(246f) Rheological Properties of Concentrated Suspensions

The paper presents results from rheological tests conducted using two commercial bentonite based products employed in geotechnical engineering applications (directional drilling, microtunneling, pipejacking, well drilling, etc.). The paper describes the different behavioral features of concentrated suspensions prepared from these two products and hypothesizes the formation of a different gel microstructure. The first material (Bent#1) is a bentonite only product, while the second product (Bent#2) contains as primary additive PHPA, as well as other proprietary additives. The tests were performed making use of a Rheologica Viscotech rheometer equipped with a 40 mm diameter, 4¢X cone and plate geometry. Different types of tests were conducted: the study of the flow properties of the suspensions relied on constant rate and stress ramp tests, while the at-rest properties were investigated using creep and recovery tests and three types of oscillatory tests, which measured the changes in viscoelastic properties (storage modulus G', loss modulus G", phase angle ?n?Ô) with varying strain, frequency, and time. Prior to all the tests all suspensions underwent a standard pre-shear phase. The behavior of the materials considered was studied as a function of the concentration of the suspension. It was observed that above some critical concentration each of the materials behaved as a gel. This was ascertained through the appearance of a yield stress, a low phase angle, and the appearance of recoverable strain in creep and recovery tests all at the same concentration of bentonite. Figure 1 shows results from stress ramp tests for both materials tested as a function of concentration. In addition to differences in the concentration in correspondence to which a yield stress is observed (60 g/l for Bent#1, 18g/l for Bent#2), the figure highlights different types of flow curves. The behavior observed in the suspension prepared from Bent#1 at 60 g/l appears to be a result of the existence (also documented by other researchers) of a local minimum in the shear stress immediately following yield, which is caused by structural breakdown. This minimum cannot be captured in a test, performed under stress control, such as that shown in the figure. In this test the occurrence of shear fracture was also detected. The unloading portion of the stress ramp for this test demonstrates that the fluid is characterized by thixotropic behavior. With the second product, a different behavior is observed: following yield the fluid is characterized by shear thinning behavior; and upon reversal, increasing rheopexy is observed with increasing concentration. Distinct behaviors are also observed in the time curves (Fig. 2) obtained from constant rate tests performed on suspensions of equal concentration. For both materials at high concentrations similar results are observed in the strain sweep tests (Fig.3): the complex modulus G* is approximately constant up to a shear strain of about 0.1, beyond which it decreases very sharply; G' is always much greater than G", indicating the material behavior is predominantly elastic. The behavior is also characterized by a nearly constant G" at lower strains. G" does begin to increase at intermediate strains, indicating an increasing viscous component to the material response. For both materials this behavior is observed to be frequency independent (Fig.4). While the majority of the tests were conducted employing deionized water, a select number of suspensions were prepared using a solution of 1000 ppm NaCl (to simulate fresh groundwater) and a 35,000 ppm NaCl "salt" water solution. Figure 5 shows the changes in the flow curve of suspensions prepared at a molar concentration of 60g/l with both materials with the presence of salt. For high salt concentrations the effects are similar for both materials, i.e. the suspensions cease to display any yield stress and show a close to Netwonian behavior. The results instead differ greatly with a small increase in the ionic strength of the solution (i.e. 1 ppt solution). With Bent#1 there is: an increase in yield strength and a more marked minimum in the flow curve (Figure5), as well as an increase in both G' and G". With the other bentonite product (#2), the yield stress, G' and G" all decrease.

These results suggest that the structure of the two gels is different. Specifically it is hypothesized that in the case of Bent#1 an attractive gel is formed, while in the case of Bent#2 the structure is that of a repulsive gel.