Impact of key deposition parameters on the morphology of silver foams prepared by dynamic hydrogen template deposition

Article abstract of DOI:10.1016/j.electacta.2010.06.054

The potentiostatic deposition of porous Ag foams from a new stable thiocyanate based bath using hydrogen bubbles as a dynamic template during deposition was investigated. The influence of the electrolyte content, deposition potential and deposition time on the micro- and nanoscale morphology of the Ag form was examined. The formation of three main morphological forms on the nanoscale level: dendrites, a framework of identical particles, and agglomerates of inhomogeneous particles with big Ag granules distributed on the foam surface, was demonstrated by the analysis of the scanning electron microscopy (SEM) data. The quality of the structures obtained was examined by energy dispersive X-ray spectroscopy (EDS) and X-ray diffractometry (XRD). It was found that the experimental parameters had a huge effect on the morphology of Ag on both the micro- and nanoscale. The structures with the most efficient geometry from the technological point of view were obtained with the correct combination of parameters, viz. high concentrations of NH₄ ⁺ and Ag⁺ in conjunction with a sufficient deposition potential and time. Foams with roughness factors as high as 1100 were obtained, showing a high geometrical area normalized double-layer capacitance and relatively high gravimetric capacitance of 22 mF cm^-2 and 2.4 F g^-1, respectively.

Full text of DOI:10.1016/j.electacta.2010.06.054

Electrochimica Acta 55 (2010) 6383–6390
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Impact of key deposition parameters on the morphology of silver foams
prepared by dynamic hydrogen template deposition
∗
Serhiy Cherevko, Chan-Hwa Chung
Advanced Materials and Process Research Center for IT, School of Chemical Engineering, Sungkyunkwan University, Suwon 440-746, Republic of Korea
a r t i c l e i n f o
a b s t r a c t
Article history:
The potentiostatic deposition of porous Ag foams from a new stable thiocyanate based bath using hydro-
gen bubbles as a dynamic template during deposition was investigated. The influence of the electrolyte
content, deposition potential and deposition time on the micro- and nanoscale morphology of the Ag
form was examined. The formation of three main morphological forms on the nanoscale level: dendrites,
a framework of identical particles, and agglomerates of inhomogeneous particles with big Ag granules
distributed on the foam surface, was demonstrated by the analysis of the scanning electron microscopy
Received 30 March 2010
Received in revised form 14 June 2010
Accepted 19 June 2010
Available online 1 July 2010
Keywords:
Silver foam
Supercapacitor
Electrodeposition
Hydrogen
(
(
SEM) data. The quality of the structures obtained was examined by energy dispersive X-ray spectroscopy
EDS) and X-ray diffractometry (XRD). It was found that the experimental parameters had a huge effect
on the morphology of Ag on both the micro- and nanoscale. The structures with the most efficient geom-
etry from the technological point of view were obtained with the correct combination of parameters,
+
+
viz. high concentrations of NH4 and Ag in conjunction with a sufficient deposition potential and time.
Foams with roughness factors as high as 1100 were obtained, showing a high geometrical area normal-
−
2
−1
,
ized double-layer capacitance and relatively high gravimetric capacitance of 22 mF cm and 2.4 F g
respectively.
©
2010 Elsevier Ltd. All rights reserved.
1
. Introduction
only 1.5% of the available surface area was utilized under certain
conditions [7]. It is clear that improving the electrode geometry is
a critical step in the development of porous materials for use as
electrodes in electrocatalysis. It has been suggested that an ideal
porous electrode should have micro- and nanoscale-features. A
non-uniform microscopic porosity in the micrometer range, where
the pore sizes of the electrode decrease with increasing distance
from the front surface to the inner space, as well as porosity
on the nanoscale which brings about an increase in the surface
area and electrode activity is needed. Recently, such electrodes
were produced by applying simple electrochemical techniques. It
was found that electrodeposition at very high overpotentials in
aqueous solutions results in the formation of porous foams with
walls composed of a huge concentration of dendrites. With this
method, highly porous Ni electrodes with microscopic morpholo-
gies appropriate for cathodes in water electrolysis were produced
[8]. More recently, Shin et al. proposed the electrodeposition of
self-supported 3D foams of copper, tin and copper-tin alloy by
using the so-called hydrogen template [9,10]. It was shown while
the hydrogen bubbles, generated at the electrode, are attached to
the electrode, deposition can occur within the interstitial spaces
between the bubbles, thus replicating them. Due to the constant
evolution of hydrogen, the bubbles continue to grow and, at some
point, detach from the electrode. The size of such bubbles increases
with time, leading to the formation of non-uniform porous struc-
tures. This technique was also applied to the deposition of porous
Due to their high surface area, porous and especially nanoporous
materials have attracted a great deal of attention over the last
decade. This interest comes from the great impact of such mate-
rials in electrocatalysis. Besides the increase in the reaction rate
brought about by their huge surface area, nanostructured porous
electrodes can also have a significant effect on the kinetics of a
reaction, which is usually related to the high activity of the nanos-
tructures and large amount of active sites [1,2]. However, the
rate of an electrochemical reaction is not always proportional to
the surface area of the electrode. Indeed, as shown by numerous
researchers [3–6], the efficient utilization of highly porous elec-
trodes can only be expected when the reaction is sluggish. For
fast, diffusion controlled reactions, usually only a small increase in
the reaction rate is observed. Even for kinetically controlled reac-
tions, an inefficient electrode geometry leads to additional ohmic
and bubble overvoltages, mass transfer limitations, and, due to the
possible hydrophobicity, a decrease of the electrolyte accessible
surface area. All of these effects decrease the efficiency when uti-
lizing porous electrodes. As an example, it was found that for the
Raney–Nickel electrode used for the hydrogen evolution reaction,
∗ Corresponding author. Tel.: +82 31 290 7260; fax: +82 31 290 7272.
E-mail address: chchung@skku.edu (C.-H. Chung).
0
013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.06.054

6
384
S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
Table 1
silica [11,12], but the main research was focused on the depo-
sition of porous copper. The mechanism of deposition and the
influence of the experimental parameters on the copper morphol-
ogy were investigated. Nikolic et al. [13–17] studied the effect
of such parameters on the formation of copper dendrites, foams,
and honeycomb and dish-like structures. Also, it was shown that
the mechanical strength and branch sizes of the dendrites of the
Composition of electrolytic solutions for deposition of porous Ag.
Reagents
Concentration, M
Ag2SO4
KSCN
NH4Cl
C6H5Na3O7·2H2O
0.01–0.06
1.5
0–2.5
0–0.03
−
copper foam can be improved by using additives such as Cl ,
NH4+, acetic acid, polyethylene glycol, 3-mercapto-1-propane sul-
fonic acid, and cetyltrimethylammonium bromide [18–20]. More
recently, the influence of the deposition regime (constant or pul-
sating overpotential) on the morphology of copper honeycomb-like
structures was presented [21,22].
Ag SO , 1.5 M KSCN, 0.5 M NH Cl, and 0.01 M C H5Na O7·2H O,
2
4
4
6
3
2
which we call the “standard bath” and any modifications of this
“standard bath” used in this work will be highlighted in the text.
The deposition was performed at room temperature from an
unstirred solution. The deposition potential was varied in the range
E = −1 to −4 V vs. Ag/AgCl. The deposition time was in the range of
t = 15–120 s. After the deposition, the electrodes with deposited Ag
foams were washed with DI water and dried under nitrogen gas
flow.
In our recent communication, we applied the concept of the
hydrogen template for porous silver deposition [23]. It was shown
that the morphology of the silver electrodes can vary with the
amount of NH4+, which was used to increase the hydrogen evo-
lution rate. The aim of the current work is to study the effect of the
deposition parameters (deposition potential and deposition time)
and electrolyte content on the morphology of the porous silver. The
final goal is to find a trend in the optimization of conditions for the
production of silver electrodes most suitable for electrochemical
applications, in areas where Ag electrodes were already employed,
such as alkaline and direct borohydride fuel cells [24–26], double
layer supercapacitors [27,28], batteries [29], sensors [30] or elec-
trodes for electrocatalysis [31–33]. This is accomplished by analysis
of the electrode morphology and by measuring the electrolyte
accessible surface area.
2.4. Roughness factor evaluation
All of the electrochemical experiments were carried out at room
temperature in a standard three-electrode system. For the mea-
surement of the surface area, two solutions were used: 1 M sodium
hydroxide and0.1 Mpotassiumfluoride aqueouselectrolytes. How-
ever, it was found that due to the presence of a large amount of
highly active sites, the typical voltammetric profile of the silver
foam in sodium hydroxide solution contains numerous Faradaic
features which make it difficult to estimate the surface area. Thus,
for the estimation of the surface area, only the data obtained from
the CV taken in the potassium fluoride electrolyte was analyzed.
Prior to the voltammetric experiments, the electrolytes were deaer-
ated by purging with pure nitrogen gas for 20 min and, during the
measurements, a stream of nitrogen gas was passed over the solu-
tion.
2
. Experimental
2.1. Reagents
Potassium thiocyanate, ammonium chloride, sodium cit-
rate, potassium fluoride, sodium hydroxide were obtained from
Sigma–Aldrich. Silver sulfate was purchased from Samchun Pure
Chemicals (Korea). All of the solutions were prepared in deionized
3. Results and discussion
(
DI) water. All chemicals were used as received.
3.1. Structure characterization
2.2. Instruments
Due to the high rate of hydrogen evolution on a Pt substrate
in the presence of a large amount of NH4+ ions in a thiocyanate
based Ag bath, porous silver foams can be deposited galvanostat-
ically, as described in our previous report [23]. However, as was
pointed out, the high cathodic currents needed to obtain a suf-
ficiently high hydrogen evolution rate result in the precipitation
of sulfur or sulfur complexes. To eliminate such undesired effects,
a small amount of sodium citrate was added to the solution in
the present study. We found that by adding sodium citrate in the
concentration range from 0.01 M to 0.03 M, the formation of precip-
itates could be avoided without any detrimental effect on the foam
morphology. Thus, the electrolyte used in our previous work was
modified by adding sodium citrate at a concentration of 0.01 M.
The mechanism of bath stabilization is not clear. We found that
the addition of sodium citrate does not change the CV profile of
Ag deposition/stripping on the Pt electrode. We believe, that the
sodium citrate stabilizes the byproducts of Ag deposition, which
would otherwise precipitate.
Before moving to the analysis of the influence of the experi-
mental conditions on the microporous structure of the deposits,
we would like to focus in detail on the evaluation of the nanoscopic
features. During the electron microscopy investigation, we found
that the nanoscale morphologies of all of the electrodeposited Ag
foams can be categorized into three basic groups, namely dendrites,
a framework of identical particles, and agglomerates of inhomo-
geneous particles with big Ag granules distributed on the foam
The electrochemical deposition and electrochemical measure-
ments were performed using an electrochemical workstation
(
Zahner® Elektrik IM6ex, Germany). Ag/AgCl and Hg/HgO refer-
ence electrodes were utilized for the electrodeposition and surface
area measurements, respectively, and a Pt plate was used as the
counter electrode. The chemical composition was measured by
energy dispersive X-ray analysis (EDS) performed in a field emis-
sion scanning electron microscope (FESEM) (JEOL JSM-7000F) with
which the morphology of the deposited films was also examined.
The structural properties were examined by X-ray diffractometry
(
XRD) (Bruker AXS with Cu K␣ radiation at 40 kV and 40 mA).
2
.3. Electrode preparation
The method of synthesizing the porous Ag foam presented
herein is similar to the one reported in our previous works [23,34].
Briefly, Ag foams were deposited on a Pt/Ti/Si 5 mm × 25 mm elec-
trode. A 10 nm thick Ti layer served as an adhesion layer and a
2
00 nm thick Pt thin film was deposited onto a Si(1 0 0) oriented
wafer using electron beam evaporation. To prepare the working
electrode, a 2 mm radius circle on the Pt surface was isolated and
the edges and back side of the electrode were covered by insulat-
ing varnish. The deposition was carried out in constant potential
mode. The electrolytes used for the deposition of the Ag foams are
summarized in Table 1. Hereafter, the electrolyte containing 0.01 M

S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
6385
Fig. 1. SEM images of three typical morphologies on microlevel: (a) dendrites, (b) small powder, (c) mixed powder obtained by deposition: from “standard bath” over (a)
5 s and (c) 90 s, and (b) from bath containing 0.75 M of NH4Cl. (d) Typical EDS spectra and (e(C)) XRD pattern taken from electrodeposited Ag foam. For comparison, the XRD
pattern taken from the substrate (e(A)) and standard XRD data for poly crystalline Ag JCPDS-004-0783 (e(B)) are also shown.
1
surface, as shown in Fig. 1(a–c), respectively. The growth of den-
3.2. Influence of deposition parameters on the microstructure of
Ag foams
drites was observed only in the initial stages of foam formation,
+
that is at a small deposition time, at a low concentration of NH4
,
and at a low overpotential. Under such deposition conditions, the
amount of hydrogen gas was not enough to produce significant stir-
ring of the solution layer close to the electrode. Thus, the deposition
of Ag was primarily controlled by diffusion rather than kinetics.
This result is in good agreement with the well known fact that the
growth of Ag dendrites usually happens as a result of diffusion con-
As shown in a previous study [23], varying the amount of NH +
in the Ag electrolyte plays a significant role in the formation of
4
the porous Ag foams. However, the foams formed were relatively
thin, which made it difficult to obtain high surface area electrodes.
Adjusting the other experimental parameters is, therefore, neces-
sary to produce highly efficient electrodes for electrocatalysis. In
+
trolled deposition. By increasing the H evolution rate, the decrease
the following sections, the influence of the NH4 concentration,
2
in the thickness of the diffusion layer makes the mass-transfer of
deposition potential, deposition time, and the amount Ag+ ions on
the microstructure of the film are described.
+
Ag ions faster, leading to the formation of compact particles. This
likely compresses the dendrites and results in the formation of
denser structures. Finally, in the kinetics controlled reaction regime
.2.1. Influence of NH4+ concentration
As in the case of the galvanostatic deposition [23], in the poten-
3
(
mainly at protrusive structures of deposited foams) the deposition
of large agglomeration becomes possible.
tiostatic deposition at E = −3 V and t = 30 s from the “standard bath”
where NH4+ concentration was varied from 0 to 2.5 M, the mor-
phology of the Ag foam varied drastically depending on the amount
of NH4+ in the electrolyte, as shown in Fig. 2. Under such deposi-
tion conditions, foams with ordered pores were only obtained at
ammonium chloride concentrations of 0.5 M and higher. For the
chosen deposition time, the formation of porous films by galvano-
static deposition also starts at this concentration. The tendency of
the pore size to decrease with increasing NH4+ concentration can
be clearly seen, as presented by the solid line in Fig. 3.
Due to the decrease of the pore size, the number of pores over
a defined area increases, as shown by the dotted line in Fig. 3.
The increase in the number of surface pores, however, eventu-
ally reaches saturation as a result of the simultaneous decrease in
the pore wall thickness (or interpore distance) and hole size. The
average pore size is in the range of 10 ␮m to 20 ␮m which is two
To ensure that the electrodeposited foams did not contain unde-
sirable contaminants, EDS analysis was performed. The typical EDS
spectra clearly showing the absence of impurities are shown in
Fig. 1(d). Further confirmation of the silver quality was obtained by
X-ray diffractometry (Fig. 1(e(C))). Except for the XRD peaks from
the Pt/Ti/Si substrate (Fig. 1(e(A))), the XRD pattern obtained from
the Ag foam contains only the peaks from pure Ag (fcc) phase. The
position of the peaks matches well with the standard XRD pattern
taken from polycrystalline Ag (JCPDS 004-0783) (Fig. 1(e(B))). As
was found from SEM analysis, the size of the dendrite branches and
the particle dimensions vary significantly with the deposition time
or deposition potential. Surprisingly, it was found that the mean
crystallite size for all of the electrodeposited foams does not vary
greatly and was equal to 22.8 ± 1.7 nm. This result is based on the
investigation of twelve samples with different microscopic mor-
phologies deposited under different plating conditions. The mean
crystallite size was estimated from the full width at half maximum
height (FWHM) of the (1 1 1), (2 0 0), and (2 2 0) peaks according to
the Debye–Scherrer equation [35]. Such a result is to be expected
when taking into account the fact that both the big particles and
dendrite branches are composed of smaller grains.
times smaller than that found for Ag foams grown by galvanostatic
−2
deposition at j = 1 A cm
[23]. The current during potentiostatic
deposition increased with increasing NH4+ concentration, as shown
in Fig. S1(a) in the Supporting Materials. The current–time tran-
sients are very noisy because of the disturbances brought about by
the change in the surface area and the change in hydrodynamic con-

6
386
S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
Fig. 2. SEM images of Ag structures deposited from electrolytes containing (a) 0.1 M, (b) 0.25 M, (c) 0.5 M, (d) 0.75 M, (e) 1 M, (f) 1.5 M, (g) 2 M, and (h) 2 M NH4Cl.
ditions due to the evolution of hydrogen. The effect of the increase
of the deposition current on the concentration of ammonium chlo-
ride is complex and consists of a few regions with different slopes,
as shown in Fig. S1(b) in the Supporting Materials. Here, the average
current density jav calculated as
It was shown that the increase of the density and the decrease of
the surface tension of the solution leads to a decrease in the hydro-
gen bubble break-off diameter, which results in the formation of
smaller holes [15]. However, we believe that the observed change
−
2
in pore size is controlled by the current variation (from 0.7 A cm
−
2
+
ꢀ
to 1.2 A cm for 0 and 2.5 M of NH4 , respectively), as the contribu-
tion of the current density to the variation of the hole size is much
larger than that of the solution density and surface tension. On the
nanoscale level, the morphology changes from dendrites to small
particles with simultaneous increase in the foam wall compactness,
as NH4+ concentration is increased.
t
1
t
jav =
j(t)dt
(1)
0
was used [15]. A significant increase in the current was observed
+
for NH4 concentrations of 0.5 M and higher which correlates with
the porous structures formation. As we showed in a previous study
[
23], the Ag deposition efficiency at very high overpotentials is very
3
.2.2. Influence of deposition potential
Fig. 4 shows the evolution of Ag foam morphology obtained
low, which means that the increase in current is almost entirely
due to the rise in the hydrogen evolution rate. As was extensively
described in previous works [15,17,23], a high rate of hydrogen
evolution results in the change of the hydrodynamic conditions in
the solution layers near the electrode, thus decreasing the diffusion
layer thickness and increasing the rate of transfer of metal ions to
the electrode.
from the “standard bath” at various deposition potentials for t = 60 s.
Pores are obtained for the potential is E ≤ −2 V. With increasing
deposition potential, the morphology of the foams, in the terms
proposed by Nikolic et al. [15,17], changes from honeycomb-like
to a combined honeycomb- and dish-like structures. The popu-
lation of the holes for the dish-like foams is lower than that for
the honeycomb-like structures. Also, it can be seen that the evo-
lution of the pore multilayers slows down for E < −3 V. The foam
deposited at higher voltages consists of very compact deposits and
has a relatively small amount of holes. It is clear that increasing
the overpotential results in an increase of the hydrogen evolu-
tion, as the current density was observed to constantly increase
(
Fig. S2(a and b) in the Supporting Materials). Due to the compli-
+
cated nature of the electrodeposition (competitive Ag reduction
and H2 evolution reactions), it is difficult to explain the observed
result. However, taking into account the high efficiency of H2 gen-
eration, we can say that the current–potential profile is governed
mainly by the hydrogen evolution rate. Under such experimental
conditions, the formation of dish-like structures was unexpected.
Increasing the current density should result in the decrease of
the break-off diameter of the hydrogen bubbles and the forma-
tion of smaller holes [15]. The opposite effect was observed as
it is evident that the formation of dish-like holes was caused
by the formation of H2 with a larger break-off diameter at the
prolonged times needed for their detachment from the electrode
surface.
Fig. 3. The dependences of the surface pore diameter (solid line) and number of
surface pores over an area of 0.09 mm (dotted line) on the concentration of NH4Cl.
2

S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
6387
Fig. 4. SEM images of Ag foams deposited from the “standard bath” at applied potentials E of (a) −1.5 V, (b) −2 V, (c) −2.5 V, (d) −3 V, (e) −3.5 V, and (f) −4 V vs. Ag/AgCl. The
deposition time was t = 60 s.
On the nanoscale level small overpotentials result in the growth
of dendrites because of diffusion controlled growth (low solution
mixing rate). Due to the increase in the hydrogen evolution rate and,
thus, the increase in the mass transfer rate of Ag to the electrode
at higher potentials, the nanoscale structure changes to a compact
particle-like one with uniform particles at moderate overpoten-
tials and mixed particles at high overpotentials. Considering their
possible electrochemical applications, we believe that the foams
deposited at E = −3 V should have the most efficient reagent acces-
sible geometry.
shown by two typical examples for films deposited for 30 s and
90 s, are shown in the inset pictures of the corresponding images,
respectively. As shown in Fig. 1(a–c), the former one consists of
dendrites and the latter one of non-uniform particles. The differ-
ence in their density is also evident. With increasing deposition
time, there is also a significant change in the microstructure. For
deposition times shorter than 60 s, pore multilayer structures are
formed, whose surface pore size increases with increasing depo-
sition time. When the deposition time is longer than 60 s, there is
also a change in the wall thickness. At this point the wall thick-
ness of a pore is comparable to its diameter. The honeycomb-like
structures formed within a short time are modified to a structure
in which the holes are covered on the edges. It should be noted that
without taking the first few seconds of deposition into account the
current profile does not change significantly over time, as shown
in Fig. S3 in the Supporting Materials. However, it is obvious that
with increasing deposition time, the amount of available Pt sites
drastically diminishes. This can result in the decrease of the H2
evolution efficiency and the hydrogen bubbles can be enclosed by
Ag.
3.2.3. Influence of deposition time
Shown in Fig. 5 are the Ag foams electrodeposited from the
“
standard bath” at an applied potential of E = −3 V and different
deposition times in the range of 15–120 s. As seen in Fig. 5(a), even
for deposition times over 15 s, a thin porous foam of Ag is formed
on the electrode surface.
The foam walls consist of relatively sparse structures. The
tendency in the morphology change on the nanoscale, and the dif-
ference in the density of the Ag structure in the wall region, as
Fig. 5. SEM images of Ag foams deposited from the “standard bath” at an applied potential E = −3 V with deposition times of (a) 15 s, (b) 30 s, (c) 45 s, (d) 60 s, (e) 90 s, and (f)
20 s.
1

6
388
S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
Fig. 6. SEM images of Ag foams deposited from the electrolytes containing NH4+ (a–c) 0.5 M and (d–f) 2.5 M, and Ag+ (a and d) 0.01 M, (b and e) 0.03 M, and (c and f) 0.06 M.
The deposition potential and time were −3 V and 30 s, respectively.
It should be noted that such a change in morphology is undesir-
able, as it can inhibit the transfer of the electroactive species to the
inner layers of the electrode. Thus, improvement of the electrode
morphology (increase of the active area) cannot be obtained by
simply prolonging the deposition time.
3.3. Roughness factor evaluation
It is obvious that because of their microporous and nanoscopic
structure, the surface area of metal foams prepared by dynamic
hydrogen template deposition is significantly higher than the
corresponding geometrical area of the electrode. So far, we con-
centrated mostly on the investigation of the microstructure by
electron microscopy analysis. Three different morphologies on the
nanoscale and various morphological forms on the micro-scale
were found. Nonetheless, the electrochemical analysis of the real
surface area or electrochemically accessible area is no less impor-
tant. The electrochemically accessible area is significant in that this
is the area that will likely be utilized in most electrochemical appli-
cations of such metal foams. Moreover, for some applications, such
as double layer capacitors or electrocatalysis with sluggish kinetics,
the improvement of device performance is directly proportional to
the area of the electrode.
3
.2.4. Influence of Ag+ amount
Lastly, the effect of the amount of silver salt was investigated.
+
It is believed that increasing the concentration of Ag can decrease
the efficiency of hydrogen evolution [16] and, thus, increase the
rate of silver deposition. It should be noted that the solubility
of Ag SO is very poor and strongly depends on the amount of
2
4
KSCN. At a KSCN concentration of 1.5 M, the highest concentra-
tion of soluble Ag was less than 0.1 M. The dependence of the foam
+
morphology on the amount of Ag was studied in the electrolytes
+
with concentrations of NH4 0.5 M and 2.5 M. The corresponding
SEM images are presented in Fig. 6(a–c) and (d–f), respectively.
At a low rate of hydrogen evolution (low concentration of NH4 ),
It is convenient to use the roughness factor, Rf, the ratio of the
real active surface area to the geometrical area of the electrode, as a
quantitative parameter of the electrode surface area. In this study,
we used a method of evaluating the roughness factor based on the
measurement of the double layer capacitance of the Ag foam in KF
solution. Even though there are other methods [36] and the current
method has some limitations related to the oxide electrodes [37],
the relatively broad so-called double-layer region on Ag electrodes
makes it possible to measure the surface roughness of the Ag foams.
Another advantage of this method is that it is a non-destructive
process. We want to emphasize that, due to the presence of a large
amount of active sites, some Faradaic features in the double-layer
region were observed when the NaOH electrolyte was utilized. On
+
the rise in the deposition rate with increasing amount of silver
ions is evident. Due to the coalescence of the hydrogen bubbles
during deposition, each new layer in the foam structure con-
sists of large pores. However, at a concentration of 0.06 M, the
hydrogen evolution rate is insufficient, resulting in sparse struc-
tures. It is likely that the deposition rate is limited by the rate
of transfer of Ag ions, which is lower at a low hydrogen evolu-
tion rate. The metal in this case tends to grow on the pore walls,
making them thicker, instead of building them up to form new
layers.
Foams with completely different morphologies were formed
when the amount of ammonium chloride was sufficiently high, as
can be seen in Fig. 6(d and e). Even though the efficiency of hydro-
gen evolution probably decreases at higher concentrations of silver,
it is still very high (the current in this case is significantly higher, as
shown in Fig. S4(a and b) in the Supporting Materials for the depo-
−
the other hand, it is likely that the specific adsorption of F ions
decreases the amount of such Faradaic processes.
The voltammetric profile in 0.1 M KF solution was undoubtedly
capacitive. Fig. 7(a) shows the CVs taken from the Ag foam electrode
+
deposited from the “standard bath” with 0.03 M Ag and 2.5 M of
+
NH4+ at different scan rates, as indicated in the figure. Evidence of
the capacitive nature of the current is provided by the linear pro-
portionality between the scan rate and current measured at the
center of the potential window, i.e. −0.25 V vs. Ag/AgCl, as shown
in Fig. 7(b). The variation of the roughness factor of the Ag foams
with the experimental parameters is presented in Fig. 8. The corre-
sition from the electrolyte with NH4 concentrations of 0.5 M and
2
.5 M, respectively), which allows for high rates of Ag transfer and
deposition. Films grown under such conditions have a pronounced
D structure with increasing pore size diameter toward the outer
3
surface. Thus, the formation of a layered structure with a highly
efficient layered morphology can be achieved.

S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
6389
−
1
Fig. 7. (a) Typical cyclic voltammetry data for the Ag foam in 0.1 M KF solution at scan rates of 10–160 mV s and (b) corresponding dependence of current j0.25 on the scan
rate.
sponding CVs are presented in Fig. S5 in the Supporting Materials.
The evolution of the surface area with the amount of Ag+ in the
electrolyte was found to be different for the two electrolytes with
The R was calculated from the CV data using the formula:
f
+
different concentrations of NH4 . It can be seen in Fig. 8(d) that,
Cexp
+
at a low concentration of Ag , the deposited foams have similar R
Rf =
(2)
(3)
f
Cspec
values. However, the situation changes drastically when the Ag+
concentration is 0.045 M and 0.06 M. In the latter case, the differ-
ence in the roughness factor is approximately a factor of two (∼600
and ∼1100).
j0.25
Cexp =
dE/dt
where Cexp is the experimentally obtained capacitance normal-
ized to the geometrical surface area, j0.25 = (janodic + jcathodic)/2 is
the current density at E = 0.25 V vs. Ag/AgCl, and Cspec is the
specific capacitance of the smooth silver/electrolyte interface
It was shown by Brevnov [27,28] that electrodeposited high
surface area silver films can be used as double-layer capacitors
with a high-frequency response. Therefore, besides the capacitance
normalized to the geometric surface area, it would be worth-
while to estimate the gravimetric capacitance. To do so, the Ag
foam electrodeposited from an electrolyte containing 0.045 M Ag
and 2.5 M NH4+ was weighed. It was found that the weight of
−
6
−2
(
Cspec = 20 × 10 F cm ) [38]. It can be seen that the increase in
the Ag deposition rate in all cases results in the deposition of films
with higher surface areas. Nonetheless, the effect of the deposition
parameters is different in each case.
−
2
the foam was around 5.6 mg cm . Thus, the gravimetric sur-
2
−1
As could be predicted from the SEM data, the Ag foams
electrodeposited from the electrolyte in the presence of a high
face area is around 11.8 m g
and the gravimetric capacitance
−
1
is 2.4 F g . The geometric surface normalized capacitance was
+
−2
concentration of NH4 have a higher surface area than the ones
approximately 13.2 mF cm . The Ag foam deposited from the
+
obtained at a low concentration of ammonia chloride, as shown in
electrolyte with an [Ag ] of 0.06 M had the highest value of the
geometric surface normalized capacitance of around 22 mF cm
+
−2
Fig. 8(a). The highest obtained R was about 180 at an NH4 concen-
.
f
tration of 2.0 M where saturation takes place. Fig. 8(b) shows the
The electrode composed of interconnected silver particles with
a size of around 200 nm described by Brevnov and Olson [28]
dependence of R on the deposition potential. It can be seen that
f
−
2
−1
the R increases almost linearly and reaches values of around 300
had corresponding capacitances of 1.7 ± 0.2 mF cm and 0.7 F g
,
f
at −4 V. The similar linear proportionality of R with the deposi-
respectively. On the other hand, using porous silver films with par-
f
tion time is shown in Fig. 8(c) with a maximum at 120 s of ∼370.
ticles 30 ± 7 nm resulted in the improvement of the capacitances
Fig. 8. Evolution of roughness factor Rf of Ag foam on the (a) NH4+ concentration, (b) time, (c) potential, and (d) Ag+ ions amount in solution with NH4+ of 0.5 M (solid line)
and 2.5 M (dotted line).

6
390
S. Cherevko, C.-H. Chung / Electrochimica Acta 55 (2010) 6383–6390
to 2.9 ± 0.1 mF cm−2 and 3.9 ± 0.1 F g , respectively [27]. It is clear
that because of the multilayer structure and very high surface area
the capacitance normalized by the geometric area obtained in our
work is significantly higher than the previously reported values.
Similarly, the value of the gravimetric capacitance obtained was
higher than that of the film of 200 nm Ag particles and lower than
that of the film of 30 ± 7 nm Ag particles obtained by Brevnov et
al. Our result is in good agreement with the works of Brevnov, as
the Ag foams consist of particles with sizes smaller than 200 nm,
but larger than 30 nm. Also, it is worth noting that the voltammet-
ric profile of a typical Ag foam is capacitive in a potential window
of at least 0.9 V, as shown in Fig. S6 in the Supporting Materials.
It is evident that to improve the gravimetric capacitance, the fea-
ture sizes of the particles and the sizes of the micropores should
be decreased, which could probably be achieved by using additives
and surfactants [18–20].
−1
Acknowledgements
This work was supported by New & Renewable Energy R&D
program (2009-T100100282) under the Ministry of Knowledge
Economy, Republic of Korea
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.electacta.2010.06.054.
References
[
[
1] A. Ahern, L. Nagle, D. Burke, J. Solid State Electrochem. 6 (2002) 451.
2] L.D. Burke, J.A. Collins, M.A. Horgan, L.M. Hurley, A.P. O’Mullane, Electrochim.
Acta 45 (2000) 4127.
[3] Y. Bai, Y. Sun, C. Sun, Biosens. Bioelectron. 24 (2008) 579.
[
[
[
[
[
4] S. Cherevko, C.-H. Chung, Sens. Actuators B: Chem. 142 (2009) 216.
5] M. DeLeo, A. Kuhn, P. Ugo, Electroanalysis 19 (2007) 227.
6] S. Park, T.D. Chung, H.C. Kim, Anal. Chem. 75 (2003) 3046.
7] K. Lohrberg, P. Kohl, Electrochim. Acta 29 (1984) 1557.
8] C.A. Marozzi, A.C. Chialvo, Electrochim. Acta 45 (2000) 2111.
4
. Conclusions
The morphology of the Ag foam showed strong dependence on
the electrolyte composition and deposition parameters. The NH4+
concentration was crucial for the formation of porous structures.
Without NH4+ and at NH4+ concentrations lower than 0.25 M, the
deposits consisted of relatively uniformly distributed sparse den-
[9] H.-C. Shin, J. Dong, M. Liu, Adv. Mater. 15 (2003) 1610.
10] H.C. Shin, M. Liu, Adv. Funct. Mater. 15 (2005) 582.
[
[11] W.-Z. Jia, K. Wang, Z.-J. Zhu, H.-T. Song, X.-H. Xia, Langmuir 23 (2007)
11896.
[
12] S. Yang, W.-Z. Jia, Q.-Y. Qian, Y.-G. Zhou, X.-H. Xia, Anal. Chem. 81 (2009)
478.
3
+
drites. At [NH4 ] of 0.25 M and 0.5 M, the structure was no longer
[13] N. Nikolic, K. Popov, L. Pavlovic, M. Pavlovic, J. Solid State Electrochem. 11 (2007)
uniform and changed to honeycomb-like at higher concentrations.
Further increasing the amount of NH4 resulted in the formation of
667.
+
[14] N.D. Nikolic, G. Brankovic, M.G. Pavlovic, K.I. Popov, Electrochem. Commun. 11
(
2009) 421.
compact pores with pore sizes varying from 20 ␮m to 10 ␮m. The
nanoscopic features changed from dendrites to relatively small uni-
form particles. All of these modifications resulted in the evolution
of the roughness factor from around 25 to 180. As the deposition
potential was increased, similar changes in the nanoscale mor-
phology were found. However, at relatively high potentials, the
appearance of large agglomerates on the surface of the Ag foams
was observed. The purely honeycomb-like microstructure changed
to a mixed one with some fraction of dish-like holes. The roughness
factor increased with increasing deposition potential to around 300
at −4 V. In the case of the variation of the deposition time, after
relatively long deposition times, the appearance of large agglom-
erates was observed on the surface of the Ag foams, as in the case
of high potentials. However, the change in the microstructure was
completely different from that observed before. A high deposition
time resulted in the trapping of the hydrogen bubbles within the
Ag matrix, which led to the overgrowth of Ag within the edges of
the hydrogen bubbles. Thus, the pore wall thickness increased and
became comparable to the sizes of the holes. The roughness fac-
tor increased gradually to about 370 at deposition time over 120 s.
The most drastic change was found when the amount of Ag was
increased from 0.01 M to 0.06 M, especially with the simultaneous
[
[
[
15] N.D. Nikolic, G. Brankovic, M.G. Pavlovic, K.I. Popov, J. Electroanal. Chem. 621
(2008) 13.
16] N.D. Nikolic, L.J. Pavlovic, M.G. Pavlovic, K.I. Popov, Electrochim. Acta 52 (2007)
8096.
17] N.D. Nikolic, K.I. Popov, L.J. Pavlovic, M.G. Pavlovic, J. Electroanal. Chem. 588
(2006) 88.
18] J.-H. Kim, R.-H. Kim, H.-S. Kwon, Electrochem. Commun. 10 (2008) 1148.
19] Y. Li, W.-Z. Jia, Y.-Y. Song, X.-H. Xia, Chem. Mater. 19 (2007) 5758.
20] H.-C. Shin, M. Liu, Chem. Mater. 16 (2004) 546.
[
[
[
[21] N. Nikolic, G. Brankovic, V. Maksimovic, M. Pavlovic, K. Popov, J. Solid State
Electrochem. 14 (2009) 331.
[
22] N.D. Nikolic, G. Brankovic, V.M. Maksimovic, M.G. Pavlovic, K.I. Popov, J. Elec-
troanal. Chem. 635 (2009) 111.
[23] S. Cherevko, X. Xing, C.-H. Chung, Electrochem. Commun. 12 (2010) 467.
[24] F. Bidault, A. Kucernak, J. Power Sources 195 (2010) 2549.
[
25] E. SanlI, H. C¸ elikkan, B. Zühtü Uysal, M.L. Aksu, Int. J. Hydrogen Energy 31 (2006)
920.
26] N. Wagner, M. Schulze, E. Gülzow, J. Power Sources 127 (2004) 264.
1
[
[27] D.A. Brevnov, J. Electrochem. Soc. 153 (2006) C249.
[
[
[
28] D.A. Brevnov, T.S. Olson, Electrochim. Acta 51 (2006) 1172.
29] X. Jin, J. Lu, Y. Xia, P. Liu, H. Tong, J. Power Sources 102 (2001) 124.
30] W. Lian, L. Wang, Y. Song, H. Yuan, S. Zhao, P. Li, L. Chen, Electrochim. Acta 54
(2009) 4334.
31] J. Geng, Y. Bi, G. Lu, Electrochem. Commun. 11 (2009) 1255.
32] M.-C. Tsai, D.-X. Zhuang, P.-Y. Chen, Electrochim. Acta 55 (2010) 1019.
33] F.-H. Yeh, C.-C. Tai, J.-F. Huang, I.W. Sun, J. Phys. Chem. B 110 (2006) 5215.
[
[
[
[34] S. Cherevko, C.-H. Chung, Talanta 80 (2010) 1371.
[
[
35] A.L. Patterson, Phys. Rev. 56 (1939) 978.
36] A.J. Motheo, S.A.S. Machado, M.H.V. Kampen, J.R. Santos, J. Braz. Chem. Soc. 4
+
increase in the concentration of NH4 from 0.5 M to 2.5 M. Even
(
1993) 122.
+
for high Ag concentration, when the NH4 concentration is low,
[37] S. Trasatti, O.A. Petrii, Pure Appl. Chem. 63 (1991) 711.
[
38] A. Bard, L. Faulkner, Electrochemical Methods: Fundamentals and Applications,
John Willey, New York, 2001, p. 233.
the film is sparse. Completely different morphological forms were
observed at high concentration Ag and NH4+
+
.