[labnetwork] Electrical properties of PECVD oxide deposited in Oxford System100

[Andrei Alamariu]

Hello,

Yes, this is always expected from PECVD oxides.
The Forming Gas 7.5% H2 in N2 Sintering is required for all thermal and deposited
oxides after metalisation. For Al you can go as high to 550C.
There are 2 issues regarding these films.
First you have to do a high temperature anneal ~>800C ( or as high as Your device/ process allows; we did for SiGe a 4 hour anneal at 600C)  before the metal processing.
This is to allow the excess Oxygen and Silicon molecules to react and to force the H2
and other gases to exdiffuse.
This will dramatically improve the oxide dielectric breakdown voltage value.
Second this PECVD process inserts  a lot of metal contaminants into the deposited thin film, mostly by sputtering of the industrial grade Al or SS gas shower / unless you use
an expensive up to date Remote Plasma PECVD machine.
You can use the MOS capacitor with HF CV plot to accurately measure the mobile
Ions concentration in the oxide by BTS technique. It is very sensitive. Also you can
measure the deep level minority carriers killer metals by deep depletion CV lifetime Zerbst test.
I would like to see those CV plots, please.
Thanks, Bernard
aandreib at gmail.com

Sent from my iPad

> On Nov 19, 2020, at 10:15 AM, Lino Eugene <lino.eugene at uwaterloo.ca> wrote:
> 
> Dear colleagues,
>  
> I have done some CV measurements on a MOS capacitor with 22 nm PECVD oxide deposited in a Oxford System100 system for the first time. I noticed that:
> ·        The as-deposited layer has bad CV characteristics with large flatband voltage, hysteresis and shoulder at low frequency, which indicates the presence of interface traps.
> ·        A post-Aluminum metallization annealing in forming gas substantially improves the CV characteristics with a lower capacitance in accumulation, low flat band voltage, inversion at low frequency and no hysteresis.
>  
> I am wondering if this is expected for PECVD oxide and if forming gas annealing is always required to improve the characteristics. Has one of you have done these measurements for PECVD oxide deposited in similar systems and could share data for comparison?
>  
> I can provide the CV curves if required.
>  
> Best,
>  
> Lino Eugene, P.Eng., Ph.D.,
> Micro/nanofabrication process engineer
> Quantum-Nano Fabrication and Characterization Facility
> QNC 1611
> University of Waterloo
> 200 University Avenue West
> Waterloo, ON, Canada  
> N2L 3G1
>  
> Ph: +1 519-888-4567 #37788
> Cell: +1 226-929-1685
> Website: https://fab.qnc.uwaterloo.ca/
>  
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20201125/c475da14/attachment.html>