A micro- and macro-scale approach to probe the dynamics of sol–gel transition in cereal β-glucan solutions varying in molecular characteristics

Abstract

Cereal β-glucans are known to display functional properties with health benefits, such as reduction of plasma cholesterol and of postprandial serum glucose levels in humans and animals; such effects have been lately attributed to viscosity and gelation potential of these water soluble fibres. The local dynamics of β-glucan solutions differing in molecular size and the molar ratio of trimers to tetramers (DP3/DP4) chain segments, which undergo gelation upon ageing, was investigated. Confocal microscopy, particle tracking microrheology and conventional bulk shear rheological measurements have been employed to study the microstructure and mechanical properties of the polysaccharide networks on various length scales. The structural features of β-glucans, such as molecular weight and ratio of DP3/DP4, were found to be important determinants of their gelling properties and microstructure. For the β-glucan gels cured at 25 °C both the microrheology and the bulk rheology revealed that with decreasing molecular weight and increasing DP3/DP4 molar ratio the gelation time decreased, while the gelation rate increased along with the storage modulus. In all samples studied, particulate clusters of β-glucan assemblies were generated with size ∼1–15 μm and the clusters were eventually interconnected to expand over the available space to form an elastic network. The pore size and the structural entities were found to increase in size at lower values of the DP3/DP4 molar ratio and with preparations of high molecular weight. The size and the compactness of the structural entities seem to play an important role in the network reinforcement. The embedded tracer particles were found to experience relatively homogeneous microenvironments at DP3/DP4 ∼2.1, reflecting a rather slow rate of chain aggregates structuring. The behaviour of the polysaccharide network dynamics at a microscopic level does not always seem to match the overall bulk macroscopic response.