Synthesis and electromagnetic wave absorption property of Ni-Ag alloy nanoparticles

Article abstract of DOI:10.1016/j.jallcom.2009.02.017

Ni-Ag alloy nanoparticles with various molar ratios were synthesized and embedded in the epoxy for the investigation of electromagnetic wave absorption properties in 2-40 GHz. The reflection loss measurement indicated Ni nanoparticles had a significant ab

Full text of DOI:10.1016/j.jallcom.2009.02.017

Journal of Alloys and Compounds 480 (2009) 674–680
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Synthesis and electromagnetic wave absorption property of
Ni–Ag alloy nanoparticles
∗
Chung-Che Lee, Ya-Yi Cheng, Hsiang Yu Chang, Dong-Hwang Chen
Department of Chemical Engineering, National Cheng Kung University, Tainan 701, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
Ni–Ag alloy nanoparticles with various molar ratios were synthesized and embedded in the epoxy for
the investigation of electromagnetic wave absorption properties in 2–40 GHz. The reflection loss mea-
surement indicated Ni nanoparticles had a significant absorption in 18–40 GHz while no significant
absorption was observed in the whole frequency range examined for Ag nanoparticles. Ni1Ag1 nanoparti-
cles exhibited a significant absorption as Ni nanoparticles did, but the absorption frequency range shifted
from 18–40 to 2–18 GHz. Interestingly and notably, both the Ni3Ag1 and Ni1Ag3 nanoparticles showed
dual-frequency absorption in 2–18 and 18–40 GHz. It was suggested that the Ni- and Ag-rich micro-
domains might be formed within the Ni3Ag1 and Ni1Ag3 nanoparticles, leading to the appearance of
the second absorption due to the lags of polarization at the interfaces as the frequency was varied. The
composition-related microstructure also affected the permittivity and permeability more significantly
than the composition itself.
Received 6 November 2008
Received in revised form 3 February 2009
Accepted 3 February 2009
Available online 20 February 2009
Keywords:
Metals and alloys
Nanostructured materials
Composite materials
Chemical synthesis
Microstructure
©
2009 Elsevier B.V. All rights reserved.
1
. Introduction
Due to the rapid development of science and technology, now
the EM wave absorbers are further required to possess more
advantages such as light-weight and thinness, wide absorption
frequency, multi-frequency absorption, high thermal stability, and
anti-oxidation. Metal magnetic nanoparticles may be quite suitable
as EM wave absorbers because of their small size, large specific
surface area, high surface atom percentage and more dangling
chemical bonds, which might lead to the interface polarization and
the activation of nanoparticles. However, each kind of metal mag-
netic nanoparticles has its intrinsic properties and usually can be
used only in a specified frequency range. To meet the above require-
ments, the development of composite metal nanoparticles which
combine two or more components in each individual particle to
modify the bulk properties [6–11] or the surface properties may be
an efficient strategy [12–20].
Ni is an important soft magnetic metal and Ag is known to
possess high electric conductivity. Their combination as an EM
wave absorber may be interesting. Recently we have prepared
Ni, Ag, and Ni–Ag core–shell nanoparticles by various methods
[21–25], and found that Ni–Ag core–shell nanoparticles showed a
novel dual-frequency absorption property in 2–18 and 18–40 GHz
although Ni nanoparticles exhibited absorption only in 18–40 GHz
and no absorption was observed for Ag nanoparticles in 2–40 GHz
Because of the advantage of large data transmission, various
electronic and communication devices using the electromagnetic
(
EM) wave in the range of 1–40 GHz are being developed rapidly
in the recent years [1]. For the normal operation of devices and
human health, the development of the corresponding EM wave
absorbers becomes more and more important [2,3]. In general, the
shielding efficiency of EM wave absorbers depends on the match-
ing frequency, layer thickness of absorbers, and the relative complex
permeability and permittivity of the fillers which are determined
by their nature, shape, size, and microstructure [2,4,5]. Owing to
the high Snoek’s limit, metallic soft magnetic materials may be
more suitable as the microwave absorbers in the frequency range
over gigahertz compared to the conventional ferrites [4,5]. How-
ever, to avoid the decrease of the relative complex permeability
due to the eddy current loss, the metallic particles must be isolated
with size less than the skin depth [4,5]. So, the studies on the metal-
lic magnetic sub-micro particles and nanocomposites embedded in
the non-conductive matrixes for the EM wave absorption received
increasingly attention in recent years [1,2,4,5].
[
26]. The additional absorption of Ni–Ag core–shell nanoparticles
in 2–18 GHz was attributed to the lags of polarization between
the core/shell interfaces as the frequency was varied, which con-
tributed to the dielectric loss [5,26]. Compared to the core–shell
nanoparticles, alloy nanoparticles are relatively easily synthesized
∗ Corresponding author at: National Cheng Kung University, Department of Chem-
ical Engineering, No 1, Ta-Hsueh Rd, Taiwan 701,Taiwan.
Tel.: +886 6 2757575x62680; fax: +886 6 2344496.
E-mail address: chendh@mail.ncku.edu.tw (D.-H. Chen).
0925-8388/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2009.02.017

C.-C. Lee et al. / Journal of Alloys and Compounds 480 (2009) 674–680
675
and their properties can be tuned by varying the composition.
So, in this work, we prepared the Ni–Ag alloy nanoparticles by
the hydrazine co-reduction of Ni and Ag ions in ethylene gly-
col and investigated their EM wave absorption properties when
embedded in the epoxy resin. The result indicated that Ni–Ag
alloy nanoparticles indeed exhibited significant mono- or dual-
frequency absorption in 2–40 GHz, depending on the composition.
From the investigations on the basic properties of alloy nanopar-
ticles as well as the frequency dependences of permittivity and
permeability, it was suggested that the Ni- and Ag-rich micro-
domains might be formed within the Ni–Ag alloy nanoparticles.
Also, the EM wave absorption properties of Ni–Ag alloy nanopar-
ticles were affected more significantly by the composition-related
microstructure than by the composition itself.
2. Experimental
Fig. 1. UV–vis absorption spectra for the colloid dispersion of Ni, Ag, and their alloy
nanoparticles with various Ni/Ag molar ratios.
2.1. Preparation of Ni–Ag alloy nanoparticles
Ni–Ag alloy nanoparticles were synthesized by the co-reduction of nickel nitrate
and ꢂ were determined in terms of S11 and S21 to be
and silver nitrate in ethylene glycol with hydrazine and polyethyleneimine (PEI) as
the reducing agent and protective agent, respectively. Firstly, 0.1 M of metal pre-
cursor (500 ␮L/mL), 0.1 M NaOH (100 ␮L/mL), and 20 M hydrazine (50 ␮L/mL) were
ꢂ
2
(S2 − S2 + 1) ±
(S2 − S2 + 1) − 4S2
11
21
11
21
11
ꢃ =
(3)
(4)
2
2S
◦
11
added into ethylene glycol in sequence and then the solution was heated to 60 C.
ꢂ
When the solution became grey-black which indicated the formation of Ni–Ag alloy
2
2
2
2
2
2
◦
1 (S11 − S21 + 1) ±
(S11 − S21 + 1) − 4S21
nanoparticles, the solution was immediately cooled in an ultrasonic bath at 0 C and
ꢂ = −
2
PEI (5 wt%) was added to prevent the agglomeration of Ni–Ag alloy nanoparticles.
After about 30 min, the product was recovered magnetically and then washed with
ethanol to remove the extra ethyl glycol and PEI. Finally, the product was dried at
room temperature in a vacuum oven. The total metal ion concentration was fixed
at 50 mM. By varying the concentration ratio of nickel nitrate to silver nitrate, the
composition of Ni–Ag alloy nanoparticles could be adjusted. In the absence of silver
nitrate or nickel nitrate, pure Ni and Ag nanoparticles could be obtained, respec-
tively. For the synthesis of pure Ag nanoparticles, the product was recovered by
centrifugation.
The morphology and particle size were observed by transmission electron
microscopy (TEM) using a JEOL Model JEM-1200EX transmission electron micro-
scope at 80 kV. The structure was determined by X-ray diffraction (XRD) on a
Shimadzu Model RX-III X-ray diffractometer at 40 kV and 30 mA with Cu K␣ radi-
ation (ꢀ = 0.1542 nm). Magnetic measurement was done using a superconducting
quantum interference device (SQUID) magnetometer (MPMS7, Quantum Design).
The UV–vis absorption spectrum of metal colloid dispersion which was obtained by
dispersing PEI-protected metal nanoparticles in ethanol was analyzed by a Hitachi U-
d
2S
2
1
where d is the absorber thickness. From the the ratios of the permittivity and
permeability of absorber (ε and ꢁ) to the permittivity and permeability of free space
ε0 and ꢁ0), relative permittivity εr (=ε/ε0) and permeability ꢁr (=ꢁ/ꢁ0) could be
obtained respectively. The RL curves at various frequencies and absorber thicknesses
could be calculated from the relative permeability and permittivity according to the
following equations [1,4,6]:
(
ꢀ
ꢁ
ꢃ ꢀ
ꢁ
ꢄ
ꢁ
r
1
2
2ꢄfd
c
1
2
Zin = Z0
tanh
j
(ꢁrεr)
(5)
(6)
εr
ꢅ
ꢅ
ꢅ
ꢅ
(Z − Z )
in
0
ꢅ
ꢅ
RL = 20 log
(
Zin + Z0)
where f is the frequency of electromagnetic wave, c is the velocity of light, Z0 is
the impedance of free space, and Zin is the input impedance of absorber.
3
000 spectrophotometer equipped with a 10 mm quartz cell. The real compositions
3
. Results and discussion
of Ni–Ag nanocrystals were determined by dissolving the sample in a concentrated
HCl/HNO3 (3:1, v/v) mixture solution and analyzing the solution composition using
a GBC Model SDS-270 atomic absorption spectrometer (AAS).
3.1. Size, structure, and optical and magnetic properties of Ni–Ag
alloy nanoparticles
2.2. Measurement of electromagnetic wave absorption property
Fig. 1 shows the typical UV–vis absorption spectra for the colloid
dispersion of Ni, Ag, and their alloy nanoparticles with various Ni/Ag
molar ratios. Obviously, Ag nanoparticles exhibited their charac-
teristic absorption band at 394 nm [28]. After incorporating Ni
element, the characteristic absorption band essentially remained
unchanged but the absorbance decreased significantly with the
increase of Ni content. Because the nanoparticles were recovered
magnetically, they must contain Ni element in each particle. Also, as
illustrated in Fig. 1, Ni nanoparticles did not show any characteristic
absorption in the examined wavelength range. So the characteris-
tic absorption of nanoparticles could be attributed to the surface
plasmon resonance of Ag. This provided an evidence for the forma-
tion of Ni–Ag alloy nanoparticles and revealed the electron cloud
oscillation of surface Ag atoms might be perturbed by Ni atoms.
Fig. 2a–e shows the typical TEM images and particle size dis-
tributions of Ni, Ag, and their alloy nanoparticles with various
Ni/Ag molar ratios. Ni3Ag1, Ni1Ag1, and Ni1Ag3 denote the Ni–Ag
alloy nanoparticles obtained at the initial Ni/Ag molar ratios of 3/1,
Epoxy resin composites were prepared as follows. Firstly, reagent A and reagent
B of epoxy resin were separately added into two ethanol solutions with the same
content of PEI-protected metal nanoparticles. Secondly, they were ultra-sonicated
to ensure the uniform dispersion and then heated to remove ethanol. When most of
ethanol was evaporated, these two solutions were mixed homogeneously. Finally, by
casting the mixture on a transparency film and drying in air, epoxy resin composite
could be obtained. The ratio of metal nanoparticles to the epoxy resin was fixed
at 4:5 by weight. The composites were cured and then cut into sheet samples of
1
5 cm × 15 cm with thicknesses of about 1.5 mm. The reflection loss (RL) of each
sheet sample backed by the same-sized reference metal plate at frequencies of 2–18
and 18–40 GHz was measured using free space method developed by Damaskos Inc.,
known as free-space antenna-based inverted arch system, and a HP8722ES network
analyzer. Both low band free space setup (DI Inverted Arch) and high band free space
setup (DI mm-Wave Arches) were used to measure the reflection characteristics of
metal-backed sheet sample at frequencies sweeping from 2 to 18 GHz and 18 to
4
0 GHz, respectively. Complex permittivity ε and permeability ꢁ were determined
from the scattering parameters S11 and S21 (corresponding to the scattering reflection
and transmission coefficients, respectively) measured by the HP8722ES network
analyzer according to the following equations [27]:
ꢀ
ꢁ
ꢂ
1 − ꢃ
1 + ꢃ
ε =
(1)
(2)
ꢂ0
1
/1, and 1/3, respectively. It was obvious that they all were dis-
ꢀ
ꢁ
ꢂ
1 + ꢃ
1 − ꢃ
crete without agglomeration. The mean diameters of Ni, Ag, Ni Ag ,
3
1
ꢁ
=
ꢂ0
Ni Ag , and Ni Ag nanoparticles were 6.5, 6.3, 6.7, 7.2, and 6.7 nm,
1
1
1
3
respectively. The composition dependence of particle size was illus-
trated in Fig. 2f, in which the error bars indicated the standard
where ꢂ0 is the propagation constant in free space, ꢂ is the propagation constant
in the absorber, and ꢃ is the reflection coefficient between air and the absorber. ꢃ

676
C.-C. Lee et al. / Journal of Alloys and Compounds 480 (2009) 674–680
Fig. 2. Typical TEM images and particle size distributions of Ni (a), Ni3Ag1 (b), Ni1Ag1 (c), Ni1Ag3 (d), and Ag (e) nanoparticles and their mean diameter as a function of
composition (f).
deviations. It was found that the mean diameters of Ni–Ag alloy
nanoparticles were slightly larger than those of pure Ni and Ag
nanoparticles and a maximum seemed to occur when the Ni/Ag
molar ratio was 1. Although this effect was not remarkable while
taking the particle size distribution into account, it still implied that
the reduction, nucleation, and growth processes might be slightly
affected by the composition of reaction solution. In addition, by
AAS the real mole fractions of Ag in the products Ni Ag , Ni Ag ,
and initial composition confirmed the formation of Ni–Ag alloy
nanoparticles.
The XRD patterns of Ni, Ag, and their alloy nanoparticles with
various Ni/Ag molar ratios were indicated in Fig. 3a. It was found
that Ni nanoparticles showed three main characteristic peaks for
the (1 1 1), (2 0 0) and (2 2 0) planes of face-centered cubic (fcc) Ni
◦
at 2Â=44.8, 52.2 and 76.8 . Also, Ag nanoparticles exhibited four
main characteristic peaks for the (1 1 1), (2 0 0), (2 2 0), and (3 1 1)
3
1
1
1
◦
and Ni Ag nanocrystals were determined to be 25.6, 56.1, and
planes of fcc Ag at 2Â=38.1, 44.3, 64.4, and 77.5 . For Ni–Ag alloy
1
3
7
3.5%, respectively. Because Ni–Ag alloy nanoparticles were recov-
nanoparticles, five characteristic peaks were observed for all of
Ni Ag , Ni Ag , and Ni Ag nanoparticles. Although the character-
ered magnetically, the consistency between the real compositions
3
1
1
1
1
3
Fig. 3. XRD patterns of metal nanoparticles with various Ni/Ag molar ratios (a) and the composition dependence of the diffraction angles for Ni(1 1 1) and/or Ag(2 0 0) planes
b).
(

C.-C. Lee et al. / Journal of Alloys and Compounds 480 (2009) 674–680
677
istic peaks for the (1 1 1) and (2 2 2) planes of fcc Ni were overlapped
with those for the (2 0 0) and (3 1 1) planes of Ag, the (1 1 1) and
2 2 0) planes of fcc Ag and the (2 0 0) plane of fcc Ni were observed
(
separately [29]. This confirmed the presence of Ni and Ag in the
resultant alloy nanoparticles. Also, they all had the fcc structure. In
addition, it was noted that the diffraction angles of Ni, Ag, and their
alloy nanoparticles for Ni(1 1 1) and/or Ag(2 0 0) planes increased
slightly with increasing Ni content. Their dependence on the com-
position was shown in Fig. 3b. According to the Vegard’s law, the
diffraction peak of the metal alloy with homogeneous composition
should lie between the two set peaks of pure metals and change
linearly [30]. Fig. 3b indicates the diffraction peak of Ni–Ag alloy
nanoparticles indeed lay between those of pure Ni and Ag, revealing
the formation of metal alloy. However, the composition dependence
deviated from the linearity slightly, implying Ni and Ag atoms might
be not homogeneously distributed throughout the bulk phase of
alloy nanoparticles. This might be resulted by the differences in the
reduction, nucleation, and growth rates of nickel and silver ions,
consistent with the suggestion from the composition dependence
of particle size [31].
Because of the presence of Ni and Ag, Ni–Ag alloy nanopar-
ticles not only showed the surface plasmon absorption but also
were expected to possess magnetic property. The typical plots of
◦
magnetization versus magnetic field (M-H loop) at 27 C for Ni
nanoparticles and the Ni–Ag alloy nanoparticles with various Ni/Ag
molar ratios were indicated in Fig. 4. It was obvious that no sig-
nificant hysteresis phenomenon was observed for each case. This
revealed that the resultant Ni and Ni–Ag alloy nanoparticles were
nearly superparamagnetic, which could be attributed to their small
sizes. From Fig. 4 and its enlargement near the origin as shown in
the inset, the saturation magnetization (Ms), remanent magnetiza-
tion (Mr), and coercivity (Hc) could be determined. For Ni, Ni Ag ,
Fig. 5. Reflection loss curves for the epoxy resin composites containing Ni, Ag, and
their alloy and core–shell nanoparticles in the frequency ranges of 2–18 (a) and
3
1
18–40 GHz (b).
Ni Ag , and Ni Ag nanoparticles, the Ms values were 47.8, 26.2,
1
1
1
3
1
8.9, and 3.1 emu/g, respectively; the Mr values were 7.7, 7.2, 2.6,
and 0.9 emu/g, respectively; the Hc values were 124, 156, 94.6, and
49 Oe, respectively. It was obvious that both the Ms and Mr val-
dependence of Hc value for Ni–Ag alloy nanoparticles implied the
presence of the difference in their microstructures at various Ni/Ag
molar ratios.
1
ues decreased with the decrease in the content of Ni. This was
reasonable because the magnetic property originated from the Ni
element. As for the coercivity, the Hc values of Ni Ag1 and Ni Ag3
3.2. Electromagnetic wave absorption properties of Ni–Ag
nanoparticles
3
1
nanoparticles were similar but they were larger than that of Ni Ag1
1
nanoparticles. As stated above, the differences in the reduction,
nucleation, and growth rates of nickel and silver ions might lead
to the inhomogeneous distribution of Ni and Ag atoms in the bulk
phase of alloy nanoparticles. Also, this effect depended on the com-
position and might result in the variation of particle microstructure
From the reflection measurements, the RL curves for the epoxy
resin composites containing Ni–Ag alloy nanoparticles in the fre-
quency ranges of 2–18 and 18–40 GHz were obtained as indicated
in Fig. 5, in which the data for Ni, Ag and Ni–Ag core–shell nanopar-
ticles were also given for comparison. Obviously, the EM wave
absorption properties of Ni–Ag alloy nanoparticles depended on the
composition. It was found that Ni Ag nanoparticles exhibited a sig-
[
32]. So, in addition to the experimental errors, the composition
1
1
nificant absorption as Ni nanoparticles did, but the frequency range
shifted from 18–40 to 2–18 GHz with a maximum reflection loss of
1
9.5 dB at 9.4 GHz. Notably, both the Ni Ag1 and Ni Ag3 nanoparti-
3 1
cles showed dual-frequency absorption in 2–18 and 18–40 GHz as
Ni–Ag core–shell nanoparticles did [26]. For Ni Ag1 nanoparticles,
3
one maximum reflection loss of 7.3 dB at 7.1 GHz and the other max-
imum reflection loss of 15.0 dB at 36.5 GHz were observed. In the
case of Ni Ag nanoparticles, two maximum reflection losses of 11.7
1
3
and 18.9 dB appeared at 7.7 and 30.3 GHz, respectively. To our best
knowledge, such a dual-frequency EM wave absorption property of
alloy nanoparticles has not been reported. Because the composition
dependences of size, diffraction angle, and coercivity all implied
that Ni and Ag atoms might be not homogeneously distributed in
the bulk phase of alloy nanoparticles as stated above, it was sug-
gested that the composition dependence of the EM wave absorption
properties of Ni–Ag alloy nanoparticles might be referred to the
difference in their microstructures. That is, the dispersed Ni- and
Ag-rich micro-domains might be formed in the bulk phases of
◦
Fig. 4. Typical plots of magnetization versus magnetic field (M-H loop) at 27 C for
Ni nanoparticles and the Ni–Ag alloy nanoparticles with various Ni/Ag molar ratios.
The inset indicates the enlargement near the origin.

678
C.-C. Lee et al. / Journal of Alloys and Compounds 480 (2009) 674–680
ꢀ
ꢀ
ꢀꢀ
ꢀꢀ
Fig. 6. Frequency dependences of the real part (εr and ꢁ ) and imaginary part (ε and ꢁ ) of the relative complex permittivity for the epoxy resin composites containing
r
r
r
Ni1Ag1 (a–b), Ni1Ag3 (c–f), or Ni3Ag1 (g–j) nanoparticles.
Ni Ag and Ni Ag nanoparticles, respectively, which led to the
appearance of the second absorption due to the lags of polarization
at the interfaces as observed for Ni–Ag core–shell nanoparticles. As
and permeability for the epoxy resin composites containing Ni–Ag
alloy nanoparticles in the frequency ranges where they exhibited
significant absorption were shown in Fig. 6. In 18–40 GHz, the fre-
3
1
1
3
ꢀ ꢀꢀ
quency dependences of εr and εr values for Ni Ag3 and Ni Ag1
1 3
for the Ni Ag nanoparticles, the distribution of Ni and Ag atoms
1
1
in the bulk phase of alloy nanoparticles might be homogeneous
relatively because their content was equal. So, only an absorption
band was observed and the shift of absorption frequency range from
nanoparticles were similar to those of Ni nanoparticles observed in
ꢀ
our previous work [26]. Furthermore, the ꢁ values for both Ni Ag
r
1
3
and Ni Ag1 nanoparticles were slightly higher than those for Ni
3
ꢀ
ꢀ
18–40 GHz to 2–18 GHz reflected the change in the nature of Ni
nanoparticles, and the frequency at which a maximum ꢁr value
nanoparticles after the incorporation of Ag element into their bulk
phase.
appearedwas variedwiththe composition. Obviously, no consistent
ꢀ
ꢀꢀ
tendency was observed for the variations of ꢁ , and ꢁ values with
r
r
ꢀ
ꢀ
The frequency dependences of the real parts (ε and ꢁ ) and
Ni content. In 2–18 GHz, it was found that the frequency depen-
r
r
ꢀ
ꢀ
ꢀꢀ
ꢀ ꢀꢀ ꢀ
dences of the ε , ε , ꢁ , and ꢁ values had no significant tendencies
r r r
ꢀꢀ
imaginary parts (εr and ꢁ ) of the relative complex permittivity
r
r

C.-C. Lee et al. / Journal of Alloys and Compounds 480 (2009) 674–680
679
ꢀ
ꢀ
Furthermore, it was noted that negative ꢁ values were observed
r
in some conditions as indicated in Fig. 6. Similar phenomenon was
also found in our previous work [26]. This might be meaningless in
physics because these values were obtained by simulation based on
the scattering parameters and the negative values might have just
arisen from noise or experimental artifice. However, theoretically,
the metallic magnetic nanoparticles might be used as the material
with negative reflection (so called left-handed materials) whose
permittivity and permeability were negative [33–38]. So, this also
might imply that Ni–Ag alloy nanoparticles had the potential as
the left-handed materials. More studies and precise measurement
were necessary.
The layer thickness of absorbers is also an important factor
influencing the shielding efficiency of absorbers. According to equa-
tions (5) and (6), the RL curves for the epoxy resin composites
containing Ni–Ag alloy nanoparticles at various thicknesses in
the frequency where they exhibited significant absorption could
be calculated as indicated in Fig. 7. For Ni Ag1 nanoparticles,
1
a maximum reflection loss of 21.7 dB at 10.6 GHz appeared at
an absorber thickness of 1.8 mm. For Ni Ag1 nanoparticles, a
3
maximum reflection loss of 16.3 dB at 8.6 GHz appeared at an
absorber thickness of 1.0 mm and a maximum reflection loss of
3
2.4 dB at 38.4 GHz at an absorber thickness of 0.8 mm. For Ni Ag3
1
nanoparticles, a maximum reflection loss of 20.3 dB at 10.5 GHz
appeared at an absorber thickness of 1.2 mm and a maximum
reflection loss of 25.7 dB at 29.0 GHz at an absorber thickness of
2.2 mm.
4. Conclusions
Discrete Ni–Ag alloy nanoparticles with various molar ratios
have been synthesized by the co-reduction of nickel nitrate and sil-
ver nitrate in ethylene glycol with hydrazine and PEI as the reducing
agent and protective agent, respectively. They showed the surface
plasmon absorption at 394 nm and were nearly superparamag-
netic, with the fcc structure and the mean diameters of 6.3–7.2 nm.
From the composition dependences of size, diffraction angle, and
magnetic property, it was suggested that Ni and Ag atoms might
not be homogeneously distributed in the bulk phase of composite
nanoparticles due to the differences in the reduction, nucleation,
and growth rates of nickel and silver ions. From the measurement of
the reflection loss in 2–18 and 18–40 GHz, it was found that Ni Ag1
1
nanoparticles exhibited a significant absorption as Ni nanoparti-
cles did, but both the Ni Ag1 and Ni Ag3 nanoparticles showed
3
1
dual-frequency absorption as Ni–Ag core–shell nanoparticles did.
It was suggested that Ni- and Ag-rich micro-domains might be
formed in the bulk phases of Ni Ag1 and Ni Ag3 nanoparticles,
3
1
respectively, but the distribution of Ni and Ag atoms in the bulk
phase of Ni Ag1 nanoparticles might be homogeneous relatively.
1
The presence of dispersed micro-domains might result in the lags
of polarization at the interfaces as the frequency was varied, and
lead to the appearance of an additional absorption band which
contributed to the dielectric loss. In addition, it was suggested
that the permittivity and permeability of Ni–Ag nanoparticles were
affected more significantly by the composition-related microstruc-
ture than by the composition itself. This result should be helpful for
the development of multi-frequency EM wave absorption materi-
als.
Fig. 7. Reflection loss curves for the epoxy resin composites containing Ni1Ag1 (a),
Ni3Ag1 (b–c), and Ni1Ag3 (d–e) nanoparticles at various absorber thicknesses.
consistent with the variation of composition. In fact, the frequency
ꢀ
ꢀꢀ
ꢀ
ꢀꢀ
dependences of the ε , ε , ꢁ , and ꢁ values for Ni Ag3 nanopar-
r
r
r
r
1
ticles were more similar to those for Ni Ag1 nanoparticles while
3
comparing with Ni Ag1 nanoparticles. Because the relative com-
1
plex permeability and permittivity are determined not only by the
nature of fillers but also by their microstructures [2,4,5], it was sug-
gested that the composition-related microstructure as mentioned
above might play a more important role than the composition itself
on influencing the relative permittivity and permeability of Ni–Ag
alloy nanoparticles. This result provides a new idea for the design
or development of alloy nanoparticles-based EM wave absorption
materials.
Acknowledgments
We are grateful to the Chung-Shan Institute of Science & Tech-
nology for the financial and technical support of this research
(contract no. XD95026P), and to Dr. Jenq-Der Tsou for his valuable
knowledge on this work.

6
80
C.-C. Lee et al. / Journal of Alloys and Compounds 480 (2009) 674–680
[20] S. Zhou, B. Varughese, B. Eichhorn, G. Jackson, K. McIlwrath, Angew. Chem. Int.
References
Ed. 44 (2005) 4539–4543.
[
21] D.H. Chen, S.H. Wu, Chem. Mater. 12 (2000) 1354–1360.
[
1] J.R. Liu, M. Itoh, T. Horikawa, E. Taguchi, H. Mori, K. Machida, Appl. Phys. A 82
[22] D.H. Chen, C.H. Hsieh, J. Mater. Chem. 12 (2002) 2412–2415.
[23] S.H. Wu, D.H. Chen, J. Colloid. Interface Sci. 259 (2003) 282–286.
[24] S.H. Wu, D.H. Chen, Chem. Lett. 33 (2004) 406–407.
[25] C.C. Lee, D.H. Chen, Nanotechnology 17 (2006) 3094.
[26] C.C. Lee, D.H. Chen, Appl. Phys. Lett. 90 (2007) 193102.
(
2006) 509–513.
[
[
2] Y. Deng, L. Zhao, B. Shen, L. Liu, W. Hu, J. Appl. Phys. 100 (2006) 14304.
3] A. Ghasemi, A. Hossienpour, A. Morisako, A. Saatchi, M. Salehi, J. Magn. Magn.
Mater. 302 (2006) 429–435.
[
[
4] J.R. Liu, M. Itoh, K.I. Machida, Appl. Phys. Lett. 88 (2006) 62503.
5] X.F. Zhang, X.L. Dong, H. Huang, Y.Y. Liu, W.N. Wang, X.G. Zhu, B. Lv, J.P. Lei, C.G.
Lee, Appl. Phys. Lett. 89 (2006) 53115.
[27] B.T. Lee, H.C. Kim, Jpn. J. Appl. Phys. 35 (1996) 3401–3406.
[28] D.H. Chen, C.J. Chen, J. Mater. Chem. 12 (2002) 1557–1562.
[29] J. Xu, T. Klassen, R.S. Averback, J. Appl. Phys. 79 (1996) 3935–3945.
[30] B.Y. Tsaur, M. Maenpaa, J. Appl. Phys. 52 (1981) 728–736.
[6] S. Sun, C.B. Murray, D. Weller, L. Folks, A. Moser, Science 287 (2000) 1989–1992.
[7] M.L. Wu, D.H. Chen, T.C. Huang, Chem. Mater. 13 (2001) 599–606.
[8] M.A. Malik, P. O’Brien, N. Revaprasadu, Chem. Mater. 14 (2002) 2004–2010.
[9] T. Gao, Q. Li, T. Wang, Chem. Mater. 17 (2005) 887–892.
[
31] B.J. Hwang, L.S. Sarma, J.M. Chen, C.H. Chen, S.C. Shih, G.R. Wang, D.G. Liu, J.F.
Lee, M.T. Tang, J. Am. Chem. Soc. 127 (2005) 11140–11145.
[32] S. Bedanta, W. Kleemann, J. Phys. D: Appl. Phys. 42 (2009) 013001.
[
[
10] T. Hirakawa, P.V. Kamat, J. Am. Chem. Soc. 127 (2005) 3928–3934.
11] X. Su, J. Zhang, L. Sun, T.W. Koo, S. Chan, N. Sundararajan, M. Yamakawa, A.A.
[
33] J.B. Pendry, A.J. Holden, W.J. Stewart, I. Youngs, Phys. Rev. Lett. 76 (1996)
4773–4776.
Berlin, Nano Lett. 5 (2005) 49–54.
[
34] R.A. Shelby, D.R. Smith, S. Schultz, Science 292 (2001) 77–79.
[
[
12] F. Caruso, Adv. Mater. 13 (2001) 11–22.
13] C.M. Niemeyer, Angew. Chem. Int. Ed. 40 (2001) 4128–4158.
14] Y. Lu, Y. Yin, B.T. Mayers, Y. Xia, Nano Lett. 2 (2002) 183–186.
15] S.T. Selvan, J.P. Spatz, H.A. Klok, M. Möller, Adv. Mater. 10 (1998) 132–134.
16] M.M. Natile, A. Glisenti, Chem. Mater. 17 (2005) 3403–3413.
[
35] R.A. Shelby, D.R. Smith, S.C. Nemat-Nasser, S. Schultz, Appl. Phys. Lett. 78 (2001)
4
89–491.
[
[
[
[
[
[
36] D.R. Smith, W.J. Padilla, D.C. Vier, S.C. Nemat-Nasser, S. Schultz, Phys. Rev. Lett.
8
4 (2000) 4184–4187.
37] S.T. Chui, L. Hu, Phys. Rev. B 65 (2002) 144407.
38] T. Kasagi, T. Tsutaoka, K. Hatakeyama, Appl. Phys. Lett. 88 (2006) 172502.
[
[
17] D. Pan, Q. Wnag, S. Jiang, X. Ji, L. An, Adv. Mater. 17 (2005) 176–179.
18] J. Jang, Y. Nam, H. Yoon, Adv. Mater. 17 (2005) 1382–1386.
[19] W. Yang, H. Araki, C. Tang, S. Thaveethavorn, A. Kohyama, H. Suzuki, T. Noda,
Adv. Mater. 17 (2005) 1519–1523.