Charge Density Waves in TaSe2
October 25, 2018 – Researchers from Ulm, Turku, Dresden and Aalto have shown that a single, freestanding layer of 1T-TaSe2 on graphene keeps its charge density wave at room temperature. AC-HRTEM at 80 kV on an FEI Titan in Ulm, with electron diffraction, revealed the stars of David formed by 13 tantalum atoms, a √13 × √13 superstructure turned by 13.9°, as in the bulk crystal but less pronounced. Image simulations from computed atom positions validate the result, and low-voltage STEM at Hokkaido University and AIST in Tsukuba had seen the same order in thin layers of TaS2.
In bulk 1T-TaSe2 the electrons do not spread evenly over the crystal. Below about 470 K they gather into a regular pattern, a charge density wave, and the tantalum atoms follow it: groups of 13 atoms pull together into small stars. Whether this order survives in a single layer, only three atoms thick, had been predicted but never observed. Researchers from Ulm University, Åbo Akademi in Turku, the Helmholtz-Zentrum Dresden-Rossendorf and Aalto University have now shown it directly, at room temperature, in a freestanding single layer of 1T-TaSe2 on graphene. Aberration-corrected high-resolution transmission electron microscopy (AC-HRTEM) at 80 kV on an FEI Titan 80-300 with a Cs corrector imaged the layer, and electron diffraction found the satellite reflections of the superstructure. Image simulations from computed atom positions validate the images: the stars form a √13 × √13 superstructure, rotated by 13.9° against the lattice, exactly as in the bulk crystal, but less pronounced. Low-voltage STEM at Hokkaido University and AIST in Tsukuba had seen the same superstructure in thin layers of the sister compound 1T-TaS2, and STEM at 80 keV in Singapore later confirmed it in thin 1T-TaSe2 layers.
Electrons in step with the lattice
A charge density wave is a periodic modulation of the electron density, found mostly in metals with layered or chain-like structures[1]. It opens a gap in the electronic spectrum and can turn a metal into an insulator. Through the coupling between electrons and lattice vibrations it is always accompanied by a periodic lattice distortion, a regular shift of the atoms away from their ideal positions. The distortion is what an electron microscope can see. Layered transition metal dichalcogenides such as TaSe2, TaS2 and NbSe2 show particularly strong charge density waves[1],[2]. Thinned down to a few layers, they are expected to change their behaviour: single-layer NbSe2 orders at a higher temperature than the bulk[3], and single-layer 1T-TaS2 forms a triclinic stripe pattern[4].
Bulk 1T-TaSe2 becomes commensurate below 473 K: its distortion then repeats exactly every √13 lattice constants. The bulk crystal is metallic, a single layer is predicted to be an insulator[5],[6], and thin films order at lower temperatures as they get thinner[7]. Whether a freestanding single layer still holds the commensurate state at room temperature had not been tested.
Stars of David
In the commensurate state, 12 of every 13 tantalum atoms move towards the 13th in the middle, by up to about 33 pm, and form a star of David (Fig. 1). The selenium atoms shift mainly out of the plane[2]. The stars sit on a new hexagonal lattice whose cell is √13 times larger than the cell of the undistorted layer and turned by 13.9° against it. In the incommensurate state the angle lies between 0 and 13.9°, and domains form.
One layer, protected by graphene
With a threshold of about 190–200 kV for knocking out selenium atoms[8], 80 kV causes no knock-on damage in TaSe2. Single layers of transition metal dichalcogenides suffer from other processes under the beam, however, from electronic excitations and from etching by residual gas[9],[10] (see Damage Mechanisms in MoSe2). Graphene underneath, or on both sides, slows this damage considerably[10]–[12] (see MoS2 Encapsulated in Graphene). The team therefore placed single-layer flakes of 1T-TaSe2 on a graphene layer over the holes of a gold-coated TEM grid, as described in the box below. The graphene also carries the flakes, which are a few micrometres wide. Heating of so thin a sample by the beam at 80 kV can be neglected[13] (see No Universal Virtual Temperature).
A faint pattern made visible
At Scherzer defocus, the heavy tantalum atoms (180.95 u) show more contrast than the selenium atoms (78.96 u), and the image of the heterostructure is dominated by the TaSe2 layer (Fig. 2a); removing the graphene lattice by filtering changes nothing visible. The superstructure is hidden in the raw image. Blurring it with a Gaussian filter, which suppresses only the fine detail of the atomic lattice, brings out a clear contrast modulation (Fig. 2b).
To prove that this modulation comes from the stars of David, the team computed the relaxed atom positions of the distorted single layer with density functional theory and simulated its image for the conditions of the experiment. The simulated image shows the stars clearly (Fig. 2c). Blurred in the same way, it gives quantitatively the same contrast modulation as the blurred experimental image. In the raw images, however, the stars stand out less in the experiment than in the simulation.
Satellites at 13.9°
The superstructure also shows in reciprocal space. The Fourier transform of the AC-HRTEM image contains, besides the main TaSe2 reflections and those of graphene, weak satellite spots around each main reflection (Fig. 3a). They are turned by 13.9° against the 100 reflections, which proves that the superstructure is commensurate with the underlying lattice. A selected-area electron diffraction pattern from a much larger area, taken with a 10 µm aperture, shows the same satellites (Fig. 3b); the graphene reflections are too weak to appear there. A simulated kinematic diffraction pattern of the distorted layer puts the main reflections and the satellites at the same positions (Fig. 3c). All three agree, as the box below sets out.
Calculations of the layer on graphene showed that graphene donates electrons to the TaSe2 sheet, but this transfer leaves the superstructure unchanged and, if anything, stabilizes it.
Earlier, a group from Hokkaido University and AIST had imaged freestanding thin layers of the sister compound 1T-TaS2 by low-voltage scanning transmission electron microscopy (STEM)[4]. The box below compares the two studies.
Challenges named by the authors
The superstructure is weak: in the raw AC-HRTEM image it is hardly visible and needs filtering, and the measured contrast modulation is smaller than in the simulation. Even a layer protected by graphene degrades under the beam, so dose and imaging time are limited. The sample is a heterostructure of TaSe2 and graphene, not an isolated layer; that the graphene does not alter the order rests on the calculations.
Outlook
“To sum up, we have unambiguously proven that the CCDW/PLD is stable in a monolayer of 1T-TaSe2 at room temperature,” the authors write. AC-HRTEM at 80 kV, combined with electron diffraction and image simulation, thus shows charge-density-wave order directly in a single layer of a two-dimensional metal, at the scale of single atoms. This opens the way to study how such order changes with thickness, with the support and at defects, in materials that are considered for ultrathin electronic and phase-change devices.
Resource:
Börner, P. C., Kinyanjui, M. K., Björkman, T., Lehnert, T., Krasheninnikov, A. V., Kaiser, U. (2018).
Observation of charge density waves in free-standing 1T-TaSe2 monolayers by transmission electron microscopy.
Appl. Phys. Lett. 113(17), 173103.
https://doi.org/10.1063/1.5052722
Update – July 2, 2020
The news article of October 25, 2018 ends above. Later work that refers to its findings is listed below.
The superstructure in grown TaSe2 layers
A team from Nanyang Technological University in Singapore grew atomically thin 1T-TaSe2 crystals by chemical vapour deposition instead of cleaving them[14]. High-angle annular dark-field STEM on a probe-corrected JEOL ARM200F at 80 keV showed the superlattice fringes of the commensurate state in the grown layers at room temperature (Fig. 4a), and the Fourier transform showed the satellite spots, turned by 13.6° against the main reflections, with the √13 length ratio (Fig. 4b). Atomic-resolution images showed the 1T structure (Fig. 4c). In layers 3 nm thick, the commensurate order persisted up to 570 K, 97 K above the bulk value.
| Feature | Ulm, 2018 (this work) | Singapore, 2020 | Agreement |
|---|---|---|---|
| Microscope | FEI Titan 80-300, Cs-corrected AC-HRTEM, 80 kV | JEOL ARM200F, probe-corrected HAADF-STEM, 80 keV | – |
| Sample | cleaved single layer on graphene | layers grown by CVD | – |
| Superstructure at room temperature | yes (Fig. 2) | yes (Fig. 4a) | yes |
| Rotation of the satellites | 13.9° | 13.6° | yes |
| Length ratio | √13 | √13 | yes |
Table 5: The commensurate superstructure of 1T-TaSe2 in Ulm and in Singapore[14].
Download this news article as PDF – for reading on paper, taking notes or filing.
-
Wilson, J. A., Di Salvo, F. J., Mahajan, S. (1975). Charge-Density Waves and Superlattices in the Metallic Layered Transition Metal Dichalcogenides. Adv. Phys., 24, 117–201. https://doi.org/10.1080/00018737500101391
-
Rossnagel, K. (2011). On the Origin of Charge-Density Waves in Select Layered Transition-Metal Dichalcogenides. J. Phys.: Condens. Matter, 23, 213001. https://doi.org/10.1088/0953-8984/23/21/213001
-
Xi, X., Zhao, L., Wang, Z., Berger, H., Forró, L., Shan, J., Mak, K. F. (2015). Strongly Enhanced Charge-Density-Wave Order in Monolayer NbSe2. Nat. Nanotechnol., 10, 765–769. https://doi.org/10.1038/nnano.2015.143
-
Sakabe, D., Liu, Z., Suenaga, K., Nakatsugawa, K., Tanda, S. (2017). Direct Observation of Mono-Layer, Bi-Layer, and Tri-Layer Charge Density Waves in 1T-TaS2 by Transmission Electron Microscopy Without a Substrate. npj Quantum Mater., 2, 22. https://doi.org/10.1038/s41535-017-0025-8
-
Suzuki, M., Harima, H. (2004). Electronic band structure of 1T-TaSe2 in CDW phase. J. Magn. Magn. Mater., 272–276, E653–E654. https://doi.org/10.1016/j.jmmm.2003.12.513
-
Darancet, P., Millis, A. J., Marianetti, C. A. (2014). Three-Dimensional Metallic and Two-Dimensional Insulating Behavior in Octahedral Tantalum Dichalcogenides. Phys. Rev. B, 90, 045134. https://doi.org/10.1103/PhysRevB.90.045134
-
Samnakay, R., Wickramaratne, D., Pope, T. R., Lake, R. K., Salguero, T. T., Balandin, A. A. (2015). Zone-Folded Phonons and the Commensurate–Incommensurate Charge-Density-Wave Transition in 1T-TaSe2 Thin Films. Nano Lett., 15, 2965–2973. https://doi.org/10.1021/nl504811s
-
Komsa, H.-P., Kotakoski, J., Kurasch, S., Lehtinen, O., Kaiser, U., Krasheninnikov, A. V. (2012). Two-Dimensional Transition Metal Dichalcogenides Under Electron Irradiation: Defect Production and Doping. Phys. Rev. Lett., 109, 035503. https://doi.org/10.1103/PhysRevLett.109.035503
-
Egerton, R. F., Li, P., Malac, M. (2004). Radiation Damage in the TEM and SEM. Micron, 35, 399–409. https://doi.org/10.1016/j.micron.2004.02.003
-
Lehnert, T., Lehtinen, O., Algara-Siller, G., Kaiser, U. (2017). Electron Radiation Damage Mechanisms in 2D MoSe2. Appl. Phys. Lett., 110, 033106. https://doi.org/10.1063/1.4973809
-
Algara-Siller, G., Kurasch, S., Sedighi, M., Lehtinen, O., Kaiser, U. (2013). The Pristine Atomic Structure of MoS2 Monolayer Protected From Electron Radiation Damage by Graphene. Appl. Phys. Lett., 103, 203107. https://doi.org/10.1063/1.4830036
-
Zan, R., Ramasse, Q. M., Jalil, R., Georgiou, T., Bangert, U., Novoselov, K. S. (2013). Control of Radiation Damage in MoS2 by Graphene Encapsulation. ACS Nano, 7, 10167–10174. https://doi.org/10.1021/nn4044035
-
Börner, P., Kaiser, U., Lehtinen, O. (2016). Evidence Against a Universal Electron-Beam-Induced Virtual Temperature in Graphene. Phys. Rev. B, 93, 134104. https://doi.org/10.1103/PhysRevB.93.134104
-
Wang, H., Chen, Y., Zhu, C., Wang, X., Zhang, H., Tsang, S. H., Li, H., Lin, J., Yu, T., Liu, Z., Teo, E. H. T. (2020). Synthesis of Atomically Thin 1T-TaSe2 With a Strongly Enhanced Charge-Density-Wave Order. Adv. Funct. Mater., 30, 2001903. https://doi.org/10.1002/adfm.202001903




