A comprehensive review of zno materials and devices

TL;DR: The semiconductor ZnO has gained substantial interest in the research community in part because of its large exciton binding energy (60meV) which could lead to lasing action based on exciton recombination even above room temperature.

Abstract: The semiconductor ZnO has gained substantial interest in the research community in part because of its large exciton binding energy (60meV) which could lead to lasing action based on exciton recombination even above room temperature. Even though research focusing on ZnO goes back many decades, the renewed interest is fueled by availability of high-quality substrates and reports of p-type conduction and ferromagnetic behavior when doped with transitions metals, both of which remain controversial. It is this renewed interest in ZnO which forms the basis of this review. As mentioned already, ZnO is not new to the semiconductor field, with studies of its lattice parameter dating back to 1935 by Bunn [Proc. Phys. Soc. London 47, 836 (1935)], studies of its vibrational properties with Raman scattering in 1966 by Damen et al. [Phys. Rev. 142, 570 (1966)], detailed optical studies in 1954 by Mollwo [Z. Angew. Phys. 6, 257 (1954)], and its growth by chemical-vapor transport in 1970 by Galli and Coker [Appl. Phys. ...

Summary

A comprehensive review of ZnO materials and devices

• This review gives an in-depth discussion of the mechanical, chemical, electrical, and optical properties of ZnO in addition to the technological issues such as growth, defects, p-type doping, band-gap engineering, devices, and nanostructures.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• This present review is distinguishable from the other reviews50–55 in that the previous ones focused mainly on material processing, doping, and transport properties, while the present one treats those topics in greater depth in addition to an in-depth discussion of the growth, optical properties, p-type doping, and device fabrication aspects.

A. Crystal structures

• Most of the group-II-VI binary compound semiconductors crystallize in either cubic zinc-blende or hexagonal wurtzite structure where each anion is surrounded by four cations at the corners of a tetrahedron, and vice versa.
• Using the LDA calculation technique, the equilibrium cohesive energy of ZnO was reported as −9.769, −9.754, and −9.611 eV for wurtzite, zinc-blende, and rocksalt structures, respectively.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The measured lattice-plane spacings as a function of pressure for the B1 phase are shown in Fig.
• Since none of the three structures described above possess an inversion symmetry, the crystal exhibits crystallographic polarity, which indicates the direction of the bonds, i.e., closed-packed 111 planes in zinc-blende and rocksalt structures and corresponding 0001 basal planes in the wurtzite structure differ from 1̄1̄1̄ and 0001̄ planes, respectively.

B. Lattice parameters

• The lattice parameters of a semiconductor usually depend on the following factors: i free-electron concentration acting via deformation potential of a conduction-band minimum occupied by these electrons, ii concentration of foreign atoms and defects and their difference of ionic radii with respect to the substituted matrix ion, iii external strains for example, those induced by substrate , and iv temperature.
• For the wurtzite ZnO, lattice constants at room temperature determined by various experimental measurements and theoretical calculations are in good agreement.
• The lattice constant measured with the RHEED technique is in very good agreement with the theoretical prediction.
• The discrepancy in the calculated values is larger than the measured ones.
• Even though the variation with pressure seems within the experimental error, this pressure coefficient is in agreement with previously published experimental −0.0007 GPa−1 Ref. 65 and predicted −0.0005 GPa−1 Ref. 56 values.

C. Electronic band structure

• The band structure of a given semiconductor is pivotal in determining its potential utility.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• A surface-related state was not identified and all the observed peaks were attributed to the bulk-band tran- sitions.
• They obtained very good results for the anion p valence bands and for the band-gap energies but no assertion concerning the d-band positions could be made.
• The electronic band structure of the other phases of ZnO has also been studied by a number of researchers.

D. Mechanical properties

• The mechanical properties of materials involve various concepts such as hardness, stiffness, and piezoelectric constants, Young’s and bulk moduli, and yield strength.
• The elastic constants C11 and C66 can be directly obtained from measurement of the phase velocity of the longitudinal mode and of the shear horizontal mode traveling parallel to the crystal surface.
• Besides the experimental investigations, many theoretical calculations have also been employed to determine the structural and mechanical properties of ZnO.
• It has been shown that the large piezoelectric tensor of ZnO is due to the low value of its clamped-ion contribution reducing the cancelation effect ; besides, the piezoelectric tensor is domi- [This article is copyrighted as indicated in the article.
• HAtomistic calculations based on an interatomic pair potential within the shell-model approach Ref. 70 .

E. Lattice dynamics

• A fundamental understanding of the thermal as well as electrical properties in terms of low- and high-field carrier transports requires precise knowledge of the vibrational modes of the single crystal.
• Infrared reflection and Raman spectroscopies have been commonly employed to derive zone-center and some zone-boundary phonon modes in ZnO.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• So far, a number of reports has appeared for the infrared and Raman modes, which have been associated with local vibrational modes of impurities, dopants, and defect complexes.

1. Thermal-expansion coefficients

• The lattice parameters of semiconductors are temperature dependent and quantified by thermal-expansion coeffi- [This article is copyrighted as indicated in the article.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The temperature dependence of the lattice constants a and c, and the thermal-expansion coefficients of hexagonal ZnO have been determined by the capacitive method.
• The c0 parameter did not give any minimum, most probably due to its less precision and uncertainty in the measurement.
• They investigated the thermalexpansion coefficient as a function of some growth parameters, such as substrate temperature, sputtering power, etc.

2. Thermal conductivity

• Thermal conductivity , which is a kinetic property determined by the contributions from the vibrational, rotational, and electronic degrees of freedom, is an extremely important material property when high-power/hightemperature electronic and optoelectronic devices are considered.
• The measurements were made at different points on each sample and the results are also shown in Table VI.
• To be consistent with the earlier models, one would argue that forming-gas annealing has resulted in surface roughness, which has considerably reduced .
• Curves of thermal conductivity versus temperature for the three sets of ZnO samples micrometer, submicrometer, and nanometer , measured as they were heated to 600 °C and cooled back to 25 °C, are shown in Fig. 17.

3. Specific heat

• The specific heat of a semiconductor has contributions from lattice vibrations, free carriers very effective at low temperatures , and point and extended defects.
• Only Lawless and Gupta145 investigated the specific heat for both pure and varistor types of ZnO samples between the temperature ranges of 1.7 and 25 K, where the latter has an average grain size of 10 m.
• This difference was attributed to the contribution by the large [This article is copyrighted as indicated in the article.
• The coefficient in front of the term R has been multiplied by 2 to take into account the two constituents making up the binary compound.
• By fitting the measured temperature-dependent heat capacity to the Debye expression, one can obtain the Debye temperature D specific to the heat capacity.

G. Electrical properties of undoped ZnO

• Advantages associated with a large band gap include higher breakdown voltages, ability to sustain large electric fields, lower noise generation, and hightemperature and high-power operation.
• The electron transport in semiconductors can be considered for low and high electric fields.
• When the electric field is increased to a point where the energy gained by electrons from the external field is no longer negligible compared to the thermal energy of the electron, the electron distribution function changes significantly from its equilibrium value.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• IP: 128.172.48.58 On: Mon, 19 Oct 2015 20:45:44 decreased to submicron range, transient transport occurs when there is minimal or no energy loss to the lattice during a short and critical period of time, such as during transport under the gate of a field-effect transistor or through the base of a bipolar transistor.

1. Low-field transport

• Hall effect is the most widely used technique to measure the transport properties and assess the quality of epitaxial layers.
• Dislocation scattering is due to the fact that acceptor centers are introduced along the dislocation line, which capture electrons from the conduction band in an n-type semiconductor.
• Experimental investigation of the temperature-dependent carrier mobility and concentration can be used to determine the fundamental material parameters and understand the carrier scattering mechanisms along with an accurate comparison with theory.
• Table VII gives the selected best values of electron mobility and corresponding carrier concentration in bulk and thin-film ZnO grown by various techniques.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

2. High-field transport

• Ensemble Monte Carlo MC simulations have been the popular tools to investigate the steady-state electron transport in semiconductors theoretically.
• The calculated electron drift velocity versus electric-field characteristics are plotted in Fig. 25 for wurtzite-phase ZnO [This article is copyrighted as indicated in the article.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• With increasing electric field the drift velocity increases and reaches a peak value of 3 107 cm/s at 250 K.
• The peak fields differ considerably, with the peak field for GaN being at a field which is about 100 kV/cm lower than that of ZnO.

III. ZnO GROWTH

• The growth of ZnO thin films has been studied for acoustical and optical devices because of their excellent piezoelectric properties and a tendency to grow with strong 0001 preferential orientation on various kinds of substrates, including glass,149 sapphire,150 and diamond.
• The improved quality of ZnO films allowed the observation of optically pumped lasing at room temperature.
• In comparison with GaN/Al2O3, ZnO/Al2O3 has approximately equivalent x-ray-diffraction XRD and photoluminescence PL linewidths, and even lower dislocation densities.

A. Bulk growth

• Growth of large-area and high-quality ZnO crystals is important not only for materials science but also for many device applications.
• The low supersaturation of the solution during hydrothermal reaction favors crystal growth.
• When used as substrates for epitaxy, proper surface preparation is necessary to evaluate the quality of hydrothermally grown ZnO.176.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

C. rf magnetron sputtering

• One of the most popular growth techniques for early ZnO investigations was sputtering dc sputtering, rf magnetron sputtering, and reactive sputtering .
• The cross-sectional transmission electron microscopy TEM revealed that the samples grown under the optimum temperature and rf power combination exhibited the largest grain size.
• A more advanced sputtering technique utilizes electroncyclotron-resonance ECR source to supply the power to the plasma.

D. Molecular-beam epitaxy

• The main advantage of molecular-beam epitaxy MBE is its precise control over the deposition parameters and in situ diagnostic capabilities.
• When the O plasma is used, the chamber pressure during growth is in the 10−5-Torr range.
• Successful growth of ZnO films by using hydrogen peroxide H2O2 vapor as a source of active oxygen has also been reported.
• For the correct rf plasma mode, the emission at 777 nm due to the atomic oxygen transition of 3p5P-3s5S0 should dominate the optical emission spectrum OES ,158,213 as shown in Fig. 31.

E. Pulsed-laser deposition

• In the pulsed-laser deposition PLD method, highpower laser pulses are used to evaporate material from a target surface such that the stoichiometry of the material is preserved in the interaction.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The properties of the grown ZnO films depend mainly on the substrate temperature, ambient oxygen pressure, and laser intensity.
• The films grown under lower oxygen pressure regimes 10−5–10−4.
• The distance between the target and the substrate was 30 mm, and the deposition time was kept at 30 min.

F. Chemical-vapor deposition

• Among other growth methods, chemical-vapor deposition CVD technology is particularly interesting not only [This article is copyrighted as indicated in the article.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• In the CVD method, ZnO deposition occurs as a result of chemical reactions of vapor-phase precursors on the substrate, which are delivered into the growth zone by the carrier gas.
• The ZnO films grown by this method show quite high crystal, electrical, and luminescence properties.
• The layers showed smooth surface morphology and high crystalline quality as demonstrated by XRD FWHM of 0002 scans for a 2.28- m-thick layer was 160 arc sec .

B. Optical transitions in ZnO

• Optical transitions in ZnO have been studied by a variety of experimental techniques such as optical absorption, transmission, reflection, photoreflection, spectroscopic ellipsometry, photoluminescence, cathodoluminescence, calorimetric spectroscopy, etc.
• The donor- and acceptorbound excitons are discussed next.
• Finally, LO-phonon replicas of the main excitonic emissions and the donor-acceptor-pair DAP transition are reviewed.
• The temperature dependence of the full PL spectrum is also explored and necessary arguments are made to support the assignments for specific transitions.

1. Free excitons and polaritons

• The wurtzite ZnO conduction band is mainly constructed from the s-like state having 7 c 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.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• A satisfactory agreement was attained when the crossing frequencies of the polariton dispersion and light with E c and E c polarizations were compared with the measured absorption line energies.
• Additionally, the reflection minima at 3.427 and 3.433 eV were assumed to be related to the second and first excited states of the A and B free excitons, respectively.
• Table X tabulates the observed excitonic peak energies reported by Teke et al.299 along with some of the other reported results for high-quality ZnO single crystals from reflectance, photoreflectance, absorption, and PL measurements.

2. Bound excitons

• Bound excitons are extrinsic transitions and are related to dopants or defects, which usually create discrete electronic states in the band gap, and therefore influence both opticalabsorption and emission processes.
• The electronic states of the bound excitons depend strongly on the semiconductor material, in particular, the band structure.
• A basic assumption in the description of the bound exciton states for neutral donors and acceptors is a dominant coupling of the like particles in the bound-exciton states.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Many sharp donor- and acceptor-bound-exciton lines were reported in a narrow energy range from 3.348 to 3.374 eV in ZnO see, for example, Ref. 314 and references therein .

3. Two-electron satellites in PL

• Another characteristic of the neutral-donor-boundexciton transition is the two-electron satellite TES transition in the spectral region of 3.32–3.34 eV.
• The proportionality constant is found to be 0.34, which is close to the 0.3 reported by Alves et al.,309 who calculated the binding energies of the donors as 43, 52, and 55 meV for the TABLE XI.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

4. DAP and LO-phonon replicas in PL

• The spectral region containing the DAP transition and LO-phonon replicas of the main transitions has not been studied widely for single-crystal ZnO.
• Due to the line broadening, the peaks corresponding to each individual bound exciton could not be resolved very well.
• Resolving the second- and higher-order LO replicas is even harder because the energy position 3.218–3.223 eV falls in the spectral region where the DAP transition and its LO-phonon replicas are expected to appear in the PL spectra.
• The relatively broad peak around 3.280 eV in Fig. 54 is the first LO-phonon replica associated with the most intense acceptor-bound-exciton line 3.3564 eV .

5. Temperature-dependent PL measurements

• In order to provide additional support for some of the peak assignments in the low-temperature PL spectrum of the high-quality ZnO substrate investigated, Teke et al.299 studied the temperature evolution of these peaks.
• The variation of both A- and B-exciton peak positions with temperature is shown in the inset of Fig. 55.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Based on this argument and the works of Reynolds et al.306 and Hamby et al.,308 Hamby et al. observed a very good agreement in terms of the temperature-dependent energy positions of this peak with that of the predicted values by taking into account the temperature broadening effect.

C. Time-resolved PL on ZnO

• Time-resolved PL TRPL is a nondestructive and powerful technique commonly used for the optical characterization of semiconductors.
• They noted that free-exciton lifetimes are determined not only by the radiative decay but also by the nonradiative decay and by capture processes leading to bound excitons.
• For the as-received sample the decay constants increase slightly with increasing excitation density, whereas the forming-gas-treated sample follows an opposite trend.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Koida et al.322 studied the correlation between the excitonic PL lifetime at room temperature and point defect density in bulk and epitaxial ZnO layers.

D. Refractive index of ZnO

• Knowledge of the dispersion of the refractive indices of semiconductor materials is necessary for accurate modeling and design of devices.
• In the early 1950s and 1960s, several workers reported the results of optical reflection measurements with light polarized parallel and perpendicular to the c axis, and used the Kramers-Kronig analysis to determine the dielectric functions.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Also shown in the figures are fits to a first-order Sellmeier equation: TABLE XIII.
• When compared to the ZnO data by Teng et al.328 Jellison and Boatner,327 Yoshikawa and Adachi,326 and Hu et al.,325 the ZnO ordinary and extraordinary refractive index values of Schmidt et al.329 are about 0.02 lower and 0.03 higher, respectively, up to 700 nm, and the birefringence is larger.

E. Stimulated emission in ZnO

• Even though n- and p-type dopings have been reported in ZnO thin-films,330 there is no demonstration of electrically pumped lasing in ZnO-based structures.
• Fabrication of low-dimensional structures such as quantum wells and dots has been the focus of semiconductor laser research to decrease the threshold for lasing.
• Excitonic emission may also be used to obtain efficient lasing, which may be realized for ZnO due to its larger exciton binding energy 60 meV 38,331 compared to other wide-band-gap semiconductors.
• In the intermediate excitation density regime emissions due to biexcitonic, exciton-exciton and exciton-carrier interactions may be observed.
• Equation 22 gives 99 meV, which is in good agreement with the experimental results which will be discussed below.

1. Thin films

• Özgür et al.333 have shown that ZnO layers deposited directly on c-plane sapphire substrates by rf magnetron sput- tering could have optical quality sufficient for excitonic laser action.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Figure 60 shows the PL data spectrally integrated between 3.1 and 3.4 eV for all the samples.
• At room temperature, for excitation intensities exceeding 400 kW/cm2 the excitonexciton-scattering-related SE peak appeared at 3.18 eV and grew superlinearly.

2. Polycrystalline ZnO films and “random lasers”

• As previously observed in Ti:sapphire and TiO2 disordered systems,342,343 there have been reports of “random lasing” in polycrystalline ZnO films as a result of strong scattering in disordered dielectric media.344–346.
• Unlike the traditional semiconductor lasers which have well-defined cavities, the random-laser cavities are “self-formed” due to strong optical scattering in the polycrystalline films.
• When the excitation intensity exceeded a threshold, very narrow peaks emerged in the emission spectra.

3. Multiple-quantum wells

• Stimulated emission has been observed also in quantumwell structures utilizing the alloys of ZnO.
• In other II-VI materials SE has been mostly demonstrated only at low temperatures and rarely at room temperature.
• They investigated ten-period MQWs grown on SCAM substrates by laser MBE.
• Figure 70 shows the variation of the SE threshold with well width for two sets of samples having 12% and 26% Mg in the barriers.
• They observed SE induced by both exciton-exciton scattering and EHP recombination.

4. Stimulated-emission dynamics

• The TRPL measurements used the same optical pulse excitation source as the TI-PL measurements mentioned before in Sec. IV E 1, and the PL was detected by an 45-ps resolution Hamamatsu streak camera.
• Figure 71 shows the TRPL data for three annealed samples at room temperature.
• The spontaneous recombination times observed by Özgür et al.333 for the rf-sputtered ZnO thin films are comparable with other values reported in the literature.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The slight decrease in the decay time at 85 K may be explained by increased absorption at low temperatures and the weak carrier density dependence of the recombination times.

A. Predictions from first principles

• As in any semiconductor, point defects affect the electrical and optical properties of ZnO as well.
• A low formation energy implies a high equilibrium concentration of the defect; a high formation energy means that defects are unlikely to form.
• The results of calculations for oxygen and zinc vacancies, interstitials, and antisities in ZnO are shown in Fig. 72 for the two limiting zinc chemical-potential values.
• Indeed, incorporation of hydrogen during growth may increase acceptor solubility and suppress formation of compensating de- Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• These authors suggest that the use of N2 or N2O gas is not effective in p-type doping since energy must be supplied to break the strong N–N bond.

B. Experimental studies of native and unintentionally introduced defects

• Most of the experimental results have been obtained from the analysis of mainly the low-temperature photoluminescence PL data.
• In undoped ZnO, the well-known25 green luminescence GL band peaking at about 2.5 eV usually dominates the defect-related part of the PL spectrum.
• Some information is available about the shallow donoracceptor-pair DAP band having its main peak at about 3.22 eV.
• There are a few reports about other PL bands, in particular, the yellow luminescence in Li-doped ZnO.

1. Shallow acceptor in ZnO

• Besides strong and rich exciton-related emissions in the photon energy range of 3.25–3.4 eV, PL spectrum of undoped high-quality ZnO usually contains a sharp peak at about 3.22 eV followed by at least two LO-phonon replicas.
• As acceptor, the activation energy of which was estimated at 180±10 meV from the temperature dependence of the PL spectrum.
• Note that simultaneous with the DAP emission line, a sharp line at 3.325 eV appeared in these samples, tentatively attributed to the As acceptor-bound exciton.

2. Green luminescence band

• The nature of the GL, appearing at about 2.5 eV in undoped ZnO, remained controversial for decades.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Dietz et al.372 concluded that the Cu2+ t2 wave function is radially expanded more than e wave function relative to the d wave function of the free Cu ion, and the t2 hole spends about 60% of its time on the Cu2+ ion, while it spends the rest of the time in the oxygen sp3 orbitals.
• Note that the g values obtained on the structureless GL band393 and on the GL with a distinct phonon structure371,383 are incompatible, and thus, these two PL bands are related to different defects.

3. Yellow luminescence band

• In contrast to the GL band, the YL band decays very slowly after switching off the excitation source, and can be observed also in the thermoluminescence spectrum.
• The YL is polarized at low temperatures, which was explained by two metastable orientations of the LiZn center in the ZnO lattice.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The YL band saturated with excitation intensity above 10−3 W/cm2, indicating low concentration of the related defects.

4. Red luminescence band

• The RL band is broad with a FWHM of about 0.5 eV, and its shape is Gaussian.
• The quenching of the RL band apparently causes the emer- gence of the GL band Fig. 81 This may be a result of competition for holes between the acceptors responsible for the GL and RL bands.

VI. DOPING OF ZnO

• ZnO has a strong potential for various short-wavelength optoelectronic device applications.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The difficulties can arise from a variety of causes.
• The majority of these studies have dealt with the manner in which dopant solubility397,420 or native defects421,422 such as vacancies, interstitials, and antisites interfere with doping.
• It then appears that perhaps the best candidate for p-type doping in ZnO is N because among the group-V impurities, N has the smallest ionization energy, it does not form the NZn antisite, and the AX center of N is only metastable.

1. Nitrogen doping

• Helped by success with ZnSe nitrogen has been explored for p-type doping of II-VI semiconductors.
• Conduction-type conversion from n-type to p-type did not occur.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Results indicated that N can be incorporated into ZnO films without plasma or high-temperature process and a high N concentration was obtained only under Zn-rich conditions as predicted by Yan et al.441.

3. Other dopants in group V

• While most efforts on p-type doping of ZnO have focused on nitrogen doping with/without codopants because of its relatively shallow level, considerably fewer studies have considered other group-V elements for substitutional doping on the O sites370,467–472.
• Only in some of these studies inversion of the film’s conductivity type was observed.
• Another group was unsuccessful in achieving p-type ZnO using phosphorus;472 however, they recently demonstrated p-type behavior in P-doped ZnMgO layers grown by pulsed-laser deposition.

VII. ZnO-BASED DILUTE MAGNETIC SEMICONDUCTORS

• The III-V and II-VI diluted magnetic semiconductors DMSs have attracted considerable attention because the spin-dependent magnetic phenomena can be manipulated in these low-dimensional tailored magnetic thin films for various spin-based devices to unprecedented capabilities.
• As a consequence, the electronic structure of the substituted 3d transition-metal impurities in semiconductors is influenced by two competing factors: strong 3d-host hybridization and strong Coulomb interactions between 3d-3d electrons.
• If accomplished, above room-temperature ferromagnetism could form the basis for charge, spin-based, or mixed spin- and charge-based devices.
• Then, a brief summary of the main experimental results on ZnO-based DMS have been put forth for the explanation of the ferromagnetism in these compound semiconductors.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

A. Theory of ZnO-based magnetic semiconductors

• There are two basic approaches in understanding the magnetic properties of dilute magnetic semiconductors.
• Therefore, Mn2+ has a relatively large magnetic moment spin S=5/2 and angular momentum L=0 with characteristic of a half-filled d shell and can be incorporated in sizable amounts up to 35% of Mn into ZnO host without affecting much the crystallographic quality of the DMS, whereas about 5% is tolerable for III-V-based hosts.
• This interaction determines the spin splitting of the band states in an external magnetic field, giving rise to interesting magneto-optical and magnetotransport responses in the DMS samples.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• By employing the mean-field theory based on the Zener ferromagnetism model510 Dietl et al.47 and Dietl511 evaluated the Curie temperatures TC for various semiconductors.

B. Experimental results on ZnO-based magnetic semiconductors

• Setting the controversy aside for the time being, what has been reported experimentally will be discussed in this section.
• As the Mn content is increased, the midgap absorption around 3 eV develops and the absorption edge blueshifts.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• The introduction of Mn into these and other materials under the right conditions is found to produce ferromagnetism near or above room temperature.

VIII. BAND-GAP ENGINEERING

• A crucial step in designing modern optoelectronic devices is the realization of band-gap engineering to create barrier layers and quantum wells in device heterostructures.
• In order to realize such optoelectronic devices, two important [This article is copyrighted as indicated in the article.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• While p-type doping of ZnO is under intensive study, the latter has been demonstrated by the development of MgxZn1−xO Refs. 527–540 and CdyZn1−yO alloys,537,541–547 allowing modulation of band gap in a wide range.
• The bowing parameter b depends on the difference in electronegativities of the end binaries ZnO and AO.

A. MgxZn1−xO alloy

• MgxZn1−xO alloy has been considered as a suitable material for the barrier layers in ZnO/ Mg,Zn O superlattice structures537 because alloying ZnO with MgO Eg 7.7 eV enables widening of band gap of ZnO.
• XRD studies show that the a-axis length gradually increases and the c-axis length decreases with increasing Mg content, and therefore the cell volume is hardly changed.
• This indicates that MgxZn1−xO is a suitable material for barrier layers in ZnO/ Mg,Zn O heterostructures with a band-gap offset up to 0.85 eV.
• The exciton binding energy is much higher than that in ZnO and needs corroboration.
• The bowing parameter of the as-grown films calculated using the band-gap values obtained from the absorption spectra is very high b =3.11 .

A. Ohmic contacts to ZnO

• An Ohmic contact can be defined as having a linear and symmetric current-voltage relationship for both positive and negative voltages and is so important for carrying electrical current into and out of the semiconductor, ideally with no parasitic resistance.
• This reduction has been [This article is copyrighted as indicated in the article.
• In addition, Ti/Au contacts showed thermal degradation after annealing at temperatures in excess of 300 °C Ref. 556 and the same behavior was observed by Akane et al.560 for In Ohmic contacts annealed at 300 °C for 5 min.
• Some of the various Ohmic-contact metallization schemes to n-type ZnO are summarized in Table XVIII together with carrier concentration, specific contact resistance, and method used to measure the specific contact resistivity.

C. Heterostructure devices

• As discussed in Sec. VI B the growth of reproducible p-type ZnO films is not developed yet, and therefore fabrication of ZnO p-n homojunction-based LEDs has not been possible.
• This subject has received a great deal of attention recently, and heterojunctions of good quality have been realized using various p-type materials such as Si,587–589 GaN,590,591 TABLE XIX.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.
• Besides, ZnO-based PDs are expected to have superior resistance to ionizing radiation and high-energy particles.

1. Light-emitting devices

• In 1967, Drapak597 fabricated ZnO-based hetererostructure LED.
• For both types intense electro luminescence EL was observed under forward bias; however, spectra obtained from sets I and II heterostructures were different, as shown in Fig. 107.
• By controlling the deposition atmosphere, n-type ZnO conductivity was achieved.
• Alivov et al.590 have reported growth, processing, and fabrication of n-ZnO/ p-GaN heterojunction LED devices, where the 1- m-thick CVD-grown n-ZnO Ga-doped and the MBE-grown p-GaN Mg-doped layers had carrier concentrations of 4.5 1018 and 3 1017 cm−3, respectively.

2. Photodiodes

• There have been many reports regarding the photoresponse properties of the ZnO-based heterojunctions.
• A photoresponsivity as high as 0.3 A W−1 was achieved at 26-V reverse bias under irradiation by 360-nm light that corresponds to the ZnO energy gap.
• Alivov et al.600,601 fabricated n-ZnO/ p-6H-SiC-type PDs growing 0.5- m-thick n-ZnO layer on p-type 6H-SiC substrates at 600 °C by plasma-assisted MBE and demonstrated high-quality p-n heterojunctions.
• Ohmic contacts to the 250- m-diameter mesastructures were made [This article is copyrighted as indicated in the article.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

D. Metal-insulator-semiconductor diodes

• In the absence of p-n junctions, electroluminescence properties of ZnO could be exploited by fabricating metalinsulator-semiconductor MIS structures which do not require p-type ZnO.
• In spite of this, there have been only a few reports on such MIS structures.
• Au metal layer, deposited thermally onto the insulating layer i-layer , was 1.8 mm in diameter.
• The I-V characteristics of a typical diode are shown in Fig. 117.

E. Transparent thin-film transistors

• There has been a great interest in transparent electronics lately,610 especially in conjunction with transparent thin-film transistors TTFTs , because it is expected that the characteristics of TTFT will not degrade on exposure to visible light due to the wide band gap of its active channel layer, whereas the characteristics of amorphous or poly-Si TFT do degrade.
• The most important electrical parameters in quantifying TFT performance are the drain-current on-to-off ratio and active channel mobility.
• A glass substrate was blanket coated with a 200-nm-thick layer of sputtered ITO and a 220-nm-thick layer of aluminumtitanium oxide ATO deposited by atomic layer deposition.
• In another report, Carcia et al.616 fabricated ZnO thinfilm transistors by rf magnetron sputtering on heavily doped n-type Si substrates held near room temperature.

X. ZnO NANOSTRUCTURES

• One-dimensional semiconductor nanowires and nanorods have attracted increasing attention due to their physical properties arising from quantum confinement such as electronic quantum transport and enhanced radiative recombination of carriers .
• Therefore, 1D ZnO structures stimulated so much attention, and a large number of publications have appeared lately reporting nanostructures of various shapes nanowires, nanobelts, nanorings, nanotubes, nanodonuts, nanopropellers, etc. grown by different methods.
• Haupt et al.639 were able to reduce the average diameter of the nanowires below 30 nm using the same technique.
• Figure 130 shows SEM images of the tubes formed at 400 °C under different reactor pressures.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

XI. SUMMARY

• A comprehensive review of properties, preparation, processing, and device applications of ZnO is presented.
• Zinc oxide ZnO material has been a subject of varying degrees of research effort over the decades.
• This is in part due to strong selfcompensation effects when p-type dopants are attempted.
• Clearly, this issue must be overcome before ZnO can be of practical use, as was the case for GaN.
• Of particular interest to nanoscale structures and devices, ZnO is very favorable as compared to other wide-band-gap semiconductors in that this material lends itself nicely to the production of nanostructures from which functional devices have already been fabricated.

ACKNOWLEDGMENTS

• The authors are also thankful to B. Nemeth and J. Nause of Cermet, Inc., for collaborations on many ZnO-related topics and supply of high-quality melt-grown ZnO substrates.
• Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions.

Figures

FIG. 119. The schematics of a typical TTFT structure. Reprinted with permission from R. L. Hoffman, B. J. Norris, and J. F. Wager, Appl. Phys. Lett. 82, 733 2003 . Copyright 2003, American Institute of Physics.

FIG. 120. Electrical characteristics of TTFT: a drain-current–drain-voltage ID-VDS characteristics; b transfer characteristics and gate leakage current for a TTFT with a width-to-length ratio of 10:1 for VDS=10 V. Reprinted with permission from R. L. Hoffman, B. J. Norris, and J. F. Wager, Appl. Phys. Lett. 82, 733 2003 . Copyright 2003, American Institute of Physics.

FIG. 1. Stick and ball representation of ZnO crystal structures: a cubic rocksalt B1 , b cubic zinc blende B3 , and c hexagonal wurtzite B4 . The shaded gray and black spheres denote Zn and O atoms, respectively.

FIG. 3. Schematic representation of a wurtzitic ZnO structure having lattice constants a in the basal plane and c in the basal direction; u parameter is expressed as the bond length or the nearest-neighbor distance b divided by c 0.375 in ideal crystal , and and 109.47° in ideal crystal are the bond angles.

FIG. 2. Total energy vs volume both per ZnO f.u. for the three phases: zinc blende squares , wurtzite diamonds , and rocksalt circles . The zero of energy is the sum of the total energy of an isolated Zn and an isolated O atom. Reprinted with permission from J. E. Jaffe and A. C. Hess, Phys. Rev. B 48, 7903 1993 . Copyright 1993 by the American Physical Society.

FIG. 7. a Off-normal-emission spectra of the clean surface recorded at h =22.5 eV and i=55° along the 12̄10 ̄X̄ direction and at h =27 eV and i=55° along the 0001 ̄X̄ direction. The incidence plane of the light was parallel to the 12̄10 direction for both detection directions. The spectra are shown with 2° interval. The peak positions, indicated by the vertical bars, were determined from the second derivative of the spectra. The position of the valence-band maximum was determined from the normalemission spectrum taken at h =17 eV by extrapolating the onset of the valence-band emission as shown in the inset of the right panel. b The measured dispersion of the O 2p dangling-bond state open circles and the bulk-band-related states filled circles along the ̄X̄ and ̄X̄ axes. The hatched area corresponds to the projected bulk-band region and the bold dashed line indicates the O 2p dangling-bond bands, both of which have been calculated using the sp3 model by Wang and Duke Ref. 93 . The thin dashed lines which are located above the projected bulk bands are the dangling-bond bands obtained from the LDA calculations. Reprinted with permission from K. Ozawa, K. Sawada, Y. Shirotori, K. Edamoto, and M. Nakatake, Phys. Rev. B 68, 125417 2003 . Copyright 2003 by the American Physical Society.

FIG. 69. The dependences of the UV CL peak intensity and FWHM as a function of oxygen partial pressure T=300 K, electron-beam current of 1 A was used for excitation after Alivov et al. Ref. 354 .

FIG. 68. Lasing spectra of the polycrystalline ZnO thin film obtained by oxidation of metallic Zn films. The threshold intensity Ith for the lasing was 9 MW/cm2. Reprinted with permission from S. Cho, J. Ma, Y. Kim, Y. Sun, G. K. L. Wong, and J. B. Ketterson, Appl. Phys. Lett. 75, 2761 1999 . Copyright 1999, American Institute of Physics.

FIG. 67. PL intensity and PL peak position vs annealing temperature. Reprinted with permission from S. Cho, J. Ma, Y. Kim, Y. Sun, G. K. L. Wong, and J. B. Ketterson, Appl. Phys. Lett. 75, 2761 1991 . Copyright 1999, American Institute of Physics.

FIG. 49. Absorption coefficient and excitonic structure for an annealed solid lines and an unannealed dotted line sample at room temperature. The inset shows the absorption coefficient for the annealed samples at 77 K. Reprinted with permission from J. F. Muth, R. M. Kolbas, A. K. Sharma, S. Oktyabrsky, and J. Narayan, J. Appl. Phys. 85, 7884 1999 . Copyright 1999, American Institute of Physics. The C exciton is misinterpreted as the B exciton and the exciton-LO-phonon complex transitions as the C exciton.

FIG. 50. Free-excitonic fine-structure region of the 10-K PL spectrum for a forming-gas-annealed ZnO substrate. Reprinted with permission from A. Teke, Ü. Özgür, S. Doğan, X. Gu, H. Morkoç, B. Nemeth, J. Nause, and H. O. Everitt, Phys. Rev. B 70, 195207 2004 . Copyright 2004 by the American Physical Society.

FIG. 42. Hall mobility and carrier concentration vs oxygen pressure for ZnO growth. For comparison, the data for the ZnO film grown at 10−1 Torr with the nucleation layer grown at 10−4 Torr are also shown. Reprinted with permission from S. Choopun, R. D. Vispute, W. Noch, A. Balsamo, R. P. Sharma, T. Venkatesan, A. Iliadis, and D. C. Look, Appl. Phys. Lett. 75, 3947 1999 . Copyright 1999, American Institute of Physics.

FIG. 43. XRD pattern of a ZnO thin film deposited under optimized conditions. The rocking curve recorded for the 002 line is shown in the inset. Reprinted with permission from V. Craciun, J. G. E. Gardeniers, and I. W. Boyd, Appl. Phys. Lett. 65, 2963 1994 . Copyright 1994, American Institute of Physics.

FIG. 23. Experimental carrier concentration triangles corrected for Hall r factor and theoretical fit solid line as a function of inverse temperature for bulk ZnO. Reprinted from D. C. Look, D. C. Reynolds, J. R. Sizelove, R. L. Jones, C. W. Litton, G. Cantwell, and W. C. Harsch, Solid State Commun. 105, 399 1998 , Copyright 1998, with permission from Elsevier.

FIG. 22. Experimental circles and theoretical solid line Hall mobilities as a function of temperature in bulk ZnO. Reprinted from D. C. Look, D. C. Reynolds, J. R. Sizelove, R. L. Jones, C. W. Litton, G. Cantwell, and W. C. Harsch, Solid State Commun. 105, 399 1998 , Copyright 1998, with permission from Elsevier.

TABLE VII. A compilation of XRD results, electron mobilities, and corresponding carrier concentration obtained in nominally undoped bulk and thin-film ZnO deposited on different substrates by various growth techniques.

FIG. 60. Room-temperature RT spectrally integrated PL for the ZnO samples normalized to the spontaneous emission. The inset shows the spectrally resolved PL for the 1000 °C sample for different excitation densities. The downward-pointing arrow in the inset indicates the exciton-exciton scattering-induced SE peak. Reprinted with permission from Ü. Özgür, A. Teke, C. Liu, S.-J. Cho, H. Morkoç, and H. O. Everitt, Appl. Phys. Lett. 84, 3223 2004 . Copyright 2004, American Institute of Physics.

TABLE XVI. Cauchy model parameters for the MgxZn1−xO alloy system Schmidt et al. Ref. 329 .

FIG. 16. Wurtzite ZnO lattice parameters as a function of temperature. Reprinted with permission from R. R. Reeber, J. Appl. Phys. 41, 5063 1970 . Copyright 1970, American Institute of Physics.

FIG. 54. 10-K PL spectrum for a forming-gas-annealed ZnO substrate in the region where the donor-acceptor-pair transition and LO-phonon replicas are expected to appear. Reprinted with permission from A. Teke, Ü. Özgür, S. Doğan, X. Gu, H. Morkoç, B. Nemeth, J. Nause, and H. O. Everitt, Phys. Rev. B 70, 195207 2004 . Copyright 2004 by the American Physical Society.

FIG. 127. Source-drain current Isd vs gate voltage Vg curves solid line as a function of source-drain voltage Vsd and log scale plot open circles of Isd-Vg Vsd=1.0 V for a ZnO single nanorod FET after polyimide coating on the device. Reprinted with permission from W. I. Park, J. S. Kim, G.-C. Yi, M. H. Bae, and H.-J. Lee, Appl. Phys. Lett. 85, 5052 2004 . Copyright 2004, American Institute of Physics.

Full text

Virginia Commonwealth University
VCU Scholars Compass
1+)86/)'1'3*42598+63-/3++6/3-9(1/)'8/437 +584,1+)86/)'1'3*42598+63-/3++6/3-

A comprehensive review of ZnO materials and
devices
Ü. Özgür
Virginia Commonwealth University94=-96:)9+*9
Ya. I. Alivov
Virginia Commonwealth University
C. Liu
Virginia Commonwealth University
See next page for additional authors
4114;8./7'3*'**/8/43'1;4607'8 .E57).41'67)425'77:)9+*9+-6+&59(7
'684,8.+ 1+)86/)'1'3*42598+63-/3++6/3-422437
=-96"1/:4:$/9+8'1)4256+.+37/:+6+:/+;4,%32'8+6/'17'3**+:/)+74963'14,551/+*.<7/)7
45<6/-.8>9(1/7./3-
D/768/)1+/7(649-.884<49,46,6++'3*45+3'))+77(<8.++584,1+)86/)'1'3*42598+63-/3++6/3-'8#" ).41'67425'778.'7(++3
'))+58+*,46/3)197/43/31+)86/)'1'3*42598+63-/3++6/3-9(1/)'8/437(<'3'98.46/=+*'*2/3/786'8464,#" ).41'67425'7746246+
/3,462'8/4351+'7+)438')8 1/()425'77:)9+*9
4;314'*+*,642
.E57).41'67)425'77:)9+*9+-6+&59(7

Authors
@?=-B6$'1/:4:/9!+0++7.)./04: 4C'3#/8'1/<:698/3 .4'3*4604A
D/7'68/)1+/7':'/1'(1+'8#" ).41'67425'77 .E57).41'67)425'77:)9+*9+-6+&59(7

APPLIED PHYSICS REVIEWS
A comprehensive review of ZnO materials and devices
Ü. Özgür,
a兲
Ya. I. Alivov, C. Liu, A. Teke,
b兲
M. A. Reshchikov, S. Doğan,
c兲
V. Avrutin,
S.-J. Cho, and H. Morkoç
d兲
Department of Electrical Engineering and Physics Department, Virginia Commonwealth University,
Richmond, Virginia 23284-3072
共Received 2 February 2005; accepted 13 June 2005; published online 30 August 2005兲
The semiconductor ZnO has gained substantial interest in the research community in part because
of its large exciton binding energy 共60 meV兲 which could lead to lasing action based on exciton
recombination even above room temperature. Even though research focusing on ZnO goes back
many decades, the renewed interest is fueled by availability of high-quality substrates and reports of
p-type conduction and ferromagnetic behavior when doped with transitions metals, both of which
remain controversial. It is this renewed interest in ZnO which forms the basis of this review. As
mentioned already, ZnO is not new to the semiconductor field, with studies of its lattice parameter
dating back to 1935 by Bunn 关Proc. Phys. Soc. London 47, 836 共1935兲兴, studies of its vibrational
properties with Raman scattering in 1966 by Damen et al. 关Phys. Rev. 142, 570 共1966兲兴, detailed
optical studies in 1954 by Mollwo 关Z. Angew. Phys. 6, 257 共1954兲兴, and its growth by
chemical-vapor transport in 1970 by Galli and Coker 关Appl. Phys. Lett. 16, 439 共1970兲兴. In terms
of devices, Au Schottky barriers in 1965 by Mead 关Phys. Lett. 18, 218 共1965兲兴, demonstration of
light-emitting diodes 共1967兲 by Drapak 关Semiconductors 2, 624 共1968兲兴, in which Cu
2
O was used
as the p-type material, metal-insulator-semiconductor structures 共1974兲 by Minami et al. 关Jpn. J.
Appl. Phys. 13, 1475 共1974兲兴, ZnO/ZnSe n-p junctions 共1975兲 by Tsurkan et al. 关Semiconductors
6, 1183 共1975兲兴, and Al/Au Ohmic contacts by Brillson 关J. Vac. Sci. Technol. 15, 1378 共1978兲兴
were attained. The main obstacle to the development of ZnO has been the lack of reproducible and
low-resistivity p-type ZnO, as recently discussed by Look and Claflin 关Phys. Status Solidi B 241,
624 共2004兲兴. While ZnO already has many industrial applications owing to its piezoelectric
properties and band gap in the near ultraviolet, its applications to optoelectronic devices has not yet
materialized due chiefly to the lack of p-type epitaxial layers. Very high quality what used to be
called whiskers and platelets, the nomenclature for which gave way to nanostructures of late, have
been prepared early on and used to deduce much of the principal properties of this material,
particularly in terms of optical processes. The suggestion of attainment of p-type conductivity in the
last few years has rekindled the long-time, albeit dormant, fervor of exploiting this material for
optoelectronic applications. The attraction can simply be attributed to the large exciton binding
energy of 60 meV of ZnO potentially paving the way for efficient room-temperature exciton-based
emitters, and sharp transitions facilitating very low threshold semiconductor lasers. The field is also
fueled by theoretical predictions and perhaps experimental confirmation of ferromagnetism at room
temperature for potential spintronics applications. This review gives an in-depth discussion of the
mechanical, chemical, electrical, and optical properties of ZnO in addition to the technological
issues such as growth, defects, p-type doping, band-gap engineering, devices, and nanostructures.
© 2005 American Institute of Physics. 关DOI: 10.1063/1.1992666兴
TABLE OF CONTENTS
I. INTRODUCTION............................ 2
II. PROPERTIES OF ZnO....................... 3
A. Crystal structures....................... 3
B. Lattice parameters....................... 6
C. Electronic band structure................. 7
D. Mechanical properties.................... 12
E. Lattice dynamics........................ 15
F. Thermal properties...................... 18
1. Thermal-expansion coefficients.......... 18
2. Thermal conductivity.................. 19
3. Specific heat......................... 21
G. Electrical properties of undoped ZnO....... 22
1. Low-field transport................... 23
a兲
Electronic mail: uozgur@vcu.edu
b兲
Present address: Balikesir University, Faculty of Art and Science, Depart-
ment of Physics, 10100 Balikesir, Turkey.
c兲
Present address: Atatürk University, Faculty of Art and Science, Depart-
ment of Physics, 25240 Erzurum, Turkey.
d兲
Electronic mail: hmorkoc@vcu.edu
JOURNAL OF APPLIED PHYSICS 98, 041301 共2005兲
0021-8979/2005/98共4兲/041301/103/$22.50 © 2005 American Institute of Physics98, 041301-1
[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:
128.172.48.58 On: Mon, 19 Oct 2015 20:45:44

2. High-field transport................... 25
III. ZnO GROWTH............................. 26
A. Bulk growth........................... 26
B. Substrates............................. 28
C. rf magnetron sputtering.................. 29
D. Molecular-beam epitaxy.................. 30
E. Pulsed-laser deposition................... 34
F. Chemical-vapor deposition................ 36
IV. OPTICAL PROPERTIES OF ZnO.............. 39
A. Prelude............................... 39
B. Optical transitions in ZnO................ 39
1. Free excitons and polaritons............ 39
2. Bound excitons...................... 42
3. Two-electron satellites in PL............ 44
4. DAP and LO-phonon replicas in PL...... 45
5. Temperature-dependent PL measurements.. 45
C. Time-resolved PL on ZnO................ 47
D. Refractive index of ZnO................. 48
E. Stimulated emission in ZnO............... 51
1. Thin films........................... 51
2. Polycrystalline ZnO films and “random
lasers”.............................. 54
3. Multiple-quantum wells................ 56
4. Stimulated-emission dynamics.......... 56
V. DEFECTS IN ZnO........................... 57
A. Predictions from first principles............ 57
B. Experimental studies of native and
unintentionally introduced defects.......... 58
1. Shallow acceptor in ZnO............... 58
2. Green luminescence band.............. 59
3. Yellow luminescence band............. 61
4. Red luminescence band................ 62
VI. DOPING OF ZnO.......................... 62
A. n-type doping.......................... 62
B. p-type doping..........................
63
1. Nitrogen doping...................... 63
2. Codoping method: Nitrogen+group III... 65
3. Other dopants in group V.............. 67
VII. ZnO-BASED DILUTE MAGNETIC
SEMICONDUCTORS....................... 68
A. Theory of ZnO-based magnetic
semiconductors......................... 69
B. Experimental results on ZnO-based
magnetic semiconductors................. 72
VIII. BAND-GAP ENGINEERING................ 76
A. Mg
x
Zn
1−x
O alloy........................ 77
B. Cd
y
Zn
1−y
O alloy........................ 78
IX. PROCESSING, DEVICES, AND
HETEROSTRUCTURES..................... 79
A. Ohmic contacts to ZnO.................. 79
B. Schottky contacts to ZnO................. 80
C. Heterostructure devices.................. 82
1. Light-emitting devices................. 83
2. Photodiodes......................... 85
D. Metal-insulator-semiconductor diodes....... 86
E. Transparent thin-film transistors............ 87
X. ZnO NANOSTRUCTURES. .................. 88
XI. SUMMARY............................... 95
I. INTRODUCTION
There has been a great deal of interest in zinc oxide
共ZnO兲 semiconductor materials lately, as seen from a surge
of a relevant number of publications. The interest in ZnO is
fueled and fanned by its prospects in optoelectronics appli-
cations owing to its direct wide band gap 共E
g
⬃3.3 eV at
300 K兲. Some optoelectronic applications of ZnO overlap
with that of GaN, another wide-gap semiconductor 共E
g
⬃3.4 eV at 300 K兲 which is widely used for production of
green, blue-ultraviolet, and white light-emitting devices.
However, ZnO has some advantages over GaN among which
are the availability of fairly high-quality ZnO bulk single
crystals and a large exciton binding energy 共⬃60 meV兲. ZnO
also has much simpler crystal-growth technology, resulting
in a potentially lower cost for ZnO-based devices.
As indicated in the abstract, ZnO is not really a newly
discovered material. Research on ZnO has continued for
many decades with interest following a roller-coaster pattern.
Interest in this material at the time of this writing is again at
a high point. In terms of its characterization, reports go back
to 1935 or even earlier. For example, lattice parameters of
ZnO were investigated for many decades.
1–9
Similarly, opti-
cal properties and processes in ZnO as well as its refractive
index were extensively studied many decades ago.
10–25
Vi-
brational properties by techniques such as Raman scattering
were also determined early on.
26–32
Investigations of ZnO
properties presumes that ZnO samples were available.
Growth methods not much different from what is employed
lately have been explored, among which are chemical-vapor
transport,
33
vapor-phase growth,
34
hydrothermal growth
35
which also had the additional motivation of doping with Li
in an effort to obtain p-type material, high-quality platelets,
36
and so on.
37
The ZnO bulk crystals have been grown by a number of
methods, as has been reviewed recently,
38
and large-size
ZnO substrates are available.
39–41
High-quality ZnO films
can be grown at relatively low temperatures 共less than
700 °C兲. The large exciton binding energy of ⬃60 meV
paves the way for an intense near-band-edge excitonic emis-
sion at room and higher temperatures, because this value is
2.4 times that of the room-temperature 共RT兲 thermal energy
共k
B
T=25 meV兲. There have also been a number of reports on
laser emission from ZnO-based structures at RT and beyond.
It should be noted that besides the above-mentioned proper-
ties of ZnO, there are additional properties which make it
preferable over other wide-band-gap materials: its high-
energy radiation stability and amenability to wet chemical
etching.
38
Several experiments confirmed that ZnO is very
resistive to high-energy radiation,
42–44
making it a very suit-
able candidate for space applications. ZnO is easily etched in
all acids and alkalis, and this provides an opportunity for
fabrication of small-size devices. In addition, ZnO has the
same crystal structure and close lattice parameters to that of
GaN and as a result can be used as a substrate for epitaxial
growth of high-quality GaN films.
45,46
ZnO has recently found other niche applications as well,
such as fabrication of transparent thin-film transistors, where
the protective covering preventing light exposure is elimi-
041301-2 Ozgur et al. J. Appl. Phys. 98, 041301 共2005兲
[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:
128.172.48.58 On: Mon, 19 Oct 2015 20:45:44

nated since ZnO-based transistors are insensitive to visible
light. Also up to 2⫻ 10
21
cm
−3
charge carriers can be intro-
duced by heavy substitutional doping into ZnO. By control-
ling the doping level electrical properties can be changed
from insulator through n-type semiconductor to metal while
maintaining optical transparency that makes it useful for
transparent electrodes in flat-panel displays and solar cells.
ZnO is also a promising candidate for spintronics applica-
tions. Dietl et al.
47
predicted a Curie temperature of ⬎300 K
for Mn-doped ZnO. n-type doping in Fe-, Co-, or Ni-alloyed
ZnO was predicted to stabilize high-Curie-temperature ferro-
magnetism. There have been a number of publications which
appear to confirm these predictions, as has been reviewed
recently,
48,49
albeit with a good deal of controversy, as will
be discussed in Sec. VII.
However, one important problem should be overcome
before ZnO could potentially make inroads into the world of
optoelectronics devices: the growth of p-type-conductivity
ZnO crystals. Despite all the progress that has been made
and the reports of p-type conductivity in ZnO films using
various growth methods and various group-V dopant ele-
ments 共N, P, As, and Sb兲, a reliable and reproducible high-
quality p-type conductivity has not yet been achieved for
ZnO. Therefore, it remains to be the most pivotal topic in
ZnO research today, and congruently most of the research
efforts are directed just to solving this problem. In order to
overcome this bottleneck and to control the material’s prop-
erties, a clear understanding of physical processes in ZnO is
necessary in addition to obtaining low n-type background. In
spite of many decades of investigations, some of the basic
properties of ZnO still remain unclear. For example, the na-
ture of the residual n-type conductivity in undoped ZnO
films, whether being due to impurities of some native defect
or defects, is still under some degree of debate. Some authors
ascribe the residual background to intrinsic defects 关oxygen
vacancies 共V
O
兲 and interstitial zinc atoms 共Zn
i
兲兴, and others
to noncontrollable hydrogen impurities introduced during
growth. The well-known green band in ZnO luminescence
spectra 共manifesting itself as a broad peak around
500–530 nm兲, observed nearly in all samples regardless of
growth conditions, is related to singly ionized oxygen vacan-
cies by some and to residual copper impurities by others.
Simply, a requisite consensus is lacking.
While p-type ZnO is difficult to attain, the advantages of
ZnO are being explored and exploited by alternative methods
such as heteroepitaxy in which p-n heterostructures can be
obtained by depositing n-type ZnO films on other p-type
materials while still utilizing ZnO as the active layer.
Progress has been made in this arena with a number of het-
erostructures fabricated wherein one of the following, Si,
NiO, GaN, AlGaN, SiC, ZnTe, CuO, CdTe, etc., plays the
role of p-type layer. In particular, high-intensity UV emission
has been observed from the n-ZnO/p-AlGaN heterojunction
in which ZnO served as the active layer. These results are
just harbingers of what can be expected of ZnO in an effort
to position it for future device applications. As in the early
developments of GaN predating the demonstration of p-type
conductivity, metal-insulator-semiconductor device struc-
tures not requiring p-type ZnO have been introduced but lack
the high efficiency.
In this paper we collate the properties of ZnO as well as
review the recent progress in ZnO research. This present
review is distinguishable from the other reviews
50–55
in that
the previous ones focused mainly on material processing,
doping, and transport properties, while the present one treats
those topics in greater depth in addition to an in-depth dis-
cussion of the growth, optical properties, p-type doping, and
device fabrication aspects. The organization of this review is
as follows: First, structural, chemical, and electrical proper-
ties of undoped ZnO are discussed in Sec. II. This is fol-
lowed by ZnO crystal growth, both bulk and film 共Sec. III兲,
and optical properties of ZnO 共Sec. IV兲. Sections V–VII are
devoted, respectively, to defects in ZnO, doping, and mag-
netic properties. Alloys of ZnO 共band-gap engineering兲 are
discussed in Sec. VIII, and this is followed by ZnO-based
devices and their applications 共Sec. IX兲. Finally, ZnO nano-
structures are reviewed in Sec. X.
II. PROPERTIES OF ZnO
In this section crystal structures, inclusive of lattice pa-
rameters, electronic band structures, mechanical properties,
inclusive of elastic contants and piezoelectric constants, lat-
tice dynamics and vibrational processes, thermal properties,
electrical properties, and low-field and high-field carrier
transports are treated.
A. Crystal structures
Most of the group-II-VI binary compound semiconduc-
tors crystallize in either cubic zinc-blende or hexagonal
wurtzite structure where each anion is surrounded by four
cations at the corners of a tetrahedron, and vice versa. This
tetrahedral coordination is typical of sp
3
covalent bonding,
but these materials also have a substantial ionic character.
ZnO is a II-VI compound semiconductor whose ionicity re-
sides at the borderline between covalent and ionic semicon-
ductor. The crystal structures shared by ZnO are wurtzite
共B4 兲, zinc blende 共B3兲, and rocksalt 共B1兲, as schematically
shown in Fig. 1. At ambient conditions, the thermodynami-
cally stable phase is wurtzite. The zinc-blende ZnO structure
can be stabilized only by growth on cubic substrates, and the
rocksalt 共NaCl兲 structure may be obtained at relatively high
pressures.
The ground-state total energy of ZnO in wurtzite, zinc-
blende, and rocksalt structures has been calculated as a func-
tion of unit-cell volume using a first-principles periodic
Hartree-Fock 共HF兲 linear combination of atomic orbitals
共LCAO兲 theory by Jaffee and Hess.
56
The total-energy data
versus volume for the three phases are shown in Fig. 2 along
with the fits to the empirical functional form of the third-
order Murnaghan equation, which is used to calculate the
derived structural properties,
041301-3 Ozgur et al. J. Appl. Phys. 98, 041301 共2005兲
[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:
128.172.48.58 On: Mon, 19 Oct 2015 20:45:44

Frequently asked questions

Q1. What are the possible scattering mechanisms in a polar semiconductor?

If the density of dislocations and native defects are high in a semiconductor, dislocation scattering and scattering through defects are also considered as possible scattering mechanisms. 

Q2. What is the main obstacle in realizing long-life operation of optical and electrical devices?

It is well known that parasitic resistance, in the form of contact resistance, is one of the major obstacles in realizing long-lifetime operation of optical and electrical devices. 

Q3. What is the effect of codoping on the acceptor level in the band gap?

It is reported that codoping of N with the reactant dopants lowers the acceptor level in the band gap due to strong interaction between N acceptors and reactive donor codopants. 

Q4. What are the important electrical parameters in quantifying TFT performance?

The most important electrical parameters in quantifying TFT performance are the drain-current on-to-off ratio and active channel mobility. 

Q5. What is the effect of the electric field on the energy distribution of electrons?

At sufficiently low electric fields, the energy gained by the electrons from the applied electric field is small compared to the thermal energy of electrons, and therefore, the energy distribution of electrons is unaffected by such a low electric field. 

Q6. What is the chain of thought to achieve p-type ZnO through N doping?

One chain of thought is that to attain p-type ZnO through N doping it may be necessary to provide oxygen also to suppress oxygen vacancy VO. 

Q7. What are the technical issues that must be resolved?

For successful incorporation of spin into existing semiconductor technology several technical issues such as efficient injection, transport, manipulation, and detection of spin polarization as well as spinpolarized currents must be resolved. 

Q8. How can the growth mode of ZnO be monitored in real time?

With the feedback from reflection high-energy electron diffraction RHEED , the growth mode of ZnO epilayers can be monitored in real time dynamically. 

Q9. Why was the shallower donor ruled out in this work?

the shallower donor that was ruled out in this work due to its smaller concentration about one order of magnitude less than the deeper donor at an energy of 31 meV was further investigated later by the same group. 

Q10. What is the way to achieve the optimum alignment of nanowires?

By choosing the optimum match between the substrate lattice and the nanowires, the epitaxial orientation relationship between the nanowire and the substrate results in the aligned growth of nanowires normal to the substrate. 

Q11. Why is MgxZn1xO considered as a suitable material for barrier?

As discussed earlier, MgxZn1−xO alloy films have been considered as a suitable material for barrier layers due to its wider band gap than that of ZnO. 

Q12. How did they calculate the formation energies of native point defects and hydrogen in ZnO?

Kohan et al.359 and Van de Walle360 recently calculated formation energies and electronic structure of native point defects and hydrogen in ZnO by using the first-principles, plane-wave pseudopotential technique together with the supercell approach. 

Q13. How was the detection sensitivity of the nanowires influenced by the gate voltage?

detection sensitivity could be modulated by the gate voltage, and it was shown to increase with decreasing nanowire radius from 270 to 20 nm. 

Q14. What is the c lattice constant for the films grown under higher oxygen pressures?

for the films deposited under higher oxygen pressures 10−2–10−1 Torr , the c lattice constant was found to approach the bulk value. 

Q15. What is the fast decay constant for the as-received sample?

The fast decay constant 1 is smaller for the as-received sample 170.4 ps , and most probably represents the effective nonradiative recombination at room temperature.