[labnetwork] Use of forming gas in cryo-pumped sputtering system

[Brent Gila]

Hello Aaron,

I my previous research, I did a lot of oxide MBE growth and we used a H2 
plasma to clean the substrate surface (GaAs, GaN,...).  The cryopump on 
the MBE system worked well for these gases.  The regen was not a big 
concern for us, H2 comes off of the cold head at 20K while O2 comes off 
at 90K.  We regened the cryo with a dry backing pump (not the standard 
flowing N2 method), so the cryopump was always under vacuum and thus we 
never had a critical build up of H2.  But I think you will be fine even 
if you use the flowing N2 regen method for the cryo (like most sputter 
tools), the dilution of the H2 will occur pretty quickly and the pump 
body will have a LOT of N2 inside.

Best Regards,
Brent

-- 
Brent P. Gila, PhD.
Director, Nanoscale Research Facility
1041 Center Drive
University of Florida
Gainesville, Florida 32611
Tel:352-273-2245
Fax:352-846-2877
email:bgila at ufl.edu




On 3/15/2019 12:12 PM, Aaron Hryciw wrote:
> Dear colleagues,
>
> Our open-access facility has a cryo-pumped, manual sputtering system 
> plumbed with argon, oxygen, and nitrogen, which is often used for 
> reactive sputtering (mostly oxides). One of our users has requested 
> the addition of forming gas (5% H₂ + balance Ar) to enable sputtering 
> of /a/-Si:H / SiO₂ multilayers, as described in this paper 
> <https://urldefense.proofpoint.com/v2/url?u=https-3A__www.osapublishing.org_ao_abstract.cfm-3Furi-3Dao-2D43-2D17-2D3548&d=DwMFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=gl_2fLZA_-_JfH_dOmx7ug&m=RQnKj8jWOJDys3S59ErGq_i0RC9WpIgTRWwy5pLQGlw&s=Y612MfcUhUpXHR-N8n8jvwfaICCA0O05B7NKvZGZvws&e=>.  
> A concern was raised, however, about the possibility of explosions in 
> the cryo during regeneration, as the frozen mixture of argon, oxygen, 
> ozone, and hydrogen is released as the pump is warmed up.
>
> Does anyone have experience with using forming gas in a cryo-pumped 
> system, and/or know of any measures that could be taken to ensure 
> these sputtering processes could be performed safely?  Doing this on a 
> system equipped with a turbo pump would be better, but we are leery of 
> putting a turbo on a manual system, where users fairly routinely crash 
> the cryo. As always, any advice would be greatly appreciated.
>
> Many thanks.
>
> Cheers,
>
>  – Aaron
>
> __
>
> Aaron Hryciw, PhD, PEng
>
> Fabrication Group Manager
>
> University of Alberta - nanoFAB
>
> W1-060 ECERF Building
>
> 9107 - 116 Street
>
> Edmonton, Alberta
>
> Canada T6G 2V4 Ph: 780-940-7938
>
> www.nanofab.ualberta.ca 
> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.nanofab.ualberta.ca_&d=DwMFaQ&c=sJ6xIWYx-zLMB3EPkvcnVg&r=gl_2fLZA_-_JfH_dOmx7ug&m=RQnKj8jWOJDys3S59ErGq_i0RC9WpIgTRWwy5pLQGlw&s=VHILdiW5UaOet2y31abkYDjTdoYOnWdfzcrFVPkiz5E&e=>
>
>
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://urldefense.proofpoint.com/v2/url?u=https-3A__mtl.mit.edu_mailman_listinfo.cgi_labnetwork&d=DwICAg&c=sJ6xIWYx-zLMB3EPkvcnVg&r=gl_2fLZA_-_JfH_dOmx7ug&m=RQnKj8jWOJDys3S59ErGq_i0RC9WpIgTRWwy5pLQGlw&s=VMscYXW1vfKJ2lHCb2UcCanpdUQks5AWKhquXQyNT-o&e=

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20190315/a9384ae0/attachment.html>