Exploring Near and Midinfrared Spectroscopy to Predict Trace Iron and Zinc Contents in Powdered Milk

Abstract

Near infrared (NIR) and mid-infrared (MIR) spectroscopy were investigated to predict iron and zinc contents in powdered milk. A hybrid variable selection method, namely, uninformative variable elimination (UVE) combined with successive projections algorithm (SPA), was applied to select the most effective wavenumber variables from full 2756 NIR and 3727 MIR variables, respectively. Finally, 18 NIR and 18 MIR variables were selected for iron content prediction, and 17 NIR and 12 MIR variables for zinc content prediction. The obtained effective wavenumber variables were input into partial least-squares (PLS) and least-squares-support vector machines (LS-SVM), respectively. The selected MIR variables obtained much better results than NIR to predict both iron and zinc contents in both the PLS and LS-SVM models. The iron content prediction results based on LS-SVM with 18 MIR spectra were as follows: coefficient of determination (r(2)) was 0.920, residual predictive deviation (RPD) was 3.321, and root-mean-square error of prediction (RMSEP) was 1.444. The zinc content prediction results based on LS-SVM with 12 selected MIR spectra were as follows:r(2) was 0.946, RPD was 4.361, and RMSEP was 0.321. The good performance shows that UVE-SPA is a powerful variable selection tool. The overall results indicate that MIR spectroscopy incorporated to UVE-SPA-LS-SVM could be applied as an alternative fast and accurate method to determine trace mineral content in powdered milk, such as iron and zinc.

Near-Infrared Spectroscopy Combined with Multivariate Tools for Analysis of Trace Metals in Environmental Matrices

“…For instance, the absorption features associated with electronic transitions of Fe 3+ and Fe 2+ ions in Fe-bearing minerals can be found in the near-infrared region at 780-1200 nm [33][34][35]. In addition, the absorption features of metals associated with molecular vibrations of organic and inorganic functional groups in organic matter, silicates, carbonates, and water can be found in near infrared at 780-2500 nm region [24,33,34,36].…”

“…Wu et al [24,36] reported the feasibility of using NIRS for monitoring and predicting trace metals in suspended solids, sediment, and soil. This was achieved by quantitative evaluations of the spectral activity of sediment and soil properties [23][24][25][26][27][28][29][30][31][32][33][34][35][36][37].…”

“…This was achieved by quantitative evaluations of the spectral activity of sediment and soil properties [23][24][25][26][27][28][29][30][31][32][33][34][35][36][37]. However, due to challenges such as the collinearity, band overlaps, and interactions for some soil properties, the spectra of soil, sediment, or suspended solids are often broad and nonspecific [25].…”

“…However, more interest is on NIRs because it is generally cheaper [23]. In NIR, many trace elements are spectrally featureless and only exhibit characteristic spectral features at high concentrations (>4000 mg kg −1 ) [23,24]. Therefore, due to high detection, low levels of heavy metals can be indirectly determined from spectra due to their association with Fe oxides, clays, and organic matter [24].…”

“…In NIR, many trace elements are spectrally featureless and only exhibit characteristic spectral features at high concentrations (>4000 mg kg −1 ) [23,24]. Therefore, due to high detection, low levels of heavy metals can be indirectly determined from spectra due to their association with Fe oxides, clays, and organic matter [24]. For this reason, cost-effective and nondestructive analytical techniques based on near-infrared (NIR) spectroscopy coupled with chemometrics have been developed to overcome the problems encounter when using traditional methods [17,25].…”