ReaxFF model

Reactive molecular dynamics (ReaxMD) is based on the Reax force field (ReaxFF) parameterized by fitting to a training set of QM data. Compared to first-principles simulations, with the ReaxFF model it is possible to speed up the calculation by several orders of magnitude. However, due to the enormous complexity of the underlying mathematical expressions, ReaxFF is -10 to 100 times more expensive computationally than simple (nonreactive) FFs. 

Compared to AIMD, with the ReaxFF model, it is possible to speed up the MD calculation by a factor of a million, allowing simulations of the bulk stmctures and chemistry of molecular heterogeneous systems with up to 1 billion atoms in massively parallel simulations. One of the most important features of the ReaxFF reaaive FF is a capability to describe chemical reactions in composite systems with a wide variety of constituents, including polymers, metals, inorganic fillers, and so on. It has been successfully used to attack problems that require accurate modeling of chemistry, including oxidation,... 

Buehler et al. presented a preliminary study on formation of water from molecular oxygen and hydrogen using a series of atomistic simulations based on ReaxFF MD method.111 They described the dynamics of water formation at a Pt catalyst. By performing this series of studies, we obtain statistically meaningful trajectories that permit to derive the reaction rate constants of water formation. However, the method requires calibrations with either ab initio simulation results in order to correctly evaluate the energetics of OER on Pt. Thus, this method is system specific and less reliable than the ab initio methods and will not replace ab initio methods. Nevertheless, this work demonstrates that atomistic simulation to continuum description can be linked with the ReaxFF MD in a hierarchical multiscale model. 

Goddard et al.18 are currently carrying out exploratory calculations on both OER and proton conduction processes using the ReaXFF MD method. They anticipate that this will establish a very realistic MD model of a tiny fuel cell and will allow determining how the performance changes as the composition, configuration, and other conditions are altered. 

In this chapter we review our recent efforts towards understanding many of the salient features of detonation using NEMD simulations. We will focus on large-scale NEMD simulations using a model interatomic potential (denoted REBO) to study generic, but complex, detonation phenomena and the use of a new, computationally more intensive, potential (denoted ReaxFF) that accurately describes a real nitramine energetic material. 

The results presented in Fig. 4 show the evolution of the number of adsorbed benzene in the different system as a function of the simulation time. The ReaxFF force field allows the creation and breaking of covalent bonds between the different atoms of the system during the molecular dynamics simulation. In fliis work, we considered that a benzene molecule was adsorbed, if it formed at least one bond with the Ni (100), Ni (111), or Raney-Nickel surface, respectively. In the first 5 ps, the benzene adsorption is comparable for all three systems evaluated, i.e. both clean Ni surfaces and Raney Ni model. After the first 5 ps, about 6-7 % of benzene molecules have been adsorbed. In the very beginning of the simulation time, the adsorption process is even faster on Ni (100) and Ni (111) surface (blue and green line in Fig. 4) compared to the catalyst (purple line). After the first few ps, the adsorption on Ni (100) surface (blue line in Fig. 4) remains rather constant and does not increase much. After 25 ps, only about 12 % of benzene have been adsorbed. In contrast to this finding, the benzene adsorption on the Ni (111) surface and the Raney Ni model system (green and purple line in Fig. 4) increases more... 

Fig. 3 Snapshot of a ReaxFF simulation of benzene adsorption in Raney Ni. The model stmcture of Raney Ni was generated as described above for an initial 50 wt% Ni/Al composition...

ReaxFF. ReaxFF allows for bond breaking and bond formation in MD simulation so that thermal decomposition can be modeled as has been shown recently for polydi-methylsiloxane [9]. ReaxxFF includes terms for traditional bonded potentials as well as nonbonded potentials (i.e., van der Waals and Coulombic). Bond breaking and bond formation are handled through a bond order/bond distance relationship. Parameterization is through high-level DFT calculations (B3LYP/6-311++G ). 

Fig. 1 Hierarchical multiscale, multiparadigm approach to materials modeling, from QM to the mesoscale, incorporating breakthrough methods to handle complex chemical processes (eFF, ReaxFF). Adapted from our multiscale group site http //www.wag.caltech.edu/multiscale...

ReaxFF, a reactive force-field approach, has been used to model the iminium-enamine conversion in the proline-catalysed self-aldol of propanol. " " Quantum mechanical methods have been used to study the same step in the proline-catalysed aldol. " ... 

Two-body potentials are typically a good approximation of neutral atoms such as noble gases, which are dominated by attractive van der Waals forces at separations and with strong repulsion at close distances due to Pauli s exclusion principle. The electronic distributions of other systems, such as covalently bonded materials, are modeled more accurately by more complicated environmentally dependent semiempirical potentials. These include bond-order potentials like Tersoff [11], Brenner [12], and ReaxFF [13] and embedded atom model (EAM) [14] potentials, which are particularly applicable to metallic systems. 

Our theoretical investigation regarding the understanding of the conversion of iminium into enamine in the framework of a proline-catalyzed aldol reaction emphasizes that the reactive force field (FF), ReaxFF, used in combination with molecular dynamics (MD) simulations is a relevant method to investigate the mechanism of proton transfers in iminium-enamine conversions. This approach should be extended to model other steps of proline-catalyzed... 

This chapter will focus on the modeling of MEA and its polymer electrolyte membrane. First, 3D modeling of PEMFC and its MEA will be discussed, and an example will be put forward. Then, dynamic modeling of PEM will be introduced. Further, this chapter will move on to the fault-embedded modeling of PEM. As an extension, application of membranes in other cases will be recommended, such as in lithium battery, vanadium redox flow battery (VRFB), chlor-alkali electrolysis, water electrolysis, and solar cell. Finally, several typical examples will be given, including Pt and Pt alloy simulation with density functional theory (DFT), water formation and Pt adsorption on carbon reactive force field (ReaxFF) simulation, and coarse-grained simulations. 

FIGURE 12.22 ReaxFF mesoscale model cartoon that simulates a whole fuel cell consists of Nafion membrane (light gray), anode (top, dark gray) and cathode (bottom, dark gray particles), and carbon support (black) particles. 

The introduction of advanced classical MD simulation techniques utilizing new force field methods designed to bridge the gap between QM and classical MD methods at the atomistic scale have recently shown notable promise for a wide range of complex materials modelling. Emerging capabilities for treating nanostructured materials in lET applications is discussed here in terms of so-called reactive force fields (as implemented in ReaxFF) - one of the most actively developed approaches for this purpose in the last few years." " ... 

---