Printed Low-dimensional Materials for Thin-film Electronic and Optoelectronic Devices

Abstract

The Internet of Things (IoT) is ushering in a future in which everyday objects, from household appliances to healthcare devices, are equipped with electronics that enable the automatic collection, sharing, and analysis of data. To meet the demand for IoT, a broad range of devices are needed, including thin-film transistors (TFTs) for driving custom display interfaces, supporting flexible electronics platforms, and emerging neuromorphic devices that enable more efficient computing architectures. Traditional silicon transistor technology falls short of meeting these various needs due to inherent limitations such as poor mechanical flexibility, high customization costs, and limited material compatibility. These limitations have spurred extensive research over the past two decades into novel additive manufacturing techniques and alternative electronic materials. This dissertation presents three pieces of work, centered around low-cost printed, nanomaterial-based electronic devices. In the first work, we explored an emerging conductor material for its compatibility with low-cost fabrication techniques, such as aerosol jet printing, and its potential as the contact material for printed carbon nanotube transistors. Ti3C2Tx MXenes have garnered significant attention for electronics applications due to their facile synthesis, tunable properties, and exceptional optical and electrical characteristics. Their stable aqueous suspension without additional surfactants enables compatibility of MXenes with various low-cost, additive manufacturing techniques, including spin coating, spraying, and direct-write printing. In this work, we developed an effective printing procedure, explored the printability of a water-based MXene ink on surfaces with different wettability, and characterized the uniformity and sheet resistance of printed thin films on both hydrophilic and hydrophobic substrates. Furthermore, we presented MXene-contacted carbon nanotube thin-film transistors (CNT-TFTs) with various device geometries, and benchmarked MXene-contacted devices with devices contacted by common contact materials such as graphene and silver nanowires. In the second work, we investigated the hysteresis effect - a physical phenomenon that commonly presents in transistor devices and degrades the device performance. Hysteresis, defined as a change in the threshold voltage (VTH) between the forward and backward sweep, is an undesired effect in transistor performance that can lead to poor reproducibility and unreliable circuit operation. Despite extensive research, the various origins and governing factors of hysteresis remain insufficiently understood, especially in emerging technologies employing low-dimensional, III-V, or organic materials. In this work, we systematically investigated the correlation between contact materials and hysteresis using printed carbon nanotube thin-film transistors (CNT TFTs) featuring silver nanowire (AgNW), MXene (Ti₃C₂Tx), and graphene Schottky-barrier contacts. When devices from each of the three contact metals were tested using a uniform gate sweep range, we observed a difference in hysteresis based on the contact materials. However, adjusting the sweep range to maintain a fixed overdrive voltage (VOV) caused the differences in hysteresis to largely disappear, revealing that the initial observation of contact-dependent hysteresis is closely tied to the variations in VTH among devices with different contacts. This was verified by studying CNT-TFTs with asymmetrical source/drain contacts and through analysis of devices at low temperature. These findings provide insight into the hysteresis effect and highlight the importance of overdrive voltage calibration for accurate hysteresis analysis. Finally, we employed a quantum dot/CNT hybrid structure and fabricated an optical neuromorphic transistor. Optical synaptic transistors have emerged as promising building blocks for neuromorphic hardware owing to their intrinsic parallelism, low-power processing of optical information, and compatibility with artificial intelligence architectures. While most reported optical synapses exhibit positive photoconductivity (PPC), devices demonstrating negative photoconductivity (NPC) remain comparatively rare due to the additional material and interface engineering typically required to realize inhibitory photoresponse. We presented a fully printed hybrid optical synaptic transistor that achieves robust NPC through the integration of solution-processed CdSe/ZnS quantum dots (QDs) with carbon-nanotube thin-film transistors (CNT-TFTs). Electrical pulses served as excitatory stimuli that potentiate the device, whereas optical illumination induced inhibitory modulation via photogating effects arising from trapped photogenerated holes. The resulting synaptic behavior was highly tunable with respect to both electrical pulse parameters and incident light intensity, enabling transitions between short- and long-term plasticity. Furthermore, we demonstrated the practical utility of this device as a self-adaptive display-control element capable of autonomously adjusting its conductance under varying ambient illumination. By applying a 4-bit electrical pulse sequence, the device accessed 16 discrete conductance states in both dark and illuminated conditions, allowing fine, user-defined brightness adjustment atop automatic light-level adaptation. These results underscored the potential of printed optoelectronic synapses as multifunctional components for adaptive electronics, neuromorphic sensing, and smart human–machine interfaces. The achievements of this dissertation present a step towards the development of low-cost printed electronics. Under the growing demand for computing capacities, the advancement of printed electronics requires the discovery of new materials, understandings of fundamental device performance, and the realization of novel devices. Combined, the contributions outlined in this dissertation make significant progress towards that goal.