Transcript Slide 1
How does diffusion affect radiative efficiency measurements? Caroline Vaughan and Tim Gfroerer, Davidson College, Davidson, NC Mark Wanlass, National Renewable Energy Lab, Golden, CO Motivation Radiative Recombination Conduction Band Conduction Band ENERGY - Defect Level HEAT Modeling Diffusion HEAT LIGHT + + Valence Band Valence Band Defect-related recombination occurs when an electron or hole is trapped in a defect energy level and recombines by releasing heat. This process takes away from the light produced by LED’s or the photocurrent generated by solar cells. Defect-related recombination can lower the efficiency of many semiconductor devices. We measure the radiative efficiency (the ratio of emitted to incident light) as a function of excitation laser power to investigate defect-related recombination in GaAs. Images of the emitted light reveal isolated dark regions where the radiative efficiency is reduced. We found a troubling result that radiative efficiency seems to be reduced by decreasing the laser excitation area. But our recombination model is incomplete because it does not account for diffusion or photon recycling. Our radiative efficiency measurements depend strongly on laser focusing, suggesting that these factors are important. We seek to model these effects computationally to explain how radiative efficiency depends on laser excitation area. 0 10 Normalized Photoluminescence Intensity Defect-related Recombination Abstract Focusing Analysis Experimental Setup 0 Radiative Efficiency 10 -1 10 -2 Even though the radiative efficiency should only depend on the density of electrons and holes (n), which is determined by the laser power per area, we find a large shift in the efficiency curve when changing the spot size of the laser. 2 4.30x10 cm -3 2 7.55x10 cm -4 2 5.21x10 cm -5 2 9.27x10 cm -6 2 7.32x10 cm -2 10 -3 10 -3 10 -2 10 -1 10 0 10 1 10 2 10 3 10 4 10 Why might this be? 2 -3 10 200 400 600 800 Distance From Center (m) By taking the integral shown above, which accounts for both laser width and diffusion length, we are able to model the predicted effects of diffusion1. But there seems to be a discrepancy between theory and experiment, especially near the laser center. By adding in a stronger generation rate due to photon recycling, we hope to find a better fit between the data and what we would theoretically expect. The recycling generation rate is equal to2: 1. “Direct imaging of anisotropic minority-carrier diffusion in ordered GaInP” N. M. Haegel, T. J. Mills, M. Talmadge, C. Scandrett, C. L. Frenzen, H. Yoon, C. M. Fetzer, and R. R. King, J. Appl. Phys. 105, 023711 (2009), DOI:10.1063/1.3068196 Photon Recycling Focused Photoluminescence Horizontal Profiles -6 2 at 7.32x10 cm 10 Photoluminescence Intensity -2 10 0 Diffusion Raditative Efficiency -1 10 2. “Measurement of 100 μm minority carrier diffusion lengths in p-GaAs by a new photoluminescence method” RJ Nelson, Inst. Phys. Conf. Ser. No. 45: Chapter 3 (1979) Power/Area (W/cm ) 0 Mathematica Diffusion Fit Experimental Data 0.011 mW 0.5216 mW 2.78 mW 32.4 mW 60.2 mW High Intensity Laser -1 10 Laser Excitation Low Intensity Laser Luminescence Absorption Event Recombination Event Electrons Holes Barrier 2m -2 10 1000 1200 1400 1600 1800 Distance (m) 100 μm At lower excitations, n is smaller and the diffusion length is longer, so the effective area of excitation is greater. There are shorter diffusion lengths for high excitation, so the effective area of excitation is smaller. Conclusions Radiative Efficiency at 298K 0 Radiative Efficiency 10 Efficiency curves are measurements of the radiative efficiency (emitted / absorbed light) as a function of laser power per area. -1 10 -2 10 Unfocused Defect NoDefect -3 10 -4 10 -2 10 -1 10 0 10 1 2 10 10 2 Power/Area (W/cm ) 3 10 • When the laser is focused on a defect, we observe a dramatic difference in the way that the radiative efficiency depends on laser power. • We observe a puzzling dependence of the radiative efficiency on laser spot size and hypothesize that it may be due to the diffusion of carriers and an added effect from photon recycling. •The diffusion could have a large effect for the small laser spot, but would have a negligible effect on the large laser spot. •With computer modeling we are working to match our experimental results to a theoretical hypothesis that would combine the effects of diffusion and photon recycling. Photon recycling is the re-absorption and possible re-emission of the photoluminescence emitted. For the focused laser, an emitted photon is more likely to move outside the small laser excitation area where n is small and re-emission is unlikely, while on the large laser spot, an emitted photon is more likely to be re-absorbed within the laser excitation area where n is large and re-emission is more probable. Re-absorption within the excitation area creates a higher net efficiency there, enhancing the photoluminescence. Acknowledgments We thank Jeff Carapella for growing the test structures and the Donors of the American Chemical Society – Petroleum Research Fund for supporting this work.