Page 6: Guidelines for Canadian Drinking Water Quality: Guideline Technical Document – Vinyl Chloride

Part II. Science and Technical Considerations (continued)

5.0 Exposure

The general population is primarily exposed to vinyl chloride from inhalation of ambient air and the ingestion of items packaged in PVC containers, from which vinyl chloride can leach (ATSDR, 2006). Canadian data on vinyl chloride levels in sediment or sewage were not located. No published quantitative data on vinyl chloride exposure from plastic toys are available.

5.1 Water

In a national survey of 30 Canadian water treatment facilities conducted in 1979, vinyl chloride was detected at < 1 µg/L in one sample each of treated and raw water collected in the months of November and December but was not present in samples collected in August or September (Otson et al., 1982).

In Alberta, out of 1622 samples collected since January 1, 2000, only two detections of vinyl chloride were reported from provincial parks treated drinking water; both detections (0.11 and 0.5 µg/L) were at or near the method detection limit (MDL), with the "largest" MDL in the data being 1.0 µg/L (Alberta Environment, 2009).

In New Brunswick, vinyl chloride was not detected in any drinking water supplies from 1994 to 2008 (New Brunswick Department of Environment, 2009).

In Nova Scotia, testing was conducted between March 2002 and April 2008 in 47 of 85 municipal drinking water treatment facilities; a total of 202 "non-detect" results for vinyl chloride were reported, which included 102 raw and 100 treated samples. The MDL ranged from 0.17 µg/L to 1 µg/L (Nova Scotia Department of Environment, 2009).

In Quebec, during the period of 2001-2009, 4824 samples from 276 distribution systems (mostly serving more than 5000 people) were analysed for vinyl chloride. In total, 4807 results (99.6%) were reported as being below the MDL, which ranged between 0.02 and 0.6 µg/L, depending on the laboratory; 17 samples contained vinyl chloride at levels ranging between 0.04 and 3.0 µg/L, with a mean of 0.75 µg/L (Ministère du Développement durable, de l'Environnement et des Parcs du Québec , 2009).

In Saskatchewan, during the period of 1996-2008, no vinyl chloride was detected in 25 samples from the distribution system (MDL ranged from 0.4 to 1 µg/L) (Saskatchewan Ministry of Environment, 2009).

The First Nations and Inuit Health Branch of Health Canada reported vinyl chloride sampling results for both raw and treated water from four provinces (Health Canada, 2009). In British Columbia, no vinyl chloride was detected in 10 raw water samples (MDL ranged from 0.5 to 1 µg/L) for the period of 2002-2008; however, vinyl chloride was detected in 2 of 920 treated samples at 0.61 and 0.54 µg/L. For Alberta, from 2004 to 2008, vinyl chloride was not detected in 177 treated drinking water samples; however, it was detected in 1 of 30 raw water samples at a level below 2 µg/L (MDL ranged from 0.2 to 2 µg/L). In Ontario, vinyl chloride was not detected in 424 raw water samples (MDL of 0.2 µg/L); however, 1 detection (0.5 µg/L) out of 1072 samples was reported for treated drinking water during the period of 2001-2009. Finally, in Manitoba, from 2001 to 2009, no vinyl chloride was detected in 460 samples of raw water (MDL ranged from 0.2 to 2 µg/L); however, 7 samples of treated drinking water out of 700 contained vinyl chloride at levels ranging from 0.9 to 1.6 µg/L.

Environment Canada (1998) monitored vinyl chloride levels in surface waters and groundwaters in eastern Canada (Nova Scotia, Prince Edward Island, Newfoundland and New Brunswick). Between 1985 and 1988, vinyl chloride levels were measured in lakes (n = 86), ponds (n = 17), reservoirs (n = 29) and rivers and streams (n = 50). The average vinyl chloride concentration in each of these surface water systems was below the MDL of 5 µg/L (Environment Canada, 1998). In 141 groundwater samples from various locations throughout the provinces, vinyl chloride levels were below the MDL of 5 µg/L in all samples (Environment Canada, 1998).

5.1.1 Leaching from Polyvinyl Chloride Pipes

Currently, PVC and chlorinated PVC (CPVC) plastic pipes and components are used for conveying potable water. In the polymerization process, some of the vinyl chloride monomers are retained in the plastic pipe matrix (PVC or CPVC) as a residue and may be released into the air or water (WHO, 2004; Richardson and Edwards, 2009). In early field investigation (Dressman and McFarren, 1978) and experimental studies (Banzer, 1979; Ando and Sayato, 1984) demonstrated the ability of residual vinyl chloride to migrate from water distribution PVC pipes into the water flowing through them. The extent of leaching is determined by the vinyl chloride concentrations in the pipe material. In the field study, the highest vinyl chloride concentrations (1.4 μg/L) consistently occurred in water from pipes manufactured in 1975 whereas the lowest level (0.03 μg/L) was found in the pipes manufactured in 1966 (Dressman and McFarren, 1978). The amount of vinyl chloride migrating from rigid PVC pipes into drinking water was found to be directly proportional to the residual level of vinyl chloride in the pipe itself. Older PVC pipes (pre-1977) may contain vinyl chloride in concentrations up to 600 mg/kg, and have been shown to leach vinyl chloride into drinking water (Beardsley and Adams, 2003).

There are several solutions to address vinyl chloride leaching from the pre-1977 PVC pipes that are still in use today, including replacement of pipes, use of drinking water treatment devices and implementation of flushing protocol. Beardsley and Adams (2003) determined that the development and implementation of an appropriate flushing protocol is a practical approach to reduce exposure from vinyl chloride in these pipes. PVC is generally used for cold water applications for distribution system and premise plumbing pipes and components (i.e., fittings). The National Plumbing Code of Canada (NPC) allows the use of PVC only for cold water plumbing applications (NRCC, 2010). CPVC has higher heat resistance and greater mechanical stability from the chlorination process. Under the NPC, CPVC is permitted for use in both hot and cold water applications (NRCC, 2010). It is also important to note that the NPC has required that all plastic pipe meet the Canadian Standards Association's (CSA) standard for plastic pipes (CSA, 2009). The CSA standard was revised in 1990 to include a leaching test for vinyl chloride into drinking water. The CSA B137 series of standards for thermoplastic pressure piping currently requires that PVC and CPVC pipes and components (e.g. tubing and fittings), used for drinking water applications, comply with the requirements of NSF/ANSI Standard 61 (CSA, 2009). In order for PVC and CPVC products and materials to be certified to NSF/ANSI Standard 61, they must contain 3.2 mg/kg or less residual vinyl chloride monomer in the product wall, which is equivalent to 0.2 µg/L of vinyl chloride diffused into water. This acceptance criterion is the single product allowable concentration (SPAC), which is derived using the contaminant regulatory values from the U.S. EPA and Health Canada (one tenth of the regulatory value of 2 µg/L) (NSF/ANSI, 2012). Pipes and components meeting NSF/ANSI Standard 61 would be expected to leach very low concentrations of vinyl chloride into drinking water.

Richardson and Edwards (2009) investigated the leaching and accumulation of vinyl chloride from static segment/fragment reactors from PVC and CPVC piping used in drinking water applications. Although the results indicated that the newer PVC reactors leached faster and accumulated higher concentrations than CPVC, the levels of vinyl chloride concentrations leached from PVC and CPVC reactors were very low. The results from this study are presented in greater detail in section 7.1.5.

When PVC pipes were exposed to sunlight at maximum ambient temperatures of 35ºC the pipes were found to release more than 2.5 µg/L of vinyl chloride into the water. In the absence of sunlight, no vinyl chloride was released which indicates that environmental conditions such heat and UV light can adversely affect vinyl chloride release (Al Malack and Sheikheldin, 2001).

5.2 Food

Vinyl chloride has been shown to migrate from PVC packaging to bottled drinking water. Benfenati et al. (1991) found that progressive migration of vinyl chloride from the bottle to the water occurred at a rate of 0.001 µg/L per day. Vinyl chloride levels of 0.013-0.083 µg/L (mean of 0.048 µg/L) were measured in Italian drinking water bottled in PVC (Benfenati et al., 1991).

There is very little information on concentrations of vinyl chloride in food. The leaching of vinyl chloride from wet food packaging materials is low; however, vinyl chloride is soluble in alcohols and mineral oil. The Canadian Food and Drug Regulations prohibit the sale of any food packaged in materials that leach detectable residues of vinyl chloride monomer, as determined by an analytical method with a detection limit of 50 µg/kg (Government of Canada, 2011). Vinyl chloride intake via food and drinks has been estimated to be 0.1 µg/day (Purchase et al., 1985).

5.3 Air

In Alberta, 24-hour average concentrations of vinyl chloride at the perimeter of a vinyl chloride plant from 1979 to 1984 were 12.9 µg/m3 (5 ppb) 86-98% of the time; 12.9-23.2 µg/m3 (5-9 ppb) 0.3-4.9% of the time; 25.8-77.4 µg/m3 (10-30 ppb) 0.1-6.5% of the time; and 77.4 µg/m3 (30 ppb) 0.3-2.4% of the time (Environment Canada, 1986). The general public may be exposed to small amounts of vinyl chloride from inhalation of ambient air in urban areas, typically in the order of 5 µg/day per person (ECETOC, 1988), with higher amounts in the vicinity of vinyl chloride and PVC plants.

Of 2560 samples collected at 47 sites in Canada in 2002, vinyl chloride was generally not detected (< 0.05 µg/m3) in 24-hour samples (Environment Canada, 2004). A maximum 24-hour concentration of 0.7 µg/m3 was reported for Sarnia, Ontario; Sarnia is an urban area with a known industrial point source. Based on annual outdoor air data from Montreal, Quebec, the average 24-hour vinyl chloride concentration was 0.02 µg/m3 for 2001 and 2002 (four sites, eight samples) (Environment Canada, 2002).

National Air Pollution Surveillance Network stations measured concentrations of vinyl chloride in ambient air across Canada (Environment Canada, 2006). Between 2003 and 2006, nearly 99% of all average vinyl chloride concentrations for both rural (4379 of 4447 samples) and urban centres (7787 of 7879 samples) were reported to be below the MDL of 0.02 µg/m3.

Health Canada measured vinyl chloride levels inside 100 non-smoking homes in Windsor, Ontario, from 2005 to 2006 (Stocco et al., 2008). The mean indoor vinyl chloride concentration was 0.062 µg/m3 in the summer and 0.028 µg/m3 in the winter. Outdoor levels of vinyl chloride were lower than indoor levels in summer and winter. The mean exposure concentration was determined to be 0.01 µg/m3 in winter and < 0.001 µg/m3 in summer (Stocco et al., 2008).

In a study by Dawson and McAlary (2009), vinyl chloride was detected in 7% of a total of 1684 North American homes sampled; concentrations up to 1.3 µg/m3 were detected (with reporting limits ranging from 0.01 to 1.3 µg/m3).

In the United States, indoor air concentrations of vinyl chloride in houses close to landfills have reached concentrations of up to 1 mg/m3 and have exceeded the maximum reported concentration in outdoor air for areas adjacent to the landfills. A California monitoring program that collected 500 samples at two outdoor and four indoor sites downwind of a landfill reported that the 120 samples containing the highest vinyl chloride concentration (25 μg/m3) were taken inside homes (Little et al., 1992). An average vinyl chloride concentration of 4 ppb (10.4 μg/m3) and a maximum of 9.3 ppb (24.2 μg/m3) were measured in over 500 samples from 69 homes near a co-disposal landfill site in California, while concentrations over the site ranged from 83 to 12,800 ppm (216-33,280 mg/m3) (Stephens et al., 1986).

5.4 Soil

To establish local background conditions, the Ontario Ministry of the Environment measured the vinyl chloride concentration in soil samples collected from sites with no known point sources (OME, 1997). A vinyl chloride concentration of 0.003 µg/g was determined to be the typical background level in Ontario for agricultural and all other land uses, based on the 90th percentile concentration measured in soil samples.

5.5 Multi-Route Exposure Through Drinking Water

A human physiologically based pharmacokinetic (PBPK) model based on Clewell et al. (2001, 2004) was developed to extrapolate the results from two high-dose oral exposure studies in rats (Feron et al., 1981; Til et al. 1991) to humans exposed to low concentrations of vinyl chloride in drinking water. The model was also used to estimate the litre-equivalent (L-eq) contributions from dermal and inhalation exposure to vinyl chloride when showering and bathing. Using the external doses generated from the human PBPK model (see Sections 8.4 and 10.1), litre-equivalent contributions from dermal and inhalation exposure during showering or bathing were estimated by running the human PBPK model for a 30-minute bathing scenario. By comparing the internal doses generated from dermal and inhalation routes of exposure with the internal dose from ingestion, the litre-equivalent contributions for dermal and inhalation exposure were determined to be 1.9 L-eq and 0.4 L-eq, respectively. When added to the standard Canadian drinking water consumption rate of 1.5 L/day, the total litre-equivalent daily exposure to vinyl chloride in drinking water was estimated to be 3.8 L-eq. This litre-equivalent daily exposure was used in both the cancer and non-cancer risk assessments in Section 10.

Page details

Date modified: