Publication Abstract

Polychlorinated dibenzo-p-dioxins, polychlorinated dibenzofurans (PCDD/Fs) and dioxin-like polychlorinated biphenyls (dl-PCBs), collectively referred to as ‘dioxins’, are widely recognised environmentally persistent, ubiquitous biotic and abiotic contaminants. Dioxins are well established priority hazardous substances and are listed by the Oslo and Paris Convention for the Protection of the Marine Environment of the North-East Atlantic (OSPAR). Since 2010, the OSPAR Coordinated Environment Monitoring Programme (CEMP) lists dioxins and furans as contaminants that are likely to be mandatorily monitored, in sediment and biota, by OSPAR contracting parties in the near future (pre-CEMP list).

 

Cefas has developed a capability in establishing the occurrence of dioxins and dioxin-like-substances in environmental samples using the DR-CALUX bio-analytical technique (Hurst et al. 2004). In addition to providing a quantitative determination of the dioxins and furans, the DR-CALUX assay also provides an integrated signal of dioxin-like activity in a sample. Cefas findings to date suggest that compounds, other than dioxins and furans, are present in the UK marine environment and are also exhibiting dioxin-like properties. The development of a chemical analytical method for dioxins analysis would allow an accurate measurement of the contribution being made by dioxins and furans to the total dioxin-like signal to be made and thus permit an accurate determination of the contribution being made by other, as yet unknown, dioxin-like-substances

 

The analysis of dioxins is expensive due to the use of high-resolution mass spectrometers (HRMS) run by highly skilled analysts. Outputs from the EU research programmes DIAC and DIFFERENCE, which focused on the development of chemical and bio-analytical techniques for dioxins analysis, have highlighted the potential for non-HRMS methods for accurate dioxins analysis. The aim of the work reported here was to evaluate GC-low resolution ion trap mass spectrometry in tandem MS mode (GC-IT-MS/MS) for the determination of dioxins and for its applicability to Defra’s monitoring and research programmes. Following instrument optimisation, the evaluation of GC-IT-MS/MS was satisfactory for all 29 toxic dioxins and dioxin-like compounds [7 dioxins (PCDDs), 10 dibenzofurans (PCDFs) and 12 dioxin-like PCBs (coplanar PCBs)], with adequate sensitivity, as stipulated in US-EPA8290 method for the analysis of dioxins by high resolution MS.

 

The successful outcome of this initial evaluation of GC-IT-MS/MS warranted further method development involving semi-automated extraction of sample followed by automated clean up and fractionation steps. The possibility of eliminating the separate, laborious clean up step traditionally used in dioxins analysis was investigated by incorporating a clean up step with both the extraction and fractionation of target analytes. This was achieved using accelerated solvent extraction (ASE) with simultaneous in-cell sample clean up followed by a PCB fractionation step using a porous graphitic carbon (PGC) high performance liquid chromatography (HPLC) column protected by a normal phase (NP-) preparative column which acts as an additional clean up. The NP-HPLC clean up was optimised first, using a preparative amino-propyl (NH2-) HPLC column and 100% n-hexane as the mobile phase. This was done successfully, and all target compounds were eluted within 8 minutes.

 

The second step involved the fractionation of the predominantly low level, non-ortho, coplanar PCBs (and possibly some low level mono-ortho coplanar PCBs) from the non-coplanar PCBs. The latter are generally present in matrices in much higher concentrations than most coplanar PCBs and would mask the presence of some of the toxic coplanar PCBs. Following optimisation of various instrument parameters, a satisfactory fractionation was developed using 100% n-hexane in forward flow to drive all non-coplanar and mono-ortho coplanar PCBs through the PGC column while retaining the non-ortho coplanar PCBs, PCDDs and PCDFs. By then reversing the flow through the PGC column using a more polar mobile phase (100% toluene), the compounds retained on the column were eluted. With the NH2-HPLC column installed in front of the PGC column, both clean up and fractionation were achieved simultaneously with a total run time of <70 minutes. This setup can process up to 20 sample extracts unattended.

 

Work subsequently progressed to the assessment of ASE with in-cell clean up as a dioxins extraction procedure. The in-cell cleanup was achieved by using sulphuric acid impregnated silica. Recoveries of PCDDs/PCDFs from procedural blanks spiked with standard solutions were satisfactory and highest using n-heptane as the extraction solvent. Consequently, the accuracy and precision of the full method, including extraction, clean up, fractionation, was evaluated using a sediment reference material (SETOC 737). However, GC analysis of the sediment sample extracts indicated the presence of interfering matrix components resulting in the over-estimation of some dioxins, but also in the rapid loss of the sensitivity of the instrument.

 

As a result, an additional clean up stage was investigated using a commercially available, multilayer silica column (Supelco ‘Dioxin Prep System’ from Sigma Aldrich, Poole, UK) and fractionation windows for optimum recovery of target analytes were established. Analyte recoveries were acceptable and, as a result, the sediment reference material SETOC 737 was analysed in triplicate using the full procedure involving ASE with in-cell sulphuric acid cleanup, ‘Dioxin Prep System’ cleanup, HPLC clean-up and PGC fractionation, and finally GC-LR-MS analysis.

 

For the 17 toxic dioxins (7 PCDDs and 10 PCDFs), recoveries were generally satisfactory (within the range 50% to 120%), with only 2,3,7,8-TCDF exhibiting a very low recovery (17%) and two determinands (1,2,3,6,7,8-HxCDD and 1,2,3,7,8-PeCDF) having recoveries >120%. Relative standard deviation were generally acceptable at <35% except for 1,2,3,4,7,8,9-HpCDF (41%). These results indicate that the methodology developed here has the potential to be applied successfully to real sediment samples. However, despite the extensive sample extract clean up that is required, the sensitivity of the GC-IT-MS/MS dropped throughout a batch of just a few samples. Cefas has recently acquired a GC with the capability of triple quadrupole tandem mass spectrometry (MS/MS), an instrument that is expected to be more robust and more sensitive than the GC-IT-MS/MS used in this study. Further work should focus on the (1) transfer of the GC-IT-MS/MS instrumental methodology to the GC-MS/MS, (2) solving issues with declining sensitivity during batch analysis, (3) extending the extraction, clean up and analytical methodologies to the analysis of dioxins from biota samples and (4) comprehensive validation of the final methods using certified reference materials to fully characterise method performances.