Using Quantitative Structure–Activity Relationships to Support the Assessment of the Environmental Fate and Aquatic Toxicity of a Series of Methacrylic Acid Esters

ABSTRACT

Methacrylic acid esters or methacrylates (with C1 to 8 alkyl side-chains) are a group of compounds used in the manufacture of resins and plastics. Releases to the atmosphere, surface waters, or soil can come about during manufacturing, use, transport, and storage activities. Given the potential for environmental exposures, an examination into factors that control the distribution, fate, and toxicity of methacrylates is warranted. Methacrylates are relatively volatile compounds that are easily and rapidly degraded in the atmosphere and in oxic surface waters and sediments, and so are not considered persistent. Aquatic organisms are also able to metabolize methacrylates as demonstrated by estimated bioconcentration factors of < 45 for the acid and the C1-4 esters, and a measured bioconcentration factor (BCF) for 2-EHMA. Thus, the methacrylates are not considered bioaccumulative. Acute and chronic aquatic toxicity data for the methacrylates were assembled. Quantitative structure–activity relationships (QSARs) were developed with the acute and chronic fish, invertebrate, and algal datasets to support the toxicity assessment of the methacrylates (with log octanol-water partition coefficients used as surrogates for structures). The data show the methacrylates to be of low to moderate toxicity with all acute LC/EC50 values and chronic no observed effect concentrations ranging from about 2 to 170 mg/l for the acid and C1-4 esters, and 0.1 mg/l up to the aqueous solubility limit (about 2 mg/l) for the C8 ester. The measured toxicity data and the data estimated from the QSARs were used to develop predicted no effect concentrations in water and sediment for methacrylic acid and the C1-C8 esters.

Key Words:

• methacrylates
• volatility
• biodegradation
• distribution modeling
• QSARs

ACKNOWLEDGMENTS

The authors are grateful to the Japanese Ministry of the Environment and the Chemicals Evaluation and Research Institute (CERI) on behalf of the Ministry of Economy, Trade and Industry (METI) in Japan. This work was sponsored by the Methacrylate Producers Association, Hamilton, VA, USA.

Notes

¹Values calculated from molecular weight, vapor pressure, and water solubility.

²Values either measured or calculated from a regression of measured data for MAA, MMA, and iBMA.

³Values calculated based on log K_ow using BCFWIN v2.17 (EPISuite v3.20, USEPA 2007).

⁴Bioconcentration factors based on studies with Brachydanio rerio, based both on steady state concentrations and uptake/depuration results.

¹For the methacrylates, atmospheric photo-degradation is a combination of hydroxyl radical and ozone-mediated reactions. First-order half-lives used for multi-media modeling are calculated assuming 1.5E6 hydroxy-radical molecules per cm³, a 12-hour sunlight period per 24-hour day, and 7E11 ozone molecules per cm³ (no sunlight adjustment made). For MMA, the measured hydroxyl-radical degradation rate was coupled with the calculated ozone-mediated rate.

²Volatilization half-lives from model rivers and lakes assume river water characteristics of 1 m depth, wind velocity of 5 m/sec, and current velocity of 1 m/sec, and for a lake of 1 m depth, wind velocity of 0.5 m/sec, and current velocity of 0.05 m/sec.

³Ready tests used OECD 301-series protocols.

⁴For 2-EHMA, the 301B test achieved 93% by day 28, but did not meet the 10-day window.

⁵Level III distribution modeling according to EPISuite v3.20, USEPA (2007). As discussed in the text, emissions were adjusted to 99.82% to air, 0.18% to water, and 0% to soil. (1) Douglas and Bell 1992; (2) Archer 1990; (3) CITI 1992; (4) Miller 1989; (5) Thiebaud and Moncel 1995a; (6) Kikushima 1998a; (7) Thiebaud and Moncel 1995b; (8) Mori 1998a; (9) Thiebaud and Moncel 1995c; (10) Kikushima 1998c; (11) Mori 1998b; (12) Mead 1995.

¹All valid studies used flow-through or renewal test systems or entirely closed vessels, and measured test concentrations. In addition, valid acute tests were of standard duration (96 h for fish, 48 h for daphnids, and ≥ 72 h for algae). Acute tests measured survival and immobilization in fish and invertebrates, respectively, and growth rate and biomass in algae. Chronic tests measured survival and growth in fish, survival, and reproduction in invertebrates and growth rate and biomass in algae. Species geometric means (GM) are given for species with multiple valid tests. A chronic QSAR could not be developed for fish.

²One species, rainbow trout Oncorhynchus mykiss.

³One species, zebrafish Brachydanio rerio.

⁴One species, water flea Daphnia magna, multiple results given if multiple valid tests were available.

⁵One species, green algae Selenastrum capricornutum, multiple results given if multiple valid tests were available.

⁶Two species, O. mykiss and bluegill sunfish Lepomis macrochirus.

⁷One species, Salmo gairdneri.

⁸Two species, medaka fish Oryzias latipes and fathead minnow Pimephales promelas.

⁹Lowest observed effect level (LOEC) not given.

¹⁰Estimated toxicity values calculated from QSARs in molar units, converted back to mg/l. (1) Bowman 1990a; (2) Schafers 2002a; (3) Burgess 1990a; (4) Putt 1995a; (5) Comber and Long 1999; (6) Bowman 1990b; (7) Bailey et al. 1985; (8) Schafers 2002b; (9) Burgess 1990b; (10) Putt 1997; (11) Smyth and Long 1999; (12) Forbis 1990; (13) Sousa 1995a; (14) Putt 1995b; Thiebaud 1994a; (16) Putt 1995c; (17) Hoberg 2002a; (18) MoE 1998a; (19) Morris et al. 1993b; (20) MoE 1998b; (21) Kent et al. 1993; (22) Thiebaud 1994b; (23) MoE 1998c; (24) Putt 1995d; (25) MoE 1998d; (26) Smyth et al. 1993; (27) Sousa 1995b; (28) Putt 1995e; (29) Thiebaud 1994c; (30) Hoberg 2002b; (31) Schafers 2003; (32) MoE 1998e.

¹The lowest LC/EC50 or NOEC value from Table 3.

²PNEC_aquatic calculated assessment factors of 10, deemed protective of long-term exposure to the compound.

³Koc values given are the mean measured or estimated values from Table 1.

⁴PNEC_sediment calculated using Eq. (1). The foc was assumed to be 2% in sediments.