Processing and purity assessment of standards for the analysis of type-B trichothecene mycotoxins

The lack of reliable, certified calibrant solutions for the Fusarium mycotoxins deoxynivalenol (DON), 3-acetyl-DON (3-Ac-DON), 15-acetyl-DON (15-Ac-DON) and nivalenol (NIV) is a serious drawback in the already problematic area of trichothecene analysis. For this reason, purified DON, 3-Ac-DON, 15-Ac-DON and NIV standards were processed, the conditions required for their isolation and purification were optimised, and the crystalline toxins were thoroughly characterised. Several complimentary analytical methods were used to evaluate the identities of the mycotoxins and the types and amounts of impurities; results obtained from ¹ H and ¹³ C NMR spectra, as well as from IR-spectra, were in agreement with the literature. Elemental analysis revealed that the isolated NIV occurs as monohydrate. If this is not known it results in a weighing error of approximately 5%. Differential scanning calorimetry (DSC) was only successful for 15-Ac-DON, as the other trichothecenes decomposed during measurements. No traces of chloride, nitrate and sulphate were found by means of ion chromatography (IC). As expected UV absorption spectra for DON, NIV, 3-Ac-DON and 15-Ac-DON yielded _max values of 216, 217, 217 and 219 nm, respectively. Minor peaks due to impurities were observed by high performance liquid chromatography (HPLC) with UV detection. The main impurity peak in the DON sample was identified by LC-tandem mass spectroscopy (LC-MS/MS) as 4,7-dideoxy-NIV (7-deoxy-DON), which occurs at levels of approximately 1.4%. Gas chromatography (GC) was performed, coupled with either an electron capture detector (ECD), a flame ionisation detector (FID), or a mass spectrometric detector (MS); however, derivatisation prior to GC analysis makes the estimation of impurities difficult. LC-MS/MS was found to be unsuitable for quantifying levels of impurities. It can be concluded that high-purity (>97%) B-trichothecene standards were successfully processed and fully characterised for the first time.Dedicated to the memory of Wilhelm Fresenius     

Determination of pesticide residues in nonfatty foods by supercritical fluid extraction and gas chromatography/mass spectrometry: collaborative study

A collaborative study was conducted to determine multiple pesticide residues in apple, green bean, and carrot by using supercritical fluid extraction (SFE) and gas chromatography/mass spectrometry (GC/MS). Seventeen laboratories from 7 countries participated in the final study, and a variety of different instruments was used by collaborators. The procedure simply entails 3 steps: (1) mix 1.1 g drying agent (Hydromatrix) per 1 g frozen precomminuted sample, and load 4-5.5 g of this mixture into a 7-10 mL extraction vessel; (2) perform SFE for 20-30 min with a 1-2 mL/min flow rate of carbon dioxide at 0.85 g/mL density (320 atm, 60 degrees C); and (3) inject the extract, which was collected on a solid-phase or in a liquid trap, into the gas chromatograph/mass spectrometer, using either an ion-trap instrument in full-scan mode or a quadrupole-type instrument in selected-ion monitoring mode. The ability of GC/MS to simultaneously quantitate and confirm the identity of the semivolatile analytes at trace concentrations is a strong feature of the approach. The selectivity of SFE and GC/MS avoids the need for post-extraction cleanup steps, and the conversion of the CO2 solvent to a gas after SFE eliminates the solvent evaporation step common in traditional methods. The approach has several advantages, but its main drawback is the lower recoveries for the most polar analytes, such as methamidophos and acephate, and the most nonpolar analytes, such as pyrethroids. Recoveries for most pesticides are >75%, and recoveries of nonpolar analytes are still >50%. The (within-laboratory) repeatability relative standard deviation (RSDr) values of the recoveries are generally <15%. More specifically, the average results from the 9-14 laboratories in the final analysis of 6 blind duplicates at 3 concentrations for each pesticide are as follows: carbofuran in apple (75-500 ng/g), 89% recovery, 7% RSDr, 9% reproducibility relative standard deviation (RSDR); diazinon in apple (60-400 ng/g), 83% recovery, 13% RSDr, 17% RSDR; vinclozolin in apple (6-400 ng/g), 97% recovery, 13% RSDr, 18% RSDR; chlorpyrifos in apple (50-300 ng/g), 105% recovery, 11% RSDr, 13% RSDR; endosulfan sulfate in apple (150-1000 ng/g), 95% recovery, 15% RSDr, 17% RSDR; trifluralin in green bean (30-200 ng/g), 58% recovery, 11% RSDr, 27% RSDR; dacthal in green bean (60-400 ng/g), 88% recovery, 11% RSDr, 17% RSDR; quintozene in green bean (60-400 ng/g), 79% recovery, 13% RSDr, 18% RSDR; chlorpyrifos in green bean (50-300 ng/g), 84% recovery, 11% RSDr, 17% RSDR; p,p'-DDE in green bean (45-300 ng/g), 64% recovery, 14% RSDr, 27% RSDR; atrazine in carrot (75-500 ng/g), 90% recovery, 11% RSDr, 15% RSDR; metalaxyl in carrot (75-500 ng/g), 89% recovery, 8% RSDr, 12% RSDR; parathion-methyl in carrot (75-500 ng/g), 84% recovery, 14% RSDr, 15% RSDR; chlorpyrifos in carrot (50-300 ng/g), 77% recovery, 13% RSDr, 19% RSDR; and bifenthrin in carrot (90-600 ng/g), 63% recovery, 12% RSDr, and 25% RSDR. All analytes except for the nonpolar compounds trifluralin, p,p'-DDE, and bifenthrin gave average Horwitz ratios of <1.0 when AOAC criteria were used. These 3 analytes had high RSDr values but lower RSDR values, which indicated that certain SFE instruments gave consistently lower recoveries for nonpolar compounds. The collaborative study results demonstrate that the method meets the purpose of many monitoring programs for pesticide residue analysis, and the Study Director recommends that it be adopted Official First Action.     

Liquid chromatographic method with immunoaffinity column cleanup for determination of ochratoxin A in barley: collaborative study

A collaborative study was conducted to evaluate a liquid chromatographic (LC) method with immunoaffinity column cleanup for determination of ochratoxin A. The method was tested at 3 concentration levels of ochratoxin A in barley, which represent possible future European regulatory limits. The test portion was extracted with acetonitrile-water by blending at high speed. The extract was filtered, diluted with phosphate-buffered saline (PBS), and applied to an ochratoxin A immunoaffinity column. The column was washed with water and the ochratoxin A eluted with methanol. The solvent was then evaporated and the residue redissolved in injection solvent. After injection of this solution onto reversed-phase LC column, ochratoxin A was measured by fluorescence detection. Eight samples of low level naturally contaminated barley and 2 samples of blank barley (ochratoxin A not found at the limit of detection of 0.2 microg/kg at the signal-to-noise ratio of 3 to 1) were sent, along with ampules of ochratoxin A, calibrant, and spiking solutions, to 15 laboratories in 13 different European countries. Test portions were spiked with ochratoxin A at levels of 4 ng/g, and recoveries ranged from 65 to 113%. Based on results for spiked samples (blind duplicates) and naturally contaminated samples (blind duplicates at 3 levels), the relative standard deviation for repeatability (RSDr) ranged from 4 to 24%, and the relative standard deviation for reproducibility (RSDR) ranged from 12 to 33%. The method showed acceptable within- and between-laboratory precision, as evidenced by HORRAT values, at the low level of determination for ochratoxin A in barley.     

Combined phenyl silane and immunoaffinity column cleanup with liquid chromatography for determination of ochratoxin A in roasted coffee: collaborative study

A collaborative study was conducted to evaluate a liquid chromatography (LC) method for ochratoxin A using sequential phenyl silane and immunoaffinity column cleanup. The method was tested at 3 different levels of ochratoxin A in roasted coffee, which spanned the range of possible future European regulatory limits. The test portion was extracted with methanol and sodium bicarbonate by shaking for 30 min. The extract was filtered, centrifuged, and then cleaned up on a phenyl silane column before being eluted from the washed column with methanol-water. The eluate was diluted with phosphate-buffered saline (PBS) and applied to an ochratoxin A immunoaffinity column, which was washed with water. The ochratoxin A was eluted with methanol, the solvent was evaporated, and the residue was redissolved in injection solvent. After injection of this solution onto a reversed-phase LC apparatus, ochratoxin A was measured by fluorescence detection. Eight laboratory samples of low-level naturally contaminated roasted coffee and 2 laboratory samples of blank coffee (< 0.2 ng/g ochratoxin A at the signal-to-noise ratio of 3:1), along with ampules of ochratoxin A calibrant and spiking solutions, were sent to 15 laboratories in 13 different European countries. Test portions of the laboratory samples were spiked at levels of 4 ng/g ochratoxin A, and recoveries ranged from 65 to 97%. Based on results for spiked blank material (blind duplicates) and naturally contaminated material (blind duplicates at 3 levels), the relative standard deviation for repeatability (RSDr) ranged from 2 to 22% and the relative standard deviation for reproducibility (RSDR) ranged from 14 to 26%. The method showed acceptable within- and between-laboratory precision, as evidenced by HORRAT values, at the low level of determination for ochratoxin A in roasted coffee.     

Determination of molar absorptivity coefficients for major type-B trichothecenes and certification of calibrators for deoxynivalenol and nivalenol

This paper presents results from the European Commission-funded project Doncalibrant, the objective of which was to produce calibrators with certified mass fractions of the Fusarium toxins deoxynivalenol (DON), 3-acetyldeoxynivalenol (3-Ac-DON), 15-acetyldeoxynivalenol (15-Ac-DON), and nivalenol (NIV), in acetonitrile. The calibrators, available in ampoules, were sufficiently homogeneous, with between-bottle variations (s _bb) of less than 2%. Long-term stability studies performed at four different temperatures between −18 and 40 °C revealed no significant negative trends (at a confidence level of 95%). Molar absorptivity coefficients (in L mol⁻¹ cm⁻¹) were determined for all four toxins (DON: 6805 ± 126, NIV: 6955 ± 205, 3-Ac-DON: 6983 ± 141, 15-Ac-DON: 6935 ± 142) on the basis of a mini-interlaboratory exercise. The overall uncertainty of the calibrators’ target values for DON and NIV were evaluated on the basis of gravimetric preparation data and include uncertainty contributions from possible heterogeneity, storage, and transport. The Doncalibrant project resulted in the production of calibrators for DON (IRMM-315) and NIV (IRMM-316) in acetonitrile with certified mass fractions of 25.1 ± 1.2 μg g⁻¹ and 24.0 ± 1.1 μg g⁻¹, respectively. Both CRMs became commercially available from the Institute for Reference Materials and Measurements (IRMM, Geel, Belgium) at the beginning of 2007.     

Determination of phytase activity in feed by a colorimetric enzymatic method: collaborative interlaboratory study.

Fourteen laboratories participated in a collaborative study (coded fyt9404) and 13 laboratories participated in a study (coded fyt9410) to validate a colorimetric assay for determination of microbial phytase activity in feed. For each study, all laboratories received 6 laboratory samples provided by one commercial supplier (phytase activity levels within the range of 200-400 per kg) to be analyzed in duplicate. Method performance was calculated and statistical calculations were executed according to AOAC guidelines. Results from 3 laboratories for study fyt9404 and from one laboratory for study fyt9410 were excluded from statistical analysis because of invalid data determined during initial review by Youden pair, value versus laboratory. For study fyt9404, repeatability relative standard deviation (RSDr) values ranged from 6.2 to 8.6%, and reproducibility relative standard deviation (RSDR) values ranged from 14.1 to 27.6%. No outliers were identified. For study fyt9410, RSDr values ranged from 3.9 to 7.9%, and RSDR values ranged from 14.0 to 20.5%. With outliers excluded, RSDr values ranged from 2.5 to 7.9%, and RSDR values ranged from 14.0 to 20.5%.     

Determination of neutral lactase activity in industrial enzyme preparations by a colorimetric enzymatic method: collaborative study

Thirteen laboratories participated in a collaborative study to validate a colorimetric assay for determining neutral lactase activity in industrial enzyme preparations. Each laboratory received 5 duplicate samples with activity levels of 2000 and 5000 neutral lactase units provided by 4 commercial suppliers. Two laboratories did not return results. Method performance was calculated according to AOAC guidelines. From the 11 remaining laboratories, 3 were excluded from statistical analysis because of invalid data determined during initial review by Youden pair, value versus laboratory. Repeatability relative standard deviation (RSDr) values ranged from 3.20 to 8.62%, and reproducibility relative standard deviation (RSDR) values ranged from 8.77 to 16.35%. With outliers excluded, RSDr values ranged from 2.94 to 5.01%, and RSDR values ranged from 7.50 to 13.84%. The colorimetric enzymatic method for determining neutral lactase activity in industrial enzyme preparations has been adopted first action by AOAC INTERNATIONAL.     

Determination of Deoxynivalenol in Poultry Feed by Three Gas Chromatographic Detection Techniques

Three different gas chromatographic detection techniques were applied for the determination of deoxynivalenol (DON) present in poultry feed samples. Extraction and cleanup procedures were kept the same for GC–FID, GC–ECD and GC–MS methods. Although all three GC methods provided good and comparable results, but more attention was focused on GC–FID due to its lower cost and easy availability in many laboratories. Therefore, a short 15 m DB-1 short column was introduced for the determination of DON in poultry feed to reduce the time of analysis and initial cost of column. An inter-laboratory study for GC–FID was performed in two laboratories using four naturally DONS-contaminated feed samples and one spiked with standard. The relative standard deviations for repeatability (RSDr) and relative standard deviations for reproducibility (RSD_R) of naturally contaminated feed were in the range 5–23 and 11–24 %, respectively. The Horwitz Ratio (HORRAT) was less than 1.0 in each sample. From the spiking test, recovery, RSDr, RSD_R and value of HORRAT were 93, 5, 11 and 0.6 %, respectively. For GC–FID method, limit of quantification was found to be 6 μg kg^?1. Thus, GC–FID method using 15 m DB-1 capillary column is sensitive and validated analytical method for the determination of DON for poultry feed.     

Determination of the acid value of instant noodles: interlaboratory study

An interlaboratory study was performed to evaluate the method for determining the acid value of instant noodles, based on the Japanese Agricultural Standard (JAS), with extraction of lipid using petroleum ether at a volume of 100 mL to the test portion of 25 g. Thirteen laboratories participated and analyzed 5 test samples as blind duplicates. Statistical treatment revealed that the repeatability (RSDr) of acid value was <6.5%, and the reproducibility (RSDR) of acid value was <9.6%. The HorRat values (RSDR/predicted RSDR) were 1.2-1.8, where the RSDR and the predicted RSDR were obtained in terms of free fatty acids in the noodles per unit weight, using the equation [acid value = percent free fatty acids (as oleic) x 1.99] and the extracted lipid contents. This method was shown to have acceptable precision by the present study.     

Preparation of a calibrant as certified reference material for determination of the Fusarium mycotoxin zearalenone

In the field of mycotoxin analysis, substantial problems shown by high between-laboratory standard deviations and noncomparable and nontraceable results have been caused by the lack of proper calibrants for external calibration. During a large-scale Standard Measurement and Testing project of the European Commission (EC) dealing with preparation and certification of reference materials for determination of the mycotoxin zearalenone (ZON) in maize, a ZON calibrant in acetonitrile was prepared and checked for purity, homogeneity, and stability. Before certification, on the basis of preparation, the calibrant was checked in a mini-interlaboratory study by UV spectrophotometric determination. The molar absorptivities of ZON in acetonitrile at 236, 274, and 314 nm were established, and as a main result, a common reference wavelength of 274 nm with molar absorptivity of 12623 +/- 111 L/mol x cm can be recommended for ZON in acetonitrile. A concentration and expanded uncertainty of the ZON calibrant of 9.95 +/- 0.08 microg/mL was calculated as a preliminary value before final evaluation through the certification panel of the EC.     

Determination of docosahexaenoic acid and n-3 fatty acids in refined fish oils by 1H-NMR spectroscopy: IUPAC interlaboratory study

A high-resolution proton nuclear magnetic resonance (NMR) method for determining the concentration (mg/g) of docosahexaenoic acid (DHA), the molar proportion (mol%) of DHA, and the molar proportion of total n-3 fatty acids in fish oils was validated by an IUPAC interlaboratory study (the Commission VI-6 on Oils, Fats, and Derivatives WG 3/98). Thirteen laboratories from 5 countries tested 6 pairs of blind duplicate fish oils: a refined tuna oil, 2 extracted tuna oils, an extracted bonito oil, an extracted salmon oil, and an extracted sardine oil ranging from 9 to 30 mol% DHA and from 20 to 35 mol% n-3 fatty acids. Before 1 D-proton NMR measurements with 300-500 MHz instruments, oil samples were weighed and diluted with deuterochloroform solution containing ethylene glycol dimethyl ether as internal standard. To achieve precise performance, a detailed procedure for signal area measurement was described in the protocol, and all participants were instructed about the critical importance of following the protocol. Statistical performances with invalid and outlier data removed were as follows: repeatability relative standard deviations (RSDr) ranged from 0.91 to 2.62% and reproducibility relative standard deviation (RSDR) ranged from 1.73 to 4.27% for DHA concentration (mg/g); RSDr ranged from 0.39 to 2.06%, and RSDR ranged from 0.59 to 3.46% for mol% DHA; RSDr ranged from 0.23 to 0.90% and RSDR ranged from 0.85 to 2.01 % for mol% total n-3 fatty acids. The method is expected to be recommended by IUPAC.     

Determination of lasalocid sodium in animal feeds and premixes by reversed-phase liquid chromatography: collaborative study

A liquid chromatographic (LC) method for the analysis of lasalocid sodium in premixes, complete animal feeds, and trace-level feeds was collaboratively studied. The method employs a 0.5% HCI acidified methanol extraction followed by 20 min sonication in a water bath heated to 40 degrees C. Samples are then shaken on a mechanical shaker for 1 h and stored overnight, followed by an additional 10 min shaking the following morning. Sample extracts are diluted if necessary with extractant, filtered, and injected onto an LC system. Determination of all lasalocid homologs is by reversed-phase LC with fluorescence detection at 314 nm excitation and 418 nm emission. Eight samples of drug premixes, medicated feeds, and mineral supplements, along with 2 samples for trace-level analysis were sent to 20 collaborators in the United States, Canada, and The Netherlands. Study data were returned by 17 laboratories. Two additional supplemental trace-level samples and a blank feed were provided to 15 of the collaborating laboratories, and test data were received from all 15 participants. For the drug premixes, medicated feeds, and mineral supplements, RSDr values (within-laboratory repeatability) ranged from 1.2 to 19.9%, RSDR values (among-laboratory reproducibility) ranged from 3.4 to 32.3%, and HorRat values ranged from 0.35 to 3.73. For the trace-level samples, only lasalocid A, the predominant homolog comprising > 90% of the sum of all homolog peak area, was quantified. All laboratories correctly identified the analyte. Although some instrument response was reported by a number of laboratories for the blank feed, all but one laboratory's results were well below the 1 mg/kg limit of quantification. RSDr values for the initial 2 trace-level samples were excessive, ranging from 51.6 to 64.4%. RSDR values ranged from 51.6 to 75.7%, and HorRat values ranged from 3.6 to 4.0. Data for the initial trace-level samples indicated that the test samples were improperly prepared to ensure homogeneity, and a new set of supplemental samples was provided to collaborators, with significantly improved results. RSDr values for the 2 supplemental trace-level samples ranged from 1.6 to 2.5%, RSDR values ranged from 5.6 to 9.2%, and HorRat values ranged from 0.43 to 0.62.     

Determination of methylcellulose and hydroxypropyl methylcellulose food gums in food and food products: collaborative study

A collaborative study was performed to determine the reproducibility of a method for the determination of methylcellulose (MC) and hydroxypropyl methylcellulose (HPMC) in food. These widely used food gums possess unusual solubility characteristics and cannot accurately be determined by existing dietary fiber methods. The new method uses the enzyme-digestion procedure of AOAC Official Method 991.43. Digestate solutions must be refrigerated to fully hydrate MC or HPMC. The chilled solutions are filtered and analyzed by size-exclusion liquid chromatography. Collaborating laboratories received 28 samples containing MC or HPMC in the range of 0-100%. The sample set included blind duplicates of 5 food matrixes (bread, milk, fish, potato, and powdered juice drink). Cochran and Grubbs tests were used to eliminate outliers. For food samples containing MC, values for within-laboratory precision, repeatability relative standard deviation (RSDr), ranged from 4.2 to 16%, and values for among-laboratories precision, reproducibility relative standard deviation (RSDR), ranged from 11 to 20%. For HPMC samples, RSDr values ranged from 6.4 to 27%, and RSDR values ranged from 17 to 39%. Recoveries of MC and HPMC from the food matrixes ranged from 78 to 101%. These results show acceptable precision and reproducibility for the determination of MC and HPMC, for which no Official AOAC Methods exist. It is recommended that this method be adopted as AOAC Official First Action.     

Determination of ginsenosides (ginseng saponins) in dry root powder from Panax ginseng, Panax quinquefolius, and selected commercial products by liquid chromatography: interlaboratory study

Twelve collaborating laboratories assayed 4 products, namely, Panax ginseng, Panax quinquefolius, and 2 ginseng products, for 6 ginsenosides: Rb1, Rb2, Rc, Rd, Re, and Rg1. Collaborators also received a negative control for the recovery study. Pure ginsenosides were provided as reference standards for the liquid chromatography (LC) analysis and the system suitability tests. The LC analyses were performed on the methanol extract using UV detection at 203 nm. For P. ginseng, individual ginsenosides were consistent in their means; repeatability standard deviations (RSDr) ranged from 4.17 to 5.09% and reproducibility standard deviations (RSDR) ranged from 7.27 to 11.3%. For P. quinquefolius, the Rb1 and Rb2 ginsenosides were higher and lower in concentration than P. ginseng, with RSDr values of 3.44 and 6.60% and RSDR values of 5.91 and 12.6% respectively, and other analytes at intermediate precisions. For ginseng commercial products, RSDr values ranged from 3.39 to 8.12%, and RSDR values ranged from 7.65 to 16.5%. A recovery study was also conducted for 3 ginsenosides: Rg1, Re, and Rb1. The average recoveries were 99.9, 96.2, and 92.3%, respectively. The method is not applicable for the determination of Rg1 and Re in ginseng product at levels <300 mg/kg.     

Evaluation of a chemiluminescence method for measuring alkaline phosphatase activity in whole milk of multiple species and bovine dairy drinks: interlaboratory study

Alkaline phosphatase (ALP) is a ubiquitous enzyme in milk with time-temperature destruction similar to that of certain pathogens destroyed in pasteurization. Measurement of ALP to indicate proper pasteurization is a common practice. Recently the public health level for ALP was decreased to 350 mU/L, a level below the sensitivity of older colorimetric ALP methods. This study was conducted within the structure of the International Dairy Federation and the International Organization for Standardization to evaluate the reproducibility of the chemiluminescence method (Charm PasLite) for ALP at 50, 100, 350, and 500 mU/L in whole milk of multiple species to meet new regulations in the United States and proposed regulations in the European Union (EU). Fifteen laboratories from 8 countries evaluated bovine, goat, sheep, and buffalo milk, bovine skim milk, 20% fat cream, and 2% fat chocolate milk. At ALP levels of 350 and 500 mU/L, the average relative standard deviation for repeatability (RSDr) was 7.5%, and the average relative standard deviation of reproducibility was (RSDR) 15%. For ALP at 100 and 50 mU/L, the average RSDr values were 10.5 and 12.6%, respectively, and the average RSDR values were 18 and 25%, respectively. The limit of detection was 20 mU/L. Results are comparable to those obtained with other enzymatic photo-activated system methods such as the fluorometric method. Results indicate that the method is suitable for measuring ALP in the milk of multiple species and in dairy drinks at U.S. and proposed EU levels.     

Determination of arsenic in seafood by electrothermal atomic absorption spectrometry after microwave digestion: NMKL collaborative study

Eight laboratories participated in an interlaboratory method performance (collaborative) study of a method for the determination of arsenic in foodstuffs of marine origin by electrothermal atomic absorption spectrometry after wet digestion using a microwave oven technique. The study was preceded by a practice round of familiarization samples. The method was tested on 8 materials (cod roe, krill, blue mussel, saithe, scampi, cod fillet, shrimp, and cod extract) ranging in As content from 2 to 75 mg/kg. The materials were sent to participants in the study as blind duplicates, and the participants were asked to perform single determinations on each sample. Repeatability relative standard deviations (RSDr) for As ranged from 6.8 to 17.4%. Reproducibility relative standard deviations (RSDR) ranged from 7.6 to 24%. The highest RSDR value was found for the sample with the highest concentration of As.     

Determination by glyphosate and aminomethylphosphonic acid in crops by capillary gas chromatography with mass-selective detection: collaborative study.

A collaborative study was conducted to validate a method for the determination of glyphosate and aminomethylphosphonic acid (AMPA) in crops. The analytes are extracted from crops with water, and the crude extracts are then subjected to a cation exchange cleanup. The analytes are derivatized by the direct addition of the aqueous extract into a mixture of heptafluorobutanol and trifluoroacetic anhydride. The derivatized analytes are quantitated by capillary gas chromatography with mass-selective detection (MSD). The collaborative study involved 13 laboratories located in 5 countries; 12 laboratories returned valid data sets. The crops tested were field corn grain, soya forage, and walnut nutmeat at concentrations of 0.050, 0.40, and 2.0 mg/kg. The study used a split-level pair replication scheme with blindly coded laboratory samples. Twelve materials were analyzed, including 1 control and 3 split-level pairs for each matrix, 1 pair at each nominal concentration. For glyphosate, the mean recovery was 91%, the average intralaboratory variance, the repeatability relative standard deviation (RSDr), was 11%, and the interlaboratory variance, the reproducibility relative standard deviation (RSDR), was 16%. For AMPA, the mean recovery was 87%, the RSDr was 16%, and the RSDR was 25% at mg/kg levels.     

Evaluation of a method to determine flavonol aglycones in Ginkgo biloba dietary supplement crude materials and finished products by high-performance liquid chromatography: collaborative study

An interlaboratory study was conducted for evaluation of a method to determine the flavonol aglycones quercetin, kaempferol, and isorhamnetin in Ginkgo biloba products. The method calculates total glycosides based on these aglycones formed after acid hydrolysis. Twelve matrixes were chosen for study by 12 collaborating laboratories in 2 countries. Test materials included crude leaf material, standardized dry powder extract, single and multiple entity finished products, ethanol and glycerol tinctures, and National Institute of Standards and Technology (NIST) standard reference materials (SRMs). Results from 11 laboratories were used for the final calculations. Eight of the 12 matrixes evaluated produced acceptable results for total flavonol glycosides, with HorRat scores ranging from 1.31 to 2.05; repeatability relative standard deviations (RSDr) from 1.46 to 4.14; and reproducibility relative standard deviations (RSDR) from 4.67 to 9.69. These 8 matrixes consisted primarily of simple dosage forms (e.g., dry powder extracts, crude leaf samples, liquid extracts, and SRMs) and a single tablet product (Ginkgo Awareness). Four additional matrixes, consisting of 3 tablets and 1 soft gel product (Ginkgold, Ginkoba, Ginkogen, and Ginkgo Phytosome, respectively), showed greater total flavonol glycoside HorRat scores in comparison, ranging from 2.39 to 5.13, with RSDr values from 2.83 to 8.16, and RSDR values from 8.53 to 20.4. Based on the results presented here, the method is recommended for Official First Action for determination of total flavonol glycosides calculated from quercetin, kaempferol, and isorhamnetin in dry powder extracts, crude leaf material, liquid extracts, and a select finished product, Ginkgo Awareness.     

Measurement of resistant starch by enzymatic digestion in starch and selected plant materials: collaborative study

Interlaboratory performance statistics was determined for a method developed to measure the resistant starch (RS) content of selected plant food products and a range of commercial starch samples. Food materials examined contained RS (cooked kidney beans, green banana, and corn flakes) and commercial starches, most of which naturally contain, or were processed to yield, elevated RS levels. The method evaluated was optimized to yield RS values in agreement with those reported for in vivo studies. Thirty-seven laboratories tested 8 pairs of blind duplicate starch or plant material samples with RS values between 0.6 (regular maize starch) and 64% (fresh weight basis). For matrixes excluding regular maize starch, repeatability relative standard deviation (RSDr) values ranged from 1.97 to 4.2%, and reproducibility relative standard deviation (RSDR) values ranged from 4.58 to 10.9%. The range of applicability of the test is 2-64% RS. The method is not suitable for products with <1% RS (e.g., regular maize starch; 0.6% RS). For such products, RSDr and RSDR values are unacceptably high.     

Determination of sodium in foods by flame atomic absorption spectrometry after microwave digestion: NMKL interlaboratory study

Nine laboratories participated in an interlaboratory method performance (collaborative) study of a method for the determination of sodium in foods by flame atomic absorption spectrometry after wet digestion, using a microwave oven technique. Before the study, the laboratories were able to practice on samples with defined sodium levels (pretrial test). The method was tested on a total of 6 foods (broccoli, carrot, bread, saithe fillet, pork, and cheese) with sodium concentrations of 1480-8260 mg/kg. The materials were presented to the participants in the study as blind duplicates, and the participants were asked to perform single determinations for each sample. The repeatability relative standard deviations (RSDr) for sodium ranged from 1.9 to 6.5%. The reproducibility relative standard deviations (RSDR) ranged from 4.2 to 6.9%. The HorRat values ranged from 0.9 to 1.6.