Fun with setbacks

Be forewarned - this is a boring post about my latest difficulties with CIGS.

The CIGS evaporator is down again, this time for a corroded thermocouple (a temperature measuring device) that's needed to regulate the copper evaporation source. I haven't been permitted to keep a spare heater on hand due to the 13000 euro price tag, which means that a replacement can't be bought until failure occurs. Delivery time for the new source: 3 months.

However, far from putting an early end to my CIGS activities at the Ångström Lab, the missing source gives me a wonderful opportunity to get really creative in making solar cells. The heaters for the three metals, copper, indium and gallium, are namely interchangeable (not selenium because it evaporates at a relatively low temperature). The crystal structure for CIGS requires about 50% selenium, 25% copper and 25% indium and/or gallium. Indium and gallium are, in fact, interchangeable in the crystal. It's thus possible to make CIS (copper indium diselenide) or CGS (copper gallium diselenide) or something in between known as CIGS with varying electronic properties that are still useable for solar cells. Copper, however, is a must and so is selenium. So, I can either ditch the gallium source and use its heater for copper to make CIS, or ditch the indium source and make CGS.

But wait, there are more possibilities. What if I keep both gallium and indium and figure out a new way of getting copper where it's needed? To refresh your memory on how CIGS is made I'll refer you to an earlier post entitled The making of a CIGS solar cell. What if I start with glass and its molybdenum coating and also precoat it with all the copper I think I'll need? Then when indium, gallium and selenium come along, these four materials will all mix together anyway and still come out with the right crystal structure. I think, anyway. Molybdenum doesn't mix in but keeps itself safely separate for its future function as metallic contact.

Problem is, I don't have any good method of precoating with copper. Sputtering would be my first choice - then I could sputter copper on right after sputtering molybdenum onto the glass. But, drat it, I don't have a copper "target" that would be needed. There is, though, a small evaporator that can handle copper. Too bad it can coat only one substrate at a time and at best it could do two in a day. Ideally I would like to have a couple dozen. But I can still have some fun doing some preliminary tests with just a few.

And who knows? Maybe it'll work so well that CIGS solar maufacturers will all end up using this method. A more likely scenario is that it won't work very well. But at least I get to have fun playing with this stuff.

Now if I could just get equally creative with setbacks in my personal life.