Publication Date


Date of Final Oral Examination (Defense)


Type of Culminating Activity


Degree Title

Master of Science in Materials Science and Engineering


Materials Science and Engineering

Major Advisor

Paul J. Simmonds, Ph.D.


Dave Estrada, Ph.D.


Eric Jankowski, Ph.D.


III-V semiconductors grown by molecular beam epitaxy (MBE) on (111) surfaces have some interesting electronic properties. For certain materials with a (111)-orientation, the Γ- and L-valleys are reasonably close in energy. This means that it may be possible to take advantage of electron conduction in the L- and Γ-valleys at the same time, allowing us to overcome the so-called “density-of-states bottleneck,” and enable transistors with large drive currents.1 We have investigated this phenomenon in GaSb- and InAs-based 2D electron gases for which the electron effective masses are low.

However, growth of materials with a (111) orientation is typically more challenging than on traditional (001) surfaces. The MBE conditions needed to grow high quality material are often poorly understood.2 We began by exploring InAs/GaSb quantum well (QW) structures,3 grown directly on GaSb(111)A substrates. This work shows that low growth rates under very high group V overpressures produce good GaSb homoepitaxy and InAs heteroepitaxy, as characterized by XRD and AFM. However, although we have been able to identify MBE conditions that lead to the growth of smooth, high-quality material, GaSb(111)A substrates are extremely expensive, as well as being intrinsically n-type, which complicates the carrier transport measurements in which we are interested. If we could instead grow our GaSb-based QW structures on cheaper, non-conductive GaAs(111)A substrates, we could overcome these issues. The challenge is the large lattice mismatch between GaSb and GaAs, which typically results in strain-driven crystallographic disorder at the heterointerface and poor material quality.

One technique that has shown promise in circumventing these problems on (001) surfaces is the use of interfacial misfit arrays (IMFs). Under specific molecular beam epitaxy (MBE) conditions it is possible to produce an array of 90° dislocations that lie in the GaSb/GaAs(001) heterointerface. These dislocations efficiently relieve the strain between the two materials without generating the high density of threading dislocations that one would ordinarily expect. As a result, it is possible to grow high quality materials and active device structures above these IMF-based heterointerfaces.

This thesis describes our work to extend a modified version of this IMF technique to (111) surfaces in order to grow our InAs/GaSb QW structures on GaAs(111)A substrates. So far, this work has produced GaSb grown on GaAs (111)A with a fullwidth- half-maximum (FWHM) XRD peak value of 124’’. For GaSb/GaAs(001) grown via an IMF approach, other groups have reported FWHM values of 240’’.4 This work shows how various MBE parameters such as growth temperature, Sb overpressure, GaSb growth initiation and GaSb growth rate affect IMF formation. This thesis also reports initial electron transport measurements from InAs/GaSb QWs grown on GaAs(001) and (111) substrates via this IMF technique.