Effects of ultraviolet treatment on the contact resistivity and electronic transport at the
Ti/ZnO interfaces
Yow-Jon Lin, Chia-Lung Tsai, W.-R. Liu, W. F. Hsieh, C.-H. Hsu, Hou-Yen Tsao, Jian-An Chu, and Hsing-Cheng Chang
Citation: Journal of Applied Physics 106, 013701 (2009); doi: 10.1063/1.3157201
View online: http://dx.doi.org/10.1063/1.3157201
View Table of Contents: http://scitation.aip.org/content/aip/journal/jap/106/1?ver=pdfcov Published by the AIP Publishing
Articles you may be interested in
Comparison of Ti Al Pt Au and Ti Au Ohmic contacts on n -type ZnCdO Appl. Phys. Lett. 88, 012109 (2006); 10.1063/1.2161927
Low-resistance ohmic contacts to p - Zn Mg O grown by pulsed-laser deposition Appl. Phys. Lett. 86, 192103 (2005); 10.1063/1.1925309
Carrier concentration dependence of Ti/Al/Pt/Au contact resistance on n-type ZnO Appl. Phys. Lett. 84, 544 (2004); 10.1063/1.1644318
Specific contact resistance of Ti/Al/Pt/Au ohmic contacts to phosphorus-doped ZnO thin films J. Vac. Sci. Technol. B 22, 171 (2004); 10.1116/1.1641060
Low-resistance Ti/Au ohmic contacts to Al-doped ZnO layers Appl. Phys. Lett. 77, 1647 (2000); 10.1063/1.1308527
Effects of ultraviolet treatment on the contact resistivity and electronic
transport at the Ti/ZnO interfaces
Yow-Jon Lin,1,a兲 Chia-Lung Tsai,2W.-R. Liu,3,4W. F. Hsieh,3C.-H. Hsu,3,4Hou-Yen Tsao,1 Jian-An Chu,1and Hsing-Cheng Chang5
1
Institute of Photonics, National Changhua University of Education, Changhua 500, Taiwan 2Department of Physics, National Changhua University of Education, Changhua 500, Taiwan
3Department of Photonics and Institute of Electro-optical Engineering, National Chiao Tung University, Hsinchu 300, Taiwan
4Research Division, National Synchrotron Radiation Research Center, Hsinchu 300, Taiwan 5Department of Automatic Control Engineering, Feng Chia University, Taichung 407, Taiwan
共Received 31 March 2009; accepted 26 May 2009; published online 1 July 2009兲
We report on the effect of ultraviolet 共UV兲 treatment on the specific contact resistance 共兲 and electronic transport at the Ti/ZnO interfaces. The experimental results show the same barrier height of Ti/ZnO samples without UV treatment as Ti/ZnO samples with UV treatment and the higherof Ti/ZnO samples with UV treatment than Ti/ZnO samples without UV treatment, suggesting the barrier-height independence of. Based on the thermionic-emission model and x-ray photoelectron spectroscopy results, we found that the induced decrease in the number of the hydroxides at the surface region of ZnO by UV treatment resulted in decreases in the electron concentration near the surface region and the excess current component related to tunneling, increasing in of Ti/ZnO samples. © 2009 American Institute of Physics.关DOI:10.1063/1.3157201兴
I. INTRODUCTION
ZnO is a versatile material. ZnO has a direct band gap at room temperature with a large exciton binding energy and a strong cohesive energy. Its applications include antireflection coatings, transparent conducting electrodes, light-emitting diodes, flexible displays, and solar cells.1–8The achievement of acceptable device characteristics relies heavily on devel-oping low specific contact resistance共兲 Ohmic metallization schemes.9 To produce low-resistance Ohmic contacts, a fairly high annealing temperature is required. However, ther-mal annealing at high temperature may cause surface rough-ening, surface decomposition, and spiky interfaces resulting in the deterioration of device performance and hence device reliability.10,11 To fabricate high-performance optoelectronic devices, low-resistance nonalloyed Ohmic contacts are es-sential. Treatment of the ZnO surface prior to any metal deposition has been used to decrease the contact resistivity.12,13 In the paper, we report on the effect of ultra-violet共UV兲 treatment on the contact resistivity of nonalloyed Ti Ohmic contacts to ZnO films grown by pulsed-laser depo-sition共PLD兲. The mechanisms of Ti Ohmic contacts to ZnO films with and without UV treatment were also investigated in this study. According to the experimental results, we found the same barrier height共B兲 of Ti/ZnO samples without UV
treatment as Ti/ZnO samples with UV treatment and the higher of Ti/ZnO samples with UV treatment than Ti/ZnO samples without UV treatment, suggesting the barrier-height independence of. Based on the thermionic-emission 共TE兲 model and x-ray photoelectron spectroscopy 共XPS兲 results, we found that the induced decrease in the number of the hydroxides at the surface region of ZnO by UV treatment
resulted in decreases in the electron concentration near the surface region and the excess current component related to tunneling, increasing of Ti/ZnO samples.
II. EXPERIMENT PROCEDURE
The 600 nm ZnO film grown on a共0001兲 sapphire sub-strate by PLD is used in this study. In previous studies,14–20 undoped and doped ZnO films have been fabricated by PLD. A KrF excimer laser共wavelength of 248 nm兲 was employed and the beam was focused to produce an energy density of 5 – 7 J cm−2 at 10 Hz repetition rate on a commercial hot pressed stoichiometric ZnO 共99.999% purity兲 target. The films were deposited at a growth rate of 94.25 nm/hr at 550 ° C substrate temperature and a base vacuum of 1.2 ⫻10−8 Torr. No oxygen gas flow was introduced during the process of growth. These grown samples were cleaned with chemical solutions of trichloroethylene, acetone, and metha-nol. Then, the samples were rinsed with deionized water and immediately blown dry with N2. In our experiments, the ZnO sample was irradiated in air for 30 min by a UV light source 共output power density of 810 W/cm2兲 with emission cen-tered at 254 nm. Next, Ti was used as the electrode and vacuum evaporated on top of the ZnO samples with and without UV treatment. Ti was deposited using the sputter coater. X-ray diffraction 共XRD兲 was employed to identify crystalline phases. Using a Ne–Cu laser 共the 248.6 nm line兲 as an excitation source, the photoluminescence 共PL兲 band was observed for ZnO samples with and without UV treat-ment at room temperature. XPS was employed to examine the changes in the surface band bending of ZnO, chemical bonding states of ZnO, and the band bending in ZnO near the Ti/ZnO interface. In this study, the surface band bending was determined by location of the energy of the valence band maximum 共VBM兲 in photoemission from the sample. XPS a兲Author to whom correspondence should be addressed. Electronic mail:
JOURNAL OF APPLIED PHYSICS 106, 013701共2009兲
0021-8979/2009/106共1兲/013701/5/$25.00 106, 013701-1 © 2009 American Institute of Physics
system was equipped with monochromatic Al K␣source. We took a Au 4f7/2 peak or a C 1s peak for energy reference purposes. All binding energies are measured relative to the Fermi level 共EF兲. was measured using the circular
trans-mission line method 共CTLM兲21 for Ti contact to ZnO with and without UV treatment at various temperatures between 300 to 400 K. The CTLM pattern was designed with a con-stant outer radius r = 100 m and spacings of 2, 4, 8, 16, and 32 m. The current-voltage 共I-V兲 characteristics of the Ti/ ZnO samples were measured using a Keithley Model-4200 semiconductor characterization system.
III. EXPERIMENTAL RESULTS AND DISCUSSION Figure1shows PL spectra of the ZnO film without UV treatment. One emission peak at room temperature was ob-served for ZnO samples. The peak at 3.42 eV is near band edge emission, the so called ultraviolet luminescence 共UL兲. In Fig.1, we do not find the green luminescence共peak po-sition at ⬃2.5 eV兲22 related to the oxygen-vacancy-related emission.23 The inset of Fig.1 shows the UL spectra of the ZnO films with and without UV treatment. Compared with the ZnO films without UV treatment, we can see that the UL slightly shifted toward the low photon energy for ZnO films with UV treatment. An explanation for this will be given later. Figure 2 shows the XRD pattern of the ZnO film.19 Diffraction peaks appearing at about 34.4° and 30.9° corre-spond to those from ZnO 共002兲 and Zn共OH兲2 共100兲 planes.24,25 However, the intensity of 共100兲 peak is very weak. These results shown in Figs. 1 and 2 provide direct evidence on the deposition of the good-quality ZnO film.
Figure 3 shows the I-V characteristics of the Ti/ZnO samples with and without UV treatment, measured between metal pads with gap spacing of 8 m. The I-V characteris-tics of the Ti/ZnO samples with and without UV treatment show linear behavior. It indicates that the Ohmic perfor-mance can be obtained for Ti/ZnO samples. Then, was measured using the CTLM applied to a structure with gap spacing. The associated resistance as a function of gap
spac-ing for Ti/ZnO samples can be obtained.was determined to be 2.7⫻10−5 共4.1⫻10−6兲 ⍀ cm2 for Ti/ZnO samples with 共without兲 UV treatment.
To study the formation mechanism of nonalloyed Ohmic contacts for Ti/ZnO samples with and without UV treatment, XPS was employed to determine the VBM position 共Ev兲 of
ZnO samples and the position 共EZnOZn 3d兲 of Zn 3d core-level peak at the ZnO surfaces before Ti deposition. Figure 4
shows the Zn 3d core-level and valence-band spectra col-lected on a ZnO sample with or without UV treatment. Evis
determined by extrapolating two solid lines from the back-ground and straight cutoff in the spectra.26,27 For ZnO samples without UV treatment, Ev was measured to be 3.52
eV and EZnOZn 3dwas observed at 10.96 eV, suggesting that the energy difference共EVC兲 between EZnOZn 3dand EVis determined
to be 7.44 eV, which agrees with the previously reported values of 7.3–7.5 eV.28–30 For ZnO samples with UV treat-ment, Ev was measured to be 2.50 eV and EZnOZn 3d was ob-served at 10.24 eV, implying that EVC is equal to 7.74 eV.
The energy difference between EVCof ZnO samples with and
without UV treatment is attributed to the presence of dipole at the ZnO surface following UV treatment. In photoelectron FIG. 1. 共Color online兲 PL spectra of the ZnO films without UV treatment.
Inset: UL spectra of the ZnO films共a兲 without and 共b兲 with UV treatment.
FIG. 2. 共Color online兲 XRD 2scan of the ZnO films.
FIG. 3.共Color online兲 I-V characteristics of the Ti/ZnO samples 共a兲 without and共b兲 with UV treatment measured between metal pads with gap spacing of 8 m.
013701-2 Lin et al. J. Appl. Phys. 106, 013701共2009兲
spectroscopy, the existence of a dipole at polar heterojunc-tion can have a substantial influence on the measured binding energies of photoelectrons, depending on the origin of the collected photoelectrons being from below or above the heterojunction.31 Hong et al.28found the difference between VBM and Zn 3d level as 7.5 and 7.3 eV, respectively, before and after Ar+ion cleaning. As a consequence, the measured binding energy can be smaller or larger, depending on the orientation of interface dipole. If the band gap energy共Eg兲 of
ZnO was assumed to be 3.42 eV, EF before UV treatment
would be located at approximately 0.1 eV above the conduc-tion band minimum 共CBM兲 共that is, the formation of the downward band bending兲 and EF following UV treatment
would be located at approximately 0.92 eV below CBM共that is, the formation of the upward band bending兲.
Figure5shows the Zn 3d core-level XPS spectra at the Ti/ZnO interface with and without UV treatment. The peaks are determined by Gaussian fitting.B at the Ti/ZnO
inter-face was determined from the position 共ETiZn 3d/ZnO兲 of Zn 3d
core-level peak at the ZnO surfaces following Ti deposition and EVCaccording to Eq.共1兲.32Equation共1兲is expressed as
B= Eg−共ETiZn 3d/ZnO− EVC兲. 共1兲
According to the result shown in Fig.5, we find that ETi/ZnOZn 3d without共with兲 UV treatment is located at 10.56 共10.86兲 eV. According to the result shown in Fig. 4, we find that EVC
without 共with兲 UV treatment is equal to 7.44 共7.74兲 eV. Therefore,Bof the Ti/ZnO sample without共with兲 UV
treat-ment was calculated to be 0.30共0.30兲 eV. We find thatBof
the Ti/ZnO sample without UV treatment is equal toB of
the Ti/ZnO sample with UV treatment, suggesting that of the Ti/ZnO sample without UV treatment may be similar to of the Ti/ZnO sample with UV treatment. However, was determined to be 2.7⫻10−5 ⍀ cm2for Ti/ZnO samples with UV treatment which is higher than the determined value 共4.1⫻10−6 ⍀-cm2兲 for Ti/ZnO samples without UV treat-ment. An explanation for this will be given latter. On the other hand, we use the TE theory to calculate the B at the
Ti/ZnO interface. According to the TE theory, the relation-ship between andBshould be expressed as
= k
qAⴱTexp
冉
B
kT
冊
, 共2兲where Aⴱ共Aⴱ= 32 A cm−2K−2兲 共Ref.19兲 is the effective Ri-chardson constant of ZnO, q is the magnitude of the electron charge, k is the Boltzmann constant, and T is the absolute temperature. In Eq. 共2兲, the contribution to current due to tunneling through the barrier is completely neglected. For the Ti/ZnO samples with and without UV treatment, B values
are 0.21 and 0.16 eV共based on the TE theory兲, respectively, which are not in agreement with the results from the direct measurement using XPS, of 0.3 and 0.3 eV, respectively. This experimental finding cannot be explained through Eq.
共2兲. We propose that the origin of this result is tunneling through the barrier and is related to the depletion-layer width 共due to the higher B determined by XPS
measure-ment than the calculated value based on the TE theory兲. An explanation for this will be given later.
Figure6 shows O 1s core-level spectra at the ZnO sur-faces with and without UV treatment. Deconvolution of the O 1s core-level peak obtained via XPS studies of the ZnO surfaces without UV treatment revealed two peaks indicative of O–Zn and O–H共or O–OH兲 bonding,33–35respectively. The peaks are determined by Gaussian fitting. The binding energy shown here has been scaled with respect to the position of the O–Zn bond peak. In Refs. 33–35, the binding energy of O–Zn 共O–H or O–OH兲 bonds had been determined to be 531.3 共532.9兲 eV by XPS measurement. Therefore, the dif-ference in the O–Zn bond and the O–H bond energies is determined to be 1.6 eV. The XPS result is in agreement with the XRD result 关shown in Fig. 2兴. Comparing the Ti/ZnO
sample without UV treatment, we can see that the O–H bond peak in the XPS spectrum becomes a smaller fraction of the total O peak for the Ti/ZnO sample with UV treatment. First-principles calculations have provided evidence that H be-haves as a shallow donor in ZnO and can be incorporated into the thin film via the formation of O–H bonds at the surface during growth.34,36 Cai et al.37 suggested that the FIG. 4. 共Color online兲 The left-hand figure presents the spectrum of the
valence-band region and the right-hand spectrum shows the Zn 3d core-level spectra on the ZnO surfaces共a兲 without and 共b兲 with UV treatment.
FIG. 5. Zn 3d core-level spectra at the Ti/ZnO interfaces共a兲 without and 共b兲 with UV treatment.
013701-3 Lin et al. J. Appl. Phys. 106, 013701共2009兲
incorporated hydrogen not only passivates most of the de-fects, but also introduces shallow donor states. Gu et al.35 pointed out that the presence of the hydroxide constituent could lead to the formation of an accumulation layer having a high conductivity on the ZnO surface. Look et al.38found that the surface sheet carrier concentration increases with increasing the intensity of the O–H bond peak in the XPS spectrum. Mönch39 pointed out that the current transport across metal-semiconductor contacts occurs by TE over the barrier for doping levels of the semiconductor up to approxi-mately 1018donors per cm3, while above this limit tunneling through the then narrower depletion layers dominates. This provides the evidence that a number of O–H bonds, acting as donors for electrons, were existed at the ZnO surfaces, re-sulting in the formation of low-resistance nonalloyed Ohmic contacts for Ti/ZnO samples without UV treatment. In addi-tion, was determined to be 2.7⫻10−5 ⍀ cm2 for Ti/ZnO samples with UV treatment, which is higher than the deter-mined value共4.1⫻10−6 ⍀ cm2兲 for Ti/ZnO samples without UV treatment. This is attributed to the presence of the wider depletion layer at the Ti/ZnO interface 共due to the lower intensity of O–H bonds of Ti/ZnO samples with UV treat-ment than Ti/ZnO samples without UV treattreat-ment, as shown in Fig. 6兲, leading to a decrease in the probability of field
emission. Consequently, we suggest that the tunneling plays an important role in the carrier transport across the barrier at the Ti/ZnO interface.
The measurement temperature dependence ofobtained on the basis of the TE model and CTLM data is shown in Fig. 7. It is worth noting that obtained on the basis of CTLM data for Ti/ZnO samples with or without UV treat-ment is less sensitive to measuretreat-ment temperature than that obtained on the basis of the TE model. Figures7共b兲and7共c兲 show that the contacts exhibit almost constant in the tem-perature range of 300–400 K. In addition, we find that obtained on the basis of the TE model is higher than that obtained on the basis of CTLM data. These results show that the dominant mechanism of current transport is field emis-sion.
The well-known equations for the electron concentration 共n3兲 and mobility 共3兲 in two-layer systems are expressed as38,40 3= n1d112+ n2d222 n1d11+ n2d22 , 共3兲 n3= 共n1d11+ n2d22兲2 共d1+ d2兲共n1d112+ n2d222兲 . 共4兲
In Eqs.共3兲and共4兲, n1and1represent the bulk, conduction-band electrons, and n2and2, the surface-layer electrons. d1 is the bulk thickness and d2 is the surface-layer thickness 共d1+ d2= 600 nm兲. However, the actual thickness of the sur-face layer is unknown. d2 was assumed to be 4 nm in this study. According to the Van der Pauw–Hall measurements, we found that the electron concentration 共n3兲 and mobility 共3兲 of the ZnO films without UV treatment were calculated to be 4.3⫻1017 cm−3and 12 cm2V−1s−1, respectively. Ac-cording to the Van der Pauw–Hall measurements, we found that the electron concentration and mobility of the ZnO films with UV treatment were calculated to be 2.5⫻1016 cm−3 and 48 cm2V−1s−1, respectively. However, the actual n
1 and 1 of the bulk are difficult to obtain in this study. We assume that UV treatment may lead to the removal of the excess surface-layer electrons, meaning that n1 共1兲 can be assumed to be the electron concentration 共mobility兲 of the ZnO films with UV treatment. Then, n2 and 2 can be ob-tained from Eqs.共3兲and共4兲. n2and2were estimated to be 5.3⫻1020 cm−3 and 1.12 cm2V−1s−1, respectively. Based on the XPS result shown in Fig.6, the large portion of O–H bond peak comparing O–Zn peak still remained after UV treatment. However, we found that the surface-layer electron concentration was higher than 1019 cm−3 even for n
1as low as 1015 cm−3. Clearly, the higher surface-layer electron con-centration than 1019 cm−3 could lead to the occurrence of carrier tunneling at the Ti/ZnO interface and an increase in
Egof ZnO determined by PL measurements. We deduce that the enlarged Eg of 3.42 eV and the detected shift in UL
FIG. 6.共Color online兲 O 1s core-level spectra for ZnO samples 共a兲 without and共b兲 with UV treatment.
FIG. 7. 共Color online兲 Variation in with temperature: 关共a兲 Ti/ZnO samples共B= 0.3 eV兲兴 obtained on the basis of the TE model and关Ti/
ZnO samples共b兲 without and 共c兲 with UV treatment兴 obtained on the basis of CTLM data.
013701-4 Lin et al. J. Appl. Phys. 106, 013701共2009兲
spectra following UV treatment共shown in the inset of Fig.1兲
are associated with a pronounced Burstein–Moss shift.41 Ac-cording to the Burstein–Moss shift, the band gap widening for n-type semiconductors with a parabolic band is attributed to the free electron concentration. When the concentration of electrons in the conduction band exceeds the effective dsity of states in the conduction band of ZnO, the Fermi en-ergy lies within the conduction band. The type of semi-conductor is called degenerate n-type semisemi-conductor.42 Ac-cording to the XPS result shown in Fig. 4, we find that EF
before UV treatment is located at approximately 0.1 eV above CBM, suggesting the formation of the downward band bending and the occurrence of degeneration. These results demonstrate that the contact type conversion involves a de-crease in the number of O–H bonds plus lowered tunneling due to UV treatment.
IV. CONCLUSIONS
The effect of UV treatment on theand electronic trans-port at the Ti/ZnO interfaces has been researched in this study. We found that the tunneling plays an important role in the carrier transport across the barrier at the Ti/ZnO inter-face. The experimental results show the sameB of Ti/ZnO
samples without UV treatment as Ti/ZnO samples with UV treatment and the higherof Ti/ZnO samples with UV treat-ment than Ti/ZnO samples without UV treattreat-ment, suggesting the barrier-height independence of . Therefore, we deduce the contribution to current due to tunneling through the bar-rier must be taken into account and is related to the depletion-layer width. The XPS results also show the in-duced decrease in the number of the hydroxides at the sur-face region of ZnO by UV treatment. This provides the evi-dence that a decrease in the number of O–H bonds existed at the ZnO surfaces may result in a decrease in the electron concentration near the surface region, increasing theof the Ti/ZnO samples.
ACKNOWLEDGMENTS
The authors acknowledge the support of the National Science Council of Taiwan 共Contract No. 97-2628-M-018-001-MY3兲 in the form of grants.
1C. Y. Chang, F. C. Tsao, C. J. Pan, G. C. Chi, H. T. Wang, J. J. Chen, F.
Ren, D. P. Norton, and S. J. Pearton,Appl. Phys. Lett.88, 173503共2006兲. 2M. C. Jeong, B. Y. Oh, M. H. Ham, and J. M. Myoung,Appl. Phys. Lett.
88, 202105共2006兲.
3S. H. K. Park, J. I. Lee, C. S. Hwang, and H. Y. Chu,Jpn. J. Appl. Phys., Part 244, L242共2005兲.
4Y. I. Alivov, J. E. V. Nostrand, D. C. Look, M. V. Chukichev, and B. M.
Ataev,Appl. Phys. Lett.83, 2943共2003兲.
5C. Yuen, S. F. Yu, S. P. Lau, and T. P. Chen,Appl. Phys. Lett.86, 241111
共2005兲.
6A. Tsukazaki, M. Kubota, A. Ohtomo, T. Onuma, K. Ohtani, H. Ohno, S.
F. Chichibu, and M. Kawasaki,Jpn. J. Appl. Phys., Part 244, L643共2005兲.
7T. L. Yang, D. H. Zhang, J. Ma, H. L. Ma, and Y. Chen,Thin Solid Films 326, 60共1998兲.
8A. W. Ott and R. P. H. Chang,Mater. Chem. Phys.58, 132共1999兲. 9A. A. Iliadis, R. D. Vispute, T. Venkatesan, and K. A. Jones,Thin Solid
Films420–421, 478共2002兲.
10Y. J. Lin and C. T. Lee,Appl. Phys. Lett.77, 3986共2000兲.
11H. K. Kim, S. H. Han, T. Y. Seong, and W. K. Choi,Appl. Phys. Lett.77,
1647共2000兲.
12T. Akane, K. Sugioka, and K. Midorikawa,J. Vac. Sci. Technol. B18,
1406共2000兲.
13J. M. Lee, K. K. Kim, S. J. Park, and W. K. Choi,Appl. Phys. Lett.78,
3842共2001兲.
14W. R. Liu, W. F. Hsieh, C. H. Hsu, K. S. Liang, and F. S. S. Chien,J. Cryst. Growth297, 294共2006兲.
15X. W. Sun and H. S. Kwok,J. Appl. Phys.86, 408共1999兲.
16J. J. Chen, T. J. Anderson, S. Jang, F. Ren, Y. J. Li, H. S. Kim, B. P. Gila,
D. P. Norton, and S. J. Pearton,J. Electrochem. Soc.153, G462共2006兲. 17M. Brandt, H. Wenckstern, H. Schmidt, A. Rahm, G. Biehne, G. Benndorf,
H. Hochmuth, M. Lorenz, C. Meinecke, T. Butz, and M. Grundmann,J. Appl. Phys.104, 013708共2008兲.
18K. Ip, Y. W. H. Baik, D. P. Norton, S. J. Pearton, and F. Ren,Appl. Phys. Lett.84, 544共2004兲.
19C. L. Tsai, Y. J. Lin, Y. M. Chin, W.-R. Liu, W. F. Hsieh, C.-H. Hsu, and
J. A. Chu,J. Phys. D: Appl. Phys.42, 095108共2009兲.
20Y. J. Lin, S. S. Chang, H. C. Chang, and Y. C. Liu,J. Phys. D: Appl. Phys. 42, 075308共2009兲.
21H. S. Yang, Y. Li, D. P. Norton, K. Ip, S. J. Pearton, S. Jang, and F. Ren, Appl. Phys. Lett.86, 192103共2005兲.
22Ü. Özgür, Y. I. Alivov, C. Liu, A. Teke, M. A. Reshchikov, S. Doğan, V.
Avrutin, S. J. Cho, and H. Morkoç,J. Appl. Phys.98, 041301共2005兲. 23Y. J. Lin, C. L. Tsai, Y. M. Lu, and C. J. Liu,J. Appl. Phys.99, 093501
共2006兲.
24S. Zhao, Z. Ji, C. Wang, and J. Du,Mater. Lett.61, 897共2007兲. 25Z. Ji, S. Zhao, C. Wang, and K. Liu,Mater. Sci. Eng., B117, 63共2005兲. 26Y. J. Lin, F. M. Yang, C. Y. Huang, W. Y. Chou, J. Chang, and Y. C. Lien,
Appl. Phys. Lett.91, 092127共2007兲.
27S. Y. Kim, J. M. Baik, H. K. Yu, and J. L. Lee,J. Appl. Phys.98, 093707
共2005兲.
28S. K. Hong, T. Hanada, H. Makino, Y. Chen, H. J. Ko, A. Tanaka, H.
Sasaki, and S. Sato,Appl. Phys. Lett.78, 3349共2001兲.
29M. Ruckh, D. Schmid, and H. W. Schock,J. Appl. Phys.76, 5945共1994兲. 30S. K. Hong, T. Hanada, H. Makino, H. J. Ko, Y. Chen, T. Yao, A. Tanaka,
H. Sasaki, S. Sato, D. Imai, K. Araki, and M. Shinohara, J. Vac. Sci. Technol. B19, 1429共2001兲.
31C. L. Wu, H. M. Lee, C. T. Kuo, S. Gwo, and C. H. Hsu,Appl. Phys. Lett. 91, 042112共2007兲.
32J. R. Waldrop and R. W. Grant,Appl. Phys. Lett.52, 1794共1988兲. 33B. J. Coppa, C. C. Fulton, P. J. Hartlieb, R. F. Davis, B. J. Rodriguez, B.
J. Shields, and R. J. Nemanich,J. Appl. Phys.95, 5856共2004兲. 34B. J. Coppa, C. C. Fulton, S. M. Kiesel, R. F. Davis, C. Pandarinath, J. E.
Burnette, R. J. Nemanich, and D. J. Smith, J. Appl. Phys.97, 103517 共2005兲.
35Q. L. Gu, C. K. Cheung, C. C. Ling, A. M. C. Ng, A. B. Djurišić, L. W.
Lu, X. D. Chen, S. Fung, C. D. Beling, and H. C. Ong,J. Appl. Phys.103, 093706共2008兲.
36Ç. Kiliç and A. Zunger,Appl. Phys. Lett.81, 73共2002兲.
37P. F. Cai, J. B. You, X. W. Zhang, J. J. Dong, X. L. Yang, Z. G. Yin, and
N. F. Chen,J. Appl. Phys.105, 083713共2009兲.
38D. C. Look, H. L. Mosbacker, Y. M. Strzhemechny, and L. J. Brillson, Superlattices Microstruct.38, 406共2005兲.
39W. Mönch,J. Vac. Sci. Technol. B17, 1867共1999兲.
40D. C. Look and R. J. Molnar,Appl. Phys. Lett.70, 3377共1997兲. 41X. H. Zhou, Q. H. Hu, and Y. Fu,J. Appl. Phys.104, 063703共2008兲. 42D. A. Neamen, Semiconductor Physics and Devices共McGraw-Hill,
Bos-ton, 2003兲.
013701-5 Lin et al. J. Appl. Phys. 106, 013701共2009兲
