  upon re-
         exposure of the sediment to the same compounds.

      •  ^-Nitrophenol was totally decomposed by soil microflora within 16 days at 25 ฐC
         (Alexander and Lustigman, 1966).  The half-life for degradation of/;-nitrophenol
         was 10 times higher under anaerobic conditions than under aerobic conditions.
         This was due to  the absence or very small  number of microorganisms in the subsoil
         and to the higher acidity of the subsoil  (pH 4.7) compared with the topsoil (pH 5.4)
         (Lokke, 1985).  The half-life for  2 rag/kg of ^-nitrophenol was 0.7 to  1.2 days in the
         topsoil under aerobic  conditions  and 14 days under anaerobic conditions.  In the
         subsoil, the  half-life of p-nitrophenol was 40 days under aerobic conditions, but
         under anaerobic conditions the degradation of ^-nitrophenol was minimal.

-------
P-Nitrophenol                                                                April, 1992.
      •  The sorption and desorption of p-nitrophenol in sediment from a Louisiana gulf
         coast estuary were studied under aerobic and anaerobic conditions (Siragusa and
         DeLaune, 1986).  The results showed that decomposition of p-nitrophenol,
         measured as release of radiolabeled carbon dioxide, was several orders of
         magnitude faster under aerobic than under anaerobic conditions (1.04 x 10"J versus
         2.95 x 10's /xg/day/g dry sediment, respectively).  The results suggested thatp-
         nitrophenol is rapidly desorbed from the water column to the bottom sediment,
         where it may persist for years owing to the slower rate  of biodegradation under
         anaerobic conditions in estuarine sediment than occurs under aerobic conditions.
         Moreover, the equilibration rate of p-nitrophenol sorption-desorption did not affect
         the re-release of p-nitrophenol.

      •  The results from a biodegradability laboratory test carried out by Tabak et al.
         (1981) on 114 organic priority  pollutants showed rapid acclimation and complete
         biodegradation of p-nitrophenol. The static culture flask-screening method utilizing
         biochemical oxygen demand (BOD) dilution water was  used.  Five dr 10 rag/L of p-
         nitrophenol and 5 mg/L of yeast  extract  as the synthetic medium were incubated for
         7 days under static conditions at 25ฐC in the dark.  Three weekly subcultures were
         done with final incorporation of wastewater as the microbial inoculum.

      •  The biodegradation of p-nitrophenol was compared in three laboratory test systems
         and in the field utilizing freshwater from a pond (Spain et al., 1984).  The water
         samples were treated simultaneously with 0.2 mg/L of radiolabeled p-nitrophenol
         and incubated at 19ฐC. The results showed that p-nitrophenol was biodegraded
         more slowly by microorganisms acclimated in  the laboratory than by those found in
         the pond. The acclimation periods in the laboratory systems that contained
         sediment were similar to  those in the pond, indicating that the results from
         laboratory test systems can be  compared with those from the field.
III. PHARMACOKINETICS

Absorption

      •  Limited information  is available on the gastrointestinal absorption of p-nitrophenol
         following oral administration.  Robinson et al. (1951) administered 200 mg/kgp-
         nitrophenol in water  to rabbits via a stomach tube. It was concluded that p-
         nitrophenol was rapidly absorbed  from the gastrointestinal tract following oral
         administration in rabbits, since it was completely excreted within 1 day.

      •  Based on the results  of a study using 60- to 80-day-old hairless  mice  (SKH-hr-1
         strain-), Jetzer et al. (1988) concluded that p-nitrophenol is absorbed  through the
         skin, and that the rate  of permeation is temperature-dependent. The results

-------
P-Nitrophenol                                                                April, 1992.
         showed that permeation rates decreased with decreasing temperatures, but were
         higher in non-isothermal studies than in low-temperature isothermal studies.
Distribution
      •  No information was available on the tissue distribution of p-nitrophenol following
         oral administration.  Groups of 200- to 350-gram male Wistar rats were
         intravenously (1.6, 4.0 or 8.0 rag/kg) and intra-arterially (4.0 and 8.0 mg/kg) injected
         with p-nitrophenol (Machida et al.,  1982).  p-nitrophenol rapidly equilibrated
         between red blood cells and plasma. This equilibrium was so rapid that
         p-nitrophenol-glucuronide and -sulfate could be identified 1 minute after dosing.
         The ratio of p-nitrophenol concentrations in red blood cells and plasma was the
         same for all dosagess at the 2-, 4- and 8-minute sample collection intervals.
Metabolism
      •  p-Nitrophenol undergoes both phase I (reduction) and phase II (glucuronidation
         and sulfation) metabolic transformation in the liver. In addition to sulfation and
         glucuronidation, p-nitrophenol undergoes reduction to panz-amino-phenol
         (Robinson et al., 1951). Rabbits (age and sex not reported) were orally
         administered 200 mg/kg of p-nitrophenol in a water suspension, and 24-hour urine
         samples were collected. Excretion was complete in 1 day.  Most of the dose (82 to
         92%) was excreted as unchanged p-nitrophenol, and 11 to 19% was excreted as the
         reduced ami no compound. Of the parent chemical, 59 to 62% was conjugated and
         excreted as glucuronide, and 13 to 21% was excreted as organic sulfates.

      •  Meerman et al. (1987) reported that there was no sex difference in conjugation of
         p-nitrophenol in the rat.  Male and female Wistar rats (60 days old) were injected
         via the lateral tail vein with 60 /imol/kg of p-nitrophenol. Both p-nitrophenol-
         sulfate and -glucuronide were detected in the 24-hour urine samples collected.
         Both female and male rats excreted  the same amount of p-nitrophenol-sulfate and
         -glucuronide (35 and 37% of the administered dose, respectively).

      •  p-Nitrophenol also undergoes extrahepatic metabolism.  Machida et al. (1982)
         found that formation of p-nitrophenol-glucuronide occurred in the lungs and
         kidneys of 200- to 350-gram male Wistar rats administered p-nitrophenol. The  rats
         were administered the p-nitrophenol via three routes as follows: 1.6, 4 and 8 mg/kg
         in physiological saline solution were given intravenously through the left femoral
         vein, 4 and 8 rag/kg were injected intra-arterially and 1.6 and 4 mg/kg were injected
         into the hepatic portal vein.  Administration of 4mg/kg of p-nitrophenol via the
         femoral vein and via the portal vein produced different p-nitrophenol plasma
         concentrations. Similarly, the area under the curve produced by intravenous
         femoral vein administration (AUV^) and the area under the curve produced by
         portal vein injection (AUV ) at 1.6  and 4 rag/kg were significantly (p < 0.05)

-------
P-Nitrophenol                                                                April, 199Z
         different.  No significant difference was seen between AUC^, values and the areas
         under the curve produced by intra-arterial dosing. Nevertheless, pulmonary
         metabolism ofp-nitrophenol was not ruled out.  The hepatic metabolic clearance
         rate for the femoral dose was 17 to 21 mL/min, which corresponded to the hepatic
         blood flow rate (10-20 mL/min). This suggested that both sulfation and
         glucuronidation ofp-nitrophenol are limited by hepatic blood flow.

      •  Livers isolated from 20- to 30-gram male Swiss mice were perfused in situ with 1 to
         100 pM p-nitrophenol (Sultatos and Minor, 1985). Preliminary experiments showed
         that less than 1% of the p-nitrophenol appeared in the bile as unchanged
         compound or metabolites, indicating that biliary excretion is not a major pathway.
         In animals perfused with up to 4 yiM. p-nitrophenol, the sulfate conjugate was the
         only metabolite. However, in animals perfused with p-netrophenal at
         concentrations of 15 /iM or higher, there was more unchanged p-nitrophenol and
         p-nitrophenol-glucuronide than p-nitrophenol-sulfate in the bile.  The steady-state
         concentrations of p-nitrophenol-glucuronide and sulfate were reached at 25 and
         15 /xM,  respectively.  This suggested that sulfation is kinetically more favorable than
         glucuronidation at the lower concentrations, while the reverse is true at higher
         concentrations.

      •  In perfused livers from female Sprague-Dawley rats (140 to 150 g), 0.06 mM
         p-nitrophenol was rapidly hydroxylated to 4-nitrocatechol; the latter competed with
         p-nitrophenol for conjugation with glucuronic acid and sulfate (Reinke and Moyer,
         1985).  The oxidation ofp-nitrophenol occurred via the reduction of nicotinamide-
         adenine dinucleotide phosphate (NADP) at pH 7.1-7.4.
Excretion
         The in vivo Specific Activity Difference Ratio (SADR) technique was used to
         measure the excretory metabolites ofp-nitrophenol in rat kidneys (Tremaine et al.,
         1984). Male Sprague-Dawley rats (335 to 400 g) were given continuous infusions of
         a solution containing 2 /imol l4C-p-nitrophenol/min/kg via the jugular vein; the
         infusions were given at a constant rate (between 0.02 and 0.08 mL/min).  Either p-
         nitrophenol-glucuronide or p-nitrophenol-sulfate was infused at 0.3 /xmol/min/kg.
         Urine and arterial blood samples were collected.  The mean p-nitrophenol plasma
         concentration was 31 jiM.  A steady state was reached, indicating first-order
         elimination kinetics for each of the infused compounds. During steady-state
         conditions, 75% of the infused radiolabel was recovered in urine.  In the urine, 99%
         of the radiolabel was found in p-nitrophenol-sulfate and p-nitrophenol-glucuronide.
         The rat kidney  formed both of these conjugates at equal rates. No unchanged
         14C-p-nitrophenol was identified in the urine. The renal clearance for p-nitrophenol
         in the rat  was 6.4 mL plasma/min/kg/kidney, which is 1.6 times the glomerular
         filtration rate.   It was reported that the kidney accounted for a minimum  of 20%  of
         the endogenously formed conjugates ofp-nitrophenol.  The bile does not  appear to

-------
P-Nitrophenol                                                              April, 1992.
         account for the significant amounts of p-nitrophenol excretion in the rat (0.9% of
         the infused radiolabel was recovered in the bile).

      •  Robinson et al., 1951 administered 200 mg/kg ofp-nitrophenol to rabbits via a
         stomach tube.  The compound was exclusively and completely excreted in the 24-
         hour urine sample. Most of the dose (82 to 92%) was excreted as unchanged p-
         nitrophenol, and 11 to 19% was excreted as the reduced amino compound.  Of the
         unchanged p-nitrophenol, 59 to 62%  was conjugated and excreted as glucuronide,
         and 13 to 21% was excreted as  organic sulfates.

      •  Gorge et al., 1987 studied the excretion and metabolism of p-nitrophenol in two
         frog species. The frogs were injected with 3 to 5 mg/kg p-nitrophenol  (2 to 3
         /xCi/animal) into the dorsal lymphatic sac through the thigh muscle.  Within 24
         hours, both frogs excreted 90 to 95% of the administered dose,  p-nitrophenol was
         metabolized primarily by glucuronidation (12.5%) and sulfation (46%). Smaller
         amounts of /vnitrophenol were  reduced to p-nitrocatechol and then conjugated.
         The kinetics of excretion fit a two-compartment model. Unchanged p-nitrophenol
         accounted for 35% of the excreted dose.
IV.   HEALTH EFFECTS

Humans

  Short-term Exposure

      •  No information was found in the available literature regarding short-term exposure
         to p-nitrophenol in humans.

  Long-term Exposure

      •  Nitro-amino compounds are thought to have nephrotoxic effects. Yoshida et al.
         (1989) conducted a clinical cross-sectional study with 62 male workers (aged 20 to
         64 years) exposed to aromatic nitro-amino compounds, including ^-nitrophenol, in a
         chemical factory.  The urinary enzyme activities of workers were measured as an
         index of renal damage; 27 office workers  served as controls.  The duration of
         exposure ranged  from 3.6 to 18 years for  both exposed and control groups.
         Information on smoking and alcohol consumption was obtained through a
         questionnaire.  The environmental concentrations of the nitro-amino compounds
         were not measured directly, but the investigators identified a probable exposure
         level of 0.3 mg/mj. This value was based  on a study conducted by the authors in
         which workers  were exposed to 0.38 mg/m3 p-nitrochlorobenzene.  In these workers,
         the authors found 0.5 mg of urinary diazo metabolites per mg of creatinine. In the
         present study, the exposed workers also excreted diazo metabolites at 0.5 mg/mg

-------
P-Nitrophenol                                                               April, 1992,
         creatinine. In addition, for two decades the amounts handled by the workers were
         constant.  Therefore, these workers were probably exposed to aromatic nitro-amino
         compounds at concentrations greater than 03 mg/m3, since the aromatic nitro-
         amino compounds can  be absorbed through the skin and the  respiratory tract of
         workers.  In exposed workers who were 50 years of age or older, there was a
         significant increase (p < 0.05) in urinary n-acetyl-beta-
-------
P-Nitrophenol                                                                 April, 1992.
         transient, increases in the rate and depth of respiration when administered to the
         normally innervated common carotid artery. The effect was absent when the same
         dose was injected into a denervated common carotid artery. The investigators
         concluded that p-nitrophenol and other nitro compounds  act indirectly to stimulate
         the carotid chemoreceptors, resulting in changes in respiration. A similar
         experiment was conducted on the aortic chemoreceptors in the dog.  Thirty dogs, 6
         to 24 kg, were given 0.2-2.0 mL of a 0.5% solution of p-nitrophenol intra-arterially
         through the left common carotid artery; injection of 5 rag of .p-nitrophenol led to an
         immediate, but transient stimulation of respiration, producing  an indirect, specific
         effect on the aortic chemoreceptors (Shen, 1962).

      •  In a subchronic inhalation toxicity study, Sprague-Dawley rats  (15/sex, 6 to 7 weeks
         of age) were exposed to 0, 1, 5 and 30 rag/m3 of p-nitrophenol  dust for 6 hours a
         day, 5 days a week, for 4 weeks (Hazleton Laboratories America, Inc., 1983). A
         control group was exposed to filtered air. Exposure to/7-nitrophenol caused no
         deaths, and no consistent exposure-related effects were noted in hematology values,
         clinical chemistry values, gross examination and histopathology, body or organ
         weights, or ophthalmoscopic examination.

  Dermal/Ocular Effects

      •  /7-Nitrophenol is absorbed from the skin and causes strong skin irritation, but it
         does not sensitize skin.  Its dermal LDM in guinea pigs is  > 1 g/kg. Repeated 10-
         day skin applications to guinea pigs caused staining of skin within 24 hours and
         caused decreased weight gain within 10 days (Eastman Kodak  Company, 1980).

      •  Six 2.5- to 3.0-kg New Zealand White albino rabbits were tested for p-nitrophenol
         induced skin irritation. A mass of 0.5 +. 0.007 g of p-nitrophenol moistened with
         saline was applied to each of two  intact and two abraded sites of the rabbit  ear for
         24 hours.  By the first day of application, the skin stained yellow on all four treated
         sites of all six animals.  Scarring occurred in four rabbits by day 14.  In three of
         these four rabbits,  dark brown discoloration of the skin was seen prior to the
         appearance of scabs or scars (Branch,  1983b).

      •  />-Nitrophenol (70 rag in 0.1 raL) was instilled into the conjunctival sac of the right
         eye of each of six 2.2- to 3.0-kg New Zealand albino rabbits. The untreated eye of
         each rabbit served as the control (Branch, 1983c). Moderate-to-severe corneal
         cloudiness was seen through day 21  in five animals.  Periocular skin and/or fur of all
         five animals was stained yellow throughout the study.  Blistered conjunctivae were
         seen during the first 3 days of application. Scabbing and alopecia of the upper and
         lower eyelids were  seen in one female rabbit. Corneal neovascularization was found
         in five animals from day 7 to  day  21. Based on the nature of the corneal
         involvement, /7-nitrophenol was classified as corrosive to the eyes.

-------
P-Nitrophenol                                                                 April, 1992
  Long-term Exposure

      •  Adult Sprague-Dawley rats (20/sex/group) were given p-nitrophenol by gavage daily
         for 13 weeks (Hazleton Laboratories America, Inc., 1989).  The doses were selected
         based on an initial pilot study with six rats/sex/dose, in which 0, 1.0, 10, 50, or 100
         mg/kg/day /7-nitrophenol was  administered for 4 weeks.  No adverse effects were
         seen at any dose, so the No-Observed-Adverse-Effect  Level (NOAEL) was 100
         mg/kg/day. On the basis of the findings from this inital  study,  the doses chosen
         were 25, 70, and 140 rng/kg/day of p-nitrophenol.  The dosages were administered
         by gavage in water, the control group was administered  water only.  End points
         measured included:  clinical observations,  body weights,  food consumption,
         hematology and clinical chemistry values, ophthalmoscopic examination, organ
         weights and histopathology. Death occurred in all groups, but was significant
         (p < 0.05) only at the high dose in both sexes.  Deaths in the high-dose group
         might have been caused by the acute toxicity ofp-nitrophenol,  including
         raethernoglobinemia.  In mid-dose females and all high-dose rats, the liver, kidneys,
         and lungs were stained dark.  The authors  suggested that this affect might be due to
         respiratory stress.  The NOAEL for this study was 25  mg/kg/day, and the Lowest-
         Observed-Adverse-Effect Level (LOAEL) was 70 mg/kg/day based on the increased
         mortality in both sexes and histopathological findings of low-to-moderate congestion
         of the liver, kidneys, and lungs.

      •  The only chronic toxicity study on p-nitrophenol was an abstract obtained from the
         U.S.S.R.  The data provided information that was qualitative but inadequate  to
         make quantitative  assessment (Makhinya,  1969).  Chronic administration (probably
         oral, dose not mentioned) of  p-nitrophenol to warm-blooded animals adversely
         affected "neurohumoral regulation." Higher doses of p-nitrophenol caused gastritis,
         enteritis,  colitis, hepatitis, neuritis and hyperplasia of the spleen, and inhibition of
         oxidation processes.

  Reproductive Effects

      •  Plasterer et al. (1985) evaluated the effects of prenatal exposure to selected
         chemicals in a screening protocol designed to detect reproductive effects, p-
         nitrophenol disserved in com  oil was administered via gavage at 0 or 400 mg/kg/day
         to groups of 50 pregnant CD-I  mice during gestational days 7 to 14.  A limited
         number of end points were studied: for maternal toxicity, they were mortality and
         body weight; for developmental toxicity, they were number of live and dead pups
         per litter, pup body weight on days 1 and 3 and gross structural abnormalities on
         day 1.  Evidence of maternal  toxicity included a significantly lower survival rate
         (80%) among  treated animals, as well as a significantly (p < 0.05) decreased weight
         gain (18.7 g versus 22.8 g in the control group). Developmental toxicity was
         manifested as  a slight (nonsignificant) reduction in the number of live pups per
         litter (9.8 versus 10.8 in the control group on day  1).

                                           10

-------
P-Nitrophenol                                                                April, 199Z
      •  Angerhofer (1985) conducted a two-generation dermal study to investigate the
         reproductive effects of p-nitrophenol (a leather fungicide incorporated into combat
         boots), p-nitrophenol dissolved in 100% ethanol was applied along the dorsal body
         line over a 4- by 10-cra area; the doses were 0 (ethanol control), 0 (saline control),
         50, 100 and 250 rag/kg/day, 5 days/week, applied to groups of 24 female  and  12
         male Sprague-Dawley rats. The F0 generation was exposed for 140 days before
         mating.  The females continued to be exposed through breeding, gestation, and
         lactation.  The Ft generation was exposed for 168 days after weaning; the females
         again were exposed through breeding, gestation, and lactation.  Toxic signs included
         a dose-related pattern of dermal irritation (erythema, scaling, and cracking).  No
         reproductive effects were observed.

  Developmental Effects

      •  Kavlock (1990) evaluated the developmental toxicity of p-nitrophenol in a structure-
         activity relationship study of phenols, p-nitrophenol (dissolved in a mixture of
         water, Tween 20ฎ, propylene glycol, and ethanol in a ratio of 4:4:1:1) was
         administered via gavage at 0, 100, 333, 667, or 1,000 rag/kg on  gestational day 11 to
         groups of 12 to 13 Sprague-Dawley rats. Maternal toxicity end points included
         overt signs of toxicity, mortality, body weight and implantation scars at the end of
         weaning.  Developmental toxicity end points included viability,  weight of offspring
         at postnatal days 1, 3 and 6, overt malformations, and perinatal loss.  Mortality
         among dams exposed to 667 and 1,000 mg/kg of />-nitrophenol was  3/13 and 4/12,
         respectively (versus 0 in all other groups). At 333, 667, and  1,000 rag/kg, the litter
         size on postnatal days 1 and 6 decreased nonsignificantly. The NOAEL was  100
         mg/kg, and the LOAEL was 333 rag/kg.

  Mutagenicity

      •  p-Nitrophenol  (10 to 500 jig/plate) was not rautagenic in the  Salmonella
         typhimurium/mammaAian rnicrosome plate incorporation assay (Ames test)
         performed by McCann et al. (1975).

      •  As part of the  National Toxicology Program (NTP), o- and p-nitrophenols were
         among 250 coded compounds evaluated by independent  laboratories in the
         preincubation modification to the standard Ames test (Haworth et  al., 1983). Over
         a concentration range of 10 to 3333 /ig/plate (para), there was  no evidence of a
         mutagenic response in S. typhimurium TA1535, TA1537, TA98, or TA100 either  in
         the absence or presence of liver S9 preparations from Aroclor 1254-induced rats or
         hamsters.  Similarly, 1 to 100 fig/plate of p-nitrophenol was negative in
         preincubation Ames tests conducted with 5. typhimurium TA98 and TA100 (Suzuki
         et al., 1983). The inclusion of the co-mutagen norharman into  the S9 mix did not
         cause an increase in mutant colony counts of either strain exposed to the three
         isoraers.  The limited data presented by Kawai et al. (1987) support the earlier
         findings  that p-nitrophenol is not mutagenic  under preincubation conditions.

                                          11

-------
P-Nitrophenol                                                               April, 1992,
      •  Under the conditions of the nitro reduction preincubation Ames test performed by
         Dellarco and Prival (1989), ?             assayed to cytotoric levels, was not
         .mutagenic in the presence of rat liver S9-mix containing flavin mononucleotide
         (FMN).  It was noteworthy that the investigators presented data demonstrating the
         ability of FMN to facilitate the nitro reduction of two nonmutagenic nitro
         compounds (i.e., nitrobenzene and p-nitrophenol).  It was concluded that the lack of
         a rnutagenic response by p-nitrophenol in this well-conducted study was not related
         to a failure of FMN to  reduce this compound.

      •  Nonactivated p-nitrophenol (125 to 2,000 jig/plate) was not genotoric in DNA
         repair-deficient strains of 5. typhimurium or Escherichia coli (Rashid and Mumma,
         1986).

      •  Major data gaps exist in this genetic  toxicology evaluation of p-nitrophenol (i.e., no
         mammalian cell gene mutation or cytogenetic assays were found in the available
         literature); however, the Ames test has an established record of correctly identifying
         mutagenic nitroaromatics as carcinogens (Klopman et al., 1990). Accordingly,
         Tennant et al. (1990) have used the weight of evidence  from well-conducted
         Salmonella  gene mutation assays showing no indication of mutagenesis to predict
         that p-nitrophenol will not be a carcinogen in the rodent carcinogenicity bioassay
         that is currently being performed by  the NTP.

  Carcinogenicitv

      •  Only one carcinogenicity study was found in the literature. Thirty-one mice
         received single biweekly 25-^iL dermal applications of a 20% dioxane solution (2.5
         mg) of p-nitrophenol applied to  the back for 12 weeks.  No skin  tumors were
         observed during the study (Boutwell  and Bosch, 1959).
V.    QUANTIFICATION OF TOXICOLOGICAL EFFECTS

      Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
and Lifetime exposures if adequate data are available that identify a sensitive noncarcinogenic
end point of toxicity. The HAs for noncarcinogenic toxicants are derived using the following
formula:

       TTA        (NOAEL or LOAEL) x (bw)           ...    .  .4
       HA =        (UF) x (_ L/day)          = — mg/L (rฐunded tO
                                          12

-------
P-Nitrophenol                                                               April, 1992.
where:

      NOAEL or LOAEL    =   No- or Lowest-Observed-Adverse-Effect Level in mg/kg
                                 bw/day.

                     BW    =   assumed body weight of a child (10 kg) or an adult
                                 (70 kg).

                      UF    =   uncertainty factor (10, 100, 1,000 or 10,000), in
                                 accordance with EPA or NAS/OW guidelines.

                    L/day    =   assumed daily water consumption of a child (1 L/day) or
                                 an adult (2 L/day).

One-day Health Advisory
      Only acute LDjo studies of p-nitrophenol were available. The acute oral LDK values in
rats ranged from 50 to 230 mg/kg (Eastman Kodak Company, 1980; Vernot et al., 1977;
Branch, 1983a).  In mice, the oral LDj,, values ranged from 200 to 470 mg/kg (Vernot et al.,
1977; Eastman Kodak, 1980). Clinical signs of toxicity observed in these studies included
salivation, lethargy, ptosis, prostration, convulsions, dyspnea, and death.  In rabbits, the oral
MLD was 600 to 900 mg/kg (Monsanto Chemical Co., 1956).  The intraperitoneal LD^ in rats
was 50 to 97 mg/kg, whereas that in mice was 50 mg/kg (Von Oettingen, 1941; Eastman
Kodak Company,  1980).  In a developmental toxicity study by Kavlock (1990), Sprague-
Dawley rats were not affected when exposed to a single oral dose of 100 mg/kg p-nitrophenol
(lowest dose tested) on day 11 of gestation.  Although this dose, 100 mg/kg, appears to  be a
NOAEL in this study, the acute oral LD^ in rats was also reported to vary between 50  and
230 mg/kg. As noted, an LD^ of 50 mg/kg is half the  dose considered to be a NOAEL in the
Kavlock (1990) study.  Therefore, in the absence of adequate short-term studies to calculate a
One-day HA for a 10-kg child, it is recommended that the Longer-term HA for a 10-kg child
calculated below, 0.833 mg/L (rounded to 800 pg/L), be used as a conservative estimate of the
One-day HA value for p-nitrophenol.

Ten-day Health Advisory

      No adequate short-term study was available to calculate a Ten-day HA for
p-nitrophenol. The range-finding experiment performed to determine the doses for the study
by Hazleton (1989) did not result in any toxicity  at the highest dose used (100 mg/kg) after 4
weeks of exposure.  However, because of the design of this study, these data are inadequate
for use  in calculating the Ten-day HA, especially since the acute oral LDj,, in rats may be as
low as 50 mg/kg. Therefore, it is recommended  that the Longer-term  HA for a 10-kg child
below, 0.833 mg/L (rounded to 800 /ig/L), be used as a conservative estimate  of the Ten-day
HA for p-nitrophenol.
                                          13

-------
P-Nitrophenol                                                               April,  1992.
Longer-term Health Advisory

      The subchronic study reported by Hazleton Laboratories America, Inc. (1989) in rats
was used to determine the Longer-term HA. ^-nitrophenol was administered daily for 13
weeks to Sprague-Dawley rats of both sexes at 0, 25, 70, and 140 mg/kg. Mortality occurred
at the mid and high doses but was significant only at the high dose.  In raid-dose females and
all high-dose rats, p-nitrophenol caused dark discoloration of the lungs, livers and kidneys.
This effect was thought to be an indirect result of respiratory stress. The NOAEL in this
study was 25 mg/kg/day, and the LOAEL value was 70 mg/kg/day based on lung, liver, and
kidney effects and on  the unscheduled mortalities.

      The Longer-term HA for the 10-kg child is calculated as follows:

      Longer-term HA =  (25  mg/kft/dav) (10 kg)   =.  0.833 rag/L (rounded to 800 Mg/L)
                            (300)  (1 L/day)
where:
             25 mg/kg/day     =   NOAEL, based on the absence of clinical signs and only
                                 minimal to moderate congestion noted in the
                                 histopathological  examination (Hazleton Laboratories,
                                 1989).

                    10 kg     =   assumed weight of a child.

                      300     =   uncertainty factor, this 100-fold uncertainty factor was
                                 chosen in accordance with EPA or NAS/OW guidelines
                                 for use of a NOAEL from an animal study.  An extra
                                 three-fold UF was used for lack of adequate
                                 reproductive/developmental data.

                  1 L/day     =   assumed water consumption of a 10-kg child.
      The Longer-term HA for the 70-kg adult is calculated as follows:

      Longer-term HA = ^/ffi8??]/'7? k^ = 2.916 mg/L (rounded to 3,000
where:
             25 mg/kg/day     =   NOAEL, based on the absence of clinical signs and only
                                 minimal to moderate congestion noted in the
                                 histopathological examination (Hazleton Laboratories,
                                 1989).
                                          14

-------
P-Nitrophenol                                                                April, 1992.
                    70 kg    =   assumed weight of an adult.

                      300    =   uncertainty factor, this 100-fold uncertainty factor was
                                 chosen in accordance with EPA or NAS/OW guidelines
                                 for use of a NOAEL from an animal study.  An extra 3-
                                 fold UF was used for lack of adequate
                                 reproductive/developmental data.

                  2 L/day    =   assumed water consumption of a 70-kg adult.


      Note that the Longer-term HA for the 70-kg adult, 2.9 rag/L, is also close to the water-
odor threshold ofp-nitrophenol of 2.5  mg/L.

Lifetime Health Advisory

      The Lifetime HA represents that portion of an individual's  total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure. The Lifetime  HA is derived in a three-step process.  Step 1
determines the Reference  Dose (RfD), formerly called the Acceptable Daily Intake (ADI).
The RfD is an estimate of a daily  exposure to the human population that is likely to be
without appreciable risk of deleterious effects over a lifetime, and is derived from the
NOAEL (or LOAEL),  identified from a chronic (or subchronic) study, divided by an
uncertainty factor(s). From the RfD, a Drinking Water Equivalent Level (DWEL) can be
determined (Step 2). A DWEL is a medium-specific  (i.e., drinking water) lifetime  exposure
level,  assuming 100% exposure from that medium, at which adverse, noncarcinogenic health
effects would not  be expected to occur. The DWEL is derived from the multiplication of the
RfD by the assumed body weight of an adult and divided by the assumed daily water
consumption of an adult.  The Lifetime HA is determined in Step 3 by factoring in other
sources of exposure, the relative source contribution  (RSC). The RSC from drinking water is
based on actual exposure data or,  if data are not available, a value of 20% is assumed.  If the
contaminant is classified as a  known, probable or possible carcinogen, according to the
Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986), then caution must
be exercised in making a decision  on how to deal with possible lifetime  exposure  to this
substance.  For human (A) or probable human (B) carcinogens, a Lifetime HA is not
recommended.  For possible human carcinogens (C),  an additional 10-fold safety factor is
used in the calculation  of the  Lifetime HA.  The risk  manager must balance this assessment
of carcinogenic potential and  the quality of the data against the likelihood of occurrence  and
significance of health effects related to noncarcinogenic end points of toxicity.  To assist the
risk manager in this process, drinking water concentrations  associated with estimated excess
lifetime cancer risks over the  range of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult
drinking 2 L of water per day are  provided in the Evaluation of Carcinogenic Potential
section.

      The study by Hazleton  Laboratory America, Inc. (1989) has been selected to serve as
the basis for calculation of the Reference Dose (RfD).  In this study, adult Sprague-Dawley
                                          15

-------
P-Nitrophenol                                                               April, 199Z
rats (20/sex/group) were dosed daily for 13 weeks with p-nitrophenol at dosages of 25, 70, and
140 mg/kg/day.  The end points measured included clinical observations, body weights, food
consumption, hematology and clinical chemistry values, ophthalmoscopic examination, organ
weights, and histopathology.  The NOAEL was 25 rag/kg/day based on the absence of clinical
signs, and the LOAEL was 70 mg/kg/day based on  miniraal-to-moderate congestion of the
liver, kidneys, and lungs.  Other studies were not available in the literature.

      Using this study,  the Lifetime HA is derived as follows:

Step 1:  Determination  of the RfD
        RfD = (25 mY  = Qm rag/kg/day
                   (3,000)

where:

       25 rag/kg/day     =   NOAEL, based on the absence of clinical signs and only
                            minimal to moderate congestion noted in the histopathologjcal
                            examination (Hazleton Laboratories, 1989).

              3,000     =   uncertainty factor, this uncertainty factor was chosen in
                            accordance with EPA or NAS/OW guide- lines for use of a
                            NOAEL from an animal study of less-than-lifetime  duration.
                            An extra three-fold UF was used for lack of adequate
                            reproductive/developmental data.

Step 2:  Determination of the Drinking Water  Equivalent Level (DWEL)

        DWEL = (0-008 mfc/kg/day) (70 kg) = Q 2g mg/L (rounded to 300 Mg/L)
                       (2 L/day)

where:

     0.008 mg/kg/day     =   RfD.

              70 kg     =   assumed weight of an adult.

            2 L/day     =   assumed water consumption of a 70-kg adult.

Step 3:  Determination of the Lifetime HA

        Lifetime  HA = 0.28 mg/L x 20% = 0.056 mg/L (rounded to 60 /tg/L)
                                          16

-------
P-Nitrophenol                                                             April, 1992.



where:

          0.28 mg/L     =   DWEL.

              20%     =   assumed relative source contribution from drinking water.


Evaluation of Carcinogenic Potential

      • Only one dermal toxicity study of/7-nitrophenol was found in the literature.
        Thirty-one mice received single biweekly 25-/iL applications of a 20% dioxane
        solution (2.5 mg) of p-nitrophenol applied to the back for 12 weeks.  No skin tumors
        were observed throughout the duration of the study (Boutwell and Bosch, 1959).

      • Applying the criteria described by the U.S. EPA (U.S. EPA,  1986), ^-nitrophenol
        may be classified in Group D:  not classifiable. This category is for agents with
        inadequate evidence of carcinogenicity in animals.


VI.   OTHER CRITERIA. GUIDANCE AND STANDARDS

      • A taste threshold of 43.4 rag/L was reported by National Academy of Sciences
        (1982), and a water-odor threshold of 2.5 mg/L was reported in the Handbook of
        environmental data on organic chemicals (Verschueren, 1983).


VII.   ANALYTICAL METHODS

      • Gas chromatography  (GC) and high-performance liquid chromatography (HPLC)
        have been used to identify and quantify /r-nitrophenol in environmental  samples and
        in human urine (Kirby et al., 1979; Alber et ah, 1989). The detection limits for GC
        and HPLC were 0.5 /zg/L. However, Alber et al. (1989) reported that the resolution
        of the GC method was better than that obtained from HPLC analysis.

VTIL  TREATMENT TECHNOLOGIES

      • To be prepared by the U.S.  EPA.
                                        17

-------
P-Nitrophenol                                                              April,  199Z.
IX.  REFERENCES

Alber, M., H.B. Bohm, J. Brodesser, J. Feltes, K. Levsen and H.F. Scholer.  1989.
     Determination of nitrophenols in rain and snow.  Fres. Z. Anal. Chern. 334:540-545.

Alexander, M. and B.K. Lustigraan.  1966.  Effect of chemical structure on microbial
     degradation of substituted benzenes. J. Agric.  Food Chera. 14:410-413.

Angerhofer,  R.A.  1985. Effect of dermal applications of para-nitrophenol on the
     reproductive functions of rats. Report no. 75-51-0047-85.  Aberdeen Proving Ground,
     MD: U.S. Army, Environmental Hygiene Agency.

Batchelder, T.L.  1975. Environmental analysis of panz-nitrophenol. Final report no. 86-
     89000113S.  Washington, DC: U.S. Environmental Protection Agency Office of Toxic
     Substances. Revised in 1986.

Boutwell, R.K. and D.K. Bosch.  1959.  The tumor-promoting action of phenol and related
     compounds for mouse  skin. Cancer Res. 19:413-424.

Branch,  D.K. Monsanto Agricultural Products Company.  1983a.  Acute oral toricity of p-
     nitrophenol to rats. Project no. ML-82-131-A.  Report no. 86-890000358. Washington,
     DC:  U.S. Environmental Protection Agency.

Branch,  D.K. Monsanto Agricultural Products Company.  1983b.  Primary skin irritation of
     p-nitrophenol  to rabbits. Project no. ML-82-131-B.  Report no. 68-890000361.
     Washington, DC:  U.S. Environmental Protection Agency.

Branch,  D.K. Monsanto Agricultural Products Company.  1983c.  Primary eye irritation ofp-
     nitrophenol to rabbits.   Project no. ML-82-131-C. Report no. 86-890000360.
     Washington, DC:  U.S. Environmental Protection Agency.

Cameron, M.A.M.  1958.  The action of nitrophenols on the metabolic rate of rats. Br. J.
     Pharraacol.  13:25-29.

Dellarco, V.L. and  MJ. Prival. 1989.  Mutagenicity of nitro compounds in Salmonella
     typhimuriwn in the presence of flavin mononucleotide in a preincubation assay.
     Environ. Mol. Mutagen. 13:116-127.

Eastman Kodak Company.  1980. Toxicology Laboratory Report no. 125877V. U.S. EPA
     Report no. 86-890000202. Rochester, New York: Health,  Safety and Human Factors
     Laboratory.

Ellis, D.D., C.M. Jones, R.A Larson and D J. Schaeffer. 1982.  Organic constituents of
     mutagenic secondary effluents from wastewater treatment plants.  Arch. Environ.
     Contam. Toxicol. 11:373-382.

                                         18

-------
P-Nitrophenol                                                               April,
Gorge, G., J. Beyer and K. Urich.  1987. Excretion and metabolism of phenol, 4-nitrophenol
     and 2-methylphenol by the frogs Rana temporaria and Xenopus laevis.  Xenobiotica
     17-1293-1298.

Hawley, G.G.  1981.  The condensed chemical dictionary, 10th ed.  New York: Van Nostrand
     Reinhold Co.

Haworth, S., T. Lawlor, K. Mortelraans, W. Speck and E. Zeiger.  1983. Salmonella
     mutagenicity test results  for 250 chemicals.  Environ. Mutagen. 5(Suppl. 1):3-142.

Hazleton Laboratories America, Inc.  1989.  Subchronic  toxicity study in rats with para-
     nitrophenol. HLA Study no. 241-221.  St. Louis, MO: Monsanto Chemical Company.

Hazleton Laboratories America, Inc.  1983.  Subacute dust inhalation study (withp-
     nitrophenol). Project no. 241-139. U.S. EPA Report no. 86-890000362.  St. Louis, MO:
     Monsanto Chemical Company.

Jetzer, W.E., S.Y.E. Hou, A.S. Huq, N. Duraiswamy, N.F.H. Ho and G.L. Flynn.  1988.
     Temperature dependency of skin permeation of waterborne organic compounds.  Pharm.
     Acta Helv. 63(7): 197-201.

Kavlock, R.J. 1990. Structure-activity relationships in the developmental toxicity of
     substituted phenols: In vivo effects. Teratology 41:43-59.

Kawai, A., S. Goto, Y. Matsumoto and H. Matsushita. 1987.  Mutagenicity of aliphatic and
     aromatic nitro compounds.  Japn. J. Ind. Health 29:34-54.

Kirby, K.W., J.E. Keiser, J. Groene and E.F. Slach.  1979.  Confirmation of /wa-nitrophenol
     as a human urinary metabolite at the nanogram level.  J. Agric. Food Chera. 27(4): 757-
     759.

Klopman, G., M.R. Frierson and H.S. Rosenkrantz.  1990. The structural basis of the
     mutagenicity of chemicals in Salmonella typhimurium: The Gene-Tox database. Mutat.
     Res. 228:1-50.

Leo, A., C. Hanch and D. Elkins.  1971. Partition coefficients and their uses.  Chem. Rev.
     71:525-616.

Lokke, H.  1985.  Degradation of 4-nitrophenol in two Danish soils.  Environ. Pollut. (series
     A) 38:171-181

Machida, M., Y. Morita, M. Hayashi and S. Awazu.  1982. Pharmacokinetic evidence for the
     occurrence of extrahepatic conjugative metabolism of ^-nitrophenol in rats.  Biochem.
     Pharmacol. 31(5):787-791.
                                          19

-------
P-Nitrophenol                                                               April, 1992-
Makhinya, A.P.  1969.  Comparative hygienic and sanitary-toxicological studies on nitrophenol
     isoraers in relation to their normalization in drinking waters.  Prom. Zagryazgneniya
     Vodoemov. 9:84-85.  Chemical Abstract 72:4723/c.

McCann, J., E. Choi, E. Yamasaki and B.N. Ames.  1975. Detection of carcinogens as
     mutagens in the Sabnonella/micTOsome test: Assay of 300 chemicals. Proc. Nat Acad
     Sci. USA 72:5135-5139.

Meerman, J.H.N., C. Nijland and GJ. Mulder.  1987.  Sex differences in sulfation and
     glucuronidation of phenol, 4-nitrophenol and n-hydroxy-2-acetylaminofluorine  in the rat
     in vivo.  Biochera. Pharmacol. 36(16): 2605-2608.

Monsanto Chemical Co.  1956.  A certificate of analysis and summary of toxicological
     investigation of ^ara-nitrophenol. Project no. Y-56-56. Report no. 86-890000356.
     Washington, DC:  U.S. Environmental Protection Agency.

National Academy of Sciences.  1982.  Drinking Water and Health, vol. 4.  Washington, DC:
     National Academy Press, p. 230.

NIOSH.  1981. National Institute for Occupational Safety and Health Information profiles
     on potential occupational hazards: Nitrophenols.  2nd draft Center for Chemical
     Hazard  Assessment, Syracuse Research Corporation. TR81-536. Cincinnati,  OH:
     NIOSH.

Plasterer,  M.R., W.S. Bradshaw, G.M. Booth and M.V. Carter.  1985. Developmental toxicity
     of nine selected compounds following prenatal exposure in the mouse:  Naphthalene, p-
     nitrophenol, sodium selenite, dimethylphthalate, ethylenethiourea and four glycol ether
     derivatives.  J. Tox. Environ. Health  15:25-38.

Rashid, K.A.  and R.O. Mumraa.  1986. Screening pesticides for their ability to damage
     bacterial DNA.  J. Environ. Sci. Health 4:319-334.

Reinke, L.A.  and MJ. Moyer.  1985. ^-nitrophenol hydroxylation:  A microsomal oxidation
     which is highly inducible by ethanol.  Drug Metabol. Disp. 13:548-552.

Robinson, D., J.N. Smith and R.T. Williams.  1951. Studies in detoxication 39.  Nitro
     compounds, (a) The metabolism of o-, m-, and^-nitrophenols in the rabbit, (b) The
     glucuronides of the mononitrophenols and observations of the anomalous optical
     rotations of triacetyl beta-o-nitrophcnylglucuronide and its methyl ester.  Biochera.  J.
     50:221-227.

Shen, T.C.R.   1962.  The stimulating effect of dinitro-ort/io-cresol, dinitro-phenol zndpara-
     nitrophenol on the aortic chemoreceptors in dogs. Arch. Int. Pharmacodyn. 140(3-
     4):521-527.
                                          20

-------
P-Nitrophenol                                                               April, 1992.
Shen, T.C.R. and W.H. Hauss.  1939.  Influence of dinitro-phenol 1,2,4-dinitro-ort/w-cresol
     and/?ara-nitrophenol upon the carotid sinus chemoreceptors of the dog.  Arch. Int.
     Pharmacodyn. 63:251-258.

Siragusa, G.R. and R.D. DeLaune.  1986.  Mineralization and sorption of p-nitrophenol in
     estuarine sediment.  Environ. Tox. Chem. 5:175-178.

Spain, J.C., P.H. Pritchard and A.W. Bourquin.  1980.  Effects of adaptation on
     biodegradation rates in sediment/water cores from estuarine and freshwater
     environments. Appl. Environ. Microbiol. 40:726-734.

Spain, J.C. and P.A. Van Veld.  1983.  Adaptation of natural microbial communities to
     degradation of xenobiotic compounds: Effects of concentration, exposure time,
     inoculum and chemical structure. Appl. Environ. Microbiol. 45:428-435.

Spain, J.C., P.A. Van Veld, CA. Monti, P.H. Pritchard and C.R. Gripe.  1984: Comparison of
     p-nitrophenol biodegradation  in field and laboratory test systems. Appl. Environ.
     Microbiol. 48(5):944-950.

Sultatos, L.G. and L.D. Minor.  1985.  Biotransformation of paraxon and/7-nitrophenol by
     isolated perfused mouse livers. Toxicology 36:159-169.

Suzuki, J., T. Koyama and S. Suzuki.  1983.  Mutagenicities  of mono-nitrobenzene derivatives
     in the presence of norharman. Mutat. Res. 120:105-110.

Tabak, H.H., S.A. Quave, C.I. Mashni and E.F. Earth. 1981.  Biodegradability studies with
     organic priority pollutant compounds. J. Water Pollu.  Control Fed. 53:1503-1518.

Tennant, R.W., J. Spalding, S. Stasiewicz and J. Ashby. 1990. Prediction of the outcome of
     rodent carcinogenicity bioassays currently being conducted on 44 chemicals  by  the
     National Toxicology Program.  Mutagenesis 5:3-14.

Tremaine, L.M., G.L. Diamond and AJ. Quebbemann.  1984. In vivo quantification of renal
     glucuronide and sulfate conjugation of 1-naphthol andp-nitrophenol in the rat.
     Biochem. Pharmacol. 33:419-427.

U.S. EPA.  1986. U.S. Environmental Protection Agency.  Guidelines for carcinogen risk
     assessment.  Fed. Reg. 51(185):33,992-34,003.  September 24.

Vernot, E.H., J.D. MacEwen, C.C.  Haun and E.R. Kinkead.  1977. Acute toxicity and skin
     corrosion data for some organic and inorganic compounds  and aqueous solutions.
     Toxicol. Appl. Pharmacol. 42:417-423.

Verschueren, K.  1983. Handbook  of environmental data on organic chemicals,  2nd ed.
     New York:  Van Nostrand Reinhold Co., p. 919.

                                          21

-------
P-Nitrophenol                                                                April, 199,2.
Von Oettingen, W.F.  1941.  The aromatic amino and nitro compounds:  Their toxicity and
     potential dangers.  A review of the literature.  U.S. Pub. Health Bull. 271:130-155.

Windholz, M., S. Budavari, R. Bluraetti and E.S. Otterbein, eds.  1989.  The Merck index:
     An encyclopedia of chemicals and drugs, llth ed.  Rahway, NJ: Merck & Co., Inc., p.
     1,047.

Yoshida, M., M. Sunaga, I. Kara, M. Katsumata and M. MJnami.  1989.  Elevation of urinary
     n-acetyl-beta-J-glucosaminidase and beta-galactosidase  activities in workers with  long-
     term exposure to aromatic nitro-amino compounds.  Bull. Environ. Contain. Toxicol.
     43:1-8.

Zaidi, B.R., Y. Murakami and M. Alexander.  1989. Predation and inhibitors in lake water
     affect the success of inoculation to enhance biodegradation of organic chemicals.
     Environ. Sci. Technol. 23:859-863.
                                          22

-------
EPA  0553

RX000027511
                                                                            .    April 1992
                                        PHENOL
                              Drinking Water Health Advisory
                                      Office of Water
                           U.S. Environmental Protection Agency
 I.     INTRODUCTION
       The Health Advisory Program, sponsored by the Office of Water (OW), provides
information on the health effects, analytical methodology and treatment technology that would
be useful in dealing with the contamination of drinking water.  Health Advisories describe
nonregulatory concentrations of drinking water contaminants at which adverse health effects
would not be anticipated to occur over specific exposure durations.  Health Advisories contain a
margin of safety to protect sensitive members of the population.

       Health Advisories serve as informal technical guidance to assist Federal, State and local
officials responsible for protecting public health when emergency spills or contamination
situations occur. They are not to be construed as legally enforceable Federal standards. The
HAs are subject to change as new information becomes available.

       Health Advisories are developed for One-day, Ten-day, Longer-term (approximately 7
years, or 10% of an individual's  lifetime) and Lifetime exposures based on data describing
noncarcinogenic end points of toxicity.  For those substances that are known or  probable human
carcinogens, according to the Agency classification scheme (Group A or B), Lifetime Health
Advisories  are not recommended. The chemical concentration values for Group A or B
carcinogens are correlated with carcinogenic risk estimates by  employing a cancer potency (unit
risk) value  together with assumptions for lifelong exposure and the ingestion of water.  The
cancer unit risk is usually derived from  a linearized multistage model with 95%  upper
confidence  limits providing a low-dose estimate of cancer risk. The cancer  risk  is usually.
derived from  the linear multistage model with 95% upper confidence limits. This provides a
low-dose estimate of cancer risk to humans that is considered unlikely to pose a carcinogenic
risk in excess of the stated values. Excess cancer risk estimates may also be calculated using the
one-hit, Weibull, logit or probit models. There is no current understanding of the biological
mechanisms involved in cancer to suggest that any one of these models is able to predict risk
more accurately than another.  Because each model is based on differing  assumptions, the
estimates that are derived can differ by several orders of magnitude.

-------
Phenol                                                                        April 1992



II.     GENERAL INFORMATION AND PROPERTIES

       CAS No. 108-95-2

       Structural Formula
                                        Phenol

       Synonyms

       •     Carbolic acid; hydroxybenzene; oxybenzehe; phenic acid; phenyl hydroxide;
             phenylic acid (Windholz et al., 1983).

       Uses

       •     Phenol is used as a disinfectant, antiseptic,  bactericide, and antimicrobial agent.
             It is also used in the manufacture of resins  and medical and industrial organic
             compounds and dyes.  In addition, phenol serves as a solvent for petroleum
             refining and as a reagent in chemical analysis (Windholz et al., 1983).

       Properties  (Leonardos et al., 1969; U.S. EPA, 1980; Windholz et al., 1983)
             Chemical Formula
             Molecular Weight                         94.11
             Physical State (25ฐC)                      Colorless, acicular crystals or
                                                       white crystals
             Boiling Point  182ฐC
             Melting Point                             43 ฐC (40.85 ฐC for ultrapure
                                                       material)
             Density                                  1.071 g/mL
             Vapor Pressure (25 ฐC)                    03513 mm Hg
             Specific Gravity (water = 1 )                 1 .0722 at 20/24 ฐC
             Water Solubility (16 ฐC)                    66.7 g/L
             Log Octanol Water Partition               —
               Coefficient (log K,,,,)
             Odor Threshold                          0.05 ppm (air); 1.0 mg/L (water)
             Taste Threshold                          0.3 mg/L
             Conversion Factor (25ฐC,                  1 mg/raj = 0.263 ppm
               760  mm Hg)                            1 ppm = 3.84 mg/mj

-------
Phenol                                                                         April  1992
       Occurrence

       •     The Organic Monitoring Survey Program (U.S. EPA, 1980) qualitatively detected
             phenol (i.e., carbolic acid) in 2 of 110 raw water supplies; the compound was
             identified by gas-liquid chromatography and mass spectrometry, but no
             quantification was made. The National Commission on Water Quality found that
             the annual mean concentration of phenol from the lower Mississippi  River was
             1.5 pg/L (U.S. EPA, 1980).

       •     In studies conducted from 1972 through 1977, water samples from Lake Huron,
             the Detroit River and the St. Clair River were analyzed for phenol (Konasewich
             et al., 1978). Phenol levels varied widely, from nondetectable to 24 mg/L;
             concentrations of phenol in the Detroit River ranged from <0.5 to 5 /xg/L.

       •     Traces of phenol (undetectable to 100 jig/L) have been reported in other rivers in
             the United States (Elder et al., 1981; Jungclaus et al., 1978; Sheldon  and Hites,
             1979).

       •     Phenol has been found in trace quantities (30 ppt) in urban/ suburban air in
             Columbus, OH; however, its concentration has been found to be considerably
             higher (520 to 44,000 ppt) in industrial areas of the United States (Brodzinsky
             and Singh, 1982).  Some of the phenol in the air might be from automobile use;
             Kuwata et al. (1980) has reported phenol at an average concentration of 0.29
             ppm in auto  exhaust.  Phenol was also found in particulate matter in the air, at a
             mean concentration of 52 ppt, during a smog episode in West Covina, CA
             (Cronn et al., 1977).

       Environmental Fate

       •     On an aqueous environment in which microbes are present (e.g., activated sludge
             or pond or river water), phenol degrades completely within one to several days
             (Kincannon et al., 1983; Petrasek et al., 1983; Richards and Shieh, 1986; Tabak et
             al., 1981); in natural estuarine water, phenol was estimated to persist for as long
             as 72 days (Lee and Ryan,  1979). Phenol has been found to degrade completely
             within  2  to 5 days when mixed with wet, nonsterile soil in air (Rizet et al., 1977;
             Walker,  1954); however, anaerobic degradation of phenol was only 20% of
             aerobic degradation after 40 days (Baker and Mayfield, 1980).  Rees and King
             (1981) found that high concentrations of phenol (>3 g/L) will inhibit microbial
             biodegradation by destroying the organisms. In contrast to these laboratory
             results, detectable concentrations of phenol have been found in water taken  from
             aquifers  15 to 18 months after accidental spills had occurred (Baker  et al., 1978;
             Delfino and Dube,  1976).

-------
Phenol                                                                          April 19921
              Phenol, at ppb to ppm concentrations in water, may be destroyed (photolyzed) by
              sunlight or converted to odorous chlorophenols if the water is chlorinated
              (Wajon et al., 1982). Phenol does not evaporate  from water (Aaberg et al., 1983)
              and is only moderately volatile as a pure material or on dry soil (Ehrlich,  1982).

              A review of the available literature revealed no information concerning the
              bioaccumulation of phenol by aquatic microorganisms or by aquatic invertebrates
              or vertebrates (Aaberg et al.,  1983; Branson, 1978;  Kobayashi et al., 1979).

              Although the fate of phenol released into the atmosphere at low levels has not
              been studied comprehensively, it has been calculated that its half-life, by reaction
              with sunlight- generated hydroxyl radicals, is 0.7 to  14 hours; this value is
              dependent on the amount of other pollutants present (Finlayson-Pitts and Pitts,
              1986; Hendry and Kenley, 1979). The nitrate radical, formed in the air from
              pollutants and sunlight, also reacts rapidly with phenol (Carter et al., 1981) with a
              calculated half-life of 0.28 to 12.5 minutes, depending on the nitrate radical
              concentrations (Finlayson-Pitts and Pitts,  1986).  Other atmospheric processes,
              including direct photolysis and adsorption on particulates, appear to be  insignifi-
              cant to the environmental fate of phenol (Callahan  et al.,  1979).
III.    PHARMACOKINETICS

       Absorption

       •      Absorption of a small single diluted oral dose of phenol by humans and rats is
              high; approximately 85 to 100% of phenol administered by this route is absorbed
              within 24 hours.  Retention of inhaled phenol ranges between 60 and 88% for
              humans exposed to vapors for 7 hours.  Dermal absorption of phenol vapors is
              proportional to the concentration of the vapor used, with about one-third of the
              phenol concentration in the air being absorbed through the skin of adult men
              and women. Regardless of route, absorption generally is rapid, as evidenced by
              the development of symptoms of toxicity within minutes after administration of,
              or exposure to, phenol.
              Single oral doses of 14C-phenol (0.01 mg/kg, 2.7 pCif person) were readily
              absorbed by the gastrointestinal tract of three healthy men (Capel et al., 1972).
              Approximately 85 to 98% of the radioactivity administered was excreted in the
              urine within 24 hours; fecal 14C was not measured.

              Absorption of phenol by three female Wistar rats was high (91 to 100%)
              following administration of a single oral dose of 25 mg l4C-phenol/kg
              (5.0 /iCi/animal) (Capel et al., 1972).  Radioactivity was recovered in urine only.

-------
Phenol                                                                         April  1992
       •     In a series of 12 experiments conducted by Piotrowski (1971), a group of seven
             healthy human volunteers (six men, one woman) inhaled approximately 5, 10 or
             25 mg phenol/m3 for 8-hour intervals, with two 30-minute breaks at 2.5 and 5.5
             hours after exposure began.  (Actual measured exposures were between 4.8 and
             26.1 mg/m3.).  To avoid absorption of phenol vapor through the skin, subjects
             remained outside the exposure chamber and inhaled the air from inside through
             a face mask connected to the interior of the chamber.  Analysis of urine collected
             during exposure and 24 hours after termination of each experiment showed that
             essentially all (99%) of the phenol inhaled was excreted. Thus, absorption of
             phenol via the inhalation route is rapid and  nearly complete.

       •     On the basis of dermal absorption/vapor exposure  data, Piotrowski (1971)
             calculated an absorption coefficient of 0.35 m3/hr, indicating that for each hour
             human subjects were exposed to phenol vapors (5,  10 or 25 mg/mj for 7 hours),
             they absorbed, through the skin, the amount of phenol contained in 0.35 m3 of
             air.  Clothing provided minimal protection against  dermal absorption of phenol
             vapors.

       Distribution

       •     Deichmann (1944) recovered phenol and/or  its metabolites  in all major organs
             and tissues from groups of five albino rabbits (sex  and strain not given).killed 1
             to 3 minutes after receiving oral doses of 500 mg phenol/kg. The liver contained
             the highest levels of phenol (up to 304 /zg/g tissue), with most (up to 293 jig/g)
             recovered as "free" phenol and the remainder (up to 31  jig/g) as conjugates of
             phenol. Intermediate levels of up to 171,  126, 103  and 71 /*g/g were found in  the
             lungs, blood, brain and spinal cord, and kidneys, respectively. Muscle tissue
             contained totals of 8 to 38 /tg/g, and urine taken from the bladder contained 17
             to 29 /ig/g concentrations of phenol and related compounds.

       Metabolism

       •     Four .metabolites were  recovered from the urine of three healthy adult men
             following administration of single oral doses of 0.01 mg 14C-phenol/kg
             (2.7 /iCi/person) (Capel et al., 1972).  Radiochromato-graphic scans of urine
             samples showed peaks that corresponded to phenyl sulfate,  phenyl glucuronide,
             quinol monosulfate  and quinol  monbglucuronide. The data indicate that humans
             can conjugate phenol with sulfate and glucuronic acid and oxidize it to quinol.

       •     Approximately 65 and 35% of the radioactivity recovered from the urine of two
             female rhesus monkeys given single oral doses of l*C-phenol (50  mg/kg,
             10 Ci/animal) were  associated with phenyl sulfate and phenyl glucuronide,
             respectively (Capel  et al., 1972). Female squirrel and capuchin monkeys
             metabolized a 25-mg/kg dose of MC-phenol (8 /xCi/animal).  The  data indicate

-------
Phenol                                                                         April  1992
             that rhesus monkeys and humans metabolize phenol in a similar manner, with
             phenyl sulfate as the primary urinary metabolite; in contrast, squirrel and
             capuchin monkeys eliminate mostly phenyl glucuronide, with phenyl sulfate and
             quinol glucuronide as secondary metabolites.

             Capel et al. (1972) reported that 30 female ICI mice given single oral doses of
             l4C-phenol (25 mg/kg, 2.5 /xCi/ animal) excreted the same four metabolites
             (phenyl sulfate, phenyl glucuronide, quinol monosulfate and quinol
             monoglucuronide) that humans dosed orally excreted.
       Excretion
             The primary route of elimination for phenol is the urine. Capel et al. (1972)
             reported that 24 hours after ingesting single oral doses of 0.01 mg 14C-phenol/kg
             (2.7 /xCi), three healthy adult men excreted an average of 90% (range, 85 to
             98%) of the administered radioactivity in urine.  Feces were not collected.

             Elimination of a single oral dose of 14C-phenol (25 to 50 mg/kg) was slower from
             female rhesus and squirrel monkeys  than from humans; these monkeys excreted,
             respectively, approximately 43 and 31% of the administered radioactivity in the
             urine within 24 hours post-dosing (Capel et al., 1972).  In contrast, the one
             female capuchin monkey used in this study excreted 73% of the 14C dose (8 /xCi)
             in the urine within 24 hours.  Fecal excretion of radioactivity from any animal
             was not measured.

             Of the seven rodent species examined by Capel et al. (1972), female rats
             eliminated the greatest amount (95%) of a 25-mg/kg oral dose of 14C-phenol in
             the urine during the first 24-hour post-dosing collection period. Two female
             golden hamsters eliminated 73 and 78% of the administered radioactivity
             (5 /iCi/animal) in the urine within 24 hours post-dosing. The elimination of 14C
             was comparable (64 to 68%) for 30 ICI female mice and 4 English guinea  pigs
             given single oral doses of 25 mg/kg 14C-phenol (2.5 to 2.7 pCi/animal). The
             24-hour urinary excretion of phenol metabolites was  lowest for female Egyptian
             jerboas (47%) and female Steppe lemmings (40%). The  authors did not collect
             feces.

             In a series of 12 experiments described earlier (Piotrowski, 1971), approximately
             60 to 88% of the phenol to which individuals were exposed was retained by the
             body, and 99% of this was excreted in urine within 24 hours after exposure was
             terminated.

             Exhaled air of albino rabbits  (five/group) (sex not specified) contained between
             100 and 700 fig phenol at 15, 90. 120, 150 or 360 minutes after administration of
             a single oral dose of 500 mg phcnol/kg to each animal (Deichmann, 1944).  The

-------
Phenol                                          •                              April 1992
             highest concentrations were measured at 2 hours post-dosing, and the total  .
             amount of phenol recovered in the expired air accounted for less than 0.1% of
             that administered.
IV.    HEALTH EFFECTS

       Humans

             Short-term Exposure

             •     The rapid systemic distribution of phenol following acute exposure
                    produces cardiac arrhythmia, wide fluctuations in blood pressure,
                    respiratory distress, reduced body temperature and death due most often
                    to respiratory failure (Bruce et al., 1987).  Ingestion of even small
                    amounts of phenol causes severe burns of the mouth and esophagus, as
                    well as abdominal pain.  The human oral LDLO for phenol was estimated
                    to be  140 mg/kg (Bruce et al., 1987):

             •     An epidemiologjcal study (Baker  et al., 1978) described the outbreak of
                    human illness resulting from the contamination of underground drinking
                    water from an accidental spill of phenol near East Troy, WI,  during July
                    1974.  In this retrospective study, 17 individuals with an average age of
                    21.7 years were estimated to have ingested 10 to 240 mg
                    phenol/person/day over approximately 1 month, primarily during July and
                    August 1974. The authors' exposure estimate cannot be  supported by the
                    data.  The exposed individuals experienced diarrhea, dark urine, mouth
                    sores  and burning of the mouth during that period.  Testing of well water
                    1 week after the spill revealed "phenol" concentrations between 0.21  and
                    126 mg/L; however, it  is  not clear whether the authors actually measured
                    pure phenol or total phenolic compounds. Over the next 6 months the
                    highest phenol concentration was 1,130 mg/L.  Physical examination  of
                    exposed individuals 6 months after the accident revealed no. significant
                    adverse health effects when compared to unexposed controls in terms of
                    incidence of skin rash, mouth lesions, conjunctivitis, and abnormal
                    sensation. Routine blood chemistry, urinalysis (including phenol levels)
                    and liver function tests were normal. A true Lowest-Observed-Adverse-
                    Effect Level (LOAEL) for phenol following  oral exposure in humans
                    cannot be determined from this study because of uncertain exposure
                    levels.

             •     Truppman and Ellenby (1979) reported that 10 of 43 (23%) patients
                    undergoing phenol face peels developed major cardiac arrhythmias when
                    50% or more of the face was painted in less than 30 minutes; the surface

-------
Phenol                                                                          April 19921
                    area of treated skin was also positively correlated with cardiac
                    abnormalities  in rapidly treated patients.  In contrast, no arrhythmias
                    occurred when the same areas of facial skin were painted in a 60-minute
                    period. Cardiac abnormalities were not correlated with age of patient,
                    dosage of intravenous analgesics used during treatment, nasal oxygen
                    administration, phenol formulation (i.e., a 50% [v/v] saponated aqueous
                    mixture or a 46% [w/v] nonsaponated glycerine-water mixture), or prior
                    cardiac status  (only 1 of the 10 affected individuals had a previous cardiac
                    history). The  data indicate that the duration and extent of dermal
                    exposure to phenol are closely related to the development of systemic
                    toxic effects, particularly cardiac arrhythmias.

              •     In a similar study by Gross (1984), cardiac arrhythmias occurred in 39%
                    (21 of 54) of patients who underwent chemical exfoliation of the face and
                    neck simultaneously and in 22% (22 of 100) of subjects who had their
                    faces and necks treated 24 hours apart. The chemical preparation used in
                    this study contained approximately 50% phenol. Serum phenol levels were
                    monitored  throughout these procedures in a total of 21  individuals; single
                    measurements were taken in an additional 15 subjects.  For all but  four
                    patients, serum phenol levels increased as the surface area of treated skin
                    increased. However,  no relationship was observed between serum phenol
                    and irregular heart rhythm. The data suggest that age, sex and cardiac
                    history were not important in predicting susceptibility to cardiac
                    arrhythmias in patients whose skin is treated with phenol.

      Animals

             Short-term Exposure

             •     Flickinger {1976) estimated an oral LDM for phenol in rats (sex and strain
                    not given) of 650 mg/kg/day. The oral LDy, for phenol in Wistar rats was
                    530 or 540 rag/kg when 2 to 10% aqueous phenol was used and 340 mg/kg
                    when 20%  aqueous phenol was used (Deichmann and Witherup, 1944).

             •     Mortality tests were carried out in groups of 20 or 30 Wistar rats of
                    differing ages  by administering a single oral dose of phenol at 600 mg/kg
                    to each animal (Deichmann and Witherup, 1944). After exposure, death
                    occurred in 90% of the 10-day-old rats within 12  to 24 hours; in 30% of
                    the 5-weekold rats within 30 to 90 minutes; and  in 60% of the adult rats
                    within  30 to 65 minutes.  Thus, young (weanling) and older rats appear to
                    be more susceptible to a single oral dose of phenol than the 5-week-old
                    rats (Deichmann and Witherup, 1944).

-------
Phenol                                                                        April 1992
             Dermal/Ocular Effects

             •     A dermal LD50 of 850 rag/kg (95% confidence limits, 600 to 1,200 mg/kg)
                    for  rabbits (strain and sex not specified) exposed to phenol for a
                    maximum of 12 hours was reported by Flickinger (1976). The
                    percutaneous LD50 for phenol in adult female Alderly Park rats was 625
                    mL/kg (approximately 650 mg/kg) (Conning and Hayes,  1970).

             •     Upon application of 100 mg phenol into the eyes of male albino rabbits,
                    the conjunctivae became inflamed, the cornea became opaque and the
                    animals experienced marked discomfort (Flickinger, 1976). At 24 hours
                    after exposure, there was severe conjunctivitis, iritis, corneal opacification
                    occluding most of the iris and corneal ulceration over the entire corneal
                    surface.  By day 14, the exposed eyes exhibited keratoconus and pannus
                    formation.

             •     Ocular and nasal irritation was observed in rats subjected to inhalation of
                    aerosolized aqueous phenol at 900 mg/m3 for 8 hours (236 ppm phenol/8
                    hours) (Flickinger,  1976).

             Long-term  Exposure

             •     No  effects on liver, kidneys or any other organs were observed in 5- to
                    6-week-old  B6C3Fi mice (10/sex/dose) administered up to 10,000 ppm
                    phenol (2,000 mg/kg) in drinking water for 13 weeks (NCI, 1980). (Other
                    groups of mice [10/sex/dose]  received 100, 300, 1,000 or 3,000 ppm phenol
                    [20, 60, 200 or 600 mg/kg/day, respectively] in drinking water.) Dose
                    conversion was done according to Tatken and Lewis (1983).  Control
                    animals (10/sex) received only tapwater. Survival was 100% for all doses
                    and feed consumption among all treatment groups was comparable to
                    controls. However, water consumption dropped to  60 and 20% that of
                    controls for mice administered 3,000 and 10,000 ppm phenol,  respectively.
                    Gains in mean body weight  in all treated groups (except the high-dose
                    group) were comparable to or greater than those in controls.  In mice
                    receiving 10,000 ppm phenol, weight gain was 80% less than control
                    values for males and 33% less for females.  The greatest differences in
                    body weight occurred during the first half of the study. Microscopic
                    examination revealed  no compound-related pathology.  This study
                    identified a No-Observed-Adverse-Effect Level (NOAEL) of 360
                    mg/kg/day (adjusted for water consumption) for mice receiving phenol in
                    drinking water.

             •     In a 90-day oral exposure study (NCI, 1980),  groups of 10 male and 10
                    female Fischer 344 rats received 0. 100, 300, 1,000, 3,000, or 10,000 ppm

-------
Phenol                                                                         April  1992
                    phenol (0, 13, 38, 125. 375 or 1,350 mg/kg, respectively) in drinking water.
                    Dose conversion was done according to Tatken and Lewis (1983).
                    Survival was 100% for all groups.  Mean body weight gains for rats
                    receiving 10,000 ppm were depressed 16% for males and 26% for females
                    in comparison to controls. No other differences in body weights were
                    observed.  Rats rejected the high-dose water; males and females
                    consumed 67 and 50% less water,  respectively, than same-sex control rats.
                    No compound-related gross or microscopic alterations were detected at
                    necropsy.  A NOAEL of 375 mg/kg/day was established in this study.

              •     Body weight gain and water consumption were reduced in B6C3Ft mice
                    (50/sex/dose) administered either 2,500 or 5,000 ppm phenol (500 and
                    1,000 mg/kg/day, respectively) in drinking water for 103 weeks (NCI,
                    1980).  High- and low-dose groups' water consumption was depressed by
                    75 and 50 to 60%, respectively. Therefore, the actual delivered dose for
                    the high- and low-dose groups translates into 750 and 275 mg/kg/day
                    (Tatken and Lewis, 1983). Animals were 5 to 6 weeks  old at the start of
                    the study.  No other clinical or histopatho-logical  signs  related to the
                    consumption of phenol were observed.  Mortality also was not affected in
                    either treatment group; approximately 84 to 96% of the male and 80 to
                    84% of the female mice  (including controls) survived until the end of the
                    2-year study. This study identified a LOAEL of 275 mg/kg/day, based on
                    depressed body weight gain.

              •     Ingestion of 2,500 or 5,000 ppm phenol (313 and  625 mg/kg/day,
                    respectively, based on dose conversions done according to Tatken and
                    Lewis,  1983) for approximately 2 years was not toxic to Fischer  344 rats
                    (50/sex/dose) (NCI, 1980). Mean body weights of males and females from
                    the high-dose group were somewhat lower than those of control rats;
                    however, these were not reported  as being statistically significant.  Food
                    consumption among treated and control animals was comparable, but
                    water consumption by the low- and high-dose groups was reduced by 10
                    and 20%, respectively. Mortality was not affected by consumption of
                    either level of phenol; after 104 to 105 weeks of treatment,  44 to 60% of
                    the males  (including controls) and 74 to 78% of the females (including
                    controls) were still alive.  The inflammatory, degenerative and
                    hyperplastic lesions seen in treated animals were  similar in  number and
                    kind to those that occur  naturally in aged F344 rats. The highest dose
                    tested, 625 mg/kg/day, was identified as the NOAEL and was based on
                    the absence of any significant adverse health effects.
                                           10

-------
Phenol                                                                          April 1992
              Reproductive Effects

              •      In a series of two-, three-, and five-generation reproduction studies by
                     Heller and Pursell (1938), no deleterious effects on growth, reproduction,
                     and rearing of young were observed in rats (number and strain not
                     reported) consuming 15 to 5,000 ppm phenol (1.9 to 625 mg/kg/day based
                     on conversion factors of Tatken and Lewis, 1983) in drinking water.  In
                     contrast, growth was stunted in dams and pups at concentations of 7,000,
                     8,000, 10,000 and 12,000 ppm phenol; reproductive capabilities were
                     retarded and absent in animals ingesting the two highest doses,
                     respectively. Mortality was increased in both parents and offspring at
                     concentrations of 8,000 ppm and higher, despite a drop in water
                     consumption at these dose levels.  The authors also noted that mothers
                     receiving  10,000 ppm phenol neglected their young.  Although not
                     specifically designed to examine the reproductive or developmental
                     toxicity of phenol, this study indicated a lack of structural abnormalites
                     following oral exposure to the test material. The study also identified a
                     NOAEL of 625 mg/kg/day.

              Developmental Effects

              •      Price et al. (1986) tested the effects of oral administration of phenol in
                     CD-I mice. Pregnant mice  were given 70, 140 or 280 mg/kg on gestation
                     days 6 through 15; no other details concerning administration  of the test
                     material were given.  Treatments produced dose-related incidences of
                     maternal  tremors, ataxia,  lethargy, and irritability; the  incidence of these
                     symptoms was reported as being statistically significant at the  highest
                     dose, but no p level was given.  In addition to exhibiting clinical signs of
                     toxicity, 11% of the mice  (4/35) dosed with 280 mg/kg  died; fetuses from
                     this group had reduced body weights and an increased incidence (p level
                     not specified) of cleft palate. The highest NOAEL in this study, based on
                     the absence of systemic and fetal toxicity, was 140 mg/kg/day.

              •      Jones-Price and coworkers (1983)  reported that phenol administered by
                     gavage to groups of 20 to 22 pregnant  CD-I rats on gestation  days
                     6 through 15 produced no maternal effects at  30, 60 or 120 mg/kg/day.
                     An increase in resorptions was noted for ail phenol-treated groups when
                     compared with controls, but the effects were statistically significant only at
                     the 30- and 60-mg/kg/day  levels; thus, the rate of resorptions was  not
                     dose-related.  Fetal body  weights decreased with increasing dose levels,
                     and the values at 120 mg/kg were significantly lower (p < 0.001) than
                     controls.  No structural abnormalities were noted at any dose  level tested.
                     The highest NOAEL in this study was  60  mg/kg/day; the LOAEL, based
                     on highly significantly reduced fetal body weights, was 120  mg/kg/day.

                                            11

-------
Phenol                                                                         April 1992
              Mutagenicity

              •     The weight of evidence suggests that phenol is not mutagenic in
                    Salmonella  typhimurium (DeMarini et al., 1987; Gocke et al., 1981;
                    Haworth et al., 1983); or Drosophila melanogaster (Gocke et al., 1981).
                    However, rat liver S9-activated phenol was  mutagenic in L5178Y mouse
                    lymphoma cells (Tennant et al., 1987).

              •     Phenol was clastogenic in Asperigillus nidulans cultured Chinese hamster
                    ovary (CHO) cells (Tennant et al., 1987) and in vivo in male mice (Lowe
                    et al., 1987) but was negative in mouse bone marrow micronuclei assays
                    (Gocke et al., 1981; Lowe et al., 1987).

              •     Houk and DeMarini (1988) reported that phenol induced prophage
                    induction in Escherichia coli WP2r  Phenol did not cause DNA strand
                    breaks in mouse lymphoma L5178YS (Pellack-Walker and Blumer, 1986);
                    however, Garberg and Bolcsfoldi (1985) listed phenol as positive at this
                    endpoint in mouse lymphoma cells. In vitro, phenol increased the
                    frequency of sister chromatid exchange (SCE) in CHO cells (Tennant et
                    al.,  1987) and human lymphocytes (Morirnoto and Wolff, 1980); however,
                    in vivo mouse studies with phenol were negative for SCE induction (Lowe
                    et al., 1987).
             Carcinogenicity

             •      The most recent chronic oral toxicity study on phenol was performed by
                    the National Cancer Institute (NCI, 1980). A total of 200 B6C3Fi mice
                    (50/sex/dose) were administered 2,500 or 5,000 ppm (500 and 1,000
                    mg/kg/day, respectively, based on conversions of Tatken and Lewis, 1983)
                    phenol (98.47% pure) in drinking water for 103 weeks.  Animals were 3
                    to 4 weeks old at the start of the study.  Control mice (50/sex) received
                    tapwater only.  Histopathological examination revealed  no phenol-related
                    toxic or carcinogenic efffects in mice under the conditions of this
                    experiment.  In addition, statistical analysis (by Cochran-Armitage and
                    Fisher exact tests) indicated that no tumor at any site could be clearly
                    associated with the administration of phenol in this bioassay.

             •      In the chronic exposure study by NCI (1980), Fischer 344 rats
                    (50/sex/dose) were administered drinking water containing 2,500 or 5,000
                    ppm  phenol (98.47% pure). These doses are equivalent to 335 and 625
                    mg/kg/day, respectively (based on conversions of Tatken and Lewis,  1983).
                    Matched controls (50/sex) received tapwater.  Animals were treated  for
                                           12

-------
Phenol                                                                       • April 1992
                    103 weeks and observed for an additional 2 weeks. Under the conditions
                    of this bioassay, phenol was not carcinogenic in F344 rats of either sex.

              •     Salaman and Glendenning (1957) reported that "S" strain albino mice
                    (four groups of 20 mice each) showed strong tumor-promoting activity
                    after initiation with 0.3 mg dimethylbenzoylanthracene (DMBA) and
                    subsequent, repeated skin applications of 20% phenol (w/v in acetone) for
                    24 to 32 weeks. This concentration of phenol produced significant skin
                    damage and was weakly carcinogenic when applied alone.  A 5% solution
                    of phenol had a moderate promoting effect but was not carcinogenic
                    without prior initiation.

              •     In tumor-promotion studies conducted by Van Duuren et al. (1968),
                    20 female ICR/Ha Swiss mice were treated  first with  a single dose of
                    benzo(a)pyrene, then with applications of 3 mg phenol in acetone, three
                    times/week, for 1 year.  Four animals developed papillomas, and one had
                    developed a squamous carcinoma at the end of the treatment period. No
                    tumors developed in any of the control groups (i.e., animals treated only
                    with initiator, phenol, or acetone, and animals given no treatment).

              •     In subsequent experiments on cocarcinogenesis, 20 female ICR/Ha Swiss
                    mice received dermal applications of 5 pg benzo(a)pyrene and 3 mg
                    phenol (in acetone) three times/week for approximately 66 weeks (460
                    days) (Van Duuren et al., 1971).  At the end of the study, three mice had
                    developed papillomas, and one had developed a carcinoma. No tumors
                    were observed in the group receiving phenol alone.  However, among the
                    animals exposed only to benzo(a)pyrene, eight had developed papillomas,
                    and one had developed a squamous carcinoma. On the basis of these
                    data, it was concluded that phenol was not a cocarcinogen; rather, at the
                    dose used in Swiss mice, phenol slightly inhibited the tumorgenic response
                    normally exhibited by benzo(a)pyrene.  The partial inhibitory effect of
                    phenol on the carcinogenic activity of benzo(a)pyrene was confirmed by
                    Van Duuren and Goldschmidt (1976) and Van Duuren et al. (1973).
                    However, these latter studies did not evaluate the effects of solvents and,
                    in some cases, of pretreatment with either DMBA or benzo(a)pyrene.

              In Vitro Cvtotoxicitv

              •     Phenol was found to be cytotoxic to BF-2 cells derived from blue gill sun
                    fish (Babich and Borenfreund, 1987), rat lung epithelial cells (Li, 1986),
                    V79/4 Chinese hamster lung fibroblasts (Hunt et al.,  1987), baby hamster
                    kidney fibroblasts (Tyas, 1978), and human amnion epithelial cells
                    (Eichhorn et al., 1987).
                                           13

-------
Phenol                                                                        April 1992
V.     QUANTIFICATION OF TOXICOLOGICAL EFFECTS

       Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
(up to 7 years) and Lifetime exposures if adequate data are available that identify a sensitive
noncarcinogenic end point of toxicity.  The HAs for noncarcinogenic toxicants are derived using
the following formula:
          (NOAEL or LOAEL) x fBW>           _  ,
            (UF) (_ IVday)             = _ mg/L (
where:
      NOAEL or LOAEL =   No- or Lowest-Observed-Adverse-Effect Level (in mg/kg
                              bw/day).

                     BW =   assumed body weight of a child (10 kg) or an adult (70 kg).

                     UF =   uncertainty factor, (10, 100, 1,000 or 10,000, in accordance with
                              EPA or NAS/OW guidelines.

                	L/day =   assumed daily water consumption of a child (1 L/day) or an
                              adult (2 L/day).
      One-day Health Advisory

      No suitable information was found in the available literature for determining the
One-day Health Advisory (HA) for phenol. The modified Drinking Water Equivalent Level
(DWEL) of 6,000 ngfL for a 10-kg child, calculated below, is recommended for use as a
conservative estimate for a one-day exposure.

      Ten-day Health Advisory

      No suitable information was found in the available literature for determining the
Ten-day HA for phenol.  The modified DWEL of 6,000 /xg/L for a 10-kg child, calculated below,
is recommended for use as a conservative estimate for a 10-day exposure.

      Longer-term  Health Advisory  ,

      Two adequate subchronic oral exposure studies with phenol have been conducted (NCI,
1980).  These 90-day drinking water studies identified NOAELs of 360 and 375 mg/kg/day for
mice and rats, respectively, based on the absence of systemic toxicity. However, since exposure
of the fetus is determined by maternal exposure, and since significant fetal effects without


                                           14

-------
Phenol                                                                         April 1992
maternal toxicity have been observed in rats given 120 mg phenol/kg/day (on gestation days 6
through 15) (Jones-Price et al., 1983), this dose level must be considered in calculating the
Longer-term HA.  It is, therefore, recommended that the modified DWEL of 6,000 /xg/L for a
10-kg child be  used as a conservative estimate for Longer-term exposure for a child.


     .,  .... . r,,,/r:T      (0.6 mg/kg/dav) (10 kg)           ,   n  ,,-fw.    n^
     Modihed DWEL =    J	  (\\IA  \	          = " m&** (ฐ'000 Mg/L)

where:

                     0.6 mg/kg/day  =     RfD  (see below).

                     10 kg         =     assumed body weight of a child.

                     1 L/day        =     assumed daily water consumption of a child.

       The DWEL of 20,000 /xg/L, calculated below, should be used as a conservative estimate
for the Longer-term HA value for the 70-kg adult.

       Lifetime Health Advisory

       The Lifetime HA represents that portion of an individual's total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure. The Lifetime HA is derived in a three-step process. Step 1
determines the Reference Dose (RfD), formerly called the Acceptable Daily Intake (ADI). The
RfD is an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily
exposure to the human population (including sensitive subgroups) that is likely to be without
appreciable  risk of deleterious health effects during a lifetime, and  is derived from the NOAEL
(or LOAEL), identified from a chronic (or subchronic) study, divided by an uncertainty
factor(s).  From the RfD, a Drinking Water Equivalent Level (DWEL) can be determined
(Step 2). A DWEL is a medium-specific (i.e., drinking water) lifetime exposure level, assuming
100% exposure from that medium, at which adverse, noncarcinogenic health effects would not
be expected to occur.  The DWEL is derived from the multiplication of the RfD by the assumed
body weight of an adult and divided  by the assumed daily water consumption of an adult.   The
Lifetime HA in drinking water alone is determined in Step 3 by factoring in other sources of
exposure, the relative source contribution (RSC). The RSC from drinking water is based  on
actual exposure data or, if data are not available, a value of 20% is assumed.

       If the contaminant is classified as a  known, probable, or possible human carcinogen,
according to the Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986), then
caution must be exercised in making a decision on how to  deal with possible lifetime exposure .
to this substance.  For human (A) or probable (B) human carcinogens, a Lifetime HA is not
recommended.  For possible (C) human carcinogens, an additional 10-fold safety factor is used
in the calculation of the Lifetime HA. The risk manager must balance this assessment of

                                            15

-------
 Phenol                                                                          April 19921
carcinogenic potential and the quality of the data against the likelihood of occurrence and
significance of health effects related to noncarcinogenic endpoints of toxicity. To assist the risk
manager in this  process,  drinking water concentrations associated with estimated excess lifetime
cancer risks over the range of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult drinking 2  L of
water/day are provided in the Evaluation of Carcinogenic Potential section.

       The developmental toxicity study in  rats by Jones-Price et al. (1983) has been selected to
serve as the basis for the Lifetime HA because it is an adequately conducted study in which the
LOAEL for fetal toxicity, 120 mg/kg/day, was lower than the NOAELs from other adequate
subchronic, chronic and reproductive/developmental studies (i.e., 140 to 625 mg/kg/day for rats
and mice, respectively) (Heller and Pursell, 1938; NCI, 1980; Price  et al., 1986).  These
subchronic and chronic studies also suggest that the rat is more  sensitive to oral exposure to
phenol than mice (Jones-Price et  al., 1983; NCI,  1980; Price et al., 1986). The epidemiological
study by Baker et al. (1978) was considered inadequate for estimation of a NOAEL because the
authors' exposure estimate was not supported by the data.

       In the study by Jones-Price et al. (1983), groups of 20 to 22  female CD-I  rats were given,
by gavage,  30, 60 or 120 mg phenol/kg/day on gestation days 6 through 15.  No maternal effects
were observed at any dose, but reductions in fetal body weights  were dose-related and
significantly lower (p < 0.001) than control values at the 120-mg/kg/day dose level. No
structural abnormalities were noted at any dose level.  Results of this study suggest a NOAEL
and LOAEL for developmental effects of 60 and 120 mg/kg/day, respectively.

       Using the Jones-Price et al. (1983) study,  the Lifetime HA is derived as follows:

Step 1:  Determination of the RfD


             RfD =     J—nno^  —     = ^ ras/'ic8/day

where:             .

                      60 mg/kg/day        =     NOAEL, based on the absence of
                                                developmental  effects in fetuses of rats
                                                exposed to phenol by gavage on gestation
                                                days 6 through 15 (Jones-Price et al., 1983).

                                100        =     uncertainty  factor, chosen in accordance
                                                with EPA's  proposed developmental toxicity
                                                guidelines.
                                            16

-------
Phenol                                          .              -             April 1992
Step 2:  Determination of the DWEL
   DWEL=  (0'6 "gffiPV70 kg)      = 21 mg^L (rounded to 20,000
where:

                    0.6 mg/kg/day       =      verified RfD.

                    70 kg               =      assumed body weight of an adult.

                    2 L/day             =      assumed daily water consumption of an
                                              adult.

Step 3: Determination of the Lifetime HA

                     Lifetime HA = (21 mg/L) (20%) = 4.2 mg/L (4,000 Mg/L)

where:

                      21 mg/kg/day       =      DWEL.

                             20%       =      assumed RSC from water.


      Evaluation of Carcinogenic Potential

      •      IARC has not evaluated the carcinogenic potential of phenol.

      •      Applying the criteria described in EPA's guidelines for assessment  of
             carcinogenic risk (U.S. EPA, 1986), phenol may be classified in Group D: not
             classifiable.  This category is for agents with inadequate animal evidence of
             carcinogenicity.


VI.   OTHER CRITERIA. GUIDANCE. AND STANDARDS

      •      The American Conference of Governmental Industrial Hygienists recommends a
             dermal Time-Weighted  Average — Threshold Limit Value (TWA-TLV) of 5 ppm
             phenol, which  is  equivalent to an inhalation exposure of 19 mg phenol/m3
             (AGGIH, 1988).
                                          17

-------
Phenol                                                                        April 1992'
       •     The National Institute of Occupational Safety and Health/Occupational Safety
             and Health Administration (NIOSH/OSHA) standard for phenol exposure is 20
             mg/m3 (approximately 5 ppm) in air, averaged over an 8-hour work shift
             (NIOSH/OSHA, 1981). NIOSH also recommends a ceiling of 60 mg phenol/m3
             for a 15-minute exposure (NIOSH/OSHA, 1981).

       •     The Public Health Service, U.S. Department of Health, Education and Welfare,
             established a standard of 1 /zg/L phenols in drinking water.  This was based on
             the taste and odor threshold as a result of chlorination of phenols (U.S. DHEW,
             1962).

VII.    ANALYTICAL METHODS

       •     Depending on the detection level desired, several methods are available for  the
             determination of phenol.

       •     The classical methods for quantifying phenol, EPA Methods 420.1 and 420.2,
             determine total phenols.  To determine specific phenols, EPA Methods 604 or
             625 should be used (U.S. EPA, 1984a,b). In these procedures the typical 1L of
             sample is extracted with methylene chloride, and the extract reduced to ImL.
             Analysis in Method 604 is by Flame lonization-Gas Chromatography (FID/GC).
             Method 625 is a Gas Chromatographic-Mass  Spectrometry Method for
             determining semi-volatile base/neutrals and acids, phenol being determined in the
             acid fraction. The method calls for using dueterated phenol (D-5 phenol) as a
             surrogate.  The detection limit for phenol by Method 604 is 0.14 /ig/L and by
             Method 625 it is 1.5 /xg/L.  Method 604 has an optional derivatization procedure
             utilizing pentafluorobenzylbromide. The resulting derivatives of many phenols
             have much lower detection levels utilizing Electron Capture-Gas Chromatography
             (EC/GC).
VIII.  TREATMENT TECHNOLOGIES

      •     Available data indicate that granular activated carbon (GAC) adsorption will
             significantly reduce phenol levels in drinking water.

      •     GAC adsorption isotherms and pilot plant tests were conducted to evaluate GAC
             performance in removing phenol (Royer, 1980).  Freundlich isotherm constants
             for phenol were determined  to be 133 mg/L ""po,"'  for K and 0.299 for 1/n. The
             pilot plant consisted of four GAC columns operated in series, each 3 inches in
             diameter  and packed with 40 inches of carbon.  A contact time of 18.7 minutes
             produced a 95% phenol reduction from an influent concentration of 938 mg/L.
             The carbon usage  rate was 44.5 Ib GAC'1.000 gal.
                                           18

-------
Phenol                                                                        April 1992
             Zogorski et al. (1976) studied the kinetics of the adsorption phenomenon for
             phenols in a batch reactor. The GAC initial concentration was kept constant at
             500 mg/L. The adsorption of phenolic compounds by GAC was extremely rapid.
             Approximately 60 to 80% of the adsorption occurred within the first hour of
             contact.

             Kim et al. (1986) reported a maximum phenol adsorption capacity of 0.13 g
             phenol/g carbon at an influent concentration of 40 mg/L by two 2-stage pilot
             scale, anaerobic GAC reactor units operated in series.  Each stage consisted of
             two reactors:  a fixed-bed (once- through mode) containing 0.5  inch Rasching
             rings and a fluidized GAC bed with recirculation.  Each reactor was designed for
             an empty bed contact time (EBCT) of 24 hours. The performance of this system
             was evaluated in terms of phenol removal by adsorption,  biomass  production and
             biogas production.

             Data were not found for the removal of phenol from drinking water by aeration.
             However, phenol may be slightly amenable to removal by aeration due  to its
             moderate Henry's Coefficient value.
                                           19

-------
Phenol                                                                         April  1992
IX.     REFERENCES

Aaberg, R.L., R.A. Peloquin, D.L. Strenge and PJ. Mellinger.  1983. Aquatic-pathways model
        to predict the fate of phenolic compounds.  Publication no.  DE 83011074.  Richland,
        WA:  U.S. Department of Energy.

ACGIH.  1988. American Conference of Governmental Industrial Hygienists. Threshold limit
        values and biological exposure indices for 1987-88.   Cincinnati, OH:  ACGIH, p. 29.

Babich, H. and E. Borenfreund.  1987. In vitro cytotoxicity of organic pollutants to bluegill
        sunfish (BF-2) cells. Environ. Res. 42:229-237.

Baker, EX., PJ. Landrigan, P.E. Bertozzi,  P.H. Field, BJ. Basteyns and H.G. Skinner.  1978.
        Phenol poisoning due to contaminated drinking water. Arch. Environ. Health 33:89-94.

Baker, M.D. and C.I. Mayfield.  1980.  Microbial. and nonbiological decomposition of
        chlorophenols and phenol in soil.  Water Air Soil Pollut. 13:411-424.

Branson, D.R. 1978. Predicting the fate of chemicals in the aquatic environment from
        laboratory data. ASTM STP G57. Philadelphia, PA:  American Society for Testing
        and Materials, pp. 55-70.

Brodzinsky, R. and H.B. Singh.  1982.  Volatile organic chemicals in the atmosphere:  an
        assessment of available data. Contract no. 68-02-452.  Research Triangle Park, NC:
        U.S. Environmental Protection Agency.

Bruce, R.M., J. Santodonato and M.W. Neal.  1987.  Summary  review of the health effects
        associated with phenol. Toricol. Ind. Health 3:535-568.

Callahan, M.A., M.W. Slimak and N.W. Gabel. 1979. Water related environmental fate of 129
        priority pollutants.  EPA 440/4-79-029b. Washington, DC:  U.S. Environmental
        Protection Agency.

Capel, I.D., M.R. French, P. Millburn, R.L. Smith and R.T. Williams.  1972. The fate of
        [l4C]phenol in various species. Xenobiotica 2:25-34.

Carter, W.P.L., A.M. Winer and J.N. Pitts Jr.  1981.  Major atmospheric sink for phenol and the
        cresols.  Reaction with the nitrate  radical.  Environ. Sci.  Technol. 15:829-831.

Conning, D.M. and MJ. Hayes.  1970. The dermal toxicity of phenol: An investigation of the
        most effective first-aid measures. Br. J. Ind. Med. 27:155-159.
                                           20

-------
Phenol                                       '                                  April 1992
Cronn, D.R., RJ. Charlson, R.L. Knights, A.L. Crittenden and B.R. Appel. 1977. A survey of
        the molecular nature of primary and secondary components of particles in urban air by
        high-resolution mass  spectrometry.  Atmos. Environ. 11:929-937.

Deichmann, W.B.  1944.  Phenol studies.  V. The distribution,  detoxification, and excretion of
        phenol in the mammalian body. Arch. Biochem. 3:345-355.

Deichmann, W.B., T. Miller and J.B. Roberts.  1950. Local and systemic effects following
        application of dilute solutions of phenol in water and in camphor-liquid petrolatum on
        the skin of animals.  Arch. Ind. Hyg. 2:454-461.

Deichmann, W.B. and S. Witherup.  1944.  Phenol studies.  VI.  The acute and comparative
        toxicity of phenol and o-, m- and p-cresols for experimental animals.  J. Pharmacol.
        Exp. Ther.  80:233-240.

Delfino, J.J. and D J. Dube.  1976.  Persistent contamination of ground water by phenol.  J.
        Environ. Sci. Health Al 1:345-355.

DeMarini, DM., T.P. Inmon, I.E. Simmons, E. Berraan, T.C. Pasley, S.H. Warren and R.W.
        Williams.  1987.  Mutagenicity in Salmonella of hazardous wastes and urine from rats
        fed these wastes. Mutat. Res. 189:205-216.

Ehrlich, G.G.  1982. Degradation of phenolic contaminants in ground water by anaerobic
        bacteria: St. Louis Park, Minnesota.  Ground Water Magazine, vol. 20.

Eichhorn, U., R. Klocking, H. Schweizer and H.-P. Klocking. 1987.  Cell membrane toxicity
        detected with the chromium-51 release test, mechanisms and models in toxicology.
        Arch. Toxicol.(Suppl. ll):334-337.

Elder, V.A., B.L. Proctor and R.A. Kites. 1981. Organic compounds found near dump sites in
        Niagara Falls, NY. Environ. Sci. Technol. 15:1237-1243.

Finlayson-Pitts, BJ. and J.N. Pitts Jr. 1986. Atomospheric chemistry: fundamentals and
        experimental techniques.  New York, NY: John Wiley & Sons, pp. 410, 506.

Flickinger, C.W. 1976. The benzenediols: catechol, resorcinol and  hydroquinone—A review of
        the industrial toxicology  and current industrial exposure limits. Am. Ind. Hyg. Assoc. J.
        37:596-606.

Garberg, P.  and G.  Bolcsfoldi. 1985.  Evaluation of a genotoxicity test measuring DNA strand
        breaks in mouse  lymphoma cells by alkaline unwinding and hydroxylapatite
        chromatography. Environ. Mutagen. 7(SupplJ):73.

-------
Phenol                                                                         April 1992
Gocke,  E., M.-T. King, K. Eckhart and D. Wild.  1981.  Mutagenicity of cosmetic ingredients
        licensed by the European Communities.  Mutat. Res. 90:91-109.

Gross, B.G.  1984. Cardiac arrhythmias during phenol face peeling. Plast. Reconstr. Surg.
        73:590-594.

Haworth,  S., T. Lawler, K. Mortelmans, W. Speck and E. Zeiger.  1983. Salmonella mutagenicity
        test results for 250 chemicals. Environ. Mutagen. 5(Suppl. 1):3-142.

Heller, V.G. and L. Pursell.  1938.  Phenol-contaminated waters and their physiological action.
        J. Pharmacol. Exp. Ther. 63:99-107.

Hendry, D.G. and R.A. Kenley. 1979. Atmospheric reaction products of organic compounds.
        Document no. EPA-560/12-79-001.  Washington, DC: U.S. Environmental Protection
        Agency, pp. 20, 46.

Houk, V.S. and D.M. DeMarini.  1988. Use of the microscreen phage-induction assay to assess
        the genotoxicity of 14 hazardous industrial wastes. Environ. Mutagen. 11:13-29.

Hunt, S.M., C. Chrzanowska, C.R. Barnett, H.N. Brand and T. K. Fawell. 1987.  A comparison
        of in vitro cytotoxicity assays and their application to water samples: Altern. Lab.
        Animals 15:20-29.

Jones-Price, C.N., T.A. Ledoux, J.R. Reel, P.W. Fischer, L. Langhoff-Paschke and M.C. Marr.
        1983. Teratological evaluation of phenol (CAS no. 108-95-2) in CD rats.  RTI project
        nos. 31U-1287 and 31U-2312. Contract nos. NO1-ES-6-2127 and PR 259231. Research
        Triangle Park, NC: National Institute of Environmental Health Sciences.

Jungclaus, G.A., V. Lopez-Avila and R.A. Hites.  1978.  Organic compounds in an industrial
        waste water:  a case study of their environmental impact.  Environ. Sci. Technol.
        12:88-96.

Kim, B.R., E.S.K. Chian,  W.H. Gross and S. Cheng.  1986.  Adsorption, desorption,  and
        bioregeneration in an anaerobic, granular activated carbon reactor for the removal of
        phenol. J. Water Pollut. Contr. Fed.  58(1):35^40.

Kincannon, D.F., E.L. Stover, V. Nichols and D. Medley. 1983. Removal  mechanisms for toxic
        priority pollutants. J. Water Pollut. Contr. Fed. 55:157-163.

Kobayashi, K., H. Akitake and K. Manabe.  1979.  Relationship between toxicity and
        accumulation of various chlorophenols in goldfish. Bull. Jap. Soc. Sci. Fish. 45:173-175.

-------
Phenol                                                                         April 1992
Konaseuich, D., W. Traversy and H. Zar.  1978.  Status report on organic and heavy metal
        contaminants in the Lake Erie, Michigan, Huron and Superior Basins.  Appendix E.
        Windsor, Ontario:  Great Lakes Water Quality Board.

Kuwata. K., M. Uebori and Y. Yamazaki.  1980.  Determination of phenol in polluted air as
        p-nitrobenzeneazophenol derivative  by reversed  phase high performance liquid
        chromatography. Anal. Chem. 52:857-860.

Lee, R.F. and C. Ryan.  1979. Microbial degradation of organochlorine compounds in estuarine
        waters  and sediments.  EPA-600/9-79-012. Washington, DC: U.S. Environmental
        Protection Agency, pp. 443-450, 462-476.

Leonardos, G.,  D. Kendall and N. Barnard.  1969. Odor threshold determinations of 53 odorant
        chemicals. J. Air Pollut. Control Assoc.  19:91-95.

Li, A.P. 1986.  An in vitro lung epithelial cell system for evaluating  the potential toxicity of
        inhalable materials. Food Chem.  Toxicol. 24:527-534.

Lowe, K.W., C.J. Holbrook, S.L. Linkous and M.R. Roberts.  1987.  Preliminary comparison of
        three cytogenetic assays for genotoxicity in mouse bone marrow cells. Environ.
        Mutagen. 9(Suppl. 8):63.

Morimoto,  K. and S. Wolff. 1980. Increase  of sister chromatid exchanges and perturbations of
        cell division kinetics in human lymphocytes of benzene metabolites.  Cancer Res.
        40:1189-1193.

NCI.  1980. National Cancer Institute.  Bioassay for phenol for possible carcinogenicity.
        Research Triangle Park, NC:  National Toxicology Program.

NIOSH/OSHA.  1981. National Institute for Occupational Safety and Health/Occupational
        Safety and Health Administration. Occupational health guidelines for chemical
        hazards.  Washington, DC:  U.S.  Department of Health and Human Services.

Pellack-Walker, P. and J.L. Blumer.  1986. DNA damage in L5178YS cells following exposure
        to  benzene metabolites.  Molec. Pharmacol. 30:42-47.

Petrasek, A.C.,  IJ. Kugelman, B.M. Austern, T.A. Pressley,  LA.  Winslow and R.H. Wise.  1983,
        Fate of toxic organic compounds in  wastewater treatment plants.  J. Water Pollut.
        Contr.  Fed. 55:1286-1296.

Pibtrowski,  T.K. 1971. Evaluation of exposure to phenol: Absorption of phenol vapor in the
        lungs and through the skin and excretion of phenol  in the urine.  Brit. T. Ind. Med.
      .  28:172-178.

-------
Phenol                                                                          April  1992
Price, C.J., T.A. Ledoux, J.R. Reed, P.W. Fisher, L.L. Paschke, M.C. Marr and C.A. KimmeK
         1986. Teratological evaluation of phenol in rats and mice. Teratology 33:92C-93C.

Rees, J.F. and J.W. King. 1981. The dynamics of anaerobic phenol biodegradation in lower
        greensand. Technol. Bioechnol. 31:306-310.

Richards, D J. and W.K. Shieh. 1986. Biological fate of organic priority pollutants in the
        aquatic environment.  Water Res. 20:1077-1090.
Rizet, M., J. Mallevialle and J.-C. Cournante. 1977. Pilot plant investigation of the evolution of
        various pollutants during artificial recharge of an aquifer by a basin. Prog. Water
        Technol. 9:203-215.

Royer, M.D.  1980. Pilot-plant and laboratory prediction of the per-formance of a mobile,
        full-scale granular activated carbon system for treating hazardous materials
        spill-contaminated water.  Proceedings of the 1980 National Conference on Control of
        Hazardous Material Spills.

Salaman, M.H.  and O.M. Glendenning.  1957. Tumor promotion in mouse skin by sclerosing
        agents.  Br. J. Cancer 11:434-444.

Sheldon, L.S. and R.A. Kites.  1979.  Sources and movement of organic chemicals  in the
        Delaware River. Environ. Sci. Technol.  13:574-579.

Tabak, H., S.A. Quave, C.I. Mashni and E.F. Barth. 1981. Biodegradability studies with
        organic priority pollutant compounds. J. Water Pollut. Contr. Fed.  53:1503.

Tatken, R.L. and RJ. Lewis, Sr., eds.  1983. Registry of toxic effects of chemical substances,
        vol. 3.  Cincinnati, OH:  NIOSH.

Tennant, R.W., B.H. Margolin, M.D. Shelby, E. Zeiger, T.K. Haseman, T. Spalding, W.
        Caspary, M. Resnick,  S. Stasiewicz,  B. Anderson and R. Minor. 1987.  Prediction of
        chemical carcinogenicity in rodents from in vitro genetic toxicity assays.  Science
        236:933-941.

Truppman, E.S. and J.D. EUenby.  1979.  Major electrocardiographic changes during chemical
        face peeling.  Plast. Reconstr. Surg. 63:44-48.

Tyas, MJ.  1978.  A histochemical study of the effect of phenol on the mitochondria and
        lysosomes of cultured cells. Histochem.  J. 10:333-342.

U.S. DHEW. 1962.  U.S. Dept. of Health,  Education and Welfare.  Public Health Service
        drinking water standards.  Rockville, MD: Public Health Service, p. 51.

                                           24

-------
Phenol                                                                        April 1992
U.S. EPA.  1980.  U.S. Environmental Protection Agency.  Ambient water quality criteria
        document for phenol.  EPA 440/5-80-066.  NTIS PB81- 117772. Washington, DC:  U.S.
        Environmental  Protection Agency.

U.S. EPA.  1984a.  U.S. Environmental Protection Agency.  EPA Method 604 - Phenols. 40
        CFR Part 136.  October 26.

U.S. EPA.  1984b.  U.S. Environmental Protection Agency.  EPA Method 625 - Base/neutrals
        and acids. 40 CFR Part 136. October 26.

U.S. EPA.  1986.  U.S. Environmental Protection Agency.  Guidelines for carcinogen risk
        assessment.  Fed. Reg. 51(185):33991-34003. September 24.

Van Duuren, B.L. and B.M. Goldschmidt. 1976. Cocarcinogenic and tumor-promoting agent in
        tobacco carcinogenesis.  J. Natl. Cancer Inst. 56:1,237-1,242.

Van Duuren, B.L., B.M. Goldschmidt, C. Katz, S. Melchionne and A. Sivak. 1971.
        Cocarcinogenesis studies on mouse skin and inhibition of tumor production.  J. Natl.
        Cancer Inst.  46:1,039-1,044.

Van Duuren, B.L., C. Katz and B.M. Goldschmidt.  1973. Cocarcinogenic agents in tobacco
        carcinogenesis.  J. Natl. Cancer Inst. 51:703-705.

Van Duuren, B.L., A. Sivak, L. Langseth, B.M. Goldschmidt and A. Segal. 1968. Initiators and
        promoters in tobacco carcinogenesis.   In: Wynder, E.  and D. Hoffman, eds. Toward a
        less harmful  cigarette. NCI Monograph 28.  Bethesda, MD:  National Cancer Institute,
        pp.  173-180.

Wajon, I.E., D.H. Rosenblatt and E.P. Burrows.  1982.  Oxidation of phenol and hydroquinone
        by chlorine dioxide.  Environ. Sci. Technol.  16:396-  402.

Walker, N. 1954.  Preliminary observations on the decomposition of chlorophenols in soil.
        Plant Soil 5:194-204.

Windholz, M., S. Budavari, R.F. Blumetti and  E.S. Otterbein,  eds. 1983. The Merck index—An
        encyclopedia of chemicals, drugs, and  biotogicals, 10th ed. Rahway, NJ: Merck and
        Company, Inc., p. 1,043.

Zogorski, J.S., S.D. Faust  and J.S. Haas.  1976. The kinetics of adsorption of phenols by
        granular activated carbon.  J. of Colloid and Interface Science 55(2):329-341.

-------
  EPA 0553

  RX000027511
                                                                                April 1992
                                         SILVER
                              Drinking Water Health Advisory
                                      Office of Water
                           U.S. Environmental Protection Agency
I.      INTRODUCTION
       The Health Advisory Program, sponsored by the Office of Water (OW), provides
information on the health effects, analytical methodology and treatment technology that would
be useful in dealing with the contamination of drinking water.  Health Advisories (HAs)
describe nonregulatory concentrations of drinking water contaminants at which adverse health
effects would not be anticipated to occur over specific exposure durations. Health Advisories
contain a margin of safety to protect sensitive members of the population.

       Health Advisories serve as informal technical guidance to assist Federal, State, and local
officials responsible for protecting the public health when  emergency spills or contamination
situations occur. They are not to be construed as legally enforceable Federal standards. The
HAs are subject to change as new information becomes available.

       HAs are developed for One-day, Ten-day, Longer-term (approximately 7 years,  or 10%
of an individual's lifetime) and Lifetime exposures based on data describing noncarcinogenic
end points of toxicity.  For those substances that are known  or probable human carcinogens,
according to the Agency classification scheme (Group A or B), Lifetime HAs are not
recommended.  For substances with a carcinogenic potential, chemical concentration values are
correlated with carcinogenic risk estimates by employing a cancer potency (unit risk) value
together with assumptions for lifelong exposure and the ingestion of water.  The cancer unit risk
is usually derived from a linearized multistage model with  95% upper confidence limits
providing a low-dose estimate of cancer risk.  The cancer risk is characterized as being  an upper
limit estimate, that is,  the true risk to humans, while not identifiable, is not likely to exceed the
upper limit estimate and in fact may be lower. While alternative risk modeling approaches  may
be presented, for example one-hit, Weibull, logit, or probit, the range of risks described by using
any of these models has little biological significance unless data can be used to support the
selection of one model over another. In the interest of consistency of approach and in
providing an upper-bound on the potential carcinogenic risk, the Agency recommends  using the
linearized multistage model.

-------
Silver                                                                            April 1992



II.     GENERAL INFORMATION AND PROPERTIES

       CAS No.  7440-22-4

       Structural Formula

       •      Not applicable.

       Synonyms

       •      Argentum; argentum crede; C • I • 77820; collargol; L-3; shell silver; silber
              (German); silver atom; silver colloidal (Weast,  1977; Stokinger, 1981).

       Uses

       •      Silver is used in coinage; the manufacture of tableware, jewelry and ornaments;
              electroplating; the manufacture of solder, brazing alloys and high-capacity silver-
              zinc and silver-cadmium batteries; the processing of food and beverages; inks and
              dyes; etching of ivory and electrical contacts. Silver is also used as a drinking
              water disinfectant and as a catalyst in hydrogenation and oxidation.  It is
              extensively used in photographic processing, in  mirror production and in dental
              alloys. Silver salts are used in the treatment of warts and burns.

       •      In addition to metallic silver, compounds of silver used are silver oxide, silver
              acetate, silver bromide, silver chloride, silver cyanide, silver iodate, silver iodide,
              silver nitrate and silver sulfate (Stokinger, 1981). The use of dilute silver nitrate
              solution (AgNO3) for prophylaxis against ophthalmia neonatorum is still a
              routine requirement in some States (Harvey, 1985).  Silver oxide is used in the
              purification of drinking water because of its toxicity to bacteria and other
              potentially pathogenic microorganisms (Budavari et al., 1989).

       Properties (Weast,  1977; Hawley, 1981)

              Chemical Formula          Ag
              Molecular Weight           107.868
              Physical State               Soft, ductile, malleable, lustrous white metal,
                                         face-centered cubic structure
              Boiling Point                2.212ฐC
              Melting Point               960i5ฐC
              Density                     10.50 g/mL at 20ฐC
              Vapor Pressure              —

-------
Silver                                                                            April 1992
              Water Solubility             Silver is insoluble in water and soluble in fused
                                          alkali hydroxides, peroxides and cyanides, hot
                                          sulfuric and nitric acids.  Most silver salts have
                                          limited solubility in water; the solubilities of silver
                                          nitrate and silver acetate are  122 and 1.02 g/100 mL
                                          water, respectively.
              Specific Gravity             10.5 at 20ฐC
       Occurrence

       •      Silver is a naturally occurring rare element, with concentrations averaging 0.1
              /ig/g in the Earth's crust (Weast, 1977).  Silver has a low solubility in water  (0.1
              to 10 mg/L, depending on the pH and the chloride concentration) (Hem, 1970).
              The average concentration of silver  in seawater is 0.15 to 0.3 ng/kg (Weast, 1977).
              Drinking water contains extremely low concentrations of silver.  It is reported
              that the concentration of silver in 380 samples of finished drinking water from
              the United States ranged from less than 1 up to 5 /ig/L (Nordberg and
              Gerhardsson, 1988).  The atmospheric concentration of silver over Southern
              California is of the order of 2 ng/m3 (Bruland et al., 1974).  It has been estimated
              that the emission of silver iodide crystals during cloud seeding result in a silver
              concentration in the air of about 0.1 ng/m3 (Nordberg and Gerhardsson,
              1988).  Soil concentrations of silver vary greatly by geological location. Granite
              igneous rocks in Nevada  contain up to 50 mg/kg silver, and coal fly ash may
              contain up to 15 mg/kg silver  (Nordberg and Gerhardsson, 1988).

       •      Trace amounts  of silver are found in natural and finished waters originating from
              natural sources and from industrial waste. Data from 1,577 samples of well and
              surface waters from 130 points in the United States showed detectable silver
              concentrations (0.1 /ig/L or greater) in only 104 samples.  The concentrations
              ranged from 0.1 to 38 /ig/L: the median  was 2.6 /ig/L (Kopp and Kroner, 1967).
              The highest concentrations were noted in the St. Lawrence and Colorado Rivers
              (Durum and Hafty, 1961).

       •      Chemical  analysis of finished  water from public supplies of the 100 largest U.S.
              cities revealed trace quantities of silver as high as 7 /ig/L, with a median
              concentration of 2.3 /ig/L (Durfor and Becker, 1964).  In another survey of
              finished water,  silver was found in 6.1%  of 380 samples, with concentrations
              ranging from 0.3 to 5 /ig/L (mean was 2.2 /ig/L) (Kopp and Kroner,  1967).

       Environmental Fate

       '•      Adsorption appears to be the dominant  process leading to partitioning of silver
              into the sediments. Silver concentrations in lake sediments and nearby soils were

-------
Silver                                                                            April 1992
              found to correlate with organic content. Silver concentrations in sediments were
              100 times the concentration of overlying waters (Freeman, 1977). The affinity of
              silver to three clays decreased in the following order:  montmorillonite > illite >
              kaolinite.  However, silver has a stronger affinity to MnO, and Fe(OH)3 (Kharkar
              et al., 1968).  Sediments from Colorado streams showed a high correlation (r =
              0.913) between silver and manganese content.  Iron oxides also adsorb silver but
              play a secondary role to MnO2 (Chao and Anderson, 1974; Dyck, 1968).  Almost
              all of the silver adsorbed to MhO, and  other sediments was released on contact
              with seawater, indicating that silver  transported in the paniculate phase of the
              water column may be released in  the marine or estuarine environment.

              Several studies have shown that silver bioaccumulates in aquatic organisms at
              relatively low concentrations because most of its compounds are sparingly soluble
              in water.  Bioconcentration ratios ranging  from 10 to 100 were reported  for
              largemouth bass and bluegill (Coleman and Cearley, 1974; Cearley, 1971).
              Activated sludge microbes bioaccumulate silver at about 100 times the
              concentration of silver present in solution (Chin, 1973).  Freeman (1977)
              reported fluctuations in silver concentrations in plankton that were closely
              correlated to changes in lake water concentration, while benthic species showed
              fluctuations more closely correlated  to concentrations in the sediments.
              Soluble silver compounds that ionize readily are quite toxic to fish. The
              ranges from 0.2 mg Ag/L for young eels to 0.003 mg Ag/L for salmon fry
              (Terhaar et al., 1972).  Silver complex, as in the thiosulfate that occurs in
              photographic processing effluents, however, is at most, only slightly toxic to fish,
              the 96-hr LCj,, is greater than 250 mg Ag (in the thiosulfate complex) (Bard, et
              al., 1976).
III.    PHARMACOKINETICS

       Absorption
              Dequidt et al. (1974) administered colloidal silver to Wistar rats orally at 1.68
              g/kg for 4 days or 0.42 g/kg for 12 days.  The rats absorbed about 2 and 5% of
              the administered dose, respectively.

              Very little absorption was observed in rats (strain and sex not given)
              administered carrier-free radioactive silver (< 1 /xg; 1 /iCi) intragastrically by
              stomach tube (Scott and Hamilton, 1950).  Within 4 days after dosing, about 99
              and 0.18% of the original dose was eliminated in the feces and urine,
              respectively.  Total tissue distribution amounted to 0.835% of the administered
              dose.

-------
Silver                                                                            April  1992
       •      Furchner et al. (1966, 1968) administered uฐAg-labeled silver nitrate via the oral
              and intravenous (iv)  routes to female RF mice (0.25 /xCi, oral; 0.25 to 0.26 pCi,
              iv), male Sprague-Dawley  rats (0.5 /iCi via either route), beagle dogs (0.6 //Ci,
              oral: 0.4 pCi, iv), and Macacca mulatto monkeys (0.6 jzCi via either route) (dose
              of Ag in mg not stated).  The total body burden and persistence of silver
              (calculated as an equilibrium factor) increased in proportion to species size and
              was greater with the  iv route than with the oral route in all of the tested species.
              Very little silver was  absorbed from the gut; in all species, cumulative  excretion
              ranged between 90 and 99% on the second  day after oral ingestion.  The extent
              of absorption was found to be directly proportional to the transit time through
              the gut in these species.

       •      In studies evaluating  therapeutic uses of silver in humans and the  effects of
              occupational exposure to silver,  silver was readily absorbed following ingestion or
              inhalation, especially  when the silver ingested was in the colloidal  form (Hill and
              Pillsbury, 1939; Newton and Holmes, 1966; Dequidt et al., 1974).

       Distribution

       •      Silver was found to be present in almost all  human body tissues and to
              accumulate over the lifetime. Tipton and Cook (1963) performed postmortem
              analyses of several metals in the tissues of 150 adults who had died
              instantaneously and who had spent their lives in the United States. The highest
              concentrations of silver were found in the thyroid,  followed by the skin, liver,
              adrenal, intestine (sigmoid colon) and other tissues.

       •      In a study involving 30 American males, minute amounts of silver  have been
              found spectrophotometrically in the blood, brain, liver, lung,  rib bone and
              intestines but not in the urine, kidney, heart, spleen, muscle,  long bones and
              stomach (Kehoe-et al., 1940).

       •      A 47-year-old woman developed argyria after she had ingested an excess amount
              (not specified) of an oral anti-smoking remedy containing 6 mg silver acetate per
              lozenge over a period of 6 months (East et al., 1980).  The tissue distribution of
              silver was evaluated by neutron activation analysis.  No evidence of silver was
              found in or between epidermal cells.  The estimated total body silver content of
              the subject was 6.4+2 g after the woman had been  taking the lozenges for 2-1/2
              years. Silver retention was measured following ingestion of tracer silver acetate
              (4.5 mg) containing 4.43 jtCi 110Ag. After approximately (he  first week, 18% of
              the radioactive silver  tracer was  retained and this retention remained constant up
              to 30 weeks. The blood level of radioactivity 2 hours after administration was
              low (0.00045% of the dose/mL).  On days 1, 2, 4 and 7, urinary silver excretion in
              a  12-hour overnight sample was  2.90, 2.87, 1.94 and 2.40 x 10"*% of the dose in
              the whole sample, respectively.  Silver was detected in various tissues;

-------
Silver                                                                            April 1992
              concentrations ranged between 71.7 +3.7 /zg/g (in the skin biopsy sample) and
              0.34 +_ 0.04 /ig/g (in pubic hair).  Uptake of silver by the skin was substantial
              (about 8,000 times normal uptake) following ingestion  of the oral silver
              preparation.

       •      Whole-body radiation monitoring was carried out on a 29-year-old man who had
              accidentally inhaled an unknown  amount of dust containing uฐAg from an
              experimental nuclear reactor (Newton and Holmes, 1966).  Radiation monitoring
              during the first 155 days after exposure revealed widespread radioactivity in the
              body,  with approximately 25% of the total radiation concentrated near a location
              corresponding to the liver.

       •      Creasey and Moffat (1973) reported silver granule (30- to 90-nm diameter)
              deposition in the kidneys of mature rats (strain not given) after 5 weeks of silver
              nitrate ingestion in drinking water (0.15%). However,  weanling rats exhibiting a
              similar distribution of silver had a much slower deposition time (12  to 14 weeks
              until visualization  by either the naked eye or by light microscopy), and the
              granules were much smaller (< 30 nm in diameter) than those found in the adult
              animals.

       •      Sprague-Dawley rats were maintained for up to 60 weeks on drinking water
              containing 6, 12, or 24 mM silver nitrate (Walker, 1971). Silver  deposition was
              observed within 6 weeks following treatment with 12 mM (1,296 mg silver/L)
              silver nitrate solution in the basement membranes of the glomerulus, colon and
              liver.  After 12 or more weeks, the choroid plexus, thyroid acinar and basement
              membranes of the skin  surface, urinary  bladder and prostatic acini showed silver
              deposition.  Silver deposition continued to appear in animals restored to silver-
              free tapwater for 4 weeks following ingestion of 12 mM silver nitrate in drinking
              water  for 10 weeks.  No silver deposition was observed microscopically in rats
              maintained on 6 mM silver nitrate (648 mg/L of silver) in drinking water for 12
              weeks.  In another study, silver was found to accumulate in the glomerular
              basement membrane of mice exposed to 6 mM silver nitrate in drinking water for
              a period of 21 weeks (Day et al.,  1976).

       •      Olcott (1947) administered silver  in drinking water (as  a 1:1,000  solution of either
              silver nitrate or silver chloride) to albino rats from weaning until death. Deposits
              of silver granules were found in the homogeneous layer of the membrane of
              Bruch, in the membrane underlying the epithelium of the ciliary body and in the
              outer layer of the  optic nerve and its  vessels as they enter  the eyeball.

       •      One day after iv administration of radioactive silver to  rats, most of the
              radioactivity was detected in the liver and spleen of the treated animals
              (Anghileri, 1969; Gammill et al., 1950).  Silver concentrations decreased gradually
              thereafter.

-------
Silver                                                                            April 1992



       Metabolism

       •      No information on the metabolism of silver was found in the available literature.

       Excretion

       •      About 90 to 99% of the 110Ag administered orally (as silver nitrate) to male
              Sprague-Dawley rats,  female RF rats, male beagle dogs and Macacca mulatto
              monkeys was eliminated in the feces; small amounts were eliminated in the urine
              (Furchner et al., 1968).  A similar elimination pattern was detected in  rats
              following iv administration of llฐAg (Gregus and Klaassen, 1986; Scott and
              Hamilton, 1950).  Most of the radioactivity found in the feces was eliminated via
              the bile (Tichy et al.,  1986; Gregus and Klaassen, 1986; Klaassen, 1979).

       •      Biliary excretion of silver was investigated in  male Sprague-Dawley rats after iv
              administration of 110Ag as silver nitrate (10 to 50 /zCi/kg,  2 mL/kg) (Gregus and
              Klaassen, 1986). Two hours after injection of 0.01, 0.03,  0.1 or 0.3 mg/kg metal
              ion, the biliary concentration of silver as a percentage of the administered dose
              was about 26, 45, 32 and 35%, respectively.  The fraction of the dose excreted
              was not markedly altered by the administered dose.  Further review of the  metal
              concentration ratios in the bile, liver and plasma indicated that silver was highly
              concentrated in the bile relative to the plasma,  and that it tended to accumulate
              in the liver.

       •      A marked variation in biliary excretion was observed in different  species
              administered uฐAg as  silver nitrate, in a single iv injection at 0.1 mg/kg of silver
              over a period of 2 hours (Klaassen,  1979).  Thirty minutes after treatment,  male
              Sprague-Dawley rats excreted silver into the bile at a rate of 0.25 /xg/min/kg, New
              Zealand White  male rabbits excreted 0.05 ng/min/kg and mongrel male dogs
              excreted 0.005 /ig/min/kg.  The concentration of silver in the plasma was
              markedly lower in the dog than in the rat or rabbit (at 2  hours, the 110Ag
              concentration in the plasma  of the dog, rabbit and rat was 0.03, 0.1 and
              0.3 /ig/mL,  respectively), indicating a larger volume of distribution in the dog.
              This variation does not appear to be attributable to differences in the transfer of
              silver from plasma to  liver, but rather from liver to bile.  The species with the
              lowest biliary excretion rate  (the dog) had the highest liver concentration of silver
              (rat 1.24, rabbit 2.13 and dog 2.9 /xg silver/g liver).  In all species, the
              concentration of silver in the bile was greater than that in plasma with no
              observable  dose gradient,  thereby indicating an active transport process and a
              saturable mechanism.

       •      Whole-body radiation monitoring was carried out on a 29-year-old man who
              accidentally inhaled an unknown amount of dust containing 110Ag (Newton  and
              Holmes, 1966) from an experimental nuclear reactor. Labeled  silver was not

-------
Silver                                                                            April 1992
              detected in the urine samples taken during the first 54 days but was detected in
              bulked fecal samples up to about day 300.  The inhaled silver was eliminated
              from the body (data for the  first 2  days were not available) following a biphasic
              exponential decay curve that corresponded to a two-compartment model. The
              effective half-life for the first phase was  about  1 day; the half-life identified for
              the slower second (terminal) phase was 43 days.  Earlier data on the inhalation
              exposure of female  beagle dogs to  radioactive silver fit a three-compartment
              model with initial, intermediate and terminal half-life phases of 1.7, 8.4 and 40
              days (Phalen  and Morrow, 1973).

              Thirty silver workers (average age  45 years) exposed to 1 to 100 /zg/m3 of silver in
              the air (based on a  2-month  environmental monitoring), for a mean exposure
              duration of 20 years, had a mean blood silver concentration of 0.011 /xg/mL
              (more than 80% had levels ranging from 0.006 to 0.026 /xg/mL) (Di Vincenzo et
              al., 1985). Matched controls (n =  35) not exposed to silver showed no detectable
              blood silver.  Silver  in urine  samples was generally  not detected in workers or in
              matched controls. Average fecal excretion of silver in the exposed workers
              (n = 29) and the controls (n = 25) was about 15 and 1.5 ttg/g, respectively.
IV.    HEALTH EFFECTS

       Humans

       •      No experimental studies providing data on the adverse health effects of silver
              ingestion in humans were found in the available literature.

              Short-term Exposure

              •      Hill and Pillsbury (1939) reported that accidental ingestion of large doses
                     of silver nitrate resulted in abdominal pain, diarrhea, vomiting, corrosion
                     of the gastrointestinal  tract, shock, convulsions and death.  A fatal single
                     oral dose of silver nitrate  was estimated to be 10 g (about 143 mg/kg for a
                     70-kg person).

              Long-term Exposure

              •      In the early part of this century, inorganic silver salts and colloidal silver
                     preparations (e.g., silver arsphenamine) were used as therapeutic agents,
                     especially in the treatment of syphilis.  The single adverse effect resulting
                     from the therapeutic use of silver  asphenamine administered by injection
                     was argyria (also called argyrosis or argyrism), which is characterized by
                     permanent discoloration and darkening of the skin on exposure to light.
                     In the early stages of argyria,  the  discoloration is observed in the

                                            8

-------
Silver                                                                            April 1992
                     conjunctiva or the nail bed, the mucous membranes of the mouth and
                     pharynx, the lips and many internal organs (Hill and Pillsbury, 1939). Of
                     the 19 reported cases of syphilitic patients, 4 developed argyria after
                     receiving, a total of 6.3 to 10.2 g silver arsphenamine  (14.5% silver) via
                     injection. Although the smallest  dose inducing argyria corresponded to
                     0.91 g of metallic silver, the actual dose is likely to be higher because
                     exposure to silver via ingestion of food and water  and via consumption of
                     other silver-containing therapeutic agents prior to the study were not
                     taken into account.  Furthermore, this estimate is conservative because
                     the observed  response occurred in a small fraction of studied individuals.
                     These individuals may have been sensitive because they were in the late
                     stages of syphilis.

              •      Gaul and Staud (1935)  reviewed  several studies pertaining to argyria in
                     humans resulting from the use of organic and colloidal silver medication.
                     From 1914 to 1928, 13 cases of argyria were  reported, and 27 additional
                     cases were reported from 1928 to 1935. Approximately 30% of the cases
                     occurred in persons under 40 years of age, and 20% occurred in children
                     under 10 years of age.  Argyria developed in 65% of the cases following
                     administration of medication via  the pharyngeal and intranasal routes,
                     and 35% occurred following administration via the oral  route. The
                     duration of treatment ranged from 1 month to 11 years  (about 1 mg
                     silver/kg from oral treatment with argyrol, a mild silver-protein complex
                     and one of three medications used).

              •      Clinical and therapeutic data were presented for 10 males, aged 23 to 64
                     years, and for one woman, aged 49 years, who were treated with iv
                     injections of silver arsphenamine over a 2- to 9.75-year period (Gaul and
                     Staud, 1935).  The total dose of silver arsphenamine injected (31 to 100
                     injections) ranged from 4.1 to 20 g (0.13 to 0.475 g/injection).  In some
                     patients, argyria  developed after  a total dose of 4, 7 or 8 g of silver
                     arsphenamine; in other patients,  argyria did not develop until 10, 15 or 20
                     g had been  injected.  The degree of discoloration was directly
                     proportional to the amount of injected silver arsphenamine and to the
                     duration of exposure.  The authors also studied 10 cases of generalized
                     argyria by biospectrometric analyses of biopsies and concluded that
                     argyria becomes  clinically apparent after an equivalent of about 8 g of
                     silver arsphenamine is retainted in the body. The authors referred to a
                     previous study in which they  established that the quantity of silver in a
                     biopsy specimen  was directly proportional to the dosage  of silver
                     arsphenamine. From the data presented, silver arsphenamine
                     administered  in about 40  doses (0.18 to 0.21  g/injection) over 2 to 4 years
                     to the males in this study corresponded to an average total dose of 8 g   *
                     silver arsphenamine.

-------
Silver                                                                           April 1992
              •      Blumberg and Carey (1934) reported the case of a 33-year-old woman
                     who had ingested capsules containing 16 mg of silver nitrate three
                     times/day during alternate weeks for a period of 1 year. The only effects
                     observed were skin pigmentation and possible asthenia and anemia.
                     Spectrographic analyses of blood samples revealed marked argyria; the
                     woman's blood silver level was 0.5  mg/L  1 week after cessation of silver
                     ingestion; only a small decrease was observed in blood silver levels 3
                     months after silver  ingestion had ceased. Heavy traces of silver in the
                     skin, moderate amounts in the urine and feces and trace amounts in the
                     saliva were also  reported in samples tested 3 months after ingestion of the
                     capsules was stopped.  The total amount of silver ingested during the 1-
                     year period was calculated to be 6.4 g (about 35  mg/day of silver, on
                     alternate weeks  over 1 year), which was identified as the lowest observed
                     cosmetic effect level for this study.

              •      Kent and McCance (1941) reported that a woman who had washed out
                     her nose for many years with an organic silver preparation suffered  from
                     generalized argyria.  The weekly intake of silver via food ingestion over
                     the 3 consecutive weeks of experimentation was 0.05, 0 and 0.7 mg.  No
                     urinary excretion of silver was detected during these weeks.  However,
                     weekly fecal excretion over the 3-week period was 1.3, 1.5  and 2.3 mg.
                     Thus, the woman was in negative silver balance (-1.25, -1.5 and -1.6
                     mg/week), which was not attributed to any physiologic effect but was due,
                     perhaps, to desquamation of silver-containing mucous membrane cells of
                     the mouth, nose and gastrointestinal tract. This study did  not provide any
                     other information from which a dose-effect relationship could be derived.

       Animals

              Short-term Exposure

              •      A single oral dose of 420 mg/kg of silver colloid did not result in any
                     deaths in rats. Mortality was noted only after repeated daily oral
                     ingestion of 1,680 mg/kg for 4 days (Dequidt et al., 1974).

              •      Diplock et al. (1967) administered  1,500 mg/L of silver acetate (970  mg/L
                     of silver) in drinking water to groups of nine weanb'ng Norwegian hooded
                     rats fed a basal vitamin E-deficient diet. All of the rats (mean age 55
                     days) died with liver necrosis 2 to 4 weeks after the addition of silver
                     acetate to the water.  In another group, 1.0 ppm of selenium was added
                     to the diet; 4 of the 9 rats (mean age 74 days) died with liver necrosis. In
                     another group 120 ppm of vitamin  E was added to the diet.  All nine rats'
                     in this group had normal livers at necropsy on day 86 (50 days of silver
                     exposure).  A No-Observed-Adverse-Effect Level (NOAEL)  and a

                                           10

-------
Silver                                                                           April  1992
                     Lowest-Observed-Adverse-Effect  Level (LOAEL) could not be
                     established based on these data.

                     Bunyan et al. (1968) reported that no liver necrosis occurred in
                     Norwegian hooded rats (about 5 weeks old) fed a vitamin E-deficient diet
                     and 1,000 mg/L of silver acetate (650 mg/L of silver) in drinking water
                     (duration not stated).  When the  selenium content of the diet was
                     reduced by omitting yeast from the diet (so that 8.3% casein was the only
                     protein in the diet) liver necrosis was induced by as little as 130 mg/L of
                     silver acetate (85 mg/L of silver) in drinking water.  Vitamin E at 120
                     mg/L reportedly prevented liver necrosis.  A NOAEL and LOAEL were
                     not identified.

                     Grasso et al. (1969) observed  a rapidly fatal liver necrosis beginning on
                     day 14 after addition of silver  acetate to the diet (130 to 1,000 ppm silver
                     acetate; 4 to 33 mg/kg/day, based on Lehman, 1959) or to the drinking
                     water (1,500 ppm silver acetate; 97.5 mg/kg/day assuming a 200-g rat
                     consumes 20 mL/day of water) of vitamin E-deficient weanling Norwegian
                     hooded rats.  LOAELs for silver administered in the diet or drinking
                     water were not identified.

                     Van Vleet (1976) reported that four weanling swine fed a diet containing
                     adequate selenium  and vitamin E and 0.5% silver acetate (3,250 ppm of
                     silver,  or 130 mg/kg/day based on the assumptions of Lehman, 1959) for 4
                     weeks  developed anorexia, diarrhea and growth depression; three of the
                     four pigs died. Hepatic lesions in all four pig* were consistent with
                     hepatosis dietetica. No lesions developed in pigs fed 0.2% silver acetate
                     (1,300  ppm  of silver, or 52 mg/kg/day based on the assumptions of
                     Lehman, 1959). Vitamin E (100 ITJ/kg diet) but  not selenium (1 ppm)
                     supplementation in the diet (two  pigs/group)  prevented development of
                     lesions and  mortality.  The LOAEL and NOAEL were  130 and 52
                     mg/kg/day of silver, respectively.

                     Wagner et al. (1975) administered silver for 52 days as silver acetate in
                     drinking water at 0, 76 or 751 mg/L to groups of ten 21-day-old Holtzman
                     strain  rats fed a vitamin E-deficient and low-selenium (0.02-ppm) diet.  A
                     similar regimen was given to vitamin E-deficient  rats whose diets
                     contained added selenium (0.5 ppm). In 4 of the rats fed the vitamin E-
                     deficient and low selenium diet, 751 mg/L of silver caused severe growth
                     depression and death within 39 to 46 days.  Although histologjcal
                     examinations were  not conducted, no gross evidence of liver necrosis was
                     observed. Dietary  selenium improved growth and survival among the rats
                     given 751 mg/L of silver and completely overcame the growth depression
                     in rats fed 76 mg/L of silver the concentration of silver in the liver  was

                                            11

-------
Silver                                                                            April 1992
                     increased in rats fed selenium. The activity of liver glutathione
                     peroxidase in the selenium-supplemented group was reduced to 30% of
                     control in the at 76-mg/L group and to 4% of control in the 751-mg/L
                     group.

                     The enzyme activity was not reduced in erythrocytes and was not detected
                     in the livers of rats fed the low-selenium (0.02 ppm) basal diet.
                     Administration of 751  mg/L of silver in drinking water for 15 weeks in
                     rats fed a diet supplemented with vitamin E (100 lU/kg) and selenium
                     (0.5 ppm as sodium selenite) decreased glutathione peroxidase activity  in
                     the liver, erythrocytes and kidneys to 5, 37 and 38% of the controls fed
                     the same diet, respectively.  In addition, a 15% decrease in body weight
                     was seen in treated rats.  It was postulated that the antagonism of silver
                     and selenium is exerted through an effect on the biosynthesis of the
                     selenium-containing enzyme glutathione peroxidase.  The 15-week  study
                     (with vitamin E and selenium) suggests  a LOAEL  for silver of less than
                     751 ppm (114.2  mg/kg/day of silver) in  rats.

              •      In a review of the literature, Ganther (1980) suggested that silver toxicity,
                     which is suppressed by the addition of low levels of vitamin E or selenium
                     to the diet, may be due to a silver-induced deficiency of selenium that
                     makes it unavailable for glutathione peroxidase synthesis.  The
                     effectiveness of selenium in  reducing toxicity is attributed to overcoming
                     this deficiency. Vitamin E, however, although effective in overcoming
                     silver- induced growth  depression, did not prevent  the depression of
                     glutathione peroxidase  activity. Thus, it appeared that the growth
                     depression was not caused by the  silver- induced decrease in glutathione
                     peroxidase activity.

              Dermal/Ocular Effects

              •      Rungby (1986) applied three drops of a 0.66% (42  ppm silver) silver
                     nitrate solution to the right eye of male Wistar rats.  Forty-five days after
                     treatment, the rats were killed. Silver deposits were found in the cornea
                     and conjuctiva and scattered in the cells of the outermost part of the
                     anterior corneal epithelium;  heavy deposits were found in Bowman's
                     layer, reticular fibers of the corneal  stroma, Descemet's membrane and
                     the posterior corneal epithelium.  No lexicological  effects were reported.
                    A NOAEL and LOAEL were not established.

              Long-term Exposure

              •     A study by Walker (1971) showed that silver deposits could be found
                    within the glomerular basement membranes in the  kidneys of Sprague-

                                           12

-------
Silver                                                                            April 1992
                     Dawley rats after the rats ingested 12 mM AgNOj (assuming a 200-g rat
                     consumes 20 mL fluid/day; 12 mM AgNO3 or 1,296 mg silver/L equals
                     130 mg/kg/day) for 4, 6, 8, 10, 12, 16, 25 or 60 weeks.  Rats remained in
                     excellent  clinical condition.  Between the 76th and 81st weeks of ingesting
                     this concentration of AgNO3, however, a rapid deterioration in clinical
                     condition was noted. By contrast, rats given 6 mM AgNO3 (65
                     mg/kg/day)  exhibited no difference in appearance, behavior or fluid
                     consumption compared with control animals.  After 12 weeks, no
                     prominent silver deposits were noted, and  the experiment was
                     discontinued.   In rats given 24 mM AgNO3, a large decline in fluid intake
                     occurred  in the first week. The rats were poorly groomed, listless and
                     continued to demonstrate reduced fluid intake during the second week, at
                     which time  the experiment was discontinued.  A LOAEL of 130
                     mg/kg/day was identified. A NOAEL was  not established because,
                     although  no effects were observed at 65  mg/kg/day, the exposure was
                     halted after 12 weeks.  Therefore, no comparison between this group and
                     the nine rats receiving 130 rag/kg/day could be made.

                     As noted by Walker (1971), silver deposits were present in the basement
                     membrane of the kidneys of mice administered 6 rnM silver nitrate (65
                     mg/kg/day of silver) for 12 days to 14 weeks (Day et al., 1976). No
                     toxicity other  than a slight reduction in water intake was noted.  The
                     NOAEL was 65 mg/kg/day.

                     Olcott (1948)  reported that a 1:1,000 solution of silver  nitrate  in drinking
                     water (63.5 mg/kg/day of silver, assuming a 200-g rat consumes  20 mL
                     water/day) given to rats for 218 days induced intense pigmentation of
                     many tissues,  most notably the basement membrane of the glomeruli, the
                     walls of the vessels between the kidney tubules, the portal vein and other
                     parts of the liver, the choroid plexus of the brain, the choroid  layer of the
                     eye and the thyroid g'and. No shortening of lifespan or reduction of body
                     weight occurred.  A NOAEL of 63.5 mg/kg/day was identified.

                     Olcott (1947)  examined the eyes  in life and from tissue sections  at
                     necropsy  of 139 albino rats administered a 1:1,000 solution of silver
                     nitrate in drinking water (63.5 mg/kg/day of silver, assuming a 200-g  rat
                     consumes 20 mL water/day). A treatment  duration of 218 days (3.2 g
                     silver total) resulted in eyes that were slightly gray (stage 1); a duration of
                     373 days  (5.7  g silver total)  resulted in eyes more gray  than pink (stage 2);
                     a duration of  447 days (6.8 g silver total) resulted in dark but translucent
                     eyes  (stage  3), and a duration of 553 days (9.4 g silver  total) resulted in
                     opaque eyes (stage 4).  Histologjcal sections showed few, but definite,
                     granules in  the membranes of Bruch at stage 1.  At stage 4, the
                                            13

-------
Silver                                                                            April 1992
                     membrane was almost uniformly black.  No toxicity was observed.  A
                     NOAEL of 63.5 mg/kg/day was identified.

              •      Olcott (1950) dosed rats (>9 months of age) with a 1:1,000 solution of
                     silver nitrate (63.5 mg/kg/day of silver, assuming a 200-g rat consumes
                     20 mL water/day). The end point of hypertrophy was defined as a ratio
                     of the left ventricle weight to the final body weight of the rat, expressed
                     as the weight of the left ventricle per 100 g of body weight ^0.2.  A total
                     of 233 rats were autopsied.  The  study author reported  that a large
                     number of spontaneous deaths occurred because of advanced pulmonary
                     lesions, but these  lesions were  not attributed to silver ingestion. After
                     long-term silver ingestion (duration not stated), a statistically significant
                     (no p value, Chi = 3.13) hypertrophy of the left ventricle was noted in
                     29% of rats given  silver nitrate compared to 12% of rats given water.
                     The authors postulated that this finding was indicative of vascular
                     hypertension produced when silver deposition caused a  thickening of the
                     basement membrane of the renal glomeruli.  A NOAEL and LOAEL
                     were not identified because the length of exposure  was  not reported.

              Reproductive Effects

              •      No information on the reproductive effects of silver was found in the
                     available literature.

              Developmental  Effects

              •      Robkin et al.  (1973) reported that the concentration of  silver in the livers
                     of 12 anencephalic human fetuses was higher (0.75  ฑ 0.15 mg/kg) than
                     those found in 9 premature  infants (0.68 ฑ 0.22 mg/kg), 12 fetuses
                     obtained through  therapeutic abortions (0.23 ฑ 0.05 mg/kg) or 14
                     spontaneously aborted fetuses (0.21 + 0.05 rag/kg). The authors were not
                     able to determine  if the higher concentrations of silver in anencephalic
                     fetuses was associated with the malformation or if the concentrations
                     were related to fetal age.

              •      Rungby  et al. (1987) treated Wistar rat pups from two litters with
                     subcutaneous injections of silver lactate monohydrate; two pups from each
                     litter received daily injections of 0.10, 0.20 or 0.35 mg during weeks 1,  2,
                     or 3 to 4, respectively. The authors reported that hyppocampal tissues
                     from the treated fetuses contained significantly (p < 0.05) smaller
                     pyramidal cells; they speculated that the findings suggest that pyramidal
                     cells are the first elements in the hyppocampus to show signs of silver
                     toxicity,  or that these cells are selective sites for silver neurotoxicity.
                                            14

-------
Silver                                                                           April  1992
              Mutagenicity

              •      Silver  nitrate (5 x 10"6 to 10"s%), in the absence of exogenous metabolic
                     activation, did not increase the frequency of reversion to streptomycin
                     nondependence in Escherichia coli (Demerec et al.,  1951).

              •      Similarly, AgNO3 (0.1 /xM) was neither mutagenic in E.  coli WP2 nor
                     comutagenic in ultraviolet-irradiated cultures of E. coli WP2 (Rossman
                     and Molina, 1986).

              •      Silver  chloride (0.05 M), at a single nonactivated dose, caused equivalent
                     inhibition of recombinational repair-proficient (H-17) and repair-deficient
                     (M-45) strains of Bacillus subtilis (Nishioka,  1975).

              Carcinogenicity

              •      Local  sarcomas have been induced after subcutaneous implantation of
                     foils and discs of silver and other noble metals. However, Furst  (1979,
                     1981)  concluded that intraperitoneal and subcutaneous implants were
                     invalid indicators of carcinogenicity because  of the phenomenon  of
                     solid-state carcinogenesis, which results in local fibrosarcomas  even with
                     insoluble solids such as smooth ivory and plastic but not with
                     crumbled tin.

              •      Schmaehl and Steinhoff (1960) reported that colloidal silver injected
                     subcutaneously into rats resulted in tumors in 8 of 26 rats surviving longer
                     than 14 months.  In six of the eight rats, the tumor was at the
                     subcutaneous injection site.  In 700 untreated rats, the rate of
                     spontaneous tumor formation was  1 to 3%; no vehicle control was
                     reported.

              •      Furst and Schlauder (1977) suspended silver powder in trioctanoin and
                     gave it once each month by intramuscular  injection to Fischer 344 rats
                     (50/sex/group).  The dose given was 5 mg each for five treatments and 10
                     mg each for five more treatments,  for a total of 75 mg silver. An inert
                     material was used as the vehicle control, and cadmium was  used  as a
                     positive control.  No fibrosarcomas appeared at the injection site in silver-
                     treated rats. Injection site sarcomas were  found only in the vehicle-
                     control (1/50) and cadmium-treated (30/50) rats.  The latent period in the
                     vehicle-control  group was 19  months, and the latent period in the
                     cadmium-treated group was as short as 4 months. The authors concluded
                     that silver was not tumorigenic when administered as a finely divided
                     powder.
                                            15

-------
Silver                                                                          April 1992
 V.     QUANTIFICATION OF TOXICOLOGICAL EFFECTS

       Health Advisories (HAs) are generally determined for One-day, Ten-day, Longer-term
 (up to 7 years)  and Lifetime exposures if adequate data are available that identify a sensitive
 noncarcinogenic end point of toxicity.  The HAs for noncarcinogenic toxicants are derived using
 the following formula:

       MA  -      (NOAEL or LOAEL)  (BW)    _
       ^  "         (UF) (_ IVday)            - _ rng^ (	Mg/L)

 where:

 NOAEL or LOAEL  =    No- or Lowest-Observed-Adverse-Effect Level (the exposure dose
                           in mg/kg bw/day).

               BW  =    assumed body weight of a child (10 kg) or an  adult (70 kg).

             UF(s)   =    uncertainty factors based upon quality and nature of data (10,
                           100, 1,000 or 10,000), in accordance with EPA or NAS/OW
                           guidelines.

             L/day  =    assumed daily water  consumption (1 L/day/for child or 2  L/day/for
                           adult).
       One-day Health Advisory

       No data suitable for determining the One-day Health Advisory (HA) for silver was
found in the available literature.  Acute toxicity studies such as Dequidt et al. (1974) provide
data on lethal doses but do not provide the dose-response data required to calculate the HA.
The cosmetic Drinking  Water Equivalent Level (DWEL) for silver of 0.2 mg/L calculated below
is recommended for use as a conservative estimate of the One-day HA for a child or an adult.

       Ten-day Health Advisory

       No data suitable for determining the Ten-day HA for silver was found in the available
literature. Short-term studies have provided information on lethal doses, but have not provided
NOAELs or LOAELs.  The cosmetic DWEL for silver of 0.2 mg/L calculated below is
recommended for use as a conservative estimate of the Ten-day HA for a child or adult.

       Longer-term Health Advisory

       No data suitable for determining the Longer-term HA for silver were found in the
available  literature.  The data found were not sufficient to establish a NOAEL and LOAEL.
                                           16

-------
Silver                                                                            April 1992
The cosmetic DWEL for silver of 0.2 mg/L calculated below is recommended as a conservative
estimate of the Longer-term HA for an adult or a child.

        Lifetime Health Advisory

        The Lifetime HA represents that portion of an individual's total exposure that is
attributed to drinking water and is considered protective of noncarcinogenic adverse health
effects over a lifetime exposure.  The Lifetime HA is derived in a three-step process. Step 1
determines the Reference Dose (RfD), formerly called the Acceptable Daily Intake (ADI).  The
RfD is an estimate of a daily exposure  level to the human population that is likely to be without
appreciable risk of deleterious effects over a lifetime, and is  derived from the NOAEL (or
LOAEL), identified  from  a chronic (or subchronic) study, divided by an uncertainty  factor(s).
From the RfD. a Drinking Water Equivalent  Level (DWEL) can be determined (Step 2).  A
DWEL is a medium-specific (i.e.,  drinking water) lifetime exposure level, assuming 100%
exposure from that medium, at which adverse, noncarcinogenic health effects would not be
expected to occur.  The DWEL is derived from the multiplication of the RfD by the assumed
body weight of an adult and divided by the  assumed daily water consumption of an adult.  The
Lifetime HA is determined in Step 3 by factoring in other sources of exposure, the relative
source contribution  (RSC).  The RSC from drinking water is based on actual exposure data or,
if data are not available, a value of 20% is assumed.  If the contaminant is classified as a known,
probable or possible human  carcinogen, according to the Agency's classification scheme of
carcinogenic potential (U.S.  EPA, 1986),  then caution must be exercised in making a decision
on how to deal with possible lifetime exposure to this substance.  For human (A) or  probable
human (B) carcinogens, a Lifetime HA is not recommended.  For possible human carcinogens
(C), an additional 10-fold  safety factor  is  used in the calculation of the Lifetime HA. The risk
manager must balance this assessment of carcinogenic potential and the quality of the data
against the likelihood of occurrence and significance of health effects related to noncarcinogenic
end points of toxicity.  To assist the risk manager in this process, drinking water concentrations
associated with estimated  excess cancer risks over the range of 1 in 10,000 to 1 in 1,000,000 for
the 70-kg adult drinking 2 L of water per day are provided in the Evaluation of Carcinogenic
Potential section.

        Specific information on adverse health effects in humans or animals of oral exposure to
graded levels of silver needed for setting Lifetime Health Advisory levels, was  not found in the
available literature.  Two reports of argyria following ingestion of silver were found (Blumberg
and Carey, 1934; East et al., 1980), but these  reports are of single cases with inadequate
documentation of the dose rate.  On the other hand, valuable clinical and therapeutic data were
presented on human cases of argyria by Gaul and Staud (1935, as cited in U.S. EPA, 1980) and
by Hill and Pillsbury (1939, as cited in U.S. EPA, 1980). These data indicate that about 0.9  to
1.5 g of silver administered over a period of 1 to 2 years as iv injections will cause argyria  in
patients. However,  the individuals are  likely to have been sensitive because they were in the
late stages  of syphilis.  Furthermore, many other patients received similar dose regimen without
developing argyria.
                                            17

-------
Silver                                                                             April 1992
        Derivation of the Lifetime Health Advisory for the Cosmetic Effect of Silver

        A Lifetime HA based on cosmetic effects is calculated assuming that 1 g of iv silver will
produce mild argyria in the most sensitive individuals (Gaul and Staud, 1935; Hill and Pillsbury,
1939). Assuming a 4% absorption rate (Furchner et al., 1968) following oral exposure, the 1-g iv
dose corresponds  to an oral dose of 25 g (1 g/ 0.04 = 25 g). This dose is then averaged over a
lifetime, assumed  to be 70 years:

                70 years   x 25 g     =   978 /ig/day
               25,550 days

        Based on  an adult body weight of 70 kg, this corresponds to the Lowest-Observed-
Cosmetic-Effect Level of  14 jig/kg/day (978 /zg/day/70 kg =  14 /xg/kg/day).  Using 14 /ig/kg/day as
the Lowest-Observed-Cosmetic-Effect level for silver, a Lifetime HA for the cosmetic effect of
silver is calculated as follows:

Step 1:  Determination of the Cosmetic RfD

                   14 ^g/kg/day
Cosmetic RfD  =        3           =4.7 /zg/kg/day (rounded to 5 /zg/kg/day)

where:

        14 /zg/kg/day   =      Lowest-Observed-Cosmetic-Effect  Level based on argyria.

                  3   =      Uncertainty factor:  An uncertainty factor of 3 is used to estimate
                             an RfD associated with lifetime exposure.  This uncertainty factor
                             was applied for the following reasons:  First, a 10-fold uncertainty
                             factor is usually applied to human  data to account for intraspecies
                             variability.  However, since this calculation has already included
                             sensitive individuals,  a 10-fold uncertainty  factor is not warranted.
                             Second, an uncertainty factor less than 10  (i.e., 3) is sufficiently
                             protective  since the estimated dose causing argyria within 1-3
                             years is being apportioned over a lifetime, and the effect is based
                             on argyria which is considered a cosmetic effect, and not an
                             adverse health effect.

Step 2:  Determination of the Cosmetic DWEL

                                 5 ug/kg/dav x 70 kg
       Cosmetic DWEL    =          2 L/day             = 175 iig/L (rounded to 200 /zg/L)
                                             18

-------
Silver                                                                           April 1992



where:

        5 /zg/kg/day   =     cosmetic RfD.

              70 kg   =     assumed weight of an adult.

            2 L/day   =     assumed water consumption of a 70-kg adult.

       The DWEL is derived on the assumption that 100% of the silver intake comes from
drinking water. As estimated by the World Health Organization (WHO, 1984), the upper-
bound intake of silver from food is 20 to 80 jig/day and is essentially negligible from air.
Therefore, the Lifetime HA for the cosmetic effect of silver can be calculated by subtracting the
amount of silver obtained in food.

Step 3: Lifetime HA for the Cosmetic Effect of Silver

Lifetime HA for Cosmetic Effect   =

       (0.005 me/kg/dav^ (70 kg) -  0.08 mg/dav       = 0.135 mg/L (rounded to 0.1 mg/L)
                   2 L/day

       Thus, a concentration of silver in water of 100 pg/L or 0.1 mg/L is considered protective
of the cosmetic effect of silver (argyria) for the general population.

       Evaluation of Carcinogenic Potential

       Applying the criteria described in the U.S. EPA's guidelines for assessment of
carcinogenic risk (U.S.  EPA, 1986), silver has been classified in Group D: not classified.  This
category is for agents with inadequate animal evidence of carcinogenicity.


VI.    OTHER CRITERIA. GUIDANCE AND STANDARDS

       •      The U.S. EPA had originally regulated silver with an MCL of  50 /ig/L.
              However, since silver caused only argyria, a cosmetic effect, the U.S. EPA (1991)
              replaced the primary standard of 50 jig/L with a value of 100 /xg/L as the
              secondary standard.


VII.   ANALYTICAL METHODS

       •      Most of the methods available for silver analysis involve atomic absorption
              spectroscopy. In these methods, the metal is dissolved and thermally
              excited.  When excited, the metal absorbs light frequencies characteristic of that

                                           19

-------
Silver                                                                            April  1992
              element.  In addition, colorimetric (dithizone) methods (American Public Health
              Association, 1976) and inductively coupled plasma atomic emission spectroscopy
              can be used to analyze silver (CFR, 1987)

        •     Direct Aspiration Atomic Absorption Spectroscopy (AA). In this technique the
              dissolved metal (silver) is aspirated into a flame source and excited to the point
              of dispersion into a mono-atomic state; a light source whose cathode is the  metal
              of interest passes through the flame and the resulting absorption of light by the
              element of interest is directly proportional to concentration.  Disadvantages of
              this technique include the inability to analyze more than one metal at a time and
              the insensitivity of the technique.

        •     Graphite Furnace Atomic Absorption (GFAA). This technique differs from AA
              in that a specific amount of liquid is dried on the thermal source, effecting a
              concentration step.  The  sample  is electrothermally excited. This technique has
              great sensitivity, but  it is  still a tedious one-metal-at-a-time determination.

VIII.    TREATMENT TECHNOLOGIES

        •     Available data indicate that the following treatment technologies have been
              effective in removing silver from drinking water supplies:  direct  filtration,
              coagulation/ filtration, lime softening, and reverse  osmosis (RO).

        •     Sorg et al. (1978) conducted laboratory-jar tests without coagulants using Ohio
              river water spiked with an initial silver concentration of 0.15 mg/L.  Direct
              filtration of this water with the pH adjusted with soda ash from 6.8 to 9.4
              reduced the silver concentration  by 49 to 56%.

        •     Sorg et al. (1978) also ran laboratory-jar tests to evaluate  the effect of pH on
              silver removal using Ohio river water and found that pH had no significant effect
              on silver removal by either alum or iron coagulation in the pH 6  to 8 range.
              Removal of silver by  alum coagulation decreased above pH 8. Poor floe
              formation was postulated as the cause of this decrease.  Further jar tests showed
              that alum removal of silver was directly proportional to the turbidity of the
              water.  Iron coagulation tests did not show a turbidity effect; 55  to 65% of the
              silver was removed from  river water in these tests.

        •     Hannah et al. (1977)  tested the effectiveness of a pilot plant utilizing
              coagulation/filtration  or excess lime treatment followed by granular-activated
              carbon (GAC) adsorption in removing silver.  The plant consisted of a rapid mix,
              flocculator, sedimentation basin and dual-media filtration. Silver was present at •
              an initial concentration of 0.5 mg/L. Ferric chloride was added at a dose of 40
              mg/L and a pH of 6.2; alum was  added at a dose of 415 mg/L and a pH of 11.5.
              Excess lime  treatment removed 97.1% of the silver, and GAC adsorption

                                            20

-------
Silver                                                                          April 1992
              increased the overall removal efficiency to 98%.  Coagulation/filtration with
              ferric chloride removed 98.2% of the silver, and GAC adsorption increased the
              overall removal efficiency to 99%.  Coagulation with alum removed 96.9% of the
              silver and GAC adsorption increased the overall removal efficiency to 99%.

              McCarty et al. (1982) reported the removal of silver by Water Factory 21, a
              wastewater treatment plant in Orange County, CA, using an excess-lime
              softening treatment process, followed by RO. The Lime (350 to 400 mg/L as
              calcium oxide) was added to raise the pH to 11.  The clarified water was
              carbonated to pH 8.0, filtered and chlorinated.  Silver was present at  an initial
              concentration of 1.6 mg/L.  The RO unit waj> composed of spiral-wound cellulose
              acetate membranes.  Lime softening alone removed 76% of the silver, and the
              RO unit  alone removed 63% of the silver, with an overall 90% silver  reduction
              by lime softening/RO treatment.
                                          21

-------
Silver                                                                           April 1992
 IX.     REFERENCES

 American Public Health Association. 1976. Standard methods for the examination of water
        and wastewater, 14th ed. Washington, D.C.

 Anderson, J.B., E.A. Jenne and T.T. Chao.  1973.  The sorption of silver by poorly crystallized
        manganese oxides.  Geochim. Cosmochim. Acta  37:611-622.  Reviewed in U.S. EPA.
        1979.  U.S. Environmental Protection Agency.  Water  related environmental fate of 129
        priority pollutants.  Vol. I. EPA-440/4-79-29a. Washington,  DC: U.S. EPA, p. 17-1.

 Anghileri, L.J.  1969.  Studies on the in vivo breakdown of insoluble halides. Acta Isot.
        9:347-356.

 Bard, C.C., JJ. Murphy, D.L. Stone and C.J. Terhaar. 1976. Silver in photoprocessing effluents.
        Jr. Water Pollution Control  Fed. 489(2) 389-394.

 Blumberg, H. and T.N. Carey.   1934. Detection of unsuspected and obscure argyria by the
        spectrographic demonstration of high blood silver. J. Am.  Med. Assoc.  103:1,521-1,524.

 Bruland, K.W., K. Bertine, M. Koide and E.D.  Goldberg. 1974. History of metal pollution in
        the Southern California  coastal  zone. Environ. Sci. #1, Technol. 8:425-432.

 Budavari, S., M.J. O'Neil, A. Smith and P.E. Heckelman.  1989. The Merck Index, llth ed.
        Rahway, NJ:  Merck and Co., Inc.

 Bunyan, J., A.T. Diplock, M.A. Cawthome and J. Green. 1968. Vitamin E and stress.
        Nutritional effects of dietary stress with silver in vitamin E-deficient chicks and rats. Br.
        J. Nutr. 22:165-182.

Carson, B.L. and I.C.  Smith. 1977.  Silver, an appraisal of environmental exposure.  In: Trace
        metals in the environment, vol. 2, silver. Ann Arbor, MI:  Ann Arbor Science.

Cearley, I.E. University of Oklahoma.  1971.  Toxicity and bioconcentration of cadmium,
        chromium and silver in Micropteris salmoides  and Lepomuis macrocluiers.  Ph.D.
        thesis.  (Abstract only.)  Diss. Abstr. 328:5281.

Chao, T.T. and BJ. Anderson.   1974. The scavenging of silver by  manganese and iron oxides in
        stream sediments collected from two drainage areas of Colorado.  Chem. Geol. 14:159-
        166.

Chin, Y. 1973. Recovery of heavy metals by microbes.  Ph.D.  thesis. London, Ontario:
        University of Western Ontario School of Engineering, pp.  18-20.
                                           22

-------
Silver                                                                           April  1992
CFR.  1987.  Code of Federal  Regulations. Inductively-coupled plasma atomic emission
        spectrometric method  for trace element analysis of water and wastes, method 200.7.
        Appendix C to part 136, pp. 512-523.

Coleman, R.L. and J.E. Cearley.  1974.  Silver toxicity and accumulation in largemouth bass  and
        bluegill.  Bull. Environ. Contam. Toxicol. 12(1):53-61.  Reviewed in U.S. EPA.  1979.
        U.S. Environmental Protection Agency.  Water related environmental fate of 129
        priority pollutants.  Vol. I. EPA-440/4-79-29a.  Washington, DC:  U.S. EPA, pp. 17-1 to
        17-9.

Cotton, F.A. and G. Wilkinson.  1972.  Advanced inorganic che'mistry.  New York, NY:
        Interscience Publishers, pp. 1,044-1,052. Reviewed in  U.S. EPA.  1979.  U.S.
        Environmental Protection Agency.  Water related environmental fate of 129 priority
        pollutants. Vol. I. EPA-440/479-29a.  Washington, DC:  U.S. EPA, pp. 17-1 to  17-9.

Creasey, M. and D.B. Moffat.  1973.  The deposition of ingested silver in the rat kidney at
        different ages.  Experientia 29:326-327.

Danscher, G. 1981.  Light and electron microscopic localization of silver in biological tissue.
        Histochemistry 71:177-186.

Day, W.A., J.S. Hunt and A.R. McGiven.  1976. Silver deposition in mouse glomeruli.
        Pathology 8:201-204.

Demerec, M., G. Bertani and J. Flint.  1951.  A survey of chemicals for mutagenic action on E.
        co/i. Am. Nat.  85:119-136.

Dequidt, J., P. Vasseur and J. Gromez-Potentier.  1974. Etude toxicologique experimentale de
        quelques derives argentiques.  I. Localisation et elimination. Bull. Soc. Pharm.  Lille
        1:23-35.  [Experimental toxicological study of some silver derivatives]  (in French).

Diplock, A.T., J. Green, J. Bunyan, D. McHale and I.R. Muthy. 1967. Vitamin E and stress.
        The metabolism of D-alpha tocopherol in the rat under dietary stress with silver. Br. J.
        Nutr. 21:115-125.

Di Vincenzo, G.D., C. J. Giordano and L.S. Schrieves.  1985.  Biologic monitoring of workers
        exposed  to silver.  Int.  Arch. Occup. Environ. Health  56:207-215.

Durfor, C.N. and E. Becker. 1964. Public water supplies  of the 100 largest cities in the United
        States, 1962.  U.S. Geological Survey Paper 1812.  Washington,  DC: U.S. Government
        Printing  Office.

Durum, W.H. and J.  Hafty. 1961. Occurrence of minor elements in water. U.S. Geological
        Survey Circular  445. Washington, DC:  National Academy of Sciences.

                                           23

-------
Silver                                                                            April 1992
 Dyck. W.  1968.  Adsorption and coprecipitation of silver on hydrous ferric oxides. Can. J.
        Chem. 46:1441-1444.

 East, B.W., K. Boddy, E.D. Williams. D. Maclntyre and A.L.C. McLay.  1980. Silver retention,
        total body silver and tissue silver concentrations in argyria associated with exposure to
        an anti-smoking remedy containing silver acetate.  Clin. Exp. Dermatol.  5:305-311.

 Freeman, R.A.  1977.  The ecological kinetics of silver in an alpine lake  eco-system.  Second
        ASTM symposium on aquatic toxicology, Oct. 31 to Nov. 1,  1977. (Preprint only.)
        Cleveland, OH: Reviewed in U.S. EPA.  1979. U.S. Environmental Protection Agency.
        Water related environmental fate of 129 priority pollutants.  Vol. I.  EPA-440/479-29a.
        Washington, DC:  U.S. EPA, pp. 17-1 to 17-9.

 Furchner, I.E., G.A. Drake and C.R. Richmond.  1966.  Retention of silver-110 by mice.
        University of California, Las Alamos:  U.S.  Atomic Energy Commission, Science
        Laboratory, pp. 186-190.

 Furchner, I.E., C.R. Richmond and G.A. Drake.  1968.  Comparative metabolism of
        radionuclides in mammals. IV. Retention of silver-110  in the mouse, rat, monkey and
        dog.  Health Physics  15:505-514.

 Furst, A.  1981.  Bioassay  of metals for carcinogenesis: Whole animals. Environ. Health
              Perspect. 40:83-92.

 Furst, A.  1979.  Problems in metal carcinogenesis.  In:  Kharasch, N., ed. Trace metals in  heart
        and disease. New York, NY: Raven Press, pp. 83-92.

 Furst, A. and M.C.  Schlauder.  1977.  Inactivity of two noble metals  as carcinogens.  J. Environ.
        Pathol. Toxicol. 1:51-57.

Gammill, J.C., B. Wheeler, E.L. Carothers and P.F. Hahn.  1950.  Distribution of radioactive
        silver colloids in tissues of rodents following injection by various  routes. Proc. Soc. Exp.
        Biol. Med.  74:691-695.

Ganther, H.E. 1980.  Interactions of vitamin E and selenium with mercury and silver. Ann.
        N.Y. Acad. Sci. 355: 212-225.

Gaul, L.E. and A.H. Staud.  1935. Clinical spectroscopy. Seventy cases of generalized argyrosis
        following organic and colloidal silver medication, including a biospectrometric analysis
        of ten cases. J. Am. Med. Assoc.  104:1,387-1,390.

Grasso,  P., R. Abraham, R. Hendy,  A.T. Diplock, L. Goldberg and J. Green.  1969. The role of
        dietary silver in the production of liver necrosis in vitamin E-deficient rats. Exp. Mol.
        Pathol. 11:186-199.

                                            24

-------
Silver                                                                           April 1992
Gregus, Z., C.D. KJaassen. 1986. Disposition of metals in rats: a comparative study of fecal,
        urinary and biliary excretion and tissue distribution of eighteen metals.  Toxicol. Appl.
        Pharmacol. 85:24-38.

Hannah, S.A., M. Jelus and J.M. Cohen.  1977.  Removal  of uncommon trace metals by physical
        and chemical treatment processes.  J. Water Pol. Con. Fed. 49(11):2,297-2,309.

Harvey, S.C.  1985. Antiseptics and disinfectants,  fungicides, ectoparasiticides.  In:  Oilman,
        A.G., L.S.  Goodman, T.W. Rail and  F. Murad, eds. Pharmacological basis of
        therapeutics, 7th ed.  New York, NY: MacMillan Publishing Co.

Hawley, G.G.  1981.  The  condensed chemical dictionary,  10th ed. New York, NY: Van
        Nostrand Reinhold Co.

Hem, J.D.  1970.  Study and interpretation of the chemical characteristics of natural waters.
        U.S. Geological Survey Paper 1473.  Washington,  DC:  U.S. Geological Survey, pp. 202-
        203.

Hill, W.R. and D.M. Pillsbury. 1939. Argyria,  the pharmacology of silver. Baltimore, MD:
        Williams and Wilkins Company.

Jackson, W.F. and  B.R. Duling.  1983.  Toxic effects of silver-silver chloride electrodes on
        vascular smooth muscles. Circulation Res. 53(1):  105-108.

Just, J. and A. Szniolis. 1938.  Germicidal  properties of silver in water. J. Am. Water Works
        Assoc.  18(4):492-506.

Kehoe, R.A., J. Cholak and R.V. Story.  1940.  A spectrochemical study of the normal ranges of
        concentration of certain trace metals in biological  materials.  J. Nutr. 19:579-592.

Kent, N.L. and R.A. McCance. 1941. The absorption and excretion  of "minor" elements by man.
        1. Silver, gold, lithium, boron and vanadium. Biochem. J.  35:837-844.

Kharkar, D.P., K.K. Turekian and K.K. Bertine. 1968.  Stream supply of dissolved silver,
        molybdenum, antimony, selenium,  chromium, cobalt,  rubidium and cesium to the
        oceans. Geochim. Cosmochim. Acta 32:285-298.

Klaassen, C.D.  1979.  Biliary excretion of  silver in the rat, rabbit and dog.  Toxicol. Appl.
        Pharmacol. 50:49-55.

Kopp, J.F. and  R.C. Kroner.  1967.  Trace  metals in waters of the United States.  A five-year
        summary of trace  metals in rivers and lakes of the United States. (October 1, 1962 to .
        September 30, 1967.) Cincinnati, OH:  U.S.  Department of Interior, Federal Water
        Pollution Control Administration, Division of Pollution Surveillance.

                                            25

-------
 Silver                                                                          April 1992
 Lehman, A.J. 1959. Appraisal of the safety of chemicals in foods, drugs and cosmetics.
        Association of Food and Drug Officials of the United States.

 Luoma, S.N. and E.A. Jenne. 1977. The availability of sediment-bound cobalt, silver and zinc
        to a deposit-feeding clam. In: Drucher,  H. and R.E. Wilding, eds.  Biological
        implications of metals in the environment. Richland, WA:  Hanford Life Sciences
        Symposium.

 McCabe, L.J., J.M. Symons, R.D.  Lee and G.G. Robeck. 1970. Survey of community water
        supply systems.  J. Am. Water Works Assoc.  62:670-687. Reviewed in NAS.  1977.
        National Academy of Sciences.  Drinking  water and health.  Washington, DC:  National
        Academy of Sciences, pp. 215, 290.

 McCarty, P.L., D. Argo, M. Reinhard, J. Giaydon, N. Goodman and M. Aieta. 1982.
        Performance of Water Factory 21 in removing priority  pollutants.  Proc. Water Reuse
        Symp. II, Vol 3,  2,325-2,349. NITS PB82-222662.

NAS. 1977.  National Academy of Sciences.  Drinking water and health. Washington, DC:
        National Academy of Sciences.

Newton, D. and A. Holmes.  1966.  A case of accidental inhalation of zinc-65 and silver-110.
        Radiat.  Res. 29:403-412.

Nishioka, H.  1975.  Mutagenic activity of metal compounds in  bacteria. Mutat. Res. 31:185-189.

Nordberg, G.F.  and L. Gerhardsson. 1988.  Silver  In: Seiler, H.G. and H. Sigel, eds.  Handbook
        on toxicity of inorganic compounds.  New  York, NY: Marcel Dekker, Inc. pp.  619-624.

Olcott, C.T.  1950.  Experimental argyrosis.  V. Hypertrophy of the  left ventricle of the heart in
        rats ingesting silver salts. Arch. Pathol.  49:138-149.

Olcott, C.T.  1948.  Experimental argyrosis.  IV. Morphologic changes in the experimental
        animal.  Am. J. Path.  24:813-833.

Olcott, C.T.  1947.  Experimental argyrosis.  III. Pigmentation  of the eyes of rats following
        ingestion of silver during long periods of time.  Am. J. Path. 23:783-789.

Phalen,  R.F. and P.E. Morrow.  1973. Experimental inhalation of metallic silver.  Health Phys.
        24: 509-518.

Robkin, M.A., D.R. Swanson and T.H. Shepard.  1973. Trace metal concentrations in human
        fetal livers.  Trans. Am. Nucl. Soc.  17:97.
                                           26

-------
 Silver                                                                            April 1992
 Rossrnan. T.G. and M. Molina.  1986.  The genetic toxicology of metal compounds:  II.
        Enhancement of ultraviolet light-induced mutagenesis in Escherichia coli WP2.  Environ.
        Mutagen.  8:263-271.

 Rungby, J.  1986.  The silver nitrate prophylaxis  of crede causes silver deposition in the cornea
        of experimental animals.  Exp. Eye Res.  42:93-94.

 Rungby, J.,  L. Slomianka, G. Danscher, A.H. Andersen  and M.J. West.  1987.  A quantitative
        evaluation of the neurotoxic effect of silver on the volumes of the components of the
        developing rat hippocampus. Toxicol.  43:261-268.

 Schmaehl, D. and  D. Steinhoff.  1960.  Versuche zur krebserzejigung mit kolloidalen silber-und
        goldlo'sungen an ratten.  Z. Krebsforsch. 63:586-591.  [Experimental carcinogenesis in
        rats with colloidal silver and gold solutions] (in German).

 Scott, K.G. and J.G.  Hamilton. 1950.  The metabolism of silver in the rat with radio-silver used
        as an indicator.  Univ. Calif. Publ. Pharmacol. 2:241-262.

 Sorg, T.J., M. Csanady and G.S. Logsdon.  1978. Treatment technology to meet the interim
        primary  drinking water regulations for inorganics: Part 3. J.  Am. Water Works Assoc.
        70:680-691.

 Stokinger, H.E.  1981.  The metals: silver. Ag. In: Clayton, G.D. and F.E. Clayton, eds.
        Patty's industrial  hygiene and toxicology, 3rd ed., vol. 2A.  New York, NY:  John Wiley
        and Sons, pp. 1,881-1,894.

Terhaar, C.J. W.S. Ewell, S.P. Dzluba,  and D.W. Fassett, 1972.  Toxicity of photographic
        processing chemicals  to fish. Photograph. Sci. Eng. 16(5):370-377.

Tichy, P., J. Rosina, K. Blaha Jr. and M. Cikrt. 1986. Biliary excretion of "ฐAg and its kinetics
        in the isolated perfused liver in rats. J. Hyg. Epidemiol. Microbiol.  Immunol.  30:145-
        148.

Tipton,  I.H. and M.J. Cook.  1963.  Trace elements in human tissue.  Part II.  Adult subjects
        from the United States.  Health Phys.  9:103-145.

 U.S. EPA. 1991. U.S. Environmental Protection Agency. Federal Register. 56:3573. January  30.

 U.S. EPA.  1988. Limiting values of radionuclides  intake and air concentration and dose
        conversion factors for inhalation, submersion and ingestion.  Federal Guidance Report
        11.  EPA 520/1-88-020.  Washington, D.C.:   U.S. EPA Office of Radiation Program, p.
        189. September 1988.
                                            27

-------
Silver                                                                           April 1992
U.S. EPA. 1986. U.S. Environmental Protection Agency, Guidelines for carcinogen risk
        assessment. Fed Reg. 51(185): 33,992-34,003.

U.S. EPA. 1980. U.S. Environmental Protection Agency.  Ambient water criteria for silver.
        EPA 440/5-80-071.  Cincinnati, OH:  U.S. EPA Office of Environmental Criteria and
        Assessment.

U.S. EPA. 1979. U.S. Environmental Protection Agency.  Water related environmental fate of
        129 priority pollutants.  Vol. I.  EPA-440/479-29a.  Washington, DC: U.S. EPA, pp. 17-
        1  to 17-9.

U.S. EPA. 1975. U.S. Environmental Protection Agency.  Chemical analysis of interstate
        carrier water supply systems.  EPA-430/9-75-005.  Washington, DC.  Reviewed in NAS.
        1977. National Academy of Sciences. Drinking water and health. Washington,  DC:
        National Academy of Sciences, p. 290.

U.S. Public Health Service.  1962. Drinking water standards. Publication no. 956.  Washington,
        DC:  U.S.  Department of Health, Education and Welfare, Public Health Service.

Van Vleet, J.F.   1976. Induction of lesions of selenium-vitamin E deficiency in pigs fed silver.
        Am. J. Vet. Res. 37:1,415-1,420.

Wagner, P.A., W.G. Hoekstro and H.E. Ganther. 1975.  Alleviation of silver toxicity by selenite
        in the rat in relation to tissue glutathione peroxidase. Proc. Soc. Exp. Biol. Med.
        148:1,106-1,110.

Wahlberg, J.E.  1965.  Percutaneous toxicity of metal compounds.  Arch. Environ. Health.
        11:201-204.

Walker, F. 1971. Experimental argyria:  A model for basement membrane studies.  Br. J. Exp.
        Pathol. 52:589-593.

Weast, R.C., ed. 1977. CRC  handbook of chemistry and physics,  58th ed.  Cleveland, OH:
        CRC Press, p. 239.

WHO, 1984.  Guidelines for drinking  water quality, vol. 2.  Health criteria and other supporting
        information.  World Health Organization. Geneva, pp. 141-144.
                                           28

-------
 EPA 0553

~RX000027511
                                                                               April 199.2.
                                       THALLIUM

                              Drinking Water Health Advisory
                                      Office of Water
                           U.S. Environmental Protection Agency
 I.  INTRODUCTION
        The Health  Advisory Program,  sponsored by  the  Office of  Water (OW),  provides
 information on the health effects, analytical methodology, and treatment technology that would
 be useful in dealing with the contamination of drinking water. Health Advisories (HAs) describe
 nonregulatory concentrations of drinking water contaminants at which adverse health  effects
 would not be anticipated to occur over specific exposure durations. Health Advisories contain
 a margin of safety to protect sensitive members of the population.

        Health Advisories serve as informal technical guidance to assist Federal, State, and local
 officials responsible for protecting public health when emergency spills or contamination situations
 occur.  They are not to be construed  as legally enforceable Federal standards.  The  HAs are
 subject to change as new information becomes available.

        HAs are developed for One-day, Ten-day, Longer-term (approximately 7 years, or 10%
 of an individual's lifetime), and Lifetime  exposures based on data describing noncarcinogenic
 endpoints of toxicity.   For  those substances that are known or probable human carcinogens,
 according to the Agency classification scheme (Group A or  B),  Lifetime Health Advisories are
 not recommended. For substances with a carcinogenic potential, chemical concentration  values
 are correlated with carcinogenic risk estimates by employing a cancer potency (unit risk) value
 together with assumptions for lifelong exposure and the ingestion of water. The cancer unit risk
 is usually derived from a linearized multistage model with 95% upper confidence limits providing
 a low-dose estimate of cancer risk.  The cancer risk is characterized as being  an upper limit
 estimate, that is, the true risk to humans, while not identifiable, is not likely to exceed the upper
 limit estimate and in fact may be lower.  While alternative risk modeling approaches may be
 presented, for example One-hit, Weibull, Logit, or Probit, the range of  risks described  by using
 any of these models has  little biological  significance unless data can be used  to support the
 selection of one model over another. In the interest of consistency of approach and in providing
 an upper-bound on the potential carcinogenic risk, the Agency recommends using the linearized
 multistage model.

-------
Thallium Health Advisory                                                     April 1992.



II.  GENERAL INFORMATION AND PROPERTIES

       CAS No.

       •  Thallium — 7440-28-0
       •  Thallium chloride — 7791-12-0
       •  Thallium sulfate — 7446-18-6

       Structural Formula

       •  TI

       Synonyms

       •  Thallium — Tl
       •  Thallium chloride — thallous chloride
       •  Thallium sulfate — eccothal, thallous sulfate

       Uses

       •  Thallium salts are used to manufacture crystals, imitation jewelry, optical systems, and
          fiberglass.  Thallium is  also used to produce  pigments, corrosion-resistant alloys,
          catalysts, low-temperature thermometers, photoelectric cells, scintillation counters, and
          other electronic equipment (Manzo and Sabbioni, 1988).

       •  ^Tl has been used in myocardial imaging (Atkins et al., 1977); nonradioactive isomers
          of thallium are used in high-temperature superconductors (Waldrop, 1988).

       Properties (Windholz et al.,  1983)

       •  The properties of thallium compounds vary with the specific compounds; examples
          follow:

                                                   Thallium            Thallium
                                 Thallium          chloride            sulfate

Chemical Formula                 Tl                T1C1                T12SO4
Atomic/Molecular Weight         204.38            239.85              504.85
Physical State                    Bluish-white metal                    White powder—
Boiling Point                     1457ฐC            —                  —
Melting Point                    303.5ฐC           430ฐC               632ฐC
Density                          11.85             7.0                 6.77
Vapor Pressure                   —                —                  —

-------
Thallium Health Advisory                                                     April 1992.
Water Solubility                   Insoluble         3.8 Mg/mL          2.70 g/100 mL
                                                   cold water          at 0ฐC; 4.87 g/100
                                                                      mL at 20ฐC
Specific Gravity                    11.85             7.0                 6.77
Log Octanol/Water
   Partition Coefficient            —               —                 —
Taste Threshold                   —               —                 —
Odor Threshold                   —               —                 —
   Occurrence

     •    Thallium (Tl) is a naturally occurring element; mean concentration in the earth's crust
          is 1 ppm (Smith and Carson, 1977).

     •    In addition to the metal, thallium also exists in  two oxidation states,' Tl*1 and Tl*3.
          Generally,  inorganic Tl*1 compounds  are more stable  than their  Tl*3 analogs.
          However, Tl*3 can also form stable complexes with many ligands (Manzo and Sabbioni,
          1988).

     •    Thallium occurs in small amounts in all living organisms; natural levels in plants are
          reported to be between 0.01 and 3,800 ppm ash weight, 0.5 ppm being typical for most
          species (Sabbioni and Manzo, 1980).

     •    Cationic and neutral Tl*3 compounds are present in seawater and freshwater, but
          concentrations have not  been reported (Ridley et al., 1977).

     •    Thallium metal is obtained from  the flue dust of copper, zinc and lead smelters and
          as a byproduct of cadmium and sulfuric acid production. It has been estimated that
          about 1,600 tons (worldwide) of  thallium are released to the environment annually
          from emission sources, from refineries (as impurities), and from other products (Smith
          and Carson, 1977).  Iron and steel production and cement industries are sources of
          thallium emission to the environment (Manzo and Sabbioni, 1988). It is estimated that
          140 tons of thallium are released from plants in the United States (Smith and Carson,
          1977).

     •    Sulfide  components  of coal  are  particularly rich in thallium, and coal burning is
          regarded as a major source of thallium  emission to the  environment (Smith and
          Carson, 1977).  Average thallium  content in coal  varies from 0.05 to 3 ppm; however,
          some brown coal has levels up to 100 ppm.

-------
Thallium Health Advisory                                                      April
    Environmental Fate

     •    Thallium  can  be removed  from  solution by adsorption (onto  clay  minerals),
          bioaccumulation or precipitation as sulfide (in  reducing environments). Most of the
          ligands common  to  aerobic waters  form soluble salts  with thallium,  so that
          precipitation is not important under oxidizing conditions (Callahan and Slimak, 1979).

     •    Thallium  is strongly adsorbed  by montmorillonitic clays.  Magorian et  al. (1974)
          demonstrated that a 1-g/L suspension of the clay hectorite could remove 97% of a 100-
          /xg/L concentration of thallium (species not given) within 24 hours. Similarly, a 1-mg/L
          concentration of thallium was  reduced to 115 Mg/L, and  a 10-/zg/L solution was
          reduced to below 1 /ig/L.  The above values  are for pH 8.1;  adsorption  is not as
          effective at pH 4.0.

     •    Mathis and Kevern (1975), in a study of heavy metal cycling in a lake in southwestern
          Michigan, were able to detect thallium only in  the sediments. Thallium levels in the
          water, plankton and fish were below the limits  of detection.

     •    The alga Ulothrix sp. was able to concentrate thallium by a factor of 127 to 220 within
          1 hour; in comparison,  the concentration factors for 2.7-hour exposures were 114 for
          lead, 30 for cadmium, 80 for zinc and 313  for copper (Magorian et al., 1974).
III. PHARMACOKINETICS

   Absorption

     •    Thallium is readily absorbed by humans and laboratory animals following oral, dermal
          or intratracheal administration (Barclay et al., 1953; Lie et al., 1960; Munch, 1934;
          Shaw, 1933).

     •    Barclay et al. (1953) reported that Tl was readily absorbed from the gastrointestinal
          (GI) tract of a female cancer patient administered 1.8 tig "Tl  NO3 (4 ng Tl/kg)
          following oral administration.  In addition to radioactive  thallium, the woman also
          received single oral doses of 45  mg thallium sulfate (approximately 0.80 mg Tl/kg)
          every 3  days for a total of five doses. Blood "Tl levels peaked 2  hours after dosing
          and amounted to approximately  3% of the administered radioactivity.  At 48 hours
          post-dosing, blood "*n concentrations were equivalent to approximately 1.5% of the
          dose. The patient excreted 15.3% of the administered radioactivity in the urine within
          5.5 days after dosing and eliminated an average of 3.2% of the body burden of Tl for
          24 days (i.e., until death).  Only 0.4% of the administered "Tl was eliminated in the
          feces during the first 3 days after compound ingestion.

-------
Thallium Health Advisory                                                       April 199Z
     •    Male Wistar-derived rats were given one dose of "Tl, as thallous nitrate, by six routes:
          oral (767 /ig Tl/kg), intravenous (iv) (38 jig Tl/kg), intramuscular (im) (96 /ig Tl/kg),
          subcutaneous (96 ng Tl/kg), intratracheal (123 /ig Tl/kg) and intraperitoneal (ip) (146
          Mg Tl/kg) (Lie et al., 1960). The body burden of "Tl, as percent of dose, was similar
          for all routes of exposure and was found  to decay with a single exponential function,
          which extrapolated to 100% at zero time, regardless of route of administration. The
          author concluded that thallium is completely absorbed  from  the GI tract.

     •    Based on the recovery of thallium from the urine, Shaw (1933) reported that at least
          61.6% of an oral dose of thallium sulfate (25 mg Tl/kg) was absorbed from the  GI
          tract of a dog.

     •    Thallium can be absorbed through the skin, as evidenced by the toxicity of topically
          applied thallium  ointments.  Munch (1934)  reviewed  51 case histories of women
          treated for thallium poisoning following external application of an ointment containing
          thallous acetate at a concentration of 3 to 8%. In 29 cases, between 2 and 24 ounces
          of the ointment (approximately 53 to 636 mg Tl/kg for a 5.5% ointment and a 50-kg
          woman) had  been applied an unspecified number of times.  By several weeks after
          application, neurological and GI symptoms and alopecia were observed.

   Distribution

     •    In mice (Andre et al., 1960), rats (Downs et al., 1960; Lie et al., 1960; Sabbioni et al.,
          1980, 1982) and  humans (Barclay  et  al., 1953; Talas et al., 1983), Tl distributes
          throughout the body, including the kidneys, liver, gonads and  brain; highest levels are
          found in the  kidney, where Tl levels may be  two-  to four-fold higher than in other
          tissues.  Several animal studies suggest that Tl also undergoes rapid transplacental
          transfer in  mice (Andre et al., 1960; Olsen and Jonsen, 1982) and rats (Gibson and
          Becker, 1970; Sabbioni et al., 1982; Ziskoven et al., 1983).

     •    Talas et al. (1983) administered to each of five female and five male patients an iv
          tracer dose of 201T1, as thallium chloride (less than or equal to 106 and 130 ng/kg for
          male and female patients, respectively). Levels of radioactivity were serially assayed
          in plasma for up to 1 day after dosing, and the data were analyzed in terms of a two-
          compartment model. For the central compartment, the distribution volume was found
          to be 0.26 L/kg, which is similar to the adult's extracellular volume of about 0.2 L/kg.
          Distribution into the peripheral compartment was very fast (Tw = 3.9 min). The total
          volume  of  distribution was found to be 4.23  L/kg, which is much larger  than the
          average total body water volume in humans  (0.6 L/kg).  This large  observed  total
          volume of distribution is consistent with a migration of Tl into the intracellular space.
          The average terminal  half-life of Tl was determined to be 2.15 days.

-------
Thallium Health Advisory                                                       April 1992.
     •    Barclay et al. (1953) also studied the tissue distribution of ^Tl in the female cancer
          patient (45.4 kg) who had been dosed orally for 15 days until a total dose of 1.8 tig of
          204TlNOj was achieved (approximately 4 ng Tl/kg). Following the death of the patient
          24 days after the initial administration of 2ฐ*T1, analysis of tissue radioactivity indicated
          that Tl was widely and unevenly  distributed throughout the organs. It was estimated
          that radioactivity in tissues amounted to 45% of the dose.  Levels of Tl in tissues were
          represented on a relative scale, as percent radioactive Tl per gram of the average body
          distribution  per gram body weight (i.e., the total body radioactivity at death/weight of
          patient in grams).   Highest tissue levels were found in scalp hair (420% per gram),
          followed by renal papilla (354%),  renal cortex (268%), heart (236%) and spleen
          (200%).  Intermediate levels were found in adrenal medulla (157%), pancreas (129%),
          liver (125%), rib marrow (124%) and adrenal cortex (109%).  Lower values were
          found in nervous tissue (70  to 13%), ovary  and capsule (54%)  and abdominal
          subcutaneous fat (6.7%).

     •    Valence  state had essentially no effect on organ or intracellular distribution of Tl.  At
          16 hours after  oral administration  of 3.15 mg MtTl (Tl*1 or Tl*3) (15.75 mg/kg) to
          groups of five  male Sprague-Dawley rats (190 to 200 g) the kidneys, liver, and testes
          contained approximately 6 to 7, 3 to 4, and 4%, respectively, of the radioactive dose;
          the salivary glands,  heart and brain each contained  <.!%  (Sabbioni et  al., 1980).
          About 36 to 42% of the total radioactivity in kidney homogenates was recovered in the
          cytosol,   23  to  27% was in  the nuclei,  and  15  to  19%  was  in  the liposomes;
          mitochondrial and microsomal fractions each contained approximately 10% of the total
          kidney 201T1.  Only trace amounts  of MtTl (i.e., a  combined total of <1% of the
          administered radioactivity)  were recovered from the six organs listed above after a
          group of five male rats were dosed orally with M1Tl-labeled dimethyl thallium bromide
          in olive  oil  (3.15 mg Tl/rat).  Although organ distribution differed,  the pattern of
          intracellular distribution of dimethylthallium was the same as that of Tl*1 and Tl*3.

     •    Lie et al. (1960) administered 204T1, as thallium nitrate, by the oral route (767 /ig Tl/kg)
          and by five other routes, as indicated above, to groups of four or five male Wistar rats.
          In orally dosed animals, over the  first 7 days, the highest levels of Tl per gram of tissue
          were found  in  kidney (4.71%  of the body burden per gram of tissue), followed in
          decreasing order by salivary glands (1.08%), testes (0.88%), muscle  (0.78%), bone
          (0.74%), GI tract (0.62%), spleen (0.56%), heart (0.54%), liver (0.52%), respiratory
          system (0.49%), hair (0.37%) skin (0.37%) and brain (0.27%). Except for hair, which
          accounted for 60% of the body burden after 21 days, the relative concentrations of Tl
          in tissues were  largely independent of time.

     •    Thallium (administered  as 10 ppm thallous sulfate,  equivalent to 740 /xg/kg bw/day)
          concentrated in the testes of  10 male Wistar rats,  who ingested  the compound in
          drinking water for 60 days (Formigli et al., 1986). Testes of treated rats contained an
          average of 6.3  /xg Tl/g.

-------
Thallium Health Advisory                                                       April 1992
     •    In a study by Downs et al. (1960), Wistar rats (of unspecified sex and number) were
          fed a diet containing thallous acetate at a concentration of 0.003% (approximately 1.4
          mg Tl/kg bw/day assuming 15 g of diet/day and a body weight of 250 g).  After 63 days
          on  the  diet, the highest levels of Tl  were found in kidney (24 /ig Tl/g wet tissue),
          followed by liver, bone, spleen, lung and brain. Levels in these other tissues decreased
          from 16 jig Tl/g wet tissue in liver to  5 Tl/g wet tissue in brain.

     •    The highest concentrations of Tl (60 /xg/g tissue) were found  in the kidney and small
          intestine of four pregnant COBS rats 72 hours after each animal was given a  single
          oral (gavage) dose of 10 mg Tl/kg on gestation day 17 (Sabbioni et al., 1982).  Liver,
          brain and placenta  tissue contained approximately 18 to 19 jig/g; salivary glands and
          blood had average thallium concentrations of 10.6 and 2.0 p.g/g, respectively.  Thallium
          was also found in the maternal brain  (18.4/ig Tl/g tissue) and in the fetal brain (10.1
          Mg Tl/g).

     •    Ziskoven  et al. (1983)  detected Tl in  maternal kidney and brain and in  fetuses of 20
          pregnant Wistar rats and 16 pregnant Kissleg mice within 10 minutes after the animals
          were administered thallous sulfate orally at a dose level of 8 mg Tl/kg.  Concentrations
          of Tl between 1 and 4 x 10"4 M  were measured in maternal kidneys at 1 through 50
          hours post-dosing.  Maternal brain levels of Tl increased slowly, peaking at 3 x  1CCS M
          Tl 60 hours after compound administration.  Fetal tissue contained up to 7 x 10"3 M
          Tl at 1 hour and within 4 hours  after  dosing, Tl concentrations  peaked at 4 x 10"3 M.
          No statistical differences existed between species.

     •    In a study by Gibson and Becker (1970), pregnant Sprague-Dawley rats were infused
          continuously on day 20 of gestation with ""Tl,  as thallous sulfate, at doses of 0.2, 0.4,
          0.8,1.6, 3.2 or 6.4 mg/min/kg. These infusion rates correspond to 0.16, 0.32, 0.64,1.3,
          2.6 and 5.2 mg Tl/min/kg. Thirty-two minutes after the initiation of infusion, maternal
          blood Tl  levels were approximately 15 times higher than those in fetal blood, on a
          weight basis, at the lowest dose, and approximately 30 times higher at the highest dose
          (e.g., approximately 450 nmoles/mL and 15 to 16 nmoles/g in maternal and fetal blood,
          respectively).

     •    Compared to  other  tissues, the  highest levels of  radioactivity (3.35% of  the
          administered dose/g wet tissue) were recovered from the kidneys of four female COBS
          rats 4 hours after ip injection of 2 tig mi\+l (50 /iCi/rat) on gestation day 13 (Sabbioni
          et al., 1982). All other tissues contained <0.60%/g; cerebellum, brain and blood each
          contained 
-------
Thallium Health Advisory                                                       April 1992.
     •    Distribution  of  radioactive  thallium was comparable  among groups of five male
          Sprague-Dawley rats (190 to 200 g) following ip administration of 0.00004, 2, 20, or
          2000 Mg ^'TTVanimal (0.21 ng to 10.3 mg Tl/kg; 50 jiCi/animal) (Sabbioni et al., 1980).
          At 16 hours post-dosing, the muscles, kidneys and testes contained approximately 28
          to 41, 5 to 6, and 2 to 3%, respectively, of the administered radioactivity.  Very low
          levels (<1%) of M1T1 were recovered from lung, heart,  brain  and blood  tissue.
          Stomach and small intestine  levels of "'Tl peaked 2 hours after administration of 2 /ig
          Tl, accounting for about 0.7 and 1.5% of the original dose, respectively. "'Tl persisted
          in the kidneys of rats in the  2-^tg/animal group,  with approximately 2.5%  of the
          radioactive  dose (per gram of tissue)  recovered at  192  hours after compound
          administration.  Brain and heart levels of "'Tl increased slightly between 2 and 192
          hours post-dosing in the l-/ig group, but 1 g  of either tissue contained no more than
          1% of  the  administered test  material.   Time-course  distribution  data for other
          experimental groups were not  reported.  Between  25 and 50% of total homogenate
          "'Tl of  liver and kidney cells of all dosed rats was recovered from the cytosolic and
          nuclear fractions.   Thus, tissue and intracellular  distribution of Tl are not dose-
          dependent, and bioaccumulation is not likely to occur, based on experimental exposure
          levels.

   Metabolism

     •    Sabbioni et al.  (1980),  in their metabolism  study with rats, recovered equivalent
          amounts of "'Tl from several tissues (kidneys, liver, testes, lung, heart, muscle, brain
          and  blood)  and subcellular fractions of liver and  testes homogenates  (nuclei,
          mitochondria, lysosomes, microsomes andcytosol) following oral administration of 3.15
          mg Tl*' or  Tl*3 (15.75  mg/kg) to groups of five male Sprague-Dawley rats.  The
          authors note, however, that the form and valence state of the thallium recovered were
          not discernable.

   Excretion

     •    Barclay et  al. (1953)  studied the excretion of thallium by  a patient female cancer
          patient dosed orally with 1.8 jig "TlNOj (4 Atg Tl/kg) and then 45 mg thallium sulfate
          (36 mg  Tl/kg) every 3 days  for a total of five doses.   In 5.5 days, the patient had
          excreted 15.3% of the radioactive dose  in urine; in  3 days, she excreted only 0.4% in
          the feces. Approximately 3.2% of the amount of Tl in the body was excreted per day.
          Based on the daily excretion of Tl, it was calculated that at the time of death, 24 days
          after dosing, residual radioactivity in tissues amounted to 45%  of the administered
          dose.

     •    Lehmann and Favari (1985) administered thallous sulfate by gavage at a dose of 12.35
          mg/kg (approximately 10 mg Tl/kg) to  female Wistar rats (4 to  8/group). By day 8

                                           8

-------
Thallium Health Advisory                                                       April 1992.
          post-dosing, 32% of the dose was eliminated in feces and 21% in urine.  Lie et al.
          (I960) reported that, in Wistar rats dosed orally with ""Tl as thallous nitrate (767 /ig
          Tl/kg), the ratio of fecal to urinary excretion of thallium increased from about 2 to 5
          between days 2 and 16 after  dosing,  and decreased thereafter to about 3 by day 20.
          The ratios of fecal to urinary excretion were similar, regardless of whether dosing was
          oral, iv, ip or im.

          Most of a single oral dose of thallous sulfate (25 mg TVkg) administered to one dog
          was eliminated via the urine (Shaw, 1933). Excretion of Tl was 32 and 61.6% of the
          dose at 3 and 36 days after dosing, respectively.  No data for excretion in feces were
          given.

          The biological half-life of Tl in rats is reported to be 3.3 days (Lie et al., 1960) and 7.3
          days (Lehmann and Favari, 1985). A mean whole-body half-life of 9.8 days for Tl has
          been reported in  humans dosed iv with MtTl (Atkins et al., 1977).
   IV.  HEALTH EFFECTS

       Humans

          Short-term Exposure

     •    The minimum lethal oral dose for Tl in adult humans is estimated to be in the range
          of 0.2 to 1.0 g (i.e., approximately 3  to 14 mg Tl/kg for a 70-kg adult) (Clayton and
          Clayton, 1981; Grunfeld and Hinostroza,  1964; Heath et al., 1983; Moeschlin, 1980).

     •    Oral  administration  (in what the  authors  reported  appeared  to be  deliberate
          poisonings) of approximately 0.72 g Tl, as the acetate, to one 56- and one 60-year-old
          male in two or three divided doses (i.e., a total of approximately 10.3 mg Tl/kg/person)
          produced neurological symptoms  (paresthesia of the hands  and feet, facial nerve
          weakness, visual impairment)  and death within 1 month of symptom development
          (Cavanagh et al.,  1974).   A  smaller dose of thallium  (0.24 g Tl, 3.4 mg Tl/kg),
          administered orally to a 26-year-old  male, produced chest pain, vomiting, weakness
          and paresthesia in both feet. Alopecia appeared 19 days after the onset of  symptoms.
          Within  several weeks,  hair  regrowth was  seen and  health returned  to normal
          (Cavanagh et al., 1974).  Postmortem examinations of the two older subjects revealed
          minimal neuronal degeneration in the sciatic nerve and spinal cord of one patient and
          marked and extensive distal degeneration of several types of  nerve  fibers and  early
          degenerative changes  in nerve  fibers of the oculomotor muscles in the second patient.
          Peripheral neuropathy was present in all three individuals. Pulmonary congestion and
          edema also were observed, as were centrilobular congestion, necrosis and fatty changes
          of  the liver.  Thallium  was  recovered  from  the tissues/remains of the first two

-------
Thallium Health Advisory                                                      April 1992,
          individuals and in the urine of the third subject; however, no  control data were
          reported for comparison.

     •    Several cases of accidental (oral) Tl poisoning have been reported. Ingestion of 0.49
          to 1.2 g Tl (7.0 to 17 mg Tl/kg), as the sulfate, by 20- to 25-year old males produced
          a wide range of neurological symptoms, ataxia, alopecia, impairment of renal function,
          tachycardia, high blood pressure, coma and death (De Groot et al., 1985; Grunfield
          and Hinostroza, 1964; Moeschlin, 1980;  Moeschlin and Condrau, 1950).  Similar
          symptoms and death occurred in a 20-year-old man who consumed approximately 3.8
          to 7.6 g thallium (54 to 109 mg Tl/kg) as the nitrate salt (Davis et al., 1980).

     •    Ingestion of an unspecified number of Tl-coated peanuts by a 3-year-old boy produced
          ptosis, dysarthria, restricted movement of the left leg, and alopecia of the scalp.  The
          patient eventually recovered, but a residual, slightly winded gait and minimal left ankle
          weakness persisted (Taber, 1964).

     •    Munch  (1934) reviewed 51 case histories of women treated  for neurological  and
          gastrointestinal symptoms and/or alopecia following external application of a depilatory
          ointment containing thallous acetate at a concentration of 3 to 8%.  In 29 cases (where
          doses could be estimated), between 2 and 24 ounces of the ointment (approximately
          53 to  636 mg Tl/kg for a 5.5% ointment and a 50-kg woman) had been applied an
          unspecified number of times.

       Long-term Exposure

     •    No information was found in the available literature.

   Animals

       Short-term Exposure

     •    The acute oral LDM values for various Tl compounds in mice and  rats range from 16
          mg Tl/kg  for thallium sulfate in  rats to 46  mg Tl/kg for  thallium sulfate in mice
          (Danilewicz et al.,  1979; Downs et al., 1960;  NIOSH, 1985; Tikhova, 1964; Truhaut,
          1959).

     •    Intraperitoneal administration of thallic chloride tetrahydrate in single doses of 50,100
          or 200 mg/kg  (26.7, 53.4 or 106.8 mg Tl/kg, respectively) to  Sprague-Dawley  rats
          produced  ultrastructural changes in  mitochondria, increased number of autophagic
          lysosomes and structural changes in the endoplasmic reticulum of the liver (Woods and
          Fowler,  1986).
                                           10

-------
Thallium Health Advisory                                                       April 1992.
     •    Subcutaneous administration of thallium acetate in two to three weekly doses of 10 to
          15 mg/kg (i.e., 7.8 to 11.6 mg Tl/kg)  for up to 16 days to Sprague-Dawley rats
          produced mitochondrial  granules in kidney and liver cells and lipofuscin bodies in
          brain neurons. Acute necrosis was seen  in the mesencephalon of two out of four rats
          (Herman and Bensch, 1967).

       Dermal/Ocular Effects

     •    No information was found in the available literature.

       Long-term Exposure

     •    Thallous acetate administered in the diet of 100-g young Wistar rats (five/sex/dose) for
          15 weeks at levels of 5, 15 or 50 mg/kg  of diet (i.e., 0.4, 1.2 or 3.9 mg Tl/kg bw/day,
          respectively)  produced high mortality:   20, 40 and 100%, respectively, for the three
          dosage groups.   Mortality in the control group was  also high:  40%  by week  15.
          Approximately 80 and 60% of the male and female rats,  respectively, in a fourth group
          of test animals  (five/sex) given a diet  containing 30  mg  thallous acetate/kg feed
          (approximately 2.4 mg Tl/kg bw/day) for  63 days died by week 8 of treatment; 70% (7/
          10) of the controls for this group survived the 9-week experiment. The only significant
          finding at necropsy was moderate to marked alopecia in rats fed a dose of 1.2 mg Tl/
          kg bw/day or greater.  No histopathologic changes were observed (Downs et al., 1960).
          Due to the high mortality rate among all groups,  a true no-effect level could not be
          established.

     •    In a comparison study, thallic oxide was administered in the diet to groups of five 100-
          g young Wistar rats for 15 weeks at levels  of 20, 35, 50, 100 and 500 mg/kg diet (i.e.,
          1.8, 3.1, 4.5, 9.0 and 44.8 mg Tl/kg bw,  respectively).  High mortality resulted at all
          dose levels: 20%  in untreated controls, 20 and 60% at the two lower doses, and 100%
          at the three  higher doses.  Necropsy of  survivors showed a  significant (p .<0.05)
          elevation in kidney weights for females at the two lower test doses and for males
          receiving  the lowest  dose.    Alopecia was observed  at  all  dose levels.   No
          histopathologic changes were found in lungs, liver, kidneys and brain (Downs et al.,
          I960). Again, the high mortality rate precluded identification of a true health effect
          level.

     •    Thallium sulfate administered in the drinking water of 80 female Sprague-Dawley rats
          at 10 mg Tl/kg (i.e., about 1.4 mg Tl/kg/day) for 36 weeks produced 21% mortality and
          a 20% incidence of alopecia. Abnormal electrophysiological parameters (e.g., decrease
          in amplitude  of motor action potentials) were observed in 10 of 16 rats. Histological
          examination of the sciatic nerve  from  six rats  revealed Wallerian degeneration of
          scattered fibers and vacuolization and lamination of the myelin sheath of about 10%
          of the fibers (Manzo- et al., 1983).  Because the study employed only one dose level.

                                           11

-------
Thallium Health Advisory                                                       April 1992*
          and because the one dose used caused a high number of deaths, a no-effect level could
          not be established.

     •    Thallium sulfate administered by gavage to Sprague-Dawley rats (20/sex/dpse) at levels
          of 0.01, 0.05 or 0.25 mg/kg bw/day (i.e., 8.1, 40.5 or 202 /xg Tl/kg bw/day, respectively)
          for 90 days resulted in significant (p <0.05) changes in serum enzymes and electrolytes
          (Stoltz et al., 1986). In males, elevations in the enzymatic activities of serum glutamic-
          oxaloacetic transaminase (SGOT) and lactic dehydrogenase  (LDH) were observed.
          Serum levels of sodium and calcium were significantly (p  <0.05) elevated  at the two
          higher dose levels.  In females, dose-related increases in sodium were apparent at the
          end of treatment,  and SGOT and  LDH  were transiently  elevated  at 30 days of
          treatment.   No histopathologic  changes were found  at the light microscopic level.
          Electron microscopy  was not performed.   A   No-Observed-Adverse-Effect  Level
          (NOAEL) of 0.20 mg/kg/day was established, based on the absence of gross and light
          microscopic histopathology.

     •    Subcutaneous  administration of thallium acetate to 15 Sprague-Dawley rats  as an
          initial  injection of 10 to 20 mg/kg  (i.e., 7.8 to 15  mg  Tl/kg) followed by weekly
          injections of 5 mg/kg (i.e., 3.9 mg Tl/kg) for up to 24 weeks produced ultrastructural
          damage in  kidneys, liver and brain.  Although no changes were visible at the light -
          microscopic level, electron microscopic examination revealed mitochondria! changes
          in kidney and liver and lipofuscin bodies in brain neurons (Herman and Bensch, 1967).
          A no-effect level  was not identified due to uncertainty in dosing protocols.

       Reproductive Effects

     •    Ten male Wistar rats exposed to  10 ppm thallous sulfate (approximately 740 /zg Tl/kg/
          day) in drinking water for 60 days exhibited adverse effects on sperm cell maturation
          and mo til icy.   In  addition,  histological  examinations showed  alterations in  the
          epithelium  of  the seminiferous  tubules and  in the Sertoli cells.  The testicular Tl
          concentration  was 6.3 /ig/g in treated rats versus less than 0.08 ng/g  in controls (p
          <0.01) (Formigli et al., 1986). The use of only one dose precluded establishment of
          a NOAEL and Lowest-Observed-Adverse-Effect Level (LOAEL) for the reproductive
          effects of Tl.

       Developmental Effects

     •    In a study by Claussen et al. (1981), pregnant NMRI mice (29 to 36/ dose level) were
          dosed by gavage with thallous acetate or thallous chloride at daily doses of 0, 3 or 6
          mg/kg (corresponding to 0, 2.3, and 4.7 mg Tl/kg/day for  the  acetate, and 0, 2.6 and
          5.1 mg Tl/kg/day for the chloride, respectively) on days 6 through 15 of gestation.  In
          groups  receiving  thalious chloride, a slight  increase in  postimplantation  loss and
          postnatal mortality  was noted only at the 6-mg/kg  dose level.  For groups receiving

                                           12

-------
Thallium Health Advisory                                                        April 1992.
          thallous acetate, slightly increased incidences of cleft palate were observed at 3 and
          6 mg/kg; a slight reduction in fetal weights was also noted at 6 mg/kg. Thus, this study
          established a NOAEL and LOAEL of 3 and 6 mg/kg, respectively, for developmental
          toxicity of thallous chloride; in contrast, all levels of thallous acetate had some type
          of adverse effect on fetal development.

     •    Claussen et al. (1981) also administered thallous acetate or thallous chloride to groups
          of 9 to 24 pregnant Wistar rats by gavage at doses of 0, 3, 4.5 or 6 mg/kg (i.e., 0, 2.3,
          3.5 and 4.7 mg Tl/kg/day for the acetate, and 0,  2.6,  3.8 and 5.1 mg Tl/kg/day for the
          chloride, respectively) on days 6 through  15  of gestation produced 100% maternal
          mortality in the two highest  dose groups. Lower doses of 3 mg/kg (2.3 or 2.6 mg Tl/kg/
          day) were associated with developmental effects including wavy ribs and dumbbell-
          shaped sternebrae in fetuses, and a slight  increase in postnatal  mortality. All doses,
          therefore, were associated with adverse developmental effects.

     •    Group of six pregnant Sprague-Dawley rats were administered thallous sulfate ip at
          2.5 mg/kg (2.0 mg Tl/kg/day) on days 8 to  10 or 12 to 14 of gestation or at 10 mg/kg
          (8.1 mg Tl/kg/day) on days 12 to 14 of gestation.  All thallium-treated groups had
          litters with reduced fetal body weights.  Increased incidences of unossified vertebral
          bodies were noted in fetuses of both test groups dosed on days 12 to 14 of gestation.
          Signs of maternal toxicity were noted in all Tl treatment groups -- diarrhea, lethargy,
          irritability and body hair loss. However, incidences of hydronephrosis were increased
          only in fetal groups receiving 2.5 mg/kg (Gibson and Becker, 1970). The data show
          that administration of either 2.0 or 8.1 mg Tl/kg/day produces adverse developmental
          effects in fetal  rats.

     •    Postnatal intravenous administration of thallous  sulfate to 6-  and 9-day-old  pups
          (number not specified) at doses of 20 and 40 jig/g (16 and 32 mg Tl/kg, respectively)
          produced defective calcification and severely hypoplastic columnar cartilage of the long
          bones by day 18. However,  the dose causing these effects was not specified (Nogami
          and Terashima, 1973).

     •    Cultured 10.5-day-old rat embryos (number not  reported), when exposed to 3, 10, 30
          or 100 Mg Tl/mL, exhibited  dose-related  growth  retardation  at all dose levels, with a
          complete growth inhibition  at 100 jig/mL.  At 3  ng/mL, no gross abnormalities were
          evident, but cytotoxicity  to  the central nervous  system was  revealed microscopically
          (Anschutz et al., 1981). Thus, all doses had some detrimental effect on developing rat
          fetuses.

       Mutagenicitv

     •    Reverse mutation assays  in Salmonella typhimurium and Escherichia coli were negative
          (Claussen et al., 1981; Kanematsu et al., 1980).  Although the results were uniformly

                                            13

-------
Thallium Health Advisory                                                      April 1992.
          negative, limitations of the Ames assay for detection of metal-induced mutagenesis
          make the results in Salmonella inconclusive (McCann et al., 1975).  Similar problems
          may be encountered with the results in E. coli.

     •    Dose-dependent increases in chromosome and chromatid breaks were observed in rat
          embryo fibroblasts incubated in the presence of 10^ to 10"5 M thallium carbonate for
          24 hours (Zasukhina et al., 1981).

     •    Thallium was positive in  a rat dominant lethal assay, using animals dosed orally with
          thallium carbonate solution (5 x 10"* to 5 x 10* mg/kg/body weight, daily for 8 months).
          The significance of these results is difficult to assess due to incomplete reporting of
          both the results and the  protocol (Zasukhina et al., 1981, 1983).

     •    Incubation of the Bacillus subtilis DNA repair-competent strain, H17 (rec*), and the
          DNA repair-deficient strain, M45(rec"), with a single dose (0.001 M) of nonactivated
          thallium nitrate caused preferential inhibition of M45, indicating that DNA damage
          had occurred (Kanematsu et al., 1980).

     •    Thallium carbonate produced decreased levels of vaccinia virus reactivation, and a
          dose-dependent increase in strand  breaks in C57BL/6 and CBA mouse embryo cells
          and rat embryo fibroblasts, following incubation with 10"4 to 10"* M thallium carbonate
          (Zasukhina et al., 1981, 1983).

     •    No significant increases in sister chromatid exchange were observed in bone marrow
          cells of Chinese hamsters administered two oral doses of 5 or  10 mg/kg thallium
          chloride in a 24-hour period (Claussen et al., 1981).

     •    Thallium acetate at 0.1 and 0.2 mM produced a significant (p < 0.01) and dose-related
          enhancement of viral-induced cell  transformation of Syrian  hamster embryo cells
          incubated in  the presence of 0.025, 0.05, 0.1 or 0.2 mM thallium acetate (Casto et al.,
          1979).

       Carcinogenicitv

     •    No studies on the carcinogenicity of thallium were identified.


V.  QUANTIFICATION OF TOXICOLOGICAL EFFECTS

       Health Advisories are based upon the identification of adverse health effects associated
with the most sensitive and meaningful noncarcinogenic end point of toxicity. The induction of
this effect is related to a particular  exposure  dose over a specified period of time, most often
determined from the results of an experimental animal study.  Traditional risk characterization

                                           14

-------
Thallium Health Advisory                                                     April 1992.
methodology for threshold toxicants is applied in HA development. The general formula is as
follows:

                     (NOAEL or LOAEL)  (BW)
                   '                              '
where:

       NOAEL =  No-Observed-Adverse-Effect Level (the exposure dose in mg/kg bw/day).

       or

       LOAEL =  Lowest-Observed-Adverse-Effect Level (the exposure dose in mg/kg bw/day).

           BW =  assumed body weight of protected individual (10 kg for child or 70 kg for
                  adult).

         UF(s) =  uncertainty factors, based upon quality and nature of data (10, 100, 1,000, or
                  10,000 in accordance with EPA or NAS/OW guidelines).

         L/day =  assumed water consumption (1 L/day for child or 2 L/day for  adult).

       One-day Health Advisory

       The available data are insufficient to develop a One-day HA for Tl. It is recommended
that the Longer-term HA for the 10-kg child, 7.0 /ig Tl/L, be used as a conservative estimate of
the One-day HA value.

       Ten-day Health Advisory

       The available data are insufficient to develop a Ten-day HA for Tl. It is recommended
that the Longer-term HA for the 10-kg child, 7.0 pg Tl/L, be used as a conservative estimate of
the Ten-day HA value.

       Longer-term Health Advisory

       The study of Stoltz et al. (1986) has been selected to serve as the basis for the Longer-
term HA for Tl. In this study, Sprague-Dawley rats (20/sex/group) were dosed by gavage with
0.008, 0.040 or 0.20 mg Tl/kg/day (dose based on the percent of Tl in T12SO4) for 13 weeks. Gross
pathologic and light microscopic examination of organs and tissues did not reveal any significant
treatment-related effects.  Apparent instances of alopecia were not confirmed following light
microscopic examination of hair follicles  for possible damage.   Treated rats also  displayed
moderate changes  in blood chemistry, which included increases  in SGOT, LDH, and sodium

                                          15

-------
Thallium Health Advisory                                                        April 1992.
levels.  However, although the enzymatic activities were significantly elevated (p < 0.05) with
respect to vehicle-treated controls, it was not possible to ascertain whether the effects were dose-
related due to scattering of data points. Furthermore, no gross or histopathologic lesions were
reported that correlated with the blood chemistry changes.  Thus, based on the absence of gross
or light microscopic histopathologic  changes, the dose  of 0.20 mg Tl/kg/day is considered a
NOAEL.

        Other studies identified toxic effects of Tl but were  not used because NOAELs were not
established and/or effects were seen  at higher doses  than the NOAEL of 0.20 mg Tl/kg/day
reported by Stoltz et al. (1986).  These studies are described below.

        True NOAEL and LOAEL values could  not be established in the Downs et al.  (1960)
study in which Wistar rats maintained for up to 15 weeks on a Tl-containing diet developed
alopecia; the significance of the results cannot be assessed  due to the high mortality seen in all
dose levels and controls. Furthermore, tests for testicular toxicity and neurotoxiqity,  which were
observed in other studies (Formigli et  al., 1986; Herman and Bensch, 1967), were not conducted.
Thus, the study by Downs et al. (1960) is not suitable for derivation of the Longer-term HA.

        Manzo et al.  (1983) reported axonal  destruction  and  increased  mortality in  rats
administered 1.4  mg Tl/kg/day in the drinking water  for 36 weeks, and  Formigli et al.  (1986)
reported decreased sperm motility, increased numbers of immature sperm cells, and cytoplasmic
vacuolization in Sertoli cells in rats administered 0.74 mg TL/kg/day in the drinking water  for 60
days.  Because both studies were performed at only one dose level, and consequently no data on
the dose-response relationship are available for extrapolating effects to lower dose levels, neither
study is suitable for derivation of the  Longer-term HA.

        Herman and Bensch  (1967) reported  histopathological changes in the kidney, liver and
brain of rats dosed subcutaneously with Tl. Following an initial subcutaneous injection (precise
dose  unspecified, in the  range  of 7.8 to 15.5 mg  Tl/kg),  rats were administered weekly
subcutaneous injections of 3.9 mg Tl/kg, generally once each week (approximately 0.6 mg  Tl/kg/
day),  for up to 24 weeks. The uncertainty in the dosing protocol, the relatively high  levels of Tl
administered in each weekly dose, and the difficulties in extrapolating between subcutaneous and
oral administration make this study unsuitable for derivation of HA values.

        In view of the uncertainty associated with the  selected NOAEL value of 0.20 mg  Tl/kg/
day(Stoltz et al., 1986), an uncertainty  factor of 3 has been introduced into the calculations  of the
Longer-term HA  to account for inadequate testing of other endpoints of toxicity.  This factor of
3 is in addition to the factor of 100 for use with a NOAEL for an animal study.

        The Longer-term HA for a 10-kg child is  calculated as follows:
                                           16

-------
  Thallium Health Advisory                                                     April 1992.
     Longer-term HA .                         "g)  • 0.0066 mg/L  (rounded  * 7
  where:

0.20 mg Tl/kg/day =  NOAEL, based on the absence of gross or light-microscopic histopathologic
                    changes in rats exposed to thallous sulfate, by gavage, for 90 days (Stoltz et
                    al.,  1986).

           10 kg =  assumed weight of a child.

             100 =  uncertainty factor, chosen in accordance with EPA or NAS/OW guidelines for
                    use with a NOAEL from an animal study.

               3 =  extra uncertainty factor to account for inadequate testing of other species,
                    endpoints, and uncertainties with associated critical study.

         1 L/day =  assumed water consumption of a 10-kg child.

  The Longer-term HA for a 70-kg adult is calculated as follows:

    T             ...     (0.20 mg Tl/kg/day) (70  kg)    nfY,i     r,   ,    A * >  ~>r\    * \
    Longer-term HA = ^ - 2 - ฐL — LLฑ - ฐL = 0.023 mg/L  (rounded to 20 iig/L)
        *                    (100) (3) (2  L/day)               *   V              **  '

  where:

0.20 mg Tl/kg/day =  NOAEL, based on the absence of gross or light-microscopic histopathologic
                    changes in rats exposed to thallous sulfate, by gavage, for 90 days (Stoltz et
                    al.,  1986).

           70 kg =  assumed weight of an adult.

             100 =  uncertainty factor, chosen in accordance with EPA or NAS/OW guidelines for
                    use with a NOAEL from an animal study.

               3 =  extra uncertainty factor to account for inadequate testing of other species,
                    endpoints, and uncertainties associated with a critical study.

         2 L/day =  assumed water consumption of a 70-kg adult.
                                             17

-------
Thallium Health Advisory                                                       April 1992,
Lifetime Health Advisory

        The Lifetime HA represents  that portion of an  individual's total  exposure that  is
attributed to drinking water and is considered protective of noncarcinogenic adverse health effects
over a lifetime exposure.  The Lifetime HA is derived in a three-step process.  Step 1 determines
the Reference Dose (RfD), formerly called the Acceptable  Daily Intake (ADI). The RfD is an
estimate (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the
human population (including sensitive subgroups) that is likely to be without appreciable risk of
deleterious health effects during a lifetime, and is derived from  the NOAEL  (or LOAEL),
identified from a chronic (or subchronic) study, divided by  an uncertainty factor(s). From the
RfD, a Drinking Water Equivalent Level (DWEL) can be determined (Step 2). A DWEL is a
medium-specific (i.e., drinking water) lifetime exposure level, assuming 100% exposure from that
medium, at which adverse, noncarcinogenic health effects would not be expected to occur. The
DWEL is derived from the multiplication of the RfD by the assumed body weight of an adult and
divided by the assumed daily water consumption of an adult.  The Lifetime HA in drinking water
alone is  determined in Step 3 by factoring  in  other sources of exposure, the'relative source
contribution (RSC). The RSC from drinking water is based on actual exposure data or, if data
are not available, a value of 20% is assumed.

       If the contaminant is classified as a known, probable  or possible carcinogen, according to
the Agency's classification scheme of carcinogenic potential (U.S. EPA, 1986c), then caution must
be exercised in making a decision on how to deal with possible lifetime exposure to this substance.
For human (A) or probable (B)  human carcinogens, a Lifetime HA is not recommended. For
possible (C) human carcinogens, an additional  10-fold safety factor  is used in the calculation of
the Lifetime HA. The risk manager must balance this assessment of carcinogenic potential and
the quality of the data against the likelihood of occurrence and significance of health effects
related to noncarcinogenic endpoints of toxicity.  To assist the  risk manager in this  process,
drinking water concentrations associated with estimated excess lifetime cancer risks over the range
of 1 in 10,000 to 1 in 1,000,000 for the 70-kg adult drinking 2 L of water/day are provided in the
Evaluation of Carcinogenic Potential section.

       No chronic study was found to determine the Lifetime HA for Tl.  The 90-day study with
rats by Stoltz et  al.  (1986) has been selected for calculation  of the Lifetime HA for Tl.  In this
study, Sprague-Dawley rats (20/sex/group) were dosed with  T12SO4 by gavage at doses of 0.008,
0.040 or 0.20 mg Tl/kg/day for 13 weeks. Moderate dose-related changes were observed in some
blood chemistry parameters:  increased SCOT, LDH and sodium levels, and decreased blood
sugar levels. The only grossly observed finding at  necropsy thought to be treatment-related was
alopecia, especially in female  rats;  however, microscopic evaluations  did not  reveal any
histopathologic alterations. Furthermore, no gross or histopathologic lesions were observed that
correlated with the blood chemistry changes. Based on the results of this study the dose of 0.20
mg Tl/kg/day is considered a NOAJEL.

       Using the study of Stoltz et al.  (1986), the DWEL is derived as follows:

                                           18

-------
  Thallium Health Advisory                                                      April 199Z
         Step 1:  Determination of the Reference Dose (RfD)
       RfD  = (0-2ฐ               = 0.066 ug Tl/kg/day  (rounded to 0.07 ug Tl/kg/day)
  where:

0.20 mg Tl/kg/day =  NOAEL, based on the absence of gross or light-microscopic histopathologic
                    changes in rats exposed to thallous sulfate, by gavage, for 90 days (Stoltz et
                    al., 1986).

            1,000 =  uncertainty factor, chosen in accordance with EPA or NAS/OW guidelines for
                    use with a NOAEL for an animal study of less-than-lifetime duration.

               3 =  additional uncertainty factor to account for inadequate  testing of other
                    species, endpoints  of toxicity, and uncertainties  associated with the critical
                    study.

         Step 2:  Determination of the Drinking Water Equivalent Level (DWEL)

        DWEL  __ (0.07 ug Tl/kg/day) (70 kg) = ^ ^ m  (rQunded to 2
                           (2  L/day)      \

  where:

0.07 Mg Tl/kg/day =  RfD.

           70 kg =  assumed weight of an adult.

         2 L/day =  assumed water consumption of a 70-kg adult.

         Step 3:  Determination of the Lifetime Health Advisory

                     Lifetime HA = (2.0 jig TVL) (20%  ) = 0.4 ug Tl/L

  where:

      2.0 Mg Tl/L =  DWEL.

            20% =  assumed relative source contribution from water.
                                             19

-------
Thallium Health Advisory                                                     April 1992.



Evaluation of Carcinogenic Potential

•      The  carcinogenic potential of Tl has not  been evaluated by the U.S. Environmental
       Protection  Agency (U.S. EPA) or the International Agency for Research on Cancer
       (IARC).

•      No quantitative assessment of excess cancer risk due to Tl has been reported.

•      Thallium was found to be genotoxic in in vitro assays with mammalian cells.


VI. OTHER CRITERIA. GUIDANCE AND STANDARDS

•      Thallium salts are designated as a hazardous substance under Section 311(b)(2)(A) of the
       Federal Water  Pollution Control Act and further regulated by the Clean Water Act
       Amendments of 1977 and 1978. These regulations apply to discharges of these substances
       (U.S. EPA, 1986a).

•      Reportable quantity of thallium salts, when discharged into or upon the navigable waters
       and  adjoining  shorelines of the United States,  is 1,000  pounds  (454 kg)  (U.S. EPA,
       1986b).


VII.  ANALYTICAL METHODS

•      Methods for metal analysis involve spectroscopy, either emission or absorption.  In all of
       the methods, the metal is dissolved and thermally excited.  All elements when excited
       emit or absorb light frequencies characteristic of that element.  Most metal spectroscopy
       is done in the ultra-violet and x-ray regions.

       Direct Aspiration Atomic Absorption Spectroscopy (AA).  In this technique the dissolved
       metals are aspirated into a flame source, and  excited to  the point that the metals are
       dispersed to a mono-atomic state, a light source whose cathode is the metal of interest
       passes  through the flame, the  resulting absorption of light by the element of interest is
       directly proportional to concentration.  The disadvantages of this technique is the one at
       a time determination of the metals and the insensitivity of the technique.  This is EPA
       method number 279.1 with a detection limit of  25
       Graphite Furnace Atomic Absorption (GFAA).  In this technique a specific amount of
       liquid is dried on the thermal source, effecting  a concentration step.  The sample is
       electrothermally excited. This technique is very sensitive.  This is EPA method number
       286.2 with a detection limit of 0.25 /ig/L.
                                          20

-------
Thallium Health Advisory                                                     April 1992
       Inductively Coupled Plasma  Atomic Emission  Spectroscopv.  This  method  utilizes
       aspiration of a liquid sample, but the flame is actually a plasma torch of argon excited to
       super hot levels by radio-frequency (RF) radiation.  Metals are excited to levels where
       they emit radiation.  By using classical dispersion grating optics a large number of metals
       can be determined simultaneously.  This is EPA method 200.7 with a detection limit of
       10/ig/L.

       Inductively Coupled Plasma Mass Spectrometrv (ICP/MS).  In this case the excitation is
       again by  an  RF plasma, but the  excited  atoms  are  then interfaced into a mass
       spectrometer.  Quantitation is achieved by computerized software programs similar to
       those in other EPA MS organic  methods.  This is  EPA method number 6020 with a
       detection limit of 0;03 /ig/L.
VIII.  TREATMENT TECHNOLOGIES

•      The available literature indicates that conventional coagulation/filtration processes will
       possibly remove Tl from contaminated drinking water.  If granular activated carbon
       (GAG) is added as a post-treatment, then the removal efficiency will be improved.

•      Hannah et al. (1977) tested the effectiveness of a pilot plant utilizing coagulation/filtration
       and excess lime treatment in removing Tl. The plant consisted of a rapid mix designed
       for a capacity of 4 ppm, a flocculator, a sedimentation basin and a dual-media filtration.
       Thallium was present in the influent at a concentration of 0.5 mg/L.  Ferric chloride was
       added at a dose of 40 mg/L and at a pH of 6.2; alum was added at a dose of 220 mg/L
       and at a pH of 6.4; and hydrated lime was added at a dose 415 mg/L and at a pH of 11.5.
       Excess lime treatment produced a Tl reduction of 60%. The alum coagulation produced
       a Tl reduction of 31%, and the ferric chloride coagulation produced a Tl reduction  of
       30%.

•      Hannah et al. (1977) also  reported the results of using GAG  adsorption  as a post-
       treatment to the conventional coagulation/filtration mentioned above, following dual-
       media filtration. Using the above Tl-containing effluents from the coagulation/ filtration
       process, two GAG columns, operated in parallel and designed as "old" and  "new" were
       tested. The "old" GAG had been in use for several months before  this evaluation was
       made. When the lime coagulation effluent was processed through the "old" GAG column,
       an additional 24% of the Tl was removed for a total of 84%, and the "new" GAG column
       removed an additional 12%  for a total of 72%.  When the ferric chloride coagulation
       effluent was  processed through the "old"  column, an additional 17% of the Tl was
       removed for a total of 47%, and the "new" GAG column removed an additional 15% for
       a total of 45%.  When the  alum coagulation effluent was processed through  the "old"
       GAG column, an additional 21% of the Tl was removed for a total of 50%, and the "new"
       GAG column removed an additional 8% for a total of 39%.

                                          21

-------
Thallium Health Advisory                                                      April 1992.
IX.  REFERENCES

Andre, T., S. Ullberg and G. Winqvist.  1960.  The accumulation and retention of thallium in
   tissues of the mouse.  Acta Pharmacol. Toxicol. 16:229-234.

Anschutz, M., R. Herken and D. Neubert.  1981.  Studies on embryo toxic effects of thallium
   using the whole-embryo culture technique.  In:  Neubert, D. and HJ. Merker, eds.  Culture
   techniques: applicability for studies on prenatal  differentiation and toxicity.  5th Symposium
   on prenatal development, Berlin, West Germany, May 7-10, 1981. New York:  Walter De
   Cruyter,  pp. 57-66.

Atkins, H.L., T.F. Budinger, E. Lebowitz, A.N. Ansari, M.W. Greene, R.G. Fairchild and K.J.
   Ellis.  1977. Thallium-201 for medical use. Part 3. Human distribution and physical  imaging
   properties. J. Nucl. Med. 18:133-140.

Barclay,  R.K.,  W.C.  Peacock and  D.A. Karnofsky.   1953.   Distribution and excretion  of
   radioactive thallium in the chick embryo, rat, and man. J. Pharmacol. Exp. Ther.  107:178-187.

Callahan, M. and M. Slimak.  1979. Water-related environmental fate  of 129 priority pollutants.
   EPA-440/14-79-0299. Washington, DC:  U.S. EPA, Office of Water Planning and Standards,
   pp. 18-1  - 18-8.

Casto, B.C.,  J. Meyers and J.A. DiPaolo.   1979.   Enhancement of viral transformation  for
   evaluation of the carcinogenic or mutagenic potential of inorganic  metal salts. Cancer Res.
   39:193-198.

Cavanagh, J.B., N.H. Fuller, H.R.M. Johnson and P. Rudge. 1974.  The effects of thallium salts,
   with particular reference to the nervous  system changes. Q. J. Med.  43:293-319.

Claussen, U., R. Roll, R. Dolgner, G. Matthiaschk, F. Majewski, B. Stoll and G. Rohrborn.  1981.
1  Zur  mutagenitat und teratogenitat von  thallium - unter  besonderer Berucksichtigung der
   situation in lengerich.  Rheinisches Arzteblatt  16:469-475.

Clayton, G.D. and F.E. Clayton.  1981.  Patty's industrial hygiene and toxicology, 3rd ed., Vol. 2A.
   New York:  John Wiley and Sons, Inc., pp. 1914-1931.

Danilewicz, M., M. Danilewicz and A. Kurnatowski. 1979. Evolution of histopathological changes
   in the kidney of experimental animals due to lethal doses of thallium.  Med. Pracy.  30:257-
   265.   (In Polish.)

Davis, L.E., J.C. Standefer, M. Kornfeld, D.M. Abercrombie and C. Butler. 1980. Acute thallium
   poisoning: toxicological and morphological studies of the nervous system.  Ann. Neurol.
   10:38-44.

                                          22

-------
Thallium Health Advisory                                                      April 1992.
De Groot, G., R. Van Leusen and A.N.P. Van Heisjst.  1985.  Thallium concentrations in body
   fluids and tissues in a fatal case of thallium poisoning.  Vet. Hum. Toxicol.  27:115-119.

Downs, W.L., J.K. Scott, L.T. Steadman and E.A. Maynard. 1960. Acute and subacute toxicity
   studies of thallium compounds.  Ind. Hyg. J. 21:399-406.

Formigli, L., R. Scelsi, P. Poggi, C. Gregotti, A. Di Nucci, E. Sabbioni, L. Gottardi and L. Manzo.
   1986. Thallium-induced texticular toxicity in the rat. Environ. Res. 40:531-539.

Gibson, I.E. and B.A. Becker.  1970. Placental transfer, embryotoxicity, and teratogenicity of
   thallium sulfate in normal and potassium-deficient rats.  Toxicol. Appl. Pharmacol.  16:120-
   132.

Grunfeld, O. and G. Hinostroza.  1964. Thallium poisoning. Arch. Intern. Med.  114:132-138.

Hannah, S.A, M. Jelus and J.M. Cohen.  1977.  Removal of uncommon trace metals by physical
   and chemical treatment processes. Journal Water Pollution Control Federation, 49(11):2,297-
   2,309.

Heath, A.,  J.  Ahlmen, B. Branegard, S. Lundstedt, I. Wickstrom and O. Andersen.   1983.
   Thallium poisoning - toxin elimination and therapy in three cases.  J. Toxicol. Clin. Toxicol.
   20:451-463.

Herman, M.M. and K.G. Bensch.  1967.  Light and electron microscopic studies  of acute and
   chronic thallium intoxication  in rats. Toxicol. Appl. Pharmacol.  10:199-222.

Kanematsu, N.,  M.  Hara  and T. Kada.  1980.  Rec assay and mutagenicity studies on metal
   compounds.  Mutat. Res.  77:109-116.

Lehmann, P.A.F. and L. Favari.  1985. Acute thallium intoxication: kinetic study of the relative
   efficacy of several antidotal treatments in rats.  Arch. Toxicol.  57:56-60.

Lie, R., R.G. Thomas and J.K. Scott.  1960. The distribution and excretion of thallium-204 in the
   rat, with suggested MFCs  and a bio-assay procedure. Health Phys. 2:334-340.

Magorian, T.R., K.G. Wood, J.G. Michalovic, S.L. Pek and M.M. Lan Lier. 1974. Water pollution
   by  thallium  and related  metals.   PB 253-333.  Springfield, VA:   National Technical
   Information Service, pp. 145-160.

Manzo, L. and E. Sabbioni. 1988. Thallium. In:  Seiler, H.G. and H. Sigel, eds. Handbook on
   toxicity  of inorganic compounds.  New York: Marcel Dekker, Inc., pp. 677-688.
                                           23

-------
Thallium Health Advisory                                                      April  199Z.
Manzo, L., R. Scelsi, A. Moglia, P. Poggi, E. Alfonsi, R. Pietra, F. Mousty and E. Sabbioni. 1983.
    Long-term toxicity of thallium in the rat. Chem. Toxicol.  Clin. Chem. Met. Proc. Int. Conf.
    2:401-405.

Mathis, BJ. and N.R. Kevern.  1975. Distribution of mercury, cadmium, lead, and thallium in an
    eutrophic lake.  Hydrobiol.  46:207-222.

McCann, J., E. Choi, E. Yamasaki and B.N. Ames. 1975. Detection of carcinogens as mutagens
    in the Sabnonella/microsome test:   assay of 300 compounds.  Proc. Natl. Acad. Sci: USA.
    72:5,135-5,139.

Moeschlin, S.  1980. Thallium poisoning.  Clin. Toxicol. 17:133-146.

Moeschlin, S. and G. Condrau.  1950.  Epileptiforme anfalle bei thalliumvergiftung. Schweiz.
    Med. Wochenschr.  20:519-522.

Munch, J.C.  1934.  Human thallotoxicosis.  J. Am. Med. Assoc. 102:1,929-1,934.

NIOSJJ. 1985:  National Institute for Occupational Safety and Health. Registry of toxic effects
    of chemical substances.    1983-84  Supplement.  Cincinnati, OH:   National  Institute for
    Occupational Safety and Health.

Nogami, H. and Y. Terashima.  1973.  Thallium-induced achondroplasia  in the rat.  Teratol.
    8:101-102.

Olsen, I. and J. Jonsen.  1982. Whole-body autoradiography of 2WT1 in embryos, fetuses and
    placentas of mice.  Toxicol. 23:353-358.

Ridley,  W.P., LJ. Dizikes and J.M. Wood.   1977.  Biomethylation of toxic elements in the
    environment.  Science 197:329-332.
f - -
Sabbioni, E., C. Gregotti, J. Edel, E. Marafante, A. Di Nucci and L.  Manzo. 1982.  Organ/tissue
    disposition of  thallium in pregnant rats. Arch. Toxicol. 5(Suppl.):225-230.

Sabbioni,  E. and  L. Manzo.  1980.  Metabolism and toxicity of thallium.  In: Manzo, L., ed.
    Advances in neurotoxicology.  Oxford:  Pergamon Press, pp.  249-270.

Sabbioni, E., E. Marafante, J. Rade, A. Di Nucci, C. Gregotti and L. Manzo. 1980.  Metabolic
    patterns of low and toxic doses of thallium in the rat. In: Holmstedt B., R. Lauwerys, M.
    Mercier and M. Roberfroid, eds.  Mechanisms of toxicity and hazard evaluation. Amsterdam:
    Elsevier, pp. 559-654.
                                           24

-------
Thallium Health Advisory                                        ""             April 1992,
Shaw, P.A. 1933. Toxicity and deposition of thallium in certain game birds. J. Pharmacol. Exp.
   Ther.  48:478-487.

Smith, C. and B.L. Carson. 1977.  Trace metals in the environment. I. Thallium.  Ann Arbor,
   MI: Ann Arbor Science.

Stoltz, M.L.,  M.A. Stedham, L.K. Brown,  L. Laber and A.M.  El-hawari. Midwest Research
   Institute.  1986.  Subchronic (90-day) toxicity of thallium (I) sulfate (CAS No. 7446-18-6) in
   Sprague-Dawley rats.  Final report. Project no. 8702-L(18).  Prepared for U.S. Environmental
   Protection Agency.

Taber, P.  1964.  Chronic thallium poisoning: rapid diagnosis and treatment. J. Pediatr. 65:461-
   463.

lalas, A., D.P. Pretschner and H.H. Wellhoner. 1983. Pharmacokinetic parameters for thallium
   (I) ions in man.   Arch. Toxicol.  53:1-7.

Tikhova, T.S.  1964.  Problems of labor hygiene in the  production of metallic ^thallium and its
   salts. Hyg. Sanit. 29:26-32.

Truhaut, R. 1959. Recherches sur la toxicologie du thallium. Paris, France: Institute Nationale
   de Securite pour  la Prevention  des Accidents du Travail.

U.S. EPA. 1986a. U.S. Environmental Protection Agency.  Code of Federal Regulations.  40
   CFR 116.4.  July 1.                                                    ...

U.S. EPA. 1986b. U.S. Environmental Protection Agency.  Code of Federal Regulations.  40
   CFR 117.3.  Julyl.
U.S. EPA.  1986c.   U.S.  Environmental Protection Agency.  Guidelines for carcinogen  risk
    assessment. Fed. Reg. 51(185):33992-34003. September 24.

Waldrop, M.M.  1988.  Thallium superconductor reaches 125K. Science 239:1243.

Windholz, M., S. Budavari,  R.F. Bluminetti  and E.S.  Ottenbcin,  eds.  1983.  The Merck
    index—An encyclopedia of chemicals, drugs, and biologicals, 10th ed. Rahway, NJ: Merck
    and Company, Inc., pp. 1324-1,325.

Woods, J.S. and  B.A. Fowler.  1986.  Alteration  of hepatocellular  structure and function by
    thallium chloride:  ultrastructural morphometric, and biochemical studies.  Toxicol. Appl.
    Pharmacol. 83:218-229.
                                           25

-------
Thallium Health Advisory                                                       April 1992.
Zasukhina, G.D., G.N. Krasovskii, I.M. Vasil'eva, N.I. Sdirkova, V.V. Sokolovskii, U.I. Kenesariev
   and L.Y. Vasyukovich.  1981.  Molecularbiological effects of thallium carbonate. Bulk Exp.
   Biol. Med. 90:1,731-1,733.

Zasukhina, G.D.,  I.M. Vasilyeva,  N.I.  Sdirkova,  G.N.  Krasovsky,  L.Y.  Vasyukovich, U.I.
   Kenesariev and P.G. Butenko.  1983.  Mutagenic effect  of thallium and mercury salts on
   rodent cells with different repair activities.  Mutat. Res.  124:163-173.

Ziskoven, R., C.  Achenbach, H.R. Schulten and R. Roll.  1983. Thallium determinations in fetal
   tissues and maternal brain and kidney. Toxicol. Lett.  19:225-231.
                                           26

-------