Nanosecond Time-Resolved Microscopic Gate-Modulation Imaging of Polycrystalline Organic Thin-Film Transistors

We develop a time-resolved microscopic gate-modulation ($\mu \mathrm{GM}$) imaging technique to investigate the temporal evolution of the channel current and accumulated charges in polycrystalline pentacene thin-film transistors (TFTs). A time resolution of as high as 50 ns is achieved by using a fast image-intensifier system that could amplify a series of instantaneous optical microscopic images acquired at various time intervals after the stepped gate bias is switched on. The differential images obtained by subtracting the gate-off image allows us to acquire a series of temporal $\mu \mathrm{GM}$ images that clearly show the gradual propagation of both channel charges and leaked gate fields within the polycrystalline channel layers. The frontal positions for the propagations of both channel charges and leaked gate fields coincide at all the time intervals, demonstrating that the layered gate dielectric capacitors are successively transversely charged up along the direction of current propagation. The initial $\mu \mathrm{GM}$ images also indicate that the electric field effect is originally concentrated around a limited area with a width of a few micrometers bordering the channel-electrode interface, and that the field intensity reaches a maximum after 200 ns and then decays. The time required for charge propagation over the whole channel region with a length of $100\mu \mathrm{m}$ is estimated at about 900 ns, which is consistent with the measured field-effect mobility and the temporal-response model for organic TFTs. The effect of grain boundaries can be also visualized by comparison of the $\mu \mathrm{GM}$ images for the transient and the steady states, which confirms that the potential barriers at the grain boundaries cause the transient shift in the accumulated charges or the transient accumulation of additional charges around the grain boundaries.

Figure 1

(a) Schematic representation of the experimental setup for the $\mathrm{TR}\text{−}\mu \mathrm{GM}$ imaging measurements. (b) GM spectra at local areas showing positive ($A$) and negative ($B$) signals in the $\mu \mathrm{GM}$ images at 1.85 eV. (c) Trigger timing of the gate voltage application, camera exposure, LED lightning, and image intensifier operation. (d) Schematic showing the measurement principle for the $\mathrm{TR}\text{−}\mu \mathrm{GM}$ imaging.

Figure 2

Optical micrograph and the corresponding $\mathrm{TR}\text{−}\mu \mathrm{GM}$ images at various delay times captured with a time resolution of 100 ns. The steady-state image is measured at a delay time of $500\mu \mathrm{s}$.

Figure 3

(a) The lateral line profiles of $\mathrm{TR}\text{−}\mu \mathrm{GM}$ images (Fig. 2) whose intensity is divided by that in the steady state. The red and blue lines indicate the positive and negative $\mu \mathrm{GM}$ signals. The black arrows indicate the frontal positions of the channel determined by extrapolating the linear fit (green line) to the zero baseline. The yellow regions correspond to the source and drain electrodes. (b) The expanded line profile near the source electrode.

Figure 4

(a) Schematic representation of the transient carrier distribution in a TFT. (b) The square of the distance from the source electrode to the frontal position of the channel plotted as a function of time. (c) Transient carrier distribution simulated by the temporal-response model of TFTs (upper graph) and a comparison with the experimental results for a positive $\mu \mathrm{GM}$ signal (lower graph).

Figure 5

(a) $\mathrm{TR}\text{−}\mu \mathrm{GM}$ image and the expanded image at a delay time of 250 ns with a time resolution of 50 ns. (b) Cross-sectional profile on the white line in the $\mathrm{TR}\text{−}\mu \mathrm{GM}$ image for the transient (red) and steady states (black). (c) $\mathrm{TR}\text{−}\mu \mathrm{GM}$ image at 250 ns whose intensity is divided by that in a steady state. (d) Cross-sectional profile on the white line the divided image.

Figure 6

Cross-sectional profiles on the white lines in crossed-Nicols polarized micrographs and the corresponding $\mathrm{TR}\text{−}\mu \mathrm{GM}$ images at a delay time of 250 ns obtained with a time resolution of 50 ns. (a) Border between negative and positive $\mu \mathrm{GM}$ signals at a grain boundary. (b) Border between positive and negative $\mu \mathrm{GM}$ signals inside a grain. (c) Border between negative and positive $\mu \mathrm{GM}$ signals inside a grain. (d) Border between positive and positive $\mu \mathrm{GM}$ signals at a grain boundary. (e) Border between negative and negative $\mu \mathrm{GM}$ signals at a grain boundary. Note that the cross-sectional profiles at 250 ns (red) are shown with those in a steady state (black).

Figure 7

Schematics showing the microscopic charge distribution in a transient state.