The formation damage due to retention of colloidal particles suspended in injection waters has been recently analyzed as consisting of four more or less overlapping steps:
deposition onto pore surface,
pore bridging, with accumulation upstream from bridged pores
internal cake building near non-percolation threshold and,
external cake formation.
A new formation damage model accounting for these different steps has been recently proposed. The results reported in this paper show the consistency between model predictions and experimental results in the convective diffusion regime in the absence of pore bridging. The experiments have been carried out by injecting both negatively and positively charged well-characterized hydrophilic latex micro spheres inside model granular packs made of negatively charged sharp-edged silicon carbide grains of narrow size distribution. To prevent any pore bridging, dilute suspensions of latexes having a size much smaller than that of pore throats were injected at low Peclet number.
In the absence of any energy barrier between particles and pore walls, i.e. with positively charged latexes or with negatively charged ones at high ionic strength, the model predictions assuming that the potential of pore surface is homogeneous are in very good agreement with experimental results. Colloidal particles are progressively deposited onto pore surface as a monolayer which reduces pore size and thus permeability. Maximum surface coverage is found to be that predicted by random sequential adsorption theory and the corresponding hydrodynamic thickness is equal to nearly 0.8 times the particle diameter. With negatively charged latexes at low and medium ionic strength, the results reveals that the pore surface of our model porous medium is definitely heterogeneous in term of surface potential. Since the surfaces of natural porous media are obviously heterogeneous, this heterogeneity must be taken into account in the model to predict both deposition kinetics and subsequent permeability reduction.
The presence of colloidal particles in fluids flowing through oil reservoirs often results in severe permeability damages around the well bore causing a decrease in well productivity and injectivity When caused by kinetically stable particles, which are not spontaneously adsorbed onto pore surface and whose size is much smaller than that of pore throats, permeability damage is observed to occur mainly (and sometimes only) at intermediate velocities, i.e. at a distance from injection well where remediation is questionable. Thus, modelling particle retention and its effect on permeability is very useful to predict and prevent injectivity or productivity damages. P. 161^