Description of batter mixing using near infrared spectroscopy

The aim of the present study was to describe the physicochemical events occurring during batter mixing at different water contents (51.8, 54.4, and 56.7 g of water/100 g of dough) using near infrared (NIR) spectroscopy. An FT-NIR spectrometer over the 1000–2500 nm range with a fibre optic probe was used to record NIR spectra in-line. The analysis of both one-dimensional statistical method (principal components analysis) and two-dimensional statistical methods (generalised two-dimensional correlation spectroscopy) was conducted to evaluate the possibilities of NIR spectroscopy to monitor physical and physicochemical modifications observed during mixing of batter. The NIR results were in agreement with the physical and physicochemical analysis traditionally used to study bread dough mixing (consistency and glutenin depolymerisation). PCA on raw NIR spectra demonstrated that PC1 describes the same traces as the dough consistency curves. PCA on raw NIR spectra can be used to monitor the batter mixing and to identify the NIR mixing time close to the tpeak.PCA on spectra after second derivative demonstrated that PC1 and PC2 traces described different traces compared to the dough consistency curves. The loading spectra associated to PC1 and PC2 suggested that almost the same physicochemical and chemical mechanisms occur during the dough mixing at 51.8 or 54.4% water contents, but with kinetic and intensity differences. The 2D COS method allowed a sequence of chemical events occurring during mixing for the batters at 51.8 and 54.4% water contents to be tentatively proposed. The 2D COS did not give clear physicochemical differences between the three batters during mixing. The NIR results for the highly hydrated batter (56.7%) were difficult to analyse due to its high water content.