Geometrical structure data of nanoporous carbon systems obtained from computer simulated pyrolysis

This article contains data on nanoporous carbon materials coming from lignocellulosic components. Such data is directly related to the research paper “Insights into the design of carbon electrodes coming from lignocellulosic components pyrolysis with potential application in energy storage devices: A combined in silico and experimental study” [1]. In this work, the geometrical parameters of nanoporous carbon systems were found with Molecular Dynamics (MD) simulations at the ReaxFF level. The tridimensional structures of such carbon systems are given in Cartesian coordinates. They were computed at different heating rates, simulating the conditions observed in pyrolysis processes of Agave angustifolia leaves, which were carried out in a solar furnace. Nanoporous carbon systems are characterized with radial distribution functions (RDF) and ring distribution profiles.


Data
Data contains the structural description and at the atomistic level of nanoporous carbon models obtained from simulated pyrolysis processes. Electronic structure properties may be found in the Supplementary Data of the related research article [1]. Calculations were performed at the ReaxFF level with the heating rates of 0.005, 0.0196 and 0.1 K/fs with simulation timing of 211.5, 175.9 and 165.0 Â 10 3 fs, respectively. Pyrolyzed lignin simulations are closely related to experimental data and their Cartesian coordinates are listed as Supplementary Data in the related research article [1] for a total of 62 systems. Fig. 1 shows the massive molecular models of lignocellulosic components and  Tables 1 and 2 show the parameters used in the Molecular Dynamics simulations at the ReaxFF level. Additionally, the reactive force field used in our Molecular dynamics simulations was included in the Supplementary data of the related research article [1].

Data from Density Functional Theory (DFT) calculations
It is important to denote that Density Functional Theory (DFT) data on the magnitude of the atomic charges was calculated for the molecular models of lignin, cellulose and hemicellulose. Such data were Specifications  [1].

Value of the Data
The nanoporous carbon Cartesian coordinates data may be used as input geometries in the theoretical multiscale modeling of carbon electrodes in energy storage devices. The data related to radial distribution functions may provide insights into the molecular morphology of lignocellulosic components obtained after a biomass pyrolysis process. The structural data may be implemented as substrates to identify ion diffusion pathways that are relevant in the understanding of electrochemical energy storage in carbon electrodes coming from pyrolyzed biomass. The nanoporous carbon tridimensional data may be used as support models to simulate the interactions in heterogeneous catalysis.
presented in Tables S4-S6 of Supplementary Information in the related research article [1]. Such data were obtained using DFT at the PBE/6-31Gþþ [2] level. Additionally, Natural Bond Order (NBO) analysis [3] was performed to localize the electronic charge at each of the atoms in the lignocellulosic components. Such data are relevant to explore possible reactivity in the lignocellulosic molecules. DFT calculations were performed using Gaussian 09 computational code [4]. The data was obtained from fully relaxed structures, which were performed at the same level of theory given above. DFT data were found in vacuum to simulate the conditions of pyrolysis in a heating pot at a solar furnace [1].

Data from Molecular Dynamics (MD) simulations
The simulated pyrolysis of lignocellulosic components was carried out using MD calculations at the ReaxFF level [5] on a massive model of lignin with the Adler's approach [6]. MD calculations were performed using LAMMPS [7] computational code. Moreover, the cellulose and hemicellulose models were implemented in accordance to Jablan et al. [8], Adel et al. [9] and Zhang et al. [10]. The pyrolyzed  systems were nanoporous carbon materials, whose Cartesian coordinates are summarized as Supplementary data in the related research article [1]. Such data were acquired by considering that all components of Adler's lignin model were present during the pyrolysis (Approach 1). Furthermore, Approach 1 was also implemented for the massive model of cellulose and hemicellulose. The initial parameters considered in the computational simulations are all presented in Tables 1 and 2 Data obtained from this pyrolysis are summarized in the RDF profiles of Figs. 3 and 4, respectively.
All MD simulations were performed with the NVT ensemble, and keeping fixed the number N of atoms in a simulation box. In the computations, the equation of motion for all atoms was obtained with a 0.25 Â 10 À15 sec time step. The chemical reactivity through the simulated pyrolysis was modeled with the reactive force field (FF) potential of Kim et al. [11], which is implemented in the ReaxFF [5] module of LAMMPS code [7]. ReaxFF potentials are capable to describe the rising and weakening of chemical bonding with the aid of localized atomic charges. Additionally, ReaxFF introduces the van der Waals (vdW)-type interactions playing a role in the CeC bonding through the simulated pyrolysis. Such attraction is modeled with a distance-corrected Morse potential. In our computations, the vdW interactions were included up to 10 Å. The ReaxFF scheme is also based upon the bond orders (BO) with different atoms to simulate the chemical reactions involved in the pyrolysis processes. Consequently, the total energy of a system modeled with the ReaxFF methodology is described in accordance to: (1) in which E bond , E over , E under , E lp , E val , E tor , E vdW , E Coulomb , correspond to bond energy, overcoordination stability, undercoordination energy, lone pair, valence, torsion, vdW interactions and Coulomb energies, respectively. Additionally, the non-bonding interactions are also contained in the ReaxFF framework, with the introduction of a shielding contribution in the Coulomb and vdW energy terms. Possible discontinuities in the non-bonding energies are avoided with the van Duin's seventh-order taper function [12]. All NVT-MD computations were carried out with a time-step of 0.25 fs by controlling the temperature with the Berendsen [13] thermostat, using a temperature damping of 0.1 ps. Finally, it was also considered that oxygen and hydrogen were released at a limit temperature of 1280K during the simulation (Approach 2). The RDF profiles for those structural data are summarized in Fig. 8, for the Adler's lignin model. The heating rates in all simulations were performed at 0.005, 0.0196 and 0.1 K/fs, starting from 300K up to 1280K during 196000, 50000 and 9800 steps. A second stage in the simulation was performed with constant temperature of 1280K during 50000 simulation steps. Consequently, a quenching rate was performed by 500000 steps until reaching room temperature. The process was followed up by a stabilization stage at room temperature for 100000 simulation steps. The final densities of the nanoporous structure data range from 0.00443 to 0.853 gr/cc.
The estimation of pressure has previously been performed in accordance to previous methodology [14], in which the pressure was evaluated from the van der Waals equation of state for a pure ndodecane model system. In the case of lignin components, no van der Waals constants values (a and b) are available in open literature. The determination of such parameters is out of the scope of the present work and represents a subject of research for future work.        Table 1 Number of atoms considered in the pyrolysis simulation of lignocellulosic precursors.

Precursors
No. of atoms used in the simulations Lignin 2700-21600 Cellulose 720-4560 Hemicellulose 750-9000 Table 2 Simulation regimes for all lignocellulosic precursors of the data.

Transparency document
Transparency document associated with this article can be found in the online version at https:// doi.org/10.1016/j.dib.2019.103874.