Summary A program was performed to study the effects of cyclic pressure/temperature fluctuations of gas storage wells on annular leakage. A review of pertinent literature was conducted, and operators of gas storage wells were surveyed to determine their experience in the area. Results indicated that for most of the gas storage wells currently in operation. Surface leakage is not a problem. Where leakage is reported, cyclic fluctuations do not appear to be a significant cause. In the majority of cases, leakage occurs within the first few cycles, indicating that it is caused more by static then by dynamic loads. The only variables found to correlate with leakage at a high level of significance were depth and bottomhole pressure (BHP). Wells that leaked tended to be deeper and had higher pressures than those that did not. The only way to stop leakage effectively at these higher pressures is the use of mechanical sealing mechanisms. Introduction and Background Gas intrusion into cemented wellbores and the resultant leakage to either the surface or porous formation below the wellhead have been persistent problems in the gas industry for many years. These problems have resulted in significant safety hazards and economic loss to operators. Through the efforts of a number of researchers in the past 2 decades, cementing techniques in general have been vastly improved and have helped reduce the magnitude of the problem. Little work has been done, however, to determine the effects of pressure/temperature cycling on the bonding characteristics of annular cement to casing. Data of this type are of particular importance to operators of gas storage wells because these wells operate on periodic injection/withdrawal cycles with associated pressure/temperature fluctuations. The current project was initiated in an effort to begin study of this problem. It was conducted in two phases. First, available literature was reviewed to identify previous work in cementing technology in general and in the area of annular leakage of oil and gas wells in particular. Then, operators of gas storage wells were surveyed to determine the magnitude of the annular leakage problem, to identify similarities or differences in wells with known annular leakage, and to establish typical environmental/operating parameters of gas storage wells. Literature Review A computerized search of the open literature was conducted by use of the COMPENDEX data base (Dialog Information Services Inc.). From this search, 24 references were selected for more detailed review. These works were thought to describe fairly well the general evolution of thought from the early 1960's until the present on the causes and means of prevention of gas leakage (also called gas migration and annular gas flow) in cemented wellbores.1–24 A more detailed review of the papers than given here can be found in Ref. 25. Past research efforts can be divided into two broad categories: methods and materials designed to minimize the leakage of formation pressures through the cement itself (i.e., the percolation of gas into the cement during setting and curing and the resultant creation of leakage paths) and methods designed to prevent gas migration at the cement/casing and cement/formation interfaces. Some researchers categorized these as two separate failure mechanisms, while others categorized them as separate cases of a single mechanism. In either case, it is evident that means taken to prevent the formation of leakage paths through the cement itself will also help prevent the formation of leakage paths at the interfaces. Additional preventive measures may be necessary, however, to minimize or to prevent leakage at the interfaces. Most research has concentrated on the prevention of leakage through the cement column itself, and one fairly well developed theory explains the need for prevention of these types of failures. This theory states that annular gas flow is caused by a hydrostatic pressure loss sometime between placement of the slurry in the wellbore and the development of sufficient static gel strength to resist the percolation of gas into it.24 Static gel strength is that internally developed rigidity in the slurry that resists a force placed on it. The gelling process will begin immediately after pumping has stopped and will continue until the cement develops a set. As gel strength increases, the cement column begins partially to support itself and its volume decreases owing to loss of filtrate to the formations and/or to hydration. Hydrostatic pressure caused by the slurry column lessens as the column begins increasingly to support itself and as volume changes occur. If volume changes are large enough and if the gel strength is not yet sufficient to resist gas intrusion, leakage to other producing zones or to the surface (directly or through a "leap-frog" approach from zone to zone) may occur.