Collaborative study of an automated method for the determination of crude protein in animal feeds.

An automated macro Kjeldahl instrument determines per cent protein at the rate of 20 samples/hr. The methodology involved is similar to the present official final action Kjeldahl method, sec. 7.016. The 2 methods were compared in a collaborative study. Sixteen animal feeds, 4 meats, tryptophan, ammonium dihydrogen phosphate NBS standard, and ammonium sulfate primary standard were analyzed by the participating laboratories. The data were agreement between the 2 methods. The automated method has been adopted as official first action for the determination of crude protein in feeds, plants, and cereal foods.     

Modified method for carbadox in feeds: collaborative study

The official first action method for carbadox in swine feed, 42.C01-42.C04, was modified in 2 respects. First, the samples were leached overnight at room temperature instead of boiled for 1 hr. This change avoided problems with overheating and excessive evaporation. Second, the dilution scheme for samples spiked with carbadox standard solution was changed to give absorbance values that were within the optimum working range of all types of spectrophotometers. The modified procedure was collaboratively studied by 21 laboratories. The repeatability standard deviation (sigma0) and reproducibility standard deviation (sigmax) were sigma0 = 0.00029% and sigmax = 0.00056% (8.9% of grand mean) for feeds containing 0.00617% carbadox; and sigma = 0.0012% and sigmax = 0.0019% (9.3% of grand mean) for feeds containing 0.0198% carbadox. The between-laboratory variance ratio was significant for feeds containing 0.0198% carbadox. The mean per cent of intent values for feeds containing 0.00617% carbadox and 0.0198% carbadox were 102% and 104%, respectively. In general, the statistical results were comparable to those previously obtained for the official first action method. Consequently, the modified procedure is not recommended as a replacement for the official first action method.     

Microbiological determination of neomycin in feeds: collaborative study

A modification of the AOAC microbiological determination of neomycin in feeds was collaboratively studied by 12 laboratories. The official method was modified by substituting a constant salt concentration diluent for the feed extract diluent, preparing the agar medium in tris buffer, and performing the test with a monolayer plating system. Each laboratory performed single assays on 8 samples in a randomized sequence. The samples included duplicates of a cattle and swine feed at 2 different marketed concentrations. The mean recovery across all laboratories was 110.7% of theory with a range of means of 69.4-128.6 across the 12 laboratories. The results of one laboratory and 2 additional values from different laboratories were deemed outliers and excluded from statistical analysis. The statistical analysis gave a confidence interval of +/- 26% for individual assays.     

Liquid chromatographic method for determination of furazolidone in premixes and complete feeds: collaborative study

A liquid chromatographic (LC) method for determining furazolidone in finished feeds and premixes was collaboratively studied. Finished feed sample is extracted with acetone-water (93 + 7) on a Goldfisch apparatus, extracting solvent is removed, and the residual material is dissolved in warm DMF. A solution of tetraethylammonium bromide is added, the fat layer is removed, and the sample is clarified by filtration and injected onto a reverse phase LC system with detection at 365 nm. Premixes, extracted by shaking with DMF and diluted so that the final furazolidone concentration is about 55 micrograms/mL, are chromatographed and detected the same as finished feed samples, using a mobile phase of acetonitrile-2% acetic acid (20 + 80). Ten commercial feed samples were preweighed and supplied to 14 collaborators. The 5 matched pairs were chosen to represent the following allowed levels: 0.0055, 0.022, 0.033, 2.2, and 22%. Two familiarization samples at the 0.0055 and 11% levels were also supplied. Instructions called for a single analysis of each sample. Two results were eliminated by the Dixon test. The coefficients of variation, following treatment by the ranking test, ranged from 2.0 at the 22% level to 6.5 at the 0.0055% level. Calculated F-values are not significant (P greater than 0.01) except for the 0.0055% level samples extracted overnight. This method has been adopted official first action.     

Pyridine extraction method for determination of bacitracin in feeds: collaborative evaluation.

Seventeen laboratories evaluated the pyridine extraction method and neomycin-sensitized agar for the determination of zinc and MD bacitracin in swine and broiler rations at 10 and 100 g/ton. The method was also applied to the analysis of 2 premixes labeled 50 g/lb (MD bacitracin) and 40 g/lb (zinc bacitracin). Bacitracin activity was determined on each of 2 days with 2 dilutions on each day. No significant difference was found between dilutions within a day or between days for each sample. The type of bacitracin or type of feed did not significantly affect results. The difference in results between MD and zinc bacitracin in premixes approached significance. The large coefficients of variation for premixes (ca 13%) and complete feeds (ca 15--30%) indicate operational problems. The main difficulty was evaporation of pyridine. Some laboratories were not able to evaporate it completely, whereas others lost bacitracin activity, probably due to high temperature of drying. The pyridine extraction method as in 42.200 and 42.204 should be discontinued.     

Spectrophotometric determination of pyrantel tartrate in swine feeds by method of standard additions: collaborative study

The spectrophotometric method for pyrantel tartrate in swine feeds was collaboratively studied. Twenty-seven laboratories assayed feeds containing 0.0103, 0.0965, and 0.7902% pyrantel tartrate. Repeatability (sigmao) and reproducibility (sigmax) standard deviations were: sigmao = 0.00068%, sigmax = 0.00105% (10% of grand mean) for 0.0103% pyrantel tartrate level; sigmao - 0.0065%, sigmax = 0.0090% (10% of grand mean) for 0.0965% pyrantel tartrate level; and sigmao = 0.0415%, sigmax = 0.0743% (10% of grand mean) for 0.7902% pyrantel tartrate level. The mean theoretical recovery values for feeds containing 0.0103, 0.0965, and 0.7902% were 100, 97, and 96%, respectively. The method was adopted as official first action for feeds or concentrates containing 0.0106-0.8811% pyrantel tartrate.     

Determination of crude protein in animal feeds, using block digestion followed by steam distillation: collaborative study.

A method consisting of digesting animal feeds in a block digestor and determining ammonia by steam distillation followed by titration has been evaluated and compared with the official final action Kjeldahl method, 7.016. Fifteen laboratories analyzed 5 feed samples and lysine monohydrochloride. Statistical analysis showed that results from the 2 methods were comparable. The distillation technique has been adopted as official first action as an alternative technique for ammonia determination from the digest of the official final action block digestor method, 7.B11.     

Determination of amprolium in feeds: collaborative study.

The official final action method, 42.028--42.032, for determining amprolium in feeds was modified by a change in the preparation of aluminum oxide for chromatography. A premix containing 0.5% amprolium was collaboratively studied by the modified and the official methods. Compared with the modified method, 87.7% of the drug was recovered from the premix by using the official method. The modification makes possible the assay of premixes as well as finished feeds. The official final action method has been modified to incorporate this change.     

Liquid chromatographic method for determination of azinphos-methyl in formulated products: collaborative study

A liquid chromatographic method for determination of azinphosmethyl (Guthion) in formulated products has been developed. Samples are dissolved in acetonitrile and analyzed by reverse-phase chromatography using n-butyrophenone as an internal standard. The method was subjected to a collaborative study involving 15 participating laboratories. Each collaborator was furnished with reference standard, internal standard, and blind duplicate samples of Guthion 50% wettable powder (50WP), 3 flowable (3F), and emulsifiable concentrate (2L and 2S) formulations. Collaborators were instructed to evaluate the method by peak height measurements only. Relative standard deviations for reproducibility (RSDR) were 1.11, 6.28, 2.47, and 1.17% for the 50WP, 3F, 2L, and 2S formulations, respectively. The method has been approved interim official first action for determination of azinphos-methyl in the 50WP, 2L, and 2S formulations.     

Enzymatic-spectrophotometric method for determination of cholinesterase activity in whole blood: collaborative study

Ten collaborating laboratories assayed 4 blind duplicate pairs of whole bovine blood for cholinesterase activity. The 4 sample pairs ranged from normal (100%) to severely organo-phosphorus-inhibited (less than 10%) activity. Collaborators also received commercially available human lyophillized serum as an external control and a chromate solution to evaluate spectrophotometer performance. The Ellman kinetic assay was performed on a 1:1000 dilution of the whole blood in pH 8.0 phosphate buffer. The method monitors the increase in absorbance at 412 nm caused by formation of 5-thio-2-nitrobenzoic acid (yellow reaction product). Repeatability standard deviations (RSDr) ranged from 4.30 to 14.2%; reproducibility standard deviations (RSDR) ranged from 6.99 to 19.3%. The lower limit of detection was estimated to be 0.10 mumole/mL/min. The method has been approved interim official first action by AOAC.     

Headspace gas chromatographic method for determination of ethanol in canned salmon: collaborative study

Six laboratories collaboratively studied a headspace gas chromatographic method for determination of ethanol in the aqueous phase of canned salmon. Ethanol is determined by a headspace sampling technique with tert-butanol as the internal standard, using a gas chromatograph equipped with a Super Q column and a flame ionization detector. With outliers excluded, the mean recoveries from samples spiked with 25.1 and 78.4 ppm ethanol were 112 and 110%, respectively. For the 4 sample pairs quantitated, repeatability coefficients of variation ranged from 1.42 to 4.25% and reproducibility coefficients of variation from 2.55 to 8.09%, with 3 of the 4 reported values less than 5%. The method has been adopted official first action.     

Liquid chromatographic-atomic absorption spectrophotometric method for determination of methyl mercury in seafood: collaborative study

A previously developed method that uses a simplified sample preparation procedure and atomic absorption detection of liquid chromatographic eluates for the determination of methyl mercury in seafood has been collaboratively studied. The unique feature of the method involves the use of a specially designed interface for the generation of mercury vapor. Methyl mercury is isolated from the blended sample by chloroform elution from a diatomaceous earth-hydrochloric acid column. The organomercury compound is then extracted into a small volume of 0.01M sodium thiosulfate solution. An aliquot of this solution is injected onto a Zorbax ODS column and eluted with methanol-ammonium acetate solution (3 + 2), pH 5.7, containing 0.01% mercaptoethanol. Mercury is detected by flameless atomic absorption spectrophotometry using the interface. The samples analyzed in the study were unspiked swordfish, unspiked and spiked lobster, and unspiked and spiked tuna. The spiked samples contained methyl mercury both above and below the U.S. Food and Drug Administration guideline level of 1 microgram Hg/g. Reproducibility relative standard deviations ranged from 10.5% at 1 microgram Hg/g to 18.2% at about 0.1 microgram Hg/g. Accuracy, measured by comparison to reference values obtained by the Associate Referee, ranged from 94.4 to 99.6%. The method has been adopted official first action.     

Atomic absorption spectrophotometric method for determination of polydimethylsiloxane residues in pineapple juice: collaborative study

An atomic absorption spectrophotometric method for determination of polydimethylsiloxane (PDMS) residues in pineapple juice was collaboratively studied by 9 laboratories. PDMS residues are extracted from pineapple juice with 4-methyl-2-pentanone and the extracted silicone is measured by atomic absorption spectrophotometry using a nitrous oxide/acetylene flame. Collaborators analyzed 5 samples including 1 blind duplicate. Reproducibility relative standard deviations (RSDR) were 13.1% at 31 ppm, 6.9% at 18 ppm, 14.8% at 7.9 ppm, and 16.1% at 4.9 ppm PDMS. The method has been approved interim official first action by AOAC.     

Determination of acid-detergent fiber and crude protein in forages by near-infrared reflectance spectroscopy: collaborative study

A collaborative study was conducted to determine the standard error of difference among laboratories for near-infrared reflectance spectroscopic (NIRS) determination of acid-detergent fiber (ADF) and crude protein in forages. The 6 participating laboratories were members of the USDA/ARS National Near-Infrared Reflectance Spectroscopy Forage Research Project. The NIRS calibration equations were developed in the Associate Referee's laboratory for crude protein and ADF and were transferred to the instrument in each of the other collaborating laboratories. The calibration set included over 650 diverse forage samples with crude protein and ADF calibration data; the validation set included 94 samples of bermudagrass. Among-laboratory reproducibility for the NIRS method, calculated as the relative standard deviation for reproducibility (RSDR), was 1.14% for ADF and 0.42% for crude protein. The variance component for among-laboratory variation (coefficient of variation) was 2.54% for ADF and 2.89% for crude protein. These results confirm that it is possible to calibrate, validate, and transfer (NIRS) equations and data among laboratories for the accurate determination of ADF and crude protein, and thereby demonstrate that NIRS can be used as a standard method for the analysis of forages. The method has been adopted official first action.     

Comparison of HgO and CuSO4/TiO2 as catalysts in manual Kjeldahl digestion for determination of crude protein in animal feed: collaborative study

Because of environmental concerns about HgO, and because of lengthy digestion requirements for HgO and CuSO4, interest in alternative catalysts for the Kjeldahl determination of animal feeds remains high. A digestion system using a mixed CuSO4/TiO2 catalyst has been found to reduce digestion times to 40 min. A collaborative study was carried out to compare this system to the official AOAC HgO method, 7.015. Thirty-eight samples, consisting of blind duplicates of closely matched pairs and 2 standard materials, were analyzed once by each method. Results were received from 13 laboratories. Means and standard deviations of individual samples were comparable, with an overall difference of grand means of 0.005% protein. With only one exception, analyses of variance showed no significant method difference at the 95% confidence level. The CuSO4/TiO2 method has been approved interim official first action as an alternative method for determination of crude protein in animal feed.     

Automated method for the determination of niacin and niacinamide in cereal products: collaborative study.

In a collaborative study, an automated method for the determination of niacin and niacinamide in cereal products was compared with the official final action microbiological (43.121-43.125) and chemical (43.044-43.046) methods. Ten samples of cereal products, including enriched flour, yeast-leavened baked products, fortified breakfast cereals, and baked pet food products, were submitted to 14 laboratories. Nine laboratories reported values by the automated method, 6 reported values by the microbiological method, and 7 reported values by the chemical method. The results from the microbiological method were not subjected to analysis of variance because of the unusually large between-laboratory variation. The between-laboratory coefficients of variation for the automated and chemical methods were 10.90 and 10.18%, on the basis of results from 7 and 4 laboratories, respectively. There was no significant (p greater than 0.05) difference between methods when results from the 4 laboratories who used both methods were compared. The automated chemical method has been adopted as official first action.     

Microbiological plate and turbidimetric assays of chlortetracycline in feeds: collaborative study.

The manual and automated turbidimetric assays and a modified official plate assay for chlortetracycline (CTC-HCl) in feed were collaboratively studied. Three feed samples (swine feed, 100 g CTC-HCl/ton; premix I, 20 g each of CTC-HCl and sulfamethazine/lb, and 10 g penicillin/lb; and premix II, 50 g CTC-HCl/lb) were analyzed at 2 dilutions. Twelve laboratories conducted the plate assay; 8 laboratories the manual turbidimetric method; and 7 laboratories, the Autoturb analysis. Within a method, there was no significant difference between dilutions. Between methods, there was a significant difference between the manual turbidimetric plate assays only for swine feed. However, the same sample dilutions or the average values of the 2 dilutions for both methods showed no statistical difference. Among the collaborators, the slope of CTC-HCl standard curve varied between about 2.0 and 3.0 for the plate method. The turbidimetric assay has been adopted as official first action for feeds containing larger than or equal to 20 g CTC-HCl/lb.     

Gas chromatographic-chemiluminescence method for determination of volatile N-nitrosamines in minced fish-meat and surimi-meat frankfurters: collaborative study

A collaborative study was carried out on a solid-phase extraction method for separating volatile N-nitrosamines, particularly N-nitrosodimethylamine (NDMA), from combination minced fish or surimi-meat frankfurters with detection by gas chromatography-chemiluminescence (thermal energy analyzer). The results from the 10 collaborators were evaluated using the most recent AOAC guidelines for determining outliers and for the analysis of variance. For NDMA, repeatability standard deviations, sr, ranged from 0.56 to 2.25; repeatability relative standard deviations, RSDr, ranged from 8.9 to 11.5%. Reproducibility standard deviations, SR, for NDMA ranged from 1.40 to 6.49, and reproducibility relative standard deviations, RSDR, ranged from 24.2 to 28.9%. Our data compared favorably to the reproducibility (RSDR) curve of Horwitz. The method has been adopted official first action by AOAC.     

Fluorometric determination of histamine in tuna: collaborative study

Six samples of canned tuna, albacore, yellowfin, and skipjack, in water or oil pack were analyzed in duplicate by a fluorometric method and the AOAC colorimetric method. For the fluorometric method, recoveries of histamine added to acceptable tuna averaged 99% with a range of 91 to 107%. Agreement between laboratories for the analyses of decomposed tuna containing 20-200 mg histamine/100 g sample was excellent. Results from the fluorometric method are comparable with those from the AOAC colorimetric method; the fluorometric method has been adopted as official first action.     

Semiautomated method for riboflavin in food products: collaborative study

A collaborative study was conducted comparing a semiautomated riboflavin method with a manual riboflavin method for 10 food products. Six laboratories provided results from the semiautomated method and 16 laboratories used the manual technique. The semiautomated method was more repeatable within a laboratory than was the manual method. The semiautomated method results compared favorably with the manual method for all 10 products.