Semiconducting single-walled carbon nanotubes (s-SWCNTs) are fascinating materials for future electronic and optical applications. Conjugated polymer wrapping is one of the most promising methods for mass production of high purity s-SWCNTs. However, its chiral selectivity is relatively inferior to other s-SWCNT production methods. In this paper, the chiral selectivity of two polymers, poly[(9,9-dioctylfluorenyl-2,7-diyl)-alt-co-(6,6′-{2,2′-bipyridine})] (PFO-BPy) and poly[9-(1-octylonoyl)-9H-carbazole-2,7-diyl] (PCz), which are representatives of widely used polyfluorene and polycarbazole families, respectively, were comparatively studied. Both polymers exhibited high selectivity for a subset of existing chiral species in each of the commercially available raw SWCNT materials (CoMoCAT, HiPco, and arc-discharge) which cover a diameter range of 0.6–1.8 nm. Less chiral species were selected by PFO-BPy from small diameter (< 1 nm) raw SWCNT materials, while more from large diameter (> 1.2 nm) raw materials. High chiral purity (6, 5) (> 99%) and (7, 5) (> 75%) solutions were extracted by PFO-BPy and PCz from CoMoCAT materials, respectively. The different chiral angle and diameter selections for different raw materials by both polymers were ascribed to their different geometrical structures and related polymer-tube interactions. Our work provides indispensable information for better understanding the mechanism of polymer wrapping method and improving extraction of single chirality s-SWCNTs.

Brain-inspired neuromorphic computing is expected for breaking through the bottleneck of the computer of conventional von Neumann architecture. To this end, the first step is to mimic functions of biological neurons and synapses by electronic devices. In this paper, synaptic transistors were fabricated by using carbon nanotube (CNT) thin films and interface charge trapping effects were confirmed to dominate the weight update of the synaptic transistors. Large synaptic weight update was realized due to the high sensitivity of the CNTs to the trapped charges in vicinity. Basic synaptic functions including inhibitory post-synaptic current (IPSC), excitatory post-synaptic current (EPSC), spike-timing-dependent plasticity (STDP), and paired-pulse facilitation (PPF) were mimicked. Large dynamic range of STDP (> 2, 180) and low power consumption per spike (~ 0.7 pJ) were achieved. By taking advantage of the long retention time of the trapped charges and uniform device-to-device performance, long-term image memory behavior of neural network was successfully imitated in a CNT synaptic transistor array.

Semiconducting single-walled carbon nanotubes (s-SWCNTs) are the foundation of CNT-based electronics and optoelectronics. For practical applications, s-SWCNTs should be produced with high purity, high structural quality, low cost, and high yield. Currently conjugated polymer wrapping method shows great potential to fulfill these requirements due to its advantages of simple operation process, high purity separation, and easy scaling-up. However, only a small portion of both CNTs and polymers go into the final solution, and most of them are discarded after a single use, resulting in high cost and low yield. In this paper, we introduce a closed-loop recycling strategy, in which raw materials (CNTs and polymers) and solvents were all recycled and reused for multiple separation cycles. In each cycle, high-purity (> 99.9%) s-SWCNTs were obtained with no significant change of structural quality. After 7 times of recycling and separation, the material cost was reduced to ~ 1% in comparison with commercially available products, and total yield was increased to 36% in comparison with 2%–5% for single cycle separation. Our proposed closed-loop recycling strategy paves the way for low-cost and high-yield mass production of high-quality s-SWCNTs.

Large area, highly uniform, and density controllable carbon nanotube (CNT) films, either well-aligned or random network, are required for practical application of CNT-based electronics. Mass production methods for such CNT films and corresponding quality metrology, which are critical for pushing the CNT-based transistor technology to manufacturing, should be developed in advance. Much progress has been made on fabrication of CNT films; however, there still lacks a metrology for thoroughly quantifying their quality until now. In this paper, through comparing study of CNT films fabricated by dip-coating (DC) and direct deposition (DD) methods, local anisotropy in the film is revealed to impact the performance uniformity of devices so fabricated in a spatial scale dependent manner. The anisotropy effect should be taken into account for the quality characterization of CNT films, which was not noticed in previous studies. Based on these findings, we propose a four-parameter metrology to quantify the overall quality of the CNT films, which includes the local tube density (D_L), global density uniformity (C_v), local degree of order (O_L), and the relative tube proportion in a certain orientation (P_θ) at a location. The four-parameter characterization and corresponding device performance confirm DC films are superior to DD films for practical application. The four-parameter metrology is not only powerful for overall quality evaluation of CNT films, but also able to predict the fluctuation of devices’ performance. Therefore, this material metrology is important for devices and circuits design and valuable for pushing the CNT-based transistor technology forward.

Carbon nanotube thin film transistors (CNT-TFTs) are a potential TFT technology for future high-performance macroelectronics. Practical application of CNT-TFTs requires the production of large-area, highly uniform, density-controllable, repeatable, and high-throughput CNT thin films. In this study, CNT films were fabricated on 4-inch Si wafers and 2.5^th generation (G2.5) backplane glasses (370 mm × 470 mm) by dip coating using a chloroform-dispersed high-purity semiconducting CNT solution. The CNT density was controlled by the solution concentration and coating times, but was almost independent of the substrate lifting speed (1–450 mm·min^-1), which enables high-throughput CNT thin film production. We developed an image processing software to efficiently characterize the density and uniformity of the large-area CNT films. Using the software, we confirmed that the CNT films are highly uniform with coefficients of variance (C_V) < 10% on 4-inch Si wafers and ~ 13.8% on G2.5 backplane glasses. High-performance CNT-TFTs with a mobility of 45–55 cm²·V^-1·s^-1 were obtained using the fabricated CNT films with a high-performance uniformity (C_V ≈ 11%–13%) on a 4-inch wafer. To our knowledge, this is the first fabrication and detailed characterization of such large-area, high-purity, semiconducting CNT films for TFT applications, which is a significant step toward manufacturing CNT-TFTs.