SPE 124409

Formation and Removal of Hydrates inside Wellhead Connectors

T. J. Akers, SPE, ExxonMobil Development Company
Copyright 2009, Society of Petroleum Engineers
This paper was prepared for presentation at the 2009 SPE Annual Technical Conference and Exhibition held in New Orleans, Louisiana, USA, 47 October 2009.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Natural gas hydrates have long plagued drilling operations in deepwater environments. Professional papers to date have
focused on hydrate formation inside the wellbore. This paper discusses the problem of hydrates that form within the BOPwellhead connector. A case history is presented of a blowout preventer (BOP) stack stuck on a subsea wellhead due to the
formation of methane gas hydrates external to the wellbore but inside of the BOP-wellhead connector. Efforts to release the
connector are discussed as well as the thermal modeling undertaken of the hydrate melting attempts. Alternative release
methods considered during the event are given along with the final resolution to the event. A discussion of the effectiveness
of both thermal and chemical methods of removing hydrates from inside the connector is presented along with expectations
of strategy effectiveness. Providences to prevent and mitigate future occurrences of hydrated BOP connectors are offered.
Introduction
Natural gas hydrates, more commonly referred to as just "hydrates" and less commonly referred to as "gas clathrates," are
solids formed of hydrocarbon gases and water. At favorable thermodynamic conditions, water traps and forms a crystalline
lattice structure around the gas. Hydrates can form with hydrocarbon gases C1-C4, H2S, and CO2. Much literature (Katz
1945; Carson and Katz 1942; Kobayashi and Katz 1949; Mehta and Sloan 1999; Sloan 1991, Sloan 2000, IADC 1998) has
been published regarding hydrates, and the reader is directed to these references for a more complete discussion with respect
to the structure and thermodynamic properties of hydrates.
Hydrates began to become a problem in drilling operations as wells were drilled in progressively deeper water depths and
where conditions are favorable for hydrate formation. Fig. 1 shows the pressure-temperature conditions that are favorable for
hydrate formation. Favorable conditions for hydrate formation are present at and near the mudline for deepwater drilling and
production operations. Barker and Gomez (1989) discuss two incidents which occurred in the United States, one in the Gulf
of Mexico and one off the U.S. West Coast, where hydrates formed inside subsea well control equipment. Lage, et al. (2002)
discuss a well control incident offshore Brazil that was also complicated by the formation of hydrates inside the well control
equipment.
The aforementioned incidents involved the formation of hydrates inside the wellbore / BOP / riser conduit to the surface.
A similar hazard exists in the potential for hydrates to form outside of the wellbore / BOP / riser conduit. In some locations,
hydrocarbons migrate along faults and through shallow sediments to the seafloor and result in hydrocarbon seeps. In some
locations, such as in the Santa Barbara Channel, offshore California, USA or in West Africa, oil sheens on the ocean surface
are caused by these seeps. In the case of gas seeps, often termed as biogenic gas, the gas is harmlessly released at the ocean
surface. However, if the gas is trapped near the seafloor and allowed to accumulate in static conditions at favorable
temperature and pressure, hydrates will form. Fig. 2 and Fig. 3 show hydrates that have formed on the outside of a BOP
stack deployed at a deepwater location. The potential can also exist for biogenic gas to enter and accumulate inside the
wellhead connector which connects and forms the pressure seal between the BOP and subsea wellhead. Gas that enters the
connector and forms hydrates can interfere with the operation of the wellhead connector. This paper discusses an incident
where hydrates did form inside a wellhead connector and impeded the release of the BOP and riser from the subsea wellhead.
Incident Initiation
The incident occurred offshore West Africa in a water depth of approximately 1500m. The drilling unit was a dynamicallypositioned semi-submersible vessel. Fig. 4 shows a generalized schematic of the well. Fig. 5 shows the temperature profile
of the water column at the wellsite and Fig. 6 shows the pressure-temperature conditions of the seawater column as compared
to the hydrate formation curves derived from Katz (1945). After running and latching the BOP and riser on the 20-in. casing,

SPE 124409

gas bubbles were observed rising around the wellhead and guide base. The gas emanated not from the wellhead itself, but
from the seafloor around the wellhead. The shallow geotechnical investigation had noted numerous shallow gas hazards in
the area of the well location. The BOP stack was equipped with a downward-facing funnel to assist in landing the BOP
stack. After temporarily abandoning the well, the next operation was recovery of the riser and BOPs. The BOPs had been
latched to the subsea wellhead for 188 days.
The wellhead connector was of an early vintage, manufactured before the drilling industry became familiar with hydrates.
Fig. 7 shows a cutaway view of the connector. The wellhead connector consists of:
1. A dog ring, which engages recesses in the wellhead.
2. 12 hydraulic pistons which drive a single cam ring.
3. A cam ring which drives in behind the dog ring and forces the dog ring to engage the wellhead.
4. An indicating rod which is attached to the cam ring and protrudes from the connector housing. It allows
measurement of the cam ring position and freedom of movement. It does not have a seal.
5. A metal-to-metal seal ring gasket which provides the wellbore pressure seal.
6. A trash-seal O-ring between the lower part of the connector and the wellhead.
The wellhead connector release was functioned. Under normal release operations, 3.5 gallons of fluid should be required,
but only 1.7 gallons was observed. The connector indicating the rod moved only 2 of the normal 4 in., confirming the
connector release was not fully functioning. An attempt to pick up on the riser / BOP was unsuccessful even with 250 kips of
overpull. The release cycle was repeated several times without success. The wellhead connector and BOPs were stuck on
the well. From the beginning of the incident, hydrates were identified as the most likely source of the connector failing to
function properly. Because of the BOP funnel and the guide base around the wellhead, the remotely-operated vehicle (ROV)
was not capable of viewing underneath the wellhead connector to check for the presence of hydrates. Even so, it was
theorized that hydrates formed inside the wellhead connector in the cavity just above the cam ring. With a hydrate plug in the
way of the cam ring reaching the open position, the cam ring could not drive away from the dog ring and allow the dog ring
to relax from the wellhead recesses. The indicating rod showed the cam ring movement never exceeded 1/4 to 1/2 open
during the entire release effort. Fig. 8 shows the suspected location of the hydrate plug inside the connector.
Release Efforts
The wellhead connector was not equipped for hydraulic access into the cam-ring cavity. This eliminated the possibility of
flushing the cavity with a solute, such as methanol or mono-ethylene glycol, to attempt to chemically decompose the hydrate
through alteration of the phase equilibria. The remaining option was to attempt to modify the environmental conditions in
which the hydrate existed. Reduction of hydrostatic pressure in the connector at the seafloor was not an option. The
remaining alternative was to attempt to heat the connector cavity and thermally decompose the hydrate. In the Gulf of
Mexico case history presented by Barker and Gomez, this approach was utilized but without success.
The drillstring was run back down to the wellhead connector and steam heated seawater (110 to 115F) was circulated
down the drillpipe and up the choke line at 240 to 420 gal/min. The connector was functioned once an hour to determine if
any progress was being made and to perhaps assist in crushing the solid hydrate. The drilling rig was also moved 50 ft to
each side of the wellhead every few hours to attempt to loosen the connector. This effort continued for 24 hrs without
success. Initial thermal modeling suggested that fluid temperature at the wellhead was not sufficient to achieve hydrate
decomposition.
Next, ambient seawater (75 to 80F) was circulated down the drillpipe and up the riser at 1200 gal/min. It was found the
higher flow rates of the lower temperature fluid resulted in higher return fluid temperatures (and temperature modeling
eventually confirmed it also resulted in higher temperatures at the wellhead connector). Functioning of the connector and
offset movement of the rig around the wellhead continued. This effort continued for 72 hrs while additional production
heaters and heat exchangers were located and mobilized to the rig. The connector still would not release. Concurrently, a
thermal model of the riser circulation and of the wellhead connector was constructed to estimate the temperature profile
across cam-ring cavity.
The drill team located and mobilized two production heaters and one hot oil treater to further heat the seawater. The heat
exchange equipment was in substandard condition and it was found the best performance that could be generated by the
equipment was ~300 gal/min of fluid at ~140F. The connector still would not release.
A plot of the drillpipe inlet and riser outlet temperatures during the heating effort is shown in Fig. 9.
Thermal Modeling of Release Efforts
Since it was expected that the melting point of hydrates at 2,100-psi hydrostatic pressure was approximately 61F, computer
thermal simulation using the operator's in-house program was used to estimate the actual temperature being delivered to the
wellhead connector and how that temperature could be maximized. Some of the principal unknowns in the simulation were
the thermal insulating properties of the riser buoyancy. While the insulating properties of new buoyancy material were
known, the riser buoyancy on the rig's riser was significantly aged and had experienced some degree of water invasion during
deployment. This riser had been deployed for 188 days. By using the available recorded drillpipe inlet and riser outlet
temperatures, the riser thermal conductivity was adjusted to match the observed field data.

SPE 124409

The computer simulation showed that the higher temperature-lower flow rate attempts using the heated seawater were
ineffectual. Circulating ambient temperature (80F) seawater at the maximum flow rate capable by the rig's mud pumps
resulted in a temperature approximately 12 degrees higher than the attempts with the surface heating equipment. Fig. 10
shows the computer predicted temperature profiles in the drilling riser.
Once the fluid temperature at the wellhead was properly modeled, the temperature gradient in the wellhead connector was
calculated using the heat diffusion, Fourier's law of heating, and Newton's law of heating. Appendix 1 provides the
calculations used in this model. The wellhead connector was modeled as a cylinder with a thermal conductivity of 26.2
BTU/hr-ft-F (steel). The model assumes that all heat transfer is one dimensional and in the radial direction. End effects
were ignored and as such, this temperature gradient is an upper bound of the actual temperature gradient. The heat transfer
from the warm circulating fluid was modeled as convection, with a convection heat transfer coefficient of 1000
BTU/hr-ft2-F. The heat transfer coefficient was calculated to be 500 BTU/hr-ft2-F assuming liquid flow in a pipe; however,
to take into account the effect of the bit jets, the value was doubled. (Note: A sensitivity was evaluated by varying the heat
transfer coefficient from 500 to 1000 BTU/hr-ft2-F, which only changed the inside surface temperature by 2F). The heat
transfer from the cool seawater was modeled as convection using empirical formulas for low Reynolds number flow past a
single cylinder.
The model results indicated the circulation of either ambient temperature or heated seawater, under conditions that could
be achieved in the field, would not increase the temperature in the wellhead connector above the hydrate formation
temperature. Fig. 11 shows the wellhead connector thermal model results for a variety of surface temperatures and flow rate.
Alternatives Considered
Several innovative alternative ideas were proposed to release the connector. A summary of these various thermal techniques
considered for hydrate removal is shown in Table 1. None of these ideas materialized, but are offered for consideration and
discussion. The inherent problem in riser circulation was the heat loss along the seawater column. The heat generated at the
surface for riser circulation is mostly lost by the time it reaches the wellhead connector and what heat remains does not
sufficiently heat through the connector. Alternatives were considered for generation of heat at the wellhead connector.
Three alternative heating methods were proposed:
1. Use of the heat of hydration of CaCl2.
2. Use of the heat of hydration of cement.
3. Use of an exothermic chemical reaction process.
Calcium chloride (CaCl2) liberates substantial energy when dissolved in water. This heat of solution is caused by the
breaking of molecular bonds as the CaCl2 goes into solution. The occurrence is well known in the oilfield as CaCl2 as a
common workover and completion fluid. Fig. 12 shows the theoretical temperature increase to be expected from the
preparation of calcium chloride solutions. If the CaCl2 could be delivered to the wellhead in flake / pellet form, the heat of
solution could be generated at the wellhead and result in a higher wellhead temperature than achieved through circulation
from the surface. An effective delivery system for approximately 5 to 6 ft3 of CaCl2 flake / pellet that would deliver the
material at the wellhead, while maintaining the material isolated from water, could not be devised, and the idea was shelved.
The generation of heat as a by-product of the hydration of cement is also familiar and measured phenomena. At the cold
temperatures of the seafloor, the reaction rate is retarded, and the resultant rate of heat release of cement is reduced. Cement
heats of hydration have been observed in the range of 35 to 55 BTU/lbm (Ravi, et al 1999). A technique to generate
downhole heat would be to fill a joint of 16-in. casing with cement and run the casing on drillpipe to a depth opposite of the
wellhead connector. Dillenbeck, et al (2002) note the amount and rate of heat liberated per unit of cement is highly
dependent upon the temperature at which the cement hydration is maintained and at very low temperatures, the amount of
temperature increase observed may be minimal. Furthermore, temperature losses would occur across the 16" casing wall and
then through the seawater between the 16-in. casing and the wellhead connector inner wall. Hence, the uncertainty of success
associated with this idea rendered it of limited merit.
A promising idea was the use of an exothermic chemical process. Similar processes have been developed and patented by
multiple companies (Mitchell, et al 1984, McSpadden, et al 1986, Ashton, et al 1986, Rocha, et al 2003). These processes
have been successful in the thermal stimulation of wells with paraffinic or waxy crude oils. The processes involve the
exothermic reaction of sodium nitrite with either ammonium nitrate or ammonium chloride and produces nitrogen gas,
sodium nitrate, water, and heat. They are reported as being capable of producing 132,500 BTU heat/bbl of solution and can
rapidly generate a temperature increase of over 200F over the ambient conditions. It was proposed that such a reaction
initiated opposite the wellhead connector would most likely generate the thermal potential to raise the temperature in the
connector above the hydrate formation temperature. Several logistical and technical issues prevented implementation of this
idea. The chemicals and personnel required in the process were not located in country and could not be mobilized into the
country in a practical period of time. Also, the exothermic nature of the reaction raises the concern of control of the
generated heat and potential damage to BOP elastomers which are located immediately above the wellhead connector or even
alteration of the metal grain structure at sufficiently high temperatures. As a footnote, some years after the wellhead
connector hydrate incident discussed in this paper, one operator has successfully used this exothermic chemical approach to
dissociate a hydrate in a subsea tree cap (Marques, et al 2002).
A few non-thermal alternatives were also proposed. If the BOP-wellhead seal made by the ring gasket could be

SPE 124409

compromised, the chance existed that a hydrate inhibitor such as methanol could be injected through the damaged seal.
Drawbacks to this plan included depth control for any cutting device to accurately locate and cut the ring gasket, damage to
the wellhead connector during the cutting operation, and the likelihood that any injected methanol would flush the cam-ring
cavity. Well control integrity would also be lost if the well had to be reentered to cut the wellhead. This idea was completely
abandoned when it was learned of a secondary gasket in place that would result in any fluid passing the ring gasket to vent to
the sea.
Final Resolution
Not all SPE papers and case histories have successful conclusions. Although some alternative techniques held attraction,
either technical or logistical issues prevented their implementation. Because the likelihood of reentering the well at a later
date was considered small, the decision was made to cut off the wellhead and retrieve riser and BOPs with the wellhead still
attached.
The surface cement plug had to be drilled out. Both 9-in. and 13-in. casing had to be cut and pulled. For each string,
the casing hanger pack-off had to be retrieved before casing was cut and pulled. A new surface cement plug was set before
the 20-in. casing was cut. Once free and the BOPs were pulled a sufficient height above the seafloor, the ROV could get a
better view of the connector / BOP funnel and observed a large amount of hydrate material underneath the BOP funnel.
While pulling the riser and BOPs with the wellhead attached, gas started decomposing and venting from the wellhead
connector when it reached 850 to 900m WD. Once on surface, the connector was function-tested and performed with no
problems.
Prevention of Reoccurrence
Since the time of the incident, the wellhead connector manufacturer has made several design upgrades to address hydrate
formation issues. The wellhead connectors are now equipped with a hydrate exclusion ring groove and elastomeric seal to
attempt to prevent potential entry of free gas into the connector cam-ring cavity (Vigeant 1998). This elastomeric gasket
seals between the wellhead connector and the subsea wellhead. A second upgrade is the addition of cavity flush ports to
allow hydraulic intervention into the cam-ring cavity.
Best practices implemented to prevent reoccurrence of a similar event include:
1. Verify installation of an elastomeric seal in hydrate exclusion ring groove and verify condition of seal if it is not
newly installed. Cases have been encountered where the ring groove is present in the connector but the
connector has been run without the elastomeric seal installed.
2. Consideration given to periodic flushing of the cavity with a hydrate inhibitor such as methanol or monoethylene glycol. Some drilling contractors have expressed reluctance to periodically flush the cam-ring cavity
reasoning that continual periodic injection may flush the cavity of grease, and any differences in objectives
should be resolved prior to implementing plan.
3. Install an elastomeric seal on the high-pressure housing that will seal and isolate the 20 x 30-in. or 20 x 36-in.
annulus.
4. If a mud mat is used, install an elastomeric gasket to seal against the structural casing that will divert gas to the
edge of the mud mat.
5. Use extension tubes on the ports of the low-pressure housing that can be closed or isolated after the casing string
attached to the high-pressure wellhead housing is cemented.
Strategy for Hydrate Dissociation with Chemicals Inside of a Wellhead Connector
Despite the preventative measures mentioned, potential still exists for hydrate formation. Multiple instances have occurred in
the field where either the hydrate exclusion seal has been omitted prior to running the BOP stack or has become dislodged
while the BOP is in the splash zone or while setting the BOP stack on the subsea wellhead. Periodic inhibitor injection may
not provide adequate and uniform inhibitor concentrations inside the cam-ring cavity. Gas diversion tubes and/or elastomeric
seals may not be effective due to currents or loss of elastomer elasticity at cold temperatures. One of the principal strategies
for hydrate remediation should a blockage form inside the cam-ring cavity has been the plan to melt the hydrate via chemical
injection. A thorough understanding of the physics and limitations of this strategy will help establish realistic expectations of
the time required to dissociate the hydrate and the likelihood of success.
The act of melting a hydrate, whether it be by thermal, pressure or chemical methods, is termed hydrate dissociation.
Hydrate dissociation is not a reaction but rather a phase change caused by a change in thermodynamic equilibrium. The
introduction of certain chemicals to mixtures of gas and water can alter the equilibria and inhibit the formation of hydrates.
These chemicals are termed thermodynamic inhibitors. Alcohols and glycols are both effective thermodynamic inhibitors for
gas hydrates. Both methanol (MeOH) and mono-ethylene glycol (MEG) are commonly used thermodynamic inhibitors.
Table 2 compares the physical properties of MeOH and MEG. Methanol is lighter than water, whereas mono-ethylene
glycol is denser than water, which often makes it the preferred choice for applications where the inhibitor is applied from
above the hydrate. Methanol's lower viscosity allows it to more readily mix in water when agitated.
The addition of thermodynamic inhibitors to the presence of an existing gas hydrate will cause the hydrate to dissociate.
Little published data exists as to the effectiveness of hydrate dissociation with thermodynamic inhibitors. A common

SPE 124409

misconception in the drilling industry is that only methanol is capable of hydrate dissociation. Glycols (such as MEG) are
also capable of hydrate dissociation and are often used in production applications. Siri, Patil and Kamath (1990) studied the
effects of both MeOH and MEG on the rate of hydrate dissociation. In the study, the inhibitor was kept in continuous contact
with hydrate material through constant injection of inhibitor into the sample chamber with the hydrate material. Their results
showed that the rate of dissociation was a function of inhibitor concentration, injection rate, pressure and temperature
differential between the inhibitor and the hydrate. Siri, Patil and Kamath summarized the inhibitor effectiveness of melting
hydrates with the following equations:
Methanol (MeOH):

Qg / (H * A) = 4.557 x 10-7 * (T)1.668

Mono-Ethylene Glycol (MEG):

Qg / (H * A) = 8.606 x 10-8 * (T)2.578

Where:
Qg = Gas production rate due to hydrate dissociation in presence of inhibitors (Gmol/s)
H = Volume fraction hydrates at the interface.
A = Cross-sectional area exposed to inhibitor (cm2).
T = True temperature driving force between inhibitor and hydrate (C).
By making the assumption that temperature driving force in the case of a hydrate in a wellhead connector is zero degrees, one
can conclude that MeOH is approximately five times more effective than MEG when in direct contact with a hydrate.
The effectiveness of thermodynamic inhibitors for hydrate dissociation is also highly dependent upon their physical
properties. The chemical must be in contact with the hydrate in order to change the equilibrium. This is complicated by the
ongoing dissociation process which liberates water that forms a film or barrier to the remaining hydrate. In order to continue
the process, the chemical must penetrate the water film be it by forced convection, diffusion, or gravity.
Li, Gjertsen and Austvik (1999, 2000) also studied hydrate dissociation with MeOH, MEG, and tri-ethylene glycol
(TEG). Their work emphasized the importance of hydrate plug properties, inhibitor properties, and the turbulence of the
inhibitor-hydrate system. In their study, Li, Gjertsen and Austvik found that MEG was a more effective inhibitor than MeOH
due to its higher specific gravity. This conclusion was reached through experiments in which the inhibitor was added to the
top of a vessel containing water and hydrate and containing water, condensate, and hydrate. When added to the vessel, the
lighter-than-water methanol remained at the top and did not contact the hydrate. When the experiment was repeated with the
heavier-than-water MEG, the MEG sank through the water and was able to contact the hydrate. Diffusion of either methanol
or glycol in the water occurred at such a slow rate that it is negligible. Li, Gjertsen and Austvik state that calculations using
Fick's second law of diffusion show that uniform diffusion of inhibitors into water can take years. Gravity and system
turbulence are the mechanisms which most affect the rate of hydrate dissociation. Inhibitor density and viscosity are the
properties that affect inhibitor efficiency of contacting the hydrate.
Valberg (2006) also studied the effectiveness of MeOH and TEG inhibitors along with new proposed chemical for
hydrate dissociation, potassium formate. The experimental setup was such that the hydrate was placed at the bottom of a
vessel filled with water, and the inhibitor chemical was added to the top. No external agitation or forced convection was
employed. Because of their heavier densities, TEG and potassium formate both outperformed MeOH in hydrate dissolution
as they were able to penetrate the overlying water column without external agitation.
The aforementioned studies uniformly confirm that effective application of chemicals to dissociate hydrates is dependent
upon effective and continuous contact of chemical to the hydrate face. As the hydrate dissociates and produces a water
barrier, the chemical must penetrate the barrier by either forced convection / agitation or by gravity. This water film
formation is the reason that hydrate dissociation by chemicals has such a varied record of successes and failures.
The cam-ring cavity of a wellhead connector is a relatively inaccessible space. The space is an annular ring of
approximately 48-in. outside diameter and 42-in. inside diameter, 4-in. height with a volume of roughly 7 gal. Access to this
cavity on some connectors is through ports (normally 4) of approximately 1-in. diameter, located in the connector cap or
bonnet. Hence, access to the cavity is from above. The cam-ring cavity access ports are plumbed as specified by the drilling
contractor (three inlets / one outlet; two inlets / two outlets; etc). For connectors plumbed with three inlets and only one
outlet, the potential exists for a dead spot in the cam-ring cavity, which may not receive adequate chemical. Connectors
plumbed with dual inlets and dual outlets are expected to have improved flow characteristics.
With access ports being located at the top of the wellhead connector, higher density fluids (MEG) will have improved
interaction with a hydrate than fluids with a density lighter than water (MeOH). In the soaking mode, where the inhibitor is
injected and left static in the cam-ring cavity, MEG will gravity-descend to the base of the cavity and continue hydrate
dissociation, whereas MeOH will remain at the top of the cavity and become ineffective once a water film has been liberated
from the hydrate. The question arises, "what is the best chemical for hydrate dissociation"? The answer is that it depends on
the situation. Methanol, when in continual contact with the hydrate face, is a superior dissociation agent to MEG. Methanol,
however, is not kept on many rigs due to the fire hazard and the cost for methanol pumping equipment. Case histories have
shown that large injection volumes of dissociation chemical may be required to melt a hydrate. If this quantity of MeOH is
not available, use of MEG should be considered as an alternative even though it has demonstrated a slower dissociation rate.

SPE 124409

MEG has some ability, through its density, to displace the water film and stay in contact with the hydrate plug longer. It
might be considered for a "soaking" attempt while sufficient MeOH is mobilized.
No comprehensive database exists within the industry of the attempts to remove hydrates from wellhead connectors
through chemical dissociation. Incidents have occurred with an array of operators and drilling contractors around the globe.
The case histories of chemical dissociation of hydrates gathered by the author showed a varying degree of difficulty in
achieving connector release. All cases have used methanol due to methanol being the more effective hydrate dissociation
agent when maintained in contact with the hydrate and due to the perception by most in the drilling industry that MEG cannot
be used for hydrate dissociation. In one case, a stuck connector released after only 4 gal of methanol injection in 15 min. In
another case, several ROV dives, multiple days, and over 700 gal of methanol were injected before the connector released.
This range of success is echoed in hydrate chemical dissociation attempts in pipelines (Sloan, 2000). Success appears to be a
function of not only the size and stability of the hydrate plug but also the ability to constantly deliver the dissociation
chemical to the face of the hydrate.
ROV Intervention of Wellhead Connector
Subsea, remotely operated vehicles (ROVs) play a critical role in diagnosing and observing a wellhead connector issue as
well as in participating in remediation efforts. In most cases, ROVs are required for chemical injection to connector cavity.
In some cases, the ROVs role may simply be to stab a chemical injection line into the connector cavity access port. The
chemical injection line may be in place as part of a spare line in the BOP hydraulic umbilical or in an Installation and
Workover Control System (IWOCS) control line. In such cases, the ROV neither delivers nor pumps the remediation fluid.
In the majority of cases, however, the ROV will be relied upon to both carry and pump the remediation fluid. Under
routine operations, ROVs are equipped with both a small chemical reservoir, with a volume capacity of several gallons, and
an injection pump. Fig. 13 shows a typical ROV injection pump and Fig. 14 shows a standard bladder-type chemical
reservoir. For hydrate remediation, the standard injection pump on the ROV is sufficient as injection rate and pressure
requirements are not critical. The injection rate of chemical must only be sufficient to flush the melt-water from the hydrate
and place fresh inhibitor at the hydrate face.
The small capacity of the chemical tank on a typical ROV renders the tank used for prevention purposes and not for
remediation applications. A well-hydrated connector may require in excess of 700 gal of chemical, based on some field case
histories, to provide enough flushing to dissociate the hydrate. ROV companies have addressed this problem with larger
tanks, up to 250 gal, that can be run with the ROV. Fig. 15 is a photograph of such a skid. Even with chemical tanks of this
size, several dives may be required to sufficiently dissociate a hydrate to allow the connector to release. These larger tanks
are not universally available and may be days or weeks away from the location should the rig be in a remote/frontier region.
Realistic expectations should be established with the duration of a chemical intervention. Once all equipment and
chemicals are in place, an intervention will still require several dives and may last several days. Twenty-four hour ROV
crews should be planned to allow crew rest and to minimize downtime. If methanol is used, proper hazards mitigations
should also be put in place.
Other Considerations
When a wellhead connector will not release from a wellhead, the impeding mechanism may not always be a gas hydrate in
the cam-ring cavity. Mechanical sticking mechanisms may be preventing release. Three such mechanisms that should be
considered are: 1) a hydraulic obstruction either not allowing opening pressure to be applied below the piston or not allowing
closing side fluid to be vented from the connector housing; 2) mechanical bind not allowing the open connector to release
from the wellhead; and 3) excessive friction between the cam ring and locking dog elements due to insufficient lubricating
grease. Many wellhead connectors are now fitted with a pilot-operated check valve (POCV) on the closing side to trap
closing pressure on the wellhead connector and prevent accidental release. Fig. 16 shows an example POCV circuit. Should
the POCV malfunction locked in the closed position or the pilot line from the unlocking side become obstructed, the closing
fluid cannot vent from the connector. In such a case, the only option becomes cutting the wellhead connector lock line
between the connector and the POCV. Before a team considers cutting the lock-side hydraulic line to the wellhead
connector, well control aspects of the operation should be carefully considered in case the well must be reentered and cement
plugs need to be drilled in order to cut off the wellhead. An alternative to consider may be to first retract the pod stabs to
ensure the locking pressure is vented upstream of the POCV.
Another technique which has proven successful is the use of unlock pressure in excess of the manufacturers
recommended operating pressure. In this case history, a connector would not release with 3,000-psi opening pressure. Even
though the connector was outside of the hydrate formation envelope, MeOH was still pumped as a precaution. The drill team
increased the unlock pressure to above 3,500 psi and the connector unlocked. This technique also has risks including
rupturing the unlock hoses upstream or downstream of the shuttle valve. Any troubleshooting plan should be carefully
planned and all parties involved before any action is taken that may not be reversible.

SPE 124409

Conclusions & Recommendations

1. Hydrates can form within the BOP-wellhead connector and interfere with proper operation of the wellhead connector.
2. The old adage, "An ounce of prevention is worth a pound of cure," rings true. Simple hydrate prevention practices can
prevent significant non-productive time and potential loss of a well.
3. The engineering department of an organization should be involved in remediation efforts early in the release process.
Thermal modeling can provide valuable insight as to the probability of success of any thermal release strategy and avoid
continued pursuit of strategies with little chance of success.
4. Pumping of heated fluids from surface to thermally decompose hydrates has been field proven to be an ineffective
technique in deepwater environments to date due to the inability to sufficiently heat the fluids at the surface and the rapid
heat loss during delivery of the fluids downhole. Heated fluids should still be considered when capable surface
equipment and pump capacity is available, and thermal modeling indicates probable success.
5. Innovative techniques to generate heat downhole hold promise to decompose hydrates. Some techniques may not be
practical to implement due to existing infrastructure that might be available in country and required up-front planning.
6. Dissociation with chemicals provides an alternative to thermal dissociation of hydrates. An understanding of the
chemical dissociation process can help provide realistic expectations for the volume of chemicals and time required to
chemically dissociate a hydrate within a wellbore connector.
Exxon Mobil Corporation has numerous affiliates, many with names that include ExxonMobil, Exxon, Esso, and Mobil. For convenience and simplicity in
this presentation, those terms and terms like corporation, company, our, we, and its are sometimes used as abbreviated references to specific affiliates or
affiliate groups. Abbreviated references describing global or regional operational organizations and global or regional business lines are also sometimes
used for convenience and simplicity.

Acknowledgments
The author wishes to thank ExxonMobil Development Company for permission to publish this paper. Gratitude is extended
to ExxonMobil Upstream Research Company which assisted in thermal modeling efforts of the case history well. Special
thanks to Dennis Hining, Miles Peroyea, Stan Christman, Nelson Tears and Carl Brooks for counsel on various aspects on
this topic. Additional thanks are offered to Drew Roberts and Art Thompson.
Nomenclature
BOP
ROV
TD
TVD
MD
BML
POCV
MEG
TEG
MeOH
IWOCS

Blowout Preventers
Remotely Operated Vehicle
Total Depth
True Vertical Depth
Measured Depth
Below Mudline
Pilot Operated Check Valve
Mono-Ethylene Glycol
Tri-Ethylene Glycol
Methanol
Installation and Workover Control System

SI Metric Conversion Factors

bbl x 1.589 873
E-01 = m3
cp x 1.0*
E-03 = Pa.s
ft x 3.048*
E-01 = m
in x 2.540*
E-02 = m
gal x 3.785 412
E-01 = m3
lbm x 4.535 924
E-01 = kg
psi x 6.894 757
E+00 = kPa
* conversion factor is exact

SPE 124409

References
Ashton, J.P., Kirspel, L.J., Nguyen, H.T. and Credeur, D.J. 1986. In-Situ Heat system Stimulates Paraffinic-Crude Producers in Gulf of
Mexico. SPEPE (May 1989) 157 - 160. SPE-15660-PA.
Barker, J.W. and Gomez, R. K. 1989. Formation of Hydrates During Deepwater Drilling Operations, JPT (March 1989), 297 - 301.
SPE-16130-PA.
Carson, D. B. and Katz, D. L 1942. Natural Gas Hydrates. 1942 Trans AIME , 146, 150-158.
Dillenbeck, R.L., Heinold, T., Rogers, M.J., Mombourquette, I.G. 2002. The Effect of Cement Heat of Hydration on the Maximum
Annular Temperature of Oil and Gas Wells. SPEDC (December 2003), 284-292. SPE-87326-PA.
International Association of Drilling Contractors, 1998. IADC Deepwater Well Control Guidelines,
Katz, D. L. 1945. Prediction of Conditions for Hydrate Formation in Natural Gases. 1945 Trans AIME 160, 140-149.
Kobayashi, R. and Katz, D. L. 1949. Methane Hydrate at High Pressure. 1949 Trans AIME 164, 66-70.
Lage, A.C.V. M., Silva, P.R.C., Gonalves, C.J. C.. Borges, F. A. T., Mansano, R.B., Imanishi, R. I. and Martins, S.B. 2002. Gas-Influx
Events in a Deep Water Exploratory Well: A Field Case History. Paper SPE 74517 presented at the 2002 IADC/SPE Drilling
Conference held in Dallas, TX, USA, 26-28 February.
Li, X., Gjertsen, L.H., and Austvik, T. 1999. Thermodynamic Inhibitors for Hydrate Plug Meltin. Annals of the New York Academy of
Sciences 912 (1), 822-831.
Li, X., Gjertsen, L.H., and Austvik, T. 2000. Hydrate Plug Properties - Formation and Removal of Plugs. Annals of the New York
Academy of Sciences 912:294-303.
Marques, L.C.C., Pedroso, C.A. and Neumann, L.F. 2002. A New Technique to Solve Gas Hydrate Problems in Subsea Christmas Trees.
SPEPF (November 2004), 253-258. SPE-77572-PA.
McSpadden, H.W., Tyler, M.L. and Velasco, T.T. 1986. In-situ Heat and Paraffin Inhibitor Combination Prove cost-Effective in NPR#3,
Casper, Wyoming. Paper SPE 15098 presented at the 56th SPE California Regional Meeting held in Oakland, California U.S.A. 2 4
April.
Mehta, A. P. and Sloan, E.D. 1999. Structure H Hydrates: Implications for the Petroleum Industry. SPEJ 4 (1), 3 8. SPE-53450-PA.
Mitchell, T.I., Donovan, S.C., Collesi, J.B. 1984. Field Application of Chemical Heat and Nitrogen Generating System. Paper SPE 12776
presented at the SPE California Regional Meeting held in Long Beach, California U.S.A., 11 - 13 April.
Ravi, K., Biezen, E, Lightford, S.C., Hibbert, A., and Greaves, C. 1999: Deepwater Cementing Challenges. Paper SPE 56534 presented at
the SPE Annual Technical Conference and Exhibition held in Houston, TX USA 3-6 October.
Rocha, N.O., Khalil, C. N., Leite, L.C.F. and Bastos, R.M. 2003. A Thermochemical Process for Wax Damage Removal. Paper SPE
80266 presented at the SPE International Symposium on Oilfield Chemistry held in Houston, TX USA 5-7 February.
Sira, J.H., Patil, S.L., and Kamath, V.A. 1990. Study of Hydrate Dissociation by Methanol and Glycol Injection. Paper SPE 20770
presented at the 65th Annual Technical Conference and Exhibition of the Society of Petroleum Engineers held in New Orleans, LA
USA 23-26 September.
Sloan Jr., E.D. 1991. Natural Gas Hydrates. JPT (December 1991), 1414 - 1417. SPE-23562-PA.
Sloan Jr., E.D. 2000. Hydrate Engineering, Monograph Series, SPE, Richardson, Texas 21.
Valberg, T. 2006. Efficiency of Thermodynamic Inhibitors for Melting Gas Hydrates. Masters Thesis, Norweigian University of Science
and Technology, Trondheim, Norway.
Vigeant, S.P. 1998. Deepwater Driven Advancements in Well Control Equipment and Systems. Paper IADC/SPE 39298 presented at the
IADC/SPE Drilling Conference held in Dallas, TX USA, 3-6 March.

SPE 124409

Appendix 1
Wellhead Connector Thermal Model

r2

r1

T1, h1

Ts1

Ts2

r
T(r)
L
T2, h2

T(r) = Temperature at any radial distance located in the wellhead connector.

Ts1 = Temperature at the ID surface of the wellhead connector.
Ts2 = Temperature at the OD surface of the wellhead connector.
T1 = Temperature of the warmed circulating fluid in riser at connector.
T2 = Temperature of seawater in ocean at mudline.
h1 = Convection heat transfer coefficient of the warm circulating fluid.
h2 = Convection heat transfer coefficient of the cold ocean seawater.
k = Thermal Conductivity of the wellhead connector.
r1 = Inside radius of the wellhead connector.
r2 = Outside Diameter of the wellhead connector.
L = Length of the connector.
As a first (simple) approximation, assume the temperature gradient in the wellhead connector is 1-dimensional and varies
only in the radial direction (end effects neglected, which will provide an optimistic result).
For steady state conditions with no heat generation, the appropriate form of the heat transfer equation is:
1/r d/dr ( kr dT/dr) = 0

(1)

The solution to this differential equation is:

T(r) = [(Ts1 - Ts2 ) / ln(r1/ r2) ] * ln(r/ r2) + Ts2

(2)

Ts1 and Ts2 can be calculated using an electrical analogy. The thermal resistance due to convection in the hot fluid,
conduction in the connector and convection in the cold seawater outside the connector can be written as:
R total = [ 1/ (2 * r1 h1 L)] + [ln(r1/ r2) / (2 * k L)] + [ 1/ (2 * r2 h2 L)]

(3)

Assuming (T1 - T2 ) is the thermal potential (voltage), the heat transfer rate (current) can be found:
Q = heat transfer rate = (T1 - T2 ) / {[ 1/ (2 * r1 h1 L)] + [ln(r1/ r2) / (2 * k L)]
+ [ 1/ (2 * r2 h2 L)]}

(4)

All of the variables on the right side of the equation (4) can be calculated with emperical formulas or with the results from the
thermal model of riser circulation. Since Q can be calculated, Ts1 and Ts2 can be found (once again from the electrical
analogy):
(5) and
Q = (T1 - Ts1) / (1/2 * r1 h1 L)
Q = (Ts2 - T2 -) / (1/2 * r2 h2 L)

(6)

Ts1 and Ts2 can be found by plugging equations (5) and (6) into equation (4). Then T(r) can be determined.
Ts1 = T1 - (T1 - T2 ) / {[ 1 + (r1 h1 ln(r2/ r1)/ k ) + r1 h1/ r2 h2)] }
Ts2 = T1 + (T1 - T2 ) / {[ 1 + (r2 h2 ln(r2/ r1)/ k ) + r2 h2/ r1 h1)] }

(7)
(8)

10

SPE 124409

Conditions Favorable for Natural Gas Hydrate

Formation (from Katz 1945)
10000

Methane
Gas Grav=0.6
Gas Grav=1.0

Pressure - PSIA

Favorable Conditions for

Hydrates to form

1000
Unfavorable Conditions for
Hydrates to form

100
20

40

60

80

100

Temperature- Degrees F

Fig. 1Conditions conducive to hydrate formation.

Fig. 2Hydrates below BOP funnel.

WD ~ 1500m
30 Casing with
LP Wellhead Housing

20 Casing with
HP Wellhead Housing

16 Liner

13-3/8 Casing

9-5/8 Casing

7 Liner
5-7/8 Open Hole

Fig. 3Hydrates accumulated on outside of a BOP Stack.

Fig. 4Casing /Cementing Schematic

SPE 124409

11

Natural Gas Hydrate Formation Conditions for

Subject Well

Temperature- deg. F
0

10

20

30

40

50

60

70

10000

80

Methane

Gas Grav=0.6
Gas Grav=1.0
200

Favorable Conditions for

Hydrates to form

Seawater Temp-Pressure Profile

400

Pressure - PSIA

Hydrostatic Head at Seabed

Depth (m)

600

800

Favorable Conditions
for Methane Hydrates
to form below 800m
of water depth
in subject well

1000
Unfavorable Conditions for
Hydrates to form

1000

1200

1400

SEAFLOOR
100
20

1600

40

60

80

100

Temperature- Degrees F

Fig. 5Seawater Temperature Profile at Wellsite.

Fig. 6Region favorable for Hydrate Formation at Wellsite

Location of
suspected
hydrates
Probable
pathway of
gas migration
into wellhead
connector

Fig. 7Cut-Away View of Wellhead Connector.

Fig. 8Probable location of Hydrates in Wellhead Connector.

12

SPE 124409

Attempt to Release Wellhead Connector Surface Temperature Data

180

Heated Water with 2 production heaters and 1

hot oil treater.

Opened BOPs and started circulating at

high rates with ambient seawater.

160

140

100

80

60

Temperature Down DP
Return Fluid Temperature

40

20

0
0

20

40

60

80

100

120

Cum. Hrs. since Start of Circ.

Fig. 9Surface Temperature Data from Attempts to Thermally Decompose Hydrates.
Temperature (Deg F)
0

20

40

60

80

100

120

140

160

Drillpipe

200

DP x Riser
Annulus

400

600

Seawater
Depth (m RKB)

Degrees F.

120

800

1000

1200

Seawater Temperature
Inside DP - 350 GPM - 110 Deg F
Riser Return - 350 GPM - 110 Deg F
Inside DP - 1200 GPM - 80 Deg F
Riser Return - 1200 GPM - 80 Deg F

1400

Inside DP - 300 GPM - 140 Deg F

Riser Return - 300 GPM - 140 Deg F

1600

Fig. 10Thermal Model Results of Riser Circulation.

140

13

100

100

90

90

80

80

Cam Ring OD Temperature (Deg. F)

Cam Ring ID Temperature (Deg. F)

SPE 124409

70
60
50
40

Flow Rate 80 deg. F

Flow Rate 100 deg. F
Flow Rate 150 deg. F
Flow Rate 200 deg. F

30
20
10

Flow Rate 80 deg. F

Flow Rate 100 deg. F
Flow Rate 150 deg. F
Flow Rate 200 deg. F

70
60
50
40
30
20
10

200

400

600

800

1000

1200

1400

200

400

Flow Rate (GPM)

600

Fig. 11Model Results for Temperatures across Wellhead Connector.

Theoretical Temperature Increase from Preparing Aqueous

Solutions of Calcium Chloride

180

160
Anhydrous
Dowflake

Temperature Increase, Degrees F

140

120

100

80

60

40

20

0
0

10

20

30

800

Flow Rate (GPM)

40

50

Weight Percent Calcium Chloride

Fig. 12Temperature Increase from Preparation

of Aqueous Solutions of Calcium Chloride.

Fig. 13ROV Injection Pump.

1000

1200

1400

14

SPE 124409

Fig. 14ROV Bladder-type chemical tank.

Fig. 15ROV Hydrate Remediation Skid with 250/gal

chemical capacity.

ROV
Intervention
- Lock
Pilot Line

Wellhead
Connector

Optional
ROV
Intervention
Pilot
Operated
Check Valve

Yellow
Blue
Pod
Pod
Secondary Unlock
Yellow
Pod

Blue
Pod
Unlock

Yellow
Pod

Blue
Pod
Lock

Fig. 16Example Pilot Operated Check Valve Circuit on Well Connector Lock Function.

SPE 124409

15

Table 1: Summary of Heat Generation Ideas Considered for Wellhead Connector Release
Idea

Method of
Heat
Generation

Hot Water
CaCl2
Cement
Exothermic
Chemical Reaction

Intrinsic Heat
applied at
surface
Heat of
Hydration
Heat of
Reaction
Heat of
Reaction

Heat
Generation
Potential
(BTU / lbm)
~ 1 BTU/(lbmF)

Issues

Heat loss during delivery of water to connector

40-70

Delivery of pelletized CaCl2 to connector

35-55

Delivery of mixed cement to the wellhead.

Low heat of reaction at low temperatures.
Availability of chemicals in remote areas.
Need for trained specialist who may not be available
in remote areas.
Criticality of job design; prevention of runaway
reaction.
Potential damage to BOP elastomers and steel if too
much heat generated.

132,000

Table 2: Select Physical Properties of Methanol and Mono-Ethylene Glycol

Methanol
Alcohol
CH3OH

Mono-Ethylene Glycol
Glycol
C2H4(OH)2

Molecular Weight

32.04 g/mol

68.068 g/mol

Density (g/cc) @ 20C

0.7915 g/cm

1.1135 g/cm

Viscosity (cp) @ 20C

Freezing Point, C

0.55 centipoise
97 C (176 K)

21 centipoise

64.7 C (337.8 K)

197.3 C (470 K)
111 C (closed cup)
Fully miscible
2.20D

Family
Chemical Formula
Representation

Boiling Point, C
Flash Point, C
Solubility in Water
Dipole Moment
NFPA 704 rating

11 deg C

Fully miscible
1.69 D

12.9 C (260 K)

3
1

1
0

Flammible
Toxic when
ingested

Toxic when
ingested