PRELIMINARY PUBLIC HEALTH ASSESSMENT
FOOTE MINERAL COMPANY
FRAZER, CHESTER COUNTY, PENNSYLVANIA
The tables in this section list the contaminants of concern. We evaluate these contaminants in the following sections of the public health assessment and determine whether exposure to them has public health significance. PADOH selects and discusses these contaminants based upon several factors, including (a) concentration of chemicals on site and off site; (b) comparison of on-site and off-site concentrations with health assessment comparison values for carcinogenic and non-carcinogenic end points, and (c) community health concerns.
Comparison values for health assessments are contaminant concentrations in specific media that are used to select contaminants for further evaluation. These values include Environmental Media Evaluation Guides (EMEGs), Cancer Risk Evaluation Guides (CREGs), and other relevant guidelines. CREGs are estimated contaminant concentrations based on a one excess cancer in a million persons exposed over a lifetime (70 years).
In the data tables under On-Site Contamination and Off-Site Contamination, the fact that a contaminant is listed does not mean that it will cause adverse health effects from exposure. Instead, the list indicates which contaminants will be evaluated further in the public health assessment. When selected as a contaminant of concern in one medium, the contaminant will be reported in all media sampled.
At least three other sites with waste releases exist within a mile of Foote Mineral. They are Transcontinental Gas 200 (benzene, toluene), A.I.W. Frank (trichloroethene, tetrachloroethene), and Malvern TCE (trichloroethene, tetrachloroethene). Also passing near the site and through the area is a petroleum pipeline carrying, according to a pipeline company official, No. 6 fuel (Figure 2).
The Toxic Chemical Release Inventory data base was accessed by PADOH through the National Library of Medicine's Toxicology Data Network and searched for estimated annual releases of toxic chemicals to the environment, from industries within a 2-mile radius of the Foote site to identify possible facilities that could contribute to groundwater and other media contamination near the site. No significant releases which would affect the quality of the groundwater or other media downgradient of the site were reported in the 1987, 1988, and 1989 data bases.
Source Areas - Quarry, Pit, and Pond Borings
The Cyprus Minerals contractor (Roy Weston, Inc.) investigated four waste source areas by performing 20 sediment/soil borings in 1988 and 1989 (F). The source areas were the two quarries, the burn pit, and the lepidolite ponds (Figure 3). The borings yielded approximately forty-two samples for analysis. For the purpose of this preliminary public health assessment, the exposure medium will be classified as sediment (F) because only one soil sample (0-12 inches) contained maximum concentrations of contaminants. In general, sediments contained 2 or 3 orders of magnitude higher concentrations than soil samples.
Groundwater - Monitoring Wells
The Cyprus contractor drilled 15 monitoring wells in November and December 1988, and 11 more wells about a year later. Eight wells were completed in the quarry sediments and the remainder were completed in bedrock. All monitoring wells and the two site production wells (PW2 and PW3, Figure 3) were sampled for metals and volatile organic compounds. Table 2 lists the contaminants detected and their maximum concentrations.
An interesting on-site contaminant distribution pattern emerges. The site topographic map (Malvern, PA Quadrangle and Figure 2) indicates that a surface water divide passes approximately north-south through the middle of the site. It is, therefore, probable that a groundwater divide passes through the site such that groundwater flows westward and eastward from the divide. A 1993 U.S. Geological Survey (U.S.G.S.) report (K) supports this conclusion. Figure 4 indicates which wells contain metals and which contain organic contaminants listed in the tables. No appreciable levels of organic (volatile) compounds have been detected east of production well PW2. Conspicuously absent from the eastern wells are benzene and tetrachloroethene, two soluble compounds which should have appeared in the well array if groundwater flow were eastward beneath the entire site as reported in site documents (E and F). Such an eastward flow would carry volatiles from the solvent burn area (MW22) toward the old quarries and off site. This eastward migration apparently did not occur. Rather, westward flow from the burn pit is probably responsible for the PCE contamination (5 ppb) in monitoring well No. 23 approximately 1,000 feet southwest of MW22.
The only contaminants of concern in groundwater in and near the old quarries are metals, most notably lithium, chromium, boron, and antimony. As shown in Figure 4, lithium and other metals occur almost universally in on-site wells. There is an insufficient number of monitoring wells west of the site to determine if contaminants have moved westward (and how far) with groundwater flow.
In December 1987, the NUS Field Investigation Team (FIT) 3 took surface water samples from both quarries. The south (wet) quarry indicated the presence of lithium (4,870 ppb) and chromium (157 ppb), and the north (dry) quarry aqueous sample indicated the presence of lithium (8,000 ppb), chromium (304 ppb), and antimony (272 ppb). No organic contaminants were detected in either quarry. Today, there is no standing water in the quarries. Some ponded water was noted in excavations during the site visit, but the quality of this water is not known to PADOH. Because the site is on a drainage divide, no through-flowing streams exist. The Cyprus contractor apparently took no on-site surface water samples in the 1988-89 investigation.
Foote Mineral Company began monitoring local groundwater near the plant in the late 1960s. Sampling has continued until the present day with PADER, U.S.G.S., and EPA contractors participating. PADOH has not been furnished historical data or accurate maps showing all sampling locations. However, a clear and consistent pattern of site related metal contamination in groundwater is evident east of the site. Figure 5 shows four of the wells sampled and Table 3 reports the maximum concentrations of contaminants of concern in all public and private wells for which data have been supplied. Data in the table are taken from the U.S.G.S. report of 1987 (J) and the NUS reports of 1988 and 1991 (E and F). The PSWC Chester Valley Well, serving about 17,000 people, is the only off-site well downgradient of the site containing a site-related VOC (TCE at 18 ppb). The distance of this well from the site (6,000 feet), local groundwater flow, and the industrial development in the area strongly suggest a source of contamination unrelated to Foote Mineral. The public well water is treated with an air stripper to remove the volatiles.
Sporadic appearances of toluene and xylene occur, but they are usually near or downgradient of the petroleum pipeline. According to the eyewitness account of a township employee, excavations near the pipeline during routine township maintenance frequently reveal petroleum residue and odor, indicating that leaks have occurred. A detailed account of alleged hydrocarbon contamination of a private well near the pipeline appears on the last page of reference E.
A 1987 U.S.G.S. report (J) states that groundwater contaminated with lithium and boron is being discharged into Valley Creek (Figure 5). The highest concentrations in surface water (780 ppb lithium, 120 ppb boron, 1985 sampling) occurred between the Route 202 overpass and Mill Lane. This reach of Valley Creek includes the tributary draining the eastern part of the Foote Mineral site where lithium and boron wastes were produced.
Surface water samples taken by NUS FIT 3 (1987) on site and immediately downstream of the site (eastward) did not show lithium present above background levels. It is, therefore, concluded that nearly all surface water contamination east of the site is attributable to groundwater discharge to receiving streams along a narrow plume approximated by the unnamed tributary to Valley Creek. No data have been given to PADOH concerning surface water quality west of the site.
In preparing this public health assessment, ATSDR relies on the information provided in the referenced documents. We assume that adequate quality assurance and quality control measures were followed regarding chain-of-custody, laboratory procedures, and data reporting. The analyses, conclusions, and recommendations in this public health assessment are valid only if the referenced documents are complete and reliable.
In the case of Foote Mineral, PADOH and ATSDR believe more surface water and groundwater information are needed for the area west of the site. A detailed summary and description of private wells and well use east of the site for the past thirty years would be helpful in evaluating exposure through groundwater. Such information should include well depths, years of use, contaminant levels, sampling frequency, and the description of any treatment systems used.D. Physical and Other Hazards
Decaying structures present most of the physical hazards on site. Most of the buildings are open to potential explorers, especially children. During the site visit, an abandoned refrigerator (with door attached) was seen. The Cyprus representative promised to have the refrigerator door detached. Several open pits and the quarries present fall hazards to site trespassers.
To determine whether nearby residents are exposed to contaminants migrating from the site, PADOH evaluated the environmental and human components that lead to human exposure. An exposure pathway consists of five elements: a source of contamination, transport through an environmental medium, a point of exposure, a route of human exposure, and an exposed population.
PADOH categorizes an exposure pathway as a completed or potential exposure pathway if the exposure pathway cannot be eliminated. In a completed exposure pathway, the five elements exist and exposure occurred, is occurring, or will occur in the future. In potential exposure pathways, however, at least one of the five elements cannot be identified with available information, but could exist. Potential exposure pathways indicate that exposure to a contaminant could have occurred in the past, could be occurring now, or could occur in the future. An exposure pathway can be eliminated if at least one of the five elements is missing and will never be present.
Contaminants have entered groundwater at the site through past waste disposal practices. The on-site production well (PW2), now abandoned, was an exposure point for workers during early years of operation when contaminants entered the well. Metals, especially lithium, boron, and chromium, have migrated from the site through groundwater to an unknown number of private residential wells. At least four wells, serving an estimated 8-12 people, contained all three metals when sampled in the past. There is uncertainty as to how many homes and businesses are still unconnected to the public water supply since the number is subject to change at any time. Foote's contractor (Weston) sampled over forty wells near and downgradient of the site in 1990. Figure 6 shows the locations of those wells. Citizens using contaminated private wells are exposed to the metals by dermal contact and ingestion. The quality of water treatment, if present, for private water supplies is unknown.
At least one residential well within 650 feet of the site is contaminated with toluene (52 ppb), ethylbenzene (11 ppb), xylenes (260 ppb), and naphthalene (10 ppb) by a source unrelated to the site. Though not used for drinking, residents were exposed to these chemicals through dermal contact and inhalation of the volatile compounds during showering and clothes washing for more than 30 years. In the late 1980s, this home was connected to the PSWC public water supply.
Contaminants have migrated from the site to a public water supply well. The PSWC Chester Valley Well is located on Church Road about 6,000 feet northeast of the site. The well is fitted with an air stripper to remove volatiles (TCE, 18 ppb). However, chromium (18 ppb), boron (200 ppb), and lithium (930 ppb) are not removed. The Chester Valley Well is part of a large water supply system and its water is mixed with that from other wells and surface water. Considerable dilution, therefore, takes place. Nevertheless, exposure to the metals of concern occurs through the use of this public well which serves an estimated 17,000 to 60,000 people (F). Samples taken by the water company from the distribution system in 1992 indicated the presence of a range of lithium concentration from 200-600 ppb (oral communication with company manager).
Contaminants migrate from the site through groundwater and discharge into Valley Creek and a tributary of the creek. People recreating in Valley Creek have been and will be exposed to lithium and boron by dermal contact and incidental ingestion.
Sediment and Soil
Contaminants have been identified in on-site soil and sediment. Site workers and visitors were exposed in the past to metals (Table 1) through incidental ingestion, skin contact, and inhalation of sediment and soil. The few employees who remain may be exposed and will be exposed to sediment and soil contaminants, particularly since there is no water in either quarry to prevent contact with and inhalation of wind blown dust. The exact time at which the quarries became totally dry is not known to PADOH.
To a lesser extent, nearby residents have and will experience some exposure to contaminants through inhalation of airborne dust. Contaminated soil is currently being treated at the site by a biodegradation system. As degradation processes continue, the amount of soil contaminants available for exposure will decrease.
A potential exposure pathway exists from the site through groundwater to wells which could be drilled in the future to serve a receptor population. Should private or public wells be constructed downgradient (east or west) of the site, the narrow geologically controlled plume(s) could be intercepted. Additionally, some drinking water wells, which could be contaminated, may not yet be identified. People using contaminated water wells could be exposed to contaminants through ingestion, direct skin contact, an inhalation of any volatile compounds that may be present in the water.
In this section, we will discuss the likely health effects in persons exposed to specific contaminants, evaluate state and local health data bases, and address specific community health concerns.
To evaluate health effects, either a Minimum Risk Level (MRL) for contaminants developed by ATSDR, or a Reference Dose (RfD) developed by EPA has been used. The MRL is an estimate of daily exposure to a contaminant below which non-cancer adverse health effects are unlikely to occur. The RfD is an estimate of a daily exposure (mg/kg/day) to the general public (including sensitive groups), which is likely to be without an appreciable risk of harmful effects during a lifetime exposure (chronic RfD) or exposure during a limited time interval (subchronic RfD).
The National Toxicology Program (NTP), the International Agency for Research on Cancer (IARC), and EPA have reviewed available information from human and/or animal studies to determine whether certain chemicals are likely to cause cancer in humans. The potential for cancer to occur in an individual or a population is evaluated by estimating the probability of an individual developing cancer over a lifetime as the result of the exposure. EPA has developed cancer slope factors for many carcinogens. A cancer slope factor is an estimate of a chemical's potential for causing cancer. If adequate information about the level of exposure, frequency of exposure, and length of exposure to a particular carcinogen is available, an estimate of excess cancer risk associated with the exposure can be calculated using the cancer slope factor for that carcinogen.
Exposure to site-related lithium has occurred in the past. The maximum documented level of lithium (13,000 ppb) to which people were exposed in the past was detected in water from a private well in 1985. Exposure to lithium is also occurring to some area residents through ingestion of contaminated public well water. The level of lithium, following dilution, in the public water supply distribution system ranges from 200 - 600 ppb. In addition to the lithium in the public water system, residents may also be exposed to lithium from other sources. Lithium is commonly found in plant and animal tissues, and it is estimated that the general population ingests 2 milligrams (mg) of lithium daily from natural sources (D). Also, lithium salts, especially lithium carbonate, are used for the prevention and treatment of affective mental illness. Therapeutically, lithium (lithium carbonate) is used to control manic episodes in manic-depressive illness in doses of 900 - 1,800 mg/d (L). These doses are equivalent to an adult consuming 2 liters of lithium contaminated water at concentrations of 84,500 - 169,000 µg/L. However, lithium treatment is not recommended for patients with significant renal or cardiovascular disease, severe debilitation or dehydration or sodium depletion and to patients receiving other medications (e.g., diuretics) because the risk of lithium toxicity is very high in such patients (L).
Lithium can be measured in people through a blood test. The serum (a part of the blood) is tested, and levels of lithium are measured in units of milliequivalents/liter of blood serum (mEq/L). Desirable serum lithium levels for long-term control are 0.6 to 1.2 mEq/L (a dose of 900 mg/day is likely to maintain this level). The occurrence and severity of toxic reactions are generally directly related to serum lithium concentrations as well as individual sensitivity to lithium, and usually occur more frequently and with greater severity at higher concentration. People who are usually sensitive to lithium may exhibit toxic signs at serum levels of 1 to 1.5 mEq/L. Some of the signs of lithium toxicity include diarrhea, vomiting, tremors, mild ataxia, drowsiness, or muscular weakness (L).
Animal experiments have shown lithium carbonate is a weak teratogen (M). Increased frequencies of cleft palate and fetal loss have been observed among the offspring of mice treated chronically during pregnancy with lithium carbonate. However, studies in rats, rabbits, and monkeys have no evidence of lithium induced teratology (L). In humans, data from lithium birth registries suggest an increase in cardiac and other anomalies, especially Ebsteins anomaly (L). Epidemiologic data are consistent with animal studies and suggest an association between therapeutic doses of lithium and congenital defects in humans (M).
There are several groups of drugs that interact with lithium causing increased levels of lithium in the serum. These include diuretics (e.g., Hydrodiuril), nonsteroidal anti-inflammatory agents (e.g., Motrin), calcium channel blocking agents (e.g., Calan), and angiotensin-converting enzyme inhibitors (e.g., Capoten) (L). Past exposure to lithium (13,000 ppb) in drinking water through private wells by children may have resulted in toxic effects. The concurrent use of any drug associated with the aforementioned groups would increase the likelihood of toxic manifestations.
ATSDR has not developed intermediate or chronic MRLs for ingestion exposure to lithium. However, EPA has established a provisional chronic RfD of 20 µg/kg/day for lithium. Also, to protect the public health, EPA estimated that a lithium concentration in a potable water supply should not exceed 700 µg/L. The concentration of lithium in the blended public water supply system which is distributed to the community is below EPA's recommended level, although the level found in the PSWC well exceeds the recommended level. Using the maximum concentration of lithium (600 ppb) detected in the public water supply, the ingestion exposure for adults (70 kg) does not exceed EPA's provisional chronic RfD of 20 µg/kg/day; however, a 10 kg child's exposure exceeds the provisional value. Because the ingestion exposure dose is below EPA's RfD for adults, non-cancer adverse health effects are not expected. Even though the estimated exposure dose for children is higher than EPA's RfD, it is unlikely that non-cancer health effects will occur. This is based, in part, upon the fact that EPA applied an uncertainty factor of 100 in determining the RfD (10 for the protection of sensitive human subpopulations, in particular pregnant women, and 10 for the use of a LOAEL). At present, based on the review of the available literature, it is determined that exposure to lithium (200-600 ppb) in the public water supply is not likely to cause adverse health effects to the general population. However, effects of those levels on sensitive subpopulations, which include children, pregnant women, people with significant cardiovascular disease, sodium depletion, and people on medications previously discussed, are less understood. Those people may experience an increase in serum lithium levels upon exposure.
Some residents were exposed to boron when they ingested contaminated water from their private wells. Exposure to this chemical is still occurring and may occur in the future. Ingestion is the main route of exposure; however, some dermal absorption can occur.
A review of literature found no evidence that boron causes cancer in animals or humans (B).
A maximum concentration of 20,000 µg/L of boron was detected in 1985 in a residential well. Ingestion of that level of boron exceeds both ATSDR's intermediate MRL and EPA's RfD. However, no adverse health effects have been reported in people exposed to that level of boron (B). No information is available on possible adverse health effects to long-term, low level exposures to boron (B). Some studies indicated that people who accidentally ingested very large amounts of boron (more than 300 times the maximum levels found in private well water) experienced adverse effects to the stomach, intestines, liver, and brain. Animal studies indicated that male reproductive organs, particularly the testes, were affected when large amounts of boron were ingested (a dose more than 50 times the estimated ingestion dose of people drinking maximum concentrations of boron found in well water) (B). Skin and eye irritation developed in rabbits exposed to large amounts of boron for 1 day (B). No subpopulations have been identified as more susceptible to toxic properties than the general population (B). Because exposures to maximum levels of boron detected exceed the intermediate MRL and the RfD, long-term exposure to boron at the highest level detected in private well water may cause adverse, non-cancer health effects.
Antimony, Arsenic, and Chromium
There is no evidence that exposure to antimony has occurred to area residents via ingestion of well water. The potential for exposure to antimony is not likely to occur in the future since antimony was detected only in the on-site water "perched" in the dry quarry fill (1,400 µg/L).
Exposure to arsenic and chromium at low levels has occurred to some area residents through ingestion of contaminated well water. Arsenic and chromium, like most metals, are poorly absorbed through skin and are not volatile organic compounds. Ingestion is the only route of exposure by the area residents. The level of exposure to arsenic and chromium by area residents through ingestion of contaminated well water is not likely to cause any health effects (O,P).
Volatile Organic Compounds (TCE, PCE, and Benzene)
Residents have not been exposed to any of the volatile organic compounds of concern through their private wells. The public water supply was contaminated with TCE (may not be site related), but this is removed before it is supplied to area residents. Therefore, exposure to TCE has been eliminated or reduced. Future exposure to these VOCs may occur if residential wells become contaminated. High levels of benzene (5,300 ppb), PCE (1,400 ppb) and TCE (420 ppb) were detected in the on-site groundwater. Exposure to the contaminants should be avoided. Routine monitoring of downgradient private wells would detect any VOCs that may enter the well water. Connection of residences and businesses to public water supply systems will eliminate the possibility of exposure to these contaminants at levels of concern because the public water supply is treated for VOCs.
One private well contained non-site related contaminants associated with petroleum products (ethylbenzene, toluene, xylenes, and naphthalene). The contaminated well water was used for household purposes, but not used for drinking water, for about 30 years. The levels of the contaminants found in the well water are not expected to result in adverse health effects upon exposure through inhalation and skin contact.
The Foote Mineral site is located in East Whiteland Township, Chester County. Twenty-one years of all cause mortality and cancer mortality (total cancer and eight cancer sites) were collected for East Whiteland Township (G). The 1980-1990 data were analyzed using Pennsylvania's 1984-1986 mortality experience as the standard and the 1990 Census population for age and sex.
Total deaths (all causes) were very close to the calculated expected number - 752 deaths observed and 742 deaths expected, producing a Standard Mortality Ratio (SMR) of 1.013. There were 191 cancer deaths observed for the period and approximately 170 expected (SMR 1.125). An "expected" death is a statistical term used for measuring mortality among a specified population. Neither of these ratios were statistically significant (C). The individual cancer sites analyzed were: (1) buccal cavity and pharynx; (2) digestive system; (3) respiratory system; (4) bone, connective tissue, skin and breast; (5) genitourinary system; (6) other and unspecified sites; (7) leukemia; and (8) other lymphatic and hematopoietic tissues.
Of the eight sites analyzed, only cancer of other and unspecified sites (SMR 1.632) was statistically significantly higher at p < .05. There were 31 cancer deaths in the 1980-1990 period and 19 would be expected.
This cancer surveillance does not indicate an unusual number or distribution of cancer mortality in East Whiteland Township; however, the analysis is not sensitive enough to answer questions concerning health in the immediate environs of the NPL site. Also, Malvern TCE, another NPL site, is located in East Whiteland Township.
Comments received during the public comment release of this document are addressed in the Attachment at the end of this document. We have addressed the community concerns expressed during availability sessions as follows:
1. Is the public water supply safe?We believe that the water is safe to drink. No adverse health effects have been seen in people exposed to the levels of the metals found in the public water supply, but studies of long-term exposures at those levels have not been done. Because some health guidelines (refer to the Toxicological Evaluation section) are exceeded, we feel that further chemical-specific research is needed before we can say conclusively that no adverse effects will occur. The health guidelines (RfDs and MRLs) are very conservative. They are developed to protect people that may be much more sensitive to toxic properties than the general population. Also, the guidelines for boron and lithium were developed mostly from animal studies. Therefore, large safety factors are incorporated into the calculations when deriving health guidelines for people.
VOCs are present in the public water supply, but the water is treated and the contaminants are removed or levels substantially reduced before the water is distributed to the people. The VOCs are not of public health concern as long as the treatment is found to be effective.
2. Will activated charcoal filters on private wells remove harmful chemicals?
Charcoal filters will remove most VOCs (e.g., TCE, PCE, benzene), but not the metals associated with the site. Sand and cartridge filters do not effectively remove the metals, either. Some other types of systems have been shown to be effective for removing some metals. Those systems include reverse osmosis and distillation units.
3. What will happen when (if) 55 gallon drums of PCB transformer oil leak their contents?
Please keep in mind that on-site disposal of 55 gallon drums of PCB transformer oil has not been confirmed. If drums have been discarded in the quarries and if they have leaked (or ever do leak) PCB oils, there will be a serious potential for groundwater contamination on and off the site. Depending upon the density of the oil, the amount of oil, and the width of rock openings (fractures), such oils could sink in the aquifer and be very difficult to remove.
4. What about the home adjacent to the site and across the railroad which was not able to receive public water?
According to all documents and water samples reviewed by PADOH, this home and its well are upgradient of the site (to the north) and unaffected by site contaminants. The contaminated groundwater is flowing away from the well at this home. Therefore, any people using the well water at the home should not be exposed to contaminants from the site that are present in groundwater.
5. What are (were) the consequences of 30 years exposure to petroleum compounds detected in a private well near the petroleum pipeline about 650 feet northeast of the site?
The contaminated well water that contained non-site related petroleum-type products was used for domestic purposes, but not as drinking water, for about 30 years. The levels of the contaminants found in the well water are not likely to result in adverse health effects from exposure through inhalation and skin contact.