# 氧化鋅晶粒大小及氧化鎂鋅之鎂含量對激子-縱模光聲子作用的影響

全文

(2) 氧化鋅晶粒大小及氧化鎂鋅之鎂含量對激子-縱模光 聲子作用的影響. Effect of grain size and Mg content of MgxZn1-xO on the exciton-longitudinal optical phonon interaction Student：Wei-Tse Hsu. 研 究 生：徐瑋澤 指導教授：謝文峰. Advisor：Wen-Feng Hsieh. 教授. 國 立 交 通 大 學 光電工程研究所 碩 士 論 文 A Thesis Submitted to Institute of Electro-optical Engineering College of Electrical Engineering and Computer Science National Chiao Tung University in partial Fulfillment of the Requirements for the Degree of Master in Electro-optical Engineering January 2007 Hsinchu, Taiwan, Republic of China. 中華民國九十七年一月.

(3) 氧化鋅晶粒大小及氧化鎂鋅之鎂含量對激子-縱模光聲子作用的影 響. 研究生：徐瑋澤. 指導教授：謝文峰. 教授. 國立交通大學光電工程研究所. 摘要. 利用量測由溶膠－凝膠製成不同晶粒大小的氧化鋅及不同鎂含量（0 ≤ x ≤ 0.05）氧 化鎂鋅粉末（MgxZn1−xO）的變溫螢光光譜，可以得到激子－縱模光聲子（LO-phonon） 之間耦合的關係。在氧化鋅量子點系統中不明顯激子的縱模光聲子映稱（LO-phonon replicas），以及較小的 13 K 至 300 K 激子能量差顯示激子－縱模光聲子耦合強度的降 低。從氧化鋅量子點系統及氧化鎂鋅粉末激子游離能的變大可得知激子波耳半徑的變小。 這個結果將導致激子結構極性降低，以致於當氧化鋅量子點縮小及氧化鎂鋅中鎂含量增 加，激子－縱模光聲子的交互作用將隨之減弱。. i.

(4) Effect of grain size and Mg content of MgxZn1-xO on the exciton-longitudinal optical phonon interaction. Student: Wei-Tse Hsu. Advisor: Prof. Wen-Fen Hsieh. Institute of Electro-optical Engineering National Chaio Tung University. Abstract. The coupling between exciton and longitudinal optical (LO) phonon was investigated by using of the temperature-dependent photoluminescence from various sizes of ZnO grain and MgxZn1−xO powders in the range of 0 ≤ x ≤ 0.05 prepared by sol-gel synthesis.. The. unapparent LO-phonon replicas of free exciton emission in ZnO quantum dots (QDs) system and the smaller free exciton energy difference between 13 K and 300 K reveal decreasing weighting of exciton−LO phonon coupling strength.. From ZnO QDs system and MgxZn1-xO. powders, the increase of the exciton binding energy resulting from the decrease of the exciton Bohr radius makes the exciton less polar that is responsible for reducing the coupling strength of exciton-LO phonon as shrinking ZnO QDs size and increasing Mg content.. ii.

(5) 誌謝 終於要結束求學的生涯啦！一年半的日子，一溜煙的就過去。 我要感謝我媽還有我的爺爺，這一路來給我經濟、精神上的支持，讓我能一路順利 的走到現在，現在我終於可以不再花他們的錢，準備工作賺錢給他們囉！還有依琳，因 為妳的幫夫運，讓我能夠進入好的研究所，也能順利的畢業。 再來要謝謝我的指導教授謝文峰老師，謝謝您不嫌棄我是從動機系來的學生，什麼 都不懂，還給了我很多的機會，到最後肯定我，讓我提早畢業，您的肯定讓我對於自己 的將來有更足夠的信心去面對所有事。 國峰學長，我的論文幾乎是你完成的，沒有你每次結結巴巴的把東西教給我，我到 現在可能還不知道我在做什麼，沒有你不屑製程的東西，我哪有畢業的東西可以寫。而 且畢業前還麻煩你載我東載我西的，讓我瞬間有變實驗室大爺的感覺，真是太感激您了！ 另外也要感謝金金大學姐，你跟學長一樣幫了我很多忙，讓老師對我有好印象，也讓我 copy 了很多你論文的東西，我能夠提早畢業，妳也有很大很大的功勞。 材料組的學長們，感謝你們過去一年半來在材料實驗上的幫忙，謝謝黃董 Raman 方面的指導，松松教我的低溫 PL，小豪陪我調過無數次的光路，以及小郭、維仁的幫 忙等等，謝謝你們，也祝福你們以後能順順利利！雷射組的智章、博濟學長，也謝謝你 們在實驗室的陪伴，雖然共事機會不多，但還是有請教你們的時候。至賢學長也謝謝你 過去一直關心我的身體狀況。 碩二的夥伴們，我們都要謝謝小賴這個 high 咖讓我們彼此那麼熟，然後小詹的認真 iii.

(6) 態度也讓我們無形中多學了一些東西，陳厚仁感謝你幫我常常幫我弄電腦（小豪也謝謝 你），林易慶要感謝你過去一年來一起分享實驗的心得以及注意的事情，IC 謝謝你幫我 們拍了這麼多好照片（小豪也謝謝你），李博士就不用提了，哈哈！ 碩一的學弟妹們，由於你們坐裡面，而且又剛過半年，所以跟你們還沒那麼熟，但 應該不只我跟你們不熟，所以你們要趕快出來跟學長姊們 social 一下，不然又一眨眼大 家都要畢業光了。 還有已故的碩二學長姊們，我可沒忘了提你們一下，尤其現在在奇美逍遙的那位梁 飛鴻，感謝當時卡丁車的陪伴，哦對還有棒球。其他學姊們也謝謝你們很照顧我這個學 弟。 該謝的都謝完了，沒想到有兩頁這麼多，我們實驗室人太多了，以前的人怎麼一頁 就打得完了。 最後感謝國科會計畫. NSC-96-2628-E-009-018. 對於此研究的支持與贊助. 老徐 iv. 于. 九十七年一月二十九日.

(7) Content Abstract (in Chinese). i. Abstract (in English). ii. Acknowledgement. iii. Content. v. List of figures. viii. Chapter 1 Introduction. 1. 1.1 Basic property of ZnO and significance of ZnO related photonic devices. 1. 1.2 Problem description and Motivation. 1. 1.3 Organization of thesis. 2. Chapter 2 Theoretical background. 3. 2.1 Sol-gel method. 3. 2.2 Quantum confinement effect. 5. 2.2.1 Weak confinement. 5. 2.2.2 Strong confinement. 7. 2.3 Fröhlich interaction. 8. 2.4 Photoluminescence characterization. 12. 2.4.1 General concepts. 12 v.

(8) 2.4.2 Exciton-related emission. 14. 2.4.2.1 Free excitons and polaritons. 14. 2.4.2.2 LO-phonon replicas. 19. 2.4.3 The information obtained from PL spectrum. 20. 2.4.3.1 Binding energy. 21. 2.4.3.2 Band gap energy variation with temperature. 21. Chapter 3 Experiment detail. 23. 3.1 Sample preparation. 23. 3.1.1 ZnO quantum dots synthesized by varying concentration of zinc precursor. 23. 3.1.2 MgxZn1-xO powders synthesized by varying molar ratio of Mg/Zn. 23. 3.2 Photoluminescence system. 23. Chapter 4 Results and discussions. 25. 4.1 ZnO quantum dots system. 25. 4.1.1 Photoluminescence spectrum for different sizes of ZnO particles. 25. 4.1.2 Temperature-dependent PL for different sizes of ZnO particles. 26. 4.1.3 The weighting of exciton-LO phonon coupling strength (α0). 27. 4.1.4 Binding energy. 28. 4.1.5 Bohr radius (aB). 28. 4.1.6 Correlation of α0 and aB. 29 vi.

(9) 4.1.7 Summary. 30. 4.2 MgxZn1-xO alloy system. 30. 4.2.1 Morphology and temperature-dependent photoluminescence. 30. 4.2.2 Binding energy. 32. 4.2.3 The weighting of exciton-LO phonon coupling strength (α0). 33. 4.2.4 Correlation of α0 and aB. 34. 4.2.5 Summary. 36. Chapter 5 Conclusion and perspective. 37. 5.1 Conclusion. 37. 5.2 Perspective. 37. References. 39. vii.

(10) List of Figures Fig. 2-1 Schematic of the rotes that one could follow within the scope of sol-gel processing. 4 Fig. 2-2 A pair excitation in the scheme of valence and conduction band in the exciton picture for a direct gap semiconductor.. 13. Fig. 2-3 The electronic band structure of wurtzite ZnO.. 15. Fig. 2-4 Free excitonic fine structure of the 10 K PL spectrum for the ZnO single crystal.. 16. Fig. 2-5 Dispersion curves of exciton and exciton polaritons.. 18. Fig. 2-6 10 K PL spectrum in the region where LO-Phonon replicas are expected to appear. 20 Fig. 3-1 PL detection system.. 24. Fig. 4-1 PL spectra of different ZnO particle sizes at 13K.. 25. Fig. 4-2 Temperature-dependence PL spectra of various sizes of ZnO particles.. 26. Fig. 4-3 Experimental and calculated exciton energies plotted against inverse temperature for different ZnO particle sizes.. 27. Fig. 4-4 The ܺܨୀଵ integral intensity as a function of the inverse temperature from 13 to 300K for various sizes of ZnO particles.. 28. Fig. 4-5 The relation of α0 QD/α0 Powders and aB QD/aB Powders with different dot sizes.. 29. Fig. 4-6 The SEM image of the 5% Mg sample.. 31. Fig. 4-7 Temperature dependence of PL spectra in Mg0.05Zn0.95O powders.. 32. Fig. 4-8 Normalized integrated intensity of ZnO sample as a function of temperature.. 33. Fig. 4-9 Dependence of PL peak energy positions on temperature for the MgxZn1−xO alloys. 34 Fig. 4-10 The coupling strength of the exciton-LO-phonon given as the average ratio α0 ratio = α0 MgxZn1−xO/α0 ZnO and aB for experiment and correction is also given as aB ratio = aB MgxZn1−xO/ab ZnO.. 35. Figure 5-1 PL spectra of different anneal temperatures for ZnO-MgO at 13K.. 38. viii.

(11) Chapter 1 Introduction 1.1 Basic property of ZnO and significance of ZnO related photonic devices During the last decade, zinc oxide (ZnO) has received much attention because of its wide band gap and large binding energy (Eb ~ 60 meV) [1].. Optical and physical. properties of semiconductor quantum dots (QDs) have also attracted considerable interests due to their potential applications to light-emitting diodes [2], optically pumped lasers [3] and other electronic devices [4].. 1.2 Problem description and Motivation Although large number of researches on II-VI QDs and III-V QDs have been published [5,6], the properties of ZnO QDs have not been studied as completely as other materials. To fully utilize ZnO-based technologies, band gap engineering and p-type doping are still the remaining issues in modern optoelectronics.. Jingbo et al. [7]. proposed that, using the first-principles band-structure calculations, the p-type dopability of ZnO can be improved by lowering the ionization energy of acceptors in ZnO by codoping acceptors with donor or isovalent atoms (e.g., Mg or Be). (Mg,Zn)O alloys have been proved feasible to realize the band gap modulation of ZnO.. Additionally, MgZnO can be used as energy barrier layers of ZnO. light-emitting devices and ZnO/Mg0.2Zn0.8O coaxial nanorod single quantum well structures for carrier confinement [8,9].. They are also very promising materials for. use as active layers in double heterostructures, e.g., Shibata et al. [10] reported Zn1-xMgxO alloys are very brilliant light emitters, even more brilliant than ZnO,. 1.

(12) particularly in the high-temperature region.. Therefore, it requires a full. understanding of material characteristic of MgZnO systems. The interaction between exciton and longitudinal-optical (LO) phonon has a great influence on the optical properties of polar semiconductors.. Ramvall, et al. [11]. reported a diminishing temperature-dependent shift of the photoluminescence (PL) energy with decreasing GaN QD size caused by a reduction of the LO-phonon coupling.. In our previous work [12], the resonant Raman scattering (RRS) of. various ZnO QDs sizes reveals that decrease of I2LO/I1LO with decreasing particle sizes gives an evidence for the reduction of exciton-LO phonon interaction with decreasing QDs size.. Chang, et al. [13] reported that the exciton LO-phonon interaction energy. |Eex-ph| theoretically evaluated as functions of electric field strength and the size of the quantum dots.. The field enhanced by reducing the separation between electron and. hole would increase |Eex-ph|; whereas, the decrease of dot size leads to delocalize the wave functions of both electron and hole in turn decreases |Eex-ph|.. Sun et al., [14]. who grew ZnO/ZnMgO multiquantum wells by laser-molecular-beam epitaxy, investigated λEx-LO by the temperature dependence of the linewidth of the fundamental excitonic peak from optical absorption spectra. However, the size dependence and Mg concentration of MgxZn1-xO powders of exciton-LO-phonon coupling is a complicated problem to be investigated.. 1.3 Organization of thesis In this work, we qualitatively compared the PL spectra of various ZnO particle sizes and Mg-doped ZnO powders with increasing Mg substitution, and quantitatively deduced the weighting of exciton-LO-phonon coupling strength.. We finally. obtained the reduction of exciton-LO phonon interaction with decreasing ZnO particle sizes and raising Mg substitution in MgxZn1-xO powders. 2.

(13) Chapter 2 Theoretical background. 2.1 Sol-gel method An aerosol is a colloidal suspension of particles in a gas (the suspension may be called a fog if the particles are liquid and a smoke if they are solid) and an emulsion is a suspension of liquid droplets in anther liquid. A sol is a colloidal suspension of solid particles in a liquid, in which the dispersed phase is so small (~1-1000 nm) that gravitational force is negligible and interactions are dominated by the short-range forces, such as Van der Waals attraction and surface charge.. All of these types of colloids can be made by sol-gel synthesis and used to. generate polymers or particles from which ceramic materials can be made. Sol-gel synthesis has two ways to prepare solution. One way is the metal-organic route with metal alkoxides in organic solvent; the other way is the inorganic route with metal salts in aqueous solution. It is much cheaper and easier to handle metal salts than metal alkoxides, but their reactions are more difficult to control.. The inorganic route is a step of polymerization reactions through. hydrolysis and condensation of metal alkoxides M(OR)Z, where M = Si, Ti, Zr, Al, Sn, Ce, and OR is an alkoxy group and Z is the valence or the oxidation state of the metal. The chemical reactions occur in the inorganic-route sol-gel synthesis including the following steps: First, hydroxylation upon the hydrolysis of alkoxy groups:. M − OR + H 2 O → M − OH + ROH .. (2.1). The second step, polycondensation process leads to the formation of branched oligomers and polymers with a metal oxygenation based skeleton and reactive residual hydroxyl and alkoxy groups.. There are 2 competitive mechanisms in the second step: 3.

(14) Oxolation: formation of oxygen bridges involving. M − OH + XO − M → M − O − M + XOH ,. (2.2). where the hydrolysis ratio (h = H2O/M) decides X=H (h >> 2) or X = R (h<2). Olation:. formation of hydroxyl bridges when the coordination of the metallic center is. not fully satisfied (N - Z > 0) involving. M − OH + HO − M → M − (OH ) 2 − M , where X = H or R. The kinetics of olation is usually faster than those of oxolation.. (2.3) Figure. 2-1 presents a schematic of the routes that one could follow within the scope of sol-gel processing [15]. The precursors (starting compounds) for preparation of a colloid consist of a metal or metalloid element surrounded by various ligands.. Figure 2-1 Schematic of the routes that one could follow within the scope of sol-gel processing.. 4.

(15) 2.2 Quantum confinement effect [16] Recently, it is possible to achieve the low-dimensional structures of materials. Because of reducing dimensional structures, the quantum confinement effects (QCE) become predominant and give rise to a lot of interesting electronic and optical properties.. The. electron energy will be quantized and varies with dot sizes that cause the variation of band gap energy, binding energy, and Bohr radius. Two regimes of quantization are usually distinguished in which the crystallite radius R is compared with the exciton Bohr radius or related quantities: (1) Weak confinement regime for R ≧ aB; and (2). Strong confinement regime for R ＜ aB.. In the weak confinement regime, the motion of center of exciton mass is quantized while the relative motion of electron and hole given by the envelope function. is hardly. affected. In the strong confinement regime, however, the Coulomb energy increases roughly with R-1, and the quantization energy with R-2, so that for sufficiently small values of R one should reach a situation where the Coulomb term can be neglected. 2.2.1 Weak confinement [17] Coulomb-related correlation between the charged particles handled through the use of a 5.

(16) vary approach involving higher-order wave function of the confined particles, thus, we cannot neglect the electron hole Coulomb potential.. The Schrödinger equation may be written as. Ψ Taking. and. Ψ. Ψ. ,. (2.4). , then the equation becomes R. R. with. E Ψ.. r Ψ. Ψ. (2.5). , and Et as the total energy of the system.. and consider Coulomb interaction first, then we get Ψ ,. R. (2.6) .. Here. If we take. is resulting from the inclusion of Coulomb interaction.. (2.7). Then we consider the. confinement potential 0, ∞, Thus, the eigenenergy is ,. 1, 2, 3, … …. and the energy of an absorbed resonant photon is. 6. (2.8).

(17) ω .. ω where. (2.9). R. is band gap energy.. 2.2.2 Strong confinement The size quantization band states of the electron and hole is dominated by the kinetic energies of electron and hole which are larger than the electron-hole Coulomb potential, therefore, the effect of the Coulomb attraction between the electron and hole can be treated as a perturbation.. The Schrödinger equation can be written as Ψ. Ψ. Ψ. (2.10). Again, the confinement potential is defined as 0, ∞,. , ,. and the energy of electron or hole is. ,. ,. 1, 2, 3, … …. (2.11). The energy of an absorbed resonant photon is ω. (2.12). 7.

(18) with. as the reduced mass of electron and hole.. 2.3 Fröhlich interaction [18] One of the most important carrier-phonon scattering mechanisms in semiconductors occurs when charge carriers interact with the electric polarization, P(r), produced by the relative displacement of positive and negative ions.. In low-defect polar semiconductors such. as GaAs, InP, and GaN, carrier scattering in polar semiconductors at RT is dominated by this polar-optical-phonon (POP) scattering mechanism. The POP-carrier interaction is referred to as the Fröhlich interaction.. Here, the potential energy associated with the Fröhlich. interaction will be denoted by φFr(r).. The polarization P associated with polar-optical. phonons and the potential energy associated with the Fröhlich interaction, φFr(r), are related by ∇ 2φFr (r ) = 4π e∇ ⋅ P (r ) .. (2.13). In terms of the phonon creation and annihilation operators, the operators act on states each having a given number of phonons.. The creation operator acting on a state of nq. phonons of wavevector q increases the number of phonons to nq + 1 and the phonon annihilation operator acting on a state of nq phonons of wavevector q decreases the number of phonons to nq − 1.. P(r) may be written as. P(r ) = ζ. d 3q (aq eiq.r eq , j + aq+ e −iq.r e*q , j ) , ∑ 3 ∫ (2 ) π j =1,2,3. (2.14). where eq,j represents the polarization vector associated with P(r) and q is the phonon wavevector; then, it follows that. 8.

(19) d 3q 4π∇ ⋅ P(r ) = 4πiζ ∑ ∫ (aq eiq.r q ⋅ eq , j − aq+ e −iq.r q ⋅ e*q , j ) . 3 j =1,2,3 (2π ). (2.15). Consider the case of a polar crystal with two atoms per unit cell, such as GaAs. The dominant contribution to P(r) results from the phonon modes in which the normal distance between the planes of positive and negative charge varies. Such modes are obviously the longitudinal optical (LO) modes since in the case of LO modes eq,j is parallel to q. However, transverse optical (TO) phonons produce displacements of the planes of charge such that they remain at fixed distances from each other; that is, the charge planes slide by each other but the normal distance between planes of opposite charge does not change. So, TO modes make negligible contributions to P(r).. For TO phonons, eq,j·q = 0. Accordingly,. 4π∇ ⋅ P (r ) = 4πiζ ∫. d 3q (aq eiq.r q − aq+ e −iq.r q) , 3 (2π ). (2.16). and the potential energy associated with the Fröhlich interaction, φFr(r), is given by H Fr = φFr (r ) = −4πieζ ∫. d 3q 1 (aq eiq.r − aq+ e − iq.r ) , 3 (2π ) q. (2.17). where φFr(r), has been denoted by HFr, the Fröhlich interaction Hamiltonian since φFr(r) is the only term contribution to it. The dependence of φFr(r) on q-1 is familiar from the Coulomb interaction; the coupling constant ζ remains to be determined. In polar semiconductors, the masses m and M carry opposite charges, e*, as a result of the redistribution of the charge associated with polar bonding. The electric polarization P(r) may be written as. 9.

(20) P(r ) =. Ne* u (r ) ε (∞ ) q. Ne* 1 = ε (∞ ) N. ∑∑ q. j =1,2,3. = × (aq eiq.r eq , j − aq+ e − iq.r e*q , j ) , mM 2( )ωLO m+M. (2.18). where the division by ε(∞) accounts for screening, the normal-mode phonon displacement has been used for u(r), q is summed over all wavevectors in the Brillouin zone, and N is the number of unit cells in the sample. By noticing that. 1 V. ∑ q. d 3q ↔∫ (2π )3. (2.19). and by comparing expressions (2.14) and (2.18), it follows that. ζ=. However,. ε (ω ) = ε (∞) +. from. the. Ne* 1 ε (∞) N. dielectric. = . mM 2( )ωLO m+M. constant. due. to. (2.20). the. electronic. response. 4π Ne*2 1 1 ( + ) evaluated for ω = ω0, it follows that 2 − ω 2 ) ε (∞ ) m M (ωTO 2 2 ωLO − ωTO Ne*2 1 1 ( ) + = 4πε (∞) ε (∞ ) 2 m M. =. 2 2 ωLO 1 1 ωTO [ ] − 2 4π ε (∞) ε (∞) ωLO. 2 ωLO 1 1 [ ], = − 4π ε (∞) ε (0). so that. 10. (2.21).

(21) ζ=. Ne* 1 ε (∞ ) N. = = mM 2( )ω LO m+M. =. 2ωLO. 2 ωLO 1 1 [ ], − 4π ε (∞) ε (0). (2.22). and H Fr = φFr (r ) = −4πieζ ∫. d 3q 1 ( aq eiq.r − aq+ e − iq.r ) (2π )3 q. 2π e 2 =ωLO 1 1 1 = −i [ ] ∑ ( aq eiq.r − aq+ e − iq.r ) . − V ε (∞) ε (0) q q. (2.23). Besides, the lattice distortion created by a phonon resulting in a local change in the crystal’s energy band; the energy associated with the change is known as the deformation potential and it represents one of the major scattering mechanisms in non-polar semiconductors.. Indeed, the deformation-potential interaction is a dominant source of. electron energy loss in silicon-based electronic devices. The piezoelectric interaction occurs in all polar crystals lacking inversion symmetry. In the general case, the application of an external strain to a piezoelectric crystal will produce a macroscopic polarization as a result of the displacements of ions.. Thus an acoustic phonon mode will drive a macroscopic. polarization in a piezoelectric crystal.. 11.

(22) 2.4 Photoluminescence characterization [19] 2.4.1 General concepts. The ground state of the electronic system of a perfect semiconductor is a completely filled valance band and a completely empty conduction band. We can define this state as the “zero” energy or “vacuum” state. If we start from the ground state and excite one electron to the conduction band, we simultaneously create a hole in the valance band. In this sense an optical excitation is a two-particle transition. The same is true for the recombination process. An electron in the conduction band can return radiatively or nonradiatively into the valance band only if there is a free space, i.e., a hole. Two quasi-particles are annihilated in the recombination process. What we need for the understanding of the optical properties of the electronic system of a semiconductor is therefore a description of the excited states of the N-particle problem. The quanta of these excitations are called excitons. Here we will consider the so-called Wannier excitons more specifically. In Wannier excitons, the Bohr radius (i.e. the mean distance between electron and hole) is larger compared to the lattice constant. This situation is well realized in narrow-gap semiconductors which have large ε (>> 1) and small μ (<<m0). Using the effective mass approximation, the Coulomb interaction between electron and hole leads to a hydrogen-like problem with a Coulomb potential term. −e2 . 4πε 0ε re − rh. Indeed excitons in semiconductors form, to a good approximation, a hydrogen or positronium like series of states below the gap. For a simple parabolic band in a direct-gap semiconductor one can separate the relative motion of electron and hole and the motion of the center of mass. This leads to the dispersion relation of exciton as shown in Fig. 2-2 12.

(23) Eex (nB , K ) = Eg − Ry R *. 1 =2 K 2 + 2M nB2. wherre nB = 1, 2, 2 3… is thhe principall quantum number, n Ryy* = 13.6eV V. (2.24)). μ 1 is the excitonn m0 ε 2. bindiing energy, M = me + mh, and K = ke + kh arre the transllational maass and wav ve vector off the exciton, e resspectively. The seriess of exciton n states in (2.24) has an effectiv ve Rydbergg energgy Ry * moodified by the t reduced mass of eleectron and hole h and thee dielectric constant off the medium m in which w thesee particles move. m. Thee radius of the t exciton equals the Bohr B radiuss. of thhe H atom m again moodified by ε and μ.. Using the material pparameters for typicall. semiconductors one finds that t the orbbits of electrron and holle around thheir commo on center off masss average ovver many unit u cells and this in turrn justifies the effectivve mass app proximationn in a self-consiste s ent way. Theses T excittons are callled Wannieer excitons.. Fig. 2-2 2 A pair exciitation in the scheme s of valence and cond duction band in i the exciton picture for a direct d gap semicconductor. [116] 13 1.

(24) 2.4.2 Exciton-related emission [20]. Exciton photoluminescence (PL) is very sensitive to the quality of crystal structures and to the presence of defects. Therefore, the regularities of the change in exciton spectra as a function of composition and structural peculiarities enable one to use them to control the quality of the crystal optically. Impurities and their associations, main lattice defects and their impurity associations, and exciton and exciton impurity complexes all play an important role in the luminescence, which significantly complicates the observation of free-exciton PL. Generally, PL spectrum of single crystal ZnO consists mainly of two bands. The one in UV region corresponds to near-band-edge (NBE) emission is attributed to exciton states; the other one in visible region is due to structural defects and impurities. 2.4.3.1 Free excitons and polaritons. The optical properties of a semiconductor are connected with both intrinsic and extrinsic effects. For a start, the intrinsic excitonic features in the 3.376–3.450 eV range are discussed. The wurtzite ZnO conduction band is mainly constructed from the s-like state having ( Γ c7 ) symmetry, whereas the valence band is a p-like state, which is split into three bands due to the influence of crystal field and spin-orbit interactions. The ordering of the crystal-field and spin-orbit coupling split states of the valence-band maximum in wurtzite ZnO was calculated by our group using tight-binding theory. The obtained electronic band structure of wurtzite ZnO is shown in Fig. 2-3. The valence-band symmetry ordering (A- Γ9 , B- Γ 7 , and C- Γ 7 ) in ZnO is the same as that observed in low temperature PL and magneto-luminescence 14.

(25) measurements by other researchers.. Fig. 2-3 The electronic band structure of wurtzite ZnO. [unpublished]. Group theoretical arguments and the direct product of the group representations of the band symmetries ( Γ 7 for the conduction band, Γ 9 for the A valence band, upper Γ 7 for the B valence band, and lower Γ 7 for the C valence band) will result in the following intrinsic exciton ground state symmetries:. Γ 7 × Γ9 → Γ5 + Γ 6 , Γ 7 × Γ 7 → Γ5 + Γ1 + Γ 2 The Γ5 and Γ 6 exciton ground states are both doubly degenerate, whereas Γ1 and. Γ 2 are both singly degenerate.. Γ 5 and Γ1 are allowed transitions with E⊥c and E║c,. respectively, but the Γ 6 and Γ 2 are not allowed.. 15.

(26) Fig. 2-4 Free excitonic fine structure region of the 10 K PL spectrum for the ZnO single crystal.. Figure 2-4 displays the PL spectrum in the range of fundamental excitonic region measured at 10 K in the E⊥c polarization geometry for a high quality ZnO crystal. A-free exciton and its first excited state emission are observed at FX An =1 = 3.3771 eV (3.3757 eV for. Γ 6 ) and FX An = 2 = 3.4220 eV for Γ 5 (3.4202 eV for Γ 6 ) band symmetry, respectively. Although, at k = 0, Γ 6 exciton is forbidden in the current measurement mode of polarization, it is still observed, due to the fact that the photon has finite momentum. Geometrical effects such as not having the sample orientation exactly perpendicular to the electric field can also be a reason for the observed Γ 6 transition. Using the energy separation of ground state and excited state, the exciton binding energy and band gap energy can be evaluated. The energy difference of about 45 meV gives A-free exciton binding energy of 60 meV and a 16.

(27) corresponding band gap energy of 3.4371 eV at 10 K. Based on the reported energy separation of the A- and B-free excitons (in the range of 9–15 meV), we assigned the weak emission centered at 3.3898 eV, which is about 12.7 meV apart from the A exciton to the B exciton transition. Additional fine structure of exciton lines was also observed in low temperature PL spectra. In strongly polar materials like ZnO transverse Γ 5 excitons couple with photons to form polaritons. Therefore, as indicated in Fig. 2-4, the FX An =1 ( Γ 5 ) exciton line has two components. The higher energy component at 3.3810 eV, which is 3.6 meV apart from the A exciton, can be assigned to the so-called longitudinal exciton (upper polariton branch—UPBA). The lower energy component at 3.3742 eV, which is about 2.9 meV apart from the A exciton, corresponds to the recombination from the “bottleneck” region, in which the photon and free-exciton dispersion curves cross (lower polariton branch—LPBA). As a result, their dispersion curves are illustrated in Figure 2-5.. We will explain the energy. diagram of a system consisting of a crystal and radiation below. Since Γ 6 excitons do not have transverse character, they do not interact with light to form polaritons, and thus have only normal free-exciton dispersion curves as seen in the PL spectra. The incident radiation in the exciton band region is converted to the exciton polariton inside the crystal and it has two modes, the upper and lower branches. The energy of the 17.

(28) polariton is shown in Fig. 2-5 as a function of wave vector k. The broken line represents the relation between the energy and wave vector of the photon in vacuum.. The absorption of a. photon creates an upper-branch polariton at a, which is then scattered to state b of the lower-branch polariton.. The polariton thus formed is thermalized from b to c through. phonon and point-defect scattering.. The thermalized polariton can be annihilated as a. photon at d (direct radiative annihilation) or move in the crystal to be ionized or annihilated nonradiatively at imperfections, since the thermalized state has a longer lifetime.. If the. incident radiation energy is below the bottom of the upper-branch polariton, it generates the lower polariton directly.. Fig. 2-5 Dispersion curves of exciton and exciton polaritons.. 18.

(29) Here two assumptions are necessary. First, most inelastic scattering is caused by the LO phonon.. Second, the probability for a thermalized exciton to be ionized or annihilated at. imperfections is larger than that for it to be annihilated as a photon at d after being scattered from b to c. Under these assumptions if the energy difference between the upper polariton state a and the lower polariton state d is an integral multiple of the LO phonon energy, the cascade scattering of the polariton into state d occurs very efficiently and it is annihilated as a photon. If this energy difference is not an integral multiple of the phonon energy, the polariton is scattered and thermalized in the state near c in the lower-branch polariton and then it is ionized or trapped at imperfections before reaching state d. Thus the LO phonon structure is expected in the excitation spectrum of the free-exciton emission.. 2.4.3.2. LO-phonon replicas. As indicated in Fig. 2-6, the bump at the higher energy side of the spectrum labeled as 1LO (FXA) has a peak around 3.306 eV, which is the expected value for the 1LO-phonon replica of the free exciton peak (about 71 meV apart from the FX An =1 free-exciton peak). Although weak, the second and the third order LO phonon replicas labeled as 2LO (FXA) and 3LO (FXA) are also observed in the PL spectrum. It should be noted that LO-phonon replicas occur with a separation of 71–73 meV, which corresponds to the LO-phonon energy 19.

(30) in ZnO. The peak at 3.2898 eV is the first LO-phonon replica of both 3.3618 and 3.3605 eV lines, whereas the first LO-phonon replica of 3.3650 eV line is seen as a shoulder on the high-energy side of this intense peak.. Fig. 2-6 10 K PL spectrum in the region where LO-Phonon replicas are expected to appear.. 2.4.3 The information obtained from PL spectrum. The temperature dependence of the PL spectrum, intensity and energy of free exciton in particular have been used to obtain information about electronic gap levels in various semiconductors. Below we will discuss two methods to obtain the binding energy and the relation of exciton-phonon coupling.. 20.

(31) 2.4.3.1 Binding energy [21]. It is recognized that a host of different processes contribute to the reduction of the PL intensity as temperature increases. In CdTe, thermal release of trapped carriers, followed by capture in a nonradiative recombination center, is proposed as the principal mechanism of thermal quenching. In some cases the thermal quench is pictured as being essentially due to a transition from the excited state directly to the ground state via the so-called internal mechanism. Irrespective of the specific quenching mechanism, the temperature dependence of the integrated intensity I(T) of PL bands is most often described by the expression. . . . with the process rate parameter α and activation energy higher temperature region (T > 60 K), and then the. (2.25). . If the expression treats for. is the binding energy of free exciton.. 2.4.3.2 Band gap energy variation with temperature [22]. Both mechanisms have the contributions to the variation of band gap energy E(T): exciton-phonon interaction and the thermal lattice expansion. individual lattice oscillators with energies ε. The contributions of the. ω are proportional to the average phonon 21.

(32) exp ε⁄k B T. occupation number, n. 1. . Therefore, E(T) can be represented as ⁄. 0. Here α. /. (2.26). is the slope of the linear T dependence in the high temperature. 0 is the normalized weighting function of the phonon oscillators,. limit, and . |. . . 1.. A convenient analytical treatment of E(T) is obtained: we reduced the integration term to a summation term 0. ∑. ⁄. ,. each term represents a phonon mode to the particular material.. (2.27). represents the weighting. of certain phonon mode. Therefore, we can obtain the contribution of certain phonon mode by deducing the weighting. .. 22.

(33) Chapter 3 Experimental procedures. 3.1 Sample preparation 3.1.1 ZnO quantum dots synthesized by varying concentration of zinc precursor ZnO quantum dots were synthesized by the sol-gel synthesis as the following procedure. Stoichiometric zinc acetate dihydrate [99.5% Zn(OAc)2·2H2O, Riedel-deHaen] was dissolved into diethylene glycol [99.5% DEG, ethylenediamine-tetra-acetic acid (EDTA)] to make 0.1 M, 0.05 M, 0.01 M solutions. The resultant solutions were put separately in a centrifuge operating at 3000 rpm for 30 minutes, and transparent solutions were obtained containing dispersed single crystalline ZnO nanocrystals. Next, the solutions were dropped onto Si (100) substrates and dried at 150°C for further characterization.. 3.1.2 MgxZn1-xO powders synthesized by varying molar ratio of Mg/Zn MgxZn1-xO powders were synthesized using the aqueous sol prepared by stoichiometric zinc acetate dihydrate [99.5% Zn(OAc)2·2H2O, Riedel-deHaen] and magnesium acetate tetrahydrate [99.5% Mg(OAc)2·4H2O, Riedel-deHaen] (Mg/Zn = 0, 3, and 5% in molar ratio) dissolved into methanol.. The sol was dried in a furnace at 900°C under air atmosphere for 1. hour, and then slowly cooled to room temperature.. 23.

(34) 3.2 Photoluminescence system For photoluminescence (PL) measurement, we used a He–Cd laser (λ = 325 nm) as the excitation source.. The schematic of the PL system is shown in Fig 3-1. The excitation. laser beam was directed normally and focused onto the sample surface with power being varied with a variable optical attenuator. The spot size on the sample is about 100 μm. Spontaneous and stimulated emissions were collected by a fiber bundle and coupled into a 0.32 cm focal-length monochromator (TRIAX 320) with a 1200 lines/mm grating, then detected by either an electrically cooled CCD (CCD-3000) or a photomultiplier tube (PMT-HVPS) detector. The temperature-dependent PL measurements were carried out using a closed cycle cryogenic system.. A closed cycle refrigerator was used to set the temperature. anywhere between 15 K and 300 K. He-Cd laser (325 nm). M1. M3 (Focus lens). Triax 320. M4 (Focus lens) M2. Figure 3-1 PL detection system.. 24. Sample holder.

(35) Chapter 4 Results and discussion. 4.1 ZnO quantum dots system 4.1.1. Photoluminescence spectra of different sizes of ZnO particles Figure 4-1 shows the PL spectra of different ZnO sizes at 13 K.. The spectra of ZnO. powders consists of the free exaction emission ( FX An=1 ) and the donor-bound exciton (D0X) emission peaks along with three obvious LO-phonon replicas [23]. The FX An=1 emission of ZnO powders is 3.377eV which behaves as ZnO bulk. The energy shift (dash line) from 3.377 eV to 3.475 eV due to quantum confinement effect (QCE) can be observed. The full width at half maximum which increases as the dot size decreases may be caused by the contribution of surface-optical phonon [24], surface-bound acceptor exciton complexes [25], and size distribution. Accordingly, we observed that LO-phonon replicas are obvious in ZnO powders but are unapparent in other QD-samples.. Duke et al. interpreted the intensities. of LO-phonon replicas depend strongly on their exciton-phonon coupling strengths [26]. This result reveals reducing exciton-phonon coupling as decreasing the particle sizes that is consistent with our previous report [12]. D0X. Intensity (a.u.). LO phonon replicas. n=1. FXA. Powder. 12 nm 7.4 nm 5.3 nm 3.0. 3.2 3.4 Photon Energy (eV). 3.6. Figure 4-1 PL spectra of different ZnO particle sizes at 13K. 25.

(36) 4.1.2. Temperature-dependent PL of different sizes of ZnO particles Figure 4-2 displays the temperature-dependent PL of different sizes of ZnO, it reveals. only single one band for T = 13 ~ 260 K in QDs system.. Due to small binding energy of. D0X, it will be ionized as T > 100 K, so we can easily attribute the single band to the emission.. We also find that the peak energy difference of. in QDs system between. 13 K and 260 K are ~ 44, 18 and 15 meV that are smaller than 65 meV of ZnO powders.. It. is known that the main contribution to the energy shift is the Fröhlich interaction [27], a result of Coulomb interaction. We will do further analysis on this matter in the following section.. P P-1LO. FXn=1 A. P-2LO P-3LO. Log Intensity (a.u.). Powders 13K 30K 50K 80K 100K 140K 180K 220K 260K. 12nm. (b). 13K 30K 50K 80K 100K 140K 180K 220K 260K. Intensity (a.u.). 0. DX. (a). 44 meV. 62 meV. Intensity (a.u.). (C). 3.2. 3.4. 3.6. 3.0. 7.4nm 13K 30K 50K 80K 100K 140K 180K 220K 260K. (d). Intensity (a.u.). 3.0. 3.2. 3.4. 3.6. 5.3nm 13K 30K 50K 80K 100K 140K 180K 220K 260K. 13 meV 18 meV 3.0 3.2 3.4 3.6 3.0 3.2 3.4 3.6 Photon Energy (eV) Photon Energy (eV) Figure 4-2 Temperature-dependence PL spectra of (a) Powders, (b) 12nm, (c) 7.4 nm, (d) 5.3 nm.. 26.

(37) 4.1.3. The weighting of exciton-LO phonon coupling strength (α0) In order to quantitatively investigate the relation between the quantum confinement size. and the exciton-LO phonon interaction, we adopted the formula (2.27).. As our previous. RRS [12] and PL results, the most promising LO-phonon involvement in RRS and PL is the one having energy of 71 to 72 meV corresponding to the A1(LO) phonon.. We therefore take. only a single one of the summation terms of Eq. (2.27) with ħω = 72 meV into account to discuss the exciton-LO phonon coupling and α0 represents the weighting of exciton-A1(LO) phonon coupling.. Although the LO-phonon energy depends on the size of QD, from our. fitting result even for 5.3 nm QD, the phonon energy shift is less than 1 meV, it is insufficient (< 4%) to change α0. We plotted the fitting results α0 = 0.59, 0.40, 0.21, and 0.19 for powders, 12 nm, 7.4 nm, and 5.3 nm QDs, respectively in Fig. 4-3.. These results are. consistent with the observations of PL spectra, weakening coupling strength of exciton-LO phonon as decreasing the particle sizes.. Photon Energy (eV). 3.48 3.44 3.40 3.36. Powders; α0 = 0.59 12 nm. 3.32. 7.4 nm ; α0 = 0.21 5.3 nm ; α0 = 0.19. 3.28 0.0. ; α0 = 0.40. 0.2. 0.4 0.6 -1 1/KT (meV). 0.8. Figure 4-3 Experimental and calculated (solid line) exciton energies plotted against inverse temperature for different ZnO particle sizes.. 27.

(38) 4.1.4. Binding energy From the temperature-dependence PL, we can obtain the exciton binding energy (Eb). from the relation (2.25). The fitting results are shown in Fig. 4-4, Eb of the ZnO powder is 60 meV, which is close to that of ZnO Bulk. respectively, for 12, 7.4, and 5.3 nm QDs.. We obtained Eb = 67, 87 and 132 meV,. The decreasing particle size would raise the. electron-hole interaction as a result of the compressing boundary to cause increasing Coulomb energy. Therefore, the binding energy increases as the particle size decreases.. (a) 60meV. Intensity (a.u.). (b). 67meV (c). 87meV (d). 132meV -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 -1 1/KT (meV). Figure 4-4 The. integral intensity as a function of the inverse temperature from 13 to 300K for (a). Powders, (b) 12nm, (c) 7.4 nm, (d) 5.3 nm.. Squares represent experiment data, while solid lines are the. theoretical fitting.. 4.1.5. Bohr radius (aB) The increasing Eb gives an indication for reduction of exciton-LO phonon interaction.. The enhancement of Eb or Coulomb potential indicates a reduction of aB.. It makes the. exciton less polar capable for efficiently interacting with LO-phonon through the Fröhlich interaction. To find out the relation between aB and α0, we calculated aB from our PL 28.

(39) spectra including the. emission energy and Eb for different dot sizes based on the. weak confinement model, Eq. (2.9), and aB2 = ħ2/(2µ∗Eb) [28], where. ω is the measured. emission energy plus Eb, Eg = 3.43 eV is the band gap energy of bulk ZnO, e is the charge of electron, ħ is Planck’s constant divided by 2π, R is the particle radius, µ* is the reduced mass of exciton, ε = 3.7 is the relative permittivity [29], and ε0 is the permittivity of free space. The calculated exciton Bohr radii aB-QD for 5.3 nm, 7.4 nm and 12 nm QDs are 0.977 nm, 1.038 nm and 1.328 nm.. The ratios of aB QD to the exciton Bohr radius for bulk. ZnO of aB bulk = 2.34 nm are 0.42, 0.46 and 0.57, respectively, which agree well with 0.42, 0.49 and 0.59 obtained by Senger, et al. [30]. 4.1.6. Correlation of α0 and aB Figure 4-5 shows similar trends of α0 QD/α0 Powders and aB QD/aB bulk against the dot size.. It represents that the exciton formation is attained by Coulomb interaction, as the particle sizes decrease, the quantum confinement effect causes increase of Eb and decrease of aB. The electric dipole, which is proportional to the distance of electron-hole pair, is then reduced. The exciton formation thus becomes less polar so that reducing the coupling strength with the polar lattice via the Fröhlich interaction.. Consequently, we demonstrated the reduction of. exciton-LO phonon interaction occurs in ZnO-QD system. 0.7. Ratio. 0.6 0.5 0.4 α0 ratio. 0.3 0.2. aB ratio 4. 6. 8 10 Dot size (nm). 12. Figure 4-5 The relation of α0 QD/α0 Powders and aB QD/aB Powders with different dot sizes. 29.

(40) 4.1.7. Summary The temperature-dependent PL of different sizes of ZnO particles showed above. unobvious LO-phonon replicas of. in ZnO QDs.. The decreasing. emission. energy difference of 13-300 K and the increasing exciton binding energy Eb with the decreasing quantum dot size can be obtained from temperature-dependent PL.. The deduced. α0 revealed the reduction of exciton-LO phonon interaction. The reduced aB with particle size obtained from Eb confirms that the exciton becomes less polar in turn reducing the Fröhlich interaction and the exciton-LO phonon interaction is reduced with decreasing ZnO QDs.. 4.2 MgxZn1-xO alloy system We have successfully deduced the relation of exciton-LO phonon interaction with shrinking ZnO QDs above.. Next, we used the same process to examine the exciton-LO. phonon interaction in the MgxZn1-xO system. 4.2.1. Morphology and temperature-dependent photoluminescence. The average crystallite sizes of MgxZn1-xO and ZnO powders were estimated from the scanning electron microscopy images in Fig. 4-6 and its inset, respectively, to be 1 ± 0.5 μm that does not reveal quantum size effect (Bohr radius of exciton in ZnO is ~2.34 nm [30, 31]). Accordingly, the exciton behavior was affected mainly by the incorporation of Mg. The near band edge (NBE) emissions of the ZnO and Mg0.05Zn0.95O samples measured at various temperatures are shown in Fig. 4-2(a) and 4-7, respectively. First, we discuss the PL spectra of the undoped sample in order to comparatively assign the PL peaks of the Mg0.05Zn0.95O powders. At T = 13 K (Fig. 4-2(a)), the A-free exciton emission (denoted by FX An=1 ) is. 30.

(41) observed at 3.373 eV. Besides the broad line at the lower energy shoulder around 3.308 eV labeled as "P", we also found several LO-phonon replicas separated by a constant interval of 71-73 meV.. The P line can be resolved into two-electron satellite [32], donor-to-acceptor. pair [33, 34], exciton-exciton scattering [35], and 1LO-phonon replica of FX An=1 and DoX, respectively.. With increasing temperature, the relative intensity of FX An=1 increases. whereas that of DoX decreases and becomes not detectable for T > 80 K. This is due to the thermal dissociation of bound exciton into free exciton at higher temperature. As further increasing temperature, the A exciton, and P line finally merge into a broad peak.. At RT, the. exciton peak position at ~ 3.309 eV.. Fig. 4-6 The SEM image of the 5% Mg sample. The inset shows the pure ZnO powders.. The NBE emission of the Mg0.05Zn0.95O sample was shown in Fig. 4-7; it would evolve from that of the ZnO sample.. According to the opposite temperature dependence of the. relative emission intensity, the peaks at 3.422 eV is reasonably assigned to the FX An=1 transitions. The NBE peak position at RT exhibits a maximum blueshift of ~ 74 meV as 5 at.% Mg was introduced into the ZnO powders. Based on the formula: E(MgxZn1−xO) = E(ZnO) + 1.64 * x (eV), the Mg content in the MgxZn1−xO thin films has been determined for 0 ≤ x ≤ 0.2, grown by PLD [36]. Here, E(MgxZn1−xO) and E (ZnO) are the NBE emission 31.

(42) peak positions of MgxZn1−xO and ZnO, respectively.. The calculated Mg contents of our. MgxZn1−xO alloys were shown in Table 4-1, e.g., the molar ratio of Mg/Zn = 5% was estimated to be ~ 4.51%. Therefore, the blueshift of the UV emission demonstrated that the Mg ions had incorporated into the ZnO host lattice.. Mg0.05Zn0.95O. Log Intensity (a.u.). 13K 50K 80K 100K 140K 180K 220K 260K. 47meV 3.0. 3.3 Photon Energy (eV). 3.6. Fig. 4-7 Temperature dependence of PL spectra in Mg0.05Zn0.95O powders. The dashed line marked the peak energies of 13 and 260K. Binding energy. 4.2.2. Figures 4-4(a) and 4-8 show the peak intensity of the FX An=1 emission as a function of reciprocal temperature for the undoped and Mg-doped samples, respectively.. The data can. be described by Eq. (2.25). The curve fitting gives rise to two binding energies (Eb) of about 60 meV for ZnO powders (Fig. 4-3(a)) and about 73 meV for Mg0.05Zn0.95O powders (Fig. 4-8).. Table 4-1 clearly shows that the binding energy of FX An=1 increases markedly in 32.

(43) samples of higher x. Therefore, it implies that the exciton localization takes place and that the degree of localization increases with increasing x.. The essential origin of the localization. is thought to be the spatial fluctuation of the local composition of Mg in the alloys, which results in the spatial fluctuation of the potential energy for the excitons.. Intensity (a.u.). Mg0.05Zn0.95O. 73meV 0.0. 0.2. 0.4. -1. 1/KT (meV). 0.6. Fig. 4-8 Normalized integrated intensity of ZnO sample as a function of temperature.. The weighting of exciton-LO phonon coupling strength (α0). 4.2.3. It is well known the increase of Eb may result from the decrease of aB or effective reduced mass. Due to the electron is more delocalized than the hole in the bulk exciton, the polar lattice experience a net negative charge in the outer regions of the exciton, while it experiences a net positive one in the inner regions.. This charge inhomogeneity couples to. the polar lattice via the Fröhlich mechanism that leads to reduce aB and to reduce the exciton-LO-phonon coupling.. Thus, we further deduced the coupling strength of free. exciton A with LO phonon from temperature-dependent energy shift of FX An=1 as increasing Mg concentration according to the expression (2.27).. Here E(0) represents the emission. energy of free exciton A at T = 0 K, =ωLO = 72 meV is the LO phonon energy, and α0i is a proportional weighting which reflects a change in the exciton-LO-phonon interaction. Figure 4-9 depicts the PL-peak energy shift as a function of the temperature with different Mg 33.

(44) concentrations. powders at 13 K.. Notice that we had set the PL-peak energy position to zero for Mg0.05Zn0.95O By fitting the experimental data for all the MgxZn1−xO and ZnO powders. from T = 13 to 260 K with Eq. (2.27), the exciton-LO-phonon coupling of each MgxZn1−xO sample are compared with ZnO powder by taking the average ratios, α0 MgxZn1−xO/α0 ZnO, which are listed in Table 4-1 clearly show decreasing as more Mg incorporation. Even though aB and α0 are not directly proportional, there is a relation: If aB reduces α0 will also reduce. Assuming that the Mg-doped ZnO powders have almost the same effective electron and hole masses as the undoped ZnO powders due to a small amount of Mg2+ substitution for Zn2+, we calculated aB from Eb.. Photon Energy (eV). 3.42 3.40 3.38 3.36 3.34. ZnO powders. ;α0=0.59. Mg0.03Zn0.97O powders ;α0=0.56 Mg0.05Zn0.95O powders ;α0=0.53. 3.32 0.0. 0.2. 0.4 0.6 -1 1/KT (meV). 0.8. Fig. 4-9 Dependence of PL peak energy positions on temperature for the MgxZn1−xO alloys (x = 5%, 3%, and zero).. Correlation of α0 and aB. 4.2.4. The correlation between α0 and aB normalized to ZnO powders are shown in Fig. 4-10 with open triangles and squares, respectively. Besides, the solid dots are obtained by the modified effective masses, which were corrected from MgZnO electronic band structure using semi-empirical tight-binding approach sp3 model [37, 38] and virtual-crystal approximation 34.

(45) method [39] done by Mr. Kuo-Feng Lin in our laboratory. Therefore, a contraction of aB will make it less polar thereby reducing the coupling to LO phonons.. The results show that. relaxation by means of LO phonon becomes less important as the more Mg incorporation. Consequently, we attribute this reducing coupling effect to increase in Eb with raising Mg mole fraction up to 5% of MgxZn1−xO powders.. The reduction of α0 Ex-LO has to taken into. account for the MgZnO-based excitonic device performance in which carrier relaxation to the exciton ground state is a crucial parameter. 1.02 0.99. Ratio. 0.96 0.93. α0 ratio. 0.90. ab ratio ab ratio. 0.87 0.84 0. 1. 2. (correction) (Experiment) 3. 4. 5. Mg content x (%) Fig. 4-10 The coupling strength of the exciton-LO-phonon given as the average ratio α0 ratio = α0 MgxZn1−xO/α0 ZnO. For comparison, the diminution of ab for experiment and correction is also given as ab ratio = ab MgxZn1−xO/ab ZnO.. Samples : MgxZn1-xO. x=0. x=3%. x=5%. Calculated Mg content (%). 0. 2.9. 4.5. Exciton binding energy (eV). 60 ± 4.7. 70 ± 6.7. 73 ± 8.4. 1. 0.95. 0.91. Exciton-LO phonon coupling strength (α0 MgxZn1−xO/α0 ZnO). Table 4-1 Summary of the results of the temperature-dependent PL characterization.. 4.2.5. Summary 35.

(46) The temperature-dependent NBE PL spectra of MgxZn1−xO powders within the range 0 ≤ x ≤ 0.05 were measured from 13 K to RT.. We deduced experimentally exciton binding. energy showing elevation in powders up to 5% Mg substitution. It is suggests that the localization of excitons, because of the compositional fluctuation, takes place in MgxZn1-xO alloys and that the degree of the localization increases with increasing x.. The reduction of. exciton-LO phonon interaction may originate from a diminution in aB making the exciton less polar, which could be explained by the dopant-induced increase of exciton binding energy.. 36.

(47) Chapter 5 Conclusion and perspectives. 5.1 Conclusion We presented temperature-dependent PL of different sizes of ZnO particles and different Mg substitution in MgxZn1−xO within the range 0 ≤ x ≤ 0.05.. The unobvious LO-phonon. replicas of the exciton A emission, FX An=1 , were observed when the ZnO particle sizes were under 12 nm in diameter.. The FX An=1 emission energy difference of 13-260 K decreases as. the particle size decreases and Mg substitution raises.. The increasing exciton binding energy. with the decreasing quantum dot size and the increasing Mg content can be obtained from temperature-dependent PL.. From the temperature-dependent change of FX An=1 emission. energy, the exciton-LO phonon coupling strength reduces as the particle size decreases and Mg substitution increases. The former is consistent with reducing LO-phonon replica in PL spectra and our previous RRS results [12].. The reduced exciton radius with particle size. obtained from exciton binding energy and PL spectra confirms that the exciton becomes less polar in turn reducing the Fröhlich interaction and the exciton-LO phonon interaction is reduced with decreasing ZnO QDs and raising Mg content in MgxZn1−xO.. 5.2 Perspectives In order to combine the advantages of QDs and Mg-doped ZnO, ZnO-MgO core-shell structure is the most interesting topic for our work.. That can overcome the effect of surface. defects in ZnO, especially in QDs system which has large surface-volume ratio.. Because the. band gap energy of MgO is much larger than that of ZnO, the ZnO-MgO core shell QDs should have better emission efficiency than ZnO QDs.. 37.

(48) We will also suggest to use simple sol-gel method to add Mg(OAc)2·4H2O in ZnO QDs solution then stir at 40°C. 300°C to 700°C.. When the prepared new solution drops on Si then is annealed for. The primary results of the low temperature PL spectra of samples. annealed at various temperatures are shown in Fig. 5-1. We can observe that a broaden emission from MgZnO alloy and two sharp emission from ZnO.. We also found that the. defect emission has been diminished from sample 300°C to 700°C. To further investigate more characteristic of ZnO-MgO core-shell QDs, the temperature dependent PL and time-resolved PL should proceed.. 3.383 eV 3.368 eV. Intensity (a.u.). 3.615 eV 300C. 400C. 3.333 eV 3.311 eV. 500C 600C 700C 3.0. 3.2 3.4 3.6 Photon Energy (eV). Figure 5-1 PL spectra of different anneal temperatures for ZnO-MgO at 13K.. 38.

(49) References [1] C. R. Ding, S. W. Li , H. Z. Wang, Appl. Phys. Lett. 90, 241918 (2007). [2] C. Y. Lee, Y. T. Huang, W. F. Su, and C. F. Lin, Appl. Phys. Lett. 89, 231116 (2006). [3] K. Tachibana, T. Someya, Y. Arakawa, R. Werner, A. Forchel, Appl. Phys. Lett. 75, 2605 (2005). [4] G. Yusa, H. Sakaki, Superlattices and Microstructures 25, 247 (1999). [5] A. Murayama, T. Furuta, K. Hyomi, I. Souma, Y. Oka, D. Dagnelund, I. A. Buyanova, and W. M. Chen, Phys. Rev. B 75, 195308 (2007). [6] Y. H. Cho, H. S. Kwack, B. J. Kwon, J. Barjon, J. Brault, B. Daudin, and L. S. Dang, Appl. Phys. Lett. 89, 251914 (2006). [7] J. Li, S. H. Wei, S. S. Li, and J. B. Xia, Phys. Rev. B 74, 081201 (2006). [8] J. H. Lim, C. K. Kang, K. K. Kim, I. K. Park, D. K. Hwang, and S. J. Park, Adv. Mater. 18, 2720 (2006). [9] J. Y. Bae, J. Yoo, and G. C. Yi, Appl. Phys. Lett. 89, 173114 (2006). [10] H. Shibata, H. Tampo, K. Matsubara, A. Yamada, K. Sakurai, S. Ishizuka, and S. Niki, and M. Sakai, Appl. Phys. Lett. 90, 124104 (2007). [11] P. Ramvall, P. Riblet, S. Nomura, and Y. Aoyagi, J. Appl. Phys. 87, 3883 (2000). [12] H. M. Cheng, K. F. Lin, H. C. Hsu, W. F. Hsieh, Appl. Phys. Lett. 88, 261909 (2006). [13] R. Chang, and S. H. Lin, Phys. Rev. B 68, 045326 (2003). 39.

(50) [14] H. D. Sun, T. Makino, Y. Segawa, M. Kawasaki, A. Ohtomo, K. Tamura, and H. Koinuma, Appl. Phys. Lett. 78, 2464 (2001). [15] C. J. Brinker and G. W. Scherer, “Sol-Gel Sience”, p. 303. [16] C. F. Klingshirn, “Semiconductor Optics”, p. 169. (1997) [17] Sun-Bin Yin, “Fabrication and Characterization of CdS and ZnSe Microcrystalline Doped Glass Thin Films by Pulsed Laser Deposition”, p. 7. National Chiao Tung University Department of Photonics, (1999). [18] M. A. Stroscio and M. Dutta, “Phonons in Nanostructures”, p. 41, Cambridge University press, (2005). [19] C. F. Klingshirn, "Semiconductor Optics" , Springer, Berlin, (1997) [20] A. Teke, U. Ozgur, S. Dogan, X. Gu, and H. Morkoc, Phys. Rev. B, 70, 195207 (2004). [21] D. S. Jiang, H. Jung, and K. Ploog, J. Appl. Phys. 64, 1371 (1988). [22] G. Ortner, M. Schwab, and M. Bayer, Phys. Rev. B, 72, 085328 (2005). [23] C. J. Pan, K. F. Lin,W. T. Hsu, and W. F. Hsieh, Appl. Phys. Lett. 91, 111907 (2007). [24] Z. D. Fu, Y. S. Cui, S. Y. Zhang, J. Chen, D. P. Yu, S. L. Zhang, L. Niu and J. Z. Jiang, Appl. Phys. Lett. 90, 263113 (2007). [25] V. A. Fonoberov and A. A. Balandin, Appl. Phys. Lett. 85, 5971 (2004). [26] C. B. Duke and G. D. Mahan, Phys. Rev. 139, 1965 (1965). [27] S. J. Sheih, K. T. Tsen, D. K. Ferry, A. Botchkarev, B. Sverdlov, A. Salvador, and H. 40.

(51) Morkoc, Appl. Phys. Lett. 67, 1757 (1995). [28] N. Zettili, Quantum Mechanics, Concepts and Applications (2004). [29] S.A. Studenikin, N. Golego, M. Cocivera, J. Appl. Phys. 84, 2287 (1998). [30] R. T. Senger, and K. K. Bajaj, Phys. Rev. B 68,045313 (2003). [31] K. F. Lin, H. M. Cheng, H. C. Hsu, L. J. Lin, and W. F. Hsieh, Chem. Phys. Lett. 409, 208 (2005). [32] A. Teke, U. Ozgur, S. Dogan, X. Gu, H. Morkoc, B. Nemeth, J. Nause, and H. O. Everitt, Phys. Rev. B 70, 195207 (2004). [33] B. P. Zhang, N. T. Binh, Y. Segawa, K. Wakatsuki, and N. Usami, Appl. Phys. Lett. 83, 1635 (2003). [34] B. P. Zhang, N. T. Binh, Y. Segawa, Y. Kashiwaba, and K. Haga, Appl. Phys. Lett. 84, 586 (2004). [35] A. Yamamoto, K. Miyajima, T. Goto, H. J. Ko, and T. Yao, J. Appl. Phys. 90, 4973 (2001). [36] M. Lorenz, E. M. Kaidashev, A. Rahm, Th. Nobis, J. Lenzner, G. Wagner, D. Spemann, H. Hochmuth, and M. Grundmann, Appl. Phys. Lett. 86, 143113 (2005). [37] A. Kobayashi, O. F. Sankey, S. M. Volz, and J. D. Dow, Phys. Rev. B 28, 935 (1983). [38] K. F. Lin, C. J. Pan, and W. F. Hsieh, submitted to Appl. Phys. Lett. [39] D. W. Jenkins, and J. D. Dow, Phys. Rev. B 39, 3317 (1989). 41.

(52)

數據

相關文件

好了既然 Z[x] 中的 ideal 不一定是 principle ideal 那麼我們就不能學 Proposition 7.2.11 的方法得到 Z[x] 中的 irreducible element 就是 prime element 了..

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

For pedagogical purposes, let us start consideration from a simple one-dimensional (1D) system, where electrons are confined to a chain parallel to the x axis. As it is well known

The observed small neutrino masses strongly suggest the presence of super heavy Majorana neutrinos N. Out-of-thermal equilibrium processes may be easily realized around the

incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement..

(1) Determine a hypersurface on which matching condition is given.. (2) Determine a

• Formation of massive primordial stars as origin of objects in the early universe. • Supernova explosions might be visible to the most

Miroslav Fiedler, Praha, Algebraic connectivity of graphs, Czechoslovak Mathematical Journal 23 (98) 1973,