Preparation of microporous nickel electrodeposits using a polymer matrix

Article abstract of DOI:10.1016/S0025-5408(99)00103-8

Nickel-mesh electrodes were embedded into polyHIPE (a generic hollow fiber polymer) by immersion in the precursor emulsion and subsequent entrapment into the solid microporous matrix produced by polymerization and drying. Immersing this Ni/polyHIPE/Ni composite into a nickel electroplating bath and passing direct current through the two electrodes resulted in growing Ni electrodeposits on the cathode and through the polymer cells and pores. When the polymer matrix was subsequently burned off, a granular microporous Ni coating was produced on the cathode. Variation of the electroplating time and current density showed that the structure of the Ni coating is determined by the local distortions of the electric field inside the tortuous microporous body of the insulating polymer.

Full text of DOI:10.1016/S0025-5408(99)00103-8

Materials Research Bulletin, Vol. 34, No. 7, pp. 1055–1064, 1999
Copyright © 1999 Elsevier Science Ltd
Printed in the USA. All rights reserved
0025-5408/99/$–see front matter
PII S0025-5408(99)00103-8
PREPARATION OF MICROPOROUS NICKEL ELECTRODEPOSITS USING A
POLYMER MATRIX
I.J. Brown, D. Clift, and S. Sotiropoulos*
School of Chemical, Environmental, and Mining Engineering, Nottingham University,
University Park, Nottingham NG7 2RD, UK
(Refereed)
(Received July 15, 1998; Accepted September 11, 1998)
ABSTRACT
Nickel-mesh electrodes were embedded into polyHIPE (a generic hollow fiber
polymer) by immersion in the precursor emulsion and subsequent entrapment
into the solid microporous matrix produced by polymerization and drying.
Immersing this Ni/polyHIPE/Ni composite into a nickel electroplating bath
and passing direct current through the two electrodes resulted in growing Ni
electrodeposits on the cathode and through the polymer cells and pores. When
the polymer matrix was subsequently burned off, a granular microporous Ni
coating was produced on the cathode. Variation of the electroplating time and
current density showed that the structure of the Ni coating is determined by
the local distortions of the electric field inside the tortuous microporous body
of the insulating polymer. © 1999 Elsevier Science Ltd
KEYWORDS: A. metals, A. microporous materials, C. electrochemical mea-
surements, C. electron microscopy, D. microstructure
INTRODUCTION
The preparation of high surface area Ni electrode materials is relevant to a number of
industrial applications of electrochemical technology. The most important applications in-
clude the use of Ni hydrogen anodes in alkaline or molten carbonate H2/O2 fuel cells [1–4],
*To whom correspondence should be addressed. Tel: ϩ44-115-9514182. Fax: ϩ44-115-9514181.
E-mail: s.sotiropoulos@nottingham.ac.uk.
1055

1056
I.J. BROWN et al.
Vol. 34, No. 7
NiO(OH)/NiO cathodes in nickel–cadmium and nickel–hydrogen batteries [5,6] and hydro-
gen-evolving Ni cathodes in alkaline water electrolysis [7,8]. Nickel foam electrodes have
been successfully used as three-dimensional porous electrodes in the electrolytic recovery of
Cu ions [9] (wastewater treatment) and in electroorganic synthesis [10,11], and Raney nickel
electrodes have been employed in the electrochemical hydrogenation of organics [12,13].
There are various types of high surface area Ni electrodes and many different methods
used for their preparation. Sintered microporous Ni coatings on electrodes are usually
prepared according to the ceramic foil-casting technology by a procedure that involves
mixing of a micrometer-size nickel powder with an organic binder, which is subsequently
thermally decomposed, and further sintering at elevated temperatures in a hydrogen atmo-
sphere [4]. Techniques used in the production of reticulated metal include the incorporation
of the metal into a porous matrix and subsequent matrix decomposition [14] or the use of
foaming agents [15]. Highly porous metal structures can also be obtained by sintering metal
microfibers [6,16,17]. Various methods for the production of nanoporous Ni coatings,
including cold rolling, plasma spraying, annealing, and cathodic codeposition of the Raney
nickel precursor alloys (Ni/Al or Ni/Zn) on a nickel support, are reviewed in ref. 18. Smooth
Raney nickel coatings have also been produced by sherardizing of nickel substrates, i.e., by
the reaction of Ni with Zn vapors at 400°C, which results in the formation of a smooth Ni/Zn
precursor coating [19].
PolyHIPE polymer (PHP) is a microporous material developed at Unilever [20] and Los
Alamos National Laboratories [21] and used since by a number of research groups [22,23].
It is produced through the formation of a high internal phase water-in-oil emulsion, in which
the volume of the aqueous dispersed phase is greater than about 75%, and the subsequent
polymerization (at 60°C) of the oil phase, which contains the monomer styrene (and
occasionally other monomers) and the cross-linker divinyl benzene. This results in the
production of a polymer matrix with an extremely high void (up to 97%) due to the
evaporation of the water droplets, which were present in the precursor emulsion. The
structure of PHP, therefore, is characterized by the presence of numerous cells (1–100 ␮m
FIG. 1
Schematic representation of the electroplating cell comprised of a rectangular Ni-mesh
cathode and a cylindrical Ni-mesh anode, embedded in a polyHIPE polymer body.

Vol. 34, No. 7
MICROPOROUS NICKEL
1057
diameter) interconnected by small pores (0.1–10 ␮m diameter). The size of both the cells and
the pores can be controlled by changing the mixing time, the surfactant (emulsion stabilizer)
concentration, and the monomer composition during the emulsification.
In a recent paper [24], we reported for the first time the successful incorporation of Ni into
the PHP matrix by electroplating through its pores and onto a thin Au-layer electrode pasted
on one side of a polymer sample. Burnout of the polymer resulted in a granular Ni structure
of a very high BET surface area (50 m2gϪ1). Those preliminary experiments were useful in
establishing the viability of the method, but their logical extension to cases of industrial
interest would be the application of the method to the production of porous Ni coatings onto
Ni or stainless steel electrodes. The work presented here introduces a modification of the
technique, whereby the Ni cathode (together with a Ni anode) is immersed into the precursor
emulsion and finally entrapped in the PHP body after polymerization. The Ni-cathode/PHP/
Ni-anode composite cell is then used in a Ni electroplating bath, and Ni is deposited through
the polymer pores onto the cathode. The effect of the electroplating current density and
duration on the resulting Ni coating (after polymer burnout) is also investigated.
EXPERIMENTAL
Preparation of Ni/polyHIPE Polymer/Ni Composite Electrochemical Cells. In this work,
polymer of an 80% void was produced; a typical production batch consisted of 500 ml of
emulsion, prepared by mixing 400 ml of water phase with 100 ml of oil phase. The
composition of the oil phase for the preparation of elastomeric polyHIPE was (by volume):
15% styrene (Aldrich, 99%), 62% 2-ethylhexyl acrylate (Aldrich), 8% divinyl benzene
(Aldrich, 80%), and 15% sorbitan monooeleate (Aldrich, 95%). The aqueous phase was
distilled water and 1% w/w potassium persulfate (Aldrich, 99ϩ%), which acted as the
polymerization initiator. The emulsion was produced in a stainless steel mixing vessel
equipped with a Janke and Kunkel (IKA Labortechnik) RW 20DZM two-blade mixer,
according to the procedure described elsewhere [24]. The emulsion was then poured into
glass vials, and a rectangular Ni-mesh cathode (26 ϫ 26 wires 0.25 mm thick per inch,
Goodfellow Ltd) of typical apparent areas of 1–1.5 cm2, together with a cylindrical piece of
the same mesh serving as the anode, was immersed in the emulsion. The inter-electrode gap
was typically in the 3.5–6 mm range. Two mesh strips protruding from each electrode and
out of the emulsion were subsequently used to make electrical contacts to the electrodes. The
entire assembly was then covered tightly with parafilm to minimize water evaporation during
polymerization, which was performed for 6 h in an oven heated at 60°C. After polymerization
had been completed, the cover was removed and the samples remained in the oven at 60°C
for another 24 h to remove residual water from the pores. The sorbitan monooleate, which
was the surfactant used as an emulsion stabilizer, was removed by washing several times with
boiling isopropyl alcohol. The Ni/PHP/Ni composite cell prepared as a result of this
procedure is depicted in Figure 1, in which typical dimensions are also shown.
Ni Electrodeposition. A standard nickel sulfamate bath [25] served as the electroplating
solution at 60°C: 600 g/l nickel sulfamate (Aldrich, 98%), 10 g/l nickel chloride (Aldrich),
and 40 g/l boric acid (Aldrich, 99.5ϩ%). A regulated 7A-35V DC GW(GPR-3060) labora-
tory power supply was used to carry out the electroplating at various current densities in the
3–300 mA cmϪ2 range. The current densities are reported per apparent area of the Ni mesh
(one face). A 1.6 cm2 true geometric electrode area per 1 cm2 apparent mesh area of the Ni

1058
I.J. BROWN et al.
Vol. 34, No. 7
FIG. 2
SEM micrograph of a cross section of a plain polyHIPE sample.
wire in the cathode mesh (two faces, due to the anode surrounding the cathode) was estimated
from the manufacturer’s specifications and SEM photographs. Hence, all current densities
reported hereafter should be divided by 1.6 to give the true current density. Prior to
electroplating, the Ni electrodes were activated by immersion into a 1:1 HCl–water solution
for a few minutes.
Burnout of the Polymer Matrix and SEM Analysis. The burnout of the polymeric matrix
of the Ni/PHP/Ni composites after electroplating was carried out under air atmosphere in a
preheated furnace at 500°C for 1 h. SEM and EDAX analyses of the resulting material have
confirmed the complete polymer burnout at the end of this procedure [24]. SEM experiments
were carried out with a Hitachi S-570.
RESULTS AND DISCUSSION
Figure 2 shows the SEM micrograph of polyHIPE polymer before plating, in which a regular
structure of closed but porous cells can readily be seen. The polymer structure consisted of
large cells of a 10–30-␮m diameter, corresponding to the water droplets of the initial
water-in-oil emulsion, interconnected by smaller pores of a 3–5-␮m diameter, corresponding
to the contact areas of the initially present water droplets.
Figure 3(a) shows the micrograph of the Ni-mesh cathode prior to its incorporation into the
polymer body and electroplating. Figure 3(b) shows that of the cathode mesh coated with a
Ni electrodeposit following electroplating for 24 h at a rate of 35 mA cmϪ2 in a nickel
sulfamate bath and after the polymer matrix of the resulting composite had been thermally
decomposed. It can be seen that a uniform microporous Ni coating was produced on the Ni
cathode. The electrodeposit appears to have a cauliflower structure organized in near-
spherical interconnected aggregates of 150–50 ␮m diameter, which consist of smaller
spherical deposits. Although the presence of the polymer in contact with the cathode is
certainly responsible for the structure of the microporous deposit (electroplating of plain Ni
mesh at same current densities results in nonporous homogeneous coatings), there is no

Vol. 34, No. 7
MICROPOROUS NICKEL
1059
FIG. 3
SEM micrographs of (a) the bare Ni-mesh wires before electroplating and (b) the Ni coating
on the Ni-mesh cathode after electroplating through polyHIPE and subsequent polymer
burnout. Electroplating at 35 mAcmϪ2 for 24 h.
straightforward matrix–product relationship between the polymer template and the metal
grown through its pores. For the latter to hold, the metal deposit should have been organized
in simple spherical aggregates of 10–30 ␮m (size of the polymer cells) filling the polymer
void completely. Instead, it seems that the distortion of the homogeneity of the electric field
around the cathode substrate, which is caused by the presence of the insulating PHP shield,
gives the Ni deposit its structure. The uneven growth of metal deposits produced by
electroplating through porous materials is a common feature of this method for reticulated
metal production, and it has recently been modeled [14] in terms of uneven current distri-
bution within porous electrodes.
Figure 4(a) and (b), which show SEM micrographs of the same deposit at higher magni-
fications, further support this suggestion. In the center of Figure 4(a), a large porous
aggregate (ca. 140 ␮m diameter) consisting of smaller aggregates (ca. 10–30 ␮m diameter)
can be seen. The larger of the latter aggregates is comprised of small spherical deposits. The

1060
I.J. BROWN et al.
Vol. 34, No. 7
FIG. 4
SEM micrographs of the microporous Ni electrodeposits shown in Figure 3(b).
sequence of events leading to the incorporation of Ni within a single PHP cell could start with
the initial growth of the electrodeposit through the pore closer to the cathode substrate and
on layers of already deposited Ni in contact with the cell. This deposit could start growing
around the pore and into the cell, resulting in one of the smaller spherical structures
mentioned above. At moderate current densities (which prevent more extreme field nonho-
mogeneity), Ni deposits could start entering the cell from neighboring pores further away
from the cathode substrate, thus creating more spherical deposits within the individual
polymer cell. This description of events takes into account only the local characteristics of the
electric field inside a single cell. However, current distribution considerations over the entire
electrodeposit growing through a microporous (i.e., locally structured) matrix lead to the
conclusion that local current densities vary significantly from pore to pore and from cell to
cell, the deposition being favored at locations closer to the facing anode, i.e., at locations that
the deposit has already started to build up. This means that the deposition preferentially
proceeds around already plated pores, rather than around new pores of the same cell. Once
an entire cell is filled with Ni, deposition preferentially starts in a neighboring cell closer to

Vol. 34, No. 7
MICROPOROUS NICKEL
1061
the anode, rather than in one lying at the same plane with the Ni-filled cell. These effects lead
to either partially filled or completely void polymer cells in accordance with Figure 4(a), in
which 10–30-␮m diameter voids can be seen in the body of the central aggregate. In Figure
4(b), much smaller pores of a few microns diameter can be identified between the small
spherical aggregates, corresponding to either partial cell filling or the walls of the polymer
matrix in the precursor Ni/PHP composite (prior to polymer decomposition). Also, the
surface of the aggregates appears to be rough and pitted to a submicron level, probably due
to local high current densities resulting in rough deposits and hydrogen embrittlement. These
properties should be beneficial to the deposit’s catalytic activity, since they increase its true
surface area.
A crude estimate of the porosity of the Ni coating can be calculated as follows. From the
micrographs in Figure 3(a) and (b), the diameter of the coated Ni wire can be seen to be
approximately 750 ␮m (b) and that of the plain wire, 250 ␮m (a). The volume of a
hypothetical cylindrical nonporous deposit, which can be found by subtracting the volume of
the plain wire from that of the coated wire, is 3.93 ϫ 10Ϫ3 cm3/cm of wire. Since there are
10.23 wires per 1 cm along each of the mesh dimensions (mesh specifications: 26 ϫ 26 wires
per inch), it follows that there are 2 ϫ 10.23 ϭ 20.5 wires in 1 cm2 of mesh, amounting to
a total wire length of 20.5 cm/cm2 of mesh. Therefore, a nonporous tubular deposit would
have a volume of 3.93 ϫ 10Ϫ3 ϫ 20.5 ϭ 80.565 ϫ 10Ϫ3 cm3/cm2 of mesh. The charge
passed during Ni electroplating at 35 mA/cm2 of mesh for 24 h corresponds to 3,024 C/cm2
of mesh. The current efficiency as found by weighing the cathode after polymer burnout was
ca. 47% (we comment on this value below), which means that 0.47 ϫ 3,024 ϭ 1,421 C were
used for Ni electrodeposition per 1 cm2 of mesh. From Faraday’s law (Ni molecular weight
and density taken as 58.015 and 8.9 g/cm3, respectively), we can calculate the actual volume
of the Ni deposit to be 48 ϫ 10Ϫ3 cm3/cm2 of mesh. Therefore, the porosity of the deposit
should be approximately (80.565 Ϫ 48)/80.565 ϭ 40.4%. The current efficiencies of ca. 50%
obtained in all of the Ni-plating experiments through polyHIPE were much lower than the
usual 98% values for plating onto bare metal substrates under similar bath, temperature, and
current density conditions [26]. This was due to the fact that the local current densities around
the polymer pores were much higher than the average current density over the entire sample,
resulting in higher rates of hydrogen evolution, competing with nickel deposition and, hence,
lowering the efficiency of the latter.
The micrographs in Figure 5 depict situations of partial coating formation due to smaller
charges passed through the cell (electroplating at shorter times and/or smaller current
densities) and are, therefore, helpful in understanding the development of the microporous
structure of the deposit. Close inspection of Figure 5(a) reveals that the initially small
spherical deposits are connected to each other by needlelike deposits of submicron width.
These very thin deposits, present after only 1 h of electroplating at 33 mA/cm2, could have
been developed through the pores at the walls of neighboring cells, linking small deposits
formed within different cells; as time lapses, both the spherical deposits and the Ni needles
thickened to finally form structures similar to those shown in Figures 3 and 4. The micro-
graphs in Figure 5(b) and (c), which correspond to very small quantities of electrodeposited
Ni, show the first stages of the coating built up where the density of Ni aggregates was small.
In particular, Figure 5(c) reveals that, before the first aggregates started to form, the surface
of the Ni wire was covered by a rough but nonporous thin Ni layer similar to that sometimes
observed on bare metal substrates. It was only after this layer had formed that the deposits
started to squeeze into the polymer matrix and, hence, acquired the porous structure discussed

1062
I.J. BROWN et al.
Vol. 34, No. 7
FIG. 5
SEM micrographs of microporous Ni electrodeposits produced after electroplating at (a) 33
mAcmϪ2 for 1 h, (b) 3.3 mAcmϪ2 for 10 min, and (c) same as in (b) but at a different
location.

Vol. 34, No. 7
MICROPOROUS NICKEL
1063
FIG. 6
SEM micrograph of microporous Ni electrodeposits produced after electroplating at 330
mAcmϪ2 for 10 min.
above. This agrees with the fact that polyHIPE is known to not adhere well on metal
substrates [22].
Finally, Figure 6 shows the structure obtained when electroplating was carried out at the
high current density of 330 mA/cm2 for 10 min. It can be seen that instead of the granular
structure obtained at lower current densities, the deposit consisted of numerous Ni needles of
submicron width, probably corresponding to deposits growing only through pores due to a
much more nonuniform current distribution at the much higher local current densities
prevailing in this case.
ACKNOWLEDGMENTS
The authors wish to thank EPSRC for a studentship to I.J.B. (Quota Ref. No. 97309830). S.S.
acknowledges The University of Nottingham for a New Lecturers’ Research Grant
(97NLRG860) and the Royal Society for a Royal Society Research Grant (RSRG19058).
REFERENCES
1. T.C. Benjamin, E.H. Camera, and L.G. Marianowski, Handbook of Fuel Cells, Institute of Gas
Technology, Chicago, IL (1980).
2. S. Ye, A.K. Vijh, and L.H. Dao, J. Electrochem. Soc. 144, 90 (1997).
3. Sleem-Ur-Rahman, M.A. Al-Saleh, A.S. Al-Zakri, and S. Gultekin, J. Appl. Electrochem. 27, 215
(1997).
4. O. Bo¨hme, F.U. Leidich, H.J. Salge, and H. Wendt, Int. J. Hydrogen Energy 19, 349 (1994).
5. C.A. Vincent and B. Scrosati, Modern Batteries, 2nd ed., Arnold, London (1997).
6. J. Francisco, D. Chiappetti, and J. Brill, in 1996 NASA Aerospace Battery Workshop, ed. J.C.
Brewer, p. 283, NASA Space Flight Center, Huntsville, AL (1997).
7. K. Wall, Modern Chlor-Alkali Technology, Vol. 3, Ellis Horwood, Chichester (1986).
8. K. Lohrberg and P. Kohl, Electrochim. Acta 29, 1557 (1984).

1064
I.J. BROWN et al.
Vol. 34, No. 7
9. D.A. Cox, PhD Thesis, Southampton, UK (1981).
10. J. Chaussard, R. Rouget, and M. Tassin, J. Appl. Electrochem. 16, 803 (1986).
11. C.J. Brown, D. Pletcher, F.C. Walsh, J.K. Hammond, and D. Robinson, J. Appl. Electrochem. 24,
95 (1986).
12. S.H. Langer and H.P. Landi, J. Am. Chem. Soc. 85, 3043 (1963).
13. V. Anantharaman and P.N. Pintauro, J. Electrochem. Soc. 141, 2729 (1994).
14. L. Cai and H.Y. Cheh, Meeting Abstracts of the 191st Meeting of The Electrochemical Society,
p. 58, No. 51, The Electrochemical Society, Inc., Pennington, NJ (1997).
15. H.-D. Kunze, J. Baumeister, J. Banhart, and M. Weber, PIM 25, 182 (1993).
16. S. Ahn and B.J. Tatarchuk, J. Appl. Electrochem. 27, 9 (1997).
17. M. Parent, S. Passerini, B.B. Owens, and W.H. Smyrl, in Meeting Abstracts of the 191st Meeting
of The Electrochemical Society, p. 840, No. 637, The Electrochemical Society, Inc., Pennington,
NJ (1997).
18. Th. Borucinski, S. Rausch, and H. Wendt, J. Appl. Electrochem. 22, 1031 (1992).
19. S. Rausch and H. Wendt, J. Electrochem. Soc. 143, 2852 (1996).
20. D. Barby and Z. Haq, European Patent 0,060,138, Unilever, 1982.
21. J. Williams, Langmuir 4, 44 (1988).
22. G. Akay, Z. Bhumgara, and R.J. Wakeman, Trans. IChemE 73, 782 (1995).
23. P. Hainey, I.M. Huxham, B. Rowatt, D.C. Sherrington, and L. Tetley, Macromolecules 24, 117
(1991).
24. S. Sotiropoulos, I.J. Brown, G. Akay, and E. Lester, Mater. Lett. 35, 383 (1998).
25. F.A. Lowenheim, Electroplating, McGraw-Hill, New York (1978).
26. A.K. Graham, Electroplating Engineering Handbook, p. 248, Van Nostrand, New York (1971).