Symposium S.NM05—1D Carbon Electronics—From Synthesis to Applications

Materials purely made up of non-magnetic atoms can display magnetic properties. One such example are aromatic hydrocarbon structures. Following Lieb’s theorem, if the number of atoms in each carbon sublattice is different, the molecular structure will hold a net spin. However, even in the absence of net spin, polyradicals can be generated on graphitic structures subject to particular edge topologies. In addition to the edge topologies, also the structure’s size determines the presence or absence of such radical states, whose corresponding spin, if present, is furthermore predicted to display a specific relative alignment. These interesting predictions are hard to verify because of the difficulty in the synthesis of appropriate nanographytic structures with atomic precision. In this respect, on-surface synthesis under vacuum conditions has appeared as an extremely promising approach. Making use of it, we have synthesized a variety of nanosized aromatic hydrocarbon structures with well-determined shapes and sizes, including graphene nanoribbons with different edge morphologies, widths and lengths. For the particular case of 5 atoms wide armchair graphene nanoribbons we describe, based on scanning tunneling microscopy and spectroscopy, and combined with theoretical calculations, their electronic properties and the magnetism associated to its topologic end-states.

Graphene can spontaneously develop intrinsic paramagnetism. Crucial examples are the magnetization of zig-zag edges in graphene, or the emergence of paramagnetism in open shell graphene nanostructures. I will show that graphene nanoribbons (GNR), fabricated with atomic precision on a metal surface exhibit fingerprints of π-paramagnetism on a metal surface, which can be detected and spatially localized using low temperature scanning tunneling [1].

Single electron spins emerge localized at certain zigzag sites of the carbon backbone. Their presence could be detected and mapped by spatially resolving the zero-energy resonance due to the Kondo effect. We found that near-by spins are coupled into a singlet ground state and quantify their exchange interaction via singlet-triplet inelastic electron excitations. Theoretical simulations picture how electron correlations result in spin-polarized radical states with the experimentally observed spatial distributions. Extra hydrogen atoms bound to radical sites quench their magnetic moment and switch the spin of the nanostructure in half-integer amounts.

I will also review other methods for activating magnetic ground states in graphene. For example, spin states can be created on a ribbon simply by substitutional doping or by incorporating magnetic species into a ribbon using on-surface synthesis routes (see included image of a Fe porphyrin contacted to chiral nanoribbons). In this last case, we proved that the molecular spin survives in the ribbon by using spin-excitation inelastic spectroscopy [2]. By proper selecting the position of the halogen functionalization, we fabricated linear GNR-FeTPP-GNR structures and performed electronic transport measurements [3], detecting spin-excitation fingerprints in transport mode.

J. Li, S. Sanz, M. Corso, D.J. Choi, D. Peña, T. Frederiksen, J.I. Pascual, “Single Spin Localization and Manipulation in Graphene Open-Shell Nanostructures”, Nature Communications 10, 200 (2019).
J. Li, N. Merino-Díez, E. Carbonell-Sanromà, M. Vilas-Varela, D. G. de Oteyza, D. Peña, M. Corso, and J.I. Pascual, “Survival of spin state in magnetic porphyrins contacted by graphene nanoribbons”
Science Advances 4, eaaq0582 (2018)
J. Li, N. Friedrich, N. Merino-Díez, D. G. de Oteyza, D. Peña, D. Jacob, and J.I. Pascual, “Electrically Addressing the Spin of a Magnetic Porphyrin through Covalently Connected Graphene Electrodes”, Nano Letters 19, 3288 (2019).

A long-standing challenge in nanoscience has been to fabricate devices of only a few nanometers with single-atom accuracy. The necessary structural control can be provided by bottom-up fabricated molecular graphene nanoribbons with atomically precise edges. Only such control of the graphene nanoribbons should allow derived field-effect transistors that exhibit magnetic and topological effects.[1,2] Here we report the observation of magnetic, topological, and electronic effects on the quantum transport of hitherto unexplored molecular graphene nanoribbons. This includes magnetic shifting of molecular energy levels, topology and time-reversal symmetry breaking in a universality-class transition, negative differential conductance, and quantum transport effects up to near-room temperature.

[1] Narita, A., Wang, X.-Y., Feng, X. & Müllen, K. New advances in nanographene chemistry. Chem. Soc. Rev. 44, 6616-6643 (2015).
[2] Ritter, K.A., & Lyding, J.W., The influence of edge structure on the electronic properties of graphene quantum dots and nanoribbons. Nature Mater. 8, 235–242 (2009).

The remarkable flexibility, stable chemical structure, and extraordinary thermal, electrical, and optical properties of carbon nanotubes (CNTs) are promising for a variety of applications in flexible and/or high-temperature electronics, optoelectronics, and thermoelectrics, including wearables, refractory photonics, and waste heat harvesting. However, the long-standing problem in the preparation of CNT ensembles is to maintain the extraordinary properties of individual CNTs on a macroscopic scale; the polydispersity and randomness remain two main challenges. Here, we will discuss three ways of creating wafer-scale aligned CNTs: direct growth of aligned CNTs by chemical vapor deposition, production of ultrahigh-conductivity CNT fibers through solution spinning and coating, and spontaneous formation of wafer-scale aligned CNT films via controlled vacuum filtration. We will then describe flexible and high-temperature applications of these materials, such as flexible CNT broadband detectors, spectrally selective thermal emitters, and thermoelectric devices.

Plasmon resonance in carbon nanotubes is an exciting resource for confining light to the nanometer scale [1]. In this talk, I will discuss how dense films of carbon nanotubes have emergent optical properties related to these resonances. We assemble our films with a simple vacuum filtration technique and find that optimal assembly leads to the nanotubes crystallizing into monolithic, two-dimensionally ordered films. Using Mueller-matrix ellipsometry, we then characterize the films’ optical properties and find a broadband, electrostatically tunable hyperbolic region in the mid-infrared [2]. Another fascinating property of dense nanotube films is that excitons in the nanotubes can couple to the nanotubes’ own plasmon resonances – that is, cavity quantum electrodynamics (cQED) without an external cavity [3]. In fact, this coupling strength can reach the ultrastrong light-matter interaction regime. Technologically, crystallized nanotube films could provide a materials-driven pathway from fundamental concepts in cQED to advanced telecommunications hardware, including tunable thermal emitters, super-resolution imaging systems, and active, nonlinear optical devices like modulators and nanolasers.

[1] Falk et al, “Coherent plasmon and phonon-plasmon resonances in carbon nanotubes,” Phys. Rev. Lett. 118, 257401 (2017).
[2] Roberts et al, “Tunable Hyperbolic Metamaterials Based on Self-Assembled Carbon Nanotubes,” Nano Lett. 19, 3131 (2019).
[3] Ho et al, “Intrinsically ultrastrong plasmon–exciton interactions in crystallized films of carbon nanotubes,” Proc. Nat. Acad. Sci. 115, 12662 (2018).

In order to achieve a sufficiently high on-current, carbon nanotube based electronic devices must be comprised of highly aligned, densely packed arrays of nanotubes. One technique that has demonstrated immense promise for aligning nanotubes at optimal packing densities across technologically relevant substrates is floating evaporative self-assembly (FESA). The FESA process is a fluid-based method to align carbon nanotubes from solution post-synthesis and electronic sorting.^1–3 FESA aligned carbon nanotube films exhibit exceptional local alignment and packing density, which has enabled the fabrication of field-effect transistors (FETs) with current densities exceeding that of GaAs and Si.⁴ However, globally, FESA carbon nanotube films are highly non-uniform, as the films are comprised of alternating stripes of aligned and unaligned nanotubes. In the work presented here, we demonstrate a new technique in order to overcome the shortcomings of the FESA method and produce uniform films of aligned carbon nanotubes at desired packing densities, in a scalable technique.

In our technique, nanotubes are continuously deposited onto planar surfaces by sweeping a nanotube ink/water interface over a target substrate. The ink/water interface serves to collect and confine the carbon nanotubes, which aids in their alignment across the substrate during deposition. By translating the substrate vertically through the flowing nanotube ink/water interface, we fabricate uniform films of highly aligned (within ± 5.7°) carbon nanotubes on the wafer-scale. Carbon nanotube ink concentration is varied from 0.1 to 100 µg mL^-1 to control the film packing density from 0.3 to 50 µm^-1, near those required for electronics. Using optimized conditions, we align carbon nanotubes 60 mm laterally across a 100 mm wide wafer, with no limit on the deposition in the vertical/lift direction. The high uniformity of nanotube alignment leads to excellent FET characteristics—high on-state current density averaging 520 µA µm^-1 at -0.6 V and FET to FET variation of only 19%. In contrast, FETs fabricated across FESA films demonstrate variation of 52%, due to the stripes of aligned and unaligned nanotubes.

1. Jinkins, K., R. et al. Nanotube Alignment Mechanism in Floating Evaporative Self-Assembly. Langmuir 33, 13407–13414 (2017).
2. Joo, Y., Brady, G. J., Arnold, M. S. & Gopalan, P. Dose-Controlled, Floating Evaporative Self-assembly and Alignment of Semiconducting Carbon Nanotubes from Organic Solvents. Langmuir 30, 3460–3466 (2014).
3. Brady, G. J. et al. Polyfluorene-Sorted, Carbon Nanotube Array Field-Effect Transistors with Increased Current Density and High On/Off Ratio. ACS Nano 8, 11614–11621 (2014).
4. Brady, G. J. et al. Quasi-ballistic carbon nanotube array transistors with current density exceeding Si and GaAs. Sci. Adv. 2, (2016).

Conjugated polymers have been extensively utilized to isolate semiconducting single-walled carbon nanotubes (s-CNTs) to greater than 99.9% purity in organic solvents creating a high-purity s-CNT “ink”. Solution deposition of s-CNTs from these inks onto target substrates has been widely studied in the literature for fabricating field-effect transistors (FETs). However, s-CNT-based FETs have so far underperformed compared to conventional silicon- and gallium arsenide-based FETs. One limiting factor is the difficulty in fabricating ideal s-CNT arrays using scalable methods. The ideal morphology of s-CNTs in array FET consists of perfectly aligned s-CNTs (i.e. 0° orientation difference between all s-CNTs in the array) with a pitch of 5-10 nm between nanotubes. Even with the plethora of alignment and spatially localized s-CNT deposition methods designed to control s-CNT deposition, fundamental understanding of the driving forces for s-CNT deposition is still lacking. The individual roles of the polymer wrapper, solvent, target substrate composition, and the s-CNT itself are not completely understood because it is difficult to decouple the deposition parameters.

In order to understand and control s-CNT deposition onto substrates, we present our studies to understand the multitude of factors driving polymer wrapped s-CNT deposition onto substrates. Chemical modification of the substrate with self-assembled monolayers (SAMs) presents an ideal platform to explore these parameters. We use poly[(9,9-dioctylfluorenyl- 2,7-diyl)-alt-co-(6,6′-{2,2′-bipyridine})] (PFO-BPy) wrapped s-CNTs and study their interactions with chemically modified substrates. Experimentally, we characterize number density of s-CNT deposition on different chemically modified surfaces from toluene and chloroform. To gain a molecular level understanding we use atomistic molecular dynamics simulations to examine the driving forces for s-CNT adsorption. Multiple parameters are considered including substrate surface energy, substrate chemistry, solvent type, and the interactions that arise from the polymer wrapper PFO-BPy. We find that the global free-energy minimum associated with favorable s-CNT adsorption occurs for a configuration in which the minimum of the solvent density around the s-CNT coincides with the minimum of the solvent density above a SAM-grafted surface. This configuration indicates that solvent structure near a SAM-grafted surface determines the adsorption free-energy landscape driving s-CNT deposition.

Knowledge gained from these simulations and experiments guides substrate surface pattern designs for spatially localized s-CNT deposition. We use scalable s-CNT alignment techniques in combination with surface patterned substrates to fabricate aligned and spatially localized s-CNT arrays for integrated device structures. These s-CNTs arrays are characterized by s-CNT alignment degree for use in s-CNT-based FETs.

Three-dimensional (3D) assemblies of super porous carbon nanotube (CNT) structure is the recent focus of nano-carbon research for the unique properties offered by the interconnected conductive network in multifunctional applications like flexible sensors, energy storage, ultra-low-density electromagnetic interference (EMI) shielding, and selective organic compound absorption. In a 3D macroscopic assembly not only the purity, size and diameter of the CNTs are major variables also the orientation, alignment, and stability of the CNT-CNT junctions play a crucial role in the final structural performance. We developed a 3D free-standing CNT sponge from high purity, similar-sized (length 1mm, diameter ~ 23 ±4 nm) spinnable CNTs with complete control over the CNT network alignment technique. And the CNT-CNT junctions within the 3D sponge were secured by conformal coating with atomic layer by layer infiltration of amorphous pyrolytic carbon (PyC) inside the CVD furnace at 800 °C. The density (as low as 6 mg cm^-3) and porosity (as high as 99.8 %) of the CNT/PyC sponge can be tuned by varying the PyC coating time and regulated consolidation of aligned CNTs. The CNT sponges are super compressible with quick recovery of the structure even after as high as 80 % compressive strain and showed negative poison ratio auxetic compression behavior. Further, nanomanipulation of the superhydrophobic surface of the CNT sponge is done by controlled pulse corona treatment, which allows initiating certain hydrophilic polymer grafting on the surface. The surface modification changes the water contact angle from 153 ° to 0° and enables us to directly incorporate different polymers (polyvinyl chloride, polycarbonate, polydimethylsiloxane) into CNT sponge to fabricate CNT/PyC-polymer sponges without interrupting the stable alignment of the CNTs. The conformal polymer wrapping around the CNT/PyC can be easily controlled from a very few nanometres to hundreds of nanometres. All the CNT/PyC and CNT/PyC-polymer sponges were fully characterized using SEM, TEM, FTIR, Raman, and TGA. The polymer coating on the CNT/PyC sponges not only help to prevent the nano-release into the environment during dynamic use also showed very promising sensing and EMI shielding capabilities. This tunable nanomanipulation of the CNT/PyC sponge surface chemistry open-up a platform for easier multi-material hybrid structure construction in bulk without disrupting the pre-formed super aligned nano conductive network.

A unique feature of single-wall carbon nanotube (SWCNT) material is its structure diversity: within the same rolled-up graphene sheet structure framework one finds SWCNTs of different electronic bandgaps: metals with zero bandgap, quasi-metals with meV bandgap and semiconductors with ~ 1 eV bandgap. However, making use of SWCNT structure diversity requires a method to obtain SWCNTs of defined structures. In this contribution, we present our current state of DNA-based sorting of SWCNTs. In particular, we show our ability to put any DNA sequence on any SWCNT species, enabling DNA-based assembly of high-density SWCNT devices. In the end, we discuss the design of a device we call Photon Perceptron Array (PPA). PPA is an array of micron-sized spectrometer units that can measure the spectrum of incoming photon in the wavelength range of 200 nm to 2000 nm (and possibly going all the way to THz), with a spatial resolution comparable to the size of a single spectrometer unit. Each spectrometer unit is composed of a set of distinct photodetectors made of SWCNTs of defined atomic structures, placed in parallel with tube-to-tube separations controlled at nanometer precision by DNA origami technology. A high-density array of identical spectrometers can be constructed by placing DNA origami/carbon nanotube complexes in photolithographically-defined locations. PPA may be used in spectral imaging in many fields of science and technology.

Supramolecular polymer sorting is a promising approach to separating single-walled carbon nanotubes (CNTs) by electronic type. Unlike conjugated polymers, they can be easily removed from the CNTs after sorting by breaking the supramolecular bonds, allowing for isolation of electronically pristine CNTs as well as facile recycling of the sorting polymer. However, little is understood about how supramolecular polymer properties affect CNT sorting. Herein we use chain stoppers to engineer the conformation of a supramolecular sorting polymer. Through NMR and UV-vis spectroscopy, SAXS, and thermodynamic modeling, we show that this supramolecular polymer exhibits ring-chain equilibrium, and that this equilibrium can be skewed towards chains by the addition of chain stoppers. We also demonstrate that by controlling the stopper-monomer ratio, we can double the sorting yield without compromising the purity or properties of sorted CNTs.

In the past decade tuning the chirality distribution of single-walled carbon nanotube (SWCNT) networks has led to rapid advances of their optoelectronic, electronic and thermoelectric applications. Nevertheless, the full potential of this design freedom in those networks is yet to be explored, both in terms of the processes governing charge transport as well as their performance in applications. Here we report on charge carrier density and temperature dependent field-effect mobility and on-chip Seebeck coefficient measurements of polymer-sorted small diameter (6,5) SWCNT (0.76 nm) networks, 1,2,4,5-tetrakis(tetramethylguanidino)benzene (ttmgb) treated (6,5) networks as well as large diameter plasma torch SWCNT (1.17-1.55 nm) networks with different network densities to provide insights into their charge transport mechanisms and potential for thermoelectric applications. The Seebeck coefficient offers insights into transport energetics, potential contributions from electron-electron and electron-phonon interactions and the relative contributions of inter- versus intra-CNT transport.¹ Contrary to previous studies we conclude that a pure percolation theory and variable-range hopping description of transport is insufficient to explain the charge density and temperature dependence of both transport coefficients. Instead intra-nanotube transport that depends on the chirality of the CNTs composing the network needs to be considered as well. All untreated networks show remarkable ambipolar behaviour in both the mobility and the Seebeck coefficient, emphasizing the purity of our samples suitable to study their intrinsic transport mechanisms. Interestingly, in the high network density regime, the network density influences neither electric nor thermoelectric transport strongly, indicating both are dominated by the tubes rather than the tube-tube junctions. Furthermore, we emphasize the importance of density of states (DoS) engineering via chirality selection for thermoelectric generators. Despite their generally lower mobilities (6,5) SWCNTs show comparable power factors to large diameter, small bandgap networks with a broad chirality distribution. These findings provide design guidelines towards narrow DoS distribution, large diameter SWCNT networks for both electronic and thermoelectric applications.²
[1] Statz, M. et al. Commun. Phys. 1, 16 (2018).
[2] Statz, M. et al., in prep. (2019).

Flexible and stretchable electronics are now widely studied for a next class of electric devices replaced from the silicon-based conventional inflexible devices. In particular, multi-functional sensor sheets can contribute to a variety of applications for Internet of Things (IoT) including wearable healthcare. For this flexible device concepts, one of the potential materials is nanocarbons including carbon nanotube and graphene due to their unique electrical and mechanical properties. Furthermore, material handling to pattern these on a macroscale flexible film is readily conducted which is an important factor for practical flexible electronics. By utilizing the advantages of nanocarbon, in this study, macroscale, multi-functional flexible sensor sheets fabricated by mainly printing methods of nanocarbon composite materials are introduced for healthcare and IoT applications. Firstly, fundamental characteristics of strain, tactile pressure, temperature, and humidity sensors are discussed, followed by the device demonstrations. Healthcare, wind flow mapping, and soft-robotic applications are mainly discussed in this talk. Finally, to move forward to building the platform of flexible electronics, stability and reliability of sensors in terms of sensitivity and long-time measurements will be briefly introduced.
Although a lot of researches and developments are still required for practical application, these studies may help to open a door for the next step of flexible electronics.

There is a movement to realize a more efficient and energy-saving society by utilizing a vast amount of data obtained by a huge number of sensors installed in various places of our society. In order to construct such a sensor network, it is essential to realize self-powered sensor devices that do not require wirings. Energy harvesting technology is expected as one of the power sources. The energy harvesting technology gains a power source for driving sensors from small energy existing in the environment. Recently, researches to realize energy harvesters using graphene and carbon nanotubes has extensively being conducted. The nanocarbon materials is a highly conductive film with excellent flexibility and transparency so as to have installation flexibility. In this presentation, we introduce our research on flexible energy harvesting technologies and ultra-low power sensing circuit technologies.

The rapid development in smart robotics and prosthetic solutions calls for large-area and low-cost tactile sensor networks with high performance in both electrical and mechanical properties. Tremendous progress has been made recently on the research of flexible pressure sensors using various nanomaterials including organic and oxide semiconductors, carbon nanotube (CNT), semiconductor nanowire, etc. However, several key challenges remain, such as the fabrication of high-performance flexible transistor arrays with high density as well as the synthesis of ultra-thin piezoresistive film with high sensitivity. In this work, we report the integration of large-area CNT active matrix and highly sensitive piezoresistive thin film into a functional flexible tactile pressure sensor network. The 16×16 CNT active matrix was fabricated using highly purified (99.9%) semiconducting single-walled CNT solution, achieving a high density of 7 pixels per inch (ppi) over 4-inch area. A novel piezoresistive thin film was made with low-temperature processed multi-walled CNT composite, showing excellent pressure sensitivity with tunable thickness and flat surface. The highly integrated flexible tactile pressure sensor network was built on ultra-smooth polyimide with a surface roughness less than 1 nm, and it can be easily delaminated from the silicon substrate while maintaining the integrity and performance of the integrated system fabricated on top. The thickness of the entire flexible tactile sensor system can be controlled around 50 μm, leading to excellent flexibility. The demonstration in this work paves the road to build large-area electronic skins with high integration density and superior performance for future applications in smart robotics.

Carbon nanotube (CNT) is a promising candidate of field emission (FE) cathode materials, due to its small tip radius, high aspect ratio, and excellent electrical, thermal and mechanical properties. The FE current density of an individual pristine CNT with 14 nm diameter is ~10³ A/cm². On the other hand, vertical CNT array with good orientation can supply a mass of dense and uniform emission sites, and is expected to have the potential applications in devices of the X-ray sources, and FE flat plane displays so on. the length-diameter ratio in the modularized VACNT array synthesized by CVD technology is usually less than 3. It remains unclear how to suit applications for the novel devices because of the lack of strategies in large length-diameter ratio and micro-scaling up nanotube block array growth, assembly and patterning. In this paper, the vertically aligned carbon nanotubes (VACNT) micro-bundle arrays with different length-diameter ratios are invented by using of the pulsed laser ablation technology. The VACNT micro-bundle arrays with different length-diameter ratios from 4:1 to 30:1. These VACNT micro-bundle arrays have a very low threshold field (0.067 V/μm), excellent time stability and high field emission current densities (137.915 A/cm²). These are superior to carbon nanotubes grown directly on a silicon substrate. The large length-diameter ratio and the minor diameter of the VACNT array micro-bundle is potential applied in the field of considered to be the main reason for obtaining the excellent field emission performance.

Building electronic circuitry with atomic precision is beyond the capability of current, and in the foreseeable future, top-down lithographic techniques. In recent years, the US Office of Naval Research (ONR) has championed several research initiatives that sought to address this key challenge. One of the most exciting development and initial success that resulted from these initiatives is a bottom-up chemical synthesis technique that has proven highly effective at assembling two-dimensional (2D) graphene nanostructures with spatial precision control down to single atom and single bond level. In this presentation, I will briefly review the current status of this exciting new research field, and also the challenges that still remain. I will share some personal perspectives and speculate on possible future directions. The motivation behind the discussion is to socialize the basic ideas and capabilities with, and get feedback from, the relevant research communities, in order to further refine and improve future research agenda.

On-surfaces synthesis has become a versatile tool for the fabrication of atomically precise graphene nanostructures. Recent studies have successfully focused on the controlled synthesis of graphene nanoribbons (GNRs) (1), which offer a wide range of physical properties depending on edge structure and width. For instance, width-control in armchair GNRs gives access to band gap tuning in a wide range and allows, owing to the deterministic selection of a single width, for the preparation of an electronically homogenous material. For the case of nine atom wide armchair GNRs (9-AGNRs), the band gap is of 1.4 eV and allows, together with the development of suitable GNR transfer procedures, for the realization of field effect transistors with high on-currents (~1 µA) and high on/off current ratios (10⁵) (2).
Some particularly appealing properties of graphene nanostructures are achieved when the sublattice symmetry is broken. For zigzag GNRs this gives rise to spin-polarized edge states and opens new opportunities for spintronic applications. For the case of six atom wide zigzag GNRs (6-ZGNRs), we have recently identified suitable molecular building blocks and have confirmed the existence of edge-localized states for electronically decoupled 6-ZGNRs (3).
Such localized states also emerge from short zigzag segments embedded in armchair GNRs. By periodically positioning these zigzag segments along the GNR axis, this can be used to design one-dimensional lattices of coupled states yielding electronic bands in the band gap of the hosting armchair GNR (4). These bands are fully described by the Su-Schrieffer-Heeger (SSH) model, from which all relevant electronic properties can be derived by considering two coupling constants. Depending on the relative strength of these coupling constants, topologically trivial and non-trivial electronic quantum phases can be realized and give rise to characteristic topological boundary states in heterostructures consisting of topologically trivial and non-trivial segments. Tuning of the coupling constants of the SSH model also allows for the design of low-bandgap GNRs. In this limit, charge carrier properties can be approximated by the dispersion of massive Dirac fermions. The experimental realization of such a system together with its electronic characterization will be presented.
Finally, recent achievements in the on-surface synthesis of magnetic nanographenes will be discussed. Using inelastic scanning tunneling spectroscopy, the unambiguous identification of robust antiferromagnetic order in specific nanographenes is achieved and opens new opportunities for carbon-based magnetism.



1. J. Cai et al., Nature. 466, 470–473 (2010).
2. J. P. Llinas et al., Nat. Commun. 8, 633 (2017).
3. P. Ruffieux et al., Nature. 531, 489–492 (2016).
4. O. Gröning et al., Nature. 560, 209–213 (2018).

Bottom-up synthesis of graphene nanoribbons (GNR) enables us to obtain atomically-precise GNRs with well-defined edges [1]. The GNR heterojunction is one of the key elements required for realizing promising electronic devices, such as tunneling field-effect transistors and backward diodes. One of the promising applications of extremely narrow GNRs is a detector diode, which rectifies THz-waves into a direct-current (DC) output voltage [2]. Due to its small junction capacitance of sub-nm wide GNR p-n junction, it was predicted that the backward diode with GNR heterojunction can outperform the state-of-the-art diodes made from III-V compound semiconductors [2]. Among various techniques of manipulating electronic states of GNRs, we focus on the edge-functionalization of GNRs in this study. While a couple of edge-functionalized GNRs have been reported to date [3, 4], the edge-functionalization is still challenging mainly due to protonation of edges during cyclo-dehydrogenation [5, 6]. In this research, we report on some of our recent attempts for realizing edge-functionalized GNRs from molecular precursors.
First, we refer to the effect of edge-functionalization on the on-surface polymerization. It was found that the on-surface reaction of newly-developed precursor named 5-bromo-11(10-bromoanthracene-9-yl)anthra[2,3-b:7,6-b']dithiophene (BABAT) with isomers, which has a similar structure as 10,10'-Dibromo-9,9'-bianthryl (DBBA) with one anthracene substituted with anthradithiophene, ends in intramolecular cyclodehydrogenation [7]. Unlike DBBA, which is widely known to form anthracene polymers and armchair-edge graphene nanoribbons on Au(111), BABAT was found to make planar nanographene with intramolecular C-C bonding after annealing at 200°C on Au(111). The mechanism was investigated using first-principle density functional theory, which revealed that on-surface polymerization is not kinetically preferred in the case of BABAT. The intramolecular C-C bonding in DBBA biradicals, on the other hand, does not take place because of a faster reverse reaction. By referring to the electron density of BABAT biradicals, it was concluded that thiophene functionalization modifies the distribution of electron density in BABAT biradicals and facilitates electrophilic addition, leading to intramolecular C-C bonding after 200°C annealing. These results indicate that the design of radical moiety is particularly important in the on-surface Ullmann-type coupling.
The second part of our presentation is dedicated to the bottom-up synthesis of GNRs with edges functionalized by methylenedioxy functional groups. Unlike BABAT, it was confirmed by STM measurement that GNRs functionalized by methylenedioxy were obtained from anthracene trimer precursor functionalized by methylenedioxy. Unlike fluorine, which was detached during cyclodehydrogenation [5], methylenedioxy groups seem to be robust and survive cyclodehydrogenation process, while co-existence of hydroxyl groups was also confirmed by XPS. The GNR formation process was studied using the nudged elastic band (NEB) method within the density functional theory (DFT). One of the important findings is that methylenedioxy substitution opens an alternative path for dehydrogenation, which protects methylenedioxy groups from protonation. Due to electron-donating nature of methylenedioxy groups, it was also confirmed, by first-principles simulations, that the ionization potential and electron affinity of Medioxy-7AGNR differ from those of pristine 7AGNR, which may lead to the formation of a Type II heterojunction of GNRs.
This work was supported by CREST JST (no. JPMJCR15F1) and MEXT/JSPS KAKENHI grant nos. 26105004, 16H02286.

[1] Nature 466, 470 (2010). [2] Appl. Phys. Express 10, 074001 (2017). [3] ACS Nano 11, 7355 (2017). [4] J. Phys. Chem. C 120, 2684 (2016). [5] ACS Nano 11, 6204 (2017). [6] ACS Appl. Mater. Inter. 10, 31623 (2018). [7] ChemPhysChem in press, DOI: 10.1002/cphc.201900510

In contrast to zero-bandgap graphene, structurally confined graphene nanoribbons (GNRs) features open bandgaps due to the quantum confinement effect, which render them highly promising for nanoelectronic and optoelectronic applications.^[1] The properties of the GNRs are determined by their chemical structures, in particular the width and the edge configuration, which makes their precise structural control essential. Whereas such precision cannot be achieved by “top-down” fabrication methods such as “cutting” of graphene, bottom-up chemical synthesis can afford atomically precise GNRs with various structures.^[1,2] The synthesis is carried out through polymerization of a tailor-made monomer leading to a polymer precursor, which can be “planarized” and “graphitized” into GNRs by oxidative cyclodehydrogenation. While it is challenging to obtain very long GNRs through transition-metal-catalyzed or metal-surface-assisted polymerization, we have demonstrated solution synthesis of GNRs with average lengths larger than ~600 nm through AB-type Diels-Alder polymerization.^[2,3] Solubilizing alkyl chains can be attached on the GNR edges through the solution synthesis, making them dispersible in organic solvents and thus allowing for the liquid-phase processing for the device integration.^[3,4] By modifying the monomer structure, laterally extended GNRs with the width of up to ~2 nm could be obtained, revealing the lowering of the optical bandgap down to ~1.2 eV.^[2,5] We have also achieved edge functionalization of the GNRs by introducing a bromo group on the monomer. For example, introduction of perylene monoimide units led modulation of the self-assembly behavior, providing intriguing “rectangular” networks of GNRs.^[6] Moreover, introduction of spin-bearing organic radicals on the GNR edges allowed for spin injection to the aromatic core, inducing a delocalized magnetic state, which might have potential for application in the quantum computers.^[7] These results form a fundamental basis for applications of such GNRs in the future nanotechnology.

References
[1] X.-Y. Wang, A. Narita, K. Müllen, Nat. Rev. Chem. 2017, 2, 0100.
[2] A. Narita, Z. Chen, Q. Chen, K. Müllen, Chem. Sci. 2019, 10, 964.
[3] A. Narita, X. Feng, Y. Hernandez, S. A. Jensen, M. Bonn, H. Yang, I. A. Verzhbitskiy, C. Casiraghi, M. R. Hansen, A. H. R. Koch, G. Fytas, O. Ivasenko, B. Li, K. S. Mali, T. Balandina, S. Mahesh, S. De Feyter, K. Müllen, Nature Chem. 2014, 6, 126.
[4] T. Preis, C. Kick, A. Lex, D. Weiss, J. Eroms, A. Narita, Y. Hu, K. Müllen, K. Watanabe, T. Taniguchi, Appl. Phys. Lett. 2019, 114, 173101.
[5] Y. Hu, P. Xie, M. De Corato, A. Ruini, S. Zhao, F. Meggendorfer, L. A. Straasø, L. Rondin, P. Simon, J. Li, J. J. Finley, M. R. Hansen, J.-S. Lauret, E. Molinari, X. Feng, J. V. Barth, C.-A. Palma, D. Prezzi, K. Müllen, A. Narita, J. Am. Chem. Soc. 2018, 140, 7803.
[6] A. Keerthi, B. Radha, D. Rizzo, H. Lu, V. Diez Cabanes, I. C.-Y. Hou, D. Beljonne, J. Cornil, C. Casiraghi, M. Baumgarten, K. Müllen, A. Narita, J. Am. Chem. Soc. 2017, 139, 16454.
[7] M. Slota, A. Keerthi, W. K. Myers, E. Tretyakov, M. Baumgarten, A. Ardavan, H. Sadeghi, C. J. Lambert, A. Narita, K. Müllen, L. Bogani,, Nature, 2018, 557, 691.

On-surface synthesis has emerged as an efficient strategy for the synthesis of 1D and 2D covalent nanostructures (1). The vast variety of graphene nanoribbons (GNR) that have been synthesized during the last decade represent the most successful case of this bottom-up approach. GNRs have indeed been synthesized with atomic scale control over the edge structure, width, or chemical composition (3). However, extending dimensionality to 2D nanostructures has revealed itself as a paramount challenge for on-surface chemistry. In this talk, I will report on some novel 2D graphene nanostructures obtained by a hierarchical onsurface method. The method relies on the stepwise synthesis of polymeric chains that turn into arrays of aligned graphene nanoribbons that then couple into nanoporous graphene (NPG), which combines 1nm size ribbons and pores (4). As demonstrated by combining scanning tunneling (STM) and density functional theory (DFT), the peculiar topology of the NPG imprints a band-gap, novel electronic pore states and high electronic in-plane anisotropy. Our calculations have also predicted that the anisotropic electron propagation within the mesh gives rise to Talbot interference effects (5), analogous to photons in coupled waveguides. Moreover, as I will show with particular examples, all these unique properties of NPG can be tuned by controlling the geometry and/or composition of the pores.
1. J. Cai et al., Nature 466, 470-473 (2010).
2. Q. Fan, J. M. Gottfried, J. Zhu, Acc. Chem. Res. 48, 2484–2494 (2015).
3. D. G. de Oteyza, C. Rogero, Eds., On-Surface Synthesis II (Springer International Publishing, Cham, 2018), Advances in Atom and Single Molecule Machines.
4. C. Moreno et al., Science, 360, 199–203 (2018).
5. G. Calogero et al., Nano Letters 19, 576-581 (2019).

Graphene nanoribbons (GNRs) exhibit an electronic bandgap due to the lateral confinement of charge carriers and edge effects. They can be fabricated by bottom-up on-surface synthesis from molecular precursors resulting in atomically precise structures [1]. This approach promises tunable optical and electronic properties [2]. Their integration into functional devices will require an in-depth understanding of their structural properties, in particular after transfer from their metallic growth substrates to the target device substrate. We use Raman spectroscopy to characterize GNRs with a large area Raman mapping approach on optimized substrates exhibiting interference enhancement [3]. This allows the investigation of low-intensity geometry dependent Raman modes. We show that a longitudinal compressive mode (LCM) is particularly suited to investigate the length of bottom-up synthesized GNR, probe the effects of device processing as well as their interaction with different types of substrates [4].

[1] Cai, J. et al., Nature 466, 470–473, (2010)
[2] Chen, Y.-C et al., ACS Nano 7, 7, 6123–6128, (2013)
[3] Overbeck, J. et al., phys. stat. sol. (b), doi: 10.1002/pssb.201900343, (2019)
[4] Overbeck, J. et al., ACS Nano, doi: 10.1021/acsnano.9b05817, (2019)

Graphene nanoribbons (GNRs) are narrow strips of graphene, with exciting properties deriving from electron confinement and related band gap tunability[1]. The ability to tune GNRs’ electronic and magnetic properties by structural design at the single atom level makes them an ideal platform for a wide range of device applications, from classical transistors to spintronics. Despite their exceptional properties, significant challenges remain for GNR fabrication, processing and characterization. This contribution addresses some of the critical challenges in the further development of GNR technologies in particular on GNR fabrication, transfer, and characterization.
Bottom-up synthesis of graphene nanoribbons is most commonly performed under ultra-high vacuum conditions, which is one of the bottlenecks in the further technological advancement of this material. Additionally, little is known about the stability of ultra-narrow GNRs under ambient conditions or during device processing. In this work we focused on 9- and 5-atom wide armchair GNRs (9-AGNR and 5-AGNR, respectively) grown under ultra-high vacuum conditions on Au(111) and on vicinal gold surfaces Au(788), which promotes GNR's unidirectional growth on the narrow (111) terraces along the step edges. We first investigate the role of halogen functionalization of the precursor monomer in the final GNR length. Using different precursor monomers and synthesis protocols we show an increase in the average length for the 9-AGNR from 15 nm to 45 nm[2] and for the 5-AGNRs from 8 nm to 20 nm. GNRs grown on Au(111) substrates are transferred using a polymer-free method[3]. Raman spectra indicate no significant degradation of GNRs' quality, reveal their homogeneous distribution on the target surface and also show that GNRs have remarkably stability under ambient conditions tracked over a 2-year period[3]. GNRs grown on Au(788) are transferred via electrochemical delamination method[4], and Polarized Raman of the G mode show that GNRs retain their orientation after transfer with maximum Raman intensity along the ribbon axis and minimum intensity perpendicular to it.
These process steps allowed us to integrate both 9- and 5-AGNR into short channel field-effect transistors (FETs). Using the 9-AGNRs as channel material we demonstrate FETs with high on-current I_on >1 μA at V_d = -1 V, high I_on/I_off ratios of ~10⁵ and 10% device yield[5]. In a next step, aligned 9-AGNR-FET devices are produced using graphene electrodes, and a performance of I_on > 6μA at V_d = 0.1 V and high I_on/I_off ratios of ~10⁴ is observed along with an increase of device yield to 80-100%, due to the GNRs' orientation along the source-drain axis of the devices. GNR-FET devices using short 5-AGNRs (2-10 nm) as channel material and graphene as electrodes show linear I-V curves (metal-like behavior) at room temperature, while long 5-AGNRs (>20 nm) integrated into FETs with Pd as contacts behave as a semiconductor. The role of GNRs' surface coverage and length on the transfer efficiency and electronic/transport properties will be discussed.

References:
[1] J. Cai et al, Nature, 466, 470-473, 2010
[2] M. Di Giovannantonio et al, ACS Nano, 12, 74-81, 2018
[3]. G. Borin Barin et al, ACS Applied Nanomaterials, 2, 2184-2192, 2019
[4]. B.V. Senkovskiy et al, Nano Letters, 17, 4029-4037, 2017
[5]. J.P. Llinas et al, Nature Comm, 8, 633, 2017

Atomically precise graphene nanoribbons (GNRs) are at the forefront of nanocarbon research and hold great promise for electronic and optoelectronic applications. Theoretical studies have shown that narrow GNRs possess interesting electronic and magnetic properties that strongly depend on the nanoribbon’s width and edge structure. GNRs with different structures can be synthesized with atomic precision and fine-tuned properties by coupling properly designed molecular precursors and planarization of the resulting polymers.
In this talk, I will discuss our recent results on solution and on-surface synthesis and characterization of atomically precise GNRs. The solution approach is based on Yamamoto coupling of pre-synthesized molecular precursors followed by cyclodehydrogenation using Scholl reaction. This approach can be applied to the synthesis of a variety of GNRs, including pristine and nitrogen-doped GNRs with the chevron structure. The on-surface approach is based on the coupling of halogenated molecular precursors on metal substrates and subsequent cyclodehydrogenation of the resulting polymers at elevated temperatures.
Both solution and on-surface approaches were used to synthesize laterally extended chevron GNRs (eGNRs). The ribbons grown on Au(111) were used for extensive scanning tunneling microscopy (STM), noncontact atomic force microscopy (nc-AFM) and scanning tunneling spectroscopy (STS) characterization. They were found to exhibit a bandgap of about 2.2 eV, which is considerably smaller than the values reported for chevron GNRs in similar studies. The solution-synthesized eGNRs were processed into uniform thin films that exhibited improved electrical conductivity compared to similar films of regular chevron GNRs. The eGNR films were employed in gas sensors that showed very high responsivity to low molecular weight alcohols compared to similar sensors based on benchmark graphitic materials, such as graphene and graphene oxide. Finally, we demonstrated the first electronic nose system based on atomically precise GNRs that could reliably recognize analytes of nearly the same chemical nature, such as methanol and ethanol. Overall, this work demonstrates the entire path from designing new atomically precise GNRs with improved properties to their bottom-up synthesis, characterization, processing and implementation in electronic devices.

Graphene nanoribbons with sub-10 nm widths and ultra-smooth armchair edges are exciting candidates for next-generation semiconductor electronics because of their potential to exhibit exceptional charge and thermal transport properties. However, widespread integration of nanoribbons into commercial applications has been limited by challenges in producing nanoribbons with a high degree of structural precision and assembling the nanoribbons into large-area arrays on technologically-relevant substrates.

Here, we present a promising scalable approach for producing nanoribbons with sub-10 nm widths, smooth armchair edges, and excellent charge transport properties directly on conventional semiconductor wafer platforms via chemical vapor deposition (CVD). We have discovered that by exploiting highly anisotropic crystal growth of graphene from hydrocarbon precursors on Ge(001), narrow nanoribbons with high-aspect ratios can be synthesized. Due to the bottom-up, self-defining nature of growth, the nanoribbons have nearly atomically smooth armchair edges. The nanoribbon width can be continuously tuned to virtually zero by controlling the growth time and growth rate, which depends on the precursor flux and substrate temperature, and nanoribbons with sub-10 nm widths can be hundreds of nanometers in length. Breaking the symmetry of the Ge(001) surface by introducing a miscut results in wafer-scale unidirectional nanoribbon alignment. Control over both the alignment as well as the placement of individual nanoribbons is realized by initiating growth from lithographically-defined nanoscale seeds, enabling large-area hierarchical nanoribbon arrays. This nanoribbon synthesis is also adapted onto Si(001) wafers by utilizing epitaxial Ge interlayers.

Nanoribbons grown on Ge(001) via CVD have sizeable bandgaps and display high-performance charge transport properties at room temperature. For example, field-effect transistors with nanoribbon channels can simultaneously exhibit on/off conductance ratio of 2×10⁴ and on-state conductance of 5 µS. Therefore, this bottom-up nanoribbon synthesis may provide a high-throughput route for realizing future state-of-the-art technologies based on semiconducting graphene.

References: Jacobberger et al., Nat. Commun. (2015), ACS Nano (2017), Nanoscale (2019); Way et al. Nano Lett. (2018), J. Phys. Chem. Lett. (2019); Saraswat et al. J. Phys. Chem. C (2019); Kiraly et al. Appl. Phys. Lett. (2016).

With the fast development of emerging technologies such as artificial intelligence and internet of things, there has been higher and higher demand for future chips in terms of computing power, speed, and energy efficiency. Amid the slowdown of Moore's law, new chip architecture is needed to alleviate or overcome the so-called von Neumann bottleneck. Monolithic three-dimensional (3D) integration of multiple layers of functional components, such as emerging memory, logic and sensors, through high-density interlayer interconnects, provides a promising approach to build highly energy-efficient system-on-chip (SoC). On one hand, carbon nanotube (CNT) has been demonstrated as a superior candidate for high-speed logic technology that can be fabricated with low-temperature process. On the other hand, resistive random-access memory (RRAM) has been extensively studied for energy-efficient computing-in-memory (CIM) by taking advantage of its analog switching characteristics. In this work, we report the monolithic integration of CNT logic circuits and analog RRAM array using back-end-of-line (BEOL) compatible low-temperature fabrication processes. The performance of CNT integrated circuits and analog RRAM devices have been carefully optimized. Hybrid circuits such as ternary content-addressable memory (TCAM) are also fabricated. Our work represents a critical step towards 3DSoC with CIM capability.

Carbon nanotubes (CNTs) is one of the most promising candidates for next-generation electronics for its extraordinary intrinsic properties, including high carrier mobility, excellent mechanical flexibility and solution processability, which are desirable in unique application scenarios, from smart sensing to biomedical implants. However, the integrated circuits (ICs) with enough complexity and integration scale for signal processing, which is the core units in modern electronic systems, are still lacking, which is mainly due to the nonuniformity of the devices, low yield manufacturing and large power consumption. This presentation would detail approaches for wafer-scale, high-yield and highly-uniform manufacturing of CNT devices and ICs. Chips using CNTs as semiconducting materials are 100% successfully transferred to water-soluble substrates with average device yield of 96.6%. Great uniformity is also demonstrated with minimum standard deviation of 55 mV in threshold voltage. The ICs could be successfully transferred onto biological surfaces, such as human wrist, biodegradable polymers and plant leaf. Integrated with other sensors, the CNT based flexible ICs could demonstrate system-level environmental monitoring. In addition, to further push forward the CNT based flexible electronic devices, we propose a gate engineering approach to boost the electrical performance of the carbon nanotube transistors, to make them more energy-efficient and fit for the low power consumption requirement in practical biological or environmental applications.

Reference
[1] L. Xiang, H. Zhang, G. Dong, D. Zhong, J. Han, X. Liang, Z. Zhang, L. Peng and Y. Hu, Nature Electronics, 2018, 1, 237-245.
[2] L. Xiang, F. Xia, H. Zhang, Y. Liu, F. Liu, X. Liang and Y. Hu, Advanced Functional Materials, 2019, 1905518
[3] L. Xiang, Y. Wang, P. Zhang, X. Fong, X. Wei, Y. Hu, Nanoscale, 2018, 10, 21857-21864
[4] L. Xiang, F. Xia, W. Jin, X. Zeng, F. Liu, X. Liang, Y. Hu, in submission.

Thermal imaging has applications in non-destructive imaging, testing, detection, and many more areas. Here, we demonstrate a thermal imaging sensor design made of carbon nanotubes (CNTs) using the thermoelectric effect. We designed the sensor device consisted of an array of p-type CNT fibers that crisscross with another array of n-type CNT fibers. Temperature differences between different p-n fiber knots produce voltage differences. Measuring the voltage differences between p and n terminals of each knot, the temperature of each knot can be
determined and the thermal image of the object can be obtained. For a 7×7 sensor array, the p-n terminal voltage difference of ∼0.22 mV was attained for a temperature difference of 100 K. As CNT gives flexibility and tunability of the Seebeck coefficient in the device, this thermal sensor device can be made wearable with nanorange thermal imaging with extreme precision.

The direct contact of vertically aligned carbon nanotube (VACNT) based field emitter arrays (VACNT-FEAs) to their growth substrate and their structurally controlled morphology are critical factors for applications in vacuum micro/nano-electronic devices. Here, we propose a method for fabricating a high-performance VACNT-FEAs by controlling the morphology of VACNT micro/nano-pillars synthesized directly on stainless steel substrates without external metal catalyst particles via nanosphere lithography (NSL). Exposure of carbon precursor gas to the limited space, created after the NSL using polystyrene spheres of different sizes, on the surface of the thermally heated bulk catalytic substrate was a key for manipulating the morphology of the VACNT-FEAs. Field emission properties of the as-synthesized VACNT-FEAs were measured and found fascinating field emission properties along with high emission current stability. The correlation between the number of emitters in a pillar structure and their field emission behavior has been investigated. The proposed method may be used to fabricate the cold cathodes for vacuum electronic devices requiring high emission current density along with extended emission current stability.

Elusive measurements of the Longitudinal optical (LO) phonon mode frequency of isolated Cn@MWCNT systems (linear carbon chains encapsulated by multi-wall carbon nanotubes) were studied under extreme high pressure via Resonance Raman spectroscopy. The pressure-dependent frequency softening with increasing pressure is discussed in terms of a simple force constant model in which the spring constants present a dependence with pressure. The model allowed us to easily obtain both the Young’s modulus and the Gruneisen parameter associated to such LO mode. In particular, the Gruneisen parameter was found to be much higher than that of Graphene and CNTs. Additionally, the LO phonon lifetime was found to be decrease with increasing stress.

Finite-thickness films of aligned single-wall carbon nanotubes (SWCNT) are recently demonstrated to have extraordinary optoplasmonic properties including a tunable negative dielectric response in a wide photon energy range[1]. We explain this theoretically using the Maxwell-Garnett (MG) mixing[2] for the dielectric responses of planar periodic arrays of identical SWCNTs. We start with the calculation of the conductivity for an individual SWCNT using the (k*p)-method of band structure calculations[3]. Then we use the low-energy plasmonic response calculation technique earlier proposed by one of us for finite-thickness metallic films with periodic cylindrical anisotropy[4], to derive an analytical expression for the dynamical dielectric response tensor of an ultrathin finite-thickness periodically aligned SWCNT array. The dielectric responses of about a dozen of SWCNT arrays with different chiralities and very close diameters are then mixed using the MG method to obtain the anisotropic optical response of the finite-thickness SWCNT film which explains the experimental ellipsometry measurements reported[1]. We also note that in such SWCNT films the inhomogeneously broadened interband plasmon resonance of individual SWCNTs overlaps with the broadened exciton resonance, thus making the exciton-plasmon coupling and associated photon bandgap formation possible[5] for these and similar cylindrically anisotropic compound systems including double-walled carbon nanotubes.

Acknowledgements: NSF DMR-1830874 (C.A.), DOE DE-SC0007117 (I.V.B.)

References:
[1] J.A.Roberts, S.-J.Yu, P.-H.Ho, S.Schoeche, A.L.Falk, and J.A.Fan, Nano Lett. 19, 3131 (2019)
[2] V.A.Markel, J. Opt. Soc. Am. A33, 1244 (2016)
[3] T.Ando, J. Phys. Soc. Jpn. 74, 777 (2005)
[4] I.V.Bondarev, Opt. Mater. Expr. 9, 285 (2019)
[5] I.V.Bondarev, Phys. Rev. B 85, 035448 (2012)

Electronic devices with configurability to providing multiple functions are of great interests for their superior adaptability to the ever-changing and multifarious application scenarios. Here, we report flexible integrated circuits possessing configurable functions constructed with dual-material gate (DMG) devices based on carbon nanotube thin films, which can serve as either transistors or diodes, on a 2-μm-thick parylene substrate. When configured as a transistor, the DMG device has great advantages over the normal-gated devices regarding the current on/off ratio , the subthreshold swing and the drain-induced barrier lowering due to the regulated energy band distribution in channel area. When operating as a diode, a typical DMG device demonstrates a sufficient rectification ratio of approximately five orders and a diode-on-current of over 26μA. Scalable manufacturing of DMG devices was also demonstrated with great uniformity both in diode and transistor configurations. Finally, multifunctional integrated circuits, which can dynamically switch their function from rectifier to follower or from OR gate to voltage adder by changing controlling signals, were constructed. The functional-configurability, together with scalable manufacturing and the realization on ultrathin flexible substrates, will open up great opportunity for the future environmentally-adaptive system in the field of flexible electronics.

Carbyne, the structure consisting of one-dimensional (1D) chain of sp-hybridized carbons is predicted to have greater tensile strength than graphene, making it a highly sought material. However, the highly reactive sp-hybridized carbon chains convertto the more stable sp²-hybridized phase through cross-linking reactions and ozonolysis by exposure to oxygen. We have developed a new class of molecules/materials, “pseudocarbynes”, that we define as sp-hybridized carbon chains that are stabilized through non-covalent interactions with small metal clusters. Here, I will present our progress in synthesizing pseudocarbynes through two distinct experimental techniques: 1) through direct laser ablation of a gold target submerged in liquid, and 2) through the self-assembly of conjugated carbon chains with a colloidal solution of goldnanoparticles. Pseudocarbynes remain stable in the solid-state (tested for up to several months). We characterize the sp-hybridized carbon structure by analyzing FT-IR and Raman spectroscopies, as well as X-ray powder diffraction directly in the solid state. By securing stability in the solid state, pseudocarbynes are expected to be a good candidate for a diverse set of optical or electronic applications.

Graphene nanoribbons (GNRs) are known for their highly tunable physical properties and great promise for a broad spectrum of electronic applications. The current study focuses on the chemical design of novel molecular precursors for the bottom-up solution synthesis of atomically precise GNRs. As semiconducting properties of GNRs are highly dependent on their width, edge geometry, and heteroatom doping, here we report on the design, synthesis, and characterization of three different GNRs, which include C4N- and C8N-GNR heterojunctions derived from the pristine with nitrogen-substituted chevron-based monomers, laterally extended chevron GNRs (LecGNRs) and edge-modulated chevron GNRs (EmcGNRs). The XPS-data for C4N- and C8N-GNR heterojunctions confirms the presence of nitrogens and suggests the random positions of pristine and nitrogen-doped GNR segments. The data on the structural, optical and electronic properties of all three GNRs will be presented. In particular, LecGNRs showed an expected lower optical bandgap than their counterparts, while EmcGNRs demonstrate a higher optical bandgap than the pristine chevron GNRs. While this study has a particular focus on the solution synthesis of nanoribbons, the developed molecular precursors may also be compatible with the established on-surface synthesis approaches for GNRs.

Electrospinning (ES) is a rapid and simple manufacturing process for producing polymer materials with high surface-area-to-volume ratios. Traditional ES fabrication involves an electrostatic force that pulls polymer from a grounded spinneret tip to a charged metallic deposition surface held at high voltage. Due to the high voltages and the charged electrodes required for ES fiber creation, ES conductive polymers presents a great challenge. During ES of conductive materials, a circuit is completed from the ES spinneret to the deposition substrate, resulting in electrical arcing and non-predictable deposition. In the work presented here, a handheld ES device that incorporates directed airflow was used for deposition of conductive polymer fiber materials directly onto substrates regardless of dielectric charges. In the portable ES device described, deposition occurs by placing a conductive, ring electrode between a spinneret and a deposition surface. The resulting electrostatic force pulls polymer from the spinneret and toward the ring electrode. Directed airflow then carries the polymer fiber through the center of the ring electrode and onto a non-conductive surface beyond the device. Such a system prevents electrical arcing, thereby improving predictability and allowing for the creation of a fiber mat capable of carrying electrical current ranging from direct to alternating. Electrically conductive polymer fiber mats are desirable due to their mechanical compliance, high surface area, potential for optical transparency, and a variety of electrical, optical, and mechanical applications. These applications include but are not limited to materials for water filtration, photocatalysis, light polarization, light energy harvesting, sensors, actuators, and miniature electrical components.

The portable, electrostatic and air driven (EStAD) ES device presented includes enclosed electrodes and a fiber propulsion system that provides several additional benefits as compared to other portable ES devices including improved safety for the user and predictable deposition with controllable directionality. A highly desirable and novel application of the EStAD device is that it can be used to electrospin conductive materials onto non-conductive substrates without the drawbacks of traditional ES. In the proposed work, the EStAD device will be used to deposit conductive sensors onto non-conductive substrates. Carbon nanotube (CNT) polymer composites will consist of multi-walled CNTs (MWCNTs) treated with amphiphilic surfactants or untreated single and double-walled CNTs dispersed in the neat polymer. Previously, amphiphilic surfactants were used to create homogenously dispersed MWCNT in an aligned nanofiber network with improved conductivity. The feasibility of using untreated single and double-walled CNTs will also be investigated for sensor creation on non-conductive substrates. The preparation process will include a solution-based purification process (ultracentrifugation) in order to inexpensively achieve high purity, metallic CNTs that have been used in high-performing electronic devices. High-purity metallic CNTs are more suitable for superior optical transparency because poorly conducting, strongly absorbing impurities, and semiconducting CNTs will be eliminated through the purification process. Both amorphous nanofiber mats and aligned fiber arrays will be created for subsequent testing. A four-point probe will be used to measure conductivity across the composite CNT nanofiber mats or arrays. The electronic response of nanofiber sensors to analytes (e.g. water vapor and ammonia) will be characterized through current-voltage (I-V) measurements. Morphological surface characterization will be performed through scanning electron microscope (SEM) imaging and analysis, and CNT dispersion within the composite matrix will be analyzed through transmission electron microscope (TEM) imaging and analysis.

Chemical functionalization of the sidewalls of semiconducting single-wall carbon nanotubes (SWCNTs) has attracted intense attention because of the localization of tightly bound excitons at defect sites on the nanotube surface, allowing single-photon emission (SPE) at room temperature. Such covalently functionalized SWCNTs hold the potentials as single-photon sources (SPSs) for the applications of quantum photonics and quantum information processing. Aryl sp³ defects has been sparsely and covalently implanted to the random positons of SWCNT sidewalls through diazonium chemistry, leading to the trapping of mobile band-edge excitons in the aryl defects and causing spectrally shifted photoluminescence (PL). Single photon has been emitted through the radiative recombination of localized excitons at the defect sites in the altered nanotube with ultrahigh single-photon purity (99%).
Previous reports about chemically functionalized SWCNTs as single-photon emitters are based on sparse and random placement of oxygen atom or aryl group on the nanotube surface. Incorporation of multiple defects with spatial pattern in nanotube sidewalls has not been achieved until recent discovery of covalently functionalizing SWCNTs with single-stranded DNA (ssDNA). The coating agents such as surfactants, polymers, or DNA are required for individualizing nanotube particles. ssDNA oligomers have been found to helically wrap around nanotube surfaces trough noncovalent interactions. Recently, a room-temperature chemical reaction in samples of ssDNA-suspended SWCNTs has been found between the guanine nucleotides in the ssDNA coating and the nanotube sidewall in the presence of singlet oxygen (¹O₂). Therefore, introducing and controlling spatially patterned guanine defects on nanotube surfaces is easily achieved by manipulating ssDNA sequences. Such DNA-templated covalent functionalization allows smooth modulation of SWCNT electronic energy levels along the nanotube axis with customized spatial pattern and depth.
Each reacted guanine nucleotides creates a defect with shallow trapping depth in the nanotubes. The axial separation between guanines in the (GT)_n ssDNA coating is estimated to be ~0.4 nm, which is at least 5 times smaller than the exciton size. Therefore, excitons in the DNA-tailored SWCNTs are capable of sensing the cumulative potential of several nearby guanine-introduced defects. On the contrary, the exciton is localized in a defect site with deep trapping depth for SWCNT samples with sparse chemical doping of oxygen atom or aryl group on the nanotube surface. We have found that single photon can be generated from the defect-state emission in DNA-altered SWCNTs. Notably, quantum coupling of multiple defect-states has been achieved, resulting from multiple guanine defects with spatial patterning on the nanotube surface.

In this study, we developed a novel approach to synthesize high-quality metal-free Nitrogen-doped graphene quantum dots (N-GQDs) with high quantum yield, via irradiation of s-triazene in a solution with benzene by using pulsed laser. The TEM, HRTEM, XPS, XRD, Raman spectroscopy and FTIR were carried out to observe the morphology, size distribution, crystalline structure and to prove successful doping of GQDs with nitrogen atoms. Furthermore, for the first time, to our knowledge, their magnetic properties were investigated. The results indicate that N-GQDs exhibit superparamagnetic behavior. The specific size, shape and zigzag edge structure of N-GQDs were considered to explain the origin of the observed magnetism. The magnetization dependence led to estimating the N-GQD material magnetic permeability for different ambient temperatures. From the zero-field-cooled (ZFC) and field-cooled (FC) magnetization measurements, carried out at 50 Oe magnetic field strength, we estimated the blocking temperature T_B to be around 300 K. Based on the experimental data analysis, the magnetic permeability, number of correlated spins per single N-GQD, and number density of superparamagnetic N-GQD per gram of material were estimated. The excellent superparamagnetic properties together with optical properties manifested by N-GQDs have the potential to lead to high performance biomedical applications.

Carbon nanotube (CNT)-based electronics has been considered one of the most promising candidates to extend Si complementary metal-oxide-semiconductor (CMOS) technology. In particular CNTs have been investigated for various electronic and optoelec-tronic device applications, such as sub-10nm CMOS devices which outperform that of state-of-the-art Si based CMOS devices in both speed and power consumption [1-5], as well as idea material for monolithic optoelectronic integration with complementary MOS-compatible signal processing circuit [6-8].
Prototype device studies on individual CNTs revealed that CNT based devices have the potential to outperform Si CMOS technology in both performance and power consumption. With a well-designed device structure and in combination with graphene, we showed that high-performance top-gated CNT FETs with a gate length of 5 nm can be fabricated. Significant progress has also been made in fabricating carbon nanotube low-power devices. An efficient way to reduce the power is to lower the supply voltage VDD, but this voltage is restricted by the 60 millivolts per decade thermi-onic limit of subthreshold swing (SS) in FETs. A Dirac source (DS) with a much narrower electron density distribution around the Fermi level than that of conventional FETs was recently proposed and demonstrated using CNT to reduce SS [2].

References
[1] C. Qiu et al., Scaling carbon nanotube complementary transistors to 5-nm gate lengths, Science 255 (2017) 271.
[2] C. Qiu et al., Dirac source field-effect transistors as energy-efficient and high-performance electronic switches, Science 10.1126/science.aap9195 (2018).
[3] D.L. Zhong et al., Gigahertz integrated circuits based on carbon nanotube films, Nature Electronics 1 (2018) 40.
[4] L.-M. Peng et al., Carbon-based electronics: A new stage for flexible nanotube devices, Nature Electronics 1 (2018) 158.
[5] L. Xiang et al., Highly uniform carbon-based low-power electronics with nio-integration capability in a transferrable form, Nature Electronics 1 (2018) 273.
[6] L.J. Yang et al., Extremely efficient photovoltage multiplication in carbon nanotubes, Nature Photonics 5 (2011) 672.
[7] Y. Liu et al., Carbon nanotube-based three-dimensional monolithic optoelectronic integrated system, Nature Communication 8 (2017) 15649.
[8] Y. Liu et al., Electrically-driven monolithic plasmonic interconnect circuits fabricated on a subwavelength scale from carbon nanotubes, Science Advances 3 (2017) e170145.

At this exact moment when future applications are demanding massive improvements in computing performance, conventional approaches to improving computing are becoming increasingly challenging. Nanosystems – leveraging unique properties of emerging nanotechnologies to realize new systems and architectures – promise to enable the next wave of computing gains for next-generation applications as well as provide the means for developing rich new functionalities that lie beyond the scope of traditional computing today. Yet despite this promise, imperfections and variations inherent in beyond-silicon emerging nanotechnologies have made realizing large-scale integrated systems infeasible. In this talk, I describe how by combining advances across the entire stack – from nanomaterial synthesis to new processing techniques, circuit design methodologies, and architectures – these challenges can be overcome to realize the first large-scale and complex electronic systems leveraging beyond-silicon emerging nanotechnologies. As a case-study, I will describe the journey of transforming carbon nanotubes from an interesting scientific nanomaterial into the foundation for realizing the first beyond-silicon microprocessor built entirely from carbon nanotube transistors – RV16X-NANO, a 16-bit microprocessor based on the RISC-V instruction set. I will conclude by describing how emerging nanotechnologies are beginning to be transferred from the “lab” to the “fab,” promising a new and exciting age for nanoelectronics and computing.

The National Research Council (NRC) is Canada’s premier organization for research and technology development. NRC’s Printable Electronics (PE) program develops materials and inks for additive manufacturing in order to enable a sustainable PE sector in Canada. Progress has been made over the last 15 years, yet many challenges remain at the materials, fabrication and integration level, limiting performance and commercialization. This presentation covers our progress in high purity semiconducting single-walled carbon nanotube (sc-SWCNT) enrichment and transistor fabrication via solution based processes as well as conductive molecular ink development ^[1-11]. I will highlight our recent progress in understanding sc-SWCNT enrichment using conjugated polymers, with special consideration given to the effects of solvent parameters and doping on the mechanism and yield/purity of the final product ^[2-4]. Removal of the wrapping polymer using a dry process on-wafer will also be described ^[5]. Developments in conductive molecular inks and progress towards the commercialization of new applications will be described such as in-mold electronics ^[6-8]. Challenges and advances associated with using polymer-based dielectrics and encapsulants will be discussed as well ^[9-10]. Such transistor packages have enabled the realization of fully inkjet-printed transistors as a result of the excellent electrical properties of sc-SWCNTs ^[2,11]. A demonstration of a fully additive process to make TFT backplanes via R2R printing will also be highlighted. The development of new functional nanomaterials and their successful integration into devices will enable additive manufacturing of TFT arrays and in-mold electronics.

[1] J. Lefebvre et al. Acc. Chem. Res. 50, 2479 (2017).
[2] J. Ouyang et al. ACS Nano 12, 1910 (2018).
[3] J. Ding et al. J. Phys. Chem. C 120, 21946 (2016).
[4] Z. Li et al. ACS Omega 3, 3413 (2018).
[5] Z. Li et al. Adv. Funct. Mater. 28, 1705568 (2018).
[6] A. J. Kell et al. ACS Appl. Mater. Interfaces 9, 17226 (2017).
[7] C. Paquet et al. Nanoscale 10, 17226 (2018).
[8] B. Deore et al. ACS Appl. Mater. Interfaces 11, 38880 (2019).
[9] J. Lefebvre et al. Appl. Phys. Lett. 107, 243301 (2015).
[10] Lapointe et al. Appl. Mater. Interfaces 11, 36027 (2019).
[11] C. Homenick et al. ACS Appl. Mater. Interfaces 8, 27900, (2016)

Giant piezoresistivity of Carbon Nanotubes (CNTs) is a unique feature promoting further miniaturization of strain sensors. In the context of condition monitoring for safety relevant applications, Industry 4.0, or robotics, especially integrated scalable solutions are overdue. CNT-based electro-mechanical transducers were already shown in proof-of-principle studies [1, 2] and recently also scalable technologies were applied [3]. However, there are still remaining challenges hindering reproducible device fabrication. To emphasize, on the material side a lack of control over chirality and length distribution of single-walled CNTs leads to a distribution of sensor properties. On the other side, an undefined initial CNT-strain condition makes reliable and controlled operation of sensors very challenging.
Here, we have addressed several in-depth studies along a reproducible CNT integration technology, such as length separation, role of CNT/metal contacts, pre-straining CNTs, and their impact on CNT device performance. This integration technology was consequently utilized on wafer-level in order to enable a scalable fabrication of CNT-based piezoresistive sensors in the perspective of forthcoming heterogeneous system-integration in e.g. safety-relevant ASICs.
We found significant improvements on the CNT-FET performance by integrating pre-sorted long CNTs which is mainly correlated to reduced network effects, since less amount of CNT to CNT intersections are formed in the transistor channel [4]. In particular, we statistically revealed improvements in terms of fabrication yield, charge carrier mobility and subthreshold controllability. Moreover, we identified tunneling-related contact effects in CNT-based piezoresistive sensors to enhance the intrinsic gauge factor up to 165% [5]. The gauge factor can further be controlled by the input drain-source and gate-source voltage.
In addition, we realized pre-straining of integrated CNTs, which was verified by a significant blue-shift of the CNT-specific phonon mode G⁺ of up to 6.2 cm^-1, with respect to un-strained CNTs. Based on previous literature [6], we assessed the built-in strain ranging from 0.4% up to 0.8%.
In conclusion, we emphasize several technological aspects of the process value-chain and their relevance for a successful manufacturing of miniaturized CNT-based piezoresistive sensors for condition monitoring of forthcoming smart systems.


References
[1] Grow R J, Wang Q, Cao J, Wang D and Dai H, Appl. Phys. Lett. 86 93104, 2005
[2] Stampfer C, Helbling T, Obergfell D, Schöberle B, Tripp M K, Jungen A, Roth S, Bright V M and Hierold C, Nano letters 6 233–7, 2006
[3] Böttger S, Schulz S E and Hermann S, Procedia Engineering 168 692–5, 2016
[4] Böttger S, Hermann S, Schulz S E and Gessner T, Nanotechnology 27 435203, 2016
[5] Böttger S, Wagner C, Lorkowski F, Hartmann M, Schuster J and Hermann S, Sensors and Actuators A: Physical 295 288–95, 2019
[6] Liu Z, Zhang J and Gao B, Chemical communications 6902–18, 2009

Carbon nanotube field-effect transistors (CNFETs) are a promising candidate for implementing highly energy-efficient future digital systems. This is due to the superior transport properties of carbon nanotubes (CNTs) together with small body thickness. Furthermore, low-temperature (< 200°C) fabrication of CNFETs enables new monolithic three-dimensional (3D) architectures with computation immersed in memory, enabling up to 1,000-fold system-level energy-delay-product benefits for abundant-data applications (such as deep learning). In this talk, we will present ways in which many challenges traditionally associated with CNFETs have been overcome in the recent years. At the device-level, hysteresis in CNFET electrical characteristics is reduced to < 0.5% of the V_GS sweep range by studying charge impurities around CNTs. CNT PFET with low contact resistance of 6.5 kΩ (10 nm contact length) has been successfully demonstrated. At the circuit-level, a combination of fabrication and design techniques (the imperfection-immune paradigm) overcomes substantial inherent imperfections and variations traditionally associated with CNTs. The imperfection-immune paradigm has been key to the demonstration of CNFET digital VLSI circuits, as well as monolithic 3D NanoSystems that will also be highlighted in this talk.

Recent advances in on-surface organic synthesis have led to the development of bottom-up graphene nanoribbon (GNR) growth with atomic precision¹. Bottom-up synthesized GNRs have promising electronic properties for high performance field effect transistors (FETs) and ultra-low power devices such as tunneling FETs². However, the short length, wide band gap, and random spatial orientation of GNRs have impeded the fabrication of devices with the expected performance and switching behavior³. Here, recent progress in integrating bottom-up synthesized GNRs into devices will be presented. High on-current and high on-off ratio FETs with a nine-atom wide GNRs as the channel material were successfully fabricated⁴. Strategies to further improve FET device yield and performance with the parallel arrays of GNRs transferred from single crystal Au (788) will be discussed. Finally, a new approach for the transfer-free fabrication of GNR devices has been developed.
References
[1] Cai, J. et al., Nature 466, 470-473 (2010).
[2] Smith, S. et al., IEEE Electron Device Lett. 39, 143-146 (2018).
[3] Bennett, P. B. et al., Appl. Phys. Lett. 103, 253114 (2013).
[4] Llinas, J. P. et al., Nat. Commun. 8, 633 (2017).

We have developed a new, simple, scalable method based on novel plasma catalytic reaction [1-3] for directly fabricating narrow graphene nanoribbons (GNRs) devices on an insulating substrate [4]. Since the establishment of our novel GNRs fabrication method, direct conversion of a Ni nanobar to a suspended GNR is now possible. Indeed, GNRs can be grown at any desired position on an insulating substrate without any post-growth treatment, and the wafer-scale synthesis of suspended GNRs arrays with a very high yield (over 98%) is realized [5]. The growth dynamics of suspended GNRs is also investigated through the systematic experimental study combined with molecular dynamics simulation and theoretical calculations for phase diagram analysis. Unique optoelectrical property, known as persistent photoconductivity (PPC), is also observed in our suspended GNRs devices. By using the PPC, GNRs-based non-volatile memory operation is demonstrated [6]. High thermoelectric performance is also shown in our as-grown suspended GNRs [7]. We believe that our results can contribute to pushing the study of atomically thin layered materials from basic science into a new stage related to the optoelectrical applications [8-11] in industrial scale.

[1] T. Kato and R. Hatakeyama, ACS Nano 4 (2010) 7395.
[2] T. Kato and R. Hatakeyama, ACS Nano 6 (2012) 8508.
[3] B. Xu, T. Kaneko, Y. Shibuta, T. Kato, Scientific Reports 7 (2017) 11149.
[4] T. Kato and R. Hatakeyama, Nature Nanotechnology 7 (2012) 651.
[5] H. Suzuki, T. Kaneko, Y. Shibuta, M. Ohno, Y. Maekawa, and T. Kato, Nature Communications 7 (2016) 11797.
[6] H. Suzuki, N. Ogura, T. Kaneko, T. Kato, Scientific Reports 8 (2018) 11819.
[7] Q.-Y. Li, T. Feng, W. Okita, Y. Komori, H. Suzuki, T. Kato, T. Kaneko, T. Ikuta, X. Ruan, K. Takahashi, ACS Nano 13 (2019) 9182.
[8] T. Kato and T. Kaneko, ACS Nano 8 (2014) 12777.
[9] T. Kato and T. Kaneko, ACS Nano 10 (2016) 9687.
[10] T. Akama, W. Okita, R. Nagai, C. Li, T. Kaneko, T. Kato, Scientific Reports 7 (2017) 11967.
[11] C. Li, T. Kameyama, T. Takahashi, T. Kaneko, T. Kato, Scientific Reports 9 (2019) 12958.

Graphene nanoribbons (GNRs) have attracted a strong interest from researchers worldwide, as they constitute an emerging class of quantum-designed materials [1-4]. Some of the major challenges towards their exploitation in electronic applications are their reliable contacting due to their small size (< 50 nm), as well as the preservation of their physical properties upon device integration.

In this combined experimental and theoretical study, we report on the first demonstration of quantum dot (QD) transport in atomically precise GNRs [5]. The devices consist of a film of aligned 5-atoms wide GNRs (5-AGNRs) transferred onto graphene electrodes with a sub 5-nm nanogap [6-8]. We demonstrate that the narrow-bandgap 5-AGNRs exhibit metal-like behavior resulting in linear IV curves for low bias voltages at room temperature and single-electron transistor behavior at 13 K. By performing spectroscopy of the molecular levels, we obtain addition energies in the range of 200-300 meV. DFT calculations predict comparable addition energies, and reveal that, due to the finite length of the 5-AGNRs, two states at the termini of the 5-AGNRs are formed within the bandgap of the infinite ribbons. The properties of the obtained QD reflect the band-structure and therefore demonstrate the preservation of the GNR properties upon device integration, a critical step in the realisation of more exotic GNR-based nano-electronic devices.

References:
[1] Cai, J. et al., Atomically precise bottom-up fabrication of graphene nanoribbons. Nature 466, 470 (2010).
[2] Ruffieux, P. et al., Electronic structure of atomically precise graphene nanoribbons. ACS Nano
6, 6930-6935 (2012).
[3] Chen, Y.-C. et al., Tuning the band gap of graphene nanoribbons synthesized from molecular
precursors. ACS Nano 7, 6123-6128 (2013).
[4] Slota, M. et al., Magnetic edge states and coherent manipulation of graphene nanoribbons. Nature 557, 691 (2018).
[5] M. El Abbassi et al., Controlled quantum dot formation in atomically engineered graphene nanoribbons field-effect transistors. To be submitted.
[6] M. El Abbassi et al., From Electroburning to Sublimation: Substrate and Environmental effects in the Electrical Breakdown. Nanoscale 2017, 9, 17312-17317.
[7] L. Posa, M. El Abbassi, et al., Multiple physical timescale and dead time rule in few-nm sized graphene-SiOx-graphene memristors. Nano letters 2017, 17, 6783-6789.
[8] M. El Abbassi, et al., Robust graphene-based molecular devices. Nature Nanotechnology, 14, 957-967, 2019.

We have recently demonstrated the rational design and experimental realization of a graphene nanoribbon (GNR) superlattices that hosts a 1D array of symmetry-protected topological states, thus generating otherwise inaccessible electronic structure. This new class of materials can be thought of as an extended form of poly-acetylene wherein highly localized half-filled topological states replace the familiar single occupied p-orbitals along a conjugated p-system. Experimental results and first-principles calculations reveal that the frontier band structure of these GNR superlattices is defined purely by the coupling between adjacent topological interface states and can be tuned all the way from a semiconductor to a metal. This novel manifestation of 1D topological phases presents an entirely new route to band engineering in 1D materials based on precise control of their electronic topology, and is a promising new platform for future studies of 1D quantum spin physics.

Bottom-up prepared carbon nanostructures appear as promising platforms for future carbon-based nanoelectronics due to their atomically precise and versatile structure.[1] An important breakthrough is the recent preparation of nanoporous graphene (NPG) as an ordered covalent array of graphene nanoribbons (GNRs).[2] Within NPG, the GNRs may be thought of as 1D electronic nanochannels through which electrons preferentially move, highlighting NPG’s potential for carbon nanocircuitry. However, the π-conjugated bonds bridging the GNRs give rise to electronic crosstalk between the individual 1D channels, leading to spatially dispersing electronic currents.[3] In this talk, we propose a chemical design of the bridges resulting in destructive quantum interference, which blocks the crosstalk between GNRs in NPG, electronically isolating them. Our multiscale calculations reveal that injected currents can remain confined within a single, 0.7 nm wide, GNR channel for distances as long as 100 nm.[4] The concepts developed in this work thus provide an important ingredient for the quantum design of future carbon nanocircuitry. In addition, our latest results demonstrating an external control of these characteristics will also be presented.

[1] Grill, L.; Dyer, M.; Lafferentz, L.; Persson, M.; Peters, M. V; Hecht, S. Nano-Architectures by Covalent Assembly of Molecular Building Blocks. Nat. Nanotechnol. 2007, 2 (11), 687–691.
[2] Moreno, C.; Vilas-Varela, M.; Kretz, B.; Garcia-Lekue, A.; Costache, M. V.; Paradinas, M.; Panighel, M.; Ceballos, G.; Valenzuela, S. O.; Peña, D.; Mugarza, A. Bottom-up Synthesis of Multifunctional Nanoporous Graphene. Science. 2018, 360, 199–203
[3] Calogero, G.; Papior, N. R.; Kretz, B.; Garcia-Lekue, A.; Frederiksen, T.; Brandbyge, M. Electron Transport in Nanoporous Graphene: Probing the Talbot Effect. Nano Lett. 2018, 19 (1), 576–581
[4] Calogero, G.; Alcón, I.; Papior, N.; Jauho, A.-P.; Brandbyge, M. Quantum Interference Engineering of Nanoporous Graphene for Carbon Nanocircuitry. J. Am. Chem. Soc. 2019, 141 (33), 13081–13088

A strong driving force for future communication systems is the improvement of transceiver electronics efficiency. Carbon nanotube-based field effect transistors (CNTFETs) are promising electronic components¹, since high intrinsic speed and linearity due to their high charge carrier mobility, current carrying capability and one-dimensional charge transport are prospected^2-4. We report on design rules for the high frequency (HF) CNTFETs device architecture with improved operating speed. Therefore, the spacer between the gate and the source electrode (S-spacer) and gate to drain electrode (D-spacer) was investigated with respect to the HF performance. It will be discussed to which extent the spacer variation is balancing the device electrostatics in the CNT-metal contact area which influences the charge carrier injection and reflection by the Schottky barriers (SBs) that form between the metal covered CNT ends and the uncovered CNT channel region. Theoretically the dependence of the S-spacer and D-spacer on speed and linearity has been shown by simulations⁴ but was not yet investigated in an experimental study. Mothes et al. point out that especially within the S-spacer the fringing fields are needed to modulate the SBs, which is essential to gain highest speed and linearity. Therefore a small finite SB is essential to reduce the impact of charge carrier back-scattering from the source electrode. In contrast to FET geometry variations reported in literature, where additional gate electrodes or asymmetrical gate oxide trenches were introduced and investigated with respect to their DC performance^5,6, we report on the impact of S-spacer and D-spacer variations on the HF performance of CNTFETs. In terms of HF performance, inspiring results were obtained with symmetrical gate arrangements, using a self-aligned T-shape top gate electrode⁷ where Cao et al. realized CNTFETs with a transit frequency f_T as well as maximum frequency of oscillation f_max of up to 70 GHz at 100 nm channel length. In contrast, a buried gate helps minimizing parasitic capacitances and thus improving the HF performance further⁴. We present HF CNTFETs with a buried gate fabricated by an additional lithography step enabling to realize devices with asymmetrical S-spacer and D-spacer. Thereby we followed scalable device fabrication procedures based on a wafer-level prototyping platform allowing to fabricate sufficiently large device counts for systematic and conclusive studies. The studies exposed a speed increase, measured by the external f_T and f_max, of up to 20% for a specific asymmetrical gate electrode position in the channel. The fabricated devices show transit frequencies up to 16 GHz for 280 nm long channels with randomly oriented CNTs and a tube density of up to 30 CNTs/µm.

[1] Schroter et al., IEEE J. Electron Devices Soc. 1 (1), 9–20. 2013
[2] Dürkop et al., Nano Lett. 4 (1), 35–39 2004
[3] Baumgardneret al., Appl. Phys. Lett. 91 (5), 52107 2007
[4] Mothes et al., SISPAD 2015
[5] Lin et al., Nano Lett. 4 (5), 947–950 2004
[6] Qiu et al., ACS nano 9 (1), 969–977 2015
[7] Cao et al., Nano Res. 9 (2), 363–371 2016

Ideal graphene nanoribbons (GNRs) have been shown to exhibit extreme chirality dependence as metals or semiconductors. Therefore, the capability to precisely produce GNRs with defined chirality at the atomic level is required in order to engineer their band gap and electrical properties. It is obvious that earlier approaches have fundamental limitations for further electronic investigation. Electronics always require scalable transfer-free approaches for growing GNRs and conducting band gap engineering. Controlled fabrication of oriented GNRs embedded on hexagonal boron nitride (h-BN) has the capability to overcome the above difficulties. With proper control, the band gap and magnetic properties can be precisely engineered. Most desired features for GNRs can be automatically attained using this approach.

Here we developed a two-step growth method and successfully achieved sub-5 nm zigzag and armchair GNRs embedded in h-BN, respectively. Further transport measurements reveal that the sub-7 nm zigzag GNRs exhibit openings of the bandgap inversely proportional to their width, while narrow armchair GNRs exhibit some fluctuation in the bandgap-width relationship. Transistors made of these GNRs with large bandgaps (>0.4 eV) exhibit excellent electronic performance even at room temperature (e.g. conductance on-off ratio more than 10E5 and carrier mobility more than 0.15m2 per volt per second ). An obvious conductance peak is observed in the transfer curve of 8-10 nm-wide zigzag GNRs while it is absent in most of armchair GNRs of similar width. Magneto-transport experiments show that zigzag GNRs exhibit relatively small magneto-conductance (MC) while armchair GNRs have much higher MC than zigzag GNRs. This integrated lateral growth of edge-specific GNRs in h-BN brings more robust and orientation-controlled edges than previously demonstrated, and will enable exciting future investigations into the physics and applications of edge-specific GNRs.

I will provide an overview of the recent progress made in my group on the understanding of the electronic and vibrational properties of 1D carbon nanostructures based on a combination of theoretical and computational methods. In the first part of the presentation, I will first focus on providing a model to account for the significant screening caused by the presence of a substrate on the electronic properties of graphene nanoribbons. In the second part of the talk, I will present a broad and detailed overview of the intrinsic vibrational properties of 1D carbon nanostructures with a variety of edge configurations, sizes, and shapes. These results will be systematically compared to experiment (notably, Scanning Tunneling Spectroscopy and Raman Spectroscopy). While theory and experiment are found to usually be in excellent agreement, I will also discuss a number of unresolved issues and will suggest a roadmap to tackle them.

Using the high energy environment of laser ablation, we have synthesized several new materials composed of 1D carbon chains stabilized through interaction with small metal clusters, that we have named “Pseudocarbynes”. These new materials are expected to approach the unprecedented strength, elastic modulus, and stiffness of carbyne, which has defied isolation and characterization for over a century. Our spectroscopic characterization suggests that pseudocarbynes self-assemble from colloidal solutions into new mesomaterials that bridge the molecular and materials realms. These measurements are supported by density functional theory (DFT) calculations that highlight a strong tendency of small metal clusters to interact with the sides of the -C≡C- chains. X-ray powder diffraction and high-resolution electron microscopy demonstrate the long-range material crystallinity. I will discuss our recent experimental results aimed at extending the pseudocarbyne library to include a variety of transition metals, cluster sizes, and lengths of carbon chain. Our results suggest that pseudocarbynes have potential to become a platform for developing an array of materials with tunable electronic properties.

Graphene nanoribbons (GNRs) are quasi-1D strips of graphene that are tunable semiconductors and can exhibit excellent electronic, thermal and mechanical properties.¹ Field effect transistors (FETs) made of sub-10 nm wide GNRs with smooth, armchair edges can outperform silicon-FETs in drive current and suppress short channel effects due to ultra-thin geometry.^{2, 3} However, in order to become a viable contender to silicon based high volume manufacturing, compatibility with existing CMOS processes is of paramount importance. Although, the synthesis of GNRs has been demonstrated on Ge(001),^{4, 5} owing to the lack of availability of 300 mm Ge wafers and relatively higher cost, a promising alternative is to synthesize graphene nanoribbons on Ge virtual substrates, which can be epitaxially grown on 300 mm Si wafers.
Here we explore the concept of transfer-free growth of armchair graphene nanoribbons on a Si(001) platform by conducting the GNR synthesis on germanium virtual substrates by CVD.⁶ By tuning the thickness of the Ge epilayer, growth temperature, methane flow rate and the synthesis duration, we are able to achieve sub-10 nm wide GNRs that possess a semiconducting bandstructure and are several hundred nanometers long. Scanning tunneling microscopy reveals that these GNRs possess smooth armchair edges, which is critical to achieving high on/off ratios and drive currents. We also find that GNR growth kinetics on Ge virtual substrates is identical to Ge(001) and that Si diffusion from the bulk can be suppressed for as long as 9 h, preventing SiC formation on the surface. Furthermore, the threading dislocations, which are inevitably present in heteropitaxially mismatched systems as Ge-Si are also not found to influence GNR nucleation and/or growth.
These results indicate that all previous advances of GNR growth on Ge(001) (e.g. seeding initiated synthesis)^{7, 8} can be readily scaled to Ge virtual substrates, overcoming a substantial barrier for adoption of GNRs in novel semiconductor technologies. This demonstration also opens up the possibility of achieving nanoribbons directly on a Si surface via selective vapor phase etching of Ge.⁹

References

1. Z. X. Guo, D. Zhang and X. G. Gong, Applied Physics Letters, 2009, 95.
2. G. C. Liang, N. Neophytou, D. E. Nikonov and M. S. Lundstrom, Ieee Transactions on Electron Devices, 2007, 54, 677-682.
3. Z. S. Geng, B. Hahnlein, R. Granzner, M. Auge, A. A. Lebedev, V. Y. Davydov, M. Kittler, J. Pezoldt and F. Schwierz, Annalen Der Physik, 2017, 529.
4. B. Kiraly, A. J. Mannix, R. M. Jacobberger, B. L. Fisher, M. S. Arnold, M. C. Hersam and N. P. Guisinger, Applied Physics Letters, 2016, 108.
5. R. M. Jacobberger, B. Kiraly, M. Fortin-Deschenes, P. L. Levesque, K. M. McElhinny, G. J. Brady, R. R. Delgado, S. S. Roy, A. Mannix, M. G. Lagally, P. G. Evans, P. Desjardins, R. Martel, M. C. Hersam, N. P. Guisinger and M. S. Arnold, Nature Communications, 2015, 6.
6. V. Saraswat, Y. Yamamoto, H. J. Kim, R. M. Jacobberger, K. R. Jinkins, A. J. Way, N. P. Guisinger and M. S. Arnold, The Journal of Physical Chemistry C, 2019, 123, 18445-18454.
7. A. J. Way, R. M. Jacobberger and M. S. Arnold, Nano Letters, 2018, 18, 898-906.
8. A. J. Way, E. A. Murray, F. Goltl, V. Saraswat, R. M. Jacobberger, M. Mavrikakis and M. S. Arnold, The journal of physical chemistry letters, 2019, 4266-4272.
9. Y. Bogumilowicz, J. M. Hartmann, R. Truche, Y. Campidelli, G. Rolland and T. Billon, Semiconductor Science and Technology, 2005, 20, 127-134.

Atomically precise graphene nanoribbons (GNRs) are at the forefront of nanocarbon research and hold great promise for electronic and optoelectronic applications. Theoretical studies have shown that narrow GNRs possess interesting electronic and magnetic properties that strongly depend on the nanoribbon’s width and edge structure. GNRs with different structures can be synthesized with atomic precision and fine-tuned properties by coupling properly designed molecular precursors and planarization of the resulting polymers.
In this talk, I will discuss our recent results on solution and on-surface synthesis and characterization of atomically precise GNRs. The solution approach is based on Yamamoto coupling of pre-synthesized molecular precursors followed by cyclodehydrogenation using Scholl reaction. This approach can be applied to the synthesis of a variety of GNRs, including pristine and nitrogen-doped GNRs with the chevron structure. The on-surface approach is based on the coupling of halogenated molecular precursors on metal substrates and subsequent cyclodehydrogenation of the resulting polymers at elevated temperatures.
Both solution and on-surface approaches were used to synthesize laterally extended chevron GNRs (eGNRs). The ribbons grown on Au(111) were used for extensive scanning tunneling microscopy (STM), noncontact atomic force microscopy (nc-AFM) and scanning tunneling spectroscopy (STS) characterization. They were found to exhibit a bandgap of about 2.2 eV, which is considerably smaller than the values reported for chevron GNRs in similar studies. The solution-synthesized eGNRs were processed into uniform thin films that exhibited improved electrical conductivity compared to similar films of regular chevron GNRs. The eGNR films were employed in gas sensors that showed very high responsivity to low molecular weight alcohols compared to similar sensors based on benchmark graphitic materials, such as graphene and graphene oxide. Finally, we demonstrated the first electronic nose system based on atomically precise GNRs that could reliably recognize analytes of nearly the same chemical nature, such as methanol and ethanol. Overall, this work demonstrates the entire path from designing new atomically precise GNRs with improved properties to their bottom-up synthesis, characterization, processing and implementation in electronic devices.

Recent advances in on-surface organic synthesis have led to the development of bottom-up graphene nanoribbon (GNR) growth with atomic precision¹. Bottom-up synthesized GNRs have promising electronic properties for high performance field effect transistors (FETs) and ultra-low power devices such as tunneling FETs². However, the short length, wide band gap, and random spatial orientation of GNRs have impeded the fabrication of devices with the expected performance and switching behavior³. Here, recent progress in integrating bottom-up synthesized GNRs into devices will be presented. High on-current and high on-off ratio FETs with a nine-atom wide GNRs as the channel material were successfully fabricated⁴. Strategies to further improve FET device yield and performance with the parallel arrays of GNRs transferred from single crystal Au (788) will be discussed. Finally, a new approach for the transfer-free fabrication of GNR devices has been developed.
References
[1] Cai, J. et al., Nature 466, 470-473 (2010).
[2] Smith, S. et al., IEEE Electron Device Lett. 39, 143-146 (2018).
[3] Bennett, P. B. et al., Appl. Phys. Lett. 103, 253114 (2013).
[4] Llinas, J. P. et al., Nat. Commun. 8, 633 (2017).

I will provide an overview of the recent progress made in my group on the understanding of the electronic and vibrational properties of 1D carbon nanostructures based on a combination of theoretical and computational methods. In the first part of the presentation, I will first focus on providing a model to account for the significant screening caused by the presence of a substrate on the electronic properties of graphene nanoribbons. In the second part of the talk, I will present a broad and detailed overview of the intrinsic vibrational properties of 1D carbon nanostructures with a variety of edge configurations, sizes, and shapes. These results will be systematically compared to experiment (notably, Scanning Tunneling Spectroscopy and Raman Spectroscopy). While theory and experiment are found to usually be in excellent agreement, I will also discuss a number of unresolved issues and will suggest a roadmap to tackle them.

We have recently demonstrated the rational design and experimental realization of a graphene nanoribbon (GNR) superlattices that hosts a 1D array of symmetry-protected topological states, thus generating otherwise inaccessible electronic structure. This new class of materials can be thought of as an extended form of poly-acetylene wherein highly localized half-filled topological states replace the familiar single occupied p-orbitals along a conjugated p-system. Experimental results and first-principles calculations reveal that the frontier band structure of these GNR superlattices is defined purely by the coupling between adjacent topological interface states and can be tuned all the way from a semiconductor to a metal. This novel manifestation of 1D topological phases presents an entirely new route to band engineering in 1D materials based on precise control of their electronic topology, and is a promising new platform for future studies of 1D quantum spin physics.

Graphene can spontaneously develop intrinsic paramagnetism. Crucial examples are the magnetization of zig-zag edges in graphene, or the emergence of paramagnetism in open shell graphene nanostructures. I will show that graphene nanoribbons (GNR), fabricated with atomic precision on a metal surface exhibit fingerprints of π-paramagnetism on a metal surface, which can be detected and spatially localized using low temperature scanning tunneling [1].

Single electron spins emerge localized at certain zigzag sites of the carbon backbone. Their presence could be detected and mapped by spatially resolving the zero-energy resonance due to the Kondo effect. We found that near-by spins are coupled into a singlet ground state and quantify their exchange interaction via singlet-triplet inelastic electron excitations. Theoretical simulations picture how electron correlations result in spin-polarized radical states with the experimentally observed spatial distributions. Extra hydrogen atoms bound to radical sites quench their magnetic moment and switch the spin of the nanostructure in half-integer amounts.

I will also review other methods for activating magnetic ground states in graphene. For example, spin states can be created on a ribbon simply by substitutional doping or by incorporating magnetic species into a ribbon using on-surface synthesis routes (see included image of a Fe porphyrin contacted to chiral nanoribbons). In this last case, we proved that the molecular spin survives in the ribbon by using spin-excitation inelastic spectroscopy [2]. By proper selecting the position of the halogen functionalization, we fabricated linear GNR-FeTPP-GNR structures and performed electronic transport measurements [3], detecting spin-excitation fingerprints in transport mode.

J. Li, S. Sanz, M. Corso, D.J. Choi, D. Peña, T. Frederiksen, J.I. Pascual, “Single Spin Localization and Manipulation in Graphene Open-Shell Nanostructures”, Nature Communications 10, 200 (2019).
J. Li, N. Merino-Díez, E. Carbonell-Sanromà, M. Vilas-Varela, D. G. de Oteyza, D. Peña, M. Corso, and J.I. Pascual, “Survival of spin state in magnetic porphyrins contacted by graphene nanoribbons”
Science Advances 4, eaaq0582 (2018)
J. Li, N. Friedrich, N. Merino-Díez, D. G. de Oteyza, D. Peña, D. Jacob, and J.I. Pascual, “Electrically Addressing the Spin of a Magnetic Porphyrin through Covalently Connected Graphene Electrodes”, Nano Letters 19, 3288 (2019).

In contrast to zero-bandgap graphene, structurally confined graphene nanoribbons (GNRs) features open bandgaps due to the quantum confinement effect, which render them highly promising for nanoelectronic and optoelectronic applications.^[1] The properties of the GNRs are determined by their chemical structures, in particular the width and the edge configuration, which makes their precise structural control essential. Whereas such precision cannot be achieved by “top-down” fabrication methods such as “cutting” of graphene, bottom-up chemical synthesis can afford atomically precise GNRs with various structures.^[1,2] The synthesis is carried out through polymerization of a tailor-made monomer leading to a polymer precursor, which can be “planarized” and “graphitized” into GNRs by oxidative cyclodehydrogenation. While it is challenging to obtain very long GNRs through transition-metal-catalyzed or metal-surface-assisted polymerization, we have demonstrated solution synthesis of GNRs with average lengths larger than ~600 nm through AB-type Diels-Alder polymerization.^[2,3] Solubilizing alkyl chains can be attached on the GNR edges through the solution synthesis, making them dispersible in organic solvents and thus allowing for the liquid-phase processing for the device integration.^[3,4] By modifying the monomer structure, laterally extended GNRs with the width of up to ~2 nm could be obtained, revealing the lowering of the optical bandgap down to ~1.2 eV.^[2,5] We have also achieved edge functionalization of the GNRs by introducing a bromo group on the monomer. For example, introduction of perylene monoimide units led modulation of the self-assembly behavior, providing intriguing “rectangular” networks of GNRs.^[6] Moreover, introduction of spin-bearing organic radicals on the GNR edges allowed for spin injection to the aromatic core, inducing a delocalized magnetic state, which might have potential for application in the quantum computers.^[7] These results form a fundamental basis for applications of such GNRs in the future nanotechnology.

References
[1] X.-Y. Wang, A. Narita, K. Müllen, Nat. Rev. Chem. 2017, 2, 0100.
[2] A. Narita, Z. Chen, Q. Chen, K. Müllen, Chem. Sci. 2019, 10, 964.
[3] A. Narita, X. Feng, Y. Hernandez, S. A. Jensen, M. Bonn, H. Yang, I. A. Verzhbitskiy, C. Casiraghi, M. R. Hansen, A. H. R. Koch, G. Fytas, O. Ivasenko, B. Li, K. S. Mali, T. Balandina, S. Mahesh, S. De Feyter, K. Müllen, Nature Chem. 2014, 6, 126.
[4] T. Preis, C. Kick, A. Lex, D. Weiss, J. Eroms, A. Narita, Y. Hu, K. Müllen, K. Watanabe, T. Taniguchi, Appl. Phys. Lett. 2019, 114, 173101.
[5] Y. Hu, P. Xie, M. De Corato, A. Ruini, S. Zhao, F. Meggendorfer, L. A. Straasø, L. Rondin, P. Simon, J. Li, J. J. Finley, M. R. Hansen, J.-S. Lauret, E. Molinari, X. Feng, J. V. Barth, C.-A. Palma, D. Prezzi, K. Müllen, A. Narita, J. Am. Chem. Soc. 2018, 140, 7803.
[6] A. Keerthi, B. Radha, D. Rizzo, H. Lu, V. Diez Cabanes, I. C.-Y. Hou, D. Beljonne, J. Cornil, C. Casiraghi, M. Baumgarten, K. Müllen, A. Narita, J. Am. Chem. Soc. 2017, 139, 16454.
[7] M. Slota, A. Keerthi, W. K. Myers, E. Tretyakov, M. Baumgarten, A. Ardavan, H. Sadeghi, C. J. Lambert, A. Narita, K. Müllen, L. Bogani,, Nature, 2018, 557, 691.

A unique feature of single-wall carbon nanotube (SWCNT) material is its structure diversity: within the same rolled-up graphene sheet structure framework one finds SWCNTs of different electronic bandgaps: metals with zero bandgap, quasi-metals with meV bandgap and semiconductors with ~ 1 eV bandgap. However, making use of SWCNT structure diversity requires a method to obtain SWCNTs of defined structures. In this contribution, we present our current state of DNA-based sorting of SWCNTs. In particular, we show our ability to put any DNA sequence on any SWCNT species, enabling DNA-based assembly of high-density SWCNT devices. In the end, we discuss the design of a device we call Photon Perceptron Array (PPA). PPA is an array of micron-sized spectrometer units that can measure the spectrum of incoming photon in the wavelength range of 200 nm to 2000 nm (and possibly going all the way to THz), with a spatial resolution comparable to the size of a single spectrometer unit. Each spectrometer unit is composed of a set of distinct photodetectors made of SWCNTs of defined atomic structures, placed in parallel with tube-to-tube separations controlled at nanometer precision by DNA origami technology. A high-density array of identical spectrometers can be constructed by placing DNA origami/carbon nanotube complexes in photolithographically-defined locations. PPA may be used in spectral imaging in many fields of science and technology.

The remarkable flexibility, stable chemical structure, and extraordinary thermal, electrical, and optical properties of carbon nanotubes (CNTs) are promising for a variety of applications in flexible and/or high-temperature electronics, optoelectronics, and thermoelectrics, including wearables, refractory photonics, and waste heat harvesting. However, the long-standing problem in the preparation of CNT ensembles is to maintain the extraordinary properties of individual CNTs on a macroscopic scale; the polydispersity and randomness remain two main challenges. Here, we will discuss three ways of creating wafer-scale aligned CNTs: direct growth of aligned CNTs by chemical vapor deposition, production of ultrahigh-conductivity CNT fibers through solution spinning and coating, and spontaneous formation of wafer-scale aligned CNT films via controlled vacuum filtration. We will then describe flexible and high-temperature applications of these materials, such as flexible CNT broadband detectors, spectrally selective thermal emitters, and thermoelectric devices.

Plasmon resonance in carbon nanotubes is an exciting resource for confining light to the nanometer scale [1]. In this talk, I will discuss how dense films of carbon nanotubes have emergent optical properties related to these resonances. We assemble our films with a simple vacuum filtration technique and find that optimal assembly leads to the nanotubes crystallizing into monolithic, two-dimensionally ordered films. Using Mueller-matrix ellipsometry, we then characterize the films’ optical properties and find a broadband, electrostatically tunable hyperbolic region in the mid-infrared [2]. Another fascinating property of dense nanotube films is that excitons in the nanotubes can couple to the nanotubes’ own plasmon resonances – that is, cavity quantum electrodynamics (cQED) without an external cavity [3]. In fact, this coupling strength can reach the ultrastrong light-matter interaction regime. Technologically, crystallized nanotube films could provide a materials-driven pathway from fundamental concepts in cQED to advanced telecommunications hardware, including tunable thermal emitters, super-resolution imaging systems, and active, nonlinear optical devices like modulators and nanolasers.

[1] Falk et al, “Coherent plasmon and phonon-plasmon resonances in carbon nanotubes,” Phys. Rev. Lett. 118, 257401 (2017).
[2] Roberts et al, “Tunable Hyperbolic Metamaterials Based on Self-Assembled Carbon Nanotubes,” Nano Lett. 19, 3131 (2019).
[3] Ho et al, “Intrinsically ultrastrong plasmon–exciton interactions in crystallized films of carbon nanotubes,” Proc. Nat. Acad. Sci. 115, 12662 (2018).

At this exact moment when future applications are demanding massive improvements in computing performance, conventional approaches to improving computing are becoming increasingly challenging. Nanosystems – leveraging unique properties of emerging nanotechnologies to realize new systems and architectures – promise to enable the next wave of computing gains for next-generation applications as well as provide the means for developing rich new functionalities that lie beyond the scope of traditional computing today. Yet despite this promise, imperfections and variations inherent in beyond-silicon emerging nanotechnologies have made realizing large-scale integrated systems infeasible. In this talk, I describe how by combining advances across the entire stack – from nanomaterial synthesis to new processing techniques, circuit design methodologies, and architectures – these challenges can be overcome to realize the first large-scale and complex electronic systems leveraging beyond-silicon emerging nanotechnologies. As a case-study, I will describe the journey of transforming carbon nanotubes from an interesting scientific nanomaterial into the foundation for realizing the first beyond-silicon microprocessor built entirely from carbon nanotube transistors – RV16X-NANO, a 16-bit microprocessor based on the RISC-V instruction set. I will conclude by describing how emerging nanotechnologies are beginning to be transferred from the “lab” to the “fab,” promising a new and exciting age for nanoelectronics and computing.

Carbon nanotube (CNT)-based electronics has been considered one of the most promising candidates to extend Si complementary metal-oxide-semiconductor (CMOS) technology. In particular CNTs have been investigated for various electronic and optoelec-tronic device applications, such as sub-10nm CMOS devices which outperform that of state-of-the-art Si based CMOS devices in both speed and power consumption [1-5], as well as idea material for monolithic optoelectronic integration with complementary MOS-compatible signal processing circuit [6-8].
Prototype device studies on individual CNTs revealed that CNT based devices have the potential to outperform Si CMOS technology in both performance and power consumption. With a well-designed device structure and in combination with graphene, we showed that high-performance top-gated CNT FETs with a gate length of 5 nm can be fabricated. Significant progress has also been made in fabricating carbon nanotube low-power devices. An efficient way to reduce the power is to lower the supply voltage VDD, but this voltage is restricted by the 60 millivolts per decade thermi-onic limit of subthreshold swing (SS) in FETs. A Dirac source (DS) with a much narrower electron density distribution around the Fermi level than that of conventional FETs was recently proposed and demonstrated using CNT to reduce SS [2].

References
[1] C. Qiu et al., Scaling carbon nanotube complementary transistors to 5-nm gate lengths, Science 255 (2017) 271.
[2] C. Qiu et al., Dirac source field-effect transistors as energy-efficient and high-performance electronic switches, Science 10.1126/science.aap9195 (2018).
[3] D.L. Zhong et al., Gigahertz integrated circuits based on carbon nanotube films, Nature Electronics 1 (2018) 40.
[4] L.-M. Peng et al., Carbon-based electronics: A new stage for flexible nanotube devices, Nature Electronics 1 (2018) 158.
[5] L. Xiang et al., Highly uniform carbon-based low-power electronics with nio-integration capability in a transferrable form, Nature Electronics 1 (2018) 273.
[6] L.J. Yang et al., Extremely efficient photovoltage multiplication in carbon nanotubes, Nature Photonics 5 (2011) 672.
[7] Y. Liu et al., Carbon nanotube-based three-dimensional monolithic optoelectronic integrated system, Nature Communication 8 (2017) 15649.
[8] Y. Liu et al., Electrically-driven monolithic plasmonic interconnect circuits fabricated on a subwavelength scale from carbon nanotubes, Science Advances 3 (2017) e170145.

The National Research Council (NRC) is Canada’s premier organization for research and technology development. NRC’s Printable Electronics (PE) program develops materials and inks for additive manufacturing in order to enable a sustainable PE sector in Canada. Progress has been made over the last 15 years, yet many challenges remain at the materials, fabrication and integration level, limiting performance and commercialization. This presentation covers our progress in high purity semiconducting single-walled carbon nanotube (sc-SWCNT) enrichment and transistor fabrication via solution based processes as well as conductive molecular ink development ^[1-11]. I will highlight our recent progress in understanding sc-SWCNT enrichment using conjugated polymers, with special consideration given to the effects of solvent parameters and doping on the mechanism and yield/purity of the final product ^[2-4]. Removal of the wrapping polymer using a dry process on-wafer will also be described ^[5]. Developments in conductive molecular inks and progress towards the commercialization of new applications will be described such as in-mold electronics ^[6-8]. Challenges and advances associated with using polymer-based dielectrics and encapsulants will be discussed as well ^[9-10]. Such transistor packages have enabled the realization of fully inkjet-printed transistors as a result of the excellent electrical properties of sc-SWCNTs ^[2,11]. A demonstration of a fully additive process to make TFT backplanes via R2R printing will also be highlighted. The development of new functional nanomaterials and their successful integration into devices will enable additive manufacturing of TFT arrays and in-mold electronics.

[1] J. Lefebvre et al. Acc. Chem. Res. 50, 2479 (2017).
[2] J. Ouyang et al. ACS Nano 12, 1910 (2018).
[3] J. Ding et al. J. Phys. Chem. C 120, 21946 (2016).
[4] Z. Li et al. ACS Omega 3, 3413 (2018).
[5] Z. Li et al. Adv. Funct. Mater. 28, 1705568 (2018).
[6] A. J. Kell et al. ACS Appl. Mater. Interfaces 9, 17226 (2017).
[7] C. Paquet et al. Nanoscale 10, 17226 (2018).
[8] B. Deore et al. ACS Appl. Mater. Interfaces 11, 38880 (2019).
[9] J. Lefebvre et al. Appl. Phys. Lett. 107, 243301 (2015).
[10] Lapointe et al. Appl. Mater. Interfaces 11, 36027 (2019).
[11] C. Homenick et al. ACS Appl. Mater. Interfaces 8, 27900, (2016)