A two-step factorial design for optimization of protein extraction from industrial rapeseed meal after ethanol-assisted reduction of antinutrients

• Autor(a) principal: Kalaydzhiev, H.
• Data de Publicação: 2019
• Outros Autores: Brandão, T. R. S., Ivanova, P., Silva, C. L. M., Chalova, V. I.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.14/28380
• topic: Ethanol-treated rapeseed meal; Protein yield optimization; Response surface methodology

Rapeseed meal is a by-product of oil production with relatively high antinutrient content which limits its application in both feed and food industries. A two-step procedure, including two-level full factorial and central composite designs, was used to assess the influence of five factors namely pH (6 and 12), temperature (20 and 40 degrees C), NaCl concentration (0 and 7.5%), rapeseed meal concentration (2.5 and 5%), and extraction longevity (15 and 60 min) on protein extraction from industrial rapeseed meal subjected to ethanol-assisted reduction of antinutrients. Data demonstrated that pH and its interaction with NaCl influenced protein extractability the most. While still significant, individual effects of NaCl, extraction longevity, temperature and its interaction with pH had a less significant effect on protein yield. Overall. higher protein yields were obtained in response to combining high pH values with low NaCl concentrations, and high pH values with high temperatures. The highest protein yield (59.56 +/- 1.29%) was achieved after extraction of 5% ethanol-treated rapeseed meal at pH 12 with no NaCl addition, at 40 degrees C and extraction longevity of 60 min. A second-order polynomial model for protein yield prediction was generated. The adequacy of the model was verified by coefficient of determination and residual analyses. A high correlation between experimentally obtained and predicted protein yields (R-2 = 0.95) was established. Since extraction conditions are strong determinants of protein characteristics, the predictive model generated in the present work is useful for the selection of factor combinations, which allows for the achievement of optimal protein yield of a product with desired techno-functional properties.