Search

Twenty-three laboratories from 13 countries in continental Europe, Scandinavia, the British Isles, and North America participated in a polychlorinated biphenyl (PCB) Check Sample Program conducted under the auspices of the International Council for Exploration of the Sea (ICES). PCBs were determined in unspiked and spiked (1.00 mg Aroclor 1254/kg oil) herring (Clupea harengus harengus) oil by the participants, each of which used his or her own cleanup and quantitation techniques; a common Aroclor 1254 mixture was used as a standard and a common quantitation technique was used for comparative purposes. Results for the unspiked oil ranged from 0.48 to 3.416 mg PCB/kg oil, while spiked oil results ranged from 0.70 to 3.891 mg/kg. Calculated spike recoveries ranged from 22 to 136%. Serious deficiencies were found in most steps in the procedures. No evidence was found to support the use of a common PCB standard or a common method of calculation using packed column chromatography. The chromatographic stationary phase used appeared to affect the PCB levels obtained. Florisil cleanup adsorbent yielded higher mean results for both unspiked and spiked oils than did alumina. Large coefficients of variation were found (25-50%), the principal source of which was systematic error (interlaboratory).

Twenty-five laboratories were sent 2 materials, one an acetone powder of lobster digestive gland, the other, the oil which had been extracted during preparation of the powder. Each laboratory was requested to measure the levels of a suite of polycyclic aromatic hydrocarbons in both materials. The response was poor with only 10 laboratories submitting results. Both intra- and interlaboratory precisions were poor; the interlaboratory error was so great as to preclude statistical analysis of the error. Relative standard deviations for oil results determined by liquid chromatography ranged from 39 to 96%.

Analysts participating in a multi-laboratory comparative study were asked to identify 4 chlorobiphenyls (CBs) supplied in neat form, to measure the amounts of these present in spiked and unspiked fish oil, to measure other CBs and organochlorines in the unspiked fish oil, and to compare results for their own standard solution and those for standard solutions prepared from the supplied CB compounds. Comparisons were done for a common supplied method and the individual methods used in each laboratory. Participants had no trouble identifying 2,2',5,5'-tetrachlorobiphenyl, 2,2',4,5,5'-pentachlorobiphenyl, and 2,2',4,4',5,5'-hexachlorobiphenyl. Most misidentified 2,2',3,4,5-pentachlorobiphenyl. Approximately one-half of the quantitative comparisons between participants' standards and those prepared from the supplied CBs differed by more than 10%. Two standards prepared from one of the CBs differed by an average of 6.6% (range 0.0-24.5%). Recoveries of added CBs from the fish oil ranged from 24 to 294% for spikes of 63-85 ng/g with no clear distinction between results for the common method vs individual laboratory methods. The common methodology gave a lower coefficient of variation (CV) for most other CBs and organochlorines, but in most cases the CVs for the individual CBs were not smaller than that for Aroclor 1254 equivalents.

A new, simple, and rapid cleanup procedure is described for di-2-ethylhexyl phthalate (DEHP) residues in fish lipids. Extracted are chromatographed on small alumina:sulfuric acid-impregnated alumina (layered) columns after gel permeation chromatography of the fish lipid extracts on BioBeads SX-3. Quantitative analyses were carried out by electron capture gas chromatography using 2 columns. Spiked DEHP recoveries varied from 79.3% in mackerel to 86.3% in herring. DEHP concentrations were determined in plaice, eel, redfish, herring, cod, and mackerel tissues and ranged from trace (less than 0.001 microgram/g) to approximately 10 microgram/g (wet wt basis). Ethanolic KOH saponification of the purified DEHP fraction, resulting in the disappearance of the DEHP peak, was used as a confirmatory test. Limited studies with dibutyl, di-n-heptyl, and diethyl phthalates suggest that the method can be made more versatile.

A method is described for determining selenium in fish tissues, meat, cereals, milk powder, and other materials by flameless atomic absorption spectrophotometry. Samples are solubilized in HNO3 and atomized in a graphite furnace in the presence of nickel nitrate. Recoveries of 0.500 and 1.000 microgram selenium added to several fish samples averaged 99.0 and 98.3%, respectively, with standard deviations of 5.3 and 4.0. Results agreed with those obtained for samples previously analyzed by fluorometry, and with results for NBS Standard Reference Material. The detection limit was 3 ng/ml solution and 50 ng/g sample.

A simple, rapid, easily automated method is described for the determination of polycyclic aromatic hydrocarbons (PAHs) in shellfish such as American lobster (Homarus americanus) and blue mussel (Mytilus edulis). PAHs are extracted from small amounts (1-8 g) of tissue by saponification in 1N ethanolic potassium hydroxide followed by partitioning into 2,2,4-trimethylpentane. This solution is evaporated just to dryness by rotary evaporation and the residue is dissolved in cyclohexane-dichloromethane (1 + 1) for gel permeation chromatography (GPC) on Bio-Beads SX-3. The GPC procedure is ideal as a screening method in the range 25-18 000 ng PAHs/g tissue. If individual PAH measurements are required, the appropriate GPC fraction is collected and PAHs are separated by reverse phase liquid chromatography (LC) with fluorometric detection. Individual PAHs at concentrations as low as 0.25-10 ng/g can be determined. Recoveries of added fluoranthene, pyrene, benz[a]anthracene, chrysene, benzo[e]pyrene, benzo[b]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, dibenz[a,h]anthracene, benzo[ghi]perylene, and indeno[1,2,3-cd]pyrene were quantitative, with relative standard deviations ranging from 0.0 to 16.9%.