Is there any correlation between the grain size of the sputtering targets towards the formation of the thin films?
I came across a research by CA Michaluk (2000), stating that the rate of erosion would be inversely proportional to the size of the grains of the bulk material perpendicular to the incident beam. Grain with favourable orientatition ( e.g. BCC crystal --> (011) > (111) > (001) ) and smaller size would erode faster.
However, I am interested to know, if the grain size of the sputtering target will also effect the size of the thin film's particles (Smaller grain size --> Small particle size?). Is there already reported research on this? I would be grateful if anyone can suggest literature reviews on this phenomena.
I agree that target grain size/distribution/orientation will have an impact of the rate of erosion and consequently the deposition rate, its uniformity across the wafer and stability in time. As far as I know there is no reason to expect a correlation between the target grain size and the grain size in the grown film. Only very extreme and rather unpractical conditions (high power - high rate) may have some correlations. Note that target grain sizes are typically described in um range while as deposited grains in thin films will be in the nm range.
I agree that target grain size/distribution/orientation will have an impact of the rate of erosion and consequently the deposition rate, its uniformity across the wafer and stability in time. As far as I know there is no reason to expect a correlation between the target grain size and the grain size in the grown film. Only very extreme and rather unpractical conditions (high power - high rate) may have some correlations. Note that target grain sizes are typically described in um range while as deposited grains in thin films will be in the nm range.
Sputtering deposition is a non-equilibrium process occurring in plasma. The films deposited under the conditions (total pressure, power supplied to the target, substrate and its temperature) for promoting coalescence will have larger grains. At a very high pressure, lower ion energy due to the frequent collisions results in very low adatom mobility. In the case of lower adatom mobility, the coalescence process is more kinetically limited than in the case of higher adatom mobility. Consequently, low adatom mobility under high sputtering pressure leads to smaller islands at the initial stage and therefore smaller grains later. However, in the moderate range of pressure, where adatoms obtain more energy by ion bombardment, the adatoms can move more actively, and larger grains are formed. So, in other terms the parametres that influence grain size of the deposited film are various and, at the first instance, not are directly lied to the grain size of the target. The rate of erosion of the target can influence the process in terms of quantity of material that arrive at the substrate (rate), but only in an indirect manner it influences the morphology of the growing film.
The grain sizes influence the rate of sputtering of target material. So if this rate increases due to grain size and the the plasma parameters the property of deposited film also changes. Targets with small grains and different plasma parameters can yield the same film quality as that of target of large grains under different plasma condition.
I dont see any reason for any relation between the film quality and the grain size of the target. Explanations of Anna and Adam are mostly correct.
Thank you all for the detailed explanation. I really appreciate it.
Since we are in the discussion on what may effect the size of the particle, I was wondering if the melting temperature of the target may effect the outcome of the particle size.
Could it be that, 2 different sputtering targets with similar grain size and orientation, but has different melting temperature, sputtered under the same sputtering condition, produce thin film with different particle size?
Since the sputtering pressure and power supply is the same, hence the ion bombardment on both target can be assumed to be similar rate. The sputtering yield (number of ejected atoms from target per one bombarded ion) will depend mostly on bonding energy (I think). However, is it possible that target with higher melting temperature would produce smaller island on the substrate in the initial stage, due to the fact that it has less kinetic energy to move and form coalescence?
I am saying that, target with higher melting temperature would produce smaller island, based on my understanding that, higher melting temperature would mean that the atoms will need higher energy to "hop" from one adsorption site to another adsorption site to produce a coalescence.
Is this agreeable? I appreciate your kind comments/ideas.
Mr. Susantyoko,
Thank you for the reference. I could not find the journal that you suggested from my university's database, but managed to obtained "Stress and Grain Growth" (J. Mech Phys. Solids, 1996) instead. It has the same explanation.
However, I did not manage to obtain the same result. Between pure Ni and NiAl alloy (NiAl has slightly lower melting temp), both sputtered under the same condition, the pure Ni thin film has coarser particle as compared to NiAl thin film. However, sicne the sputtering time is more than 30 minutes, I believe the mechanics of stress and grain growth applies?
Dr. Piramanayagam,
I really appreciate your kind reply. After re-reading my previous question, I understand your point. Sharing findings of my experiment (Ni vs. NiAl) does deviate from my actual questions. The data should be accessed and analyzed separately.Thank you for pointing that out.
- Badges
- Science topic
- Similar topics
- Chemistry
- Nanotechnology
- Nanostructures
- Thin Films