Nanoenergetic materials for MEMS: A review

Abstract

New energetic materials (EMs) are the key to great advances in microscale energy-demanding systems as actuation part, igniter, propulsion unit, and power. Nanoscale EMs (nEMs) particularly offer the promise of much higher energy densities, faster rate of energy release, greater stability, and more security (sensitivity to unwanted initiation). nEMs could therefore give response to microenergetics challenges. This paper provides a comprehensive review of current research activities in nEMs for microenergetics application. While thermodynamic calculations of flame temperature and reaction enthalpies are tools to choose desirable EMs, they are not sufficient for the choice of good material for microscale application where thermal losses are very penalizing. A strategy to select nEM is therefore proposed based on an analysis of the material diffusivity and heat of reaction. Finally, after a description of the different nEMs synthesis approaches, some guidelines for future investigations are provided.

Full text

Open Archive Toulouse Archive Ouverte (OATAO)
OATAO is an open access repository that collects the work of Toulouse researchers an
d
makes it freely available over the web where possible.
This is an author-deposited version published in: http://oatao.univ-toulouse.fr/
Eprints ID : 2460
To link to this article :
URL :
http://dx.doi.org/10.1109/JMEMS.2007.893519
To cite this version :
Rossi, Carole and Zhang, Kaili and Estève, Daniel and
Alphonse, Pierre and Tailhades, Philippe and Vahlas, Constantin ( 2007)
N
anoenergetic Materials for MEMS: A Review. IEEE/ASME Journal of
Microelectromechanical Systems, vol. 16 (n° 4). pp. 919-931. ISSN 1057-7157
Any correspondence concerning this service should be sent to the repository
administrator: staff-oatao@inp-toulouse.fr

JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007 919
Nanoenergetic Materials for MEMS: A Review
Carole Rossi, Kaili Zhang, Daniel Estève, Pierre Alphonse,
Philippe Tailhades, and Constantin Vahlas
Abstract—New energetic materials (EMs) are the key to great
advances in microscale energy-demanding systems as actuation
part, igniter, propulsion unit, and power. Nanoscale EMs (nEMs)
particularly offer the promise of much higher energy densities,
faster rate of energy release, greater stability, and more security
(sensitivity to unwanted initiation). nEMs could therefore give
response to microenergetics challenges. This paper provides a
comprehensive review of current research activities in nEMs for
microenergetics application. While thermodynamic calculations of
flame temperature and reaction enthalpies are tools to choose de-
sirable EMs, they are not sufficient for the choice of good material
for microscale application where thermal losses are very penal-
izing. A strategy to select nEM is therefore proposed based on an
analysis of the material diffusivity and heat of reaction. Finally, af-
ter a description of the different nEMs synthesis approaches, some
guidelines for future investigations are provided. [2006-0259]
Index Terms—Materials processing, materials science and tech-
nology, microelectromechanical devices, microelectromechanical
system (MEMS), micropyrotechnics, nanoscale energetic materi-
als (nEMs), propulsion, synthesis, thermal power generation.
I. INTRODUCTION
E
NERGETIC materials (EMs) are substances that store
chemical energy. Conventional EMs can be classified into
different classes, i.e., propellants, explosives, and pyrotechnics.
Propellants and pyrotechnics release their energy through rela-
tively slow deflagration processes that last for several seconds
(combustion). Explosives release their energy in fast detonation
processes (microsecond timescale). Basically, traditional EMs
are produced by the following.
1) The mixing of oxidizer and fuel constituents into one
molecule to produce monomolecular EMs (for example,
nitrocellulose, nitroglycerine, and trinitrotoluene).
2) The mixing of oxidizer powders (potassium or am-
monium nitrate or perchlorate, ...) and fuel powders
(carbon, sulfur, ...) to produce composite EMs (black
powder for example). These composites exhibit high en-
ergy density, but their energy release rates are slower than
monomolecular material because the mass transport rate
is limited by the granulometry of the reactants.
Manuscript received November 19, 2006; revised January 12, 2007. Subject
Editor S. M. Spearing.
C. Rossi, K. Zhang, and D. Estève are with Microsystems and Systems
Integration (MIS), Laboratoire d’Analyse et d’Architecture des Systèmes
(LAAS)–Centre National de la Recherche Scientifique, 31077 Toulouse Cedex
4, France (e-mail: rossi@laas.fr; kaili_zhang@hotmail.com).
P. Alphonse, P. Tailhades, and C. Vahlas are with Centre Interuniversitaire
de Recherche et d’Ingénierie des Matériaux (CIRIMAT), 31062 Toulouse
Cedex 4, France (e-mail: alphonse@chimie.ups-tlse.fr; chane@chimie.ups-
tlse.fr; tailhade@chimie.ups-tlse.fr; constantin.vahlas@ensiacet.fr).
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/JMEMS.2007.893519
Traditional EMs (monomolecular or composite) are rela-
tively easy to prepare (by mixing), and their performances can
be predicted and tailored by adjusting the stoichiometry of the
chemical reactants. For years, they have been widely used in
military, mining, and demolition applications. For two decades,
EMs have drawn a growing interest in the scientific community
and civil industry since they are very attractive sources of
onboard energy to generate gas, heat, and power. Typically,
the combustion of propellant produces 5 MJ/kg, while a mod-
ern chemical lithium battery used in new laptops only stores
0.5 MJ/kg. Therefore, EMs could have a major impact on mi-
croenergetic field. Applications include microthrust [1]–[10],
microinitiation [11]–[15], gases for actuation (including injec-
tion or moving fluid) [16]–[20], gases for chemical reaction
[21]–[24], heating (power) and welding [25], and switching
[26]. Various teams have successfully investigated chemical
and technological ways to tailor EMs to the targeted applica-
tions either to produce gas, heat, or chemical compound. An
important challenge that is not really addressed is the com-
patibility of the EMs with MEMS technologies. This implies
two exigencies. First, the thin film of EMs must be grown
or deposited at low temperature (below 250
◦
C). For some
applications for which the EMs film is deposited in freestanding
microstructures, its stress has to be controlled without any heat
treatments. Some well-known semiconductor processing tech-
nique such as PVD followed by liftoff [27], [28], screen printing
[1], or chemical reaction [13] have been already investigated
to prepare and deposit EMs on silicon surface. The second
exigency is to minimize heat losses which is one of the most
challenging issues to be solved in MEMS applications.
Despite the significant improvements that have been per-
formed through chemical formulation by combining known
chemical compounds/molecules [29], the traditional composite
and monomolecular EMs feature a relatively slow reaction rate
for microscale applications [30]. For example, HMX defla-
gration will quench in steel tubes of several millimeters in
diameter at atmospheric pressure. GAP/AP combustion will
quench in glass tubes of 1.4 mm in diameter at atmospheric
pressure [31]. To satisfy energy demanding systems of mi-
crometer size, more reactive EMs than traditional ones are
required. Significant efforts have been made to introduce metal
powder with micro and nanosized particles into traditional
EMs. Faster combustion velocities have been demonstrated
[32]–[34]. However, difficulties in handling such powders and
their incorporation into existing formulations have also been
reported. Besides traditional EMs, inorganic energetic compos-
ite combining metal oxidizer and metal also called metastable
intermolecular composite (MIC) are good candidates as well.
They undergo a solid-state redox reaction that is rapid and very
1057-7157/$25.00 © 2007 IEEE
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

920 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007
exothermic (nearly twice that of the best monomolecular EM).
First results demonstrated mainly empirically that the initiation
and combustion properties of EMs are strongly influenced by
their microscopical properties. They suggest that reducing the
particle size to the nanoscale may result in reduction of the
mass-transport rate and therefore would increase the burning
rates making these nanoscale EMs (nEMs) attractive alterna-
tives to monomolecular structures.
This paper presents a review of nEMs, a young but very
active field of research. To begin, we introduce the recently
synthesized nEMs and give the main performance reported by
the authors. A section is dedicated to MIC: it is an important
class of materials that combine excellent performances with in-
teresting synthetic approaches for MEMS compatibility. While
thermodynamic calculations of adiabatic flame temperature and
reaction enthalpies are tools to help the choice of desirable
EMs, they are not sufficient, in general, for the choice of
good material for microscale application where thermal losses
have to be taken into account. Section IV will discuss design
considerations for nEMs for pyrotechnic systems of micrometer
size. Section V will review the different approaches to the
synthesis of nEMs. An analysis of the compatibility of the
different synthesis routes with MEMS will be given. Lastly,
some guidelines for future investigations are provided in the
conclusions.
A list of abbreviations used in the review paper is given
in Table I.
II. R
EVIEW OF EMS FOR MICROSCALE APPLICATION
The first route explored by researchers was to adapt tradi-
tional EMs to microscale combustion by incorporating nanopar-
ticles of Al into traditional propellants or explosives. Doping
composite propellant with thermal conductive nanoscale parti-
cles has demonstrated a slight enhancement in the reactivity of
the mixture [31], [35]. Then, with the progress of nanoengineer-
ing, researchers preferred to combine or synthesize inorganic
energetic nanocomposite composed of particles of oxidizer
(typically metal oxide) and fuel (mainly Al) to produce MIC.
A. Nanoparticle-Doped Propellant
Nano-Al is the most widely used metallic doping particle for
the following reasons.
1) Al is a common metal used in technology and is relatively
low cost.
2) The formation of a thin oxide layer around each Al par-
ticle prevents Al powder from spontaneous combustion
[33], [36].
3) Al can be easily produced in nanoparticles form
(50–120 nm) and nano-Al is commercially available for
example from NovaCentrix, formerly known as Nano-
technologies, Inc.
4) Al enhances the reactive power of the material by in-
creasing the combustion velocity due to its high thermal
conductivity.
In 1997, Brown et al. [34] investigated the influence of the
incorporation of microsized Sb particles into KMnO
4
and found
TAB LE I
A
CRONYMS,ABBREVIATIONS, AND NOMENCLATURE
that reducing particle size from 14 to 2 µm increased the burn-
ing rate by a factor of 4. Ivanov and Tepper [37] reported that
by adding aluminium nanoparticles in a propellant formulation,
the mixture burning rate could be enhanced by a factor of
5–10. Chiaverini et al. [38] demonstrated an increase of 70%
in burning rate of HTPB-based solid propellant doped with
20% of nanoparticles of Al. Armstrong et al. [39] confirmed
that the burning rate of the AP-based propellants increases
when conventional Al powder is replaced by nano-Al. The
burning rate increases from 1 to more than 100 mm/s when the
aluminum particle size decreases from 10 µm to 100 nm. This
tendency has been confirmed for several types of propellants
(HTPB/AP, GAP/AP, and GAP/AN) by numerous authors and
studies [31], [40]–[43]. The ignition time is also reduced when
the propellant is doped with Al nanoparticles [41]. However,
for explosives, the use of metallic nanopowder has not proven
to be effective [43].
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

ROSSI et al.: NANOENERGETIC MATERIALS FOR MEMS: A REVIEW 921
Fig. 1. SEM image of micrometer-sized PETN grains among CNTs. Repro-
duced with permission from [44]. Copyright 2005 American Chemical Society.
In conclusion, doping propellant with Al nanoparticles im-
proves slightly its reactivity and ignition capability; the max-
imum burning enhancement seems to be reached when the
Al content represents 20% of the propellant mass. However,
recent studies reveal that the incorporation of nanosized metal
particles into propellants does not increase the reaction temper-
ature and burning rate sufficiently to compensate the thermal
losses occurring in microscale devices. These nano-Al doped
propellants suit mesoscale applications but are not applicable
for microscale applications. The quenching diameter is around
the millimeter, and the combustion is unstable in the centimeter
scale. Furthermore, some authors have reported [40] that an AP-
based propellant with pure nano-Al has unstable and nonrepro-
ducible combustion because of the difficulty to uniformly mix
the ultrafine Al particles into the propellant. Other approaches
have been therefore considered. One is the incorporation of
carbon nanotubes (CNT), having very high thermal conductiv-
ity, into traditional propellant, and the other is developed by
Pivkina et al. to produce nanopowder of AN and RDX. Very
recently, Manaa et al. [44] presented a study of ignition and
combustion of explosive-nanotube mixture. They incorporated
single-wall CNT produced by Carbon Nanotechnologies, Inc.
inside a PETN EM and demonstrated that the mixture (see
Fig. 1) can burn rapidly with local temperature estimated be-
tween 1500
◦
C–2000
◦
C. Shock wave with an average speed of
6.8 km/s has been measured.
According to the authors, in a longer term, there is hope
that CNTs could be used to encapsulate energetic ingredi-
ents/molecules to produce an EM that not only has good
energetic performance but also has much improved handling
and long-term storage capabilities.
An interesting vacuum codeposition technique was presented
by Pivkina et al. to synthesize nanostructured AN, RDX,
and AN/RDX composites [4], [45]. The typical process of
synthesizing nano-RDX and AN/RDX consists of realizing
nanopowder of the constituent (AN, RDX) by vacuum con-
densation of evaporated pure bulk substrates onto the cooled
substrates to obtain AN and RDX powder with 50 nm in
diameter. Conventional AN and RDX powder with 200 and
50 µm in diameter, respectively, have been compared with
nano-AN and RDX powder with 50 nm in diameter. According
to the results, the burning rate can be doubled for nanoscale
RDX: 15.1 mm/s for the conventional RDX and 30 mm/s for the
nanoscale one. Pure AN could not have a sustained combustion;
only thermal analysis has been performed and showed that the
temperature of maximun heat release for AN nanopowder is
less than that for conventional one. Vasylkiv et al. [23], [24]
proposed nanoreactors produced by nanoblasts impregnated
with particles of C
3
H
6
N
6
O
6
. The described technique opens
the door to the synthesis of a wide range of multimetal oxide
ceramic and metal-ceramic composite nanopowders with pre-
cise stoichiometries and uniform morphologies.
B. Nanoscale Thermite Material or MIC
Thermite reaction is a highly exothermic reaction that in-
volves a metal reacting with a metallic or a nonmetallic oxide
to form a stable oxide and the corresponding metal or nonmetal
of the reactant oxide [46]. This is a form of oxidation–reduction
reaction that can be written as
M + AO → MO + A +∆H
where M is a metal or an alloy, A is either a metal or a nonmetal,
MO and AO are their corresponding oxides, and ∆H is the heat
of reaction. The thermite reactions exhibit fast reaction rates
that make their use extremely energy efficient.
Nano-Al is most widely used as the fuel for the reasons
detailed in Section II-A plus additional advantages for thermite
reactions.
1) Al has low vapor pressure: Al, unlike calcium or magne-
sium, does not require a pressure-tight reaction vessel for
the reaction.
2) Al melting temperature is low (approximately 660
◦
C)
resulting in a low ignition temperature.
Among a large number of possible oxidizers [52], only a
few have been investigated by different teams: Fe
2
O
3
,MoO
3
,
KMnO
4
,CuO,NiO,MnO
2
,WO
3
, SnO
2
, and SiO
2
.
In 1995, Aumann et al. [33] produced MoO
3
/Al MIC being
20–50 nm in diameter. When the mixture is stoichiometric,
the energy density reached 16 kJ/cm
3
, and the mixture can
burn 1000 times faster than macroscale thermite material.
Bockmon et al. [47] compared different samples of Al/MoO
3
MIC fabricated using ultrasonic mixing where nano-Al was
used. The measured average combustion velocity increases
from approximately 685 to 990 m/s when the Al particle size
is decreased from 121 to 44 nm. As a comparison, micrometer
scale Al/MoO
3
burns at 10 mm/s [34]. Authors also note that
the reaction rate becomes independent of Al particle diameter
below a critical diameter, which is 40 nm in their case. This
could be explained by the fact that, when the Al particle
diameter decreases, the proportion of Al
2
O
3
over Al increases
resulting in a reduction of the volume of active material and
possible inhibition of the thermite reaction.
Granier and Pantoya [48], [49] observed that the burning
rate of nano-Al/MoO
3
MIC is enhanced by a factor of 10
when the average Al particle size decreases from 20 000 to
50 nm. Moreover, they noted that by decreasing the Al particle
size from micrometer to nanometer scale in the composite
consistently decreases the ignition time by up to two orders
of magnitude (from 6 s down to 20 ms) and improves the
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

922 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007
repeatability of the composite’s response to ignition. The in-
creased sensitivity to ignition may be explained by the re-
duction in melting temperature associated with nanoparticles.
Sun et al. [50] studied the reactions of nano-Al with O
2
with the
average Al particle diameters ranging from 30 to 160 nm. It was
shown that Al/O
2
reacts below the Al melting temperature. The
heat released by the reaction (approximately 10 kJ/g) is below
the theoretical one (25 kJ/g). Bhattacharya et al. [51] investi-
gated the burning rate of nanoscaled CuO/Al and Bi
2
O
3
/Al.
The burning rate reaches the average values of 440 and 150 m/s
for nanoscaled CuO/Al and Bi
2
O
3
/Al, respectively.
Results of empirical studies reported in Section II establish
clearly that the initiation temperature, reaction properties, and
propagation rate are strongly influenced by the microscopical
properties of the EMs, including the size of constituents and
intimacy of the contact.
III. S
TRATEGY TO DESIGN nEMs FOR
MICROSCALE APPLICATION
Basically, the synthesis of EMs and the choice of mole-
cules or compounds are mainly guided by molecular model-
ing and thermochemical codes. This approach has proven its
efficiency for traditional EMs and macroscale/mesoscale ap-
plications, and significant improvements have been performed
through chemical formulation, i.e., combining known chemical
compounds/molecules. However, for microscale applications,
besides the optimization of energy density given by thermo
chemical codes, other considerations have to be taken into
account to design novel nEMs.
1) The rate of reaction has to be enhanced to minimize heat
losses, not only by increasing the heat of reaction.
2) The technological process has to be controlled to adapt
the stoichiometry of reactants, the size of particles, the
purity of the final mixture, and the intimacy of the contact
between the fuel/oxidizer.
3) The contribution of the mass diffusion rate has to be
evaluated since it could limit the combustion rate.
For the first point, we propose a theoretical evaluation that is
based on the diffusivity parameter, to guide the selection of any
EMs for microscopic application. The second point, which is
related to the structural control of the material, is very difficult
to be theoretically predicted yet, and it is mainly investigated
empirically by numerous research teams. Main results will be
presented and discussed in Section IV. For the third point, a
simple intuitive reasoning can estimate the scale of the particle
size below which diffusion process can be neglected. For that,
we consider a spherical particle coated with a thin oxide layer
that burns from the external surface to the center, as shown in
Fig. 2. The diameter of the particle decreases over the time as
follows π × d
2
= r × t with r as the combustion rate (mm
2
/s),
and d as the particle diameter. Therefore, with r =1.5 mm
2
/s
(which is the low limit), particles of 100 nm in diameter
would be consumed in 0.01 µs. This time is sufficiently short
compared to mass diffusion delay
1
to consider the mass dif-
1
Mass diffusion time of a particle of Al of 100 nm in diameter: hundreds of
microns.
Fig. 2. Schematic of the combustion of a nanoparticle coated with a
thin oxide.
fusion process negligible for particles having diameter lower
than 100 nm.
For the rest of the section, the mass diffusion phenomena
will be neglected, and three parameters will be considered to
propose a classification of nEMs: the thermal diffusivity α,the
heat of reaction ∆H, and the adiabatic temperature T
ad
.
1) Diffusivity: α = k/(ρ × Cp) where k is thermal con-
ductivity, and ρ × Cp is the volumetric heat capacity.
Substances with high thermal diffusivity conduct heat
quickly. Tables II and III give the calculated diffusiv-
ity for well-known organic EMs and seven thermite
EMs: Al/Fe
2
O
3
, Al/MoO
2
, Al/CuO, Al/NiO, Al/SnO
2
,
Al/SiO
2
, and Al/TiO
2
.
2) Heat of reaction: for each thermite couple, thermody-
namic calculations of heat of reaction (∆H) and adi-
abatic flame temperature (T
ad
) are given in Table IV.
The values of T
ad
are from [52]. For the purpose of
comparison, the values of ∆H are from COACH [53],
which is an in-house code, and [52], respectively.
3) The adiabatic reaction temperature given in Tables II
and IV is the temperature that results from a complete
combustion process that occurs without any heat transfer
or changes in kinetic or potential energy.
A. Calculation of the Diffusivity of Thermite Materials
We suppose that the aluminum (Al) and metallic oxide
(MO)
2
are mixtures of nanoscale powders with d
Al
and d
MO
being the Al and MO particle diameter. We also assume that
the Al-to-MO mass ratio (m
Al
/m
MO
) is the stoichiometric one
for each reaction. The thermal conductivity of Al/MO MIC at
300 K is a function of the diameter of the Al and MO particles
as well as thermal conductivities of both Al and MO. In a first
approximation, we suppose that the MIC that is composed of
Al and MO particles in intimate contact can be modeled as an
AL/MO bilayer material. The Al and MO layer thickness would
be equal to the Al and MO particle diameter.
The resulting bilayer thermal conductivity can be therefore
calculated as the following:
k
Al/MO
=
k
Al
k
MO
×

1
d
Al
+
1
d
MO

k
Al
d
Al
+
k
MO
d
MO
2
MO is Fe
2
O
3
,MoO
2
,CuO,NiO,SnO
2
,SiO
2
,TiO
2
.
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

ROSSI et al.: NANOENERGETIC MATERIALS FOR MEMS: A REVIEW 923
TAB LE II
D
IFFUSIVITY,HEAT OF REACTION, AND ADIABATIC TEMPERATURE FOR THREE DIFFERENT ORGANIC EMs
TABLE III
D
IFFUSIVITY FOR SEVEN DIFFERENT MICs CALCULATED FOR THREE FUEL-OVER-METALLIC OXIDE DIAMETER RAT IOS
TAB LE IV
S
TOICHIOMETRIC MASS RATI O,HEAT OF REACTION (∆H), AND ADIABATIC TEMPERATURE (T
ad
) FOR SEVEN DIFFERENT MICs
with k
Al
and k
MO
as the thermal conductivities at 300 K of Al
and MO, respectively, and d
Al
and d
MO
as the diameters of Al
and MO particles, respectively.
The heat capacity at 300 K of Al/MO can be calculated as
the following:
Cp
Al/MO
=
m
Al
× Cp
Al
+ m
MO
× Cp
MO
m
Al
+ m
MO
with m
Al
and m
MO
as the masses of Al and MO contained
in the MIC, respectively, and Cp
Al
and Cp
MO
as the heat
capacitances of Al and MO, respectively.
B. Results and Comments
In Fig. 3, it can be clearly shown that the Al/MO thermite
diffusivities decrease when the MO-to-Al particle diameter
ratio increases. In consequence for the rest of the study, d
Al
is taken which is equal to d
MO
.
Fig. 4 classifies the different EMs in a 2-D diagram: heat
of reaction in y-axis and diffusivity in x-axis. The graph
indicates that any EMs placed in the left bottom corner of
the graph could not suit microscale application because of
too low heat of reaction and insufficient diffusivity. However,
the EMs in the upper right parts of the diagram are con-
sidered to be well adapted to MEMS because they release
high amount of heat, and the rate is expected to be high. To
conclude this short study, among investigated MICs, the best
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

924 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007
Fig. 3. Example of diffusivity evolution as a function of the MO-to-Al
diameter ratio for three MICs: Al/Fe
2
O
3
, Al/NiO, and Al/MnO
2
.
Fig. 4. Classification of EMs in a 2-D diagram (∆H , α).
ones are: Al/MnO
2
, Al/Fe
2
O
3
, and Al/NiO. From Fig. 4, MICs
appear to be a promising category of nEMs that could be a
good candidate to realize thin layer of EMs for microscale
applications since they have excellent diffusivity, and they can
be very reactive compared with traditional organic EMs, and
also, they can exhibit high exothermicity. For organic EMs, the
best one appears to be PETN that nevertheless presents poor
diffusivity.
IV. S
YNTHESIS OF nEMs FOR
MICROSCALE APPLICATIONS
There are different approaches to synthesize nanosized ther-
mite composite or MIC. Basically, they can be classified into
traditional approaches, which include MEMS compatible ap-
proaches, nanopatterning, and “bottom-up” or molecular engi-
neering approaches. In traditional approaches, the nanoparticles
of fuel and oxidizer are first synthesized or purchased and
then mixed together, i.e., powder mixing. In MEMS com-
patible approaches, the fuel/oxidizer molecules or particles
are synthesized and put together in a single process such as
sol–gel chemistry and vapor deposition (VD) techniques. In
molecular engineering approaches, the atoms or molecules of
fuel and oxidizer are first synthesized and then combined using
molecular engineering in solution or via a polymer chain. The
“bottom-up” approaches are mainly investigated with the aim at
developing a fundamental understanding of the evolution of the
EMs properties with the size and intimacy of the constituents.
The main synthesis approaches of MIC are presented and
discussed in the following.
Fig. 5. SEM images of 80-nm Al particles mixed with MoO
3
particles.
Reproduced with permission from [48]. Copyright 2004 Elsevier.
A. Powder Mixing
The easiest method is the physical powder mixing. In recent
works, the ultrasonic mixing is mostly employed to com-
bine nano-Al and oxidizer powder [40], [47]–[50], [54]–[56].
Typically, the nano-Al and oxidizer are dispersed in solvents
(e.g., hexane) and mixed together with ultrasonic wave. The
sonication process is used to break up agglomerates and mix
both constituents. After sonication, the mixture is heated to
evaporate the solvent. The mixture is often passed through
a fine mesh to break any agglomerates formed during the
evaporation process and produce submicronic powder. Fig. 5 is
a representative SEM image of the Al/MoO
3
MIC which, thus,
obtained mixture.
Mixing nanopowders of fuel with oxidizer is a simple
method. Nevertheless, it presents limitations. First, one is the
difficulty to mix intimately ultrafine powder, the second one is
the difficulty to obtain a homogeneous distribution of oxidizer
and fuel nanoparticles, and thirdly, the manipulation of some
powder can be dangerous, and the deposition in thin film or
into a microsystem is difficult [30].
B. Sol–Gel/Aerogel Process
A wide variety of glass or glass–ceramic monoliths and
nanostructured powders are synthesized through sol–gel tech-
nique, since it has the advantages of low temperature process-
ing, high chemical homogeneity, etc. Researchers at Lawrence
Livermore National Laboratory have first introduced the use of
the sol–gel chemistry to synthesize nEMs [57], [58].
The sol–gel process involves reactions in solution to produce
dispersion of nanoparticles in a liquid phase, called “sol”
(colloidal solution). By condensation, the sol gives a 3-D solid
network, called a “gel,” with the open pores being occupied
by the solvent. Solvent removal by evaporation produces the
collapse of the open pore structure of the gel and leads to
a xerogel. Supercritical drying allows removing the solvent
without collapsing the gel structure. This leads to a highly
porous and lightweight material called aerogel, with excellent
uniformity given that the particles and the pores are both in
nanoscale. The sol–gel process is illustrated in Fig. 6.
Tillotson et al. [58] reported about successful sol–gel syn-
thesis of Fe
2
O
3
/Al MIC. It consists of clusters of Fe
2
O
3
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

ROSSI et al.: NANOENERGETIC MATERIALS FOR MEMS: A REVIEW 925
Fig. 6. Schematic diagram of sol–gel methodology.
Fig. 7. Al/Fe
2
O
3
MIC prepared by sol–gel process. Reproduced with permis-
sion from [59]. Copyright 2002 AMMTIAC.
being 3–10 nm in diameter, which are in contact with nano-
Al particles of about 30 nm in diameter, as shown in Fig. 7.
The obtained specific surface of the Fe
2
O
3
nanoporous matrix
is around 300–400 m
2
/g. The authors proceeded to standard
safety tests (spark, friction, ...) showing a relatively insensi-
tivity. Ignition and thermal analysis gave a heat of reaction of
1.5 kJ/g, whereas the theoretical value is 3.9 kJ/g. The authors
suggested that the reasons for reduction of the total measured
energy could be the substantial passivation of Al powder by a
native oxide coating and the presence of organic impurities that
could represent 10% in mass of the sample.
The sol–gel chemistry provides process advantages over
mixing methods: it is MEMS compatible, low cost, safe, and
can potentially produce nEMs with interesting and special prop-
erties due to the high porosity. Sol–gel process can also offer
good opportunity to tailor the nEMs properties by controlling
the process. However, two disadvantages can be noted: First is
the random distribution of the particles that can lead to locally
a separation between the fuel and oxidizer and, therefore,
inhibition of self-sustaining reaction; Second is that sol–gel
mixture often has organic impurities that can reach up to 10%
of the sample mass leading to a significant reduction of the
reaction performance.
Besides the sol–gel chemistry used in [57], [58], and [60],
the aero-sol–gel chemistry was also employed to synthesize
nano-KMnO
4
,CuO,Fe
2
O
3
, and the corresponding MICs with
nano-Al as the fuel [55], [61]–[63]. Clapsaddle et al. [63], for
example, reported that the particle size and properties of the
aerogel preparation can be tailored easily by controlling the
epoxide addition procedure.
Prakash et al. [64] reported a novel aerogel-based process
to create a pure nanosized MIC core-shell nanostructure that
is obtained by coating a strong oxidizer nanoparticle with a
thin layer of relatively mild oxidizer. The authors use a two-
temperature aerosol spray-pyrolysis method, as shown in Fig. 8.
The core is KMnO
4
, and the shell is Fe
2
O
3
. In this process,
an aqueous solution of Fe(NO
3
)
3
9H
2
O and KMnO
4
is sprayed
into droplets (1 µm in diameter). The environment is dried. The
aerosol is then thermal cured. First, it is maintained above the
decomposition temperature of iron nitrate (typically 120
◦
C),
and then, the temperature is increased to about 240
◦
C (approxi-
mately melting point of permanganate). The 0.6-µm composite
particles are then collected on a filter. At 120
◦
C, iron nitrate
decomposes into Fe
2
O
3
enrobing with intimately the solid
permanganate. At 240
◦
C, the permanganate melts, and the
Fe
2
O
3
aggregates around the KMnO
4
particles. The resulting
composite nanoparticle of KMnO
4
(150 nm in diameter) coated
with Fe
2
O
3
(4-nm thick) is shown in Fig. 9. Authors performed
reactivity experiments and announced that the reactivity can
be moderated over a relatively large range by changing the
thickness of the iron oxide. Besides the resulting intimate
contact between the reactants, the oxide coating thickness on
each nanoparticle is nanoscale and could be precisely controlled
to lower the ignition temperature and also not to inhibit the
thermite reaction.
Prakash’s process is interesting to precisely investigate the
influence of the nEMs structural properties (oxide thickness
for example) on its reactivity. However, it is complex and not
currently applicable for mass production process.
C. Vapor Deposition
An alternative route to sol–gel/aerogel chemistry is to deposit
alternatively a nanolayer of oxidizer and fuel under vacuum
by vapor deposition or sputtering. VD is a chemical process
often used in the semiconductor industry for the deposition of
thin films of various materials. In a typical VD process, the
substrate is exposed to one or more volatile precursors, which
react and/or decompose on the substrate surface to produce
the desired deposition. Basically, the resulting layer’s thickness
ranges from 20 nm to 2 µm. Sputtering is a physical process
whereby atoms in a solid target material are ejected into the gas
phase due to bombardment of the material by energetic ions. It
is also commonly used for thin-film deposition.
VD technique has been employed to deposit PETN thin film
by Tappan et al. [27], [28]. Vine et al. [65] reported the VD
of Ti/C and Ti/Al multilayers. An increase in the burning rate
has been found for multilayered composition compared with
the conventional one. Blobaum et al. [66] carried out a study on
Al/CuOx multilayer nanofoils deposited by magnetron sputter-
ing onto silicon substrates. The process of deposition is simple:
a CuO target is RF sputtered on silicon in alternance with Al
target. To avoid metal interdiffusion, the silicon wafer is cooled
down during the deposition process. The deposited composi-
tions exhibit a layered structure Al (0.3 µm)/Cu
4
O
3
(0.7 µm),
as shown in Fig. 10. The measured heat of reaction of the
fabricated nanofoils is 3.9 ± 0.9 kJ/g that is similar to that
calculated for the reaction of Al and CuO, as listed in Table IV.
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

926 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007
Fig. 8. Aerosol system for the synthesis of core-shell nanocomposite oxidizer. Reproduced with permission from [64]. Copyright 2005 American Chemical
Society.
Fig. 9. TEM graph and STEM elemental map of nano-KMnO
4
coated with
Fe
2
O
3
. Reproduced with permission from [64]. Copyright 2005 American
Chemical Society.
Fig. 10. TEM image of Al/CuOx multilayer nanofoils. Reproduced with
permission from [66]. Copyright 2003 American Institute of Physics.
An important dependence of the burning rate with the layer
thickness and intermixing conditions has been noticed.
Gavens et al. in 1999 [67] has proposed Al/Ni nanofoils
deposited using high vacuum dc and RF sputtering. The total
thickness of the multilayers foils is 11 µm. Each layer thickness
ranges from 12.5 to 200 nm. Then, the foils are annealed to
intermix the fuel and the oxidizer by atomic interdiffusion. The
rate and heat of reaction are measured for different bilayer
thicknesses and intermixing conditions. The 200-nm-thick bi-
layer has a burning rate of 1 m/s. Authors mention that when
the bilayer thickness decreases, the burning rate increases and
then drops to zero. For a thick layer (200 nm), the intermixing
conditions are negligible, whereas for a thin layer, the reaction
rate is controlled by the thickness of the intermixed region.
In a fuel/oxidizer multilayer foil deposited by vapor deposi-
tion method, the diffusion distances between fuel and oxidizer
are reduced 10–1000 times compared to powders, thereby
enhancing atomic mixing and reactivity. The presence of im-
purities and Al oxidation are much smaller in multilayer foils
than in sol–gel or powder mixing process because the foils are
deposited in high vacuum.
D. Nanostructuring
1) Atomic Layer Deposition (ALD): Among investigated
nanostructuring methods, ALD is an ideal process to deposit
ultrathin films with high conformal uniformity and precise
thickness control. ALD utilizes sequential precursor gas pulses
to deposit a film one layer at a time. ALD has been already
utilized to deposit many metal oxides as WO
3
[68], Co
3
O
4
[69], and NiO [70]. The maximum thickness is a few tens
of nanometers. Ferguson et al. [71] employed it to deposit
SnO
2
oxidizer coatings on nanoparticles of Al. The Al/SnO
2
MIC prepared by ALD process is shown in Fig. 11. The first
experimentations on the Al−SnO
2
samples show a quick and
violent reaction despite the low Sn percent by mass and low
Al/O molar ratio.
2) Nanoporous Silicon With Oxidizer: Porous silicon (PSi)
has been known as a reactive material when McCord et al.
[72] discovered a combustion reaction of PSi immersed in nitric
acid. The exothermic reactions of PSi with liquid oxygen and
gadolinium nitrate were then presented in [73] and [74]. The
PSi pores (2–10 nm) are filled with a liquid oxidizer. Because
of these findings, PSi-based nEMs were proposed for industrial
applications as a component of an airbag initiator [13], [14].
The PSi (Fig. 12) can be produced by electrochemical etch-
ing of bulk silicon in solutions containing fluoride (e.g., HF)
and H
2
O
2
[75]. The porosity and the structure size of the
PSi can be adjusted from 2 to 1000 nm by selecting suit-
able etching parameters such as fluoride concentration, current
density, treatment duration, and starting material [14]. The
nanopores are then filled with a liquid oxidizer (Ca(ClO
4
)
2
,
KClO
4
,NaClO
4
, Sulfur, etc.). To enhance the intimacy between
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

ROSSI et al.: NANOENERGETIC MATERIALS FOR MEMS: A REVIEW 927
Fig. 11. Al/SnO
2
MIC prepared by ALD process. Reproduced with permis-
sion from [71]. Copyright 2005 Elsevier.
Fig. 12. SEM images of nano-PSi. (a) Cross section. (b) Top view. Re-
produced with permission from [75]. Copyright 2000 American Institute of
Physics.
the Si and O
2
, the PSi matrix can also be filled with the
oxidizer by CVD and PVD. The activation energy of the PSi/O
2
reaction can be tailored by controlling the SiO
2
native layer
before the oxidizer loading. The results are very promising:
Laucht et al. [13] tested PSI/Ca(ClO
4
)
2
and PSI/Na(ClO
4
)
2
in a calorimetric bomb. Authors noted heats of reaction of
8.6 kJ/g and 9.2 kJ/g for PSI/Ca(ClO
4
)
2
and PSI/Na(ClO
4
)
2
,
respectively, which are well above that of ZPP (6.3 kJ/g).
One particular advantage of PSi-based nEMs is that silicon
is one of the base materials for MEMS. Therefore, the nEMs
can be conveniently incorporated into silicon-based MEMS
devices. This provides the advantage of low cost production and
MEMS compatibility. In particular, the PSi-based nEMs can be
easily integrated into semiconductor circuitries.
3) Other Approaches: Menon et al. [76] realized Al/Fe
2
O
3
MIC by embedding an array of Fe
2
O
3
nanowires inside a
thin Al film, as described in Fig. 13(a). The process started
with electrochemical anodization of Al foil to form nanoporous
alumina templates [77], [78]. Then, Fe nanowire was fabricated
inside the nanopores by electrochemical deposition. After par-
tial etching of the alumina walls from the top, Fe nanowires
were revealed and then oxidized. Finally, the remaining alu-
mina walls were etched away, and at this stage, all of the Fe
Fig. 13. (a) Schematic of Al/Fe
2
O
3
MIC. (b) SEM image of Al/Fe
2
O
3
MIC.
Reproduced with permission from [76]. Copyright 2004 American Institute of
Physics.
Fig. 14. TEM images of Al/Fe
2
O
3
MIC by (a) random assembly and
(b) electrostatic assembly. Reproduced with permission from [79]. Copyright
2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.
nanowires were converted into Fe
2
O
3
nanowires by annealing.
A typical SEM image of a high-density nanowire array embed-
ded in Al film is shown in Fig. 13(b). The diameter of a single
nanowire is about 50 nm, and the density of nanowires for all
observed specimens is about 10
10
wires/cm
2
.
The main advantage of this approach is the precise control of
oxidizer and fuel dimensions at nanoscale. The method avoids
both the incorporation of impurities and the oxidation of the Al
at the fuel-oxidizer interface prior to ignition.
E. Molecular Engineering or “Bottom-Up” Approaches
Molecular engineering has been investigated very recently by
only a few teams to control the fuel/oxidizer interfacial contact
area and improve the homogeneity of the final material.
One interesting technique is the molecular self-assembly. It
means that the fuel nanoparticles will arrange by themselves
or enhanced by external forces, in a controlled manner around
the oxidizers or vice versa. Fuel or oxidizer molecules can
be arranged in inorganic solution or manipulated by using
a polymer. Kim and Zachariah [79] presented a method to
synthesize MIC in which the Al and Fe
2
O
3
self-assembly
is controlled by the electrostatic forces which exist between
charged aerosol particles. The authors compared the reactivity
of resulting Al/Fe
2
O
3
with a randomly assembled Al/Fe
2
O
3
.
As shown in Fig. 14(b), the nano-Al particles surround perfectly
the surface of Fe
2
O
3
particles and the contact between Al and
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

928 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007
Fig. 15. Schematic of nano-Al self-assembly around CuO nanorods. Reproduced with permission from [80]. Copyright 2006 Materials Research Society.
Fig. 16. TEM image of nano-Al assembly around CuO nanorods. Reproduced
with permission from [80]. Copyright 2006 Materials Research Society.
Fe
2
O
3
is intimate. Fig. 14(a) is a TEM image of nano Al/Fe
2
O
3
synthesized by random Brownian assembly. It can be seen that
nano-Al particles form linear chains in contact with the Fe
2
O
3
particles. The authors’ results showed that Al/Fe
2
O
3
MIC ma-
terials electrostatically assembled are more powerful than those
produced via random assembly due certainly to the enhanced
fuel/oxidizer interfacial surface. The total measured heat of
reaction is 1.8 kJ/g against 0.7 kJ/g for a random assembly that
remains lower than the theoretical heat of reaction of Al/Fe
2
O
3
MIC (3.9 kJ/g).
Subramaniam et al. [80] employed a polymer chain on which
the nanoparticles (CuO and Al) are binded, as shown schemat-
ically in Fig. 15. The polymer is therefore a structural link
for the nanoparticles assembly process. In a typical process,
CuO nanorods are first prepared from a chemical route (by
surfactant templating method). Then, the CuO nanorods are
coated with the poly(4-vinyl pyridine) polymer from Aldrich
in a sonic bath. After that, the CuO nanorods are removed
from the solution, washed, dried, and mixed with nano-Al
particles (80 nm in diameter with 2 nm of passivation layer).
The resulting Al/CuO MIC is shown in Fig. 16. The highest
burning rate measured by the authors is 2000 m/s for Al/CuO
MIC synthesized by the polymer-based nano-self-assembly.
LAAS–Centre National de la Recherche Scientifique and
CIRIMAT are exploring an innovative approach to realizing
MIC directly on silicon making it very easy to integrate MIC
into MEMS devices. First, metal oxide foam is achieved by
self-assembly of primary individual metal-oxide nanoparticles
onto silicon substrate. The self-assembly results in nanopores
inside the metal oxide foam. Then, nano-Al is deposited inside
the nanopores so as to have an intimate contact between Al and
metal oxide.
V. C
ONCLUSION
The R&D of nano-EMs is a quite young field but is very
active, and promising results could lead to interesting break-
through in the field of microenergetic devices. For years, differ-
ent synthesis approaches to enhance the performances of EMs
and make them applicable to microscopic systems have been
investigated and reported in this review.
The first approach consists of doping traditional EMs (HTBP,
PETN, RDX, ...) with nanoparticles of Al and, more recently,
CNT. The results are quite interesting but suffer from two main
disadvantages. Besides the difficulty to get a final homogeneous
mixture, the procedure of elaboration can be hazardous, and the
finally obtained material has to be manipulated and reported on
the initiator which is, if possible, very difficult.
The second approach widely investigated consists of elabo-
rating MICs that exhibit very powerful reaction mainly because
of its high redox capability and high thermal diffusivity. Differ-
ent nano-based processes have been investigated to reduce the
diffusion distances between reactant species and increase the
surface-over-volume ratio. These MICs can be synthesized in
different ways.
Powder mixing is a widely used method but not really
applicable to microenergetic applications. Sol–gel is an inter-
esting approach mainly because it presents process advantages
in terms of homogeneity, cost, and safety. Fabrication of the
nano-PSi-based EM is also very innovative and promising
because it is totally integratable into standard semiconductor
technologies. In particular, the nano-PSi-based EMs can be
easily integrated on a silicon chip.
Modern technologies as nanopatterning provide nice oppor-
tunities to get a new class of tailored EMs. The elaboration
of energetic multilayer nanofoils from PVD or sputtering is
also a promising approach for the integration of nano-EMs
within microsystems because of the compatibility with mass
production process, the good quality of the fabricated layer, and
the possibility to control precisely each metal and oxide thick-
ness layer and interdiffusion region. More recently, a couple
of “bottom-up” approaches have been investigated and seem to
provide a good means to better understand the structural mech-
anisms governing the EMs thermal and dynamic properties.
This “bottom-up” route would enable molecularly manipulated
energetic substances and formulations having tailored chemical
and physical properties, high performance, low sensitivity, and
multifunctional capabilities [16], [21], [81]. This is a good route
to access to a fundamental understanding of the evolution of the
EMs properties with the size and intimacy of the constituents.
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

ROSSI et al.: NANOENERGETIC MATERIALS FOR MEMS: A REVIEW 929
Although there has been considerable success in formulating
new MIC with enhanced energy-release rates, the subject of
achieving a precise control over the reactivity of nEMs is an
opportunity for further research. To achieve a better understand-
ing of nEMs properties, progress in computational method and
tools is requested to assess the reactivity to predict the stability
of these new classes of nEMs. Fundamental computations
aimed toward the simulation of the behavior of nEMs are in
the primary stage.
R
EFERENCES
[1] C. Rossi, D. Briand, M. Dumonteuil, T. Camps, P. Q. Pham, and N. F.
de Rooij, “Matrix of 10 × 10 addressed solid propellant microthrusters:
Review of the technologies,” Sens. Actuators A, Phys., vol. 126, no. 1,
pp. 241–252, Jan. 2006.
[2] D. W. Youngner, S. T. Lu, E. Choueiri, J. B. Neidert, R. E. Black, III,
K. J. Graham, D. Fahey, R. Lucus, and X. Zhu, “MEMS mega-pixel
micro-thruster arrays for small satellite stationkeeping,” presented at the
14th Annu. AIAA/USU Conf. Small Satellites, Logan, UT, Aug. 21–24,
2000, AIAA Paper SSC00-X-2.
[3] D. H. Lewis, Jr., S. W. Janson, R. B. Cohen, and E. K. Antonsson, “Digital
micropropulsion,” Sens. Actuators A, Phys., vol. 80, no. 2, pp. 143–154,
Mar. 2000.
[4] D. Teasdale, V. Milanovic, P. Chang, and K. Pister, “Microrockets for
smart dust,” Smart Mater. Struc., vol. 10, no. 6, pp. 1145–1155, 2001.
[5] W. Lindsay, D. Teasdale, V. Milanovic, K. Pister, and C. F. Pello, “Thrust
and electrical power from solid propellant microrockets,” in Proc. Tech.
Dig. 14th IEEE Int. Conf. MEMS, Piscataway, NJ, 2001, pp. 606–610.
[6] K. Takahashi, H. Ebisuzaki, H. Kajiwara, T. Achiwa, and K. Nagayama,
“Design and testing of mega-bit microthruster arrays,” presented at the
Nanotech, Houston, TX, Sep. 9–12, 2002, Paper AIAA 2002-5758.
[7] S. Tanaka, R. Hosokawa, S. Tokudome, K. Hori, H. Saito, M. Watanabe,
and M. Esashi, “MEMS-based solid propellant rocket array thruster with
electrical feedthroughs,” Trans. Jpn. Soc. Aeronaut. Space Sci., vol. 46,
no. 151, pp. 47–51, 2003.
[8] P. Q. Pham, D. Briand, C. Rossi, and N. F. De Rooij, “Downscaling of
solid propellant pyrotechnical microsystems,” in Proc. 12th Int. Conf.
Solid-State Sensors and Actuators (Transducers), Boston, MA, Jun. 8–12,
2003, pp. 1423–1426.
[9] Y. Zhao, B. A. English, Y. Choi, H. DiBiaso, G. Yuan, and M. G. Allen,
“Polymeric microcombustors for solid-phase conductive fuels,” in Proc.
17th IEEE Int. Conf. MEMS, Maastricht, The Netherlands, Jan. 25–29,
2004, pp. 498–501.
[10] K. L. Zhang, S. K. Chou, S. S. Ang, and X. S. Tang, “A MEMS-based solid
propellant microthruster with Au/Ti igniter,” Sens. Actuators A, Phys.,
vol. 122, no. 1, pp. 113–123, Jul. 2005.
[11] T. Troianello, “Precision foil resistors used as electro-pyrotechnic initia-
tors,” in Proc. 1st Electron. Compon. and Technol. Conf., Orlando, FL,
May 29–Jun. 1, 2001, pp. 1413–1417.
[12] H. DiBiaso, B. A. English, and M. G. Allen, “Solid-phase conductive fuels
for chemical microactuators,” Sens. Actuators A, Phys., vol. 111, no. 2/3,
pp. 260–266, Mar. 2004.
[13] H. Laucht, H. Bartuch, and D. Kovalev, “Silicon initiator, from the idea
to functional tests,” in Proc. 7th Int. Symp. and Exhib. Sophisticated Car
Occupant Safety Syst., 2004, pp. 12–16.
[14] A. Hofmann, H. Laucht, D. Kovalev, V. Y. Timoshenko, J. Diener,
N. Kunzner, and E. Gross, “Explosive composition and its use,” U.S.
Patent 6 984 274, Jan. 10, 2006.
[15] T. W. Barbee, R. L. Simpson, A. E. Gash, and J. H. Satcher, “Nano-
laminate-based ignitors,” U.S. Patent WO 2005 016 850 A2, Feb. 24,
2005.
[16] J. C. Hinshaw, “Thermite compositions for use as gas generants,” Interna-
tional Patent WO 95/04672, Aug. 8, 1995.
[17] C. Rossi and D. Estève, “Pyrotechnic micro actuators,” in Proc. 11th
EUROSENSORS XI, Varsovie, Pologne, 1997, vol. 2, pp. 771–774.
[18] C. Rossi, D. Estève, and C. Mingués, “Pyrotechnic actuator: A new gener-
ation of Si integrated actuator,” Sens. Actuators A, Phys., vol. 74, no. 1–3,
pp. 211–215, Apr. 1999.
[19] C. C. Hong, S. Murugesan, G. Beaucage, J. W. Choi, and C. H. Ahn, “A
functioning on-chip pressure generator using solid chemical propellant for
disposable lab-on-a-chip,” Lab Chip, vol. 3, no. 4, pp. 281–286, 2003.
[20] A. A. Norton and M. A. Minor, “Pneumatic microactuator powered by the
deflagration of sodium azide,” J. Microelectromech. Syst., vol. 15, no. 2,
pp. 344–354, Apr. 2006.
[21] D. Li, X. B. Lu, Z. Z. Zhou, M. X. Qiu, and C. G. Lin, “Laser-
initiated aluminothermic reaction applied to preparing Mo–Si film
on silicon substrates,” in Proc. Mater. Res. Soc. Symp., 1988, vol. 101,
pp. 487–490.
[22] “Pile àgrave; combustible pour l’alimentation d’appareils électroniques,
notamment portables,” Patent FR2 818 808, 2005. (in French).
[23] O. Vasylkiv and Y. Sakka, “Nanoexplosion synthesis of multimetal ox-
ide ceramic nanopowders,” Nano Lett., vol. 5, no. 12, pp. 2598–2604,
Dec. 2005.
[24] O. Vasylkiv, Y. Sakka, and V. V. Skorokhod, “Nano-blast synthesis of
nano-size CeO2-Gd2O3 powders,” J. Amer. Ceram. Soc., vol. 89, no. 6,
pp. 1822–1826, Jun. 2006.
[25] D. S. Stewart, “Miniaturization of explosive technology and microdeton-
ics,” in Proc. 21st ICTAM, Varsaw, Poland, Aug. 15–21, 2004.
[26] P. Pennarun, C. Rossi, D. Estève, and D. Bourrier, “Design, fabrication
and characterization of a MEMS safe pyrotechnical igniter integrating
arming, disarming and sterilization functions,” J. Micromech. Microeng.,
vol. 16, no. 1, pp. 92–100, Jan. 2006.
[27] A. S. Tappan, G. T. Long, A. M. Renlund, and S. H. Kravitz, “Microen-
ergetic materials-microscale energetic material processing and testing,”
presented at the 41st AIAA Aerospace Sciences Meeting and Exhibit,
Reno, NV, Jan. 6–9, 2003, Paper AIAA-2003-0242.
[28] A. S. Tappan, G. T. Long, B. Wroblewski, J. Nogan, H. A. Palmer, S. H.
Kravitz, and A. M. Renlund, “Patterning of regular porosity in PETN mi-
croenergetic material thin films,” in Proc. 36th Int. Conf. ICT, Karlsruhe,
Germany, Jun. 28–Jul. 1, 2005, pp. 134–135.
[29] L. E. Fried, M. R. Manaa, P. F. Pagoria, and R. L. Simpson, “Design and
synthesis of energetic materials,” Annu. Rev. Mater. Res., vol. 31, no. 1,
pp. 291–321, 2001.
[30] C. Rossi and D. Esteve, “Micropyrotechnics, a new technology for mak-
ing energetic microsystems: Review and prospective,” Sens. Actuators A,
Phys., vol. 120, no. 2, pp. 297–310, May 2005.
[31] C. Rossi, B. Larangot, D. Lagrange, and A. Chaalane, “Final character-
izations of MEMS-based pyrotechnical microthrusters,” Sens. Actuators
A, Phys., vol. 121, no. 2, pp. 508–514, Jun. 2005.
[32] R. A. Rugunanan and M. E. Brown, “Combustion of binary and ternary
silicon/oxidant pyrotechnic systems, Part I: Binary systems with Fe
2
O
3
and SnO
2
as oxidants,” Combust. Sci. Technol., vol. 95, no. 1–6, pp. 61–
83, 1994.
[33] C. E. Aumann, G. L. Skofronick, and J. A. Martin, “Oxidation behavior of
aluminum nanopowders,” J. Vac. Sci. Technol. B, Microelectron. Process.
Phenom., vol. 13, no. 2, pp. 1178–1183, May 1995.
[34] M. E. Brown, S. J. Taylor, and M. J. Tribelhorn, “Fuel-oxidant particle
contact in binary pyrotechnic reactions,” Propellants Explos. Pyrotech.,
vol. 23, no. 6, pp. 320–327, 1998.
[35] J. Zhi, L. S. Fen, Z. F. Qi, L. Z. Ru, Y. C. Mei, L. Yang, and L. S. Wen,
“Research on the combustion properties of propellants with low content
of nano metal powders,” Propellants Explos. Pyrotech., vol. 31, no. 2,
pp. 139–147, 2006.
[36] K. Park, D. Lee, A. Rai, D. Mukherjee, and M. R. Zachariah, “Size-
resolved kinetic measurements of aluminum nanoparticle oxidation with
single particle mass spectrometry,” J. Phys. Chem., B, vol. 109, no. 15,
pp. 7290–7299, Apr. 2005.
[37] G. V. Ivanov and F. Tepper, “Activated aluminium as a stored
energy source for propellants,” in Proc. 4th Int. Symp. Spec.
Topics Chem. Propulsion, Stockholm, Sweden, May 27–28, 1996,
pp. 636–644.
[38] M. J. Chiaverini, N. Serin, D. K. Johnson, Y. C. Lu, and K. K. Kuo,
“Challenges,” in Propellants and Combustion 100 Years After Nobel,
K. K. Kuo et al., Ed. New York: Begell House, 1997, pp. 719–733.
[39] R. W. Armstrong, B. Baschung, D. W. Booth, and M. Samirant, “En-
hanced propellant combustion with nanoparticles,” Nano Lett.,vol.3,
no. 2, pp. 253–255, 2003.
[40] V. N. Simonenko and V. E. Zarko, “Comparative studying of the com-
bustion behavior of composite propellant containing ultrafine aluminum,”
in Proc. 30th Int. Annu. Conf. ICT, Karlsruhe, Germany, Jun. 29–Jul. 2,
1999, p. 21/1.
[41] M. M. Mench, C. L. Yeh, and K. K. Kuo, “Propellant burning rate en-
hancement and thermal behavior of ultra-fine aluminum powders (Alex),”
in Proc. 29th Int. Annu. Conf. ICT, Karlsruhe, Germany, Jun. 30–Jul. 3,
1998, p. 30/1.
[42] S. K. Poehlein, D. Burch, R. Johnson, M. Beyard, and S. Larson, “Re-
qualification of demilitarized HMX for military use,” in Proc. 36th Int.
Annu. Conf. ICT & 32nd Int. Pyrotechnics Seminar, Karlsruhe, Germany,
Jun. 28–Jul. 1, 2005, pp. 31–33.
[43] A. Pivkina, P. Ulyanova, Y. Frolov, S. Zavyalov, and J. Schoonman,
“Nanomaterials for heterogeneous combustion,” Propellants Explos. Py-
rotech., vol. 29, no. 1, pp. 39–48, 2004.
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

930 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 16, NO. 4, AUGUST 2007
[44] M. R. Manaa, A. R. Mitchell, R. G. Garza, P. F. Pagoria, and B. E.
Watkins, “Flash ignition and initiation of explosives-nanotubes mixture,”
J. Amer. Chem. Soc., vol. 127, no. 40, pp. 13 786–13 787, Oct. 2005.
[45] Y. Frolov, A. Pivkina, P. Ulyanova, and S. Zavyalov, “Nanomaterials and
nanostructures as components for high-energy condensed systems,” in
Proc. 28th Int. Pyrotechnics Seminar, Adelaide, Australia, Nov. 2001,
p. 305.
[46] L. L. Wang, Z. A. Munir, and Y. M. Maximov, “Thermite reactions: Their
utilization in the synthesis and processing of materials,” J. Mater. Sci.,
vol. 28, no. 14, pp. 3693–3708, 1993.
[47] B. S. Bockmon, M. L. Pantoya, S. F. Son, B. W. Asay, and J. T.
Mang, “Combustion velocities and propagation mechanisms of metastable
interstitial composites,” J. Appl. Phys., vol. 98, no. 6, pp. 064 903/1–
064 903/7, Sep. 2005.
[48] J. J. Granier and M. L. Pantoya, “Laser ignition of nanocomposite ther-
mites,” Combust. Flame, vol. 138, no. 4, pp. 373–383, 2004.
[49] M. L. Pantoya and J. J. Granier, “Combustion behavior of highly en-
ergetic thermites: Nano versus micron composites,” Propellants Explos.
Pyrotech., vol. 30, no. 1, pp. 53–62, 2005.
[50] J. Sun, M. L. Pantoya, and S. L. Simon, “Dependence of size and size
distribution on reactivity of aluminum nanoparticles in reactions with
oxygen and MoO
3
,” Thermochim. Acta, vol. 444, no. 2, pp. 117–127,
2006.
[51] S. Bhattacharya, Y. Gao, S. Apperson, S. Subramaniam, R. Shende,
S. Gangopadhyay, and E. Talantsev, “A novel on-chip diagnostic method
to measure burn rates of energetic materials,” J. Energ. Mater., vol. 24,
no. 1, pp. 1–15, Jan.–Mar. 2006.
[52] S. H. Fischer and M. C. Grubelich, “Theoretical energy release of ther-
mites, intermetallics, combustible metals,” in Proc. 24th Int. Pyrotechnics
Seminar, Monterey, CA, Jul. 1998, pp. 1–6.
[53] COACH: COmputer Aided CHemistry, Saint Martin d’Heres,
France: Thermodata. [Online]. Available: http://thermodata.online.fr/
coachang.html
[54] K. Moore and M. L. Pantoya, “Combustion of environmentally altered
molybdenum trioxide nanocomposites,” Propellants Explos. Pyrotech.,
vol. 31, no. 3, pp. 182–187, 2006.
[55] A. Prakash, A. V. McCormick, and M. R. Zachariah, “Synthesis and reac-
tivity of a super-reactive metastable intermolecular composite formulation
of Al/KMnO
4
,” Adv. Mater., vol. 17, no. 7, pp. 900–903, 2005.
[56] L. Duraes, “Radial combustion propagation in Iron (III) oxide/
aluminum thermite mixtures,” Propellants Explos. Pyrotech., vol. 31,
no. 1, pp. 42–49, 2006.
[57] A. E. Gash, T. M. Tillotson, J. H. Satcher, Jr., J. F. Poco, L. W. Hrubesh,
and R. L. Simpson, “Use of epoxides in the sol–gel synthesis of porous
Iron(III) oxide monoliths from Fe(III) salts,” Chem. Mater., vol. 13, no. 3,
pp. 999–1007, 2001.
[58] T. M. Tillotson, A. E. Gash, R. L. Simpson, L. W. Hrubesh, J. H.
Satcher, Jr., and J. F. Poco, “Nanostructured energetic materials using
sol–gel methodologies,” J. Non-Cryst. Solids, vol. 285, no. 1, pp. 338–
345, Jun. 2001.
[59] A. W. Miziolek, “Nanoenergetics: An emerging technology area of na-
tional importance,” AMPTIAC, vol. 6, no. 1, pp. 43–48, 2002.
[60] A. E. Gash, T. M. Tillotson, J. H. Satcher, Jr., L. W. Hrubesh, and R. L.
Simpson, “New sol–gel synthetic route to transition and main-group metal
oxide aerogels using inorganic salt precursors,” J. Non-Cryst. Solids,
vol. 285, no. 1, pp. 22–28, Jun. 2001.
[61] A. Prakash, A. V. McCormick, and M. R. Zachariah, “Aero-sol–gel
synthesis of nanoporous iron-oxide particles: A potential oxidizer for
nanoenergetic materials,” Chem. Mater., vol. 16, no. 8, pp. 1466–1471,
2004.
[62] D. Prentice, M. L. Pantoya, and B. J. Clapsaddle, “Effect of nanocom-
posite synthesis on the combustion performance of a ternary thermite,” J.
Phys. Chem., B, vol. 109, no. 43, pp. 20 180–20 185, 2005.
[63] B. J. Clapsaddle, D. W. Sprehn, A. E. Gash, J. H. Satcher, Jr., and
R. L. Simpson, “A versatile sol–gel synthesis route to metal-silicon mixed
oxide nanocomposites that contain metal oxides as the major phase,” J.
Non-Cryst. Solids, vol. 350, no. 1, pp. 173–181, 2004.
[64] A. Prakash, A. V. McCormick, and M. R. Zachariah, “Tuning the reactiv-
ity of energetic nanoparticles by creation of a core-shell nanostructure,”
Nano Lett., vol. 5, no. 7, pp. 1357–1360, Jul. 2005.
[65] T. A. Vine, S. Tinston, and R. Fairhutst, “Application of physical vapor
deposition to the manufacture of pyrotechnics,” in Proc. 28th Int. Py-
rotechnics Seminar, Adelaide, Australia, Nov. 4–9, 2001, p. 725.
[66] K. J. Blobaum, M. E. Reiss, J. M. P. Lawrence, and T. P. Weihs, “Deposi-
tion and characterization of a self-propagating CuOx/Al thermite reaction
in a multilayer foil geometry,” J. Appl. Phys., vol. 94, no. 5, pp. 2915–
2922, 2003.
[67] A. J. Gavens, D. V. Heerden, A. B. Mann, M. E. Reiss, and T. P.
Weihsa, “Effect of intermixing on self-propagating exothermic reactions
in Al/Ni nanolaminate foils,” J. Appl. Phys., vol. 87, no. 3, pp. 1255–1263,
Feb. 2000.
[68] P. Tagststrom, P. Martensson, U. Jansson, and J. O. Carlsso, “Atomic layer
epitaxy of tungsten oxide films using oxyfluorides as metal precursors,”
J. Electrochem. Soc., vol. 146, no. 8, pp. 3139–3143, Aug. 1999.
[69] H. Seim, M. Niemminen, L. Niinisto, A. Kjekshus, and L. S. Johansson,
“Growth of LaCoO3 thin film from beta-diketonate precursors,” Appl.
Surf. Sci., vol. 112, no. 1, pp. 243–250, 1997.
[70] T. S. Yang, W. Cho, M. Kim, K. S. An, T. M. Chung, C. G. Kim, and
Y. Kim, “Atomic layer deposition of nickel oxide films using Ni(dmamp)
2
and water,” J. Vac. Sci. Technol. A, Vac. Surf. Films, vol. 23, no. 4,
pp. 1238–1243, 2005.
[71] J. D. Ferguson, K. J. Buechler, A. W. Weimer, and S. M. George, “SnO
2
atomic layer deposition on ZrO
2
and Al nanoparticles: Pathway to en-
hanced thermite materials,” Powder Technol., vol. 156, no. 2/3, pp. 154–
163, 2005.
[72] P. McCord, S. L. Yau, and A. J. Bard, “Chemiluminescence of anodized
and etched silicon: Evidence for a luminescent siloxene-like layer on
porous silicon,” Science, vol. 257, no. 5066, p. 68, Jul. 1992.
[73] D. Kovalev, V. Y. Timoshenko, N. Künzner, E. Gross, and F. Koch,
“Strong explosive interaction of hydrogenated porous silicon with oxygen
at cryogenic temperatures,” Phys. Rev. Lett., vol. 87, no. 6, pp. 068 301/1–
068 301/4, Aug. 2001.
[74] F. V. Mikulec, J. D. Kirtland, and M. J. Sailor, “Explosive nanocrystalline
porous silicon and its use in atomic emission spectroscopy,” Adv. Mater.,
vol. 14, no. 1, pp. 38–41, Jan. 2002.
[75] X. Li and P. W. Bohna, “A metal-assisted chemical etching in HF/H
2
O
2
produces porous silicon,” Appl. Phys. Lett., vol. 77, no. 6, pp. 2572–2574,
2000.
[76] L. Menon, S. Patibandla, K. Bhargava Ram, S. I. Shkuratov,
D. Aurongzeb, M. Holtz, J. Berg, J. Yun, and H. Temkin, “Ignition studies
of Al/Fe
2
O
3
energetic nanocomposites,” Appl. Phys. Lett., vol. 84, no. 23,
pp. 4737–4737, 2004.
[77] H. Masuda and K. Fukuda, “Ordered metal nanohole arrays made by a
two-step replication of honeycomb structures of anodic alumina,” Science,
vol. 268, no. 5216, pp. 1466–1468, Jun. 1995.
[78] H. Masuda, H. Yamada, M. Satoh, H. Asoh, M. Nakao, and T. Tamamura,
“Highly ordered nanochannel-array architecture in anodic alumina,” Appl.
Phys. Lett., vol. 71, no. 19, pp. 2770–2772, Nov. 1997.
[79] S. H. Kim and M. R. Zachariah, “Enhancing the rate of energy release
from NanoEnergetic materials by electrostatically enhanced assembly,”
Adv. Mater., vol. 16, no. 20, pp. 1821–1825, 2004.
[80] S. Subramaniam, S. Hasan, S. Bhattacharya, Y. Gao, S. Apperson,
M. Hossain, R. V. Shende, S. Gangopadhyay, P. Redner, D. Kapoor, and
S. Nicolich, “Self-assembled nanoenergetic composite,” in Proc. Mater.
Res. Soc. Symp., 2006, vol. 896, pp. 0896-H01-05.1–0896-H01-05.6.
[81] V. Sundaram, K. V. Logan, and R. F. Speyer, “Aluminothermic reac-
tion path in the synthesis of a TiB
2
−Al
2
O
3
composite,” J. Mater. Res.,
vol. 12, no. 7, pp. 1681–1684, Jul. 1997.
Carole Rossi received the degree in engineering
from the National Institute of Applied Science,
Toulouse, France, in 1994 and the Ph.D. degree in
electrical engineering and physics in 1997.
She is currently a Researcher at the French
National Center for Scientific Research (CNRS),
Toulouse, where she is a Member of the Microsys-
tems and Systems Integration (MIS) Group, Lab-
oratoire d’Analyse et d’Architecture des Systèmes,
in the MEMS department. Her research interests
include microenergetic for MEMS, micropyrotech-
nical systems, and power MEMS for electrical generation. She is currently
leading the power MEMS research area at LAAS, and her team proposes new
concepts for actuation and energy on a chip.
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.

ROSSI et al.: NANOENERGETIC MATERIALS FOR MEMS: A REVIEW 931
Kaili Zhang received the B.S. degree in mechanical
engineering from Dong Hua University, China, in
1997 and the Ph.D. degree in microsystems from
the National University of Singapore, Kent Ridge,
Singapore, in 2005.
From 2005 to 2006, he was with the National
University of Singapore as a Postdoctoral Research
Fellow. He is currently with the Microsystems
and Systems Integration (MIS) Group, Laboratoire
d’Analyse et d’Architecture des Systèmes (LAAS),
French National Center of Scientific Research
(CNRS). His current research interests include nanoenergetic materials,
nanometal oxides, nanocatalysis, micropropulsion, and thin films.
Daniel Estève was born in 1941. He received the
Ph.D. degree in 1966 and the Doctorat d’Etat degree
in 1972.
He joined the French National Center of Scien-
tific Research (CNRS) in 1966, where he is cur-
rently the Director of Research and a member of the
Microsystems and Systems Integration (MIS) Group,
Laboratoire d’Analyse et d’Architecture des Sys-
tèmes (LAAS). He was one of the first members
of the LAAS-CNRS, which was founded in 1968.
During 1982-1986, he was the Director of LAAS-
CNRS. He has been with LAAS since the beginning of microsystems activities
there. His fields of interest have covered a wide range from electronic circuit
and electronic device modeling to microsystems modeling and simulation,
design, and characterization.
Pierre Alphonse received the Ph.D. degree from
University of Toulouse, Toulouse, France, in 1986.
He is currently a Research Engineer at the Centre
Interuniversitaire de Recherche et d’Ingénierie des
Matériaux (CIRIMAT), Toulouse. His research inter-
ests include the design and analysis of materials for
catalysis, especially in the field of pollution control
and energy conversion.
Philippe Tailhades was born on July 19, 1960,
at Mazamet, France. He received the Ph.D. degree
in Material Science in 1988 and the Habilitation à
Diriger les Recherches in 1994.
He is currently the Vice Director of the Centre
Interuniversitaire de Recherche et d’Ingénierie des
Matériaux (CIRIMAT), Toulouse, France. His re-
search interests include the preparation of original
metallic oxides, especially spinel ferrites, in the form
of fine powders, thin films, or bulk ceramics and the
study of the magnetic, electric, and optical properties
of these materials. He also works on the preparation of special metallic
powders.
Dr. Tailhades received the silver medal of CNRS in France in 2000.
Constantin Vahlas received the Bachelor’s degree
in chemical engineering from the National Technical
University of Athens, Athens, Greece, and the Mas-
ter’s degree and the doctoral degree in metallurgy
from the National Polytechnic Institute of Grenoble,
Grenoble, France.
He is currently the National Center for Scientific
Research (CNRS) Director of Research at the Centre
Interuniversitaire de Recherche et d’Ingénierie des
Matériaux (CIRIMAT), Toulouse, France, where he
is the Head of the Chemical Vapor Deposition Group.
He is the Director of a European Integrated Laboratory on Advanced Coating
Technologies in the frame of the Network of Excellence on Complex Metallic
Alloys. He has been working with CNRS for 21 years, where his early
studies centered on thermochemistry of the gas-solid processes and on CVD
of ceramic and pyrolytic carbon films for microelectronics and composite
materials applications.
Authorized licensed use limited to: INP TOULOUSE. Downloaded on May 11, 2009 at 03:01 from IEEE Xplore. Restrictions apply.