Preparation and characterization of a novel nickel-palladium electrode supported by silicon nanowires for direct glucose fuel cell

Article abstract of DOI:10.1016/j.electacta.2012.01.017

Nickel-palladium and silicon nanowires (SiNWs) fabricated by chemical etching and electroless plating (EP) are evaluated as anodes in glucose fuel cells (FC). The effects of processing parameters and structural characteristics such as concentrations of KOH and glucose, medium temperature, and palladium electroless plating on the FC performance are determined. The nickel/silicon electrode exhibits significant activity even without the palladium catalyst and noticeable improvement is observed by incorporating SiNWs. The ordered channels in the SiNWs and coherence rendered by Ni-Pd are important to the enhanced electrode activity. The glucose oxidation current measured from the Ni-Pd/SiNWs electrode is approximately 4.8 mA higher than that on the Ni-Pd/Si one. Our results reveal superior electrooxidation performance and long-term stability and Ni-Pd/SiNWs are useful glucose FC anodes in integrated fuel cell applications.

Full text of DOI:10.1016/j.electacta.2012.01.017

Electrochimica Acta 65 (2012) 149–152
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Preparation and characterization of a novel nickel–palladium electrode
supported by silicon nanowires for direct glucose fuel cell
Bairui Tao , Fengjuan Miao , Paul K. Chub
a,∗
a
a
College of Communications and Electronics Engineering, Qiqihar University, 42 Wenhua Street, Qiqihar, Heilongjiang 161006, China
Department of Physics and Material Sciences, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Nickel–palladium and silicon nanowires (SiNWs) fabricated by chemical etching and electroless plating
Received 26 September 2011
Received in revised form 8 January 2012
Accepted 8 January 2012
(EP) are evaluated as anodes in glucose fuel cells (FC). The effects of processing parameters and structural
characteristics such as concentrations of KOH and glucose, medium temperature, and palladium elec-
troless plating on the FC performance are determined. The nickel/silicon electrode exhibits significant
activity even without the palladium catalyst and noticeable improvement is observed by incorporat-
ing SiNWs. The ordered channels in the SiNWs and coherence rendered by Ni–Pd are important to the
enhanced electrode activity. The glucose oxidation current measured from the Ni–Pd/SiNWs electrode is
approximately 4.8 mA higher than that on the Ni–Pd/Si one. Our results reveal superior electrooxidation
performance and long-term stability and Ni–Pd/SiNWs are useful glucose FC anodes in integrated fuel
cell applications.
Available online 16 January 2012
Keywords:
Fuel cell
Electrode
Ni–Pd/SiNWs
Ordered array
Alkaline solution
© 2012 Elsevier Ltd. All rights reserved.
1
. Introduction
Direct glucose fuel cells (DGFCs) have attracted much attention
metals such as Ag, Au, and Bi [18,19] can be used as the electrode
catalysts and OH ions in the electrolyte can expedite the kinetics
in the ORR.
The main materials (substrate) requirements are large surface
−
as renewable energy sources since glucose is abundant in nature,
non-toxic, nonexplosive, and nonvolatile. Furthermore, they can be
mass produced and have a relatively high energy content (theoret-
area, electrical conductivity, suitable porosity for reactant flux,
and stability during the FC operation. The application of three-
dimensional materials to the design of FC is an active research area.
For instance, silicon nanowires (SiNWs) are a promising substrate
with excellent characteristics such as large surface area, relatively
high mechanical stability, low cost, and compatibility to commer-
cial microfabrication processes and microelectronics technology.
Here, an ordered SiNWs structure is employed as the electrode
backbone and Ni–Pd nanoparticle catalysts are electroless-plated
on the sidewalls of the SiNWs. This ordered porous electrode
exhibits very negative onset potential and strong current response
to glucose even in the long-term oxidation of glucose. The excellent
performance suggests the feasibility of this composite electrode in
nonenzyme glucose catalysis.
−1
ical energy density of 4430 W h kg based on complete oxidation
to CO2 via 24-electron transfer) [1–5]. Hence, extensive research
is conducted on glucose electrooxidation and associated develop-
ment of DGFCs [6–8]. Studies on microbial and enzymatic glucose
fuel cells are also quite prevalent [9,10], but the low current den-
sity, short life-time, poor stability, high cost, and immobilization
of bio-catalysts on the electrodes are critical barriers to commer-
cial applications of DGFCs. Noble metal catalysts may enhance the
long-term stability and biocompatibility. Although Pt electrodes
are widely used in the electrooxidation of glucose, serious poison-
ing effects introduced by absorbed intermediates, low sensitivity,
and poor selectivity act as drawbacks [11]. Hence, identification of
better catalysts is vital to the development of DGFCs. The kinetics of
glucose oxidation on a metallic electrode depends strongly on the
effective surface area [12] and it has also been demonstrated that
in alkaline media, the glucose oxidation reaction (GOR) and oxygen
reduction reaction (ORR) are enhanced compared to those under
acidic and neutral conditions [13–15,8,16,17]. In alkaline DGFCs,
2. Experimental details
A 100-mm single polished ꢀ1 0 0ꢁ silicon wafer with a resistivity
–8 ꢀ cm and thickness of 525 ␮m was used as the substrate to fab-
2
ricate the SiNWs. NH F, NiSO ·6H O, PdCl , (NH ) SO , HF, AgNO ,
4
4
2
2
4
2
4
3
sodium citrate, ammonia, glucose, and other chemical reagents
were all analytical grade and used without further purification. The
solution was prepared with >18 Mꢀ deionized water and all the
∗ Corresponding author. Tel.: +86 452 2742787; fax: +86 452 2738748.
E-mail address: tbr sir@163.com (B. Tao).
◦
experiments were conducted at ∼25 C and 1 atm.
0
013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2012.01.017

1
50
B. Tao et al. / Electrochimica Acta 65 (2012) 149–152
Fig. 1. (a) SEM image of the microstructure of the SiNWs and (b) magnified pictures of the cross-sectional SEM images of the microstructure.
The silicon wafer was cleaned using a standard microelectron-
The surface morphology and structure were assessed by scan-
ning electron microscopy (SEM, JSM 5610) at an accelerating
voltage of 10 kV and the deposited materials were characterized
by XRD (X-ray diffraction) using Cu K␣ radiation. The electro-
chemical properties of the Ni–Pd/SiNWs were determined by
cyclic voltammetry (CV) and chronoamperometric technique. The
electrochemical experiments were performed on a LK3200A elec-
trochemical workstation (Tianjin, China). The modified SiNWs
served as the working electrode and a platinum wire formed the
counter electrode. All the potentials were referenced to the SCE
reference electrode.
ics process and cut into 1 cm × 1 cm pieces. Before fabrication of the
SiNWs, the unpolished side was protected by photoresist and used
for marking the electrode pad. The silicon nanowires were synthe-
sized by chemical etching based on the procedures described in
Refs. [20,21]. A solution [HF (20%): AgNO3 (35 mM) = 1:1 (v/v)] was
used as the etchant and after the formation of SiNWs, the silicon
pieces were removed from the solution and rinsed with DI water
and diluted nitric acid (∼30%) to remove surface residues.
Prior to electrodeposition of the Ni–Pd nanoparticles, the SiNWs
were dipped in a solution of Triton X-100 30 s to decrease the
internal stress and enhance wetting. A thin nickel film was coated
on the SiNWs by electroless plating to serve as the electrically
conductive layer and to obtain better resistance against alkaline
media. More details about the Ni electroless plating process can
be found in Ref. [22]. The Pd nanoparticles were produced on the
sidewall of the SiNWs by electroplating. The Ni/SiNWs sample was
dipped in the electroplating solution [23]. PdCl2 was used as the
metal source and EDTA as the complexing agent. The plating bath
3. Results and discussion
Fig. 1 depicts the top-view (a) and cross-sectional (b) images of
the SiNWs before electroless plating. The SiNWs have a bundled
morphology and are aligned vertically to the substrate. The inter-
face between the SiNWs and bulk silicon is clear and layered. The
length of the SiNWs is uniform and about 90 ␮m, which can be con-
trolled by choosing a different etching time. The diameters of the
SiNWs vary between 60 and 300 nm. The structure has good surface
quality and provides more effective surface area than a planar elec-
trode. Fig. 2(a) shows the surface topography of the cross-section
of the Ni–Pd/SiNWs structure. An additional thin layer of Ni–Pd
nanoparticles can be observed on the SiNWs. The diameter of the Pd
◦
was kept at 60 C and rigorously stirred. The pH value of the bath
was kept at 8.0–9.0 by adding ammonia and the current density
−
2
was 50 mA cm
[22]. The Ni–Pd/SiNWs were then rapid ther-
◦
mal annealed at 400 C in argon. Copper wires were connected to
the silver conductive adhesive for the subsequent electrochemical
experiments.
Fig. 2. (a) Cross-sectional SEM image of the Ni–Pd/SiNWs after electroplating nickel–palladium and (b) X-ray diffraction pattern of the Ni–Pd/SiNWs electrode.

B. Tao et al. / Electrochimica Acta 65 (2012) 149–152
151
Fig. 4. Cyclic voltammograms of the 1 M glucose oxidation reaction on the
Ni–Pd/SiNWs electrode at various concentrations of KOH and the variations of the
anodic peak current for glucose electrooxidation versus KOH concentrations on the
Fig. 3. Cyclic voltammograms of the Ni–Pd/SiNWs (solid curve) electrode and
Ni–Pd/Si (dashed curve) in the absence (inset of Fig. 3) and presence of 1 M glucose
−1
in a 1.0 M KOH solution at a scanning rate of 50 mV s
.
−
1
Ni–Pd/SiNWs electrode in the inset of Fig. 4 at a potential scanning rate of 50 mV s
.
and Ni particles ranges from tens to hundreds of nanometer and the
palladium and nickel particles cover the entire silicon nanowires
even at the bottom. The structure is characterized by X-ray diffrac-
tion (XRD). Fig. 2(b) shows the representative XRD patterns of the
peak current. Furthermore, the peak potential shows a continuous
negative-shift similar to that of glucose oxidation.
At first the glucose oxidation reaction on the Ni–Pd/SiNWs elec-
◦
−
sample annealed at 400 C. Two peaks emerge from the pattern of
trodeis acceleratedat higherOH concentrations. This increasesthe
◦
◦
Ni–Pd/SiNWs at 44.12 and 51.66 . The former is the characteristic
anodic peak current, but at higher KOH concentrations, adsorption
of hydroxyl ions may be dominant on the Ni–Pd/SiNWs elec-
trode and it blocks adsorption of glucose onto the electrode. The
imbalance reduces the anodic peak current [26]. The relationship
between the KOH concentration and peak current in the forward
scan is shown in the inset of Fig. 4.
◦
peak of Ni (1 1 1) with a slight shift and the peak at 51.66 is ascribed
to the Pd (2 0 0) plane. The results confirm the Ni–Pd structure.
The electrocatalytic activity of the Ni–Pd/SiNWs electrode
towards glucose oxidation under alkaline conditions is investi-
gated by cyclic voltammetry and chronoamperometric technique.
Fig. 3 displays the electrochemical response of Ni–Pd/Si (dashed
curve) and Ni–Pd/SiNWs electrodes in a 1 M KOH solution contain-
The CVs of the Ni–Pd/SiNWs electrode in the glucose solutions
with different concentrations from 0.5 to 2.5 M in 1 M KOH solution
are displayed in Fig. 5. The oxidation current increases monotoni-
cally with glucose concentration. Furthermore, the peak potential
shows a continuous positive-shift as the glucose concentration is
increased. It appears that at a lower glucose concentration, the
−
1
ing 1 M glucose measured at a scanning rate of 50 mV s . The cyclic
voltammograms of the Ni–Pd/SiNWs (solid curve) and Ni–Pd/Si
(
dashed curve) electrodes in 1 M KOH in the absence of glucose are
shown in the insets of Fig. 3. The oxidation or reduction peaks can
be observed and the current density of the Ni–Pd/SiNWs is clearly
greater than that of the Ni–Pd/Si implying that the two electrodes
have catalytic capability and the Ni–Pd/SiNWs have higher catalytic
activity. The hydrogen adsorption/desorption peaks are observed at
a negative potential region due to oxidation of hydrogen adsorbed
−
glucose concentration and coverage of OH are independent of
each other. As the glucose concentration is increased, adsorption
of glucose onto the electrode becomes dominant and adsorption of
−
on the metal. In the positive scan, Ni–Pd adsorbs OH and an inac-
tive oxide film is formed on the electrodes. The two reduction peaks
are found at potentials of −0.66 V and 0 V and they arise from reduc-
2
+
3+
2+
tion of Ni or Ni and reduction of Pd , respectively [23–25].
In the presence of 1 M glucose, electrooxidation is character-
ized by two well-defined current peaks in the forward and reverse
scans. An oxidative peak in the anodic scan is mainly attributed to
the electrocatalytic oxidation of glucose by the Pd nanoparticles.
The reoxidation peak is primarily associated with the removal of
incompletely oxidized species formed in the forward scan [26]. The
current densities are higher at the corresponding potentials on the
Ni–Pd/SiNWs electrode compared to the Ni–Pd/Si electrode. The
results suggest that Ni–Pd nanoparticles have superior electrocat-
alytic activity towards glucose oxidation in an alkaline solution, and
the Ni–Pd/SiNWs electrode has much higher catalytic activity for
glucose electrooxidation than Ni–Pd/Si under alkaline conditions.
The effect of different KOH concentrations (0.4–6 M) on the
Ni–Pd/SiNWs electrode in 1 M glucose is investigated by CV. Fig. 4
shows the CVs in 1 M glucose solution from −0.6 to 0.4 V. The oxida-
tion current increases as the KOH concentration is increased from
Fig. 5. Cyclic voltammograms of the glucose oxidation reaction on the Ni–Pd/SiNWs
electrode in 1 M KOH solution containing glucose of various concentrations at a
−
1
0.4 to 1 M and further increase leads to an obvious decrease in the
scanning rate of 50 mV s
.

1
52
B. Tao et al. / Electrochimica Acta 65 (2012) 149–152
oxidation current on the Ni–Pd/SiNWs electrode is evidently higher
than that on the Ni–Pd/Si. The chronoamperogram also shows that
electrocatalytic oxidation of glucose is maintained at high activ-
ity and is very stable for over 600 s. This provides the evidence
that most CO species can be oxidized and removed from the Ni–Pd
catalysts and our results indicate that the Ni–Pd/SiNWs nanocom-
posites have the better electrocatalytic properties.
4. Conclusion
A glucose fuel cell prototype is fabricated by electrodeposi-
tion plating Ni–Pd nanoparticles on ordered aligned Si nanowires
arrays. The Ni–Pd/SiNWs electrode possesses many desirable fea-
tures such as high electrocatalytic activity and good stability due
to the large effective surface area and electrocatalytic activity of
Ni–Pd nanoparticles. The Ni–Pd/SiNWs electrode is attractive in
direct glucose fuel cell.
Acknowledgments
Fig. 6. Cyclic voltammograms of the Ni–Pd/SiNWs in 1 M KOH containing 1 M glu-
−
1
cose at various temperatures at a potential scanning rate of 50 mV s
.
This work was jointly supported by Natural Science Foundation
of Heilongjiang Province (no. F201008 and QC2011C092), Program
for Young Teachers Scientific Research in Qiqihar University (Grant
no. 2010k-Z02 and 2011k-Z01), Excellent Young Scholars of Higher
University of Heilongjiang Province no. 1251G067, Science and
Technology Project of Qiqihar, Grant GYGG2010-03-1 and City Uni-
versity of Hong Kong Research Grant 9360110.
References
[
[
1] S.K. Chaudhuri, D.R. Lovley, Nat. Biotechnol. 21 (2003) 1229–1232.
2] J. Ryu, H.-S. Kimb, H. Thomas, D. Lashmore, Biosens. Bioelectron. 25 (2010)
1
603.
3] S. Kerzenmacher, J. Ducree, R. Zengerle, F. von Stetten, J. Power Sources 182
2008) 66.
4] N. Fujiwara, S. Yamazaki, Z. Siroma, T. Ioroi, H. Senoh, K. Yasuda, Electrochem.
Commun. 11 (2009) 390.
[
[
[
(
5] S. Kerzenmacher, J. Ducree, R. Zengerle, F. von Stetten, J. Power Sources 182
(
2008) 1.
[
[
[
[
6] L.H. Li, W.D. Zhang, J.S. Ye, Electroanalysis 20 (2008) 2212.
7] F. Xiao, F. Zhao, D. Mei, Z. Mo, B. Zeng, Biosens. Bioelectron. 24 (2009) 3481.
8] H.F. Cui, J.S. Ye, X. Liu, W.D. Zhang, F.S. Sheu, Nanotechnology 17 (2006) 2334.
9] V. Soukharev, N. Mano, A. Heller, J. Am. Chem. Soc. 126 (2004) 8368.
Fig. 7. Chronoamperogram of electroactivity of (a) Ni–Pd/SiNWs and (b) Ni–Pd/Si
electrode at an oxidation potential −0.4 V for glucose electrooxidation in 2 M KOH
[
[
10] P. Kavanagh, S. Boland, P. Jenkins, D. Leech, Fuel Cells 9 (2009) 79.
11] Z. Liua, L. Huangb, L. Zhanga, H. Maa, Y. Dinga, Electrochim. Acta 54 (2009)
◦
and 1 M glucose solution at 25 C.
7
286.
12] Y. Chen, W. Schuhmann, A.W. Hassel, Electrochem. Commun. 11 (2009)
036.
13] H.F. Cui, J.S. Ye, X. Liu, D.W. Zhang, F.S. Sheu, Nanotechnology 17 (2006)
334–2339.
[14] J. McGinley, F.N. McHale, P. Hughes, C.N. Reid, A.P. McHale, Biotechnol. Lett. 26
2004) 1771.
15] D. Basu, S. Basu, Electrochim. Acta 55 (2010) 5775.
16] M. Besson, P. Gallezot, Catal. Today 57 (2000) 127.
[17] C.P. Wilde, M. Zhang, J. Chem. Soc. Faraday Trans. 89 (1993) 385.
18] Z.N. Liu, L.H. Huang, L.L. Zhang, H.Y. Ma, Y. Ding, Electrochim. Acta 54 (2009)
286–7293.
19] D. Basu, S. Basu, Electrochim. Acta 56 (2011) 6106–6113.
[20] B.R. Tao, J. Zhang, S.C. Hui, L.J. Wan, Sens. Actuators, B 142 (2009) 298–303.
[21] F.J. Miao, B.R. Tao, L. Sun, T. Liu, J.C. You, L.W. Wang, P.K. Chu, Sens. Actuators,
A 160 (2010) 48–53.
22] J. Shi, P.L. Ci, F. Wang, H. Peng, P.X. Yang, L.W. Wang, Electrochim. Acta 56 (2011)
197–4202.
[23] Y.Y. Liao, T.C. Chou, Electroanalysis 12 (2000) 55.
[
[
2
hydroxyl ions is blocked, thereby leading to reduced anodic peak
current.
2
The effect of the medium temperature on electrooxidation of
glucose is investigated. Fig. 6 depicts the CV of the Ni–Pd/SiNWs
electrode during glucose oxidation at different temperatures. A
higher temperature increases the anodic peak current but the rate
is nonlinear. Some other special characteristics are also observed,
such as a positive shift in the anodic oxidation and re-oxidation
peak potential and a negative shift in the onset potential. As the
temperature is increased, the difference in the onset potentials
becomes smaller and a higher temperature facilitates electrooxi-
dation of glucose.
(
[
[
[
7
[
[
4
The chronoamperometric technique is an effective method to
evaluate the electrocatalytic activity and stability of catalysts.
Fig. 7 shows the chronoamperogram of the electroactivity of the
Ni–Pd/SiNWs (a) and Ni–Pd/Si (b) electrodes at an oxidation poten-
[
[
24] Y.L. Lo, B.J. Hwang, Langmuir 14 (1998) 944.
25] J. Munk, P.A. Christensen, A. Hammnett, E. Skou, J. Electroanal. Chem. 401 (1996)
215.
[26] Z.X. Liang, T.S. Zhao, J.B. Xu, L.D. Zhu, Electrochim. Acta 54 (2009) 2203.
◦
tial –0.1 V in 1 M KOH/1 M glucose at 25 C. As expected, the glucose