Thin-Film solar Market Soaring

According to a new report from NanoMarkets LC the global market for thin-film photovoltaics (TFPV) is forecast to reach $7.2 billion by 2015, compared to just over $1.0 billion today.

The report states that the market is being driven by the inherent advantages of TFPV. . . . most manufacturers are ramping up production capacity and several -- including First Solar, Fuji Electric, Nanosolar, Sanyo, Uni-Solar and G24i -- are building plants with more than 100 MW in capacity.

Some of the findings of the report include:

• PV could eventually account for as much as 20 percent of the U.S. market's energy needs.
• TFPV was only five percent of the entire PV market recently , but it is expected to account for 35 percent of the photovoltaics market by 2015.
• TFPV can be manufactured using simple printing or other R2R machines.  Printing TFPV has the potential for lowering capital costs by as much as 75 percent, reducing waste and increasing throughput.

Comments

[Luis]

Great post!

If the economics don't work, recycling efforts won't either.
As our little contribution to make this economics of recycling more appealing, http://LivePaths.com blogs about people and companies that make money selling recycled or reused items, provide green services or help us reduce our dependency on non renewable resources.

[Don Scherer]

Eight year forecasts are rarely worth the ink (or pixels) required to print them. The technologies that will exist in 2015 are unknown, making "forecasts" such as this little more than WAGS. And if TFPV can truly "save 75% of the capital costs", why would it only grab 35% of the market? In fact, with such a price advantage, it would elimate all other competitors.

[Ewout]

Don: sometimes other charactaristics are more limiting than just price, surface area is one of them - not everyone has 60m2 of roof to spare, and when non-TFPV uses 3x less surface area, that could very well be a good argument. More importantly though, "75% of what?"... If that's compared to the current non-TFPV prices, how much will they save?

[Marco de Salvo]

Finally a specialized blog on energy!
I'd like to signal my web www.marcodesalvo.com and my blog http://templeoftechnology.wordpress.com/ with hi-tech products reviews.
I'm interested also in light ducted waveguides and solar energy, so I'd like to cooperate in this field of interest.
Best Regards
Marco

[Freddie Sirmans]

Just browsing the internet, your blog is very, very interesting.

[Calamity]

Finally a specialized blog on energy!

Uhh actually there have been quite a few around for quite some time. Where have you been lurking lad? :)

Up to 20% market share seems plausible. With intermittency mitigation methods ( geographical distribution, TOU, V2G, DSM and various other nifty acronyms) this can be pushed a bit further.

However at some point storage will be needed. Solar thermal excells in this department, with simple inexpensive and durable storage, easy to scale up. That, combined with the simple fact that a solar thermal field is far less costly than a TFPV solar field, gives solar thermal a higher potential than TFPV in terms of potential market share.

If no inexpensive storage for TFPV is developed quickly enough, it will be a serous bottleneck for TFPV's market share. Then solar thermal may be more significant.

[mds]

Don Sherer,
"rarely worth the ink (or pixels) required to print them"
Yes, it's hard to predict out that far with any accuracy. I do think the study provides a good picture of the most likely future trend in PV. If this study is even roughly correct, then TFPV will continue to capture more market share after 2015, ...more than 35%.

Ewout,
I agree "75% of what?" It's not clear. I assume they mean 75% of current prices.
www.solarbuzz.com puts average current price of PV modules 125 W or larger at $4.85 per Peak Watt (Wp). Low end of this average price is $4.30/Wp, $3.96/Wp, and $3.00/Wp for mono-crystalline Si, multi-crystalline Si, and ThinFilm respectively.
A number of TFPV companies claim they will be profitable at prices below $1.00/Wp. The same claim is being made for large scale Si PV production and for ThinFilm CdTe PV production. It's going to be a very interesting race. 35% of market share by 2015 for TFPV has to be a very rough prediction. Radically reduced prices for PV in general is an easy prediction.

These prices do not include installation costs. Installed PV cost from www.solarbuzz.com:
"In order for the solar industry to make a systematic penetration in to the electricity segment, installed solar system costs will need to drop from around $8-10/Wp to $3/Wp. This would continue the trend shown above of falling solar electricity costs over the last twenty-five years. A push to $3/Wp would bring solar energy costs from the present 30 cents per kilowatt-hour to around 10 cents per kilowatt-hour, which would allow it to compete more strongly with other renewables and capture a significant share of the electricity market."

So if current prices for PV panels are about $5/Wp ($4.85) then installation costs must be $8 to $10 minus $5 or $3 to $5 for installation. Most TFPV companies are going after Building Integrated PV (BIPV) in addition to regular flat panel production. This will radically reduce installation costs. The light weight and flexibility of TFPV makes BIPV easier to accomplish. A 75% reduction from $8 to $10 installed would bring the cost down to $2.00 to $2.50 per Peak Watt (Wp). This is below the $3/Wp threshold mentioned. Kaboom, we have market share occuring, ...and the price of TFPV is not done falling at this point. In the sun-belt of the USA you can save anything over average power prices of 10 cents per kilowatt-hour. Peak power prices, primarily used for AC during the summer can be twice this high.

"sometimes other charactaristics are more limiting than just price, surface area is one of them - not everyone has 60m2 of roof to spare"
You have a point here. This is why I think companies with concentrator panels, like Delta Electronics using SpectroLab PV cells, may be able to keep a market niche. In spite of this is seems likely TFPV will dominate. The cost/Wp will be lower. The installation cost when integrated into roof shingles, windows, and wall material will be lower. Market penetration will be greater. I could be wrong. Again, it will be an interesting race.

Calamity,
Why would you use TFPV in a "solar field"? Way more cost effective to put on roof of home or business. Higher price differential at the end user site. No environmental impact mitigation required. No utility scale financing or large site planning.

Much of storage for AC in USA sun-belt can be accomplished at low cost with ice, by cooling rocks down, or by combination of the two.

I understand geographical distribution and V2G storage. (These and ice are enough storage for PV to claim more than 20% of USA energy generation.) What are TOU and DSM?

[Kit P]

TOU – time of use (metering)
DSM – demand side management

[mds]

Kit P
Thank You

[Kit P]

“Up to 20% market share (US) seems plausible.”

Not even credible. It is really hard to get solar believers to understand the limitations. Let say that you have $4000 budgeted for a vacation trip. The plane tickets costs $2000. The hotel is costs $800. Along the way, two quarters are dropped into a parking meeter. The $0.50 is the represent share of solar to the contribution of the US electricity market.

The biggest limitation on solar is environmental regulations. Ironic!!! Every large energy project requires an EIS. A 5 MWe solar project has the same environmental foot print of a 1500 MWe nuke plant and the same number of public hearings.

Making electricity is inherently dangerous. Construction projects are inherently dangerous. Maintenance is inherently dangerous. Power projects are inherently economically risky.

It is called paper work.

[Calamity]

mds said: Why would you use TFPV in a "solar field"? Way more cost effective to put on roof of home or business. Higher price differential at the end user site. No environmental impact mitigation required. No utility scale financing or large site planning.

Yes I did post with that in mind. It's all true. Perhaps I should have said modules instead of solar field for TFPV. My point was that TFPV is limited by not having an inexpensive storage method. Even if all roofs were covered with TFPV, and intermittency mitigation methods are utilised to the maximum, there will still be a point where storage is needed. Otherwise TFPV remains bounded.

For solar thermal this bound is stretched much further; it can mostly deal with electricity demand at night and in the early morning/late evening.

Storage for solar thermal is already under $20/KWh and quite durable, fairly easy to scale up. The solar field for a compact linear fresnel reflector is about 50 cents/Wp for a large plant. This means that adding more solar field plus inexpensive storage increases the economics of the plant; the relatively expensive turbine is utilised more often.

Compare that to TFPV storage. Pumped hydro? Nice but likely to be insufficient for this scale of storage. Flow batteries? Over $150/KWh usually, and uses rather rare materials i.e. difficult to scale up quickly.

What do TFPV modules cost/Wp? It may be quite some time before they reach 50 cents/Wp. (Note: such a low cost/Wp is not that important for roof mounted consumer systems).

Point is not that TFPV is no good; it is that there are bottlenecks to it's growth if there is no good storage available.

I'm optimistic that such storage will get much cheaper in the near term, but it's good that we already have a "solar backup" until they do.

Just for the record, I don't see things like "thermal vs TFPV". That's dogmatic. We'll have both for a long time, at least I hope so.

Frankly I'd be bored if all electricity comes from one source ;)

[mds]

Calamity,
Sounds like we basically agree. You are a little more bullish on solar thermal. Right now both forms of solar (all three if you want to classify as PV, CPV, and CST) are starting to make economic sense in a high demand energy market.
http://www.stirlingenergy.com/default.asp (Stirling Energy Systems – Parabolic dish with helium sterling engine, < 8 cents per kilowatt/hr)
http://thefraserdomain.typepad.com/energy/2007/03/nevada_solar_on.html Solargenix - Built and comes on line next month, 10-12 cents per kilowatt/hr – March 2007news.
http://www.ishitech.co.il/0306ar4.htm Solel - (claims 10 cents / kWh now – predicts 6 or 7 cents / kWh in 5 years)
Think these numbers bear out your $0.50/Wp claim. Note: The Stirling Energy system does not provide thermal storage.

My view is that TFPV will grow at same time that electric transportation is coming on the scene. USA sun-belt will initially use TFPV to reduce peak daytime energy costs. Then ice/rocks (thermal mass) will be used for cheap storage to handle nighttime AC loads as well. Battery storage will also be used some. (Ice and Batteries are already being used for this now.) Finally, PHEV and EV cars will provide V2G energy storage for night-time use. This will play out over a 20 to 30 year time frame, unless we have a national crisis and push it faster. I don't think Solar Thermal can come down in cost as much as TFPV, but you have good point about nighttime power use and I could be wrong. Both will probably play a role for some time into the future, maybe permanently. In 10 years solar will be making a more noticeable contribution. In 20 years it will be a major contributor to our national power. Geometric growth. PV production has been doubling in close to every three years. This is getting ready to accelerate.

Kit P
I disagree. 20% of USA power by 2015 is not credible, but they say "COULD EVENTUALLY account for as much as 20 percent of the U.S. market's energy needs".
Given the cost trends for solar, and the plethora of new approaches to reducing this cost further, and a totally undeclared time frame, I don't see how this prediction can miss. Maybe if there is a huge nuclear fission revival or if we start mining the moon for H3 to use in nuclear fussion. Even then I think solar is here and will be part of the mix.
"The biggest limitation on solar is environmental regulations. Ironic!!!"
This is not true for PV, CPV panels, and Solar Thermal Panels on home and business rooftops. TFPV with BIPV will be even more cost effective with same huge advantage. Don't know why you always fail to see this advantage. It is a considerable one.

If PV production is roughly doubling every three years, then installations are roughly tripling every three years ...and it's going to accelerate. TFPV is one of the big reasons why. Order of magnitude cheaper production facilities, better than order of magnitude faster production rates, and two orders of magnitude less material. Long term result is $5/Wp reduced $0.50/Wp or lower and this is at the end user site. The home computer or cell phone of power sources...and we're scaling production way up right now. Get ready to be shocked at the transformation,...if you're not already.

[Calamity]

mds yes we agree. If I appear a bit leaning towards solar thermal, it really is because of storage. It's a good thing that we already have inexpensive storage for this technology, no need to wait for technological breakthroughs. This does contrasts with TFPV's ability to grow faster than solar thermal.

However, storage doesn't matter much right now, so my argument isn't very relevant for now. If a good inexpensive electricity storage device is developed quickly enough (or V2G maybe, if that is sufficient by itself), then the storage barrier is gone. This is what I hope; yet hope is not all we have because there's solar thermal's storage which is ready right now.

Imagine affordable BI-TFPV with cheap durable storage on respectively in your own home. For people in sunny areas, 100% solar could be done without compromises.

Whether or not good storage will be available for TFPV, solar thermal may in either case be good for morning/evening/night demand and also additional daily peaking when needed.

In any case there's room enough for solar thermal, PV flatties (Sliver Cells?), TFPV and also concentrating PV.

About that 50 cents/Wp claim:

http://www.physics.usyd.edu.au/app/research/solar/clfr.html

Costs are kept low by using water as a heat transfer fluid, a low cost structure with reflectors close to the ground, a low cost receiver which is composed of mild steel pipe, and exceptionally low reflector costs due to advanced laminated construction. The installed array and heat exchanger cost is about $A900 peak electrical kilowatt, about $US500. O&M is low because cleaning can be done manually at ground level. This is less costly than an automatic cleaning system.

The turbine etc. needs to be added to this, I've heard claims about total system costs as low as $ 2/Wp and that's with reasonable capacity factors already in the price.

I've always thought dish/sterling is bit odd. It's solar thermal but it doesn't have the cheap storage option. That means it essentially competes with CPV more than solar thermal. It's modularity is a clear advantage though.

Each of the solar techs seems to have it's own strengths and weaknesses. I'm very curious about which ones will succeed more than others.

[Calamity]

Spelled Stirling wrong. Had to happen sometime. Of course, you could make a Stirling engine out of Sterling silver, but that would be confusing and probably not the least cost option ;)

[Kit P]

MDS, wrote “This is not true for PV, CPV panels, and Solar Thermal Panels on home and business rooftops. TFPV with BIPV will be even more cost effective with same huge advantage. Don't know why you always fail to see this advantage.”

I fail to see the advantage because I am a skeptic. MDS is correct that small projects do not need an EIS. However, small projects lose the economy of scale. I am also skeptical of the claims of Bill Clinton and Arnold the Governator for million roofs goals. Even if you can find that many gullible people willing to part with large sums of cash, gullible people are not very good at making electricity.

[mds]

Calamity,
Very interesting link. Some real design elegance there. Thank you.
You've made me think here. I wonder if solar thermal could have a future in home power generation also, in combination with PV? Thanks again!
Think the Aussies will commercialize this CST design? Sliver solar does not seem to be storming the market. Problem with Australian business development environment or with competitiveness of Sliver technology?

Kit P,
Yes, you're a cynic ;-) You're entitled.
You will have economies of scale in production of PV, more so than for large industrial PV or CST plants. This is why I keep bringing up PCs and cell phones. PV is another consumer product. It will benefit from the same mass production that saw phenominal growth of these consumer products. BIPV will similarly make installation a factional cost increase over regular roof, window, and/or wall installation. PV will have the advantage in economies of scale AND in EIS. TFPV even more so.
I was laid up a few weeks ago and read an interesting article on group behavior. Bees aren't very smart but arrive at good nesting decisions by working together. Think you will be surprised exactly how good these "gullible people" turn out to be at generating electricity.

[Calamity]

Hi mds

Now you know why I think CLFR has so much potential. I posted some more info (although a bit outdated) in another thread.

About those sliver cells, yea they appear a bit sluggish. There is a video from ABC if you haven't seen it already.

Notice how they do have automated production facilities, so this makes me hopeful. This simple idea has vast potential. High efficiency and yet proven reliability. They do mention in the video that it will take a few years for them to penetrate larger markets. We'll find out more about them I'm sure. More players and technologies in the market is a good thing for PV.

The Australian National University is very optimistic about this technology.

http://solar.anu.edu.au/level_1/pubs/papers/Hawaii_2006/Sliver_Cells.pdf

Notice the sub 1$ module manufacturing costs. If they succeed, then flat plates can be pushed a lot farther than most people think.

[Calamity]

Kit P said: [...]gullible people are not very good at making electricity.

Your logic is strange as always. Unless you want to put those people on treadmills, the relevance of your point is approaching zero.

[Calamity]

Perhaps combining high efficiency concentrator cells with CLFR could bring an interesting, low cost option for CPV?

CLFCPV? or CLFRPV?

No, those would be a torture for the dyslectic. Could lead to a lawsuit.

On a more serious note, I don't think CLFR would be suitable for residential microscale generation. Maintenance gives too much overhead and perhaps there will also be steam/fire hazards. Potential eye damage from the reflection of the absorber as well. Maybe for industrial CHP apps. That's what the Lidell coal plant's 40MWp CLFR is being used for; coal saving through steam reheating.

[Calamity]

Industrial companies in sunbelt areas that use large amounts of heat (and maybe also considerable amounts of electricity for CHP - electricity production could be grid tied for net metering) might be an excellent niche for small scale systems complementing electricity and natural gas from the grid.

[mds]

Calamity,
Thanks again.
Very interesting to see that Sliver Si technology is still alive and well. Video was very informative. I had not realized Sliver cells had the light weight and flexibility advantages of TFPV. Lacks R2R without (vacuum) vapor deposition that some TFPV has. Still pricing target is impressive and efficiency could be an advantage. Will be interesting to see if this, or one of other reduced cost Si approaches (string ribbon, ultra-thin wafer, aSi, other?), takes a significant portion of the market. Two things are for sure:
(1) Traditional monocrystalline and polycrystalline Si PV cannot remain dominant for long.
(2) Cost of PV will go down dramatically when supply finally catches up with demand.
...and if CPV and CST are still strong runners in the race, WOW!

Travis Bradford’s book “The Solar Revolution” was excellent contribution. It uses a very conservative prediction, based on economies of scale from growth of traditional Si PV, to show PV will soon reach grid competitive prices. Clearly it is going to happen much faster than his conservative prediction. Way cool!

[Calamity]

mds,

No problem. Time wasted on talking about solar is time wasted very well ;)

(1) Traditional monocrystalline and polycrystalline Si PV cannot remain dominant for long.
(2) Cost of PV will go down dramatically when supply finally catches up with demand.

With new dedicated PV grade silicon production coming on line over the next few years, there may be a PV grade silicon price drop. Traditional flat plates actually benefit more from this than other silicon PV technologies. That's because they use more silicon so a silicon price drop has a relatively large impact on prices. Things like Sliver Cells will benefit less from this price drop, obviously. This fact may very well help lengthen the dominance of traditional PV, along with large capital inertia and other factors.

Or maybe not. If silicon prices remain high then traditional flat plates may be much less competitive.

Turning this around, Sliver Cells can be fabricated from the most expensive (purest) high efficiency silicon while adding relatively little to the sticker price.

[mds]

Calamity,
Silicon is a common material. The Si shortage is a shortage of purification/crystalization equipment. Billions are being invested in this. There's no question that PV grade prices will fall when production catches up with demand.

http://thefraserdomain.typepad.com/energy/2006/12/how_long_will_t.html (sort reference to Si shortage – Dec 2006)
http://www.redherring.com/Article.aspx?a=20157
“the average of global prices for solar-grade silicon has grown from $35 per kilogram in 2004 to $74 per kilogram this year, and is expected to reach $95 per kilogram next year”
“Polysilicon only costs $25 a kilo to make” “all you need to do is put up more polysilicon plants”
“next-generation polysilicon technology that promises to make polysilicon for $15 to $20 a kilo”

http://thefraserdomain.typepad.com/energy/2007/04/lower_priced_si.html (SRI International – new Si – April 2007)
“SRI International has licensed technology to produce lower cost solar-grade silicon to three Asian companies and that pilot plants could be up and running in 18 months.”
“promises to make solar-grade silicon for $14 per kilogram”
“contract prices rising up to $85 per kilogram and non-contract prices skyrocketing to $200 per kilogram”

Many Si PV companies have invested in:
(a) technologies like Sliver that reduce the required silicon.
(b) large scale automated production to reduce Si PV manufacturing costs.

My point was companies that do not pursue both of these improvement strategies(i.e. "Traditional Si PV") will be at a considerable economic disadvantage, even when the price of bulk Si drops. It is likely to drop below 2004 price of $35 per kilo-gram because of new, more efficient, manufacturing processes for bulk Si. It will still not be an insignificant cost and will still cut into margin for companies using "traditional Si PV" processes.
When supply of PV catches up with demand we'll have a dramatic drop in PV prices and a restructuring of the industry players. Result will be more efficient PV manufacturing overall and continued growth. Classic disruptive new technology growth pattern, right?

[Calamity]

Let's hope so! And there's more advancements being made now that could lead to the next generation(s) PV:

Silicon nanocrystals for high efficiency with minimal materials used. Or plastic solar cells for inexpensive flexible cells. Or maybe photonic crystals. Concentrator cells are getting more efficient.

There's also a special on solar power in Spain.

You're correct, silicon will probably be the material of choice in the future. CIGS is a bit on the rare side, makes more sense to solve some of the silicon manufacturing problems (both quality and quantity). That does indeed seem to be advancing quite well. That, combined with new minimised material techs not only allows more Wp per unit of silicon, but also (in most cases at least) drives down total energy requirements of manufacturing, improving EROI.

[Kit P]

mds, there is a difference between being a skeptic and a cynic. The reason you keep bringing up 'PCs and cell phones' is because you do not understand physics and thermodynamics. I did not buy either a cell phone or a PCs until they were cheap and practical.

Springerville Generating Station: http://www.greenwatts.com/pages/SolarOutput.asp