Band gap engineering and spatial confinement of optical phonon in ZnO quantum dots

(1)

Band gap engineering and spatial confinement of optical phonon in ZnO quantum dots

Kuo-Feng Lin, Hsin-Ming Cheng, Hsu-Cheng Hsu, and Wen-Feng Hsieh

Citation: Applied Physics Letters 88, 263117 (2006); doi: 10.1063/1.2218775

View online: http://dx.doi.org/10.1063/1.2218775

View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/88/26?ver=pdfcov Published by the AIP Publishing

Articles you may be interested in

Influence of Mn doping on structural, optical, and magnetic properties of Zn 1 x Mn x O nanorods J. Appl. Phys. 108, 044910 (2010); 10.1063/1.3478709

Raman-active Fröhlich optical phonon mode in arsenic implanted ZnO Appl. Phys. Lett. 94, 011913 (2009); 10.1063/1.3067997

Self-assembled ZnO quantum dots with tunable optical properties Appl. Phys. Lett. 89, 023122 (2006); 10.1063/1.2221892

Size dependence of photoluminescence and resonant Raman scattering from ZnO quantum dots Appl. Phys. Lett. 88, 261909 (2006); 10.1063/1.2217925

Origin of the optical phonon frequency shifts in ZnO quantum dots Appl. Phys. Lett. 86, 053103 (2005); 10.1063/1.1861509

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP: 140.113.38.11 On: Thu, 01 May 2014 01:57:16

(2)

Band gap engineering and spatial confinement of optical phonon in ZnO

quantum dots

Kuo-Feng Lin, Hsin-Ming Cheng,a兲Hsu-Cheng Hsu, and Wen-Feng Hsiehb兲

Department of Photonics, Institute of Electro-Optical Engineering, National Chiao Tung University, 1001 Tahsueh Road, Hsinchu, Taiwan 30050, Republic of China

共Received 9 March 2006; accepted 27 May 2006; published online 30 June 2006兲

Both band gap engineering and spatial confinement of optical phonon were observed depending upon the size of ZnO quantum dots at room temperature. Size-dependent blueshifts of photoluminescence and absorption spectra reveal the quantum confinement effect. The measured Raman spectral shift and asymmetry for the E2共high兲 mode caused by localization of optical phonons agree well with that calculated by using the modified spatial correlation model. © 2006

American Institute of Physics. 关DOI:10.1063/1.2218775兴

Recent remarkable crystal growth techniques for the fab-rication of nanostructured optical devices operating in the ultraviolet共UV兲, using zinc oxide 共ZnO兲 as the constituent material,1–4have prompted studies into the properties of this promising material in its powder and nanocrystalline forms. In nanocrystals, the quantum confinement effect becomes a predominant investigation field and gives rise to many inter-esting electronic and optical properties.5–11The phenomena concerning the quantum confinement effect in ZnO nano-crystals were rarely studied due to the relatively large elec-tron effective mass caused by the large band gap共3.37 eV兲 of this material. Additionally, the Coulomb interaction of electron and hole has reduced the exciton confinement en-ergy partly due to the small dielectric constant ␧_⬁bulk in this system, which is also related to large band gap. Thus, the ZnO binary semiconductor material will have a small quan-tum confinement effect in quanquan-tum dot 共QD兲 and quantum well共QW兲 structures.

ZnO has a wurtzite crystal structure that belongs to the space group C₆4_vand group theory predicts zone-center opti-cal phonon modes, which are A1, 2B1, E1, and 2E2. The A1 and E1modes and the two E2modes are Raman active while the B modes are silent. The nonpolar E₂phonon modes have two frequencies: E2共high兲 is associated with the vibration of oxygen atoms and E2共low兲 is associated with the Zn sublat-tice. All described phonon modes have been reported in the Raman spectra12,13of bulk ZnO. The Raman spectra always show a shift of phonon frequencies in ZnO nanostructures.14–17Whether the origin of this shift is due to strain, intrinsic defects or the size of QDs is still the subject of debates. Nevertheless, by examining ZnO nanocrystals with average sizes of 8.5 and 4.0 nm, Rajalakshmi et al.17 explained the shift of phonon frequency as due to optical phonon confinement in ZnO nanostructures, without consid-ering the effects of crystallite size distribution共CSD兲 on the Raman spectra in ZnO nanostructures.

In this letter, we present a quantum confinement effect revealing a blueshift in absorption and PL spectra with a small defect emission for ZnO QDs having the average size

varying from 12 to 3.5 nm in diameter. Furthermore, we ob-served spatial confinement of the optical phonon induced spectral shift, broadening, and asymmetry of the E2共high兲 phonon mode by using typical Raman spectroscopy. We also used the modified spatial correlation 共SC兲 model, which takes the CSD into consideration, that gave a good fit to the measured Raman spectra.

The synthesis of ZnO QDs was carried out using the sol-gel method, which is similar to those published elsewhere.18,19Stoichiometric zinc acetate dihydrate关99.5% Zn共OAc兲2·2H2O, Riedel-deHaen兴 was first dissolved into di-ethylene glycol 关99.5% DEG, ethylenediamine-tetra-acetic acid共EDTA兲兴. The resultant solution was put in a centrifuge operating at 3000 rpm for 30 min and a transparent solution was then obtained containing dispersed single crystalline ZnO QDs. Finally, the supernatant was dropped on a Si共001兲 substrate with native oxide and dried at 150 ° C. The average size of ZnO QDs ranging from 3.5 to 12 nm can be tailored under a well-controlled concentration 共0.04–0.32M兲 of the precursor. The average crystallite size was determined as in the previous report18 using a Bede D1 x-ray diffractometer with grazing incidence and a JEOL JEM-2100F field emis-sion transmisemis-sion electron microscope共FETEM兲 operated at 200 keV. Micro-Raman spectroscopy was carried out by a frequency-doubled Yb:YAG 共yttrium aluminium garnet兲 la-ser共␭=515 nm兲 as a pump source and detected by a Jobin-Yvon T64000 microspectrometer with a 1800 grooves/ mm grating in the backscattering configuration.

Figure 1 shows typical photoluminescence共PL兲 and ab-sorption spectra of the samples with different average QD sizes at room temperature. The UV emission represents a relaxed state of the exciton near the band edge in the ZnO QDs. The nature of the UV-PL from ZnO QDs itself is still a matter of controversy. Some authors attributed the UV-PL to the recombination of confined excitons,20 while others ar-gued that the emission comes from surface impurities or defects.21 In our case, high efficient UV emission near the band edge is attributed to confined exciton emission, similar to that described in Ref. 20, with high density of states that shifts to the higher energies from 3.30 to 3.43 eV as the size of QDs decreases from 12 to 3.5 nm, which are comparable or smaller than the diameter共4.68 nm兲 of the exciton 共Bohr radius of bulk ZnO is 2.34 nm兲.6

Additionally, the slight wid-ening in the full width at half maximum 共FWHM兲 of UV

a兲_{Also at: Material and Chemical Research Laboratories, Industrial}

Technol-ogy Research Institute, Hsinchu, Taiwan 310, Republic of China.

b兲_{Author to whom correspondence should be addressed; electronic mail:}

[email protected]

APPLIED PHYSICS LETTERS 88, 263117共2006兲

(3)

emission with decrease in the size of ZnO QDs may be caused by the recombination of surface-bound acceptor ex-citon complexes as evident from the lower energy shoulder.22 In general, quantum confinement will widen the energy band gap and give rise to a blueshift in the transition energy as the crystal size decreases. Such a phenomenon is also revealed in the absorption spectra, although the faint excitonic absorp-tion peaks are due to the moderate size distribuabsorp-tion of ZnO QDs. From this figure it can clearly be seen that the absorp-tion onset exhibits a progressive blueshift from 3.43 to 3.65 eV as the size of ZnO QD decreases from 12 to 3.5 nm. We calculated the band gaps by using the effective-mass model18 for different sizes of ZnO QDs and the results showed good agreement with the experimental data.

In order to observe the optical phonon confinement ef-fect, the measured micro-Raman spectra with different sizes of ZnO QDs are shown in Fig. 2 under a fixed excitation laser power of 3.1 mW. We can see that the Raman peak at 300 cm−1_{, which comes from the Si substrate,}23

stays un-shifted in frequency. On the other hand, we found the spec-tral peak of E2共high兲 optical phonon around 435 cm−1 to shift to the lower frequency as the size of ZnO QDs de-creases. Compared with that of the ZnO bulk, a redshift

ranging from 0.8 to 4.7 cm−1 and an asymmetry 共⌫a/⌫b兲

from 1.36 to 1.95 were obtained as the QD size decreases from 12 to 3.5 nm. Note that⌫aand⌫bare, respectively, the

half widths on the low- and high-energy sides of the E2共high兲 mode. Such a pronouncing shift, broadening, and the asym-metry of the E₂共high兲 peak could result from three main mechanisms:15 共1兲 phonon localization by intrinsic defects, 共2兲 laser heating in nanostructure ensembles, and 共3兲 the spa-tial confinement within the dot boundaries. The frequency shift of the phonon resulting from defects should not depend upon the size of quantum dots as also indicated by the small defect PL emission observed in our samples; therefore, we may exclude the defect phonon localization. Additionally, no shift to the E2共high兲 peak was observed in all ZnO QDs as the laser power has been varied almost an order of magnitude from 1.5 to 12 mW with a fixed laser spot size of about 2␮m2. It is therefore concluded that the Raman shift is mainly due to the spatial confinement of the optical phonon. The phonon eigenstates are plane waves with infinite correlation lengths in an ideal crystal, therefore, the Raman scattering can only be observed with phonons around the Brillouin zone center共q=0兲 due to the momentum conserva-tion law. As the crystallite is reduced to nanoscale sizes, the momentum conservation law associated with the Raman scattering can be relaxed and that leads to the spectral shift, broadening, and asymmetry of the Raman modes. The Ra-man shift and broadening of ZnMnO nanoparticles24 had been evaluated based on the SC model.25 Because the pho-non wave function is partially confined to the volume of the crystallite and if a spherical shape of finite size ZnO QDs is assumed, the first-order Raman spectrum I共␻兲 can be de-scribed by the following equation:25

I共␻兲 ⬀

冕

0

1₄_␲_q2_exp_{共− q}2_L2_/4_兲dq

关␻−␻共q兲兴2₊_共⌫/2兲2 , 共1兲 where q is expressed in units of 2␲/ a, a is the lattice con-stant, ␻共q兲 is the phonon dispersion relation, ⌫ is the line-width of E2共high兲 phonon of the bulk ZnO, and L is spatial correlation length corresponding to grain size. Furthermore, Islam et al.26,27reported that CSD influenced both the shifts in Raman scattering frequencies and line shapes in silicon nanostructures. They modified the Raman intensity expres-sion, I共␻兲 of Eq. 共1兲, to I共␻, L0,␴兲 by using a Guassian CSD of an ensemble of spherical crystallites with mean crystallite size L0and standard deviation␴. After integrating the results over the crystallite sizes L under the condition L0⬎3␴, the total Raman intensity expression for the whole ensemble of nanocrystallites becomes I共␻兲 ⬀

冕

0 1_f_共q兲q2_exp_{共− q}2_L 0 2_/4_兲dq 关␻−␻共q兲兴2+共⌫/2兲2 , 共2兲 where f共q兲=1/

冑

1 + q2␴2/ 2 is the characteristics of the CSD. The calculated normalization Raman profiles from an en-semble of ZnO QDs having a mean crystallite size L0 = 6.5 nm with varying ␴ to illustrate the effect of ␴ on the Raman line shape are plotted in Fig. 3. It is clear that a single crystalline component with ␴= 0.27 describes the Raman spectra of 6.5 nm ZnO QDs quite well. Additionally, the CSD of all samples were about 27%, which agrees with the TEM result,18e.g., the obtained crystal size of 4.3± 1.1 nm. The frequency shift ⌬␻ and the asymmetry, ⌫a/⌫b, of

FIG. 1. PL共solid line兲 and absorption 共dashed line兲 spectra near the band edge of various ZnO QD sizes.

FIG. 2. Typical Raman spectra of different sizes of ZnO QDs:共a兲 12 nm, 共b兲 6.5 nm,共c兲 5.3 nm, and 共d兲 3.5 nm.

263117-2 Lin et al. Appl. Phys. Lett. 88, 263117共2006兲

(4)

E2共high兲 mode from the ZnO bulk 共439 cm−1兲 as a function of diameter or correlation length with␴= 0.27 were plotted in Fig. 4, in which the solid curves indicate the calculated results of the modified SC model and the hollow circles the experimental results. We found that the measured frequency shift and asymmetry agrees very well with the ones calcu-lated by the modified SC model and the mean values of crystallite sizes obtained from our fitting are also in good agreement with the XRD results.

In summary, size dependence of efficient UV photolumi-nescence and absorption spectra of various ZnO QD sizes give evidence for the quantum confinement effect. We have observed the spectral shift, broadening, and asymmetry of the optical phonons for different sizes of ZnO QDs and clari-fied that the origin of these effects is spatial confinement of phonon in ZnO QDs. Using the modified spatial correlation

model to analyze the broadening and asymmetry of the first-order E₂共high兲 phonon mode, we further confirmed the pho-non confinement based on the finite correlation length of a propagating phonon.

The authors would like to gratefully acknowledge partial financial support from the National Science Council 共NSC兲 of Taiwan under Contract No. NSC-94-2112-M-009-021. One of the authors共H.C.H.兲 acknowledges NSC of Taiwan for providing a fellowship. The authors also thank Lai of the TEM group of Material Research Laboratory in ITRI for the help on electron microscopy measurements.

1_{L. B. Kong, F. Li, L. Y. Zhang, and X. Yao, J. Mater. Sci. Lett. 17, 769}

共1998兲.

2_{R. L. Hoffman, B. J. Norris, and J. F. Wager, Appl. Phys. Lett. 82, 733}

共2003兲.

3_{R. Könenkamp, R. C. Word, and C. Schlegel, Appl. Phys. Lett. 85, 6004}

共2004兲.

4_{J. B. Baxter and E. S. Aydil, Appl. Phys. Lett. 86, 053114}_共2005兲. 5_{T. Kawazoe, K. Kobayashi, and M. Ohtsu, Appl. Phys. Lett. 86, 103102}

共2005兲.

6_{R. T. Senger and K. K. Bajaj, Phys. Rev. B 68, 045313}_共2003兲. 7_{T. A. Klar, T. Franzl, A. L. Rogach, and J. Feldmann, Adv. Mater.}

共Wein-heim, Ger.兲 17, 769 共2005兲.

8_{Y. Gu, Igor L. Kuskovsky, M. Yin, S. O’Brien, and G. F. Neumark, Appl.}

Phys. Lett. 85, 3833共2004兲.

9_{R. P. Wang, G. Xu, and P. Jin, Phys. Rev. B 69, 113303}_共2004兲. 10_{V. V. Ursaki, I. M. Tiginyanu, V. V. Zalamai, E. V. Rusu, G. A.}

Emelchenko, V. M. Masalov, and E. N. Samarov, Phys. Rev. B 70, 155204共2004兲.

11_{H. M. Cheng, K. F. Lin, H. C. Hsu, C. J. Lin, L. J. Lin, and W. F. Hsieh,}

J. Phys. Chem. B 109, 18385共2005兲.

12_{J. F. Scott, Phys. Rev. B 2, 1209}_共1970兲.

13_{N. Ashkenov, B. N. Mbenkum, C. Bundesmann, V. Riede, M. Lorenz, D.}

Spemann, E. M. Kaidashev, A. Kasic, M. Schubert, M. Grundmann, G. Wagner, H. Neumann, V. Darakchieva, H. Arwin, and B. Monemar, J. Appl. Phys. 93, 126共2003兲.

14_{L. Bergman, X. B. Chen, J. L. Morrison, and J. Huso, J. Appl. Phys. 96,}

675共2004兲.

15_{K. A. Alim, V. A. Fonoberov, and A. A. Balandin, Appl. Phys. Lett. 86,}

053013共2005兲.

16_{K. A. Alim, V. A. Fonoberov, M. Shamsa, and A. A. Balandin, J. Appl.}

Phys. 97, 124313共2005兲.

17_{M. Rajalakshmi, A. K. Arora, B. S. Bendre, and S. Mahamuni, J. Appl.}

Phys. 87, 2445共2000兲.

18_{K. F. Lin, H. M. Cheng, H. C. Hsu, L. J. Lin, and W. F. Hsieh, Chem.}

Phys. Lett. 409, 208共2005兲.

19_{H. M. Cheng, H. C. Hsu, S. L. Chen, W. T. Wu, C. C. Kao, L. J. Lin, and}

W. F. Hsieh, J. Cryst. Growth 277, 192共2005兲.

20_{D. W. Bahnemann, C. Kormann, and M. R. Hoffmann, J. Phys. Chem. 91,}

3789共1987兲.

21_{L. Guo, S. Yang, C. Yang, P. Yu, J. Wang, W. Ge, and G. K. L. Wong,}

Appl. Phys. Lett. 76, 2901共2000兲.

22_{V. A. Fonoberov and A. A. Balandin, Appl. Phys. Lett. 85, 5971}_共2004兲. 23_{R. P. Wang, G. W. Zhou, Y. L. Liu, S. H. Pan, H. Z. Zhang, and D. P. Yu,}

Phys. Rev. B 61, 16 827共2000兲.

24_{J. B. Wang, H. M. Zhong, Z. F. Li, and Wei Lu, J. Appl. Phys. 97, 086105}

共2005兲.

25_{H. Richter, Z. P. Wang, and L. Ley, Solid State Commun. 39, 625}_共1981兲;

I. H. Campbell and P. M. Fauchet, ibid. 58, 739共1986兲.

26_{Md. N. Islam and S. Kumar, Appl. Phys. Lett. 78, 715}_共2001兲. 27_{Md. N. Islam, A. Pradhan, and S. Kumar, J. Appl. Phys. 98, 024309}

共2005兲. FIG. 3. Fitting of the modified spatial correlation model with␴= 0, 0.2, and

0.27, respectively, to the measured result for average size of 6.5 nm ZnO QDs.

FIG. 4. Raman shift⌬␻ and asymmetric broadening ⌫a/⌫b of E2共high兲

phonon as a function of correlation length L or average size of nanocrystal. The solid curves indicate the calculated results of the modified SC model and the hollow circles the experimental results.

263117-3 Lin et al. Appl. Phys. Lett. 88, 263117共2006兲