Clean Power

Published on December 3rd, 2014 | by Guest Contributor


New Solar Cell Efficiency Record Set At 46%

December 3rd, 2014 by  

Originally published on RenewEconomy.

A new world record for the conversion of sunlight into electricity has been established in Europe, after a multi-junction solar cell developed through a French-German collaboration achieved 46 per cent efficiency – up from 43.6%.

silex-CPV-dishes-300x186The record was achieved using a four-junction cell, developed by Soitec and CEA-Leti in France, together with the Fraunhofer Institute for Solar Energy Systems ISE in Germany as one of a new generation of multi-junction solar cells, developed specifically for concentrator PV plants, and expected to have an efficiency potential as high as 50 per cent under concentrated sunlight.

Each of the cell’s four sub-cells converts precisely one quarter of the incoming photons into electricity, thanks to precise tuning of the composition and thicknesses of each layer inside the cell structure.

The new record of 46 per cent efficiency – the cooperation’s second world record in a year – has been confirmed by the Japanese National Institute of Advanced Industrial Science and Technology, one of the leading centres for independent verification of solar cell performance results under standard-testing conditions.

“We are very proud of this new world record,” said Jocelyne Wasselin, vice president of solar cell product development at French semicomductor company Soitec.

“It confirms we made the right technology choice when we decided to develop this four-junction solar cell and clearly indicates that we can demonstrate 50 per cent efficiency in the near future.

“To produce this new generation of solar cells, we have already installed a line in France. It uses our bonding and layer-transfer technologies and already employs more than 25 engineers and technicians,” added Wasselin.

“I have no doubt that this successful cooperation with our French and German partners will drive further increase of CPV technology efficiency and competitiveness.”

Dr Frank Dimroth, project manager for the cell development at the German Fraunhofer Institute for Solar Energy Systems ISE describes CPV as “the most efficient solar technology today”, suitable for most countries with decent solar resources.

Reprinted with permission.

Complete our 2017 CleanTechnica Reader Survey — have your opinions, preferences, and deepest wishes heard.

Check out our 93-page EV report, based on over 2,000 surveys collected from EV drivers in 49 of 50 US states, 26 European countries, and 9 Canadian provinces.

Tags: , , ,

About the Author

is many, many people. We publish a number of guest posts from experts in a large variety of fields. This is our contributor account for those special people. :D

  • SirMefist

    where can I buy this PV cells?

  • Andre

    The bottom line question is NOT the price. The bottom line is the Energy Return on Energy Invested(EROEI). That is the most important piece of information that determines the feasibility of an energy resource. Solar has an average EROEI of about 5:1 as a very conservative estimate. That alone proves the resource is of value. The Energy balance of solar panels is about 3.5 years. Essentially, from that point on the Panels have “paid for themselves” and the return is all profit – in terms of energy. I would probably get flack about this but I could care less about the price put into building them as it is a completely arbitrary number. How much would the price of a kWh be in 10 years? If it’s going to be a lot larger, then I would most likely make bank if I invested in Solar since the begining.

    • Bob_Wallace


      Price is the point.

      And your ERoEI number is very badly flawed. (And I bet I know the POS paper where you got it.)

      Here’s the deal. A silicon solar panel pays back embedded energy in less than two years. A thin film solar panel pays back embedded energy in less than one year.

      A conservative estimate of panel life is 20 years. (Ridiculously conservative)

      We have lots of reports of 30 year old solar panels continuing to work just fine and the oldest installed array is 40 years old and going strong.

      ERoEI = Energy Out / Energy In

      Silicon panels
      Low end = 20 yr / <2 yr = 10+ ERoEI
      High end = 40 yr / <2 yr = 20+

      Thin film panels
      Low end = 20 yr / <1 yr = 20+
      High end = 40 yr / <1 yr = 40+

      ERoEi is an interesting number if one is dealing with energy inputs that are 1) finite and 2) running out. Oil is a place where ERoEI is important to consider. But we are not going to use up all the solar energy hitting the Earth's surface in a given day and our supply will hold out for a few billion years.

      What is important is price. Price is a composite number made of material, labor, energy and other costs that take us from raw materials laying around on the Earth to the final panel coming out the factory door.

      Manufacturers are now producing solar panels for about $0.50/watt. How much embedded energy can there be at that price? All the energy to mine, transport, refine the raw materials and to turn the aluminium, glass, plastic and refined silicon into a solar panel is paid for as part of that 50 cents.

      • Andre

        I think you misunderstood my argument. I was saying that even at the numbers I was discussing Solar is the logical path to take even if you took away the analysis with money.

        • Bob_Wallace

          Solar is one of the technologies we will use to replace fossil fuels.

          Cost is very definitely an issue. Had solar not become inexpensive we would be in a world of hurt.

          Obviously the logical path is to quit using fossil fuels, the cost of unlimited climate change is extremely high. But, in reality, much of the world would not pay for solar if it cost what it did ten years ago.

  • Pieter Siegers

    CPV sounds very promising!

  • CuriosityKilledTheCat

    …so the efficiency goes up when there is extra light hitting it, or it’s just able to make more electricity because there’s more light hitting it?

    • Bob_Wallace

      These cells are very efficient compared to the type we put on our roofs. (46% vs. ~18%) They are very much more expensive as well.

      Because they are expensive it makes more sense to feed them more sunshine (energy) rather than run them with the amount their surface area alone would collect (1 Sun).

      We can take a mirror with about the same surface area and use it to dump “1 more Sun” onto the surface of the cell. Now it can change the energy in 2 Suns into electricity at 46% efficiency.

  • Tom G.

    Well this is certainly a good thing but to me it is sort of cheating. Please let me explain.

    Concentrated solar takes some area of sq. meters of solar radiation and focuses that energy on a solar cell. Something like compressing air to put it into a smaller tank.

    While this is certainly important since less silicon is needed and higher efficiencies are important, its not really the breakthrough technology we need. What we need is a flat plate solar panel that captures and converts 45-55% of every proton that hits the surface into energy. We are getting better and some cells can reach 25-35% efficiency but we need MORE.

    So come on all you bright young minds our there; go to the lab and make something like this happen. Yes I know it takes time but at the age of 74 just how much longer do you think I can afford to wait, LOL.

    Have a great day everyone.

    • Bob_Wallace

      It’s all about trying to figure out the least expensive way to produce clean electricity. I’m not convinced concentrated solar will be the route but it’s interesting and has applications such as space where “installation” costs are very high.

    • timpster

      Or, instead of using so much power for light, why not just install skylights everywhere and need a whole lot less power while having MUCH better and natural light?

      I think it’s a great idea!

      • Ronald Brakels

        I don’t know what it’s like where you are, but rooftop solar is a much cheaper option for internal lighting than installing a skylight. Here a skylight plus installation is typically going to cost at least $600 Australian dollars and usually $1,000 or more. That’s to light one room. As part of a typical install, $1,000 will buy about two thirds of a kilowatt of rooftop solar capacity which will produce enough electricity to brightly light 10 rooms for 12 hours a day.

        • timpster

          Sometimes you gotta spend the big bucks to get real quality things in life. Can’t always go the “cheap” route.

          • Ronald Brakels

            Sounds like a credit card company slogan to me. Fortunately we are above such crass materialistic sentiment in Australia. Now if you’ll excuse me I have to go help my sister park her eight cars.

          • timpster

            It’s not about materialism, sometimes you gotta do a lot of work inside, and having that extra light is slightly (not a magnitude– OK maybe it is we actually don’t know yet) better for health, again, not about fancy things although they are certainly nice.

            Just like I use an “old” Incandescent with orange wrapped around it so I get slightly more sleepy instead of just using a white or slightly beige one. Sleep is really important.

            In fact I’ve been wanting to spread awareness of Nat Geo’s new thing they have called “Sleepless in America” #sleeplessinamerica is what you can use on social media, it airs December 7th, Sunday at 9 AM in the morning! Go tell your cable box to record it! (because we both know you won’t be up that early hahah)

    • Bill Kalahurka

      I’m not sure if you are completely understanding. This solar cell does convert 46% of all incoming light into electricity. Even if you don’t use any concentrators at all, it will be 46% efficient. In principal, you could make a bunch of 46% efficient solar panels using this technology. So, you might ask, why not do just that? Why bother with the concentrating mirrors? Answer: Cost. It would be extremely expensive to take this bleeding edge quadruple junction solar technology and make a bunch of regular solar panels out of it. The idea is that you can just make a small 46% solar cell, and then use mirrors (cheap) and concentrate a bunch of light on it, so you don’t need to several square meters of this solar cell to get a good bang for your buck.

      • Tom G.

        Yup this seems like the ticket; 46% efficiency and someday cheap enough to replace existing cells that are 15-22% efficient. Sounds like all we need to do is find ways to make them cheap enough to use without concentration or the need for tracking.

  • spec9

    That’s nice and they’ll be great for spacecraft. But what we really need is cheaper cells (although they are cheap already).

    • solarone

      Nice – yes I agree….this is great news. Science doing what it is supposed to do – pushing the boundaries of what is possible. The market will sort out how practical it turns out to be, but all involved should be proud of their achievement.

  • Bill Kalahurka

    Four-junction? Pretty darn impressive. I don’t think I have seen that before. Maybe I’m not paying enough attention.
    My question is this: Are we ever going to see multi-junction cells come down in price enough for them to become the norm in non-concentrating PV? Or will multi-junction forever only be relevant in CPV?

    • JamesWimberley

      There’s work going on now on “tandem cells”, with just two layers with different spectra. For instance, Oxford PV with a thin-film transparent perovskite layer on top of conventional silicon (link). They hope to have prototype cells next year. The aim is very definitely to get cells at a competitive price to mass-market silicon, with an initially modest 20% increase in efficiency (say from 20% to 24%). That’s not much seeing that pure perovskite lab cells are up to 20% by themselves.

      The conservatism and lack of large in-house research departments of the big Chinese manufacturers means, if I read the tea-leaves right, that researchers on new pv technologies have to do most of the work to solve manufacturing problems for them.

  • Marion Meads

    Morgan Solar could use this very efficient solar cells installed at the edge of their solar ray tracer collectors.

  • JamesWimberley

    The concentration was 508 suns. So you have a large array of expensive Fresnel lenses focussing light on a small, red-hot, super-engineered solar cell. C’est magnifique mais ce n’est pas la guerre.

    • Jenny Sommer

      I was thinking Randy Mills right away…
      Sorry for bringing this up but I love to read that sometimes.

    • Neptune

      Mirrors are inherently cheaper than PV, especially flat like fresnel.

      Concentrated PV may one day become cheaper than non-concentrated PV.

      • timpster

        You know what else is cheaper: skylights, and they put off MUCH better light than any artificial one (excluding fire, because fire is amazing).
        Then we wouldn’t need so much Photo Voltaic panels.

      • Shane 2

        Randy says he can generate power from producing a state of hydrogen he calls hydrinos. He hopes to use cells designed for CPV to turn the radiant energy produced into electricity. There are some videos on Youtube showing this process. The process does not currently have efficiencies required to compete with current energy technologies. He has raised over 90 million dollars over the last twenty years pursuing this technology. I wish him well.

    • Ronald Brakels

      Well the idea was to set up mirrors to reflect light onto the cell and when technology improves the cell could be replaced with a more efficient one. It might have panned out for utility scale solar. At the time it wasn’t exactly an unreasonable idea. And the replacable cell system is still being used as in the picture above. Perhaps as electricity prices crater in the middle of the day, tracking system which favour more efficient concentrating PV will gain in popularity. Or not. I wouldn’t bet the farm on it. Maybe an injured wallaby. Definitely a disorientated green ant, but then I’d want to get rid of that anyway before it orientated itself.

      And the concentrated PV cells can’t get hot. That’s one of the drawbacks. They have to be cooled. If the cooling fails too far the cell can be cooked. Of course for tracking systems it’s generally just a matter of, “Don’t look at the sun!”

  • Larmion

    Why is there a picture of a solar stirling in an article about PV?

    • Marion Meads

      Maybe they will convert the stirling engine in focal point to use these new super efficient cells…

      • Joseph Dubeau

        No, that’s a picture from a different project.
        These solar cells are more than like used in with concentrated solar.
        They are expensive.

        • Marion Meads

          That’s why the solar concentrator used in solar Stirling engine is shown. The key is concentrating the sunlight, and it can be done many ways. And of course, no one can stop anybody from reusing the structures of failed projects.

      • Shane 2

        That picture is NOT of stirling engine systems. They are Silex concentrator PV units installed in Saudi Arabia. Silex is an Australian company. Google “Silex solar” using Google Images.

      • Steven F

        I did a google search for the image and found this

        The picture isof a Concentrating Photovoltaic (CPV) Solar Power Station located in Mildura, Victoria.

        Any design sutable for a stirling engine is also sutable for consentrated PV. The only visible difference between the two is that the sterling engine is larger in size than the concentrated PV cell.

    • Ronald Brakels

      They look like Silex concentrating PV to me:

      “Silex solar, the Australian concentrating solar system so good, we’re trying to flog it off to concentrate on uranium enrichment. Or we were until General Electric and Hitachi dropped us because demand for uranium isn’t exactly increasing. Now I don’t think we know what we’re doing.”

      • Ronald Brakels

        The picture above would almost certainly be the one megawatt demonstration plant in Saudi Arabia.

        • Joseph Dubeau

          A lot of these pictures are mess up.
          I saw one that had fake smoke stacks.

  • Omega Centauri

    CPVis still a longshot. Generally cost of the PV chip isn’t dominant for high concentration CPV, optics and cooling don’t come free however. My guess is low to moderate efficiency panels will win the cost race.

  • timpster

    Instead of getting MORE energy from the sun, to use artificial light; how about we USE the sun for light.

    Why don’t most buildings have skylights and lots of windows like VERY FEW (and quickly increasing) stores like Home Depot, Sports Academy, and other similar areas?
    Also why the fuck is it not FORCED BY REGULATION to have skylights in schools, that’s where I think we really messed up.

    Fluorescent lighting is a piece of junk compared to the sun, so why not just use THE SUN!

    • David in Bushwick

      I agree with you completely. We block the sun to add lights.
      But please stay away from gasoline.

      • timpster

        I don’t like this, why do you want to block the sun so badly, why not just work with it like a REAL man?

    • fuu

      Buildings would be less isolated so the energy we saved would be spent on heating.

      • Najeeb Ullah

        agreed. we could use more of the sun direcly

      • timpster

        This is not true, see the discussion on the page, they discussed that with another commenter! Heat is NOT a problem! What else you got?

    • Marion Meads

      Except that with skylights, you have to deal with excess heat load during the summer or loss of heat during the winter. Our city has regulations for the proportion of double paned windows and skylights used in residential and commercial buildings.

      • Bob_Wallace

        That can be minimized. My skylights are dual pane, argon-filled, and have a heat reflecting (essentially invisible) film installed on their interior.

        • timpster

          Well this is great! Wouldn’t stores (like “gas station” stores) and small grocery stores look SO much better with the sun shining in instead of fluorescent lights?

        • Ronald Brakels

          That sun you people seem to like so much? Well let me tell you something about it. It tried to take my father’s gosh darned ear! Skin cancer isn’t caused by skin you know. And you want to go inviting that sun into your homes? Let me tell you, that sun belongs in a prison! It belongs in a cell! But I’ll get my revenge. That’s why I’m encouraging the use of solar cells everywhere! When my plans come to fruition the sun will be imprisoned in cells on almost every roof in Australia! And then the world! Mwah-ha-ha-ha-haa!

          • Bob_Wallace

            Pieces of our ears are the required sacrifice to the Great Lord our Sun.

            I gave. Have you?

          • Ronald Brakels

            Of course I haven’t given! You don’t think I live underground for my health, do you?

            No wait, that’s exactly why I live underground. That and the fact I am not allowed to live like normal people.

            But I will never give my skin! Never, do you hear? Never! I keep it all in a box! Guarded by a hideous three headed dog with three bodies. Now where’s my manservant? Al Bino! Bring me my epidermal repository! I have some lovely new flakes to add.

      • Larmion

        Heat loss: not really, the heat conduction of modern glazing work (multiple layers, inert intermediary) is barely higher than that of the most insulating walls, and there is a small amount of passive heat gain from the sun.

        Heat gain is a bigger one. Reflective films are a partial solution, but there is still a lot of heat gain. Fortunately, there is a cheap and simple solution: blinds or something similar. Sure, you lose the light during the hottest part of the year. But in most climates, daylight during the cooler parts of the year is already enough for a huge amount of energy savings.

      • Marion Meads

        Yes, there are several ways of dealing with the additional heat load from the windows and skylights during the summer. Some use prismatic approach so that more lights and radiation load are reflected out from the angle of the sun made during the summer, and to take in more radiation from the lower angle of the winter.

      • Kevin McKinney

        In warmer climates, putting your skylight on the north slope of a gable roof works great–we did that for our Atlanta home. Exterior shading can have a similar effect.

      • timpster

        Heat…. just go outside and see how hot it is when this becomes a problem, but I have south facing windows which adds at **LEAST** 10 degrees warmth to my room, and I love it!

      • Gerald Katz

        California energy code has a formula for number of required skylights for industrial roofs, new high bay lighting systems like HOT 5 fluorescent can be ordered equipped with daylight sensors to automatically turn off when sufficient sunlight is available. Solatube and similar daylight devices can have led lamps inside for night illumination.

    • harisA

      If we are really going to force people by regulation, why do we not just say that all new construction should have 1-2Watts of PV for sq-ft of living or office space etc.

      That being said, using govt. stimulus funds, schools in our area are getting skylights as well as PV. My son’s primary school got about 75kw (150W/student) and the high school behind our house got about a MW (400W/student) of solar car ports. I believe the High School project which completed in summer 2011 also got some rebate from the State of CA.

      So, progress in happening slowly with a lot of carrot and not enough stick:-)

      • timpster

        Alright well that’s great!

      • Larry

        Do those schools have any data on how much money they saved (or made) on reduced electricity purchases? My guess is it’s substantial

    • Building insulation was sub standard for years; too inefficient to compensate for skylights.
      Industrial units seldom have any insulation , just a steel roof with tar and gravel.
      Its scary, anything at eye level over analyzed and stuff in the ceiling and basement, not.
      But then you have to look at buildings from the new age seventies, second class real esate now.

    • Gerald Katz

      In California all new large buildings with high ceilings have to have a certain number of skylights and usually high efficiency HOT 5 fluorescent lamps with daylight sensors are installed to maximize energy efficiency.

      • timpster

        Awe man I **HATE** fluorescent lights, they suck so much, why don’t they move to Philips warm white LED? It’s so much better!

      • timpster

        I’ll say FUCK fluorescent, LED because of VASTLY better color spectrum and higher (standard – not 4200k) color temperatures.

    • flyfaster

      Why not clean up your language!

  • Ken

    NASA is glad.

    • Omega Centauri

      But does it perform well without concentration?

  • David in Bushwick

    “It uses our bonding and layer-transfer technologies and already employs more than 25 engineers and technicians…”
    How many scores of engineers and technicians worked on the new iPod headphones?
    Cheap, glue-on panels will be the norm.

  • Marion Meads

    I hope the cost of production per Watt has reduced. But of course, instead of reducing, it could have doubled the price. There are applications for this, especially when space area becomes a premium such as in very expensive urban cities.

    • Shane 2

      These types of cells really only make sense in concentrating PV applications. CPV can only compete with conventional panels in very sunny locations.


      Well, with such high efficiency rate, you need less panels, therefore I doubt the price per panel will be anywhere close to what a panel costs now..

    • Gerald Katz

      Concentrating pv would have more value if the heat is also used for commercial industrial processes or to run a thermal engine for additional power.

      • Pieter Siegers

        That’s the other 50% then.

Back to Top ↑