2. AOACRIChemContMethods-2018Awards

1448 Pang et al.: J ournal of AOAC I nternational V ol. 98, N o. 5, 2015

Table 11. Comparison of method efficiency for determination of 20 pesticides in aged oolong tea by GC/MS/MS with and without correction

Avg. C., μg/kg

S r

, μg/kg

RSD r

, %

S R

, μg/kg

RSD R

, %

HorRat

No. of labs

Without correction

With correction

Without correction

With correction

Without correction

With correction

Without correction

With correction

Without correction

With correction

Without correction

With correction

No.

Pesticide

1

Trifluralin

14

325.8 386.3

19.1

19.1

5.9

4.9

113.0 112.0

34.7

29.0

1.8

1.6

2

Tefluthrin

14

154.5 195.1

8.9

8.9

5.8

4.6

48.7

47.4

31.5

24.3

1.5

1.2

3

Pyrimethanil

14

167.0 203.2

9.2

9.2

5.5

4.5

49.2

48.1

29.4

23.6

1.4

1.2

4

Propyzamide

13

192.0 216.9

10.9

10.9

5.7

5.0

42.5

41.3

22.1

19.1

1.1

0.9

5

Pirimicarb

13

179.9 220.9

9.3

9.3

5.2

4.2

46.4

44.5

25.8

20.2

1.2

1.0

6

Dimethenamid

14

72.4

89.1

3.9

3.9

5.4

4.4

20.4

19.9

28.1

22.3

1.2

1.0

7

Fenchlorphos

14

298.8 373.8

19.0

19.0

6.4

5.1

101.9 97.8

34.1

26.2

1.8

1.4

8

Tolclofos-methyl

13

169.7 204.6

8.6

8.6

5.1

4.2

45.5

44.4

26.8

21.7

1.3

1.1

9

Pirimiphos-methyl

14

163.1 200.8

15.6

15.6

9.6

7.8

53.4

51.8

32.7

25.8

1.6

1.3

10

2,4’-DDE

13

631.5 747.5

33.3

33.3

5.3

4.5

196.0 189.2

31.0

25.3

1.8

1.5

11

Bromophos-ethyl

14

155.6 191.3

9.0

9.0

5.8

4.7

47.0

44.9

30.2

23.4

1.4

1.1

12

4,4’-DDE

14

630.1 743.1

30.8

30.8

4.9

4.1

183.4 176.3

29.1

23.7

1.7

1.4

13

Procymidone

14

186.1 208.5

9.8

9.8

5.3

4.7

54.2

53.6

29.2

25.7

1.4

1.3

14

Picoxystrobin

14

373.6 425.9

19.4

19.4

5.2

4.6

85.3

82.4

22.8

19.3

1.2

1.1

15

Quinoxyfen

14

182.4 209.6

9.2

9.2

5.0

4.4

54.1

53.5

29.6

25.5

1.4

1.3

16

Chlorfenapyr

13 1511.9 1685.2 84.7

84.7

5.6

5.0

332.7 323.4

22.0

19.2

1.5

1.3

17

Benalaxyl

14

194.1 224.0

8.9

8.9

4.6

4.0

48.1

47.4

24.8

21.2

1.2

1.1

18

Bifenthrin

14

183.8 212.9

9.3

9.3

5.0

4.4

51.0

49.9

27.7

23.5

1.3

1.2

19

Diflufenican

13

191.3 216.7

11.6

11.6

6.1

5.4

41.6

40.1

21.7

18.5

1.1

0.9

20

Bromopropylate

14

387.2 438.2

22.1

22.1

5.7

5.0

84.9

82.4

21.9

18.8

1.2

1.0

differences in test results for two parallel samples of No. 1 and No. 2 green tea by GC/MS/MS analysis, with normal recoveries for 20 pesticides of No. 1 sample but with no detection of any test results for 20 pesticides of No. 2 sample. Likewise, regarding LC/MS/MS, there are four samples fortified with 20 pesticides, of which No. 1 and No. 2 are green tea parallel fortification samples and No. 6 and No. 7 are oolong tea parallel samples. Recoveries for 20 pesticides of No. 1 sample are normal, but none of the 20 pesticides has been detected in No. 2 sample. The same is true of No. 6 and No. 7 fortification samples, with no detection of any pesticides either, belonging to overall deviations. Nineteen outliers from Laboratory 18 all came from No. 6 and No. 7 oolong tea fortification samples, and the test results from these two samples are 37% on the average compared with those from other laboratories. As for this issue, the collaborator has failed to find out the cause. Seven outliers from Laboratory 27 came from the five pesticides in No. 1 and No. 2 green tea fortification samples, and it is judged tentatively that these pesticides were interfered with, causing accidental errors. To summarize the above-mentioned analysis, in the GC/MS/MS test results, overall deviations occurred with several specific samples for Laboratory 21 and Laboratory 18, resulting in two-thirds of the outliers concentrated in these two laboratories. Errors from other laboratories are accident deviations distributed in multi-kinds of pesticides. (c)  By LC/MS/MS.— Supplemental Tables 21–23 show that 2957 effective data were obtained from determination of 20 pesticide residues in eight samples (excluding two blank samples) by 24 laboratories; Grubbs and Dixon test application

was adopted, and 57 outliers were discovered (making up 1.9%). These 57 outliers came from 13 laboratories. There are 12 outliers from Laboratory 24, accounting for 21.1%; 11 from Laboratory 28, accounting for 19.3%; nine from Laboratory 25, making up 15.8%; and six from Laboratory 30, accounting for 10.5%. Outliers from nine other laboratories total 19, accounting for 33.3%. Outliers from each of these nine laboratories are less than six, caused by accidental errors. Eleven of 12 outliers were reported by Laboratory 24 from No. 9 and No. 10 oolong tea age samples, and the analytical results from these two samples are 40% lower than those from the other laboratories. There are 11 outliers reported by Laboratory 28. These outliers came from seven pesticides in four fortification samples (No. 1, 2, 6, and 7) of two kinds of tea. They are mainly traced to accidental errors in the determination of certain pesticides. There are nine outliers reported by Laboratory 25. They came from seven pesticides in five samples (No. 1, 2, 6, 7, and 10) traced to accidental errors with certain pesticides. Six outliers reported by Laboratory 30 came from green tea fortification samples No. 1 and 2, and the test results of acetochlor, benalaxyl, bensulide, kresoxim-methyl, and picoxystrobin in these two samples are about 40% greater than those from other laboratories, which also belong to the category of accidental errors with certain pesticides. To summarize the above-mentioned descriptions, there are 57 outliers from the 2957 effective data by LC/MS/MS, accounting for 1.9%. Fifty-seven outliers came from 13 laboratories,