Well, since normal NiAl is nonmagnetic, it seems that your films either built magnetic clusters or anything else is included in them.

What happens if you change the sputter rate to allow for more / less diffusion on the substrate surface ? Normally the two-atom structure of NiAl should inhibit the atoms' mobility, allowing only for small grains to form, but maybe there is any other effect influencing the layer structure. Oxidation normally shouldn't occur, either - but since you see a magnetic effect, we should also think about unexpected effects.

Or can you start with much thinner layers? 2 microns are already quite a thick layer.

Can you check for large clusters / any additional material by SEM with EDX or even by TEM?

And quite a different idea - is there any chance that the substrate itself has any magnetic contaminants?

Change in stoichiometry of the deposited film which might be having a higher nickel (magnetic) content than 80%. Formation of nanometer sized nickel oxide particles during film deposition may be the cause.

Hi Erhmann and Syed Zameer,

Thank you for the response.

I sputtered the thin film under 3 different sputtering time : 30 mins, 1 hour and 1 hour 30 mins, in order to get a thick enough thin film for VSM, XRD and FE-SEM. The sputtering time is long, as I am comparing it with pure Ni, in order to describe the effect of Al contents. At shorter sputtering time, NI thin film did not form.

The glass substrates were ultra-sonic cleaned with acetone - methanol - acetone - D.I. water, then dried before being out inside the chamber.

All layers (Ni20Al) showed a clear hysteresis loop, However ;

a) The Hc increased with increased thickness (sputtering time)

b) The Ms (emu/cc) reduced with increased thickness

The XRD did not detect any oxides, however the EDX showed that the Al contents of the thin film is lower (16 to17 at%). Syed Zameer, is this what you meant by the change in stoichiometry?

I can use Debye-Scherer to get the size of the particles, but how do I differentiate between the nickel and aluminum particles?

Good morning,

if your EDX shows that you have indeed an increased Ni content in the film, that should be the reason for this effect, as Syed Zameer has already mentioned. However, it's interesting that the magnetic fraction decreases with higher sputtering times.

Just thinking - do you have the possilility to sputter on a shutter for much longer times than normal before opening the shutter? Normally I would expect some minutes to be enough for the sputter target to stabilize, but maybe it's worth a try with half an hour of sputtering against the closed shutter.

On the other hand, you could try changing the substrate temperature - if the undesired effect decreases for longer sputtering times, maybe decreasing the temperature slightly can help decrease it further.

Talking about Debye-Scherrer, pure Ni and pure Al particles as well as NiAl particles should be differentiatable by the their lattice constants. But I am not sure whether this is really visible in the experiment or if the overlapping signals will only results in a picture which cannot be evaluated.

Hi Dr. Ehrmann,

Thank you for the advice. I will try to do as you suggested.

The XRD on the thin films showed various AlNI3 peaks, the strongest being (111), followed by (200). I suppose this means it is not possible to differentiate the between the Al and Ni particles?

When you say magnetic fraction, I assume you are referring to the magnetic saturation (Ms, emu/cc)? Pardon me for the question, it is just that I am a bit confused on the similar sounding jargons used in various literature reviews.

Could there be a correlation between the particle size and the Ms? Since a magnetic domain would consist of several particles, smaller particles would mean lower Ms?

Hello Mr Reza,

to the best of my knowledge, AlNi3 tends to being ferromagnetic - apparently the stoichometry changes during growth. For me, your results sound as if at the beginning you get mostly AlNi3, while the stoichiometry changes more and more to NiAl for thicker layers.

Here http://arxiv.org/ftp/cond-mat/papers/0406/0406184.pdf, e.g., is an article about the particle size dependent saturation magnetization in Fe3O4 (with the highest Ms for a middle particle size); I assume that similar effects may occur in AlNi3 - sounds like an interesting experiment for you :-)

Dr Erhmann,

Thank you for the link! I really appreciate it! :)

Hello Mr Reza,

you're welcome, I hope you get some more insight into your interesting samples.

In order to understand what phases one should expect you should refer to the phase diagram Ni-Al. It is important to clarify the 20% Al - atomic or weight% ?

Depending on the response the result will differ qualitatively.

Diagram Ni-Al (in at%)

http://www.google.ru/imgres?imgurl=http%3A%2F%2Fwww.crct.polymtl.ca%2Ffact%2Fdocumentation%2FBinary%2FAl-Ni.jpg&imgrefurl=http%3A%2F%2Ffacts-pictures.vidzshare.net%2Fthis-diagram-shows-the-electron-shell-of-a-fluorine-atom%2F0.tqn.com*d*chemistry*1*7*z*8*1*fluorineatom.jpg%2F&h=504&w=640&tbnid=bOfq0qV57t_MkM%3A&zoom=1&docid=cRfjh7ls99TdKM&ei=uZSPU_XxFfP34QSktIEY&tbm=isch&ved=0CDEQMygIMAg&iact=rc&uact=3&dur=1602&page=1&start=0&ndsp=29&biw=1708&bih=760

Diagram Ni-Al (in wt%)

http://commons.wikimedia.org/wiki/File:Raney_nickel_phase_diagram.png

Hi Komogortsev,

I am referring to at%. Hence, the composition of the bulk will be AlNi3 + AlNi.

However, the XRD on the bulk only detected AlNi3.

The bulk samples were analyzed using VSM. The result did not show a hysteresis loop, indicating that it is non-magnetic.

According the diagram that I'd found 20 at% Al will result in equilibrium mix Ni3Al+fcc solid solution Ni(Al). I have marked the content by red asterisk. The fcc solid solution Ni(Al) will contain about 8-9 at% Al. According the paper at following link it is non magnetic in bulk but magnetic near the surface. I think this may be the reason of effect that you observed. The specific grain boundary surface contribute in you film may be the reason. Often it is much than the inner surface in the bulk.

http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.74.4059

Hi Komogortsev,

Thank you for the link.

The XRD that I did was done only on the the surface of the bulk sample. Hence, based on this paper, I need to check the orientation of the inner surface of the bulk, to check if it consists of magnetic fccNiAl structure, right?

Here is the paper.

I think you just can refer to this paper for interpretation of your observations. For example you may say:

The specific surface in the film always exceed it in the bulk. According ref[**] in Ni(Al) solid solution - the much specific surface the much Tc. So if the sample is non magnetic in the bulk at room temperature it may be magnetic in the shape of thin film.