• Home
• About ATSDR
• Press Room
• A-Z Index
• Glossary
• Employment
• Contact Us
• CDC

ATSDR en Español

Search:

Skip directly to: content | left navigation | search

---

Find a Publication

1. ATSDR > Public Health Assessments & Consultations

PUBLIC HEALTH ASSESSMENT

January 5, 2002 Air Release

DIAZ CHEMICAL CORPORATION
(a/k/a FMC C/O DIAZ CHEMICAL C/O FMC)
VILLAGE OF HOLLEY, ORLEANS COUNTY, NEW YORK

APPENDIX C: NEW YORK STATE DEPARTMENT OF HEALTH REPORT ON JANUARY 14 AND JANUARY 15 SAMPLING

REPORT ON SAMPLING CONDUCTED IN THE VILLAGE OF HOLLEY, NEW YORK, NEAR THE DIAZ CHEMICAL CORPORATION FACILITY
NEW YORK STATE DEPARTMENT OF HEALTH
JANUARY 14-15, 2002

1.0 Background

On January 5, 2002, at approximately 10:45 PM, a pressure build-up in a chemical process tank at the Diaz Chemical Corporation facility resulted in a chemical discharge that visibly contaminated surfaces in the nearby neighborhood and produced odors that were reported as far as 12 miles away. According to Diaz, approximately 80 gallons of liquid material were released. The mixture was reported to be mostly water, toluene and 2-chloro-6-fluorophenol. The product 2-chloro-6-fluorophenol was being stored in the tank for purification. Droplets of the material deposited on cars, houses, and other surfaces to the east of Diaz. Fifteen to twenty families voluntarily relocated by the evening of January 6, 2002, with assistance from Diaz.

On January 8, 2002, citizens began contacting the Orleans County Health Department (OCHD) and the Western Region New York State Department of Health (NYS DOH) office with health concerns. Staff also responded to calls from a medical laboratory, and several physicians made inquiries about what medical tests should be considered in response to the release. The OCHD and Western Regional staff also met with relocated families and other interested parties on January 10.

2.0 NYS DOH Investigation

NYS DOH collected environmental samples 1) to determine whether the 2-chloro-6-fluorophenol could be measured in soil and air (both indoors and outdoors) and on surfaces and 2) to provide additional information about potential exposures.

On January 14 and 15, 2002, staff from the NYS DOH Central and Western Region Offices and Orleans County Health Department sampled air, soil, and surfaces (see Appendix A, Figure 1 for sampling locations). Samples were collected at locations with visible contamination, locations from which complaints were received, locations at different distances downwind of the Diaz facility and also locations believed to be outside of the downwind plume. All samples were delivered to the NYS DOH Laboratory at the Wadsworth Center on January 16.

Staff noticed a chemical-type odor while sampling at all locations on Jackson Street, including at the corner of Jackson Street and South Main Street. Odors were not detected at any of the sampling locations east of South Main Street. Visible evidence of the contamination was present at all of the properties sampled on Jackson Street. No visible evidence of contamination was observed at any of the locations sampled east of South Main Street.

(The wind direction at the time of the release is shown by the arrow; in general, the wind was moving from Diaz toward the Erie Canal.)

3.0 Sample Collection

A. Storage Vessel Material

After the release of material from the storage vessel, Diaz staff washed the vessel with toluene in order to recover any remaining product. At the request of NYS DOH, a 0.5 liter sample of this toluene rinse solution was provided by Diaz for state laboratory analysis. The rinse solution was reported by Diaz to contain mostly water and toluene, in addition to the product of 2-chloro-6-fluorophenol. This sample was taken to Wadsworth Center along with the environmental samples.

B. Air

Air sampling was conducted at three homes bordering Jackson Street (see map locations 2, 3 and 4) and a fourth home on South Main Street (map location 5). Four-hour duplicate samples were collected in the living room, the basement, and in an outside yard at each residence. Silica gel adsorbent cartridges were used as the collection medium as described in the National Institute Occupational Safety and Health's (NIOSH) Method 2014. Samples rates were in the range of 0.133 - 0.153 liter per minute (L/min). Total sampled air volumes ranged from 31.3 - 37.3 liters. Samples from the silica gel adsorbent cartridges were extracted using a method based on NIOSH 2014 and the extracts were analyzed by gas chromatography/flame ionization detector (GC/FID) and gas chromatography/electron capture (GC/EC). The method detection limit, which is the lowest concentration that can be found using this method, was 30 micrograms per cubic meter (µg/m³) for these samples.

C. Surface Wipes

Water-moistened wipe samples were collected at four homes bordering Jackson Street (map locations 1, 2, 3, and 4), one home from South Main Street (map location 5), and at a park gazebo (map location 7) located near the corner of East Street and Perry Street. Wipe sampling was done on surfaces that residents are likely to contact (i.e. doorknobs, railings, etc.). Wipes were extracted and analyzed using gas chromatography/mass spectrometry (GC/MS).

D. Soils

Soil samples were collected from four homes bordering Jackson Street (map locations 1, 2, 3, and 4), and a home on Perry Street. (map location 6). For each sample, approximately 250 grams of soil was collected from a two square-foot area. Scrapes of soil were taken from about the top one-quarter inch of the soil to determine what surface contamination may exist.

E. Vegetation

Two samples of vegetation with visible signs of contamination from the release by Diaz were collected. One sample was leaves from a rhododendron (map location 1) located on a property on Jackson Street, and the other sample was birch bark collected from a tree located on a different property on Jackson Street (map location 3).

4.0 Laboratory Analysis

Quantitation of the target compound (2-chloro 6-fluorophenol) for soil and wipe samples was performed by modifications to routine US Environmental Protection Agency (EPA) methods using GC/MS following sample extraction, clean-up, concentration, and instrumental analysis. Air samples were collected and analyzed using a method based on NIOSH Method 2014, with the analysis done using a GC/FID and GC/EC. Quantitation for all targeted analytes was based on external standardization using the purified standard of 2-chloro-6-fluorophenol provided by Diaz. This sample was found by GC/FID to be greater than (>) 99% pure. Method detection limits based on signal responses for soil samples was less than one microgram per kilogram (< 1 µg/kg). For wipes, the detection limit was less than 0.1 micrograms per square meter (< 0.1 µg/m²) for a 0.25 m² wipe, and for air, less than 30 micrograms per cubic meter (< 30 µg/m³) for a 30-L air sample. Quality Assurance/Quality Control measures done during the analysis all were acceptable. More details on the laboratory performance are provided in Appendix 1.

5.0 Findings

A. Storage Vessel Material

Although the product, 2-chloro-6-fluorophenol, is the most abundant compound present, (excluding the solvents, water and toluene), numerous unknown compounds were also detected. Analysis of the release mixture (provided by Diaz) indicates that other halogenated phenols, along with various dimers and trimers of halogenated phenols and benzyl-substituted halogenated phenols, were also in the mixture. The results are summarized in Table 1.

B. Air

All air samples collected were analyzed for only 2-chloro-6-fluorophenol, and none was detected. The lowest amount that could have been detected was 30 µg/m³ (see Table 2).

C. Surface Wipes

The highest level found in a wipe sample was from a tree on Jackson Street (Table 3). Two samples did not show any contamination: one was collected from a park gazebo (the furthest location from Diaz sampled), and the other was from sampling location 5, the second furthest distance at which a wipe was collected. In most cases the wipe samples that had visual evidence of contamination also were the samples with the greatest concentrations of 2-chloro-6-fluorophenol. The wipe samples also contained several other compounds associated with the 2-chloro-6-fluorophenol. Surface wipes from areas reported to have been washed still showed the presence of 2-chloro-6-fluorophenol and related contaminants.

D. Soils

The analysis of soil sample extracts showed similar qualitative results to that of the solute portion from the storage vessel. The quantitative results for 2-chloro 6-fluorophenol in these samples ranged from 8900 µg/kg (wet weight) to a much smaller amount. Sample locations closest to the release point had the highest concentrations, while those soil samples collected further from the release point (see Appendix A, Figure 1) had lower concentrations (see Table 4).

E. Vegetation

The visible material extracted from plant surfaces (birch bark and rhododendron leaves) contains 2-chloro 6-fluorophenol, dichlorofluorophenol and other halogenated phenols, along with various dimers and trimers of halogenated phenols and benzyl-substituted halogenated phenols that were also in the process tank residue.

6.0 Health Implication

There is very little information on the health effects from exposure to 2-chloro-6-fluorophenol and the associated chemicals in Table 1. Also, this investigation only provides limited information that describes the contamination from the release.

A. Air

As previously indicated, we did not find any 2-chloro-6-fluorophenol in the air, even though odors were detected by the sampling team. This suggests that either very low levels of the 2-chloro-6-fluorophenol exist in the air, or some other compound or combination of compounds is the source of the odor. Although the level appears to be relatively low, information from the Material Safety Data Sheet (MSDS) and the toxicological literature indicates that people will likely smell 2-chloro-6-fluorophenol and associated compounds at very low concentrations in air. When slight or moderate odors are present, some people may experience irritation and effects such as headaches and dizziness. Exposure to high levels of 2-chloro-6-fluorophenol and similar chemicals is likely to cause irritation of the nose, respiratory tract, eyes and skin. We expect that the irritating health effects resulting from short-term exposure to 2-chloro-6-fluorophenol should clear up once the exposure stops.

B. Soil and Surface Wipes

In most samples, the chemical present in the greatest amount in soil and on surfaces is 2-chloro-6-fluorophenol. Other chemicals, dichlorofluorophenols, chloro- and fluorobenzylphenols, and various chlorinated and fluorinated phenol dimers and trimers were also detected in samples of soil and wipes. The levels of these chemicals were generally lower than the levels of 2-chloro-6-fluorophenol. The results of the soil and wipe samples taken from locations west of South Main Street ranged from lower levels of contaminants (several micrograms) to higher levels (several thousand micrograms). Based on these results, the potential for exposure exists. However, significant exposures through dermal contact and incidental ingestion are unlikely at this time of year. It is possible that contamination on surfaces and in the soil contributes to the air contamination through a process called volatilization. Therefore, the contamination on surfaces and soils may contribute to odors and possible health effects from inhalation exposures.

7.0 Conclusions

2-Chloro-6-fluorophenol was not detected in the air samples indicating that either the odor threshold is below the detection limit of 30 µg/m³ or that the odor is due to another associated compound or a combination of compounds. Odors were noted by NYS DOH staff during sampling at all locations on Jackson Street.

Wipes from some of the surfaces reported to have been cleaned contained 2-chloro-6-fluorophenol and the other contaminants. Because no samples were collected from these surfaces prior to the washing, we can not evaluate how much the washing reduced the contamination.

In most cases the wipe samples that had visual evidence of contamination also were the samples with the greatest concentrations of 2-chloro-6-fluorophenol. The laboratory found very low levels of 2-chloro-6-fluorophenol on a swing set and a previously washed handrail that did not have any visible contamination.

In most cases, the ratio of the more volatile 2-chloro-6-fluorophenol to the other less volatile contaminants in samples of soil, vegetation, and wipes is lower than that from the material sampled from the reactor vessel, suggesting that volatilization of this compound occurred and may be occurring.

In general, the soil and surface wipes from locations closest to the release point had the highest concentrations, while those collected further from the release point had lower concentrations.

8.0 Recommendations

Minimizing your contact with visible contamination will help to minimize your exposure. Washing your hands with soap and water after contacting surfaces with contamination should also help reduce exposure.

Removing or cleaning footwear prior to entering your home may reduce tracking of the contamination into the home.

Any cleanup program should consider the merits of using visual inspection as part of the criteria to delineate areas for cleanup and confirmation that cleaning is done.

Any additional work must recognize that the contamination is not just 2-chloro-6-fluorophenol but includes several associated compounds.

Appendix 1. Analytical Methodology

Quantitation of the target compound (2-chloro 6-fluorophenol) for soil and wipe samples was performed by modifications to routine US Environmental Protection Agency (EPA) methods using GC/MS following sample extraction, clean-up, concentration and instrumental analysis. Air samples were collected and analyzed using a method based on NIOSH Method 2014 with the analysis done using a GC/FID and GC/EC. Quantitation for all targeted analyses was based on external standardization using the purified standard of 2-chloro-6-fluorophenol provided by Diaz Chemical. This sample was found by GC/FID to be > 99% pure. Method detection limits based on signal responses for soil samples was < 1 µg/kg; for wipes, < 0.1 µg/m² for a 0.25 m² wipe; and for air, < 30 µg/m³ for a 30-L air sample.

Each batch of samples was analyzed along with quality control samples (blanks, matrix spikes, and quality control standards). For soils the low and high spikes were at 360 and 3600 µg/kg and gave 63 and 67% recovery respectively. A matrix spike of sample 200200125 at 3600 µg/kg gave recovery of 107% of total theoretical value. Wipe samples also gave a 72% recovery of spiked 2-chloro-6-fluorophenol at 100 mcg/wipe and field blanks contain no detectable 2-chloro-6-fluorophenol at <0.2 µg/m². Air cartridges spiked with 2-chloro-6-fluorophenol at 1 µg (equivalent to 33 µg/m3) gave 80% recovery through the analytical procedure.

Two samples of vegetation and the mixture remaining in the process tank following the pressure valve release were also analyzed. The vegetation samples were bark from a birch tree with visible areas of splatter marks, and a rhododendron leaf with similar markings on the surface. The vegetation was selectively sampled by punching out several small areas with visible contamination. These sub-samples were allowed to soak in methanol for 30 minutes, with occasional shaking. The extracts were analyzed by GC/MS full scan. A portion of the toluene-based liquid from the process tank was also diluted 1:100 in methanol and analyzed using the same GC/MS condition to develop chromatograms for qualitative analysis. The mass spectral data were also evaluated to determine the identity of the storage products present in the process tank and environmental samples. Tentative identifications and estimated quantities of these additional major compounds found in the soil and wipes were performed based on routine laboratory procedures. Qualitative comparison of the chromatographic pattern of the vegetation samples and the process tank liquid showed an excellent match, indicating the surface contamination was from the liquid released during the venting of the tank.

Table 1. Analysis of toluene wash solution of residue remaining in process tank after material release.

Compound	Percent composition of solution (%)
2-chloro-6-fluorophenol	55
Dichlorofluorophenol^	21
Chlorofluorophenoxymethylbenzene^	3
Dichlorofluorophenoxymethylbenzene^	2
(Chlorofluorophenoxy)2-H*	10
Dichlorofluorophenoxychlorofluorophenol*	6
(Chlorofluorophenoxy)3-H*	3
Dichlorofluorophenoxy(chlorofluorophenoxy)2-H*	1

^The specific isomer of these compounds has not been determined.

*The exact molecular structure of these compounds has not been determined.

Table 2. Concentration of 2-chloro-6-fluorophenol in air samples in micrograms per cubic meter (mcg/m³).

Location	Date	Sample Site	Concentration
2	1/14/02	outdoor	< 30
		living space	< 30
		basement	< 30
3	1/15/02	outdoor	< 30
		living space	< 30
		basement	< 30
4	1/14/02	outdoor	< 30
		living space	< 30
		basement	< 30
5	1/15/02	outdoor	< 30
		living space	< 30
		basement	< 30

Table 3. Soil results from samples collected by NYS DOH in the village of Holley, NY
January 14-15, 2002
(concentrations in µg/kg)

	Location	1		2	3		4	6
	Sample	under grapevines	roof dripline	rear yard	front yard	under swingset	under swingset	front yard
Compound	2-chloro-6-fluorophenol	8900	2000	960	520	5.3	470	4
	Dichlorofluorophenol^	4200	1600	510	330	2.7	230	1.3
	Chlorofluorophenoxy-methylbenzene^	880	350	410	170	ND	130	ND
	Dichlorofluorophenoxy-methylbenzene^	1200	410	520	240	ND	150	ND
	(Chlorofluorophenoxy)2-H*	2600	730	720	750	1.3	200	2.7
	Dichlorofluorophenoxy-chlorofluorophenol*	1100	450	340	350	1.3	130	ND
	(Chlorofluorophenoxy)3-H*	780	300	520	410	ND	110	ND
	Dichlorofluorophenoxy-(chlorofluorophenoxy)2-H*	160	87	100	190	ND	24	ND

^The specific isomer of these compounds has not been determined.
*The exact molecular structure of these compounds has not been determined.
ND: The compound was not detected.

Table 4. Moist wipe results from samples collected by NYS DOH in the village of Holley, NY
January 14-15, 2002
(concentrations in µg/m²)

	Location	1	2			3		4			5	7
	Sample	hand rail, front porch	hand rail, rear entrance	tree, front yard	door handle, front	swingset and slide	hand rail, side entrance	swingset and slide	car door, front	hand rail, rear entrance	hand rail, front left	hand rail, gazebo
	Washed?	yes	no	no	yes	unknown	yes	yes	no	unknown	unknown	no
	Visual Evidence?	yes	yes	yes	yes	no	yes	yes	yes	yes	no	no
Compound	2-chloro-6-fluorophenol	49	170	360	46	0.7	0.43	170	87	7.4	ND	ND
	Dichlorofluorohpenol^	120	80	870	12	0.23	0.21	82	120	4.3	ND	ND
	Chlorofluorophenoxy-methylbenzene^	5.6	26	260	ND	ND	ND	27	160	0.86	ND	ND
	Dichlorofluorophenoxy-methylbenzine^	10	29	430	ND	ND	ND	ND	210	1	ND	ND
	(Chlorofluorophenoxy)2-H*	17	48	2200	3.9	ND	ND	130	680	1	ND	ND
	Dichlorofluorophenoxy-chlorofhuorophenol*	5.6	17	650	ND	ND	ND	33	190	0.43	ND	ND
	(Chlorofluorophenoxy)3-H*	2.7	7.2	310	ND	ND	ND	14	110	ND	ND	ND
	Dichlorofluorophenoxy-(chlorofluorophenoxy)2-H*	0.54	ND	56	ND	ND	ND	ND	23	ND	ND	ND

"Washed" refers to whether the sampled area was reported to have been powerwashed or cleaned prior to sampling.
"Visual Evidence" refers to the presence or absence of staining, residues, or other discoloration thought to have resulted from the discharge.
^The specific isomer of these compounds has not been determined.
ND: The compound was not detected.

APPENDIX D: NEW YORK STATE DEPARTMENT OF HEALTH PROCEDURES FOR EVALAUTING POTENTIAL HEALTH RISKS FOR CONTMAINANTS OF CONCERN

To evaluate the potential health risks from contaminants of concern associated with the accidental release of CFP, the New York State Department of Health assessed the risks for cancer and non-cancer health effects.

Increased cancer risks were estimated by using site-specific information on exposure levels for the contaminant of concern and interpreting them using cancer potency estimates derived for that contaminant by the US EPA or, in some cases, by the NYS DOH. The following qualitative ranking of cancer risk estimates, developed by the NYS DOH, was then used to rank the risk from very low to very high. For example, if the qualitative descriptor was "low", then the excess lifetime cancer risk from that exposure is in the range of greater than one per million to less than one per ten thousand. Other qualitative descriptors are listed below:

An estimated increased excess lifetime cancer risk is not a specific estimate of expected cancers. Rather, it is a plausible upper bound estimate of the probability that a person may develop cancer sometime in his or her lifetime following exposure to that contaminant.

There is insufficient knowledge of cancer mechanisms to decide if there exists a level of exposure to a cancer-causing agent below which there is no risk of getting cancer, namely, a threshold level. Therefore, every exposure, no matter how low, to a cancer-causing compound is assumed to be associated with some increased risk. As the dose of a carcinogen decreases, the chance of developing cancer decreases, but each exposure is accompanied by some increased risk.

There is general consensus among the scientific and regulatory communities on what level of estimated excess cancer risk is acceptable. An increased lifetime cancer risk of one in one million or less is generally not considered a significant public health concern.

For noncarcinogenic health risks, the contaminant intake was estimated using exposure assumptions for the site conditions. This dose was then compared to a risk reference dose (estimated daily intake of a chemical that is likely to be without an appreciable risk of health effects) developed by the US EPA, ATSDR and/or NYS DOH. The resulting ratio was then compared to the following qualitative scale of health risk:

Noncarcinogenic effects unlike carcinogenic effects are believed to have a threshold, that is, a dose below which adverse effects will not occur. As a result, the current practice is to identify, usually from animal toxicology experiments, a no-observed-effect-level (NOEL). This is the experimental exposure level in animals at which no adverse toxic effect is observed. The NOEL is then divided by an uncertainty factor to yield the risk reference dose. The uncertainty factor is a number which reflects the degree of uncertainty that exists when experimental animal data are extrapolated to the general human population. The magnitude of the uncertainty factor takes into consideration various factors such as sensitive subpopulations (for example, children or the elderly), extrapolation from animals to humans, and the incompleteness of available data. Thus, the risk reference dose is not expected to cause health effects because it is selected to be much lower than dosages that do not cause adverse health effects in laboratory animals.

The measure used to describe the potential for non-cancer health effects to occur in an individual is expressed as a ratio of estimated contaminant intake to the risk reference dose. A ratio equal to or less than one is generally not considered a significant public health concern. If exposure to the contaminant exceeds the risk reference dose, there may be concern for potential non-cancer health effects because the margin of protection is less than that afforded by the reference dose. As a rule, the greater the ratio of the estimated contaminant intake to the risk references dose, the greater the level of concern. This level of concern depends upon an evaluation of a number of factors such as the actual potential for exposure, background exposure, and the strength of the toxicologic data.

APPENDIX E: INTERIM PUBLIC HEALTH HAZARD CATEGORIES

*Such as environmental and demographic data; health outcome data; exposure data; community health concerns information; toxicologic, medical, and epidemiologic data; monitoring and management plans.

APPENDIX F: ESTIMATION OF EXPOSURE TO CFP BASED ON URINE LEVELS

BACKGROUND

Urine was collected from residents of Holley, New York during four sampling periods between January 17 and May 22 and analyzed for the presence of 2-chloro-6-fluorophenol (CFP). These analyses showed that CFP was present in urine of several individuals. Analyses of urine from individuals providing more than one sample during the January 17 to May 22 period showed that, for most cases, the concentration of CFP decreased over time. In some cases where the concentration of CFP in urine increased, individuals returned to a home within the area of greater impact after having been away.

The observed pattern of CFP urine concentrations in Holley residents could reflect continuing (although gradually decreasing) exposure to CFP and/or a relatively long residence time of the chemical in the body. Alternatively, CFP in urine could be the result of the breakdown of other chemicals, such as tentatively identified compounds, which may have been absorbed by individuals. The likelihood of each of these possibilities was evaluated, and it was concluded that continuing, but decreasing, exposures to CFP is the most likely explanation for the presence of CFP in urine.

Information on Toxicokinetics of Structurally Related Chemicals Suggests that CFP is Probably Quickly Absorbed and Relatively Quickly Eliminated in Urine

The degree to which continuing exposure or long residence time explains the pattern of CFP detected in the urine depends on the toxicokinetics (absorption, distribution, metabolism, elimination) of this chemical. Urinary levels of any chemical depend upon how well the chemical is absorbed and distributed in the body and how rapidly and completely it is metabolized and excreted in urine. After a single exposure, urine levels of chemicals that are readily absorbed, distributed, metabolized, and excreted in the urine tend to rapidly increase and then fairly rapidly and steadily decrease. If exposures to these types of chemicals are repeated or continuing, however, urine levels may not decrease immediately and the rate of decrease may be slowed down. This is because the chemical is continually being absorbed, distributed, metabolized, and excreted (Rozman and Klaassen 2001). Therefore, if exposures are repeated or continuing, these types of chemicals may be detected in the urine for a longer period of time. The amount detected in urine would depend upon the amount of exposure, how often exposure occurred, and the timing and duration of exposure with respect to urine collection.

Although information on the toxicokinetics of CFP in humans or animals was not found in the scientific literature, some information on structurally related chemicals is available. Structurally related chemicals are frequently characterized by similar toxicokinetics (Rozman and Klaassen 2001). Urine concentrations of the very closely related chemical 4-chloro-2-fluorophenol were determined in rats given a single oral dose via olive oil gavage (Soffers et al. 1994). About 95% of a single dose was detected in urine within 24 hours of dosing. About 2 - 4 % of the amount in urine was the parent compound; 30 - 37% was the sulfate conjugate (4-chloro-2-fluorophenylsulfate) and 58 - 59% was the glucuronide conjugate (4-chloro-2-fluorophenylglucuronide). This suggests that after a singleexposure to 2-chloro-6-fluorophenol, elimination in the urine is likely to be rapid (i.e., the elimination half-life is short, in this case around 4 - 5 hours), and most of the absorbed dose is eliminated as sulfate and glucuronide conjugates.

Mono-, and dichlorophenols are the chemicals most structurally similar to CFP for which there is some toxicokinetic information in humans. The CFP molecule is structurally the same as mono- and dichlorophenol except that either a hydrogen or a chlorine atom has been replaced by a fluorine atom. Fluorine has characteristics that make it behave similarly to hydrogen and/or chlorine (Park and Kitteringham 1994). These chlorophenols are generally recognized as being well absorbed via all routes of exposure and being nearly completely and rapidly metabolized and excreted in urine (ATSDR, 1999).

This is consistent with what has been observed in humans occupationally exposed to trichlorophenol (Pekari et al. 1991), which is also similar in chemical structure to CFP. Among sawmill workers exposed to trichlorophenol in a wood treatment solution, urine concentrations declined with an elimination half-life of 18 hours, after lumber treatment stopped for the season. Nearly all the trichlorophenol present in workers urine was present as the sulfate conjugate. Although the authors note residual contamination of the workplace with trichlorophenol probably contributed to continuing exposure and that therefore the half-life may be an overestimate,this is not known with certainty. However, based on scientific information available for structurally related chemicals, we expect that CFP would have been fairly quickly eliminated in the urine, assuming no further exposure occurred.

Delayed Elimination as an Explanation for Continued Presence of CFP in Urine Appears Unlikely

The possibility that continued detections of CFP in urine resulted from delays in urinary excretion following the initial exposure on January 5 rather than continued exposure to CFP was also considered. Delayed elimination of chemicals sometimes results from enterohepatic circulation. Enterohepatic circulation occurs when chemicals or metabolites are excreted from the liver into the bile and are then emptied into the intestine where they are either reabsorbed (and go back to the liver) or are excreted in the feces. Glucuronide conjugates, in particular, are often excreted into the bile, hydrolyzed by the enzyme ß-glucuronidase of the gut microflora, reabsorbed as parent compound, and returned to the liver for metabolism. This process tends to slow down appearance of metabolites in the urine and/or decrease their levels.

The possibility that the continued presence of CFP in urine was due to delayed elimination was considered because Bollard et al. (1996) found only about 30% of the structurally related chemical 3-chloro-4-fluorophenol in urine as sulfate and glucuronide conjugates within 8 hours after a single intraperitoneal dose (in saline) in rats. None of the administered dose was detected in urine after eight hours. The authors suggested that the remainder of the dose was not present in urine as either glucuronide or sulfate conjugates, but instead may have been excreted in bile, not reabsorbed in the intestine, and eliminated in feces. However, this hypothesis was not confirmed by analysis of bile or feces. The findings of the Bollard et al. study raise the possibility that chlorofluorophenols might be excreted in bile. If chlorofluorophenols were excreted in bile, reabsorbed and metabolized to metabolites excreted in urine, elimination might be delayed. Therefore, the possibility that elimination of CFP absorbed near the time of the release might be delayed as a result of enterohepatic circulation was thoroughly evaluated.

Although only about 30% of the dose given rats in the Bollard et al. (1996) study was recovered by 8 hours, delayed elimination of CFP is not necessarily indicated by these data. First, if there were delayed elimination, we would be expect that the administered compound would have been detected in urine after eight hours, albeit at lower levels; however, no 3-chloro-4-fluorophenol was detected in urine samples collected after eight hours. Second, chemical characteristics of CFP (e.g., its size (molecular weight) and likely metabolic pathways) suggest it is unlikely to be excreted in bile. As noted above, CFP probably appears in urine as either the glucuronide or sulfate conjugated metabolites (small amounts of parent compound are probably also present) (see e.g., Pekari et al. 1991 (human); Soffers et al. 1994 (rat)). Glucuronide conjugates, especially those with molecular weight greater than 350 (parent compound) are generally more likely to be excreted in bile. Those with molecular weight less than 250 are more likely to be excreted in urine. Sulfate conjugates are more likely to be excreted in urine (Rozman and Klaassen 2001). Since the molecular weight of CFP is about 147 we expect that it would be excreted in urine rather than bile, regardless of whether it is conjugated to sulfate or glucuronide. Third, even if biliary excretion were occurring in humans, it would likely account for a smaller proportion than the extent that occurs in rats because rats are recognized as being "better" biliary excreters than humans. Humans are very poor biliary excretors particularly of low molecular weight compounds (Calabrese 1983; Rozman and Klaassen 2001).

In addition to the above, Pekari et al. (1991) reported that all the trichlorophenol detected in urine of sawmill workers was present as the sulphate conjugate, suggesting again the likelihood that chlorophenol metabolites would be most likely to be excreted in urine and not bile, since sulfate conjugates are nearly always eliminated in urine without entering the enterohepatic circulation (Rozman and Klaassen 2001). Further, the enzyme catalyzing formation of glucuronides (glucuronylsyl transferase) is characterized by low affinity, high capacity, whereas the enzyme catalyzing the formation of sulfate conjugates (sulfotransferase) is characterized by high affinity, low capacity. This means that at low exposure levels, the sulfotransferase (i.e., the high affinity enzyme) would tend to be most active and comparatively more sulfate conjugates would be formed. At higher exposures sulfotransferase would more quickly become saturated. The glucuronylsyl transferase would become more active and comparatively more glucuronides would be formed. If comparatively more sulfate conjugates were formed than glucuronide conjugates in individuals exposed to CFP, then the likelihood that the continued presence of CFP in urine was due to delays and/or decreases in excretion due to enterohepatic circulation is lessened.

We do not know for certain why more 3-chloro-4-fluorophenol was not detected in urine of treated rats. One possible explanation for the apparent lower detections of chlorofluorophenol in urine in the Bollard et al. (1996) study relates to a potential analytical problem. Bollard et al. added enzymes to urine to hydrolyze the glucuronide and sulfate conjugates prior to analysis and quantification. This enzyme catalyzed procedure was performed by incubating urine samples at 37°C for 18 hours. This procedure may have resulted in the loss of analyte since phenols tend to be volatile. There is no information in the assay description in Bollard et al. (1996) indicating that special care was taken to avoid such a possibility. Enzymatic hydrolysis of conjugates in urine was also performed by Soffers et al. (1994), but since 95% of the administered dose was recovered, loss of phenolic metabolites seems to have been avoided.

Another possible explanation for the continued presence of CFP in urine, apart from continued exposure, is that metabolic pathways other than those that conjugate CFP with sulfate or glucuronide might exist and result in metabolic products with a longer residence time in the body than glucuronide or sulfate conjugates. Therefore, alternative metabolic pathways (other than conjugation) for CFP were explored to assess the likelihood that they might exist. Possible alternative metabolic pathways involve the oxidation of CFP by mixed function oxidases to hydroquinones, quinones, or semiquinones, potentially leading to the formation of protein adducts in the liver. Such protein adducts might be stored in the liver until destroyed by proteolysis. At that point, the CFP derivative would be in the form of cysteine conjugates that would then be available for transportation to the kidney. From the kidney, these conjugates could be eliminated in urine as N-acetylcysteine conjugated CFP. Alternatively, cysteine conjugates could be converted by kidney ß-lyase to a thiol (e.g., 2-fluoro-4-mercapto-6-chlorophenol). The 4-sulfhydryl group could then be glucuronidated and the conjugate excreted in urine. None of these urinary products are likely to be detected using the analytical method for urinary CFP developed by NYS DOH Wadsworth laboratories. The N-acetylcysteine conjugate is unlikely to be cleaved by acid hydrolysis. The glucuronidated conjugates would be hydrolyzed to (as an example) 2-fluoro-4-mercapto-6-chlorophenol (mw 178). During gas chromatography - mass spectrometry (GC-MS) analysis, a molecular ion formed from this compound might fragment to give a peak at (m/z⁺ = 145) but no peak at (m/z⁺ = 146). The GC-MS-selective ion mass (SIM) method used for these analyses only detects ions giving a peak at (m/z⁺ = 146), which is the molecular mass (weight) for CFP. Therefore, this possible metabolite would not have been detected and could not account for CFP detected in urine. Moreover, the rate at which such hypothesized storage might occur would probably be very small compared to the expected rate of reactions forming the major glucuronide and sulfate conjugates (Ahlborg and Larsson 1978; Monks et al. 1994; Waidyanatha 1996; Lin et al. 1999; Tsai et al. 2001). Thus it appears unlikely that the CFP detected in urine reflects metabolic products resulting from alternative metabolic pathways.

Finally, the possibility that other chemicals possibly absorbed by those exposed were converted to CFP metabolites subsequently detected in urine was considered. Environmental sampling in January indicated the presence of some other chemicals (referred to as tentatively identified compounds or TICs). The levels of these compounds were generally each lower than those of CFP. There is no information on how these classes of chemicals are changed and excreted in humans, but there are some studies of similar chemicals in animals (Bray et al. 1953; Tulp et al. 1979; Law et al. 1983; Chui et al. 1987; Akhtar 1990; von Meyerinck et al. 1990; Orn and Klasson-Wehler 1998). These studies suggest that minor amounts of urinary CFP might be formed from the TICs depending on the level of exposure and whether or not metabolism of these types of chemicals is similar in animals and humans. However, the potential for these types of chemicals to form CFP appears to be very limited. They appear much more likely to be converted to other metabolites. Since the available analytical methods do not allow for an unambiguous identification of these compounds, evaluation of their potential metabolism is associated with a greater degree of uncertainty than the evaluation of CFP.

Conclusion

Based on the discussion above, repeated detections of CFP in urine appear most likely to be associated with continuing (although gradually decreasing) exposures. This conclusion is supported by the fact that urine levels of CFP for some residents appeared to increase slightly when they returned home after having been away.

Since CFP is likely to be well absorbed via all pathways of exposure and relatively quickly metabolized and eliminated in urine, urinary levels of CFP are probably a reflection of current CFP intake resulting from all possible pathways of exposure (e.g., oral, inhalation, dermal). Therefore, levels of current CFP exposure can be estimated from CFP levels detected in urine. Estimated levels of CFP exposure can then be compared to levels of chlorophenol exposure that may cause adverse effects to evaluate whether a public health concern exists. Additionally, since chlorophenols (especially mono- and dichlorophenols) are likely to be absorbed, metabolized and eliminated in urine like CFP, chlorophenol urine levels associated with chlorophenol exposures having adverse effects can be estimated. Estimated chlorophenol urine levels associated with adverse effects can then be compared to estimated CFP urine levels to evaluate whether a public health concern exists.

ANALYTICAL DETERMINATION OF CFP LEVELS IN HUMAN URINE

At the time of the January 5 release there was no analytical method for the determination of CFP in urine. To determine levels of CFP in urine of individuals potentially impacted by the release, an analytical method had to be developed and validated. NYSDOH laboratories at the Wadsworth Center obtained a pure 2-chloro-6-fluorophenol standard from Diaz Chemical a few days after the chemical release and experiments began with the goal of developing an analytical method to detect trace levels of CFP in urine.

Acid Extraction Method. The initial method developed involved detection of CFP in urine using gas chromatography/mass spectrophotometry (GC/MS) following extraction of CFP from acidified urine. This method involved acidification of urine that released CFP from conjugated sulfate and glucuronide CFP metabolites present in urine followed by solid phase extraction (SPE) to separate CFP and an added internal standard from the urine matrix. The internal standard used was deuterated chlorophenol (chlorophenol with a radioactively labeledhydrogen). Under acid conditions CFP and the internal standard are retained on a C₁₈ SPE cartridge and, after washing to remove urine components, are eluted with organic solvent (dichloromethane). The organic solvent containing the CFP and internal standard is dried by passing through sodium sulfate and is then concentrated to a final volume for GC/MS analysis. Several spiked urine samples were analyzed using this procedure and the method detection limit studies and recoveries were determined. This method was able to quantify levels as low as 1 ng CFP/ml urine. All urine samples provided by Holley residents were analyzed using this acid extraction analytical method. CFP levels in urine were corrected using levels of creatinine in urine (to account for differences in urine volume) and reported in terms of mcg CFP/g creatinine.

Enzymatic Deconjugation Method. Although the acid extraction method described above was releasing measurable amounts of free CFP from urine, there was concern that the sulfate and glucuronide conjugated CFP metabolites were not being fully deconjugated. This would mean that quantification of CFP in urine using the acid extraction method may have underestimated the amount of CFP in urine samples provided by Holley residents. There was also concern that some of the deuterated chlorophenol used as the internal standard was being reduced by deuterium exchange and by vaporization. Therefore, an additional method was developed which used enzymes, rather than acid release, to de-conjugate conjugated CFP metabolites and which used 2-fluoro-4-chlorophenol and 4-fluoro-3-chlorophenol as internal standards.Naturally occurring enzymes which specifically deconjugate sulfate or glucuronide conjugates, and an enzyme preparation which deconjugates both sulfate and glucuronide conjugates were used. Portions of eight urine samples with detectable levels of CFP (based on the acid extraction method), were incubated with either sulfatase, glucuronidase, or the sulfatase/glucuronidase enzymes. CFP was then quantified using GC/MS as in the acid extraction method. CFP levels in urine were corrected using levels of creatinine in urine to account for differences in urine volume and reported in terms of mcg CFP/g creatinine. In all but one case, the enzymatic deconjugation method resulted in more CFP being detected in urine than acid extraction did. CFP levels in urine were about 1 to 8 times greater following enzymatic deconjugation than following acid extraction. Average CFP level after enzymatic conjugation for all eight urine samples evaluated was about four times the CFP level after acid extraction.

ESTIMATION OF CFP EXPOSURE BASED ON URINE LEVELS OF CFP

The highest level of CFP in urine by April 2002 based on the acid extraction method did not exceed 20 mcg/g creatinine for either children or adults. Based on the enzymatic deconjugation method, this level may actually be equivalent to about 160 mcg/g creatinine (20 mcg/g x 8).

By making several assumptions, CFP levels in urine (mcg CFP/g creatinine) can be used to estimate a range of individuals' potential exposures to CFP using the above observations and the following relationship:

Intake (mcg/kg/day) = ((mcg CFP/g creatinine) x (g creatinine/d))/(body weight)

These assumptions are as follows:

• CFP, like mono-, di- and trichlorophenols, is well absorbed via all routes of potential exposure (ATSDR 1999);
• CFP, like 4-chloro-2-flurophenol, is nearly completely metabolized in the liver to glucuronide and sulfate conjugated metabolites which are relatively quickly eliminated in the urine (as discussed above);
• CFP levels in urine at the time of urine collection had reached an equilibrium (or steady state). This means that the rates of absorption, distribution, metabolism, and elimination of CFP were constant and in equilibrium with environmental exposures;
• All the CFP (parent compound) and CFP metabolites present in urine were detected during laboratory analysis;
• Creatinine elimination per day could possibly range from 0.6 to 1.6 g creatinine/day for adults and from 0.12 to 0.33 g creatinine/day for children (Wyngaarden and Smith 1985)
• Body weight for an adult was 50 kg; body weight for a child was 15 kg;
• To ensure that potential exposures are not underestimated, it is assumed that the highest CFP urine level by April 2002 in either adults or children did not exceed 160 mcg/g creatinine.

Estimates of exposure based on these assumptions are summarized in Table F-1. Estimates based on the acid extraction method suggest that by April exposures would not have exceeded about 0.2 to 0.6 mcg/kg/day. Average exposures, among those with detectable CFP in urine, would probably have been about 0.07 to 0.1 mcg/kg/day. Estimates based on the enzymatic deconjugation method suggest that exposures would not have exceeded about 1 to 5 mcg/kg/day. Average exposures, among those with detectable CFP in urine, would probably have been about 0.3 to 0.4 mcg/kg/day.

These estimated ranges do not include the maximum possible exposure that may have occurred immediately after the release. The release occurred on January 5, 2002 whereas the earliest urine samples were obtained on January 15 or later. Additionally, all of the assumptions noted above contribute some uncertainty to these estimates. In particular, if significant amounts of absorbed CFP are not eliminated in urine, or if the analytical method is unable to detect a large percentage of CFP present in urine (e.g., if recovery is markedly lower than 100%) then these estimates may underestimate exposure.

ESTIMATION OF CHLOROPHENOL URINE LEVELS ASSOCIATED WITH TOXICITY IN ANIMAL STUDIES

Animal studies have identified chlorophenol (CP) exposure levels that are associated with adverse effects when experienced over a long time (U.S.EPA 2002 a,b,c). Levels of long-term chlorophenol exposure unlikely to be associated with adverse effects (e.g., RfDs) have also been derived. Additionally, an exposure level associated with an increased lifetime cancer risk of one-in-one-million can be derived for 2,4,6-trichlorophenol (TCP) based on an animal study. If the same assumptions about absorption, metabolism, and elimination of CFP are made for CPs then levels of CP exposure associated with these long term effects can be expressed as CP urine levels using the following relationships.

For non-cancer effects:

For cancer effects:

Estimates of CP urine levels based on these relationships and assumptions are summarized in Tables F-2 (non-cancer effects) and F-3 (cancer effects).

The maximum estimated CFP urine level in Holley residents, determined using the enzymatic deconjugation method (160 mcg/g creatinine), exceeds some estimated urine levels for 2-chlorophenol and 2,4-dichlorophenol corresponding to chronic RfDs. However, urine levels for CP corresponding to the exposures causing adverse effects (LOELs) are two orders of magnitude (one hundred times) or more greater than maximum estimated CFP urine levels in Holley residents. Thus the urinary levels of CFP in Holley residents are not likely to represent exposures that would result in adverse non-cancer health effects.

Estimates of TCP urine levels associated with a one-in-one million cancer risk (2.9 to 7.6 mcg/g creatinine) are 20 times or more lower than a maximum estimated CFP urine level in Holley residents determined with the enzyme deconjugation method (160 mcg/g creatinine). If it is assumed that CFP can cause cancer in humans, has carcinogenic potency equal to 2,4,6-TCP, and that exposures continue for an entire lifetime, these estimates might indicate that a carcinogenic risk might exist. The cancer risk associated with a urine level of about 160 mcg TCP/g creatinine would be about two in one hundred thousand, which is considered to be low. However, there is no evidence that CFP causes cancer in animals or human. Furthermore, a computer-based analysis concluded that CFP did not share chemical characteristics typical of chemicals known to cause cancer and predicted that CFP would not be carcinogenic.

Table F-1. Estimated CFP Intake Based on Urine Level

Estimates based on Acid Extraction Method⁽³⁾

Maximum Exposure

	Surrogate Level Based on Acid Extraction (mcg/g creatinine)	Creatinine Excretion (Range) (g/day) (1)	Body Weight (kg)	Estimated Exposure Based on Acid Extraction (mcg/kg/day) (2)
Child	20	0.12	15	0.2
	20	0.33	15	0.4
Adult	20	0.6	50	0.2
	20	1.6	50	0.6

Average Exposure

	Surrogate Level Based on Acid Extraction (mcg/g creatinine)	Creatinine Excretion (Average) (g/day) (1)	Body Weight (kg)	Estimated Exposure Based on Acid Extraction (mcg/kg/day) (2)
Child	5	0.2	15	0.07
Adult	5	1.0	50	0.09

Estimates Based on Enzymatic Deconjugation⁽⁴⁾

Maximum Exposure

	Surrogate Level Based on Enzymatic Deconjugation (mcg/g creatinine)	Creatinine Excretion (Range) (g/day)(1)	Body Weight (kg)	Estimated Exposure Based on Enzymatic Deconjugation (mcg/kg/day) (2)
Child	160	0.12	15	1.3
	160	0.33	15	3.5
Adult	160	0.6	50	1.9
	160	1.6	50	5.1

Average Exposure

	Surrogate Level Based on Enzymatic Deconjugation (mcg/g creatinine)	Creatinine Excretion (Average) (g/day)(1)	Body Weight (kg)	Estimated Exposure Based on Enzymatic Deconjugation (mcg/kg/day) (2)
Child	20	0.2	15	0.3
Adult	20	1.0	50	0.4

(1) Wyngaarden and Smith (1985)
(2) Intake = ((ug CFP/g creatinine) x (g creatinine/d))/(body weight)
(3) Urine treated with concentrated hydrochloric acid prior to analysis
(4) Urine treated with de-conjugating enzymes prior to analysis.

Table F-2. Estimated Range of 2-Chloro-6-fluorophenol (CFP) Urine Levels "Equivalent" to Chlorophenol (CP) RfD's, NOELs, LOELs

RfD = reference dose
NOEL = no-observed-effect level
LOEL = lowest-observed-effect level
(1) Wyngaarden and Smith (1985)
(2) CFP Urine Level (mcg/g creatinine)= (RfD, NOEL or LOEL (mcg/kg/d) x Body weight (kg)/(g creatinine/d)

Table F-3 Estimates of 2,4,6-Trichlorophenol (TCP) Urine Levels "Equivalent" to an Intake Associated with an Increased Lifetime Cancer Risk of One-in-one Million.

CPF = cancer potency factor for 2,4,6-trichlorophenol.
Calculation on child parameters not included since CPF is based on lifetime exposure of adult animals.

(1) Wyngaarden and Smith (1985)
(2) CFP Urine Level (mcg/g creatinine)= (One-in-one million dose (mcg/kg/d) x Body weight (kg)/(g creatinine/d)

References

Agency for Toxic Substances and Disease Registry (ATSDR). 1999. Toxicological Profile for Chlorophenols.

Ahlborg, UG; Larsson, K. 1978. Metabolism of tetrachlorophenols in the rat. Arch Toxicol. 40:63-74.

Akhtar, M.H. 1990. Fate of 3-phenoxybenzaldehyde: Diphenyl ether cleavage, a major metabolic route in chicken. J. Agric. Food Chem. 38: 1417-1422

Bollard, ME; Holmes, E; Blackledge, CA; Lindon, JC; Wilson, ID; Nicholson, JK. 1996. 1H and 19F-nmr spectroscopic studies on the metabolism and urinary excretion of mono- and disubstituted phenols in the rat. Xenobiotica, 26:255-273.

Bray, H.G., S.P. James, W.V. Thorpe and M.R. Wasdell. 1953. The metabolism of ethers in the rabbit. Biochem. Journal 54: 547-551.

Calabrese, EJ. 1983. Principles of Animal Extrapolation. John Wiley and Sons:NY

Chui, Y.C., R.F. Addison, F.C.P. Law. 1987. Studies on the pharmacokinetics and metabolism of 4-chlorodiphenyl ether in rats. Drug Metab. Depos. 15: 44-50.

Law, F.C.P., Y.Y. Song and S. Chakrabart. 1983. Disposition and metabolism of diphenyl ether in rats. Xenobiotica 13: 627-633.

Lin, P-H; Waidyanatha, S; Pollack, GM; Swenberg, JA; Rappaport, SM. 1999. Dose-specific production of chlorinated quinone and semiquinone adducts in rodent livers following administration of pentachlorophenol. Toxicol. Sci. 47; 126-133.

Monks, TJ; Lo, H-H; Lau SS. 1994. Oxidation and acetylation as determinants of 2-bromocystein-S-ylhydroquinone-mediated nephrotoxicity. Chem Res toxicol 7:495-502.

Orn, U. and E. Klasson-Wehler. 1998. Metabolism of 2,2',4,4'-tetrabromodiphenyl ether in rat and mouse. Xenobiotica 28: 199-211.

Park, K; Kitteringham, NR. 1994. Effects of fluorine substitution on drug metabolism: pharmacological and toxicological implications. Drug Metab Rev 26:605-643.

Pekari, K.; Muotamo, M; Jarvisalo, J; Lindroos, L; Aitio, A. (1991) Urinary excretion of chlorinated phenols in saw-mill workers. Int. Arch Occup Environ Health 63:57-62.

Rozman, KK; Klaassen, DC. 2001. Absorption, Distribution and Excretion of Toxicants. In: Casaret & Doull's Toxicology. The Basic Science of Poisons (Klaasen, CD, ed.). 6^th Edition. McGraw-Hill.

Soffers, AEMF; Veeger, C; Rietjens. 1994. Influence of the type of halogen substituent on in vivo and in vitro phase II metabolism of 2-fluoro-4-halophenol metabolites formed from 3-halo-fluorobenzene. Xenobiotica, 24:759-774.

Tsai, C-H; Lin, P-H; Waidyanatha, S; Rappaport, SM. 2001. Chracterization of metabolic activation of pentachlorophenol to quinones an dsemiquinones in rodent liver. Chem-Biol Interct 134:55-71.

Tulp, M.T.M., G. Sundstrom, L.B.J.M. Matron and O. Hutzinger. 1979. Metabolism of chlorodiphenyl ethers and Irgasan DP 300. Xenobiotica 9:65-77.

U.S. EPAa. 2-chlorophenol (CASRN 95-57-8). Integrated Risk Information System (accessed 2/27/2002)

U.S. EPAb. 2,4-dichlorophenol (CASRN 120-83-2). Integrated Risk Information System (accessed 2/27/2002)

U.S. EPAc. 2,4,5-trichlorophenol (CASRN 95-95-4). Integrated Risk Information System (accessed 2/27/2002)

U.S. EPAd. 2,4,6-trichlorophenol (CASRN 88-06-2). Integrated Risk Information System (accessed 2/21/02)

von Meyerinck, L., B. Hufnagel, A. Schmoldt; H.F. Benthe. 1990. Induction of rat liver microsomal cytochrome P-450 by the pentabromo diphenyl ether Bromkal 70 and half-lives of its components in the adipose tissue. Toxicology 61: 259-274.

Waidyanatha, S; Lin, P-H; Rappaport, SM. 1996. Characterization of chlorinated adducts of hemoglobin and albumin following administration of pentachlorophenol to rats. Chem Res Toxicol 9:647-653.

Wyngaarden, J.B.; Smith, L.H. (Eds.) 1985. Cecil Textbook of Medicine. 17^th Edition. W.B. Saunders Company. Philadelphia PA.

Next Section     Table of Contents

Agency for Toxic Substances and Disease Registry, 1825 Century Blvd, Atlanta, GA 30345
Contact CDC: 800-232-4636 / TTY: 888-232-6348 

Department of Health and Human Services