Lattice-mismatched epitaxy of AlInP and characterization of its microstructure and luminescence

The synthesis of high-quality III-V ternary alloy semiconductors is vital to the success of major technologies such as LEDs, laser diodes, high-efficiency solar cells and power electronics. However, the epitaxy of ternary alloys can be complicated due to the dissimilar behavior of constituent atoms on the growth surface. Historically, lattice matching of the ternary alloy to a substrate is another important constraint that limits access to all but a few alloy fractions. A technologically important ternary alloy in which both surface kinetics and lattice matching is crucial is the wide band-gap AlxIn1-xP system. This alloy is commercially used at one specific composition, a random solid solution of Al0.5In0.5P lattice-matched to GaAs, as a high indirect band-gap cladding/barrier layer for the ubiquitous red LEDs and laser diodes. Little is known about AlxIn1-xP at other compositions and with non-random microstructures despite the potential benefit of gaining access to the highest direct band-gap semiconductor amongst all non-nitride III-V semiconductors. In this thesis, InyGa1-yAs compositionally graded buffers are used to bridge lattice mismatch, leading to the synthesis of AlxIn1-xP at a range of compositions with low threading dislocation densities (105-06/cm2) and low oxygen levels (2x1016/cm3). The high-quality of these films result in the first report of room-temperature yellow-green luminescence from AlxIn1-xP comparable in brightness to lattice-matched films. An accurate band-gap vs composition map of the AlxIn1-xP alloy space is created with Al0.43In0.57P identified as having the highest direct band-gap of 2.33 eV. The formation of non-random microstructures in AlxIn1-xP due to phase separation and atomic ordering is studied in detail. Phase separation into aluminum-rich and indium-rich domains is found to evolve from random compositional perturbations via a positive-feedback process limited by aluminum surface-diffusion. A reduction in band-gap by more than 200 meV is obtained by converting a random microstructure to a non-random one using growth temperature. Random/non-random interfaces are designed to use this large band-gap change to improve the efficiency of the first double-heterostructure yellow-green and amber AlxIn1-xP LEDs. Finally, we observe and explain how strain fields from an inhomogeneous distribution of misfit dislocations result in surface roughness and composition fluctuations in lattice-mismatched AlxIn1-xP LEDs. The results obtained in this work will be useful not only in providing control over nanometer-scale structures but also over wafer-scale features. This is crucial in transitioning novel lattice-mismatched devices from the lab to the marketplace.

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Materials Science and Engineering.

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2014

Author dc:creator

• Mukherjee, Kunal, Ph. D. Massachusetts Institute of Technology

Advisor dc:contributor.advisor

• Eugene A. Fitzgerald.

• Materials Science and Engineering.

Statement dc:rights

• MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission.

Licence dc:rights.uri

• http://dspace.mit.edu/handle/1721.1/7582

Language dc:language.iso

eng

Handle dc:identifier.uri

http://hdl.handle.net/1721.1/112386

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/112386