Subscriber access provided by McMaster University Library
Article
Atomistic Molecular Dynamics Simulations of Carbohydrate-Calcite Interactions in Concentrated Brine Hsieh Chen, Athanassios Z. Panagiotopoulos, and Emmanuel P. Giannelis Langmuir, Just Accepted Manuscript • DOI: 10.1021/la504595g • Publication Date (Web): 09 Feb 2015 Downloaded from http://pubs.acs.org on February 18, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Atomistic Molecular Dynamics Simulations of Carbohydrate-Calcite Interactions in Concentrated Brine Hsieh Chen,1,* Athanassios Z. Panagiotopoulos,2 and Emmanuel P. Giannelis3 1
Aramco Services Company: Aramco Research Center – Boston, Cambridge, MA 02139
2
Department of Chemical and Biological Engineering, Princeton University, Princeton, NJ 08544
3
Department of Materials Science and Engineering, Cornell University, Ithaca, NY 14853
*[email protected]
1
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 21
Abstract We report atomistic Molecular Dynamics simulations to study the interactions of a model carbohydrate monomer (Glucopyranose) and calcite slabs in brine. We show that the interactions between the sugar molecules and the mineral decrease with increasing salinity. The decrease is due to the formation of salt layers on the calcite surfaces, which screen the carbohydrate-calcite hydrogen bonding. This screening effect depends on the affinities of calcite surface to specific ions as well as to the carbohydrate molecules.
2
ACS Paragon Plus Environment
Page 3 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Introduction Calcite (CaCO3) is one of the most abundant minerals on earth and is present in nearly all geological systems. It is the main constituent of chalk and limestone, which serve as hosts for oil reservoirs and drinking water aquifers, and it is an important industrial material with uses in cement, pigments, and fillers. In addition, calcite is the preferred biomineral by many organisms.1 The interaction of water and organic molecules with calcite surfaces controls a broad range of natural and industrial processes such as ion exchange,2 contaminant migration,3,4 enhanced oil recovery,5 biomineralization,1,6,7 and flocculation.8 In the past, research on carbohydrate-calcite interactions has mainly focused on their implications in biomineralization.1,9–11 For example, the unicellular algae, Emiliania huxleyi, forms a body cover of calcified platelets called coccoliths. These coccoliths consist of interlocking calcite single crystals, each of which has highly complicated yet strictly defined structure.12 Complex carbohydrate molecules, coccolith-associated polysaccharides (CAP), are generally believed to control the biomineralization process by adsorbing preferentially onto particular surfaces of the calcite crystal.13 Molecular Dynamics (MD) simulations have been used to study the carbohydrate-calcite interactions in pure water, and wide range of adsorption energies were observed depending on different carbohydrate monomers and calcite surfaces.9 In enhanced oil recovery applications, many carbohydrate polymers have been proposed as additives for waterflood (i.e., a method in which water is injected into oil reservoirs to displace residual oil) to decrease water mobility and improve the sweep efficiency.14,15 In spite of its immediate relevance in such applications, however, to the best of our knowledge there have been no thorough studies on the carbohydrate-calcite interactions in reservoir environments. A fundamental difference for carbohydratecalcite interactions in oil reservoirs relative to the well-studied biomineralization process is that oil reservoirs represent a much harsher environment with high temperature (>100 °C), high pressure, and 3
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 21
high salinity (>120,000 ppm of total dissolved solids, which here we use the convention that ppm corresponds to mg of solute / L of solution). The objective of this study is to provide a molecular-level understanding of the carbohydrate-calcite interactions in such harsh reservoir environment, especially to understand how different salinity (brine concentration) mediate the carbohydrate-calcite interactions. It is well-established that carbohydrates form hydrogen bonds with the carbonate groups on calcite surfaces.9 However, there have been no studies on how the salinity in solution can mediate these interactions. In this study we performed MD simulations of dilute carbohydrate monomers between calcite slabs with salinity from 0 ppm to 277,000 ppm, covering a broad range of salinities from DI water to highly saturated brine. In order to have a realistic representation of the carbohydrate-calcite interactions in an oil reservoir, we extend the study to different calcite crystallographic surfaces.16 Because of computational limitations, it has been difficult to simulate large carbohydrate polymers with atomistic details. Thus, in this study we only focus on the interactions between calcites and a simple carbohydrate monomer (Glucopyranose). Nevertheless, the atomistic mineral/carbohydrate monomer interactions studied here can translate to coarse-grained carbohydrate polymer models with well-defined matching observables. We believe the insights gained from this study will further the development of optimum carbohydrate polymers for enhanced oil recovery and other applications with superior reservoir stabilities.
Methods The system studied consists of dilute solution of carbohydrate monomers (α-D-Glucopyranose) between calcite slabs in brines of different concentrations. We have chosen to study calcite slabs terminated by (10
1 4), (11 2 0), (10 1 0), or (10 1 1) planes because these low index surfaces have been shown to exhibit lower surface energies and are the stable cleavage plans for calcites.16–18 Notice that the (10 1 1) surface can be terminated in two different ways: (i) the surface plane containing carbonate oxygens and calcium 4
ACS Paragon Plus Environment
Page 5 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
ions [(10 1 1)Ca], or (ii) the surface plane containing protruded carbonate groups [(10 1 1)CO3]. In order to remove surface dipoles from the (10 1 1) surfaces, half of the surface ions are placed on both sides of the slab, obtaining slab surfaces which are 50% vacant in either the Ca ions or the carbonate groups.17,18 For our studies a base connate fluid with total salinity of 213,000 ppm (213 g/L) with ionic components 59,500 ppm (59.5 g/L) Na+, 19,000 ppm (19 g/L) Ca2+, 2,500 ppm (2.5 g/L) Mg2+ and 132,000 ppm (132 g/L) Cl- was considered. For the simulations, we vary the brine concentration relative to connate fluid from 0 (DI water) to 1.3 (277,000 ppm), covering a broad range of the salinities of the oil fields with different degree of production. In the simulations, the calcite and calcite-water interactions are described by the recently developed force field of Xiao et al,19 which all the non-bonded van der Waals interactions are described by Lennard-Jones 12-6 (LJ) potentials. In this force field, water is treated using the TIP3P model. The CaCO3 force field has been used to calculate the hydration layers and the adsorption of short peptides and other biomolecules on calcite surfaces.19,20 In studying the calcite-organic molecule interactions, organic molecules are modeled with the OPLS_AA force field,21 and a geometric-average mixing rule is used to deduce all the pairwise LJ potentials from atom-wise CaCO3 or OPLS_AA parameters, consistent with the convention used in the model development.19 In addition, the OPLS_AA force field for carbohydrates developed by Damm et al.22 was used for the carbohydrate monomers. The cations and anions are modeled by the force field included in OPLS_AA parameter set originally developed by Åqvist and Chandrasekhar et al.23,24 All Molecular Dynamics simulations were performed at temperature T = 377 K and pressure P = 24.1 MPa (3500 psi), which represent standard reservoir parameters in Saudi Arabia oil fields.25,26 Note that the temperature and pressure values used here deviated from standard conditions (usually 300 K and 1 atm) where most of the force fields were parameterized. Nevertheless, we believe the simulation results obtained here still capture the essential physics of the studied system. Periodic boundary conditions were used in 3 spatial dimensions, with an extended calcite slab perpendicular to the z axis. The calcite surface 5
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 21
areas and slab thicknesses are 5.85-8.51 nm2 and 1.00-1.68 nm depending on the terminated crystallographic planes. Initially, the brine solutions containing 5 carbohydrate monomers are confined between calcite slabs separated by ~3.65-5.48 nm gaps. (For a summary for the dimensions and atom numbers of the calcite systems with different termination planes, see Supplementary Information Table S1.) The geometry is sized such that for each salinity condition all solution has constant number of water molecules and salts. For the salinity of connate fluid (213,000 ppm) there are ~830 water molecules, 48 Na+, 9 Ca2+, 2 Mg2+, and 70 Cl- ions. All simulations were carried out using GROMACS version 4.6.5.27 Electrostatic interactions were calculated using the particle-mesh Ewald (PME) summation, with a real-space cutoff of 1.0 nm, a grid spacing of 0.16 nm and fourth-order interpolation. The van der Waals and neighbor-list cutoffs were both set to 1.0 nm. We used Nose-Hoover temperature coupling with a time constant of 0.5 ps and ParrinelloRahman semi-isotropic pressure coupling with a time constant of 1 ps. The simulation time step was set to 2 fs. Initially, simulation boxes with calcite slabs and randomly-placed carbohydrate monomers are filled with water molecules. Some of the randomly-chosen water molecules are then replaced by cations or anions to reach the desired brine concentrations. After a steepest-descent energy minimization to remove atomic overlaps, systems are equilibrated with 0.1 ns NVT and 1 ns NPT simulations following 100 ns production runs in the NPT ensemble (T = 377 K and P = 24.1 MPa) with data collection every 2 ps. Notice that the calcite slabs are restrained to their lattice positions to prevent any drastic conformational fluctuation in the high temperature and pressure.
Results and Discussion Carbohydrate-calcite interactions in different brine concentrations
6
ACS Paragon Plus Environment
Page 7 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
To understand the carbohydrate-calcite interactions in different brine concentrations, we performed MD simulations of dilute carbohydrate monomers between calcite slabs with salinity from 0 ppm (DI water) to 277,000 ppm (1.3 times connate fluid salinity). Within 100 ns simulations, the carbohydrate monomers are able to sample all the space between calcite slabs. Figure 1 shows the monomer vertical positions as a function of simulation time between calcite slabs terminated with a (10 1 4) plane for three different salinities. Different colors represent the trajectories for different monomers. Interestingly, we observe that high brine concentrations prevent the carbohydrates from adsorbing to calcite surfaces, as can be clearly seen from Fig. 1. In DI water, the monomers which diffuse to the calcite surfaces stick there for prolonged time [Fig. 1(A)]. In contrast, in brine with salinity of 107,000 ppm, the monomers diffuse within the gap partially [Fig. 1(B)], and with salinity of 213,000 ppm, the monomers freely diffuse within the gap most of the time [Fig. 1(C)]. The decreased carbohydrate adsorption with increasing brine concentration is caused by the formation of adsorbed salt layers on calcite surfaces in higher brine concentrations. These salt layers then screen the carbohydrate-calcite hydrogen bonding. Figure 2 shows representative snapshots for the simulation systems described in Fig. 1. Note that the water molecules are removed for clarity. Salt layers (mainly sodium and chloride ions; blue and yellow beads) are clearly seen on top of the calcite slabs in brine with salinity 107,000 and 213,000 ppm [Fig. 2(B) and (C)]. The densities of the brine components as well as the calcite oxygens along the z axis for the above mentioned systems [DI water, 107,000 and 213,000 ppm brine solutions on calcite (10 1 4) surface] are plotted in Fig. 3. The outermost calcite planes that contain calcium atoms are used as the origin of the distance to surface measurement. The (10 1 4) plane has high density of calcium and carbonate ions that the brine solutions never penetrate the outermost calcite oxygens (red curves in Fig. 3). However, it may not be the case for the surfaces with other terminations (discussed in detail later). For water distributions (gray curves in Fig. 3), we see three distinct peaks that indicate three hydration layers exist over the calcite (10 1 4) surface, agreeing well with experimental data and previous MD simulation results.28–31 For 7
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 21
107,000 and 213,000 ppm brines [Fig. 3(B) and (C)], we see salt layers on top of the calcite surfaces which are mainly sodium and chloride ions (blue and yellow curves in Fig. 3). These sodium and chloride ions play an important role in mediating the carbohydrate-calcite interactions in our simulations. The distributions of divalent Ca and Mg ions are not clear here probably because of their small numbers comparing to Na and Cl ions. In Fig. 3(B) and (C), the peaks of the first salt layer are between the peaks of the first and second hydration layers. It may be misinterpreted that the ions are unable to penetrate the first hydration layer. However, as can be clearly seen from simulation snapshots in Fig. 4, where the same simulation frames in Fig. 2 are shown again but with the calcites and ions drawn with the sizes of their ionic radii, the ions are in fact in direct contact with the calcite slabs.
Carbohydrate-calcite interactions on different calcite crystallographic planes Figure 5(A) shows the percentage of adhesion time for carbohydrate monomers on different calcite surfaces [(10 1 4), (11 2 0), (10 1 0), (10 1 1)Ca, and (10 1 1)CO3; Fig. 5(B)] as a function of brine concentration. The carbohydrate adhesion on calcite is defined as when they form hydrogen bonds. For each data point, the adhesion time is averaged from the trajectories of the 5 monomers during the 100 ns simulation. Extra simulations for conditions at which large fluctuations are present were also performed to improve the statistics. The decreased adhesion time with increasing brine concentration is clearly seen in Fig. 5(A) for all simulated calcite surfaces. In high brine concentrations, adsorbed salt layers are observed on all calcite surfaces, as seen in Fig. 6, which shows the representative simulation snapshots for carbohydrates in between different calcite surfaces in salinity of 213,000 ppm. The adsorbed salts screen the carbohydrate-calcite interactions and decrease the carbohydrate adhesions on all calcite surfaces.
8
ACS Paragon Plus Environment
Page 9 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Although the salt dependence for the carbohydrate-calcite interactions is qualitatively the same for all surfaces, there are interesting differences between surfaces as well. On different surfaces, the adhesion time decreases with different rates, when increasing brine concentration. For example, on (10 1 4) plane, the percentage adhesion time decreases slowly from 0.9 to 0.3 when salinity increases from 0 to 277,000 ppm, while on (10 1 1)Ca plane, the percentage adhesion time decreases rapidly from 0.7 to
Article
Atomistic Molecular Dynamics Simulations of Carbohydrate-Calcite Interactions in Concentrated Brine Hsieh Chen, Athanassios Z. Panagiotopoulos, and Emmanuel P. Giannelis Langmuir, Just Accepted Manuscript • DOI: 10.1021/la504595g • Publication Date (Web): 09 Feb 2015 Downloaded from http://pubs.acs.org on February 18, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Atomistic Molecular Dynamics Simulations of Carbohydrate-Calcite Interactions in Concentrated Brine Hsieh Chen,1,* Athanassios Z. Panagiotopoulos,2 and Emmanuel P. Giannelis3 1
Aramco Services Company: Aramco Research Center – Boston, Cambridge, MA 02139
2
Department of Chemical and Biological Engineering, Princeton University, Princeton, NJ 08544
3
Department of Materials Science and Engineering, Cornell University, Ithaca, NY 14853
*[email protected]
1
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 21
Abstract We report atomistic Molecular Dynamics simulations to study the interactions of a model carbohydrate monomer (Glucopyranose) and calcite slabs in brine. We show that the interactions between the sugar molecules and the mineral decrease with increasing salinity. The decrease is due to the formation of salt layers on the calcite surfaces, which screen the carbohydrate-calcite hydrogen bonding. This screening effect depends on the affinities of calcite surface to specific ions as well as to the carbohydrate molecules.
2
ACS Paragon Plus Environment
Page 3 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Introduction Calcite (CaCO3) is one of the most abundant minerals on earth and is present in nearly all geological systems. It is the main constituent of chalk and limestone, which serve as hosts for oil reservoirs and drinking water aquifers, and it is an important industrial material with uses in cement, pigments, and fillers. In addition, calcite is the preferred biomineral by many organisms.1 The interaction of water and organic molecules with calcite surfaces controls a broad range of natural and industrial processes such as ion exchange,2 contaminant migration,3,4 enhanced oil recovery,5 biomineralization,1,6,7 and flocculation.8 In the past, research on carbohydrate-calcite interactions has mainly focused on their implications in biomineralization.1,9–11 For example, the unicellular algae, Emiliania huxleyi, forms a body cover of calcified platelets called coccoliths. These coccoliths consist of interlocking calcite single crystals, each of which has highly complicated yet strictly defined structure.12 Complex carbohydrate molecules, coccolith-associated polysaccharides (CAP), are generally believed to control the biomineralization process by adsorbing preferentially onto particular surfaces of the calcite crystal.13 Molecular Dynamics (MD) simulations have been used to study the carbohydrate-calcite interactions in pure water, and wide range of adsorption energies were observed depending on different carbohydrate monomers and calcite surfaces.9 In enhanced oil recovery applications, many carbohydrate polymers have been proposed as additives for waterflood (i.e., a method in which water is injected into oil reservoirs to displace residual oil) to decrease water mobility and improve the sweep efficiency.14,15 In spite of its immediate relevance in such applications, however, to the best of our knowledge there have been no thorough studies on the carbohydrate-calcite interactions in reservoir environments. A fundamental difference for carbohydratecalcite interactions in oil reservoirs relative to the well-studied biomineralization process is that oil reservoirs represent a much harsher environment with high temperature (>100 °C), high pressure, and 3
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 21
high salinity (>120,000 ppm of total dissolved solids, which here we use the convention that ppm corresponds to mg of solute / L of solution). The objective of this study is to provide a molecular-level understanding of the carbohydrate-calcite interactions in such harsh reservoir environment, especially to understand how different salinity (brine concentration) mediate the carbohydrate-calcite interactions. It is well-established that carbohydrates form hydrogen bonds with the carbonate groups on calcite surfaces.9 However, there have been no studies on how the salinity in solution can mediate these interactions. In this study we performed MD simulations of dilute carbohydrate monomers between calcite slabs with salinity from 0 ppm to 277,000 ppm, covering a broad range of salinities from DI water to highly saturated brine. In order to have a realistic representation of the carbohydrate-calcite interactions in an oil reservoir, we extend the study to different calcite crystallographic surfaces.16 Because of computational limitations, it has been difficult to simulate large carbohydrate polymers with atomistic details. Thus, in this study we only focus on the interactions between calcites and a simple carbohydrate monomer (Glucopyranose). Nevertheless, the atomistic mineral/carbohydrate monomer interactions studied here can translate to coarse-grained carbohydrate polymer models with well-defined matching observables. We believe the insights gained from this study will further the development of optimum carbohydrate polymers for enhanced oil recovery and other applications with superior reservoir stabilities.
Methods The system studied consists of dilute solution of carbohydrate monomers (α-D-Glucopyranose) between calcite slabs in brines of different concentrations. We have chosen to study calcite slabs terminated by (10
1 4), (11 2 0), (10 1 0), or (10 1 1) planes because these low index surfaces have been shown to exhibit lower surface energies and are the stable cleavage plans for calcites.16–18 Notice that the (10 1 1) surface can be terminated in two different ways: (i) the surface plane containing carbonate oxygens and calcium 4
ACS Paragon Plus Environment
Page 5 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
ions [(10 1 1)Ca], or (ii) the surface plane containing protruded carbonate groups [(10 1 1)CO3]. In order to remove surface dipoles from the (10 1 1) surfaces, half of the surface ions are placed on both sides of the slab, obtaining slab surfaces which are 50% vacant in either the Ca ions or the carbonate groups.17,18 For our studies a base connate fluid with total salinity of 213,000 ppm (213 g/L) with ionic components 59,500 ppm (59.5 g/L) Na+, 19,000 ppm (19 g/L) Ca2+, 2,500 ppm (2.5 g/L) Mg2+ and 132,000 ppm (132 g/L) Cl- was considered. For the simulations, we vary the brine concentration relative to connate fluid from 0 (DI water) to 1.3 (277,000 ppm), covering a broad range of the salinities of the oil fields with different degree of production. In the simulations, the calcite and calcite-water interactions are described by the recently developed force field of Xiao et al,19 which all the non-bonded van der Waals interactions are described by Lennard-Jones 12-6 (LJ) potentials. In this force field, water is treated using the TIP3P model. The CaCO3 force field has been used to calculate the hydration layers and the adsorption of short peptides and other biomolecules on calcite surfaces.19,20 In studying the calcite-organic molecule interactions, organic molecules are modeled with the OPLS_AA force field,21 and a geometric-average mixing rule is used to deduce all the pairwise LJ potentials from atom-wise CaCO3 or OPLS_AA parameters, consistent with the convention used in the model development.19 In addition, the OPLS_AA force field for carbohydrates developed by Damm et al.22 was used for the carbohydrate monomers. The cations and anions are modeled by the force field included in OPLS_AA parameter set originally developed by Åqvist and Chandrasekhar et al.23,24 All Molecular Dynamics simulations were performed at temperature T = 377 K and pressure P = 24.1 MPa (3500 psi), which represent standard reservoir parameters in Saudi Arabia oil fields.25,26 Note that the temperature and pressure values used here deviated from standard conditions (usually 300 K and 1 atm) where most of the force fields were parameterized. Nevertheless, we believe the simulation results obtained here still capture the essential physics of the studied system. Periodic boundary conditions were used in 3 spatial dimensions, with an extended calcite slab perpendicular to the z axis. The calcite surface 5
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 21
areas and slab thicknesses are 5.85-8.51 nm2 and 1.00-1.68 nm depending on the terminated crystallographic planes. Initially, the brine solutions containing 5 carbohydrate monomers are confined between calcite slabs separated by ~3.65-5.48 nm gaps. (For a summary for the dimensions and atom numbers of the calcite systems with different termination planes, see Supplementary Information Table S1.) The geometry is sized such that for each salinity condition all solution has constant number of water molecules and salts. For the salinity of connate fluid (213,000 ppm) there are ~830 water molecules, 48 Na+, 9 Ca2+, 2 Mg2+, and 70 Cl- ions. All simulations were carried out using GROMACS version 4.6.5.27 Electrostatic interactions were calculated using the particle-mesh Ewald (PME) summation, with a real-space cutoff of 1.0 nm, a grid spacing of 0.16 nm and fourth-order interpolation. The van der Waals and neighbor-list cutoffs were both set to 1.0 nm. We used Nose-Hoover temperature coupling with a time constant of 0.5 ps and ParrinelloRahman semi-isotropic pressure coupling with a time constant of 1 ps. The simulation time step was set to 2 fs. Initially, simulation boxes with calcite slabs and randomly-placed carbohydrate monomers are filled with water molecules. Some of the randomly-chosen water molecules are then replaced by cations or anions to reach the desired brine concentrations. After a steepest-descent energy minimization to remove atomic overlaps, systems are equilibrated with 0.1 ns NVT and 1 ns NPT simulations following 100 ns production runs in the NPT ensemble (T = 377 K and P = 24.1 MPa) with data collection every 2 ps. Notice that the calcite slabs are restrained to their lattice positions to prevent any drastic conformational fluctuation in the high temperature and pressure.
Results and Discussion Carbohydrate-calcite interactions in different brine concentrations
6
ACS Paragon Plus Environment
Page 7 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
To understand the carbohydrate-calcite interactions in different brine concentrations, we performed MD simulations of dilute carbohydrate monomers between calcite slabs with salinity from 0 ppm (DI water) to 277,000 ppm (1.3 times connate fluid salinity). Within 100 ns simulations, the carbohydrate monomers are able to sample all the space between calcite slabs. Figure 1 shows the monomer vertical positions as a function of simulation time between calcite slabs terminated with a (10 1 4) plane for three different salinities. Different colors represent the trajectories for different monomers. Interestingly, we observe that high brine concentrations prevent the carbohydrates from adsorbing to calcite surfaces, as can be clearly seen from Fig. 1. In DI water, the monomers which diffuse to the calcite surfaces stick there for prolonged time [Fig. 1(A)]. In contrast, in brine with salinity of 107,000 ppm, the monomers diffuse within the gap partially [Fig. 1(B)], and with salinity of 213,000 ppm, the monomers freely diffuse within the gap most of the time [Fig. 1(C)]. The decreased carbohydrate adsorption with increasing brine concentration is caused by the formation of adsorbed salt layers on calcite surfaces in higher brine concentrations. These salt layers then screen the carbohydrate-calcite hydrogen bonding. Figure 2 shows representative snapshots for the simulation systems described in Fig. 1. Note that the water molecules are removed for clarity. Salt layers (mainly sodium and chloride ions; blue and yellow beads) are clearly seen on top of the calcite slabs in brine with salinity 107,000 and 213,000 ppm [Fig. 2(B) and (C)]. The densities of the brine components as well as the calcite oxygens along the z axis for the above mentioned systems [DI water, 107,000 and 213,000 ppm brine solutions on calcite (10 1 4) surface] are plotted in Fig. 3. The outermost calcite planes that contain calcium atoms are used as the origin of the distance to surface measurement. The (10 1 4) plane has high density of calcium and carbonate ions that the brine solutions never penetrate the outermost calcite oxygens (red curves in Fig. 3). However, it may not be the case for the surfaces with other terminations (discussed in detail later). For water distributions (gray curves in Fig. 3), we see three distinct peaks that indicate three hydration layers exist over the calcite (10 1 4) surface, agreeing well with experimental data and previous MD simulation results.28–31 For 7
ACS Paragon Plus Environment
Langmuir
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 8 of 21
107,000 and 213,000 ppm brines [Fig. 3(B) and (C)], we see salt layers on top of the calcite surfaces which are mainly sodium and chloride ions (blue and yellow curves in Fig. 3). These sodium and chloride ions play an important role in mediating the carbohydrate-calcite interactions in our simulations. The distributions of divalent Ca and Mg ions are not clear here probably because of their small numbers comparing to Na and Cl ions. In Fig. 3(B) and (C), the peaks of the first salt layer are between the peaks of the first and second hydration layers. It may be misinterpreted that the ions are unable to penetrate the first hydration layer. However, as can be clearly seen from simulation snapshots in Fig. 4, where the same simulation frames in Fig. 2 are shown again but with the calcites and ions drawn with the sizes of their ionic radii, the ions are in fact in direct contact with the calcite slabs.
Carbohydrate-calcite interactions on different calcite crystallographic planes Figure 5(A) shows the percentage of adhesion time for carbohydrate monomers on different calcite surfaces [(10 1 4), (11 2 0), (10 1 0), (10 1 1)Ca, and (10 1 1)CO3; Fig. 5(B)] as a function of brine concentration. The carbohydrate adhesion on calcite is defined as when they form hydrogen bonds. For each data point, the adhesion time is averaged from the trajectories of the 5 monomers during the 100 ns simulation. Extra simulations for conditions at which large fluctuations are present were also performed to improve the statistics. The decreased adhesion time with increasing brine concentration is clearly seen in Fig. 5(A) for all simulated calcite surfaces. In high brine concentrations, adsorbed salt layers are observed on all calcite surfaces, as seen in Fig. 6, which shows the representative simulation snapshots for carbohydrates in between different calcite surfaces in salinity of 213,000 ppm. The adsorbed salts screen the carbohydrate-calcite interactions and decrease the carbohydrate adhesions on all calcite surfaces.
8
ACS Paragon Plus Environment
Page 9 of 21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Langmuir
Although the salt dependence for the carbohydrate-calcite interactions is qualitatively the same for all surfaces, there are interesting differences between surfaces as well. On different surfaces, the adhesion time decreases with different rates, when increasing brine concentration. For example, on (10 1 4) plane, the percentage adhesion time decreases slowly from 0.9 to 0.3 when salinity increases from 0 to 277,000 ppm, while on (10 1 1)Ca plane, the percentage adhesion time decreases rapidly from 0.7 to
