We demonstrate that the atom chain structure of Te allows it to be exfoliated as ultra-thin flakes and nanowires. Atomic force microscopy of exfoliated Te shows that thicknesses of 1–2 nm and widths below 100 nm can be −1 exfoliated with this method. The Raman modes of exfoliated Te match those of bulk Te, with a slight shift (4 cm ) due to a hardening of the A and E modes. Polarized Raman spectroscopy is used to determine the crystal orientation of exfoliated Te flakes. These experiments establish exfoliation as a route to achieve nanoscale trigonal Te while also demonstrating the potential for fabrication of single atom chains of Te. Keywords: Atom chain, Tellurium, Exfoliation, 1D layered material Background dimensional weakly bonded materials may be separated Dominated by carbon nanotubes and semiconductor to produce small diameter nanowires, as has been done nanowires, one dimensional (1D) materials have been with Li Mo Se [16, 17]. We argue that 1D weakly 2 6 6 extensively investigated for their extraordinary properties bonded materials present an opportunity to revisit 1D for electronics, photonics, and optoelectronics [1, 2]. materials, with a new possibility to achieve single atom Opportunities provided by 1D materials include transis- chains with atomic-scale diameters and an expectation tors scaled to the smallest possible dimensions [3, 4], ex- of new physical properties stemming from crystal struc- tremely sensitive chemical and biological sensors [5, 6], tures that are distinct from both carbon nanotubes and and unique electronic phenomena originating from the semiconductor nanowires. The anisotropic structure of similarity of optical fibers and ballistic electrons inside a 1D weakly bonded materials may allow single atom 1D wire [7, 8]. Progress with carbon nanotubes for most chains to be created by exfoliation, or possibly directly applications has been hampered by chirality randomness, grown by molecular beam epitaxy or chemical vapor and at the smallest diameters, semiconductor nanowire deposition. properties are degraded by surface dangling bonds. Con- Two exemplary 1D weakly bonded materials are tri- sequently, the focus of low-dimensional material re- gonal Se and Te, which have lattices consisting of spiral search has shifted primarily to two-dimensional (2D) chains oriented along the c-axis, each spiral having three layered materials, which combine atomic-scale thickness atoms per turn with adjacent chains arranged hexago- and high-performance physical properties by virtue of nally (Fig. 1). The chains are bound together to form a weak bonding in one direction [9–13]. single crystal through the van der Waals force  or The layered material concept may be generalized from perhaps more accurately as a weakly bonded solid . 2D materials, with weak bonds in one direction, to 1D In this letter, we report mechanical exfoliation of tri- materials, with weak bonds in two directions. Many 1D gonal Te single crystals to obtain nanoscale Te flakes weakly bonded solids are now known [14, 15]. One- and wires, which demonstrate the potential for fabrica- tion of single atom chains and a new platform for 1D electronics and photonics. * Correspondence: email@example.com While there are many 1D weakly bonded materials Department of Physics, University of Arkansas, Fayetteville, AR 72701, USA from which to choose, several properties of isolated Se Institute for Nanoscience and Engineering, University of Arkansas, Fayetteville, AR 72701, USA Full list of author information is available at the end of the article © The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Churchill et al. Nanoscale Research Letters (2017) 12:488 Page 2 of 6 Fig. 1 a Schematic of Te single crystal formed by single atom chains bonded by van der Waals force (top) and side view of Te chain structure (bottom). Note: 2 Å is the height of the triangular cross-section of a chain while the inter-chain distance is 3.4 Å. b Te single crystal used for exfoliation and Te semiconductor atom chains set them apart from self-assembly of single chains inside zeolite pores [30, other 1D atomic layered materials. For example: 31] and carbon nanotubes , the growth of 2D mono- layer trigonal Te on graphene , and solution-growth 1. They are predicted to have direct semiconducting of 2D Te [34, 35]. This earlier work demonstrates the band gaps of 1 and 2 eV for Te and Se, respectively, tendency of Te to form chains and nanowires that are with strongly thickness-dependent band gaps , relatively stable mechanically and chemically outside the creating new opportunities for tiny, wavelength- bulk Te crystal structure. Our objective is to use exfoli- tunable detectors and emitters. ation of solid Te as a route to obtain single atom chains. 2. The helical structure of Se and Te chains is expected to confer unique electrical, optical, and mechanical Methods properties, including novel spin-orbit coupling ef- To provide evidence for the potential for fabrication of fects boosted by heavy Se and Te atoms , nega- single atom chains, we investigated Te rather than Se tive compressibility and band gap narrowing under because of the availability of large, high-quality Te single pressure and strain , and extraordinary flexibility crystals . Prior to exfoliation, silicon substrates with greater than typical elastic polymers . 90 or 300 nm of thermal oxide were sonicated in acet- 3. Since they are composed of a single element, an one and isopropanol, then treated with oxygen plasma isolated Se or Te atom chain would have the to improve adhesion of Te. Trigonal Te single crystals smallest diameter of any known 1D material. The were mechanically exfoliated, without tape, directly on height of the triangular spiral cross-section is 2 Å, the silicon substrates  by manually sliding a freshly and the inter-chain distance is 3.4 Å . cleaved facet of Te on the substrate. We obtained the best results with the c-axis perpendicular to the direc- Experimental demonstration of the atom chain con- tion of motion. For Te exfoliation, we have found this cept originates with STM manipulation of individual method to be significantly superior to tape exfoliation, atoms on a substrate to achieve linear and planar arrays which likely reflects an important difference in the of coupled atoms [24, 25]. In addition to atom-by-atom bonding between 1D and 2D layered materials. Thin Te assembly on surfaces, step edges of substrates have been flakes were identified by contrast in an optical micro- decorated with atom chains , and self-assembled scope (Fig. 2a). Thin Te flakes show up with a progres- growth has been used to create large-area arrays of atom sion of colors in reflected light microscopy with the chains . However, depending on the approach, all thinnest crystals appearing as darker greens and blues these pioneering experiments do not allow 1D structures on this silicon substrate. to be created over large scales, choice of materials is lim- ited, or the structure is strongly bound to the substrate. Results and Discussion In principle atom chains derived from 1D weakly Tellurium was exfoliated in anisotropic linear bands with bonded materials could overcome these limitations. lengths up to 50 μm (Fig. 2a). Atomic force microscopy To date, the anisotropic structure of Se and Te has of some of these bands reveals heights in the 10–15 nm permitted growth of small diameter nanowires [28, 29], range (Fig. 2c), with ridges running along the length of Churchill et al. Nanoscale Research Letters (2017) 12:488 Page 3 of 6 Fig. 2 a Te exfoliated on a Si/SiO2 substrate, imaged immediately after exfoliation. b The same sample as in (a) after storage in air for 3 weeks. c AFM height image of the area inside the red square in (a). d Height profile along the white line shown in (c) the bands that are evident in both the height image and a degrades in air . This observation is consistent with height profile taken perpendicular to one of the bands as the observation that the timescale for degradation of Te shown in Fig. 2d. The modulated surface pattern and vari- nanowires in various solvents such as water is not indef- ation in wire width are evidence that the atom chains ran- inite but quite long, from hours to days . domly break away from the bulk crystal both laterally and We further characterize the exfoliated Te by Raman vertically, unlike 2D layered materials such as graphene spectroscopy. The Raman spectrum of bulk Te at room which exfoliate with mostly flat surfaces whether a tape or temperature is dominated by two sets of modes: an A −1 sliding technique is used. We were able to obtain wires of singlet at 120 cm and a pair of E doublets at 92 (104) 1–2 nm thickness using this sliding technique. and 141 (141) for transverse (longitudinal) phonons . For example, the atomic force images of the second The A and E modes of trigonal Te may be visualized as sample reveal a similar anisotropic structure of the exfo- symmetric and antisymmetric breathing modes of the liated material (Fig. 3a), as well as significantly narrower triangular cross-section of the Te chain . This Te nanowires with heights in the subnanometer range spectrum is reproduced in Fig. 4a for an excitation wave- (Fig. 3b–d) or at least corresponding to two to four length of 633 nm, with the lower E mode absent because chains for an inter-chain distance of 3.4 Å . These of the polarization direction of the incident light . ultrathin Te nanowires have lengths of 100–200 nm Peak positions agree with those reported in Ref.  to −1 (Fig. 3a). A height profile taken along the c-axis direction better than 1 cm . We note that excitation at 633 nm is (green line in Fig. 3b, green curve in Fig. 3d) indicates near a resonance with the dielectric function of bulk Te; that the surface roughness along the top of this 2–3-nm off-resonant excitation at 532 nm produces significantly tall nanowire is comparable to or less than that of the less Raman scattering intensity . SiO substrate. The Raman spectrum of an approximately 30-nm- Stability in ambient environment is a concern for any thick Te flake (red circle in Fig. 3a) shows the same two newly exfoliated material because surface reactions that peaks, shifted to slightly higher frequencies (Fig. 4a). are negligible in bulk materials can dominate the proper- The measured Raman peak of the silicon substrate at −1 ties of ultrathin exfoliated materials. An optical image of 520.9 cm (not shown) indicates that the spectrometer −1 the same Te sample in Fig. 2a is shown in Fig. 2b after is calibrated to better than 1 cm . We also note that storage for 3 weeks in air. Aside from differences in the exfoliated Te spectrum shown in Fig. 4a, which was color contrast due to camera settings, the aged sample measured in air several weeks after exfoliation, is not appears virtually the same as when it was freshly exfoli- consistent with the Raman spectra of either amorphous ated. In particular, we note a complete absence of the  or oxidized Te , which also establishes the en- blistering that occurs when 2D black phosphorus vironmental stability of ultrathin exfoliated Te. Despite a Churchill et al. Nanoscale Research Letters (2017) 12:488 Page 4 of 6 ab cd Fig. 3 a Optical micrograph of a second exfoliated Te sample. The red circle indicates the region used for Raman spectroscopy. b AFM height and (c) tapping mode amplitude images of the region indicated by the black square in (a). d Height profiles along the red, orange, and green lines in (b), perpendicular to the c-axis direction for red and orange, parallel for green. The orange and green profiles are offset vertically for clarity slight asymmetry in the Raman peaks for both bulk and One interpretation of this mode hardening is a flake- exfoliated Te, a pair of Lorentzians fits the spectra rea- substrate interaction, for example, if the Te is strained as sonably well (black curves in Fig. 4a). Peak parameters it is exfoliated on the SiO substrate. Interaction with extracted from the fits indicate a mode hardening for the substrate also generically hardens the radial breath- −1 the exfoliated flake relative to the bulk crystal of 4 cm ing modes of carbon nanotubes . Another possibility −1 for the A mode and 2 cm for the E mode. is that inter-chain interactions are reduced in ultrathin ab Fig. 4 a Raman scattering spectrum of bulk Te crystal (blue) and an exfoliated flake (red), under the same excitation conditions (633 nm, polarization parallel to c-axis). Spectra are normalized to the height of the dominant A1 peak. Fits (black curves) are a sum of two Lorentzians. b Polar plot of Raman intensity averaged over the spectral range in (a) as a function of linear excitation polarization angle relative to the c-axis (plot origin is zero intensity). The fit is a sine function plus a constant. The black arrow indicates the c-axis direction (see text) Churchill et al. Nanoscale Research Letters (2017) 12:488 Page 5 of 6 Te because a significant fraction of chains is missing one assumption for the unidirectional rubbing technique or more neighbors. A naïve expectation would be that used here. These observations demonstrate that polar- weaker inter-chain coupling would soften the A mode; ized Raman spectroscopy is sufficient to determine the however, applying pressure to Te crystals is known to re- crystal orientation of nanoscale exfoliated Te. This tech- duce the A frequency . Further, the A frequency of nique is useful in practice given that optical and atomic 1 1 isolated Te chains inside zeolite nanopores, where inter- force microscopy do not provide unambiguous informa- chain coupling is zero (or significantly less than for bulk, tion about crystal orientation. As the thickness and considering the 6.6 Å nanopore diameter), is much width of exfoliated Te approaches the single atom chain −1 higher than in bulk Te at 172 cm . The observa- limit, we expect a cross-over in the crystal direction as- tion that reduced inter-chain coupling hardens Te Ra- sociated with maximum Raman scattering because iso- man modes is explained by a competition between inter- lated Te chains inside nanopores have maximum Raman and intra-chain forces in Ref. . Our measurement of intensity for polarization parallel to the c-axis . a smaller shift for the E mode than the A mode (Fig. 4a) is also consistent with the pressure dependence re- Conclusions ported in Ref. , but substrate-induced strain may be We have introduced trigonal Te as a weakly bonded ma- expected to produce similar behavior. We are unable to terial capable of being exfoliated to produce ultrathin Te conclude within the scope of this work whether sub- single crystals. We demonstrate that the atom chain strate interaction or reduced inter-chain interactions are structure of Te allows it to be exfoliated as two- responsible for the spectral shifts we observe. dimensional flakes and one-dimensional nanowires. For the sample shown in Fig. 3, both optical and Atomic force microscopy of exfoliated Te shows that atomic force microscopy display elongated, horizontally thicknesses of 1–2 nm and wires of about 100 nm width aligned Te flakes, which suggests that the c-axis of the can be exfoliated with this method. The Raman modes Te crystal is horizontal in these images. However, the of exfoliated Te match those of bulk Te, with a slight AFM images (Fig. 3b, c) also show that a significant frac- −1 shift (4 cm ) due to a hardening of the A and E modes. tion of the exfoliated flakes, particularly the thinnest Polarized Raman spectroscopy is used to determine the ones, are tilted 45° away from horizontal. To confirm crystal orientation of exfoliated Te flakes. These experi- the crystal orientation of this sample, we use ments establish exfoliation as a route to achieve nano- polarization-resolved Raman spectroscopy. The scale trigonal Te while demonstrating the potential for polarization of the excitation beam was rotated with a fabrication of single atom chains of Te. Our current ef- half-wave plate, and the integrated Raman intensity from forts are focused on producing Te or Se single atom −1 85 to 170 cm is shown in Fig. 4b. The intensities were chains by molecular beam epitaxy or by improving normalized by the laser power under the microscope ob- exfoliation. jective measured at each polarization angle. The Raman intensity shows two maxima within one full rotation, lo- Acknowledgements cated at 45° and 225° with respect to the X and Y axes We acknowledge J. M. grant for graphical assistance. This work was supported by AFOSR award numbers FA9550-14-1-0205 (S.-Q.Y.) and FA9550- defined in the microscope images (Fig. 3). The intensity 16-1-0203 (H.C.). H. C. acknowledges support from the University of Arkansas varies approximately sinusoidal (black curve in Fig. 4b), Connor Faculty Fellowship. with an amplitude of +/−15% over a constant background. Availability of Data and Materials The corresponding author (HC) should be contacted regarding requests for Meanwhile, the optical absorption of bulk Te at data and materials. 633 nm is stronger for light polarized perpendicular to the c-axis than for parallel polarization . For Te Authors’ Contributions flakes with nearly bulk-like optical properties (Fig. 4a), HC, GS, and SQY conceived of the experiment and directed the research. IS grew the Te single crystals. HC and TH produced exfoliated Te samples. HC, we therefore expect Raman intensity to be higher for TH, XH, and JS and acquired the data. HC, TH, and GS analyzed the data. HC, light polarized perpendicular to the c-axis. Based on the GS, and SQY wrote the manuscript with input from all authors. All authors angle of the Raman maximum in Fig. 4b, we conclude read and approved the final manuscript. that the Te nanowires oriented at 45° in Fig. 3b, c are elongated parallel to the c-axis for that sample. Because Ethics Approval and Consent to Participate Not applicable. different Te flakes on the same substrate were used for Raman spectroscopy and AFM, an assumption of this Consent for Publication conclusion is that the crystal axes are the same for all Not applicable. exfoliated flakes shown in Fig. 3a. This assumption would not be appropriate for flakes prepared by the Competing Interests traditional tape exfoliation method, but it is a reasonable The authors declare that they have no competing interests. Churchill et al. Nanoscale Research Letters (2017) 12:488 Page 6 of 6 Publisher’sNote 27. Yeom HW et al (1999) Instability and charge density wave of metallic Springer Nature remains neutral with regard to jurisdictional claims in quantum chains on a silicon surface. Phys Rev Lett 82:4898 published maps and institutional affiliations. 28. Qian HS et al (2006) High-quality luminescent tellurium nanowires of several nanometers in diameter and high aspect ratio synthesized by a poly (vinyl Author details pyrrolidone)-assisted hydrothermal process. Langmuir 22:3830 Department of Physics, University of Arkansas, Fayetteville, AR 72701, USA. 29. Xi G et al (2006) Large-scale synthesis, growth mechanism, and Institute for Nanoscience and Engineering, University of Arkansas, photoluminescence of ultrathin Te nanowires. Cryst Growth Des 6:2567 Fayetteville, AR 72701, USA. Department of Electrical Engineering, University 30. Bogomolov VN (1978) Liquids in ultrathin channels (Filament and cluster of Arkansas, Fayetteville, AR 72701, USA. Department of Electrical and crystals). Sov Phys Usp 21:77 Computer Engineering, McGill University, Montreal, QC H3A 0G4, Canada. 31. Li IL et al (2005) Resonant Raman study of confined Se single helix and Se rings. Appl Phys Lett 87:071902 Received: 15 June 2017 Accepted: 27 July 2017 32. Medeiros PVC et al (2017) Extreme Te nanowires encapsulated within ultra- narrow single-walled carbon nanotubes. https://arxiv.org/abs/1701.04774 33. Huang X et al (2017) Epitaxial growth and band structure of Te film on graphene. https://arxiv.org/abs/1703.07062 34. Wang Y et al (2017) Large-area solution-grown 2D tellurene for air-stable, References high-performance field-effect transistors. https://arxiv.org/abs/1704.06202 1. Jorio A, Dresselhaus G, Dresseldhaus MS, editors (2008) Carbon nanotubes. 35. Du Y et al (2017) 1D van der Waals material tellurium: Raman spectroscopy Springer, New York under strain and magneto-transport. https://arxiv.org/abs/1704.07020 2. Zhang A, Zheng G, Lieber CM (2016) Nanowires. Springer, Switzerland 36. Shih I, Champness CH (1978) Czochralski growth of tellurium single crystals. 3. Tans SJ, Verschueren ARM, Dekker C (1998) Room-temperature transistor J Cryst Growth 44:492 based on a single carbon nanotube. Nature 393:49 37. Navarro-Moratalla E et al (2016) Enhanced superconductivity in atomically 4. Cui Y, Duan X, Hu J, Lieber CM (2000) Doping and electrical transport in thin TaS . Nat Comm 7:11043 silicon nanowires. J Phys Chem B 104:5213 38. Wood JD et al (2014) Effective passivation of exfoliated black phosphorus 5. Kong J et al (2000) Nanotube molecular wires as chemical sensors. Science transistors against ambient degradation. Nano Lett 14:6964 287:622 39. Lan W-J et al (2007) Dispersibility, stabilization, and chemical stability of 6. Cui Y, Wei Q, Park H, Lieber CM (2001) Nanowire nanosensors for highly ultrathin tellurium nanowires in acetone: morphology change, sensitive and selective detection of biological and chemical species. Science crystallization, and transformation into TeO in different solvents. Langmuir 293:1289 23:3409 7. Liang W et al (2001) Fabry - Perot interference in a nanotube electron 40. Pine AS, Dresselhaus G (1971) Raman spectra and lattice dynamics of waveguide. Nature 411:665 tellurium. Phys Rev B 4:356 8. Refael G, Heo J, Bockrath M (2007) Sagnac interference in carbon nanotube 41. Lucovsky G (1967) The structure of amorphous selenium from infrared loops. Phys Rev Lett 98:246803 measurements. In: Cooper WC, editor. The physics of selenium and 9. Novoselov KS et al (2004) Electric field effect in atomically thin carbon films. tellurium. Canada. P. 255–267 Science 306:666 42. Oyobuturi TS (1971) Physical properties of tellurium. J Soc Appl Phys 40:594 10. Novoselov KS et al (2005) Two-dimensional gas of massless Dirac fermions 43. Richter WJ (1972) Extraordinary phonon Raman scattering and resonance in graphene. Nature 438:197 enhancement in tellurium. Phys Chem Solids 33:2123 11. Zhang Y, Tan Y-W, Stormer HL, Kim P (2005) Experimental observation of 44. Brodsky MH et al (1972) The Raman spectrum of amorphous tellurium. the quantum Hall effect and Berry's phase in graphene. Nature 438:201 Phys Stat Sol B 52:609 12. Novoselov KS et al (2005) Two-dimensional atomic crystals. Proc Nat Acad 45. Mirgorodsky AP et al (2000) Dynamics and structure of TeO polymorphs: Sci USA 102:10451 model treatment of paratellurite and tellurite; Raman scattering evidence 13. Avouris P, Heinz TF, Low, T (2017) 2D materials: properties and devices. for new γ- and δ-phases. J Phys Chem Solids 61:501 Cambridge, New York 46. Araujo PT et al (2010) Resonance Raman spectroscopy of the radial 14. Cheon G et al (2017) Data mining for new two- and one-dimensional breathing modes in carbon nanotubes. Physica E 42:1251 weakly bonded solids and lattice-commensurate heterostructures. Nano 47. Richter W et al (1973) Hydrostatic pressure dependence of first-order Raman Lett 17:1915 frequencies in Se and Te. Phys Stat Sol B 56:223 15. Island JO et al (2017) Electronics and optoelectronics of quasi-one dimensional 48. Poborchii VV (1996) Polarized Raman and optical absorption spectra of the layered transition metal trichalcogenides. https://arxiv.org/abs/1702.01865 mordenite single crystals containing sulfur, selenium, or tellurium in the 16. Venkataraman L, Lieber CM (1999) Molybdenum selenide molecular wires as one-dimensional nanochannels. Chem Phys Lett 251:230 one-dimensional conductors. Phys Rev Lett 83:5334 49. Stuke J, Keller H (1964) Optical properties and band structure in the system 17. Venkataraman L, Kim P (2006) Electron transport in a multichannel one- Se-Te. Phys Status Solidi 7:189 dimensional conductor: molybdenum selenide nanowires. Phys Rev Lett 96:076601 18. Hippel ARvon (1948) Structure and conductivity in the VI group of the periodic system. J Chem Phys 16:372 19. Joannopoulos JD, Schlüter M, Cohen ML (1975) Electronic structure of trigonal and amorphous Se and Te. Phys Rev B 11:2186 20. Entin MV, Magarill LI (2002) Electrons in a twisted quantum wire. Phys Rev B 66:205308 21. Ren W, Ye J-T, Shi W, Tang Z-K, Chan CT, Sheng P (2009) Negative compressibility of selenium chains confined in the channels of AlPO -5 single crystals. New J Phys 11:103014 22. Flory PJ (1969) Statistical mechanics of chain molecules. Interscience, New York 23. Martin RM, Lucovsky G, Helliwell K (1976) Intermolecular bonding and lattice dynamics of Se and Te. Phys Rev B 13:1383 24. Eigler DM, Schweizer EK (1990) Positioning single atoms with a scanning tunnelling microscope. Nature 344:524 25. Yazdani A, Eigler DM, Lang ND (1996) Off-resonance conduction through atomic wires. Science 272:1921 26. Gambardella P et al (2002) Ferromagnetism in one-dimensional monatomic metal chains. Nature 416:301
Nanoscale Research Letters – Springer Journals
Published: Aug 10, 2017
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud