Synthesis and melting behaviour of Bi, Sn and Sn-Bi nanostructured alloy

Article abstract of DOI:10.1016/j.jallcom.2014.08.122

Lead-free solders based on Bi-Sn bimetallic nanoclusters with eutectic composition (Bi₄₃Sn₅₇) were synthesized at low temperature by simultaneous reduction reaction from aqueous solution containing bismuth and tin chlorides, using potassium borohydride as a reducing agent. By the same processing route, pure bismuth and tin nanoparticles have also been prepared. Microstructure, morphology and composition of the samples were characterized by X-ray powder diffraction (XRD), transmission (TEM) and scanning electron microscopy (SEM). TEM images of Bi-Sn nanoparticles show average size ranging from 30 to 100 nm. Thermal behaviour of Bi-Sn nanopowders was studied by DSC (differential scanning calorimetry) and a melting temperature (135 °C) lower than that of the corresponding microcrystalline sample (139 °C) was observed. SEM micrographs of the thermally treated sample up to 400 °C show fine spherical grains in the micrometer range with finer powder particles on the surface. XRD powder diffraction analysis indicates the formation of bismuth and tin nanophases with an average particle size of 85 and 126 nm, respectively. The oxidation behaviour of the samples was also investigated. The results obtained have been analyzed in view of theoretical models describing the melting temperature depression of nanoparticles.

Full text of DOI:10.1016/j.jallcom.2014.08.122

Journal of Alloys and Compounds 623 (2015) 7–14
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Synthesis and melting behaviour of Bi, Sn and Sn–Bi nanostructured
alloy
F. Frongia a, M. Pilloni a, A. Scano a, A. Ardu a, C. Cannas a, A. Musinu a, G. Borzone b, S. Delsante b,
R. Novakovic c, G. Ennas a,
⇑
a Università di Cagliari, Dipartimento di Scienze Chimiche e Geologiche and Cagliari Research Unit of the National Consortium of Materials Science and Technology (INSTM),
Cittadella Universitaria di Monserrato, 09042 Monserrato, CA, Italy
b Department of Chemistry and Industrial Chemistry, Genoa University and Genoa Research Unit of the National Consortium of Materials Science and Technology (INSTM), Via
Dodecaneso 31, I-16146 Genoa, Italy
c National Research Council (CNR), Institute for Energetics and Interphases (IENI), Via De Marini 6, 16149 Genoa, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 July 2014
Received in revised form 11 August 2014
Accepted 12 August 2014
Available online 16 September 2014
Lead-free solders based on Bi–Sn bimetallic nanoclusters with eutectic composition (Bi43Sn57) were syn-
thesized at low temperature by simultaneous reduction reaction from aqueous solution containing bis-
muth and tin chlorides, using potassium borohydride as a reducing agent. By the same processing
route, pure bismuth and tin nanoparticles have also been prepared. Microstructure, morphology and
composition of the samples were characterized by X-ray powder diffraction (XRD), transmission (TEM)
and scanning electron microscopy (SEM). TEM images of Bi–Sn nanoparticles show average size ranging
from 30 to 100 nm. Thermal behaviour of Bi–Sn nanopowders was studied by DSC (differential scanning
calorimetry) and a melting temperature (135 °C) lower than that of the corresponding microcrystalline
sample (139 °C) was observed. SEM micrographs of the thermally treated sample up to 400 °C show fine
spherical grains in the micrometer range with finer powder particles on the surface. XRD powder diffrac-
tion analysis indicates the formation of bismuth and tin nanophases with an average particle size of 85
and 126 nm, respectively. The oxidation behaviour of the samples was also investigated.
The results obtained have been analyzed in view of theoretical models describing the melting temper-
ature depression of nanoparticles.
Keywords:
Bi–Sn alloy
Nanoparticles
One-pot synthesis
Borohydride reducing agent
Melting temperature depression
Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction
In, Bi, Zn, Ag, Sb, Cu, and Mg have been proposed as Pb-free candi-
date materials [2,3]. Among them, the eutectic Bi43Sn57 (at.%) alloy
The eutectic Pb–73.9Sn (at.%) alloy has been extensively used in
the microelectronic industry due to the unique combination of
properties such as low melting point, good wettability on different
substrates and high strength of joints, together with the low cost of
this material. Despite all of these advantages, because of the inher-
ent toxicity of lead, environmental regulations around the world
have aimed to eliminate the usage of Pb-bearing solders in elec-
tronic assemblies. This has prompted the development of Pb-free
solders, and has enhanced the research activities in this field.
Although many of these alloys possess higher strength than the
traditional Pb–Sn ones, the solderability of many Pb-free alloys is
still lower than that of the Pb–Sn eutectic and any improvement
will be welcomed by industry [1]. In the last twenty years a large
number of Sn-based alloys containing minor constituents such as
is one of the promising alternatives [4]. A well-defined Bi–Sn
eutectic microstructure can be obtained at different cooling rates
and in contact with Cu substrate it exhibits similar wetting behav-
iour to that observed in the Pb–Sn/Cu system. Indeed, for both sys-
tems at the interface between the eutectic alloys and Cu substrate,
the formation of two intermetallic phases, i.e.
Sn6, takes place [5,6].
e-Cu3Sn and g-Cu5-
To design the sought-after microstructure and properties, it is
possible to act both on composition and particle size. In recent
years a growing interest in nanomaterials has arisen because of
their peculiar properties often quite different from those of the
same bulk material. The particular properties of nano-sized metals
and alloys can provide solutions to engineering problems that can-
not be solved with conventional alloyed metals. Reviews on the
general structure, properties and potential applications of nanoal-
loys have been published [7–11]. Due to their widespread applica-
tions, a large variety of physical (sputtering, thermal evaporation,
⇑
Corresponding author. Tel.: +39 0706754364; fax: +39 070 6754387.
E-mail address: ennas@unica.it (G. Ennas).
http://dx.doi.org/10.1016/j.jallcom.2014.08.122
0925-8388/Ó 2014 Elsevier B.V. All rights reserved.

8
F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
value changes from 11 (value of the KBH4 solution) to 8.5 at the end of reaction.
Bidistilled water purged with nitrogen (ALPHAGAZ, 99.9992%) was used. The ratio
of the reagents was selected on the basis of the nominal Bi43Sn57 eutectic composi-
tion. In the first preparation (CR_L) the quantity of KBH4 was in excess respect to the
stoichiometric quantity ([KBH4] = 0.9 M), while in the second preparation (CR_H)
KBH4 solution was added in a larger excess ([KBH4] = 3 M).
In order to verify the results obtained with bimetallic preparations, single Sn
(CR_Sn) and Bi (CR_Bi) metallic nanoparticles were also prepared using 3 M potas-
sium borohydride solution.
The reaction was carried out under a nitrogen atmosphere in a three-neck
round-bottom flask equipped with a dropping funnel; the borohydride solution
was vigorously stirred and maintained approximately at 0 °C with an ice bath.
The resulting fine black precipitate was filtered under nitrogen and washed first
with distilled water and then with acetone in order to remove the unreacted
reagents. The recovered powder was dried under flowing nitrogen for 12 h. All steps
of the synthesis (reaction, washing and drying) were carried out in nitrogen atmo-
sphere using a controlled flux in order to repeat the synthesis in the same condi-
laser ablation, mechanical attrition, plasma deposition, etc.) and
chemical (precipitation from solutions, electrodeposition, chemical
vapour deposition, sol–gel, spray pyrolysis, etc.) preparation meth-
ods have been developed. Among them the precipitation from
aqueous or non-aqueous solutions continues to be thoroughly
investigated [12–16], and in particular the chemical reduction
technique has been found to be the most suitable for industrial
manufacture. In fact, it is a comparatively inexpensive and easy
way to produce nanoparticles.
In the present work Sn, Bi and Sn/Bi metallic nanoparticles were
synthesized by a low temperature chemical reduction method in
aqueous solution adopting potassium borohydride as reducing
agent [7,8]. Microstructure, morphology and composition of the
samples were characterized by X-ray powder Diffraction (XRD),
Transmission (TEM, HRTEM) and Scanning Electron Microscopy
(SEM). XRD was used to determine the crystalline nature of the
samples, the grain size and the average size distribution, which
were compared with those obtained by TEM. The thermal behav-
iour of the samples has been investigated by Differential Scanning
Calorimetry (DSC) and by Thermogravimetry Analysis (TGA). After
heat treatment the surface morphology of the samples were stud-
ied by SEM.
tions. In order to prevent
a possible pyrophoric reaction, the powders were
further treated under a slow flow of nitrogen containing about 10 ppm of oxygen.
The Bi–Sn eutectic bulk alloy was also prepared to be used as a reference sam-
ple. The alloy was obtained by repeated fusion and casting of mixed tin and bismuth
powders (Bi 99.99 wt%, Sn 99.99 wt% Aldrich) into a graphite crucible under an
argon atmosphere at 350 °C. Grey metallic flakes showing the characteristic XRD
pattern of microcrystalline Bi–Sn alloy and having a melting point of 139 °C were
obtained.
2.2. Characterization
A theoretical approach to the examination of the phase stability
of nanoparticles was adopted for the first time by Pawlow in 1909
[17]. He postulated that the melting point depression occurs due to
a strong influence of the surface energy on nanoparticles proper-
ties. In particular, the large surface/volume ratio in nanosized sys-
tems has significant effects on their thermodynamic properties and
phase relations. It took almost fifty years to experimentally verify
Pawlow’s theoretical approach [18]. The first experimental verifi-
cation of the melting temperature depression was obtained for
small particles (<50 nm) by electron diffraction on Sn, Pb and Bi
thin films [19]. Subsequently Buffat and Borel [20] calculated the
size of Au particles using XRD analysis and determined their melt-
ing temperature by observing the disappearance of electron dif-
fraction patterns. Coombes [21] used a transmission electron
microscope and determined a melting temperature depression of
about 200 °C for particles of Pb with radius of 3 nm. A home-made,
highly sensitive nano-calorimeter employed by Zhang et al. [22]
showed that a significant melting point depression of 0.1–10 nm
thick discontinuous indium films occurs when the size of the nano-
structures approaches the sub-20 nm range. Various empirical
models have been developed to estimate the melting point depres-
sion of pure metals and alloys as a function of particle size [17,20–
25]. The aforementioned models are relatively simple non-linear
equations that involve the quantities such as the thermodynamic
functions on mixing or formation, bulk melting temperature, pres-
sure, particle size, molar volume or density, interfacial tensions
between the solid–liquid, liquid–vapour and solid–vapour and
some fitting parameters. These parameters are calculated from
the available experimental data providing better agreement
between theoretical and experimental curves of investigated sys-
tems. The experimental data obtained in the present study
together with the corresponding literature data were analyzed by
using the model reported by Wronski [23].
XRD data were collected on a Seifert X3000 powder diffractometer using Cu K
a
radiation and a graphite monochromator on the diffracted beam. XRD patterns were
recorded in step scan mode typically in the range 10° 6 2h 6 85° (and
10° 6 2h 6 100° only for CR_H sample) with step size 2h = 0.05°, collecting at least
1000 counts for each step. The divergence and receiving slits were chosen in order
to ensure a high resolution mode for the crystalline phases. Qualitative analysis of
XRD spectra was determined using the PDF database [26], and average crystallite
sizes of the samples were calculated using the Scherrer equation. The calibration
was performed by means of a standard silicon sample and using the Warren correc-
tion. Quantitative phase analysis of selected samples were also evaluated by the
Rietveld method using the MAUD software [27] and recommended fitting proce-
dures were adopted [28]. Structural models of the identified phases were obtained
by inorganic crystal structure database [29]. Lattice parameters, average crystalline
size, and weight content of each phase were refined. For all fitted samples weighted
pattern agreement index Rwp was less than 0.1.
Thermal behaviour was examined by differential scanning calorimeter using a
Perkin–Elmer DSC7 apparatus with a heating rate of 10 °C/min (the powders were
closed in aluminium pans under pure Argon) and by thermogravimetry using a bal-
ance TGA-SDTA 851 Mettler-Toledo with platinum sample pan, under a mix of
argon and oxygen (50:50 vol%). The calibration of the apparatuses was carried
out with In, Sn and Bi–Sn eutectic bulk alloy. The nanoparticles were characterized
using a TEM (JEOL 200CX), operating at 200 kV. High-resolution TEM images were
obtained with a JEM 2010 UHR equipped with a Gatan Imaging Filter (GIF) with a
15 eV window and a slow scan CCD camera. For the TEM observations, samples
were dispersed in octane, submitted to an ultrasonic bath and the suspension
was then dropped on carbon coated copper grids.
After DSC measurements the surface morphology of the nanopowder samples
was investigated by scanning electron microscopy using a Zeiss EVO 40 SEM (Carl
Zeiss SMT Ltd., Cambridge) operating at 20 kV and equipped with electron probe
microanalysis (EPMA, INCA 300). Both secondary electron (SE) and back-scattered
electron (BSE) signals were employed.
3. Results and discussion
On the basis of the above mentioned chemical reduction pro-
cess, Bi–Sn, Bi and Sn nanoparticles were prepared, see Table 1. A
summary of the DSC, TGA and XRD results is inserted in Table 2.
X-ray diffraction spectra of CR_H and CR_L as prepared samples
are shown in Fig. 1a and b, respectively. The pattern of the Bi–Sn
CR_H sample obtained with large excess of the reducing agent,
exhibits the characteristic features of a mixture of Sn and Bi metal-
lic nanoparticles, while the pattern of CR_L sample corresponds to
that of metallic Bi and neither trace of metallic crystalline Sn nor of
boron-containing crystalline phases were visible. This is in agree-
ment with [30,31], who observed that in a strongly basic environ-
ment due to borohydride decomposition, the formation of boron-
containing phases is not obtained and the boron content is
negligible.
2. Experimental procedures
2.1. Synthesis
The list of the prepared samples together with experimental details is given in
Table 1.
The processing route used to prepare the samples is a well known wet chemical
method [13,14]. In a typical preparation procedure, 100 ml of 0.25 M of metal
chlorides (BiCl3 Fluka 97%; SnCl2 2H2O Fluka 97%) in aqueous HCl solution (37%,
Riedel-de Haen, pure reagent ACS, ISO), was added dropwise to 250 ml of aqueous
potassium borohydride (KBH4 Aldrich 98%) solution. During the synthesis the pH

F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
9
Table 1
In CR_Bi sample XRD pattern of the as prepared sample shows
the presence of metallic bismuth nanoparticles with an average
coherent domain of 75 nm (obtained by Rietveld refinement).
Microcrystalline metallic flakes were obtained for the CR_Sn
sample having an XRD pattern exhibiting sharp reflections in the
positions of the strongly textured tin pattern. Therefore, the reduc-
tion of mixed components results more efficient for the production
of nanosized particles than that of the single salts, especially for
Sn(II). In both cases no peaks related to boron-containing phases
were observed.
Transmission electron microscopy and high-resolution electron
microscopy studies of the samples indicate that they consist of
agglomerated nanoparticles whose mean size and size distribution
depend on the experimental conditions. However, in the CR_H
sample, both single nanoparticles (diameters ranging from 30 to
100 nm, Fig. 2a and b) with spheroidal shape and rare agglomer-
ates (Fig. 2c and d) are present. Fig. 2a images reveal core–shell
and Janus type structures of nanoparticles confirmed by High Res-
olution TEM analyses.
Bi, Sn and Bi–Sn eutectic synthesized samples.
Sample Metals and
composition
(at.%)
BHꢀ
metal
ions
/
Detected phases
by XRD after
synthesis
Tm by DSC Notes
analysis
(°C)
4
ratio
CR_Bi
CR_Sn
CR_L
Pure Bi
Pure Sn
Bi43Sn57
30:1
30:1
9:1
Bi
Sn
Bi
265
232
267
See
Fig. 4a
See
CR_H
Bi43Sn57
30:1
Bi and Sn
135
Fig. 4b
Table 2
Summary of the XRD analysis carried out on CR_H, CR_L, CR_Bi and CR_Sn samples
after different heat treatments in DSC under Ar atmosphere.
Sample DSC heat
treatments
XRD phases
Notes
CR_H
160 °C
240 °C
300 °C
350 °C
450 °C
520 °Ca
800 °Ca
Sn and Bi
See
Fig. 6a
See
Fig. 6b
See
Fig. 6c
See
Fig. 6d
See
Fig. 6e
See
Fig. 6f
See
Fig. 6g
See Fig. 8
–
The HR TEM images of the CR_H sample (see Fig. 3a–c) clearly
show that the surface of the sample is covered with a continuous,
amorphous in appearance, dense and uniform layer having a thick-
ness of few nanometers. In Fig. 3c a complex composite containing
a mixture of nanocrystalline Sn and Bi coated with a SnO shell is
presented; Fig. 3a shows that composite nanoparticles are also
present with a Bi (hR6, As-type) core (012 planes) coated with a
SnO shell whereas in Fig. 3b, nanocrystals are observed with the
core showing lattice fringes (0.1292 nm) matching those of b-Sn
(tI4, Sn-type) (planes 411). These results are in agreement with
the presence of Janus or quasi-Janus nanoparticles, in which the
regions occupied by bismuth are covered by a thin shell of the
other element, similarly to what has recently been found in several
nanoalloys [33].
Bi, Sn, and SnO
Bi, Sn (almost disappeared) and
SnO
Bi and SnO
Bi, SnO2 and Sn
Bi, Sn, SnO2 and Bi64SnO98
Bi2Sn2O7 and SnO2
CR_L
CR_Bi
CR_Sn
450 °C
450 °C
450 °C
Bi, Bi2O3 and Bi64SnO98
Bi2O3, Bi2O2CO3
SnO2 and trace of Sn
–
DSC traces of the CR_L and CR_H samples are shown in Fig. 4a
and b, respectively and the thermal effects observed are summa-
rized in Table 1. Thermogram of the CR_H sample (Fig. 4b) exhibits
a sharp endothermic peak at 135 °C, due to the melting of the sam-
ple; this temperature is 4 °C lower than the melting point of Bi–Sn
eutectic microcrystalline alloy, as determined in our reference bulk
alloy [34]. This peak is followed by a weak broad exothermic effect
in the 200–320 °C temperature range, which can be related to the
oxidation of Bi–Sn nanoparticles [4,35].
a
Sample after TGA measurements performed in O2/Ar (50:50 vol.%) mixture
(heating rate 10°/min up to 800 °C).
Rietveld refinement of CR_L sample confirms the presence of
bismuth as unique phase with an average crystallite size of
80 nm, without microstrain. The calculated lattice parameters
(a = 0.4546, c = 0.1187 nm) are in agreement with the crystallo-
graphic literature data (a = 0.4547, c = 0.1186 nm) [29].
DSC of the CR_Sn sample exhibits a sharp endothermic peak at
232 °C corresponding to the melting temperature of microcrystal-
line Sn. For this sample a melting temperature depression was
not observed due to the relatively large size and low uniformity
of the particles [25,36,37].
On the contrary, for the CR_Bi sample, a large endothermic peak
at 265 °C (versus 271 °C, melting point of pure microcrystalline Bi)
was observed.
Rietveld analysis of CR_H sample shows the presence of Bi (hR6,
As-type) and b-Sn (tI4, Sn-type) with an average coherent domain
of 85 and 126 nm, respectively, without microstrain, indicating the
nanostructured nature of the metals. The lattice parameters obtained
(Bi a = 0.4544, c = 0.1186 nm; Sn a = 0.5843, c = 0.3185 nm) are in
agreement with the crystallographic data [29]. The composition Bi43-
Sn56 obtained by Rietveld’s refinement is in agreement with the nom-
inal composition.
An endothermic peak centred around 267 °C corresponding to a
depression of the Bi melting point was observed for the CR_L sam-
ple (see Fig. 4a), in agreement with the XRD results indicating the
presence of Bi nanosized particles.
After DSC measurements performed on the CR_H sample up to
400 °C in sealed tantalum crucible under argon flow (see for details
[38]), microstructural observations by SEM were carried out.
Fig. 5a shows the SE (Secondary Electron signal) image of the sam-
ple in which Bi–Sn nanoparticles form shiny spherical agglomer-
ates, in micrometric scale, with smaller nanopowders on the
surface. At higher magnification the typical eutectic microstructure
consisting of Sn-rich (grey) and Bi-rich (bright) phases is clearly
visible, see Fig. 5b.
These results indicate that a large excess of KBH4 is necessary to
reduce both metals. This can be due to the negative reduction stan-
dard potential of tin. Although the reaction mechanism is complex
and still unclear, it is possible to summarize the stoichiometry of
the reduction reactions by the following equations:
3BHꢀ4 þ 8Bi3þ þ 9H2O ¢ 3H3BO3 þ 8Bi þ 21Hþ
ð1aÞ
BHꢀ4 þ 4Sn2þ þ 3H2O ¢ H3BO3 þ 4Sn þ 7Hþ
ð1bÞ
Accordingly, 1 mol of potassium borohydride can supply 8 mol
of electrons [30,32]. On the basis of Eqs. (1a) and (1b), in the prep-
aration of CR_L sample, BH4ꢀ/metal ions ratio adopted gives a
slight excess in comparison to the stoichiometric ratio, while in
the preparation of CR_H sample there is a very large excess of
KBH4 (30:1).
Additional five different DSC measurements were carried out in
open crucible under argon flow on the same batch of the CR_H
sample; each one was heated up at five different temperatures in

10
F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
Fig. 1. XRD spectra of (a) CR_H and (b) CR_L Bi–Sn samples.
Fig. 2. TEM images of Bi–Sn CR_H two-phase sample: (a) bright field of single nanoparticles, (b) dark field of single nanoparticles, (c) bright field of agglomerated
nanoparticles, (d) dark field of agglomerated nanoparticles.

F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
11
(a)
(b)
100
150
200
250
300
Temperature, °C
Fig. 4. DSC thermograms of Bi–Sn samples (a) CR_L, only the Bi nanoparticles are
present (b) CR_H, Bi–Sn nanoparticles.
spectra with respect to those of the as prepared one. On the con-
trary, in the XRD spectra of the sample heated up to 240 °C weak
peaks of crystalline SnO (romarchite) appeared [26], as shown in
Fig. 6b. In the range 240–350 °C, an increase in the oxidation reac-
tion was observed: the peaks corresponding to the Sn tetragonal
phase progressively disappeared, whereas the intensity of the
peaks of SnO increased (Fig. 6c); at 350 °C, XRD patterns showed
the presence of only Bi and SnO (see Fig. 6d). On further heating
of the samples up to 450 °C, SnO peaks disappeared giving rise to
the formation of new peaks corresponding to SnO2 (cassiterite)
together with the peaks of Sn (see Fig. 6e). This can be explained
by the higher stability of SnO2 at this temperature.
TGA experiments have been carried out up to 800 °C on the
CR_H powders under an O2/Ar atmosphere (50:50 vol.%) with a
heating rate of 10 °C/min. TGA trace and its derivative (Fig. 7) show
a first step centered at 200 °C, and a second wide step between 300
and 500 °C. The first step can be associated to the oxidation of Sn
giving SnO, in agreement with XRD patterns observed and
described above (see Figs. 6b–e). The second wider step indicates
two different phenomena: in fact dTGA curve shows a sharp peak
around 450 °C with a shoulder on the left. The last one is due to
the transformation of SnO to SnO2 and the contemporary reappear-
ance of metallic tin peak, as observed by XRD, see Fig. 6e. Indeed, it
is generally accepted that SnO, which is formed at the initial stage
of Sn oxidation, is metastable and decomposes, even in the absence
of oxygen, in the wide 250–600 °C temperature range, according to
the following disproportionation reaction, i.e. 2SnO ? SnO2 + Sn
[39]. This reaction depends on many factors, such as initial oxygen
concentration, temperature and humidity. This disproportionation
reaction explains the reappearing of metallic tin in the XRD spectra
of CR_H sample after 450 °C.
The XRD pattern of the CR_H sample heated at 520 °C in TGA
shown in Fig. 6f, highlights the presence of SnO2 and an additional
Bi64SnO98 phase. Indeed, at this temperature bismuth metallic
peaks begin to disappear, and at 800 °C (Fig. 6g) neither tin nor bis-
muth peaks are visible, but a mixture of oxides Bi2Sn2O7 and SnO2
can be detected.
A DSC heat treatment at 450 °C was also performed on CR_Bi,
CR_Sn and CR_L samples and then they have been analyzed by
XRD; these results are also inserted in Table 2. It should be noticed
the formation of Bi2O3 for the CR_Bi and CR_L samples; besides, the
Fig. 3. HRTEM of Bi–Sn CR_H sample: (a) Bi (hR6, As-type) nanoparticle (012 lattice
fringes) coated by an amorphous SnO layer, b) b-Sn (tI4, Sn-type) nanoparticle (411
lattice fringes) coated by an amorphous SnO layer, (c) Bi, Sn nanoparticles coated by
an amorphous SnO layer.
the range 160–450 °C and subsequently cooled to room tempera-
ture. XRD patterns of these samples are shown in Fig. 6(a–e) and
the detected phases are summarized in Table 2. The samples trea-
ted up to 160 °C (Fig. 6a) do not exhibit any difference in XRD

12
F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
Fig. 5. SEM images of Bi–Sn CR_H sample after heat treatment at 400 °C: (a) Secondary electron image, (b) back scattered image.
120
"
"
+
+
(g)
(f)
+
+
+
"
"
115
110
105
100
"
"
"
"
*
+
*
*
*
*
+
^
+
+
*
°
*
+
*
*
*
^
+
*
*
°
°
°
*
*
*
(e)
°
°
+
x
*
*
*
*
** +
+
*
+
*
*
*
*
+
*
°
*
*
°
*
x
x
*
*
(d)
x
x
x
*
x
x
x
x
x
0
100
200
300
400
500
600
700
800
* *
*
(c)
°
x
*
*
Temperature °C
*
x
°
x
x
*
*
°x
*
x
°
°
*
°
*
*
*
°
Fig. 7. TGA curve (continuous line) performed under O2/Ar mixture flow for the
CR_H sample and its derivative (dotted line).
*
(b)
°
°
*
°
°
*
*
*
*
x
x
°
x
x
°
x
*
*
# Bi2O3
Bi
*
*
^
Bi64SnO98
*
° °
*
(a)
°
*
° °
°
*
*
*
*
*
*
°°
*
°
*
*
20
30
40
50
60
70
2ϑ
Fig. 6. XRD spectra of Bi–Sn CR_H sample heat processed in DSC with argon at: (a)
160 °C, (b) 240 °C, (c) 300 °C, (d) 350 °C, (e) 450 °C. XRD spectra of CR_H heat
*
#
processed in TGA with argon/oxygen mixture at: (f) 520 °C, (g) 800 °C. Marks: Bi,
*
00
s Sn, x SnO, + SnO2, ^ Bi64SnO98
,
Bi2Sn2O7.
#
#
*
*
^
^
#
*
*
*
#
*
*
# *
^
#
*
*
tetragonal form of the oxycarbonates (tetragonal Bi2O2CO3), [26]
probably formed after air exposure, is also visible in CR_Bi. In
CR_L thermally treated sample peaks corresponding to Bi64SnO98
phase are also observed together with Sn weak peaks probably
due to the presence of Sn trace amounts in amorphous SnO shell
or dissolved in the Bi structure. The XRD pattern of CR_Sn sample
after the heat treatment shows sharp SnO2 peaks and weak peaks
ascribable to the residual metallic Sn.
#
#
#
20
30
40
50
2ϑ
60
70
80
Fig. 8. XRD spectra of the CR_L sample after annealing at 450 °C carried out in DSC
under argon flow.
ꢄ
ꢂ
ꢃꢅ
The melting behaviour of Bi nanoparticles obtained in the pres-
ent work was analyzed by the Wronski model [23]. The melting
temperature depression as a function of particle size is given by
ꢀ
ꢁ
2T0
r1
r2
r
1
1
T0 ꢀ Tr ¼
þ
ꢀ
q1 q2
ð2Þ
H0 q1ðr ꢀ tÞ

F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
13
LR.2007 Regione Autonoma della Sardegna for financial support
(LR.7/2007 Project No. F71J11001220002-year 2010).
References
[1] C.M.L. Wu, D.Q. Yu, C.M.T. Law, L. Wang, Properties of lead-free solder alloys
with rare earth element additions, Mater. Sci. Eng. R 44 (2004) 1–44.
[2] M. Abtew, G. Selvaduray, Lead-free solders in microelectronics, Mater. Sci. Eng.
27 (2000) 95–141.
[3] M.F. Ashby, K. Johnson, Materials and Design: The Art and Science of Materials
Selection in Product Design, Butterworth-Heinemann, Oxford, 2002.
[4] H. Chen, Z. Li, Z. Wu, Z. Zhang, A novel route to prepare and characterize Sn–Bi
nanoparticles, J. Alloys Comp. 394 (1–2) (2005) 282–285.
[5] F. Gnecco, E. Ricci, S. Amore, D. Giuranno, G. Borzone, G. Zanicchi, R. Novakovic,
Wetting behaviour of the lead free solder of Au–In–Sn and Bi–In–Sn alloys on
copper substrate, Int. J. Adhes. Adhes. 27 (2007) 409–416.
[6] B.J. Lee, N.M. Hwang, H.M. Lee, Prediction of interface reaction products
between Cu and various solder alloys by thermodynamic calculation, Acta
Mater. 45 (5) (1997) 1867–1874.
[7] K.E. Gonsalves, S.P. Rangarajan, J. Wang, Chemical synthesis of nanostructured
metals, metal alloys, and semiconductors, in: H. Nalwa (Ed.), Handbook of
Nanostructured Materials and Nanotechnology, vol. 1, Academic Press, San
Diego, 2000, p. 40.
[8] R. Ferrando, J. Jellinek, R.L. Johnston, Nanoalloys: from theory to applications of
alloy clusters and nanoparticles, Chem. Rev. 108 (3) (2008) 846–872.
[9] J.N. Hu, H. Tanaka, T. Narita, Aging behavior of a Sn–Bi eutectic solder at
temperatures between 233 and 373 K, Mater. Trans. JIM 42 (5) (2001) 769–775.
[10] M. McCormack, S. Jin, G.W. Kammlott, Enhanced solder alloy performance by
magnetic dispersions, IEEE Trans. Comp. Packag. Manufact. Technol. Part A 17
(3) (1994) 452–457.
Fig. 9. Size dependence of the melting point of Bi nanoparticles: the curve is
calculated using the Wronski model [23]. The present results ꢀ are compared with
the literature data j [39], d [40], . [41].
where T is the temperature of equilibrium between a solid sphere
c
of radius (r ꢀ t) and a concentric liquid shell of inner radius, given
by a thickness t, T0 is the bulk melting temperature of the solid,
[11] M. Hosokawa, K. Nogi, M. Naito, T. Yokoyama, Nanoparticle Technology
Handbook, Elsevier, Oxford, 2012.
[12] H.S. Nalwa, Handbook of Nanostructured Materials and Nanotechnology,
Academic Press, San Diego, USA, 2000.
H0 is the latent heat of fusion,
r1 and r2 are the interfacial tensions
between the solid and liquid phases and between the liquid and
vapour phases, respectively.
and liquid phases.
q1 and q2 are the densities of the solid
[13] B.L. Cushing, V.L. Kolesnichenko, C.J. O’Connor, Recent advances inthe liquid-phase
syntheses of inorganic nanoparticles, Chem. Rev. 104 (9) (2004) 3893–3946.
[14] A. Corrias, G. Ennas, G. Marongiu, A. Musinu, G. Paschina, Amorphous
transition metal-boron ultrafine particles prepared by chemical methods,
Chem. Mater. 5 (1993) 1722–1726.
[15] D. Peddis, D. Rinaldi, G. Ennas, A. Scano, E. Agostinelli, D. Fiorani,
Superparamagnetic blocking and superspin-glass freezing in ultra small d-
(Fe0.67Mn0.33)OOH particles, Phys. Chem. Chem. Phys. 14 (2012) 3162–3169.
[16] M. Pilloni, J. Nicolas, V. Marsaud, K. Bouchemal, F. Frongia, A. Scano, G. Ennas,
C. Dubernet, PEGylation and preliminary biocompatibility evaluation of
magnetite–silica nanocomposites obtained by high energy ball milling, Int. J.
Pharm. 401 (1–2) (2010) 103–112.
[17] P.Z. Pawlow, Über die Abhängigkeit des Schmelzpunktes von der Oberflächenenergie
eines festen Körpers (Zusatz), Z. Phys. Chem. 65 (1909) 545–548.
[18] T.L. Hill, Thermodynamics of Small Systems, Courier Dover Publications, New
York, 1994.
[19] M. Takagi, Electron-diffraction study of liquid–solid transition of thin metal
films, J. Phys. Soc. Jpn. 9 (1954) 359–363.
Preliminary analysis of the literature data and theoretical mod-
els [17,20,23,25,40–42] indicates that the results obtained by the
models differ slightly up to 50 nm particle size, while the effect
of a small particle size on the melting temperature depression
becomes significant for nanoparticles with particle size <30 nm,
as recently reported by Zou et al. [36].
The effect of the particle size on the melting temperature
depression of Bi nanoparticles obtained in the present work is
shown in Fig. 9 in agreement with literature data. Due to the large
size of synthesized Sn particles the above mentioned calculations
were not performed.
[20] P.A. Buffat, J.P. Borel, Size effect on the melting temperature of gold particles,
Phys. Rev. A 13 (6) (1976) 2287–2298.
4. Conclusions
[21] C.J. Coombes, The melting of small particles of lead and indium, J. Phys. F: Met.
Phys. 2 (3) (1972) 441–449.
[22] M. Zhang, M.Y. Efremov, F. Schiettekatte, E.A. Olson, A.T. Kwan, S.L. Lai, T.
Wisleder, J.E. Greene, L.H. Allen, Size-dependent melting point depression of
The Bi43Sn57 eutectic nanoparticles were successfully synthe-
sized by chemical reduction of inorganic precursors with potas-
sium borohydride in aqueous solution. Its melting point (135 °C)
was 4 °C lower than the melting point of the corresponding bulk
alloy, in agreement with the nanometric size of the sample. The
nanoparticles were almost of spherical shape and their size distri-
bution was not uniform. Bi and Sn synthesized nanoparticles show
an average dimension of 85 and 126 nm, respectively.
The results obtained in the present work by the well-concerted
employed techniques (TGA/DSC/XRD/HRTEM) can lead to several
observations. Comparing the XRD analyses performed after the
heat treatment at 450 °C for the CR_H, CR_L and CR_Bi samples,
the presence of SnO shell on the CR_H nanoparticles seems to pro-
tect bismuth from the oxidation.
nanostructures: nanocalorimetric measurements, Phys. Rev.
10548–10557.
B 62 (2000)
[23] C.R.M. Wronski, The size dependence of the melting point of small particles of
tin, Brit. J. Appl. Phys. 18 (1967) 1731–1737.
[24] P.R. Couchman, W.A. Jesser, Thermodynamic theory of size dependence of
melting temperature in metals, Nature 269 (1977) 481–483.
[25] S.L. Lai, J.Y. Guo, V. Petrova, G. Ramanath, L.H. Allen, Size-dependent melting
properties of small tin particles: nanocalorimetric measurements, Phys. Rev.
Lett. 77 (1996) 99–102.
[26] PCPDF-WIN, JCPDS – International Center for Diffraction Data, Swarthome, PA,
1998.
[27] L. Lutterotti, S. Matthies, H.R. Wenk, MAUD: a friendly Java program for
material analysis using diffraction, IUCr: Newslett. CPD 21 (1999) 14–15.
[28] R.A. Young, The Rietveld Method, IUCr Monographs on Crystallography 5,
Oxford University Press, Oxford, 1995.
[29] Inorganic Crystal Structure Database ICSD, Karlsruhe, Germany, 2012.
[30] G.B. Glavee, K.J. Klabunde, C.M. Sorensen, G.C. Hadjipanayis, Borohydride
reduction of nickel and copper ions in aqueous and nonaqueous media.
Controllable chemistry leading to nanoscale metal and metal boride particles,
Langmuir 10 (12) (1994) 4726–4730.
The melting behaviour of Bi nanoparticles has been analyzed
also by theoretical model and good agreement with experimental
data was observed.
[31] D.G. Kim, H. Kim, H.J. Sohn, T. Kang, Nanosized Sn–Cu–B alloy anode prepared
by chemical reduction for secondary lithium batteries, J. Power Sources 104
(2002) 221–225.
Acknowledgements
[32] J.M. Lu, D.B. Dreisinger, W.C. Cooper, Cobalt precipitation by reduction with
sodium-borohydride, Hydrometallurgy 45 (3) (1997) 305–322.
[33] D. Bochicchio, R. Ferrando, Morphological instability of core–shell metallic
nanoparticles, Phys. Rev. B 87 (2013) 165435. 1–13.
This research is a contribution to the European COST Action
MP0903 ‘‘Nanoalloys as advanced materials: from structure to
properties and applications’’. The authors would like to thank the

14
F. Frongia et al. / Journal of Alloys and Compounds 623 (2015) 7–14
[34] C.T. Schamp, W.A. Jesser, Two-phase equilibrium in individual nanoparticles of
Bi–Sn, Metall. Mater. Trans. A 37 (6) (2006) 1825–1829.
[38] Y. Yuan, S. Delsante, J. Yi, G. Borzone, A revision of the Sm-rich region of the
Sm–Co system, J. Alloys Comp. 508 (2) (2010) 309–314.
[39] P. Song, D. Wen, Experimental investigation of the oxidation of tin
nanoparticles, J. Phys. Chem. C 113 (31) (2009) 13470–13476.
[40] E.A. Olson, M.Y. Efremov, M. Zhang, Z. Zhang, L.H. Allen, Size-dependent
melting of Bi nanoparticles, J. Appl. Phys. 97 (2005) 034304–034312.
[41] S.J. Peppiatt, Melting of small particles. II. Bismuth, Proc. R. Soc. Lond. A-Math.
Phys. Eng. Sci. 345 (1975) 401–412.
ˇ
ˇ
[35] J. Sopoušek, J. Vreštál, A. Zemanova, J. Bursik, Phase diagram prediction and
particle characterisation of Sn–Ag nano alloy for low melting point lead-free
solders, J. Min. Metall. Sect B-Metall. 48 (3) (2012) 419–425.
[36] C.D. Zou, Y.L. Gao, B. Yang, Q.J. Zhai, Size-dependent melting properties of Sn
nanoparticles by chemical reduction synthesis, Trans. Nonferr. Met. Soc. China
20 (2) (2010) 248–253.
[37] Y.H. Jo, I. Jung, C.S. Choi, I. Kim, H.M. Lee, Synthesis and characterization of low
temperature Sn nanoparticles for the fabrication of highly conductive ink,
Nanotechnology 22 (2011) 225701–225708.
[42] G.L. Allen, R.A. Bayles, W.W. Gile, W.A. Jesser, Small particle melting of pure
metals, Thin Solid Films 144 (2) (1986) 297–308.