Environ. Sci. Technol. 2001, 35, 4170-4173
Gaseous Elemental Mercury as an Indoor Air Pollutant ANTHONY CARPI* AND YUNG-FOU CHEN John Jay College, The City University of New York, 445 West 59th Street, New York, New York 10019
Mercury is not commonly considered a household air pollutant; however, a number of potential sources of the metal exist in residential settings. Eleven of 12 indoor sites sampled in this study showed levels of airborne mercury that were significantly elevated over outdoor concentrations (range 6.5-523 ng m-3). In addition, this and other published research suggest that up to 10% of households may have levels of airborne mercury above the U.S. EPA reference concentration (300 ng m-3) due to historic accidents with mercury containing devices. Exposure to mercury via indoor air is seen as second only to fish consumption as a source of mercury in the general population. Large seasonal changes in indoor mercury levels were identified in this study suggesting that short-term monitoring of mercury-contaminated sites is not sufficient to adequately assess the potential health risks and effectiveness of remediation strategies.
Introduction While the adverse health effects associated with exposure to mercury have been known since antiquity, the unique chemical and physical properties of the metal have entrenched its use in modern society. As a result, mercury contamination in the environment is widespread. Mercury pollution is the leading cause of health advisories on fishing resources in the United States, and it is the only pollutant for which the number of advisories continues to increase (1). Fish consumption is the primary source of exposure to mercury in the general population of the United States and many other countries because of the high bioaccumulation rate of methyl mercury in the aquatic food chain. Exposure to mercury in drinking water is minor, and exposure to mercury via inhalation is thought to be insignificant due to the low levels of mercury in outside air. Aside from isolated, occupational exposures, exposure to mercury in indoor air has been assumed to be relatively minor. However, indoor air pollution is increasingly garnering attention in the United States, and a number of potential sources of mercury exist in residential settings that raise questions as to their significance (2). Elemental mercury (Hg0) is readily absorbed in the respiratory tract and can adversely affect the central nervous system resulting in symptoms including tremors, increased excitability, and delirium. Unique to the heavy metals, Hg0 has a relatively high vapor pressure (∼1.1 × 10-3 Torr at 20 °C), and the saturated atmosphere concentration (∼12 mg m-3 at 20 °C) is almost 3 orders of magnitude greater than the time-weighted average threshold limit value (0.025 mg m-3) for occupational exposure (3, 4). Thus any source of Hg0 in a confined indoor * Corresponding author phone: (212)237-8944; fax: (413)431-1654; e-mail:
[email protected]. 4170
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 21, 2001
environment can result in airborne concentrations that raise significant health concerns. Despite the common use of mercury in household products, little research exists regarding the relevance of this potential source of indoor air pollution. Common sources of mercury in residential settings can be separated into two loosely defined categories: materials that contain salts of mercury either as an intentional additive or an accidental contaminant and devices that contain free Hg0. Within this first category, indoor latex paints represent one of the most prominent examples of mercury sources. Phenylmercuric acetate and other mercury compounds were common additives in latex paints through the late 1980s because of their efficacy as fungicides and bactericides (5). To a lesser extent, contact lens solutions, nasal sprays, and other home medications have been manufactured with trace concentrations of phenylmercuric acetate or phenylmercuric nitrate to inhibit microbial growth. Mercury is commonly found as a contaminant in many alkali-based detergents and cleansers because of the extensive use of mercury electrodes in the chlor-alkali industry (6). Similarly, chlorine-based cleansers and household products may contain mercury as a contaminant. While these materials contain mercury salts, volatile Hg0 may be formed either during the manufacture of these materials or due to the decomposition of the compounds (7). Free Hg0, because of its unique physical and chemical properties, has also been commonly used in a range of household devices. Among the most recognizable of these is the common mercury body temperature thermometer. Significant amounts of mercury are also used in fluorescent light bulbs, electrical tilt switches commonly used in household thermostats, float controls in sump pumps, barometers, and gas flow meters. These devices generally include mercury in a self-contained reservoir, unfortunately these containers are commonly made of glass and can be broken easily thus releasing the liquid metal into the residential environment. Mercury can adsorb onto a number of common household surfaces, thus preventing complete clean up and removal of the metal following an accidental spill (8). A number of recent events highlight the potential significance of indoor mercury contamination. Increasingly, primary and secondary schools have reported unsafe conditions resulting from accidental spills of liquid mercury (9). The Nicor power company of Chicago is currently inspecting over 200 000 households because of suspect mercury spills resulting from the routine replacement of natural gas meters (10). Several municipalities, including Boston, San Francisco, and Duluth, have banned the sale of mercury thermometers altogether as a result of potential health risks, and the state of New York is currently considering proposed legislation to limit mercury use in all consumer products (11). Despite the significance of this prospective problem, little data exists on the potential for household products and devices to contaminate indoor air with Hg0. In an effort to examine the potential for mercury exposure in common living environments, we conducted sampling for indoor Hg0 in residential and business dwellings from June 2000 through March 2001.
Materials and Methods Twelve indoor sites were selected to represent a cross-section of building types, locations, and ages in the New York metropolitan area. No information regarding a known or suspect contamination with mercury or mercury products was used in identifying these locations. Nine residential 10.1021/es010749p CCC: $20.00
2001 American Chemical Society Published on Web 09/21/2001
FIGURE 1. Indoor airborne Hg0 concentrations at 12 sites in the New York metropolitan area. settings were chosen as follows: R1 is a studio apartment in a four-story, prewar building in mid-town Manhattan; R2 is a two-bedroom condominium in a high-rise, newly built apartment building in mid-town Manhattan; R3 is a studio apartment in a high-rise, prewar apartment building in midtown Manhattan; R4 is a three-bedroom apartment in a highrise, postwar building on the upper east side of Manhattan; R5 is a one-bedroom apartment in a high-rise, prewar building on the upper west side of Manhattan; R6 is a threebedroom, two-story house in Forest Hills, Queens; R7 is a three-bedroom apartment in a four-story, postwar building in Flushing, Queens; R8 is a two-bedroom apartment in a turn-of-the-century brownstone in Park Slope, Brooklyn; and R9 is a four-bedroom, two-story, 80 year old house in western Connecticut. Three business environments were chosen as follows: B1 is a postwar, high-rise office building in midtown Manhattan; B2 is a prewar, six-story building in midtown Manhattan; and B3 is a postwar, four-story building in mid-town Manhattan that houses college classrooms and science laboratories. All mercury sampling was conducted at a height of approximately 1 m above the floor for a 3-5 h period in common areas of the living quarters or work settings (i.e. living rooms or hallways) with the windows closed. Mercury vapor sampling was conducted continuously in 5-min intervals with a Tekran Model 2537A mobile Cold Vapor Atomic Fluorescence Spectrometer. The instrument has a detection limit of < 1 ng Hg0 m-3, and instrument calibrations were conducted using an internal mercury permeation source and an external standard injection source. Calibrations and zero-air blank analyses were conducted almost every day during the sampling period. Internal permeation source calibrations remained consistent over the entire duration of this research with the standard deviation of the calibrations equaling less than 2% of the mean and the 95% confidence interval of the calibrations equaling less than 0.5% of the mean calibration value. Blank analyses remained consistently low over the entire sampling period with a mean value of 0.14 ( 0.02 ng m-3 and a maximum of 0.72 ng m-3. Periodically, the instrument permeation source was calibrated using an external, mercury standard addition injection source. The internal and external calibration sources agreed closely with a mean difference between the calibrations of 4.7% ( 1.8%. As an additional measure of analytical accuracy, background outdoor Hg0 concentrations measured over the course of the sampling period (1.8-4.2 ng m-3) agreed closely with known, published background outdoor Hg0 concentrations (12).
Results Figure 1 summarizes average indoor Hg0 concentrations at each of the 12 sampling sites. Indoor concentrations of Hg0 were significantly elevated over mean outdoor Hg0 levels at all except one of the locations sampled.
Indoor mercury concentrations were highly elevated over outdoor concentrations at sites R1, R5, R6, R7, R8, R9, B2, and B3, suggesting an indoor source of Hg0 existed at these locations. Table 1 summarizes site information and average indoor Hg0 concentrations at each of the sites monitored. In an effort to identify possible sources of elevated Hg0 at the sites monitored, extensive interviewing and additional sampling was conducted at the indoor locations. Site B3 is a college building that houses classrooms and science laboratories. The concentrations reported at B3 occurred in a highly trafficked, public hallway at least 100 m from the nearest laboratory; however, it is likely that laboratory contamination contributes to the high Hg0 levels identified at this site. At site R5, the current tenant recalled breaking a mercury thermometer within the 6-month period prior to our sampling. The spill occurred on a tiled bathroom floor, and the tenant stated that the spill was cleaned up with paper towel. The tenant further inspected the floor with a magnifying glass, and all small droplets of mercury were removed using the adhesive side of masking tape. While these are not recommended cleanup methods for liquid Hg0, they were all that was available to the tenant at the time, and they likely represent the methods used in most households under similar circumstances. Airborne mercury sampling was conducted in the bedroom at this location, which is adjacent to the bathroom where the spill was identified. The highly elevated levels of Hg0 at this location (523 ( 6 ng m-3) suggest that residual mercury from the broken thermometer is a significant source of indoor air pollution. In an effort to confirm whether residual mercury in the bathroom was the source of elevated mercury at this residence, we applied a thin dusting of powdered sulfur to the floor of the bathroom. Powdered sulfur forms a film over metallic mercury, which reduces the emission of Hg0; the technique has been used extensively to treat mercury spills in industrial and commercial settings (13). Immediately following (