Dear Edward,

Thanks for your help. I have deposited Ti followed by Au (sputtering
system) on the same substrate and did not have any problem. It seems
that the poor adhesion does not due to contamination from cleaning
process. I will check whether that is because of system contamination. I
am just wondering could that be posssible that Cr or Pt does not have
good adhesion to my substrates which are boron compounds and do not have
excellent crystalline quality? Thank you very much.

Yinyan

Edward Sebesta wrote:
> Your substrate can have a layer of contamination that isn't visible, but
> will prevent good adhesion and you can have rough metal.
>
> This contamination can be on the substrate prior to placing it in your
> deposition equipment or it can be from the deposition equipment itself.
>
> 1. Cleans: Oxidizing acid cleans are good. HF dips can be good. Resist
> stripper cleans should be followed by D.I. water and rinsed very
> thoroughly.
>
> I am always concerned about cleans that are dips in alcohols, acetone,
> alkanes, and aromatics. The solvent needs to be electronic grade and
> fresh so you don't leave a layer of organics from the solvent. The
> solvent itself can leave a molecular layer of itself. Having your
> substrate heated in vacuum can be helpful to remove organics.
>
> Descums in barrel ashers that are clean can be helpful in not only
> removing organics but providing bonding sites on the surface.
>
> 2. Pump Systems: Mechanical and Diffusion pumps can back stream
> organics. This will coat your sample and result in poor adhesion. Dry
> pumps are used just for that reason. Cryo pumps are great.
>
> 3. Chamber Contamination: The walls of the chamber can be a resevoir for
> contamination. You might want to have the chamber heated up under pump
> down and baked out for an hour before loading your samples. Make sure
> your carriers are clean and baked out as well. It might be a good idea
> to have a leak up rate done weekly on the chamber as well.
>
> 4. Wrong Lubrication: Sometimes a grease can be used that isn't a vacuum
> grease. For a vacuum system it would be preferable to have no grease.
>
> 5. Wrong O-rings: The O-ring may have been handy when someone did a
> repair but it is outgassing like crazy.
>
> 6. Substrate: Is it really a solid substrate or is it something that
> isn't 100% solid, or can have outdiffusion. Some ceramic parts look
> solid but they have some small percentage void space. Plastics and other
> seemingly solid materials may outgas. The solution would be to have a
> sit time for 10 to 30 minutes to allow outgass. Don't start deposition
> just because you reach base pressure. Your pumping speed could outpace
> the outgassing of your substrate but you still have a need to degass
> your part.
>
> 7. Target cleanliness: Was it cleaned before installation and baked out?
>
> Edward Sebesta
> Independent Semiconductor and MEMS Engineer.