Epitaxial lateral overgrowth of AlN layers on patterned sapphire substrates

1
Epitaxial lateral overgrowth of AlN layerson
patterned sapphire substrates

• ???? ? ? ??
• ? ? ???
• ? ?97.12.01

2
Outline

• Introduction
• Experiments
• Results and discussion
• Conclusion
• References

3
Introduction

• The realization of high-performance UV
  light-emitting devices is one of the most
  important targets for group III nitride
  semiconductors.
• To realize high-performance UV devices such as
  light emitting diodes . the growth of
  high-quality AlN is essential. It is still a
  critical issue to obtain high-quality AlN .
• A combination of high-temperaturegrowth and ELO
  will lead to the reduction of threading
  dislocation density and thus achieve to much
  higher quality AlN

4
Experiments

• Two different directions of pattrned sapphire
  substrates (0001) have been used in this
  investigation. In the first type of substrate,
  linear trenches have been formed along the (1010)
  direction. In the secondtype, trenches have been
  formed along the (1120) direction. The width and
  depth of the trenches are both 500 nm. The
  adjacent terrace width is 3 µm.
• The ELO-AlN layers were grown on these substrates
  by high-temperature MOVPE technique under H2
  atmosphere.

5

• The growth is carried out at a constant pressure
  of 100 Torr
• the substrate is annealed at a temperature of
  1330 C for 5
• minutes.
• the growth is carried out at a temperature of
  1300 C, for 7 120 minutes.
• For comparison, the growth of AlN at a relatively
  low temperature of 1100 C on a trench pattern
  along (1120) direction was also performed.

6
Results and discussion

• Figure 2 shows a SEM cross-sectional image of AlN
  grown at a temperature of 1100 C, which is the
  temperature commonly used for the growth of AlN
  on sapphire. Although AlN was coalesced, the
  surfaceis quite rough. In comparison, the surface
  of AlN grown at temperatures higher than 1300 C
  is
• atomically flat.

7

• Figure 3 shows the SEM cross-section image of the
  AlN layer grown on the sapphire substrate with
  grooves formed along (1010) direction. As clearly
  seen in the SEM image, AlN layer was not
  coalesced. There is a gap between the adjacent
  layers. analyze the dislocations in the
  layersespecially in the regions between the
  vertical portions,

8

• Figure 4 shows the TEM image of portion
  surrounded by black dotted lines in the Fig. 3.
  Dislocations are found
• to propagate vertically from the seed region
  through the top surface. From these observations,
  it can be concluded that the grooves formed on
  sapphire substrates along the (1010) direction is
  not suitable for growing AlN layers with reduced
  dislocation density.

9

• The SEM image of the cross-section of the AlN
  layers grown on the sapphire substrates with the
  linear grooves formed along the (1120) direction
  is presented in Fig. 5. As seen in the SEM image.
  TEM analysis was done on a prominent portion of
  the AlN layer indicated by the white dotted lines
  in order to analyze the dislocation propagation
  behaviour in detail. substrates.

10
Conclusion

• High temperature MOVPE growth of AlN on the
  sapphire substrate having trench pattern was
  conducted.
• We found a strong difference of the property of
  ELO between two trench directions. In order to
  obtain coalesced AlN, (1120) trench is essential.
  The average dislocation density of a coalesced
  AlN was 6.7 108 cm2, which is less than a half
  that of AlN grown on planer sapphire.
• (1120) trench (0002) XRD rocking curve FWHM 314
  aresec is better than (1010) 352 aresec. These
  results show that ELO technique is also suitable
  for reducing dislocation density of AlN on
  sapphire.

11
References

• 1 S. Nitta, M. Kariya, T. Kashima, S.
  Yamaguchi, H. Amano, and I. Akasaki, Appl. Surf.
  Sci. 159/160, 421 (2000).
• 2 N. Fujimoto, T. Kitano, G. Narita, T. Fuse,
  K. Balakrishnan, M. Iwaya, S. Kamiyama, H. Amano,
  I. Akasaki, K.
• Shimono, T. Noro, T. Takagai, and A. Bandoh, 52nd
  JSAP 30p-L-9 (2005).
• 3 M. Imura, K. Nakano, T. Kitano, T. Fuse, N.
  Fujimoto, K. Balakrishnan, M. Iwaya, S. Kamiyama,
  H. Amano, T.Noro, and T. Takagi, 52nd JSAP
  30p-L-10 (2005).
• 4 T. M. Katona, M. D. Craven, J. S. Speck, and
  S. P. DenBaars, Appl. Phys. Lett. 85, 1350 (2004).