Ab Initio Molecular Dynamics and Lattice ... - ACS Publications

Mar 12, 2018 - readily combined with force field (FF) models for other materials, such as .... r r. q q r. 4. 4 i. N j i ij ij ij ij ij. i j ij nonbon...
5 downloads 5 Views 861KB Size
Subscriber access provided by UNIV OF DURHAM

Surfaces, Interfaces, and Catalysis; Physical Properties of Nanomaterials and Materials

Ab Initio Molecular Dynamics and Lattice Dynamics Based Force Field for Modeling Hexagonal Boron Nitride in Mechanical and Interfacial Applications Ananth Govind Rajan, Michael S. Strano, and Daniel Blankschtein J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.7b03443 • Publication Date (Web): 12 Mar 2018 Downloaded from http://pubs.acs.org on March 13, 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 23 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 Letters

Ab Initio Molecular Dynamics and Lattice Dynamics Based Force Field for Modeling Hexagonal Boron Nitride in Mechanical and Interfacial Applications Ananth Govind Rajan1, Michael S. Strano1, and Daniel Blankschtein1* 1

Department of Chemical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States *Corresponding Author: Daniel Blankschtein (E-mail: [email protected]) ABSTRACT

Hexagonal boron nitride (hBN) is an upcoming 2D material, with applications in electronic devices, tribology, and separation membranes. Herein, we utilize density-functional-theory-based ab initio molecular dynamics (MD) simulations and lattice dynamics calculations to develop a classical force field (FF) for modeling hBN. The FF predicts the crystal structure, elastic constants, and phonon dispersion relation of hBN with good accuracy, and exhibits remarkable agreement with the interlayer binding energy predicted by random phase approximation calculations. We demonstrate the importance of including Coulombic interactions but excluding 1–4 intrasheet interactions to obtain the correct phonon dispersion relation. We find that improper dihedrals do not modify the bulk mechanical properties and the extent of thermal vibrations in hBN, although they impact its flexural rigidity. Combining the FF with the accurate

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry Letters 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 23

TIP4P/Ice water model yields excellent agreement with interaction energies predicted by quantum Monte Carlo calculations. Our FF should enable an accurate description of hBN interfaces in classical MD simulations.

TOC Graphic

ACS Paragon Plus Environment

2

Page 3 of 23 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 Letters

Hexagonal boron nitride (hBN), the inorganic analogue of graphene, is an upcoming twodimensional (2D) material, consisting of boron (B) and nitrogen (N) atoms arranged in a hexagonal lattice.1 Researchers are interested in exploiting the different friction properties2,3 and extents of interaction of ions with graphene and hBN surfaces,4,5 to enable osmotic power generation.6 Recently proposed applications of hBN also include gas separation7 and seawater desalination8 membranes made of rolled-up boron nitride nanotubes (BNNTs) or few-layered nanoporous hBN surfaces. In these applications, the mechanical and interfacial properties of hBN play an important role in modulating device and process performance and robustness. For example, the use of nanoporous hBN as a membrane for desalination, similar to graphene,9 is contingent upon the hBN surface: (i) maintaining mechanical stability over extended periods of time, and (ii) interacting differently with water and ions.10 Similarly, understanding the exfoliation of bulk hBN crystals into few-layered sheets,11,12 mediated by the presence of solvents, requires models which can adequately describe both the intralayer and interlayer interactions in hBN. Classical molecular dynamics (MD) simulations, compared to ab initio MD (AIMD) simulations, offer a computational framework to model and understand the mechanical and interfacial properties of materials, up to relatively longer length (tens of nanometers) and time (tens of nanoseconds) scales. However, there is a lack of available classical MD force fields, which adequately account for the Coulombic and interlayer interactions in hBN, while providing nonbonded (dispersion) interactions which can be readily combined with force field (FF) models for other materials, such as, liquids, gases, and other 2D materials.

The most popular FF used to model the intralayer (bonded) interactions in hBN is the modified Tersoff potential14 adapted by several authors in the literature.15,16 Although Tersoff-like

ACS Paragon Plus Environment

3

The Journal of Physical Chemistry Letters 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 23

potentials can represent the mechanical properties of solids quite well, they have some deficiencies which may be remedied. Specifically, they: (i)

lack Coulombic interactions, which are important for describing the interactions of polar molecules with hBN surfaces, e.g., in separation membranes and liquid-phase exfoliation applications. Moreover, the differences in electronegativities of B and N imply that there is some charge separation in hBN, which is not captured by Tersoff-like potentials.

(ii)

do not account for interlayer dispersion forces which dominantly hold together the layers of hBN in the bulk solid.17 Therefore, Tersoff-based models, by themselves, are inadequate for simulating multi-layered hBN and multi-walled BNNT systems, which require describing how the layers of hBN interact with each other.

In terms of describing hBN-adsorbate interactions, most studies in the literature have utilized the nonbonded force-field parameters for B and N atoms proposed by Mayo et al. in the generic DREIDING FF,18 along with combining rules. However, the DREIDING parameters were developed about 3 decades ago and the interaction energies that they yield with adsorbate (most importantly, water) molecules have not been benchmarked against advanced ab initio calculations, e.g., the quantum Monte Carlo (QMC) dataset by Al-Hamdani et al.19 Nevertheless, the DREIDING parameters have been used to model the interactions of flat and curved hBN surfaces with liquid molecules, in a number of applications, including: (i) the desalination performance of single layer nanoporous hBN membranes,8 (ii) liquid-phase exfoliation of fewlayered hBN by conventional solvents, such as, dimethyl sulfoxide (DMSO) and isopropyl alcohol (IPA),20 as well as by ionic liquids,21 (iii) the flow of water22,23 and ions10 through BN nanotubes, and (iv) the encapsulation of C60 buckyballs by BN nanotubes.24 Therefore, there is an urgent need to develop force-field parameters, specifically for B and N atoms in hBN, and

ACS Paragon Plus Environment

4

Page 5 of 23 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 Letters

benchmark them against high-quality ab initio data. In terms of describing interlayer interactions in hBN, Hod et al. recently proposed a potential for modeling the interactions between the layers in hBN.25 However, this FF lacks Lennard-Jones parameters, for describing nonbonded dispersion interactions, which may be readily combined with other FF models. In view of the above challenges regarding the use of currently available FF models in simulations of hBN interfaces, in this work, we focus on the development of a new transferrable FF model for modeling the mechanical and interfacial properties of hBN, including obtaining physical insight from the resulting FF. Figure 1(A) depicts the molecular model of an hBN monolayer, including the equilibrium bond lengths and angles. The proposed classical model for hBN includes both intralayer (bonded),  , and interlayer (nonbonded),  , interactions as follows:

 =  +  



!"#$





(1)

1 1 =   −   +   2 2  +



'()*)$*



!8( 7





 



1  & − &  2 %  3



005 5= −4,- ./ 2 − / 2 4 + 1146, 1

-

(2)

(3)

where  is the potential energy of the system, 9 , 9:; , 9=> , and 9:?