The temperature‐dependent electron‐spin‐resonance linewidth ΔH(T) may be used to investigate the effect of the geometry and interlayer material on the magnetic properties of multilayered systems. We compare ΔH(T) in CuMn/Al2O3 multilayers with previous measurements of CuMn/Cu samples. CuMn/Al2O3 samples with CuMn thicknesses, WSG, from 40 Å to 20 000 Å obey the same form as the CuMn/Cu system, but show quantitative differences in the fitting parameters. The linewidths of the CuMn/Al2O3 samples, even in the bulk, are systematically larger than the linewidths for the CuMn/Cu samples, suggesting that the ESR linewidth is sensitive to differences in sample growth and structure. The value of the minimum linewidth decreases with decreasing WSG in the CuMn/Al2O3 series, but remains constant in the CuMn/Cu series. Although susceptibility measurements of the freezing temperature Tf do not differentiate between samples with WSG≥5000 Å, the ESR linewidth is sensitive to changes at larger length scales. This experiment emphasizes the importance of considering both the total sample thickness, as defined by the range of the conduction electrons, and the spin‐glass layer thickness in analyzing the ESR linewidth in multilayers.

1.
J. A.
Cowen
,
G. G.
Kenning
, and
J. M.
Slaughter
,
J. Appl. Phys.
61
,
4080
(
1987
);
G. G.
Kenning
,
J. M.
Slaughter
, and
J. A.
Cowen
,
Phys. Rev. Lett.
59
,
2596
(
1988
).
2.
G. G.
Kenning
,
Jack
Bass
,
W. P.
Pratt
, Jr.
,
D.
Leslie-Pelecky
,
Lilian
Hoines
,
W.
Leach
,
M. L.
Wilson
,
R.
Stubi
, and
J. A.
Cowen
,
Phys. Rev. B
42
,
2393
(
1990
).
3.
A.
Gavrin
,
J. R.
Childress
,
C. L.
Chien
,
B.
Martinez
, and
M. B.
Salamon
,
Phys. Rev. Lett.
64
,
2438
(
1990
).
4.
D. L.
Leslie-Pelecky
and
J. A.
Cowen
,
Phys. Rev. B
46
,
9254
(
1992
).
5.
D. L.
Leslie-Pelecky
and
J. A.
Cowen
,
Phys. Rev. B
48
,
7158
(
1993
).
6.
The range of WSG over which Eq. (1) is applicable has been debated in the literature. See Refs. 2 and 3 for details.
7.
P.
Granberg
,
P.
Nordblad
,
P.
Svdlindh
,
L.
Lundgren
,
R.
Stubi
,
G. G.
Kenning
,
D. L.
Leslie-Pelecky
,
J.
Bass
, and
J.
Cowen
,
J. Appl. Phys.
67
,
5252
(
1990
);
L.
Sandlund
,
P.
Granberg
,
L.
Lundgren
,
P.
Nordblad
,
P.
Svedlindh
,
J. A.
Cowen
, and
G. G.
Kenning
,
Phys. Rev. B
40
,
869
(
1989
).
8.
J.
Mattson
,
P.
Granberg
,
L.
Lundgren
,
P.
Nordblad
,
G.
Kenning
, and
J. A.
Cowen
,
J. Magn. Magn. Mater.
104–107
,
1621
(
1992
).
9.
J. Bass and J. A. Cowen, in Recent Progress in Random Magnets, edited by P. H. Ryan (World Scientific, Singapore, 1992),
10.
S. E.
Barnes
,
Adv. Phys.
30
,
801
(
1981
).
11.
R. H.
Taylor
,
Adv. Phys.
24
,
681
(
1975
).
12.
For the T = 0 fit, the values of A and B are unchanged and the prefactor for the divergence is not directly comparable with C.
13.
A. Gavrin (unpublished). The values of κ in the CuMn/Al2O3 increase faster with WSG than the corresponding CuMn/Cu samples. One explanation may be that the Al2O3 interlayered samples approach 2D behavior at larger values of WSG than CuMn/Cu samples.
14.
A minor difference between the samples is that the film from the CuMn/Cu series shown in Fig. 3 is capped with 100 Å of Cu on the top and bottom of the film to protect the sample from oxidation and diffusion of silicon from the substrate. Measurements of films without the protective layer show linewidths that are slightly lower than in the sample with the protective layers.
15.
H.
Nagashima
and
H.
Abe
,
J. Phys. Soc. Jpn.
32
,
1507
(
1972
).
This content is only available via PDF.
You do not currently have access to this content.