Friday, July 24, 2009

Well Construction and Mechanical Integrity Testing

Synthesis, Evaluation and Assessment of
USEPA Technical Workshop on Geosequestration:
Well Construction and Mechanical Integrity Testing
Jerry Thornhill, P.G.
Consultant to Shaw Environmental, Inc.
Randall Ross, Ph.D.
Steven Acree, P.G.

National Risk Management Research Laboratory
Prepared under contract to Shaw Environmental, Inc.
Contract Number 68-C-03-097

Injection and monitoring wells employed for the geologic sequestration of CO2 may be
required to operate for timescales that go beyond the operating lifespan traditionally
considered by the oil and gas industry. Lessons learned from pilot studies and
conventional injection practices may provide some general guidance for the geologic
sequestration of CO2. However, the special problems related to the unique properties of
supercritical CO2 and proposed massive injection volumes may ultimately require the
development of new materials and tools to reduce the risk of failure and, possibly, new
regulations to manage the risks.
Several key issues regarding well construction and MIT with respect to geological
sequestration of CO2 were identified during this workshop.
Well Completion
Although well integrity is the cornerstone for successful injection well projects, an
integral part of a successful project is the initial well completion program (i.e., drilling
the hole, setting and cementing casing, setting tubing, etc.). This is vital to the successful
operation of any injection well project. The choice of casing and tubing material, cement
type, amount, and proper emplacement is the starting point for the success of the injection

During the workshop, it was stated that,
“Wellbore integrity problems do exist in oil and gas operations and are often due
to cementing practices.”

This statement was apparently based on information from meetings of the International
Energy Agency (IEA) Greenhouse Gas R&D Programme. Key findings from the IEA
March 2006 meeting (IEA, 2006) included, “Well integrity may be a current issue within the oil and gas industry. A detailed study on production wells in the Gulf of Mexico indicated that up to 60% of wells had casing pressure problems, which could indicate that the integrity of the wells
had been compromised.

Experience from the Permian basin in the USA indicated that when fields were changed over to CO2 floods that significant remedial work was needed to pull and re-cement wells that had not seen exposure to CO2. It was considered that many of the problems in both the Gulf of Mexico and the Permian basin resulted from poor well completions at the outset.”

An injection well must be completed in such a manner that underground sources of
drinking water (USDWs) are protected initially, during long-term injection operations,
and following the period of injection. This includes well completion to provide protection
of ground water from naturally occurring salt-water zones; well completion to keep the
injectate in the proposed injection zone and capability to detect any equipment failure
resulting from such things as corrosion, inadequate or unsuccessful cement emplacement,
channeling around an initially successful cement sheath, or mechanical failure.

A specific concern of the participants in the Well Construction breakout session was the
casing metallurgy/coatings in view of the corrosive nature of the CO2. This is especially
critical for the long string casing, tubing and packers that would be in contact with the

Participants in the Well Construction breakout section generally felt that the
Underground Injection Control (UIC) Class I requirements “may” be sufficient for CO2
geologic sequestration. It is recommended that Class I requirements be the minimum
standard considered for CO2 injection.

It has been determined that Portland-based cements react with CO2, leading to cement
degradation. Research has indicated that interaction between cement and CO2 follows a
three-step process - carbonic acid diffusion, dissolution/carbonation, and leaching. This
generally leads to loss of density and strength and an increase in porosity.

From the Los Alamos National Laboratory
Key issues identified by the Network include:
• Wellbore integrity problems exist in oil and gas operations and are often related to
cementing practices.
• Research is needed on reactivity of CO2 and cement to reconcile effects of key
• Methods for determining performance of new CO2-resistant cements are needed.
• Corrosion of tubulars and casing can be more rapid than cement degradation.
• More sensitive field monitoring tools for diagnosing well integrity are needed.
• Numerical models of wellbore geochemistry and geomechanics are needed.
• Numerical models incorporating realistic well permeability distributions are
needed to evaluate leakage potential.
• Evaluation of existing fields with long term CO2 exposure are needed to develop
more effective methods for logging/monitoring for evaluating mechanical
• Mining of existing data from private companies and regulatory authorities should
be a priority for development of a statistical basis for evaluating wellbore

CO2-Cement Interaction: From the Lab to the Well
Matteo Loizzo, Schlumberger Carbon Services Engineering
The presentation summarized research on CO2/cement reactions and the development of CO2-resistant cement. The interaction between Portland cement and CO2 is a 3-step process:
• Carbonic acid diffusion,
• Cement (portlandite) dissolution and carbonate precipitation, and
• Leaching (calcium carbonate dissolution).
Cement sheath defects would cause acceleration of the degradation process. Potential
defects include:
• Inadequate placement of cement resulting in channels or mud films,
• Channels caused by gas migration during cement hydration,
• Cracks caused by cement failure in compression/traction, and
• Microannuli caused by lack of bonding at the interfaces with casing and/or rock.

Research is being conducted on a CO2-resistant cement formulation. It was concluded that sound cement design is required, both for the placement and post-placement phases.

Selecting Sites for Geological Sequestration: Wellbore Integrity and Other Criteria Jason Heath, New Mexico Institute of Mining and Technology
The presenter, representing the Southwest Carbon Sequestration Partnership, described the efforts of the Partnership regarding selection of sites for geological sequestration. One of the key aspects for site selection is the identification of the best sink for each CO2 source.The presenter provided an example of the well integrity analysis at a test site, the Aneth Unit in southern Utah, where well construction deficiencies may potentially affect a pilot test of CO2 injection.
Analysis of well construction deficiencies included:
• Calculation of the top of cement,
• Temperature and cement bond logs, and
• Information on the depth of surface or intermediate casing.
Wells vulnerable to interformational migration of fluids were identified by the screening analysis described above. However, no monitoring has been performed in wells identified as vulnerable.

Well Construction
• Industry has developed recommended practices and protocols for well
construction. However, much of the research upon which the protocols are based
is confidential.
• Experience from EOR and acid gas operations provides a good working basis for
well construction.
• Pilot tests with real-world volumes of CO2 are needed.
• Performance-based construction standards may be appropriate.
Research Need: Development of lower cost materials that perform as well as high-cost materials.

• Abandonment procedures may need to be more stringent for geological
• Casing specifications depend on possible impurities, formation brine, pressure,
temperature, and operational conditions.
• Casing options include chrome tubing, expandable tubing, titanium casing,
fiberglass casing, and inhibited packer fluid for additional protection.
Research Need: Study the impacts of injection at varying depths.
• Cement specifications depend on CO2 impurities; formation brine; and pressure,
temperature, and operational conditions.
• Cement should run the entire length of the wellbore.
Research Need: Alternative (non-Portland) cements.

The research needs determined from this group primarily centered on a review of
laboratory, field, and modeling studies concerning:

• Cement-related microannuli self-enhancing (enlarging) vs. self-healing (sealing),
that have been conducted by the industry,
• Impact of CO2 phase changes on mechanical integrity testing of wells,
• Impact of injectate impurities on the mechanical integrity of wells,
• MIT failure rates for new vs old wells,
• The phenomenon of a cold injection fluid opening up or enlarging gaps within the well system,
• Impact of large temperature differentials between injectate and well system/formation on
pressure tests,
• Monitoring methods/MITs that could detect rates and volumes of fluid movement along the
casing, and
• Time frames of MI changes and necessary MIT frequency.

Mr. Kobelski noted that EPA is currently assessing options for a management framework for
CO2 injection for the purposes of GS. GS presents many technical challenges that go beyond
those associated with CO2 injection for enhanced oil and gas recovery (EOR/EGR). For
example, GS will involve a variety of geologic settings apart from oil and gas reservoirs (e.g.,
saline aquifers and unmineable coal seams). In addition, the CO2 from coal-fired power plants will contain impurities (i.e., sulfur and nitrogen oxides, and metals such as mercury) that are not typically found in the CO2 used in EOR/EGR operations, and GS will involve significantly greater volumes and longer storage times.

Schlumberger Carbon Services - Schlumberger Public
CO2 reaction effects on well integrity
• Carbonation
• Matrix reacts: Portlandite/CSH → Calcite
•Water release
• At an early stage, may affect marginally matrix permeability (10-4→10-3 mD)
• May lead to mechanical instability (Calcite molar volume increase) •
¾” in 7-10 months, 1 m in 2000 years
•CO2 diffusion in water: ¾” in 3 days, 1 m in 20 years
• Leaching
• Strong dependency on local Ca2+ concentration gradient
Cement effectively dissolves

Cement sheath defects – effects on scale
• Fluid flow vs. matrix diffusion
• Preferential path of fluid flow bridges the scales
• Issue not limited to CO2: 15%-20% of wells may show hydraulic communication to surface
• Carbonation healing/plugging may be effective only at small scales
• Karst
• Positive feedback effect from enhanced leaching on defect walls

Assuring cement integrity over the well life
• Risk factors and scales
• Casing corrosion
• Leakage to shallower formations or to surface
• Multiple layers of risk mitigation
• Especially when repair is difficult
• Cement system selection and optimization
• Minimize or eliminate cement sheath defects
• Minimize or eliminate cement degradation
•Not necessarily cement reaction!

Well Construction: Potential Effects on Pilot Test
Potential impact of construction deficiencies:
Construction deficiencies could “provide a potential pathway for fluid migration between aquifers where there exists a differential in hydraulic head between aquifers.”
“Because the De Chelly aquifer hydraulic head exceeds the Navajo aquifer head in much of the Aneth Field area, saline water from the De Chelly Aquifer could potentially migrate upward into the Navajo aquifer through the partially cemented wellbores.”

How “risky” for CO2 migration are the wells that are vulnerable to communication between the Upper Paleozoic Aquifer and the Navajo Aquifer?
We think that the integrity and reactivity of the cement at/above/below the target reservoir (e.g., at the Paradox Formation in this case) is very important. If CO2 can leak through these “vulnerable” cement zones (e.g., the Paradox Formation here), then superjacent groundwater reservoirs may be impacted. Well cements must be sampled and characterized, and the conditions recorded and implemented in associated reservoir models for quantifying potential risk.

Notes about mechanical integrity testing:
The current portfolio of Regional Partnership pilot tests are small enough, in terms of injection rates, that special mechanical integrity testing is not necessary. Only “routine” mechanical testing is being done for these tests.

For Phase III, which will involve injection of over 1,000,000 tons/year in relatively few wells, plans are in place to include in situ tiltmeters and strain gauges (San Juan Basin). Water
injection pressure transient tests will be carried out prior to CO2 injection to characterize state-of-stress and response.