SIMULATION OF FEW-LAYER GRAPHENE GROWTH ON THE CU?M (M= CO, NI) ALLOY CATALYST FOR SOLAR CELL APPLICATION USING DENSITY FUNCTIONAL THEORY METHOD

• Main Author: Bhekti Yutomo, Erik
• Format: Dissertations
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/75986
• Institution: Institut Teknologi Bandung

Due to its superior physical properties, few-layer graphene (FLG) is suitable for transparent conductive electrodes (TCE) in solar cells. However, FLG growth with a controlled number of layers is still challenging in chemical vapor deposition (CVD) growth using conventional catalysts. Recently, experimental studies demonstrated that using Cu-based alloy catalysts with other transition metals having high C solubility is an effective technique for growing FLG with a controlled number of layers. However, these experimental studies have not been supported by theoretical studies, so the mechanism of graphene growth on Cu-based alloy catalysts is still not well understood. In this study, the mechanism of graphene growth on a Cu-based alloy catalyst was studied using the density functional theory (DFT) method combined with atomic thermodynamic theory and the ab initio molecular dynamics (AIMD) method. First, the effect of the types of alloying atoms in the Cu?M (M=Co, Ni) alloy catalyst on the graphene growth mechanism was evaluated. In this study, the atomic fraction of M was set at 18.75 at.%. The adsorption energy of the carbon source species on the surface and subsurface of the Cu?Co catalyst is lower than that of the Cu?Ni catalyst. However, the high activation energy makes the C monomer difficult to diffuse to the Cu?Co catalyst subsurface. On the other hand, diffusion of the C monomer to the catalyst subsurface can be activated at a lower energy in the Cu?Ni catalyst. Therefore, the C monomer in the surface is the active species on the Cu?Co catalyst, and the C monomer in the subsurface is the active species on the Cu?Ni catalyst. Thus, FLG can only grow on the Cu?Ni catalyst because a segregation growth mechanism will occur. Besides the types of alloying atoms, the effect of the Ni atomic fraction in the Cu?Ni alloy catalyst on the graphene growth mechanism was also studied. The higher Ni atomic fraction causes the adsorption energy of carbon source species to be reduced. Furthermore, the relative population of carbon source species is highly dependent on the structure of the catalyst. The CH and C monomer on the surface of the catalyst competed to become the dominant species on the Cu?Ni-1 catalyst (6.25 at.% Ni atoms). Therefore, the CH and C monomer on the catalyst surface are the active species on the Cu?Ni catalyst with a low Ni atomic fraction, similar to that of a pure Cu catalyst. On the other hand, the C atom on the catalyst subsurface is the active species in the Cu?Ni alloy catalyst with a high Ni atomic fraction. Parallel to this, diffusion of the C atom into the Cu?Ni-3 (18.75 at.% Ni atoms) catalyst subsurface can be activated at a low energy of 0.16 eV. This energy can be compensated with thermal energy during CVD growth. Therefore, the surface-mediated growth mechanism will occur on a Cu?Ni catalyst with a low Ni concentration, which tends to produce monolayer graphene, and the segregation growth mechanism will occur on a Cu?Ni catalyst with a high Ni concentration (> 18.75 at. % Ni atoms), which tends to produce FLG. Because FLG is predicted to grow only on the Cu?Ni catalyst with a high Ni atomic fraction, the segregation of C adatoms and the initial step of FLG formation were studied using the AIMD method. The C adatoms present in the Cu?Ni catalyst tend to bond with Ni atoms. This tendency is indicated by the stability of the bond overlap population between Ni atoms and C atoms over a long simulation time. Then, almost all of the C adatoms undergo segregation or bulk diffusion processes since ~7 ps. This condition is due to the segregation and bulk diffusion processes in the Cu?Ni catalyst, which can be activated at a lower energy than in the pure Cu catalyst. Ni atoms facilitate bulk diffusion and segregation by forming oriented channels. The formation of C dimer was also observed due to diffusion out of the bulk of C adatoms. C dimer segregates to the interface region between the graphene top layer and Cu?Ni catalyst since ~10 ps and becomes crucial species in initiating nucleation of add-layer graphene. The C adatom segregation process reported here supports the segregation growth mechanism of FLG on the Cu?Ni catalyst proposed by a previous experimental study. In applying FLG as a TCE material, increasing the number of layers causes a slight decrease in the transmittance of FLG. On the other hand, the electronic conductivity of FLG increases significantly when the number of layers is increased. FLG performance for TCE applications will be better than ITO and AZO as conventional TCE materials if the number of layers of FLG is more than three layers. Based on all the results obtained, this study provides important and crucial insights into the effort to grow FLG with a controlled number of layers and apply FLG as a TCE material in solar cells.