Vacuum-free and solvent-free fabrication of organic field-effect transistors
Organic field-effect transistors (OFETs), which contain semiconductor layers, are based on small molecules, oligomers, and polymers.1 They can be used as switches and/or signal-processing elements for many practical applications, e.g., for driving circuits in active matrix displays, or for physical biological, and medical sensors, actuators, and radio-frequency identification tags. In the past, two different processes—vacuum vapor deposition and solution processing—have mainly been used to fabricate the organic semiconductor layers within OFETs. There are problems associated with both these processes, however, and there is a need to find a better fabrication technique.
Vacuum vapor deposition is a costly process because of the long processing times it involves. In addition, during this process, the majority of the expensive organic materials are deposited on the chamber rather than on the substrate itself. Large-scale production of vacuum vapor deposition is also difficult. In some recent studies, alternative methodologies for the production of OFET semiconductor layers have been considered. These include solution processes such as spin coating, ink-jet printing, and blade coating.2–4 It is thought that solution processing can significantly reduce the manufacturing costs because it overcomes the problems associated with vacuum vapor deposition. In solution processing, however, organic materials need to be dissolved within organic solvents. Yet, it is preferable not to use harmful organic solvents (particularly halogenated solvents) for industrial manufacturing.
We have recently demonstrated that cold and hot isostatic pressing (CIP and HIP, respectively) can be used as novel methods for fabrication of vacuum-and-solvent-free organic semiconductor layers for OFETs.5 CIP and HIP techniques are commonly used to compress and mold metal, ceramic, plastic, and composite powders into defined forms.6–8 In our work, however, we use CIP and HIP to compress an organic powder of 2,7-dioctylbenzothieno[3,2-b]benzothiophene (C8-BTBT) directly onto a silicon substrate with gold source-drain electrodes (see Figure 1).
We find that there a number of powder particles, as well as gaps between the powder particles, in our uncompressed powder sample. After CIP at 200MPa—see Figure 2(a)—the individual particles are less visible and the gaps have been crushed. Furthermore, our HIP-compressed powder—see Figure 2(b)—has a more continuous surface, with few particle boundaries. This indicates that HIP, which combines pressurization and heating, is better for gap compression. The uncompressed C8-BTBT powder does not carry a current because of the spatial gaps that are present between the powder particles. Both our CIP- and HIP-compressed powders, however, exhibit typical p-type OFET characteristics (see Figure 3).5 The hole mobility, threshold voltage, and current on/off ratio for our CIP-compressed powder are (1.6±0.4)×10−2cm2v−1s−1, −39±7V, and (1.3±1.8)×108, respectively. The same parameter values for our HIP-compressed powder are 0.22±0.07cm2v−1s−1, −65±3V, and (4.2±4.5)×108. We attribute the higher hole mobility of the HIP-compressed powder to an improvement in the continuity between the powder particles and to the growth of small crystallites into larger ones.5
There is a downside, however, to our compressed powders. We find that it can be difficult to control their molecular orientation. Carrier mobility in the direction parallel to the substrate plane can be enhanced when molecules—which are vertically oriented along their molecular axes—are stacked parallel to one another on the substrate surface. This is caused by the enhanced π-coupling that occurs between neighboring molecules. In contrast, random crystal axis directions give rise to low carrier mobility in the parallel direction. The hole mobility of our HIP-compressed powder (i.e., with random crystal-axis directions) is therefore about an order of magnitude lower than that of a vacuum-deposited C8-BTBT film (with vertically oriented crystal-axis directions).5
The densities of vacuum-deposited and solution-processed organic films are generally lower than those of single crystals. Organic films that are fabricated with these methods therefore possess a number of microscopic and macroscopic gaps. These gaps impede carrier injection and transport because the carrier movement must proceed along indirect routes. We have therefore attempted to remove the gaps within organic films with the use CIP and HIP. We thus aim to enhance the operational stability and electrical characteristics of the films. From our initial results, we find that CIP increases the hole mobility of a vacuum-deposited metal-free phthalocyanine layer by about 2000 times. This is also accompanied by a decrease in the thickness of the film.9 With a modified CIP technique, we are also able to incorporate a very high concentration of oxygen (on the order of 4.4×1020cm−3) into the organic films.10 In addition we have used HIP to improve the morphology and crystallinity of perovskite, and thereby increase the efficiency of organic-inorganic perovskite solar cells by about 1.5 times.11
We have demonstrated a new technique for the fabrication of OFET semiconductor layers. We have used CIP and HIP (at 200MPa) to compress C8-BTBT powder into films and thus reduce the gaps between powder particles, the substrate, and the electrodes. With HIP, we achieve a hole mobility of 0.22cm2v−1s−1, which is close to the value for amorphous silicon. In addition, with CIP, we have demonstrated vacuum-free, solvent-free, and heating-free fabrication of organic semiconductor layers for the first time. By using our CIP and HIP approach, we avoid the need for costly vacuum deposition systems and harmful organic solvents. Our methodology is therefore applicable for the future manufacture of low-cost OFETs. In the next steps of our research, we aim to control molecular orientation and improve the continuity between the powder particles so that we can further enhance the carrier mobilities and stress stability of our compressed powder transistors.
This work was supported through the Adachi Molecular Exciton Engineering project of the Japan Science and Technology Agency's Exploratory Research for Advanced Technology (ERATO) program.
Toshinori Matsushima obtained his PhD in engineering from Kyushu University in 2005. He was then a postdoctoral fellow at several universities until 2008 and subsequently an assistant professor at the Japan Advanced Institute of Science and Technology from 2008 to 2014. He is currently an associate professor.