Hello,

How is a capacitor normally implemented in CMOS integrated circuits. Is it through done indirectly using CMOS transistors or there is another way to do this?

 

Oxide layer between two layers of metalization if it is a series capacitor, else a single metal layer on oxide to substrate, if it is a shunt capacitor.

 

Depending on the aspect ratio of metal lines, you can also use sidewall capacitance from interleaved metal lines on the same layer.

For supply decoupling, some processes use MIM caps to replace the more traditional device cap (which you referenced) that uses gate capacitance and is more "costly" because it consumes area that could be used for more useful transistors.

 

Physical capacitors are very expensive area-wise on integrated circuits but transistors (both kinds) are relatively cheap. Often a small capacitor is made and attached to a capacitance multiplier.

users.ece.gatech.edu/rincon-mora/publicat/trade_jrnls/pmdl_0706_cx.pdf

 

Thanks a lot ...

 

Hello,

How is a capacitor normally implemented in CMOS integrated circuits. Is it through done indirectly using CMOS transistors or there is another way to do this?

There are many ways to implement capacitors on CMOS ICs. If you want high capacitance and can tolerate high nonlinearities then you make FET caps using the gate capacitance of a FET where the gate is one plate and the drain/source (which are usually tied together) form the other. You can also use interconnect layers to get linear capacitors and there are a number of ways to do that depending on your needs and the area constraints you are working under.

 

Thanks. Probably this is in the domain of Analog engineers. I am just taking baby steps in Analog design ...