Die dimensions impact on fibrous plant protein formation during high moisture extrusion

Autor: Valérie Guyony, Francine Fayolle, Vanessa Jury
Jazyk: angličtina
Rok vydání: 2022
Předmět:
Zdroj: Applied Food Research, Vol 2, Iss 2, Pp 100228- (2022)
Druh dokumentu: article
ISSN: 2772-5022
DOI: 10.1016/j.afres.2022.100228
Popis: Plant-based meat analogues that mimic the structure and texture of meat are becoming very popular. They can be produced using high moisture extrusion (HSE). . High protein materials first undergo intense thermomechanical treatment in the extruder barrel and then are textured and simultaneously cooled in the die. The mechanisms underlying this structuring process are not sufficiently understood and depend on numerous and interconnected factors, including raw materials composition and process parameters. This study focuses on the effects of die height and die length on extruder responses (torque, pressure), textural properties and fibration of soy concentrate extrudate obtained at high moisture (60% w/w). In order to evaluation the quality of fibration, and to complete texture analysis, a visual method base on angle of fibration was developed. By comparing extrudate obtained with die of 2 mm and 5 mm height, this study shows that a minimum height is necessary for temperature gradient and velocity profile creation in the die. In order to study die length higher than the one observed in the literature, a modular die with different lengths was specially developed for this study, going from 150 to 600 mm for a extrusion barrel length of 640 mm. Results show a linear correlation between fibration and die length: the longer is the die, the more elongated is the internal flow pattern. This change in flow pattern leads to an improvement in the quality of the extrudate fibration.It confirms that die length is one of the main influencing factor for fiber formation, and that a minimum length is required to obtain extrudats which structure really mimics meat fibres.
Databáze: Directory of Open Access Journals