Deriving a Process Viscosity for Complex Particulate Nanofibrillar Cellulose Gel-containing Suspensions

Phase-separable particulate-containing gel structures constitute complex fluids. In many cases they may incorporate component concentration inhomogeneities within the ensemble matrix. When formulated into high consistency suspensions, these can lead to unpredictable time-dependent variations in rheological response, particularly under shear in simple parallel plate and cylindrical rotational geometries. Smoothing function algorithms are primarily designed to cope with random noise. In the case studied here, namely nanocellulose-based high consistency aqueous suspensions, the system is not randomised but based on a series of parallel and serial spatial and time related mechanisms. These include: phase separation, wall slip, stress relaxation, breakdown of elastic structure and inhomogeneous time-dependent and induced structure rebuild. When vacuum dewatering is applied to such a suspension while under shear, all these effects are accompanied by the development of an uneven solid content gradient within the sample, which further adds to transitional phenomena in the recorded rheological data due to spatial and temporal differences in yield stress distribution. Although these phenomena are strictly speaking not noise, it is nevertheless necessary to apply relevant data smoothing in order to extract apparent/process viscosity parameters in respect to averaging across the structural ensemble. The control parameters in the measurement of the rheological properties, to which smoothing is applied, are focused on parallel plate gap, surface geometry, shear rate, oscillation frequency and strain variation, and relaxation time between successive applications of strain. The smoothing algorithm follows the Tikhonov regularisation procedure.