Creosote-impregnated Waste Materials - PSL1

2.0 Summary of Information Critical to Assessment of "Toxic"

2.1 Identity, Properties, and Uses

Creosote is a complex and variable mixture produced from coal that is made up of more than 300 compounds. The American Wood Preservation Association describes creosote (CAS Registry Number 8001-58-9) as:

"a distillate of coal-tar produced by high temperature carbonization of bituminous coal; it consists principally of liquid and solid aromatic hydrocarbons and contains appreciable quantities of tar acids and tar bases; it is heavier than water, and has a continuous boiling range of approximately 275°C, beginning at about 175°C" (AWPA, 1977).

There are five major classes of compounds in creosote:

• Aromatic Hydrocarbons including PAHs, alkylated PAHs, benzene, toluene, and xylene (PAHs can constitute up to 90% of creosote);
  
  
• Phenolics including phenols, cresols, xylenols, and naphthols (1 to 3% of creosote);
  
  
• Nitrogen-containing Heterocycles including pyridines, quinolines, acridines, indolines, carbazoles (1 to 3% of creosote);
  
  
• Sulphur-containing Heterocycles including benzothiophenes (1 to 3% of creosote); and
  
  
• Oxygen-containing Heterocycles including dibenzofurans (5 to 7.5% of creosote) (U.S. EPA, 1987).

"Pure" creosote is denser than water. For some wood preservation uses, creosote is mixed 1:1 with fuel oil. In these uses, the density will be less than pure creosote, but will still be heavier than water (Hoffman and Hrudey, 1990). Creosote is insoluble in water (Romanowski et al., 1983), although the components have a wide range of solubilities from the readily soluble tar acids and bases (i.e., phenols, cresols, acridines) to the insoluble six-ring PAHs (i.e., naphtho[2,3-e]pyrene) (CRC Press, 1973; Merck Index, 1976; Clement Int. Corp., 1990a; Syracuse Research Corp., 1989). Creosote is soluble in many organic solvents, including oil and diesel fuel (U.S. EPA, 1984; Bos et al., 1983).

The vapour pressure (Pv) of creosote is variable because of the number of compounds involved and is difficult to characterize. Vapour pressures range from 2.0 x 10-10 Pascals (Pa) for ibenzo[ghi,pqr]chrysene to 11.6 Pa for naphthalene (Clement Int. Corp., 1990). The range of log K_ow values for PAHs is from 3.29 to 7.19 (Clement Int. Corp., 1990). Other components of creosote have widely varying log K_ow values, from 0.65 for pyridine (Leo et al., 1971) to 3.95 for biphenyl (Miller et al., 1985). The range of log K_oc values for PAHs is from 2.97 to 6.74 (Clement Int. Corp., 1990).

Creosote-impregnated waste materials can arise from two separate sources, creosote waste products and creosote-contaminated sites. These sources have been estimated to comprise 71% and 29% by weight, respectively, of CIWM in the Canadian environment (Konasewich et al., 1991).

There are five creosote pressure-treating facilities operating in Canada, two in British Columbia, one in Ontario, and two in Quebec (Konasewich et al., 1991). One facility in Ontario and one in Newfoundland stopped using creosote in 1992 (Constable, 1992). There are also 20 small facilities in Quebec using dip tanks and vapour chambers (Quebec Ministry of the Environment, 1989) and two dip tank facilities in Saskatchewan (Ertman, 1992). These facilities collectively use 21x 10⁶ kg of creosote per year. Preservation of railway ties uses 54% of the creosote, marine pilings use 37%, and bridge deckings, timbers, and utility poles use the remaining 9% (Konasewich et al., 1991).

2.2 Entry into the Environment

Railway ties constitute the largest number of creosote waste products generated in Canada. The major railways decommission 4.5 x 10⁶ ties per year (450 000 m³ of wood) containing an estimated 20.2 x 10⁶ kg of creosote. It is estimated that 90% of all railway ties removed each year are reused. This leaves roughly 2.02 x 10⁶ kg/yr of creosote in discarded railway ties as creosote waste products (Konasewich et al., 1991). Some of the waste ties are burned by railway companies under permits from provincial environment authorities. Little is known about what happens to the rest of the waste-treated wood, although some of it is landfilled. Since the concentrations of PAHs found in waste railway ties vary, generalizations cannot be made about the composition of the CIWM arising from creosote waste products (Sproull and Gurprasad, 1992).

Many of the marine pilings removed from service are also reused. Out-of-service marine pilings and utility poles do not represent a significant source of creosote waste products compared to the volume of creosote waste products from discarded railway ties.

A study of water soluble leachates from out-of-service railway ties found many PAHs and associated compounds. One gram of wood was shaved from the surface of the railway ties and agitated in water for 24 hours. Up to 88.9 µg/L of naphthalene, 92.7 µg/L of dibenzofuran, 120 µg/L of fluorene, 119 µg/L of phenanthrene, and 58.9 µg/L of carbazole were found in the water. Other compounds were detected at lower concentrations (Rotard and Mailahn, 1987). Little other information is available to determine the leaching potential of creosote components remaining in creosote waste products.

Estimates of the amounts of waste creosote entering the Canadian environment from creosote-contaminated sites are not available for many sites. At most of the sites where hydrogeological surveys have been done to track subsurface contamination, however, high levels of compounds from CIWM have been discovered in soil, groundwater, and some surface waters. Estimating quantities of waste creosote at a site and amounts leaching from a site is complex and expensive, and has been attempted in detail at only two sites in Canada. There is an estimated 256 000 m³ of soil that is moderately to highly contaminated with waste creosote at 11 abandoned or operating creosote-treating facilities in Canada (see Table 1). There are at least 13 other potentially contaminated sites in Canada, both operational and non-operational, but no information was obtained on these sites. It is therefore likely that creosote-contaminated sites are a more significant source of waste creosote to the Canadian environment than are creosote waste products, but the data are not available to confirm this.

There are at least 28 creosote-treating facilities in Canada for which site information is not available. The only province that apparently does not have a creosote-contaminated site is Prince Edward Island.

2.3 Exposure-related Information

2.3.1 Fate

Elevated levels of PAHs from CIWM have been found in both Canada and the United States. Almost all of the information gathered pertains to environmental contamination from non-operational wood preservation facilities (i.e., creosote-contaminated sites). Although some studies were found on the transport of PAH components from in-service treated products, it appears that no studies have been undertaken to determine the mobility of creosote components from creosote waste products.

The environmental transport, transformation, and accumulation of the components of creosote are strongly influenced by the components' physical and chemical properties. As a result of the chemical complexity of waste creosote, studies on the behaviour of individual PAHs in the environment may not reflect the behaviour of the compound when it is a component of creosote. Consequently, observed contaminant distributions at creosote facilities are the best indicators of how PAHs from CIWM will behave in the environment (Hoffman and Hrudey, 1990).

At some sites, there appears to have been a mass transport of creosote, usually at sites where the soil was close to being saturated. This may have been due to a combination of gravity and groundwater flow. Polycyclic aromatic hydrocarbon levels in subsoils vary depending on the amount of creosote in the soil, the type of soil, the slope of the land, and the amount of groundwater present. Waste creosote in soil can occur as lighter- and heavier-than-water fractions, and consequently can be found above and below the water table, or even as a free liquid pool. The light fraction includes the nitrogen-, oxygen-, and sulphur-substituted PAHs, naphthalenes, acenaphthene, fluorene, phenols, and hydrocarbons from the oil with which the creosote was diluted. Low molecular weight PAHs are more water soluble than other creosote components and are dissolved and transported in groundwater and surface water. The light fraction has also been found to move with fluctuating water levels and, as a result, can contaminate the whole soil layer. The heavy fraction is indistinguishable from "pure" creosote. It tends to travel downwards until it encounters an impervious layer. It will flow along the impervious interface through the more porous soil in the downslope direction. This environmental behaviour has been responsible for the contamination of surface waters and groundwaters at many creosote-contaminated sites (W.L. Wardrop & Assoc., 1977; Thompson et al., 1978; Ehrlich et al., 1980; Black, 1982; Hickok et al., 1982; Goerlitz et al., 1985; Rostad et al., 1985; Hult and Stark, 1986; Berard, 1988; Coover et al., 1988; Elder and Dresler, 1988; U.S. EPA, 1988; Golder Associates, 1988; 1990b; 1991; Cherry and Smith, 1990; Reitman et al., 1990).

Table 1 Summary of Maximum Polycyclic Aromatic Hydrocarbon Concentrations at Wood Treatment/Storage Sites in Canada

To enlarge table

"Weathered" waste creosote has been found at several contaminated sites in Canada and the United States and consists of PAH components that remain after light components, such as phenol, cresol, naphthalene, phenanthrene, anthracene, and quinoline, have either degraded or been removed through evaporation or dissolution into water. Typically, weathered creosote is composed primarily of the three-, four-, and higher-ring PAHs (Merrill and Wade, 1985; Bieri et al., 1986). At one Canadian site, soil samples contained up to 7200 mg/kg dry weight of phenanthrene and 390 mg/kg dry weight of benzo[a]pyrene (Kieley et al., 1986). Researchers studying creosote-contaminated sediments from the Elizabeth River in Virginia found that, after an initial period of dissolution of the light PAHs, the PAH composition of creosote remaining in the sediments did not change over an 80-year period (Bieri et al., 1986).

In some aquatic systems, sedimentation can isolate creosote-contaminated layers from the water, thereby slowing and eventually halting the dissipation of the more water soluble PAHs. This was noted in studies of sediments from Thunder Bay Harbour where pools of waste creosote were later found with a cap of sediment forming over them (Superior Diving, 1988; Beak Consultants Ltd. and Dominion Soil Investigation Inc., 1988; de Geus, 1990; Pugh, 1989). The same situation was reported for the Elizabeth River in Virginia where creosote was quickly covered with sediment at a rate of 2 cm/ yr (Bieri et al., 1986).

2.3.2 Concentrations

The maximum levels of five PAHs from waste creosote measured in soils, sediments, groundwater, and air are shown in Table 1. The highest PAH concentration in soil at Canadian creosote-contaminated sites was 8700 mg/kg dry weight of naphthalene. Total PAH concentrations in soil have measured up to 39 630 mg/kg of dry weight. Groundwater has been severely contaminated with PAHs at several Canadian sites. Naphthalene has been found in concentrations of up to 66 mg/L and benzo[a]pyrene up to 4 mg/L. Total PAH concentrations in groundwater have been found at up to 950 mg/L, the creosote having displaced much of the groundwater. In Thunder Bay Harbour, pools of creosote have been found containing an estimated 292 m³ of creosote; surrounding the pools was a "creosote globule field approximately 8800 m2 in extent (Beak Consultants Ltd. and Dominion Soil Investigation Inc., 1988; Superior Diving, 1988; O'Connor Assoc., 1989; de Geus, 1990; Reitman et al., 1990; Golder Associates, 1990b).

Although exposure of mussels and lobsters to creosote originating from in-service products was examined in a small number of studies, no exposure information exists on concentrations of waste creosote originating from creosote waste products. In contrast, data for creosote-contaminated sites does exist, and exposure for certain aquatic biota may be established (Table 2). Most of the information presented is from the United States.

With the exception of laboratory studies, no relevant data were found concerning exposure of terrestrial populations to waste creosote or its components arising from creosote-contaminated sites.

Preliminary information from the Domtar Sunalta/Canada Creosote site in Calgary indicates that mountain whitefish (Prosopium williamsoni) are taking in more benzo[a]pyrene and phenanthrene from the site than fish from a control site upstream. Benzo[a]pyrene metabolites were detected in whitefish bile at levels of up to 200 µg/mL while fish from a control area contained 25 µg/mL. Phenanthrene levels in whitefish bile from the contaminated area ranged up to 1600 µg/mL, while whitefish from the control area contained 75 µg/mL. These analyses were done using a fluorescence technique, so the confirmation of the PAHs is tentative (Environmental Management Assoc., 1993).

In the United States, shellfish have been used to show that PAHs (phenanthrene, fluoranthene, pyrene, benzo[a]pyrene, naphthalene, anthracene, and benzo[b]fluoranthene) from creosote wood preservation operations and spill sites are readily accumulated.

Oysters (Crassostrea virginica) taken from a relatively pristine river were exposed to PAH-contaminated sediments near a creosote wood-treating facility on the Elizabeth River in Virginia. After three days, oysters had accumulated total PAHs from non-detectable levels to 10.1 to 11.7 µg/g wet weight (Pittinger et al., 1985). Clams (Rangia cuneata) exposed to water flowing over a creosote spill near Bayou Bonfouca, Louisiana accumulated approximately five times more anthracene, five times more fluoranthene, thirty times more benzo[a]pyrene, and twice as much naphthalene and phenanthrene as did clams upstream from the spill (DeLeon et al., 1988). Snails (Thais haemastoma) taken downstream from a creek running through a creosote-contaminated site in Florida had accumulated fluoranthene and phenanthrene to levels significantly higher than snails from an uncontaminated area. Only fluoranthene was accumulated in oysters (C. virginica) from the same area (Elder and Dresler, 1988).

Sediment-associated PAHs from creosote-contaminated sites have been shown to accumulate in English sole (Parophrys vetulus) (Malins et al., 1985), guppies (Poecilia reticulata) (Schoor et al., 1991), brown trout (Salmo trutta), lamprey (species unknown), and white suckers (Catostomus commersoni) (Black et al., 1980). Concentrations of metabolites of PAHs in bile from English sole were at levels roughly 40 times greater than English sole from an unpolluted area (2.10 µg/g wet weight versus 0.067 µg/g wet weight) (Malins et al., 1985). Guppies exposed to creosote-contaminated sediments accumulated significant quantities of anthracene, fluoranthene, benz[b]fluoranthene, benz[a]pyrene, and phenanthrene in their muscle tissues (Schoor et al., 1991). Brown trout and white suckers from the creosote-contaminated Hersey River in Michigan had tissue levels of phenanthrene roughly an order of magnitude greater than unexposed fish, and lamprey had levels of phenanthrene over two orders of magnitude greater than their unexposed counterparts (Black et al., 1980). Benthic insects (species unknown) taken from the Hersey River concentrated phenanthrene to levels slightly higher than those found in sediments (5.49 vs 4.10 µg/g wet weight) at the contaminated site and were found to have benzo[a]pyrene concentrations 725 times greater than non-exposed insects (Black et al., 1980).

Table 2 Concentrations of Selected Polycyclic Aromatic Hydrocarbons in Biota from Canadian and American Creosote-contaminated Sites

To enlarge table

These studies show that aquatic organisms living close to creosote-contaminated sites absorb PAHs above the background concentrations found elsewhere. The types of species that are likely to absorb PAHs are those that are in intimate contact with the sediments or those that feed on species in contact with sediments.

2.4 Effects-related Information

Ecotoxicology. No information was found on the toxicity of waste creosote to aquatic or terrestrial biota resulting from creosote waste products.

The exposure information presented in Subsection 2.3.2 and the effects data in this subsection are the only available data for Canadian sites. The characterization of Canadian sites has generally not included biotic environmental exposure and effects data, except at the Northern Wood Preservers (NWP) site in Thunder Bay and the Domtar Sunalta/Canada Creosote site in Calgary. Since Canadian environmental effects data are limited, data from the United States are also presented.

Gross autopsies of whitefish taken from the Bow River near the Domtar Sunalta/Canada Creosote site did not detect any abnormalities. Plecoptera (stoneflies) Trichoptera (caddisflies) and chironomids (midges) were generally absent from an area on the south side of the Bow River approximately 1 km in length. They were largely replaced by Gastropods (snails) and Diptera (crane flies) that are known to be more tolerant of pollutants (Environmental Management Assoc., 1993). Undiluted pore water taken from Bow River sediments from the cut-off berm to 250 m downstream from the berm was determined to be toxic in 19 of 45 samples using the Microtox test (endpoint was a 20% reduction in light generation) (Shaw, 1992). These "toxic" pore water samples were associated with a "creosote odour" and the presence of shallow Non-Aqueous Phase Liquid (i.e., waste creosote) (Hamilton, 1992).

In 1986, benthos distribution surveys were done at the Northern Wood Preservers facility in Thunder Bay Harbour. The results of these surveys indicated that benthic habitat in the vicinity of the facility wharf was degraded, with the most severe degradation located closest to the wharf. Habitat alteration, sediment contamination with PAHs, and organic enrichment caused reduced diversity of benthic invertebrates and increased the dominance of sludgeworms (Beak Consultants Limited and Dominion Soil Investigation Inc., 1988). A bioassay study on the sediments indicated that they were lethal to aquatic organisms (leeches, fathead minnows, and mayflies). Sediments from stations near the facility were lethal to all species during a 10-day exposure study; sediments from other locations were non-lethal, indicating the existence of a 150-m wide toxic zone emanating from the facility (Metcalf and Hayton, 1989).

Studies exposing the amphipod Rhepoxynius abronius to aerated water from the sediments of Eagle Harbor, Washington showed that none survived a four-day exposure at sediment water concentrations of 5%. Most of the amphipods exposed to undiluted sediment water immediately displayed abnormal swimming behaviour, a few managed to burrow into the sediment, and all died within 10 to 60 minutes (Swartz et al., 1989).

Long-term, or chronic effects of waste creosote-contaminated sediments have been observed in populations of English sole from Eagle Harbor and Puget Sound, Washington (Maims et al., 1985; 1988; Myers et al., 1987; 1990; Stein et al., 1990). There is strong evidence that the high rates of abnormal alterations of the liver, including tumors and cancers, found in the fish inhabiting the waste creosote-contaminated areas of Eagle Harbor are the result of exposure to PAH-contaminated sediments. The abnormal alterations observed in the livers of English sole closely parallel the changes and tumors that have been induced experimentally in the rat, mouse, and in certain fish by chemicals that are known to be toxic to their livers (Myers et al., 1990).

A study was conducted to determine whether contaminant exposure was associated with altered ovarian development in English sole from four areas of Puget Sound, including Eagle Harbor (Johnson et al., 1988). The results suggested that exposure to PAHs from Eagle Harbor had a significant effect on reproductive processes in English sole. Polycyclic aromatic hydrocarbons appeared to be most closely associated with inhibited ovarian development and depressed blood hormone levels in these fish. There is evidence that populations of English sole in the Sound have been declining, but the role of contaminant exposure in this decline is not known (Johnson et al., 1988).

A study conducted in the Elizabeth River in Virginia found that mummichog (Fundulus heteroclitus) had a very high prevalence of liver cancer in a population located at a site contaminated with waste creosote. Grossly visible liver lesions were present in 93% of the fish and 33% had liver cancers. Sediment PAH concentrations were 2200 mg/kg dry weight. Mummichog at two other sites having low levels of PAHs (730 and 35 times less) in the sediments showed no such indication of disease (Vogelbein et al., 1990).

Macrophages from oyster toadfish (Opsanus tau) taken from the creosote-contaminated Elizabeth River in Virginia have altered abilities to migrate towards bacteria, to engulf them, and to generate reactive oxygen species required for the degradation of engulfed material, as compared with macrophages of fish taken from the relatively non-polluted York River, Virginia. Adult oyster toadfish and sediments were sampled from four locations in the Elizabeth River. The sediment-bound PAH levels were highest near an operating creosote wood-treatment facility. Macrophage function was also most severely affected at this site (Seeley and Weeks-Perkins, 1991).

The effects of an eight-day exposure of the bottom-feeding fish, Leiostomus xanthurus, to waste creosote-contaminated sediments from the Elizabeth River have been studied under laboratory conditions (Hargis Jr. et al., 1984; Roberts et al., 1989). Exposed fish developed skin lesions, pancreatic and liver alterations, and experienced reduced weight gain, reduced numbers of red blood cells, and increased mortality. No effects were observed in fish exposed to clean sediment. Analyses of these sediments showed heavy contamination with PAHs compared to uncontaminated sediment controls. Phenanthrene and fluoranthene were the two most abundant PAHs in the sediment, each accounting for 5 to 12% of the total PAH load. Benzo[a]pyrene was detected at 43 mg/kg dry weight in the Elizabeth River sediments compared to 0.009 mg/kg dry weight in the uncontaminated sediments. L. xanthurus are largely bottom-feeders, actively agitating the surface of the sediments with their fins and body movements while foraging. This action would account for the observed high incidence of severe fin erosion of the pectoral, caudal, and pelvic fins and dilation of the blood vessels around fins in those fish exposed to Elizabeth River sediments (Hargis Jr. et al., 1984).

Another approach for estimating effects of pollutants on benthic dwelling organisms is to correlate known effects from polluted areas with the concentrations of pollutants in the sediments. The United States National Oceanic and Atmospheric Administration (NOAA) uses the apparent effects threshold (AET) approach to estimate biological effects from sediment-associated PAHs on marine organisms (NOAA, 1990). An AET is defined as the lowest concentration of a compound in sediment at which biological effects (usually changes in composition of benthic invertebrate communities) are observed to occur. The Ontario Ministry of the Environment (OMOE) uses Sediment Quality Guidelines for pollutants in sediments, including total PAHs, to estimate biological effects levels (Persaud et al., 1992). The data generated by the NOAA on creosote-associated effects on marine organisms and the OMOE data on total PAHs can be used to estimate a relationship between PAH sediment concentrations and potential effects on Canadian aquatic ecosystems where these sediments are found. The marine effects data may not be directly comparable to Canadian freshwater ecosystems, but they do at least provide a criterion against which to judge the potential effects of CIWM in freshwater sediments. Table 3 lists the AET concentrations of many PAHs in marine sediments on a dry weight basis and compares them with maximum PAH concentrations from around the Northern Wood Preservers dock in 1988. The Ontario Ministry of the Environment's tentative Lowest-Effect-Level for total PAHs in sediment is 2 mg/kg dry weight, and their tentative Severe-Effect-Level is 11 000 mg/kg dry weight (Persaud et al., 1992). The AET for total PAHs is 22 mg/kg dry weight (NOAA, 1990). In 1984, sediments in Thunder Bay Harbour close to the Northern Wood Preservers facility, contained 26 388 mg/kg dry weight of total PAHs (Berard and Tseng, 1986). Ecological impacts observed at the Northern Wood Preservers facility (reduced benthic diversity and a community shift towards Oligochaetes) would be expected at the concentrations of PAHs present.

No mammalian toxicology data were identified for creosote-impregnated waste materials. Toxicity data for creosote have been derived for mammals from laboratory exposure tests, but little data exists for environmental exposures or effects to mammals outside of this context. In many cases, testing has been performed on organic extracts of creosote, creosote wastes, or contaminated sediments. This laboratory data will not be used to estimate the toxicity of CIWM to wildlife as it does not sufficiently resemble the exposure to wildlife.

Table 3 Summary of Apparent Effects Threshold Concentrations for Polycylic Aromatic Hydrocarbons Sorbed to Marine Sediments and Polycyclic Aromatic Hydrocarbon Concentrations Found at Northern Wood Preservers, Thunder Bay, Ontario, 1984

PAH	AET Concentration* (mg/kg dry weight)	Maximum Sediment Concentration at NWP** (mg/kg dry weight)
Acenaphthene	0.150	15
Anthracene	0.300	120
Benzo[a]anthracene	0.550	600
Benzo[a]pyrene	0.700	450
Chrysene	0.900	600
Dibenz[a,h]anthracene	0.100	61
Fluoranthene	1.000	780
Fluorene	0.350	25
2-Methylnaphthalene	0.300	NA
Naphthalene	0.500	75
Phenanthrene	0.260	250
Pyrene	1.000	338
Total PAHs***	220.0	4331a 26 388b

* adapted from NOAA (1990)s

** Beak Consultants Ltd. and Dominion Soil Investigation Inc. (1988)

*** Total PAHs present in NWP sediments only

^a Total PAHs in 1984

^b Total PAHs in 1986