SouthFacing - The problem here is that different manufactures have different definitions for cell and module. Ascent’s efficiency value according to press releases is at the module level.
James - You are correct about Ascent and their 9.64% NREL approved efficiency value - I’ve adjusted the chart. Thanks for pointing that out.
Richard - Yes FSLR at 10.x% compares with Global Solar, et al.  Though Global Solar expects to be at 11% this year and 12% next year.  Is FSLR’s efficiency trajectory as steep as that?

Eric,

You say that if CIGS firms manufacture with efficiencies of 10-12 percent, they’ll be doing better than CdTe. Are you sure that you are correct on this point? I recollect that in recent quarterly conference calls, First Solar is manufacturing modules with 10.6 percent efficiency.

Adam Starr,
Absolutely, the challenge is scaling and doing it in production.
It’s not just meeting FSLR’s efficiency numbers that these CIGS firms have to do but doing it at FSLR costs which are forecasted to be at $0.75/W by 2011 or 2012.
I understand that lab records are just that, lab records.
My point was that some of the investment frenzy was due to those optimistic numbers.
Regards, Eric

Ascent Solar’s 9.6% efficiency was validated by NREL in December 08.

Please note that evry company has a different deposition method.  The best seems to be the 3 stage method developed at NREL.  Some companys such as Miasole have tried sputtered which NREL investigated but did little research due to limited results(Beck).  Other companies have tried a hybred-sputter/ evaporation with relative good results(~14.0% EPV solar) Other companies have tried an spray ink process with limited results.  Some have invented a new process(fasst) and had limited success.  Nrel’s researchers were never searching for a manufacturable process only high efficiency. So there are two problemms that face manufacturing CIGS.  One is developing a manufacturable process and the other is scaling the process up for large area.

Depending on the process I would agree that a determining factor of efficiency will be the level of vacuum.( I stand corrected.)

Please credit the first photo. This is a picture of Global Solar Energy, Inc.‘s flexible thin film solar cells.

Eric you mention the current efficiencies for the companies are your using cell or module numbers or both not sure?

Looks like Pluvia and his smart money wasn’t very smart after all. As long as you have billions and billions in subsidies being poured into the market, I would say it shows bad risk/reward-judgement from the VC’s to put all this money into an unproven technology.

A technical quibble maybe, but when you’re writing a whole article about CIGS I think you ought to get this right:
The CIGS in the solar cells is not a four-layer stack (and more particularly, di-selenide is not a material per se)!  Rather, it is a chemical compound, a crystal where one copper atom, one indium or gallium atom and two selenium atoms (hence DI-selenide) make up each basic unit.
You CAN talk about the challenge of getting four-layer structures right, but then it’s the structure back contact—absorber—buffer—window.

Relating more to the point of the article: I would deem it slightly offensive toward the venture capitalists to suggest that they all have swallowed NREL’s 20% hook, line and sinker and believe that those 0.2cm²-cell results are what can be achieved on 1m² modules!  It seems a rather basic law of industrial manufacturing that you lose a considerable amount of efficiency on your way scaling up from the lab to the factory and I would think VCs are familiar with it.  It isn’t different with Si cells, is it? The poly-c Si record is at 20%, what you can buy has around 16%. Given a few more years for the industry-scale technologies to mature, I don’t see why the more high-quality approaches (say, HV evaporation as opposed to ink printing) shouldn’t get into the same ballpark.

Sebastian,
Thanks for the correction.  I’ve added your clarification to the article.
As for your second post - I said “partly to blame”, not hook, line, and sinker.
This article is an excerpt from a larger piece and one of the points of that piece is that it is going to take a long time and a lot of effort to reach the potential of CIGS. Perhaps longer than the time horizon of a typical venture fund.
Eric

Ah. Thanks Eric both for acknowledging the correction and for putting my critical reading of your article’s point into perspective. Yeah, it’s true that it may take longer for at least some of the approaches than the investors may have reckoned.
Best, Sebastian

Your reader is absolutley WRONG!!  The Cleaniness of the vacuum has very little to do with efficiency.  The NREL champion cells were deposited at e-6 starting vacuum.  The challenge in making high efficiency CIGS has to do with large area uniformity.

Don’t know who Weldon of “Solar Junction” is or what his agenda is, but he’s 100% wrong about vacuum level. James K is exactly right: 10 e-6, modest vacuum. He’s also right that the challenge is scaling and doing it in manufacturing.

Your efficiency numbers are a mish-mash of cell and module results. The important thing is that three of the references are in production (Global, Shell, now Avancis, and Wuerth). So are Showa Shell and Honda. The numbers for their modules are 11 to 13%. (Except Global who is not as good but they are kind of flexible.) If someone can make 13% MODULES in production, they will be competitive with First Solar and really beat any AMAT factory (6-8%). That’s what all the other companies on your list are shooting for.

The ‘complaint’ about production coming short of lab records is irrelevant. First Solar is 6% short of the CdTe record. Most silicon is 8% or more short of the silicon record. Modules will always be short of the single small cell records for a given material.