Open Access
Add to BibSonomy
Abstract
GaN-based films grown on sp3-bonded single-crystalline substrates can maintain the coherent growth character. However, exfoliating III-nitride films from these substrates proves difficult because of the strong sp3-type covalent bonds between the substrates and epilayers. The sp2-bonded two-dimensional (2D) materials exhibit hexagonal in-plane lattice arrangements and weakly bonded layers, so the GaN epilayer grown on 2D materials can be transferred onto foreign substrates with ease. In this paper, graphene is used as the inserted layer (IL) on freestanding GaN substrate and the van der Waals coherent epitaxy of GaN-based single-crystalline films on such graphene/GaN templates is investigated. Density functional theory computations are performed to probe the transmission of crystallographic information of wurtzite GaN through the graphene IL. The appropriate layer numbers of graphene IL and GaN growth temperature are optimized to demonstrate the coherent epitaxy character. Both theoretical and experimental results support that the coherent epitaxy of GaN can only be achieved by using a monolayer graphene IL, and the crystalline quality of optimized GaN film can reach the same level of that grown directly on GaN freestanding substrates.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Growing GaN-based single-crystalline films generally requires a well-defined global epitaxial relationship between the film and the substrate [1,2]. At present, sp3-bonded single-crystalline substrates (e.g. sapphire, Si, SiC, etc.) have mainly been used as the substrates in commercial processes [3–6]. However, heteroepitaxy of III-nitrides will still introduce high dislocation density resulting from the lattice and thermal mismatch between epilayer and substrate [7]. Besides, freestanding GaN substrates have not been widely used commercially because of the high cost [8,9]. Materials with sp2-bonded two-dimensional layered structures, such as graphene, possess excellent thermal and chemical stability and an in-plane lattice arrangement similar to that of III-nitrides [10,11]. The weak van der Waals (vdW) interaction between the sp2-bonded graphene and III-nitride epilayer can forgoes the requirements of lattice matching for homo- or hetero-epitaxy [3]. What’s more, freestanding GaN substrates below graphene can still interact with the GaN-based epilayer due to the so-called wetting transparency documented for graphene [12,13]. Hence, growing GaN-based films on freestanding GaN substrates via graphene inserted layer (IL) is a feasible method for fabricating high-performance device. Such vdW coherent epitaxy not only maintains the high quality of epilayers with low dislocation density the same to the homogeneous substrate, but also has potential for releasing and transferring the GaN-based epilayer onto foreign substrates preferred by device fabrication or application through mechanical exfoliation [14].
J. Kim et al. first demonstrated that single-crystalline thin films can be synthesized on 2D-material-coated substrates due to the remote atomic interaction through 2D materials [15]. The penetration of the potential fluctuation in GaN substrates was simulated using the density functional theory (DFT) calculations and only mono- or bi-layer graphene sheets were considered to allow the remote interaction penetration [15]. However, bi-layer graphene sheets seem to isolate the transmission of crystallographic information from quartz substrate to GaN epilayer to a considerable extent according to our recent results [16]. Even single-crystalline GaN thin films can also be obtained on amorphous substrate (such as silica) through the graphene IL [17–19]. Therefore, the direct evidence of the coherent epitaxy character at the interface of GaN substrate and GaN epitaxial layer, such as high-resolution transmission electron microscopy (HRTEM) image, is needed for further confirmation of the remote interaction. T. Malinauskas et al. presented the growth of a 2.4-µm-thick GaN epilayer using a GaN/sapphire template covered with monolayer graphene [20]. Multi-step growth temperature protocols were employed to promote the crystalline quality of the GaN epilayer. However, there is a lack of rationale for selecting monolayer graphene in the GaN remote epitaxy. Besides, the combination of small-size (1.3 × 0.8 cm2) monolayer graphene sheet and large-scale (2-inch) sapphire substrate may lead to epitaxial lateral overgrowth of GaN, in which monolayer graphene sheet can also act as a mask. Hence it is difficult to ensure the pure coherent epitaxy character between GaN/sapphire template and GaN epilayer. Therefore, it is necessary to confirm the appropriate number of graphene sheets and the coherent epitaxy character in theory and experiment.
In this work, vdW coherent epitaxy of GaN-based films has been investigated theoretically and experimentally. Density functional theory (DFT) are employed for investigating the substrate-epilayer remote interaction with different layers of graphene ILs. It has been proven that only monolayer graphene IL allows the transmission of crystallographic information from substrate to adatoms completely. GaN growth temperature is also optimized to obtain GaN film with high quality in the vdW coherent epitaxy using metal organic chemical vapor deposition (MOCVD). The experimental results from electron backscatter diffraction (EBSD) and transmission electron microscope (TEM) confirm the prediction of coherent epitaxy growing GaN (001) on GaN (001) substrate via monolayer graphene IL. And high-quality blue InGaN/GaN multi-quantum-well (MQW) is also demonstrated and characterized by high-resolution X-ray diffractometer (HRXRD) and photoluminescence (PL).
2. Method
2.1 DFT simulations
The DFT calculations were performed to probe the transmission of crystallographic information from GaN substrate to GaN epilayer. The coherent interaction between GaN-GaN (substrate-epilayer) was explored using the Vienna Ab initio Simulation Package (VASP) with the generalized gradient approximation (GGA) and Perdew, Burke, and Ernzerh (PBE) function of the hybrid function [21,22]. Each primitive cells of hexagonal wurtzite GaN contains 2 Ga atoms and 2 N atoms. The orbit of Ga 3d electrons was treated as a core state using the projector-augmented wave method (PAW) with a 400-eV kinetic-energy cutoff [16,23]. The Monkhorst and Pack scheme of k-point sampling was used for integration over the Brillouin zone [24]. The k-point sampling grid was 2 × 2 × 4 for a primitive cell. Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm was used during the optimization of the initial geometry [25]. Conjugate gradient (CG) algorithm was applied to calculate the minimum value of the total energy [26].
The following convergence parameters were set: < 1 × 10−5 eV/atom energy change, < 0.005 eV/nm force, and < 2 × 10−6 eV/atom convergence tolerance of single atomic energy. Planar averaged electron density was calculated as a function of the distance between GaN substrate and GaN epilayer.
2.2 Growth details
To verify the prediction that vdW coherent epitaxy of GaN on freestanding GaN substrate is possible only via monolayer graphene IL, monolayer, bilayer and trilayer graphene sheets were fabricated on 4-inch Cu foil by CVD synthesis. The synthesis temperature, synthesis pressure, precursor, and carrier gas were 1050 °C, 104 Pa, CH4, and H2, respectively. The layer number of graphene sheets was determined by the synthesis time. Then wet transfer technique was used to transfer the graphene sheets onto Nanowin commercial freestanding GaN (001) substrates (10.0 × 10.5 mm2). As the substrates have been received polished to a vendor-defined “epi-ready” surface, no additional cleaning procedures were undertaken prior to transfer. The PMMA solution of 30 mg/mL was spun onto the graphene/Cu substrate, which was prepared using PMMA powder and methylbenzene. And the ammonium persulfate ((NH4)2S2O8) solution of 4 g/mL and acetone (CH3-CO-CH3) are used to remove the Cu foil and PMMA, respectively. Annealing of transferred CVD graphene on GaN substrates was performed at 300 °C for 30 min in N2 ambient to promote better adhesion.
All the samples were grown on such graphene/GaN (001) substrates (10.0 × 10.5 mm2) using an AIXTRON 2000HT MOCVD system. Trimethylgallium (TMGa) and ammonia (NH3) were used as Ga and N precursors for GaN growth, respectively. Triethylgallium (TEGa), trimethylindium (TMIn), and NH3 were used as Ga, In and N precursors for InGaN/GaN MQW growth, respectively. The whole growth process was monitored by an in-situ filmetrics system at 650 nm.
2.3 Characterization
Raman measurements of as-transferred graphene IL were performed using a confocal Raman microscope (Horiba HR-800). Raman spectra were measured using a 514.5 nm laser excitation source. The laser beam was focused to a 2-µm diameter spot on the sample surface. The crystalline characterization was performed using EBSD (Zeiss Merlin) and XRD (Philips X’Pert MRD). For microstructural characterization, a cross-sectional TEM (XTEM) observation of the sample was implemented using the in-situ FIB lift-out technique on a FEI Strata 400 Dual Beam FIB/SEM. The sample was coated with sputtered Ir and E-Pt/I-Pt prior to milling. The microstructure was investigated using a TEM (FEI Tecnai TF-20 FEG/TEM) operated at 200 kV in the high-resolution (HR) TEM mode. The surface morphology was studied by an optical microscope (Olympus BX51M). The PL of the samples was measured using a CW He-Cd laser (325 nm) for excitation with a 4 W/cm2 power density. The luminescence signal was analyzed by a monochromator (Jobin Yvon Triax 550) and detected by a GaAs photomultiplier (Hamamatsu R928). All the measurements were performed at room temperature.
3. Results and discussion
A supercell model is shown in Fig. 1(a) and there is a gap between GaN substrate (left side) and GaN epilayer (right side). Figure 1(b) illustrates the charge density between the GaN (001) substrate and a GaN (001) epilayer separated by different gaps. Significant charge density is seen between the separated GaN slabs. When the gap is increased beyond about 4.6 Å, the charge density drops to one tenth of its maximum. Therefore, we believe that there is no significant interaction between GaN substrate and GaN epilayer in this case. This demonstrates a vdW interaction between the slabs, and suggests that coherent epitaxy is possible within a 4.6-Å gap between GaN substrate and GaN epilayer. According to our previous results in Ref. [16], the total gap between GaN/monolayer graphene IL/GaN heterostructure (d1 + d2 in Fig. 2(a)) is not more than 4.3 Å, while the total gap between GaN/bilayer graphene IL/GaN heterostructure (d1 + d2 + d3 in Fig. 2(b)) is not less than 7.1 Å. In practice, interaction between GaAs slabs may be damped by the vertical vdW force exerted by interlayer graphene, although it is about an order of magnitude weaker than that of covalent interactions [12]. Thus, the true charge interaction gap between the substrate and epilayer may be less than that estimated from the calculations [12]. Hence, only monolayer graphene IL shows potential to allow the transmission of crystallographic information from GaN substrate to GaN epilayer in principle. Using the same criterion, similar result can also be drawn from Fig. 1(f) in Ref. [15].
Fig. 1. (a) Schematic presentation of a supercell model in the DFT calculations. Periodic boundary conditions are imposed along the dashed lines of the simulation model. (b) Calculation results of averaged electron density along separated slabs of GaN with different interaction gaps, which shows the existence of significant electron charge density between the separated slabs within a gap of about 4.6 Å.
Fig. 2. Schematic structure of natural slab separation with (a) monolayer graphene IL and (b) bilayer graphene IL between GaN slabs.
The number of layers of as-transferred graphene IL on GaN substrate was demonstrated using Raman spectroscopy. The Raman spectra of the graphene-covered GaN substrates are presented in Fig. 3. The major Raman features of graphene and graphite are the so called G band (∼ 1580 cm−1) and 2D band (∼ 2670 cm−1), which originate from in-plane vibration of two neighboring sp2 hybridized carbon atoms and a two phonon double resonance Raman process [27], respectively. It can be seen in Fig. 3 that the 2D band becomes broader and blueshifted when the graphene thickness increases from monolayer to trilayer graphene, and the presence of a sharp and symmetric 2D band is widely used to identify monolayer graphene [28]. In addition to the differences in the 2D band, the intensity of the G band increases almost linearly as the graphene thickness increases [29], as shown in Fig. 3. Therefore, the monolayer, bilayer and trilayer graphene ILs can be confirmed according to the Raman results [27]. What’s more, there is no defect-related D band (∼ 1340 cm−1), indicating the high crystal quality and few defects in the as-transferred graphene ILs.
The GaN epilayers were deposited on top of the graphene/GaN substrates by MOCVD. In order to inhibit freestanding GaN substrate from decomposing rapidly as the temperature rises up to above 1000 °C, two-step growth process is used in this work. A 10-nm-thick GaN nucleation layer was firstly grown at low temperature (920 °C-930 °C) to cover the freestanding GaN substrate surface. Following the growth of the nucleation layer, the growth temperature was ramped up to 1040 ◦C. The ramp step duration was 5 min while keeping the flow of TMGa and NH3 uninterrupted. Then a 1-µm-thick undoped GaN bulk layer was grown at high temperature (1040 °C-1060 °C) to achieve the high crystal quality. Figure 4(a)-(c) shows EBSD maps of GaN grown on GaN (001) substrates through monolayer, bilayer, and trilayer graphene ILs. The EBSD maps consists of domains with different colors from which c-orientations in the GaN epilayer can be identified. We found that as-grown GaN epilayer grown on monolayer graphene IL exhibits (001) orientation, as indicated in red by the inverse pole figure (IPF) color triangle (left of Fig. 4). There are a large portion of black domains and scattered green or blue dots which represent amorphous or rough GaN and (010) or (120) orientation GaN [30], respectively. Such results confirm that crystallographic information of GaN (001) substrate can only transmit through monolayer graphene IL. Figure 4(d)-(f) demonstrates that the surface of GaN grown on GaN (001) substrates via monolayer graphene IL shows smooth morphology, but those via bilayer or trilayer graphene ILs are rough and with many hillocks.
Fig. 4. EBSD maps of GaN grown on GaN (001) substrates through (a) monolayer graphene IL, showing (001) single-crystallinity, and of GaN grown on GaN (001) substrates through (b) bilayer and (c) trilayer graphene ILs showing (010)- and (120)-dominant polycrystallinity. On the left is the inverse pole figure (IPF) color triangle for crystallographic orientations. Microscopy photos of GaN grown on GaN (001) substrates through (d) monolayer, (e) bilayer and (f) trilayer graphene ILs.
The HRXRD (105) ϕ scan (Fig. 5(a)) of the GaN epilayer on monolayer-graphene-coated GaN substrate shows only one set of six-fold symmetric diffraction peaks. This phenomenon indicates that there is no azimuthal intermixing of orientation between GaN epilayer and GaN substrate, which confirms monolayer graphene’s transparency to the GaN potential field. The HRXRD (105) ϕ scan (Fig. 5(b)) of the GaN epilayer on bilayer-graphene-coated GaN substrate shows four sets of six-fold symmetric diffraction peaks with evident azimuthal misalignments from each other. It shows that coexistence of four different crystalline orientations within the GaN epilayer and GaN substrate. The substrate potential field is partially screened by bilayer graphene IL. These experimental and calculation results clearly indicate that vdW coherent epitaxy of GaN can only occur in the situation of monolayer graphene IL case. However, bilayer graphene IL is thought to also satisfy the demand of vdW coherent epitaxy of GaN in Ref. [15]. The difference in conclusions may be ascribe to the formation method of graphene IL. In Ref. [15], the bilayer graphene sheets were formed by repeating the same transfer procedure on a monolayer graphene sheet. Multiple transfer processes may damage the integrity of graphene sheet, hence epitaxial lateral overgrowth of GaN may occurs on the exposed monolayer graphene site using bilayer graphene as a mask, which has the same effect as a monolayer graphene IL. However, in this work bilayer graphene sheets were grown and transferred at one time. Hence the integrity of bilayer graphene sheets is even better, and the conclusion is more reliable in this respect. To investigate the impact of the growth temperatures, additional samples were grown on freestanding GaN (001) substrates through monolayer graphene IL using two-step growth process and varying the parameters: the nucleation layer growth temperature in the range of 920 °C-930 °C and the bulk layer growth temperature in the range of 1040 °C-1060 °C. The crystalline quality of the GaN epilayers was compared by using XRD rocking curve measurements with the results summarized in Fig. 5(c). All of the GaN epilayers grown on the monolayer graphene IL exhibited rocking curve full width at half maximum (FWHM) in the range of 148-201 arcsec, depending on the growth temperatures. Although the XRD signal mainly comes from the bottom freestanding GaN substrates, the deterioration degree of rocking curve FHWM can also demonstrate the crystalline quality of the GaN epilayer. The reference lines at the top and the bottom indicates the rocking curve FWHM of the GaN epilayer grown directly on the patterned sapphire substrates (PSS) and on the freestanding GaN (001) substrates, respectively. It can be seen that vdW coherent epitaxy of GaN can achieve good crystalline quality, which is better than the results of heteroepitaxy on traditional PSS. According to Fig. 7, we can obtain the single-crystalline GaN film whose crystalline quality is very close to the results of homoepitaxy directly on GaN freestanding substrates when the optimized growth temperatures of nucleation layer and bulk layer are 925 °C and 1055 °C, respectively.
Fig. 5. The HRXRD (105) ϕ scan of the GaN epilayer on (a) monolayer-graphene-coated and (b) bilayer-graphene-coated GaN substrates. (c) XRD rocking curve FWHM of GaN epilayers grown on freestanding GaN (001) substrates through monolayer graphene IL. The reference lines at the top and the bottom indicates the rocking curve FWHM of the GaN epilayer grown directly on the PSS and on the freestanding GaN (001) substrates, respectively. Colored symbols correspond to the different nucleation layer growth temperatures.
Fig. 6. PL spectra of InGaN samples grown on freestanding GaN (001) substrates through monolayer, bilayer, and trilayer graphene ILs. Near-band-edge emission of GaN, blue InGaN/GaN MQW emission, and yellow luminescence band are marked out at the appropriate location.
Fig. 7. (a). Arrhenius plots of the normalized integrated PL intensities of InGaN/GaN MQW grown on freestanding GaN (001) substrates through monolayer, bilayer and trilayer graphene ILs. (b). The evolution of quenching strength (A1, A2) and activation energy (E1, E2) of the 2-channels Arrhenius fitting results with the layer number of graphene ILs.
Hence, such GaN/graphene/GaN templates of high quality can be used to grow InGaN/GaN MQW for light-emitting devices. The InGaN/GaN MQW samples were prepared on freestanding GaN (001) substrates through monolayer, bilayer, and trilayer graphene ILs. The two-step growth process and the optimized growth temperatures were employed to deposit the 1-µm-thick undoped GaN bulk layer. After that, the growth temperature was adjusted to 740 °C for growing 5 periods of MQW sandwiched by nominal 3-nm-thick In0.12Ga0.88N QW and 9-nm-thick GaN barrier, followed by a nominal 12-nm-thick GaN capping layer. The optical properties of the InGaN/GaN MQW samples are studied using PL measurement at room temperature. As shown in Fig. 6, all the characteristic PL peaks can be observed: a near-band-edge emission of GaN near 364 nm, a blue InGaN/GaN MQW emission near 429 nm, and a yellow luminescence band near 550 nm. Several peculiarities can be noticed: in the sample grown on monolayer graphene IL, the blue InGaN/GaN MQW emission is considerably stronger relative to the near-band-edge emission of GaN, while yellow luminescence band cannot be observed. This result indicates a high optical quality of the InGaN/GaN MQW sample. But for the sample grown on bilayer or trilayer graphene IL, the blue InGaN/GaN MQW emission is relatively weak and the strong yellow luminescence band is gradually appeared. This might be related to the increased point defects and poor crystalline quality [31], without the coherent interaction on the condition of bilayer or trilayer graphene ILs.
Figure 7(a) shows the temperature dependence of the integrated PL intensity of InGaN/GaN MQW grown on freestanding GaN (001) substrates through monolayer, bilayer and trilayer graphene ILs, wherein the internal quantum efficiencies (IQEs) at 300 K are 69.89%, 45.97%, and 9.15%, respectively. Experimental data are presented by solid symbols, and the solid lines are their best Arrhenius fitting results. The following 2-channels Arrhenius fitting can be used for calculation of the activation energy in thermally quenching processes [32,33]:
For structural characterization, a XTEM observation of the InGaN/GaN MQW sample grown on freestanding GaN (001) substrates through monolayer graphene IL is implemented. The bright-field (BF) TEM image is shown in Fig. 8(a). The top inset is high-resolution HRTEM image of the InGaN/GaN MQW. The actual thicknesses of the InGaN QW, GaN barrier, and GaN capping layer all have a little deviation compared with the nominal values. It can be seen that each InGaN QW is slightly uneven in thickness, which may result from the heating and cooling processes during the growth of InGaN QW and GaN barrier. For this reason, the full width at half maximum (FWHM) of InGaN/GaN MQW’s PL spectra is over 30 nm (see Fig. 6). MOCVD growth conditions should be further optimized to improve the uniformity of InGaN/GaN MQW thickness. What’s more, the coherent alignment between a GaN (001) epilayer and a GaN (001) substrate was also atomically resolved by performing HRTEM, as shown in the bottom inset. The results reveal that the GaN (001) epilayer is epitaxially aligned with the GaN (001) substrate through the gap created by the monolayer graphene IL, which is visible between the epilayer and the substrate (indicated by the arrow). And there are no clear point defects and dislocation in the inspected area. The measured gap between the GaN epilayer and the substrate is about 4.5 Å, which is below the critical gap calculated with DFT. It confirms that vdW coherent epitaxy of GaN through monolayer graphene IL on GaN substrates does occur. However, there is no direct evidence of the coherent epitaxy character in the case of GaN/bilayer graphene IL/GaN heterostructure. Hence, in this respect, the possibility of vdW heteroepitaxy case cannot be ruled out in Ref. [15]. HRXRD ω−2θ scan and its fitting result is shown in Fig. 8(b). Several satellite peaks around the GaN 0002 diffraction peak can be observed in the ω−2θ scan, indicating the smoothness heterointerfaces in the MQW. The fitting of the experimental XRD curves to the theoretical curves revealed that the MQW consisted of 2.9-nm-thick In0.114Ga0.886N QWs and 8.8-nm-thick GaN barriers. These results represent a small and acceptable deviation from the design values. It demonstrates feasibility toward vdW coherent epitaxy of GaN-based luminescent thin films on reusable freestanding GaN substrates, and the release and transfer process may be addressed in future work. According to Jia et al. [35,36], graphene sheet will completely remain on the substrate surface after the transfer process of III-nitride epilayer due to the van der Waals interaction. Hence, the high-cost GaN substrate and CVD graphene sheet can be used repeatedly in production.
Fig. 8. (a) XTEM images of the InGaN/GaN MQW sample grown on freestanding GaN (001) substrates through monolayer graphene IL. The images are taken along the GaN [$11\bar{2}0$] direction. The HRTEM images of the InGaN/GaN MQW (top) and the GaN/graphene/GaN interfaces (bottom) are shown in the insets, respectively. The excellent coherent alignment of the GaN (001) lattices through the monolayer graphene IL is exhibited. (b) HRXRD ω−2θ scan (blue line) and its fitting result (red line) of the InGaN/GaN MQW sample grown on freestanding GaN (001) substrates through monolayer graphene IL.
4. Conclusion
In summary, we have demonstrated the vdW coherent epitaxy of GaN-based single-crystalline films through monolayer graphene IL on freestanding GaN substrate. The coherent interaction between GaN-GaN (substrate-epilayer) is explored by employing DFT calculations, and only monolayer graphene IL shows potential to allow the transmission of crystallographic information from GaN substrate to GaN epilayer. On this basis, monolayer, bilayer, and trilayer graphene sheets are fabricated on Cu foil by CVD synthesis and transferred onto freestanding GaN substrates, and GaN-based epilayers are grown on such graphene/GaN templates by MOCVD using two-step growth process. By optimizing the growth temperatures of nucleation layer and bulk layer, single-crystalline wurtzite phase GaN film whose crystalline quality is very close to the results of direct homoepitaxy on GaN freestanding substrates is obtained. The vdW coherent epitaxy of GaN through monolayer graphene IL on GaN substrates is experimentally demonstrated through HRTEM and HRXRD, which agree well with the DFT results. InGaN/GaN MQW sample grown on such GaN/graphene/GaN template shows good material quality and strong blue emission without yellow luminescence band. These results clarify that vdW coherent epitaxy of GaN can only occur in the situation of monolayer graphene IL case, and the optimized growth temperatures of nucleation layer and bulk layer are 925 °C and 1055 °C, respectively.
Funding
National Key Research and Development Program of China (2021YFA0716400); National Natural Science Foundation of China (61904093, 61975093, 61991443, 61974080, 61927811, 61822404, 61875104); Key Lab Program of BNRist (BNR2019ZS01005); China Postdoctoral Science Foundation (2018M640129, 2019T120090).
Acknowledgments
All the authors gratefully acknowledge the National Key Research and Development Program (Grant No. 2021YFA0716400), the National Natural Science Foundation of China (61904093, 61975093, 61991443, 61974080, 61927811, 61822404, and 61875104); the Key Lab Program of BNRist (BNR2019ZS01005); the China Postdoctoral Science Foundation (2018M640129 and 2019T120090); and the Collaborative Innovation Centre of Solid-State Lighting and Energy-Saving Electronics.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. J. H. Choi, A. Zoulkarneev, S. I. Kim, C. W. Baik, M. H. Yang, S. S. Park, H. Suh, U. J. Kim, H. B. Son, and J. S. Lee, “Nearly single-crystalline GaN light-emitting diodes on amorphous glass substrates,” Nat. Photonics 5(12), 763–769 (2011). [CrossRef]
2. J. Yu, J. Wang, W. Yu, C. Wu, B. Lu, J. Deng, Z. Zhang, X. Li, Z. Hao, L. Wang, Y. Han, Y. Luo, C. Sun, B. Xiong, and H. Li, “Low-temperature and global epitaxy of GaN on amorphous glass substrates by molecular beam epitaxy via a compound buffer layer,” Thin Solid Films 662, 174–179 (2018). [CrossRef]
3. J. Yu, L. Wang, Z. Hao, Y. Luo, C. Sun, J. Wang, Y. Han, B. Xiong, and H. Li, “Van der Waals epitaxy of III-nitride semiconductors based on 2D materials for flexible applications,” Adv. Mater. 32(15), 1903407 (2020). [CrossRef]
4. T. Fujii, Y. Gao, R. Sharma, E. L. Hu, S. P. DenBaars, and S. Nakamura, “Increase in the extraction efficiency of GaN-based light-emitting diodes via surface roughening,” Appl. Phys. Lett. 84(6), 855–857 (2004). [CrossRef]
5. A. Dadgar, J. Christen, T. Riemann, S. Richter, J. Bläsing, A. Diez, A. Krost, A. Alam, and M. Heuken, “Bright blue electroluminescence from an InGaN/GaN multiquantum-well diode on Si (111): Impact of an AlGaN/GaN multilayer,” Appl. Phys. Lett. 78(15), 2211–2213 (2001). [CrossRef]
6. V. Härle, B. Hahn, H. J. Lugauer, S. Bader, G. Brüderl, J. Baur, D. Eisert, U. Strauss, U. Zehnder, and A. Lell, “GaN-based LEDs and lasers on SiC,” Phys. Status Solidi A 180(1), 5–13 (2000). [CrossRef]
7. B. J. Skromme, H. Zhao, D. Wang, H. S. Kong, M. T. Leonard, G. E. Bulman, and R. J. Molnar, “Strain determination in heteroepitaxial GaN,” Appl. Phys. Lett. 71(6), 829–831 (1997). [CrossRef]
8. O. Moutanabbir and U. Gösele, “Bulk GaN ion cleaving,” J. Electron. Mater. 39(5), 482–488 (2010). [CrossRef]
9. D. F. Storm, T. O. McConkie, M. T. Hardy, D. S. Katzer, N. Nepal, D. J. Meyer, and D. J. Smith, “Surface preparation of freestanding GaN substrates for homoepitaxial GaN growth by rf-plasma MBE,” J. Vac. Sci. Technol. B 35(2), 02B109 (2017). [CrossRef]
10. K. Watanabe, T. Taniguchi, and H. Kanda, “Direct-bandgap properties and evidence for ultraviolet lasing of hexagonal boron nitride single crystal,” Nat. Mater. 3(6), 404–409 (2004). [CrossRef]
11. D. Li, M. B. Müller, S. Gilje, R. B. Kaner, and G. G. Wallace, “Processable aqueous dispersions of graphene nanosheets,” Nat. Nanotechnol. 3(2), 101–105 (2008). [CrossRef]
12. Y. Kim, S. S. Cruz, K. Lee, B. O. Alawode, C. Choi, Y. Song, J. M. Johnson, C. Heidelberger, W. Kong, and S. Choi, “Remote epitaxy through graphene enables two-dimensional material-based layer transfer,” Nature 544(7650), 340–343 (2017). [CrossRef]
13. H. Kim, K. Lu, Y. Liu, H. S. Kum, K. S. Kim, K. Qiao, S. Bae, S. Lee, Y. J. Ji, K. H. Kim, H. Paik, S. Xie, H. Shin, C. Choi, J. H. Lee, C. Dong, J. A. Robinson, J. Lee, J. Ahn, G. Y. Yeom, D. G. Schlom, and J. Kim, “Impact of 2D–3D heterointerface on remote epitaxial interaction through graphene,” ACS Nano 15(6), 10587–10596 (2021). [CrossRef]
14. J. Yu, Z. Hao, J. Deng, L. Wang, Y. Luo, J. Wang, C. Sun, Y. Han, B. Xiong, and H. Li, “Transferable InGaN quantum well grown at low temperature on amorphous substrates by plasma-assisted molecular beam epitaxy,” Cryst. Growth Des. 21(7), 3831–3837 (2021). [CrossRef]
15. W. Kong, H. Li, K. Qiao, Y. Kim, K. Lee, Y. Nie, D. Lee, T. Osadchy, R. J. Molnar, and D. K. Gaskill, “Polarity governs atomic interaction through two-dimensional materials,” Nat. Mater. 17(11), 999–1004 (2018). [CrossRef]
16. J. Yu, Z. Hao, L. Wang, Y. Luo, J. Wang, C. Sun, Y. Han, B. Xiong, and H. Li, “First-principle calculations of adsorption of Ga (Al, N) adatoms on the graphene for the van-der-Waals epitaxy,” Mater. Today Commun. 26, 101571 (2021). [CrossRef]
17. J. W. Shon, J. Ohta, K. Ueno, A. Kobayashi, and H. Fujioka, “Fabrication of full-color InGaN-based light-emitting diodes on amorphous substrates by pulsed sputtering,” Sci. Rep. 4(1), 5325 (2015). [CrossRef]
18. W. S. A. J. Jeong, “Structural properties of GaN films grown on multilayer graphene films by pulsed sputtering,” Appl. Phys. Express 7(8), 085502 (2014). [CrossRef]
19. J. Yu, Z. Hao, J. Deng, X. Li, L. Wang, Y. Luo, J. Wang, C. Sun, Y. Han, B. Xiong, and H. Li, “Low-temperaturevan der waals epitaxy of GaN films on graphene through AlN buffer by plasma-assisted molecular beam epitaxy,” J. Alloy Compd. 855, 157508 (2021). [CrossRef]
20. K. Badokas, A. Kadys, J. Mickevičius, I. Ignatjev, M. Skapas, S. Stanionytė, E. Radiunas, G. Juška, and T. Malinauskas, “Remote epitaxy of GaN via graphene on GaN/sapphire templates,” J. Phys. D: Appl. Phys. 54(20), 205103 (2021). [CrossRef]
21. J. Hafner, “Ab-initio simulations of materials using VASP: density-functional theory and beyond,” J. Comput. Chem. 29(13), 2044–2078 (2008). [CrossRef]
22. K. Burke, M. Ernzerhof, and J. P. Perdew, “Generalized gradient approximation made simple,” Phys. Rev. Lett. 77(18), 3865–3868 (1996). [CrossRef]
23. L. Li, J. Yu, Z. Hao, L. Wang, J. Wang, Y. Han, H. Li, B. Xiong, C. Sun, and Y. Luo, “Influence of point defects on optical properties of GaN-based materials by first principle study,” Comp Mater. Sci. 129, 49–54 (2017). [CrossRef]
24. J. D. Pack and H. J. Monkhorst, “Special points for Brillouin-zone integrations,” Phys. Rev. B 13(12), 5188–5192 (1976). [CrossRef]
25. J. D. Head and M. C. Zerner, “A Broyden-Fletcher-Goldfarb-Shanno optimization procedure for molecular geometries,” Chem. Phys. Lett. 122(3), 264–270 (1985). [CrossRef]
26. M. P. Teter, D. C. Allan, T. A. Arias, J. D. Joannopoulos, and M. C. Payne, “Iterative minimization techniques for ab initio total-energy calculations: molecular dynamics and conjugate gradients,” Rev. Mod. Phys. 64(4), 1045–1097 (1992). [CrossRef]
27. Z. Ni, Y. Wang, T. Yu, and Z. Shen, “Raman spectroscopy and imaging of graphene,” Nano Res. 1(4), 273–291 (2008). [CrossRef]
28. J. S. Bunch, A. M. Van Der Zande, S. S. Verbridge, I. W. Frank, D. M. Tanenbaum, J. M. Parpia, H. G. Craighead, and P. L. McEuen, “Electromechanical resonators from graphene sheets,” Science 315(5811), 490–493 (2007). [CrossRef]
29. Y. Y. Wang, Z. H. Ni, Z. X. Shen, H. M. Wang, and Y. H. Wu, “Interference enhancement of Raman signal of graphene,” Appl. Phys. Lett. 92(4), 043121 (2008). [CrossRef]
30. J. Yu, Z. Hao, J. Wang, J. Deng, W. Yu, L. Wang, Y. Luo, Y. Han, C. Sun, and B. Xiong, “Study on AlN buffer layer for GaN on graphene/copper sheet grown by MBE at low growth temperature,” J. Alloy Compd. 783, 633–642 (2019). [CrossRef]
31. M. A. Reshchikov and H. Morkoç, “Luminescence properties of defects in GaN,” J. Appl. Phys. 97(6), 061301 (2005). [CrossRef]
32. M. Hao, J. Zhang, X. H. Zhang, and S. Chua, “Photoluminescence studies on InGaN/GaN multiple quantum wells with different degree of localization,” Appl. Phys. Lett. 81(27), 5129–5131 (2002). [CrossRef]
33. A. Yasan, R. McClintock, K. Mayes, D. H. Kim, P. Kung, and M. Razeghia, “Photoluminescence study of AlGaN-based 280 nm ultraviolet light-emitting diodes,” Appl Phys. Lett. 83(20), 4083–4085 (2003). [CrossRef]
34. X. H. Zheng, H. Chen, Z. B. Yan, D. S. Li, H. B. Yu, Q. Huang, and J. M. Zhou, “Influence of the deposition time of barrier layers on optical and structural properties of high-efficiency green-light-emitting InGaN/GaN multiple quantum wells,” J. Appl. Phys. 96(4), 1899–1903 (2004). [CrossRef]
35. Y. Jia, J. Ning, J. Zhang, C. Yan, B. Wang, Y. Zhang, J. Zhu, X. Shen, J. Dong, D. Wang, and Y. Hao, “Transferable GaN enabled by selective nucleation of AlN on graphene for high-brightness violet light-emitting diodes,” Adv. Optical Mater. 8(2), 1901632 (2020). [CrossRef]
36. Y. Jia, H. Wu, J. Zhao, H. Guo, Y. Zeng, B. Wang, C. Zhang, Y. Zhang, J. Ning, J. Zhang, T. Zhang, D. Wang, and Y. Hao, “Growth mechanism on graphene-regulated high-quality epitaxy of flexible AlN film,” CrystEngComm 23(42), 7406–7411 (2021). [CrossRef]
