AOAC SPIFAN Nutrients ERP (December 7, 2022)

Fluor-06 (SLV) 10-2022 FOR ERP USE ONLY DO NOT DISTRIBUTE

5.8 Robustness To evaluate the robustness, eight samples were measured by two technicians: six test samples in duplicate plus the QC-LOQ and QC-LOW in six-fold. The results are shown in Table 12 and 13. Table 12. Robustness experiment in duplicate by two technicians spread over different days. SRM1869 RM 8260 RM8261 SRM1849a NIST10175 CM-62

(mg/kg)

(mg/kg)

(mg/kg)

(mg/kg)

(mg/kg)

(mg/kg)

Day 2 (System A) Technician 1 A verage 1.199

0.192

0.906

1.472

0.358

1.654

RSD (%)

0.0

2.2

0.5

2.4

0.0

0.4

Day 2 (System B) Technician 2 A verage 1.287

0.196

1.006

1.509

0.360

1.734

RSD (%)

2.0

0.4

1.2

0.2

0.6

0.4

Day 4 (System B) Technician 2 A verage 1.255

0.193

0.983

1.484

0.340

1.684

RSD (%)

1.5

1.5

0.2

0.7

0.4

1.9

Table 13. Recoveries for QC-LOQ and QC-LOW samples on two different systems by two different technicians. Day 1 Day 2 System A System B System A System B Recovery Recovery Recovery Recovery QC-LOQ-1 99.5 94.5 QC-LOW-1 97.9 104.9 QC-LOQ-2 100 94.3 QC-LOW-2 95.2 104.3 QC-LOQ-3 102.3 97.2 QC-LOW-3 100.1 104.4 QC-LOQ-4 102.4 94.5 QC-LOW-4 97.0 106.1 QC-LOQ-5 97.9 92.0 QC-LOW-5 98.4 104.7 QC-LOQ-6 96.6 97.6 QC-LOW-6 99.2 104.1 Average (%) 99.8 95.0 Average (%) 98.0 104.8 STDEV 0.023 0.021 STDEV 0.017 0.007 RSD (%) 2.0 2.2 RSD (%) 1.8 0.7 The maximum RSD value found for robustness is 2.4% (Table 11). Similar results were obtained for measurements on different days and by different technicians (Table 12). Both systems can measure correctly at low levels as predetermined by AOAC. The results of the QM62 samples were added to a control card using Datalyzer Spectrum 3.21 (Datalyzer International, Eindhoven, NL), displayed in Figure 7.