Energy band alignment of SiO2/ZnO interface ...

32 downloads 0 Views 420KB Size Report
J. B. You,a) X. W. Zhang, H. P. Song, J. Ying, Y. Guo, A. L. Yang, Z. G. Yin, N. F. Chen, and Q. S. ... electroluminescence from n-ZnO/SiO2/p-GaN was attained.
JOURNAL OF APPLIED PHYSICS 106, 043709 共2009兲

Energy band alignment of SiO2 / ZnO interface determined by x-ray photoelectron spectroscopy J. B. You,a兲 X. W. Zhang, H. P. Song, J. Ying, Y. Guo, A. L. Yang, Z. G. Yin, N. F. Chen, and Q. S. Zhu Key Laboratory of Semiconductor Materials Science, Institute of Semiconductors, CAS, Beijing 100083, People’s Republic of China

共Received 18 May 2009; accepted 14 July 2009; published online 26 August 2009兲 Thin SiO2 interlayer is the key to improving the electroluminescence characteristics of light emitting diodes based on ZnO heterojunctions, but little is known of the band offsets of SiO2 / ZnO. In this letter, energy band alignment of SiO2 / ZnO interface was determined by x-ray photoelectron spectroscopy. The valence band offset ⌬EV of SiO2 / ZnO interface is determined to be 0.93⫾ 0.15 eV. According to the relationship between the conduction band offset ⌬EC and the valence band offset ⌬EV: ⌬EC = ESiO2 − EZnO g g − ⌬EV, and taking the room-temperature band-gaps of 9.0 and 3.37 eV for SiO2 and ZnO, respectively, a type-I band-energy alignment of SiO2 / ZnO interface with a conduction band offset of 4.70⫾ 0.15 eV is found. The accurate determination of energy band alignment of SiO2 / ZnO is helpful for designing of SiO2 / ZnO hybrid devices and is also important for understanding their carrier transport properties. © 2009 American Institute of Physics. 关DOI: 10.1063/1.3204028兴 INTRODUCTION

ZnO has a wide and direct band-gap of 3.37 eV at roomtemperature and a higher exciton binding energy 共60 meV兲 compared with other wide band-gap semiconductors, such as GaN 共28 meV兲 and ZnSe 共19 meV兲, which leads to great potential applications in optoelectronic devices and ultraviolet 共UV兲 laser devices.1–4 Although ZnO-based p-n homojunctions light emitting diodes 共LEDs兲 have been fabricated and weak electroluminescence were obtained.5–7 ZnO still suffers from the lack of a reproducible, high-quality, p-type epitaxial growth technology, which seriously hampers the progress of LEDs based on ZnO homojunctions.8 As an alternative approach, there have been numerous attempts to develop p-n heterojunction LEDs with ZnO as the n-type layer, such as n-ZnO/ p-GaN,9 n-ZnO/ p共n兲-Si.10 For these heterojunctions, in order to confine electrons in n-ZnO layers and guarantee that the radiative recombination of electronhole pairs occurs in the ZnO layers, a high electron barrier existing at the side of ZnO is required. As a consequence, a SiO2 interlayer with wide band-gap was usually inserted in ZnO-based heterojunctions as an electron block layer.11–14 For example, Chen et al.11 demonstrated that the 394 nm electroluminescence from n-ZnO/ SiO2 / p-GaN was attained by controlling the recombination at the ZnO/ SiO2 interface using SiO2 interlayer. Chen et al.12 and Sun and coworkers13,14 showed that the obvious electroluminescence including UV and visible emission can be achieved from n-ZnO/ Si after inserting SiO2 into the heterojunctions. These results indicated that the thin SiO2 interlayer is the key for improving the electroluminescence characteristics of LEDs based on ZnO heterojunctions. In addition, SiO2 was also utilized as an insulation or a passivation layer for ZnO-based a兲

Author to whom correspondence should be addressed. Electronic mail: [email protected].

0021-8979/2009/106共4兲/043709/5/$25.00

light emitting devices.15–17 For these SiO2 / ZnO related heterostructures, determination of interface band alignment is important to understand their electronic transport mechanism since the carrier transport and confinement properties at the interface depends on the valence and conduction band offsets 共CBOs兲.18 In previous works, the energy band alignments of SiO2 / GaN and SiO2 / Si have been extensively investigated.18–20 However, up to now, the energy band alignment of SiO2 / ZnO has not been studied. In this letter, energy band alignment of SiO2 / ZnO interface was determined by x-ray photoelectron spectroscopy 共XPS兲. The valence band offset 共VBO兲 is determined to be 0.93⫾ 0.15 eV. According to the relationship between the CBO ⌬EC and the − EZnO VBO ⌬EV: ⌬EC = ESiO2 g g − ⌬EV, and taking the roomtemperature band-gaps of 9.0 and 3.37 eV for SiO2 and ZnO, respectively, a type-I band-energy alignment of SiO2 / ZnO interface with a CBO of 4.70⫾ 0.15 eV is found.

EXPERIMENT

Three samples were used in our XPS experiments, namely, a 300-nm-thick SiO2 film, a 1000-nm-thick ZnO film, and a 5-nm-thick SiO2 layer on a 1000-nm-thick ZnO 共SiO2 / ZnO heterojunction兲, and all of them were grown on Si 共111兲 substrates. The SiO2 and the ZnO films were deposited by radio frequency 共rf兲 sputtering high-purity 共99.99%兲 SiO2 and ZnO ceramic target, respectively. The SiO2 films were grown with an rf power of 80 W at 600 ° C, and the thicknesses were controlled by varying deposition time. The detailed growth conditions of the ZnO films can be found in our previous reports.21,22 Carrier concentration and mobility of the ZnO films on insulated sapphire substrates were 4 ⫻ 1016 cm−3 and 40 cm2 / V s, respectively, which were determined by Hall measurements using the Van der Pauw configuration. These results indicate that the densities of intrin-

106, 043709-1

© 2009 American Institute of Physics

Downloaded 13 Oct 2009 to 159.226.228.19. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

043709-2

You et al.

J. Appl. Phys. 106, 043709 共2009兲

FIG. 1. 共Color online兲 Si 2p CL spectra for 共a兲 SiO2 and 共e兲 SiO2 / ZnO samples and Zn 2p3/2 CL spectra for 共c兲 ZnO and 共f兲 SiO2 / ZnO samples. Experimental data points are fitted by Voigt 共mixed Lorentzian–Gaussian兲 line shapes 共solid lines, red兲 after the application of a Shirley background 共solid line, blue兲. Also VB spectra for 共b兲 SiO2 and 共d兲 ZnO. The peak and VBM positions are summarized in Table I.

sic defect and grain boundary in the ZnO films are much low, and thus the ZnO films are high crystal quality.2,22 The XPS measurements were carried out on a PHI Quantera SXM instrument with Al K␣ 共energy of 1486.6 eV兲 as the x-ray radiation source, which had been carefully calibrated utilizing Fermi energy level 共EF兲 and work function 共⌽sp兲. Here, the position of the Fermi level 共EF兲 was calibrated by measuring the Fermi edge of the polycrystalline gold metal film. Also, the work function of the spectrometer 共⌽sp = 4.5 eV兲 was calibrated by using the positions of Au 4f 7/2 共84.0 eV兲 of the polycrystalline gold metal film. During the measurements, because a large number of electrons are excited and emitted from the sample, the sample is always positively charged. Since the electric field caused by the charge can affect the measured kinetic energy of the photoelectron, charge neutralization was performed with an electron flood gun, where the filament current and the emitting

electron energy is about 20 ␮A and 1–2 eV, respectively. To compensate the possible residual charge effect, all XPS spectra were calibrated by the C 1s peak at 284.8 eV from surface contamination.23 The details can be described as follows, first, the deviation of the C 1s peak from 284.8 eV was recorded for each sample, and then all the XPS spectra for each sample were calibrated by the corresponding deviations. The surfaces of the specimens were examined initially by low-resolution survey scans to determine which elements were present. High-resolution spectra were acquired to determine the binding energy 共i.e., chemical state兲 in the survey spectra. RESULTS AND DISCUSSION

According to Kraut’s method,24 the VBO of SiO2 / ZnO interface can be calculated by the following formula:

Downloaded 13 Oct 2009 to 159.226.228.19. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

043709-3

J. Appl. Phys. 106, 043709 共2009兲

You et al.

TABLE I. Binding energies 共in electron volt兲 of the XPS peaks and VBM for the SiO2, ZnO, and SiO2 / ZnO samples. Energies are referenced to the Fermi level 共0 eV兲. The errors in the peak positions and VBM are ⫾0.05 and ⫾0.10 eV, respectively.

Sample SiO2

ZnO SiO2 / ZnO

State

Binding energy 共eV兲

Si 2p Si 2s VBM Zn 2p3/2 VBM Si 2p Si 2s Zn 2p3/2

103.18 154.19 4.47 1021.03 1.90 102.84 153.90 1022.36

SiO2/ZnO SiO2/ZnO ZnO ⌬EV = 共ESi − EZn 兲 + 共EZn 2p 2p 2p 3/2

SiO2 SiO2 − 共ESi 2p − EVBM兲,

3/2

ZnO − EVBM 兲

共1兲

where Esi denotes the energy of feature i in sample s. The Zn 2p3/2 and Si 2p core levels 共CL兲 have been used in Eq. 共1兲, but equally the Zn 2p3/2 and Si 2s peaks can be used.25 Figure 1 shows Si 2p CL spectra from the SiO2 film and the SiO2 / ZnO interface, Zn 2p3/2 CL spectra from the ZnO film and the SiO2 / ZnO interface, as well as valence band 共VB兲 spectra of the SiO2 and the ZnO films. The results with Si 2s CL spectra will be shown afterward. The CL spectra were fitted to Voigt 共mixed Lorentzian–Gaussian兲 line shape by employing a Shirley background.25,26 The binding energy for the CL peaks were taken as the energy corresponding to their maximum intensity.25 The VB maximum 共VBM兲 positions were determined by linear extrapolation of the leading edges of the VB spectra recorded from the SiO2 and the ZnO films to the base lines in order to account for the tail induced by instrument resolution.25,26 Since considerable accordance of the fitted lines to the original measured data has been obtained, the uncertainty of the CL positions should be lower than 0.05 eV, as evaluated by numerous fittings with different parameters. The main uncertainty of the VBO value comes from the difficulty in determining the VBM position exactly, the scatter of the data relative to the fit are estimated as an uncertainty in VBM positions of less than 0.1 eV, as also estimated from fitting with different parameters. The param-

eters deduced from Fig. 1 are summarized in Table I for clarity. The CL spectrum of Si 2p measured from the SiO2 film is shown in Fig. 1共a兲, and the unique symmetric peak located at 103.18 eV corresponds to the Si–O bond indicating the uniform bonding state. A VBM value of 4.47⫾ 0.10 eV is deduced from the VB spectra of the SiO2 film 关Fig. 1共b兲兴 by linear fitting depicted above. For a good insulator, the Fermi energy level is located in the middle of the forbidden energy gap. Thus the VBM, representing the distance between the Fermi energy level and the top of the VB, will be one-half of the band-gap of insulators. Specifically, the VBM of SiO2 should be 4.50 eV based on its bandgap of 9.0 eV, this value is good consistent with the measured value 共4.47⫾ 0.10 eV兲 in the present experiment. The CL spectrum of Zn 2p3/2 from the ZnO film is shown in Fig. 1共c兲, which can be well fitted using Voigt fitting method. The only Zn 2p3/2 peak located at 1021.03 eV is attributed to the Zn–O bond.26 According to the VB spectrum of the ZnO film in Fig. 1共d兲, a value of VBM of 1.90⫾ 0.10 eV is obtained by using the same method as SiO2. These values from the ZnO film are agreement with the results reported by Yao et al.27 The energy difference between Zn 2p3/2 and VBM of ZnO ZnO the ZnO film 共EZn 2p3/2 − EVBM兲 is 1019.13 eV, which is also close to our previous reported results where the ZnO films were deposited by metal-organic chemical vapor deposition.26 The CL spectra of Si 2p and Zn 2p3/2 from the SiO2 / ZnO interface are shown in Figs. 1共e兲 and 1共f兲, respectively. Compared with the spectra of the SiO2 and the ZnO films, the Si 2p peak corresponding to Si–O bond in the SiO2 / ZnO interface is shifted 0.34 eV to a binding energy of 102.84 eV, while the Zn 2p3/2 CL peak is shifted to 1022.36 eV. The VBO is calculated to be 0.90⫾ 0.15 eV by substituting those values obtained in the experiment into Eq. 共1兲. In most previous reports, only one combination of CLs was used to determine the VBO. To improve the accuracy of the VBO value determined by XPS, combination of various CLs was adopted by Kinget al.,28 where the VBO value of the InN/GaN heterojunction was determined by a number of In and Ga CLs and Auger peaks.28 In this study, the combination of Zn 2p3/2 and Si 2s peaks is also used to determine the VBO of SiO2 / ZnO interface. The Si 2s CL spectra recorded from the SiO2 film and the SiO2 / ZnO interface are shown in Figs. 2共a兲 and 2共b兲, respectively. As above men-

FIG. 2. 共Color online兲 Si 2s CL spectra for 共a兲 SiO2 and 共b兲 SiO2 / ZnO samples. Experimental data points are fitted by Voigt 共mixed Lorentzian–Gaussian兲 line shapes 共solid lines, red兲 after the application of a Shirley background 共solid line, blue兲. The peak positions are also summarized in Table I.

Downloaded 13 Oct 2009 to 159.226.228.19. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

043709-4

J. Appl. Phys. 106, 043709 共2009兲

You et al.

interfaces are responsible for the small difference between our measured value and the results deduced from the transitive property. CONCLUSIONS

In summary, the VBO of the SiO2 / ZnO interface was determined by XPS. It was found that a type-I heterojunction forms between SiO2 and ZnO with a VBO value of 0.93⫾ 0.15 eV and a CBO of 4.70⫾ 0.15 eV. Knowing the band alignment parameters of the SiO2 / ZnO interface will facilitate the understanding their electronic transport mechanism and the design of SiO2 / ZnO hybrid devices. ACKNOWLEDGMENTS FIG. 3. 共Color online兲 Energy band diagram of SiO2 / ZnO interface. A type-I heterojunction is formed in the staggered arrangement.

tioned, the CL peaks were fitted using a Shirley background and Voigt line shapes. The corresponding positions of Si 2s CL in the SiO2 film and the SiO2 / ZnO interface are also summarized in Table I. Similarly, the VBO of SiO2 / ZnO is calculated to be 0.95⫾ 0.15 eV. It can be seen that the VBO values calculated for the two combinations of CLs are well within experimental error, indicating the results attained in our experiment are reliable. To reduce the experimental error, the VBO value is calibrated to be 0.93⫾ 0.15 eV by averaging the two VBO values. As XPS measurements are spatially averaged due to the finite mean free path of elastic electrons, band bending could induce a systematic error in our measurements. The error is checked to be much smaller than the average standard deviation of 0.15 eV given above.26 Finally, the CBO ⌬EC can be estimated by the formula ⌬EC = ESiO2 g − EZnO g − ⌬EV. Taking the room-temperature band-gaps of 9.0 eV 共Ref. 18兲 and 3.37 eV1–3 for SiO2 and ZnO, respectively, the SiO2 / ZnO interface is found to have a type-I band lineup, as shown in Fig. 3, with a corresponding CBO of 4.70⫾ 0.15 eV. The large CBO is sufficient to provide significant electron confinement potentials for hybrid SiO2 / ZnO heterojunctions. To further confirm the reliability of the experimentally obtained values, it is necessary to compare our data with other experimental results. Although the VBO of SiO2 / ZnO interface has not been measured by XPS in the past, it can be deduced from other VBO values by transitive property which was proposed by Wei and Zunger.29 For heterojunctions formed between three materials 共A, B, and C兲, in the case of neglecting the effects of interfaces on the band offset, if ⌬EV 共A-B兲 and ⌬EV 共B-C兲 are known, ⌬EV 共A-C兲 can be deduced from the difference between ⌬EV 共A-B兲 and ⌬EV 共B-C兲. From the reported experimental results, ⌬EV 共SiO2 – GaN兲 is 2.0⫾ 0.20 eV,18 ⌬EV 共GaN–ZnO兲 is 1.37⫾ 0.20 共Ref. 25兲 or 1.0 eV,30 and then the ⌬EV 共SiO2 − ZnO兲 is deduced to be 0.63⫾ 0.20 or 1.0⫾ 0.20 eV, which is comparable to our measured value 共0.93⫾ 0.15 eV兲. Since the samples were prepared under the different growth conditions, the different

This work was financially supported by the National Natural Science Foundation of China 共Grant Nos. 50601025 and 60876031兲 and the “863” project of China 共Grant No. 2009AA03Z305兲. One of the authors 共J.B.Y.兲 thanks the CAS Special Grant for Postgraduate Research, Innovation and Practice. D. C. Look and B. Claflin, Phys. Status Solidi B 241, 624 共2004兲. D. K. Hwang, M. S. Oh, J. H. Lim, and S. J. Park, J. Phys. D 40, R387 共2007兲. 3 U. Ozgur, Ya. I. Alivov, C. Liu, A. Teke, M. A. Reshchikov, S. Dogan, V. Avrutin, S. J. Cho, and H. Morkoc, J. Appl. Phys. 98, 041301 共2005兲. 4 P. Wang, N. F. Chen, Z. G. Yin, R. X. Dai, and Y. M. Bai, Appl. Phys. Lett. 89, 202102 共2006兲. 5 A. Tsukazaki, T. Onuma, M. Ohtani, T. Makino, M. Sumiya, K. Ohtani, S. F. Chichibu, S. Fuke, Y. Segawa, H. Ohno, H. Koinuma, and M. Kawasaki, Nature Mater. 4, 42 共2005兲. 6 J. H. Lim, C. K. Kang, K. K. Kim, I. K. Park, D. K. Hwang, and S. J. Park, Adv. Mater. 共Weinheim, Ger.兲 18, 2720 共2006兲. 7 H. S. Kim, F. Lugo, S. J. Pearton, D. P. Norton, Y. L. Wang, and F. Ren, Appl. Phys. Lett. 92, 112108 共2008兲. 8 R. Deng, B. Yao, Y. F. Li, Y. M. Zhao, B. H. Li, C. X. Shan, Z. Z. Zhang, D. X. Zhao, J. Y. Zhang, D. Z. Shen, and X. W. Fan, Appl. Phys. Lett. 94, 022108 共2009兲. 9 Ya. I. Alivov, J. E. Van Nostrand, D. C. Look, M. V. Chukichev, and B. M. Ataev, Appl. Phys. Lett. 83, 2943 共2003兲. 10 P. L. Chen, X. Y. Ma, and D. R. Yang, J. Appl. Phys. 101, 053103 共2007兲. 11 C. P. Chen, M. Y. Ke, C. C. Liu, Y. J. Chang, F. H. Yang, and J. J. Huang, Appl. Phys. Lett. 91, 091107 共2007兲. 12 P. L. Chen, X. Y. Ma, D. S. Li, Y. Y. Zhang, and D. R. Yang, Appl. Phys. Lett. 94, 061110 共2009兲. 13 J. L. Zhao, X. W. Sun, S. T. Tan, G. Q. Lo, D. L. Kwong, and Z. H. Cen, Appl. Phys. Lett. 91, 263501 共2007兲. 14 S. T. Tan, X. W. Sun, J. L. Zhao, S. Iwan, Z. H. Cen, T. P. Chen, J. D. Ye, G. Q. Lo, D. L. Kwong, and K. L. Teo, Appl. Phys. Lett. 93, 013506 共2008兲. 15 P. L. Chen, X. Y. Ma, and D. R. Yang, Appl. Phys. Lett. 89, 111112 共2006兲. 16 X. Y. Ma, P. L. Chen, D. S. Li, Y. Y. Zhang, and D. R. Yang, Appl. Phys. Lett. 91, 021105 共2007兲. 17 Y. L. Wang, H. S. Kim, D. P. Norton, S. J. Pearton, and F. Ren, Appl. Phys. Lett. 92, 112101 共2008兲. 18 T. E. Cook, Jr., C. C. Fulton, W. J. Mecouch, K. M. Tracy, R. F. Davis, E. H. Hurt, G. Lucovsky, and R. J. Nemanich, J. Appl. Phys. 93, 3995 共2003兲. 19 B. R. Tuttle, Phys. Rev. B 70, 125322 共2004兲. 20 J. L. Alay and M. Hirose, J. Appl. Phys. 81, 1606 共1997兲. 21 J. B. You, X. W. Zhang, Y. M. Fan, S. Qu, and N. F. Chen, Appl. Phys. Lett. 91, 231907 共2007兲. 22 J. B. You, X. W. Zhang, Y. M. Fan, Z. G. Yin, P. F. Cai, and N. F. Chen, Appl. Surf. Sci. 255, 5876 共2009兲. 23 J. J. Chen, F. Ren, Y. J. Li, D. P. Norton, S. J. Pearton, A. Osinsky, J. W. Dong, P. P. Chow, and J. F. Weaver, Appl. Phys. Lett. 87, 192106 共2005兲. 1 2

Downloaded 13 Oct 2009 to 159.226.228.19. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp

043709-5 24

You et al.

E. Kraut, R. Grant, J. Waldrop, and S. Kowalczyk, Phys. Rev. Lett. 44, 1620 共1980兲. 25 T. D. Veal, P. D. C. King, S. A. Hatfield, L. R. Bailey, C. F. McConville, B. Martel, J. C. Moreno, E. Frayssinet, F. Semond, and J. Zúñiga-Pérez, Appl. Phys. Lett. 93, 202108 共2008兲. 26 A. L. Yang, H. P. Song, H. Y. Wei, X. L. Liu, J. Wang, X. Q. Lv, P. Jin, S. Y. Yang, Q. S. Zhu, and Z. G. Wang, Appl. Phys. Lett. 94, 163301 共2009兲. 27 Y. F. Li, B. Yao, Y. M. Lu, B. H. Li, Y. Q. Gai, C. X. Cong, Z. Z. Zhang,

J. Appl. Phys. 106, 043709 共2009兲 D. X. Zhao, J. Y. Zhang, D. Z. Shen, and X. W. Fan, Appl. Phys. Lett. 92, 192116 共2008兲. 28 P. D. C. King, T. D. Veal, C. E. Kendrick, L. R. Bailey, S. M. Durbin, and C. F. McConville, Phys. Rev. B 78, 033308 共2008兲. 29 S. H. Wei and A. Zunger, Appl. Phys. Lett. 72, 2011 共1998兲. 30 S. K. Hong, T. Hanada, H. Makino, H. J. Ko, Y. Chen, A. Tanaka, H. Saski, S. Sato, D. Imai, K. Araki, and M. Shinohara, J. Vac. Sci. Technol. B 19, 1429 共2001兲.

Downloaded 13 Oct 2009 to 159.226.228.19. Redistribution subject to AIP license or copyright; see http://jap.aip.org/jap/copyright.jsp