Supplementary information for Optimizing capacitance performance: Solar pyrolysis of lignocellulosic biomass for homogeneous porosity in carbon production [Dataset]
S0.1.1. Simulations with Molecular Dynamics: In the corresponding simulation box, the NVT ensemble was used to calculate a fixed number of N atoms. To maintain this fixed number, the isobaric ensemble was also considered. During the simulations, a time step of 0.25 ×10−15 s was implemented. The Reax...
| Autores: | , , , , , , , , , , |
|---|---|
| Tipo de recurso: | conjunto de datos |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Consejo Superior de Investigaciones Científicas (CSIC) |
| Repositorio: | DIGITAL.CSIC. Repositorio Institucional del CSIC |
| OAI Identifier: | oai:digital.csic.es:10261/356089 |
| Acceso en línea: | http://hdl.handle.net/10261/356089 |
| Access Level: | acceso abierto |
| Palabra clave: | Agave Molecular dynamics simulations Radial distribution function http://metadata.un.org/sdg/7 Ensure access to affordable, reliable, sustainable and modern energy for all Ensure sustainable consumption and production patterns hemicellulose lignocellulose |
| Sumario: | S0.1.1. Simulations with Molecular Dynamics: In the corresponding simulation box, the NVT ensemble was used to calculate a fixed number of N atoms. To maintain this fixed number, the isobaric ensemble was also considered. During the simulations, a time step of 0.25 ×10−15 s was implemented. The ReaxFF methodology incorporates a reactive force field (FF) to describe the chemical interactions in each system. In the present work, the FF developed by Kim et al [1] was applied, in which the interactions of the lignocellulosic components, namely carbon, hydrogen, and oxygen are considered. This FF was implemented due to its suitability for analogous lignocellulosic systems [2–4]. The MD calculations were performed with the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) [5] computational code and the Reax-module. The interactions among carbon atoms and the possible reactivity with the rest of the components were studied with a distance-corrected Morse potential [6]. This potential incorporates the Van der Waals (vdW) dispersion description and the computation of the bond orders (BO) to evaluate chemical interactions through the pyrolysis process. In the computational simulation models, the total energy of the system was calculated by Eq. 1: Esystem = Ebond + Eover + Eunder + Elp + Eval + Etor + EvdW + ECoulomb (1). In Eq. 1, Ebond corresponds to the bonding energy, Eover refers to the overcoordination stability, Eunder includes the energy to regulate the energy of undercoordinated atoms, and the Elp term is the lone pair energy. The Eval is termed the valence energy. The torsional energy among the atomic components inside the system is Etor, while EvdW includes the description of the vdW interactions. Additionally, the Coulomb energy is given by the ECoulomb term in Eq. 1. Those atomic interactions with non-bonding nature are fully described with a seventh-order function [7]. The morphology of the porous structure was theoretically modeled by computing the pore size distribution (PSD) of the final carbon materials resulting from the simulated pyrolysis. The calculations were performed with the Zeo++ computational code [8], which is intended to map the geometry of the pores via the Voronoi decomposition. The method incorporates a probe molecule with a radius size of 1.79˚A to evaluate the volume accessible to this probe. This radius corresponds to the approximate size of a nitrogen molecule, which is used in the physisorption experiments. The cellulose and hemicellulose precursors were studied as polymeric models in accordance to Martínez-Casillas et al [4]. Such molecular components were incorporated in the present work due to their successful incorporation into analogous biomass systems. The massive model systems were formed by tracking the specific rate contents for each lignocellulosic component with the Packmol code [9]. It produces random distributions in a closed simulation box with a side size of 140˚A, which is shown in Fig. 2 of the main text. The model systems of the agave fibers BA and BAC are based upon the experimental data, in which the following contents were found: 17% lignin, 38% cellulose, and 24% hemicellulose for BA, while for the latter, 13%, 44% and 10% for lignin, cellulose and hemicellulose respectively for BAC. The remaining components are byproducts, mainly ashes, which are not considered in the present calculations. Such compositions were based on the NREL characterization method implemented in the experiments [10]. To fulfill the total content of the sample, the molecular models were further normalized. The molecular weights of each model system were introduced to build the massive molecular system depicted in Fig. 2 (a), depicted in the main text. The lignin Adler’s model was set as the reference system, with a molecular weight of 2407.10 g/mol. The simulation box comprised 40 macromolecules and an accumulated molecular weight of 96,284 g/mol. It was then fixed as 24% of the complete molecular weight of the lignocellulosic components, after normalization. The polymer models for cellulose and hemicellulose corresponded to normalized contents of 45% and 31% [11]. Consequently, molecular weights of 180,532.5 g/mol and 124,366.83 g/mol for these two components, respectively, were considered for the BA model. In terms of polymer units, 69 hemicellulose and 109 cellulose units were introduced. The BAC model was also normalized by using the same criteria. The pyrolysis processes were evaluated with two different heating rates; namely, at 0.027 K/fs, and also at a slower rate of 0.005 K/fs for selected cases. The top temperature at which. the pyrolysis simulations were performed corresponds to the temperatures of 500 °C and 700 °C, which were reached with the experimental procedure [3, 4]. These temperatures remained constant upon annealing during 100,000 steps with a step size of 0.25 fs. When this stage of the simulation is finalized, a quenching process was applied to meet the initial room temperature of 300 K. A subsequent series of 50,000 MD steps at this fixed temperature evolved the system to reach a thermal equilibrium. It is worth mentioning that a methodology to compute the pressure values in annealing conditions, as that developed byWang et al [12] was impossible to incorporate in the simulations, since the vdW equation of state was applied. S0.1.2. Radial distribution function computation The X-ray diffraction data obtained between 2θ of 1° to 120° was used to compute the radial distribution function (RDF) that is aimed to elucidate the crystal structure of solids or amorphous materials [13]. The RDF is given in accordance with the eq. 2: G(r) = 4πrρ0[g(r) − 1] (2) in which r stands for the radial distance, and ρ0 is the atomic number density in average. Finally, the term g(r) corresponds to the macroscopic atomic pair density [13]. The physical significance of eq. 2 corresponds to the distribution of probability to find atomic pairs at a certain distance r. Moreover, G(r) is further assessed with a sine Fourier transform. It refers to the reciprocal space of the structure factor S(Q), containing the scattering structure distribution that comes from the experimental XRD data. Using the structure factor, Eq. 2 is re-written as: G(r) = 2/π Z ∞ 0 Q[S(Q) − 1] sin (Qr)dQ (3) Upon combination of eq. 2 and 3, the RDF can be obtained as given by g(r). Such values were found with the aid of the PDFGETX2 computational code [14]. The G(r) function was assessed via a Fourier transform using the structure factor S(Q) as input. A value of 16.0˚A−1 was used as the Qmax parameter [1] to apply the Fourier transform. As a final statistical treatment, the PDFGUI code was implemented to adjust the experimental RDF. The PDFGUI is a software with a user graphical interface based upon the PDFFIT2 computational code [15]. The software requires a starting guess to refine the structural data such as atomic positions, lattice constants, correlated atomic motion, and anisotropic atomic displacement. S0.1.3. Envelope Density assessment The envelope density methodology [16] was used to evaluate the densities of the final carbon materials. This study was intended to compare the densities given in the experiment with those computed with the theoretical approach. The ratio of the carbon material to the total volume is defined as the envelope density, and it was used in the experimental analysis. It is worth mentioning that a volume of 0.2mL was considered in the evaluation, which is given by Corning Inc. PCR R®tubes.-- Under a Creative Commons license BY-NC-ND 4.0 |
|---|