Tellurium, one of the rarest elements on earth, was once not thought to be very useful. According to USGS and a Mining Journal Review article, half of its traditional use is as an alloying agent in iron and steel to improve machinability; 25% of it is in catalysts and chemical use; 10% of it in alloying with non-ferrous metals like copper and lead; 8% of it is in electronic application and the remaining 7% in other applications, including as pigment agent in ceramics.
Annual global tellurium production is about 170 tons to 200 tons, based on various different estimates. It's mainly produced from the anode slime accumulated during electrolytic process of copper refining. According to this detailed analysis, copper produced from different places contain vastly different tellurium content. Typically, one ton of copper contains 100 grams of tellurium and only 33 grams are extracted and produced using existing technology. That caps the current global tellurium production at no more than 400 tons, without major investment to improve the tellurium extraction efficiency, assuming globally 12.4 million tons of copper is produced using electrolytic process per year. Because of the low quantity and thin revenue of tellurium in comparison to the revenue from main copper product, copper refineries are UNLIKELY to invest money, time and effort to improve the tellurium extraction rate, unless tellurium price goes up a lot from here, approaching gold price levels.
According to USGS, tellurium price started 2004 at $10 a pound. By the year-end it reached $22.50 a pound. In 2005 the price quickly rallied to $130-$180 a pound in mid year, then flat down to $100-$130 a pound. In 2006, it once again ran up to $155 a pound and then settled for the year at $50-$75 a pound.
So what prompted the rapid price raise? First Solar's (FSLR) CdTe solar panel was one of the demand factors. FSLR produced 60 MW in 2006 and about 20 MW in 2005. At 8 grams of tellurium per panel and 60 watts per panel, that's 8 tons tellurium consumed in 2006 and 2.7 tons in 2005. That's barely 4% and 1.4% of the supply. It's a factor in demand increase, but not the major factor.
The main Te demand increase was from other applications. CD-RW discs use tellurium, as do DVD-RW discs. And even later, ReWritable Blu-Ray DVD discs were developed by Panasonic, using a material called tellurium-suboxide-palladium.
Recently, Intel (INTC) announced a new type of phase change memory chips to start mass production in later 2007. This also uses tellurium. To understand the background of this break through, you need to read an old article: A 30-year memory problem solved?
All those electronic applications mentioned above, CD, DVD and memory chips, relate to the same phase change material called chalcogenide, which contains tellurium. Chalcogenide, the material used in ovonics, is really the cultimation of decades of scientific research on amorphous materials, which was once considered of very little practical use, but of academic interest only.
I originally connected these dots from an article I read entitled "Is There a Tellurium Rush in the Making?". Kudos to Sergio Garcia de Alba for submitting the article Energy Conversion Devices: An Amorphous Gem, which finally allowed me to connect the dots. It's all related to tellurium. Energy Conversion Devices Inc. (ENER) was funded by Stanford Ovshinsky, who invented the amorphous semiconductor materials and coined the name Ovonics. He has numerous important inventions. Chalcogenide, which is a tellurium containing material, is one of them.
Today, after many decades of quiet research and perfection, chalcogenide has suddenly become a very very useful material, having revolutionized and is continuely revolutionizing the whole electronic industry. I like to compare the precious element tellurium to a supernova. It's been quiet and ignored for so long, but all of a sudden it erupts into an extremely bright superstar. The tellurium supernova shines so brightly that it could KILL.
Let's follow the ENER article a little bit:
...a little company called Ovonyx...licensed its memory technology to semiconductor giants like Intel (which is a partner in Ovonyx), Samsung, Elpida, Hynix, Qimonda, and ST Microelectronics. The memory technology is based on chalcogenidephase change memory. It is expected to replace NOR flash memory, and could also eventually replace DRAM and NAND flash. Samsung has announced production of a 512Mbit part in 2008, and Intel a 128Mbit part that could arrive as soon as the end of this year or early 2008 (Intel's part codename is Alverstone). ...What's interesting for investors is that phase change memory could become a $40 billion+ market in a few years.
This phase change material thing is really HUGE! $40B+ market just in memory chips and we have not even included all the rewritable CDs, DVDs and Blu-Ray DVDs. And it ultimately could also replace the hard disk drives so future computers would no longer need a hard disk drive.
If you are not shocked so far, then read this article about phase change memory. Let me quote from page 2:
A person using a computer with PRAM could turn it off and back on and pick up right where he left off -- and he could do so immediately or 10 years later. Such computers would not lose critical data in a system crash or when the power went out unexpectedly. 'Instant-on' would become a reality, and users would no longer have to wait for a system to boot up and load DRAM. PRAM memory could also significantly increase battery life for portable devices.
How wonderful it would be!
Needless to say, this whole new chalcogenide based electronic industry will consume a lot of tellurium, probably all the global tellurium production and then some more. It will drive the price of tellurium to a crazily high level, maybe at gold price, maybe at platinum price, and in doing so, it will kill a lot of trivial, low added value industry users of tellurium.
The kill of this tellurium supernova will probably include FSLR. This company is most vulnerable because it uses a lot of tellurium in its solar panels, any dramatic tellurium price will increase its cost to the level that it can no longer have a profit margin. When a business no longer has a profit margin, it cannot survive. But the tellurium kill probably will go behind that. Tellurium as an alloying agent will probably have to end, as will tellurium used in portable electronic beverage coolers.
My advice to people would be to buy and hoard some tellurium metal ingots if you can still find anything at decent price. If you have FSLR long positions, I recommend that you sell them before it is too late. Tellurium, such a scarce and precious natural resource, is one of nature's best gifts to human kind, and it was never meant to be used on trivial things like generating a few watts of solar electricity; rather, it should only be used in high value added and far more useful things like advanced computer memory chips.
FSLR started on the wrong technology using the wrong material, the extremely toxic cadmium plus the extremely rare tellurium, a deadly combination leading this otherwise aggressively growing company onto a death march. It's a tragedy of nature's making, not the management's fault. But the FSLR management really need to wise up and realize that their sole CdTe product will lead them to nowhere. They must diversify into other technologies, or they may have to shut down business just a few years down the road.
Full disclosure: I currently do not have any ENER position but is looking for opportunity to buy some. I have short positions in FSLR and am planning to short more when it starts the eventual collapse. I am also actively buying physical tellurium metal ingots as investment.
Related Articles
|
This article has 56 comments:
- borisb
- 339 Comments
Dec 02 09:32 AM- NorCaliDave
- 1 Comment
Dec 02 11:00 AMJust a brilliant article, I am so impressed...
- calisdad
- 6 Comments
Dec 02 11:07 AM- Mark Anthony
- 294 Comments
My Website
Dec 02 11:40 AMAccording to this latest article on Arizona Republic:
www.azcentral.com/ariz...
The CEO suggested that the company "has found opportunities to buy tellurium from national and international sources, which it can then provide to its regular suppliers to make the compound."
That is so ODD. I was so surprised that they needed to go to that extreme to secure tellurium supply, trying to buy tellurium themselves and then pass it to their CdTe compound suppliers to make CdTe. They did not need to do this in the past. They do not need to do this with cadmium or any other material supply. They do not need to go buy sand and pass it to their glass suppliers to make the glass.
So if you can read between lines it really shows how bad a problem the global tellurium shortage has become! It's like you are now required to bring your own fresh cut raw beef to a restaurant to have steaks served to you! It surely looks like a beef shortage is going on. They did exactly that kind of thing in North Korea during the famine years, asking people to bring in their own meat if they want to eat meat in a restaurant!
- Pouting Thomas
- 3 Comments
Dec 03 04:37 AMThink twice before taking physical delivery of tellurium. When even a microscopic amount gets inside the human body, the victim develops a horrific stench that can last for the rest of their life. Which tends not to be all that long due to suicide.
I question the effect of the digital memory market on tellurium demand. Start with some absurdly conservative assumptions: that a bit cell is a 150 nanometer square, that the phase-change film contains tellurium equivalent to a 75 nanometer layer of elemental tellurium, that the memory uses 10 cells to store 8 bits, that the memory will sell for US$0.10/GB, and that sales will be US$100 billion/year. A gig of memory would contain 1.7E-11 cubic meters of tellurium, or 1.1E-7 kg. Multiply that by 1E+12 GB/year, and you get 105 tonnes/year of tellurium, or 116 U.S. tons/year. That's only half the current world production. First Solar doesn't have to worry about memory hogging the tellurium. The worry, if any, would be phase-change optical disks, which have huge bits and thick layers.
Regarding securing supplies, CdTe compounding is a low-margin, low-volume specialty business. The compounders are contractors, CdTe creation is a minor business, and they probably have no interest in learning to be international tellurium brokers.
"They do not need to go buy sand and pass it to their glass suppliers to make the glass."
Oh? Where I work, we are currently having impurity problems with a specialty glass product. We would LOVE to talk sand with the production folks, instead of working our way through the layers of an international commodity distribution chain.
- Mark Anthony
- 294 Comments
My Website
Dec 03 02:53 PMYour estimates used some ridiculous numbers. Clearly you don't have any idea of material usage in the EE industry. Look at indium as an example. Most chips are made of silicon, not indium. Indium has very limited usage as trace doping material, and used in light emitting diodes. That's all. But the electronics industry uses a couple thousand tons of indium per year. In comparison I expect the electronic industry to use a thousand ton level of tellurium per year. They will jack up tellurium price to make sure the copper refineries have enough incentive to provide that much. But a lot of low end tellurium users, like FSLR, will be killed in the way.
- Pouting Thomas
- 3 Comments
Dec 03 11:30 PMThe comparison to indium is bogus. Indium is not widely used as a trace dopant. It is used for bulk materials like solders and semiconductor substrates, and in indium-tin-oxide transparent electrodes on LCD displays. LCDs apparently use a lot: the layers are thin, but they make 'em literally by the acre. So yeah, they zip through a ton at a time.
By comparison, the phase-change material in memory devices is only used on small dice in an extremely thin layer.
Incidentally, my numbers are ridiculous, but not for the reasons you think. They are way, way conservative. The market for memory devices is a lot less than US$100G/year. Memory cost will be a lot more than US$0.10/GB, at least for the next couple of years. Practical memory cells are likely to be considerably smaller than 150 nm. Only around 25% of the chip area will be covered with phase-change material; the balance being etched off and recoverable from the waste stream.
If you disagree with the assumptions I listed, please state which ones, what you think they should be, and why. This is physics. If you're not calculating hard numbers from first principles, you're lying.
"As for FSLR, from head to toe it started as a CdTe based company, and has no other product. There is no feasible way for it to switch to a different technology, other than liquidate all existing assets and start over again from scratch on a different technology."
No. First Solar is a glass panel company having a fling with CdTe. All plausible high-efficiency materials will be rather fancy and expensive. To keep costs in hand, they will HAVE to be used as thin films, sandwiched between sheets of inexpensive glass for protection and mechanical strength. And First Solar has that problem SOLVED. Can they handle acres of glass? Check. Can they get the glass extremely smooth and clean? Check. Can they put electrodes on that glass that don't block too much sunlight? Check. Can they make electrodes that have acceptable electrical resistance? Check. Can they make reasonably rugged assemblies that stand up to weather? Check. Do they have a customer base and distribution channels? Check. That's no guarantee they will pull off the transition to a different film composition, but they have as good a chance as you could ask for.
One-trick ponies like CdTe are generally regarded as a stop gap solution. It's just not plausible that any simple material will have a good match to the entire visible spectrum. To get light conversion efficiencies up in the 50-75% range will take layers of multiple materials, or a fancy structures (keep your eyes on the folks growing vertical nanotubes on surfaces). Those technologies are being developed right now and will become increasingly practical over the next 10 years. Whoever has a glass panel plant sitting around waiting for them will have tremendous first mover advantages.
The real structural risk is that someone could invent laminated plastic PV cells that are distributed on a roll. The glass panel business would be toast.
- David Lentz
- 353 Comments
Dec 02 12:28 PM- saxonglass
- 3 Comments
Dec 03 07:23 AMFirst, the term "chalcogenide&quo... does not just refer to Te. Here is a lengthy definition - with caveats.
he chalcogens (with the "ch" pronounced with a hard "c" as in "chemistry") are the name for the periodic table group 16 (old-style: VIB or VIA) in the periodic table. It is sometimes known as the oxygen family. It consists of the elements oxygen (O), sulfur (S), selenium (Se), tellurium (Te), the radioactive polonium (Po), and the synthetic ununhexium (Uuh). The compounds of the heavier chalcogens (particularly the sulfides, selenides, and tellurides) are collectively known as chalcogenides. Unless grouped with a heavier chalcogen, oxides are not considered chalcogenides.
So, chalcogenides need not contain Te, and some of the possible phase change materials do not. (see patent literature)
On the other hand, other uses of Te, particularly for thermoelectrics are becoming more important.
Clearly Te has the potential for increased use. However, as with any such high-priced material, research continues in order to find alternatives. And alternatives exist.
- Mark Anthony
- 294 Comments
My Website
Dec 03 02:28 PMIt's true that the dictionary definition of chalcogenide includes a broad range of materials beyond just tellurium. When patenst are applied they want the definition to be as broad as possible so as to cover as many things as possible. But the fact remains that in all existing applications of chalcogenide, tellurium is the material that is being used. It is a fact of reality.
No one can predict future technology progress. The tellurium based chalcogenide material has been researched for 4 decades and improved upon for 4 decades. They MIGHT find a substitute for tellurium, but it might happen in another 4 decades, definitely not tomorrow morning, definitely not 5 or 10 years down the road. Definitely not soon enough to save FSLR from its innevitable demise.
- alpha24seven
- 170 Comments
Dec 03 06:02 PMIt seems the entire thesis is based on a one way street. It is technology and technology moves so fast that the minutiae of this argument is pointless. For example, if NanoSolar can do what they say the can, then the production/consumption of Te is a moot point.
- saxonglass
- 3 Comments
Dec 03 09:25 PMSorry Mark - but you are wrong. It's not just a dictionary definition. Go to any scientific journal - Chalcogenide is not a product - - it is a family of glasses/crystals that contain Se, Te, or S, or each. I have researched chalcogenide glasses for more than 3 decades, and rarely did they contain Te. There are dozens of technical uses beyond memory - for instance infrared lenses, infrared fiber optics, Hundreds of compositions can be used - but it is true that right now glasses containing Te makes the best phase change memory material. That does not mean that Te cannot be replaced. It is not a magic material that imparts some heretofore unknown property to the glass.
If the cost goes up - it will be replaced.
- Mark Anthony
- 294 Comments
My Website
Dec 03 06:43 PMTechnology is NOT playing 5 minutes magic show. Progress of technology takes extremely long time to get a break through. They have researched the chacogenide/tellurium thing for 4 decades. Intel alone poundered this idea for 3 decades and now they are ready to start massive application, it's not going to change overnight any time soon.
As for FSLR, from head to toe it started as a CdTe based company, and has no other product. There is no feasible way for it to switch to a different technology, other than liquidate all existing assets and start over again from scratch on a different technology. That is not much different from a Chapter 11.
FSLR will be DOOMed, the demise will come within two years.
- Saturn V
- 5 Comments
Dec 03 08:16 PMI am skeptical about the claims.i have seen so many technologies claim that they were going to dislodge DRAMs, and waste a lot of investors money in the process.The pace of progress in the semiconductor memory is so fast, the volumes and the economies of scale are so huge, that it has been impossible for any challenger to make an impact.Unless the traditional memory technologies run into a major roadblock, the prognosis for a challenger is grim.
NAND FLASH memory is also on a tear, but it has some performance limitations like number of erase cycles etc.That may present an opening for Ovonics style devices to exploit, since the new technology will inevitably cost more, and it needs a strong foothold in the performance area to have any chance of survival.
- Saturn V
- 5 Comments
Dec 03 08:25 PMOvonics has had the same "potential" for the last 35 years.Intel worked with Ovonics in the early 70's, and scrubbed it. Almost 10 years ago, Intel went back to Ovonics, and it is still a no-show.
I am skeptical about the claims.I have seen so many technologies claim that they were going to dislodge DRAMs, and waste a lot of investors money in the process.The pace of progress in the semiconductor memory is so fast, the volumes and the economies of scale are so huge, that it has been impossible for any challenger to make an impact.Unless the traditional memory technologies run into a major roadblock, the prognosis for a challenger is grim.
NAND FLASH memory is also on a tear, but it has some performance limitations like number of erase cycles etc.That may present an opening for Ovonics style devices to exploit, since the new technology will inevitably cost more, and it needs a strong foothold in the performance area to have any chance of survival.If it can meet the write erase cycle challenge at almost the same cost of NAND FLASH, it may end up replacing rotating memories for ultraportable computers.
But I would not buy Tellurium stocks as yet.
- Mark Anthony
- 294 Comments
My Website
Dec 03 10:46 PMThere is no tellurium stock to buy. The only way to speculate on tellurium is to purchase and hoard physical tellurium metal ingots.
Tellurium contained in chalcogenide material as used on CD-RW and DVD-RW discs is a commercial reality. You can buy them at any electronics store. And I see a whole lot more in stores than a few years ago. That alone is a huge tellurium demand.
Intel already announced and probably released the phase change memory. It is a reality NOW, not in the remote future. Samsung also released phase change memory. So things are happening right now.
Tellurium price has been going up fast, way much mroe than 2% up per week. That is a fact. So we are experiencing global tellurium shortage right now and see a rapid price raise.
- User 127383
- 1 Comment
Dec 03 08:22 PMMarket Cap: 17.8 billion
EPS (ttm): 1.37
Overvalued and bloated and bound to fall. When is the question...Smart guys will take profits now and wait for dip
- JSY
- 2 Comments
Dec 03 09:52 PMwww.nrel.gov/pv/cdte/
- Mark Anthony
- 294 Comments
My Website
Dec 03 10:55 PMI am sorry, the tellurium documents at the NREL site are NOT about FACTs, but about speculations, a pretty naive and amateurish speculation at that. They have already been proven wrong by the tellurium market trend in recent years.
This research paper is way much more credible, and it questions the original NREL assessment:
www3.interscience.wile...
Of course the actual global tellurium market itself is way much mroe credible. Price has been going up fast and furious. Supply is tight and hard to come by. I wanted to buy a couple hundred pounds and I need to wait several months to get it.
- Augustus
- 159 Comments
Dec 04 03:40 PM- FSLR Analysts
- 6 Comments
Dec 05 12:38 PM- Mark Anthony
- 294 Comments
My Website
Dec 08 12:37 AMwww.thomasnet.com
Please note, do NOT order powders, order metal ingots for investment. You probably need to wait a few months to receive your order.
- parker
- 13 Comments
My Website
Dec 05 06:04 PM- Mark Anthony
- 294 Comments
My Website
Dec 08 01:34 PMcc.talkpoint.com/LEHM0...
I noticed this portion of the talk:
"(8:05)What we are TRYING to achieve, is a 90% confidence that we have 5 years of supply assured post full ramp of these facilities. And that confidence is demonstrated not only through our strategic inventory that we are holding, but also through a long term contract we are engaging with our supply base. ..."
Note he said they are TRYING to reach the 90% confidence level, that mean they still do not have that confidence. And they are probably counting on maybe some tellurium they hoared from some years ago.
No business can operate sustainably if it rely on strategic hoarded stockpile for regular supply. Such hoarding is only meant to be used during emergencies, and should never be touched to supplement regular supply. I do not think FSLR has tellurium source to sully its Malaysia factories. And it certainly does not look like they can grow at all, because there is no possibility they can add more factories and find more supply, if they are already relying on strategic hoarding of materials.
That, combines with the CEO's admission that they now need to find tellurium themselves and pass it to CdTe compound suppliers, in order to obtain CdTe, despite already have a supply contract with the CdTe suppliers, it really shows how bad a shape FSLR is in, regarding tellurium.
I have send this article to FSLR, and many people have also forwarded the article to FSLR seeking their comment. So far they have avoided dressing the public concern head on, in recent presentations they avoided using the word "tellurium" altogether. That really tells you something!
I reported the CFO to SEC on his outrageous "terawatts of tellurium" claim, and I made it publically known that I did report the CFO to SEC, and that I got a response from SEC. So far FSLR said nothing about it.
Why do they let me beat them on tellurium like that and never responded to me so far? The absolute silence from FSLR is deafening!
- fslrte
- 2 Comments
Dec 09 05:06 PMOn the consumption of raw material, comparing silicon and tellurium, I was wondering how you come up with 8 grams of tellurium per 60 Watt panel.
If we take a 300 microns wafer technology, the use of silicon is around 10 g / Watt.
Suntech is in the range of 200 microns and is reporting 7 grams / Watt.
First solar CFO is saying that there usage of material is reduced because they use only a 3 micron layer of material. I would be tempted to conclude that they use 100 times less material, leading to 0.1 grams / Watt or 6 grams / panel.
Now, we still need to take into account the fact that tellurium is only 1 part of the CdTe compound. It is actually 6 grams of compound, not 6 grams of tellurium.
I am not sure of this but I believe the CdTe compound is made of a weight proportion of 47 % of Tellurium and 53 % of Cadmium. Can anybody verify this?
This would mean that the use of Tellurium per panel is 0.47 x 6 grams = 2.82 grams or 0.047 grams / Watt
We can continue with pricing comparison. Spot market of polysilicon goes up to $300 / kilograms but Suntech CEO says that they now have new long-term contract at $40 / kilograms. At 7 grams / Watt, we then have 28 cents of silicon per Watt.
If First solar is using 0.047 grams / Watt (I.E. 2.8 grams per panel), to reach cost equivalence with the silicon (28 cents), they would need to pay the tellurium $5,957 per kilograms.
It is obvious that while polysilicon trend is to go lower in price, Tellurium is taking the opposite direction. Investing in tellurium might be a good idea but it seems a long way before getting to 6,000 $.
- Mark Anthony
- 294 Comments
My Website
Dec 09 09:21 PMFSLR meantioned in many occasions they use 3 microns thick CdTe. As late as Dec. 6 conference call the CFO re-iterated 3 microns, at 00:05:10. So that has not been changed. Using density of CdTe being 5.85 grams per CC, and the area of 2x4 feet panel is 0.75 m^2, Te is slightly heavier than Cd so it's 53% of the CdTe weight. it's easy to calculate. Another clue is in a NREL cadmium fact sheet they meantion 3-9 grams for 1 to 3 micron CdTe layer. So for 3 microns you have 9 grams Cd per m^2, or 10.3 grams Te per m^2, or about 8 grams per panel.
My original calculation assumed 60 watts per panel. They now raised to 70 watts per panel. That results in about 114 kilograms of Te per MW products, or 114 metric tons per GW.
That's about 0.114 gram per watt. If silicon is paying 28 cents per watt, to beat silicon, FSLR needs to have tellurium priced below $2.46 per gram, or $2460 per kilogram.
Note FSLR does not use pure tellurium, they use high purity CdTe compound, which has a big price premium over low purity tellurium. As of today, GoodFellow is selling 99.95% pure tellurium for $830 per kilogram. It has a high premium but it looks like the going price.
The economy 101 of supply demand says that when there is a shortage, the price must keep going up until there is eitehr increased supply, or demand destruction to eliminate the shortage. Demand destruction in this case means FSLR going out of business.
- fslrte
- 2 Comments
Dec 10 09:27 AMOne more iteration on this subject:
First Solar Panel size 2 by 4 feet: 0.6096 x 1.2192 = 0.7432 square meter.
Thickness 3 microns or 3 x 10 E-6 meters.
Volume used : 0.7432 x 3 x 10 E-6 = 2.2296 x 10 E-6 cubic meters or 2,2296 cc (cubic centimeter)
CdTe weight 5.85 g per cc
Weight of compound per panel 2.2296 x 5.85 =13.04 grams
Te is 53 % of compound so Tellurium weight per panel is 13.04 x 0.53 = 6.91 grams.
If panel generate 70 Watt, Tellurium used per Watt is 6.91 / 70 = 0.09876 grams.
To reach the polysilicon equivalence of 26 cent in cost, the Tellurium price would then need to reach $2835 per kilograms.
I think that Tellurium can and will reach that price rapidly especially when a second and third CdTe panel manufacturer will start production.
I have ignored in this calculation, the cost of cadmium and the cost of preparing the CdTe compound from the separate elements.
If any Analyst is reading this, can anyone of them ask, during the next First Solar public presentation, the question of the cost of the CdTe compound per Watt or the cost of the Tellurium Per Watt at this stage. Polysilicon cells manufacturers are discussing openly this matter, I think First Solar should do the same.
I think that the price of tellurium is more critical that its availability.
I read that the total resource of Tellurium on earth is evaluated at 800,000 tons but it is still in the ground !
As it is a by-product, its extraction rate is dependent on the other metals (mainly copper) extraction. Therefore, price will definitely go up.
- Mark Anthony
- 294 Comments
My Website
Dec 10 10:47 PMOf course silicon solar panel maker can openly discuss silicon supply, because they know it's a temporary short term issue, not a long term issue. FSLR so far could not frankly discuss the tellurium issue publically because they are well aware this is a huge problem for them. I send my article to them to seek comment, so far the response is deafening silence.
I am going to buy more tellurium in a month or so. It looks now that few can even argue against the case of tellurium bull. So invest in physical tellurium is definitely a good idea. Do a Yahoo search for the NREL document using "tellurium availability". I guess it's the 6th entry. It has all you need to know to find out where to buy tellurium.
Your 800,000 tons figure is incorrect. You need to check out the USGS documents:
minerals.er.usgs.gov/m...
- WTF
- 1 Comment
Jan 02 03:48 PMwhile the article and in particular counter-points are quite interesting I think this is much ado about nothing. Why don't you site all the evidence how basically ever since the history of man there has only been ~20 years of refinable oil left. Then they discover (read "open") a new deep sea bed once the price goes up. It's more about what's practical/profitable at the market price that dictates the supply, and yes prices of most materials increased heavily in 2007 and will probably continue to do so in 2008. I mean look at PET, is there a global PET shortage in the wake since prices increased 20% on the year (>50% over 2 years) which is way out of line with traditional variance of << +/-5%?!? I doubt it. My understanding is the shortage here is primarily from the oil refinery capacity which has been heavily manipulated (at least in the US) from the supply side to maximize profitability.
Also your source on 12/03 about why the government is wrong about Cd-Te supply/uses is a site selling a book of papers that doesn't really have any free published information there... Say what you will about government corruption and lies, a release from NREL in 2003 is more up to date and credible than a series of papers written in BEFORE 1995 and then compiled as a new book in 2006.
The bigger concern is, as stated, that the (printed) flexible film PV circuits are arriving. Advantage there is the production output increases while costs decrease. I believe capital investment also decreases for this approach so it's not like FSLR, Miasole or anybody else can't get into this new approach quite easily if they pay licensing or find a work around...
If you can reach 90% 5-yr supply stability IMHO you're doing pretty damn good. I think if you find numbers for other companies they would be in similar positions. Even massive established companies rarely have more than a 1-year contract for the supply of materials or finished parts. In the case where a material is critical or the price is exceptionally volatile they might structure a longer term contract to minimize inflation exposure and guarantee a minimum supply but I think you're reading into this way too far.
Finally in science terms, when they say "in 20 years" it means the same as when your contractor says it will "take 2 weeks." It means they don't have an idea but hope it will be done by then. I trust from your silly cowboy hat that you've worked with contractors before.
If you look at the history of "inventions in 20 years" from the previous 40+ years you will laugh. Also just because something was proved in the lab doesn't mean it will make it to market anytime soon, typically this is another 5~10+ years of production development before the have a beta process that can be used.
- Mark Anthony
- 294 Comments
My Website
Jan 10 07:53 PMThe latest quarterly filing from 5N Plus (VNP.TO), the dorminant CdTe supplier to FSLR, failed to show any significant increase of sales revenue, despite of FSLR recently more than double its production as its Germany factory reached full production, bringing production lines from 3 to 7, hence needing more than double the CdTe material.
My educated