I am working on thin film deposition currently, just wondering if there is a software to simulate the sputtering process. For example, if I specify the process parameters the software should simulate and tell me what will be the deposition rate, composition etc..
Well, you could try to write an FEM simulation or something like that but I would be surprised if the results were quantitatively apllicable. I mean, let's just start with the fact that the sputtering target gets deformed in the process so the discharge zone will probably start moving during the deposition process.
Apart from that, at least in our setup parts of the sputtered material start covering parts of the focussing area (or are you depositing neutral particles?) so the focussing voltages have to be adapted, too.
Start by TRIM simulations: http://www.srim.org/
At least you can get an idea of how ions interact with the solids involved.
Diederik Depla's group in Ghent have developed software (RSD2013) for this, please refer to their page:
http://draftugentbe.webhosting.be/index.php?p=137
In addition to previous answers, the software NASCAM is also a nice and user friendly tool if you are interested on the growth of the film.
https://www.unamur.be/sciences/physique/larn/logiciels/nascam
Emile Haye Just tried the software. I am sputtering NiTi films, is there a possibility to choose the species as NiTi. I could not see any alloys in the dropdown for species
As far as I know, you can try to simulate plasma processes but not sputtering process.
Interesting discussion.
The first course of action for simulating the magnetron (or any kind of) sputtering, would be to simulate the plasma physics.
I found this paper very relevant to sputtered metal deposition dynamic simulation
Article Molecular dynamics simulation of gold cluster growth during ...
The deposition rate depends intimately on your system: type of magnetron source (cooling and magnetic field strength and geometry), thickness and usage of the target, working distance, gas pressure and power and some times base pressure or pumps/pumping speed if the target is sensitive to oxidation for instance.
It is by far much easier to measure the deposition rate, either directly or indirectly, for instance it in situ with a QCM or by running calibrations.
Regarding composition for alloy targets, this could be probably done using sputtering yields for different materials, but you can also work it out using practical knowledge on the difference between the composition of your alloy and that you will get in your coating. Here again target usage, power and working distance will play an important role in a similar way as for reactive deposition.
Overall I would agree with Muhammad and Xiao, it is almost impossible to simulate on a quantitative level, and I only know of softwares that simulate the magnetism and plasma part of the process.
Even though I am not insider in modeling as many other members of this discussion, the question looks very relevant in my opinion.
In my opinion, even though the magnetron sputtering is widely used, the most common approach in general sense is experimental experience. As far as I know there is (at least within acceptable simplicity) no general theoretical approach how to determine e.g. even common parameters of the deposition process like deposition rate or composition deviation in case of multielement targets.
Pure bottom up (atomistic/microscopic) approaches arising from particle interactions will probably be very source (computer performance & time, input data (parameters describing particle interactions)) requiring and thus practically unapplicable. On the other side, the macroscopic (overall) approach (e.g. thermodynamics, chemical kinetics, etc.) will finally be extremely data requiring (at least partially due to non-equlibrium nature of the process) as well.
In my opinion, the general problem is complexity of the process which is connected with large number of involved variables depending on process parameters, while moreover many of these variables are not constant, but are spatially or timely dependent. Sometimes, this can be cleared and solved with acceptable accuracy for some cases (e.g. classes of materials), but as far as I know general model was not developed yet.
With respect to the topic, recently I have dealt with possibility of deposition parameters (applied powers) determination with respect to the required composition of co-sputtered thin films. For (nearly) ideal case (independent sources) approach is very simple (material balance in fact) and solution feasible. Some remarks to this topic can be found there (both the same article, just different source) (https://www.mdpi.com/2079-4991/12/11/1830) or (https://www.researchgate.net/publication/360942812_Tailoring_of_Multisource_Deposition_Conditions_towards_Required_Chemical_Composition_of_Thin_Films?_sg%5B0%5D=0GN5yLBDZjL-YUcQxzmvwcHW5Yx3eNUjX7DFuWXnrqNISwkRmmJU3m_9Is0eshbbjpr8BwfVfV-a6DzkI5FmP6LjOgxAw8Q8VVD13G-6.kZBVq9xFHrsts3LAOMKP6VSt4SUWfkBLL5XQZKD2LPbJari5uQkzB34aZlT_V5A1y0cnafnm6G9TOC6zVbQO1g).
As I mentioned above, again, this is just one shred of mosaic far far away from any generalized model not to say (general) fully theoretical (experimental data free) model.
It just to be remarked that in more general case (i.e. case of interacting sources), the solution is more challenging that already used simple material balance. Unfortunately, I have lack of data to confirm or deny my ideas.
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Cell Biology
- Culture Cells
- Cell Line