NMR characterization of hydrocarbon adsorption on calcite surfaces: A first principles study Rochele C. A. Bevilaqua, Vagner A. Rigo, Marcos Veríssimo-Alves, and Caetano R. Miranda Citation: The Journal of Chemical Physics 141, 204705 (2014); doi: 10.1063/1.4902251 View online: http://dx.doi.org/10.1063/1.4902251 View Table of Contents: http://scitation.aip.org/content/aip/journal/jcp/141/20?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Ethanol adsorption on the Si (111) surface: First principles study J. Chem. Phys. 136, 114703 (2012); 10.1063/1.3691892 Characterization of stearic acid adsorption on Ni(111) surface by experimental and first-principles study approach J. Appl. Phys. 109, 07C115 (2011); 10.1063/1.3562040 Ab initio study of EMIM- BF 4 crystal interaction with a Li (100) surface as a model for ionic liquid/Li interfaces in Li-ion batteries J. Chem. Phys. 131, 244705 (2009); 10.1063/1.3273087 Calculation of NMR chemical shifts in organic solids: Accounting for motional effects J. Chem. Phys. 130, 104701 (2009); 10.1063/1.3081630 Atomic and molecular adsorption on RhMn alloy surface: A first principles study J. Chem. Phys. 129, 244711 (2008); 10.1063/1.3046691

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

THE JOURNAL OF CHEMICAL PHYSICS 141, 204705 (2014)

NMR characterization of hydrocarbon adsorption on calcite surfaces: A first principles study Rochele C. A. Bevilaqua,1 Vagner A. Rigo,1,2 Marcos Veríssimo-Alves,1,3 and Caetano R. Miranda1 1

Centro de Ciências Naturais e Humanas, Universidade Federal do ABC, UFABC, Santo André, SP, Brazil Universidade Tecnológica Federal do Paraná, UTFPR, Cornélio Procópio, PR, Brazil 3 Departamento de Física, ICEx, Universidade Federal Fluminense, UFF, Volta Redonda, RJ, Brazil 2

(Received 2 September 2014; accepted 10 November 2014; published online 26 November 2014) The electronic and coordination environment of minerals surfaces, as calcite, are very difficult to characterize experimentally. This is mainly due to the fact that there are relatively few spectroscopic techniques able to detect Ca2+ . Since calcite is a major constituent of sedimentary rocks in oil reservoir, a more detailed characterization of the interaction between hydrocarbon molecules and mineral surfaces is highly desirable. Here we perform a first principles study on the adsorption of hydrocar¯ The simulations were based on Density Functional bon molecules on calcite surface (CaCO3 (1014)). Theory with Solid State Nuclear Magnetic Resonance (SS-NMR) calculations. The Gauge-Including Projector Augmented Wave method was used to compute mainly SS-NMR parameters for 43 Ca, 13 C, and 17 O in calcite surface. It was possible to assign the peaks in the theoretical NMR spectra for all structures studied. Besides showing different chemical shifts for atoms located on different environments (bulk and surface) for calcite, the results also display changes on the chemical shift, mainly for Ca sites, when the hydrocarbon molecules are present. Even though the interaction of the benzene molecule with the calcite surface is weak, there is a clearly distinguishable displacement of the signal of the Ca sites over which the hydrocarbon molecule is located. A similar effect is also observed for hexane adsorption. Through NMR spectroscopy, we show that aromatic and alkane hydrocarbon molecules adsorbed on carbonate surfaces can be differentiated. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4902251] I. INTRODUCTION

Calcite is the most thermodynamically stable anhydrous polymorphic form of calcium carbonate (CaCO3 ), and as a mineral carbonate, is present in many geologic environments.1, 2 It also plays a role in global CO2 balance,3 carbon sequestration,4, 5 biomineralization,6–9 and is of particular interest for industrial scale processes.4, 10–13 Moreover, calcite stands out as the main constituent of sedimentary rocks1 and, therefore, is important to oil exploration processes.14–16 Since petroleum hydrocarbons are the primary constituents in oil,17 which include hexane, benzene, toluene, xylenes, naphthalene, and fluorine, the understanding of the interaction between these organic molecules and calcite surface is crucial. ¯ surAmong the calcite surfaces, we highlight the (1014) face of the hexagonal calcite structure, due to its greater stability, both observed by experimental methods18 and predicted by theory.19–22 Also, this surface has been extensively investigated by means of experimental characterization techniques such as low-energy electron diffraction (LEED),23 x-ray photoelectron spectroscopy (XPS),23 xray scattering,24 x-ray reflectivity,25 grazing incidence xray diffraction (GIXRD),26 and atomic force microscopy (AFM).27–37 Therefore, a wealth of information about the structure of this surface, where the top few atomic layers ex-

0021-9606/2014/141(20)/204705/8/$30.00

hibit long range order as well strong constraints on the chemical speciation and reactivity at the mineral-water interface, is available. In line with this, theoretical calculations38–42 have shown the calcite relaxation processes include rotations and distortions of the carbonate group, accompanied by a displacement of the calcium ion. Moreover, Refs. 38–42 also demonstrate that when water molecules are placed on the surface to produce complete monolayer coverage, the calcite surface is stabilized and the amount of relaxation in the surface layers substantially reduced. A major difficulty in characterizing the interactions of Ca ions with adsorbed molecules is that the experimental techniques mentioned above are neither element selective nor well suited to probe the structure and motion dynamics at the atomic level.43 Additionally, they do not provide information about the chemical environment surrounding the calcite surface, especially because the doubly charged cationic form of calcium, i.e., Ca2+ , is very difficult to detect since it is closed shell d0 metal cation.44 Nuclear Magnetic Resonance (NMR), particularly solid-state Magic Angle Spinning (MAS) NMR, becomes then a competitive candidate method to determine structural features in ordered systems,45–53 allowing for the measurement of chemical shifts (which can be related to structural parameters), the resolution of multiple crystallographic sites, and detection of changes in the

141, 204705-1

© 2014 AIP Publishing LLC

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

204705-2

Bevilaqua et al.

number of inequivalent sites that accompany symmetry changes.54 Furthermore, it has also been proven to be a very promising technique for direct investigation of the chemical properties of calcite surfaces.46–49 Laurencin et al.49, 55 demonstrated the utility of triplequantum43 Ca MAS-NMR technique in distinguishing between the two different crystallographic calcium sites in hydroxyapatite (Ca10 (PO4 )6 (OH)2 ). This study also showed that natural abundance 43 Ca-1 H rotational-echo double-resonance (REDOR) experiments are feasible, thereby opening the possibility of distance measurements involving calcium. Bryce et al.56 combined an experimental-theoretical 43 Ca NMR approach to provide insight on the structure of the vaterite polymorph of calcium carbonate. Their results suggest that the calcium chemical shift tensor is an important novel probe of calcium binding environments in a range of calciumbased materials. Such results suggest that Solid-State NMR (SS-NMR) should be a spectroscopic tool sufficiently sensitive to probe chemical environments of 43 Ca, and would be a very good complementary tool to x-ray crystallographic methods. Thanks to recent developments in SS-NMR, detailed characterization of the interactions between small molecules and inorganic surfaces can now be achieved.57 Besides being suitable for the fine description of interfaces57 as well as a key analytical tool for mineral surfaces58–60 due to its sensitivity to internuclear distances, SS-NMR turns out to be a suitable technique of investigation for organic-inorganic surfaces in nanomaterials.61 In spite of advances in experimental techniques, accurate interpretation of NMR spectra data from ordered and disordered systems increasingly relies on first principles calculations.59, 62–72 Recently, significant progress in Density Functional Theory (DFT)73, 74 methods enabled the calculation of NMR parameters of crystalline compounds for a variety of chemical systems. By using periodic boundary conditions to model a solid, NMR tensors can be calculated accurately using the Gauge-Including Projector Augmented Wave (GIPAW) formalism.75, 76 For a large range of nuclei, the reliability of GIPAW DFT has been assessed by comparing experimental and calculated NMR parameters,59, 62–69, 71, 72, 77–79 and as a result, extensive calculations have been carried out to derive trends which now help in the interpretation of the spectra of compounds of unknown structure.80 In this study, we propose the use of SS-NMR within the formalism of DFT as a means to understand the interaction of hydrocarbons and calcite, which is important to ultimately probe the interactions of molecules with the surfaces present in carbonate oil reservoirs. The paper is structured as follows: first, we briefly describe the theoretical methods and parameters used in our calculations. Next, we proceed to present theoretical results for NMR chemical shifts obtained with GIPAW method computing SS-NMR parameters in order to characterize calcite surfaces, in which calculations for both pristine surfaces and with adsorbed hydrocarbons, such as benzene and hexane, are performed. Finally, we discuss our results and possible applications for studies of adsorbed molecules on mineral surfaces.

J. Chem. Phys. 141, 204705 (2014)

II. METHODOLOGY

¯ surface model was constructed and exThe calcite (1014) plored in our previous study81 by a slab composed by 3 × 1 unit cells along the [42-1] and [010] directions. The Brillouin zone is sampled using a 3 × 1 × 4 Monkhorst-Pack grid,82 which amounts to 12 k-points. Three layers were employed for all supercells and the bottom-most atomic positions of the slab were kept fixed at their bulk positions. For the NMR calculations, a cutoff of 90 Ry was used for the plane-wave basis set, which ensures energy convergence down to 1 mRy/atom. Since periodic boundary conditions have been applied to our slab model, the supercells include an 11 Å vacuum layer, perpendicular to the surface. This vacuum layer, which is defined as the smallest distance between the adsorbate and the bottom of the calcite slab of the upper periodic image, is large enough to accommodate the slab and the adsorbed molecules while avoiding spurious interactions with neighboring images perpendicular to the surface plane. DFT73, 74 calculations were carried out with the Quantum ESPRESSO83 package, which performs self-consistent calculations using a plane-wave basis set to expand the oneelectron wavefunctions of the Kohn-Sham equations.74 The exchange and correlation (XC) energy functional used is the Generalized Gradient Approximation in the Perdew, Burke, and Ernzerhof (GGA-PBE) parametrization.84 The periodicity of the system is explicitly taken into account using a planewave basis set to expand the wave functions using the GIPAW pseudopotential approach.75, 76 The advantage of the GIPAW approach over other pseudopotential methods85, 86 is the possibility of keeping the nodal properties of the wave functions in the neighbourhood of the core in the presence of a magnetic field. Considering the rigid contribution of core electrons with respect to NMR parameters,87 accuracy comparable to all-electron calculations can be achieved.75 The absolute chemical shielding parameters yielded by our first-principles GIPAW calculations are compared to the experimental isotropic chemical shifts through the standard expression, δ iso = σ ref – σ iso , where σ ref and σ iso are the shieldings of the chosen references and the calculated ones. 43 Ca and 17 O chemical shifts were referenced to lime (CaO) and 13 C ones were tetramethylsilane (TMS). The resulting ab initio NMR spectra were obtained by feeding the SIMPSON56 code with the theoretically calculated chemical shifts and quadrupolar interaction parameters, using the experimental magnetic field intensities well as the same experimental parameters to obtain the best comparison with experimental lineshapes.88 In order to analyze the signals of 43 Ca, 13 C, and 17 O nuclei of the calcite surface, three specific sites are monitored, with two of them belonging to the first layer of the calcite surface, labeled near and far from the adsorbate, and the other one belonging to the innermost part of the surface, labeled bulk. Through these, it is possible to distinguish the differences in the NMR signals between atoms of the surface and of the bulk in the presence or absence of adsorbed hydrocarbon molecules. Figure 1 shows the top view of first layer of the optimized surface used for calcite, as well as a side view of the crystal cell.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

204705-3 x y

[42-1]

Bevilaqua et al.

J. Chem. Phys. 141, 204705 (2014)

C_far O_far

(a)

(b)

(c)

(a) Ca_near

C_far

(b)

[010]

O_far

O_near ear C_near

Ca_near O_near

Ca_far

C_near (c)

3.04 Å

Ca_far (d)

3.42 Å

bulk

Ca C O H

FIG. 1. (a) Top and side view of the pristine calcite surface used to study the adsorption of benzene and hexane molecules. (b) and (c) are the top and side views of the benzene and hexane adsorption on calcite surface, respectively. Only the topmost layer of the cell is displayed on top views for clarity. Large spheres represent calcium atoms, medium-sized dark gray and red spheres represent carbon and oxygen atoms, respectively, and small light gray spheres represent hydrogen atoms.

III. RESULTS AND DISCUSSION A. Pristine calcite surfaces

NMR is a very powerful tool for probing short-range ordering and local structure around the nucleus of interest due to its high sensitivity.89 As a consequence, when studying adsorption of molecules over surfaces such as hydrocarbons on calcite surface, changes in the NMR spectra for individual sites on the surface, induced by the interactions with the adsorbed molecules, can be expected. It is therefore essential to probe different sites on the pristine surface, in order to determine changes to the signals from these sites in the presence of molecules and, therefore to assess the usefulness of NMR as a tool for studying the interactions of the adsorbates with mineral surfaces. With this aim, we choose three sites for each of the atomic species (43 Ca, 13 C, and 17 O) present in the calcite crystal, two on the surface and one in the immediately adjacent lower layer, shown in Figure 2(a). We label those sites as follows: (Ca/C/O)_near represent the Ca/C/O, on the surface layer, closest to the adsorbate. Also on the surface layer, we define the Ca/C/O atoms farthest from the molecule, within the boundaries of our supercell, as (Ca/C/O)_far. The Ca/C/O in the immediately adjacent inner layer are labeled (Ca/C/O)_bulk. NMR signals from individual sites of the pristine calcite slab, as described above, are presented in Figures 2(b)–2(d), for Ca, C, and O, respectively, along with the total spectra for each nuclei. For all three elements, as expected, we observe no difference in the signals from the near and far surface atoms. However, we do observe important differences between NMR signals originated from atoms at the calcite surface, and from those of the bulk layer, as shown in Figures 2(b)–2(d). The most dramatic difference is for the Ca signal, which becomes narrower and more intense in the bulk layer, in comparison to that from the surface atoms.

FIG. 2. (a) Top view of the calcite surface used for the calculations, with only the surface atoms shown, and labels as described in the text. (b)–(d) 43 Ca, 13 C, and 17 O theoretical MAS-NMR signals of the pristine calcite surface onto which the hydrocarbon molecules are later adsorbed.

Surface and bulk C atoms also display a very discernible difference in their NMR signals, expressed by the positions and relative heights of the peaks in their surface and bulk NMR signals. Signals from surface and bulk O atoms display less dramatic changes, but one can still notice a change between the relative height of the most pronounced peaks of the NMR signals (see Fig. 2(d)) as well as an overall shift in the position of the curve and a slight narrowing of the curve width for the bulk O atom, compared to that for the surface O atom. This change, which is certainly due to the different atomic environment experienced by the surface atoms with lower coordination numbers, already point out that the presence of adsorbates can possibly be detected by NMR spectroscopy.

B. Calcite surface with adsorbates

Once NMR signals have been characterized for each representative site of the pristine calcite surface, it becomes feasible to investigate the effects of hydrocarbon adsorption. In this section, we present results for the adsorption of the benzene and hexane molecules. Our previous study81 showed a weak (∼0.1 eV) adsorption of hydrocarbons molecules, such as benzene and hexane, ¯ surface. These molecules are representative on calcite (1014) of aromatic and alkane hydrocarbons, and the calculations of Ref. 81 suggested that Ca sites are the most energetically favorable for adsorption of hydrocarbons on the calcite surface. An analysis of the partial density of states (PDOS) for each of the chemical elements, shown in Figure 3, displays no significant difference between the far and near sites of the calcite surface, providing no information about which molecule is adsorbed on the surface. While the PDOS of Ca and C atoms are very similar for both sites and adsorbates, those of O atoms show a small difference between near and far sites but are still insensitive to the kind of adsorbate present on the surface.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

Bevilaqua et al.

204705-4

J. Chem. Phys. 141, 204705 (2014)

Since the analysis of the PDOS for the different chemical species present is unable to provide information about the kind of adsorbate present, we turn to the analysis of NMR spectra, which is more sensitive to the local chemical environment. Figure 4 shows a top view of the lowest-energy configuration of a benzene molecule adsorbed on the calcite surface and 43 Ca/13 C and 17 O MAS-NMR signals of near, far, bulk atoms, as well as the total spectrum of each nuclei of our calcite slab for the lowest-energy configuration. From the spectra presented in Figs. 4(b)–4(d), we can see that spectra for Ca atoms display the most noticeable changes, for both bulk and surface atoms, with small displacements for all three atoms probed. The MAS-NMR curve for Ca_near, while keeping its shape, presents a large negative displacement of 10 ppm with respect to the same atom in the pris-

tine slab. On the other hand, Ca_far and Ca_bulk spectra did not show displacements with respect to those of the pristine osurface. The changes in the NMR spectra are subtle due to the weak interaction of the benzene molecule with calcite, reflected by the large height of the benzene molecule relative to the surface. An analysis for C and O atoms shows changes also to the MAS-NMR spectra for those atoms. This time, we see no displacement in the peaks or change in the relative height; instead, the C_near atom spectrum displays two extra peaks when benzene is present, each at the extreme of the curve. C_far and C_bulk NMR spectra, on the other hand, are unchanged by the presence of the benzene molecule over the calcite surface, with the number of peaks, positions, and relative heights indistinguishable from those for the pristine surface. Changes in the spectra for all three probed O atoms, when benzene is present, happen mainly in the positions of peaks. With respect to the spectrum of the pristine calcite surface, the spectrum from O_near atoms display a slightly negative displacement, while O_far atoms display a positive displacement, larger in absolute value than that of the O_near atom (5 ppm). The peaks in the O_bulk spectrum are unchanged with respect to those for the pristine surface. The most striking changes due to the presence of the adsorbates is seen in the spectra of Ca atoms: the peaks in the spectra of Ca_near and Ca_far atoms, in the presence of the molecules, are considerably lower than those for the pristine surface, and the spectrum is overall much broader. It is also possible to distinguish the spectrum of near and far atoms through a sizeable, almost rigid displacement of the spectra from the two atoms. Therefore, our results suggest that the dominant interaction of the benzene molecule with the surface is with Ca atoms, whose spectra are strongly modified when it adsorbs. At the same time, NMR spectra observed for hexane adsorption, displayed in Figure 5, also show distinct features

(a)

(a)

FIG. 3. PDOS of Ca/C and O sites of calcite surface with (a) benzene and (b) hexane adsorbates. The blue lines represent the PDOS for surface atomic species just below the hydrocarbon, labeled near, while red lines represent surface atomic species farthest from the hydrocarbon, labeled far.

Ca_near

C_far

(b)

Ca_near

(b)

O_far

O_far

O_near

O_near

C_near (c)

C_far

C_near

Ca_far (d)

FIG. 4. (a) Top view of the calcite surface + benzene molecule, with only surface atoms shown, and labels as described in the text. (b)–(d) 43 Ca, 13 C, and 17 O theoretical MAS-NMR signals of the calcite surface + benzene molecule. Curves are labeled as in (a).

(c)

Ca_far

(d)

FIG. 5. (a) Top view of the calcite surface + hexane molecule, with only surface atoms shown, and labels as described in the text. (b)–(d) 43 Ca, 13 C, and 17 O theoretical MAS-NMR signals of the calcite surface + hexane molecule. Curves are labeled as in (a).

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

204705-5

Bevilaqua et al.

when compared to the signals of the pristine calcite surface, albeit less evident than those observed for benzene. The analysis now has added complexity, since hexane not only encompasses more than one Ca site, but also interacts more weakly with the calcite surface, as can be inferred by the larger distance (∼3.42 Å) to the calcite surface, when compared to benzene (∼3.04 Å). Looking at the spectrum for the Ca_near atom, the curve presents a far smaller displacement (5 ppm) than that for the case when benzene is adsorbed on the calcite surface (15 ppm), which we believe to be an indication of the weaker interaction of the molecule with the Ca atoms. It is also interesting to see that the spectra of Ca_near and Ca_far atoms present a smaller relative displacement among them, when compared to the case of benzene adsorption. This is certainly due to the fact that hexane extends over two Ca atoms, while benzene encompasses only one. In spite of that, MAS-NMR is sufficiently sensitive to display differences between the signals of near and far atoms. C and O NMR spectra, on the other hand, do not present any relevant difference with respect to the spectra of the same atoms when benzene is adsorbed. The present results provide yet more evidence that the organic molecules present in oil should interact with the calcite surface primarily through Ca atoms. Additionally, we have performed first principles NMR characterization for structures relaxed with van der Waals (vdW) dispersion corrections, using the same total energy calculation protocol as originally reported in Ref. 81 to verify the effects of the dispersion correction on the NMR spectra. For hexane, the inclusion of vdW corrections leads to an approximation of the whole molecule towards the calcite surface. Therefore, only a rigid displacement on the chemical shift is observed in the Ca NMR spectra for all Ca sites studied (bulk, far and near the hexane molecule), as detailed in the supplementary material.91 For benzene adsorbed on calcite surface, the vdW corrections provide an increase in the asymmetry of the placement of the benzene ring on top of the Ca site. This leads to some variation on the Ca chemical shift in the calculated NMR spectra, with respect to the non-vdW calculations, presented in Fig. 4. For the other C and O spectra, no significant modifications were observed with the inclusion of vdW corrections. While systematic work for the other distinct hydrocarbons and dispersion correction schemes is under way, we can safely affirm that it is possible to differentiate the hydrocarbon species adsorbed, regardless of whether the total energy calculations take vdW corrections into account. This result is consistent with the work of Colas et al.,66 where the authors show the similarities of the experimental NMR spectra with that obtained with non-vdW GIPAW calculations to the CaC2 O4 system. The results for the NMR spectra for benzene are shown in Fig. S1 of the supplementary material.91 In order to better quantify the differences in the NMR signals for Ca, C, and O atoms, two-dimensional (2D) NMR technique was also computed. Although standard, onedimensional (1D) NMR is sufficient to observe distinct peaks for the various functional groups of small molecules, for larger systems many overlapping resonances can make interpretation of NMR spectrum difficult. Two-dimensional (2D)

J. Chem. Phys. 141, 204705 (2014) (a)

(b)

(c)

(d)

FIG. 6. Contour plots of theoretical two-dimensional MAS-NMR 43 Ca (surface)-13 C (molecule) spectra of benzene and hexane adsorbed on calcite surface. (a) 2D MAS-NMR spectrum for Ca_near site in the presence of benzene. (b) 2D MAS-NMR spectrum for Ca_near site in the presence of hexane. (c) 2D MAS-NMR spectrum for Ca_far site in the presence of benzene. (d) 2D MAS-NMR spectrum for Ca_far site in the presence of hexane. In the top left of the figures, the interaction schemes are available.

NMR, however, allows one to circumvent this challenge by adding additional experimental variables and thus introducing a second dimension to the resulting spectrum, providing data that are easier to interpret and often more informative. In this way, an analysis of the intensities and positions of isosurfaces to assess changes in local chemical environments becomes possible. Figures 6–8 show the 2D plots for the interaction of nuclear spins 43 Ca/13 C/17 O with 13 C of the adsorbates, respectively. From the spectra in Fig. 6, the interaction between the calcite surface and the hydrocarbon molecules can be seen through the presence of cross peaks in the diagonal (Figs. 6(a) and 6(c)). It is important to notice that the (a))

(b)

(c))

(d)

FIG. 7. Contour plots of theoretical two-dimensional MAS-NMR 13 C (surface)-13 C (molecule) spectra of benzene and hexane adsorbed on calcite surface. (a) 2D MAS-NMR spectrum for C_near site in the presence of benzene. (b) 2D MAS-NMR spectrum for C_near site in the presence of hexane. (c) 2D MAS-NMR spectrum for C_far site in the presence of benzene. (d) 2D MAS-NMR spectrum for C_far site in the presence of hexane. In the top left of the figures, the interaction schemes are available.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

204705-6

(a)

Bevilaqua et al.

J. Chem. Phys. 141, 204705 (2014)

(b)

IV. CONCLUSIONS

Through first-principles calculations, we investigated the Ca, 17 O, and 13 C MAS-NMR properties for calcite with benzene and hexane molecules adsorbed on the surface. Theoretical 1D and 2D MAS-NMR spectra for these adsorbed molecules were obtained which allow to proposing peak assignments in the spectra of both molecules. Atoms on the surface and in the bulk of calcite can present sizeable displacements of 10 ppm in the 1D MAS-NMR spectra for benzene adsorbed on calcite surface and less expressive, but still important, differences of 4 ppm in the case of hexane adsorbed on the surface. According to our results, the 2D MAS-NMR technique introduces an additional spectral dimension which provides us with further information regarding the nature of the molecules adsorbed on the calcite surface, as seen from the different responses of the analyzed surface sites. In other words, the 2D NMR spectra make data interpretation easier and possibly unambiguous, as shown in the analysis of the spectra for the carbon and oxygen atoms of the calcite carbonate groups, which do not directly interact with the molecules around the calcite surface unlike calcium atoms. The results also show that interactions as weak as those of organic molecules with the calcite surface do generate changes in the MAS-NMR spectra which could be experimentally measured. We can conclude that the high sensitivity of the technique makes it indeed useful for probing the interactions of the organic molecules present in oil and the calcite surface, allowing for the distinction of the molecule adsorbed over a given calcite surface site, with implications for improved oil extraction from mineral reservoirs. In order to improve enhanced oil recovery (EOR) processes90 and increase the amount of oil recovered, the fundamental interactions of oil-mineral should be well understood and characterized. A good understanding of these interactions should provide useful information to devise materials and processes for improving the yield of oil extraction. In this direction, we have shown that MAS-NMR is able to provide information not accessible through other experimental techniques, therefore showing great potential for use in both applied and fundamental science. Our study shows that SS-NMR method can provide complementary information on the nature of chemical interactions of organic molecules with mineral surfaces, such as calcite, otherwise not available from the more common analysis of electronic densities of states. With theoretical SS-NMR, valuable information on the (weak) interaction of organic molecules adsorbed on top of mineral surfaces can be obtained. Thus, SS-NMR constitutes a powerful tool to study the interactions of adsorbates with mineral surfaces, with great potential to contribute to the understanding of the interactions between adsorbates and surfaces which are not yet wellestablished or difficult to characterize experimentally.

43

(c)

(d)

FIG. 8. Contour plots of theoretical two-dimensional MAS-NMR 17 O (surface)-13 C (molecule) spectra of benzene and hexane adsorbed on calcite surface. (a) 2D MAS-NMR spectrum for O_near site in the presence of benzene. (b) 2D MAS-NMR spectrum for O_near site in the presence of hexane. (c) 2D MAS-NMR spectrum for O_far site in the presence of benzene. (d) 2D MAS-NMR spectrum for O_far site in the presence of hexane. In the top left of the figures, the interaction schemes are available.

intensity of these peaks is higher for the case of benzene (Fig. 6(a)) than in the case of hexane (Fig. 6(c)), in agreement with the one-dimensional spectra (Figs. 4 and 5(b)). In the two-dimensional spectra, differences in position and intensity of isosurfaces for each molecule (Figs. 6(a) and 6(b)), allow for an easier determination of the type of hydrocarbon adsorbed on the surface. Additionally, changes in the intensity signals due to the sites near (Figs. 6(a) and 6(b)) to adsorbates remain more discernible than those of far ones (Figs. 6(c) and 6(d)). The 2D NMR spectra indicate that the geometry of the hydrocarbon is a determining factor on the changes in the number and intensity of isosurfaces, and the introduction of an additional dimension to the spectra allows us to better resolve signals which would normally overlap in 1D NMR. For the C atoms, changes in the spectra due to the presence of different molecules become more evident (Figs. 7(a)– 7(d)). While 1D, available in Figs. 3(c) and 4(c), showed a slight displacement for C_near and C_far, the 2D spectra are capable of distinguishing the sites near and far from adsorbates as well the type of hydrocarbon present. As seen in Figs. 7(a)–7(c) and 7(b), there are changes in the positions and number of cross peaks for benzene and hexane, facilitating the distinction between the two molecules. Nevertheless, Figs. 7(c) and 7(d) show less intense isosurfaces, suggesting that the local chemical environment is scarcely changed for the surface C atoms. In turn, the 2D spectrum for oxygen shows small differences with respect to the intensity of isosurfaces, not clearly discernable (Figs. 8(a)–8(d)). Nevertheless, some details in the spectra of Figs. 8(a) and 8(b) might allow for experimentally distinguishing the type of hydrocarbon by analysis of the oxygen 2D MAS-NMR spectrum: in Fig. 8(a), the isosurfaces located in the middle are less intense than the outer ones. The opposite happens in Fig. 8(b), suggesting a different adsorbate.

ACKNOWLEDGMENTS

This work was supported by the Advanced Energy Consortium: http://www.beg.utexas.edu/aec/ Member companies include BP America Inc., BG Group, Petrobras, Repsol,

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

204705-7

Bevilaqua et al.

Schlumberger, Statoil, Shell, and Total. The authors also acknowledge the financial support provided by the Brazilian agencies CAPES, CNPq, and Fapesp. The calculations have been performed at CENAPAD-SP supercomputer facilities. 1 R. J. Reeder, Carbonates (Mineralogical Society of America, 1983), p. 394. 2 J.

W. Morse, R. S. Arvidson, and A. Lüttge, Chem. Rev. 107, 342–381 (2007). 3 S. R. Dickinson, G. E. Henderson, and K. M. McGrath, J. Cryst. Growth 244, 369–378 (2002). 4 C. R. Usher, A. E. Michel, and V. H. Grassian, Chem. Rev. 103, 4883–4940 (2003). 5 M. Vinoba, M. Bhagiyalakshmi, A. N. Grace, D. H. Chu, S. C. Nam, Y. Yoon, S. H. Yoon, and S. K. Jeong, Langmuir 29, 15655–15663 (2013). 6 N. K. Dhami, M. S. Reddy, and A. Mukherjee, Front. Microbiol. 4, 1–13 (2013). 7 J. H. Harding, D. M. Duffy, M. L. Sushko, P. M. Rodger, D. Quigley, and J. A. Elliott, Chem. Rev. 108, 4823–4854 (2008). 8 H. H. Teng, P. M. Dove, C. A. Orme, and J. J. De Yoreo, Science 282, 724–727 (1998). 9 N. R. Chevalier, C. Chevallard, M. Goldmann, G. Brezesinski, and P. Guenoun, Cryst. Growth Des. 12, 2299–2305 (2012). 10 T. R. Soderling and J. T. Stull, Chem. Rev. 101, 2341–2352 (2001). 11 M. P. Hofmann, U. Gbureck, C. O. Duncan, M. S. Dover, and J. E. Barralet, J. Biomed. Mater. Res. B 83, 1–8 (2007). 12 K. Sato, S. Yamaguchi, T. Nemizu, S. Fujita, K. Suzuki, and T. Mori, J. Ceram. Soc. Jpn. 115, 370–373 (2007). 13 H. V. K. Diyabalanage, R. P. Shrestha, T. A. Semelsberger, B. L. Scott, M. E. Bowden, B. L. Davis, and A. K. Burrell, Angew. Chem. Int. Ed. 46, 8995–8997 (2007). 14 S. Morad, I. S. Al-Aasm, M. Sirat, and M. M. Sattar, J. Geochem. Explor. 106, 156–170 (2010). 15 J. Bourdet, J. Pironon, G. Levresse, and J. Tritlla, Mar. Petrol. Geol. 27, 126–142 (2010). 16 V. M. Sánchez and C. R. Miranda, J. Phys. Chem. C 118, 19180–19187 (2014). 17 T. J. Mayer, Anal. Chem. 43, 174–185 (1971). 18 D. D. Archibald, S. B. Qadri, and B. P. Gaber, Langmuir 12, 538–546 (1996). 19 N. H. de Leeuw and S. C. Parker, J. Phys. Chem. B 102, 2914–2922 (1998). 20 J. O. Titiloye, N. H. de Leeuw, and S. C. Parker, Geochim. Cosmochim. Acta 62, 2637–2641 (1998). 21 M. Bruno, F. R. Massaro, M. Rubbo, M. Prencipe, and D. Aquilano, Cryst. Growth Des. 10, 3102–3109 (2010). 22 S. Kerisit, A. Marmier, and S. C. Parker, J. Phys. Chem. B 109, 18211– 18213 (2005). 23 S. L. Stipp and M. F. Hochella, Jr., Geochim. Cosmochim. Acta 55, 1723– 1736 (1991). 24 P. Geissbühler, P. Fenter, E. DiMasi, G. Srajer, L. B. Sorensen, and N. C. Sturchio, Surf. Sci. 573, 191–203 (2004). 25 P. Fenter, P. Geissbühler, E. DiMasi, G. Srajer, L. B. Sorensen, and N. C. Sturchio, Geochim. Cosmochim. Acta 64, 1221–1228 (2000). 26 U. Magdans, X. Torrelles, K. Angermund, H. Gies, and J. Rius, Langmuir 23, 4999–5004 (2007). 27 J. Schütte, P. Rahe, L. Tröger, S. Rode, R. Bechstein, M. Reichling, and A. Kühnle, Langmuir 26, 8295–8300 (2010). 28 S. Rode, N. Oyabu, K. Kobayashi, H. Yamada, and A. Kühnle, Langmuir 25, 2850–2853 (2009). 29 A. L. Rachlin, G. S. Henderson, and M. C. Goh, Am. Mineral. 77, 904–910 (1992). 30 M. Jin, E. Shimada, and Y. Ikuma, J. Ceram. Soc. Jpn. 107, 1166–1170 (1999). 31 S. L. S. Stipp, C. M. Eggleston, and B. S. Nielsen, Geochim. Cosmochim. Acta 58, 3023–3033 (1994). 32 Y. Liang, A. S. Lea, D. R. Baer, and M. H. Engelhard, Surf. Sci. 351, 172– 182 (1996). 33 F. Ohnesorge and G. Binning, Science 260, 1451–1456 (1993). 34 P. Rahe, J. Schütte, and A. Kühnle, J. Phys.: Condens. Matter 24, 354007 (2012). 35 J. Baltrusaitis and V. H. Grassian, J. Phys. Chem. A 116, 9001–9009 (2012). 36 M. Ricci, P. Spijker, F. Stellacci, J.-F. Molinari, and K. Voïtchovsky, Langmuir 29, 2207–2216 (2013).

J. Chem. Phys. 141, 204705 (2014) 37 C.

M. Pina, R. Miranda, and E. Gnecco, Phys. Rev. B 85, 073402 (2012). Kristensen, S. L. S. Stipp, and K. Refson, J. Chem. Phys. 121, 8511 (2004). 39 A. L. Rohl, K. Wright, and J. D. Gale, Am. Mineral. 88, 921–925 (2003). 40 K. Wright, R. T. Cygan, and B. Slater, Phys. Chem. Chem. Phys. 3, 839– 844 (2001). 41 J. S. Lardge, D. M. Duffy, M. J. Gillan, and M. Watkins, J. Phys. Chem. C 114, 2664–2668 (2010). 42 T. Akiyama, K. Nakamura, and T. Ito, Phys. Rev. B 84, 1–10 (2011). 43 Y.-C. Huang, Y. Mou, T. W.-T. Tsai, Y.-J. Wu, H.-K. Lee, S.-J. Huang, and J. C. C. Chan, J. Phys. Chem. B 116, 14295–14301 (2012). 44 D. Laurencin and M. E. Smith, Prog. Nucl. Magn. Reson. Spectrosc. 68, 1–40 (2013). 45 K. J. D. MacKenzie and M. E. Smith, Multinuclear Solid-State NMR of Inorganic Materials (Pergamon, 2002). 46 R. Dupree, A. P. Howes, and S. C. Kohn, Chem. Phys. Lett. 276, 399–404 (1997). 47 Z. Lin, M. Smith, F. Sowrey, and R. Newport, Phys. Rev. B 69, 224107 (2004). 48 A. Wong, A. P. Howes, R. Dupree, and M. E. Smith, Chem. Phys. Lett. 427, 201–205 (2006). 49 D. Laurencin, A. Wong, R. Dupree, and M. E. Smith, Magn. Reson. Chem. 46, 347–350 (2008). 50 K. Shimoda, Y. Tobu, K. Kanehashi, T. Nemoto, and K. Saito, Solid State Nucl. Magn. Reson. 30, 198–202 (2006). 51 K. Shimoda, Y. Tobu, Y. Shimoikeda, T. Nemoto, and K. Saito, J. Magn. Reson. 186, 156–159 (2007). 52 K. Shimoda, Y. Tobu, K. Kanehashi, T. Nemoto, and K. Saito, J. Non-Cryst. Solids 354, 1036–1043 (2008). 53 F. Angeli, M. Gaillard, P. Jollivet, and T. Charpentier, Chem. Phys. Lett. 440, 324–328 (2007). 54 B. L. Phillips, Rev. Mineral. Geochem. 39, 203–240 (2000). 55 D. Laurencin, A. Wong, J. V. Hanna, R. Dupree, and M. E. Smith, J. Am. Chem. Soc. 130, 2412–2413 (2008). 56 D. L. Bryce, E. B. Bultz, and D. Aebi, J. Am. Chem. Soc. 130, 9282–9292 (2008). 57 C. Bonhomme, C. Coelho, N. Baccile, C. Gervais, T. Azaïs, and F. Babonneau, Acc. Chem. Res. 40, 738–746 (2007). 58 F. Pourpoint, C. C. Diogo, C. Gervais, C. Bonhomme, F. Fayon, S. L. Dalicieux, I. Gennero, J.-P. Salles, A. P. Howes, R. Dupree, J. V. Hanna, M. E. Smith, F. Mauri, G. Guerrero, P. H. Mutin, and D. Laurencin, J. Mater. Res. 26, 2355–2368 (2011). 59 F. Pourpoint, C. Gervais, L. Bonhomme-Coury, T. Azaïs, C. Coelho, F. Mauri, B. Alonso, F. Babonneau, and C. Bonhomme, Appl. Magn. Reson. 32, 435–457 (2007). 60 E. Davies, M. J. Duer, S. E. Ashbrook, and J. M. Griffin, J. Am. Chem. Soc. 134, 12508–12515 (2012). 61 N. Folliet, C. Gervais, D. Costa, G. Laurent, F. Babonneau, L. Stievano, J.-F. Lambert, and F. Tielens, J. Phys. Chem. C 117, 4104–4114 (2013). 62 C. Bonhomme, C. Gervais, F. Babonneau, C. Coelho, F. Pourpoint, T. Azaïs, S. E. Ashbrook, J. M. Griffin, J. R. Yates, F. Mauri, and C. J. Pickard, Chem. Rev. 112, 5733–5779 (2012). 63 J. R. Yates, T. N. Pham, C. J. Pickard, F. Mauri, A. M. Amado, A. M. Gil, and S. P. Brown, J. Am. Chem. Soc. 127, 10216–10220 (2005). 64 M. Benoit, M. Profeta, F. Mauri, C. J. Pickard, and M. E. Tuckerman, J. Phys. Chem. B 109, 6052–6060 (2005). 65 M. d’Avezac, N. Marzari, and F. Mauri, Phys. Rev. B 76, 165122 (2007). 66 H. Colas, L. Bonhomme-Coury, C. C. Diogo, F. Tielens, F. Babonneau, C. Gervais, D. Bazin, D. Laurencin, M. E. Smith, J. V. Hanna, M. Daudon, and C. Bonhomme, Cryst. Eng. Commun. 15, 8840 (2013). 67 M. Profeta, F. Mauri, and C. J. Pickard, J. Am. Chem. Soc. 125, 541–548 (2003). 68 C. Gervais, R. Dupree, K. J. Pike, C. Bonhomme, M. Profeta, C. J. Pickard, and F. Mauri, J. Phys. Chem. A 109, 6960–6969 (2005). 69 C. Gervais, C. Coelho, T. Azais, J. Maquet, G. Laurent, F. Pourpoint, C. Bonhomme, P. Florian, B. Alonso, G. Guerrero, P. H. Mutin, and F. Mauri, J. Magn. Reson. 187, 131–140 (2007). 70 F. Pourpoint, A. Kolassiba, C. Gervais, T. Azaïs, L. Bonhomme-Coury, C. Bonhomme, and F. Mauri, Chem. Mater. 19, 6367–6369 (2007). 71 S. E. Ashbrook, A. J. Berry, D. J. Frost, A. Gregorovic, C. J. Pickard, J. E. Readman, and S. Wimperis, J. Am. Chem. Soc. 129, 13213–13224 (2007). 72 M. Profeta, M. Benoit, F. Mauri, and C. J. Pickard, J. Am. Chem. Soc. 126, 12628–12635 (2004). 73 P. Hohenberg and W. Kohn, Phys. Rev. 136, B864 (1964). 38 R.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48

204705-8 74 W.

Bevilaqua et al.

Kohn and L. J. Sham, Phys. Rev. 140, A1133 (1965). Pickard and F. Mauri, Phys. Rev. B 63, 245101 (2001). 76 J. Yates, C. Pickard, and F. Mauri, Phys. Rev. B 76, 024401 (2007). 77 D. L. Bryce and E. B. Bultz, Chem. Eur. J. 13, 4786–4796 (2007). 78 S. E. Ashbrook, L. Le Pollés, R. G. Gautier, C. J. Pickard, and R. I. Walton, Phys. Chem. Chem. Phys. 8, 3423 (2006). 79 L. Truflandier, M. Paris, C. Payen, and F. Boucher, J. Phys. Chem. B 110, 21403–21407 (2006). 80 C. Gervais, D. Laurencin, A. Wong, F. Pourpoint, J. Labram, B. Woodward, A. P. Howes, K. J. Pike, R. Dupree, F. Mauri, C. Bonhomme, and M. E. Smith, Chem. Phys. Lett. 464, 42–48 (2008). 81 V. A. Rigo, C. O. Metin, Q. P. Nguyen, and C. R. Miranda, J. Phys. Chem. C 116, 24538–24548 (2012). 82 H. J. Monkhorst and J. D. Pack, Phys. Rev. B 13, 5188–5192 (1976). 83 P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. Dal Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer, U. Gerstmann, C. Gougoussis, 75 C.

J. Chem. Phys. 141, 204705 (2014) A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. P. Seitsonen, A. Smogunov, P. Umari, and R. M. Wentzcovitch, J. Phys.: Condens. Matter 21, 395502 (2009). 84 J. P. Perdew, K. Burke, and M. Ernzerhof, Phys. Rev. Lett. 77, 3865–3868 (1996). 85 F. Mauri, B. G. Pfrommer, and S. G. Louie, Phys. Rev. Lett. 77, 5300–5303 (1996). 86 D. Sebastiani and M. Parrinello, J. Phys. Chem. A 105, 1951–1958 (2001). 87 T. Gregor, F. Mauri, and R. Car, J. Chem. Phys. 111, 1815 (1999). 88 D. Laurencin, C. Gervais, A. Wong, C. Coelho, F. Mauri, D. Massiot, M. E. Smith, and C. Bonhomme, J. Am. Chem. Soc. 131, 13430–13440 (2009). 89 T. Charpentier, Solid State Nucl. Magn. Reson. 40, 1–20 (2011). 90 V. Alvarado and E. Manrique, Energies 3, 1529–1575 (2010). 91 See supplementary material at http://dx.doi.org/10.1063/1.4902251 about effects of van der Waals (vdW) dispersion corrections on the NMR spectra for benzene and hexane adsorbed on calcite surface.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 131.170.6.51 On: Fri, 13 Mar 2015 11:59:48