Solar Energy Conversion Efficiency of 47% Demonstrated

A new six-junction solar cell, developed by NREL (National Renewable Energy Laboratory), converts 47.1% of incident light into electricity when combined with optical concentration. Indications thus far suggest solar cells of this type can reach an efficiency rate of 50%.

143 sun concentration means they use mirrors or lens to focus light onto a smaller cell. So if it is a cubic inch cell then the mirror is about one square foot. 144 square inches. Instead of 24% conversion of sunlight to energy for regular solar panels they have about double. Part of the higher efficiency is getting the extra concentration of light. More brightness for a more expensive but higher efficiency multi-layer cell.

There are a number of obstacles to commercialization, though NREL believes that will probably be overcome in the near future. One of these is the presence of a resistive barrier inside the cell, which prevents the flow of a high percentage of current. This problem does not allow to achieve a 50% efficiency. Another obstacle to consider is the high cost for the production of the materials needed for the devices.

One way to lower costs is to reduce the active lighting area involved. You could, for example, use a mirror or concentrator to capture light and concentrate it on a specific point.

Solar concentration could reduce the amount of photosensitive material needed by as much as a factor of a hundred or even a thousand. It is well known that efficiency increases when light is concentrated. Prior to this, four-junction solar cells have demonstrated the highest solar conversion efficiency levels, but now, with the adoption of six junctions, the results have greatly improved. A further reduction in series resistance within this structure could realistically allow efficiency levels in excess of 50%.

Nature Energy – Six-junction III–V solar cells with 47.1% conversion efficiency under 143 Suns concentration

Single-junction flat-plate terrestrial solar cells are fundamentally limited to about 30% solar-to-electricity conversion efficiency, but multiple junctions and concentrated light make much higher efficiencies practically achievable. Until now, four-junction III–V concentrator solar cells have demonstrated the highest solar conversion efficiencies. Here, we demonstrate 47.1% solar conversion efficiency using a monolithic, series-connected, six-junction inverted metamorphic structure operated under the direct spectrum at 143 Suns concentration. When tuned to the global spectrum, a variation of this structure achieves a 1-Sun global efficiency of 39.2%. Nearly optimal bandgaps for six junctions were fabricated using alloys of III–V semiconductors. To develop these junctions, it was necessary to minimize threading dislocations in lattice-mismatched III–V alloys, prevent phase segregation in metastable quaternary III–V alloys and understand dopant diffusion in complex structures. Further reduction of the series resistance within this structure could realistically enable efficiencies over 50%.

36 thoughts on “Solar Energy Conversion Efficiency of 47% Demonstrated”

  1. If you’re getting more 2x more energy from 143x the sunlight… you’re *NOT* 2x more efficient. Your 72% less efficient.

  2. It is but now you have something collecting it so you might as well try and put it to good use. I was in Jamaica last year on vacation. Solar water heaters on many roofs. Not many solar power units. Maybe they are still too expensive.

  3. True, but at 1 sun concentration they still get 39%, which is rather good. It would be even better if they could produce this kind of panel out in space for Dan’s LSP, but it will probably be some centuries before we can do six-junction PV out of regolith…

  4. “under the direct spectrum at 143 Suns concentration”
    So do I need 143 sq m. of mirror to light up 1 sqm. of panel? With direct sun light? 
    As an aside, at present I am in New Zealand. Wellington the capital, had the grand total of 6 min of sunshine last week…. that’s what they said on the weather report last night.

  5. NREL has had this 6 junction cell on their best pv cell efficiency chart for a year now, indicating they achieved this in early 2019. sad that it took a year to get published.

  6. Solar + some wind + a lot of batteries could become the dominant energy supply eventually. (In high latitudes without reliable wind, they’d better have hydro or nuclear power though.)

    We probably won’t keep enough grid batteries to handle more than about 24 hours without solar – including the normal day-night cycle. So we’d be looking at a future that is much more affected by weather. For heavily clouded days, major electric energy consumers could shut down, keeping only ‘essential maintenance’ items like freezers powered.

    We’d probably have to do rolling black-outs – or more likely rolling ‘power hours’. If you need more continuous power, you’d better have your own battery or generator. A cloudy week in an area where that is unusual would be an emergency situation, with shelters opening for those who need assistance to survive.

    So – if that’s a future we are willing to accept, yep, let’s go all solar and wind. It wouldn’t be so different from when I was a kid in a rural area – ice or wind storms would knock out the power for most of a day several times a year. You can get used to it, keep candles and lanterns on hand, a supply of fuel for a fireplace or grill, etc.

  7. Applying this same research to thermophotovoltaics gets you some of that wasted thermal energy back by absorbing, and re radiating it as EM spectrum that can be converted into electrical current by the photovoltaic cells.

    Thermophotovoltaics are the ultimate in solar conversion efficiency – unless you can make a cell that just converts it all directly to electricity without requiring a thermal to EM stage.

  8. You are certainly right about the death rays. My theory is that after some countries start doing this, others will follow. Japan is already starting serious plans. We don’t quite yet have total Earth domination by the irrational.

  9. Space applications yes. No night to require storage.
    There are 2 factors that limit the usefulness of solar on earth. 1) night & clouds interrupting power production. 2) low power per unit area requiring large collector areas.
    Both of these are less of a problem for using solar in space. For 1) even in LEO where you are in earth’s shadow almost half the time, the time in shadow is only about 45 minutes, so the amount of storage needed is relatively small. For 2) aluminum foil concentrating mirrors can be really large cheap & flimsy since there is no gravity or wind to collapse them.

    I’m still unconvinced space solar for powering stuff on earth can work out. There are problems with power beaming. If you use microwaves to get through clouds, the required antenna size is km across. This *might* be OK for Gigawatt scale power plants, but I don’t see an advantage over nuclear.

  10. Oh, I agree we need it – but I remember the environmentalists fighting it the first time it was proposed in the ’70s-80s or so. Transmission of power to rectenna ground station arrays was characterized as the equivalent of the interior of a microwave oven at full power. Flock of birds fly in? Fried birds would drop out. And heaven forbid anyone get caught in it…

    (That, and other things like cost to get to orbit doomed the first iteration of this idea.)

    Unless you’re talking about really long extension cords, you’re going to have to beam the power via microwave or laser. The beam density for either of those will be a target for the environmental lobby. It will be blamed for weather changes, global warming, ozone hole depletion, atmospheric ionization, animal habitat loss, and anything else they can shove out to a gullible public. And don’t forget the pollution from launching all that mass – that’ll be a talking point also. It doesn’t matter how true it’ll be – it’s ALL about making sure it doesn’t come about.

  11. When using mirrors to concentrate light – If I am not mistaken , heat from concentrated light causes all sorts of problems, durability is decreased,…

  12. Solar concentrators, excessive waste heat management etc….
    I’m sure Stirling engines can have a role here and remedy the situation in a hybrid setup. The heat engine can use a good chunk of the waste heat and turn it into power. Concentrated solar and Stirling engine power generation has some sort of unofficial record for solar to grid real world conversion efficiency at 32%. This is for a parabolic dish commercial system you can actually buy.

    Check out Swedishstirling .com to see the state of the art in commercial stirling tech.

  13. The concentrated solar cells you see in (for example) the wikipedia article tend to have water cooling.

  14. Yes, they collect the sunlight and focus it onto a small solar cell.
    And yes, this needs direct sunlight to allow good focusing.

  15. Forced cooling is in order for any substantial solar concentration by mirror or lens. Seventy to One-hundred centigrade is a common range of operational temperature. Thermal fluctuation driven materials fatigue shortens the operational lifetime.

  16. How is the efficiency depending on the temperature .. High light concentration (100 or even 1000x) means super high temeratures — even if 50% is turned into electricity, a lot excess heat would reduce efficiency significantly .. Sorry, this makes me skeptical

  17. But how does that work “143 sun concentration”? Does one collect the sunlight from 143 square feet, to light up 1 square ft of CPV panel? How will it work if you don’t have direct sunlight on your concentrator?

  18. Solar definitely has a future, especially in space, but I don’t know that the future will mostly be solar. Nuclear has a great future too. If it weren’t for irrational fear and regulations, I think nuclear could be on a similar path to getting cheaper as solar has been. There’s still plenty of innovation to be had with nuclear at the molecular and atomic level that can make it significantly better than it already is. In the interim, wind and natural gas are quite capable as well. The future will be energy abundant.

  19. Concentrator + multijunction have been allowing over 40% for more than a decade, 46% was reached 6 years back. This will see the same kinds of limited usage(space + extreme niche) due to the cost.

  20. Pretty sure we need 20-200 TWe to make it worthwile for global heating purposes, 10 minimum then expand as needed. So we can go ahead directly to Starship ISRU instead of wasting time launching finished product. ISRU will be for any Space project, not just Space Solar. Criswell redirectors, perhaps the first thing to build along with rectennae, solve both ends of the variability problem, by allowing the matching of intermittant sources to irregular loads, needed for base load.

  21. John Mankins has some *ill informed* objections to Criswell LSP, unaware of the advantage of delivering the power anywhere desired. Mankins uses a single laser beam, I believe, and thinks *the* beam from the Moon will be too big. We need a “fresh look” (to borrow a phrase), at LSP using Starship pricing. The 20-200 TWe at 1 cent per KWh was before these savings.

  22. We already have done the experiment about *standard* radio power beaming by holding cell phones against our brain for quite a few total hours. Criswell LSP solves the light pollution problem by putting the cells on the Moon, altho that may be a little harder to use mirrors on. But not that bad, as the Moon has very little inclination to the Sun, so the mirrors can be fixed. Please consider the problems with NOT doing Space Solar, which includs slowing the depopulation of the Earth as it is returned to Nature.

  23. The book *The Case for Space Solar Power* does some detailed cost analysis. With pre-SpaceX prices they came to 15 cents/kWh for a 2GW sat. I plugged in Starship pricing and it came to under 5 cents/kWh.

  24. Isn’t this the same amount of heat that the area is already getting from the sun? MINUS the proportion turned into electricity?

  25. There are some great places to retire that are away from the maddening crowds. All of the conveniences but none of the noise and pollution.

  26. I believe any attempt to make a SPS will face so many environmentalist legal challenges that it’ll fail.

    You’re talking about beaming power through the atmosphere. Whatever method you use will need to be proven beyond a shadow of any doubt that it’s impossible to be redirected AND it’s totally safe for any critter that may pass through that beam.

    We won’t even talk about the passionate Dark Skies advocates complaining about how much light pollution such satellites would cause. There’s already a lot of heartburn over Starlink.

  27. The waste heat could be used for home heating, hot water, home cooling, refrigeration and snow removal/deicing. In arid regions or were water quality is bad the waste heat could be used to produce clean water.

  28. Yep, things are slowly but surely moving in favor of Solar Power Satellites.

    More conversion efficiency = less mass for equal power = less mass to orbit and thus, less cost of launch for a same power generation (or more power sats launched for the same money).

    And SPSs do have base load capability, whatever power level we need, we can assure we have it just by adding more. Plus whatever improvements SpaceX Starships make in cost per pound to LEO, they make SPSs even more feasible.

  29. The nice thing in space is that, because the incoming light isn’t diffuse, you can do all sorts of things that aren’t feasible in an atmosphere. Using dichrotic mirrors, you don’t need to stack junctions, you can just split up the light and use one or two junction cells exposed to the idea frequency range.

    I expect that large scale SPS, instead of just exposing their cells directly to sunlight, will extensively sort it, first; Besides efficiency, the cells can run cooler if they’re not exposed to light they can’t convert, and cool cells live longer.

  30. I think what you mentioned about houses with the conveniences of modern life but off-grid is vastly underestimated. I’m in the process of buying some rural ag land and building an off-grid home. It’s really not that expensive up-front especially when the land is so cheap.

  31. This is particularly good for Solar Power Sats, as free Space is the ideal location for large concentrating mirrors. “inner Solar System activities” includes the Earth’s electric and H power.

  32. The higher the efficiency the most likely the future will be -mostly- solar, and not by force or feel good reasons, but by sheer convenience.

    EVs that can feed themselves for the daily commute, houses in the great spaces with all the conveniences of modern life but fully off the grid and thus, placed wherever you like as long as the weather allows, all become possible with sufficiently good solar power.

    And the same applies for space applications, where solar energy is the energy source of choice for inner Solar System activities.

Comments are closed.