6. AOACSPIFANMethods-2018Awards

284

S CHIMPF ET AL . : J OURNAL OF AOAC I NTERNATIONAL V OL . 101, N O . 1, 2018 267

Figure 4. Chromatographic separation of lutein, b-carotene, and lycopene in a working standard.

Single-Laboratory Validation Experimental

prepared on day 1 were analyzed each day during the single- laboratory validation. Between days, the reconstitutions were stored at 2 – 8°C in glass containers protected from light. One container of each liquid product was opened on day 1 and used for all precision analyses. Liquid products were transferred to glass jars, capped, and stored at 2 – 8°C between analyses. Method accuracy was established by spiking an aliquot of three representative SPIFAN matrixes with lutein, b -carotene, and lycopene dissolved in ethanol (lutein) or THF ( b -carotene and lycopene) with either a relatively low, medium, or high level of trans -lutein, -lycopene, and - b -carotene and determining carotenoid recoveries. As noted above, the spiked matrixes included an adult high-fat RTF nutritional, an infant soy- based powder, and an infant partially hydrolyzed milk-based powder. Each spiked sample was stored at 2 – 8°C in capped glass jars. Spiked samples were prepared and analyzed in duplicate on 6 days. Method linearity was evaluated by injecting four standards with trans -lutein, - b -carotene, and -lycopene concentrations ranging from approximately 10 to 250, 25 to 500, and 5 to 100 µg/L, respectively, before and after every set of validation samples. Calibration curves were constructed from these standard injections, and the regression parameters from least-squares fittings were used to back calculate the concentration of each working standard in order to determine calibration errors at each level.

To verify the applicability of this method for the determination of carotenoids in infant and adult nutritionals, a single-laboratory validation with several SPIFAN infant, pediatric, and adult matrixes was completed. It should be noted that most of the SPIFAN matrixes were not fortified with carotenoids and did not contain significant levels of inherent carotenoids, so several matrixes were spiked with carotenoids and used to generate precision and accuracy data. To establish method precision, six SPIFAN matrixes; two non-SPIFAN matrixes; and three SPIFAN matrixes spiked with lutein, b -carotene, and lycopene were prepared and analyzed in duplicate on 6 days. The remaining SPIFAN matrixes were prepared and analyzed in duplicate on 3 days. All SPIFAN powder matrixes were reconstituted. SRM 1849a and SRM 1869 were reconstituted by mixing the entire contents of the sachet (10 g) in 90 mL water. All other powders were reconstituted by mixing 25 g powder in 200 mL laboratory water. The SPIFAN matrixes that were spiked with carotenoids included an adult high-fat ready-to- feed (RTF) sample, a reconstituted infant soy powder, and a reconstituted infant hydrolyzed milk powder. Each of these matrixes was spiked with either a relatively low, medium, or high level of trans -lutein, -lycopene, and - b -carotene. The same powder reconstitutions and spiked matrixes that were

Figure 5. Chromatographic separation of carotenoids in a liquid infant nutritional (38300).