Although solar cell modeling is quite well developed, the optimization of solar cells is much more challenging. This is partly due to the unavailability of derivatives, which are crucial to efficient high-dimensional optimization.
Developing new solar cells has generally been a tedious process of making small changes to one of these parameters at a time. Computational simulators can evaluate these changes without building each new variation for testing. Still, the process remains slow.
Now, scientists at MIT and Google Brain have developed a system that makes it possible to evaluate one proposed design at a time. At the same time, it gives information about which changes will provide the desired improvements.
This newly developed system is known as a differentiable solar cell simulator. Unlike conventional solar cell simulators, this new differentiable solar cell simulator predicts the efficiency of solar cells. It shows how much that output is affected by any one of the input parameters.
Research scientist Giuseppe Romano of MIT’s Institute for Soldier Nanotechnologies said, “It tells you directly what happens to the efficiency if we make this layer a little bit thicker, or what happens to the efficiency if we, for example, change the property of the material.”
“In short, we didn’t discover a new device, but we developed a tool that will enable others to discover more quickly other higher performance devices. Using this system, we are decreasing the number of times that we need to run a simulator to give quicker access to a wider space of optimized structures.”
“In addition, our tool can identify a unique set of material parameters that have been hidden so far because it’s very complex to run those simulations.”
MIT junior Sean Mann said, “The traditional approaches use a random search of possible variations essentially. With this new tool, we can follow a trajectory of change because the simulator tells you what direction you want to be changing your device. That makes the process much faster because instead of exploring the entire space of opportunities, you can follow a single path” that leads directly to improved performance.”
Advanced solar cells are composed of multiple layers. These layers are interlaced with conductive materials to carry electric charges from one to the other.
This newly developed tool shows how changing the relative thicknesses of these layers will affect the device’s output.
Mann explains, “This is very important because the thickness is critical. There is a strong interplay between light propagation and the thickness of each layer and the absorption of each layer.”
It can also evaluate other variables such as the amount of doping each layer receives, the dielectric constant of insulating layers, the bandgap, a measure of the energy levels of photons of light that can be captured by different materials used in the coatings.
Romano said, “This simulator is now available as an open-source tool that can be used immediately to help guide research in this field. It is ready and can be taken up by industry experts. To make use of it, researchers would couple this device’s computations with an optimization algorithm, or even a machine learning system, to rapidly assess a wide variety of possible changes and home in quickly on the most promising alternatives.”
As the new tool is end-to-end, it can compute the sensitivity of the efficiency by considering light absorption.
Currently, the simulator is based on just a one-dimensional version of the solar cell. In further studies, scientists will expand its capabilities to include two- and three-dimensional configurations.
Romano said, “An appealing future direction is composing our simulator with advanced existing differentiable light-propagation simulators to achieve enhanced accuracy.”
“Moving forward, because this is an open-source code, that means that once it’s up there, the community can contribute to it. And that’s why we are excited.”
- Sean Manna, Eric Fadel, Samuel S.Schoenholz, Ekin D.Cubuk, Steven G.Johnson, Giuseppe Romano. ∂PV: An end-to-end differentiable solar-cell simulator. DOI: 10.1016/j.cpc.2021.108232