Oil·Based Drilling Mud as a Gas·Hydrates Inhibitor R.B. Grigg, * SPE, and G.L. Lynes, Conoco Inc. Summary. Gas-hydrates fonnation must be considered when petroleum reservoirs are developed in arctic regions and deepwater environments. This paper demonstrates that gas hydrates can fonn in oil-based muds, but that two major components-oil and dissolved solids in the aqueous phase-significantly inhibit this fonnation. This work identifies two major components in oil-based drilling mud that affect gas-hydrates fonnation. The temperature and extent of gas-hydrates fonnation both can be inhibited significantly, but not necessarily prevented, in oil-based drilling muds. A system that contained 20-vol % water and had an oil-continuous phase inhibited gas-hydrates fonnation 5 to lOOP. Dissolved solids in a 19.22-wt% calcium chloride (CaCI 2 ) brine inhibited gas-hydrates formation 20 to 25°P and significantly reduced the extent of fonnation. Gas-hydrates fonnation in an oil-based drilling mud, prepared with 20- vol%, 19.22-wt% brine, was inhibited more than 30 0 P over the pressure range studied, 500 to 4,500 psig. In most cases, oil-based mud can be prepared with sufficient concentrations of dissolved solids to prevent gas-hydrates formation under downhole conditions. Mud samples should be tested to determine the temperature of gas-hydrates fonnation before field use. Introduction Gas hydrates have been known and studied for more than 160 years. l The petroleum industry first noted them in the 1930's when they were found to be the cause of plugging in natural gas lines. 2 Until recently, most research has been related to natural gas trans- portation. More recently, gas hydrates have been found in situ in both arctic and deepwater drilling. 2-6 As the industry moves into deeper water, the conditions (higher pressures and lower tempera- tures) become more favorable for the fonnation of gas hydrates in systems containing water and hydrocarbon gases. In fact, such occurrences in water-based muds and their detrimental effects have been noted. 7 In this paper, we do not review the properties or structure of gas hydrates because these are well documented. 2 ,g-n It suffices to say that many gases (carbon dioxide, nitrogen, hydrogen sulfide, methane, ethane, propane, and isobutane) that are common in the presence of water under pressure induce the formation of a solid phase at temperatures above the freezing point of the aqueous so- lution. The formation of gas hydrates in drilling mud can result in block- age of flowlines and valves, with potentially serious safety and ec- onomic consequences. In deepwater drilling, the drilling fluid may pass through regions during circulation where the temperature is low enough and the pressure high enough to fonn gas hydrates. In 1986, Sloan** tested an oil-based mud and detected no gas hy- drates. The mud was 20-vol%, 30-wt%, CaCl 2 brine. The question of whether oil-based muds could be used without concern for the fonnation of gas hydrates arose. In general, oil- based muds are not water-free and are as much as 30-vol % aque- ous phase. We chose 20 vol % for the aqueous phase. In each oil- based mud, oil was the continuous phase. It was believed that this would prevent gas-hydrates formation. Another known gas-hydrates- formation inhibitor is dissolved salts in the aqueous phase-in this work, CaCI 2 . A series oftests was proposed. This series was made up off our basic groups: pure water as the baseline, brine, oil-based mud pre- pared with pure water, and oil-based mud prepared with brine. Each system is discussed in detail. Experimental Equipment and Method Figs. 1 and 2 shows schematics of the two experimental appara- tus. Pig. 1 shows the blind cell, which was a 0.07-gal glass-lined autoclave reaction vessel rated to 5,400 psi. The system volume included the volume of the cell and tubing connected to the pres- sure transducer and pressure relief devices. The autoclave was fit- ted with a magnetically driven impeller to mix the gas and liquid. Mixing was essential because gas-hydrates fonnation is a surface • Now at Core Laboratories. "Personal communication with E.D. Sloan, Colorado School of Mines, Golden (1986). Copyright 1992 Society of Petroleum Engineers 32 phenomenon. If not disturbed, a gas-hydrates crust will fonn at the gaslliquid surface, retarding further fonnation of gas hydrates. A high-capacity temperature control unit, modified to control from 5 to 95°F, circulated an ethylene-glycol/water mixture through coils in the cell. A programmer controlled the bath temperature. The coil, impellers, and stainless-steel thennocouple well were the only metal in direct contact with the liquid phase. The relatively small mass of this system allowed rapid temperature response. A per- sonal computer was used in both setups for data acquisition and processing. Time, pressure, and temperature were the parameters of interest. Fig. 2 shows a windowed cell rated to 3,000 psi. The mixing system consisted of a tubular housing threaded into the top of the cell and a rod with a mixing bar on one end and a bar magnet fastened to the opposite end. The magnetic end of the rod was placed in the tubular housing. A ring magnet around the tubular housing was fastened to a reciprocating air cylinder that moved the stirring bar up and down in the windowed cell, creating a mixing action. This mixing system was vigorous enough for viscous drilling-mud mixtures. The window cell was placed in a liquid bath. The same temperature control system was used as described above, except the cooling coils cooled the liquid bath. Both test systems were pres- surized by injection of test gas. During each test, the system had a constant volume and compo- sition. A test series was run from lowest to highest starting pres- sure by the addition oftest gas. Gas was not removed during a test series because the composition would not be identical to the inject- ed gas. This would be especially evident in an oil-based mud be- cause gas solubility in oil is significant and varies between components. Also, there was the danger of losing liquid during depressurization. To observe gas-hydrates fonnation, a 1 'h-in. Lexan® plate was installed in the liquid bath. Safety glass on each side of the Lexan plate protected the surface from scratching. Observation was en- hanced greatly by installing a borescope with fiber optics, a video camera, and monitor. Results Each test apparatus was a closed, constant-volume system, and each test was done at constant composition. Gas-hydrates fonnation was detected by slope changes· in the temperature/pressure traces dur- ing cooling and heating of the system (Fig. 3). In Pig. 3, the rela- tively straight line from Point A to Point B has no gas hydrates fonning; the pressure drop was caused by the gas cooling. From Point B to Point C, the change in slope was caused by gas being encapsulated in the solid structure, forming the gas hydrates. From Point C to Point D, the system was heated and gas-hydrates for- mation continued. This was evident when either the pressure con- tinued to decrease or the pressure increase was less than predicted . From Point D to Point E, rapid dissociation of the gas hydrates occurred. At Point E, the last hydrates were dissociated. SPE Drilling Engineering, March 1992