Exciton recombination dynamics in $\mathrm{In}\mathrm{As}∕\mathrm{In}\mathrm{P}$ self-assembled quantum wires

In this work we investigate the exciton recombination dynamics in $\mathrm{In}\mathrm{As}∕\mathrm{In}\mathrm{P}$ semiconductor self-assembled quantum wires, by means of continuous wave and time resolved photoluminescence. The continuous wave photoluminescence results seem to indicate that the temperature quenching of the emission band seems to be more probably due to unipolar thermal escape of electrons towards the InP barrier. On the other hand, the analysis of time resolved photoluminescence reveals that the temperature dependence of the radiative and nonradiative recombination times is mainly determined by the dynamics of excitons localized by disorder (high energy tail of the PL band) and strongly localized (low energy tail of the PL band) in local size fluctuations of the quantum wires.

Figure 1

(a) Continuous lines stand for PL spectra measured at $12\mathrm{K}$ and at room temperature, and dotted lines correspond to PL-Gaussian components (P1–P4); (b) Arrhenius plot of each PL-Gaussian component integrated intensity (symbols), and the best fitting curves according to Eq. (1) (dotted lines).

Figure 2

Gaussian height distribution obtained from atomic force microscopy (AFM) measurements of an uncapped InAs QWRs sample. A $500\times 500\mathrm{nm}$ AFM image is shown as an inset.

Figure 3

(a) Dotted lines correspond to experimental PL transients at $10\mathrm{K}$ for the P2 component measured under different excitation densities. (b) Measured decay time as a function of the detection energy at different excitation densities: 1, 10, and $100\mathrm{W}∕{\mathrm{cm}}^2$ represented by squares, circles, and triangles respectively. Dotted line reproduces the PL spectrum shown in Fig. 1a on the energy range where PL transients have been measured. As an inset, the level scheme applying for the free and localized exciton dynamics with temperature; label “∗” in the rate ${\Gamma }_{BF}$ and the time ${\tau }_{FL}$ indicates that they are affected with the thermal factor, as written in Eqs. (4, 5).

Figure 4

(a) Temperature evolution of the decay time deduced from the monoexponential PL transients measured for P2 Gaussian component, and (b) estimation of the radiative and nonradiative recombination times as explained in the text.

Figure 5

(a) Logarithmic plot of the radiative recombination time against temperature for P1 (squares), P2 (circles), and P3 (triangles) detecting energies at 10 (hollow symbols) and $100\mathrm{W}∕{\mathrm{cm}}^2$ (solid symbols); dotted lines stand for the best fit to a $T$-power law. (b) Arrhenius plot of the nonradiative time for P1–P2–P3 Gaussian components (solid symbols) and the best fit to an energy activated law (continuous lines).