3D Solar Cell simulation

solarex20.in : 3D Solar Cell simulation

Requires: Victory Process, Victory Mesh, Victory Device
Minimum Versions: Victory Process 7.30.4.R, Victory Mesh 1.4.6.R, Victory Device 1.14.1.R

By default Victory Process and Device run on just one processor. To ensure better perfomance on your computer the following simulation condition simflags="-P all" could be specidied in the go line starting Victory Process or Device. This means that all processors available will be used. If you want to use a smaller number of processors you can substitute "all" with a desired number, e.g. simflags="-P 4".

This example compares simulation results of a 3D solar cell with and without lenses.

This example demonstrates:

  • 3D process simulation of the solar cell using Victory Cell.
  • 3D delaunay mesh generation of the 3D Solar cell structure.
  • 3D device simulation of IV characteristics using Victory Device.
  • Display of the results in TonyPlot and TonyPlot 3D

The 3D process simulation starts from a mask and consist of a deposition and angled etch of an oxide layer on top of silicon in order to define oxide lenses. The shape of the lenses can be adjusted as a function of mask size and angle used during the etch. At the end of the process simulation a 3D structure is saved and meshed using a full 3D delaunay mesh. 3D Refinement of the mesh is done as a function of Net Doping.

The 3D structure is then pass to Victory Device where optical and electro-optical simulations are performed.

3D Ray-tracing is used during the simulation. This is a method by which beams of light are traced through a structure, taking into account reflection, refraction and attenuation. The calculated light intensities and absorptions are then used to calculate the photogeneration rates. Whenever a ray encounteres a region interface, or the incident ray encounters a device boundary, the ray is split into a reflected and a transmitted beam. For complex structures (especially in 3D) this can result in a large number of rays. Fortunately, ray-tracing is fast, can fully take advantage of parallel processing, and only needs to be done once. The number of rays produced by this algorithm is automatic and is the minimal number of rays required to fully resolve a given structure.

The ANGLE (or PHI) parameter specifies the direction of propagation of the beam relative to the x-axis. In 3D, you may also specify the angle THETA , which is the rotation about the y-axis. By default the beam will automatically be oriented and sized to illuminate the entire device from the top (i.e. stright down). It is extremely important to make sure that the origin window of the beam is outside of the structure. Otherwise, you get incorrect results.

It is not always easy to make sure the beam is defined correctly. In the runtime output of deckbuild (lower window of deckbuild), during the first solve B1= some information about the beam direction is given to the user:

Ray Tracing Beam 1: Origin(0.5,0.5,-1), Direction(0,0,1)

In this case it means that we have a vertical illumination from the top.

The optical and electrical simulations are done using an iterative multi-threading domain decomposition based solver called PAM.MPI . Combined with a multi-threading ray tracing algorithm, the electro-optical simulation is very fast. This example takes only few minutes to run on a multi-cpu machine.

Optical simulation reveals that reflectivity (absoption) is higher (lower) without lenses as expected. It explains why when doing an electro-optical simulation Jsc is higher with lenses. It can also be confirmed by examining the rays and photogeneration rate in TonyPlot3D .

To load and run this example, select the Load button in DeckBuild > Examples. This will copy the input file and any support files to your current working directory. Select the Run button in DeckBuild to execute the example.