Learning Forum

[Kawshik Nath]

I've started to simulate perovskite-silicon tandem solar cell. But during simulation in fdtd i saw that using solar generation monitor i got about 22 mA/cm2 current for top perovskite cell and got only 0.5 mA/cm2 current for bottom silicon cell. That means no light is absorbed in the bottom cell but why this happened? I used the structure which is given in a paper. Please help me where is the problem and how i got that kinds of result. As the two cells are series connected so the overall current is 0.5 mA/cm2 and for that the cell efficiency is too low. What should i do now to overcome this problem? use ITO as a interconnected layer between these two cells.

I'll give the structure which i used to simulate.

[Guilin Sun]

You may  several monitors to mean the transmission in different material and see if the transmision is reasonable. If the top layers absorb most light then there is not much for the bottom.

What is the total thickness? if it is thick, you may need to have one time monitor at the bottom and check if the signal decays sufficiently small.

BTW: is the bottom the air, or it has a substrate? usually it should have a substrate. without substrate the resonance might be strong and does not match the real device.

since this is thin-film type, you can also use stackfield to get the field intensity:

https://optics.ansys.com/hc/en-us/articles/360034406294-stackfield-Script-command

then you can use script to postprocess the data.The total absorption within a unit cell is just to multiply the unit area. 

If you have questions for stackfield and postprocess please write a new post.

[Kawshik Nath]

The bottom cell is crystalline silicon a substrate. I used several monitors to see the transmitted light in the bottom cell and I found that most of the light is being absorbed by perovskite. Where is the problem here in nk data of perovskite? BUt I'd taken it from the refractive index.info website.

 

 

 

[Guilin Sun]

 

 

Do you meean below n-Si there is another substrate? If no, the structure is suspended in air. I am not sure if it is phyiscal device.

 It is hard for us to explain the device physics. As long as you use correct source (plane wave), periodic conditions, long enough simulation time and small enough autoshutoff min, as well as proper PML layers, the result should be ok. I only convern the substrate.

”Where is the problem here in nk data of perovskite? BUt I’d taken it from the refractive index.info website.”

FDTD depends on input material properties. However we cannot comment on the data you use. It might work since you may not be able to find another set of data. Its refractive index may not change much from different sources.

 

 

 

[sumit.20dr0152]

Dear Kawshik, how have you solved the above problem? I am also facing a similar problem in tandem solar cells simulation.

[Guilin Sun]

As long as you use correct materials, source, boundary conditions and simulation time/autoshutoff min, the result should be correct if it does not diverge.