Carbon Dioxide Binding at Dry FeOOH Mineral Surfaces: Evidence for

Article pubs.acs.org/est

Carbon Dioxide Binding at Dry FeOOH Mineral Surfaces: Evidence for Structure-Controlled Speciation Xiaowei Song* and Jean-François Boily Department of Chemistry, Umeå University, SE-901 87 Umeå, Sweden S Supporting Information *

ABSTRACT: Interactions between CO2(g) and mineral surfaces are important to atmospheric and terrestrial settings. This study provides detailed evidence on how differences in mineral surface structure impact carbonate speciation resulting from CO2(g) adsorption reactions. It was achieved by resolving the identity of adsorption sites and geometries of (bi)carbonate species at surfaces of nanosized goethite (α-FeOOH) and lepidocrocite (γ-FeOOH) particles. Fourier transform infrared spectroscopy was used to obtain this information on particles contacted with atmospheres of CO2(g). Vibrational modes of surface hydroxo groups covering these particles were first monitored. These showed that only one type of the surface groups that are singly coordinated to Fe atoms (−OH) are involved in the formation of (bi)carbonate species. Those of higher coordination numbers (μ−OH, μ3−OH) do not participate. Adsorption geometries were then resolved by investigating the C−O stretching region, assisted by density functional theoretical calculations. These efforts provided indications leaning toward a predominance of monodentate mononuclear species, −O−CO2Hx=[0,1]. In contrast, monodentate binuclear species of (−O)2-COHx=[0,1], are expected to form at particle terminations and surface defects. Finally, calculations suggested that bicarbonate is the dominant species on goethite, while a mixture of bicarbonate and carbonate species is present on lepidocrocite, a result stemming from different hydrogen bonding patterns at these mineral surfaces.

■

INTRODUCTION The impact of current and future atmospheric carbon dioxide levels on climate change, nature, and society has been debated extensively in the scientific literature,1−8 as well as in press. While substantial progress has been made in offering plausible scenarios on Earth’s future climate, the impact of increased levels of atmospheric carbon dioxide, one that has experienced a rise from 280 ppm in 18002 to above 400 ppm in 2013,9 on carbon speciation and reactivity also needs to be specifically addressed. This aspect is important for mineral (nano)particles in Earth’s atmosphere and in oceans as well as in terrestrial and freshwater environments. Mineral surfaces in these settings are sinks and transformation centers for CO2(g) toward various forms of carbonate species and solids,10−14 and even for photochemically or microbially reduced forms of carbon.15−18 Such interactions are moreover highly relevant to technological applications where accelerated geochemical weathering,19 oceanic/geological storage,20 and solidification11,13,14 are considered as CO2(g) mitigation approaches. Interactions with iron oxide minerals are of particular relevance in many of these contexts given the widespread occurrence and reactivity of these compounds in nature. These have, for instance, been focal points of research to understand processes taking place at surfaces of hematite (α-Fe2O3), goethite (α-FeOOH), and ferrihydrite (Fe8.2O8.5(OH)7.4· 3H2O)21 particles exposed to aqueous or gaseous environments.22−30 Of particular relevance is the ability of CO2(g) at © 2013 American Chemical Society

converting to stable carbonate and bicarbonate complexes at the surfaces of these minerals. The structures and coordination environments of these complexes in the gas and aqueous phases have been chiefly assessed by vibrational spectroscopy given the high sensitivity of this technique to changes in bonding environments of (bi)carbonate species, and given direct possibilities for comparing results with theoretical simulations.23−25,29,30 Despite numerous efforts made along such fronts, much remains to be learned on how mineral surface structure affects (bi)carbonate formation and speciation, as well as how mineral surface affect the stability of these important species. In this study we address this issue by resolving interactions between CO2(g) and three different synthetic iron oxyhydroxide (FeOOH) particles with distinct and well-defined surfaces (Figure 1). A molecular-level understanding of adsorption reactions at these surfaces was developed starting from an a priori knowledge of the crystallographically expected dispositions and densities of reaction sites on these different particle surfaces (Figure 1).31−33 FeOOH mineral surfaces are, under ambient atmospheric conditions, terminated by hydroxo groups that can be singly- (≡FeOH, −OH), doubly- (≡Fe2OH, Received: Revised: Accepted: Published: 9241

May 8, 2013 July 25, 2013 July 25, 2013 July 25, 2013 dx.doi.org/10.1021/es4020597 | Environ. Sci. Technol. 2013, 47, 9241−9248

Environmental Science & Technology

Article

Figure 1. Band assignment of distinct surface hydroxyls (SI Table S1) on LL, RL, and G. FTIR spectra (left column), idealized particle morphologies (middle column) are shown for all particles. Crystal structures of the dominant surfaces (right two columns) are shown as side and top views (Fe = brown, O = red, H = black, hydrogen bonds = black dash lines).

procedures and experiments were carried out at 25 °C and at a total pressure of 760.4 Torr. For each loading, CO2(g) was equilibrated with the sample for twenty minutes during which time CO2(g) levels in the end flow were monitored by a nondispersive infrared analyzer (LI-7000, Licor Inc.). FTIR spectra were collected in situ with a Bruker Vertex 70/ V FTIR spectrometer, equipped with a DLaTGS detector. All spectra were collected in the 4500−1200 cm−1 range at a resolution of 4.0 cm−1 and at a forward/reverse scanning rate of 10 Hz. Frequencies below 1200 cm−1 could not be measured due to the overwhelmingly strong absorbance of the CaF2(s) windows in the reaction cell. Background spectra were obtained in vacuo for the same tungsten mesh used for the mineralCO2(g) experiments. Each spectrum was an average of 50 scans. A sequence of numerical treatment procedures was applied to the raw spectra to obtain spectral features revealing molecular details of CO2(g) interactions with FeOOH surfaces. These procedures, which are similar to those carried out in a previous study32 on water vapor adsorption, are detailed further in the SI. Ferric Oxyhydroxide Theoretical Cluster Calculations. Density functional theory (DFT) was used to calculate theoretical vibrational frequencies for various configurations of carbonate species bound to a dimeric ferric oxyhydroxide cluster. The cluster consisted of edge-sharing iron octahedra complexed by hydroxo, water and carbonate. The latter either coordinated to one (monodentate mononuclear; MM) or two (monodentate binuclear; MB) iron atoms. Valence shell electrons of the iron atoms were completely unpaired (M = 2S + 1 = 11) during these calculations. These clusters were geometry-optimized at the B3LYP/6-31G(d) level of theory35 using Cartesian d and f basis functions, and a pruned (99,590) integration grid. Various starting configurations of hydroxo and water molecules were tested. Those associated with substantial departures from iron octahedral geometry were discarded. Vibration frequencies were calculated from the optimized geometries of selected clusters at the same level of theory and integration grids. These values were scaled by a factor of 0.96, which has previously been shown to improve the prediction of

μ−OH), or triply coordinated (≡Fe3OH, μ3−OH).34 In the gas phase, many of these groups can be experimentally resolved as narrow, nearly discrete (e.g., 10−25 cm−1 peak full width at half-maximum), O−H stretching vibration bands using Fourier transform infrared (FTIR) spectroscopy (Figure 1). A detailed account of the band assignments can be found in the earlier publication from our group (Supporting Information (SI) Table S1).31 As O−H stretching modes are strongly responsive to any change in bonding environment, and namely in the order of 150−175 cm−1 per pm change in O−H bond length,34 they can be used to identify hydroxyl groups interacting with CO2(g). Binding modes of the resulting (bi)carbonate species can, on the other hand, be resolved by inspecting the C−O stretching region. In this work, we show how even seemingly minor differences in surface structure can induce important differences on the speciation of (bi)carbonate species at FeOOH particle surfaces.

■

EXPERIMENTAL SECTION Mineral Synthesis and Characterization. All FeOOH particles were synthesized in a N2(g)-filled glovebox to minimize CO2(g) contamination. Product solids were dialyzed with preboiled and N2 (g)-degassed deionized water (18.2 MΩ· cm) for the same purpose. Detailed synthesis and characterization procedures are presented in the SI. Salient physical properties of the resulting particles are also presented in the SI (Figure S1−S3, Table S2). CO2 Adsorption and FTIR Spectroscopy. After centrifugation of the mineral suspensions, the wet pastes were transferred on to a fine tungsten mesh (Unique wire weaving, 0.002 in. mesh diameter) squeezed into a copper sample holder and dried to a thin film under dry N2 (g) flow overnight. CO2(g) adsorption experiments were then carried out on the dry mineral films in an airtight/vacuum cell (AABSPEC #2000A) equipped with CaF2(s) windows. Atmospheres of 0−16 000 ppm CO2(g) were generated by mixing CO2(g) with a carrier N2(g) flow at predetermined ratios controlled by mass flow controllers (MKS, 179A), maintaining the total flow rate at 400 SCCM (standard cubic centimeter per minute). All gas mixing 9242

dx.doi.org/10.1021/es4020597 | Environ. Sci. Technol. 2013, 47, 9241−9248


Article

findings concur with previous ones supporting the stronger reactivity of −OH with gases given its stronger propensity for donating and accepting hydrogen bonds as well as for ligand exchange, compared to functional groups of greater iron nuclearity.31−33 This was also confirmed for the case of goethite, as will now be detailed. Much like in the (001) plane of lepidocrocite, the predominant (110) plane of goethite (G) exhibits rows of −OH, μ−OH, μ3-O(H) sites.31,34 An important distinction however lies in the hydrogen bonding patterns adopted by −OH groups. Molecular dynamics simulations31 showed that every eighth −OH group on the (001) plane of lepidocrocite donates a hydrogen bond to a vicinal group, while every second −OH group is a H-bond donor on the (110) plane of goethite (Figure 1). The greater hydrogen bonding population between −OH groups of goethite is caused by the weakening of the O− H bond strength by the donating μ3,I−OH group (Figure 1). This can be effectively seen through the lower O−H stretching frequency in −OH in G (3661 cm−1), compared to that in RL (3667 cm−1), and through strong correlations with variations in band intensities of μ3,I−OH (3491 cm−1) when surfaces are reacted to various foreign species. In this case, exposing G surfaces to CO2(g) redshifts the 3661 cm−1 band of −OH 8 cm −1 while the 3491 cm −1 band of μ 3,I −OH group concomitantly responds through a strengthening of its donating hydrogen bond, as manifested by a redshift to 3425 cm−1 (Figure 2). The spectra also reveal an additional peak at 3678 cm−1, suggesting the appearance of a purely isolated (not hydrogen bonded) −OH (Figure 2) resulting from the partial breakage of the hydrogen bonding network due to the coordination of (bi)carbonate species at −OH sites. A similar but even weaker band in RL at 3676 cm−1 denotes the same type of isolated −OH groups. These features suggest, in the same token, that the formation of (bi)carbonate does not consume all −OH groups at these surfaces. Although only the major crystallographic planes were discussed in this section, we must emphasize that the terminal (100) plane of lepidocrocite and (021) plane of goethite particlesas well as other similar planes or surface imperfectionscannot be negligible in these adsorption reactions. These surfaces are generally rich in a diversity of hydrogen bonding patterns, which typically attenuate the narrow intensities of the O−H stretching region. The (100) plane of lepidocrocite, for instance, exhibits bare Fe sites, which are strong ligand complexation sites (e.g., Fe-(OH2)2 in water). The (021) plane of goethite also exposes binding sites at edges of iron octahedra. As these sites represent ∼5−15% of the total activity −OH site on these particle surfaces,31,33,38 their contributions in (bi)carbonate will also be tentatively resolved in this work by investigating the C−O stretching region of (bi)carbonate surface species. Carbonate Speciation at FeOOH Surfaces. The full set of spectra of the C−O stretching region collected for this work is reported in SI Figure S5. These data will be used to identify the species formed at lepidocrocite and goethite surfaces based on the notion that only −OH groups are involved. It should however be stressed that the elucidation of coordination geometries from the splitting and frequency of symmetric (νs) and asymmetric (νas) stretching modes is not a necessarily straightforward task, as it will be further emphasized in this section. Band assignments will be guided by theoretical frequency calculations of carbonate-bound iron oxyhydroxide

experimental frequencies.25,36 All calculations were carried with Gaussian 09.37

■

RESULTS AND DISCUSSION In this work, the identity of surface hydroxo groups responsible for binding (bi)carbonate species is revealed by the O−H stretching region (3350−3750 cm−1), as will be discussed in the following section. Binding modes of the resulting species will be discussed in the latter part by comparing the C−O stretching region (1200−1750 cm−1) of the FTIR spectra with values obtained by DFT. Surface Site Activity for CO2(g). Lath (LL) and rod (RL) lepidocrocite particles have the same crystal structure but expose surface species in different proportions (Figure 1). LL is thus an ideal material to resolve interactions with twodimensional arrays of μ−OH groups of the (010) plane, while RL should shed insight into interactions with more structurally complex (001) surface containing adjacent rows of −OH, μ−OH and μ3−OH sites (Figure 1, SI Table S1). Interactions with the (010) plane of lepidocrocite will thereby be specifically probed by monitoring changes in the 3626 cm−1 band of μ−OH. Interactions with the (001) plane will be monitored through the 3667 cm−1 (−OH) and the 3551 cm−1 (μ3−OH) bands. Those of μ−OH cannot be specifically resolved on this plane due to a strong overlap with the 3626 cm−1 band of its counterpart species of the (010) plane. Exposing dry LL and RL particles to atmospheres of up to 16000 ppm CO2(g) induced substantial and systematic changes in the positions and intensities of limited ranges of the O−H stretching frequencies. The entire sequence of spectra can be found in SI Figure S4, whereas Figure 2 shows spectra only at

Figure 2. Representative CO2(g)-subtracted FTIR spectra of LL, RL and G particles in surface O−H stretching region. The entire sequence of spectra in the 0−16 000 ppm region is reported in the SI. The spectra of the original FeOOH surface at 0 ppm CO2(g) (blue) and at the 16 000 ppm CO2(g) (green) are shown for LL and RL. A representative spectrum at an intermediate CO2(g) loading (red) is also included for G. Arrows (↑↓) denote intensity changes while black rectangles (▼) denote stable bands.

representative pressures. Bands of the μ−OH (3626 cm−1) were unaffected by the presence of CO2(g), and therefore provided evidence for the negligible reactivity of these groups. The same can be said for the μ3−OH groups (3551/3532 cm−1) of the (001)/(100) plane. In strong contrast, the 3667 cm−1 (−OH) band underwent a 5 cm−1 redshift with a concomitant loss in intensity with CO2(g) loading. These 9243



Article

clusters, as well as comparison with well-known materials when possible. Previous theoretical frequency studies of (bi)carbonate complexes on ferric oxyhydroxide clusters23,24,28,29 and of their surrogates25,27 have already underscored the great sensitivity of C−O stretching frequencies to species nuclearity, denticity (i.e., mono- vs bidentate complexation), protonation state (i.e., on carbonyl and on surface (hydr)oxo groups), and hydrogen bonding environments. A number of studies have, moreover, largely favored monodentate binuclear (MB) coordination for these species on hematite (α-Fe2O3) surfaces of undetermined surface structure.23,25 The predominance of such complexes on iron oxide surface was also justified from bond valence arguments.39 Some of the literature’s theoretical C−O stretching frequencies23−25 appear to correspond to our experimental values, and reported in SI Figure S5 and Table S3. To explain all major peaks, we however need to consider results from hydrated clusters (e.g., Hausner et al.,24) although the presence of water is highly unlikely in our systems. This result in itself already illustrates the profound effects that hydrogen bonding can play on absolute C−O stretching frequencies. Similar effects can be seen in, for instance, how hydration of metal carbonates (e.g., MgCO3·xH2O, CaCO3·xH2O)40 or how replacing Na+ by C(NH2)3+ in the Na6[Ce(CO3)5]·12H2O and Na6[Th(CO3)5]·12H2O solids41,42 increase the degeneracy of C−O stretching (νas and νs) frequencies. Another notable difficulty involves the consideration of MB complexes as viable candidates to explain our experimental data. While theoretical frequencies generated from previous studies23−25 would appear to support such a concept, they are challenged from a sterical point of view. Inspection of the spacing along rows of −OH on the (001) plane of lepidocrocite and the (110) plane of goethite (0.30 nm) in fact tends to mismatch plausible O−O distances adopted in (bi)carbonate.43−46 A MB complex would therefore require a substantial degree of surface relaxation, one involving the tilting of two adjacent iron-octahedra toward one another. This structure can be found in many carbonate-bearing minerals (e.g., Dawsonite, alkaline earth metal carbonates) where the carbonate ion is of relatively imperturbable trigonal planar geometry and with a limited range of O−O distances.47 A highly comparable example to FeOOH lies in the Dawsonite (e.g., KAl(CO3)(OH)2) bulk where the ∼82° angle formed between two adjacent octahedra (∠ O−Al−Al) accommodates the relatively inflexible ∼0.23 nm wide spacing of the C−O2 moiety of carbonate. Furthermore, this geometry results in zigzagged chains of edge sharing Al (Al−Al spacing of 0.28 nm), shown in Figure 3. Those geometries would require substantial surface relaxation at FeOOH surface that, in our experience in simulating these materials by molecular dynamics,31,46 appears to be unjustified. Interestingly, the very DFT frequency calculations that were used to support the MB configuration on hematite involved distorted binuclear clusters, somewhat akin to the Dawsonite configuration.23,25,26,39 Recognizing this shortcoming, Chernyshova et al.28 recently suggested that neighboring μ3−OH groups could be involved in MB complexation given the shorter O−O spacing of such a configuration. However, as shown in the previous section, μ3− OH groups are unreactive in the formation of (bi)carbonate (Figure 2). It is also notable that previous studies of carbonate adsorption in aquatic solutions suggested monodentate mononuclear carbonate as important inner-sphere species.48

Figure 3. Crystal structure of Dawsonite (top) and representative structure of iron oxyhydroxide surfaces (bottom). O−O distances are labeled in nm. (Al-octahedra = blue; Fe-octahedra = brown; O = red; C = black.).

In an effort to support our band assignment further, we performed DFT calculations on dimeric ferric octahedral clusters that are very comparable to those of Hausner et al.24 and Bargar et al.23 These calculations consider monodentate binuclear (MB; −O−CO2Hx=[0,1]) as well as monodentate mononuclear (MM; (−O)2-COHx=[0,1]) (bi)carbonate complexes. Figure 4, Table 1 and SI Table S3 in fact show that our

Figure 4. Cluster geometries and calculated frequencies of (bi)carbonate species (cf. Table 1 for details). O−O distances are labeled in nm. Possible positions for proton and hydrogen bonds are denoted in green. (Fe = blue; O = red; C = black; H = pink.) In the right column, theoretical frequency predictions generated at the B3LYP/631G(d) (solid points) level of theory are plotted alongside literature values from Bargar et al.,23 Hausner et al.,24 and Baltrusatis et al.25 (hollow diamond points). Star symbols denote the δCOH modes of bicarbonate species.

calculations are generally in line with those of corresponding clusters from the literature.23−25 Our new calculations however also consider effects of protonation sites and different hydrogen bonding environments. These, for instance, show that by protonating the monodentate binuclear carbonate (MBC) complex at the carbonate oxygen site, the resulting MBB1 species possesses the highest symmetry of all those considered in this work, and thus the greatest degeneracy in the C−O 9244



Article

Table 1. Summary of Cluster Frequency Calculations (B3LYP/6-31G(d))a calculated frequencies clusters monodentate binuclear carbonate monodentate binuclear bicarbonate monodentate mononuclear carbonate monodentate mononuclear bicarbonate

a

νas

νs

Δνα

1708 1590 1846 1452 1665 1589 1575 1701

1162 1431 1225 1271 1139 1204 1372 1325

546 159 621 181 526 385 203 376

δCOH 1193 1151

1038 1177 1173

H-bonding from −OHβ

acronymγ

to Fe-O-C isolated isolated to O−C-O to Fe-O-C to O−C-OH/isolated OH to Fe-O-C/isolated OH to Fe-O-C & O−C-OH

MBC MBB1 MBB2 MMC1 MMC2 MMB-a MMB-b0 MMB-b1

Δν = νas - νs. β. The hydrogen accepting O sites are shown in bold letters. γ. These geometries are shown in Fig. 4.

stretching (i.e., small Δν = νas-νs) region. Protonating the Febound oxo group (Fe−O-C) of MBC induces, in contrast, much less degeneracy (i.e., greater Δν), an effect also reported in Baltrusaitis et al.25 Similar phenomena can be observed by altering H-bonding sites. The monodentate mononuclear carbonate (MMC1) complex, for example, generates less band splitting because carbonate is stabilized by hydrogen bonding. Moving this hydrogen bond to the adjacent Fe-bound oxo group (Fe−O-C) decreases the symmetry of the complex (MMC2) and thereby splits the C−O stretching frequencies to Δν = νas-νs = 526 cm−1 (Figure 4, Table 1). It is notable that the calculated frequencies for MMB clusters in the literature have larger deviations than MBC and MBB clusters, whereas our three MMB clusters fall in line with the reported range that νas ranges from 1779 to 1627 cm−1 and νs ranges from 1390 to 1182 cm−1 (Δν = 534−237 cm−1, SI Table S3).23−25 Our band assignment procedures will thus be guided by inspection of Δν values, as well as absolute C−O stretching (νas and νs) and COH bending (δCOH) frequencies. We first note that Δν values beyond 500 cm−1 are not present in the experimental spectra of all three minerals considered in this work. MBC, MBB2, and MMC2 complexes should thereby be discarded as viable species. The Δν values of the remaining possible five species considered for this work are thereby presented with the corresponding experimental values in Figure 5. MMB-a and MMB-b1 produce Δν values exceeding 300 cm−1, while those of MBB1, MMC1, and MMB-b0 species are more degenerate with Δν values in the 159−203 cm−1 range. MMB-a (1589/1204 cm−1) and MMB-b1 (1701/1325 cm−1) can thereby effectively explain the FTIR bands at 1570− 1580, 1221 cm−1 and 1680, 1327−1331 cm−1 of both LL and RL. The more degenerate band set at 1614, 1418−1420 and 1221 cm−1 of LL and RL is, on the other hand, chiefly assigned to MMB-b0 as MBB cannot form along rows of −OH groups. This being said, we cannot entirely discard the possibility that bands at 1570−1580 and 1418 cm−1 (Δν = 152−162 cm−1) correspond to minor MBB species (Δν = 162−196 and 150− 194 cm−1) formed at particle terminations or at surface defects. Finally, we note that MMC1 produces νas and νs values at 1452 and 1271 cm−1 (Δν= 181 cm−1) that could be compared to the 1500 and 1327 cm−1 (Δν= 173 cm−1) bands of LL and RL, given their highly similar values. As for goethite, the C−O stretching modes at 1530 and 1263 cm−1 are generally lower than their corresponding values in lepidocrocite (1580/1570 and 1331/1327 cm−1). Although the clusters used for this work could not specifically test for differences in hydrogen bonding patterns between lepidocrocite and goethite, namely those involving μ3−OH, we suspect that these differences in C−O stretching modes are a result of this

Figure 5. Difference spectra for the C−O stretching (νas and νs) and COH bending (δCOH) regions (left column) of LL (blue), RL (green) and G (red) exposed to atmospheres containing up to 16 000 ppm CO2(g). The dash and solid line spectra represent the lowest and highest CO2(g) levels considered in this work. Values of Δν = νas − νs of the dominant (color-coded) experimental bands are compared to theoretical values obtained at the B3LYP/6-31G(d) level of theory in the diagram on the right. The abbreviations for the clusters and their calculated frequencies are detailed in Table 1 and the band assignments summarized in Table 2.

effect. The 1530 and 1263 cm−1 (Δν= 267 cm−1) bands of goethite are therefore assigned to MMB (203−385 cm−1). The minor bands at 1618 and 1416 cm−1 could, on the other hand, be assigned to MBB1, as was done in lepidocrocite. Another minor band in the 1200 cm−1 region could also be explained by stretching modes of hydrogen bonded CO2(g), as previously noted.24 Finally, it is interesting to note that our band assignments (Figure 5) suggest a predominance of bicarbonate species on goethite, but a coexistence of bicarbonate and carbonate on lepidocrocite. When considering MM complexes of the dominant planes of these minerals, this finding could suggest that the ability of −OH groups in forming hydrogen bonds with one another should dictate the protonation states of carbonate complexes at mineral surfaces. The relative instability of carbonate at the goethite surface could, for instance, arise from the resilience of mutually hydrogen bonded −OH groups in stabilizing carbonate oxo groups, which lie outside the surface plane (Figure 6). Once more, we stress that the hydrogen bond strength between −OH groups (−OH···−OH) of G is driven by interactions with μ3−OH. The weaker hydrogen bonded −OH groups of lepidocrocite would, in contrast, facilitate the stabilization of carbonate, and even provide a mechanistic pathway toward bicarbonate formation 9245



Article

can interact with CO2(g) and that their hydrogen bonding patterns with vicinal groups can dictate chemical speciation. Such mechanisms are expected to play particularly important roles in atmospheric and terrestrial systems where CO2(g) is interacted with mineral surfaces. The same forms of interactions are expected to occur in capture and storage technologies, and heterogeneous catalysis mechanisms used to synthesize products49,50 including methanol51 or dimethyl carbonate.52 Further studies along such lines should thereby involve surfaces of varied yet controlled structures to test the validity of the ideas conveyed in this work to a broader range of minerals of environmental and technological importance. A forthcoming article from our group will in fact show how surface structural differences in these FeOOH minerals affect carbonate speciation when exposed to water vapor.

Figure 6. Schematic representation of CO2(g) adsorption mechanisms and (bi)carbonate formation at lepidocrocite and goethite surfaces.

■

by acid- and/or thermally induced proton hopping from neighboring −OH and/or −OH2 species. A schematic representation of suggested reactions leading to these complexes on both minerals considered in this work is presented in Figure 6.

* Supporting Information (1) Mineral synthesis and characterization. (2) Spectral analysis. (3) Spectra in full CO2 range from 0 to 16 000 ppm. This material is available free of charge via the Internet at http://pubs.acs.org.

■

■

SIGNIFICANCE AND IMPLICATIONS This work furthered our pursuit for building a fundamental understanding of the interactions at FeOOH mineral surfaces,

*Phone: + 46 90 786 5361; e-mail: [email protected]. se. Notes

The authors declare no competing financial interest.

experimental data νs

Δνα

MBB1 MMC1 MMB-b0 MMB-b1 MMB-a

1580 1500 1614 1680 1580

1418 1327 1418 1327 1221

162 173 196 353 359

MBB1 MMC1 MMB-b0 MMB-b1 MMB-a

1570 1500 1614 1680 1570

1420 1331 1420 1331 1221

150 169 194 349 349

MBB1 MMB-b0

1618 1530

1416 1263

202 267

LL

RL

G a

νas

species

AUTHOR INFORMATION

Corresponding Author

Table 2. FTIR Band Assignments of (Bi)Carbonate Groups at FeOOH Surfaces (Figure 5)a FeOOH

ASSOCIATED CONTENT

S

■

δCOH

ACKNOWLEDGMENTS This work was supported by the Swedish Research Council (2009-3110; 2012-2976), as well as the Knut and Alice Wallenberg, the J.C. Kempe, and the Carl Tryggers Foundations.

1221

■

REFERENCES

(1) Doney, S. C.; Fabry, V. J.; Feely, R. A.; Kleypas, J. A. Ocean acidification: The other CO2 problem. Annu. Rev. Mar. Sci., 2009; Vol. 1, pp 169−192. (2) Etheridge, D. M.; Steele, L. P.; Langenfelds, R. L.; Francey, R. J.; Barnola, J. M.; Morgan, V. I. Natural and anthropogenic changes in atmospheric CO2 over the last 1000 years from air in Antarctic ice and firn. J. Geophys. Res.: Atmos. 1996, 101 (D2), 4115−4128. (3) Hansen, J.; Sato, M.; Ruedy, R.; Lo, K.; Lea, D. W.; MedinaElizade, M. Global temperature change. Proc. Natl. Acad. Sci. U.S.A. 2006, 103 (39), 14288−14293. (4) Keeling, C. D.; Chin, J. F. S.; Whorf, T. P. Increased activity of northern vegetation inferred from atmospheric CO2 measurements. Nature 1996, 382 (6587), 146−149. (5) Raupach, M. R.; Marland, G.; Ciais, P.; Le Quere, C.; Canadell, J. G.; Klepper, G.; Field, C. B. Global and regional drivers of accelerating CO2 emissions. Proc. Natl. Acad. Sci. U.S.A. 2007, 104 (24), 10288− 10293. (6) Thomas, C. D.; Cameron, A.; Green, R. E.; Bakkenes, M.; Beaumont, L. J.; Collingham, Y. C.; Erasmus, B. F. N.; de Siqueira, M. F.; Grainger, A.; Hannah, L.; Hughes, L.; Huntley, B.; van Jaarsveld, A. S.; Midgley, G. F.; Miles, L.; Ortega-Huerta, M. A.; Peterson, A. T.; Phillips, O. L.; Williams, S. E. Extinction risk from climate change. Nature 2004, 427 (6970), 145−148. (7) Cox, P. M.; Betts, R. A.; Jones, C. D.; Spall, S. A.; Totterdell, I. J. Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model. Nature 2000, 408 (6809), 184−187.

1221

Δν = νas − νs.

here by focusing on interactions with CO2(g). As a great majority of studies dedicated to mineral-CO2(g) interactions were predominantly focused on spectroscopic attributes of (bi)carbonate species 22−27,48 less knowledge has been developed on the identity of reactive surface (hydr)oxo groups involved in the transformation of CO2(g) into (bi)carbonate. This work thereby attempted to alleviate this dearth of knowledge by resolving changes in the O−H stretching region to identify hydroxyl functional groups involved in the adsorption reactions. Our efforts supported the concept for hydrogen bonding as a key factor controlling the fate of CO2(g) once interacted with FeOOH surfaces, one that can certainly be extended to other hydroxo-terminated mineral surfaces. These results also showed that only singly coordinated −OH groups 9246



Article

(8) Zak, D. R.; Pregitzer, K. S.; King, J. S.; Holmes, W. E. Elevated atmospheric CO2, fine roots and the response of soil microorganisms: A review and hypothesis. New Phytol. 2000, 147 (1), 201−222. (9) Jones, N. Troubling milestone for CO2. Nat. Geosci. 2013, 589. (10) Villalobos, M.; Trotz, M. A.; Leckie, J. O. Surface complexation modeling of carbonate effects on the adsorption of Cr(VI), Pb(II), and U(VI) on goethite. Environ. Sci. Technol. 2001, 35 (19), 3849−3856. (11) Bachu, S.; Gunter, W. D.; Perkins, E. H. Aquifer disposal of CO2Hydrodynamic and mineral trapping. Energy Convers. Manage 1994, 35 (4), 269−279. (12) Hellevang, H.; Aagaard, P.; Oelkers, E. H.; Kvamme, B. Can Dawsonite permanently trap CO2? Environ. Sci. Technol. 2005, 39 (21), 8281−8287. (13) Lackner, K. S.; Wendt, C. H.; Butt, D. P.; Joyce, E. L.; Sharp, D. H. Carbon-dioxide disposal in carbonate minerals. Energy 1995, 20 (11), 1153−1170. (14) Oelkers, E. H.; Gislason, S. R.; Matter, J. Mineral carbonation of CO2. Elements 2008, 4 (5), 333−337. (15) Dhillion, S. S.; Roy, J.; Abrams, M. Assessing the impact of elevated CO2 on soil microbial activity in a Mediterranean model ecosystem. Plant Soil 1996, 187 (2), 333−342. (16) Meldrum, F. C.; Colfen, H. Controlling mineral morphologies and structures in biological and synthetic systems. Chem. Rev. 2008, 108 (11), 4332−4432. (17) Rochette, P.; Gregorich, E. G. Dynamics of soil microbial biomass C, soluble organic C and CO2 evolution after three years of manure application. Can. J. Soil Sci 1998, 78 (2), 283−290. (18) Dentener, F. J.; Carmichael, G. R.; Zhang, Y.; Lelieveld, J.; Crutzen, P. J. Role of mineral aerosol as a reactive surface in the global troposphere. J. Geophys. Res.: Atmos. 1996, 101 (D17), 22869−22889. (19) Kump, L. R.; Brantley, S. L.; Arthur, M. A. Chemical, weathering, atmospheric CO2, and climate. Annu. Rev. Earth Planet. Sci. 2000, 28, 611−667. (20) Sabine, C. L.; Feely, R. A.; Gruber, N.; Key, R. M.; Lee, K.; Bullister, J. L.; Wanninkhof, R.; Wong, C. S.; Wallace, D. W. R.; Tilbrook, B.; Millero, F. J.; Peng, T. H.; Kozyr, A.; Ono, T.; Rios, A. F. The oceanic sink for anthropogenic CO2. Science 2004, 305 (5682), 367−371. (21) Michel, F. M.; Barron, V.; Torrent, J.; Morales, M. P.; Serna, C. J.; Boily, J. F.; Liu, Q.; Ambrosini, A.; Cismasu, A. C.; Brown, G. E., Jr. Ordered ferrimagnetic form of ferrihydrite reveals links among structure, composition, and magnetism. Proc. Natl. Acad. Sci. U.S.A. 2010, 107 (7), 2787−92. (22) Russell, J. D.; Paterson, E.; Fraser, A. R.; Farmer, V. C. Adsorption of carbon-dioxide on goethite (Alpha-Feooh) surfaces, and Its implications for anion adsorption. J. Chem. Soc., Faraday Trans. 1 1975, 71 (7), 1623−1630. (23) Bargar, J. R.; Kubicki, J. D.; Reitmeyer, R.; Davis, J. A. ATRFTIR spectroscopic characterization of coexisting carbonate surface complexes on hematite. Geochim. Cosmochim. Acta 2005, 69 (6), 1527−1542. (24) Hausner, D. B.; Bhandari, N.; Pierre-Louis, A. M.; Kubicki, J. D.; Strongin, D. R. Ferrihydrite reactivity toward carbon dioxide. J. Colloid Interface Sci. 2009, 337 (2), 492−500. (25) Baltrusaitis, J.; Jensen, J. H.; Grassian, V. H. FTIR Spectroscopy combined with isotope labeling and quantum chemical calculations to investigate adsorbed bicarbonate formation following reaction of carbon dioxide with surface hydroxyl groups on Fe2O3 and Al2O3. J. Phys. Chem. B 2006, 110 (24), 12005−12016. (26) Hiemstra, T.; Rahnemaie, R.; van Riemsdijk, W. H. Surface complexation of carbonate on goethite: IR spectroscopy, structure and charge distribution. J. Colloid Interface Sci. 2004, 278 (2), 282−290. (27) Baltrusaitis, J.; Schuttlefield, J. D.; Zeitler, E.; Jensen, J. H.; Grassian, V. H. Surface reactions of carbon dioxide at the adsorbed water-oxide interface. J. Phys. Chem. C 2007, 111 (40), 14870−14880. (28) Chernyshova, I. V.; Ponnurangam, S.; Somasundaran, P. Linking interfacial chemistry of CO2 to surface structures of hydrated metal oxide nanoparticles: Hematite. Phys. Chem. Chem. Phys. 2013, 15 (18), 6953−64.

(29) Kubicki, J. D.; Kwon, K. D.; Paul, K. W.; Sparks, D. L. Surface complex structures modelled with quantum chemical calculations: Carbonate, phosphate, sulphate, arsenate and arsenite. Eur. J. Soil Sci. 2007, 58 (4), 932−944. (30) Pierre-Louis, A. M.; Hausner, D. B.; Bhandari, N.; Li, W.; Kim, J.; Kubicki, J. D.; Strongin, D. Adsorption of carbon dioxide on Al/Fe oxyhydroxide. J. Colloid Interface Sci. 2013, 400, 1−10. (31) Song, X. W.; Boily, J. F. Structural controls on OH site availability and reactivity at iron oxyhydroxide particle surfaces. Phys. Chem. Chem. Phys. 2012, 14 (8), 2579−2586. (32) Song, X.; Boily, J.-F. Water vapor interactions with FeOOH particle surfaces. Chem. Phys. Lett. 2013, 560, 1−9. (33) Ding, X.; Song, X.; Boily, J.-F. Identification of fluoride and phosphate binding sites at FeOOH surfaces. J. Phys. Chem. C 2012, 116 (41), 21939−21947. (34) Rustad, J. R.; Boily, J. F. Density functional calculation of the infrared spectrum of surface hydroxyl groups on goethite (alphaFeOOH). Am. Mineral. 2010, 95 (2−3), 414−417. (35) Delbene, J. E.; Person, W. B.; Szczepaniak, K. Properties of hydrogen-bonded complexes obtained from the B3lyp functional with 6-31g(D,P) and 6-31+G(D,P) basis-sets - comparison with Mp2/631+G(D,P) results and experimental-data. J. Phys. Chem. 1995, 99 (27), 10705−10707. (36) Wong, M. W. Vibrational frequency prediction using density functional theory. Chem. Phys. Lett. 1996, 256 (4−5), 391−399. (37) Frisch, M. J. T., G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J. Gaussian 09 Revision A.1; Gaussian Inc.: Wallingford, 2009. (38) Barron, V.; Torrent, J. Surface hydroxyl configuration of various crystal faces of hematite and goethite. J. Colloid Interface Sci. 1996, 177 (2), 407−410. (39) Irikura, K. K.; Johnson, R. D.; Kacker, R. N. Uncertainties in scaling factors for ab initio vibrational frequencies. J. Phys. Chem. A 2005, 109 (37), 8430−8437. (40) Bearat, H.; McKelvy, M. J.; Chizmeshya, A. V. G.; Sharma, R.; Carpenter, R. W. Magnesium hydroxide dehydroxylation/carbonation reaction processes: Implications for carbon dioxide mineral sequestration. J. Am. Ceram. Soc. 2002, 85 (4), 742−748. (41) Jolivet, J. P.; Thomas, Y.; Taravel, B.; Lorenzelli, V. Vibrational study of coordinated CO3 ions. J. Mol. Struct. 1980, 60 (Jan), 93−98. (42) Jolivet, J. P.; Thomas, Y.; Taravel, B.; Lorenzelli, V.; Busca, G. Infrared-spectra of cerium and thorium penta-carbonate complexes. J. Mol. Struct. 1982, 79 (1−4), 403−408. (43) Tribe, L.; Kwon, K. D.; Trout, C. C.; Kubicki, J. D. Molecular orbital theory study on surface complex structures of glyphosate on goethite: Calculation of vibrational frequencies. Environ. Sci. Technol. 2006, 40 (12), 3836−3841. (44) Kaplun, M.; Nordin, A.; Persson, P. On the anomalous adsorption of [Pd(edta)](2−) at the water/goethite interface: Spectroscopic evidence for two types of surface complexes. Langmuir 2008, 24 (2), 483−489. (45) Lo, C.; Tanwar, K.; Chaka, A.; Trainor, T. Density functional theory study of the clean and hydrated hematite (1102) surfaces. Phys. Rev. B 2007, 75, 7. (46) Effenberger, H.; Mereiter, K.; Zemann, J. Crystal-structure refinements of magnesite, calcite, rhodochrosite, siderite, smithonite, 9247



Article

and dolomite, with discussion of some aspects of the stereochemistry of calcite type carbonates. Z. Kristallogr. 1981, 156 (3−4), 233−243. (47) Fernández-Carrasco, L.; Puertas, F.; Blanco-Varela, M. T.; Vázquez, T.; Rius, J. Synthesis and crystal structure solution of potassium Dawsonite: An intermediate compound in the alkaline hydrolysis of calcium aluminate cements. Cem. Concr. Res. 2005, 35 (4), 641−646. (48) Villalobos, M.; Leckie, J. O. Surface complexation modeling and FTIR study of carbonate adsorption to goethite. J. Colloid Interface Sci. 2001, 235 (1), 15−32. (49) Sakakura, T.; Choi, J. C.; Yasuda, H. Transformation of carbon dioxide. Chem. Rev. 2007, 107 (6), 2365−2387. (50) Sakakura, T.; Kohno, K. The synthesis of organic carbonates from carbon dioxide. Chem. Commun. 2009, 11, 1312−1330. (51) Carnes, C. L.; Klabunde, K. J. The catalytic methanol synthesis over nanoparticle metal oxide catalysts. J. Mol. Catal A: Chem. 2003, 194 (1−2), 227−236. (52) Bhanage, B. M.; Fujita, S.; Ikushima, Y.; Arai, M. Synthesis of dimethyl carbonate and glycols from carbon dioxide, epoxides, and methanol using heterogeneous basic metal oxide catalysts with high activity and selectivity. Appl. Catal., A 2001, 219 (1−2), 259−266.

9248


Carbon Dioxide Binding at Dry FeOOH Mineral Surfaces: Evidence for

Recommend Documents