Why n+ and p+ are used by welltap？

[wizard]

Why substrates are not directly connected to VDD/GND? N+ is between n- and VDD. P+ is between p- and GND? Thanks.

https://en.wikipedia.org/wiki/File:Cmos_impurity_profile.PNG

 

[erikl]

The p+ bulk p-substrate connection is always connected directly to GND (or the most negative potential). The n+ bulk connection of an n-well, in the majority of cases is directly connected to VDD (or the most positive potential), but not necessarily so: in order to avoid the substrate effect, it can be connected to a lower potential, e.g. connected directly to the source of a cascode-connected pmos transistor (which needs its own n-well, in such case).

 

[wizard]

Sorry, maybe I hadn't made myself understood. I mean why p- substate is not connected to VSS directly, but a p+ implant has to be between them?

The p+ bulk p-substrate connection is always connected directly to GND (or the most negative potential). The n+ bulk connection of an n-well, in the majority of cases is directly connected to VDD (or the most positive potential), but not necessarily so: in order to avoid the substrate effect, it can be connected to a lower potential, e.g. connected directly to the source of a cascode-connected pmos transistor (which needs its own n-well, in such case).

 

[frankaurora]

the reason is to reduce contact resistor.

 

[baenisch]

Directly connecting metal to a weakly doped n or p area usually ends in
a schottky contact and not in an ohmic contact. Therefore a strongly
doped contact area is formed to reduce this effect.

Best Regards

Andi

 

[sat]

All the answers are perfect.

Adding my 2 cents -
The tap points really need to have highest doping density at the surface level - in addition to P+ Implant, there must be PDIFF as well.

Due to finite velocity [finite ion energy range 50 - 500 KeV] of Ions, in the implant procedure, the dopants does penetrate inside the silicon substrate/surface they do not reside right on the surface, even after RTA [ thermal annealing] this does not give optimal density at the surface.

To have desired density, they usually always do a P+ diffusion on the top of the P+ implantation to P-substrate regions, wherever a PTAP is needed.

The case is similar for NTAP [NWELL TAP] as well, they have NIMP + NDIFF over NWELL.