-
Formation of Intermetallics in Lead-Free Systems.
M. Braunovic MB Interface, 5975 Place de lAuthion, Suite 503,
Montral. QC, Canada, H1M 2W3
E-mail: [email protected]
D. Gagnon IREQ, Institut de Recherche dHydro Qubec, Varennes,
QC, Canada, J3X 1S1
cole de technologie Suprieure 1100 Notre Dame Ouest Montral
(Qubec) E-mail: [email protected]
Abstract Formation and growth of intermetallics is one of the
most important problem in the search for reliable lead-free alloys.
Since the majority of the commercially available lead-free alloys
contain high tin concentration, there is a long-term reliability
concern due to tendency of tin to form intermetallics. A prolonged
exposure to higher temperatures of lead-free alloys results in the
continuous growth of brittle intermetallic layer that is prone to
facture thus leading to mechanical and electrical failure of joint.
In this work, the formation and growth of intermetallics between
lead-free alloys and contact materials such as copper, tin-plated
and silver-plated copper materials were studied. For this purpose
bimetallic couples formed between commercially available lead free
alloys and selected contact materials were subjected to diffusion
annealing using thermal gradients and heating by electrical
current. Following diffusion annealing, the contact interfaces were
subjected to a detailed metallographic, Scanning Electron
Microscope (SEM), Energy Dispersive X-Ray (EDX) analyses. In
addition, electrical resistance were used for electrical
characterization of the intermetallics formed at the contact
interfaces. The results of the work enabled to make a comparative
assessment of the susceptibility of the lead free alloys to the
formation of intermetallics
I. INTRODUCTION
Increasing global concern about the environment and awareness of
lead-free activities has prompted users and suppliers to
investigate lead-free solder systems in detail. The European
community has a proposed Waste in Electronic and Electric Equipment
(WEEE) Directive that restricts the intentional use of lead in
electronic products after January 1, 2006. Although this Directive
has not been approved yet by the European Parliament, competitive
pressures in consumer electronics and concerns about the lead in
discarded electronic products, prompted considerable movement to
reduce or completely eliminate the use of lead in products. In
Japan, the Japan Electronic Industry Development Association
(JEIDA) published a road map to achieve lead replacement by
2005.
In the U.S., however, there is no pending legislation to ban
lead-bearing alloys. Nevertheless, search for the lead-free
alternatives has been given considerable attention as demonstrated
the development work on lead-free solders that has been launched by
a number of organizations and institutions. As a result of
concentrated efforts, alternates to lead-bearing solder alloy have
been identified [1-3].
However, despite considerable advances made so far in our
understanding of the behavior of the lead-free alloys, the
complexity of formation and properties of these alternatives
requires further work, especially in the area of the effect of
alloying elements on the aging behaviour and temperature cycle
conditions on the component and board level reliability.
Since most of the developed and commercially available lead-free
alternatives are tin-base alloys with melting point in the range of
200-240C, there is a possibility that the temperatures used during
soldering processes may lead to the formation of thick
intermetallic layers. The intermetallics formed are brittle and may
compromise the mechanical integrity of a joint, leading to failure
at unacceptably low mechanical stresses [4, 5]. Hence, it is
important to determine as to whether lead-free solder joints are
susceptible to the formation and growth of intermetallic compounds
and prone to fracture.
The main objective of this work is to investigate the
susceptibility of lead-free alloys to the formation of
intermetallics when in contact with copper base and their impact on
the electrical integrity of a joint. For this purpose a number of
commercially available lead-free alloys were used.
The work is carried out as a preliminary to a more general
investigation of the effect of intermetallics formation on the
quality and reliability of electrical connections involving
lead-free alloys.
II EXPERIMENTAL DETAILS
A. Samples and Surface Preparation
All test samples used in this work were 200 x 6 x 25 mm busbars,
cut from the ETP-grade copper bars. The size of the busbars used
corresponds to the terminals commonly used for 2/0 size
conductor.
Prior to coating, the copper busbars were first cleaned using a
commercial degreaser/defluxer solvent Ayarel 2200. Following
cleaning and drying for 2 hours, the busbars were dipped in RMA2002
commercial liquid flux, placed in an environmental chamber and
heated to a temperature corresponding to that of molten bath. The
ends of busbars,30 mm long, were then dipped in the molten alloy
bath at 50C higher than the melting point of the alloys used, held
for 30s,
267
9.4
0-7803-7862-8/04/$20.00 (C) IEEE.
-
removed from the bath and cooled down at room temperature.
Depending on the lead-free alloy used, the thicknesses of the
solder coatings were in the range 20- 50 m.
The composition, electrical and mechanical properties of five
different lead-free solder alloys used in this work are shown in
Table.1.
Table 1 Selected properties of the lead-free alloys used[1].
Alloy Melting
point (C) Microhardness
HV (10 grf)
Electrical resistivity (cm)
Sn-Ag2.5/Cu.8-Sb0.5 217 17.6 13.8 Sn-Ag3.8/4-Cu0.5-0.7 217 15.2
12.4 Sn-Cu0.7 227 12.9 11.4 Sn 48-Bi58 138 17.2 30
B. Diffusion Annealing
Diffusion annealing was realized by heating the samples with a
DC electrical current and in an environmental chamber. The busbar
joints were tightened to 50Nm force torque, which corresponds to an
initial contact force of 16 kN. Each joint comprises a combination
of steel bolt, two disc-spring (Belleville) and two thick flat
washers placed on each side of the joint. This joint combination
minimizes the effect stress relaxation and thermoelastic ratcheting
[6,7].
The busbars intended for diffusion annealing with an electrical
current were assembled in cross-rod configuration and connected to
a current source. A calibrated shunt mounted in the circuit loop
monitored the current level required to bring the joint temperature
to pre-set values that is 100 and 150C. Since the joint resistances
were slightly different, the control system, HP 3852A
microprocessor, computed the maximum temperature in the loop and
adjusted the source current to maintain either 100 or 150C. This
procedure provided conditions for diffusion annealing where the
rate of formation of the intermetallic phases is relatively
rapid.
The joint temperature was measured by chromel-alumel
thermocouples (Type K) inserted in holes drilled in the busbars
about 5mm from the contact interfaces. At any given time, the
difference between the maximum and minimum temperature in the loop
was within 10 - 15C. Figure 1 shows a schematic of the set up used
for diffusion annealing by an electrical current.
Two identical series of samples were produced: one series was
heated electrically while the other in an environmental chamber at
the same temperatures i.e. 100 and 150C. Since the selected
lead-free alloys have different melting temperatures, diffusion
annealing was carried out on two sets of coated busbars.
The busbar coated with alloys Sn-Ag2.5-Cu0.8-Sb0.5,
Sn-Ag3.8-4-Cu0.5-0.7 and Sn-Cu0.7 with melting points around 215C
were subjected to diffusion annealing at 150C for 25 days. The
busbars coated with Sn42-Bi58 alloys were exposed to diffusion
annealing at 100 C also for 25 days. A control
group of samples, not subjected to heat treatment, was used to
compare the final state to the initial state.
After diffusion annealing, the samples were sectioned and
prepared for metallographic examination by optical and scanning
electron microscope (SEM) and X-ray diffraction analysis (EDX) for
composition analysis.
1
2
3 4
6
7
T1
T2
T3 T4
T5
T6
T7
SHUNT DCSOURCE
I
T
HP UNIXHP 3852DATA LOGGER
V7
V6
V3 V4
5
V2 V5
V1
V
Figure 1: Schematic of experimental current cycling set up
C. Four-Point Probe Resistance Measurements
Following diffusion annealing by electrical current and thermal
gradient, a four-point probe DMO 350 microohmetre, with a precision
of 0.01 , was used to measure the resistance changes developed
between the plating and the copper base within the volume sampled
by current penetration. The probe spacing was 1 mm. The
microohmetre operates on current pulses of 10 A and 17 ms
duration.
The advantage of the four-probe resistance measuring technique
lies in its ability to detect minute structural change in the
material within a very small volume penetrated by current. The size
of the current penetration is determined by the probe spacing.
Typically, current penetration is limited to 1.5 times the probe
spacing [8]. This technique has been successfully used for studying
semiconductor materials as well as to monitor the case-,
precipitation-, and strain-hardening processes in metallic systems
[9].
Another advantage of four-point probe resistance measurement
method is the possibility to correctly measure the resistance of
the sample without any interference from the contact resistance at
the probe contacts. This is because no current flows through the
inner pair of contacts thus no voltage drops is generated at the
probe contacts. The resistance changes in the selected zones were
sufficiently large to be measured between the potential probe
points. A simplified
2680-7803-7862-8/04/$20.00 (C) IEEE.
-
schematic of the assembly used for the resistance measurements
is shown in Fig. 1.
DMO-350 MICROHMETRE
Fig. 1 Simplified schematic of four-point probe resistance
measurements assembly.
When the probe spacing (s) is considerably smaller than the
thickness (t) of the busbar, the resistivity ( ) of the volume
sampled by the four-point probe is given as
= 2 s Rm (1) where s is spacing between the probes ( 1mm ) and
Rm is the resistance measured between the potential probes. Since
the volume sampled by the four-point probes encompasses both bulk
and the plating materials measures, the resistivity measured
comprises the contributions from the plating and the bulk (copper
busbars).
III RESULTS A Four-Point Probe Resistance Measurements
The results of four-point probe resistance measurements are
shown in Table 2. The resistance data were obtained by averaging
the readings from ten different measurements made on the plated
surfaces at random locations. The data are shown with standard
deviation and 95% confidence limit.
Also shown in Table 2 are the resistance measurements made on
unplated sections of the busbars and some other materials such
aluminum alloy 6061, nickel an silver. These measurements were used
as a reference for the reproducibility and accuracy of the
four-point probe measurement technique. The resistivity values for
these materials derived from these measurements agree with the ones
generally attributed to these type of materials.
The characteristic feature of the resistance data is that
platings exert an appreciable effect on the resistance. The effect
appears to be more pronounced in the diffusion annealed than in
non-annealed.
Different platings in their non-annealed (initial) state,
increase the resistance by 2-3% in respect to that of non-plated
sections of the copper busbars. It is also apparent that different
platings increased the resistance by more or less the same amount.
In other words, this implies that the compositions of the platings
showed no appreciable influence on the resistance.
Table 2 Four-point probe resistance data for different platings
and copper in the initial state and after diffusion annealing by
electrical current and thermal gradient.
DIFFUSION ANNEALED BY ELECTRIC CURRENT
SnAgCu SnBi SnAgCuSb SnCu
Mean () 2.98 2.93 2.85 2.88 Std. Dev. 0.21 0.20 0.12 0.09 95%
Conf. 0.15 0.14 0.09 0.06 Min 2.57 2.64 2.71 2.71 Max 3.17 3.19
3.09 3.01 R () 0.30 0.25 0.24 0.20 R (%) 10.99 8.87 6.16 7.19
DIFFUSION ANNEALED BY THERMAL GRADIENT
Mean () 2.83 2.81 2.79 2.78 Std. Dev. 0.31 0.17 0.13 0.26 95%
Conf. 0.22 0.12 0.09 27.8 Min 2.51 2.65 2.71 2.47 Max 3.53 3.26
3.14 3.20 R () 0.15 0.13 0.11 0.10 R (%) 5.57 4.85 3.91 3.69
INITIAL STATE AS PREPAREDBY HOT DIPPING
Mean () 2.73 2.76 2.76 2.71 Std. Dev. 0.11 0.12 0.19 0.14 95%
Conf. 0.05 0.06 0.09 0.07 Min 2.53 2.58 2.53 2.56 Max 2.94 2.98
3.03 3.01 R () 0.05 0.08 0.08 0.03 R (%) 1.87 3.01 2.97 1.25
REFERENCE RESISTIVITY MEASUREMENTS
Cu ETP Ni Al-6061 Ag Mean () 2.68 9.99 5.57 2.48 Std. Dev. 0.06
0.26 0.20 0.15 95% Conf. 0.04 0.19 0.14 0.11 Min 2.57 9.62 5.18
2.20 Max 2.77 10.43 5.81 2.66 ( cm) 1.68 6.27 3.50 1.56
On the other hand, the resistance data of diffusion annealed
sampled showed completely different situation. This difference is
manifested not only by the magnitude of the resistance changes but
also by the type of diffusion annealing and the plating
composition. The diffusion annealed samples, plated
2690-7803-7862-8/04/$20.00 (C) IEEE.
-
with SnAgCu, showed the highest increase in the resistance while
those plated with SnAgCuSb and SnCu the least.
Another very important feature of these results is that the
increase in resistance of samples subjected to diffusion annealing
by electrical current is higher than that of diffusion annealed
ones in the thermal gradient. It appears that the intermetallic
phases formed under the influence of electrical field are more
resistive than those formed by thermal gradient. Some plausible
explanation for the observed difference will be discussed later in
the paper.
B. SEM and EDX Analyses
The width of the intermetallic layers was determined with a
scanning electron microscope (SEM). The advantages of the SEM
analysis of diffusion studies are basically the ease of operation,
direct observation, and possibility for determining the
concentration penetration data and thus calculation of the
diffusion coefficients.
The thickness of the intermetallic phases formed at the
plating-copper interface was determined from the SEM micrographs.
The results of these measurements are shown in Table 3. Figure 2
illustrates some typical morphology features of the intermetallic
phases formed during diffusion annealing by electrical current and
thermal gradient. Basically in all samples tested two phases were
detected.
From these results it can be inferred that the thicknesses of
the phases formed are apparently not affected by the type of
diffusion annealing used. In other words, it appears that the
formation is apparently independent of the diffusion annealing
treatment used.
The EDX elemental composition analysis of the intermetallic
phases formed indicated that the phase 1 corresponds to Cu3Sn while
phase 2 to Cu6Sn5. However, it should be pointed out that in the
cases of Sn-Ag-Cu and Cu-Bi alloys, silver and bismuth were also
found. Another interesting feature of the SEM examination of the
copper-plating interface is that in some cases like Sn-Cu alloys
cracks were observed at
Table 3 Thicknesses of intermetallic phases formed at the
interface between copper base and plating of samples diffusion
annealed by electrical current and thermal gradient.
DIFFUSION ANNEALING BY ELECTRIC CURRENT
SnAgCu SnBi SnAgCuSb SnCu
Phase 1 (m) 1 2 1 2 2 - 3 1 - 2
Phase 2 (m) 3 4 3 4 4 - 5 4 - 5
Plating (m) 40 60 50 50
DIFFUSION ANNEALING BY THERMAL GRADIENT
Phase 1 (m) 2 - 3 1 2 1 - 2 1 - 2
Phase 2 (m) 4 - 6 3 4 3 - 4 4 - 5
Plating (m) 40 60 50 50
the interface copper-intermetallics. The cracks were propagating
jot only along the boundary between copper and intermetallic phase,
but also across the intermetallic phase as well.
Sn Ag Cu Electrical Diffusion Annealing
Sn-Bi Thermal DiffusionAnnealing
Sn Cu Electrical Diffusion Annealing
Fig. 2 Typical morphology of the intermetallic phases formed at
the interface between copper and plating. Phase 1 corresponds to
Cu3Sn while phase 2 to Cu6Sn5
This is clearly illustrate din Fig.,3 depicting the morphology
of the copper-intermetallic phase interface of Sn-Cu alloys
subjected to diffusion annealing by electrical current. Note the
presence of extensive cracking not only along the intermetallic
interfaces but also across the phases. This feature, however, was
not present in the case of samples diffusion annealed in the
thermal gradient.
2700-7803-7862-8/04/$20.00 (C) IEEE.
-
Fig. 3. Cracks at the intermetallic boundaries in diffusion
annealed SnCu by electrical current.
IV. DISCUSSION
The results of numerous studies showed that the mechanical
properties of copper-tin are strongly affected by the presence of
the intermetallics when their thickness at the interface exceeds
the critical value of 2 microns. At this thickness, the interface
between the two metals in contact becomes brittle, thus making the
interface more highly porous and more susceptible to adverse
environmental effects due to the generation of numerous fissures in
the interdiffusion layer [10- 14]. This is clearly illustrated in
Fig. 9 showing extensive cracking at the interface between Cu3Sn
and Cu6Sn5 phases.
The formation of the intermetallic phases has an extremely
detrimental effect on the electrical resistance as manifested by
dramatic increases in the contact resistance. A direct consequence
of increased resistance is heating of the contact, which will
result in an increased rate of formation of intermetallic phases
and also other degradation processes such as creep, stress
relaxation, fretting, oxidation, corrosion etc.
The evolution of the intermetallic phases at the
lead-free-copper interface I schematically illustrated in Fig.4.
The results lf the SEM analysis indicated that diffusion annealing
either by thermal gradient or by electrical current for 25 days
resulted in the formation of basically Cu3Sn and C6Sn5 phases. For
practically all alloys used, the thickness of the Cu3Sn phase was
in the range 1-3 m whereas that of the C6Sn5 phase within 3-5
m.
AFTER PLATING EARLY STAGES LATER STAGES
Cu6Sn5
Cu
Cu3Sn
Cu
Cu6Sn5Cu3Sn
Cu
Sn
OXIDE
Sn
OXIDE OXIDE
Cu6Sn5
Sn
Cu6Sn5
Fig. 4 Schematic of the intermetallic phase formation at he
lead-free - copper interface after plating and in the early and
later stages of diffusion annealing.
The resistivity of the intermetallics formed at the
lead-free-copper interface can be calculated from the four-point
probe resistance measurements made on the cross-sectioned pads of
flexible connectors. Since the resistance measurement of the
microohmetre is based on the current penetration into the material
[8], a rectangular conductor, such as shown in Fig. 5 can be
envisaged as a volume of the material sampled by the microohmetre
probes. The cross-section area A = a b (b = 1.5 S) of such a
conductor was derived from the resistance measurements made on the
annealed copper samples using the known resistivity value for hard
drawn copper Cu = 1.80 cm and the probe spacing S.
R = (1/A) Cu S (2)
A = Cu S / R The results of resistance measurements made by
four-point
probe (Table 1) can be used to calculate the overall resistivity
(i) of the intermetallic layers formed. The calculations were made
from the weighted averages of resistors in series using the
following expression:
R = (1/A) n xn (3) i = (1/t ) [RA Cu (S - p) p (p t)]
b Cu SnIMA
S
A
pd
R RRCu Sni
R
t
a
Fig. 5 Schematic illustration of the conductor sampled by the
microohmetre probes.
where R is the total resistances of the conductor sampled by the
the potential probes of the microohmetre measured across the
intermetallic layer, copper base and plating and A is the
cross-section area of the conductor, n and xn are the resistivity
and the thickness of each of the components. The thickness of
intermetallic layer was determined from the SEM images of the
cross-sectioned busbars. Since the thickness of the
2710-7803-7862-8/04/$20.00 (C) IEEE.
-
intermetallic phases formed were within 4-6m, the value of 5 m
fwas used in these calculations. The results are shown in Table.
3.
Table 3. Calculated resistivity values for the intermetallic
layers formed at the lead-free-copper interface after diffusion
annealing by thermal gradient and electrical current.
SnAgCu SnBi SnAgCuSb SnCu
Plating (m) 40.00 60.00 50.00 50.00 IM (m) 5.00 5.00 5.00
5.00
Thermal 2.83 2.81 2.79 2.78
i ( cm) 19.06 16.80 13.54 12.54 Electric 2.98 2.93 2.85 2.88
i ( cm) 37.61 30.72 21.31 24.45 The obtained values for the
resistivities of the intermetallic
phases formed under the influence of thermal gradient are within
the values generally found in the literature for the Cu3Sn and
C6Sn5 phases. However, the resistivities of samples diffusion
annealed by electrical current are significantly higher than those
treated by temperature gradient. Hence, if the assumptions used to
make these calculations are correct, then the obtained results are
very intriguing, indeed.
Although a specification of the exact mechanism for the observed
difference is beyond the scope of this work, some plausible
explanations can be put forward to account for the effect of
electrical current on the resistivity of diffusion annealed
samples.
On possibility is that the observed difference is an indication
of different stresses generated during diffusion annealing. It is
now well established that as a result of rapid interstitial and
grain boundary diffusion, significant mechanical stresses can be
developed in the intermetallic layers that impair mechanical
integrity of the layered contact interface [12, 15].
Indeed, the microhardness of the phases formed is much harder
than that of copper and lead-free plating and are very are brittle
[1, 4]. Hence, it would appear that under the influence of
electrical current, higher mechanical stresses are generated
leading to crack formation at the intermetallic interfaces. Indeed,
as seen in Fig. 3, cracks were observed in the samples subjected to
diffusion annealing by electrical current.
In addition, since an electric current was used to generate the
intermetallic growth, it may be argued that the observed
acceleration might be associated with the materials transport by
electromigration [16]. In electromigration the material transport
occurs via interaction between the atoms of a conductor and a high
density current of the order of 104 - 105 A/cm2. However,
electromigration is not expected under the low-current densities or
AC conditions. The AC current densities used in this work (
-
[7]. M. Braunovic and M. Marjanov, Thermoelastic Ratcheting
Effect in Bolted Aluminum-t-Aluminum Connections, IEEE Trans.
CHMT-11, vol. 11, No. 1 (1988) p. 54.
[8]. W. Hain, Monitoring Material Property Transformations with
Electrical Resistivity, Mater. Eval., vol. 47, June (1989) p.
619
[9]. F.M. Smits, Measurements of Sheet Resistivities with the
Four-Point Probe, Bell Syst. Tech. J., vol. 37, (1958) p. 711
[10]. R.J Fields and S.R. Low, Physical and Mechanical
Properties of Intermetallics Compounds Commonly Found in solder
Joints:+ NSIT Publication, www.metallurgy.nist.gov
[11]. W. Urquhart, Interdiffusion Studies on Contact Plating
Materials, Electrical Contacts -1976, ITT, September (1976) p.
185.
[12]. C.L. Bauer and G.G. Lessmann, "Metal-Joining Methods",
Annual Review of Materials Sci., 6, (1976), pp. 361-387.
[13]. U. Lindborg, B. Asthner, L. Lind and L. Revay,
"Intermetallic Growth and Contact Resistance of Tin Contacts after
Aging", Electric Contacts-1975, IIT, Chicago, (1975), p. 25.
[14]. L. Zakraysek, Intermetallic Growth in Tin-Rich Solders.
Welding Research Suppl., November (1972) p. 537s.
[15]. S. F. Dirnfeld and J.J. Ramon, Microstructure
Investigation of Copper-Tin Intermetallics and the Influence of
Layer Thickness on Shear Strength, Weld. Res., October (1990), p.
373-s.
[16]. H.B. Huntington, "Electromigration in Metals" in Diffusion
in Solids: Recent Developments, Eds. A.S. Nowick, and J.J. Burton,
NY, Academic Press, (1974), Chapter 6
[17]. N.A. Gjosten, Diffusion, ASM (1973) p. 241
[18]. M. Braunovic and N. Alexandrov, Intermetallic Compounds at
Aluminum-to-Copper Electrical Interfaces: Effect of Temperature and
Electric Current, IEEE Trans. CPMT, Part A, vol 17, (1994) p.
78
[19]. A.K. Bandyopadhyay and S.K. Sen, "A Study of Compound
Formation in a Copper-Tin Bimetallic Couples", J. Appl. Phys., 67,
(1990), p. 3681.
Dr. Milenko Braunovic received his Dipl. Ing Degree in Technical
Physics from the University of Belgrade, ex-Yugoslavia (now
Serbia-Montenegro), in 1962 and the M. Met. and Ph. D. degrees in
Physical Metallurgy from the University of Sheffield, England in
1967 and 1969 respectively. From 1971 until 1997 he was working at
IREQ, Hydro-Quebec Research Institute, Varennes, Quebec as a senior
member of the scientific staff. He retired from IREQ in 1997 and
established his
own scientific consulting company, MB Interface. During the last
30 years, he has been responsible for the development and
management of a broad range of research projects for Hydro-Qubec
and the Canadian Electrical Association (now CEA Technology) in the
areas of electrical power contacts, connector design and
evaluation, tribology and accelerated test methodologies. He also
initiated and supervised the R&D activities in the field of
shape-memory alloy applications in power systems
Dr. Braunovic is the author of more than 100 papers and
technical reports, including contributions to encyclopaedias and
books, in his particular areas of scientific interests. In addition
he frequently lectures at numerous seminars worldwide and has
presented a large number of papers at various international
conferences. For his contributions to the science of electrical
contacts, Dr. Braunovic is recipient of the Ragnar Holm Scientific
Achievement Award, the Ralph Armington Recognition Award and the
IEEE CPMT Best Paper Award. He successfully chaired the 15th
International Conference on Electrical Contacts held in Montreal
1990 and was a Technical Program Chairman of the 18th International
Conference on Electrical Contacts held in Chicago 1996. He is a
Senior Member of IEEE, and a member of American Society for Metals
(ASM), Materials Research Society (MRS), Planetary Society,
American Society for Testing of Materials (ASTM) and The Minerals,
Metals & Materials Society (TMS).
Daniel Gagnon received his B.Eng. and M.Eng. degrees in
mechanical and systems engineering from cole de Technologie
Suprieure in Montreal Canada in 1995 and 1998 respectively. He is
presently a project manager at Hydro Qubec Research Institute,
Canada in the mechanical, metallurgical and civil engineering
department. He is also a Ph. D. candidate at
the cole de Technologie Suprieure. His research interests are
electrical contacts, composite materials and aerial conductors.
2730-7803-7862-8/04/$20.00 (C) IEEE.