|
We study the electron-hole dynamics of c-plane GaN/AlN quantum dots (QDs) emitting above the bulk GaN bandgap [1] by means of time-resolved photoluminescence (TRPL) [2].
The PL dynamics displays a bi-exponential decay with a short time-constant (0.3 ns), independent of photon energy, and a longer one that follows the expected evolution of the radiative lifetime with the QD size. We attribute the fast dynamics to a non-radiative energy transfer to deeper levels, which have sufficiently low density and long lifetime such that their population can be easily saturated. A very long component, showing up as a constant background in our experiment, can thus be explained by the corresponding reverse process: carriers having experienced an ionization and/or trapping event will remain in that state during a long time period before slowly relaxing to their original state.
The intrinsic radiative recombination process is recognizable thanks to the characteristic dependence of its lifetime with the emission energy, due to the presence of the large built-in electric field typical of such polar quantum heterostructures [3]. By analytically modeling TRPL spectra, we show that the QD luminescence undergoes a linear redshift with increasing time-delay accompanied by a decrease of its linewidth, while the initial Gaussian lineshape is conserved. From this analysis, we not only demonstrate that the experiment is performed in such excitation conditions such that the dynamical screening of the built-in field by free carriers is negligible, but we also determine the energy-dependence of the QD radiative lifetime with an improved precision when compared to the usual procedure that consists in fitting the PL decays for different QD energies. Beyond the confirmation that the dependence of the effective PL lifetime with emission energy in GaN/AlN QDs is dominated by the built-in electric field [3,4], we also show, using variational calculation, that changes in the QD emission energy by several hundreds of meV within a given QD sample can be understood through fluctuations in the lateral dot dimensions within a given QD subset corresponding to a well-defined dot height. Numerical simulations, that show good qualitative agreement with experimental data over an energy range of 1 eV and six orders of magnitude for the effective PL lifetime when assuming an electric field of 7 MV/cm, enable to evaluate the dot height in our sample to be distributed around a mean value of 1.5 nm [2].
Bibliography:
[1] S. Tamariz, G. Callsen, and N. Grandjean, Applied Physics Letters 114, 082101 (2019).
[2] M. Hrytsaienko, M. Gallart, M. Ziegler, O. Crégut, S. Tamariz, R. Butté, N. Grandjean
B. Honerlage and P. Gilliot, Journal of Applied Physics 129, 054301 (2021)
[3] T. Bretagnon, P. Lefebvre, P. Valvin, R. Bardoux, T. Guillet, T. Taliercio, B. Gil, N. Grandjean, F. Semond, B. Damilano, A. Dussaigne, and J. Massies, Phys. Rev. B 73, 113304 (2006).
[4] S. Kako, M. Miyamura, K. Tachibana, K. Hoshino, and Y. Arakawa, Applied Physics Letters 83, 984 (2003)
|
