Hi Andy,

How about platinum?  Pt resistance temperature detectors (RTDs) are already
used extensively (just Google "Pt RTD"), and I don't see any reason why you
couldn't scale one down to fit on a chip.  I had an idea for one of these
about a year ago, but never implemented it - email me directly if you'd like
to discuss it further.

Best regards,

Brian C. Stahl
Graduate Student Researcher
UCSB Materials Research Laboratory
brian.stahl@gmail.com / bstahl@mrl.ucsb.edu
Cell: (805) 748-5839
Office: MRL 3117A


On Thu, Dec 16, 2010 at 5:13 AM, Andrew Irvine  wrote:

> Hi all.
>
> I'd like to design a chip with a very crude resistance thermometer on it
> (as well as lots of things that do wildly exciting science).  Using optical
> lithography and thermal evaporation, and I'd connect by overlaying Cr/Au
> bond pads.
>
> It occurs to me that someone on the list must have done it before.
>
> Any tips on metal system, or technical pitfalls that I should know about?
>
> Nichrome makes a good resistor, but has poor temperature co-efficient,
> whereas Au, Al, Cu, Ag etc are horribly good conductors, but vary nicely.
>  I'd rather keep away from toxic or otherwise difficult metals (W, Zn, Ni,
> Fe etc) as I'm giving it to students as a project.
>
> It can be pretty crude accuracy-wise.  I'd be delighted with +/-5K at
> nitrogen temps, and it's a simple two-terminal resistance measurement such
> that the resistance is dominated by the thermometer.
>
> I'm not interested in separate sensors, just on-chip resistance.
>
> All tips gratefully received!
>
> Andy