Salt-Free Cement An Alternative to Collapsed Casing in, Plastic Salts(includes associated papers 12952 and 13012 )

Abstract

Summary The high incidence (20 to 25%) of collapsed ordoglegged casing in the Williston basin led to an investigation of cementing systems and practices employed in that area. The vast majority of the collapse problems are occurring across the Charles formation. The Charles contains numerous salt sections over a 600- to 700-ft [183- to 213-m] gross interval. These salt sections exhibit plastic tendencies when exposed to aqueous drilling fluids-i.e., the saltsections flow. Under plastic conditions, the saltsections flow against the casing before the cementsheath can attain sufficient compressive strength toprotect the casing. These salt sections can be seen on conventionally cemented casing strings as 100% bondedsections in what otherwise appears to be poorlycemented intervals. The plastic salt exhibits a horizontal stress against the casing equivalent to the overburdenstress (about 1 psi/ft [22.6 kPa/m] of depth). The use of high concentrations (20) t(37.2%)of saltin cement slurries exhibit a retarding effect on thickening time and the rate of compressive strength development. These excessive development times appear to be contributing to the collapse problems defined. Laboratory investigation of salt dissolution rates in cement slurries, at various annular velocities show that salt-free cementslurries can be placed across the salt sections with finalsalt concentrations of less than 10% as a result of saltdissolution. This prompted an investigation involving the possible use of salt-free cement slurries, which exhibit controlled thickening times and much faster compressive strength development, across the Charles. To date, the wells that have been cemented in the Willistonbasin with the salt-free cement systems have shown no incidence of casing collapse, except as listed in the last paragraph under Field Experience. Introduction The use of salt in oil well cementing has been practiced for several years, especially for cementing casing with salt domes along the U.S. gulf coast, and has been included in cement slurries for various other reasons. Saltis used in cement slurries to obtain bonding to acrystalline salt, such as that existing in a salt dome. This practice has been extended to using salt slurries acrossany salt section, including those encountered in the Williston basin. Several salt-bearing formations are encountered in the Williston Basin-e.g., the Prairieevaporite, the Charles, the Pine, and the Dunham salts. Investigation of cementing practices in the Williston basin indicated a severe problem with collapsed ordoglegged casing in the Charles salt zones, with some instances of collapse reported in the Pine and Prairie salts as well. These collapse problems can occur within a timeframe of 4 weeks to 3 years. The salt sections creating the collapse problems exhibit plastic tendencies when exposed to aqueous fluids for extended periods of time. They can generally be seen on bond logs (Fig. 1) as four to six separate sections of apparent 100% bond in what otherwise appears to be a poorly bonded interval. These plastic salt sections apparently move against the casingprior to development of a competent cement sheath, subsequently exhibiting a horizontal stress against the casing equivalent to the overburden stress (about 1 psi/ft[22.6 kPa/m] of depth). The prevalent cementingslurries are composed of 50/50 or 65/35 pozzolanic cementscontaining varying amounts of gel (bentonite), 85 to95% mix water, and 18 to 37.2%salt. A comparison ofthe compressive strength development of a 65/35 Pozzolan system and Class G cement is given in Fig. 2. Laboratory Investigation The effects of varying salt concentrations on cementslurries have been demonstrated. As the salt concentration increases in a cement slurry(2 to 8%), it acts as an accelerator. Ten to 18% salt is a transition zone exhibiting no appreciable effect on the slurry. However, at concentrations greater than 18%, an increasingly greater retarding effect is exhibited on both setting time and the rate of compressive strength development (Fig. 3). Laboratory investigations were conducted to determine the dissolution rates of salt by cement slurries atlaminar and plug flow annular velocities. Reynold'snumbers of 100 for plug flow and 3,000 for turbulent flow were used in the annular velocity calculations. The test apparatus consists of a centrifugal pump capable of circulating a cement slurry through a 1-in.[2.54-cm] line and then through a compressed salt block. The pump is capable of achieving high or low displacement rates, depending on the flow characteristics desired for testing(plug, laminar, or turbulent flow). These flow rates are regulated by a rheostat control and valve system. The apparatus, being a closed system, circulates a known volume of cement in a loop from a reservoir tank across the salt section and back into the reservoir (see Fig. 4). JPT P. 320^