Print 73 comment(s) - last by gfredsen.. on Jul 13 at 9:42 AM

The new solar cells from the University of California, Berkley use nanopillars to create cheap and efficient cells. An optimized cost could cut solar power costs to a third of current levels.  (Source: Ali Javey, UC Berkeley)
New technology may unlock some massive savings

One of the biggest factors in solar power remaining an expensive power source, despite constantly improving efficiencies, is the inherent cost of materials and processing for solar cells made of polysilicon.  Cutting these costs could make the solar power the preferred energy source for mankind, but thus-far there has been little high-performance designs made with cheaper processes or materials.

Now University of California, Berkeley researchers have created a new type of solar cells that may offer exactly that -- lots of solar energy with low processing and materials costs.  The new type of solar cells are composed of tiny nanopillars in a thin film layer atop aluminum foil.  The foil is enclosed in a protective layer of transparent, rubbery polymer.

The total materials costs are quite low, and the production costs, while not fully determined also look promising.  Ali Jarvey, an electrical-engineering and computer-sciences professor who led the work, cheers, "You won't know the cost until you do this using a roll-to-roll process, but if you can do it, the cost could be 10 times less than what's used to make [crystalline] silicon panels."

The cells use a nanofilm of cadmium telluride with uniform 500-nanometer-high pillars of cadmium sulfide laid on top of it.  Other thin-film solar cells with pillars have been made before, says Professor Jarvey, but they have relied on more expensive deposition techniques.  Further, the new cells have an efficiency of 6 percent in transforming sunlight into electricity, where past designs had efficiencies of less than 2 percent.

Silicon-based photovoltaics still have the cell beat in efficiency with 20 percent or more in commercially available designs; however, they are extremely pure, expensive crystalline silicon.  Impurities can cause electrons to get trapped in the semiconductor, so the expensive process of making this high quality crystal material is unavoidable for that design.  Purity is much less of a cost concern in the new design.

Creating an equivalent amount of power would require three times the area (panels) of photovoltaic cells, given their respective efficiencies.  This means that given the cost estimates, solar power costs could be cut to a third of the current levels.

Another key advantage of the new design over traditional photovoltaic panels is flexibility.  Traditional crystalline panels would break if flexed.  The thin film nanopillar cells, though, can be rolled and unrolled with ease.

The new design essentially divides silicon's responsibilities.  The thin film material absorbs light and generates electrons, while the pillars conduct the electrons to the circuit and help to trap light.  As electrons have a shorter distance to travel to reach the pillars they're less likely to get trapped by defects, and thus crystal quality is less of a concern.

Currently the cells are produced using a relatively cost-effective anodizing design to grow the pillars on a thin aluminum foil film, the bottom electrode.  The thin semiconductor film is then layered over the pillars and a top electrode of copper and gold is layered thinly to complete the circuit. 

Two key areas of improvement are the top film and the production process.  Adopting a roll-to-roll production system could speed up the assembly and make it cheaper.  Also, currently the gold only allows half the sunlight to enter the cell as its semi-opaque.  Replacing the gold with a transparent material like indium oxide could double the efficiency to 12 percent or more, while not significantly impacting the cost of materials.

States Professor Yang, "(The) architecture is most important--materials we can continue working on. The beauty of this paper is the demonstration of how well the architecture works."

The research appears in this month's edition of the journal Nature Materials.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

RE: well good
By mdogs444 on 7/7/2009 9:03:54 AM , Rating: 1
Seeing that these things are still less than 20% efficient, who cares. I don't want forced into solar and wind that is going to grab more of out my pocket book for someones dream of social engineering.

RE: well good
By glenn8 on 7/7/2009 9:47:27 AM , Rating: 3
So what's the average efficiency of the internal combustion engine? Especially at the beginnings of the tech?

RE: well good
By randomposter on 7/7/2009 9:53:47 AM , Rating: 5
And don't forget to add in the (in)efficiencies associated with drilling for crude, transporting and refining crude, securing (often militarily) supplies to crude, etc.

RE: well good
By Sdaas on 7/7/2009 10:14:55 AM , Rating: 2
Well if you consider all of that then petrol would have been more efficient in its earlier life then today. However, it produces 44 MJ per kilogram. Even with all of the cost and resources required to acquire the fuel for a combustion engine it is still incredibly more efficient then anything we can think of other then maybe nuclear.

However, Lower cost > then efficiency. There is no such thing as 100% efficiency solar energy, and especially if its on the surface of the earth. What needs to be considered is most energy per <currency here>. Not how much energy it produces. Places like California, Arizona, Nevada, and any other desert location would have quite a load taken off of their grids if even 1/5 of the houses and/or businesses produce 1/10th of their power.

RE: well good
By hduser on 7/7/2009 2:31:22 PM , Rating: 3
It'd be hard to compare efficiency of engines, because the quality of the fuel back in the 50-80 years ago doesn't compare to the quality of fuel now. I think the octane rating was quite low then.

RE: well good
By Type2 on 7/7/2009 10:27:23 AM , Rating: 2
Agreed, although 20% is still too inefficient for me. It'll have to be at least twice as efficient as the most efficient power source currently used, one tenth the cost to design, manufacture, transport and integrate into our delicate load balanced power grid vs the current market leader, a one year ROI, the power companies will have to buy the excess power generated at competitive market rates, it must create a substantial net amount of jobs with zero government intervention AND I better not get a whiff of patchouil (aka social engineering) or my highly-conditioned, moving-target support will be even more evasive, untenable and mind-blowingly impossibly unrealistic than ever before. grumble grumble...

RE: well good
By mmcdonalataocdotgov on 7/7/2009 11:10:19 AM , Rating: 3
Where do you get 20%?

the new cells have an efficiency of 6 percent in transforming sunlight into electricity

The arithmetic is: Jason promised 1/3 the cost based 6% efficiency as opposed to 20% for other methods. That is based on 10X less cost but 3X greater area required to match the current efficiency.

So this is MUCH less efficient, but costs MUCH less to make, but requires MUCH more real estate to deploy them on.

Jason forgot to figure in the real estate costs.

If you had to have 3 acres of land at 6% efficiency to deploy these on as opposed to 1 acre of land at 20% efficiency, um lessee, higher land costs, less efficient production rate per acre... I would say this article is crap.

RE: well good
By mmcdonalataocdotgov on 7/7/2009 11:10:57 AM , Rating: 2
Oh, but they're nano, I forgot that part. Nano = better.

"We basically took a look at this situation and said, this is bullshit." -- Newegg Chief Legal Officer Lee Cheng's take on patent troll Soverain

Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki