Huiyu Gao1, Guodong Wang1, Zhu Wang1. 1. Key Laboratory of Trace Element Nutrition of National Health Commission, National Institute for Nutrition and Health, Chinese Center for Disease Control and Prevention, Beijing 100050, China.
Abstract
OBJECTIVE: Near-infrared(NIR) spectroscopy combined with partial least squares(PLS) were applied to establish a rapid method for green direct determination of mineral elements(calcium, phosphorus and potassium) in wheat flour samples. METHODS: NIR spectra and analytical measurements of calcium, phosphorus and potassium were collected from 117 wheat flour samples with different processing levels(whole grain wheat, special grade No. 1 wheat and wheat core flour). Principal components analysis(PCA) was developed to assign 81 wheat flour samples to build models and 36 samples as the validation set to evaluate the performance of the developed models. The influence of wavelength range and spectral preprocessing method on the predictive ability of the model were discussed, and the best models were selected. RESULTS: For calcium, the best NIR model showed a good prediction performance(r~2=0. 7907, RMSEP=5. 35, RPD=2. 19); the best NIR model for phosphorus gave an excellent prediction performance(r~2=0. 9777, RMSEP=15. 3, RPD=6. 71); the best model for potassium also gave an excellent prediction performance(r~2=0. 9777, RMSEP=18. 9, RPD=6. 84). CONCLUSION: NIR spectroscopy can realize the rapid prediction of mineral elements(calcium, phosphorus and potassium) in wheat flour. By selecting the wavelength range and spectral preprocessing method, the prediction ability of the NIR model can be significantly improved.
OBJECTIVE: Near-infrared(NIR) spectroscopy combined with partial least squares(PLS) were applied to establish a rapid method for green direct determination of mineral elements(calcium, phosphorus and potassium) in wheat flour samples. METHODS: NIR spectra and analytical measurements of calcium, phosphorus and potassium were collected from 117 wheat flour samples with different processing levels(whole grain wheat, special grade No. 1 wheat and wheat core flour). Principal components analysis(PCA) was developed to assign 81 wheat flour samples to build models and 36 samples as the validation set to evaluate the performance of the developed models. The influence of wavelength range and spectral preprocessing method on the predictive ability of the model were discussed, and the best models were selected. RESULTS: For calcium, the best NIR model showed a good prediction performance(r~2=0. 7907, RMSEP=5. 35, RPD=2. 19); the best NIR model for phosphorus gave an excellent prediction performance(r~2=0. 9777, RMSEP=15. 3, RPD=6. 71); the best model for potassium also gave an excellent prediction performance(r~2=0. 9777, RMSEP=18. 9, RPD=6. 84). CONCLUSION: NIR spectroscopy can realize the rapid prediction of mineral elements(calcium, phosphorus and potassium) in wheat flour. By selecting the wavelength range and spectral preprocessing method, the prediction ability of the NIR model can be significantly improved.
Entities:
Keywords:
mineral elements; near-infrared spectroscopy; wheat