I prepared several Ni-5.5Al thin films, then checked the magnetic properties using VSM. I expected that with increased sputtering time, the hysteresis loop would change proportionally; Ms will increase with increased sputtering time.
However, the Ms did not follow that trend. The sample sputtered mid-duration has the lowest Ms (see attached).
When analyzed using FE-SEM, the thin film of all samples are porous. The particles are not "joined" as what normally observed. All thin films showed similar trend.
Is there any correlation between the porosity of the thin film with the peculiar hysteresis loop?
First think about your deposition method: sputtering is an out-of-equilibrium technique. Check what factors may have changed the trend: temperature effects, shadowing, etc. The porosity of the samples avoid the filed magnetization on the surface therefore the higher the porosity the smaller the lateral limit to the size of the locally joined particles associated to the porous morphology.
Hi Mr Domingues-Canizares,
Thank you for the feedback.
I sputtered this target, Ni-5.5Al (at%), in the same day as another sputtering target, pure Ni. I sputtered the pure Ni target first, change the target to Ni-5.5Al, then continue sputtering.
On both targets, the substrate temperature, base pressure and working pressure, the Ar gas flow and power supply are similar. However, the pure Ni thin film is not porous, and the Ms increases with increased sputtering time.
Hence I was wondering, could it be that the porosity of the Ni-5.5Al thin film is caused by the composition (Al contents)? Meaning, this phenomena is strictly material related and not sputtering process related?
Regarding your statement - "The porosity of the samples avoid the filed magnetization on the surface therefore the higher the porosity the smaller the lateral limit to the size of the locally joined particles associated to the porous morphology"...
Just to make sure that I understand it correctly, are you saying that, with increase porosity, the size of the grains are reduced, hence the Ms is reduced?
Dear Muhammad Reza
The magnetization in your plot is in emu/cc. i.e. you have reduced the moment (in emu) by the film volume (in cc). In what way do you determine the film volume?
Do you change the target by opening the chamber? This out-of-equilibrium technique is very sensitive to the substrate temperature, as well as off-axis magnetron and pressure (ballistic-diffusive regime). It may depend more on the synthesis technique than the material itself, but one way is to check the film before deposit Ni-Al in order to observe it: is it planar? it has islands at any grade of coalescence? is formed by columns?
If you reduce the grains/particle size then this may reduce the magnetization for the cristallite domains are lower (guess).
Hi Komogortsev, Domingues-Canizares,
1. The substrates are small glass slides, that were cur using diamond cutter from a big-sized glass slides. Each sides were measured and rounded to mm, then converted to nm.
To get the thickness, the cross section of the thin film were analyzed using FE-SEM.
Using all 3 data ; Side (nm) x Side (nm) x Average thickness (nm), the volume was determined.
I prefer to use emu/cc instead of emu/g as there are almost no weight difference between a clean and an coated substrate.
2. Yes, the target was changed by opening the chamber. The standard procedure that I used when sputtering new targets, after changing the targets are as follows;
a) Vacuum chamber
b) Purge with Ni gas
c) Vacuum chamber + start substrate heater
d) Wait for 30 minutes in order to ensure substrate is thoroughly heated. Plus, when the heater is switched on, the vacuum level tends to rise, hence it needs around 30 minutes to stabilize back to preferred level.
e) Pre-sputter target for 5 minutes (substrates not exposed to the plasma)/
e) Start sputter process.
Hence, I would assume that the substrate temperature may not have a significant effect on the thin film formation (?).
I sputtered the NiAl thin film directly on the substrate, as I wanted to observe the thin film growth without adhesion layer. Due to analysis equipment limitation, the minimum thickness that I can produce is 100nm and above. Above that thickness, the controlled sample (pure Ni) is already fully coalescence. This is what makes me believe that the porous nature of the NI-5.5Al thin film is more related to the composition of the material.
I am also thinking to measure the crystallite size from the XRD peaks using the Debye-Scherer method. However, since it is porous, I am not sure if the calculation result is correct?
Dear Muhammad Reza
As I see from the SEM image the film is very porous. Then you don't know the total NiAl volume. You measured the volume of porous media that consist of Ni-Al alloy and the numerous voids. As a result the magnetzation in emu/cc is characteristic of porous media not of the Ni-Al alloy.
I think your can determine the magnetization of NiAl if you take in to account the voids volume (it should be extracted from total volume that you estimate). The voids volume may be estimated for example from SEM images.
Hi Komogortsev.
I see your point. However, I am not familiar with any method to calculate the volume.
I used the "Linear Intercept" method to measure the grain size, but it is not suitable to estimate the voids. Do you have suggestions on any particular methods?
The volume Ni-Al can be determined for example in following way.
1. Weigh the substrate together with the film.
2. Remove NiAl by etching
3. Weigh remainder of the substrate.
4. Difference in mass is the mass of the film.
5. Dividing the resulting mass by density of NiAl obtain the volume of the alloy.
- Badges
- Science topic
- Similar topics
- Chemistry
- Nanotechnology
- Nanostructures
- Thin Films