AOAC Final Action Methods in 2017

560 P ACQUETTE & T HOMPSON : J OURNAL OF AOAC I NTERNATIONAL V OL . 101, N O . 2, 2018

Table 7. Comparison of straight RSD percentages measured in two independent MLT studies using ICP-MS on similar sample sets a

Parameter or Product

Na Mg

P

K

Ca

Cr

Mn

Fe

Cu

Zn

Se Mo

SPIFAN matrix: Method 2011.19 MLT (3)

No. of labs

6

6

6

6

6

8

7

7

7

7

8

8

19.2 b

Adult Milk Powder

6.6 6.9 6.8 6.8 6.4 6.5 6.6 7.3 6.2 6.5

7.6 8.1 8.3 7.3 4.4

3.9 3.3 3.6 4.2 4.0

5.1 5.2 5.7 5.1 5.5

3.6 3.2 3.0 3.4 3.5

4.7 3.4 4.8 4.6 4.8

3.0 2.1 2.4 2.6 2.3

5.8 1.9 5.3 2.4 5.9

6.0 7.9 2.6 6.9 6.9 3.1 7.0 4.7 8.9 8.1 7.7 9.3 4.9 3.7 2.0 2.8 4.1 2.7 3.8 2.3 6.5 3.5 5.9 3.8 5.9 2.9 5.5 2.3 4.2 3.1 3.0 2.1 8 8

Large c

Infant Powder Hydrolyzed Milk

Adult Powder, Low-Fat

6.8

Child Powder

11.4 14.3

Infant Elemental Powder Adult RTF, High-Protein

7.6 8.1 14.0 b 8.5 7.9 10.9 b

4.8 33.8 b 5.1 48.1 b

25.5 b 26.1 b

11.9 b

2.8 14.2 b

7.9 6.1 2.2

Downloaded from https://academic.oup.com/jaoac/article-abstract/101/2/536/5653941 by guest on 05 May 2020

Adult RTF, High-Fat

8.9 3.8

3.4 2.0

9.7 2.0

SRM 1849a

2.4 3.0

1.9

1.7

1.5

4.0

SPIFAN matrix: Method 2015.06 MLT d

No. of labs

10

10

10

10

10

8

10

10

10

10

Large c Large c

Infant Milk Powder

3.0 2.5 2.6 4.7 4.1 2.9 3.1 2.3 2.8 2.4 4.7 4.0

2.7 4.2 4.5 2.7 4.1 6.8

3.4 4.4 3.8 2.7 2.7 4.9

4.7 4.1 4.1 3.1 3.7 4.6

5.5 2.8 3.6 3.8 2.4 4.2

4.3 5.4 3.3 2.7 5.1 5.5 7.0 1.1

4.6 6.9 4.8 5.2 6.7 3.9 4.6 2.3

3.7 5.8 4.3 4.6 3.7 4.9 4.4 2.0

Infant Powder Hydrolyzed Milk

Adult Powder, Low-Fat

4.7 3.8 6.0 4.3 3.9 1.5

Child Powder

Infant Elemental Powder Adult RTF, High-Protein

4.9 4.5 10.6 b

5.2 20.1 b

13.7 b

Adult RTF, High-Fat

SRM 1849a

1.4 1.3

1.5

1.3

2.1

1.6

a Matrixes having the same name are two different batches of the same product made by the same manufacturer a few years apart. b These RSD percentages are larger than the allowed SMPR limits for RSDR. c “ Large ” represents >50% RSD; concentration was

R reported in Table 4. They can be calculated from Table S1 by pooling the

raw data for both replicates.

Ryan Connelly, Richard Zywicki, and Kayla Donhowe, Covance Laboratories (Madison, WI) Yoshihiro Ikeuchi, Shin-ichi Totsuka, Mariko Nagatoshi, and Naoto Hieda, Megmilk Snow Brand Co. (Kawagoe, Japan) Shigeki Kimura, Nobuyuki Kanno, Yoko Nakamoto, Kaori Akabane, and Chihiro Saitou, Meiji Co., Ltd (Odawara, Japan) Yi Ding, Abbott Nutrition (Singapore) Ashutosh Mittal, Abbott Nutrition (Bangalore, India) Frea Woltjes, Qlip (Zutphen, The Netherlands) Atsushi Miura and Tetsuo Kubota, Morinaga Milk Industry Co. (Zama, Japan) Bharathi Sadipiralla, Marc Connelly, and Brendon Gill, Fonterra Co-operative Group Ltd (Waitoa, New Zealand) Hendrik Veltman, Eurofins (Heerenveen, The Netherlands) We also acknowledge the efforts of Hans Cruijsen (Friesland Campina), and Eric Poitevin (Nestl´e), the Co-Study Directors of Method 2011.14 MLT, for working closely with us to match the protocols and other aspects of our two independent studies to make the comparison of results as meaningful as possible. We also thank Harrie van den Bijgaart (ISO TC34/SC5 Milk and Milk Products group), Aurelie Dubois-Lozier (IDF) and Marcel de Vreeze (Netherlands Standardization Institute) for first alerting us to the need for an ISO/IDF standard in this area and then being very helpful in guiding us through the process. Lastly, we wish to thank the ISO/IDF statistics group of Silvia Orlandini, Martin Alewijn, and Rob Crawford for reviewing the large amounts of data.

determine the major elements, Na, Ca, P, Mg, and K. Because this 12-element method is identical to Method 2011.19 , exhibits equivalent reproducibility for Cr, Mo and Se, and also covers more matrixes within its scope, we recommend that it replace Method 2011.19 to avoid any confusion in having two very similar methods available. The fortunate timing of the MLT for AOAC Official Method 2011.14 enabled us to work in close collaboration with the method ’ s team from the start to conduct an accuracy check of both methods by comparing mean results collected under very similar protocols. A very powerful set of data were amassed on both sides to show convincingly that the emission lines/masses, ISTDs, and other instrumental parameters chosen yield accurate results in these kinds of materials. There should be little doubt that ICP-MS with modern CRCs can test low-mass elements with the same accuracy, precision, and sensitivity as high-mass elements. As expected, the sensitivity of the ICP-MS affords better robustness and reproducibility in testing the lower concentrations of Cu, Fe, Zn, and Mn, but both methods ’ results should be viewed as interchangeable at higher concentration levels.

Acknowledgments

We wish to thank the following collaborators and their laboratories for completing this study: Isabelle Malaviole and Gauthier Demangeon, Laboratory Aquanal (Pessac, France)