Normas trabajos finales XXIV Encuentro GEF

Transcription

On characterizing microscopically the adhesion interphase for the adhesion
between brass-plated steel cords with different Cu content and rubber compound
Gyung Soo Jeon1
1
Dept. of Fire-Safety Man., Jeonnam Provincial College
152 Juknokwonro Damyangyangeup damyanggun Jeonnam 517-802 South Korea.
[email protected]
ABSTRACT
Brass-plated steel cords inserted in the belt and carcass of tires have long been used as a reinforcing material to
provide sufficient mechanical strength and stability to endure cars themselves and their loads. Good adhesion of a
rubber compound to brass-plated steel cord is very important in radial tires. Brass plating on the surface of steel
cords reacts with sulfur in the rubber compound during the curing process of tire manufacturing, forming an adhesion
interphase between the rubber compound and the steel cord. Copper and zinc also react with oxygen and water in the
rubber, forming oxides and hydroxides of copper and zinc. Therefore, the brass oxide thickness and composition play
an important role in determining the adhesion characteristics between rubber compound and brass plated steel cord.
Therefore, the adhesion interphase is very complex in terms of components and content, so good adhesion can be
achieved when the adhesion interphase is formed with a sufficient thickness and a stable structure. In this study, three
kind of brass-plates steel cord with different Cu content and the same brass plating amount were prepared and
adhesion properties between rubber compound and brass-plated steel cord were measured in order to study the effect
of Cu concentration of brass-plated steel cord on the structure and thickness of adhesion interphase between rubber
compound and brass-plated steel cord against hostile aging treatments. The adhesion interphases of adhesion samples
were analyzed by the Ar+ sputtering of Auger electron spectroscopy (AES). The adhesion force after only cure
decreased with increasing copper content of brass plating. But adhesion force after humidity aging treatment
decreased with increasing copper content of brass plating. The enhanced adhesion durability between rubber
compound and brass-plated steel cord with lowest copper content may be explained as the controlled growth of
adhesion interphase as confirmed by depth profiling measurement of AES.
KEY WORDS: rubber-to-brass bonding; adhesion interphase; AES; depth profile; dezincification, copper sulfide, zinc oxide; humidity aging; adhesion
stability
and cure condition of the rubber compound are required
for a strong and stable adhesion interphase [2]. Brass
has been used as a plating material for steel cord for a
long time because of its excellent processability and
adhesion properties. But the additional growths of
copper sulfide and zinc oxide, which cause adhesion
degradation, is inevitable due to the heat generated
from a tire during service and the contact with moisture
and oxygen in air. The excessive growths of sulfides
and oxides result in cohesive rupture of the adhesion
interphase, so a continuous search has been made for
substitutes which maintain adhesion interphase
integrity under severe service conditions for a long
period. The major components of the adhesion
interphase are sulfides, oxides and hydroxides of copper
and zinc [3]. A large content of zinc oxide at the outer
surface of the brass induces a cohesive failure, since its
mechanical strength is very weak. On the other hand,
the coexistence of zinc oxide with zinc in the
interphase is helpful to control the mass transfer rate of
reacting species in the formation of the adhesion
interphase. This contributes to the stability of the
adhesion interphase by preventing excessive growth of
1. INTRODUCTION
Adhesion between rubber compound and metal has
been widely used in many industrial fields. Good
adhesion of a rubber compound to brass-plated steel
cord is very important in radial tires. Steel cords, the
surfaces of which are plated with brass for adhesion
with rubber compound, are inserted into the rubber
compounds of belt and carcass in order to enhance the
mechanical stability of a tire. The adhesion between
the rubber compound and the brass-plated steel cord is
obtained by the reaction of copper in the brass and
sulfur in the rubber compound during vulcanization,
forming copper sulfide in the adhesion interphase [1].
Besides copper sulfide, oxides and hydroxides of
copper and zinc are formed as well in the adhesion
interphase due to the reaction of brass with residual
water and air in the rubber compound. Therefore, the
adhesion properties between rubber compound and
brass-plated cord are strongly dependent on the
composition and status of the adhesion interphase.
Thus, elaborated controls of the composition and
plating weight of the brass, as well as the composition
121
its components. In this research, three brass-plated steel
cords with different copper content were prepared. The
adhesion properties between the rubber compound and
the brass-plated cord were investigated, and adhesion
stability was also studied under thermal and humidity
aging treatments.
For humidity aging, specimens were placed in a
humidity chamber at 85 oC under 85% relative
humidity for 5, 10, and 15 days. Also, the adhesion
samples were aged thermally at 90 oC for 5, 10, and 15
days under air atmosphere. Pull-out force was
determined as the maximum force exerted by the tensile
tester on the T-test adhesion sample during the pull-out
test, at a crosshead speed of 10 mm/min. Rubber
coverage, defined as the percentage of rubber-adhered
area on cord surface, was also noted. Each value
reported is an average of six specimens tested. The
morphology of the pulled-out steel cord surface after
measuring pull-out force was studied using an image
analyzer.
A brass plated steel cord was covered with a filter
paper (pore size: 5 m; catalog no LSWP 142 50,
Millipore Co., USA), sandwiched between two uncured
pads of rubber compound, and then placed in a pad
mold [1,3]. Curing and aging conditions for the rubber
compound/brass plate samples were the same as in the
preparation of the T-test specimens. After the various
treatments, samples for the surface analysis of the
adhesion interphase were obtained by peeling away the
filter paper. Sulfur from the rubber compound migrated
through the pores of the filter paper and reacted with
the copper and zinc of the brass-plated steel cord,
forming an adhesion interphase. After removing the
rubber and filter paper from the brass-plated steel cord,
the adhesion interphase, including copper sulfide and
zinc oxide, remained on the brass-plated steel cord.
The depth profiles from the interphase in contact with
the rubber compound to the bulk of the brass were
recorded on a Ulvac-PHI Auger spectrometer (model
Ulvac-PHI 7000, Ulvac-PHI Ins., U.S.A.). An area of
2. EXPERIMENTAL
A rubber compound was prepared. Masterbatch
components were as follows; natural rubber (Lee
Rubber Co., Malaysia, SMR-100), 100 phr; carbon
black N330 (Lucky Co., Korea), 45 phr; aromatic
processing oil (Michang Co., Korea, A#2), 4 phr; zinc
oxide (Hanil Co., Korea), 10 phr; antioxidant
(Monsanto Co., USA Kumanox-RD, 2,2,4-trimethyl1,2-dihydroquinone), 1 phr; and cobalt salt (Rhone
Pouluenc Co., France, Manobond 680C), 0.43 phr;
adhesion promoter (Indspec Co., USA, B-18S), 2 phr.
Final rubber compound components were as follows:
masticated rubber masterbatch, 100 phr; stearic acid
(Pyungwha Co., Korea), 3 phr; accelerator (Monsanto
Co., USA, Santocure MOR, 2-(morpholinothio)- thiobenzothizole), 0.8 phr; and insoluble sulfur (Akzo Co.,
The Netherlands, Crystex HS OT 20), 4 phr; adhesion
promoter (Cytec Co., USA, Cyrez-964), 4 phr; PVI, 0.2
phr. All the rubber compounds were mixed as described
in ASTM D-3184 using an internal mixer (Banbury
Mixer model 82, Farrel Co., USA). Ingredients for the
masterbatch were mixed for 10 min at a rotor speed of
40 rpm and discharged at 150 oC.
After the
masterbatch had cooled to room temperature, the final
mixing components were mixed for 5 min at 30 rpm
and discharged at 90 oC. After mixing, the compounds
were carefully remilled into flat sheets on a two-roll
mill (model MKIII, Farrel Co. USA). Based on the
procedure described in ASTM D-2229, T-test
specimens were cured at 150 oC on a cure press.
Curing was continued for 5 min more than t90 time.
The three kind of brass-plated steel cord were prepared
with having different Cu content in brass-plating as
shown in Table 1. The brass-plated steel cords with 3 x
0.30 construction in which 3 steel wires having the
same diameter of 0.30 mm were twisted together,
manufactured by Hyosung T&C Co., Korea, were used.
10 ×10 ㎛ 2 was examined using an ion beam with a
potential of 3.0 keV, a current of 10 nA, and an
incident angle to the specimen of 60 o. Surface
concentrations were determined every 0.5 min from the
Auger peaks of detected elements with compensation
for their sensitivities. A sputter gun with an argon ion
beam rastered a 2×2 mm2 area for depth profiling. The
sputtering rate for the brass film was determined to be
14 nm/min.
3. RESULTS and DISCUSSION
Table 1. Specification of brass-plated steel cords employed with
having different Cu content
Specification
LC
Cord diameter (mm)
Pitch (mm)
Tensile load (N)
Elongation (%)
Weight (g/m)
Cu content (wt%)
Brass-plating (g/kg)
0.650
13.8
677
2.4
1.699
51.4
3.83
Steel Cord
MC
0.647
13.8
680
2.2
1.687
64.2
3.88
The surface composition for three kind of brass-plated
steel cords was examined by depth profiling of AES
(Fig.1). The carbon peak of steel cord decreased
abruptly with increasing sputter time. At the 1 min of
sputter time, the carbon concentration is nearly trace
level. The concentration of copper and zinc is linearly
increased with sputter time up to 2 min. Further
increase of sputter time of 2 min, the concentration of
copper and zinc is nearly constant indicating the
existence of brass. The iron detected after the sputter
time of 1 min and its concentration was linearly with
HC
0.646
13.8
673
2.5
1.683
75.5
3.86
122
LC3)
MC4)
HC5)
sputter time. The nitrogen detected in the outermost
surface of brass-plated steel cord. This is may be
resulted from the residual lubricant on steel cord which
may be arisen from the manufacturing process of
drawing of filament. The concentration and width of
copper increased with increasing the copper content as
shown in Fig.1.
Concentration (atom. %)
(A)
C
O
N
Cu
Zn
Fe
100
80
(B)
C
O
N
Cu
Zn
Fe
100
80
80
60
60
60
40
40
40
20
20
20
0
0
0
2
4
6
8
10
The adhesion samples were humidity aged at 85 oC
and 85% relative humidity. 2) Aging period (days).
The pull-out force of the HC cord was reduced to 40%
of the unaged force after humidity aging of 15 days,
and the rubber coverage abruptly decreased to 20%.
The adhesion properties are degraded by thermal
aging as shown in Table 3. But the degradation rate of
adhesion properties is significantly different from the
copper content.
With increasing copper content,
adhesion properties after thermal aging worsen. Also,
HC cord showed a large degradation of adhesion
properties. The excellent adhesion retention after
thermal aging exerted for LC cord.
Another important point relating to adhesion for tire
manufacturing is a stability resisting heat, air, salt and
humidity. The adhesion stability of brass-plated cord
with low copper content against thermal aging was
considerably high compared with that of the brassplated cord. The unaged adhesion properties of LC
cord were superior to the MC and HC cords and the
degradation rate was slightly slow for LC cord; i.e., the
pull-out force of the LC cord with the lowest copper
content after thermal aging of 15 days was recorded as
275 N, 54% of the unaged force, and the rubber
coverage was noted as 95%, which was the same as the
unaged rubber coverage of 95%. On the other hand,
the unaged pull-out force of the HC cord was as high as
298 N, but after thermal aging of 15 days it was
reduced to 138 N, which was 46% of the unaged force.
It is worth noting that at the
0
0
2
4
6
8
10
0
2
4
6
8
10
Figure 1. Sputter
AEStimedepth
profiles Sputter
of raw
steel
cord with
(min)
time (min)brass-plated
Sputter
time (min)
respect to Cu content: (A) LC; (B) MC; (C) HC.
The adhesion properties of the brass-plated cords
having different copper content before and after
humidity aging treatments, as shown in Table 2. The
unaged adhesion properties of LC cords were superior
to those of the other cords. The adhesion properties
were high when the copper content was low. The LC
cord, which had the lowest copper content, showed 506
N, while the unaged pull-out force of the HC cord was
298 N. The rubber coverage also became higher as the
copper content was lower. Rubber coverage was found
to be 40% on the HC cord, but on the LC cord it was
95%. The pull-out force and rubber coverage are better
on brass-plated steel cord with a small amount of
copper content; therefore, the LC cord having the
lowest copper content shows the best adhesion
properties among the brass-plated cords prepared in
this study. The adhesion stability of brass-plated cord
with low copper content was more exceptional than the
brass-plated cord with high copper content against
humidity aging, as shown in Table 2. The pull-out
force and rubber coverage of the LC cord after humidity
aging of 15 days were both superior to those of the MC
and HC cords. With humidity aging, the pull-out
forces of three cords decreased regardless of copper
content, but the degradation rate was extremely slow on
the MC cord. The rubber coverage of the HC cord was
significantly decreased with humidity aging, while that
of the LC cord was rather improved and/or slightly
decreased. The rubber coverage of the HC cord at the
unaged state was poor; and, after humidity aging, they
were half decreased. This result shows the exceptional
adhesion stability of LC cord compared with the
adhesion degradation of MC and HC cords.
Table 3. Adhesion test results for thermally-aged1) adhesion samples
between rubber compound and brass-plated steel cord having
different Cu content
Steel Cord
LC3)
MC4)
HC5)
1)
2)
Pullout force(N)/Rubber coverage (%)
02)
5
10 15
506/95
311/70
298/40
376/100 285/95
361/90 276/95
256/30 160/30
275/95
261/95
138/30
The adhesion samples were thermally aged at 90 oC.
Aging period (days).
unaged state the pull-out force of the LC cord was
about two times of the HC cord, and adhesion retention
after thermal aging of 15 days, it was almost the same
as that of HC cord. The rubber coverage of the LC cord
was nearly constant with thermal aging, while that of
the HC cord slightly lowered.
The analysis of the surface layer on the brass plate
provides information on the formation and
disappearance of chemical components in the adhesion
interphase during curing process. Fig. 2 shows AES
Table 2. Adhesion test results for humidity-aged1) adhesion samples
between rubber compound and brass-plated steel cord having
different Cu content
Steel Cord
427/100 355/100 336/85
353/75 302/85 285/80
157/20 125/20 120/20
1)
(C)
C
O
N
Cu
Zn
Fe
100
506/95
311/70
298/40
Pullout force(N)/Rubber coverage (%)
02)
5
10 15
123
conditions since zinc is easily dissolved in the presence
of water. The copper atom in brass becomes active due
to the loss of the zinc atom, and results in excessive
growth of copper sulfide during humidity aging. In
contrast, there is a controlled formation of copper
sulfide on the LC cord, resulting in exceptional stability
during humidity aging.
depth profiles of the unaged adhesion interphase
formed between rubber compound and MC cord. At the
outer surface of the brass plated steel cord adhered to
the rubber compound, carbon, copper and sulfur were
detected. Beneath these elements, zinc, oxygen and
iron were detected. The affluence of carbon at
outermost interphase is due to surface contamination.
With increasing sputter time, carbon concentration
decreased exponentially. Iron detected significantly
from 1 min of sputtering and increased linearly up to
10 min of sputtering. After 4 min of sputtering, copper
and zinc contents were constant with depth, indicating
non-reacted brass. This depth profile shows that
copper sulfide is formed on the outer surface of the
brass plated steel cord and zinc oxide on the inner side,
although the oxidation states of these elements are not
clearly identified. The adhesion properties of the brassplated
4. CONCLUSIONS
Brass-plated cord (LC cord) with low copper content
about 51.4 %, shows good adhesion with the rubber
compound. The adhesion properties between brassplated steel cord and the rubber compound improved
with the lowering in the copper content. At the unaged
Cu
S
80
80
40
40
40
20
20
80
4
6
8
10
0
0
Zn
O
2
4
6
8
10
0
Zn
O
30
20
20
10
10
10
0
0
2
4
6
8
10
2
4
6
8
10
4
6
8
10
Zn
O
30
20
0
(C)
20
0
2
30
S
C
O
Fe
Cu
Zn
80
60
0
Cu
S
(B)
60
0
100
Cu
S
(A)
60
0
0
2
Sputter time (min)
4
6
8
10
0
Sputter time (min)
2
Sputter time (min)
Figure 3. AES depth profiles of Cu, S (top) and Zn, O (bottom) for the
adhesion interphases of unaged adhesion samples between the rubber
compound and brass plated steel cord with respect to Cu content: (A)
LC; (B) MC; (C) HC.
60
40
Figure 2. AES depth profiles of C, O, S, Fe, Cu and Zn for the
adhesion interphase of unaged adhesion samples between rubber
compound and brass plated steel cord with medium Cu content.
20
0
2
4
6
8
(A)
Cu
S
80
10
0
Sputter time (min)
cords are somewhat different with respect to copper in
brass-plated steel cord. The brass plated steel cord
exhibited strong dependence on the copper content, and
excellent stability in the humidity and thermal aging
treatments.
First, the pull-out force and rubber
coverage of the brass-plated steel cord were improved
with the decrease in copper content from the HC cord
to the LC cord. Second, the adhesion properties of the
LC cord which had the lowest copper content was
superior to the other brass-plated cords after humidity
or thermal aging, and the unaged adhesion properties
were superior. The adhesion properties of LC cord are
strongly related to the structure and stability of the
adhesion interphase, and the specificity of LC cord is
due to the difference in the adhesion interphase from
those of MC and HC cords.
The excessive formation of copper sulfide, which
induces its own cohesive failure leading to deterioration
of the adhesion layer, is limited by the copper content
on LC cord (Fig. 3). The primary reason for better
adhesion on LC cord with a small amount of copper
content is the suppression of excessive growth of copper
sulfide. The average content of the LC cord is 51.4
wt%, so the copper content is small enough to form
copper sulfide as a separate layer resulting in cohesive
failure.
The suppression of copper sulfide formation on the
LC cord also contributes to the excellent adhesion
stability during
humidity aging
(Fig,
4).
Dezincification of brass is accelerated in humid
Cu
S
(C)
60
40
40
40
20
20
0
0
8
16
24
32
40
Zn
O
30
20
20
0
0
8
16
24
32
40
Zn
O
30
20
10
8
16
24
32
Sputter time (min)
40
8
16
24
32
40
Zn
O
10
0
0
0
30
20
10
0
Cu
S
80
60
0
(B)
80
60
0
0
8
16
24
32
Sputter time (min)
40
0
8
16
24
32
40
Sputter time (min)
Figure 4. AES depth profiles of Cu, S (top) and Zn, O (bottom) for the
adhesion interphases of humidity aged adhesion samples between the
rubber compound and brass plated steel cord with with respect to Cu
content: (A) LC; (B) MC; (C) HC.
Humidity aging : 15 days, 85 oC and 85% relative humidity.
compound, the adhesion properties between brassplated steel cord and the rubber compound improved
with the lowering in the copper content. At the unaged
state, the adhesion properties of LC cord were superior
to those of the MC and HC cords, and after humidity
and thermal aging treatments those properties were also
superior. The exceptional adhesion properties of the
LC cord is ascribed to the suppression of excessive
growth of the copper sulfide layer which induces
adhesion failure.
The excellent stability against
humidity or thermal aging is explained by the
suppression of additional formation of copper sulfide,
and dezincification due to the lack of metallic copper
and zinc on the outer surface of the LC cord
ACKNOWLEDGEMENTS
This research was supported by Basic Science Research
Program through the National Research Foundation of
124
Korea (NRF) funded by the Ministry of Education,
Science and Technology(2011-0024214).
REFERENCES
[1] G.S. Jeon, G.S., J. Adhesion Sci. Technol, 23, 913
(2009).
[2] Hotaka,T., Ishikawa, Y. and Mori, K., Rubber
Chem. Technol., 80, 61 (2007).
[3] Jeon, G.S., J. Adhesion Sci. Technol., 19, 445
(2005).
125
126
ADHESION IMPROVEMENT OF VULCANISED RUBBER CONTAINING
NOTICEABLE AMOUNT OF PROCESSING OILS BY TREATMENT WITH
LOW PRESSURE PLASMA WITH DIFFERENT SHELVES
CONFIGURATION IN THE REACTOR CHAMBER
Beatriz Cantos-Delegido, José Miguel Martín-Martínez
Adhesion and Adhesives Laboratory, University of Alicante, 03080 Alicante, Spain
[email protected], [email protected]
ABSTRACT
Vulcanised rubber containing intentionally noticeable excess of processing oils in its formulation (L3 rubber) was
treated with RF low pressure Ar-O2 (2:1, (v/v)) plasma for achieving a satisfactory level of adhesion to waterborne
polyurethane adhesive. The effectiveness of the plasma treatment depends on of the configuration of the plasma
chamber shelves and the length of treatment. Surface modifications were assessed by ATR-IR and XPS spectroscopy,
contact angle measurements, and scanning electron microscopy (SEM). Low pressure Ar-O2 plasma treatment in direct
configuration provided the most aggressive and effective surface modification of the L3 rubber. The increase in the
length of the plasma treatment improves the extent of the surface modifications on the L3 rubber surface. However,
even important surface modifications are produced by plasma treatment, adhesion of treated L3 rubber was not
improved due to the creation of weak boundary layer at the polyurethane-rubber interface after joint formation.
Heating at 80ºC for 12 hours of the as-received L3 rubber prior plasma treatment enhances the extent of the surface
modifications and improved adhesion is obtained as the formation of the interfacial weak boundary layer is avoided.
KEY WORDS: Vulcanized rubber, processing oils, Ar-O2 low pressure plasma treatment, adhesion, waterborne polyurethane adhesive.
vulcanised ethylene propylene diene polymethylene
(EPDM) compounded with natural rubber (NR) during
co-vulcanization surface was treated with low pressure
Ar-oxygen plasma to improve its adhesion. Although
the plasma treatment changed both the surface
composition and roughness, a small increase in peel
strength was obtained. Therefore, in this study, Aroxygen low pressure plasma is used to improve
noticeably the adhesion of a vulcanised styrenebutadiene (SBR) intentionally formulated with an
excess of paraffin oil (a well-known antiadherent
moiety in rubber).
1. INTRODUCTION
Vulcanised rubber is commonly used as sole material in
shoe industry because of its lightness, comfort and
thermal insulation. However, the non-polar nature of
rubber inhibits good wettability and therefore poor
adhesion to polar polyurethane (PU) adhesive used in
shoe manufacturing is obtained. Furthermore, some
low-molecular weight additives in the rubber
formulation such as antiozonants, processing oils and
mold-release agents contribute in decreasing its surface
energy. For increasing the adhesion of vulcanized
rubber, surface treatment is required to chemically
modify the surface and produced suitable joints. The
most common and effective surface treatment for
vulcanized rubber is the halogenation with
trichloroisocyanuric acid solutions in organic solvents;
however, this treatment produces chemical residues and
the organic solvents are harmful for the environment
and unhealthy for humans. To overcome the problems
associated to the chemical surface treatment of rubber,
treatments based on cold low-pressure gas plasma have
proposed earlier1-4.
2. EXPERIMENTAL
2.1. Materials
Vulcanised SBR rubber containing intentionally double
amount of processing oils in its formulation – L3 rubber
- was provided by Cauchos Arnedo S.A. (La Rioja,
Spain). This rubber contains, among other additives,
zinc stearate, precipitated silica, antiozonant
(microcrystalline paraffin wax) and paraffin oil.
The surface treatment of vulcanised rubbers with
oxygen and air low-pressure plasmas have been shown
effective but the formation of weak boundary layer by
oils and wax migration during or after joint formation
was not completely avoided1-3. In recent study4
2.2. Surface treatment with Ar-O2 low
pressure plasma
127
RF plasma (13.56 kHz) Digit Concept NT1 (BSET EQ,
Antioch Ca, USA) equipment was used. The residual
gas pressure during plasma treatment was lower than 50
mTorr. Two different shelves configurations of the
plasma reactor – etching and direct - were studied
(Figure 1) for changing the aggressiveness of the
treatment. The power of the plasma was varied between
100 and 400 W, and the lengths of treatment were
varied between 60 and 800 seconds.
Preliminary studies of Ar-oxygen plasma treatment in
etching configuration by changing the power between
100 and 400 W were carried out. 400 W was selected
because of the greater surface modification produced in
the L3 rubber.
Two shelves configurations (direct, etching) in the
plasma reactor chamber were tested (power: 400 W) by
treating the L3 rubber with Ar-oxygen plasma for 5
minutes. Etching configuration provided the least
aggressive treatment. Surface energy values for Aroxygen plasma treated L3 rubber in configuration direct
(Table 1) increase, due to the increase in its polar
component mainly. Lower increase in the surface
energy of the plasma treated L3 rubber value is obtained
by using the configuration etching.
Table 1. Surface energy (γs), and polar (γsp) and dispersive (γsd)
component of the surface energy of as-received and Ar-oxygen plasma
treated L3 rubber (25 oC). Incidence of the shelves configuration.
Figure 1. Shelves configurations in the plasma reactor chamber.
2.3. Experimental techniques
γsp
(mJ/m2)
0.3
16.1
21.6
Treatment
The chemical composition of the nearer L3 rubber
surface (about 1 µm) was analysed by ATR-IR
spectroscopy by using Alpha spectrometer (Bruker
Optics, Etilinger, Germany) working in attenuated total
reflexion mode (Ge prism). The chemical changes
produced in the most external surface layer of the L3
rubber (2 nm) were analysed by X-Ray phoelectron
spectroscopy (XPS) using K-Alpha instrument (Thermo
Scientific, West Palm Beach, USA) using AlKα
radiation.
Abs (a.u.)
As-received
Etching
Direct
1.0 L3 RUBBER – PLASMA
POWER = 400 W
0.8 TIME = 300 s
CONF = DIRECT
0.6
γsd
(mJ/m2)
30.0
34.1
35.2
γs
(mJ/m2)
30.3
50.2
56.8
1102
963
2909
2841
0.4
698
908
727
1635
2998
1535 1448
991
802
0.2
The wettability at 25 ºC of the treated L3 rubber was
evaluated by contact angle measurement with RaméHart 100 (Netcong, NJ, USA) goniometer. Two liquids
- water (polar) and diiodomethane (non-polar) - were
used. Surface energy and its components were obtained
by using the Owens-Wendt-Kaelble approach (1).
Abs (a.u.)
0.0
1.0
L3 RUBBER – PLASMA
POWER = 400 W
0.8 TIME = 300 s
CONF = ETCHING
0.6
727
0.0
1.0
Abs (a.u.)
0.8
The surface topography of the rubbers was studied by
scanning electron microscopy (SEM) using Jeol JSM840 (Jeol Ltd, Tokyo, Japan) microscope. An electron
beam of 20 kV was employed.
2841
0.4
1468 1458
988
2947
0.2
(1)
2909
1635 1535
L3 RUBBER
As-received
908
718
698
2909
0.6
2841
0.4
727
1468 1458
2947
0.2
0.0
4000
963
1102
1635 1535
3000
2000
1102 988
718
908
1000
Wavenumber(cm-1)
Adhesive joints produced with L3 rubber treated with
low-pressure plasma pieces and aqueous polyurethane
adhesive (Desmocoll U54, Bayer, Leverkusen,
Germany) were made. T-peel tests were carried out in
Instron 4411 (Instron, Ldt., Buckinghamshire, UK)
universal testing machine after 72 hours of adhesive
joint formation and using a cross-head speed of 100
mm/min.
Figure 2. ATR-IR spectra (Ge prism) of as-received and low pressure
Ar-O2 plasma (400 W) treated L3 rubber. Incidence of the shelves
configuration in the reactor chamber.
ATR-IR spectra support the higher effectiveness in the
treatment of L3 rubber with plasma in configuration
direct with respect to configuration etching (Figure 2).
In configuration direct, an increase in the relative
intense of the bands of silica (1102 cm-1) and of the
band at 963 cm-1 corresponding to the rubber are
produced. In addition, a decrease of the relative
3. RESULTS AND DISCUSSION
128
migration of wax and/or oils to the rubber-polyurethane
interface after joint formation causing the formation of a
weak boundary layer.
intensity of the methylene bands due to paraffin oil and
wax (2909, 2841, 727 and 718 cm-1) was markedly
produced. According to these results, Ar-oxygen plasma
in configuration direct was selected for continuing the
research.
Abs (a.u)
1.0
The influence of the length of treatment (10-600
seconds) of low pressure argon-oxygen plasma (400 W)
in configuration direct on the surface modification of
L3 rubber was studied. By increasing the length of the
plasma treatment, an increase of the surface energy of
the L3 rubber and its polar component is produced
meanwhile its dispersive component does not vary
substantially (Figure 3).
0.5
L3 RUBBER - PLASMA
POWER= 400 W
TIME = 600 s
CONF = DIRECT
Abs (a.u.)
Abs (a.u.)
2841
1635
908
727
797
1102
698
2947
L3 RUBBER – PLASMA
POWER = 400 W
TIME = 60 s
CONF = DIRECT
908
727
2841
L3 RUBBER
As-received
1535
1448
2841
797
991
1468
1635
1535
1458
988
1102
727
718
908
2909
2841
0.5
727
2947
0.0
4000
698
2909
2947
0.0
1.0
963
2909
1635
0.5
1535
1448
991
L3 RUBBER / PLASMA
POWER = 400 W
TIME = 300 s
CONF = DIRECT
0.5
0.0
1.0
Abs (a.u.)
963
2909
2947
0.0
1.0
Surfaceenergy (mJ/m2)
1102
1635
3000
1468 1458
988
1102
1535
2000
Wavenumber (cm-1)
908
718
1000
Figure 4. ATR-IR spectra (Ge prism) of as-received and low pressure
Ar-O2 plasma (400 W, configuration direct) treated L3 rubber.
Incidence of the length of treatment.
Time (s)
Figure 3. Surface energy (polar component - γsp -, dispersive
component - γsd - and total -γs) of the L3 rubber (25 oC) treated with
low pressure Ar-O2 plasma in configuration direct. Incidence of the
length of treatment.
As-received
On the other hand, the ATR-IR spectra of L3 rubber
treated with low pressure argon-oxygen plasma (Figure
4) show an increase in the relative intensity of the bands
of silica and rubber, and a decrease in the methylene
bands of oil and wax, more notoriously when the length
of plasma treatment increases. Therefore, the Ar-oxygen
plasma treatment removes contaminants from the L3
rubber surface even though, low degree of surface
oxidation is obtained.
LPP - 5 min
10 µm
10 µm
LPP - 10 min
SEM micrograph of the as-received L3 rubber (Figure
5) shows the presence of wax crystallites on the surface.
The application of low-pressure Ar-O2 plasma (400 W,
configuration direct) for 5 minutes (Figure 5) removes
partially wax crystallites and produces the migration of
processing oils and the exposure of silica covered by a
thin rubber layer (noticed as bumps on the surface).
When the length of the plasma treatment is extended to
600 s (10 min) (Figure 5) both antiozonant wax and
paraffin oil are removed producing a different surface
topography with higher roughness and apparently free
of contaminants.
10 µm
Figure 5. SEM micrographs of the as-received and Ar-O2 low
pressure plasma treated (400 W, configuration direct) L3 rubber.
Influence of the length of treatment.
In order to avoid the migration of low-molecular weight
additives to the polyurethane-rubber interface, prior to
plasma treatment the L3 rubber was heated at 80 oC for
12 hours. XPS analysis of the heated L3 rubber surface
shows the creation of new carbon-oxygen moieties,
removal of carbon-rich moieties and surface oxidation
(C-O polar species appears) (Table 2). A noticeable
increase in peel strength (3 kN/m) was obtained in the
adhesive joints made with L3 rubber pretreated by
heating followed by Ar-oxygen plasma treatment (400
W, configuration direct, 800 s).
Despite the surface modifications produced on the L3
rubber surface by treatment with Ar-oxygen plasma, the
adhesive strength of treated L3 rubber/waterborne
polyurethane adhesive/treated L3 rubber was very low
(<1 kN/m). This poor adhesion could be caused by the
129
[4] Ganesh C. Basaka, G.C., Bandyopadhyayb, A.,
Neogic, S., Bhowmick, A.K. “Surface modification
of argon/oxygen plasma treated vulcanized ethylene
propylene diene polymethylene surfaces for
improved adhesion with natural rubber”. Applied
Surface Science, 257, pp. 2891–2904, 2011.
Table 2. Chemical surface composition of as-received, heated (80 oC,
12 h) and low pressure Ar-O2 plasma treated L3 rubber.
Treatment
A -receive
Thermal
Thermal
+ Ar-O2 plasm
C-H (at%)
(285.0 eV)
93.2
eV)93.6
C-O (at%)
(286.8 eV)
6.8
6.4
75.2
24.8
4. CONCLUSIONS
Paraffin wax and processing oils in the as-received L3
rubber surface unfavoured its adhesion. Different
shelves configurations in the plasma reactor chamber
determined the effectiveness of the surface treatment,
the configuration direct was the most suitable. The
increase in the length of the Ar-oxygen plasma
treatment increased its effectiveness. Even the surface
of the L3 rubber was modified by treatment with Ar-O2
plasma, an improvement in adhesion to waterborne
polyurethane was not obtained due to the formation of a
weak boundary layer at the polyurethane-rubber
interface. Finally, the extent of the surface
modifications and the adhesion of the L3 rubber were
more marked when it was heated at 80ºC for 12 hours
prior to Ar-O2 plasma treatment. These improvements
were due to the removal of antiozonant wax and
processing oils from the L3 rubber surface, an increase
of the surface energy and wettability, creation of
roughness and surface polarity (C-O groups) were
produced.
5. REFERENCES
[1] Romero-Sánchez, M.D., Pastor-Blas, M.M., MartínMartínez, J.M. “Environmental friendly surface
treatments of SBS rubber: Alternatives to the
solvent-based
halogenation
treatment”,
International Journal of Adhesion and Adhesives,
25, pp. 19-29, 2005.
[2] Ortiz-Magán, A.B., Pastor-Blas, M.M., MartínMartínez, J.M. “Surface modifications and adhesion
of vulcanized SBR rubber treated with RF plasmas
of different gases”, The Journal of Adhesion, 80, pp.
613-634, 2004.
[3] Torregrosa-Coque, R., Martín-Martínez, J.M.
“Influence of the configuration of the plasma
chamber on the surface modification of synthetic
vulcanized rubber treated with low-pressure oxygen
RF plasma”, Plasma Processes and Polymers, 8,
pp.1080-1092, 2011.
130
THE STABILITY OF SURFACE PRETREATMENTS ON DIFFERENT STAINLESS STEELS
B. Weller1, C. Kothe1, J. Wünsch1
1
Technische Universität Dresden / Institute of Building Construction
Dresden, Germany
[email protected]
ABSTRACT
The trend to the transparency just as to the dematerialization of the building envelope is of continuous interest in
architecture. Point fittings supported these topics and will therefore often used for glass facades. Novel applications of
adhesive joints represent an alternative to the currently used mechanical fixed point fittings.
Permanent adhesive joints can only be realized with special qualities of the substrate surfaces. Thereby the surface
structure, inherent moisture and contaminants and their reactivity have a decisive influence on the adhesion properties.
The surfaces of the materials and thus their adhesive behaviour can be modified by specific cleaning and surface
treatments. Due to deactivation processes on the modified surfaces, a fast bonding after the pretreatment is
recommended. Since the value chain of the manufacturing of a building envelope is distributed to multiple companies,
it makes sense to separate the pretreatment as special process from the adhesive bonding. Hence, to increase quality and
economic efficiency of the industrial production of adhesive glass fixings, a temporally and locally separation of the
individual manufacturing steps should be aimed. So a long-term stability of the surface modifications or an appropriate
surface protection has to be ensured.
Therefore, the stability of the surface pretreatment with atmospheric plasma used on different stainless steels have been
investigated before and after long term storages under specified atmospheric conditions and vacuum. The experimental
study includes contact angle measurements for the determination of the wettability and the surface energy of untreated
and pretreated as well as of unstored and stored stainless steel surfaces. Furthermore, the adhesion of pretreated and
stored specimens will be determined by the tensile-strength-test of butt joints (according to EN 15780).
KEY WORDS: surface pretreatment, long-term stability, metal joints
To increase quality and economic efficiency of the
industrial production of adhesive glass fixings, , the aim
should be a temporal and local separation of the
individual manufacturing steps. Hence, a long-term
stability of the surface modifications or an appropriate
surface protection has to be ensured.
1. INTRODUCTION
State of the art for the integration of glass elements in
buildings are point and linear fixings which are usually
designed mechanically. However, these mechanical
fixings are unsuitable for the fragile material of glass. In
particular, the mechanical point fixings require unfavorable glass processing, like boreholes which leads to
high stress concentrations and increases the risk of glass
breakage. Adhesive joints avoid these problems. A wide
range of adhesive products exists for the bonding of
glass and other materials, but for a very good adhesion
especially the quality of the substrate surfaces is
important. Thereby the surface structure, inherent
moisture and contaminants and their reactivity have a
decisive influence on the durability of the adhesive
bond. The surfaces of the materials and thus their
adhesive behaviour can be modified by specific
cleaning and surface treatments.
This study investigates the long term stability of
surfaces which were pretreated with atmospheric
pressure plasma by determination of the surface
wettability as well as the tensile strength of bonded
joints.
2. SURFACE PRETREATMENT
Due to the long life expectancy of buildings, a
permanent aging resistance of the adhesive joints is
necessary. Therefore, generating the highest possible
adherency between adhesive polymer and adherend
surface is essential. Special surface treatment processes
ensure a better wettability of the surfaces and also
create energetically active sites that can interact with the
adhesive molecules. However, many of the industrially
Due to deactivation processes on the modified surfaces,
a fast bonding after the pretreatment is recommended.
However, the pretreatment and the bonding are special
processes that cannot be realized in any manufactory.
131
established surface pretreatments, especially those for
metallic materials, are not sustainable, since these
methods often use corrosive, highly toxic and
environmentally hazardous substances. In various
industries, such as automotive, electrical engineering
and dentistry, newly developed surface treatment
methods based on plasma and deposition technologies
are already used. This raises the question of the
applicability of such methods on materials for glass
constructions and of their benefit to the optimization of
structural adhesive joints.[1]
3.3. Test specimens
To produce a tensile adhesive bond, butt-sealed test
specimens were prepared in accordance with the
standard EN 15870. For this purpose, two metal
cylinders with the nominal diameter of 24 mm were
respectively bonded to each other. The joining device
allows a vertical arrangement of the cylinders and an
offset-free adherence. The surfaces of the substrates of a
specimen were each purified and pretreated the same
way. Subsequently the adhesive was dosed on the
bottom-fixed metal cylinder and the second cylinder
was placed on top. The nominal thickness of the
adhesive layer is 1 mm.
In this study the atmospheric pressure plasma method
was used for the surface pretreatment. The plasma leads
to cleaning, neutralization and formation of chemical
reactive centers on the surfaces. These change the
surface energies of various materials by a significant
factor and thereby adhesive properties can be
improved.[2]
3.4. Characterization methods
For analyzing the wettability of pretreated surfaces,
contact angle measurements are suitable indicators. A
test liquid, e.g. water, is applied to a specimen surface
and the angle between adjacent tangent and surface is
measured. Therefore, the contact angle depends on the
energetic interaction between surface and the chemical
compositions and the topography of the solid.
Generally, small contact angles less than 30° indicate a
good wetting behaviour of the surface. Pretreatments
influence the surface energy of specimens so that the
contact angle and the wettability will change. In
addition to the measurement of contact angles, the
analyzing system allows the determination of surface
energy by using different test liquids, e.g.
diiodomethane or ethylene glycol. The surface energy is
divided into a polar and a dispersive part due to the
existence of polar and dispersive interactions between
adhesive and surface. Since only similar interactions
can take place and adhesives often have chemical
compositions with many polar functional groups a high
polar part of the surface energy of the substrate is
desired.[5],[6]
3. EXPERIMENTAL
3.1. Materials
The metals are narrowed down to four stainless steels
with general technical approval [3] and good
availability. The stainless steels with the alloys
X5CrN18-10
(material
number
1.4301)
and
X6CrNiTi18-10 (material number 1.4541) have a
moderate corrosion resistance. These two materials are
mainly used for indoor applications or in industrially
unpolluted outdoor areas. The stainless steels with the
alloys X2CrNiMo17-12-2 (material number 1.4404)
and X6CrNiMoTi17-12-2 (material number 1.4571)
have a mean corrosion resistance. These two materials
are mainly used for outdoor applications and in
structurally inaccessible areas. All investigated
materials belong to the strength class S235.[4]
The adhesive, which was used for the investigation on
the cylindrical butt joints, is the Hysol 9483 from
Henkel-Loctite (Germany). This adhesive is a twocomponent epoxy resin system. The visual appearance
is high transparent as well as colourless. The
components of the adhesive were mixed in the
proportions specified by the manufacturer. The curing
process lasted 2 hours at a temperature of 60 °C.
For the investigation of the long-term stability of
plasma pretreated surfaces, they were stored under
various environmental conditions. Long term storages
were performed under specified atmospheric conditions
(23°C and 30% relative humidity) and under vacuum.
After defined periods of time specimens were removed
and their surface energies were determined by contact
angle measurements. In addition their adhesion
properties were determined by the tensile-strength-test
of butt joints.
3.2. Surface pretreatment
The metallic surfaces were pretreated with atmospheric
pressure plasma. For plasma generation a gas flow is
ionized by high-voltage discharge in the plasma jet and
then directed on specimen surface. For this pretreatment
the Plasmatreater 400 from Plasmatreat GmbH was
used. The process parameters of jet distance and speed
were selected to 10 mm and 50 m/min.
The destructive testing with the butt-jointed samples
was performed under the standard climate 23/50 (ISO
291). The samples were loaded with a constant strain
rate to ultimate tensile strength. The strain rate was
selected to 25% per minute, referring to the nominal
thickness of the adhesive layer. Each test run comprises
of five samples. The nominal tensile stresses were
analyzed.
132
results are shown in Figure 3 and Figure 4 using the
example of stainless steels 1.4301 and 1.4404.
4. RESULTS
4.1. Wettability
With the atmospheric pressure plasma process surface
cleaning and formation of active sites can be achieved.
The cleaning effect of the plasma has been already
studied else-where and the mechanism of the removal of
contamination could be explained with oxidative
reactions. Also, the use of this pretreatment for
improving the wettability could be shown there (see
also Figure 1). Here, however, a very accurate and
substrate-dependent selection of device parameters, e.g.
plasma jet distances and speeds, must be made.[1],[7]
Figure 3. Surface energy on stainless steel 1.4301
before and after surface pretreatment and storage.
Figure 1. Water drop on an untreated (left) and
a plasma-treated (right) metal surface.
The surfaces of the investigated stainless steels showed
similar surface energies of about 40 mN/m after
cleaning (Figure 2). Between the different metals there
were only small differences in the dispersive and polar
fractions.
Figure 4. Surface energy on stainless steel 1.4404
before and after surface pretreatment and storage.
The surfaces of the two different stainless steels showed
a very similar behavior. The surface energies decreased
during the storages in dependence of storage periods.
This occurs under atmospheric condition as well as
under vacuum. It should be noted that the decrease in
surface energy after 2 hours of storage was about the
same under both storage conditions. This is due to that
the contact angle measurement were carried out under
atmospheric conditions and therefore the samples stored
under vacuum could also react for a short time with the
ambient air and humidity. This time-dependent effect
affects not so much for the specimens which were
stored for 24 hours. In contrast, the decrease of the
surface energy was significantly lower for those
specimens which were stored under vacuum than for
those stored under atmospheric conditions. This applies
particularly to the polar fraction of the surface energy.
Especially this part is important for bonding processes.
Figure 2. Surface energy on different stainless steels
before and after surface pretreatment.
After the atmospheric pressure plasma process a
significant improvement of the wettability was achieved
on all stainless steels. Here almost identical surface
energies resulted with equal distributions of polar and
dispersive fraction.
4.2. Long-term stability of surface pretreatment
4.3. Tensile-strength-test
For the determination of the long-term stability of the
surface pretreatment the specimens were stored under
specified atmospheric conditions (23°C and 35%
relative humidity) or under vacuum for 2 and 24 hours.
Afterwards the surface energies were measured. The
The adhesion properties of pretreated and stored
specimens should be examined by the tensile-strengthtest of butt joints. There were bonded five specimens
for each of the different pretreatment and storage
133
conditions. However, the tensile strengths varied
widely, so that large standard deviations occurred in the
results. Thereby no significant differences between the
individual test series could be observed. This is shown
in Figure 5 using the example of stainless steel 1.4301.
continued with increasing the sample size and the
selection of an additional adhesive to achieve evaluable
results.
ACKNOWLEDGEMENTS
We thank the German Federal Ministry of Education
and Research (BMBF) for funding these studies in the
research project GLASKONNEX Transfer.
REFERENCES
[1] Kothe, C., „Oberflächenvorbehandlung von Fügeteilen zur Optimierung adhäsiver Verbindungen im
Konstruktiven Glasbau - Surface Modification
Methods for Improving Adhesive Joints in Glass
Structures“, Ph.D. Thesis, Technische Universität
Dresden, Germany.
Figure 5. Tensil stress of bonded specimens after
surface pretreatment and storage with the indication of
the standard deviation.
[2] Lommatzsch, U. et al., “Atmospheric pressure
plasma jet treatment of polyethylene surfaces for adhesion improvement”, Plasma Process. Polym. 4, pp.
1041-1045, 2007.
Obviously, the bonding process with the selected
adhesive was not reproducible enough to obtain
analogous results as from the wettability tests. In
addition to the maximum tensile strength also the type
of fracture was recorded in the investigations. It was
either a breakage with adhesive and cohesive fractions
or a complete cohesive failure. Here too, the results
were very different from each other within the
individual test series. No relationship between the
failure mode and height of tensile stress could be found.
These results can be explained with an irregular stress
distribution in the adhesive joint.
[3] Z-30.3-6, April 2009, “Erzeugnisse, Verbindungsmittel und Bauteile aus nichtrostenden Stählen.
Allgemeine
bauaufsichtliche
Zulassung“,
Düsseldorf, Informationsstelle Edelstahl Rostfrei,
2009.
[4] Dehn, F., König, G., Marzahn, G., „Konstruktionswerkstoffe im Bauwesen“, Berlin, Ernst & Sohn,
2003.
[5] Habenicht, G., „Kleben: Grundlagen, Technologien;
Anwendungen“, Berlin: Springer, 2006.
5. CONCLUSION
[6] Thomsen, F., „Benetzung und Haftung normgerecht
messen“, Adhäsion 57 (10), pp. 26-29, 2013.
The atmospheric pressure plasma causes good wettable
surfaces at all investigated stainless steels. Independent
of the material almost identical surface energies were
achieved after the pretreatment. On the basis of these
results the time-dependent stability of the plasma
activation was closer examined. It could be
demonstrated by analysis of the surface energy that it is
possible to reduce the deactivation by storage of
pretreated specimens under vacuum. For a more
accurate understanding of the deactivation, the changes
in the chemical composition of pretreated surfaces
during storage have to be investigated in further studies.
[7] Weller, B., Kothe, C., “Investigation of surface
modification methods to improve adhesive joints in
glass construction”, Proceedings of Glass Performance Days 2011, pp. 677-680, Tampere, 2011.
To investigate the adhesive properties of the pretreated
surfaces in dependence to the storage conditions, buttsealed test specimens were prepared and subjected to a
tensile test. Consistently high tensile stresses were
achieved, which shall be evaluated as very positive for
the application in glass constructions. However, the
obtained results did not allow more precise conclusions
for the research question. Therefore, these tests are
134
OPTIMISING INTERFACIAL STRENGTH BETWEEN STEEL AND EPOXY
RESIN THROUGH ATMOSPHERIC PLASMA AND WET CHEMISTRY
TREATMENTS
Amit Kumar Ghosh1 +, Ellen Bertels3 +, Gabriella Da Ponte4 +, Danny Van Hemelrijck2, Bart Goderis3, Bert
Verheyde4, Bruno Van Mele1
1
Department of Physical Chemistry and Polymer Science & 2Department of Mechanics of Material and Constructions,
Vrije Universiteit Brussel, Pleinlaan 2, Brussels, Belgium, [email protected].
3
Department of Polymer Chemistry and Materials, KU Leuven, Heverlee, Belgium
4
VITO NV, Mol, Belgium.
+
SIM researcher, SIM-Flanders vzw, Technologiepark, Zwijnaarde, Belgium
ABSTRACT
This paper presents the improvement of adhesion strength of a stainless steel and epoxy hybrid material by silane
surface treatment. Two different routes, a wet chemical dipping process and an atmospheric plasma route were
chosen to deposit the silane on the stainless steel surface. The conditions of both wet chemical dipping process and
atmospheric plasma deposition were optimised to get the most optimised deposition condition. Pull-off testing (dolly
testing, in accordance with ISO 4624:2002) was used to evaluate the interfacial strength of different surface-treated
samples with an epoxy resin. In both cases, the wet chemical dipping and the atmospheric plasma route, a huge
improvement of the interfacial strength of the stainless steel-epoxy system was observed after optimisation of the
surface treatment conditions.
KEY WORDS: Stainless steel, epoxy resin, silane, pull-off testing, interfacial strength.
get the most optimised deposition condition. This
includes the cleaning procedure of the stainless steel
substrates, the silane solution concentration, the
dipping condition in silane solution, the rinsing
condition of the surface treated samples and the
condensation condition of the silane layer (time,
temperature of the oven and effect of vacuum). Also the
process conditions of the atmospheric plasma treatment
were optimised. This included changing the type of
precursor, the plasma powers, and the number of
passes. Pull-off testing (dolly testing, in accordance
with ISO 4624:2002) was used to evaluate the
interfacial strength of different surface-treated samples.
1. INTRODUCTION
Stainless steel is one of the mostly used materials in
fields like construction, transportation and process
industry due to its excellent mechanical properties,
corrosion resistance and recyclability [1]. Recently, it
gained a lot of interest for composite applications.
These hybrid materials offer: higher impact resistance,
higher strength-to weight ratio, resistance to corrosion,
better dimensional stability, and greater design freedom
[2]. Apart from its superior mechanical properties, one
of the drawbacks of stainless steel for polymer
composite application is that it has comparatively poor
adhesion strength towards epoxy resin and leads to an
adhesive failure [3]. Therefore it is very important to
improve the interface for a better performance of steelepoxy hybrids [4]. In this work, silane coupling agent
based surface treatment of stainless steel was studied to
see the effect of surface treatment on the adhesion
strength of stainless steel [4]. The properties and
quality of silane layers are strongly dependent on the
deposition conditions, like the preparation of the silane
solution and the condensation conditions [5]. Proper
control of the deposition conditions is very important to
get the desired properties from the silane layer. In this
work, γ-aminopropyltriethoxysilane (γ-APS) was
chosen to modify the stainless steel surface. Two
different routes, a wet chemical dipping process and an
atmospheric plasma route were chosen to deposit the
silane on the stainless steel surface [6]. The conditions
of the wet chemical dipping process were optimised to
2. EXPERIMENTAL DETAILS
2.1 Materials:
Coupling agent used was γ –APS, supplied by SigmaAldrich with 99% purity. The metal substrate was 0.8
mm thick 304 type (mirror polished) stainless steel and
it was cleaned ultrasonically (fifteen minutes in an
ethanol bath) before the sample preparation. The epoxy
resin was diglycidyl ether of bisphenol A (DGEBA)
based resin EPIKOTE 828 LVEL, supplied by
HEXION chemicals and the the diamine curing agent
was 1,2 Diamino Cyclohexane (DAC) supplied by
Sigma-Aldrich chemicals. The resin was prepared by
mixing epoxy and amine stoichiometrically.
Aluminium dollies of 20 mm in diameter were used to
prepare the dolly sample for pull-off testing.
135
glass beads. The curing cycle of the epoxy-amine
system was overnight at room temperature, followed
with a partial cure at 80°C for two hours and
afterwards a final cure at 180°C for one hour. The
thickness of the coating after cure was about 0.2 mm.
One of the specimens of the dolly samples was shown
in figure 2.
2.2 Surface treatments
2.2.1 Atmospheric plasma treatment:
Coating experiments were carried out in a typical
parallel plate dielectric barrier discharge (DBD)
atmospheric pressure reactor shown in figure 1. One of
the most distinctive feature of this apparatus is the
mobility of the upper high voltage electrode which
enables the deposition of homogeneous coating on the
samples, placed on the botton grounded electrode. The
gap between the electrode was fixed at 2 mm in order to
assure stable plasma operation. The deposition process
was performed by using nitrogen as main gas: nitrogen
is injected in the palsma reactor as carrier gas to ignite
and sustain the discharge but it is also used to generate
the aerosol of the organic precursor by using an home
made atomizer based on the Venturi’s principle. After
plasma coating, the samples were goes throgh a
thermal annealing process (four hour at 100°C).
Figure 2: Dolly-plate-dolly sample and dolly sample preparation
setup (respectively)
2.3 Adhesion strength measurement (Dolly testing)
The pull-off adhesion tests were performed in Instron
5885H (with a jaw speed of 0.5 mm/min) in accordance
with ISO 4624. The testing methodology was modified
for the thin plate stainless steel plate samples to avoid
the effect of bending and stress concentration during
testing by giving geometric symmetry. The testing
scheme is shown in the figure 3. The aluminium dolly
was pulled vertically away from its stainless steel plate
substrate where the lower jaw was kept fixed. All the
samples were tested in room temperature condition.
The load at break was recorded after each testing. Six
measurements were performed for each sample to
calculate the average. The stress at break was
calculated by using the equation 1.
S = L * 4 / (π * D2)
(1)
Where S is stress at break (in MPa), L is the load at
break (in Newton) and D is the diameter of dolly (in
mm).
Figure 1: Schematic representation of the atmospheric plasma setup
2.2.2 Wet chemical deposition technique:
Silane solution was prepared by mixing γ-APS into a
water/ethanol mixture solvent. The ratio of water to
ethanol was set to 90/10. After that, the solution was
kept for thirty minutes for the full hydrolysis of the
silane. This hydrolysis process was established after
carrying out several experiments and by following the
hydroxyl kinetics of silanes by H-NMR. This silane
solution was applied on the steel surface by the wet
deposition method. The cleaned stainless steel sample
was immersed immediately into the silane solution to
avoid recontamination of the surface. After that it went
through a rinsing process in an ethanol bath. Finally,
the silane layer was condensed in a vacuum enabled
oven.
Threaded
steel screw
Al – dolley
(Dia. 20 mm)
Sample plate
Epoxy film
(apx. 200 µm)
Al – dolley
(Dia. 20 mm)
Threaded
steel screw
Fixed
Figure 3: Schematic representation of the dolly-plate-dolly testing
methodlogy and the actual view of the test setup
2.3 Dolly-plate-dolly sample preparation:
3. RESULT AND DISCUSSIONS
For adhesion strength measurement dolly samples were
prepared by using aluminium dollies and the stainless
steel samples. The epoxy resin adhesive layer was first
applied on the surfaces of the aluminium dollies
(surfaces of the dollies were sandblasted) and the
sheets, and joints were stuck together and held in a
place by specially designed sample preparation setup
(figure 2). The coating thickness was maintained by the
3.1 Optimisation of wet chemistry deposition
3.1.2 Optimisation of the condensation condition:
To optimise the condensation condition, three different
temperature (50°C, 70°C, 90°C) and three different
time interval (30 minutes, 1.5 hour and 4 hour)
136
combinations were studied to find the most suitable
condensation condition. Effect of vacuum on
condensation was also studied (sample denotation:
‘NV’ and ‘VAC’ were used for the samples which were
condensed without vacuum or with vacuum
respectively). Other deposition conditions like, silane
solution concentration (2%), dipping time (30 seconds),
rinsing condition (brief rinse- rinsing for 4-5 seconds)
were kept constant for the sample preparation. The
result is shown in figure 4. From the results, it can be
conclude that temperature of the condensation
condition has a strong effect on the adhesion strength
and also vacuum has a positive influence. It is clear
from the results that most favourable condensation
condition is 1.5 hour at 70°C in vacuum). This
condition was used for further optimisation of the other
deposition condition.
Stress at break (MPa)
50
2% Silane
5% Silane
30
20
10
in
m
5
m
3
1
m
in
in
Br
ie
f
se
c
15
se
Br
ie
15 f
se
c
1
m
in
3
m
in
5
m
in
rin
No
Br
ie
15 f
se
c
1
m
in
3
m
in
5
m
in
0
Figure 5: Effect of silane concentartion and rinsing time on adhesion
strength
3.1.4 Optimisation of dipping time:
In this work, different dipping time like, ten seconds,
thirty seconds and three minutes were used to find the
effect of dipping on the final adhesion value, while
other conditions like, silane solution concentration
(2%), rinsing condition (one minute of rinsing),
condensation condition ( 1.5 hour at 70°C in VAC)
were kept constant. The result is shown in figure 6.
From the results, it can be conclude that the shorter
dipping time is not favourable and also longer dipping
time is not helpful either. The optimum dipping time is
thirty seconds.
40
30
20
10
50
Bl
an
k
°C
-N
50 V- 3
°C 0M
50 -V - I N
°C 3 0
-N M
IN
50 V-1.
°C 5 H
50 V -1 r
°C .5
-N hr
50 V- 4
°C
H
-V r
-4
Hr
70
°C
-N
70 V- 3
°C 0M
70 -V - I N
°C 3 0
-N M
IN
70 V-1.
°C 5 H
70 V -1 r
°C .5
-N hr
70 V- 4
°C
H
-V r
-4
Hr
90
°C
-N
90 V- 3
°C 0M
90 -V - I N
°C 3 0
-N M
IN
90 V-1.
°C 5 H
90 V -1 r
°C .5
-N hr
90 V- 4
°C
H
-V r
-4
Hr
0
50
Figure 4: Effect of condensation condition on adhesion strength
3.1.3 Optimisation of silane concentration and rinsing
time:
40
30
20
10
0
In this study, different silane concentrations (1%, 2%
and 5%) were used to find the most optimised silane
concentration. For optimisation of rinsing condition,
the samples were rinsed in an ethanol bath for different
time interval like, no rinsing at all, brief rinsing (rinse
for four to five seconds), fifteen seconds of rinsing, one
minute of rinsing, three minute of rinsing and five
minute of rinsing to see its effect on the final adhesion
towards epoxy resin of the different samples, while
other conditions like, dipping time (30 seconds),
condensation condition ( 1.5 hour at 70°C in VAC)
were kept constant. The result is shown in figure 5. We
found that the optimum rinsing condition is one minute
rinsing in ethanol bath. The effect of silane
concentration is also very prominent from the figure 5.
It can be conclude that the optimum silane
concentration is 2%, where 1% silane concentration
might be giving incomplete covering of the surface and
5% silane concentration gives a thicker and non
uniform coating.
10 sec
30 sec
3 min
Figure 6: Effect of dipping time on adhesion strength
3.2 Optimisation of Atmospheric plasma deposition
3.2.1 Optimisation of precursor:
Two different types of precursor, the pure silane and
the pre-hydrolysed silane (APS+1% water) were
studied. The other conditions like, plasma power
(350W), number of passes (6 pass) were used for
sample preparation. The result is shown in figure 7. It
can be conclude that the most favourable precursor is
pre-hydrolysed silane. This precursor was used for
further optimisation of the other process parameters.
20
18
S tress at break (MP a)
1% Silane
40
16
14
12
10
8
6
4
2
0
Blank
APS
APS+1% water
Figure 7: Effect of type of precursour on adhesion strength
137
3.2.2 Optimisation of plasma power:
Effect of plasma power on the adhesion strength was
studied. Different level of plasma power, 350 W and
200 W were used for the sample preparation. While,
the other conditions like, plasma precursor (APS+1%
water), number of passes (6 pass) were kept constant.
The result is shown in figure 8. From the result, it can
be notice that adhesion strength is increased with
lowering of the power and the most favourable plasma
power is 200W. At higher level of plasma power, the
plasma might be destroying the functionality of the
silane, which could be the possible reason for this lower
adhesion values at high plasma power condition.
Blank
Atmospheric plasma
Figure 10: Faliure pattern of different dolly samples
4. CONCLUSIONS
Stainless steel surface was successfully modified by
silane treatment. Both, wet chemical deposition
technique and atmospheric plasma deposition
conditions were optimised. In both cases, a huge
improvement of the interfacial strength of the stainless
steel-epoxy system was observed after optimisation of
the surface treatment conditions. Both treatments show
a similar level of improvement of the interfacial
strength of the steel-epoxy system but they have their
specific advantages or disadvantages.
25
Wet chmistry
20
15
10
5
0
Blank
350W
200W
ACKNOWLEDGEMENTS
Figure 8: Effect of plasma power on adhesion strength
SIM-Flanders is acknowledged for the financial
support.
ArcelorMittal Global R&D Gent - OCAS NV is
acknowledged for providing steel plate samples.
3.2.2 Optimisation of number of passes:
In this study, the number of passes was optimised.
Different numbers of passes were used for the sample
preparation. While, the other conditions like, plasma
precursor (APS+1% water), plasma power (200 W)
were kept constant. The result is shown in figure 9. It
was found that most optimised number of passes is four
pass, where two passes might be giving incomplete
covering of the surface and six passes or more passes
gives a thicker and non uniform coating.
REFERENCES
[1] Davis, J. R., “Stainless Steels”, ASM International,
Ohio, 1999.
[2] Honkanen, M., Hoikkanen, M., Vippola, M.,
Vuorinen, J., Lepistö, T., “Metal-plastic adhesion in
injection-molded hybrids”, Joournal of Adhesion
Science and Technolgy, 23, pp. 1747-1761, 2009.
50
45
40
35
30
[3] De Morais, A.B., Pereira, A.B., Teixeira, J.P.,
Cavaleiro, N.C., “Strength of epoxy adhesive-bonded
stainless-steel joints”, International journal of adhesion
& adhesives, 27, 679-686, 2007.
25
20
15
10
5
0
Blank
6 pass
4 pass
2 pass
[4] Possart, W., “Modern state of models for
fundamental adhesion – a review extended abstract”,
Adhesion and Interface, 3(1), 43-51, 2002.
Figure 9: Effect of number of passes on adhesion strength
3.3 Failure pattern of dolly samples
[5] Plueddemann, E. P., “Silane Coupling Agents”,
Plenum press, New York, 1991.
Failure pattern of different samples were recorded by a
digital camera and analysed further. The failure
patterns of different sample are shown in figure 10. It
can be seen that in case of blank/unmodified surface it
is purely an adhesive failure as the interface is the
weakest link. However, when the interface was
modified by the surface treatment it goes through a
more cohesive or mix failure mode. Both, wet
chemistry deposition and atmospheric plasma
deposition shows similar kind of failure pattern after
the optimisation.
[6] Heyse, P., Dams, R., Paulussen, S., Houthoofd, K.,
Janssen, K., Jacobs, P. A., Sels, B. F., “Dielectric
Barrier Discharge at Atmospheric Pressure as a Tool to
Deposit Versatile Organic Coatings at Moderate Power
Input”, Plasma Process and Polymers, 4, pp.145157,2007.
138
Low‐power laser ablation of metal substrates
for enhanced adhesive bonding
Giovanni Chiodo2, Marco Alfano1, Stefano Pini2, Alessandro Pirondi3, Franco Furgiuele1, Roberto Groppetti3
1
Department of Mechanical, Energy and Management Engineering, University of Calabria,
Via P. Bucci, 44C, 87036 Rende (CS), Italy
2
Università di Parma - Centro Interdipartimentale SITEIA.PARMA
Parco Area delle Scienze 181/A, 43124 Parma, Italy
3
Università di Parma - Dipartimento di Ingegneria Industriale
Parco Area delle Scienze 181/A, 43124 Parma, Italy
ABSTRACT
The present work addresses the bond toughness of epoxy bonded joints with laser ablated metal substrates. An
ytterbium-doped pulsed fiber laser was employed to perform low power laser ablation on AA6082-T4 aluminium alloy
substrates. Surface profilometry, scanning electron microscopy (SEM) and X-Ray photoelectron spectroscopy (XPS)
were employed to evaluate morphological and chemical modifications induced by the laser treatment. Double
Cantilever Beam (DCB) specimens were prepared and tests carried out in order to evaluate the bond toughness of
adhesive joints. Results are compared with those obtained on samples prepared using a classical surface degreasing
treatment. An almost three-fold increase in fracture toughness was recorded with respect to samples with degreased
substrates.
KEY WORDS: adhesive bonding; laser treatment; fracture toughness.
shear joints [10]. The enhancement of joint strength was
associated to the increased mechanical interlocking
between the adhesive and the substrates and to the
improvement of chemical bonding between adhesive
molecules and the surface oxide, respectively.
The present work aims to extend the scope of the
previous study and to assess the variation in bond
toughness. Low power laser ablation was herein carried
out on aluminum (AA6082-T4) substrates using an
ytterbium pulsed fiber laser (LaserPoint YFL 20P). The
morphological and chemical modifications induced by
the laser process were examined by means of surface
profilometry, scanning electron microscopy (SEM) and
X-Ray photoelectron spectroscopy (XPS). Mechanical
tests were carried out on Double Cantilever Beam
(DCB) samples. The DCB test was selected because the
samples can be easily manufactured and accurate data
reduction schemes for the determination of fracture
toughness are available [11]. The obtained results
shows that low power laser ablation can effectively be
employed to increase the toughness of Al/epoxy joints.
1. INTRODUCTION
Adhesive bonding is widely employed for several
structural applications, ranging from microelectronics
devices to civil infrastructures due to the welldocumented advantages provided over the traditional
joining technologies [1-2]. However, the strength of
adhesive joints strongly depends on surface preparation
and surface conditions. Mechanical grinding or
sandblasting have been employed in order to increase
surface roughness and joint strength [3]. Chemical
etching can even provide better performances, because
it promotes the formation of rough fibrillated surface
improving mechanical interlocking [4]. Finally,
anodizing processes affect adhesion and durability by
mean of an electrochemically formed porous oxide film
that improves chemical interaction between the
adhesive and the adherends [5]. Additional works has
also focused on laser irradiation with excimer [6] or
solid state lasers [7]. More recently, fiber lasers, e.g.
ytterbium (Yb) fiber laser, have also been employed [8].
A pulsed laser surface pre-treatment can induce a
beneficial action on the strength of adhesive joints. It
allows to remove contaminants and produces favorable
changes in both the chemical composition and the
morphology of the surface [8-9-10].
Previous work by the authors showed that low power
pulsed laser irradiation of steel and Al substrates can
lead to improved mechanical performance of single lap
2. MATERIALS AND METHODS.
2.1. Materials
The specimens analyzed in this work are Double
Cantilever Beams (DCBs) composed by aluminum alloy
(AA6082-T4) substrates and general purpose epoxy
139
adhesive for industrial applications (Hysol 9466,
Henkel, Germany). Substrates were cut at a length L =
120 mm, width b = 25 mm and thickness h = 6 mm.
Young’s modulus and Poisson’s ratio for a AA6082-T4
aluminum alloy are E = 70 GPa and υ = 0.29. The
adhesive properties were obtained from [12].
assessment, as defined in ISO/FDIS 25178-2. In
particular, the Sk field parameter (core height) was
chosen as a measure of the surface roughness. Sk
represents a measure of the nominal roughness and may
be used to replace the mean roughness, Ra, because it is
a more robust parameter, especially when localized
singularities (e.g., very high peaks or deep pits) are
present on the surface.
XPS measurements were carried out in an ultra-high
vacuum (UHV) chamber equipped for standard surface
analysis with a base pressure in the range of 10-9 torr.
Non-monochromatic Mg-Ka X-ray (hυ=1253.4 eV) was
used as excitation source. The XPS spectra were
calibrated with the C1s peak of a pure carbon sample
(energy position 284.6 eV). All XPS spectra have been
corrected with analyzer transmission factor and the
background was subtracted using the straight line
subtraction mode. Moreover, the XPS data were fitted
assuming a Gaussian distribution for high resolution
analysis.
2.2. Surface pre-treatments
Laser ablation of aluminum alloy substrates was carried
out using a LaserPoint YFL 20P ytterbium fiber laser
operated in pulsed mode (1055–1070nm wavelength,
20–80 Hz pulse repetition frequency, 120ns minimum
pulse Full Width at Half Maximum, FWHM). The laser
is able to achieve continuous power levels up to 20W
with adjustable line spacing (LS) and feed rate (V). The
LS and V parameters can be adjusted by controlling the
X-Y stage where the sample under treatment is housed.
In this work, LS and V have been set equal to 60 mm
(50% larger than spot size) and 5mm/s (maximum feed
rate of the system), respectively. Laser power was set
equal to 18W because this value provides the best
performance in terms of surface morphology and
surface chemistry [9]. A summary of the laser process
parameters employed herein is given in Tab. 1. Note
that the process was carried out at room temperature
and in atmospheric environment.
2.4. Joint preparation and mechanical testing
The specimens analyzed in the paper are Double
Cantilever Beam (DCB). Joints preparation and testing
were carried out according to the ASTM D 3433-99
standard. After surface preparation, the adhesive was
applied to substrate surfaces and an un-bonded area was
introduced using a Teflon film. Adhesive bond-line
thickness was set equal to t = 0.3 mm (representative of
practical applications) using a brass foil as spacer.
Specimens were cured at 70° for 30 minutes. An initial
pre-crack a0 = 25 mm was subsequently created by
means of repeated loading-unloading cycles. Tests were
performed at room temperature using a servohydraulic
universal testing machine (MTS 810). The crack mouth
opening displacement (δ) was monitored using a clipgage.
The strain energy release rate was evaluated using the
following equation [13]
Table 1. Laser processing parameters.
Parameter
Laser nominal power
Emission wavelength
Emission line width
Mode of operation
Pulse frequency range
Minimum pulse width
FWHM
Beam quality factor, M2
Laser beam mode
Spot size
[kHz]
[ns]
Value
1-20
1055-1070
< 10
pulsed
20-80
120
[µm]
1.8
TEM00
40
[W]
[nm]
[µm]
G
2
P 2 dC Pa  
1 
1 


2b da
bEI 
a 
2.3. Analysis of surface morphology and chemistry
2
(1)
In order to analyze surface morphological
modifications, SEM analyses (Cambridge Stereoscan,
20 keV electron beam, current 3.4 mA, spot size
1.4mm2) were carried out on as-produced and pretreated samples. Moreover, contact profilometry was
also employed using a SM RT-150 stylus-based
profilometer in order to perform a quantitative
assessment of surface morphology. Each specimen was
sampled 12 times. Each resulting data set, consisting of
a height map sized 64x64 points with uniform spacing
(dx=dy=12.5 mm), was leveled by subtraction of the
least-squares mean plane and used to evaluate the
principal 3D field parameters for surface finish
C

P

2 t
E a' b
2

2
3
1  2 a   2 a   3  a  


(2)
 
4
4k
6 E'a
 3
EI h t E
(3)
where a is the crack length, P is the load, E the Young’s
modulus and I is the second moment of area of the
beam section, Ea’ the plane strain Young’s modulus of
the adhesive. Therefore, Eq. (1) allows one to determine
the fracture energy of a test sample when the evolution
140
of crack growth is known from Eq. (2), which is in turn
fed by compliance values measured at given points
during the test by unloading-reloading.
3.2. Fracture toughness tests
DCB tests results are summarized in Fig. 2.
Specifically, the average strain energy release rate (G)
is reported for laser-treated and degreased samples.
Notice that six degreased and three laser treated
specimens were tested. Laser ablation consistently
improved fracture toughness of the joints with respect
to simple degreasing. Indeed a +170% increase was
observed. Moreover, it should be
noted that a
significantly lower scatter was recorded in the
experimental results for the laser-treated specimens.
3. RESULTS
3.1. Analysis of surface morphology and chemistry
During the laser irradiation process a fraction of the
laser beam energy is absorbed, promoting the removal
of material by vaporization and erosion. In order to
analyze surface morphological modifications, contact
profilometry, as well as scanning electron microscopy,
were performed. SEM images of as produced and laser
irradiated (@18 W) aluminum substrate are shown in
Fig. 1. It is quite apparent that the low power laser
ablation significantly affects surface morphology.
Fig. 2. Strain energy release rate for degreased and laser-treated
DCB speciemen. Standard deviation is also shown for both specimen
sets.
Fig. 1. SEM images of as produced (ap) and laser treated (18W)
substrates.
Fracture surfaces have been also observed by means of
scanning electron microscopy. Typical cohesive
fractures occurred in laser treated specimens (Fig.3).
The whitening of fracture surfaces indicates a high
degree of inelastic deformation, which may be
attributed to the enhancement of mechanical
interlocking associated to the surface asperities induced
by laser ablation, and shown in Fig. 1. Several voids are
also clearly visible on the adhesive.
The survey spectrum for the as-received sample showed
that the surface is characterized by the presence of
aluminum, carbon and oxygen. Contaminants such as
tungsten and fluorine were also observed. Laser
irradiation at power levels up to 11 W was not able to
remove the contaminants from the surface. However,
when the power level achieved values equal o or higher
than 12W, surface contaminants were no longer
detected. This result is further demonstrated by the
quantitative chemical analysis reported in Tab. 3. It is
shown that the oxygen content decreased with laser
intensity, while the aluminum percentage increased,
thereby confirming the ability of the laser irradiation
process to clean the substrate surface.
Table 3. Surface chemical composition.
Substrate
AA 6082-T4
aluminum
alloy
As produced
Laser 8W
Laser 18W
Element (at. %)
Al Fe O
C
29.2 29.7 39.8 -
Cr W
17.5 49.6 19.3 47.6 12.4 47.8 -
F
1.5 2.2
1.3 2.1
ND ND
Fig. 3. Typical fracture surface of laser ablated speciemen observed
using SEM.
On the other hand, mechanical interlocking was not
obersved in as produced specimens since adhesive
fracture occurred in all samples tested. A typical
141
fracture surface for this batch of samples is given in
Fig.4.
[3] Baldan, A., (2004). J. Mat. Sci. 39(1).
[4] Critchlow, G.W., Brewis, D.M., (1996)
Adhesion and Adhesives 16(4), 255.
Int. J.
[5] Spadaro, C., Dispenza, C., Sunseri, C., (2006). J.
Physics Cond. Matter 18(33), S2007.
[6] Palmieri, F. L., Watson, K. A., Morales, G.,
Williams, T., Hicks, R., Whol, C. J., Hopkins, J. W.,
Connel J. W. (2012). Int. SAMPE Tech. Conf.,
Baltimore, USA.
[7] Langer ,M., Rechner, R., Thieme, M., Jansen, I.,
Beyer, E. (2012). Solid State Sci. 14, 926-935.
Fig. 4. Typical fracture surface of a specimen prepared using
classical surface degreasing observed using SEM.
4. CONCLUSIONS
WORKS
AND
[8] Alfano, M., Ambrogio, G., Crea, F., Filice, L.,
Furgiuele, F. (2011). J. Adhes. Sci. Tech. 25, 1261–
1276.
a
[9] Alfano, M., Lubineau, G., Furgiuele, F., Paulino, G.
H., (2012). Study on the role of laser surface
irradiation on damage and decohesion of Al/epoxy
joints", Int. J. Adhes. Adhes. 39, 33-41.
FUTURE
[10]Alfano, M., Pini, S., Chiodo, G., Pirondi, A.,
Furgiuele, F., Groppetti, R., in press. Surface
Patterning of Metal Substrates Through Low Power
Laser Ablation for Enhanced Adhesive Bonding, J
Adhesion.
The present work focused on the evaluation of bond
toughness of aluminum/epoxy bonded joints with lasertreated substrate. An ytterbium-doped pulsed fiber laser
was employed to perform low power laser ablation on
aluminum alloy substrates. The morphological and
chemical modifications induced by the laser process
were examined using surface profilometry, scanning
electron microscopy and X-Ray photoelectron
spectroscopy. Finally, mechanical tests were conducted
on Double Cantilever Beam (DCB) samples. The results
obtained indicated that low power laser ablation can
increase in a efficient manner the fracture toughness of
epoxy adhesive bonded joints. An almost three-fold
increase in fracture toughness was recorded with respect
to samples with degreased substrates.
[11]Alfano, M., Furgiuele, F., Pagnotta, L., Paulino,
G.H., (2011). Analysis of fracture in aluminum
joints bonded with a bi-component epoxy adhesive
J Test and Eval 39(2), 2753.
[12]Pirondi, A., Fersini, D., Perotti, E., Moroni, F.,
(2007). Applicabilità del modello di zona coesiva in
simulazioni della frattura per diverse geometrie di
giunti incollati, XIX Italian Group on Fracture
Conference, Milan, Italy.
[13]Krenk, S., (1992). Energy Release Rate of
Symmetric Adhesive Joints, Engineering Fracture
Mechanics.
ACKNOWLEDGEMENTS
The work of Giovanni Chiodo was partially supported
by the Emilia-Romagna Region, Italy through the grant:
POR FSE 2007-2013.
REFERENCES
[1] Kinloch, A.J., (1986). Adhesion and Adhesives,
Science and Technology. Chapman & Hall, London,
UK.
[2] Adams, R.D., (2010). Adhesive Bonding. Science,
Technology and Application. Woodhead Publishing
Limited, Cambridge, UK.
142
Self-assembled nanostructures on titanium surfaces by Nd:YAG-laser
treatment for durable adhesion
Specht, Uwe 1, Ihde, Jörg 1, Mayer, Bernd 1,2
1
Fraunhofer Institute for Manufacturing Technology and Advanced Materials (IFAM)
Wiener Str. 12, 28359 Bremen, Germany
[email protected]
2
University of Bremen
Bibliothekstr. 1, 28359 Bremen, Germany
ABSTRACT
Titanium was treated in ambient atmosphere using a pulsed Nd:YAG (1064 nm) laser. Repeated laser treatments
induce a removal of surface contaminants as well as the formation of a nano-structured top layer exhibiting a large
effective surface. The laser induced oxidation leads to the presence of a surface layer with strongly improved,
hydrothermally stable adhesion when joined to a one-component, hot-curing epoxy-based adhesive. Changes in the
material properties have been characterized with respect to the topography using SEM, cross-beam FIB and EFTEM
analysis in order to correlate the adhesion behaviour with the structural characteristics of the surface.
KEY WORDS: titanium, laser oxidation, adhesive bonding, pre-treatment, hot-wet stability
tartrate [3]. Usually these chemical surface treatments
require several process steps and can produce harmful
waste.
1. INTRODUCTION
Titanium and titanium alloys exhibit a high specific
strength and good corrosion stability; therefore they are
widely used in lightweight constructions. Especially the
alloy Ti6Al4V with a market share of 50 % [1] is used
in air- and spacecrafts or in corrosive environments,
such as for ship propellers or turbines. In the latest
generation of aircrafts titanium composes more than 10
per cent by weight of the construction. In conventional
structural lightweight design titanium is mechanically
joined with other lightweight materials such as
aluminum and carbon fibre reinforced plastics (CFRP).
Beside the associated economic and manufacturing
disadvantages contact corrosion is a critical point.
Adhesive bonds can avoid these problems for efficient
lightweight design without damaging the base materials.
Laser treatments are an alternative to conventional wet
chemical methods. With lasers the essential cleaning
and texturing of the surface can be realized in a single
process step. Furthermore the elaborate waste disposal
and needed huge bathes and energy are omitted.
2. Experimental
For the single lap shear tests commercially pure
Titanium - Ti(CP) has been used as well as Ti6Al4V
samples with a size of 100 mm x 25 mm and 1 mm
thickness. The untreated surface has an average
roughness of Ra = (0,8 ± 0,1) µm.
However, it is difficult to realize long-term stable
adhesive joints with titanium, since the titanium surface
is covered with a natural oxide that changes under
interaction with temperature and humidity on the
nanometer scale [2]. This is caused by the different
spatial arrangements of the various crystallographic
oxide structures (brookite, anatase, rutile). Due to this
surface effect adhesive bonds can be detached and the
hybrid structure fails. To reduce these mechanisms a
variety of wet-chemical surface treatments have been
tested. Commonly applied are etching processes, such
as the alkaline pickling with TURCO 5578 from
Henkel. In addition, anodizing can be used as the
NaTESi process of sodium hydroxide and sodium
Laser treatments were carried out using a Q-switched
CL 250 Nd:YAG ( = 1064 nm) laser from CleanLasersysteme GmbH. The system generates pulse
frequencies in the range of 10 kHz - 40 kHz. Best
adhesion results could be achieved at a frequency of 10
kHz, pulse duration of 80 ns and a Gaussian shaped
laser spot with a diameter of 680 µm (FWHM). The
laser intensity during the pulse is 723 GW/m² and the
spot meanders over the surface with a line distance of
0.17 µm and a scan velocity of 1715 mm/s. Each
treatment consists in a first step of a horizontal and
afterwards in vertical meander direction. This parameter
(LASER A) results in a surface treatment rate of 1.5
cm²/s.
143
As a reference treatment the industrial used alkaline
cleaning with SurTec 152 and pickling by TURCO
5578 [4] has been used. SurTec 152 contains
carbonates, non-ionic surfactants, phosphates, borates,
and corrosion inhibitors. The application took 10
minutes and was conducted at 75 °C in an ultrasonic
bath. The concentration of SurTec 152 in deionized
Water was 48 g/l. Subsequently, the specimens were
rinsed three times for 5 minutes each time with
deionized water at room temperature. TURCO 5578
contains caustic soda, sodium meta-silicate and
pyrophosphate. This anisotropic etchant leads to a
macro- and micro-roughness on the surface [2]. The
etching was performed for 10 minutes at a temperature
of 90°C in a solution containing 480 g/l TURCO 5578
in deionized water. After this procedure, the specimens
were rinsed three times with deionized water for 5
minutes.
Figure 2. SEM images of laser induced nano-porous oxide layers on
Ti (CP) – left: unaged samples [6], right: after hot wet aging for 100 h
at 120 °C, 100 % r.h. and 2 bar[7].
The results of the lap shear tests in figure 3 show the
effect of the improved cleaning and structuring of the
Ti6Al4V surfaces by the laser treatment and TURCO
5578-pickling. Thus, a higher adhesive strength is
achieved already before artificial aging, compared to
IPA or SurTec 152 cleaning. After hot-wet-exposure the
only cleaned samples (IPA / SurTec) showed an even
greater decline in strength of about 90 %.
The adhesive performance was tested with the onecomponent, hot-curing epoxy-based adhesive DELO
Monopox AD 286, which consists in cured condition
mainly of the bisphenol A diglycidyl ether. The
aluminum-filled adhesive has been cured at 150 °C for
40 minutes. Single lap shear strength tests were
performed using a Zwick/Roell Z 020 tensile test
machine with respect to DIN EN 1465 (see figure 3).
The adhesive performance was tested after curing of the
epoxy as well as after hydrothermal aging with an
autoclave at a relative humidity of 100 % and a
temperature of 120°C for 100 hours at 2 bar (pressurecooker-test PCT; DIN EN 60749-33). The structural
properties of the titanium substrates were characterized
using a scanning electron microscopy (SEM Zeiss Leo
Gemini 1530) and a FEI Helios Nanolab 600 for the
generation of cross sectional lamellas by focused ion
beam (FIB). For the energy-filtered transmission
electron microscopy (EFTEM) analysis a FEI Tecnai
F20 S-TWIN microscope was used.
Figure 1. test geometry of single lap joints (DIN EN 1465)
3. Results and Discussion
Figure 3. results of the lap shear tests according to DIN EN 1465 –
bottom: tensile shear strengths for different Ti6Al4V pre-treatments,
top: fracture patterns of aged samples with isopropyl alcohol and
laser treatment
The laser treatment of pure titanium (CP) in ambient
atmosphere caused a clean and nano-structured surface
layer with high surface enlargement [5]. Analysis of the
surface chemistry by XPS measurements shows the
removal of organic and residual contaminations due to
laser treatment [6]. The nano-porous structure is ideal
for durable adhesive joining because even after a hot
wet aging procedure, this structure can be retained
(figure 2).
Because of a lack of micromechanical interlocking the
adhesive separates completely from the surface
(adhesive failure). Both the laser pre-treated and stained
samples show a decrease of 50 % after ageing. But the
failure mode is still cohesive and can be explained by
the adhesive degradation due to hot-wet-exposure. The
144
energy absorption is directly proportional to the amount
of evaporated material, measured in form of particles.
This self-amplifying process depends in addition to the
number of repetition on the laser power. The amount of
the evaporated substrate material can be controlled by
the applied laser power. For the chosen parameter
following laser treatment cycles results in a saturation
region of the particles produced. With this selfamplifying and self-regulating process the laser
parameters for the adhesive experiments have been
chosen [7].
laser-treated titanium-oxide surface, however, is still
stable, as can be seen in Figure 3.
The nano-porous layer generated by the laser treatment
leads to a long-term stable adhesion since the adhesive
can infiltrate the structures and generates a
micromechanical interlocking, which remains stable
even after wet-hot-aging. Via energy filtered
transmission electron microscope (EFTEM) it is evident
by the carbon signal that the EP-adhesive infiltrates into
the laser-produced nanostructures (see cross sectional
view in figure 4).
In further studies, the application of laser nanostructures could be shown on CFRP-titanium-laminates,
fine titanium wire structures as well as for durable paint
coating adhesion [9],[10]. The fact that a laser induced
nano-structuring leads to durable adhesion even better
than the established NaTESi method, was shown by
[11].
Figure 5. scheme of laser surface interaction (adapted from[12])
4. Summary
It could be shown that treatment of titanium surfaces
with a Nd:YAG-laser is a very good environmentally
friendly and economical alternative to wet-chemical
processes for adhesive bonding of titanium. The reason
for this is a laser-induced formation of nano-structured
titanium-oxide layer, which allows hydrothermally
stable bonds. The origin of the formation of this layer is
the generation of TiO2 nanoparticles in an oxygen
containing atmosphere as a result of laser-induced
material evaporation and condensation. With this
understanding of the process mechanism, an ecofriendly pre-treatment of titanium substrates for durable
adhesive bonding or painting is possible.
Figure 4. top: SEM of laser induced nano-porous layer on Ti6Al4V,
bottom: EFTEM-image of the filtered electron signal of the adhesive
infiltrated into the nano-porous layer (brightness is associated to the
carbon content)
The reason of the formation of the nano-porous layer
was suspected in the evaporation of the substrate
material, followed by condensation-reaction and
redeposition on the surface under oxygen containing
atmosphere [8]. The mechanism of local evaporation of
the substrate material, and finally the formation of
plasma close to the surface by the laser irradiation is
shown in Figure 5.
The laser pulse supplies the energy required for
evaporation of the titanium substrate. When the laser
pulse stops, titanium-oxide nano-particles condense
under interaction with the gas atmosphere and deposit
on the surface. If a second pulse hits the surface with
the condensed particles, those act like defects on the
surface and leading to a higher laser absorption and
hence to the material evaporation. This increase of
ACKNOWLEDGEMENTS
The authors gratefully acknowledge financial support of
this work by DFG (FOR1224) and our partners within
the research group “Schwarz-Silber”. We thank Dr.
Karsten Thiel for TEM-measurements.
145
REFERENCES
[1] M. Peters, “Titan und Titanlegierungen”, 3. Aufl.
Weinheim: Wiley-VCH, 2007.
[2] S. Nouri Shirazi, “Wet chemical surface
modifications of Titanium and Ti6Al4V alloy and
their effect on the hydrothermal aging mechanisms
and adhesion properties”, Dissertation, Universität
Bremen, Bremen, 2011.
[3] G. Habenicht, „Kleben: Grundlagen, Technologien,
Anwendungen”, 6. Aufl. Springer, 2008.
[4] G. W. Critchlow und D. M. Brewis, “Review of
surface pretreatments for titanium alloys”, Int. J.
Adhes. Adhes., Bd. 15, Nr. 3, S. 161–172, Juli 1995.
[5] S.
Zimmermann,
“Badfreie
Oberflächenmodifikation
von
Titan
durch
thermische und lasergestützte Verfahren“, Thesis,
Technische Universität Ilmenau, Ilmenau, 2011.
[6] S. Zimmermann, U. Specht, L. Spieß, H. Romanus,
S. Krischok, M. Himmerlich, und J. Ihde, “Improved
adhesion at titanium surfaces via laser-induced
surface oxidation and roughening”, Mater. Sci.
Eng. A, Bd. 558, S. 755–760, Dez. 2012.
[7] U. Specht, J. Ihde, und B. Mayer, „Erhöhte
Langzeitstabilität von Titan-Epoxidklebungen durch
Laser-Nanostrukturierung“,
in
Tagungsband
Verbundwerkstoffe, Karlsruhe, 2013, Bd. 19, S.
440–449.
[8] A. Pereira, P. Delaporte, M. Sentis, A. Cros, W.
Marine, A. Basillais, A. L. Thomann, C. Leborgne,
N. Semmar, P. Andreazza, und T. Sauvage, “Laser
treatment of a steel surface in ambient air”, Thin
Solid Films, Bd. 453–454, S. 16–21, Apr. 2004.
[9] U. Specht, S. Zimmermann, M. Himmerlich, J. Ihde,
und B. Mayer, “Oberflächenvorbehandlung von
CFK-Aluminium-Übergangsstrukturen“, Dresden,
2012, Bd. DVS-Berichte-292, S. 15–20.
[10]U. Specht, Ihde, Jörg, N. Shirazi, R. Wilken, und B.
Mayer, “Stabile Titan-Klebungen durch LaserNanostrukturierung“, WOMag, Bd. 3, Jan. 2014.
[11]A. Kurtovic, E. Brandl, T. Mertens, und H. J.
Maier, “Laser induced surface nano-structuring of
Ti–6Al–4V for adhesive bonding”, Int. J. Adhes.
Adhes., Bd. 45, S. 112–117, Sep. 2013.
[12]D. Bäuerle, “Laser processing and chemistry", 4.
Aufl. Berlin; New York: Springer, 2011.
146
Evaluation of the durability of new “eco-efficient and REACH compliant”
surface pre-treatments on metallic substrates prior to adhesive bonding
Jean-Pierre Jeandrau
CETIM
7 rue de la presse, 42952 Saint-Etienne Cedex 1, France
[email protected]
ABSTRACT
Based on a previous bibliographic review, for each metallic alloy, different eco-efficient
surface pre-treatments were performed (silane coupling, sol-gel treatment, free-chromate
commercial solutions…) and compared to the best established procedures used as a
reference (i.e. Chromic Acid Anodising for aluminium alloy).
Surface energies (dispersive and polar components) and surface roughness
measurements were performed on treated substrates prior to adhesive bonding.
Then wedge-tests specimens (according to EN 14444 standard) were bonded with a twopart epoxide adhesive (3M Scotch weld 9323 B/A) and tested after immersion in hot
(60°C) demineralised water. For each substrate, the different “eco-efficient” surface
preparations were compared with the conventional ones
For the three metallic alloys studied, “eco-efficient” surface preparations are proposed in
this paper: a sol-gel procedure and the use of a silane coupling agent (Ɣ GPS) are the
most interesting in terms of performances and durability of the bonded joints
KEY WORDS: adhesive-bonding,
1
metallic substrates, surface pre-treatment, eco-efficient, wedge test.
-the ferric chloride/hydrochloric acid etch, 2 stage
oxalic/sulphuric acid and sulphuric/chromic acid etch,
or other chemical etching or anodising solutions for
stainless steels.
INTRODUCTION
It is well known that the surface preparation of the
substrates prior to adhesive-bonding is a very important
step for the performances and for the long-term
durability of the bonded joints in hot/wet environment.
Unfortunately, most of the best established procedures
successfully applied on metallic substrates involve
solvent cleaning and chemical treatments which are not
compliant to new legislations relating to health and
environmental considerations (VOC limitations, RoHS
and REACH European regulations). Moreover, there is
a need for simpler, more robust and economically
cheaper processes.
A literature survey informs us that lot of work was
made to identify and compare the suitable surface pretreatments prior to adhesive bonding for metallic
materials [1-4].
The best and most cited pre-treatments in terms of
adherence and long term durability of bonded joints in
hot/wet environment are:
-the chromic-sulfuric acid etch process or other
proprietary chromate conversion coatings, the chromicacid anodising process (CAA) or the phosphoric-acid
anodising process (PAA) for aluminium alloys,
-the chromic-acid anodising process (CAA) or the
sodium hydroxide anodising (SHA) and liquid hone
Pasa-Jell 107® etching process for titanium alloys,
More recently, alternative more “eco-efficient” and
robust surface pre-treatments on these three metallic
materials have been tested, analysed and compared with
conventional pre-treatments in literature:
-the use of silane coupling agents [5,6],
-sol-gel process [7-10]
-chromate-free conversion coatings [11-16].
147
adhesive bonding (Alodine 5700® provided by
Henkel): Turco 5578 + Alodine 5700.
The conventional surface pre-treatments used as
references for this study were:
-the chromic acid anodizing process for the aluminium
alloy: CAA
-the Pasa-Jell® etching process: PJ 107
-the 2 stage oxalic/sulphuric acid and sulphuric/chromic
acid etching process for the stainless steel alloy: OS/SC.
2 EXPERIMENTAL
In this paper, new “friendly and eco-efficient” surface
preparations were evaluated on three metallic alloys
(aluminium alloy 2024T3, stainless-steel 304L and
titanium alloy Ti4V6Al) and compared to conventional
and well established procedures (most of them
involving the use of chromate compounds).
2.1 Materials
2.3 Surface characterization
Commercially produced aluminium alloy 2024-T3,
stainless-steel 304L and titanium alloy Ti4V6Al were
used for this study. For the wedge tests, according to
NF-EN 14444 standard, coupons of 25 x 150 mm², 3
mm thick were bonded with a two-part epoxide
adhesive (Scotch-Weld 9323 B/A provided by 3M
Company) cured for two hours at 80°C.
Roughness measurements were made using a
profilometer Marh Perthometer PGK 120. The Ra
parameter gives the mean deviation from the mean line
whereas the Rt parameter gives the maximum peak-tovalley distance on the surface.
Static contact angle measurements were performed with
a PGX goniometer using two liquids (water and
diiodomethane). The surface energies were calculated
with the method of Owens-Wendt, including both the
dispersive and polar components.
2.2 Surface pre-treatments
The following surface pre-treatments were employed on
the 3 substrates prior to adhesive bonding of the wedge
test samples:
1.
2.
3.
4.
5.
6.
2.4 Mechanical test methods
Wedge testing was performed according to the
procedure described in standard NF-EN 14444 (see the
specimen dimensions on figure 1). Spacers (made of
PTFE) were inserted at the edges to obtain a glue-line
thickness of 0.2 mm between the two metallic surfaces.
After applying and curing the adhesive, the cut edges
were polished to facilitate measurement of the crack
length under an optical microscope Tesa Vision V300.
Specimens were immersed in DI water at 60°C up to
1500 hours.
Degreased in a commercial hydrocarbon
solvent: D
Degreased as in 1, grit blasted with Al2O3
particles (65 μm size) and re-degreased: D +
GB
Degreased, grit blasted with Al2O3 particles
(65 μm size) and re-degreased, then applying
and drying a proprietary system based onƔglycidoxypropyltrimethoxysilane (Ɣ GPS) : D
+ GB + Ɣ GPS
(65 μm size) and re-degreased, then applying
and drying a proprietary system based on a solgel coat (Socogel B0102® provided by
Socomore: D + GB + SG B0102
(65 μm size) and re-degreased then applying
and drying a proprietary system based on a
chromate-free conversion coating containing
organo-metallic zirconate complex based
elements specially formulated for preparing
aluminium and titanium alloys before adhesive
bonding (Alodine 5700 ® provided
by
Henkel): D + GB + Alodine 5700. This
treatment was not applied on the stainless steel
alloy.
Degreased and chemical etch in a proprietary
system based on sodium hydroxide (Turco
5578® provided by Henkel) for Titanium alloy,
then applying and drying a proprietary system
based on a chromate-free conversion coating
containing organo-metallic zirconate complex
based elements specially formulated for
preparing aluminum and titanium alloys before
Figure 1. Wedge test specimen according to NF-EN 14444
148
3 RESULTS AND DISCUSSION
3.2 Titanium Ti4V6Al alloy
3.1 Aluminium 2024-T3 alloy
Table 2 summerises the surface energy measurements
made on pre-treated surfaces of Ti4V6Al titanium alloy
(within a 30 minutes delay).
made on pre-treated surfaces of 2024-T3 aluminium
alloy (within a 30 minutes delay).
Surface pre-treatment
Surface energy mJ/m²
Es
Disp
45
Es Pol
D
Es
Total
53
D + GB
65
47
18
D + GB + Ɣ GPS
71
45
26
D
Es
Total
46
Es
Disp
38
Es
Pol
8
D + GB
60
47
13
D + GB + SG B0102
81
45
36
D + GB+ Ɣ GPS
74
45
29
D + GB+ SG B0102
80
45
35
D + GB + Alodine 5700
CAA*
75
73*
50
50*
25
23*
D + GB + Alodine 5700
Turco 5578 + Alodine 5700
PJ 107**
74
66
NA**
50
39
NA**
24
27
NA**
8
Table 2.surface energies on pre-treated titanium alloy
** Results not yet available.
Table 1. surface energies on pre-treated aluminium alloy
*The CAA pre-treatment process was not operated in our laboratory
for safety reasons. The delay between the end of the pre-treatment and
the surface energy measurements is not optimal in this case.
Figure 3 shows mean crack length as a function of time
during the first stage of environmental exposure.
Figure 3.crack length vs time for Ti4V6Al titanium alloy wedge test
specimens
Figure 2. crack length vs time for 2024-T3 aluminium alloy wedge test
specimens
The results indicate high crack propagation already after
2 hours on degreased only specimens, leading to a
complete interfacial delamination, associated to a low
surface energy.
Grit blasting leads to slower crack propagation as
expected, associated to a higher polar component of the
surface energy.
The best durability results were obtained with the
application of the sol-gel after grit blasting associated to
the highest polar component of the surface energy.
Similar results are obtained on grit blated surfaces and
post-treated with the Ɣ GPS silane or Alodine 5700
after grit blasting. Grit blasting may be replaced by a
chemical etching (Turco 5578) with a slight beneficial
effect.
Except for only degreased specimens, the test
specimens didn’t fail during the 1500 hours immersion
in hot DI water.
Results with the Pasa-Jell® 107 etching process are not
yet available.
surface energy.
surface energy.
application of the sol-gel after grit blasting, associated
to the highest polar component of the surface energy.
post-treated with the Ɣ GPS silane after grit blasting.
Post-treated specimens with Alodine 5700 after grit
blasting give slightly higher crack propagation than the
previous pre-treatments.
Not surprisingly, the CAA pre-treated specimens show
the lowest crack extentions.
in hot water.
149
GPS) are the most interesting in terms of performances
and durability of the bonded joints
3.3 Stainless-steel 304L
made on pre-treated surfaces of 304L stainless steel
alloy (within a 30 minutes delay).
REFERENCES
[1] Cognard, P., in “Handbook of adhesives and
sealants”, vol 1, ch 4 and 5, 2005
Es Total
Es Disp
Es Pol
D
71
44
27
D+GB
65
49
16
D+GB+ Ɣ GPS
74
46
28
D+GB+SG B0102
76
46
30
OS/SC***
NA***
NA***
NA***
[2] Adams, R.D., in “Adhesive Bonding – Science,
Technology and Applications”, ch 4 and 6, 2005.
[3] «Adhesives and sealants – Engineered materials
Handbook», vol 3, section 4, 1990.
[4] Critchlow, G.W., Lee, R.J., Bull S.J., Wingfield,
J.R.J.,Brewis,
D.M.,
Hutchinson,
A.R,
“characterisation of surface condition-Review of
substrate surface treatment”, Report No 2, 1993.
Table 3. surface energies on pre-treated stainless steel
*** Results not yet available.
[5] Special issue “Silane Coupling Agents”, Int J Adh
Adh, vol 26, n° 1-2, 2006.
[6] Bistac, R., Kopp, J.B., Brogly, M., Delaite, C.,
“Surface treatment of stainless steel with silane:
Infrared reflection-absorption spectroscopy analysis”,
Structural Adhesive Bonding conference, Porto, 2011.
[7] Lunder, O., “Chromate-free pre-treatment of
aluminium for adhesive bonding”, report URN-NBN,
no-7238, May 2003.
[8] “Non-chromate Aluminium pre-treatments” -NCAP
Project # PP0025-Phase I report, August 2003, Phase II
interim report, September 2004
Figure 4.crack length vs time for 304L stainless steel alloy wedge test
specimens
surface energy.
surface energy.
application of the sol-gel after grit blasting, associated
to the highest polar component of the surface energy.
post-treated with the Ɣ GPS silane after grit blasting.
in hot DI water.
Results with the 2 stage oxalic/sulphuric acid and
sulphuric/chromic acid etching process are not yet
available.
[9] Blohowiak, K., “Improvement in surface
preparation methods for adhesive bonding”, Material
and Process Technology, The BOEING Company,
Seattle, Washington, USA-SERP/ESTCP Workshop, 27
February 2008.
[10] Wong, J.T., “Evaluation of Titanium Bonding
Surface Preparation Method: Sol-gel AC-130-2”,
Conference presented at the American Helicopter
Society , 68th Annual Forum, Fort Worth, Texas, May
1-3, 2012.
[11] Mertens, T, Gammel F.J, Kolb M, Rohr O, Kotte
L, Tschöcke S, Kaskel S, Krupp U“Investigation of
surface pre-treatments for the structural bonding of
titanium”, Int J Adh Adh. 34, pag 46-54, 2012.
[12] Museux, F, Theilmann, R, « Introducing more
eco-efficient chemical treatments for aircraft structure
towards a chromate-free Airbus », F.A.S.T Airbus
technical Magazine N° 45, Pag 2-10, 2009.
4 CONCLUSIONS
For the three metallic alloys studied, “eco-efficient”
surface preparations are proposed in this paper: a solgel procedure and the use of a silane coupling agent (Ɣ
150
Atmospheric pressure plasma jet treatment for structural adhesive bonding of
thermoplastic composites
J. Schäfer123,
J. Holtmannspötter , M. Frauenhofer2, T. Hofmann1, J. von Czarnecki1, H.-J. Gudladt3
1
1
Bundeswehr Research Institute for Materials, Fuels and Lubricants (WIWeB)
Institutsweg 1, 85435 Erding, Germany
2
AUDI AG
Auto Union Str. 1, 85045 Ingolstadt, Germany
[email protected]
3
Bundeswehr University Munich
Department of Aerospace Engineering – Institute for Material Science
Werner-Heisenberg-Weg 39, 85577 Neubiberg, Germany
ABSTRACT
A surface treatment process using atmospheric pressure plasma jet (APPJ) treatment was investigated for the structural
(τB > 15 MPa) adhesive bonding of carbon fiber reinforced plastics with a polyamide 6 (PA6) matrix. The treated
surfaces were examined by contact angle measurement, x-ray photoelectron spectroscopy, and atomic force
microscopy. Additionally, the bond strengths of single-lap shear samples as a function of different plasma intensities
were determined for two room temperature curing two-component polyurethane adhesives. The results show, that the
surface chemistry is heavily influenced by APPJ. The relative concentration of bound oxygen on the surface was found
to increase; in particular oxygen containing functional groups were created causing the surface free energy to rise.
Furthermore, the topography of the surface can be significantly modified by APPJ. Special attention has to be paid to
the intensity of the APPJ to avoid melting and flattening of the PA6 surface on the nanometer scale. Deliberately
optimized, lap shear strengths of 20 MPa were achieved for the first time for this material system.
KEY WORDS: atmospheric pressure plasma jet, surface technology, structural adhesive bonding, polyamide, XPS, AFM
This study investigates the effects of APPJ on PA6
surfaces and its relevance for structural adhesive
bonding. The chemical composition of the surface, the
wettability, the topography and the bond shear strength
were evaluated to generate a deeper understanding of
surface effects created by exposure to atmospheric
plasma. Special attention was paid to the intensity of the
plasma jet due to the sensibility of the polymer surface.
1. INTRODUCTION
Carbon fiber reinforced plastics (CFRP) with
thermoplastic matrix systems are of major interest for
automotive applications due to their weight saving
potentials. Particularly for automotive mass production
CFRP with a polyamide 6 (PA6) matrix are able to
fulfill the necessary requirements in terms of its price,
processing time, and mechanical properties [1]. For
structural usage in future light weight multi material car
bodies, adhesive bonding is the preferred joining
method [1-3].
However, polymer materials are known for their poor
wettability and adhesive properties [3-5]. Especially
PA6 exhibits poor adhesive properties with room
temperature curing two-component polyurethane (PU)
adhesives. Thus far, structural (τB > 15 MPa) adhesive
bonding was not possible with PA6 and PU adhesives.
To overcome this barrier, surface treatment of polymers
can be used to form durable bonds of high strength. A
widespread surface treatment method for adhesive
bonding is low-pressure plasma, however its application
is not appropriate for industrial manufacturing due to
the high processing cost and incompatibility to inline
processes. A practicable alternative without the above
disadvantages is APPJ treatment.
2. EXPERIMENTAL
2.1. Material
An unidirectional carbon fiber reinforced polyamide 6
(BASF Ultramid B) with 60 weight percent of carbon
fibers, produced by a film stacking process, was used.
All samples were dried and conditioned according to
EN ISO 1110 [6]. Single overlap joints were bonded
with two different room temperature curing twocomponent polyurethane adhesives (Table 2) developed
by DOW Automotive. The adhesive layer thickness was
adjusted with calibrated glass beads (195-210 µm) and
mechanical testing was performed after a curing time of
7 days under laboratory conditions (23 °C, 50% rel.
humidity).
151
mm x 25 mm. The test rate was 10 mm/min at room
temperature.
2.2. Surface treatment
The samples were cleaned (wiped with n-heptane) and
treated with a Plasmatreater 400 atmospheric pressure
plasma jet (APPJ) generator from Plasmatreat GmbH.
The frequency (21 kHz), plasma cycle time (100),
plasma voltage (315 V), and gas flow (2400 l/h) were
kept constant. The APPJ intensity was adjusted to avoid
visible surface effects, like melting. Three different
intensities for the plasma treatment were chosen and set
by adjusting nozzle velocity and distance. The different
intensities are categorized as high, medium, and low. As
a forth treatment, the low intensity treatment was
performed consecutively five times with an intermission
of 30 seconds in between to reduce thermal influences.
3. RESULTS
3.1. Contact angle measurement
The selected parameters of the APPJ treatments result in
a strong increase of surface free energy (SFE) (Figure 1)
and an improved wetting ability. Especially noteworthy
is the polar component of SFE, which indicates an
increase of polar functional groups on the surface,
which are most likely oxygen-related. Wetting however
only represents a necessary condition for adhesion but
not sufficient condition to gauge adhesive bond
strength.
2.3. Analytical techniques
2.3.1. Contact angle measurement
The contact angle was measured using the static sessile
drop method to obtain the wettability of cleaned and
treated PA6 samples. The contact angles of water and
diiodomethane were measured with a Krüss DSA30
drop shape analyzer. The polar, dispersive components
and total surface free energy were evaluated by
application of the Owens, Wendt, Rabel and Kaelble
method (OWRK). [2, 5]
2.3.2. X-ray photoelectron spectroscopy
The chemical composition of cleaned and APPJ-treated
PA6 surfaces, along with the chemical environment of
carbon species were examined by x-ray photoelectron
spectroscopy (XPS). The experiments were performed
in an Escalab 220i-Xl (Thermo Fisher Scientific) using
the Mg Kα radiation (1253.56 eV) of a dual anode x-ray
source at a base pressure below 5.0x10-9 mbar. Sample
charging was compensated by shifting all spectras to the
binding energy position for C-C / C-H species at 285.0
eV. Peak fitting was performed using a GaussianLorentzian peak shape and a linear background for all
peaks.
Figure 1. Surface free energy of APPJ treated samples
Figure 2. Theoretical PA6 structure
3.2. Surface chemical composition
2.3.3. Atomic force microscopy
XPS was used to analyze the surface chemical
composition of the cleaned PA6 surface and samples
treated with varying APPJ intensities. The
photoemission peaks around 285 eV, 400 eV and 532
eV correspond to the C 1s, N 1s and O 1s levels,
respectively. To compare the effect of different APPJ
treatments with the cleaned sample, the data was
normalized to the elemental composition ratio of C 1s to
N 1s and O 1s. After the different APPJ treatments the
O/C and N/C ratio strongly increase (Table 1). This
phenomenon is a result of the presence of atmospheric
oxygen and nitrogen during the treatment process. APPJ
emit a mixture of highly energetic vacuum-UV radiation
(6.2-12.4 eV), energy-rich particles, ions, electrons, and
Atomic force microscopy (AFM) was used to analyze
and quantify the surface topography. All investigations
were performed with a Cypher AFM (Asylum
Research) in tapping mode with a scan area of 10 µm x
10 µm at a frequency of 0.5 Hz.
2.3.4. Mechanical testing
Bond strength and fracture mode of untreated and
treated samples were determined using lap shear
strength tests. The tests were carried out using a tensile
testing machine (Zwick Roell Z020) and a setup
according to DIN EN 1465 [7] with an overlap of 12.5
152
radicals. UV radiation and energy-rich particles are able
to break up various carbon bonds, e.g. ozone radicals
(O3) react with the aliphatic CH2 and amide groups
(CONH) [8]. This reaction forms new functional
groups.
surface. This is related to an etching process on the
surface in the absence of major thermal effects.
Table 1. XPS analysis – chemical composition
chemical composition [at. %]
C 1s
O 1s
N 1s
theoretical
(Figure 2)
cleaned
high
medium
low
5x low
atomic ratio [-]
O/C
N/C
75.0
12.5
12.5
0.17
0.17
72.2
53.9
56.3
58.5
56.9
18.2
33.3
31.1
30.6
31.3
9.0
12.7
12.5
10.9
11.8
0.25
0.62
0.55
0.52
0.55
0.12
0.23
0.22
0.18
0.20
The XPS peak series (Figure 3) of the C 1s spectra show
that atmospheric pressure plasma jets lead to the
incorporation of different oxygen containing functional
groups into the polymer surface. The amount of newly
incorporated species is depending on the APPJ intensity.
Figure 3 shows the qualitative trend of C-OH, C-OOH
and C=O species.
Figure 4. AFM surface images (10 µm x 10 µm) - topography
modification of different APPJ-treated PA6 samples
In contrast to parameters previously determined for
thermoset materials, the workable parameters found for
PA6 turned out to be vastly different. The results
suggest that APPJ treatments require material specific
tuning and surface analysis to achieve the desired
surface properties.
3.4. Mechanical testing
Single overlap shear strength was determined with two
different PU adhesives (Table 2) to investigate the
effects of APPJ treatments on adhesive bonding. Figure
5 shows the lap shear strength as a function of APPJ
treatment. By decreasing the plasma intensity, bond
strength increased, however the failure mode was still
adhesive for both PU adhesives. Only after applying the
low intensity treatment five times in a row, the cohesive
strength of the adhesives was reached.
Figure 3. Modification of XPS C 1s peaks of PA6 treated with
different APPJ intensities. Oxidized species (arrow) are located at
higher binding energies.
Table 2. Polyurethane adhesives
PU adhesive 1
PU adhesive 2
Young’s
modulus
[MPa]
21
300
Tensile
strength
[MPa]
8
18
Elongation
[%]
150
80
3.3. Surface topography
Figure 4 presents AFM topography images of PA6
surfaces after APPJ treatment with varying intensities.
The PA6 surface with the high intensity treatment
shows a drastically reduced surface roughness as
compared to the cleaned surface. (Figure 4 a, b) The
root mean square (RMS) roughness value is reduced
from about 116 nm to 4 nm, due to the melting of the
polymer surface. By decreasing the plasma intensity, the
surface roughness increases compared to the cleaned
Figure 5. Lap shear strength of cleaned and APPJ surface treated
PA6 samples
153
4. Discussion
5. Conclusion
The surface analytical techniques show, that the
polymer surface is heavily influenced by APPJ
treatment. Surface free energy is drastically increased
after all treatments. Surface chemical composition, as
determined by XPS, shows a strong rise in oxygen
content through formation of polar oxide groups.
Furthermore, AFM measurements suggest, that the
surface topography is strongly dependent on the APPJ
intensity, allowing surfaces of higher roughness to
surfaces with strongly decreased roughness to be
created in the process. The resulting lap shear strength
of the high, medium, and low intensities can be
explained by the intensity-dependent chemical
interaction of PA6 and APPJ.
The structure of non-treated PA6 has a variety of
chemical bonds of which most can be broken up by the
vacuum-UV radiation ( = 100-200 nm) and energyrich particles emitted by APPJ. Especially chain
scissions at the C-N bond and oxidation of the aliphatic
CH2 bond can occur [8]. Figure 3 shows the
modification of the C 1s peak with APPJ intensity.
Especially highly oxidized functional groups (like
hydroxyl C-OH, aldehyde R-CHO and carboxylic acid
R-COOH) were generated on the surface. These
functional groups represent different oxidation states
(hydroxyl < aldehyde < carboxylic). Due to the increase
of these groups, which is usual a positive aspect for
adhesive bonding, there could also be a strong oxidation
leading to degradation of the polymer surface and lower
layers in case of overtreatment. For the latter case there
will likely be loosely bonded polymer fragments and
low molecular weight oxidized material (LMWOM) on
the surface. However, our results show that when
functionalization with oxidized groups is optimized and
degradation minimized, structural adhesive bonds can
be created.
A feasible process to treat carbon fiber reinforced
polyamide 6 with atmospheric pressure plasma jet was
demonstrated, achieving stable structural adhesive
bonds with about τ = 20 MPa lap shear strength. The
improved adhesive strength can be explained as a result
of the surface enlargement due to plasma etching and
strong functionalization with oxygen containing groups.
Surface melting, polymer degradation and generation of
LMWOM however need to be avoided to achieve
improved structural adhesive bonds.
This can be accomplished by paying special attention to
the intensity of the APPJ. Further experiments will be
performed to determine the main influencing factors to
adhesive bonding.
In summary, polarity, oxygen concentration, and the
amount of functional groups increase for all treatments
but bond strength only reaches the cohesive strength of
the adhesive in the case of the five-fold treatment with
low intensity. All other lap shear samples show an
adhesive failure mode and lower adhesive bond
strengths. In case of polymer degradation, adhesion can
occur between the overtreated surface but there is no
bonding strength between the LMWOM and nondegraded bulk material.
Another possibility for the decrease of bond strength
and the adhesive failure mode is the influence of the
surface melting and the associated loss of effective
surface area. Further experiments are on the way to shed
more light on the underlying reasons for the decrease of
bond strength at high and medium APPJ treatments.
[4] DIN Deutsches Institut für Normung e.V. (2009),
Klebstoffe - Bestimmung der Zugscherfestigkeit von
Überlappungsklebungen, Vol. 83.180 DIN EN 1465,
Beuth Verlag GmbH, Berlin.
ACKNOWLEDGEMENTS
This work has been supported by the Bundeswehr
Research Institute for Materials, Fuels and Lubricants
(WIWeB) and the AUDI AG in the framework of a
cooperation doctoral program with the Bundeswehr
University Munich.
REFERENCES
[1] Braess, H.-H. and Seiffert, U. (Eds.) (2013), Vieweg
Handbuch Kraftfahrzeugtechnik, 7th ed., Springer
Vieweg.
[2] Brockmann, W., Geiß, P.L., Klingen, J. and Schröder, B.
(2005), Klebtechnik: Klebstoffe Anwendungen und
Verfahren, Wiley-VCH, Weinheim.
[3] DIN Deutsches Institut für Normung e.V. (1998),
Kunststoffe - Polyamide - Beschleunigte Konditionierung
von Probenkörpern, Vol. 83.080.20 DIN EN ISO 1110,
Beuth Verlag GmbH, Berlin.
[5] Habenicht, G. (2009), Kleben: Grundlagen Technologien
Anwendungen, 6th ed., Springer, Berlin.
[6] Ickert, L., Thomas, D., Eckstein, L., and Tröster, T.
(2012), Beitrag zum Fortschritt im Automobilleichtbau
durch belastungsgerechte Gestaltung und innovative
Lösungen für lokale Verstärkungen von
Fahrzeugstrukturen in Mischbauweise.
[7] Silva, L.F.M.d. (Ed.) (2011), Handbook of adhesion
technology, Springer, Berlin.
[8] Friedrich, J. (2012), The Plasma Chemistry of Polymer
Surfaces, Wiley-VCH, Weinheim
154
Improvement in the adhesion property and ink-jet printing of
polypropylen, PET, silicone resin, and other polymeric materials
Hitoshi Kanazawa1, Aya Inada2
1
Fukushima University ,
1 Kanayagawa, Fukushima 960-1296, Japan
[email protected]
2
Fukushima University,
[email protected]
The improvement in the adhesive property of chemically stable polymeric materials such as polypropylene, ultrahigh
molecular weight PE, polycarbonate, polyester, silicone and fluorocarbon resins, was carried out by the combination
method which is composed of an activation process and a chemical reaction process. We further developed the
method, and made the modification of engineering plastics such as polyacetal, polybutyleneterephthalate,
polyamides, polysulfone, polyimide polyether ether ketone, and fiber-reinforced plastics (FRP). The modified
materials gave a durable adhesion property, using general kinds of adhesives. In addition, the surface modification
of chemically stable polymeric materials suitable for the ink-jet printing with water-based ink was studied.
KEY WORDS: surface modification, adhesive property, polyacetal, CFRP, ink-jet printing, water-based
1. INTRODUCTION
2.1 Materials
Polymeric materials: polypropylene(PP), polyethylene (PE),
ultrahigh molecular weight PE
(UHMWPE),
polycarbonate(PC),
polyethyleneterephthalate (PET), silicone resin, fluorocarbon resins
(PTFE, PETFE).
Engineering plastics: polyacetal
(POM), polybutylene-terephthalate (PBT), polyamides
(PA), polysulfone (PS), polyimide (PI), polyether ether
ketone (PEEK). Fiber-reinforced plastics (FRP) ; carbon
fiber reinforced epoxy resin (CFR-EPO) and carbon
fiber reinforced PEEK (CFR-PEEK). The materials of
films, sheets, fibers, boards, rods and tubes were used
after washing in methanol. Reagents: hydrophilic
monomers, polymers and catalysts were used after a
general purification.
Adhesives: polyvinylpyrrolidone (PVP), starch, a
woodwork bond (polyvinylacetate (PVAC)-water
mixture : Konishi Co. Jp), cyanoacrylate adhesive (CA;
Aron alpha: Toa Gosei Kagaku Co.), epoxy resin
adhesive (Quick 5 ; mixture of epoxy resin and
polythiol, Konishi Co. Ltd.), a cyanoacrylate-primer set
(Cemedine PPX; primer : organic amine 1% and
heptane 99%; Cemedine Co. Ltd.), etc. were used.
Paints: commercial water-based paints (Asahipen, Co.
Ltd., Japan) which contain synthetic resins (acrylic,
silicone and fluorocarbon resins), pigments and organic
solvents, and water-based urethane car-use paints were
used.
A corona discharge treatment was extensively used for
the improvement of the adhesive property of polymeric
materials such as polypropylene and polyester films, etc.
But, the corona process is known to be not effective for
the materials such as silicone resin and
polymethylpentene (polyacetal). In addition, the corona
treated materials must be processed quickly after the
treatment. Thus, the corona process is not preferable for
large and heavy materials.
We investigated the modification of polyorefins such
as polypropylene and polyethylene and found that the
combination of an active treatment and chemical
reactions [1,2]. Polypropylene fabrics which absorb
water were obtained by the combination method. We
named the KANA1 surface modification method to this
process.
The method was developed to improve the adhesive
property of chemically stable polymeric materials such
as silicone and fluorocarbon resins. In addition, we
improved the adhesion property of several engineering
plastics, and tried FRPs which are useful for the vehicle
and aircraft materials. We aimed at the durable adhesion
improvement of the polymeric materials using general
kinds of adhesives. The modified polymeric materials
were bonded to each other or other materials using usual
adhesives. In addition, we studied the ink-jet printing
with water-based ink of several polymeric materials
which are impossible by usual methods.
2.2 Treatment
The present method (the KANA1 surface
modification method ) is composed of two processes,
2. EXPERIMENTAL
155
improvement by the present method. This fact indicates
the availability for the adhesion improvement of PP by
the present method.
an activation process and a chemical reaction
process. First, polymeric materials were activated by a
chemical oxidation, or energy irradiations such as UV,
corona, plasma, high-voltage electric discharge
treatments. Subsequently, the activated polymeric
materials were treated with solutions of polymers,
monomers or other chemical reagents in the presence of
catalysts or initiators.
Table 1 Tensile shear bond strength test of PP
boards bonded to aluminum boards
Adherend /
adhesive
Strength or
failure
Untreated /
CA
Untreated /
CA+ primer
Present
method / CA
2.3 Measurements
Adhesion strength: a tensile shear bond strength test
and a T peel test were carried out using a Shimadzu
AGS-H5KN.
IR spectra: polymeric materials were observed by a
Shimadzu IRPrestige-21 equipped with a Smiths
DuraSampl IR II (ATR accessory).
Cross-cut test: JIS 5400 (Japanese Industrial Standards
Committee) was employed for the peeling test of paints
and inks.
PP-Al / CA
Tensile shear
strength (MPa)
0.01
1.69
3.04
Failure mode
PP interface
Aluminum
interface
Cohesive of
CA
3.2 Adhesion of silicone resin, PC, PI, PETFE,and
PTFE
The modified silicone rubber sheets got wet in water
and were bonded to wooden boards using a PVAC
adhesive (woodwork bond). Modified PC, PI boards
were bonded to paper with starch or PVA glue.
2.4 Ink-jet printer
Ink-jet printers, Dainichi-seika printer, Canon MG8230,
and Epson PX105 were used with water-based pigment
inks.
PETFE (polyethylene tetrafluoroethylene copolymer)
materials modified by the present method were bonded
to aluminum boards using epoxy resin adhesives.
PTFE resin sheets modified by the present method
were bonded to aluminum boards using a PVP glue (see
Figure 2).
3. RESULT AND DISCUSSION
3.1 Comparison of the present method with the
primer-adhesive
The adhesive property of almost polymeric materials
was improved well by the present method. Modified PP
boards were bonded to wooden boards or aluminum
boards using any type of adhesives (see Figure 1).
Fig. 1 Modified PP board bonded to a wooded
board with a woodwork glue
Fig. 2 PTFE sheet bonded to an
aluminum board with PVP glue
The adhesive property of almost polymeric materials
was improved well by the present method. Recently,
primer-adhesive sets are commercially available for the
adhesion of PP, PE and other materials of which
adhesion is difficult. Thus, the CA adhesion of the PP
boards modified by our method was compared with the
primer-CA adhesion (Cemedine PPX) of untreated PP
boards. Table 1 gives the tensile shear strength test of
the adhesion of PP boards to aluminum boards.
The tensile shear strength of untreated PP board to
aluminum board using the primer-CA is 1.69 MPa ; the
adhesive at the PP interface was peeled. On the other
hand, the cohesive failure of CA was observed at 3.04
MPa in the test of the adhesion of the modified PP
board bonded to aluminum board with only CA. This
fact indicates the availability for the adhesion
3.3 Adhesion of engineering plastics and CFRPs
PEEK resin is one of engineering plastics of which
adhesion is very difficult. PEEK boards were modified
well by the present method. Untreated PEEK boards
were bonded to each other using a film type epoxy resin
adhesive(product of 3M Co. Ltd.). The PEEK boards
modified by the present method were bonded to each
other using the same adhesive. The tensile shear bond
strength test of these specimens was carried out.
Though untreated PEEK boards were not bonded to
each other, the modified ones gave a material failure at
6.3MPa.
The specimens used in the tensile shear test of the
untreated and modified CFR-PEEK boards bonded to
156
Modified flurocarbon resin, PETFE rods were also
coated well with water-based paints.
each other using the film type adhesive are given in
Figure 3. The modified one gave a tensile shear strength,
25 MPa, and gave a cohesive failure. But, untreated one
gave a tensile shear strength 4.8 MPa, and an interface
peeling of the adhesive was observed.
3.6 Ink-jet printing
It is impossible to print on PP, PC, PET and silicone
resin materials by an ink-jet printer with water-based
ink. But, the materials modified by the present method
can be printed by usual ink-jet printers with water-based
ink. A PET film printed by an ink-jet printer with
water-based ink is given in Figure 5. The results are
presented in a poster presentation in the EURADH 2014.
Fig.3 CFR-PEEK boards used in the
tensile shear test; the upper and under
boards are untreated and modified ones
CFR-EPO boards were modified by the present
method and they were bonded to each other using the
film type epoxy resin adhesive. Untreated CFR-EPO
boards gave around tensile shear strength 20MPa, and
the adhesive gave an interface peeling, On the other
hand, modified ones gave 50MPa, and a cohesive failure
was observed.
Fig. 5 Ink-jet printing on modified PET film
3.7 Mechanism
The mechanism of the present process was speculated
with reference to the old studies [3] . Several oxidized
polymer structures were proposed. Oxidized polymers
(P-OOH or P=CO) formed in the activation process are
considered to react with P-X (functional compound) to
form product “Polymer-OP”. The Polymer-OP gives
the modification.
3.4 Solvent bonding
PP, PE, PMP and silicone resin tubes modified by the
present method are strongly connected to each other by
the solvent bonding. It is preferable for the medical use.
The results are shown in a poster presentation in the
EURADH 2014.
3.5 Coating with water-based paints
The polymeric materials, PP, PE, UHMWPE, PC,
various FRP, silicone resins, and fluorocarbon resins
(PETFE,etc.) modified by the present method were
coated well with water-based paints.
Figure 4 gives the result of the cross cut test of the PP
boards coated with water-based. Although the untreated
PP board gave 0/100, the modified one gave 100/100 in
the cross-cut test.
4.
CONCLUSIONS
KANA1 surface modification method) The modified
polymeric materials are useful in many fields.
Especially, the present method is useful for reducing the
weight of automobiles and airplanes by applying in
composite materials (CFRP).
REFERENCES
[1] Kanazawa, USA Patents No.7294673 and No.
6830782B2.
[2] H. Kanazawa, Japanese Patent No.4229421, etc.
[3] T. Matsui, A. Yamaoka, and Y. Yamaguchi, Nippon
Kagaku Kaishi, 732 (1992).
Fig.4 Cross-cut test of PP boards
painted with water-based paints.
untreaed
(upper;0/100)
and
modified (lower;100/100)
157
158
SURFACE TREATMENTS FOR ADHESIVE BONDING OF COMPOSITE
MATERIALS
Gaëlle M. Roger1, Mai T. Nguyen1
Onera – The French Aerospace Lab
F-92322 CHATILLON, France
[email protected]
1
ABSTRACT
In the aeronautics, the lightning of structures in a very important objective and adhesives are one solution among
others to replace the use of heavy bolts and rivets. The final goal of our project is to bound adhesively two pieces of
composite materials made of epoxy matrix and carbon fibers.
The aim of the present study is to characterize various simple surface treatments (mechanical treatments like
sandpapering, thermo-oxidation, chemical treatments) thanks to physico-chemical methods. The chemical
modifications of the surfaces are followed by FT-IR measurements. Contact angle measurements give also very useful
additional information on the surface state and allow determining the surface energy  of the material. AFM pictures of
the different samples provide further information on the topography and the roughness of the samples.
KEY WORDS: composites, epoxy, AFM, FT-IR, oxidation
The effects of surface treatments are evaluated with
different experimental techniques: Atomic Force
Microscopy (AFM), contact angle measurement and
Fourier Transform Infra-Red spectroscopy (FT-IR).
1. INTRODUCTION
A brief review of literature shows that plenty of works
exist on adhesion and surface treatments in general but
only a few on epoxy based composite materials. A study
of Zaldivar et al. [1], [2] indicates that the mechanical
quality of adhesion between two composite samples
with an epoxy adhesive joint is directly related to the
oxidation state of the surface. Carboxy groups (-COO)
appear on the surface of materials treated with
atmospheric plasma with oxygen as active gas. The
mechanical strength of the bonded joint is strongly
correlated with the amount of –COO groups on the
surface. Indeed, carboxy groups can easily create strong
covalent bonds with the epoxy joint and thus reinforce
the mechanical strength of the assembly.
The atomic force microscope used in this project is a
Topometrix II. Pictures are recorded with a size of
10x10 µm to determine the nanorugosity of the samples.
The arithmetic rugosity RA is used in this study:
RA 
1
l
l
 y( x) dx
(1)
0
The measure of contact angle allows finding out the
surface energy S of the composite material. Two
different liquids are used (here distilled water and
glycerol) and the surface energy is given by the Owens
and Wendt formula [3]:
2. MATERIALS AND TECHNIQUES
The composite materials of this study are processed
from prepreg sheets furnished by Hexcel Composites.
They are composed of carbon fibers (T700) and epoxy
matrix (M21).
 L (1  cos  )  2  Sd  Ld  2  Sp Lp
with  S   Sd   Sp
These materials are compared in this study:
- mechanical treatments: sandpapering
- chemical treatments with a solution of oxalic
acid,
- thermal treatments to enhance thermooxidation of the surface.
159
(2)
100
90
Rawx and sand papered samples are plunged during one
minute in a saturated solution of oxalic acid. First
observations with the AFM clearly indicate that oxalic
acid has induced a decrease of the rugosity of the
samples (either raw and sandpapered composites).
80
70
Raw
Raw + oxi. 1d
Raw + oxi. 2d
Raw + oxi. 4d
Raw + oxi. 5d
 (mN/m)
60
0,400
50
40
0,350
0,300
Raw + Oxalic Acid
Sandpapering + Oxalic Acid
Raw
Sandpapering
30
20
Nanorugosity (µm)
10
0,250
0
0,200
Figure 3. Evolution of surface energy on raw composite samples after
several days of thermo-oxidation (100°C)
0,150
100
0,100
90
0,050
80
0,000
70
Sandpaper
Sandpaper + oxi. 1d
Sandpaper + oxi. 2d
Sandpaper + oxi. 4d
Sandpaper + oxi. 5d
Figure 1. Nanorugosity of the different samples after acidic treatment
 (mN/m)
60
Indeed, this effect is also seen on the value of surface
energy, obtained through contact angle measurement.
50
40
35
30
30
Raw + Ac.Ox.
Sandpaper + Ac.Ox.
Raw
Sandpaper
20
10
25
 (mN/m)
0
Figure 4. Evolution of surface energy on sandpapered composite
samples after several days of thermo-oxidation (100°C)
20
15
For raw composite materials, the surface energy
globally increases with the time of exposition to
elevated temperatures but it is not so obvious with
sandpapered samples. However, the rugosity of the
samples is not affected by thermo-oxidation and
remains constant either for raw and sandpapered
materials.
Thus the variation of surface energy rugosity is
complex and a compromise has to be found between
chemical and mechanical surface treatments.
10
5
0
Figure 2. Surface energy of treated and non-treated samples after
acidic treatment
Nanorugosity and surface energy of the samples
decrease after chemical treatment with oxalic acid.
These variations are not favorable for further adhesion
on these materials, the aim is to obtain a surface energy
as high as possible.
CONCLUSION
In order to compare efficiently the different surface
treatments presented in this abstract, mechanical tests
with single lap joint will be performed as well as further
analyses with FT-IR in order to identify the chemical
modifications that occurred on the surface.
Another treatment performed on these raw and
sandpapered composite samples is thermo-oxidation.
Samples are placed several days in an oven at 100°C
under ambiant air (no specific atmosphere conditions).
Thermo-oxidation is usually feared for composite
materials because it causes degradation of the surface
but a short exposition to elevated temperatures allows to
modify surface composition without damaging too
much the surface. Both raw and sandpapered samples
are placed in the oven and their surface energy is
followed thanks to contact angle measurements.
160
REFERENCES
[1] R. J. Zaldivar, J. Nokes, G. L. Steckel, H. I. Kim,
and B. A. Morgan, “The Effect of Atmospheric
Plasma Treatment on the Chemistry, Morphology
and Resultant Bonding Behavior of a Pan-Based
Carbon Fiber-Reinforced Epoxy Composite,” J.
Compos. Mater., vol. 44, no. 2, pp. 137–156, 2010.
[2] B. A. Morgan, R. J. Zaldivar, H. I. Kim, G. L.
Steckel, J. A. Chaney, and J. P. Nokes, “Effect of
isopropanol rinse on adhesion of plasma-treated
carbon-fiber reinforced epoxy composites,” J.
Compos. Mater., vol. 45, no. 12, pp. 1331–1336,
2011.
[3] D. K. Owens and R. C. Wendt, “Estimation of the
surface free energy of polymers,” J. Appl. Polym.
Sci., vol. 13, no. 8, pp. 1741–1747, 1969.
161
162
Incidence of wood content on the surface modifications and adhesion of wood
plastic composites (WPCs) treated with oxygen-argon low pressure plasma
Andrés J. Yáñez-Pacios, José Miguel Martín-Martínez
Adhesion and Adhesives Laboratory
University of Alicante, 03080 Alicante, Spain
[email protected]
ABSTRACT
Wood plastic composites (WPCs) are made of lignocellulosic materials from renewable sources and polyolefins, being
considered an alternative to wood. WPCs are used in building, furniture, automotive and engineering applications. In
some applications adhesion of WPCs to coatings, adhesives and varnishes is limited. In this study, a surface treatment
with secondary downstream oxygen/argon low pressure plasma of WPCs with different wood contents is suggested.
Low pressure plasma treatment increases the surface energy of the WPCs due to oxidation and formation of new C=O
and C-O functional groups. Furthermore, the polyethylene on the surface covering the wood fibres of the WPCs was
removed, and ablation was produced. For both WPCs, a noticeable increase in adhesion to Magic Scotch® adhesive
tape was produced, irrespective of the wood content in the composition of the WPC.
Key words: Composites (WPC), low pressure plasma, surface treatment, adhesion.
wood index9 of 59 - and the other contains higher
amount of PE (WPC-High PE) - wood index9 of 332.
The WPCs were cut into pieces of 3x10cm and 2.5x7
cm (low and high PE content respectively) for
characterization.
1. INTRODUCTION
Wood plastic composites (WPCs) are used in decking,
automotive industry and outdoor furniture. Depending
on their polyolefin content, their adhesion to coatings
and adhesives can be limited, and thus a surface
treatment is needed to improve WPC adhesion,
durability and outdoor resistance.
Several surface treatments for improving WPC
adhesion have been proposed in the existing literature15
, including chemical (i.e. chromic acid) and physical
(i.e. corona discharge, roughening, flame, water
immersion) treatments6-11. Although these treatments
were effective, in this study, an argon-oxygen low
pressure plasma treatment is proposed as new surface
treatment for improving adhesion of WPCs
manufactured with different polyethylene contents.
This treatment was selected because it is
environmentally friendly and effective in improving
the adhesion of polyolefins.
A comparative study on the effectiveness of the low
pressure plasma treatment of two WPCs with different
polyethylene content was carried out. The surface
modifications were obtained by contact angle
measurements, ATR-IR spectroscopy and scanning
electron microscopy (SEM). The adhesion properties
were obtained from 180º peel tests in joints made with
Scotch Magic® adhesive tape.
WPC-Low PE
WPC-High PE
Figure 1. WPC samples with different polyethylene content.
2.2. Experimental techniques
Argon:Oxygen (2:1, v/v) low-pressure plasma was
used for surface treatment of the WPCs. Digit Concept
NT1 (BSET EQ, Antioch, CA, USA) radiofrequency
plasma (13.57 MHz) equipment was used.
RF SOURCE
Direct shelf
GAS IN
Valve
PLASMA
VACUUM
Sample
2. EXPERIMENTAL
2.1. Materials
Floating shelf
Ground shelf
The WPC materials used in this study were extruded
wood flour filled with polyethylene (PE) commercial
composites provided by Condepols Company (Jaén,
Spain). According to the manufacturer, the wood
content in the WPCs varies between 70 and 30%, i.e.
one contains lower amount of PE (WPC-Low PE) -
Figure 2. Secondary downstream low pressure plasma scheme.
The plasma chamber has three shelves (floating, power
and ground) and depending on their position direct,
downstream and etching plasmas can be produced. In
163
Low PE produces noticeable etching but ablation is
mainly produced in WPC-High PE. The plasma
treatment decreases the surface roughness in both
WPCs.
this study secondary downstream plasma (sample
placed between the power and the ground shelves) was
selected (Figure 2). Plasma power was 200 W, the
residual gas pressure was 800 mbar (80kPa) and the
length of treatment was varied between 20 and 90
seconds.
WPC-Low PE
0.80
2.3. Characterization techniques
0.70
Absorbance (a.u.)
2.3.1.
ATR-IR
spectroscopy.
The
chemical
modifications in the WPC surface produced by low
pressure plasma treatment were assessed by attenuated
total reflectance infrared spectroscopy (ATR-IR) in
Alpha spectrometer (Bruker Optiks, Etlinger,
Germany) using germanium prism. An angle of
incidence of the IR beam of 45º was used and 60 scans
were averaged with a resolution of 4 cm-1.
2.3.2. Contact angle. The wettability of the WPCs was
obtained by contact angle measurements, using a
Ramé-Hart 100 goniometer (Netcong, NJ, USA). 4µl
drops of two different liquids (water, polar, and
diiodomethane, non-polar) were placed on the WPC
surface, obtaining the contact angle values after 1
minute of deposition. For surface energy calculation,
the Owens-Wendt-Kaelbe approach was used.
2.3.3. Scanning Electron Microscopy (SEM). The
topography changes of the WPCs were measured by
scanning electron microscopy using Jeol JSM-840
(Jeol Ltd., Tokyo, Japan) microscope working at 15kV.
For good contrast, the WPC samples were gold coated
before SEM analysis.
2.3.4. 180º peel adhesion tests. Adhesion was obtained
from 180º peel tests of treated and non-treated WPC
samples and Magic Scotch® adhesive tape joints, using
TA-XT2i texture analyser (Stable Micro Systems,
Godalming, UK); a peeling rate of 1mm/s was used.
The adhesive tape was applied over the WPC by means
of a rubber roller passing it 30 times over the adhesive
tape.
Downstream-60 s
0.60
Downstream-30 s
0.50
0.40
0.30
As-received
0.20
0.10
2950
2900
2850
2800
Wavenumber (cm-1)
WPC-High PE
0.090
0.080
Absorbance (a.u.)
0.070
Downstream-30s
As-received
0.060
0.050
0.040
0.030
Downstream-60s
0.020
0.010
0.000
3600
3400
3200
Wavenumber (cm-1)
Total surface energy
(mJ/m2)
Figure 3. ATR-IR spectra of Ar-O2 downstream low pressure plasma
treated WPCs. Germanium prism.
Whereas in WPC-Low PE, the plasma treatment
increases the relative intensity of polyethylene (2950
and 1450 cm-1), in the WPC-High PE a decrease of the
relative intensity of the cellulose bands is produced
(Figure 3). Furthermore, for both WPCs the plasma
treatment causes an enrichment of the surface in
polyethylene and new polar groups (C-N at 1650 cm-1,
C=O at 1730 cm-1, C-O at 1260 cm-1) are created. The
creation of polar groups is favoured more in the treated
WPC-Low PE.
80
75
70
65
60
55
50
45
40
35
Low PE content
High PE content
0
20
40
60
80
100
RF treatment time (s)
Polar component surface
energy (mJ/m2)
Figure 4a. Variation of the surface energy of the Ar-O2 downstream
low pressure plasma treated WPCs as a function of the length of
time.
The presence of the new polar groups causes an
increase in the wettability and the surface energy of the
Ar-O2 low pressure plasma treated WPCs (Figure 4a),
irrespective of the polyethylene content and the length
of treatment, i.e. a treatment of 20-30 seconds is
sufficient for improving the surface energy of the
WPCs. For both plasma treated WPCs, the increase in
the surface energy is due to the increase in the polar
component of the surface energy mainly (Figure 4b).
45
40
35
30
25
20
15
10
5
0
Low PE content
High PE content
0
20
40
60
80
100
RF treatment time (s)
Figure 4b. Variation of the polar component of the surface energy of
the Ar-O2 downstream low pressure plasma treated WPCs as a
function of the length of time.
SEM micrographs (Figure 5) show the changes in the
topography of the WPCs. Plasma treatment of WPC-
164
Because of the increase in wettability, the increase of
surface energy, the creation of new polar groups and
the reduction in roughness caused by the plasma
treatment of the WPCs, improved adhesion can be
expected. Figure 6 shows the peel adhesion of joints
made with adhesive tape and the as-received and
plasma treated WPCs. The selected conditions of
plasma treatment for each WPC were the optimal for
each material, i.e. 60 seconds for WPC-Low PE and 90
seconds for WPC-High PE. The adhesion of both
plasma treated WPCs was notably improved, and
higher adhesion was obtained in the joints made with
the plasma treated WPC-High PE. Therefore, the
adhesion was determined mainly by the reduction in
roughness and the ablation of the WPC surface rather
than by the increase in surface energy and the creation
of polar groups on the surface.
As-received Low PE
content WPC
4. CONCLUSIONS
New Ar-O2 downstream low pressure plasma surface
treatment has been shown effective in improving the
surface properties and the adhesion of WPCs with
different polyethylene content. The treatment created
new surface polar groups (C-O, C-N, C=O) and
reduced the cellulose content on the WPC surface
(irrespective of the polyethylene content in the WPC).
Wettability of the WPCs was increased after plasma
treatment and the surface energy too, due to improved
polar component of the surface energy mainly. On the
other hand, the roughness of the WPC was reduced
after plasma treatment and ablation was dominant.
Adhesion measured with 180º peel tests of WPCacrylic adhesive tape was increased after plasma
treatment, mainly in the WPC-High PE, likely due to
the dominant effect of the reduction in roughness and
ablation caused. In summary, the surface treatment
with Ar-O2 downstream low pressure plasma increased
the adhesion of WPC materials containing
polyethylene.
60s downstream LPP
Low PE content
5. ACKNOWLEDGEMENTS
Financial support by Economy and Competitiveness
Ministry of Spain, INNPACTO project IPT-20111454-020000 is acknowledged.
100 μm
As-received High PE
content WPC
100 μm
6. REFERENCES
[1] Dányádi, L., Móczó, J., Pukánszky, B. “Effect of
various surface modifications of wood flour on the
properties of PP/wood composites”. Composites: Part
A, 41, pp. 199-206, 2010.
60s downstream LPP
High PE content
[2] Gramlich, W.M., Gardner, D.J., Neivandt, D.J.
“Surface treatments of wood-plastic composites
(WPCs) to improve adhesion”. Journal of Adhesion
Science and Technology, 20, pp. 1873-1887, 2006.
[3] Jeske, H., Schirp, A., Cornelius, F. “Development
of a thermogravimetric analysis (TGA) method for
quantitative analysis of wood flour and polypropylene
in wood plastic composites (WPC)”, Thermochimica
Acta, pp. 165-171, 2012.
100 μm
100 μm
Figure 5. SEM micrographs of the as-received and Ar-O2
downstream low pressure plasma treated WPCs.
180º peel strength (N/m)
200
As-received
[4] Kim, M., Kim, H.S., Lim, J.Y., “A study on the
effect of plasma treatment for waste wood
biocomposites”. Journal of Nanomaterials, 2013,
Article ID 138083.
Optimal LPP treatment
[5] Kuruvilla, J., Sabu, T., Pavithran, C. “Effect of
chemical treatment on the tensile properties of short
sisal fibre-reinforced polyethylene composites”.
Polymer, 37, pp. 5139-5149, 1996.
150
100
[6] Li, Y. “Effect of woody biomass surface free energy
on the mechanical properties and interface of
wood/polypropylene composites”. Journal of Adhesion
Science and Technology, 28, pp. 215-224, 2013.
50
0
Low PE content
[7] Moghadamzadeh, H., Rahimi, H., Asadollahzadeh,
M., Hemmati, A.R. “Surface treatment of wood
polymer composites for adhesive bonding”.
High PE content
Figure 6. 180º peel adhesion values of as-received and Ar-O2
downstream low pressure plasma treated WPC/Magic Scotch® tape
joints.
165
International Journal of Adhesion and Adhesives, 31,
pp. 816–821, 2011.
[8] Oporto, G.S., Gardner, D.J., Bernhardt, G.,
Neivandt, D.J. “Characterizing the mechanism of
improved adhesion of modified wood plastic composite
(WPC) surfaces”. Journal of Adhesion Science and
Technology, 21, pp. 1097-1116, 2007.
[9] Stark, N.M., Matuana, L.M., Clemons C.M.,
“Effect of processing method on surface and
weathering characteristics of wood–flour/HDPE
composites”, Journal of Applied Polymer Science, 93,
pp. 1021–1030, 2004.
[10] Wingfield, J.R.J. “Treatment of composite
surfaces for adhesive bonding”. International Journal
of Adhesion and Adhesives, 13, pp. 151-156, 1993.
[11] Wolkenhauer, A., Avramidis, G., Hauswald, E.
Militz, H., Viöl, W., “Plasma treatment of WoodPlastic Composites to enhance their adhesion
properties”. Journal of Adhesion Science and
Technology, 22, pp. 2025-2037, 2008.
166
Improvement in the adhesion property of CFRP materials
and the water-based paint coating and ink-jet printing of
silicone rubber, PP and PET films, etc
Hitoshi Kanazawa1*, Aya Inada1 , Takuto Tanaka1, Yuki Yamaguchi 1
1
Fukushima University,
*[email protected]
The improvement in the adhesive property of carbon fiber reinforced plastics (CFRPs) was carried out by the
combination method which is composed of an activation process and a chemical reaction process. In addition, the
water-based paint coating and ink-jet printing of silicone rubber, polypropylene and PET films, etc. were studied by
developing the process. Modified carbon fiber reinforced epoxy resin boards were well bonded to each other using a
film adhesive. Carbon fiber reinforced poly(ether ether ketone) boards were also modified and the adhesion property
was improved well. Our method gave good results for the improvement in the water-based paint coating and the ink-jet
printing of silicone rubber, polypropylene and PET films, etc.
KEY WORDS: surface modification, adhesive property, CFRP, ink-jet printing, water-based paint coating, silicone resin
based paint coating and ink-jet printing of silicone
rubber, polypropylene and PET films, etc. were studied.
1. INTRODUCTION
Polyolefins such as polypropylene (PP), polyethylene
(PE) or ultra-high molecular weight polyethylene
(UHMWPE) and poly(methyl pentene) (PMP), etc. have
both high tensile strength and resistance to chemical
reagents. However, these materials cannot be bonded to
each other or other materials with usual adhesives.
Many techniques were examined to improve the
adhesive property,[1,2] but, the modified surface
property was changed with time.
A corona discharge treatment was extensively used for
the improvement of the adhesive property of polymeric
materials such as polypropylene and polyester films,
etc. But, the corona process is known to be not effective
for the materials such as silicone resin and
polymethylpentene (polyacetal). In addition, the corona
treated materials must be processed quickly after the
treatment. Thus, the corona process is not preferable for
large and heavy materials.
We tried to combine two or three methods, and found
that the combination of an activation process by one of
old techniques and a chemical treatment with monomer
or polymers was effective for the modification of many
kinds of chemical stable polymeric materials.[3,4] This
method is composed of an activation process and a
chemical reaction process, and is named as “KANA1
surface modification method”.
We improved
furthermore our method, and tried the improvement in
the adhesive property of carbon fiber reinforced
plastics (CFRPs) which are useful for the materials
of automobiles and aircrafts. In addition, the water-
2. EXPERIMENTAL
2.1 Materials
Polymeric materials: polypropylene(PP), polyethylene (PE),
ultrahigh molecular weight PE
(UHMWPE),
polycarbonate(PC),
polyethyleneterephthalate (PET), silicone resin, fluorocarbon resins
(PTFE, PETFE).
Engineering plastics: polyacetal
(POM), polybutylene-terephthalate (PBT), polyamides
(PA), polysulfone (PS), polyimide (PI), polyether ether
ketone (PEEK). Fiber-reinforced plastics (FRP) ; carbon
fiber reinforced epoxy resin (CFR-EPO) and carbon
fiber reinforced PEEK (CFR-PEEK). The materials of
films, sheets, fibers, boards, rods and tubes were used
after washing in methanol. Reagents: hydrophilic
monomers, polymers and catalysts were used after a
general purification.
Adhesives: polyvinylpyrrolidone (PVP), starch, a
woodwork bond (polyvinylacetate (PVAC)-water
mixture : Konishi Co. Jp), cyanoacrylate adhesive (CA;
Aron alpha: Toa Gosei Kagaku Co.), epoxy resin
adhesive (Quick 5 ; mixture of epoxy resin and
polythiol, Konishi Co. Ltd.), a cyanoacrylate-primer set
(Cemedine PPX; primer : organic amine 1% and
heptane 99%; Cemedine Co. Ltd.), etc. were used.
Paints: commercial water-based paints (Asahipen,
Co. Ltd., Japan) which contain synthetic resins (acrylic,
167
treatment. On the other hand, the silicone rubber sheets
treated by our method gave good coating property even
five years after the processing.
silicone and fluorocarbon resins), pigments and organic
solvents, and water-based urethane car-use paints were
used.
2.2 Treatment
The present method (the KANA1 surface modification
method ) is composed of two processes, an activation
process and a chemical reaction process. First,
polymeric materials were activated by a chemical
oxidation, or energy irradiations such as UV, corona,
plasma, high-voltage electric discharge treatments.
Subsequently, the activated polymeric materials were
treated with solutions of polymers, monomers or other
chemical reagents in the presence of catalysts or
initiators.
2.3 Measurements
Adhesion strength: a tensile shear bond strength test
and a T peel test were carried out using a Shimadzu
AGS-H5KN.
IR spectra: polymeric materials were observed by a
Shimadzu IRPrestige-21 equipped with a Smiths
DuraSampl IR II (ATR accessory).
Cross-cut test: JIS 5400 (Japanese Industrial Standards
Committee) was employed for the peeling test of paints
and inks.
3.2 Ink-jet printing with water-based paints
Polymer films (PP, PE, PET, PC, silicone resin, etc.)
were modified by the present method, and a postprocessing suitable for the inkjet printing with waterbased ink was investigated.
Figures 2 gives the results for PP films ; the used ink
has the best printing property among the three inks. The
untreated PP film was writable little, but it is peeled by
rubbing. But, the PP film modified by our method
indicated a good printing property.
Figure 3 gives the results for PET films. The PET film
modified by our method indicated a good printing
property. But, it was impossible to print on the untreated
PET film.
Figure 4 gives the results for silicone rubber sheets.
The silicone rubber sheet modified by the present
method indicated a good printing property. But, it was
impossible to print on the untreated silicone sheet.
Thus, modified polymeric materials were printed by the
ink-jet printer with water-based inks. It is preferable to
make a coating on the surface of printed materials using
transparent coating materials in order to protect the
surface.
2.4 Ink-jet printer
Ink-jet printers, Dainichi-seika printer (P1), Canon
MG8230 (P2), and Epson PX105 (P3) were used with
water-based pigment inks. The ink of the P1 is the most
excellent for the printing on several plastics.
3. RESULT AND DISCUSSION
3.1 Water-based paint coating
Polymeric materials modified by the present
technique were coated well with water-based paints.
Modified PP boards were coated with a water-based
paint (for a car use) and dipped in water at 40˚C for 20
days. Subsequently, the cross-cut test was carried out.
The PP boards gave a full-mark, 100/100 in the crosscut test. These requirements are available for the car
use. The specimens are given in Figure 1.
Silicone rubber sheets cannot be coated with a waterbased paint. Corona-discharged silicon rubber sheets
have to be coated with water-based paints immediately
after the treatment; we confirmed that the coating was
impossibl
e 5 hours
after the
corona
discharge
Fig.1 Cross-cut test resut of PP car use
boards; untreated PP (upper; 0/100) and
modified one (lower; 100/100)
168
Fig.2 Modified PP (lower) and untreated
PP (upper) films printed by an ink-jet
printer with a water-based resin ink
(Dainichi Seika Co. Ltd.)
Fig.4 Modified silicone rubber (lower) and
untreated PP (upper) sheets printed by the
printer, P1(Dainichi Seika Co. Ltd.)
3.3 Adhesion of CFRPs
CFR-EPOXY boards were modified by the
present method and they were bonded to each other
using the film type epoxy resin adhesive. Untreated
CFR-EPO boards gave around tensile shear
strength 20MPa, and the adhesive gave an interface
peeling, On the other hand, the modified one gave
50MPa, and a cohesive failure was observed. The
resuts is given in Figure 5.
Fig.5 CFR-EPOXY boards used in the
tensile shear test. Upper :untreted one
and lower: modified one
Fig.3 Modified PET film printed by an inkjet printer, Epson PX-105 with a waterbased resin ink.
169
3.4 Solvent bonding
PP, PE, PMP and silicone resin tubes modified by the
present method are strongly connected to each other by
the solvent bonding. This technique is preferable for the
medical use. Figure 5 gives a PP intravenous drip
chamber modified by the present method connected to
polybutadiene tube by the solvent bonding.
Modified and connected
part
Fig.6 PP intravenous drip chamber
modified by the present method
connected to polybutadiene tube by
the solvent bonding
3.5 Mechanism
The mechanism of the present process was speculated
with reference to the old studies [3] , and it is given in
Scheme 1. Several oxidized polymer structures were
proposed. Oxidized polymers (P-OOH or P=CO)
formed in the activation process are considered to react
with P-X (functional compound) to form product
“Polymer-OR”.
The
Polymer-OR
gives
the
modification.
4.
Scheme 1
REFERENCES
[1] R.H. Young, Sr., et. al., United State
Patent, No.5432000 (1995).
[2] M. Kinoshita, Japanese Patent
Application, No.09012752 (1997).
[3] H. Kanazawa, USA Patents No.7294673
and No. 6830782B2.
[4] H. Kanazawa, Japanese Patent
No.4229421, 3729130, 4941794.
[5] T. Matsui, A. Yamaoka, and Y. Yamaguchi,
Nippon Kagaku Kaishi, 732 (1992).
CONCLUSIONS
The modified polymeric materialswre printed by ink-jet
printers using water-based ink. The present, KANA1
method is useful for the modification of many kinds of
polymeric materials. Especially, the present method is
useful for reducing the weight of automobiles and
airplanes by applying in composite materials (CFRP).
170
SURFACE CHARACTERIZATION AND COMPARISON OF SURFACE
PRETREATED THERMOPLASTIC POLYOLEFINS –
RELEVANCE TO ADHESION FOR WATERBORNE PAINTS
K. Grundke1, K. Estel1, A. Marschner1, C. Bellmann1, B. Joos-Müller2 and A. Stoll3
1
2
Leibniz-Institut für Polymerfoschung Dresden e.V.
Hohe Straße 6, D-01069 Dresden, Germany
[email protected]
Fraunhofer-Institut für Produktionstechnik und Automatisierung IPA
Nobelstraße 12, 70569 Stuttgart, Germany
[email protected]
3
Forschungsinstitut für Leder und Kunststoffbahnen Freiberg
Meißner Ring 1, 09599 Freiberg, Germany
[email protected]
ABSTRACT
Three surface pretreatment techniques- flame, plasma jet treatment and gas phase fluorination- were compared to
study the modification effects of commercially available thermoplastic polyolefin materials (TPO). The pretreatment
parameters were systematically changed under laboratory conditions to simulate under- and overtreatments and
partly compared with pretreatment techniques carried out in a paint line for TPO parts in the industry. The surfaces
have been characterized with regard to their chemical composition (X-ray photoelectron spectroscopy-XPS,
gravimetric analysis in combination with infrared spectroscopy), surface-energetic properties (advancing and
receding water contact angles), acid-base characteristics (zeta potential vs. pH) and surface roughness (confocal
optical and scanning force microscopy). The adhesion of a waterborne paint system was estimated by vapor jet tests.
Depending on the treatment parameters oxygen- and nitrogen-containing functional groups were introduced in the
surface region of the pre-treated TPO samples. Whereas flaming and gas phase fluorination had only a minor
influence on the surface roughness and heterogeneity of the TPO surfaces, they were dramatically increased after
plasma jet treatments. By optimizing the treatment conditions surface roughness could be kept unaltered. Vapour jet
tests showed a relation between the lowest and highest oxygen content in the surface region and paint adhesion. But,
in several cases, no relation between oxygen content and paint adhesion was observed. XPS analysis of the fracture
surfaces showed that interfacial failure predominated when the modification effect was too small. Cohesive failure in
the TPO material predominated when the surface was over-treated. Higher amounts of additives in the surface region
of TPO materials caused insufficient paint adhesion.
KEY WORDS: thermoplastic polyolefin, flame treatment, plasma jet treatment, gas phase fluorination, X-ray photoelectron spectroscopy, contact angle,
zeta potential, confocal optical microscopy, waterborne paint, vapor jet test.
95 % of the TPO bumpers for the automotive industry
in Europe are flame treated [1]. The polymer surface is
oxidized by creating a wide range of oxygen containing
chemical groups in a very thin surface region.
Additionally, cross-linking and chain scission
phenomena occur. While flame treatment is widely
used several disadvantages in a commercial volume
production environment are discussed, such as an
overtreatment of parts, an incipient melting as well as
the hazards associated with combustible gas [2].
Plasma pretreatments have also been used in the
automotive industry to modify the surface of TPO parts
[2-4]. The vacuum type of the plasma process requires
the parts to be treated under low vacuum pressure in a
1. INTRODUCTION
Thermoplastic polyolefins (TPO) are widely used in the
automotive industry for the manufacturing of bumpers.
For esthetic and protective reasons, the decorative
painting of bumpers is desirable. However, untreated
TPO have inherent poor adhesion properties due to
their low surface free energy and inert chemical nature.
Effective surface modification techniques are, therefore,
needed to improve the adhesion properties of TPO. In
industrial applications, the flame treatment of complex
3-D automotive shapes, such as bumpers, has been used
for many years to obtain good paint adhesion. Almost
171
the TPO materials were also flame treated under
industrial conditions. The plasma jet treatment was
carried out under laboratory conditions using a static or
a rotating nozzle. Gas phase fluorination was also
performed under laboratory conditions in a tunnel
machine using a gas mixture of F2/N2/O2
(oxyfluorination) and in a batch process using a
mixture of F2/N2. The fluorine concentration and time
of exposure were varied.
XPS measurements were carried out using an Axis
Ultra spectrometer (Kratos Analytical, Manchester,
UK) to quantify the chemical surface composition. The
maximum information depth was about 8 nm. Highresolution XPS spectra were carried out to analyze the
binding states of the elements. Advancing ( adv) and
receding ( rec) water contact angles were measured by
the sessile drop technique using a commercially
available device (OCA 35XL, DataPhysics, Germany).
For each TPO sample the contact angle data obtained
from six individual drops on different surface areas
were averaged. Electrokinetic measurements were
carried out to determine the zeta-potential of the TPO
surface as a function of the pH values of an electrolyte
solution using the streaming potential method. A pair
of equally treated samples of a size of 1 x 2 cm2 were
analyzed with an electrokinetic analyzer EKA of Anton
Paar (Graz, Austria). The surface topography was
investigated by light microscopy based on a confocal
technique
(Nanofocus-µSurf,
NanoFocus
AG,
Oberhausen, Germany) and by scanning force
microscopy (SFM, Dimension 3100, Veeco Instr., Inc.,
USA)). Arithmetic mean roughness parameters (Ra)
were determined from topographical images to quantify
the
surface
roughness
(µSurf:
scan
size
1600 x 1600 µm; SFM: scan size 50 x 50 µm2). Ra
values were averaged from six up to nine different
surface areas on each TPO sheet. The amount of
soluble components in the surface region was
determined gravimetrically before and after surface
pretreatments. Their chemical characterization was
carried out by infrared spectroscopy.
The adhesion of the paint systems was estimated by a
vapour jet test (DIN 55662 draft standard) after
conditioning (7 day at 23 °C and 50 % rel. humidity)
and after an additional TWT test (3 cycles each with
15 h 105 °C, 30 min RT, 8 h -40 °C, 30 min RT) of the
coated TPO sheets which contained a cross-cut. The
adhesion level was assessed by a number between 0 (no
damage to the coating) and 5 (a large area of
destruction of the lacquer layer).
chamber. A gas, such as oxygen, is introduced and
ionized resulting in complex chemical and
topographical changes in the near surface region of the
TPO parts. This method is limited by the size of the
chamber and is a batch process. Atmospheric plasma
pretreatment techniques are more attractive since the
parts are treated in situ rather than in a chamber [5].
Despite these surface modifications problems with the
paintability of the TPO parts cannot be avoided up to
now, especially when conventional solvent-borne paint
systems are replaced by waterborne systems. Often, a
partial detachment of the lacquer is described. Thereby,
it is not clear whether it is an interfacial failure
between lacquer and TPO or a cohesive failure in a
near-surface layer of the TPO material.
Here, we compare three surface pretreatment
techniques- flame, plasma jet treatment and gas phase
fluorination - to study the modification effects of
commercially available TPO materials.
The
pretreatment conditions were systematically changed
under laboratory conditions to simulate under- and
overtreatments and partly compared with pretreatment
techniques carried out under industrial conditions.
It was the aim of this work to compare the modification
effects of the three gas-phase methods, especially with
regard to the improvement of the adhesion of
waterborne paint systems on TPO and to a better
understanding of the relations between chemical
composition, surface energetic properties and surface
morphology of the pretreated TPO and its paintability.
2. MATERIALS AND METHODS
Commercially available TPO materials are blends that
typically contain an additional rubber, inorganic fillers
such as talc, and various additives and are developed
for each specific customer. We used commercially
available materials from Basell, Total, Borealis and
Sabic. These materials are blends that contain
polypropylene (PP) and an additional rubber (ethylene
propylene-diene rubber- EPDM) as well as additives
and talc. The talc content varied between 10 %, 20 %
and 30 %. The exact chemical composition of the
blends was not known. In the text, the different
materials are designated with A, B, C, D, E, and F. For
the present study, test panels of TPO were prepared by
injection molding.
The lacquer system consisted of a 2K hydroprimer
(light gray, conductive), a base coat (iridiumsilver MB9775), and a 2K clear coat (high gloss finished)
provided by Wörwag company, Germany. Primer and
base coat were waterborne paints, the clear coat was
solvent borne.
The flame treatment was performed using a laboratory
equipment. An air/ propane (22:1) mixture was used
and the exposure time (flame velocity), the burner
capacity and the flame distance were varied to simulate
weak and strong modification effects. For comparison,
Due to the complexity of the TPO materials their
surface composition might be influenced in each step of
the paint line for the plastic parts, from injection
molding conditions to surface cleaning and surface
pretreatment up to paint application. Serious troubles
172
may be caused by incorrect surface pretreatments. The
application of suitable surface characterization
techniques can help to reveal these problems.
In a first step, the initial surface properties of the
untreated TPO materials have been characterized. The
vertical surface roughness varied between 200 nm and
500 nm (Ra values, scan size 1600 x 1600 µm) for the
untreated test panels and different types of TPO
material. With increasing talc content (from 10 % up to
30 %), a small increase in the mean height roughness
(from 180 nm up to 230 nm) was observed. All
untreated materials were hydrophobic with mean
advancing water contact angles between 100° and 106°
and receding angles between 83° and 88°. The small
oxygen content in the surface region ([O]:[C] < 0.01)
confirmed the hydrophobic nature of the untreated
surfaces. The estimated surface free energy is about
23 mJ/m2.
untreated and flame treated TPO (E, 20 % talc). No
significant differences in the surface properties were
detected after a laboratory cleaning procedure using
isopropanol/ water mixtures, an industrially applied dry
ice blasting and a power wash cleaning. When the
differently cleaned surfaces were flame treated, no
influence of the cleaning procedure could be revealed.
Independent of the cleaning procedure the wettability
was improved by the flame treatment, the oxygen
content in the surface region increased as well as the
availability of acidic functional surface groups. No
changes in surface roughness were detected.
atomic concentration [%]
2,5
2,0
1,5
0,5
Zetapotential [mV]
u
a
b
w
BFa02
0,07
[O]:[C]
0,06
0,05
untreated, unwashed
isopropanol/ water
dry ice blasting (BMW)
power wash (Wörwag)
flame treatment 2
0,04
u BFa BFa BFa BFw
01 02 03 x
u BFa BFa BFa BFw
01 02 03 x
u BFa BFa BFa BFw
01 02 03 x
u BFa BFa BFa BFw
01 02 03 x
material A
material B
material C
material E
25
20
15
10
5
0
-5
-10
-15
-20
-25
-30
-35
-40
BFa02 flame treatment 2
BFwx industrial flame treatment
A-BFwx
B-BFwx
C-BFwx
E-BFwx
A-BFa02
B-BFa02
C-BFa02
E-BFa02
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
pH
0,03
0,02
Figure 2. Effect of different flame procedures under laboratory (BFa)
and industrial (BFw) conditions on the surface properties of four
commercially available TPO materials
0,01
0,00
Zetapotential [mV]
untreated, unwashed
flame treatment 1
flame treatment 2
flame treatment 3
industrial flame treatment
1,0
0,0
0,08
nitrogen content of flame treated surfaces
u
BFa01
BFa02
BFa03
BFwx
u
25
20
15
10
5
0
-5
-10
-15
-20
-25
-30
-35
-40
0a
0b
0w
0a
BFa02
0b
BFa02
0w
BFa02
It is well known that by changing the flame treatment
conditions the kind and amount of functional surface
groups, such as C-O, C=O, O-C=O, C-NH2, C-NH3+,
C-NOx, can be influenced. Figure 2 shows exemplarily
the chemical surface composition (here the nitrogen
content) and acid-base surface characteristics (zeta
potential vs. pH) after a flame treatment under
laboratory conditions and under industrial conditions
for four TPO materials. The decrease in the advancing
water contact angles to values of about 80° was
comparable for both procedures, just as the insertion of
oxygen containing groups. However, remarkable
differences were found regarding the amount of
nitrogen containing functional surface groups. Under
the specific industrial conditions for the flame
treatment, a higher content of nitrogen could be
detected in the surface region of all TPO materials.
Streaming potential measurements confirmed this
behavior by revealing a more basic surface character.
0a
0b
0w
0a-BFa02
0a-BFa02
0a-BFa02
a isopropanol/ water
b dry ice blasting (BMW)
w power wash (Wörwag)
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
pH
Figure 1. Effect of different cleaning procedures on the surface
properties of untreated and flame treated TPO (E) determined by
contact angle, XPS and streaming potential measurements
Figure 1 shows exemplarily the influence of different
cleaning procedures on the surface properties of
173
Figure 3 shows results of the vapor jet test regarding
the adhesion of a waterborne paint system on flame
treated TPO materials. In the case of TPO materials A
and C, destruction of the lacquer layer was observed.
This behaviour correlated with a higher amount of
soluble constituents in the surface region after the
flame treatment of these materials. It was found that the
soluble constituents contain additives that migrate to
the outermost surface. Obviously, these additives
contribute to the observed effects after the vapour jet
tests. There were several hints that the filler content in
the surface region had also an influence on the paint
adhesion.
25
quantity
1
4
10
13
quantity
BFa
01
10
10
C
D
E
F
gas phase fluorination
plasma jet
static nozzle
plasma jet
rotating nozzle
Ra [µm]
1,5
1,0
0,5
0,0
UG, BFa BFa BFa BFw GFa GFa GFa GFa GFa GFa GFa GFa PLa PLa PLa PLa PLa PLa RPLa RPLa RPLa RPLa RPLa RPLa
UB 01 02 03
x
01 02 03 04 06 07 08 09 02 03 04 08 10 14 15 16 17 18 19 20
0,5
[O]:[C]
0,4
flame
treated
BFa
02
BFa
03
15
15
7
GFa
02
GFa
03
PLa
08
15
4
RPLa
16
In Figure 4, the modification effects of the flame
treatments of the TPO material E are compared with
the other two surface treatment techniques. Flaming
and gas phase fluorination had only a minor influence
on the surface roughness and heterogeneity. In the case
of the plasma jet treatment, the vertical roughness and
surface heterogeneity of the TPO material was
dramatically increased depending on the treatment
parameters. However, by optimizing the treatment
conditions surface roughness could be kept unaltered.
High fluctuations were also observed regarding the
oxygen content in the outermost surface region. The
vapor jet tests showed a relation between the oxygen
content in the surface region and the adhesion. For the
lowest and the highest oxygen content, the adhesion
was insufficient (Figure 5). In the first case, the
modification effect was too small. In the second case,
the surface was obviously over-treated. However, in
several cases where the oxygen content varied in the
surface region, no effect on the adhesion has been
observed. XPS analysis of the fracture surfaces after
removing the lacquer showed that interfacial failure
predominated when the modification effect was too
small. When the surface was over-treated a cohesive
failure in the TPO material predominated.
Figure 3. Results of vapour jet tests using different TPO materials that
were flame treated (BFa02: flame treatment 2) under laboratory
conditions
flame
treated
10
Figure 5. Results of the vapour jet tests for TPO material E after
different surface pretreatments; BFa: flame treatment, GFa: gas
phase fluorination, PLa: plasma jet static nozzle, RPLa: plasma jet
rotating nozzle
material
2,0
10
5
0
B
10
0
15
12
A
3
10
3
17
5
15
5
3
1
characteristic value 3-5
characteristic value 2
20
characteristic value 2
20
15
25
gas phase fluorination
plasma jet
static nozzle
plasma jet
rotating nozzle
0,3
0,2
0,1
ACKNOWLEDGEMENTS
0,0
UB BFa BFa BFa BFw GFa GFa GFa GFa GFa GFa GFa GFa PLa PLa PLa PLa PLa PLa RPla RPla RPla RPla RPla RPla
01 02 03
x
01 02 03 04 06 07 08 09 02 03 04 08 10 14 15 16 17 18 19 20
This project (no. 387 ZBG) was funded via AiF by the
German Federal Ministry for Economic Affairs and
Energy within the governmental R&D-support-measure
“Industrial cooperative research”. The authors want to
be grateful to the companies of the project board
especially BMW, Daimler, Plasmatreat and Wörwag
for making available the materials investigated as well
as their support in carrying out the experiments.
Figure 4. Comparison of different surface pretreatments with regard
to surface roughness (Ra value) and chemical surface composition
(oxygen content)
174
REFERENCES
[1] Cremers, J., paper at DFO-Tagung “Kunststofflackierung 2007”, 6th and 7th of March, 2007,
Hof, Germany
[2] Stewart, R., Goodship, V., Guild, F., Green, M.
and Farrow J., Internat. J. Adhes. & Adhesives 25
(2005) 93-99.
[3] Green, M.D., Guild, F.J. and Adams, R.D.,
Internat. J. Adhes. & Adhesives 22 (2002) 81-90
[4] Nihlstrand, A., Hjertberg, T. and Johansson K.,
Polymer 38 (1997) 3591-3599
[5] Noeske, M., Degenhardt, J., Strudthoff, S. and
Lommatzsch U., Internat. J. Adhes. & Adhesives
24 (2004) 171-177
175
176
EVALUATION OF PRIMER-LESS METHYL-METHACRYLATE
ADHESIVE TECHNOLOGY AND SURFACE TREATMENTS FOR
ADHESION PROMOTION
Gavin Creech1, Victoire Bresset2, Heather Puddephatt1, Dean Bugg1
1
Scott Bader Company Ltd.,
Wollaston, Northamptonshire, NN29 7RL, UK.
[email protected]
2
Cranfield University,
Bedfordshire, MK43 0AL, UK.
ABSTRACT
Two component toughened acrylic adhesives are becoming increasingly attractive in assembly processes, achieving
structural bonding performance on a range of substrate materials with short fixture times and ambient temperature
processing. Typically, acrylic adhesives can require some surface treatment of the substrates to be bonded, in
addition to solvent cleaning. Treatment of metal substrates can include abrasion or application of an adhesion
promoter, wetting agent or corrosion inhibitor if long-term performance or ultimate bond strength is required.
However, recent developments have led to „primer-less‟ formulations which can achieve similar performance to that
obtained by pre-treating the substrate surface, but negate this processing step and thus reduce manufacturing costs
and labour requirements.
The performance effects of applying four classes of adhesion promoter to aluminium substrates are evaluated for
adhesion performance and compared to a commercial primer-less adhesive. The comparison includes evaluation of
the ultimate mechanical performance through lap-shear and peel testing of bonded aluminium substrates, in addition
to the effects of environmental aging by salt-spray. Liquid contact angle measurements to the primed substrate
surface are investigated for correlation of surface wettability to the measured mechanical performance.
KEY WORDS: acrylic adhesive, surfactant, coupling agent, wettability, lap-shear, peel
which cure by free-radical polymerisation, offering
ambient temperature processing and rapid curing while
being able to bond most substrates, including metals,
plastics and composites.
1. INTRODUCTION
Adhesives are a commercially attractive and rapidly
growing area of manufacturing technology due to
reducing costs, increasing manufacturing speed and
imparting performance benefits when comparison to
conventional joining technologies, such as mechanical
fastening and welding. Automotive, aerospace, wind
energy, land transport and marine are all examples of
applications where primary bonded structures are
required to withstand demanding load conditions, for the
lifespan of the structure. Presently, the range of
adhesives suitable for bonding of primary structures is
vast, with epoxy, toughened acrylic, urethane acrylate
and polyurethane chemistries showing potential to
replace traditional joining methods.
Toughened acrylic adhesives (and many other
chemistries of adhesive) commonly require some surface
treatment of the substrates to be bonded to achieve long
term durability, especially when bonded joints are
exposed to harsh environmental conditions. Treatments
can include abrasion to improve mechanical interlocking
of the adhesive at the substrate surface, however, further
performance increases can be achieved through
application of an adhesion promoter, wetting agent or
corrosion inhibitor to the substrate surfaces prior to
bonding [1]. The aim of this substrate treatment is to
increase the amount of chemical bonding at the
adhesive-to-substrate interface, increase the surface
wetting of the substrate or reduce degradation of the
substrate material at the adhesive interface.
Toughened acrylic adhesives are one class of material
that is currently receiving rapid growth within
manufacturing industries due to achieving a balance of
mechanical performance and processing simplicity.
Typically, this adhesive type is supplied as two
components, comprising an adhesive and hardener,
2. SCOPE
177
the adhesive and an inorganic functionality to attach
with the substrate. Ideally the layer applied would be one
molecular layer thick. They can be applied by brushing,
wiping, spraying or dipping of the whole surface and are
commonly used commercially to enhance the adhesion
properties of resin matrixes with glass and carbon fibres.
The current study examines the performance of a
commercial toughened acrylic adhesive forumulation in
bonding substrates treated with a selection of common
coupling agents and surfactants. These are compared
directly through mechanical testing of bonded joints both
before and after periods of simulated environmental
conditioning, to ascertain the surface treatment effect on
long term joint durability. Surface wettability of the
treated substrates is also measured.
Table 2. Details of the surface preparation options evaluated within this
study
3. MATERIALS
Adhesion
promoter /
Surfactant
type
3.1. Adhesive selection
Silane
Titanate
The commercial grade of toughened acrylic adhesive, is
Crestabond M1-20, supplied by Scott Bader Company.
This adhesive is a fast curing, two-component
methacrylate based structural adhesive with a 10 to 1
mix ratio (adhesive to activator). This adhesive was
developed with the aim of reducing the need for excess
substrate preparation by including within the formulation
a combination of adhesion prompters to improve the
mechanical performance and durability of bonded joints
for a range of substrate materials. This has additional
advantages of reducing the number of processing steps
necessary in use and therefore reduces cost and improves
the robustness of the joining process.
Zirconate
Chemical type
A
B
A
A
B
Amino silane
Epoxy silane
Monoalkoxy titanate
Neoalkoxy zirconate
Coordinate zirconate
Non-ionic
flourosurfactant
Non-ionic
flourosurfactant
A
Surfactant
B
No surface
preparation
No-primer
N/A
Control
Crestabond
M1-20
Not disclosed due to
commercial
sensitivity
The most common commercial grade of coupling agent
is silanes. Their usual form is R’-Si-(OR)3 where R’ is
the group reacting with organic materials, and can be a
vinyl, epoxy, amino or acrylic function. The OR group
reacts with the inorganic materials and is usually a
methoxy, ethoxy or acetoxy group.
The adhesive formulation or adhesion promoters used
within this adhesive is not defined within this paper for
reasons of commercial sensitivity. A summary of the cast
adhesive mechanical performance is included in table 1.
Table 1. Cast tensile properties of the toughened acrylic adhesive,
tested to standard method ISO 527-2:1996.
Material Property
Modulus of elasticity
Tensile elongation at break
Tensile peak stress
Designation
Metal-based coupling agents such as titanium or
zirconium have a similar chemistry to silanes and can
also be used as adhesion promoters. Their action on the
adhesive-to-adherand interface is very similar to the
mechanisms attributed to silane coupling agents,
however when coupled with a metal substrate the
chemical bond is typically more hydrolytically stable and
can further improve joint durability.
Value
0.6 - 0.85 GPa
65 - 70 %
17 - 19 MPa
Within this study, the commercial grade of adhesive is
used as the control sample, while a laboratory
manufactured batch of adhesive without the adhesion
promoters included has been applied to substrates with,
or without, alternative substrate surface pre-treatment.
Surfactants are amphililic molecules which, once
aggregated or adsorbed onto a surface, modify its
properties by lowering the surface energy. This leads to
improved surface wettability during application of an
adhesive and can lead to reduced diffusion of chemicals
at the adhesive-to-substrate interface to enhance the
resistance of a bonded joint to chemical attack. Fluorosurfactants are the most common commercially available
surfactants for use on bonded substrates. They provide
excellent wetting, spreading, chemical and thermal
stability properties to many formulations such as
coatings, inks and adhesives. Fluoro-surfactants
drastically reduce the surface tension at very low
concentrations.
3.2. Adhesion promoters and surfactants
Four common, commercially available types of coupling
agent / surfactant have been investigated within this
study, namely silanes, titanates, zirconates and flourosurfactants. Details of the samples tested within this
work are given in table 2.
Coupling agents are made of a short organic chain,
having an organo-functional group in order to bond with
178
strength in accordance to the T-peel method for metal
substrates ASTM D-1876. Peel specimens were prepared
with a bondline thickness of 0.26mm ±0.1mm,
controlled by inserting 1%wt. of glass spheres in the
adhesive during application to the substrate.
4. TESTING AND EVALUATION
The aim of the test programme has been to ascertain the
mechanical performance of bonded joints pre- and postenvironmental conditioning and investigate the
correlation of the surface wettability of the adhesive to
these results. Each of the different adhesion promoters
defined in table 2 has been applied to the substrates to be
joined, prior to adhesive bonding. Samples made this
way are compared to the performance of the adhesive
with no substrate surface preparation and to the
commercial ‘control’ adhesive, which includes adhesion
promoters in the formulation itself. A summary of the
testing and standard methods used can be seen in table 3.
For all tests, five specimens were evaluated for each
sample. Bond strengths were measured by means of an
Instron 3369 electro-mechanical test frame. The ambient
temperature was 23°C for all tests. Adherands used for
the manufacture of lap-shear and peel joints were
aluminium 6061-T6 and aluminium L61-31032,
respectively. All substrate pieces were cleaned using
acetone soaked absorbent paper prior to any surface
treatment or bonding. Once bonded, all test samples have
been cured for 24 hours at ambient temperature followed
by a post-cure of 16 hours at 40°C, in an attempt to
ensure a similar degree of cross-linking of the adhesive.
Table 3. Details of the test programme utilised in the current study
500 hours saltspray, ASTM
B117:2003
T-Peel
ASTM D-1876:2008
Surface wettability
4.2. Surface preparation methodology
N/A
Lap-shear
ISO
4587:2003
A
B
A
A
B
A
B






















No surface
preparatio
n
No-primer




Ensuring the adherand surface has the optimum
concentration or film thickness of each adhesion
promoter or surfactant is critical to achieving a good
adhesive bond. This can be controlled, in part, by the
concentration of the additive in a suitable solvent, prior
to application on the substrate surface [1]. Consequently,
in this study dilute solutions of the adhesion promoter or
surfactant were produced in propan-2-ol at what is
reported in the literature and supplier datasheets as the
optimum concentration for each. Relative to each
category this optimum concentration would be 2%wt. for
silanes [2,3], 1%wt. for titanates and zirconates [4] and
0.1%wt. for the flouro-surfactants. To prepare the
aluminium substrates, each adherand was cleaned in
acetone before being dipped in the required dilute
priming solution and dried vertically for 10 minutes
prior to bonding.
Control
Crestabon
d M1-20




4.3. Contact angle measurement
Silane
Titanate
Zirconate
Surfactant
Designation
Adhesion
promoter /
Surfactant
type
The ability of an adhesive to wet, or penetrate in to the
microstructure of the adherand surface is critical to
achieve good mechanical interlocking as a mechanism of
adhesion. Contact angle measurement is a common
measure of the wettability, however direct measurement
is complicated for many adhesives due to the high
viscosity of the adhesive formulation itself. Since the
adhesive used in this study is high in viscosity and
thixotropic, only the lower viscosity base acrylic resin
and monomer constituents of the adhesive have been
tested. The contact angle of this resin blend has been
measured using a Dataphysics OCA goniometer, for each
of the cleaned and/or primed aluminium substrate
surfaces.
4.1. Mechanical testing
The performance of adhesive joints can vary
dramatically depending on the geometry and load path of
the specimen to be tested. Shear is commonly considered
to be the most mechanically ideal mechanism of joint
loading, hence this work has primarily utilised the
single-sided lap shear specimens in accordance to
method BS EN ISO 4587. For each surface treatment
samples have been tested both before and after 500 hours
of environmental conditioning in a salt-spray chamber,
in accordance to ASTM B-117. Lap-shear specimens
were manufactured in a specific mould to control the
bondline thickness to 0.53mm ± 0.1mm.
Selected adhesion promotion and surfactants from each
chemical type have subsequently been assessed for peel
5. RESULTS
179
outperforms samples with alternative surface substrate
treatments in peel. The neoalkoxy zirconate and flourosurfactant produced similar peel strength to substrates
with no surface treatment and therefore showed no
beneficial effect.
The peak strength measured in lap-shear testing, both
pre- and post- salt-spray conditioning for 500 hours, is
shown in Figure 1. Error bars indicate the standard
deviation from the mean average result.
In summary, many of the adhesion promoters and
surfactants were able to improve upon the lap-shear
strength of the adhesive (when comparing to the ‘noprimer’ results) prior to environmental conditioning,
with the exception of the monoalkoxy titanate and epoxy
functional silane. The commercial ‘control’ formulation
proved to have the highest strength. However, after saltspray conditioning, many samples showed little
improvement in comparison to the non-primed surface,
with only the titanate, coordinate zirconate and flourosurfactants maintaining any improvement in strength.
Failure modes were mixed cohesive/adhesion failure
with the exception of the control adhesive, which
produced full cohesive failure prior to environmental
conditioning.
6. DISCUSSION
Aluminium is the easiest metal to be treated by the
commonly utilised silane solutions, due to the fact that
chemical bonds are easily formed between the
aluminium and a hydrolysed silane. However, this bond
is not hydrolytically stable, meaning the salt-spray
solution can degrade the interfacial bonds of the silane.
This is confirmed by considering the failure mechanism
also changed, with a higher proportion of adhesion
failure for all conditioned samples, progressing form the
exposed edges of the bonded area. The monoalkoxy
titanate, coordinate zirconate and flouro-surfactants
demonstrated improved hydrolytic stability, not only
with improved strength retention after conditioning, but
with a higher proportion of cohesive failure than
equivalent conditioned silane treated samples. The
commercial ‘control’ adhesive demonstrated the highest
joint strengths and that incorporation of adhesion
promoters within an adhesive formulation can
significantly improve adhesion performance. However,
this formulation also proved susceptible to long term
aging in salt-spray conditions and therefore has potential
to be further improved upon for use in environmentally
harsh applications.
The negative correlation between peel testing and surface
contact angle measurement of the adhesive constituents
demonstrates that the contact angle measurement could
prove a simple method to screen and identify suitable
surface treatment methods. The extent to which this
comment is true would require further investigation.
Figure 1. Results of lap-shear bonded joint testing
7. CONCLUSIONS
Bonded joint testing demonstrates that incorporating
adhesion promoters within a commercial acrylic
adhesive formulation is suitable to significantly improve
joint strength and failure mode.
Flouro-surfactants, monoalkoxy titanates and coordinate
zirconates can improve environmental resistance of
acrylic bonded aluminium joints.
Figure 2. Results of peel and contact angle testing
The average propagation strengths of T-peel testing and
surface contact angle from wettability testing are shown
in figure 2. The results demonstrate a strong negative
correlation between these different tests, indicating that
as surface contact angle decreases (meaning the surface
is more easily wet by the base resins of the adhesive), the
peel strength increases. These results demonstrate that
the commercial ‘control’ adhesive significantly
Simple contact angle measurement with the acrylic
adhesive or primary resin constituents, can prove
suitable to screen surface treatments and give an
indication of relative joint performance.
REFERENCES
180
[1] Wake, W.C., “Adhesion and the formulation of
adhesives”, 1976
[2] da Silva, L.F.M., Öchsnere, A., Adams, R.D.,
“Handbook of Adhesion Theory”, 2011
[3] Rothon, R. (2003), “Bridging the gap with coupling
agents”, Plastics, Additives and Compounding, vol.
5, no. 3, pp. 40-45, 2003
[4] Kinloch, A.J., “Adhesion and adhesives: science and
technology”, 1987
181
182
OPTIMIZATION OF HYDROPHOBIC RECOVERY OF PDMS BY
TREATMENT WITH LOW PRESSURE PLASMAS MADE WITH
DIFFERENT MIXTURES OF ARGON AND
OXYGEN PLASMAS
Maribel Butrón-García, José A. Jofre-Reche, José Miguel Martín-Martínez
Adhesion and Adhesives Laboratory, University of Alicante, 03080 Alicante, Spain
[email protected] , [email protected], [email protected]
ABSTRACT
Poly (dimethyl siloxane) (PDMS) is an amorphous, hydrophobic and low surface energy polymer commonly used in
medical devices, wound dressing and drug delivery systems. In some applications, some polarity (i.e. hydrophilicity) in
PDMS is needed. Low-pressure plasma treatments have been shown effective in increasing the surface energy of PDMS
but the surface modifications last for short time due to hydrophobic recovery. In this study, the adhesion and durability
of PDMS treated with argon-oxygen low-pressure plasmas was optimized by means of experimental design tools
considering the water contact angle values obtained 24 hours after treatment as response variable. The low-pressure
plasma treated PDMS was characterized by XPS and ATR-IR spectroscopy, AFM and SEM. Adhesion was measured by
T-peel tests of treated PDMS/medical acrylic tape. The residual pressure, the power, the length of treatment and the
gas composition were the most influencing variables in inhibiting hydrophobic recovery after plasma treatment. The
optimal conditions of low-pressure plasma treatment were low residual gas pressure of 200-300 mbar, mixture of 97%
oxygen+7% argon as plasmogen gas, low power of 20 watts and long length of treatment (150 seconds).
KEY WORDS: Poly (dimethylsiloxane), PDMS, low-pressure plasma, hydrophobic recovery, experimental design.
ageing and hydrophobic recovery after treatment. To the
best of our knowledge, the optimization of the lowpressure plasma treatment conditions by means of
multivariable methodology is novel in the existing
literature and the use of this methodology for inhibiting
hydrophobic recovery too.
1. INTRODUCTION
Due to its particular properties (biocompatibility,
chemical inertness, thermal stability, optical clearness,
flexibility) PDMS (polydimethylsiloxane) is used in
medical applications including wound dressing, drug
delivery systems and prosthesis or implants construction
[1]. However, the low surface energy and high
hydrophobicity of PDMS gives poor adhesion. In some
biomedical applications (cell adhesion, tissue
engineering) [2] adhesion of PDMS is important.
2. EXPERIMENTAL
Materials: Two polydimethylsiloxanes with different
degree of crosslinking (Elastosil E41 and Elastosil
E43N - Wacker Silicones AG, Munich, Germany) were
used. For measuring adhesion, an acrylic adhesive tape
for medical use (3M Co. – St. Paul, MN, USA) was
used.
Although the use of chemical primers has been proved
effective to impart adhesion to PDMS, these chemicals
restrict their use in biomedical applications. In this way,
the use of plasmas has been revealed as clean, effective
and biocompatible surface treatments for enhancing the
surface energy and improve the adhesion of PDMS
[3,4]. However, due to the high amorphous structure of
PDMS, most of the surface modifications produced by
plasma treatment are not stable and fast hydrophobic
recovery is produced [5]. One way to optimize the
surface modification but inhibiting hydrophobic
recovery in PDMS is the optimization of the different
experimental variables during plasma treatment.
Power electrode shelf
Plasma gas
stream
RF power
source
PDMS sample
Floating
shelf
In this study RF low-pressure plasma treatment of
PDMS have been optimized by using multi-variable
experimental design procedure in order to balance the
increase in surface energy and hydrophilicity, and the
improvement in adhesion of PDMS but minimizing
Ground electrode shelf
Vacuum
pump
Figure 1. Low-pressure plasma reactor showing the shelves
configuration. Direct configuration.
183
treated PDMS measured 24 hours after treatment was
used.
Figure 2 shows that for residual gas pressure values
higher than 300 mbar, the wettability of the treated
PDMS drops noticeably. Therefore, the residual gas
pressure was fixed at 300 mbar and a second screening
experimental design was carried out. According to
Figure 3, apart of the residual pressure, the length of the
plasma treatment, the plasma gas composition and the
power determines noticeably the effectiveness of the
low-pressure plasma treatment.
Low-pressure plasma treatment: Surface treatment of
PDMS was carried out with low-pressure RF plasma
equipment BSETEQ NT-1 Supersystem (BSETEQ,
Pittsburg, USA) by using different oxygen-argon
mixtures. Different parameters able of determining the
effectiveness of the low-pressure plasma treatment were
considered, including the gas composition, the power,
the length of treatment, the residual pressure, the
distance between the power and the ground shelf, the
distance between the sample surface and the plasma
source, and the degree of crosslinking of PDMS. The
scheme of the plasma reactor chamber (Figure 1) shows
the power and ground electrodes shelves in direct
configuration, i.e. the sample is placed on the floating
shelf between the ground and power shelf.
Characterization of the treated PDMS surfaces: Changes
in the wettability of PDMS were monitored by water
contact angle measurements at 25 ºC using ILMS 377
goniometer (GBX Instruments, Bourg de Péage,
France); 4 µl drops of deionized and bidistilled water
were placed on the treated PDMS surface and at least
four replicates were measured and averaged. Water
contact angle values were measured just after treatment
and 24 hours later for monitoring the stability of the
surface modifications.
Figure 2. Variation of the water contact angle values (measured 24
hours after plasma treatment) as a function of the residual pressure in
the plasma chamber.
The topography of the treated PDMS was examined by
scanning electron microscopy (SEM) in Jeol JSM-840
microscope (Jeol Ltd., Tokyo, Japan). Samples were
gold coated before analysis, and the energy of the
electron beam was 12 kV. Quantitative analysis of the
nanoroughness on the PDMS surface after plasma
treatment was measured by atomic force microscopy
(AFM) using a NTEGRA Prima microscope (NT MDT,
Moscow, Russia). Samples were measured in semicontact mode and a scanning area of 25x25 µm was
analyzed.
The chemical modifications on the surface of the treated
PDMS were analyzed by XPS (X-ray photoelectron
spectroscopy)
using
K-Alpha
Thermoscientific
spectrometer (Thermo Fischer Scientific, Waltham,
MA, USA) provided with Al-Kα X-Ray source (1486.6
eV) and twin crystal monochromator, operating at 3 mA
and 12 kV; the residual pressure during the experiments
was lower than 2·10-9 torr.
Figure 3. Relative weight of experimental parameters on the
effectiveness of the low-pressure plasma treatment. Residual pressure:
300 mbar.
A Doehlert experimental plan was performed for
studying simultaneously the incidence of the length of
treatment, the power and the gas composition on the
stability and ageing inhibition of the treated PDMS
surface. According to Figure 4, the length of treatment
and the power were not independent in the low-pressure
plasma treatment of PDMS, and the most important
increase in hydrophilicity is obtained for low power (25
W) and long length of treatment (120 seconds). On the
other hand, a composition of 93% oxygen + 7% argon
mixture in the low-pressure plasma treatment shows the
highest hydrophilic stability of the PDMS.
PDMS adhesion was obtained from T-peel tests of
treated PDMS/acrylic adhesive joints in TA-TX2i
Texture Analyser (Stable Micro Systems, Surrey, UK)
by using a pulling rate of 1 mm/s. The dimensions of the
test samples were 20×50 mm.
An experimental screening design was carried out to
find the key experimental parameters determining the
effectiveness of the plasma treatment of PDMS. As
response variable, the water contact angle values on
184
of 102.1eV, 102.8eV and 103.4eV that correspond to
(CH3)2SiO2, (CH3)SiO3 and SiO4 species respectively.
(CH3)SiO3 and SiO4 species are produced by the
replacement of CH3 groups by OH groups during
plasma treatment [6]. Figure 6 shows that the content of
oxidized silicon species is more noticeable when PDMS
is treated at low power and high length of treatment. On
the other hand Figure 7 shows the effect of power for
mid-time treatment showing an increase in the SiO4
species content at 10 and 50W.
20W 150s
Figure 4. Wettability isoresponse surface of treated PDMS under
different power and length of treatment. 50% O2 – 50% Ar lowpressure plasma treated PDMS.
20W 30s
40W 150s
40W 30s
Figure 6. Si2p curve fitting of PDMS treated with 93% O2 + 7% Ar
low-pressure plasma. Influence of the power and length of treatment.
Figure 5. Wettability level curves of low-pressure plasma treated
PDMS as a function of the plasma gas composition and the length of
treatment
Table 1 shows the chemical composition (obtained by
XPS) on the PDMS surfaces treated at different power
and length of treatment. The highest oxygen content on
the PDMS surface is achieved by treatment at low
power and long length of treatment. When the power is
increased, the effectiveness of the plasma treatment for
long time decreases but the stability of the surface
modifications increases.
Time
(s)
30
20
150
40
40
C (at%)
O (at%)
Si (at%)
O/C
44.5
27.8
27.8
0.64
36.9
33.5
29.6
0.91
30
44.8
26.6
28.6
0.59
150
45.2
26.7
28.1
0.59
10W
30W
50W
Figure 7. Si2p curve fitting in PDMS treated with 93% O2 + 7% Ar
low-pressure plasma for 90s. Effect of the power.
Figure 8 shows the AFM micrographs of the PDMS
treated at different power for 90 seconds, and the
creation of different nano-roughnesses on the PDMS
treated at high (50W) and low (10W) power is
observed; furthermore, the ablation suffered at
intermediate (30W) power are evidenced.
Table 1. Chemical surface composition on PDMS treated with 93% O2
+ 7% Ar low-pressure plasma. Residual pressure: 300 mbar. XPS
experiments
Power
(W)
20
Untreated
Hydrophobic recovery of the treated PDMS surfaces
was monitored by water contact angle measurements at
different times elapsed after plasma treatment. The gas
composition of the plasma affects noticeably the extent
of PDMS hydrophobic recovery and the most stable
water contact angle values were obtained by treatment
with 93% O2 + 7% Ar plasma (Table 2).
The effect of the gas composition of the plasma on the
stability of PDMS surface modifications was studied by
XPS, and the highest content of SiO4 species on the
PDMS treated surface for two weeks after treatment was
obtained with 93% O2 + 7% Ar plasma (Figure 9).
Figures 6 and 7 show the curve fitting of Si2p
photopeak of PDMS treated with low-pressure plasma
at different power and length of treatment. Different
silicon species can be distinguished at binding energies
185
Untreated
Ra= 33 nm
10W
Ra= 45 nm
30W
Ra= 17 nm
50W
Ra= 50 nm
the low-pressure plasma treatment of PDMS, and the
optimal conditions to allow the minimal hydrophobic
recovery at 300 mbar of residual pressure were 25W
and 120s. Other important factor was the composition of
the plasma gas stream, being 93% O2 + 7% Ar the
optimal with respect to the surface modifications and
inhibition of hydrophobic recovery of treated PDMS.
Adhesion improvement of PDMS was obtained and it
was maintained for at least 2 weeks after plasma
treatment.
Figure 8. AFM micrographs of 25x25nm of untreated and lowpressure 93% O2 + 7% Ar plasma treated PDMS for 90s at different
power.
Table 2. Water contact angle values (measured 14 days after
treatment) of PDMS treated with low-pressure plasma at 25W,
300mbar and 120s varying the gas composition.
Gas composition
100% Ar
50% O2 – 50% Ar
93% O2 – 7% Ar
100% O2
Water contact angle (degrees)
67 ± 4
66 ± 2
59 ± 2
68 ± 3
Figure 10. Peel strength values of treated PDMS/acrylic tape
obtained at different times after low pressure plasma treatment. 25W,
300mbar, 2min, 93% O2+7% Ar.
ACKNOWLEDGEMENTS
Authors thank 3M, Wacker Silicones GMBH and
Innovaciones DISRAS S.L. for supplying some
materials used in this study.
REFERENCES
[1] Mata A., Fleischman A.J., Roy S., “Characterization
of polydimethylsiloxane (PDMS) properties for
biomedical
micro/nanosystems”.
Biomed
Microdevices”, 7:4, pp. 281-293. 2005.
[2] Siow K.S., Britcher L., Kumar S., Griesser H.J.,
“Plasma methods for the generation of chemically
reactive surfaces for biomolecule immobilization
and cell colonization”. Plasma Process. Polym., 3,
pp.392-418. 2006.
[3] Zhou J., Ellis A.V., Voelcker N.H., “Recent
developments in PDMS surface modification for
microfluidic devices”. Electrophoresis, 31, pp. 2-16.
2010.
[4] Kim H.T., Jeong O.C., “PDMS surface modification
using atmospheric pressure plasma”. Microelectron
Eng., 88, pp. 2281-2285. 2011.
[5] Encinas N., Dillingham R.G., Oakley B.R.,
Abenojar J., Martínez M.A., Pantoja M.,
“Atmospheric
pressure
plasma
hydrophilic
modification of a silicone surface”. J. Adhesion, 88,
pp. 321-336. 2012.
[6] Hegemann D., Brunner H., Oehr C., “Plasma
treatment of polymers for surface and adhesion
improvement”. Nuclear Instruments and Methods in
Physics Research, Part B: Beam Interactions with
Materials & Atoms, 208, pp. 281-286. 2003.
Figure 9. Percentages of silicon species on PDMS treated with lowpressure plasma at 25W for 120s, 14 days after treatment. Si2p
photopeaks.
The effect of the low-pressure plasma treatment on the
adhesion and stability over time of PDMS treated under
the optimal conditions is shown in Figure 10. A huge
improvement (about 1000%) in T-peel adhesion
strength of treated PDMS to acrylic adhesive from 22
N/m to 226 N/m is obtained. Moreover, the adhesion
properties of the PDMS remains almost constant up to
more than two weeks elapsed between the low-pressure
plasma treatment and joints formation.
4. CONCLUSIONS
The residual gas pressure was the main factor
determining the surface modifications of PDMS, the
optimal results are obtained for 200-300mbar. RF power
and length of treatment were interdependent factors in
186

Normas trabajos finales XXIV Encuentro GEF

Transcription

Similar documents

Sikaflex®-292i FOR LONG LASTING BONDS

Brochure

Plasma Thawer - kizlonmedical.com

tecbond hotmelts are STRONGER. FASTER.

UvoNova

PH TOFAB® Ultra

Adhesive Tips - Sally Lynn MacDonald

Misc. User Notes: - Magnepan Users Group

Temporary Tattoos