(Invited) Optimizing Material Systems for All-Inkjet-Printed Organic Thin-Film Transistors

Organic transistors have the advantages of mechanical flexibility and low-temperature printability, thus having attracted considerable attention. However, printed organic transistors typically underperform electrical characteristics and stability, which are usually worse in all-printed devices. Here, we demonstrate all-inkjet-printed organic transistors with a steep subthreshold slope of 60.4 mV/decade and good stability under bias stress, which were achieved by reducing trap density in the fabricated devices. The transistors showed good uniformity, with an average subthreshold slope of 62.4±14.0 mV/decade and an average threshold voltage of -0.042±0.026 V, which were superior to other reported organic transistors and thin-film device technologies. For all-inkjet-printed organic transistors, it is important to use a proper device structure to reduce device trap density. The traps in organic transistors can be found in semiconductor bulk materials and semiconductor/dielectric interfaces. We used small molecule materials for semiconductors that could generate large crystals and used polymers as the smooth bases for semiconductor/dielectric interfaces that could reduce interfacial traps. Given these considerations, the bottom-gate bottom-contact device architecture was chosen. The fabricated devices demonstrated a low density of states in the deep states of 6.59×1014 cm- 3 eV- 1. As a result, the devices showed such a steep subthreshold slope that approached the theoretical limit, and as well as a close-to-zero threshold voltage. Besides device architecture, we also present the significance of dielectric material selections to enhance device stability. It has been reported that using a fluoropolymer as the dielectric could significantly improve the stability of organic transistors. However, the issue of fluoropolymers is that they are too hydrophobic to print other functional materials on top of them. Thus, they cannot be used as a dielectric material for all-inkjet-printed transistors, since the dielectric is typically sandwiched by the gate and source/drain/semiconductor and there is an inevitable layer on top of the dielectric. Bipolar materials could be used to solve the printability issue, but they induce device instability. We found that a Lewis-acid monopolar polymer could solve these two issues at the same time. Due to the monopolar behavior, the polymer does not attract bipolar water molecules; due to its nature of being a Lewis acid, it wets common organic solvents that are mostly Lewis bases (such as ethanol, toluene, anisole). We compared two types of fabricated devices using poly-(4vinyl phenol) (bipolar) and polyvinyl cinnamate (Lewis-acid monopolar) as dielectrics. The two types of transistors demonstrated very similar static device characteristics, while the devices with polyvinyl cinnamate as the dielectric showed much superior bias-stress stability. Together with the advantage of trap density reduction, the fabricated all-inkjet-printed organic transistors demonstrated negligible changes in their electrical characteristics for one-year storage in the ambient (with a threshold voltage shift under 1 mV) or for one-hour bias stress (with a threshold voltage shift under 30 mV). Based on these transistors, we demonstrate an ultra-low-power amplifier with a high voltage gain for electrophysiology monitoring. The steep subthreshold slope gave the advantages of high intrinsic gain (> 1000 V/V) and high transconductance efficiency (38.2 S/A) in the subthreshold regime, where the devices operate with a low current at a sub-nanoampere level. As a result, a common-source amplifier showed a high voltage gain of 260 V/V, and more importantly ultra-low power consumption of < 1 nanowatt. We implemented this amplifier to track eye movements by monitoring electrooculogram signals. These signals were originally below 1 mV and amplified to around 0.3 V. This amplifier has the potential for subtle eyeball movement analysis and virtual and augmented reality.