Reactivity of Silver Iodide (β-AgI) Surfaces: A DFT Study - The Journal

2 days ago - Support. Get Help · For Advertisers · Institutional Sales; Live Chat. Partners. Atypon · CHORUS · COPE · COUNTER · CrossRef · CrossCheck ...
0 downloads 0 Views 801KB Size
Subscriber access provided by University of Winnipeg Library

C: Surfaces, Interfaces, Porous Materials, and Catalysis

Reactivity of Silver Iodide (#-AgI) Surfaces: A DFT Study Houssam Hijazi, Laurent Cantrel, and Jean-Francois Paul J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b06931 • Publication Date (Web): 26 Oct 2018 Downloaded from http://pubs.acs.org on October 29, 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

Reactivity of Silver Iodide (-AgI) surfaces: A DFT Study Houssam Hijazia, b, Laurent Cantrelb, Jean-François Paula* (a)Univ. Lille, CNRS, ENSCL, Central Lille, Univ. Artois, UMR 8181 – UCCS, Unité de Catalyse et Chimie du Solide, F-59000 Lille, France (b)Institut de Radioprotection et de Sûreté Nucléaire (IRSN), PSN-RES, 13115 Saint-Paul-Lez-Durance Cedex, France Email [email protected]

Abstract

AgI aerosols may be produced in severe nuclear accident and are typically used to favour the formation of droplet inducing rain. To elucidate their behaviour, we study at the DFT (PBE+D) level -AgI surface stability and reactivity. The most stable surfaces are the (110), (100) and (120) ones. These three surfaces are non-polar. On the opposite, the polar (001) surface is less stable. The water adsorption on these stable surfaces is molecular. The adsorption energy is small (0.30 eV) but increases with the surface coverage due to the formation of hydrogen bonds network between the adsorbed molecules. These surfaces are not very reactive and the formation of volatile compounds (I2 or IO°) is only possible in presence of strong oxidants such as OH° that may be produced by water radiolysis.

ACS Paragon Plus Environment

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

1. Introduction Silver iodide (AgI) is an inorganic compound, which is widely used in cloud seeding1 and in manufacturing photosensitive materials such as photocatalyst 2,3. Beside these applications, AgI is one of the possible iodide aerosols produced after a severe accident

4

in a nuclear power plant and may contribute to

radioactive contamination. The most stable iodide aerosol in severe accident condition is composed of CsI

5

which is very soluble and hygroscopic. Unlike the CsI compounds, AgI is almost insoluble. On the opposite to CsI in which the Cs+ cation is not reducible, the Ag+ ion may react as an oxidant and may induce a different behavior leading to favor formation of volatile iodine species such as I2. Understanding the chemical evolution of AgI aerosol is suitable to propose an accurate evaluation of the radioactive contamination. A large number of theoretical and experimental works were carried out to investigate the electronic structure and properties of silver halides6–13. Gordienko et al.6 performed theoretical study, using linear response method, on lattice dynamics of three silver halides including AgI. Two phase of AgI have been studied: the γ- (a zinc blende structure11) and β-phases (wurtzite structure). The most stable phase under atmospheric pressure and for temperature below 420 K is the β-phase. In order to describe the chemical process that may occur at low temperature, we have considered only this phase in our study. The interactions of water molecules with AgI surfaces have been studied experimentally and theoretically. Experimental works comparing the adsorption of water vapor on AgI prepared either by direct reaction (between metallic silver and iodine) or by precipitation (starting by AgNO3 and NH4I) at 30°C are detailed in ref 14. The results showed that a large amount of water molecules are adsorbed on AgI prepared by precipitation containing hygroscopic impurities. However, only a small amount of water is adsorbed on the AgI prepared by direct reaction between Ag and I2, without any hygroscopic impurities. The latter demonstrates the importance of

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

AgI purity on its properties as well as low chemical affinity of H2O with pure AgI. This first result was confirmed by Corrin and Neslon.14 From the theoretical point of view, most of the studies are based on classical molecular dynamics. Shevkunov13 studied the behavior of steam on flat surfaces split in β-AgI crystal. On one side, (001) surface model contains Ag+ cations while the other side contains I- anions. Interaction of water molecules with the surfaces differs between the two terminations. On the former, the cohesion between water molecules and crystal surface takes place by interaction between hydrogen atoms and iodine anions located in the second layer. Whereas in the latter one, it takes place by interaction between oxygen atoms and silver cations in the second layer as well. The study demonstrates the role of the electrostatic interaction between the water molecule and the surfaces. Taylor et al.15 explored the nucleation process on AgI surfaces. The authors concluded that the interaction depends on the surface orientations. In addition, on the basal (001) plane, two layers of liquid water may be adsorbed even at temperatures below 0°C. Glatz et al.16 studied the effect of the iodide location on and underneath the AgI surface by classical molecular dynamics. They observed that water molecules can form a thin film (two layers) on the surface. To the best of our knowledge, only Hiemstra17 studied the water/surface interactions at the DFT level. However, this study used only small clusters to model the surface and to estimate the surface charge distribution in presence of water. These charges were used in a classical model in a second step.

In order to improve the knowledge of the -AgI surface chemistry, we have computed at DFT level, surface energies to determine the most stable surface. In a second step, we have studied water adsorption on this stable surface and AgI surface oxidation to estimate the possible iodine revolatilisation.

2. Theoretical methods

ACS Paragon Plus Environment

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 4 of 27

2.1 Computation parameters All DFT calculations were carried out with VASP program (Vienna Ab initio Simulation Package)18,19. The wave function is constructed on a set of plane waves and the electron-ion interactions were described using the Projector Augmented Wave method (PAW )20. The energy is calculated using generalized gradient approximation as parameterized by Perdew et al.,21 and PBE functional was employed with a 0.1 eV Gaussian 22 smearing. PBE has shown to give good results on ionic systems if the Van der Walls corrections are taken into account23. In this study, as the iodine oxidation state change during reactions, the Van der Walls corrections are calculated according to the iterative Tkatchenko Scheffler model24 that is well adapted to ionic systems25,26. The self-consistent solving of the Kohn−Sham equations is performed until the energy difference between two successive electronic steps is lower than 10−5 eV. The atomic positions were optimized until the forces become less than 0.03 eV/Å. The electron configurations [Kr] 4d10 5s1, [Kr] 4d10 5s2 5p5, [He] 2s2 2p4 and 1s1 were used for silver, iodine, oxygen and hydrogen respectively. Test calculations were done by varying cut-off energy from 200 to 600 eV and the k-points mesh from 1×1×1 to 11×11×11 following Monkhorst-Pack27 schemes to define calculation parameters. From these evaluations, all the bulk and surface calculations were performed with a 450 eV cut-off energy, using a 5×5×5 k-point mesh for bulk calculations whereas 5×5×1 was used for all surface calculations. Using these parameters, the accuracy of the DFT calculation is expected to be better than 0.1 eV for the calculated reaction energy. Dipolar corrections in the z direction have not been used in this study. Test calculation show that their effects are very small on non-polar surfaces (below 2 mJ.m-2) as well as on polar ones (below 10 mJ.m-2) and do not affect the conclusion of the study. -AgI crystal has a hexagonal structure which belongs to the P63mc space group, with four atoms in the unit cell (see Figure 1). The volume of the cell is 137.2 Å3 (cell size is 4.59 x 4.59 x7.52 Å3 while α, β and γ are 90°, 90° and 120° respectively).

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Figure 1: Unit cell of -AgI crystal (Ag and I are respectively colored in bright blue and purple, the same color will be used in the article). A (1x1) supercell composed of eight AgI layers was used in surface energy calculations. The four outermost layers allow taking into account the surface formation, while the other layers were kept fixed to mimic bulk constraints. We added 15 Å of vacuum between two consecutive slabs to avoid interactions. Larger cells (2x1) are used for water adsorption calculations. The computed surfaces energies are equal in the two cells. The smaller unit vector used for water adsorption is 3.98 Å long.

The following non-equivalent low index surfaces: (110), (120), (100) and (001) were optimized. Surface energies for symmetric surfaces were calculated as

γ= (Eslab – nEbulk)/2A (1)

where

is the slab energy, nEbulk denotes the energy of n unit cells of AgI (n represents the number of

unit cells in the given slab), and A is the area of the considered surface unit cell. The relaxation energy was added to the average surface energy to include surface relaxation in the calculations when only one side of the slab is relaxed. The adsorption energy is defined by the following formula (2):

ACS Paragon Plus Environment

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 6 of 27

Eads= EAgI + n*E(H2O)gas –E(AgI+ H2O) (2)

Where EAgI, E(H2O)gas, E (AgI+

H2O)

are, respectively, the energies of the relaxed AgI substrate, the energy of

one water molecule in gas phase and the energy of the total system. ‘n’ is the number of water molecules adsorbed on the surface per cell. Using this definition, a positive adsorption energy means it is an exothermic adsorption reaction.

2.2 Thermodynamics model Statistical thermodynamic models were used to describe water adsorption on the AgI surfaces at defined pressure and temperature. Gas phase presents the reservoir in equilibrium with the surface and the adsorbed molecules. We defined afterwards the Gibbs free energy of the adsorption reaction (reaction (R1)) (∆rG) as a function of temperature, pressure and the gas phase chemical potential using the following expression (3):

Surface+ n Gas  Surface-(Gas)n (R1) ∆rG = [∆E0 + ∆Ezpe - n∑∆µ(T,p)]

(3)

Where ∆E0 is the adsorption energy obtained from DFT energies, ∑∆µ(T,p) is the chemical potential difference between the gas phase and adsorbed molecules, while ∆Ezpe is the zero point energy difference. The chemical potential depends on temperature and pressure dependent to of the water molecule, it can be expressed as follows (4):

∆µ(T, p) = ∆µ0(T) + RT ln [P/P0] (4)

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Where ∆µ0(T) = ∆[Evib + Erot + Etrans ] + RT –T(Svib + Srot + Strans). This expression includes thermal contributions of the change in vibrational, rotational, and translational degrees of freedom28,29 and is evaluated using perfect gas partition function. The harmonic vibrations and rigid rotator formula have been used to compute the partition functions. The effect of these approximations on the thermodynamic corrections is small 30.

2.3 Transition state calculations Using Climbing Image Nudged Elastic Band (CI-NEB)31,32, we determine the minimum energy path (MEP) between two minima i.e. a reactant and a product. In our case, 8 images, located between the reactant and the product of elementary reactions defined the reaction path. The highest point on the reaction path is optimized to the transition state (TS). After optimization the transition state is characterized by one and only one imaginary frequency.

3. Results and discussion In order to study the surface reactivity of AgI particles, we determine the most stable surfaces that will be exposed to the gas phase and we then define the shapes of these particles based on these surface energies. The water adsorption will be considered in the second part while the reactivity will be discussed in the third one. 3.1 Exposed Surface We have showed that the most stable low index planes for AgI are (100), (110), (120) and (001), see Table 1. Surface energy of the exposed surfaces . The first three surfaces are non-polar and end up with the two types of ions Ag+ and I-. On the opposite, the (001) surfaces terminate either by only Ag+ cation or I- anion

ACS Paragon Plus Environment

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 8 of 27

forming polar surfaces. The latter are expected to be less stable unless large reconstruction are performed or surfactant added33. After relaxation, we have noticed large reconstruction of the three non-polar surfaces. The ions tend to form new bonds with surface and subsurface atoms to compensate the formation of dangling bonds. However, the polar (001) surface does not relax and its surface energy remains high compared to the other ones. The high atomic density of the silver cation plane prevents the insertion of the iodine anion into the cation plane to cancel the surface dipole. Table 1. Surface energy of the exposed surfaces (110)

(120)

(100)

(001)

300

307

286

416

Relaxed γ(mJ.m )

109

119

114

413

Relaxation Percentage (%)

64

61

60

1

-2

Rigid γ(mJ.m ) -2

Figure 2 shows side views of the mentioned surfaces before and after relaxation. All surfaces were modeled by orthorhombic cell except for (001) surface which was studied in a hexagonal (γ=120°) cell. The non-polar (110), (100) and the polar (001) surfaces contain four atoms (two silver and two iodine atoms) while the (120) has six atoms on the surface. Plane Rigid (110)

Relaxed

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

(120)

(100)

(001)

Figure 2: Side view of AgI low index surfaces. (Ag and I are respectively colored in bright blue and purple). The computed surface energies are sumerised in Table 1. Surface energy of the exposed surfaces. Surface energies of the relaxed planes range in the following order: (110) < (100) < (120) < (001), (110), (100) and (120) are the most sfigure surfaces of -AgI. These surfaces show high relaxation energy percentage (~ 60%) which is due to the formation of stronger bond between the surface atoms. Based on these surface energies, one can deternime the crystal shape in vacuum using Wulff formalism.

ACS Paragon Plus Environment

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 10 of 27

Figure 3 Shape of AgI particle in vacuum based on Wullf construction. (110, 100 and 001 are respectively colored in yellow, blue and violet) Figure 3 Shape of AgI particle in vacuum based on Wullf constructionepresents the shape of the particle according to Wullf model and, (110) and (100) plans are the main ones. 3.2 Water adsorption To study the surface reactivity, we start by adsorbing one water molecule on the surface. Molecular and dissociative adsorption modes were tested on the three stable surfaces of AgI. For each adsorption mode, various geometries were tested to get the most favored one. In the second step, we add the water molecules successively until we reach one monolayer of water on top of the surfaces. The monolayer will be achieved when the number of adsorbed water molecule is equal to the number of silver cation on the first layer. We present in this part the results for one water molecule and half and full monolayer.

3.2.1 MOLECULAR ADSORPTION OF ONE WATER MOLECULE ON AGI SURFACES

Starting by molecular adsorption of one water molecule on the surface, we have tested different interaction types. In the first one, hydrogen atom of the water molecule interacts with iodide ion (H-I) on the surface. We have studied in a second step the interaction between oxygen atom of the water molecule and one silver ion (O-Ag bond) from the surface. For the O-Ag interaction, the water molecule is parallel to the surface with an O-Ag distance of about 2.6 Å, the O atom is on top position of Ag+ ion. Geometrical parameters of optimized water molecule are very similar to the gas phase ones (H-O-H angle about 105° and H-O bond 0,98 Å). Under the effect of water adsorption, no major relaxation in the upper layer of AgI crystal were noticed in the H-I interaction case. Whereas in the other case, interacting Ag+ ion has moved up around 0.5 Å

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

above the three surfaces. Adsorption energy for the H-I interaction is about 0.1 eV for the three surfaces (Figure 4: Geometries and energies for one water molecular adsorption on stable AgI surfaces (Ag, I, H and O are respectively colored in bright blue, purple, red and white). However, the adsorption energies are higher than 0.2 eV for the second orientation. This indicates that the main interaction is O-Ag+ one.

Surface direction

(110)

(100)

(120)

0.10

0.12

0.10

0.32

0.24

0.20

H-I interaction (in eV)

O-Ag interaction (in eV) Figure 4: Geometries and energies for one water molecular adsorption on stable AgI surfaces (Ag, I, H and O are respectively colored in bright blue, purple, red and white). In conclusion, the adsorption is more exothermic on the (110) surface than on the (100) one. This first surface is expected to be stabilized in water which explains the hexagonal shape of the crystals synthesized in aqueous solutions.17,33

3.2.2

DISSOCIATIVE ADSORPTION OF ONE WATER MOLECULE ON AGI SURFACES

ACS Paragon Plus Environment

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 12 of 27

Dissociative adsorption has also been studied for the three surfaces. Before optimization, one proton was placed about 1.6 Å on top of I while the OH group was set on top of Ag at distance around 2.6 Å. After optimization, the calculated adsorption energy is about -3.0 eV, see Figure 5. Test calculations using larger cells (4x2 ones) prove that these values do not depend on the size of the cell. These energies are similar on all surfaces and correspond to a very endothermic reaction that will not be favored thermodynamically. The dissociation of water on AgI stable is not expected in agreement with the very small basicity of the iodine anions.

Surface

(110)

(100)

(120)

-3.21

-2.79

-2.83

direction Energy of dissociative adsorption (in eV)

Figure 5. Geometries and energies for one water dissociative adsorption on stable AgI surfaces (Ag, I, H and O are respectively colored in bright blue, purple, red and white). 3.2.3 TEMPERATURE AND PRESSURE EFFECTS ON THE ADSORPTION In order to study the effect of temperature and pressure on the adsorption of water on AgI surfaces, we have calculated the Gibbs free energy starting by the adsorption energy of one water molecule on (110) surface in the case of Ag-O interaction (Eads= 0.32 eV). As it can be seen from the plot in Figure 6, that adsorption can happen only at very low temperature (200 K curve) while it is not possible at room temperature. As the most stable adsorptions occur on the (110) surfaces, these results demonstrate that no water will be adsorbed on

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

the AgI surface at temperature and pressure representatives of the nuclear containment building in severe accident conditions.

Figure 6: Gibbs free energy plot for water adsorption on AgI (110) surface (temperature in Kelvin, P°=1.00 bar)

3.2.4 HALF AND FULL WATER MONOLAYER ADSORPTION Water molecules were added successively on each AgI surfaces to reach half and monolayer coverages. The most stable configurations are presented in Figure 7. The adsorption interaction energies were calculated according to the following equation: Eads = nEH2O +EAgI -E(AgI+ n H2O) Where E(AgI+H2O), n  EH2O, EAgI are the energies of the adsorption cell, ‘n’ separate water molecules in gas phase and energy of AgI substrate respectively. In addition, water-water and surface-water interactions were calculated according to the following formulas respectively: Ew_w = n*E(H2O)gas - Ewater polymer Es_w = Eads - Ew_w

ACS Paragon Plus Environment

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 14 of 27

Table 2 contains the calculated energies for half and one monolayer water coverage of the AgI stable surfaces. The corresponding geometries are represented in Figure 7. Starting by the half monolayer case, two water molecules are adsorbed on the surface interacting together and with the surface. Adsorption on (100) is the most favored one (rE = 0.71 eV) followed by the (110) and (120) surfaces with interaction energies equal to 0.60 and 0.41 eV, respectively. Water-water interaction energy is nearly the same for the three surfaces with energy close to 0.20 eV. The differences between the surfaces are due to the surface-water interaction contributions. Adding more water molecules on the surface would favor the water-water interaction instead of the surfacewater part as demonstrated by the monolayer case. (Table 2. Adsorption and interaction energies for half monolayer and monolayer coverage on the stable AgI surfaces.) The (100) surface has the stronger interaction. The adsorption energy is 2.09 eV with a water-water energy of 1.37 eV and a surface water one of 0.72 eV, which means that the interaction between water molecules is twice larger than the water surface ones. Similar conclusion can be drawn for all other surfaces since the energies are comparable with (100) surface. The orientation of water molecules presented in Figure 7 (d, b and f) shows that each proton of a water molecule is directed towards the oxygen of the neighboring one with a distance of 1.8 Å which is the typical distance between interacting water molecules

Half monolayer

Monolayer

(110)

a

b

(100)

c

d

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

(120)

e

f

Figure 7. Top view of water adsorption geometries for half monolayer and monolayer on stable AgI surfaces (Ag, I, H and O are respectively colored in bright blue, purple, red and white).

Table 2. Adsorption and interaction energies for half monolayer and monolayer coverage on the stable AgI surfaces. Half monolayer

Monolayer

(2 H2O)

(4 H2O)

Surface

(110)

(100)

(120)

(110)

(100)

(120)

Eads(eV)

0.60

0.71

0.41

1.74

2.09

1.50

Ew-w(eV)

0.25

0.22

0.18

1.36

1.37

1.33

Es-w(eV)

0.35

0.49

0.23

0.38

0.72

0.17

From all our previous results, it can be concluded that the interactions between water molecule and AgI surfaces are weak. Adding more water molecules on the surface would enhance the interaction between water molecules itself instead of interaction with the surface (Ew-w is about 40% of the Eads for the half-monolayer case while it represents about 80% of Eads in the one monolayer adsorption). From a reactivity point of view,

ACS Paragon Plus Environment

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 16 of 27

we can say that chemistry of AgI crystal, i.e. formation of volatile iodine species that may happen in severe accidental conditions, will happen on the bare surface without any water molecules.

3.3 Reactivity of AgI surfaces The second part of our study is dedicated to the possible formation of volatile iodine species from the stable surfaces of AgI. Different species are taken into account such as AgI(g), HI(g), IOH(g) and I2(g). In this part, we restrict this study on AgI(110) surface which is the dominant one either from the experimental or theoretical point of view. We will discuss the formation of chemical species from clean surface without and with oxidizing it (representative of radiation conditions) separately to draw a general conclusion at the end.

3.3.1 REACTIVITY WITHOUT OXIDANT Starting from bare surfaces, AgI molecule can be released from the surface directly to the gas phase in a reaction opposite to the crystal formation. As expected, this reaction is very endothermic (∆rE= 2.38 eV) and it is not probable thermodynamically. The investigate in a second step the formation of I2(g). During this reaction, the iodine is oxidized while the silver is reduced. The reaction is very endothermic (∆rE= 3.5 eV) and I2(g) will not be formed directly. An alternative is to consider the participation of one water molecule in the mechanism that could lead to HI formation by acid-base reaction. Dissociation of water way on the surface is very endothermic (∆rE= 3.2 eV on the (110) surface) while the formation of gaseous HI(g) molecule in the second step, is a exothermic reaction (∆rE= -0.38 eV). Thermodynamically, this reaction is not possible due to the first step (water dissociation).

3.3.2 REACTIVITY WITH OXIDANTS

ACS Paragon Plus Environment

Page 17 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

In the absence of oxidants, we have showed that the formation of gaseous iodine species from the surface is not favored. Thus, we oxidize the surface with one or two electrons in order to promote reaction pathway without very endothermic steps. One of the possible oxidants that may exist during a severe accident is the OH° radicals, produced from steam radiolysis.

3.3.2.1 Formation of I2 after oxidizing the surface with one OH° radical Starting with a clean surface, one OH° radical is adsorbed on top of Ag atom (d(O-Ag)= 2.07 Å).then, I2 is formed from the surface and released into the gas phase. Addition of one OH° group on the surface is exothermic (∆rE= -1.52eV) while the departure of gaseous I2 is still an endothermic step (∆rE= 2.20 eV). Thus, this two electrons oxidation reaction pathway is not favor on the surface when one one electron oxidant such as OH° is involved. The re-volatilization of I2 in presence of only one OH° is not possible. The surface has to be oxidized twice to release I2(g) in the gas phase.

3.3.2.2 Formation of I2 after oxidation of the surface by two OH° radicals In this section, reaction pathways including two OH are investigated. Different mechanisms have been explored to find the favored one either from the thermodynamic or kinetic point of view. Figure 8 shows the pathway in which we oxidize the surface twice in a first step (A to B figures). This reaction is exothermic (∆E= -2.83 eV). The second step corresponds to the displacement of one iodine from the surface to form IOH on top of Ag atom (C figure). Transition state of this step is shown in Figure 9. The activation energy is 0.45 eV. The next step is the formation of I2 on the surface. I2 is located in a bridging position between the OH group and a Ag cation of

ACS Paragon Plus Environment

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 18 of 27

the surface (geometry D). Activation energy of this step is 0.39 eV The transition state is represented in

Figure 10. The next step is the displacement of I2 on the surface to break the Ag-I2 bond. The I2 molecule is positioned perpendicularly to the surface (geometry E). The activation energy for this last step D to E is very small (Eact= 0.15 eV) (Figure 11). The I2 departure from the surface (E to F reaction) requires activation energy of 0.43 eV. This value is similar to the reaction energy as the I2(g) absorption on the surface is non-activated. The final step of the reaction mechanism, F to G, corresponds to the displacement of OH in order to fill the vacancy created by the missing iodine.

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Figure 8: Proposed reaction mechanism leading to the formation of I2 in presence of two OH° radicals (Ag, I, H

and

O

are

respectively

colored

in

bright

ACS Paragon Plus Environment

blue,

purple,

red

and

white).

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 20 of 27

Figure 9: Activation energy and Geometry of the transition state between geometries B and C (Ag, I, H and O are respectively colored in bright blue, purple, red and white).

Figure 10: Activation energy and Geometry of the transition state between geometries C and D (Ag, I, H and O are respectively colored in bright blue, purple, red and white).

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Figure 11: Activation energy and Geometry of the transition state between geometries D and E (Ag, I, H and O are respectively colored in bright blue, purple, red and white). 3.3.2.3 Formation of IOH after oxidation of the surface by two OH° radicals Since the formation of IOH occurs during the formation of the I2 molecule, we investigated its possible volatilization. The reaction mechanism is presented in Figure 12. The first two steps are similar to the previous case (Figure 8). Formation of IOH, in step D’, corresponds to the simple desorption. The activation energy of this step is 0.21 eV (D’). As for I2, IOH adsorption on the surface is not activated. Finally, the second OH° move on the surface to fill the anionic vacancy.

ACS Paragon Plus Environment

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 22 of 27

Figure 12: Proposed reaction mechanism leading to the formation of IOH in presence of two OH° radicals (Ag, I, H and O are respectively colored in bright blue, purple, red and white). 4. Conclusion In the present work, we have studied the chemistry of AgI nanocrystal in dry and moist atmospheres. The Wulff construction of AgI particles, based on the relaxed energies of the low indices surfaces (100), (110) and (120) surfaces, have been established. The water molecules are adsorbed associatively on the surface by interaction between oxygen and silver on the surface with adsorption energy of 0.32, 0.24, 0.20 eV for (110), (100) and (120), respectively. Adsorption of half and one monolayer of water show that water-water interactions are more important than water-surface interactions. This indicated that the interactions between water layer and surface were very weak. It can be noticed that the formation of hydrogen bonds network between the adsorbed water molecules tends to stabilize water layer on the surface. Taking into account temperature and pressure effects, we have demonstrated that water could only be adsorbed at very low temperature.

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

Possible release of iodine species from AgI(110) surface were explored in a second step. Formation of I2 and AgI(g) is not possible without the participation of oxidants species. The formation of I2 after oxidizing the surface once by OH° involves some very endothermic step and is not favored (∆rE= 2.20 eV). On the contrary, oxidation mechanisms involving the participation of two OH° lead to the volatilization of iodine species (I2 and IOH) with low activation energies, i.e. below 0,50 eV for both species. Consequently, AgI would be a source of volatile iodine species at low temperature in presence of oxidant. In the frame of OECD STEM2 project (https://www.oecd-nea.org/jointproj/stem2.html) conducted by IRSN, labelled silver iodide aerosols previously deposited on a glass coupon will be -irradiated in a steam/air atmosphere at 80°C. Online iodine revolatilisation will be measured to check and quantify these phenomena. Acknowledgements H. H thanks IRSN for PhD grant. This work has been supported by the French State under the program "Investissements d’Avenir MiRE managed by the ANR under grant agreement ANR-11-RSNR-0013-01. The authors thank the Lille University Computational Centre (CRI) partially funded by Feder for CPU allocation.

References (1) (2) (3) (4) (5)

Corrin M. L.; Edwards, H. W.; Nelson, J. A. The Surface Chemistry of Condensation Nuclei: 2. The Preparation of Silver Iodide Free of Hygroscopic Impurities and Its Interaction with Water Vapor. Journal of the Atmospheric Sciences 1964, 21 (5), 565–567. Cao, J.; Xu, B.; Luo, B.; Lin, H.; Chen, S. Preparation, Characterization and Visible-Light Photocatalytic Activity of AgI/AgCl/TiO2. Applied Surface Science 2011, 257 (16), 7083–7089. Wen, X.-J.; Niu, C.-G.; Ruan, M.; Zhang, L.; Zeng, G.-M. AgI Nanoparticles-Decorated CeO 2 Microsheets Photocatalyst for the Degradation of Organic Dye and Tetracycline under Visible-Light Irradiation. Journal of Colloid and Interface Science 2017, 497, 368–377. Simondi-Teisseire, B.; Girault, N.; Payot, F.; Clément, B. Iodine Behaviour in the Containment in Phébus FP Tests. Annals of Nuclear Energy 2013, 61, 157–169. Beahm, E. C.; Weber, C. F.; Parker, G. W. Iodine Chemical Forms in LWR Severe Accidents. U.S.NRC Report NUREG/CR-5732, Oak Ridge National Laboratory 1992.

ACS Paragon Plus Environment

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

(6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26)

Page 24 of 27

Gordienko, A. B.; Kravchenko, N. G.; Sedelnikov, A. N. Ab Initio Calculations of the Lattice Dynamics of Silver Halides. Russian Physics Journal 2010, 53 (7), 692–697. Vogel, D.; Krüger, P.; Pollmann, J. Ab Initio Electronic Structure of Silver Halides Calculated with Self-Interaction and Relaxation-Corrected Pseudopotentials. Physical Review B 1998, 58 (7), 3865. Wilson, D. J.; Sokol, A. A.; French, S. A.; Catlow, C. R. A. Defect Structures in Silver Chloride. Journal of Physics: Condensed Matter 2004, 16 (27), S2827–S2838. Gordienko, A. B.; Zhuravlev, Y. N.; Poplavnoi, A. S. Electronic Structure of AgCl, AgBr, and AgI. phys. stat. sol. (b) 1991, 168 (1), 149–156. Catti, M. First-Principles Landau Potential for the Rocksalt to KOH to TlI-Type Phase Transitions of AgI. Physical Review B 2006, 74 (17). Catti, M. Kinetic Mechanisms of the Pressure-Driven Phase Transitions of AgI. Physical Review B 2005, 72 (6). Hull, S.; Keen, D. A. Pressure-Induced Phase Transitions in AgCl, AgBr, and AgI. Physical Review B 1999, 59 (2), 750. Shevkunov, S. V. Structure of Water in Microscopic Fractures of a Silver Iodide Crystal. Russian Journal of Physical Chemistry A 2014, 88 (2), 313–319. Corrin, M. L.; Nelson, J. A. Energetics of the Adsorption of Water Vapor on “Pure” Silver Iodide. The Journal of Physical Chemistry 1968, 72 (2), 643–645. Taylor, J. H.; Hale, B. N. Monte Carlo Simulations of Water-Ice Layers on a Model Silver Iodide Substrate: A Comparison with Bulk Ice Systems. Physical Review B 1993, 47 (15), 9732–9741. Glatz, B.; Sarupria, S. The Surface Charge Distribution Affects the Ice Nucleating Efficiency of Silver Iodide. The Journal of Chemical Physics 2016, 145 (21), 211924. Hiemstra, T. Variable Charge and Electrical Double Layer of Mineral–Water Interfaces: Silver Halides versus Metal (Hydr)Oxides. Langmuir 2012, 28 (44), 15614–15623. Hafner, J. Ab-Initio Simulations of Materials Using VASP: Density-Functional Theory and Beyond. Journal of Computational Chemistry 2008, 29 (13), 2044–2078. G.Kresse; J.Furthmuller. Efficiency of Ab-Initio Total Energy Calculations for Metals and.Pdf. Coputational Materials Science 1996, 6, 15–50. Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B 1999, 59 (3), 1758–1775. John P. Perdew, Kieron Burke, Matthias Ernzerhof. Generalized Gradient Approximation Made Simple.Pdf. Phys. Rev. Lett. 1996, pp 3865–3868. Hendrik J. Monkhorst and James D. Pack. Special Points for Brillonin-Zone Integrations.Pdf. Physical Review B. 1989, p 3616. Kebede, G. G.; Spångberg, D.; Mitev, P. D.; Broqvist, P.; Hermansson, K. Comparing van Der Waals DFT Methods for Water on NaCl(001) and MgO(001). The Journal of Chemical Physics 2017, 146 (6), 064703. Tkatchenko, A.; Scheffler, M. Accurate Molecular Van Der Waals Interactions from Ground-State Electron Density and Free-Atom Reference Data. Phys. Rev. Lett. 2009, 102 (7), 073005. Bučko, T.; Lebègue, S.; Hafner, J.; Ángyán, J. G. Improved Density Dependent Correction for the Description of London Dispersion Forces. J. Chem. Theory Comput. 2013, 9 (10), 4293–4299. Bučko, T.; Lebègue, S.; Ángyán, J. G.; Hafner, J. Extending the Applicability of the TkatchenkoScheffler Dispersion Correction via Iterative Hirshfeld Partitioning. J Chem Phys 2014, 141 (3), 034114.

ACS Paragon Plus Environment

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

The Journal of Physical Chemistry

(27) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Physical Review B 1976, 13 (12), 5188–5192. (28) Reuter, K.; Scheffler, M. Composition and Structure of the RuO 2 ( 110 ) Surface in an O 2 and CO Environment: Implications for the Catalytic Formation of CO 2. Physical Review B 2003, 68 (4). (29) Geysermans, P.; Finocchi, F.; Goniakowski, J.; Hacquart, R.; Jupille, J. Combination of (100), (110) and (111) Facets in MgO Crystals Shapes from Dry to Wet Environment. Physical Chemistry Chemical Physics 2009, 11 (13), 2228. (30) Tougerti, A.; Méthivier, C.; Cristol, S.; Tielens, F.; Che, M.; Carrier, X. Structure of Clean and Hydrated α-Al2O3 (11̄02) Surfaces: Implication on Surface Charge. Physical Chemistry Chemical Physics 2011, 13 (14), 6531. (31) Henkelman, G.; Uberuaga, B. P.; Jónsson, H. A Climbing Image Nudged Elastic Band Method for Finding Saddle Points and Minimum Energy Paths. The Journal of Chemical Physics 2000, 113 (22), 9901. (32) Henkelman, G.; Jónsson, H. Improved Tangent Estimate in the Nudged Elastic Band Method for Finding Minimum Energy Paths and Saddle Points. The Journal of chemical physics 2000, 113 (22), 9978–9985. (33) Kuang, Q.; Zheng, X.; Yang, S. AgI Microplate Monocrystals with Polar {0001} Facets: Spontaneous Photocarrier Separation and Enhanced Photocatalytic Activity. Chemistry - A European Journal 2014, 20 (9), 2637–2645.

ACS Paragon Plus Environment

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

TOC graphic

ACS Paragon Plus Environment

Page 26 of 27

Page 27 of 27

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

ACS Paragon Plus Environment