One-dimensional II–VI nanostructures: Synthesis, properties and optoelectronic applications

Nano Today (2010) 5, 313—336

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/nanotoday

REVIEW

One-dimensional II—VI nanostructures: Synthesis, properties and optoelectronic applications Jiansheng Jie a,b, Wenjun Zhang a,∗, Igor Bello a, Chun-Sing Lee a, Shuit-Tong Lee a a

Center of Super-Diamond and Advanced Films (COSDAF) and Department of Physics and Materials Science, City University of Hong Kong, Hong Kong SAR, PR China b School of Electronic Science and Applied Physics, Hefei University of Technology, Hefei, Anhui 230009, PR China Received 11 May 2010; received in revised form 22 June 2010; accepted 28 June 2010 Available online 27 July 2010

KEYWORDS II—VI nanostructures; Synthesis; Doping; Optoelectronic devices

Summary The distinct properties of II—VI nanostructures have opened new opportunities for the applications of II—VI semiconductor materials in electronics and optoelectronics. Herein, we present a comprehensive review on the recent advances in the synthesis, properties and optoelectronic applications of one-dimensional II—VI nanostructures. In particular, the approaches to manipulate the electronic, optoelectronic, and transport properties of II—VI nanostructures by controlled doping and the latest progresses in fabricating high-performance II—VI nanoelectronic and nano-optoelectronic devices are discussed. © 2010 Elsevier Ltd. All rights reserved.

Introduction Discovery of carbon nanotubes (CNTs) in the early 1990s [1] has inspired a great interest in exploiting the promising potentials of one-dimensional (1D) semiconductor nanostructures in new electronic and optoelectronic device applications based on their specific geometries and distinct properties. Enormous efforts have been made to synthesize and characterize 1D nanostructures in the past two decades; and a host of nanostructures such as group IV, groups III—VI, groups II—VI, and oxide—semiconductor nanowires (NWs), nanoribbons (NRs)/nanobelts, and nanotubes (NTs), etc. have been successfully achieved. Among these nanostructures, 1D II—VI semiconductor nanostructures have shown

∗

Corresponding author. E-mail address: [email protected] (W. Zhang).

to be an important group with considerable progresses in the synthesis and utilization of their unique properties in extensive and novel applications [2]. We will start with a brief historical retrospect to the developments of II—VI materials, which may offer a conceptual understanding to the question why II—VI nanostructures have attracted such broad research interests. Wide bandgap II—VI semiconductors have been intensively studied for many years due to their great potentials for a variety of applications, in particular in the areas of electronic devices, phosphor, light-emitting and light-detecting devices, photovoltaic conversion (solar cells), X-ray and ␥-ray detection, etc. [3]. Nevertheless, the industrial application of II—VI compounds is still at its infancy, and a part of the problems is associated with the synthesis techniques and the lack of understanding and control over the properties of II—VI compounds. For instance, CdTe has a band-gap of 1.5 eV which was suggested to be the optimum band-gap for photovoltaic solar energy conversion [4]. The direct band-gap of CdTe also

1748-0132/$ — see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.nantod.2010.06.009

314 leads to a high optical absorption coefficient for photons in the whole solar energy spectrum (˛ > 105 cm−1 ), enabling the absorption of the most incident light in a relative thin CdTe layer of a few microns thick. As a result, the materials costs could be minimized and the requirement on the purity and crystallinity of CdTe films is not so strict due to the reduced diffusion length of photo-generated carriers. CdTe is a very promising material for high-efficiency thin film solar cells; and an efficiency up to 16.5%, against a theoretical maximum of 30%, has been indeed achieved on CdTe-based solar cells in labs [5,6]. In comparison, silicon has an indirect energy band-gap of ∼1.12 eV and a low absorption coefficient of ˛ ≈ 100 cm−1 . A thick silicon layer of several hundred microns (1/˛) is needed, and the electron—hole pairs generated have to diffuse up to that distance to reach the electric field in the depletion region to contribute to the photocurrent [7]. Based on above, CdTe is expected to be a superior material to silicon in photovoltaic applications. However, the relatively poor reproducibility and uniformity of CdTe films over large areas result in a large difference between lab efficiencies of ∼16.5% and the best module efficiencies of ∼10.7% for CdTe solar cells [5], which hinders the scale-up market of CdTe-based solar cells. II—VI compounds are also promising materials for optoelectronic applications, in particular for fabricating short-wavelength light-emitting diodes (LEDs) and laser diodes (LDs). Based on the achievement of p-type conduction via plasma-activated nitrogen doping, the first blue-green LD was developed from ZnSe in 1991 [8], which was about five years earlier than the report of the first GaNbased blue LD in 1996 [9]. The ZnSe LDs emitted coherent light at a wavelength of 490 nm, but it could work only at low temperature of 77 K with a short lifetime. Extensive efforts have been made to improve the output power and device lifetime of LDs. ZnCdSe continuous-wave (cw) LDs with a lifetime of about 100 h and LEDs with a lifetime up to 400 h under high current injection were achieved [10], and the use of ZnSe lasers in high-density CD players was also demonstrated [11]. However, some inherent problems such as poor electronic properties, low thermal conductivity, poor thermal stability, large Ohmic contact resistances, and low damage threshold have restricted ZnSe-based lasers to short lifetimes [12]. On the other hand, III-nitride industry has undergone an explosive development in the past decade since the demonstration of the first commercial blue GaN laser operating at 405 nm in 1998, which also resulted in a near halt to the then promising laser technology based on II—VI materials. Owing to their high sensitivity and high quantum efficiency, II—VI photodetectors are ideally suited for light detection ranging from near infrared to ultraviolet, however II—VI photodetectors have thus far limited applications in some simple devices such as photoswitches in street lamps and photoresistors in some cameras. The most common detectors in the visible range are still fabricated from silicon even though Si suffers from the obvious disadvantages such as indirect band-gap in the infrared range. In addition to the visible light detecting, II—VI compounds, e.g., CdTe and ZnTe, have shown more attractive potentials in nuclear radiation detection at room temperature [13]; and the ternary CdZnTe compound also possesses favorable properties for X-ray and ␥-ray detection, such as high atomic number (Z)

J. Jie et al. for efficient radiation—atomic interactions, high resistivity and low leakage current due to high purity and large bandgap, and high intrinsic  ( is the carrier mobility,  is the carrier lifetime) for efficient charge collection. These characteristics make CdZnTe an excellent candidate among a number of materials for high-energy radiation detection and high-performance spectrometers have been developed [14]. There are still more examples to illustrate the potentials of II—VI semiconductors in a variety of important applications. Nevertheless, difficulties in obtaining highquality and homogeneous II—VI semiconductor films/bulks with high reproducibility, stability, and reliability pose serious obstacles to their practical utilization. Electronic and optoelectronic applications require still higher quality II—VI semiconductors with controlled defects and impurities. Therefore, the obtainment of device-quality and singlecrystal II—VI semiconductor bulks/films is essential. The advances in nanoscience and nanotechnology open new opportunity for the application of II—VI semiconductor materials. Because the requirement on the substrates for growth of high-quality, single-crystalline 1D nanostructures is not as strict as for epitaxial thin film growth, nanomaterials with high crystalline perfection, reduced defects and controlled doping are easier to fabricate. The crystal quality of the 1D nanostructures can be significantly improved compared to their thin film/bulk counterparts. In addition, the 1D semiconductor nanostructures exhibit distinct electronic and optical properties arising from unique geometries and size-confinement effects [15]. For example, singlecrystalline wide band-gap 1D semiconductors can serve as nanoscale lasers, where the high refractive index contrast between the nanostructure and the surroundings defines a sub-wavelength-sized optical cavity [16,17]. They are also expected to function as waveguides for guiding and manipulating light on the sub-wavelength scale. The 1D nanostructures facilitate device fabrication since their length is usually in the range of micrometers, enabling processing and manipulation with the conventional technologies such as photolithography. The 1D system provides a fresh domain for basic physical studies as well as a new material system for next-generation electronic and optoelectronic devices [18]. Thus far, research related to II—VI nanomaterials has made great progress, ranging from synthesis of various nanostructures via different growth methods to applications in different fields. Since it is beyond the scope of this brief report to summarize all those previous works, we will restrict ourselves to first address the growth and doping of 1D II—VI nanomaterials, and then the recent advances in their electronic and optoelectronic applications. A comprehensive literature survey and bibliography are also included. On the other hand, we note that the oxide II—VI nanomaterials such as ZnO have already been extensively studied and reviewed [19]. Therefore, this review will be focused on the non-oxide II—VI nanomaterials instead.

Growth and doping of II—VI nanostructures General approaches to grow II—VI nanostructures The II—VI 1D semiconductor nanostructures are mostly synthesized by chemical vapor deposition (CVD), wet-

One-dimensional II—VI nanostructures chemical routes, and template-assistant methods [20]. Vapor—liquid—solid (VLS) growth mechanism has been well accepted for the synthesis of 1D nanostructures, in which usually a metal nanoparticle functions as a catalyst to direct 1D anisotropic growth [21,22]. During the VLS growth, the gas-phase reactants react with the metal nanoparticle at elevated temperatures to form a supersaturated melt, and the nucleation and subsequent axial elongation of a crystalline NW occur via reactants precipitation. The catalyst droplet then guides the NW’s growth direction and defines the diameter of the NWs. The VLS mechanism offers the capability to rationally fabricate the NWs via precise control of the catalyst type, size, and distribution. On the other hand, if the growth of the NWs in the longitudinal direction is accompanied by lateral growth via direct vapor—solid surface deposition, normally NRs are obtained instead of NWs. Vapor—solid (VS) mechanism is usually adopted to account for the growth of NRs while no catalyst is used or no catalyst head can be found at the end of the NRs. In particular, the VS mechanism is able to interpret the growth of a variety of oxide NRs such as ZnO, Ga2 O3 , SnO2 , PdO2 , etc. [23]. It seems that direct vaporization of the solid at a higher temperature and subsequent nucleation and anisotropic growth at lower temperature dominate the VS process. Beside VLS and VS mechanisms, oxide-assisted growth (OAG) method was also developed to grow Si NWs at first [24], and then extended to the growth of II—VI and III—VI semiconductor nanostructures [25], in which the oxide helps the nucleation and growth of 1D nanostructures. This method is particularly useful for the large-scale growth of 1D nanostructures without catalyst contamination. The crystallinity and controllability of the morphologies, structures, components of 1D nanostructures are crucial to their applications in electronic and optoelectronic devices. 1D nanostructures grown by CVD techniques usually have better crystal quality than those via wet-chemical method (the residual solvent may affect the quality) owing to the growth process at high temperature. In addition, the CVD growth allows the precise control of the components of the product by adjusting the compositions of the vapor, facilitating the doping process and the growth of nanostructures with homo-/hetero-junctions or superlattice structures. An additional merit of the CVD methods is that they are mature and sophisticated techniques widely used in the present semiconductor industries to grow thin films such as SiO2 , Si3 N4 , and GaN, which facilitates the integration of nanostructures with thin films for future applications. We will discuss in details the achievements in the growth and doping for some II—VI nanostructures with important application potentials below, and the emphasis will be placed on the CVD growth methods.

Binary II—VI nanostructures ZnS ZnS NWs were normally synthesized by thermal evaporation [26]. Meng et al. reported the ultrafine ZnS NWs (diameter 10—20 nm) with a sphalerite structure; the phase of the ZnS NWs was controlled by the deposition temperature, and the diameter of NWs via the thickness of the gold film [27]. Ding et al. grew high-density ZnS NWs on anodic alumina oxide (AAO) templates, and the photoluminescence

315 spectra of the NWs showed stimulated emission as well as narrow resonant cavity modes [28]. Growth of ordered ZnS NW arrays was achieved by using various crystal substrates, including CdSe [29], ZnS [30], and Zn3 P2 [31]. Using twostep vapor deposition process, Jiang et al. demonstrated that single-crystal ZnS NRs could be promising substrates for homoepitaxial growth of ZnS NW and NR arrays [32]. As shown in Fig. 1, well-aligned ZnS NW arrays with high density were grown on both the top and side surfaces of the ZnS NR substrates. Interestingly, the cross-arrays of NWs could function as Fabry—Perot cavities, which leads to bandgap lasing emission in two specific directions. More intense green emission from the <2 1 0> ZnS NWs cross-arrays than the [0 0 1] ones was observed in CL measurements, revealing the optical properties of the ZnS NWs are orientation dependent. This method was also employed to grow ordered ZnS NW arrays on CdS NRs [33], implying that it might be a general method for homoepitaxial and heteroepitaxial NW arrays growth. Catalyst-assisted post-annealing treatment was also used to obtain nanocantilever arrays on the edges of the ZnS NRs [34]. Synthesis of ZnS NRs was realized by using various thermal evaporation techniques [35,36]. The high-temperature synthesized ZnS NRs usually have perfectly rectangular, single-crystal 2H structure, as shown in Fig. 2. The growth direction of the ZnS NR in Fig. 2(a) is [1 2 0], different from the usual ZnS NRs grown in [0 0 1] orientation [37]. The high-quality facets with a refraction index larger than air can serve as reflecting mirrors. The rectangular ZnS NRs thus can form potential optical waveguides/resonant cavities for lasing emission. In 2003, lasing in a single ZnS NR was reported by Lee’s group [38]. Measurements on single ZnS NR showed that NRs serve as excellent optical cavities and gain medium with high lasing modes free of PL background even for a low pumping power density of 9 kW/cm2 (Fig. 3). Besides ZnS, lasing emission was also observed in other II—VI nanostructures such as CdS NRs [39]. Moreover, the emission wavelength of lasing could be tuned from near infrared to ultraviolet by using ternary alloy including Znx Cd1−x S and CdS1−x Sex [15,40,41]. These results highlight the promising applications of groups II—VI nanostructures as the building block for room-temperature high-performance optoelectronic nanodevices. Periodically polytypic and twinned structures were often observed in ZnS NWs and NRs [36,42]. The formation of these structures can be attributed to modulating mass diffusion procedures in the catalyst droplet or on the NW side. It was also demonstrated that the strain caused by impurity incorporation and plasma bombardment was an important factor responsible for the formation of heterocrystal and bicrystal structures in ZnS NWs [43]. The incorporation of Si impurity in ZnS could even result in the growth of ZnS NRs with novel dart-shape [44]. By combining ZnS with other semiconductors, various heterojunction structures were formed, and these structures are of interest to optoelectronic applications. For example, ZnS/ZnO heterojunction NRs were synthesized by feeding a small amount of O2 during ZnS NR growth [45], and X-ray excited optical luminescence (XEOL) study revealed that the luminescence from ZnO/ZnS NRs was dominated by the ZnO component of the NRs [46]. Other heterojunction structures that have been investigated include CdS/ZnS, CdSe/ZnS, Zn3 P2 /ZnS, SiO2 /ZnS and so on.

316

J. Jie et al.

Figure 1 (a) Scanning electron microscope (SEM) overall views of the nanostructure arrays grown on the top and side surfaces of the substrate nanoribbons at 800—850 ◦ C. (b) Enlarged view of the nanowire cross-array formed on the top surface. The nanowires grow along two distinct directions at 60◦ with respect to each other. (c) Schematic diagram showing the alignment of nanowires grown on different surfaces of the substrate. (d) The scheme of the basic cell of hexagonal structure showing the crystallographic relations of the lattice planes and directions involved in epitaxial growth. (e) Molecular structure model showing the orientation of the cross-array nanowires with respect to the substrate. Reprinted with permission from Ref. [32].

Doping with metal elements has been used to control the optical and electrical properties of II—VI nanostructures. In particular, doping by transition-metal ions is of special interest owing to the nanoscale spintronic applications. So far, various metal elements such as Mn, Cu, Fe, Co, and Ga have been selected as the dopants for ZnS nanostructures. Among them, Mn-doped ZnS nanostructures were widely studied due to their potential applications in both phosphors and spinelectronics. For instance, Radovanovic et al. synthesized Mn-doped II—VI and III—V semiconductor NWs based on metal nanocluster-catalyzed CVD route [47]. Li et al. reported Mn doping of ZnS NRs by post-annealing; and the NRs emitted a new PL peak at 585 nm after doping, which was attributed to Mn-related defects [48]. Kang et al. demonstrated the Mn/Fe doped and Co-doped ZnS by using metal chloride as a metal carrier via a chemical vapor transport method [49]. Magnetic property measurements showed that the doped nanostructures have room-temperature ferromagnetism [49]. On the other hand, Cheng and Wang reported Eu2+ -doped wurtzite ZnS NWs with long-lasting phosphorescence [50]. Lu et al. synthesized Ga-doped ZnS nanowalls with better photo-catalytic and photoresponse properties than the film counterpart [51]. In spite of these progresses, doping in ZnS nanostructures with high controllability and reproducibility remains a challenge due to the complexity of the growth process. Both n- and p-type dop-

ing are essential to device applications, nevertheless very few reports dealt with p-type doping. Recently, Yuan et al. reported the synthesis of p-type ZnS NRs by using NH3 as the dopant source and a post-annealing process was shown to be required to activate N acceptors [52]. Control of the electrical and photoelectrical properties of ZnS nanostructures via doping is still far from satisfactory. ZnSe ZnSe is regarded as one of the most important II—VI semiconductors due to its numerous applications in optoelectronic devices. Actually, the first blue-green laser diode was developed based on ZnSe in 1991 [8]. The synthesis and properties of ZnSe nanostructures have been intensively studied in the past decade. Wang and Chan et al. reported the epitaxial growth of ZnSe NWs by molecular beam epitaxy (MBE) on GaAs and GaP substrates [53]. The ultrathin NWs showed diameter dependence of growth rates, which was quite different from classical VLS or other growth models. It was found later that surface incorporation and diffusion played an important role in controlling the diameter-dependent growth [54]. The relationship of growth direction with NW diameter and growth temperature was also revealed. It was suggested that NW growth direction was controlled by the liquid—solid interface structure at the catalytic tips of NWs [55]. Zhang et al. showed the growth of oriented arrays of

One-dimensional II—VI nanostructures

317

Figure 2 Transmission electron microscopic (TEM) characterization of ZnS nanoribbons: (a) low-resolution TEM image. Note the transparency indicating a thin nanoribbon (100 nm and below). (b) Selected area electron diffraction (SAED) pattern of the [0 0 1] zone axis. (c) Energy dispersive spectroscopy (EDS) indicating a pure stoichiometric ZnS nanoribbon. (d) Schematic diagram of the nanoribbon crystallographic directions. (e) TEM cross-sectional images of two nanoribbons. (f) SAED pattern of the [1 2 0] zone axis. High-resolution TEM image of (g) the lower side, (h) the right-hand side, (i) the lower right-hand side corner of the cross-section of the short nanoribbon in (e). Reprinted with permission from Ref. [38].

ZnSe NWs in zinc-blende structure by metal—organic chemical vapor deposition (MOCVD) on ZnSe epilayers on selected GaAs surfaces [56], and the optical properties of the ZnSe NWs showed a dependence on the growth pressure [57]. Moreover, the solvothermal method was employed to synthesize ZnSe nanosheets, NWs, and NRs arrays [58]. Selfcatalytic growth of ZnSe nanorod arrays on ZnSe grains were investigated by Choy et al. [59], who also revealed the formation of ZnSe nanorings due to the electrostatic polar charge [60]. Femtosecond laser ablation of ZnSe crystal in air led to the growth of ZnSe NWs on the ablation crater on crystal surface [61]. Panda et al. revealed the spontaneous assembly of ultranarrow ZnSe nanorods and NWs into highly ordered two-dimensional (2D) supercrystals [62], and the polarization properties of the NWs were investigated

[63]. Jiang et al. demonstrated the growth of ZnSe NRs with wurtzite-2H single-crystal structure by using laser ablation assisted vapor growth method [64], whereas ZnSe NRs in zinc-blende structure was obtained by Zhang et al. by using MOCVD [65]. The optical properties of ZnSe nanostructures were intensively studied, which showed high sensitivity to the crystal structures, surfaces, defects, and components of the nanostructures [66]. Li et al. reported the growth of size-dependent periodically twinned ZnSe NWs, and suggested that the periodic insertion of stacking faults into the lattice was responsible for the twin formation [67]. Wire- and ribbon-like nanocables composed of periodically twinned ZnSe core and amorphous SiO2 shell were investigated by Fan et al. [68]. It was proposed that the periodic twinning was mainly

318

J. Jie et al. [72]. Interestingly, it was found that CdSe quantum dots in ZnSe could function as efficient source for single photons [73]. As compared with the research on material synthesis and structure characterizations, doping in ZnSe 1D nanostructures has received relatively less attention so far. Mn, Ag, and Cu doping were used to modulate the optical and structural properties of ZnSe NWs [74—76]. The Mn-doped ZnSe NWs showed an Mn2+ emission at 590 nm and the photoluminescence (PL) quantum yield of the doped NWs was as high as 40% after passivation with a thin CdSe shell [74]. Recently, Song et al. reported p-type doping in ZnSe NWs by using Zn3 As2 as the dopant source [77], which is an important step for developing ZnSe-based nanoelectronic devices.

Figure 3 (a) PL of a single ZnS nanoribbon excited by a 266 nm laser beam with different power densities. Left inset in (a): the single ZnS nanoribbon dispersed on a TEM grid used for the measurements. Right inset in (a): PL intensity vs. input power density. Lasing starts at ∼60 kW/cm2 . PL acceptance angle was ∼33◦ . (b) PL obtained with an acceptance angle of ∼2◦ and an angle of ∼5◦ between the optical fiber and the plane of the TEM grid. The optical fiber is directed toward the long axis of the ZnS nanoribbon. The modes are very cleanly resolved with no broad PL contribution. The full width at half maximum (FWHM) of the modes is <0.1 nm. Left inset in (b): an individual emission pulse indicating sharp modes with ∼0.4 nm spacing. PL acceptance angle was 33◦ . Right inset in (b): the PL intensity vs. the detection angle with respect to the nanoribbon long axis. PL acceptance angle was 33◦ . Reprinted with permission from Ref. [38].

attributed to the effects of periodical surface compressive stress, while periodic change of ZnSe concentration in the catalyst droplet could also facilitate the formation of twinned NWs. As for the composite nanostructure, atomic layer deposition (ALD) was employed to grow NWs composed of ZnSe/CdSe superlattices [69]. Wang et al. showed the growth of ZnO/ZnSe core/shell NW arrays on transparent conducting oxide substrates, which had potential applications in solar cells [70]. The formation of side-byside ZnSe/ZnCdSe bicrystalline NRs was accomplished by a two-step process employing MOCVD [71], and ZnSe/Si bicoaxial NWs were formed by one-step thermal evaporation

CdS CdS NWs have been synthesized using various methods, including template-assisted electrochemical deposition [78], laser ablation [79], AAO template [80], and solvothermal [81]. Aligned CdS NW arrays were grown on Cd foils by solvothermal process [82] or liquid reaction route [83]. As for CdS NRs, since Dong et al. synthesized CdS NRs on tungsten substrates by using CdS powder as the source material [84], multifarious methods have been developed to grow CdS NRs. Duan et al. have synthesized CdS NRs by vapor transport method [85]. Rapid growth method, either by rapid heating or microwave irradiation, was adopted to grow CdS NRs in a shorter time [86]. Liu et al. showed the growth of high-quality CdS NRs capable of forming lasing cavities by a thermal evaporation process, and PL measurements revealed a threshold value of ∼35 kW/cm2 for the laser emission upon 266 nm light pumping [89]. Stimulated emission from individual CdS NR through the cooperative emission of the 2LO phonon-bound excitons was observed by Pan et al. [87]. Lieber and co-workers demonstrated that CdS NWs can work as effective waveguides in the sub-wavelength scale through the quantitative detection of optical intensity at different propagation distances [88]. Pan et al. investigated waveguide in CdS NRs; they found that the optical losses in NRs was larger than that in NWs, and the emission bands showed a significant red-shift with respect to increasing traveling distances [89]. Fan et al. reported the synthesis of bicrystalline CdS NRs by using thermal evaporation [90]. Coaxial CdS/ZnS NWs was fabricated by a one-step MOCVD process [91], and coaxial CdS/Si NWs could be formed by vapor transport [92]. Mn-doped CdS nanorods were synthesized by using both high-temperature solution phase chemistry and vapor transport method [93,94]. Magnetic measurements demonstrated the robust room-temperature ferromagnetism of CdS:Mn nanorods. Mg, Zn, and Hg doping in CdS NWs was conducted, and it was found that the NW emission was controlled by the introduction of dopants [95]. On the other hand, Ma et al. reported the synthesis of In-doped n-type CdS NRs via a CVD method and field-effect transistors (FETs) and nanodiodes were successfully fabricated based on the n-CdS NRs [96]. CdSe Solution—liquid—solid (SLS) method has been widely used to grow CdSe quantum NWs [97]. The wet-chemical SLS approach has utilized low-melting-point nanocatalysts, usu-

One-dimensional II—VI nanostructures ally bismuth (Bi), and is based on well-established methods used to prepare semiconductor nanocrystals [98]. CdSe NW arrays could be grown in AAO templates by electrochemical or photochemical methods [99]. NWs prepared via this method were mainly polycrystalline in structure consisting of a large amount of nano-grains, while single-crystal CdSe NWs were also reported recently [100]. On the other hand, single-crystalline CdSe nanostructures could also be realized by thermal evaporation of CdSe powder [101]. Duan et al. obtained CdSe NWs via a laser-catalyst growth (LCG) approach [102]. Shan et al. synthesized oriented CdSe nanoneedles by MOCVD [103]. Wang et al. showed that the morphologies of CdSe nanostructures could be tuned by adding Si in the starting materials in thermal evaporation process [104]. Venugopal et al. investigated the optical properties of CdSe NRs that were grown by a CVD method assisted with laser ablation, and revealed a lattice contraction in the NRs and a partially polarized dependence in PL measurements [105]. Giblin et al. further studied the excitation polarization anisotropy of single CdSe NWs [106]. CdS/CdSe nanorod heterostructures were realized by employing a SLS mechanism with the assistance of bismuth nanocrystals [107]. Si/CdSe heterostructures were obtained by one-step metal-catalyzed thermal evaporation method and their time-resolved X-ray-excited optical luminescence was characterized [108,109]. Dai et al. showed the growth of hierarchical CdSe/CdS nanostructures by a two-step thermal evaporation method [110]. However, doping in CdSe nanostructures remains problematic and very few reports concerned about this issue, while the doped CdSe nanos-

319 tructures do show interesting properties. For instance, giant Zeeman splitting has been investigated in Mn2+ -doped CdSe quantum NRs [111]. Recently, He et al. demonstrated for the first time the tuning of electrical and photoelectrical properties of CdSe NWs by indium doping [112]. Two methods have been applied in this study, including in situ doping by co-evaporating the dopant (indium) and CdSe powder during NW growth (Fig. 4), and post-growth doping via a thermal diffusion process. Both methods were shown to be effective in tuning the electronic properties of CdSe NWs over a wide range. The conductivity of CdSe NWs was increased by nearly five orders of magnitude from ∼10−4 to tens of S cm−1 via doping, and carrier concentration as high as ∼1019 cm−3 were achieved for the heaviest doped NWs.

ZnTe ZnTe has a direct band gap of 2.26 eV and a Bohr exciton radius of 6.2 nm [113]. It has been suggested as a useful material in optoelectronic and thermoelectric devices, such as the first unit in a tandem solar cell, a buffer layer for an HgCdTe infrared detector, or a part of the graded p-Zn(Te)Se multiquantum-well structure in a blue-green laser diode [114]. Compared with other II—VI semiconductors, reports on ZnTe and CdTe nanostructures are relatively scarce. Li et al. reported the synthesis of ZnTe nanorods by a solvothermal process [115]. Single-crystalline ZnTe NW arrays in AAO template were obtained by pulsed electrochemical deposition technique [116]. Wang et al. reported the synthesis of ZnTe NWs by using the SLS method, in which Bi nanoparticles

Figure 4 (a) Schematic illustration of the experimental set-up for co-evaporation of In and CdSe sources in a three-zone furnace. (b) Single intrinsic NW (denoted as ICS), with a typical SEM image of single NW I—V measurement shown in the inset. (c) Four kinds of doped single NWs with different doping levels. (d) Distribution of conductivity values for 100 two-terminated devices based on single CdSe NW, 20 devices each for intrinsic CdSe NWs (ICS) and the samples synthesized at different In source temperature of 380 ◦ C (CIS 380), 500 ◦ C (CIS 500), 700 ◦ C (CIS 700), and 800 ◦ C (CIS 800). Reprinted with permission from Ref. [112].

320

J. Jie et al.

were used to seed the 1D growth of ZnTe [117]. Other alterative solution methods were also used to fabricate ZnTe NWs [118]. Periodically twinned ZnTe NWs were synthesized by thermal evaporation [119]. Janik et al. and Kirmse et al. prepared ZnTe and axial ZnTe/CdTe NWs on GaAs substrates by Au-catalytic MBE method [120,121]. On the other hand, ZnTe NRs were obtained by a chemical process, and the p-type nature of the ZnTe NRs was confirmed by electrical transport measurements [122]. The p-type characteristic of the unintentionally doped ZnTe NRs (also for ZnTe in other forms) was suggested to originate from the acceptor defects such as Zn vacancies in the materials [122]. Cao et al. demonstrated the one-step growth of coaxial p-ZnTe/SiOx nanocables by thermal evaporation [123]. The SiOx shell was insulative and could sever directly as a dielectric layer, facilitating the integration of ZnTe NWs in FETs and chemical and biological sensors. The p-type conduction of ZnTe NWs could be further enhanced by Cu+ doping in solution [124], and the Ni/Au multilayer electrodes showed good ohmic contact with ZnTe NWs [125]. However, reliable n-type doping for ZnTe nanostructures remains a problem, and there is no report thus far. CdTe CdTe has important applications in diverse fields such as LEDs, solar cells, photonic crystals, and bioimaging [126]. CdTe NWs have been grown from solution by several groups [127,128]. Besides the solution methods, Neretina et al. prepared vertically aligned wurtzite CdTe NWs on (0 0 0 1) sapphire substrate by pulsed laser deposition (PLD) [129]. The use of bismuth catalyst and polyvinyl alcohol (PVA) mediator on substrates has been shown to be critical to the nucleation and growth of CdTe NWs [129]. Park et al. reported the synthesis of CdTe NWs by the gas-phase transformation of the pre-grown ZnTe NWs using Cd vapor transport [126]. Similar to ZnTe, the unintentionally doped CdTe usually exhibits p-type characteristics due to the existence of Cd vacancies. Nevertheless, there are only few reports on the control of the electronic and optoelectronic properties of CdTe nanostructures via intentional doping [130].

Ternary II—VI nanostructures The capability of tuning the emission wavelength is important for optoelectronic applications. It is known that II—VI materials can form a series of solid solutions [131], allowing continuous variation of the band-gap and subsequently the emission wavelength by adjusting the composition. The growth and color tunable photoluminescence from ternary CdSx Se1−x NWs, NRs and nanosheeets have been reported [132—134], and stimulated emission from CdSx Se1−x NRs by optical excitation has been observed [41,135]. Lee and co-workers reported the synthesis and lasing properties of a variety of ternary II—VI nanostructures [15,41,42]. They found that lasing of II—VI ternary NRs at room temperature could cover the entire NIR—vis—UV spectra range (Fig. 5) [42]. Photoluminescence measurements of CdSx Se1−x NRs and Zny Cd1−y S revealed lasing with tunable wavelength from 1.75 eV (band-gap of CdSe) via 2.43 eV (band-gap of CdS) to 3.65 eV (band-gap of ZnS). Moreover, copper-doping was

Figure 5 (a) SEM image of CdSx Se1−x NRs. The inset presents a typical EDS spectrum from a composition x ∼0.9 (the Cu signal is from the TEM grid). (b) X-ray diffraction (XRD) patterns of CdSx Se1−x NRs; composition x calculated from the XRD data is indicated. (c) Lasing from NIR to UV (710—340 nm) from two ternary II—VI NRs Zny CdS1−y (spectra A—C) and CdSx Se1−x (spectra D—F). Reprinted with permission from Ref. [41].

One-dimensional II—VI nanostructures Table 1

321

Current status of the research on doping in one-dimensional II—VI nanostructures. n-type doping Al

ZnS ZnSe CdS CdSe ZnTe CdTe

Ga

p-type doping In

[51]

Cl

Br

I

N

P

Transition metal doping

As

Sb

Ag

Li

Na

Cu

[52] [77]

[76]

[75]

[96] [112]

employed to further tailor the electrical and optical properties of Zny Cd1−y S NRs, and the Cu-doped NRs showed four orders of magnitude larger photocurrent than the undoped ones [136].

n- and p-type doping in II—VI nanostructures For the electronic and optoelectronic applications of II—VI nanostructures, controlled doping to achieve bipolar electrical conduction, i.e. efficient n- and p-doping, is essential. Nevertheless, research on doping, electronic and optoelectronic properties of 1D II—VI nanostructures is still in the initial stage. Table 1 summaries the present status of the doping research in 1D II—VI nanostructures. The possible doping elements for realizing efficient n-type, p-type, and magnetic doping are listed. Surprisingly, the reports on II—VI nanostructures doping are quite few, particularly for n- and p-type doping. The device application of wide band-gap semiconductors requires efficient doping both with donors and acceptors up to high concentrations of >1018 /cm3 , desirably above 1019 /cm3 . However, it is difficult to achieve both n- and ptype doping in such a high doping concentration for wide band-gap semiconductors, particularly for II—VI materials. Normally they exhibit strong unipolar electrical conductivity, i.e., n-type conductivity for ZnS, ZnSe, CdS, CdSe and p-type conductivity for ZnTe and CdTe. In II—VI films, the efficient doping is mainly limited by the self-compensation by the native defects such as vacancies, interstitials, and antisites, etc. [3]. It is difficult to conclude that whether n- and p-type doping would be more or less efficient in II—VI nanostructures due to the absence of full survey on the doping behaviors of II—VI nanostructures. However, we note that n-type doping in CdSe NWs using In dopant has led to a high electron concentration as high as ∼1019 /cm3 [112], whereas p-type doping in ZnS using N dopant to a hole concentration of ∼1018 /cm3 [52], implying that better doping efficiency might be achieved in II—VI nanostructures. Compared with II—VI films, II—VI nanostructures have the following merits: (1) single-crystal nanostructures with fewer defects are easier to be obtained; (2) nanostructures are able to sustain larger lattice strain due to the nanoscale size. As a result, higher doping concentration can be achieved without inducing the defects such as layer defaults and dislocations; and (3) in contrast to bulk doping, the large surface-to-volume ratio of nanostructures allows controlling the electronic properties by surface charge transfer doping

Mn [47—49] [74] [93,94] [111]

[124]

[137,138], thus overcoming the difficulties in obtaining high bulk doping concentration.

Electronic applications of II—VI nanostructures A variety of nanoelectronic devices such as FETs, logic circuits, Schottky diodes have been constructed based on semiconductor nanostructures. Owing to the superior crystallinity and size-confined transport properties of the nanostructures, the relevant nanodevices have shown significantly improved performance with respect to their thin film counterparts. In particular, due to the distinct electronic structures of II—VI semiconductors, II—VI nanostructures are expected to have promising applications in new electronic and optoelectronic nanodevices which are thus far not possible with II—VI thin films.

Field-effect transistors (FETs) based on II—VI nanostructures Metal—oxide—semiconductor FETs (nano-MOSFETs) FETs are the basic elements in many devices such as integrated circuits, flat panel displays, and data storage due to their high yield and capability to scale down in dimensions. Among different structures of FETs, metal—oxide—semiconductor FETs (MOSFETs) are most widely used and play a core role in current IC techniques. A normal MOSFET is composed of five major elements: conduction channel, source electrode, drain electrode, gate oxide, and gate electrode. MOSFETs arise from silicon IC techniques, and have now been applied to various semiconductor materials. The unique geometry of 1D nanostructures makes them a nature candidate for the conduction channels in MOSFETs (nano-MOSFETs). Fig. 6(a)—(c) shows the structure of a nano-MOSFET fabricated from single CdSe NR [139]. The device was constructed on a SiO2 /p+ -Si substrate in a globe back-gate configuration. 300-nm thick SiO2 and degenerately doped Si substrate served as the gate dielectric and the back-gate electrode, respectively. In a typical fabrication process, the singlecrystal CdSe NRs were dispersed on a SiO2 /p+ -Si substrate at a desired density, and then electrodes were deposited on the two ends of the NRs by e-beam evaporation using a shadow mask or by photolithography and lift-off processes. The gate-dependent current (Ids ) vs. voltage (Vds ) curves measured on the CdSe NRs are shown in Fig. 6(d). The device shows a pronounced gating effect. As the gate

322

J. Jie et al.

Figure 6 (a) SEM image of CdSe single NR FET device. (b) HRTEM of the CdSe NR. The NR has a growth direction of [109]. (c) Schematic diagram of FET and photoconductive measurements. (d) Ids vs. Vds curves (a—f) for different Vg that range from 0 V to 30 V. (e) Ids vs. Vg plots for different Vds . The linear regimes of the Ids —Vg curves are fitted by the dash lines. The inset shows the logarithmic plot of the transfer characteristics at Vds = 0.8 V. Reprinted with permission from Ref. [139].

voltage (Vg ) increases, the conductance of the NR increases correspondingly, which is typical of ‘‘normally off’’ MOFETs with an n-channel. The n-type conductivity of the CdSe NRs is most likely attributed to the selenium vacancies in the CdSe NRs. From the transfer characteristics (Ids —Vg ) of the device (Fig. 6(e)), the mean threshold voltage (VT ), which is deduced from the linearly extrapolated value at the Vg axis (dash lines in Fig. 6(e)), is 20.9 V for this device. In addition, the electron mobility (n ) can be estimated from the channel transconductance (gm ) of the FET. In the linear regime (Vg from 25 V to 30 V), gm =(∂ Ids /∂ Vg ) = (Z/L)n C0 Vds , where Z/L is the width-to-length ratio of the channel, and C0 is the gate capacitance per unit area. For the SiO2 dielectric layer, C0 = εε0 /h, where ε is the dielectric constant of SiO2 (3.9), ε0 is the permittivity of free space, and h is the thickness of SiO2 . Based on the above equations, gm = 21 nS and n = 9.6 cm2 /(V s) are obtained. The on—off current ratio (Ion /Ioff ) >104 can also be deduced from the logarithmic plot of the Ids —Vg curves. Light irradiation with energy above the band-gap of II—VI nanostructures results in the generation of electron—hole pairs in the nanostructures, leading to the increase of conductivity. Therefore, the device characteristics of the II—VI nano-MOSFETs are remarkably influenced by the incident light. Upon light irradiation, VT shifts to negative direc-

tion along with the decrease of on—off current ratio due to the increase of carrier concentration in the nanostructure channel [139]. Moreover, the transport properties of the nanostructures are also affected by the surface absorption and desorption of atmosphere gas. In air, O2 molecules adsorbed on n-type II—VI nanostructures trap electrons via the chemisorption of O (g) + e → O− (ad) [140]. As a result, a depletion layer is formed near the surface and the conductivity decreases. On the contrary, oxygen molecules are desorbed from the NR surface in vacuum, leading to the increase of carrier concentration and decrease of carrier surface scattering. Correspondingly, VT will decrease and n will increase for the device. The high sensitivity of the nano-MOSFETs to ambient gas originates from the high surface-to-volume ratio of semiconductor nanostructures. Surface-sensitive properties of nanostructures form the basis of high-sensitivity gas, chemical and biological detectors. However, such high surface sensitivity also presents a serious challenge to device fabrication as it leads to difficulty in the reliability and reproducibility of device performances. Therefore, proper encapsulation and surface modification are in particular important for fabricating nanostructure-based devices with controlled performance. Although the nano-MOSFETs based on single CdSe and CdS NRs show obvious gating effect, the devices constructed

One-dimensional II—VI nanostructures from intrinsic (undoped) II—VI nanostructures usually possess poor device performance, such as high work voltage, small transconductance, low carrier mobility, and small on—off ratio [139—141], which deviate from the expectation for the nanostructures due to their superior properties. By analyzing these devices, the following factors are considered to cause the poor performance of the devices: (1) low conductivity of the undoped II—VI nanostructures. Intrinsic II—VI nanostructures have only limited carrier concentration, which results in the low conduction current and as well hinders the formation of Ohmic contacts between electrodes and nanostructures. (2) Poor electrode contact. Ohmic contacts between source/drain electrodes and the nanostructures are essential for reducing the contact resistance and increasing the channel current. However, Schottky contact instead of Ohmic contact is usually formed due to the mismatch of work functions and the surface/interface defects. In particular, Ohmic contacts are difficult to achieve for wide band-gap II—VI nanostructures such as ZnS and ZnSe NWs/NRs because of the serious surface Fermi energy level pinning. (3) Thick oxide dielectrics. Large thickness and small dielectric constant of SiO2 dielectric layer result in small channel capacitance, which weakens the coupling of gate electrode to the channel and leads to increased work voltage and decreased transconductance. (4) Backgate device configuration. Such a device structure has weak channel coupling which limits the device performance. Moreover, the fully overlapping of back-gate electrode with source/drain electrodes induces a large parasitic capacitance and consequently reduces the switching speed of the nano-MOSFETs. On the other hand, the globe back-gate configuration is also unfavorable for the device integration.

High-performance nano-MOSFETs with high-␬ dielectrics and top-gate geometry Based on the above discussion, it is obvious that all components of nano-MOSFETs, such as channel materials, contact electrodes, gate dielectrics, and device structures should be systematically optimized to achieve high-performance devices. Firstly, the II—VI nanostructures should be doped appropriately to achieve a high channel conduction and low contact resistance with electrodes [97]. In this case, even Schottky barrier is formed between nanostructures and electrodes, it is thinner and lower, and carriers could pass through the barrier easier by tunneling effect. As the series resistance decreases, the work current of nano-MOSFETs can be enhanced [142,143]. However, it should also be note that heavily doping would induce metallic-like semiconductor characteristics of nanostructures; the utilization of such channel would abate the gate effect. The selection of a proper dielectric material and optimization of device structure are crucial for improving the device performance [144,145]. For instance, by using the high-␬ gate insulators, the gate voltage can modulate the channel conductance more effectively even with a relatively thick dielectric, as a result, the work voltage and the directtunneling leakage current can be drastically decreased. On the other hand, top-gate nano-FETs have also been studied, and the top-gate configuration has demonstrated many advantages over the conventional global back-gate devices, including localized gate biasing with low voltage, high-speed

323 switching, and high integration density [146,147]. The topgate dielectrics can fully sheath the NWs/NRs and provide enhanced coupling to the conduction channel [148—150]. Thus far, omega-shaped gate and organic dielectrics gate have also been employed to improve the device performance [151,152]. Fig. 7 shows top-gate nano-MOSFETs built with CdS:P NRs [153]. Both back-gate and top-gate nano-FETs are constructed for comparison. In contrast to the back-gate devices, the top-gate nano-MOSFETs with high-␬ HfO2 dielectric (ε ∼ 25) exhibit a substantially improved performance: the work voltage reduced from about ±25 V to ±5 V, the subthreshold swing (S) decreased from >25 V/dec to 200 mV/dec, the transconductance enhanced from 7.3 nS to 0.87 ␮S, and the Ion /Ioff ratio remarkably increased from ∼10 to ∼107 . Wu et al. have also reported the fabrication of CdS NR MOSFETs with small subthreshold swing of 62 mV/dec and high on/off ratio of 2 × 109 by using HfO2 dielectric and top-gate geometry [151]. He et al. have fabricated a dual-gate nano-FET based on the same single In-doped CdSe NW using non-oxide high-␬ Si3 N4 and SiO2 as top- and back-gate dielectrics, respectively [154]. The dual-gate FET minimized materials-dependent property fluctuation and enabled direct comparison of the device performance of FETs in both top- and back-gate configurations. Remarkably, the field-effect mobility, peak transconductance, and Ion /Ioff ratio of the Si3 N4 top-gate FET were 52, 142, and 2.81 × 105 times larger than the respective values of the SiO2 back-gate FET. Meanwhile, the threshold voltage and the subthreshold swing of the topgate FET decreased to −1.7 V and 508 mV/dec, respectively. A more significant finding of this work is the demonstration of the predominant effect of FET configuration on the evaluation of electric and transport properties of nanostructures. Thus far, due to the limitation of the nanostructure dimensions, the electric and transport properties of nanostructures are usually studied based on single-object FETs, which has become a standard and well accepted approach. Most nano-FETs are constructed in a simple back-gate configuration with SiO2 as the dielectric layer. However, as demonstrated in [154], the ‘real’ overall electronic and transport properties of the nanostructures studied cannot be evaluated reliably, and confusing results may be obtained. For example, the electron mobility of the same NW deduced from the FET in top-gate configuration is two orders of magnitude lager than that from SiO2 back-gate configuration. Correspondingly, the electron concentration calculated from the two device configurations based on the same single NW also differs by two orders of magnitude. The electron mobility obtained from top-gate configuration is closer to that reported for CdSe single crystals, and thus is believed to be more reliable, reflecting the high-quality single-crystal nature of the CdSe:In NWs. The work demonstrates the need of a proper fabrication protocol of FETs for accurate evaluation of the electronic characteristics of the nanostructures [154].

Metal—semiconductor FETs (nano-MESFETs) MESFET is another important type of FETs and has been used in GaAs microwave devices [155]. Different from a MOSFET, the channel conduction is modulated by a Schottky gate in a

324

J. Jie et al.

Figure 7 Schematic illustration of the back-gate (a) and the top-gate (b) CdS NRs FET. (c) SEM image of a representative top-gate device. The inset shows the enlarged SEM image of the FET channel. (d) Electrical characteristics of a typical top-gate CdS NR FET. Ids —Vds curves of the FET measured with varied gate voltages from −2 to 8 V in a step of +2 V. (e) Ids —Vg curve at Vds = 1 V. Reprinted with permission from Ref. [153].

MESFET. The depletion layer formed at the interface of the metal/semiconductor junction in a MESFET functions similarly to the oxide dielectric in a MOSFET. When a gate voltage is applied to the Schottky gate, the width of the depletion layer is manipulated and thus the effective conduction channel is varied. Schottky diodes based on GaN NWs and ZnO NWs/NRs have been successfully demonstrated [156—158], and Park et al. reported nano-MESFETs from ZnO nanorods [159]. Recently, Ma et al. demonstrated high-performance nano-MESFET by employing the Schottky junction formed between Au and CdS:In NRs, as shown in Fig. 8 [160]. The Au/CdS NR Schottky diodes have an ideality factor of n = 1.14 and low reverse current density. Further characterizations revealed that the CdS NR MESFETs working in the n-channel normally-on (depletion) mode have high transconductance of ∼3.5 ␮S, low subthreshold swing of ∼45 mV/dec, and high on/off current ratio of ∼2 × 108 . Based on the Schottky structures, Qin and Dai et al. constructed a series of nanodevices, including photovoltaic devices, logic circuits, and inverters based on CdS NWs/NRs [161—163]. In MOSFETs, oxide/semiconductor interface is known to be an important issue to seriously influence the device performance due to the carrier traps, defects, and metal ions at the interface. In contrast, no gate dielectric is needed in MESFETs, and the device performance is determined by the quality of Schottky and Ohmic contacts. The Schottky

gate can provide large capacitive coupling with the semiconductor channel, leading to excellent device performance. In addition, nano-MESFETs have a simple structure and can be controlled individually by the local Schottky gate, which facilitates the integration of nanodevices into more complex functional circuits. However, the leakage current cannot be neglected when the forward bias of the Schottky gate is increased to a certain extent. As a result, MESFETs have in general low upper limitations of the on-state gate voltage and on current [151]. Electrode/semiconductor contact Reliable electrode/semiconductor contact is vital for obtaining high-performance electronic and optoelectronic devices. Large contact resistance may result in degeneration of device performance in terms of limiting the work current, increasing energy consumption, deteriorating heat dissipation, and shortening device lifetime. Due to the small contact area, the large contact resistance make the electric contact problem in nanostructure-based devices more serious. For n-type II—VI nanostructures, metals with low work functions are used for achieving Ohmic contact. Among them, indium has shown ideal Ohmic contact with n-CdS and n-CdSe NWs/NRs [112,97]. Metals such as Al, Ti, and Ag have also been examined, but they are usually inferior to In [139,140]. As for p-type II—VI nanostructures, such as p-ZnS,

One-dimensional II—VI nanostructures

325

Figure 8 Performances of a single CdS NR MESFET. (a) Two-terminal I—V curves measured between the source-drain (black curve), source-gate (red curve), and drain-gate (green curve). (b) Ids —Vds curves measured at various gate voltages. (C) Ids —Vg curves measured at various Vds on an exponential scale (black, red, and green curves corresponding to Vds = 1.0, 0.8, and 0.6 V, respectively). Inset shows the FESEM image of a single CdS NR MESFET. The scale bar is 10 ␮m. (d) Zoomed in Ids —Vg curve in the region around Vth (Vds = 1.0 V). (e) The gm —Vg curve (Vds = 1.0 V). (f) Ig —Vg curves measured at various Vds (black, red, and green curves corresponding to Vds = 1.0, 0.8, and 0.6 V, respectively). Reprinted with permission from Ref. [160].

p-ZnSe, p-CdTe, p-ZnTe, metals with high work functions, i.e., Au, Pd and Ni have exhibited reliable Ohmic contacts [52,77,123—125]. Electrical contact with n-type wide band-gap II—VI materials such as n-ZnS and n-ZnSe nanostructures remains a

challenging issue. Schottky contacts are normally formed for various electrode metals such as Ag [52], Cr [164,165], Au [166—168], and Pt:Ga [169]. Although several reports indicated that Ti could form Ohmic contact with unintentionally doped n-ZnSe NWs [170], we note that reliable

326 and reproducible Ohmic contact of Ti to n-ZnS and n-ZnSe nanostructures is still hard to achieve. Surface states and consequent surface Fermi-level pinning might be responsible for the non-Ohmic contact. ZnS and ZnSe nanostructures have in general high-density defects and dangling bonds on their surfaces, which may lead to the pinning of surface Fermi level in the band-gap. In this case, a negative surface is resulted due to the formation of acceptor states; the surface energy band bends upwards and thus induces a large contact barrier at the interface. To resolve this problem, reducing the surface states and avoiding surface Fermi-level pinning are the keys. To employ transparent conductive oxide (TCO) electrodes such as In2 O3 :Sn (ITO) and ZnO:In (IZO) instead of conventional metal electrodes might be a feasible way. The oxide layer saturates the surface states, and a highly conductive interfacial layer is likely formed, which may avoid surface Fermi-level pinning and promote the carrier injection.

Integrated electronic devices As we discussed above, various individual devices based on single II—VI nanostructures have been demonstrated. High-performance integrated devices, such as logic, storage, and drive, have also been fabricated on the basis of these II—VI nanodevices. The integration of individual nanodevices enables implementing advanced or multifunctions. Ma et al. demonstrated that nanoinverters with a voltage gain as high as 83 could be fabricated by combing two CdS NW MESFETs, and NOR and NAND gates by integrating three transistors [161]. Wu et al. reported high-performance nanoinverters with a voltage gain up to 1000 based on CdS NR MOSFET with high-␬ HfO2 dielectric [163]. Recently, He et al. reported the realization of logic circuits by combing two top-gate CdSe:In NW FETs, which are also capable of incorporating of the inherent photoresponse properties of CdSe NWs [154]. The CdSe:In NWs showed high sensitivity to visible light, and illumination using an incandescent light source increased the threshold voltage of the transistor, as depicted in Fig. 9(a). The Ids —Vgs curves measured in dark revealed slight hysteresis at a sweeping rate of 6 V/s, which, however, became negligible under illumination. Fig. 9(b) shows the Vout —Vin (Vg ) dependences for two individual top-gate FETs connected in the circuit as depicted in the inset in Fig. 9(b). For both transistors, Vout showed sharp responses to Vin , and the performance gain (defined as dVout /dVin in the inflection regions of the Vout —Vin curves) is nearly six at a sweeping rate of 6 V/s. Two top-gate transistors were connected in series and in parallel to construct two basic logic circuits, ‘AND’ and ‘OR’, respectively, as shown in Fig. 9(c). The logic ‘AND’ and ‘OR’ circuits demonstrated great performance gain and stability. Fig. 9(d) and (e) reveal the reasonable operation speeds and high reproducibility of both ‘AND’ and ‘OR’ circuits. In addition, by combining the photoelctronic response characteristics of CdSe NW transistors, comprehensive photo-sensitive switches were achieved. Studies on integrated devices are expected to promote the practical applications of II—VI nanostructures. The following two aspects are considered to be critical to realize high-performance integrated circuits: (1)

J. Jie et al. the integrated circuits are built on the basis of individual nanodevices such as nano-FETs and nanodiodes; the improvement of the performance of individual is therefore a prerequisite for realizing high-performance integrated circuits. (2) To achieve high-density integrated nanodevices, the tools/methods for reliably and facilely manipulating and assembling the individual nanoobjects to specific position, alignment, and density have to be developed. The low-cost technology for fabricating integrated nanodevices in a large scale is still a bottleneck.

Optoelectronic applications of II—VI nanostructures Photodetectors II—VI semiconductors have direct band-gaps from 1.5 eV (CdTe) to 3.7 eV (ZnS), and the band-gaps of II—VI nanostructures could be further extended via quantum confinement effects. Therefore II—VI nanostructures are very promising candidates for applications such as nanophotodetectors (nano-switches) available to work from NIR to UV. Distinctively, owing to the high quantum efficiency and high crystallinity of II—VI nanostructures, such nano-photodetectors are expected to have much improved performance and lower cost as compared with the conventional II—VI film and bulk devices. Photoconductors Photoconductor is a kind of photodetectors in a simple device structure. Fig. 10 shows a nano-photoconductor fabricated on single-crystal CdS NRs [171]. Ti/Au electrodes were deposited on the two ends of the CdS NR forming a metal—semiconductor—metal (MSM) structure. It was revealed that the photocurrent was dependent strongly on excitation wavelength and showed the highest light sensitivity at 490—495 nm. A power-law dependence of photocurrent on light intensity and four orders of magnitude change in conductance was observed for the CdS NR photoconductor. Significantly, the photoconductor had the response speed (response time ∼hundreds of microseconds) substantively faster than those ever reported for conventional film and bulk CdS photoconductors (>tens of milliseconds). The size of NRs had a significant influence on the response speed; and smaller CdS NRs showed higher response speed. Moreover, it was observed that the absorption of ambient gas (mainly oxygen) could significantly change the photoconductivity of CdS NRs. The following physical processes should be considered to understand the photoconductive properties of semiconductor nanostructures: (1) exciton generation by incident light with energy larger than the band-gap, which determines the spectral response of photoconductors. (2) Electrons and holes separation by applied electrical field. The process induces photocurrent, and a strong electrical field is desired to facilitate exciton dissociation and consequently increase the response speed. (3) Exciton recombination at the surface and bulk of nanostructures (leading to the decrease of photocurrent). The lifetime of the photo-generated carriers determines the gain and the response speed. However, large gain will reduce the response speed. A compromise

One-dimensional II—VI nanostructures

327

Figure 9 (a) Hysteresis loops of Ids —Vgs curves of a top-gate transistor, measured in dark () and under illumination (red ), respectively. Vgs sweeps from −5 to 3 V and back to −5 V at a sweep rate of 6 V/s. (b) Vout —Vin dependencies for two different top-gate FETs connected in the circuit as depicted in the inset. The two FETs are denoted as transistors 1 and 2, respectively. R is a constant resistor of 100 M. The measurements were conducted at a constant Vcc = 2.5 V. Vin sweeps from −3 to 3 V and back to −3 V at sweeping rates of 6 and −6 V/s, respectively. The solid line is forward and the dashed line is backward for each transistor. Although the turn-on voltages for the transistors are different, both transistors can be considered in ‘ON’ states (logic 1) at Vin = 2.5, and in ‘OFF’ state (logic 0) at Vin = −2.5 V, where Vout > 2 V is defined as ‘ON’ and Vout < 0.5 V as ‘OFF’ state, respectively. (c) The schematic configuration of the typical ‘‘AND’’ and ‘‘OR’’ circuits design based on two transistors. The inset table summarizes the experimental results for the two logic gates. (d) A typical switch response from the ‘‘AND’’ circuit. (e) A typical switch response from the ‘‘OR’’ circuit, where the final change shows the light-induced switch on when the two inputs are set as logic 0. Reprinted with permission from Ref. [154].

between gain and response speed has to be made to get a reasonable overall performance of the device. (4) Carrier trapping. Carrier trapping and releasing play key roles in determining the rise time and decay time. In other words, high crystallinity (fewer defects) of nanostructures could benefit a high response speed. (5) For wide band-gap semiconductors, surface band bending usually occurs due to Fermi-level pinning, which results in an energy barrier for electron—hole recombination and is responsible for the persistent photocurrent in GaN NWs [172]. The barrier height

is size-dependent, and thinner NWs have a lower barrier. (6) Ambient effect. Absorption of ambient gas, particularly O2 and moisture, is able to withdraw electrons from n-type nanostructures and thus leads to the decrease of photocurrent. On the contrary, desorption of gas molecules will contribute to the increase of photocurrent. Based on the above discussion, it is understandable that the high photosensitivity and high photoresponse speed of CdS NR photoconductors were attributable to the large surface-tovolume ratio and high single-crystal quality of CdS NRs, and

328

J. Jie et al.

Figure 10 (a) I—V curves of a CdS single-nanoribbon illuminated with light of different wavelength. The light intensity is kept constant at 1.8 mW/cm2 . The insets show the optical microscopic image of the single-nanoribbon device (left) and the schematic diagram of the configuration for photoconductive measurements (right). (b) Spectral response of the nanoribbon () measured at a constant bias of 1 V. The normalized room-temperature PL (red line) and absorption (blue line) spectra of CdS nanoribbons are also illustrated for reference. (c) Time response of the CdS nanoribbon to the pulsed incidence light (490 nm, 1.8 mW/cm2 ). The bias voltage is 1 V. The arrows indicate the 10% and 90% points of the peak value used for calculating the rise and fall time. The solid lines are obtained by smoothing the data points (open circles). The inset shows the relative balance of current ((Imax − Imin )/Imax %) vs. the frequency of pulsed light. (d) Natural logarithmic plot of the time response spectrum. Note the rise and fall edges are well fitted with straight lines. Reprinted with permission from Ref. [171].

the reduction of recombination barrier in nanostructures [171]. Besides CdS NR photoconductor, nano-photoconductors based on CdSe [173], ZnTe [125], ZnSe [165], and ZnS [164] nanostructures have also been demonstrated recently. However, it is noted that the devices fabricated from undoped II—VI nanostructures are usually insensitive to incident light and have a low photocurrent. Moreover, the large spacing between electrodes of the reported devices resulted in weak electrical field in the nanostructures and a long diffuse distance of the photo-generated carriers to electrodes, thus reducing the sensitivity and response speed of the nano-photoconductors. To achieve high-performance

nano-photodetectors, further optimization of the materials and device structures is necessary. Phototransistors and photodiodes Upon light irradiation, the conventional nano-FETs and nanodiodes can function as nano-photodetectors as well. The merits of the so-called phototransistors and photodiodes are obvious: Duo to the strong electrical field (applied or self-built) at the oxide/semiconductor interface for nanoFETs, metal/semiconductor interface for Schottky diodes, or p/n interface for p—n junctions, the photo-generated electron—hole pairs could be quickly separated and transferred to electrodes. As a result, the devices may afford

One-dimensional II—VI nanostructures a high photosensitivity and high response speed. Moreover, the existence of depletion layers in the junctions can suppress the dark current and thus enhance the on/off ratio remarkably. Fig. 11 shows the device characteristics of a phototransistor fabricated on CdS:P NRs [153]. From the transfer characteristics measured in dark (solid line) and in white light illumination (dashed line), it was revealed that the Vth shifted dramatically from −1.45 to −11 V due to increased carrier concentration in the NR upon light irradiation. A high response ratio of Ilight /Idark = 106 was obtained at a fixed Vg of −3 V, which was improved drastically from ∼10 when no gate voltage was applied. Moreover, due to the negative gate voltage, the accumulated carriers could deplete as soon as the incident light was turned off, leading to a fast photoresponse speed and the fall time was <5 s for the phototransistor. In contrast, the photoconductor built on the same NR showed a much longer fall time of several tens seconds. A photodiode fabricated on a single CdS:P NRs is shown in Fig. 12 [174]. The Au/CdS:P NR contact formed an ideal Schottky diode and a high on/off current ratio of ∼106 was deduced from the rectifying characteristic curve (Fig. 12(c)). Measurements on the photoresponse revealed a high Ilight /Idark current ratio of ∼103 with a reasonable response speed under the illumination of white light. Recently, Wei et al. reported that the sensitivity and response speed of photon detection could be remarkably enhanced by employing Pt/CdS NW Schottky photodiodes [175]. Remarkably, as a high back bias was applied on the p—n nanophotodiodes, avalanche photodiodes were demonstrated on nanoscale p—n diodes consisting of crossed Si/CdS NWs [176]. The NW-based avalanche photodiodes (nanoAPDs) exhibited ultrahigh sensitivity with a detection limit less than 100 photons and a sub-wavelength spatial resolution of 250 nm. Based on above progresses, it is optimistic to anticipate that practical high-performance nano-photodetectors based on II—VI nanostructures may be realized in the near future.

Light-emitting diodes (LEDs) and laser diodes (LDs) Lieber and co-workers have made an impressive progress in fabricating electrically driven light-emitting nanodevices including nano-LEDs and nano-LDs from various semiconductor nanostructures. They utilized a crossed p- and n-type NW scheme instead of the epitaxial junction growth to obtain p—n diodes [177—179]. The scheme was first demonstrated for p—n crossed InP NW junction [177] and subsequently extended to other NW systems such as p-Si/n-GaN, p-GaN/n-GaN, p-Si/n-CdS, and p-Si/n-CdSe [178]. Electroluminescence (EL) of a crossed p-InP/n-InP diode under a forward bias was investigated, as shown in Fig. 13(a). The 3D EL intensity mapping indicated that the emission originated from a point-like source, which corresponded to the band-edge emission of InP from the cross-point. Emission wavelength of the crossed NW diodes can be modulated by using semiconductor nanostructures with varied band-gap, and wide emission wavelength covering the whole spectral regime from UV to near IR has been illustrated (Fig. 13(b) and (c)). Due to the intrinsically low

329 surface state density of II—VI materials, the CdS nano-LED exhibited a higher quantum efficiency (0.1—1%) than that of the InP (∼0.001%) and GaN nano-LEDs. Moreover, the nano-LEDs have a near-field power densities greater than 100 W/cm2 , which is actually sufficient to excite molecular and nanoparticle chromophores, such as CdSe quantum dots and propidium iodides [178], and implies the potential applications of nano-LEDs in integrated chemical and biological analysis. Duan et al. demonstrated laser emission from CdS NW by using a hybrid structure (Fig. 13(d)—(f)), in which a high carrier injection density along the length of a CdS cavity was achieved [17]. Low-temperature measurements made on the devices showed the preferential pumping into single mode. The dominant emission line at 493 nm has a linewidth of 0.8 nm (Fig. 13(g)), even comparable to the instrument resolution. In spite of the large progress in nano-LEDs and nano-LDs on the basis of p-Si/n-II—VI hybrid nanostructures, it is worth to point out that nano-emission devices based on pure II—VI p—n diodes (both p and n parts from II—VI materials) have not been realized thus far. II—VI p—n diodes made of II—VI p—n homojunctions are expected to be superior to the pSi/n-II—VI hetero-junctions diodes in terms of efficiency and device performance owing to better energy band matching and less interface defects. However, II—VI semiconductors typically suffer from strong self-compensation and unipolar conduction. Complementary n- and p-type doping is important for the realization of fully II—VI-based nano-LEDs and nano-LDs, for which further research is needed.

All-NW image sensors A contact NW-printing method has been proposed by Fan and Javey et al. to obtain highly aligned, parallel NW arrays in large area on receiver substrates [180]. Based on this technique, they demonstrated an all-NW image sensor via the integration of nano-photodetectors and nano-transistors, as shown in Fig. 14 [181]. Each sensing circuitry was composed of a CdSe NW light sensor (NS), an impedance-matching NW transistor (T1), and a buffer transistor (T2). A light signal is detected by the NS and the current output signal is then amplified by the two transistors, which are effectively configured as a voltage divider (Fig. 14(a) and (b)). The nano-transistors (T1 and T2) were constructed with parallel arrays of Gi/Si core/shell NWs. The image sensor consisted of arrays (13 × 20) of the all-NW circuitry, in which each circuit element operated as a single pixel with a functional pixel yield of ∼80% (Fig. 14(c)). A 2D circular-light intensity map imaged by the all-NW sensor arrays was demonstrated in Fig. 14(d), in which the spatial intensity variation from the center to the outer edge of the circuit matrix could reveal the intensity profile of the projected circular-light pattern. The work provided an excellent example for applying II—VI nanostructures in integrated nano-optoelectronic devices.

Future challenges Recent advances in nanoscience and nanotechnology open up myriad opportunities for new-generation nanodevice applications of II—VI semiconductors. While the practical

330

J. Jie et al.

Figure 11 Device characteristics of a phototransistor fabricated on a single CdS:P NR with high-␬ HfO2 dielectric and top-gate configuration. (a) Ids —Vg curves measured in dark (solid line) and under light irradiation (dashed line), respectively, at Vds = 0.3 V. (b) Time—response curves of the CdS:P top-gate FET to pulsed white light measured at Vg = −3 V (red line) and 0 V (black line), respectively. Inset shows the plots in linear scale. A and B mark the first and second decay edges of the response curve measured at Vg = 0 V, respectively. Reprinted with permission from Ref. [153].

applications of II—VI-based thin film/bulk devices are seriously obstructed by the difficulties in achieving high-quality crystals and complementary doping, II—VI nanomaterials with high-crystal quality have been readily synthesized by the ‘‘bottom-up’’ growth techniques. It has been demonstrated that the II—VI nanostructures could be grown without relying on the use of lattice-matched substrates, thus over-

coming one of the most difficult problems faced by their thin film counterparts. In particular, the high-crystal quality as well as the capability in defect relaxation towards surface may alleviate self-compensation effects and allow more efficient doping in II—VI nanostructures. Moreover, the unique electronic and optical properties of the II—VI nanomaterials arising from the nanoscale effects such as

Figure 12 Metal—semiconductor Schottky photodiode fabricated on a single CdS:P NR. (a) SEM image of the photodiode. (b) Zoomed in image of the photodiode. (c) Rectification characteristic curve of the Au/CdS:P NR Schottky diode. (c) Time response curve of the photodiode at a backward bias of −0.3 V. White light from microscope serves as the light source.

One-dimensional II—VI nanostructures

331

Figure 13 (a) False color scanning electron micrograph of an n-InP/p-InP crossed NW device, overlaid with corresponding spatially resolved EL image showing the light emission from the cross-point. (b) Schematic of a tricolor nano-LED array assembled by crossing n-GaN, n-CdS, and n-CdSe NWs with a p-Si NW. The array was obtained by fluidic assembly and photolithography with ca. 5 ␮m separation between NW emitters. (c) Normalized EL spectra and color images from the three junctions. (d) Schematic showing the cross-section of a nano-LD. In this structure, electrons and holes can be injected into the CdS NW along the whole length from the top metal layer and the bottom p-Si layer, respectively. (e) Optical image of a nano-LD. The arrow highlights the exposed CdS NW end. Scale bar is 5 ␮m. (f) An EL image recorded at room temperature with an injection current of about 80 ␮A. The arrow highlights emission from the CdS NW end. (g) Emission spectra from a CdS NW device with injection currents of 200 ␮A (red) and 280 ␮A (green) recorded at 8 K. The spectra are offset by 0.10 intensity units for clarity. Reprinted with permission from Ref. [17,178,179].

size-confinement effects and surface effects offer the possibility for a variety of new nanodevices with more powerful functions, lower cost, and less energy consumption. In the past decade, researchers have achieved significant progresses in the synthesis and characterization of various II—VI nanostructures. Single-crystal II—VI nanostructures in the forms of NWs/NRs/NTs have been readily grown via the welldeveloped physical and chemical methods such as VLS, VS, OAG, and SLS, etc. Laser emission with tunable wavelength has been realized from single II—VI nanostructures such as ZnS and CdS NWs/NRs. A host of novel nanodevices such as FETs, diodes, logic circuits, LEDs, LDs, photodetectors, and solar cells, etc. based on II—VI nanostructures have been successfully demonstrated. These nanodevices showed superior performance to those made of their thin film counterparts, revealing their tremendous application potentials in nanoelectronics and nano-optoelectronics. The followings are considered to be important issues for promoting the practical applications of II—VI nanostructures:

(1) The properties of nanostructures are determined by their geometries, stoichiometric compositions, structures, and surface states. Although various methods have been employed to synthesize II—VI nanostructures, reliable synthesis of nanostructures with high uniformity and controlled dimension, structure, composition, location, orientation, and alignment remains a challenge, and will inevitably influence the feasibility and reproducibility of the nanodevices. Therefore new methodologies are still needed to further improve the synthesis techniques. (2) Controlled doping of II—VI structures with bipolar nand p-type conduction is another important issue for achieving high-performance II—VI nanodevices. Studies on doping techniques and doping mechanisms for II—VI nanostructure are relatively few, and acutely needed in future research. In addition to the conventional bulk doping, surface charge transfer doping is a promising approach to tune the electronic and optoelectronic

332

J. Jie et al.

Figure 14 (a) Diagram of a proof-of-concept NW photosensor circuitry, consisting of a CdSe NW light sensor (NS), an impedancematching NW transistor (T1), and a buffer transistor (T2). (b1—b4) Optical and SEM images of an all-NW sensor circuitry with the above mentioned circuit elements. (c) A perspective picture showing the imaging function of the circuit array. (d) 2D circular-light intensity map imaged by the all-NW sensor arrays, demonstrating a proof-of-concept image sensing. The contrast represents the normalized photocurrent, with the gray pixels representing the defective circuits. Reprinted with permission from Ref. [181].

properties of nanomaterials due to their distinctive large surface-to-volume ratios. (3) The realization of advanced device structures such as p—n junctions, superlattices, and quantum wells is the basis of functional nanodevices, and needs to be further addressed. As compared with the film counterparts, an important feature of nanodevices is high flexibility. Diverse device architectures, such as crossed junction and hierarchical junction, which are difficult in the film format, can be readily obtained for nanodevices. Nevertheless, it remains a challenging task to obtain nanostructures with desired structures due to the complexity in nanostructure growth. (4) Surface-sensitive properties of nanostructures form the basis of high-sensitivity gas, chemical and biological detectors. However, such high surface sensitivity also presents a serious challenge to device fabrication as it leads to controversial interpretation of the experimental observations and difficulty in the reliability and reproducibility of device performances. It is also considered a possible origin for the inferior device performance to that expected for nanostructures. Therefore, understanding the nanomaterial/metal, nanomaterial/dielectric, nanoma-

terial/nanomaterial interfaces, and the proper design of device structures are in particular important for fabricating nanostructure-based devices with controlled performance. (5) Various individual nanodevice prototypes have been demonstrated thus far, they have presented some properties superior to their film counterparts in electronic and optoelectronic devices. The capability to assemble and integrate individual nanodevices into functional devices on a large scale is essential for their practical applications. To achieve high-density integrated nanodevices, the tools/methods for reliably and facilely manipulating and assembling the individual nanoobjects to specific position, alignment, and density have to be developed. The low-cost technology for fabricating integrated nanodevices in a large scale is still a bottleneck.

Acknowledgements The work was supported by the Research Grants Council of Hong Kong SAR, China—–CRF Grant (No. CityU5/CRF/08) and GRF Grant (No. CityU110209), the National High Technology Research and Development Program of China (No. 2007AA03Z301), the National Natural Science Foundation

One-dimensional II—VI nanostructures of China (Nos. 60806028 and 20901021), National Basic Research Program of China (No. 2006CB933000), and Program for New Century Excellent Talents in University of the Chinese Ministry of Education (NCET-08-0764).

References [1] [2] [3] [4] [5] [6]

[7] [8] [9]

[10] [11]

[12] [13] [14]

[15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31]

S. Iijima, Nature 354 (1991) 56. S. Kumar, T. Nann, Small 2 (2006) 316. U.V. Desnica, Prog. Cryst. Growth Charact. 36 (1998) 291. J.J. Loferski, J. Appl. Phys. 27 (1956) 777. M. Green, Third-generation Photovoltaics: Advanced Solar Energy Conversion, Springer, Berlin, 2006. T. Aramoto, S. Kumazawa, H. Higuchi, T. Arita, S. Shibutani, T. Nishio, J. Nakajima, M. Tsuji, A. Hanafusa, T. Hibini, K. Omura, H. Ohyama, M. Murozono, Jpn. J. Appl. Phys. 36 (1977) 6304. R.W. Miles, K.M. Hynes, I. Forbes, Prog. Cryst. Growth Charact. Mater. 51 (2005) 1. M. Hasse, J. Qui, J. DePuydt, H. Cheng, Appl. Phys. Lett. 59 (1991) 1272. S. Nakamura, M. Senoh, S. Nagahama, N. Iwasa, T. Yamada, T. Matsushita, H. Kiyoku, Y. Sugimoto, Jpn. J. Appl. Phys. 35 (1996) L174. S. Taniguchi, T. Hino, S. Itoh, K. Nakano, N. Nakayama, A. Ishibashi, M. Ikeda, Electron. Lett. 32 (1996) 552. K.W. Haberern, P.F. Baude, S.J. Flamholtz, M. Buijs, J.J. Horikx, K.K. Law, M.A. Haase, T.J. Miller, G.M. Haugen, Proc. SPIE-Int. Soc. Opt. Eng. 3001 (1997) 101. H. Morkoc, S. Strite, G.B. Gao, M.E. Lin, B. Sverdlov, M. Burns, J. Appl. Phys. 76 (1994) 1363. T.E. Schlesinger, J.E. Toney, H. Yoon, E.Y. Lee, B.A. Brunett, L. Franks, R.B. James, Mater. Sci. Eng. 32 (2001) 103. T.E. Schlesinger, R.B. James (Eds.), Semiconductors for Room Temperature Nuclear Applications, vol. 43, Academic Press, New York, 1995, Chapter 9, p. 335. Y.K. Liu, J.A. Zapien, Y.Y. Shan, C.Y. Geng, C.S. Lee, S.T. Lee, Adv. Mater. 17 (2005) 1372. M. Law, D.J. Sirbuly, J.C. Johnson, J. Goldberger, R.J. Saykally, P.D. Yang, Science 305 (2004) 1269. X. Duan, Y. Huang, R. Agarwal, C.M. Lieber, Nature 421 (2003) 241. B. Yu, M. Meyyappan, Solid-State Electron. 50 (2006) 536. Z.L. Wang, J. Phys.: Condens. Matter 16 (2004) R829. Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F. Kim, H. Yan, Adv. Mater. 15 (2003) 353. R.S. Wagner, W.C. Ellis, Appl. Phys. Lett. 16 (1964) 761. A.M. Morales, C.M. Lieber, Science 279 (1998) 208. Z.L. Wang, Adv. Mater. 15 (2003) 432. Y.F. Zhang, Y.H. Tang, N. Wang, D.P. Yu, C.S. Lee, I. Bello, S.T. Lee, Appl. Phys. Lett. 72 (1998) 1835. W.S. Shi, Y.F. Zheng, N. Wang, C.S. Lee, S.T. Lee, Adv. Mater. 13 (2001) 591. C.J. Barrelet, Y. Wu, D.C. Bell, C.M. Lieber, J. Am. Chem. Soc. 125 (2003) 11498. X.M. Meng, J. Liu, Y. Jiang, W.W. Chen, C.S. Lee, I. Bello, S.T. Lee, Chem. Phys. Lett. 382 (2003) 434. J.X. Ding, J.A. Zapien, W.W. Chen, Y. Lifshitz, S.T. Lee, X.M. Meng, Appl. Phys. Lett. 85 (2004) 2361. D.F. Moore, Y. Ding, Z.L. Wang, J. Am. Chem. Soc. 126 (2004) 14372. M. Wang, G.T. Fei, X.G. Zhu, B. Wu, M.G. Kong, L.D. Zhang, J. Phys. Chem. C 113 (2009) 4335. G.Z. Shen, Y. Bando, D. Golberg, C.W. Zhou, J. Phys. Chem. C 112 (2008) 12299.

333 [32] Y. Jiang, W.J. Zhang, J.S. Jie, X.M. Meng, J.A. Zapien, S.T. Lee, Adv. Mater. 18 (2006) 1527. [33] Y.Q. Li, J.X. Tang, H. Wang, J.A. Zapien, Y.Y. Shan, S.T. Lee, Appl. Phys. Lett. 90 (2007) 093127. [34] Y.Q. Li, K. Zou, Y.Y. Shan, J.A. Zapien, S.T. Lee, J. Phys. Chem. B 110 (2006) 6759. [35] C. Ma, D. Moore, J. Li, Z.L. Wang, Adv. Mater. 15 (2003) 228. [36] Y. Jiang, X.M. Meng, Z.Y. Xie, C.S. Lee, S.T. Lee, Adv. Mater. 15 (2003) 323. [37] D. Moore, Z.L. Wang, J. Mater. Chem. 16 (2006) 3898. [38] J.A. Zapien, Y. Jiang, X.M. Meng, W. Chen, F.C.K. Au, Y. Lifshitz, S.T. Lee, Appl. Phys. Lett. 84 (2004) 1189. [39] Y.K. Liu, C.Y. Geng, J.A. Zapien, Y.Y. Shan, C.S. Lee, Y. Lifshitz, S.T. Lee, Appl. Phys. Lett. 85 (2004) 3241. [40] Y.K. Liu, J.A. Zapien, Y.Y. Shan, H. Tang, C.S. Lee, S.T. Lee, Nanotechnology 18 (2007) 365606. [41] J.A. Zapien, Y.K. Liu, Y.Y. Shan, H. Tang, C.S. Lee, S.T. Lee, Appl. Phys. Lett. 90 (2007) 213114. [42] Y.F. Hao, G.W. Meng, Z.L. Wang, C.H. Ye, L.D. Zhang, Nano Lett. 6 (2006) 1650. [43] J.S. Jie, W.J. Zhang, Y. Jiang, X.M. Meng, J.A. Zapien, M.W. Shao, S.T. Lee, Nanotechnology 17 (2006) 2913. [44] X. Fan, X.M. Meng, X.H. Zhang, W.S. Shi, W.J. Zhang, J.A. Zapien, C.S. Lee, S.T. Lee, Angew. Chem. Int. Ed. 45 (2006) 2568. [45] X. Fan, M.L. Zhang, I. Shafiq, W.J. Zhang, C.S. Lee, S.T. Lee, Adv. Mater. 21 (2009) 2393. [46] M.W. Murphy, X.T. Zhou, J.Y.P. Ko, J.G. Zhou, F. Heigl, T.K. Sham, J. Chem. Phys. 130 (2009) 084707. [47] P.V. Radovanovic, C.J. Barrelet, S. Gradecak, F. Qian, C.M. Lieber, Nano Lett. 5 (2005) 1407. [48] Y.Q. Li, J.A. Zapien, Y.Y. Shan, Y.K. Liu, S.T. Lee, Appl. Phys. Lett. 88 (2006) 013115. [49] T. Kang, J. Sung, W. Shim, H. Moon, J. Cho, Y. Jo, W. Lee, B. Kim, J. Phys. Chem. C 113 (2009) 5352. [50] B.C. Cheng, Z.G. Wang, Adv. Funct. Mater. 15 (2005) 1883. [51] M.Y. Lu, M.P. Lu, Y.A. Chung, M.J. Chen, Z.L. Wang, L.J. Chen, J. Phys. Chem. C 113 (2009) 12878. [52] G.D. Yuan, W.J. Zhang, W.F. Zhang, X. Fan, I. Bello, C.S. Lee, S.T. Lee, Appl. Phys. Lett. 93 (2008) 213102. [53] S.K. Chan, Y. Cai, I.K. Sou, N. Wang, J. Cryst. Growth 278 (2005) 146. [54] Y. Cai, T.L. Wong, S.K. Chan, I.K. Sou, D.S. Su, N. Wang, Appl. Phys. Lett. 93 (2008) 233107. [55] Y. Cai, S.K. Chan, I.K. Sou, Y.F. Chan, D.S. Su, N. Wang, Adv. Mater. 18 (2006) 109. [56] X.T. Zhang, Z. Liu, Q. Li, Y. Leung, K. Ip, S. Hark, Adv. Mater. 17 (2005) 1405. [57] X.T. Zhang, Z. Liu, K.M. Ip, Y.P. Leung, Q. Li, S.K. Hark, J. Appl. Phys. 95 (2004) 5752. [58] L. Shi, Y. Xu, Q. Li, J. Phys. Chem. C 113 (2009) 1795. [59] W.C.H. Choy, C.F. Guo, Y.P. Leung, G.Z. Wang, T.I. Yuk, Appl. Phys. A 83 (2006) 301. [60] Y.P. Leung, W.C.H. Choy, I. Markov, G.K.H. Pang, H.C. Ong, T.I. Yuk, Appl. Phys. Lett. 88 (2006) 183110. [61] T.Q. Jia, H.X. Chen, M. Huang, X.J. Wu, F.L. Zhao, M. Baba, M. Suzuki, H. Kuroda, J.R. Qiu, R.X. Li, Z.Z. Xu, Appl. Phys. Lett. 89 (2006) 101116. [62] A.B. Panda, S. Acharya, S. Efrima, Adv. Mater. 17 (2005) 2471. [63] S. Acharya, A.B. Panda, S. Efrima, Y. Golan, Adv. Mater. 19 (2007) 1105. [64] Y. Jiang, X.M. Meng, W.C. Yiu, J. Liu, J.X. Ding, C.S. Lee, S.T. Lee, J. Phys. Chem. B 108 (2004) 2784. [65] X.T. Zhang, K.M. Ip, Z. Liu, Y.P. Leung, Q. Li, S.K. Hark, Appl. Phys. Lett. 84 (2004) 2641. [66] A. Othonos, E. Lioudakis, U. Philipose, H.E. Ruda, Appl. Phys. Lett. 91 (2007) 24113.

334 [67] Q. Li, X.G. Gong, C.R. Wang, J. Wang, K. Ip, S. Hark, Adv. Mater. 16 (2004) 1436. [68] X. Fan, X.M. Meng, X.H. Zhang, M.L. Zhang, J.S. Jie, W.J. Zhang, C.S. Lee, S.T. Lee, J. Phys. Chem. C 113 (2009) 834. [69] R. Solanki, J. Huo, J.L. Freeouf, B. Miner, Appl. Phys. Lett. 81 (2002) 3864. [70] K. Wang, J.J. Chen, W.L. Zhou, Y. Zhang, Y.F. Yan, J. Pern, A. Mascarenhas, Adv. Mater. 20 (2008) 3248. [71] Z. Liu, C.X. Shan, S.K. Hark, L.P. You, J. Chen, J. Phys. Chem. C 111 (2007) 16181. [72] C.R. Wang, J. Wang, Q. Li, G.C. Yi, Adv. Funct. Mater. 15 (2005) 1471. [73] T. Aichele, A. Tribu, G. Sallen, J. Bocquel, E.B. Amalric, C. Bougerol, J.P. Poizat, K. Kheng, R. André, S. Tatarenko, H. Mariette, J. Cryst. Growth 311 (2009) 2123. [74] P.T.K. Chin, J.W. Stouwdam, P.A.J. Jassen, Nano Lett. 9 (2009) 745. [75] B.J. Xi, D.C. Xu, S.L. Xiong, C.M. Wang, X.M. Feng, H.Y. Zhou, Y.T. Qian, J. Phys. Chem. C 112 (2008) 5333. [76] X.T. Zhang, K.M. Ip, Q. Li, S.K. Hark, Appl. Phys. Lett. 86 (2005) 203114. [77] H.S. Song, W.J. Zhang, G.D. Yuan, Z.B. He, W.F. Zhang, Y.B. Tang, L.B. Luo, C.S. Lee, I. Bello, S.T. Lee, Appl. Phys. Lett. 95 (2009) 033117. [78] D.S. Xu, Y.J. Xu, D.P. Chen, G.L. Guo, L.L. Gui, Y.Q. Tang, Adv. Mater. 12 (2000) 520. [79] X. Duan, C.M. Lieber, Adv. Mater. 12 (2000) 298. [80] S.P. Mondal, K. Das, A. Dhar, S.K. Ray, Nanotechnology 18 (2007) 095606. [81] P. Yang, Y. Xie, Y.T. Qian, X.M. Liu, Chem. Commun. 14 (1999) 1293. [82] A. Datta, P.G. Chavan, F.J. Sheini, M.A. More, D.S. Joag, A. Patra, Cryst. Growth Des. 9 (2009) 4157. [83] B.L. Cao, Y. Jiang, C. Wang, W.H. Wang, L.Z. Wang, M. Niu, W.J. Zhang, Y.Q. Li, S.T. Lee, Adv. Funct. Mater. 17 (2007) 1501. [84] L.F. Dong, J. Jiao, M. Coulter, L. Love, Chem. Phys. Lett. 376 (2003) 653. [85] X.F. Duan, C.M. Niu, V. Sahi, J. Chen, J.W. Parce, S. Empedocles, J.L. Goldman, Nature 425 (2003) 274. [86] Z.Q. Wang, J.F. Gong, J.H. Duan, H.B. Huang, S.G. Yang, X.N. Zhao, R. Zhang, Y.W. Du, Appl. Phys. Lett. 89 (2006) 033102. [87] A.L. Pan, R.B. Liu, B.S. Zou, Appl. Phys. Lett. 88 (2006) 173102. [88] C.J. Barrelet, A.B. Greytak, C.M. Lieber, Nano Lett. 4 (2004) 1981. [89] A.L. Pan, D. Liu, R.B. Liu, F.F. Wang, X. Zhu, B.S. Zou, Small 1 (2005) 980. [90] X. Fan, M.L. Zhang, I. Shafiq, W.J. Zhang, C.S. Lee, S.T. Lee, Cryst. Growth Des. 9 (2009) 1375. [91] Y.-J. Hsu, S.-Y. Lu, Chem. Commun. 2102 (2004). [92] X.L. Fu, L.H. Li, W.H. Tang, Solid State Commun. 138 (2006) 139. [93] D.S. Kim, Y.J. Cho, J. Park, J. Yoon, Y. Jo, M.-H. Jung, J. Phys. Chem. C 111 (2007) 10861. [94] C.W. Na, D.S. Han, D.S. Kim, Y.J. Kang, J.Y. Lee, J. Park, D.K. Oh, K.S. Kim, D. Kim, J. Phys. Chem. B 110 (2006) 6699. [95] Y.C. Zhang, W.W. Chen, X.Y. Hu, Cryst. Growth Des. 7 (2007) 580. [96] R.M. Ma, L. Dai, H.B. Huo, W.Q. Yang, G.G. Qin, P.H. Tan, C.H. Huang, J. Zheng, Appl. Phys. Lett. 89 (2006) 203120. [97] H. Yu, J. Li, R.A. Loomis, L.W. Wang, W.E. Buhro, Nat. Mater. 2 (2003) 517. [98] C.B. Murray, D.J. Norris, M.G. Bawendi, J. Am. Chem. Soc. 115 (1993) 8706. [99] M. Schierhorn, S.W. Boettcher, S. Kraemer, G.D. Stucky, M. Moskovits, Nano Lett. 9 (2009) 3262.

J. Jie et al. [100] H.Y. Sun, X.H. Li, Y. Chen, D.F. Guo, Y.W. Xie, W. Li, B.T. Liu, X.Y. Zhang, Nanotechnology 20 (2009) 425603. [101] C. Ma, Y. Ding, D. Moore, X.D. Wang, Z.L. Wang, J. Am. Chem. Soc. 126 (2004) 708. [102] X.F. Duan, C.M. Lieber, Adv. Mater. 12 (2000) 298. [103] C.X. Shan, Z. Liu, S.K. Hark, Appl. Phys. Lett. 87 (2005) 163108. [104] Z.Y. Wang, Q.F. Lu, X.S. Fang, X.K. Tian, L.D. Zhang, Adv. Funct. Mater. 16 (2006) 661. [105] R. Venugopal, P.-I. Lin, C.-C. Liu, Y.-T. Chen, J. Am. Chem. Soc. 127 (2005) 11262. [106] J. Giblin, V. Protasenko, M. Kuno, ACS Nano. 3 (2009) 1979. [107] L. Ouyang, K.N. Maher, C.L. Yu, J. McCarty, H. Park, J. Am. Chem. Soc. 129 (2007) 133. [108] X.H. Sun, T.K. Sham, R.A. Rosenberg, G.K. Shenoy, J. Phys. Chem. C 111 (2007) 8475. [109] R.A. Rosenberg, G.K. Shenoy, X.H. Sun, T.K. Sham, Appl. Phys. Lett. 89 (2006) 243102. [110] G.Z. Dai, Z.W. Peng, Q.L. Zhang, W.C. Zhou, M.X. Xia, H.X. Li, A.L. Pan, Q. Wan, B.S. Zou, Nanotechnology 20 (2009) 125601. [111] J.H. Yu, X.Y. Liu, K.E. Kweon, J. Joo, J. Park, K.-T. Ko, D.W. Lee, S.P. Shen, K. Tivakornsasithorn, J.S. Son, J.-H. Park, Y.W. Kim, G.S. Hwang, M. Dobrowolska, J.K. Furdyna, T. Hyeon, Nat. Mater. 9 (2010) 47. [112] Z.B. He, J.S. Jie, W.J. Zhang, W.F. Zhang, L.B. Luo, X. Fan, G.D. Yuan, I. Bello, S.T. Lee, Small 5 (2009) 345. [113] T. Mahalingam, V.S. John, S. Rajendran, P. Sebastian, J. Semiconduct. Sci. Technol. 17 (2002) 465. [114] D.W. Kisker, J. Cryst. Growth 98 (1989) 127. [115] Y.D. Li, Y. Ding, Z.Y. Wang, Adv. Mater. 11 (1999) 847. [116] L. Li, Y.W. Yang, X.H. Huang, G.H. Li, L.D. Zhang, J. Phys. Chem. B 109 (2005) 12394. [117] F. Wang, A. Dong, J. Sun, R. Tang, H. Yu, W.E. Buhro, Inorg. Chem. 45 (2006) 7511. [118] Y.-W. Jun, C.-S. Choi, J. Cheon, Chem. Commun. 1 (2000) 101. [119] Q.F. Meng, C.B. Jiang, S.X. Mao, J. Cryst. Growth 310 (2008) 4481. [120] E. Janik, J. Sadowski, P. Dłu˙zewski, S. Kret, L.T. Baczewski, A. Petroutchik, E. Łusakowska, J. Wróbel, W. Zaleszczyk, G. Karczewski, T. Wojtowicz, A. Presz, Appl. Phys. Lett. 89 (2006) 133114. [121] H. Kirmse, W. Neumann, S. Kret, P. Dłu˙zewski, E. Janik, G. Karczewski, T. Wojtowicz, Phys. Stat. Sol. (c) 5 (2008) 3780. [122] J. Zhang, P.-C. Chen, G.Z. Shen, J.B. He, A. Kumbhar, C.W. Zhou, J.Y. Fang, Angew. Chem. Int. Ed. 47 (2008) 9469. [123] Y.L. Cao, Y.B. Tang, Y. Liu, Z.T. Liu, L.B. Luo, Z.B. He, J.S. Jie, R. Vellaisamy, W.J. Zhang, C.S. Lee, S.T. Lee, Nanotechnology 20 (2009) 455702. [124] H.B. Huo, L. Dai, C. Liu, L.P. You, W.Q. Yang, R.M. Ma, G.Z. Ran, G.G. Qin, Nanotechnology 17 (2006) 5912. [125] Q.F. Meng, C.B. Jiang, S.X. Mao, Appl. Phys. Lett. 94 (2009) 043111. [126] W.I. Park, H.S. Kim, S.Y. Jang, J. Park, S.Y. Bae, M. Jung, H. Lee, J. Kim, J. Mater. Chem. 18 (2008) 875. [127] Z.Y. Tang, N.A. Kotov, M. Giersig, Science 297 (2002) 237. [128] I. Gur, N.A. Fromer, M.L. Geier, A.P. Alivisatos, Science 310 (2005) 462. [129] S. Neretina, R.A. Hughes, J.F. Britten, N.V. Sochinskii, J.S. Preston, P. Mascher, Nanotechnology 18 (2007) 275301. [130] S.M. Zhou, X.H. Zhang, X.M. Meng, S.K. Wu, S.T. Lee, Appl. Phys. A 81 (2005) 1647. [131] T. Yokogawa, T. Ishikawa, J.L. Merz, T. Taguchi, J. Appl. Phys. 75 (1994) 2189. [132] A. Pan, H. Yang, R. Liu, R. Yu, B. Zou, Z. Wang, J. Am. Chem. Soc. 127 (2005) 15692.

One-dimensional II—VI nanostructures [133] Y.-J. Choi, I.-S. Hwang, J.-H. Park, S. Nahm, J.-G. Park, Nanotechnology 17 (2006) 3775. [134] Y.L. Kim, J.H. Jung, K.H. Kim, H.S. Yoon, M.S. Song, S.H. Bae, Y. Kim, Nanotechnology 20 (2009) 095605. [135] A. Pan, R. Liu, F. Wang, S. Xie, B. Zou, M. Zacharias, Z. Wang, J. Phys. Chem. B 110 (2006) 22313. [136] T.Y. Lui, J.A. Zapien, H. Tang, D.D.D. Ma, Y.K. Liu, C.S. Lee, S.T. Lee, S.L. Shi, S.J. Xu, Nanotechnology 17 (2006) 5935. [137] W. Chen, D.C. Qi, X.Y. Gao, A.T.S. Wee, Prog. Surf. Sci 84 (2009) 279. [138] J.S. Jie, W.J. Zhang, K.Q. Peng, G.D. Yuan, C.S. Lee, S.T. Lee, Adv. Funct. Mater. 18 (2008) 3251. [139] J.S. Jie, W.J. Zhang, Y. Jiang, S.T. Lee, Appl. Phys. Lett. 89 (2006) 133118. [140] J.S. Jie, W.J. Zhang, Y. Jiang, S.T. Lee, Appl. Phys. Lett. 89 (2006) 223117. [141] A. Khandelwak, D. Jena, J.W. Grebinski, K.L. Hull, M.K. Kuno, J. Electron. Mater. 35 (2006) 170. [142] A. Javey, H. Kim, M. Brink, Q. Wang, A. Ural, J. Guo, P. Mcintyre, P. Mceuen, M. Lundstrom, H.J. Dai, Nat. Mater. 1 (2002) 241. [143] B. Xiang, P.W. Wang, X.Z. Zhang, S.A. Dayeh, D.P.R. Aplin, C. Soci, D.P. Yu, D.L. Wang, Nano Lett. 7 (2007) 323. [144] J.-M. Lee, I.-T. Cho, J.-H. Lee, W.-S. Cheong, C.-S. Hwang, H.-I. Kwon, Appl. Phys. Lett. 94 (2009) 222112. [145] X.H. Zhang, B. Domercq, X.D. Wang, S. Yoo, T. Kondo, Z.L. Wang, B. Kippelen, Orga. Electron. 8 (2007) 718. [146] Y.K. Park, A. Umar, S.H. Kim, J.H. Kim, E.W. Lee, M. Vaseem, Y.B. Hahn, J. Nanosci. Nanotechnol. 8 (2008) 6010. [147] P.C. Wu, R.M. Ma, C. Liu, T. Sun, Y. Ye, L. Dai, J. Mater. Chem. 19 (2009) 2125. [148] L.J. Lauhon, M.S. Gudiksen, D. Wang, C.M. Lieber, Nature 420 (2002) 57. [149] H.T. Ng, J. Han, T. Yamada, P. Nguyen, Y.P. Chen, M. Meyyappan, Nano Lett. 4 (2004) 1247. [150] N. Singh, A. Agarwal, L.K. Bera, T.Y. Liow, R. Yang, S.C. Rustagi, C.H. Tung, R. Kumar, G.Q. Lo, N. Balasubramanian, D.L. Kwong, IEEE Electron. Dev. Lett. 27 (2006) 383. [151] K. Keem, D.Y. Jeong, S. Kim, M.S. Lee, I.S. Yeo, U.I. Chung, J.T. Moon, Nano Lett. 6 (2006) 1454. [152] S. Ju, K. Lee, D.B. Janes, M.H. Yoon, A. Facchetti, T.J. Marks, Nano Lett. 5 (2005) 2281. [153] D. Wu, Y. Jiang, L. Wang, S.Y. Li, B. Wu, Y.Q. Yu, C.Y. Wu, Z.B. Wang, J.S. Jie, Appl. Phys. Lett. 96 (2010) 123118. [154] Z.B. He, W.J. Zhang, W.F. Zhang, J.S. Jie, L.B. Luo, G.D. Yuan, J.X. Wang, C.M.L. Wu, I. Bello, C.S. Lee, S.T. Lee, J. Phys. Chem. C 114 (2010) 4663. [155] W.W. Hooper, W.I. Lehrer, Proc. IEEE 55 (1967) 1237. [156] J.-R. Kim, H. Oh, H.M. So, J.-J. Kim, C.J. Lee, S.C. Lyu, Nanotechnology 13 (2002) 701. [157] Y.W. Heo, L.C. Tien, D.P. Norton, S.J. Pearton, B.S. Kang, F. Ren, J.R. LaRoche, Appl. Phys. Lett. 85 (2004) 3107. [158] C.S. Lao, J. Liu, P. Gao, L. Zhang, D. Davidovic, R. Tummala, Z.L. Wang, Nano Lett. 6 (2006) 263. [159] W.I. Park, J.S. Kim, G.-C. Yi, H.-J. Lee, Adv. Mater. 17 (2005) 1393. [160] R.M. Ma, L. Dai, G.G. Qin, Nano Lett. 7 (2007) 868. [161] R.M. Ma, L. Dai, H.B. Huo, W.J. Xu, G.G. Qin, Nano Lett. 7 (2007) 3300. [162] Y. Ye, L. Dai, P.C. Wu, C. Liu, T. Sun, R.M. Ma, G.G. Qin, Nanotechnology 20 (2009) 375202. [163] P.C. Wu, Y. Ye, T. Sun, R.M. Peng, X.N. Wen, W.J. Xu, C. Liu, L. Dai, ACS Nano. 3 (2009) 3138.

335 [164] X. Fang, Y. Bando, M. Liao, U.K. Cautam, C. Zhi, B. Dierre, B. Liu, T. Zhai, T. Sekiguchi, Y. Koide, D. Golberg, Adv. Mater. 21 (2009) 2034. [165] X. Fang, S. Xiong, T. Zhai, Y. Bando, M. Liao, U.K. Gautam, Y. Koide, X. Zhang, Y. Qian, D. Golberg, Adv. Mater. 21 (2009) 5016. [166] Y.G. Wang, B.S. Zou, T.H. Wang, N. Wang, Y. Cai, Y.F. Chan, S.X. Zhou, Nanotechnology 17 (2006) 2420. [167] B. Liu, Y. Bando, M. Wang, C. Zhi, X. Fang, C. Tang, M. Mitome, D. Golberg, J. Appl. Phys. 106 (2009) 034302. [168] S.X. Xhou, Y.G. Wang, W. Han, N. Wang, J. Phys. Chem. C 112 (2008) 18644. [169] J.H. He, Y.Y. Zhang, J. Liu, D. Moore, G. Bao, Z.L. Wang, J. Phys. Chem. C 111 (2007) 12152. [170] J. Salfi, U. Philipose, C.F.D. Sousa, S. Aouba, H.E. Ruda, Appl. Phys. Lett. 89 (2006) 261112. [171] J.S. Jie, W.J. Zhang, Y. Jiang, X.M. Meng, Y.Q. Li, S.T. Lee, Nano Lett. 6 (2006) 1887. [172] R. Calarco, M. Marso, T. Richter, A.I. Aykanat, R. Meijers, A.V.D. Hart, T. Stoica, H. Lüth, Nano Lett. 5 (2005) 981. [173] Y. Jiang, W.J. Zhang, J.S. Jie, X.M. Meng, X. Fan, S.T. Lee, Adv. Funct. Mater. 17 (2007) 1795. [174] D. Wu, J.S. Jie, L. Wang, Y.Q. Yu, Y. Jiang (unpublished). [175] T.Y. Wei, C.T. Huang, B.J. Hansen, Y.F. Lin, L.J. Chen, Appl. Phys. Lett. 96 (2010) 013508. [176] O. Hayden, R. Agarwal, C.M. Lieber, Nat. Mater. 5 (2006) 352. [177] X.F. Duan, Y. Huang, Y. Cui, J. Wang, C.M. Lieber, Nature 409 (2001) 66. [178] Y. Huang, X.F. Duan, C.M. Lieber, Small 1 (2005) 142. [179] Y. Li, F. Qian, J. Xiang, C.M. Lieber, Mater. Today 9 (2006) 18. [180] Z. Fan, J.C. Ho, Z.A. Jacobson, R. Yerushalmi, R.L. Alley, H. Razavi, A. Javey, Nano Lett. 8 (2008) 20. [181] Z. Fan, J.C. Ho, T. Takahashi, R. Yerushalmi, K. Takei, A.C. Ford, Y-L. Chueh, A. Javey, Adv. Mater. 21 (2009) 1. Jiansheng Jie is a professor at Hefei University of Technology. He obtained his PhD in University of Science and Technology of China in 2004 and spent the subsequent three years as a postdoctoral fellow at City University of Hong Kong. He joined the Hefei University of Technology as a professor since 2006. His research interest includes the novel semiconductor nanostructures, transport properties of nanomaterials, nanoelectronic and nanooptoelectronic devices, and nanodevice assembly and integration for large scale applications. Wenjun Zhang obtained his PhD degrees in 1994 from Lanzhou University, PR China. He was a post doc at the Fraunhofer Institute for Surface Engineering and Thin Films (FhG-IST), Germany, from 1995 to 1997, and then spent one year at the City University of Hong Kong as a research fellow. From 1998 to 2000, he worked as a recipient of Science and Technology Agency (STA) Fellowship at National Institute for Research in Inorganic Materials (NIRIM), Japan. He joined City University of Hong Kong in December 2000 again as a senior research fellow. He is currently an associate professor in The Department of Physics and Materials Science; and he is also a core member of the Center Of Super Diamond and Advanced Films (COSDAF). His

336 research focuses on thin films, semiconducting nanomaterials, surface science and modification, and ions/materials interactions. He has published over 160 referred articles and holds 5 patents. He received the Japan Society of Applied Physics (JSAP) Best Paper Award in 2002, and the Friedrich Wilhem Bessel Research Award of Alexander von Humboldt Foundation, Germany, in 2003. Bello is professor in Physics and Materials Science at City University Hong Kong. He is a founding core member of the Center of Super-diamond and Advanced Films (COSDAF) and the Advanced Coatings Applied Research Laboratory (ACARL) in Hong Kong. His research interest involves thin films, particularly diamond, cubic boron nitride and related materials, semiconducting nanomaterials and photovoltaic and light emitting devices. He has published over 310 referred articles and holds 13 patents. He has been invited speaker in many world conferences particularly referring to diamond and related materials. He has been adjunct professor in Condense Matters and Materials Engineering, and Industrial Consultant, at the University of Western Ontario, Canada. Before arriving to Canada he was associated professor at the Microelectronic Department of the Slovak University of Technology and Vice-chairman of the Czechoslovak Expert Assembly for Vacuum Technology and Applications. He obtained Leverhulme Fellowship in England. He earned PhD and masters in microelectronics focusing on the technology of device fabrication particularly ion implantation and plasma processes.

J. Jie et al. Chun-Sing Lee obtained his doctor of Philosophy degree in 1991 from the University of Hong Kong. He then moved to the University of Birmingham to carry out postdoctoral research with the support of a Croucher Foundation Fellowship. He jointed the faculty of the City University of Hong Kong in 1994 and is currently a chair professor in materials science. He co-founded the Center Of Super-Diamond and Advanced Films (COSDAF) in 1998 and is the center’s Associate Director. Prof Lee has published over 400 journal papers, which received over 10,000 citations. His main research interest is on surface and interface physics, organic electronics and nanomaterials. Shuit-Tong Lee is a member (academician) of Chinese Academy of Sciences (CAS) and fellow of the Academy of Sciences for the Developing World (TWAS). He is a chair professor of Materials Science, and the founding director of the Center of Super-Diamond and Advanced Films (COSDAF) at City University of Hong Kong. He is also the founding director of Nano-Organic Photoelectronic Laboratory at the Technical Institute of Physics and Chemistry, CAS, and Institute of Functional Nano & Soft Materials (FUNSOM) at Soochow University. His major research areas include organic light-emitting diode (OLED) display technologies, functional nanomaterials and devices, nano-biosensors, as well as diamond and other super-hard thin film technologies.