Effect of Water Breakthrough on Sand Production: Experimental and Field Evidence

Abstract

Effect of Water Breakthrough on Sand Production: Experimental and Field Evidence Anne Skjaerstein and Johan Tronvoll, IKU Petroleum Research, Frederic J. Santarelli, Oilfield Production Consultants, Havard Joranson, Statoil Norway. Abstract It has been reported from different field operators that the onset of sand production in a particular field coincided with water breakthrough at the production wells. Several mechanisms have been proposed to explain these observations such as pore pressure gradient increase near perforations due to relative permeability effects, reduced capillary bond due to increased water saturation, and loss of rock cohesion because of chemical incompatibility between produced water and the reservoir rock. Yet, there does not seem to exist any systematic study of such a phenomenon on a large scale, neither capable of supporting any of the proposed mechanisms nor the statement that the onset of water breakthrough may trigger the onset of sand production. As a consequence, a vast campaign of physical model experiments of perforation cavities using field and outcrop material was undertaken to observe directly the effect of water breakthrough upon the onset of sand production. The experiments, have convincingly demonstrated that the two phenomena could not be linked directly, but that once sand production has initiated, water production could accelerate the sand influx thus making it more noticeable and problematic for daily operations. Finally, the problem of the coincidence of water breakthrough and onset of sand production has been analysed on a series of more than 40 oil and gas wells in weak sand reservoirs. This analysis of field data further confirms the laboratory and theoretical findings described in the paper. Introduction Over the years, a 'common knowledge' has developed within the petroleum industry stating the direct relationship between the onset of sand production and this of water breakthrough. However, despite this knowledge there are very few hard data to back up such a statement. Two noticeable exceptions are:a data set presented by Shell that seemed to indicate such a coincidence;another one presented by Agip on a series of gas wells of the Adriatic sea that demonstrated exactly the contrary. Also, few laboratory simulations have been reported that clarify possible mechanisms leading to sand production triggered by water breakthrough. From a theoretical point of view, various explanations to such observations may be considered such as:reduction of apparent cohesion of the rock due to reduced capillary bonding, which for some rocks may be quite considerable, while in sandstones the capillary suction is in the order of 0 – 1 MPa;decohesioning of the rock due to instability of pore lining minerals and cement bond chemically incompatible with the formation water;fluid flow erosion of the grain matrix due to increased fluid drag during two-phase flow;thermally induced stress increase due to transport of warm water from deeper reservoir zones;erosion of sand bed in production liner due to two-phase flow;plugging of parts of the pay zone due to relative permeability effects or fine particle depositions, resulting in concentration of fluid inflow at permeable (and often weak) intervals leading to hydrodynamically induced instability; P. 565^