next up previous contents
Next: Mixed structures Up: Phototransductionthe Doorway to Previous: Quantum Efficiency of a

Quantum Efficiency of a Lateral Bipolar Photodetector

 

The structure of a lateral bipolar device is shown in Figure 7.9. Before analyzing the devices photoresponse we make a few simplifying assumptions. We assume that only the area between the emitter and collector diffusions, and the depletion regions are exposed to the light. Because otherwise, there will be a large contribution from the vertical bipolar components formed by p-diffusion/n-well/p-substrate. In reality the photogenerated electron-hole pairs will diffuse to other areas. We also assume that the effective depth of the device is only tex2html_wrap_inline7810 . Again there will be some currents diffusing through other areas.

Note that in the equations x denotes the horizontal axis and y the vertical axis. Also tex2html_wrap_inline7810 is the depth of the collector/emitter junctions.

   figure2112
Figure 7.9: The structure of a lateral bipolar detector in an N-Well CMOS process.

The diffusion equations in three regions can be written as:

eqnarray2120

The diffusion length in the collector and emitter regions is very short. Therefore, we make another simplifying assumption that these junctions extend four times the diffusion length in these regions. We set the origin at tex2html_wrap_inline7840 before the start of the emitter junction, to be able to reuse the derivations for the vertical bipolar transistor.

The boundary conditions for the three regions are:

eqnarray2149

eqnarray2181

The diffusion currents at collector and emitter are:

eqnarray2261

The drift currents in depletion regions of the collector and emitter junctions are:

eqnarray2284

The emitter and collector currents can be obtained by integrating the corresponding drift and diffusion components of each current over the range tex2html_wrap_inline7842 . Notice that the current density is per unit width of the device. These should be divided by the junction depth tex2html_wrap_inline7810 to yield a current density per unit area. The simulation result for a PNP device with minimum diffusion spacing ( tex2html_wrap_inline7846 ) in a 2 tex2html_wrap_inline7217 m CMOS process is shown in Figure 7.10. The general shape of the quantum efficiency is very similar to that of a lateral photodiode (Figure 7.6).

eqnarray2296

   figure2367
Figure 7.10: Simulated quantum efficiency of a vertical bipolar transistor in a 2 tex2html_wrap_inline7217 m CMOS process.


next up previous contents
Next: Mixed structures Up: Phototransductionthe Doorway to Previous: Quantum Efficiency of a

Alireza Moini,
Centre for High Performance Integrated Technologies and Systems (CHIPTEC),
Adelaide, SA 5005,
March 1997