Origin of Blue Shifts in Quantum-Well Wires Unrelated to Lateral Confinement
- PDF / 467,658 Bytes
- 6 Pages / 420.48 x 639 pts Page_size
- 40 Downloads / 170 Views
ORIGIN OF BLUE SHIFTS IN QUANTUM-WELL WIRES UNRELATED TO LATERAL CONFINEMENT
ANDERS GUSTAFSSON, X. LIU, I. MAXIMOV, L. SAMUELSON AND W. SEIFERT Department of Solid State Physics, Lund University, Box 118, S-221 00 Lund, Sweden.
ABSTRACT Experimentally observed blue shifts of the peak position of the luminescence from quantum-well-wire and -dot structures are often significantly larger than the calculated shifts induced by lateral confinement in the structures. In this work we have used high-quality InGaAs/lnP multi-quantum-wells for the fabrication of wires. The quantum wells are in the range 3 to 17 monolayers (ML) nominally. The thinnest well, 3 ML, shows a clearly resolved split into two luminescence peaks from areas with a thickness difference of 1 ML. In the case of the wires, the luminescence from the thicker wells show a blue shift, as well a significant broadening. However, the thinnest well shows no blue shift, but a different ratio of the two peaks, with the high energy peak favoured in the wire case. We interpret these effects in terms of a reduced transfer of excitons from thinner to thicker areas of the well in the wire as compared to the unpatterned areas. This due to a reduction of the transfer from 2 dimensional to 1 dimensional in the wires. The peaks originating in areas of different ML thicknesses are not spectrally resolved in the thicker wells and the reduced transfer therefore results in a blue shift as well as a broadening of the luminescence peak. INTRODUCTION Blue shifts of the peak energy position of the luminescence from quantum wells (QWs) in etched wire or dot structures have been reported over the past few years [1-7]. The blue shifts have been attributed to lateral confinement. This even though the reported shifts are often larger than what can be expected from theoretical calculations of lateral confinement, based on the physical width of the wire. Even more surprising is the fact that the shift is more or less independent of the physical width of the etched wire structures. Usually no explanation for the discrepancy between theory and experiment is given. Often a blue shift of the luminescence is used as an evidence of lateral confinement. In figure 1 the theoretical curve for the additional blue shift in a wire of InGaAs/InP is shown, together with the blue shifts reported in ref 1-7. As an example, a 1 000 A wide wire is expected to have an additional lateral confinement of 1 meV and it takes a wire width of 350 A to introduce a 10 meV confinement.
100 •> FIGURE 1. The solid line
represents the theoretical the calculation of additional lateral confinement in a QW
wire. The calculationsare based on a semi-infinite barrier. The different symbols are the experimentally observed
blue shifts in wire structures,from ref 1-7.
"refl ,
3 ref 3 ref 4
+ ref 7
N
E=3 0
0
A 0r"
0)
(3
a A
0
m
A ref5 A ref6
• ref2
1 100
+ ++ 3
E
1000
Wire wit00 Wire width
10000
(A)
Mat. Res. Soc. Symp. Proc. Vol. 283. @1993 Materials Research Society
796
SAMPLE FABRICATION In this s
Data Loading...
