Geochemical Implications of Gas Leakage associated with Geologic CO 2 Storage—A Qualitative Review

Geochemical Implications of Gas Leakage associated with GeologicCO2 StorageA Qualitative ReviewOmar R. Harvey, †,‡,* Nikolla P. Qafoku, † Kirk J. Cantrell, † Giehyeon Lee,§ James E. Amonette,∥

and Christopher F. Brown †

†Geosciences Group, Pacific Northwest National Laboratory, 902 Battelle Blvd, K6-81, Richland, Washington 99354, United States‡Department of Geography and Geology, The University of Southern Mississippi, 118 College Drive, #5051, Hattiesburg, Mississippi,39406, United States§Department of Earth System Sciences, Yonsei University, 134 Shinchon-dong, Seodaemun-gu Seoul 120-749, Korea∥Chemical and Materials Sciences, Pacific Northwest National Laboratory, 902 Battelle Blvd, K8-96, Richland, Washington 99354,United States

ABSTRACT: Gas leakage from deep storage reservoirs is a major risk factorassociated with geologic carbon sequestration (GCS). A systematic under-standing of how such leakage would impact the geochemistry of potableaquifers and the vadose zone is crucial to the maintenance of environmentalquality and the widespread acceptance of GCS. This paper reviews the currentliterature and discusses current knowledge gaps on how elevated CO2 levelscould influence geochemical processes (e.g., adsorption/desorption anddissolution/precipitation) in potable aquifers and the vadose zone. Thereview revealed that despite an increase in research and evidence for bothbeneficial and deleterious consequences of CO2 migration into potableaquifers and the vadose zone, significant knowledge gaps still exist. Primaryamong these knowledge gaps is the role/influence of pertinent geochemicalfactors such as redox condition, CO2 influx rate, gas stream composition,microbial activity, and mineralogy in CO2-induced reactions. Although these factors by no means represent an exhaustive list ofknowledge gaps we believe that addressing them is pivotal in advancing current scientific knowledge on how leakage from GCSmay impact the environment, improving predictions of CO2-induced geochemical changes in the subsurface, and facilitatingscience-based decision- and policy-making on risk associated with geologic carbon sequestration.

■ INTRODUCTION

The capture and storage of CO2 in deep geologic formations (orgeologic CO2 sequestration) is widely considered a feasibleapproach to reducing industrial loadings of greenhouse gases tothe atmosphere.1−5 Although oil and gas reservoirs6−8 andunmineable coal seam formations9−11 have been identified aspotential geologic repositories for CO2, deep (often saline)nonpotable aquifers are preferred and are the most widelystudied. Reasons include ubiquity, availability of maturetechnology, high storage capacities, and potential for CO2

conversion to carbonate minerals.3,12−14 Estimates for CO2

storage capacity in a single deep nonpotable aquifer range from10−2 to 104 Gt2,13−17 and would be sufficient for storing decadesto centuries of future CO2 emissions.14

Despite apparent promise, a major risk factor and potentialbarrier to widespread deployment of geologic CO2 sequestrationis leakage of gas from the storage aquifer.4,18−21 Shaffer18

suggested that gas leakage of 1% or less per thousand years from astorage reservoir, or continuous resequestration, would berequired to maintain atmospheric CO2 concentrations close tothose projected for alternative approaches (e.g., lowering globalemissions by 2050 to 60% compared to 1990). Concerns with gas

leakage may be considered globally or locally.4,19 On a globalscale, increased atmospheric CO2 concentration due to leakageof previously sequestered CO2 is of greatest concern. However,current trends in storage assurances suggest that a significantincrease in atmospheric CO2, or any subsequent effects onclimate change, due to leakage from geologic storage is unlikely.For example, the International Panel on Climate Change notedthat with respect to global risk, the fraction of CO2 retained inappropriately selected and managed reservoirs is likely to exceed99%.4 Such storage assurances are well within the 60−95% CO2

retentions suggested to make impermanent storage valuable forthe mitigation of climate change. 4

Locally two gas leakage scenarios are of concern.4,19 The first iswhere there is a sudden, fast and short-lived release of gas aswould occur in the case of well failure during injection orspontaneous blowouts.22−25 In general, this leakage scenario is

Special Issue: Carbon Sequestration

Received: July 19, 2012Revised: October 18, 2012Accepted: October 23, 2012Published: October 23, 2012

Critical Review

pubs.acs.org/est

© 2012 American Chemical Society 23 dx.doi.org/10.1021/es3029457 | Environ. Sci. Technol. 2013, 47, 23−36

pubs.acs.org/est

anticipated to be relatively rare (on the order of 1 every 105 wellsyear−1) with environmental damage confined to the vicinity ofthe accident.22,23 However, in cases where CO2 concentrationsare high enough there is the potential for loss of animal andhuman life.19,26 The second local leakage scenario is where theleak is more gradual occurring from abandoned wells, alongundetected faults, fractures or the injection-well lining.27 It is thistype of leakage that is of greatest concern because, withoutrigorous monitoring, such diffusive leakage may go undetectedfor prolonged periods of time, and has the greatest potential tocause broad scale environmental impacts including decreases indomestic, agricultural, or industrial water quality and soilfertility.4,19,24

In this paper, the current literature on the geochemicalimplications associated with gas leakage from deep CO2 storageformations to near-surface environments is reviewed. Emphasisis placed on CO2-induced effects on dissolution/precipitationand adsorption/desorption reactions in the near subsurface, andpotential beneficial or deleterious consequences on the geo-chemistry of potable aquifers or the vadose zone. Gaps in currentknowledge and research needs related to geochemicalimplications of CO2 intrusion into the near subsurface are alsoidentified and discussed.

■ CARBON DIOXIDE IN SUBSURFACEENVIRONMENTS

In deep geologic formations, suitable for CO2 sequestration, thetemperature and pressure conditions typically exceed those ofthe critical point of CO2 (31.1 °C/7.38 MPa). Carbon dioxide instorage aquifers will therefore exist primarily as a supercriticalfluid.19,27,28 Several mechanisms for the sequestration of thissupercritical CO2 (scCO2) in deep aquifers include geologictrapping through physical containment by geologic features,solubility trapping through dissolution in the formation water orresidual oil, mineral trapping through the formation of carbonateminerals, hydrodynamic trapping due to differences in viscositybetween the CO2 plume and formation water, and capillarytrapping where CO2 is held in formation due to capillaryforces.3,29 In uneconomical coal-bed formations, sorption of CO2to the coal surface is the primary sequestration mechanismidentified.30−32

It is difficult to ascertain how much CO2 is sequestered by aspecific mechanism; however, estimates of coal-bed storagecapacity suggest accessibility to sorption sites of less than 60% forCO2 sequestration.30,32 For deep aquifers/reservoirs, geologictrapping of CO2 is expected to be dominant during the earlyyears with solubility, hydrodynamic and mineral trappingbecoming increasingly significant with time.4,19,33 Severalstudies34−37 suggest that even after centuries of operation, CO2sequestration via mineral trapping might be extremely small (lessthan 0.5%) in sedimentary formations.Within 2 km of the surface, scCO2 will be less dense than the

formation water.28 Any CO2 not sequestered via solubility ormineral trapping would therefore be expected to form a buoyantscCO2 plume within the storage reservoir. The distribution ofCO2 between the supercritical plume and aqueous phase (CO2sequestered via solubility trapping) will depend on reservoirtemperature and pressure, as well as the chemical composition ofthe formation water.27,28,38,39 The reactivity of both phasestoward geologic and man-made materials (e.g., cement and steelused in well construction) has been extensively studied underconditions consistent with deep storage formations.40−44

Although, results from some of these studies will be mentioned

throughout this paper, reactions in the deep storage reservoir arenot the primary focus of this review. Gaus45 recently reviewed theliterature pertinent to CO2-rock interactions in deep geologicformations suitable for CO2 sequestration.It is the CO2 that would migrate diffusively (in the case of a

leak) from the storage aquifer, through various leakage paths(e.g., fractures, faults or well-bores) and into overlying potableaquifers or the vadose zone that is of primary concern in thisreview. Celia and Nordbotten27 provide a good overview of theseleakage pathways and their significance. In contrast to deepstorage reservoirs, temperature and pressure conditions in nearsurface environments, where most potable water aquifers can befound, are below the critical point of CO2. Under these subcriticalconditions, CO2 exists predominantly in the gaseous phase. Inthe event that CO2 migrates from deep storage, it is the lowerpressure and temperature conditions (associated with nearsurface environments) that drive the transition of scCO2 to CO2gas.The fate of the leaked CO2 gas will depend largely on the

physical and chemical characteristics of the receiving subsurfaceenvironment. From a physical perspective, the presence of arestricting layerin the near subsurface (e.g., a confined aquiferor thick aquitard)reduces the risk of CO2 migrating back intothe atmosphere. Otherwise, a significant amount of the leakedCO2 may eventually diffuse/migrate into the atmosphere.46,47

Klusman47 estimate that about 170 tons of CO2 is lost annuallythrough leakage from deep storage to the atmosphere, at a CO2-EOR site in Rangely, Colorado. If a confined aquifer or thickaquitard was present above the injection zone and reservoircaprock at Rangely, losses of CO2 to the atmosphere would likelybe significantly less due to physical containment of the gas, aswell as continued dissolution of CO2 into the groundwater.From a chemical perspective, the partitioning of the CO2 into

the aqueous phase is crucial in determining the fate and impact ofleaked CO2 in the near subsurface. In fact, it is the effect of theaqueous phase CO2 on aqueous phase pH, and subsequent rock/mineral-solution interactions that drives current thoughts onhow the leakage of CO2 from deep storage reservoirs wouldimpact the geochemistry of near-surface environments. In thefollowing sections, we synthesize the current literature on howthe partitioning of CO2 into aqueous solution and associatedchanges in aqueous pH may effect beneficial or deleteriousgeochemical changes in potable aquifers and the vadose zone.

Partitioning of CO2 into Solution and Changes inAqueous Phase pH. The dissolution of CO2 in water to formcarbonic acid, and its subsequent dissociation, is known to causea decrease in pH as a result of aqueous phase proton enrichment:

+ ↔ ↔ +− +CO (g) H O H CO HCO (aq) H (aq)2 2 2 3 3(1)

Changes in pH and the production of HCO3− will influence or

control the dissolution of minerals and the subsequent release ofchemical elements and contaminants into the aqueous phase, aswell as precipitation reactions and formation of neophases. Inaddition, these changes may significantly and even dramaticallyaffect the extent and rate of chemical, biological, and hydrologicalprocesses and reactions, which may control contaminantmobility in the subsurface.Experimental andmodeling studies suggest that CO2 intrusion

into the vadose zone or potable aquifers may induce a decrease inaqueous pH on the order of 1−3 units.48−56 For a given system,the magnitude of the decrease in pH will likely be a function ofthe solubility of CO2 in the aqueous solution, and the buffering

Environmental Science & Technology Critical Review

dx.doi.org/10.1021/es3029457 | Environ. Sci. Technol. 2013, 47, 23−3624

capacity of the system. The solubility of CO2 in aqueous solutionis known to decrease with increasing temperature and solutionionic strength, but increase with pressure.38,57 Hence, for systemsof a given buffering capacity, decrease in the pH at the lower endof the range (1−3 pH units) would be favored under conditionsof lower CO2 solubility (e.g., warm, shallow potable aquifers ormore saline arid soils). On the other hand, ceteris paribus, agreater magnitude of pH change would be expected in cooler/deeper aquifers and nonsaline soils.For systems with similar CO2 solubility, a decrease in pH

closer to 1 pH units would be typical of well buffered systems,where CO2-induced dissolution of reactive carbonates (eq 2),feldspars (eq 3), and/or the precipitation/dissolution of clays(eqs 3 and 4) would provide enough buffering capacity (viaHCO3

− alkalinity) to resist drastic changes in pH.

+ + → ++ −CaCO CO (g) H O Ca 2HCO3 2 22

3 (2)

+ +

→ + + ++ −

2NaAlSi O 11H O 2CO (g)

Al Si O (OH) 2Na 2HCO 4H SiO3 8 2 2

2 2 5 4 3 4 4(3)

+ +

→ + ++ −

Al Si O (OH) 5H O 6CO (g)

2Al 6HCO 2H SiO2 2 5 4 2 2

33 4 4 (4)

Poorly buffered systems (e.g., sandy soils/aquifers) are devoid ofsufficient quantities of alkalinity-producing minerals and there-fore lack the ability to resist changes in pH. A major concern forsystems with high CO2 solubility, and/or low buffering is that anygeochemical change due to CO2 intrusion is likely to be moreapparent and the risk for pH-induced perturbation to environ-mental quality is more significant and prolonged compared towell buffered systems.13,52,54

■ BENEFICIAL CO2-INDUCED INTERACTIONS IN THENEAR SURFACE ENVIRONMENTS

The migration of CO2 from deep geologic storage into nearsurface environments could be considered beneficial if it resultedin resequestration (the trapping of CO2 that was previouslystored in deep storage) and/or reduced mobility andbioavailability of contaminants.CO2 Resequestration. Estimates of up to 96% retention of

leaked CO2 over a 1000 year period have been reported in thevadose zone.46 Processes identified for such CO2 resequestrationin the subsurface include the accumulation of the CO2 gas at thewater table (due to CO2 being denser than soil air), permeabilitytrapping (due to anisotropy favoring the horizontal flow of CO2)and solubility trapping (due to dissolution of CO2 intoinfiltrating or residual soil water).The mineralogical trapping of CO2 in carbonate minerals

provides another potential mechanism for CO2 resequestrationin near surface environments. From an environmental andsequestration perspective, this would be the most desirableoutcome for CO2 intrusion into near surface environments,because of the immobility of carbonate minerals compared toother CO2 trapping mechanisms. Several studies have demon-strated mineralogical trapping of CO2 under conditionsassociated with deep geologic CO2 storage, but such assessmentshave not been widely considered in near surface environmentsdue to thermodynamic and kinetic limitations.40,46,58,59

Thermodynamic limitations to carbonate precipitation in thenear surface environments are due to the fact that many soil

solutions and potable groundwater are undersaturated (satu-ration index, SI < 0) with respect to most carbonate minerals(except calcite).60,61 Even in cases where soil solutions orgroundwater are supersaturated, some carbonates will notprecipitate due to kinetic limitations. Magnesite and dolomiteare well-known examples.62−64 Saldi et al.63 estimated that for SI= 1 (10 times saturation) the growth rate of magnesite at 25 °Cwas at least 6 orders of magnitude lower than that of calcite underthe same conditions. They further estimated that it would take atleast 340 000 years to precipitate a 1 mm layer of magnesite frombulk solution.Jimenez-Lopez et al.65 found a similar situation for the

precipitation kinetics of siderite (FeCO3) compared to calcite.They found that for a given SI the precipitation rate of FeCO3 at25 °C was about 8 orders of magnitude lower than that of calciteunder the same conditions. The higher ion surface-chargedensities of Fe2+ and Mg2+ relative to Ca2+ have been used bySaldi et al.,63 Jimenez-Lopez et al.65 (and references therein) toexplain the markedly different precipitation rates between calcite(which forms readily at low temperatures) and other carbonates.The smaller ionic radius of Fe2+(0.064 nm) andMg2+(0.065 nm),compared to Ca2+ (0.074 nm), means that a higher activationenergy is required to initiate dehydration of Fe2+ and Mg2+, andsubsequently the precipitation of siderite, magnesite or dolomite.Activation energies of 38, 129, and 159 kJ mol−1 have beenreported for the formation of calcite, siderite, and magnesite,respectively at 25 °C.63,65

The preceding discussion on thermodynamic and kineticlimitations to carbonate formation may raise some questionsabout the validity of mineral trapping as an important mechanismin CO2 resequestration in near surface environments. There ishowever evidence to suggest that some level of discussion iswarranted. First, while some carbonates do not form readilyunder conditions consistent with near surface environments,precursors of these minerals are known to form readily underthese conditions. For example, there is an increasing number ofstudies to suggest that nesquehonite (MgCO3•3H2O) formsreadily in low temperature CO2-rich solutions that aresupersaturated with respect to magnesite.66−68Jimenez-Lopezet al.65 also noted a series of metastable precursors that formedreadily and eventually lead to the formation of well-crystallizedsiderite. It is therefore plausible that these precursors could play asignificant role in mineral trapping in near surface environmentsimpacted by CO2.A second reason to consider mineral trapping mechanisms in

near-surface environments is the evidence that microorganismsare capable of overcoming the thermodynamic and kineticbarriers to produce carbonates in systems where they would notbe predicted on the basis of temperature or saturationindex.69−71Roberts et al.71 observed and demonstrated thatmethanogens were able tomediate the formation of dolomite in ashallow freshwater basalt aquifer on time scale of weeks tomonths. They suggested that the methanogens overcame thethermodynamic barrier by releasing Mg2+ and Ca2+ from thebasalt. This was consistent with the findings of Kenward et al.,69

who showed that methanogenesis increased the dolomitesaturation state of their solutions by 2 orders of magnitude.Roberts et al.,71 Kenward et al.,69 and other researchers72,73 allsuggest that carbonate-forming microorganisms overcome thekinetic barrier by using their cells as a seeding/nucleation surface.The presence of a suitable seeding surface is known to increasecarbonate precipitation rates by up to 3 fold.74The high pCO2 in



CO2-impacted environments could also plausibly enhancemicrobially mediated carbonate formation.75

According to Gislason et al.,15 the rate of dissolution for silicateminerals will be the limiting step in carbonation/mineral

trapping. Since the dissolution of a rock is likely to be a functionof its composition and crystallinity, its reactive surface area, andtemperature and composition of the solution phase15then itfollows that a broad array of carbonation environments are

Table 1. Summary of Modeling and Experimental Studies Examining the Impact of CO2 Intrusion on the Geochemistry of PotableAquifers and the Vadose Zone

descriptiona key findings on CO2 effects

ModelingStudies

Altevogt andJaffe48

•2-D variably saturated hypothetical vadose zone exposed to CO2 for up to 40 days •potential mobilization of toxic metals, due to acidification of thevadose zone

•variables: pH, O2(aq), organic matter, NO3−, NH4

+, Mn(II), Mn(IV), Fe(II), Fe(III),SO4

2−, H2S, H2CO3

•decrease in organic matter degradation rates, and oxidized species,due to CO2 displacing O2

Vong et al.51 •2-/3-D glauconitic-sandstone aquifer (with trace metal - bearing sulfides added) exposed toCO2 over 10 years

•predicted acidification increased dissolution of metal- bearingsulfides and subsequently aqueous concentration of trace metals

•variables: pH, Cd, Pb, Zn

Wang andJaffe52

•2-D buffered/unbuffered hypothetical potable aquifers with galena (PbS) as source of Pband exposed to CO2 for 8 years

•increased aqueous Pb from acidic dissolution of galena

•variables: pH, total carbonate, Pb •magnitude of change in pH/aqueous Pb controlled by carbonatesbuffering

Zheng et al.55 •3-D Eastern Coastal Plain model aquifer exposed to CO2 for 100 years •acidification increased galena and arsenian pyrite dissolutionresulting in increased aqueous Pb and As

•variables: pH, Pb, As •aqueous As regulated by suitable sorbents

Jacquemetet.al.89

•similar to Vong et al.51 (without sulfides) but with SOx and NOx (as impurities in CO2 gasstream)

•increased aqueous Fe and Mn mineral dissolution

•variables: pH, Fe, Mn, SOx(aq), NOx(aq) • lower pH and increased aqueous Fe and Mn with SOx and NOx,(<2%) as impurities in CO2 gas stream.

Zheng et al.90 •modeling of field data from Kharaka et al.56 •Ca2+ and CO3− (from calcite dissolution) induced ion exchange

likely increased cations/anions concentration

•variables: pH, HCO3−, Ca, K, Sr, Cu, Zn, Fe, Se, Na, Cd, P, Pb •increased dissolution of reactive Fe-bearing minerals

BatchExperiments

Little andJackson49

•yearlong CO2-nanopure water-sediment study with sediments from aquifers, which overlaypotential geologic carbon storage sites in the U.S.

•aqueous concentrations of some species (e.g., Mn, Fe, Co, Ni, andZn) increased, others (e.g., Mo) decreased, and some remainedunaffected

•variables: pH, Li, Mg, Ca, Rb, Sr, Co, Se, Ba, U, As, Al, V, Cr, Mn, Fe, Ni, Cu, Zn, Mo, Cd, B •trends for changes in species concentrations may vary acrossaquifers

Lu et al.50 •sediments from representative potable aquifers within the US Gulf Coast region.Background electrolyte (∼1 mMNaCl) used as groundwater medium. Exposed to CO2 for2 weeks, preceded by Ar purging (2 weeks)

•“Type I” cations (e.g., Ca, Mg, Si, K, Sr, Mn, B, Zn) rapidlyincreased and reached stable concentrations

•variables: pH, Ca, Al, Zn, Mg, Fe, Cs, Na, B, Ni, K, Co, Rb, Si, Cu, As, Mn, Mo, Cr, Sr, U, Ba,V, Cl, SO4, Br, F, NO3

•“Type II” cations (including Fe, Al, Mo, U, V, As, Cr, Cs, Rb, Niand Cu) increased, but declined in most cases, to levels lowerthan pre-CO2 concentration

Wei et al.53 •variably saturated soils exposed to CO2 (at 25 bar) for 3 days •aqueous concentration of some species (e.g., Mg, K, Al, V, Cr, Mn,Fe, Co, Cu, Rb, Sr, Ba, Pb and U) increased, and others (e.g., Znand Cd) decreased

•variables: pH, Mg, K, Al, Ti, V, Cr, Mn, Fe, Co, Cu, Rb, Sr, Mo, Cs, Ba, Pb, Th, U, Zn, Cd •changes in metal concentration still tolerable by plants

Field Studies

Smyth et al.76 •laboratory (See Lu et al.50) and field study. Field study involved potable groundwatersamples from within and around a CO2/EOR site in Scurry County, Texas

•Lab experiments (Same as Lu et al.50)

•variables for field study: pH, broad suite of cations and anions, DIC, DOC, CH4, and CO2 •No evidence from field study to suggest mobilization of chemicalspecies due to CO2/EOR

•CO2/EOR initiated in 1972; most of approximately 7.5 × 107 tons of unrecovered CO2possibly trapped in subsurface

Kharaka et al.56

•CO2 injected, for 30 days, into the shallow groundwater at the ZERT field site in Bozeman,Montana

•increase in EC, TDS, alkalinity (as HCO3−) and some metals and

trace elements(e.g., Ca, Mg. Mn. Pb, As, Zn)

•variables: pH, EC, TDS, HCO3, Na, K, Mg, Ca, Sr, Ba, Mn, Fe, F, Cl, Br, NO3, PO4, SO4,SiO2, Al, As, B, Cd, Co, Cr, Cu, Li, Mo, Pb, Se, U, Zn

•no effect on some species including Na and K

aVariables were written “as is” in cited reference.



possible. For example, the higher divalent metal content ofultramafic and mafic rocks would make them better candidatesfor carbonation than a sandstone. Gislason et al.15 suggested thatsuch compositional variations can result in about 2 orders ofmagnitude difference in the rate of carbonation. Similarly, CO2intrusion into an aquifer is expected to cause a significant increasein the rate of dissolution for aquifer minerals. It is this potentialfor CO2-enhanced weathering of soil and aquifer minerals thatunderpins the third reason for discussing mineral trapping as apotential mechanism for carbon sequestration in near surfaceenvironments.A generalized reaction for the CO2-induced weathering of a

plagioclase mineral to carbonate is shown in eq 5 (modified fromMcGrail et al.13).

+ + +

→ + +

+•

+

n

n

2Na (M) AlSi O (10 )H O CO (g)

(M)CO H O Al Si O (OH) Na

4H SiO

0.5 0.5 3 8 2 2

3 2 2 2 5 4

4 4 (5)

where, M could be Ca2+, Mg2+, Fe2+ or a combination of Ca/Mgor Ca/Fe. The dissolution of CO2 in water is anticipated to causea decrease in solution pH leading to enhanced dissolution of theplagioclase and increased chemical activity of Ca, Mg, or Fe andHCO3

−. Estimates for CO2-induced increases in dissolution ratescan be around 2 orders of magnitude.15The increasing activity ofCa,Mg, or Fe andHCO3

− could lead to supersaturation (SI≫ 0)and precipitation of carbonates.Contaminant Immobilization. Contaminant immobiliza-

tion in subsurface systems may occur via adsorption,precipitation as separate neophases, precipitation on mineralsurfaces or incorporation into solid solutions or mineralassemblages. There is some modeling and direct experimentalevidence to suggest that the introduction of CO2 into potableaquifers or the vadose zone may reduce the mobility of somecontaminants.48,50,76 In simulating the effect of intruding CO2 onsoil biogeochemistry, Altevogt and Jaffe48 concluded that thedisplacement of O2 by CO2 in soil gas and the decrease in soilsolution pH (due to CO2 dissolution) would favor the reductionof soil NO3

− to NH4+. Because of the greater tendency for NH4

+

(compared to NO3−) to be adsorbed to negatively charged soil

minerals and organic matter, such CO2-induced changes innitrogen speciation would be useful in reducing nitrate mobilityin the vadose zone and contamination of groundwater inagricultural areas.Based on laboratory experiments, Lu et al.50 and Smyth et al.76

concluded that introduction of CO2 into a fresh water aquifercould reduce the aqueous concentration of some contaminants(including As, Cr, U and V). The behavior of these contaminantswas attributed to CO2-induced changes in dissolution ofcarbonate minerals and adsorption characteristics of organicmatter, clays and hydrous oxides.Recent studies show that, in addition to carbonates, the

enhanced weathering effect of CO2 on primary soil and aquiferminerals could also yield secondary clay minerals or metal(hydr)oxides.77−80 The formation of these new mineral phasescould serve to reduce the mobility of contaminants in near-surface environments via incorporation into the mineralstructure and/or sorption to the mineral surface. For example,the ability of secondary minerals (e.g., allophane,79 goethite 80)and highly weathered soils with pH-dependent positivelycharged reactive surface groups to adsorb oxyanion contaminants(e.g., nitrate, arsenate, arsenite, chromate, selenate) under acidic

conditions is well-known.81−84 In addition, clays (e.g., illite andsmectites 79) with permanent negatively-charged surfaces arealso known to adsorb cationic contaminants.85,86 In basalticsystems, the weathering effect of the intruding CO2 on basalts toform the ferruginous smectite minerals, such as nontronite, couldalso serve to enhance the immobilization of redox-sensitivecontaminants.87,88

■ DELETERIOUS CO2 INTERACTIONS IN NEARSURFACE ENVIRONMENTS

Degradation of water quality stemming from CO2-inducedreactions is one of the greatest concerns associated withmigration of CO2 gas from deep geologic storage to near surfaceenvironments. Such degradation may involve increased mobi-lization of contaminants or changes in other water qualityparameters such as alkalinity, salinity or total dissolved solids(TDS). Research in this area is relatively new, and at the time ofthis review, applicable work included several modeling studies,batch experiments, and short-term field studies (Table 1).

Mobilization of Contaminants. The potential for CO2-induced mobilization of contaminants has been reported inexperimental and modeling studies and has been attributedlargely to the activity the H+ and HCO3

−.Wang and Jaffe,52 modeled potential changes in aqueous Pb

concentration as a result of intrusion of CO2 into hypotheticalaquifers comprising of PbS/CaCO3or PbS/quartz. For an 8 yearsimulation, they predicted that a CO2-induced decrease in pHwould increase dissolution of PbS and the aqueous concentrationof Pb. Vong et al.51 modeled the potential for the intrusion ofCO2 into a glauconitic-sandstone aquifer system to induceexcessive release of heavy metals into solution. They alsopredicted that a decrease in aqueous pH associated with CO2intrusion into the aquifer would adversely affect potablegroundwater quality due to CO2-enhanced dissolution of galenaand other trace-metal bearing minerals resulting in increasedaqueous concentrations of Pb, Cd, Cu, Fe, Mn, and Zn. Similarpredictions were made by Zheng et al.55 who also considered thesignificance of sorption on the mobility of Pb and As in a typicalEastern United States coastal plain aquifer impacted by theintrusion of CO2 gas.Laboratory and field studies have also highlighted the potential

for CO2-induced changes in near-surface environments tonegatively affect water quality. Wei et al.53 found that, althoughstill tolerable by plants, exchangeable concentrations of a widerange of metals (including Cu, Pb, Cr, V, and U) in anagricultural soil increased by up to 500% during a three dayincubation period with CO2 (25 °C and 25 bar). Little andJackson49 tracked metal release from various aquifer sediments(exposed to CO2) and reported that CO2 intrusion resulted inincreased aqueous concentrations of several metals (e.g., Al, V,Cr, Mn, Zn, and Co) by up to 3 orders of magnitude. Kharaka etal.,56 during a 30 day field study, found that the injection of CO2into a shallow freshwater aquifer (2−2.3 m) resulted in increasedaqueous concentrations of Fe,Mn, As, Cu, Zn, Cd, Pb, Al, and Se.In all these cases, the increase in aqueous concentration of thecontaminants was attributed to enhanced dissolution and/ordesorption from the surfaces of soil and aquifer minerals.Wilkin and Digiulio54 cautioned that although the increased

HCO3− concentration resulting from weathering of soil and

aquifer minerals by dissolved CO2 may promote carbonateprecipitation it may also promote desorption of somecontaminants from mineral surfaces. The ability of HCO3

− todesorb negatively charged species of arsenic and naturally



occurring uranium from metal oxides, soils, and aquifersediments is well recognized.89−91 Mason et al.89 found that inthe presence of HCO3

−, 75−90% of the sorbed uranium could bedesorbed from contaminated soils. Anawar et al.90 found thatarsenic desorption from Bangladeshi aquifer sediments wasenhanced in the presence of HCO3

−. Saalfield and Bostick91

found that desorption of As from ferrihydrite increased by 1−3orders of magnitude with increasing HCO3

− concentrations.Other Water Quality Parameters. In addition to decrease

in pH and enhanced mobilization of contaminants, the intrusionof CO2 into near surface environments can also induce changesin other pertinent water quality parameters such as alkalinity,salinity/TDS and total metal concentration. Kharaka et al.56

reported that in addition to a change in pH (from 7.0 to 5.6), theintroduction of CO2 into a shallow groundwater aquifer over atwo week period caused HCO3

− alkalinity to increase from 400to 1300 mg/L, salinity (measured as electrical conductivity) toincrease from 600 to 1800 μS cm−1 and TDS from 600 to 1500mg/L. Kharaka et al.56 and other studies,49,50,76 have also shownthe potential for the aqueous concentrations of nontoxic metalssuch as Ca, Mg, K, Ba, and Na to increase significantly inresponse to CO2 introduction. Excessive amounts of thesemetals, although not a serious environmental risk, couldcompromise potable water quality due to their contributions tototal dissolved solids.

■ KNOWLEDGE GAPS AND RESEARCH NEEDS

Despite an increase in research and some evidence to supportboth beneficial and deleterious impacts of CO2 intrusion on thegeochemistry of near surface environments, there are severalcritical aspects of research that are still lacking. Primary amongthese knowledge gaps is the role of pertinent geochemical factorssuch as CO2-induced shifts in redox condition, gas intrusion rate,composition of the intruding gas stream, organic constituents/microbial activity, and sediment mineralogy, in determiningadsorption/desorption and dissolution/precipitation in CO2-impacted aquifers or vadose zone. Although the above factors byno means represent an exhaustive list (and could be expanded to

include physical factors such as mixing/dispersion, contact areaand residence time) we believe that a systematic evaluation oftheir influence on adsorption/desorption and dissolution/precipitation dynamics in CO2-impacted systems, as well as thedevelopment of rigorous experimental/assessment protocols,would represent crucial advances in scientific-knowledge,improve predictability of geochemical outcomes of CO2intrusion into near-surface environments, and facilitate betterscience-based decision- and policy-making.

Why Is Redox Condition Important? Like pH, the redoxcondition (often measured as oxidation−reduction potential,Eh) is known to control many geochemical processes in thesubsurface. Eh is especially important in understanding thegeochemical behavior of redox-sensitive minerals/elements. Forexample, sulfide minerals (a likely source of, or sink for, tracemetal contaminants in the subsurface) are known to exhibit verydifferent dissolution/precipitation characteristics under anaero-bic (low Eh) versus aerobic (high Eh) conditions.92,93 Similarly,the speciation of redox sensitive elements such as As, Fe, Cr, Mn,U, N, and S and subsequently their precipitation and/oradsorption/desorption behavior will be dictated by Eh (Figure1). However, in contrast to pH, Eh is far less prevalent as aquantitative variable in current studies concerning the geo-chemical effects of CO2 intrusion into the vadoze zone or potableaquifers. Instead, these studies (e.g., Zheng et al.55 and Vong etal.51) attribute CO2-induced geochemical chemical changes inthe subsurface solely to changes in solution pHeven in caseswhere the contaminants of interest are redox-sensitive orcontained in redox-sensitive minerals. Carroll et al.,94 inrecognizing the potential significance of changes in redoxconditions in CO2-impacted systems, noted that robustpredictions of the geochemical impact of a CO2 leak will requirethe coupling of CO2 plume modeling with laboratory experi-ments under a range of redox conditions.Work by Birkholzer et al.55,60 has shown that many potential

trace metal contaminants of concern (e.g., As, Pb, U, Hg, Cd,Zn), in potable aquifers across the United States are redox-sensitive and/or contained in redox-sensitive minerals (including

Figure 1. Eh-pH diagrams for (a) Fe and (b) As in a Fe−As−S−HCO3−H2O system. Points on the plot indicate the median pH and Eh of oxidized(open circle) and reduced (filled circle) groundwater in the contiguous U.S. Arrows indicate general direction of change in the event of CO2 intrusioninto a potable aquifer or the vadose zone. The area bound by the red broken line is representative of typical pH-Eh conditions in natural soils and potableaquifers. All data used in developing the figure were taken from Birkholzer et al.60 and references therein.



pyrite, galena, uraninite, sphalerite, and various solid solutionassemblages). The intrusion of CO2 gas into these aquifers, or thevadose zone, is likely to induce changes in redox condition due toCO2-induced reactions resulting in the redistribution of oxidizedand reduced aqueous species, O2 depletion/displacement withthe CO2 gas,48,95,96 or changes in biological activity.97,98 Theextent and direction of Eh change will control the extent and rateof mineral dissolution, precipitation of neophases, as well assorption/desorption of contaminants onto pertinent sorbentsuch as organic matter, clays, and metal (hydr)oxides.In reality, pH and Eh in the subsurface will be interrelated. This

is reflected in the fact that most redox reactions involve protons(oxidation releases H+ and reduction consumes H+) and willinduce a simultaneous change in solution pH. A change in pHwill also impact Eh due to pH effects on dissolution andsubsequently the distribution of reduced versus oxidized speciesin solution. The impact of CO2 on chemical speciation and thedissolution/precipitation characteristics of minerals in thesubsurface will therefore be a combined effect of Eh and pH.With respect to coupled changes in pH and Eh, one of twoscenarios are plausible in the event of CO2 intrusion into thevadose zone or a potable aquifer. Scenario 1 is where a decline inpH is accompanied by a decline in Eh (e.g., due to thedisplacement of O2 by CO2) and scenario 2 is where a decline inpH is accompanied by an increase in Eh (e.g., due to an increasein the aqueous activity of oxidized to reduced species). Figure 1suggests that while a CO2-induced change in pH of 1−2 units49,55and Eh of 0.1 V, would have little to no impact on Fegeochemistry (irrespective of scenario), similar changes in pHand Eh would favor a shift in As speciation fromHAsO2 toHAsS2(scenario 1; reduced groundwater) or HAsO4

2‑ to H2AsO4−

(scenario 1/2; oxidized groundwater). Such shifts in the chemicalspeciation of As have significant implication for its mobility,reactivity and toxicity in the environment.99,100Shifts inspeciation, resulting in changes in the mobility, reactivity ortoxicity of other redox-sentive contaminants (such as Cr and U)are also plausible outcomes of CO2-induced changes in the pHand Eh of the vadose zone or potable aquifers. We contend that,along with pH, it is crucial that Eh be routinely considered as amajor variable in studies seeking to decipher how thegeochemistry of aquifers and the vadose zone would likely beimpacted by CO2 migration from deep storage reservoirs.Why is CO2 Intrusion Rate Important? The flux of the

migrating gas stream into a potable aquifer or the vadose zonewill influence the rate at which the system pH and Eh changes.The rate at which the pH and Eh of the system changes willdirectly impact mineral dissolution/precipitation kinetics,101,102

the nature and properties of the neophases precipitated,103 aswell as the kinetics of (or tendency for) adsorption/desorptionreactions (to occur).104The strong dependence of the dissolutionand precipitation rate, for many minerals, on pH and Eh dictatesthat gas streamflow rate would be crucial in predictingcontaminant release, how rapidly potential sorbents may formor, how rapidly a systemmay become supersaturated with respectto carbonate minerals. Similarly, the often strong dependence ofsorbent surface chemistry, and aqueous speciation of contami-nants, on pH and Eh also dictates that CO2 intrusion rate willimpact the fate and transport of contaminants in CO2-impactedaquifers or the vadose zone.One criticism of current experimental studies is that the rate of

CO2 intrusion used is unrealistically high and would overwhelmany chemical buffering provided by the minerals; as aconsequence, these studies yield unrealistic mineral dissolution

rates and trace element concentrations.105 Although such anargument is refutable on the basis that the solubility of CO2 inwater is limited and excess CO2 in the experiments would havebeen vented,49 the reality is that no experimental data currentlyexist to show how CO2 intrusion rate would likely influencegeochemical changes in a potable aquifer or the vadose zone. Inour search of the literature, we found that only the modelingstudies by Zheng et al.55 and Vong et al.51 considered the likelyimpact of intrusion rate on aquifer geochemistry. In both studies,it was concluded that increasing CO2 intrusion rates led to lowerpH which, enhanced dissolution of sulfide minerals such asgalena (PbS), sphalerite (ZnS) and arsenopyrite (FeAsS) andsubsequently compromised groundwater quality due toincreased aqueous concentrations of trace-metals.Based on the results of Zheng et al.55 and Vong et al.51, a

reasonable argument could be made that increasing CO2intrusion rates, resulting in the lowering of pH, could enhancethe dissolution of aquifer minerals and increase aqueousconcentrations of trace metals in groundwater. However, basedon the Nernst equation,106 it is also plausible (at leastanecdotally) that a decrease in pH could result in an increasein Ehdue to a redistribution of aqueous species favoringoxidized over reduced species. Hence, if the increase in Eh wassuch that the oxidation of aqueous Fe2+ and subsequentprecipitation of Fe-oxides was favorable, then the fate andmobility of anionic contaminants (e.g., arsenate) would becontrolled by the rate of change in Eh as dictated by the CO2intrusion rate. Understanding the potentially delicate balancebetween CO2 intrusion rate, system properties, pH, Eh,precipitation/dissolution, and adsorption/desorption is there-fore vital to assessing geochemical impact of CO2 on theenvironment.

Why Is Gas Stream Composition Important? Variationsin the composition of the intruding gas stream will influence theextent to which the pH and Eh of the system change in responseto gas leakage, and subsequently, the extent to which a gas-leakage-induced reaction would proceed. Variations in intrudinggas streams are likely because of differences in the composition ofinjected gas streams, or the composition of native reservoir gas.Such differences also dictate that the composition of intrudinggas streams will vary between sequestration sites and from that ofthe food-grade CO2 used in many experimental studies.Determining the identity/amount of co-injected (or nativereservoir gases) likely to be present in an intruding gas streamand understanding how their combined effect with CO2influences the geochemistry of the subsurface is thereforeimportant to understanding the impact of CO2 leakage onpotable aquifers or the vadose zone.The co-injected gases most likely to be present in an intruding

gas stream will depend on preinjection gas capture/treatmentactivities, and/or regulations on gas injections. In casesemploying postcombustion CO2 capture technology, andwhere regulations allow for the co-injection of limited quantitiesof other gases, studies will need to consider the combined effectof CO2 with other acid gases such as H2S, NOx, and SOx. Theoccurrence of any significant quantities (>2%)of H2S, NOx, andSOx in an intruding gas stream is unlikely where regulationsprohibit their co-injection and/or alternative capture technolo-gies to postcombustion (e.g., precombustion or oxy-combus-tion) are employed. In such cases any co-injected gas present inthe intruding gas stream would have been derived from thepreinjection separation/stripping activities. For example, NH3



and some volatile organic compounds are known to be producedduring amine-based gas stripping/separation activities.107

At the time of this review only the modeling efforts ofJacquemet et al.108 and Gislason et al.15 were found to considerthe potential effects of co-injected gases on near surfaceenvironments. Jacquemet et al.108 predicted that presence ofco-injected SO2 (∼1.5%) and NO (∼0.5%), could lower pH byan additional 1 unit (compared to CO2 alone) and increase therelease of metals from aquifer minerals. Such a finding isconsistent with the acidification effects associated withdissolution of NOx and SOx in water,109−111 but ignores anypotential beneficial consequences of NOx and SOx. For example,there is some evidence to suggest that aqueous SOx and NOxspecies may serve as an important oxidant in microbiallymediated mineral precipitation.112−116 Gislason et al.15 alsonoted that the presence of SO4 (from SO2 oxidation) and F(from HF in flue gas) may serve to enhance mineralogicaltrapping through the complexation of Al3+ and the enhancedissolution of Mg2+, Ca2+, and Fe2+ bearing alumino-silicates.Accounting for such beneficial outcomes, in terms of thepotential for mineralogical trapping of CO2 or contaminantimmobilization, is important and highlights the need forsignificant experimental research to validate current modelingefforts as well as assess both beneficial and deleterious outcomesof other co-injected gases (including those from strippingactivities).Studies of gases in the subsurface suggest that geologic

formation suitable for CO2 storage are likely to contain nativegases that have different partitioning/solubility characteristicsthan CO2. Kharaka et al.

39 found that gas samples collected fromaround 1500 m in the Frio formation, Texas contained 95 ± 3%CH4 with the bulk of the remaining gas comprised of N2, C2, andhigher hydrocarbon gases. High concentrations of CH4 and H2have also been observed in basaltic aquifers down to 1200 m.117

Any leakage of injected gas from the CO2 storage reservoir willtherefore occur in conjunction with these native reservoir gases.Considering that native aquifer gases will exhibit very differentgeochemical behavior than CO2, accounting for their potentialimpact on geochemical changes associated with leakage of CO2from deep storage aquifers is also important. To our knowledge,none of the current studies have attempted to address this issue.Establishing research looking at native reservoir gases and howtheir combined effect with CO2 is likely to influence subsurfacegeochemistry is pivotal. The establishment of research thatconsiders native gases, their transformation and subsequentimpact on potable aquifer and vadose zone geochemistry istherefore very important.Why is Microbial Activity Important? In the event of gas

leakage from geologic storage, microbial activity will impactsubsurface geochemistry largely through its influence on the fateof the gases comprising the intruding gas stream, changes in pHand Eh, and subsequently, changes in dissolution/precipitationor adsorption/desorption reactions.Several recent studies have considered microbial activity in the

storage reservoir,97,118,119 but to our knowledge, only themodeling effort of Onstott98 and a preliminary assessment byJones et al.120 have considered how CO2 leakage might influencemicrobial life in the near subsurface. However, several additionalstudies support the need for systematic assessments of the role ofmicroorganisms and microbial activity on the geochemicaloutcomes of gas migration in potable aquifers or the vadose zone.For example, Klusman47,121 found that most of the CH4 (andother hydrocarbons) in soils above exploited oil fields, will

undergomethanotrophic oxidation (to produce CO2) under oxicconditions. In the event of gas leakage at a CO2 sequestration site,the methanotrophic transformation of native hydrocarbonswould increase CO2 concentration in a migrating gas streamand enhance its impact on geochemical processes in thesubsurface.From the perspective of CO2 sequestration, many micro-

organisms likely to be encountered in near-surface environmentsabove geologic sequestration sites (e.g., sulfate-reducers, nitrate-reducers, iron-reducers, urea-degraders, and methanogens) arecapable of biogenic carbonate formation.71−73,122−125 Forexample, the formation of carbonate minerals by urea-degradershas been extensively studied.122−125 In a recent study122 it wasshown that, under conditions consistent with CO2 storageconditions, urea-degraders incorporated up to 37% of totalcarbon into carbonate minerals. Since the high pressureconditions of deep storage formations could significantly inhibitcellular activity126 a higher incorporation of CO2 into carbonatesmight be expected in near surface environments.Roberts et al.71 found that iron-bearing dolomite formed after

3 months of suspending sterilized Columbia River basaltfragments into a fresh water aquifer (to a depth of 7 m) wasassociated with the colonization of the basalt by a consortium ofmethanogens and fermenters. Dolomite formation was sug-gested to occur via local microbially enhanced weathering of thebasalt coupled with methanogenesis (with dissolved CO2 as asource of carbon) and crystal nucleation on the cell walls of themicroorganism (eq 7).

+ + + +

→ + + +

− + + +

+

3HCO 4H Ca 0.9Mg 0.1Fe

Ca Mg Fe (CO ) CH 3H O H3 2

2 2 2

1.0 0.9 0.1 3 2 4 2 (7)

Interestingly, there is also ample evidence in the literature tosuggest that the CH4 produced in eq 3 could be reoxidized viaanaerobic methane oxidation (AMO) by reducing bacteria(including iron-, nitrate- and sulfate reducers), resulting infurther biogenic carbonate formation (eq 8).112−116

+ + → + + +− + − +CH SO M MCO HS H O H4 42 2

3 2(8)

where M2+, could be any carbonate forming divalent metal (e.g.,Ca2+, Mg2+, or Fe2+). If M2+ was Cd2+ or Pb2+, biogenic carbonateformation could also potentially serve to enhance naturalattenuation of toxic metals through the formation of CdCO3and PbCO3.Theoretically, eq 7 suggests that mineral trapping alone could

account for 66% of dissolved inorganic carbon. In cases where eq8 is thermodynamically and kinetically favorable, and if thesystem is not limited by oxidant (e.g., SO4

2‑, NO3−, or Fe3+) or

dissolved metal (e.g., Ca2+, Mg2+, or Fe2+) concentrations, thenumber would be closer to 100%. It is therefore plausible that atgeologic sequestration sites where biogenic carbonate formationis favorable, mineral trapping could be very important indetermining the fate of leaked CO2 in the near surfaceenvironments. At potential geologic sequestration sites (e.g.,enhanced CO2-EOR operations, exploited oil reservoirs orbasaltic formations) where anaerobic subsurface microbialecosystems have been shown to be comprised of significantquantities of methanogens and reducing bacteria,117,127,128

microbial activity coupled to biogenic carbonate formationshould be considered. From a contaminant mobilizationperspective, there is also some evidence to suggest that microbialmetabolism of migrating gases may result in enhanced



mobilization of some contaminants.129,130 Considering multifoldeffects of the impact of migrating gases on subsurface microbialpopulations and subsequently microbial-mediated processes istherefore important.Why Is Sediment Mineralogy Important? Geology and

subsequently mineralogy at geologic sequestration sites will varyon an intra- and intersite basis. In the event of a CO2 gas streamintruding into a potable aquifer or the vadose zone, it is themineralogy of the system that will dictate how well the pH or Ehis buffered, the type and amount of contaminants likely to bemobilized, and what neophases/sorbents are likely to precipitate.A systematic understanding of the minerals present, their relativedistributions and their chemical and physical behavior is crucialto assessing the impact of CO2 on near surface environments.In many of the studies reviewed, the significance of mineral

type to outcomes of CO2-intrusion was apparent. However, thequestion on how geochemical outcome is influenced by mineraldistribution/heterogeneity has not been answered. In the studiesby Lu et al.50 and Smyth et al.,76 the dissolution of carbonateminerals and adsorption/desorption to hydrous oxides, organicmatter, and clayminerals were highlighted as controlling the totalaqueous concentration of various metals; but no information wasprovided on the distribution of the pertinent minerals or theirorigin. This information is relevant to answering the question ofwhether the minerals were precipitated before or after theintroduction of CO2. Kharaka et al.

56 also attributed observedincreases in aqueous metal concentrations at the ZERT field siteto CO2-induced dissolution of carbonates, iron oxides and/ordesorption-proton exchange reactions possibly from clayminerals but, again, the distribution and origin of the mineralsresponsible for the observed changes were unknown.Knowledge of mineral distribution/heterogeneity in aquifers

or the vadose zone may also be crucial in interpreting anddesigning laboratory experiments with material from field sites.

For example, Little and Jackson49 found that CO2 intrusion hadno effect on aqueous Cd concentration in calcite-rich sedimentsbut induced a 1000% increase in Cd for the calcite-deficientsediments. In contrast, concentrations of Al, V, and Cr increasedafter CO2 intrusion in chalcopyrite-containing sediments butdecreased in other sediments. The lack of an effect on Cdconcentrations in the calcite-rich sediments of Little andJackson49 is attributable to CO2-induced dissolution of calcitewhich would serve to buffer the pH and thereby preventdissolution of Cd-bearing minerals. Similarly, the oxidativedissolution of chalcopyrite would decrease solution pH therebyenhancing dissolution of Al, V, or Cr-bearing minerals andincreasing aqueous concentrations of Al, V, and Cr. Thus, fromthe perspective of experimental design, knowledge of mineralogyis crucial in matching laboratory conditions (as closely aspossible) to those that could be expected under field conditions.We believe that the lack of matching laboratory (e.g., solution

and redox) conditions to those observed in the field is one areawhere current laboratory studies could be improved. Forexample, Little and Jackson49 noted that all their experimentswere conducted in deionized water under oxidized conditions.However, the presence of sulfide minerals in some of theirsediments wasmore representative of sediments originating froma reduced environment. Hence despite Little and Jackson49

pointing to pH as the primary factor responsible for thediscrepancies between their study and that of Lu et al.,50 it ismore plausible that a change in redox conditions was the factorcontrolling metal release (especially since pH values between thetwo systems were comparable). Rather than the oxic conditionsof Little and Jackson,49 conditions existing in the Lu et al.50

experiments were likely anoxic (an Ar purge was used prior toCO2 introduction). This difference in redox condition is aplausible reason why, in contrast to Little and Jackson,49 Lu etal.50 observed no significant changes in the aqueous concen-

Figure 2. Conceptual framework for assessing geochemical impact of CO2 on near surface environments.



trations of redox-sensitive metals (e.g., Fe, Cu, and U) or metals(e.g., Ni) that were likely contained in redox-sensitive minerals.Why Are Experimental and Assessment Protocols

Crucial? A revisiting of current experimental approaches andthe establishment of rigorous protocols is crucial to thealleviation of discrepancies between experimental studies, aswell as the reconciliation of information from “analog” or“comparable” systems with leakage at geologic sequestrationsites. The likelihood of experimental discrepancies can beminimized if experiments are conducted under protocols thatwere developed from a process-based understanding of geo-chemical reactions in the subsurface. For example, inconsisten-cies such as those between Lu et al. and Little and Jacksonmay beprevented by defining specific protocols/guidelines for redoxcondition and solution chemistry used in experiments gearedtoward understanding how gas intrusion is likely to impact thegeochemistry of potable aquifers and the vadose zone. Suchprotocols could also provide a framework within which tocompare findings from different studies and conduct researchaimed at closing current knowledge gaps.One possible framework within which to assess the geo-

chemical impact of CO2 on near-surface environment ispresented in Figure 2. The framework in Figure 2 is centeredon the master variables (pH and Eh). It is the magnitude and/ordirection of change in pH and Eh, as controlled by the propertiesof the impacted system (e.g., mineralogy or initial solutioncomposition) and the characteristics of the intruding gas stream(e.g., composition or influx rate), that would determine if gasleakage would have beneficial or deleterious impact on thegeochemistry of the receiving potable aquifer or vadose zone. Forexample, gas-leakage-induced changes in system pH and Ehfavoring precipitation reactions could be beneficial providing thatsuch precipitation results in the formation of carbonate mineralsthereby enhancing CO2 sequestration, the formation of sorbentswith significant sorption capacity to immobilize potentialcontaminants from solution, and/or incorporation of contami-nants into the structure of the neophase(s) thereby reducingtheir mobility. On the other hand, gas-leakage-induced changesin system pH and Eh favoring dissolution could be deleteriousespecially in the absence of a suitable sorbent.There is also a wealth of information from “analog” or

“comparable” systems that needs to be reconciled, to ascertaintheir geochemical value in understanding the impacts of gasleakage from GCS. Although a detailed reconciliation of analogor comparable studies is really beyond the scope of this review,we do believe that a set of rigorous experimental and assessmentprotocols would be useful in the process. For example, it has beenargued that natural CO2 analogs highlight the low risk for gas-induced contamination associated with GCS. As evidence for thislow risk, proponents have pointed to the fact that water fromthese systems are often potable and are consumed with noadverse health effects.105 There is also evidence from studies atanalog sites that have shown no, or only localized, deteriorationin water quality due to intrusion of CO2 from naturalsources.131−134 However, before conclusions on risk can bemade, it is important to consider pertinent questions such as:How comparable are the gas stream characteristics (e.g.,composition, flow rate or duration) at natural analogs to thatexpected at GCS? Are (or were) the characteristics of thereceiving environment (e.g., mineralogy, solution composition,sediment exposure to gas) comparable?It is important to note that asking these questions should not

negate the value of natural analogs in understanding the impact

of gas migration from geologic storage. Instead, they should seekto qualify such value. For example, although a given naturalanalog may have little value in predicting short-term geochemicalconsequences of gas leakage associated with GCS, it does notnegate the fact that it could provide valuable information onpotential flow paths, leakage size, reaction kinetics, equilibrationtime and equilibrium mineral phases. Such information could beimportant in terms of policy and strategic decisions onremediation approaches in the event of significant gas-leakage-induced deterioration in potable water or soil quality. We believethat questions and concerns with natural analogs or comparablesites would be best addressed using a set of rigorous protocolsthat clearly defines what constitutes an analog or comparable site,for what reasons, and what stage in the life cycle of the gas leak arethe use of these sites appropriate?

■ SUMMARY AND CONCLUSIONSGas leakage from the storage reservoir is a major risk factor andpotential barrier to the widespread acceptance of geologic CO2sequestration. Different schools of thought exist, concerning howsuch leakages would impact the geochemistry of critical nearsurface environments such as potable aquifers and the vadosezone. On one hand it has been argued that the intrusion of CO2into the near surface would have little to no negative impact ongeochemistry. Proponents of this thought point to the fact thatpotable sources of CO2 rich groundwater exist across the worldand have been consumed for centuries without adverse healtheffect. Another common school of thought is that CO2-inducedacidification of groundwater will enhance the dissolution ofcontaminant-bearing soil and rock minerals, resulting in adegradation of environmental quality. In recent years a significantamount of scientific research has been conducted to assess thepotential geochemical implications of CO2 migration into thevadose zone or potable aquifers. These studies include 1- to 3-Dmodeling efforts and laboratory experiments, as well as field andnatural analog studies.We conducted a comprehensive review of the recently

published literature on how elevated CO2 levels may impactgeochemical processes under low-temperature, low-pressureconditions characteristic of near surface environments. Emphasiswas placed on CO2-induced effects on dissolution/precipitationand adsorption/desorption reactions, and consequences for thegeochemistry of the vadose zone and potable aquifers. Thereview revealed that, 1) there is a significant amount of newscientific evidence that suggests that CO2 intrusion into potableaquifers or the vadose zone may have both beneficial anddeleterious outcomes, 2) despite an increase in recent effortssignificant knowledge gaps still exist. From the perspective ofbeneficial outcomes there is strong evidence to suggest that CO2intrusion may result in the immobilization of certaincontaminants by influencing their chemical speciation, enhanc-ing their incorporation into stable mineral phases, or enhancingthe precipitation of suitable sorbents. On the other hand, there isalso strong evidence to indicate that CO2 intrusion may bedeleterious due to mobilization of some contaminants as a resultof CO2-induced dissolution of contaminant-bearing minerals, ordesorption of contaminants from sorption sites.From the perspective of knowledge gaps, we have identified

and discussed several areas of research that we believe require asignificant amount of effort (particularly experimental work).These include the development of a systematic understanding ofhow CO2 impacts both pH and Eh and, subsequently, theircoupled effect on precipitation/dissolution and adsorption/



desorption reactions in CO2-impacted systems, how microbesare impacted by or may impact pertinent geochemical processes(e.g., precipitation/dissolution) in CO2 impacted systems, howmineral heterogeneity and distribution influences geochemicaloutcome, and how specific geochemical processes are influencedby gas stream characteristics (such as composition and fluxcharacteristics). We are cognizant that the knowledge gapsdiscussed in this manuscript by nomeans represent an exhaustivelist and could be expanded to include other relevant factors suchas mixing/dispersion, contact area and residence time. However,we do believe that a systematic closing of these knowledge gapswould most efficiently advance scientific knowledge, improvepredictability and risk assessment associated with CO2 leakageand facilitate better science-based decision- and policy-making tosupport GCS.

■ AUTHOR INFORMATIONCorresponding Author*Phone: (601) 266-4529; fax: (601) 266-6219; e-mail: [email protected] authors declare no competing financial interest.

■ ACKNOWLEDGMENTSFunding for this research was provided by the National RiskAssessment Partnership (NRAP) through the U.S. DOE Officeof Fossil Energy's Division of Crosscutting Research undercontract DE-AC05-76RL01830. We are grateful to Drs. HongboShao (PNNL), Wooyong Um (PNNL), Susan Carroll (LLNL),and Nic Spycher (LBNL) who reviewed an earlier version of themanuscript and provided valuable comments and suggestions.Comments and suggestions from the Associate Editor andanonymous reviewers also improved this manuscript.

■ REFERENCES(1) Bergman, P. D.; Winter, E. M. Disposal of carbon dioxide inaquifers in the U.S. Energy Convers. Manage. 1995, 36 (6−9), 523−526.(2) Herzog, H. J. What future for carbon capture and sequestration?Environ. Sci. Technol. 2001, 35 (7), 148A−153A.(3) Bachu, S.; Adams, J. J. Sequestration of CO2 in geological media inresponse to climate change: Capacity of deep saline aquifers to sequesterCO2 in solution. Energy Convers. Manage. 2003, 44 (20), 3151−3175.(4) Metz, B.; Davidson, O.; de Coninck, H.; Loos, M.; Meyer, L.Carbon dioxide capture and storage-special report of the inter-governmental panel on climate change; Cambridge University Press:Cambridge, 2005.(5) Benson, S. M.; Cole, D. R. CO2 sequestration in deep sedimentaryformations. Elements 2008, 4 (5), 325−331.(6) Kovscek, A. R. Screening criteria for CO2 storage in oil reservoirs.Petrol. Sci. Technol. 2002, 20 (7−8), 841−866.(7) Hovorka, S. D.; Choi, J.-W.; Meckel, T. A.; Trevino, R. H.; Zeng,H.; Kordi, M.;Wang, F. P.; Nicot, J.-P., Comparing carbon sequestrationin an oil reservoir to sequestration in a brine formation-field study. InGreenhouse Gas Control Technologies 9; Gale, J. H. H. B. J., Ed.; 2009;Vol. 1, pp 2051−2056.(8) Zhang, L.; Ren, S.; Ren, B.; Zhang, W.; Guo, Q.; Zhang, L.Assessment of CO2 storage capacity in oil reservoirs associated withlarge lateral/underlying aquifers: Case studies from China. Int. J.Greenhouse Gas Control 2011, 5 (4), 1016−1021.(9) White, C. M.; Smith, D. H.; Jones, K. L.; Goodman, A. L.; Jikich, S.A.; LaCount, R. B.; DuBose, S. B.; Ozdemir, E.; Morsi, B. I.; Schroeder,K. T. Sequestration of carbon dioxide in coal with enhanced coalbedmethane recoveryA Review. Energy Fuel 2005, 19 (3), 659−724.(10) Faiz, M. M.; Saghafi, A.; Barclay, S. A.; Stalker, L.; Sherwood, N.R.; Whitford, D. J.; Flagship, C. E. T. Evaluating geological sequestration

of CO2 in bituminous coals: The southern Sydney Basin, Australia as anatural analogue. Int. J. Greenhouse Gas Control 2007, 1 (2), 223−235.(11) Kronimus, A.; Busch, A.; Alles, S.; Juch, D.; Jurisch, A.; Littke, R. Apreliminary evaluation of the CO2 storage potential in unminable coalseams of the Munster Cretaceous Basin, Germany. Int. J. Greenhouse GasControl 2008, 2 (3), 329−341.(12) Goldberg, D. S.; Kent, D. V.; Olsen, P. E. Potential on-shore andoff-shore reservoirs for CO2 sequestration in Central Atlantic magmaticprovince basalts. Proc. Nat. Acad. Sci. U.S.A 2010, 107 (4), 1327−1332.(13) McGrail, B. P.; Schaef, H. T.; Ho, A. M.; Chien, Y. J.; Dooley, J. J.;Davidson, C. L., Potential for carbon dioxide sequestration in floodbasalts. J. Geophys. Res., [Solid Earth Planets] 2006, 111, (B12).(14) Goldberg, D. S.; Takahashi, T.; Slagle, A. L. Carbon dioxidesequestration in deep-sea basalt. Proc. Nat. Acad. Sci. U.S.A 2008, 105(29), 9920−9925.(15) Gislason, S. R.; Wolff-Boenisch, D.; Stefansson, A.; Oelkers, E. H.;Gunnlaugsson, E.; Sigurdardottir, H.; Sigfusson, B.; Broecker, W. S.;Matter, J. M.; Stute, M.; Axelsson, G.; Fridriksson, T. Mineralsequestration of carbon dioxide in basalt: A pre-injection overview ofthe CarbFix project. Int. J. Greenhouse Gas Control 2010, 4 (3), 537−545.(16) Donda, F.; Volpi, V.; Persoglia, S.; Parushev, D. CO2 storagepotential of deep saline aquifers: The case of Italy. Int. J. Greenhouse GasControl 2011, 5 (2), 327−335.(17) Medina, C. R.; Rupp, J. A.; Barnes, D. A. Effects of reduction inporosity and permeability with depth on storage capacity and injectivityin deep saline aquifers: A case study from the Mount Simon Sandstoneaquifer. Int. J. Greenhouse Gas Control 2011, 5 (1), 146−156.(18) Shaffer, G. Long-term effectiveness and consequences of carbondioxide sequestration. Nat. Geosci. 2010, 3 (7), 464−467.(19) Bachu, S. CO2 storage in geological media: Role, means, statusand barriers to deployment. Prog. Energy Combust. Sci. 2008, 34 (2),254−273.(20) Wilson, E. J.; Friedmann, S. J.; Pollak, M. F. Research fordeployment: Incorporating risk, regulation, and liability for carboncapture and sequestration. Environ. Sci. Technol. 2007, 41 (17), 5945−5952.(21) Celia, M. A.; Bachu, S., Geological sequestration of CO2: Isleakage unavoidable and acceptable? In Greenhouse Gas ControlTechnologies6th International Conference, Gale, J.; Kaya, Y., Eds.Pergamon: Oxford, 2003; pp 477−482.(22) Jordan, P.; Benson, S. Well blowout rates and consequences inCalifornia Oil and Gas District 4 from 1991 to 2005: Implications forgeological storage of carbon dioxide. Environ. Geol. 2009, 57 (5), 1103−1123.(23) Holloway, S.; Pearce, J. M.; Hards, V. L.; Ohsumi, T.; Gale, J.Natural emissions of CO2 from the geosphere and their bearing on thegeological storage of carbon dioxide. Energy 2007, 32 (7), 1194−1201.(24) Damen, K.; Faaij, A.; van Bergen, F.; Gale, J.; Lysen, E.Identification of early opportunities for CO2 sequestration-worldwidescreening for CO2-EOR and CO2-ECBM projects. Energy 2005, 30(10), 1931−1952.(25) Skinner, L. CO2 Blowouts: An Emerging Problem: Well Control andIntervention; Gulf: Houston, TX, ETATS-UNIS, 2003; Vol. 224, p 4.(26) Hepple, R. P., Human health and ecological effects of carbondioxide exposure. In Carbon Dioxide Capture for Storage in Deep GeologicFormations; Elsevier Science: Amsterdam, 2005; pp 1143−1172.(27) Celia, M. A.; Nordbotten, J. M. Practical modeling approaches forgeological storage of carbon dioxide. Ground Water 2009, 47 (5), 627−638.(28) Bachu, S. Screening and ranking of sedimentary basins forsequestration of CO2 in geological media in response to climate change.Environ. Geol. 2003, 44 (3), 277−289.(29) Saadatpoor, E.; Bryant, S.; Sepehrnoori, K. New trappingmechanism in carbon sequestration.Transport PorousMed. 2010, 82 (1),3−17.(30) Bachu, S. Carbon dioxide storage capacity in uneconomic coalbeds in Alberta, Canada: Methodology, potential and site identification.Int. J. Greenhouse Gas Control 2007, 1 (3), 374−385.



mailto:[email protected]

mailto:[email protected]

(31) Ross, H. E.; Hagin, P.; Zoback, M. D. CO2 storage and enhancedcoalbed methane recovery: Reservoir characterization and fluid flowsimulations of the Big George coal, Powder River Basin, Wyoming, USA.Int. J. Greenhouse Gas Control 2009, 3 (6), 773−786.(32) Faiz, M. M.; Saghafi, A.; Barclay, S. A.; Stalker, L.; Sherwood, N.R.; Whitford, D. J. , Evaluating geological sequestration of CO2 inbituminous coals: The southern Sydney Basin, Australia as a naturalanalogue. Int. J. Greenhouse Gas Control 2007, 1 (2), 223−235.(33) Han, W. S.; McPherson, B. J.; Lichtner, P. C.; Wang, F. P.Evaluation of trapping mechanisms in geologic CO2 sequestration: Casestudy of SACROC northern platform, a 35-year CO2 injection site. Am.J. Sci. 2010, 310 (4), 282−324.(34) Han, W. S.; McPherson, B. J.; Lichtner, P. C.; Wang, F. P.Evaluation of trapping mechanisms in geologic CO2 sequestration: Casestudy of SACROC northern platform, a 35-year CO2 injection site. Am.J. Sci. 2010, 310 (4), 282−324.(35) Gunter, W. D.; Wiwehar, B.; Perkins, E. H. Aquifer disposal ofCO2-rich greenhouse gases: Extension of the time scale of experimentfor CO2-sequestering reactions by geochemical modelling.Miner. Petrol.1997, 59 (1), 121−140.(36)Wilkinson,M.; Haszeldine, R. S.; Fallick, A. E.; Odling, N.; Stoker,S. J.; Gatliff, R. W. CO2−Mineral reaction in a natural analogue for CO2

storageImplications for modeling. J. Sediment. Res. 2009, 79 (7), 486−494.(37) Moore, J.; Adams, M.; Allis, R.; Lutz, S.; Rauzi, S. Mineralogicaland geochemical consequences of the long-term presence of CO2 innatural reservoirs: An example from the Springerville−St. Johns Field,Arizona, and New Mexico, U.S.A. Chem. Geol. 2005, 217 (3−4), 365−385.(38) Duan, Z. H.; Sun, R. An improved model calculating CO2

solubility in pure water and aqueous NaCl solutions from 273 to 533K and from 0 to 2000 bar. Chem. Geol. 2003, 193 (3−4), 257−271.(39) Kharaka, Y. K.; Cole, D. R.; Hovorka, S. D.; Gunter, W. D.;Knauss, K. G.; Freifeld, B. M. Gas-water-rock interactions in FrioFormation following CO2 injection: Implications for the storage ofgreenhouse gases in sedimentary basins.Geology 2006, 34 (7), 577−580.(40) Loring, J. S.; Thompson, C. J.; Wang, Z.; Joly, A. G.; Sklarew, D.S.; Schaef, H. T.; Ilton, E. S.; Rosso, K. M.; Felmy, A. R. In situ infraredspectroscopic study of forsterite carbonation in wet supercritical CO2.Environ. Sci. Technol. 2011, 45 (14), 6204−10.(41) Shao, H.; Ray, J. R.; Jun, Y.-S. Dissolution and precipitation of clayminerals under geologic CO2 sequestration conditions: CO2−brine−phlogopite interactions. Environ. Sci. Technol. 2010, 44 (15), 5999−6005.(42) Jacquemet, N.; Pironon, J.; Saint-Marc, J. Mineralogical changesof a well cement in various H2S-CO2(-brine) fluids at high pressure andtemperature. Environ. Sci. Technol. 2008, 42 (1), 282−288.(43) Kutchko, B. G.; Strazisar, B. R.; Dzombak, D. A.; Lowry, G. V.;Thaulow, N. Degradation of well cement by CO2 under geologicsequestration conditions. Environ. Sci. Technol. 2007, 41 (13), 4787−4792.(44) Regnault, O.; Lagneau, V.; Catalette, H.; Schneider, H.Experimental study of pure mineral phases/supercritical CO2 reactivity.Implications for geological CO2 sequestration. C. R. Geoscience 2005,337 (15), 1331−1339.(45) Gaus, I. Role and impact of CO2-rock interactions during CO2

storage in sedimentary rocks. Int. J. Greenhouse Gas Control 2010, 4 (1),73−89.(46) Oldenburg, C. M.; Unger, A. J. A. On leakage and seepage fromgeologic carbon sequestration sites: Unsaturated zone attenuation.Vadose Zone J. 2003, 2 (3), 287−296.(47) Klusman, R. W. Rate measurements and detection of gasmicroseepage to the atmosphere from an enhanced oil recovery/sequestration project, Rangely, Colorado, USA. Appl. Geochem. 2003, 18(12), 1825−1838.(48) Altevogt, A. S.; Jaffe, P. R., Modeling the effects of gas phase CO2

intrusion on the biogeochemistry of variably saturated soils. WaterResour. Res. 2005, 41, (9).

(49) Little, M. G.; Jackson, R. B. Potential impacts of leakage from deepCO2 geosequestration on overlying freshwater aquifers. Environ. Sci.Technol. 2010, 44 (23), 9225−9232.(50) Lu, J. M.; Partin, J. W.; Hovorka, S. D.; Wong, C. Potential risks tofreshwater resources as a result of leakage from CO2 geological storage:A batch-reaction experiment. Environ. Earth Sci. 2010, 60 (2), 335−348.(51) Vong, C. Q.; Jacquemet, N.; Picot-Colbeaux, G.; Lions, J.;Rohmer, J.; Bouc, O. Reactive transport modeling for impact assessmentof a CO2 intrusion on trace elements mobility within fresh groundwaterand its natural attenuation for potential remediation. Energy Procedia2011, 4, 3171−3178.(52) Wang, S.; Jaffe, P. R. Dissolution of a mineral phase in potableaquifers due to CO2 releases from deep formations; Effect of dissolutionkinetics. Energy Convers. Manage. 2004, 45 (18−19), 2833−2848.(53) Wei, Y.; Maroto-Valer, M.; Steven, M. D. Environmentalconsequences of potential leaks of CO2 in soil. Energy Procedia 2011,4 (0), 3224−3230.(54) Wilkin, R. T.; DiGiulio, D. C. Geochemical impacts togroundwater from geologic carbon sequestration: Controls on pH andinorganic carbon concentrations from reaction path and kineticmodeling. Environ. Sci. Technol. 2010, 44 (12), 4821−4827.(55) Zheng, L. G.; Apps, J. A.; Zhang, Y. Q.; Xu, T. F.; Birkholzer, J. T.On mobilization of lead and arsenic in groundwater in response to CO2

leakage from deep geological storage. Chem. Geol. 2009, 268 (3−4),281−297.(56) Kharaka, Y. K.; Thordsen, J. J.; Kakouros, E.; Ambats, G.;Herkelrath, W. N.; Beers, S. R.; Birkholzer, J. T.; Apps, J. A.; Spycher, N.F.; Zheng, L. E.; Trautz, R. C.; Rauch, H. W.; Gullickson, K. S. Changesin the chemistry of shallow groundwater related to the 2008 injection ofCO2 at the ZERT field site, Bozeman,Montana. Environ. Earth Sci. 2010,60 (2), 273−284.(57) Spycher, N.; Pruess, K. CO2-H2O mixtures in the geologicalsequestration of CO2. II. Partitioning in chloride brines at 12−100°Cand up to 600 bar. Geochim. Cosmochim. Acta 2005, 69 (13), 3309−3320.(58) Bearat, H.; McKelvy, M. J.; Chizmeshya, A. V. G.; Gormley, D.;Nunez, R.; Carpenter, R. W.; Squires, K.; Wolf, G. H. Carbonsequestration via aqueous olivine mineral carbonation: Role ofpassivating layer formation. Environ. Sci. Technol. 2006, 40 (15),4802−4808.(59) 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 halfthe story been neglected? In Greenhouse Gas Control Technologies 9;Gale, J.; Herzog, H.; Braitsch, J., Eds.; Elsevier Science Bv: Amsterdam,2009; Vol. 1, pp 3415−3419.(60) Birkholzer, J. T.; Apps, J. A.; Zheng, L.; Zhang, Y.; Xu, T.; Tsang,C.-F. Research project on CO2 geological storage and groundwaterresources: Water quality effects caused by CO2 intrusion into shallowgroundwater. http://www.osti.gov/bridge/product.biblio.jsp?query_id=0&page=0&osti_id=944435 (accessed March 06, 2012),(61) Langmuir, D., Aqueous Environmental Geochemistry; Prentice-HallInc.: Upper Saddle River, NJ, 1997.(62) Lippmann, F. Sedimentary Carbonate Minerals. Springer-Verlag:New York, 1973.(63) Saldi, G. D.; Jordan, G.; Schott, J.; Oelkers, E. H. Magnesitegrowth rates as a function of temperature and saturation state. Geochim.Cosmochim. Acta 2009, 73 (19), 5646−5657.(64) Arvidson, R. S.; Mackenzie, F. T. The dolomite problem: Controlof precipitation kinetics by temperature and saturation state. Am. J. Sci.1999, 299, 257−288.(65) Jimenez-Lopez, C. n.; Romanek, C. S. Precipitation kinetics andcarbon isotope partitioning of inorganic siderite at 25A°C and 1 atm.Geochim. Cosmochim. Acta 2004, 68 (3), 557−571.(66) Loring, J. S.; Thompson, C. J.; Zhang, C.; Wang, Z.; Schaef, H. T.;Rosso, K. M., In situ infrared spectroscopic study of brucite carbonationin dry to water-saturated supercritical carbon dioxide. J. Phys. Chem. A116, (19), 4768-4777.



http://www.osti.gov/bridge/product.biblio.jsp?query_id=0&page=0&osti_id=944435

http://www.osti.gov/bridge/product.biblio.jsp?query_id=0&page=0&osti_id=944435

(67) Harrison, A. L.; Power, I. M.; Dipple, G. M., Acceleratedcarbonation of brucite in mine tailings for carbon sequestration. Environ.Sci. Technol..(68) Beinlich, A.; Austrheim, H. k., In situ sequestration of atmosphericCO2 at low temperature and surface cracking of serpentinized peridotitein mine shafts. Chem. Geol. 332a €“333, (0), 32-44.(69) Kenward, P. A.; Goldstein, R. H.; GonzALez, L. A.; Roberts, J. A.Precipitation of low-temperature dolomite from an anaerobic microbialconsortium: The role of methanogenic Archaea. Geobiology 2009, 7 (5),556−565.(70) Power, I. M.; Wilson, S. A.; Small, D. P.; Dipple, G. M.; Wan, W.;Southam, G. Microbially mediated mineral carbonation: Roles ofphototrophy and heterotrophy. Environ. Sci. Technol. 2011, 45 (20),9061−8.(71) Roberts, J. A.; Bennett, P. C.; Gonzalez, L. A.; Macpherson, G. L.;Milliken, K. L. Microbial precipitation of dolomite in methanogenicgroundwater. Geology 2004, 32 (4), 277−280.(72) Dong, H.Mineral-microbe interactions: A review. Front. Earth Sci.China 2010, 4 (2), 127−147.(73) Douglas, S.; Beveridge, T. J. Mineral formation by bacteria innatural microbial communities. FEMSMicrobiol. Ecol. 1998, 26 (2), 79−88.(74) Lin, Y.-P.; Singer, P. C. Effects of seed material and solutioncomposition on calcite precipitation. Geochim. Cosmochim. Acta 2005,69 (18), 4495−4504.(75) Mozley, P. S.; Burns, S. J. Oxygen and carbon isotopiccomposition of marine carbonate concretions; An overview. J. Sediment.Res. 1993, 63 (1), 73−83.(76) Smyth, R. C.; Hovorka, S. D.; Lu, J. M.; Romanak, K. D.; Partin, J.W.; Wong, C.; Yang, C. B., Assessing risk to fresh water resources fromlong term CO2 injectionLaboratory and field studies. In GreenhouseGas Control Technologies 9, Gale, J., Herzog, H., Braitsch, J., Eds.; ElsevierScience Bv: Amsterdam, 2009; Vol. 1, pp 1957−1964.(77) Shao, H.; Ray, J. R.; Jun, Y.-S. Dissolution and precipitation of clayminerals under geologic CO2 sequestration conditions: CO2-brine-phlogopite interactions. Environ. Sci. Technol. 2010, 44 (15), 5999−6005.(78) Shao, H.; Ray, J. R.; Jun, Y.-S. Effects of salinity and the extent ofwater on supercritical CO2-induced phlogopite dissolution andsecondary mineral formation. Environ. Sci. Technol. 2011, 45 (4),1737−1743.(79) Lu, P.; Fu, Q.; Seyfried, W. E.; Hereford, A.; Zhu, C. NavajoSandstone-brine-CO2 interaction: Implications for geological carbonsequestration. Environ. Earth Sci. 2011, 62 (1), 101−118.(80) Hu, Y.; Ray, J. R.; Jun, Y.-S. Biotite-Brine interactions under acidichydrothermal conditions: Fibrous Illite, goethite, and kaoliniteformation and biotite surface cracking. Environ. Sci. Technol. 2011, 45(14), 6175−6180.(81) Opiso, E.; Sato, T.; Yoneda, T. Adsorption and co-precipitationbehavior of arsenate, chromate, selenate and boric acid with syntheticallophane-like materials. J. Hazard. Mater. 2009, 170 (1), 79−86.(82) Arai, Y.; Sparks, D. L.; Davis, J. A. Arsenate adsorptionmechanisms at the allophane−water interface. Environ. Sci. Technol.2005, 39 (8), 2537−2544.(83) Gimenez, J.; Martínez, M.; de Pablo, J.; Rovira, M.; Duro, L.Arsenic sorption onto natural hematite, magnetite, and goethite. J.Hazard. Mater. 2007, 141 (3), 575−580.(84) Rovira, M.; Gimenez, J.; Martínez, M.; Martínez-Llado, X.; dePablo, J.; Martí, V.; Duro, L. Sorption of selenium(IV) and selenium(VI)onto natural iron oxides: Goethite and hematite. J. Hazard. Mater. 2008,150 (2), 279−284.(85) Bradbury, M. H.; Baeyens, B. Modelling the sorption of Zn andNion Ca-montmorillonite. Geochim. Cosmochim. Acta 1999, 63 (3−4),325−336.(86) Poinssot, C.; Baeyens, B.; Bradbury, M. H. Experimental andmodelling studies of caesium sorption on Illite. Geochim. Cosmochim.Acta 1999, 63 (19−20), 3217−3227.

(87) Vingiani, S.; Righi, D.; Petit, S.; Terribile, F. Mixed-layer kaolinite-smectite minerals in red-black soil sequence from basalt in Sardinia(Italy). Clay Clay Miner. 2004, 52 (4), 473−483.(88) Jaisi, D. P.; Dong, H.; Plymale, A. E.; Fredrickson, J. K.; Zachara, J.M.; Heald, S.; Liu, C. Reduction and long-term immobilization oftechnetium by Fe(II) associated with clay mineral nontronite. Chem.Geol. 2009, 264 (1−4), 127−138.(89) Mason, C. F. V.; Turney, W.; Thomson, B. M.; Lu, N.; Longmire,P. A.; ChisholmBrause, C. J. Carbonate leaching of uranium fromcontaminated soils. Environ. Sci. Technol. 1997, 31 (10), 2707−2711.(90) Anawar, H. M.; Akai, J.; Komaki, K.; Terao, H.; Yoshioka, T.;Ishizuka, T.; Safiullah, S.; Kato, K. Geochemical occurrence of arsenic ingroundwater of Bangladesh: Sources and mobilization processes. J.Geochem. Explor. 2003, 77 (2−3), 109−131.(91) Saalfield, S. L.; Bostick, B. C. Synergistic effect of calcium andbicarbonate in enhancing arsenate release from ferrihydrite. Geochim.Cosmochim. Acta 2010, 74 (18), 5171−5186.(92) Gerson, A. R.; O’Dea, A. R. A quantum chemical investigation ofthe oxidation and dissolution mechanisms of Galena. Geochim.Cosmochim. Acta 2003, 67 (5), 813−822.(93) Floroiu, R. M.; Davis, A. P.; Torrents, A. Kinetics and Mechanismof As2S3(am) Dissolution under N2. Environ. Sci. Technol. 2004, 38 (4),1031−1037.(94) Carroll, S.; Hao, Y.; Aines, R. Geochemical detection of carbondioxide in dilute aquifers. Geochem. Trans. 2009, 10 (1), 4.(95) Ardelan, M. V.; Steinnes, E. Changes in mobility and solubility ofthe redox sensitive metals Fe, Mn and Co at the seawater-sedimentinterface following CO2 seepage. Biogeosciences 2010, 7 (2), 569−583.(96) Huesemann, M. H.; Skillman, A. D.; Crecelius, E. A. Theinhibition of marine nitrification by ocean disposal of carbon dioxide.Mar. Pollut. Bull. 2002, 44 (2), 142−148.(97) Kirk, M. F. Variation in energy available to populations ofsubsurface anaerobes in response to geological carbon storage. Environ.Sci. Technol. 2011, 45 (15), 6676−6682.(98) Onstott, T. C. Impact of CO2 injection on deep subsurfacemicrobial ecosystems and potential ramifications for the surfacebiosphere. In The CO2 Capture and Storage Project; Thomas, D. C.,Benson, S. M., Eds.; 2005; Vol. II, pp 1207−1239.(99) Goldberg, S. Competitive adsorption of arsenate and arsenite onoxides and clay minerals contribution from the George E. Brown Jr.,Salinity Laboratory. Soil Sci. Soc. Am. J. 2002, 66 (2), 413−421.(100) Wilkin, R. T.; Wallschlager, D.; Ford, R. G. Speciation of arsenicin sulfidic waters. Geochem Trans 2003, 4 (1), 1−7.(101) Kamei, G.; Ohmoto, H. The kinetics of reactions between pyriteand O2-bearing water revealed from in situ monitoring of DO, Eh andpH in a closed system. Geochim. Cosmochim. Acta 2000, 64 (15), 2585−2601.(102) Brantley, S. L., Reaction kinetics of primary rock-formingminerals under ambient conditions. In Treatise on Geochemistry;Heinrich, D. H., Karl, K. T., Eds.; Pergamon: Oxford, 2003; pp 73−117.(103) Thompson, A.; Rancourt, D. G.; Chadwick, O. A.; Chorover, J.Iron solid-phase differentiation along a redox gradient in basaltic soils.Geochim. Cosmochim. Acta 2011, 75 (1), 119−133.(104) Raven, K. P.; Jain, A.; Loeppert, R. H. Arsenite and arsenateadsorption on ferrihydrite: Kinetics, equilibrium, and adsorptionenvelopes. Environ. Sci. Technol. 1998, 32 (3), 344−349.(105) Gilfillan, S. M. V.; Haszeldine, R. S. Comment on “potentialimpacts of leakage from deep CO2 geosequestration on overlyingfreshwater aquifers. Environ. Sci. Technol. 2011, 45 (7), 3171−3174.(106) McBride, M. B. Environmental Chemistry of Soils; OxfordUniversity Press: Oxford, 1994.(107) Reynolds, A. J.; Verheyen, T. V.; Adeloju, S. B.; Meuleman, E.;Feron, P. Towards commercial scale postcombustion capture of CO2

with monoethanolamine solvent: Key considerations for solventmanagement and environmental impacts. Environ. Sci. Technol. 2012,46 (7), 3643−3654.(108) Jacquemet, N.; Picot-Colbeaux, G.; Vong, C. Q.; Lions, J.; Bouc,O.; Jeremy, R. Intrusion of CO2 and impurities in a freshwater aquifer –



Impact evaluation by reactive transport modelling. Energy Procedia2011, 4, 3202−3209.(109) Fowler, D.; Smith, R. I.; Muller, J. B. A.; Hayman, G.; Vincent, K.J. Changes in the atmospheric deposition of acidifying compounds in theUK between 1986 and 2001. Environ. Pollut. 2005, 137 (1), 15−25.(110) Goyer, R. A.; Bachmann, J.; Clarkson, T. W.; Ferris, B. G.;Graham, J.; Mushak, P.; Perl, D. P.; Rall, D. P.; Schlesinger, R.; Sharpe,W.; Wood, J. M. Potential human health-effects of acid-rain- Report of aworkshop. Environ. Health Perspect. 1985, 60 (MAY), 355−368.(111) Matson, P. A.; McDowell, W. H.; Townsend, A. R.; Vitousek, P.M. The globalization of N deposition: Ecosystem consequences intropical environments. Biogeochemistry 1999, 46 (1−3), 67−83.(112) Aloisi, G.; Bouloubassi, I.; Heijs, S. K.; Pancost, R. D.; Pierre, C.;Damste, J. S. S.; Gottschal, J. C.; Forney, L. J.; Rouchy, J. M. CH4-consuming microorganisms and the formation of carbonate crusts atcold seeps. Earth Planet. Sci. Lett. 2002, 203 (1), 195−203.(113) Blair, N. E.; Aller, R. C. Anaerobic methane oxidation on theAmazon shelf. Geochim. Cosmochim. Acta 1995, 59 (18), 3707−3715.(114) Chen, Y.; Matsumoto, R.; Paull, C. K.; Ussler, W., III; Lorenson,T.; Hart, P.; Winters, W. Methane-derived authigenic carbonates fromthe northern Gulf of MexicoMD02 cruise. J. Geochem. Explor. 2007,95 (1−3), 1−15.(115) Michaelis, W.; Seifert, R.; Nauhaus, K.; Treude, T.; Thiel, V.;Blumenberg, M.; Knittel, K.; Gieseke, A.; Peterknecht, K.; Pape, T.;Boetius, A.; Amann, R.; Jorgensen, B. B.; Widdel, F.; Peckmann, J.;Pimenov, N. V.; Gulin, M. B. Microbial reefs in the Black Sea fueled byanaerobic oxidation of methane. Science 2002, 297 (5583), 1013−1015.(116) Peckmann, J.; Reimer, A.; Luth, U.; Luth, C.; Hansen, B. T.;Heinicke, C.; Hoefs, J.; Reitner, J. Methane-derived carbonates andauthigenic pyrite from the northwestern Black Sea.Mar. Geol. 2001, 177(1−2), 129−150.(117) Stevens, T. O.; McKinley, J. P. Lithoautotrophic microbialecosystems in deep basalt aquifers. Science 1995, 270 (5235), 450−454.(118) Morozovaa, D.; Zettlitzerb, M.; Leta, D.; Wurdemanna, H.;group, t. C. S. Monitoring of the microbial community composition indeep subsurface saline aquifers during CO2 storage in Ketzin, Germany.Energy Procedia 2011, 4, 4362−4370.(119) Wandreya, M.; Pellizaria, L.; Zettlitzerb, M.; Wurdemanna, H.Microbial community and inorganic fluid analysis during CO2 storagewithin the frame of CO2SINKLong-term experiments under in situconditions. Energy Procedia 2011, 4, 3651−3657.(120) Jones, D. G.; Lister, T. R.; Smith, D. J.; West, J. M.; Coombs, P.;Gadalia, A.; Brach, M.; Annunzialellis, A.; Lombardi, S. In SalahGas CO2

storage JIP: Surface gas and biological monitoring. Energy Procedia 2011,4, 3566−3573.(121) Klusman, R.W. Detailed compositional analysis of gas seepage atthe National Carbon Storage Test Site, Teapot Dome, Wyoming, USA.Appl. Geochem. 2006, 21 (9), 1498−1521.(122) Mitchell, A. C.; Dideriksen, K.; Spangler, L. H.; Cunningham, A.B.; Gerlach, R. Microbially enhanced carbon capture and storage bymineral-trapping and solubility-trapping. Environ. Sci. Technol. 2010, 44(13), 5270−5276.(123)Mitchell, A. C.; Ferris, F. G. The coprecipitation of Sr into calciteprecipitates induced by bacterial ureolysis in artificial groundwater:Temperature and kinetic dependence. Geochim. Cosmochim. Acta 2005,69 (17), 4199−4210.(124) Ferris, F. G.; Phoenix, V.; Fujita, Y.; Smith, R. W. Kinetics ofcalcite precipitation induced by ureolytic bacteria at 10 to 20°C inartificial groundwater. Geochim. Cosmochim. Acta 2004, 68 (8), 1701−1710.(125) Tobler, D. J.; Cuthbert, M. O.; Greswell, R. B.; Riley, M. S.;Renshaw, J. C.; Handley-Sidhu, S.; Phoenix, V. R. Comparison of rates ofureolysis between Sporosarcina pasteurii and an indigenous groundwatercommunity under conditions required to precipitate large volumes ofcalcite. Geochim. Cosmochim. Acta 2011, 75 (11), 3290−3301.(126) Wu, B.; Shao, H.; Wang, Z.; Hu, Y.; Tang, Y. J.; Jun, Y.-S.Viability and metal reduction of Shewanella oneidensis MR-1 underCO2 stress: Implications for ecological effects of CO2 leakage from

geologic CO2 sequestration. Environ. Sci. Technol. 2010, 44 (23), 9213−9218.(127) Lin, X.; Kennedy, D.; Fredrickson, J.; Bjornstad, B.; Konopka, A.Vertical stratification of subsurface microbial community compositionacross geological formations at the Hanford Site. Environ. Microbiol.2011.(128) Chapelle, F. H.; O’Neill, K.; Bradley, P. M.; Methe, B. A.; Ciufo,S. A.; Knobel, L. L.; Lovley, D. R. A hydrogen-based subsurfacemicrobial community dominated by methanogens. Nature 2002, 415(6869), 312−315.(129) Lloyd, J. R. Microbial reduction of metals and radionuclides.FEMS Microbiol. Rev. 2003, 27 (2−3), 411−425.(130) Lovley, D. R. Dissimilatory metal reduction. Annu. Rev.Microbiol. 1993, 47 (1), 263−290.(131) Aiuppa, A.; Federico, C.; Allard, P.; Gurrieri, S.; Valenza, M.Trace metal modeling of groundwater−gas−rock interactions in avolcanic aquifer: Mount Vesuvius, Southern Italy.Chem. Geol. 2005, 216(3−4), 289−311.(132) Keating, E. H.; Fessenden, J.; Kanjorski, N.; Koning, D. J.; Pawar,R. The impact of CO2 on shallow groundwater chemistry: Observationsat a natural analog site and implications for carbon sequestration.Environ. Earth Sci. 2010, 60 (3), 521−536.(133) Keating, E. H.; Hakala, J. A.; Viswanathan, H.; Capo, R.; Stewart,B.; Gardiner, J.; Guthrie, G.; William Carey, J.; Fessenden, J. Thechallenge of predicting groundwater quality impacts in a CO2 leakagescenario: Results from field, laboratory, andmodeling studies at a naturalanalog site in New Mexico, USA. Energy Procedia 2011, 4, 3239−3245.(134) Flaathen, T. K.; Gislason, S. R.; Oelkers, E. H.; Sveinbjornsdottir,A. E. Chemical evolution of the Mt. Hekla, Iceland, groundwaters: Anatural analogue for CO2 sequestration in basaltic rocks. Appl. Geochem.2009, 24 (3), 463−474.



Geochemical Implications of Gas Leakage associated with Geologic CO 2 Storage—A Qualitative Review

Documents