Solar energy represents a vast potential clean, cheap energy. Harnessing this vast source of energy in an economical way such that it can compete with conventional fossil fuels is not a trivial matter. Current state-of-the-art solar cells are triple junction devices based on GaP/InGaAs/Ge or InGaP/GaAs/InGaAs technologies, with conversion efficiencies of over 40%. To increase the conversion efficiency over 50%, it will be necessary to increase the number of cells in multijunction solar cells and to find materials with larger band gap energies than is currently attainable in the traditional III-P or III-As material systems. The III-N material system (including alloys of InN, AlN, and GaN) has several characteristics which give it key advantages over the existing solar cell materials. One advantage is the wide range of band gap energies available in III-N materials, which range from <0.7 eV for InN to 6.2 eV for AlN. This energy range spans nearly the entire solar spectrum allowing for design of devices whose absorption is matched to the solar spectrum for optimum efficiency. The III-N materials are also direct band gap and thus have exceptionally high absorption coefficients on the order of 1x105 cm-1 for GaN, which is an order of magnitude greater than for GaAs and several orders of magnitude higher than for an indirect band gap material such as silicon.

While III-Ns are appealing, there are challenges for device design which are unique to this material system, namely: polarization charges at hetero-interface; high defect density; and significant lattice mismatch. This dissertation covers our work on the design, fabrication and characterization of III-N based solar cells. The unique design challenges are outlined and our progress on improving InGaN device performance is covered.

We demonstrate p-i-n devices with fill factors of 75-80 %, V OC of greater than 2 V, and peak external quantum efficiency (EQE) of over 80%. MQW solar cells are reported with EQE response beyond 500 nm and VOC of 2 V. Positive thermal power coefficients are also reported for both p-i-n and MQW solar cells. These results highlight the potential for future III-N based solar cell progress.