Type-I band alignment at MoS2/In0.15Al0.85N lattice matched heterojunction and realization of MoS2 quantum well

The valence and conduction band offsets (VBO and CBO) at the semiconductor heterojunction are crucial parameters to design the active region of contemporary electronic and optoelectronic devices. In this report, to study the band alignment parameters at the In0.15Al0.85N/MoS2 lattice matched heterointerface, large area MoS2 single layers are chemical vapor deposited on molecular beam epitaxial grown In0.15Al0.85N films and vice versa. We grew InAlN having an in-plane lattice parameter closely matching with that of MoS2. We confirm that the grown MoS2 is a single layer from optical and structural analyses using micro-Raman spectroscopy and scanning transmission electron microscopy. The band offset parameters VBO and CBO at the In0.15Al0.85N/MoS2 heterojunction are determined to be 2.08 ± 0.15 and 0.60 ± 0.15 eV, respectively, with type-I band alignment using high-resolution x-ray photoelectron spectroscopy in conjunction with ultraviolet photoelectron spectroscopy. Furthermore, we design a MoS2 quantum wel...

The valence and conduction band offsets (VBO and CBO) at the semiconductor heterojunction are crucial parameters to design the active region of contemporary electronic and optoelectronic devices. In this report, to study the band alignment parameters at the In 0.15 Al 0.85 N/MoS 2 lattice matched heterointerface, large area MoS 2 single layers are chemical vapor deposited on molecular beam epitaxial grown In 0.15 Al 0.85 N films and vice versa. We grew InAlN having an in-plane lattice parameter closely matching with that of MoS 2 . We confirm that the grown MoS 2 is a single layer from optical and structural analyses using micro-Raman spectroscopy and scanning transmission electron microscopy. The band offset parameters VBO and CBO at the In 0.15 Al 0.85 N/MoS 2 heterojunction are determined to be 2.08 6 0.15 and 0.60 6 0.15 eV, respectively, with type-I band alignment using high-resolution x-ray photoelectron spectroscopy in conjunction with ultraviolet photoelectron spectroscopy. Furthermore, we design a MoS 2 quantum well structure by growing an In 0.15 Al 0.85 N layer on MoS 2 /In 0.15 Al 0.85 N type-I heterostructure. By reducing the nitrogen plasma power and flow rate for the overgrown In 0.15 Al 0.85 N layers, we achieve unaltered structural properties and a reasonable preservation of photoluminescence intensity with a peak width of 70 meV for MoS 2 quantum well (QW). The investigation provides a pathway towards realizing large area, air-stable, lattice matched, and eventual high efficiency In 0. 15  Group III-V semiconductors have been extensively studied due to their potential applications in high efficiency electronic and optoelectronic devices such as high electron mobility transistors, light emitting diodes, and laser diodes. [1][2][3][4] Alongside, group VI transition metal dichalcogenides (TMDs) in the form of MX 2 have recently emerged as an atomic layered material system with promising electronic and optoelectronic properties. [5][6][7][8] Among the TMDs, molybdenum disulfide (MoS 2 ) in a single layer form is of potential interest for such devices owing to its direct bandgap and its prominent transport properties. 9,10 Recent advances in the integration of 2D-layered materials with wide band gap group-III nitride semiconductors are exciting due to their variety of applications in high current tunnel diodes. [11][12][13] In this context, several efforts were made to grow GaN on closely lattice matched TMDs. Yamada et al. reported the growth of GaN on bulk MoS 2 by plasma-enhanced molecular beam epitaxy (MBE). 14 Furthermore, there are a few reports on the growth of GaN on large area MoS 2 and WS 2 layers, 15,16 layered MoS 2 on GaN epilayers 17 by chemical vapor deposition (CVD) growth techniques, and layer transferred p-MoS 2 on GaN. 11 The band offset parameters (junction type: valence band offset-DE v and conduction band offset-DE c ) are measured for various heterojunctions. [18][19][20][21] Recently, we have reported the growth of GaN on single layer (SL)-MoS 2 and SL-WSe 2 with type-II band alignment. 22,23 In spite of the smaller in-plane lattice mismatch (%0.8%) of GaN with MoS 2 , 15 the type-II heterojunction formed by them can be solely utilized for electronic devices. [11][12][13] In contrast, optoelectronic devices formed by 2D/ 3D heterojunctions require a type-I band alignment which can be realized by using the group III-nitride alloys with higher bandgap as a constituent semiconducting layer of the 2D/3D heterojunction. Thus, In x Al 1-x N with a low In composition (12%-18%) exhibiting higher bandgap (>4.5 eV) and lattice matched to the MoS 2 layer with high contrast of the refractive index (%30%) may be employed to achieve the type-I 2D/3D junction. Hence, the determination of the band offset parameters (VBO and CBO) and the type of junction for epitaxially formed MoS 2 /In x Al 1-x N (or In x Al 1-x N/MoS 2 ) heterointerfaces is necessary to provide a route towards the integration of group-III nitrides with TMDs for designing 2D/3D based optoelectronic devices. Furthermore, the introduction of nitrogen plasma quenches the photoluminescence intensity due to plasma-induced damage, and thus, a modified epitaxial process was utilized.
In this study, CVD grown SL-MoS 2 on Si and In 0.15 Al 0.85 N/Si and MBE grown In 0.15 Al 0.85 N on Si and SL-MoS 2 /Si and In 0.15 Al 0.85 N epilayers were used to determine the band offsets at the In 0.15 Al 0.85 N/SL-MoS 2 heterointerface. Later, the structural and optical properties of the designed In 0.15 Al 0.85 N/MoS 2 /In 0.15 Al 0.85 N quantum well structure were studied.
The large area MoS 2 layers were prepared on Si(111) and In 0.15 Al 0.85 N/Si(111) substrates using a CVD system. The growth experiments of In 0.15 Al 0.85 N on Si and MoS 2 /Si substrates were carried out using the Veeco GEN 930 plasma-assisted molecular beam epitaxy (PAMBE) system at a substrate temperature of 500 C. For In 0.15 Al 0.85 N growth, the nitrogen plasma source was operated with a flow rate of 0.5 sccm and a radio frequency (RF) power of 200 W. In and Al metals were evaporated using a standard Knudsen cell with the beam equivalent pressure (BEP) values of %1 Â 10 À8 and %4 Â 10 À8 Torr. The crystallinity and composition of indium of the In x Al 1-x N epilayer were investigated using a CuK a High-Resolution X-ray Diffractometer (HRXRD). High-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) was utilized by operating a probe-corrected FEI Titan at an acceleration voltage of 300 kV. Using Horiba Aramis, room temperature (RT) Raman measurements were performed on In 0.15 Al 0.85 N and MoS 2 layers with the excitation line of the He-Cd laser of 325 nm and the cobolt laser of 473 nm, respectively, whereas lPL emission was acquired on MoS 2 /Si and MoS 2 / In 0.15 Al 0.85 N using a 473 nm excitation source. To probe the deep UV emission of In 0.15 Al 0.85 N, a Varian spectrometer with Xenon flash lamp as an excitation source was used. The high-resolution XPS measurements were carried out using a Kratos Axis Ultra DLD spectrometer equipped with a monochromatic Al K a X-ray source (h ¼ 1486.6 eV). To acquire electron affinities of MoS 2 and In 0.15 Al 0.85 N, UPS measurements were performed using the He I (h He-I ¼ 21.2 eV) photon source.
It is important to point out that the indium composition (x) can be chosen in such a way that the in-plane lattice mismatch between In x Al 1-x N and MoS 2 is vanishingly low (<0.5%). In order to accomplish such lattice matched In x Al 1-x N thin films, several kinetically controlled growth parameters were utilized having a prior idea of the growth conditions. 24 Figure 1(a) presents the HRXRD 2h-x scans obtained on the symmetric (on-axis) and asymmetric (offaxis) planes of the In x Al 1-x N film. Asymmetric planes were examined in skew symmetric geometry. 25 Figure 1(a) displays c-oriented peaks: InAlN(0002) at 35.28 along with the substrate Si(111) at 28.55 and Si(222) at 58.98 and the asymmetric reflection InAlN (10 11) at 37.13 . Using these on-axis and off-axis orientations of InAlN, the lattice parameters were extracted to be a ¼ 0.3178 and c ¼ 0.5087 nm. Thereby, the In composition was estimated to be %15% for epitaxially grown In x Al 1-x N. 26 To inquire the surface morphology and root mean square (rms) roughness of InAlN/Si, atomic force microscopy scans were obtained in the tapping mode using an Agilent 5400 AFM. From Fig. 1(b), the root mean square (rms) surface roughness of the InAlN film is observed to be %1.5 nm. Figure 1(c) shows a nonlinear dependence of the bandgap of the In x Al 1-x N alloy on the In composition (X-axis) and the corresponding in-plane lattice parameter (alternate X axis). The inset of Fig. 1(c) shows the PL spectrum of the In x Al 1-x N/Si sample exhibiting a peak position at %4.76 eV which is consistent with the estimated bandgap curve with the most accepted band bowing parameter b % 5 eV. 27,28 As a result, we observe that the in-plane lattice parameter of In x Al 1-x N with an indium composition of 15% is well lattice matched with that of the MoS 2 layer.
Micro-Raman spectroscopy was performed to investigate the optical quality and to ascertain the number of layers that accommodated in the CVD grown MoS 2 on Si and In 0.15 Al 0.85 N. Figure 1(d) (i-iii) shows the Raman spectra of MoS 2 /Si, In 0.15 Al 0.85 N/MoS 2 /Si, and MoS 2 /In 0.15 Al 0.85 N/Si, which were obtained by employing a 473 nm laser. In order to obtain the enhanced Raman signal from In 0.15 Al 0.85 N of the heterojunction, the measurements were carried out using 325 nm excitation by which the signal can overcome the reduced resonant excitation effect as shown in Fig. 1(d) (iv). 22,24 The two characteristic peaks in Fig. 1(d) observed at 385.5 6 0.5 and 405.0 6 0.5 cm À1 are attributed to E 1 2g and A 1g phonon modes of MoS 2 , respectively. These modes correspond to the in-plane vibration of Mo and S atoms (E 1 2g ) and out-of-plane vibration of S atoms (A 1g ) in MoS 2 . The separation between E 1 2g and A 1g phonon modes is found to be 19.5 6 0.5 cm À1 , which reveals that the MoS 2 layers deposited by CVD on Si and In 0.15 Al 0.85 N/Si templates are single layers. 29  of long wavelength optical phonons for In 0.15 Al 0.85 N is due to the condition that either m Al <l (In,N) or m In <l (Al,N) , where m is the mass and l is the reduced mass. 31,32 In contrast, the phonon modes of In 0.15 Al 0.85 N were not observed when the sample was excited with the 473 nm line, as shown in Fig. 1(d) (ii and iii), which is ascribed to the reduced resonant excitation effect. 22 Moreover, as a consequence of this resonant effect, in Fig. 1(d) (iv), the intensity of E 1 2g and A 1g of MoS 2 observed to be extremely lower than E H 2 , InN-like, and AlN-like A 1 (LO) phonon modes of In 0.15 Al 0.85 N with 325 nm excitation. Thus, these results show the sustainability of the MoS 2 layer during the growth of In 0.15 Al 0.85 N. The inset of Fig. 1(d) shows the cross-sectional STEM image acquired at the In 0.15 Al 0.85 N/SL-MoS 2 /Si heterojunction having a MoS 2 layer thickness of %0.8 nm, which is in good agreement with the thickness of the S-Mo-S single-layer. 33 34 Furthermore, the surface area of MoS 2 is sufficiently large to carry out the XPS measurements. In particular, due to the non-continuity of the MoS 2 layer on Si and In 0.15 Al 0.85 N templates, the region of interest on MoS 2 /Si and MoS 2 / In 0.15 Al 0.85 N samples was selectively chosen within the spatial resolution of HRXPS. 21,35 This was executed by comparing the intensity of Mo 3d and In 3d and Si 2p core-levels, which allowed us to collect the photoemission signal solely from MoS 2 /Si and MoS 2 /In 0.15 Al 0.85 N (In 0.15 Al 0.85 N/MoS 2 ), respectively. To determine the VBO, the core levels Mo 3d and In 3d were used for the analysis. The VBO for the MoS 2 /In 0.15 Al 0.85 N heterojunction can be calculated by the method provided by Kraut et al., 36 which is expressed as From Fig. 2(a), the first term of Eq. (1) deduced by taking into account the position of the Mo 3d 5/2 core-level referenced with respect to the VBM, which is measured to be 228. 88  Here, the In 3d core-level is deconvoluted with the In-N bonding, and no other states are observed. Figure 2(b) shows the fitting of the Mo 3d core-level equipped with three chemical states as similar to the states in the Mo 3d core-level of MoS 2 in Fig. 2(a). Thereby, the binding energy separation between Mo 3d 5/2 and In 3d 5/2 core-levels for MoS 2 / In 0.15 Al 0.85 N is found to be 215.30 6 0.10 eV, which is well corroborated with that of the In 0.15 Al 0.85 N/MoS 2 heterojunction (215.40 6 0.10 eV) as shown in Fig. 2(c). The 3d 3/2 corelevels of Mo and In have similar deconvolutions at higher binding energy values differed by %3.10 and %7.50 eV, respectively, from Mo 3d 5/2 and In 3d 5/2 states. The deconvolution of the Mo 3d core-level for the heterojunction sample infers the absence of the octahedral phase in MoS 2 which is in contrast to the previous report 22 Fig. 2(d). The In 3d 5/2 and 3d 3/2 core-levels are deconvoluted with the In-N bonding states at ( Thus, from Eq. (2), the CBO (DE c ) is determined to be 0.60 6 0.15 eV. Furthermore, the measured CBO value is verified by Anderson's affinity rule which is defined as the difference between the electron affinity values of constituent semiconductors of the heterojunction. 39 where h HeÀI is the photon energy of the He-I line, X is the spectrum width, and E g is the electronic band gap. The spectrum width (X) is the energy separation between the VBM and cutoff energy of secondary electrons with a sharp photoemission onset as shown in Figs Figure  4(b) shows the magnified STEM image of (a). Figure 4(c) displays the FFT image acquired on the In 0.15 Al 0.85 N barrier layer, which shows the wurtzite crystal structure with a c-oriented growth corroborated by the HRXRD measurements. Thus, the HAADF-STEM analysis is a clear evidence of the formation of a single layer MoS 2 QW sandwiched between In 0.15 Al 0.85 N layers. Furthermore, the free exciton peak at %1.87 in lPL spectra as shown in Fig. 4(d) for the MoS 2 QW is slightly blue shifted with respect to the PL peak at 1.84 eV of MoS 2 /In 0.15 Al 0.85 N. Here, we compare PL spectra of MoS 2 /In 0.15 Al 0.85 N and MoS 2 QWs to rule out the peak shifts associated with the strain. Thus, the blue shift of %30 meV is attributed to the quantum size induced confinement effect. This observed blue shift is far less than the reported blue shift for MoS 2 QDs, 40 which can be due to the van der Waals epitaxy with reduced confinement effects in the QW structure. The peaks at 1.87 and %2.03 eV are the exciton resonances corresponding to the transitions from the conduction band to two spin-split valence sub-bands that originated from the broken inversion symmetry of the crystal lattice. 41,42 The low intensity of PL for the QW is ascribed to the non-resonant excitation, which means that the In 0.15 Al 0.85 N barrier layers are not excited along with the MoS 2 QW. 43 However, our previous reports show the enormous quenching (100-150 times) of PL for nitrogen plasma irradiated MoS 2 layers and 2D/3D heterojunctions. 16,23 In this study, we preserved the PL signal of the QW with relatively high intensity (quenched by 5 times) and a comparable peak width of 70 meV by using the low nitrogen flow rate and low forward power of plasma source for the growth of In 0.15 Al 0.85 N barrier  layers. The respective inset shows the Raman spectra for the QW sample having the phonon mode separation of 19.5 6 0.5 cm À1 , which clarifies that the MoS 2 QW is a single layer.
In conclusion, MoS 2 single layers were chemical vapor deposited on the PAMBE grown In 0.15 Al 0.85 N template to study the band alignment at the lattice matched MoS 2 /In 0.15 Al 0.85 N heterointerface. We confirm that the CVD deposited MoS 2 is a single layer by means of micro-Raman phonon modes and STEM. The determination of band offset parameters at the MoS 2 /In 0.15 Al 0.85 N heterostructure was carried out by utilizing the HRXPS and UPS measurements. We determine the valence band and conduction band offset values of 2.08 6 0.15 eV and 0.6 6 0.15 eV, respectively, with type-I band alignment at the MoS 2 /In 0.15 Al 0.85 N heterostructure. Furthermore, we prepared a MoS 2 QW by growing the In 0.15 Al 0.85 N top barrier layer on the MoS 2 /In 0.15 Al 0.85 N template. The blue shift in PL spectra with respect to the MoS 2 single layer confirmed that the MoS 2 well exhibits the quantum confinement effect. The PL properties of the QW were preserved with a peak width of 70 meV by lowering the nitrogen plasma power and N 2 flow rate of overgrown In 0.15 Al 0.85 N layers. Therefore, our investigation indicated a feasible route for large area integration of 3D group III nitride materials with 2D transition metal dichalcogenide layers for future high efficiency, air-stable, and reliable applications of 2D/ 3D based photonic and optoelectronic applications.