Summary

This work identifies a mechanism that could be responsible for substantial gas invasion into a cemented column after placement. A model is developed to predict the risk for gas invasion and to aid in the design of a cement job suggesting a "tail" and a delayed "lead" slurry. The model provides the absolute theoretical migration height when cement properties and formation parameters are known. The work accounts for the calculation of the hydrostatic pressure drop, for the estimation of the inflow rates resulting from shrinkage and filtration, and for the description of the bubble-rise process. A sensitivity study on important design and well parameters is included.

Introduction

The hydrostatic pressure of a slurry column decreases because of gelation, fluid loss into permeable layers, and chemical shrinkage. 1–3 After pressure balance between the cemented column and the gas formation, the slurry has already developed and continues to develop a certain gel structure around the gas zone, and substantial gas migration may be initiated subsequently.

Many special cementing techniques and cement systems have been designed to combat the problem. Some of these developments have proved relatively successful, but the problem of gas migration is still not under control. Until now, no model has existed that could describe explicitly the process of gas invasion into a homogeneous slurry during the transition time or plastic state of gelation. In most of the published work, the mechanism of early gas migration has been related to heterogeneities, such as the effect of filter cakes on the cement/formation interface,4 channels left from mud or spacer during the displacement process,5,6 nonhomogeneities within the slurry itself,7 and to flow through the high permeability of the "liquid" slurry matrix. 8,9

To minimize these obvious risks, certain guidelines were presented.4,7,9-14 However, some actual and more pessimistic reports on failed cement jobs that were performed under the recommended conditions prompt the following questions. Is there any mechanism that allows gas to invade into a cemented annulus even when the slurry is homogeneous and in the plastic state of gelation? Is the problem related to a not-understood interaction at the cement/formation interface or within the heterogeneous state of the slurry?

This work tries to find answers to these questions. We have developed a model that allows the understanding of early-time gas invasion after cement placement. While there is no clear solution to the gas migration problem, optimizing the conditions around the gas zone may resolve the problem. Thus it becomes a completion issue.

This content is only available via PDF.
You do not currently have access to this content.