Previous   |    Next

doi:10.2204/iodp.proc.327.109.2011

Continuum of development

Initial CORK design

In simple terms, the initial CORK design consisted of a steel casing that was open to the formation at depth and was sealed at the seafloor (Fig. F2) (Davis et al., 1992). The sealing mechanism consisted of a removable plug with a pressure sensor and data logger. A cable with 10 thermistors was suspended below the plug, and fluid sampling was accessible through a ball valve attached to polytetrafluoroethylene (PTFE) tubing, which was attached to the thermistor cable inside the borehole. The first CORK deployments occurred at Sites 857 and 858 in an active hydrothermal system. Pressures at depth were expected to be greater than hydrostatic. Thus, it was expected that once the ball valve was opened, fluids from depth would ascend the PTFE tubing to be sampled at the seafloor. However, the PTFE tubing was not stiff enough and kinked upon deployment.

The subsequent CORK deployment on Hydrate Ridge, Cascadia margin (Holes 889C and 892B), used the same basic design without PTFE tubing (Westbrook, Carson, Musgrave, et al., 1994). However, this design required fluids to ascend through the mild steel casing before being sampled, leading to contamination, especially for transition metals (e.g., Wheat et al., 2004). For 403 days an instrument package was connected to the sampling port to monitor flow driven by natural overpressure. This package included two continuous fluid samplers (OsmoSamplers; more details are given below), pressure sensors, flow meters, and chambers for gas hydrate and carbonate precipitation from which microbial communities could be extracted. The formation of gas hydrate began 45 days into the experiment and reduced the flow rate considerably.

This initial CORK design was also used on the next four deployments (ODP Legs 156, 168, 174B, and 195) (Shipley, Ogawa, Blum, et al., 1995; Davis, Fisher, Firth, et al., 1997; Becker, Malone, et al., 1998; Salisbury, Shinohara, Richter, et al., 2002). To minimize contamination with the casing, OsmoSamplers were attached to the thermistor cable during each expedition, with the exception of Leg 174B (Fig. F2). These samplers were placed in open holes whenever possible, below the mild steel casing.

Advancement in CORK design: ACORK and CORK-II

A major change in CORK design, known as the advanced CORK (ACORK), was used during ODP Leg 196 (Mikada, Becker, Moore, Klaus, et al., 2002). Instead of having all of the sensors in the borehole, pressure sensors for the ACORK were attached to multiple stainless steel tubes that were bundled in a protective hard-plastic coating (called an umbilical) and attached to the outside of the casing. An umbilical is designed to terminate at a depth horizon of interest. This horizon is isolated by the use of packers, which are like giant tires that are inflated to fill the annulus of the borehole. Individual tubes within the umbilical terminate at miniscreens, which are 1–2 m long and ~3 cm in diameter, with a wire wrap that provides an effective open cross section of ~15%. The umbilical and miniscreens are attached to the outside of the CORK casing using plastic zip ties, stainless steel hose clamps, and duct tape.

The Leg 196 ACORK design was further expanded on with the CORK-II design, which included an umbilical attached to the outside of the casing and an open borehole where instrument strings can be placed in a hydrologic horizon (Fig. F3) (Morris, Villinger, Klaus, et al., 2003; see figs. F3, F4 in Fisher et al., 2005). From deepest to shallowest, borehole instrument strings consist of a sinker bar, OsmoSampler packages, Spectra cable, self-recording temperature loggers, and sealing plugs. The plugs were modified to seal with a tapered O-ring seat and gravity instead of by latching in place within the CORK, which made the plugs difficult to remove. Gravity plug seals were used to avoid issues with the pulling tool, which is heavy and difficult to manipulate. These plugs also require a separate submersible dive to release the latch and often were not effective, requiring additional dives. Another change was to place gravity seals near the bottom of the casing with a redundant seal at the seafloor. The lower seal, located just above the OsmoSampler package, limited exposure of borehole fluids to mild steel casing.

Two novel downhole packages were deployed in CORK-IIs during ODP Leg 205 on the Costa Rica margin (Morris, Villinger, Klaus, et al., 2003; Jannasch et al., 2003; Solomon et al., 2009). The OsmoSampler packages deployed in Hole 1255A were housed within a steel casing with a titanium stinger. This stinger protects sample intakes and penetrates a seal that opens to a portion of perforated and wrapped casing (a “screened” section), allowing fluid exchange among the formation, borehole, and sample intakes. When the stinger is removed, the valve is pulled closed, maintaining the pressure of the formation at in situ conditions. This string was recovered in 2004 and redeployed with the R/V JOIDES Resolution, but it was not properly seated when redeployed. During recovery attempts in 2009 the line parted at the steel package.

In contrast, the OsmoSampler packages in Hole 1253A were housed in steel casing and exposed to the formation through the open portion of the borehole and within the protective casing, similar in style to those deployed during Leg 168 that used a polycarbonate housing. When the downhole packages were recovered by the JOIDES Resolution in 2004, one of the two identical packages was lost during handling procedures in the moonpool. Two packages were deployed in 2004; however, only one package—the package protected by the casing—was recovered using the DSRV Alvin (Woods Hole Oceanographic Institution) in 2009. A new string with a polycarbonate instead of steel housing was deployed in 2009.

CORK-II systems were deployed in two new holes during IODP Expedition 301 (Holes U1301A and U1301B), and the original-style CORK deployed in ODP Hole 1026B was recovered, providing substrate for microbial studies (Nakagawa et al., 2006; Steinsbu et al., 2010), and was replaced with a CORK-II (Fisher et al., 2005). OsmoSampler packages were deployed within the boreholes with a gravity seal at depth and at the seafloor. Additional fluid samplers were deployed on the wellhead, which was partitioned into pressure, chemical, and microbiology bays, each with its own system of valves (Fig. F4). These CORK-IIs also included a dedicated umbilical for microbiological sampling and downhole experimentation (Fisher et al., 2005; Orcutt et al., 2011).

Current CORK advancements: L-CORK and genius plug

The lateral CORK (L-CORK) was modified from the CORK-II design to add a 10.2 cm (4 inch) diameter ball valve connected to a break-out tube in the CORK body at the wellhead to hold a novel inductive flow meter, which provides an additional way to access downhole samplers, sensors, and fluids. L-CORKs were deployed during IODP Expedition 327 on the Juan de Fuca Ridge flank (Fisher, Wheat, et al.) and will be deployed during IODP Expedition 336 (Edwards et al., 2010) (Fig. F3). A second major change in the L-CORK design was the addition of a top valve that is latched in place. This design was instituted to avoid deployment issues with the top plugs that were encountered during Expedition 301. A third modification included redundant borehole seals. A combination of inflatable and water-swellable packers was employed. This was the first scientific use of swellable packers, which are 1.5 m in length and made of FREECAP FSC11 polymer (TAM International). A fourth modification was a deployment scheme that placed all of the borehole OsmoSampler packages within perforated epoxy-coated drill collars. This modification was implemented because ~50% of all downhole packages that extend beyond the protection of the cased hole are now entombed in the crust. Drill collars serve three purposes: (1) they maintain tension on the 11.4 cm (4.5 inch) diameter CORK casing; (2) perforations allow fluid circulation within a protected environment; and (3) the baked coating on the inside and outside of the collars minimizes contamination for microbiological and geochemical studies (Orcutt et al., 2010). Structurally, these drill collars were deemed critical because casing is strong in tension and relatively weak in compression. This probably contributed to the failed CORK deployments during Legs 195 and 206 and Expedition 301. A final modification was a new valve and coupler system to access fluids from the umbilical (Fig. F5). This change from the Aeroquip standard was necessary because the Aeroquip connection is mechanically difficult to complete for submersible pilots, it is difficult to discern whether the connection has been made, and constant maintenance is needed (Fisher, Wheat, et al.).

The L-CORK design will be employed during Expedition 336, but with the use of fiberglass casing (Edwards et al., 2010). Fiberglass was chosen to minimize the amount of potential contamination from mild steel casing. Fiberglass is strong in tension (as is steel) but weaker than steel in compression, yet it is more flexible than steel.

No fluid-sampling capabilities were incorporated into CORKS deployed during IODP Expeditions 319 (Saffer, McNeill, Byrne, Araki, Toczko, Eguchi, Takahashi, and the Expedition 319 Scientists, et al., 2010) and 328 (Davis et al., 2010). However, during IODP Expedition 332 an OsmoSampler and a microbial colonization experiment (Orcutt et al., 2010; described in more detail below) were deployed in Hole C0010A as part of the “genius plug” to detect changes in borehole fluid and microbial conditions in the screened zone that bisects a major fault in the Nankai accretionary system (Fig. F6) (Kopf et al., 2010). The sampler and microbial experiments were designed to be as compact as possible and fit within a small extension below the pressure port. This small size limits the amount of sample that can be collected.

Top of page   |    Previous   |    Next