Porous membranes for the preparation of magnetic nanostructures

Article abstract of DOI:10.1016/S0304-8853(02)00533-4

This paper is a short review on the use of microporous membranes as templates for the preparation of nanomaterials using different synthetic approaches. Magnetic nanowires and multilayers have been synthesized by electrochemical deposition techniques. Hereby, we also report preliminary results to show the advantage of the chemical route as an alternative to template other nanostructures as for instance Co nanotubes.

Full text of DOI:10.1016/S0304-8853(02)00533-4

Journal of Magnetism and Magnetic Materials 249 (2002) 214–219
Porous membranes for the preparation of magnetic
nanostructures
1
P. Aranda*, J.M. Garc !ı a
Instituto de Ciencia de Materiales de Madrid, CSIC, Cantoblanco, E-28049 Madrid, Spain
Abstract
This paper is a short review on the use of microporous membranes as templates for the preparation of nanomaterials
using different synthetic approaches. Magnetic nanowires and multilayers have been synthesized by electrochemical
deposition techniques. Hereby, we also report preliminary results to show the advantage of the chemical route as an
alternative to template other nanostructures as for instance Co nanotubes.
r 2002 Elsevier Science B.V. All rights reserved.
PACS: 75.70.Ày; 81.20.Zx; 85.40.u
Keywords: Nanowires; Template synthesis; Nanostructures; Nanotubes; Magnetic materials
1
. Introduction
use of the organized porosity of certain micro-
porous membranes. C.R. Martin’s group [2a–c]
was pioneer in the use of this methodology and
nowadays other teams are using it to prepare
nanoparticles, nanotubes, nanofibrils and nano-
wires of different materials such as polymers,
metal oxides, metals, etc.
Magnetic materials and other related structures
have also been prepared at the nanoscale level
using lithographic methods and template mem-
branes [3]. Filling of pores of anodized Al with
magnetic materials was firstly reported in 1975 [4]
although the advantage of polymeric track-etch
membranes for the preparation of arrays of
magnetic nanowires was not reported till the
Nowadays one of the most exciting areas in
Materials Science is the study of nanomaterials
due to their potential applications in fields as
diverse as optics, electronics, catalysis, magnetism,
electrochemistry, information processing and sto-
rage, etc. Preparation of inorganic, organic or
organic–inorganic hybrid materials in the nan-
ometer scale can be achieved either by physical or
chemical methods and in many cases it requires the
use of solids presenting voids or cavities in which
the material can be synthesized [1]. Among the
different preparation procedures, one of the most
widespread is the template method, based on the
1
990s [5]. Afterwards, the groups of Piraux [6]
and Ansermet [7], and more recently others
8,9a,b,10a,b], have used track-etch and anodized
Al O membranes to prepare arrays of nanowires
*
Corresponding author. Tel.: +34-91-334-9000; fax: +34-91
72 0623.
E-mail address: aranda@icmm.csic.es (P. Aranda).
3
[
1
Presently at: Laboratoire de Physique des Solides,
Universit e! Paris-Sud & CNRS, Orsay, France
2
3
or multilayers of magnetic metals (Fe, Ni, Coy).
0304-8853/02/$ - see front matter r 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 3 0 4 - 8 8 5 3 ( 0 2 ) 0 0 5 3 3 - 4

P. Aranda, J.M. Garc !ı a / Journal of Magnetism and Magnetic Materials 249 (2002) 214–219
215
This paper is organized starting with an
introduction about membranes and methodologies
to template nanomaterials including magnetic
nanowires, then preliminary results are introduced
on the use of the chemical route as an alternative
approach to grow magnetic nanomaterials in
membranes and finally some conclusions and
prospects are included.
Track-etch mica membranes present higher che-
mical and thermal stability with diamond-like
cross-section pores (Fig. 1b). Anodic Al O mem-
2
3
branes are prepared by electrochemical oxidation
of Al producing pores of asymmetric structure
(Fig. 1c and d). Whatman (Anodisc) and Merck
(Anotec) commercially sell anodic Al O mem-
2
3
branes but they are restricted to a very limited
range of pore diameters. Consequently, many
researchers prepare their own templates. Recently,
the preparation of polycrystalline and monocrys-
talline pore arrays with large interpore distance in
anodic Al O has been reported [11]. Nanochannel
2
. Membranes as templates of nanomaterials
2
3
Porous membranes are generally employed in
glass templates (Corning) and porous silicon are
two other types of porous membranes also used as
templates for the synthesis of nanomaterials.
filtration technologies for the separation of differ-
ent species (polymers, colloids, molecules, salts,
etc.), depending on their pore radii that may vary
from mm to nm. Although they exhibit in general
heterogeneous porosity, a few of them can be
prepared with well-defined shape pores of a
narrow distribution of diameters. Among these it
is worth mentioning track-etch membranes and
anodized Al O membranes. Polycarbonate (PC)
The synthetic approaches used for the prepara-
tion of nanomaterials inside the pores of mem-
branes are variable, due to the large variety of
materials and leading purposes. Martin [2a–c] and
more recently Hucko [12] have extensively re-
viewed the subject, and therefore we will just
shortly refer to the most commonly employed
methodologies. The strategy consists in finding out
the right pathway leading to the wanted nanoma-
terial (nanorods, nanotubes, nanowires, nanocom-
posites, multilayersy) presented in a particular
final form, i.e., as an ordered array with individual
components more or less separated, single nano-
particles, etc. Strategies of synthesis must consider:
2
3
track-etch membranes (commercially available
from Nucleopore, Poretics, Millipore) show cy-
lindrical pores (Fig. 1a), mainly perpendicular to
the membrane sheet although they could be tilted
up to 341 (o51 in ‘‘home-made’’ membranes).
(i) nature of precursors and affinity to interact
with the pore wall (e.g., hydrophobic/hydrophilic
character of solvents); (ii) characteristics of the
deposition reaction (rate, sequence of steps, etc.)
and (iii) chemical and thermal stability of the host
membrane. The synthetic approaches can be
summarized as follows:
*
Electrochemical deposition is perhaps the most
employed method and requires the coating of
one face of the template with a metal film to act
as cathode for electroplating. The volume of the
pore is continuously filled from the pore bottom
up allowing the control of the total length of
wires or even the preparation of multilayers. A
large variety of metal nanowires (Au, Ag, Ni,
Co, Cu, etc.) have been prepared following
this method although preparation of metal
Fig. 1. SEM images of track-etch polycarbonate (a) and mica
(b) membranes and both surfaces (c) and (d) of an anodized
Al membrane.
2 3
O

216
P. Aranda, J.M. Garc !ı a / Journal of Magnetism and Magnetic Materials 249 (2002) 214–219
nanotubes requires previous modification, e.g.,
silanization of Al O pore walls [2a–c]. Con-
mation, etc.) only inorganic membranes such as
Al O templates can be used.
2
3
2
3
*
ducting polymers as polypyrrole (PPy), polyani-
line (PANI), etc. have been electropolymerized
using different templates resulting either nano-
fibers or nanotubes depending on factors like
experimental conditions, polymer nature or
template membrane [2a–c].
Chemical vapor deposition (CVD) is a techni-
que widely used for preparation of solid thin
films, but barely applied to produce templated
nanostructures because of its usually fast
deposition rates. However, carbon nanotubes
of variable wall thickness can be produced by
CVD passing a gas (ethene or propene) through
Al O membranes placed inside a high-tem-
*
Chemical deposition usually requires either a
chemical reducing agent, to plate a metal from
solution onto a surface, or a catalyst, which
promotes the reaction of the reagents filling the
pores. As the deposition starts at the pore wall,
it is possible to obtain hollow tubes of selected
inside diameter with length and outside dia-
meter determined by the characteristics of the
template membrane. Other methods to prepare
metal arrays (e.g., Au) or more complicate
structures as LiMn O nanotubes have been
2
3
perature furnace [15].
A large number of nanocomposites as metal
multilayers, concentric-tubular nanostructures,
etc. [2a–c,3,12,16] can be synthesized by combining
properly the anterior strategies enlarging in this
way the scope of applications of nanomaterials.
Anodized Al O was the first porous structure
used to grow magnetic nanostructures [4] although
the limitations to reach complete filling of the
pores determined the use of nanoporous Al O
2
3
2
4
reported [12]. In certain cases, it is also possible
the incorporation of fluids or melts (In, Sn,
Al, Se, Te, GaSb, Bi Te ) by high-pressure injec-
2
3
sheets detached from the electrode, i.e., mem-
branes. In this way, Al O , track-etch PC and mica
2
3
tion [13].
2
3
*
Chemical polymerization is accomplished by
simply immersing the membrane into a solution
containing the monomer and a polymerization
reagent. The polymer growth starts on the pore
walls allowing the preparation of either tubes or
fibers. Conducting polymers as PPy, PANI,
etc., showing enhanced conductivities have been
prepared by this method. PPy nanotubes can be
used for bioencapsulation of enzymes to pre-
pare, for instance, biosensors of glucose [2a–c].
Insulating polymers as polyacrylonitrile can be
grown into Al O membranes and thermally
membranes, and nanochannel glass templates have
been employed to prepare a large variety of
magnetic nanostructures as arrays of nanowires
and multilayers [3–8,9a,b,10a,b,17,18]. PC track-
etch membranes is the template most extensively
employed, as they are easy to handle and exhibit
relatively low porosity limiting magnetic interac-
tion among individual wires, but defects (di-, tri-
pores, surface roughness, tilting of poresy) could
cause detriment in the searched properties. Mica
track-etch membranes [9a,b] are fragile and show
pores, i.e., nanowires, with a curious shape but its
higher thermal stability allows the study of
magnetic properties with temperature and nano-
wires modified by previous thermal treatments.
Al O membranes are also thermally stable and
show better parallel alignment of perpendicular
pores but their higher porosity and smaller
interpore distance could introduce magnetic inter-
action among the nanowires. The general proce-
dure to produce arrays of nanowires (Co, Ni, Fe,
CoNi, NiFey) is electrochemical deposition,
which also allows production of multilayers (Co/
Cu, Ni/Cu, etc.) [6,7,17]. Magnetic nanowires
(Ni, Fe) have also been encapsulated on hollow
2
3
transformed into carbon nanotubes and fibrils
2a–c]. Poly(methylmetacrylate) has been pre-
pared inside the pores of Al O membranes to
[
2
3
produce a ‘‘negative’’ polymeric membrane by
dissolution of the Al O template. This ‘‘new
2
3
2
3
template’’ can be then used for the synthesis
of nanoporous structures of metals as Au or
Pt [14].
*
Sol–gel deposition is a relatively easy way to
prepare metal oxides (TiO , ZnO, WO , V O ,
2
3
2
5
etc.) tubules and fibers [2a–c,12]. As the sol–gel
method involves thermal treatments among
other steps (hydrolysis of precursors, gel for-

P. Aranda, J.M. Garc !ı a / Journal of Magnetism and Magnetic Materials 249 (2002) 214–219
217
nanotubes of conducting polymers (PPy, PANI)
chemically prepared on PC or Al O membranes
is minimum in such direction as a consequence of
the dipolar interaction among the wires increasing
for arrays templated on PC membranes because
their interpore distance is greater (porosity: 25%
for PC and 80% for Al O ) [10a,b].
2
3
[
19a,b]. Structures such as carbon nanotubes [20],
biomolecular entities [21] or mesoporous silica [22]
have also been used to grow magnetic metal
nanowires and nanotubes. These last approaches
could be of interest as they introduce methods of
synthesis that might be alternatives to electro-
deposition for growing other type of nanostructres
than nanowires.
2
3
In a new approach we have used the chemical
route to template Co nanostructures in Al O
2
3
(Anodisct) and track-etch PC (Nucleoporet)
membranes. As it is well known, certain organo-
metallic complexes and coordination compounds
are used to generate metal particles for catalytic
purposes [23]. We have selected the Co (CO) as
2
8
3
. Chemical preparation of Co nanostructures
precursor. This compound is very sensitive to
moisture and thermally decomposes to Co
(o901C) in an inert atmosphere [24]. Incorpora-
tion of the precursor inside the pores of the
membrane is achieved by two procedures, working
As we have reported elsewhere [10a,b] electro-
chemical synthesis allows the preparation of Co
nanowires (Fig. 2a) whose crystalline structure,
cubic or hexagonal, depends on the nature of the
membrane support, Al O or PC templates,
respectively. Despite these differences both types
of arrays show similarities in the magnetic
behavior. The easy direction is along the wire axis
due to the strong shape anisotropy. The coercivity
always in a dry box (o0.2 ppm H O, Ar atmo-
2
sphere). The membranes are previously treated
under dynamic vacuum for 24 h to remove
adsorbed water. In the first procedure a membrane
(e.g., Al O 200 nm pore diameter) is immersed
2
3
2
3
À2
E5 min in 10 ml of 5 Â 10 M Co (CO) /n-hexane
2 8
Fig. 2. SEM images of Co nanostructures templated in Al
hollow nanocylinder (c).
2 3
O membranes by chemical (a) and electrochemical methods (b), detail of a

218
P. Aranda, J.M. Garc !ı a / Journal of Magnetism and Magnetic Materials 249 (2002) 214–219
solution and then dried at room temperature. The
process is repeated several times, typically 5, to
increase the loading of the pores. The second
procedure profits from the relatively high vapor
pressure of the carbonyl complex to pass it
through the pores using the experimental setup
shown in Fig. 3. In both the cases the precursor is
deposited inside the pores and the membrane turns
brown-black. Once the membrane had the pre-
cursor incorporated, the system was alternatively
left to evolve several days at room temperature or
heated at 901C to produce the formation of Co
particles. The SEM study coupled with EDX
microanalysis indicates the presence of Co inside
the pores of the membrane. From the FTIR and
DRX data we deduce that the amount of adsorbed
compound must be very small, since we could not
detect the precursor, metallic Co or cobalt oxide.
After dissolution of the membrane the formation
of Co hollow cylinders could be observed (Fig. 2a
and c), in contrast to the nanowires resulting from
electrochemical deposition (Fig. 2b). Magnetism
of the arrays of Co nanotubes prepared by the two
procedures, measured by means of VSM using a
LDJ magnetometer, is undetectable. It could be
that the strong interaction of Co particles with the
Al O substrate induces oxidation of the metal but
that oxidize when exposed to air during the
magnetic measurements. Besides, we believe that
the small amount of deposited Co might also be a
reason of this lack of magnetic properties. The
method of synthesis, including the use of other
precursors in order to prepare nanocomposite
structures, are currently being optimized.
4. Conclusions and perspectives
The versatility of the membrane template
method allows the exploration of new routes to
prepare new magnetic systems. In this way, we
have showed that the chemical route is an
alternative to electrochemical deposition to obtain
Co nanotubes instead of nanowires. This new
approach opens a way to the preparation of
composite structures, as for instance ferromag-
netic/non-magnetic nanotubes, with potential no-
vel properties, although there are still experimental
difficulties to obtain nanostructures with reprodu-
cible magnetic behavior.
2
3
Acknowledgements
the same behavior is showed by arrays prepared
using PC membranes. Probably the method
induces formation of very small particles of metal
Authors thank financial support from the
CICYT (Spain).
References
[
[
1] G.A. Ozin, Adv. Mater. 4 (1992) 612.
2] (a) C.R. Martin, Science 266 (1994) 1961.
(
(
b) C.R. Martin, Chem. Mater. 8 (1996) 1739.
c) J.C. Hulteen, C.R. Martin, J. Mater. Chem. 7 (1997)
1075; and references therein..
3] D. Routekevich, A.A. Tager, J. Haruyama, D. Alamwlawi,
[
M. Moskovits, J.M. Xu, IEEE Trans. Electron Devices 43
(
1996) 1646.
[
[
4] S. Kawai, R. Ueda, J. Electrochem. Soc. 19 (1975) 32.
5] T.M. Withney, J.S. Jiang, P.C. Searson, C.L. Chien,
Science 261 (1993) 1316.
[6] L. Piraux, J.M. George, J.F. Despres, C. Leroy, E. Ferain,
R. Legras, K. Ounadjela, A. Fert, Appl. Phys. Lett. 65
(
1994) 2484.
Fig. 3. Experimental setup for chemical growth of Co nano-
tubes inside microporous membranes.
[7] A. Blondel, J.P. Meier, B. Doudin, J.-Ph. Ansermet, Appl.
Phys. Lett. 65 (1994) 3019.

P. Aranda, J.M. Garc !ı a / Journal of Magnetism and Magnetic Materials 249 (2002) 214–219
219
[
[
8] W. Schwarzacher, K. Attenborough, A. Michel, G.
Nabiyouni, J.P. Meir, J. Magn. Magn. Mater. 165 (1997)
[17] W. Schwarzacher, D.S. Lashmore, IEEE Trans. Magn. 32
(1996) 3133.
2
3.
9] (a) L. Sun, P.C. Searson, C.L. Chien, Appl. Phys. Lett. 74
1999) 2803.
b) L. Sun, P.C. Searson, C.L. Chien, Phys. Rev. B 61
2000) R6463.
[18] P.P. Nguyen, R.J. Tonucci, in: R.F. Lobo et al. (Eds.),
Microporous, Macroporous Materials, MRS Symposium
Proceedings, Vol. 431, Materials Research Society, Pitts-
burgh, 1996, p. 421.
(
(
(
[19] (a) L. Piraux, S. Dubois, J.L. Duvail, A. Radulescu, S.
Demoustier-Champagne, E. Ferain, R. Legras, J.
Mater. Res. 14 (1999) 3042.
[
10] (a) J.M. Garc !ı a, A. Asenjo, J. Vel a! zquez, D. Garc !ı a, M.
V a! zquez, P. Aranda, E. Ruiz-Hitzky, J. Appl. Phys. 85
(
1999) 5480.
(b) H. Cao, Z. Xu, D. Sheng, J. Hong, H. Sang, Y. Du, J.
Mater. Chem. 11 (2001) 958.
(
b) J.M. Garc !ı a, A. Asenjo, M. V a! zquez, P. Aranda, E.
Ruiz-Hitzky, IEEE Trans. Magn. 36 (2000) 2981.
11] A.P. Li, F. M u. ller, U. G o. sele, Electrochem. Solid-State
Lett. 3 (2000) 131.
[20] N. Grobert, W.K. Hsu, Y.Q. Zhu, J.P. Hara, H.W. Kroto,
M. Terrones, H. Terrones, Ph. Redlich, M. R u. hle, R.
Escudero, F. Morales, Appl. Phys. Lett. 75 (1999) 3363.
[21] R. Kirsch, M. Mertig, W. Pompe, R. Wahl, G. Sadowski,
K.J. B o. hm, E. Unger, Thin Solid Films 305 (1997) 248.
[22] L. Chen, P.J. Klar, W. Heimbrodt, F. Brieler, M. Fr o. ba,
Appl. Phys. Lett. 76 (2000) 3531.
[
[
[
12] A. Huczko, Appl. Phys. A 70 (2000) 365.
13] C.A. Huber, R.E. Huber, M. Sadoqi, J.A. Lubin, S.
Manalis, C.B. Prater, Science 263 (1994) 800.
14] H. Masuda, K. Fukuda, Science 268 (1995) 1466.
15] T. Kyotani, L. Tsai, A. Tomita, Chem. Mater. 8 (1996)
[
[
[23] C. Moreno-Castilla, J. Chem. Soc. Faraday Trans. 91
(1995) 3519.
2109.
[
16] V.M. Cepak, J.C. Hulteen, G. Che, K.B. Jirage, B.B.
Lakshmi, E.R. Fisher, C.R. Martin, J. Mater. Res. 13
[24] G. Bor, U.K. Dietler, P. Pino, A. Poe, J. Organomet.
Chem. 154 (1978) 301.
(1998) 3070.