Abstract:

Raman spectroscopy was applied to replace the current methods for the determination of quality parameters of
flour (including moisture, ash, and gluten content), since the conventional methods are time-consuming and laborious. Raman
spectra of 100 samples from different flour types obtained from Guchuan flour mill were collected. Quantitative models
were established using different pretreatment methods combined with partial least squares. The correlation coefficients of
the established models for flour moisture, ash, and wet gluten were 0.945 66, 0.993 39, and 0.981 65, respectively, and the
root mean standard error of calibration (RMSEC) were 0.145, 0.012 6, and 0.456, respectively. The results showed that
Raman spectroscopy could be use for rapid, non-destructive, and efficient determination of flour quality. Qualitative models
of different flours produced in 2013 and 2014 were also established based on distance matching analysis. The results showed
that expired flour was indentified with 100% accuracy.

Key words: flour, Raman spectroscopy, partial least squares (PLS), quantitative analysis, distance matching analysis