A critical issue in matrix acidizing of vertically extensive carbonate reservoirs is the acid distribution along the wellbore. This estimation is very important especially for the case where the reservoir properties (permeability and damage distribution) vary along the wellbore. Several acid placement models for oil reservoirs have been developed and applied in the field successfully. However, when evaluating acidizing in a gas well, the models are not adequate because of the viscosity contrast between the reservoir fluid (gas) and the injected fluid (acid) and because of relative permeability effects. We have developed an acid placement model for acid injection into vertically extensive gas wells in heterogeneous carbonate reservoirs that includes gravity segregation in the wellbore, viscosity contrast, relative permeability effects, and a wormhole model. The effects of all of these factors on the local injectivity during acid injection are incorporated into a dynamic skin factor model.
To test this new model, we analyzed coreflooding data in the literature. The analysis showed that the differential pressure across a gas saturated core did not linearly drop during acidizing, but instead increased during a large part of the acid injection period. From this analysis, we determined that the pressure drop in a spent-acid zone ahead of the wormholes dominates the overall pressure drop behavior.
Applying this model to typical field conditions, we find that a natural viscous diversion takes place in heterogeneous gas reservoirs because of the viscosity contrast between the gas and the injected acid. In particular, we have modeled thick carbonate gas reservoirs having a few thin, very high permeability thief zones. Before wormholes break through the damage zone in the high permeability parts of the reservoir, this viscous diversion effect distributes acid more uniformly than expected. However, after wormholes break through the damaged zone, thief zones will take a large portion of the injected acid. Another important prediction with this model is that the injection pressure often increases during acid injection into gas wells. Without an understanding of the role of mobility contrast on the injection pressure response, operators may incorrectly conclude that the injected acid is not stimulating the formation.
Introduction
Matrix acidizing is a widely used well stimulation method in carbonate reservoirs to increase well productivity. In the process, acid flows down the well casing or tubing into the formation. Then the acid reacts with rock so that damage near the wellbore can be removed or bypassed by creating large channels known as wormholes. Effective stimulation requires that sufficient volume of acid to create long enough wormholes is placed in all target zones. To predict the acid distribution along the wellbore, acid placement models have been developed.
Extensive research has been done on modeling acid placement in oil wells (Jones and Davies, 1996; Glasbergen and Buijse, 2006; Mishra et al, 2007). Those models have been applied in the field successfully. However, when evaluating acidizing in gas wells, the models are not adequate because the mobility difference between the reservoir fluid (gas) and the injected fluid (acid) possibly affects the acid distribution.
Zhu et al. (1998) presented a real-time monitoring model for acidizing of gas wells. In their model, they estimated damage skin evolution during an acid treatment by evaluating viscous skin effect separately. On the basis of their model, Fadele et al. (2000) modeled acid treatments for gas reservoirs. However, their model didn't account for wellbore phenomena that are important to model acid placement.
In this work we develop an apparent skin factor model based on acid coreflooding results in a gas saturated core (Shukla et al, 2006). The model accounts for damage removal, the viscosity contrast, and relative permeability effects.
