Mechanical shear properties of adhesives

E.M.P. Huveners, F. van Herwijnen, F. Soetens and H. Hofmeyer

Department of Architecture, Building and Planning, Structural Design Group, Technische Universiteit Eindhoven
Eindhoven, The Netherlands

New Products, Applications and Nanotechnology

Keywords

1=shear modulus 2=glued joint 3=stiffness 4=adhesive

Abstract shear stiffnesses (k2 and k3). The stiffness The formulae need the Young’s
is determined by the modulus (Young’s modulus and shear modulus of the
A glued joint is a suitable connection
modulus and shear modulus) and the adhesive, but these properties are not
between a glass pane and a framework.
joint’s thickness (formula 1 and 2). given by most suppliers. To determine
The mechanical properties of adhesives
the shear properties, tests have
are not given in specifications. The
been carried out in conformity with
most important property of adhesives
(1) standards [1]. Two different types of
is the shear - strain relation. Tests are
adhesives are tested, polyurethane
carried out to determine this relation
and epoxy. Polyurethane is SIKAFLEX-
under prescribed conditions. Curing
252 and is a one-component moisture
conditions, short-lived increase of
(2) curing adhesive. Epoxy is Scotch Weld
temperature, storing time and testing
9323 B/A and is a two-component
protocol are the same for each adhesive.
adhesive curing chemically. This paper
Two adhesives, polyurethane and epoxy,
In which: discusses the experimental research for
are tested at a constant deformation
k1 is the normal stiffness in N/mm3 determining the shear properties of
rate and at three different temperatures.
k2/3 is the shear stiffness in N/mm3 these two adhesives [2].
Polyurethane has a low stiffness which
Ea is the Young’s modulus of the
slightly increases with larger strains. The
adhesive in N/mm2 Test set-up
stiffness and shear strength decrease
Ga is the shear modulus of the adhesive
with increasing temperature. Epoxy The tests are carried out in conformity
in N/mm2
shows a stiff and toughened shear with the German standard DIN 54 451
tj is the thickness of the joint in mm
- strain relation at room temperature. (test specimens, preparations, etc.)
Under frozen conditions, the epoxy has [3] and accompanying standards and
a larger stiffness and shear strength, but ETAG 002 [4,5] using aluminium test
fails brittle. The stiffness significantly specimens. Figure 2 shows the geometry
decrease at high temperature of the test specimens for epoxy and
accompanying with large strains. Both polyurethane. The joint’s length of test
adhesives are subjected to a short-lived specimens for polyurethane deviates
increase of temperature three days from the standard. This length is
after making the specimens. This had enlarged from 5 mm to 18 mm, because
favourable effect on epoxy (post curing), this adhesive deforms more than the
leading to a more toughened behaviour. prescribed length. The thickness of
Polyurethane shows smaller values for the joint is 1.0 mm and 0.5 mm for
shear stress and larger values for strain. polyurethane and epoxy respectively.
After longer storing time each adhesive The specimens are treated to increase
shows a slightly increase of the shear Figure 1 the adhesion between adhesive and
strength. The shear modulus of the aluminium. The treatment exists of
Joint stiffnesses
adhesives has been determined. The immersing the aluminium specimens
shear modulus of polyurethane can be
simplified by a bi- or tri- linear diagram
and for epoxy by a linear diagram.
 WM  
Finally, the found shear strength is not
the ultimate shear strength, because all  WM   
test specimens fail adhesively instead of


cohesively.





Introduction






Glass has the possibility to act as a shear

 


wall to stabilise a building. Therefore,

  



the glass pane has to be structurally




connected to the framework to transfer




the in-plane load. A circumferentially



glued joint is a suitable connection.











Because this spreads stresses and

prevents glass-metal contact. The glued




joint can be modelled as bedding and

has three linear stiffnesses (figure 1),
namely a normal stiffness (k1) and two Figure 2
Test specimens (aluminium) for epoxy (left) and polyurethane (right)

GLASS PERFORMANCE DAYS 2007 | www.gpd.fi 367

 in a chemical bath (chromium short-lived increase of temperature. This
sulphuric acid solvent). The bath effect has no influence on the shear-
(4)
has a temperature between 60°C to strain relation at -20°C. However, at
65°C and the duration of immersion 23°C and at 80°C, the shear stress and
is 30 minutes. After immersing, the shear strain decrease compared with the
In which:
specimens are rinsed by distillated water same test conditions without a short-
T is the calculated average shear stress
and dried in a dust free chamber at lived increase of temperature.
in N/mm2
New Products, Applications and Nanotechnology

60°C for 30 minutes. After drying the Figure 5 shows the shear-strain
F is the measured force in N
specimens are fixed on a template and relation for epoxy batches 10, 12, 14
Aj is the joint area in mm2
the joints are made within 15 minutes. and 16 tested at temperature 23°C,
G is the shear strain [-]
Each adhesive has 7 batches -20°C, 80°C and 23°C respectively. At
v is the measured shear deformation in
consisting of 4 specimens. After making -20°C the relation is linear and fails
mm
the test specimens, the samples are immediately without warning at a strain
tj is joint thickness in mm
stored in a climate chamber with of 0.12 with an ultimate shear stress
constant temperature (20°C) and 37.68 N/mm2. The relation at 23°C is
relative humidity (60%). Epoxy has been
Results bi-linear. The first part is also linear till
tested after 7 days and polyurethane This paragraph deals with the a strain of 0.12 and shear stress 26.71
after 23 days. The enlarged joint area of experimentally found results. Figure N/mm2. The second part shows a slightly
the moisture cured polyurethane needs 3 shows the shear-strain relation for decrease of shear stress from 26.71
a longer curing time. Three batches of polyurethane batches 03, 05, 08 and N/mm2 to 25.43 N/mm2 at a strain of
each adhesive have a twenty four hours 09 tested at temperature 23°C, -20°C, 0.50. At 80°C the relation is linear with
short-lived increase of temperature 80°C and 23°C respectively. The shear- a large ultimate strain (0.70) and a
of 80°C to investigate temperature strain relation of this adhesive clearly small ultimate shear stress (7.6 N/mm2).
history (see table 1 denoted to (2)). has two parts at -20°C and 23°C. The Batch 16 has a longer storing duration
Each adhesive has been tested at first part shows a linear relation till a and gives the same results at test
three temperatures: -20°C, 23°C and shear strain 0.75 with shear stress 0.65 temperature 23°C than batch 10 with
80°C. The time between placing the N/mm2 for -20°C and 0.42 N/mm2 for storing time of 8 days.
specimen in the climate chamber 23°C. The second part, the shear stress Figure 6 shows the shear-strain
at room temperature and testing at rapidly increases to 3.34 N/mm2 for relation for epoxy batches 11, 13 and 15
-20°C or 80°C is 10 minutes. The -20°C and to 1.73 N/mm2 for 23°C. tested at temperature 23°C, -20°C and
joint’s temperature is measured with The relation for 80°C is different. The 80°C respectively. These batches have a
a thermocouple. Finally, one batch of shear stress gradually increases with short-lived increase of temperature. The
each adhesive has been tested after large shear strain. Batch 09 has a longer relation at -20°C is linear till a strain of
approximately 8 weeks to investigate storing duration and gives slightly better 0.24 and than the relation becomes a
aging (see table 1 denoted to (3)). The results at a test temperature 23°C than curve till a strain of 0.28. The ultimate
displacement rate is 2.5 mm/min. The batch 03 with 23 days storing. shear stress is 36.63 N/mm2 and 49.72
specimens are fastened with a universal Figure 4 shows the shear-strain N/mm2 respectively. The adhesive fails
joint in the tensile bench. relation for polyurethane batches 04, without warning. The relation at 23°C is
06 and 07 tested at temperature 23°C, almost comparable to tests at the same
Adhesive Batch Testing -20°C and 80°C respectively. These temperature without short-lived increase
temperature batches have a
[°C]
Polyurethane 03(1) 23 Figure 3
04(2) 23 T-G-relation PU tested at
23 days(1) and 60 days(3).
05(1) -20
06(2) -20
07(1) 80
08(2) 80
09(3) 23
Epoxy 10(1) 23
11(2) 23
12(1) -20
13(2) -20
14(1) 80
15(2) 80
16(3) 23

(1)
no short-lived increase of temperature, curing
time PU 23 days and epoxy 8 days.
(2)
short-lived increase of temperature of 80°C at Figure 4
day 3 after making the joint, curing time see (1). T-G-relation PU with
(3)
curing time 60 days and no short-lived in- short-lived increase of
crease of temperature. temperature tested at 23
days(2).
Table 1
Oversight test program

The average shear stress is the

measured tensile force divided by the
joint area (formula 3). The shear strain
can be calculated with formula 4.

(3)

368 GLASS PERFORMANCE DAYS 2007 | www.gpd.fi

of temperature. The first part is also elastomers. The first part is stretching The short-lived increase of
linear till a strain of 0.12 with a lightly the entangled chains to one direction. temperature has a positive influence
smaller shear stress (24.01 N/mm2). The Once the chains have one direction, on the mechanical properties of epoxy.
second part shows a slightly increase of they are able to carry load. The adhesive This is post curing. The brittle behavior
the ultimate shear stress from 24.01 N/ is sensitive for a short-lived increase of at -20°C becomes more toughened and
mm2 to 28.36 N/mm2 at a strain of 0.50. temperature. the ultimate shear stress increases. Tests
The relation at 80°C is a curve which Epoxy at 23°C behaves stiff followed at 23°C also have larger ultimate shear

New Products, Applications and Nanotechnology

gradually decreases. The ultimate strain by a plastic-like zone in which the stress and the second part of the stress
(0.42) is smaller and the ultimate shear stiffness slightly decreases. Toughened – strain diagram slightly increases. Tests
stress is larger (8.54 N/mm2) compared behavior is a property of this epoxy. At at 80°C shows slightly larger ultimate
with the same tests without short-lived -20°C it behaves slightly stiffer than shear stresses and the shear stress-strain
increase of temperature. the same epoxy at 23°C, but without a diagram is more convex.
Table 2 gives an oversight of the plastic-like zone. It also fails brittle. The
experimental results of each batch for epoxy shows a very small stiffness at T [°C] G [-] Ga [N/mm2]
polyurethane and epoxy. The measuring 80°C in comparison to the stiffness at -20(1) 0.06 387
data has been truncated at the given the other two temperatures. Mechanical -20(2) 0.08 290
strain, because the specimens start properties quickly drop at temperatures
23(1) 0.04 280
rotating with pealing off the joint from larger than the glass transition
23(2) 0.06 230
the aluminium (adhesion problems). temperature [1]. For this epoxy the glass
The shear-strain relation of transition temperature is 60°C. 23(3) 0.66 273
polyurethane is characteristic for 80(1) 0.04 11
80(2) 0.04 45
PU Epoxy Table 3
Ttest [°C] Tave [N/mm2] G [-] Standard Tave [N/mm2] G [-] Standard error
error [%] [%] Ga-modulus epoxy till given strain G.
-20(1) 3.34 1.35 3.0 37.68 0.14 7.6
-20(2) 3.35 1.35 6.7 49.72 0.28 2.0 The shear stress - strain relations
23(1) 1.73 1.35 2.7 25.42 0.50 2.8 of polyurethane and epoxy under
23(2) 1.33 1.30 6.6 28.36 0.50 4.9 prescribed conditions have been
23(3) 1.77 1.35 1.2 25.43 0.50 1.0
determined. The next step is to
determine the shear modulus. The shear
80(1) 0.78 1.25 6.9 7.60 0.70 12.8
stress – strain relation for polyurethane
80(2) 0.65 1.15 2.9 8.54 0.42 5.3
is simplified to a bi- or tri- linear diagram
(1)
no short-lived increase of temperature, curing time PU 23 days and epoxy 8 days.
(figure 7 and 8). The relation for epoxy
(2)
is simplified to a linear diagram (table
short-lived increase of temperature of 80°C at day 3 after making the bond line, curing time see (1).
3). The shear stiffness of the joint
(3)
curing time 60 days and no short-lived increase of temperature. can be calculated with (formula 2).
Table 2 The shear stiffness of polyurethane is
Experimental results just before pealing off the joint from the aluminium. small, because of the very small shear
modulus and the thickness of the joint
of several millimetres. Conversely, the
Figure 5 shear stiffness of epoxy is large, because
T-G-relation epoxy of the shear modulus is large and the
tested at 8 days(1) and 60 thickness of the joint is very small (0.1
days(3). mm).

Conclusions
The shear stress - strain relation and
shear modulus have been determined
experimentally for Sikaflex 252 and
Scotch Weld 9323 B/A at prescribed
conditions. So, the linear as well as
non-linear shear stiffness of a glued
joint can be used in numerical models.
The experimentally found values are
applicable till the given strain. The real
values are larger than experimentally
found, because the specimens fail
adhesively and not cohesively. And
Figure 6 therefore, no ultimate shear strength
T-G-relation epoxy with and strain relation has been found.
short-lived increase of The shear stress - strain diagram
temperature tested at 8 of polyurethane is tri-linear. The first
days(2).
part shows a straight line with small
shear stresses and than gradually the
stresses increase at larger strains. The
testing temperature influences the
value of the ultimate shear stresses.
Larger temperatures lead to lower
ultimate shear stresses. The short-lived
increase of temperature has a negative
effect on the ultimate shear stresses
(degradation), but there is no effect
at -20°C. The shear modulus at small
strains is small, but increases at larger

GLASS PERFORMANCE DAYS 2007 | www.gpd.fi 369

 Figure 7 strains. The shear stiffness with regard
Ga-modulus PU tested at to the thickness of the glued joint is very
23 days(1) and 60 days(3). small.
On the other side, epoxy shows
larger values for ultimate shear stresses
and small values for ultimate strains
than polyurethane. This epoxy shows a
New Products, Applications and Nanotechnology

toughened behavior at 23°C. At frozen

conditions epoxy has larger ultimate
shear stresses than at 23°C, but it
behaves brittle. Temperatures larger
than the glass transition temperature
lead to significantly smaller values for
the ultimate shear stresses and sliding
off of the joint from the aluminum such
as with this epoxy. The shear stress
and strain are favorable at a short-lived
increase of temperature (post-curing).
Figure 8 The shear stiffness of epoxy with respect
Ga-modulus PU(2) with to the very small thickness of the glued
short-lived increase of
temperature tested at joint is larger than of polyurethane.
23 days(2).
References
[1] Habenicht, G (2006), Kleben, Grundlagen,
Technologien, Anwendungen, 5. aktualisierte
Auflage, Berlin, Germany
[2] Huveners, E.M.P., Koggel, B.A. (2006), Tensile
Shear Tests for the Determination of the Shear-
Strain Diagram of several Adhesive Types,
Report Technische Universiteit Eindhoven, the
Netherlands
[3] DIN 54 451, Schubspannungs-
Gleitungsprüfung, Deutsches Institut für
Normung, Berlin, Germany
[4] ETAG 002 (2003), Guideline for European
Technical Approval of Structural Sealant
Glazing Systems, European Organisation for
Technical Approvals, Brussels, Belgium
[5] Wellershoff, F. (2006) Nutzung der Verglasung
zur Aussteifung von Gebäudehüllen, Shaker
Verlag GmbH , Aachen, Germany

370 GLASS PERFORMANCE DAYS 2007 | www.gpd.fi