OTC 18839
Reverse-Circulation Cementing To Seal a Tight Liner Lap
Tim Marriott, Steve Chase, and Isaac Khallad, Halliburton, and Richard Bolt* and Peter Whelan, Devon Canada Corp.
Copyright 2007, Offshore Technology Conference
This paper was prepared for presentation at the 2007 Offshore Technology Conference held in
Houston, Texas, U.S.A., 30 April–3 May 2007.
This paper was selected for presentation by an OTC Program Committee following review of
information contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Offshore Technology Conference and are subject to
correction by the author(s). The material, as presented, does not necessarily reflect any
position of the Offshore Technology Conference, its officers, or members. Papers presented at
OTC are subject to publication review by Sponsor Society Committees of the Offshore
Technology Conference. Electronic reproduction, distribution, or storage of any part of this
paper for commercial purposes without the written consent of the Offshore Technology
Conference 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 where and by whom the paper was presented. Write Librarian, OTC, P.O.
Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
Abstract
Drilling in the Arctic can present a number of challenges even
before the drilling begins. Among the challenges are logistic
difficulties, weather extremes, and environmental sensitivities.
To offset the production decline for natural gas in North
America operators must confront these challenges by
resuming exploration in the Arctic.
The target in this case study is a well in the Paktoa field.
The design challenge presented by this well was to drill the
well to total depth (TD) with the lowest number of casing
strings while avoiding remedial cementing operations. Given
that the mud-weight window predicted for this wellbore was
quite narrow, well planners determined that no fewer than five
casing strings were needed to reach TD. Multiple casing
strings can lead to tighter annuli and more challenging
cementing operations.
Using modeling software as an aid to cement design,
planners determined that the cement on the first liner, from
1,300 to 2,600 ft, could not be circulated conventionally
without breaking down the formation because of the high
equivalent circulating density (ECD). The greatest contributor
to the high ECD is the tight annulus of the liner lap. The
model parameters were reversed and the model predicted that
a reverse-circulation cementing (RCC) operation would be
successful.
RCC is a method of pumping cement down the annulus
and receiving returns inside the casing. One advantage of
reverse circulating is that the ECD is reduced and less pressure
is exerted on the formation. This will help reduce or eliminate
cement losses into weak formations.
Placing the cement down the annulus appeared to be
feasible with the computer model but to conduct the operation
in the field, other challenges were addressed. In a reverse
operation, the float valve is removed or sheared out before
pumping cement. This is not of concern with a normal casing
*
Formerly employed by Devon-Canada
string but with a liner, the drillpipe is detached and the cement
will drop until the pressure is balanced at the shoe. To prevent
the cement drop, two flapper valves were installed such that
the liner running tool and slick stinger could stab in and
circulation could be established.
The RCC operation on the liner was successfully
conducted. This paper presents a case history of the reverse-
circulation cementing operation.
Background/Introduction
The Beaufort Sea, located between Alaska and the Northwest
Territories, is the focus of recent exploration, with Devon
Canada returning to the Canadian Beaufort Sea to drill the first
offshore well started since 1991. Using 3D seismic mapping
technology, Devon has narrowed its emphasis to the Paktoa
field, an area that indicated significant gas reserves.
The drilling platform used by Devon was designed for
Beaufort Sea drilling and christened in 1982; the SDC (steel
drilling caisson) was built from the hull of a retired oil
supertanker (Fig. 1). Since 2002, the SDC drilling rig has been
in cold storage waiting for exploration to resume. The rig was
refurbished, upgraded, and towed to its present location
(during the summer months). Once on-site, it was lowered to
the shallow seafloor of the Beaufort where it waited for the
winter freeze and drilling to commence. Fig. 2 shows that
environmental impact is an important concern to operators in
the Beaufort operating area.
To drill safely to the targeted 7,700 ft, Devon Canada
engineers designed the well with five casing strings (Fig. 3).
The design resulted in tight annuli between the casings and
more challenging cement designs. The Devon engineers asked
Halliburton for a design to ensure that cement returns were
achieved.
The most challenging casing string to cement in this well
design was the 11 ¾-in. intermediate liner from 1,300 to 2,600
ft, inside a 13 ⅝ in. surface casing string. At this depth, an
unconsolidated formation with a low fracture gradient
prevented a conventional cement job to be conducted, so
planners chose to reverse circulate the cement in place. The
advantage of reverse circulation cementing for this casing
string is lower ECD, helping ensure a competent cement
sheath at the liner lap. Cement returns in the casing were not
an issue since the cement in this stage would be drilled out for
the next string.
Originally modeled with the beta version of a proprietary
Halliburton cement-job planning software package, the cement
design was then reconfirmed with a later release. The
capability to model this reverse cement job and review the
2 OTC 18839
positives and negatives of each design element helped assure
team confidence.
Drilling, completion, and production operations in the
Beaufort Sea involve meeting multiple challenges of weather,
logistics, and environmental sensitivity. This paper describes
how the combined team of Devon Canada and Halliburton
overcame an added major problem: cementing an intermediate
casing string across an unconsolidated interval having a low
fracture gradient that would not support the ECD’s of a
conventional cement job. Therefore, cement could not be
pumped down the casing; the combined team elected to apply
the cement by the RCC method. In the RCC application
method, the slurry is pumped down the annulus and brought
back a short distance up into the pipe. ECD reached by this
method are low, usually enabling weak formations to support
the weight of the cement column and the light pumping
pressures that are applied.
Reverse-Circulation Cementing
General
RCC can provide the following advantages in wells meeting
the requirements for the method:1,2
• Shorter slurry thickening times.
• Improved mud displacement.
• Reduced ECD.
• Improved compressive strength development.
• Improved safety and environmental management.
• Improved formation production due to less risk of
cement invasion into the producing zone.
• Easier cement slurry selection.
Shorter Thickening Time
The concentration of retarder used in these slurries can be
significantly reduced or even eliminated because fill slurries
will not be required to go around the shoe. If necessary,
sections of the fill-cement blends may contain accelerators
where the shoe cement and the deeper parts of the fill contain
retarders. In addition to the engineered design of cement
blends based on interval temperatures that they will be
exposed to, the minimized displacement volume can provide a
shorter pumping time. On most RCC jobs, the displacement
volumes are less than 6.9 bbl. The displacement volume is a
calculated small volume of water that is pumped behind the
fill cement to clear the surface lines and to provide enough
clear casing-to-casing length to ensure the setting of casing
slips. Rig-time saving could add up to tens of thousands of
dollars on deep-well rigs with larger, deep casing strings.
Also, when using the RCC method, the need for cement plugs
is eliminated because the cement slurries are pumped on the
backside.
Improved Displacement
Since RCC can be placed at higher rates, an inherent
improvement of mud displacement over conventional jobs
placed at lower rate (to reduce the ECD of slurry placement)
can be obtained. In addition to the above discussion of higher
flow rates, Griffith et al. indicated through large-scale tests
that mud displacement is slightly improved by the RCC
method under similar rates and conditions.3 Briggs also saw
improvement of mud displacement on multi-string
completions.4 This is because of the larger drilled particles
falling to the bottom of the hole that are lifted into the casing,
as opposed to being lifted in the annulus.
Spacer and marker design is extremely important and
entirely dependent on the drilling fluid properties. Density,
plastic viscosity, yield point, and rheology of the drilling fluid
form the basis for designing the proper spacer and marker
fluids. If an oil-based mud is used, the system wettability
determination will become very important as well. A series of
tests should be performed to help the operator select the
optimum surfactant package. A wettability meter1 can help
determine the effectiveness of a surfactant package in the
spacer and marker fluids.
Reduced ECD
ECD can be significantly reduced in RCC method. The gravity
force is working in favor of the slurry flow; therefore, the
hydraulic horsepower required to place the cement slurry is
greatly reduced, which in turn reduces the friction pressures
and yields low ECDs. Lower ECDs can provide the following
benefits:
• Higher top of cement (TOC).
• Elimination of stage tool(s).
• Higher flow rate.
Higher TOC is a desirable feature because the targeted
zone(s) are covered with cement. This advantage can also help
the abandonment procedure since more of the annulus is filled
with cement. Elimination of the stage tool is another benefit.
Higher flow rate (pump rate) is another benefit to lower
ECDs. When the critical velocity targets are approached and
met, a higher percent of mud cleaning efficiency is realized.5
This means that when pumping speed increases, the
“circulatable hole” value increases, which in turn enables a
better cement job and a better cement bond log (CBL) results
with more mud removed from the hole.6
Compressive Strength
Compressive-strength development of fill slurries has always
been an area of concern when dealing with cooler
temperatures in the uphole sections. Compressive strength of
shoe slurries when dealing with higher densities and at
moderate to high temperatures is typically not a concern.
However, in moderate to deep wells that have under-
pressured and potential upper hydrocarbon-bearing zones,
compressive strength development becomes a concern.
Shallower potential producing zones in a deep well have to be
cemented properly and according to regulatory organization
requirements.
In conventional cementing, fill cement is retarded to
provide enough pumping time to pass around the casing shoe.
However, the retarded fill slurry may be unable to meet the
regulations if it has to cover shallower zones of interest in the
upper hole sections. In the RCC process, the fill cement will
not cross the shoe bottomhole temperatures (BHTs); therefore,
a faster-setting cement can be designed.
OTC 18839
Safety and Environment
In general, RCC can be a safer method of cementing oil and
gas wells. Placement and displacement pressures are much
lower than pressures observed on conventional cement jobs.
RCC is also an environmentally friendly process. Positive
environmental impact is realized through less time and
equipment used on location.
Cement Invasion
Production rates may suffer if cement invades the producing
zone. Cement damage can occur with whole losses to the
formation or by filling the micro-fractures initiated with the
additional pressure on the formation. Whole losses will inhibit
long-term production by reducing the drainage pattern.
Cementation of microfractures will increase the near-wellbore
stresses and can complicate the completion of the reservoir.
Cement Slurry Selection
Cement slurry design parameters and considerations are
crucial to the successful completion of an RCC job. All
potential producing zones and all under-pressured zones
should be clearly identified, and cement slurry properties
should be tailored to suit the zones. Water tables should also
be of primary concern if they are present.
Some of the cement slurry properties that are engineered
include:
• Thickening time (proper pumping time).
• Compressive strength (500 psi or more in 48 hours).
• Static gel strength (crucial to pressure maintenance).
• Fluid loss (follow conventional method).
• Free water (follow conventional method).
• Rheology (follow conventional method).
RCC can allow the design engineer to select different
slurry densities with shorter pumping time as designers work
their way toward the upper sections of the annulus. This
usually means less or no retarder in cement slurries that will
cover upper sections of the hole. A good understanding of the
dynamics, fluid behavior, cement slurry properties, and
drilling fluid properties will help achieve a successful RCC
job.
Details of the Paktoa RCC Job
General
The casing design for this well was restricted by the 13 5/8-in.
BOP stack on the SDC, and the narrow mud-weight window
that was determined by the predicted pore pressures and
fracture gradients. No fewer than five strings were required to
drill this exploration well. Fig. 3 shows the planned casing
diagram for the entire casing program; however, the remainder
of the paper is devoted to the upper two sections of casing.
The first, third, and fourth strings were cemented in
conventional fashion; the second (11 ¾-in.) section was
cemented by the RCC method. Table 1 gives the predicted
fracture gradient and pore-pressure profile. Table 2 lists the
planned pumping schedule. The fifth section (lowest) is an
optional string that would be used in the event another hole
section was needed to reach the planned TD. Figs. 4 and 5
3
show the cementation of the surface casing (13 5/8 in.) and
intermediate casing (11 ¾ in.) strings.
The necessity of planning for five casing sections to reach
TD dictated the narrow annular clearances and the tight liner
lap between the surface casing and the first intermediate liner.
However with the RCC process, the pressure buildup from
forcing slurry through the small liner-lap clearance was
confined to the inside of the surface casing. Fig. 4 shows the
wellbore, casing, and liner hanger (schematically only and not
to scale). At this point, the cased well is full of drilling fluid.
Fig. 6 is a job chart showing details of the job execution. Fig.
7 compares the designed pressure placed on the formation by
conventional circulation vs. reverse circulation. The curves
show that conventional circulation would have broken the
formation down, resulting in lost circulation.
Propriatary modeling software was used in the design of
all casing and liner strings of this wellbore. The parameters
required to model these strings accurately included well bore
and liner hanger geometries, fluid properties, fracture/pore
pressures, and planned pump rates.
The Conductor Pipe
The conductor pipe is a 24-in. section that was hammer-driven
to refusal depth. The 13 5/8-in. surface casing was run and
cemented conventionally.
The Surface String
An 8 ½-in. pilot hole was drilled to accommodate LWD tools.
The open hole for this section was drilled to 14 ¾ in. to allow
installation of the 11 ¾-in. surface casing. Casing grade was
P-110, 66.3 lb/ft.
The Intermediate String
This section of the hole contained an unconsolidated
sandy/shale formation that proved to be challenging to drill. In
addition, the drilling was delayed due to unexpected warm
weather and the associated environmental regulations of
drilling in the Arctic. With the wellbore being exposed to
drilling fluid longer than anticipated, hole conditions were not
ideal and the objectives of the RCC job were re-prioritized.
The objectives were to successfully seal the liner lap,
prevent losses to the formation, and eliminate chances of
bridging off at the liner lap with the well bore rubble. The
shoe could be squeezed in the event cement did not circulate
as planned.
The liner was run in the hole to setting depth on drill
collars and drill pipe. Once on depth, the pipe was
reciprocated for the duration of the cementing job. Fig. 5
illustrates the planned RCC process. A seawater spacer was
the first fluid pumped, displacing the well fluid, followed by a
variable-viscosity spacer, then a small volume of scavenger
cement.
The cement volume was calculated before the job based on
caliper log, excess, and shoetrack volume. This cement
volume (less 63 bbl due to a valve failure) was mixed and
pumped down the surface casing-intermediate liner annulus
and displaced until the cement turned the corner and entered
the casing from the bottom.
At this time, pumping was halted, the liner running tool
was disconnected from the liner, and the 10,000-psi liner top
4 OTC 18839
packer was set, sealing the annulus at the liner top. The string
was then picked up so that the slick stinger raised up above the
spring-loaded flapper valves, allowing the flapper valves to
return to their normally closed position, acting as a float valve,
preventing the U-tubing of the cement up into the liner.
Results
The cement was allowed to cure and the drillstring was tripped
in the hole to determine the top of the cement returns inside
the liner. Scavenger cement was found approximately 200
feet above the shoe. It was determined that a valve failure
occurred during the job and prevented the planned returns
inside the casing. Nonetheless, the objectives of the
intermediate reverse circlulation cementing job were achieved:
1. The CBL showed good cement bond throughout the
length of the liner.
2. The liner lap was successfully cemented.
3. Conducting the job with RCC prevented cement
losses and damage to the formation.
Conclusions
1. Reverse circulation cementing (RCC) of a liner is an
option when wellbore conditions dictate.
2. RCC can prevent well bore damage that may inhibit
production.
3. RCC can successfully seal a tight liner lap.
4. An accurately placed indicator can identify where the
top of cement would be upon completion of the job.
It was decided that the indicator was not required for
this case study.
5. A good cement bond can be achieved with the RCC
process.
Acknowledgement
The authors thank the management of Devon Canada and
Halliburton for their permission to present this paper.
Nomenclature
TD = total depth
ECD = equivalent circulating density
RCC = reverse circulation cementing
BOP = blow out preventor
SDC = steel drilling caisson
TOC = top of cement
BHT = bottom hole temperature
LWD = logging while drilling
CBL = cement bond log
References
1. Davies, Jason et al.: “Reverse Circulation of Primary
Cementing Jobs—Evaluation and Case History,” paper
IADC/SPE 87197 presented at the 2004 IADC/SPE
Drilling Conference, Dallas, Texas, 2-4 March.
2. Spielman, Paul, Hernandez, Rafael, and Nguyen, Hao:
“Reverse Circulation of Foamed Cement in Geothermal
Wells” paper presented at the 2006 Annual Meeting,
Geothermal Resources Council (GRC), San Diego,
California 10-13 September.
3. Griffith, J.E., Nix, D.Q., and Boe, G.A.: “Reverse
Circulation of Cement on Primary Jobs Increases Cement
Column Height Across Weak Formations,” paper SPE
25440 presented at the 1993 Production Operations
Symposium, Oklahoma City, Oklahoma, 21-23 March.
4. Briggs, H.G.: “Abu Dhabi Triple Completion Wells are
the First in the Mideast,” Oil and Gas Journal, Volume 9,
No. 6, June, 1969.
5. Virginillo, B. and Tahmourpour, F.: “Reverse Circulation
of Primary Cementing Jobs – Evaluation,” paper
CADE/CAODC 2003-036 presented at the 2003 CADE
Conference, Calgary, Alberta, Canada, 20-23 October.
6. Griffith, J.E. and Ravi, K.M.: “Monitoring Circulatable
Hole with Real-Time Correction: Case Histories,” paper
SPE 29740 presented at the 1995 Production Operations
Symposium, Oklahoma City, Oklahoma, 2-4 April.
OTC 18839 5

Table 1 - Predicted Fracture Gradient & Pore Pressure Profile

True Vertical Pore Reservoir Reservoir Fracture Fracture Fracture
Depth, Pressure, Gradient, Density, Gradient, Density, Pressure,
ft psi psi/ft lb/gal psi/ft lb/gal psi
328 159 0.441 8.48 0.633 12.18 222
656 330 0.480 9.23 0.675 12.98 458
1,148 587 0.498 9.58 0.709 13.62 828
1,640 866 0.519 9.98 0.737 14.17 1,224
2,132 1,255 0.581 11.18 0.778 14.97 1,675
2,624 1,541 0.581 11.18 0.797 15.32 2,106

Table 2 - Planned Pumping Schedule

No. Description Density, Rate, Volume, Duration,
lb/gal bbl/min bbl min
1 Paktoa B-60 (PHPA Predicted) 12.50 6.3 0.0 0
2 Sea Water 8.53 5.0 12.6 2.5
3 Tuned Spacer 13.75 5.0 37.8 7.5
4 Scavenger Cement 14.17 5.0 37.8 7.5
5 0:1:0 Class G + 0.6% Halad 23 + 0.2% CFR-3 15.00 5.0 512.5 101.69
6 Sea Water 8.53 5.0 72.5 14.38
Total 106.86 133.57

Fig.1—This steel drilling cassion (SDC) was built from the hull of a retired oil supertanker. It was towed to the drilling location during the
summer months; drilling began after the ship was frozen in place. (Courtesy Sean McDermott - Ice Monitor with Horizon)
6 OTC 18839

Fig. 2—This visitor helps emphasize the importance of environmental protection during exploration and production. This picture was taken
from the drill platform. (Courtesy Don Connelly – Offshore Installation Manager, Devon-Canada)
OTC 18839 7

Fig 3—To drill safely to the targeted 7,700-ft depth, Devon Canada engineers designed the well with five casing strings. The design resulted in
tight annuli between the casings and required more challenging cement designs. This view shows the plan for all five casing strings;
however, the lowermost string is optional, for use in the event that another hole section was needed to reach the planned TD.
8 OTC 18839

Fig. 4—Wellbore, casing, and liner hanger are shown schematically and not to scale. At this point, the cased well is full of drilling fluid.
OTC 18839 9

Fig. 5—The RCC process is illustrated. The liner has been run in on drillpipe and the liner running assembly. Note that reverse circulation can
be achieved while the two flapper valves are held open by the liner running assembly. A seawater spacer was the first fluid added to the well
fluid, followed by a variable-viscosity spacer, then a small volume of scavenger cement. Pumping was halted when the planned cement
volume was pumped down the casing-wellbore annulus, turned the corner, and entered the casing through the shoe.
10 OTC 18839

Fig. 6 – This job chart captures the details of the job execution. The surface pressure during the RCC job was very low.

Fig. 7—This chart compares the designed pressure on formation of conventional circulation vs. reverse circulation. As indicated the pressure
on formation during conventional circulation was too high and losses would occur.