Macroporous germanium by electrochemical deposition

Article abstract of DOI:10.1039/b206970a

Germanium is electrodeposited in a template formed from a dried suspension of silica spheres. The germanium completely fills the pores of the silica matrix. The semiconductor, as deposited, is amorphous but can be crystallized by annealing. Selective dissolution of the silica template gives a macroporous germanium-air sphere matrix, which offers interesting possibilities for photonic applications.

Full text of DOI:10.1039/b206970a

Macroporous germanium by electrochemical deposition
a
a
a
b
b
a
L. K. van Vugt, A. F. van Driel, R. W. Tjerkstra†, L. Bechger†, W. L. Vos†, D. Vanmaekelbergh
a
and J. J. Kelly*
a
Physics and Chemistry of Condensed Matter, Debye Institute, Universiteit Utrecht, P.O. Box 80000, 3508
TA Utrecht, The Netherlands. E-mail: J.J.Kelly@Phys.uu.nl
Van der Waals –Zeeman Instituut, Universiteit van Amsterdam, Valckenierstraat 65, 1018 XE Amsterdam,
b
The Netherlands
Received (in Cambridge, UK) 16th July 2002, Accepted 5th August 2002
First published as an Advance Article on the web 14th August 2002
Germanium is electrodeposited in a template formed from a
dried suspension of silica spheres. The germanium com-
pletely fills the pores of the silica matrix. The semiconductor,
as deposited, is amorphous but can be crystallized by
annealing. Selective dissolution of the silica template gives a
macroporous germanium-air sphere matrix, which offers
interesting possibilities for photonic applications.
selectively with respect to the plated semiconductor to produce
a germanium replica of the template.
As electrode or plating base, cleaned glass slides (1 3 3 cm)
were provided with a 50 nm thick gold or copper film by
evaporation. Templates were formed by spreading evenly a thin
layer of a concentrated suspension of silica particles in ethanol
on the gold surface. The particles had an average diameter of
2
26 nm with a polydispersity of 2%. After drying, the layers
A widely used approach to the fabrication of macroporous
systems is to fill the voids of a template, e.g. an artificial opal,
with the solid of interest and subsequently to remove selectively
the template material. Filling methods that have been employed
include chemical vapour deposition (CVD), sol–gel chemistry
and infiltration using suspensions of nanoparticles.1 Recent
work has shown that electrodeposition of a solid in the porous
template is a relatively simple and attractive method for
producing macroporous systems.4 Electrodeposition has the
advantage that the new phase is grown from the plating base; the
voids are filled ‘from the bottom up’, giving volume-templated
structures. This enables a higher volume fraction of solid than
that generally obtained by other methods in which deposition
occurs on the internal surface of the template (“surface
templating”). Electrodeposition has been used to make ordered
were annealed for 3 h at 600 °C under nitrogen to create necks
between the particles and make the templates more robust. The
thickness of the silica layers was varied between 0.8 and 2 mm.
In this work we did not attempt to make well-ordered colloidal
crystals.
The solvent used for the plating bath was either anhydrous
ethylene glycol (99.8% from Aldrich) or anhydrous propylene
glycol (99.8% from Merck). GeCl (99.99% from Acros) was
4
added to the solvent to give a 5–10 vol% solution; this was
carried out in a nitrogen-flushed glove box to prevent hydrolysis
–3
–6
4
of the GeCl . The baths also contained KCl and acetic acid (both
p.a. quality from Merck) to increase the conductivity and
improve the plating properties. Electrodeposition was carried
out at 60 °C. Either a two-electrode cell with a large area
graphite counter electrode or a three-electrode cell with a Ag/
AgCl reference electrode was used for the electrochemical
experiments. A nitrogen overpressure was maintained in the cell
during operation.
4–6
macroporous structures of gold, nickel, CdS and CdSe. The
4,6
resulting materials have interesting optical and, in the case of
nickel, magnetic properties.5
One of the most exciting fields of application of macroporous
solids is in photonics. Random porous systems can give strong
light scattering and, in the ultimate case, Anderson localiza-
tion.7 In a photonic bandgap material, the porous matrix must
be crystallographically ordered. In both cases solids are
required with a refractive index that is large with respect to that
of air.
The germanium layers were characterized by X-ray diffrac-
tion (with a Philips PW 1820 diffractometer), by scanning
electron microscopy (SEM) and energy dispersive X-ray
analysis EDAX (with a Philips XL 305 field emission gun
scanning electron microscope). The germanium layer thickness
was measured with a Dektak 303 OST auto texture profiler.
Fig. 1 shows a cyclic voltammogram for an experiment with
a gold working electrode in an ethylene glycol solution. A scan
to negative potentials gives a cathodic peak at about 20.5 V vs.
Ag/AgCl (range I). If the scan direction is reversed immediately
after this peak, an anodic peak (not shown in Fig. 1) is observed.
Germanium deposition does not occur in this range; therefore,
–9
Germanium, with a refractive index of 4 and a bandgap of
.67 eV, is an interesting candidate for photonic applications in
0
the IR spectral range. Gómez Rivas et al. have studied light
scattering in a random germanium matrix based on micron-
8
sized particles obtained by pulverizing single crystals. Míguez
et al.10 have made networks of air spheres in germanium by a
three-step approach involving hydrolysis of tetramethoxy-
germane in a crystalline silica template, the reduction of the
2
GeO product of this reaction, and the selective removal of the
11
template. Subsequent work by Míguez et al., in which
germanium was deposited from digermane by CVD, gave
improved results. SEM images show clearly that this is a
surface-templated process with the semiconductor being de-
posited on the silica spheres throughout the layer. Fill factors up
to 78% were reported, depending on the deposition temperature.
Interesting optical results have been observed.11
In this paper we show that germanium can be electro-
deposited directly in a porous silica matrix. For this purpose we
used a plating bath based on GeCl in ethylene or propylene
4
glycol; the bath is an adaptation of one previously reported by
Szekely.12 We also show that it is possible to etch the silica
Fig. 1 Cyclic voltammogram of a gold working electrode in a 10 wt%
†
Current address: Universiteit Twente, P.O.Box 217, 7500 AE, Enschede,
4
ethylene glycol solution of GeCl . The bath also contained 1.66 wt% KCl
The Netherlands.
and 1.66 wt% acetic acid.
2
054
CHEM. COMMUN., 2002, 2054–2055
This journal is © The Royal Society of Chemistry 2002

the peaks are due to the Ge(IV)/Ge(II) couple.12 At more
negative potentials (range II) the cathodic current increases due
to deposition of germanium and reduction of protons to
hydrogen. The germanium can be stripped anodically at
potentials more positive than +0.1 V vs. Ag/AgCl. Although the
current efficiency is low ( ~ 1%) micron-thick germanium
layers could be deposited.
The electrochemistry of Ge in silica templates showed results
similar to those obtained with the bare gold working electrode.
Clearly, the plating solution permeates the porous template and
has access to the gold base. The semiconductor could thus be
deposited under conditions similar to those for a bare gold
electrode.
The silica matrix could be selectively removed in 5% HF
solution to expose the plated replica. Fig. 2(a) is a SEM image
of an unannealed sample deposited in a two-electrode cell with
a cell voltage of 2.3 V, after removal of the silica. An EDAX
spectrum confirmed that germanium had indeed been deposited
and that the template had been removed. From the absence of
peaks in the X-ray diffractogram we must conclude that the as-
deposited layer is amorphous. The morphology seen in Fig. 2(a)
is typical of the whole sample. Clearly, on removal of the silica
a three-dimensional network of air-spheres in germanium is
formed. The size of the domains in the silica matrix is rather
small (1–2 mm). Germanium is also deposited in vacancies and
domain boundaries (not shown); the filling of such large voids
provides strong evidence for volume templating, with the
germanium growing uniformly from the plating base. Amor-
phous films are not suitable for photonic applications; conse-
quently we heat-treated the germanium after deposition. The
morphology seen in Fig. 2(a) does not change when the sample
is annealed for 30 min at 400 °C in a nitrogen atmosphere. From
the X-ray diffractogram of Fig. 2(b), however it is clear that the
film becomes crystalline during annealing. Peaks in the 2q
diffractogram are observed at 27.28, 45.30 and 53.68° corre-
sponding to the (111), (220) and (311) planes of germanium.
Application of the Scherrer equation to the peak widths yields
crystallite sizes in the 20–50 nm range. For the electrochemical
approach, deposition of an amorphous or a nanocrystalline layer
is desirable; formation of larger crystallites with facetted
surfaces is clearly unfavourable for perfect filling of the
voids.13
This electrochemical method provides an interesting alter-
native to the CVD approach of Míguez et al.11 High fill factors
are obtained. Two requirements have to be met in order to
produce a photonic bandgap in germanium: (i) a thicker
template crystal with long-range order is needed; (ii) because
the bandgap of germanium corresponds to the IR spectral range,
the silica spheres forming the artificial opal must have a
diameter larger than those used in the present work. In principle,
it is quite possible to meet both of these requirements.
Our results show that electrodeposition in silica templates
provides an interesting route to the formation of both random
and ordered networks of air spheres in germanium. Such
8
networks can act as a high scattering medium for IR light or as
a photonic bandgap material.
We would like to thank Dr Ralf Wehrspohn of the Max
Planck Institute, Halle for suggesting germanium electrodeposi-
tion. The work of L. B. and W. L. V. is part of the research
programme of the ‘Stichting voor Fundamenteel Onderzoek der
Materie (FOM)’, which is financially supported by the
‘
Nederlandse Organisatie voor Wetenschappelijk Onderzoek
(NWO)’. The work of A. F. v. D. and R. W. T. was financed by
NWO-CW.
Notes and references
1
2
A. Stein, Microporous Mesoporous Mater., 2001, 44–45, 227–239.
Y. A. Vlasov, X.-Z. Bo, J. C. Sturm and D. J. Norris, Nature, 2001, 414,
2
89.
3
4
J. E. G. J. Wijnhoven and W. L. Vos, Science, 1998, 281, 802.
J. E. G. J. Wijnhoven, S. J. M. Zevenhuizen, M. A. Hendriks, D.
Vanmaekelbergh, J. J. Kelly and W. L. Vos, Adv. Mater., 2000, 12,
8
88.
5
L. Xu, W. L. Zhow, C. Frommen, R. H. Baughman, A. A. Zakhidov, L.
Malkinski, J.-Q. Wang and J. B. Wiley, Chem. Commun., 2000, 997.
P. V. Braun and P. Wiltzius, Adv. Mater., 2001, 13, 482.
F. J. P. Schuurmans, M. Megens, D. Vanmaekelbergh and A. Lagendijk,
Phys. Rev. Lett., 1999, 83, 2183.
6
7
8
9
J. Gómez Rivas, R. Sprik, A. Lagendijk, L. D. Noordam and C. W.
Rella, Phys. Rev. E, 2000, 62, 4540R.
F. J. P. Schuurmans, D. Vanmaekelbergh, J. van de Lagemaat and A.
Lagendijk, Science, 1999, 284, 141.
1
0 H. Míguez, F. Meseguer, C. López, M. Holgado, G. Andreasen, A.
Mifsud and V. Fornés, Langmuir, 2000, 16, 4405.
Fig. 2 (a) SEM image of a macroporous amorphous germanium layer
electrodeposited from a propylene glycol solution after removal of the
template (b). The X-ray diffractogram indicates that after annealing for 30
min at 400 °C under nitrogen crystalline germanium is formed. The gold and
aluminium peaks are due to the substrate and sample holder. The large
background is due to the glass substrate.
11 H. Míguez, E. Chomski, F. García-Santamaría, M. Ibisate, S. John, C.
López, F. Meseguer, J. P. Mondia, G. A. Ozin, O. Toader and H. M. van
Driel, Adv. Mater., 2001, 13, 1634.
12 G. Szekely, J. Electrochem. Soc., 1951, 98, 318.
13 P. E. de Jongh, D. Vanmaekelbergh and J. J. Kelly, Chem. Mater., 1999,
11, 3512–3517.
CHEM. COMMUN., 2002, 2054–2055
2055