Formation of Hydrates During Deepwater Drilling Operations

Abstract

Summary Two deepwater wells in widely separated geographical areas have experienced natural gas hydrate formation during drilling operations. In both cases, hydrates, ice-like mixtures of natural gas and water, plugged subsea equipment, causing difficulties in subsequent operations. The potential for hydrate formation merits consideration during planning and conducting of deepwater drilling operations. Introduction During recent years, the petroleum industry has continued to extend and to expand deepwater exploration drilling efforts in many areas of the world. As drilling water depths increase, the potential for natural gas hydrate formation during drilling operations also increases. Hydrates are solid mixtures of natural gas and water that resemble dirty ice in appearance. Unlike ice, however, they can form at temperatures well above 32°F [0°C] when sufficient pressure is present. The higher seafloor hydrostatic pressures and lower environmental temperatures encountered in deepwater drilling increase the likelihood of hydrate formation in chokelines, drilling risers, blowout preventers (BOP's), and subsea wellheads. This paper describes two cases in which gas hydrates formed in subsea equipment during drilling operations, leading to costly delays and special remedial procedures. Properties of Hydrates Hydrates were first observed by Davy in 1810.1 They were introduced to the petroleum industry in 1934 by Hammerschmidt,2 who determined that these substances were responsible for the freezing of gas transmission lines. Hydrates are part of a group of substances known as clathrates because they consist of "host" molecules (water) that form a lattice structure that acts like a cage to entrap "guest" gas molecules. C1, C2, C3, C4, H2S, and CO2 are known to produce hydrates with water. Hydrocarbons larger than nC5 cannot form hydrates with water because of limited host-molecule cage size. A noteworthy property of hydrates is the amount of gas trapped in a given volume. One ft3 [0.028 m3] of hydrate can contain as much as 170 scf [4.8 std m3] of gas. When hydrates decompose because of reduced pressure and/or increased temperature, they can produce a large volume of gas. If hydrates decompose in a limited-volume, sealed container, such as a core barrel, very high pressures that can rupture the container can be generated. Combinations of pressure and temperature that allow natural gases and water to form stable hydrates have been determined accurately. A major factor that affects hydrate formation is the natural gas composition. Fig. 1,3 a result of early hydrate research, illustrates how gas composition, i.e., specific gravity, can affect the pressure/temperature range where hydrates are stable. A second factor that affects hydrate formation is the liquid-phase composition. Many substances, when added to the liquid phase, will depress hydrate formation temperature at a given pressure. Pressure/temperature conditions under which hydrates are stable are well-established in comparison with present knowledge concerning time and conditions needed to form a hydrate initially. Recent work in this area determined that the time to form a hydrate depends on several factors, including the degree of supercooling.4 Supercooling occurs when the actual temperature of a gas/liquid mixture is lower than hydrate-equilibrium temperature at a given pressure. Deepwater Drilling Drilling water-depth records increased greatly during the last decade; the number of wells drilled worldwide in deep water, however, is still limited (see Figs. 2 and 35-12). Through 1986, about 150 wells worldwide have been drilled safely in water more than 2,000 ft [610 m] deep. The potential of gas-hydrate formation during deepwater drilling operations has been recognized.1 The actual formation of hydrates, however, has not been reported. Possible explanations for the absence of reports are thatthe industry's deepwater-well drilling experience is limited;the likelihood of conditions favorable for hydrate formation - i.e., sufficient quantities of natural gas and water combined with the right pressure and temperature conditions - is low; andthe lack of understanding and recognition of hydrates leads to the attribution of well difficulties to other causes - e.g., barite settling and mechanical failures. Temperature Gradients With increasing water depth, static seawater temperature normally declines in a roughly parabolic manner. For the Gulf of Mexico,14 the average seawater temperature (see Fig. 4) drops rapidly to about 48°F at 1,500 ft [9°C at 457 m]. Below this depth, the water temperature drops more slowly to about 40°F at 3,000 ft [4°C at 914 m]. Case Histories of Deepwater Hydrate Occurrences Case 1. Case 1 describes an occurrence of hydrates in a well located at 1,150-ft [350-m] water depth off the U.S. west coast with a seawater temperature of 45°F [7°C] at the mudline. Fig. 5 shows the casing strings in this wellbore. After 7-in. [17.78-cm] casing was run and cemented, the subsea casing hanger packoff was set and pressure tested. The BOP's were round-tripped from the ocean floor to change BOP ram sizes. While cement was drilled at the casing shoe, low gas units were detected. This increased sporadically. The BOP's were closed and casing pressure was increased to 1,300 psi [8964 kPa]. A through-drillstring noise log indicated that a gas influx was occurring from a sand at 7,750 ft [2362 m]. Gas was entering from the formation, channeling through the primary cement column, and migrating up the 7×9 5/8-in. [17.78×24.45-cm] casing annulus. The wellhead hanger packoff was leaking (even though it had been pressure tested), allowing the migrating gas to enter the freshwater mud at the subsea wellhead. An unsuccessful attempt to recover the BOP wear bushing while holding pressure on the well was made. Hydrates may have prevented the retrieval of the wear bushing. Recovery of the wear bushing would have allowed access to the leaking casing-hanger packoff. Attempts to pump mud into the 7×9 5/8-in. [17.78×24.45-cm] casing annulus with pressures up to 3,000 psi [20.7 MPa] were unsuccessful. (Subsequent investigation of the casing-hanger packoff revealed that the packoff had not reached its correct setting position, probably because of debris in the wellhead. The result of the improperly set packoff was a seal that held only from above and not from below.) To stop the gas influx, we decided to perforate the casing just above the gas sand and to pump heavyweight mud into the formation. To ensure a successful casing perforating job, the drillstring was severed and stripped up through the BOP's until the severed drillstring end was above the gas sand. A through drillstring perforating gun was then run to shoot the 7-in. [17.78-cm] casing just above the gas sand. The gas influx was killed by pumping 14.2-lbm/gal [1702-kg/m3] mud down the drillstring and into the formation at surface pressures up to 3,100 psi [21.4 MPa]. At the conclusion of the kill operation, approximately 7 days after gas was first detected, both the chokeline and the kill line were found plugged. Subsequent operations were hampered by the inability to use either line. Numerous attempts to unplug the chokeline and the kill line by applying pressure surges at the surface were unsuccessful. After cementing operations, which secured the wellbore, the BOP's were recovered. Hydrates and trapped gas were found in the chokeline and the kill line of the bottom eight riser joints. Case 1. Case 1 describes an occurrence of hydrates in a well located at 1,150-ft [350-m] water depth off the U.S. west coast with a seawater temperature of 45°F [7°C] at the mudline. Fig. 5 shows the casing strings in this wellbore. After 7-in. [17.78-cm] casing was run and cemented, the subsea casing hanger packoff was set and pressure tested. The BOP's were round-tripped from the ocean floor to change BOP ram sizes. While cement was drilled at the casing shoe, low gas units were detected. This increased sporadically. The BOP's were closed and casing pressure was increased to 1,300 psi [8964 kPa]. A through-drillstring noise log indicated that a gas influx was occurring from a sand at 7,750 ft [2362 m]. Gas was entering from the formation, channeling through the primary cement column, and migrating up the 7×9 5/8-in. [17.78×24.45-cm] casing annulus. The wellhead hanger packoff was leaking (even though it had been pressure tested), allowing the migrating gas to enter the freshwater mud at the subsea wellhead. An unsuccessful attempt to recover the BOP wear bushing while holding pressure on the well was made. Hydrates may have prevented the retrieval of the wear bushing. Recovery of the wear bushing would have allowed access to the leaking casing-hanger packoff. Attempts to pump mud into the 7×9 5/8-in. [17.78×24.45-cm] casing annulus with pressures up to 3,000 psi [20.7 MPa] were unsuccessful. (Subsequent investigation of the casing-hanger packoff revealed that the packoff had not reached its correct setting position, probably because of debris in the wellhead. The result of the improperly set packoff was a seal that held only from above and not from below.) To stop the gas influx, we decided to perforate the casing just above the gas sand and to pump heavyweight mud into the formation. To ensure a successful casing perforating job, the drillstring was severed and stripped up through the BOP's until the severed drillstring end was above the gas sand. A through drillstring perforating gun was then run to shoot the 7-in. [17.78-cm] casing just above the gas sand. The gas influx was killed by pumping 14.2-lbm/gal [1702-kg/m3] mud down the drillstring and into the formation at surface pressures up to 3,100 psi [21.4 MPa]. At the conclusion of the kill operation, approximately 7 days after gas was first detected, both the chokeline and the kill line were found plugged. Subsequent operations were hampered by the inability to use either line. Numerous attempts to unplug the chokeline and the kill line by applying pressure surges at the surface were unsuccessful. After cementing operations, which secured the wellbore, the BOP's were recovered. Hydrates and trapped gas were found in the chokeline and the kill line of the bottom eight riser joints.