Blog Post

Is the 50% efficient solar cell on the horizon?

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

The summer sunshine is upon us, at least here on the East Coast, which means it’s the perfect time to think about how best to harness those golden rays for clean energy. The solar industry’s power players put their newest tech on display at last week’s Intersolar North America conference in San Francisco, despite the industry’s slightly depressed showing this year (see Ucilia Wang’s report from the trade show floor).

Nonetheless, despite these recent business shadows, there have been glimmers of research progress. At the conference Alta Devices, for example, presented technology that can boost solar cell efficiency to between 30 percent and an unprecedented 50 percent or more, using both materials and optical advances.

Best research photovoltaic cell efficiencies Rev. 06-2013
Best research photovoltaic cell efficiencies Rev. 06-2013

Besides the solar market slump, constraints imposed purely by physics have also been the bane of solar cell manufacturers. Many modern commercial cells hover around the 10-20 percent efficiency mark, and boosting efficiency is a major R&D focus for many companies. There are lots of reasons why solar cells can’t reach 100 percent efficiency: blackbody radiation (think of it as ambient energy evaporation), the PV materials used, and their capacity to accommodate electrons. Oddly, photons themselves can also be a roadblock to optimal efficiency.

Traditional solar cells are best at collecting only a narrow band of wavelengths, depending on their PV material, and are constrained by the Shockley-Queisser limit, which dictates that the maximum efficiency of an ideal solar cell can never exceed 33.7 percent. That’s because in traditional single-junction solar cells, the bandgap between two semiconductor materials defines how well the photons are converted into electrons within the cell. In this scenario, capturing photons whose energy is well-matched to the materials’ specific bandgap is crucial.

Alta Devices military 3

Fortunately, silicon and other semiconducting materials are pretty well-matched in their bandgaps to harness the sun’s natural spectral distribution. But spreading your photonic net across wavelengths could yield more energy-producing photons. That’s what multi-junction or tandem cells, like those from Alta Devices, aim to do.

In March, Alta Devices announced their solar cells had exceeded 30 percent efficiency, and at Intersolar last week founder and Caltech professor Harry Atwater outlined how the company plans to break the 50 percent efficiency barrier. Photon recycling and epitaxial lift-off (check out this earlier story for an explainer) are the two main factors distinguishing Alta’s cells, whose super thin gallium arsenide films are currently more efficient than traditional PV materials like silicon. They are, however, also more expensive, and may thus best serve niche markets where performance requirements trump cost. Alta is focusing on mobile deployments of its tech, from unmanned aerial vehicles to transportable solar arrays.

Alta Devices CEO Chris Norris shows a sample of solar cells. According to Atwater, simulations indicate that efficiencies in multi-junction cells can continue to increase, provided the structure of the cells is appropriately tweaked. The stacking of thin film layers and using tuned materials to cover the entire wavelength spectrum are some of the main considerations. By iterating and improving the PV design over these parameters, efficiencies of 50 percent or greater should be achievable.

Spectrum splitting – using optical methods to reflect and redirect incident light to appropriate layers – and using lenses or mirrors as concentrators are two accessory ways to further improve solar efficiency that Alta and others are pursuing. With light concentrators, the argument is that fewer solar cells are then needed, leading to potential cost and area savings. The snag is that you need a device, like the QBotix robot, to track the sun, and you need to funnel the energy of hundreds of suns into the system. Companies that use solar concentrators, like Solar Junction, have achieved over 40 percent efficiency with their cells in this way.

The future for solar efficiency is thus bright, in theory, but materials costs and technical hurdles related to manufacturing intricate multi-junction cells may keep these advances from being fully realized for the time being.

This post was updated at 6:30pm to clarify how traditional solar cells collect wavelengths.

4 Responses to “Is the 50% efficient solar cell on the horizon?”

  1. so then the question becomes, why has Alta laidoff or fuloughed most of their staff and the remaining folks are yield engineers.
    Plus, the question of how much capital will it take to scale up to 100 MW or 500 or 1 G ?
    these are fantastic objectives to put out there and great goals but the question of why this comes from a solar zombie company has to be asked ?

  2. Nice article. A bit confusing about the Shockley-Queisser limit. What happens in a single junction cell is that a cutoff energy per photon is needed to create an electron. Photons with less energy than the cutoff are list and any energy above the cutoff is lost for the high energy photons. So either you collect many photons and little energy per photon or few photons but a lot of energy per photon.

  3. Corey Carroll

    Here’s the thing: if you’re going to go through the hassel of adding tracking mechanisms for the X and Y axis (altitude and azimuth) why not just use cheap mirrors and go CST? The thing about CST is the energy storage is essentially “free” because it just makes so much sense to add it to increase the capacity factor of the turbine genset. Utilities are also a big fan of CST because it is dispatchable…. something you can’t get with PV or CSP unless you add batteries.