Environ. Sci. Technol. 2008, 42, 8740–8746
Bioaccessibility and Bioavailability of Cu and Zn in Sediment Contaminated by Antifouling Paint Residues ANDREW TURNER,* NIMISHA SINGH, AND LEIGH MILLARD School of Earth, Ocean and Environmental Sciences, University of Plymouth, Drake Circus, Plymouth PL4 8AA, United Kingdom
Received July 11, 2008. Revised manuscript received September 22, 2008. Accepted October 1, 2008.
embedded into the matrix as oxidic grains. Particles also contain considerable quantities of other trace metals (e.g., Cd, Cr, Pb) as contaminants or additional pigments and organic and organometallic cobooster biocides, including zinc pyrithione, Irgarol 1051, diuron, and dichlofluanid (8, 9). With regard to the oxidic pigments, Cu release from paint particles into seawater proceeds as follows (9) 0.5Cu2O(s) + H+ + 2Cl- T CuCl2 + 0.5H2O CuCl2
-
+ Cl T
CuCl23
(1) (2)
where eq 1 is kinetically constrained and eq 2 is instantaneous and, in the presence of dissolved oxygen, univalent copper is rapidly oxidized to Cu2+. The mechanism governing dissolution of ZnO in seawater is not fully understood, but the following, overall reaction has been postulated (10) ZnO(s) + H2O + 2Cl- T 0.5ZnCl24 + 0.5Zn(OH)3 +
The bioaccessibility and bioavailability of the principal metallic constituents of spent antifouling particles ([Cu] ) 300 mg g-1; [Zn] ) 100 mg g-1) have been evaluated by in vitro incubations and in microcosms containing the marine depositfeeder Arenicola marina. In mixtures of sediment and paint, metalaccessibilitytotheprotein,bovineserumalbumin,asurrogate for the gut fluids of deposit feeders, increased as the proportion of paint particles in the sample decreased. This effect was attributed to solubility constraints on metal salts and complexes and resulted in estimates of bioaccessibility in paint residues ranging from about 0.3% to 1.7% for Cu and 0.2% and 2.3% for Zn. A. marina maintained in sediment-paint cores and in paint leachate accumulated Cu with accumulation factors of about 0.1% and 0.5%, respectively, suggesting that both diet and aqueous exposure contribute to the uptake of this metal. In contrast, Zn was not measurably accumulated by either exposure route, suggesting that A. marina is able to regulate this metal. Through burial and conveyor-belt feeding, A. marina also accelerated both the subduction of antifouling residues and the mobilization of metals into the interstitial and overlying waters. The findings of this study have important implications regarding the cycling of trace metals in coastal waters impacted by boating activities.
Introduction Fragments of antifouling paint represent a significant, heterogeneous source of contamination in many coastal and marine environments whose impacts are poorly understood (1). Particles are widely generated during the maintenance of boats through, for example, the scraping, sanding, and hydroblasting of hulls (2-4). Despite legislation or codes of practice concerning their safe disposal, considerable quantities of particles are transported into local sediment with washdown and runoff or as airborne dust (5, 6). Antifouling residues are also generated during the grounding of ships (7) and readily flake off stationary and moored structures in situ, including buoys, nets, cages, piers, and abandoned boats. The chemical characteristics of spent particles largely reflect those of the original (dried) antifouling formulations. Consequently, residues are highly enriched in Cu and Zn, the principal metallic biocidal pigments that are usually * Corresponding author phone: +44 1752 584750; fax: +44 1752 584710; e-mail:
[email protected]. 8740
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 23, 2008
0.5OH- (3) Thus, release of both Cu and Zn from their respective oxides appears to be dependent on the square of the concentration of chloride. Given the chemical makeup and reactivity of antifouling paint, it would be reasonable to hypothesize that contaminants in spent residues are highly amenable to marine benthic organisms, especially those that are exposed to or ingest large amounts of sediment. To this end, we evaluate both the bioaccessibility and bioavailability of the metallic biocidal constituents (Cu and Zn) of antifouling paint residues. Bioaccessibility is defined as the fractional concentration of a particulate chemical or contaminant that is accessible in the gastrointestinal environment and is considered an upper bound of the proportion of the chemical that is bioavailable upon ingestion (11). Here, bioaccessibility is evaluated in vitro by incubating mixtures of antifouling paint particles and clean sediment with a commercially available protein as a surrogate for the digestive fluids of deposit-feeding animals (12, 13). Bioavailability is defined as the fractional concentration of a particulate chemical that is available for absorption by an organism upon ingestion and generally related to (but not equal to) the bioaccessible fraction. Here, bioavailability to the marine polychaete worm, A. marina, is determined in microcosms to which a controlled quantity of antifouling paint particles is administered. A. marina is an abundant, infaunal conveyor-belt feeder of temperate latitudes that is often used to study contaminant accumulation and processing in near-shore environments (14, 15). It lives in a J-shaped burrow and prefers fine sand or muddy sand with a low organic content.
Experimental Section All plasticware and glassware employed in the experiments and for sample processing and storage were first soaked in 10% HCl for 24 h and subsequently rinsed with Milli-Q water (MQW). Reagents were purchased from BDH/VWR, Fisher Scientific, and Sigma Aldrich and were of analytical grade or equivalent. Seawater (S ≈ 33; pH ≈ 7.5) was available in the laboratory on tap having previously been collected in bulk from Plymouth Sound at high water, and was used after being vacuum filtered through a 0.45 µm Whatman membrane. Collection and Processing of Source Materials. Fragments of boat paint, of between about 5 and 50 mm in length, were collected by hand from the hard standings and slipways of a boat maintenance facility, catering for mainly (but not 10.1021/es801923e CCC: $40.75
2008 American Chemical Society
Published on Web 11/06/2008
exclusively) leisure craft, in Plymouth, SW England. Although we implicitly refer to the sample and fragments thereof as antifouling in nature, it is important to appreciate that some paint particles may have been derived from parts of the boat not associated with the hull (e.g., decking, cabin) but undergoing general maintenance. These particles have a different chemical makeup to antifouling fragments, but the net sample is representative of the signature of particulate contamination derived from the general, contemporary practice of leisure boat maintenance. In the laboratory, visible extraneous particulates (e.g., grit, macroalgae) were removed and the fragments pooled and subsequently ground with a pestle and mortar, a process aided by the occasional addition of a few milliliters of liquid nitrogen. The composite was then sieved through a 1 mm Nylon mesh and the fine fraction stored in a polyethylene canister in the dark until required. About 10 L of sand were taken at low tide from the marine reaches of a protected estuary (Erme, SW England) that is densely populated by A. marina (14). The sample was wet sieved on site through a 1 mm Nylon mesh, and the fraction passing through the sieve was transported to the laboratory in a lidded polyethylene bucket. The sample was divided and stored in a series of zip-locked plastic bags at -18 °C until required. Samples of A. marina were collected from the Erme Estuary on a separate occasion (i.e., when laboratory microcosms had been established; see below). Individuals were located by their casts and carefully retrieved from their burrows by digging to a depth of about 30 cm using a garden fork. Organisms were transported in a lidded bucket containing about 4 L of estuarine water, and in the laboratory they were allowed to void their guts for 48 h in 5 L Perspex tanks half-filled with aerated laboratory seawater (15). Mature individuals of similar length (9-12 cm) and wet weight (3-6 g) were selected for further experimental use. In vitro Studies. The bioaccessibility of Cu and Zn in paint particles, sediment, and mixtures thereof was evaluated by digesting solids in the commercially available protein, bovine serum albumin (BSA). Solutions of this protein do not contain active digestive agents, such as enzymes and surfactants, but mimic chemical characteristics (and in particular complexation capacity) of the gut fluids of depositfeeding marine invertebrates that are instrumental to the mobilization of trace metals from ingested solids (12, 13). Solutions were prepared, as required, by dissolving 5 g of BSA (fraction V as lypophilized power) in 1 L of laboratory seawater in a 1.5 L polyethylene bottle. Given the molecular mass of the protein (66 400 g mol-1) and number of component amino acids (AA ) 583), working solutions are equivalent to an AA concentration of about 45 mM, close to the median concentration encountered in the digestive environments of a range of temperate zone deposit-feeding organisms (16). Digestion was undertaken in foil-wrapped 125 mL polyethylene bottles and at a solid to fluid ratio of about 250 g L-1 (typical of that occurring in the digestive environments of marine deposit feeders; (13)). Different sediment to paint ratios and, therefore, w/w metal concentrations were attained by weighing out appropriate quantities of source materials (note that sediment was weighed out in a thawed, wet state with prior knowledge of its water content). One hundred milliliters of BSA solution were added to each bottle and the contents agitated on a lateral shaker at about 200 rpm at room temperature (19 ( 1 °C). Subsamples of 15 mL, taken at different time intervals up to 20 h, were centrifuged in polypropylene tubes for 15 min at 2100 g and subsequently filtered through 0.45 µm. Triplicate 4 mL aliquots were stored in individual Sterilin tubes at 4 °C before being analyzed (within 3 days). Time-dependent controls were undertaken for each sediment to paint ratio in seawater without BSA amendment.
Microcosms with A. marina. The approach employed to examine the accumulation of Cu and Zn by A. marina is based on that described elsewhere (14, 15, 17). Thus, five types of microcosm were prepared in quadruplicate in a series of clear, 1 L Perspex core tubes whose bases had been capped. To the first type of microcosm (M1) we added 500 mL of Erme estuarine sediment, 500 mL of laboratory seawater, and a single worm. After casting was observed the sediment surface was leveled using a plastic spatula, and 45 mg of the fractionated paint composite were allowed to settle through the water column. Equivalent controls (M2) were prepared likewise but without paint particles, while a third type of microcosm (M3) contained paint and sediment but no A. marina. Microcosms were incubated at 15 °C and under fluorescent lighting and constant aeration for a period of 10 days. In selected cases, daily 5 mL water samples were abstracted, filtered through 0.45 µm, and stored in 10 mL Sterilin tubes after acidification to pH < 2 using HCl. At the end of the incubation period remaining water was siphoned off and the sediment transferred to a polyethylene tray. Here, it was divided using a plastic spatula into six sections of increasing thickness from the surface (1.5 cm) to the base (3 cm). Sections were filtered through 0.45 µm to remove interstitial water and air dried in individual Petri dishes. During sectioning, worms were carefully retrieved from their burrows and placed in filtered seawater for 24 h to void their guts before being freeze dried. Incubations were also performed in the absence of sediment. Thus, in M4 individuals of A. marina were exposed to solutions of paint leachate for a period of 5 days (animals did not survive for 10 days under these conditions). Leachate was prepared by equilibrating 22.5 mg of paint particles in 100 mL of laboratory seawater for 5 days under occasional agitation and subsequently diluting to 500 mL with clean seawater. Corresponding controls (M5) were undertaken in particle-free seawater under otherwise identical conditions. Sample Digestion and Analysis. Total digestion of source materials was undertaken in aqua regia as follows. Thus, three 10 mg subsamples of the antifouling paint composite and three 100 mg subsamples of the dried estuarine sediment and a certified reference sediment (LGC 6137; Laboratory of Government Chemists, Teddington, U.K.) were weighed into individual 50 mL Pyrex beakers. Five milliliters of 3 parts HCl to 1 part HNO3 were added to each beaker, and after about 1 h the contents were covered with watch glasses and heated on a hot plate to about 75 °C for a further 2 h. The cooled contents and MQW rinsings were transferred to individual 25 mL Pyrex volumetric flasks and diluted to mark with 0.1 M HNO3. Procedural blanks were performed likewise but in the absence of solids. For the total dissolution of A. marina, individuals were digested in HNO3 under otherwise identical conditions. Acidified filtrates and total and in vitro digests were analyzed for Cu and Zn by inductively coupled plasma-optical emission spectrometry (ICP-OES) using a Varian 725 ES (Mulgrave, Australia). The instrument was calibrated using mixed, acidified standards, and internal standardization was achieved by addition of yttrium. Analysis of reference material digests revealed concentrations that were within 10% of certified values. Concentrations of carbon and nitrogen in the dried sediment sample (3.1 ( 0.3% and 0.20 ( 0.01%, respectively) were established in triplicate 5 mg aliquots by flash combustion using a Carlo Erba EA 1110 elemental analyzer. The instrument was calibrated using EDTA standards, and analysis of various reference sediments revealed that C determinations were always within 5% of certified values. VOL. 42, NO. 23, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8741
FIGURE 1. Concentrations of Cu in mixtures of estuarine sediment and antifouling paint particles that are accessible to seawater (0) and 5 g L-1 of BSA in seawater (9) as a function of time. Error bars represent the standard deviation about the mean of three measurements but were smaller than the size of the symbols in most cases. [CuT] is the total concentration of Cu present, and fAFP denotes the fractional contribution of paint particles to each preparation. The inset exemplifies accessibility to seawater on an expanded y axis.
FIGURE 2. Concentrations of Zn in mixtures of estuarine sediment and antifouling paint particles that are accessible to seawater (0) and 5 g L-1 of BSA in seawater (9) as a function of time. See caption to Figure 1 for further details. centrations accessible to seawater increased in a convex 3. Results and Discussion fashion, an effect evident on expansion of the y axis (see Accessibility of Cu and Zn to Seawater and BSA. Conceninsets). Where paint was the sole solid phase (fAFP ) 1) and trations of Cu and Zn in antifouling paint particles, estuarine metal mobilization is governed by reactions 1, 2, and 3, the sediment, and mixtures thereof that are accessible to (or time distributions of Cu and Zn concentration were more mobilized by) seawater and BSA are shown as a function of complex with evidence of rapid mobilization followed by a time and on a w/w basis in Figures 1 and 2. Annotated on period in which readsorption or precipitation took place. each panel are the mass fraction of paint particles in the In all experiments the presence of BSA enhanced metal sample, fAFP, and the total concentration of metal present, accessibility, presumably because component amino acids [MeT], either measured directly or calculated from the relative of the protein act as additional ligands for complexation of proportions and compositions of source materials. In sediCu (in both valence states; (18)) and Zn (11) in eqs 1, 2, and ment and sediment-paint mixtures (fAFP < 1) metal con8742
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 23, 2008
FIGURE 3. Bioaccessibilities of Cu and Zn measured in mixtures of estuarine sediment and antifouling paint particles (solid bars) and calculated for paint particles alone using eq 5 (hatched bars). Bioaccessibilities of Cu and Zn in the sediment end member (about 0.8% and 1.6%, respectively) are lower than equivalent values reported for finer, more contaminated sediment (1, 11, 13) but considerably greater than bioaccessibilities measured directly in the antifouling paint composite (
