De Graef, Veerle × Foubert, Imogen Agache, Eveline Bernaert, Herwig Landuyt, Alex Vanrolleghem, Peter Dewettinck, Koen #
Wiley-v c h verlag gmbh
European journal of lipid science and technology vol:107 issue:5 pages:297-306
The objective of this study was to investigate whether it is possible to predict migration fat bloom based on measurements shortly after production. At different storage times shortly after production (0, 1, 4 h), the chocolate batches, varying in tempering method, tempering degree and amount of added butter oil, were evaluated by DSC, pNMR and texture analysis. Discriminant analysis and principal component analysis were combined to investigate the potential towards prediction. The batches were classified into groups depending on the time when white spots appeared (<8wk, 8-13wk, >13wk). A good separation (100% correct classifications, 100% using cross-validation) was obtained using the afore-mentioned analyses and storage times. It was also shown that it is possible to exclude DSC analyses or analyses at 0 h storage time without compromising the classification performances too drastically. The study further elucidated that the tempering method has no significant effect on visual fat bloom development. Furthermore, undertempered chocolates bloomed quicker than well-tempered ones, while fat bloom was delayed on overtempered chocolates. Addition of 6% butter oil promoted fat bloom development, while no significant difference was detected between chocolate with no added butter oil and chocolate with 3% butter oil added.