Surface Properties of Fluorite in Presence of Water: An Atomistic

Subscriber access provided by Kaohsiung Medical University

B: Glasses, Colloids, Polymers, and Soft Matter

Surface Properties of Fluorite in Presence of Water: An Atomistic Investigation Yann Foucaud, Michael Badawi, Lev O. Filippov, Inna V. Filippova, and Sébastien Lebègue J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.8b02717 • Publication Date (Web): 30 Apr 2018 Downloaded from http://pubs.acs.org on May 2, 2018

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 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 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.

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 27 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

The Journal of Physical Chemistry

Surface Properties of Fluorite in Presence of Water: An Atomistic Investigation Yann Foucaud1,*, Michaël Badawi2,*, Lev O. Filippov1, Inna V. Filippova1 and Sébastien Lebègue2,* 1

Université de Lorraine, Laboratoire GeoRessources, UMR 7359 – CNRS, 2 rue du Doyen

Marcel Roubault, 54 505 Vandœuvre-lès-Nancy-Cedex, France. E-mail: [email protected]. 2

Université de Lorraine, Laboratoire Physique et Chimie Théoriques, UMR 7019 – CNRS,

BP239, Boulevard des Aiguillettes, 54 506 Vandoeuvre-lès-Nancy-Cedex, France. E-mail: [email protected]; [email protected].

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 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 27

Abstract

Density functional theory simulations including a correction for dispersive interactions were performed to investigate the adsorption of water on the main cleavage plane of the fluorite, namely the (111) surface. In the case of a single molecule of water, we observe that the molecular form is preferred over the dissociated one, and absorbs on the surface with an energy of -55 kJ.mol-1, including a significant contribution from the dispersion forces. Also, we show that the substitution of a fluorine atom by a hydroxyl on the surface of fluorite is not energetically favorable. Then, the hydration of the surface in function of the coverage by water molecules was studied in a systematic way. It was shown that the geometries involving the formation of a cluster of water molecules on the surface, with half of the molecules adsorbed, are the most favorable. Finally, ab initio molecular dynamics conducted at 300 K confirms the trends observed at 0 K, albeit the adsorption energies are reduced by about 10 kJ.mol-1. Also, we observe that once put in interaction with a large number of water molecules, half of the calcium atoms at the surface are in close interaction with a water molecule, while the rest of the molecules are further away but present a relatively well defined structure showing similarities with the one of water clusters.


2

Page 3 of 27 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


Introduction Fluorite (CaF2) is the most important source of all the industrial uses of fluorine in the world, as hydrofluoric acid (HF).1,2 HF is synthesized by acid treatment of fluorite, consuming the majority of world fluorite production.3 It serves as a basis for the production of many mineral and organic fluorine-based chemicals (fluoropolymers, fluorocarbon, aluminum fluoride…). Some of them are crucial in the chemical industry, as aluminum fluoride, which is strongly involved in the fabrication of metallic aluminum by electrolysis.1,2,4 Fluorite is also used in the glass and ceramic production, as a flux and a desulphurizing agent in the iron industry and in public health, for drinkable water and toothpastes. Recent applications of fluorite have been studied as its use for epitaxial thin film growth5 or for the fabrication of ionic superconducting materials.6 Some of these applications, as the production of hydrofluoric acid, require high-grade products (> 97 % CaF2, < 1 % SiO2)2 and fluorite needs to be separated from gangue minerals. The technique of froth flotation is widely used,7–11 permitting to process fine-grained and low-grade ores. However, in tungsten or phosphate ores, fluorite does not bring any added value compared to the extracted materials and its grade is too low to exploit it as a by-product. In such cases, fluorite is considered as a gangue mineral and its elimination in the flotation process is mandatory.12–14 Removal of fluorite in flotation permits to enhance the metal grade in the concentrate, allowing to process it by hydrometallurgy with reduced operating costs. This step has become crucial over the past decades: many current tungsten deposits contain fluorite in the gangue, while tungsten has been classified as a Critical Raw Material in the European Union in 2010.15 Nevertheless, elimination of fluorite in the flotation process is difficult.16


3


Page 4 of 27

Overall, the chemical transformations, purifications or elimination of fluorite are performed in aqueous conditions. Hence, it appears crucial to characterize as well as possible the surface hydration of fluorite as it can impact its reactivity. Many authors reported that the most exposed surface of fluorite is the main cleavage plane, the (111) surface.17–20 Moreover, some authors proved that water molecules adsorb in the molecular form onto the (111) surface, with a tilted configuration and with calculated adsorption energies of 41-53 kJ.mol-1.20 A global surface coverage of 50 % seems the most stable hydrated surface condition.20 In this paper, we present density functional theory (DFT) investigations on the interaction of water with the (111) surface of a fluorite crystal. First, the adsorption of a single dissociated water molecule is compared to molecular adsorption. The influence of the surface coverage on the water adsorption is then investigated, towards the building of a water monolayer. Finally, ab initio molecular dynamics calculations are performed on hydrated surfaces, to study the effect of temperature on the hydration layer above the surface. Overall, the main aim of this work is to understand the molecular mechanisms involved in the hydration of the (111) fluorite surface. At the moment, these aspects cannot be tackled directly by experiments but theoretical calculations can provide important insights.

Computational and Structural Details Calculation Settings The total energy and structure of the systems were determined by density functional theory calculations,21,22 using the Vienna ab initio simulation program (VASP).23 The semilocal Perdew−Burke−Ernzerhof (PBE) exchange-correlation functional proposed by Perdew and co-


4

Page 5 of 27 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


workers24, in the generalized gradient approximation (GGA), was employed. The electron-ion interactions were described using the projector augmented wave (PAW) method25,26 and the Kohn-Sham equations22 were solved self-consistently27 until the energy difference between cycles became lower than 10-7 eV. The plane wave cut-off energy was set to 500 eV. A Methfessel Paxton smearing28 of σ = 0.1 eV was applied to occupations, to help the total energy convergence. The structural relaxations have been performed until all forces were smaller than 0.03 eV/Å. All the calculations were realized using the Γ-point only due to the large size of the cell. To describe precisely the interactions involved in the adsorption of molecules, van der Waals (vdW) forces have been taken into account. Since they are not included in the PBE functional, the D229 and the D330 corrections of Grimme were used. The ab initio molecular dynamics calculations were performed using the same computational parameters. A NoséHoover thermostat was used,31–33 with a temperature set to 300 K. The time step was 1 fs and 100 000 steps were realized permitting to reach a total simulation time of 100 ps. For the ab initio molecular dynamics calculations, only the D2 correction method of Grimme29 was used. Structural Model In the present study, a primitive cubic cell of fluorite (CaF2) has been used as the starting point.34 The cell was optimized through full relaxation to obtain lattice parameters a = b = c = 5.462 Å. It is known that the (111) surface is the most exposed surface for fluorite crystals.17–20 Following previous studies, a (111) cleavage surface has been created from the fully relaxed bulk. We have set a supercell containing 144 atoms (48 Ca and 96 F), made of 4 layers of calcium atoms, each of them separated by 2 successive layers of fluorine atoms (Figure 1). Therefore, the (111) surface consists of a plane of 12 seven-coordinate calcium ions, each one being in a hexagonal array (Figure 1). Also, the cell was constructed so that a vacuum of 15 Å was kept between the


5


Page 6 of 27

surface and the upper limit of the cell, following the z axis. It permitted to avoid any unwanted interaction due to the periodicity of the cell. Four layers of bottom atoms in the cell (1 layer of calcium ions, 3 layers of fluorine ions) were frozen to their bulk positions. As ab initio molecular dynamics simulations are numerically costly, in this case, the fluorite slab was reduced to a thickness of 6 layers.

Figure 1. (111) fluorite surface seen from side (left) and from top (right). The blue balls represent the calcium atoms while the grey balls represent the fluorine atoms.

Energy Calculation To understand the interaction of water molecules onto the fluorite surface, three calculations have been performed for each system, giving the following energies: -

EFluo: the total energy of the fluorite slab alone;

-

EX: the total energy of the water molecule alone in vacuum;

-

EFluo-X: the total energy of the fluorite slab with adsorbed molecule(s).


6

Page 7 of 27 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


To establish the surface state of fluorite (111) in presence of water, two types of chemical reactions have been considered: 1. Adsorption: Surface + H2O = Surface-H2O 2. Exchange F/OH: Surface + H2O = SurfaceF->OH + HF The adsorption energy is determined using: ∆ = − −

(1)

The contribution of dispersion energy in the adsorption energy is determined in a similar way: ∆ = − −

(2)

When n water molecules are adsorbed simultaneously, the adsorption energy per molecule is determined using:

∆/ =

∆/ =

×

!" !" × !"

(3)

(4)

On ab initio molecular dynamics simulations, calculations of internal adsorption energies were realized using equations (1) and (3), based on the average energy of each system computed for the last 90 ps of the simulation to exclude the thermalization period.


7


Page 8 of 27

Results and Discussion At the moment, very few authors investigated the adsorption of water onto fluorite surface. DFT studies were conducted by de Leeuw and co-workers using VASP with the PW91 exchangecorrelation functional in the generalized gradient approximation (GGA).20,35 They found that water adsorbs under molecular form even if dissociated water (H+ and HO-) is set as input.20 They reported average adsorption energies of -53.4 kJ.mol-1 per water molecule for 50 % surface coverage, decreasing to -41.4 kJ.mol-1 per water molecule for 100 % surface coverage.20 However, this study didn’t take into account dispersive interactions and didn’t investigate in details the influence of the coverage.20,35 In the present paper, we investigated the adsorption of water molecules onto the (111) fluorite surface with the PBE exchange-correlation functional and including dispersive interactions correction. The global aim is to evaluate the mechanisms and the energy linked to the water adsorption. Fluorite is known36–38 to be a semi-soluble salt, its solubility product, Ksp, being between 5.3 x 10-9 and 7 x 10-11. The quite high Ksp of fluorite could lead to a partial substitution of F- by HO- onto the surface. Therefore, the energetic stability of this phenomenon was also investigated. Isolated Water Molecule on Fluorite a. Adsorption of Dissociated Water For the dissociated case, several configurations have been explored. The D3 correction method was used for all these calculations. A hydroxyl group was placed above a calcium ion on the surface, with the O close to the calcium and the H oriented up, away from the surface. The lone


8

Page 9 of 27 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


H+ ion was set at the vicinity of a fluorine ion. Various distances between the OH and H groups were tested. 1. Short OH-H Distance Firstly, the H+ and the HO- were set on respectively a fluorine atom and a calcium atom which are part of the same octahedron (i.e. Ca-F bond length is 2.37 Å). The H+ ion is captured by the HO- and the water molecule is reformed. De Leeuw and co-workers have reported the same phenomena of recombination of molecular water.20 After relaxation, the O-H1 and O-H2 bonds lengths are both 0.98 Å. The calculated adsorption energy is ∆Eads = -55.6 kJ.mol-1 including ∆Edisp = -13.2 kJ.mol-1. The O atom adsorbs on the calcium atom, the Ca-O bond length being 2.46 Å. The water molecule is in a plane sub-parallel to the surface (Figure 2). One H atom (H1) is oriented towards one of the three surface fluorine atoms being around the Ca2+, the H1-F distance being 1.64 Å. Due to the vicinity of H1, the Ca-F bond is broken, the distance between Ca and F is now 2.92 Å instead of 2.36-2.38 Å without H2O adsorption. This fluorine atom is then bonded to only two Ca atoms, instead of three. The second hydrogen atom (H2) is slightly pointing away from the surface, inducing a tilted configuration of the molecule (Figure 2). These results are in very good accordance with the DFT calculations performed previously, which reported -43 to -51 kJ.mol-1.20 The differences in the adsorption energies and in the H-F distances can be explained by the non-inclusion of dispersive energy correction in the previous DFT calculations and by the use of a different functional (PW91).35


9


Page 10 of 27

Figure 2. Adsorption of dissociated water molecule with H+ and HO- close on the (111) fluorite surface after DFT relaxation, in side view (left) and in top view (right). The water molecule is reformed.

2. Medium OH-H Distance Then, the distance between HO- and H+ was increased. The two ions were not set on atoms of the same octahedron. One fluorine atom was left between H+ and HO-, for an initial OH-H distance of 4.73 Å. The distance was large enough to avoid the reformation of the water molecule. However, the hydrogen ion forms an H-F bond with the unoccupied fluorine ion, and both tilt so that the hydrogen ion becomes closer to the hydroxyl group. This latter moves and replace the unoccupied fluorine ion, forming a hydrogen bond with the hydrogen ion and two Ca-O bonds which lengths are 2.49 Å and 2.47 Å. This configuration is a local energy optimum found by the DFT calculation. The water molecule is not reformed and the calculated adsorption energy is ∆Eads = +37.83 kJ.mol-1. It constitutes a high endothermic value, indicating that the adsorption of dissociated water is disfavored at 0 K.


10

Page 11 of 27 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


3. High OH-H Distance When the proton is placed at 6 Å from the hydroxyl group, there is no interaction between them. The hydrogen ion forms H-F bonds with two surface fluorine atoms, which lengths are 1.21 Å and 1.08 Å (Figure 3). The HO- forms a Ca-O bond with a length of 2.04 Å. This value is small compared to the Ca-O bond length when the molecular water is reformed, which is 2.46 Å. It can be explained by the availability of three lone pairs of electrons for the hydroxyl group oxygen whereas the molecular water oxygen has only two available lone pairs. This configuration is very much disfavored as the calculated adsorption energy is ∆Eads = +247.80 kJ.mol-1.

Figure 3. Adsorption of dissociated water molecule with H+ and HO- far on the (111) fluorite surface after DFT relaxation, in side view and in top view (right).

b. Adsorption of Molecular Water As the molecular adsorption of water appears to be more stable than the dissociative adsorption, we have tested several configurations of adsorbed water where a water molecule was directly set


11


Page 12 of 27

above a surface calcium ion. The D3 correction method was also used for this case. Three different spatial orientations were tested. The H of the water molecules were oriented towards the vacuum (“up” configuration), towards the fluorite crystal (“down” configuration) or within a plane parallel to the surface (“side” configuration). As fluorite is part of cubic system, a 3-fold axe is constituted by the calcium ions alignment, perpendicularly to the (111) surface. Then, the 6 fluorine ions coordinating each calcium ion are composed of 2 different fluorine ions being repeated 3 times by the 3-fold axis. In these two different fluorine ions, one is part of the plane located below the calcium layer and the other is part of the plane located above the calcium layer. Thus, two different cases had to be tested in terms of spatial configuration in the (111) plane, even if all the Ca-F bond lengths are the same. All the different cases tested (“up”, “down”, “side” configurations and two planar orientations for each configuration) lead to the same final geometry. Also, the final configuration of the adsorbed water molecule is the same than after the reformation of the water molecule in the dissociative case: the molecule is adsorbed on a surface calcium, the oxygen atom forming a bond with the surface calcium (Figure 2). The molecule is nearly parallel to the (111) plane, being slightly tilted. The mean Ca-O bond length is 2.47 Å and the hydrogen atoms are oriented toward two fluorine ions of the plane located above the calcium layer (Figure 2). One of the two H is pointing away from the surface. The H-F bond lengths are respectively 1.79 Å and 2.59 Å. The calculated adsorption energy is loosely the same than for the reformed water molecule, being ∆Eads = -55.9 kJ.mol-1. Substitution of F- by HOFinally, the substitution of F- by HO- on the (111) fluorite surface was investigated, still with the D3 correction method. A hydroxyl group was set at the place of a surface fluorine atom. The substitution of F- by HO- on the (111) fluorite surface has been described by Shi and Lang,


12

Page 13 of 27 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


2012.39 They found that the most stable spatial configuration for a substitution of F- by HO- is when the HO- is placed in the first fluorine sublayer, with the H above the O.39 We then studied the following reaction for only one spatial configuration, the more energetically stable one after Shi and Lang, 2012,39 calculating the energy of each component: Surface + H2O = SurfaceF->OH + HF When the hydroxyl group is introduced in the fluorite structure, it forms Ca-O bonds with the three surrounding calcium ions (Figure 4). These calcium atoms, that coordinate the anion, are distending and the Ca-O bond lengths are 2.46 Å, while the usual Ca-F bond lengths are 2.38 Å. This difference can be explained by the difference of electronegativity between fluorine and oxygen atoms. The calculated reaction energy is ∆Er = +71.1 kJ.mol-1, including ∆Edisp = 8.4 kJ.mol-1. It indicates that the substitution of F- by HO- on the surface is not thermodynamically favored at 0 K. However, this result is crucial to understand the influence of the pH on the surface condition of fluorite. A high concentration of HO- in the solution will probably make possible the substitution of F- by HO- in the first layer. Indeed, the free energy of this reaction depends on the H2O/HF concentration ratio. The solubility of fluorite is constant from pH 5 to pH 11, i.e. when the H+ and HO- concentrations are still low.40 At pH < 5, HF will be formed in solution as F- will be substituted by a solution anion onto the surface (SO42- if sulfuric acid is used for example) and the Ca2+ cation substituted by several protons.40,41 Also, in alkaline medium (pH > 11), the hydroxyl group, in high concentration, will substitute more easily the surface fluorine ions, leading to a progressive destabilization of the calcium cations and their solvation.40,41 Dissolution of fluorite has been intensively investigated by researchers who showed that the defects percentage on the surface is, with the pH, one of the main parameter involved in the dissolution rate.42–45 Our study, performed on a perfect (111) cleavage plane and


13


Page 14 of 27

with pure molecular water permits to conclude that the dissolution of fluorite, through the substitution of a fluorine ion by a hydroxyl group or the substitution of calcium ions by protons, needs an activation energy (brought by temperature) and/or high concentrations of pHdetermining ions.

Figure 4. Side view (left) and top view (right) of the (111) fluorite surface after DFT relaxation of the system where a fluorine atom has been replaced by a hydroxyl group. The hydroxyl establishes three bonds with surrounding calcium atoms.

Towards a Full Hydration of the Surface Once the most favorable configuration for a single molecule of water determined, the progressive hydration of the surface was investigated. Then, 2, 3, 6 and 12 H2O were placed on the surface composed of 12 calcium atoms. For each number of molecules, three different spatial configurations were tested:


14

Page 15 of 27 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


1. A “surface” layout, where each water molecule is set on a calcium atom with dCaO

~ 2.5 Å (Figure 5a).

2. A “partial clusters” layout, where the total number of water molecules is divided between an adsorbed layer (dCa-O ~ 2.5 Å) and a layer further from the surface, involved in hydrogen bonds with the first layer (Figure 5b). 3. A “full clusters” layout, where the geometry of the water molecules is set up following the cluster spatial organization described by Temelso and co-workers46 (Figure 5c).


15


Page 16 of 27

Figure 5. Side views (left) and top views (right) of the (111) fluorite surface showing the different spatial configurations investigated when the water coverage is increased, for instance 100 % coverage here. “Surface” (top), “full clusters” (middle) and “partial clusters” (down) layouts were tested. Dashed lines represent H-bonds.

For each case, the water molecules can be set following different geometries, and we present only the most stable configuration and the corresponding adsorption energy per water molecule (see Figure 6). Also, the PBE functional is compared with the D2 and D3 correction methods for each case and each number of water molecules.


16

Page 17 of 27

Number of water molecules 0

1

2

3

4

5

6

7

8

9

10

11

12

-30

-30

∆Eads (kJ.mol-1)

-35

Surface layout

PBE

Partial clusters layout

PBE-D2

Full clusters layout

PBE-D3

-35

-40

-40

-45

-45

-50

-50

-55

-55

-60

-60

-65

-65

-70

-70

Figure 6. Adsorption energies per water molecule for the different spatial configurations tested when the water coverage is increased from 1 water molecule to 12 water molecules. D2 and D3 correction methods were tested for each case, as well as no correction method.

For the “surface” layout, the most stable case is, for each coverage, the configuration where water molecules are close of each other. The hydration energy per water molecule becomes less negative when the number of water molecules increases to 12. Hence, the configuration with 6 molecules, i.e. a 50 % coverage of the surface, is the most stable, which is in accordance with what previous authors have shown.20 However, the variation between all the coverages is limited: for instance, with the PBE functional, the calculated adsorption energies is ranging between -40 and -45 kJ.mol-1 while for the corrected functionals it varies between -55 and 63 kJ.mol-1. The “full clusters” layout is the less favoured case since a small number of water molecules is adsorbed directly on the surface, while the hydrogen bonds between water molecules do not


17

∆Eads (kJ.mol-1)

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 18 of 27

compensate. While increasing the number of water molecules, the adsorption energy per molecule goes from a minimum of -37 to a maximum of -45 kJ.mol-1 with the PBE functional. However, using the D2 and D3 corrections, it ranges between -52 to -59 kJ.mol-1. This corresponds to a progressive stabilization of the system for an increasing number of molecules. Nevertheless, the “partial clusters” layout is the most stable configuration. The adsorption energies per water molecule decreases systematically when the number of molecules is increased. The calculated adsorption energies are going from -61 kJ.mol-1 for 2 molecules to 64 kJ.mol-1 for 12 molecules, including a contribution of -12 to -14 kJ.mol-1 of dispersive energy. It shows a stabilization of the adsorbed molecules involved in some water clusters. Overall, the energies obtained with the PBE functional are, in absolute values, significantly lower than the ones obtained with the corrected methods, which highlights their importance. Moreover, the energy ordering between the various layouts is sometimes different (see for instance the energy for 12 molecules in “full clusters” and “surface” layouts). Also, the two dispersion correction methods lead to very similar results.


18

Page 19 of 27 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


Ab Initio Molecular Dynamics Calculations of the Hydrated (111) Fluorite Surface To complete the results obtained at T = 0 K, the hydration of the (111) fluorite surface was studied using ab initio molecular dynamics simulations at T = 300 K using PBE+D2 correction method. Molecular dynamics calculations were performed on the surface with 1, 2, 3, 6 and 12 water molecules, as before. The results show that the temperature does not change the hydration mechanisms determined at T = 0 K (Figure 7): the water molecules form spontaneously the “partial clusters” layout predicted at T = 0 K. Indeed, half of the water molecules are adsorbed on calcium atoms with dCa-O = 2.4-2.6 Å. The other half is a bit further from the surface with dSurface-Molecule = 3-3.5 Å. All the molecules are involved in clusters with 4, 5 or 6 water molecules in each cluster, where the stabilization of the structure is due to H-bonds, with bond lengths ranging between 1.4 and 2.5 Å. For each number of water molecules, the computed internal energy of adsorption at T = 300 K follows the same trend than the adsorption energy at T = 0 K, albeit with a shift of roughly -10 kJ.mol-1 due to the temperature (Figure 7).


19


Number of water molecules 0

0

1

2

3

4

5

6

7

8

∆Eads (kJ.mol -1)

-10

9

10

11

0K 300 K

-20

12

0 -10 -20

-30

-30

-40

-40

-50

-50

-60

-60

-70

-70

Figure 7. Internal adsorption energies calculated from the AIMD calculations for different water coverages, confirming the results obtained at 0 K. Then, configurations with 24 molecules, 36 molecules, and the cell filled with water molecules have been simulated (Figure 8). For this last case, the number of water molecules was adapted to obtain a calculated water density close to 1 g.cm-3. For all the configurations investigated, we have found that only 50 % of the surface calcium ions are occupied by an adsorbed water molecule, with a distance of 2.46 Å (Figure 8). The other molecules are further from the surface with distances ranging 2.80 Å to more than 3 Å. Moreover, the presence of multiple water layers does not affect the specific geometric configuration of the first layer identified at T = 0 K previously.


20

∆Eads (kJ.mol -1)

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 20 of 27

Page 21 of 27 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


Figure 8. Snapshots of side views with 90° between each other of the (111) fluorite surface with the vacuum above the surface completely filled with water molecules so that the density is 1 g.cm-3. Dashed lines represent H-bonds. Lone atoms and repeated boundaries are due to the periodicity of the cell. Note that only 6 water molecules are adsorbed on the surface, representing only half of the calcium atoms occupied by water molecules.

Water molecules that are close to the surface are often part of non-planar pentamers from which one or two molecules are adsorbed onto a surface calcium. For the bulk of water, the geometries are structured in a way that the two H of two molecules point towards an oxygen of another molecule (Figure 8). The H of the two other molecules are establishing hydrogen bonds with other multi-molecular structures. This result is in accordance with water clusters that have been


21


Page 22 of 27

intensively described before47–51 : clusters involving from 2 to 100 water molecules are well known in the literature.48,52–55 The average H-bond length was calculated on the 96 water molecules set on the (111) fluorite surface. The mean H-bond length is 1.837 Å, all the H-bonds lengths ranging between 1.44 Å and 2.46 Å. Also, 100 % of the oxygen atoms in the bulk establish one H-bond with another water molecule, and 90 % of them establish two H-bonds with other water molecules.

Conclusions In this publication, the interplay between the (111) surface of fluorite and water molecules was investigated using density functional theory calculations including a correction for van der Waals interactions at 0 K and at 300 K. The studies showed that water adsorbs in its molecular form on a calcium atom with ∆Eads ≈ -55 kJ.mol-1 including ∆Edisp ≈ -13 kJ.mol-1 and dCa-O = 2.46 Å, as HO- and H+ set on the surface reforms the water molecule during the calculation. Also the substitution of F- by HO- on the surface was investigated and was found to be not energetically favored. Then, the interaction of the surface with several water molecules was studied. Different configurations were tested and the most favorable is the one where half of the surface calcium atoms are occupied by an adsorbed water molecule. Then, molecular dynamics calculations at T = 300 K confirmed the results obtained at T = 0 K concerning the structure of the interface, although the interaction energy is reduced due to the temperature. The results presented in this work contribute to a better understanding of the properties of fluorite and will motivate further theoretical and experimental works, in particular for future adsorption studies in aqueous phase. For instance, we plan to study the adsorption of organic


22

Page 23 of 27 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


molecules onto different mineral surfaces, the comprehension of the corresponding mechanisms being crucial for hydrometallurgy and flotation.

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Notes There are no conflict of interest to declare.

Acknowledgements This work was granted access to the HPC resources of TGCC under the allocation 2017A0030910306 made by GENCI. The research leading to these results has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 641650 for the FAME project.


23


Page 24 of 27

References (1) (2) (3) (4) (5) (6) (7) (8)

(9) (10)

(11) (12)

(13) (14) (15)

(16)

(17) (18) (19)

Emsley, J. Nature’s Building Blocks: An A-Z Guide to the Elements, New ed., completely rev. and updated.; Oxford University Press: Oxford ; New York, 2011. Féraud, J. Mémento Des Roches et Minéraux Industriels : La Fluorine Ou Spath Fluor (in French). Rap. BRGM R 40825, 102 p., 2 fig., 10 tabl., 2 ann., 1 carte h.t. 1999. Free, M. L.; Miller, J. D. Kinetics of 18-Carbon Carboxylate Adsorption at the Fluorite Surface. Langmuir 1997, 13 (16), 4377–4382. US Patent 4,865,701. Klumpp, S.; Dabringhaus, H. Experimental Study of the Adsorption of Lithium Fluoride on the (111) Surface of CaF2. Surf. Sci. 1998, 417 (2–3), 323–336. Patro, L. N.; Hariharan, K. Fast Fluoride Ion Conducting Materials in Solid State Ionics: An Overview. Solid State Ion. 2013, 239, 41–49. Mielczarski, J. A.; Mielczarski, E.; Cases, J. M. Dynamics of Fluorite−Oleate Interactions. Langmuir 1999, 15 (2), 500–508. Mielczarski, E.; Mielczarski, J. A.; Cases, J. M.; Rai, B.; others. Influence of Solution Conditions and Mineral Surface Structure on the Formation of Oleate Adsorption Layers on Fluorite. Colloids Surf. Physicochem. Eng. Asp. 2002, 205 (1), 73–84. Zhang, Y.; Song, S. Beneficiation of Fluorite by Flotation in a New Chemical Scheme. Miner. Eng. 2003, 16 (7), 597–600. Song, S.; Lopez-Valdivieso, A.; Martinez-Martinez, C.; Torres-Armenta, R. Improving Fluorite Flotation from Ores by Dispersion Processing. Miner. Eng. 2006, 19 (9), 912– 917. LIN, D.; NIE, G.; LUO, G.; TANG, Z. Collect Mechanisms of Oleic Acid on Fluorite and Calcite Minerals. 2016. Liu, L.; Xue, J.; Zhu, J. Removing Fluorite and Calcite from Scheelite During Floatation Separation Process with Calcium-and Sodium-Containing Reagents. In EPD Congress 2014; John Wiley & Sons, Inc., 2014; pp 431–439. Chen, W.; Feng, Q.; Zhang, G.; Yang, Q.; Zhang, C. The Effect of Sodium Alginate on the Flotation Separation of Scheelite from Calcite and Fluorite. Miner. Eng. 2017, 113, 1–7. Zhang, Y.; Li, Y.; Chen, R.; Wang, Y.; Deng, J.; Luo, X. Flotation Separation of Scheelite from Fluorite Using Sodium Polyacrylate as Inhibitor. Minerals 2017, 7 (6), 102. European Commission. Critical Raw Materials for the EU: Report of the Ad-Hoc Working Group on Defining Critical Raw Materials. http://ec.europa.eu/enterprise/policies/rawmaterials/documents/index_en.htm. 2010. Filippova, I. V.; Filippov, L. O.; Duverger, A.; Severov, V. V. Synergetic Effect of a Mixture of Anionic and Nonionic Reagents: Ca Mineral Contrast Separation by Flotation at Neutral PH. Miner. Eng. 2014, 66–68, 135–144. Parks, T. C.; Barker, W. W. The Ordered Dispersal of Point Defects over Cubic Lattices: Application to Fluorite-Related Structures. J. Solid State Chem. 1977, 20 (4), 397–407. Tasker, P. W. The Structure and Properties of Fluorite Crystal Surfaces. J. Phys. Colloq. 1980, 41 (C6), C6–488. Bennewitz, R.; Reichling, M.; Matthias, E. Force Microscopy of Cleaved and ElectronIrradiated CaF2(111) Surfaces in Ultra-High Vacuum. Surf. Sci. 1997, 387 (1–3), 69–77.


24

Page 25 of 27 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


(20) De Leeuw, N. H.; Purton, J. a A.; Parker, S. C.; Watson, G. W.; Kresse, G. Density Functional Theory Calculations of Adsorption of Water at Calcium Oxide and Calcium Fluoride Surfaces. Surf. Sci. 2000, 452 (1), 9–19. (21) Hohenberg, P.; Kohn, W. Inhomogeneous Electron Gas. Phys. Rev. 1964, 136 (3B), B864–B871. (22) Kohn, W.; Sham, L. J. Self-Consistent Equations Including Exchange and Correlation Effects. Phys. Rev. 1965, 140 (4A), A1133–A1138. (23) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B 1993, 47 (1), 558–561. (24) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77 (18), 3865–3868. (25) Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50 (24), 17953– 17979. (26) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B 1999, 59 (3), 1758–1775. (27) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy Calculations Using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54 (16), 11169–11186. (28) Methfessel, M.; Paxton, A. T. High-Precision Sampling for Brillouin-Zone Integration in Metals. Phys. Rev. B 1989, 40 (6), 3616–3621. (29) Grimme, S. Semiempirical GGA-Type Density Functional Constructed with a LongRange Dispersion Correction. J. Comput. Chem. 2006, 27 (15), 1787–1799. (30) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu. J. Chem. Phys. 2010, 132 (15), 154104. (31) Nosé, S. A Molecular Dynamics Method for Simulations in the Canonical Ensemble. Mol. Phys. 1984, 52 (2), 255–268. (32) Nosé, S. A Unified Formulation of the Constant Temperature Molecular Dynamics Methods. J. Chem. Phys. 1984, 81 (1), 511–519. (33) Hoover, W. G. Canonical Dynamics: Equilibrium Phase-Space Distributions. Phys. Rev. A 1985, 31 (3), 1695–1697. (34) Cheetham, A. K.; Fender, B. E. F.; Cooper, M. J. Defect Structure of Calcium Fluoride Containing Excess Anions: I. Bragg Scattering. Journal of Physics C. 1971. (35) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C. Atoms, Molecules, Solids, and Surfaces: Applications of the Generalized Gradient Approximation for Exchange and Correlation. Phys. Rev. B 1992, 46 (11), 6671– 6687. (36) Marinakis, K. I.; Shergold, H. L. The Mechanism of Fatty Acid Adsorption in the Presence of Fluorite, Calcite and Barite. Int. J. Miner. Process. 1985, 14 (3), 161–176. (37) Ball, J. W.; Nordstrom, D. K. User’s Manual for WATEQ4F, with Revised Thermodynamic Data Base and Test Cases for Calculating Speciation of Major, Trace, and Redox Elements in Natural Waters. 1991. (38) Lange’s Handbook of Chemistry, 15. ed.; Dean, J. A., Lange, N. A., Eds.; McGraw-Hill handbooks; McGraw-Hill: New York, NY, 1999. (39) Shi, H.; Chang, L.; Jia, R.; Eglitis, R. I. Ab Initio Calculations of Hydroxyl Impurities in CaF 2. J. Phys. Chem. C 2012, 116 (10), 6392–6400.


25


Page 26 of 27

(40) Roche, M. Caractérisation de l’état de Surface Des Fluorines et Application à La Flottation Des Minerais de Fluorine Du District Du Tarn (Montroc - Le Burc), Université de Nancy I, 1973. (41) Zhang, R.; Hu, S.; Zhang, X. Experimental Study of Dissolution Rates of Fluorite in HCl– H2O Solutions. Aquat. Geochem. 2006, 12 (2), 123–159. (42) Godinho, J. R. A.; Piazolo, S.; Evins, L. Z. Effect of Surface Orientation on Dissolution Rates and Topography of CaF2. Geochim. Cosmochim. Acta 2012, 86, 392–403. (43) Maldonado, P.; Godinho, J. R. A.; Evins, L. Z.; Oppeneer, P. M. Ab Initio Prediction of Surface Stability of Fluorite Materials and Experimental Verification. J. Phys. Chem. C 2013, 117 (13), 6639–6650. (44) Godinho, J. R. A.; Putnis, C. V.; Piazolo, S. Direct Observations of the Dissolution of Fluorite Surfaces with Different Orientations. Cryst. Growth Des. 2014, 14 (1), 69–77. (45) Godinho, J. R. A.; Piazolo, S.; Balic-Zunic, T. Importance of Surface Structure on Dissolution of Fluorite: Implications for Surface Dynamics and Dissolution Rates. Geochim. Cosmochim. Acta 2014, 126, 398–410. (46) Temelso, B.; Archer, K. A.; Shields, G. C. Benchmark Structures and Binding Energies of Small Water Clusters with Anharmonicity Corrections. J. Phys. Chem. A 2011, 115 (43), 12034–12046. (47) Liu, K.; Cruzan, J. D.; Saykally, R. J. Water Clusters. Science 1996, 271 (5251), 929–933. (48) Ludwig, R. Water: From Clusters to the Bulk. Angew. Chem. Int. Ed. 2001, 40 (10), 1808– 1827. (49) Roy, R.; Tiller, W. A.; Bell, I.; Hoover, M. R. The Structure Of Liquid Water; Novel Insights From Materials Research; Potential Relevance To Homeopathy. Mater. Res. Innov. 2005, 9 (4), 98–103. (50) Perera, A. On the Microscopic Structure of Liquid Water. Mol. Phys. 2011, 109 (20), 2433–2441. (51) Skinner, L. B.; Benmore, C. J.; Neuefeind, J. C.; Parise, J. B. The Structure of Water around the Compressibility Minimum. J. Chem. Phys. 2014, 141 (21), 214507. (52) Maheshwary, S.; Patel, N.; Sathyamurthy, N.; Kulkarni, A. D.; Gadre, S. R. Structure and Stability of Water Clusters (H 2 O) n , n = 8−20: An Ab Initio Investigation. J. Phys. Chem. A 2001, 105 (46), 10525–10537. (53) Møgelhøj, A.; Kelkkanen, A. K.; Wikfeldt, K. T.; Schiøtz, J.; Mortensen, J. J.; Pettersson, L. G.; Lundqvist, B. I.; Jacobsen, K. W.; Nilsson, A.; Nørskov, J. K. Ab Initio van Der Waals Interactions in Simulations of Water Alter Structure from Mainly Tetrahedral to High-Density-Like. J. Phys. Chem. B 2011, 115 (48), 14149–14160. (54) Anacker, T.; Friedrich, J. New Accurate Benchmark Energies for Large Water Clusters: DFT Is Better than Expected. J. Comput. Chem. 2014, 35 (8), 634–643. (55) Gillan, M. J.; Alfè, D.; Michaelides, A. Perspective: How Good Is DFT for Water? J. Chem. Phys. 2016, 144 (13), 130901.


26

Page 27 of 27 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


Table of Contents

The hydration of the (111) surface of fluorite has been investigated by ab initio molecular dynamics and dispersion-corrected DFT. Dimensions voulues: 8.5 cm x 4.75 cm


27

Surface Properties of Fluorite in Presence of Water: An Atomistic

Recommend Documents