Mike,
What about gold, is anybody looking at that, isn't that the obvious next step after Cu?
electrical conductivity: silver > copper > gold > aluminum > magnesium > zinc > iron > tin > chromium > lead.
Copper is "it". Gold is not as good a conductor and, since it is highly desired by people (mostly women?) for some reason, ;-), its price has been driven sky high since biblical times, or earlier. Silver, as Patrick points out, has serious migration problems. In layman terms, it goes everywhere and shorts out everything. Your chips would end up useful for not much more than as a fuse. Copper, I understand, is tricky and can contaminate in processes where the process people don't know what they are doing yet. It is, according to IBM (and MOT?) handle-able if you do it right. For reference, Intel is going to do copper along with their 0.13 micron process, which should be in a year and a half or so. They are just getting into 0.18 right now, and their upcoming, this month-next month Coppermine chips will be in 0.18 (but not in copper, they selected the code name quite a while ago and didn't think about the name confusion, apparently.).
I do like Wilf's philosophy of not trying to be first in a lot of the technology pushes, like copper and 300 mm. Let someone else take the risk first. LSI's value is in their ASIC, SOAC, coreware, etc. innovations.
Tony |
