New Photovoltaics Made with Titanium Foil

Memorize writes "A company called Daystartech has released a new type of photovoltaic cell which, unlike almost all the cells currently in use, does not silicon. This is based on a thin titanium film. Given the current shortage of solar-grade silicon, and all-time high oil prices, maybe titanium solar panels are here at the right time. The questions are, will they release it as a consumer solar product, and what will be the price per kilowatt hour?"

can't get something for nothing

[kebes]

Food for thought: if your solar sail is using photon pressure, then by coating it in a photoelectric, you're halving its efficiency as a solar sail. Why? Well if your solar sail is a perfect reflector, then the photons bounce off and reverse direction, so the momentum change is twice the initial photon momentum (yes photons are massless but they do have momentum). If the sail is absorbing the photons for electricity, then they are not reflecting, so you merely absorb their momentum, making your forward impulse half what it would otherwise have been.

But, as we all know, solar sails work both by exploiting photon pressure, and solar wind (particles emitted by the sun), so the situation is maybe not that bad.

Re:You know...

[nmb3000]

Ha!

Try copying and pasting that paragraph into Word (I used 2003). Guess what? No grammar errors!

Actually, looking at a diagram on their website...

[Anonymous Coward]

Take a look at this diagram [daystartech.com]. There's clearly a layer of SO2 in there. I'm not sure what that means though as far as their no-silicon claims.

Re:I gotta say...

[DoubleD]

http://www.geek.com/news/geeknews/2003Jun/bch20030 616020429.htm

Re:Slicon Shortage

[dbIII]

Like, you think that titanium, and the equipment required to work titanium comes cheap? Cheaper than sand?

Titanium is also available in sand, most commonly in the form of rutile and ilmanite. Most readers here have probaly eaten titanium dioxide taken from sand, it is frequently used as a white food colouring and paint pigment.

It costs a lot to do anything with titanium because the oxide forms quickly on any exposed surface and takes a lot of energy to break down.

Re:Slicon Shortage

[Rei]

Slicon?

The interesting thing here is that the fastest growing solar cell market is not silicon: it's organic solar cells. They're incredibly cheap, but currently inefficient. However, their efficiency has been growing dramatically. One company, nanosolar [nanosolar.com], claims to have achieved almost the efficiency of amorphous silicon cells. Their patent [uspto.gov] is rather interesting, and well worth a read.

Re:Slicon Shortage

[Rei]

Monocrystalline silicon is incredibly expensive. Polycrystalline silicon (which has largely taken over in the solar cell market) is simply "very expensive". Silicon is common, but pure silicon crystals require clean-room conditions to grow.

Titanium isn't that rare. The ore isn't the primary cost component (like, say, gold). Instead, like aluminum, the main costs are in refining. Unlike aluminum, however, there is currently no continuous production process - only an expensive batch production process. Even the inventor of the process, William Kroll expected to have it be replaced within decades of its implementation in 1940; no suitable replacement was found, however.

Fortunately, it looks like there are some on the horizon. Most interestingly, it appears that electrolysis can be conducted directly on titanium oxide (this has huge potential applications for other hard-to-refine metals as well, and may allow for the creation of new alloys). There's also a aluminum-style molten-salt electrolysis process (FFC-Cambridge) in testing.

Titanium isn't inherently hard to work with, persay; you just need to be properly equipped to work with it and experienced with it. You have to use *very* pure argon in welding, and you have to keep the argon going for longer after you take the heat off. You also have to avoid working it with aluminum tools, which can alloy with the metal and weaken it. Etc.

There are some benefits, though. Impurities in titanium are very easy to spot, as they tend to discolor. Also, titanium is *very* fatigue resistant, and aircraft with titanium structural components have sometimes even been found to be stronger after being flown a few times than when they were built.

OT, Chemical Databases-NIST

[Anonymous Coward]

http://pages.pomona.edu/~wsteinmetz/PChem.htm [pomona.edu]

Re:I gotta say...

[FLEB]

Therefore... [geek.com]

$/W

[EmbeddedJanitor]

The main cost in PV is the energy rerquired to make the silicon. You need a lot of energy to melt the sand, purify it and dope it. That energy costs money.

PV will not be a viable alternative until the input energy is reduced significantly (ie. by a factor of 5 or so).

Re:Slicon Shortage

[Rei]

I should also add that titanium is really just the backing. It's a great backing, given it's strength and condition-tolerances compared to its mass, but it's not what generates the power The cell itself is actually a copper-indium-gallium-diselenide cell - not that it's cheap, either ;)

Re:Slicon Shortage

[Qzukk]

Yes, actually. This isn't just some sand scooped off a beach. Solar panel grade silicon comes from the leftovers after semiconductor grade silicon users have picked through their crystal wafers [worldenergy.org], which is why there is a shortage in the first place, since there is a narrow range of quality ("almost" good enough for semiconductors). As for titanium, my 30 year old encyclopeda says its one of the 10 most common metals on the planet. Titanium Oxide is cheaply produced and used liberally in paint.

Titanium is malleable when hot [speclab.com] (meaning you can flatten it into foil [answers.com]). So producing titanium foil is probably not a difficult task, depending on how hot "hot" is. (Though the article mentions that the titanium foil used is thinner than household aluminum foil. The process [azom.com] looks like it would be easy anyway, but time consuming.)

As for your post on waste products, the most common smelting procedure in use [tms.org] works without catalyst or flux to produce pig-iron and Titanium Oxide, though this process is common because of its use in paint. This process [itponline.com] was recently developed for producing metallic titanium, its outputs are salt (NaCl), titanium, and whatever impurities get washed into the liquid sodium stream and removed later.

Re:Slicon Shortage

[ikeleib]

Also, titanium is *very* fatigue resistant, and aircraft with titanium structural components have sometimes even been found to be stronger after being flown a few times than when they were built.

The above refers to one aircraft in particular. The SR-71/A-12 was found to have a stronger airframe after flight. This is not really due to titanium itself, but rather the gentle heating and cooling that the aircraft underwent with each flight. It annealed the metal, thereby making it stronger and helping to eliminate the fatigue that is normally problematic in airplane structures.

Re:Slicon Shortage

[Anonymous Coward]

Electric arc smelting != electrolysis.

It takes much less energy to melt metallic aluminum than it does bauxite.

Electrolysis, however, is used to make bullion [sp]. Smelt down gold ore. Electrolysis the gold from that ingot. Resmelt the electrolysis product to make .9999% pure gold bullion.

Re:This has all been gone over before...

[jgoemat]

The amount of energy is easy to find, yet you spend time writing seven paragraphs instead of looking it up? You seem to just assume that we couldn't get all the power we need from solar cells.

Well, the amount of solar energy hitting us is around 1.5 kilowatts per square meter at our distance, that would be when the sun is directly overhead (and through the atmosphere). That drops off as a cos of the angle away from the point facing the sun. So if the sun passed directly overhead at noon, at 9:00 am and 3:00 pm (45 degrees away) we would be getting about 70.71% of the energy, or about 1 kilowatt. At 30 degrees lattitude, we would still be getting 75% of the maximum energy as early as 10:00 am and as late as 2:00 pm. So let's say we have 35% cloud cover (some areas could be much more sunny), that should account about for the rest of the hours in the day if we ignore them, but let us go ahead and take an hour off our peak time. So we'd have just three hours of sunlight at 80% (on average lets say) of 1.5 kilowatts, or 3.6 kilowatt hours per square meter per day. let's assume a solar cell that is 20% efficient, so we only get 0.72 kilowatt-hours per square meter per day.

Statistics [usgs.gov] show that hte US used 94.27 quadrillion BTUs of energy from all sources in 1998. From the conversion factors [infoplease.com], that comes out to 27 trillion kilowatt hours. Divide by 365 and that's 74 billion kilowatt hours per day that we need. So we end up needing 103 billion square meters at 30 degrees lattitude to power the entire U.S. That's an area 320.5 kilometers to a side, about 1/7th the size of Texas.

And that's using conservative estimates. Plug in 30% efficency for solar cells, take into account the whole day and not just three hours like I did, and that area will shrink considerably.

WTF? It is still Si

[svis]

Only one person bothered to read the article so far!!!! Well, I am second :) Their "Schematics" clearly show that active ingredient is still SiO, Silicone. They designed a way to put it on flexible substrate. So did many other people. Perhaps they deliver excellent performance cells. However, it does not change the fact that it is still Silicone that moves electrons. It is a clear marketing ploy that conveniently ommits using Si in the marketing blurb.

Re:This has all been gone over before...

[rcw-home]

Photoelectric won't work, won't solve even a small fraction of our power needs, not remotely.

Don't confuse photoelectrics with photovoltaics.

For example, Sandia Labs has a plant currently in operation [sandia.gov] that produces 5MW in 9 acres, by focusing light onto a tower that heats molten salt which drives turbines. It can produce energy 24 hours a day.

The United States' generating capacity a few years ago was 813 gigawatts [geni.org], so at .55 MW per acre you'd need 1.4 million acres for all of the US's energy needs. That's about 2300 square miles or 6000 square kilometers, or about 1.5 Rhode Islands. We have many deserts that are larger than that.

Realistically, you don't need a power generation mechanism to be able to handle the entire United States energy needs before you put it in production. You just need it to be cheap (and cheap after the costs of fighting NIMBY lawsuits are factored in).

Sandia's web site doesn't say what their cost per megawatt hour is, but they do say the entire facility is currently worth $120 million. Since this type of system uses nothing exotic, I would expect economies of scale to change the numbers quite a bit. Assuming a life of 30 years, they'd have to be able to reduce the cost by about a factor of 10 to be competitive with today's rates. It could happen.

It DOES use silicon

[Animats]

This isn't a new non-silicon solar cell technology. It's just a metal base under a thin layer of silicon, instead of a thick silicon wafer. This reduces weight, but it doesn't help cost or performance. It may have space applications.

Their solar cells are made in a wafer fab and have no more than 15% efficiency, like everybody else's.

So this isn't the Great Solar Breakthrough. Sorry.

Re:This has all been gone over before...

[rcw-home]

Sandia's web site doesn't say what their cost per megawatt hour is

But this Department of Energy page [energy.gov] does. They say such systems are currently at 9-12 cents/kWh, but expect 4-5 cents/kWh in a few decades. Which is certainly competitive.

For those wanting to use solar for everything...

[UlfGabe]

This isn't a knock down, but some simple numbers.

1.74×10^17 W : Earths solar constant.(level 1 civ)

3.86×10^26 W : Energy output of our sun. (level 2 civ)

0.82 current level of civilization. (kardashev scale)

solar energy will probably be the only way to go from a civ 1 to civ 2, involving a dyson sphere,

why not get some expertise now, and cover unsightly texas with those solar panels?

BECAUSE SOLAR PANELS are EASILY Damaged, just use maddox's 1000000 penny bomb, and spread them over the solar fields...

The USA and other military countries will not tolerate an easily attackable energy infrastructure. Look at nuke plants. I have seen test video of jets travelling in excess of mach 3 barely denting the outer concrete shell.

solar is good, but first we need peace between all peoples on earth

Re:solar schmolar -- CROPS are the real solar ener

[Fear the Clam]

Remember that "calorie" in American food parlance is actually a kilocalorie in terms of heating up water.

Please carry on.

Reference to Advanced Solar Cells

[Midnight Thunder]

From looking at this page [energy.gov], its says the following:

A variety of advanced approaches to solar cells are under investigation. Dye-sensitized solar cells use a dye-impregnated layer of titanium dioxide to generate a voltage, rather than the semiconducting materials used in most solar cells. Because titanium dioxide is relatively inexpensive, they offer the potential to significantly cut the cost of solar cells. Other advanced approaches include polymer (or plastic) solar cells (which may include large carbon molecules called fullerenes) and photoelectrochemical cells, which produce hydrogen directly from water in the presence of sunlight.

Re:Slicon Shortage

[dhovis]

IIRC, the problem with titanium is not so much that the raw material is expensive. The problem is not even so much that it oxidizes readily (aluminum does too). The problem is that it has a high melting point, and is very difficult to forge and to machine.

Pure Ti-metal has a hexagonal close packed microstructure (HCP). Most other metals have a cubic structure (either face centered cubic:FCC or body centered cubic:BCC). FCC and HCP have the same packing effficincy, but it is much easier to form and move dislocations in a lot of different directions in either FCC or BCC than for HCP. Dislocations are necessary for forging, and forging creates such a tangle of dislocations that it actually strengthens the material.

That is why Apple moved away from Ti for Powerbooks, IMHO. It impossible to economically bend the titanium to form the laptop shell without making the metal so thin that it is way to flexible. So the old Ti-Powerbooks had a Ti top and bottom, with Ti-painted plastic in between. This paint invariably started to flake, which led to lots of complaints. Apple wisely switched to an aircraft grade of aluminum, which can be sufficiently bent and machined to form the entire shell of the laptop, not just the top and bottom.

Anyway, that is the basics. IAAMSBTDNCMA (I am a materials scientist, but this does not constitute materials advice)

Re:Slicon Shortage

[theshowmecanuck]

They don't use sand to produce silicon, they use quartz rock. They reduce (redox reaction) the SiO2 using coal and charcoal to produce the initial Silicon metal(oid). Or to put it in layman's terms, it is smelted in a reaction similar to reducing iron in a blast furnace (except with silicon, it is done in a three phase AC powered arc furnace). The reactions happen in the gas phase at over 1400 degrees C. Chunks of quartz are more suitable since the gases can move between them. Sand just clogs things up... kind of like smothering a fire.

Si02 + 2C = 2CO + Si

Once this silicon is produced, it is refined into super-pure semiconductor grade silicon, or more usually, into silicone rubber pre-cursors. I used to work in silicon smelting R&D and so I have some idea about what I'm talking about. (We built and ran the worlds largest direct current arc furnace during a series of pilot runs in the early 90's to research making lower cost silicon. That was before Russia opened up. After they did, they flooded the market with cheaper silicon, and there was no point trying to create lower cost silicon.) The biggest use of silicon is in making silicone rubber (but not so many boobs any more). The raw material for ultra-pure silicon is taken from the raw material (not so pure silicon) used for silicone production.

Anyway, smelting silicon creates large volumes of CO. CO (carbon monoxide) is highly flammable, on the order of natural gas, and usually burns off to C02 at the top of the furnace bed. (CO could be used as a fuel like natural gas, but it is so poisonous it is not really safe to do so.) Since coal and charcoal are used in the process, other carbon by-products are also released, mostly in gaseous form. E.g. like the stuff that makes up tars and such... a little nasty... but quite small relative to CO and CO2 since the high temperature tends to atomize them. However, some of the coal and charcoal does burn away in the upper part of the furnace (where it is relatively cooler) and before it gets a chance to react. As well as producing some not so nice gases, it is a very energy intensive process. Silicon is never found in elemental form in nature. It must be separated from SiO2, which requires a lot of power, which in turn needs to be produced at generating stations.

As far as silicon used in semi-conductors goes, I'm not sure if they use electrolysis to refine it to ultra-pure levels. Maybe in some sort of deposition process from a gasous phase, but I am just guessing from what I have read in general chemistry related articles. The details of that type of processing are usually very top secret so I am not sure who could or would comment on that. And I mean either industrial secrets and likely in a military sense as well (it is probably of strategic value).

Re:Slicon Shortage

[dbIII]

The problem is that it has a high melting point, and is very difficult to forge and to machine.

It is difficult to forge and machine due to the oxide layer - which is very hard and one of the reasons we use it in the first place (it's mostly used in chemical plants). It isn't really a good choice for a laptop since it costs so much to make and is very difficult to do anything with - and aluminium conducts heat better and can be formed while soft for the aircraft grades - the stuff the early 20th century airships were made out of.

The metal itself has a high strength and hardness, but there are plenty of steels harder than it. The oxide layer is very hard, and as soon as you chip some away it forms again. A slightly harder compound, titanium nitride, is the gold coloured stuff you see plating the tips of cutting tools.

If the oxide is being used in these cells the process may be surprisingly cheap, since the hard bit is reducing the oxide to metal. If it's something else, there may be ways of making it cheaply from an ore - a mineral sand. If a vapour is being sprayed onto a substrate it might not cost a lot either.

I'm not a materials scientist anymore, but for a while when I was I used to teach engineering students how to break things under controlled circumstances - and find out why stuff broke under uncontrolled circimstances.

Re:OT but serious question

[theshowmecanuck]

I wished I was in the post breast implant inspection team, but that was a highly coveted job that I didn't have the seniority to get. So I can't help you with the hardness thing. :-(

Silicon is a metaloid element (sits on the boundary of metal and non-metal). In pure form it is non-conductive, but if you heat it to around at least 1000 degrees, it starts to conduct.

Silicone is a rubber. Simply put, silicon has similar properties to carbon (being in the same family) like being able to form chains. However since it is a much bigger atom, it is a little too heavy to be able to form long chains. When it gets a little too long it pulls itself apart. So you form a chain interspersed with oxygen (which forms very strong bonds) ...Si-O-Si-O-Si-O... and so on... polysiloxane. Then they start hanging other side chains and cross linking, etc. and you get different types of synthetic rubber. Anyway, I switched to programming and IT about 10 years ago (after the silicon project ended), so I would have to pull out my books to any deeper anyway. :-)

Re:Slicon Shortage

[theshowmecanuck]

It's possible. It could be a type of co-generation. The idea is used in a lot of places, but usually it involves using excess heat to produce steam, or waste steam. It would be a good idea.

Something to think about: in order to be flammable you need concentrations of at least 5% CO in air (about the same as needed for natural gas). That's 50000 ppm. To put it in perspective if you were in a room with 800 - 1000 ppm CO for several hours, you would likely end up dead. If you walked into a room with 4000 to 5000 ppm CO, you might not even know what hit you as you hit the floor. It wouldn't be long before you died. So basically, if you used it for a fuel source, it would really suck if the pilot light went out. Maximum OSHA allowable limits in the workplace is 35 ppm. In the middle of typical rush hour traffic (I measured it with a portable meter): 50 ppm! Mind you in industry you are usually indoors where it can concentrate, and often there are very high levels behind it (our offgas lines had 75 to 80 % pure CO... even small leaks were dangerous... we had monitors and venting systems and escape air bottles everywhere).

I'm a big solar booster, but...

[WOV]

...these guys are nothing special. Here's the deal:

88%+ of the world's solar panels are still cut crystals of mono - or poly - crystalline silicon. People know how to work it, they get a reliable if uninspiring 5 - 8% annual decrease in prices from it, and they've been able to ride it through quite a bit of market growth - up over 1200 MW in 2004, up from 750 the previous year, 400-some in 2002, etc. Good stuff.

The thin-film solar people have always made these claims that they're going to cut solar from $2.50 / Watt (mfg. cost) to like $1. And theoretically, there seems to be no reason they shouldn't. But their factories, which are always supposed to just run like printing presses or coated auto glass factories, always end up being much much more finicky and expensive and labor intensive than initial projections, and they end up - not with ridiculous costs, but right back in that $2 / Watt range. Hence the sub 5% market share.

DayStar's technology is not markedly different from any of the other thin-film silicon people (or thin-film CiGS or CiS or the other materials) - their big deal is that they have that superlight titanium foil. It does jack up their manufacturing costs hugely from using like a stainless steel (Uni-Solar) or a plastic / roofing material backer (Uni-Solar / Solar Integrated Technologies) or putting it into a normal framed module (First Solar, Shell Solar,) etc. And thier new little factory in NY there maxes out at I think 30 MW / year (2.5% of annual world production) So why would they do it?

Weight-conscious applications. It costs $10,000 per pound, still, to launch things into space, and people are honestly starting to look at airships again. Even though Boeing Spectrolab has essentially owned the high-value-add high efficiency to weight ratio solar market for a long time , there's still serious money to be had there - they may either settle for being a big player there, or, take DARPA money and use it to work the kinks out of their stuff for two, three years and go to market with a cheaper substrate and a roll-out roofing product, using much less silicon than a conventional process.