A tool to accelerate the development of new solar cells | MIT News

In the ongoing race to develop ever better materials and configurations for solar cells, there are many variables that can be adjusted to try and improve performance, including material type, thickness, and geometric layout. Developing new solar cells has generally been a tedious process of making small changes to one of these parameters at a time. While computer simulators have made it possible to evaluate such changes without actually having to build each new variation for testing, the process remains slow.

Now, researchers at MIT and Google Brain have developed a system that not only evaluates one proposed design at a time, but also provides information on changes that will bring about the desired improvements. This could significantly increase the rate of discovery of new and improved configurations.

The new system, called a differentiable solar cell simulator, is described in an article published today in the journal Computing Physics Communications, written by Sean Mann junior at MIT, researcher Giuseppe Romano of the Institute for Soldier Nanotechnologies at MIT, and four others at MIT and Google Brain.

Traditional solar cell simulators, explains Romano, take the details of a solar cell setup and produce as the output a predicted efficiency – that is, what percentage of the energy in the incoming sunlight is actually converted into Electric power. But this new simulator both predicts efficiency and shows how much that output is affected by any of the input parameters. “It tells you directly what happens to the efficiency if we make that layer a little thicker, or what happens to the efficiency if we change the property of the material for example,” he says.

In short, he says, “we haven’t discovered a new device, but we have developed a tool that will allow others to discover other more efficient devices more quickly. Through this system, “we are reducing the number of times we need to run a simulator to give faster access to a larger space of optimized structures.” Additionally, he says, “our tool can identify a unique set of material parameters that have been hidden so far because it is very complex to run these simulations.

While traditional approaches essentially use a random search for possible variations, Mann says, with his tool, “we can follow a trajectory of change because the simulator tells you in which direction you want to change your device. This makes the process much faster because instead of exploring the entire space of opportunity, you can just follow one path “that leads directly to improved performance.

Since advanced solar cells are often made up of multiple intertwined layers of conductive material to carry electrical charge from one to the other, this calculator reveals how changing the relative thicknesses of these different layers will affect the output of the device. “It’s very important because the thickness is critical. There is a strong interaction between the propagation of light and the thickness of each layer and the absorption of each layer, ”says Mann.

Other variables that can be evaluated include the amount of doping (the introduction of atoms from another element) that each layer receives, or the dielectric constant of the insulating layers, or the bandgap, a measure of the levels of energy of light photons that can be captured by different materials used in the layers.

This simulator is now available as an open source tool that can be used immediately to guide research in this area, explains Romano. “It’s ready and can be picked up by industry experts. To use it, the researchers would couple the calculations of this device with an optimization algorithm, or even a machine learning system, to quickly assess a wide variety of possible changes and quickly focus on the most promising alternatives.

At this point, the simulator is based only on a one-dimensional version of the solar cell, so the next step will be to expand its capabilities to include two-dimensional and three-dimensional configurations. But even this 1D version “can cover the majority of cells currently in production”, explains Romano. Some variations, such as so-called tandem cells using different materials, cannot yet be simulated directly by this tool, but “there are ways to approach a tandem solar cell by simulating each of the individual cells,” Mann explains.

The simulator is “end-to-end”, explains Romano, which means that it calculates the sensitivity of the efficiency, also taking into account the light absorption. , to obtain increased precision.

Moving forward, says Romano, because this is open source code, “it means that once it’s up there the community can contribute to it. And that’s why we’re so excited. While this research group is “only a handful of people,” he says, anyone working in the field can now make their own improvements and improvements to the code and introduce new features.

“Differentiable physics will provide new capabilities for simulations of engineering systems,” says Venkat Viswanathan, associate professor of mechanical engineering at Carnegie Mellon University, who was not associated with this work. “The differentiable solar cell simulator is an incredible example of differentiable physics, which can now provide new capabilities to optimize the performance of solar cell devices,” he said, calling the study an “exciting step forward. “.

Besides Mann and Romano, the team included Eric Fadel and Steven Johnson from MIT, as well as Samuel Schoenholz and Ekin Cubuk from Google Brain. The work was supported in part by Eni SpA and the MIT Energy Initiative, and the MIT Quest for Intelligence.

Comments are closed.