Gas content of Gladys McCall reservoir brine

Abstract

On October 8, 1983, after the first full day of production from Sand No.8 in the Gladys McCall well, samples of separator gas and separator brine were collected for laboratory P-V-T (pressure, volume, temperature) studies. Recombination of amounts of these samples based upon measured rates at the time of sample collection, and at reservoir temperature (290 F), revealed a bubble point pressure of 9200 psia. This is substantially below the reported reservoir pressure of 12,783 psia. The gas content of the recombined fluids was 30.19 SCF of dry gas/STB of brine. In contrast, laboratory studies indicate that 35.84 SCF of pure methane would dissolve in each STB of 95,000 mg/L sodium chloride brine. These results indicate that the reservoir brine was not saturated with natural gas. By early April, 1987, production of roughly 25 million barrels of brine had reduced calculated flowing bottomhole pressure to about 6600 psia at a brine rate of 22,000 STB/D. If the skin factor(s) were as high as 20, flowing pressure drop across the skin would still be only about 500 psi. Thus, some portion of the reservoir volume was believed to have been drawn down to below the bubble point deduced from the laboratory recombination of separator samples. When the pressure in a geopressured geothermal reservoir is reduced to below the bubble point pressure for solution gas, gas is exsolved from the brine flowing through the pores in the reservoir rock. This exsolved gas is trapped in the reservoir until the fractional gas saturation of pore volume becomes large enough for gas flow to commence through a continuous gas-filled channel. At the same time, the gas/brine ratio becomes smaller and the chemistry of the remaining solution gas changes for the brine from which gas is exsolved. A careful search was made for the changes in gas/brine ratio or solution gas chemistry that would accompany pressure dropping below the bubble point pressure. Changes of about the same magnitude as the scatter in the data appear to have occurred in mid-1985 when calculated flowing bottomhole pressure was in the range of 9400 to 9700 psi. After the amount of brine flowing through the rock near to the wellbore has exsolved enough gas for onset of gas mobility through a continuous gas-filled channel, another test for whether the reservoir is below its bubble point becomes possible. This ''bubble test'' consists of suddenly increasing flow rate so that bottomhole pressure drops. Gas expansion then results in a small portion of the free gas from near the wellbore being produced in a short period of time. The resulting ''bubble'' of gas has a higher natural gas liquids content than gas produced before and after the transient. ''Bubble tests'' were performed in February 1986 and April 1987. Neither test liberated enough additional gas to provide a detectable change in produced gas/brine ratio. However, observed small transients in Ethane/Methane and Propane/Methane ratios indicate that some free gas was produced from the near wellbore region. These results suggest that the bubble point pressure must have been in the vicinity of the calculated 9500 psi flowing bottomhole pressure during the second of 1985. They conclude that: (1) Sand No.8 in the Gladys McCall well was not saturated with natural gas at the reported initial reservoir pressure of 12,873 psia; (2) flowing bottomhole pressure became less than the bubble point pressure during 1985; and (3) bubble point pressure was in the range of 9200 to 10,000 psi.