Surrogate Recoveries Research Articles

Collection of two-ring aromatic hydrocarbons, chlorinated phenols, guaiacols, and benzenes from ambient air using polyurethane foam/Tenax-GC cartridges

Two-ring aromatic hydrocarbons and chlorinated phenols, guaiacols, and benzenes and were measured in ambient air at three locations using a high volume air sampler equipped with a glass fiber filter followed by a polyurethane foam/Tenax-GC (PUF-TX) sandwich cartridge. Isotopically-labeled surrogate compounds were spiked on to the front of the cartridge before sampling to assess collection efficiency in the field. Surrogate recoveries from ten sample extracts averaged 87 ± 19% for 2,4-dichlorophenol-d 3; 111 ± 26% for pentachlorophenol- 13C 6; 40 ± 22% for naphthalene-d 8; and 75 ± 18% for biphenyl-d 10. With the exception of the most volatile surrogate, naphthalene-d 8, collection efficiency of the PUF-TX traps was acceptable. The highest average concentrations at all three locations were observed for the aromatic hydrocarbons followed by the chlorinated benzenes, chlorinated phenols and the chlorinated guaiacols. This technique allows for ambient monitoring of these more volatile compound classes with commercially available high volume air sampling equipment.

A multi‐laboratory evaluation of analytical methods for estimating bioconcentratable contaminants in effluents, tissues and sediments

AbstractThe U.S. Environmental Protection Agency has developed draft guidance for the assessment and control of bioconcentratable contaminants in surface waters that are not covered by water quality criteria. The guidance document contains analytical procedures for identification and quantification of bioconcentratable contaminants in effluents, sediments, and tissues. These analytical procedures were the subject of a multilaboratory round‐robin study in 1992. The results reported in this article are from five of the 12 laboratories that completed the round‐robin study. Analyses of data from the five laboratories show that analytical variance explains the large observed intralaboratory variability in surrogate recoveries. Sample variance is minor relative to analytical and laboratory variances. All five laboratories isolated and positively identified only three of eight spiked compounds at >50% frequency using the proposed analytical procedures and gas chromatography–mass spectrometry library search method. The inability of the library search method to detect some of the reference spike chemicals is a concern because the methods are intended to detect unknown bioconcentratable chemicals. In general, the multistep procedure for each matrix was complex, prone to analyte loss, time intensive, and analytically demanding. Specific procedural problems and alternative methods are discussed, and recommendations for improvement are presented.

A MULTI-LABORATORY EVALUATION OF ANALYTICAL METHODS FOR ESTIMATING BIOCONCENTRATABLE CONTAMINANTS IN EFFLUENTS, TISSUES AND SEDIMENTS

The U.S. Environmental Protection Agency has developed draft guidance for the assessment and control of bioconcentratable contaminants in surface waters that are not covered by water quality criteria. The guidance document contains analytical procedures for identification and quantification of bioconcentratable contaminants in effluents, sediments, and tissues. These analytical procedures were the subject of a multilaboratory round-robin study in 1992. The results reported in this article are from five of the 12 laboratories that completed the round-robin study. Analyses of data from the five laboratories show that analytical variance explains the large observed intralaboratory variability in surrogate recoveries. Sample variance is minor relative to analytical and laboratory variances. All five laboratories isolated and positively identified only three of eight spiked compounds at >50% frequency using the proposed analytical procedures and gas chromatography–mass spectrometry library search method. The inability of the library search method to detect some of the reference spike chemicals is a concern because the methods are intended to detect unknown bioconcentratable chemicals. In general, the multistep procedure for each matrix was complex, prone to analyte loss, time intensive, and analytically demanding. Specific procedural problems and alternative methods are discussed, and recommendations for improvement are presented.

An improved analysis for chlorinated pesticides and polychlorinated biphenyls (PCBs) in human and bovine sera using solid-phase extraction.

Chlorinated pesticides and polychlorinated biphenyls (PCBs) remain public health concerns because of their unresolved health impact and their persistence in humans. Current epidemiological studies of cancer, non-Hodgkins lymphoma, and endocrine disruption in National Center for Environmental Health (NCEH) laboratories require exposure assessment of many analytes in thousands of people. Previous methods of analyzing pesticides and PCBs in serum have proven inadequate for timely processing of the number of samples required for epidemiological studies. A new method that involves solid-phase extraction (SPE) and cleanup followed by dual-column gas chromatographic separation and electron capture detection has been developed. Nine surrogate compounds were added to the serum prior to sample workup to provide quality assurance for the SPE steps. These surrogates mimic the chemistry of the analytes in the extraction, cleanup, and gas chromatographic analysis steps. To increase selectivity, extracts were injected onto two gas chromatographs with different capillary columns, a DB-1701 and a DB-5. Recoveries of 17 pesticides, 28 PCB congeners, and one polybrominated biphenyl congener ranged from 40 to 80%. Recoveries from this procedure were found to be similar to those from the previously used liquid-liquid extraction method. Correlation of analyte and surrogate recoveries were compared to examine the ruggedness of the technique. The SPE method was found to provide improved sample throughput by a factor of 15.

Evaluation of the goulden large-sample extractor for acidic compounds in natural waters

The Goulden Large-Sample Extractor has received extensive use for monitoring and surveillance surveys of natural waters impacted by pulp and paper mills and agricultural runoff water. However, there are concerns about whether this sampler, which was originally developed for extractions of hydrophobic polycyclic aromatic hydrocarbons, polychlorinated biphenyls, and other organochlorines, is suitable for sampling polar acidic compounds. The sampler was evaluated for recovery of surrogates for resin acids, fatty acids, herbicide acids, and chlorophenols from natural waters. Performance tests conducted in this work indicated that three surrogate compounds with K p (C DCM /C water pH2 ) values from 16,700 to 1,260 were extracted from pH 2-adjusted 20-L water samples with an average recovery of 83.6%. The surrogate compounds with K p values less than 1,000 were extracted with significantly lower recoveries. The variability ranged from 10 to 36% relative standard deviation. Recoveries and variability compared favorably with reported recoveries and variabilities for neutral pesticide surrogates. Specific performance criteria (percent recoveries ± standard deviation, number of determinations in parentheses) observed for the surrogates 2,4,6-tribromophenol, heptadecanoic acid, O-methylpodocarpic acid, dichlorophenylacetic acid, and 4-bromophenol were 89.5 ± 24.0 (17), 82.8 ± 21.7 (18), 78.4 ± 14.8 (18), 41.9 ± 8.5 (16), and 22.1 ± 8.1 (19), respectively. Low recoveries of the 4-bromophenol surrogate may be due in part to side reactions with diazomethane. As a result, 4-bromophenol is not recommended as a surrogate. These values can be used to provide guidelines for acceptable surrogate recoveries and validation of extractions of polar acidic compounds.

An MSD-based method for the detection of chlorinated dibenzo-p-dioxins and chlorinated dibenzofurans in fish

A stream-lined method for the detection of polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) in fish samples is described. The samples were mechanically blended with anhydrous sodium sulphate and then packed in a glass column. The lipid fraction was eluted from the column with dichloromethane. The described extraction technique recovered 13C-labelled surrogate spikes as effectively as a method based on acidic digestion and liquid phase extraction. Lipids were removed from the fish extracts by size exclusion chromatography followed by chromatography on a mini acid/base silica column. The extract was further enriched by high performance liquid chromatography (HPLC) on basic alumina and activated carbon. The analytes were identified and quantified by gas chromatography combined with mass selective detection. Two sets of fish samples (n = 16 & n = 6) were used to evaluate the method: surrogate recoveries ranged from 67.5% to 75.5%, and analytical precision ranged from 13.9 to 26.4%. Tritiated 2,3,7,8-T 4CDD was used to conveniently trouble shoot the method.

Non-statistically-based scoring system to evaluate interlaboratory performance for the analysis of dioxins and furans in fish tissues

Many regulatory and other organizations are concerned with the quality of analytical data generated at ultra-trace levels by contract laboratories. These organizations are often forced to contract out work because of a lack of inhouse expertise to perform sample analysis, especially for the determination of the chlorinated dioxins and furans. Further, such organizations may have limited capability to select a competent laboratory to perform the required work, or to evaluate the quality of analytical data generated by contract laboratories. The problem is magnified for some determinations, such as dioxins/furans in fish tissue, where no certified reference materials exist and therefore analytical accuracy is difficult to establish. A non-statistically-based laboratory data scoring system was developed to address this problem. Use of the scoring system was tested by application to results of a five-laboratory fish tissue round-robin for dioxins/furans. The scoring system proved to be flexible and effective, as it incorporated laboratory detection limits and surrogate standard recoveries in addition to the dioxin/furan concentration data.

The analysis of human blood for seventy-five PCB congeners by dual column capillary-gas chromatography

The method presented here is a good routine method for the determination of PCBs in blood. Analysis of total PCBs showed good coefficients of variation and good surrogate recoveries. The maximum standard deviation found in the trplicates is 14%; this can be considered the precision of the method. The coefficients of variation for individual congeners were also good. However, verification of congener identity needs further assessment.

Analysis of pulp mill effluent contaminants in water, sediment, and fish muscle— chlorophenols and related compounds

ABSTRACT: The need for multimedia environmental studies on pulp mill effluent contaminants was addressed by refining and validating analytical protocols for the analysis of chlorinated phenols, guaiacols, catechols, phenolic aldehydes, and polymethoxybenzenes in water (µg/L range), sediment, and fish tissue samples (tenths of µ/g range). The methods, with common features for the different media, involved extraction with t‐butyl methyl ether/dichloromethane solutions, with ascorbic acid added as an antioxidant. In the case of fish tissue, sample cleanup by size exclusion chromatography was carried out to reduce lipid interference. Derivatization with acetic anhydride /pyridine in organic solution yielded the acetates, which were analyzed by GC‐SIM‐MS. Excellent precision (SD = 2–15%) and satisfactory recoveries (42–115%) were achieved for most of the 32 contaminants evaluated. Application of these methods to samples from the northern Alberta region yielded satisfactory results (surrogate recoveries = 94–134%).

Analysis of pulp mill effluent contaminants in water, sediment, and fish bile—fatty and resin acids

ABSTRACT: Validation work was carried out to develop analytical protocols for fatty and resin acids in water, sediment, and fish bile samples in the µg/L, tenths of µg/g, and µg/g, range respectively. The methods, with similar steps for the different media, involved extraction from a pH 5 medium with t‐butyl methyl ether/dichloromethane solutions, with previous enzymatic hydrolysis in the case of bile samples. Concentration of the extract and derivatization with diazomethane in organic solution yielded the methyl esters, which were analyzed by GC‐SIM‐MS. Good recoveries (60–110%) and excellent precision (SD = 2–11%) were obtained for the twelve target compounds. Application of these methods to samples from the northern Alberta region yielded satisfactory results (surrogate recoveries = 75–110%).

Oxidation of phenolic acid surrogates and target analytes during acid extraction of natural water samples for analysis by GC/MS using EPA method 625

Zero or low phenolic acid surrogate recoveries were occasionally observed in natural water samples analyzed by EPA method 625. These low recoveries were usually caused by the oxidation of the phenolic acid surrogates during acid extraction. Three oxidation products of the phenolic acid surrogates were identified. Iodocyclohexanol, an artifact formed during extraction with cyclohexene-inhibited methylene chloride, was usually present in samples that contained oxidizing agents. Matrix effects on the recoveries of phenolic target analytes were investigated in samples having low phenolic acid surrogate recoveries. The results showed that recoveries for phenols with strong electron-withdrawing substituents, such as a nitro group, were not affected. However, decreased recoveries were observed for phenols having electron-donating substituents or weak electron-withdrawing substituents. It is demonstrated that manganese(III,IV) oxides are most likely the oxidizing agents for the groundwater samples studied. If samples were extracted according to EPA method 625, in the presence of a large amount of manganese, Mn{sup 2+} dissolved in the sample would be oxygenated to manganese(III,IV) oxides during base extraction, which in turn would oxidize the phenols during acid extraction.

Determination of 2,3?7,8-Tetrachlorodibenzo-p-Dioxin in Fish, Using Electron Capture Gas Chromatography with Confirmation by Mass Spectrometry: Interlaboratory Study

An interlaboratory study of the determination of 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) in fish was conducted by 6 analysts in 4 laboratories using high resolution gas chromatography with electron capture detection (HRGC-EC) for quantitative screening analysis. Samples consisted of 3 Great Lakes channel catfish homogenates containing different levels of bioincurred 2,3,7,8-TCDD; 1 of these was prepared in duplicate and another was prepared both with and without standard 2,3,7,8-TCDD fortification for a total of 5 samples per set. All methods used included addition of 1,3,7,8-TCDD surrogate (to correct for procedural losses) followed by ethanolic KOH digestion and hexane extraction. Certain cleanup steps used, including sulfuric acid washing and multidimensional column liquid chromatographic procedures, varied among laboratories. Mean HRGC-EC results for the bioincurred residues were 56.6, 25.2, and 7.7 pg/g (ppt) with corresponding relative standard deviations (RSDs) of 9.1, 18.6, and 53.2%. Average determination of standard 2,3,7,8-TCDD from the fortified sample (corrected for surrogate recoveries averaging 74.6%) was 106% of the added amount (30.9 pg/g) with 11.0% RSD. HRGC-multiple ion detection mass spectrometry (MS), monitoring 12 ions, was used for confirmation. With the exception of several results from 1 analyst, HRGC-MS and HRGC-EC quantitations were in good agreement. All but 1 result reported met all of the MS identity criteria.

Evaluation of a Large-Volume Extractor for Determining Trace Organic Contaminant Levels in the Great Lakes

Abstract The design, shipboard operation and performance of a large-volume extractor is described. A total of 224 water samples were extracted at up to 1 L.min−1 with dichloromethane during monitoring cruises on lakes Ontario, Erie, Huron and Superior in the spring of 1986. To evaluate the performance of the extractor, a spiking solution containing surrogate standards was added during the course of the extraction. Five surrogate spikes were added in the field: endrinketone, 1,3-dibromobenzene, 1,3,5-tribromobenzene, 1,2,4,5-tetrabromobenzene and 2,3,5,6-tetrachlorobiphenyl. Extractor performance was evaluated for both ambient and centrifuged water samples using recovery and reproducibility as criteria. Analytical results could then be excluded on the basis of surrogate recoveries exceeding two standard deviations from the mean. Recoveries ranged from 89–130% and coefficients of variation were generally ≤25%.