Available via license: CC BY 3.0
Content may be subject to copyright.
IOP Conference Series: Materials Science and Engineering
PAPER • OPEN ACCESS
Contribution to the aluminum–tin–zinc ternary system
To cite this article: J Drápala et al 2017 IOP Conf. Ser.: Mater. Sci. Eng. 266 012002
View the article online for updates and enhancements.
This content was downloaded from IP address 191.101.89.109 on 23/11/2017 at 14:41
1
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Contribution to the aluminum–tin–zinc ternary system
J Drápala, G Kostiuková and M Losertová
Vysoká škola báňská – Technical University of Ostrava, Faculty of Metallurgy and
Materials Engineering, 17. listopadu 15, 708 33 Ostrava – Poruba, Czech Republic
Jaromir.Drapala@vsb.cz
Abstract. The Sn–Zn–Al alloys are one of significant candidates in the proposal of alternative
lead-free solders for higher temperature soldering. This paper deals with the study of the
aluminum–tin–zinc system. Twenty Sn–Zn–Al alloys together with six binary Sn–Zn alloys
were prepared and studied experimentally. Alloys were prepared from pure Sn, Zn and Al
(melting and cooling in a vacuum resistance furnace). The specimens were studied
metallographically including the micro-hardness measurements, complete chemical analysis
(ICP-AES, OES), X-ray micro-analysis of alloys by SEM and EDX in order to determine the
composition and identification of individual phases. Significant temperatures and enthalpies of
phase transformations were determined by DTA. After long-term annealing of selected alloys
in vacuum followed by quenching the structural and chemical microanalyses of the present
phases and their limit concentrations were carried out. The achieved results were compared
with the thermodynamic modelling of the ternary Sn–Zn–Al system (computer programs
THERMOCALC, MTDATA, PANDAT and databases CALPHAD, COST). Electrical
resistivity, density, magnetic susceptibility and wettability of Sn–Zn–Al solders were measured
as well.
1. Introduction
The Sn–Zn–Al ternary system can be one of applicable alternative alloys to high lead content Pb–Sn
solders, when the soldering process reaches temperatures as high as 300 °C. These solders can be used
above all in automotive industry, in heat exchangers and in special applications. Tin – zinc alloys are
used as solders for aluminum.
The tin – zinc binary system diagram is of a eutectic type. The eutectic reaction proceeds at the
temperature of 198.5 °C and the concentration of 85.1 at.% Sn. The zinc limit solubility in tin at this
temperature corresponds to 0.6 at.% Zn, while on the other hand, the tin maximum solubility in
hexagonal zinc is only 0.039 at.% Sn [1].
The binary diagram of the aluminum – tin system [1, 2] is also of a eutectic type with
a considerably limited solubility in the area of solid solution (Al) and tetragonal (Sn). The tin
solubility in (Al) solid solution exhibits a retrograde character with a maximum of approximately
0.026 at.% Sn at the temperature of 625±20 °C [2].
The Al–Zn equilibrium phase diagram [1] represents a eutectic system involving another reaction
and the existence of a miscibility gap area in solid phase α (Al).The eutectic reaction proceeds at
381 °C closer to Zn side (the eutectic point corresponds to 88.7 at.% Zn). (Al) solid solution has an
extended area of homogeneity interrupted at the temperature of 351.5 °C and the content of 39.5 at.%
Zn. Between this temperature and the temperature of 277 °C, (Al) disintegration into two solid
solutions occurs: (Al) with a lower Zn content and (Al”) with a high zinc content (up to 59 at.%). The
2
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Zn solubility in (α Al) increases from 2.2 at.% at 110 °C to 16.5 at.% at 277 °C. Above temperature
277 °C the zinc solubility increases from 59±1 at.% Zn at 277 °C to 67±1 at.% Zn at the eutectic
reaction. The maximum solubility of Al in (Zn) is 2.8±0.2 at.% Al at the eutectic temperature and
decreases along with a decreasing temperature to 1.6 at.% Al at 277 °C and down to 0.07 at.% Al at
the room temperature. Then, (Al) + (Zn) two-phase structure can be expected in the equilibrium
crystallization of Al–Zn alloys. In Table 1 thermodynamic values and reactions in Al–Sn, Sn–Zn and
Al–Zn binary systems are listed.
Table 1.Thermodynamic values of phase transformations of Sn, Zn and Al pure metals and
equilibrium compositions of solidus and liquidus at the eutectic or eutectoid reaction in Al–Sn, Sn–Zn
and Al–Zn binary systems [1, 2].
Metal A T
mA
(°C) ∆H
mA
(J/mol)
Sn 232 7029
Zn 419.6 7322
Al 660.5 10711
T
mA
– melting point of metal A
∆H
mA
– molar enthalpy of fusion for metal A
T
E
– eutectic temperature
X
SE
– maximal solubility of element B in A at T
E
X
LE
– concentration of element B in liquidus at T
E
* – eutectoid reaction
The Al–Sn–Zn ternary system was studied in details by Prowans [3] and later by Nayak [4] and other
authors [5, 6]. A revision of the Al–Sn–Zn ternary system and a thermodynamic calculation was
performed by Fries [7] (see Figure 1). The latest findings on this system are mentioned on
www.msiport.com [8] and www1.asminternational.org [9] proposed by M. Hubert-Protopopescu [10].
When researching this system
it was found out that the
liquidus areas in the Al–Sn–Zn
ternary system including their
isotherms differ according to
data of authors [3-10] both in a
position of the eutectic valley
and in a character of the
isoliquidus courses. So, there
are significant disproportions
in describing this system
according to various authors.
The thermodynamically shaped
liquidus area according to Fries
[7] is closer to the diagram
designed experimen-tally by
Nayak [4]. In Table 2 are
summarized the invariant
equilibria in the Al–Sn–Zn
ternary system.
Figure 1.Thermodynamic calculated liquidus surface of the Al–Sn–Zn ternary system by [7, 10] with
the denotation of nine sections forimage of isopleths.
Binary X
SE
X
LE
T
E
system (at.%)
(at.%) (°C)
Sn–Al 1.0 2.4 228
Al–Sn <0.01
97.6 228
Sn–Zn 0.6 14.9 198.5
Zn–Sn 0.04 85.1 198.5
Zn–Al 2.8 11.3 381
Al–Zn 67.0 88.7 381
Zn–Al*
1.6* 41* 277*
Al–Zn*
16.5* 59* 277*
Al
Sn
Zn
a)
b)
c)
d)
f)
e)
g)
h)
i)
3
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Table 2.Thermodynamic invariant equilibria in the Al–Sn–Zn ternary system.
On the basis of structural analyses, SEM and x-ray analysis EDX and DTA, the aim of this paper
was to examine validity of the published diagrams, to perform thermodynamic calculations of phase
equilibria after long-term isothermic annealing of specimens of selected alloys. Other properties of the
selected Al–Sn–Zn alloys, such as resistivity, density, magnetic susceptibility and wettability are
tested as well.
2. Assessment of the Al–Sn–Zn ternary system
Figure 2 shows isothermal cross-sections through the ternary Al–Sn–Zn system at the temperatures of
300 °C to 270 °C according to [7-11]. The diagram in Figure 2 b) is beside the character of the
existence of (Al) + (Al”) phases area, which penetrate deep inside the diagram at the expense of three-
phase (Al) + (Al”) + L area in the cross-section at 290 °C. Comparing the cross-sections at the
temperatures of 300 °C and 280 °C, it is obvious that the cross-section at 290 °C is improperly plotted
in term of the coexistence of phases. Sebaoun [11] studied the Al–Sn–Zn phase diagram using the
reactive diffusion with preparing three diffusion pairs: Zn-10 wt.% Sn/Al, Zn-28 wt.% Al/Sn and Al-
10 wt.% Sn/Zn (see lines g, h and i in Figure 1), which were annealed for a period of 3 up to 6 hours at
the temperatures of 270, 280 and 290 °C. Following annealing, the specimens were water-quenched
rapidly. From structural and chemical microanalyses (SEM/EDX) the respective diffusion routes were
found out, on the basis of which a modified Al–Sn–Zn diagram was created (see the isothermal cross-
sections in Figure 2 b), c) and d).The following phases were found out: (Al), (Al”), (Zn) and
a solidified tin-rich melt. In all cases the tin solubility in (Al”) phase was < 0.8 at.% Sn. That’s why
the isothermal cross-section in Figure 2 b) has been explicated incorrectly.
U
1
L + (Al) ↔ (Al)” + (Zn) E
1
L ↔ (Al) + (Zn) + (Sn)
T
(°C) Phase Sn
(at.%)
Zn
(at.%)
Al
(at.%) Ref. T
(°C) Phase Sn
(at.%)
Zn
(at.%)
Al
(at.%) Ref.
277.8
L 61.76 31.79 6.45
[7] 195.4
L 84 13.86 2.14
[7]
(Al) 0.01 14.17 85.82 (Al) 0 5.59 94.41
(Al)” 0.02 58.97 41.01 (Zn) 0.05 99.35 0.6
(Zn) 0.13 98.24 1.63 (Sn) 98.74 0.61 0.65
196 L 86.6 10.7 3.1 [5]
196 L 77.7 16.6 5.7 [6]
197.3 L 84.7 12.9 2.4 [3]
198 L 82.5 15 2.5 [4]
Al
Zn
Sn
(Zn)
(Al)
290 °C
b)
Al
Zn
Sn
(Zn)
300 °C
a)
(Al) + L
(Al) + L
(
Zn
)
+ L
(
Zn
)
+ L
(Al”)
(Al) +(Al”) + L
(Zn) +(Al”) + L
(Al
”)
+ L
(Zn) +(Al”)
(Al”)
(Zn) +(Al”) + L
(Al
”)
+ L
(Al) +(Al”)
(
Al) +
(Al”) +
L
(Zn) +(Al”)
(Al) +(Al”)
4
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Figure 2. Isothermal sections in Al–Sn–Zn ternary system at temperatures 300 °C by [7, 10]
and at 290 °C, 280 °C a 270 °C by [11].
Figure 3. Vertical sections in Al–Sn–Zn ternary system a) red line „a” in Figure 1 [7, 10]; b) green
line„b” in Figure 1 [3]; c) light blue line „e” in Figure 1 [3]; d) orange line „f” in Figure 1 [11].
Sn
(Zn)
(Al)
280 °C
c)
Al
Zn
Al
Zn
Sn
(Zn)
(Al)
270 °C
d)
(Al) + L
(Al) + L
(
Zn
)
+ L
(
Zn
)
+ L
(Al) +(Zn) + L
(Al) +(Zn)
(Zn) +(Al”)
(Al”)
(Al) +(Al”)
(Al) +(Al”) + L
(Zn) +(Al”) + L
(Al
”)
+ L
c)
26N
(Al) + (Al”) + L
(Al) + (Al”)
d
)
22N
(Al) + (Al”) + L
(Al”) + L
(Al”)
(Al”) + L
a)
b
)
16N
23N
4N
24N
K
(Al”) + L
277 °C
194 °C
(Al) + (Zn) + L
(Al) + (Sn) rt + Zn
(Al) + (Sn) rt + L
(Al) + (Al”) + L (Zn) + (Al”) + L
(Al”) + L
5
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Figure 3 depicts 4 different isoplethic sections in the ternary Al–Sn–Zn system which has been
marked out in Figure 1. Figure 3 a) shows a cross-section for 50 at. % Sn according to authors [7, 10].
On confronting this diagram with isothermal cross-sections in Figure 2, we may state that the given
vertical cross-section corresponds to reality. In the diagram in Figure 3 b) Prowans [3] has marked out
two areas with a congruent designation (Al) + (Zn) + L, though a phase transformation occurs at the
temperature of 275 °C. However, this contradicts the Gibbs phase rule as well as laws of
thermodynamics. Therefore we added two thermodynamically conforming curves into the diagram in
the temperature area of 275 to 350 °C. The succession of phase formation in Figure 3 b) left to right is
as follows:(Al) + L → (Al) + (Al”) + L → (Al”) + L→ (Zn) + (Al”) + L → (Zn) + L. The isopleth
section in Figure 3 b) conforms to the Sebaoun ideas – see Figure 2 b), however, considering the
diagrams in Figures 2 a), c) and d), the tin solubility in the area of (Al) + (Al”) phases is negligible, so
there is a disproportion here again. In Figure 3 b), the authors’ scheme is plotted as the red curve and
(Al) + (Al”) + L area includes both the existing yellow tinged area above the temperature of 278 °C
and the white area (marked with a green arrow), which has been considered an area of the coexistence
of (Al) + (Al”) phases according to [3, 11]. In Figure 3 d) all the existing phases are not complemented
either. For example for the temperature of 300 °C, the succession of phase formation left to right is as
follows: (Al”) → (Al”) + L → (Al) + (Al”) + L→ (Al) + L → L. In Figure 3 b), the authors have
complemented the missing curves for the area of the coexistence of (Al”) phase – the red curve and for
(Al”) + L area – the violet curve. These complemented isopleth sections have a considerable
significance for understanding the succession of phase formation during crystallization of particular
alloys.
3. Experiment
3.1. Experimental samples
Six binary Sn-Zn and fourteen ternary Al-Sn-Zn alloys with 0.6 to 3 wt.% Al were prepared
experimentally (melting in the electrical resistance furnace and casting into graphite mould). The
purity of the initial metals was 3N5 at minimum. Next alloys with different ratios of individual
elements Al, Sn and Zn (marked with the letter N – see Table 3) were prepared experimentally by
smelting of pure metals in an electrical resistance furnace in evacuated quartz ampoules. These Al–
Sn–Zn alloys were melted at the temperature of 750 °C in a muffle furnace, held on this temperature
for 15 minutes, overturned for homogenization and held on the temperature for another hour. Finally
were the samples overturned again and left in the furnace to cool down until the temperature of 70 °C
was reached [12 - 15].
The selected specimens were also subjected to long-term annealing in evacuated ampoules at
temperatures of 250, 300 and 350 °C for periods of 3, 7, 14 and 28 days followed by water quenching.
The alloys were studied metallographically after casting and after homogenization annealing, their
micro-hardness was measured, their overall chemical composition was analysed (ICP-AES and OES
on the SPECTROMAX device) and X-ray micro-analysis (EDX, WDX) of the coexisting phases was
performed, too.
3.2. Differential thermal analysis
The differential thermal analysis (DTA) of all the specimens was performed on the SETARAM
SYSTEM 18
TM
apparatus. The analyses were carried out twice at the speed of heating as well as
cooling 4 K/min. Table 3 presents the final results of the DTA measurements of liquid temperatures T
L
and eutectic transformation T
E
of all the specimens including the determination of phase
transformation enthalpies. The eutectic reaction temperature in binary alloys was 199 °C, while in all
the ternary alloys it was T
E
= 197.7 ± 0.7 °C. Another invariant reaction was found at the temperature
of T
U
= 278.6 ± 0.7 °C in some specimens. It was caused by the aluminum presence in the area of
medium and high Zn concentrations. Further transformations T
1
and T
2
occur in ternary alloys between
the temperatures 278 °C and liquid, which requires a more detailed analysis utilizing further methods.
6
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
The experimentally obtained transition temperatures are presented in the calculated isopleth for 3
wt.% of Al in Figure 4. Excellent agreement was achieved in the case of ternary invariant reactions
temperatures [16-18].
Table 3. Nominal chemical composition of Al–Sn–Zn alloys (charge) and results of DTA analysis
(heating rate 4 °C/min).
Specimen Sn
(at.%)
Zn
(at.%)
Al
(at.%)
T
E
(°C)
T
U
(°C)
T
1
(°C)
T
2
(°C)
T
L
(°C)
1 5.5 87.2 7.3 197.1 279.3 360.1 385.0
2 18.2 73.8 8 197.8 279.0 346.8 355.5
3 33.7 57.4 8.9 198.9 278.3 337.3 342.3
4 52.9 37.1 10 198.8 278.7 294.1 324.0
5 70.8 18.2 11 198.6 241.0 360.0
6 5.8 94.2 0 199.1 397.0
7 19.1 80.9 0 199.4 382.0
8 35.5 64.5 0 198.9 357.3
9 56.2 43.8 0 198.8 323.5
10 75.7 24.3 0 198.9 272.6
11 82.5 15 2.5 196.6 207.1 263.0
12 60 33 7 197.4 279.8 294.2 342.4
13 70.5 24.8 4.7 197.5 258.1 324.3
14 10 80 10 197.2 278.1 355.7 363.1
15 39 51.8 9.2 197.8 278.7 324.0 348.5
16 50 43 7 197.6 279.1 310.2 333.0
17 91 6.5 2.5 196.9 215.1 268.0
18 85 15 0 198.0
19 80 12 8 198.4 207.8 379.4
20 25 65 10 198.1 278.2 341.4 346.0
1N 85 12.5 2.5 198.8
3N 70.4 24.7 4.9 197.6 253.8
4N 10 80 10 196.7 277.3 327.1 358.8
10N 70 20 10 197.6 239.0
12N 60 33 7 197.9 296.5
15N 39 52 9 197.3 277.5 325.0
16N 50 43 7 197.9 277.5 313.6
21N 20 65 15 197.5 277.6 325.3 363.4 439.6
22N 40 35 25 197.5 277.8 297.0 318.6 406.0
23N 40 42 18 197.6 277.8 311.6 337.0
24N 9.8 57.9 32.3 196.9 277.4 344.5 ~412 446.9
25N 18.7 29.8 51.5 197.4 277.4 333.8 ~439 536.4
26N 20.5 39 40.5 197.8 277.4 302.9 ~332 506.5
7
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Figure 4. Vertical section in Al–Sn–Zn ternary diagram for 3 wt.% Al with the experimental values
of temperatures (DTA analysis).
3.3. Microhardness
The Vickers microhardness test HV was performed. The results of HV 0.1 measurements (load 100 g)
are plotted in Figure 5 in dependence on the tin content for both the binary Sn–Zn system and ternary
Al–Sn–Zn alloys containing 3 wt.% Al. It can be concluded from measurements that the
microhardness decreased with the rising tin content in all hypoeutectic alloys. However, aluminum
increased microhardness substantially, especially with high zinc contents, when the microhardness
value was nearly double compared to the binary tin-zinc system.
Figure 5. Dependence of the
microhardness HV 0.1 on the
tin content in binary Sn–Zn
system and ternary Al–Sn–Zn
alloys containing 3 wt.% Al.
T[C]
w%(SN)
0
100
200
300
400
500
0 10 20 30 40 50 60 70 80 90 100
w%(SN)
T[C]
LIQUID
LIQUID+FC C_A1
LIQUID+FCC_A1
LIQUID+FCC_A1+FCC_A1
LIQUID+HCP_ZN+FCC_A1
LIQUID+HCP_ZN
HCP_ZN+FCC_A1
HCP_ZN+FCC_A1+LIQUID
BCT_A5+HCP_ZN+FCC_A1
LIQUID+FCC_A1+BCT_A5
0 10 20 30 40 50 60 70 80 90 100
0
100
200
300
400
500
Liquid
L + (Zn)
(Zn) + (Al’’)
(Sn) + (Zn) + (Al)
L + (Zn) + (Al)
L + (Zn) + (Al’’)
L + (Al’’)
L + (Al)
L + (Al) + (Al’’)
L + (Al) + (Sn)
(Sn) + (Al)
Temperature
(°C)
Wt.%Sn
278.6°C
197
.
7°C
Alloy:1 14 2 20 3 15 16 4 12 5 19
T
L
T
2
T
U
T
E
T
1
Sn
Zn
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
80
90
Sn (wt.%)
HV
0.1
Sn-Zn-3 wt.% Al
Sn-Zn alloys
Z
n
8
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
3.4. Metallographic study
The metallographic analyses showed basically fine-grained structure, mostly of a eutectic type, often
with present dendritic formations. A certain portion of fine lamellar ternary eutectics (Al)+(Sn)+(Zn)
with a minority Al presence was identified in all the samples. Needles of primarily precipitated Zn and
the secondary Al–Zn alloy containing 9÷12 at.% Zn were found in the structures – see Figures 6 and 7.
Figure 6.
SEI/EDX structural and phase analysis of the sample 16N (7 at.% Al, 43 at.% Zn, 50 at.% Sn),
liquid temperature T
L
= 314 °C.
Annealing 250 °C/3 days Annealing 250 °C/ 28 days
Figure 7. Microstructural and phase analysis of the sample 16N after annealing at 250 °C in evacuated
ampoule followed quenching into water.
In Figure 6 on the lefts shown a microstructure of the alloy 16N (charge: 7 at.% Al, 43 at.% Zn,
50 at.% Sn) after casting – initial state. In Figure 6 on the right is presented a microstructure with
eutectic composition. Main part was formed by pure tin, on the thin lamellas of pure zinc were
sometimes seen particles of (Al) solid solution. Area fraction of the Zn + (Al-Zn) phase in eutectic was
9.6 %. Microhardness of the eutectic was HV
0.01
= 15. In Figure 7 is shown microstructure of the alloy
16N after long-time heat treatmentat the temperature 250 °C – 3 and 28 days. It can see a matrix with
eutectic and two phases (Al) with 10 – 12 at.% Zn and (Zn) with 3 at.% Al. Presence of tin in these
phases has not been observed. In Figure 8 is shown a microstructure of the alloy 15N (charge: 9 at.%
Al, 52 at.% Zn, 39 at.% Sn) after annealing at the temperatures 300 °C/14 days and 250 °C/28 days. In
the microstructure were found: ternary eutectic with high content of tin, then solid solution (Zn) with
aluminum (3 and 6 at. %). At temperature 250 °C was in structure present a phase (Al) with 11 at.%
Zn and at 300 °C was found a phase (Al”) with 62.3 at.% Zn.
Al-Zn
88.4 at.% Al
Sn
Zn
Area analysis:
Zn needles:15.8 %
Phase
Al
-
Zn:11
.
5
%
200 µ
µµ
µm 50
µ
µµ
µ
m
Al-Zn
Sn
Zn
Ternary eutectic
Spectrum Al (at.%) Zn (at.%)
1, 2, 4 90.78 9.22
3 2.97 97.04
Al–Zn
88 at.% Al
500
µ
µµ
µ
m
Zn–Al
97 at.% Zn
9
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Annealing 300 °C/14 days Annealing 250 °C/ 28 days
Figure 8.
SEI/EDX structural and phase analysis of the sample 16N (7 at.% Al, 43 at.% Zn, 50 at.% Sn),
liquid temperature T
L
= 314 °C.
3.5. Isothermal diffusion in ternary system Al–Sn–Zn
Figure 9. SEM/EDX microanalysis of the diffusion couple: pure tin-alloy 24N (32 at.% Al, 58 at.%
Zn, 10 at.% Sn) after vacuum annealing 300 °C/6 hours followed by quenching into water.
1
1
1
Spectrum Al (at.%) Zn (at.%)
1 2.44 97.56
2, 3, 4, 5 37.70 62.30
Spectrum Al (at.%) Zn (at.%)
1 89.0 11.0
2 6.0 94.0
Sn
24N
A
BC
Name
Analysis
Al
Zn
Sn
Region A area 4.75 25.61 69.64
Region B area 41.17 20.31 38.52
Region C area 46.60 51.08 2.32
Spectrum 1 (Zn) 1.82 98.18
Spectrum 2 Zn 100.00
Spektrum 3 (Al) 81.40 17.86 0.74
Spektrum 4 eutectic 9.66 90.34
Spektrum 5 (Al) 85.52 14.48
Spectrum 6 Zn 100.00
Spectrum 7 eutectic 8.62 91.38
Spectrum 8,9 (Al) 83.71 15.62 0.68
Spectrum 10 (Al") 51.18 48.34 0.48
atomic %
10
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Phases formed by isothermal diffusion in the Al–Zn–Sn ternary system have been studied at 300 °C
for the three various couples (Al–Zn–Sn)-Sn. The experiment completes a study using direct together
with differential thermal analysis, and provides better understanding of isothermal equilibria.
Comparing the results allows different possible diffusion paths to be proposed, and EDX microprobe
analysis of phases in isothermal equilibrium gives detailed information on the Al–Zn–Sn phase
diagram. As an example we show a profile of the diffusion couple: pure tin-alloy 24N (32 at.% Al, 58
at.% Zn, 10 at.% Sn) after vacuum annealing 300 °C/6 hours followed by quenching into water – see
Figure 9. Region A was before the heat treatment a pure tin. During annealing diffused here Zn and Al
from alloy 24N. We found in the structure: eutectics, needles of pure Zn and phase (Al) with c. 18
at.% Zn. Transition region B contained: ternary eutectic, pure Zn and phase (Al) with c. 15 at.% Zn.
Region C with high concentration of Zn contained minority portion of eutectic and two phases: (Al)
with c. 15 at.% Zn and (Al”) 48 at.% Zn. It means that in region C were at the temperature 300 °C in
equilibrium three phases: (Zn) + (Al) + (Al”).
4. Thermodynamic modeling of Al–Zn–Sn ternary system
Theoretically calculated isothermal sections (software MTDATA) at 300 and 250 °C in the Al–Sn–Zn
ternary system are presented in Figure 10. The positions of the chemical compositions of some alloys
are noted in these diagrams together with the results of area structural analyses of individual phases.
We found in all the studied alloys the presence of the ternary eutectic (Sn) + (Zn) + (Al) with high
concentration of tin. Then long needles of pure zinc were found in the microstructure of alloys. Next
phases were: solid solutions (Al) with content about 8-12 at.% Zn and in some cases a solid solutions
(Al”) with 50-60 at.% Zn.
On the basis of the DTA, structural and EDX analyses of Al–Sn–Zn alloys after crystallization and
after long-time annealing in evacuated ampoules (250, 300 and 350 °C; times 3, 7, 14 and 28 days)
followed by quenching into water were observed the proportions of the individual phases in the
structure. The results of the phase equilibria for annealing temperatures 300 and 250 °C are presented
in Figure 10 (examples for alloys 4N, 15N and 16N). The positions of the chemical compositions of
some alloys are noted in these diagrams on the basis of results of area structural analyses of individual
phases [16, 17].
11
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Figure 10. Isothermal sections in Al–Sn–Zn ternary diagram for temperatures 300 and 250 °C with the
positions of experimental alloys and phase equilibrium (area analyses of structures after long-time
annealing). Notes: fcc(Al) = fcc(1) is solid solution (Al), fcc(2) is solid solution (Al”),
hcp(Zn) is solid solution (Zn).
5. Physical and technological properties of Al–Sn–Zn alloys
5.1.Electrical resistivity
Resistivity of the selected specimens of Al–Sn–Zn alloys was measured by two methods: the contact
method (two-point and four-point) in accordance with DIN/IEC 468 [19] and the contactless method
for measuring in a rotating magnetic field. The specimens were adapted to a unified shape in order to
ensure reproducibility of the measurements. The results of the contact method measurements are
shown in Table 4. It was found out that resistivity in the Al–Sn–Zn alloys was lower than in P5 solder
of SAC type commonly used today.
Resistivity of metals depends on purity of particular elements and on structural indefectibility of
materials. It can be changed significantly even by a negligible amount of impurities. Comparing the
results of our own measurements of resistivity with literary data, for example for Sn–Pb alloy
ρ
= 14.4 µΩ·cm according to [20] and 15.0 µΩ·cm according to [21], or 14.27 µΩ·cm [22] compared
to 15.3 µΩ·cm resistivity measured in our experiment (see Table 4).
Table 4. Electrical resistivity (µΩ·cm) of solders
Notes: ChU – measured at Charles University in Prague, CTU –
Czech Technical University in Prague
Table 5. List of alloys for measurements.
Alloy
Composition (wt.%)
Resitivity
ChU
Resistivity
CTU
P5
Sn
95.5 Ag3.8 Cu0.7 11.98 11.7
P6
Sn
63 Pb37 15.26 16.2
5 Sn84 Al2.7 Zn13.3 10.8
15 Sn56 Al3 Zn41 9.34
Sample
no.
Zn
(at.%)
Sn
(at.%)
Al
(at.%)
T
L
(°C)
17 6.5 91 2.5 268
19 12 80 8 379
13 25 70 5 324
16 43 50 7 333
20 65 25 10 346
12
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
From this follows that the values measured in our experiment were by 12 % higher than literary
data in average. This may be caused by an initial condition of our specimens which were deformed by
cold rolling to foils of 0.2 mm thickness and were not annealed for internal defects reduction before
the measurements. Nevertheless, the results can be considered objective and comparable to one
another because all the specimens had the same geometric shape, dimensions and processing history.
The resistivity temperature dependences (see Figure 11) were linear in all the Al–Sn–Zn alloy
specimens and were fitted with a linear function:
[
]
)(1
LT
TT −+=
βρρ
, where
ρ
T
stands for the molten
metal resistivity at T
L
temperature, β stands for the temperature coefficient of specific resistance.
Temperature dependences
ρ
(T) were also linear for the solid phase and were fitted with a similar linear
function.
ρ
(T) curves almost merged with each other during heating as well as cooling in the liquid
state (molten metal). Certain hysteresis was observed in some specimens during the change from the
solid to liquid state and vice versa. The total uncertainty for determination of resistivity values was
determined ± 5 % [23, 24].
Figure 11. Temperature dependence of the electrical resistivity of some Al–Sn–Zn alloys.
5.2.Density
Figure 12 shows the temperature dependence of density for five Sn–Zn–Al alloys. Alloy no. 17 (high
Sn content) exhibits the highest density. It is not without interest that a pair of alloys no. 19 and 13 and
alloys no. 16 and 20 have approximately the same density at different zinc contents (see Table 5).
Figure 12. Temperature dependence of density of some Al–Sn–Zn alloys in liquid state.
0
10
20
30
40
50
60
70
80
0 100 200 300 400 500 600
Temperature (°C)
Resistivity (µ
Ω
Ω
Ω
Ω
.cm)
heating No. 17 cooling No. 17
heating No. 13 cooling No. 13
heating No. 16 cooling No. 16
heating No. 20 cooling No. 20
heating No. 19 cooling No. 19
6100
6200
6300
6400
6500
6600
6700
6800
250 350 450 550 650 750
Temperature (°C)
Density (kg / m
3)
)
No. 17
No. 19
No. 13
No. 16
No. 20
13
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
5.3.Magnetic susceptibility
Magnetic susceptibility χ was measured using the Faraday method during heating and subsequent
cooling at a rate of 3 K/min in a helium atmosphere in BeO crucibles [23, 24]. Pressure in the vacuum
chamber was 10
-3
Pa. In all the measurements it was found out that the Al–Sn–Zn alloys featured only
low susceptibility to the outer magnetic field, therefore the uncertainty in susceptibility absolute values
was on ± 10 % level. For all the specimens χ > 0 up to T = 200 to 210 °C (the solid phase area) and
above this temperature χ < 0 (for specimen no. 20 χ < 0 at T > 140 °C, however, a decrease was also
observed at 210 °C). Other anomalies were not found at liquid temperatures, however, worth noting
are temperature regions T = 380 to 460 °C and 530, or 580 °C, where susceptibility was constant,
which was an evidence of certain structural rearrangements. The graph (see Figure 13) shows here
evidently a step change of χ during the transition from the solid state to liquid.
Figure 13. Temperature dependence of the magnetic susceptibility of Al–Sn–Zn alloy no. 17.
5.4.Wettability
Wettability is defined as an ability of a molten solder to adhere to a surface of the material being
joined in the presence of a flux at a specified temperature. A formation of an intermetallic compound
is an essential condition of good wettability and the bond between a solder and a wetted metal.
The wetting balance method was used, enabling to obtain accurate quantitative results and
observing also a non-equilibrium situation. A tested component was suspended on a dynamometer
above a pot filled with a molten solder. The measurement of wettability was carried out on
Meniscograph Solderability Tester GEC in CTU in Prague. The apparatus consists of a measuring
head with a spring microbalance, a holder for clamping the tested object, a solder bath and an
electronic control unit [25].
Regarding relatively small amounts of solder specimens, the measuring apparatus was modified in
the part of the molten bath, so that the alloy amount needed to perform the test was reduced to
approximately 8 cm
3
. Crucibles, made of stainless steel and graphite, were used for this purpose. A
small crucible with a specimen of the Al–Sn–Zn alloy was lowered and dipped into the main Pb–Sn
solder bath (the eutectic composition) of a volume of c. 250 cm
3
in the tester pot and clamped in
holders made of stainless material laid over the original pot.
Eight specimens were selected for the Al–Sn–Zn alloys wettability testing. Three types of fluxes
were used for testing of wettability of solders with various substrates on a meniscograph. Almost for
all the tested combinations: a solder, a material, a flux – a perfect wetting of a specimen by a solder
has not occur. Non-wetting of the specimens of Cu and Ni wires resulted from the used type of fluxes,
which appeared insufficient in term of the used measurement temperatures of molten solders. An
example of a non-wettable course is depicted in Figure 14.
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
0 100 200 300 400 500 600
Temperature (°C)
x*10
7
(emu / g)
No. 17 - heating
No. 17 - cooling
14
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
Figure 14. Wetting forces of the alloy 17.8 wt.% Sn, 78.2 wt.% Zn, 4 wt.% Al; Ni wire ø 1 mm;
flux RXZ (94-RXZ-M – flux on the basis of synthetic resin formachine soldering).
6. Discussion
When testing resistivity of solders, it was found out that the Sn–Zn–Al alloys had ρ values lower than
Pb–Sn solder and even lower than the best known alternative solder SAC, whereas the presence of
aluminum supported the resistivity decrease. Our prepared specimens were deformed by rolling before
the testing, which resulted in an increase of ρ by approximately 12 % in comparison with the non-
deformed condition. A linear increase in resistivity in the solid as well as in the liquid state (molten
metal) was observed from temperature dependences ρ(T) in five investigated alloys – see Figure 11.
According to expectations, a resistivity step increase occurred in all the solders in the area of the
eutectic reaction temperature. Complicated reactions at phase transformations occurred in specimens
of alloys no. 16 and 20 with high zinc contents (Table 5). A certain hysteresis degree was evident on
the curves.The lowest ρ value was exhibited by alloy no. 20 with a high Al content (10 at.%).
A linear decrease ρ
L
(T) was observed from temperature dependences of molten metal densities in
the above mentioned five specimens – see Figure 11 which was in conformity with theoretical
assumptions. An interesting finding was that alloys no. 19 and 13 and alloys 16 and 20 nearly overlaid
one another, which was caused by the presence of aluminum in a suitable ratio to Sn and Zn.
The magnetic susceptibility measurement confirmed that this method was sensitive to structural
and phase transformations, too. Here also, a jump of χ values at the eutectic temperature and a certain
degree of hysteresis behavior during heating and cooling of the specimens became evident (Figure 13).
Three types of fluxes were used for testing of wettability of solders with various substrates on a
meniscograph. Cu and Ni wires (1 mm diameter) were applied as substrates and 8 Sn–Zn–Al alloys
were tested. The measurements proved that nickel exhibited higher wettability with the solders used,
however, the measurements results showed low wettability in general compared to common Pb–Sn,
SAC and Sn-Cu solders. Here the presence of aluminum in the alloy had a negative effect; aluminum
has a tendency to form oxides on solder surfaces and their dissolution using the particular fluxes
failed. The results of measured courses of wetting forces “on plateau” (Figure 14) imply that the Sn–
Zn binary alloys can be preferred to aluminum containing alloys. These conclusions were confirmed
also by the results of a subjective image analysis. It can be stated that solders not containing aluminum
are better solderable. The solderability decreases with an increasing proportion of aluminum in a
soldering alloy. An issue remains to find an applicable flux even for these types of alloys. In this case,
it will be advisable to use nitrogen as the inert atmosphere for testing.
F
(mN)
-4
-3
-2
-1
0
1
2
3
4
5
0 2000 4000 6000 8000 10000
Time (ms)
15
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
7. Conclusion
The presented paper deals with a complex study of the Al–Sn–Zn ternary system both in term of phase
equilibria and structural and mechanical characteristics depending on the alloy chemical composition
and a method of heat treatment (a condition after crystallization or long-term annealing in vacuum at
three different temperatures and time periods followed by subsequent rapid cooling – quenching).
Further, measurements of resistivity, density, magnetic susceptibility and wettability of the selected
alloys were carried out. Some results are presented in relevant tables and graphs. The alloys were also
subjected to various corrosion tests not mentioned in this paper [26, 27].
For further applications it will be necessary to perform other technological tests of solderability to
find out an optimal flux and to optimize the chemical composition of the Al–Sn–Zn based lead-free
solder for particular soldered materials, especially for aluminum alloys.
Acknowledgements
This work was supported by Ministry of Education, Youth and Sport of the Czech Republic in the
project EU – COST MP0602 “Advanced solders materials for high-temperature applications” No. OC
08032 and in the project No. LO1203 "Regional Materials Science and Technology Centre –
Feasibility Program".
References
[1] Massalski, TD 1996 Binary Alloy Phase Diagrams. Second Edition Plus Updates on CD ROM
(Ohio: ASM International, Metal Park)
[2] Dorward RC 1976 Metal Trans. 7A 308
[3] Prowans S and Bohatyrewicz M 1968 Arch. hutn. 13 217
[4] Nayak AK 1975 Trans. Indian Inst. Met. 28 148
[5] Plumbridge DV 1911 On the binary and ternary alloys of aluminum, zinc, cadmium and tin.
Thesis (Munich: University of München Germany)
[6] Losana L and Carozzi E 1923 Gazetta Chimica Italiana 53 546
[7] FRIES, SG, LUKAS, HL, KUANG S, EFFENBERG, G 1991 Calculation of the Al–Zn–Sn
ternary system. In "User Aspects of Phase Diagrams", Ed. F. Hayes, The Institute of Metals,
London, 280-286
[8] http://www.msiport.com/ MSI Eureka Content - Materials Science International, GmbH.
[9] http://www1.asminternational.org/asmenterprise/apd(Ohio: ASM Alloy Phase Diagrams Center)
[10] Hubert-Protopopescu M 1993Aluminium – Tin – Zinc www.msiport.com
[11] Sebaoun A, Vincent D and Tréheaux D1987Materials Science and Technology 3 241
[12] Drápala J et al. 2011 Lead-free solders. Theoretical and experimental study of phase equilibria
and properties of new types of solders. Monograph (Ostrava: ES of VŠB – TU Ostrava) p
218
[13] Kostiuková G 2016 Low-temperature alloys for lead-free solders. Study of physico-chemical
characteristics. Dissertation. (Ostrava: VŠB – Technical University of Ostrava) p. 87
[14] Drápala J, Kroupa A, Smetana B, Burkovič R, Lasek S and Musiol J 2009 Study of Zn–Sn–Al
alloys for high-temperature solders. In Metal 2009 (Ostrava:Tanger, Ltd.) 380
[15] Drápala J, Musiol J, Petlák D, Burkovič R and Vodárek V 2010 Study of behavior of tin–zinc–
aluminium alloys for high–temperature applications In Metal 2010 (Ostrava:Tanger, Ltd.)
776
[16] Drápala J, Kostiuková G, Smetana B, Madaj M and Kroupa A 2015 Advanced Science,
Engineering and Medicine 7 291
[17] Villain J, Klima S, Drapala J, Smetana B, Zla S, Vodarek V, Lasek S, Petlak D, Konecna K and
Kostiukova G 2012 Experimental and theoretical study of the Al-Sn-Zn system In COST
MP0602 action - Handbook of high-temperature lead-free soldering systems Group Project
Reports (Brno: ed. A. Kroupa, COST Office) 3 243-280
16
1234567890
27th Joint Seminar Development of Materials Science in Research and Education IOP Publishing
IOP Conf. Series: Materials Science and Engineering 266 (2017) 012002 doi:10.1088/1757-899X/266/1/012002
[18] Smetana B, Drápala J, Zlá S and Kroupa A 2009 Proc. of 38
th
Int. Conf. „Calphad 2009“
(Prague) (Brno: Masaryk University and Institute of Physics of Materials AS of the Czech
Republic 175
[19] DIN/IEC Standart 768 „Method of measurement of resistivity of metallic materials“
[20] Database for solder properties with emphasis on new lead-free solders 2002 (Golden: National
Institute of Standards and Technology and Colorado School of Mines)
[21] Cook BA, Anderson IE, Harringa JL and Terpstra RL 2002 J. Electron. Mater. 31 1190
[22] Brorson G and Nylén M 1997 Examination of enviromentally adapted tin-base alloys for
replacement of tin-lead solders 5 AFR Report (Sweden) 177
[23] Sidorov VE, Drápala J, Uporov S, Sabirzyanov A, Popel P, Kurochkin A and Grushevskij K
2011 Some physical properties of Al–Sn–Zn melts In EPJ Web of Conferences 15 01022
http://www.epj-conferences.org or http://dx.doi.org/10.1051/epjconf/20111501022
[24] Drápala J, Musiol J, Petlák D, Vodárek V, Smetana B, Zlá S, Kostiuková G, Kroupa A and
Sidorov VE 2011 Structural, thermodynamics and physical properties of Zn – Sn – Al alloys.
In Metal 2011, Brno, 18.5.-20.5.2011 (Ostrava:Tanger, Ltd.) on CD ROM 1091
[25] Dušek K, Urbánek J and Drápala J 2009 Hutnické listy LXII 106
[26] Lasek S, Drápala J, Titz K and Blahetová M 2009 Hutnické listy LXII 101
[27] Lasek S, Drápala J, Blahetová M and Vrbický J 2009 Hutnické listy LXII 20