IADC/SPE-199663-MS

A Comprehensive Test Pad for Bradenhead Pressure Mitigation in the DJ

Basin

Ashley Belvin and Brian Cocchiere, SRC Energy; Meghan Jacobs, Halliburton; Adonis Ichim, Tenaris

Copyright 2020, IADC/SPE International Drilling Conference and Exhibition

This paper was prepared for presentation at the IADC/SPE International Drilling Conference and Exhibition held in Galveston, Texas, 3–5 March 2020.

This paper was selected for presentation by an IADC/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 International Association of Drilling Contractors or the Society of Petroleum Engineers and are subject to correction
by the author(s). The material does not necessarily reflect any position of the International Association of Drilling Contractors or 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 International Association of Drilling
Contractors or 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 IADC/SPE copyright.

Abstract
Bradenhead pressure, or sustained casing pressure, is pressure build up in the annular space between the
surface casing and the next smaller diameter casing string within the wellhead. The objective of the test pad
was to determine if increasing the physical flexibility of cement and rotating the casing string to increase
displacement efficiency would help improve the cement bond to casing, decrease cement channeling, and
help eliminate future bradenhead pressure accumulation.
A twelve well pad housed three different cement slurries: four latex-type jobs, four resin jobs, and four
foam jobs. A rotating cement head was used to enable mud circulation, dropping plugs, and rotating the
string of casing during the cementing process on two of the four wells of each slurry type. For the production
casing string, a threaded and coupled connection with a wedge thread profile was used to withstand the high
torque experienced during rotation operations.
Results were determined by evaluating pre- and post-stimulation logs along with continued bradenhead
pressure monitoring. According to the outcomes from this test pad, recommendations were made on
cementing practices within the Denver-Julesburg (DJ) Basin, based on regional gas-oil ratios (GOR).
This interdisciplinary work determined whether the deployment of advanced cement slurries and casing
rotation would help eliminate a potential health, safety, and environment (HSE) risk and help improve well
integrity as related to bradenhead pressure.

Introduction
The Colorado Oil and Gas Conservation Commission (COGCC) identifies bradenhead pressure as a
potential HSE risk. Based on this classification, testing and reporting requirements with suggested
procedures and documentation submittal are required. Recently, the COGCC issued an operator guidance
notice to promote the consistency of monitoring and reporting, including mitigation strategies in the latest
notice to operators (COGCC 2019).
Mitigating sustained casing pressure once the well is in the production phase of its life is a retrograde
solution that may not be achievable during the first attempt. Financial ramifications to the operator can be
severe to continue to attempt correction or even abandonment. The goal of the test pad was to determine
2 IADC/SPE-199663-MS

if bradenhead pressure could be mitigated by rotating the production casing string and/or implementing
advanced cement slurries that not only focused on annular coverage and zonal isolation by helping eliminate
cement channeling, but also the flexibility and shear bond to casing to help prevent micro-annulus formation.
Each scenario was compared to the traditional cement program (or the slurry normally pumped without
rotation on jobs other than this test pad) to determine effectiveness.

Theory and Definitions

Sustained casing pressure, referred to regionally as bradenhead pressure, is pressure buildup in the annular
space between the surface casing and the next smaller diameter casing string within the wellhead (Nelson
and Guillot 2006). In the DJ Basin, a monobore design has been adopted almost exclusively, meaning there
are no intermediate casing strings installed between the surface and production strings. Fig. 1 shows this
configuration and where bradenhead pressure is measured.

Figure 1—Bradenhead valve on a monobore configuration. (not to scale, vertical representation)

These measurements are recorded by a pressure gauge before stimulation within 60 days of pad drilling
completion, within 60 days of first sales, and annually for the remainder of the life of each well. There is a
noticeable increase in bradenhead pressure occurrence from the pre-stimulation test to the post-stimulation
test, and herein lies the theory on the cause.
Through general research, it has been concluded that sustained casing pressure can occur when a micro-
annulus is formed, creating a pathway for fluid travel. The fluid consists of either gas or liquid hydrocarbons,
water, or both. Testing confirmed that the hydrocarbons are from the targeted hydrocarbon-bearing zones
within the Niobrara Formation. It is yet to be determined if the fluid migration via the micro-annulus is
directly from the Niobrara, or if there are permeable and/or depleted zones up-hole being charged with fluid
during operations. As for the increased occurrences post-stimulation, void space in the annulus might be
created or exacerbated by the cyclical pressure loading experienced during wellbore stimulation. To help
eliminate the formation of a travelable pathway, the test pad was planned to focus on cement flexibility and
annular bond during cyclical compression.
IADC/SPE-199663-MS 3

In addition to the pre-stimulation to post-stimulation trend, there appears to be a regional pattern of

instances based on GOR. In Fig. 2, pressure recordings are shown both before and after stimulation over a
GOR map. The size of each white bubble is directly proportional to the amount of pressure recorded.

Figure 2—Pre and post stimulation bradenhead occurrences (white) underlain by a regional GOR map.

Note that all data have been recorded from horizontal wells and only recently, as newly required by the
COGCC. The density of recorded pressure tests has increased vastly throughout the previous year. Even so,
there is a pre-stimulation trend of higher pressure recorded on wells closer to the center of the map, or the
higher GOR area. At the center, where there are seemingly no occurrences, little horizontal development
has occurred because of existing urban development. Additionally, it is apparent that, post-stimulation, the
density of instances increases as they near the higher GOR area, suggesting that current cement programs
might require some alteration as they near this region.

Application of Equipment and Processes

The test location was chosen based on its proximity to the higher GOR region of the Wattenberg Field; for
reference, it is located on the northwest edge of the reddest region in Figure 2. The test pad was comprised
of twelve wells where three different cement slurries were implemented: four latex-type jobs, four resin
jobs, and four foam jobs. Each solution was designed to stop gas or liquid migration and help improve
mechanical integrity. The pad layout can be seen in Fig. 3.
4 IADC/SPE-199663-MS

Figure 3—Pad layout by solution; each rectangle represents one well location.

The production casing string was rotated during cementing in two of the four wells of each slurry type
using a rotating cement head. To accommodate the high torque while rotating, a proprietary threaded and
coupled connection with a wedge thread profile was also used.
Regionally, the surface casing setting depth varies from 1,000 to 3,000 ft, depending on the local aquifer
depth and operator preference. The production casing setting depth varies from 6,000 to 8,000 ft true vertical
depth (TVD) and 11,000 to 20,000 ft measured depth (MD). For this test pad, the surface casing was set at
approximately 1,800 ft for all wells, and the production casing was set at approximately 7,300 ft TVD and
17,650 ft MD. Both the surface and production casing strings were cemented to surface.
For the duration of this study, the slurry normally pumped on jobs other than this test pad was referred
to as the traditional cement program. This program included a top plug, an 11.5 pound per gallon (ppg)
spacer pumped ahead of the lead cement, a 13.2 ppg lead cement from landing point (the measured depth
where the well entered the production horizon and is approximately horizontal) to surface, a 13.2 ppg tail
cement from total depth (TD) to landing point, a bottom plug, and a fresh water displacement. To accurately
compare the traditional cement program with the three solutions on this test pad, the spacer volume and
weight, circulation time before and after the production casing run, and mud properties of a 10 pgg oil-
based mud system were all kept the same. As for centralization, solid-body bow-spring centralizers were
used throughout the pad for optimal standoff.
The three solutions were designed to help prevent both pre-stimulation and post-stimulation fluid
migration from occurring. Pre-stimulation fluid migration can occur if the cement loses volume allowing
formation fluids to enter the wellbore. For fluid to flow, there should be flow potential, which occurs
when the pore pressure of the formation is greater than the hydrostatic pressure of the cement in the well
bore. Additionally, gas can migrate into the wellbore during the phases of the critical static gel strength
(CSGS) hierarchy. CSGS is reached when the hydrostatic pressure of the cement and well fluids are equal
to formation pressure. The cement slurry then enters its CSGS period, or transition time, until it reaches 500
lbf/100 ft2. According to the American Petroleum Institute (API), a gel strength of 500 lbf/100 ft2 indicates a
relatively low probability of fluid influx during cement hydration. Therefore, the transition time is the most
vulnerable time for gas to migrate through cement because the further gelation is creating an underbalanced
condition within the wellbore. To help prevent fluid migration, the cement closest to the surface of the
wellbore requires hydrostatic pressure exertion for a longer duration against the formation before it starts to
set or reach its CSGS. Once the transition period of the lower cement ends, the lead cement still transmits
hydrostatic pressure, known as zero gel time, and the gas is then theoretically stopped (API 2010).
IADC/SPE-199663-MS 5

Post-stimulation fluid migration is induced through three primary modes of failure that can occur in the
cement, as well as the inherent permeability the cement has when set. These failures include radial cracking
(tensile failure), deformation (compressive failure), and debonding. The swelling and contracting of the
casing causes axial and radial strain on the cement, causing the cement to crack or shrink away from the
casing. The cyclic loading on the casing and cement during hydraulic stimulation and production, causes
poor or a complete lack of interfacial bonding. This can enable the formation of a micro annulus thereby
creating a path for the fluids to migrate to surface. A highly permeable cement can also allow fluids to
migrate.
Cement characteristics necessary to decrease the likelihood of fluid migrating are clearly understood.
However, in the DJ Basin, the cause of fluid migration is speculative. API Standard 65-2 provides only
generalized, qualitative recommendations for related, testable parameters and solutions on how to manage
fluid influx once it occurs. When designing and selecting the three cementing solutions for the test pad,
special attention was paid to the parameters listed in Table 1. Each recommendation relates to both pre and
post-stimulation fluid migration patterns being the cause.

Table 1—Description of cement mechanical properties.

Parameter Definition Recommendation Relation

Shrinkage Cement matrix shrinking in Negative shrinkage - Pre- and post-stim

size internally expansion

Transition Time Time from CSGS to Under 45 minutes Pre-stim

500lb/100ft2

Young's Modulus Longitudinal stress over strain Low, withstand changes in Post-stim
length

Poisson's Ratio Ratio of transverse strain to High; elasticity Post-stim

axial strain

Tensile Strength Resistance to break under High Post-stim

tension

Shear Bond Adhesion to casing and High Post-stim

formation

Fluid Loss Dehydration of cement to Low; dehydration leads to Pre-stim

formation increased permeability

The design of each solution was planned and executed to allow for direct comparison. In the DJ Basin,
target zones for horizontal drilling include the three tiers of the Niobrara and the Codell, between 6,700
and 7,200 ft. The shallowest producing zone is the Sussex formation at approximately 4,500 ft TVD. To
help ensure coverage of all producing zones and provide complete isolation of said zones, it was designed
to place each solution from 2,500 to 7,400 ft MD, or through the near vertical and curve portions of the
wellbore. The one exception was the resin slurry which was a 100-bbl pill placed approximately from 2,500
to 5,000 ft MD. This slurry had to be batch mixed in 50-bbl increments in a 100-bbl capacity mixer. The
placement of each of these slurries aimed at providing a barrier for any gas that would migrate upward in
the annulus from any of these zones. The cement program was consistent, including the traditional cement
program spacer circulated to surface, a cap cement placed from 2,500 ft to surface, and the tail slurry placed
from the landing point to TD. The tail slurry was kept the same throughout the test pad and the traditional
cement program.

Latex-Type Slurry
The latex-type slurry was a traditional 13.2 ppg slurry with a liquid styrene-butadiene copolymer additive
resembling latex characteristics. This latex-type additive was designed to decrease the overall fluid lost,
6 IADC/SPE-199663-MS

slurry permeability, and transition time while increasing mechanical properties by lowering Young's
modulus, increasing tensile strength, and increasing Poisson's ratio. The latex-type additive, stabilized
with surfactants, was an emulsion that hydrated the cement, leaving behind a small-polymerized styrene-
butadiene polymer that decreased existing pore space within the slurry; in other words, it decreased the
surface area over which the gas could migrate. Comparison of these characteristics is shown in Table 2
below.

Table 2—Mechanical properties of latex-type additive and resin

compared to traditional cement (cured for 7 days at 150°F and 2,025 psi).

Young's Compressive
Poisson's Ratio (-) Tensile Strength (psi)
Modulus (Mpsi) Strength (psi)

Traditional 0.41 0.15 1236 189

Latex-Type 0.66 0.40 2380 209

Resin 0.16 0.42 2482 257

Resin Slurry
The resin slurry was a mixture of 90% cement at 13.8 ppg and 10% resin at 9.1 ppg, bringing the
combination to a 13.2 ppg cement slurry. The addition of resin was intended to decrease fluid loss, transition
time, permeability, and Young's Modulus while increasing Poisson's ratio, tensile strength, and shear bond
strength. The addition of resin exhibited a decrease in the overall permeability by filling pore spaces that,
in turn, decreased the likelihood of gas migration. Additionally, it increased the strength of the bond to
casing and formation while also being elastic in nature to maintain integrity during stimulation cycles. A
comparison of these characteristics can also be seen in Table 2.

Foam Slurry
The foam slurry had a 15.4% quality with nitrogen for a final density of 13.2 ppg. This slurry was designed
to decrease the likelihood of fluid intake and permeability while providing optimum hole cleaning. The
likelihood of fluid intake was decreased because the nitrogen expanded the cement to fill any lost volume.
The cement had a decreased permeability because the nitrogen bubbles did not interconnect to form a
channel because the quality was under 35%. Another benefit of keeping the quality at approximately 16%
was that the cement maintained 50% of its overall compressive strength, while decreasing its Young's
modulus by 50% (Kutchko et. al. 2014). The nitrified foam slurry was designed to provide more efficient
hole cleaning, because of the higher viscosity compared to traditional cement slurries. In theory, the pressure
of the nitrogen bubbles in the cement helped push the left-over mud out of the hole to bond to a cleaner
formation and casing (Ahmed et. al. 2008).

Rotation
For cementing in general, there are industry-accepted best practices that can provide a solution to gas
migration, in addition to slurry design. These recommendations include centralizing the casing to ensure
the highest standoff possible, circulating well fluid before the cement job for two complete hole volumes,
pumping the cement job as fast as the pumps and formation allow, and using enough spacer to see at least a
10-minute contact time with 1,000 feet of annular fill. Another recommendation is to rotate and reciprocate
the casing during the cement job. As previously stated, the production casing string was rotated within
half of the wells during the entire cement job. Rotation was intended to increase displacement efficiency
of unwanted wellbore fluids that were static, which, in turn, could help eliminate void spaces and more
accurately place cement where designed, as represented in the modeling in Fig. 4.
IADC/SPE-199663-MS 7

Figure 4—Displacement efficiency modeling representing a spacer, cap, lead then tail pump order.

The y-axis represents the total measured depth of the well, and the x-axis provides a full 360-degree view
of the annular space. The top of each fluid is more defined with rotation, and there does not appear to be
any evident channeling, according to the modeling software.
For this test pad, the rotation of the production casing string (at 30 RPM) started as spacer was pumped.
The rotation stopped at approximately 20 bbl from the end of displacement to work the remaining torque
out of the string. A rotating cement head rated to 30,000 ft-lbf of torque, which dropped the bottom and top
plug while pumping using a remote-controlled device, enabled the rotation of the casing. An example of
one of the jobs’ torque and rotary speeds can be seen in Fig. 5.

Figure 5—Torque and rotary speed during cementing operations.

8 IADC/SPE-199663-MS

To enable rotation at the specified 30 RM and under the given set of conditions, a proprietary threaded
and coupled wedge-type connection was used in the 5 ½″ 20 lb/ft P110-IC production casing string (Fig.
6). This connection has an operating torque of 32,000 ft-lbs. and a yield torque of 38,000 ft-lbf, provided
by two primary features: the dovetail thread profile and the pin-to-pin back up mechanism.

Figure 6—Threaded and coupled connection.

Before field deployment, this connection was subjected to a thorough design process, including
verification, from the drawing board to the rig. Finite element analysis was used to asses possible design
alternatives and optimize specific features of the connection. Connection make-up (torque-turn), pin-to-pin
contact, hoop stresses, sealability, and structural capacity were evaluated. The operating and yield torque
were confirmed using make-and-break tests with drift verification.
Finally, the connection design was verified using a full-scale testing protocol representing one possible
demanding scenario for shales that encompassed the casing installation, stimulation, and production phases.
This testing included make-up, fatigue testing, and sealability testing (Blanc et al. 2019).
For the test pad, the average running speed of the production casing string was 24 joints/hr, with no
rejected make-ups. The maximum torque experienced during the rotating while cementing operation was
26,000 ft-lbf at 30 RPM. The maximum dogleg severity observed was 16°/100 ft.

Data and Results

Results were determined by evaluating post-processed, ultrasonic logs both pre and post stimulation while
monitoring bradenhead pressure continuously. The logs were run solely under hydrostatic pressure with
zero applied pressure to best mimic conditions later in the life of the well.
Fig. 7 and 8 show the pre-stimulation logs in the near vertical section. The logs were visually evaluated
to determine the differences between the three slurry types. In the right-most track of each log, orange
represents bonded pipe and light blue represents void space or free pipe.
IADC/SPE-199663-MS 9

Figure 7—Vertical section logs – without rotation – from surface to 6400 ft MD.

Figure 8—Vertical section logs – with rotation – from surface to 6400 ft MD.
10 IADC/SPE-199663-MS

As seen in Fig. 8, the slurry tops were much more defined, suggesting improved displacement efficiency
with rotation. Without rotation, the resin slurry seems to have channeled but appears exceptional once
rotation was used. Again, this was a 100-bbl pill; therefore, placement is from approximately 2,500 to 5,000
ft MD. The latex slurry had minimal void space before rotation and even less once it was rotated. The foam
slurry had significant void space both before and after rotation.
Pre-stimulation logs were also run in the lateral on one well of each job type. Fig. 9 shows these logs.

Figure 9—Horizontal section logs from 6600 ft MD to TD.

Overall, rotation helped improve cement coverage throughout the wellbore. Upon initial testing, there
were no bradenhead pressure occurrences on the test pad pre-stimulation.
The lateral portion of each well on this test pad was stimulated after the initial logs were acquired. Once
this phase was complete, logging was performed a second time on nine of the twelve near verticals for
IADC/SPE-199663-MS 11

comparison to the pre-stimulation logs. Fig. 10 demonstrates the most representative findings from this
assessment; each "A" labeled log is a pre-simulation log. and each "B" labeled log is a post-stimulation log.

Figure 10—Pre-Stimulation and post-stimulation logs from surface to 6000 ft MD.

The center tracks on each log represent the pre vs. post-stimulation cement coverage. In the right-most
tracks of each log, orange represents bonded pipe and light blue represents void space or free pipe both
pre and post-stimulation. Based on this visual comparison, any existing void space was exacerbated after
the cyclical loading of hydraulic stimulation. Void space was much less prevalent when casing was rotated.
Improved placement of each cement slurry was observed using rotation, which was in accordance with
the initial modeling. From these findings, it was determined that the possible development of bradenhead
pressure later in the life of a well was likely due to increasing the void space and free pipe areas within
the cement coverage. With adequate coverage pre-stimulation, there was little effect observed in the post-
stimulation logs. Overall, rotation minimized the amount of pre-stimulation void space. This effect was
even more prevalent with a resin additive mixed in the slurry.
To determine if the traditional cement program was still a viable option, an additional ultrasonic log was
run after this test pad was complete. The production casing was not rotated during the cement job on this
well. Fig. 11 is a clip from 2,500 to 4,500 ft MD of the base case log; this was used as the control point for
12 IADC/SPE-199663-MS

comparison during this test. Based on the existing coverage with minimal void space, the traditional cement
program could still be a viable option with rotation.

Figure 11—Traditional cement program vertical section log – without rotation – from 2,500 to 4,500 ft MD. (right)

Conclusions
This work discussed possible methods of helping eliminate sustained casing pressure both pre and post
stimulation. Three cement types were pumped, with and without rotating the production casing string,
and compared to the traditional cement program. Based on cement bond logs before and after hydraulic
stimulation and continuous bradenhead pressure monitoring, recommendations were made for solutions fit
for regional GOR. Rotation during cementing operations improved the cement placement and displacement
efficiency in every case, showing defined cement tops and less void space or free pipe. The resin slurry,
with rotation, showed the most consistent coverage. The traditional cement program is a viable option, with
rotation, based on the coverage observed in the pre-stimulation log and will be implemented in areas with
lower GOR and lower risk of bradenhead pressure development. In areas with higher GOR and higher risk
of bradenhead pressure development, the resin slurry with rotation will be executed for the most effective
cement coverage. This interdisciplinary work provided conclusive options when addressing how hydraulic
fracturing and GOR may affect the possibility of bradenhead pressure occurrences.

Acknowledgements
We express special thanks to the Drilling and Completions teams at SRC Energy, and gratitude to Kenneth
Gathings and Chris Rodriguez for coordinating the operations needed for this project.
We acknowledge the Cement and Wireline and Perforating services of Halliburton, with special attention
to Christopher Chavez, Ryan Keeran, Dale Hopwood, and Paul Osbourne.
The support of Tenaris’ management, Unconventional Technologies team, and Denver office is much
appreciated. A special thank you to Mollye Hinds and her creativity in the collaborative efforts in this paper.
In her name, we were able to make this a reality.

References
Ahmed, R. M., Takach, N., Khan, U., Taotaou, S., James, S., Saasen, A., Godoy, R., (2008, December 9). Rheology of
foamed cement. Society of Petroleum Engineers. doi:10.1016
American Petroleum Institute, API Standard 65-2, 5-2, Isolating Potential Flow Zones During Well Construction, Second
Edition/December 2010
Blanc, C., Lewis, J., Ichim, A., Mutis, D., Zestran, A. A., Lucca, C. I., Perello, L.A., New OCTG Developments to
Overcome Challenges in Unconventional Plays. Society of Petroleum Engineers. doi:10.2118/196596-MS
Colorado Oil and Gas Conservation Commission (2019, January 3). Operator Guidance for Order No. 1-232, Bradenhead
Pressure Monitoring and Reporting
Kutchko, B., Crandall, D., Moore, J., Gill, M., Haljasmaa, I., Spaulding, R., Harbert, W., Benge, G., DeBruijn, G., Shine,
J. (2014, September 10). Assessment of Foamed Cement Use in Deep Offshore Wells. 170298-MS.
Nelson, E.B., Guillot, D. (2006). Well Cementing, 2nd Edition. Schlumberger