+LJK7HPSHUDWXUH7HQVLOHDQG&UHHS%HKDYLRURI/HDG)UHH6ROGHUV

Mohammad S. Alam, Jeffrey C. Suhling, Pradeep Lall

Center for Advanced Vehicle and Extreme Environment Electronics, and
Department of Mechanical Engineering
Auburn University
Auburn, AL 36849
Phone: +1-334-844-3332
FAX: +1-334-844-3124
E-Mail: [email protected]

$%675$&7 ,1752'8&7,21
The mechanical behavior of lead free solders is highly The reliability of electronics used in various devices
dependent on the testing temperature. Previous investigations depends on the environmental conditions experienced during
on mechanical characterization of conventional and doped field use. In oil and gas exploration, avionics, automotive,
lead free SAC solders have mainly emphasized stress-strain and defense applications, electronics typically experience very
and creep testing at temperatures from 25 to 125 °C. harsh environments compared to consumer electronics [1-2].
However, solders are exposed to very high temperatures from The electronic systems used in under-the-hood automotive
125-200 °C in several harsh environment applications applications are operated at temperatures over 150 °C [3-4].
including well boring, geothermal energy, and aerospace In addition, engine control modules mounted directly onto the
engines. In the current work, we have extended our previous engine experience high-temperature excursions while in
studies to explore mechanical properties for SAC305, operation.
SAC_Q, SAC_R, and Innolot solders at temperatures from The electronic modules used in oil and gas exploration
125-200 °C at a strain rate of 0.001 (sec-1). The Anand applications experience ambient temperatures above 150 °C,
constitutive model with parameters measured previously using and often up to 200 °C [5]. Wireline logging represents a
test data from 25-125 has been shown to fit the high typical application in oil and gas drilling where the electronic
temperature stress-strain curves reasonably well. In addition, modules are exposed to very high temperatures. During this
high temperature creep behavior of SAC305 was explored. application, the logging tool is lowered into the wellbore, and
Finally, the high temperature tensile properties of the above- the electronics experience extremely low or high temperatures
mentioned solders have been compared. depending on the location and type of the oil well. The
Our results show a significant degradation of mechanical reliability of electronics is an important concern here since the
properties of lead-free solders at higher temperatures. Also, a wireline logging usually lasts 2 to 6 hours. In addition,
noteworthy increase in the secondary creep strain rate has electronic systems used in commercial and defense aircraft,
been observed. Comparison of the results for different solders ground military vehicles, high-speed civil transport, and
has shown that the addition of dopants (e.g. Bi, Ni, and Sb) in supersonic aircraft, can be exposed, most often, to
the traditional SAC alloys improve their properties temperatures up to 200 °C [2]. Also, electric propulsion
significantly. systems in battle tanks and other mobile ground defense
applications can experience ambient temperatures above 200
.(<:25'6Lead-Free Solder, Aging, Stress-Strain °C [6].
Curve, Modulus, Yield Stress, Ultimate Tensile Strength, Thus, electronic products experience very harsh
Anand Model, Constitutive Relations environments, e.g., temperatures up to 200 °C, which are very
 close to the melting points of the various Sn-Ag-Cu (SAC)

V
120(1&/$785( solders used by industry. Moreover, longer exposures at such

H
Uniaxial Stress high temperatures can result in extreme aging induced
Uniaxial Strain changes to their properties [7]. In particular, significant
T Temperature degradations have been reported in ball shear strength [8],
s Internal Variable elastic modulus [9], drop reliability [10], fracture behavior
ȟ Multiplier of Stress [11], microstructure [12], creep behavior [13-17], thermal
A Pre-Exponential Factor
cycling reliability [18-24], Anand model parameters [21-22],
Q Activation Energy
R Universal Gas Constant nanoindentation joint modulus and hardness [23-25], high
ho Hardening Constant strain rate mechanical properties [26], uniaxial cyclic stress-
so Internal Variable Initial Value strain curves and fatigue life [27-28], and shear cyclic stress-
ŝ Coefficient for Deformation Saturation Value strain curves and fatigue life [29-30].
m Strain Rate Sensitivity of Stress SAC305 (96.5Sn–3.0Ag–0.5Cu) is one of the most
n Strain Rate Sensitivity of Saturation Process popular lead free solders used in surface mount assembly due
a Strain Rate Sensitivity of the Hardening Process to its relatively high melting point, attractive mechanical
 properties, and thermal cycling reliability. Addition of
 dopants such as Bi, Ni, In, Mg, Mn, Zn, La, Ce, Co, and Ti to

978-1-5090-2994-5/$31.00 ©2017 IEEE 1229 16th IEEE ITHERM Conference

the traditional SAC alloys is predicted to improve the alloys at extreme high temperatures from 125 to 200 °C is still
wettability, melting temperature, shock/drop reliability, creep unknown. Therefore, investigation on the mechanical and
properties, and microstructure [17]. For example, Bismuth microstructural behavior of these alloys at high temperatures
(Bi) can reduce the solidification temperature, increase up to 200 °C is necessary to support several harsh
strength by means of precipitation hardening, and also environment electronics applications in automotive, avionics,
facilitate the reduction of IMC (Intermetallic Compound) and oil-exploration industries, as well as for military uses.
layer thicknesses in lead free solder materials [32]. Matahir Previously, we investigated the tensile properties of water
and coworkers [33] studied the Effect of Bi on the mechanical quenched (WQ) SAC305 at temperatures up to 200 °C [47].
properties of a SAC (Sn3.5Ag0.9Cu) alloy and reported that In the current work, we have extended our previous studies to
an increase of Bi up to 2% (wt) improves the shear strength of explore mechanical properties for reflowed (RF) SAC305,
the alloy. Beyond that point, the shear strength reduced with SAC_Q, SAC_R, and Innolot solders at temperatures from
increasing Bi% (wt). This is due to the fact that higher Bi 125-200 °C (e.g. 125, 150, 175, and 200 °C). For each
content facilitates the evolution of Bi rich phase and elevated temperature, stress-strain data were measured at a
fragmentation of the IMC. Pandher, et al. [34] also reported strain rate of 0.001 sec-1. The experimental results were then
the similar phenomena. compared to the predictions of the Anand constitutive model.
Cai, et al. [17, 34] reported that addition of 0.1% Bi in In addition, high temperature creep behavior of WQ SAC305
SAC307 solder significantly decreases aging effects. Witkin was explored at two stress levels (7.5 MPa and 10 MPa).
[35] also studied the mechanical and microstructural 
properties of SAC305 and some Bi-doped alloys, and also (;3(5,0(17$/352&('85(
reported the improved properties with Bi addition. He further
concluded that Bi exists as a single phase in the Test Matrix
microstructure or goes into solid solution with Sn. It doesn’t The experimental test matrix (tensile and creep) for high
form any intermetallic compounds with other constituents, temperature characterization of SAC solders (SAC305 and
thus, reduces IMC layer thickness [36-37]. doped SAC alloys) are listed in Table 1 and 2. Testing
Dopants added to alloys in very small amounts have been conditions for the uniaxial tensile tests included four testing
shown to have significant effects on the microstructure of the temperatures of 125, 150, 175, and 200 °C and a strain rate of
alloys. Zhao, et al. [38] studied the addition of 0.02% Ni to 0.001 sec-1. For the tensile tests, samples were reflowed in a
SAC105 and reported an increase of NiCuSn IMC and as well reflow oven following a certain temperature profile used in
as a reduction of localized grain size at solder/NiAu pad electronics packaging industries. These samples are called
interfaces. In addition, Sousa, et al. [39] investigated the reflowed (RF) samples, and have coarser microstructure.
effects of adding various doping elements (i.e. Co, Fe, In, Ni, The creep tests were performed for SAC305 at two stress
Zn, and Cu) in small amounts to SAC305 BGA solder joints levels of 7.5 and 10 MPa throughout the same temperature
on Cu pads, and concluded that addition of low levels of Zn range. For the creep tests, a water quenched (WQ)
significantly improves the properties. Lee and coworkers [40] solidification profile has been used for the sample preparation,
concluded that addition of low levels of Ni and Bi to SAC and the specimens were aged for 10 days at room temperature.
alloys accelerates thermal fatigue life and reduces impact This solidification process leads to fine microstructures and
resistance. Finally, Yeung, et al. [41] reported a superior provides the upper limits of the mechanical properties for
solder joint reliability of SAC_Q (Bi-doped SAC alloy) each alloy.
during thermal cycling.
Innolot, an alloy consists of Sn, Ag, Cu, Bi, Ni, and Sb, Table 1 - Tensile Text Matrix
has been introduced by several solder vendors. This alloy
provides enhanced reliability at high temperature. Dudek, et Temperature
SAC305 SAC_Q SAC_R Innolot
al. [42] compared the creep responses of Innolot with some (°C)
125 ¥ ¥ ¥ ¥
other SAC alloys and concluded that Innolot yields lowest 150 ¥ ¥ ¥ ¥
creep strain rates at both T = 20 °C and T = 150 °C. Miric 175 ¥ ¥ ¥ ¥
[43] showed that ceramic chip capacitors made with Innolot 200 ¥ ¥ ¥ ¥
solder ensures higher thermal cycling reliability than several
other lead free solders. Tao, et al. [44] performed lap shear Table 2 - Creep Experiment Test Matrix (SAC305)
tests for Innolot specimens at temperatures up to 125 °C and
proposed a creep model to fit their data. Temperature
Stress
Prior work on mechanical characterization of lead-free Sn- (MPa)
(°C)
7.5 10
Ag-Cu (SAC) solders are restricted mostly to the operating
100 ¥ ¥
temperatures ranges from 25 to 125 °C. For example, the 125 ¥ ¥
authors [13-17, 21-22] and Zhu, et al. [45] have investigated 150 ¥ ¥
the mechanical properties of SAC305 at various strain rates 175 ¥ ¥
and temperatures up to 125 °C. Ahmed, et al. [46] measured 200 ¥ ¥
stress-strain curves for three doped SAC alloys referred to as
Ecolloy (SAC_R), CYCLOMAX (SAC_Q), and Innolot at Ahmed, et al. [46] previously determined the chemical
temperatures up to 125 °C. Mechanical behavior of these compositions of SAC_Q, SAC_R, and Innolot using Energy
Dispersive X-Ray Spectroscopy (EDX). These compositions
are tabulated in Table 3. The compositions of the traditional
SAC105 and SAC305 alloys are also listed in Table 3 for
comparison. SAC_R and SAC_Q contain Bi along with Sn,
Ag, and Cu, while Innolot is formulated with a combination of
six elements such as Sn, Ag, Cu, Bi, Ni, and Sb). The
compositions of SAC_R and SAC105 are very similar.
However, the SAC_R does not contain any measurable silver
(Ag) content, unlike other SAC alloys. For this reason,
SAC_R is regarded widely as a low cost solder alloy. SAC_Q
and Innolot contain similar amount of silver (3.41% and
3.80% respectively) to that of SAC305 (3.00%).

Table 3 - Chemical Compositions of the Solder Alloys

Figure 1 - Mechanical Test System with Solder Sample Inside
Alloy Sn Ag Cu Bi Ni Sb
a High Temperature Oven
SAC_R 96.62 0.00 0.92 2.46 0.00 0.00
SAC105 98.50 1.00 0.50 0.00 0.00 0.00
SAC_Q 92.77 3.41 0.52 3.30 0.00 0.00
For all of the listed alloys, about 8-10 uniaxial stress-strain
SAC305 96.50 3.00 0.50 0.00 0.00 0.00 tests were performed for each unique set of test conditions
Innolot 90.95 3.80 0.70 3.00 0.15 1.40 (temperature and strain rate). Raw data were extracted from
the experiments and fitted with an empirical model. Several
Uniaxial Test Sample Preparation different empirical models are available to represent the
Our sample preparation setup includes a heating system stress-strain results. However, a hyperbolic tangent empirical
with a temperature controller, a portable vacuum pump, a model having four constants, shown in equation 1, is found to
cylindrical glass crucible and rectangular glass tubes. Bulk represent the stress-strain behavior of SAC solders efficiently.

V C 1 tanh( C 2 H )  C 3 tanh( C 4 H )
SAC solder samples available as a large bars. These were cut
into small pieces. Solid materials were melted into a quartz (1)
crucible. The temperature controller controls the temperature
of the heating system based on the melting point of the This model was used to fit the experimental stress-strain
solders. A vacuum suction method allowed the molten solder curves (usually a set of 10) and an “average” experimental
to be drawn into rectangular glass tubes [13-17]. The samples stress-strain representation was generated.
were solidified using both water quenched (WQ) and reflowed
(RF) solidification profiles. Initially, all the samples were $1$1'9,6&23/$67,&02'(/
solidified with water quenched method. In this study, water The Anand viscoplastic model [48] is one of the most
quenched specimens were used for the creep tests where these popular constitutive models for solder materials. This model
specimens were aged at room temperatures for 10 days before is built into ANSYS and ABAQUS finite element codes and
testing. For tensile tests, some specimens were reflowed in a has been widely adopted to predict solder joint reliability. For
reflow oven where these were re-melted in the glass tube and example, Che, et al. [49] studied multiple constitutive models
then re-solidified in a controlled reflow profile similar to that and used Anand viscoplastic model to predict the fatigue life
used for SMT assembly. Before testing, samples were of lead-free solders accurately. Details of the Anand model
collected by breaking the glass tubes. The final test samples are explained in the literature [e.g., 21-22, 50-51]. The Anand
typically have dimensions of 80 x 3 x 0.5 mm. The gage model provides three equations: (1) stress equation, (2) flow
length of the testing specimen was about 60 mm. The test equation, and (3) evolution equation. There are nine material
samples were preserved in low temperature freezer at -40 °C parameters (constants) in the model that can be determined by
to avoid any unintentional aging effects on their properties. using the stress-strain results obtained at several different
temperatures and strain rates. Finally, a least-squares
Mechanical Testing System and Data Processing regression method is used to extract these parameters [21-22,
The stress-strain testing in this work was performed with a 49, 50-51]. Anand model parameters for SAC305 and
micro-tension/torsion thermo-mechanical test system (MT- SAC_R have previously been measured using stress-strain
200) from Wisdom Technology Inc. (Figure 1). This system data recorded from 25-125 °C [46, 50-51].
offers an axial displacement resolution of 0.1 micron. This 
test system includes a universal 6-axis load cell that can (;3(5,0(17$/5(68/76
simultaneously monitor three forces and three moments
during sample mounting and testing. An additional heating Stress-Strain Data for Various Temperatures and Strain Rates
chamber was used (Figure 1), which allowed samples to be The plots in Figure 2 illustrate the recorded high
tested up to +200 °C. temperature stress-strain curves for the SAC305, SAC_R,
SAC_Q, and Innolot alloys (RF microstructures) at a strain
rate of 0.001 sec-1. In each plot, the four different colored
 curves represent the 4 testing temperatures (T = 125, 150,
175, and 200 °C). Each curve in these plots is an “average”
H 0 . 001 sec 1
60
experimental stress-strain curve representing the fit of the SAC_Q RF o
125 C
o
150 C
empirical model in eq. (1) to the 10 recorded experimental 50 175 oC
stress-strain curves for the particular strain rate and o
200 C

Stress, V MPa
temperature. In each plot, the top curve represents the 40
average stress-strain curve at T = 125 °C, and the bottom
curve is the average stress-strain curve at T = 200 °C. These 30
results clearly show a significant reduction in the mechanical
properties with increase in temperature. For example, the 20
yield stress and ultimate tensile strength of these alloy
typically drop by 50-60% between T = 125 °C and T = 200 10
°C.
0

Strain, H
0.000 0.005 0.010 0.015 0.020

H 0 .001 sec  1
60
SAC305 RF o
125 C
150 oC (c) SAC_Q
50 o
175 C
o
200 C
Stress, V MPa

40

H 0 . 001 sec 1
60
Innolot RF o
125 C
30 o
150 C
50 175 oC
o
200 C
20
Stress, V MPa 40

10
30

0
20

Strain, H
0.000 0.005 0.010 0.015 0.020

10
(a) SAC305
0

Strain, H
0.000 0.005 0.010 0.015 0.020

0 .001 sec 1
60
İ SAC_R RF o
125 C
o
150 C (d) Innolot
50 175 oC
o
200 C
Figure 2 - Experimental Stress-Strain Curves
Stress, V MPa

40
(RF, No Aging)
30
A comparison of the uniaxial tensile test results for the
four different solder alloys at 125 °C is shown in Figure 3. It
20
can be seen that SAC_Q possesses the best properties whereas
SAC305 shows to have lowest strength which is about 50% of
10
that of SAC_Q. The strength of Innolot is also significantly
higher than SAC305 and very close to SAC_Q. The strength
0
of SAC_R is higher than SAC305, even though SAC_R does

Strain, H
0.000 0.005 0.010 0.015 0.020
not contain any Ag. Analogous results were also obtained at
other testing temperatures.
(b) SAC_R
 H 0 .001 sec 1 H 0 .001 sec 1
60 60
T = 125 °C SAC_R RF SAC_Q

Effective Modulus, E (GPa)

SAC_Q Innolot
50 Innolot 50 SAC305
SAC305
Stress, V MPa

40 40

30 30

20 20

10 10

0 0
0.000 0.005 0.010 0.015 0.020

Strain, H
100 120 140 160 180 200 220
o
Temperature, T ( C)
(a) Effective Elastic Modulus
Figure 3 - Comparison of Stress-Strain Curves
(T = 125 °C, Strain Rate = 0.001 sec-1)

H 0 .001 sec 1
60

Ultimate Strength, UTS (MPa)

The mechanical properties for the SAC solder alloys have RF SAC_Q
Innolot
been extracted from the average stress-strain data. Plots of 50 SAC305
the initial effective elastic modulus and ultimate tensile
strength (UTS) as a function of temperature, for example, are 40
presented in Figures 4a and 4b respectively. All other
properties are tabulated in Table 4. These plots indicate a 30
decreasing trend for the properties as expected and these
variations are nearly linear over the 125-200 °C temperature 20
range.
10
Correlation of the Experimental Results with Anand Model
Predictions
0
The experimental results have been compared with the 100 120 140 160 180 200 220
predictions of the Anand model, and example comparisons for o
Temperature, T ( C)
SAC305 and SAC_R are presented in this paper. Anand
model parameters for SAC305 and SAC_R have previously (b) Ultimate Tensile Strength
been determined using stress-strain data recorded from T =
25-125 °C [46, 50-51]. These values are listed in Table 5. Figure 4 - Mechanical Property Variation with Temperature
These parameters were used to predict stress-strain behavior (RF, No Aging)
of SAC solder alloys tasted at 125-200 °C. Figure 5
illustrates both experimental results and Anand model Table 4 - Material Properties of the Solder Alloys
predicted results in the same plot. In these plots, solid curves
represent experimental stress-strain data and the dashed Properties
SAC
curves are the corresponding model predictions. For all the Temperature (°C) E UTS YS
Alloys
SAC alloys, fairly good correlations between experimental (GPa) (MPa) (MPa)
results and Anand model predicted results were observed at 125 28.1 23.5 21.9
150 22.9 19.6 18.5
all the temperatures. Therefore, Anand model can predict the SAC305
175 13.9 15.5 14.4
stress-strain behavior of these four SAC alloys accurately 200 11.2 12.6 11.7
even at extreme high temperatures near the melting point of 125 29.7 24.6 23.3
the material. 150 23.7 23.3 22.0
SAC_R
175 15.9 19.7 18.4
200 13.8 17.8 17.4
125 34.5 42.7 38.8
150 27.3 35.5 34.4
SAC_Q
175 20.7 29.5 28.4
200 13.9 20.4 20.0
125 29.9 40.3 36.9
150 23.5 32.1 30.9
Innolot
175 20.1 27.7 27.2
200 14.1 24.8 24.0
 Table 5 - Anand Model Parameters for SAC305 and Figure 6. When a uniaxial constant load is applied to a lead
SAC_R Solders (RF) free solder, the creep response begins with a quick transition
to the initial “elastic” strain level. And, the remaining
Par. Anand responses can be expressed by three regions of creep, i.e.,
Units SAC305 SAC_R
No. Parameter primary, secondary, and tertiary creep. The secondary region
1 so MPa 21 30.59
represents a nearly constant slope over time. Due to its
2 Q/R 1/K 9320 11100 constant value the slope is called “steady state” secondary
-1
3 A sec 3501 2000 creep strain rate. This secondary creep strain rate is one of the
4 ȟ Dimensionless 4 6 key material parameters for solders to predict solder joint
5 m Dimensionless 0.25 0.12 reliability.
6 ho MPa 180,000 141,180
7 ŝ MPa 30.2 61.2
8 n Dimensionless 0.0100 0.0009
9 a Dimensionless 1.78 1.63

H 0 .001 sec 1
60
SAC305 o
125 C, Experimental
150 oC, Experimental
50 175 oC, Experimental
200 oC, Experimental
Stress, V(MPa)

125 oC, Model Prediction

40 150 oC, Model Prediction
175 oC, Model Prediction
200 oC, Model Prediction
30

20

10 Figure 6 - Typical SAC Solder Creep Curve

0 Creep Data Processing

A four parameter Burger’s (spring-dashpot) model has
Strain, H
0.000 0.005 0.010 0.015 0.020
been used to fit the experimental creep data. This model
provides a functional representation of the data where the
(a) SAC305
secondary creep rate is extracted from. The Burger’s (spring-
dashpot) model is represented by the following equation:

60
H 0 .001 sec 1 SAC_R o
125 C, Experimental
İ C 0  C1 t  C 2 (1  e  C 3 t ) (2)
150 oC, Experimental
50 175 oC, Experimental
200 oC, Experimental where C0, C1, C2 and C3 are fitting constants. Note that
constant C1 represents the “steady state” creep strain rate. 
Stress, V(MPa)

125 oC, Model Prediction

40 150 oC, Model Prediction
175 oC, Model Prediction
200 oC, Model Prediction &5((37(675(68/76
30

fixed stress levels of V = 7.5 and 10 MPa. The five colored

Figures 7 illustrates creep curves of SAC305 (WQ) under
20
curves in each plot represent five different testing
temperatures (T = 100, 125, 150, 175, and 200 °C). In each
10
plot, we can see variations of secondary creep strain rate as
temperature changes. However, the variations appear to be
0
more significant at higher stress levels. For example, in
Strain, H
0.000 0.005 0.010 0.015 0.020
Figure 7b, the creep rate increases dramatically at higher
temperatures of 150-200 °C. At T = 200 °C, the samples
(b) SAC_R experiences tertiary creep when the load was applied for about
2 hours. Increased dislocation movements occur for the
Figure 5 - Correlation of the Anand Model Predictions with higher creep rates at higher stress levels. At higher
Experimental Stress-Strain Data temperatures, the dislocation movements occur more easily,
resulting in more creep for the same applied loading. The
Typical Creep Test Data average secondary creep strain rates are tabulated in Table 4.
A typical SAC solder creep curve obtained with the strain Also, the variations of creep rate with temperature are shown
vs. time response for a constant applied stress is shown in in Figure 8.
 Table 6 - Average Secondary Creep Strain Rates
0.006
ı = 7.5 MPa SAC305 Secondary Strain Rates
Temperature Stress (x 10-8 sec-1)
0.005 (°C) (MPa)
SAC305
0.004 7.5 1.3
100
Strain, H

10 1.8
7.5 2.3
0.003 125
10 3.1
7.5 3.9
150
0.002 10 15.5
100 oC 7.5 7.5
125 oC 175
10 61.0
0.001 150 oC
o
175 C 7.5 18.5
200
200 oC 10 114.0
0.000 
0 1000 2000 3000 4000 5000 6000 7000 8000 6800$5<$1'&21&/86,216
Time, t (Sec) In this investigation, we have extended our previous
(a) 7.5 MPa studies to explore the mechanical properties of SAC305,
SAC_R, SAC_Q, and Innolot solders at temperatures from T
= 125-200 °C (e.g. 125, 150, 175, and 200 °C). At each
elevated temperature, stress-strain data were measured at a
0.006
ı = 10 MPa SAC305 strain rate of 0.001 sec-1 for all of the alloys. The variations
of mechanical properties such as effective elastic modulus,
0.005
yield stress, and ultimate tensile strength (UTS) with
temperature were extracted. The stress-strain results of these
0.004
alloys were compared to the predictions of the Anand
Strain, H

constitutive model. Also, high temperature creep tests were

0.003
performed for the water-quenched SAC305 samples that were
aged at room temperature for 10 days.
0.002
100 oC Our results show that mechanical properties of the
125 oC abovementioned solders degrade significantly at higher
0.001 150 oC
175 oC temperatures. For example, the yield stress and ultimate
o
200 C tensile strength for these alloys typically drop by 40-50%
0.000 from T = 125 °C to T = 200 °C. Similar behavior is expected
0 1000 2000 3000 4000 5000 6000 7000 8000
at other strain rates. Correlations between the experimental
Time, t (Sec) results and the Anand model predictions suggest that Anand
(b) 10 MPa model can represent the experimental data relatively
accurately over a wide range of temperatures and strain rates.
Figure 7 - Average Creep Curves The creep curves for SAC305 show a significant increase of
secondary creep strain rates at higher temperatures.
Additional testing of the high temperature creep behavior of
10-5 the doped alloys SAC_R, SAC_Q, and Innolot is underway.
SAC305 7.5 MPa
10 MPa
$&.12:/('*0(176
This research was supported by the NSF Center for
Strain Rate, H (sec-1)

10-6
Advanced Vehicle and Extreme Environment Electronics
(CAVE3).
•

10-7
5()(5(1&(6
1. Ganesan, S., Pecht, M., /HDG)UHH (OHFWURQLFV, -RKQ
:LOH\DQG6RQV, 2006.
10-8
2. McCluskey, P., Grzybowski, R., Podlesak, T., +LJK
7HPSHUDWXUH(OHFWURQLFV, CRC Press, 1997.
3. Hattori, M., “Needs and Applications of High-
10-9
100 120 140 160 180 200 220
Temperature LSIs for Automotive Electronic Systems,”
o 3URFHHGLQJV RI WKH 7KLUG (XURSHDQ &RQIHUHQFH RQ +LJK
Temperature, T ( C)
7HPSHUDWXUH (OHFWURQLFV +,7(1 , Berlin, Germany, pp.
37-43, 1999.
Figure 8 – Comparison of Secondary Creep Strain Rates 4. Johnson, R., Evans, J., Jacobsen, P., Thompson, J.,
Christopher, M., “The Changing Automotive
 Environment: High-Temperature Electronics,” ,((( Effects Using Doped SAC Alloys,” 3URFHHGLQJV RI WKH
7UDQVDFWLRQV RQ (OHFWURQLFV 3DFNDJLQJ 0DQXIDFWXULQJ, WK(OHFWURQLF&RPSRQHQWVDQG7HFKQRORJ\&RQIHUHQFH,
Vol. 27(3), pp. 164-176, 2004. pp. 1493-1511, 2010.
5. Parmentier, B., Vermesan, O., Beneteau, L., “Design of 18. Zhang, J., Hai, Z., Thirugnanasambandam, S., Evans, J.
High Temperature Electronics for Well Logging L., Bozack, M. J., Sesek, R., Zhang, Y., Suhling, J. C.,
Applications,” 3URFHHGLQJV RI WKH ,QWHUQDWLRQDO “Correlation of Aging Effects on Creep Rate and
&RQIHUHQFH RQ +LJK 7HPSHUDWXUH (OHFWURQLFV +,7(1 , Reliability in Lead Free Solder Joints,” 607$ -RXUQDO,
Oxford, UK, July 8-11, 2003. Vol. 25(3), pp. 19-28, 2012.
6. Rolfe, R., Brown, H., Nash, S., “Defense Electronics 19. Zhang, J., Hai, Z., Thirugnanasambandam, S., Evans, J.
Product Reliability Requirements,” $'$,QVWLWXWH L., Bozack, M. J., Zhang, Y., Suhling, J. C., “Thermal
IRU'HIHQVH$QDO\VHV, Alexandria, VA, 1997. Aging Effects on Thermal Cycling Reliability of Lead-
7. Ma, H., Suhling, J. C., “A Review of Mechanical Free Fine Pitch Packages,” ,((( 7UDQVDFWLRQV RQ
Properties of Lead-Free Solders for Electronic &RPSRQHQWVDQG3DFNDJLQJ7HFKQRORJLHV, Vol. 3(8), pp.
Packaging,” -RXUQDO RI 0DWHULDOV 6FLHQFH, Vol. 44, pp. 1348-1357, 2013.
1141-1158, 2009. 20. Hai, Z., Zhang, J., Shen, C., Evans, J. L., Bozack, M. J.,
8. Chou, G. J. S., “Microstructure Evolution of SnPb and Basit, M. M., Suhling, J. C., “Reliability Comparison of
SnAgCu BGA Solder Joints during Thermal Aging,” Aged SAC Fine-Pitch Ball Grid Array Packages Versus
3URFHHGLQJV RI WKH WK 6\PSRVLXP RQ $GYDQFHG Surface Finishes,” ,((( 7UDQVDFWLRQ RQ &RPSRQHQWV
3DFNDJLQJ0DWHULDOV, pp. 39-46, 2002. 3DFNDJLQJ DQG 0DQXIDFWXULQJ 7HFKQRORJ\, Vol. 5(6),
9. Hasegawa, K., Noudou, T., Takahashi, A., Nakaso, A., pp. 828-837, 2015.
“Thermal Aging Reliability of Solder Ball Joint for 21. Motalab, M., Cai, Z., Suhling, J. C., Zhang, J., Evans, J.
Semiconductor Package Substrate,” 3URFHHGLQJV RI WKH L., Bozack, M. J., Lall, P., “Improved Predictions of Lead
607$,QWHUQDWLRQDO, pp.1-8, 2001. Free Solder Joint Reliability That Include Aging Effects,”
10. Chiu, T. C., Zeng, K., Stierman, R., Edwards, D., Ano, 3URFHHGLQJV RI WKH QG ,((( (OHFWURQLF &RPSRQHQWV
K., “Effect of Thermal Aging on Board Level Drop DQG 7HFKQRORJ\ &RQIHUHQFH, pp. 513-531, San Diego,
Reliability for Pb-Free BGA Packages,” 3URFHHGLQJV RI CA, May 30 - June 1, 2012.
WKH WK (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ 22. Motalab, M., Cai, Z., Suhling, J. C., Zhang, J., Evans, J.
&RQIHUHQFH, pp. 1256-1262, 2004. L., Bozack, M. J., Lall, P., “Correlation of Reliability
11. Ding, Y., Wang, C., Li, M., Bang, H. S., “Aging Effects Models Including Aging Effects with Thermal Cycling
on Fracture Behavior of 63Sn37Pb Eutectic Solder during Reliability Data,” 3URFHHGLQJV RI WKH UG ,(((
Tensile Tests under the SEM,” 0DWHULDOV 6FLHQFH DQG (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp.
(QJLQHHULQJ, Vol. A384, 314-323, 2004. 986-1004, Las Vegas, NV, May 28-31, 2013.
12. Pang, J. H. L., Low, T. H., Xiong, B. S., Xu, L., Neo, C. 23. Hasnine, M., Mustafa, M., Suhling, J. C., Prorok, B. C.,
C., “Thermal Cycling Aging Effects on Sn–Ag–Cu Bozack, M. J., Lall, P., “Characterization of Aging
Solder Joint Microstructure IMC and Strength,” 7KLQ Effects in Lead Free Solder Joints Using
6ROLG)LOPV, Vol. 462-463, pp. 370-375, 2004. Nanoindentation,” 3URFHHGLQJV RI WKH UG ,(((
13. Ma, H., Suhling, J. C., Lall P., Bozack, M. J., “Reliability (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp.
of the Aging Lead-Free Solder Joint,” 3URFHHGLQJRIWKH 166-178, Las Vegas, NV, May 28-31, 2013.
WK(OHFWURQLF&RPSRQHQWVDQG7HFKQRORJ\&RQIHUHQFH, 24. Hasnine, M., Suhling, J. C., Prorok, B. C., Bozack, M. J.,
San Diego, California, pp. 849-864, 2006 Lall, P., “Exploration of Aging Induced Evolution of
14. Ma, H., Suhling, J. C., Zhang, Y., Lall, P., and Bozack, Solder Joints Using Nanoindentation and
M. J., “The Influence of Elevated Temperature Aging on Microdiffraction,” 3URFHHGLQJV RI WKH WK ,(((
Reliability of Lead Free Solder Joints,” 3URFHHGLQJV RI (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp.
WKH WK ,((( (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ 379-394, Orlando, FL, May 28-30, 2014.
&RQIHUHQFH, pp. 653-668, Reno, NV, May 29-June 1, 25. Hasnine, M., Suhling, J. C., Prorok, B. C., Bozack, M. J.,
2007. Lall, P., “Nanomechanical Characterization of SAC
15. Zhang, Y., Cai, Z., Suhling, J. C., Lall, P., Bozack, M. J., Solder Joints - Reduction of Aging Effects Using
“The Effects of Aging Temperature on SAC Solder Joint Microalloy Additions,” 3URFHHGLQJV RI WKH WK ,(((
Material Behavior and Reliability,” 3URFHHGLQJV RI WKH (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp.
WK ,((( (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ 1574-1585, San Diego, CA, May 27-29, 2015.
&RQIHUHQFH, pp. 99-112, Orlando, FL, May 27-30, 2008. 26. Lall, P., Shantaram, S., Suhling, J., Locker, D., “Effect of
16. Zhang, Y., Cai, Z., Suhling, J. C., Lall, P., Bozack, M. J., Aging on the High Strain Rate Mechanical Properties of
“The Effects of SAC Alloy Composition on Aging SAC105 and SAC305 Leadfree Alloys,” 3URFHHGLQJV RI
Resistance and Reliability,” 3URFHHGLQJV RI WKH WK WKH UG ,((( (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\
,((( (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp. 1277-1293, Las Vegas, NV, May 28-31,
&RQIHUHQFH, pp. 370-389, San Diego, CA, May 27-29, 2013.
2009. 27. Mustafa M., Cai Z., Suhling J. C., Lall P., “The Effects of
17. Cai, Z., Zhang, Y., Suhling, J. C., Lall, P., Johnson, R. Aging on the Cyclic Stress-Strain Behavior and
W., Bozack, M. J., “Reduction of Lead Free Solder Aging Hysteresis Loop Evolution of Lead Free Solders,”
 3URFHHGLQJV RI WKH VW (OHFWURQLF &RPSRQHQWV DQG Pad Structures,” 3URFHHGLQJ RI WKH WK (OHFWURQLF
7HFKQRORJ\&RQIHUHQFH, pp. 927-939, 2011. &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, San Diego,
28. Mustafa, M., Roberts, J. C, Suhling, J. C, Lall, P., “The CA, pp. 1454-1461, 2006.
Effects of Aging on The Fatigue Life of Lead Free 40. Lee, J. H., Kumar, S., Kim, H. J., Lee, Y. W. and Moon,
Solders,” 3URFHHGLQJV RI WKH WK ,((( (OHFWURQLF J. T., “High Thermo-Mechanical Fatigue and Drop
&RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp. 666-683, Impact Resistant Ni-Bi Doped Lead Free Solder,”
Orlando, FL, May 28-30, 2014. 3URFHHGLQJV RI WKH WK (OHFWURQLF &RPSRQHQWV DQG
29. Mustafa, M., Cai, Z., Roberts, J. R., Suhling, J. C., Lall, 7HFKQRORJ\ &RQIHUHQFH, pp. 712-716, Orlando, FL, May
P., “Evolution of the Tension/Compression and Shear 27-30, 2014.
Cyclic Stress-Strain Behavior of Lead-Free Solder 41. Yeung, T-S., Sze, H., Tan, K., Sandhu, J., Neo, C-W.,
Subjected to Isothermal Aging,” 3URFHHGLQJV RI ,7KHUP Law, E., “Material Characterization of a Novel Lead-Free
, pp. 765-780, San Diego, CA, May 30 - June 1, Solder Material-SACQ,” 3URFHHGLQJV RI WKH WK
2012. (OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp.
30. Mustafa M., Suhling J. C., Lall P., “Experimental 518-522, Orlando, FL, May 27-30, 2014.
Determination of Fatigue Behavior of Lead Free Solder 42. Dudek, R., Faust, W. Ratchev, R., Roellig, M., Albrecht,
Joints in Microelectronic Packaging Subjected to H-J., Michel, B., “Thermal Test and Field Cycling
Isothermal Aging,” 0LFURHOHFWURQLFV5HOLDELOLW\, Vol. 56, Induced Degradation and its FE-Based Prediction for
pp. 136-147, 2016. Different SAC Solders,” 3URFHHGLQJV RI ,7KHUP ,
31. Huang, M. L. and Wang, L., “Effects of Cu, Bi and In on pp. 668-675, 2008.
Microstructure and Tensile Properties of SnAg (Cu, Bi, 43. Miric, A. Z., “New Developments in High Temperature,
In) Solders,” 0HWDOOXUJLFDO DQG 0DWHULDOV 7UDQVDFWLRQV High Performance Lead Free Solder Alloys,”
$, Vol. 36(6), pp. l439-1446, 2005. 3URFHHGLQJV RI WKH 607$ ,QWHUQDWLRQDO &RQIHUHQFH,
32. Matahir, M., Chin, L.T., Tan, K. S., Olofinjana, A. O., Orlando, FL, October 24-28, 2010.
“Mechanical Strength and its Variability in Bi-modified 44. Tao, Q. B., Benabou, L., Tan, K. L., Morelle, J. M.; and
Sn-Ag-Cu Solder Alloy,” -RXUQDO RI $FKLHYHPHQW LQ Ben Ouezdou, F., “Creep Behavior of Innolot solder
0DWHULDOV DQG 0DQXIDFWXULQJ (QJLQHHULQJ, Vol. 46, pp. Alloy Using Small Lap-Shear Specimens,” 3URFHHGLQJV
50-56, 2011. RI WKH WK (OHFWURQLFV 3DFNDJLQJ 7HFKQRORJ\
33. Pandher, R. S., Lewis, B. G., Vangaveti, R., and Singh, &RQIHUHQFH, Singapore, Dec 2-4, 2015.
B., “Drop Shock Reliability of Lead-Free Alloys - Effect 45. Zhu, F., Zhang, H., Guan, R., and Liu, S., “Effects of
of Micro-Additives,” 3URFHHGLQJV RI WKH WK ,((( Temperature and Strain Rate on Mechanical Property of
(OHFWURQLF &RPSRQHQWV DQG 7HFKQRORJ\ &RQIHUHQFH, pp. Sn96.5Ag3Cu0.5,” -RXUQDO RI $OOR\V DQG &RPSRXQGV,
669-676, Reno, NV, May 29-June 1, 2007. Vol. 438(1), pp.100-105, 2007
34. Cai, Z., Suhling, J. C., Lall, P., Bozack, M. J., 46. Ahmed, S., Basit, M. M., Suhling, J. C., Lall, P.,
“Mitigation of Lead Free Solder Aging Effects using “Characterization of Doped SAC Solder Materials and
Doped SAC-X Alloys,” 3URFHHGLQJVRI,7KHUP, pp. Determination of Anand Parameters,” 3URFHHGLQJV RI
896-909, San Diego, CA, May 30 - June 1, 2012. ,QWHU3$&., pp. 1-14, San Francisco, CA, July 6-9,
35. Witkin, D. B., “Influence of Microstructure on 2015.
Mechanical Behavior of Bi-Containing Pb-Free Solders,” 47. Alam, M. S., Basit, M. M., Suhling, J. C., Lall, P.,
3URFHHGLQJV RI ,3& $3(; (;32 &RQIHUHQFH DQG “Mechanical Characterization of SAC305 Lead Free Sold
([KLELWLRQ, pp. 540-547, San Diego, CA, Feb 19-21, er at High Temperatures,” 3URFHHGLQJV RI ,7KHUP 
2013. pp. 755-760, Las Vegas, NV, May 31- June 3, 2016.
36. Li, G., Shi, X., “Effects of Bismuth on Growth of 48. Anand, L., “Constitutive Equations for the Rate-
Intermetallic Compounds in Sn-Ag-Cu Pb-Free Solder Dependent Deformation of Metals at Elevated
Joints,” 7UDQVDFWLRQ RI 1RQIHUURXV 0HWDOV 6RFLHW\ RI Temperatures,” -RXUQDO RI (QJLQHHULQJ 0DWHULDOV DQG
&KLQD, Vol. 16, pp. 739-743, 2006. 7HFKQRORJ\, Vol. 104(1), pp. 12-17, 1982.
37. Hodulova, E., Palcut, M., Lechovic, E., Simekova, B., 49. Che, F., Pang, H., Zhu, W., Sun, W., and Sun, A.,
Ulrich, K., “Kinetics of Intermetallic Phase Formation at “Modeling Constitutive Model Effect on Reliability of
the Interface of Sn–Ag–Cu–X (X = Bi, In) Solders with Lead-Free Solder Joints,” 3URFHHGLQJV RI WKH WK
Cu Substrate,” -RXUQDO RI $OOR\V DQG &RPSRXQGV, Vol. ,QWHUQDWLRQDO &RQIHUHQFH RQ (OHFWURQLFV 3DFNDJLQJ
509, pp. 7052-7059, 2011. 7HFKQRORJ\pp. 1-6, 2006.
38. Zhao, Z., Wang, L., Xie, X., Wang, Q., Lee, J., “The 50. Basit, M. M., Motalab, M., Suhling, J. C., Lall, P., “The
Influence of Low Level Doping of Ni on the Effects of Aging on the Anand Viscoplastic Constitutive
Microstructure and Reliability of SAC Solder Joint,” Model for SAC305 Solder,” 3URFHHGLQJVRI,7KHUP,
3URFHHGLQJV RI WK ,QWHUQDWLRQDO &RQIHUHQFH RQ pp. 112-126, Orlando, FL, May 27-30, 2014.
(OHFWURQLF 3DFNDJLQJ 7HFKQRORJ\  +LJK 'HQVLW\ 51. Basit, M. M., Motalab, M., Suhling, J. C., Lall, P.,
3DFNDJLQJ, pp. 1-5, Shanghai, China, 2008. “Viscoplastic Constitutive Model for Lead-Free Solder
39. Sousa, I., Henderson, D. W., Patry, L., Kang, S. K., Shih, Including Effects of Silver Content, Solidification Profile,
D. Y., “The Influence of Low Level Doping on The and Severe Aging,” 3URFHHGLQJVRI,QWHU3$&., pp.
Thermal Evolution of SAC Alloy Solder Joints with Cu 1-18, San Francisco, CA, July 6-9, 2015.