Spectrometric technologies are widely used for analytical studies of valuable characteristics of materials, including as an indicator of the levels and related properties of graphene doping. Graphene has engineering appeal due to the high crystalline organization of the material structure. Beams et al. concluded that fine tuning of the added impurities allows for controlling the useful properties of graphene, which creates favorable conditions for modifying the material depending on practical tasks (23). It is also worth specifying that the thermal conductive properties of graphene due to anharmonic phonon-phonon scattering are highly valued in applied engineering. Raman spectroscopy studies the trends of light scattered from atoms and particles and characterizes the polarizability of the bonds between them. In the context of studying the effect of doping on the temperature dependences of graphene material, it is essential to cite the study by Kolesov et al., who studied the differentiation of N-doped, based on SiO2/Si, and pure graphene. The primary conclusion made by the researchers in this work was the non-universal distribution of the charge carriers in the pure and modified graphene. In particular, the authors reported that in pure graphene, adsorption led to an increase in the heterogeneity of the charge carrier distribution — whether holes or electrons — while for doped graphene, the relationship was reversed: the more intense the adsorption, the less heterogeneous is the distribution (“Atmospheric Adsorption on Pristine” 7). More specifically, the decrease in heterogeneity in p-doped graphene was due to electron-hole recombination in the system in an effort to compensate for the local heterogeneity caused by the added nitrogen atoms. By adsorption, Kolesov et al. understood the absorption by graphene of atmospheric air components, including molecular oxygen and water; in other words, doped graphene had greater homogeneity when stored in the air than pure.
An important issue is the determination of temperature-dependent Raman trends as functions of the nature of the graphene used. Graphene can be used in its pure form, grown on substrates, transferred artificially to substrates, and doped with impurity materials — each of these states, according to Kolesov et al., has a number of unique temperature-dependent Raman tendencies (“Phonon Anharmonicities” 192). The positions of the G-peak characterizing the crystallinity of the material and the 2D peak characterizing the layering of graphene were used as measures of such patterns. The main conclusion of this study postulates that the nature of graphene modification — grown or doped — significantly affects the temperature properties of the material. In particular, it was reported that graphene grown on copper had a more substantial bias increase than pure graphene (“Phonon Anharmonicities” 191). In addition, if graphene was doped with SiO2/Si, this maximized the intensity of the D-band, which characterizes the edge disorder of the material; in other words, impurity components increase the defect density for graphene. The choice of the substrate material also has an influence; it was found that the energy change of the E2G mode was about 60% higher for the graphene/copper system than for graphene on a glass substrate (“Low-Temperature Anharmonic” 590). In other words, the choice of the substrate has been demonstrated to be an essential predictor influencing the temperature-dependent behavior of the propagating quasi-particle lattice oscillations and therefore has value for studying the thermal conductivity properties of graphene used as an applied layer.
The vast possibilities of modifying the substrate on which the graphene layer is applied create temperature-dependent patterns of the G-band energy distribution in the spectrogram of the material. Kolesov et al. aimed their work at studying these dependencies for copper as one of the most popular materials of industrial metallurgy, as well as for glass. The research parameter was the position of the G-band, which in classical Raman spectroscopy characterizes the measure of crystallinity of the material. The authors found that with increasing temperature, the Raman shift of the G-band showed a linear downward dependence — in other words, with increasing temperature, there was a change in the planes of the vibrational mode (“Low-Temperature Anharmonic” 589). Several potential causes of this change were named, including phonon and electron-phonon contributions, as well as the resulting deformation due to differences in the thermal expansions of the graphene and the substrate. It is noteworthy that no similar effect was found for the “graphene/glass” system, and the G-band position was constant or changed insignificantly as the temperature increased. The deformations also occurred when growing graphene on copper but were not observed for the material transferred to copper, which means that the mechanism of substrate formation —growing or transfer — significantly affects the intensity of the G- and D- bands, as well as their overtones, which in turn can lead to the formation of undesirable conducting effects. Interestingly, the authors also cite the difference in the temperature expansion coefficients of the graphene and the substrate material as the reason for the formation of such a deformation (“Phonon Anharmonicities” 191). The anharmonic constants for the substrates made of aluminum oxide and SiO2/Si were also investigated: the research result postulates that the change in the G and D peaks is significantly lower than the similar changes for the system grown on copper. Chemical interactions between the carbon atoms and copper during the growing process, as well as thermal deformations resulting from cooling of the system during its synthesis, are cited as the probable cause of this effect. More specifically, the interaction energy of graphene and copper during growth was about 60% higher than the energy for graphene and SiO2/Si, and graphene and Al2O3 (“Phonon Anharmonicities” 193). Thus, it was found that the nature of graphene modification plays an essential role in shaping the useful electrical and thermal conductive properties of the material, so it is critical to be selective in the choice of modification mechanism.
Beams et al. used Raman shift principles to determine the defect and doping properties of graphene depending on the type of impurities added. The general conclusion reached by the authors is that the presence of defects in the graphene material, which includes chipping, uneven surfaces, roughness, and areas of depression, can have an ambiguous effect on the useful characteristics of the material, including the semiconductor conductivity (Beams et al. 23). On the one hand, defects have traditionally been perceived as problematic areas preventing the formation of useful application properties. On the other hand, the authors found that for some devices — for the use of graphene in gas sensors — the presence of defects determines the increase of useful properties. These results may correlate with the findings of Kolesov et al., who found that the roughness of copper and glass substrates had no statistically significant effect on the Raman shift G-band (“Low-Temperature Anharmonic” 588). It is worth clarifying that defect formation is often a product of doping but can also be realized for other reasons. In particular, Kolesov et al. studied the effect of doping time on the functionality of modified graphene. It was shown that the time of 90 seconds was ineffective since it led to more defects compared to 60 seconds. It is worth noting that the occurrence of defects with increasing doping time was not due to doping materials but the removal of nitrogen atoms from the graphene (“Atmospheric Adsorption on Pristine” 4). These results lead to the conclusion that two-dimensional doping is extremely sensitive for graphene, which means that the doping processes should be finely controlled and take into account the effects of electron-hole recombination.
Among the additional results of the literature search should be noted that Beams et al. found a violation of the symmetry of the graphene intensity distribution if the material had defects. In the corresponding spectrogram, additional peaks were observed in the region of D-bands and G-bands, probably due to either overtones or Raman modes (Beams et al. 13). The effect on these bands was also studied by Kolesov et al., who offered equally intriguing results in their study of atmospheric adsorption on pure and doped graphene. More specifically, doping with atomic nitrogen was found to intensify the G-peak and D-peak widths (“Atmospheric Adsorption on Pristine” 3). It follows that nitrogen doping increases the probability of defects in the graphene composition, which can but not necessarily have a negative impact on the valuable properties of the conducting material. Under such doping, the hole conductivity is dominated in graphene, and the hole concentration increases smoothly with the addition of nitrogen. In addition, it was found that the energy of the ground state of graphene — Fermi level — tends to increase with doping: the difference is found both with the state without impurities and with p-doped graphene, and the energy difference is maximal just with hole doping (Beams et al. 16). The increase in this parameter can indicate the level of bonding strength between the electrons, which reports the conductive properties of graphene.
Works Cited
Beams, Ryan, Luiz Gustavo Cançado, and Lukas Novotny. “Raman Characterization of Defects and Dopants in Graphene.” Journal of Physics: Condensed Matter, vol. 27, no. 8, 2015, pp. 1-26.
Kolesov, E. A., et al. “Low-Temperature Anharmonic Phonon Properties of Supported Graphene.” Carbon, vol. 111, 2017, pp. 587-591.
Kolesov, Egor A., et al. “Atmospheric Adsorption on Pristine and Nitrogen-Doped Graphene: Doping-Dependent, Spatially Selective.” Journal of Physics D: Applied Physics, vol. 53, no. 4, 2019, pp. 1-9.
Kolesov, Egor A., et al. “Phonon Anharmonicities in Supported Graphene.” Carbon, vol. 141, 2019, pp. 190-197.