Validation studies for multicomponent quantitative NMR analysis: the example of apple fruit juice

2014 
Laboratories intending to work as official laboratories for food control have to be accredited according to ISO/IEC 17025. This necessitates the use of validated analytical methods. In this study, we present validation results of the recently commercialized “JuiceScreener” based on nuclear magnetic resonance (NMR) spectroscopy with apple juice as application example. The quantitative analysis included 29 compounds such as major sugars, amino acids, organic acids, as well as acetoin, arbutin, benzaldehyde, hydroxymethylfurfural, acetaldehyde, methanol, and ethanol. Limit of detection (LOD), limit of quantification (LOQ), coefficients of variation (CV) for replicated measurements, repeatability, linear range, and recoveries were determined. The LOD and LOQ values varied in the 0.48–16 mg/L and in the 1.9–122 mg/L ranges with the lowest values for shikimic acid and highest for the principal sugars. The 1H NMR assays were linear in broad concentration ranges (R > 0.99), encompassed typical concentration in apple juices, and are sufficient to control the requirements of the code of practice of the European fruit juice association. Recoveries between 92 and 109 % on average for five separate standard additions were obtained. The average CVs were found to be 3.0 % (intraday) and 3.6 % (interday) excluding sample preparation (by measuring five time one solution) and 5.5 % (intraday) and 6.2 % (interday) including sample preparation (by preparing and analyzing five separate samples). The NMR method was judged as suitable for the simultaneous quantification of compounds in apple juice for official food control purposes. Our results show that multiparameter NMR methods can be successfully validated with standard instrumentarium and that they are fit for the purpose of official food control.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    38
    References
    30
    Citations
    NaN
    KQI
    []