InGaAs/GaAs Quantum Dot Interdiffusion Induced by Cap Layer Overgrowth
- PDF / 1,934,981 Bytes
- 6 Pages / 414 x 635.4 pts Page_size
- 49 Downloads / 207 Views
growth a fully strained so-called wetting layer (WL) is formed. When its thickness reaches the critical value, which is of the order of a few monolayers, the two-dimensional growth mode switches into the three-dimensional one and small coherent islands known as self-assembled QDs start to appear. Part of the strain energy is then replaced by surface energy and the total energy is minimized. To obtain an active QD layer it has to be overgrown by a capping layer of barrier material. One of the biggest problems in device-oriented QDs applications is their uniformity. Usually QD size distribution is broad for at least 10%, which results in a serious inhomogeneous broadening of QD photoluminescence (PL) and other optical spectra [9]. Another important factor related with optoelectronic applications of self-assembled QDs is tuning their interband transition energy. It was shown that both size distribution and intersubband transition energy can be changed by modification of QDs' size, shape or composition. These changes can be achieved using different growth conditions [10], substrate orientations [11] and post-growth annealing [12-14]. In particular, the post-growth annealing results in a relatively large blue-shift of the PL emission and narrowing of the full width at half-maxima (FWIHM) of the PL line [12-13,15-17]. It is belived that this is due to interdiffusion occurring after a QDs growth. This process is an important mechanism in QD-based device engineering. Most studies dealing with this issue have focused on QD interdiffusion caused by post-growth annealing. On the other side, less work has been done on the effect of the capping layer growth [18-23]. In this study, we show that growth of the capping layer leads to a significant QD interdiffiusion and that the effect of such growth can be similar to the effect of post-growth annealing. 179 Mat. Res. Soc. Symp. Proc. Vol. 618 ©2000 Materials Research Society
EXPERIMENTAL PROCEDURE Samples investigated in this study were grown by metal organic vapour phase epitaxy (MOVPE) on misoriented (20 off) GaAs n+ (100) substrates. Sample #328 consisted of a GaAs buffer layer, a layer of self-assembled In 0 .5Gao.5As QDs, a 30nm thick GaAs layer, 40nm of Al0.2 Gao0.As and 30nm of GaAs. A 40nm layer of GaAs instead of Al0 .2Gao.&As was grown in sample #350. All these three different materials were grown at significantly different temperatures: Ino. 5Gao.5As QDs at 550 0C, GaAs at 650TC and Al0.2Gao.sAs at 750 0 C (growth of the Al0 .2Gao.gAs layer took about 4min). The QDs growth was repeated at the top of the structure in order to reveal their morphology using atomic force microscopy (AFM). In order to investigate the effect of post-growth annealing three samples of the #350 structure were annealed in the MOVPE reactor under AsH 3 ambient at 750 0 C for 1min (#350a), 3min (#350b) and 5min (#350c), respectively. Temperature dependent PL measurements were performed in a continuos flow CF-1204 Oxford cryostat with excitation with a semiconductor laser ()= 780nm) at a temperatu
Data Loading...
