Topological insulator nanostructures: Materials synthesis, Raman spectroscopy, and transport properties

Abstract

Nanostructured topological insulator materials such as ultrathin films, nanoplates, nanowires, and nanoribbons are attracting much attention for fundamental research as well as potential applications in low-energy dissipation electronics, spintronics, thermoelectrics, magnetoelectrics, and quantum computing due to their extremely large surface-to-volume ratios and exotic metallic edge/surface states. Layered Bi₂Se₃ and Bi₂Te₃ serve as reference topological insulator materials with a large nontrivial bulk gap up to 0.3 eV (equivalent to 3600 K) and simple single-Dirac-cone surface states. In this mini-review, we present an overview of recent advances in nanostructured topological insulator Bi₂Se₃ and Bi₃Te₃ from the viewpoints of controlled synthesis and physical properties. We summarize our recent achievements in the vapor-phase synthesis and structural characterization of nanostructured topological insulator Bi₂Se₃ and Bi₂Te₃, such as nanoribbons and ultrathin nanoplates. We also demonstrate the evolution of Raman spectra with the number of few-layer topological insulators, as well as the transport measurements that have succeeded in accessing the surface conductance and surface state manipulations in the device of topological insulator nanostructures.

Graphical abstract

Keywords

topological insulator / nanostructure / synthesis / Raman / transport / surface state manipulation

1 Introduction

Stoichiometric compounds Bi₂Se₃, Bi₂Te₃, and Sb₂Te₃ are predicted to be three-dimensional (3D) topological insulators (TIs) with an insulating bulk state and robust surface states consisting of a single Dirac cone at the \Gamma point [1]. Among these compounds, Bi₂Se₃ is found to have a large nontrivial bulk gap of about 0.3 eV (equivalent to 3600 K), making it an attractive reference material for exploring the physical properties of the topological surface states [1, 2]. The Dirac-cone surface states of 3D TIs have been mainly investigated by angle-resolved photoemission spectroscopy (ARPES) and scanning tunneling microscopy (STM) [3–6]. These studies have focused mainly on cleaved surface of bulk crystals grown by self-flux method [7]. However, TI bulk crystals are unintentionally and uncontrollably doped by anion vacancies and anti-sites defects, which give rise to the presence of large bulk electron pockets near the Fermi level in these samples. As a result, it is difficult to probe the transport properties and manipulate topological surface states in bulk crystal samples because the surface states are significantly overwhelmed by the unwanted bulk carriers [6, 8, 9].

It is desirable to find effective and convenient methods to obtain true TI samples with the Fermi level lying in the bulk gap, suitable for studying the transport properties of surface states. The material quality of TI bulk crystals can be further improved by compensating the anion vacancies or reducing the defects via an annealing process in excess of Se or Te. However, this is challenging due to the high volatility of Se and Te. Doping the bulk crystals with elements such as Sn, Ca, Sb or Pb can also effectively tune the Fermi level into the bulk gap [4, 5, 10–12], but this method usually carries the caveat of lowering the crystalline quality and carrier mobility. Chemical doping can also tune the Fermi level in TIs by adsorbing NO₂ molecules on the surface of Bi₂Se₃ crystal [7], although such a technique may be incompatible with device fabrication processes since NO₂ is not the most user-friendly gas and easily desorbs from the surface.

Compared with bulk materials, TI nanostructures with large surface-to-volume ratios such as thin films, nanoplates, nanowires, and nanoribbons are expected to significantly enhance topological surface conduction, and can also facilitate external manipulation of the surface states. Furthermore, TI nanostructures can provide excellent geometries for investigating the transport properties of exotic surface states. In particular, quasi-two-dimensional (quasi-2D) nanoribbons, nanoplates, and thin films of TIs may have superior applications in electronic devices because it is relatively easy to fabricate complex structures from them.

Recently, we have demonstrated direct transport evidence of the robust, conducting surface states in TI nanoribbons. In the initial work, pronounced Aharonov–Bohm oscillations have been observed in Bi₂Se₃ nanoribbons, which present the first demonstration of quantum interference in the transport properties of TIs [13]. Following this work, the Kondo effect has also been revealed below \sim30 K in magnetically-doped Bi₂Se₃ nanoribbons [14]. Several groups have demonstrated that the surface states of TI nanostructures can be easily manipulated by an external electric field [15–19]. For example, in Bi₂Te₃ nanoplates, we observed a clear gate effect and tunable chemical potential by the combination of a strong electrical field from the top gate using a high-k dielectric layer and ultrathin thickness of nanoplates [17]. More recently, Shubnikov–de Haas (SdH) oscillations [15], surface ambipolar effects [16], and weak antilocalization behavior [18] have been observed in electric-field-modulated TIs nanostructures, attractive for probing topological surface states and future device applications.

In this mini-review, we present the recent progress in the controlled synthesis, structural characterizations, Raman spectroscopy, and transport measurements of TI nanostructures. We first summarize the recent advances in the production of TI nanostructures. Then we give Raman characterizations of TI nanostructures. Finally, we present studies on the transport properties of surface conductance and surface state manipulations in TI nanostructure devices.

2 Results and discussion

2.1 Production of TI nanostructures

To date, controlled production of TI nanostructures with high crystalline quality and high carrier mobility is still lacking. Layered Bi₂X₃ (X = Se, Te), the reference TIs, share the same rhombohedral crystal structure in the space group D_{3d}^5(R\overline{3}m), with each quintuple layer (QL) consisting of X–Bi–X–Bi–X sheets. The planar, covalently bonded sheets are held together by weak van der Waals interactions [Fig. 1(a)]. Bulk materials of Bi₂X₃ usually exist in the layered form of quasi-2D flakes or platelets owing to this anisotropic bonding. On the other hand, the anisotropic bonding characteristics may play an important role in the production of TI nanostructures.

Various methods have been used to produce Bi₂X₃ nanostructures, which can be generally classified into “bottom–up” and “top–down” approaches. The “top–down” approach includes micromechanical exfoliation by scotch tape [20, 21], peeling with an AFM tip [22], and chemical exfoliations [23] of few-layer Bi₂X₃ sheets from bulk crystal samples. This simple micromechanical exfoliation has been used to yield high quality, isolated nanosheets [21]. However, this method still has some limitations such as low yield, irregular shapes, and poor controllability. In the “bottom–up” approach, few-layer Bi₂X₃ are synthesized from atoms or molecules via chemical reactions. Chemical vapor transport [13, 24], metal-organic chemical vapor deposition [25], molecular beam epitaxy (MBE) [26, 27], and wet chemical synthesis [28, 29] are some prominent examples of the “bottom-up” approach. MBE is a promising technique to grow high-quality TI films in a very controlled manner. Ultrathin films of TIs down to 1 QL were obtained by MBE [12, 26, 30], and Landau quantization of the topological surface states was observed in these high-quality films [31]. However, the state-of-art MBE technique is expensive with limited accessibility.

We have developed facile vapor-phase deposition methods to grow TI nanostructures with high crystallinity and diverse morphologies [13, 24, 32]. Quasi-1D nanowires and nanoribbons have been synthesized via a catalyst-assisted vapor–liquid–solid (VLS) growth. Ultrathin quasi-2D nanoplates of Bi₂Se₃ and Bi₂Te₃ with uniform thickness down to one QL and the lateral dimensions up to 100 µm have been grown via van der Waals epitaxy.

2.1.1 Synthesis of Quasi-1D nanostructures of TIs

VLS growth has proven to be an effective method to obtain quasi-1D nanostructures with high crystalline quality and clean surfaces [33–40]. We synthesized high-quality Bi₂Se₃ nanowires and nanoribbons by means of a gold-catalyzed VLS growth [13, 24]. The schematic diagram of the growth system is shown in Figs. 1(b) and (c).

Scanning electron microscope (SEM) images of as-grown quasi-1D nanostructures of Bi₂Se₃ with various morphologies are shown in Fig. 2. We found that morphologies of Bi₂Se₃ nanostructures were controlled by the growth temperature and gas flow rate. Narrow nanoribbons were the dominant products when the substrate temperature was about 450 °C [Fig. 2(a)]. Nanoribbons with various morphologies were obtained when the substrate temperature were about 375 °C, as shown in Fig. 2(b). The growth rate of synthesized products increased dramatically with the increase in source temperature and flow rate. Wide nanoribbons with lateral dimensions of several tens of microns were obtained by using higher source temperature and flow rate [Fig. 2(c)]. We have also observed quasi-1D nanostructures with other diverse morphologies, such as nanowires formed from stacked nanoplates grown along the c-axis [Figs. 2(d) and (e)], nanowires grown off along the c-axis [Fig. 2(f)], zig-zag nanoribbons [Fig. 2(g)], sheet-like wide nanoribbons [Fig. 2(h)], and nanoplates grown on the substrate via catalyst-free vapor-solid (VS) mechanism [Fig. 2(i)]. High-resolution transmission electron microscope (HRTEM), selected area electron diffraction (SAED), and energy-dispersive X-ray spectroscopy (EDX) studies were used to evaluate the structure and chemical composition of the as-grown quasi-1D nanostructures. Extensive HRTEM and SAED studies have confirmed that the quasi-1D nanostructures have single-crystalline phases with atomically smooth surface/edges. EDX analyses have revealed uniform chemical composition of the TI nanostructures with a Bi: Se atomic ratio of 2:3, indicating stoichiometric Bi₂Se₃ free of detectable impurities within the sensitivity of the EDX analyses (<2%) [13, 24].

2.1.2 Synthesis of Quasi-2D nanostructures of TIs

Van der Waals epitaxy (VDWE) is an effective strategy for growing designed structures of quasi-2D layered materials with van der Waals gaps onto a substrate without surface dangling bonds [41, 42]. In layered TI Bi₂X₃, low surface energy and the absence of unsaturated chemical bonds at the surface can relax stringent lattice matching conditions, which facilitate the growth of high-quality quasi-2D nanostructures of TIs on various substrates such as oxidized silicon [17], graphene [32], Si(111) [43], and SrTiO₃ [18].

We have achieved the growth of ultrathin Bi₂Se₃ and Bi₂Te₃ nanoplates via a catalyst-free VS process on an oxidized silicon substrate [17]. A diagram for the growth system is shown in Figs. 1(b) and (d). Triangular and hexagonal quasi-2D Bi₂Se₃ and Bi₂Te₃ nanoplates with thicknesses down to \sim3 nm (3 QLs) and lateral dimensions up to 20 microns have been obtained on 300 nm SiO₂/Si substrates with a quasi-VDWE method [Fig. 3(a)] [17]. Most of the nanoplates have a flat surface with uniform lateral thickness. Analyses of the thickness distribution indicated that the major nanoplate products have the thicknesses between 3 QLs and 6 QLs [Figs. 3(b) and (c)]. HRTEM and SAED results revealed the single crystalline nature of the Bi₂Se₃ and Bi₂Te₃ nanoplates. Both Bi₂Se₃ and Bi₂Te₃ nanoplates had (0001) facets as top and bottom surfaces, and (01-10) facets as side surfaces. EDX analyses confirmed the uniform chemical composition of the nanoplates with a Bi: (Se, Te) atomic ratio of 2:3. In addition, we have speculated on the growth mechanism of ultrathin nanoplates on SiO₂/Si substrates. The high anisotropic bonding nature of Bi₂Se₃ and Bi₂Te₃ is believed to play a major role in the growth of ultrathin nanoplates. The growth anisotropy factor is several thousand, estimated by taking the ratio of the lateral to vertical dimensions. The direct growth of ultrathin TI nanoplates onto SiO₂/Si substrates facilitates device fabrications and transport measurements.

However, the growth of large-area, ultrathin nanoplate/sheets was restricted by the presence of the surface dangling bonds on the SiO₂/Si substrate. In contrast, a layered, chemically inert substrate favors the growth of large-area and high-quality epitaxial layers via the VDWE process. Wu et al. has pioneered quintuple-layer epitaxy of TI Bi₂Se₃ thin films on a Si(111) substrate in an MBE apparatus [44]. Following this work, Xie et al. has achieved the VDWE of large-area high-quality TI Bi₂Se₃ epifilms on a vicinal Si(111) substrate with the surface dangling bonds saturated by hydrogen atoms [43]. The crystalline Bi₂Se₃ thin films have a low-temperature resistivity of \sim1 mΩ cm, a carrier mobility of \sim 2000{\mathrm{cm}}^2\cdot {\mathrm{V}}^{-1}\cdot {\mathrm{s}}^{-1}, and an electron density of \sim 3\times {10}^{18}{\mathrm{cm}}^{-3}. Recently, Wu et al. have achieved high quality, gate-tunable Bi₂Se₃ thin films on SrTiO3 substrates by MBE [18, 19]. Meanwhile, Xue group have succeeded in the growth of high-quality ultrathin TI Bi₂Se₃ films on a graphitized-terminated 6H-SiC (0001) substrate using MBE [26]. Landau quantization in such high-quality Bi₂Se₃ thin films was directly observed by STM [31]. In addition, the same group have reported major achievements on the epitaxy of high-quality Bi₂Te₃ film on Si(111) substrate using MBE [27, 30].

Following our initial work on vapor-phase epitaxy of Bi₂Se₃ and Bi₂Te₃ nanoplates grown on an oxidized silicon substrate [17], we have extended this epitaxy method to the growth of large-area, high crystalline TI nanoplates on VDWE substrates including graphene [32]. Arrays of Bi₂Se₃ and Bi₂Te₃ nanoplates with good orientations were also obtained. More importantly, lateral dimensions of single nanoplates of Bi₂Se₃ and Bi₂Te₃ on the VDWE substrate can reach up to \sim0.1 mm. The high quality nanoplates with large lateral dimensions epitaxially grown on an insulating VDWE substrate provided a straightforward platform for spectroscopic characterizations, device fabrications, and transport measurements. Electrodes can be easily patterned on individual nanoplates using standard micro-photolithography techniques. The combination of spectroscopic characterizations with electrical transport measurements allowed us to correlate directly the electrical properties and the structure of the same individual nanoplate (unpublished data).

2.1.3 Epitaxial heterostructure of TI nanoplate and graphene

Graphene, a 2D material consisting of a flat and flexible monolayer of carbon atoms without surface dangling bonds [45, 46], is an excellent VDWE substrate that favors the epitaxial growth of large-area, ultrathin TI nanoplates/sheets. Recently, we have achieved epitaxial heterostructures of ultrathin TI nanoplate and graphene using a simple vapor-phase deposition method [32]. The hexagonal periodicity of graphene surface has a lattice mismatch of \sim2.9% with Bi₂Se₃ [Fig. 4(a)]. However, lattice matching conditions can be relaxed drastically in VDWE, allowing the formation of a large variety of high-quality heterostructures [41, 42].

Optical photographs and SEM images of Bi₂Se₃ nanoplates epitaxially grown on a graphene substrate show that the Bi₂Se₃ nanoplates have regular triangular or hexagonal shapes [Figs. 4(b) and (c)]. The lateral dimension of the nanoplates were in the range of several hundred nanometers to several microns. Each nanoplate has uniform color contrast, suggesting a uniform thickness through the whole nanoplate surface. The Bi₂Se₃ nanoplates were aligned in the same orientation on the graphene substrate. From the histogram of edge orientation distribution shown in Fig. 4(d), the orientation of the nanoplates’ edges occur at multiples of {60}^{°}, suggesting the epitaxial nature of the growth mechanism.

It is worth noting that the thickness of single crystalline Bi₂Se₃ nanoplates can be as small as 1 QL on pristine graphene substrates. As shown in Fig. 5, continuous, ultrathin Bi₂Se₃ nanoplates with uniform thicknesses from 1 QL to 9 QLs epitaxially grown on graphene substrates were characterized by AFM. The atomically flat Bi₂Se₃ nanoplates can grow over the steps of graphene. The size of the Bi₂Se₃ nanoplates greatly depends on the lateral dimension of the graphene substrate since the growth terminates at the edges of graphene.

The structure and chemical compositions of the Bi₂Se₃ nanoplates have been studied by TEM and EDX. EDX revealed the uniform composition of the nanoplate with an exact Bi:Se atomic ratio of 2:3. A typical bright-field TEM image of the triangle Bi₂Se₃ nanoplates with the same orientation epitaxially grown on graphene is shown in Fig. 6(a). A HRTEM image [Fig. 6(b)] and the corresponding SAED pattern [Fig. 6(c)] confirmed the single crystalline nature of the nanoplates. The HRTEM image revealed the expected hexagonal lattice fringes, with a lattice spacing of 0.21 nm, consistent with the lattice spacing of Bi₂Se₃ (11–20) planes [13, 24]. Nanoplates have (0001) facets as top and bottom surfaces, and (01–10) facets as side surfaces.

Interestingly, Moiré-type modulation structures have been observed in the epitaxial heterostructures of Bi₂Se₃ nanoplates and graphene. In principle, a moiré pattern can be induced by the superposition of the epifilm and substrate due to the lattice mismatch and/or different orientations, which is a useful tool for characterizing interfacial structures and strain development in lattice mismatched semiconductor heterostructures [47, 48]. Take Bi₂Se₃/graphene nanoplate heterostructures for example, Fig. 7(a) shows the planar-view TEM image of two triangular Bi₂Se₃ nanoplates on suspended few-layer graphene membrane recorded along the 0001 zone axis. From the close-up view [Fig. 7(b)], the sets of intersecting translational moiré fringes can be clearly observed in the three main directions ⟨11-20⟩. The well-defined moiré pattern in the left and right nanoplate has a period of about 4.2 and 5.5 nm, respectively [Fig. 7(c)]. The period difference presumably originates from different orientation angles between the Bi₂Se₃ nanoplates and underlying graphene. The highly ordered, defect-free moiré patterns throughout the nanoplates indicate that the Bi₂Se₃ nanoplate has high crystalline quality.

Moiré patterns can be used to determine the plastic relaxation in individual nanoplates. For example, planar-view TEM images of the Bi₂Se₃/graphene nanoplate heterostructure [Figs. 7(d) and (g)] reveal a typical highly ordered and continuous moiré pattern with a period of \sim7.3 nm, consistent with the period of the simulated moiré pattern by assuming a fully relaxed Bi₂Se₃ on graphene with a rotational angle of {29}^{°} [49]. The corresponding FFT image [inset of Fig. 7(g)] shows two sets of hexagonal spots with a rotational angle, which confirms that the moiré pattern results from the superposition of two structures with different lattice constants and/or orientations. Interestingly, the few-layer Bi₂Se₃ nanoplate are relatively more stable than few-layer graphene against the electron-beam irradiation in a TEM operated at 300 kV. We have carried out layer-by-layer peeling for Bi₂Se₃/graphene nanoplate heterostructures using high-energy electron irradiation. The suspended graphene substrate was first removed by high-energy electron irradiation until a nanometer-sized hole formed in the graphene sheet. The moiré pattern of Bi₂Se₃ nanoplate disappears in the hole region [Figs. 7(e) and (f)], which indicates that the moiré pattern results from the superposition of Bi₂Se₃ and graphene. HRTEM images of the suspended Bi₂Se₃ layer reveal a simple hexagonal lattice of Bi₂Se₃ [Figs. 7(h) and (i)]. The corresponding FFT image [inset of Fig. 7(i)] shows a single diffraction pattern, indicating the disappearance of the moiré pattern and the plastic relaxation in the individual nanoplates.

Another interesting observation was that a high-resolution nanofabrication of TI nanoplates with the help of a high energy, highly focused electron beam has been achieved. As shown in Figs. 7(e) and (f), extremely small nanoholes, as small as 5 nm in diameter, was reproducibly fabricated in TI nanoplates. From the HRTEM images, the holes have a quasi-hexagonal shape with atomically smooth edges [Figs. 7(h) and (i)]. Furthermore, the TEM images recorded near the edges of the nanoholes in the nanoplate display strain contrast presumably induced by corrugating or bending of the suspended nanoplates.

2.2 Raman characterizations of TI nanostructures

Investigations on phonons and electron–phonon interactions in topological insulators, especially down to a few-layer regime, are interesting topics. Raman scattering spectroscopy is a powerful and sensitive tool for the study of phonons in 2D nanomaterials [32, 50], such as graphene and few-layer TIs. We have systemically studied the relationship between Raman spectra and layer numbers of 2D Bi₂Se₃ nanoplates at room temperature in the backscattering geometry [32]. The Raman intensity and frequency shifts have a close relationship with the layer number of TI. Take Bi₂Se₃ nanoplates epitaxially grown on graphene, for example, the layer number of Bi₂Se₃ nanoplates was first identified by optical color contrast [Fig. 8(a)] and confirmed by AFM measurements. Figure 8(b) shows the evolution of Raman spectra produced from a 633 nm laser with the number of Bi₂Se₃ layers. The in-plane vibrational mode (E_{\mathrm{g}}^2) and out-of-plane vibrational mode (A_{1\mathrm{g}}^2) of Se–Bi–Se–Bi–Se lattice vibration modes [51] were detected at \sim131 and \sim174 cm^-1, respectively. Figure 8(c) shows the dependence of the E_{\mathrm{g}}^2 and A_{1\mathrm{g}}^2 vibration peaks intensity on the layer number. Bi₂Se₃ nanoplates with a thickness of 3 QLs exhibit the strongest Raman intensity, presumably due to the presence of an enhancement factor.

Strain effects in the Bi₂Se₃/graphene nanoplate heterostructures have also been studied by Raman spectroscopy [32]. Figure 8(d) plots the evolution of the Raman frequency position of the E_{\mathrm{g}}^2 and A_{1\mathrm{g}}^2 modes with the number of layers. Both single-QL and 2-QL nanoplates reveal a significant down shift in frequency of the E_{\mathrm{g}}^2 modes, and an upward shift for the frequency of A_{1\mathrm{g}}^2 modes. These results suggest that a tensile stress and a compressive stress exist in the ab-plane and along the c-axis direction of Bi₂Se₃ nanoplates due to lattice mismatch, respectively [52–56]. Meanwhile, the tensile stress of Bi₂Se₃ nanoplates might lead to a compressive stress in the few-layer graphene substrate. Indeed, upward shifts of 4\sim10 cm^-1 for the G-band and 15\sim 25{\mathrm{cm}}^{-1} for 2D-band of graphene were observed, indicating the presence of the compressive stress in the decorated single-layer graphene substrate, consistent with tensile stress in the 1-QL and 2-QL Bi₂Se₃ nanoplates. In contrast, no obvious shifts of Raman position for multilayer (⩾3 QLs) Bi₂Se₃ nanoplates were detected, indicating that the stress inside the multilayer Bi₂Se₃ epifilm was relaxed [52, 56].

2.3 Transport measurements

Quasi-1D and quasi-2D nanostructures of TI Bi₂Se₃, with their large surface-to-volume ratios, provide excellent geometries for manipulating and probing the transport properties of surface states [13, 15, 17, 18]. We have performed transport measurements of TI nanostructures grown by the vapor-phase deposition method [13]. Pronounced Aharonov–Bohm (AB) oscillations associated with the robust topological surface states in Bi₂Se₃ nanoribbons have been observed, which is the first demonstration of quantum interference in the transport properties of TIs [13].

To observe the quantum oscillations caused by the AB effect [Fig. 9(a)], an external magnetic field (B) was applied along the Bi₂Se₃ nanoribbon length [Fig. 9(b)]. Four-point probe devices have been fabricated by depositing Ti/Au ohmic contacts onto the nanoribbon [Fig. 9(b)]. For the 2D states covering the entire surface, the TI nanoribbon can be considered as a hollow conducting cylinder. Owing to the quantum interference effect, the magnetoconductance depends on the magnetic flux enclosed by the cylinder and oscillates with a period given by the magnetic flux quantum h/e, where h is Planck’s constant and e is the elementary charge. The weak anti-localization effect characterized by a sharp cusp was observed at low magnetic fields, indicating the presence of the strong spin–orbit coupling in Bi₂Se₃ nanoribbons [Fig. 9(c)]. Pronounced and reproducible resistance oscillations, the hallmark of the well-known AB effect, have been observed with the magnetic field ranging from 0.15 to 2 T [Fig. 9(c)]. Although the perfect periodicity slightly deteriorated at higher fields up to 9 T, the resistance minima has a well defined linear relationship with an approximate period of 0.6 T [Fig. 9(d)]. A weak second harmonic period located at h/(2e) was also observed except for the prominent h/e oscillation period [Fig. 9(e)] in the Bi₂Se₃ nanoribbon by performing FFT of the derivative dR/dB. Most oscillatory features of the AB effect persisted up to 20 K. The transport measurements also revealed that the surface states exist not only on the top and bottom (0001) surfaces but on the side walls of the Bi₂Se₃ nanoribbon as well. The observation of AB oscillations in nanoribbons provides important insights into the topological surface states of Bi₂Se₃ and opens the route for electronic transport experiments in nanostructures of 3D TIs.

The surface states and Fermi level of TI nanostructures can be easily manipulated by an external electric field. Recently, we have observed a clear gate effect and tunable chemical potential in ultrathin Bi₂Te₃ nanoplates by applying a strong electric field from the top gate [17]. In that work, ultrathin Bi₂Te₃ nanoplates grown by VS process on 300 nm SiO₂/Si substrate have an extremely large surface-to-volume ratio and can be electrically gated more effectively than the bulk materials. To enhance the electrical gating effect, a high-k dielectric top gate was deposited onto the individual ultrathin nanoplate. Tunable chemical potentials and pronounced weak antilocalization effect have been observed. Figure 10(a) shows the relationship between the sheet resistance R_{\mathrm{S}} and the top gate voltage V_{\mathrm{G}} measured at 2 K. The R_{\mathrm{S}} showed a quasi-linear increase with negative V_{\mathrm{G}} up to –17.5 V, beyond which the R_{\mathrm{S}} increased more rapidly when the V_{\mathrm{G}} was further increased in the negative direction. The slope of R_{\mathrm{S}} dependence of V_{\mathrm{G}} manifested the sharper increase of R_{\mathrm{S}} beyond –17.5 V [inset of Fig. 10(a)]. The relationship between R_{\mathrm{S}} and V_{\mathrm{G}} suggests that the tunable Fermi level is approaching the bottom of the conduction band at V_{\mathrm{G}} of –17.5 V. The magnetoconductance of the same Bi₂Te₃ nanoplate has been measured with an applied external magnetic field perpendicular to the nanoplate’s plane. A weak anti-localization effect characterized by a sharp peak in the magnetoconductance [Fig. 10(b)], resulting from the strong spin–orbit coupling in TI materials, was observed at low magnetic field [13]. More recently, several other groups have also reported on the manipulation of surface states of TI nanostructures by external electric fields [15, 16, 18]. For example, Shubnikov–de Haas (SdH) oscillations [15] and surface ambipolar effects [16] have been observed in electric-field-modulated TI nanostructures, which indicate that TI nanostructures provide promising materials for future device applications.

3 Summary

In this mini-review, we have made an overview of recent advances in the controlled synthesis and physical property measurements of nanoscale 3D TI materials. We have succeeded in the controlled growth of quasi-1D nanostructures such as nanowires and narrow nanoribbons via a gold-catalyst-assisted vapor-liquid-solid growth. Beyond, we have achieved the growth of ultrathin nanoplates of Bi₂Se₃ and Bi₂Te₃ with uniform thickness down to one quintuple layer on various substrates via van der Waals epitaxy. We have investigated the evolution of Raman intensity and peak positions with the thickness of TI nanostructures. The strain evolution in few-layer TI nanostructures was studied. We found that the nanoribbons and nanoplates were ideal geometries for direct measurements and manipulation of the topological surface states. Pronounced Aharonov-Bohm-type quantum interference effects were observed in Bi₂Se₃ nanoribbons, providing direct transport evidence of the robust conducting surface states. The surface states of the nanoplates were directly manipulated with a high-k dielectric top gate. Our studies suggest that topological insulator nanostructures hold great promise for the study of topological surface states and future electronic applications.

References

[1] H. J. Zhang, C. X. Liu, X. L. Qi, X. Dai, Z. Fang, and S. C. Zhang, Nat. Phys., 2009, 5: 438

[2] J. E. Moore, Nat. Mater., 2010, 464: 194

[3] M. Z. Hasan and C. L. Kane, Rev. Mod. Phys., 2010, 82: 3045

[4] Y. L. Chen, J. G. Analytis, J. H. Chu, Z. K. Liu, S. K. Mo, X. L. Qi, H. J. Zhang, D. H. Lu, X. Dai, Z. Fang, S. C. Zhang, I. R. Fisher, Z. Hussain, and Z. X. Shen, Science, 2009, 325: 178

[5] Y. Xia, D. Qian, D. Hsieh, L. Wray, A. Pal, H. Lin, A. Bansil, D. Grauer, Y. S. Hor, R. J. Cava, and M. Z. Hasan, Nat. Phys., 2009, 5: 398

[6] J. G. Checkelsky, Y. S. Hor, M. H. Liu, D. X. Qu, R. J. Cava, and N. P. Ong, Phys. Rev. Lett., 2009, 103: 406601

[7] D. Hsieh, Y. Xia, D. Qian, L. Wray, F. Meier, J. H. Dil, J. Osterwalder, L. Patthey, A. V. Fedorov, H. Lin, A. Bansil, D. Grauer, Y. S. Hor, R. J. Cava, and M. Z. Hasan, Phys. Rev. Lett., 2009, 103: 146401

[8] A. A. Taskin and Y. Ando, Phys. Rev. B, 2009, 80: 085303

[9] J. G. Analytis, J. H. Chu, Y. Chen, F. Corredor, R. D. McDonald, Z. X. Shen, and I. R. Fisher, Phys. Rev. B, 2010, 81: 205407

[10] D. Hsieh, D. Qian, L. Wray, Y. Xia, Y. S. Hor, R. J. Cava, and M. Z. Hasan, Nature, 2008, 452: 970

[11] Z. Y. Wang, T. Lin, P. Wei, X. F. Liu, R. Dumas, K. Liu, and J. Shi, Appl. Phys. Lett., 2010, 97: 042112

[12] Y. Zhang, C. Z. Chang, K. He, L. L. Wang, X. Chen, J. F. Jia, X. C. Ma, and Q. K. Xue, Appl. Phys. Lett., 2010, 97: 194102

[13] H. L. Peng, K. J. Lai, D. S. Kong, S. Meister, Y. L. Chen, X. L. Qi, S. C. Zhang, Z. X. Shen, and Y. Cui, Nat. Mater., 2010, 9: 225

[14] J. J. Cha, J. R. Williams, D. S. Kong, S. Meister, H. L. Peng, A. J. Bestwick, P. Gallagher, D. Goldhaber-Gordon, and Y. Cui, Nano Lett., 2010, 10: 1076

[15] F. X. Xiu, L. A. He, Y. Wang, L. N. Cheng, L. T. Chang, M. R. Lang, G. A. Huang, X. F. Kou, Y. Zhou, X. W. Jiang, Z. G. Chen, J. Zou, A. Shailos, and K. L. Wang, Nat. Nanotechnol., 2011, 6: 216

[16] H. Steinberg, D. R. Gardner, Y. S. Lee, and P. Jarillo-Herrero, Nano Lett., 2010, 10: 5032

[17] D. S. Kong, W. H. Dang, J. J. Cha, H. Li, S. Meister, H. L. Peng, Z. F. Liu, and Y. Cui, Nano Lett., 2010, 10: 2245

[18] J. Chen, H. J. Qin, F. Yang, J. Liu, T. Guan, F. M. Qu, G. H. Zhang, J. R. Shi, X. C. Xie, C. L. Yang, K. H. Wu, Y. Q. Li, and L. Lu, Phys. Rev. Lett., 2010, 105: 4

[19] G. H. Zhang, H. J. Qin, J. Chen, X. Y. He, L. Lu, Y. Q. Li, and K. H. Wu, Adv. Funct. Mater., 2011, 21: 2351

[20] V. Goyal, D. Teweldebrhan, and A. A. Balandin, Appl. Phys. Lett., 2010, 97: 133117

[21] D. Teweldebrhan, V. Goyal, and A. A. Balandin, Nano Lett., 2010, 10: 1209

[22] S. S. Hong, W. Kundhikanjana, J. J. Cha, K. J. Lai, D. S. Kong, S. Meister, M. A. Kelly, Z. X. Shen, and Y. Cui, Nano Lett., 2010, 10: 3118

[23] J. N. Coleman, M. Lotya, A. O’Neill, S. D. Bergin, P. J. King, U. Khan, K. Young, A. Gaucher, S. De, R. J. Smith, I. V. Shvets, S. K. Arora, G. Stanton, H. Y. Kim, K. Lee, G. T. Kim, G. S. Duesberg, T. Hallam, J. J. Boland, J. J. Wang, J. F. Donegan, J. C. Grunlan, G. Moriarty, A. Shmeliov, R. J. Nicholls, J. M. Perkins, E. M. Grieveson, K. Theuwissen, D. W. McComb, P. D. Nellist, and V. Nicolosi, Science, 2011, 331: 568

[24] D. S. Kong, J. C. Randel, H. L. Peng, J. J. Cha, S. Meister, K. J. Lai, Y. L. Chen, Z. X. Shen, H. C. Manoharan, and Y. Cui, Nano Lett., 2010, 10: 329

[25] Y. F. Lin, H. W. Chang, S. Y. Lu, and C. W. Liu, J. Phys. Chem. C, 2007, 111: 18538

[26] Y. Zhang, K. He, C. Z. Chang, C. L. Song, L. L. Wang, X. Chen, J. F. Jia, Z. Fang, X. Dai, W. Y. Shan, S. Q. Shen, Q. Niu, X. L. Qi, S. C. Zhang, X. C. Ma, and Q. K. Xue, Nat. Phys., 2010, 6: 584

[27] H. W. Liu, H. T. Yuan, N. Fukui, L. Zhang, J. F. Jia, Y. Iwasa, M. W. Chen, T. Hashizume, T. Sakurai, and Q. K. Xue, Cryst. Growth Des., 2010, 10: 4491

[28] H. M. Cui, H. Liu, J. Y. Wang, X. Li, F. Han, and R. I. Boughton, J. Cryst. Growth, 2004, 271: 456

[29] S. H. Yu, J. Yang, Y. S. Wu, Z. H. Han, J. Lu, Y. Xie, and Y. T. Qian, J. Mater. Chem., 1998, 8: 1949

[30] Y. Y. Li, G. A. Wang, X. G. Zhu, M. H. Liu, C. Ye, X. Chen, Y. Y. Wang, K. He, L. L. Wang, X. C. Ma, H. J. Zhang, X. Dai, Z. Fang, X. C. Xie, Y. Liu, X. L. Qi, J. F. Jia, S. C. Zhang, and Q. K. Xue, Adv. Mater. (Deerfield Beach Fla.), 2010, 22: 4002

[31] P. Cheng, C. L. Song, T. Zhang, Y. Y. Zhang, Y. L. Wang, J. F. Jia, J. Wang, Y. Y. Wang, B. F. Zhu, X. Chen, X. C. Ma, K. He, L. L. Wang, X. Dai, Z. Fang, X. C. Xie, X. L. Qi, C. X. Liu, S. C. Zhang, and Q. K. Xue, Phys. Rev. Lett., 2010, 105: 076801

[32] W. H. Dang, H. L. Peng, H. Li, P. Wang, and Z. F. Liu, Nano Lett., 2010, 10: 2870

[33] C. M. Lieber, MRS Bull., 2003, 28: 486

[34] P. D. Yang, MRS Bull., 2005, 30: 85

[35] P. Gao and Z. L. Wang, J. Phys. Chem. B, 2002, 106: 12653

[36] M. T. Bjork, B. J. Ohlsson, T. Sass, A. I. Persson, C. Thelander, M. H. Magnusson, K. Deppert, L. R. Wallenberg, and L. Samuelson, Appl. Phys. Lett., 2002, 80: 1058

[37] S. Meister, H. L. Peng, K. McIlwrath, K. Jarausch, X. F. Zhang, and Y. Cui, Nano Lett., 2006, 6: 1514

[38] J. S. Lee, S. Brittman, D. Yu, and H. Park, J. Am. Chem. Soc., 2008, 130: 6252

[39] H. L. Peng, S. Meister, C. K. Chan, X. F. Zhang, and Y. Cui, Nano Lett., 2006, 7: 199

[40] H. L. Peng, C. Xie, D. T. Schoen, and Y. Cui, Nano Lett., 2008, 8: 1511

[41] A. Koma, Thin Solid Films, 1992, 216: 72

[42] A. Koma, J. Cryst. Growth, 1999, 201: 236

[43] H. D. Li, Z. Y. Wang, X. Kan, X. Guo, H. T. He, Z. Wang, J. N. Wang, T. L. Wong, N. Wang, and M. H. Xie, New J. Phys., 2010, 12: 11

[44] G. H. Zhang, H. J. Qin, J. Teng, J. D. Guo, Q. L. Guo, X. Dai, Z. Fang, and K. H. Wu, Appl. Phys. Lett., 2009, 95: 053114

[45] A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6: 183

[46] C. Stampfer, S. Fringes, J. Güttinger, F. Molitor, C. Volk, B. Terrés, J. Dauber, S. Engels, S. Schnez, A. Jacobsen, S. Dröscher, T. Ihn, and K. Ensslin, Front. Phys., 2011, 6(3): 271

[47] S. Miao, J. Zhu, X. Zhang, and Z. Y. Cheng, Phys. Rev. B, 2001, 65: 052101

[48] T. Kehagias, P. Komninou, G. Nouet, P. Ruterana, and T. Karakostas, Phys. Rev. B, 2001, 64: 195329

[49] C. L. Song, Y. L. Wang, Y. P. Jiang, Y. Zhang, C. Z. Chang, L. L. Wang, K. He, X. Chen, J. F. Jia, Y. Y. Wang, Z. Fang, X. Dai, X. C. Xie, X. L. Qi, S. C. Zhang, Q. K. Xue, and X. C. Ma, Appl. Phys. Lett., 2010, 97: 143118

[50] M. S. Dresselhaus, A. Jorio, M. Hofmann, G. Dresselhaus, and R. Saito, Nano Lett., 2010, 10: 751

[51] W. Richter and C. R. Becker, Phys. Status Solidi B, 1977, 84: 619

[52] Y. I. Yuzyuk, R. S. Katiyar, V. A. Alyoshin, I. N. Zakharchenko, D. A. Markov, and E. V. Sviridov, Phys. Rev. B, 2003, 68: 104104

[53] P. S. Dobal, S. Bhaskar, S. B. Majumder, and R. S. Katiyar, J. Appl. Phys., 1999, 86: 828

[54] G. Wedler, C. M. Schneider, A. Trampert, and R. Koch, Phys. Rev. Lett., 2004, 93: 236101

[55] A. Fillon, G. Abadias, A. Michel, C. Jaouen, and P. Villechaise, Phys. Rev. Lett., 2010, 104: 096101

[56] G. Springholz and K. Wiesauer, Phys. Rev. Lett., 2002, 88: 015507

Rights & permissions

Higher Education Press and Springer-Verlag Berlin Heidelberg 2012