THE ENVIRONMENTAL RELEASE OF DIOXINS ASSOCIATED WITH THE USE OF PESTICIDES

Polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs or dioxins) are persistent organic pollutants (POPs) that can bioaccumulate and elicit adverse effects in organisms at relatively low exposure levels. To protect human health and the environment, the international Stockholm Convention treaty on persistent organic pollutants (POPs) aims to eliminate or reduce the sources of these compounds on a global scale. In order to achieve these goals, POP sources need to be identified and quantified, to facilitate their prioritisation for national and international actions. This study investigated the role of pesticides as contemporary sources of PCDD/Fs to the environment, and aimed to i) estimate the primary release of PCDD/Fs production impurities during pesticide use, and ii) evaluate secondary PCDD/F releases via photolytic formation from pesticide derived precursors. The presence of pesticide derived PCDD/F precursors was investigated in eight environmental samples containing elevated levels of PCDD/Fs, particularly octachlorodibenzo-p-dioxin (OCDD). Higher chlorinated phenols (CPs) and /or phenoxy phenols (PCPPs) were identified in four soil and two sediment samples at levels comparable to wood treatment sites where high loads of chlorophenol pesticides had been released. Among these, known precursors to OCDD formation, specifically pentachlorophenol (PCP; 0.03 - 18 μg kg-1 dry weight (dw)) and nonachlorophenoxy phenol (nonaCPP; 0.2 - 30 μg kg-1 dw), were present at the highest concentrations. In addition, PCPP profiles were similar to those identified in OCDD contaminated, PCP based, agrochemical formulations. These results indicate that pesticide derived PCDD/F precursors may be a possible source origin for PCDD/F (and particularly OCDD) contamination in soil and sediments. The potential for photolytic PCDD/F formation from pesticide derived precursors was evaluated by exposing currently registered formulations, containing 2,4-dichlorophenoxyacetic acid (2,4-D; n=1) and pentachloronitrobenzene (PCNB; n=2), to natural sunlight. Considerable formation of PCDD/Fs was observed in all formulations resulting in up to 450 and 3,080% increased PCDD/F concentrations in sunlight exposed 2,4-D and PCNB, respectively. Toxic equivalency (TEQ) concentrations also increased to relatively high levels (max: 14 μg kg-1) in the PCNB formulations, predominantly due to formation of OCDD. In contrast, the main PCDD/F congeners formed in 2,4- D included 2,4,6,8-tetrachlorofuran (2,4,6,8-TCDF), 1,3,6,8-tetrachlordibenzo-p-dioxin (1,3,6,8- TCDD) and 1,3,6,8-TCDF. The congener-specific PCDD/F changes during sunlight exposure allowed theoretical identification of the PCDD/F precursors PCP and nonaCPP, as well as, 2,4- Dichlorophenol (2,4-diCP), 2,4,6-trichlorophenol (2,4,6-TriCP), 2,2'-dihydroxy-3,3‗5,5'- tetrachlorobiphenyl (tetraCDOHB), and 4,6-dichloro-2-(2,4,6-tetrachlorophenoxy)phenol ix (pentaC2PP). Photolytic formation from these precursors may represent a significant source of PCDD/Fs to the environment. In addition to precursors, both 2,4-D and PCNB formulations contained elevated levels of PCDD/Fs as manufacturing impurities. To investigate the presence of PCDD/F impurities formed during pesticide production further, a wider range of both currently used (n=23) as well as banned or redundant (n=4) pesticide formulations were analysed. PCDD/Fs were detected in all formulations analysed (0.020 to 2,100 μg ΣPCDD/F kg-1 active ingredient (AI)) with highest levels observed in pentachloronitrobenzene (PCNB) (max: 2,000 μg ΣPCDD/F kg-1 AI; 5.6 μg TEQ kg-1 AI). PCDD/F release associated with the use of pesticides was estimated for currently used formulations containing PCNB (average 27 g TEQ annum-1) and 2,4-D (average 0.87 g TEQ annum-1), as well the banned pesticide 2,4,5-trichlorophenoxyacetic acid (2,4,5-T; maximum 27 g TEQ annum-1). These release estimates are relatively high compared to other, previously quantified PCDD/F sources, and thus indicate that currently used pesticides are ongoing and important sources of these POPs to the environment. Full PCDD/F congener profiles in pesticides were quantified to establish reference fingerprints for future source identification. In a case-study approach, these fingerprints were used to evaluate PCDD/F contamination in Australian soils from agricultural, urban and forestry land-use. The receptor model Positive Matrix Factorisation (PMF) was applied to resolve individual sources and apportion their contributions to these soils. PCDD/F fingerprints in several current use pesticides, including commonly used herbicides, e.g. 2,4-dichlorophenoxybutyric acid (2,4-DB), and sunlight exposed pesticides, 2,4-D and PCNB, were found to closely match PCDD/F contamination profiles in Australian soils. Sources with the highest contributions (max: 98%) to these samples included the OCDD dominated current use pesticides imazamox, PCNB and mecoprop, as well as, banned pesticides PCP and chlordane. Overall, the results from this research demonstrate that pesticide use is an ongoing and potentially significant source of PCDD/Fs to the environment. These results have important implications for the regulation of impurities in pesticide manufacture, and have been used to inform health risk assessment and associated changes to pesticide regulation, and serve as proposed new emission factors in the UNEP toolkit for identification and quantification of dioxin sources under the Stockholm Convention