Non-colloidal Suspensions Rheology – An Experimental Study

To get a complete scenario of the rheological behaviour of non-colloidal suspension it is important to explore the transient shearing, elongational, and oscillatory shearing as well as simple steady shearing. The thesis includes new experiments and comparisons with the few theories available. The simple steady shearing and elongation flow behaviour of non-colloidal suspensions were studied. Simple shear-thinning effects were discussed and the elongational phenomena in non-colloidal suspensions were investigated, where transient shearing was also considered. The thinning effects in non-colloidal suspension were studied from different experiments and simulations. Shear-thinning was clear on 40% volume fraction suspensions. The shear-thinning depends on the variation of the interparticle friction coefficient with the shear rate or shear stress. The transient result did not replicate Gadala‐Maria and Acrivos (1980) results. They performed their test 40 years ago using similar suspensions and we believe the variation may be due to the better performance/sensitivity of the newer rheometers. We developed a new setup to perform the elongational test and the motion of the fluid drops was filmed with a high-speed FASTCAM PCI R2 (Photron) camera. The reduction in diameter of the drop with time indicates the elongation rate; pure elongation was seen where the diameter was minimum. To investigate the changes in elongational viscosity of 12500 cs Silicone oil suspension, a steel ball was introduced with the suspensions this improves the filament shape – it is more uniform. We assumed the rate of elongation (ε ̇) was constant over the cross-section (but varied at different times); the rate of elongation was then found using the variation in diameter with time (Dai and Tanner, 2017). Friction plays a significant role in non-colloidal suspension and it becomes more complex at higher concentration of suspension. A “bootstrap” feedback model was proposed (Tanner et al., 2018) which shows how friction amplifies the stress and increases the complexity of the suspension behaviour. The bootstrap mechanism of friction enhancement in non-colloidal suspensions appears to be a useful idea. In view of the confusing picture reported by others, it was decided to make further measurements of small and medium strain oscillatory flows keeping in mind possible friction between particles. Because of the observed tendency to hysteresis, the oscillatory flow was combined with the parallel steady shear flow to control or eliminate hysteresis. The response of concentrated suspensions under oscillatory shear is surprisingly complex, and it becomes more complex when oscillatory shearing and steady shearing are combined. The non-zero storage modulus experienced due to frequency sweep tests in a glycerine matrix was an apparent storage modulus which was due to inertia effects (which were very small). We observed a slight decline in G^' and G^'' with an increase of strain amplitude and we believe settling is not responsible. Surprising hysteresis effects were experienced by both moduli with increased and decreased strain amplitude. The hysteresis was more dominant in the storage modulus than in the loss modulus and at higher concentrations the effect was severe. In simple steady shearing, friction increases the suspension viscosity, and we believe both G^' and G^'' also increase due to friction. The effect of variable oscillatory shear stress and steady shear stress was analysed and a model for superposed storage modulus, loss modulus and viscosity responses was proposed. Frictional effects were considered in the proposed model and satisfactory fitting with the experimented data was observed. The ‘bootstrap’ feedback mechanism was adapted to explain frictional effects. We proposed a universal model for non-colloidal suspensions. This is a modified Reiner-Rivlin inelastic model (eq. 5.9) which is a reasonable description of available tests of suspensions with Newtonian matrices in several flows for larger strains. Further checks are needed, and the quest must continue for more basic understanding. Nevertheless, the modified Reiner-Rivlin model is simple to apply and should be useful for numerical work on the concentrated suspensions