very low leakage currents

[kars]

Hi all,
for work (yes yes, i know) ive been diving into the wonderfully frustrating world of femto/pico amperes.
For the past half year i've been trying to find the source of a leakage current that is both temperature dependent and different from device to device which makes pin pointing the source a pain.

i am fairly certain the issue lies in our sensor bit i'm trying to read out. Unfortunately this makes finding information about potential issues rather difficult as most publications focus on the circuit board.
What i think i can share: my main area of focus currently is a metal->ceramic->metal structure we use as an insulator.

Does anybody have any experience in this area and/or can point me to more suitable literature?
Are there any obscure effects that i need to take into account?

Kind regards,
Kars

[Dr. Frank]

Hello Kars,
that sounds similar to the technology of ceramic capacitors, MLCC, where you have the same layer structure.
leakage currents are known there, as well, depending on the thickness, composition and quality / purity of the ceramic.
The ceramic firing and annealing process as well determines the isolation.

I propose that you search for technical papers of the known good ceramic capacitor manufacturers.

Frank

[jwet]

I worked on Ion Mobility Spectrometers for Bomb and Dope detection many years ago- 90's  A radioactive isotope of Nickel bombarded the test sample and then would enter an electrode structure.  This consisted of alternating sets of electrodes at progressively higher voltages up to a few kV that were used to accelerate the ions.  There were some papers on this at the time- IEEE Instrumentation or Nuclear Methods.  We settled on platinum plated electrodes and sapphire balls assembled by very patient assemblers.  This is what is still used as far as I know. These techniques were published at the time- funding was DARPA, DEA or FAA, companies were SAIC, Oakridge, Berkeley, U of Tennessee and UW Pullman, IIRC.  Spanned late 80's through mid 90's.