220
7 Composites
7.1.2 Cu(Co) Alloys
Granular Cu(Co) alloys were essentially all deposited with sodium citrate as bath
additive. The works on d.c.-plated Cu–Cu alloys can be divided into two distinct
groups from the viewpoint of the structure of the deposits. If either no sodium citrate
was added to the bath (a few samples reported in [6]) or the sodium citrate concentration was less than the total metal ion concentration [7, 8], the hcp phase of the cobalt
always appeared beside the fcc Cu (or, beside a Cu-rich alloy that might contain
traces of Co). However, in all cases when the concentration of the citrate ions was
large enough [6, 9–12], alloys with fcc structure could be produced without the indication of the formation of the hcp cobalt phase. As deducted from the XRD pattern,
the purely fcc alloys obtained appear to be homogeneous; at least, the major XRD
peak can be fitted with one single function (mostly Lorentzian) and the mean lattice
distance varies systematically with the Co content as expected from the difference in
the atomic volume of the Cu and Co atoms. For the correct elucidation of the result,
one has to take into account that the sensitivity of the XRD method is usually claimed
to be about 2 vol.%, and even this limit can be achieved only if one has a mixture of
well-crystallized distinct phases. For a mixture of crystals with continuously varying
composition, the sensitivity of XRD is much worse.
In contrast, the magnetization and magnetoresistance properties of the d.c.-plated
deposits tell a different information on the structure and component distribution of
the Co–Cu alloys. Even the alloys found to be “homogeneous” with XRD show
either a significant saturation magnetization [6, 9], a measurable GMR [10] or both
[12] in their as-received state, although the mean composition would not allow it if
the sample were truly homogeneous. Therefore, it is obvious that the XRD patterns
obtained show a mean lattice distance for the distribution of which the width of the
XRD peak is somewhat indicative. However, the local fluctuation of the concentration
can be rather high and the Co-rich grains can grow large enough to become magnetic
already in the as-received state.
The solution composition used for the deposition of Cu(Co) granular alloys is
similar to those used for multilayer deposition in the sense that the Co
2+ concentration
is significantly larger than that of the Cu
2+ concentration. Concerning the deposit
composition as a function of bath operation parameters, the influence of essentially
all parameters can be taken as the natural consequence of the principles described
in Chap. 2. These are as follows [11]: (i) The increase in the current density (in
excess of the Cu diffusion-limited current density) leads to larger Co mole fraction
in the deposit. (ii) If the bath temperature is increased under otherwise unchanged
deposition conditions, the Co mole fraction in the deposit decreases, which is due to
the enhanced Cu deposition rate. (iii) The increase in Co
2+ concentration results in Co
enrichment in the deposit. This latter trend does not originate from electrochemical
principles but is the result of the competition between the growth of Co-rich and
Cu-rich zones.
7 Composites
7.1.2 Cu(Co) Alloys
Granular Cu(Co) alloys were essentially all deposited with sodium citrate as bath
additive. The works on d.c.-plated Cu–Cu alloys can be divided into two distinct
groups from the viewpoint of the structure of the deposits. If either no sodium citrate
was added to the bath (a few samples reported in [6]) or the sodium citrate concentration was less than the total metal ion concentration [7, 8], the hcp phase of the cobalt
always appeared beside the fcc Cu (or, beside a Cu-rich alloy that might contain
traces of Co). However, in all cases when the concentration of the citrate ions was
large enough [6, 9–12], alloys with fcc structure could be produced without the indication of the formation of the hcp cobalt phase. As deducted from the XRD pattern,
the purely fcc alloys obtained appear to be homogeneous; at least, the major XRD
peak can be fitted with one single function (mostly Lorentzian) and the mean lattice
distance varies systematically with the Co content as expected from the difference in
the atomic volume of the Cu and Co atoms. For the correct elucidation of the result,
one has to take into account that the sensitivity of the XRD method is usually claimed
to be about 2 vol.%, and even this limit can be achieved only if one has a mixture of
well-crystallized distinct phases. For a mixture of crystals with continuously varying
composition, the sensitivity of XRD is much worse.
In contrast, the magnetization and magnetoresistance properties of the d.c.-plated
deposits tell a different information on the structure and component distribution of
the Co–Cu alloys. Even the alloys found to be “homogeneous” with XRD show
either a significant saturation magnetization [6, 9], a measurable GMR [10] or both
[12] in their as-received state, although the mean composition would not allow it if
the sample were truly homogeneous. Therefore, it is obvious that the XRD patterns
obtained show a mean lattice distance for the distribution of which the width of the
XRD peak is somewhat indicative. However, the local fluctuation of the concentration
can be rather high and the Co-rich grains can grow large enough to become magnetic
already in the as-received state.
The solution composition used for the deposition of Cu(Co) granular alloys is
similar to those used for multilayer deposition in the sense that the Co
2+ concentration
is significantly larger than that of the Cu
2+ concentration. Concerning the deposit
composition as a function of bath operation parameters, the influence of essentially
all parameters can be taken as the natural consequence of the principles described
in Chap. 2. These are as follows [11]: (i) The increase in the current density (in
excess of the Cu diffusion-limited current density) leads to larger Co mole fraction
in the deposit. (ii) If the bath temperature is increased under otherwise unchanged
deposition conditions, the Co mole fraction in the deposit decreases, which is due to
the enhanced Cu deposition rate. (iii) The increase in Co
2+ concentration results in Co
enrichment in the deposit. This latter trend does not originate from electrochemical
principles but is the result of the competition between the growth of Co-rich and
Cu-rich zones.