Nonisovalent Alloys

Visible wavelength optoelectronic devices, including solar cells, photodetectors, and light emitting diodes (LEDs), predominantly use isovalent compound semiconductors of the form (AB)y(CD)1-y where AB and CD are binary compounds of the same valency, meaning a III-V mixed with another III-V (e.g., GaAs-AlAs, GaP-InP), or a II-VI with a II-VI (e.g., HgTe-CdTe, CdTe-ZnTe). Combining these various elements in alloy form lends to characteristic material properties. However, a problem facing visible wavelength optoelectronics is that there are no isovalent alloys that lend well to the green part of the spectrum, around 550 nm. AlGaAs and AlGaInP work well from the red to yellow and GaN and InGaN work well from the blue to blue-green, but a gap exists (known as the “green gap”) between these various material systems. My interest, then, is in exploring outside of the traditional spread of isovalent alloys to include nonisovalent alloys. The primary difference is that the constiuent binaries of the alloy, (AB)y(CD)1-y, are not of the same type (e.g., ZnSe-GaAs, ZnS-GaP). Using these new alloys lends to a vast new spread of accessible material properties, many of which fit well within what is needed for the green gap. In my lab at UC Davis, we are in the process of refurbishing a molecular beam epitaxy (MBE) system for the sole purpose of growing ZnSe-GaAs alloys, which are particularly interesting due to their lattice match to commercially-abundant GaAs substrates across the entire compositional range.

The Bottom Line

Green is a nice color. We need alternative materials that make nice green light. Nonisovalent alloys have the potential to meet that need. We are building equipment with lots of knobs and tubes to make these alloys.