Optical properties of $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ monolayer quantum wells and self-assembled quantum dots

The optical properties of $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ monolayer (ML) quantum wells (QW’s) and self-assembled quantum dots (QD’s) grown by molecular beam epitaxy are investigated by continuous-wave (cw) and time-resolved (TR) photoluminescence (PL) spectroscopy. The cw-PL spectra clearly show the growth development from ML-QW to QD’s as the CdSe thickness increases from 1 to beyond 3 ML. The carrier dynamics are characterized chiefly by TRPL. For ML-QW’s, it is determined that carrier localization in the potential fluctuations and scattering by phonons significantly influence the optical properties. On the other hand, TRPL reveals the carrier dynamics for QD’s and a wetting layer (WL); some of the carriers photogenerated in the WL radiatively emit, while others are captured by QD’s via a carrier-carrier scattering on a $\sim 5\mathrm{ps}$ time scale. Within the next $10–50\mathrm{ps}$, the carriers relax to the ground level via a carrier–LO–phonon scattering. Finally, the carriers radiatively recombine with a recombination lifetime of $100–300\mathrm{ps}$, depending on the emission energy.

Figure 1

Spinodal curve for MgCdSSe alloy at a typical growth temperature of $240°\mathrm{C}$. The shaded region indicates the immiscible region.

Figure 2

Heavy-hole exciton binding energy for $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ and $\mathrm{Cd}\mathrm{Se}∕\mathrm{Zn}\mathrm{Se}$ QW’s.

Figure 3

Contour plots of transition energy, binding energy, and spontaneous emission lifetime of $1s$ heavy-hole excitons in $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ quantum disks.

Figure 4

Low-temperature $\left(5\mathrm{K}\right)$ PL spectra of ML-QW’s [(a) and (b)] and (c) QD’s. The nominal CdSe thicknesses are (a) 1 ML, (b) 2 ML, and (c) $>3$ ML. The ZnSe heavy-hole (HH) emission is from the buffer layer. The inset shows the polarization PL spectra for 2 ML QW and QD’s acquired along the [110] and $\left[1\overline{1}0\right]$ directions.

Figure 5

Typical AFM image of QD samples without MgS caps. The marker represents $1\mu \mathrm{m}$.

Figure 6

Temperature dependence of (a) the PL peak energy, (b) the integrated intensity, and (c) the linewidth of the 2-ML QW emission. In (a), the dashed line is the band gap of CdSe (Ref. 20) shifted to a higher energy for comparison with the data, and the inset shows a detailed variation of the PL peak energy below $50\mathrm{K}$. The solid line in (b) is a fit to the data as described in the text (reproduced from Ref. 12).

Figure 7

(a) Schematic of potential fluctuations used for the explanation of the PL temperature dependence shown in Fig. 6. (b) An example of the layer growth that could realize the potential fluctuations in (a), where I and II are CdSe islands and III is either a CdSe ML or the MgS barrier layer. (Reprinted from Ref. 12).

Figure 8

(a) Time-integrated PL spectrum and PL decay time $\left({\tau }_{\text{fast}}\right)$ of the $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ ML-QW. The dashed line is the result of a fit; see the text for details. (b) PL decay curve at the emission peak. The inset illustrates the variation of the fast- and slow-decay components as a function of the excitation power, which was extracted by the fit using Eq. (1).

Figure 9

TRPL spectra. The origin of time is when the laser intensity for excitation reaches the maximum.

Figure 10

Decay curves of the PL peak at various temperatures.

Figure 11

Temperature dependence of the PL decay times. The solid and dashed lines are the results of fitting assuming an exciton-phonon interaction.

Figure 12

Summary of the temperature dependence measurement for the QD sample with the PL spectrum at $5\mathrm{K}$ shown in Fig. 4c. The dashed line in (a) is the shifted band gap of CdSe. The solid lines in (b) are the result of the simulations based on the rate equations. (Reproduced from Ref. 12).

Figure 13

PL decays taken at the energy positions designated by arrowheads in the inset, where the time-integrated PL spectrum is shown. The excitation power was $0.3\mu \mathrm{J}∕{\mathrm{cm}}^2$.

Figure 14

PL decay times as a function of the emission energy. The open circles and triangles represent slow and fast PL decay, respectively, evaluated using Eq. (1).

Figure 15

Intensity ratio between the fast- and slow-decay components and fast-decay time in the energy range of the QD emission. The dependences on the excitation power are shown.

Figure 16

cw PL spectra resonantly excited by the energies shown in the figure.

Figure 17

Temperature dependence of the QD PL decay time. The open circles and triangles represent the slow and fast decay, respectively. The solid curve is the result of a fit. See the text for the detail.

Figure 18

Time-integrated PL spectra of WL (and ZnSe) under different excitation powers. The spectra were normalized with respect to the ZnSe emission. The inset is the intensity ratio between the fast and slow decays as a function of the excitation power.

Figure 19

PL and PLE spectra of the QD sample.

Figure 20

TRPL spectra of a $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ QD.

Figure 21

Schematic illustration of the carrier dynamics in the present $\mathrm{Cd}\mathrm{Se}∕\mathrm{Mg}\mathrm{S}$ QD’s.