HAL will be down for maintenance from Friday, June 10 at 4pm through Monday, June 13 at 9am. More information
Skip to Main content Skip to Navigation

Développement de stratégies originales d'extraction des protéines du tourteau de colza pour la production d'isolats par une démarche de modélisation / optimisation multicritères

Abstract : Proteins are major nutrients for nutrition: they are consumed by the ingestion of products naturally rich in proteins, or by food formulation, where proteins are ingredients. Since the 1980s, research projects are carried out to increase the value of rapeseed meal and to use its proteins in human nutrition, as an ingredient, instead of being used for animal nutrition only. In the context of the ever-increasing demography, the use of rapeseed protein has been investigated as a mean to fulfill the needs in proteins worldwide. Rapeseed meal is composed of two main storage proteins, albumin and globulin, with an amino acid profile balanced regarding sulfuric and lysine residues. Moreover, to be used as ingredient, it is crucial that these proteins are soluble and functional (to form foams, emulsions, gels). The physico-chemical properties of albumin and globulin are different. That is why, it is of interest to make two distinct ingredients. The main objective of the thesis was to develop strategies to extracts these proteins to use it as ingredients (isolate grade). To do so, an approach based on modelling by design of experiments, and on multicriteria optimization through genetico-evolutionary algorithms, was applied to better understand the phenomenon at play during extraction. The first part of the thesis consisted in the development of an analytical method that would allow the separate quantification of both protein contents. This method, developed by High Performance Liquid Chromatography, Size-Exclusion, is original and is calibrationless. Statistics were used to prove the possible substitution of the reference techniques (Kjeldahl, SDS-PAGE densitometry) by this method. Classical processes of extraction are carried out under alkaline conditions to reach high yields. Then, proteins are separated and purified to give two isolates with distinct functional properties. However, these steps imply non-negligible losses and costs. Hence, the second part of the thesis presents a strategy based on selective extraction of proteins with the pH. Modelling and optimization tools were used to identify operating conditions (pH, salt added) resulting in a two-step process. Two scenarios were considered: the production of a unique isolate, a mix of both proteins; or the production of two isolates, albumin and globulin-rich. An original feature of the work was the consideration of the quality of the residual meal. At the end of this process, the meal is very poor in proteins and thus, has low value for animal nutrition. That is why, the third part of the thesis consisted in developing a new strategy to optimize the selective extraction of albumin at acidic pH, to yield two products: a protein isolate, for human nutrition, and a meal rich in globulins for animal nutrition. Modelling and optimization tools were used to assess the impact of three factors: pH, salt content and temperature of the extraction medium. High yields were obtained at 55°C, and the optimal conditions resulted in rapeseed albumin with comparable properties (functional and structural) than at 20°C. Moreover, the residual meal was still valuable with a protein content similar to the initial meal and with less phytic acid, an antinutritional molecule, co-extracted with albumin.
Complete list of metadata

Contributor : Thèses Ul Connect in order to contact the contributor
Submitted on : Tuesday, March 24, 2020 - 3:19:53 PM
Last modification on : Wednesday, November 3, 2021 - 7:09:31 AM


  • HAL Id : tel-02517398, version 1



Claire Defaix. Développement de stratégies originales d'extraction des protéines du tourteau de colza pour la production d'isolats par une démarche de modélisation / optimisation multicritères. Génie des procédés. Université de Lorraine, 2019. Français. ⟨NNT : 2019LORR0176⟩. ⟨tel-02517398⟩



Record views