Two Particle Type Blending for the Rheological Modification of Yield Stress Sediments

The rheology of concentrated particulate suspensions is frequently described as a yield strength fluid of shear thinning behavior. The magnitude of the yield stress can be a function of many parameters including the initial particle concentration, particle-particle interaction strength (particularly for colloidal particles), and sample aging (i.e. time-dependent consolidation and strengthening via chemical reaction). In the nuclear industry yield stress sediments have formed in nuclear waste legacy storage tanks due to the corrosion of metals and the time-dependent consolidation of the corroded products. These yield stress sediments are encountered on UK and US nuclear sites and prove particularly challenging to re-suspend for their removal to interim (long-term) storage facilities. In these interim facilities time-dependent consolidation is again anticipated, although ideally mechanisms to alleviate the build-up of yield strength are being considered.

In the current study, alumina (Al2O3) colloids were the main component of the particulate suspension, an ideal test material for legacy wastes encountered on US nuclear sites. Alumina is an interesting colloid since the iso-electric point (i.e.p.) of the particle is pH = 8.6, and the particle charge can be strongly positive and negative in low and high pH solutions, respectively. At conditions away from the i.e.p. the yield stress of an alumina colloid suspension increases as a power-law function with particle concentration, attaining a yield stress at 30 vol% of ~175 Pa and ~ 90 Pa when the solution pH is 8, ζ = 13.7 mV and 10, ζ = -26.0 mV, respectively.

One approach to modify the yield stress and flow properties of concentrated alumina suspensions is to blend the suspension with a second colloid. We have considered silica spheres as a tool to “lubricate” yield stress suspensions, while maintaining a high solids content. The spherical nature of the silica colloid is desirable, along with the negative surface charge across a wide pH range (pHi.e.p. = 2). As such, when the pH is below pHi.e.p.(alumina) the silica and alumina particles will be strongly attractive (pH 8, ζalumina = 13.7 mV, ζsilica = -36.0 mV), while at pHs above pHi.e.p.(alumina) the particles will be repulsive (pH 10, ζalumina = -26.0 mV, ζsilica = -37.5 mV). Since the d50 of the alumina was 520 nm, two silica particle sizes were considered i) d50 = 100 nm, ii) d50 = 800 nm, such that the particle size effect on the lubrication potential can be elucidated.

While the HHF colloidal theory confirmed hetero-aggregation of the particle species at pH 8 and colloidal dispersion at pH 10, drop casting and plunge freezing scanning electron microscopy was used to validate the aggregation potential. The alumina particles were blended with 100 nm silica at pH 8 and 10 at a particle number ratio (alumina to silica) of 1:100. The SEMs confirmed hetero-aggregation of alumina and silica at pH 8, although the silica coverage on the alumina remained quite sparse. At pH 10 no visible interaction between the two particles was observed and confirmed electrical double layer stabilization between the two negatively charged particles. This behavior was also observed when the alumina particles were blended with the larger silica spheres (d50 = 800 nm).

The rheology of the particle suspensions was studied using a fixed solids concentration, 30 vol% (based on the slurry volume). The particle blends were considered as follows (alumina vol% + silica vol%) i) 29.9 + 0.1, ii) 29 + 1, iii) 28 + 2 and iv) 25 + 5. At pH 8, blending the alumina slurry with the 100 nm silica particles, thus promoting hetero-aggregation due to the positive interaction between the two particle types, resulted in minimal variation in the sediment yield stress, 𝜏𝑦 ~ 175 Pa. However, at pH 10 the yield stress was measured to decrease from ~ 90 Pa (30 vol% alumina) to ~ 20 Pa for the 25 vol% + 5 vol% particle blend. Quite interestingly this reduction in yield stress matches the yield stress reduction for alumina only (i.e. particle concentration effect). Therefore, while the total solids concentration remains at 30 vol%, the observed reduction mostly likely results from dilution of the main slurry component (alumina), and the silica nanoparticles have minimal influence on the bulk rheology. For the larger silica particles (d50 = 800 nm) the same blending protocol once again resulted in little modification of the sediment yield stress when the two particle types aggregated (pH = 8). However, at pH 10 the yield stress of the 30 vol% slurry was observed to decrease slightly with the yield stress of the 25 vol% + 5 vol% particle blend equal to 70 Pa. This equates to a ~ 20 Pa reduction in the overall yield stress, but significantly this yield stress is greater than the yield stress of 25 vol% alumina (𝜏𝑦 ~ 25 Pa). Hence the minor component, colloidal silica, contributes to the overall rheology of the particle blend.

The research is ongoing to establish a particle system that provides effective lubrication and rheological modification of dense yield stress sediments.