AOAC Final Action Methods in 2017

J OSEPH ET AL . : J OURNAL OF AOAC I NTERNATIONAL V OL . 101, N O . 3, 2018 837

(i) Injection volume. — 100 µL. (j) Gradient program. — See Table 2016.02B .

For ready-to-feed liquid samples, the sample weight used for extraction is used for the calculation. Report the results to three significant figures (i.e., µg/100 g), or convert to other units as required.

H. QC

J. Repeatability

(a) Check system suitability by injecting Standard 3 five times. The RSD should be ≤ 2%. (b) Run the calibration standards at the beginning and end of the sequence (slope drift ≤ 2%). (c) The six-point calibration should give a correlation coefficient ≥ 0.997. (d) Test one in five samples in duplicate. The duplicates should be within the method repeatability. (e) Inject one of the calibration standards after every five sample injections. (f ) Analyze a reference sample (e.g., NIST SRM 1849a) in duplicate. (g) Identification of biotin peak is based on absolute retention time. Spectrum scan can be used for peak purity confirmation if required. (h) Perform three high-level recoveries with every new batch of IACs. The chromatography software will automatically calculate the concentration of the sample (µg/100 g), provided the concentration of the standards (µg/100 mL), sample weight (g), and dilution are entered correctly. Manual calculation can also be performed by using the following equation: Biotin or Biocytin ð µ g = 100 g Þ = ð Sample Area × Dilution Þ ð Slope × Sample weight in grams Þ where Dilution = 10 (i.e., 100 × 1 ÷ 10): sample is diluted with water to 100 mL, with 10 mL used for IAC cleanup, to a final volume of 1 mL for HPLC analysis (the dilution will be 20 if 5 mL is used for IAC cleanup); Slope = the valid slope calculation based on concentration on the x -axis and peak area on the y -axis; and Sample weight = the calculated powder equivalent (g) using the following equation for reconstituted powder samples: Sample weight ð powder equivalent Þ in grams = ð W 1 × W 2 Þ = W 3 where W 1 = the weight of powder sample (g); W 2 = the weight of powder and water or total slurry weight (g); and W 3 = the weight of slurry taken for analysis (g). I. Calculation and Reporting

The difference between the results of duplicate portions of the same sample tested at the same sequence should not exceed 6%of the mean result.

K. Reproducibility

Downloaded from https://academic.oup.com/jaoac/article-abstract/101/3/831/5653969 by guest on 05 May 2020

The difference between the results of duplicate determinations tested on different days should not exceed 12% of the mean result.

L. Uncertainty of Measurement

Uncertainty of the method was calculated as 7%, using an appropriate statistical procedure (i.e., square root of the sum of squares of the errors, expressed as a percentage).

M. LOQ

LOQ was calculated based on the lowest working standard concentration and the dilution factor as follows: LOQ = ð 1 × 100 Þ = ð 20 × 50 Þ = 0 : 1 µ g = 100 g ð 1 ppb Þ where 1 = the lowest working standard concentration of 1 µg/100 mL; 100 = the volume (mL); 20 = the 20 g sample; 50 = the volume (mL) loaded on the IAC; and 1 = the final volume (mL). The laboratories were requested to analyze biotin and biocytin in the samples; however, no sample was found to have biocytin, and only biotin results were reported and evaluated in the study. All the biotin results presented in the report are expressed on the basis of micrograms per 100 g sample as received from AOAC. Three of the 12 laboratories (Laboratories 10, 11, and 12) could report only the results of practice samples before the due date of the study (January 27, 2017). In general, the results from Laboratory 7 showed low bias, which was investigated further. The possible reasons for the low bias could be the practice that was followed in sample preparation, especially the steps related to loading of the IAC and the elution technique, in which triple backflush is specified. The investigation from Laboratory 7 reported that the analysts did not follow the backflush technique effectively, which may have caused the incomplete elution of biotin from the cartridges. Laboratory 7 also reported that they had deviated from the protocol in the reconstitution of powder samples by taking 10 g powder in 80 mL. The results from Laboratories 7, 10, 11, and 12 are included in the MLT report and used for statistical evaluation of the results in order to calculate precision of the analytical method. Results and Discussion

Table 2016.02B. Gradient program

Flow rate, mL/min

Mobile phase A, %

Mobile phase B, %

Mobile phase C, %

Time, min

0.0

0.6 0.6 0.8 0.8 0.6 0.6

90 90

10 10

0 0

18.0 18.5 24.0 24.5 27.0

0 0

0 0

100 100

90 90

10 10

0 0