Close look at charge carrier injection in polymer field-effect transistors

Abstract

Parasitic contact resistance effects are becoming a major issue in organic transistors in that they can severely limit or even dominate their overall transistor performance. We present a systematic study of the contact resistance in bottom-contact polymer field-effect transistors made from poly(3-hexylthiophene) (P3HT) as well as $\mathrm{poly-}{\text{9,9}}^{\prime }\mathrm{dioctyl-fluorene-co-bithiophene}$ (F8T2). A microscopic approach based on noncontact scanning-probe potentiometry was used to directly separate the transport properties of the transistor channel and the electrode/polymer contacts, giving very accurate experimental access to both the source and drain contact resistance. The influence of the relevant parameters (temperature, electrode work function, ionization potential of the polymer, charge carrier mobility) on the source/drain contact resistance is investigated. We find that for “good” source/drain contacts that give rise to relatively small overall contact resistances (⩽50 kΩ cm), e.g., P3HT with chromium–gold electrodes, the source and the drain contact resistances are almost identical and are governed by bulk transport through the conjugated polymer. However, for “bad” contacts with a Schottky barrier for hole injection ${\phi }_b\text{⩾0.3\hspace{0.17em}}\mathrm{eV},$ e.g., F8T2 with gold electrodes, the source contact resistance is considerably larger than the drain contact resistance and is dominated by charge-carrier injection at the source. Surprisingly small activation energies of 60–140 meV have been found for the source contact resistance, which are smaller than both ${\phi }_b$ and the activation energy of the mobility. From this we conclude that the commonly assumed (diffusion-limited) thermionic-emission models do not adequately describe the charge injection process in bottom-contact polymer transistors. On the basis of our results we propose a simple model, in which the source contact resistance is assumed to be the sum of resistance arising from the injection process and resistance due to bulk transport through a depletion region, whereas only the latter contributes to the drain contact resistance.