I have Co/Pt multilayer systems with individual layer thickness ~ 1nm. The XRD suggests coherency strain in the multilayers. The magnetic properties suggest interface sharpening on annealing. But, on the other hand, there are many studies which reported intermixing of Co-Pt when annealed.
Some intermetallic ordered compounds (CoPt3, CoPt, Co3Pt) are stable, thus intermixing may occur. Annealing of Co/Pt MLs leads to intermixing – we have done such experiments. Back-diffusion – a process which sharpens interfaces in multilayers was reported in Au/Co system (e.g. by H.P. Oepen). You should also be aware that the quality of interfaces depends on fabrication method (MBE vs. sputtering). In Pt/Co multilayers you should observe perpendicular magnetization with rectangular hysteresis loop. How do magnetic properties change upon annealing in your MLs? It may happen that an ordered alloy is formed and exhibits strong perpendicular anisotropy.
@Andrzej Wawro: Thanks for the reply. The multilayers has been deposited by using DC magnetron sputtering with base pressure of ~5x10^-9 torr. The perpendicular magnetic anisotropy is increased in Co/Pt after annealing. Especially the Hard axis, which is in-plane in this case, became harder (the field required to attain the magnetic saturation increased). Though increase in Ms is observed, it is within error bar- But I'm sure Ms is not decreasing on annealing.
Further, the XRR data suggests multilayer structure after annealing, with not very much difference to XRR of as deposited samples.
In most of the Co/Pt studies I've seen, decrease in perpendicular anisotropy is reported and mentioned that it is due to intermixing at interface. I've seen one or two papers in which though increase in perpendicular anisotropy is mentioned but no reason is given.
In sputtering method you should obtain rougher interfaces than in MBE one. What is annealing temperature? In our case annealing at 250 and 400 C leads to magnetic changes, suggesting alloying at interfaces (we have studied Pt/Co/Pt trilayers). It may happen that your interface is not alloyed but rough. Then a delicate annealing may smoothen your interfaces with no substantial interdiffusion due to annealing. Such explanation is consistent with experimental observation – perpendicular anisotropy is stronger for smoother interfaces. You have to remember that XRR does not differentiate between roughness and alloying. What changes in XRR spectra do observe after annealing? In my feeling it should be a lower rate decrease of reflected beam intensity with increasing angle. TEM observations of the multilayer cross-section can be another way to solve your problem.
Strain relaxation will take place in all the multilayer systems during annealing. In this case hexagonal Co and cubic Pt should have more strain. If your temperatures are not high or long time then you can reduce strains without inter-diffusion.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Thin Films and Nanotechnology