Efficient Photothermoelectric Conversion in Lateral Topological

Dec 5, 2016 - Strong Photothermoelectric Response and Contact Reactivity of the Dirac Semimetal ZrTe5. François Léonard , Wenlong Yu , Kimberlee C...
0 downloads 0 Views 4MB Size
Subscriber access provided by UNIVERSITY OF SOUTH CAROLINA LIBRARIES

Communication

Efficient photo-thermoelectric conversion in lateral topological insulator heterojunctions Soudabeh Mashhadi, Dinh Loc Duong, Marko Burghard, and Klaus Kern Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.6b03851 • Publication Date (Web): 05 Dec 2016 Downloaded from http://pubs.acs.org on December 5, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Efficient photo-thermoelectric conversion in lateral topological insulator heterojunctions

Soudabeh Mashhadi, 1,* Dinh Loc Duong,2,3 Marko Burghard,1 and Klaus Kern1,4 1

Max Planck Institute for Solid State Research, Heisenbergstrasse 1, D-70569 Stuttgart,

Germany 2

Center for Integrated Nanostructure Physics, Institute for Basic Science (IBS), Suwon 16419,

Republic of Korea 3

Sungkyunkwan University (SKKU), Suwon 16419, Republic of Korea

4

Institut de Physique, Ecole Polytechnique Fédérale de Lausanne, CH-1015 Lausanne,

Switzerland

ABSTRACT Tuning the electron and phonon transport properties of thermoelectric materials by nanostructuring has enabled improving their thermopower figure of merit. Three-dimensional topological insulators, including many bismuth chalcogenides, attract increasing attention for this purpose, as their topologically protected surface states are promising to further enhance the thermoelectric performance. While individual bismuth chalcogenide nanostructures have been studied with respect to their photo-thermoelectric properties, nanostructured p-n junctions of these compounds have not yet been explored. Here, we experimentally investigate the room temperature thermoelectric conversion capability of lateral heterostructures consisting of two different three-dimensional topological insulators, namely the n-type doped Bi2Te2Se and the ptype doped Sb2Te3. Scanning photocurrent microscopy of the nanoplatelets reveals efficient

*

To whom correspondence should be addressed: [email protected]

1 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

thermoelectric conversion at the p-n heterojunction, generated by hot carriers of opposite sign in the two materials. From the photocurrent data, a Seebeck coefficient difference of ΔS = 200 µV/K was extracted, in accordance with the best values reported for the corresponding bulk materials. Furthermore, it is in very good agreement with the value of ΔS = 185 µV/K obtained by DFT calculation taking into account the specific doping levels of the two nanostructured components. KEYWORDS:

thermopower,

thermoelectric

conversion,

topological

insulator,

lateral

heterostructure, p-n heterojunction

Direct heat-to-electricity conversion has attracted great interest in the field of renewable energy. A major challenge is to identify high-performance thermoelectric materials that effectively compete with conventional materials employed in energy conversion approaches. Well-established thermoelectric materials such as Bi2Te3 display a large Seebeck coefficient of up to 150 μV/K at room temperature1. Interestingly, they belong to the material class of threedimensional (3D) topological insulators (TIs), which are electrical insulators in the bulk, but possess metallic surface states. The latter are protected by time reversal symmetry in the absence of the non-magnetic impurities2. TIs share several common features with topologically trivial thermoelectric materials, including the presence of heavy elements which impart strong spinorbit coupling, and a small bulk band gap3,4. Theory suggests that hybridization between the nontrivial (helical) surface states from the top and bottom surfaces of a 3D TI can enhance the thermoelectric performance5. A positive impact of the topological surface state on the thermoelectric transport has been corroborated by ab initio electronic structure calculations on Sb2Te36. Moreover, acoustic phonon-mediated cooling of the Dirac fermions in the 3D TI might 2 ACS Paragon Plus Environment

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

favor the creation of long-lived hot carriers and hence enable high efficiency photothermoelectric applications7 . Experimentally, various types of 3D TI nanostructures have been investigated with respect to their thermoelectric performance. Scanning photocurrent microscopy (SPCM) of step-terraced two-dimensional (2D) crystals of Bi2Te3 and Sb2Te3 has revealed efficient thermoelectric conversion at monolayer steps8 . This observation has been attributed to optothermal motion of hot carriers through the surface states of the 3D TIs. In addition, an enhancement of the photo-thermoelectric effect in Bi2Se3 nanoribbons has been reported due to the excitation of spin-polarized currents by circularly polarized light9. Supporting evidence for the role of the spin-helical surface states in 3D TIs has recently been gained for ultrathin sheets of (BixSb1-x)2Te2, which display an enhanced spin Seebeck effect signal10. It has furthermore been demonstrated that the thermoelectric performance of ultrathin Bi2Se3 nanosheets can be optimized through thickness control11. Despite the promising thermoelectric conversion efficiency of single-component 3D TI nanostructures, thermoelectric devices that integrate the corresponding n- and p-type components at the nanoscale have not yet been achieved. In fact, while both lateral and vertical TI p-n heterostructures have been successfully prepared by molecular beam epitaxy12, solvothermal synthesis13 or vapor phase growth14, only their general electrical properties have been studied. Here, we report the photo-thermoelectric response of individual lateral heterostructures composed of two different 3D TIs, specifically Bi2Te2Se (BTS) as n-type component and Sb2Te3 as the p-type component. In the BTS crystal, ordered hexagonal Te-Bi-Se-Bi-Te quintuple layers are stacked via van der Waals interactions, while Sb2Te3 is made of stacked quintuple layers of Te-Sb-Te-Sb-Te. The two tetradymite crystal structures belong to the same space group R3� m, rendering the two materials suitable candidates for fabricating lateral TI heterojunctions15,16. 3 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The BTS/Sb2Te3 nanoplatelets were synthesized through two consecutive steps of vaporsolid growth. In the first step, BTS nanoplatelets were grown on a Si/SiO2 substrate, as this compound has a higher melting point than Sb2Te3, and is hence able to sustain the growth conditions required for the latter compound. The BTS growth was performed using a source temperature of 582°C with the deposition substrate kept at 480°C. After cooling to room temperature, the substrate was transferred into another furnace where Sb2Te3 was grown using a source temperature of 570°C and a deposition substrate temperature of 420°C. For shorter Sb2Te3 deposition times, lateral heterostructures were obtained, while prolonged deposition yielded vertical heterostructures wherein a closed Sb2Te3 film completely covers the underlying BTS nanoplatelets. For the lateral heterostructures, the thickness ratio between the central BTS nanoplatelet and the Sb2Te3 frame at the periphery could be adjusted by the growth parameters (see Supporting Information Fig.S2). The respective n- and p-type doping character of the singlecomponent BTS and Sb2Te3 nanoplatelets could be confirmed by Hall measurements on individual platelets (see Supporting Information, Fig. S6). The atomic force microscopy (AFM) image in Fig.1a shows the topography of a BTS/Sb2Te3 nanoplatelet comprising BTS and Sb2Te3 regions of notably different thickness. The line profile in Fig. 1b evidences that the central platelet with a thickness of 15 nm is surrounded by a ~40 nm thick frame region of Sb2Te3. AFM characterization of individual BTS nanoplatelets before and after Sb2Te3 growth revealed that the thickness of the former is largely preserved, signifying an edge-induced growth mechanism for the Sb2Te3. That the obtained heterostructures indeed have the desired composition could be verified by transmission electron microscopy (TEM)-based energy dispersive X-ray (EDX) analysis, as exemplified for the nanoplatelet in Fig. 1c. The corresponding EDX line profiles across the interface between the two regions of different 4 ACS Paragon Plus Environment

Page 4 of 28

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

thickness testify that Bi is present only in the central region, whereas Sb occurs only in the outer frame (Fig. 1d). The heterostructure composition could be further confirmed by confocal Raman spectroscopy (Fig. 2). The Raman spectrum of BTS consists of two isolated modes at 65 cm-1 and 109.5 cm-1 corresponding to Ag1 and Eg modes, respectively. In addition, there are two overlapping peaks at 140 cm-1 and 155 cm-1 associated with Ag2 mode17,18. While the origin of 140 cm-1 peak is somewhat controversial it is commonly attributed to anti-site defects. The Sb2Te3 Raman spectrum shows three peaks at 71 cm-1, 115 cm-1 and 168 cm-1 , which can be assigned to the A1g1, Eg2, A1g2 vibrational modes, respectively19,18. The Raman maps in Fig. 2 demonstrate that the A1g1 mode characteristic of Sb2Te3 (peak at ~71 cm-1, see the spectrum in Fig. 2b) appears predominantly within the frame region, while the corresponding mode belonging to BTS (peak at ~65 cm-1, see Fig. 2d) resides within the central region. Further, indirect proof for the chemical composition of the heterostructures was obtained by Kelvin probe force microscopy (KPFM), as illustrated in Fig. 3. The line profile in Fig. 3b, taken along the white line in the KPFM image in Fig. 3a, indicates a surface potential difference of ~130 mV between the inner BTS and the outer Sb2Te3 region. In accordance with the work function of BTS (~5.2 eV20) exceeding that of Sb2Te3 (~4.6 eV20,21), a lower surface potential is detected on the frame region composed of Sb2Te3. For electrical characterization, individual BTS/Sb2Te3 nanoplatelets were provided with two separate metal contacts on the inner BTS and the outer Sb2Te3 region. The contacts were defined by e-beam lithography, followed by thermal evaporation of Ti/Au (4 nm/50 nm). In order to avoid contact between the BTS electrode and the Sb2Te3, the former was fabricated on top of an 80 nm thick, thermally evaporated SiOx layer. The representative current(I)-voltage(V) characteristics measured across the p-n heterojunction in Fig.4 disclose fully linear behavior at 5 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

both high (T = 290 K) and low (T = 1.4 K) temperatures. The resistance decrease upon cooling reflects a metallic character, which arises from the high doping degree of both heterostructure components, with the Fermi energy positioned within in the bulk conduction (valence) band for BTS (Sb2Te3). Similar metallic behavior has been documented for nanostructures made of the single components22. Remarkably, the fully linear I/V behavior of the devices was preserved even for a large bias window of ±1 V at T = 1.4 K. As the heterojunction is composed of two degenerately doped semiconductors, the devices would be expected to behave like an Esaki diode, exhibiting a negative differential resistance (NDR) peak in the backward bias direction23. However, the presence of the topological surface states is likely to cause deviation from the NDR characteristics in the tunneling regime, which could explain the smooth, linear I-V curves observed for the present samples. Another contribution may arise from the interface between BTS and Sb2Te3 being not sharp at the atomic scale, similar to observations made on other TI heterostructures24.. It is noteworthy that a small number of devices showed a weak rectifying behavior along with tiny wiggles, however, in all cases the origin of these features could be traced back to asymmetric Schottky barrier contacts on the two materials (see Supporting Information, Fig.S3). Moreover, the two-probe magnetoresistance across the p-n junction (inset of Fig. 4) displays a close-to-quadratic dependence for higher B-fields, along with a dip around zero B-field. The latter is due to the weak-antilocalization (WAL) effect, which is characteristic of both, n- and p-type 3D TIs25,26, and manifests itself also in the present single component devices (see Supporting Information, Fig.S6). Having established the chemical composition and basic electrical behavior of the p-n heterojunctions, we investigated their photo-thermoelectric properties. As apparent from Fig. 5, when a laser spot (λ = 514 nm, laser power of 35 µW) is located at the p-n region within a lateral 6 ACS Paragon Plus Environment

Page 6 of 28

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

heterostructure, the original I-V curve (black line) is slightly down-shifted along the y-axis, indicating a very small open circuit voltage Voc on the order of 10 µV. It is noteworthy that the illumination causes a substantial resistance increase of ~230 Ω, which renders these devices into suitable photodetectors. This prominent resistance increase under the laser illumination and applied bias indicates the coexistence of the thermoelectric and bolometric effect. The photocurrent dependence on light power (see inset of Fig. 5) shows a close-to-linear increase. In order to gain spatially resolved information about the photo-electric behavior of the devices, we used scanning photocurrent microscopy (SPCM), whose principle is illustrated in Fig. 6a. Our SPCM set-up is operated under ambient with a focused green laser and a spatial resolution (laser spot size) of about 400 nm. The photocurrent generated during a scan is collected through the two electrodes as a function of the laser spot position. In Fig. 6b, the zero bias photocurrent map of a representative p-n heterojunction device (see upper inset) is shown. It displays pronounced photocurrents at the source and drain contacts, as well as at the p-n junction. The photocurrents generated at the two contacts have identical sign, which is opposite to that of the photocurrent at the junction. In general, two major mechanisms are responsible for photocurrent generation in such devices. The first one is the photovoltaic effect which involves the separation of photoexcited electron-hole pairs by a built-in electric field27 , while the second one is the photothermoelectric effect arising from the temperature gradient across an interface between two materials with different Seebeck coefficients28. We attribute the photocurrent signals to be dominated by the photo-thermoelectric effect for the following reasons. First, as apparent from the overlap between the photocurrent map and the reflection image, the signals close to the contact regions extend several µm away from them, which is in contrast to the more localized response typically detected in case of the photovoltaic effect29 (See Supporting Information

7 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Fig.S4). Second, the full linearity of the I-V curves over a wide bias range (Fig. 4) speaks against the presence of sizeable Schottky barriers at the contacts. Third, we have reproducibly observed a stronger photocurrent signal at the contact to the BTS, which is in accordance with Seebeck coefficient difference between gold and BTS being larger compared to that between gold and Sb2Te3. Furthermore, based upon the electron affinity of Sb2Te3 of between 4.1 and 4.5 eV15 and gold’s work function of ~5 eV , the photovoltaic mechanism would be expected to result in an opposite sign of the photocurrent at the electrode than observed by experiment. Taking these arguments together, it follows that due to the laser-induced local heating, electrons are diffusing from the p-type Sb2Te3 to the n-type BTS, leading to negative photocurrent in the depicted experimental setup. The photothermal current generated upon illumination at the junction is proportional to the difference between the Seebeck coefficients of the components, according to the following equation: S1-S2 = ∆V/∆T,

(1)

where S1 (S2) is the Seebeck coefficient of the p (n)-type material, ∆V is the generated voltage (∆V = ∆I⋅R, with ∆I as the induced photothermal current and R as the sample resistance), and ∆T is the temperature difference induced through local heating by the laser spot30. In order to gain access to the temperature increase ∆T, we investigated the electrical behavior of individual BTS nanoribbons contacted by two gold electrodes. Measurements of the dark resistance of the BTS nanoribbons as a function of temperature (between 1.4 and 298 K) yielded a temperature coefficient of dR/dT = 3.2 Ω/K (see Supporting Information, Fig.S5). In complementary experiments performed under ambient, the middle section of the BTS nanoribbon between the

8 ACS Paragon Plus Environment

Page 8 of 28

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

two contacts was illuminated by the green laser. Assuming that the laser-induced resistance change ∆R under this condition originates from local heating, the temperature difference ∆T between the illuminated spot and the electrodes is estimated using ∆T ≈ ∆R⋅(dR/dT)-1 to be 0.3 ± 0.05 K for a laser power of 5.2 µW (0.4 ± 0.02 K for a power of 8.7 µW) averaged over different samples. It should be emphasized that the above estimation is limited to a maximum laser power of 20 µW. For higher power, the appreciable concentration of photo-excited carriers enhances the optical absorption and thereby favors further heating, which would lead to an overestimation of temperature.. For the lateral heterostructure device (Fig. 6), inserting the measured resistance R, the magnitude of the photocurrent ∆I detected at the p-n junction, and the estimated temperature ∆T difference into equation(1) yields a Seebeck coefficient difference of ~200 µV/K. This value exceeds those reported for single component bismuth chalcogenide nanostructures such as Bi2Te3 and Sb2Te3 nanowires (~150 µV/K)1,9,30. To investigate the accuracy of our estimation, we further performed a calculation of the Seebeck coefficient by a semi-classical transport approach using BoltzTrap code (See Supporting Information). The Seebeck coefficient is dependent on the Fermi energy as shown in Fig. 7. From separate Hall measurements on individual BTS and Sb2Te3 nanoplatelets, we obtained an electron concentration of n = 1.2⋅1019 cm-3 for BTS, and a hole concentration of p = 7.4⋅1020 cm-3 for Sb2Te3, confirming the strong (degenerate) doping degree of both compounds (see Supporting Information, Fig. S6). Based upon the effective mass of 0.11me31 for electrons in BTS and 0.78me for holes in Sb2Te332, and assuming for both compounds an approximately quadratic band structure close to the Fermi level, we estimate from the equation

𝐸𝐸𝐹𝐹 =

ћ2

2𝑚𝑚

(3𝜋𝜋 2 𝑛𝑛)2⁄3 (2) 9 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

the Fermi level of BTS to be 0.17 eV above the conduction band edge, and the Fermi level of Sb2Te3 to be 0.23 eV below the valence band edge. Combined with the calculated data in Fig. 7, this yields a Seebeck coefficient of about -120 µV/K for BTS and +65 µV/K for Sb2Te3. The corresponding difference S2-S1 = 185 µV/K is in excellent agreement with the experimental observation. In summary, we have successfully synthesized lateral p-n heterojunctions composed of the ntype BTS and the p-type Sb2Te3 as two different 3D TIs of opposite doping characteristic. The well-defined composition and electronic structure of the BTS/Sb2Te3 nanoplatelets endows them with a high thermoelectric performance, as demonstrated by our photo-thermoelectric measurements. In future experiments, the device performance might be further improved by tuning the Fermi levels through local molecular surface doping. The heterostructures are amenable to implementation into a more elaborate device configuration that allows directly harnessing environmental temperature gradients. Moreover, they are promising as components of novel, high performance photodetectors.

Associated content Supporting Information Available: Description of the growth of heterostructures, rectifying I-V curves, estimation of laser-induced heating, magnetotransport data and information about DFT calculations. This material is available free of charge via the internet at http://pubs.acs.org.

Author Information Corresponding Author *E-mail: [email protected]

10 ACS Paragon Plus Environment

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Notes The authors declare no competing financial interest.

Acknowledgments We would like to acknowledge Peter Kopold and Peter van Aken for assistance with the transmission electron microscopy measurements. We are grateful to Pascal Gehring for valuable discussions.

References (1)

Kim, J.-H.; Choi, J.-Y.; Bae, J.-M.; Kim, M.-Y.; Oh, T.-S. Mater. Trans. 2013, 54 (4), 618–625.

(2)

Hasan, M. Z.; Kane, C. L. Rev. Mod. Phys. 2010, 82 (4), 3045–3067.

(3)

Shi, H.; Parker, D.; Du, M.-H.; Singh, D. J. Phys. Rev. Appl. 2015, 3 (1), 14004.

(4)

Müchler, L.; Casper, F.; Yan, B.; Chadov, S.; Felser, C. Phys. status solidi – Rapid Res. Lett. 2013, 7 (1–2), 91–100.

(5)

Ghaemi, P.; Mong, R. S. K.; Moore, J. E. Phys. Rev. Lett. 2010, 105 (16), 166603.

(6)

Hinsche, N. F.; Zastrow, S.; Gooth, J.; Pudewill, L.; Zierold, R.; Rittweger, F.; Rauch, T.; Henk, J.; Nielsch, K.; Mertig, I. ACS Nano 2015, 9 (4), 4406–4411.

(7)

Wang, Y. H.; Hsieh, D.; Sie, E. J.; Steinberg, H.; Gardner, D. R.; Lee, Y. S.; Jarillo-Herrero, P.; Gedik, N. Phys. Rev. Lett. 2012, 109 (12), 127401.

(8)

Sung, J. H.; Heo, H.; Hwang, I.; Lim, M.; Lee, D.; Kang, K.; Choi, H. C.; Park, J.-H.; Jhi, S.-H.; Jo, M.-H. Nano Lett. 2014, 14 (7), 4030–4035.

11 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(9)

Yan, Y.; Liao, Z.-M.; Ke, X.; Van Tendeloo, G.; Wang, Q.; Sun, D.; Yao, W.; Zhou, S.; Zhang, L.; Wu, H.-C.; Yu, D.-P. Nano Lett. 2014, 14 (8), 4389–4394.

(10)

Jiang, Z.; Chang, C.-Z.; Masir, M. R.; Tang, C.; Xu, Y.; Moodera, J. S.; MacDonald, A. H.; Shi, J. Nat. Commun. 2016, 7, 11458.

(11)

Hong, M.; Chen, Z.-G.; Yang, L.; Han, G.; Zou, J. Adv. Electron. Mater. 2015, 1 (6), 1500025.

(12)

Lanius, M.; Kampmeier, J.; Weyrich, C.; Kölling, S.; Schall, M.; Schüffelgen, P.; Neumann, E.; Luysberg, M.; Mussler, G.; Koenraad, P. M.; Schäpers, T.; Grützmacher, D. Cryst. Growth Des. 2016, 16 (4), 2057–2061.

(13)

Min, Y.; Park, G.; Kim, B.; Giri, A.; Zeng, J.; Roh, J. W.; Kim, S. Il; Lee, K. H.; Jeong, U. ACS Nano 2015, 9 (7), 6843–6853.

(14)

Li, Y.; Zhang, J.; Zheng, G.; Sun, Y.; Hong, S. S.; Xiong, F.; Wang, S.; Lee, H. R.; Cui, Y. ACS Nano 2015, 9 (11), 10916–10921.

(15)

Zhang, H.; Liu, C.-X.; Qi, X.-L.; Dai, X.; Fang, Z.; Zhang, S.-C. Nat. Phys. 2009, 5 (6), 438–442.

(16)

Misra, S.; Bever, M. B. J. Phys. Chem. Solids 1964, 25 (11), 1233–1241.

(17)

Tian, Y.; Osterhoudt, G. B.; Jia, S.; Cava, R. J.; Burch, K. S. Appl. Phys. Lett. 2016, 108 (4), 41911.

(18)

Akrap, A.; Tran, M.; Ubaldini, A.; Teyssier, J.; Giannini, E.; van der Marel, D.; Lerch, P.; Homes, C. C. Phys. Rev. B 2012, 86 (23), 235207.

(19)

Sosso, G. C.; Caravati, S.; Bernasconi, M. J. Phys. Condens. Matter 2009, 21 (9), 95410.

(20)

Niesner, D.; Otto, S.; Fauster, T.; Chulkov, E. V; Eremeev, S. V; Tereshchenko, O. E.; Kokh, K. A. J. Electron Spectros. Relat. Phenomena 2014, 195, 258–262.

12 ACS Paragon Plus Environment

Page 12 of 28

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(21)

Jin, K.-H.; Yeom, H. W.; Jhi, S.-H. Phys. Rev. B 2016, 93 (7), 75308.

(22)

Wang, J.; DaSilva, A. M.; Chang, C.-Z.; He, K.; Jain, J. K.; Samarth, N.; Ma, X.-C.; Xue, Q.-K.; Chan, M. H. W. Phys. Rev. B 2011, 83 (24), 245438.

(23)

Sze, S. M.; K.Ng, K. Physics of semiconductor devices; Wiley : New Jersey, 2007.

(24)

Bathon, T.; Achilli, S.; Sessi, P.; Golyashov, V. A.; Kokh, K. A.; Tereshchenko, O. E.; Bode, M. Adv. Mater. 2016, 28 (11), 2183–2188.

(25)

Liu, M.; Zhang, J.; Chang, C.-Z.; Zhang, Z.; Feng, X.; Li, K.; He, K.; Wang, L.; Chen, X.; Dai, X.; Fang, Z.; Xue, Q.-K.; Ma, X.; Wang, Y. Phys. Rev. Lett. 2012, 108 (3), 36805.

(26)

Chiu, S.-P.; Lin, J.-J. Phys. Rev. B 2013, 87 (3), 35122.

(27)

Xia, F.; Wang, H.; Xiao, D.; Dubey, M.; Ramasubramaniam, A. Nat. Phot. 2014, 8 (12), 899–907.

(28)

Koumoto, K.; Mori, T. Thermoelectric Nanomaterials; Springer: Berlin, 2013.

(29)

Buscema, M.; Barkelid, M.; Zwiller, V.; van der Zant, H. S. J.; Steele, G. A.; Castellanos-Gomez, A. Nano Lett. 2013, 13 (2), 358–363.

(30)

Nolas, G. S.; Sharp, J.; Goldsmid, H. J. Thermoelectrics, Basic principles and new materials developements; Springer, 2001.

(31)

Li, Y.; Wu, K.; Shi, J.; Xie, X. Front. Phys. 2012, 7 (2), 165–174.

(32)

Zuev, Y. M.; Lee, J. S.; Galloy, C.; Park, H.; Kim, P. Nano Lett. 2010, 10 (8), 3037–3040.

13 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1

Figure 1: (a) Topographic AFM image of a lateral Bi2Te2Se/Sb2Te3 heterostructure, obtained by two consecutive vapor-solid growth steps. The central Bi2Te2Se nanoplatelet is surrounded by a significantly higher frame of Sb2Te3. (b) Cross-sectional profile along the white line in the AFM image of panel (a). (c) Overview TEM image of a Bi2Te2Se/Sb2Te3 nanoplatelet. (d) Corresponding EDX profiles for Sb, Bi, Te and Se, taken across the Sb2Te3/Bi2Te2Se interface along the white line in panel (c).

14 ACS Paragon Plus Environment

Page 14 of 28

Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2

Figure 2: (a) Raman map of a lateral BTS/ Sb2Te3 heterostructure, displaying the scattered intensity at ~71 cm-1 associated with the A1g1 mode of Sb2Te3. (b) Raman spectrum (λ = 514 nm) recorded at the position marked in panel (a), which lies within the Sb2Te3 region at the periphery. The blue line in the Lorentzian fit of the lowest energy peak. (c) Raman map of the same nanoplatelet as in panel (a), displaying the scattered intensity at ~ 65 cm-1 belonging to the Ag1 mode of Bi2Te2Se. (d) Raman spectrum collected from the center of Bi2Te2Se region of the heterostructure (open circle) .The blue line is a Lorentzian fit of the lowest energy peak. All of the above maps and spectra were acquired under ambient with λ = 514 nm.

15 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3

Figure 3: (a) Kelvin probe force microscopy image of the periphery of a Bi2Te2Se/Sb2Te3 nanoplatelet. The image was taken under ambient conditions. (b) Corresponding surface potential profile along the white line in panel (a). The starting point belongs to the right end of the line.

16 ACS Paragon Plus Environment

Page 16 of 28

Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 4

Figure 4: Current-voltage characteristics of a Bi2Te2Se/Sb2Te3 nanoplatelet at 1.4 and 290 K. The current was measured across the interface between the two components, using separate electrical contacts on the central Bi2Te2Se region and the Sb2Te3 frame at the periphery. The inset shows the resistance of the same heterostructure nanoplatelet as a function of the magnetic field recorded at T = 1.4 K. The B-field is oriented perpendicular to the sample plane.

17 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 28

Figure 5

Figure 5: Current-voltage curves of Bi2Te2Se/Sb2Te3 nanoplatelet, acquired in the dark (black curve) and under laser illumination (λ = 514 nm) of the interfacial region (red curve). The curve was measured using Sb2Te3 as the drain contact. All measurements were performed under ambient conditions. The inset is a plot of photocurrent vs. laser power (λ = 514 nm) for local illumination of the p-n junction.

18 ACS Paragon Plus Environment

Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 6

Figure 6: (a) Schematic depiction of a scanning photocurrent microscopy (SPCM) measurement, where a confocal laser spot is scanned across a Bi2Te2Se/Sb2Te3 device. (b) SPCM map of a Bi2Te2Se/Sb2Te3 nanoplatelet device with separate contacts on the two different components. The inset shows an optical image of the device, where the bright region below the electrode leading to the Bi2Te2Se is the SiOx layer used for avoiding contact between the electrode and the underlying Sb2Te3 region. The map was recorded under ambient conditions using λ = 514 nm.

19 ACS Paragon Plus Environment

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 7

Figure 7: DFT calculation-derived Seebeck coefficients of (a) Sb2Te3 and (b) BTS as a function of energy.

20 ACS Paragon Plus Environment

Page 20 of 28

BTS

b 40 Nano Letters Height (nm)

a Page 21 of 28

30 20

Atomic %

1000nm

1 2 10 Te 3 3 Sb 2 0 4 1000nm 0 0.2 0.4 0.6 0.8 1 1.2 Length (µm) c5 d 6 Sb-LA Bi-LA 60 7 Te-LA Se-KA 8 40 9 10 BTS 20 3 11 Te ACS Paragon Plus Environment 2 12 300 nm Sb 0 13 0 200 400 600 14 Length (nm)

b Nano Letters Counts (a.u.)

a

Page 22 SbofTe28 2

3

Counts (a.u.)

1 2 3 1 µm 50 100 150 200 4 Raman Shift (cm-1) c5 d 6 BTS 7 8 9 10 ACS Paragon Plus Environment 11 121 µm 50 100 150 200 13 Raman Shift (cm-1)

Page 23 of 28 Sb 650nm

1 2 3 4 5

Te 3

b Nano Letters -400

2

BTS

Δ Surface potential (mV)

a

-450

~130 meV -500

ACS Paragon Plus Environment -550 710nm

0.0

0.5 1.0 1.5 Length (µm)

2.0

Nano Letters 4

Page 24 of 28

T = 1.4 K T = 290 K

Rxx (Ω)

Current (µA)

1 2 2 3 4 0 496 T = 1.4 K 5 492 6 7 -2 488 8 484 9 -8 -4 0 4 B (T) 10 -4 Plus -1.0Paragon -0.5 0.0 Environment 0.5 1.0 11 -1.5 ACS Voltage (mV) 12 13

8

1.5

Page 25 of 28 10

Nano Letters

Laser OFF 5

Laser ON

0

-5

1.4

PC (mA)

Current (µA)

1 2 3 4 5 6 7 8 9 10 11 12 13

1 0.6 0.2

0

-10

20

40

60

Power (µw)

ACS -6 Paragon -4 4 -2Plus 0Environment 2 Voltage (mV)

6

80

a

Nano Letters

Page 26 of 28

1 2 A 3 Sb2Te3 Bi2Te2Se Sb2Te3 4 5 c b 6 0.51µA 7 8 9 10 2.0µm 11 -0.73 µA Te 3 12 Sb 2 13 BTS 14 15 16 A 17 2.0µm 18 19 ACS Paragon Plus Environment 20 21

aPage 27 of 28

Sb2Te3

0.4

0.4

BTS

E - EF (eV)

0.2

E - EF (eV)

1 20.2 3 40.0 5 6 7-0.2 8 9 -0.4 10 11 12 13

b Nano Letters

0.0

-0.2

-0.4

ACS Paragon Plus Environment -150

0

S ( µV/K )

150

-400 -200

0

200 400

S ( µV/K )

Nano Page Letters 28 of 28

1 Paragon Plus Environmen ACS 2 3