Discussion Forum

Hi all,

Currently I am trying to simulate a measurement that I will perform on a assembly of two test structures to simulate the coupling between sensor/read-out pixels (for CERN ATLAS and CLIC next pixel detectors R&D).

The test structure is composed by two chips: one chip with two terminals connected to two pads, and a second chip with two terminals (matching the footrpint of the first chip) connected with each other.
A cross section of the test structure is attached. The vertical scale is amplified by a factor of ~200. This picture corresponds to something like 2mm wide and 10 um.

[Fig_1]

The problem: The idea is to measure C_glue (should be about 4pF) by measuring the total capacitance of the system. C_glue_2 and C_ox are small. Currently the silicon substrate is about 0.1 Ohm*cm (ultra low resistivity). With that, the bottom silicon substract acts like a "wire", making possible to the pads to couple to it, as you can see in the attached picture. As the SiO2 layer is thin (500nm) the C_sub is high (about 80pF). Indeed, it was measured something about 44 pF for the system

Another group has simulated that if they increase the resistivity of the substrate (making it with quartz with 10E18 Ohm*cm, for example) the total capacitance of the system will decrease to about C_glue, as the higher resistivity material will act like an insulator, making the system looks like the next picture.

[Fig_2]

I am able to simulate the structure in COMSOL and extract the capacitance using both Electrostatic and Electric Currents modules. Also, I was able to simulate the two extremes cases (conductive and insulator substrate) by changing the boundary condition of the substrate, applying the Floating Potential for its boundary (to simulate a metal).

My question: I would like to know if it is possible to apply some boundary condition on the substrate to simulate the "circuit being opened" when the resistivity of the material goes to +infinite. I already tried to change the resistivity of the boundary (having it with the potential floating) but it doesn't "open the circuit" as you still have a "metalic boundary" allowing the pads to couple with it.
To simulate the capacitance the Electrostatic module usually is enough. I wounder if in this case I should use the Electric currents (maybe coupled to Electric circuit module?)...

Any help will be much appreciated.
Thanks!
Cheers!