Modeling of recombination lifetimes in charge-separation device structures
- PDF / 1,126,478 Bytes
- 5 Pages / 612 x 792 pts (letter) Page_size
- 108 Downloads / 217 Views
Z4.5.1
Modeling of recombination lifetimes in charge-separation device structures Jamiyanaa Dashdorj1,2, Richard Ahrenkiel1 and Wyatt Metzger1 1 National Renewable Energy Laboratory, 1617 Cole Blvd., CO 80401, U.S.A 2 Department of Physics, Colorado School of Mines, CO 80401, U.S.A ABSTRACT Recombination processes in semiconductors are inhibited by charge-separation mechanisms that are produced in structures such as pn junctions. In this work, we calculated the recombination rates in Si and GaAs n +p junction structures by both analytical and numerical techniques. Transport effects can be divided into three physical phenomena. The first is the flow of carriers that results in the build up of a junction voltage or charge-separation. The time constant of this effect is determined by the junction dopant concentrations and the mobilities. The second is the flow of carriers due to diffusion. The time constant of this event is a function of the carrier diffusion coefficient and physical width of the absorbing medium. In most semiconductor structures, this time constant is relatively short relative to the recombination lifetime. The third effect is the discharge of the open-circuit voltage of the junction by minority carrier charge injection. A heuristic analysis of the long-term decay in this regime indicates that the decay time varies as CkT /qJ 0 , where C is the junction capacitance and J 0 is the dark current. In this regime, the decay time is a function of recombination lifetime through the dependence of J 0 , but the measured excess carrier decay time is often not the true recombination lifetime of the underlying materials. † † † INTRODUCTION
†
Minority carrier lifetime is one of the most important parameters that affects the performance of semiconductor devices. Many lifetime techniques analyze the response of a device after a fast laser pulse by measuring a physical response, such as photoconductivity or photoluminescence that is proportional to the number of carriers in the device. Interpretation of these measurements becomes very complicated when a junction is present, making it difficult for a complete device processing sequence to be monitored with regard to contamination or defects using carrier lifetime measurements. We consider an n+p junction structure with a thin emitter where most of the light is absorbed in the base. Spatial doping profiles both in the emitter and in the base layers are assumed to be constant. Free carriers near the space-charge region drift from the emitter into the base and from the base into the emitter due to the electric field in the space-charge region. The corresponding charge-separation time will be discussed first. Because of the optical excitation, the generated minority carriers can diffuse into the bulk or can be extracted at the depletion edge. The diffusion transit time in the bulk of the junction device will be discussed second. Then the time required to restore equilibrium will be analyzed. Low-level injection and a one-dimensional configuration are assumed throughout
Data Loading...
