In previous papers we investigated high-molecular-weight polyacrylamide adsorption under high shear rates in low-permeability media and found that, above a critical value ?c, adsorbed macromolecules can reduce the permeability by factors over hundred, suggesting a mechanism of pore throat bridging. New lab experiments have been designed specifically to elucidate the mechanism at the origin of this "bridging adsorption". Polymer injections were carried out over a wide range of shear rates in homogeneous high-permeability granular packs having hydrodynamic pore throats too large to be bridged by polymer macromolecules. When polymer is injected at low shear rate (?<?c) the adsorbed layer thickness eHS (calculated from permeability reduction values) does not depend on injection rate. When the injection occurs at higher shear rates (?>?c), eH increases slowly up to reach maximum values eHM increasing with injection shear rate. These eHM values were found to be large enough to explain the very high reductions in permeability obtained previously in low-permeability packs.
These results show that polymer adsorption in porous media can be increased significantly by the hydrodynamic forces normal to the pore wall, as soon as they become high enough to "push" additional macromolecules into the already adsorbed polymer layer. This mechanism increases both polymer adsorption density and adsorbed layer thickness. We propose to refer to this mechanism as a "flow-induced adsorption". This new interpretation is consistent with all results previously attributed to "bridging adsorption" and the new results reported in this paper. It provides an important conceptual tool to model polymer placement in water shutoff and to design conformance treatment.
In a series of previous papers,1–4 several experimental results were presented, dealing with polymer behavior in porous media under near-wellbore flow conditions. These experiments showed that, when a large number of pore volume is injected in low-to-medium permeability cores beyond a critical shear rate ?c, a plugging tendency is observed as well as a strong increase in adsorption density. 3 The plugging does not occur either in non-adsorbing conditions or in high permeability cores. To explain these observations, a mechanism called "bridging adsorption" was proposed, consisting of two steps:
the stretching of macromolecules by high elongation stresses and
the adsorption of these stretched macromolecules by forming numerous bridges accross pore throats.
Such a mechanism is consistent with several observations such as:
A rate of plugging increasing as permeability decreases. 1,3
A rate of plugging increasing with the presence of residual oil (which reduces pore throat size). 3
A rate of plugging increasing with adsorption energy (from neutral1,3 to cationic4 polyacrylamides).
A rate of plugging decreasing in cores having a broad pore size distribution. 3
However, such a mechanism did not fit very well with some other results, namely:
The critical shear rate (around 70 s-1) is significantly lower than the onset of coil-stretch transition (> 200 s-1). 5
The fact that more than 99% of polymer could flow easily (with very small apparent viscosity) through highly bridged pore throats.
The fact that oil could also flow easily with a very low resistance factor, through these strongly plugged cores.
These discrepancies with the proposed mechanism of "bridging adsorption" called for additionnal experiments. We decided to run the same type of coreflood experiments as before in cores with very high permeability, for which pore throat size was too large to allow polymer bridging, even with strongly stretched macromolecules. The experimental procedure was designed to measure very precisely the evolution of the thickness of the adsorbed polymer layer after each polymer slug.