Solvothermally Synthesized Sb2Te3 Platelets Show Unexpected

Letter pubs.acs.org/NanoLett

Solvothermally Synthesized Sb2Te3 Platelets Show Unexpected Optical Contrasts in Mid-Infrared Near-Field Scanning Microscopy Benedikt Hauer,† Tobias Saltzmann,‡ Ulrich Simon,‡ and Thomas Taubner*,† †

Institute of Physics (IA) and JARA − Fundamentals of Future Information Technologies, RWTH Aachen University, 52056 Aachen, Germany ‡ Institute of Inorganic Chemistry (IAC) and JARA − Fundamentals of Future Information Technologies, RWTH Aachen University, 52056 Aachen, Germany S Supporting Information *

ABSTRACT: We report nanoscale-resolved optical investigations on the local material properties of Sb2Te3 hexagonal platelets grown by solvothermal synthesis. Using midinfrared near-field microscopy, we find a highly symmetric pattern, which is correlated to a growth spiral and which extends over the entire platelet. As the origin of the optical contrast, we identify domains with different densities of charge carriers. On Sb2Te3 samples grown by other means, we did not find a comparable domain structure. KEYWORDS: Nanoplatelet, screw dislocation, crystal growth, free charge carriers, topological insulator

A

ntimony telluride (Sb2Te3) has gained lots of attention due to its exotic physical properties and the resulting potential for new device applications.1 It is related to the class of phase change alloys2,3 located on the tie line between Sb2Te3 and GeTe (i.e., the GeSbTe compounds). At the same time, it can be used in thermoelectric devices4,5 and was shown to be a three-dimensional topological insulator.6−8 In its crystalline state at ambient conditions, Sb2Te3 has a 5 rhombohedral structure [space group R3̅m (D3d ), lattice constants a = 4.264 Å, c = 30.458 Å] with five atoms in a unit cell.6,9 In Figure 1a and b, the crystal structure is shown. The three-dimensional view in Figure 1a shows the rhombohedral unit cell (black). Along the z axis (crystal c axis), the material is characterized by a layer structure with hexagonal close-packed atomic layers. It can be divided into strongly coupled quintuple layers, which are terminated by a tellurium layer. Between the quintuple layers, the coupling is predominately of van der Waals type and therefore much weaker than the other bonds.6 Figure 1b shows a top view on a quintuple layer along the z axis, which reveals a 3-fold rotation symmetry of the system. Looking from the side at one quintuple layer, one can count the uppermost atoms at the exposed surfaces. The ratio for Sb:Te evaluates to 3:3 and 2:3, respectively, as can be seen in Figure 1c. Reflectivity measurements have shown that Sb2Te3, which is not intentionally doped, has a plasma edge around 1000 cm−1.10−14 This plasma edge originates from an excess hole concentration that develops due to negatively charged antisite defects (Sb on a Te position).15 Typical free charge carrier concentrations for Sb2Te3 are between 1019 cm−3 and 1021 cm−3.11−13 We performed nano-optical investigations on Sb 2 Te 3 hexagonal platelets grown from solution via solvothermal synthesis.16 In brief, the reaction was performed in a Teflon © XXXX American Chemical Society

Figure 1. Crystal structure of Sb2Te3. The three-dimensional view (a) shows a rhombohedral unit cell and the hexagonal close-packed atomic layers aligned along the z axis. The atomic layers can be divided into strongly coupled quintuple layers. The top view on a quintuple layer (b) reveals a 3-fold rotation symmetry of the system. Te1 and Te2 denote different chemical states. The lateral surface of one quintuple layer is illustrated in (c) (solid lies: first row, dashed lines: second row). Depending on the crystallographic orientation [A or B in (b)], the Sb concentration is different. (a) and (b) are modified from ref 6.

lined autoclave by heating antimony- and telluriumdioxide in the presence of polyvinylpyrrolidone (PVP), which is applied as a capping ligand, under basic solvothermal conditions. The resulting platelets are interesting due to their large surface-tovolume ratio and the resulting multitude of possible applications17 including nanoelectroics.18 A scanning electron Received: September 26, 2014 Revised: March 23, 2015

A

DOI: 10.1021/nl503697c Nano Lett. XXXX, XXX, XXX−XXX

Letter

Nano Letters

to an extend that the steps become visible. The pyramidal shape of the growth spiral has therefore been subtracted. Figure 3a shows the optical images recorded with s-SNOM of the exact same region as the topography image shown in Figure 2b for five different illumination wavelengths in scattering amplitude S2 and phase ϕ2 (tip oscillation amplitude: 30 nm). The measured signals are normalized to the respective values measured on the silicon substrate. All optical images show a very distinct pattern with trigonal rotation symmetry. Depending on the wavelength, the contrasts change and can even be inversed. In case of the amplitude, there are two inversion points between 957 and 1150 cm−1 and between 1600 and 1850 cm−1, whereas the phase images only show one inversion point between 1250 and 1600 cm−1. A schematic overlay of the optical pattern with the topography of the growth spiral is illustrated in Figure 3b. Obviously, both the optical pattern and the growth spiral show a trigonal rotation symmetry around the screw dislocation in the center. Apart from that, however, the optical and topographical information differ a lot. Similar near-field optical structures have been found on various hexagonal platelets that were grown by solvothermal synthesis. It has been verified that the measured contrast does not depend on the orientation of the sample with respect to the direction of illumination, that is, it is not affected by far-field polarization or an interference effect with the direct illumination. Also a nonlinear behavior with the illumination power has been excluded within the experimentally applicable range (about 1 order of magnitude). With complementary methods such as visible optical microscopy including crossed polarizers24 and transmission electron microscopy (TEM, c.f. Supporting Information), we did also not find any comparable pattern. In combination with the strong spectral variation of the observed near-field contrasts, this leads us to the conclusion that the feature can only be observed in the energy range of mid-IR light. Before comparing different growth mechanisms for Sb2Te3, first the near-field contrasts in Figure 3a will be analyzed in more detail. Besides the obvious optical pattern there is a smooth long-range gradient observable toward the edge of the platelet. This could be due to an independent sample property or due to an effect of indirect illumination of the tip.25 A reliable reference to the silicon substrate, as commonly used for nanometer-sized objects, is therefore hampered due to the huge lateral extend of the platelet. Because the focus here is on the origin of the optical pattern with its sharp boundaries, the average relative contrast between the two domains marked with A and B in Figure 3b have been extracted within 500 nm radius around the screw dislocation in every image. The results are plotted in Figure 3c and d for the scattering amplitude and phase, respectively (dots, upper plots). The inversion of contrast can be seen here as an intersection with the horizontal lines at S2(A)/S2(B) = 1 and ϕ2(A) − ϕ2(B) = 0, respectively. To explain the observed optical behavior, we give a quantitative interpretation in terms of regions with different charge carrier densities. The sharp boundaries within the optical pattern in Figure 3a suggest the presence of two different domains. However, we emphasize that our SAED investigations do not indicate the presence of crystalline domains with different crystallographic orientations (c.f. Supporting Information). Because the mid-IR dielectric properties of Sb2Te3 are dominated by the Drude contribution, and the plasma frequency can be supposed within the investigated

micrograph of a platelet is shown in Figure 2a. The diameter of the platelets can be up to 10 μm, whereas their height is only in

Figure 2. (a) SEM image of a Sb2Te3 hexagonal platelet with a diameter of more than 10 μm. The topography (AFM) in (b) reveals a growth spiral. The recorded area is marked by the black square in (a). (c) shows the geometry under which the platelets were imaged. The red arrow marks the illumination of the tip. The near-fields (red lines) are dominantly oriented parallel to the crystal c axis.

the order of 300 nm. Selective-area electron diffraction (SAED) measurements were performed to ensure that the crystallographic orientation within the platelets is uniform (c.f. Supporting Information). The quasi-two-dimensional structure of the platelets can be explained by the van der Waals coupling between the quintuple layers, which causes the growth along the crystal c axis to be much slower than in other directions.9,17 Due to the size of the platelets, a detailed analysis of material properties and possible substructures on individual hexagons is challenging compared to bulk material or extended films. We imaged the platelets using scattering-type scanning nearfield optical microscopy (s-SNOM). This method is based on atomic force microscopy (AFM) and makes use of the nearfields that appear at the AFM tip under illumination, as sketched in Figure 2c. In addition to the topographic information from AFM, s-SNOM offers a wavelengthindependent optical resolution in the order of 20 nm.19 The fields scattered by the tip are distinguished from the unavoidable background by oscillating the tip vertically above the sample at a frequency Ω and demodulating the detector signal at higher harmonics nΩ.20 Additionally, a pseudohyterodene interferometric detection enables the background-free determination of the near-field scattering amplitude Sn and phase ϕn, separately.21 We used a commercially available microscope (Neaspec GmbH) and standard metalized AFM tips (NanoWorld Arrow NCPt) in combination with quantum cascade lasers (Daylight Solutions) and a CO2 laser (Edinburgh Instruments) as mid-infrared (mid-IR) light sources. On regular platelets, one or more growth spirals with a very regular trigonal shape can be found in the topography image (AFM). They indicate the presence of a screw dislocation with a Burgers vector along the crystal c axis in the center.22,23 A topography map of a growth spiral is shown in Figure 2b. The height of the steps is about 1 nm, which corresponds to the height of one quintuple layer. Note that the image is flattened B


Letter

Nano Letters

Figure 3. (a) Near-field optical images in scattering amplitude S2 (first row) and phase ϕ2 (second row), normalized to silicon (upper left corner). The five columns show measurements for different illumination frequencies. A schematic overlay of the optical pattern from (a) with the topography of the growth spiral from Figure 1e is shown in (b). In (c) and (d), the relative contrast between the domains, as labeled in (b), is plotted in amplitude and phase (dots). In the inset of (c), the area of 500 nm radius around the screw dislocation that is used for the evaluation of the relative contrasts is sketched. The lines are calculated near-field contrasts under the assumption of different Drude terms for the domains A and B: NA = 2.1 × 1020 cm−3, NB = 1.5 × 1020 cm−3. In the lower plots the respective contrast to silicon is shown. (e) and (f) Illustrations of the effect of a variation of ±30% around the extracted damping factor γ = 540 cm−1 while N is kept constant at 1.8 × 1020 cm−3. Crossing points which indicate an inversion of the contrast between two spectra are marked with vertical dashed lines.

where c is the speed of light, e is the elementary charge, and ϵ0 is the vacuum permittivity. The optical parameters ϵ∞ = 29.5 and m* = 0.25m0 (for E∥c) are taken from ref 12. In our calculation, the only variable parameter is the free charge carrier density NA,B of the domains A and B. NA and NB were fitted in a way that the resulting curves reproduce the contrast inversion of the experiments correctly in both amplitude and phase. The resulting values are NA = 2.1 × 1020 cm−3 and NB = 1.5 × 1020 cm−3. The error of NA,B lies within about 10%, otherwise the frequencies where the contrast is inverted are not correctly reproduced. In the lower panels of Figure 3c and d, the corresponding calculated near-field contrasts with respect to silicon are shown. To grasp the spectral width of the optical signature, the damping constant is set to γ = 540 cm−1. We have to note that for bulk Sb2Te3, the damping constant is usually much lower with γ = 200...300 cm−1.12,13 Nanostructures, however, have a much larger surface-to-volume ratio, and thus, it is not surprising that the scattering rate can be dramatically increased, for example, due to surface scattering. Experimentally, a larger damping constant has been already reported in literature during

spectral range,10−14 the optical contrasts are interpreted as domains with different charge carrier density.26 The solid lines in Figure 3c and d are calculated near-field contrasts using the finite dipole model,25 which describes the optical interaction between the tip and the sample. To this end, the tip is modeled with a dipole that is polarized both by the illuminating field and by the Fresnel reflection of the resulting near-fields at the sample surface, that is, by its mirror image. The field that is scattered by the tip is proportional to its polarization and therefore depends on the reflectivity of the sample. The dielectric function of Sb2Te3 is described by the Drude model ⎛ ν2 ⎞ ϵ(ν) = ϵ∞⎜1 − 2 P ⎟ ν + iνγ ⎠ ⎝

(1)

with νP =

1 2πc

Ne 2 ϵ0ϵ∞m*

(2) C


Letter

Nano Letters

Figure 4. Near-field images of differently grown Sb2Te3 samples. (a) Solvothermally synthesized hexagonal platelet that exhibits a more complicated optical pattern due to the presence of several growth spirals in scattering amplitude. In (b), the region marked by the square in (a) is shown in higher magnification. (c) Image of a MOVPE film (bright) on silicon (dark). Growth spirals were not found on the film. The upper scale bar refers to (a)− (c). (d) Topography of an MBE film. Note that the appendant scale bar is saturated. The highest feature is in the range is 230 nm. (e) Optical image corresponding to (d). The optical image of a crystalline sputtered film is shown in (f). The lower scale bar refers to (e) and (f).

the very first infrared near-field optical analysis of free charge carriers in semiconductor nanowires.27 In Figure 3e and f, we demonstrate the effect of the damping factor on the calculated near-field spectra for a variation of ±30% around γ = 540 cm−1. Obviously, a lower damping factor leads to a sharper and more pronounced resonance, which shows that also the damping factor strongly affects the near-field contrast. However, we assured that the measured data cannot be explained with local variations in γ and a constant value for N. The vertical dashed lines in Figure 3e and f show that in this case, there would be only one crossing point (inversion of contrasts) in the amplitude spectra, whereas there would be two inversions of contrasts in the phase spectra. This is in clear contradiction to our measured data. A different supposable contrast mechanism that needs to be excluded is based on birefringence effects and strain. The strong optical anisotropy12 of Sb2Te3 suggests that the material exhibits birefringence in the mid-IR spectral range. Birefringence effects have been observed in near-field optical images.28−30 However, this would require domains with substantially different orientation of the crystal c axis, which is not expected for the platelets. A different source of birefringence is strain, which can be observed around defects. For example, in silicon carbide, birefringence can be observed around micropipes,24 which are hollow core dislocations with a large Burgers vector. However, because the Burgers vector in case of the screw dislocations in Sb2Te3 is only one or a few quintuple layers in the z-direction, strain fields should be much lower. A further argument that stands against any effect associated with strain, such as a modification of the damping of free charge carriers,31 is that strain-fields relax with increasing distance from their source,31 which causes the contrast to blur out. Additionally, strain in van der Waals crystals relaxes over very short distances.32 The images in Figure 3a, however, show sharp boundaries over distances of several micrometers from the center of the pattern. This argument also excludes stressinduced defects15 as a reason. Alternatively, effects associated with the surface conductivity are possible. Defects like screw dislocations and grain boundaries have a local effect on the surface states in

topological insulators.32−34 Also, the presence of low contents of substitute atoms and the thickness of the crystal can have drastic influences on the topological surface states.35 In principle, the surface conductivity of topological insulators should be observable in s-SNOM because they support surface waves36 similar to the sheet conductivity of graphene.37−40 However, due to the high intrinsic (bulk) charge carrier density of Sb2Te3 the Fermi level usually lies within the valence band, which causes the surface states to be unoccupied.7 As for other three-dimensional topological insulators like Bi2Se341 and Bi2Te3,42 a possible optical signature of surface states, therefore, is most likely masked by bulk effects. This leads to the assumption that the optical pattern in Figure 3a is not explicitly related to the surface states of Sb2Te3. Other electronic causes, such as recently reported intraband transitions in the frequency range of our observations,14 cannot be ruled out completely. Only if the plasma frequency is slightly below this transition, it becomes discernible in the IR reflection spectra. Because the near-field optical microscope is extremely sensitive to small variations in the permittivity close to a resonance (ϵ ≈ −1...−3), indeed a localized modeonly occurring in certain areas of the samplecould potentially lead to a shifted response, even with a fixed charge carrier concentration. However, such a localized mode suggested in ref 14 is not observed for pure Sb2Te3. Instead, it only appears in solid solutions of Sb2Te3 with Bi2Te3 or when doping with Sn. Because in our sample neither Bi nor Sn is present, we think we can exclude also the explanation of our results by such a localized intraband transition mode. Due to these arguments and the fact that the first-mentioned interpretation leads to a good agreement between theory and experimental data, we conclude that solvothermally synthesized platelets exhibit a domain structure with different free charge carrier densities. The pattern shown in Figure 3 could be qualitatively reproduced on most solvothermally synthesized platelets from various syntheses. Differences are that the angles enclosed by domains A and B vary from platelet to platelet. Also, the precise values of NA and NB are not necessarily identical for all platelets. The 3-fold symmetry, however, is always the same. In many cases the structures are more complicated because two or more patterns can be overlaid, D


Letter

Nano Letters which can be attributed to the presence of more than one growth spiral. A representative example is shown in Figure 4a and b (detail). Additionally, in Figure 4 the near-field optical behavior of the Sb2Te3 platelets is compared to Sb2Te3 films grown by other means. All images were taken at 1250 cm−1, where the domain contrast according to Figure 3a is expected to be high. Figure 4c shows an Sb2Te3 film (bright) grown on a silicon substrate via metalorganic vapor phase epitaxy (MOVPE). A hexagonal shape of the crystallites can also be found throughout the sample. However, neither growth spirals nor any comparable optical contrast has been found, which leads to the conclusion that the charge carrier density is uniform within the MOVPE layer. The relative contrast to silicon is comparable to the one of solvothermally synthesized hexagons. Therefore, it can be assumed that the charge carrier density is in a similar range. The wavy modulation with a period of about 5 μm that is overlaid to Figure 4c is most likely a far-field illumination artifact due to an insufficiently suppressed background. In Figure 4d and e, an Sb2Te3 film grown by molecular beam epitaxy (MBE) is shown in topography and near-field scattering amplitude. Because the film is closed, the signal is not referenced to the substrate. Also here, hexagonally shaped crystallites can be found, which in this case also exhibit small pyramidally shaped growth spirals. However, the spirals are not accompanied by characteristic optical patterns. Interestingly, some crystallites are tilted and, therefore, presumably have a different crystallographic orientation. In the optical image these crystallites appear significantly darker which could be evidence for the optical anisotropy. Figure 4f shows the scattering amplitude of a crystalline Sb2Te3 film that was sputtered. The crystallites here are too small to be examined in detail. The dynamic range, however, is the same as for the MBE film. Solvothermally synthesized Sb2Te3 hexagonal platelets reveal a pattern in mid-IR near-field images that is not found on samples grown by other means (MOVPE, MBE, and sputtering). The pattern is always correlated with a growth spiral. According to the evaluation of the observed contrast the optical contrast mechanism is most likely due to a domain structure with different intrinsic charge carrier densities. The open question that remains is what causes this domain structure. It can be presumed that the explanation lies in the specific growth mechanism of platelets in solution. Recently, Zhuang et al. investigated the screw-dislocationdriven growth of (isomorphous) Bi2Se3 platelets.43 An important result of their work is that the development of platelets from solution happens via bidirectional spiral growth, that is, both surfaces of the platelets develop equivalently via spiral growth from one common screw dislocation. The chirality of this screw determines the rotational direction of both spirals. Figure 5a shows a three-dimensional view of a platelet where also the growth spiral on the backside is illustrated (red, dashed) according to ref 43. Interestingly, the orientation of corners and edges of the two spirals is exchanged. This can also be seen in the top view in Figure 5b. It is presumably owed to this behavior that the platelets develop as hexagons rather than triangles. The bidirectional spiral growth is a fundamental difference between platelets grown from solution and any other technique where a substrate is involved. Both sides of the chemically synthesized platelets are geometrically equivalent. Therefore, the platelet itself, and the unit cell of Sb2Te3 have the same inversion symmetry. Due to the interchanged roles of edges and

Figure 5. (a) Three dimensional view of a hexagonal platelet showing the two growth spirals. The red dashed lines illustrate the spiral on the backside of the hexagon. Both spirals emerge from one screw dislocation (thick black line). The top view in (b) reveals that the position of corners and edges of the triangular screws on front side and backside are exchanged. (c) Sketch of the relation between one growth spiral and hypothetic twin domains (gray and white). As illustrated by the hexagonal symbols in the dashed box to the right, the unit cell is rotated by 180° with respect to one another. The symbols A and B refer to the lateral surfaces of the a quintuple layer, as labeled in Figure 1b and c.

corners, however, it seems likely that in an early stage of development the lateral growth in each of the six directions is dominated by either of the two spirals. This might lead to the development of so-called growth twins, which are domains with a unit cell that is rotated by 180° about the c axis with respect to one another (or equivalently the stacking order in z direction is reversed).44 For the example of cadmium telluride (zincblende lattice), it has been shown that depending on the growth conditions and the crystallographic direction of step growth, different twin domains can develop in a reproducible way.45 The twin domains reported in ref 45 geometrically resemble the optical signature in Figure 3a. In Figure 5d, the growth of one spiral after the formation of the twin domains (gray versus white) is sketched. During the development of a hexagon, the crystallographic structure that is determined by the already existing layers is further pursued, resulting in six single-domain regions.45 Because these growth twins must develop already in the initial nucleation process, it seems likely that differences between the growth dynamics of the two spirals cause the angles enclosed by the regions to vary from one platelet to another. As shown in Figure 5c, the growth of the two different regions A and B at the investigated surface is dominated by either edges or corners of the growth spiral. This may lead to different enclosed angles in the observed pattern. Boundaries between growth twins are known to have an effect on thermal transport due to phonon scattering without degrading the electronic properties.44,46 It is, however, not clear how a domain structure of differently oriented twins could directly lead to an optical contrast, especially since the frequency of infrared active phonons is below the measured spectral range, as mentioned above. Also, a charge polarization cannot be expected from the crystal structure. Even single quintuple layers, which consist of five hexagonal layers in the stacking order Te−Sb−Te−Sb−Te are not polar. E


Letter

Nano Letters

Helpful commends of G. Roth and M. Morgenstern (Aachen) are acknowledged. This study was financially supported by the DFG in the framework of the SFB 917, the Excellence Initiative of the federal and state governments, and the Ministry of Innovation of North Rhine-Westphalia.

However, the sketch in Figure 5c reveals that depending on the domain, a different crystallographic surface is laterally exposed at the edges of the spiral (A or B in Figure 1c). In epitactically grown films, it has been reported that different directions of growth go along with slight deviations from stoichiometry,47 which can be explained with the different surface stoichiometry of the laterally exposed crystal faces (c.f. Figure 1c). This effect could potentially be much stronger for solvothermally grown platelets because the crystallites that develop into a hexagon vary significantly in stoichiometry.16 Twin domains in Sb2Te3 platelets might therefore lead to the formation of regions with different densities of antisite defects (regions A or B in Figure 3). Because antisite defects act as dopants in Sb2Te3, this effect could explain the observed optical contrast. Our suggested explanation therefore links an electronic effect (the charge carrier density) to a crystallographic phenomenon (growth twins) with sharp domain boundaries. In conclusion, the s-SNOM images reveal information that might be related to the specific growth process of solvothermally synthesized Sb2Te3 hexagonal platelets. Even though the origin of the observed domain structure could not be definitely determined, the s-SNOM images clearly show regions with different optical properties. According to our analysis this optical contrast can be traced back to different charge carrier densities. This unexpected finding might be relevant for the application of chemically synthesized platelets as building blocks for novel electronic devices, such as nanoscale electronic switches or data storage devices. It has been shown that s-SNOM provides a very direct and convenient way for the visualization of the domain structure on Sb2Te3 hexagonal platelets. The strength lies in the high spatial resolution in combination with the low requirements to the sample fabrication. Especially, a thin layer (∼1 nm) of PVP that covers the synthesized platelets16 can be an obstacle for any alternative method that relies on electric contacts. However, further studies are required for a definitive explication of the origin of contrast. In the future, the interpretation of the sSNOM results in terms of free charge carriers could possibly be approved by PEEM48 or high-resolution EELS.49 Also, confocal Raman spectroscopy could, in the case of Sb2Te3, provide a way to address the charge carrier density via coupled plasmon− phonon modes.12

■

■

ASSOCIATED CONTENT

S Supporting Information *

Selective-area electron diffraction (SAED) data. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Wuttig, M. Phys. Status Solidi B 2012, 249, 1843−1850. (2) Wuttig, M.; Yamada, N. Nat. Mater. 2007, 6, 824−832. (3) Lencer, D.; Salinga, M.; Grabowski, B.; Hickel, T.; Neugebauer, J.; Wuttig, M. Nat. Mater. 2008, 7, 972−977. (4) Wright, D. A. Nature 1958, 181, 834−834. (5) Snyder, G. J.; Toberer, E. S. Nat. Mater. 2008, 7, 105−114. (6) Zhang, H.; Liu, C.; Qi, X.; Dai, X.; Fang, Z.; Zhang, S. Nat. Phys. 2009, 5, 438−442. (7) Hsieh, D.; Xia, Y.; Qian, D.; Wray, L.; Meier, F.; Dil, J. H.; Osterwalder, J.; Patthey, L.; Fedorov, A. V.; Lin, H.; Bansil, A.; Grauer, D.; Hor, Y. S.; Cava, R. J.; Hasan, M. Z. Phys. Rev. Lett. 2009, 103, 146401. (8) Plucinski, L.; Herdt, A.; Fahrendorf, S.; Bihlmayer, G.; Mussler, G.; Dö ring, S.; Kampmeier, J.; Matthes, F.; Bü rgler, D. E.; Grützmacher, D.; Blügel, S.; Schneider, C. M. J. Appl. Phys. 2013, 113, 053706. (9) Shi, W.; Zhou, L.; Yang, J.; Zhang, H. Adv. Mater. 2008, 20, 1892−1897. (10) Horák, L.; Tichý, L.; Vaško, A.; Frumar, M. Phys. Status Solidi A 1972, 14, 289−298. (11) Langhammer, H. T.; Stordeur, M.; Sobotta, H.; Riede, V. Phys. Status Solidi B 1982, 109, 673−679. (12) Richter, W.; Krost, A.; Nowak, U.; Anastassakis, E. Z. Phys. B 1982, 49, 191−198. (13) Dordevic, S. V.; Wolf, M. S.; Stojilovic, N.; Lei, H.; Petrovic, C. J. Phys. (Paris) 2013, 25, 075501. (14) Stepanov, N. P.; Kalashnikov, A. A.; Ulashkevich, Y. V. Opt. Spectrosc. 2014, 117, 401−405. (15) Thonhauser, T.; Jeon, G. S.; Mahan, G. D.; Sofo, J. O. Phys. Rev. B 2003, 68, 205207. (16) Zhang, G.; Wang, W.; Lu, X.; Li, X. Cryst. Growth Des. 2009, 9, 145−150. (17) Morin, S. A.; Forticaux, A.; Bierman, M. J.; Jin, S. Nano Lett. 2011, 11, 4449−4455. (18) Kong, D.; Dang, W.; Cha, J. J.; Li, H.; Meister, S.; Peng, H.; Liu, Z.; Cui, Y. Nano Lett. 2010, 10, 2245−2250. (19) Taubner, T.; Hillenbrand, R.; Keilmann, F. J. Microsc. 2003, 210, 311−314. (20) Keilmann, F., Hillenbrand, R. In Nano-optics and near-field optical microscopy; Zayats, A., Richards, D., Eds.; Artech House: Boston/London, 2009; Chapter 11, pp 235−265. (21) Ocelić, N.; Huber, A.; Hillenbrand, R. Appl. Phys. Lett. 2006, 89, 101124. (22) Burton, W. K.; Cabrera, N.; Frank, F. C. Nature 1949, 163, 398−399. (23) Frank, F. C. Adv. Phys. 1952, 1, 91−109. (24) Ouisse, T.; Chaussende, D.; Auvray, L. J. Appl. Crystallogr. 2010, 43, 122−133. (25) Cvitković, A.; Ocelić, N.; Hillenbrand, R. Opt. Express 2007, 16, 8550−8565. (26) The frequency of the infrared-active phonons of Sb2Te3 are below12 and the indirect electronic band gap is above6 the spectral range covered in this study. (27) Stiegler, J. M.; Huber, A. J.; Diedenhofen, S. L.; Gómez Rivas, J.; Algra, R. E.; Bakkers, E. P. A. M; Hillenbrand, R. Nano Lett. 2010, 10, 1387−1392. (28) McDaniel, E. B.; Hsu, J. W. P. J. Appl. Phys. 1996, 80, 1085− 1093. (29) Kehr, S. C.; Cebula, M.; Mieth, O.; Härtling, T.; Seidel, J.; Eng, L. M.; Winnerl, S.; Stehr, D.; Helm, M. Phys. Rev. Lett. 2008, 100, 256403.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: + 49 (0)241 80 20260. Fax: + 49 (0)241 80 620260. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the following persons for providing samples used in this study: S. Rieß, H. Hardtdegen (Jülich − MOVPE), J. Kampmeier, G. Mussler (Jülich − MBE), F. Lange, M. Wuttig (Aachen − sputter deposition). M. Bornhöfft (Aachen) is acknowledged for performing TEM measurements on platelets. F


Letter

Nano Letters (30) Jones, A. C.; Berweger, S.; Wei, J.; Cobden, D.; Raschke, M. B. Nano Lett. 2010, 10, 1574−1581. (31) Huber, A. J.; Ziegler, A.; Köck, T.; Hillenbrand, R. Nat. Nanotechnol. 2009, 4, 153−157. (32) Jia, J. Nat. Phys. 2014, 10, 247−248. (33) Ran, Y.; Zhang, Y.; Vishwanath, A. Nat. Phys. 2009, 5, 298−303. (34) Liu, Y.; Li, Y. Y.; Rajput, S.; Gilks, D.; Lari, L.; Galindo, P. L.; Weinert, M.; Lazarov, V. K.; Li, L. Nat. Phys. 2014, 10, 294−299. (35) Wu, L.; Brahlek, M.; Valdés Aguilar, R.; Stier, A. V.; Morris, C. M.; Lubashevsky, Y.; Bilbro, L. S.; Bansal, N.; Oh, S.; Armitage, N. P. Nat. Phys. 2013, 9, 410−414. (36) Karch, A. Phys. Rev. B 2011, 83, 245432. (37) Fei, Z.; Andreev, G. O.; Bao, W.; Zhang, L. M.; McLeod, A. S.; Wang, C.; Stewart, M. K.; Zhao, Z.; Dominguez, G.; Thiemens, M.; Fogler, M. M.; Tauber, M. J.; Castro-Neto, A. H.; Lau, C. N.; Keilmann, F.; Basov, D. N. Nano Lett. 2011, 11, 4701−4705. (38) Fei, Z.; Rodin, A. S.; Andreev, G. O.; Bao, W.; McLeod, A. S.; Wagner, M.; Zhang, L. M.; Zhao, Z.; Thiemens, M.; Dominguez, G.; Fogler, M. M.; Castro-Neto, A.; Lau, C. N.; Keilmann, F.; Basov, D. N. Nature 2012, 487, 82−85. (39) Chen, J.; Badioli, M.; Alonso-González, P.; Thongrattanasiri, S.; Huth, F.; Osmond, J.; Spasenović, M.; Centeno, A.; Pesquera, A.; Godignon, P.; Zurutuza Elorza, A.; Camara, N.; García de Abajo, F. J.; Hillenbrand, R.; Koppens, F. H. L. Nature 2012, 487, 77−81. (40) García de Abajo, F. J. ACS Photonics 2014, 1, 135−152. (41) Di Pietro, P.; Vitucci, F. M.; Nicoletti, D.; Baldassarre, L.; Calvani, P.; Cava, R.; Hor, Y. S.; Schade, U.; Lupi, S. Phys. Rev. B 2012, 86, 045439. (42) Chapler, B. C.; Post, K. W.; Richardella, A. R.; Lee, J. S.; Tao, J.; Samarth, N.; Basov, D. N. Phys. Rev. B 2014, 235308. (43) Zhuang, A.; Li, J.; Wang, Y.; Wen, X.; Lin, Y.; Xiang, B.; Wang, X.; Zeng, J. Angew. Chem. 2014, 126, 6543−6547. (44) Medlin, D. L.; Ramasse, Q. M.; Spataru, C. D.; Yang, N. Y. C. J. Appl. Phys. 2010, 108, 043517. (45) Maruyama, K.; Yoshikawa, M.; Takigawa, H. J. Cryst. Growth 1988, 93, 761−765. (46) Cahill, D. G.; Ford, W. K.; Goodson, K. E.; Mahan, G. D.; Majumdar, A.; Maris, H. J.; Merlin, R.; Phillpot, S. R. J. Appl. Phys. 2003, 93, 793−818. (47) Ferhat, M.; Tecenac, J. C.; Nagao, J. J. Cryst. Growth 2000, 218, 250−258. (48) Günther, S.; Kaulich, B.; Gregoratti, L.; Kiskinova, M. Prog. Surf. Sci. 2002, 70, 187−260. (49) Bell, G. R.; McConville, C. F. Phys. Rev. B 1996, 54, 2654−2661.

G


Solvothermally Synthesized Sb2Te3 Platelets Show Unexpected

Recommend Documents