Negative Bias Temperature Instability: Still an issue?

JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

What is the most important degradation mechanism that limits IC lifetime

Hot carrier degradation

Gate dielectric breakdown

Negative bias temperature instability

radiation induced damage

stress and electro-migration

Clear selection

How important is NBTI as a degradation mechanism

Extremely important

Very important

Somewhat important

Not important for most applications

Not important for any application

Clear selection

For NBTI analysis, what type of measurement data do you use

Standard stress-measure-stress measurement

On-the-fly measurement

Ultrafast measurement

All of the above

None of the above

Clear selection

Regarding NBTI lifetime project, we use

Internal proprietary model

Model developed by our fourndry

Models from the literature

A pass-fail criterion

We rely on vendor certification

Clear selection

NBTI may be enhanced by self-heating of transistors

We already account for self-heating in our design

We account for it through guard-banding

We rely on IC performance measurement to see if self-heating is important

We believe self-heating is unimportant, and therefore we do not account for it.

We are unsure regarding the role of self-heating, and therefore we are exploring options

Clear selection

The main uncertainty in NBTI lifetime projection comes from

Measurement delay and uncertain time exponent

Voltage acceleration factor

Temperature acceleration factor

Statistical distribution of parameters

All of the above

Clear selection

In our design, we account for duty-cycle and frequency-dependent lifetime improvement

Always

Most of the time

Only when DC lifetime is insufficient.

Almost never, the projection models are untrustworthy

Never

Clear selection

Many groups are working on new channel materials (e.g. Ge). Do you feel that NBTI will be

much worse

somewhat worse

comparable to Si transistors

somewhat lower

We do not have sufficient data to decide one way or the other

Clear selection

IC applications have evolved from periodically-off devices such as desktops and cellphones to always on devices such as Google Home and Amazon Echo. For these applications, we need to

Fundamentally rethink NBTI degradation

Tailor existing models appropriately

NBTI is not an important issue for these new systems

If the IC is already qualified for NBTI, no additional qualification is necessary

Clear selection

I learn about NBTI primarily from

Tutorial and talks at various conferences, e.g. IRPS

From reading journal papers, review articles, and books

By talking to my expert colleagues

By doing my own measurement and simulations

All of the above

Clear selection

I feel that today NBTI is a

Fundamentally well-understood phenomena

Well-understood phenomena for practical work

A generally-understood phenomena, but still requires considerable work

A confusing phenomena that depends on who you talk to

Clear selection

Submit

Clear form

This content is neither created nor endorsed by Google. Report Abuse - Terms of Service - Privacy Policy

Forms