7851 - Laserfiche WebLink

Home Browse Search

Selenium in the Razorback Sucker Evaluation of Muscle Plug Technique For the muscle plug technique to be useful, selenium concentrations in muscle plugs collected from dissimilar areas of a fish must not be significantly different. Because taking multiple plugs for comparison from an endangered razorback sucker was not pennitted, common carp (Cyprinus carpio) and f1annelmouth sucker (Catostomus laIipinnis) were collected from the same geographical area and used for this purpose.. ~ese species were collected by electroshocking and pre- served WIth Ice, and muscle plugs were taken in the laboratory. MusCle plugs were also taken from a frozen razorback sucker that had been killed accidentally in 1991. Muscle plugs were taken from mid-dorsal (1-2 cm from dorsal fin), posterior (near the tail), and anterior Gust behind the gills) areas of each fish. These tissues were collected and handled in the manner described above. It was of interest to determine if the selenium concentration found in a muscle plug, which is typically <50 mg including skin and scales, was representative of the concentration found in a conventional muscle fillet sample. To make this comparison, about 20 g of muscle tissue wi~ attached skin and scales was taken adjacent to each muscle plug usmg a clean scalpel. Determination of Selenium in Fish Muscle Plugs Neutron activation was used for the analysis of muscle plugs because of the e~tremely small sample mass (average -20 mg). All sample preparatIon prior to neutron activation analysis was conducted by the Midwest Science Center (MSC). Each muscle plug was transferred ~rom its ~riginal container into a small polyethylene vial (0.7 g capac- Ity) proVided by the University of Missouri Research Reactor (MURR) staff. Vials were precleaned by stepwise washing with acetone, nitric acid. and deionized water. Each muscle plug was positioned and pressed flat against the vial bottom with a clean glass rod. All vials were left open and placed in the tray chamber of a Virtis 20-SRC lyophilizer and frozen to - 30oC. Samples were lyophilized to constant weight, because lyophilization greatly reduces 190 in the irradiated sample and significantly enhances measurement precision. Dried mus- cle plug weights averaged 21.4 mg (range: 5.6 - 42.8 mg). After recording of final sample weight, an expandable, clean polyethylene plug was inserted into the vial, against which the vial lid was com- pressed shut. The polyethylene plug served to maintain constant sam- ple geometry. Samples were transported to MURR for the determina- tion of the radionuclide Se77m (McKown and Morris 1978). Because the major contributions of gross activity in irradiated tissue are produced from sodium and chlorine, standard solutions cOntaining nonnallevels of selenium, sodium, and chlorine were prepared for the fish m~le matrix. Irradiation standards were prepared by placing small allquots of the standard solutions onto cellulose. Samples and standards were successively placed in a shuttle rabbit and irradiated for 5 see at a thermal neutron flux of about I X 1014n X cm-2 X see-I. The pneumatic transfer facility has a delivery time to the counting station of about 7 see. The returned shuttle rabbit was opened quickly, and the sample vial was transferred to a 45-cm3 germanium-lithium ~~. ray spectrometer system. All samples were analyzed by as-see irradiauon, IS-see decay, and 20-see count, with a sample to detector dis~ giving less than 10% deadtime at the analyzer (about 3 cm). Selemum values <JLg) were obtained by direct comparison of peak areas obtained for the samples to the average peak areas obtained for a set of standards. Determination of Selenium in Fish Muscle Because fish muscle samples were not limited in mass, as were muscle plugs, a more conventional approach was used for preparation and 323 analysis, which was conducted by the Environmental Trace Substances Research Center (ETSRC) in Columbia, MO. Each sample of about 20 g of{"ISh muscle tissue was lyophilized in a LabconcoFreeze Drier 8 unit until constant weight was achieved. Perrentmoisture was deter- mined from water loss by lyophilization. After drying, samples were homogenized with a blender or Spex Industries Model 8000 mixer/mill with tungsten carbide vial and balls. About 0.5 g of tissue was placed into a 100 mL Kjeldahl flask to which 15 mL of concentrated nitric acid and 2;5mL of concentrated perchloric acid were added (sub-boiled quality). The flask was placed on low heat until the evolution of dark red fumes ceased. The heat was increased until the nitric acid began refluxing, and refluxing was continued overnight. Heat was increased until nitric acid was driven off and perchloric reaction occurred, as evidenced by the formation of dense white fumes. At this time, sam- ples were removed from the heat and cooled. Two mL of concentrated hydrochoric acid were added to each flask, and the flasks were heated until the contents reached the boiling point. Flasks were then removed from the heat, cooled, and the contents diluted with deionized water to a fmal volume of 50 mL. Digestates were stored in 50 mL polyethylene bottles. Selenium was detennined by hydride generation atomic absorption spectroscopy (Clinton 1977), using a Perkin-Elmer Model 3030 spec- trophotometer equipped with a Varian VGA-76 hydride generation accessory. Reagents were 50% sub-boiled hydrochloric acid and 0.6% sodium borohydride in 0.5% sodium hydroxide. Samples were diluted, when necessary, with 10% sub-boiled hydrochloric acid. Standards were prepared by dilution of Fisher 10000ppm stock with 10% hydro- chloric acid in the range of 0-20 ppb. The instrument was standardized to read directly in ppb using S I = 5.0 and S2 = 20.0. Standardization was checked with a quality control solution of known selenium concen- tration. The detection limit was determined by reading the 0 standard ten times and computing twice the standard deviation of the mean. Samples were analyzed by taking an integrated reading for 3 see after a plateau was reached (about 45 see). A recorder was used to monitor peak shape and to assist in determining when the integration should be started. Standardization was checked after every 8-15 samples, and about 10% of the samples were analyzed by the method of additions to check for interference from the sample matrix. Statistical Treatment The variability among dorsal, posterior, and anterior muscle plugs and tissue from the same fish was expressed by computing the coefficient of variation. The analysis of variance (ANOV A) was a randomized complete block design in which the linear statistical model contains the effects of species, location, muscle type, and the interaction oflocation X muscle type. Species represented the complete block. Differences among treatment means were ascertained using Fisher's Least Signifi- cant Difference (Snedecor and Cochran 1972). Results and Discussion Quality Control - Selenium by Neutron Activation Accuracy and precision of the irradiation method were esti- mated in two ways. First, for an independent blind accuracy check, four samples of tissue reference or research materials (-50 mg each) were prepared, sealed, and submitted with the samples to MURR. Each sample was identified only bya four- digit number. Second, MURR conducted its own accuracy and method precision check by the irradiation of four samples of NIST (National Institute of Standards and Technology) Bovine Liver SRM (Standard Reference Material) 1577. Results of