• Home
  • Sitemap
  • Contact us
Article View

Research Paper

Applied Science and Convergence Technology 2023; 32(6): 151-154

Published online November 30, 2023


Copyright © The Korean Vacuum Society.

Electroluminescent MgZnO/ZnO Heterojunction Diode

Byeong-Hyeok Kima and Jang-Won Kangb , *

aGIST Central Research Facilities, Gwangju Institute of Science and Technology, Gwangju 61005, Republic of Korea
bDepartment of Semiconductor and Applied Physics, and MNU Semiconductor Nanotechnology Institute, Mokpo National University, Muan 58554, Republic of Korea

Correspondence to:kangjw@mnu.ac.kr

Received: August 11, 2023; Revised: October 12, 2023; Accepted: October 13, 2023

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (https://creativecommons.org/licenses/by-nc-nd/4.0/) which permits non-commercial use, distribution and reproduction in any medium without alteration, provided that the original work is properly cited.

We report the fabrication and characterization of a heterojunction diode with a MgZnO/ZnO structure grown on a GaN/Al2O3 substrate. A radio frequency sputtering method was employed to achieve a high Mg composition in the MgZnO alloy layer, whereas the ZnO layer was epitaxially grown by metal-organic chemical vapor deposition. The MgZnO/ZnO-based heterojunction diode exhibited rectifying current-voltage (I−V) characteristics under both forward and reverse bias. However, white electroluminescence (EL) emission was observed only under high forward bias. To understand these phenomena, we thoroughly investigated the I−V characteristics of the heterojunction diode, which revealed that the tunneling of holes through the MgZnO barrier is an important mechanism for EL emission. In particular, the Fowler-Nordheim (FN) tunneling of holes is mainly responsible for EL emission under high forward bias. An explanatory schematic of the band diagram based on the I−V characteristics suggests that the EL emission is primarily attributed to the injection of holes via FN tunneling through the MgZnO barrier. This study provides a potential application for optoelectronic devices using MgZnO/ZnO-based heterostructures as ZnO-based light emitters.

Keywords: Heterojunction diode, MgZnO, Tunneling, Electroluminescence

Oxide semiconductors have been widely used in semiconductor devices such as field-effect transistors, solar cells, and piezoelectric devices [14]. The zinc oxide (ZnO) semiconductor, for example, is used in a wide range of optoelectronic applications [18]. The direct bandgap and large exciton binding energy (60 meV) of ZnO are favorable properties as an active layer of light emitters [911]. In addition, MgZnO and CdZnO layers, adopted through bandgap engineering by alloying MgO and CdO, enable the fabrication of ZnO-based quantum heterostructures and heterojunction devices [1215]. However, achieving a high Mg composition in ZnO-based alloy layers such as MgZnO, is challenging owing to the limited alloy solubility (25.0 and 8.5 % [16,17] for Mg and Cd, respectively). Moreover, the fabrication of MgZnO/ZnO heterostructured layers is limited owing to the difference in the crystalline structures of ZnO (wurtzite) and MgO (cubic), which makes it challenging to achieve epitaxial layer-by-layer growth between ZnO and MgZnO with a high Mg composition. Therefore, a thin-film-based platform of a MgZnO/ZnO heterostructure requires an alternate approach to grow a MgZnO layer of high Mg content on the ZnO layer. The high Mg composition in the MgZnO layer may be achieved using non-equilibrium thermodynamic growth conditions, such as the radio frequency (RF) sputtering system [18,19]. In this study, we fabricated a MgZnO/ZnO heterojunction diode with white-light emission properties using MgZnO/ZnO layers grown on a GaN/Al2O3 substrate using two methods. To grow the MgZnO/ZnO heterostructure, a ZnO layer was epitaxially grown on a ZnO-buffered GaN substrate by metal-organic chemical vapor deposition (MOCVD). The MgZnO layer with a high Mg content was grown via RF sputtering to serve as a barrier for the tunneling of carriers. We study the I−V characteristics of the heterojunction diode to draw conclusions upon the rectifying I−V characteristics under forward and reverse bias, as well as electroluminescence (EL) observations.

For epitaxial growth of the ZnO active layer, a ZnO buffer layer with a thickness of approximately 200 nm was deposited on the GaN/Al2O3 substrate under an Ar/O2 gas ambient at 950 °C using RF-sputtering. Subsequently, a ZnO film with a flat surface morphology was grown by MOCVD using diethylzinc (DEZn) and oxygen gas (99.999 % purity) as the sources of Zn and O. The growth temperature and working pressure were 800 °C and 50 torr, respectively. A MgZnO layer with a high Mg composition was grown at 500 °C using the same RF sputtering method employing a ZnO target mixed with 20 wt% MgO. The thickness of each layer was controlled through the growth time and estimated using scanning electron microscope (SEM) images. The absorbance and photoluminescence (PL) spectra confirmed the Mg composition of the MgZnO layers. An area of 5 mm × 5 mm was patterned using photolithography and the oxide layers, including the MgZnO and ZnO layers, were selectively etched via wet-etching using diluted hydrochloric (HCl) acid to fabricate the MgZnO/ZnO-based heterojunction diode. Using a shadow mask with an open area of 0.5 mm × 0.5 mm, the contact metal electrodes of Ti (20 nm)/Au (80 nm) were deposited on both the MgZnO and GaN layers via e-beam evaporation.

3.1. Structural and optical properties

Structural properties

Figure 1(a) displays a schematic of the film structure obtained via layer-by-layer growth. Empirically, growth of a highly crystalline ZnO thin film on a GaN substrate is challenging because the single crystalline phase Ga2O3 interface layer must be deposited by epitaxial growth [20]. Thus, the ZnO layer was grown via RF-sputtering as a buffer layer to support the epitaxial growth of the ZnO film. An additional ZnO active layer was grown via MOCVD. A MgZnO layer with high Mg composition forms a high barrier height from a bandgap larger than that of ZnO, which is necessary to fabricate tunnel diodes. However, MgZnO with a Mg content above 25 % is challenging to obtain under equilibrium thermodynamic growth conditions [16]. Therefore, the MgZnO barrier layer with a high Mg content was deposited using RF sputtering. At each step of the layer-by-layer growth, we confirmed the thickness of the layers using SEM measurements, as displayed in Fig. 1(b).

Figure 1. (a) Film structure of MgZnO/ZnO/GaN heterostructure via layer-by-layer growth. (b) Cross-sectional SEM image of the MgZnO/ZnO/GaN heterostructure. (c) Absorption spectrum of MgZnO layer grown on the glass substrate by RF sputtering. The absorption edge of the MgZnO layer corresponds to approximately 5.4 eV. (d) PL spectrum of the ZnO layer grown on the ZnO-buffered GaN substrate via MOCVD.
Optical properties

The absorption spectra were measured using a ultraviolet-visible spectrometer, as illustrated in Fig. 1(c), to estimate the Mg composition and bandgap of the MgZnO layer. The absorption edge can be estimated at approximately 5.4 eV, which corresponds to a Mg composition of 45 % [21], indicating that a large barrier height against carrier transport can be provided at the MgZnO/ZnO interface. To verify the optical properties of the ZnO layer, the PL spectrum was measured using a home-built spectroscopy system. Figure 1(d) demonstrates that the PL intensity near the band edge was significantly improved in the MOCVD-grown ZnO layer compared to that deposited via RF sputtering.

3.2. Device characteristics

Figure 2(a) displays a schematic of the MgZnO/ZnO/GaN heterojunction device fabricated via layer-by-layer growth using RF sputtering and MOCVD. To form separate contact electrodes for MgZnO and GaN, the MgZnO/ZnO layers were selectively etched after prepatterning using diluted HCl wet etching without damaging the GaN layer. Figure 2(b) displays the I−V curve of the MgZnO/ZnO/GaN heterojunction device with the rectifying diode characteristics for both forward and reverse bias, which are analogous to back-to-back Schottky or tunnel diodes [22,23]. In the case of a forward bias, a positive bias is applied to the contact on the MgZnO layer, whereas a negative bias is applied to the GaN side. The turn-on voltage under forward bias was measured at approximately 5.4 V, whereas the turn-on voltage under reverse bias was approximately 2.5 V, implying that the barrier height for electron injection through the MgZnO layer was lower than that for the holes. Importantly, light emission was detected only under a forward bias. The EL spectra as a function of the injection current are shown in Fig. 2(c) at 20 mA intervals. In Fig. 2(d), the deconvolution fit of the EL spectrum measured at 100 mA indicates that the origin of emission is related to the transition of both the band-edge and defectrelated traps. The broad peak in the visible range can be deconvoluted into two Lorentzian fits centered at approximately 520 and 670 nm, which indicates that the visible emission can be attributed to oxygen vacancies (VO) and oxygen interstitials (Oi) [24].

Figure 2. (a) Device structure of MgZnO/ZnO/GaN heterojunction diodes. (b) I−V curve of MgZnO/ZnO/GaN heterojunction diodes, showing the rectification under both forward and reverse bias. (c) EL spectra as a function of the injection current with an interval of 20 mA under the forward bias. (d) Deconvolution fits of EL spectrum at the injection current of 100 mA.

3.3. Discussion

Figure 3(a) displays a log-scale plot of the I−V curve under forward bias with the turn-on voltage estimated at 5.358 V. The linear dependence is dominant in the low-bias range below 5.358 V, indicating that direct tunneling through the MgZnO barrier was the main source of carrier transport [25]. At a high bias above 5.358 V, the trend is no longer expressed through monotonic linear dependence, as indicated by the red and blue dotted lines. For a bias range above 6 V, the inverse voltage (1/V) versus ln(I/V2) plot in Fig. 3(b) illustrates a linear dependence. This suggests that the carrier transport is dominated by Fowler-Nordheim (FN) tunneling [26]. In contrast, under a reverse bias, the log-scale I−V plot features three regimes described by different linear dependencies, as shown in Fig. 3(c). The reverse bias confirms that the electrons were injected from the contact metal into ZnO through the MgZnO barrier. The three different regions correspond to the direct, trap-assisted, and FN tunneling transitions [27].

Figure 3. (a) Log plot of I−V curve for the forward bias. (b) The inverse voltage (1/V) versus ln(I/V2) plot for the positive bias range above 6 V. (c) Log plot of I−V curve under reverse bias.

The relationship between light emission from the MgZnO/ZnO/GaN diodes and the corresponding carrier transport mechanism explains why light is emitted only under a forward bias. The band alignments under forward and reverse bias are illustrated in Fig. 4. Under forward bias, while the electrons are confined at the interface between the MgZnO and ZnO layer, the holes tunnel through the MgZnO barrier under a positive bias, which causes the electrons and holes to recombine at the band edge or defect traps at the MgZnO/ZnO interface. However, under a reverse bias, the electrons which are provided via tunneling through the MgZnO barrier are difficult to confine to the MgZnO/ZnO interface. In the ZnO, due to the higher mobility of electrons compared to holes, the band alignment under the reverse bias makes it more challenging for electrons and hole to recombine in the ZnO active layer [28]. These phenomena explain why EL emission was not observed under a reverse bias. The band alignment under different bias conditions allows us to understand the relationship between carrier transport and the light emission mechanism in MgZnO/ZnO/GaN diodes.

Figure 4. Band diagram of MgZnO/ZnO/GaN heterojunction diode under both forward and reverse bias conditions. The holes injected into the MgZnO/ZnO interface via the tunneling can recombine with the electrons confined by the MgZnO barrier only under forward bias.

We successfully fabricated a MgZnO/ZnO/GaN heterojunction diode via a layer-by-layer growth approach using RF sputtering and MOCVD. The MgZnO/ZnO-based heterojunction diode exhibited rectifying I−V characteristics for both forward and reverse bias, indicating that the tunneling current of the electrons and holes through the energy barrier is the primary transport mechanism. In addition, weak band-edge and strong visible emissions were observed in the EL spectrum only under forward bias. Our systematic analysis revealed that holes injected into the MgZnO/ZnO interface via tunneling can effectively recombine with electrons confined by the MgZnO barrier. Our results provide a potential approach for optoelectronic devices using MgZnO/ZnO-based heterostructures and their applicability to ZnObased light emitters.

This research was supported by the Basic Science Research Program (No. 2022R1C1C1004981) through the National Research Foundation of Korea and by a grant (00144108) funded by the Ministry of Trade, Industry, and Energy of the Korean government. B.-H. K. acknowledges the GIST Central Research Facilities (GCRF).

  1. M. A. Borysiewicz, Crystals 9, 505 (2019).
  2. R. Vittal and K.-C. Ho, Renew. Sust. Energ. Rev. 70, 920 (2017).
  3. Y.-S. Choi, J.-W. Kang, D.-K. Hwang, and S.-J. Park, IEEE Trans. Electron. Devices 57, 26 (2010).
  4. B. Kumar and S.-W. Kim, Nano Energy 1, 342 (2012).
  5. H. Hakkoum, A. Moumen, M. Ghougali, N. Sengouga, and E. Comini, J. Mater Sci. Mater. Electron. 33, 26604 (2022).
  6. B. Chakraborty, R. Saha, S. Chattopadhyay, D. De, R. D. Das, M. K. Mukhopadhyay, M.Palit, and C. RoyChaudhuri, Appl. Surf. Sci. 537, 147895 (2021).
  7. S. H. Ko, D. Lee, H. W. Kang, K. H. Nam, J. Y. Yeo, S. J. Hong, C. P. Grigoropoulos, and H. J. Sung, Nano Lett. 11, 666 (2011).
    Pubmed CrossRef
  8. K. Jeronimo, V. Koutosos, R. Cheung, and E. Mastropaolo, Sensors 21, 5873 (2021).
    Pubmed KoreaMed CrossRef
  9. S. Kim, G. Singh, Y. Jeong, S. Lee, and K. Lee, Adv. Mater. Technol. 8, 2201895 (2023).
  10. A. S. Alshammari, Z. R. Khan, M. Gandouzi, M. Mohamed, M Bouzidi, M Shkir, and H. M. Alshammari, Opt. Mater. 126, 112146 (2022).
  11. T.-H. Kim, M.-C. Jeong, W. Lee, and J.-M. Myeong, Appl. Surf. Sci. 240, 280 (2005).
  12. Y. Duan, S. Zhang, M. Cong, D. Jiang, Q. Liang, and X. Zhao, J. Mater. Chem. C 8, 12917 (2020).
  13. Y.-S. Choi, J.-W. Kang, B.-H. Kim, and S.-J. Park, ECS J. Solid State Sci. Technol. 2, R21 (2013).
  14. V. Awasthi, S. K. Pandey, S. Verma, and S. Mukherjee, J. Lumin. 180, 204 (2016).
  15. M. A. Khan, P. Kumar, G. Siddharth, M. Das, and S. Mukherjee, IEEE Trans. Electron. Devices 66, 5097 (2019).
  16. D. Thapa, J. Huso, J. Lapp, N. Rajabi, J. Morrison, M. D. Mc-Culskey, and L. Bergman, J. Mater. Sci. Mater. Electron. 29, 16782 (2018).
  17. J. Zuniga-Perez, V. Munoz-Sanjose, M. Lorez, G. Benndorf, S. Heitsch, D. Spemann, and M. Grundmann, J. Appl. Phys. 99, 023514 (2006).
  18. R. Pandey, A. Munshi, T. Shimpi, A. Shah, A. Bothwell, D. Kuciauskas, and J. R. Sites, Sol. RRL 5, 2100126 (2021).
  19. S. Shiau Jr, S. Brahma, C.-P. Liu, and J.-L Huang, Thin Solid Films 620, 170 (2016).
  20. S.-K. Hong, T. Hanada, H.-J. Ko, Y. Chen, T. Yao, D. Imai, K. Araki, and M. Shinohara, Appl. Phys. Lett. 77, 3571 (2000).
  21. S. Choopun, R. D. Vispute, W. Yang, R. P. Sharma, T. Venkatesan, and H. Shen, Appl. Phys. Lett. 80, 1529 (2002).
  22. R. Labar and T. K. Kundu, J. Electron. Mater. 51, 223 (2022).
  23. J.-H. Chen, K.-C. Chen, and J.-G. Hwu, AIP Adv. 12, 045116 (2022).
  24. N. H. Alvi, K. Hasan, O. Nur, and M. Willander, Nanoscale Res. Lett. 6, 130 (2011).
    Pubmed KoreaMed CrossRef
  25. Z. Cui, Y. He, H. Tian, A. Khanaki, L. Xu, W. Shi, and J. Liu, ACS Appl. Electron. Mater. 2, 747 (2020).
  26. J.-W. Kang, Y.-S. Choi, B.-H. Kim, C. G. Kang, B. H. Lee, C. W. Tu, and S.-J. Park, Appl. Phys. Lett. 104, 051120 (2014).
  27. C. Schleich, D. Waldhor, T. Knobloch, W. Zhou, B. Stampfer, J. Mich, M. Walt, and T. Grasser, IEEE Trans. Electron. Devices 69, 4479 (2022).
  28. L.-C. Chen, C.-H. Hsu, X. Zhang, and J.-R. Wu, Int. J. Photoenergy 2014, 959620 (2014).

Share this article on :

Stats or metrics