Electrochemical investigation of electrodeposited Fe-Pd alloy thin films

Article abstract of DOI:10.1016/j.electacta.2010.09.022

In the present study, the electrodeposition of Fe, Pd and Fe-Pd alloys, in alkaline solutions, has been investigated. Using ammonium hydroxide and trisodium citrate as the complexing agents, it has been shown that the co-deposition of Fe and Pd is achieved due to diminishing the difference between the reduction potentials of these two metals. Cyclic voltammetry results clearly show that the electrodeposition processes are diffusion-controlled and the diffusion coefficients of Fe²⁺ and Pd²⁺ are 1.11 × 10^-6 and 2.19 × 10^-5 cm² s ^-1, respectively. The step potential experiments reveal that nucleation mechanism is instantaneous with a typical three-dimensional (3D) growth. At low overpotentials, addition of Pd²⁺ to Fe²⁺ solution leads to a dramatic reduction in the number of nucleation sites, due to this fact that at such overpotentials, the electrodeposition behavior of Pd²⁺ governs on the overall process. The analysis of chemical composition of the electrodeposited films and the number of nucleation sites indicate that at higher overpotential, Fe²⁺ is deposited preferentially, thus the electrodeposition of iron-palladium alloys was classified as an anomalous co-deposition.

Full text of DOI:10.1016/j.electacta.2010.09.022

Electrochimica Acta 56 (2010) 483–490
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Electrochemical investigation of electrodeposited Fe–Pd alloy thin films
M. Rezaei, M. Ghorbani∗, A. Dolati
Department of Materials Science and Engineering, Sharif University of Technology, Azadi Ave., P.O. Box 11155-9466, Tehran, Iran
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 10 July 2010
Received in revised form 6 September 2010
Accepted 7 September 2010
Available online 16 September 2010
In the present study, the electrodeposition of Fe, Pd and Fe–Pd alloys, in alkaline solutions, has been
investigated. Using ammonium hydroxide and trisodium citrate as the complexing agents, it has been
shown that the co-deposition of Fe and Pd is achieved due to diminishing the difference between the
reduction potentials of these two metals. Cyclic voltammetry results clearly show that the electrodepo-
sition processes are diffusion-controlled and the diffusion coefficients of Fe2+ and Pd2+ are 1.11 × 10−6
and 2.19 × 10−5 cm2 s−1, respectively. The step potential experiments reveal that nucleation mechanism
is instantaneous with a typical three-dimensional (3D) growth. At low overpotentials, addition of Pd2+
to Fe2+ solution leads to a dramatic reduction in the number of nucleation sites, due to this fact that at
such overpotentials, the electrodeposition behavior of Pd2+ governs on the overall process. The analysis
of chemical composition of the electrodeposited films and the number of nucleation sites indicate that at
higher overpotential, Fe2+ is deposited preferentially, thus the electrodeposition of iron–palladium alloys
was classified as an anomalous co-deposition.
Keywords:
Fe–Pd alloy
Electrodeposition
Kinetic
Diffusion-controlled process
Nucleation and growth mechanism
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
sition process from electrochemical point of view has been less
attended.
Fe–Pd thin films have many potential applications due to
their specific chemical composition. Shape memory properties at
alloys including 28–31 at.% Pd [1–6] and high magnetic anisotropy
at about 50 at.% Pd [7–13] make them as promising materi-
als for microelectromechanical systems (MEMS). Moreover, invar
anomalies (30–35 at.% Pd) [14,15] and catalytic properties of
palladium-rich thin films (including high hydrogen absorption
capacity [16–18] and dechlorination of organic compounds [19,20])
can be considered as reasons of increased attention to the Fe–Pd
thin films.
Different methods have been proposed to synthesize Fe–Pd
thin films; including electron-beam evaporation and sputter-
ing [15,21,22], melt spinning [5,23], arc melting [3–5,12] and
electrodeposition [6–9,24]. Among these techniques the electrode-
position is more desirable both economically and practically [8,9].
There are some reports on the electrodeposition of Fe–Pd alloys
in ammonium citrate [8,24], ammonium tartrate [6,9] and sulfos-
alicylic acid solutions [7,16]. Although all of the electrolytes are
effective to co-deposition of Fe–Pd alloys, it is difficult to achieve
a wide range of chemical composition of the thin layers with-
out varying the electrolyte composition [8,16]. In addition, most
studies on electrodeposition of Fe–Pd alloys have been dedicated
to investigation of their physical properties whereas their depo-
Characterization of alloy electrodeposition is useful to recognize
the relation between thin films properties and the mechanism of
electrodeposition [25,26]. Therefore, the need of a comprehensive
study with a focus on the mechanism of Fe–Pd alloy electrodepo-
sition seems to be essential.
The aim of present study is to develop a stable electrolyte to
achieve a wide range of Fe–Pd alloy thin films. Moreover, cyclic
voltammetry and chronoamperometry techniques were applied to
study the mechanism of the electrodeposition process. The effect
of cathodic potential on overpotential deposition (OPD) and also on
the number of nucleation sites has been investigated.
2. Experimental
2.1. Materials
Electrolytes were prepared using analytical grade of
FeSO4·7H2O and PdCl2 as the parent metal ions with the dif-
ferent concentrations ratio. NH4OH and Na3C6H5O7·2H2O were
used as the complexing agents for palladium and iron ions,
respectively. H3BO3 and (NH4)2SO4 were utilized as additive and
supporting electrolyte. The chemical compositions of the solutions
are listed in Table 1. The pH of the electrolytes was adjusted to 8
using ammonia solution. At such levels of pH, hydrogen evolution
on the cathode surface is not an issue of concern. The working
electrode was a copper wire with surface area about 1 cm2 and
roughness less than 20 nm. The counter electrode was a 316L
∗
Corresponding author. Tel.: +98 21 66165219; fax: +98 21 66005717.
E-mail address: ghorbani@sharif.edu (M. Ghorbani).
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.09.022

484
M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
Table 1
Chemical composition of the solutions (mmol/dm3).
Solution
FeSO4·7H2O
PdCl2
H3BO3
(NH4)2SO4
Na3C6H5O7·2H2O
NH4OH
1
2
3
4
5
6
7
8
9
200
200
–
–
–
–
–
200
200
200
–
–
–
250
–
250
250
250
250
250
250
250
–
600
–
600
600
600
600
600
600
600
–
300
–
300
300
300
300
300
300
300
–
900
–
300
450
750
900
900
900
900
20
20
20
20
20
10
20
40
10
Regarding the value of ˇ4 = 1030.5 [28], if whole Pd2+ ions are
present in the form of [Pd(NH3)4]2+ complexes, the value of E20 will
be −0.24 V vs. SCE.
stainless steel plate with surface area about 8 times larger. A
saturated calomel electrode (SCE, +0.241 V vs. SHE) was used as
reference electrode.
Addition of NH4OH into the solution leads to an increase in pH.
Under this condition, the possibility of Fe precipitation and/or the
formation of Fe(OH)+ enhances significantly. Such these reactions
can be hindered using citrate ions to form iron citrate complex [8]
and boric acid to reduce the (OH)− concentration near the surface
[29], resulting in stability of Fe2+ ions.
2.2. Methods and analysis
All experiments were performed in
a deaerated three-
electrode electrochemical cell without agitation using a potentio-
stat/galvanostat Autolab® model PGSTAT302N equipped by Nova
1.5 software. The reduction regions of Fe2+ and Pd2+ ions at the
different concentrations of the complexing agents were deter-
mined applying linear sweep voltammetry (LSV). In addition, cyclic
voltammetry was employed to identify the type of reduction pro-
cess. The scan rate variety was in the range of 20–150 mV s−1. The
potentiostatic j–t transients were recorded every 100 mV in the
range of −800 to −1100 mV for a pulse time of 10 s. The films were
deposited for 10 min and then removed from the substrate by a 2:1
mixture of HCl and HNO3 solutions as the acidic solvent. The con-
centrations of Fe and Pd in the obtained solutions were performed
using atomic absorption spectroscopy (AAS) (model Perkin Elmer
Instruments, Analyst 800).
For the evaluation of Fe2+ and Pd2+ reduction process, the lin-
ear sweep voltammetry curves of the electrolytes number 1 and 3
(without any complexing agents) and the electrolytes 2, 4–7 (with
additives and different concentrations of ammonium hydroxide)
are depicted in Fig. 1(a) and (b), respectively. Based on the obtained
results, complexing agents decrease the reduction potential of both
Fe (from −0.11 to −0.27 V vs. SCE) and Pd (from +0.64 to −0.25 V
vs. SCE); and at the same time, the difference between reduction
potential of these two metals shifts to a lower value. The thermody-
namically determined values of reduction start potential for Pd2+
and [Pd(NH3)4]2+ show acceptable agreements with the empirical
results (Table 2). According to Nernst equation, it is expected that
the reduction of Fe2+ in solution (1) (200 mM Fe2+) start at −0.70 V
vs. SCE, while based on the results of Fig. 1(a), the start point occurs
at about −0.11 V vs. SCE. This difference can be related to the under-
potential deposition (UPD) of Fe which is a common phenomenon
in the deposition process of iron group metals on a noble substrate
[30,31]. For instance, Mallett et al. [32] have observed the UPD pro-
cess of Fe2+ on platinum substrate. In addition, Leistner et al. [33]
have shown that the underpotential deposition of a 100 mM Fe2+
solution on copper substrate starts at about E = −0.1 V vs. SCE.
The same calculation procedure can be employed in the case of
reduction start potential of iron citrate complex using Eqs. (6)–(9):
3. Results and discussion
3.1. Thermodynamics of Fe and Pd co-deposition
As iron is more active than palladium, for co-deposition of Fe2+
and Pd2+, it is thermodynamically essential to make near the reduc-
tion potential of these two ions. In order to achieve such these
conditions, addition of the given complexing agents and/or chang-
ing of Fe2+ and Pd2+ ions concentrations seems to be effective
methods [8].
It has been shown that the difference between the stan-
dard reduction potential of Pd2+/Pd (+0.95 V vs. SHE) and Fe2+/Fe
(−0.44 V vs. SHE) can be reduced, if [Pd(NH3)4]2+ complex form of
Pd is present in the medium [27]. It is due to the lower standard
reduction potential of [Pd(NH3)4]2+ according to:
Fe2+(aq) + Cit(aq) ⇒ [Fe–Cit]2+(aq) ꢀG10 = −RT ln K
[Fe–Cit]2+(aq) + 2e− ⇒ Fe(s) + Cit(aq) ꢁG20 = −nFE20
Fe2+(aq) + 2e− ⇒ Fe(s) ꢀG30 = −nFE30
(6)
(7)
(8)
RT
E20 = E30
−
ln K
(9)
Pd2+(aq) + 4NH3 ⇒ [Pd(NH3)4]2+(aq) ꢀG10 = −RT ln ˇ4
[Pd(NH3)4]2+(aq) + 2e− ⇒ Pd(s) + 4NH3 ꢀG20 = −nFE20
Pd2+(aq) + 2e− ⇒ Pd(s) ꢀG30 = −nFE30
(1)
(2)
(3)
nF
Assuming that E30 = −0.11 V vs. SCE (according to Fig. 1(a)) and
K = 104.4 [34], the value of E20 is estimated as −0.24 V vs. SCE. This
value is close to that of presented in Table 2.
Fig. 1(b) indicates that electrodeposition of Fe2+ from solution
(2) and Pd2+ from solution (7) initiates from nearly close potential.
As it was mentioned before, addition of ammonium hydroxides
leads to an increase in [Pd(NH3)4]2+ concentration and conse-
quently, this event resulted in a decrease in the both palladium
reduction potential and deposition rate (Fig. 1(b), curves 1–4). As a
consequence, an overlapping in the reduction regions of Fe (curve
4) and Pd (curve 5) is occurred. Such a solution can be potentially
introduced as an effective electrolyte solution for Fe–Pd alloy elec-
trodeposition.
where ˇ4 is the constant stability of [Pd(NH3)4]2+ complex, F is the
Faraday’s constant and Eq. (3) is the sum of Eqs. (1) and (2). Using
Nernst equation for 20 mM concentration of Pd2+, the value of E30
is calculated +0.66 V vs. SCE. Accordingly, the value of E20 can be
determined using following equations:
ꢀG10 + ꢀG20 = ꢀG30
(4)
(5)
RT
E20 = E30
−
ln ˇ4
nF

M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
485
Table 2
The value of reduction start potential vs. SCE/V.
Electroactive species
Fe2+ (sol. 1)
Pd2+ (sol. 3)
[Fe–Cit]2+ (sol. 2)
[Pd(NH3)4]2+ (sol. 7)
E0 (calculated value)
E0 (empirical value)
−0.70
−0.11
+0.66
+0.64
−0.24
−0.27
−0.24
−0.25
The films were potentiostatically electrodeposited in a solution
(9) which contains both species of solutions (2) and (7). The effect
of applied cathodic potential on Fe and Pd contents of the films is
shown in Fig. 2. The Fe content of the deposits increases from 12.1
to 93.2 at.% when the potential changes from −650 to −1200 mV.
It can be related to underpotential co-deposition of Fe–Pd same
as what occurs in Fe–Pt co-deposition [33]. Also, the other reason
can be the drop in concentration of Pd2+ near the electrode surface
due to long time of 10 min or anomalous co-deposition of Fe–Pd
same as Fe–Ni co-deposition [25]. It should be mentioned that the
underpotential co-deposition only in low overpotentials (less than
−700 mV vs. SCE) can occur. Since in whole range of applied poten-
tials, the Fe content of the films (as less noble metal) increases by
an increase in overpotential and, also, the analysis of the number
of nucleation sites shows anomalous nucleation (Section 3.3), this
phenomenon cannot be only due to diffusion of Pd2+. Because the
nucleation process occurs only in the initial times when there is no
diffusion process. Therefore, it is more probable and rational that
the coupling of anomalous co-deposition together with diffusion
process is responsible for the increasing of Fe content due to an
increasing in overpotential.
a
1) Fe Sol. (1)
3.2. Kinetics of electrodeposition
40
2) Pd Sol. (3)
Fig. 3(a) and (b) illustrates the reduction behavior of Fe and Pd
at various scan rates from electrolytes 2 and 7. In Fig. 3(a), there is a
single wide peak in Fe voltammograms. It is specific property of the
Fe voltammetric curves when the deposition is diffusion-controlled
[25]. In Fig. 3(b), two peaks (a and b) for Pd electrodeposition
is observed. Peak (b) completely disappears when the scan rate
increases from 20 to 150 mV s−1. This behavior shows that two
serial reactions may have occurred [35]. Since the electrodeposi-
tion of some metals such as Ni and Pt (platinum group metals),
which shows similar behaviors like Pd, obeys two-step reduction
mechanism [29,36,37], it may be related to the two-step reduction
of Pd2+ as follows:
30
20
10
Pd2+ + e− ⇒ Pd+(slow)
Pd+ + e− ⇒ Pd(fast)
(10)
(11)
0
This two-step reaction during the palladium electrodeposition has
been already suggested by Kravtsov [38]; the first reaction is con-
sidered as the rate determining step, while the second reaction is a
fast reaction [37].
700 500 300 100 -100 -300 -500 -700 -900 -1100 -1300 -1500
E vs. SCE / mV
b
1) Pd Sol. (4)
2) Pd Sol. (5)
3) Pd Sol. (6)
4) Pd Sol. (7)
5) Fe Sol. (2)
100
40
Solution(9)
90
80
70
30
Fe at%
Pd at%
60
50
40
30
20
10
0
20
10
0
600
700
800
900
1000
1100
1200
1300
-100
-300
-500
-700
-900
-1100
-1300
E vs. SCE / mV
-E vs. SCE / mV
Fig. 1. Linear sweep voltammetry curves for electrodeposition of Fe and Pd: (a)
without complexing agents and (b) with complexing agents, scan rate = 2 mV s−1
Fig. 2. Influence of cathodic potential on the Fe and Pd contents of Fe–Pd alloy;
deposition times 10 min.
.

486
M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
-0.012
-0.025
-0.02
-0.015
-0.01
-0.005
0
a
b
Pd-Solution(7)
Fe-Solution(2)
-0.009
-0.006
-0.003
0
-0.25
-0.45
-0.65
-0.85
-1.05
-1.25
-0.25
-0.45
-0.65
-0.85
-1.05
-1.25
E vs. SCE / V
E vs. SCE / V
-0.09
d
-0.015
-0.005
0.005
0.015
0.025
c
C Fe
SR=50 mV s-1
SR=50 mV s-1
1) Sol. (8)
C2Pd
C1Pd
-0.07 2) Sol. (9)
3) Sol. (10)
Second zone
-0.05
A2Pd
-0.03
first zone
A Fe
A1Pd
-0.01
0.3
0.1
-0.1
-0.3
-0.5
-0.7
-0.9
-1.1
-1.3
-0.25
-0.45
-0.65
-0.85
-1.05
-1.25
E vs. SCE / V
E vs. SCE / V
Fig. 3. Cyclic voltammograms for deposition of Fe, Pd and Fe–Pd alloys: (a) Fe in solution (2), (b) Pd in solution (7), (c) Fe and Pd at more positive potentials range and (d)
Fe–Pd in solutions (8–10).
At reverse path, for both Fe and Pd voltammograms (Fig. 3(c)),
there are long distances between the reduction peaks (CFe and C2Pd
where Ep, Ep/2, n˛ and ˛ are the peak potential, the potential where
j = jp/2, the number of electrons in the rate determining step and the
charge transfer coefficient, respectively. Charge transfer coefficient
is independent of the concentration of the solution but a function
of potential [39], therefore using the intercept of Fig. 4(a) curves,
these values for iron and palladium can be obtained [40].
Fig. 4(b) shows the linear dependenceofpeakcurrentagainstthe
square root of scan rate. Thus, it can be confirmed that the reduction
of Fe2+ and Pd2+ on copper electrode are irreversible and diffusion-
controlled processes. The diffusion coefficient can be, therefore,
calculated from Randles–Sevick equation as following [39]:
)
and oxidation peaks (AFe and A2Pd), so that the electrodeposition of
Fe and Pd can be considered as the irreversible reactions [35].
In the case of Fe–Pd voltammograms (Fig. 3(d)), two distinct dif-
fusion regions are observed. By an increase in Pd2+ concentration
in solution (from 10 to 40 mM) the reduction current increases at
first zone and the increase in current at second zone is related to
the current increase in first zone. As a consequence, the first peak
in Fig. 3(d) can be related to Pd2+ reduction and the second one cor-
responded to Fe2+. Moreover, the voltammograms of Fe–Pd alloys
show that in the presence of Fe2+ ions, the reduction peak of Pd2+
shifts to less cathodic potentials.
ꢀ
ꢁ
1/2
˛n˛
RT
jp = −0.496nF3/2C∞D1/2ꢂ1/2
(13)
The kinetic parameters of Fe2+ and Pd2+ reduction can be deter-
mined using Nicholson and Shain analysis [39]. Fig. 4(a) shows
the variation of Ep − Ep/2 vs. the S.R.1/2. For an irreversible reac-
tion, charge transfer coefficient can be calculated using Delahay
equation as following [39]:
where n, C∞, D and ꢂ are the number of exchanged electron
involved in the electrode process, the solution bulk concentration,
the diffusion coefficient and the potential scan rate, respectively.
Also, the kinetic constant of the reduction reactions can be deter-
mined using Gokhshtein equation from intercept of curves in
Fig. 4(c) as following [39]:
ꢀ
ꢁ
−1.857RT
−˛n˛F
ꢀ
Ep − Ep/2
=
(12)
jp = −0.227nFC∞K0 exp
(Ep − E0
)
(14)
˛n˛F
RT

M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
487
0.022
a
b
0.3
0.25
0.2
Diffusion coefficient determination
_ Fe
_ Pd
Charge transfer coefficient determination
_
_
_ Fe
_
○
_
_ Pd
○
0.017
0.012
0.007
0.002
Slope: 0.0481
2 = 0.9977
R
Intercept: 0.1636
R2 = 0.9971
Slope: 0.0251
R2 = 0.9903
0.15
0.1
Intercept: 0.0592
R
2 = 0.9973
0.05
0
0.1
0.2
0.3
0.4
0.5
0
0.1
0.2
0.3
0.4
0.5
S.R.1/2 / (V s-1)1/2
S.R.1/2 / (V s-1)1/2
-2.5
c
Kinetic constant determination
-3
-3.5
-4
_
_ Fe
_
_ Pd
○
Intercept: - 10.33
2 = 0.9904
R
Intercept: - 10.482
R
2 = 0.9981
-4.5
-5
-5.5
-6
0.4
0.45
0.5
0.55
0.6
(Epeak-Eo') / V
Fig. 4. Analyzed curves extracted from cyclic voltammograms for deposition of Fe and Pd: (a) Ep − Ep/2 vs. S.R.1/2, (b) −jp vs. S.R.1/2 and (c) ln(−jp) vs. Ep − E0ꢀ solutions (2) and
(7).
The kinetic parameters of Fe2+ and Pd2+ electrodeposition
obtained by voltammetric analysis are listed in Table 3. It should be
pointed out that, based on Eq. (10), as the rate determining step, the
stoichiometric coefficient of electron (e−) is 1. Assuming this value,
Pd2+. These explanations can shown that at higher overpotentials,
iron is deposited with higher rate than palladium.
3.3. Nucleation and growth mechanism
DPd can be calculated and compared with the value obtained from
2+
chronoamperometry analysis (Section 3.3). Additionally, the result
of Table 3 reveals that the Fe2+ charge transfer coefficient (as coef-
ficient of the activation energy) is lower than that of Pd2+ and the
kinetic constant of Fe2+ electrodeposition is higher than that of
Chronoamperometric analysis was applied, in order to study the
mechanism of Fe, Pd and their binary alloy electrodeposition. Fig. 5
shows a series of potentiostatic current transient obtained from
OCP (−0.25 V vs. SCE) to potential of −0.8, −0.9, −1.0 and −1.1 V
Table 3
The kinetic parameters of the Fe and Pd electrodeposition.
Ion
˛
D/cm2 s−1
K0/cm s−1
Voltammetric analysis
Chronoamperometric analysis (Cottrel equation)
jmax, tmax analysis
Fe2+ (sol. 2)
Pd2+ (sol. 7)
0.1460
0.8057
1.11 × 10−6
2.19 × 10−5
1.11 × 10−6
2.27 × 10−5
1.07 × 10−6
2.20 × 10−5
6.7 × 10−4
6.4 × 10−5

488
M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
Table 4
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
a
The effect of overpotential on the number of nucleation sites.
0.024
0.019
0.014
0.009
The number of nucleation sites, (N0/10+6)/cm+2
Cathodic potential vs.
SCE/mV
2
R
= 0.9993
Fe (sol. 2)
Pd (sol. 7)
Fe–Pd (sol. 9)
−800
−900
19.1
26.8
32.2
38.3
1.99
3.81
8.72
11.5
2.21
5.06
10.0
23.5
−1000
−1100
4
0
0.2 0.4 0.6 0.8
1
3
2
1
t-1/2/ s-1/2
vs. SCE. The j–t transients have a normal dependence to overpo-
tential, whereas the current density increases with an increase in
overpotential. Based on the characteristic shape of the chronoam-
perometry curves, it seems that a nucleation and growth process
can describe the electrodeposition process; and the rate of growth
is limited by electroactive species diffusion toward the surface
[25,26,40,41]. Fig. 5 shows that for all step potentials at initial times,
the current increases due to the nucleation and growth process
[25,41]. Afterwards, the current drops by increasing the concentra-
tion gradient of electroactive species on the surface [42], according
to Cottrel equation:
-0.1
0.1
0.3
0.5
0.7
0.9
1.1
1.3
1.5
t / s
0.06
0.05
0.04
0.03
0.02
0.01
0
b
0.025
0.02
2
j = −0.564nFC∞D1/2t−1/2
(15)
R
= 0.9991
The encountering of the nucleation and growth with the dif-
fusion phenomenon has resulted in the creation of a maximum
current at all overpotentials [41]. The critical time (tmax) for the
transient of Fe decreases from 113 to 62 ms, for Pd from 113 to
22 ms and for Fe–Pd from 319 to 65 ms as the potential is made
more cathodic (dash lines).
0.015
0.01
0
0.5
1
1.5
2
t-1/2/ s-1/2
According to the linear plots of Fig. 5 (j vs. t−1/2) at the high
overpotentials, the diffusion coefficients can be calculated from the
sum of least square method using Levenberg–Marquardt algorithm
(LMA). The values of D were 1.11 × 10−6 and 2.27 × 10−5 cm2 s−1
for Fe and Pd, respectively. As seen in Table 3, there is a good
agreement between the calculated values from chronoamperom-
etry and those obtained form voltammetry analysis. On the other
hand, to determine diffusion coefficient by voltammetry analysis, it
was assumed that the controlling step in palladium reduction is Eq.
(10) (n = 1). Since the obtained values of D from chronoamperom-
etry and voltammetry analysis are close together, this assumption
is now reasonable.
The partial differences between the values of D obtained in this
work and those reported in the literature may be related to different
concentration of the applied electrolytes [25,43].
In order to investigate the nucleation type of the deposition, j
vs. t1/2 curves were plotted for Fe, Pd and Fe–Pd alloys at initial
times (Fig. 6). The linear relationship in this region approves the
applicability of the proposed model for instantaneous nucleation
and diffusion-controlled growth [44]:
4
3
2
1
-0.1
0.1
0.3
0.5
0.7
t / s
0.9
1.1
1.3
1.5
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
c
0.037
0.032
0.027
0.022
0.017
2
R
= 0.9991
0
0.2 0.4 0.6 0.8
1
t-1/2/ s-1/2
4
j = −5.012nFD3/2
C
∞ 3/2Vm1/2N0(t − t0)1/2
(16)
3
2
1
where t0 is the representative of the induction period; Vm and N0
are the molar volume and the number of nucleation sites, respec-
tively. The values of N0, obtained from slope of j–t1/2, are listed in
Table 4. The number of nucleation sites for Fe is higher than Pd
which could be related to the lower concentration of Pd2+ in the
electrolyte. By an increase in the applied electrode potential from
−800 to −1100 mV, the Pd nucleation sites rises to 6 times higher
than initial value, while in the case of Fe this value is 2. Therefore,
N0 for Pd electrodeposition is more affected by an increase in over-
potential compared with Fe. According to the results presented in
Table 4, addition of the Pd2+ to Fe2+ electrolyte leads to a dramatic
reduction in the number of nucleation sites, especially at low over-
potentials. A comparison between the values of N0 for solutions (2),
(7) and (9) reveals that the value of N0 for solution 9 (deposition of
Fe–Pd alloy) is between N0 of solution (2) (deposition of pure Fe)
-0.1
0.1
0.3
0.5
0.7
0.9
1.1
1.3
1.5
t / s
Fig. 5. Potentiostatic j–t transients and linear dependence of j–t−1/2 accordance to
Cottrell relation for deposition of: (a) 200 mM Fe (sol. 2), (b) 20 mM Pd (sol. 7) and (c)
200 mM Fe + 20 mM Pd (sol. 9) at the different cathodic potentials: (1) E = −800, (2)
E = −900, (3) E = −1000 and (4) E = −1100 mV vs. SCE. The inset figure corresponds
to E = −1100 mV vs. SCE.

M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
489
1.2
a
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
R2 = 0.9975
2 = 0.9997
-Sol.(2)
1) E=-800 mV vs. SCE
2) E=-900
3) E=-1000
4) E=-1100
Fe
1
0.8
0.6
0.4
0.2
0
R
R
2 = 0.9999
4
3
2
1
E=-1100 mV vs. SCE
□
Fe- Sol. 2
O Pd- Sol. 7
∆ Fe-Pd Sol. 9
Inst. _____
Prog. __ __
R2 = 0.9985
0
0.5
1
1.5
2
2.5
3
0
0.2
0.4
0.6
0.8
1
1.2
1.4
t / tmax
t1/2/ s1/2
Fig. 7. Non-dimensional (j/jmax)
2 vs. t/tm curves for deposition of: Fe (sol. 2), Pd (sol.
7) and Fe–Pd (sol. 9) at E = −1100 mV vs. SCE.
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0
b
-Sol.(7)
1) E=-800 mV vs. SCE
2) E=-900
3) E=-1000
4) E=-1100
Pd
R2 = 0.9985
nucleation pattern of alloy is very similar to that of Pd whereas to
that of Fe at the higher cathodic potentials. This behavior confirms
the anomalous co-deposition of Fe–Pd alloys.
According to the model of diffusion-controlled growth of 3D
hemispherical particles presented by Scharifker and Hills [44], the
deposition transients for progressive and instantaneous nucleation
can be expressed by the non-dimensional equations (17) and (18):
R2 = 0.9995
R2 = 0.9969
4
ꢂ
ꢃ
ꢄꢅ
2
ꢀ
ꢀ
ꢁ
ꢁ
ꢀ
ꢁ
2
2
j
1.2254
t/tm
t
=
1 − exp −2.3367
(17)
(18)
3
2
jmax
tm
prog
2
R2 = 0.9999
ꢆ
ꢀ
ꢁꢇꢈ
2
j
1.9542
t/tm
t
1
=
1 − exp −1.2564
jmax
tm
inst
2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Fig. 7 shows the dimensionless theoretical plots of (j/jmax
) vs.
t1/2/ s1/2
t/tm in both form of instantaneous and progressive nucleation. It
can be seen that the nucleation process of Fe, Pd and Fe–Pd at
E = −1100 mV, obeys an instantaneous nucleation with diffusion
controlled growth. This behavior can be confirmation of the results
obtained through analysis of j–t1/2 in Fig. 6.
Verification of the mechanism of nucleation and growth can be
performed through determination of the diffusion coefficient, and
comparison with previously calculated values. For instantaneous
nucleation, these values calculated by following equation [44]:
0.12
0.1
c
-Sol.(9)
1) E=-800 mV vs. SCE
2) E=-900
FePd
R
2 = 0.9998
3) E=-1000
4) E=-1100
0.08
0.06
0.04
0.02
0
jm2 axtmax
0.1629(nfC)2
D =
(19)
R
2 = 0.9989
4
From Eq. (19), the diffusion coefficients at potential −1100 mV
are calculated 1.07 × 10−6 and 2.20 × 10−5 cm2 s−1 for Fe and Pd,
respectively. A comparison between these diffusion coefficients
and values obtained from the slope of j–t−1/2 can be further evi-
dence for the proposed mechanism (Table 3).
R
2 = 0.9999
3
2
1
R2 = 0.9988
0
0.2
0.4
0.6
t1/2/ s1/2
0.8
1
1.2
1.4
4. Conclusions
In the presence of complexing agents, the reduction potential of
Fe2+ and Pd2+ shift to more negative potentials which is more severe
for Pd2+; therefore Pd2+ was electrodeposited in the range of Fe2+
reduction region. The Fe content of the deposits increased from 12.1
to 93.2 at.% with an increase in potential from −650 to −1200 mV
vs. SCE. The mechanistic analysis of the cyclic voltammetry clearly
showed that the electrodeposition of the Fe, Pd and Fe–Pd alloys
on copper substrates are diffusion-controlled processes and the
reduction of palladium is a two-step reaction. Regarding to j–t tran-
Fig. 6. Potentiostatic j–t1/2 curves for deposition of: (a) Fe (sol. 2), (b) Pd (sol. 7) and
(c) Fe–Pd (sol. 9) at the different cathodic potentials.
and N0 of solution (7) (deposition of pure Pd). It should be noted
that, at the lower cathodic potentials, N0 for the alloy is close to that
of pure Pd (N0 of solution (7)), while at the higher cathodic poten-
tials this value approaches to that of pure Fe (N0 of solution (2)). This
phenomenon can confirm that at the lower cathodic potentials the

490
M. Rezaei et al. / Electrochimica Acta 56 (2010) 483–490
sient analysis, the nucleation mechanism is instantaneous with a
3D hemispherical growth. The number of nucleation sites increased
with an increase in overpotential. By analyzing the number of
nucleation sites, it was found that at low overpotentials, the N0 for
the alloy is close to that of Pd and at high overpotentials as a conse-
quence of the anomalous co-deposition of iron–palladium, the N0
becomes larger and closer to its values for single Fe deposition.
[18] T.B. Flanagan, D. Wang, J. Alloys Compd. 488 (2009) 78.
[19] N. Lingaiah, P.S.S. Prasad, P.K. Rao, L.E. Smart, F.J. Berry, Appl. Catal. A 213 (2001)
196.
[20] N. Lingaiah, P.S.S. Prasad, P.K. Rao, F.J. Berry, L.E. Smart, Catal. Commun. 3 (2002)
397.
[21] K. Sato, Y. Hirotsu, H. Mori, Z.G. Wang, T. Hirayama, J. Appl. Phys. 97 (2005)
024308.
[22] G. Vanderlaan, K.C. Chesnel, M. Belakhovsky, A. Marty, F. Livet, S.P. Collins, E.
Dudzik, A. Haznar, J.P. Attane, Superlattice Microst. 34 (2004) 107.
[23] D. Vokoun, C.T. Hu, J. Alloys Compd. 346 (2002) 153.
[24] D. Pecˇko, K.Z. Rozˇman, P.J. McGuiness, B. Pihlar, S. Kobe, J. Appl. Phys. 107 (9)
(2010), 09A712.
Acknowledgment
[25] A. Afshar, A.G. Dolati, M. Ghorbani, Mater. Chem. Phys. 77 (2003) 358.
[26] A. Dolati, M. Sababi, E. Nouri, M. Ghorbani, Mater. Chem. Phys. 102 (2007) 124.
[27] A.J. Bard, R. Parsons, J. Jordan, Standard Potentials in Aqueous Solutions, Marcel
Dekker, New York, 1985.
We gratefully acknowledge helpful discussion with Dr. D.F.
Haghshenas.
[28] R.N. Goldberg, L.H. Hepler, Chem. Rev. 68 (1968) 229.
[29] N. Zech, D. Landolt, J. Electrochim. Acta 45 (2000) 3471.
[30] L.H. Mendoza-Huizar, J. Robles, M. Palomar-Pardavé, J. Electroanal. Chem. 521
(2002) 106.
[31] K. Ozoemena, T. Nyokong, Electrochim. Acta 47 (2002) 4043.
[32] J.J. Mallett, E.B. Svedberg, S. Sayan, A.J. Shapiro, L. Wielunski, T.E. Madey, W.F.
Egelhoff Jr., T.P. Moffat, Electrochem. Solid-State Lett. 7 (2004) C124.
[33] K. Leistner, S. Oswald, J. Thomas, S. Fähler, H. Schlörb, L. Schultz, Electrochim.
Acta 52 (2006) 199.
[34] E.M. Martell, R.M. Smith, Critical Stabilities Constants, Plenum Press, New York,
1989.
[35] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Fundamentals and Applica-
tions, John Wiley and Sons, New York, 2001.
References
[1] T. Okazaki, T. Ueno, Y. Furuya, M. Spearing, N.W. Hagood, Acta Mater. 52 (2004)
5175.
[2] E.C. Oliver, T. Mori, M.R. Daymond, P.J. Withers, Scr. Mater. 53 (2005) 612.
[3] J. Cui, T.W. Shield, R.D. James, Acta Mater. 52 (2004) 47.
[4] H. Kato, Y. Liang, M. Taya, Scr. Mater. 46 (2002) 475.
[5] C.T. Hu, T. Goryczka, D. Vokoun, Scr. Mater. 50 (2004) 542.
[6] F. Wang, S. Doi, K. Hosoiri, H. Yoshida, T. Kuzushima, M. Sasadaira, T. Watanabe,
Electrochim. Acta 51 (2006) 4254.
[7] F.M. Takata, G. Pattanaik, W.A. Soffa, P.T.A. Sumodjo, G. Zangari, Electrochem.
Commun. 10 (2008) 571.
[8] S.C. Hernandez, B.Y. Yoo, E. Stefanescu, S. Khizroev, N.V. Myung, Electrochim.
Acta 53 (2008) 5627.
[9] S. Doi, F. Wang, K. Hosoiri, T. Watanabe, Mater. Trans. 44 (2003) 652.
[10] T. Mohri, Y. Chen, J. Alloys Compd. 383 (2004) 31.
[11] H. Yabe, R.C. O’Handley, T. Kuji, J. Magn. Magn. Mater. 310 (2007) 2502.
[12] A.R. Deshpande, J.M.K. Wiezorek, J. Magn. Magn. Mater. 270 (2004) 166.
[13] T. Mohri, T. Horiuchi, H. Uzawa, M. Ibaragi, M. Igarashi, F. Abe, J. Alloys Compd.
317–318 (2001) 18.
[14] F. Ono, Y. Chimi, N. Ishikawa, H. Kanamitsu, Y. Matsushima, A. Iwase, T. Kam-
bara, Nucl. Instrum. Methods B 257 (2007) 405.
[15] M. Foos, C. Frantz, M. Gantois, J. Phys. 43 (1982) 394.
[16] K.J. Bryden, J.Y. Ying, J. Electrochem. Soc. 145 (1998) 3339.
[17] W. Zhang, S. Luo, T.B. Flanagan, J. Alloys Compd. 293–295 (1999) 6.
[36] K. Leistner, A. Krause, S. Fähler, H. Schlörb, L. Schultz, Electrochim. Acta 52 (1)
(2006) 170.
[37] C.R.K. Rao, D.C. Trivedi, Coord. Chem. Rev. 249 (2005) 631.
[38] V.I. Kravtsov, Russ. J. Electrochem. 40 (2004) 1286.
[39] R.S. Nicholson, I. Shain, Anal. Chem. 36 (1964) 723.
[40] L. Soleimany, A. Dolati, M. Ghorbani, J. Electroanal. Chem. 645 (2010) 34.
[41] A. Dolati, M. Ghorbani, M.R. Ahmadi, J. Electroanal. Chem. 577 (2005) 8.
[42] D. Pletcher, Southampton Electrochemistry Group, Instrumental Methods in
Electrochemistry, Ellis Horwood, Chichester, 2001.
[43] V.I. Kravtsov, R.K. Astakhova, E.G. Tsventarnyi, O.Y. Kurtova, N.V. Peganova,
Russ. J. Electrochem. 38 (2002) 164.
[44] B. Scharifker, G. Hills, Electrochim. Acta 28 (1983) 879.