Naslov
COCOA POWDER DRINK MIX: THE APPLICATION OF A NON-INVASIVE NIR SPECTROSCOPY TECHNIQUE FOR THE FAST IN-LINE DETERMINATION OF MIXTURE COMPOSITION

Autori
Benković, Maja ; Valinger, Davor ; Jurina, Tamara ; Jurinjak Tušek, Ana ; Gajdoš Kljusurić, Jasenka

Vrsta, podvrsta i kategorija rada
Poglavlja u knjigama, znanstveni

Knjiga
Diversified benefits of cocoa and chocolate

Urednik/ci
Zayas Espinal, Bonifacia

Izdavač
Nova Science Publishers

Grad
New York (NY)

Godina
2018

Raspon stranica
93-114

ISBN
978-1-53613-258-8

Ključne riječi
cocoa ; sugar ; NIR spectroscopy ; principal component analysis ; partial least squares regression

Sažetak
Tropical tree Theobroma cacao L. is a source of one of the most highly appreciated raw materials in the food industry: cocoa beans. Raw cocoa beans are used for the production of chocolate, cocoa mass, cocoa butter and cocoa powder. Cocoa powder is a product which is usually lower priced in comparison to cocoa butter and chocolate, but it is further used in many more foodstuffs as a raw material. One of the widely used product which includes cocoa powder as a raw material is cocoa powder drink mix. It is produced by mixing cocoa powder with sugar, usually in a 30:70% ratio. Besides cocoa and sugar, milk powder and instantizers (soy lecithin) are added to the mixtures prior to agglomeration and drying. In order to achieve and maintain a constant mixture composition, it is of great importance to use adequate mixing equipment, as well as to monitor the mixing process at all stages. Since industry mixing devices contain large tonnages of powdered material, constant sampling and mixture composition analysis can be somewhat difficult and time consuming. Because of that, the need for a fast, in-line, build-in mixture composition monitoring technique arises. Near infrared spectroscopy (NIRs) is a valuable analytical tool in agriculture and the food industry for rapid and accurate analysis of certain product constituents (Kays et al., 1997). In addition to being rapid, NIRs requires very little sample preparation, requires no chemicals, and produces no chemical waste (Kays et al., 2002). In this work, the objectives were to explore the potential of NIR usage for monitoring the cocoa powder and sugar mixtures composition. Cocoa powder and powdered sugar mixtures were produced in different ratios (0 – 100%) with the aim of creating a calibration curve based on the NIR spectra recorded for each mixture. The recorded NIR spectra were then subjected to the Principal Component Analysis (PCA) to obtain the factor values of principal components. The factor values were then plotted in relation to the mixture composition to obtain a calibration curve. The Partial Least Squares regression (PLS) was performed to obtain and validate models based on which the concentration of sugar at any given moment in the mixture can be quantitatively and/or qualitatively predicted. Based on the obtained results, NIRs analysis in combination with the PCA and the PLS has proved to be an adequate, fast and non-invasive method for the cocoa powder drink mix composition prediction.

Izvorni jezik
Engleski

Znanstvena područja
Biotehnologija, Prehrambena tehnologija

POVEZANOST RADA