Graphene nanoribbons created using bottom-up methods can be arranged in atomic detail, and their physical property can be controlled precisely. In quantum technology, applications that require manipulating photons, spins, charges, or charges are necessary. However, doing this on the scale of one graphene nanoribbon can be experimentally tricky because of the difficulties of contacting individual nanoribbons, mainly synthesized on the surface. This paper describes the contact and electrical characterization of graphene synthesized on-surface nanoribbons in a multigate design using carbon nanotubes that have a single wall as electrodes. The method is based on the self-aligned nature of both nanotubes with dimensions as small as 1 nm and on the nanoribbons’ growth on each growth substrates. The resulting nanoribbon-nanotube devices are characterized by quantum transport phenomena – including the Coulomb blockade and excited states with vibrational origin and the Franck-Condon blockade. These signal the contact between the individual graphene nanoribbons.
Main
Bottom-up synthesized graphene nanoribbons (GNRs) are a tunable class of quantum material with a wide range of electronic, magnetic, and optical properties, including variable bandgaps, single-photon emission, and spin-polarized/topologically protected states1,2,3,4,5. These materials provide greater chemical flexibility than materials made by top-down methods six that control the size and shape of the edges is limited and may result in additional localized states caused by disorder at the edges. Using the materials to create quantum devices requires controlling their chemical structure as well, and they must be integrated into structures 7.8. The integration and contact of a single GNR with atomic precision can make semiconductor quantum dots (QDs) that capture individual charges and their spins. They can be used to make spin or charge qubits and single-photon emitters.
Contacting GNRs individually, particularly on-surface synthesized ones–is a problematic job 5-7,8. Bottom-up synthesized GNRs were previously approached using various methods (Fig. 1a,b), with the electrode material either a noble metal (gold, platinum or palladium)9,10,11,12,13,14,15,16,17,18 or graphene19,20,21,22,23,24,25. Electrodes can be made before or after GNR transfer and are referred to as ‘GNR last’ and ‘GNR first approaches in the respective. The GNR-first approach is preferred for ultrashort channel lengths when metallic electrodes are used9,10,11,12,13,14,15,16,17,18. The electrodes are made by using electron beam technology. (EBL) techniques that may cause damage and contamination to the GNRs when made (Fig. 1b, left). Graphene is a desirable alternative since it is flat atomically, making it the ideal material for the GNR-last method. Graphene electrodes can be defined with the EBL-defined nanogaps 19-20 (Fig. 1b middle) or fabricated by an electrical breakdown (EB) method, which produces ultra-narrow nanogaps within the range of 1-5 nanometers (refs. 22,23,24,25) (Fig. 1b, right).
Fig. 1 Scaling of size in the GNR bottom-up transistors with various geometries.
A, Comparison of the physical size of transistors from GNRs with different contact strategies: metal electrodes9,10,11,12,13,14,15,16,17,18 (orange), EBL-defined graphene electrodes19,20,21 (blue), EB-formed graphene electrodes22,23,24,25 (green) and EBL-defined SWNT electrode (red; this work). The squares represent GNRs with surface-polymerization in an ultrahigh vacuum. The triangles represent GNRs that are solution-polymerized, and the circles represent CVD-synthesized GNRs. B Schematic of transistors typical of bottom-up GNR transistors using metallic electrodes (left) and graphene electrodes that are EBL-defined (middle) and graphene electrodes that have been shaped by EB (right). C is ultimately used to connect bottom-up GNRs. The schematic of the ultrahigh-vacuum (UHV)-synthesized GNR array parallel to the Au(788) terraces (top left). Diagram of Parallel SWNT electrode array on a SiO 2. surface (top left). Schematic of a single GNR-based transistor using SWNTs as the final scaled electrodes (bottom). Only the GNRs nearest to nanogaps are shown to make it more transparent.
Full size image
However, for both metal and graphene electrodes, it is still challenging to get in contact with the individual GNR due to their small dimensions and their lateral separation (Fig. 1a), typically between 1-2 nm is ess than the capabilities of current EBL 26.27 (Fig. 1a; grey dashed lines). The distance between the inter-ribbons GNRs may be increased. However, it is typically achieved by reducing the number of precursor molecules present on the growth substrate. This results in smaller GNRs 28.
Individual GNRs were previously contacted with a graphene-based breakdown gap. However, this technique needs better-defined electrode geometries. It also only works for highly tiny GNRs comparable to electrode separation (around 5 nanometers). The longer GNRs enable the creation of superlattices in which localized states, or spins, are arranged periodically on the GNRs. This makes creating spin chains 29 or topologically protected states four possible. In these GNRs, it could be beneficial to place the functional portion between the electrodes rather than above them. However, as GNR length increases, the chance of electrodes joining multiple GNRs will also increase. Furthermore, the absence of precise control over the nanogap’s location hinders the creation of devices with multiple gates essential for controlling multi-QD systems.
At present, long GNRs have been only used to contact graphene or other metal electrodes, where bridging several GNRs in series and parallel is probable. This results in the formation of non-closing, irregular Coulomb diamonds 30 and makes using the electrical structure of one GNR for use in device applications extremely difficult. Thus, different methods of contacting GNRs with long lengths, like the one-dimensional electrode 31, must be considered.
In this article, we discuss the contacting of individually synthesized on-surface long GNRs in a multiple-gate transistor with single-walled carbon Nanotube (SWNT) electrodes (Fig. 1c). Our method is based on the self-aligned character of both SWNTs, with dimensions as small as 1 millimeter, and grow over their growth substrates. The structure of the SWNT-GNR-SWNT devices is confirmed using a spectrum of molecular levels that are performed at cryogenic temperatures, which highlights several aspects that define the flow through a distinct GNR, including Coulomb blockade, the presence of vibrational states in the single electron tunneling (SET) regime, and the blockade of Franck-Condon. Multiple gates allow the conductivity of GNRs and SWNT electrodes to be adjusted and also for the cause of the various states seen in spectroscopic studies to be discovered. The capability to connect long GNRs precisely within the multigate design could facilitate the control of dual- or multiple-QD systems shortly.
Device design
The devices that have been studied (Fig. 2.) comprise two SWNT electrodes that are separated by 15-25 nanometers. Below the nanogap is a 100-nm wide Finger gate made of Cr/Pt (FG), finely patterned along with the two side gates (SG1 and SG2). Multiple gates are necessary to regulate the densities of states (DOS) in the lead SWNT. Due to the quantum restriction of charge carriers because of the single-dimensional nature of the SWNTs, Sharp high-frequency Van Hove singularities occur at the beginning of every sub-band 32 and 33. Furthermore, SWNTs can be found in metal SWNTs (M-SWNTs) and semiconducting SWNTs (S-SWNTs). While M-SWNTs have a uniform and non-zero DOS near Fermi energy, their semiconducting counterparts show an extensive bandgap. Figure 2a,b shows the band diagrams of SWNT-GNR-SWNT junctions, with the distinct energetic levels of GNR along with their Van Hove singularities in the DOS of the S-SWNT and M-SWNT leads as well as the Van Hove singularities in the DOS of the S-SWNT leads. They are separate from one another via an nm-thick Al 2O 3 layer. A layer of GNRs is then placed onto the substrate of the device. Figure 2c illustrates the schematic for the system. A complete description of the material and fabrication process can be found in the Methods section and Supplementary Part 1.