Photoluminescence Study on O-plasma Treated ZnO Thin Films
Photoluminescence Study on O-plasma Treated ZnO Thin Films
Journal of the Optical Society of Korea. 2013. Dec, 17(6): 543-547
Copyright © 2013, Journal of the Optical Society of Korea
  • Received : August 20, 2013
  • Accepted : October 16, 2013
  • Published : December 25, 2013
Export by style
Cited by
About the Authors
Jaewon, Cho
Department of Electrophysics, Kwangwoon University, Seoul 139-701, Korea
Jinsung, Choi
Department of Electrophysics, Kwangwoon University, Seoul 139-701, Korea
SeGi, Yu
Department of Physics, Hankuk University of Foreign Studies, Yongin 449-791, Korea
Seuk Joo, Rhee
Department of Physics, Hankuk University of Foreign Studies, Yongin 449-791, Korea

A temperature dependent (10K-290K) photoluminescence (PL) study for two differently prepared ZnO thin films (as-grown and O-plasma treated) is presented. Four characteristic peaks were identified for both samples: (i) neutral donor-bound excitons (D o X), (ii) two electron satellites (TES), (iii) phonon replica of D o X (D o X-1LO), and (iv) donor-acceptor pair transition (DAP). As the sample temperature increased, D o X-1LO and DAP transitions became indistinct. This was accompanied by newly-rising emission of free electron-acceptor transitions (e,A o ). The spectral evolution with temperature for as-grown samples also showed the optical emission from free excitons, which became dominant at higher temperatures. Some features related to O-plasma were identified in PL spectra: (i) different positions of TES transitions (28meV lower than D o X for as-grown samples and 35meV for O-plasma treated samples), (ii) the decrease of spectral intensity in both emissions of D o X and DAP after O-plasma treatment, and (iii) no noticeable transition from free excitons after the O-plasma treatment.
ZnO is a semiconductor with direct band structure (band gap energy of 3.37 eV at room temperature), which is similar in physical characteristics to GaN. Carrier doping is the first barrier to be overcome for the application of ZnO as electronic and photonic devices. While n-type doping can be easily realized to the level up to 10 21 cm -3 [1] , reports on successful p-type doping are rare but gradually increasing [2 , 3] . Both bipolar (n-and p-type) doping and wide band gap could maximize the possibility of developing ZnO as transparent electronic devices and light emitting devices from visible to uv regions [4 - 7] . In addition, the high exciton binding energy of ZnO (~60 meV, cf. ~25 meV for GaN) allows near-band-edge excitonic emission at room and higher temperatures [5 , 6 , 8 , 9] . Recently, the technology enabling us to exploit ZnO is developing fast, however, the fundamental understanding for optical characteristics is still under debate. Besides such a physical understanding, various physical and chemical treatment processes should be linked in order to realize promising devices of ZnO.
Photoluminescence (PL) investigation is a suitable tool to determine the crystal quality and the presence of impurities in the material. In addition, the study at low temperatures gives more detailed comprehension of the physical properties of the sample. In this article, we studied the effects of O-plasma treatment on the physical characteristics of ZnO thin films with PL spectroscopy, varying sample temperatures from 10 K to 290 K.
The sample used in this experiment was ZnO thin films of thickness ~350 nm grown on c-plane sapphire at 650℃ by rf-magnetron sputtering in an Ar+O 2 ambient atmosphere. After chemical cleaning, the sample was showered at 300℃ in O-plasma generated by an RF plasma system. The experimental conditions for the plasma treatment are shown in the Table 1 . PL measurement was performed as the sample was kept in the closed cycle cryostat. The sample temperature was varied from 10 K to 290 K. The excitation source was a CW He-Cd laser (λ=325 nm).
Experimental conditions for the O-plasma treatment
PPT Slide
Lager Image
Experimental conditions for the O-plasma treatment
The experimental results of this work are based on the analysis of PL spectra from ZnO thin films. The spectra were taken at the varying temperatures of the sample from 10 K to 290 K from two differently prepared samples, one of which is “as grown” and the other is “O-plasma treated”. There was spectral similarity between the two samples, as shown in Fig. 1 , where four characteristic peaks (Peak I, II, III and IV) were identified for each sample. Now the physical origins for each peak and the resulting effects of O-plasma treatment on ZnO films will be examined.
PPT Slide
Lager Image
The comparison of PL spectra from two differently prepared ZnO samples (as grown and O-plasma treated). The spectra shown in the figure were taken at the sample temperature of 20 K. The PL data are shown on a log intensity scale but are displaced vertically for clarity.
In the undoped bulk ZnO, the neutral shallow donor bound exciton often dominates because of the presence of donors resulting from unintentional impurities and/or shallow donor-like defects [10] . Acceptors like Na or Li could be present in ZnO, too [11] . However, the chemical origin and binding energy of the most underlying donor and acceptor atoms remain unclear. It is known that the low-temperature PL spectra are dominated by several bound excitons in the narrow energy range from 3.348 to 3.374 eV [12] . The higher energy part of the range is mainly due to the recombination of neutral donor-bound excitons (D 0 X), while the lower part comes from neutral acceptor bound excitons (A 0 X) [12] . Peak I in Fig. 1 , appearing at 3.360 eV for both samples (as grown and O-plasma treated), is believed to be mainly the recombination of D 0 X, which is in good agreement with the results reported by other groups [12 - 15] .
When D 0 X decays, all the energy of the exciton is converted into the energy of a photon. However, the decay process could involve TES (Two Electron Satellites) transition, where the donor atom is excited to a higher energy state. When this happens, the outcoming photon will have less energy by an amount equal to the energy difference between the first excited and ground states of the donor. For ZnO, the difference is known to be around 30 meV, which varies with the donor atoms that localize D 0 X [12 , 13 , 16] . The TES transition could be identified in our study. Peak II of our spectra depicted in Fig. 1 is considered to be related to the TES transition. Unlike Peak I (D 0 X), however, Peak II appears at different positions according to the sample preparation. The positions are as follows: 28 meV lower than Peak I for as-grown samples and 35 meV for O-plasma treated samples. What this observation implies will be discussed later.
Peak III, appearing at 3.289 eV at 20 K, whose energy is lower than Peak I by about 71 meV, is believed to be due to a phonon replica of Peak I, as the longitudinal optical (LO) phonon energy of ZnO is known to be 71-73 meV [17] .
When there exists both donor and acceptor impurities together, the recombination of an electron and a hole can take place via donor-acceptor pair (DAP) transitions. The DAP transition, which is observed at low temperatures, happens through a quantum tunneling process when impurities are further away from each other than effective Bohr radius. The photon energy for the DAP transition is as follows; hν = E g - E D - E A + q2 /ϵr [18] . The experimental evidence for the DAP transition is found in many other works, and the outgoing photon energy related to the DAP transition is known to be near 3.22 eV for ZnO [12 , 16 , 19 , 20] . The DAP transition could be identified in our experiment, too. Peak IV, appearing at 3.217 eV for both samples, is considered to be the emission from the DAP transitions.
Figure 2 and Fig. 3 show the evolution of PL spectra with sample temperatures for O-plasma treated and as-grown samples, respectively. Common features can be found in the both figures. The Peak III and IV, which are distinct in spectra at low temperatures of 10 K-30 K, become obscured at higher temperatures. This, instead, accompanies the generation of a new peak positioned at 3.23-3.24 eV which is designated by an arrow in the figures. The new peak was distinguishable up to 200 K for as-grown samples and 180 K for O-plasma treated samples. This experimental observation indicates the thermal ionization of neutral donors. Donors in ZnO get thermally ionized more and more as the temperature gets higher. Free electrons from the neutral donors are expected to recombine with the neutral acceptors, which is denoted by the free electron-acceptor transition (e, A 0 ). This (e, A 0 ) peak will appear at higher energies than the DAP peak, and similar results were reported in other works [12 , 16 , 19] . Hence, the newly-rising peak at 3.23-3.24 eV is believed to be the emission from the (e, A 0 ) transition.
PPT Slide
Lager Image
These PL spectra were selected at several representative temperatures (20 K, 30 K, 70 K, 150 K and 200 K) for the O-plasma treated ZnO film. At 70 K, a new peak (e, A0) around 3.24 eV (depicted byarrows) arises while both D0X-1LO and DAP peaks vanish.
PPT Slide
Lager Image
The PL spectra recorded at different sample temperatures (20 K, 30 K, 70 K, 150 K, 200 K, and 250 K) for the as-grown ZnO film. The plots are vertically shifted, each for a better visibility.The intensities of D0X-1LO and DAP peaks decrease and a free electron-acceptor transition peak (e, A0) arises around 3.24 eV at 70 K.
Figure 4 shows spectral evolution with temperatures for the as-grown sample, where a new peak, designated by arrows, could be identified at the right shoulder of Peak I (D 0 X). The peak gets more distinct as temperature increases. The position of the peak is 3.369 eV at 90 K, which is higher by 14.1 meV than the D 0 X peak. We believe that this peak comes from the decay of the free exciton, considering the relative position of the peak to D 0 X along with characteristic features of temperature evolution. The localization energy of bound excitons in ZnO is known to be 10-20 meV [12 , 14 , 15 , 20] , which is in a good agreement with our result. This indicates that bound excitons change into free excitons as the temperature increases. The free exciton peak gets larger with the increase of the temperature, and starts to surpass the D 0 X peak at 120 K, as shown in the figure. This tendency continues, and free excitons are dominant from 150 K.
PPT Slide
Lager Image
The changes in the PL spectra with increasing temperature for the as-grown ZnO film.Transitions due to free excitons are depicted by arrows.
Let us consider the effects of O-plasma treatment on the properties of ZnO film. The plasma treatment process can induce both physical and chemical effects on the sample. The atoms and/or ions in the plasma will penetrate into the sample, damaging the crystal and making chemical bonds with the atoms in the sample. As a result, the plasma process induces crystal defects as well as impurities in the sample. Oxygen atoms and ions in the O-plasma are expected to make chemical bonds with donor-like impurities and defects which are known to exist in an undoped ZnO. They can also fill O-vacancies which are known to play an important role as donors in ZnO along with interstitial Zn atoms. As a result, the density of the donor-like defects decreases in the plasma process, which results in decrease of the intensity of D 0 X and DAP transitions. This was identified in our experimental result. Comparing the intensities of the D 0 X peak and the DAP peak for both samples at 10 K, we could find that the intensity decreased by about 40% for the D 0 X peak and 60% for the DAP peak after the plasma treatment. Oxygen atoms and defects, which are generated in the sample during the plasma process, can become new centers that localize excitons. Some of those centers will act as neutral donors to capture excitons, which results in another D 0 X. The difference in the position of Peak II between both samples, as stated above in Fig. 1 , can be understood to be due to newly generated neutral donors after the O-plasma treatment.
Another feature, to be noted, is that the remarkable recombination of free excitons was not identified for the O-plasma treated sample. It is difficult to determine the physical origin of the phenomena based solely on our data, but it is suggestive that the higher defect and impurity concentrations induced in the sample by O-plasma can make the formation of free excitons difficult and the lifetime of the free exciton shorter. Hence, the peak from free excitons seems to be shaded after the plasma treatment.
The PL intensity concerning D o X and DAP decreased after the O-plasma treatment by 40% and 60% respectively at 10 K. These two emissions are closely related to neutral donors in ZnO, which indicates that the O-plasma process induces a remarkable decrease of neutral donors. Considering that the O-vacancies and interstitial Zn atoms play an important role as donors in ZnO, the result implies that oxygen atoms and ions in the O-plasma penetrate into the film, filling O-vacancies and making chemical bonds with interstitial Zn atoms. And crystal defects and impurities caused during the O-plasma process are expected to become new centers that localize excitons, which is responsible for the different position of TES emissions (28 meV lower than D o X for as-grown samples and 35 meV for O-plasma treated samples). In our study, no significant trace of free excitons has been observed in the PL data after the O-plasma treatment. This could be attributed to higher defect and impurity concentrations generated in the sample by the O-plasma treatment.
The present research had been conducted by the Research Grant of Kwangwoon University in 2012.
Ellmer K. 2000 “Magnetron sputtering of transparent conductive zinc oxide: relation between the sputtering parameters and the electronic properties,” J. Phys. D: Appl. Phys. 33 R17 - R32    DOI : 10.1088/0022-3727/33/4/201
Look D.C. , Clafin B. , Alivor Y.I. , Park S.J. 2004 “The future of ZnO light emitters,” Phys. Stat. Sol. A 201 2203 - 2212    DOI : 10.1002/pssa.200404803
Ryu Y. R. , Zhu S. , Look D. C. , Wrobel J. M. , Jeong H. M. , White H. W. 2000 “Synthesis of p-type ZnO films,” J. Cryst. Growth 216 330 - 334    DOI : 10.1016/S0022-0248(00)00437-1
Chen Y. F. , Bagnall D. M. , Koh H. , Park K. , Hiraga K. , Zhu Z. , Yao T. 1998 “Plasma assisted molecular beam epitaxy of ZnO on c-plane sapphire: Growth and characterization,” J. Appl. Phys. 84 3912 - 3918    DOI : 10.1063/1.368595
Liang W. Y. , Yoffe A. D. 1968 “Transmission spectra of ZnO single crystals,” Phys. Rev. Lett. 20 59 - 62    DOI : 10.1103/PhysRevLett.20.59
Reynolds D. C. , Look D. C. , Jogai B. , Litton C. W. , Cantwell G. , Harsch W. C. 1999 “Valence-band ordering in ZnO,” Phys. Rev. B 60 2340 - 2344    DOI : 10.1103/PhysRevB.60.2340
Yu D. P. , Bai Z. G. , Ding Y. , Hang Q. L. , Zhang H. Z. , Wang J. J. , Zou Y. H. , Qian W. , Xiong G. C. , Zhou H. T. , Feng S. Q. 1998 “Nanoscale silicon wires synthesized using simple physical evaporation,” Appl. Phys. Lett. 72 3458 - 3460    DOI : 10.1063/1.121665
Zu P. , Tang Z. K. , Wong G. K. L. , Kawasaki M. , Ohtomo A. , Koinuma H. , Segawa Y. 1997 “Ultraviolet spontaneous and stimulated emissions from ZnO microcrystallite thin films at room temperature,” Solid State Commun. 103 459 - 463    DOI : 10.1016/S0038-1098(97)00216-0
Bagnall D. M. , Chen Y. F. , Shen M. Y. , Zhu Z. , Goto T. , Yao T. 1998 “Room temperature excitonic stimulated emission from zinc oxide epilayers grown by plasma-assisted MBE,” J. Cryst. Growth 184/185 605 - 609    DOI : 10.1016/S0022-0248(98)80127-9
Özgür Ü. , Alicov Ya. I. , Liu C. , Teke A. , Reshchikov M. A. , Dogan S. , Avrutin V. , Cho S. J. , Morkoç H. 2005 “A comprehensive review of ZnO materials and devices,” J. Appl. Phys. 98 041301-1 - 041301-3    DOI : 10.1063/1.1992666
Kobayashi A. , Sankey O. F. , Dow J. D. 1983 “Deep energy levels of defects in the wurtzite semiconductors AlN, CdS, CdSe, ZnS, and ZnO,” Phys. Rev. B 28 946 - 956    DOI : 10.1103/PhysRevB.28.946
Teke A. , Özgür Ü. , Dogan S. , Gu X. , Morkoc H. , Nemeth B. , Nause J. , Everitt H. O. 2004 “Excitonic fine structure and recombination dynamics in single-crystalline ZnO,” Phys. Rev. B 70 195207-1 - 195207-10    DOI : 10.1103/PhysRevB.70.195207
Alves H. , Pfisterer D. , Zeuner A. , Riemann T. , Christen J. , Hofmann D. M. , Meyor B. K. 2003 “Optical investigations on excitons bound to impurities and dislocations in ZnO,” Optical Materials 23 33 - 37    DOI : 10.1016/S0925-3467(03)00055-7
Hamby D. W. , Lucca D. A. , Klopfstein M. J. , Cantwell G. 2003 “Temperature dependent excitation photoluminescence of bulk ZnO,” J. Appl. Phys. 93 3214 - 3217    DOI : 10.1063/1.1545157
Wang L. , Giles N. C. 2003 “Temperature dependence of the free-exciton transition energy in zinc oxide by photoluminescence excitation spectroscopy,” J. Appl. Phys 94 973 - 978    DOI : 10.1063/1.1586977
Thonke K. , Gruber Th. , Teofilov N. , Schonfelder R. , Wagg A. , Sauer R. 2001 “Donor-acceptor pair transitions in ZnO substrate material,” Physica B 308-310 945 - 948    DOI : 10.1016/S0921-4526(01)00877-8
Damen T. C. , Porto S. P. , Tell B. 1966 “Raman effect in Zinc oxide,” Phys. Rev. 142 570 - 574    DOI : 10.1103/PhysRev.142.570
Pankove J. I. Optical Processes in Semiconductors Dover Publications, Inc. 143 -
Tamura K. , Makito T. , Tsukazaki A. , sumiya M. , Fuke S. , Furumochi T. , Lippmaa M. , Chia C. H. , Segawa Y. , Koinuma H. , Kawasaki M. 2003 “Donor-acceptor pair luminescence in nitrogen-doped ZnO films grown on lattice-matched ScAlMgO4(0001) substrates,” Solid State Communications 127 265 - 269    DOI : 10.1016/S0038-1098(03)00424-1
Meyer B. K. , Alves H. , Fofmann D. M. , Kriegseis W. , Forster D. , Bertram F. , Christen J. , Hoffmann A. , Strassburg M. , Dworzak M. , Haboeck U. , Rodina A.V. 2004 “Bound exciton and donor-acceptor pair recombinations in ZnO,” Phys. Stat. Sol. (b) 241 231 - 260    DOI : 10.1002/pssb.200301962