June 16, 2007
Metamorphic Materials Increase Solar Photovoltaic Efficiency

MIT's Technology Review reports on progress in raising solar photovoltaic cell efficiency.

The cell, which employs new "metamorphic" materials, is designed for photovoltaic systems that use lenses and mirrors to concentrate the sun's rays onto small, high-efficiency solar cells, thereby requiring far less semiconductor material than conventional solar panels. Last month Spectrolab published in the journal Applied Physics Letters the first details on its record-setting cell, initially disclosed in December, which converts 40.7 percent of incoming light into electricity at 240-fold solar concentration--a healthy 1.4 percent increase over the company's previous world-record cell. Other groups are developing promising cells based on the new type of materials, including researchers at the Department of Energy's National Renewable Energy Laboratory (NREL), in Golden, CO. The NREL researchers will soon publish results in the same journal showing that their NREL's designs are tracking Spectrolab's, improving from 37.9 percent efficiency in early 2005 to 38.9 percent efficiency today.

The Boeing subsidiary Spectrolab is keeping ahead of NREL. Are private sector workers more motivated or better funded? How about a hefty X Prize for the first group to exceed 50%, 51%, and so on?

Combine higher efficiency with solar concentrators for lower cost. Cheap solar power anyone?

Such high output may be just the beginning. Raed Sherif, director of concentrator products at Spectrolab, says there is every reason to believe that these metamorphic solar cells will top 45 percent and perhaps even 50 percent efficiency. Sherif says those efficiencies, combined with the vast reduction in materials made possible by 1,000-fold concentrators, could rapidly reduce the cost of producing solar power. "Concentrated photovoltaics are a relatively late entry in the field, but it will catch up very quickly in terms of cost," he predicts. (See "Solar Power at Half the Cost.")

Cheap solar power would favor a shift to electric cars. Combine cheap solar power with cheap, high capacity, and safe car batteries and market forces alone would make our environment much cleaner. We'd get cleaner air. But we'd also get cleaner water as the use of oil (leaks of which run off in rain) gradually dropped.

Cheap solar power would also lower costs and therefore accelerate economic growth and raise living standards. Cheap solar power with high conversion efficiency would use probably two orders of magnitude less land to produce the same amount of energy as biomass produces. So cheap solar power would reduce the demand for land for biomass energy. That would reduce habitat loss, especially in tropical lands such as Brazil, Malaysia, and Indonesia.

The rising price of oil and the expectation that oil production won't rise as fast as demand has powerfully concentrated many minds to look for alternative cheaper sources of energy. Cheap solar power is an achievable goal and I expect we will see it achieved by the 2020s at the latest.

Share |      Randall Parker, 2007 June 16 08:18 AM  Energy Solar

Archives said at June 16, 2007 9:24 AM:

The problem's not really about efficiency or cost of solar cells. The problem's lack of storage, and the sun only being available an average of 6 hours a day.

If redox flow cells ever shape up, solar will be great for fixed location installations.

Bigelow said at June 16, 2007 9:35 AM:

Re: the question “Are private sector workers more motivated or better funded?”

“America’s largest exporter, Boeing is also the Pentagon’s second largest contractor, eclipsed only by Lockheed Martin.” http://www.corpwatch.org/article.php?list=type&type=10

Most people call it some variation of Government-Military-Industrial-Complex, not private enterprise.

K said at June 16, 2007 12:05 PM:

There is a lot of fantasy about defense and commercial work. Defense contracts do not make you a commercial winner. Most other big US defense contractors have very little commercial product. Many have failed miserably, more than once, in commerce.

Boeing commercial competes against an entire world of companies who receive government support and are just as involved in defense sales as Boeing.

Commercial and defense work are well separated in law. There is very little financial interaction. Auditing and oversight is strict. The real advantage is not in financial flim-flam. It lies elsewhere and can be substantial.

With defense and commercial work you have a bigger pool of talent and upper management has more information. The cost of training charged to the defense side benefits the commercial side when employees transfer. Resources are more easily shifted to handle difficulties. Expiring defense contracts may leave behind excess facilities which are nearly useless to outsiders and can be acquired cheaply. Defense contracts also bring political support.

Nick said at June 16, 2007 3:55 PM:

Not to mention that NREL has been under attack almost it's entire life by entrenched interests.

This was the area that was defunded so badly by the Bush administration just a year or two ago, that when Bush planned a PR event there, they had to backtrack and restore funding to reinstate employees, lest it be a PR disaster.

Randall, I know you were being conservative when you predicted price parity by the 20's but I think you can also comfortably add that there is a real possibility within the next 5 years. PV costs have continued to fall lately, though prices have stayed high because of the excess of demand over supply. Silicon supply is likely to catch up in the next year or two, and prices will fall sharply. More importantly, CIGS thinfilm is accelerating, and thinfilm companies predict parity by 2009 or 10. That may be a bit ambitious, but 5 years seems to me doable.

Randall Parker said at June 16, 2007 3:56 PM:


If solar panels become dirt cheap (e.g. they can get molded to serve as roof tiles and cost no more than existing roof tiles) then they'll certainly get used to supply daytime power. That'll cut use of natural gas for peak electric and also cut use of coal during the day.

Add in cheap batteries for cars and photovoltaics will get used to charge up cars. The fact that the sun doesn't shine at night is not a problem for cars if the cars can get fully charged during the day.

Make a power source cheap some of the time and more expensive other times and people will find ways to shift some of their demand to when the power is cheapest. A large variety of methods enable that shift in demand. For example:

1) Make freezers colder during the day to take them thru the night.

2) Use liquids or solids in tanks that get heated in the winter or cooled in the summer during the day to use to heat buildings during winter nights or cool them during summer nights.

3) Run energy intensive metal processing processes only during the day (e.g. aluminum smelters) and much more during the summers.

4) Move water in the California Aqueduct and similar aqueducts only during the day. Just move more water when moving it during the day so that it doesn't need to get moved during the night.

5) Program clothes washers and dish washers to come on around noon time.

6) Program automated vacuums to come on and clean a house during the mid day.

Nick said at June 16, 2007 4:01 PM:

Archives, the storage problem is greatly exaggerated. Solar insolation matches demand pretty closely. By the time solar gets to, say, 20% market share, and needs storage, PHEV's will be widespread and managed charging will be able to effectively provide all the storage needed.

Not to mention that PV is a consumer controlled item, largely beyond the control of utilities. If solar makes demand more variable, they'll just have to learn to cope.

Nick said at June 16, 2007 4:05 PM:

Randall, you took the time to detail some more load management options - a good list. You can add long-distance transmission, which would join the time-zones in the US and smooth out solar generation; pumped storage like Ludington, MI; gasified biomass in peaker generators...there are many options.

Randall Parker said at June 16, 2007 4:07 PM:


Yes, the demand run-up in recent years combined with the increased demand for silicon crystals for semiconductors has hidden the real production cost drops for making PV devices. Plus, yes, new processes that do not use expensive silicon crystals will cause much bigger price drops.

Yes, I'm being conservative by predicting lost cost PV by the 2020s. I'm expecting low cost PV to come much sooner than that.

I'm expecting batteries good enough for PHEV within 5 years tops and good enough for long range electric cars some time in the 2010s. I'm also expecting cheap PV in the 2010s.

Randall Parker said at June 16, 2007 4:25 PM:


Here's a thought that hadn't occurred to me before: Long distance transmission with superconductors would work better in Eurasia than in North America or South America. Why? Eurasia has more time zones in which to place photovoltaic panels. Superconductor cables will lower electric costs more in Eurasia than in the Americas?

You might argue that transmission of electricity between north and south would work due to different seasonal exposures. But South America is too narrow.

Given cheap long distance superconductors the US and Canada will do very well with solar during the summer when the days are very long in the far north. But that's only true if it is worth putting solar panels up there and if the panels won't get damaged during winter.

Of course, cheap storage tech (including tech to convert electricity into liquid hydrocarbons) will change the economics of all this considerably.

Nick said at June 16, 2007 5:24 PM:

I suppose Eurasia will be helped by their size, if they learn to cooperate to that extent across national boundaries. OTOH, the continental US & Canada's 4 time zones would go a long way toward smoothing out solar insolation: that's a 4 hour longer window.

AFAIK wind is a little stronger in winter than in summer, so that's a nice synergy.

Again, solar matches demand awfully nicely. Keep in mind that much night-time, so-called baseload demand is really Industrial daytime demand (like the aluminum smelters that you mentioned) that's been shifted to the night by primitive demand-management (day time peak demand charges, etc).

So, with relatively little demand management (primarily of PHEV's) I don't think much storage will be needed for diurnal cycles, and it would be relatively expensive for seasonal cycles. I suspect biomass that starts out as biomass will make a better source for supplementary generation in winter and the very occasional extended period of low output from all renewables. As transportation electrifies, inefficient liquid hydrocarbons will be less important.

Nick said at June 16, 2007 5:35 PM:

On batteries: GM's engineers are saying that their candidate batteries (A123systems & LG Chem) are good enough right now. What's needed is battery pack integration and power electronics, which is scheduled for completion June 08, only 1 year away.

You never know until it's actually out the door - just look at Toyota's recent panic about cobalt based li-ion safety - but A123systems batteries have a fundamentally much safer chemistry, and have had a lot of tough testing by power-tool users. There's an extremely good chance that PHEV batteries will be ready in just a year!

I suspect/hope that GM is hoping to get the Volt out in just 24 months. They were saying 2010, and lately they're saying that there is a specific target date, but that it's a corporate secret. Production planning normally follows the completion of design work, and in this case GM is doing them in parallel, which is a high-risk, high reward strategy which is unusual for GM. With Toyota's li-ion Prius delay, they suddenly have a chance to leapfrog Toyota. GM is acutely aware of it's PR problems, and is hoping to take the high-tech green mantle away from Toyota. I think they've become aware that it may make the difference in their survival...

TTT said at June 16, 2007 7:46 PM:

Tesla Motors will have the $50,000 electric car out by 2010. They will have a $30K one out by 2013.

Even if PV costs drop by 5% a year (which they have since the 1970s), solar power becomes cost-effective by 2013 anyway.

Jim said at June 16, 2007 10:02 PM:

All this is sort of wasteful. All that effort into designing and manufacturing more plastic...

All you need to do is put several hundred high altitude electric generators at 30,000 feet and we could generate most of the power needed to power the entire country cheaply.

Don't think so?
These few paragraphs may help sway your thinking on this:

Oh, then toss in some nice big algae ponds in some god forsaken back country, like the entire useless waste of space state of Wyoming with it's entire population of 500,000 people, and we have all the bio diesel we need to get around (just move 'em all to Idaho and "Algae" the place-- no one would ever notice...).

Just a little quote, "At current levels — algae can produce about 1,200 barrels of oil a day — a company would need 4 million acres of land to replace the 5 billion gallons of jet fuel used annually."

All this is doable at low cost within a decade, without further gazillions spent on "research," or ruining our economy by promoting ethanol alternatives... oh man, another of this decades biggest "joke's on you" scams.

What the heck huh?

OK, solar is cool for it's sheer geekiness, and it seems so green doesn't it. But it's still so incredibly wasteful in terms of production and waste over time... We are just moving one problem around to another (into land fills when all those solar cells are "upgraded," as well as the increased GHG's in order to produce all the extra plastics and exotic materials needed to fabricate those "solar roof tiles").

Can you say: "Outgassing"

See ya,

Randall Parker said at June 16, 2007 10:30 PM:


The United States uses about 145 billion gallons of gasoline per year plus more for airplanes. Well, translate your 4 million acres for 5 billion barrels into enough to run all the cars: 145/5 times 4 miilion equals about 112 million acres. That is more than the 90 million acres of corn planted in 2007.

That much area as algae ponds would be enormously more difficult than as corn fields. The ground would need to be sealed to prevent leakage over an enormous area. During the rainy season the water would overflow the ponds. The pounds would need to be flat on terrain that won't mostly be flat to start with or the ponds will need to have very high walls on lower areas and more water to cover their more elevated areas. Or each pond would need to be much smaller and they'd need to be layed out like terrace steps.

During the dry season the ponds would need water pumped in to replace evaporation or they'd need to be covered over in glass that would let the light through.

Solar photovoltaics are easy to use in comparison.

Phil said at June 17, 2007 4:31 AM:


Aluminium smelting cannot be a stop-start process. It has to be running as near as possible to 24/7:

"If production is interrupted by a power supply failure of more than four hours, the metal in the pots will solidify, often requiring an expensive rebuilding process."

( http://www.world-aluminium.org/production/smelting/index.html )

That's one of the reasons why most smelters are powered by hydro-electricity.


Nick said at June 17, 2007 5:36 PM:

Jim, where did you get the impression that photovoltaics require much plastic? Think silicon (both purified for the PV cells, and in the form of glass substrates and covers), and metals.

Phil, I believe that article is talking about catastrophic complete loss of power - the power required to maintain temperature and prevent solidification is a small fraction of normal operational power requirements to actually separate the pure aluminum. Aluminum smelting is extremely sensitive to the price of electricity, and I'm sure they operate preferentially at night wherever daytime peak power is at a cost premium.

Innovation Catalyst said at June 19, 2007 7:33 AM:

The problem with use of concentrating solar panels at the home is that they have to be pointed at the sun at all times. That means a motorized tracking system. Unless your home faces north you won't be able to orient the panels to allow east-west panel element rotation. I have seen one solar panel application that uses waveguides so the panels don't have to be pointed directly at the sun - this is the best solution I've seen yet.

Nick said at June 19, 2007 10:10 AM:

I suspect that concentrating solar is intended more for utility-scale generation.

Industrial/Commercial flat rooftops have some potential - for example, Sunflower.com is intended for I/C roofs.

Kralizec said at June 19, 2007 9:27 PM:

Large cash prizes for advances in the conversion of solar energy would be worthy contributions to American national security. That thought leads me to realize that, for a couple of reasons, it seems the Americans should beware of the possible deployment of cash and other sorts of influence on the other side, as well. Anything that contributes to a large relative reduction in the Americans' consumption of oil will be in opposition to the financial and theologico-political interests of a family of foreign oil producers. It'll be interesting to see how the family responds, if they do eventually come to see themselves as threatened by emerging alternatives to the oil over which they've been unwisely accorded control.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Go Read More Posts On FuturePundit
Site Traffic Info
The contents of this site are copyright ©