New Insights into Phase Equilibria of the Sb-Sn System

 

 

J. Phase Equilib. Diffus. (2021) 42:63–76
https://doi.org/10.1007/s11669-020-00849-7

New Insights into Phase Equilibria of the Sb-Sn System

G. Borzone1,2

• S. Delsante1,2

• D. Li3

• R. Novakovic2

Submitted: 29 January 2020 / in revised form: 5 October 2020 / Accepted: 21 October 2020 / Published online: 3 January 2021
(cid:2) ASM International 2021

Abstract A new experimental investigation of the Sb-Sn
phase diagram has been performed. Special attention was
focused on the phase equilibria within SbSn-Sb3Sn4 com-
position range. Phase equilibria have been investigated by
DSC measurements combined with a thorough optical and
electronic microscopy examination with a special attention

Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s11669-020-00849-7) contains sup-
plementary material, which is available to authorized users.

This invited article is part of a special tribute issue of the Journal of
Phase Equilibria and Diffusion dedicated to the memory of Gu¨nter
Effenberg. The special issue was organized by Andrew Watson,
Coventry University, Coventry, United Kingdom; Svitlana Iljenko,
MSI, Materials Science International Services GmbH, Stuttgart,
Germany; and Rainer Schmid-Fetzer, Clausthal University of
Technology, Clausthal-Zellerfield, Germany.

& G. Borzone

borzonegabriella@gmail.com

S. Delsante
simona.delsante@unige.it

D. Li
lidajian@gmail.com

R. Novakovic
rada.novakovic@ge.icmate.cnr.it

1 Department of Chemistry and Industrial Chemistry, Genoa
University and Genoa Research Unit of INSTM, Via
Dodecaneso 31, 16146 Genoa, Italy

2

3

Institute of Condensed Matter Chemistry and Technologies
for Energy, National Research Council (ICMATE-CNR), Via
de Marini 6, 16149 Genoa, Italy

Institute for Applied Materials-Applied Materials Physics
(IAM-AWP), Karlsruhe Institute of Technology (KIT),
Hermann-von-Helmholtz-Platz 1,
76344 Eggenstein-Leopoldshafen, Germany

to the SbSn-Sb3Sn4 phase transformation. Both DSC and
metallographic results have been critically evaluated and
debated. The presence of
reactions
L ? Sb3Sn4 (cid:2) (bSn)
at 243 ± 1 (cid:3)C and L ? (Sb)
(SbSn) at 420 ± 1 (cid:3)C have been confirmed. The invariant
reaction at 321 ± 1 (cid:3)C has been observed and an addi-
tional small effect at about 330-340 (cid:3)C was systematically
recorded. These results have been extensively discussed.

two peritectic

Keywords antimony-tin (cid:2) catatectic reaction (cid:2) differential
scanning calorimetry (cid:2) microstructure (cid:2) Sb-Sn phase
diagram

1 Introduction and Literature Review

Phase diagrams play an important role in the understanding
of many scientific and technological disciplines, and are
important guidelines in the design, processing, production
and application of materials. Application oriented research
of the Sb-Sn system mainly includes Sn-rich binary and
multicomponent alloys as high temperature lead-free sol-
ders[1-9] and intermediate Sb-Sn compositions used for
anode materials in Li-ion batteries.[10-12] In the first case,
the surface and wetting properties of liquid Sb-Sn and Sb-
Sn-based alloys are key properties required for design and
development of both lead-free solder alloys,[2,4,6] as well as
fluxes, that are necessary to prevent the oxidation during
industrial soldering processes. Indeed, the Sb-Sn is a sub-
system present
lead-free solders
in several commercial
such as CASTIN (0.8Cu-0.5Sb-2.5Ag-96.2Sn wt.%, melt-
ing T = 216 (cid:3)C) and J-alloy (Sn-25Ag-10Sb wt.%, melting
T = 228 (cid:3)C).[13,14] Similarly to CASTIN alloy, the present
authors[15] have developed Sn-1.9Ag-0.4Cu-2.1Sb, wt.%

123

64

J. Phase Equilib. Diffus. (2021) 42:63–76

(SAC194Sb) taking into account the SAC187 reference
alloy. Detailed investigation of the effects of Sb on its
surface properties, wetting characteristics and microstruc-
tural evolution of the interfaces formed with Cu and Ni-
substrates together with subsequent mechanical character-
ization of solder joints is reported in Giuranno et al.[15]
Concerning lithium batteries, Sn, Sb, and Si are the most
promising elements
forming high lithium-containing
intermetallic phases.[12] In particular, among Sb-Sn alloy
compositions, it seems that the use of (SbSn) intermediate
phase at the equiatomic composition is beneficial during
lithiation due to its separation into the more lithiated and/or
not
lithiated phases, helping to relieve mechanical
strain.[16] Another way to produce anode materials for high
performance lithium ion batteries is the use of nanostruc-
tured materials,
and
Srivastava.[11]

reviewed

in Roy

recently

Since 1891, many authors have investigated the Sb-Sn
alloy system and its phase diagram, resulting in at least
twenty assessments.[17-22] Controversial versions of the Sb-
Sn phase diagram have been proposed[23-28] and the most
widely accepted was reported in the Massalski compila-
tion,[29] which was based mainly on the experimental work
by Predel and Schwermann.[24] This version shows the
presence of three peritectic reactions (see Table 1) with the
formation of the (SbSn) phase that exhibits a wide homo-
geneity range shifting towards Sb-rich compositions at
higher temperatures, and the Sb2Sn3 compound that was
considered to be stable between 242 and 324 (cid:3)C.[24] An
assessment of the Sb-Sn was prepared by Dreval.[30] Sch-
metterer et al.[31] published a very thorough analysis of the
primary literature about the Sb-Sn system, and this paper
will be considered as the basis for literature data in the
subsequent discussion.

According to most authors, the effect observed around
310-326 (cid:3)C is associated to the peritectic formation of the
Sb2Sn3 phase
there was

(see Table 1). However,

Table 1 Invariant reactions (upon cooling) in the Sb-Sn system

(DTA) measurements

disagreement about the stability range of this compound.
Recently Chen et al.,[32] who performed Differential
Thermal Analysis
in the 8.2-
93.2 at.% Sb composition with a scanning rate of 8 (cid:3)C/
min, ruled out the decomposition of the Sb2Sn3 around
240 (cid:3)C and proposed a new version of the phase diagram
with Sb2Sn3 stable up to room temperature. While our
experiments were ongoing, Schmetterer et al.[33] performed
a new experimental study of the Sb-Sn alloys focused on
their complex crystallography and, also on the basis of the
DTA and Energy-Dispersive x-ray Spectroscopy (EDS)
data, provided a new version of the Sb-Sn phase diagram.
The Authors established the existence of the Sb3Sn4 phase
instead of Sb2Sn3, and suggested the existence of the
Sb3Sn4 phase down to room temperature. Therefore, in the
following, instead of Sb2Sn3 we will refer to the Sb3Sn4
phase.

It is worth noting that very few Differential Scanning
Calorimetry (DSC) curves and micrographs were reported
in the literature. Numerous structural investigations have
been carried out by different authors.[22,25,34,35] Lidin et al.
investigated the SbSn phase by single crystal x-ray
Diffraction (XRD) and EDS analysis; according to their
findings,[34,35] the SbSn has a rhombohedral distorted and
incommensurately modulated NaCl-type crystal structure.
Lidin and Folkers[36] carried out a single crystal syn-
chrotron XRD investigation in order to solve the structure
of the ‘‘ht-Sb2Sn3’’ phase. At the same time, Schmetterer
et al.[33] found that the Sb3Sn4 phase crystallizes in a
rhombohedral structure with pronounced structural simi-
larities to the (SbSn) phase. The authors stated that, using
XRD analysis, both structures share the main reflexes and
the difference between them only manifests itself in the
angular positions of the superstructure reflexes. Recently,
three more papers have been published on the Sb-Sn sys-
tem. Kroupa et al.[37] studied Sb-Sn alloys experimentally
and through theoretical modelling; they prepared three bulk

Reaction

Reaction type

at.% Sb
invariant point

T, (cid:3)C
TW

T, (cid:3)C
[18]

T, (cid:3)C
[21]

T, (cid:3)C
[24]

T, (cid:3)C
[27]

T, (cid:3)C
[28]

T, (cid:3)C
[32]

T, (cid:3)C
[33]

L ? (Sb) (cid:2) (SbSn)
L ? (SbSn) (cid:2) Sb3Sn4

Peritectic

Peritectic

L ? Sb3Sn4 (cid:2) (bSn)
Sb3Sn4 (cid:2) (SbSn) ? (bSn)

Peritectic

Eutectoid

65.2[24]
40.0[24]

10.0

…

420 ± 1

321 ± 1(a)

243 ± 1

…

430

310

243

…

425

325

246

…

425

324

250

242

425

325

245.5

242.5

424

322

246

244

424

323

243

…

425 ± 2

326 ± 3

244 ± 1

…

Present DSC results (TW) compared to literature experimental data
Before[33] the Sb3Sn4 compound was described as Sb2Sn3
In TW the temperatures were determined as the average of the DSC results reported in Table 3

(a) See discussion in section 3.3

123

J. Phase Equilib. Diffus. (2021) 42:63–76

65

alloys, namely at 14.7, 24.7 and 44.4 at.% Sb using con-
ventional method and nanoalloys using a wet synthesis.
After melting, the bulk samples were annealed at 200 (cid:3)C
for 168 h and quenched. Scanning Electron Microscopy
(SEM) and DSC were used as the main tools to analyze
both bulk and nanoalloy samples. The combination of the
CALPHAD (CALculation of PHAse Diagram) method and
ab initio calculations was used to model the effect of the
particle size on the behaviour of complex intermetallic
phases in the system. Concerning bulk alloys, they deter-
mined phase boundaries and invariant reaction tempera-
tures. However, they omitted the recent data reported by
Schmetterer et al.[31,33] and presented a calculated phase
diagram, which included the (SbSn) phase and the Sb2Sn3
compound stable in a narrow temperature range. The cal-
culated phase diagram is therefore very similar to that
reported by Predel and to Kroupa’s own previous ver-
sion.[38] Lysenko,[39] on the basis of a thermodynamic
description of the Sb-Sn system, proposed a modified
version of the Sb-Sn phase diagram that includes both
Sb2Sn3 and Sb3Sn4 phases together with the (SbSn).
However, this version does not seem consistent taking with
the available experimental information. Gierlotka[40] pro-
posed a new thermodynamic model of the binary Sb-Sn
system by using a combination of the CALPHAD method
and ab initio calculation. The proposed phase diagram
includes the existence of
(SbSn) phase and Sb3Sn4
stable up to room temperature instead of Sb2Sn3. Never-
theless, the phase transformation involving the Sb3Sn4 and
(SbSn) phases has not been well established yet, and the
question remains whether Sb3Sn4 and (SbSn) are inde-
pendent phases. Therefore, an experimental re-investiga-
tion of the Sb-Sn phase diagram seems necessary. Hence,
this paper considers what is known about the Sb-Sn system
and is aimed to clarify the reaction mechanism for the
formation of the Sb3Sn4 and (SbSn) phases by applying
DSC combined with
accurate metallographic
an
investigation.

2 Experimental

Around 60 alloy samples in the 2.5-80.5 at.% Sb compo-
sition range were prepared starting from high purity
materials. Pure elements (Sb and Sn 99.999 mass%,
Newmet Koch) were sealed inside quartz tubes and melted
in a resistance furnace under an argon atmosphere followed
by water quenching. The alloy buttons (about 3 g) were
divided into several pieces and employed for further
characterization and measurements. Differential Scanning
Calorimetry (DSC), Light Optical Microscopy (LOM) and
Scanning Electron Microscopy (SEM) coupled with EDS
(Energy-Dispersive x-ray Spectroscopy) analysis were

employed for the consistent investigation of both annealed
and DSC samples. Before DSC measurements, most of the
samples were homogenized at different temperatures for
different durations and either cooled down slowly or
quenched to room temperature.

as

DSC measurements were performed on a DSC 111
SETARAM apparatus,
a Calvet-type
designed
calorimeter, containing cylindrical reference and working
cells surrounded by two differentially connected thermal
fluximeters. The sample and reference Calvet sensors are
inserted in a calorimetric block and are composed of 120
thermocouples mounted in a cylinder that surrounds the
measurement zone, which provides a high sensitivity sen-
sor. Pieces of the samples weighing 300-500 mg were
sealed under argon in quartz tube crucibles. The DSC
instrument was used in a continuous mode and calibrated at
the melting points of high purity In, Sn, Pb and Zn metals
sealed in quartz crucibles; the accuracy of measurement is
considered to be within ± 0.5 (cid:3)C. Usually, samples were
heated up to 200 (cid:3)C at 1 (cid:3)C/min followed by an isotherm
of 60-120 min for thermal equilibration, then a scanning
rate of 0.3(cid:3) C/min was generally applied up to the liquid
temperature. With the aim to clarify the stability of the
Sb3Sn4 phase and the Sb3Sn4-SbSn phase transformation,
many samples were also equilibrated at selected tempera-
tures for different durations.

During the investigation of a phase diagram, character-
istics of the micrography are very useful to unravel the
phase transformation mechanisms. For this reason, all
prepared samples were cold mounted in epoxy resin and
polished using standard metallographic techniques. The
samples were characterized by LOM and SEM/EDS to
examine microstructures, identify the phases, and measure
their composition on polished surfaces. A Leica Digital
Microscope and a Zeiss Evo 40 (Carl Zeiss SMT Ltd,
Cambridge, England) equipped with a Pentafet Link
Energy Dispersive x-ray Spectroscopy system managed by
the INCA Energy software (Oxford Instruments, Analytical
Ltd., Bucks, U.K) were employed. The microscope was
generally operated at an acceleration voltage of 20 kV and
calibration for the quantitative measurements was per-
formed by a Co standard. Resulting compositions were
finally corrected for ZAF (Z = atomic number, A = ab-
sorption, and F = fluorescence) effects using pure elements
as standards. Due to the close atomic number of Sn and Sb
and similar scattering factors, it was difficult to have a
good image contrast using Back Scattered Electron (BSE)
detector. It was therefore considered advisable to make a
careful optical microscopic examination before the SEM
encountered in obtaining
analysis. Difficulties were
scratch-free surfaces of Sn-rich alloys, due to the large
difference in hardness of the different phases. Furthermore,
the (Sn) soft matrix was severely etched by all reagents

123

66

J. Phase Equilib. Diffus. (2021) 42:63–76

tried, except for the nital 1.0 solution, while the inter-
metallic phases showed great chemical resistance to most
reagents. The most suitable etching reagent for alloys
containing the Sb3Sn4 and SbSn phases was a FeCl3-HCl
solution.

Due to the complexity of the Sb3Sn4 and SbSn crystal
structures, as described by different authors,[33-36] and the
limited resolution of XRD patterns, only a few samples
were investigated by XRD, and the investigation was dis-
continued because the diffractometer available in our lab-
oratory does not resolve the reflexes of the two Sb-Sn
intermediate phases.

3 Results and Discussion

3.1 Sb-Sn Phase Diagram

Based on the DSC and micrographic analyses, the new
experimental results are shown in Fig. 1, together with
literature data for comparison. For the Sb-Sn phase dia-
two versions are proposed that may be used to
gram,

account for the present results. However, as highlighted in
the Discussion section, we do not reach a conclusive rep-
resentation. The list of the invariant reactions identified in
this work through DSC is given in Table 1 together with
experimental literature data.

3.2 Annealed Sb-Sn Samples

Pieces of the synthesized alloys of various compositions
have been annealed at temperatures from 200 to 380 (cid:3)C for
different periods and generally quenched in icy water.
Details of these treatments are given in Table 2. In several
cases,
the samples were powdered either directly after
alloying or after a first annealing step.

Micrographs of samples in the 30.0 to 39.0 at.% Sb
composition range, annealed at 280-300 (cid:3)C, show Sb3Sn4
crystals (43.0-44.0 at.% Sb) surrounded by b(Sn),
the
residual melt. Some LOM and SEM images might help the
discussion. Figure 2 displays the Secondary Electron (SE)
image of sample #17A (30.0 at.% Sb) annealed at 280 (cid:3)C
in comparison with the LOM image of the same sample,
after DSC analysis (#5). Figure 3 reproduces the BSE

Fig. 1 Sb-Sn phase diagram. New experimental results (TW) and literature data.[27,33] Two graphical representations are sketched, see
discussion in section 3.3

123

J. Phase Equilib. Diffus. (2021) 42:63–76

67

Table 2 Annealed and icy quenched Sb-Sn samples

at.% Sb (SEM/
EDS)

Thermal treatment T, (cid:3)C, days (d)

Phases

at.% Sb (SEM/EDS)

Remarks

#

1A

2A

3A

4A

5A

6A

7A

8A

9A

10A

11A

12A

13A

14A

15A

16A

17A

18A

19A

20A

21A

22A

23A

24A

25A

26A

16.0

37.5

38.0

41.0

41.5

44.4

45.0

45.7

49.0

56.0

59.0

79.8

37.0

38.4

39.0

14.2

30.0

30.0

37.0

37.9

38.0

38.7

39.0

41.5

44.0

44.5

205, 60 d

205. 60 d

205, 63 d

205, 63 d

205, 63 d

205, 60 d

205, 84 d

205, 109 d

205, 63 d

200, 84 d

205, 60 d

205, 19 d

225, 10 d

225, 6 d

300, 13 d

280, 22 d

300, 13 d

305, 15 d

300, 13 d

295, 90 d

280, 8 d and 225, 60 d

280 (cid:3)C, 19 d

280, 14 d

280, 14 d and 300, 60 d

280, 14 d and 305, 13 d

Sb3Sn4
bSn
Sb3Sn4
bSn
Sb3Sn4
bSn
Sb3Sn4
bSn
Sb3Sn4
bSn
Sb3Sn4
Sb3Sn4
Uneven crystals

Geometric crystals

SbSn

SbSn

SbSn

Sb

SbSn

Sb
Sb3Sn4
bSn
Sb3Sn4
bSn
Sb3Sn4
bSn
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4
L
Sb3Sn4

54-57

88.0

62.5-63.0

43.2

5.0

43.5

8.0

44.0

8.0

43.5
8.0

44.0

11.0

44.4

45.0

48.0

44.0

49.0

56.0

90.0

43.0

8
43.0

7.5

43.0

9.0

44.0

11.0

43.0

8.0

43.5

10.0

43.5

6.0

6.0

43.0

10.0
43.5

—

43.0

8.0

43.3

4.0

44.0

14.0

One-phase sample

One-phase sample

Two-phase field

See Fig. 6

One-phase sample

One-phase sample

Composition fluctuation

Droplets, see Fig. 4

Droplets

Droplets, see Fig. 5

Droplets, see Fig. 2b

Droplets

Droplets

Droplets

Droplets

Droplets

Droplets, see Fig. 3

44.0-45.0

One-phase sample

123

340, 3 h and 310, 14 d

43.0-44.0

Droplets

Table 2 continued

at.% Sb (SEM/
EDS)

Thermal treatment T, (cid:3)C, days (d)

Phases

at.% Sb (SEM/EDS)

Remarks

68

#

27A

28A

29A

30A

31A

32A

33A

34A

35A

36A
37A

38A

39A

300, 100 d from crushed powders

280, 22 d and 305, 12 d

45.7

48.5

49.0

54.0

56.0

58.0

61.0

80.5

41.0

44.0
44.0

45.0

54.5

300, 48 d

300, 18 d

300, 13 d

300, 13 d

300, 13 d

360, 13 d

360, 20 d
360, 27 d

360, 13 d

380, 13 d

crystals

precipitate

SbSn

SbSn

SbSn

SbSn

SbSn

SbSn

Sb

Sb

Sb

L

SbSn

SbSn

L

SbSn

J. Phase Equilib. Diffus. (2021) 42:63–76

Two phases, see text and Fig. 7

One-phase sample

48.0-50.0

One-phase sample, composition
fluctuations

One-phase sample

One-phase sample

Few spots

54.0-58.0

Composition fluctuations

53.0-59.0

Composition fluctuations

Reaction underway
Reaction underway

53.0-56.0

One-phase with composition
fluctuations

46.0

44.8

48.5

54.0

56.0

88.5

89.5

91.0

48.0

9.0

47.5

9.0

300, 100 d from crushed powders

SbSn

60.0-61.0

Small composition fluctuations

Fig. 2 (a) Sb-Sn alloy #5 (30.0 at.% Sb) after DSC, without etching;
LOM image shows the large Sb3Sn4 crystals (43.0 at.% Sb) in which
Sn-rich particles (* 7 at.% Sb) formed during cooling are visible,

and (bSn). (b) Shows the SE image for sample (# 17A) (30.0 at.%
Sb), annealed 14 days at 280 (cid:3)C

image of sample #21A (38.0 at.% Sb) annealed at 280 (cid:3)C,
whereas the microstructure of sample #13A (37.0 at.% Sb)
from both LOM and SEM (BSE signal) is reported in
Fig. 4(a) and (b). Figure 5 displays the LOM photomicro-
graph of sample #15A (39.0 at.% Sb); the composition of

the phases obtained by EDS analysis, Sb3Sn4 (43.0 at.%
Sb) together with (bSn) 9.0 at.% Sb and droplets of 9.0
at.% Sb, was also confirmed by means of Wavelength
Dispersive x-ray Spectroscopy (WDS) for quantitative
analysis, employing a Jeol JSM 6460, working at 20 kV,

123

J. Phase Equilib. Diffus. (2021) 42:63–76

69

beam current of 10 nA, and pure elements as standard
materials (see Fig. 5b). All samples shown in the micro-
graphs in Fig. 2, 3, 4 and 5 contain droplet-shaped parti-
cles, which appear to be inside the large Sb3Sn4 grains.
This observation indeed is hardly explainable, and would
to microstructures expected after a catatectic reac-
fit
tion.[41-43] According to the thermal effects recorded by
DSC analysis (see section 3.3), for the Sb-Sn system, a
catatectic reaction could consist in the decomposition of
the (SbSn) solid phase into the liquid and solid Sb3Sn4
phase, upon cooling, as shown in the sketch of Fig. 1(a).
Whenever the droplet dimensions allowed EDS analysis,
compositions between 6 and 10 at.% Sb were determined.
This composition can be ascribed to an ‘‘unquenchable’’
liquid due to the steep slope of the liquid phase. Indeed,

Fig. 3 BSE image of the Sb-Sn sample # 21A (38.0 at.% Sb)
annealed at 280 (cid:3)C for 22 days and quenched. Large Sb3Sn4 crystals
(43.0 at.% Sb) with Sn-rich particles (* 6 at.% Sb) and (bSn)

this composition belongs to the peritectic 243 (cid:3)C reaction,
in which (bSn) forms from the reaction between the liquid
and Sb3Sn4. This of course means that compositions have
changed during cooling to room temperature, while,
according to the phase diagram,[29,32,33] the droplets should
contain about 17-18 at.% Sb at the peritectic or possible
catatectic reaction at T = 321 (cid:3)C.

The catatectic reaction, also called metatectic,[41] has
been widely described in Ref 42, 43 as ‘‘the inverse
melting as leading to the formation, on cooling, of liquid by
nucleation inside the crystals and therefore to the appear-
ance of a number of typical droplets having an approxi-
mately spherical geometry’’. Furthermore, a catatectic
reaction is a rare case found in few binary metal systems
such as Ag-In,[44,45] Sm-Al,[46] Cu-Sn,[29,47,48] and in bin-
ary systems in which one component shows an allotropic
transformation as in the Fe-Zr,[29] Nd-Au[43] or Mn-Sb.[49]
Since these Sn-rich droplet-shaped particles appear to be
inside the large Sb3Sn4 grains, we will discuss this topic in
depth in section 3.3.

In samples annealed at 300 (cid:3)C for 13 days (from #18A
up to #24A), large regular crystals with an average com-
position of 43.5 at.% Sb were observed. Furthermore, the
sample #8A (45.7 at.% Sb) annealed at 205 (cid:3)C for
109 days, shows in Fig. 6(a) Sb3Sn4 ? (SbSn) two-phase
field formed by geometric crystals (44.0 at.% Sb) together
with uneven crystals (48.0 at.% Sb). This is in line with the
results by Chen et al.,[32] as it was shown in Fig. 1(c) of
their paper (alloy N.11, 45.0 at.% Sb, annealed at 300 (cid:3)C
for 12 weeks), in which a two-phase sample formed by
crystals of ‘‘Sb3Sn2’’ (43.3 at.% Sb) together with (SbSn)
(46.5 at.% Sb) were observed.

Figure 7 shows the LOM microphotograph of alloy
#27A (45.7 at.% Sb) annealed at 300 (cid:3)C for 48 days, with
the presence of a precipitate having 44.8 at.% Sb

Fig. 4 Sb-Sn sample #13A (37.0 at.% Sb) annealed at 225 (cid:3)C for 10 days. (a) LOM image, Sb3Sn4 crystals (SEM analysis 43.0 at% Sb), dark-
gray bSn (8 at.%Sb). (b) SE image: Sb3Sn4 crystals (43.0 at.% Sb) with small Sn-rich particles (* 8 at.% Sb) inside and (bSn)

123

70

J. Phase Equilib. Diffus. (2021) 42:63–76

Fig. 5 Sb-Sn sample # 15A (39at%Sb), annealed 8 days at 280 (cid:3)C and 60 days at 225 (cid:3)C. (a) LOM photomicrograph after etching: large Sb3Sn4
(43.0 at.%Sb) crystals with droplet-shaped particles occurring inside (* 9.0 at.% Sb) and (bSn). (b) SE image and WDX analysis (see text)

Fig. 6 SE and BSE images of
the Sb-Sn alloy # 8A (45.7 at.%
Sb) annealed at 205 (cid:3)C for
109 days. Sb3Sn4-SbSn two-
phase field formed by uneven
crystals (48.0 at.% Sb) and
geometric crystals of Sb3Sn4
(44.0 at%Sb), see text

composition inside the (SbSn) crystals. The microstructure
can be explained as the boundary between the (Sb3Sn4-
? SbSn) two-phase field and the SbSn one-phase field.
Therefore at 300 (cid:3)C, beyond the alloy compositions con-
taining around 46 at.% Sb, we enter the (SbSn) one-phase
region. This is further confirmed by the EDS analysis
performed on sample #28A (48.5 at.% Sb, annealed at
18 days). As
300 (cid:3)C for
different
underlined
authors,[34,35]
the EDS analysis evidenced a somewhat
variable local composition for the (SbSn) phase, which
cannot be fully equilibrated at 300 (cid:3)C, even after long

by

annealing time (up to 100 days). Sample #29A (49.0 at.%
Sb) annealed at 300 (cid:3)C for 100 days, gives only crystals of
(SbSn) with a composition varying between 48 and 50 at.%
Sb. Only in a few cases, we were able to obtain homoge-
neous (SbSn) samples (see Table 2). Accordingly, we
conclude that the (SbSn) saturation limit at 200 (cid:3)C must be
situated at about 48 to 60 at.% Sb, whereas, at 300 (cid:3)C it
occurs between 46 and 60 at.% Sb. Most authors agree with
the opinion that Sb can dissolve about 10 at.% Sn. We
found similar values in the annealed samples (#33A and

123