new record for solar cells efficiency

[dragon]

New world record for solar cell efficiency at 46%
French-German cooperation confirms competitive advantage of European photovoltaic industry

Press Release 26/14, December 1, 2014
Bernin, France and Freiburg, Germany



New record solar cell on a 100 mm wafer yielding approximately 500 concentrator solar cell devices. ©Fraunhofer ISE/Photo Alexander Wekkeli

IV characteristics of the new 4-junction solar cell with an efficiency of 46% at 50.8 W/cm2 which corresponds to a concentration ration of 508 times the solar AM1.5d (ASTM E927-10) spectrum.

A new world record for the direct conversion of sunlight into electricity has been established. The multi-junction solar cell converts 46% of the solar light into electrical energy and was developed by Soitec and CEA-Leti, France, together with the Fraunhofer Institute for Solar Energy Systems ISE, Germany. Multi-junction cells are used in concentrator photovoltaic (CPV) systems to produce low-cost electricity in photovoltaic power plants, in regions with a large amount of direct solar radiation. It is the cooperation’s second world record within one year, after the one previously announced in September 2013, and clearly demonstrates the strong competitiveness of the European photovoltaic research and industry.

Multi-junction solar cells are based on a selection of III-V compound semiconductor materials. The world record cell is a four-junction cell, and each of its sub-cells converts precisely one quarter of the incoming photons in the wavelength range between 300 and 1750 nm into electricity. When applied in concentrator PV, a very small cell is used with a Fresnel lens, which concentrates the sunlight onto the cell. The new record efficiency was measured at a concentration of 508 suns and has been confirmed by the Japanese AIST (National Institute of Advanced Industrial Science and Technology), one of the leading centers for independent verification of solar cell performance results under standard testing conditions.

A special challenge that had to be met by this cell is the exact distribution of the photons among the four sub-cells. It has been achieved by precise tuning of the composition and thicknesses of each layer inside the cell structure. ”This is a major milestone for our French-German collaboration. We are extremely pleased to hear that our result of 46% efficiency has now been independently confirmed by AIST in Japan”, explains Dr. Frank Dimroth, project manager for the cell development at the German Fraunhofer Institute for Solar Energy Systems ISE. “CPV is the most efficient solar technology today and suitable for all countries with high direct normal irradiance.”

Jocelyne Wasselin, Vice President Solar Cell Product Development for Soitec, a company headquartered in France and a world leader in high performance semiconductor materials, says: “We are very proud of this new world record. 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% efficiency in the near future.” She adds: “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. I have no doubt that this successful cooperation with our French and German partners will drive further increase of CPV technology efficiency and competitiveness.”

About Soitec
Soitec is an international manufacturing company, a world leader in generating and manufacturing revolutionary semiconductor materials at the frontier of the most exciting energy and electronic challenges. Soitec’s products include substrates for microelectronics (most notably SOI: Silicon-on-Insulator) and concentrator photovoltaic systems (CPV). The company’s core technologies are Smart Cut™, Smart Stacking™ and Concentrix™, as well as expertise in epitaxy. Applications include consumer and mobile electronics, microelectronics-driven IT, telecommunications, automotive electronics, lighting products and large-scale solar power plants. Soitec has manufacturing plants and R&D centers in France, Singapore, Germany and the United States. www.soitec.com

About CEA-Leti (France)
By creating innovation and transferring it to industry, Leti is the bridge between basic research and production of micro- and nanotechnologies that improve the lives of people around the world. Backed by its portfolio of 2,200 patents, Leti partners with large industrials, SMEs and startups to tailor advanced solutions that strengthen their competitive positions. It has launched more than 50 startups. Its 8,000m² of new-generation cleanroom space feature 200mm and 300mm wafer processing of micro and nano solutions for applications ranging from space to smart devices. Leti’s staff of more than 1,700 includes 200 assignees from partner companies. Leti is based in Grenoble, France, and has offices in Silicon Valley, Calif., and Tokyo. www.leti.fr

About Fraunhofer ISE
With a staff of 1300, the Fraunhofer Institute for Solar Energy Systems ISE, based in Freiburg, is the largest solar energy research institute in Europe. Fraunhofer ISE is committed to promoting energy supply systems which are sustainable, economic, safe and socially just. It creates the technological foundations for supplying energy efficiently and on an environmentally sound basis. To this end, the institute develops materials, components, systems and processes for energy efficiency, energy conversion, energy distribution and energy storage. The Institute is a member of the Fraunhofer-Gesellschaft, Europe’s largest application-oriented research organization. www.ise.fraunhofer.de

link

[salm]

We are going to be told that this is useless junk because lol no storage.

[Nova Andromeda]

We are going to be told that this is useless junk because lol no storage.

Can someone explain to me why cracking water into hydrogen and oxygen isn't a perfectly good way to store the energy? Is 18% efficiency not worth it? Are the costs too high (even assuming bulk manufacture?)? Here is a Solar Energy Map. Average daily power usage in the US is 30 kWh (US .gov source). Given 18% efficiency, you'd need ~35m2 (6mx6m) of sun light to power your house for 1 day off of H2 generated from these solar cells. Of course, you don't need to power your house on stored energy all the time so won't need that much sun light area as direct solar power if 2.5x more efficient.

Final efficiently calculation:
0.48 (solar cell) * 0.75 (H2&O2 generation*) * 0.5 (fuel cells?) = 0.18
* (Does keeping and reacting pure 02 help with this efficiency? I just pulled the 1st Google thing I found.)

Costs:
No idea what the three components above would cost if manufacturing was ramped up to the level of solar cells.

[aerius]

Actually the problem is it requires highly concentrated sunlight to work, according to the article it needs 508 times normal sunlight. The other issue is it's a quad junction cell, ain't easy to make and will always be far more expensive than conventional cells. We've actually had high efficiency multi-junction cells in the lab stage for well over a decade now, they all require concentrated sunlight and none of them are cheap since they all require complicated fabrication methods and/or expensive rare earth metals. It's an interesting technology but it's not something that's going to revolutionize solar power.

[Nova Andromeda]

Actually the problem is it requires highly concentrated sunlight to work, according to the article it needs 508 times normal sunlight. The other issue is it's a quad junction cell, ain't easy to make and will always be far more expensive than conventional cells. We've actually had high efficiency multi-junction cells in the lab stage for well over a decade now, they all require concentrated sunlight and none of them are cheap since they all require complicated fabrication methods and/or expensive rare earth metals. It's an interesting technology but it's not something that's going to revolutionize solar power.

If we sum the costs of this and the necessary mirrors (maybe a parabolic dish like they use with solar thermal?) would the total costs be comparable? We'd only need a solar cell covering a ~0.07m2 area if we use the 35m2 total sun light area figure from above.

[madd0ct0r]

We are going to be told that this is useless junk because lol no storage.

Can someone explain to me why cracking water into hydrogen and oxygen isn't a perfectly good way to store the energy? Is 18% efficiency not worth it? Are the costs too high (even assuming bulk manufacture?)? Here is a Solar Energy Map. Average daily power usage in the US is 30 kWh (US .gov source). Given 18% efficiency, you'd need ~35m2 (6mx6m) of sun light to power your house for 1 day off of H2 generated from these solar cells. Of course, you don't need to power your house on stored energy all the time so won't need that much sun light area as direct solar power if 2.5x more efficient.

Final efficiently calculation:
0.48 (solar cell) * 0.75 (H2&O2 generation*) * 0.5 (fuel cells?) = 0.18
* (Does keeping and reacting pure 02 help with this efficiency? I just pulled the 1st Google thing I found.)

Costs:
No idea what the three components above would cost if manufacturing was ramped up to the level of solar cells.

close.
hydrogen is a bugger to store since the atoms are so small it leaks through most things, and has other nasty effects like embrittlement of stainless steel. One trick I've been investigating is the use of hydrogen as a feedstock for making methane, which is easier to store.

Unless you are storing the hydrogen at atmospheric pressure, there would be further efficiency lost in the energy used to compress they hydrogen during storage. You should be able to figure out the upper size of the storage tank you might need fairly easily.

[Nova Andromeda]

We are going to be told that this is useless junk because lol no storage.

Can someone explain to me why cracking water into hydrogen and oxygen isn't a perfectly good way to store the energy? Is 18% efficiency not worth it? Are the costs too high (even assuming bulk manufacture?)? Here is a Solar Energy Map. Average daily power usage in the US is 30 kWh (US .gov source). Given 18% efficiency, you'd need ~35m2 (6mx6m) of sun light to power your house for 1 day off of H2 generated from these solar cells. Of course, you don't need to power your house on stored energy all the time so won't need that much sun light area as direct solar power if 2.5x more efficient.

Final efficiently calculation:
0.48 (solar cell) * 0.75 (H2&O2 generation*) * 0.5 (fuel cells?) = 0.18
* (Does keeping and reacting pure 02 help with this efficiency? I just pulled the 1st Google thing I found.)

Costs:
No idea what the three components above would cost if manufacturing was ramped up to the level of solar cells.

close.
hydrogen is a bugger to store since the atoms are so small it leaks through most things, and has other nasty effects like embrittlement of stainless steel. One trick I've been investigating is the use of hydrogen as a feedstock for making methane, which is easier to store.

Unless you are storing the hydrogen at atmospheric pressure, there would be further efficiency lost in the energy used to compress they hydrogen during storage. You should be able to figure out the upper size of the storage tank you might need fairly easily.

The volume at one Atm is about (2.5m)^3. Also, the auto industry may have solved many of the problems you've raised in their pursuit of hydrogen powered cars (link).

Interestingly, 1 gallon of gas is basically ~33kWh which is about the same as the daily household power usage. One could just double the size of their solar installation and get both free electric and ~100 miles worth of fuel (2 kg H2 / day + quick Google of ~50+ miles / kg H2 = 100 miles).

[Simon_Jester]

hydrogen is a bugger to store since the atoms are so small it leaks through most things, and has other nasty effects like embrittlement of stainless steel. One trick I've been investigating is the use of hydrogen as a feedstock for making methane, which is easier to store.

Although then you have to burn the methane and you have a carbon dioxide problem?

Or am I confused?

[Terralthra]

Presumably you'd be fixing atmospheric carbon to make the methane in the first place.

[aerius]

If we sum the costs of this and the necessary mirrors (maybe a parabolic dish like they use with solar thermal?) would the total costs be comparable? We'd only need a solar cell covering a ~0.07m2 area if we use the 35m2 total sun light area figure from above.

Last time I checked the cost of conventional solar cells was about $2 per watt, at a typical 15% or so efficiency that's about $300/m2 so call it $10,000 in round numbers. The high end multi-junction cells are $100,000/m2 or thereabouts, no, that's not a typo, it really does cost that much. Considering that this is a lab stage proof of concept it likely costs a hell of a lot more, but let's use $100k for now. That brings the panel cost to $7000, but then we need to add in the cost of the reflector plus the cost of the cooling system* for the solar cells, neither of which is going to be cheap.

*35kW on a 1' diameter circle is serious business. About half it gets turned into electricity so let's say 18kW of waste heat, the large heating element on my electric range is about the same size but it's only 2600W. Keeping those cells from burning up takes serious cooling.

[Korvan]

We are going to be told that this is useless junk because lol no storage.

Can someone explain to me why cracking water into hydrogen and oxygen isn't a perfectly good way to store the energy? Is 18% efficiency not worth it? Are the costs too high (even assuming bulk manufacture?)? Here is a Solar Energy Map. Average daily power usage in the US is 30 kWh (US .gov source). Given 18% efficiency, you'd need ~35m2 (6mx6m) of sun light to power your house for 1 day off of H2 generated from these solar cells. Of course, you don't need to power your house on stored energy all the time so won't need that much sun light area as direct solar power if 2.5x more efficient.

Final efficiently calculation:
0.48 (solar cell) * 0.75 (H2&O2 generation*) * 0.5 (fuel cells?) = 0.18
* (Does keeping and reacting pure 02 help with this efficiency? I just pulled the 1st Google thing I found.)

Costs:
No idea what the three components above would cost if manufacturing was ramped up to the level of solar cells.

Right now, it is cheaper to get your hydrogen from methane steam reformation which takes natural gas and water to produce hydrogen and carbon dioxide. So unless you heavily subsidize "green" hydrogen you'd just get your hydrogen from hydrocarbon sources. However, if its energy you are after, you're likely already using natural gas directly to generate your power.

The hydrogen economy can't really take off until the carbon economy starts to die off. But it would be a really good thing to have most of the technology worked out so its ready and waiting for the day it becomes economically viable. The research group I'm doing my PhD with is working on water oxidation catalysts using the more abundant elements such as iron, nickel and the like. It's pretty slow going as it can take months of work just to see a small % increase and that's if you're lucky.

[madd0ct0r]

hydrogen is a bugger to store since the atoms are so small it leaks through most things, and has other nasty effects like embrittlement of stainless steel. One trick I've been investigating is the use of hydrogen as a feedstock for making methane, which is easier to store.

Although then you have to burn the methane and you have a carbon dioxide problem?

Or am I confused?

Presumably you'd be fixing atmospheric carbon to make the methane in the first place.

Bingo - the specific context was improving the 40% vol of biogas that's carbon dioxide, but inorganic process could probably cope with any co2 feedstock. It's less carbon capture and more carbon recycling. Similar experiments were using the flue gas off a power station to feed algae that could then be burnt in the power station

[Nova Andromeda]

If we sum the costs of this and the necessary mirrors (maybe a parabolic dish like they use with solar thermal?) would the total costs be comparable? We'd only need a solar cell covering a ~0.07m2 area if we use the 35m2 total sun light area figure from above.

Last time I checked the cost of conventional solar cells was about $2 per watt, at a typical 15% or so efficiency that's about $300/m2 so call it $10,000 in round numbers. The high end multi-junction cells are $100,000/m2 or thereabouts, no, that's not a typo, it really does cost that much. Considering that this is a lab stage proof of concept it likely costs a hell of a lot more, but let's use $100k for now. That brings the panel cost to $7000, but then we need to add in the cost of the reflector plus the cost of the cooling system* for the solar cells, neither of which is going to be cheap.

*35kW on a 1' diameter circle is serious business. About half it gets turned into electricity so let's say 18kW of waste heat, the large heating element on my electric range is about the same size but it's only 2600W. Keeping those cells from burning up takes serious cooling.

Yeah, I completely forgot about where the other 18kW of energy was going to go . I wonder what the operating temperature of these cells are. If someone came up with a practical idea to solve this problem would it be worth anything? What if they worked out this whole system (solar cell, cooling, reflector, electrolysis, pressurizer, H2 storage, and H2 electric generator)? It would depend on price correct? What is the price to beat - a quick Google suggest $4/watt installed * 35kW = $140k?

I also noticed that energy transfer to lithium batteries is highly efficient (86%). It seems like a lithium battery buffer is the way to go for now, but for long term bulk energy storage and transfer I still like hydrogen.

[Nova Andromeda]

hydrogen is a bugger to store since the atoms are so small it leaks through most things, and has other nasty effects like embrittlement of stainless steel. One trick I've been investigating is the use of hydrogen as a feedstock for making methane, which is easier to store.

Although then you have to burn the methane and you have a carbon dioxide problem?

Or am I confused?

Presumably you'd be fixing atmospheric carbon to make the methane in the first place.

Bingo - the specific context was improving the 40% vol of biogas that's carbon dioxide, but inorganic process could probably cope with any co2 feedstock. It's less carbon capture and more carbon recycling. Similar experiments were using the flue gas off a power station to feed algae that could then be burnt in the power station

Is there any chance that this process can be done at a residential scale that is cost effective? Creating hydrogen is 'easy', but I have no idea how much trouble it is to create methane at home from say my leaves and wood chips or a bag of coal.

[LaCroix]

In theory, you could use the heat by using the cooling cycle to drive some kind of steam turbine. That would increase energy output by a few percent.

[Nova Andromeda]

In theory, you could use the heat by using the cooling cycle to drive some kind of steam turbine. That would increase energy output by a few percent.

Actually, I have a much better idea along the same lines and you'd improve the total efficiency by quite lot in addition. One might even be able to combine this with methane generation (maybe madd0ct0r can comment on whether high temps could be used with such a process).

[Borgholio]

In theory, you could use the heat by using the cooling cycle to drive some kind of steam turbine. That would increase energy output by a few percent.

How about a Sterling-type engine?

[madd0ct0r]

In theory, you could use the heat by using the cooling cycle to drive some kind of steam turbine. That would increase energy output by a few percent.

Actually, I have a much better idea along the same lines and you'd improve the total efficiency by quite lot in addition. One might even be able to combine this with methane generation (maybe madd0ct0r can comment on whether high temps could be used with such a process).

Methane is slightly less energetic then the equivalent amount of hydrogen and carbon dioxide, so mix the two of them up and have enough energy to get over the initial 'hump' of the reaction you'd get methane. This is getting slightly beyond what I know.

The specific method I was looking at was using the same bacteria culture that turn biomass into sludge, co2 and methane. If you feed small amounts of hydrogen into the chamber, they can also 'eat' the reaction of that with c02 as it turns into more methane. it's a net energy loss, but one that works well when you are already producing methane as a product. The organic catalysts wouldn't like high temperatures.

There's processes such as wet hydolosis (which is basically heating lots of wet organic shit up, and leaving it to cook for a while) - that'll give you some solids and more gas and liquid then a traditional dry hydrolysis method (like making charcoal). People were looking at microwaves as a hyper effeient way to heat the water. Again, there is a net energy loss.

[Irbis]

We are going to be told that this is useless junk because lol no storage.

No, because lol laws of physics that still won't let even 99.9% efficient solar cells walk around the pesky problems of nights, clouds, or even air at high latitudes stopping most of useful energy before you can capture it. Solar energy might be good close to equator, but most of highly developed, populated and industrialized regions that do need electricity are (surprise!) far away from it. Let me quote:

“CPV is the most efficient solar technology today and suitable for all countries with high direct normal irradiance.”

I also don't see costs in that snippet, breaking records is all well and good but it's easy to do on unlimited budget. I also don't see anything about resources used, new renewable energy technologies unfortunately require big amount of rare earths much better spent elsewhere, like making electric devices more efficient for tenth of the cost.

Can someone explain to me why cracking water into hydrogen and oxygen isn't a perfectly good way to store the energy? Is 18% efficiency not worth it? Are the costs too high (even assuming bulk manufacture?)?

In short - yes.

Frankly, at such pathetic efficiencies we're talking about with "green" energy I'd honestly expect biobutane synthesising bacteria (and rest of biomass exploitation) research effort to end up much less expensive, and, since it can be used with current infrastructure, for it to sink solar/wind and lead to humanity to happily continue burning conventional fuels. Making problem worse, not better.

And to think same research effort put into nuclear power plants would give us - today - much cleaner and smaller footprint energy than any future "renewables" even dare to promise us. Sadly, irrational and idiotic scare of Baltic tsunamis and Alpine earthquakes killed it, no matter how many times you beat Greenpeace imbeciles over the head with the fact that Euratom countries had ZERO fucking serious accidents over last 70 years, with technology far more primitive than that we have today

[Enigma]

How much better is this than using tech derived from DVD\Blu-Ray discs to increase solar cell efficiency? I'm talking about that article Mr Friendly Guy posted a while ago that noone replied to. In the article it was mentioned that they used a Blu-Ray disc to increase solar cell efficiency by 22% to an efficiency level of 44%.

[Nova Andromeda]

How much better is this than using tech derived from DVD\Blu-Ray discs to increase solar cell efficiency? I'm talking about that article Mr Friendly Guy posted a while ago that noone replied to. In the article it was mentioned that they used a Blu-Ray disc to increase solar cell efficiency by 22% to an efficiency level of 44%.

A quick Goolge only showed an increase of 20% efficiency. Are you sure that is additive (not 22% efficiency * 1.22 = ~27%)? If so do you have a reference I can read?

[LaCroix]

How much better is this than using tech derived from DVD\Blu-Ray discs to increase solar cell efficiency? I'm talking about that article Mr Friendly Guy posted a while ago that noone replied to. In the article it was mentioned that they used a Blu-Ray disc to increase solar cell efficiency by 22% to an efficiency level of 44%.

A quick Goolge only showed an increase of 20% efficiency. Are you sure that is additive (not 22% efficiency * 1.22 = ~27%)? If so do you have a reference I can read?

Still, a lot of 'ass-cheap to make' 27% cells that only need direct sunlight do trump a stupendously expensive 46% effective cell that needs (sun-tracking) mirrors amplifying the sunlight by a factor of 500...

[Enigma]

How much better is this than using tech derived from DVD\Blu-Ray discs to increase solar cell efficiency? I'm talking about that article Mr Friendly Guy posted a while ago that noone replied to. In the article it was mentioned that they used a Blu-Ray disc to increase solar cell efficiency by 22% to an efficiency level of 44%.

A quick Goolge only showed an increase of 20% efficiency. Are you sure that is additive (not 22% efficiency * 1.22 = ~27%)? If so do you have a reference I can read?

Just go down about halfway on this page and you'll see Mr. Friendly Guy's post on this.

[Nephtys]

We are going to be told that this is useless junk because lol no storage.

Are you an Engineer? No? Shut up.

Efficiency is only one of many, many, many different attributes that goes into making an effective overall system. We've had III-V PVs that have had efficiencies in the 40s for years. Doesn't mean they're remotely viable for say... USE ON EARTH.

[Sky Captain]

Storage is main problem. Solar cells could be free and even then solar power station that includes storage so it can last through night and few cloudy days are going to be very expensive. IIRC even now solar cells make only about 30 - 40 % of the total cost of solar panel farm and that is not inlcuding storage which currently is handled by fossil fueled stations adjusting theit output. Installation, maintanance, connections to grid make up majority of the costs. If we also included strorage requirements for every new solar power station the solar cells would be tiny fraction of overall cost.