Synthesis and characterization of Sn-rich Ni-Sb-Sn nanosolders

Article abstract of DOI:10.1016/j.jallcom.2011.10.023

Nano-crystalline samples of pure Sn and of Sn-rich ternary Ni-Sb-Sn alloys, with compositions ranging from 80 to 97.5 at% Sn and a Ni to Sb molar ratio of 1:1, were synthesized by reduction of stoichiometric metal chloride solutions with NaBH₄ at 0 °C in alkaline medium. The particle sizes of the obtained alloys, measured by TEM/SEM, were found to be in the range of 40-350 nm. A relative decrease in melting temperature of up to 15 °C was observed for these alloys compared to a bulk sample. A sample with 95 at% Sn was subjected to heat treatment at 180 °C for various time periods up to 5.0 h to study particles with well defined average sizes between 50 and 135 nm. The melting temperatures of the annealed samples were found to increase progressively with increasing particle size. At the same time, the excess surface energy of these heat treated samples was measured as a function of particle size using a Calvet-type calorimeter. From the differences of the (H₅₇₃-H₂₉₉) values between nano and bulk samples, the excess enthalpies for nano-sized samples were derived to be between 18.8 ± 1.9 and 0.8 ± 1.4 kJ g^-1 for particle sizes between 50 and 135 nm.

Full text of DOI:10.1016/j.jallcom.2011.10.023

Journal of Alloys and Compounds 513 (2012) 224–229
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Synthesis and characterization of Sn-rich Ni–Sb–Sn nanosolders
Ratikanta Mishraa, Adela Zemanovab, Ales Kroupab, Hans Flandorfera, Herbert Ipsera,∗
a University of Vienna, Department of Inorganic Chemistry/Materials Chemistry, Waehringerstr. 42, A-1090 Wien, Austria
b Institute of Physics of Materials, Academy of Sciences of the Czech Republic, CZ-61662 Brno, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 August 2011
Received in revised form 9 October 2011
Accepted 10 October 2011
Available online 18 October 2011
Nano-crystalline samples of pure Sn and of Sn-rich ternary Ni–Sb–Sn alloys, with compositions ranging
from 80 to 97.5 at% Sn and a Ni to Sb molar ratio of 1:1, were synthesized by reduction of stoichiometric
metal chloride solutions with NaBH4 at 0 ◦C in alkaline medium. The particle sizes of the obtained alloys,
measured by TEM/SEM, were found to be in the range of 40–350 nm. A relative decrease in melting
temperature of up to 15 ◦C was observed for these alloys compared to a bulk sample. A sample with
95 at% Sn was subjected to heat treatment at 180 ◦C for various time periods up to 5.0 h to study particles
with well defined average sizes between 50 and 135 nm. The melting temperatures of the annealed
samples were found to increase progressively with increasing particle size. At the same time, the excess
surface energy of these heat treated samples was measured as a function of particle size using a Calvet-
type calorimeter. From the differences of the (H573–H299) values between nano and bulk samples, the
excess enthalpies for nano-sized samples were derived to be between 18.8 1.9 and 0.8 1.4 kJ g−1 for
particle sizes between 50 and 135 nm.
Keywords:
Metals and alloys
Nanostructured materials
Nanosolders
Lead-free soldering
Chemical synthesis
Surface enthalpy
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
entropy is important in predicting the above properties of nano-
materials.
The physical, electronic and thermodynamic properties of alloys
in their nanoform are significantly different from those of the
corresponding bulk materials. It is the surface energy in these
nanomaterials that assumes significance, due to the presence of
a higher number of atoms at the surface compared to the bulk
material. The difference in properties of materials with particles
in the nanometer size regime (1–100 nm) compared to bulk mate-
rials can be related to the excess surface energy [1–3]. For example,
a depression of the melting temperature of nanoparticles has been
reported by many authors [4–8]. Various quantitative relation-
ships between the melting temperature and the particle size have
been developed, among others, by Pawlow [9], Buffat and Borel
[10], Borel [11], Dick et al. [12], Nanda et al. [13], Jiang et al. [14],
and Qi and Wang [15]. Likewise, a decrease of the Curie tem-
perature, the reduction of magnetic moments, and an improved
magneto-caloric effect for magnetic refrigeration has been reported
for magnetic nanoparticles [16–18]. In semiconductors, the band
gap increases with the decrease of particle size, and it is highly
significant when the particle sizes are comparable to or below the
exciton Bohr radius [19,20]. Nanocrystals exhibit an intermediate
behavior between bulk and isolated molecules. The knowledge of
excess thermodynamic parameters such as excess enthalpy and
Sb–Sn based alloys are possible candidates for new high-
temperature solders. The melting temperatures, mechanical
properties and quality of joints formed by these materials with
different substrates can be modified by alloying with additional
elements. There are continuing attempts to reduce the particle size
in solder powders in order to reduce the melting temperature of
solder alloys, especially in reflow soldering. Currently, the smallest
commercial particle size for lead-free solders seems to be in the
range between 2 and 8 ␮m (Type 8) [21] which is still far from the
nanometer size where a size dependent reduction of the melting
temperature would come into play. In this work we report syn-
thesis and characterization of Sn-rich Ni–Sb–Sn nanoalloys with
melting temperatures that are noticeably reduced compared to the
bulk alloy. We also report the determination of the excess surface
energy of the Ni–Sb–Sn nanoalloys as a function of particle size.
2. Experimental
Nanopowders of Sn and Sn-rich Ni–Sb–Sn alloys were synthesized following
the procedure described by Yang et al. [22]. In the first step aqueous stock solutions
of SnCl2 (0.08 M), SbCl3 (0.03 M), and NiCl2 (0.06 M) were prepared by dissolving
SnCl2·2H2O and SbCl3 (Alfa Aesar, 99.99% purity) in diluted hydrochloric acid (5 M),
and NiCl2·6H2O (Alfa Aesar, 99.9985% purity) in distilled water. Appropriate volumes
of the above metal chloride solutions were mixed together to make compositions
with Sn contents varying from 80 to 97.5 at% and with a Ni to Sb molar ratio of 1:1.
To the resulting solutions required amounts of a 1 M solution of tri-sodium citrate
was added as chelating reagent. A 0.5 M NaBH4 stock solution was prepared at a
pH value of 14 from a 4.4 M NaBH4 stable solution obtained from Alfa Aesar. The
two separate aqueous solutions were cooled to 0 ◦C in ice water before reaction.
∗
Corresponding author. Tel.: +43 1 4277 52906; fax: +43 1 4277 9529.
E-mail address: herbert.ipser@univie.ac.at (H. Ipser).
0925-8388/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jallcom.2011.10.023

R. Mishra et al. / Journal of Alloys and Compounds 513 (2012) 224–229
225
40 ml of metal chloride solution was added drop wise within 10 min to 40 ml of the
NaBH4 solution under strong magnetic stirring. With continued addition, the color
of the solution turned brown with formation of some bubbles which was followed
by the appearance of some black suspensions. The mixed solutions were stirred
for another 30 min, and for the entire period the reaction vessel was cooled in ice
water. The resultant suspensions were separated by centrifuge and rinsed several
times with distilled water and acetone. The precipitate was dried under vacuum at
room temperature.
The fine black powder samples obtained in this way were characterized by X-ray
diffraction (XRD), differential scanning calorimetry (DSC), and electron microscopic
techniques. To make particles with different crystallite sizes, an alloy with com-
position Ni2.5Sb2.5Sn95 was synthesized following the procedure described above.
Individual parts of it (about 200 mg) were sealed in evacuated quartz ampoules
and annealed at 180 ◦C for 0.5, 1.0, 1.5, 2.0, 3.0, and 5.0 h. During sealing, extreme
care was taken in order not to increase the temperature of the samples but to
keep them at room temperature. After the annealing experiments, the change in
morphology, crystal size, particle size, and the melting temperatures of the sam-
ples were determined using optical microscopy, XRD, scanning electron microscopy
(SEM), transmission electron microscopy (TEM), and DSC techniques. The powder
patterns of the samples were recorded on a Bruker D8 diffractometer in ꢀ/2ꢀ geome-
try (reflection setting) equipped with a one-dimensional silicon strip detector using
Cu K␣1 radiation. The unit cell parameters were refined using the program TOPAS
V3 [23]. High purity silicon was used as internal standard for the lattice parameter
determination.
Scanning electron microscopy (SEM) equipped with FIB (focused ion beam) was
used to observe the morphologies of the synthesized nanoparticles (Tescan Lyra 3
FEG FIB, Tescan, Brno, Czech Republic), applying an accelerating potential of 15 kV.
SEM specimens were prepared by a two-stage replica method. The Ni–Sb–Sn nanoal-
loys were dispersed in collodion on a slide. After drying, a carbon film supported by
a copper grid (150 meshes) was applied onto the sample. Then the sample with
the copper grid was carburized and the collodion was dissolved in an amyl acetate
solution.
DSC measurements were carried out on a DSC 404S/3 (Netzsch, Selb, Germany)
in alumina crucibles under the flow of Ar gas. The apparatus was evacuated and
flushed with purified Ar several times before heating. About 50 mg of the powder
samples was pressed into a pellet with 5 mm diameter at a pressure of 10 bars.
Two heating- and cooling curves were recorded for each sample using a heating
rate of 10 K min−1. The temperature calibration of the DSC instrument was done at
the melting points of high purity In, Sn, Ag and Au under identical experimental
conditions.
Measurements of the heat content (H573–H299) of the samples were carried out
in a Calvet-type micro-calorimeter (SETARAM, Lyon, France). The heat flow signal in
the calorimeter is measured by the help of a thermopile with more than 200 ther-
mocouples. The wire wound resistance furnace is suitable for measurements up to
1000 ◦C, and the calorimeter is equipped with an automatic sample drop device for
30 drops. Data acquisition, control, and data evaluation were performed using the
computer programs LabView and HiQ as described by Flandorfer et al. [24]. The
samples, in the form of pellets, were maintained at 299 0.5 K in the sample holder
and were then dropped into the sample cell equilibrated at the experimental tem-
perature of 573 K. During the experiments a flowing Ar atmosphere (approximately
60 cm3 min−1) was maintained to prevent oxidation. At the end of each series the
calorimeter was calibrated by five additions (approximately 40 mg each) of standard
Al2O3 (NIST SRM-720) supplied by the National Institute of Standards and Technol-
ogy (NIST), Gaithersburg, MD, USA. The temperature of the isothermal block (DT)
and the sample temperature in the furnace T were measured using a Pt–Pt 10%
Rh thermocouple and thermoresistors for each drop. The accuracy is usually better
than 1 ◦C. The heat flow signal for each experiment was recorded for 30 min before
dropping a new sample.
Fig. 1. XRD patterns of four nano-sized samples containing 80, 90, 95, and 100 at%
Sn; samples with less than 100 at% Sn contain besides Sn also traces of Ni3Sn4 and
of the ␤-phase (SbSn).
to (Sn). The intensities of the XRD peaks due to Ni3Sn4 and the ␤
phase were found to increase with a decrease in Sn content indi-
cating the formation of higher amounts of these phases at lower
Sn contents. Fig. 1 shows the XRD patterns of samples with 80, 90,
95, and 100 at% Sn and a molar ratio 1:1 of Ni to Sb, demonstrating
the evolution of Ni3Sn4 and ␤-phase with decreasing Sn content.
Table 1 gives cell parameters and the relative average crystallite
size of the different co-existing phases derived from the refine-
ment of the XRD data under similar refinement conditions. It can be
observed that the crystallite size of the major phase (Sn) decreases
with a decrease in the Sn content of the sample.
3. Results and discussion
As an example, SEM images of Ni–Sb–Sn nanoalloys with four
different Sn contents are shown in Fig. 2a–d. The particles have
approximately spherical shape, and their average sizes are found to
be 45, 60, 85, and 195 nm, respectively, for Sn contents of 80, 87.5,
92.5, and 100 at% Sn. Considering all error limits, the agreement
with the sizes determined from XRD in Table 1 is good, except for
the one sample with pure Sn.
Fig. 3 shows the DSC melting peaks for the sample with 95 at%
Sn recorded at a heating rate of 10 K min−1. This relatively fast heat-
ing rate was chosen in order to avoid sintering of the alloy as far
as possible. In the first heating segment, a clear endothermic peak
appeared with an onset at 221 ◦C which was related to the first
appearance of a liquid phase. On repeating the thermal cycle, an
increase of 11 ◦C was observed for the onset temperature of this
peak. (Note that the onset temperature was taken for the appear-
ance of the first liquid for better comparison although the melting
behavior of alloys in the Sn-rich corner is rather complicated [25].)
3.1. Characterization by XRD and DSC
The XRD analyses of the Ni–Sb–Sn nanoalloy samples show that
the major phase for all compositions is tetragonal (Sn) with space
group I41/amd (see Table 1). However, for samples with Sn con-
tents of less than 95 at%, two additional co-existing phases were
found, i.e. monoclinic Ni3Sn4 (space group C2/m), and the trigo-
nal ␤ phase (∼SbSn, space group R-3m) which is in agreement with
our recent experimental phase diagram investigation of the Sn-rich
corner [25]. The appearance of the ␤ phase in a sample with an Sb
content of 3.75 at% might indicate that the solubility of Sb in (Sn)
has decreased from about 6 at% at 200 ◦C [25] to less than 3.75 at%
at 0 ◦C although this sample, prepared at 0 ◦C, is not necessarily in
an equilibrium state. The XRD lines for Ni3Sn4 and the ␤ phase are
found to be broader and more diffuse compared to the lines due

226
R. Mishra et al. / Journal of Alloys and Compounds 513 (2012) 224–229
Table 1
Results of XRD and DSC measurements of Ni–Sn–Sb nanoalloys.
Composition
Phases
Amounta (%) Cell parameters (Å)
Crystallite size (XRD) (nm) Melting temp.
Melting temp.
(onsetb) 2nd
heating (◦C)
Difference between
1st and 2nd
(onsetb) 1st
heating (◦C)
heating (◦C)
Sn100
Sn
100
100
∼97
a = 5.82988(4)
c = 3.18001(2)
a = 5.83117(4)
c = 3.18093(3)
a = 5.83161(7)
c = 3.18121(4)
–
a = 5.8329(4)
c = 3.1820(2)
a = 12.98(1)
b = 3.844(2)
c = 5.326(2)
ˇ = 107.9◦
a = 5.8316(1)
c = 3.1805(9)
a = 13.37(3)
b = 3.818(3)
c = 5.246(3)
ˇ = 108.8◦
a = 5.8328(1)
c = 3.1811(1)
a = 12.174(9)
b = 4.062(5)
c = 5.080(6)
ˇ = 104.7◦
345
150
75
227.5
225
231.5
232
4
Sn97.5Ni1.25Sb1.25 (Sn)
7
(Sn)
221.0
232
11
Sn95Ni2.5Sb2.5
Ni3Sn4
(Sn)
∼3
–
65
95
217.5
217.5
217
232
232
232
14.5
14.5
15
Sn92.5Ni3.75Sb3.75
Ni3Sn4
5
20
(Sn)
80
20
65
15
Sn90Ni5Sb5
Ni3Sn4
(Sn)
65
20
60
Sn87.5Ni6.25Sb6.25
Ni3Sn4
<10
␤-Phase
(Sn)
15
50
30
a = 4.392(3)
c = 5.399(7)
a = 5.833(1)
c = 3.1823(6)
a = 11.86(1)
b = 4.145(4)
c = 5.920(4)
ˇ = 99.7◦
30
85
217.5
216.5
232
232
14.5
15.5
Sn82.5Ni8.75Sb8.85
Ni3Sn4
<10
␤-Phase
(Sn)
20
45
30
a = 4.389(6)
c = 5.41(1)
40
45
a = 5.8322(2)
c = 3.1807(1)
a = 12.217(9)
b = 4.033(2)
c = 5.118(3)
ˇ = 104.7◦
Sn80Ni10Sb10
Ni3Sn4
<10
␤-Phase
25
a = 4.39(3)
40
c = 5.38(9)
a
The amount was derived from the refinement of the XRD data of the samples; the corresponding error is estimated to be about 10%.
Although the melting behavior of Sn-rich ternary Ni–Sb–Sn samples is rather complicated [25] the onset of the first effect on heating was taken in order to allow for a
b
better comparison.
However, no difference in onset temperature could be observed for
the corresponding peak in the cooling cycle.
XRD analyses of individual parts of the alloy Sn95Ni2.5Sb2.5,
annealed at 180 ◦C for periods of 0.5, 1.0, 1.5, 2.0, 3.0, and 5.0 h,
showed that the samples were almost single phase with small
amounts of the Ni3Sn4 phase. The crystallite size obtained from the
refinement of the XRD data of the annealed samples also indicated
that the average crystal size was progressively increasing with the
annealing period. Fig. 4 shows the SEM pictures for this particular
alloy after different periods of annealing. The average particle sizes
for these samples were found to be 55, 70, 95, 105, 120 and 135 nm,
respectively, which agrees quite well with the sizes obtained from
the XRD measurements.
The difference in the onset temperature of this peak in the first
and second heating segment can be explained by the fact that, in
first heating, most of the particles are still in a nanocrystalline form
which has a lower melting temperature compared to the bulk sam-
ple. During cooling and reheating, the samples have re-crystallized
as bulk material, and the thermal effect is due to the melting of
bulk samples with normal crystallite sizes. On the other hand, in
the cooling stage samples freeze in both cycles from the same
liquid state, so they exhibit the same (somewhat super cooled)
onset temperature. This phenomenon of a decrease of the melt-
ing temperature of nanoparticles compared to their corresponding
bulk materials, regardless of the nature of the materials, has been
reported in the past by several authors [4–8]. Table 1 gives the abso-
lute temperatures as well as the difference in the onset temperature
of the melting peak between first and the second cycle for samples
with Sn contents between 80 and 100 at%. It was further observed
that, as the compositions of the samples move farther away from
pure Sn, the intensity of the melting peak decreases progressively.
Fig. 5 gives the DSC plots for the corresponding samples. The
onset temperatures of the melting peak for the annealed samples
were found to increase progressively with an increase in parti-
cle size, i.e. 221, 222.5, 223.5, 225.5, 227.5, 228.5 and 230.0 ◦C for
samples annealed at 180 ◦C for periods of 0, 0.5, 1.0, 1.5, 2.0, 3.0,
and 5.0 h, respectively. The average onset temperature for all the
heat treated samples in the second heating cycle was found to be
232 0.5 ◦C. Thus, the depression in the onset temperature of the
melting peak (ꢁT) for Sn95Ni2.5Sb2.5 was 11, 9.5, 8.5, 6.5, 4.5, 3.5,

R. Mishra et al. / Journal of Alloys and Compounds 513 (2012) 224–229
227
Fig. 2. SEM pictures as examples of four nano-sized samples containing 80, 87.5, 92.5, and 100 at% Sn, as prepared and without any further treatment.
and 2 ◦C according to average particle sizes of 50, 55, 70, 95, 105,
120, and 135 nm, respectively. Fig. 6 shows the variation of ꢁT with
particle size. This is in quite good agreement with previous reports
on the reduction of the melting temperature with decreasing par-
ticle size for pure Sn [26] or different lead-free solder alloys, e.g.
Sn–3.0Ag–0.5Cu [27] or Sn–0.4Co–0.7Cu [28].
It is also interesting to compare the observed melting tempera-
ture depression with the equation given by Qi and Wang [15]:
ꢀ
ꢁ
r
Tm = Tmb 1 − 6˛
(1)
D
where Tm and Tmb are the melting temperatures of the nanopar-
ticles and of the bulk material, ˛ is a shape factor, r is the atomic
radius, and D is the diameter of the nanocrystals. Assuming spher-
ical particles (˛ = 1), and taking the melting point (505 K) and the
atomic radius of pure metallic Sn (r = 0.140 nm [29]) – although it is
clear that the samples contain some small amounts of Ni3Sn4 and
that Sn has dissolved some Sb – a curve can be calculated which is
included in Fig. 6. Considering the fact that the particle sizes esti-
mated from SEM micrographs may contain errors in the range of
10–15% and that the melting temperatures are not for pure Sn,
the agreement with the experimental results is still quite good.
DSC curve for Sn95Ni2.5Sb2.5 nano alloy
0.6
0.4
229.3 °C
cooling
0.2
231.9 °C
heating
0.0
3.2. Determination of the excess surface energy of nanoalloys
221 °C
To determine the excess surface energy of nano-crystalline sam-
ples with different particle sizes, individual parts of the sample
Sn95Ni2.5Sb2.5 that had been annealed at 180 ◦C for different peri-
ods of time (see above) were dropped from room temperature
(299 0.5 K) into the calorimeter which was kept at 573 K. A piece
of the sample that had been heated twice up to 773 K during ther-
mal cycling in the DSC was used as the reference bulk sample after
surface cleaning.
-0.2
-0.4
150
175
200
225
250
275
300
Temperature °C
Fig. 3. DSC curves on first and second heating for a nano-crystalline sample with
the composition Sn95Ni2.5Sb2.5
.

228
R. Mishra et al. / Journal of Alloys and Compounds 513 (2012) 224–229
Fig. 4. SEM pictures of a nano-crystalline sample with the composition Sn95Ni2.5Sb2.5 heated at 180 ◦C for 0.5, 1.0, 1.5, 2.0, 3.0, and 5.0 h.
Dropping of a sample into the calorimeter resulted in an
of the different nano-alloys were obtained by integrating the area
under the curve over a time scale between 100 and 1800 s after
the sample drop. Table 2 gives the (H573–H299) values in J g−1 for
the Sn95Ni2.5Sb2.5 alloy with average particle sizes ranging from 50
endothermic peak which was observed in the heat flux versus time
plot. This endothermic effect is due to the difference in enthalpy
content (H573–H299) which is itself composed of the enthalpy
difference of the solid sample between 299 K and the melting
to 135 nm. From this table it can be observed that the (H573–H299)
ꢂ
T
m
temperature Tm
(
ꢁCpsolid · dT), the enthalpy effect due to the
values decrease with a decrease in the particle size of the alloys.
This is attributed to the exothermic effect arising from the con-
version of nano-particles to bulk material which has to become
larger if the particles become smaller. The excess surface enthalpy
stored in the nano-particles is released when the alloy is heated
299
melting process itself, the enthalpy difference of the liquid sample
between the end of the melting process and the calorimeter tem-
perature 573 K, and an additional exothermic heat effect (exꢁH)
due to the conversion of nano-size samples to the bulk samples.
A steady base line in the heat flux versus time curve after 30 min
indicated that the above thermal effects were completed within
this time period. The values of the enthalpy change for dropping
Fig. 5. DSC curves for a nano-crystalline sample with the composition Sn95Ni2.5Sb2.5
Fig. 6. Depression of the onset temperature of the first melting peak versus particle
size for a nano-crystalline sample with the composition Sn95Ni2.5Sb2.5..
heated at 180 ◦C in quartz ampoules for periods of 0.5, 1.0, 1.5, 2.0, 3.0, and 5.0 h.

R. Mishra et al. / Journal of Alloys and Compounds 513 (2012) 224–229
229
Table 2
Depression in the onset temperature of the melting peak (ꢁT) of nano-crystalline Sn95Ni2.5Sb2.5 samples as a function of particle size (obtained from SEM micrographs) and
their excess surface energy.
Annealing time (h)
Average particle
size (nm)
ꢁT (K)
(H573–H299) (J g−1
)
Excess enthalpy (J g−1
)
Bulk
50
55
70
95
105
120
135
–
11
9.5
8.5
6.5
4.5
3.5
2
129,342 1275
110,496 1438
111,310 1249
117,510 658
124,576 1602
127,838 723
128,297 85
–
0
0.5
1
1.5
2
3
−18,846 1921
−18,032 1784
−11,832 1434
−4766 2047
−1504 1465
−1045 1277
−819 1414
5
128,523 613
and undergoes melting. The average excess energy of the differ-
ent nano-crystalline samples was obtained by subtracting their
(H573–H299) values from that of the bulk sample. It was found
to be 18.8 1.9, 18.0 1.8, 11.8 1.4, 4.8 2.0, 1.5 1.5, 1.0 1.2
and 0.8 1.4 kJ g−1 for Sn95Ni2.5Sb2.5 samples with average parti-
cle sizes of 50, 55, 70, 95, 105, 120 and 135 nm, respectively (see
Table 2).
References
[1] H. Gleiter, Prog. Mater. Sci. 33 (1989) 223–315.
[2] R. Dingreville, J. Qu, M. Cherkaoui, J. Mech. Phys. Solids 53 (2005) 1827–1854.
[3] R.S. Ningthoujam, N. Sudhakar, K.P. Rajeev, N.S. Gajbhiye, J. Appl. Phys. 91
(2002) 6051–6056.
[4] J.R. Sambles, Proc. R. Soc. Lond. Ser. A 324 (1971) 339–351.
[5] A.N. Goldstein, C.M. Echer, A.P. Alivisatos, Science 256 (1992) 1425–1427.
[6] H.W. Sheng, K. Lu, E. Ma, Nanostruct. Mater. 10 (1998) 865–873.
[7] K.F. Peters, J.B. Cohen, Y.-W. Chung, Phys. Rev. B 57 (1998) 13430–13438.
[8] C.L. Cleveland, W.D. Luedtke, U. Landman, Phys. Rev. B 60 (1999) 5065–5077.
[9] P. Pawlow, Z. Phys. Chem. 65 (1909) 545–548.
4. Conclusions
[10] Ph. Buffat, J.-P. Borel, Phys. Rev. A 13 (1976) 2287–2298.
[11] J.-P. Borel, Surf. Sci. 106 (1981) 1–9.
[12] K. Dick, T. Dhanasekaran, Z. Zhang, D. Meisel, J. Am. Chem. Soc. 124 (2002)
2312–2317.
[13] K.K. Nanda, S.N. Sahu, S.N. Behera, Phys. Rev. A 66 (2002) 13208/1–13208/8.
[14] Q. Jiang, S. Zhang, M. Zhao, Mater. Chem. Phys. 82 (2003) 225–227.
[15] W.H. Qi, M.P. Wang, Mater. Chem. Phys. 88 (2004) 280–284.
[16] Y.-W. Du, M.-X. Xu, J. Wu, Y.-B. Shi, H.-X. Lu, J. Appl. Phys. 70 (1991) 5903–5905.
[17] D. Zhang, K.J. Klabunde, C.M. Sorensen, G.C. Hadjipanayis, Phys. Rev. B 58 (1998)
14167–14170.
[18] F. Shir, L. Yanik, L.H. Bennett, E. Della Torre, R.D. Shull, J. Appl. Phys. 93 (2003)
8295–8297.
[19] J. Del Castillo, V.D. Rodríguez, A.C. Yanes, J. Mendez-Ramos, M.E. Torres, Nano-
technology 16 (2005) S300–S303.
[20] A.C. Yanes, J. Del Castillo, M. Torres, J. Peraza, V.D. Rodríguez, J. Mendez-Ramos,
Appl. Phys. Lett. 85 (2004) 2343–2345.
[21] http://heraeus-contactmaterials.com/en/products/solderpowder/productp
solderpowder.aspx.
Ni–Sb–Sn nanoalloys with compositions between 80.0 and
97.5 at% Sn were synthesized by reduction of metal chloride solu-
tions with NaBH4 in alkaline medium. In the obtained products,
the presence of Sn and of small amounts of the binary phases
Ni3Sn4 and SbSn (␤-phase) could be established at 0 ◦C. The par-
ticle size of the alloys progressively decreases with a decrease
in Sn content. A relative decrease in melting temperature of up
to 15 ◦C compared to the bulk sample was observed for the dif-
ferent alloys depending on their particle size. Finally, the excess
surface energy of the nanoalloys was determined as a function of
particle size and was found to increase with decreasing particle
size.
[22] R. Yang, J. Huang, W. Zhao, W. Lai, X. Zhang, J. Zheng, X. Li, J. Power Sources 195
(2010) 6811–6816.
[23] TOPAS V3, General Profile and Structure Analysis Software for Powder Diffrac-
tion Data, Bruker AXS, Karlsruhe, Germany, 2005.
Acknowledgements
This paper is a contribution to the European COST Actions
MP 0602 and MP 0903. Financial support by the European Union
through a Marie Curie Fellowship for R.M. (FP7-PEOPLE-IIF-2008,
Project No.234920), by the Ministry of Education, Youth and Sports
of the Czech Republic (Grants No. LD 11024 and MEB 061005), as
well as additional support by the Scientific-Technical Cooperation
Austria-Czech Republic (Project CZ 08/2010) is gratefully acknowl-
edged.
[24] H. Flandorfer, F. Gehringer, E. Hayer, Thermochim. Acta 382 (2002) 77–87.
[25] R. Mishra, A. Zemanova, A. Kroupa, H. Ipser, in preparation.
[26] Ch.-D. Zou, Y.-L. Gao, B. Yang, Q.-J. Zhai, Trans. Nonferrous Met. Soc. China 20
(2010) 248–253.
[27] Ch.D. Zou, Y.L. Gao, B. Yang, Q.J. Zhai, X.Zh. Xia, Q.J. Zhai, C. Andersson, J. Liu, J.
Electron. Mater. 38 (2009) 351–355.
[28] Ch.D. Zou, Y.L. Gao, B. Yang, Q.J. Zhai, Q.J. Zhai, C. Andersson, J. Liu, Sold. Surf.
Mount Technol. 21 (2009) 9–13.
[29] C.E. Housecroft, A.G. Sharpe, Inorganic Chemistry, 3rd ed., Pearson Education
Ltd., Harlow, UK, 2008, p. 381.