The airport at Fornebu 1998. The undersigned was newly graduated engineer and had sought a job at a new factory that would make “solar components.” Now called ScanWafer AS factory was far away from some places I had been before, Glomfjord in Nordland, and even for the entrepreneur Alf Bjørseth matched it best to take job interview at Fornebu.
Over a cup of tea with lemon was I was told that solar was the futures industry. I knew so far already, that was why I had sought the job. The technology was called “multicrystalline silicon”, which was a well-known technology, but now would be the most cost effective, so why should they be able to compete even with the big ones. He said that Norway was ideal for such production because we had cheap energy and plentiful supply of cooling water. Finally said Bjørseth that we really ought to do in Norway was to create “monocrystalline silicon” because it required a lot more energy, but in return gave better and more efficient solar cells.
Alf Bjørseth is a very persuasive person and not long after I had pulled myself up by the roots and lived in scenic Meloy. Next door was the sister company SiNor started and quite right, here was made monocrystalline silicon.
The most perfect man has created
monocrystalline silicon or Czochralski Si, are, quite objectively , the most perfect we humans have managed to create. 4 billion atoms in a row, without any deviation. Large sausage-shaped “ingot” which is a single crystal from bottom to top, that is, the atoms are placed with the same pattern and the same distance throughout.
There are no “grain boundaries”, the boundary where two crystals having different directions meet, and no small displacements in the crystal that gives imperfect lines in the material, known as dislocations, which leads to much hassle for the electrons in the solar cell.The silicon starting with To grow the crystal has a purity of 99.9999999 percent; contamination level corresponds to the amount of caffeine in a cup of decaffeinated coffee mixed with 10 tons of water. This product is the heart of all microelectronics, all computers, phones and gadgets we use. And it thus provides the best solar cells: Up to 22.5 percent of the energy in the light is converted to electricity in such panels.
In 2004, the solar industry really taken off, in Norway and internationally. It is still multicrystalline silicon apply. New, large factories are starting up in Glomfjord and in Porsgrunn, ScanWafer was REC, which also has production of all that is, from silicon feedstock to finished solar cells. It was harder with monocrystalline silicon: SiNor went bankrupt, but were resurrected under the name SiTech.
Much of the reason why it is not easy to break through with monocrystalline silicon, is that it is terribly difficult to make. The fact that we actually manage to produce these absolutely perfect crystals is our time really large industrial feats Silicon melted in a crucible made of quartz with a purity level that matches silicon using.
Then immerse a seed crystals, a stake in around one centimeter in diameter, perfectly carved from a crystal you have created earlier into the melt. The temperature is controlled extremely carefully while seed crystals are rotated and pulled out of the crucible. And presto! Hanging from seed crystals there is a new crystal, just perfect, which can be both long and thick (two meters high, 20 centimeters in diameter and 150 kg heavily) ..
Golden Age of Norwegian solar
2008 golden age for Norwegian solar manufacturers. REC was very, both at home and abroad. ELKEM was in full swing to build its solar silicon factory in Kristiansand. SiTech had also become part of the REC, and major developments were headed. Alf Bjørseth still had faith in monocrystalline silicon and started up now Norsun Årdal. But the world market is dominated by multicrystalline silicon.
Most of the new production coming from this technology, both in Norway and in the rest of the world. Some Chinese companies had begun to show themselves on the major exhibitions. I once asked a key person in the solar industry about what they thought about these, and the answer was that they came with, they knew what they were doing, but they were not worried about them yet.
key to Golden Age
multicrystalline silicon is basically a simpler technology than monocrystalline silicon. Compared with other materials, such as steel or aluminum for construction purposes, it has very few defects – grain boundaries and dislocations, which are absolutely necessary to get the properties needed to harvest electricity from light.
But here one does not attempt to achieve a perfectly material. The silicon is melted in a square støpedigel so, lower the temperature of the bottom of the crucible below the melting point. Sooner or later forcing one or more crystals forward where it suits them.
The essay continues below.
If you have used a hand warmer traveling some once, you’ve seen the phenomenon. Hand heaters are liquid inside a plastic case, with a small metal gizmo inside the liquid. If you break the metal thingy, solidifies suddenly liquid and turns into a hard lump – crystals – while head is hot. The phenomenon called supercooling. The temperature of the hand warmer is below the melting point, but since there are no sharp edges crystal growth can start, the liquid will remain liquid until abrupt changes of this metal spline, or until the temperature becomes so low that it is not possible to keep it in melt anymore. Then spurts crystals front and the liquid is solid! For multicrystalline silicon occurs when the temperature in the bottom of the crucible has been low enough – without having any control on exactly when this happens. After a while calms growth down, and you get stable growth of crystals upward in the crucible, the entire weld pool is converted into a lump (or, as it’s called, ingot) which consists of many crystals.
Since one has given up the material perfectly, one can also give up to get it completely clean. And the solar cells from multicrystalline silicon are inferior to those of monocrystalline, in such panels is converted up to 19.5 percent of the energy in light into electricity.
The end of the golden age
2012, everything goes doing well. Norwegian manufacturing jobs desperately to get back to the vision of cost in order to compete with the big Chinese producing multicrystalline silicon in huge quantities, and saturate the market completely. It is almost unbelievable that it is possible to cut production costs in Norway 30 percent almost overnight, but it is not enough, and REC declared bankrupt in Norway. Little brother Norsun are left to carry solar tab in Norway.
Today is still power and coolant cheap in Norway, and we still have top expertise in silicon and making crystals. English Crystals have made production after REC SiTech during business idea that we should do what we have competitive advantage in Norway, the rest must entrust to the rest of the world. No one has ventured on multicrystalline silicon again, after the collapse in 2012.
Expensive is cheap, but cheap is fast
The market is controlled by how much power you get for your money invested, and the simpler multi-process gives so much cheaper cells that compensate for that they can not extract as much power. Yet there are many who predict that monocrystalline silicon is going to take the lead and become more and more dominant in the coming years. The reason is, ironically, these being the cheapest option.
In order to retrieve solar to your house, it is not just the actual solar cell you need: you need a roof, wiring and other electronics, called inverter to create AC power, and glass and aluminum to protect and maintain the solar cells in place. All this costs money.
And since the solar cell has been so wonderfully cheap in recent years, it is this which constitutes the main cost of the entire system. When it pays to have solar cells that derives much energy per area, and prioritize cells that cost more, but perhaps provide 20 percent more energy, to save these expenses.This is a difficult calculation that has many variables, and it is not good to say how the solution to the math will change as the industry matures.
However, there is another very important factor that must be mentioned, which perhaps explains why it is still multicrystalline silicon which is the fastest growing technology and which perhaps will be more and more important in the years and decades to come. Production Takt. A standard furnace for multicrystalline silicon produces about as much in a week as of monocrystalline in a month, and multicrystalline ingots weighing 800 kg against 150 for a CZ ingot. Productivity for mc can relatively easily be increased by increasing the surface area of ingots. Recently it was shown off an ingot of 2 tons produced by a company in Taiwan.
Heaters for REC standing fallow at Porsgrunn, has well, although they were built for completely different sizes, the potential to nearly double yet again. For comparison, there is a physical limit to how big CZ crystals can be created before they simply become too heavy and fall down. Why is it possible that this factor becomes more important in the years ahead? The reason is that if solar cells will become an important energy source globally and become the tool we hope the work to limit climate change, energy production increased by a factor of between 100 and 1000 and 2050. Then you have to simply roll out massive amounts photovoltaics in the years ahead, and this requires a technology that can withstand a brutal upscaling.
Absolutely perfect or great production?
So what philosophy should choose: to make the product as perfect as possible, avoid all defects and get high efficiency at the lowest price, or learn to live with the problems and limitations imposed by a technology that enables us to produce fast?
In Norway, with all our expertise in silicon both industrial and research, it is probably wise to do both.
Published:
No comments:
Post a Comment