Environment

Tandem silicon-perovskite solar cells could provide solar power on the cheap

View 2 Images
A new tandem silicon-perovskite monolithic solar cell could achieve high performance while keeping costs at bay (Photo: MIT)
A new tandem silicon-perovskite monolithic solar cell could achieve high performance while keeping costs at bay (Photo: MIT)
The two sub-cells absorb light on different parts of the spectrum (Image: MIT)

By combining silicon solar cells with their cheap and efficient perovskite-based counterparts, researchers at Stanford and MIT are creating a new type of "tandem cell" that could reach efficiencies up to 35 percent.

Out of all the solar cell technologies discovered so far, perovskite certainly takes the cake as the fastest-growing and one of the most promising for the future. In just a few years of development, perovskite-based cells have gone from efficiencies in the low single digits to a respectable 20 percent. The more traditional silicon based single-junction cells are still in the lead on the efficiency front, but progress has slowed to a crawl as their output approaches the theoretical maximum.

On the other hand, the quick pace at which perovskite cells are improving, together with cheap production costs and the fact they can even be sprayed onto windows and roofs, makes them a very interesting proposition.

Researchers Tonio Buonassisi, Jonathan Mailoa and team have now found a way to blend silicon and perovskite cells into a single "tandem cell" that could prove both cheaper and more efficient than current technology. The new device improves on the scientists’ previous work in which the two cells were simply stacked on top of each other and needed separate control circuits to function, which made them harder to build and operate.

The two sub-cells absorb light on different parts of the spectrum (Image: MIT)

In their new device the top sub-cell, made of the semitransparent perovskite, absorbs the higher-energy photons in the near-UV spectrum. The photons that aren’t transduced into electricity here travel toward the bottom of the cell through a silicon "tunnel junction" and to the bottom silicon cell, which absorbs some of the remaining light (mostly the lower-energy, near-infrared photons).

The new cell achieved a power conversion efficiency of 13.7 percent. This is poor compared to a standard perovskite or silicon cell, but it is mostly due to the fact that neither sub-cell was optimized for performance. Moreover, the amount of current produced in this design is limited by the capacity of the least efficient layer. The effects of this can be mitigated by matching the outputs of the two sub-cells as closely as possible. By doing so, the researchers believe the cell could achieve a 30 or even 35 percent efficiency.

The idea of using separate layers to absorb different parts of the spectrum isn’t new,  it’s the principle on which multijunction solar cells are based, but one important advantage to this new design would be cost. While multi-junction solar cells can reach efficiencies of over 40 percent, they often use rare, expensive materials that simply can’t scale. Manufacturing a tandem silicon-perovskite cell, however, should be relatively straightforward because the materials are cheap to source and the process would be similar to building standard silicon cells.

The researchers are now working on optimizing the cell’s power output by improving the quality of the perovskite sub-cell (currently the main bottleneck) and experimenting on how to make the perovskite composite, a combination of an inorganic material with a water-soluble organic material, more durable in the open air without sacrificing performance.

The advance is described in the journal Applied Physics Letters.

Source: MIT

  • Facebook
  • Twitter
  • Flipboard
  • LinkedIn
3 comments
grtbluyonder
With wonderous solar cell technology per the many announcements over decades comes the important qualifier "could" that never seems to turn into "does" . I designed satellite solar arrays 40 years ago. At that time pundits proclaimed that "soon", silicon monolithic cells would be available for 25 cents per watt. Like the kids say "are we there yet?". My advice, if you are waiting to implement one of the many modern miracle solar cells, you will be waiting a long long time so do something else.
Observer101
Why haven't I read about the "experiments" of utilizing solar HEAT, combined with other technologies to create electric, on a smaller scale? Maybe something like the old trick of starting a fire using a magnifying glass, focused on a piece of paper? But, rather than starting a fire, perhaps the HEAT generated by that type of system, focused on Voltaic cells (up close and personal), could increase efficiency to areas close to 80%????
dek0609
If you correct for inflation 25 cents in 1975 is $1.10 today. I can buy solar panels all day long for 65 cents/watt. In 1975 dollars that's 14.5 cents/watt. I power my cabin with solar power for cheaper than it would cost to just pay the monthly service fee. Are we there yet? yes.