RHEOLOGICAL EVALUATION OF DENSE SUSPENSIONS: SIMULATION OF A FRESH CEMENT PASTE

Marble dust suspended in a xanthan gum solution was evaluated, through use of a pipe flow viscometer and a Rheometrics Pressure Rheometer, RPR (concentric coaxial cylinder viscometer), to study the rheological behavior of a dense suspension similar to a fresh cement paste. The marble dust allowed the study of rheological phenomena exhibited by cement pastes, unhindered by complications due to cement hydration. The size distribution of the marble dust particles closely approximated that of cement. The concentrations of marble dust in the suspensions ranged from 19 wt % up to 80 wt %.;Correlation of flow curves generated from the two viscometers was difficult due to an apparent "slippage" that occurred at the boundary surfaces, which became more significant as marble dust concentration was increased. The use of a grooved bob on the RPR considerably reduced the amount of slip. Large annular gap sizes (2.5 mm gap width) also contributed to inconsistent results on the RPR.;The Reiner-Riwlin equation, derived for a Bingham fluid was used to approximate the flow curves generated on the RPR for the concentrated marble dust suspensions. The Bingham model has often been applied to fresh cement pastes in the past. The marble dust suspensions displayed antithixotropic (rheopectic) behavior (representative of a system which is building structure with applied shear) regardless of the water/solids ratio tested. The duration of the RPR tests had an effect on data only when sedimentation occurred while a test was in progress. Sedimentation became apparent when testing marble dust concentrations above 50 wt %, and was indicated by a buildup of the marble dust in the bottom of the viscometer cup that was observed at the completion of the test. The sedimentation produced an irregular shear thinning within the suspension which (if not recognized to be due to sedimentation) could have been misinterpreted to be a part of the structural buildup/breakdown processes occurring. Such misinterpretation of data may account for the many contradictions in literature regarding the antithixotropic/thixotropic behavior exhibited by fresh cement pastes. The marble dust suspensions generally provided a good model of a system similar to a fresh cement paste.