DOE OSTI2012
All the milestones for this program were achieved and in most cases, particularly those related to cell performance, more samples were supplied than was required, and the deliverables substantially exceed the performance milestones. Moreover, the cell deliverables sent to NREL for evaluation were cells 4 mm or 5.5 mm, nearly 5 to 10 times larger in area than the 2 mm cells called for in the subcontract, and one of the most common sizes in the CPV industry. For the entire program there were 12 separate cell performance milestones, the most important of which was the 3J cell efficiency. Phase 1 had 2 3J efficiency milestones and Phase 2 had three. For the Phase 1 milestone calling for 38% efficiency, NREL measured a peak efficiency of 41.1%, about 2% higher than the data sheet values of commercial cells at that time. The efficiencies measured in Phase 1 exceed the Phase 2 cell performance milestones in the subcontract’s original Statement of Work. Cells submitted to NREL as part of this subcontract, although not as deliverables, were measured to have a new World Record cell efficiency of 43.5%, breaking the previous World Record by 1.2% absolute. In addition to improved cell performance, during Phase 1 a modeling capability for designing ARCs was developed and verified by comparing the measured current from a cell using a specific ARC with the current predicted by the model. This model was then used to design a new ARC giving lower reflectivity and a corresponding higher cell current. Again the prediction of the model was verified by measured cell properties. In Phase 2, the modeling capability was used to design a glass matched ARC, the type most in use in the CPV industry, which was tuned to the band structure of a 3J cell with a dilute nitride bottom cell. Finally, tools and test methodologies were developed that enable HALT to be applied to cells to determine the operating and destruct limits and to determine or set bounds on the lifetime of the cells in typical CPV operating conditions. As part of this effort a novel HALT system was built that enabled cells to be subjected to light at concentrations up to 80 W/cm 2 , (800 suns) while the cell temperature could be held or cycled between -100C and +180C. As part of the Phase 2 effort the system was modified to enable the light to the cell to be interrupted at a maximum rate of 10 times/second. Tests using interrupted light at >500 suns on cells held at 180C showed no evidence of failure after a total of >22 sun-years of operation at 100C. Another key outcome of the reliability efforts of the program was from high temperature storage tests done independently of the HALT test. There were two key outcomes of this activity: 1) single junction GaInNAs cells showed no evidence of failure under the equivalent of at least 100 years of on-sun time operating at 100C and 2) we can conclude with 95% confidence that the probability of device failure after 30 sun-years of operation at 100C is less than 1.5%, and mostly less than 0.5%. Based on these results there is no reason to believe that dilute nitrides in and of themselves pose a reliability problem for CPV applications and that 3J cells using dilute nitrides have very low probability of failure during the expected 20-30 year life of a CPV power plant.