[labnetwork] ALD TEMAHf precursor and pump oil loss

[Elena Cianci]

Dear Mac,
thank you for your suggestions.

The pump has been installed on last July and the chamber pressure with 
20 sccm N2 flow is 0.25Torr.

We do not have any trap on the pump, I will look for the oil mist trap 
you mention and i will check the exhaust. Do oxide powders in the oil 
change the oil appearance?

HfO2 deposition temperature is 175°C, and I tried longer purge time to 
improve material properties, but it didn't work. I will check the film 
chemical composition by ToFSIMS in the next days.

Elena


Il 2016-11-21 18:14 Mac Hathaway ha scritto:
> Hi Elena,
> 
>  We have the same tools and process here (Harvard CNS), and your
> pump/oil should be fine.
> 
>  What is your base pressure (with 20 sccm of N2 flowing)? It should be
> between 200mT and, say, 500mT. If it is 600mT or higher, it could well
> be your pump is dying, and in need of a rebuild. If base pressure is
> suddenly higher than previously, that is particularly suspicious.
> 
>  The pump oil is probably leaking from a seal, which can be fixed with
> a rebuild. If the pump runs low on oil, that will effect it's pumping
> ability. Assuming your exhaust is vented (and well sealed), it is less
> likely that the oil is coming out of the exhaust, (but possible). See
> if you can tell if the exhaust side of the pump has a lot of oil in
> it.
> 
>  At CNS, we used to go through pumps at a rather alarming clip (less
> than 6 months between rebuilds), due to the creation of oxide powders
> in the pump oil (excellent grinding slurry). After we started putting
> oil mist traps on our pumps (as an in-line reaction surface), our
> pumps starting lasting 2 years or more between rebuilds.
> 
>  What temp are you depositing your HfO2 at? Below 240C should give
> more consistent results, as above that you start getting close to the
> decomposition temperature of the TDMAHf. Also, if your pump rate is
> dropping (as evidenced by higher base pressure), you might find your
> wait times between pulses is no longer long enough to keep out of CVD
> mode (just a bit), which can also effect your refractive index.
> 
>  Mac Hathaway
> 
> 	*
> 	*
> 	*
> 
> Mac Hathaway
> 
> Senior Process and Systems Engineer
> 
> Harvard Center for Nanoscale Systems
> 
> 11 Oxford St.
> 
> Cambridge, MA 02138
> 
> 617-495-9012
> 
>  On 11/21/2016 11:23 AM, Elena Cianci wrote:
> 
>> Good day all,
>> 
>> We have a Cambridge Nanotech Savannah200, and few days ago we
>> mounted TEMAHf cylinder.
>> 
>> The system is pumped by a rotary vane pump Pascal2021 with Fomblin
>> oil. After few processes using TEMAHf and ozone, a large oil loss
>> has been observed, with oil present in the vessel under the pump.
>> 
>> Furthermore, HfO2 films resulted with lower refractive index than
>> expected.
>> 
>> Are there incompatibility issues of precursors like TEMAHf with the
>> pump in use?
>> 
>> or are there other issues that we should take into account to
>> prevent such problem?
>> 
>> We really approeciate any suggestions you might have
>> 
>> Elena Cianci
>> 
>> -------------------------------------- Elena Cianci, PhD
>> Research associate
>> CNR-IMM, Unità di Agrate Brianza
>> via C. Olivetti 2
>> 20864 Agrate Brianza (MB), Italy
>> 
>> _______________________________________________
>> labnetwork mailing list
>> labnetwork at mtl.mit.edu
>> https://www-mtl.mit.edu/mailman/listinfo.cgi/labnetwork [1]
> 
> 
> Links:
> ------
> [1] https://www-mtl.mit.edu/mailman/listinfo.cgi/labnetwork