2.3 C OAXIAL - CYLINDERS RHEOMETER – T HE BML V ISCOMETER
2.3.2 Limitations in measurement of rheological parameters of fresh concrete
to 4.75 mm according to ASTM C125) and cement particles. Due to the nature of fresh concrete, there are some limitations in the measurement of rheological parameters of fresh concrete that are not present in classical fluid rheology. In order to have meaningful measurements and results, it is essential to minimise spatial segregation in the concrete. This can be done by ensuring that the concrete mixtures have continuous grading size in the aggregates and sufficient fine particle proportions (Ferraris and Brower, 2001). The particle size distribution of the aggregates in the current study is presented in Chapter 3, Section 3.2 (page 58) and Table 3.4.
When a fluid is under shear, it is normal to assume that the velocity gradient in the test sample is continuous up to the interface. This means that the velocity between two parallel plane surfaces in the fluid varies with distance between the surfaces. The assumption is valid for Newtonian fluids such as some oil and water. However, with granular suspensions such as fresh concrete, an interfacial layer will form in the vicinity of the wall due to exclusion of coarse particles. This causes the material to have a higher fluid content than the bulk of the suspension. Thus, this facilitates a sliding motion during shearing of the concrete in a rheometer, known as slippage.
When this happens at low shear rate, the concrete moves as a whole mass known as a plug flow. In the case of a coaxial-cylinders rheometer, the slippage is likely to occur along the interior wall of the outer cylinder and the surface of the inner cylinder, both of which are in contact with the test sample in the shear zone. This is shown in Fig.2.11 (left) indicated by the region between the inner and outer cylinder where possible plug flow may occur. Various authors have proposed the following model to describe this phenomenon (Morinaga, 1973; Browne and Bamforth, 1977), which is well known in the pumping technology of concrete:
τ = τ0,i + ηi. vg (2.18)
where τ is the shear stress, τ0,i is the interfacial yield stress, ηi is the interfacial viscosity, and vg is the sliding velocity. The interfacial yield stress is always smaller than the yield stress of the concrete. Thus, when a shear force is applied to the concrete, the bulk of the material will start sliding along the wall with no shear within the concrete. The concrete will start to shear only when the shear stress exceeds the yield stress after the sliding velocity increases beyond a certain level. In general, the measured rheological parameters will be lower than the actual values when slippage occurs during the measurement.
Besides this, particle migration within the concrete mass during shearing is another factor that will affect the accuracy of the measurement of the rheological parameters. The degree of the particle migration is possibly affected by three factors.
These include the collision rate between the particles, dilatancy of the concrete, and the confinement effect (Ferraris and Brower, 2001). The discussion that follows is applicable to the BML rheometer. Firstly, the rate of collision between the particles is proportional to applied shear rate. From Equations (2.15) and (2.16), the shear rate is highest at the shear zone next to the inner cylinder. During shear of concrete, the particles will tend to be pushed away from the region of high collision rate to region of low collision rate. This means that the coarser particles will move outward towards the outer cylinder, as well as inward into the serrated region of the inner cylinder.
There is also a possibility that the coarser particles may migrate downward to the bottom of the outer cylinder. When this happens, the concrete in the shear zone next to the inner cylinder (region of highest shear rate) will have a higher mortar content.
Another physical phenomenon that could be present at the same time during shearing is the effect of dilatancy in the concrete. During particle migration, there is an exchange of moving particles within the concrete mass. Thus, for the coarse
aggregates to migrate towards the regions of lower shear rate, the mortar matrix from these regions will have to move into the zone of highest shear rate to replace the coarse aggregates. This phenomenon is confirmed by Mork (1994). The degree of dilatancy in the concrete is dependant on the volume of dead zone, which is the region outside the shear zone. Generally, dilatancy in the concrete will increase during the measurement when the volume of dead zone is increased.
Finally, the confinement effect becomes apparent when the ratio of the gap between the outer and inner cylinders (i.e. the shear zone) relative to the maximum size of the coarse aggregate (Dgap / Dmax) is low. When the D / Dgap max is below 4 to 5, perturbation may take place in the measurements of rheological parameters (Ferraris and Brower, 2001). When the gap in the shear zone is narrow, bridging of coarse aggregates can occur during shearing of the concrete. This happens when the coarse aggregates are close enough to form an interlocking link across the shear zone between the outer and inner cylinders (Fig.2.14). This causes an abnormally high torque to be registered during shearing of the concrete, giving rise to higher measured rheological parameters than the actual ones. This is especially so when the measurement is performed under confined condition. A confined condition occurs when the ratio between volume of the concrete in the shear zone and total volume of the tested concrete (V / Vs t) is high. This also means that the amount of dead zone (i.e.
the region outside the shear zone) available for the migration of the coarser aggregates is low. Thus, in a confined condition, the aggregates in the concrete have lower tendency to migrate out of the shear zone. On the other hand, the measured rheological parameters obtained under an unconfined condition may be lower than the actual values due to significant migration of the coarser aggregates into the dead zone.
In view of the limitations in the measurement of rheological parameters of fresh concrete, certain measures were taken to minimise the possible inaccuracies of the measured rheological parameters in the current study. This will be presented in Section 3.4.1 (page 63).
Bridging of aggregates
Inner cylinder
Outer cylinder
Fig.2.14 – Bridging of coarse aggregates during shearing of fresh concrete in a coaxial-cylinders rheometer with rotating outer cylinder