Partial Least Square (PLS) multivariate calibration associated to Near Infrared Reflection Spectroscopy (NIRRS) or Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) were used to establish methods for simultaneous determination of protein and ash content on commercial wheat flour samples of Triticum aestivum L. Duplicate spectra of 100 samples with protein content between 8.85-13.23% (Kjeldahl method) and ash content between 0.330-1.287% (gravimetric method) were employed to build calibration methods. The spectra were used in log (1/R) format as well as in the first and second derivatives and they were also pre-processed by mean centering (MC) or variance scaling (VS) or both. The samples were divided in two sets, one with 55 samples and the other with 45 used as calibration and validation sets, respectively. Calibration methods were obtained by using minimum values of root mean square error of calibration (RMSEC) and root mean square error of validation (RMSEV). Methods of calibration with RMSEC and RMSEV values lower than 0.33 in the determination of protein content and lower than 0.07 in the determination of ash content. The proposed methods are environmentally non aggressive as they avoid the use of reagents and do not generate hazardous waste. Furthermore, they allow fast and non destructive simultaneous and direct determination of ash and protein content in wheat flour samples.
infrared spectroscopy; PLS regression; wheat flour; protein content; ash content; NIR spectroscopy; chemometrics
