Nanostructures: The Influence of

Apr 9, 2013 - (13) However, the theme of other works on ZnTe nanobelts is still limited ..... the New Initiative Fund (M58110100) from Nanyang Technol...
0 downloads 0 Views 4MB Size
Article pubs.acs.org/crystal

The Growth of Ultralong ZnTe Micro/Nanostructures: The Influence of Polarity and Twin Direction on the Morphogenesis of Nanobelts and Nanosheets Muhammad Iqbal Bakti Utama,†,# Maria de la Mata,‡,# Qing Zhang,† Cesar Magen,§ Jordi Arbiol,*,‡,∥ and Qihua Xiong*,†,⊥ †

Division of Physics and Applied Physics, School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore 637371 ‡ Institut de Ciència de Materials de Barcelona, ICMAB-CSIC, E-08193 Bellaterra, CAT, Spain § Laboratorio de Microscopías Avanzadas (LMA), Instituto de Nanociencia de Aragon (INA) - ARAID and Departamento de Fisica de la Materia Condensada, Universidad de Zaragoza, 50018 Zaragoza, Spain ∥ Institució Catalana de Recerca i Estudis Avancats (ICREA), 08010 Barcelona, CAT, Spain ⊥ Division of Microelectronics, School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore 639798 S Supporting Information *

ABSTRACT: Although ZnTe nanobelts present an intriguing platform to study various optical properties and phenomena in semiconductors, there was very limited study regarding the crystalline structure and defects of ZnTe nanobelts. Here, we correlate the structural properties and features in the crystal of ZnTe nanobelts with the resulting as-synthesized morphology. Ultralong ZnTe nanobelts were synthesized to reach the subcentimeter length scale. Two types of nanobelts were identified according to whether tapering was present and discerned on the basis of crystallinity and polarity of the structure. We conclude that tapered sheetlike nanobelts have Te-terminated lateral facets that induced lateral growth, whereas untapered nanobelts have facets that are nonpolar and nonreactive. Axial and transversal twins were also observed, where the polarity was conserved across twinning boundaries. ZnTe nanobelts commonly possess p-type conductivity,12 which makes ZnTe attractive for optoelectronic devices in the green wavelength. We recently showed that ZnTe nanobelts have excellent optical properties and are an interesting platform to study electron−phonon interactions.13 However, the theme of other works on ZnTe nanobelts is still limited to the synthesis and electrical transport measurements.12,14−19 Details on the morphogenesis of ZnTe nanobelts in relation to their crystallinity, presence of planar defects, such as twins, and atomic polarity are not yet reported. We believe that the understanding of the structural aspects of ZnTe nanobelts is critical to achieving reliable and reproducible device fabrication and applications based on ZnTe nanobelts. Hence, we report herein the synthesis of ZnTe nanobelts and the characterizations of their crystallinity and polarity. We aim to correlate the structural properties and features of the nanobelts with the morphology, in the context of the synthesis environment during the growth. Specifically, we will show herein that ZnTe

1. INTRODUCTION Nanobelts, or also referred to as “nanoribbons”, are quasi onedimensional (1D) nanostructures with a high aspect ratio that have an unconfined length and a rectangular-like cross-section, with a thickness being in the nanometer scale.1 Compared to their bulk counterpart, nanostructures, such as nanobelts from semiconductor compounds, are relatively easier to produce in good crystallinity while improving the tunability of the material (electrical, optical, mechanical, or chemical) properties of the compound.2,3 Nanobelts have thus deservedly become a focal point of intensive research in the past decade owing to their potential applications in optoelectronics.4 For example, among the materials that can be synthesized into nanobelts, proof-ofconcept applications from II−VI semiconductor nanobelts are abundant, such as for field emitters, solar cells, photodetectors, and devices to probe emergent nanoscale phenomena.5−10 Recent breakthroughs of laser cooling of semiconductors using CdS nanobelts further promulgate the applicability of such nanomorphology owing to their strong exciton−longitudinal optical phonon coupling strength.11 Zinc telluride (ZnTe) is a II−VI semiconductor with a 2.26 eV direct band gap at room temperature. Besides, as-grown © 2013 American Chemical Society

Received: March 6, 2013 Revised: April 8, 2013 Published: April 9, 2013 2590

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596

Crystal Growth & Design

Article

nanobelts may contain axial and transversal twins. We will then discuss the influence of the twins and the atomic polarity on the emergence of two different shapes of nanobelts.

2. EXPERIMENTAL SECTION The synthesis of the nanobelts was conducted with a home-built vapor transport thermal evaporation setup using a single zone furnace (Lindberg/Blue M TF55035C-1), which we have reported to be effective in the growth of nanostructures.20,21 The source material was high-purity ZnTe powder (99.99%, Alfa Aesar), and the substrate is a ⟨100⟩ p-Si chip with native oxide. The substrate was coated using a thermal evaporator (Elite Engineers) with 6 nm thick gold (Au) film to act as the catalyst. The source powder was put at the center of the furnace, whereas the substrate was positioned in the downstream region. The carrier gas was 30 sccm Ar with 5% H2. The central temperature of the furnace (850 °C) and the pressure inside the reactor (50 Torr) were set and stabilized for 90 min. The as-grown sample was characterized by field emission scanning electron microscopy (FE-SEM, JEOL JSM-7001F), and X-ray powder diffraction (XRD, Bruker D8 advanced diffractometer, Cu Kα radiation) in θ−θ geometry. The segment of the as-grown sample containing nanobelts was ultrasonicated in isopropanol to separate the nanobelts from the substrate. The resulting suspension was drop-casted to a clean Si chip for Raman spectroscopy and atomic force microscopy (AFM, Park Systems NX10), and drop-casted to a lacey carbon grid (Electron Microscopy Sciences) for scanning and high-resolution transmission electron microscopy (STEM and HRTEM, respectively). The Raman spectroscopy was conducted on an individual nanobelt at room temperature using a micro-Raman spectrometer (Horiba-JY T64000) in a backscattering geometry. The backscattered signal was collected by a 100× objective and dispersed by a 1800 g/mm grating in a single mode with a spectral resolution of ∼1 cm−1, and recorded by a liquidnitrogen-cooled charge-coupled device detector. The nanobelts in the lacey carbon grid were observed by means of HRTEM (JEOL 2010F with a field emission gun operated at 200 kV) and aberration-corrected high-angle angular dark-field STEM (HAADF-STEM; FEI Titan 60− 300 keV, operated at 300 kV).

Figure 1. As-grown ZnTe sample. (a) Photograph of the sample with visible ultralong nanostructures next to a centimeter-scale ruler. The white arrow denotes the direction of the increase in local temperature during growth. (b) XRD pattern of the sample in the semilogarithmic scale. (c−e) SEM images of nanostructures found on the sample corresponding to different positions of the substate, as shown in differently colored arrows in (a): (c, blue) kinked nanowires; (d, green) ultralong nanowires; (e, red) ultralong nanobelts. Inset in (e): edge-on view of a nanobelt. (f) Zoomed-in SEM images of an ultralong wire, showing the tip and midregion of the wire.

The trend in the dependence of the ZnTe nanostructure morphology to the local substrate temperature is welldocumented and in agreement with the previously published literature.14,15,18 The most commonly accepted explanation of such an observation relates the local growth temperature to the extent of the accompanying lateral broadening of nanobelts via the vapor−solid (VS) mechanism.22 The high local temperature promotes 2D nucleation in the lateral facet of the structure with the VS mechanism alongside the axial elongation from the VLS mechanism, such that nanobelts will be formed. On the other hand, lateral growth at lower temperature is not as extensive, which promotes the growth of nanowires instead. In accordance to this conclusion, we have also reported elsewhere the growth of CdS nanobeltsgrown with similar methods with the present ZnTe nanobeltswhere the lateral broadening of the structure has a noticeable positive correlation with the local temperature of the substrate.23 Similarly, nanobelts from other compounds, such as ZnS24 and Al2O3,25 are also observed to grow at a higher local growth temperature than that for nanowire growth. An inset in Figure 1e clarifies the morphological distinction between nanobelt and nanowire: while nanobelts have a submicrometer level thickness that is comparable to the diameter of micro/nanowires, the width of nanobelts could reach tens of micrometers. Meanwhile, the structure of a wire is elucidated in greater detail with zoomed-in images in Figure 1f. At the tip of most micro/nanowires (and also nanobelts; see Figures 3e and 4b), we observed a spherical particle that might be composed of the coalesced Au film from the substrate due to the high temperature during the synthesis process; this result is typical of a catalyzed nanowire in accordance to vapor−liquid− solid (VLS) growth.26 On the other hand, nanowiresor even nanobeltswere not produced when no Au film was coated on the substrate, indicating the catalytic function of the Au particle.

3. RESULTS AND DISCUSSION Figure 1 shows an overview of our synthesis result. The asgrown sample carried very long structures that are visible to the naked eye (Figure 1a). The XRD pattern from the sample (Figure 1b) reveals intense peaks that can be assigned to diffraction from a ZnTe crystal in cubic zincblende (ZB) phase and the Si substrate. We also observed that the sample is divided into three segments with a discernible appearance, as has been marked with three differently colored arrows in Figure 1a. Interestingly, the three segments contained vastly different morphologies of nanostructures, as observed with SEM: The left region marked with a blue arrow, which is the downstream region with a lower local temperature during the synthesis, has kinked micro/nanowires (Figure 1c); the middle region with a green arrow, which occupied a relatively higher local synthesis temperature, has randomly oriented long micro/nanowires (Figure 1d); and the right region marked with a red arrow, which is the high-temperature upstream region, has very long nanobelts (Figure 1e). Given that the location at which the different structures are present is already distinct, it is thus possible to isolate a specific structure that one desires by simply cutting the respective segment of the substrate (originally ∼3 cm long). For example, one could cut the first 0.5 cm of the upstream region of the substrate to get nanobelts with a negligible quantity of wires or kinked wires. Alternatively, a smaller piece of substrate can be used such that only a specific structure will be obtained, separate from other structures. 2591

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596

Crystal Growth & Design

Article

3d). Moreover, the overall features of spectra from both morphologies are identical. Both spectra show seven peaks

Whereas the body of the wire close to the Au particle appears to be cylindrical, the region further away from the particle exhibited a faceted morphology with an aperiodic segment length. Radial growth is also suggestive, as the diameter far from the Au particle at the tip is much larger than that near the tip. Such “microfaceted” morphology is very well-known in multitwinned wires with the ZB phase grown in the ⟨111⟩ direction:27 microfacets in nanowires are caused by the emergence of segments in the shape of truncated octahedron with {111} facets;28 a twin caused a 180° change in the orientation of microfacets. Thus, nucleation of multiple twin planar defects caused alternating microfaceted segments, as seen in Figure 1f. Readers are referred to our previous work on the structure of ZnTe nanorods29 and the other relevant literature for more in-depth discussions regarding twinning phenomena in nanowires.30−34 Overall, the nanostructures in all segments of the as-grown sample can be grown to reach the centimeter length scale, which we believe could still be lengthened if the growth time was increased with an adequate amount of source powder. The high-yield synthesis of ultralong ZnTe 1D nanostructures presents an opportunity for a large upscaled production of the material, which will be suitable for industrial purposes. Inspection of as-transferred single nanobelts (Figure 2a) reveals that the ZnTe nanobelts could still be categorized into

Figure 3. Characterizations of nanobelts. (a) Mosaic optical micrograph of an ultralong nanobelt. (b, c) AFM images of the left and right part of the nanobelt in (a), respectively. The height profiles are superimposed in the respective images. (d) RRS spectra with 532 nm excitation from three different locations of the nanobelt: left, middle, and right, as illustrated by differently colored circles in (a). (e) STEM image of a nanobelt. (f) EDX compositional mapping of Zn, Te, and Au edges near the nanobelt tip as marked with a yellow square in (e). (g, h) HRTEM images of a nanobelt, revealing the presence of nearly periodic twins. White arrows in (e, g, h) denote the growth direction of the nanobelt. (i) FFT from the nanobelt in (h).

located at 206, 408, 615, 823, 1029, 1223, and 1440 cm−1 (albeit weakly) that arein averagespaced by (206 ± 7) cm−1, which is the phonon frequency of the longitudinal optical (LO) phonon of ZB ZnTe. Thus, we assign those seven peaks to the nth order (n = 1−7) LO Raman scattering, respectively. We believe that the ability to observe the high-frequency, seventh order LO Raman scattering is partly enabled due to the high quality of our samples; no other groups have reported the high-order phonon observation in ZnTe nanobelts yet. Detailed discussions regarding the implication of the multiphonon Raman spectroscopy in ZnTe nanobelts and the time-resolved PL characterization of the structure have been published elsewhere.13 Meanwhile, the peak at 125 cm−1 can be assigned to the mode from crystalline Te phase, similar to that observed in ZnTe and CdTe nanorods,35,36 and the peak at 520 cm−1 originates from the Si substrate. Figure 3a shows a representative appearance of an ultralong ZnTe nanobelt with a length of ∼600 μm after being transferred to a Si chip. The ultrasonication process to remove the nanobelt from the substrate may also break the nanobelt, shortening the nanobelt from its as-grown length. AFM imaging of the nanobelts confirmed the rectangular cross section of the nanobelt (Figure 3b,c). The lateral encroachment in the nanobelt served as the distinction of a nanobelt from a nanowire (Figure 1f). The cross-sectional shape of the nanobelt throughout its length was fairly uniform, where the thickness remained in the same order (cf. Figure 3b,c, which were acquired a few hundred micrometers apart), unlike nanowires,

Figure 2. (a) Optical micrographs of nanobelts and nanosheets. (b) RRS spectra of nanobelts and nanosheets with 514 nm excitation.

two groups: (1) nanobelts with the classically expected morphology of a 1D nanostructure with parallel sides and (2) nanobelts with a sheetlike morphology that has pronounced tapering and 2D growth. The strongly tapered nanobelts will be called “nanosheets” hereafter. We then conducted intensive characterizations on the structural and crystalline properties for each of these two groups of ZnTe nanobelts, which are the structures-of-interest in this work, to provide an understanding of the difference and morphogenesis of the two morphologies. In terms of the optical properties, we characterized both nanobelts and nanosheets with the resonant Raman scattering (RRS) spectroscopy. RRS was shown to be effective in probing the electronic structure and electron−phonon coupling in ZnTe nanorods, where strong coupling was observed.13 Under a 514 nm laser excitation (Figure 2b), the most noticeable difference between the RRS spectrum of nanobelts and nanosheets is in terms of their photoluminescence (PL) intensity, in which the PL of nanobelts is stronger. However, the PL in both spectra had a very similar line shape and was peaked at the identical position (546 nm ≈ 2.27 eV; cf. Figure 2592

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596

Crystal Growth & Design

Article

since ZnTe is more thermodynamically stable in the ZB phase rather than the WZ phase.40 Nanobelts from ZnTe are thus relatively unique since nanobelts from other II−VI, III−V, and oxide compounds are either polytypic or growing in the WZ phase completely. Instead, we observed the coexistence of two ZB patterns, namely, in [1̅1̅0]ZB and [110]ZB zone axes. We remark that the observation of the sample in the [110] zone axis of the crystal is a reliable and unambiguous method to verify that the crystal is indeed of ZB phase instead of WZ phase. Should the crystal be in WZ phase, one would observe no [110] zone axis and instead the crystal will exhibit a [112̅0] zone axis behavior at the same direction. The primary reason in observing the sample in the [110] or [112̅0] zone axis is because the atomic stacking arrangement between ZB and WZ phases (ABCABC and ABABAB, respectively) can be conclusively distinguished at those zone axes. Given the difference in the stacking arrangement in these zone axes, there will be an associated difference in the diffraction pattern produced by the samples in those zone axes. Indeed, the patterns match well with a ZB pattern in the [110] zone axis with twinning features instead of a WZ pattern in the [112̅0] zone axis. In the indexing, we also use the finding in our previous work,21,29 where we established that Te2−-terminated (anion) planes of ZB ZnTe (i.e., 1̅1̅1̅, 1̅11, 11̅1, 111̅) are more reactive than are Zn2+-terminated (cation) planes (i.e., 111, 111̅ ,̅ 11̅ 1,̅ 1̅1̅1). Thus, ZnTe nanostructures, such as nanorods and nanotripods, grown along the polar direction will select the Tedirected −⟨111⟩, which has a higher growth rate than Zndirected +⟨111⟩. We, therefore, conclude that the present nanobelt is growing in the [11̅1] direction with {111} twin planes. The annotated STEM in Figure 4a summarizes our characterizations of the nanosheet-like morphology of ZnTe as we shall substantiate subsequently. In contrast to the nanobelt, the tapering of the nanosheet was evident. We also see that the tip of the nanosheet is also occupied by a particle (Figure 4b). Compositional mapping with EDX showed that the particle is mainly composed of Au, suggesting a Aucatalyzed VLS growth mechanism. We also noticed small residual traces of Zn that were mainly localized on the tip surface. Meanwhile, the nanosheet is composed of Zn and Te in a 1:1 ratio. The FFT from the HRTEM image of the nanosheet showed that the nanosheet contained the signature of twin features, showing two sets of ZB diffraction spots in [1̅1̅0]ZB and [110]ZB zone axes each (Figure 4c). However, dissimilar to the case in the nanobelt, the growth direction of the nanosheet is along the nonpolar [11̅2̅] direction, with the lateral facet on the righthand side in Figure 4a being constructed from a {111} plane. We confirmed the presence of twins with an atomic resolution HAADF-STEM study of the nanosheet, which revealed the presence of axial {111} twin boundaries (Figure 4d). The twin boundaries originated at the stepped Au particle−nanobelt interface and propagated through the entire length of the nanosheet. Here, the diffraction pattern from the particle at the tip (Figure 4d, inset) has been indexed to Au along the [110] zone axis, which is in line with the [110]ZB zone axis of the ZnTe nanosheet. We noted that the size of the Au particle (∼200 nm) was similar to that found at the tip of the nanobelt (Figure 3f), which is reasonable since both structures are found at a similar local temperature regime during growth within the same substrate to allow coalescence of the Au film into isolated

which experienced a conspicuous radial growth. The RRS spectra taken with a 514 nm laser excitation from three different sites at the nanobelt (dashed circles in Figure 3a) exhibited nearly identical features. These results suggest that the properties of an ultralong nanobelt are consistent over a large length scale. The height profiles in Figure 3b,c show that the thickness of a long nanobelt that we observe is ∼300 nm. We remark that there is a significant number of similar structures within our samples that have dimensions within the micrometer-sized level, for which categorization as “microstructures” will be more appropriate (for instance, the wire in Figure 1f). However, given that the catalyst used in our synthesis is a Au film, the coalescence of the Au film at high temperature during the synthesis might result in a broader size distribution of Au catalyst particles. Thus, variation of sizes in nanowires/ nanobelts is to be expected as the broad distribution of the coalesced Au film dictates the diameter/thickness of nanowires/nanobelts in VLS growth. Correspondingly, there are structures within our sample with a thickness ≤ 100 nm, which can then be appropriately categorized as a “nanostructure” (see Figure S1 in the Supporting Information). We used HAADF-STEM to view the general morphology of a ZnTe nanobelt (Figure 3e). Similar to the instance of the nanowire shown in Figure 1f, we also observed a spherical particle at one end of the nanobelt (Figure 3e). Compositional mapping with energy-dispersive X-ray (EDX) was conducted in situ with the STEM to analyze the composition of the particle (Figure 3f). The spherical particle contained concentrated Au atoms and some Zn and Te constituents as well, with the Zn composition being slightly higher than Te. In comparison, almost no Au was present on the nanobelt, whereas Zn and Te were distributed evenly (in a 1:1 ratio) within the nanobelt. This result is in agreement with the VLS growth mechanism, where Au particles, which behaved similarly to a “catalyst” of the crystal growth, served as the preferential adsorption sites of vaporized precursor material.37 In the VLS model, the elevated temperature coalesces the Au film into liquid droplets and promotes alloying between the continuously adsorbed precursor vaporsZnTe in our case. Once the alloy is supersaturated, nucleation occurs at a single site and continued with 1D crystal growth of nanostructures. Although twin-free monocrystalline nanobelts were also observed, we found that the nanobelt morphology might contain nearly periodical twin planes that are perpendicular to the axis of the nanobelt growth direction (Figure 3g). Because of the twins, the nanobelt is divided into segments with alternating fringe contrast; we have placed pseudocolored twin domains side-by-side with the HRTEM image in Figure 3h for ease of visualization. The transversal twins in the nanobelt are spontaneously formed, as we did not adjust the growth environment manually during the growth process. Although transveral twins are common in nanobelts from other compounds,38 reports of twinned nanostructures in ZnTe were limited only to nanowires and tripods;29,39 virtually all other works on ZnTe nanobelts showed a monocrystalline structure without twins.12,14−19 The fast-Fourier transform (FFT) pattern from the nanobelt (Figure 3i) showed that the ZnTe nanobelt is of ZB phase, similar to the result in other works.12,14−19 Unlike our results in randomly twinned nanorods,29 we did not identify any pattern of the hexagonal wurtzite (WZ) phase in nanobelts nor there was any extended WZ phase. The absence of WZ is reasonable 2593

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596

Crystal Growth & Design

Article

surface, which can be assigned to (11̅ 1)̅ as is implied from the FFT indexing in Figure 4c. Meanwhile, the other lateral facet was inclined from the growth direction of the nanosheet, creating the tapered structure. The inclination angle was also changing along the length of the nanosheet, while remaining positive, which indicates that the lateral facet might have been an active growth front. As the polarity of ZnTe is conserved across the twin boundaries, we can then safely assign that the inclined lateral facet has a Te polarization. We clarify the influence of the polarity of the crystals to explain how nanobelts and nanosheets have a distinct morphology despite the fact that the axial growth in both structures is initiated by a similar VLS mechanism. To do so, we invoke the growth model as discussed for tapered ZnS nanobelt by Hao et al.32 to explain the presence of nanosheets. Although the ZnS nanobelt was partially in the WZ phase with the lateral expansion being in the ZB phase without any clarification on whether axial twinning was present, the nanobelt structure shared a similarity to our ZnTe nanosheet in which the lateral facets are polar surfaces. In applying the growth model to our case, we use our previous result in which the Te-terminated cation surface was assumed to be chemically and catalytically more active than that of the Zn-terminated surface.29 Under high supersaturation, the Te-terminated lateral facet could serve as a preferential diffusion direction of arriving adatoms other than the Au catalyst particle, whereas the Zn-terminated surface remained less reactive. Thus, the Te-terminated lateral facet in ZnTe becomes a “sink” of adatoms and becomes an active front for crystal growth that is concurrent with the axial elongation growth of the nanobelt. The lateral encroachment of a particular crystal plane closer to the particle is narrower than that further from the particle as a shorter time has elapsed ever since the plane is formed, resulting in a tapered structure. Here, we see that our assumption of a chemically active Teterminated surface is in agreement with the structural data from STEM (Figure 4a,d,e), which shows that the lateral facetthat is, Te-polarizedwas the facet experiencing lateral expansion. Interestingly, our results demonstrate that ZnTe is a unique compound since nanobelts with polar lateral facets from other II−VI compounds (e.g., CdO, CdS, CdSe, ZnS, ZnSe) would have the cation-terminated surface that is being chemically reactive,4 which would give rise to a sawlike structure at the cation-terminated lateral facet of the nanobelt.43 On the other hand, we observe that the influence of twin directions in the morphogenesis of the structures is rather subtle and nonobvious, yet significant, since the ZnTe nanobelts and nanosheets contain an appreciable number of twins. As shown in Figures 3g,h and 4d, the twin planes in the ZnTe is in the {111} planes, which are exactly parallel to (for nanobelts) or perpendicular to (for nanosheets) the growth plane. Moreover, all the twins that we observed (e.g., in Figure 4e) are of the orthotwin type, since the polarity of the crystal is conserved across the twin. Therefore, the direction of twins in the nanobelts and nanosheets is such that it is suitable to preserve the polar/nonpolarity of the lateral facets, which, in turn, will affect whether or not the structure is subjected to lateral growth (i.e., whether a nanobelt or nanosheet is being formed). Consider the case where there exists at least a twin plane that is located on a plane that is at another arbitrary angle that is neither perpendicular nor parallel to the growth plane. Such twin directions may change or reorient the growth direction of the nanobelts so as to create a kinked structure.44,

Figure 4. Characterizations of a nanosheet. (a) HAADF-STEM image. (b) Zoomed-in STEM image close to the gold-capped tip. Inset: EDX compositional mapping. (c) FFT of the nanosheet. (d) Highresolution HAADF-STEM image at the region in (b) indicated with an orange square. Yellow dashed lines mark the location of twin boundaries. Inset: FFT of gold particle. (e) High-magnification HAADF-STEM showing individual dumbbells close to the twin boundaries (yellow dashed line) shown in blue and purple squares in (d). The images are superimposed with balls representing Zn and Te atoms to clarify the dumbbell orientations around the twins.

particles of similar size during the annealing. Hence, we believe that the tapering of the nanosheet could not be attributed to the shrinkage of the Au particle catalyst during the growth. The Z-contrast imaging of STEM in HAADF mode made HAADF-STEM sensitive to sample composition, where the observed intensity from the sample image is approximately proportional to Z2. Hence, given that Te has a reasonably higher atomic number than Zn (ZTe = 52; ZZn = 30), we could distinguish the constituent atomic dumbbells of ZnTe via aberration-corrected HAADF-STEM, with Te atoms appearing brighter than Zn atoms. HAADF-STEM also allows determination of the stacking arrangement of the dumbbells and the orientation of the dumbbells without phase interference. The stacking arrangement of dumbbells at a region away from the twin boundaries exhibits the ABCABC stacking that is a defining characteristic of the ZB phase. As was the case in nanorods,41 the presence of twin boundaries along the {111} direction changes the dumbbell orientation. However, the polarity of the dumbbells is still conserved across the twin boundaries. The ZnTe nanosheet also adhered to similar characteristics; details of the dumbbell orientation in the vicinity of twin boundaries are shown in the zoomed-in image in Figure 4e. The atomic resolution magnified images in Figure 4e have been improved applying a probe deconvolution by using the STEM-CELL Software.42 The ZnTe dumbbells can be resolved well, with the heavier Te atoms appearing brighter than Zn atoms. Consequently, as seen in Figure 4a, the right-hand-side lateral facet that has a steady straight shape is a Zn-terminated 2594

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596

Crystal Growth & Design

Article

the Universitat de Barcelona and INA-LMA at the University of Zaragoza.

With such an unsuitable twin direction, it will be impossible to obtain an ultralong straight nanobelt as we show in our article.



4. CONCLUSIONS We have shown that ZnTe micro- and nanostructures can be synthesized in large quantities, with the length of the structures reaching the milli- and centimeter scales. We were able to distinguish two types of nanobelts: (1) nanobelts with parallel lateral facets and (2) tapered nanosheets, according to their structure and crystallinity. The two types of nanobelts had different growth directions and twin directions (w.r.t. the growth plane) that result in the different facet polarity characteristics, which we correlate with the lateral growth that is responsible for the tapering in nanosheets. We believe that our characterization methodologies and line of reasoning may also be extended to explain the morphological phenomenon in other nanobelt materials. We also believe that the possibility to tune the crystalline and morphological characteristics of ZnTe nanobelts may have potential applications to adjust the electronic and optoelectronic properties of the structure. For example, we may use the fact that a bilayer of the WZ phase is created at every twin boundary of the ZB crystal.21,45 The nearly periodic occurrence of twins in the ZnTe nanobelts can be considered to create periodic ZB−WZ polytypism in the crystal. Interestingly, theoretical prediction and experimental works on ZnTe show that the ZB and the less-stable WZ phases have dissimilar optoelectronic properties (e.g., band gap, carrier effective masses, and refractive index).46,47 By rationally adjusting the emergence and direction of the ZB−bilayer WZ heterojunction that is formed with the creation of twins, it is possible to engineer the band structure of the nanostructure for the creation of advanced electronic and optoelectronic structures, such as quantum wells or single electron transistor structures.45,48



ASSOCIATED CONTENT

S Supporting Information *

Additional AFM images and height profiles of ZnTe nanobelts. This material is available free of charge via the Internet at http://pubs.acs.org.



REFERENCES

(1) Pan, Z. W.; Dai, Z. R.; Wang, Z. L. Science 2001, 291, 1947. (2) Xia, Y. N.; Yang, P. D.; Sun, Y. G.; Wu, Y. Y.; Mayers, B.; Gates, B.; Yin, Y. D.; Kim, F.; Yan, Y. Q. Adv. Mater. 2003, 15, 353. (3) Utama, M. I. B.; Zhang, J.; Chen, R.; Xu, X.; Li, D.; Sun, H.; Xiong, Q. H. Nanoscale 2012, 4, 1422. (4) Ma, C.; Moore, D.; Ding, Y.; Li, J.; Wang, Z. L. Int. J. Nanotechnol. 2004, 1, 431. (5) Li, D.; Zhang, J.; Xiong, Q. H. ACS Nano 2012, 6, 5283. (6) Li, D.; Zhang, J.; Zhang, Q.; Xiong, Q. H. Nano Lett. 2012, 12, 2993. (7) Jie, J. S.; Zhang, W. J.; Jiang, Y.; Meng, X. M.; Li, Y. Q.; Lee, S. T. Nano Lett. 2006, 6, 1887. (8) Hosono, E.; Fujihara, S.; Honna, I.; Zhou, H. S. Adv. Mater. 2005, 17, 2091. (9) Fang, X. S.; Bando, Y.; Liao, M. Y.; Gautam, U. K.; Zhi, C. Y.; Dierre, B.; Liu, B. D.; Zhai, T. Y.; Sekiguchi, T.; Koide, Y.; Golberg, D. Adv. Mater. 2009, 21, 2034. (10) Zhao, L.; Hu, L.; Fang, X. S. Adv. Funct. Mater. 2012, 22, 1551. (11) Zhang, J.; Li, D.; Chen, R.; Xiong, Q. H. Nature 2013, 494, 504. (12) Zhang, J.; Chen, P. C.; Shen, G. Z.; He, J. B.; Kumbhar, A.; Zhou, C. W.; Fang, J. Y. Angew. Chem., Int. Ed. 2008, 47, 9469. (13) Zhang, Q.; Liu, X.; Utama, M. I. B.; Zhang, J.; de la Mata, M.; Arbiol, J.; Lu, Y.; Sum, T. C.; Xiong, Q. H. Nano Lett. 2012, 12, 6420. (14) Fasoli, A.; Colli, A.; Hofmann, S.; Ducati, C.; Robertson, J.; Ferrari, A. C. Phys. Status Solidi B 2006, 243, 3301. (15) Meng, Q. F.; Jiang, C. B.; Mao, S. X. J. Cryst. Growth 2008, 310, 4481. (16) Yim, J. W. L.; Chen, D.; Brown, G. F.; Wu, J. Q. Nano Res. 2009, 2, 931. (17) Li, S.; Jiang, Y.; Wu, D.; Wang, L.; Zhong, H.; Wu, B.; Lan, X.; Yu, Y.; Wang, Z.; Jie, J. J. Phys. Chem. C 2010, 114, 7980. (18) Davami, K.; Kang, D.; Lee, J. S.; Meyyappan, M. Chem. Phys. Lett. 2011, 504, 62. (19) Wu, D.; Jiang, Y.; Zhang, Y. G.; Li, J. W.; Yu, Y. Q.; Zhang, Y. P.; Zhu, Z. F.; Wang, L.; Wu, C. Y.; Luo, L. B.; Jie, J. S. J. Mater. Chem. 2012, 22, 6206. (20) Utama, M. I. B.; Peng, Z. P.; Chen, R.; Peng, B.; Xu, X. L.; Dong, Y. J.; Wong, L. M.; Wang, S. J.; Sun, H. D.; Xiong, Q. H. Nano Lett. 2011, 11, 3051. (21) Utama, M. I. B.; Zhang, Q.; Jia, S. F.; Li, D. H.; Wang, J. B.; Xiong, Q. H. ACS Nano 2012, 6, 2281. (22) Dai, Z. R.; Pan, Z. W.; Wang, Z. L. Adv. Funct. Mater. 2003, 13, 9. (23) Li, M.; Wu, B.; Ekahana, S. A.; Utama, M. I. B.; Xing, G.; Xiong, Q. H.; Sum, T. C.; Zhang, X. Appl. Phys. Lett. 2012, 101, 091104. (24) Fang, X. S.; Ye, C. H.; Zhang, L. D.; Wang, Y. H.; Wu, Y. C. Adv. Funct. Mater. 2005, 15, 63. (25) Fang, X. S.; Ye, C. H.; Peng, X. S.; Wang, Y. H.; Wu, Y. C.; Zhang, L. D. J. Mater. Chem. 2003, 13, 3040. (26) Wagner, R. S.; Ellis, W. C. Appl. Phys. Lett. 1964, 4, 89. (27) Johansson, J.; Karlsson, L. S.; Patrik, T.; Svensson, C.; Martensson, T.; Wacaser, B. A.; Deppert, K.; Samuelson, L.; Seifert, W. Nat. Mater. 2006, 5, 574. (28) Caroff, P.; Dick, K. A.; Johansson, J.; Messing, M. E.; Deppert, K.; Samuelson, L. Nat. Nanotechnol. 2009, 4, 50. (29) Utama, M. I. B.; de la Mata, M.; Magen, C.; Arbiol, J.; Xiong, Q. H. Adv. Funct. Mater. 2013, 23, 1636. (30) Xiong, Q. H.; Wang, J.; Eklund, P. C. Nano Lett. 2006, 6, 2736. (31) Algra, R. E.; Verheijen, M. A.; Borgstrom, M. T.; Feiner, L.-F.; Immink, G.; van Enckevort, W. J. P.; Vlieg, E.; Bakkers, E. P. A. M. Nature 2008, 456, 369. (32) Hao, Y.; Meng, G.; Wang, Z. L.; Ye, C.; Zhang, L. Nano Lett. 2006, 6, 1650.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (J.A.), [email protected] (Q.X.). Author Contributions #

These authors contributed equally to this paper.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Q.X. acknowledges the strong support from Singapore National Research Foundation via a fellowship grant (NRF-RF2009-06) and Ministry of Education via two Tier2 grants (MOE2011-T22-051 and MOE2012-T2-2-086). This research is also supported, in part, by the start-up grant (M58110061) and the New Initiative Fund (M58110100) from Nanyang Technological University. J.A. acknowledges the funding from the Spanish MICINN projects MAT2010-15138 (COPEON) and CSD2009-00013 (IMAGINE) and Generalitat de Catalunya (2009 SGR 770 and NanoAraCat). M.d.l.M. thanks CSIC JAE Pre-Doc program. The authors thank the TEM facilities at 2595

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596

Crystal Growth & Design

Article

(33) Xiong, Q. H.; Gupta, R.; Adu, K. W.; Dickey, E. C.; Lian, G. D.; Tham, D.; Fischer, J. E.; Eklund, P. C. J. Nanosci. Nanotechnol. 2003, 3, 335. (34) Arbiol, J.; Morral, A. F. i.; Estrade, S.; Peiro, F.; Kalache, B.; Cabarrocas, P. R. i.; Morante, J. R. J. Appl. Phys. 2008, 104, 064312. (35) Zhang, Q.; Zhang, J.; Utama, M. I. B.; Peng, B.; de la Mata, M.; Arbiol, J.; Xiong, Q. H. Phys. Rev. B 2012, 85, 085418. (36) Xi, L.; Chua, K. H.; Zhao, Y.; Zhang, J.; Xiong, Q. H.; Lam, Y. M. RSC Adv. 2012, 2, 5243. (37) Wu, Y. Y.; Yang, P. D. J. Am. Chem. Soc. 2001, 123, 3165. (38) Fang, X. S.; Ye, C. H.; Zhang, L. D.; Xie, T. Adv. Mater. 2005, 17, 1661. (39) Janik, E.; Dłużewski, P.; Kret, S.; Presz, A.; Kirmse, H.; Neumann, W.; Zaleszczyk, W.; Baczewski, L. T.; Petroutchik, A.; Dynowska, E.; Sadowski, J.; Caliebe, W.; Karczewski, G.; Wojtowicz, T. Nanotechnology 2007, 18, 475606. (40) Yeh, C. Y.; Lu, Z. W.; Froyen, S.; Zunger, A. Phys. Rev. B 1992, 46, 10086. (41) de la Mata, M.; Magen, C.; Gazquez, J.; Utama, M. I. B.; Heiss, M.; Lopatin, S.; Furtmayr, F.; Fernández-Rojas, C. J.; Peng, B.; Morante, J. R.; Rurali, R.; Eickhoff, M.; Fontcuberta i Morral, A.; Xiong, Q. H.; Arbiol, J. Nano Lett. 2012, 12, 2579. (42) (a) Grillo, V. Microsc. Microanal. 2011, 17, 1292. (b) Grillo, V.; Rotunno, E. Ultramicroscopy 2013, 125, 97. (c) Grillo, V.; Rossi, F. Ultramicroscopy 2013, 125, 112. (43) Ma, C.; Ding, Y.; Moore, D.; Wang, X.; Wang, Z. L. J. Am. Chem. Soc. 2003, 126, 708. (44) (a) Dayeh, S. A.; Wang, J.; Li, N.; Huang, J. Y.; Gin, A. V.; Picraux, S. T. Nano Lett. 2011, 11, 4200. (b) Tian, B.; Xie, P.; Kempa, T. J.; Bell, D. C.; Lieber, C. M. Nat. Nanotechnol. 2009, 4, 824. (c) Conesa-Boj, S.; Russo-Avershi, E.; Dalmau-Mallorqui, A.; Trevino, J.; Pecora, E. F.; Forestiere, C.; Handin, A.; Ek, M.; Zweifel, L.; Wallenberg, L. R.; Rüffer, D.; Heiss, M.; Troadec, D.; Dal Negro, L.; Caroff, P.; Fontcuberta i Morral, A. ACS Nano 2012, 6, 10982. (45) Bao, J.; Bell, D. C.; Capasso, F.; Wagner, J. B.; Mårtensson, T.; Trägårdh, J.; Samuelson, L. Nano Lett. 2008, 8, 836. (46) Karazhanov, S. Z.; Ravindran, P.; Kjekshus, A.; Fjellvag, H.; Grossner, U.; Svensson, B. G. J. Appl. Phys. 2006, 100, 043709. (47) Kshirsagar, S. D.; Ghanashyam Krishna, M.; Tewari, S. P. Mater. Sci. Semicond. Process. DOI: dx.doi.org/10.1016/j.mssp.2013.02.015. (48) Dick, K. A.; Thelander, C.; Samuelson, L.; Caroff, P. Nano Lett. 2010, 10, 3494.

2596

dx.doi.org/10.1021/cg4003498 | Cryst. Growth Des. 2013, 13, 2590−2596