I am trying to deposit DC sputtered GeTe thin films. I need to anneal the films to crystallize them. I am using RCA1 cleaned substrates with chamber vacuum > 2e-8 Torr. After deposition, films look perfectly fine under optical microscope, but after annealing, pin holes starts to form starting from the edges of the substrate. I tried it on Alumina Oxide substrate, glass and Silicon substrates. Same issues. Tried lowering the deposition power, annealing in chamber or outside the chamber, but still has same issues. Pin holes size starts to decrease in the middle of the substrate.
Moreover, I tried two different sputtering systems but same problem. If this is the inherited property of the material, I should see the pin holes across the wafer. After annealing, even I can see the peeled off material landed on different areas of the wafer.
Pinholes are due to stress at the interface, i.e. lattice mismatch between GeTe and substrate. If you have substrate heating in your system, try to get single crystal material at higher substrate temperature. May be you need different substrate with better correspondence of GeTe and substrate lattice constants. Substrate treatment before deposition is also important.
I experienced a similar problem with sputtered CdTe films following a post-growth, CdCl2-activated, re-crystallization process (~2 micron thick CdTe films on SnO2-coated glass). We determined that the Ar sputter pressure was important, and can change film stress from tensile to compressive. I have mentioned this to several people since, and most have found that sputter pressure is indeed a more critical parameter for stress in many sputtered films than often assumed. See J. Vac. Sci. Technol. 12(4) pp. 1608-1613 (1994).
Thank you V. L. Gurtovoi and Timothy, I understand heating might mitigate the problem but for my process, I need amorphous GeTe films. Going beyond 150 deg C substrate temperature would lead to crystallization. I'll try to deposit at more elevated temperatures and will see if the problem is sorted out, I might need to change my process.
Please see attached image showing holes in GeTe (top) and Tungsten (bottom). If I deposit tungsten under GeTe and anneal the GeTe at around 100 deg C, tungsten layer starts peeling off. Larger pin-holes in tungsten as well.
I'll try some samples at different Ar pressure. I tried 50%, 70% and 100% pressure before, I was thinking more or less that only affect thickness. I'll inspect those samples or may be deposit again.
The artifacts in the image you are calling "pinholes" appear to me more of what I would call "blisters" that may have formed during the post-depositon treatment. Blisters form due to accommodation of excessive compressive stress in the film. Sometime later (maybe during post-process handling or cleaning?), the top of the blister have fallen off or mechanically sheared away - leaving holes through the the GeTe film. This "blistering process" is different from the typical formation process for what are generally called "pinholes" - which more typically result from accommodation of tensile stress (leading to micro-tears in the film). You may want to look at the holes you have with a oblique-angle SEM to get a better idea of what you are seeing. If the SEM shows uplifting near the edge of the hole feature, you most likely have blisters.
Also, I am not sure what you mean by 50, 70, 100% pressure. Is this Ar gas mixed with another gas? The pressure change that you would want to change is in absolute pressure, such as increasing sputtering pressure from 3 mTorr, to 5 mTorr, to 10 mTorr (as measured with a calibrated capacitive manometer)
Thank you so much Timothy! I agree with you that I should call these blisters as material pops open rather than creating holes. Sorry for the confusion. I also agree that I have noticed material on top of blister fallen off and landed on other parts of the substrate. About the percentage, I assumed you mentioned gas flow in sccm (that's why I mentioned Ar percentage). So far, I have tried 2 mTorr, 3 mTorr (I commonly used) and 5 mTorr. Degree of blisters do vary with pressure change, so as the thickness. I have also tried to get similar thickness at different pressure recently, still blisters are there, around the edges of wafer, and the size of blisters vary though.
I never see this blistering with amorphous films but always in annealed crystalline films. Moreover, as I have mentioned, Silicon substrate, glass and alumina oxide substrates do show this blistering. I use two different sputtering systems. Blistering is worse in one sputtering chamber.
The only thing, I haven't tried is elevating the temperature during deposition of GeTe films. I'll give it a shot in one or two days.
Pinhole formation is the result of the reduction of phase volume associated with crystallization.Try to deposit fully or partly crystallized film in order to avoid or to reduce the change in the phase volume..
For that purpose you have to sputter on a hot substrate. The temperature should bef about +/-25% of your crystallization anneal temperature.
If you know the thermal characteristics of a crystallization (from DTA/DTG), you can try also depositing a partly crystallized (nucleated) film with the following thermal crystal growth.Partial crystallization requires lower deposition temperature.
Hi Tejinder,
It is my experience that many of these adhesion issues are a result of bad intrinsic adhesion of the sputtered material on the substrate combined with thermal or intrinsic stress as a result of the anneal. You're base pressure is fine for sputtering but do you execute a sputter etch of the substrate prior to the deposition? Normally this sputter etch removes residual water and oxygen from the substrate and and a short etch helps film adhesion tremendously. And are your targets cleaned of oxygen by executing a target clean using shutters?
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Thin Films and Nanotechnology