04.09.2013 Views

tesi R. Miscioscia.pdf - EleA@UniSA

tesi R. Miscioscia.pdf - EleA@UniSA

tesi R. Miscioscia.pdf - EleA@UniSA

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Chapter 2 79<br />

The research on the OE field has to face all these challenges,<br />

and more other ones, so that this new technology can assert itself and<br />

become competitive covering the market sectors left still partially<br />

unexplored by the inorganic electronic technology.<br />

Relying on the literature information reported in the State of<br />

the Art of the present doctoral thesis and on the just exposed<br />

Conclusions, future technological guidelines have to be defined for the<br />

development of this research activity.<br />

The guidelines are defined setting the target of integrating<br />

organic and/or polymeric and inorganic commercial materials in<br />

optimized structures in terms of electrical performances and<br />

fabrication processing.<br />

In particular, the topological factor is a fundamental<br />

component of the future organic transistor development: as disclosed,<br />

bottom-gate (BG) architectures are particularly advisable for<br />

optimizing the channel-semiconductor interface. Moreover, at the aim<br />

of obtaining the highest profit from the patterning techniques<br />

developed in inorganic electronic and from emerging, not<br />

conventional methods, structures with source and drain contacts<br />

placed under the channel will be realized for reducing the issues<br />

related to gate-overlap, loss currents and parasitic effects due to the<br />

diffusion of the vapour-phase deposited metals towards organic<br />

polycrystalline or amorphous semiconductor.<br />

In conclusion, the BGBC architecture appears particularly<br />

suitable for the future improvements of the presented activity. The<br />

drawback for using of this structure lies in the formation of interfacial<br />

dipoles between source/drain and channel thus increasing the contact<br />

resistance, producing potential barriers and reducing the device<br />

saturation current. At this aim, new processes will be experimented to<br />

optimize the contact-semiconductor interface putting down<br />

undesirable phenomena (injection barrier and parasitic resistances).<br />

As the choice of commercial materials to be employed in<br />

OTFT is wide and the electrical performances in static and dynamic<br />

regimes are often correlated to technological factors, such as<br />

processing, architectural and interface physics factors, the future<br />

research will consist in using of available materials and in optimizing<br />

the structure through the study and the modeling.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!