1996 Fiscal Year Final Research Report Summary
Study of microcavity semiconductor laser which allows the quantum optical confinement by vertical multiple reflectors
| Project/Area Number |
07837004
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 時限 |
|---|
| Research Field |
極微細構造工学
|
|---|
| Research Institution | Yokohama National University |
|---|
Principal Investigator |
BABA Toshihiko Yokohama National University Faculty of Engineering Associate Professor, 工学部, 助教授 (50202271)
|
|---|
| Project Period (FY) |
1995 – 1996
|
| Keywords | semiconductor laser / reflector / GaInAsP / InP / optical communication / short cavity / reactive ion beam etching / photonic integration |
|---|
| Research Abstract |
Next era optical fiber communications such as FTTH and optical interconnect require a semiconductor laser that enables the low power consumption, high efficiency and adaptability to mass production. In this study, a novel monolithic short cavity laser simultaneously satisfying these requirements was proposed and demonstrated. This laser has vertical multiple reflectors. Since this type of reflector is equivalent to the multilayr mirror with semiconductor and air, it achieves a high reflectivity over 0.99 even with a small number of pairs. Thus it allows the extreme short cavity, resulting in the ultra-low threshold. It has been shown theoretically that, when we assume GaInAsP/InP compressive strained quantum wells as the active layr, threshold current as low as 100muA will be possible with cavity length of 20mum and stripe width of 1mum. To fabricate this type of reflector, we optimized the condition of electron beam lithography and that of reactive ion beam etching (RIBE). It realized
… More
the semiconductor linewidth of 0.3mum in the multiple reflector, which corresponds the condition of the third diffractional grating. We observed the threshold current normalized by stripe width of 2.6mA/mum for dabricated laser with the refractor for one end and cavity length of 100mum. The comparison between the experiment and theoretical estimation of threshold current indicates that the effective reflectivity of the reflector in the fabricated device was nearly 0.6. This is slightly lower than the theoretical value obtained by taking the diffraction loss into account. It seems to be caused by the roughness of semiconductor sidewalls induced during the etching process. It was measured to be 20 nm from the topographic image by a multi-dimensional scanning electron micrograph. We have also observed that the increase of the number of semiconductor walls simply improves the threshold current. Theoretically it was found that (1) diffraction loss is reduced and effective reflectivity is increased to over 0.95 by changing the air space between semiconductor walls from the third order grating design to the first order, and (2) the first order design for both semiconductor linewidth and air space provides relatively small improvements to the effective reflectivity. However, the air space of the first order design requires the extremely fine etching condition. The trial fabrication of the air space by the RIBE etching concluded that the increase of acceleration voltage of ion beam allows upto the second order design but does not the first order design. We expect the following research to realize the first order design and drastically improve the laser performance by introducing the chemically assisted ion beam etching, which exhibits the smooth semiconductor sidewalls and high aspect ratio. It will not only contribute to the laser performance but also to the integration technology of laser and other photonic devices such as detectors and modulators. Less
|
