Constitutive modeling of chalk – application to waterflooding

Abstract

The compaction of chalky reservoirs during oil extraction and other important problems like the "casing collapse" or the "chalk production" are related to the mechanical properties of chalk. Controlling compaction is very important because reservoir deformations imply seabed subsidence that endangers the offshore stations. The first explanation of subsidence links the compaction to the pore pressure decrease in the reservoir. The solution was the injection of gas and water into the oilfield in order to repressurise the reservoir. But the waterflooding induced additional subsidence. Though many studies have been already performed on chalks, the basis mechanism of the water sensitivity was not defined. Obviously, no satisfactory constitutive law can be written without this deep insight of the phenomenon. This is the scope of the ongoing EC Research Program Pasachalk. The origin of the research is in the comparison of experimental results obtained on Lixhe chalk and on Jossigny silt which showed that the influence of water on pure high porosity chalk is similar to that on partially saturated soils (Delage & al. 1996). Another extensive experimental analysis of the influence of the saturating fluid on Lixhe chalk behaviour concluded that the water-weakening effect might be suction related. Hence the idea appeared to apply the knowledge, the approach, and the tools of the partially saturated soil mechanics to the understanding, description, and modelling of chalk behaviour during changes in saturation fluids, such as when waterflooding. This paper presents the developed constitutive model, which is a cap type plasticity model coupled with the Barcelona one (Alonso & al. 1990) for taking the suction effect into account. The model parameters are calibrated based on the experimental results. The validation of the model is performed on a waterflooding experiment. We show that the model is able to reproduce qualitatively and quantitatively the observed basic phenomena. As viscous effects may be important for the reservoir exploitation time, some first insights in the development of an elastoviscoplastic constitutive model are presented. The idea is to combine the frameworks of the Perzina viscoplasticity and of the preceding developed elastoplastic model.