Developing plating baths for the production of reflective Ni-Cu films

Article abstract of DOI:10.1016/j.electacta.2011.12.049

Electrolytic baths have been developed to obtain bright nickel-rich nickel-copper micrometric films with good mechanical properties and high reflectivity, valid as components of optical mirrors. As metal source, the developed baths contain nickel (II) sulfamate and copper (II) acetate. Citrate was used as buffer and complexing agent to favour the simultaneous electrodeposition of copper and nickel. The influence of citrate concentration, solution pH, [Ni(II)]/[Cu](II) ratio and applied potential on the composition was studied. In order to achieve constant composition throughout the deposit thickness, potentiostatic technique was selected. The optimization process was followed according to a simple factorial design. By adjusting the conditions, compact, uniform and fine grained nickel-copper (Ni-Cu) electrodeposits with high Ni percentages and low roughness were obtained. The deposits show mechanical properties suitable for electroforming. The optical properties (reflectivity and gloss) of the Ni-Cu films make them adequate to be elements of optical mirrors.

Full text of DOI:10.1016/j.electacta.2011.12.049

Electrochimica Acta 62 (2012) 381–389
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Developing plating baths for the production of reflective Ni–Cu films
P. Callejaa, J. Esteveb, P. Cojocaruc, L. Magagninc, E. Vallésa, E. Gómeza,∗
a Electrodep. Departament de Química Física, Universitat de Barcelona, 08028 Barcelona, Spain
b Departament de Física Aplicada i Òptica, Universitat de Barcelona, 08028 Barcelona, Spain
c Dipartimento di Chimica, Materiali e Ingegneria Chimica “Giulio Natta”, Politecnico di Milano, 20131 Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Electrolytic baths have been developed to obtain bright nickel–rich nickel–copper micrometric films with
good mechanical properties and high reflectivity, valid as components of optical mirrors. As metal source,
the developed baths contain nickel (II) sulfamate and copper (II) acetate. Citrate was used as buffer and
complexing agent to favour the simultaneous electrodeposition of copper and nickel. The influence of
citrate concentration, solution pH, [Ni(II)]/[Cu](II) ratio and applied potential on the composition was
studied. In order to achieve constant composition throughout the deposit thickness, potentiostatic tech-
nique was selected. The optimization process was followed according to a simple factorial design. By
adjusting the conditions, compact, uniform and fine grained nickel–copper (Ni–Cu) electrodeposits with
high Ni percentages and low roughness were obtained. The deposits show mechanical properties suitable
for electroforming. The optical properties (reflectivity and gloss) of the Ni–Cu films make them adequate
to be elements of optical mirrors.
Received 27 July 2011
Received in revised form
12 December 2011
Accepted 12 December 2011
Available online 19 December 2011
Keywords:
Electrodeposition
Ni–Cu alloy
Gloss
Reflectivity
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
several requirements: surface finish, dimensional tolerances and
flatness or radius of curvature have to closely match the original.
Aerospace industry requires lightweight mirror shells as mir-
ror modules of the telescopes [1]. Various technologies have been
developed for the production of these such as galvanoplastic repli-
cation or direct polishing [2]. Different studies demonstrated that
electroforming allows mirror shell production with large opti-
cal surface area, enabling the production of low cost, isotropic,
precision optical shell-like reflectors highlighting excellent perfor-
mances. In principle, any reflector thickness can be electroformed
with the desired curvature and surface characteristics close to the
optical quality of the mandrel. Electroforming offers the ability
to precisely replicate in a hard, wear-resistant metal a pattern
by electrodeposition in a suitable electrolyte. Another benefit of
this technology is the ability to replicate sequentially, thus pro-
ducing mirrors from only one master. Replicated nickel optics
have been used extensively in X-ray astronomy. In the electro-
formed replication (ER) fabrication process, nickel mirror shells
are electroformed onto a figured and superpolished mandrel.
Afterwards, mirror shells are released from the mandrel using
differential thermal contraction [3]. This technique involves form-
ing an extremely lightweight mirror by electroplating nickel and
nickel-based alloys onto a highly polished precision mandrel [4].
For such an optical electroform to be usable, it does have to meet
This can only be accomplished if internal stresses in the elec-
troformed layer are kept to a minimum. These can be explained
as the tendency of the deposited layer to either contract (tensile
stress) or expand (compressive stress) relative to the substrate
upon separation from it. This is determined by a number of fac-
tors, such as deposition rate, bath chemistry, additives, agitation,
etc. To ensure compositional uniformity of the finished product,
solution conditions must be controlled within a narrow range, the
parameters being regularly checked and adjusted. The electrode-
position of the Ni–Cu system has attracted much attention and it
has been widely treated due to the high versatility of applications
of this alloy. These applications derive from their electrocatalytic
properties [5–7], magnetic or magnetoresistive properties [8–13],
their anticorrosion properties [14], the recent exploited thermo-
electric properties [15–17], as well as their mechanical strength
[18,19] and reasonably good wear resistance [20–22]. Depending
on the final application, different electrolytic baths has been pro-
posed, which led to variable percentages of the metals in the alloy
[18,23–25].
The main interest of this contribution is the study of the
process of development of nickel–copper baths able to prepare
nickel–copper (Ni–Cu) alloys which could be used in optical mir-
ror production. The combination of good corrosion resistance and
interesting mechanical properties makes Ni–Cu alloy suitable. Low
copper proportions are desirable to improve nickel mechanical
properties while maintaining high film gloss and reflectivity.
∗
Corresponding author. Tel.: +34 934021234; fax: +34 934021231.
E-mail address: e.gomez@ub.edu (E. Gómez).
0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.12.049

382
P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
In order to facilitate codeposition, citrate was selected as a com-
conditions were as follows: 250 mN peak load, 10 s load-
ing/unloading time and 5 s holding time at peak load. The reported
values are the average of 5 measurements taken on three different
samples prepared in the same conditions from the same bath. To
analyse the roughness of the deposits an interferometric surface
analysis microscope Plu Neox 3D optical Profiler of Sensofar was
used.
mon complexing agent for Ni–Cu baths [26–33], since it not only
has a low toxicity but does also act as a brightening [23], levelling
[34] and buffering agent [35], thus eliminating the need for other
bath additives. Despite the numerous studies performed regarding
obtaining nickel–copper alloy, are few ones devoted to prepare rich
nickel deposits, and their results make difficult to draw conclusions
in acidic media. Moreover, as some citrate nickel–copper baths pro-
posed in the literature show instability [26,27] related in some
cases to specific complex formation [36,37], a step by step process
was considered to develop the basic bath formulation. Firstly, for
the selected salts, single-metal deposition behaviour was studied
at different citrate concentrations and solution pH. Our objective
was producing, from solutions with low [Cu(II)]/[Ni(II)] ratio, Ni–Cu
films with no less than 75 wt.% Ni, having in mind the bath sta-
bility. The deposit quality and compositional variation obtained
from each bath was evaluated prior to selecting conditions. This
optimization process was followed according to a simple factorial
design [38], in which we found the significant effects that exert the
main interaction in the general behaviour in the screening step. This
was followed by an optimization process in which the conditions
leading to the best response were defined.
Electrical resistivity of the different alloys has been measured
after the thin foils have been detached from their inox substrate.
The foils have been cut to an approximate square shape and the
electrical measurements have been done according to Van der Paw
[39] with four electrical contacts located at the four corners of
the square sample. DC current (100 ␮A) was injected between two
adjacent contacts and DC voltage was measured at the two opposite
contacts. This procedure has the benefit of not requiring cutting a
precise shape stripe from the small foil piece and furthermore it
does not require perfect electrical contacts: pressure springs are
enough (measurement results have been contrasted with mea-
surements on strip-shaped pieces with two copper wire electrical
contacts welded at the edges). Four contacting schemes are possi-
ble by cycling the square corners. For each contacting scheme two
measurements are done by reversing the current polarity and two
more measurements by reversing the voltage contacts. In total this
meant 16 voltage measurement values that were added together to
obtain sample’s resistivity. It is necessary, in addition, to precisely
know the foil thickness. This was measured by two methods: metal-
lographic embedment of the foil and measurement of the foil edge
thickness from its SEM micrograph. And also by direct mechanical
contact: by using a precision Mitutoyo micrometer fitted with a
spherical anvil.
Gloss measurements were conducted with a BYK-Gardner
micro-gloss meter with an extended beam white light, and a 60◦
measurement angle. Sample surface should be carefully cleaned
before measurement, by wiping it with an optical cleaning cloth
and ethanol, in order to obtain consistent results. The calibra-
tion was performed automatically by means of a highly polished
black standard integrated in the glossmeter. Multiple measure-
ments with each coating were made in order to evaluate data
statistically.
Spectral reflectivity measurements were conducted with
the help of an OceanOptics fiber-optic spectrophotometer
USB4000 with a reflection probe designed for quasi-perpendicular
reflectance (3◦ measurement angle). The collected reflectivity spec-
tra extend in the visible range (400–750 nm) and were normalised
against two reference standards: a high reflectivity (96%) aluminum
standard mirror and a flat spectrum magnesium oxide standard
plate.
2. Experimental
The electrochemical measurements were performed in
conventional three-electrode cell using an Autolab with
a
PGSTAT30 equipment and GPES software. Chemicals used were
Ni(H2NSO3)2·4H2O (nickel sulfamate), (CH3COO)2Cu·H2O (copper
acetate) and Na3C6H5O7·2H2O (sodium citrate), all of analytical
grade. Sulfamate, citrate and acetate were selected as being
less aggressive with the environment. All solutions were freshly
prepared with water first doubly distilled and then treated with
a Millipore Milli Q system. Before and during the experiments,
solutions were de-aerated with argon. The temperature was
maintained at 25 ◦C.
Working electrodes were vitreous carbon (Metrohm) used for
the basic electrochemical study and inox steel to grow the films.
The vitreous carbon electrode was polished to a mirror finish with
alumina of different grades (3.75 and 1.87 ␮m) and ultrasonically
cleaned for 2 min in water before each experiment. Prior to the
electrodeposition inox steel (stainless steel AISI 316 with maxi-
mum percentages of C, Mn, Cr, Ni and Mo of 0.05, 2, 17, 12 and
2.5 wt.% respectively) was polished with grit paper (FEPA p =/
4000) and washed with abundant water. The reference electrode
was an Ag/AgCl/1 mol dm−3 NaCl electrode mounted in a Luggin
capillary containing 0.25 mol dm−3 sodium sulfamate solution. All
potentials were referred to this electrode. The counter electrode
was a platinum spiral in the basic study on vitreous carbon or a Ni
sheet during the preparation of deposits.
3. Results and discussion
Deposit morphology was examined with a Hitachi 2300 scan-
ning electron microscopy and the resolution of the grain with a
Hitachi H-4100 FE field emission (FE-SEM). Elemental composi-
tion was determined with an X-ray analyzer incorporated in Leica
Stereoscan S-360 equipment. The structure was resolved with X-
ray powder diffraction (XRD), using a conventional Bragg–Brentano
3.1. Ni and Cu deposition behaviour
Ni and Cu deposition was analysed as a function of the citrate
concentration and the solution pH. Attending to the wide difference
between the standard potentials of nickel and copper, citrate was
used as complexing agent to decrease the difference between the
deposition potentials of both metals; the influence of the citrate
concentration was analysed between 0 and 0.25 M. The pH was
varied between 3 and 6 because more acidic media could favour
hydrogen codeposition and higher ones lead to instabilities. Solu-
tions with analytical concentration of 0.025 M for [Cu(II)] and 0.5 M
for Ni(II) were tested for this study.
˚
diffractometer Siemens D-500. The Cu K␣ radiation (ꢀ = 1.5418 A)
was selected using a diffracted beam curved graphite monochro-
mator. The X-ray powder diffraction diagrams were measured in
the 5–100◦ 2ꢁ range with a step range of 0.016◦ and a measuring
time of 100 s per step.
Ni–Cu coatings were characterised in terms of micromechanical
properties. Vickers microhardness (HV) data were obtained from
penetration depth-load curves by means of a FISCHERSCOPE®
H100 microhardness measurement system. Measurements
The percentage of each species in solution was calculated using
the MEDUSA-Chemical Diagrams (2.0) and HYDRA-Hydrochemical
Database (2.0) Softwares [40,41]. From the selected concentrations

P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
383
3.2. Ni–Cu deposition
3.2.1. Screening step
In the screening step, different parameters have been fixed: a
[Ni(II)]/[Cu(II)] ratio of 20, as copper is nobler than nickel, citrate
concentration of 0.25 M and a temperature of 25 ◦C. Solution pH
(between 3 and 6) and applied potential were considered as vari-
able parameters. For each one of the solutions analysed, the applied
potential range was selected from the corresponding voltammetric
study.
For all solutions studied, negative voltammetric scan shows a
first peak due to copper (II) reduction (Fig. 3A, curve a); following
the scan, an important current increase appeared when simulta-
neous nickel codeposition began (Fig. 3A, curve b), followed by
hydrogen evolution (Fig. 3A, curve c). In the positive scan, indepen-
dently of the cathodic limit, an oxidation peak centred on 50 mV
appeared. A second oxidation peak, at 175 mV is observed only at
intermediate deposition potentials (Fig. 3A, curve b), when the scan
was reversed near the onset of nickel deposition. When the scan
was reversed before Ni codeposition (curve a), the oxidation peak
corresponds only to Cu oxidation. When Ni begins to electrodeposit
(curve b), the oxidation scan reflects the Cu oxidation peak of the
initial deposited Cu and the peak corresponding to the oxidation of
the Cu present in the alloy or directly to Cu-rich alloy, at more posi-
tive potentials due to the presence of Ni. When the scan is reversed
at more negative potentials (curve c) Ni content in the alloy must
increase and moreover, significant hydrogen evolution occurs. The
formation of hydroxides, due to the increase of the local pH, hin-
ders the oxidation and the oxidation peak recorded corresponds
to the copper not covered/not affected by the hydroxides. The Ni
alloy oxidation is not observed in a similar way that occurs in pure
electrodeposited nickel.
Fig. 1. Diagrams of M(II)–citrate complexes present in solution for 0.25 M
Na3C6H5O7 and M(II): (A) 0.025 M Cu(II) and (B) 0.5 M Ni(II).
The voltammetric curves were sensitive to the solution stirring:
an increasing stirring causes a continuous increase in the reduction
charge (Fig. 3B), as occurs in other baths [37]. Only the Cu oxidation
peak is observed as expected for a cathodic limit of −1.4 V; the
observation of nickel or alloy oxidation is not expected in these
conditions in according to the results shown in Fig. 3A. The charge
involved in the Cu oxidation peak increases with the stirring rate
because more Cu is deposited.
of sodium citrate and copper acetate (or nickel sulfamate), the ionic
strength, the bath temperature and the stability constants of the
metallic cations with the citrate species were considered. From
the Hydra Database, the proportion of the different metal-citrate
complexes as a function of the pH of solution was estimated. Fig. 1
shows the predominant species of copper (II) (Fig. 1A) and nickel (II)
(Fig. 1B) complexes in citrate solution as a function of the solution
pH.
Electrochemical behaviour of the separate metallic cations was
followed at all conditions by means of voltammetric experiments.
Fig. 2A shows the effect of successive citrate additions to a copper
solution. On increasing citrate content, reduction process onset was
delayed to more negative potentials. For a fixed analytical citrate
concentration, reduction current appeared at more negative poten-
tials as pH increased (Fig. 2B). In both cases, the shift of the onset
of the deposition process to negative potentials was a consequence
of a higher copper (II) complexation.
For each solution, the potential range at which Ni and
Cu codeposit was selected. Samples of −25 mC were prepared
using potentiostatic technique in order to evaluate composi-
tion. Galvanostatic technique was not used since no constant
potential was attained, which could lead to a variation in the
composition–thickness profile. The solutions were gently stirred
(300 rpm) during the deposition in order to prevent copper (II)
depletion, the less concentrated electroactive species, in electrode
environment during the deposition process. Greater stirring rate
was discarded to avoid excessive copper incorporation into the
deposits, because the increase in the agitation led to an increase
in the copper content [42]. Stirring is also useful to prevent the
possible both local pH variation caused by the possible hydrogen
coevolution, and the subsequent precipitation of hydroxides. j–t
profiles of deposits preparation under stirring conditions showed
quasi-stabilization of the current at each potential applied (Fig. 4).
From all solutions studied, obtained deposits were metallic grey
in colour at nickel percentages high enough. Nickel-rich coatings
prepared were homogeneous, compact, smooth and fine-grained
as the manner that grains were hardly observed by SEM, Fig. 5 cor-
responds to a representative image of this kind of deposits. Table 1
shows the compositional analysis of some deposits prepared. Nickel
content increased significantly as the applied potential decreased,
but levelling off at the more negative applied potentials. Deposits
obtained at fixed electrodeposition conditions but at variable depo-
sition times showed the same composition. This proves that the
Citrate also complexes with nickel, but in lower proportion
than with copper. Then, at fixed pH, moderate shift of the nickel
deposition onset by increasing citrate concentration was observed
(Fig. 2C).
In view of this, citrate confirms as a good election indeed. High
citrate concentrations and pH values in the selected 3–6 range,
favoured the approximation of the Cu deposition to the Ni(II)
reduction. Nevertheless, the 0.5 M nickel sulfamate solution con-
taining 0.25 M citrate did show instability, especially at pH 6, since
precipitates were observed one week after preparation. Similar
occurs at pH 6 when copper acetate is present. Reason because it
was considered to reduce the [Ni(II)], solutions containing 0.25 M
of nickel sulfamate were selected to analyse the Ni–Cu deposi-
tion in the screening step, in order to avoid the formation of
precipitates.

384
P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
Fig. 2. Cyclic voltammograms at v= 50 mV s−1, vitreous carbon, ω = 0 rpm for solutions: (A) at pH = 4, [Cu(II)] = 0.025 M and [Na3C6H5O7]: (curve a) 0 M, (curve b) 0.1 M, (curve
c) 0.25 M; (B) [Cu(II)] = 0.025 M and [Na3C6H5O7] = 0.25 M: (curve a) pH = 3, (curve b) pH = 4, (curve c) pH = 6; (C) at pH = 4, [Ni(II)] = 0.50 M and [Na3C6H5O7]: (curve a) 0 M,
(curve b) 0.1 M, (curve c) 0.25 M.
Fig. 3. Cyclic voltammograms of 0.25 M Ni(II) + 0.0125 M Cu(II) + 0.25 M Na3C6H5O7 solution, at v= 50 mV s−1, vitreous carbon. (A) pH = 4, ω = 0 rpm and different cathodic
limits: (curve a) −800, (curve b) −1100 and (curve c) −1400 mV. (B) pH = 5: (curve a) ω = 0, (curve b) ω = 300 and (curve c) ω = 550 rpm. Inset: detail of currents.

P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
385
Nickel deposits were crystalline, showing a cubic face-centred (fcc)
structure and 200 preferential orientation (Fig. 6). Copper incor-
poration into the deposit maintains the fcc phase but shifts the
diffraction peaks to lower angles and induces an 111 preferred
orientation (Fig. 6); the diffraction peaks appear at intermediate
positions between those of pure Ni and Cu, the specific position
depending on the Cu percentage in the deposits. A solid solution is
always formed between both metals.
The results of this screening step indicate that the deposit com-
position could be modulated as a function of electrodeposition
conditions. For a fixed solution, nickel percentage increases both as
the applied potential was made more negative and upon increas-
ing solution pH. However, using the selected both [Ni(II)]/[Cu(II)]
ratio and citrate concentration, the expected nickel percentages
are lower than the proposed as objective, even at the higher pH
solution. On the other hand, in the solution of pH = 6 precipitates
were observed after 15 days of preparation at difference that occurs
when other salts were used to develop the electrolytic bath [37].
3.2.2. Optimization step
According to these results optimization step was made pursu-
ing two objectives: to avoid instabilities in the solutions and to
ensure greater nickel percentages. Solutions prepared at pH = 6
were discarded due to the instabilities observed in the screening
step. Thus, solution pH was maintained in the 4 ≤ pH ≤ 5 range.
In order to prevent precipitations, the citrate concentration was
reduced to 0.18 M in this optimization step. Simultaneously, to
enhance the Ni percentage into the deposits, the ([Ni(II)]/[Cu(II)])
ratio was increased, as the manner that the nickel concentration
was ranged between 0.30 ≤ Ni(II) ≤ 0.40 M, maintaining the copper
concentration. Applied potentials were selected according to the
corresponding voltammetric study, and were constrain to those
that led to rich nickel deposits. Samples of −50 mC on vitreous
carbon were prepared. Potentiostatic curves showed the general
profile obtained previously (Fig. 7). Fig. 8 shows the dependence of
the nickel percentage with the potential for deposits prepared from
the solutions containing 0.3 or 0.4 M of Ni(II) at pH 4.5 and 5. The
increase in Ni(II) actually confirms the expected increase of nickel
percentage at the more negative applied potentials, in spite of the
diminution of citrate concentration. As it was observed also in the
screening step, nickel content increased as solution pH increased
and applied potential decreased. Stabilization of nickel percentage
was attained at the more negative applied potentials, more neg-
ative values than those applied using the conditions tested in the
screening step. This stabilization of nickel percentage would facil-
itate the preparation of constant composition layers when no flat
substrates are used. Deposits prepared from the solutions tested in
this step were also homogeneous, compact and showed a metallic
sheen.
According to the overall results obtained, in the 4 ≤ pH ≤ 5
range, solutions containing [Ni(II)]/[Cu(II)] > 20 and 0.18 M citrate
concentration were useful to obtain the higher Ni percentages,
maintaining the necessary quality of deposits. In order to assure
stability, avoiding precipitation processes we select the solutions
at pH 4.5 to test the preparation of Ni–Cu deposits. This solution
pH allows a wide margin to prepare deposits of less of 100 wt.% of
nickel. These solutions remained stable at least three months after
preparation.
For later application on industrial substrates, these bath com-
positions established from the optimization step, using vitreous
carbon electrode as substrate, were tested on metallic ones. Com-
pact deposits up to dozens of microns thick were obtained on
brass and inox steel. The inox steel electrode was used as test
substrate for the electroforming procedure, taking advantage of
its non-adherence to the deposits. The Ni–Cu deposits can be
easily detached of the inox steel, which facilitated the layer
Fig. 4. Potentiostatic curves of 0.25 M Ni(II) + 0.0125 M Cu(II) + 0.25 M Na3C6H5O7
solution, pH = 5, at ω = 300 rpm, vitreous carbon at different deposition potentials:
(a) −980, (b) −1000, (c) −1020, (d) −1030 and (e) −1050 mV.
Fig. 5. SEM micrograph of a Ni–Cu deposit (28 wt.% Cu) obtained at pH = 5 from
0.25 M Ni(II) + 0.0125 M Cu(II) + 0.25 M Na3C6H5O7 solution. Inset of the figure shows
an AFM detail.
selected stirring conditions lead to deposits of constant compo-
sition, a constant alloy composition throughout deposit thickness
was obtained.
The crystalline structures of Ni–Cu deposits of different thick-
ness and composition were analysed using XRD and compared with
those of Ni deposits obtained from similar baths without Cu(II).
Table 1
Percentage of nickel in Ni–Cu deposits prepared potentiostatically under stirred
conditions (ω = 300 rpm) over vitreous carbon from the 0.25 M Ni(II) + 0.0125 M
Cu(II) + 0.25 M Na3C6H5O7 solution at different pH.
pH = 4
pH = 5
pH = 6
−E (mV)
Ni (wt.%)
−E (mV)
Ni (wt.%)
−E (mV)
Ni (wt.%)
960
980
12.5
42.2
51.6
62.4
65.2
69.2
69.7
950
980
12.4
43.4
53.2
64.9
68.7
71.7
71.9
950
960
970
990
1000
1020
1040
1050
16.4
21.9
28.7
43.4
51.7
67.3
72.8
73.0
1000
1020
1030
1040
1050
1000
1020
1030
1040
1050
All potentials are referred to a Ag/AgCl/1 mol dm−3 NaCl reference electrode.

386
P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
Fig. 6. Three details of the X-ray diffractograms of deposits obtained from 0.25 M Ni(II) + yM Cu(II) + 0.25 M Na3C6H5O7 solution (a) y = 0, pure Ni (black line) and (b) y = 0.0125,
Ni–Cu (33 Cu wt.%) (grey line).
characterization without substrate interference. When the deposits
prepared from the developed baths were detached from the inox,
tensile stress was observed because they were slightly bent.
Saccharine was selected as agent to reduce the tensile stress
of the deposits; its concentration was varied in the 0.3–1 g/dm3
range. Saccharine did not affect the onset of the deposition pro-
cess in voltammetric experiments and no significant variation in
film composition was observed with respect to the corresponding
condition in a saccharine-free bath, but its presence inhibits stress,
detached deposits obtained on planar substrates maintaining pla-
narity. 0.5 g/dm3 saccharine concentration was sufficient to ensure
deposit quality. The addition of saccharine did not modify solution
stability, which remained also stable at least three months after
preparation.
Ni–Cu deposits obtained, a good current efficiency, ranged between
68 and 75%, was obtained (Fig. 9). Current efficiency (Á) has been
evaluated by comparing the charge involved in the deposit produc-
tion (Qeffect), calculated from the real deposit weight, with the total
charge passed (Qflow) during deposition process.
Qeffect
Á =
× 100
(1)
Qflow
3.3. Characterization of Ni–Cu deposits obtained on inox steel
Samples of 5 cm × 2.5 cm were electrodeposited at different
conditions and once prepared Ni–Cu deposits were easily detached
from the substrate due to their low adherence.
Ni–Cu deposits of variable composition were prepared on inox
electrode, from solutions in which the [Ni(II)]/[Cu(II)] was moved
between 24 and 32, applying different and sufficiently negative
potentials. In these conditions shiny, silvery-bright and smooth
Ni–Cu deposits up to 92 wt.% of Ni were obtained. Moreover, for all
3.3.1. Structural characterization
X-ray diffraction was used also to characterise the Ni–Cu
deposits. Pure-nickel electrodeposits obtained in similar conditions
Fig. 7. Potentiostatic curves of 0.40 M Ni(II) + 0.0125 M Cu(II) + 0.18 M Na3C6H5O7
solution, pH = 4.5 at ω = 300 rpm, vitreous carbon at different deposition potentials:
(a) −950, (b) −980, (c) −1010, (d) −1040, (e) −1070, (f) −1100 and (g) −1130 mV.
Fig. 8. Composition of deposits obtained from xM Ni(II) + 0.0125 M Cu(II) + 0.18 M
Na3C6H5O7 solution: (ꢀ) x = 0.30, pH = 4.5, (ꢁ) x = 0.30, pH = 5, (᭹) x = 0.40, pH = 4.5,
(ꢂ) x = 0.40, pH = 5.

P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
387
Table 2
Position of the main diffraction peaks and estimated grain size (D) for electrodeposited Ni and Ni–Cu alloys with low Cu percentage.
1 1 1
2 0 0
3 1 1
2 2 2
2ꢁ (◦)
D (nm)
2ꢁ (◦)
D (nm)
10
9
9
2ꢁ (◦)
D (nm)
2ꢁ (◦)
D (nm)
Electrodeposited Ni
Ni–Cu (14 wt.% Cu)
Ni–Cu (19 wt.% Cu)
44.57
44.34
44.25
19
17
18
51.83
51.60
51.50
92.92
92.52
92.34
8
6
6
98.45
97.90
97.72
8
8
8
when moderate percentages of copper were included in the Ni fcc
lattice.
3.3.2. Mechanical characterization
Measurements of both HV hardness and Young’s modulus of
some of the films prepared revealed that the incorporation of
copper in nickel deposits does not cause any significant varia-
tion in mechanical properties. HV hardness values (Fig. 10A) were
higher than those corresponding to the electrodeposited nickel
(HV = 670 MPa) obtained from a similar bath containing only the
single metal. All them were higher than that of annealed poly-
crystalline bulk Ni (HV = 638 MPa). The high hardness values can
be attributed to the fine grain size that has been obtained by elec-
trodeposition of these series of Ni–Cu alloys. Composition may have
effects on the hardness, but predominantly, a very small grain size
can increase the polycrystalline film hardness as compared to that
of the micron-sized polycrystalline bulk material. The increase in
hardness gives consistency to the film, always beneficial in coating
applications. The deposits prepared showed similar values of the
Young’s modulus to those of electrodeposited Ni (67 GPa) (Fig. 10B)
demonstrating that alloying Ni with moderate percentages of Cu
does not cause notable changes in material elasticity.
3.3.3. Electrical characterization
Fig. 9. Dependence of current efficiency as a function of copper content in the
deposit. Deposits obtained on inox substrate at pH = 4.5 from xM Ni(II) + 0.0125 M
Cu(II) + 0.18 M Na3C6H5O7 + 0.5 g dm−3 saccharine, 0.3 ≤ x ≤ 0.4 M solutions.
Coatings to be used in astronomy instruments may be valuable
if they have a good electrical conductivity. Fig. 11 shows the resis-
tivity of different Ni–Cu alloys and pure Ni deposits. The resistivity
of the electrodeposited nickel (90 nꢂ m) appears to be somewhat
higher than the standard resistivity of bulk nickel (72.5 nꢂ m). This
bulk value is given for well-annealed Ni with micron-sized grains
[44]. Higher electrical resistivity of electrodeposited Ni with respect
to the bulk Ni value has been reported, and found to be depen-
dent of the preparation method [45], more precisely; it has been
attributed to the nanocrystalline grain size of the electrodeposited
Ni [46]. The incorporation of copper atoms in our materials causes
a resistivity increase of the foil samples. Resistivity increases lin-
early with Cu content within the range examined (up to 29 wt.%
using a free-copper nickel bath were also analysed as reference.
Crystalline deposits of fcc structure were obtained as over vitre-
ous carbon electrode. Their grain size (D) was estimated from the
Debye–Sherrer analysis of the main diffraction peaks [43]. Table 2
shows the values obtained from some representative samples. Ni
and Ni–Cu prepared deposits were nanocrystalline, showing a cer-
tain crystalline anisotropy. Values of D in the 8–19 nm range were
obtained. No significant variation in the grain size was observed
Fig. 10. Dependence of (A) microhardness and (B) Young modulus for Ni–Cu deposits as a function of copper content. Deposits obtained on inox substrate at pH = 4.5 from
xM Ni(II) + 0.0125 M Cu(II) + 0.18 M Na3C6H5O7 + 0.5 g dm−3 saccharine, 0.3 ≤ x ≤ 0.35 M solutions.

388
P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
Fig. 11. Dependence of electrical resistivity of Ni–Cu deposits as a function of copper
content, measured by four electrical contacts with van der Paw geometry. Deposits
obtained on inox substrate at pH = 4.5 from xM Ni(II) + 0.0125 M Cu(II) + 0.18 M
Na3C6H5O7 + 0.5 g dm−3 saccharine, 0.3 ≤ x ≤ 0.4 M solutions.
copper). Other samples, with very high Cu, content showed low
electrical resistivity values. This experimental behaviour was to be
expected: alloy resistivities are always higher than the resistivities
of the pure metals. The incorporation of a different element in the
crystalline network of a pure metal hinders the electron mobility
and consequently the electrical resistance increases. All the differ-
ent commercial nickel alloys show an electrical resistivity higher
than that of pure nickel. For the electrodeposited alloys, the small
crystalline grain size of the deposited materials may additionally
contribute to increase their resistivity [46].
Fig. 13. Spectral reflectivity of the pure Ni sample (curve a) and the Ni–Cu alloys
(curve b) 12.4, (curve c) 14.3, (curve d) 17, (curve e) 20.7, (curve f) 21.7 and (curve
g) 29.4 Cu wt.% measured in the visible range. Inset: Total reflectivity of the Ni and
Ni–Cu alloy samples as calculated from the integration of the spectral reflectiv-
ity data in the range 400–750 nm. Deposits obtained on inox substrate at pH = 4.5
from xM Ni(II) + 0.0125 M Cu(II) + 0.18 M Na3C6H5O7 + 0.5 g dm−3 saccharine solu-
tion, 0.3 ≤ x ≤ 0.4 M solutions.
surface roughness was low (profile parameters with 50–80 nm
values for rms and 40–62 nm for Ra were obtained).
Spectral reflectivity measurements, between 450 nm and
750 nm, confirmed that the colour of nickel was not changed signif-
icantly by copper alloying. Pure Ni electrodeposit showed (Fig. 13,
curve a) a spectral reflectivity that is in close agreement with that
of polished nickel metal samples [47]. The reflectivity curves for the
Ni–Cu samples (Fig. 13, curve b–g) showed a very similar shape to
that the obtained for pure nickel sample and slightly lower reflec-
tivity, except for that with 29.4 wt.% Cu for which the loss was
significant. Similar shape means that the hue of these samples is
very similar to that of the pure Ni. The spectral reflectivity decrease
was less pronounced for the short wavelength colours, meaning a
slight shift from yellowish to bluish hue, but this was not appre-
ciable by visual observation. From the reflectivity spectra of each
sample, we can calculate a “total reflectivity” in the visible range by
integrating the spectra over the 450–750 nm range, and normalise
the result to the same result calculated for the 100% reflectivity flat
spectrum obtained with the standard reference samples. This inte-
grated reflectivity (inset in Fig. 13) has been closely compared to
the relative gloss values measured with white light with the gloss-
meter. A good concordance between both kinds of measures was
observed, as it expected.
3.3.4. Optical characterization
Visually, the deposits were silvery-bright. The presence of
copper (<30 wt.%) did not modify qualitatively the colour or the
brightness aspect of the material. The gloss measurements of
deposits were carried out. Fig. 12 shows that by increasing copper
percentage, gloss decreased smoothly. These low differences in
brightness can be explained taking into account the preparation
method. Whatever was the solution used, samples with low
copper percentages were obtained by applying the more negative
potentials, consequently nucleation is favoured, grain size is
diminished and light reflection is enhanced. Nevertheless, the
4. Conclusions
Useful formulations leading electrochemically produced Ni–Cu
deposits with good gloss, reflectivity, mechanical properties and
which could be implemented in an electroforming process have
been developed.
For each separate metal, the study of the influence of both cit-
rate concentration and pH solution put the bases for the design
of a first general formulation ([Ni(II)] = 0.25 M, [Cu(II)] = 0.0125 M,
[citrate] = 0.25 M]). An initial study over vitreous carbon electrode
and different pH values has permitted us to select the codeposition
potential range for each formulation tested and to establish the pH
Fig. 12. Dependence of gloss (in gloss units) of electrodeposited Ni–Cu with copper
percentage. Depositsobtainedon inoxsubstrateat pH = 4.5 fromxMNi(II) + 0.0125 M
Cu(II) + 0.18 M Na3C6H5O7 + 0.5 g dm−3 saccharine, 0.3 ≤ x ≤ 0.4 M solutions.

P. Calleja et al. / Electrochimica Acta 62 (2012) 381–389
389
and applied potential effect in the electrodeposition processes and
to prepare Ni-rich Ni–Cu deposits. Solid solution formation occurs,
which allows one modulating the composition of the deposit as
a function of both bath composition and electrodeposition condi-
tions.
(Eds.), Proceedings of SPIE: Optics for EUV, X-ray and gamma-ray, vol. 6688,
29–30 August, San Diego, CA, 2007, p. 668804, doi:10.1117/12.736063.
[4] R. Banham, G. Valsecchi, A. Valenzuela, Precision Replication by nickel elec-
troforming, in: ASPE Proceedings. ASPE Precision Fabrication and Replication,
October 31–November 5, Monterey, CA, 1999.
[5] H.-M. Zhang, Y.-G. Guo, L.-J. Wan, C.-L. Bai, Chem. Commun. 1 (24) (2003) 3022.
[6] M. Jafarian, F. Forouzandeh, I. Danaee, F. Gobal, M.G. Mahjani, J. Sol. State Elec-
trochem. 13 (8) (2009) 1171.
In a subsequent step, a simple factorial analysis was used to
optimise the bath. Because of the instabilities observed at pH 6
in the screening step, the analysis was restricted to pH values in
the 4–5 range and, in view of the not high enough nickel percent-
ages attained, the nickel (II) concentration was raised. The results
obtained over vitreous carbon electrode were extended to metallic
substrates in order to check for electroforming possibilities. Finally,
stable solutions with [Cu(II)] = 0.0125 M [sodium citrate] = 0.18 M,
[Ni(II)] in the 0.3–0.4 M range and [saccharin] = 0.5 g/dm3 led to
deposits with 10–30 wt.% of Cu and low stress over an inox steel
flat substrate. For each solution, there is a potential range at suf-
ficient negative potential values in which the Ni–Cu composition
remains almost constant. This could facilitate the maintenance of
a constant composition during the plating over a non-planar sub-
strate. Since at these conditions the composition was only slightly
dependent on the applied potential, constant composition of the
deposits is expected for samples with no flat geometrical shapes.
The optical, mechanical and electrical properties of the opti-
mised Ni–Cu deposits make them an alternative to pure nickel for
reflective applications. Ni–Cu alloy with 10 wt.% Cu shows reflec-
tivity similar to that of pure electrodeposited Ni and Ni–Cu with
20 wt.% of Cu shows only a slight decrease. Moreover, the mechan-
ical properties of these alloys improve with respect to those of the
corresponding nickel coating.
[7] J.L. Pinilla, I. Suelves, M.J. Lázaro, R. Moliner, J.M. Palacios, Int. J. Hyd. Energy 35
(18) (2010) 9801.
[8] G. Xiao, C.L. Chien, J. Appl. Phys. 61 (1987) 4061.
[9] J. Toth, L.F. Kiss, E. Toth-Kadar, A. Dinia, V. Pierron-Bohnes, I. Bakonyi, J. Magn.
Magn. Mater. 198–199 (1999) 243.
[10] E. Toth-Kadar, L. Peter, T. Becsei, J. Toth, L. Pogany, T. Tarnoczi, P. Kamasa, I.
Bakonyi, G. Lang, A. Cziraki, W. Schwarzacher, J. Electrochem. Soc. 147 (2000)
3311.
[11] I. Kazeminezhad, W. Schwarzacher, J. Solid State Eletrochem. 8 (2004) 187.
[12] M. Alper, M.C. Baykul, L. Peter, J. Toth, I. Bakonyi, J. Appl. Electrochem. 34 (2004)
841.
[13] D. Bazavan, R. Bazavan, I. Enculescu, E. Matei, C. Necula, L. Ion, S. Antohe, Opto-
electron. Adv. Mater. Rapid Commun. 3 (5) (2009) 484.
[14] W.A. Badawy, K.M. Ismail, A.M. Fathi, J. Appl. Electrochem. 35 (2005) 879.
[15] R.G. Delatorre, M.L. Sartorelli, A.Q. Schervenski, A.A. Pasa, S. Güths, J. Appl. Phys.
93 (2003) 6154.
[16] S. Serrano-Guisan, L. Gravier, M. Abid, J.-Ph. Ansermet, J. Appl. Phys. 99 (8, Pt.
3) (2006), 08T108/1.
[17] W. Glatz, E. Schwyter, L. Durrer, Ch. Hierold, J. Microelectromech. Syst. 18 (3)
(2009) 763.
[18] I. Mizushima, M. Chikazawa, T. Watanabe, J. Electrochem. Soc. 143 (1996)
1978.
[19] E. Pellicer, A. Varea, S. Pané, K.M. Sivaraman, B.J. Nelson, S. Surin˜ach, M.D. Baró,
J. Sort, Surf. Coat. Technol. 205 (2011) 5285.
[20] S.K. Ghosh, G.K. Dey, R.O. Dusane, A.K. Grover, J. Alloys Compd. 426 (2006) 235.
[21] I.Z. Shehata, O.S. Ayob, E.A. Bull, Electrochemistry 22 (3) (2006) 103.
[22] N. Rajasekan, S. Mohan, Trans. Inst. Met. Finish. 89 (2) (2011) 83.
[23] M. Ishikawa, H. Enomoto, M. matsuoka, C. Iwakura, Electrochim. Acta 40 (11)
(1995) 1663.
[24] M. Zhou, N. Myung, X. Chen, K. Rajeshwar, J. Electroanal. Chem. 398 (1995) 5.
[25] P. Bradley, S. Roy, D. Landolt, J. Chem. Soc. Faraday Trans. 92 (20) (1996) 4015.
[26] J.R. Roos, J.P. Celis, C. Buelens, D. Goris, Proc. Metall. 3 (1984) 177.
[27] K. Vu Quang, E. Chassaing, B. Le Viet, J.P. Celis, J.R. Roos, Met. Finish. 10 (1985)
25.
Nevertheless, the high [Ni(II)]/[Cu(II)] ratios supposes a peri-
odic control of the solution to replenish the consumed copper (II),
but this is easily attainable in view of the large number of analysis
methods available for these electroactive metals.
[28] R.Y. Ying, J. Electrochem. Soc. 135 (1988) 2957.
[29] E.J. Podlaha, Ch. Bonhote, D. Landolt, Electrochim. Acta 39 (1994) 2649.
[30] C. Bonhote, D. Landolt, Electrochim. Acta 42 (15) (1997) 2407.
[31] H. Natter, M. Schmelzer, R. Hempelmann, J. Mater. Res. 13 (1998) 1186.
[32] S.K. Ghosh, A.K. Grover, G.K. Dey, M.K. Totlani, Surf. Coat. Technol. 126 (2000)
48.
Acknowledgements
This paper was supported by contract CTQ2010-20726 and con-
tract FUNCOAT CSD2008-0023, from the Ministerio de Ciencia e
Innovación (MICINN) and the FEDER fund. This work has also finan-
cial assistance from the Bilateral Project between Politecnico di
Milano (Italy) and Universitat de Barcelona (Spain) (HI2008-0058).
The authors wish to thank the Centres Científics i Tecnològics –
CCiTUB (UB) for the use of this equipment.
[33] I. Tabkovic, S. Riemer, S. Ming, A.V. Vladyslav, M. Kief, J. Electrochem. Soc. 152
(2005) C851.
[34] A.R. Despic, V.D. Jovic, S. Spaic, J. Electrochem. Soc. 136 (1989) 6.
[35] M. Pushpavam, K. Balakvushuan, J. Appl. Electrochem. 26 (1996) 1065.
[36] E. Beltowska-Lehman, P. Ozga, Electrochim. Acta 43 (5–6) (1998) 617.
[37] T.A. Green, A.E. Russell, S. Roy, J. Electrochem. Soc. 145 (1998) 875.
[38] P.W. Araujo, R.G. Brereton, Trend. Anal. Chem. 15 (1996) 26.
[39] L.J. Van der Pauw, Philips Res. Rep. 13 (1958) 1.
[40] MEDUSA-Chemical Diagrams (2.0) and HYDRA-Hydrochemical Database (2.0)
Softwares of I. Puigdomenech. http://www.kemi.kth.se/medusa/.
[41] Stability constants of metal-ion complexes, in: D. Douglas, Perrin (Eds.), Section
B: Organic Ligands. IUPAC. Chemical Data Series 22, Pergamon Press, Exeter,
1983.
[42] B.H. Priscott, Trans. Inst. Met. Finish. 36 (1959) 1958.
[43] B.D. Cullity, Debye–Sherrer equation, in: Elements of X-ray Diffraction, Addison
Wesley, Reading, MA, 1978, 283 p.
[44] M.J. Laubitz, T. Matsumura, P.J. Kelly, Can. J. Phys. 54 (1976) 92.
[45] ASM Metals Handbook, vol. 2, ASM International, 10th edition, OH, 1990.
[46] A. Cziraki, B. Fogarassy, I. Gerocs, E. Toth-Kadar, I. Bakonyi, J. Mater. Sci. 29
(1994) 4771.
References
[1] W.J. Egle, H. Bulla, P. Kaufmann, B. Aschenbach, H.W. Braeuninger, Opt. Eng. 29
(10) (1990) 1267.
[2] R. Hudec, L. Pina, L. Sveda, A. Inneman, V. Semencova, V. Brozek, M. Mika, M.
Skulinova, R. Kacerovsky, J. Sik, A. Zentko, in: A. Wilson (Ed.), Proceedings of
X-ray Universe 2005 (ESA SP-604), El Escorial, Madrid, 2005, p. 969.
[3] S. Romaine, S. Basso, R.J. Bruni, W. Burkert, O. Citterio, V. Cotroneo, D. Engel-
haupt, M. Freyberg, J. Gorenstein, P. Gubarev, G. Hartner, F. Mazzoleni, S. O’Dell,
G. Pareschi, B.R. Ramsey, C. Speegle, D. Spiga, Development of a prototype nickel
optic for the Constellation-X hard X-ray telescope, in: S.L. O’Dell, G. Pareschi
[47] D.J. Arrowsmith, Gold Bull. 19 (4) (1986) 117.