Fractal porosity in metals synthesized by a simple combustion reaction

Article abstract of DOI:10.1039/c2ra22441k

A simple modification of a combustion method has been used for the production of ultraporous metals in air. Nitrates of different metallic elements were reacted with glycine as a reducing fuel. The glycine to nitrate ratio can be simply used to control the formation of oxides or, in the case of fuel-rich mixtures, the formation of metals such as Ni, Co, Cu or Ag. Furthermore, the metallic monoliths obtained present a remarkable porosity of fractal nature (from macro to nano scales) with pores ranging from many microns down to at least 5 nm. This exceedingly simple approach shows the way for the design and synthesis of complex porous microstructures of metals for the wide variety of applications where interface optimization is crucial.

Full text of DOI:10.1039/c2ra22441k

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Fractal porosity in metals synthesized by a simple
combustion reaction3
Cite this: RSC Advances, 2013, 3,
2351
ab
Pedro Gomez-Romero,* Julio Frailec and Belen Ballesterosa
´
A simple modification of a combustion method has been used for the production of ultraporous metals in
air. Nitrates of different metallic elements were reacted with glycine as a reducing fuel. The glycine to
nitrate ratio can be simply used to control the formation of oxides or, in the case of fuel-rich mixtures, the
formation of metals such as Ni, Co, Cu or Ag. Furthermore, the metallic monoliths obtained present a
remarkable porosity of fractal nature (from macro to nano scales) with pores ranging from many microns
down to at least 5 nm. This exceedingly simple approach shows the way for the design and synthesis of
complex porous microstructures of metals for the wide variety of applications where interface
optimization is crucial.
Received 8th October 2012,
Accepted 5th December 2012
DOI: 10.1039/c2ra22441k
www.rsc.org/advances
selective leaching or dealloying,2,4,5,7–10 application of sacrifi-
cial templates11–13 or the combination of combustion and
Introduction
chemical techniques.14 The surface area of these materials
depends on the preparation conditions and can vary between
One of the greatest challenges facing materials science for the
coming decades is the management of complexity in multiple
length scales (macro, meso, micro). But this ever-growing
complexity should not be accompanied by a matching level of
intricacy in the fabrication processes. In recent decades a
wealth of work has been devoted for instance to the control of
macro- and meso-porosity through the use of pore-formers
and surfactant-templated micelles, respectively.1 Different
tools for different dimensions. And when the desired porosity
reaches the micro domain, still different molecular
approaches need to be used. But in order to keep advancing
the field of complexity, it will be necessary to use simple
preparation methods and engineering techniques. In the case
of porous materials for instance, the goal would be to induce
hierarchically structured porosity without hierarchically varied
tools. Furthermore, even better than hierarchical structures
would be to get a continuous porosity going from the macro to
the nanoworld. The reason lies in the wealth of applications
which could benefit from an improved integral surface
engineering which would include micro (nano-) meso and
macro surfaces. Applications such as ultraporous metallic
materials are of high relevance due to their technological
applications in catalysis or electrocatalysis2, sensing3, biome-
dical engineering4,5 or artificial muscles.3,6 Methods to
prepare them involve multi-step synthesis techniques like
0.5 m2 g2113 up to 20–300 m2 g21 15
Other relevant strategies
.
for the synthesis of porous metals reported in literature
include displacement of porous silicon,16,17 hydrothermal
synthesis with fatty acid templates,18 sol–gel fabrication of
hollow bimetallic microfibers,19 or even the use of diatomites
as support for catalytic bimetallic alloys.20
Some of these applications are truly limited by a lack of
understanding and control of surface and porosity, while they
require optimization and maximization of complex interfaces
(solid–gas solid–electrolyte) even triple interfaces (solid–
electrolyte–gas) in the case of fuel cells for instance.
As part of our ongoing research on the development of
materials for fuel cells we have systematically worked on the
synthesis of NiO, Ni and Ni-based cermets and have found,
and report here, surprising results which are more far-
reaching and with possible applications beyond fuel cells.
We report a fast, simple, reproducible, single-step method to
obtain metallic nickel sponges with fractal porosity (from
microns to nanometre size). The synthesis is based on the
well-known glycine–nitrate combustion technique where a
metal nitrate is used as oxidizing agent and glycine as the fuel.
This method has been frequently used to prepare high surface
area metal oxide materials,21 but no pure-metallic structures
had been obtained just through this technique when we
carried out this work. Very recently (note added in proof by
suggestion of a referee) Deraz reported a similar synthesis of
Ni in a narrow synthetic conditions range.22,23 Remarkably
and surprisingly, the simple control of the amount of glycine
provides the means to obtain not only oxides but also even
pure metals as it will be shown.
a
´
´
Centro de Investigacion en Nanociencia y Nanotecnologıa, CIN2 (CSIC-ICN),
Bellaterra, Barcelona, Spain, E-08193. E-mail: pedro.gomez@cin2.es; Fax: +34 93
5868020; Tel: +34 93 5868010
bMATGAS Research Center, Campus UAB, Bellaterra, Barcelona, Spain, E-08193
cICMAB (CSIC), Campus UAB, Bellaterra, Barcelona, Spain, E-08193
Electronic supplementary information (ESI) available: BET adsorption
3
isotherms and BJH pore analyses for each sample. See DOI: 10.1039/c2ra22441k
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 2351–2354 | 2351

View Article Online
Paper
We used different glycine-to-nitrate ratios and found that a
RSC Advances
fuel-rich stoichiometry, that is, an excess of the reducing
agent, was the simple key to produce monoliths of nickel in its
metallic state. These Ni sponges were thus produced in a
single step reaction open to the air, challenging conventional
wisdom for the need of reducing atmospheres (normally
hydrogen) in order to reach the reduced metal state.13
Furthermore, we expanded our work to show the generality
of the method to other metals, as it will be shown below.
Experimental
Hydrated metal nitrates and glycine from Aldrich were used as
received. In a typical reaction, Ni(NO3)2?6H2O and glycine
(mole ratio 1 : 2) were dissolved in the minimum amount of
deionized water with stirring, The solution was heated in a 2
litre beaker on a hot plate until a viscous green gel was formed
and eventually dried out. The heating was maintained until
ignition of the mixture took place leading to a spontaneous
combustion reaction proceeding smoothly at ca. 250 uC
(heating plate) with an orange flame spreading in a few
seconds from a single ignition point until the whole mixture
had reacted leaving a porous ash-like residue of metallic
nickel.
SEM analyses were performed in a Stereoscan S-360
equipped with EDX, INCA Energy 200 Oxford Instruments
with Be detector. The equipment was fixed for high resolution
images at 6 mm distance, 10 kV and 50 pA. The samples were
mounted on aluminium stubs. The materials were analyzed
after being sputter-coated with gold. In cases where the
material presented good metallic properties (like metallic
nickel) no coating was required. High resolution images were
obtained with a Hitachi H-4 100 FE SEM.
X-Ray diffraction analyses were performed using a Rigaku,
Rotaflex Ru-200 B with Cu-Ka (l = 1.5418), from 5 to 80u and
step intervals of 0.02u. Elemental analysis was performed in a
Carlo Erba CHN EA 1108 equipment at maximum temperature
of 1800 uC. Pore size analyses were carried out in a 9Pore
Master9 Porosimeter from Quantachrome Instruments which
analyzes pore volume of pores with diameter between 0.0070–
400 microns by intrusion between 5% to 95% of 0.5 cc of Hg.
BET analyses were carried out in a Micrometrics ASAP 2000,
the samples were degased in vacuum for 14 h before testing.
Fig. 1 SEM images of metallic nickel showing a microstructure with porosity
apparent at different length scales. In the higher magnification photo (f)
(recorded with Field Emission SEM) it becomes apparent that the structure is
made of nickel nanoparticles much smaller than 100 nm sintered together but
retaining an open porosity.
surrounded by smaller ones, the latter coming out in turn as
large pores in the next magnification scale.
This fractal nature of porosity in our metallic samples
extends through a remarkable range spanning from milli-
metres to microns to nanometres. In order to determine the
nanoporosity in a more precise way that Fig. 1e could provide,
we recorded additional high-resolution images with a field
emission SEM microscope. Fig. 1f shows both the nanoparti-
culate nature of the solid sample as well as the minimal pore
size we were able to detect with this technique corresponding
to pores of ca. 50 nm. The morphology of the whole nickel
monoliths encompass metallic nickel nanoparticles of sizes
between 15–40 nm sintered together (Fig. 1f) at the high local
temperatures reached during synthesis, generating nickel
aggregates of about 200–400 nm (Fig. 1e). Porosity is shaped
by the combustion gases that break out during the reaction
precluding the 200–400 nm nickel micro particles from
sintering and leading to a hierarchical pore distribution.
The isolation of pure nickel was confirmed by X-ray powder
diffraction (Fig. 2a). Three major peaks with d spacings 2.03,
1.76 and 1.24 Å can be indexed as fcc Ni, a phase stable at high
temperature, in good agreement with the high local tempera-
tures characteristic of the method. Marginally visible peaks at
2.40, 2.08, 2.47 Å (corresponding to NiO) and at 2.3, 2.15, 1.57
and 1.33 (from Ni3N) can also be detected. NiO has been
proposed as intermediate species produced in the course of
Results and discussion
The synthesis was carried out as described above and led to an
ash-looking material. Scanning electron microscopy (SEM)
images of that material (Fig. 1a) show the macroscopic spongy
morphology of the millimetre-sized nickel monoliths
obtained. As we analyse the same sample under higher and
higher magnifications we detect
a recurrent pattern, a
distribution of pores with different sizes. Thus, in going from
Fig. 1b through e, we can see in every case how large pores are
2352 | RSC Adv., 2013, 3, 2351–2354
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 2 X-Ray diffraction patterns of the products obtained for different metal
ions through the glycine combustion method in air.
this type of reactions.22,23 However, our final sample is
essentially NiO-free. Elemental analyses showed 0.39% N and
0.13% C. Both impurities are close to the detection limit
confirming an essentially pure metallic sample.
In order to test the generality of the method, the synthesis
of other materials was also carried out. The same procedure
leading to essentially pure Ni metal led to metals such as Cu
and Ag, or to pure oxides like Fe2O3, or Al2O3. X-Ray analyses
showed the formation of Ag, Cu and Ni in the metallic state
and pure c-Fe2O3 in its maghemite, form (Fig. 2). Fittingly,
cobalt represents an intermediate case for which mixtures of
the metal and its oxides where obtained ( Fig. 2). In all cases
the syntheses proceeded in the same single-step conditions
and followed a smooth and non-violent path. Only in the case
of copper the combustion reaction was brisk and took place
with spewing out part of the product, leaving a plate-like
morphology settled on the bottom of the beaker and well-
defined spherical particles deposited on the beaker walls.
Fig. 3 shows the porous nature of all these materials closely
resembling that of Ni described in more detail in Fig. 1.
In order to ascertain the fractal hierarchical nature of
porosity we also used BET gas absorption (Brunauer–Emmet–
Teller) and mercury porosimetry analyses. Preliminary mea-
surements for our original Ni foams revealed a wide pore-
diameter distribution with the presence of pore diameters as
small as ca. 5 nm, a surface area of 4.2 m2g21 and an average
pore fraction of 56% of the sample.
Fig. 3 SEM images of the products resulting from the nitrate–glycine reaction
starting from iron(III) nitrate (a), cobalt(II) (b), Ni(II) (c), Ag(I) (d) and Cu(II) (e). The
latter includes two photographs corresponding to the plates (e) and spherical
particles on the walls (e1) (see text).
metallic silver with 1.18 m2 g21. The exception for both, pore
volume and surface area, was observed for copper with the
highest surface area of all materials (77 m2 g21) and wild pore
size distribution, the latter possibly due to the dual morphol-
ogy observed for the final product resulting from the vigorous
reaction behaviour. In Fig. 4 this results in data for Cu
featuring an order of magnitude larger pore volume than the
other materials.
Fig. 4 plots pore volume as a function of pore diameter for
the whole series of materials studied. In all cases there is a
wide range of pore diameters ranging from 2–5 nm up to 100
nm with a minimum for pore diameters of ca. 10 nm. Thus,
fractal porosity characterizes all the materials studied.
A trend of pore volume and surface area as a function of the
material prepared is also clear from Fig. 4.
Thus, pore volume increase according to Fe . Co . Ni .
Ag. The same trend is also detected for surface area: iron oxide
shows a surface area of 17 m2 g21, followed by cobalt/cobalt
oxides, 6.32 m2 21, metallic nickel 4.21 m2 g21 and finally
g
Fig. 4 Pore diameter vs. pore volume measured by BET found in different
samples obtained by the glycine–nitrate combustion method. All data except Cu
are referred to the left y axis. The vertical axis on the right is just for Cu. The
anomalous values found for Cu must be related to the fact that its synthesis was
the only one taking place in an abrupt way (see text).
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 2351–2354 | 2353

View Article Online
Paper
RSC Advances
3 J. Weissmu¨ller, et al., Science, 2003, 300, 312–315.
4 D. V. Pugh, A. Dursun and S. G. Corcoran, J. Mater. Res.,
2003, 18, 216–221.
5 A. Dursun, D. V. Pugh and S. G. Corcoran, J. Electrochem.
Soc., 2003, 150, B355–B360.
6 R. H. Baughman, Science, 2003, 300, 268–269.
7 J. Erlebacher, et al., Nature, 2001, 410, 450–453.
8 M. Rajamati, et al., J. Mater. Chem., 2001, 11, 2489–2492.
9 M. Stratmann and M. Rohwerder, Nature, 2001, 410,
420–422.
Conclusions
We have demonstrated a simple and effective modification of
the glycine–nitrate combustion synthesis to obtain metals or
metal oxide materials with fractal porosity. The novelty of our
work is double. First we are able to obtain metals (and not just
noble metals such as Ag) very easily in air. Secondly, the
materials prepared present a remarkable hierarchically-graded
microstructure with effective fractal porosity. The materials
are prepared in a single-step reaction without the need of
further treatments. The resulting structures are formed by the
partial sintering of metal or metal oxide particles between 150–
500 nm in size. The voids left behind produce a structure with
56% porosity and more than about 4 m2 g21 of surface area for
the case of metallic nickel. The hierarchical pore size range
from microns to nanometers making these materials suitable
for a wide variety of applications for which interfaces should
be optimized. Furthermore, the method could also be explored
for the design of not only pure metals or oxides but also of
ceramic-metal nanocomposite materials.
10 Y. Ding and J. Erlebacher, J. Am. Chem. Soc., 2003, 125,
7772–7773.
11 K. M. Kulinowski, et al., Adv. Mater., 2000, 12, 833–838.
12 O. D. Velev and E. W. Kaler, Adv. Mater., 2000, 12, 531–534.
13 D. Walsh, et al., Nat. Mater., 2003, 2, 386–390.
14 M. Panda, M. Rajamathi and R. Seshadri, Chem. Mater.,
2002, 14, 4762–4767.
15 S. Raush and H. Wendt, J. Electrochem. Soc., 1996, 143,
2852–2862.
16 M. Hayase, T. Matsuzaka and J. G. Alves Brito-Neto, ECS
Trans., 2008, 16, 231–235.
17 H. Bandarenka, S. Redko, A. Smirnov, A. Panarin,
S. Terekhov, P. Nenzi, M. Balucani and V. Bondarenko,
Nanoscale Res. Lett., 2012, 7, 477.
18 A. K. Patra, A. Dutta and A. Bhaumik, Catal. Commun.,
2010, 11, 651–655.
Acknowledgements
Partial funding from the Spanish Ministry MAT2011-28931 is
acknowledged. We thank MLC for the preparation of some
samples.
19 X. Shen, L. Guo, M. Liu, F. Song and Y. Zhu, Mater. Lett.,
2011, 65, 2841–2843.
20 Y. Chen, M. Xiao, S. Wang, D. Han, Y. Lu and Y. Meng, J.
Nanomater., 2012, 2012, 610410.
21 R. D. Purohit, B. P. Sharma, K. T. Pillai and A. K. Tyagi,
Mater. Res. Bull., 2001, 36, 2711–2721.
Notes and references
1 M. Kruk, Acc. Chem. Res., 2012, 45(10), 1678–1687.
2 A. Katagiri and M. Nakata, J. Electrochem. Soc., 2003, 150,
C585–C590.
22 N. M. Deraz, Curr. Appl. Phys., 2012, 12, 928–934.
23 N. M. Deraz, Int. J. Electrochem. Sci., 2012, 7, 4608–4616.
2354 | RSC Adv., 2013, 3, 2351–2354
This journal is ß The Royal Society of Chemistry 2013