11.07.2015 Views

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

Copyright 2004 by Marcel Dekker, Inc. All Rights Reserved.

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

statistics for the off times are insensitive to the different characteristics (size,shape, defects, environment) of each dot. Initial experiments at RT show thatthe same blinking statistics are also observed in CdTe QDs, demonstratingthat this power-law phenomenon is not restricted to CdSe QDs.A. Temperature DependenceTo develop a physical model from this phenomenological power-lawbehavior, we probe the temperature dependence of the blinking statistics;this dependence should provide insight into the type of mechanism (tunnelingversus hopping) and the energy scales of the blinking phenomenon.Qualitatively, the time traces in Figs. 2a and 2b suggest that at a lowtemperature, the QDs blink less and stay in the on state for a larger portionof the time observed. However, when we plot the off-time probabilitydistributions at temperatures ranging from 10 to 300 K, as shown in Fig.8b, the statistics still show power-law behavior regardless of temperature.Moreover, the average exponents in the power-law distributions are statisticallyidentical for different temperatures (10 K: 1.51 F 0.1; 30 K: 1.37F 0.1; 70 K: 1.45 F 0.1; RT: 1.41 F 0.1). Such a seemingly contradictoryconclusion is resolved <strong>by</strong> plotting the on-time probability distribution at 10 Kand RT, as shown in Fig. 9c. Unlike the off-time distribution, the on timeshave a temperature dependence that is qualitatively observed in the raw dataof Fig. 2.B. On-Time TruncationThe on-time statistics also yield a power-law distribution with the sameexponent* as for the off-times, but with a temperature-dependent ‘‘truncationeffect’’ that alters the long time tail of the distribution. This truncationreflects a secondary mechanism that eventually limits the maximum on timeof the QD. The truncation effect can be seen in the on-time distribution of asingle QD in Figs. 9a and 9b, and in the average distribution of many singleQDs as a downward deviation from the pure power law. At low temperatures,the truncation effect sets in at longer times and the resulting timetrace shows ‘‘long’’ on times. The extension of the power-law behavior forlow temperatures on this logarithmic timescale drastically changes the time* The power-law distribution for the on times are difficult to fit due to the deviation from powerlaw at the tail end of the distribution. The power-law exponent with the best fit for the ontimes is observed for low excitation intensity and low temperature.<strong>Copyright</strong> <strong>2004</strong> <strong>by</strong> <strong>Marcel</strong> <strong>Dekker</strong>, <strong>Inc</strong>. <strong>All</strong> <strong>Rights</strong> <strong>Reserved</strong>.

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

Saved successfully!

Ooh no, something went wrong!