the role of silicate surfaces on calcite precipitation ... et al. gca the rolæe of silica... ·...

Available online at www.sciencedirect.com

www.elsevier.com/locate/gca

ScienceDirect

Geochimica et Cosmochimica Acta 135 (2014) 231–250

The role of silicate surfaces on calcite precipitation kinetics

Gabrielle J. Stockmann a,b,⇑, Domenik Wolff-Boenisch a,e, Nicolas Bovet c,Sigurdur R. Gislason a, Eric H. Oelkers a,b,d

a Institute of Earth Sciences, University of Iceland, Sturlugata 7, 101 Reykjavık, Icelandb GET-Universite de Toulouse-CNRS-IRD-OMP, 14 Avenue Edouard Belin, 31400 Toulouse, France

c Nano-Science Center, Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmarkd Earth Sciences, University College London, Gower Street, London WC1E 6BT, United Kingdom

e Department of Applied Geology, Curtin University, GPO Box U1987, Perth 6845 Western Australia, Australia

Received 7 January 2013; accepted in revised form 14 March 2014; Available online 26 March 2014

Abstract

The aim of this study is to illuminate how calcite precipitation depends on the identity and structure of the growth sub-strate. Calcite was precipitated at 25 �C from supersaturated aqueous solutions in the presence of seeds of either calcite or oneof six silicate materials: augite, enstatite, labradorite, olivine, basaltic glass and peridotite rock. Calcite saturation wasachieved by mixing a CaCl2-rich aqueous solution with a NaHCO3–Na2CO3 aqueous buffer in mixed-flow reactors containing0.5–2 g of mineral, rock, or glass seeds. This led to an inlet fluid calcite saturation index of 0.6 and a pH equal to 9.1.Although the inlet fluid composition, flow rate, and temperature were identical for all experiments, the onset of calcite pre-cipitation depended on the identity of the seeds present in the reactor. Calcite precipitated instantaneously and at a constantrate in the presence of calcite grains. Calcite precipitated relatively rapidly on labradorite, olivine, enstatite, and peridotite(mainly composed of Mg-olivine) surfaces, but more slowly on augite and basaltic glass. Calcite precipitation rates, however,became independent of substrate identity and mass over time, and all rates approach 10�9.68 ± 0.08 mol/s for �10 day longexperiments and 10�9.21 ± 0.2 mol/s for �70 day long experiments. Scanning Electron Microscope images showed olivine,enstatite and peridotite surfaces to be covered extensively with calcite coatings at the end of the experiments. Less calcitewas found on labradorite and augite, and the least on basaltic glass. In all cases, calcite precipitation occurs on the mineral,rock or glass surfaces. Calcite precipitation on these surfaces, however, negligibly affects the dissolution rates of the silicategrains. These results support ultramafic and basalt carbonation as a long-term carbon storage strategy, as calcite readily pre-cipitates on the surfaces of minerals contained in these rocks without inhibiting their dissolution.� 2014 Elsevier Ltd. All rights reserved.

1. INTRODUCTION

Several past studies reported that calcite nucleation andgrowth is affected by the identity and state of available min-eral substrates (e.g. Suda et al., 2000; De Yoreo and

http://dx.doi.org/10.1016/j.gca.2014.03.015

0016-7037/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author. Present address: Department of Geo-logical Sciences, Stockholm University, 106 91 Stockholm, Sweden.Tel.: +46 (0)8 164878.

E-mail address: [email protected] (G.J. Stock-mann).

Vekilov, 2003; Cubillas et al., 2005; Diao et al., 2011). Twoof our recent studies (Stockmann et al., 2011, 2013) demon-strated a dramatic difference in the location of calcite precip-itates in the presence of basaltic glass versus diopside; calciteprecipitates readily on diopside surfaces, but avoids basalticglass surfaces. The present study was motivated to furtherour understanding of the effect of substrate identity on calcitenucleation and growth through a systematic study of calciteprecipitation rates in the presence or absence of 7 distinctmineral, rock or glass surfaces. The purpose of this manu-script is to report the results of this experimental study and


mailto:[email protected]


232 G.J. Stockmann et al. / Geochimica et Cosmochimica Acta 135 (2014) 231–250

to use these to illuminate the role of substrate compositionand structure on calcite nucleation and growth.

Calcite precipitation is a prerequisite for successful carbonstorage via carbonate mineralization (e.g. Seifritz, 1990;Lackner et al., 1995; Xu et al., 2005; Marini, 2007;Wakahama et al., 2009; Flaathen et al., 2011). Understandingthe processes that control calcite precipitation on silicates,therefore, may be critical for the success of carbon minerali-zation efforts in basaltic rocks, such as those as currentlybeing carried out in Iceland (Oelkers et al., 2008;Alfredsson et al., 2008, 2011, 2013; Matter et al., 2009,2011; Gislason et al., 2010; Aradottir et al., 2011, 2012a,b;Gysi and Stefansson, 2011, 2012; Galeczka et al., 2014).

2. THEORETICAL BACKGROUND

The standard state adopted in this study is that of unitactivity of pure minerals and H2O at any temperature andpressure. For aqueous species other than H2O, the standardstate is unit activity of species in a hypothetical 1.0 mol/kgsolution referenced to infinite dilution at any temperatureand pressure. Thermodynamic calculations reported in thisstudy were performed using the PHREEQC version 2.17computer code (Parkhurst and Appelo, 1999) together withits llnl.dat database. Thermodynamic properties for calcite,enstatite, diopside, hedenbergite, anorthite, and forsterite(Mg-olivine) are included in this database. No thermody-namic data are available in this database, however, for augiteand labradorite. Therefore, the thermodynamic properties ofdiopside/hedenbergite and anorthite were used to estimate thestability of augite and labradorite, respectively.

Calcite precipitation can be described by (e.g. Stummand Morgan, 1996):

Ca2þ þ CO2�3 ¼ CaCO3ðsÞ ð1Þ

Constant pH calcite precipitation rates are commonlyquantified using the semi-empirical relation (e.g. Shirakiand Brantley, 1995):

rs ¼ kðX� 1Þn ð2Þ

where rs represents the precipitation rate normalized to areacting surface area, k refers to an apparent rate constant,X defines the saturation state of solution, and n designatesan empirical reaction order. X is equal to the ratio of theion activity product (Q) to the thermodynamic solubilityproduct (K). Consistent with the standard state, the satura-tion state for calcite is given by

X ¼ QK¼

aCa2þ � aCO2�3

Kð3Þ

where ai refers to the activity of the subscripted aqueousspecies. The logarithm of X (logX) is commonly referredto as the saturation index (SI).1

1 The equations provided in this manuscript are generallyapplicable so long as one uses a consistent set of units on theparameters. In the present study the units adopted for rs and k aremol/cm2/s. Fluid concentrations adopted in this study are in unitsof mol/kg and fluid flow rates in kg/s. X, n, Q, K and ai aredimensionless.

3. MATERIALS AND METHODS

3.1. Silicate minerals, rocks, and glass

Six different silicate materials: augite, enstatite, olivine,labradorite, basaltic glass, and peridotite rock, were usedin this study as substrates for calcite precipitation. Millime-tre-sized crystals of enstatite from Bamble, Norway; augitefrom Harcourt, Ontario, Canada; and labradorite fromSonora, Mexico were purchased from Ward’s Science andcrushed with agate mortar and pestle. The 45–125 lm sizefraction of these minerals were obtained by dry sieving.The resulting powders were cleaned ultrasonically in ace-tone over several cleaning cycles. After each cycle, the fineparticles were discarded and the cleaning cycle repeateduntil the acetone remained clear after cleaning. Powdersof olivine (Fo93) from Aheim, Norway; and Almklovdalenperidotite from Gusdal, Norway were obtained as 65–125and 45–125 lm size fractions, respectively, and cleaned inacetone as described above. The peridotite is mainlycomposed of Mg-rich olivine with minor amounts ofclinochlore, and is identical to the material described byWolff-Boenisch et al. (2011). A detailed petrologicaldescription of the Almklovdalen peridotite is provided inKostenko et al. (2002). The basaltic glass was collectedfrom the Stapafell Mountain in Southwest Iceland and isthe same as that described by Oelkers and Gislason(2001), Gislason and Oelkers (2003), Wolff-Boenisch et al.(2011), and Stockmann et al. (2011, 2012). The preparationof the 45–125 lm size fraction of the basaltic glass used inthis study is described in detail by Stockmann et al.(2011). After cleaning, all silicates were dried overnight at60 �C. The chemical composition of the six silicate materi-als, determined by X-ray Fluorescence spectrometry, islisted in Table 1. All silicates were examined by SEM–EDS prior to the experiments. All of the surfaces werefine-particle free and no secondary phases were observedto be present. It should be noted, however, that trace quan-tities of other phases, including calcite, could be present atconcentrations below the analytical detection limits. Thespecific surface area, ABET, of each cleaned silicate powderwas determined by the multi-point krypton adsorption BETmethod and is listed in Table 2, together with the calculatedgeometric surface area, Ageo, and the roughness factorequal to the ratio of the BET to the geometric surface area(ABET/Ageo).

3.2. Calcite

The calcite powder used in this study as a growth sub-strate originated as large transparent Iceland spar calcitecrystals collected from hydrothermal veins in basaltic trapsin Central Siberia. This calcite is identical to that describedby Pokrovsky et al. (2005) and Flaathen et al. (2011). Theseauthors reported that electron microprobe and total chem-ical analysis showed that this calcite contained less than0.5% impurities and that no other phases were detectedby X-ray diffraction (XRD). The size fraction of this calcitepowder was 100–200 lm and the specific surface area of thispowder was 370 cm2/g (Pokrovsky et al., 2005) as

Table 1Chemical composition of investigated minerals and rocks (in wt.%) obtained from XRF analysis.a

Mineral SiO2 Al2O3 Fe2O3 MgO CaO Na2O K2O TiO2 MnO P2O5 LOI Total

Augite 53.15 0.87 10.86 11.74 23.15 0.48 0.003 0.029 0.266 0.004 �0.78 99.76Basaltic glass 48.55 14.43 12.08 9.51 11.94 1.91 0.274 1.570 0.192 0.199 �0.66 100.00Enstatite 57.59 0.70 7.73 30.05 1.75 0.23 0.024 0.073 0.048 1.022 0.49 99.71Labradorite 53.54 29.42 0.40 0.00 12.15 4.28 0.370 0.093 0.005 0.015 0.03 100.30Olivine 42.76 0.27 7.81 48.15 0.36 0.01 0.019 0.021 0.110 0.010 0.45 99.97Peridotite 43.06 0.46 7.01 48.99 0.04 0.00 0.000 0.013 0.101 0.008 0.24 99.92

a Analyzed at the Grant Institute, University of Edinburgh, 2011.

Table 2Specific surface area of the mineral and rock powders used in this study.

Minerala Specific surface area (cm2/g) Surface roughness Density Molar volume (Vm)e

ABETc Ageo

d (ABET/Ageo) (g/cm3) (cm3/mol)

Augite 1528 ± 20 225 7 3.40 �65Basaltic glass 5878 ± 40 251 23 3.05 �42Enstatite 4505 ± 30 239 19 3.20 31.47Labradorite 694 ± 20 285 2 2.69 �106Olivine 3672 ± 40 200 18 3.27 43.786Peridotite 3286 ± 40Calciteb 370 ± 20 153 2 2.71 36.934

a The powders are the 45–125 lm size fraction except for peridotite (65–125 lm) and calcite (100–200 lm).b ABET from Pokrovsky et al. (2005).c Determined by multi-point krypton adsorption using the BET method.d Calculated using equations reported by Wolff-Boenisch et al. (2004a).e Source: Calcite, olivine and enstatite values are taken from Weast and Astle (1986), and labradorite and basaltic glass from Marini (2007).

No value was found for augite, but it is assumed close to the Vm of diopside and hedenbergite of �66 cm3/mol (Weast and Astle, 1986).

Fig. 1. Schematic illustration of the 30 mL reactors used in thepresent study. Calcite saturation in the experiments is obtained bymixing two aqueous inlet solutions comprised of Na2CO3/NaHCO3 and CaCl2, respectively, inside the reactor.

G.J. Stockmann et al. / Geochimica et Cosmochimica Acta 135 (2014) 231–250 233

determined by multi-point krypton absorption using theBET method. For a detailed description of the preparationand cleaning of this calcite powder see Pokrovsky et al.(2005) and Flaathen et al. (2011).

3.3. Dissolution–precipitation experiments in mixed-flow

reactors

Three types of experiments were performed in this study:(1) short-term calcite precipitation experiments in the pres-ence and absence of solid substrates, (2) short-term silicatedissolution experiments in the absence of secondary phaseprecipitation, and (3) long-term calcite precipitation exper-iments in the presence of solid substrates. All experimentswere performed in mixed-flow polypropylene reactors.The short-term calcite precipitation experiments were per-formed in the 30 mL reactors illustrated in Fig. 1. The otherexperiments were performed in 300 mL reactors of similardesign as that shown in Fig. 1. A weighed mass of selectedpowdered substrate was placed into the mixed-flow reactorwhich was then placed into a water bath kept at 25 ± 2 �C.The reactor was filled with the reactive fluid selected for theexperiment, the reactor sealed, and a constant inlet fluidflow was initiated. Reactors were continuously stirred withfloating Teflon stirring bars, and reactive fluid left the reac-tor through a 2.5 lm polypropylene filter. Additional filter-ing through a 0.2 lm cellulose acetate membrane filter wasperformed on the outlet fluids before chemical analysis.

Two types of inlet fluids were used in the experiments.Experiments designed to precipitate calcite were run using

an inlet fluid created by mixing two aqueous fluids insidethe reactor: (1) a Na-carbonate solution and (2) a CaCl2solution. The mixed fluid inside the reactor, prior to reac-tion, had a pH of 9.1, and a calcite saturation index of0.6. The Na-carbonate solution was comprised of deminer-alised H2O, NaHCO3 (99.5% purity), and Na2CO3 (>99%purity) from Sigma–Aldrich. The Ca-bearing solution wascomprised of H2O, CaCl2 (1000 ppm standard solution)from Merck, alkalized to pH �9 using a 1 mol/kg MerckNaOH solution. Concentrations of NaHCO3 and Na2CO3


in the carbonate solution were 1.68 � 10�2 and 1.29 � 10�3

mol/kg, respectively, and the CaCl2 solution had a Ca con-centration of 1.39 � 10�4 mol/kg. As these two fluids areinjected into the reactor at equal rates, the inlet reactor fluidconcentration is equal to half that of each of these two aque-ous solutions after mixing. Experiments designed to measuresilicate substrate dissolution in the absence of calciteprecipitation were run using an aqueous Na-carbonate solu-tion with a pH of 9.1. The inlet fluids for these experi-ments were comprised of 8.36 � 10�3 mol/kg NaHCO3 and0.64 � 10�3 mol/kg Na2CO3, so that the carbonate concen-tration in these experiments was equal to that in the reactorduring the calcite precipitation experiments.

Table 3aExperimental conditions and steady-state results of the mineral/rock diss

Exp.namea

Mass of solidsubstrate (g)

Init. BET surfacearea (m2)

Fluid flowrate (g/min)

Durationexp. (day

Substrate: Calcite

C-1 0.23 0.0085 0.42 7.1C-3 0.57 0.021 0.44 2.8C-4 0.18 0.0067 0.46 2.8C-6 0.26 0.010 0.43 7.0

Control 0.00 0.00 0.46 3.0

Substrate: Olivine

O-2 0.27 0.098 0.44 7.1O-3 0.25 0.091 0.44 5.7O-1-C 1.98 0.73 0.92 14.0O 100-P 1.98 0.73 �1.0 66.1

Substrate: Basaltic glass

BG-2b 0.26 0.15 0.43 8.7BG-3b 0.26 0.15 0.45 4.0

Substrate: Enstatite

E-1 0.27 0.12 0.42 8.7E-2b 0.26 0.12 0.47 4.0E-3b 0.25 0.11 0.40 5.5E-1-C 1.99 0.90 0.96 14.0E-100-P 1.99 0.90 �1.0 66.1

Substrate: Augite

A-1b 0.25 0.037 0.47 9.0A-2b 0.26 0.040 0.45 2.7A-3b 0.25 0.038 0.24 1.5A-4b 0.25 0.039 0.42 1.9A-1-C 1.98 0.30 0.88 14.0A-100-P 1.98 0.30 �1.0 66.1

Substrate: Labradorite

L-1 0.26 0.018 0.47 9.0L-1-C 2.01 0.14 0.90 14.0L-100-P 2.01 0.14 �1.0 66.1

Substrate: Peridotite

P-1-C 2.05 0.67 0.90 14.0P-100-P 2.05 0.67 �1.0 66.1

a The type of experiment is designated by its name. The first letter(s) interm calcite precipitation experiments are denoted by names that haveexperiments are denoted by names that have a “-C” suffix. Long-term ca

b The experiment did not attain a steady state [Ca2+] concentration prepresents the final measured outlet fluid concentration.

3.4. Analytical methods

The outlet fluids were collected and analyzed regularlyto monitor reaction progress. Outlet fluid pH was measureddirectly after sampling at 23 ± 2 �C. Inlet and outlet fluidscollected from the short-term calcite precipitation experi-ments were analysed using a Varian SpectrAA 300 atomicabsorption spectrometer for calcium, the “molybdate bluemethod“ using a Varian Spectrophotometer for silicon(Koroleff, 1976), and a Perkin Elmer Zeeman 5000 AtomicAbsorption Spectrometer (AAS) for magnesium. The majorelement concentrations of inlet and outlet fluids from allother experiments were determined using a Spectro Ciros

olution experiments performed at 25 ± 2 �C.

ofs)

Outlet fluidpH (25 �C)

Outlet fluid composition

[Si](lmol/kg)

[Mg](lmol/kg)

[Al](lmol/kg)

[Ca](lmol/kg)

9.11 22.509.11 41.009.10 41.909.11 40.06

9.12 67.28

9.14 0.69 1.05 43.649.15 41.579.13 2.04 2.429.17 0.59 1.23 23.70

9.13 1.98 0.50 66.139.13 65.63

9.11 0.64 0.36 33.049.17 57.809.17 42.739.14 1.77 1.539.15 0.66 1.00 26.20

9.14 0.73 64.609.14 65.739.14 69.789.17 68.039.14 1.49 1.029.24 0.93 0.90 63.20

9.14 0.61 40.989.14 1.81 1.479.17 1.15 0.86 37.20

9.15 1.29 1.789.17 0.55 1.00 27.90

the experiment name refers to the identity of the substrate. Short-a single number after the first dash. Names of silicate dissolutionlcite precipitation experiments are denoted by a “-P” suffix.rior to the end of the experiment. The [Ca2+] concentration listed


Vision Inductively Coupled Plasma Optical Emission Spec-trometer (ICP-OES).

Selected solid samples were examined by a HITACHIS3400N Scanning Electron Microscope (SEM) after goldcoating. Energy Dispersive X-ray Spectroscopy (EDS) per-formed via an Oxford Instruments spectrometer was usedtogether with the SEM to identify primary and secondarymineral phases. Cleaned silicate powders were analyzedby X-ray Fluorescence spectroscopy, X-ray diffraction,and SEM–EDS prior to experiments to determine theirchemical composition and purity. Additional analyses ofsolids with and without calcite precipitates were performedusing X-ray Photoelectron Spectroscopy (XPS). These

Table 3bSteady-state results of the mineral/rock dissolution experiments perform

Exp.namea

Experimentduration (days)

SI calciteoutlet fluid

�DGrb

(kj/mol)Calcite precipitatiLog r (mol Ca/s)

Calcite (CaCO3)

C-1 7.1 0.10 �9.55C-3 2.8 0.35 �9.69C-4 2.8 0.36 �9.62C-6 7.0 0.34 �9.71

Olivine (Mg1.75Fe0.14Si1.04O4)

O-2 7.1 0.40 �9.74O-3 5.7 0.39 �9.73O 1-C 14.0 �3.03 50.44O 100-P 66.1 0.13 55.35 �9.07

Basaltic glass (Si1.00Al0.35Fe0.19Mn0.003Mg0.29Ca0.26Na0.076K0.007Ti0.024P0.

BG-2 8.7 0.53BG-3 5.0 0.53

Enstatite (Mg0.78Fe0.10Si1.01O3)

E-1 8.7 0.26 �9.61E-2 4.0 0.54E-3 5.5 0.41 �9.82E 1-C 14.0 �2.36 27.74E 100-P 66.1 0.21 30.11 �9.37

Augite (Ca0.94Mg0.66Fe0.31Si2.00O6)

A-1 9.0 0.47A-2 2.7 0.56A-3 1.5 0.59A-4 1.9 0.61A 1-C 14.0 �1.06 49.76/84.27A 100-P 66.1 0.10 40.69/69.84 �9.25

Labradorite (Ca0.59Na0.38Al1.57Si2.42O8)

L-1 9.0 0.34 �9.67L 1-C 14.0 �1.59 63.08L 100-P 66.1 0.21 55.06 �9.12

Peridotite

P 1-C 14.0 �4.23 52.46P 100-P 66.1 0.17 56.59 �9.25

a The type of experiment is designated by its name. The first letter(s) interm calcite precipitation experiments are denoted by names that haveexperiments are denoted by names that have a “-C” suffix. Long-term ca

b Chemical affinity (�DGr) of the dissolving silicate taken as forsteritaugite, and as anorthite for labradorite, whereas enstatite log Q/K value

analyses were conducted using a Kratos Axis UltraDLD fittedwith a monochromated AlKa X-ray source. XPS is a surfacesensitive technique that quantifies the chemical compositionof the top 10 nm of the surface with a detection limit of0.1% (atomic%), thus allowing the detection of minutequantities of surface precipitates. The analysed spot size isapproximately 700 � 300 lm. Interpretation of XPS resultswas made with the commercial software CasaXPS, using aShirley background. Uncertainties of the analysis methodsused in this study, based on the reproducibility of the anal-ysis of standard materials, are estimated to be 2% for theAAS analysis, 3% for the spectrophotometer, 3–5% forthe ICP-OES analyses, and 10% for the XPS analysis.

ed in this study. All experiments were performed at 25 �C.

on rates Silicate dissolution rateLog r+,BET (mol/cm2/s)based on the release of:

Element release ratesLog r+BET (mol/cm2/s)

Si Mg Al Si Mg Al

�14.30 �14.34 �14.28 �14.10

�14.38 �14.54 �14.37 �14.29

004O3.37)

�14.04 �14.10 �14.04 �14.63

�14.44 �14.57 �14.43 �14.68

�14.51 �14.47 �14.50 �14.56

�14.12 �13.82

�14.44 �14.13 �14.14 �14.31

�13.95 �13.56�14.09 �14.00 �13.71 �13.80

�14.54 �14.40

the experiment name refers to the identity of the substrate. Short-a single number after the first dash. Names of silicate dissolutionlcite precipitation experiments are denoted by a “-P” suffix.

e log Q/K for olivine and peridotite, as diopside/hedenbergite fors exist.

Table 4Results of X-ray photoemission spectroscopy (XPS) analysis on silicate surfaces.

Sample:a Atomic ratio

C/Si O/Si Fe/Si Mg/Si Ca/Si Na/Si Ti/Si Cl/Si Al/Si N/Si

Olivine

Original powder 0.43 3.23 0.08 1.59 0.00 0.00 0.00 0.00 0.00 0.00O-2 MilliQ 1.31 3.46 0.10 1.44 0.03 0.02 0.00 0.03 0.00 0.12O-2 Cc 1.84 3.69 0.09 1.40 0.04 0.22 0.00 0.04 0.00 0.14O-100P MilliQ 0.72 3.23 0.12 1.52 0.02 0.00 0.03O-100P Cc 0.71 3.39 0.12 1.43 0.05 0.03 0.02

Augite

Original powder 0.35 2.15 0.09 0.37 0.16 0.01 0.00 0.00 0.00 0.00A-1 MilliQ 0.85 2.18 0.10 0.33 0.17 0.02 0.00 0.04 0.00 0.03A-1 Cc 1.25 2.13 0.08 0.28 0.15 0.04 0.00 0.03 0.00 0.07A-100P MilliQ 0.67 2.42 0.12 0.27 0.20 0.02 0.00A-100P Cc 0.51 2.32 0.11 0.27 0.19 0.06 0.00

Basaltic glass

Original powder 0.22 2.52 0.09 0.16 0.10 0.02 0.02 0.00 0.34 0.00BG-2 MilliQ 0.99 2.37 0.11 0.08 0.10 0.01 0.02 0.25 0.32 0.07BG-2 Cc 0.90 2.66 0.16 0.10 0.10 0.14 0.03 0.11 0.33 0.05

Enstatite

Original powder 0.13 2.15 0.07 0.60 0.00 0.00 0.00 0.00 0.00 0.00E-1 MilliQ 1.04 2.14 0.06 0.59 0.07 0.00 0.00 0.14 0.00 0.08E-100P MilliQ 0.42 2.32 0.10 0.62 0.03 0.00 0.00E-100P Cc 0.35 2.24 0.08 0.59 0.02 0.03 0.00

Labradorite

Original powder 0.61 2.35 0.00 0.00 0.12 0.04 0.00 0.00 0.59 0.00L-1 MilliQ 1.12 2.25 0.00 0.00 0.10 0.17 0.00 0.00 0.57 0.00L-1 Cc 0.95 2.35 0.00 0.00 0.10 0.22 0.00 0.00 0.54 0.02L-100P MilliQ 0.57 2.32 0.00 0.00 0.13 0.07 0.00 0.58L-100P Cc 1.51 2.51 0.00 0.00 0.10 0.16 0.00 0.55

Peridotite

Original powder 0.13 2.82 0.05 1.55 0.00 0.00 0.00 0.00P-100P MilliQ 0.58 3.11 0.14 1.43 0.04 0.00 0.00 0.00P-100P Cc 0.67 3.29 0.15 1.39 0.03 0.06 0.00 0.00

a Duplicate analysis were performed on each sample, and it is the average values that are listed in this table. The terms “MilliQ” and “Cc”

refer to the post-experimental powders which were cleaned with either MilliQ water or with a calcite saturated solution.


4. EXPERIMENTAL RESULTS

An overview of the experimental conditions and resultsobtained from all experiments is provided in Tables 3a and3b. The type of experiment is designated by its name. Thefirst letter(s) in the experiment name refers to the identityof the substrate; calcite (C), augite (A), enstatite (E), olivine(O), labradorite (L), basaltic glass (BG), and peridotite (P).Short-term calcite precipitation experiments are denoted bynames that have a single number after the first dash. Namesof silicate dissolution experiments are denoted by names thathave a “–C” suffix. Long-term calcite precipitation experi-ments are denoted by a “–P” suffix. The results of XPS sur-face analysis of the pre- and post-experimental silicatepowders are listed in Table 4. The steady-state calcite precip-itation rates listed in Tables 3a and 3b and the average calciteprecipitation rates reported in Table 5 were calculated fromthe measured Ca concentration using:

r ¼ jCCa;outlet � CCa;inlet � DCCa;silicatej � fr ð4Þ

where r refers to non-surface area normalized calcite precip-itation rates, CCa,outlet and CCa,inlet stand for the Ca concen-tration of the outlet and inlet fluid, DCCa,silicate designatesthe change in fluid Ca concentration due to the dissolutionof silicate substrate, and fr refers to the fluid flow rate. TheDCCa,silicate term in Eq. (4) was determined from measuredoutlet fluid Si, Mg, and/or Al concentrations assuming stoi-chiometric element release from the dissolving silicate sub-strate; this term was set equal to zero in the absence ofsilicate substrates. Note that Eq. (4) yields calcite precipita-tion rates that are not normalized to the substrate surfacearea in the reactor. As there may be a link between sub-strate surface area and precipitation rates, the degree towhich they are proportional to one another in these exper-iments will be discussed below.

Measured outlet fluid concentrations of Si, Mg, and/orAl from experiments containing silicate glass, rock, or min-eral substrates were used to generate surface area normal-ized dissolution rates based on the release of the jthelement (rs,i,j) using:

Table 5Average calcite precipitation rates during short-term precipitation experiments at pH 9.1, 25 �C and SIcalcite = 0.6.a

Solidphase

Ca2+contribution fromsolid phase

ntotal

calcitemtotal

calciteExp.duration

Average calcite (cc)growth rate

Average calcite precipitation ratec

Log r Log r,BET Log r,geo

(mmol) (mg) (days) (mg cc/day) mol/s mol/cm2/s mol/cm2/s

Calcite 1 mol Ca/mol calciteC-1 0.20 20.02 7.1 2.82 �9.49 �11.43 �11.04C-3 0.050 5.00 2.8 1.79 �9.68 �12.01 �11.63C-4 0.057 5.71 2.8 2.04 �9.63 �11.45 �11.07C-6 0.11 11.01 7.0 1.57 �9.74 �11.73 �11.35

Olivine NoneO-2 0.085 8.51 7.1 1.20 �9.86O-3 0.065 6.51 5.7 1.14 �9.88

Basaltic glass 0.263 mol Ca/mol BGBG-2 0.015 1.50 8.7 0.17 �10.70BG-3 0.029 2.90 5.7 0.51 �10.23

Enstatite 0.03 mol Ca/mol enstatiteE-1 0.15 15.01 8.7 1.72 �9.70E-2 0.076 7.61 4.7 1.62 �9.73E-3 0.047 4.70 5.5 0.86 �10.00

Augite 0.94 mol Ca/mol augiteA-1 0.037 3.70 9.0 0.41 �10.32A-2 0.005 0.50 2.7 0.19 �10.69A-3 0.0005 0.05 1.5 0.03 �11.41A-4 0.003 0.30 1.1 0.27 �10.50

Labradorite 0.59 mol Ca/mol labradoriteL-1 0.12 12.01 9.0 1.33 �9.81

a The table list total moles of calcite (n), total mass of calcite (m), duration of experiment, growth of calcite mass per day, and average calciteprecipitation rates of non-normalized and surface-normalized (rBET) substrate areas. Molar volume and density of calcite are taken fromTable 2.

c Average rates from dividing total moles of calcite precipitated by the duration of experiment.

2 Note the use of the term steady state in this context is pragmaticand refers to a condition where outlet fluid concentrations do notvary within analytical uncertainty for more than 10 residence times.Residence time is equal to the volume of the reactor divided by thefluid flow rate.


rs;i;j ¼DCj � frmj � Ai � m

ð5Þ

where DCj stands for the difference in concentration of thejth element between the outlet and the inlet fluid, mj refers toa stoichiometric factor equal to the number of moles of thejth element in one mole of the substrate, Ai designates thespecific surface area of this substrate, and m denotes the ini-tial mass of the substrate used in the experiment. UsingBET or geometric surface areas (ABET or Ageo) in Eq. (5)yields rBET,i,j or rgeo,i,j, respectively. No corrections havebeen made in Eq. (5) for the change in substrate mass asa result of its dissolution. Calculations show the changein mass affects negligibly the dissolution rates obtained inthis study.

4.1. Short-term calcite precipitation experiments

Short-term calcite precipitation experiments were per-formed in the presence of either calcite, five different silicatesubstrates (augite, enstatite, olivine, basaltic glass, labra-dorite) or in the absence of mineral substrates under identi-cal experimental conditions. Each experiment lasted for 2–9 days. The temporal evolution of the Ca outlet fluid con-centration in each experiment is shown in Fig. 2a–f. Inletfluid Ca concentrations are represented as a solid line onthese plots, and a decrease in Ca outlet concentration com-pared to the inlet is interpreted to be due to calcite precip-

itation. In the experiments where calcite was the substrate,the Ca outlet fluid concentration attained a steady-state2

value within 1 day, as shown in Fig. 2a. In contrast, asteady-state Ca concentration is attained after 4–5 days inthe experiments where olivine and enstatite were substrates(Fig. 2b and d, respectively). A steady-state outlet fluid Caconcentration was not attained during �10 day experimentswhen basaltic glass, augite, or labradorite was the growthsubstrate (see Fig. 2c, e, and f, respectively). In Fig. 3, thecalcite precipitation rates of the experiments with calciteas the substrate (Fig. 3a) and with silicates as the substrate(Fig. 3b) are plotted as a function of elapsed time. The dataillustrated in these figures indicate that the calcite precipita-tion rates for all the calcite substrate experiments and forsome of the silicate substrate experiments (labradorite, oliv-ine and enstatite) attain a steady-state value of�10�9.68±0.08 mol/s. In addition, the temporal evolutionof the calcite precipitation rates of the experiments per-formed in the presence of basaltic glass and augite suggeststhat the steady-state calcite precipitation rates in theseexperiments will eventually be similar to those obtained in

O-1O-2

BG-1

C-1C-3C-4C-6

BG-3

0 1 2 3 4 5 6 7 8 9 10

Ca

[mm

ol/k

g]

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

Ca inlet

Calcite SI = 0.0

a

0 1 2 3 4 5 6 7 8 9 10

Ca

[mm

ol/k

g]

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

Ca inlet

Calcite SI = 0.0

b

0 1 2 3 4 5 6 7 8 9 10Elapsed time [days]

Ca

[mm

ol/k

g]

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

Ca inlet

Calcite SI = 0.0

c Calcite precipitation on basaltic glass

Calcite precipitation on calcite

Calcite precipitation on olivine

Calcite precipitation on enstatite

Calcite precipitation on augite

Calcite precipitation on labradoriteC

a [m

mol

/kg]

Ca

[mm

ol/k

g]C

a [m

mol

/kg]

Calcite SI = 0.0

Ca inlet

Calcite SI = 0.0

Ca inlet

Calcite SI = 0.0

Ca inlet

Elapsed time [days]

E-1E-2E-3

A-1A-2A-3A-4

L-1

0 1 2 3 4 5 6 7 8 9 100.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

d

0 1 2 3 4 5 6 7 8 9 100.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

e

0 1 2 3 4 5 6 7 8 9 100.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

f

Fig. 2. Temporal evolution of Ca concentration in inlet and outlet fluids from the short-term mixed-flow reactor experiments containingcalcite and silicate substrates at pH 9.1 and 25 �C. The solid lines represent the Ca concentration of the inlet fluids, and the dashed linesrepresent the Ca concentration of calcite saturation (SI = 0). The error bars on the plots correspond to ±0.005 mmol/kg.


the presence of calcite and the other silicate substrates (seeFig. 3b). This observation implies, that once nucleation iscomplete, growth continues at an approximately constantand similar rate regardless of the substrate identity. Thisobservation (1) justifies the use of non-substrate surfacearea normalized rates to describe calcite precipitation inthis study and (2) suggests minor amounts of Si, Mg, Fe,and Al in the fluid phase does not significantly affect calciteprecipitation rates at these conditions. The induction timefor calcite precipitation on the different substrates can beestimated from Fig. 3a and b. For calcite and all silicatemineral substrates, precipitation is measurable within halfa day, whereas for basaltic glass precipitation is not evidentfor up to 5 days.

No difference in the Ca concentration was observed forinlet versus outlet solution in the substrate-free experimentreferred to as “control” in Table 3a. This is consistent withthe absence of calcite precipitation in this experiment andwith previous results showing that calcite will not precipi-tate in the absence of growth substrates at this low degreeof supersaturation (e.g. Lioliou et al., 2007; Hu et al.,2012). This result indicates that either the induction timefor calcite precipitation in these substrate-free systems islonger than the experiment durations or that calcite

precipitation will not occur in the absence of suitable sub-strates at these low saturation states.

The Si and Mg concentrations in the outlet fluids werealso measured during the silicate substrate experiments,and the calculated mineral/glass dissolution rates basedon Si and Mg release are shown in Fig. 4a–e. These ratesare calculated taking into account the mineral or glass stoi-chiometric formulas provided in Table 3b. Element releaserates are also listed in Table 3b. Results shown in Fig. 4suggest that the dissolution of all silicate substrates attaineda steady-state rate during the experiments.

4.2. Short-term silicate mineral/rock dissolution experiments

These experiments were run to assess the dissolutionrates of the silicate substrates in the absence of calcite pre-cipitation. The majority of these silicate substrates containonly small amounts of Ca (see Table 1), thus release of Cafrom the silicates during these experiments would not leadto calcite precipitation. This was confirmed by PHREEQCmodelling of the outlet solutions; all outlet fluids of theseexperiments were undersaturated with respect to calcite(see Table 3b). Fig. 5a–d illustrate the temporal evolutionof the dissolution rates of olivine, enstatite, augite, and

-11.0

-10.8

-10.6

-10.4

-10.2

-10.0

-9.8

-9.6

-9.4

-9.2

-9.0

0 1 2 3 4 5 6 7 8

Elapsed time [days]

C-1 (0.23 g)

C-3 (0.57 g)

C-4 (0.18 g)

C-6 (0.26 g)

Log

r [ m

ol C

a / s

]

a

b

Log

r [ m

ol C

a / s

]

-12.5

-12.0

-11.5

-11.0

-10.5

-10.0

-9.5

0 1 2 3 4 5 6 7 8 9 10

Elapsed time [days]

C-4 C-6 O-2 O-3 BG-2 BG-3 E-1 E-3 L-1 A-1

+/-0.1

Fig. 3. (a) Calcite precipitation rates for the short-term experiments, where calcite is the substrate. Rates become independent of substrate andmass after 0.2–3 days. (b) Calcite precipitation rates for the short-term experiments, where silicate substrates are present; olivine (O), basalticglass (BG), enstatite (E), labradorite (L) and augite (A). Two calcite substrate experiments (C) are added to this plot for comparison. Theerror bars on the plots correspond to ±0.10 log units uncertainty on the rates.


labradorite, respectively, and Fig. 5e shows the Si and Mgelement release rates from peridotite dissolution. Both dis-solution rates and element release rates for each silicatephase are listed in Table 3b. Silicate dissolution ratesobtained from these calcite-free experiments are close tothe corresponding rates obtained from the calcite precipi-tating experiments (see Table 3b). Mineral saturation statescalculated from PHREEQC modelling confirmed that allexperiments were conducted at far-from-equilibrium condi-tions with respect to the dissolving primary solid phases.

4.3. Long-term calcite precipitation on silicate minerals/rock

The main purpose of these experiments was to exposethe silicate powders to calcite-saturated fluids for sufficienttime to allow for substantial calcite growth on the silicatesurfaces. For this purpose, at the conclusion of the calciteprecipitation-free experiments described above in Sec-tion 4.2, a new inlet fluid, identical to that added for theshort-term calcite precipitation experiments was passedthrough the reactors. These ‘long-term’ experiments ranfor approximately 70 days; the temporal evolution of the

fluids in these experiments is illustrated in Fig. 6. The ele-ment concentrations of Si, Mg, and Al are generally stableor increase somewhat after �50 days from the start of theseexperiments (Fig. 6a–e); these observations indicate thatcalcite precipitation did not slow element release from thesilicates. In contrast, outlet fluid Ca concentrationsdecrease with elapsed time and attain a steady-state concen-tration after �40 days corresponding to a calcite SI of 0.1–0.2 in most of these experiments (Fig. 6f). As such, thecalcite precipitation rates found in these long-term experi-ments attained a steady-state rate with the exception ofaugite; these rates are close to those of the short-term exper-iments with an average rate of �10�9.21±0.2 mol/s (seeFig. 7). Chemical modelling of outlet solutions show all pri-mary silicate phases to be at far-from-equilibrium duringthese experiments (see Table 3b).

4.4. Results of SEM and XPS analysis

All pre- and post-experimental powders recovered fromthe short-term and long-term calcite precipitation experi-ments were analyzed by SEM and XPS with the exception

SiMg

SiMg

O livine d isso lution ra tes

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

0 1 2 3 4 5 6 7 8 9 10

Log

r +, B

ET [

mol

/ cm

2 / s

]SiMg

a

Basaltic glass d isso lution ra tes

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

0 1 2 3 4 5 6 7 8 9 10

Log

r +, B

ET [

mol

/ cm

2 / s

]

b

Enstatite d isso lution ra tes

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

0 1 2 3 4 5 6 7 8 9 10Elapsed tim e [days]

Log

r +, B

ET [

mol

/ cm

2 / s

]

c

-12.5

-13.5

Si

Aug ite d isso lution ra tes

0 1 2 3 4 5 6 7 8 9 10

Log

r +, B

ET [

mol

Si /

cm

2 / s

]

Sid

Labradorite d isso lution ra tes

0 1 2 3 4 5 6 7 8 9 10

Log

r +, B

ET [

mol

Si /

cm

2 / s

]

e

-15.0

-14.5

-14.0

-13.0

-12.5

-15.0

-14.5

-14.0

-13.5

-13.0

E lapsed tim e [days]

Fig. 4. Temporal evolution of silicate substrate dissolution rates based on Si and/or Mg release of (a) olivine, (b) basaltic glass, (c) enstatite,(d) augite, and (e) labradorite during short-term calcite precipitation experiments at pH 9.1 and 25 �C. Black circles represent Si rates, andopen triangles represent Mg rates. The error bars on all plots correspond to ±0.15 log units uncertainty on the rates.


for the labradorite from the short-term experiment whichonly was analyzed by XPS. Both techniques confirmed calciteprecipitation on the silicate substrates during the calcite pre-cipitation experiments. The most extensive calcite growth wasfound on enstatite, olivine, and peridotite, and least on basal-tic glass as illustrated by the SEM images in Fig. 8 and assummarized in Table 6. Labradorite and augite showed inter-mediate calcite coverage. Surface analysis of the silicates byXPS (Table 4) shows an increase in the Ca/Si ratio of the sur-faces, most notable from olivine, enstatite and peridotite, con-firming the results of fluid chemistry analyses and SEMobservations. Interestingly, calcite growth on labradoritewas not detected by XPS although calcite did form accordingto the SEM images. In the case of labradorite, calcite growthoccurred via formation of a few, large (�60 lm) calcite crys-tals in contrast to the smaller but more numerous calcite crys-tals on enstatite and olivine (see Fig. 8). No secondary phasesother than calcite were detected by SEM-EDS analysis onthese post-experimental samples.

4.5. Calcite precipitation rates

The calcite precipitation rates obtained during theexperiments described in Section 4.1 are listed in Table 5.

Rates were calculated from the decrease in reactive fluidCa concentration after taking account of the Ca added tothe fluid phase by silicate seed dissolution. In cases wherecalcite substrates were present, rates normalized to sub-strate surface area are also provided. What is apparent fromthese data is that with time, calcite precipitation rates becomeindependent of the identity of the substrate and its surfacearea, which again supports the use of non-substrate surfacearea normalized rates to describe calcite precipitation.

5. DISCUSSION

5.1. Comparison with silicate dissolution rates from the

literature

Silicate dissolution rates obtained from this study arecompared to those from the literature in Table 7. Numer-ous dissolution rate data have been reported in the litera-ture for olivine (e.g. Grandstaff, 1977; Kuo andKirkpatrick, 1985; de Leeuw et al., 2000; Rosso andRimstidt, 2000; Pokrovsky and Schott, 2000a,b; Oelkers,2001; Morales and Herbert, 2002; Olsen and Rimstidt,2008; Prigiobbe et al., 2009), diopside (e.g. Knauss et al.,1993; Chen and Brantley, 1998; Golubev et al., 2005;

SiMg

SiMg

SiMg

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

Log

r +, B

ET [

mol

/ cm

2 / s

]

E lapsed tim e [days]0 2 4 6 8 10 12 14 16

Augite d isso lutionc

Log

r +, B

ET [

mol

/ cm

2 / s

]

0 2 4 6 8 10 12 14 16

Enstatite d isso lutionb

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

Log

r +, B

ET [

mol

/ cm

2 / s

]

0 2 4 6 8 10 12 14 16

Olivine d isso lutiona

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.0

-13.2SiMg

SiAl

Log

r +, B

ET [

mol

/ cm

2 / s

]0 2 4 6 8 10 12 14 16

Peridotite element releasee

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

Log

r +, B

ET [

mol

/ cm

2 / s

]

0 2 4 6 8 10 12 14 16

Labradorite d isso lutiond

Elapsed tim e [days]

-15.0

-14.8

-14.6

-14.4

-14.2

-14.0

-13.8

-13.6

-13.4

-13.2

-13.0

Fig. 5. Temporal evolution of silicate mineral dissolution rates based on Si, Mg and/or Al release: (a) olivine, (b) enstatite, (c) augite, and (d)labradorite, and (e) element release rates of peridotite during silicate substrate dissolution experiments at pH 9.1 and 25 �C. These results arefrom pure dissolution experiments with no calcite or other secondary phases precipitating. The inlet fluids for these experiments consisted of8.36 mM NaHCO3/0.64 mM Na2CO3. The dashed lines on the plots represent steady-state dissolution rates obtained from these experiments.The error bars on all plots correspond to ±0.15 log units uncertainty on the rates.


Golubev and Pokrovsky, 2006; Dixit and Caroll, 2007;Daval et al., 2010), enstatite (e.g. Oelkers and Schott,2001; Halder and Walther, 2011) and basaltic glass (e.g.Oelkers and Gislason, 2001; Gislason and Oelkers, 2003;Wolff-Boenisch et al., 2004a,b, 2006, 2011; Stockmannet al., 2011, 2012), whereas only limited data are availablefor augite (e.g. Siegel and Pfannkuch, 1984; Schott andBerner, 1985; Sverdrup, 1990; McAdam et al., 2008) andlabradorite (Sjoberg, 1989; Welch and Ullman, 1993;Taylor et al., 2000; McAdam et al., 2008) and none at basicpH. It can be seen in Table 7 that measured basaltic glassand olivine dissolution rates are identical to previous find-ings. Measured augite dissolution rates are �0.8 log unitsfaster than those of diopside (CaMgSi2O6), a composition-ally simpler Ca-clinopyroxene. This comparison suggeststhat augite dissolution might be faster than diopside.Hoch et al. (1996) compared the dissolution rates of diop-side and augite at pH 5.8 and 25 �C with varying dissolvedO2 concentrations and observed diopside dissolutionrates to be independent of dissolved O2, whereas augite

dissolution rates increased with increasing dissolved O2

concentration becoming up to three times faster than thecorresponding diopside rates. They concluded that theseobservations could be due to a rate enhancement originat-ing from the oxidation of iron in the pyroxene structure.In contrast, Schott and Berner (1983) reported bronzite dis-solution rates at acidic pH and ambient temperature andfound that they were slower in oxic rather than anoxic con-ditions. They ascribed this difference to the formation of aFe3+–silicate surface layer, which impeded bronzite dissolu-tion at oxic conditions. Similarly, Saldi et al. (2013)reported that the formation of Fe(III)–Si-rich protectivelayers at oxic conditions slowed substantially olivine disso-lution rates. Enstatite dissolution rates measured in thisstudy are �0.7 log units faster than corresponding ratesreported in the literature. Such differences could be relatedto how this silicate was prepared for the experiments.Schott et al. (2012) emphasised the potential role of grind-ing, leading to additional reactive surface sites and thus fas-ter dissolution rates.

Enstatite

0.000

0.001

0.002

0.003

0.004

0.005

0 10 20 30 40 50 60 70

Con

c. [m

mol

/ kg

]

b

Olivine

0.000

0.001

0.002

0.003

0.004

0.005

0 10 20 30 40 50 60 70

Con

c. [m

mol

/ kg

]

a

Augite

0.000

0.001

0.002

0.003

0.004

0.005

0 10 20 30 40 50 60 70Elapsed tim e [days]

Con

c. [m

mol

/ kg

]

c

Periodotite

0.000

0.001

0.002

0.003

0.004

0.005

0 10 20 30 40 50 60 70C

onc.

[mm

ol /

kg]

e

Labradorite

0.000

0.001

0.002

0.003

0.004

0.005

0 10 20 30 40 50 60 70

Con

c. [m

mol

/ kg

]

d

0.000

0.020

0.040

0.060

0.080

0.100

0.120

0 10 20 30 40 50 60 70

Con

c. [m

mol

/ kg

]f

E lapsed tim e [days]

SiMg

SiMg

SiMg

SiMg

SiAl

LabradoritePeridotiteAugiteEnstatiteOlivine

Fig. 6. Temporal evolution of Si, Mg, and Al concentration in the outlet fluids from the long-term calcite precipitation experiments in thepresence of (a) olivine, (b) enstatite, (c) augite, (d) labradorite, and (e) peridotite at pH 9.1 and 25 �C, and (f) the temporal evolution of Caconcentration in all experiments. These experiments used two aqueous inlet fluids containing CaCl2 and NaHCO3/Na2CO3, respectively, tocreate calcite saturation inside the reactors. The error bars on the plots correspond to ±0.0003 mmol/kg on a–e, and to ±0.003 mmol/kg on f).


5.2. The effect of calcite precipitation on silicate dissolution

rates

Several studies have investigated the effect of carbonatecoatings on different silicate substrates (Giammar et al.,2005; Bearat et al., 2006; Andreani et al., 2009; Davalet al., 2009a,b; Stockmann et al., 2011, 2013; Fernandez-Martinez et al., 2012; Hovelmann et al., 2012; Saldi et al.,2013). Some of these past studies observed decreasing sili-cate dissolution rates in the presence of precipitated carbon-ates (Bearat et al., 2006; Andreani et al., 2009; Daval et al.,2009a,b). In these studies, carbonates (calcite or magnesite)were intergrown with amorphous Si in continuous layers,which might account for a slowing of the dissolution rateof the underlying mineral. Saldi et al. (2013) found that nei-ther magnesite precipitation nor pure silicate surface layersaffected olivine dissolution rates. As mentioned above, apassivating effect was found, however, by the formationof a Fe(III)–Si surface layer. Hovelmann et al. (2012)reported carbonation experiments on Almklovdalenperidotite from the same locality as that used in the present

study, and observed the growth of magnesite and amor-phous silica around olivine grains. The silica precipitates,however, had spherical shapes, which did not influence oliv-ine dissolution rates. Note there was no evidence of amor-phous Si layer growth during the experiments performed inthe present study. Calcite growth in this study was observedmainly as individual rhombohedral crystals of up to 60 lmin size growing on kinks, edges, and corners of the silicatesubstrate surfaces according to SEM analysis (see Fig. 8).Especially in the case of labradorite, calcite growth was aslarge single crystals. For the case of enstatite, these calcitecrystals were observed to completely cover some of the sub-strate grains (Fig 8d). For the most part, however, as evi-denced by SEM observations, the calcite precipitated indiscrete locations leaving much of the original silicate sub-strate surfaces in contact with the reactive fluid. Theseobservations are validated by the results of XPS analysis,which shows substantial Si to be present at the near surfaceof substrate grains recovered after the experiments. Theseobservations in our study are consistent with those ofCubillas et al. (2005), who concluded that the slowing of

Log r

[ m

ol C

a / s

]

-10.5

-10.3

-10.1

-9.9

-9.7

-9.5

-9.3

-9.1

-8.9

-8.7

-8.5

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70Elapsed time [days]

AugiteEnstatiteLabradoriteOlivinePeridotite

+/- 0.2

Fig. 7. Calcite precipitation rates of the long-term experiments. The rates approach an average steady-state rate of log r = �9.21 ± 0.2 mol/s.


the dissolution rates of a mineral substrate requires a closematch between the crystallographic structure of the sub-strate and the surface precipitate. None of the silicates usedas calcite growth substrates in the present study has a hex-agonal structure, as does calcite. As such, the poor crystalmatch between the substrate and precipitating calcite leavessufficient porosity near the substrate surface to permit itsdissolution. Moreover, a comparison of the rates compiledin Table 7 shows that in no case are the silicate ratesobtained from experiments where calcite was precipitatingin the present study substantially lower than correspondingrates reported in the literature in calcite-precipitation-freeexperiments. This observation supports the conclusion thatcalcite precipitates at silicate mineral surfaces do notdecrease silicate dissolution rates substantially, at least forthe minerals and conditions considered in this study.

5.3. Calcite precipitation rates

A large number of past studies have been aimed at quan-tifying calcite nucleation and precipitation rates (e.g.Meyer, 1984; Zhong and Mucci, 1989, 1993; Dove andHochella, 1993; Davis et al., 2000; Vavouraki et al., 2008;Larsen et al., 2010; Flaathen et al., 2011; Rodriguez-Blanco et al., 2011, 2012). Many of these studies noted thatthe presence of aqueous trace metals and certain anions canstrongly influence calcite growth rates. Examples of aque-ous species that can inhibit calcite growth, at least at certainconditions, include inorganic phosphate (e.g. Plant andHouse, 2002), Mg2+ ions (e.g. Zhang and Dawe, 2000;Rodriguez-Blanco et al., 2012), and natural organic materi-als (e.g. Lin et al., 2005; Lakshtanov et al., 2011; Nielsenet al., 2012). Such inhibition, however, was not observedin this study. All non-surface area normalized calcite pre-

cipitation rates trended to the same value despite the pres-ence of dissolving Mg bearing substrates in some of theseexperiments (see Table 5). This observation suggests thatthe presence of aqueous Mg did not significantly affect cal-cite precipitation at the conditions adopted in this study.The lack of Mg inhibition on rates in this study may bedue to a number of factors including the low reactive fluidMg concentration. Some have reported that the degree towhich an aqueous species affects mineral surface reactivitymay also depend on fluid pH (e.g. Alkattan et al., 2002).

5.4. Parameters controlling calcite nucleation

A number of recent studies have attempted to interpretcalcite nucleation in terms of (1) a reduction in the thermo-dynamic barrier due to decreased interfacial free energyconsistent with classic nucleation theory or (2) throughthe formation of nucleation clusters and an amorphous cal-cite carbonate (ACC) precursor (e.g. Rodriguez-Blancoet al., 2011; Hu et al., 2012; Teng, 2013). Hu et al. (2012)showed that at relatively low solute concentrations, nucle-ation occurs ion-by-ion, consistent with classic nucleationtheory, as the kinetic barrier to building an ordered nucleusfrom ions is much less than that of desolvating and orderingclusters. There is no indication in the results of this studythat an amorphous calcite carbonate (ACC) precursorformed, and all experiments were performed in mildlysupersaturated fluids. It is reasonable, therefore, to assumethat calcite nucleation in this study can be interpreted usingclassical nucleation theory. According to this theory, twomain parameters control the height of the thermodynamicbarrier for nucleation: (1) the interfacial energy, and (2)the degree of fluid supersaturation with respect to the nucle-ating solid.

Fig. 8. Scanning Electron Microscope images of pre- and post-experimental solid samples from short-term and long-term calcite precipitationexperiments of; (a) augite, (b) basaltic glass, (c) diopside, (d), enstatite, (e) labradorite, (f) olivine, (g) peridotite, and (h) calcite crystals afterhaving been exposed to reactive fluids supersaturated with respect to calcite for 7–9 days in the short-term experiments and 2–3 months in thelong-term experiments.


For the case of heterogeneous nucleation, three differentinterfacial energies come into play; (1) liquid–nucleatingcrystal, (2) liquid–substrate, and (3) the substrate–crystalinterfacial energy (Fernandez-Martinez et al., 2012). Thepresence of a mineral surface can lower the free energy bar-rier for heterogeneous nucleation if the interfacial freeenergy of the substrate–crystal interface is lower than thatof the liquid–substrate interface (De Yoreo and Vekilov,2003). This can be seen as a competition between the fluidand the nucleating crystal for the substrate, and impliesthat substrate hydrophilicity/hydrophobicity and latticemismatch are the key parameters controlling heterogeneous

nucleation (Fernandez-Martinez et al., 2012). Direct quan-titative application of these formalisms to the experimentsin this study is, however, not currently possible due to thedearth of quantitative information on these interfacial ener-gies in the calcite-silicate systems. This theoretical formal-ism is, nevertheless, consistent with the observation that acritical requirement for heterogeneous nucleation and sub-sequent crystal growth is some degree of molecular recogni-tion between the seeding substrates and the overgrowthmaterial (e.g. Stumm and Morgan, 1996). If the crystalstructure of the substrate surface closely matches a particu-lar plane in the nucleating phase, then the interfacial free

Fig 8. (continued)


energy becomes low, and nucleation occurs preferentiallyon that plane (De Yoreo and Vekilov, 2003). The resultsof the present study indicate that the matching of the crystalstructures between calcite and a variety of silicate minerals,including olivine, enstitite and laboradorite was sufficientfor catalyzing calcite precipitation in pH 9.1 solutions hav-ing a saturation index of 0.6 at 25 �C. Catalysis by the sub-strate is demonstrated by the presence of precipitatedcalcite in the experiments containing these substrates within3 days and lack of calcite precipitation in substrate free con-trol experiments over this time frame. Curiously, Lin andSinger (2005) report that the presence of quartz and dolo-mite seeds was insufficient to promote the precipitation ofcalcite at 25 �C in a fluid having a saturation index of0.72 and a pH �8. Similarly, calcium carbonate precipita-tion was not observed by Lioliou et al. (2007) in the pres-ence of quartz at 25 �C in fluids at pH = 8.5 having a

saturation index of up to 0.95; at saturation indexesgreater than �1 calcite was observed to nucleate sponta-neously in the absence of growth substrates. Theseauthors nevertheless argued, based on lattice compatibilityconsiderations (Lonsdale, 1968) and the Royer–Friedelrule (Turnbull and Vonnegut, 1952) that quartz surfacesare sufficiently similar to be favorable for epitaxial calcitegrowth.

Another factor that could influence the nucleation on asubstrate is the presence in the substrate of elementsrequired for the precipitation of the secondary phase (e.g.Putnis, 2009). This process could be driven by enhancedsecondary phase supersaturation near the substrate surface.This, however did not appear to be the case in this study;the calcium-free substrates olivine and enstatite promotedcalcite nucleation to a greater extent than the calcium-bearing basaltic glass and augite.

Table 6Calcite coverage of silicates from visual observations of SEM images.

Experiment Short-term Short-term Long-term Long-termRange of calcite coverage (%) Average coverage (%) Range of calcite coverage (%) Average coverage (%)

Augite 0–15 2 0–5 1–2Basaltic glassa 0–1 0 0 0Diopsidea No exp. No exp. 10–90 50Enstatite 10–90 10 1–5 2Labradoriteb No SEM data No SEM data 0–2 1Olivine 2–30 5–10 1–50 5Peridotite No exp. No Exp. 0–10 2

a Long-term precipitation data taken from Stockmann et al. (2011, 2013) are included for comparison. These ran at calcite saturation,SI = 1.6 and pH �8.

b XPS analysis of this sample shows no increase in Ca concentration on the surface (see Table 4).

Table 7Comparison of steady-state dissolution rates (Log r+) based on Si release in mol/cm2/s at 25 �C.

Mineral/rocka This study(pH 9.1)Log r+,BET

a

Literature(pH 9)Log r+,BET

b

Comments References

Basaltic glass �14.0 �14.2 Gislason and Oelkers(2003)

Augiteb �14.3 �15.1 This is the literature rate reported for diopside(Ca0.99Mg0.98Fe0.02Cr0.01Si2O6) at pH 8.7

Golubev et al. (2005)

Enstatite �14.5 �15.2 This is the literature rate for an enstatite with the formulaMg0.849Fe0.136Ca0.004Si1.002O3

Oelkers and Schott(2001)

Labradoriteb �14.0Peridotitec �14.5 As olivine is the dominant mineral phase, it can be compared

to the olivine value below from Pokrovsky and Schott (2000a,b)Olivine (Fo93) �14.3 �14.5 This is the rate reported for olivine with the composition

Mg1.82Fe0.18SiO4 (Fo91) at pH 9.3Pokrovsky and Schott(2000a,b)

a Chemical composition of solid phases provided in Table 1, Table 3a and Table 3b.b No literature data available for dissolution of augite and labradorite at alkaline pH.c Si element release rate. The peridotite is mainly composed of Mg-rich olivine and minor amounts of clinochlore.


5.5. Implications for mineral carbonation efforts

A large current effort is being made to determine if sub-surface mineral carbonation could be a viable option forthe long-term storage of anthropogenic carbon dioxide(c.f. Lackner et al., 1995, 1997; Goldberg et al., 2001;Oelkers and Schott, 2005; McGrail et al., 2006; Matteret al., 2007, 2009, 2011; Kelemen and Matter, 2008;Oelkers et al., 2008; Gislason et al., 2010; Wolff-Boenischet al., 2011). This carbonation process involves the releaseof Ca, Mg, and Fe from silicate minerals by dissolution.These divalent metals can then react with dissolved CO2

in the fluid phase to form stable carbonate minerals provid-ing the safe long-term storage of injected carbon. It hasbeen argued that silicate mineral dissolution is the slowestand thus rate-limiting step of this coupled process. Thisstudy demonstrates that all of the silicate substrates consid-ered in this study, including basaltic glass, olivine, augite,labradorite, and enstatite (1) can promote calcite precipita-tion on their surfaces and (2) have dissolution rates that arenot significantly slowed by the calcite precipitation. All of

these phases are components of natural basalts, such asthose currently being considered for CO2 storage viain situ mineral carbonation in Iceland (Oelkers et al.,2008; Alfredsson et al., 2008, 2011, 2013; Gislason et al.,2010; Aradottir et al., 2011, 2012a,b; Gudbrandssonet al., 2011; Gysi and Stefansson, 2011, 2012; Galeczkaet al., 2014) and in the Northwestern United States(McGrail et al., 2006; Schaef and McGrail, 2009; Schaefet al., 2009, 2010, 2011). Moreover previous work reportedthat carbonate precipitation does not affect basaltic glassnor diopside dissolution rates (Stockmann et al., 2011,2013). Taken together, the observations support the poten-tial use of basaltic and ultramafic rock carbonation as along-term carbon storage solution.

6. CONCLUSIONS

The results of this study indicate that the identity ofmineral or glass substrate influences calcite precipitationin mildly supersaturated fluids. In the presence of calciteseeds, calcite nucleation and growth starts immediately.


In the presence of silicate seeds, the fastest nucleation andmost extensive growth of calcite was found on enstatite,olivine, and peridotite, representing minerals belonging tothe orthorhombic crystal structure. Calcite was slowest tonucleate on basaltic glass, which has a non-ordered silicateframework. After nucleation occurs, however, the fluid sat-uration state with respect to calcite and calcite precipitationrates tend to a common value regardless of substrate. Fur-thermore, the presence of precipitated calcite on the sur-faces of all studied silicates does not inhibit theirdissolution. Taken together, these observations favour thestorage of carbon dioxide via the subsurface carbonationof mafic or ultramafic rocks.

ACKNOWLEDGEMENTS

This study is part of the CarbFix project (www.carbfix.com) inIceland, and we would like to thank all colleagues and co-workerswithin this project. In addition, we would like to thank several col-leagues for their help, in particular Carole Causserand, Alain Cas-tillo, and Phillippe de Parseval for technical assistance, and ThereseK. Flaathen, Allison Stephenson, Sigurdur Markusson, SvavaArnardottir, Sam Parry, Quentin Gautier, Giuseppe Saldi, ChrisPearce and Vasileios Mavromatis for lab assistance. We are mostgrateful for the SEM technical support provided by David Cornelland Johan Hogmalm at the Department of Earth Sciences, Univer-sity of Gothenburg in Sweden. Finally, we would sincerely like tothank Erik Sturkell for graphical assistance and continued support.CarbFix (Collaborative project-FP7-ENERGY-2011-1-283148)and the Environmental and Energy Fund of Reykjavık Energy,the Research Fund of the University of Iceland, the Nordic Councilof Ministers through NORDVULK, and the European Commu-nity through the MIN-GRO Research and Training Network(MRTN-CT-2006-035488) and the ERASMUS student mobilityprogram are gratefully acknowledged for their financial support.Constructive comments provided by the associate editor JiwcharGanor, reviewer Damien Daval and two anonymous reviewersgreatly improved the quality of the manuscript.

REFERENCES

Alfredsson H. A., Hadrarson B. S., Franzson H. and Gislason S.R. (2008) CO2 sequestration in basaltic rock at the Hellisheidisite in SW Iceland: stratigraphy and chemical composition ofthe rocks at the injection site. Mineral. Mag. 72, 1–5.

Alfredsson H. A., Wolff-Boenisch D. and Stefansson A. (2011) CO2

sequestration in basaltic rocks in Iceland: development of apiston-type downhole sampler for CO2 rich fluids and tracers.Energy Proc. 4, 3510–3517.

Alfredsson H. A., Oelkers E. H., Hadrarson B. S., Franzson H.,Gunlaugsson E. and Gislason S. R. (2013) The geology andwater chemistry of the Hellisheidi, SW-Iceland carbon storagesite. Int. J. Greenhouse Gas Control 12, 399–418.

Alkattan M., Oelkers E. H., Dandurand J.-L. and Schott J. (2002)An experimental study of calcite dissolution rates at acidicconditions and 25� C in the presence of NaPO3 and MgCl2.Chem. Geol. 190, 291–302.

Andreani M., Liquot L., Gouze P., Godard M., Hoise E. andGibert B. (2009) Experimental study of carbon sequestrationreactions controlled by the percolation of CO2-rich brinethrough peridotites. Eviron. Sci. Technol. 43, 1226–1231.

Aradottir E. S., Sonnenthal E., Bjornsson G. and Jonsson H.(2011) Multidimensional reactive transport modeling of CO2

mineral sequestration in basalts at the Hellisheidi geothermalfield, Iceland. Int. J. Greenhouse Gas Control 9, 24–40.

Aradottir E. S. P., Sonnenthal E. L. and Jonsson H. I. (2012a)Development and evaluation of a thermodynamic dataset forphases of interest in CO2 sequestration in basaltic rocks. Chem.

Geol. 304–305, 26–38.Aradottir E. S. P., Sonnenthal E. L., Bjornsson G. and Jonsson H.

(2012b) Multidimensional reactive transport modeling of CO2

mineral sequestration in basalts at the Hellisheidi geothermalfield, Iceland. Int. J. Greenhouse Gas Control 9, 24–40.

Bearat H., McKelvey M. J., Chizmeshya A. V., Gormley D.,Nunez R., Carpenter R. W., Squires K. and Wolf G. H. (2006)Carbon sequestration via aqueous olivine mineral carbonation:role of passivating layer formation. Eviron. Sci. Technol. 40,4802–4808.

Chen Y. and Brantley S. L. (1998) Diopside and anthophyllitedissolution at 25 and 90 �C and acid pH. Chem. Geol. 147, 233–248.

Cubillas P., Kohler S., Prieto M., Causserand C. and Oelkers E. H.(2005) How do mineral coatings affect dissolution rates? Anexperimental study of coupled CaCO3 dissolution – CdCO3

precipitation. Geochim. Cosmochim. Acta 69, 5459–5476.Daval D., Martinez I., Corvisier J., Findling N., Goffe B. and

Guyot F. (2009a) Carbonation of Ca-bearing silicates, the caseof wollastonite: experimental investigations and kinetic model-ing. Chem. Geol. 265, 63–78.

Daval D., Martinez I., Guigner J.-M., Hellmann R., Corvisier J.,Findling N., Dominici C., Goffe B. and Guyot F. (2009b)Mechanism of wollastonite carbonation deduced from micro-to nanometer length scale observations. Am. Min. 94, 1707–1726.

Daval D., Hellmann R., Corvisier J., Tisserand D., Martinez I. andGuyot F. (2010) Dissolution kinetics of diopside as a functionof solution saturation state: macroscopic measurements andimplications for modeling of geological storage of CO2.Geochim. Cosmochim. Acta 74, 2615–2633.

Davis K. J., Dove P. M. and De Yoreo J. J. (2000) The role ofMg2+ as an impurity in calcite growth. Science 290, 1134–1137.

de Leeuw N. H., Parker S. C., Catlow C. R. A. and Price G. D.(2000) Modelling the effect of water on the surface structure andstability of forsterite. Phys. Chem. Miner. 27, 332–341.

De Yoreo J. J. and Vekilov P. G. (2003) Principles of crystalnucleation and growth. Rev. Min. Geochem. 54, 57–93.

Diao Y., Myerson A. S., Hatton T. A. and Trout B. L. (2011)Surface design for controlled crystallization: the role of surfacechemistry and nanoscale pores in heterogeneous nucleation.Langmuir 27, 5324-2334.

Dixit S. and Caroll S. A. (2007) Effect of solution saturation state

and temperature on diopside dissolution. Trans, Geochem,10.1186/1467-4866-8-3.

Dove P. M. and Hochella, Jr., M. F. (1993) Calcite precipitationmechanisms and inhibition by orthophosphate: in situ obser-vations by scanning force microscopy. Geochim. Cosmochim.

Acta 57, 705–714.Fernandez-Martinez A., Hu Y., Lee B., Jun Y.-S. and Waychunas

G. A. (2012) In situ determination of interfacial energiesbetween heterogeneously nucleated CaCO3 and quartz sub-strates: thermodynamics of CO2 mineral trapping. Environ. Sci.

Technol. 47, 102–109.Flaathen T. K., Oelkers E. H., Gislason S. R. and Aagaard P.

(2011) The effect of dissolved sulphate on calcite precipitationkinetics and consequences for subsurface CO2 storage. Energy

Proc. 4, 5037–5043.Galeczka I., Wolff-Boenisch D., Oelkers E. H. and Gislason S. R.

(2014) An experimental study of basaltic glass–H2O–CO2

http://refhub.elsevier.com/S0016-7037(14)00179-3/h0005







































































































interaction at 22 and 50 �C: implications for subsurface storageof CO2. Geochim. Cosmochim. Acta 126, 123–145.

Giammar D. E., Bruant, Jr., R. G. and Peters C. A. (2005)Forsterite dissolution and magnesite precipitation at conditionsrelevant for deep saline aquifer storage and sequestration ofcarbon dioxide. Chem. Geol. 217, 257–276.

Gislason S. R. and Oelkers E. H. (2003) Mechanism, rates andconsequences of basaltic glass dissolution: II. An experimentalstudy of the dissolution rates of basaltic glass as a function ofpH and temperature. Geochim. Cosmochim. Acta 67, 3817–3832.

Gislason S. R., Wolff-Boenisch D., Stefansson A., Oelkers E. H.,Gunnlaugsson E., Sigurdardottir H., Sigfusson G., Brocker W.S., Matter J., Stute M., Axelsson G. and Fridriksson T. (2010)Mineral sequestration of carbon dioxide in basalt: a pre-injection overview of the CarbFix project. Int. J. Greenhouse

Gas Control 4, 537–545.Goldberg P., Chen Z.-Y., O’ Connor W., Walters R., and Ziock H.

(2001) CO2 mineral sequestration studies in U. S. Presented at

the First National Conference on Carbon Sequestration. Wash-ington DC, May 14–17, 2001.

Golubev S. V. and Pokrovsky O. S. (2006) Experimental study ofthe effect of organic ligands on diopside dissolution kinetics.Chem. Geol. 235, 377–389.

Golubev S. V., Pokrovsky O. S. and Schott J. (2005) Experimentaldetermination of the effect of dissolved CO2 on the dissolutionkinetics of Mg and Ca silicates at 25 �C. Chem. Geol. 217, 227–238.

Grandstaff D. E. (1977) Some kinetics of forsterite olivinedissolution. Trans. Am. Geophys. Union 58, 539.

Gudbrandsson S., Wolff-Boenisch D., Gislason S. R. and OelkersE. H. (2011) An experimental study of crystalline basaltdissolution from 2 6 pH 6 11 and temperatures from 5 to75 �C. Geochim. Cosmochim. Acta 75, 5496–5509.

Gysi A. P. and Stefansson A. (2011) CO2–water–basalt interaction.Numerical simulation of low temperature CO2 sequestrationinto basalts. Geochim. Cosmochim. Acta 75, 4728–4751.

Gysi A. P. and Stefansson A. (2012) CO2–water–basalt interaction.Low temperature experiments and implications for CO2

sequestration into basalts. Geochim. Cosmochim. Acta 81,129–152.

Halder S. and Walther J. V. (2011) Far from equilibrium enstatitedissolution rates in alkaline solutions at earth surface condi-tions. Geochim. Cosmochim. Acta 75, 7486–7493.

Hoch A. R., Reddy M. M. and Drever J. I. (1996) The effect of ironcontent and dissolved O2 on dissolution rates of clinopyroxene atpH 5.8 and 25 �C: preliminary results. Chem. Geol. 132, 151–156.

Hovelmann J., Austrheim H. and Jamtveit B. (2012) Microstruc-ture and porosity evolution during experimental carbonation ofa natural peridotite. Chem. Geol. 334, 254–265.

Hu Q., Nielsen M. H., Freeman C. L., Hamm L. M., Tao J., Lee J.R. I., Han T. Y. J., Becker U., Harding J. H., Dove P. M. andDe Yoreo J. J. (2012) The thermodynamics of calcite nucleationat organic interfaces: classical vs. non-classical pathways.Faraday Discuss. 159, 509–523.

Kelemen P. B. and Matter J. (2008) In situ carbonation ofperidotite for CO2 storage. Proc. Natl. Acad. Sci. U.S.A. 105,17295–17300.

Knauss K. G., Nguyen S. N. and Weed H. C. (1993) Diopsidedissolution kinetics as a function of pH, CO2, temperature, andtime. Geochim. Cosmochim. Acta 57, 285–294.

Koroleff F. (1976) Determination of silicon. In Methods of

Seawater Analysis (ed. K. Grasshoff). Springer Verlag, NewYork, pp. 149–158.

Kostenko O., Jamtveit B., Austrheim H., Pollok K. and Putnis C.(2002) The mechanism of fluid infiltration in peridotites atAlmklovdalen, western Norway. Geofluids 2, 203–215.

Kuo L. C. and Kirkpatrick R. J. (1985) Kinetics of crystaldissolution in the system diopside–forsterite–silica. Am. J. Sci.

285, 51–90.Lackner K. S., Wendt C. H., Butt D. P., Joyce E. L. and Sharp D.

H. (1995) Carbon dioxide disposal in carbonate minerals.Energy 20, 1153–1170.

Lackner K. S., Butt D. P. and Wendt C. H. (1997) Progress onbinding CO2 in mineral substrates. Energy Convers. Manage.

38, S259–S264.Lakshtanov L. Z., Bovet N. and Stipp S. L. S. (2011) Inhibition of

calcite growth by alginate. Geochim. Cosmochim. Acta 75,3945–3955.

Larsen K., Bechgaard K. and Stipp S. L. S. (2010) The effect of theCa2+ to CO3

2� activity ratio on spiral growth on the calcite 10�14surface. Geochim. Cosmochim. Acta 74, 2099–2109.

Lin Y.-P. and Singer P. C. (2005) Effects of seed material andsolution composition on calcite precipitation. Geochim. Cos-

mochim. Acta 69, 4495–4504.Lin Y.-P., Singer P. C. and Aiken G. R. (2005) Inhibition of calcite

precipitation by natural organic material: kinetics, mechanism,and thermodynamics. Environ. Sci. Technol. 39, 6420–6428.

Lioliou M. G., Paraskeva C. A., Koutsoukos P. G. and PayatakesA. C. (2007) Heterogeneous nucleation and growth of calciumcalcite on calcite and quartz. J. Colloid Interface Sci. 308, 421–428.

Lonsdale K. (1968) Epitaxy as a growth factor in urinary calculiand gallstones. Nature 217, 56–58.

Marini L. (2007) Geological Sequestration of Carbon Dioxide:

Thermodynamics, Kinetics, and Reaction Path Modeling. Else-vier, Amsterdam, p. 470.

Matter J. M., Takahashi T. and Goldberg D. S. (2007) Experi-mental evaluation of in situ CO2–water–rock reactions duringCO2 injection in basaltic rocks: implications for geological CO2

storage. Geochem. Geophys. Geosyst. 8(Q02001), 1–19.Matter J. M., Broecker W. S., Stute M., Gislason S. R., Oelkers E.

H., Stefansson A., Wolff-Boenisch D., Gunnlaugsson E.,Axelsson G. and Bjornsson G. (2009) Permanent carbondioxide storage into basalt: the CarbFix pilot project Iceland.Energy Proc. 1, 3641–3646.

Matter J. M., Broecker W. S., Gislason S. R., Gunnlaugsson E.,Oelkers E. H., Stute M., Sigurdardottir H., Stefansson A.,Alfredsson H. A., Aradottir E. S., Axelsson G., Sigfusson B.and Wolff-Boenisch D. (2011) The CarbFix pilot project –storing carbon dioxide in basalt. Energy Proc. 4, 5579–5585.

McAdam A. C., Zolotov M. Y., Sharp T. G. and Leshin L. A.(2008) Preferential low-pH dissolution of pyroxene in plagio-clase–pyroxene mixtures: implications for martian surfacematerials. Icarus 196, 90–96.

McGrail B. P., Schaef H. T., Ho A. M., Chien Y. J., Dooley J. J.and Davidson C. L. (2006) Potential for carbon dioxidesequestration in flood basalts. J. Geophys. Res.-Solid Earth 111, 13.

Meyer H. J. (1984) The influence of impurities on the growth rateof calcite. J. Cryst. Growth 66, 639–646.

Morales T. A. and Herbert R. (2002) Surface chemistry and acidicdissolution of forsterite. Geochim. Cosmochim. Acta 66, A521.

Nielsen J. W., Sand K. K., Pedersen C. S., Lakshtanov L. Z.,Winther J. R., Willemoes M. and Stipp S. L. S. (2012)Polysaccaride effects on calcite growth: The influence ofcomposition and branching. Cryst. Growth Des. 12, 4906–4910.

Oelkers E. H. (2001) An experimental study of forsterite dissolu-tion rates as a function of temperature and aqueous Mg and Siconcentrations. Chem. Geol. 175, 485–494.

Oelkers E. H. and Gislason S. R. (2001) The mechanism, rates andconsequences of basaltic glass dissolution: I. An experimentalstudy of the dissolution rates of basaltic glass as a function ofaqueous Al, Si and oxalic acid concentration at 25 �C andpH = 3 and 11. Geochim. Cosmochim. Acta 65, 3671–3681.






























































































































































Oelkers E. H. and Schott J. (2001) An experimental study ofenstatite dissolution rates as a function of pH, temperature, andaqueous Mg and Si concentration, and the mechanism ofpyroxene/pyroxenoid dissolution. Geochim. Cosmochim. Acta

65, 1219–1231.Oelkers E. H. and Schott J. (2005) Geochemical aspects of CO2

sequestration. Chem. Geol. 217, 183–186.Oelkers E. H., Gislason S. R. and Matter J. (2008) Mineral

carbonation of CO2. Elements 4, 333–337.Olsen A. A. and Rimstidt J. D. (2008) Oxalate-promoted forsterite

dissolution at low pH. Geochim. Cosmochim. Acta 72, 1758–1766.

Parkhurst D. L. and Appelo C. A. J. (1999), User’s guide toPHREEQC (version 2) – A computer program for speciation,batch-reaction, one-dimensional transport, and inverse geo-chemical calculations. Washington, DC: U.S. Geological Sur-vey Water-Resources Investigations, Report 99-4259, p. 312.

Plant L. J. and House W. A. (2002) Precipitation of calcite in thepresence of inorganic phosphate. Colloids Surf. A 203, 143–153.

Pokrovsky O. S. and Schott J. (2000a) Forsterite surface compo-sition in aqueous solutions: a combined potentiometric, elec-trokinetic, and spectroscopic approach. Geochim. Cosmochim.

Acta 64, 3299–3312.Pokrovsky O. S. and Schott J. (2000b) Kinetics and mechanism of

forsterite dissolution at 25 �C and pH from 1 to 12. Geochim.

Cosmochim. Acta 64, 3313–3325.Pokrovsky O. S., Golubev S. V. and Schott J. (2005) Dissolution

kinetics of calcite, dolomite and magnesite at 25 �C and 0 to 50atm pCO2. Chem. Geol. 217, 239–255.

Prigiobbe V., Costa G., Baciocchi R., Hanchen M. and MazzottiM. (2009) The effect of CO2 and salinity on olivine dissolutionkinetics at 120 �C. Chem. Eng. Sci. 64, 3510–3515.

Putnis A. (2009) Mineral replacement reactions. Rev. Min.

Geochem. 70, 87–124.Rodriguez-Blanco J. D., Shaw S. and Benning L. G. (2011) The

kinetics and mechanisms of amorphous calcium carbonate (ACC)crystallization to calcite, via vaterite. Nanoscale 3, 265–271.

Rodriguez-Blanco J. D., Shaw S., Bots P., Roncal-Herrero T. andBenning L. G. (2012) The role of pH and Mg on the stabilityand crystallization of amorphous calcium carbonate. J. Alloys

Comp. 536, S477–S479.Rosso J. J. and Rimstidt J. D. (2000) A high resolution study of

forsterite dissolution rates. Geochim. Cosmochim. Acta 64, 797–811.

Saldi G., Daval D., Morvan G. and Knauss K. G. (2013) The roleof Fe and redox conditions in olivine carbonation rates: Anexperimental study of the rate limiting reactions at 90 and150 �C in open and closed systems. Geochim. Cosmochim. Acta

118, 157–183.Schaef H. T. and McGrail B. P. (2009) Dissolution of Columbia

River basalt under mildly acidic conditions as a function oftemperature: experimental results relevant to the geologicalsequestration of carbon dioxide. Appl. Geochem. 24, 980–987.

Schaef H. T., McGrail B. P. and Owen A. T. (2009) Basalt–CO2–H2O interactions and variability in carbonate mineralizationrates. Energy Procedia 1, 4899–4906.

Schaef H. T., McGrail B. P. and Owen A. T. (2010) Carbonatemineralization of volcanic province basalts. Int. J. Greenhouse

Gas Control 4, 249–261.Schaef H. T., McGrail B. P. and Owen A. T. (2011) Basalt

reactivity variability with reservoir depth in supercritical CO2

and aqueous phases. Energy Proc. 4, 4977–4984.Schott J. and Berner R. A. (1983) X-ray photoelectron studies of

the mechanism of iron silicate dissolution during weathering.Geochim. Cosmochim. Acta 47, 2233–2240.

Schott J. and Berner R. A. (1985) Dissolution mechanisms ofpyroxenes and olivines during weathering. In The Chemistry of

weathering (ed. J. I. Drever). Reidel, Dordrecht, pp. 35–53.Schott J., Oelkers E. H., Benezeth P., Godderis Y. and Franc�ois L.

(2012) Can accurate kinetic laws be created to describe chemicalweathering?. C. R. Geosci. 344 568–585.

Seifritz W. (1990) CO2 disposal by means of silicates. Nature 345,486.

Shiraki R. and Brantley S. (1995) Kinetics of near-equilibriumcalcite precipitation at 100 �C: an evaluation of elementaryreaction-based and affinity-based rate laws. Geochim. Cosmo-

chim. Acta 59, 1457–1471.Siegel D. I. and Pfannkuch H. O. (1984) Silicate mineral dissolu-

tion at pH 4 and near standard temperature and pressure.Geochim. Cosmochim. Acta 48, 197–201.

Sjoberg L. (1989) Kinetics and stoichiometry of labradoritedissolution. In Proceedings of the 6th International Symposium

on Water-Rock Interaction (ed. D. L. Miles). Malvern, UK, pp.639–642.

Stockmann G. J., Wolff-Boenisch D., Gislason S. R. and Oelkers E.H. (2011) Do carbonate precipitates affect dissolution kinetics?1: basaltic glass. Chem. Geol. 284, 306–316.

Stockmann G. J., Shirokova L. S., Pokrovsky O. S., Benezeth P.,Bovet N., Gislason S. R. and Oelkers E. H. (2012) Does thepresence of heterotrophic bacterium Pseudomonas reactans

affect basaltic glass dissolution rates? Chem. Geol. 296–297, 1–18.

Stockmann G. J., Wolff-Boenisch D., Gislason S. R. and Oelkers E.H. (2013) Do carbonate precipitates affect dissolution kinetics?2: diopside. Chem. Geol. 337–338, 56–66.

Stumm W. and Morgan J. J. (1996) Aquatic Chemistry. Wiley-Interscience.

Suda S., Ichikawa S., Wada N. and Umegaki T. (2000) Morphol-ogy of calcium coating on amorphous silicate powder. J. Mater.

Sci. 35, 3023–3028.Sverdrup H. U. (1990) The Kinetics of Base Cation Release Due to

Chemical Weathering. Lund University Press, Lund, Sweden.Taylor A. S., Blum J. D. and Lasaga A. C. (2000) The dependence

of labradorite dissolution and Sr isotope release rates onsolution saturation state. Geochim. Cosmochim. Acta 64, 2389–2400.

Teng H. H. (2013) How ions and molecules organize to formcrystals. Elements 9, 189–194.

Turnbull D. and Vonnegut B. (1952) Nucleation catalysis. Ind.

Eng. Chem. 44, 1292–1298.Vavouraki A., Putnis C., Putnis A. and Koutsoukos P. G. (2008)

An atomic force microscopy study of the growth of calcite inthe presence of sodium sulfate. Chem. Geol. 253, 243–251.

Wakahama H., Mito S., Ohsumi T., Ueda A., Yajima T., Satoh H.,Sugiyama K., Ozawa A., Ajima S., Todaka N., Sato T., KatoM., Kaji Y., Tokumarui T., Kaieda H. and Kubota K. (2009) Aconcept of CO2 Georeactor sequestration at the Ogachi HDRsite, NE Japan. Energy Procedia 1, 3683–3689.

Weast R. C. and Astle M. J. (1986) CRC Handbook of Chemistry

and Physics, 63rd edition. CRC Press Inc, Boca Raton, Florida.Welch S. A. and Ullman W. J. (1993) The effect of organic acids on

plagioclase dissolution rates and stoichiometry. Geochim. Cos-

mochim. Acta 57, 2725–2736.Wolff-Boenisch D., Gislason S. R., Oelkers E. H. and Putnis C. V.

(2004a) The dissolution rates of natural glasses as a function oftheir composition at pH 4 and 10.6, and temperatures from 25to 74 �C. Geochim. Cosmochim. Acta 68, 4843–4858.

Wolff-Boenisch D., Gislason S. R. and Oelkers E. H. (2004b) Theeffect of fluoride on the dissolution rates of natural glasses atpH 4 and 25 �C. Geochim. Cosmochim. Acta 68, 4571–4582.













































































































































Wolff-Boenisch D., Gislason S. R. and Oelkers E. H. (2006) Theeffect of crystallinity on dissolution rates and CO2 consumptioncapacity of silicates. Geochim. Cosmochim. Acta 70, 858–870.

Wolff-Boenisch D., Wenau S., Gislason S. R. and Oelkers E. H.(2011) Dissolution of basalts and peridotite in seawater, in thepresence of ligands, and CO2: implications for mineral seques-tration of carbon dioxide. Geochim. Cosmochim. Acta 75, 5510–5525.

Xu T. F., Apps J. A. and Pruess K. (2005) Mineral sequestration ofcarbon dioxide in a sandstone–shale system. Chem. Geol. 217,295–318.

Zhang Y. and Dawe R. A. (2000) Influence of Mg2+ on the kineticsof calcite precipitation and calcite crystal morphology. Chem.

Geol. 163, 129–138.Zhong S. and Mucci A. (1989) Calcite and aragonite precipitation

from seawater solutions of various salinities: precipitation rateand overgrowth compositions. Chem. Geol. 78, 283–299.

Zhong S. and Mucci A. (1993) Calcite precipitation in seawaterusing a constant addition technique: a new overall reactionkinetic expression. Geochim. Cosmochim. Acta 57, 1409–1417.

Associate editor: Jiwchar Ganor
























the role of silicate surfaces on calcite precipitation ... et al. gca the rolæe of silica... ·...

Documents