"Solar Cells - Cheaper Than Milk" by Chris Devaney page 4

By far, the major share of the PV solar industry is dominated by single crystal and poly-crystalline silicon panels or modules. That may change imminently if Nanosolar is true to their word. Even before Nanosolar, the writing on the wall was clear, and competing technologies were visibly maturing. The public was also demanding lower cost modules, even to the point of looking to China for relief. In answer to this, the silicon labs were busy developing second and third generation solar technologies. I suspect the pace is much more frantic these days. Out of this effort came the first silicon multi-layered thin-film flexible panels, a second generation product that uses multiple PN junctions and new light management schemes to offset the poor solar absorption characteristics of silicon.

Most promising of all, however, and what may end up being the last vestige of silicon in the solar cell market was some technology pioneered very early on by the Japanese (and then forgotten for nearly two decades) on amorphous silicon thin films, generally referred to as a-Si or a-Si:H2 to denote the use of hydrogen as a necessary passivation agent. But as we’ll see, there is still more development work to be done.

Stepping back for a moment, one of the major flaws of silicon for use in conventional solar cells stems from its relatively poor absorption of solar radiation. It takes about 100 to 300 microns or more of silicon depth to absorb sunlight before the light generates the electronic characters (electrons and holes) that are responsible for the electricity we get from the solar cell device. This is a direct consequence of what physicists and engineers refer to as the band gap of silicon; an inherent feature of all semiconductors that dictates the minimum amount of energy light needs to possess in order to free an electron from the material and thus be available for conduction. Like water, silicon has some peculiar idiosyncrasies. It enjoys (?) what is known as an indirect band gap as opposed to a normally behaved direct band gap. With an indirect band gap things are just a bit more convoluted. With this type of semiconductor, as sunlight penetrates the silicon surface, before it can be absorbed and free an electron for conduction, the electron needs an additional kick in the pants to free itself from the mother atom’s grasp... a phonon (think mechanical vibration on an electron scale). Not only that, but both the phonon and the photon (the packet of sunlight) must arrive simultaneously for absorption to take place. Even though at room temperature phonons are plentiful, the statistical chances of a simultaneous arrival of photon and phonon means that the sunlight has to travel that much deeper into the material before things happen. Absorption length increases. It’s just the way things are... or so they thought.

In developing the amorphous silicon cell, one of the most delightful surprises of all was that the silicon, when in amorphous form where there is no crystallinity at all and silicon bonds are arranged willy-nilly as opposed to neat and orderly, its band gap all of a sudden becomes direct instead of indirect... a physicist’s delight! All those nasty things associated with indirect band gaps disappeared, most noticeably, the poor absorption of light. Silicon scientists were ecstatic at that time.

What a breakthrough! And with that advance, very thin films on the order of 10 to 25 microns thick could be produced with all the inherent blessings of a thin-film process. The issues that remained such as electronic inefficiencies could either be dealt with or balanced against the massive cost reduction this new technology should bring about.

To get to market, however, another nagging issue had to be solved involving the difficulty in forming the necessary electronic junction, the PN junction. The issue stemmed from the bizarre behavior on the part of intentionally added impurities typically added to silicon. The impurity atoms sometimes were and sometimes were not electrically active as they should have been. This was once again in response to the willy-nilly arrangement of atoms in the amorphous structure. Dangling bonds they called them. It was cleverly rectified using hydrogen to passivate any unsatisfied silicon or impurity bonds much the same way a tantrum is avoided with a rubber pacifier. They shoved hydrogen atoms in the structure where bent, broken and chaotic bonds were left to throw electronic tantrums in the device. Things couldn’t be any simpler, couldn’t be any better than this, life was good to the silicon people, but still, this was a new technology where anything could happen.

And then it did. They’d put me in jail if I didn’t mention...

The Staebler-Wronski Affair

It sure didn’t escape anyone’s attention when the first amorphous silicon solar cells were put out in the field. After a very short period, about 30 days or so, efficiency of the cells dropped remarkably. Fifteen, twenty, even as much as fifty percent degradation after sitting in the sun for a month. Sherlock Holmes in chatting with Dr. Watson in A Scandal in Bohemia, summed up the situation perfectly for the silicon scientists when he said, “It could be, Watson, that our plans have been menaced!” Menaced indeed, Sherlock!

First described by Mr. Staebler and Mr. Wronski, these poor individuals are now associated forever, even in name, with an extremely nasty defect. So vile is this effect that you can only whisper “Staebler-Wronski” in the presence of other silicon technologists.

Still to this day the “Staebler-Wronski” effect is not quite understood nor the problem solved, even though amorphous silicon solar panels are in the marketplace today. However, great advances have been made in both these areas. The oddest feature of this mess is that the efficiency comes back to the original specifications (or very close to) after a short high temperature anneal at around 300 degrees centigrade! This led the scientists to delve into the role of hydrogen passivation. Being such a small and active atom, thermal collisions involving the hydrogen atoms could dislodge it temporarily leaving unsaturated silicon bonds (the dangling bonds) that are as bad for a silicon solar cell as poly-unsaturated Crisco oil is to the face of an oily teenager. Still, unless amorphous panels are shipped with a rather large annealing furnace attached, something still needs to be done.

Consumer beware: Reputable manufacturers and sales outlets dealing with amorphous silicon panels will generally state in the specifications whether the claimed output wattage is before or after stabilization. Some don’t. If it’s not clear, ASK!

At present, amorphous silicon can be more or less routinely produced with minimum, but still present, light induced degradation of about 5 to 10 percent loss and it is commercially available in a variety of building-friendly formats like flexible roofing panels, translucent panels and more. Because of this, it may still exist after Nanosolar takes over the world.

A Silicon Swan Song and a Parting Prayer

With this new technology emerging before it’s time, it could be the last gasp for silicon as the dominant player in the solar cell market, only time will tell. I, for one, will miss those blue eyes passively looking upwards toward the sky, faithfully giving me electrons in return for a bath in the sun. Still, we must consider that silicon enjoys the highest overall efficiency of all despite the aforementioned problems, and nowhere in the Nanosolar press releases is there any mention of an overall efficiency. There must be a reason why it is likely buried in the private specification sheet that we aren’t privy to yet. But in a cost versus efficiency struggle, the consumer has waited too long for reasonable pricing and few would buy into a solar roof installation that costs three to four times what Nanosolar can supply it for, even if it takes a few more square feet of roof to get the same power output and even though it has the prettiest blue eyes. Cheaper-than-milk solar panels easily offsets the real estate discrepancy. Throw in the not-too-impossible scene where silicon solar manufacturing ceases, plants close, and silicon modules are not to be found anymore. Where does that leave us pioneers with 20-year warranties on ol’ blue eyes? How safe is our 20-year warranty now?

Still, clinging to the past, one final view to consider is the history-repeats-itself scenario where for some reason Nanosolar fails to deliver on the pitched goods, just like Bruno did, just like Bell Labs did. Say, for instance, there is an after-production glitch akin to the Staebler-Wronski affair or some kind of government interference with price controls to protect the myriad industry fallout that might result. Price doesn’t come down. Investment sours. No more Google money, government grants directed toward solar research gets cut back, who is going to invest in an idea that has repeatedly failed to materialize? It’s a possibility that must be considered, although unlikely given the hope of getting something for nothing like solar promises. We might still see silicon solar cells but most likely not in the blue-eyed form of single crystal or even polycrystalline panels anymore, they are just too darn expensive.

My guess is that this time, the miracle of affordable solar energy is truly upon us. Tomorrow will be much different than today, thanks to Nanosolar. Get ready to gear up for solar energy. It just may be cheaper than milk! But, look out Nanosolar, another third generation technology, organic semiconductor solar cells, just may be the Hell-hound on your tail, breathing fire down your neck and searching for massive capital to threaten you with extinction in a year’s time if you slip up! In the meantime...

May God bless the whippersnapper, God bless ’em, one and all.

Further Reading

Practical Photovoltaics - Elctricity form Solar Cells, Richard J. Komp, AATEC Publications, Ann Arbor, MI, 2002.

The Physics of Solar Cells, Jenny Nelson, Imperial College Press, London, Eng., 2003.

Thin-Film Solar Cells Next Generation Photovoltaics and its Applications, Yoshisiro Hamakawa, (ed.), Springer-Verlag, NY, NY, 2003.

Previous 1 2 3 4 Home

Hit Counter