Jukka Pellinen

Multiresidue method for the analysis of emerging contaminants in wastewater by ultra performance liquid chromatography–time-of-flight mass spectrometry

A multiresidue method for screening of emerging contaminants in aquatic environments was developed. The method was based on sample pretreatment with solid phase extraction (SPE) and analysis with an ultra performance liquid chromatograph-time-of-flight mass spectrometer (UPLC-TOF-MS). The method was optimized and tested with standard solutions of model compounds containing 84 pesticides and pharmaceuticals. Four different SPE sorbents were evaluated to gain maximum recovery for the analytes. For the final procedure a combination of two different sorbents was chosen. In spite of high matrix suppression, the method quantification limits (MQLs) were acceptable. Therefore, the method can be used for screening known target compounds. The applicability of the method for posttarget and nontarget screening will be reported later. To preliminarily assess the quantitative performance of the method, some compounds in wastewater effluent were quantified using the standard addition method. Three pesticides and eight pharmaceuticals were found in concentrations up to ∼2200 ng/L.

Critical evaluation of screening techniques for emerging environmental contaminants based on accurate mass measurements with time-of-flight mass spectrometry.

Emerging contaminants from wastewater effluent samples were analysed, using posttarget and nontarget analysis techniques. The samples were analysed with an ultra performance liquid chromatograph-time-of-flight mass spectrometer (UPLC-TOF-MS), and the resulting data were processed with commercial deconvolution software. The method works well for posttarget analysis with prior information about the retention times of the compounds of interest. With positive polarity, 63 of 66 compounds and with negative polarity, 18 of 20 compounds were correctly identified in a spiked sample, while two compounds of a total of 88 fell out of the mass range. Furthermore, a four-stage process for identification was developed for the posttarget analysis lacking the retention time data. In the process, the number of candidate compounds was reduced by using the accurate mass of selected compounds in two steps (stages 1 and 2), structure-property relationships (stage 3) and isotope patterns of the analytes (stage 4). The process developed was validated by analysing wastewater samples spiked with 88 compounds. This procedure can be used to gain a preliminary indication of the presence of certain analytes in the samples. Nontarget analysis was tested by applying a theoretical mass spectra library for a wastewater sample spiked with six pharmaceuticals. The results showed a high number of false identifications. In addition, manual processing of the data was considered laborious and ineffective. Finally, the posttarget analysis was applied to a real wastewater sample. The analysis revealed the presence of six compounds that were afterwards confirmed with standard compounds as being correct. Three psycholeptics (nordiazepam, oxazepam and temazepam) could be tentatively identified, using the identification process developed. Posttarget analysis with UPLC-TOF-MS proved to be a promising method for analysing wastewater samples, while we concluded that the software for nontarget analysis will need improvement before it can be used in environmental analytical work with LC-TOF-MS systems.

Binding of organic xenobiotics to dissolved organic macromolecules: comparison of analytical methods

Partition coefficients (Koc) between water and dissolved organic material (DOM) were measured for benzo(a)pyrene (BaP), 3,3′,4,4′-tetrachlorobiphenyl (TCBP) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). DOM consisted of both natural humic substances from lake water and chlorolignin discharged from a bleached kraft pulp mill. The Koc values were determined by equilibrium dialysis and reverse-phase chromatography (SepPak). In addition, Koc values for BaP sorption to chlorolignin preparation were measured by thin layer chromatography (TLC), benzene extraction and fluorescence quenching methods. In all cases, equilibrium dialysis gave considerably higher Koc s compared to the reverse-phase measurement. In some samples, the difference was 10-fold for BaP and from threefold to fourfold for TCBP and TCDD. The Koc values for BaP binding to chlorolignin measured by benzene extraction, TLC and SepPak gave similar values while equilibrium dialysis and fluorescence quenching were similar. Thus, there are clear differnces between the methods which interfere with equilibrium between freely dissolved and bound xenobiotic fractions in the solution (extraction, TLC, SepPak) and those that do not (equilibrium dialysis and fluorescence quenching). Koc values were also used to predict the bioavailability (expressed as bioconcentration factors, BCF) of the xenobiotics to waterfleas (Daphnia magna) in the presence of DOM. Predicted BCF values were then compared to the results obtained in bioaccumulation experiments with Daphnia. Equilibrium dialysis gave a better prediction of the xenobiotic bioavailability than the reverse-phase chromatography measurements.

High-performance size-exclusion chromatography of lignin and its derivatives

Abstract High-performance size-exclusion chromatography of different lignins was studied using styrene—divinylbenzene copolymer gel columns. Acetylated and silylated model compounds were compared with underivatized samples to see the effect of possible adsorption or intermolecular association on the chromatography. The results showed that these phenomena did not cause any major interference in the chromatographic system described here. Derivatization increased the average molecular weights of lignins by more than was calculated from the hydroxyl content. We believe that the derivatization changed the hydrodynamic volume of the molecules, so the calibration compounds should also be changed. Alternatively, this might reflect some kind of association between the non-polar groups in the lignin molecules. For this reason the average molecular weights of lignin cannot be calculated by subtracting the mass of derivative groups from the average molecular weight of derivatized samples. Polystyrene calibration was found unsuitable in the analysis of lignin derivatives. The lignin model compound calibration seemed to be promising.

Aqueous size exclusion chomatography of industrial lignins

Abstract Aqueous size exclusion chromatography of lignin in effuents from the pulp industry was studied using several mobile phases at pH 8.9–12.0 and an organic semi-rigid high-performance gel (TSK PW type). The calibration and resolution calculations were based on the elution of sodium polystyrenesulphonates. The plate counts nad plate heights of the columns were also determined. The best results were obtained by using as eluent a low-ionic-strength bicarbonate buffer, pH 10.5, with addition of polyethylene glycol. Molecular weights obtained for spent bleaching liquor lignins and lignosulphonate were compared with results obtained with other chromatographic systems. The distribution curves were symmetrical and narrow. The chromatograms of spent E1-stage bleaching liquor obtained with system described and with a Bio-Gel P-60 column were very similar, but differed markedly from the obtained with a Sephadex LH-60 column.

Utilization of dimeric lignin model compounds by mixed bacterial cultures

SummaryThe degradation of dimeric phenylpropanoid lignin model compounds using mixed bacterial cultures was studied. The six model compounds contained the most common linkages of lignin: β-O-4, β-β, β-5, and β-1. The results indicate that it is possible to enrich bacteria which are able to degrade all these compounds. Bacteria were also able to use these dimers as the sole source of carbon for growth. In view of these results it seems probable that bacterial inability to degrade polymeric lignin is due to the physical properties such as the molecular size of lignin.

Biodegradation of two tetrameric lignin model compounds by a mixed bacterial culture

SummaryThe ability of a mixed bacterial culture to decompose two tetrameric lignin model com-pounds as a sole source of carbon and energy was investigated. The mixed bacterial culture con-sisted mainly of Gram negative rods. The tetram-ers contained two types of lignin substructures, namely the most abundant β-O-4 ether structure in lignin and also the 5-5 biphenyl structure.The tetramer (MW 638) containing two phe-nolic hydroxyls was decomposed readily; after 13 days of incubation, all intermediate products formed were almost totally decomposed. The non-phenolic tetramer (MW 666) was decom-posed much more slowly; after 53 days of incuba-tion, 5% of the substrate was unchanged. When both tetramers were degraded simultaneously, the non-phenolic tetramer was decomposed similarly to the phenolic tetramer.Determination of molecular weights of cata-bolic products showed that the degradation of the non-phenolic tetramer had proceeded at least to dimer level.SKF 525A, inhibitor of cytochrome P-450, caused one catabolic product to accumulate in the culture medium. This indicates involvement of cy-tochrome P-450 in the degradation pathway of the model compounds used.We conclude that this mixed bacterial culture was able to degrade the lignin model compounds used and that free phenolic groups seem to in-crease the biodegradability significantly.

Degradability of different lignins by bacteria

1 High-performance size exclusion chromatography was used to study metabolism of different lignins in liquid culture of bacteria. The culture degraded well tetrameric lignin model compounds but could not efficiently degrade lignins with a similar MW. The lignin concentration (measured äs absorbance at 280 nm) decreased at most by 24% (chloroethanol-acetic acid lignin) from that in uninoculated control. Kraft lignin, a second chloroethanol-acetic acid lignin preparation and chlorinated lignin from spent bleaching liquor (alkali stage) were the most recalcitrant. The lignin tetramer degrading culture (LTD-culture) mineralized 6% of C-(side chain)-labeled dehydrogenation polymer of lignin precursors (DHP). The molecular weight of the lignin preparations seemed not to be of major importance.

Identification of organic xenobiotics in urban aquatic environments using time-of-flight mass spectrometry.

Qualitative non-target and post-target analysis methods using gas chromatography-time-of-flight mass spectrometry were applied for analysing neutral and acidic organic xenobiotics in urban and suburban water samples. Ten water samples representing wastewater, stormwater and surface water matrices were collected and concentrated using solid phase extraction. Compound identification was performed using a spectral deconvolution program, accurate mass measurements and comparisons with library spectra. The non-target and post-target analyses identified 36 and 18 compounds, respectively. The identification of 10 compounds was afterwards confirmed with standard compounds. Organophosphate esters were the most abundant compound group detected. The combination of non-target and post-target analyses proved a useful tool in the tentative identification of xenobiotics in water samples. Post-target analysis can complement non-target analysis results at low analyte concentrations. Results showed that several organic xenobiotics originate in urban areas and accumulate in the environment. The wastewater sample produced the highest number of identified compounds, but most of these compounds were also found in stormwater samples from the city centre. Nearly all the compounds present in wastewater were additionally detected in the surface water sample taken 3 km downstream from the wastewater effluent discharge point. Only a few xenobiotics were otherwise detected in the surface water samples.

Qualitative nontarget analysis of landfill leachate using gas chromatography time-of-flight mass spectrometry

Nontarget analysis means that a sample is analysed without preselection of the studied analytes. While target analysis attempts to determine whether certain selected compounds are present in the sample, nontarget analysis is performed to explore what unknown compounds can be found. We developed a nontarget method using a landfill leachate sample as a complex test sample. The method was based on the use of a gas chromatograph-time-of-flight mass spectrometer (GC-TOF-MS) for final analysis and a deconvolution computer application for data processing. This nontarget analysis method was tested and validated by applying it to a landfill leachate sample spiked with 11 organic pollutants that were treated as unknowns. Sensitivity was found to be the most critical parameter affecting the success of nontarget analysis. The limit of identification (LOI) was 2500 ng L(-1) for four of the 11 compounds, 500 ng L(-1) for three compounds and 100 ng L(-1) for one compound. Three compounds were not detected in any of the spiked samples. A six-stage identification process was developed based on the spiking experiments. The process was based on the forward fit value of the library hit, the number of deconvoluted ions and the accurate mass scoring of the measured ions. The process was applied to an unspiked leachate water sample. Altogether, 44 compounds were tentatively identified in the sample. Elemental compositions of 36 components were additionally determined for which an unequivocal compound identification could not be given. Nontarget analysis with GC-TOF-MS is a promising method for the qualitative analysis of complex water samples. However, we conclude that the computer application for nontarget analysis needs improvement to decrease the amount of manual work needed in the identification process.