ReaxFF molecular dynamics study on the influence of temperature on

Research Article Cite This: ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

www.acsami.org

ReaxFF Molecular Dynamics Study on the Influence of Temperature on Adsorption, Desorption, and Decomposition at the Acetic Acid/ Water/ZnO(101̅0) Interface Enabling Cold Sintering Mert Y. Sengul,†,‡ Clive A. Randall,†,‡ and Adri C. T. van Duin*,‡,§ †

Department of Materials Science and Engineering, ‡Materials Research Institute, and §Department of Mechanical and Nuclear Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802, United States

ACS Appl. Mater. Interfaces Downloaded from pubs.acs.org by UNIV OF WINNIPEG on 10/17/18. For personal use only.

S Supporting Information *

ABSTRACT: The reaction dynamics of a liquid−solid interface with the example of an acetic acid/water solution interacting with a ZnO(101̅0) surface was investigated using ReaxFF reactive force field-based molecular dynamics. The interactions were studied over a broad temperature range to assess the kinetics and reaction pathways. Two different acetic acid dissociation mechanisms are observed in the simulations: (1) deprotonation to surface cation, which produces a terminal hydroxyl and (2) deprotonation to a bridging hydroxyl, which results in water production. An increase in temperature promotes the dissociation of acetic acids and its adsorption to surface at first, but as the temperature increase continues, the surface coverage by acetates decreases due to evaporation from the surface or decomposition. The acetate decomposition starts with a nucleophilic attack of oxygen to methyl carbon and results in the production of carbon dioxide, which is consistent with experimental findings in the literature. The coverage of the surface by water molecules decreases as the system is heated up, which is also observed in other molecular dynamics studies. At elevated temperatures, acetate molecules are more stable than water molecules or bridging hydroxyls on the surface. These simulations validate the ReaxFF method for the water/organic mixture and metal oxide surface interactions and provide insights into structure and reactivity of aqueous solvents on metal oxide surfaces at elevated temperatures. Adsorption trends that are observed in this study are consistent with phenomenological Langmuir models. The reaction of acetic acid decomposition to smaller molecules such as CO2 and CH2O agrees with experimental observations. Understanding the details of these dynamic surface reactions are critical to understand important new cold sintering processes that utilize transient liquid and solid reactions, and the latter could be used to predict solvent selection for cold sintering. KEYWORDS: atomistic simulation, Langmuir adsorption, zinc oxide, interfacial reaction dynamics, ReaxFF, surface coverage, decomposition, cold sintering

1. INTRODUCTION

the macroscopic properties. For example, optoelectronic properties can be tuned via attachment of organic molecules on the ZnO surface. However, at the same time, the presence of organic molecules in the interface can affect macroscopic properties negatively such as the decrease in electrical conductivity due to the change in interface chemistry.10 Therefore, several studies have been conducted to understand the interaction of ZnO surfaces with organic molecules. Bowker et al. have investigated the adsorption of acetic acid and acetaldehyde on polycrystalline ZnO materials using a mass spectrometer and observed that these organic molecules were adsorbed as acetate and then decomposed with increasing temperature.11 Vohs et al. have studied the interaction of acetic

There has been an increasing interest on semiconductor metal oxide materials over the last years because of their unique optoelectronic properties.1 In particular, zinc oxide (ZnO) materials have attracted the technological interest because of having a direct wide band gap (Eg ≈ 3.3 eV at room temperature) semiconductor and having a large exciton binding energy (∼60 meV). The combination of these novel optical and electrical properties of ZnO have led to applications such as UV lasers,2 field-effect transistors,3 varistors,4 dye-sensitized solar cells,5 and gas sensors6 and make this material one of the most studied ceramics and is a model system for many sintering studies including the cold sintering process.7−9 Depending on the application, the ZnO can be in the form of a nanomaterial or bulk ceramics. When ZnO is synthesized as a nanomaterial, the characteristics of its interface dominates © XXXX American Chemical Society

Received: August 9, 2018 Accepted: October 3, 2018 Published: October 3, 2018 A

DOI: 10.1021/acsami.8b13630 ACS Appl. Mater. Interfaces XXXX, XXX, XXX−XXX

Research Article

ACS Applied Materials & Interfaces

minimization simulations, 12 layers of ZnO slabs, each composed of 360 zinc (Zn) and oxygen (O) atoms, were used. The amount of coverage on the surface by acetic acid molecules was indicated as a normalized order parameter, θ. The number of acetic acid molecules for θ = 1.0 is 30 and θ = 0.5 is 15. The minimum energy configurations are shown in Figure 1. A dissociative adsorption is

and propionic acids with polar ZnO(0001) surfaces and observed carbon dioxide (CO2) release due to the decomposition at elevated temperatures.12 Moezzi et al. have investigated transition of zinc hydroxyacetate to zinc oxide and observed decomposition of acetate to CO2 at elevated temperatures.13 Some other studies have been conducted to explain the interaction of ZnO surfaces with different organic molecules.14−17 In addition to adsorption−decomposition studies, some others have been carried out to investigate the functionalization of ZnO nanomaterials by organic molecules, 10,18−20 including the carboxylic acid groups (−COOH).21 Moreira et al. have examined the stabilization of carboxyl group on nonpolar ZnO surfaces at room temperature in vacuum using density-functional theory (DFT).22 They have concluded that a fully covered surface is favorable in ligand-rich environments. Several other studies have been conducted to investigate the influence of acetate and water molecules on ZnO crystal growth, and it has been understood that the co-presence of water and acetate supports the ZnO coarsening23 and is important for the compaction of ZnO particles into densified ceramics and composites through a cold sintering process.8 Chemical interactions at metal oxide/aqueous interfaces are in general complex. For metal oxides like ZnO, which have properties that are sensitive to interface chemistry, there is a need to better understand the interaction dynamics between water, acetic acid, and ZnO surface. The chemical interactions can be examined in detail using first-principles-based methods. However, the computational cost of these methods to model a metal oxide surface in aqueous medium is high and therefore limits the system size and simulation time. As an alternative to first-principles-based methods, ReaxFF24,25 is an empirical potential for molecular dynamics (MD) based on bond orders and is able simulate chemical reactions. Therefore, in this work, we used reactive MD simulations with ReaxFF potential to model the interactions between respective acetic acid and water molecules and a ZnO(101̅0) surface. The (101̅0) surface has been widely studied experimentally and theoretically22,26−32 as it is the main cleavage plane, has one of the lowest formation energies, and is one of the most commonly observed surface in ZnO powders.33 Thus, this paper focuses on the effect of temperature on adsorption, desorption, and decomposition reaction dynamics in the acetic acid/water/ ZnO(101̅0) system.

Figure 1. Minimum energy configuration of ZnO(101̅0) covered by acetic acid molecules. Top: [0001] view, bottom: [121̅0] view. Oxygen, hydrogen, carbon, and zinc atoms are visualized as red, white, cyan, and silver spheres, respectively, using VMD software.37 (a) 100% coverage (θ = 1.0) configuration, (b) 50% coverage (θ = 0.5) configuration. favored for acetic acid molecules on ZnO(101̅0) surface. For full coverage (θ = 1.0), the hydrogen bonds of the methyl hydrogens with oxygens of other acetates affect the orientation of acetates with respect to the surface normal. Energy minimization simulations for two different coverages were conducted, and adsorption energies were calculated using eq 1. Eadsorption =

Ecovered − Eclean − NEmolecule N

(1)

where Ecovered is the total energy of covered surface, Eclean is the energy of clean surface, N is the number of acetic acid molecules adsorbed on ZnO(101̅0) surface, and Emolecule is the energy of a neutral molecule in vacuum. The calculated adsorption energy values using ReaxFF have been compared with the DFT calculation results of22 in Table 1. The results are in a very good agreement with those from earlier DFT study.

Table 1. Adsorption Energies, Eadsorption (eV) of Acetic Acid Molecules on ZnO(101̅0) Surfaces

2. METHODS 2.1. ReaxFF Validation. A new ReaxFF potential was used to model the acetic acid/water/ZnO interfacial reactions by combining two previously developed potentials. The potential for H/O/Zn interaction was taken from ref 34. This potential has been developed to study water/ZnO interactions and validated by describing the allimportant adsorption and dissociation mechanisms of water on different ZnO surfaces. The potential for C/H/O interaction was taken from ref 35, which has been developed to model the conformational dynamics of biomolecules in solution. The combined potential has been optimized to model deprotonation of acetic acid in water.36 The combined potentials are from the aqueous branch of ReaxFF and therefore can be merged by only optimizing the newly created interaction terms. We note that because the Zn/C pair interactions are not expected in this system, bonding interaction terms between these atoms were disabled by using dummy parameters during force field optimization process. The interaction between solvent molecules and surfaces can be investigated by adsorption energies; therefore, energy minimization simulations were conducted to calculate the adsorption. In the

coverage (θ)

ReaxFF

ref 22

0.5 1.0

2.040 1.396

2.07 1.39

2.2. ReaxFF MD Simulation Details. ZnO has a wurtzite global energy minimum structure under ambient conditions. The cations (Zn2+) are coordinated tetrahedrally with oxygen ions (O2−). The nonpolar ZnO(101̅0) surface is one of the low-index surfaces of the wurtzite structure. Each cation on this surface is missing the oxygen ligand in the fourth tetrahedral position and each oxygen is missing the nearest cation. In this study, the ZnO slab used in the simulations consists of 14 layers containing 3024 Zn2+ and O2− ions. The thickness of the slab is more than the cutoff value of the interactions between atoms. There are a total of 432 cation and anion sites on both surfaces of the slab. All of the slabs were energy minimized before using in the preparation of the initial configurations of simulations. Experimentally, it is known that ZnO is insoluble in B


Research Article


Figure 2. Visualization of atomistic system at different temperatures. Oxygen, hydrogen, carbon, and zinc atoms are visualized as red, white, cyan, and silver spheres, respectively, using VMD software.37 water; however, this changes with the addition of an organic acid.8 Here, an aqueous mixture was used in the simulations contains 800 water and 200 acetic acid molecules. A high acid concentration was used in simulations to mimic the low pH conditions used in cold sintering process. The solution was prepared by positioning each molecule randomly into the simulation box, followed by an energy minimization. The density of the aqueous mixture was determined by keeping water density as 1 g/cm3 and acetic acid density as 1.05 g/ cm3 by adjusting the simulation box dimensions. Then, the prepared equilibrated liquid slab was placed on top of the ZnO surface and the vacuum region was created by enlarging the simulation box in the direction perpendicular to the surface (Figure 2). The final simulation box dimensions are 60 × 65 × 300 Å3 including the vacuum region. The ReaxFF reactive potential discussed in Section 2.1 was used for the MD simulations.24 Reactive MD calculations were performed using the ADF software.38 We used a 0.25 fs time step to integrate the Newtonian equations of motion by using a velocity-Verlet algorithm. A weak Berendsen thermostat39 was used to maintain temperatures at desired values. The temperature-damping constant used in all simulations was 0.1 ps.

with respect to the total number of available sites for adsorption on surface. 3.1. Acetic Acid/Water/ZnO(101̅0) Interaction at T = 300 K. The fractions of the respective molecular species populations in the gas phase and on the ZnO surface at room temperature are shown in Figure 3. According to our ReaxFF

Figure 3. Fractions of the populations of the species in the gas phase and on the surface at room temperature. The fractions were calculated by dividing by the total number of corresponding species in the simulation box. Acetic acid, acetate, and water indicated as CH3COOH, CH3COO, and H2O, respectively. The relative surface coverage was calculated with respect to the total number of available sites for adsorption on surface.

3. RESULTS AND DISCUSSION To investigate the reaction dynamics between acetic acid molecules and ZnO(101̅0) surface in the presence of water, ReaxFF MD simulations were performed at effective temperatures ranging between 300 and 1200 K to drive the reactions within the computational time. All simulations were heated up to their desired temperatures with 100 K increments, and after each increment, a 50 ps equilibration simulation was performed. An additional 1 ns equilibration simulation was performed at the final temperature and the population analysis for molecular species was done by averaging at every 25 fs over the last 0.5 ns of each simulation. The population of the molecular species were calculated in the gas phase and on the surface separately. The molecules with a distance of 3.0 Å or less from surface cations were counted as surface species, and the rest were counted as gas species. The fractions were calculated by dividing by the total number of corresponding species: the fractions of acetate, acetic acid, and carbon species were calculated with respect to the total number of acetic acids in the system; the fractions of water and hydroxyls were calculated with respect to the total number of water molecules in the system; and the relative surface coverage was calculated

MD simulations, 33.3% of water molecules are adsorbed on the surface either molecularly or dissociatively and the remaining 66.7% are in the solution phase. Of the adsorbed water molecules, half display a molecular adsorption and the other half dissociate into terminal and bridging hydroxyls. Similar dissociation ratios have been reported by DFT and previous ReaxFF studies.27,40 An example of the dissociation reaction for water is given in eq 2, where the first term on the righthand side of the equation is bridging hydroxyl and the second term is the terminal hydroxyl. The molecular adsorption of water molecules is observed through the oxygen atom of water on-top Zn cation site. We observed the (2 × 1) half-dissociated pattern between water molecules, which has also been observed by previous ReaxFF MD, DFT,27,40 and experimental41 studies. In addition to molecular adsorption, some water molecules remain close to the surface via hydrogen bonding with terminal hydroxyls (Figure 4). C


Research Article


Figure 4. Snapshots from our ReaxFF MD simulations showing the different adsorption states of water and acetic acid on ZnO surface. (a) Bridging and terminal hydroxyls, (b) molecularly adsorbed water, and (c) adsorbed acetate and terminal hydroxyl. Oxygen, hydrogen, carbon, and zinc atoms are visualized as red, white, cyan, and silver spheres. Oxygen and hydrogen atoms that are of interest are visualized as blue and yellow, respectively. All visuals are prepared by using VMD software.37

Approximately 58% of the acetic acid molecules in the system are adsorbed on the surface. Dissociation of Brønsted acids (i.e., acetic acid) is a very common mechanism on metal oxide surfaces17 and is also readily observed in our simulations. In the simulations, two types of dissociation mechanisms are observed. The first dissociation path results in the protonation of negatively charged surface oxygens and coordination of positively charged surface cation by acetate (eq 3). The second type of dissociation is the protonation of a bridging hydroxyl, which produces a water molecule (eq 4). The second mechanism can also be thought as an exchange mechanism between acetate and the bridging hydroxyl. Depending on the environment, acetate may bind to empty sites left by the bridging hydroxyl or remain in the gas phase. This observation explains why over 10% of the acetic acid molecules are present as an acetate in the gas phase (Figure 3). H 2O(g) + 2ZnO(s) F ZnO(OH)(s) + ZnOH(s)

(2)

CH3COOH(g) F CH3COO(ad) + H(ad)

(3)

CH3COOH(g) F OH(ad) → CH3COO(g) + H 2O(g)

(4)

Figure 5. Schematic representation of adsorption, desorption, and decomposition reactions at the acetic acid/water/ZnO(101̅0) interface with respect to temperature.

species in the gas phase and on the surface at different temperatures are shown in Figures 6 and 7. The average values

According to our simulations, there are three different adsorption configurations for acetic acid molecules; (1) the binding of carbonyl oxygens to same cation (chelating bidentate); (2) the binding of carbonyl oxygens to separate cations (end-bridged or on-top bidentate); and (3) the binding of one of the carbonyl oxygens to cation (monodentate). Molecular acetic acid adsorption via methyl hydrogen is not observed here. The surface coverage presented in Figure 3 is calculated by taking hydroxyls and adsorbed water and acetic acid molecules into account. According to our simulations, approximately 80% of the surface sites are covered, which is similar to the results obtained by Raymand et al.27 for water species coverage on the ZnO (101̅0) surface. Most of the surface is covered by water species, but given that the amount of water is more than that of the acetic acid, our results indicate that the acetic acid molecules are more likely to dissociate and are relatively more stable on the surface compared with a water molecule. 3.2. ZnO(101̅0)/Acetic Acid/Water Interaction at Elevated Temperatures. Several studies have been conducted to investigate the molecular desorption at the ZnO surfaces. It was found that the increase in temperature changes the surface chemistry and initiates the decomposition of the organic molecules (Figure 5).11,14−16 In this study, to investigate the respective effect of temperature on adsorption, desorption, and decomposition of acetic acid and water molecules on ZnO(101̅0) surfaces, ReaxFF MD simulations were performed ranging from 400 to 1200 K with 200 K increments. The fractions of the populations of the molecular

Figure 6. Fractions of the populations of the species in gas phase at different temperatures. The fractions were calculated by dividing by the total number of corresponding species in the simulation box. Acetic acid, acetate, water, carbon dioxide, and formaldehyde indicated as CH3COOH, CH3COO, H2O, CO2, and CH2O, respectively.

for molecular species are calculated with respect to the total number of corresponding species in the system as discussed in Section 3. As the system is heated up, the percentage of acetic

Figure 7. Fractions of the populations of the species on the surface at different temperatures. The fractions were calculated by dividing by the total number of corresponding species in the simulation box. The relative surface coverage is calculated with respect to the total number of available sites for adsorption on surface. D


Research Article


Figure 8. (a) Plot of ln[C − 1(θ/1 − θ)] vs θ of data obtained from simulations (blue dots) and fitted equation (red line). (b) Plot of interaction energy between neighboring molecules for corresponding temperatures.

decrease in relative surface coverage is observed for the acetic acid/water/ZnO(101̅ 0 ) interface. The ZnO surface is composed of anions and cations which create a certain number of well-determined adsorption sites for water and organic acids. The surface area and the concentration are constant, and relative surface coverage values are given in Figure 7. After adsorption, there are lateral interactions between adsorbed species. Under these circumstances, the adsorption mechanism in this system satisfies sufficient conditions for a modified Langmuir model with lateral interactions, also called Frumkin−Fowler−Guggenheim adsorption.42 The linearized form of the modified Langmuir equation is as follows ÅÄÅ 1 i θ yÑÉÑ Å zzÑÑÑ = ln((K ) + nE L θ lnÅÅÅ jjj L ÅÅÇ C k 1 − θ z{ÑÑÑÖ RT (5)

acid molecules in the gas phase decreases initially and then increases, which is similar to experimental observations.11,16 As can be seen from Figure 7, the percentage of adsorbed carbon species start increasing until T = 800 K, which is the reason for the decrease in acetic acid or acetate amount in the gas phase (Figure 6). The reason of the increase in the adsorbed acetate amount is the exchange mechanism between acetic acid and bridging hydroxyl (eq 4). After T = 800 K, there is a decrease in carbon species amount on the surface and increase in the gas phase. Also, at T = 800 K, we begin to see CO2 molecules appear in the system. This suggests that after T = 800 K, the carbon species on the surface evaporated to a gas phase as an acetate and/or acetic acid or decomposed to CO2 and formaldehyde (H2CO). In experiments,11 the temperature where the increase in acetic acid amount in the gas phase and CO2 release starts are reported to be approximately equal, which is also the case observed in our simulations. It should be noted that the temperature values reported in the experiments are lower than those in our simulations. This is because the temperature increase during the system heat up is faster in the simulations, compared with the experimentsas such our simulation tend to over-estimate exact temperatures and needs to be considered as a trend, not an absolute. As the system is heated up, there is an increase in the water amount in the gas phase because of the evaporation of molecularly adsorbed water molecules. At the same time, temperature increase enables the reassociation of the terminal and bridging hydroxyls to form free water molecules in the gas phase. The decrease in water, bridging hydroxyl, and terminal hydroxyl amounts on surface also explains these reactions (Figure 7). The decrease in the bridging hydroxyl amount is more than that of the terminal hydroxyl amount because in addition to reassociation with terminal hydroxyls, bridging hydroxyls re-associate with acetic acids to produce an acetate and water. It has been reported that the dissociated and molecularly adsorbed water ratio (1:1) remains constant at elevated temperatures.27 As discussed in Section A, we observed the same ratio and the (2 × 1) pattern at room temperature and T = 400 K as well. As the temperature increases, because of the increase in adsorbed acetates on surface, this pattern is disrupted and results in the change of 1:1 ratio between hydroxyls and adsorbed water molecules as shown in Figure 7. After T = 800 K, nearly all bridging hydroxyls are desorbed from the surface. The temperature and surface coverages are inversely proportional. At room temperature, nearly 80% of the surface is covered. The decrease of coverage with increasing temperature for water/ZnO(101̅ 0 ) interface has been previously observed.27 Similarly, in this study, an exponential

where θ is the relative surface coverage, C is the concentration, n is the neighboring binding sites, R is the gas constant, T is temperature, KL is Langmuir isotherm, and EL is the interaction energy between neighboring molecules. A plot of ln[C − 1(θ/1 − θ)] versus θ is given in Figure 8. Using the values for C = 0.2, n = 4, R = 8.63 × 10−5 eV/K−1 atom−1, and the constants from fitted equation (Figure 8a), the Langmuir constant is calculated as 0.575 and the interaction energies for corresponding temperatures is shown in Figure 8b. Although the absolute values might not be correct due to the overpredicted temperatures, the adsorption trend is in agreement and the interaction energies are positive. The positive interaction energy indicates that there is an attraction between neighboring molecules, which creates an environment available for the initiation of decomposition reactions that are discussed in more details below. 3.2.1. Reaction Pathway for Carbon Dioxide and Formaldehyde Production. The change in coordination environment of surface cations might enable cation-adsorbate reactions. CO2 desorption from ZnO and acetic acid interaction has been observed by several researchers.11−16 Bowker et al.11 have examined the decomposition of acetic acid on polycrystalline ZnO and observed that CO2 is the major product of the decomposition. Moezzi et al. have observed acetate to CO2 decomposition at elevated temperatures.13 In another study, it was reported that CH2O production is decomposed as a result of a nucleophilic attack of an oxygen on methyl carbon of methyl formate (HCOOCH3).14 In our simulations, the CO2 production is observed on ZnO(101̅0) as a result of a sequence of reactions. These reactions are similar to the ones suggested by Bowker et al.14 for decomposition of methyl formate into formaldehyde. The first step is the abstraction of methyl hydrogen of acetate E


Research Article


Figure 9. Snapshots from our ReaxFF MD simulations showing the acetic acid decomposition to formaldehyde and carbon dioxide. Both products are observed by experimental studies as a result of acetic acid and ZnO surface interactions. (a) Snapshot of acetate and water prior to the decomposition reactions. (b) Snapshot of deprotonation of methyl hydrogen by surface oxygen. (c) Snapshot of water molecule dissociated into bridging hydroxyl and terminal hydroxyl. (d) Snapshot of a nucleophilic attack to methyl carbon, resulting in C2H3O3 radical formation. (e) Snapshot of dissociation of C2H3O3 radical. (f) Snapshot of the decomposition to carbon dioxide and formaldehyde. Oxygen, hydrogen, carbon, and zinc atoms are visualized as red, white, cyan, and silver spheres. Oxygen and hydrogen atoms that participate into decomposition reaction are visualized as blue and yellow, respectively. All visuals are prepared by using VMD software.37 Time frame for each snapshot is given on the top right corner.

(Figure 9a,b), followed by a nucleophilic attack of hydroxyl oxygen to methyl carbon, which results in the production of C2H3O3 radical intermediate (Figure 9c,d) (eq 5). In the second step, the radical is deprotonated (Figure 9e) and converted into CO2 and CH2O (Figure 9f) (eq 6). In the simulations, CH2O is not adsorbed to the surface; instead, directly goes to the gas phase. However, CO2 is likely to be adsorbed on the surface, which explains the population difference in the gas phase between CH2O and CO2. CH3COO(ad) + HO(ad) → COH 2COO(ad) + 2H(ad)

(6)

COH 2COO(ad) → CH 2O(ad) + CO2(ad)

(7)

According to our simulations, the CO2, water, and CH2O are the major decomposition products. Bowker et al. have observed CH2O decomposition into CO2 on the ZnO surface.15 This decomposition was not observed in our simulations because the CH2O molecules tend to stay in the gas phase. However, by extending the simulations times, it may be possible to observe such decomposition. To investigate the energy barrier of the decomposition reaction of CH2O into CO2 on a ZnO(101̅0) surface, we applied a bond-restraint approach. The initial configuration was composed of a CH2O molecule close to surface and bridging hydroxyl close to it (Figure 10a). To initiate and maintain the decomposition reaction at T = 0.25 K, we applied three successive restraints. The applied restraint energy that is a function of distance between two atoms (rij) is given by Erestraint = k1{1 − exp( −k 2(rij − R 0))2 }

Figure 10. Potential energy profile of CH2O decomposition to CO2 on ZnO(101̅0) surface and snapshots from our ReaxFF MD simulations showing the decomposition reaction pathway. (a) Snapshot of CH2O and bridging hydroxyl prior to the decomposition reaction. (b) Snapshot of deprotonation of CH2O and hydrophilic attack of bridging hydroxyl oxygen to carbon. (c) Snapshot of CH2O2 approach to surface oxygen. (d) Snapshot of deprotonation of CH2O2. (e) Snapshot of deprotonation of CHO2, resulting in CO2. Oxygen, hydrogen, carbon, and zinc atoms are visualized as red, white, cyan, and silver spheres. All visuals are prepared by using VMD software.37

(8)

where k1 and k2 are constants and R0 is the initial distance between two atoms. First, we started a nucleophilic attack by adding restrain to the distance between bridging hydroxyl oxygen and carbon, which results in CH2O2 (Figure 10b). Then, we applied another restraint to drive molecule to adjacent surface oxygen F


Research Article

ACS Applied Materials & Interfaces (Figure 10c) and force the deprotonation of hydrogen to the oxygen which produces a CHO2 molecule (Figure 10d). The third restraint was applied to CHO2 for deprotonation to the closest surface oxygen and results in production of a CO2 molecule (Figure 10e). The energy barriers for entire decomposition reaction are 0.47, 5.89, and 4.52 kcal, respectively (Table 2). These energies are small enough to account for the decomposition reactions.

■

k1

k2

R0 (Å)

0.47 5.89 5.52

50.0 50.0 50.0

5.0 5.0 5.0

2.94 3.96 5.91

*E-mail: [email protected]. ORCID

Mert Y. Sengul: 0000-0002-5309-0316 Adri C. T. van Duin: 0000-0002-3478-4945 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The information, data, or work presented herein was funded in part by the Advanced Research Projects Agency-Energy (ARPA-E), U.S. Department of Energy, under Award Number DE-AR0000766. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

■

4. CONCLUSIONS The interaction between acetic acid and ZnO(101̅0) surface in the presence of water was investigated at different temperatures using ReaxFF MD simulations. A ReaxFF potential is used in this work that is the combination of previously developed O/H/Zn and O/H/C potentials, which are, respectively, developed to describe ZnO/water interactions and dynamics of biomolecules in solution. The adsorption energy acetic acid for this combined potential was found to be in good agreement with DFT calculations from the literature. According to our simulations, water molecules are dissociated on the surface as either molecularly or dissociatively. The half-dissociated pattern between adsorbed and dissociated water molecules was observed, which agrees with previous theoretical and experimental studies. An increase in temperature promotes the acetate adsorption, which results in the breakdown of this pattern. Acetic acid molecules are dissociated on surface by being deprotonated either by surface oxygen or bridging hydroxyl. As the system is heated up, acetate molecules are adsorbed initially, but then desorption starts, which is in agreement with experimental studies. At elevated temperatures, acetate molecules are more stable than water molecules or bridging hydroxyls. At room temperature, 80% of surface is covered; this coverage amount decreases with the increase of the temperature. This adsorption trend matches with a modified Langmuir model. As the temperature is increased, acetate molecules are decomposed to CO2 and CH2O which is in agreement with experimental observations. Our study demonstrated that the ReaxFF force field is capable of modeling the reaction in the acetic acid/water/ZnO interface. Using this force field and taking the advantage of computational efficiency of MD, it is possible to model different systems involving ZnO, acetate, and water at high temperatures and pressures. Results from such simulations enable an atomistic level investigation of reactions (e.g., crystal growth, dissolution) under ambient and supercritical conditions such as cold sintering process.

■

AUTHOR INFORMATION

Corresponding Author

Table 2. Restraint Energy Equation Parameters for Each Restraint and Corresponding Energy Barrier Values energy barrier (kcal/mol)

(TXT)

REFERENCES

(1) Yu, X.; Marks, T. J.; Facchetti, A. Metal Oxides for Optoelectronic Applications. Nat. Mater. 2016, 15, 383−396. (2) Huang, M. H.; Mao, S.; Feick, H.; Yan, H.; Wu, Y.; Kind, H.; Weber, E.; Russo, R.; Yang, P. Room-temperature Ultraviolet Nanowire Nanolasers. Science 2001, 292, 1897−1899. (3) Kim, H.-J.; Lee, C.-H.; Kim, D.-W.; Yi, G.-C. Fabrication and Electrical Characteristics of Dual-gate ZnO Nanorod Metal Oxide Semiconductor Field-effect Transistors. Nanotechnology 2006, 17, S327−S331. (4) Gupta, T. K. Application of Zinc Oxide Varistors. J. Am. Ceram. Soc. 1990, 73, 1817−1840. (5) Seow, Z. L. S.; Wong, A. S. W.; Thavasi, V.; Jose, R.; Ramakrishna, S.; Ho, G. W. Controlled Synthesis and Application of ZnO Nanoparticles, Nanorods and Nanospheres in Dye-Sensitized Solar Cells. Nanotechnology 2009, 20, 045604. (6) Wang, L.; Kang, Y.; Liu, X.; Zhang, S.; Huang, W.; Wang, S. ZnO Nanorod Gas Sensor for Ethanol Detection. Sens. Actuators, B 2012, 162, 237−243. (7) Nie, J.; Zhang, Y.; Chan, J. M.; Huang, R.; Luo, J. Water-Assisted Flash Sintering: Flashing ZnO at Room Temperature to Achieve ∼ 98% Density in Seconds. Scr. Mater. 2018, 142, 79−82. (8) Funahashi, S.; Guo, J.; Guo, H.; Wang, K.; Baker, A. L.; Shiratsuyu, K.; Randall, C. A. Demonstration of the Cold Sintering Process Study for the Densification and Grain Growth of ZnO Ceramics. J. Am. Ceram. Soc. 2017, 100, 546−553. (9) Gonzalez-Julian, J.; Neuhaus, K.; Bernemann, M.; da Silva, J. P.; Laptev, A.; Bram, M.; Guillon, O. Unveiling the mechanisms of cold sintering of ZnO at 250 °C by varying applied stress and characterizing grain boundaries by Kelvin Probe Force Microscopy. Acta Mater. 2018, 144, 116−128. (10) Wang, L.; Feng, L.; Cui, C.; Xu, S.; Liu, J.; Sun, X.; Gao, X.; Wang, W. Study on the Surface Functionalization of ZnO Nanorods and Their Tunable Electrochemiluminescence Properties. ECS J. Solid State Sci. Technol. 2016, 5, R74−R81. (11) Bowker, M. The Interaction of Acetaldehyde and Acetic Acid with the ZnO Surface. J. Catal. 1983, 79, 431−444. (12) Vohs, J. M.; Barteau, M. A. Reaction Pathways and Intermediates in the Decomposition of Acetic and Propionic Acids on the Polar Surfaces of Zinc Oxide. Surf. Sci. 1988, 201, 481−502. (13) Moezzi, A.; McDonagh, A.; Dowd, A.; Cortie, M. Zinc Hydroxyacetate and Its Transformation to Nanocrystalline Zinc Oxide. Inorg. Chem. 2013, 52, 95−102. (14) Bowker, M.; Houghton, H.; Waugh, K. C. TemperatureProgrammed Reaction Studies of the Interaction of Methyl Formate and Ethanol with Polycrystalline Zinc Oxide. J. Chem. Soc., Faraday Trans. 1 1982, 78, 2573−2582.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.8b13630. Adsopbed species on ZnO(101̅0) surface at different temperatures (PDF) G


Research Article

ACS Applied Materials & Interfaces (15) Bowker, M.; Houghton, H.; Waugh, K. C. Mechanism and Kinetics of Methanol Synthesis on Zinc Oxide. J. Chem. Soc., Faraday Trans. 1 1981, 77, 3023−3036. (16) Vohs, J. M.; Barteau, M. A. Spectroscopic Characterization of Surface Formates Produced via Reaction of HCOOH and HCOOCH3 on the (0001) Surface of Zinc Oxide. Surf. Sci. 1988, 197, 109−122. (17) Barteau, M. A. Organic Reactions at Well-Defined Oxide Surfaces. Chem. Rev. 1996, 96, 1413−1430. (18) Burger, A.; Srikantharajah, R.; Peukert, W.; Hirsch, A. Individualization and Stabilization of Zinc Oxide Nanorods by Covalent Functionalization with Positively Charged Catechol Derivatives. Chem.Eur. J. 2017, 23, 17257−17268. (19) Niepelt, R.; Schröder, U. C.; Sommerfeld, J.; Slowik, I.; Rudolph, B.; Möller, R.; Seise, B.; Csaki, A.; Fritzsche, W.; Ronning, C. Biofunctionalization of Zinc Oxide Nanowires for DNA Sensory Applications. Nanoscale Res. Lett. 2011, 6, 511. (20) Baei, M. T. Covalent Functionalization of Zn12O12 Nanocluster with Thiophene. J. Cluster Sci. 2013, 24, 749−756. (21) Taratula, O.; Galoppini, E.; Wang, D.; Chu, D.; Zhang, Z.; Chen, H.; Saraf, G.; Lu, Y. Binding Studies of Molecular Linkers to ZnO and MgZnO Nanotip Films. J. Phys. Chem. B 2006, 110, 6506− 6515. (22) Moreira, N. H.; Domıinguez, A.; Frauenheim, T.; da Rosa, A. L. On the Stabilization Mechanisms of Organic Functional Groups on ZnO Surfaces. Phys. Chem. Chem. Phys. 2012, 14, 15445−15451. (23) Dargatz, B.; Gonzalez-Julian, J.; Guillon, O. Anomalous Coarsening of Nanocrystalline Zinc Oxide Particles in Humid Air. J. Cryst. Growth 2015, 419, 69−78. (24) Senftle, T. P.; Hong, S.; Islam, M. M.; Kylasa, S. B.; Zheng, Y.; Shin, Y. K.; Junkermeier, C.; Engel-Herbert, R.; Janik, M. J.; Aktulga, H. M.; Verstraelen, T.; Grama, A.; van Duin, A. C. T. The ReaxFF Reactive Force-field: Development, Applications and Future Directions. npj Comput. Mater. 2016, 2, 15011−15025. (25) van Duin, A. C. T.; Dasgupta, S.; Lorant, F.; Goddard, W. A. ReaxFF: A Reactive Force Field for Hydrocarbons. J. Phys. Chem. A 2001, 105, 9396−9409. (26) Raymand, D.; van Duin, A. C. T.; Goddard, W. A.; Hermansson, K.; Spångberg, D. Hydroxylation Structure and Proton Transfer Reactivity at the Zinc Oxide−Water Interface. J. Phys. Chem. C 2011, 115, 8573−8579. (27) Raymand, D.; van Duin, A. C. T.; Spångberg, D.; Goddard, W. A.; Hermansson, K. Water Adsorption on Stepped ZnO Surfaces from MD Simulation. Surf. Sci. 2010, 604, 741−752. (28) Moreira, N. H.; da Rosa, A. L.; Frauenheim, T. Covalent Functionalization of ZnO Surfaces: A Density Functional Tight Binding Study. Appl. Phys. Lett. 2009, 94, 193109. (29) Duke, C. B.; Meyer, R. J.; Paton, A.; Mark, P. Calculation of low-energy-electron-diffraction intensities fromZnO(101̅0). II. Influence of calculational procedure, model potential, and second-layer structural distortions. Phys. Rev. B 1978, 18, 4225−4240. (30) Parker, T. M.; Condon, N. G.; Lindsay, R.; Leibsle, F. M.; Thornton, G. Imaging the Polar (0001) and Non Polar (10-10) Surfaces of ZnO with STM. Surf. Sci. Lett. 1998, 415, L1046−L1050. (31) Jaffe, J. E.; Harrison, N. M.; Hess, A. C. Ab initio study of ZnO (101̅0) surface relaxation. Phys. Rev. B 1994, 49, 11153−11158. (32) Wander, A.; Harrison, N. M. An Ab Initio Study of ZnO(1010). Surf. Sci. 2000, 457, L342−L346. (33) Scarano, D.; Spoto, G.; Bordiga, S.; Zecchina, A.; Lamberti, C. Lateral Interactions in CO Adlayers on Prismatic ZnO Faces: a FTIR and HRTEM Study. Surf. Sci. 1992, 276, 281−298. (34) Raymand, D.; van Duin, A. C. T.; Baudin, M.; Hermansson, K. A Reactive Force Field (ReaxFF) for Zinc Oxide. Surf. Sci. 2008, 602, 1020−1031. (35) Monti, S.; Corozzi, A.; Fristrup, P.; Joshi, K. L.; Shin, Y. K.; Oelschlaeger, P.; van Duin, A. C. T.; Barone, V. Exploring the Conformational and Reactive Dynamics of Biomolecules in Solution Using an Extended Version of the Glycine Reactive Force Field. Phys. Chem. Chem. Phys. 2013, 15, 15062−15077.

(36) Sengul, M. Y.; Randall, C. A.; van Duin, A. C. T. ReaxFF Molecular Dynamics Simulation of Intermolecular Structure Formation in Acetic Acid-Water Mixtures at Elevated Temperatures and Pressures. J. Chem. Phys. 2018, 148, 164506. (37) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual molecular dynamics. J. Mol. Graph. 1996, 14, 33−38. (38) Baerends, E. J.; Ziegler, T.; Atkins, A. J.; Autschbach, J.; Bashford, D.; Baseggio, O.; Brces, A.; Bickelhaupt, F. M.; Bo, C.; Boerritger, P. M.; Cavallo, L.; Daul, C.; Chong, D. P.; Chulhai, D. V.; Deng, L.; Dickson, R. M.; Dieterich, J. M.; Ellis, D. E.; van Faassen, M.; Ghysels, A.; Giammona, A.; van Gisbergen, S. J. A.; Goez, A.; Gtz, A. W.; Gusarov, S.; Harris, F. E.; van den Hoek, P.; Hu, Z.; Jacob, C. R.; Jacobsen, H.; Jensen, L.; Joubert, L.; Kaminski, J. W.; van Kessel, G.; Knig, C.; Kootstra, F.; Kovalenko, A.; Krykunov, M.; van Lenthe, E.; McCormack, D. A.; Michalak, A.; Mitoraj, M.; Morton, S. M.; Neugebauer, J.; Nicu, V. P.; Noodleman, L.; Osinga, V. P.; Patchkovskii, S.; Pavanello, M.; Peeples, C. A.; Philipsen, P. H. T.; Post, D.; Pye, C. C.; Ramanantoanina, H.; Ramos, P.; Ravenek, W.; Rodrguez, J. I.; Ros, P.; Rger, R.; Schipper, P. R. T.; Schlns, D.; van Schoot, H.; Schreckenbach, G.; Seldenthuis, J. S.; Seth, M.; Snijders, J. G. Sol, ADF2017, SCM, Theoretical Chemistry; Vrije Universiteit: Amsterdam, The Netherlands, 2017 https://www.scm.com. (39) Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; DiNola, A.; Haak, J. R. Molecular Dynamics with Coupling to an External Bath. J. Chem. Phys. 1984, 81, 3684−3690. (40) Meyer, B.; Rabaa, H.; Marx, D. Water adsorption on ZnO(1010): from single molecules to partially dissociated monolayers. Phys. Chem. Chem. Phys. 2006, 8, 1513−1520. (41) Meyer, B.; Marx, D.; Dulub, O.; Diebold, U.; Kunat, M.; Langenberg, D.; Wöll, C. Partial Dissociation of Water Leads to Stable Superstructures on the Surface of Zinc Oxide. Angew Chem. Int. Ed. 2004, 43, 6642−6645. (42) Butt, H.-J.; Graf, K.; Kappl, M. Physics and Chemistry of Interfaces, 3rd, Revised and Enlarged; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2013.

H


ReaxFF molecular dynamics study on the influence of temperature on

Recommend Documents