IODP Proceedings    Volume contents     Search

doi:10.2204/iodp.proc.327.107.2011

Expedition 327 CORK holes and CORK mechanical and hydraulic features

Creating ridge-flank basement borehole observatories during Expedition 327 began with setting a reentry cone at the seafloor and installing 53 m of large-diameter conductor casing into the uppermost sediments (Fig. F2; Table T1). The conductor casing helps hold the reentry cone in position, preventing the cone from sinking in unconsolidated sediments until a longer casing string can be cemented at depth. The conductor casing deployed at Site U1362 (and at Site U1301 during Expedition 301) had a 20 inch OD. Once the cone and conductor casing were installed in the seafloor, a hole was drilled through the underlying sediment and into the top few meters of basement using an 18½ inch OD tricone bit. Sixteen inch OD casing was installed and cemented to or across the sediment/basement interface. We subsequently drilled into basement with a 14¾ inch tricone bit to allow installation of a third casing string (10¾ inch OD) within upper basement. We omitted any attempt to core the uppermost basement section prior to casing it off because (1) uppermost basement tends to be rubbly and unstable in this area, contributing to difficulties with drilling, coring, and other operations, and (2) recovery within this interval tends to be poor. Hole U1362A was deepened by rotary coring once the 10¾ inch casing string was installed and cemented into place, whereas Hole U1362B was deepened with a 9⅞ inch tricone (noncoring) bit to make enough room for CORK installation.

Expedition 327 CORKs were built around a central 4½ inch OD casing system that extends deep below the seafloor into a previously drilled and cased borehole. The part of the CORK system that sits above the seafloor when installed, referred to as the “wellhead,” was constructed from concentric 4½ and 10¾ inch casing sections, with parts of the larger casing omitted or cut away between horizontally oriented, 30 inch OD bulkheads (Fig. F3). Seafloor sampling and valve manifolds, sensor packages, data loggers, and samplers were arranged within three 1.5 m high bays, offset by 120°, and separated by vertical gussets. The gussets provide strength to the wellhead and help to guide the ship’s camera system and the submersible platform around the bays during CORK installation and later operations, protecting instrumentation and valves. During Expedition 327, one bay was dedicated to monitoring and logging pressure data, another was dedicated to fluid sampling, and a third was dedicated for a flowmeter and to microbiological sampling or auxiliary pressure monitoring or other experiments. Cutouts on the bulkheads and gussets are designed to allow a submersible or ROV to hold on for stability and leverage, and signs attached to the gussets indicate the directions that valves should be turned during operation.

CORK systems are designed to hydraulically isolate the formation at depth from the overlying ocean, requiring the use of multiple seal components (Figs. F2, F4). The 10¾ inch OD casing systems deployed in Holes U1362A and U1362B included three independent mechanisms for sealing the formation at depth and inside 16 inch casing. First, tapered casing seal rings were designed so that the 10¾ inch casing would be sealed against the 16 inch casing immediately upon installation, using rubber O-rings and metal sealing surfaces in the casing hangers held in place by gravity and a mechanical latch (Fig. F4A). Second, swellable packer elements were bonded to the outside of sections of 10¾ inch casing (Fig. F4B). These elements use an FSC-11 Freecap elastomer (developed by TAM International, www.tamintl.com/) that expands in contact with seawater and which had an initial external diameter of 14¾ inches. Full expansion of the swellable packer elements to the inner diameter of the 16 inch casing (15⅛ inch) will likely require several weeks to months. Thus, the swellable packer element does not provide a seal immediately following casing installation but should provide a reliable casing seal over subsequent months to years. Finally, both the 16 and 10¾ inch casing strings in both Holes U1362A and U1362B were cemented using cement having a density of 15.9 lb/gallon (1.9 kg/L). One-inch square pieces of cellophane (“Cello Flake”) were added to the cement to clog pores and fractures adjacent to the borehole (lost circulation material [LCM]). In addition, 1.6% CaCl2 was added to the cement as a setting accelerant. The use of LCM and an accelerant was new for scientific ocean drilling and was considered necessary given the difficulty of achieving cement objectives during earlier work in fractured ocean crust. Cement was deployed around the shoe of the 10¾ inch casing without a cement retainer by backfilling the hole after the casing was landed and latched mechanically into place.

The use of a casing seal ring (Fig. F4A) might have prevented the flow of fluid and cement up the annulus between 10¾ inch casing and the borehole wall, but it seems likely that the extremely fractured, rubbly, and permeable formation allowed significant lateral flow of cement near the base of the hole. Although this cement may not have formed a complete hydrologic seal between the 10¾ inch casing and the formation, the cement should have helped to separate the main borehole area from the dead (annular) space outside the 10¾ inch casing, and this should improve the quality of future geochemical and microbiological samples. If the cement also provides a hydrologic seal, this should additionally improve measurements of formation pressure, particularly transient responses to tides and seismic events. Postinstallation monitoring of pressure within the annular gap of the cased interval will allow quantitative assessment of the quality of the cement seal at depth.

The main CORK seal (Fig. F4C) is located at the base of the CORK wellhead, where it seals against the 10¾ inch casing in the throat of the reentry cone. Holes were drilled and tapped through the CORK seal sub for packer inflation, pressure monitoring, and fluid sampling lines. Below the seal sub are crossover subs and 4½ inch casing that extends to depth below seafloor. The length of 4½ inch casing that hangs below the CORK body is adjustable at sea depending on hole depth and conditions, available joint lengths, and the dimensions of other CORK components.

Expedition 327 CORKs used two packer systems for sealing 4½ inch casing in the open hole and in casing (Fig. F4D, F4E). At each CORK packer depth, a hydraulic packer was deployed below a pair of swellable packer elements. The hydraulic packer, inflated by pumping seawater down the pipe once the CORK was lowered into position, was intended to provide an immediate seal, as needed for monitoring short-term formation pressure response and limiting the continued inflow of cold bottom water. The swellable packer elements should provide a reliable long-term seal but will take several months to expand against 10¾ inch casing or the wall of the open borehole. The final CORK seal component uses the weight of the top plug, instrument string, and a sinker bar to hold an O-ring against a tapered seal area in the top of the wellhead (Figs. F2, F4F).

Expedition 327 CORKs included multiple perforated and screened components to permit pressure monitoring and fluid sampling while protecting downhole instrumentation from basement collapse (Fig. F5). Three perforated drill collars were deployed above a bullnose and below the deepest inflatable packer in each hole, with lines of 2 inch holes separated by 9 inches running vertically up four sides of the collars (Fig. F5A, F5B). The collars provide weight needed to pull the lower end of the CORK into the hole during deployment. Additional sections of perforated 5½ inch casing were deployed above the collars, with lines of 1½ inch holes separated by 8 inches running vertically up four sides of the casing (Fig. F5C). The perforated collars and casing, bullnoses, and crossover subs needed to attach these components to the rest of the CORK casing string were coated internally with TK-34XT (Tuboscope, www.tuboscope.com/) and externally with Amerlock 400 (Amercoat, www.amercoatcanada.com/). Both of these epoxy-based compounds should reduce the extent of reaction between the steel collar and casing alloys and the warmed borehole fluids (Orcutt et al., 2010), improving our understanding of the composition and microbiology of subseafloor fluids.

Formation pressure is monitored and borehole fluids are sampled at the CORK wellheads using wire-wrap miniscreens installed at depth (Fig. F5D–F5F). Stainless steel screens are used for pressure monitoring and standard geochemical sampling, whereas microbiological sampling is completed through titanium screens.

Three forms of tubing umbilical were prepared for use during Expedition 327 (Fig. F6). One plastic-jacketed flat pack contained two ¼ inch OD stainless steel tubes for pressure monitoring and a single ½ inch OD hydraulic packer inflation line (Fig. F6A). A second plastic-jacketed flat pack contained three ½ inch OD stainless steel tubes intended mainly for geochemical sampling (Fig. F6B). A final plastic and woven metal–jacketed umbilical was constructed around a ½ inch polytetrafluoroethylene (PTFE) tube for use with microbiological sampling from the deepest monitored interval (Fig. F6C). Umbilical tubes were deployed as each CORK was constructed, from the bottom to the top, with tubes being passed through the inflatable and swellable packers as needed (Fig. F6D, F6E). Final connections were made at the top of the CORK casing, connecting sampling and monitoring lines to pigtails that were preinstalled to pass through the seafloor CORK seal (Fig. F6F).

Each CORK installed in Holes U1362A and U1362B included a lateral 4½ inch casing section that extended at a ~20° angle up from below the lower bulkhead (shown in Fig. F2), leading to the name “lateral CORK” or “L-CORK.” The angled lateral casing penetrated the lower bulkhead with an offset of 9¼ inches from the CORK center line and was terminated at the top with a large-diameter (4 inch) ball valve (Fig. F7). The ball valve was modified to include welded valve handle stops, and holes were drilled through the valve casing to avoid trapping air in dead space around the ball that could lead to development of a large differential pressure and damage during deployment. A custom ring clamp on top of the ball was deployed with a dust cover to prevent fouling of the valve. The dust cover will be removed and replaced with a flowmeter at one of the Expedition 327 CORKs in summer 2011 so that when the ball valve is opened the natural overpressure in basement will drive fluid out of the borehole at an anticipated rate of 5–10 L/s. This flow will be allowed to continue for at least 12 months, comprising a controlled cross-hole experiment that will allow assessment of directional basement properties at a crustal scale. Because we wished to be able to run the long-term free-flow experiment from either of the new CORKs, no lower plug was placed on the instrument strings deployed in these observatories. However, plug seats (3⅜ inch ID) were placed between the swellable and inflatable packer elements at depth in both systems so that plugs could be installed in the future if this were desired following recovery of the initial instrument strings.

Unlike Expedition 301 CORKs, which were deployed with most valves open and flowing freely, the Expedition 327 CORKs were deployed with valves either closed (in flowmeter/microbiology and geochemistry bays) or sealed and being monitored (in the pressure bay). No submersible or ROV expedition was planned to install instruments immediately after Expedition 327, and we wished to have a sustained period of sealed conditions to allow recovery of formation pressure to a predrilling state.