Microwave irradiation assisted rapid synthesis of Fe-Ru bimetallic nanoparticles and their catalytic properties in water-gas shift reaction

Article abstract of DOI:10.1016/j.materresbull.2008.12.001

Fe-Ru bimetallic nanoparticles were prepared by a microwave irradiation assisted glycol reduction method using poly-N-vinyl-2-pyrrolidone (PVP) as protective agent. The structure and morphology of the nanoparticles were characterized by X-ray diffraction

Full text of DOI:10.1016/j.materresbull.2008.12.001

Materials Research Bulletin 44 (2009) 1347–1351
Contents lists available at ScienceDirect
Materials Research Bulletin
journal homepage: www.elsevier.com/locate/matresbu
Microwave irradiation assisted rapid synthesis of Fe–Ru bimetallic nanoparticles
and their catalytic properties in water-gas shift reaction
Jian-Quan Du a, Yong Zhang b, Tian Tian b, Shan-Cheng Yan b, Hai-Tao Wang b,
*
a Department of Chemistry, Taishan University, Taian, 271021, PR China
b State Key Laboratory of Bioelectronics, School of Biological Science and Medical Engineering, Southeast University, Nanjing, 210096, PR China
A R T I C L E I N F O
A B S T R A C T
Article history:
Fe–Ru bimetallic nanoparticles were prepared by a microwave irradiation assisted glycol reduction
method using poly-N-vinyl-2-pyrrolidone (PVP) as protective agent. The structure and morphology of
the nanoparticles were characterized by X-ray diffraction (XRD), energy-dispersive X-ray analysis
(EDXA) and high-resolution transmission electron microscopy (HRTEM). EDXA and XRD analysis
confirmed the presence of Fe and Ru. The bimetallic nanoparticles were subsequently loaded onto an
MgAl2O4 supporter with K2O as promoters and used as catalyst for water-gas shift reaction. The results
indicated that the FeRu bimetallic nanoparticles exhibit high catalytic activity for water-gas shift
reaction due to the synergistic effect between iron and ruthenium. Potassium oxide can enhance the CO
selectivity of the catalyst significantly besides increasing the catalyst activity.
Received 1 December 2007
Received in revised form 10 November 2008
Accepted 12 December 2008
Available online 25 December 2008
Keywords:
A. Alloys
A. Nanostructures
B. Chemical synthesis
D. Catalytic properties
ß 2008 Elsevier Ltd. All rights reserved.
1. Introduction
the reduction the CO levels in the ammonia synthesis. WGSR has
aroused renewed interest with the respect of supplying highly pure
The design and controlled fabrication of nanoscaled materials
with functional properties have attracted much interest due to
their various applications such as catalysts, magnetic devices,
single electron transistors, and optoelectronics [1–4]. Recently,
core–shell and alloy bimetallic nanoparticles have been exten-
sively studied because they can exhibit unique electronic, optical,
and catalytic properties that are absent in corresponding mono-
metallic nanoparticles [5–8]. Furthermore, the addition of second
metal provides a method to control the chemical and physical
properties of nanoparticles.
hydrogen as a combustible to fuel cell power generation [15].
Ruthenium deposited on alkali promoted supporters makes
catalysts for WGSR very active that they have been accepted as
alternatives to the commercially used iron catalysts [16–19].
In this paper, we report a simple method to prepare Fe–Ru
bimetallic nanoparticles using 1,2-propanediol to reduce the
mixture of FeCl3 and RuCl3 with the presence of poly-N-vinyl-2-
pyrrolidone (PVP) as protective agent. Additionally, the catalytic
properties of the Fe–Ru bimetallic nanoparticles supported on
MgAl2O4 for the water-gas shift reaction were investigated.
The polyol method was used for the preparation of nanopar-
ticles [9–14] in the past decade. In this method, a polyol (for
example, ethylene glycol (EG)) was used as both a reducing reagent
and a solvent. The glycol method using conventional heating
requires a relatively long preparation time (usually several hours
or longer). Compared with the conventional heating, microwave
heating is promising due to its unique effects, such as rapid
volumetric heating, higher reaction rates and shorter reaction
time, higher reaction selectivity and energy saving and leads to the
formation of uniform nanoparticles.
2. Experimental
2.1. Prepare FeRu bimetallic nanoparticles
The Fe–Ru nanoparticles stabilized by poly-N-vinyl-2-pyrroli-
done were prepared by a modified polyol method reported by
Feldmann and Metzmacher [12]. In brief, PVP (0.8325 g, 7.5
 10À3 mol), RuCl3ÁnH2O (0.0371 g, 1.5  10À4 mol) and FeCl3Á6H2O
6H2O (0.06 g, 1.5 Â 10À4 mol) were dissolved in 45 mL 1,2-
propanediol under stirring to form a dark red solution, then 5 mL
1,2-propanediol solution of ammonia (0.1 M) was added dropwise
under vigorous stirring. The mixture was then transferred to a
homemade Microwave Oven. A transparent dark brown homo-
genous colloidal solution of Fe–Ru bimetallic nanoparticles was
obtained without any precipitate after microwave heating at an
The water-gas shift reaction (WGSR) is an industrially important
process that can be employed in the Fischer–Tropsch reaction and in
* Corresponding author. Tel.: +86 25 83790820; fax: +86 25 83790310.
E-mail address: htwang315@yahoo.com.cn (H.-T. Wang).
0025-5408/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.materresbull.2008.12.001

1348
J.-Q. Du et al. / Materials Research Bulletin 44 (2009) 1347–1351
output of 300 W for 3 min. As prepared polyol dispersions of Fe–Ru
bimetallic nanoparticles were precipitated using excess acetone.
The precipitates were washed with anhydrous acetone in order to
remove chloride ions and dried under vacuum and re-dispersed into
ethanol–water mixture prior to characterization or use. A series of
bimetallic nanoparticles were prepared by vary the molar ration of
iron to ruthenium and that of PVP to metal ions.
a constant pressure of 0.1 MPa. Steam was introduced by flowing
the gas mixture, comprising about 25% CO, 25% H2, and 50 vol% N2,
through double distilled water. The molar ratio of steam to gas (S/
G) was controlled by changing the temperature of water. In current
experiments, the steam to gas ratio (S/G) was maintained at 1.0
and space velocity (SV) was 2000 hÀ1. The catalyst was reduced by
the gas mixture at 100 8C for 2 h before the reaction. The gas
composition was analyzed with two on-line gas chromatographs
(GC-122) equipped with TCD and FID, respectively. The activity of
the catalyst was evaluated by the carbon dioxide formation.
2.2. Prepare supported catalysts
The bimetallic colloidal solution diluted with deionized water
was mixed with a certain amount of MgAl2O4 spinel and the
mixture was stirred for 1 h at room temperature. The colloidal
particles adsorb instantly onto the MgAl2O4 spinel. Potassium
acetate was added as promoter. The solvent was removed with a
rotary evaporator, and then the sample was dried at 120 8C
overnight and then calcined at 350 8C in Ar for 1.5 h in order to
remove organic materials (PVP). The catalyst is labeled as FeRu/
MA. Catalysts Fe/MA and Ru/MA were also prepared by using the
same method. Catalyst FeRu/MA(im) was prepared by co-impreg-
nation of MgAl2O4 with a mixture of FeCl3 and RuCl3 solutions.
3. Results and discussion
Due to the size effects, it is important to prepare bimetallic
nanoparticles with precise size control. The formation of bimetallic
nanoparticles was first confirmed by TEM. The typical TEM images
of three colloidal FeRu samples thus formed with their corre-
sponding particle size distribution are shown in parts (a–c) in
Fig. 1. The molar ratio of iron to ruthenium is 1:1 in all three
samples. The TEM image indicated that all the bimetallic
nanoparticles are spherical in shape with uniform and narrow
size distribution. When the molar ratio of PVP to metal ions is 5:1,
it is found that the nanoparticles mainly consist of particles with a
diameter of ꢀ8 nm and some large particles with a diameter of
ꢀ10 nm are also observed, as shown in Fig. 1a. When the molar
ratio is increased to 10:1, it is clearly seen that most of the
nanoparticles are ꢀ5 nm in diameter, as shown if Fig. 1b. When the
molar ratio is increased to 50:1, the nanoparticles further become
smaller (ꢀ1.4 nm), as shown in Fig. 1c. Obviously, the average size
of such colloidal FeRu bimetallic nanoparticles can be controlled by
varying the molar ratio of PVP to metal ions.
2.3. Characterization
Transmission electron microscopy (TEM) studies were con-
ducted on a Tecnai G2 S-TWIN transmission electron microscope
equipped with an energy-dispersive X-ray analysis (EDXA)
operating at 200 kV. X-ray diffraction (XRD) patterns were
obtained with
a Rigaku D/MAX-3B X-ray diffractometer at
35 mA, 30 kV, employing Cu K radiation at a scan rate of 38/
1
a
min. The surface morphology of the samples was analyzed with a
Hitachi S-570 scanning electron microscope (SEM).
High-resolution transmission electron microscopy (HRTEM)
and EDAX were used to analyze the structure and composition of
FeRu bimetallic nanoparticles. Fig. 2 shows the typical HRTEM
images of three bimetallic nanoparticles with different molar ratio
of iron to ruthenium. The molar ratio of PVP to metal ions is 10:1
in all samples. Judging from the lattice fringes, the bimetallic
2.4. Catalytic reactions
The WGS reaction was carried out in a continuous flow fixed-
bed microreactor filled with 0.5 g catalyst and operated at
Fig. 1. Typical TEM images of FeRu (1:1) bimetallic nanoparticles with PVP/[Fe + Ru] molar ratio 5:1 (a), 10:1 (b), 50:1 (c). Below (a–c) are their corresponding particle size
distribution.

J.-Q. Du et al. / Materials Research Bulletin 44 (2009) 1347–1351
1349
Fig. 2. Typical high-resolution TEM images of FeRu nanoparticles with Fe/Ru molar ratio 0.5:1 (a) 1:1 (a) and 1.5:1 (c). The molar ratio of PVP to metal ions for all is 10:1. The
scale bar is 5 nm.
nanoparticles appeared to be single crystals. EDAX results indicate
that Fe/Ru (0.5:1), Fe/Ru (1:1) and Fe/Ru (1.5:1) nanoparticles have
the composition of iron and ruthenium in the molar ratio of 0.43,
0.97 and 1.29, respectively, close to the initial molar ratio in the
preparation process. It is difficult to obtain homogenous bimetallic
nanoparticles when the Fe/Ru molar ratio is too high.
The formation of bimetallic nanoparticles is possibly due to a
strong protective ability of PVP on both Fe and Ru monometallic
clusters. In the preparation of bimetallic catalysts, both iron and
ruthenium ions could be attached to the C O group or N atom of
PVP. The glycol reduction of these ions then resulted in the
formation of bimetallic nanoparticles, which were stabilized by
PVP. We have also tried to use PVA and PEG as protective agents to
prepare bimetallic nanoparticles. Agglomerated nanoparticles and
submicro-particles other than monodipersed nanoparticles were
obtained through this microwave irradiation method.
Due to its thermal and steam stability, less acidic surface and
attrition resistance, MgAl2O4 spinel was selected as the carrier of
supported catalyst for WGS reaction. The MgAl2O4 spinel was
prepared according to literature [20]. The formation of MgAl2O4
spinel was confirmed by X-ray diffraction. The X-ray diffraction
patterns of fresh and spent Ru/MgAl2O4, Fe/MgAl2O4 and Fe–Ru/
MgAl2O4 catalysts are shown in Fig. 3. X-ray diffraction analysis
revealed the presence of MgAl2O4 phase in all fresh and spent
samples and the presence of K2O phase at the content 8 wt%. In the
range of active species loading from 0.6 to 1.2 wt%, there is no other
diffraction peaks except that of MgAl2O4 and K2O. It can be observed
from SEM micrograph of the samples that the supporter is formed by
agglomerates of 30–80 nm MgAl2O4 spinel grains. The size of
bimetallic nanoparticles was so small that cannot be detected by
SEM. Small irregular holes having no characteristic size or shape
among agglomerates were also detected. There is no obvious
morphological difference between the fresh and spent samples.
The catalytic activity of the bimetallic catalysts prepared by the
colloid method for water-gas shift reaction as a function of FeRu
content is plotted in Fig. 4. The FeRu bimetallic nanoparticles with
FeRu molar ratio 1:1 and average diameter 5 nm were used in all
samples. The changed trend of catalyst activity at 300 8C was
similar to that at 265 8C, carbon mono-oxide conversion was first
increased with increasing the amount of bimetallic nanoparticles
loading up to 1.0 wt% and then decreased with the further increase.
The decrease of catalysts activity at high active species content
may be caused by the agglomeration of bimetallic nanoparticles. So
an optimum bimetallic nanoparticles loading for preparation of the
supported catalysts with high catalytic activity is 1.0 wt%.
It is well known that potassium was used as electronical
promoters in iron-based water-gas shift catalysts. Potassium can
enhance the CO2 selectivity as well as catalytic activity of
supported catalysts in this study significantly. It is well known
that ruthenium belongs to platinum-group metals exhibiting
strong hydrogenating property [21]. In the case of WGSR, the
mixture of substrates and products creates the atmosphere
favorable for hydrogenation reaction whose main product is
methane [22–24], an undesirable product. According to literature
data, ruthenium supported on oxides with acidic character gives
catalysts of high activity in CO hydrogenation to hydrocarbons
[22–25], while that supported on basic oxides show lower activity
in CO hydrogenation [26–28]. Adding potassium as promoter can
enhance the surface basicity of MgAl2O4 spinel and affect the
reaction direction. The influence of potassium loading on catalytic
activity and CO2 selectivity of supported catalysts at temperature
265 8C was shown in Fig. 5. It was obvious that potassium loading
Fig. 3. X-ray diffraction patterns of catalysts Ru/MA (1), FeRu/MA (2), Fe/MA (3) and FeRu/MA(im) before (a) and after used (b). The diffraction peaks of MgAl2O4 are marked
with asterisk.

1350
J.-Q. Du et al. / Materials Research Bulletin 44 (2009) 1347–1351
Fig. 4. Dependence of the conversion of CO on the loading of FeRu nanoparticles.
has great influence on the catalytic activity. As potassium oxide
concentration increases from 0 to 6.0 wt%, the catalytic activity get
a rapid rise from 2.5% to 87.3%. And then the catalytic activity
decreases with further increasing the content of potassium. The
CO2 selectivity increases from 46% to 99% when the potassium
oxide concentration increasing from 0 to 4.0 wt%, and reaches
100% with further increase of potassium loading. Thus, an
optimum concentration of K2O is 6.0 wt% under the comprehen-
sive actions of the above two factors. Besides the effects of property
of support, the metal crystallite size, the metal dispersion and the
nanoparticles-support interaction may have influence on the
activity and selectivity of the supported catalysts.
The catalytic activities of catalysts FeRu/MA with Fe/Ru molar
ratio 1.5:1, 1:1 and 0.5:1, FeRu/MA(im), Fe/MA and Ru/MA as a
function of reaction temperature are plotted in Fig. 6. The activity
of catalyst FeRu/MA increases slowly when the reaction tempera-
ture increasing from 120 to 235 8C, and then there is an abrupt
increase in the catalytic activity when the reaction temperature
increasing from 235 to 265 8C. The CO conversion on catalyst FeRu/
MA at 300 8C is obviously higher than that over monometallic
catalysts Fe/MA and Ru/MA. The comparison clearly indicates that
the bimetallic catalyst has much higher activity for water-gas shift
reaction due to the synergistic effect between iron and ruthenium.
The unique catalytic properties for water-gas shift reaction are due
to the unique electronic and atomic structures of the bimetallic
surfaces, which are quite different from those of the pure metals. It
is interesting to compare the activity of catalyst FeRu/MA with that
of the catalyst FeRu/MA(im). By comparison, it is evident that the
catalyst prepared by colloid method is superior in catalytic activity
than the catalyst prepared by impregnation method. This can be
attributed to the fact that not all of iron and ruthenium species in
catalyst FeRu/MA(im) can form alloy particles.
Fig. 6. Dependence of the conversion of CO on the reaction temperature over
catalyst Fe/MA (a), Ru/MA (b), FeRu/MA(im) (c), FeRu/MA with Fe/Ru 1.5:1 (d),
FeRu/MA with Fe/Ru 0.5:1(e) and FeRu/MA with Fe/Ru 1:1 (f).
The catalyst FeRu/MA with Fe/Ru molar ratio 1:1 shows superior
catalytic activity than catalysts with Fe/Ru molar ratio of 1.5:1 and
0.5: 1(see Fig. 6). The electronic properties of bimetallic nanopar-
ticles have close relationship with their compositions. It has been
reported that the carbon nanotubes can be synthesized from CoMo
nanoparticles with different compositions and the catalyst activity
can be modulated by changing the Mo content [29]. The author
suggested that this phenomenon is related to the production rate of
carbon atoms as Mo is inferior to Co in respect of adsorption and
decomposing CO to release carbon atoms. In the present study, this
would be consistent with the adsorption of CO being an important
factor and it is possible that the composition may influence
the catalyst activity. However, more detailed study is needed to
elucidate the relationship between the properties of nanoparticles
and their catalytic activity.
In conclusion, Fe–Ru bimetallic nanoparticles with different
diameters and compositions have been prepared by a microwave
assisted polyol method. Most of the bimetallic nanoparticles
appeared to be single crystals based on the results of HRTEM. The
bimetallic nanoparticles were loaded onto MgAl2O4 and tested for
water-gas shift reaction. The supported catalyst shows very high
catalytic activity due to the synergistic effect between iron and
ruthenium and the optimum loading of FeRu nanoparticles was
1.0 wt%. The compositions of nanoparticles have influence on the
catalysts activity, and catalyst with Fe/Ru molar ratio 1:1 gave the
best result. Potassium can enhance the CO selectivity of catalysts
significantly besides increasing the catalysts activity, and the
optimum content of K2O was 6.0 mol%.
Reference
[1] X. Duan, Y. Huang, R. Agarwal, C.M. Lieber, Nature 421 (2003) 241–245.
[2] I. Arnaud, J.P. Abid, C. Roussel, H.H. Girault, Chem. Commun. (2005) 787–788.
[3] M.M. Miller, A.A. Lazarides, J. Phys. Chem. B 109 (2005) 21556–21565.
[4] P.V. Kamat, M. Flumiani, A. Dawson, Colloids Surf. A 202 (2002) 269–279.
[5] J.P. Abid, H.H. Girault, P.F. Brevet, Chem. Commun. (2001) 829–830.
[6] U. Jeong, J.U. Kim, Y. Xia, Nano Lett. 5 (2005) 937–942.
[7] C.S. Ah, S.D. Hong, D.J. Jang, J. Phys. Chem. B 105 (2001) 7871–7873.
[8] A. Rai, M. Chaudhary, A. Ahmad, S. Bhargava, M. Sastry, Mater. Res. Bull. 42 (2007)
1212–1220.
[9] C. Feldmann, H.-O. Jungk, Angew. Chem., Int. Ed. 40 (2001) 359–362.
[10] P.Y. Silvert, R. Herrera-Urbina, K. Tekaia-Elhsissen, J. Mater. Chem. 7 (1997) 293–
299.
[11] G. Viau, R. Brayner, L. Poul, N. Chakroune, E. Lacaze, F. Fie´vet-Vincent, F. Fie´vet,
Chem. Mater. 15 (2003) 486–494.
[12] C. Feldmann, C. Metzmacher, J. Mater. Chem. 11 (2001) 2603–2606.
[13] Q. Li, Y. Wei, Mater. Res. Bull. 33 (1998) 779–782.
Fig. 5. Dependence of the CO2 selectivity and CO conversion on the loading of K2O
The reaction temperature is 265 8C.
[14] H. Yang, C. Huang, A. Tang, X. Zhang, W. Yang, Mater. Res. Bull. 40 (2005) 1690–
1695.

J.-Q. Du et al. / Materials Research Bulletin 44 (2009) 1347–1351
1351
[15] Y. Liu, Q. Fu, M. Flytzani-Stephanopoulos, Catal. Today 93–95 (2004) 241–246.
[16] A. Basin´ ska, F. Domka, Catal. Lett. 43 (1997) 59–61.
[23] M.A. Vannice, J. Catal. 50 (1977) 228–236.
[24] M.A. Vannice, Catal. Rev. Sci. Eng. 14 (1976) 153–191.
´
[17] A. Basinska, F. Domka, Appl. Catal. A 179 (1999) 241–246.
[25] N.M. Gupta, V.P. Londhe, V.S. Kamble, J. Catal. 169 (1997) 423–437.
[26] M.E. Dry, G.J. Oosthuizen, J. Catal. 11 (1968) 18–24.
[27] K. Askura, Y. Iwasawa, J. Chem. Soc., Faraday Trans. 86 (1990) 2657–2662.
[28] T. Ishihara, K. Harada, K. Eguchi, H. Arai, J. Catal. 136 (1992) 161–169.
[29] X.Z. Liao, A. Serquis, Q.X. Jia, D.E. Peterson, Y.T. Zhu, H.F. Xu, Appl. Phys. Lett. 82
(2003) 2694–2696.
[18] A. Venugopal, M.S. Scurrell, Appl. Catal. A 245 (2003) 137–147.
[19] A. Venugopal, J. Aluha, D. Mogano, M.S. Scurrell, Appl. Catal. A 245 (2003) 149–158.
[20] J. Guo, H. Lou, H. Zhao, X. Wang, X. Zheng, Mater. Lett. 58 (2004) 1920–1923.
[21] E.A. Seddon, K.R. Seddon, The Chemistry of Ruthenium, Elsevier, New York, 1984.
[22] M.A. Vannice, J. Catal. 37 (1975) 449–461.