Quantification of photoinduced and spontaneous quantum-dot luminescence blinking

Photoluminescence blinking of ${\mathrm{Zn}}_{0.42}{\mathrm{Cd}}_{0.58}\mathrm{Se}$ colloidal quantum dots was investigated experimentally as a function of the excitation intensity and modeled via a Monte Carlo method. In order to test the current blinking model of electron tunneling to states external to the quantum dot, experiments were performed with quantum dots on (insulating) glass and (conducting) indium-tin-oxide substrates. Comparison of simulated and experimental data allows one to extract characteristic parameters of the blinking free of artifacts and shows that the blinking can be understood in terms of two independent processes. First, a process is active for switching between the on and off states and has a power-law probability distribution for the length of both on and off periods. Second, a single-rate transition from the on to the off state is evoked. The power-law blinking process is found to be insensitive to excitation intensity and the electronic states of the substrate and presents identical statistics for the two switching directions, pointing to an intrinsic symmetry. The single-rate on→off transition can account for all experimentally observed photoinduced and environmental (substrate) effects. On glass, its behavior is well described by a one-photon process in agreement with earlier studies on insulators. In contrast, a different behavior is found on the conducting substrates.

Figure 1

(a) Histogram of the photon-arrival times corresponding to the fluorescence emission of a single QD on glass (Excitation intensity $P=2.04\mathrm{kW}∕{\mathrm{cm}}^2$). (b) 0.4-s detail of the data shown in (a). The horizontal line marks the threshold used to classify the bins as on and off in order to determine the length of the on- and off-periods ($t_{\text{on}}$ and $t_{\text{off}}$, respectively). (c) Histogram of the detected $t_{\text{on}}$ and $t_{\text{off}}$. The dashed line is a power-law $(\propto t^{-m_h})$ fit to the distribution of $t_{\text{off}}$.

Figure 2

Histograms of the experimental and simulated $t_{\text{on},h}$ and $t_{\text{off},h}$ for QDs on glass and ITO under different excitation powers $I[\mathrm{kW}∕{\mathrm{cm}}^2]$; each computed from $\sim 15$ traces with total time $T$ as indicated in the graphs. A power law $(y=A{x}^{-m};A=\mathrm{const})$ was fitted to the experimental $t_{\text{off},h}$ histograms. For clarity, $t_{\text{on},h}$ and $t_{\text{off},h}$ histograms are separated by a vertical offset.

Figure 3

Experimental and simulated cycles per second (a) and on-time fraction (b). The curves are slightly shifted horizontally for clarity. The error bars mark the extreme values obtained from 10 simulations with all parameters fixed.

Figure 4

Photoinduced lifetime of the on-state $\left({\tau }_{\mathrm{PI}}\right)$ used to reproduce the glass and ITO experimental data as a function of excitation power $I$ in log-linear and log-log scales.