LA-UR-07-7795

Science-Based Risk Assessment and Wellbores and in a CO2 Storage System

George GuthrieFossil Energy & Environment Programs

Los Alamos National Laboratory

Wellbores Factors in CCS Operations…

• Placement of CO2economic (number of wellbores needed; maintenance, P&A, etc.)

• Potential release of CO2 (penetration through primary seal)

poor completion or abandonmentmechanical damagechemical damage (corrosion of cement; corrosion of casing)

• Potential conduits/fastpaths for CO2 movement within the geologic site

must address phase-change effects

LA-UR-07-7795

Wellbore integrity is important in long-term CO2 storage.

• Wellbores are typically completed & plugged with portland-based cement

hydrated portland cements contain calcium hydroxide (a base) and other acid sensitive materials

• Portland cements can degrade in the presence of CO2 & water

CO2 + water => carbonic acid batch experiments suggest rapid degradation by carbonic acid

LA-UR-07-7795

How can we scale fundamental processes to system-level behavior?

CO2 +brine dissolveshydrated cement

?chemical reactions involving fluids and solids…• in a heterogeneous porous medium• coupled to fluid flow• field kinetics often differ from lab determined

LA-UR-07-7795

Quantitative risk assessment is a formal process to minimize potential consequences of long-term storage.

CO2 storage reservoir

Risk assessment must consider the potential for CO2 release and subsequent movement from storage reservoir to various receptors.

atmosphere; anthropogenic systems

groundwatersystem

LA-UR-07-7795

CO2 -PENS is a probabilistic system model relying on several modules associated with wellbore release.

CO2CO2

CO2 Captureat Power Plants

CO2 Transportand Injection

CO2 Storagein Geologic Reservoir

CO2

CO2 Movementfrom Reservoir

StorageSystem

Potential ReleaseMechanism

Potential Receptors

Transport

Satu

rate

dPo

rous

Flo

wW

ell B

ore

Flow

Wel

l Bor

eR

elea

seSe

alR

elea

seLa

tera

lM

igra

tion

Uns

atur

ated

Poro

us F

low

Atm

osph

eric

Syst

ems

Wat

erW

ells

Oth

erR

eser

voirs

Car

bona

teR

eser

voir

Frac

ture

Flow

LA-UR-07-7795

Science based prediction of natural system performance requires system-level probabilities based on process level phenomena.

theory,experiment,computation

observation(analog sites)

time

Wel

lbor

ePe

rmea

bilit

y

time

i ii

iii iv

Possible scenarios for wellbore fatei. wellbores degrade rapidlyii. wellbores degrade slowlyiii. wellbores improve over timeiv. wellbores are unaffected by CO2 +brine

LA-UR-07-7795

Several pathways have the potential to contribute to CO2 release from wellbores.

Gasda, Bachu, Celia, 2004Princeton CMI

1. flow between cement & casing (Gasda et al. mechanisms A & B)

2a. diffusion through cement (Gasda et al. mechanism C)

0. dissolution/corrosion of cement

2b. fracture flow through cement (Gasda et al. mechanism E)

3. flow between cement & shale (Gasda et al. mechanism F)

LA-UR-07-7795

CO2 -EOR operations routinely utilize wellbore technology to place (and to contain) fluids within the reservoir.

SACROC is one of several industrial-scale examples in the Permian Basin

• ~13.5 million tonnes of CO2 /yr injected• (~6-7 million t/yr of new CO2 )• ~ 70 million tonnes CO2 accumulated

(>30 million tonnes anthropogenic)• CO2 injection since 1972

Carey et al., 2007

49-6

LA-UR-07-7795

Whipstock drilling at SACROC 49-6 provided recovery of core through cemented annulus to within 12’ of top of pay.

Drilled/completed 1950

Water flood initiated 1954

First direct CO2 exposure 197510 yrs as injector; 7 yrs as producer

Topof

Pay

Carey et al., 2005, SPE;Guthrie et al., 2005, Midland CO2 Conf.

LA-UR-07-7795

Observations suggest initial flow along interfaces followed by precipitation of silica and carbonate phases.

casing shalefragment zone

carbonatedcement

“pristine”cement

silica-carbonateprecipitation

shale caprockw

ellb

ore

fluid flow along interface into sandy unit in shaleprecipitation of silica and carbonate from brine along interfacial zonesdiffusive carbonation of cement to form orange zone

Guthrie et al., 2005, Midland CO2 Conf.; Carey et al., 2007

LA-UR-07-7795

Cement sample from SACROC 49-6 (6550’ near top of pay) was exposed to CO2 for ~ 30 years (~110,000 tonnes).

casing shalefragment zone

carbonatedcement

“pristine”cement

silica-carbonateprecipitation

shale caprockw

ellb

ore

“pristine” hydrated cement (portlandite (Ca(OH)2) in matrix/veins)limits both complete dissolution & Gasda et al. mechanisms C&Ethin carbonated zone between cement and casing (Gasda et al. mechanism A)orange carbonated cement (“popcorn” texture) (Gasda et al. mech.s C&F)gray carbonated vein fluid flow followed by precipitation of silica/carbonate

Carey et al., 2007, Int. J. GHG Control, 1:75-85

LA-UR-07-7795

Summary of major processes occurring at wellbore 49-6.

Gasda, Bachu, Celia, 2004Princeton CMI

1. flow between cement & casing (Gasda et al. mechanisms A & B)

2a. diffusion through cement (Gasda et al. mechanism C)

0. dissolution/corrosion of cement

2b. fracture flow through cement (Gasda et al. mechanism E)

3. flow between cement & shale (Gasda et al. mechanism F)

• precipitation along interface

• minimal—Ca(OH)2 preserved

• minimal—Ca(OH)2 along fractures

• cement alteration;• precipitation filling voids

LA-UR-07-7795

Field observations have elucidated fluid-rock reactions, some of which can now be modeled at the continuum scale.

Carey et al., 2007, Int. J. GHG Control, 1:75-85Lichtner & Carey, 2006, GCA, 70:1356-1378.

LA-UR-07-7795

Field observations have elucidated fluid-rock reactions, some of which can now be modeled at the continuum scale.

Carey et al., 2007, Int. J. GHG Control, 1:75-85Lichtner & Carey, 2006, GCA, 70:1356-1378.

“Relative” Permeability(mD; air dried)(B. Svec, NMT)

Upper Cement (~5160’)

Lower Cement (~6550’)

Orange Zone (~6550’)

0.09

0.090.070.11

0.380.190.11

LA-UR-07-7795

Flow-through experimental models can reproduce many of the observations from the field samples.

Wigand, Carey, Hollis, Kaszuba, 2007 (LA-UR-07-1358)

cement

alteration zone

Orange ZoneMineralization

Mineralization along casing-cement interface

LA-UR-07-7795

Assessing Wellbore Integrity Following CO2 Exposure by a Combination of Field, Experiment, Theory, and Computation

Analysis of field samples of cement exposed to CO2(in collaboration with Kinder-Morgan CO2 company)

Laboratory experiments on cement exposure to CO2

Numerical simulation of cement degradation Field observations on gas

migration through wellbore cements in Alberta (in collaboration with Alberta Electric Utilities Board)

Application of non-linear acoustic methods to detect fractures in wellbore cements

LA-UR-07-7795

LA-UR-07-7795

Approximations used in traditional approaches preclude the ability to capture coupled wellbore-reservoir processes in large simulations.

• Finite Element Heat & Mass (FEHM) numerical simulatorMulti-dimensional, multi-phase porous media heat & mass transfer simulator (brine,water, CO2, methane hydrates)Coupled flow, reactive transport, and stress capabilitiesFinite volume method captures complex geologyIncorporation of wellbores and near-wellbore radial flow computationally efficient

60 km

60 k

m

Modeling of reservoir-scale observations can be used to assess effective permeability.

Pawar, 2007 (LA-UR-07-1358)

LA-UR-07-7795

Example application of CO2 -PENS: Impact of release through wellbores

Primary CO2 storage reservoir: 20 m thick

Impermeable layer : 200 m thick

Permeable layer : 5 m thick

Injection well

Hypothetical Sequestration Site

Pawar, 2007 (LA-UR-07-1358)

LA-UR-07-7795

In reservoir

50 years

In top aquifer

Time (years)

Mas

s of

CO

2(%

of i

njec

ted

CO

2)

CO2 Fraction near Wellbore in Hypothetical Storage System (Base Case)

Pawar, 2007 (LA-UR-07-1358)

LA-UR-07-7795

Numerous variables can impact the processes that control wellbore integrity.

• Hydration State of Cement (porosity, permeability, mineralogy)

“slurry” characteristics(cement type; admixtures; fluids; water:cement ratio; drilling muds)hydration conditions (reservoir fluids; P, T, time)

• Physical State of Wellbore Systemwellbore fractures and flow pathwayseffective permeability of wellbore(cement permeability; flow along interfaces and fractures)materials properties (rock, casing, hydrated cement)P, T, …

• Chemical and Biological State of Wellbore Systemfluid chemistry (PCO2, T, brine composition, reservoir rocks, …)sulfur cycle?

LA-UR-07-7795

Cements used in well completions have varied over time.

1975 1950 1925190018751850 2000

first oil well

first portland cement in U.S.

first use of cement slurry

no additives; 1 cement type3 additives; 2 cement types

38 additives; 8 cement types44 additives; 4 cement types

first use of pozzolansfirst standards committee

after Smith (1976)

LA-UR-07-7795

Additives can have a profound impact on cement properties.

Accelerating agentssalts (CaCl2 , NaCl)

High density additivesbaritehematite

Dispersing agentssulfonateslignin

Retarding agentslignosulfonatesphosphonatesCMHECdiesel

Low density additivesbentonitesNa-silicatespozzolans (e.g., fly ash,

tuff, diatomaceous earth)walnut shellscoalcellophane

LA-UR-07-7795

Reservoir fluid chemistries can impact the hydration of cement and subsequent carbonated-brine interactions.

LA-UR-07-7795

Summary

• Wellbores are critical components of a CO2 storage systemPlacement of CO2Potential release pathways from reservoirPotential conduits/fastpaths for CO2 movement

• Predicting the integrity of wellbores/cement requires integration of experiments, field observations, theory to upscale fundamental processes to system level

field, field, fieldmore observations needed

• In some examples, CO2 +brine+cement interactions may improve integrity of wellbore