Effect of charging on CdSe/CdS dot-in-rods single-photon emission

The photon statistics of CdSe/CdS dot-in-rods nanocrystals is studied with a method involving postselection of the photon detection events based on the photoluminescence count rate. We show that flickering between two states needs to be taken into account to interpret the single-photon emission properties. With postselection we are able to identify two emitting states: the exciton and the charged exciton (trion), characterized by different lifetimes and different second-order correlation functions. Measurements of the second-order autocorrelation function at zero delay with postselection shows a degradation of the single-photon emission for CdSe/CdS dot-in-rods in a charged state that we explain by deriving the neutral and charged biexciton quantum yields.

Figure 1

(a) Transmission electron microscope (TEM) image of the investigated DRs before dilution. (b) DR geometrical features determined from statistical analysis performed on TEM images. Shell length $l=35$ nm, shell thickness $t=7$ nm, and core diameter $d=2.7$ nm. (c) Experimental setup. Wide field microscopy can be realized by exciting the sample with a UV lamp. In this case, a CCD camera is used to image the excited area of the sample. A zoom on a typical CCD camera image is presented with DRs photoluminescence highlighted by a red circle. Circularly polarized light from a picosecond-pulsed excitation laser diode with a wavelength of 404 nm, a pulse width of 100 ps, and a repetition rate of $2.5$ MHz is used to excite single DRs. A confocal microscope collects the photoluminescence of the DRs. A high pass filter (cutoff 570 nm) removes the remaining blue light from the optical path. Single-photon events are recorded by two avalanche photodiodes in a Hanbury-Brown and Twiss configuration. $\lambda /4$ = quarter wave plate, BS = beam splitter, APD = avalanche photodiode, TCSPC = Time-correlated single-photon counting.

Figure 2

Study of DR1 photoluminescence. (a) PL intensity of DR1 on second (left) and millisecond (right) time scales for a mean number of excitons in the nanocrystal of $\langle Neh\rangle =0.4$ and a bin time of 250 $\mu$s. (b) Histogram of the PL intensity shown in (a) (the number of events corresponding to a given intensity per time bin) together with a fit with two Poissonian distributions. The bright state in red (above 70 counts/ms) has a mean emission intensity $I_X=86$ counts/ms, the grey state in green (below 40 counts/ms) has a mean emission intensity $I_{X^{-}}=30$ counts/ms. (c) $g^{\left(2\right)}$ of the PL intensity on several decades of delays between photons. (Top) $g^{\left(2\right)}$ of the grey state [green intensity window in (a) and (b)]. (Middle) $g^{\left(2\right)}$ of the bright state [red intensity window in (a) and (b)]. (Bottom) $g^{\left(2\right)}$ of the whole PL intensity. (d) Decay of the PL intensity. Blue diamonds: PL decay of the whole intensity distribution in (b). Red squares: PL decay for the photons in the red intensity window in (b), a monoexponential decay fit gives ${\tau }_X=65$ ns. Green circles: PL decay for the photons in the green intensity window in (b), a monoexponential decay fit gives ${\tau }_{X^{-}}=11.6$ ns.

Figure 3

Schematic of the different relaxation pathways after a pulsed excitation of the shell state continuum. Green arrows symbolize the possible Auger relaxation with their associated rates ${\gamma }_{A^{+}}$ if the relaxation energy is given to a hole or ${\gamma }_{A^{-}}$ if it is given to an electron. Dashed green arrows mean that different Auger relaxation possibilities are in competition. Orange arrows symbolize radiative recombinations. For the low-power excitation considered, $\langle Neh\rangle =0.4$, five relaxation cases are possible. For a neutral DR, (a) single-photon emission after an Auger nonradiative decay of the biexciton and (b) biexciton binding and two-photon emission with the given probabilities. For an ionized DR, (c) nonradiative relaxation of the biexciton and exciton, (d) single-photon emission after Auger nonradiative decay of the biexciton, and (e) biexciton binding and two-photon emission.

Figure 4

ACF for DR1 at short time scales. (a) Grey state ACF. (b) Bright state ACF. (c) Whole PL intensity ACF.

Figure 5

(a) Histogram of the PL intensity of DR2 (number of events corresponding to a given intensity per time bin) for a mean number of excitons in the crystal of $\langle Neh\rangle =0.4$, bin time $250\mu$s. Count rates below 30 counts/ms in green are associated with the grey state, while the part of the histogram above 100 counts/ms is attributed to the bright state. (b) PL decay curves for grey state photons (green circles) and bright state photons (red squares). (c) $g^{\left(2\right)}$ for DR2 on several decades of delays between the photons. From top to bottom: $g^{\left(2\right)}$ for the grey state photons, bright state photons together with the $g^{\left(2\right)}$ of the whole PL intensity. (d) ACF for DR2 at short time scales. From top to bottom: ACF for the grey state photons, bright state photons together with the ACF of the whole PL intensity.