One-minute synthesis of crystalline binary and ternary metal oxide nanoparticles

Article abstract of DOI:10.1039/b717334b

Highly crystalline metal oxide nanoparticles such as CoO, ZnO, Fe ₃O₄, MnO, Mn₃O₄, and BaTiO ₃ were synthesized in just a few minutes by reacting metal alkoxides, acetates or acetylacetonates with benzyl alcohol under microwave heating. The Royal Society of Chemistry.

Full text of DOI:10.1039/b717334b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
One-minute synthesis of crystalline binary and ternary metal oxide
nanoparticlesw
Idalia Bilecka, Igor Djerdj and Markus Niederberger*
Received (in Cambridge, UK) 9th November 2007, Accepted 6th December 2007
First published as an Advance Article on the web 19th December 2007
DOI: 10.1039/b717334b
Highly crystalline metal oxide nanoparticles such as CoO, ZnO,
Fe3O4, MnO, Mn3O4, and BaTiO3 were synthesized in just a
few minutes by reacting metal alkoxides, acetates or acetylace-
tonates with benzyl alcohol under microwave heating.
condensation step for the Zn–O–Zn bond formation. This
result is of utmost significance, because it gives strong indica-
tion that microwave irradiation can be used as a powerful tool
to accelerate the formation of metal oxide nanoparticles
through directly influencing the organic reaction pathways.
The synthesis of the metal oxide nanoparticles involved the
dispersion of the precursors in benzyl alcohol (Scheme 1),
followed by heating in a microwave of 2.45 GHz frequency,
with the power modulated as a function of temperature (for
details see ESIw). The reaction mixture was continuously
stirred and then thermally quenched by a compressed air jet.
The temperature profile during the reaction can be divided
into three regions (see ESIw) that all play an important role for
nanoparticle formation. In the first stage, the mixture is heated
from room temperature to 60 1C to allow the precursor to
completely dissolve. Thereafter, fast heating to 200 1C leads to
homogeneous nucleation, probably following the same model
as recently reported for the preparation of monodisperse iron
oxide nanocrystals by a heating-up process.24 Crystal growth
occurred then upon further heating at this temperature,
enabling control over the final crystallite size.
Metal oxide nanoparticles constitute an outstanding class of
functional materials with potential applications in almost all
fields of technology.1 Among the different synthesis ap-
proaches developed in the last few years, nonaqueous or
nonhydrolytic processes were particularly successful with re-
spect to achieving control of crystallite size, shape, and
assembly behavior.2–5 But in spite of all the progress, the
preparation of nanocrystalline metal oxides remains rather
time- and also energy-consuming. One possibility to accelerate
the synthesis process is the use of microwave irradiation,
which allows a more efficient and homogeneous heating of
the reaction mixture. Microwave-assisted preparation routes
for inorganic materials are not yet as popular as for organic
compounds. However, the growing number of publications
dealing with that topic gives a first impression about the great
potential of the method.6–8 The composition of inorganic
nanoparticles obtained via microwave heating ranges from
metals (Au, Pd, Ag, Pt, AuPb)8,9 to oxides (TiO2, BaTiO3,
tungstates, ZnO),10–15 chalcogenides [M2S3 (M = Bi, Sb),
ZnS, ZnSe, CdSe, PbTe, PbSe, CdS],16–19 and phosphates
(LaPO4:Ce,Tb)20 and the solvents used include water, organic
solvents, or ionic liquids.
Powder X-ray diffraction (XRD) patterns of the obtained
samples (Cu Ka or Co Ka radiation) are shown in Fig. 1. All
patterns correspond to the respective phase-pure metal oxide
with broad reflections that point to nanosized crystallites. The
XRD pattern of the brownish CoO sample can be assigned to
the cubic rocksalt-type phase (ICDD PDF No. 01-089-7099)
with a crystallite size of 6 nm calculated by Scherrer analysis of
the 200 reflection (Fig. 1a). In the case of ZnO (Fig. 1b),
independent of the precursor Zn(ac)2 or Zn(acac)2 the final
white powder consisted of zincite ZnO (ICDD PDF No. 36-
1451). However, according to Scherrer analysis of the 100
reflection the crystallite size amounted to 20 nm for the
acetylacetonate as precursor, whereas for the acetate the sizes
are in general larger (around 25–30 nm). The reaction of
Here we present a broadly applicable synthesis strategy that
enables the preparation of highly crystalline metal oxide
nanoparticles in good yields within minutes and even seconds
using a combination of nonaqueous sol–gel chemistry and
microwave heating. Thus, the use of microwave heating offers
an immense reduction of the reaction times in comparison to
traditional heating in an oil bath or in an autoclave. Further-
more, the pronounced dependence of the crystallite size on the
heating time (which was not found in analogous solvothermal
experiments in the autoclave21–23) provides a precious tool to
tailor this parameter. Comparison of the reaction mechanism
of ZnO, prepared either in the microwave or in the autoclave,
proved that both processes involved ester elimination as
Laboratory for Multifunctional Materials, Department of Materials,
ETH Zurich, Wolfgang-Pauli Strasse 10, 8093 Zurich, Switzerland.
¨
¨
E-mail: markus.niederberger@mat.ethz.ch; Fax: 41 44 632 11 01;
Tel: 41 44 633 63 90
w Electronic supplementary information (ESI) available: experimental
details, temperature/pressure profile, XRD powder pattern of MnO
and Mn3O4, TEM images of ZnO. See DOI: 10.1039/b717334b
Scheme 1 General reaction scheme displaying the metal oxide
precursors used, the solvent, the experimental conditions, and the
resulting metal oxide nanoparticles.
ꢀc
This journal is The Royal Society of Chemistry 2008
886 | Chem. Commun., 2008, 886–888

View Article Online
Fig. 1 X-ray diffraction patterns of (a) CoO, (b) ZnO, (c) Fe3O4, and
(d) BaTiO3 (a) and (c) Co Ka, (b) and (d) Cu Ka radiation.
Fe(ac)2, Fe(acac)2 or Fe(acac)3 yielded black powders that
showed magnetic behavior under the influence of a permanent
magnet. Although the XRD powder patterns were similar for
all three samples (Fig. 1c) and could clearly be assigned to
magnetite Fe3O4 (ICDD PDF No. 19-629), their crystallite
sizes (extracted from the 311 reflection) differed considerably
in dependence of the precursor used. 11 nm-sized crystals were
obtained with Fe(acac)2, 7 nm with Fe(ac)2 and Fe(acac)3 gave
the smallest crystallites of just 5 nm. In the case of the
manganese oxide nanoparticles, the choice of the precursor
does not only influence the crystallite size, but the final phase
composition (see ESIw). Mn(ac)2 favoured the formation of
10 nm sized MnO (calculated from the 200 reflection) in the
manganosite structure (ICDD PDF No. 7-230), wheras
Mn(acac)2 resulted in 7 nm sized (from the 224 reflection)
Mn3O4 hausmannite (ICDD PDF No. 24-734). The XRD
pattern in Fig. 1d corresponds to BaTiO3 (ICDD PDF No.
31-174). In this case it is worth highlighting again that there is
no indication of BaCO3 as crystalline by-product. The crystal-
lite size extracted from the (110) reflection amounts to about
12 nm. Due to the small crystal size the reflections are too
broad to discriminate between the cubic and the tetragonal
structure.23
Fig. 2 TEM and HRTEM images (insets) of (a) CoO, (b) MnO, (c)
Fe3O4, and (d) BaTiO3.
6–18 nm (Fig. 2d), also matching the XRD data. All these
metal oxides exhibit well-defined lattice fringes in the corre-
sponding HRTEM images (Fig. 2, insets), proving the high
crystallinity of these nanomaterials.
Different organic reaction pathways have been identified to
play a crucial role in the nonaqueous liquid-phase synthesis of
nanocrystalline metal oxides particles.4,5 The investigation of
the organic reaction mechanism involved in the formation of
ZnO starting from Zn(ac)2 provides a unique opportunity to
compare the crystallization process of the nanoparticles either
under microwave heating (for a few minutes) or in the auto-
clave (for several hours). Thus, the final reaction solutions
after removal of the inorganic precipitates were analyzed by
coupled gas-chromatography–mass spectrometry (GC-MS).
In addition to the solvent benzyl alcohol, the only organic
species detected was benzyl acetate. Based on this result, one
can clearly state that autoclave as well as microwave heating
led to ZnO formation via an ester elimination process.25 At
this point, we can only speculate about the underlying effect,
which results in the great acceleration of the microwave
heating process in comparison to the autoclave synthesis. In
general, the acceleration of reactions by microwave exposure is
due to thermal effects (dielectric heating) and specific non-
thermal effects.26 Decrease in the activation energy is the most
important part of the specific nonthermal microwave effect,
which is a consequence of the greater stabilization of the
transition state during the reaction course compared to the
ground state.26 The effect is particularly pronounced, when the
transition state is more polar than the ground state, i.e., when
the dipole moment is enhanced from the ground to the
transition state.26 Considering that the formation of benzyl
acetate by nucleophilic attack on the acetate ligand proceeds
via a highly polar transition state, the acceleration can be
explained on the basis of the specific nonthermal microwave
effect.
Representative transmission electron microscopy (TEM)
images were used to study the particle morphologies and size
distributions (Fig. 2). All the images display well-dispersed
nanoparticles, however with different morphologies and vary-
ing uniformity. The size and shape of the CoO (Fig. 2a) and
ZnO (see ESIw) nanoparticles is not very well defined, but
roughly agrees with the XRD data. On the other hand, MnO is
characterized by a relatively homogeneous cube-like shape
(Fig. 2b). The side lengths of these cubes (30–50 nm) is
considerably larger than the crystallite size measured by
XRD, which points to a polycrystalline nature of these
particles. The magnetite sample in Fig. 2c [prepared from
Fe(ac)2] exhibits a rather uniform spherical particle morphol-
ogy with a small size distribution in the range of 5 to 10 nm,
which agrees well with XRD data. The BaTiO3 nanoparticles
are particularly well dispersed, with sizes in the range of
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 886–888 | 887

View Article Online
propose a way to control the reaction rate by applying
microwave irradiation to influence (i.e., to accelerate) the
organic reaction pathway occurring in parallel to nanoparticle
formation. In comparison to the corresponding synthesis
procedures performed in the autoclave, the reactions are much
faster and the crystallite size is tunable. The fact that nano-
particle formation is based on the same mechanism in the
autoclave and in the microwave strongly supports the propo-
sition that microwave irradiation has a great potential to
control the growth of inorganic nanoparticles through influen-
cing the organic reaction pathways.
Fig. 3 (a) X-ray diffraction pattern (using Co Ka radiation) of CoO
obtained after reaction times ranging from 30 s to 20 min (0.1 M
precursor concentration). (b) Evolution of the average crystallite size
(calculated from the 200 reflection) with the reaction time and in
dependence of the precursor concentration.
Financial support by the ETH Zurich is gratefully acknowl-
¨
edged. We made use of the facilities provided by EMEZ
(Electron Microscopy ETH Zurich).
¨
References
1 J. A. Rodriguez and M. Fernandez-Garcia, Synthesis, Properties
and Applications of Oxide Nanomaterials, John Wiley & Sons, New
Jersey, 2007.
2 Y. W. Jun, J. S. Choi and J. Cheon, Angew. Chem., Int. Ed., 2006,
45, 3414.
Table 1 Crystallite sizes of CoO nanoparticles calculated by Scherrer
analysis of the 200 reflection in dependence of the heating time (0.1 M
precursor concentration)
Thermal
treatment/s
XRD
particle size/A
Reaction
yield [%]
3 J. Park, J. Joo, S. G. Kwon, Y. Jang and T. Hyeon, Angew. Chem.,
Int. Ed., 2007, 46, 4630.
Sample no
4 M. Niederberger and G. Garnweitner, Chem.–Eur. J., 2006, 12,
7282.
5 M. Niederberger, Acc. Chem. Res., 2007, 40, 793.
6 D. M. P. Mingos, Adv. Mater., 1993, 5, 857.
7 K. J. Rao, B. Vaidhyanathan, M. Ganguli and P. A. Ramakrish-
nan, Chem. Mater., 1999, 11, 882.
1
2
3
4
5
30
60
180
600
1200
45
54
58
61
75
60.8
68.8
66.2
70
77
8 M. Tsuji, M. Hashimoto, Y. Nishizawa, M. Kubokawa and T.
Tsuji, Chem.–Eur. J., 2005, 11, 440.
9 F. K. Liu, Y. C. Chang, F. H. Ko and T. C. Chu, Mater. Lett.,
2004, 58, 373.
10 W. W. Wang and Y. J. Zhu, Inorg. Chem. Commun., 2004, 7, 1003.
11 S. Baldassari, S. Komarneni, E. Mariani and C. Villa, Mater. Res.
Bull., 2005, 40, 2014.
12 A. V. Murugan, V. Samuel and V. Ravi, Mater. Lett., 2006, 60,
479.
13 K. L. Ding, Z. J. Miao, Z. M. Liu, Z. F. Zhang, B. X. Han, G. M.
An, S. D. Miao and Y. Xie, J. Am. Chem. Soc., 2007, 129, 6362.
14 J. M. Lee, D. P. Amalnerkar, Y. K. Hwang, S. H. Jhung, J. S.
Hwang and J. S. Chang, J. Nanosci. Nanotechnol., 2007, 7, 952.
15 A. Michailovski, R. Kiebach, W. Bensch, J. D. Grunwaldt, A.
Baiker, S. Komarneni and G. R. Patzke, Chem. Mater., 2007, 19,
185.
16 R. Kerner, O. Palchik and A. Gedanken, Chem. Mater., 2001, 13,
1413.
17 A. B. Panda, G. Glaspell and M. S. El-Shall, J. Am. Chem. Soc.,
2006, 128, 2790.
18 Y. Jiang and Y. J. Zhu, J. Phys. Chem. B, 2005, 109, 4361.
19 A. V. Murugan, R. S. Sonawane, B. B. Kale, S. K. Apte and A. V.
Kulkarni, Mater. Chem. Phys., 2001, 71, 98.
20 G. Buehler and C. Feldmann, Angew. Chem., Int. Ed., 2006, 45,
4864.
21 N. Pinna, S. Grancharov, P. Beato, P. Bonville, M. Antonietti and
M. Niederberger, Chem. Mater., 2005, 17, 3044.
22 I. Djerdj, D. Arcon, Z. Jaglicic and M. Niederberger, J. Phys.
Chem. C, 2007, 111, 3614.
23 M. Niederberger, G. Garnweitner, N. Pinna and M. Antonietti, J.
Am. Chem. Soc., 2004, 126, 9120.
24 S. G. Kwon, Y. Piao, J. Park, S. Angappane, Y. Jo, N.-M. Hwang,
J.-G. Park and T. Hyeon, J. Am. Chem. Soc., 2007, 129, 12571.
25 G. Clavel, M.-G. Willinger, D. Zitoun and N. Pinna, Adv. Funct.
Mater., 2007, 17, 3159.
26 A. Loupy, Microwaves in Organic Synthesis, Vol. 1, Wiley-VCH,
Weinheim, 2006.
In the solvothermal nonaqueous sol–gel synthesis of metal
oxide nanoparticles in benzyl alcohol, especially the choice of
the precursor influences the final crystal size, morphology, and
composition.4,5 Parameters like reaction time and precursor
concentration just play a minor role. However, in the case of
microwave heating these parameters become more important.
Therefore, time dependant studies in the CoO system were
performed using two different precursor concentrations. XRD
analysis (Fig. 3a) of the CoO samples obtained after different
reaction times in the range of 30 s to 20 min revealed that
already after 30 s the main reflections of CoO are present, in
addition to other not yet assignable peaks. After one minute,
the CoO sample is phase-pure and longer heat treatment just
results in crystallite growth from about 5 to 8 nm, which can
easily be monitored by the narrowing of the 200 reflection
(Table 1). Furthermore, at the same temperature the yield
increases with reaction time from about 60% to nearly 80%.
Enhanced crystal growth cannot only be achieved by extend-
ing the heating time, but also by increasing the precursor
concentration from 0.1 to 0.2 M (Fig. 3b). The crystallite sizes
change then from about 3 to 9 nm as the reaction time
increases. A closer look at the two curves in Fig. 3b shows
two distinct kinetic regimes, i.e., during the first minutes the
average crystallite size increases rapidly, followed by a slower
growth after about 3 min.
The results reported here on the one hand offer a fast
synthesis route to a variety of binary and ternary metal oxide
nanoparticles with high crystallinity, and on the other hand
ꢀc
This journal is The Royal Society of Chemistry 2008
888 | Chem. Commun., 2008, 886–888