Recombination of excitons bound to oxygen and silicon donors in freestanding GaN

Abstract

The neutral donor bound exciton recombination processes in freestanding GaN have been studied. The photoluminescence spectrum shows emission lines related to silicon and oxygen donors. Time-resolved luminescence allows us to correlate the principal donor bound exciton lines with their two-electron satellites. The magnetic field splitting of the two-electron lines is well described by the theory of the hydrogen atom in a magnetic field. For the oxygen donor a 1.5 meV chemical shift and a 30.8 meV effective Rydberg have been evaluated. Two-electron satellites involving excitations to the 2p and 2s donor states are separated by an energy of 1.0 and 1.3 meV for O and Si impurity, respectively. The temperature dependence of the two-electron emission clearly shows that this separation arises from a splitting of the ground state of the neutral donor bound exciton complex. The nature of this splitting is discussed and it is suggested that it is due to rotational states of donor bound excitons.

Reply to “Comment on ‘Recombination of excitons bound to oxygen and silicon donors in freestanding GaN’ ”

“…It is worth noting, that the rise time observed for the L-E wing is longer than that observed for the D 2 0 X A line, and is similar to that of the L1 and L2 lines corresponding to the oxygen donor. 2 On the other hand, the rise time observed for the L4 line is similar to that of the D 2 0 X A line. A two-peak fitting procedure gives curves, plotted in We regret that our presentation of intradonor transitions and donor chemical shifts is misleading for the reader.…”

“…We present a discussion of the intradonor transition energies and donor chemical shifts to avoid misunderstanding. In their Comment 1 on our publication, 2 Freitas, Moore, and Shanabrook ͑FMS͒ discuss differences between our paper 2 and the paper 3 by Freitas et al ͑FEA͒. The major difference between our analysis and that of FMS is in the assignment of the bound exciton recombination lines.…”

“…2 Thus the statement of FMS concerning this point seems to result from a misunderstanding of our data presentation.…”

“…As can be seen in the photoluminescence spectrum, 2 are at least three donors present in freestanding GaN. It is worth noticing that the shallowest donor observable in the PL experiment is not the shallowest uncompensated donor, as is the case for the IR absorption.…”

ParamagneticGaN:Feand ferromagnetic(Ga,Fe)N: The relationship between structural, electronic, and magnetic properties

“…15 is assigned to a two electron satellite D 0 X A,n=2 , where the recombination of the D 0 X A leaves the donor in its excited n = 2 state. 54,55 From the separation between the principal D 0 X A and the D 0 X A,n=2 maximum, the donor-ionization energy can be obtained. 56 In our case it results in E D ≈ 30 meV, pointing to oxygen or silicon as principal binding site for the D 0 X A .…”