Maximum Vgs voltage?

[AMSA84]

Hi guys,

I am working with UMC 130nm technology and I want to know if there is someone here that has worked with this technology.

Taking into account that this technology works with 1.2V with high speed mosfets and 3.3V for high gain; focusing on the high speed, I would like to know if I can use and higher voltage to drive the mosfet (switch) like 2V or so? Is this possible? Even knowing that the operating voltage on this technology is 1.2V?

Kind regards.

 

[rfsystem]

In most technologies the NMOS threshold shift over the lifetime at bias Vds=Vmax and Vgs=0.75-1.00*Vmax set the maximum voltage. Typical lifetime of the gate is in the range of hours around 2*Vmax. The simple Vmax=1.2V, as in this tech, simplfies design of margins. If you want to exceed this simple rules you need realiability models and have to check every circuit conditions for lifetime.

 

[AMSA84]

Hi rfsystem and thank you so much for the reply.

Can you show me, please, the reference where I can read about what you have mentioned?

What you mean is that the typical lifetime of the gates are made with a margin of 2*Vmax? I didn't understood what you means with "the range of hours around 2*Vmax. Are giving the nº of horus in Volts? There is a mistake there or maybe I am not understanding quite well.

Thanks in advance.

 

[jkatic]

You are limited by the oxide breakdown, which usually appears when Eox=Vox/tox reaches 1 V/nm. Here Eox is electrical field across the oxide, Vox is voltage (higher of Vgs and Vdg) and tox is oxide thickness in your tech (somewhere between 2-3nm I guess). Therefore, for Eox<1 V/nm you are usually safe.