3
8
G.A. Jones et al. / Journal of Magnetism and Magnetic Materials 184 (1998) 28—40
Ehrenfest’s formula [19] viz.,
pared by ED. Moreover, as in the case of Cu Co ,
ꢉꢄ
ꢆ
it is observed that lattice planes are generally more
distinct within patches, but they can encroach upon
neighbouring lighter contrast regions. Thus, it
would be an over-simplification to state that the
dark patches consist of islands of crystalline matter
contained within an amorphous matrix, as is the
case described by Hong et al. [2] in their Co/poly-
mer system. An alternative explanation is that the
patches represent some sort of clustering process;
the magnetic results confirmed this hypothesis in
the case of Cu Co . With regard to the CoRe
sin h "1.23j/2d,
ꢀ
where d is the interatomic distance and j is the
wavelength of the radiation. (This angle may not
necessarily coincide with the low-angle Bragg peak
of the corresponding crystalline phase.) Assuming
a random, dense packing system, which is a reason-
able assumption for Co and Re, then d may be
taken as the atomic diameter (0.125 and 0.137 nm
for Co and Re, respectively). In the case of complete
segregation of the two elements, the first amorph-
ous maxima would occur at 0.51° (Re) and 0.55°
ꢉꢄ
ꢆ
alloys, the patches or clusters in Co Re are small,
ꢄ
ꢉꢆ
(Co) for 300 kV electrons. We cannot demonstrate
ca. 3 nm. On the other hand, the putative clusters
from our electron diffraction patterns that segrega-
tion into two distinct amorphous phases occurs in
these films, although it is well attested [18,19] in
some alloys. Two broad maxima at close angles
would be difficult to resolve in the microscope and,
in any event, only a single amorphous phase may
be present. A significant comparison is with the
are much bigger in Co Re , and of a size such
that they could give rise to discernible magnetic
effects. The X-ray detector on the microscope was
unable to resolve any meaningful compositional
difference between the clusters and adjacent areas:
this was true for both compositions.
ꢇꢁ ꢇꢁ
corresponding scattering angles for the +1 0 1
ꢀ
0,
4.3. Magnetic measurements
and +1 0 1 1, reflections, both of which have been
identified in the diffraction patterns. In the case of
ꢀ
In the electrodeposition of our samples, we have
attempted, by working close to equilibrium condi-
tions, to deposit a crystalline inhomogeneous alloy.
Clearly, this was only partially successful, since our
TEM measurements show that in all samples both
a crystalline and an amorphous phase were present.
We assume that the crystalline phase is the granu-
lar alloy with inclusions rich in Co; there is some
evidence for this in the dilute Co alloys in the TEM
results. If this is the case, the magnetic behaviour
that we would expect is that of superparamag-
netism. Part of the evidence for this is that, in the
dilute 4% Co alloy, we see hysteresis at low temper-
atures. For the 4% Co alloy this disappears at
about 170 K, Fig. 3, a temperature which we would
assign to the maximum blocking temperature of the
film. Also, ZFC/FC curves for that alloy, Fig. 5,
show a bifurcation at that temperature. We also
expect, for pure superparamagnetism, to observe
a scaling of the magnetization when plotted as
a function of H/¹ at temperatures above ¹ . The
fact that we do not observe such a scaling effect,
Fig. 8, does not preclude superparamagnetism
since such non-scaling can be attributed to interac-
tion effects. Indeed, scaling in thin films where
Co Re these are found, using Bragg’s law, to be
ꢇꢁ ꢇꢁ
0
.50° and 0.56°, respectively, for 300 kV electrons,
i.e. they straddle the values obtained above from
the Ehrenfest relation. Thus, we may conclude that
the inner halo in the diffraction patterns occurs at
an angle in the vicinity of that expected from an
amorphous phase of CoRe or segregated amorph-
ous phases of either Co or Re. Electron micro-
graphs of both alloys support this assertion. While
acceding that the imaging conditions must be ideal
for the formation of lattice plane images, best at-
tained in a thin sample, we believe that the dearth
of fringes in Fig. 1b, Fig. 1c and Fig. 2 is not just
a reflection of instrumental deficiency, but indicates
that the films comprise some regions which are
crystalline and others which are not.
One final aspect of the micrographs should be
noted, namely the mottling effect. This could rep-
resent gradations in thickness arising from the ion-
thinning procedure but it is not apparent why the
artefacts of such a process should differ between
samples or even within a single sample. In fact, we
have previously reported [20] a mottling phenom-
enon in an alloy of Cu Co which was also pre-
ꢉꢄ
ꢆ
Jones
