Transmission-Gate DC ON-resistance Temperature Coefficient Positive?

[JohnLai]

Dear Folks,

I observed that for a pair of complementary transmission-gates (PMOS//NMOS), the DC ON-resistance increases as temperature rises. It seems contradicting that the threshold voltage of transistors (Vth) becomes smaller as temperature rises such that the overdrive across each of the transistors becomes larger. The larger overdrive should result in larger conducting currents, which is smaller ON-resistance.

Can someone explain why TG DC ON-resistance has positive temperature coefficient?

Thanks.

 

[timof]

A good question.

Vt does becomes smaller at higher temperatures.
However, mobility goes down at higher temperature.
So, depending on the value of Vgs, the current can be higher or lower, at high temperature.
At low Vgs (slightly above Vt), current goes up, and at higher Vgs (nominal Vg) - it goes down, with temperature increase.
There is a Vgs point on Ids-Vgs characteristic that is (almost) temperature insensitive.

Max
---------

 

[JohnLai]

timof,

Thanks for the answer. I need to study a bit why mobility goes down when temperature rises.

 

[timof]

When mobility is dominated by phonon scattering (acoustical phonons at low fields, and optical phonons at higher fields, in silicon), higher temperature leads to higher scattering rate (higher concentration of phonons - Bose distribution), and thus - lower mobility. When mobility is dominated by ionized impurity scattering (very low temperatures) - the trend is opposite (higher temperature --> higher mobility).

For most of the practical purposes (room temperature and higher) - carrier (both electron and hole) mobility in silicon decreases with temperature.

This link might be useful:

https://en.wikipedia.org/wiki/Electron_mobility

 

[Junus2012]

I think the mobility go down also with higher vgs, isnt it???