Environ. Sci. Technol. 2006, 40, 988-993
Release of Arsenic to the Environment from CCA-Treated Wood. 1. Leaching and Speciation during Service B E R N I N E I . K H A N , †,| H E L E N A M . S O L O - G A B R I E L E , * ,† T I M O T H Y G . TOWNSEND,‡ AND YONG CAI§ Department of Civil, Architectural, and Environmental Engineering, University of Miami, P.O. Box 248294, Coral Gables, Florida 33124-0630, Department of Environmental Engineering Sciences, University of Florida, Gainesville, Florida 32611-6450, and Department of Chemistry & Biochemistry and Southeast Environmental Research Center (SERC), Florida International University, Miami, Florida 33199
Insufficient information exists about the speciation of arsenic leaching from in-service chromated copper arsenate (CCA)-treated products and the overall impact to soils and groundwater. To address this issue, two decks were constructed, one from CCA-treated wood and the other from untreated wood. Both decks were placed in the open environment where they were impacted by rainfall. Over a one-year period, rainwater runoff from the decks and rainwater infiltrating through 0.7 m of sand below the decks was collected and analyzed for arsenic species by HPLC-ICP-MS. The average arsenic concentration in the runoff of the untreated deck was 2-3 µg/L, whereas from the CCA-treated deck it was 600 µg/L. Both inorganic As(III) and As(V) were detected in the runoff from both decks, with inorganic As(V) predominating. No detectable levels of organoarsenic species were observed. The total arsenic concentration in the infiltrated water of the treated deck had risen from a background concentration of 3 µg/L to a concentration of 18 µg/L at the end of the study. Data from the deck study were combined with annual CCA-treated wood production statistics to develop a mass balance model to estimate the extent of arsenic leaching from in-service CCA-treated wood structures to Florida soils. Results showed that during the year 2000, of the 28 000 t of arsenic imported into the state and utilized for in-service CCA-treated wood products, approximately 4600 t had already leached. Future projections suggest that an additional 11 000 t of arsenic will leach during in-service use within the next 40 years.
Introduction Chromated copper arsenate (CCA) is a chemical preservative added to wood to protect the wood from biological dete* Corresponding author tel: 305-284-3489; fax: 305-284-3492; e-mail:
[email protected]. † University of Miami. ‡ University of Florida. § Florida International University. | Present address: U.S. Environmental Protection Agency, Office of Water/Office of Science and Technology Health and Ecological Criteria Division, Washington DC 20460. 988
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 3, 2006
rioration. The amount of chemical added to the wood, or retention level, is typically expressed in units of kilograms of CCA chemical per cubic meter of wood. Lower retention wood is used for above ground applications (4 kg/m3), while higher retention wood is used for utility poles (9.6 kg/m3) and poles used in marine environments (40 kg/m3) (1). This study focused on evaluating the arsenic component only, although CCA also contains chromium which can be highly toxic in the hexavalent state and net impacts of arsenic releases from the wood can be potentially enhanced by the chromium and copper components of the CCA chemical. Arsenic is added to the wood in its pentavalent oxidation state. Arsenic accounts for 22% of the CCA chemical by weight and the concentration in the treated wood can range from 1900 to 19 000 mg/kg. Given the high concentrations, arsenic leaching from CCA-treated wood structures to surrounding soils poses a possible threat to groundwater supplies. Studies have shown that soils (2-8) and groundwater (9) in close proximity to in-service CCA-treated wood products measured at levels well above natural background concentrations and risk-based regulatory levels. Recent surveys show the “actual” in-service life of low retention CCA-treated wood products such as decks to vary from 9 to 13 years (10, 11) instead of the design life of 20 to 25 (12) years. This early retirement of the wood is attributed to aesthetics due to the effects of natural weathering. For higher retention treated wood, the “actual” in-service life is approximately 40 years (13) or more (12). While a significant fraction of the arsenic remains fixed into the wood after prolonged environmental exposure, small losses may be of environmental significance due to the large concentrations in the wood and the toxicity of the components (14). Arsenic toxicity varies with speciation, with the inorganic species (As(III) and As(V)) considered to be more toxic than the organic species, monomethylarsonic acid (MMAAV) and dimethylarsinic acid (DMAAV) (14). Although recent studies have found that trivalent forms of the organic species can be more toxic than the inorganic species (15) these forms are believed to be relatively short-lived in the environment and thus measurements of organic forms of arsenic focused on the pentavalent species. At the end of 2003, arsenic-treated wood was phased out in favor of new alternative wood preservatives for the U.S. residential market and for publicly used facilities (16). This transition affected CCA-treated wood structures used for playgrounds, decks, picnic tables, landscaping timber, residential fencing, patios, walkways, and boardwalks. While the phase-out provides a decrease in production rates, the mass of arsenic already released to the soil from preexisting and present-day leaching of CCA-treated wood products is of concern. Past laboratory leaching studies provide useful insight into preservative leaching from CCA-treated wood products (17-20), however, very few of these studies measure arsenic species (21-23) and few evaluate leaching under field conditions (24, 25). None have measured arsenic species released under field conditions. The objective of the current study was to evaluate the overall quantities and species of arsenic leached from in-service CCA-treated wood subjected to natural rainfall conditions. Two decks were constructed and placed outdoors as part of the current study, one made from CCA-treated wood and the other made from untreated wood. Rainwater runoff from the decks and infiltrated water percolating through soil beneath the decks were collected periodically and analyzed for arsenic species (As(III), As(V), MMAA, and DMAA). Results obtained from the decks were 10.1021/es0514702 CCC: $33.50
2006 American Chemical Society Published on Web 12/21/2005
combined with annual CCA-treated wood production statistics in Florida to develop a mathematical mass balance model to approximate the annual input of arsenic to the environment as a result of the in-service use of CCA-treated wood. This manuscript is part of a two-paper series. The second paper focuses on evaluating the quantities of arsenic released after in-service use, once the wood is disposed.
Materials and Methods Deck Design. Two 1.8-m × 1.8-m decks were constructed from wood purchased at a local retail store. One deck was composed of CCA-treated Southern Yellow Pine (SYP) wood and the other, a control, was made from untreated SYP wood. Only the deck boards (not the support beams and columns) were constructed with CCA-treated wood and contained arsenic. The retention level of the CCA-treated wood was measured at 3.5 kg/m3, resulting in a total mass of 43 g of arsenic within the deck boards. The surface area of the deck impacted by rainfall was 2.92 m2, which corresponded to a wood volume of 0.056 m3. Each deck was placed outside within an open area located on the University of Miami, Coral Gables campus. Each deck was fitted with a plexiglass rain gauge. Analysis of the water samples collected from each rain gauge indicated that the concentration of arsenic in the rainfall was less than 2 µg/L. Water samples were routinely collected from two leachate collection systems; the first was designed to collect runoff from the deck surface and the second was designed to collect infiltrated water (See Figure A in the Supporting Information). The surface runoff collection system consisted of a gutter located along the downstream edge of the deck. Runoff from the gutter was diverted toward a reservoir located immediately below the deck. The gutter was covered such that direct rainfall could not enter the gutter. Collection of infiltrated water was facilitated by placing each deck within a 2.4-m × 2.4-m × 1-m untreated resin-coated wooden box situated upon on a 5° inclined aluminum base. The wood used for the box was made of untreated plywood and the inside walls of the box were lined with plexiglass. The decks were placed in each box on top of a 0.7-m layer of Florida native sand characterized by less than 1% organic content. The 0.7-m depth was chosen because, in Florida, the depth to groundwater can be as shallow as 0.7 m (26). The 0.7-m sand layer was underlain by a drainage collection system that consisted of a permeable geotextile, gravel, an impervious liner, and a perforated pipe that diverted infiltrated water toward a reservoir located outside the wooden box. Runoff and infiltrated water samples were collected from the corresponding reservoirs for a 1-year period (September 2002 to August 2003). These samples were analyzed for arsenic species. Arsenic Analysis. Speciation analysis was performed by use of continuous high performance liquid chromatography (HPLC) coupled with inductively coupled plasma-mass spectrometry (ICP-MS) detection. The species measured using this method were inorganic As(III) and As(V), and the organic species MMAA and DMAA. The HPLC (SpectraSYSTEM P4000 with a SpectraSYSTEM AS 3000 Autosampler) was fitted with an anion-exchange column (Hamilton PRP-X100, 250 mm × 4.6 mm i.d., 10 µm particle size) through which the mobile phase (15 mM KH2PO4/15 mM K2HPO4, pH 5.81) was passed at a flow rate of 1.0 mL min-1. Sample injection volume was 100 µL. The ICP-MS (HP4500 Plus) was controlled using HP ChemStation Software WinNT for ICP-MS. The plasma and auxiliary gas flow rate for the ICP-MS were maintained at 15.9 and 1.04 L/min, respectively. The signal-to-noise ratio was
