Search tips
Search criteria 


Logo of scirepAboutEditorial BoardFor AuthorsScientific Reports
Sci Rep. 2017; 7: 7457.
Published online 2017 August 7. doi:  10.1038/s41598-017-07568-z
PMCID: PMC5547142

Radiative recombination of confined electrons at the MgZnO/ZnO heterojunction interface


We investigate the optical signature of the interface in a single MgZnO/ZnO heterojunction, which exhibits two orders of magnitude lower resistivity and 10 times higher electron mobility compared with the MgZnO/Al2O3 film grown under the same conditions. These impressive transport properties are attributed to increased mobility of electrons at the MgZnO/ZnO heterojunction interface. Depth-resolved cathodoluminescence and photoluminescence studies reveal a 3.2 eV H-band optical emission from the heterointerface, which exhibits excitonic properties and a localization energy of 19.6 meV. The emission is attributed to band-bending due to the polarization discontinuity at the interface, which leads to formation of a triangular quantum well and localized excitons by electrostatic coupling.


Oxide heterojunctions are a subject of strong current interest because they provide a wealth of novel functionalities that can be exploited in a broad range of currently emerging technologies involving ultrafast electronics, high-sensitivity chemical sensors and quantum technology14. The electrical transport properties of the MgZnO/ZnO interface are particularly exciting following the observation of two dimensional electron gas (2DEG) formation at MgZnO/ZnO heterointerfaces due to a strong built-in potential arising from polarization mismatch5. These 2DEG samples show remarkably high electron mobilities (in excess of 106 cm2/V·s at 40 mK6) and exhibit the fractional quantum Hall effect5. While the electronic transport properties of MgZnO/ZnO heterojunctions have been investigated by various groups7,8, few studies have focused on its optical properties. To date, only optical investigations of indirect excitons have been carried out in MgZnO/ZnO quantum wells7,9,10. However, optical characterization of the MgZnO/ZnO interface using such double heterojunction systems has two significant disadvantages. First, the recombination of electrons from sub-bands in excited quantum wells usually take place because of short recombination lifetimes, leading to exciton blue shift and well size-dependent electron-optic effects11. Second, the electron mobility in quantum wells is typically lower than in single heterojunctions.

An optical interface emission (called the H-band) has been observed in AlGaAs/GaAs12 and AlGaN/GaN1315 single heterostructures at low temperatures (T < 20 K), which has been attributed to radiative recombination between photoexcited holes and electrons confined at the interface. Similarly, due to its large band offset and strong strain induced piezoelectric field, the MgZnO/ZnO interface is expected to produce a 2DEG and an associated H-band emission that is more robust at higher temperatures because of the high exciton binding energy in ZnO. In this work, MgZnO/ZnO single heterostructures grown by Pulsed Laser Deposition (PLD) are studied using correlative luminescence and electrical characterization techniques. Notably, depth-resolved cathodoluminescence (CL) spectroscopy was employed to investigate the optical emission phenomenon at the MgZnO/ZnO interface. The optical properties of MgZnO/ZnO core-shell nanowires have been investigated previously by other workers16,17. The authors reported an enhancement of the near-band-edge (NBE) luminescence; however, the exact cause of the increase could not be identified unambiguously due to the difficulty of spatially resolving the interface geometry in nanowire core-shell structures. Identification of optical signature for MgZnO/ZnO heterostructures constitutes a seminal step in the understanding of localized excitons at the oxide interface and can have major implications for the interpretation of optical and electrical measurements.

Results and Discussion

Enhanced conductivity in the MgZnO/ZnO bilayer

The four-point resistivity of the MgZnO/ZnO/sapphire bilayer is 0.006 Ω.cm, which is more than two orders of magnitude lower than that of the constituent MgZnO film and an order of magnitude lower than that of the constituent ZnO (ρ = 3.0 Ω.cm and 0.1 Ω.cm for the single-layer MgZnO and ZnO films, respectively). This resistivity value is exceptionally low in absolute terms for a ZnO film without intentional shallow donor doping18. The reproducibility of the measured resistivity value was confirmed by collecting data at several separate areas of sample with various forward and reverse injection currents and repeating the analysis in three different laboratories. This enhanced conduction is attributed to electron confinement at the smooth MgZnO/ZnO interface. The surface roughness of the ZnO film was measured to be in the range 0.3–0.9 nm over an atomic force microscopy (AFM) scan area of 5 × 5 μm2 (shown in Supplementary Fig. S1(b)). Figure 1 shows temperature-dependent Hall mobility and carrier concentration for each of the films. All samples show n-type behavior. For the ZnO single layer, the mobility is about 20 cm2/V·s at room temperature (RT) and decreases linearly with temperature to approximately 16 cm2/V·s at 77 K. The corresponding carrier concentration rises from a RT value of 4.0 × 1018 cm−3 to a plateau of 4.4 × 1018 cm−3 from 180 K to 77 K. For the MgZnO single layer, the RT mobility of 3.5 cm2/V·s is significantly lower than that of the ZnO single layer. In a similar manner to the ZnO layer, the MgZnO mobility decreases linearly with temperature to ~0.8 cm2/V·s at 77 K. The corresponding carrier concentration is an order of magnitude lower than for the ZnO and rises from a RT value of 1.2 × 1018 cm−3 to 1.5 × 1018 cm−3 at 77 K. At 42 cm2/V·s, the RT mobility for the MgZnO/ZnO bilayer is significantly higher than for the ZnO or MgZnO single layers. As the temperature decreases, the mobility of the bilayer remains fairly constant down to ~200 K, after which it drops to 31 cm2/V·s at 77 K. This temperature-dependent behavior is different from those of the ZnO or MgZnO single layers, which exhibit monotonic decreases in mobility with decreasing temperature. This could be caused by the presence of highly mobile electrons and interface roughness scattering. The carrier concentration of the bilayer is higher than for the ZnO single layer and drops from a RT value of 5.0 × 1018 cm−3 to a minimum value of 4.8 × 1018 cm−3 at 180 K.

Figure 1
Temperature dependent Hall effect measurements conducted on the MgZnO/ZnO bilayer and constituent single-layer ZnO and MgZnO films on c-sapphire substrate, which yield electron mobility and concentration. At RT, the bilayer exhibits two orders of magnitude ...

Luminescence band originating from the heterostructure interface

Figure 2 shows photoluminescence (PL) spectra for the ZnO, MgZnO, and MgZnO/ZnO samples. The ZnO spectrum exhibits a dominant donor-bound exciton (D0X) peak at 3.365 eV, attributed to an I3 ionized donor, while a weaker peak on the right shoulder at 3.385 eV is due to free exciton transitions (FX)19,20. The peaks at 3.295 eV and 3.221 eV are the first and second order longitudinal optical (LO) phonon replicas of the FX, respectively, with a constant spacing equal to the ZnO LO phonon energy of 72 meV. The presence of the phonon replica peaks indicates high optical quality of the film. The broad NBE emission of the MgZnO and MgZnO/MgO films at 3.570 eV are blue shifted with respect to that of the ZnO due to the enlarged bandgap caused by Mg substitution on Zn sites21. The broadening of the MgZnO emission is due to potential fluctuation effects, arising from variation in the Mg composition and lattice disorder in the MgZnO alloy22. The absorption edges in RT optical transmission spectra (Supplementary Fig. S3) corresponded to a bandgap energy of 3.61 eV for the MgZnO/ZnO. This corresponds to an Mg content of 16.4 at%23,24, which is 2.7 at% more than the nominal composition of the MgZnO PLD target. Such an Mg enhancement compared to the target concentration has been reported elsewhere for PLD growth of MgZnO and was attributed to a difference in the vapor pressures of Mg and Zn18. The two sharp peaks at 3.671 eV and 3.743 eV, located at 72 meV and 144 meV from the laser excitation line, are E1(LO) and 2E1(LO) Raman modes of ZnO. The most significant feature in Fig. 2 is the emergence of a broad emission band at ~3.2 eV (denoted H-band) in the MgZnO/ZnO bilayer. This luminescence band is absent in both the single ZnO and MgZnO films, suggesting that it originates from the heterojunction interface.

Figure 2
High-resolution PL spectra of three samples of ZnO/c-sapphire, MgZnO/c-sapphire, and MgZnO/ZnO/c-sapphire at 7 K using a 325 nm laser excitation source. The bilayer exhibits a luminescence H-band at ~3.2 eV, which is absent in ...

Depth-resolved CL microanalysis of the MgZnO/ZnO interface H-band

To verify the origin of the 3.2 eV band, depth-resolved CL was conducted on each of the three films. In this measurement the electron beam power and hence the electron-hole (e-h) pair generation rate in the sample was kept constant as the CL excitation depth was increased by raising the electron beam energy. The ZnO and MgZnO films exhibit identical CL spectra at all accelerating voltages confirming the in-depth homogeneity of the layer (Supplementary Fig. S3a). To obtain depth-dependent dependences of the CL in the MgZnO/ZnO bilayer, Monte Carlo modeling was carried out using the CASINO simulation package25 to determine the electron energy loss profile with the electron beam excitation energy. Figure 3a shows the energy loss profiles of electrons in the bilayer for acceleration voltages between 2 kV and 10 kV obtained from the CASINO simulations. These energy loss profiles are comparable with those obtained for ZnO by other workers26,27. The Monte Carlo simulation of the bilayer reveals that the electron beam reaches the 260 nm deep ZnO layer at 5 kV and starts penetrating into the sapphire substrate from 7 kV. The depth-resolved CL measurements of the bilayer (Supplementary Fig. S3b) are consistent with the CASINO modeling results. Figure 3b shows in-depth CL spectra of the bilayer from 7 kV to 10 kV. The bilayer spectra exhibit an additional broad peak positioned at ~3.2 eV, which is not observed in either of the constituent ZnO or MgZnO single layers. The shape of the H-band was obtained by subtracting the ZnO luminescence contribution from the bilayer CL emission at 7 kV, yielding a broad spectrum peaked at 3.2 eV (Supplementary Fig. S4). These depth-resolved CL results confirm the H-band emission originating from the MgZnO/ZnO heterointerface.

Figure 3
Depth-resolved CL measurements of the MgZnO/ZnO/c-sapphire bilayer. (a) CASINO simulated electron energy loss curves for acceleration voltages from 2 kV to 10 kV. The vertical axis corresponds to the percentage of CL generated at that ...

Excitation-power resolved CL analysis of the bilayer at 7 kV was measured by increasing the beam current (I B) from 0.03 nA to 70 nA and analyzed using the power-law model I CL [proportional, variant] I B k, where I CL is the CL intensity and I B is beam current28. Since the minimum energy of the electron beam necessary to cause knock-on damage is 400 keV for Zn ions and 250 keV for O ions, the electron beam is not expected to damage the bilayer film. The log-log plots based on the power-law model are displayed in the inset of Fig. 3b, which yields the exponent k = 1.01 and k = 0.91 for the D0X and H-band peaks, respectively. These results indicate that the H-band is not related to a lattice coupled defect, as defect-related emissions in ZnO typically exhibit sublinear dependence on the beam power29,30. In addition, as the excitation density is increased, the H-band is strongly blue shifted by ~40 meV, while the energy position of the ZnO D0X peak remains unchanged (Supplementary Fig. S4). This is likely caused by the band bending field being screened by excess carriers that are generated by the electron beam, providing further support for its assignment of the H-band to localized excitons at the MgZnO/ZnO interface.

Temperature dependent CL was performed to determine the activation energy of the interface emission. Figure 4a shows the temperature-resolved CL spectra of the MgZnO/ZnO bilayer acquired at temperatures from 80 K to 300 K at 7 kV. With increasing temperature, the NBE CL intensity is quenched because of the thermal detachment of bound excitons from shallow donors and the high probability of non-radiative recombination. The H-band shows remarkable thermal stability up to 150 K. The activation energies of the D0X and H-band can be determined according to the Arrhenius law equation:


where I(0) is the luminescence intensity at 0 K, E a is the activation energy of the emission, and C is a scaling factor. Figure 4b shows the D0X and H-band intensities of the bilayer against temperature; fitting these data to the Arrhenius equation gives E a = 19.6 ± 0.4 meV for the H-band and 16.0 ± 0.3 meV for the D0X line. The E a value for the D0X line is within the range of reported activation energies of bound excitons (10–20 meV)31. Because the E a of the MgZnO/ZnO H-band is significantly greater than that of GaN heterostructures (~10 meV)32, the emission from the ZnO is thermally stable at higher temperatures.

Figure 4
Temperature-resolved CL measurements of the MgZnO/ZnO bilayer. (a) Temperature-dependent CL spectra of the bilayer acquired at an acceleration voltage of 7 kV. The H-band is identifiable at temperatures up to 160 K. (b) Variations of the ...

Radiative recombination at the heterointerface

The H-band emission in the ZnO/MgZnO heterostructure can be explained by the interface band structure (Fig. 5). Using the reported band offset ratio of ΔE vE c = 0.533, the band-edge discontinuities at the interface can be estimated to be ΔE c = 253 meV for the conduction band and ΔE v = 127 meV for the valence band. In contrast to AlGaAs/GaAs12 modulation-doped heterostructures where a large barrier height at the interface together with the doping level determines the electron distribution at the interface, it is the polarization discontinuity at the MgZnO/ZnO that induces the band bending at its interface, leading to the formation of a triangular electron well26, where the electron is confined within the potential well and spatially separated from the hole in the flat-band region of the ZnO valence band. Thus the nergy position of the H-band localized exciton can be estimated as ref.34:

EH-bandEg − Ee,loc − Ee,hole

where E g is the ZnO band gap, E e,loc the localization energy of electrons in the interface potential well and E e,hole the energy required to disassociate the hole from the excitonic complex (i.e. the binding energy of the hole). Additionally, it has been pointed out that the activation energy of the H-band exciton corresponds to E e,hole, rather than the localization energy of electron in the potential well34. According to these results, we can estimate the binding energy of the hole in the localized exciton to be ~20 meV. From the measured values for the heterostructure, the ground state of electrons in the potential well can be calculated from Eq. (2), E e,loc = 110 meV. The polarization sheet charge at the MgZnO/ZnO interface can be estimated as follows35,36:

σPsp(x) − Psp(0) − Ppe(x)

where P sp(x) is the spontaneous polarization in the MgxZn1−xO layer and P pe is the piezoelectric polarization, respectively. Analytical solutions for each of the polarization terms have previously been determined by solving Poisson and Schrödinger equations37, which yield Psp(x) = −0.0322 + 0.024x C/m2 and Ppe(x) = −0.0584x. For the MgZnO/ZnO heterostructure in this work (x = 0.164), σ = 0.0135C/m2, which is consistent with the value obtained from the interface scattering model for MgZnO/ZnO heterostructures38.

Figure 5
Schematic band diagram of the MgZnO/ZnO heterostructure, showing the recombination channels responsible for the emissions in the bilayer film. Dotted lines represent the excitonic transitions accounting for the observed NBE emissions in ZnO and MgZnO. ...

In conclusion, optical studies were conducted on the MgZnO/ZnO heterointerface which showed significantly increased conductivity and electron mobility compared with the constituent layers. Depth-resolved luminescence spectroscopy revealed a radiative recombination H-band at 3.2 eV arising from the heterointerface, which was not observed for the constituent single-layer ZnO or MgZnO films. This interfacial H-band with a large activation energy of 19.6 meV is stable up to 150 K and attributed to localized excitons that arise due to the presence of confined electrons at the interface as a result of band bending. These findings may provide new opportunities to optically access hidden junctures and to utilize the optical emission arising from recombination of localized excitons in optoelectronic interface devices.


MgZnO thin films were grown simultaneously on c-sapphire and O-polar ZnO (0001)-coated c-sapphire substrates using pulsed laser deposition (PLD) with a Coherent LPX 100 KrF (248 nm) excimer laser and a commercial sintered MgZnO (13.4 atomic % Mg) target in oxygen, using fabrication conditions described elsewhere39. The film thicknesses determined from scanning electron microscope (SEM) cross-sections in an FEI field emission gun XL30S system were 110 nm and 260 nm for the ZnO and MgZnO films, respectively. X-Ray Diffraction (XRD) was performed in a Panalytical MRD Pro system using Cu Kα1 radiation. Electrical resistivity was measured using a Signatone co-linear four-point probe system. Temperature dependent Hall measurements were made using a Van der Pauw configuration and indium contacts. The chemical composition of the films was determined using a Zeiss Evo LS15 SEM equipped with a Bruker XFlash 5030 Silicon Drift Detector EDX spectrometer. RT optical transmission studies were performed using an Ocean Optics system comprising a halogen and deuterium lamps plus a Maya spectrometer. CL microanalysis of the films was conducted in an FEI Quanta 200 Environmental SEM equipped with a diamond machined parabolic light collector and a Hamamatsu S7011–1007 CCD spectrometer. Depth-resolved CL measurements were conducted under constant beam power (I o V o = 28 µW) by increasing the accelerating voltage while adjusting the electron beam current. Under such excitation conditions the e-h pair generation in the sample was kept constant. Additionally, the samples were analyzed by micro-PL spectroscopy in a liquid He bath at 6 K. The samples were excited by 325 nm He-Cd laser line and the emitted light was dispersed by a Spex-1404 double monochromator. The spectra were acquired with an optical power density of 100 kW/cm2.

Electronic supplementary material


This work was supported by Australian Research Council Discovery, grant DP150103317. The authors would like to thank (i) V. Pillard, T. Maroutian and P. LeCoeur at the Institut d’Electronique Fondamentale (IEF) of the University of Paris Sud for the AFM measurements, and (ii) F. Bayle, also of the IEF, for the SEM measurements.

Author Contributions

Author Contributions

C.T.-T., S.C., M.R.P. and D.J.R. co-wrote the manuscript, which was reviewed by all authors. S.C. and A.G. carried out the cathodoluminescence and optical measurements; C.N. performed photoluminescence experiments and analysed the data with the assistance of A.H.; D.J.R. carried out XRD and resistivity measurements; E.V.S conducted SEM experiments. D.L. performed Hall effect measurements. E.V.S., P.B., F.H.T., R.M. and M.R. contributed to the conception, design and fabrication of the oxide heterostructures.


Competing Interests

The authors declare that they have no competing interests.


A correction to this article is available online at

Electronic supplementary material

Supplementary information accompanies this paper at doi:10.1038/s41598-017-07568-z

Change History: A correction to this article has been published and is linked from the HTML version of this paper. The error has been fixed in the paper.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.


1. Först, M. et al. Spatially resolved ultrafast magnetic dynamics initiated at a complex oxide heterointerface. Nat. Mater. 14, 883 (2015). [PubMed]
2. Yang, Z., Ko, C. & Ramanathan, S. Oxide electronics utilizing ultrafast metal-insulator transitions. Annu. Rep. Mater. Res.41, 337 (2011).
3. Kumar, A., Sanger, A., Kumar, A. & Chandra, R. Highly sensitive and selective CO gas sensor based on a hydrophobic SnO2/CuO bilayer. RSC Adv.6, 47178 (2016).
4. Hwang, H. Y. et al. Emergent phenomena at oxide interfaces. Nat. Mater.11, 103 (2012). [PubMed]
5. Tsukazaki, A. et al. Quantum Hall effect in polar oxide heterostructures. Science315, 1388 (2007). [PubMed]
6. Falson, J. et al. MgZnO/ZnO heterostructures with electron mobility exceeding 1 × 106 cm2/Vs, 6 (2016). [PMC free article] [PubMed]
7. Bian, J. et al. Room temperature electroluminescence from the n-ZnMgO/ZnO/p-ZnMgO heterojunction device grown by ultrasonic spray pyrolysis. Chem. Phys. Lett.430, 183 (2006).
8. Tampo, H. et al. Two-dimensional electron gas in Zn polar ZnMgO/ZnO heterostructures grown by radical source molecular beam epitaxy. Appl. Phys. Lett.89, 2113 (2006).
9. Kuznetsova, Y. et al. Transport of indirect excitons in ZnO quantum wells. Opt. Lett.40, 3667 (2015). [PubMed]
10. Coli, G. & Bajaj, K. Excitonic transitions in ZnO/MgZnO quantum well heterostructures. Appl. Phys. Lett.78, 2861 (2001).
11. Makino, T., Segawa, Y., Kawasaki, M. & Koinuma, H. Optical properties of excitons in ZnO-based quantum well heterostructures. Semicond. Sci. Technol.20, S78 (2005).
12. Bergman, J. et al. Time-resolved measurements of the radiative recombination in GaAs/Al x Ga 1− x As heterostructures. Phys. Rev.B43, 4771 (1991). [PubMed]
13. Bergman, J. et al. Photoluminescence related to the two‐dimensional electron gas at a GaN/AlGaN heterointerface. Appl. Phys. Lett.69, 3456 (1996).
14. Jiang, H. & Lin, J. AlGaN and InAlGaN alloys-epitaxial growth, optical and electrical properties, and applications. Opto-electron.Rev271 (2002).
15. Romero MF, et al. Luminescence from two-dimensional electron gases in InAlN/GaN heterostructures with different In content. Appl. Phys. Lett. 2012;100:4. doi: 10.1063/1.4720087. [Cross Ref]
16. Cao, B. et al. Homogeneous core/shell ZnO/ZnMgO quantum well heterostructures on vertical ZnO nanowires. Nanotechnology20, 305701 (2009). [PubMed]
17. Ahn, C. H., Mohanta, S. K., Kong, B. H. & Cho, H. K. Enhancement of band-edge emission of ZnO from one-dimensional ZnO/MgZnO core/shell nanostructures. J. Phys. D: Appl. Phys.42, 115106 (2009).
18. Lorenz, M. et al. Optical and electrical properties of epitaxial (Mg, Cd) x Zn 1− x O, ZnO, and ZnO:(Ga, Al) thin films on c-plane sapphire grown by pulsed laser deposition. Solid State Electron.47, 2205 (2003).
19. Wang, L. & Giles, N. Temperature dependence of the free-exciton transition energy in zinc oxide by photoluminescence excitation spectroscopy. J. Appl. Phys.94, 973 (2003).
20. Teke, A. et al. Excitonic fine structure and recombination dynamics in single-crystalline ZnO. Phys. Rev. B70, 195207 (2004).
21. Wang, F., Zhao, C., Liu, B. & Yuan, S. Synthesis and photoluminescence of quasi-arrayed ZnMgO nanorods, J. Phys. D: Appl. Phys.42, 115411 (2009).
22. Park, W., Yi, G.-C. & Jang, H. Metalorganic vapor-phase epitaxial growth and photoluminescent properties of Zn1− xMgxO (0≤ x≤ 0.49) thin films. Appl. Phys. Lett.79, 2022 (2001).
23. Liu, J. et al. Degenerated MgZnO films obtained by excessive zinc. J. Cryst. Growth347, 95 (2012).
24. Kim, D. C. et al. Selective Crystalline Seed Layer Assisted Growth of Vertically Aligned MgZnO Nanowires and Their High-Brightness Field-Emission Behavior. Cryst. Growth Des.9, 4308 (2009).
25. Drouin, D., Hovington, P. & Gauvin, R. CASINO: A new monte carlo code in C language for electron beam interactions—part II: Tabulated values of the mott cross section. Scanning19, 20 (1997).
26. Ruane, W. et al. Defect segregation and optical emission in ZnO nano-and microwires. Nanoscale8, 7631 (2016). [PubMed]
27. Brillson, L. et al. Spatially-resolved cathodoluminescence spectroscopy of ZnO defects. Mater. Sci. Semicond. Process.57, 197 (2017).
28. Phillips, M. R. et al. Cathodoluminescence efficiency dependence on excitation density in n-type gallium nitride. Microsc. Microanal.9, 144 (2003). [PubMed]
29. Ton-That C, Weston L, Phillips MR. Characteristics of point defects in the green luminescence from Zn- and O-rich ZnO. Phys. Rev. B. 2012;86:115205. doi: 10.1103/PhysRevB.86.115205. [Cross Ref]
30. Ton-That, C. & Phillips, M. R. In Semiconductor nanowires: Materials, Synthesis, Characterization and Applications, edited by Arbiol & Q. Xiong (Woodhead Publishing, Cambridge, England, 2015, p. 393).
31. Wagner, M. et al. Bound excitons in ZnO: Structural defect complexes versus shallow impurity centers. Phys. Rev. B84, 035313 (2011).
32. Martínez-Criado, G. et al. Photoluminescence of Ga-face AlGaN/GaN single heterostructures. Mater. Sci. Eng. B82, 200 (2001).
33. Yin HT, et al. Composition dependent band offsets of ZnO and its ternary alloys. Sci Rep. 2017;7:41567. doi: 10.1038/srep41567. [PMC free article] [PubMed] [Cross Ref]
34. Martinez-Criado G, et al. Two-dimensional electron gas recombination in undoped AlGaN/GaN heterostructures. Jpn. J. Appl. Phys. Part 1 - Regul. Pap. Brief Commun. Rev. Pap. 2004;43:3360. doi: 10.1143/JJAP.43.3360. [Cross Ref]
35. Tampo, H. et al. Polarization-induced two-dimensional electron gases in ZnMgO/ZnO heterostructures. 93, 202104 (2008).
36. Malashevich, A. & Vanderbilt, D. First-principles study of polarization in Zn1-xMgxO. Phys. Rev.75, 045106 (2007).
37. Khan, M. A., Singh, R., Mukherjee, S. & Kranti, A. Buffer Layer Engineering for High (>=10(13) cm(−2)) 2-DEG Density in ZnO-Based Heterostructures. IEEE Trans Electron Devices64, 1015 (2017).
38. Wang P, et al. Theoretical investigation of the impact of barrier thickness fluctuation scattering on transport characteristics in undoped MgZnO/ZnO heterostructures. Jpn. J. Appl. Phys. 2015;54:5.
39. Rogers, D. J. et al. In Integrated Optoelectronic Devices 2005 (International Society for Optics and Photonics, 2005, p. 412).

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group