13.07.2015 Views

Proceedings - Viện Vật lý

Proceedings - Viện Vật lý

Proceedings - Viện Vật lý

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Advances _________________________________________________________________________________in Optics, Photonics, Spectroscopy and Applications. Aug. 2006, Cantho, Vietnamthe emitter stripe, which has not been affected by the high current density during the55 h of operation; see circles in Fig.4 (a). NOBIC-scans are taken at an excitationwavelength of 633 nm and 730 nm. The first one provides interband excitation ofthe QW, whereas 730 nm matches the maximum of the absorption band of thedefect band, which starts at 450 meV below the QW band edge. Comparing thenormalized Fig. 4 (c) and (d), the most striking feature is a 2.3 fold increase in the deeplevel defect contribution to the photocurrent in the „aged area‟ below the emitter stripeswith respect to that in the „pristine‟ reference area.Figure 4. a) Schematic diagram of the front view of a red-emitting HPLD and epitaxial layersequence of the laser structure; b) LBIC scans taken at 730 nm along the front facet of the devicemonitored after certain hours of regular operation at 0.9A cw; c) and d) NOBIC scans parallel tothe growth direction across the active region of the device. Full circles and open circles representdata obtained at excitation wavelengths of 633 nm and 730 nm respectively. The scan was extendedacross the solder layer to the heat sink in order to ensure that even the highly doped GaAs-contactlayer is completelyThe raw data for spectroscopic strain analysis at the considered emitters (thatshown in Fig.3) are introduced in Fig. 5 (a) and (b). These data are extracted fromthe µPC spectra as described in Ref. [8]. Figure 2 (a) shows µPL data taken fromtraces along the substrate. Here we plot the PL peak position, which is influencedby both, the hh-e and the lh-e bulk transitions in the substrate. The µPL scansclearly show that there are no residual strain fields in the substrate. This finding isimportant for our further discussion, excluding that a dislocation line, e.g. related toa V-defect in the substrate, might have caused the defect accumulation in theemitter [9]. Strain values in Fig.5 (c), (d) and (e) are calculated from the shift of1hh-1e transition with method described in Ref. [10]. Fig. 5 shows a completely295

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!