Wafer-scalable, aligned carbon nanotube transistors operating at frequencies of over 100 GHz
The emerging carbon nanotube high-frequency transistor technology has the potential to be a star contributor in our increasingly wireless future. Here we present breakthrough results that already surpasses the incumbent technology, RF-CMOS, in key metrics.
Beginning in the early-2000s, carbon nanotubes (CNTs) rapidly grew a reputation as a leading emerging material for electronics due to its impressive single-CNT transport characteristics. Quasi-ballistic transport has been observed for tube lengths of 300 nm as a result of the reduced scattering degrees of freedom associated with being a one-dimensional material. Scaling single CNT device performance to that of a densely aligned array will result in current densities and transconductance values exceeding that of the incumbent technologies. Furthermore, 1D materials such as CNTs are expected to have superior linearity meaning it can reproduce and amplify an input signal without distortion over a wider range of input power levels. In today’s increasingly ubiquitous wireless environment, high linearity devices are at a premium as it allows for savings in power, frequency-bandwidth, and size.
Although much of the attention for CNTs had been and still remains with its digital application as a Si CMOS replacement, in 2009 we suggested in Nature Nanotechnology that a “more realistic point of insertion into the market may be high-performance analogue radiofrequency (RF) devices, where manufacturing tolerances are relaxed and the performance metrics required for commercial systems are more suitable to the materials and device properties of nanotubes.” However, even under these comparatively “relaxed” requirements, challenges remained, particularly with scaling from a single CNT device to that of a massively arrayed set of thousands of semiconductor CNTs all operating together in parallel. This challenge dogged the research community for over a decade and interest gradually waned as those early overhyped expectations were not met.
Fast forward to today, we report in Nature Electronics a wafer scalable platform that has overcome these twin engineering challenges (i.e. high purity semiconducting CNTs that are assembled in densely aligned arrays) and have delivered device performance that exceeds the incumbent technology RF-CMOS in both cutoff frequency (fT) and linearity (OIP3/Pdc) metrics. As further engineering improvements are introduced, the ultimate performance is expected to exceed that of the top-tier RF device technologies currently available while being manufacturable using a simpler process and at a lower material input cost. This work sets a new standard and expectation for what is achievable with carbon nanotube RF transistor technology and offers a path forward as a high-performance RF solution that is compatible with CMOS integration without the performance trade-off typically associated with a full RF-CMOS implementation.
For some of us, this journey began in the early 2000s researching carbon nanotubes as part of our graduate work back when nanotubes were all the rage. Although generally out of favor by 2014 -overtaken by the popularity of graphene and 2D materials- the underlying technology had quietly advanced sufficiently that a market opportunity was identified to commercialize carbon nanotube transistor RF technology for high-frequency 5G & mmWave applications. With that, our company Carbonics Inc. was formed with the backing of private capital and a multidisciplinary team of eight. Today, incumbent semiconductor industry and DoD contractors & component suppliers are trending towards higher performance and higher levels of CMOS integration. This is driven by the push towards mmWave operating frequencies as well as increasingly complex RF-front-ends that necessitate more CMOS control circuits for advanced functionality. Examples of this market demand are DARPA’s multi-million dollar Dynamic Range-enhanced Electronic Materials (DREaM) program where they look to push the boundaries of device linearity. Also, the rollout of 5G mmWave communication systems around the world is driving considerable innovation in the field. As the CNT RF technology matures and scales, we see it finding its place within the RF technology suite (RF CMOS, GaAs, GaN, SiGe) that powers our wireless world.
These results were recently published in Nature Electronics: "Wafer-scalable, aligned carbon nanotube transistors operating at frequencies of over 100 GHz." 2, pages 530–539 (2019) https://www.nature.com/articles/s41928-019-0326-y