Deciding Periods for Solar Cell

spacelover17spacelover17 Member Posts: 10

I'm trying to simulate a solar cell using both FDTD and CHARGE, and I am confused about how to decide how many periods to simulate, and also what exactly constitutes a "period". For example, in the Si Solar Cell w/ TiO2 Pyramid Array example, there are 12 periods, but the x-span of the FDTD simulation is 0.75 um but the x-span of the CHARGE simulation is 10.1 um (which amounts to 13.466... FDTD periods in a single CHARGE period). Plus the solar generation object has an x span of 25 um. So I am quite confused.


  • gsungsun Posts: 751Ansys Employee

    Thank you for bring this interesting topic. In my understanding, the optical and the electrical periodicity governs different physics, so they are not necessarily directly linked. For optical simulation, the periodic BCs mean the unit has infinite duplicates in order to simplify the simulation; whereas the boundaries in CHARGE simulation mean there is no current passing through the boundaries. While it is mostly true for CHARGE simulation since there is an applied voltage to force the charges flow through nearest contacts, the end units in optical simulation will not have the same behavior as the center one. But this difference is quite small considering the simulation errors in both optical and electrical simulations, as well as manufacturing errors. In addition, the size of the contacts and the distance between the contact and the end optical unit can be arbitrary, so it is not necessary to have integer number of optical periods for electrical period.

    I agree that to make it looking nicer, using integer numbers of the optical periods is better for electrical simulation.

Sign In or Register to comment.