Exchange Bias Manifesto

Ivan K. Schuller and G. Guntherodt

There is a time in the life of scientists when it is important to consider:

What is the Grand Purpose of it all?.

In this case, we started wondering, why all papers in Exchange Bias start by stating something like:

“Although exchange bias was discovered more than 40 years ago, the origin of this phenomenon is still not clear”.

This brief document attempts to put some order into this issue and raise some relevant questions. We would appreciate any corrections, comments and additions. However we must insist on one requirement: Be brief!.

Figure 1. Status of the Exchange Bias field, based on the introduction of literally hundreds of papers.

Figure 2. One possible good (i.e.brief) way to summarize the status of research for a particular system. In this case for FeF2 (110)-Fe (polycrystalline). From M. Fitzsimmons et al, Phys Rev. B 65, 134436(2002).
 

“Philosophical” Considerations

• There are many apparently confusing, contradictory (?) facts. See figure 1.
• There are very diverse systems, with various degrees of control on the structure.

Because of this it is important to have some brief concise way of classifying the main results, as indicated in figure 2 above. This should be amplified and other similar tables be constructed.

• The purpose is to find out which facts are essential, which are just side issues
• It is probably useful (?) to categorize the systems:
1. Crystalline AF – Interface - Crystalline F
2. Complex AF- Interface – Complex F
3. Disordered Magnet – Interface – F
4. Hard F – Interface – Soft F (Spring Magnet)
 
• The various parameters needed to characterize an exchange biased system are:
1. Growth method (thin film, natural oxidation, …)
2. Characterization tools (quantitative, qualitative?)
3. AF crystal structure
4. F crystal structure
5. Interfacial roughness
6. Neel temperature
7. Curie Temperature
8. Thin film crystal structures
9. Bulk and thin film anisotropies
10. Bulk and thin film spin structure
11. Blocking temperature
12. Cooling field
13. Cooling procedure
14. Exchange Bias
15. Coercivity
16. Left and Right hand dependence of M-H as a function of T
17. Reversal Mode
18. Training?

Definitions

1. “Parallel”, “perpendicular” to the interface.
2. “90°”, “collinear” with respect to the AF spins.

 

ESTABLISHED(?) EXPERIMENTAL FACTS TO BE EXPLAINED
(More or less ordered systems)

 

Experiment	System
1) Spin orientation	FeF2-Fe, Co/Pt-CoO (?), FeF2-Co/Pt
Parallel (F-AF) Spins Favored
2) Compensated vs uncompensated	depends on system? maximized
	Compensated (FeF2, MnF2, FeMn)
	Uncompensated (CoO)
3) Positive He	FeF2-, MnF2-Fe, Co-CoO
4) Coupling: A) 90° B) Collinear	LaFeO3- Co, FeMn- Co
5) Roughness non-monotonic	MnF2-Fe
6) Ion irradiation initial increase	NiFe/FeMn
7) AF dilution increase	CoO-Co, FeF2-Fe
	Metallic (Py-IrMn+Rh,Pt..)
8) Loose spins  ???	Py-CoO
9) Crystallinity
Single Xtals small Epitaxial varied Twin large Poly intermediate	CoO, NiO, FeF2 ???   FeF2 FeF2, CoO
10) He maximized (w/r AF anisotropy)
Along  ?	FeF2-Co?, MnF2-Fe
Close	CoO-FeNi
11) AF anisotropy	MnF2<<<<FeF2<<CoO
12) 3rd order anisotropy	MnF2-Fe, NiFe-CoO
13) Asymmetric loop	FeF2-, MnF2-Fe, CoO-Co,
	NiMn-NiFe, PtMn-NiFe
14) Asymmetric Reversal	FeF2 (twinned), CoO
15) Training	Fe, Co, Ni, Ni50Fe50, Ni81Fe19 - FeMn
	La(2/3)Ca(1/3)MnO,La(1/3)Ca(2/3)MnO3
	Co, -CoO,Ni, NiFe - NiO, NiFe - aFe2O3,
	Co - LaFeO3,Co – IrMn
No	FeF2-, MnF2- Fe
16) Rotation in  F, some	FeF2-, MnF2-Fe, CoO-Co
17) Vertical shift	MnF2-,FeF2-Fe, CoO-Co
18) Coercivity  enhancement	FeF2-,MnF2-Fe,CoO-Co, NiFe-FeMn
	La(2/3)Ca(1/3)MnO3 -
	La(1/3)Ca(2/3)MnO3 -, all(?)
19) Reversible measurements bigger	Co-CoO (AMR, BLS)
20) Thickness dependence
i)   AF (up+1/d)	FeF2-Fe, CoO-Co, FeMn-FeNi
AF (up+flat)	CoO-Co, FeMn-FeNi
ii)   F (1/d)	FeF2-Fe
21) Cooling field	FeF2-, MnF2-Fe, CoO-Co
22) Magnetic history, cooling from T

RANDOM QUESTIONS

1) Is more than one theory possible? This is the case after all for the resistivity.

2) Is it possible that different structural parameters are responsible for H_e and H_c? The angular symmetries are different apparently?

3) Is there a correlation between H_e and asymmetry in reversal? Does large asymmetry imply large H_eb?

4) Is the correlation between asymmetry and 3rd order anisotropy important?

5) What about the crystalline orientation?

6) Are domain walls important for H_e, H_c, both or one of them?

7) Is crystallinity (rocking curve width) important?

8) Are uncompensated spins needed, do they exist, ...?

9) Is the phenomenology of Exchange Bias the same independent of system: bilayer, multilayer, disordered AF, spring magnet,…?

10) Will we ever solve this or we will get tired before and die?. This is more relevant to me than some of you young lads.

1. Numerical simulations show anisotropic reversal in M either from:

a) Third order anisotropy.
b) Stoner-Wohlfart + Domains.