Influence of Chemical, Mechanical, and Transport Processes on

Jul 25, 2017 - This is of particular concern for storage of CO2 in depleted oil and gas reservoirs with numerous legacy wells engineered to variable s...
0 downloads 13 Views 4MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article pubs.acs.org/accounts

Influence of Chemical, Mechanical, and Transport Processes on Wellbore Leakage from Geologic CO2 Storage Reservoirs Published as part of the Accounts of Chemical Research special issue “Chemistry of Geologic Carbon Storage”. Susan A. Carroll,*,† Jaisree Iyer,† and Stuart D. C. Walsh‡ †

Lawrence Livermore National Laboratory, Livermore, California 94550, United States School of Petroleum Engineering, University of New South Wales, UNSW, Sydney, NSW 2052, Australia



CONSPECTUS: Wells are considered to be high-risk pathways for fluid leakage from geologic CO2 storage reservoirs, because breaches in this engineered system have the potential to connect the reservoir to groundwater resources and the atmosphere. Given these concerns, a few studies have assessed leakage risk by evaluating regulatory records, often self-reported, documenting leakage in gas fields. Leakage is thought to be governed largely by initial well-construction quality and the method of well abandonment. The geologic carbon storage community has raised further concerns because acidic fluids in the CO2 storage reservoir, alkaline cement meant to isolate the reservoir fluids from the overlying strata, and steel casings in wells are inherently reactive systems. This is of particular concern for storage of CO2 in depleted oil and gas reservoirs with numerous legacy wells engineered to variable standards. Research suggests that leakage risks are not as great as initially perceived because chemical and mechanical alteration of cement has the capacity to seal damaged zones. Our work centers on defining the coupled chemical and mechanical processes governing flow in damaged zones in wells. We have developed process-based models, constrained by experiments, to better understand and forecast leakage risk. Leakage pathways can be sealed by precipitation of carbonate minerals in the fractures and deformation of the reacted cement. High reactivity of cement hydroxides releases excess calcium that can precipitate as carbonate solids in the fracture network under low brine flow rates. If the flow is fast, then the brine remains undersaturated with respect to the solubility of calcium carbonate minerals, and zones depleted in calcium hydroxides, enriched in calcium carbonate precipitates, and made of amorphous silicates leached of original cement minerals are formed. Under confining pressure, the reacted cement is compressed, which reduces permeability and lowers leakage risks. The broader context of this paper is to use our experimentally calibrated chemical, mechanical, and transport model to illustrate when, where, and in what conditions fracture pathways seal in CO2 storage wells, to reduce their risk to groundwater resources. We do this by defining the amount of cement and the time required to effectively seal the leakage pathways associated with peak and postinjection overpressures, within the context of oil and gas industry standards for leak detection, mitigation, and repairs. Our simulations suggest that for many damage scenarios chemical and mechanical processes lower leakage risk by reducing or sealing fracture pathways. Leakage risk would remain high in wells with a large amount of damage, modeled here as wide fracture apertures, where fast flowing fluids are too dilute for carbonate precipitation and subsurface stress does not compress the altered cement. Fracture sealing is more likely as reservoir pressures decrease during the postinjection phase where lower fluxes aid chemical alteration and mechanical deformation of cement. Our results hold promise for the development of mitigation framework to avoid impacting groundwater resources above any geologic CO2 storage reservoir by correlating operational pressures and barrier lengths.

1. MOTIVATION 3

4

Carbon capture and storage (CCS) is a viable and demonstrated solution to this problem. There are currently 15 large scale CCS plants in operation around the globe with another 6−7 anticipated to come online by the end of 2017.4,7 Nevertheless, commercialization of CCS is limited by both the absence of carbon policies and perceived risks to long-term storage.

5

Recent energy forecasts by the IEA, Statoil, and Exxon Mobil predict increases of 20−30% in energy demand by the year 2040. At the same time, the fastest growing renewable energy sectors, solar and wind, currently constitute a small proportion of the world’s energy supply and are anticipated to remain so for many years into the future.3,4 Consequently, to limit global warming in the short to medium term, reductions in atmospheric CO2 emissions must be found in a world that continues to rely heavily on carbon-based energy sources.6 © 2017 American Chemical Society

Received: February 21, 2017 Published: July 25, 2017 1829

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research

portlandite (blue), a zone enriched in calcium carbonate (green), and a residual amorphous silicate zone (red). An issue perplexing those investigating chemical alteration of cement for secure geologic CO2 storage was the observed contradiction of increases in cement porosity and corresponding decreases in permeability following chemical alteration of the cement. Figure 1a shows enhanced porosity in the depleted layer caused by the dissolution of calcium hydroxides and in the amorphous silicate layer caused by dissolution of calcium carbonates and restructuring of calcium-bearing cement minerals. Increases in porosity are usually associated with increases in permeability, not the reverse as commonly observed in reactive-transport experiments on CO2-cement systems. Explanations for decreased permeability included changes in molecular volume of carbonate minerals, precipitation of calcium carbonate in the flow path, increased compressibility of the altered layers, and low density of amorphous silicate.13,15,19,20,22,25−28 Our group investigated how reaction zone growth affects the hydraulic and mechanical properties of fractures and interfaces in wellbore cement in core-flood experiments. The experiments tested the hypothesis that plastic deformation of chemically altered asperities maintaining the fracture aperture are responsible for observed decreases in permeability. Three idealized fracture geometries were considered: a regular grid of interlinked apertures of circular cross-section (gridded), a roughened surface created by bead blasting without masking (roughened), and a single straight channel (channel) (Figure 2). The experiments are described in greater detail in our papers.18,19,22

Wellbore leakage in particular, tops the lists of these perceived risks. Wells are a direct, engineered path from storage reservoir to surface, piercing the intervening geologic layers. Properly constructed wells provide a virtually impervious barrier to any unintended subsurface transmission. However, damage to the well during completion or CO2 injection can provide a direct path between the CO2 storage reservoir and the atmosphere. The long-term integrity of such wells is vital to the success of any CCS operation. Understanding the long-term integrity of CCS wells is a complex prospect due to the reactivity of alkaline well cement with carbonated brine. The impact of these reactions on the well integrity is compounded by competition between dissolution and precipitation reactions and their effect on the mechanical strength of the cement. In the following, we describe experimental studies and numerical models developed by Lawrence Livermore National Laboratory to address leakage risks associated with wells in CO2 storage sites.

2. EXPERIMENTS IDENTIFYING CHEMICAL AND MECHANICAL PROCESSES Our key contribution toward understanding leakage risk is the identification of how chemical alteration, caused by the reaction of ordinary Portland cements with CO2 storage fluids, leads to mechanical deformation and sealing of flow pathways. Understanding the response of Portland cements, commonly used in well completions and abandonment, is especially important because depleted oil and gas fields provide good candidates for CO2 storage.8 In this section, we highlight the fundamental reactions between CO2 and Portland cement and how they modify cement’s mechanical properties. We do not discuss details involving the reactions of cement or CO2 with O2, H2S, and SO2 impurities in the CO2 waste stream, with common additives to improve cement performance, or with steel. There is a large body of work describing chemical alteration of Portland cements described in Carey et al.,9 Carroll et al.,2 and other contributions.10−21 Diffusion of carbonic acid from carbonated brine through cement results in a series of reaction fronts that divide the cement into distinct zones.10−13,18,22,23 The chemically distinct zones are seen as variations in X-ray tomography image gray scale, emphasized with false colors in Figure 1a. From the inner unreacted cement to the cement−brine interface, the cement consists of an unaltered zone (no false color), a zone depleted in

Figure 2. Artificial fractures: (a) gridded, (b) roughened, and (c) channel. Reproduced with permission from Walsh et al.19 Copyright 2014 Elsevier.

The first piece of evidence supporting our hypothesis is the distinct permeability evolution estimated, despite similar

Figure 1. (a) Reaction fronts seen in an X-ray tomography image for a reacted cement sample with flow direction perpendicular to the image. (b) Distribution of fronts along a flow pathway and aperture reduction due to mechanical deformation (dashed lines) and calcium carbonate (modeled as calcite) precipitation. The hashed areas indicate the constriction of the original fracture aperture. Adapted with permission from Iyer et al.24 Copyright 2016 Elsevier.

Figure 3. Comparison of the change in the hydraulic aperture, b, for the gridded, channel, and roughened fracture surfaces subjected to the same inlet flow rate and brine composition. Reproduced with permission from Walsh et al.19 Copyright 2014 Elsevier. 1830

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research

Figure 4. XRCT-PIV cross sections with color overlay showing regions of compressive plastic deformation (top) and extent of portlandite-depleted (blue), carbonate (green), amorphous silicate (red), and fracture pathway (neutral) for the roughened, channel, and gridded fractures (bottom, left to right). Reproduced with permission from Walsh et al.19 Copyright 2014 Elsevier.

3. COUPLED CHEMICAL, MECHANICAL, AND TRANSPORT MODEL FOR WELL CEMENT

chemical alteration for the three fracture geometries (Figures 3 and 4). The net aperture, b, is related to permeability, k, through

k=

b2 12

Our wellbore model incorporates both mechanical deformation of chemically altered cement and calcium carbonate (modeled as calcite) precipitation in the fracture, because both processes play significant roles in reducing the fracture aperture and leakage. Reduction in aperture is shown in Figure 1b as upstream mechanical deformation of the altered zones and downstream calcium carbonate precipitation. Simulating the wellbore system involves four coupled models that account for multiphase flow of the brine and supercritical CO2 mixture, advective transport of the dissolved chemical species along the fracture, geochemical propagation of the reaction fronts into the cement, and mechanical deformation of the fracture in response to the reaction. Descriptions of these models can be found in our earlier papers.18,19,24,29

(1)

The roughened samples had the smallest initial aperture because the topography was the most subdued. As a result, there was minimal change in the aperture with a slight increase toward the second half of the experiment, indicative of channeling. The channel sample had the highest initial aperture reflecting the deepest and largest flow path. This sample resulted in little changes in the net aperture upon reaction with CO2-rich brine, because the aperture is supported by wide, unreacted sidewalls. The sidewalls are analogous to very large asperities whose strength is uncompromised by chemically altered areas. The greatest reduction in aperture is observed in the gridded sample where the asperities were subject to reaction. In these reacted samples, the asperities consist largely of portlandite-depleted cement. Both the depleted and the amorphous zones are much more compressible than the unaltered or carbonate zones.13,22 The second important piece of evidence is the plastic deformation of the chemically altered asperities shown in Figure 4. We used a form of digital imaging correlation, XRCT-PIV, to measure the displacement field and local strain by correlating the positions of minerals in the pre- and postreaction tomography.19 The result shows that portlandite-depleted pillars in the gridded samples experienced the greatest compressive deformation, in agreement with the greatest hydraulic aperture reduction. Little compression was observed in the roughened or channel experiments, despite similar chemical alteration along the length of each core.

3.1. Flow and Transport Model

The flow of two-phase mixtures of brine and supercritical CO2 is modeled by solving mass balances for brine and CO2. The velocities of the two phases are calculated using the extension of Darcy’s law to multiphase flow. The local cubic law used to compute the single-phase permeability of the fracture is modified by a relative permeability model to determine the permeability of the individual phases. We used the linear and Corey models to test the sensitivity of our simulations, because universally accepted relative permeability models for flow through fractures are lacking.30 We assume that the rate of dissolution of CO2 in brine is significantly faster than the rate at which it diffuses into or reacts with the cement. Therefore, the model assumes the dissolved carbon concentration is equal to the solubility of carbon dioxide at the simulation pressure and temperature, even for brine-filled fractures. A mass balance on calcium, accounting 1831

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research

Figure 5. Response of the fracture aperture to changes in the effective stress is simulated with a mechanical model that couples the overall stiffness of the fracture with the extent of reaction. Reproduced with permission from Walsh et al.19 Copyright 2014 Elsevier.

3.3. Geomechanical Model

for its advection, diffusion, and reaction with cement, is solved to calculate its concentration in brine.

The empirical model simulates the relationship between a changing hydraulic aperture and chemical alteration (details in Walsh et al.19). Briefly, the model assumes that the fracture aperture is supported by cylindrical cement asperities and that the chemically altered cement layers grow radially within the asperities (Figure 5). Changes to the hydraulic aperture due to the change in the cement’s mechanical properties are modeled using a set of parallel spring systems, each representing the contribution of either an alteration zone or the unreacted cement. The individual contributions are weighted by the areas of each of the reaction zones, such that the overall stiffness constant K for the fracture is

3.2. Geochemical Model

Walsh et al.18,29 developed a reduced-physics numerical model to efficiently simulate the chemical interactions between cement and carbonated brine. The model simulates the growth of the reaction zones by assuming that the dissolution and precipitation reactions occur at discrete reaction fronts between layers. The reactions controlling the movement of each front are based on those originally identified by Kutchko and co-workers11 to control the formation of reaction fronts in wellbore cement: Innermost Front: Ca(OH)2(s) → Ca 2 + + 2OH−

(2)

K = (K uA u + KdeplAdepl + KamA am )/(A u + Adepl + A am ) (5)

Middle Front: Ca 2 + + CO32 − → CaCO3(s)

where Ku, Kdepl, and Kam and Au, Adepl, and Aam represent the stiffnesses and the areas of the unreacted cement, the depleted region, and the amorphous region, respectively. Testing revealed that the calcium carbonate layer was generally thin and had similar mechanical properties to the unreacted cement.19 To limit the number of parameters in the model, the calcite layer was not explicitly represented in the mechanical model but rather included as a component of the other layers. For simplicity, the unreacted cement is represented by a simple linear spring, because we assume that its stiffness is independent of the loading history. Simple strain-hardening models were used to represent the amorphous and depleted regions (Figure 5a): regions with significantly lower yield strengths than the unreacted cement. These are modeled as springs of stiffness K′ arranged in series with a parallel spring/frictional slider system with stiffness K″ and yield σy. The extent of the reaction is coupled to the mechanical response through the contact areas as shown in Figure 5, where R0 represents an effective contact radius for the asperities holding the fracture open and Δxam and Δxdepl represent the depth of penetration of the reaction fronts into the asperities. In this manner, the extent of reaction and asperity size are linked to mechanical deformation and fracture closure under load. The model was calibrated against the experimental

(3)

Outermost Front: CaCO3(s) → Ca 2 + + CO32 −

(4)

The speed of each front depends on the flux of reactants and products into or out of the fronts, which in turn depends on diffusion and reaction rates. The front between the unreacted cement and portlandite depleted layer is defined as portlandite equilibrium, because portlandite dissolution is typically much faster than calcium carbonate (modeled as calcite) precipitation or dissolution.31 The concentrations at the other fronts depend on relative speed of diffusion and reaction.24 When the rate of reaction is significantly faster than the diffusion rate, equilibrium front concentrations are used, and ion fluxes are calculated based on diffusion across the front.29 When the layers are thin and the rate of diffusion exceeds the reaction rate, front concentrations equal the brine concentration, and the flux is calculated based on the slower reaction rates (which depend on the brine chemistry).24 Kinetic and thermodynamic equations used are reported in Iyer et al.24 Front velocities are derived from the fluxes and are integrated to obtain the front positions. 1832

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research

stops. Leakage risk associated with geologic CO2 storage reservoirs is thought to match the overpressure profile and decrease significantly after injection stops and the site is closed. Although the reservoir response to pressure should limit the potential for brine leakage to the injection period, this should not be the case for CO2. Buoyancy should continue to move CO2 up leakage pathways even in the absence of an overpressure. The only way to limit CO2 transport is to reduce the aperture and seal the fracture. The oil and gas industry mitigates leakage risk by completing wells with enough cement to isolate the reservoir fluids from the overburden. Requirements vary from ∼9 m (30 ft) in the United Kingdom to ∼15 to 30 m (50 to 100 ft) in the United States and 100 m in Norway (see references and appendices in IEA GHG36). It is important to know if the standards developed for natural gas and reservoir brines are adequate for the much more reactive CO2 and brine mixtures associated with geologic CO2 storage. In addition to overpressure, it is important to understand the role of stress on wellbore leakage. The influence of mechanical deformation of the altered cement on fracture permeability depends on the magnitude of the compressive stresses on the fractures and the manner in which the fracture aperture is supported.37,38 We consider two scenarios, because neither variable has been sufficiently quantified. In the first, the effects of stress on the fracture are ignored. This may result, for example, from minimal confining pressures on the fracture or the presence of large asperities holding the fracture open (similar to the channel fracture experiments). For this scenario, permeability reduction occurs only by calcium carbonate precipitation in the fracture; there is no compression of chemically altered cement asperities. The other scenario considers an applied effective stress perpendicular to the fracture that compresses the altered cement asperities and reduces permeability (similar to the gridded fracture experiments). We show results for a small average effective stress of 6 MPa to illustrate the role of mechanical deformation in limiting leakage. We focus our analysis on defining the aperture range that can be sealed by chemical and mechanical alteration, and how much cement and time is needed to effectively seal the leakage pathway for overpressures associated with peak and post injection. For this discussion, we define 10 MPa to be the overpressure just prior to the end of the injection period and 1 MPa to be the overpressure at 4 times the injection period. A fracture is assumed to be sealed when the minimum aperture along the length of the fracture has reduced to 1 μm. We first calculate the maximum fracture aperture that would seal as a result of chemical precipitation for cement barriers of 10 and 100 m to capture the range of oil and gas industry practices.36 In our calculations, the fractured pathways occur within the cement where chemical alteration is greatest and calcium carbonate precipitates on both sides of the fracture. We then illustrate how chemistry, mechanics, and transport reduce gas and brine fluxes as a function of time. We provide a few examples showing the reduction in sealing efficiency if leakage occurred along fractures between the cement and an inert caprock.

observations of the hydraulic aperture as a function of the extent of the amorphous layer in the gridded samples (Figure 6). The values of the experimentally calibrated parameters used in the mechanical model are listed in Iyer et al.24

Figure 6. Comparison of predicted and observed relative change in hydraulic aperture η versus the extent of the amorphous layer, a proxy for chemical alteration of the cement. Reproduced with permission from Walsh et al.19 Copyright 2014 Elsevier.

4. WELL INTEGRITY FOR GEOLOGIC CO2 STORAGE EXAMINED THROUGH OIL AND GAS PRACTICES We use our multiphase coupled chemical, mechanical, and transport model to address when, where, and in what conditions fracture pathways seal in CO2 storage wells so that leakage risk to groundwater resources is minimized. For the sake of simplicity, we consider one-dimensional fractures, although the model can be applied to more complex fracture geometries. The assessment is done in the context of oil and gas industry standards for leak mitigation and applied to the physical conditions for CO2 storage projects. Our analysis assumes that some water is present to drive chemical alteration of cement and precipitation of calcium carbonate in fractures. Our simulated results are consistent with previous work noting a correlation between residence time, aperture, and fracture sealing.32−34 Figure 7 provides a conceptual model of project risk over the lifetime of a geologic storage site.35 Increased reservoir pressure

Figure 7. Conceptual model for leakage risk over the lifetime of a geological storage site. Adapted with permission from Benson et al.35 Copyright 2012 Austria and Cambridge Press.

4.1. What Fracture Aperture Can Be Sealed by Calcium Carbonate Precipitation?

caused by the injection of supercritical CO2 is referred to as an overpressure and is the driving force for brine and CO2 leakage up fractured pathways, such as those within the cement barrier and along interfaces between the well casing, cement, and rock. Over the storage lifetime, the reservoir overpressure is greatest during injection, returning to ambient pressure after injection

First consider peak injection with a 10 MPa overpressure, in which mechanical deformation plays no role in sealing the fracture. In Figure 8a, each curve defines the maximum aperture that can be sealed by calcium carbonate precipitation as a 1833

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research

Figure 8. Apertures sealed by calcium carbonate precipitation in fractures through 10 and 100 m cement barriers assuming 10 MPa (a) and 1 MPa (b) overpressure and linear and Corey relative permeability models.

Figure 9. CO2 leakage rates for injection and postinjection overpressures (10 and 1 MPa). Green indicates fluxes below the industry threshold and do not require remediation.

excess of 50−350 μm are likely to remain open and not be sealed by chemical alteration for a 10 m cement barrier. Increasing the cement interval to 100 m, the length required in Norway, adds another level of mitigation sealing fractures with apertures up to 200−1400 μm. Longer brine residence times associated with longer cement fractures result in more chemical alteration of the cement and more calcium carbonate precipitation in the fracture. Longer residence times are also associated with lower brine permeability predicted by the Corey model, especially as the brine fraction decreases, sealing larger apertures with calcium

function of the initial amounts of CO2 and brine for two cement lengths and two relative permeability models. Fractures below the curve seal, and those above remain open. The curves are extracted from simulations that track sealing behavior as a function of initial brine residence time and aperture.24 For a given brine−CO2 mixture, permeability model, and fracture length, the initial residence time is converted to specific pressure drops. The range of maximum aperture values reflect differences in brine and CO2 saturations and the relative permeability models used in the calculations. Our calculations show apertures in 1834

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research

higher than the industry threshold. Addition of compressive stress reduces the flux to 3 times the threshold, but the fracture does not completely seal. In contrast, cement−cement fractures can seal in ∼200 days by calcium carbonate precipitation alone. The role of the 6 MPa effective compressive stress is to reduce the fracture aperture in a relatively short time as marked by the steep drop in gas leakage rates in the dashed curves in Figure 9a,b. For these examples, the 6 MPa compressive stress is not enough to reduce the fracture aperture to 1 μm, and calcium carbonate precipitation is required to completely seal the fracture. Like the peak injection scenarios, smaller fractures seal rapidly as a result of calcium carbonate precipitation in the fracture, with mechanical deformation playing little role in reducing the aperture. Generally, brines fluxes follow a similar trend with time to the CO2 fluxes discussed above, but are around 2−10 times lower than the CO2 fluxes due to lower mobility.

carbonate minerals. These calculations assume reactive surface area is independent of the brine saturation, based on the premise that a small amount of water is enough to sustain the reaction.39,40 Concerns for wellbore leakage extend beyond the injection phase, not only because of CO2’s buoyancy but also because the plume may intersect legacy oil and gas wells as it grows over time. In a postinjection environment with an overpressure of 1 MPa, residence times are longer, because brine flow along the fracture is slower, and chemical alteration is greater. As a result, fractures seal with apertures up to 100−700 μm over a 10 m cement interval and 350−2700 μm over a 100 m cement interval, depending on the relative permeability model (Figure 8b). 4.2. How Quickly Do the Fractures Seal?

We assess the time it take for fractures to seal, because rapidly sealing fractures reduce risk to groundwater resources. Industry guidelines in Alberta, Canada, require repairs to any leak that has a gas flow rate of over 300 m3/day or has any form of brine flow.1 We compare our model predictions against the gas flow limit corresponding to 540 kg/day of CO2 at 25 °C and atmospheric pressure. Figure 9 compares the time to seal fractures for peak and postinjection overpressures. The plots compare the CO2 leakage rates through the fractures in the presence and absence of compressive stress, using a linear relative-permeability model. Our calculations assume a 100 m cement barrier with cement− cement and cement−caprock, where the caprock is inert. Fracture width corresponds to 10% debonding along the circumference of a 6 in. diameter hole, 47 mm. Large 500 μm apertures never seal at peak overpressures because brine flow is too fast and its concentration too dilute for calcium carbonate to precipitate in cement−cement or cement− caprock fractures. A 6 MPa compressive stress perpendicular to the well significantly lowers leakage from ∼20000 kg/day to less than 3000 kg/day. Despite the significant reduction, the CO2 leakage rate is still ∼5 times higher than the 540 kg/day threshold for cement−cement and cement−caprock fractures. Pathways with smaller aperture readily seal even when the overpressure is elevated. For the cement−cement example shown, a 100 μm fracture seals within 10 days (Table 1). Cement−caprock

5. IMPLICATIONS FOR WELL LEAKAGE AT CO2 STORAGE SITES In this Account, we have summarized key operational and physical processes that help limit the risk of brine and CO2 leakage from wells found within storage reservoirs using our experimentally calibrated chemical, mechanical, and transport model. The simulations support the hypothesis that leakage risk is greatest during sustained injection when reservoir overpressure is greatest (Figure 7). During peak injection, chemical alteration and mechanical deformation reduce risk by closing fracture apertures, but the extent of closure depends on the fracture dimensions. Leakage risk is significantly reduced postinjection, because lower reservoir pressures yield lower fluid fluxes and result in the chemical and mechanical closure of leakage pathways. One metric for site closure could be the absence of well leakage at some defined minimum overpressure in the storage reservoir. Leakage at later times would be unlikely, as long as reservoir pressure was maintained or continued to decrease. This assessment does not consider long-term pressure effects due to the buildup of buoyancy driven flow that might change the state of stress in the well. Risk mitigation involves properly engineering the well to ensure long-term CO2 storage. Figure 8 provides a preliminary framework describing an acceptable amount of wellbore damage to avoid impacting groundwater resources overlying any geologic CO2 storage reservoir. In the United States, permits for the development of geologic CO2 storage are regulated under the Environmental Protection Agency as Class VI Wells.41 There is flexibility in the regulations, but generally proper abandonment includes a fully cemented casing with 100 foot plugs at the base of the well and below the deepest groundwater resource. While our analysis suggests that longer barriers are typically better than shorter ones, the length required strongly depends on the bonding between cement and the casing and rock surfaces. A few field studies suggest that the extent of the damage can be minimal (with evidence of 100 μm zones of carbonation27) to extensive (with carbonation zones between 1000 to 10000 μm thick10). Other field studies have observed cement bonding that was intact42 or completely degraded.43 Field studies such as these emphasize the importance of testing well integrity of injection and legacy wells to develop mitigation plans for storage sites.

Table 1. Sealing Time in Days cement−cement fracture

cement−caprock fracture

aperture

risk

σeff = 0

σeff = 6 MPa

σeff = 0

σeff = 6 MPa

500 μm

high risk low risk high risk low risk

never 220 9 10

never 216 9 10

never never 36 39

never never 36 39

100 μm

fractures also seal but require more time (∼36 days) as only one fracture surface is reactive compared to two in cement−cement fractures. The smaller fractures seal solely by calcium carbonate precipitation, with no additional reduction in leakage rates due to mechanical deformation. Leakage rates, even without sealing, are significantly below the industry threshold. As expected, fluxes for the postinjection scenarios are lower; however the ability of the fracture to seal depends on the type of fracture and if stress compresses the chemically altered cement asperities. Fractures with larger apertures of 500 μm fail to seal through calcium carbonate precipitation alone if they are at the cement−caprock interface and have leakage rates ∼8 times



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. 1835

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research ORCID

(7) Global CCS, The global status of CCS: 2016, Summary report; 2016; p 28. (8) Watson, T. L.; Bachu, S. Identification of wells with high CO2leakage potential in mature oil fields developed for CO2-enhanced oil recovery. 2008, SPE-112924-MS. (9) Carey, J. W. Geochemistry of wellbore integrity in CO2 sequestration: Portland cement-steel-brine-CO2 interactions. Rev. Mineral. Geochem. 2013, 77, 505−539. (10) Carey, J. W.; Wigand, M.; Chipera, S. J.; WoldeGabriel, G.; Pawar, R.; Lichtner, P. C.; Wehner, S. C.; Raines, M. A.; Guthrie, G. D., Jr. Analysis and performance of oil well cement with 30 years of CO2 exposure from the SACROC unit, West Texas, USA. Int. J. Greenhouse Gas Control 2007, 1, 75−85. (11) Kutchko, B. G.; Strazisar, B. R.; Dzombak, D. A.; Lowry, G. V.; Thaulow, N. Degradation of well cement by CO2 under geologic sequestration conditions. Environ. Sci. Technol. 2007, 41, 4787−4792. (12) Kutchko, B. G.; Strazisar, B. R.; Lowry, G. V.; Dzombak, D. A.; Thaulow, N. Rate of CO2 attack on hydrated class H well cement under geologic sequestration conditions. Environ. Sci. Technol. 2008, 42, 6237−6242. (13) Kutchko, B. G.; Strazisar, B. R.; Huerta, N.; Lowry, G. V.; Dzombak, D. A.; Thaulow, N. CO2 reaction with hydrated class H well cement under geologic sequestration conditions: Effects of flyash admixtures. Environ. Sci. Technol. 2009, 43, 3947−3952. (14) Duguid, A.; Scherer, G. W. Degradation of oilwell cement due to exposure to carbonated brine. Int. J. Greenhouse Gas Control 2010, 4, 546−560. (15) Huerta, N. J.; Hesse, M. A.; Bryant, S. L.; Strazisar, B. R.; Lopano, C. L. Experimental evidence for self-limiting reactive flow through a fractured cement core: Implications for time-dependent wellbore leakage. Environ. Sci. Technol. 2013, 47, 269−275. (16) Scherer, G. W.; Kutchko, B.; Thaulow, N.; Duguid, A.; Mook, B. Characterization of cement from a well at Teapot Dome Oil Field: Implications for geological sequestration. Int. J. Greenhouse Gas Control 2011, 5, 115−124. (17) Zhang, M.; Bachu, S. Review of integrity of existing wells in relation to CO2 geological storage: What do we know? Int. J. Greenhouse Gas Control 2011, 5, 826−840. (18) Walsh, S. D. C.; Mason, H. E.; Du Frane, W. L.; Carroll, S. A. Experimental calibration of a numerical model describing the alteration of cement/caprock interfaces by carbonated brine. Int. J. Greenhouse Gas Control 2014, 22, 176−188. (19) Walsh, S. D. C.; Mason, H. E.; Du Frane, W. L.; Carroll, S. A. Mechanical and hydraulic coupling in cement-caprock interfaces exposed to carbonated brine. Int. J. Greenhouse Gas Control 2014, 25, 109−120. (20) Huerta, N. J.; Hesse, M. A.; Bryant, S. L.; Strazisar, B. R.; Lopano, C. Reactive transport of CO2-saturated water in a cement fracture: Application to wellbore leakage during geologic CO2 storage. Int. J. Greenhouse Gas Control 2016, 44, 276−289. (21) Wolterbeek, T. K. T.; Hangx, S. J. T.; Spiers, C. J. Effect of CO2induced reactions on the mechanical behaviour of fractured wellbore cement. Geomech. Energy Environ. 2016, 7, 26−46. (22) Mason, H. E.; Frane, W. L. D.; Walsh, S. D. C.; Dai, Z.; Charnvanichborikarn, S.; Carroll, S. A. Chemical and mechanical properties of wellbore cement altered by CO2-rich brine using a multianalytical approach. Environ. Sci. Technol. 2013, 47, 1745−1752. (23) Zhang, L.; Dzombak, D. A.; Kutchko, B. G. Wellbore cement integrity under geologic carbon storage conditions. Novel Materials for Carbon Dioxide Mitigation Technology 2015, 333−362. (24) Iyer, J.; Walsh, S. D. C.; Hao, Y.; Carroll, S. A. Incorporating reaction-rate dependence in reaction-front models of wellbore-cement/ carbonated-brine systems. Int. J. Greenhouse Gas Control 2017, 59, 160− 171. (25) Wigand, M.; Kaszuba, J. P.; Carey, J. W.; Hollis, W. K. Geochemical effects of CO2 sequestration on fractured wellbore cement at the cement/caprock interface. Chem. Geol. 2009, 265, 122−133.

Susan A. Carroll: 0000-0002-6456-3318 Notes

The authors declare no competing financial interest. Biographies Dr. Susan A. Carroll is a researcher at the Lawrence Livermore National Laboratory. She received her B.S. in geology from the University of Kansas and her Ph.D. in geochemistry from Northwestern University. Dr. Carroll’s research areas of interest include carbon sequestration, geothermal energy production, and radioactive waste disposal. Dr. Jaisree Iyer is a postdoctoral researcher at the Lawrence Livermore National Laboratory. She received her B.Tech. from the Indian Institute of Technology, Bombay, and Ph.D. from the Massachusetts Institute of Technology, both in chemical engineering. Following that she worked as an associate scientist at the Schlumberger Doll Research Center. Her recent work has focused on reactive flow in porous and fractured media for wellbore integrity and acid stimulation applications. Dr. Stuart D. C. Walsh is a Senior Lecturer at the University of New South Wales, and a former member of the Lawrence Livermore National Laboratory. His current work focuses on the development of tightly coupled geomechanical, hydrodynamic, and geochemical multiscale models and their application to geoscientific and engineering problems. Dr. Walsh’s research areas of interest include carbon sequestration, geothermal energy production, hydraulic fracturing, and the development of novel stimulation and drilling technologies.



ACKNOWLEDGMENTS This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. This work was completed as part of the National Risk Assessment Partnership (NRAP) project (Field work proposal no. FEW0182). Support for this project came from the DOE Office of Fossil Energy’s Carbon Capture and Storage program. In addition, a part of this work was conducted using resources provided by the Pawsey Supercomputing Centre with funding from the Australian Government and the Government of Western Australia. The authors acknowledge Traci Rodosta (NETL Carbon Storage Program) and Mark Ackiewicz (DOE Office of Fossil Energy) for programmatic guidance, direction, and support. The authors thank LLNL researchers Wyatt Du Frane, Harris Mason, Joe Morris, and Pratanu Roy, as well as the broader NRAP team for contributions towards understanding wellbore integrity.



REFERENCES

(1) Watson, T. L.; Bachu, S. Evaluation of the potential for gas and CO2 leakage along wellbores. SPE Drill. Completion 2009, 24, 115−126. (2) Carroll, S.; Carey, J. W.; Dzombak, D.; Huerta, N. J.; Li, L.; Richard, T.; Um, W.; Walsh, S. D. C.; Zhang, L. Review: Role of chemistry, mechanics, and transport on well integrity in CO2 storage environments. Int. J. Greenhouse Gas Control 2016, 49, 149−160. (3) International Energy Agency, World energy outlook 2016; 2016; p 684. (4) Statoil, Energy perspectives: Long-term macro and market outlook; 2015; p 60. (5) Exxon Mobil Corporation, The outlook for energy: A view to 2040; 2017; p 52. (6) IPCC, Climate change 2014: Mitigation of climate change. Contribution of working group III to the fifth assessment report of the Intergovernmental Panel on Climate Change. 2014. 1836

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837

Article

Accounts of Chemical Research (26) Newell, D. L.; Carey, J. W. Experimental evaluation of wellbore integrity along the cement-rock boundary. Environ. Sci. Technol. 2013, 47, 276−282. (27) Crow, W.; Carey, J. W.; Gasda, S.; Williams, D. B.; Celia, M. Wellbore integrity analysis of a natural CO2 producer. Int. J. Greenhouse Gas Control 2010, 4, 186−197. (28) Abdoulghafour, H.; Luquot, L.; Gouze, P. Characterization of the mechanisms controlling the permeability changes of fractured cements flowed through by CO2-rich brine. Environ. Sci. Technol. 2013, 47, 10332−10338. (29) Walsh, S. D. C.; Du Frane, W. L. D.; Mason, H. E.; Carroll, S. A. Permeability of wellbore-cement fractures following degradation by carbonated brine. Rock Mech. Rock Eng. 2013, 46, 455−464. (30) Walsh, S. D. C.; Carroll, S. A. Fracture-scale model of immiscible fluid flow. Phys. Rev. E 2013, 87, 013012. (31) Galí, S.; Ayora, C.; Alfonso, P.; Tauler, E.; Labrador, M. Kinetics of dolomite-portlandite reaction: Application to Portland cement concrete. Cem. Concr. Res. 2001, 31, 933−939. (32) Cao, P.; Karpyn, Z. T.; Li, L. Dynamic alterations in wellbore cement integrity due to geochemical reaction in CO2-rich environments. Water Resour. Res. 2013, 49, 4465−4475. (33) Brunet, J.-P. L.; Li, L.; Karpyn, Z. T.; Huerta, N. J. Fracture opening or self-sealing: Critical residence time as a unifying parameter for cement-CO2-brine interactions. Int. J. Greenhouse Gas Control 2016, 47, 25−37. (34) Wolterbeek, T. K.; Peach, C. J.; Raoof, A.; Spiers, C. J. Reactive transport of CO2-rich fluids in simulated wellbore interfaces: Flowthrough experiments on the 1−6 m length scale. Int. J. Greenhouse Gas Control 2016, 54, 96−116. (35) Benson, S.; Bennaceur, K.; Cook, P.; Davison, J.; de Coninck, H.; Farhat, K.; Ramirez, A.; Simbeck, D.; Surles, T.; Verma, P.; Wright, I.; Ahearne, J. Carbon capture and storage. Global energy assessment Toward a sustainable future 2012, 993. (36) Benedictus, T.; van der Kuip, M.; Kronimus, R.; Huibr, J.; Yavuz, F.; Remmelts, G.; Stam, J. Long Term Integrity of CO2 Storage - Well Abandonment. 2009, IEA Greenhouse Gas R&D Programme, Report No. 2009/08. (37) Barlet-Gouédard, V.; Rimmelé, G.; Porcherie, O.; Quisel, N.; Desroches, J. A solution against well cement degradation under CO2 geological storage environment. Int. J. Greenhouse Gas Control 2009, 3, 206−216. (38) Oyarhossein, M.; Dusseault, M. Wellbore Stress Changes and Microannulus Development Because of Cement Shrinkage. Presented at the 49th US Rock Mechanics/Geomechanics Symposium. 2015; ARMA-2015-118. (39) McGrail, B. P.; Schaef, H. T.; Glezakou, V.-A.; Dang, L. X.; Owen, A. T. Water reactivity in the liquid and supercritical CO2 phase: Has half the story been neglected? Energy Procedia 2009, 1, 3415−3419. (40) Schaef, C. F.; Windisch, H. T., Jr.; McGrail, B. P.; Martin, P. F.; Rosso, K. M. Brucite [Mg(OH)2] carbonation in wet supercritical CO2: An in situ high pressure X-ray diffraction study. Geochim. Cosmochim. Acta 2011, 75, 7458−7471. (41) U.S. EPA. Federal Requirements under the Underground Injection Control (UIC) Program for Carbon Dioxide (CO2) Geologic Sequestration (GS) Wells. Fed. Regist. 2010, 45, 77230−77303. (42) Hawkes, C. D.; Gardner, C. Pressure transient testing for assessment of wellbore integrity in the IEAGHG Weyburn-Midale CO2 monitoring and storage project. Int. J. Greenhouse Gas Control 2013, 16, S50−S61. (43) Duguid, A.; Carey, J. W.; Butsch, R. Well integrity assessment of a 68 year old well at a CO2 injection project. Energy Procedia 2014, 63, 5691−5706.

1837

DOI: 10.1021/acs.accounts.7b00094 Acc. Chem. Res. 2017, 50, 1829−1837