Infrared Spectroscopy and Density Functional ... - ACS Publications


Infrared Spectroscopy and Density Functional...

0 downloads 207 Views 6MB Size

Article pubs.acs.org/JPCA

Infrared Spectroscopy and Density Functional Theory Investigation of Calcite, Chalk, and CoccolithsDo We Observe the Mineral Surface? M. P. Andersson,* C. P. Hem, L. N. Schultz, J. W. Nielsen, C. S. Pedersen, K. K. Sand, D. V. Okhrimenko, A. Johnsson, and S. L. S. Stipp ABSTRACT: We have measured infrared spectra from several types of calcite: chalk, freshly cultured coccoliths produced by three species of algae, natural calcite (Iceland Spar), and two types of synthetic calcite. The most intense infrared band, the asymmetric carbonate stretch vibration, is clearly asymmetric for the coccoliths and the synthetic calcite prepared using the carbonation method. It can be very well fitted by two peaks: a narrow Lorenzian at lower frequency and a broader Gaussian at higher frequency. These two samples both have a high specific surface area. Density functional theory for bulk calcite and several calcite surface systems allows for assignment of the infrared bands. The two peaks that make up the asymmetric carbonate stretch band come from the bulk (narrow Lorenzian) and from a combination of two effects (broad Gaussian): the surface or near surface of calcite and line broadening from macroscopic dielectric effects. We detect water adsorbed on the high surface area synthetic calcite, which permits observation of the chemistry of thin liquid films on calcite using transmission infrared spectroscopy. The combination of infrared spectroscopy and density functional theory also allowed us to quantify the amount of polysaccharides associated with the coccoliths. The amount of polysaccharides left in chalk, demonstrated to be present in other work, is below the IR detection limit, which is 0.5% by mass.



INTRODUCTION Calcium carbonate is a common biomineral, used by marine organisms as aragonite for shells and nacre and calcite as coccoliths, which are produced by some species of algae. Coccoliths are calcite platelets that interlock around the single cell of the coccolithophorids. Depending on the species, these discs have a diameter ranging from 5 to 15 μm or more and are composed of 20 to 60 individual, submicrometer calcite crystals. Coccoliths from modern species are associated with polysaccharides,1,2 and it is assumed that these complex sugars play an important role in controlling calcite growth during coccolith biomineralization.3−6 Chalk, such as that found in the rock formations of the North Sea Basin, the Gulf of Mexico, and many other localities, is composed chiefly of coccoliths and is of economic interest because it can serve as oil and gas reservoirs as well as groundwater aquifers. The surface properties of calcite in chalk determine the wetting properties and surface reactivity of the rock, both of which are important factors in enhanced oil recovery as well as for drinking water. Synthetic and biogenic calcite have been widely studied using a broad range of techniques, including infrared spectroscopy (IR)7−9 and density functional theory (DFT),10 but much more information could be gained by combining the two approaches within the same investigation, as we do in this paper. Infrared spectroscopy of minerals has the advantage that in addition to distinguishing between different polymorphs,11 it can also detect amorphous mineral phases8 and organic compounds. DFT calculations of vibrational properties for bulk minerals have been shown to give accurate results not only for the vibrational frequencies but also for infrared intensities.10,12−15 © XXXX American Chemical Society

The morphology of calcite varies tremendously depending on the conditions under which the mineral formed. The crystal shape and size differ between calcite grown inorganically in the lab and biogenic calcite such as is found in chalk or coccoliths from coccolithophorids cultured in the lab (Figure 1). There is also variation in crystal size and shape of the calcite produced in different biogenic systems, such as for the various species. This paper aims to describe how crystal morphology relates to material purity and crystallinity of the calcite by comparing the vibrational properties of the mineral samples and associated compounds. We have examined natural calcite, in the form of Cretaceous chalk and Iceland spar, three species of coccoliths from cultured coccolithophorids, and two types of synthetic, inorganically precipitated calcite (Table 1) using IR and compared the results using DFT calculations. There are similarities and differences in the infrared spectra. In particular, high surface area samples yield data where the strongest infrared absorption peak for calcite is clearly asymmetric. Our DFT calculations help us assign this feature at least partly to the calcite surface or very near surface.



COMPUTATIONAL DETAILS The electronic structure calculations were carried out using DFT16,17 in a plane wave pseudopotential implementation using the Quantum Espresso suite of programs.18 We used norm conserving pseudopotentials to represent the ionic cores. The plane wave cutoff in all DFT calculations was 70 Ry, except Received: May 31, 2014 Revised: September 17, 2014

A

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

Approximation (LDA).19 Although more sophisticated approaches exist, this was adequate for our purpose. We used the conventional hexagonal unit cell for bulk calcite (Figure 2a) and calculated the transverse optical−longitudinal optical (LO− TO) splitting using the dynmat.x program with the acoustic sum rule for crystals applied. Several model systems for the calcite surface were used, most of which were based on the thermodynamically most stable surface, the {10.4} plane. This is the most common face in material synthesized in pure solutions and is also a common face on coccolith elements.4,20 All surfaces were created from the LDA relaxed bulk structure and lattice parameters with a 100 Ry plane wave cutoff. The errors in the calculated hexagonal lattice parameters a and c were −0.1% for a and −2.9% for c. Variations included a bare {10.4} surface, water adsorbed on one side of a {10.4} slab, water adsorbed on both sides of a {10.4} slab, and periodicially repeating 1−4 mannose adsorbed on one side of a {10.4} slab (Figure 2b−e). Polymannose was chosen as a model polysaccharide for several reasons. It is a logical choice because the sugar chain backbone of the Emiliana huxleyi polysaccharide consists of mannose.1 The periodicity for 1−4 linked polymannose closely matches the lattice parameters of the calcite {10.4} surface unit cell, thus preventing too much strain in the chain. We also performed calculations for water adsorbed on an acute step, which had a terrace two atomic rows wide, in order to probe a surface termination different from the {10.4}. All results presented here are made assuming surface calcite slabs three molecular layers thick and molecules adsorbed on only one side of the slab. Supplementary calculations with water on both sides of a {10.4} slab and with water on a four layer {10.4} slab confirm that our assumption is reasonable because changes in carbonate peaks for calculated spectra for the larger systems are minor. The uppermost, close-packed layer was in all cases fully relaxed together with the adsorbed molecules during the geometry optimizations, which were considered converged when all forces were less than 0.001 Ha/bohr. We used the equilibrium lattice constants determined using LDA and the same pseudopotentials. We used at least 10 Å of vacuum between the slabs. A Monkhorst−Pack k-point sampling of 4 × 4 × 121 was used for bulk DFT calculations and a 2 × 4 × 1 mesh was used for all slab calculations. The vibrational frequencies were checked for convergence for bulk calcite with respect to the plane wave cutoff. Pseudopotentials for the DFT Calculations. The pseudopotentials used for carbon, oxygen, and hydrogen were C.pz-vbc.UPF, O.pz-mt.UPF, and H.pz-vbc.UPF. All were taken from the Quantum Espresso pseudopotential library. The LDA pseudopotential for calcium in this library only has the 4s electrons as valence electrons, which can be inadequate for highly ionic systems such as calcite. Vibrational calculations using this pseudopotential on the CaO dimer gave a Ca−O stretch frequency of 583 cm−1, which we judged to be too far from the experimental value of 732 cm−1.22 We generated a new pseudopotential for calcium using the ld1 module of Quantum Espresso and included the 3s, 3p, and 3d electrons as valence states (semicore states). The reference configuration was ionic (+1), 3s23p63d1 and the maximum transferability error for nine configurations was below 7 mRy for total energy differences and below 10 mRy for any single atomic energy level. A +2 reference state worked almost as well but gave slightly higher transferability errors for neutral configurations. The test configurations were neutral and ionic (+1 and +2) and included populating the 4s and 4p atomic levels. With this

Figure 1. Scanning electron microscopy (SEM) images of (a) synthetic calcite precipitated from solutions of Na2CO3 and CaCl2; (b) synthetic calcite prepared by carbonation of Ca(OH)2 solutions; (c) E. huxleyi cultured coccoliths; (d) chalk from the North Sea Basin, from a formation that has never contained oil. The scale is the same for all. The SEM images were obtained on a Quanta 200F with a field emission gun (FEG). The samples were coated with 2−3 nm of Au; spot size was 3 nm in diameter, working distance was 3−5 mm and beam energy was 2 to 5 keV.

Table 1. Calcite Samples sample synthetic calcite (precipitated) nanocalcite G. oceanica P. carterae E. huxleyi Iceland Spar North Sea chalk

description synthesized inorganically from CaCl2 and NaCO3 solutions synthetic calcite grown by carbonation of Ca(OH)2 coccoliths extracted from cultured algae coccoliths extracted from cultured algae coccoliths extracted from cultured algae natural calcite from a vein, Chihuahua, Mexico, optical, museum quality natural calcite composed predominantly of Cretaceous (Maastrictian) coccoliths, core sample drilled from a horizon that did not contain oil

the variable cell minimizations, which used 100 Ry. The DFT calculations were performed using the Local Density B

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

Figure 2. Structures used for the density functional theory (DFT) calculations: (a) bulk calcite; (b) polymannose on the calcite {10.4} surface; (c) side view of polymannose on calcite {10.4}; (d) water on an acute step; and (e) side view of water on an acute step.

sampled by extracting 20 mL after 1 day of reaction. The collected solution was vacuum filtered using a cyclopore track etched membrane with a pore size of 0.4 μm. The experiments were done at 24 °C ± 1 °C. The water was ultrapure deionized (DI) water (Milli-Q, resistivity >18 MΩ cm−1; Millipore Corporation). Nanocalcite. Submicrometer calcite (nanocalcite) was prepared according to the method by Schultz et al.23 A 23 mM Ca(OH)2 slurry was prepared in a 2 L Erlenmeyer flask with Milli-Q water. We introduced CO2 gas, bubbling at approximately 0.1 L/min at 24 °C through a gas distribution tube (Pyrex, porosity grade 1) into a vigorously stirred reactor. Solution pH was monitored continuously, and the CO2 source was removed to stop the reaction when the solution pH dropped below 8, suggesting that all Ca(OH)2 had been transformed to calcium carbonate. The solids were removed by centrifugation and freeze-dried prior to storage. Analyses presented in Schultz et al.23 show that the product was chemically and morphologically pure calcite with crystal sizes less than 1 μm. Iceland Spar. Single crystals of optical quality Iceland spar calcite were purchased from Ward’s Natural Science, USA. Immediately before analyzing the calcite, a crystal was cleaved on all faces to remove the material that is invariably

pseudopotential, the Ca−O stretch frequency in the CaO molecule changed to 797 cm−1, which is closer to the experimental value obtained by Hultin et al.22 The suggested cutoff for our pseudopotential is 70 Ry. We calculated the vibrational frequencies for bulk calcite using pseudopotentials with various cutoff radii and suggested kinetic energy cutoff. The frequencies were converged to within 10 cm−1 for kinetic energy cutoff ≥ 70 Ry, and were independent of the choice of cutoff radii when generating the Ca pseudopotential. This is most likely related to the hardness of the oxygen pseudopotential, and we therefore chose to use a Ca pseudopotential with a recommended energy cutoff of 70 Ry, so we could increase the transferability.



EXPERIMENTAL DETAILS The samples used in the study are described in Table 1. Details about preparation methods are presented below. Synthetic Calcite (Inorganically Precipitated). Analytical reagent grade CaCl2 and Na2CO3 were purchased from Sigma-Aldrich and Merck KGaA, respectively. The calcite was prepared by mixing stock solutions of equimolar Na2CO3 and CaCl2 to a final concentration of 25 mM Ca and CO3. The reactor was sealed and placed on a shaking board running at 80 circular motions per minute (MOT). The precipitates were C

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

10% between neighboring points because of the lower signal-tonoise ratio. All spectra were baseline corrected by adding a straight line. Before quantification of the various components by integrating the infrared spectra, we used the extended ATR correction in OPUS 6.0 to convert the MIR ATR spectra to transmission spectra. In ATR, longer wavelengths penetrate further into the sample and to quantify components based on relative intensities at different wavelengths, we need a regular transmission spectrum. We integrated the converted infrared absorption spectrum for the C−C and C−O stretch signal from 900 to 1200 cm−1 as a measure of the amount of polysaccharide and integrated the signal from 1200 to 1600 cm−1 as a measure of the amount of calcite. We did the same for our DFT calculations by summing the intensities for all frequencies in the same frequency regions and converting from intensity per mole to intensity per mass. BET Surface Area Analysis. The specific surface area was determined for all samples from N2 adsorption isotherm measurements at 77 K by means of the volumetric adsorption techniques provided by Quantachrome Nova 2000 Instrument and Quantachrome Autosorb-1 System. The BET area was calculated in the relative pressure range 0.05 < P/P0 < 0.3. The BET plots were linear for all samples in the relative pressure range examined. The r2 values for the regression were higher than 0.99, which confirms the applicability of the BET equation. Degassing treatment before adsorption experiments consisted of heating in vacuum ( 0.1 cm−1 are presented, with the exception of the symmetric carbonate stretch, ν1.

F

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

ascribed to the mineral surface is also a result of line broadening from macroscopic electrostatic interactions from depolarization fields,12,13,34 in which particle shape can strongly influence IR spectra. We suggest that the calcite frequencies obtained from DFT slab calculations mimic the macroscopic dielectric effects by changing the dielectric constant in one direction, by the insertion of vacuum to create the slab, and this is why any calcite surface signal is indistinguishable from broadening due to macroscopic dielectric effects. IR spectra of the 900−1200 cm−1 region are shown in Figure 9 for the coccoliths, along with a model polysaccharide, Naalginate, which does not contain any calcite. We also show DFT calculated spectra for two relevant slabs: water on an acute step and polymannose on the calcite {10.4} surface. Our DFT calculations with polymannose on calcite {10.4} provide relative IR intensities of the organic compound and the surface and thereby allow us to estimate the amount of polysaccharides associated with each species of coccolith (Table 4). The nanocalcite contains no polysaccharide and should have no IR absorption in the 900−1200 cm−1 region, except for a tiny peak at 1084 cm−1, which comes from the ν1 symmetric carbonate stretch of atoms close to the surface or other sites where the symmetry is broken and the infrared transition is not forbidden. This vibration is Raman active and has been measured previously.35 The peak position also matches our DFT calculations very well (Table 3). The DFT corrected ratio of the integrated intensity of the 900−1200 cm−1 region to the 1200−1600 cm−1 region is 0.5%, which we use as an error range for all polysaccharide quantification estimates. The nonzero integration area has at least two causes. One is a result of a varying baseline, the other comes from, for example, carbonate ions in low symmetry sites. DFT calculations for water on a stepped surface show that there are infrared absorption peaks in the 900−1200 cm−1 region even without the presence of any polysaccharides. They are weaker than polysaccharide vibrations but their predicted presence shows that the existence of a surface itself can give rise to weaker adsorption features in this region of the infrared spectrum. The amount of polysaccharide associated with the cultured coccoliths increases in the order G. oceanica < E. huxleyi < P. carterae (Table 4) but it also varies with the preparation procedure. A 2% to 8% weight of polysaccharide is significant, but the values are slightly overestimated by the polysaccharide location, at the surfaces of the mineral. ATR-IR has a penetration depth of ∼1 μm, which is comparable to the average particle size in chalk (Figure 1d). This means that some signal from bulk calcite is not detected using the ATR-IR technique. It is also possible that the amount of polysaccharide that remains associated with the calcite in the cultured coccoliths depends on how rigorous the cleaning procedure was. The quantity of polysaccharide remaining in the chalk was below the detection limit, so we conclude that polysaccharides contribute < 0.5 % by mass in chalk. This low level of organic matter is consistent with previous studies.36,37 A thermogravimetric mass spectrometry analysis of chalk found only 0.1− 0.2% by mass of extracted organic matter.37 X-ray photoelectron spectroscopy showed that much of this organic matter was located at the surface of the calcite.37 The concentration of organic matter of chalk is significantly smaller than for cultured coccoliths and shows that most of the organic matter associated with the chalk was lost during sedimentation, burial, diagenesis, and the ∼60 million years of geological history; however, only a

Figure 6. Comparison of a calculated bulk IR spectrum to an experimental spectrum of North Sea chalk in the wavenumber region for (a) 200−600 cm−1 and (b) 600−2000 cm−1.

observe that the fit for the E. huxleyi coccoliths is not as good as for nanocalcite, and that there are noticeable differences in the transmission spectra for the three species of coccoliths (Figure 4a). This suggests that there is additional structure in the IR absorption peaks for coccoliths that we have not yet been able to model. The additional structure is probably a result of coccolith morphology, which is much more intricate than the rhombohedral nanocalcite, Figure 1. It is however beyond the capabilities of our DFT calculations to analyze it further. From Figure 8 we estimate that about 50% of the infrared intensity comes from the broad Gaussian for nanocalcite. If we assume that nanocalcite is made up of 100 nm cubes (Figure 1b) and estimate that the volume of the outermost 10% of a cube is about half the total volume, we get a surface thickness of about 10 nm. The value seems too large to be entirely consistent with our assumption that the surface gives rise to the Gaussian peak. This implies that not the entire signal from the broad Gaussian comes from the surface. It is worth noting that the magnitudes of the measured surface shifts are of the same magnitude as LO−TO splitting for the corresponding mode. This is probably a result of the sensitivity of a certain mode to the surroundings. If the LO−TO splitting and surface shifts are correlated, any mode with large LO−TO splitting is thus expected to exhibit a larger shift at a surface. It is therefore quite likely that the observed line shape that we first tentatively G

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

Figure 7. Comparison of calculated IR spectrum for bulk and surface calcite compared with North Sea chalk and cultured G. oceanica coccoliths. The composite calculated spectrum (red) consists of 65% bulk and 35% water on an acute step. For all surface systems, the peak width has been set to 50 cm−1 for all peaks.

very thin layer of organic matter needs to be present in order for the surface composition, thus the behavior of chalk, to be significantly different from that of pure calcite. Even the 0.1− 0.2% of organic matter by mass has been shown to be enough to alter the adsorption properties of chalk compared with pure calcite.37,38 Using IR spectroscopy on the nondissolved sample allowed us to estimate the polysaccharide content including nonsoluble sugars that would be missed by water extraction methods. We know that a nonsoluble portion exists,39 but its detailed composition remains unknown because there is not enough material to analyze it with any confidence using standard IR spectroscopy. To find a good model system for chalk, more investigations are needed using other techniques. Our results clearly demonstrate that inorganically precipitated calcite is a poor model for chalk, based on dissimilarities in both SEM and IR data. Depending on which chalk properties one wants to represent with a model system, either the nanocalcite (for physical properties) or cultured coccoliths (for surface chemical properties) are better.

our experimental data with density functional theory calculations, which allowed for assignment of the infrared absorption bands, and which are consistent with both the infrared signal from the surface of calcite on the high surface area samples and line broadening from macroscopic dielectric effects. We cannot assign our observed spectra to either effect alone. We do observe the 15 Å thick water film on the nanocalcite, which permits observation of the surface chemistry of thin films adsorbed to calcite using regular transmission IR spectroscopy on high surface area calcite. Nanocalcite and cultured coccoliths consist of pure, highly crystalline calcite. We were also able to quantify the amount of polysaccharides associated with three species of cultured coccoliths. Soluble polysaccharide concentration in G. oceanica is 1.3% (by weight), 2.6% for E. huxleyi ,and 8.0% for P. carterae. The concentration of soluble polysaccharides in the chalk samples tested is below the IR detection limit, that is, < 0.5% by mass.

CONCLUSIONS We measured infrared spectra from chalk, freshly cultured coccoliths, natural calcite (Iceland spar), and two types of synthetic calcite. The coccoliths and synthetic calcite prepared with the carbonation method (nanocalcite) exhibit an absorption band for the asymmetric carbonate vibration that can be fitted well by two peaks: a low frequency, narrow, Lorenzian peak, and a broader, Gaussian peak at a higher frequency. The coccoliths and the nanocalcite samples share the characteristic of a high specific surface area. We combined

Corresponding Author





AUTHOR INFORMATION

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS We thank Karen Henriksen, Maersk Oil and Gas, for the loan of the chalk samples, Lisbeth Thygesen at the Faculty of Science, University of Copenhagen, for access to the infrared spectrometer, Helene Almind, Institute of Geography and H

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

Table 4. Mass % of Polysaccharides from the Coccolith Species Determined by Density Functional Theory and Infrared Spectroscopy sample

mass % sugars

G. oceanica E. huxleyi P. carterae

1.3 ± 0.5 2.6 ± 0.5 8.0 ± 0.5

Geology, University of Copenhagen for the XRD analyses. We are also grateful for support from MAX-lab staff and to Kai Neufeld and Kim Dalby of the NanoGeoScience group, Copenhagen University, for the SEM images of our chalk samples. The project was funded by the Danish National Advanced Technology Foundation (HTF) and Maersk Oil and Gas A/S.



REFERENCES

(1) Fichtingerschepman, A. M.; Kamerling, J. P.; Versluis, C.; Vliegenthart, J. F. G. Structural Studies of the Methylated, Acidic Polysaccharide Associated with Coccoliths of Emiliania-Huxleyi (Lohmann) Kamptner. Carbohydr. Res. 1981, 93 (1), 105−123. (2) Borman, A. H.; Dejong, E. W.; Thierry, R.; Westbroek, P.; Bosch, L.; Gruter, M.; Kamerling, J. P. Coccolith-Associated Polysaccharides from Cells of Emiliania-Huxleyi (Haptophyceae). J. Phycol. 1987, 23 (1), 118−123. (3) Marsh, M. E. Regulation of CaCO(3) Formation in Coccolithophores. Compar. Biochem. Physiol. B 2003, 136 (4), 743− 754. (4) Young, J. R.; Henriksen, K. Biomineralization within vesicles: The calcite of coccoliths. In Biomineralization; Dove, P. M., De Yoreo , J. J., Weiner, S., Eds.; Mineralogical Society of America: Washington, DC, 2003; Vol. 54, pp 189−215. (5) Crudeli, D.; Young, J. R.; Erba, E.; de Lange, G. J.; Henriksen, K.; Kinkel, H.; Slomp, C. P.; Ziveri, P. Abnormal Carbonate Diagenesis in Holocene-Late Pleistocene Sapropel-Associated Sediments from the Eastern Mediterranean; Evidence from Emiliania Huxleyi coccolith Morphology. Mar. Micropaleontol. 2004, 52 (1−4), 217−240. (6) Henriksen, K.; Stipp, S. L. S. Controlling Biomineralization: The Effect of Solution Composition on Coccolith Polysaccharide Functionality. Cryst. Growth Des. 2009, 9 (5), 2088−2097. (7) Andersen, F. A.; Brecevic, L. Infrared-Spectra of Amorphous and Crystalline Calcium-Carbonate. Acta Chem. Scand. 1991, 45 (10), 1018−1024. (8) Politi, Y.; Arad, T.; Klein, E.; Weiner, S.; Addadi, L. Sea Urchin Spine Calcite Forms via a Transient Amorphous Calcium Carbonate Phase. Science 2004, 306 (5699), 1161−1164. (9) Brusentsova, T. N.; Peale, R. E.; Maukonen, D.; Harlow, G. E.; Boesenberg, J. S.; Ebel, D. Far Infrared Spectroscopy of Carbonate Minerals. Am. Mineral. 2010, 95 (10), 1515−1522. (10) Prencipe, M.; Pascale, F.; Zicovich-Wilson, C. M.; Saunders, V. R.; Orlando, R.; Dovesi, R. The Vibrational Spectrum of Calcite (CaCO3): An ab initio Quantum-Mechanical Calculation. Phys. Chem. Miner. 2004, 31 (8), 559−564. (11) Farmer, V. C. The Infrared Spectra of Minerals; Mineralogical Society: London, 1974. (12) Balan, E.; Saitta, A. M.; Mauri, F.; Calas, G. First-principles modeling of the infrared spectrum of kaolinite. Am. Mineral. 2001, 86 (11−12), 1321−1330. (13) Balan, E.; Saitta, A. M.; Mauri, F.; Lemaire, C.; Guyot, F. Firstprinciples calculation of the infrared spectrum of lizardite. Am. Mineral. 2002, 87 (10), 1286−1290. (14) Valenzano, L.; Torres, F. J.; Klaus, D.; Pascale, F.; ZicovichWilson, C. M.; Dovesi, R. Ab initio study of the vibrational spectrum and related properties of crystalline compounds; the case of CaCO3 calcite. Z. Phys. Chem. 2006, 220 (7), 893−912.

Figure 8. Best fit uses a combination of two peaks: one Lorenzian to represent the bulk and one Gaussian to represent the surface and macroscopic dielectric effects for the infrared absorption peaks from E. huxleyi cultured coccoliths and nanocalcite for the (a) ν2 and (b) ν3 vibrations.

Figure 9. Calculated and experimental infrared absorption spectra in the C−C and C−O stretch region for coccoliths and the model polysaccharide, Na alginate. I

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX

The Journal of Physical Chemistry A

Article

Stretching Modes of Bayerite. Phys. Chem. Miner. 2008, 35 (5), 279− 285. (35) Rutt, H. N.; Nicola, J. H. Raman Spectra of Carbonates of Calcite Structure. J. Phys. C 1974, 7 (24), 4522−4528. (36) Pacey, N. R. Organic Matter in Cretaceous Chalks from Eastern England. Chem. Geol. 1989, 75 (3), 191−208. (37) Okhrimenko, D. V.; Dalby, K. N.; Skovbjerg, L. L.; Bovet, N.; Christensen, J. H.; Stipp, S. L. S. The Surface Reactivity of Chalk (Biogenic Calcite) With Hydrophilic and Hydrophobic Functional Groups. Geochim. Cosmochim. Acta 2014, 128 (0), 212−224. (38) Okhrimenko, D. V.; Nissenbaum, J.; Andersson, M. P.; Olsson, M. H. M.; Stipp, S. L. S. Energies of the Adsorption of Functional Groups to Calcium Carbonate Polymorphs: The Importance of −OH and −COOH Groups. Langmuir 2013, 29 (35), 11062−11073. (39) Hassenkam, T.; Johnsson, A.; Bechgaard, K.; Stipp, S. L. S. Tracking Single Coccolith Dissolution with Picogram Resolution and Implications for CO2 Sequestration and Ocean Acidification. Proc. Natl. Acad. Sci. U.S.A. 2011, 108 (21), 8571−8576.

(15) Valenzano, L.; Noel, Y.; Orlando, R.; Zicovich-Wilson, C. M.; Ferrero, M.; Dovesi, R. Ab Initio Vibrational Spectra and Dielectric Properties of Carbonates: Magnesite, Calcite and Dolomite. Theor. Chem. Acc. 2007, 117 (5−6), 991−1000. (16) Hohenberg, P.; Kohn, W. Inhomogeneous Electron Gas. Phys. Rev. 1964, 136 (3B), B864. (17) Kohn, W.; Sham, L. J. Self-Consistent Equations Including Exchange and Correlation Effects. Phys. Rev. 1965, 140 (4A), A1133. (18) Gianozzi, P.; Baroni, S.; Bonini, N.; Calandra, M.; Car, R.; Cavazzoni, C.; Ceresoli, D.; Chiarotti, G. L.; Coccocioni, M.; Dabo, I.; Dal Corso, A.; de Gironcoli, S.; Fabris, S.; Fratesi, G.; Gebauer, R.; Gerstmann, U.; Gougoussis, C.; Kokalj, A.; Lazzeri, M.; Martin-Samos, L.; Marzari, N.; Mauri, F.; Mazzarello, R.; Paolini, S.; Pasquarello, A.; Paulatto, L.; Sbraccia, C.; Scandolo, S.; Sclauzero, G.; Seitsonen, A. P.; Smogunov, A.; Umari, P.; Wentzcowitch, R. M. QUANTUM ESPRESSO: A Modular and Open-Source Software Project for Quantum Simulations of Materials. J. Phys.: Condens. Matter 2009, 21 (39), 395502. (19) Perdew, J. P.; Zunger, A. Self-Interaction Correction to DensityFunctional Approximations for Many-Electron Systems. Phys. Rev. B 1981, 23 (10), 5048. (20) Henriksen, K.; Stipp, S. L. S.; Young, J. R.; Marsh, M. E. Biological Control on Calcite Crystallisation: AFM Investigation of Coccolith Polysaccharide Function. Am. Mineral. 2004, 89, 1709− 1716. (21) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B: Condens. Matter 1976, 13 (12), 5188−5192. (22) Hultin, M.; Lagerqvist, A. An Infra-red Band-System of Calcium Oxide. Arkiv for Fysik 1951, 2 (6), 471−507. (23) Schultz, L. N.; Andersson, M. P.; Dalby, K. N.; Muter, D.; Okhrimenko, D. V.; Fordsmand, H.; Stipp, S. L. S. High Surface Area Calcite. J. Cryst. Growth 2013, 371, 34−38. (24) Marsh, M. E.; Chang, D. K.; King, G. C. Isolation and Characterization of a Novel Acidic Polysaccharide Containing Tartrate and Glyoxylate Residues From the Mineralized Scales of a Unicellular Coccolithophoroid Alga Pleurochrysis-Carterae. J. Biol. Chem. 1992, 267 (28), 20507−20512. (25) Guillard, R. R.; Ryther, J. H. Studies of Marine Planktonic Diatoms. 1. Cyclotella nana hustedt, and Detonula confervacea (Cleve) Gran. Can. J. Microbiol. 1962, 8 (2), 229. (26) Keller, M. D.; Selvin, R. C.; Claus, W.; Guillard, R. R. L. Media for the Culture of Oceanic Ultraphytoplankton. J. Phycol. 1987, 23 (4), 633−638. (27) Dalby, K. N.; Bovet, N.; Stipp, S. L. S. The Same but Different: Surface Characterization of Biogenic and Inorganic Calcium Carbonates. Unpublished work, 2014. (28) Bohr, J.; Wogelius, R. A.; Morris, P. M.; Stipp, S. L. S. Thickness and Structure of the Water Film Deposited from Vapour on Calcite Surfaces. Geochim. Cosmochim. Acta 2011, 74 (21), 5985−5999. (29) Marsh, M. E. Coccolith Crystals of Pleurochrysis carterae: Crystallographic Faces, Organization, And Development. Protoplasma 1999, 207 (1−2), 54−66. (30) Saruwatari, K.; Nagasaka, S.; Ozaki, N.; Nagasawa, H. Morphological and Crystallographic Transformation from Immature to Mature Coccoliths, Pleurochrysis carterae. Mar. Biotechnol. 2011, 13 (4), 801−809. (31) Wilbur, K. M.; Watabe, N. Experimental Studies on Calcification in Molluscs and Alga Coccolithius Huxleyi. Ann. N.Y. Acad. Sci. 1963, 109 (1), 82. (32) Young, J. R.; Didymus, J. M.; Mann, S. On the Reported Presence of Vaterite and Aragonite in Coccoliths of Emiliania-Huxleyi. Bot. Mar. 1991, 34 (6), 589−591. (33) Ozaki, N.; Okazaki, M.; Kogure, T.; Sakuda, S.; Nagasawa, H. Structural and Functional Diversity of Acidic Polysaccharides from Various Species of Coccolithophorid Algae. Thalassas 2004, 20 (1), 59−68. (34) Balan, E.; Blanchard, M.; Hochepied, J. F.; Lazzeri, M. Surface Modes in the Infrared Spectrum of Hydrous Minerals: The OH J

dx.doi.org/10.1021/jp5053858 | J. Phys. Chem. A XXXX, XXX, XXX−XXX