Classification and quantification analysis of peach kernel from different origins with near-infrared diffuse...

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Background:

• Peach kernels are rich in fatty acids crucial for physiological functions.
• Accurate analysis of peach kernel composition is important for various applications.

Purpose of the Study:

• To develop a rapid diffuse reflectance near-infrared spectroscopy (DR-NIR) method for peach kernel analysis.
• To utilize chemometric techniques for qualitative and quantitative determination of peach kernel varieties and fatty acid content.

Main Methods:

• Diffuse reflectance near-infrared spectroscopy (DR-NIR) in the 1100-2300 nm range.
• Spectral pre-processing using Savitzky-Golay and first derivative methods.
• Chemometric analysis including Principal Component Analysis (PCA) for classification and Partial Least Squares Regression (PLSR) for quantification.

Main Results:

• DR-NIR spectra successfully classified two peach kernel varieties using PCA.
• PLSR models demonstrated high predictive ability for linoleic and oleinic acids.
• Correlation coefficients exceeded 0.99, with low Root Mean Square Error of Deviation (RMSED) values of 1.266% and 1.412% for linoleic and oleinic acids, respectively.

Conclusions:

• The DR-NIR method, combined with PCA and PLSR, offers an efficient approach for identifying and quantifying peach kernels.
• This technique can effectively differentiate between peach kernel varieties.
• The study highlights the potential of DR-NIR spectroscopy in quality control and authentication of peach kernel products.