Odds and Ends

It has been really difficult to find time to write lately, but I have run across some stories of interest this week that I wanted to comment briefly on.

Wall Street Journal Energy Blog

First, I got an e-mail last night from an editor at WSJ saying that they had started an energy blog (and that they regularly read this blog). I obviously believe energy is going to be an incredibly important subject going forward, so I think this is a good move by them. There is a lot of interest in the topic, and we have a great deal more gas shocks/ethanol legislation/gas tax debates to come. Anyway, here is a link to their new blog, and I have added it to my blogroll.

WSJ’s Energy Roundup

Genetic Engineering and Cellulosic Ethanol

I was in graduate school in about 1993 when I first came to believe that commercial cellulosic ethanol was going to be highly dependent on advances in genetic engineering. I still believe that is the case, and an article this week caught my attention because it really highlights some of the issues:

Scientists working out bugs for use in biofuel production

In fact, some of the things mentioned in the article were things I did in graduate school:

In the hunt for efficiently destructive bugs, scientists for such companies as Diversa Corp. are reaching into the stomachs of cows, tapping hot volcanic vents and hiking deep into Costa Rica’s jungles to trap some of the worlds most ravenous termites. They are finding huge communities of bacteria, fungi and protozoa work together.

I have had my arm in a cow’s stomach all the way to my shoulder (I have a picture somewhere) where I was pulling the contents out to extract the cellulose digesting microorganisms inside. I also conducted some experiments where I attempted to use termite gut bacteria to produce chemicals. To my knowledge, I was the first to attempt this.

But the subtitle of the article, surely to be glossed over by many, is:

Genetically engineered energy crops could still be years away

I must admit to being very surprised at how slowly genetic engineering has advanced. I underestimated the opposition to it years ago, and Monsanto didn’t really help advance the cause with their approach. Still, I think advances will slowly be made in the biofuels arena. These advances will take at least two forms.

First, microorganisms will be genetically engineered that are more efficient at converting biomass. Others will be genetically engineered to provide a different product spectrum. Butanol and higher alcohols are a promising target, as (unlike ethanol) they can be sent down current fuel pipelines and the energy content is comparable to gasoline.

However, this alone will not be enough. The other advance that will need to take place is development of crops that give up their cellulose more easily, or that potentially even start the conversion process internally upon harvest. Right now, a substantial fraction of available biomass is in the form of lignin. This can’t be converted by microorganisms, and it also makes the cellulose harder to get at. Lowering the lignin content would be a good target for selective breeding and genetic engineering if the goal is to produce dedicated energy crops. Of course this problem is precisely why I favor gasification over fermentation, as gasification does convert the lignin into usable syngas.

We have to keep in mind, though, that genetic engineering is not magic. Too often I see people fall into the trap of thinking that there is no problem that genetic engineering can’t solve. In that case, why don’t we just engineer a microorganism that consumes the carbon in dirt and excretes gasoline? Because, while this is probably possible in theory, there is a tremendous amount that we do not know – and may not know for many, many years – about things like metabolic pathways and what specific genes even do. Some of these systems are mind-bogglingly complex, and can’t even be effectively modelled by present computer technology.

A Nice Solar Story

I hold out a lot of hope for solar power. After all, what is biomass anyway? It is very inefficiently captured solar energy. Ditto for gasoline, coal, natural gas, and almost all of our energy sources. It would be much more efficient to directly capture that solar energy and turn it into electricity. If the solar cells could be made cheaply enough, you could start to cover rooftops around the world and use the produced electricity to drive an electric transportation grid.

So, I was happy to see this story:

Cheap solar power poised to undercut oil and gas by half

Of course I am not knowledgeable enough about developments in solar power to know which claims are realistic and which are exaggerated. For instance, having seen some of the more unrealistic claims in the biofuels arena, I have to wonder if this bit isn’t exaggerated a bit:

Within five years, solar power will be cheap enough to compete with carbon-generated electricity, even in Britain, Scandinavia or upper Siberia. In a decade, the cost may have fallen so dramatically that solar cells could undercut oil, gas, coal and nuclear power by up to half. Technology is leaping ahead of a stale political debate about fossil fuels.

Anil Sethi, the chief executive of the Swiss start-up company Flisom, says he looks forward to the day – not so far off – when entire cities in America and Europe generate their heating, lighting and air-conditioning needs from solar films on buildings with enough left over to feed a surplus back into the grid.

The secret? Mr Sethi lovingly cradles a piece of dark polymer foil, as thin a sheet of paper. It is 200 times lighter than the normal glass-based solar materials, which require expensive substrates and roof support. Indeed, it is so light it can be stuck to the sides of buildings.

The article paints a very promising picture, but then again so did early press releases of Xethanol. Maybe someone who is knowledgeable about solar power could weigh in on this story. So how about it? Will we start to cover rooftops around the world ten years from now? Or is there a reason that we might never achieve this?

23 thoughts on “Odds and Ends”

  1. I’ve never heard of Flisom, but CIGS is the great white hope of thin film PV. Nanosolar is the best known CIGS startup, others include Miasole, HelioVolt, Solibro/Q-Cells and DayStar. All talk about sub-$1 per watt (some off the record). When one company makes big claims about a new technology I’m highly suspicious, but when half a dozen companies start building big factories I start believing.

    The other hope for PV is the 40% efficiency cells from Boeing’s SpectroLab and competitors. These are way too expensive for flat plate (unless you’re building a Mars Rover or Solar Raycer), but in concentrator systems cell efficiency trumps cell cost. CPV is not really cost effective with silicon because the “3 Ms” (motors, mirrors and maintenance) cost too much. Since a 40% cell delivers three times as much power from the same motors and mirrors, your overall cost/watt is slashed. There is a large CPV system being built in Australia with SpectroLab cells that claims $2-3/Wp. With refinement $1/Wp or better should be possible.


  2. “when half a dozen companies start building big factories I start believing.” don’t believe the hype, it’s all dumb venture money. Photovoltaic is unlikely to ever provide cost-effective electricity except remote from the grid. The real-world efficiency of solar panels is much much less than advertised even under ideal conditions, and conditions are never ideal. The thin film cells are totally unproven, but i’ve heard that they deteriorate rapidly requiring a replacement cycle of 3 years. Of course, storage is a huge problem and can more than quintuple the cost of a solar system. The goal for these companies is to go for an IPO before they have to prove their technology. Solar thermal has a lot more potential.

  3. You have to see who is talking the price down. It is the startups that want to raise tons of money and need to sell dreams to the fools.

  4. I work for a German photovoltaic system integrator that has been installing thin-film solar power plants for about 5 years. In 2006, about 30% of our installations (and I am talking about multiple tens of megawatts) were made with thin-film: amorphous silicon and more recently Cadmium-Telluride from First Solar (Arizona and soon Germany).

    Our experience with thin-film has been overwhelmingly positive. While the first-generation thin-film did degrade rapidly (leaving lasting bad impressions as evidenced in the previous comment) the current wave of thin-film modules is just as good (and considerably lower cost!) than silicon modules.

    Our tests have shown (and our existing installations continue to prove) that thin-film produces *more* power than silicon modules despite its (for the moment) lower efficiency compared to silicon. That is because thin-film produces also in diffuse sunlight. It begins producing earlier and stops producing later than silicon, yielding a wider, flatter output curve over the course of the day.

    In Germany, investors in large-scale solar power plants demand excellent return on investment. True: their return is driven by subsidies. But the subsidies are 100% performance-linked. No power output, no income, no excuses. These investors are experienced and sceptical – definitely *not* “dumb money”. And they overwhelmingly choose thin-film for it’s superior return on investment.

    So will solar power undercut fossil-fuel-burning power plants in 5 years? Well it certainly will in sunny regions where electricity is already expensive. Think of Hawaii with it’s power-hungry tourism trade and nearly total dependence on imported petroleum. But on the mainland we don’t expect solar power to reach grid-parity until about 2014. And that will also be in sunny regions like Spain and Italy, not Great Britain.

    I do believe thin-film on glass window panes such as described by Flisom will happen but my gut feeling is it will take a decade and Flisom probably won’t be the company to bring it to market. They do indeed sound like entrepreneurs trying to make a fast buck, which annoys me because hype like that hinders the credibility of an industry that actually *is* progressing (steadily and professionally) to grid parity.

  5. To the German Poster, what large scale photovoltaic power plants ion Germany are you referring to involving tens of megawatts?? I have not heard of any existing photovoltaic plants producing even a megawatt of power (at peak sunlight), so please enlighten me? I know there are some on the drawing board – there is google’s proposed 1.6 megawatts and a plant is planned for 8 megawatts in nevada i believe, but those hardly qualify as “large-scale.”

    Also, I know subsidies in Germany are astronomical – something like 50 cents per kWh?

    And what do you do about storage?

  6. OK, I found one 10 MW (peak) plant in Germany that is supposed to be the largest in the world and it doesn’t involve thin films. And 10 MW pv plant in Germany probably produces no more than a 1 MW coal plant over the course of a year. And it certainly won’t save any CO2, since the cost of producing those cells requires more carbon than they will replace in their lifetimes in Germany.

  7. Robert,
    Thanks for another thoughtful article.

    I would like to point out that, as inefficient as it is, biomass has two distinct advantages:
    1. It is self-replicating.
    2. It provides an easy way to store energy, especially once you have converted it into a liquid fuel. Electricity still has a long way in front of it on the battery front.

  8. I would like to point out that, as inefficient as it is, biomass has two distinct advantages:
    1. It is self-replicating.
    2. It provides an easy way to store energy, especially once you have converted it into a liquid fuel. Electricity still has a long way in front of it on the battery front.

    I agree with you on 2, but I don’t think 1 is an advantage over solar. Biomass is self-replicating, but each copy still requires solar power. And that solar power is inefficiently converted to energy via photosynthesis. The real advantage in my opinion is that it does provide an easy way to store energy, as you say.

    Cheers, RR

  9. See Nanosolar’s “Technology” page listing their “7 Areas of Innovation” and their 1/2007 announcement of securing 647,000 square feet to begin manufacturing in San Jose and Germany.

    Anyone have other perspectives on Nanosolar’s technology?

    In the near term storage doesn’t matter if the grid provides other dispatchable generation.

    Longer term, using plug-in hybrid vehicles also as grid-connected distributed storage shows promise. See for example UC Davis professor Dr. Andrew Frank’s presentation to the 10/06 ASPO-USA conference.

  10. Nanosolar is a racket to sucker in fools. Their biggest coup was their company name – combining two words VC’s just couldn’t resist. Of course, storage is a huge problem – if you have one day of clouds, it means all the power has to be generated by other sources, which means you still have to invest all the capital into building fossil fuel plants as you would have without solar.

  11. To the anonymous German poster: Can you provide any studies that compare thin film to polycrystaline performance in real-world conditions. I have only seen one such study, and the results were not compelling, but this was about five years ago.

    To the anonymous poster who hates Nanosolar: If you’re going to talk trash like that, don’t you think you should offer up some evidence, or a basis for your opinion, or even your name? I am not personally close to the company, but I know folks who are, who’s judgement I tend to trust, and they think that Nanosolar’s the real deal. I’d love to see evidence to the contrary.

  12. On the subject of thin-film photovoltaics, there are roughly three general areas: hydrogenated amorphous and microcrystalline silicon indirect bandgap semiconductor, organics, and the Copper-Indium-Gallium-Sellenide direct bandgap semiconductor.

    The silicon solutions have had some problems with relatively low absorption but light-trapping techniques (patterning the cell surfaces) has improved matters greatly. Amorphous silicon suffers from some inherence metastable defect generation, but the microcrystalline cells generally do not. (Aside: microcrystalline is just a profusion of ~20 nm crystallites in an amorphous matrix.) Now companies like Sharp are stacking triple-layer junctions of amorphous and microcrystalline material to get efficiency similar to that of much thicker wafer cells but at a much lower price point.

    Organics aren’t like traditional semiconductor junctions. They have a much lower conductivity so they need to use a basically random arrangement of positive and negatively doped regions all intermixed. The big problem with organic PV is getting the charge out once you’ve generated an electron-hole pair. Actually, that’s generally the problem with thin-film PV. Generally I think that the random deposition method is a bit of a quality control nightmare.

    They also have issues with solvents used for the deposition not fully evaporating and then attacking the material over time. They’re always going to have a shorter lifespan than inorganic semiconductors as well.

    The direct bandgap materials, typically CIGS, generally can’t be made via chemical vapour deposition (i.e. the method for thin-film silicon) to my knowledge. As such, they don’t have a well developed industrial production method. What nanosolar is doing is producing nanoparticles, printing them, and then sintering the particles together. I imagine they may also end up with charge transfer problems but that remains to be seen. In this case, “show me the product,” is probably a good line to invest by.

    All of these methods use an order of magnitude or less semiconductor than traditional cut wafers. At the same time, they require somewhat more advanced production facilities. Still, zone refined semiconductor material is very expensive, so that should put thin-films ahead.

    Your ubiqitious anomymous concern trolls not withstanding, solar has been heavily energy positive for a few decades (typically energy payback is 2-3 years, depending on installation site) and prices have been dropping 20 % for every doubling in production for just as long. This technology will allow this trend to continue.

  13. Have to love ad hominem attacks. I didn’t say the energy payback from pv wasn’t energy positive anywhere in the world – I said it wasn’t in Germany, where subsidies are actually above 65 cents/kWh. In California, the energy payback time may be three years.

    I post anonymously because I know people in the industry and it wouldn’t be beneficial for me if they were to think I thought they were stupid. Although, frankly, I don’t think they are stupid. VC’s like Khosla et al. are smart and mostly earnest, but they are also incredibly naive about energy and damaging our energy policy. Instead of the multibillion califoria solar roofs initiative, for instance, which will cost the state’s taxpayers over $3 billion dollars (and transfer it to the hands of VC’s and corporations), the state should be promoting solar hot water heaters which would produce more useful energy and cost dramatically less (as is done in austria and other european countries).

    It’s not up to me to prove nanosolar is worthless, it’s up to them to prove their claims or that they have something unique at all. So far all they have done is generated a lot of hype. PV technology has been around for a long time and has been making only very slow and very incremental advances in cost and efficiency, and the reality is that it needs breakthrough advances to compete with other sources of energy. At this time, no breakthroughs have been demonstrated by any of these overhyped startups. And, as I mentioned earlier, the storage problem with PV is huge and unadressed.

  14. Concern troll:

    You are, once again, completely wrong. You must think Germans are all mushrooms and get no sun there. The reality is, they get less, not none. Please see my blog post on the issue and in particular reference #1 and references therein which details solar payback for Germany versus ‘southern Europe’:


    You’ll note with interest they’re talking about payback periods of 0.8-1.5 years for thin-film CdTe.

    Also, check out the RETScreen resource which contains insolation data for much of the world:


    Please compare Stugart to Los Angelus for us.

  15. Robert,
    I still think self-replication is important. Think about the difference of harvesting algae from several (hundreds of) square miles vs. covering that area with solar collectors.

    On a related note: Is there a way that you can think of to calculate the available biomass in the Dead Zone in the Gulf of Mexico? That would allow some basic calculations on how feasible it would be harvest that (and clean up the environment, free of charge).

  16. From the “German anonymous poster”:

    A couple of additions to clarify my earlier post.

    Firstly although I am posting from Germany and work in the German market, I am actually American. But “German anonymous poster” is fine as a title…or GAP if you prefer.

    Regarding multi-megawatt plants in Germany: there are quite a lot of them. Until recently, these have been based on silicon modules because that’s what was available in the market place and because the subsidies were indeed very high. But they are lower for ground-scale systems (currently about 42 eurocents or 55 US cents per kWh, guaranteed for 20 years) and that rate decreases 6.5 percent per year. So plants built next year will get 51.5 US cents per kWh for 20 years.

    This decreasing subsidy rate (plus the lower insolation levels in Germany compared to places like Spain, Italy, and most of the USA) has forced investors to find ways to maintain a sufficiently positive return on investment. There are two main ways of doing this: tracking systems and thin-film. Tracking systems increase output by tracking the sun throughout the day. But they also increase the 3 “M”s mentioned in the earlier post. For this reason, we focus on fixed-tilted thin-film installations. No moving parts, dramatically lower costs, and improved power output combine to improve ROI at lower risk.

    To greenengineer: I’m looking for an english version of the study we reference in our company slides. It’s based on data from 2004 and 2005 and comes from the ISET (Institut for Solar Energy and Technik) in Kassel, Germany. The results are not that stark – thin film shows only a slightly higher average output…but at *significantly* lower costs. Hence the higher return on investment.

    Regarding what constitutes “large scale”: currently in Germany systems above 1 MWp are generally considered large scale. Begining around 10 MWp they are thought of as “large scale”. This sounds odd compared to large fossil power plants (or even wind parks) that generate hundreds of MWp. But bear in mind that the scale is increasing rapidly. Just a few years ago a 1 MWp solar plant of any kind was a sensation. Today, it a plant of that size only just barely justifies a press release. Google’s announcement, had it been made in Germany, would have received surprisingly little press coverage.

    If you like “real” large-scale generation, the technology to look into is concentrating solar-thermal, where heat is concentrated on a fixed point and used to drive a stirling engine or heat water to drive a steam turbine. The later method also allows storage, because the heat carrying substance (for example molten salt) retains its heat through the night and can drive the turbine after dark. This type of technology gets WAY too little press coverage. It is reportedly quite efficient, low maintenance, and has been running successfully in California’s Mojave desert since the 1980s. Now it is finally coming back, with larger next-generation plants planned in Nevada and the middle-east. [Off topic: Personally I don’t see why we don’t offer to sponsor plants like these for Iran. If all they want is power, here is a green way to produce lots of it. If they accept, we save money (cheaper than war); if they reject, their claims of “peaceful nuclear power” are invalidated and the US gains urgently needed political clout.]

    Photovoltaics currently has little to offer in the way of power storage, other than small-scale batteries suitable for off-grid systems in telecommunication stations and maybe a luxury vacation home. But seriously who cares?? It is well known that the power is mainly needed when it’s hot and sunny. When the sun goes down or the clouds come out power consumption drops and other sources (wind, for example) cover the load. Why fire up a gas-burning power plant — or worse – an outdated coal-burning plant — when photovoltaic and concentrating solar thermal plants will fire up *automatically* producing clean energy?

    I would also like to refute the idea that solar power “consumes” land. Sorry, but that idea is patently absurd. Solar power on roof-tops (even multi-megawatt systems) consumes NO land. Ground-based fixed systems are quite easily easily removed at the end of their useful life. The posts holding up a fixed-tilted ground system are just like those holding up fencing or highway guard rails. They can be mounted (and de-mounted!) using the same equipment used for freeway and fence construction. It is fast and efficient.

    This is one reason why farmers in Germany love solar power: land that has been over farmed (read: drenched with petroleum based fertilizers and pesticides) can lay fallow and heal, while still being put to productive, profitable use. Where it get’s a bit murkier is tracking plants, where large tracking devices are mounted on huge concrete blocks. This is one reason why my company doesn’t beleive in trackers. In my opinion, trackers (provided it can be demonstrated that the increased output will compensate for the increase in costs and maintenance) should only be installed on “conversion” land, e.g. old military bases or used-up coal mines (also called “brown-field to bright-field” projects) where the land probably can never be used for farming again. In fact, ground-based systems have been shown in Germany to *reclaim* land once thought to be unsuitable for agriculture. The modules give shelter to small plants, which like ground that isn’t baked quite so hard from the sun. Small animals also benefit from the shelter and they bring seeds and dropping, further improving soil quality. Far from “consuming” land, solar power merely “borrows” land it gives it back in better than original condition. And finally, when comparing solar power to fossil power, please remember that oil and coal don’t just fall from the heavens. They require vast amounts of land for mines, refinieries, depots and waste dumps. Have a look at the tar sands in Canada or the coal mines of the Rhein-Ruhr Valley in Germany. Or how about mountain-top shaving in the Osarks. Land committed to fossil fuel production is truly “consumed” – worthless for almost any purpose except maybe solar power and even then I’m not sure because any residual chemicals would have 30 years or more to eat away at the modules and cabling.

  17. Thanks, “GAP” for the extensive posts!

    On the subject of subsidies, I was just talking to a German-American who recalled large subsidies to German coal producers. Do you know anything about how they compare to solar subsidies?

    Second, I just heard German MP Hermann Scheer speak on a book tour for his new “Energy Autonomy”. Sounds like he has it right; I’m looking forward to reading it.

    And to RR: Could you add comment posting DATE to your Blogger profile? I just realized they only show the time.

  18. And to RR: Could you add comment posting DATE to your Blogger profile? I just realized they only show the time.

    Done. And thanks for this suggestion, by the way. I have been annoyed that the date wasn’t shown, but didn’t realize it was one of the options until you mentioned it.

  19. From the “German Anonymous Poster”:

    Sorry folks – I’m still searching for the english version of that German study I cited.

    Darrel: thanks for the kind word. I regret that I had not the time to make my posts shorter (grin). And Hermann Scheer is the MAN!

    Samu: the company with the current bragging rights for the largest planned photovoltaic plant is Juwi. They have just gained approval for a FORTY (40!) megawatt plant on the site of an ex-military base. It will be built exclusively with _american_ *thin-film* panels from First Solar. This is still (grumble grumble) not the 100s of megawatts that a coal-plant or concentrating solar thermal plant can produce, but it does drive solar power another step toward grid parity.

    Now someone will probably complain that that much real estate is wasted on “today’s” solar technology and that it would make more sense to wait until the technology improves. But solar power (like most, if not all technologies) only improves through experience. You can only make progress by doing what you can NOW, learning as much as possible from it, and applying that experience as fast and as intelligently as possible. IMHO the brilliant thing about the German renewable energy law (aside from it’s astonishing simplicity compared to all other solar incentive programs I’ve heard of) is that it sets *today’s* investor on relatively equal footing with tommorow’s investor. The visionary who invests today produces less power but earns a high (read *motivating* financial return) due to today’s higher feed-in tariff. Tomorrow’s investor can earn the same return, despite lower feed-in tariffs (lower, but still guaranteed for 20 years) thanks to the higher power output from tomorrow’s solar panels. And of course the demand for that level of return forces everyone in the value chain (from silicon producers right down to the system installers)to lower costs and pass the benefits on to the customer. Besides, when the modules stop producing, one can apply the “R-Squared” strategy: “Recycle and Re-Power”. (Sorry, Robert: I couldn’t resist).


  20. Robert Rapier said…
    I agree with you on 2, but I don’t think 1 is an advantage over solar. Biomass is self-replicating, but each copy still requires solar power. And that solar power is inefficiently converted to energy via photosynthesis.

    Robert, I’m bringing this thread back up to the top as I only reviewed it a few days ago and noticed the self-replicating idea.

    While I’m not going to get into the plant-it-and-forget-it part of self-replication (although nanotech grey-goo is a fun idea) one part I’d like to broach is the self-powering feature: Solar Breeders.

    You wouldn’t happen to be familiar with the concept would you? Back in the 80s or so, the concept of using the output of solar panels to power further panel manufacture was introduced.

    Aluminum and silicon electrolysis, electromagnetic casting and all of the other energy inputs can easily (relative term) be done with solar electricity. After a certain critical threshold, the fossil inputs go the way of whale oil. 🙂

    A non-expert website has some ideas about it:


Comments are closed.