sum-.0 .

M913? m

“2...

.
‘
.
(
.

. "n.3,“ "m... »q

 

 

TE’EQTS
1.

55393.61 LIBRARY

Michigan State

University

 

 

 

 

This is to certify that the

thesis entitled

Ultraviolet Light Surface Treatment of Polymer
Composites For Enhancement of Adhesion

presented by

Praveen Tummala

has been accepted towards fulfillment
of the requirements for

M. 3. degree in Materials Science
Engineering

Major professor fl

Date 12/13/02

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requeSted.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 CLICIRCv’DatoDUOpBS-DJS

ULTRAVIOLET LIGHT SURFACE TREATMENT OF POLYMER

COMPOSITES FOR ENHANCEMENT OF ADHESION

By

Praveen Tummala

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Chemical Engineering and Materials Science

2002

ABSTRACT

Ultraviolet light Treatment of Polymer Composites
for Enhancement in Surface Adhesion
By
Praveen Tummala

In spite of having desirable bulk mechanical properties suitable for all
engineering applications, polymers and polymer composites are limited for their
use in certain areas it is essential to adhesively bond them. This is due to their
poor surface adhesion. Hence, they require some kind of surface preparation
before they can be bonded. Ultraviolet light treatment has emerged out to be an
efficient technique in order to prepare the polymer surfaces for adhesive bonding.
It is simple to construct, easy to operate, faster and environmentally benign. The
current research involves the use of ultraviolet light from two different sources
viz. a lamp and an excimer laser, to treat the polymer surfaces. The polymer
composites investigated are polyethylene terepthalate/polypropylene — glass fiber
composites. Significant enhancement in surface adhesion was obtained. as a result
of the ultraviolet light treatment. This is due to a combination of cleaning the
surface and functionalizing the surface by the ultraviolet light. To explore the
practicality of the ultraviolet light treatment process, automobile engine valve
covers from Daimler Chrysler have been UV treated, adhesively bonded and
tested. The test results show a tremendous increase in the surface adhesion
properties of the valve covers and the failure modes are completely cohesive.

Thus this technique has all the potential for being applied in an industrial scale for

treating massive and large number of parts that are to be bonded.

ACKNOWLEDGEMENTS

Many individuals provided support and assistance so that this work may
be completed. I greatly appreciate and thank the scientific and technical advise
and support provided by the staff and students at the Composite Materials and
Structures Center. I would like to thank my committee members, Dr.
Krishnamurthy Jayaraman and Dr. Andre Lee for being on my committee.
Finally, I wish to express an enormous amount of appreciation and thanks to Dr.

Lawrence Drzal for inspiring, financing and advising this research work.

iii

Table of Contents

List of Tables
List of Figures
Chapter One
Introduction ........................................................................ 1
Chapter Two
Literature Review ................................................................. 5
Importance of Adhesive Bonding ..................................... 5
Reasons for Poor Adhesion of Polymers ............................. 9
Various Methods to Enhance Surface Adhesion ................... 10
UV Treatment of Surfaces ............................................. 15
Sources of UV light .................................................... 21

Advantages of the UV Process over Traditional Techniques. . .. 23
Variable Parameters Influencing the UV Treatment Process. . ...25

Chapter Three
Experimental Methods
Experimental Materials ................................................. 27
UV Treatment Process ................................................. 31
Laser Treatment ......................................................... 33
Silane Treatment ........................................................ 34
Adhesion Tests .......................................................... 35
X-Ray Photoelectron Spectroscopy .................................. 38
Environmental Scanning Electron Microscopy ..................... 40
Chapter Four
Results & Discussion
Adhesion Tests .......................................................... 41
Surface Chemistry Characterization .................................. 48
Microscopy .............................................................. 51
Chapter Five
UV Treatment of Polymer Composite Valve Covers
Valve Covers ............................................................ 55
UV Treatment of Flat Valve Cover Specimens ..................... 58
Design of a Conveyor System for Treating the Valve Covers. . ..62
Treatment of Valve Covers ............................................ 68
Bonding the Valve Covers ............................................. 69
Testing the Valve Covers .............................................. 71
Results and Discussion ................................................. 74
Chapter Six
Conclusions ....................................................................... 85

iv

List of Tables

Table 4.1: XPS results for UV treated Ticona ............................... 48
Table 4.2: XPS results for laser treated Ticona .............................. 50

Figure 2.1:
Figure 2.2:
Figure 2.3:
Figure 3.1:
Figure 3.2:
Figure 3.3:
Figure 3.4:
Figure 3.5:
Figure 4.1:
Figure 4.2:

Figure 4.3:
Figure 4.4:

Figure 4.5:
Figure 4.6:

Figure 4.7:
Figure 4.8:
Figure 4.9:
Figure 4.10:
Figure 4.11:
Figure 5.1:
Figure 5.2:
Figure 5.3:
Figure 5.4:
Figure 5.5:
Figure 5.6:
Figure 5.7:
Figure 5.8:
Figure 5.9:
Figure 5.10:
Figure 5.11:
Figure 5.12:
Figure 5.13:

Figure 5.14:
Figure 5.15:
Figure 5.16:
Figure 5.17:

Figure 5.18:
Figure 5.19:

List of Figures

Stiffening effect — bonding and riveting ................................... 8
Stress distributions in loaded joints ........................................ 8
Electromagnetic Spectrum ................................................. 15
RC-747 UV Lamp ........................................................... 28
KrF Excimer laser ............................................................ 28
Schematic representation of the UV treatment process ................ 32
Schematic representation of the Excimer laser treatment process. . ..34
Schematic representation of the PATTI system ......................... 37
PATTI results for UV treated Ticona ..................................... 41
Schematic representation of the silane reaction
on the polymer surface ................. 43

PATTI results for UV/silane treated Ticona ............................. 44
PATTI results for UV treated Ticona through silane ................... 45
PATTI results for laser treated Ticona .................................... 46
Maximum values of adhesive strengths for various

treatment conditions. . . . . . . . ....47
ESEM of control Ticona sample .......................................... 51
ESEM of UV treated Ticona ............................................... 52
ESEM if Excimer laser treated Ticona .................................... 52
ESEM of UV and silane treated Ticona .................................. 53
ESEM of Excimer laser and silane treated Ticona ...................... 53
Valve cover location in an engine ......................................... 55
Daimler Chrysler valve cover .............................................. 57
PATTI results for UV treated Ticona (Ashland Pliogrip) .............. 59
PATTI results for UV treated Hivalloy (AshlandPlio grip) ............ 59
Tested PATTI stubs and substrates ....................................... 6O
Conveyor system for UV treating the DC valve covers ................ 62
Worm and worm gear ....................................................... 64
Ozone delivery system ...................................................... 64,
Calibration graph of the conveyor system ................................ 65
PATTI results for UV treated Hivalloy using the conveyor system...67
Setup for the UV treatment of valve cover ............................... 68
Valve covers adhesively bonded and clamped together ................ 7O
Schematic representation of the UTS testing of

valve cover specimen .......... 72
Picture showing the cut section of the valve cover

and the fixtures .................. 73

UV treated and adhesively bonded valve cover test results;
Adhesive used: Ashland Plio grip. . ....74
UV treated and adhesively bonded valve cover repetition

test results; Adhesive used: Ashland Plio grip ......... 75
ESEM of as received (untreated and non-bonded) valve cover ....... 76
ESEM of untreated and bonded valve cover ............................. 76
Cured Ashland Pliogrip adhesive .......................................... 76

vi

Figure 5.20:
Figure 5.21:
Figure 5.22:
Figure 5.23:

Figure 5.24:
Figure 5.25:
Figure 5.26:
Figure 5.27:
Figure 5.28:
Figure 5.29:

Figure 5.30:
Figure 5.31:

UV treated and bonded valve cover ....................................... 76
UV treated and bonded valve cover ....................................... 77
UV treated and bonded valve cover ....................................... 77
UV treated and adhesively bonded valve cover test results;

Adhesive used: Dow LESA ............ 78

UV treated and adhesively bonded valve cover repetition

test results; Adhesive used: Dow LESA. . ......79

UV treated and adhesively bonded valve cover test results;
Adhesive used: ESSEX Betamate. . ....80

ESEM pictures of failure surfaces when ESSEX Betamate
adhesive is used for bonding the valve covers .......... 8]

Strain energy comparisons of valve cover test results using
different adhesives. . ....82

Optical Microscopy - Sample number 10 non-polished ................ 84
Optical Microscopy - Sample number 6 non-polished .................. 84
Optical Microscopy - Sample number 10 polished ..................... 85
Optical Microscopy — Sample number 6 polished ....................... 85

vii

Chapter One

INTRODUCTION

The versatility of polymers and polymer composites is creating a growing
worldwide demand for these materials in a wide variety of applications. Polymers
and polymer composites are very extensively used in today’s industry due to their
superior and excellent balance of material properties. They are attractive for
various engineering and manufacturing applications. They are highly competitive
with the traditional metals and alloys due to easy and excellent processability by
high-volume/high—rate techniques such as extrusion and injection molding and an
inherent lack of corrosion related problems. Their low cost and less weight are
added advantages. An attempt to minimize the number of materials employed in
various applications in order to facilitate recycling is another important reason for
the interest in these materials. In spite of their desirable bulk properties, the use of
many polymers is hindered by their poor surface adhesion properties. Strong and
stable adhesive bonding of polymers and polymer composites plays an important
role in extending their applications in various fields of industry. Adhesive
bonding has also been in practice to overcome the drawbacks of mechanical
fastening. Polymers may have very poor adhesive properties because they are
inert and hydrophobic due to various other reasons. Therefore they require some
kind of surface preparation before they can be bonded adhesively. There are
various techniques available to modify surfaces and thereby enhance adhesion.
Mechanical abrasion, chemical treatment, flame treatment, corona discharge and

plasma treatment are the some of the surface pretreatment techniques that are

being used extensively. Each of the above has its own limitations. This work
focuses on a new surface treatment technique using Ultra-violet light. This
technique overcomes the scientific, technological, environmental and economical

limitations of the traditional treatment processes.

Ultra-violet light is capable of chemically modifying the surfaces by
adding polar groups onto the surface, if the correct wavelengths, intensity and
ambient environment are chosen. This process can not only add the desired
functional groups to the surface, but also remove the surface contaminants thereby
cleaning the surfaces. This is a very versatile process where a supplemental gas or
a solution can also be added to the treatment process depending on the chemical
groups desired on the surface. UV treatment is a highly controlled and
reproducible surface modification technique. The following work primarily
focuses on understanding the treatment process in terms of the process

parameters.

The sources for the Ultra—violet light used in this work are a pulsed UV
lamp and an Excimer Laser. The lamp generates UV light over a range of
wavelengths whereas the Excimer Laser generates UV light at a particular
wavelength in the UV range. Studies are made in comparing the two UV

treatment pI‘OCCSSCS.

Two polymeric composite materials are mainly emphasized in this work.
One of them is a glass fiber reinforced polyethylene terepthalate (GF-PET) and

the other is a glass fiber reinforced polypropylene (GF—PP). These materials were

chosen due to their advantageous commercial potential in the durable goods,
construction and transportation industries. Their applications can be extended

provided their adhesion properties are improved.

Apart from using the UV light for modifying the surfaces, silanes are also
used in the work. It has been shown that chemically grafting coupling agents like
silanes onto a surface not only act as a protective coating but can also
functionalize the surface thereby improving its adhesion properties. Even though
the UV treatment improves the adhesion of the polymers, in many cases it is
required to further enhance the adhesion to the point of cohesive fracture of the
substrate. But untreated polymeric materials are non-receptive to these coupling
agents due to the absence of the required surface functionalities. Therefore a
pretreatment is required for the polymeric materials in order that the coupling
agents can be grafted on top of their surface. This work discusses the

functionalization of polymers using both UV light and silanes.

Adhesion tests are performed on the UV treated polymeric materials to
assess the improvement in surface adhesion. The adhesion test is called a PATTI
(Pneumatic Adhesion Tensile Testing Instrument) test, where an aluminum stub is
bonded to the substrate using a room temperature curing adhesive and the tensile
strength for pulling the stub from the substrate is measured. XPS (X-Ray
Photoelectron Spectroscopy) is used to study the chemical changes introduced by

the treatment on the polymeric surfaces. ESEM (Environmental Scanning

Electron Microscopy) is used to characterize the failure surfaces as well as to find

out if the UV light has abraded the surface.

Chapter two presents a review of the literature. A summary of traditional
methods for modifying surfaces is provided. The UV treatment process is
described and compared with the older techniques. Some results from the
literature are discussed. Chapter three discusses the experimental methods right
from sample preparation to the testing. Chapter four includes the results and a
discussion of the results. Chapter five discusses the UV treatment of a production
composite part — a Daimler Chrysler valve covers along with the bonding, testing
and analysis of the effectiveness of the process for this material. Chapter six

draws conclusions and gives suggestions for future work.

Chapter Two

LITERATURE REVIEW

2.1) Importance of Adhesive Bonding

Adhesives are a critical part of many manufacturing processes. A large
number of adhesive formulations are available to cover a broad array of
applications. Adhesive bonding is a widespread joining method in industrial
applications of polymers and polymer composites. With this technique
satisfactory joints can be produced. This has been in practice to overcome the
drawbacks of mechanical fastening. Adhesive joining, properly done creates
strong stable joints with superior mechanical and durability characteristics to
mechanically fastened structure. It has been discussed in the literature that the
creation of covalent bonds across the substrate/adhesive interface is sufficient for
creating viable adhesion in composites. The use of adhesives not only increases
the joint strength but also distributes the load more evenly, and enables alternating
joining methods to be reduced or eliminated. Dissimilar materials can be joined
together by bonding even where it is impossible to gain access to either side of the

joint, thereby increasing design flexibility.

The adhesive bond is continuous. On loading, there is a more uniform
distribution of stresses over the bonded area. The local concentrations of stresses
present in spot-welded or mechanically fastened joints are avoided. Moreover, its

not possible to weld thermoset polymers. Bonded structures can consequently

offer a longer life under load. The bonded joint being continuous produces a
stiffer structure. Alternatively, if increased stiffness is not needed, the weight of
the structure can be decreased while maintaining the required stiffness. Adhesive
bonding gives a smooth appearance to designs: there are no protruding fasteners
such as screws or rivets, and no spot-weld marks. The bonded structure is a safer
structure because, owing to the fewer and less severe concentrations of stresses,
fatigue cracks are less likely to occur. A fatigue crack in a bonded structure will
propagate more slowly than in a riveted structure or even in a machined profile
because the bond lines act as a crack stopper. Adhesive bonding does not require
high temperatures. It is a suitable means for joining together heat sensitive
materials prone to distortion or to a change in properties from the heat of brazing
or welding. Complex assemblies that cannot be joined together in any other way
are feasible with adhesives. Adhesives can join different materials together,
materials that may differ in composition, moduli, coefficients of expansion, or
thickness. The continuous adhesive bond forms a seal. The joint is consequently
leak-proof and less prone to corrosion. Adhesive joints are impermeable to air and
water. The adhesive bond can provide an electrically insulating barrier between
the surfaces. Bonded joints tend to be damage tolerant due to high damping
behavior of adhesive layer. This capacity may be useful for reducing sound or
vibration. Adhesive bonding can simplify assembly procedures by replacing
several mechanical fasteners with a single bond, or by allowing several
components to be joined in one operation. Adhesive bonding may be used in

combination with spot welding or riveting techniques in order to improve the

performance of the completed structure. All these advantages may be translated
into economic advantages: improved design, easier assembly, lighter weight, and

longer life in service.

Insufficient joint strength and lack of durability have been major obstacles
to the adhesive bonding. However, these limitations of adhesive bonding are
being addressed by the photochemical surface modification method explored in

this research.

 

Figure 2.2

 

Figure 2.1: Stiffening effect-

bonding and riveting

Figure 2.2: Stress distributions in

loaded joints

 

 

The diagram shows how a joint may
be designed to take advantage of the
stiffening effect of bonding.
Adhesives form a continuous bond
between the joint surfaces. Rivets and
spot-welds pin the surfaces together
only at localized points. Bonded
structures are consequently much
stiffer and loading may be increased

(by up to 30-100%) before buckling

OCCUI'S.

 

The riveted joint on the left is highly
stressed in the vicinity of the rivets.
Failure tends to initiate in these areas
of peak stress. A similar distribution
of stress occurs with spot welds and
bolts. The bonded joint on the right is
uniformly stressed. A continuous
weld joint is likewise uniformly
stressed but the metal in the heated

zone will have undergone a change in

strength.

2.2) Reasons for Poor Adhesion of Polymers

Adhesion problems are commonly encountered in the commercial use of
polymers and polymer composites. Adhesion is largely controlled by the
conditions on the surface of the adherend. Surface factors such as wettability,
surface free energy, the functional groups on the surface, surface contaminants
and surface roughness influence the adhesion properties of a material. Processed
polymers and polymer composites can also contain undesirable compounds or
environmental or processing additives that can reduce or limit adhesion. In
addition, the inherent low surface energy of polymers makes it difficult to bond
these materials. Polymer surfaces are highly inert and exhibit good chemical
resistance. The absence of polar chemical functional groups on the surface adds
to the poor adhesion of polymers. They are water repellant and have very poor
wetting properties, which make them hydrophobic. Polymer surfaces can be very
smooth which means that they lack surface roughness which can enhance
mechanical interlocking and adhesion. Due to all the above reasons, polymers
exhibit very poor affinity to adhesives, paints and printing inks. which frequently

limit their applications unless they are surface treated.

2.3) Various Methods to Enhance Surface Adhesion

It is necessary to pre-treat the polymer surfaces prior to their bonding,
printing or painting. There are various methods to improve the adhesion
properties of the polymers. The prominent ones are mechanical abrasion,
chemical treatment. flame treatment, corona discharge, plasma treatment and

ultraviolet radiation treatment.

2.3.1) Mechanical Abrasion: Mechanical abrasion is the process of roughening
the surface of a polymer by mechanical means to improve the adhesion. This
includes grit blasting, wire brushing, sanding, abrasive scrubbing, scraping or
filing. Mechanical abrasion creates valleys and crevices on the adherend. An
adhesive must fill these valleys and crevices of the adherend and displace trapped
air to work well. The mechanical interlocking of the adhesive and the adherend
together creates a strong bond between them, and the overall strength of the bond
is dependent upon the quality of this interlocking interface. Abrading the
adherend enhances the mechanical interlocking and increases the bonding surface
area. Though it improves adhesion, mechanical abrasion is a surface destructive

technique.

2.3.2) Chemical Treatment: Chemical treatment commonly includes acid or
alkaline etching of the adherend surface. Etching the adherend removes stubborn
oxides and roughens the surface on a microscopic scale. The desired functional
group can also be added to the surface by treating it with an acid. For example,

oxidation of surfaces can be done in chromic acid and sulfonation in

10

chlorosulfonic acid. Wet chemical and solvent treatment, if effective, often add
numerous additional processing steps such as neutralization, washing, rinsing and
drying. These solvents and chemicals are usually hazardous or designated

hazardous, constituting a toxic waste disposal problem and cost.

2.3.3) Flame Treatment: Flame treatment is ideal for treating polymer films.
Flame treatment involves exposing the surface of the substrate to a suitable
oxidizing flame for a period of time. This treatment brings about a change to the
polymer surface that makes it wettable and permits a strong adhesive bond
between the surface and the adhesive. During treatment, one side of the polymer
film is exposed to a laminar gas air flame while the back side is cooled by contact
with a water-cooled aluminum roll. The flame burner is usually a stainless steel
ribbon mounted in a cast-iron housing. Dust filtered compressed air is premixed
with natural gas in a venturi mixer. The flame power is quantified and is simply a
function of the volume of natural gas burned per minute. The input energy to the
substrate is a complex function of the flame power, the exposure time and the

burner to flame gap.

No detailed mechanism of the process has yet been developed. However,
the specific flame properties are responsible for the observed changes in the
characteristics of the polymer. The optimization of the flame treatment process is
to be done with control of variables such as fuel flow rate, film treatment velocity,
choice of fuel, non-fuel additives, substrate to flame separation distance and the

burner configuration.

11

2.3.4) Corona Discharge: Corona treatment of a material for better adhesion is
accomplished by exposing the air near the material surface to a high-voltage
electrical discharge, a corona that causes the oxygen molecules in the discharge
area to divide into their atomic form. These oxygen atoms are then available to
bond with the molecules on the surface of the material being treated, thereby
changing the surface molecular structure to one that is extremely receptive to
inks, coatings, and various adhesives. Corona treatment, in effect, chemically
modifies the surface (raising the surface tension), allowing it to grab onto the ink,

coating, or adhesive being applied.

The components of a corona treating system include a power supply, a
high-voltage transformer, and the treater station through which the material to be
treated passes. The station itself typically comprises an electrode, an electrical
insulator or dielectric, and a return path (ground), and it can be configured in a

number of ways to accommodate different materials.

The power supply has a very simple function: to raise the frequency and
voltage of the incoming power to levels sufficient to generate a corona in the
station. Power supply needs to be monitored and controlled because delivering the
proper energy level is important to the characteristics of the resulting corona
discharge and the surface-energy level attained on the surface of the material. In
general, the higher the frequency (kHz) rating of the power supply, the lower the
voltage required to deliver a given power to the corona discharge. A high-

frequency/low-voltage combination is ideal, as a lower-voltage corona is less

12

damaging to the insulators and dielectrics in the station and to the material being

treated.

There are several basic corona treatment system configurations used for treating
materials. They are defined essentially by the location of the dielectric material in
the station: conventional, bare roll, double dielectric, and convertible. The
configuration that is best for a given application depends mainly on the material
being processed. Corona treatments work best on flat surfaces. Control of the

treatment is by time of exposure.

2.3.5) Plasma Treatment: The physical definition of plasma is a partially ionized
gas consisting of electrons, ions and neutral atoms or molecules, with an
essentially equal density of positive and negative charges. These species in
plasma exist at many different levels of excitement. Plasma can exist over an
extremely wide range of temperatures and pressures. The solar corona, a
lightening bolt, a flame and a "neon" sign are all examples of plasma. The plasma
atmosphere consisting of different excited particles depends on the gas used in the
system. By changing the gas used, plasma surface treatment can be used to
modify the chemical and physical structure of a material. Because of its properties

similar to both liquid and gas, plasma is often referred as "fourth state of matter".

To enable the gas to be ionized in a controlled and qualitative manner, the process
is carried out under vacuum conditions. A vacuum vessel is first pumped down
using rotary and roots blowers, sometimes in conjunction with high-vacuum

pumps to a low to medium vacuum pressure in the range of 10‘2 to 10'3 mbar. The

13

gas is then introduced into the vessel by means of mass flow controllers and
valves. Although many gases can be used, commonly selected gases or mixtures
of gases for plasma treatment of polymers include oxygen, argon, nitrous oxide,
tetrafluoromethane, and air. A high-frequency generator, which can be in the
kHz, MHz, or microwave range is then used to ionize the gas into a plasma,
forming an environment that has been referred to as "the fourth state of matter"
(i.e., in the presence of sufficient energy, a solid can be melted to a liquid, a liquid
vaporized into a gas, and a gas ionized into a plasma). The formed reactive
particles, mainly free radicals react in a direct way with the surface without
damaging the bulk properties of the treated part. The surface modification is
limited to the outermost 10 to 1000 A of the substrate. One distinguishing
characteristic of plasma is a visible glow discharge, with colors ranging from
blue-white to dark purple, depending on the type of gas. This visible glow is

caused by the relaxation of excited gas particles that emit UV radiation.

Plasma systems generally consist of five main components: the vacuum
vessel, a pumping group, a gas-introduction and gas-control system, a high-
frequency generator, and a microprocessor-based system controller. Various
optional parts can then be added to adapt a base system to handle particular

applications or substrates.

14

2.4) UV Treatment of Surfaces

Ultra Violet light has been in use for many years to clean organic
contaminants from various surfaces. It has also been proved successful in
modifying surfaces by introducing polar groups on the surface. The traditional use
of UV light for cleaning surfaces combined with the newly discovered surface

modification technique using UV light makes it more advantageous.

Light is the most common form of electromagnetic radiation.
Electromagnetic radiation is characterized by its wavelength and frequency. The
electromagnetic spectrum covers a broad range, from radio waves with
wavelengths of meter or more, down to X-rays with wavelengths of less than a
billionth of a meter. UV light is part of the electromagnetic radiation with
wavelengths ranging from 100nm to 400 nm approximately. The UV band lies
between the X-rays and the visible light in the electromagnetic spectrum. Figure 3

depicts the optical portion of the electromagnetic spectrum.

Ultraviolet
109.400 nm

    

2 3 456789 2 3 456789

100 Wavelength}. 1.000 (nanometers) 10.000

figure 3.1: Elecromagnetic Spectrum

light can be divided into three categories. UV-A which ranges from 315-400nm,
UV-B which ranges from 280-315nm and UV-C which ranges from 100—280nm.
There are various sources that generate the UV light which will be discussed in

the following chapters.

The entire UV range produces photons of varying energies that can be
useful in modifying various surfaces. A variety of interactions take place between
the UV photons and substrate as well as between the UV photons and the
surrounding environment. Traditionally UV light is used to clean surfaces in
presence of an inert gas. This inert environment can be replaced by a very
oxidative environment. The various reactions initiated by the UV light will be

discussed in detail.

UV light has photons of sufficient and varying energies that can break
almost all the bonds on the surface of a polymer thereby creating active sites on
the surface. These active sites are highly receptive to the reactive species in the
environment surrounding the surface. The desired functional groups can thus be
attached onto the polymer surfaces by UV irradiation in the presence of reactive

gases.

There are various reactions that take place on the surface of a substrate in
presence of UV light. Photooxidation, photosubstitution, photografting.

photocrosslinking and photodegradation are the important ones.

16

Photoablation: Photoablation is the ability of UV light to break the bonds on the
surface of the polymer by bombardment with high-energy photons. This affects
the outermost surface molecular layers exposed to the UV light. As a result of
this, the organic contaminants on the polymer surface are oxidized to C02 and
H20. Ablative photodecomposition is the etching of the polymer surface to a
depth of 1000A0 or more. This occurs if UV light of very high intensity, higher
than the ablation threshold energy of a polymer strikes its surface. When photons
of high intensity are delivered onto the surface it results in spontaneous ejection
of fragments into the gas phase. The ablated material carries away a high
percentage of the energy of the photon that is not consumed in the bond breaking.
Ablative photodecomposition takes place mainly when the UV light source is an
excimer laser operating in the UV range. This causes roughness on the surface.

which in turn enhances adhesion.

Photocrosslinking: Photocrosslinking is the formation of chemical links between
the molecular chains of polymers in presence of UV light. When the surface of a
polymer is exposed to UV light in presence of inert gases, crosslinking occurs on
the surface of the polymer. Crosslinking produces a stronger and harder substrate
microsurface. Under certain circumstances. crosslinking through UV treatment
can also render additional wear or chemical resistance to a material.
Photocrosslinking of polymers finds its application mainly in the field of

photolithography.

l7

Photosubstitution: Photosubstitution is the introduction of the desired functional
groups onto the surface of the polymer using reactive gases. For example sulfonic
acid groups (-SO3H) can be introduced onto the surface by using 802 as a reactive

gas. The reaction can be depicted as follows.

R-H-I-SOz +1/202—)R-803H

Similarly chloride groups, amino groups, cyanide groups, nitride groups, boride
groups can be added to the surface in presence of the respective reactive gaseous

atmosphere.

Photografting: Photografting is the introduction of hydrophilic functionality to a
hydrophobic polymer surface by treating with UV light and using dilute aqueous
or solvent based mixture solutions. Organosilanes and other coupling agents like
isocyanates can be grafted onto the polymer surface in presence of UV light.
Untreated polymeric materials are not receptive to these due to the absence of the
appropriate active sites. Another example of photografting is the introduction of
acrylic groups using UV light in presence of a photoinitiator like benzophenone.
Similarly epoxy and amino functionalities can be grafted using appropriate

aqueous chemical solutions.

Photooxidation: Photooxidation is the introduction of oxygen-based functional
groups onto the surface of a polymer using UV light. This reaction takes place
when the surface is treated with UV light in presence of either air or oxygen or

ozone. UV light creates a very high oxidative environment around the substrate in

18

presence of any one of these gases. Ozone formation and ozone decomposition
are the important reactions in this process. For this reactions to occur UV light

having wavelengths of 184.9 nm and 253.7 nm are essential.

Molecular oxygen absorbs the 184.9 nm light to form an excited oxygen

molecule.

02 + hv (184.9 nm) —> 02*

The excited state can then dissociate to form two ground-state oxygen atoms.

02* —+ 20

The ground state oxygen atom then reacts with molecular oxygen to form ozone.

04-02-403

The quantum yield for ozone formation is approximately 2, which means that for

every photon absorbed. two ozone molecules are formed.

Ozone has a strong absorption at 253.7 nm and two weak ones at 305 and 440 nm.

Ozone absorption at the 253.7 nm is responsible for its decomposition.

03 + I“ (253.7 run) --> O + 02

The quantum yield for ozone photolysis at 253.7 nm is approximately 0.9, which
means that for each ozone molecule destroyed, there is one oxygen atom formed.

These are the main pathways of ozone formation and decomposition. The overall

19

quantum yield of ozone taking into account both formation and decomposition
reactions are 0.5, meaning that it takes two photons of light to generate one ozone

molecule.

UV light not only interacts with these gases but also oxidizes the surface
organic contaminants and creates receptive sites on the surface by breaking the
molecular bonds on the surface. The oxygen related species like ozone and
nascent oxygen generated by the UV light interact with the surface. The combined
effect of UV light and ozone thus functionalizes the surface with oxygen related
groups like OH, COOH, CHO and CO. Photooxidation using UV light is a very
effective and economical technique for modifying polymer surfaces in order to

improve their wettability and increase adhesion.

Although Ultra Violet light by itself, has been in use for many years to
clean organic contaminants from various surfaces, usually the time of exposure
has been long, several tens of minutes. The traditional use of UV light for
cleaning surfaces combined with the newly discovered surface modification

technique using UV light makes it more advantageous.

20

2.4) Sources of UV Light:

The sources for UV light can be divided into three categories. They are 1) UV

lamps generating UV light in the entire UV range. 2) Monochromatic UV lamps

and 3) Lasers operating in the UV range.

1)

2)

3)

UV Lamps — UV lamps generate light in the entire UV range. These lamps
operate over a range of wavelength from approximately 100 nm to about
1000 nm. They have varying light intensities and energies. The operating
medium can be different in different lamps. Mercury vapor lamps and
Xenon lamps are the most commonly used ones.

Monochromatic UV Lamps — These lamps generate incoherent UV
radiation that has a unique wavelength. Depending on the type of gas
(operating medium) used, different lamps generate monochromatic UV
light at different wavelengths. The operating medium is usually an
excimer (excited dimmer) in these lamps. For example lamps operating in
pure xenon (Xez) generate light at 172 nm, in a gas mixture of krypton/
chlorine (KrCl) generate light at 222 nm and in a gas mixture of xenon/
chlorine (XeCl) generate light at 308nm.

Lasers — The lasers that operate in the UV range are the excimer lasers.
Similar to the excimer UV lamps, the excimer lasers also operate at
different wavelengths depending on the gas used in the laser. The light
from the lasers is of higher intensity and smaller cross section when

compared to that from lamps. Examples are Kr; laser operating at 146 nm,

21

Xe; laser operating at 172 nm, ArF laser operating at 193 nm, KrF laser
operating at 248 nm, XeF laser operating at 351 nm, KrCl laser operating

at 222 nm, XeCl laser operating at 308 nm etc.

22

2.6) Advantages of the UV Process over the Traditional

Techniques:

UV surface treatment process can have numerous advantages when

compared to other surface modification techniques. Some of them are listed

below.

UV treatment is a very controllable and versatile process.

It is capable of treating all polymer surfaces even surfaces that are temperature

or vacuum sensitive.

UV can treat three-dimensional substrates and it is adaptable to treat flat or
complex geometries.

The times of treatment are relatively shorter and are not labor intensive.

The process produces no emission of volatile organic compounds (VOC’s).
Hence it is environmentally benign.

It is simple and easy to construct and inexpensive to Operate.

This operation does not require evacuation of the surrounding environment

during the treatment.

Because of the simplicity of the process, it is inherently low in cost to

construct as well as to operate.

It is a non-destructive technique.
It is a dry process.

Specific functionalities can be produced on the surface.

23

The UV treatment process is highly reproducible.

Unlike other treatment processes, the UV process can be operated either as a

batch or as a continuous process.

Optimization of the surface properties can be achieved without alteration of
the bulk properties.

Surface cleaning and surface modification takes place in a single step.

24

2.7) Variable Parameters Influencing the UV Treatment

Process:

There are various factors that influence the UV treatment process. The

process needs to be optimized in terms of these factors.

Type of Lamp: The configuration of the UV lamp used plays a key role in the
treatment process. The intensity of the lamp (power of the lamp), size of the lamp,
and the pulsing frequency of the lamp are the influential factors in the treatment
process. The type of the laser used also affects the treatment because different

lasers operate at different wavelengths.

Lamp to Sample Distance: Because of the use of elliptical reflectors in most
lamps, the light is focused at a particular distance from the lamp and it diverges
after that. The intensity of the lamp decreases as we go far away from the lamp.
Hence the lamp to sample distance needs to be optimized depending on the

material that is being treated.

Gas Concentration and flow rate: The concentration of the supplemental gas
supplied and its flow rate are variables that need to be optimized for treating a

particular surface.

Time of Treatment: The time of UV treatment differs from material to material.
So, depending on the material of the substrate. the treatment time has to be

optimized.

25

Temperature of the Substrate: The temperature to which a substrate reaches
during a treatment needs attention because some polymers are temperature
sensitive. Also temperatures higher than the glass transition (Tg) in polymers may
have adverse effects on their properties. Treating them for longer times or with

high intensity lamps can degrade the material or affect the bulk properties.

Chemical Composition of the Substrate: The physical and chemical characteristics
of the substrate play a key role in the UV treatment process. The chemical
composition of a substrate determines the absorption of the light and its reactions
with the surrounding environment. The type of the UV process parameters to be

used for a treatment depends entirely on the surface chemistry of the substrate.

Laser fluence and pulse repetition: The energy of the laser beam is a variable
quantity when the laser treatment is considered. Also the area of the rectangular
laser beam is variable. Laser fluence is obtained by dividing the laser energy with
the area of the beam. Both these variables are to be adjusted to attain the desired
laser fluence. Since the beam is of high energy, care has to be taken while treating
polymers because high-energy laser can ablate the surface away. The laser fluence
was can be varied in order to optimize the polymer for the maximum adhesive
performance without damaging it. The laser pulses also can be varied to achieve

maximum adhesion.

26

Chapter Three

Experimental Methods

3.1) Experimental Materials

Two UV lamps and an excimer laser have been used in order to conduct
the UV treatment experiments. The lamps are designated as RC-500 and RC-747.
Xenon is the manufacturer of the UV lamps. Both the lamps are pulsing lamps.
RC-500 has a total power output of 300 w and pulsing frequency of 120 Hz. Its
dimensions are 5 in x 1 in. RC-747 can operate in two modes. Both the modes
have the same total power output of 2000 w have the same lamp size of 16 in x
3.5 in. The only difference between the two modules is their pulsing frequency.
One of them pulses with a frequency of 120 Hz and the other with a frequency of
3 Hz. A KrF excimer laser, COMPex 301 manufactured by the Lambda Physik®.
has been used for the laser UV treatment. The excimer laser has KrF as the lasing
medium in it. It has a rectangular pulsed beam. It operates at a wavelength of 248
nm. The average maximum power of the laser is 15 w and the maximum pulse
energy is 1500 ml. It has pulse duration of 25 ns and the pulsing frequency is
variable between 1 and 10 Hz.

Figure 3.1 is the picture of the RC-747 UV lamp and figure 3.2 is the

picture of the KrF Excimer Laser.

27

 

747 UV Lamp

RC-

Figure 3.1

. ls.
.. l

:3:

as
as

:5:

am“
an.

4.2
m. m

 

: KrF Excimer Laser

.2

Figure 3

28

The materials that are treated with UV light for increased adhesive
performance are Ticona Impet EKX-215 and Hivalloy X83-21-5. Ticona is a 20%
glass fiber reinforced Polyethylene Terepthalate (GF-PET). This is a
thermoplastic composite material and has a T8 of 89°C. Hivalloy is a blend of
Polypropylene and Polystyrene with 35% glass fibers reinforced in it (GF-PP).
This is also a thermoplastic composite with a T8 of 118°C. Both these materials
have got a wide array of applications, mainly in the automotive industry.

Four different adhesives are used for bonding the UV treated samples and
testing them for enhanced adhesive strength. They are the Araldite 2015, Ashland
Pliogrip 7779, Dow AH5643A (LESA) and Essex Betamate 73005. The
formulations and the curing times of the adhesives are discussed in the results
chapter.

The UV lamp treatment chamber is an aluminum chamber whose height is
2 inches. It has an inlet and outlet for gas flow. An ozone generator is used in
order to supply supplemental ozone during the treatment process. The inlet gas
flow rate to the ozone generator and the voltage can be varied. A mass flow
controller is used to maintain constant inlet gas flow rate to the ozone generator.
The ozone coming out of the ozone generator has the same flow rate as the inlet
gas. The concentration of the ozone in the outlet stream of gas is dependent on the
inlet gas flow rate and the voltage applied. The lower the gas flow rate, the higher
the concentration of ozone. The higher the voltage applied, the higher the
concentration of ozone. The inlet gas to the ozone generator for all the

experiments is medical grade oxygen that is 99% pure.

29

The materials are cut to l in x 1 in flat samples using a band saw. The

surface of the samples is wiped with isopropanol before treatment.

30

3.2) UV Treatment Process

The blower that supplies cooling air around the UV lamp is switched on.
The power supply to the UV lamp is switched on next. The valve of the oxygen
gas cylinder is opened and the flow rate is set at 10 scflr. The oxygen from the
cylinder is fed to the gas inlet of the ozone generator using a Teflon tube. The
ozone generator is switched on and the voltage is set at the desired value. A
Teflon tube connects the outlet of the ozone generator and the inlet of the
aluminum chamber. The gas flows through the chamber, over the sample and
exits through the outlet. The cut and solvent cleaned sample is placed in the
chamber. The time of treatment and mode of treatment (continuous or
intermittent) is set on the dial of the power supply and the lamp is switched on.
Once the treatment is complete, the sample is taken out of the chamber and its
surface temperature is measured using an infrared pyrometer. The rest of the
process parameters are kept constant and only the UV treatment time is varied for
a particular set of experiments. The figure below shows a schematic

representation of the UV treatment procedure.

31

838m Eel—«8H >5 3:: gouging hem £35320. "MM flank

c3830
“Sansone

  

Seaman—m

. s... a u q
....q .c.......‘s..s..... . ..s
u . .u.uvu.......... .o.v>.......r.......«. nun-n. .mg....

.n....... .....:.....-...
.u.‘¢...r .....c... . .
..uo..... .....>..n-
. .r .-
...n

 

 

32

3.3) Laser Treatment

Laser treatment is carried out in air. The sample is mounted on the sample
holder and the laser parameters are set up. The laser parameters can be controlled
using a joystick. The power supply to the laser and the guiding laser (He-Ne
laser) are switched on. The lens covers are removed. The pulse repetition rate is
set to the desired value using the joystick. The laser energy is also set to the
desired value on the joystick. Moving the sample holder up and down can change
the area of the rectangular laser beam hitting the sample. The laser energy divided
by the area of the beam gives the value of laser fluence. All the remaining
variables are kept constant and the number of pulses hitting the substrate is varied
for a particular set of experiments. Figure 7 shows a schematic representation of

the laser treatment process.

4 T! 4
.W

 

 

 

 

   

 

 

+ Planar

Convex
Lens

 

Substrate

Figure 3.4: Schematic Representation of the Excimer Laser Treatment
Process

33

3.4) Silane Treatment

Both the UV treated and laser treated samples were also treated with silane
for further improvement in adhesive performance. The silane used is a Dow
Corning Z—6020 silane. It is a N- (B-aminoethyl)- y —
aminopropyltrirnethoxysilane. The chemical formula of the silane is
(CH30)3SiCHzCH2CH2NHCH2CH2CH2NH2. The silane is hydrolyzed prior to its
use for treatment. The Z-6020 silane is diluted by mixing it with deionized water
in the volume ratio 1:3 (1 part silane + 3 parts DI. water). The silane is left to
hydrolyze in water over night. Then the solution is made to 1% by weight of
silane by mixing it with isopropanol and left for 24 hours in a closed container
before it is used.

The hydrolyzed silane solution can be applied to the samples in two ways.
One way is to dip the samples in the silane solution for about five minutes after
they have been treated with the UV. The other way is to treat the samples with
UV with a thin layer of silane solution on top of their surface. Silane is applied to
the samples in both the ways mentioned above and the results are compared. Once
the samples are removed from the silane solution, they are washed in deinonized
water in order to make sure there is no physical adsorption of silane onto the

surface. The samples are then dried and tested or analyzed.

34

3.5) Adhesion Tests

The adhesion tests are carried using a Pneumatic Adhesion Tensile Testing
Instrument. It will be referred to as the PATTI tester. The UV treated samples are
adhered to a sandblasted aluminum pull-stub using an adhesive. These samples
are called the PATTI samples. The adhesive is allowed to cure and the adhesion
between the stub and the treated sample is tested using the PATTI system.

The sandblasted aluminum stubs are cleaned using isopropanol to remove
any contaminants. The adhesive is mixed as recommended by the manufacturer.
The adhesive is then applied to the bottom of the pull-stub and pressed against the
test surface. While holding stub in place, a cut-off ring is pressed around the stub
onto the test surface. The cut-off ring displaces the excess adhesive away from the
stub. A light uniform clamping pressure in the form of a weight is applied on the
stub during the curing time of the adhesive. The clamping weights and the cut-off
rings are removed after the curing time of the adhesive. The excess adhesive
surrounding the stub is scored using a knife. The PATTI sample is now ready to
be tested.

The PATTI system applies a true axial tensile strength (relative to stub
axis) pull test. The tensile values obtained quantitatively measure the bond
strength between the adhesive and the substrate surface. The pneumatic control
module of the PATTI system is connected to a pressurized gas supply (nitrogen or
air) and a hose to the pulling piston. The piston is attached to the pull-stub

previously glued to the test surface. Upon detachment of the pull-stub, the control

35

module will register and indicate the maximum pressure (burst pressure) attained
in psig. This pressure is then converted into psi by using the below formula.

POTS = (BP x Ag) — C/Aps

Where:

POTS = Pull-off tensile strength

BP = Burst pressure (psig)

Ag = Contact area of gasket with reaction plate = 2 sq. in. for F-2 piston

and 8 sq. in. for F-8 piston.

C = Piston contact = 0.3 pounds for F-2 piston and 0.8 pounds for F-8

piston.

Aps = Area of pull-stub = 0.196 sq. in.

The PATTI system has two different size pistons. The pistons are called
F-2 and F—8. The type of piston used for testing depends on the bond strength of
the PATTI sample, which in turn depends on the substrate being tested and the
adhesive being used. F-2 piston is used for lower bond strengths and F-8 is used if
the bond strengths are higher. The gas supply to the PATTI system is nitrogen
from a cylinder. The gas regulator is turned on and the digital reading on the
control module is zeroed before testing. The reaction plate is threaded onto the
pull-stub until light contact is made with the piston. Then the pressure is applied
by pressing the RUN button until piston assembly with attached pull-stub
detaches from the test surface. The burst pressure reading on control module, the
type of piston used and the mode of failure are noted. Figure 8 is a schematic

representation of the PATTI piston assembly.

36

 

stub

Adhesive

 

Figure 3.5: Schematic Representation of the PATTI System

37

3.6) X-Ray Photoelectron Spectroscopy

X-Ray Photoelectron spectroscopy is used to determine the chemical
composition of the sample surfaces.

X—ray photoelectron spectroscopy (XPS, also called electron spectroscopy
for chemical analysis, ESCA) is an electron spectroscopic method that uses X-
rays to eject electrons from inner-shell orbitals. The kinetic energy, Ek, of these
photoelectrons is determined by the energy of the X-ray radiation, hv, and the

electron binding energy, Eb, as given by:

E1, = h V - Eb
The experimentally measured energies of the photoelectrons are given by:
Ek = hv - Eb - Ew
Where Ew is the work function of the spectrometer.
The electron binding energies are dependent on the chemical environment of the

atom, making XPS useful to identify the oxidation state and ligands of an atom.

XPS instruments consist of an X-ray source, an energy analyzer for the
photoelectrons, and an electron detector. The analysis and detection of
photoelectrons requires that the sample be placed in a high-vacuum chamber.
Since the photoelectron energy depends on X-ray energy, the excitation source
must be monochromatic. An electrostatic analyzer analyzes the energy of the
photoelectrons, and an electron multiplier tube or a multichannel detector such

as a microchannel plate detects the photoelectrons.

38

XPS surface analysis is performed using a Perkin-Elmer Physical
Electronics PH15400 ESCA Spectrometer equipped with a standard magnesium
X-ray source operated at 300 W (15 kV and 20 mA). Data is collected in the
Fixed Analyzer Transmission mode utilizing a position sensitive detector and
hemispherical analyzer. The elemental composition of the surface is determined
from survey spectra collected using a pass energy of 89.45 eV. High-resolution
spectra of the elements are obtained using a pass energy of 17.9 eV and a step size
of 0.1 eV. Binding energies are referenced to adventitious carbon (C ls = 284.6

eV) and were measured with a precision of i 0.1 eV.

Chemical information indicating changes in the surface treated polymers
is elucidated by curve fitting the carbon ls (Cls) spectra. Curve fitting defines
and interprets the carbon chemistry as detected at the sample surface by allowing
the user to distinguish overlapping features within the C13 spectral envelope. The
Cls spectra are fit with a Lorentzian-Gaussian mix Voigt profile function using a
nonlinear least squares curve-fitting program. The model used is based on the
assumption that there is an approximate 1.5 eV chemical shift per bond to oxygen.

The resulting curve fits have levels of experimental error around 5-10%.

39

3.7) Environmental Scanning Electron Microscopy

Environmental Scanning Electron Microscopy is used to look at the
surface of the samples to see whether the UV treatment has caused any
topographical changes on the surface. The ESEM retains all the performance
advantages of a conventional scanning electron microscopy, but replaces the high
vacuum sample environment with one that contains ~2-3 torr of water vapor. It
allows the examination of any specimen, wet or dry, insulating or conducting. The
depth of field, resolution and microanalysis are superior to that of a light
microscope. This technique does not require the samples to be vacuum tolerant
and electrically conductive like in the SEM.

The ESEM used for this work is manufactured by Electroscan Corporation
(model number — 2020). It is equipped with a Lanthium Hexaboride filament.
Water vapor is used as the imaging gas. The samples to be examined are placed
on the sample holder located in the sample chamber. The imaging pressure
(chamber pressure) is set between 2 and 3 Torr. The working distance between the
detector and the sample is set between 8 and 10 mm. The accelerating voltage is
set at 20 kV. The sample is focused at different points on its area and micrograph

pictures are taken.

40

Chapter Four

Results and Discussions

This chapter discusses the adhesion test results, surface chemistry and
surface topography changes for the RC-500 UV lamp and the Excimer Laser
treated Ticona at different conditions.

4.1) Adhesion Tests:

The RC-500 UV treated Ticona materials are tested using the PATTI
tester for enhanced adhesive performance. The samples are exposed continuously
to the UV light for the time of treatment. Ciba-Geigy Araldite 2015 adhesive,
which is a two part epoxy based is used for making the PATTI samples. This

adhesive has a curing time of 24 hours.

 

 

 

 

 

 

 

 

 

 

 

Lhtreaied

Time (secs)

No supplemental Ozone . Supplemental Ozone supplied

Figure 4.1: PATTI Results for UV Treated Ticona

41

 

The PATTI test results for the UV treated Ticona without using any
supplemental ozone and using supplemental ozone are provided in Figure 4.1.

For the treatment in air maximum adhesive strength is obtained when the
samples were treated for 180 seconds and for the treatment in ozone maximum
adhesive strength is obtained when the samples were treated for 90 seconds. All
the failures in both the treatment conditions (without ozone and with ozone) are
adhesive.

In order to further improve the adhesive performance of Ticona, the UV
treated samples are post-treated with a silane. The silanes are grafted to the
surface of the Ticona by immersing the samples in hydrolyzed silane solution for
5 minutes. They are then dried and bonded to the PATTI stubs and tested after the
adhesive is cured. Dow Corning Z-6020 Silane is used. The chemical name of the
Z-6020 silane is N- (B-aminoethyl)- y —aminopropyltrimethoxysilane. The silanes
have methoxy groups that hydrolyze in water or a solvent. Once they hydrolyze,
the methoxy groups are converted to hydroxyl groups, which are very reactive. Z-
6020 silane has 3 methoxy groups, which on hydrolysis get converted to hydroxyl
groups. These hydroxyls can undergo a condensation reaction, with hydroxyl
groups on both the polymer substrate and the adhesive, thereby enabling the
substrate to be receptive to the adhesive. Thus, the silanes act as connector
molecules between the substrate and the adhesive. The end group of the silane can
be selected depending on the adhesive used for bonding. The hydrolysis of the Z-

6020 silane can be shown as follows.

(CH3O)3SICHQCH2CH2NHCH2CH2NH2 WHOhSiCHgCHzCHzNHCflgCflzNHz

42

Figure 4.2 is a schematic of the grafting of the Z-6020 Silane onto
the surface of UV treated polymer. ‘R’ in figure 4.2 represents the end group of

the silane, which is ‘CHgCHgCHzNHCH2CH2NH2’ in this case.

OH 0 OH
\ gi/ \ SIi/R
IO 6

I R \. o\
/O\ Si\R 751/

\ o

\S/ R

R

O/

\r .4

 

 

 

 

 

 

V
UV treated
Polymer

Figure 4.2 Schematic Representation of the Silane Reaction on
the Polymer Surface

43

 

 

 

 

 

 

Tlme(secs)

No supplemental Ozone . Supplemental Ozone supplied

Figure 4.3: PATTI Results for UV/Silane Treated Ticona

The PATTI results for RC-500 UV treated and then silane immersed
Ticona are shown in Figure 4.3. The surface grafted samples are tested for pre-
UV treatments both with and without supplemental ozone.

The tensile strength required to detach the PATTI stub increased
considerably for the UV treated silane grafted Ticona surfaces. The untreated
samples showed low adhesive performance because silane could not be grafted
onto them due to the lack of receptive sites on the surface. The failure mode for
the UV treated samples is a cohesive failure. By using the silane treatment
technique, the failure mode for the UV treated Ticona samples changed from

adhesive to cohesive.

Another set of experiments were conducted in which the Ticona samples
are first immersed in silane solution and then treated by exposure to the UV light
through the silane solution. The PATTI results for this treatment procedure are

shown in Figure 4.4.

 

1300
900
830
700
600
500
400
330
230
00
0
Lhireated threated & so 60 90 20 30 240
Silane
immersed for
nins
Tlme (secs)

Figure 4.4: PATTI Results for UV Treated Ticona through Silane

Longer UV treatment times are required in this case in order to reach the
same values of tensile strength as in the earlier silane treatment procedure. The
failures are adhesive upto the 120 seconds treatment condition and the failure
mode changed to cohesive for the 180 and 240 seconds treatment conditions.

The Ticona samples were also treated with the Excimer Laser for
enhanced adhesive performance. The samples are treated using the 248 nm

wavelength KrF Excimer Laser with increasing pulse rates keeping the energy

45

fluence constant at 100 mJ/sq.cm. The treated samples were tested using the
PATTI tester. Some of the Laser treated samples are post-treated with silane
following the same procedure like the UV-silane treatment. Figure 4.5 compares
the PATTI test results between the laser treated Ticona and the laser treated and

silane immersed Ticona samples.

 

 

 

 

 

 

 

 

No. of pulses

Laser treated . Laser treated and silane immersed

Ih'gure 4.5 PATTI Results for Laser Treated Ticona

Maximum adhesive strength is obtained when the Ticona samples are
treated with just for 1 pass of the Laser. The adhesive strength decreases
thereafter as the pulse rate increases. When silane is used along with the laser
treatment maximum adhesive strength is obtained at a pulse rate of 2 and

decreases thereafter.

46

 

2 pulses

 

 

 
       

180 sees 90 secs

 

 

r:
(n
a.
V
.:
a
U)
c
g 600 -
m
g 500 -
g 400 —
I; 300 .
o 200 ~ 7
3 .
a 100
0 2
untreated laser treated laser treated uv treated in uv treated in uvtreated uv treated
without silane and silane a'r without air and silane with with
immersed silane immersed supplemental supplemental
ozone ozone and
silane
immersed

[figure 4.6: Maximum values of Adhesive Strengths for various
Treatment Conditions

The PATTI test results using different experimental procedures are
summarized in Figure 4.6 which displays the maximum adhesive strength

obtained using various treatment techniques.

47

4.2) Surface Chemistry Characterization:

X-Ray Photo Electron Spectroscopy is used to examine the surfaces of the
treated samples and analyze the changes in surface chemistry. The carbon,
oxygen, nitrogen and silicon atomic percentages on the surface are compared.
Table 4.1 gives the XPS results for UV and silane treated Ticona at different
treatment conditions. Ticona samples treated for the optimum time that gave the

maximum adhesive strength for a particular set of treatment conditions was

 

 

 

 

 

investigated.
Table 4.]: XPS Results for UV Treated Ticona
Control 60 secs UV 240 secs UV
(%atomic treated and treated
conc) silane through
immersed silane
(% atomic (% atomic
cone) conc)
c 71.2 61.8 73.4
o 18.7 — 21.7 16.5
Si 10.2 12.6 4.9
0.0 3.8 5 .1
N

 

 

 

 

 

 

48

By comparing the atomic concentrations of the the control sample with the
UV/silane treated ones, it is obvious that silane is grafted onto the surface of
Ticona. Also the increase in oxygen atomic concentration in the case of UV
treated first and then silane treated Ticona suggests that the surface has been
oxidized by the UV/Ozone treatment. The decrease in oxygen concentration in the
case of UV treated Ticona through the silane must have been the result of the
increased concentration of the grafted silane on the surface. In order to better
understand the results, nitrogen atomic concentration can be compared. There is
no nitrogen present on the surface of the control sample. Whereas, the UV/silane
treated samples have nitrogen present on their surface. This nitrogen on the
surface results from the silane that is grafted on the surface. The amino groups in
the end group of the silane are responsible for the Nitrogen atomic concentration
on the UV/silane treated samples.

Table 4.2 gives the XPS results for the Excimer Laser treated Ticona at

various conditions.

49

Table 4.2: XPS Results for Laser Treated Ticona

 

 

 

 

 

Control 1 pulse 2pulses 1 pulse 2 pulses
(% at (% at (% at and silane and silane
cone) conc) cone) immersed( immersed(
% at % at
conc) conc)
C 71.2 90.7 89.7 86.9 85.5
0 18.7 8.8 9.8 11.1 11.7
Si 10.2 0.5 0.4 1.2 1.6
N O 0 0.0 O O O 8 1 2

 

 

 

 

 

 

 

 

It can be observed that there is a decrease in Oxygen content as well as the
Silicon content on the surface of the Laser treated samples. This could be the
result of the Laser degrading away the surface and the evolving products from the
Laser degradation settling down back on the surface. This degradation can be
responsible for creating surface roughness, which in turn improves adhesive

performance.

50

4.3) Microscopy:
Environmental Scanning Electron Microscopy is used to look at the
surfaces of the UV and Laser treated Ticona samples. The following figures are

the ESEM pictures of the Ticona samples, untreated as well as treated.

 

Figure 4.7: ESEM of Control Ticona Sample

51

 

Figure 4.9: ESEM of Excimer Laser Treated
Ticona

52

 

Figure 4.10. ESEM of UV and Silane Treated
Ticona

 

Figure 4.11: ESEM of Excimer Laser and Silane
Treated Ticona

53

The ESEM pictures show that neither the UV treatment nor the UV/silane
treatment changed the topography of the surface at the level detected by the
scanning electrion microscope. This means, that the UV treatment did not alter the
surface of Ticona i.e. the UV treatment did not cause any damage to the surface of
the material. The Laser treated samples however, have cracks on their surface
located mainly around the fibers. This implies that the adhesion improvement in
case of the Laser treatment could be due to the mechanical interlocking coming
about as a result of the adhesive penetrating into the surface cracks created on the
surface.

From the XPS and ESEM results, it can be inferred that the improvement
in surface adhesion in case of the UV treatment is due the surface oxidation
whereas the improvement in surface adhesion in case of the Laser treatment is

mainly due to mechanical interlocking.

54

Chapter Five

UV Treatment of Polymer Composite Valve Covers

5.1) Valve Covers

The valve cover in an engine covers the valve train. The valve train
consists of rocker arms, valve springs, push rods, lifters and cam. Oil is pumped
up through the pushrods and dispersed underneath the valve cover, which keeps
the rocker arms lubricated. Holes are located in various places in the engine head
so that the oil re-circulates back down to the oil pan. For this reason, the valve

cover must be oil-tight. The valve covers are usually made out of metals or alloys

 

Engine Block

 

Figure 5.1: Valve Cover location in an Engine

55

like aluminum, steel or iron. Since there is always lubricating oil circulating in the
interior of the valve covers, the valve covers have to be oil-tight and for this
purpose gaskets are used. These gaskets are sealants which prevent oil leakage.
Figure 5.1 is a picture showing the location of the engine valve covers and their

Sti’UCtUI‘C.

Reducing the overall weight of an automobile is much sought after way to
increase its fuel efficiency. Hence, plastics and composites are finding increasing
importance in the manufacture of automotive structural components. Fiber
reinforced plastic composites are desirable in this respect as they have structural
mechanical properties comparable to that of metals and are of low weight. The
automotive structural systems made up of these fiber reinforced plastics satisfy
the needs of safety, strength and durability in addition to being affordable,
manufacturable with desired fit and finish and recyclable. The composite
components made from fiber-reinforced plastics are capable of carrying structural
loads under static and dynamic conditions. A recent development in the use of
polymer composites is the Daimler Chrysler program to replace the metallic valve
covers of an engine with plastic and composite valve covers. These valve covers
are made out of Hivalloy X83-21-5 composite. Hivalloy X83-21-5 is a 35% glass
fiber reinforced polypropylene/polystyrene composite. Along with the use of
composite valve covers adhesive bonding is being evaluated as an efficient
technique to join these valve covers to the engine block. Adhesive joining of the
valve covers eliminates the use of gaskets, distributes the load more uniformly

and eliminates the need for alternative joining methods.

56

The valve covers made from fiber reinforced plastic are selected for their
structural and environmental performance (thermal stability, oil exposure stability
and mechanical properties) but have a surface with qualities or contaminants that
make it difficult to adhere paints of adhesives. Hence, if adhesive joining is to be
used, the bonding surfaces of these valve covers require surface preparation or
pre-treatment before they are bonded. The MSU UV surface treatment method has
been used to prepare these surfaces for a satisfactory adhesive bond. The valve
covers are treated by the UV light, bonded together using an adhesive after
treatment with a RC—747 UV lamp manufactured by Xenon Corporation and then

tested. Figure 10 is a picture of the Daimler Chrysler valve cover.

 

Lips of the Valve Cover (lips are
to be bonded)

Figure 5.2: Daimler Chrysler Valve Cover

57

5.2) UV Treatment of Flat Valve Cover Specimens

The first experiments were done initially done on flat sample plaques of
the valve cover materials as the screening experiments. These screening tests are
used to establish the process parameters for the valve cover treatment. Flat 1 in x
1 in samples are cut using the band saw. The 1 sq. in flat samples were treated in
the UV treatment chamber both in air and in ozone under various treatment
conditions in order to optimize the process parameters. For treatment in air,
supplemental air at a flow rate of 30 scfh was supplied to the chamber. For
treatment in ozone, supplemental ozone at a flow rate of 30 scfli was supplied to
the chamber while maintaining the voltage to the ozone generator at 50% V and
using oxygen as the input gas to the ozone generator. For a particular set of
experiments all other parameters such as the type of lamp, gas flow rate, ozone
concentration and lamp to sample distance were kept constant and only the
treatment time was varied. PATTI samples are made using the Ashland Pliogrip
7779 adhesive and the samples are tested. The increase in bond strength of the
treated samples can be determined using the PATTI system. Figure 5.3 shown

below gives the PATTI results of Ticona at various treatment conditions.

58

Pull off tensile strength(

 

Pull off tensile strength(psi)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6“ 00“ Cd” 00” 00% co“
0 09 o o o Q09. -0 -<\ «\c
Q o 090 09% 0,90 9 (fax Ci; &\ d?) c?\
<0 ‘b ‘b 0.5 (to (090 090 99° 09% 9°
N Q: Q) 9 {19
Figure 5.3: PATTI Results for UV Treated Ticona
1600 - "E
1400 .B T}
1200 {I
1000
800
600
400
200 2
0 “ T 1 r

 

 

Figure 5.4: PATTI Results for UV Treated Hivalloy

59

The black bars in the above figure represent an adhesive failure and the
white bars represent a cohesive failure. The untreated samples failed at very low
adhesive strengths. Maximum adhesive strength was achieved using the 120
second UV treatment condition in air and alternatively using the 90 second
treatment in ozone. The failures at these conditions are completely cohesive.

Figure 5.4 gives the PATTI results for Hivalloy at various treatment
conditions. The black bars represent an adhesive failure and the white bars
represent a cohesive failure. The untreated samples failed at very low values of
bond strengths. Maximum adhesive strength is also obtained after the 120 second
UV treatment in air and after the 90 second treatment in ozone. The failures at

these treatment conditions are also completely cohesive.

 

Untreated UV Treated

Figure 5.5: Tested PATTI stubs and substrates

PATTI tests show that UV treatment increases the bond strength to both
Ticona and Hivalloy to a point of generating a failure completely in the substrate.
Figure 5.5 shows the failure mode difference between an untreated sample

(adhesive failure) and a UV treated sample (cohesive failure). Treatments in air

60

and ozone gave comparable results. But treatment in air requires a longer time to
reach the same strength as in ozone. Even though supplemental ozone is not
supplied, the UV lamp can generate ozone by interacting with the oxygen in air.
Longer treatments are required in air as it takes some time for the creation of
sufficient ozone to interact with the surface. This transient effect would not be
present if this method was used in a manufacturing environment since the lamps

would be on all of the time.

61

5.3) Design of a Conveyor System for Treating the Valve
Covers:

In order to simulate a manufacturing operation in which the UV treatment
was used to modify the valve cover surface, a simple conveyor system was made
to treat the valve cover parts. The valve covers are 35 inches long and the distance
between the two lips of the valve cover, that are to be treated is 7 inches. The RC-
747 UV lamp is 16 inches long and 3.5 inches wide (3.5 inches is the width of the
quartz glass plate covering the lamp). In order to treat the entire valve cover, a
long lamp would have to be manufactured to provide sufficient exposure to the
entire valve cover. The valve cover could be treated in parts under the lamp but
the treatment will be non-uniform with some areas being over treated and some
areas untreated at all. Hence a conveyor system. which can transport the valve
cover under the lamp in a direction perpendicular to the length of the lamp (along
the width of the lamp) and can carry the valve cover on it was designed. In this

way the entire length of the valve cover could be treated uniformly by passing the

Power Supply & Conveyor Speed Controller

 

Motor .
Trolley Valve Cover Chain

Figure 5.6: Conveyor System for UV treating the DC valve covers

62

cover under the fized lamp. The UV treatment time can also be varied by varying
the speed of the conveyor. The conveyor system designed for treating the valve
covers is as shown in the figure 5.6.

In order that the entire valve cover is UV treated for the required time, the
conveyor was powered by a motor and chain arrangement to move it at a very low
speed. A rheostat controls the rotational speed of the motor shaft to vary the speed
of the trolley. A combination of a worm and a worm gear was added to obtain the
desired speed reduction. A 30 tooth worm gear was used to produce a speed
reduction ratio of 30:1 can be obtained. The worm gear drives a shaft and the
other spur gear sitting on the same shaft drives the chain, which passes over
another spur gear that moves on another shaft at the other end reducing the speed
appropriately enabling the entire valve cover to be treated by UV in the required
time. The rheostat, which varies the speed of the motor shaft, can be used to vary
the speed of the trolley. Variation in the speed of the trolley results in variation of
the UV treatment time. In order to supply supplemental ozone a glass delivery
system with diffusers has been designed. This gas delivery system sits on the
lamp , and is positioned so that the gas is delivered to the lips of the valve cover.
The pictures below show the mounting of the worm and the worm gear and the

ozone delivery system.

63

Worm Worm Gear

 

Figure 5.7: Worm and Worm Gear

 

Diffuser

Figure 5.8: Ozone Delivery System

64

The efficacy of the UV treatment conditions was tested using the PATTI
system on 1 sq. in samples of valve cover material. The samples were placed on
the trolley and passed under the lamp for various times (at various speeds) to
develop a new set of treatment conditions different from the previous static
treatment conditions. The following chart (figure 5.9) is a calibration curve for

setting up the desired time of treatment with a reading on the power supply.

 

 

 

 

 

 

 

 

 

 

 

 

 

900
l
g. 800 I
'6
a 700
0
5
ts ... 600
:5 m
in § 500
2* r I
8 .5 400
§ .5 \
c 300
e \
.9. 200
E
i: 100
0 I T T
20 40 60 80 100

Reading on the power supply(% rpm of motor)

Figure 5.9: Calibration Graph of the Conveyor System

The ozone concentration used with the static treatment (1 sq. in. sample
treatment in the closed chamber) was measured. The same concentration was
maintained under the lamp for the dynamic treatment and the ozone concentration

was kept constant for all the treatments. shroud of thin aluminum sheets was built

65

around the lamp in order to maintain the desired ozone concentration under the
lamp during this dynamic treatment.

In order to measure ozone concentration, a Perkin Ehner Lamda 900
Spectrometer was used. Ozone has a very strong absorption at 254 nm. This
property of UV light absorption by ozone at 254 nm was used to calculate the
ozone concentration. The spectrometer measures the absorbance of ozone at 254
nm and from the Beer’s law absorbance is directly proportional to concentration.
The ozone from under the lamp is drawn into the spectroscopy cell using suction
through teflon tubing. The flow rate of oxygen to the ozone generator is kept
constant at 30 scfh in both the static and dynamic experiments. In static
experiments a voltage of 50% V is used. At these conditions the absorbance of
ozone is measured to be 0.0239. In the dynamic case, the voltage to the ozone
generator is varied. Increasing the voltage increases the ozone concentration. It is
found from the absorbance measurements that at 30 scflr oxygen flow rate and
75% V the ozone absorbance is approximately 0.0239 as in the static case.

Figure 5.10 gives the PATTI results for UV treated Hivalloy treated using
the conveyor system. Three different adhesives are used. The black (dark) colored
bars represent the PATTI results when Essex Betamate 73005 adhesive is used.
The slate colored bars represent the PATTI results when Dow AH5643A (LESA)
adhesive is used and the gray (light) colored bars represent the PATTI results
when Ashland Pliogrip 7779 adhesive is used. The curing times of the adhesives

are listed in the results section.

66

 

 

 

Pull oft Tensile Strength(pel)

 

 

 

Untreated 3O 60 90 110 145

11me(eece)
Figure 5.10: PATTI Results for UV treated Hivalloy using the
conveyor system

Based on the above PATTI results, the 30-second treatment condition is
chosen when Dow LESA adhesive is used to bond the valve covers. The 110-
second treatment condition is used when Essex Betamate adhesive is used to bond
the valve covers and the l45-second treatment condition is used when the

Ashland Pliogrip adhesive is used to bond the valve covers.

67

5.4) Treatment of Valve Covers

Based on the PATTI test results, the optimum conditions for UV treatment
of valve covers was determined. The lips of the valve covers were cleaned using
isopropanol before the valve covers were placed on the trolley so that they can be
passed through the shroud surrounding the lamp. The speed of the trolley was
controlled depending on the treatment time required. The ozone generator was set
at 303th gas flow and the setting on the ozone generator was set at 75% V. These
conditions allowed the treatment of the entire valve cover emerging from the lamp
area that was shrouded in aluminum foil.

Figure 5.11 shows the set up for UV treatment of valve cover.

 

UV Lamp Aluminum Shroud Valve Cover

Figure 5.11: Set up for the UV treatment of valve cover

68

5.5) Bonding the Valve Covers

The UV treated valve covers are adhesively bonded together to so that
they can be tested for enhanced adhesive performance. However there are certain
problems in bonding them due to their shape. The valve covers have a bow shape
that results from their mold. The bow shape is convex upwards i.e. the lips to be
bonded bulge outward like a bow. Apart from poor surface adhesion, their bow
shape makes it more difficult to bond the valve covers.

In order to fabricate specimens for adhesive testing, two UV treated valve
covers treated at the same set of conditions were bonded to each other. Adhesive
was applied to the bottom valve cover along its lip and was spread uniformly
using a thin Teflon sheet. The lips of the top valve cover were aligned with the
lips of the bottom valve cover and then clamped together. Small spacers were
used to insure a uniform adhesive thickness. Uniform clamping force is
maintained along the length of the valve covers during bonding by using
aluminum bars as shown in the figure below. The aluminum bars are placed so
that they align with the lips on the non-bonded side (reverse side) along the length
and clamped. They are clamped at the location of each spacer. Figure 5.12 shows

adhesively bonded and clamped valve covers.

69

 

Aluminum Bars Clamp

Valve Covers

Figure 5.12: Valve covers adhesively bonded and clamped together

70

5.6) Testing the Valve Covers

The bonded valve covers are kept clamped for the curing time of the
adhesive. The clamps are then unclamped and the aluminum bars are removed.
The extra adhesive is trimmed off. The bonded pair of valve covers was cut into 2
inch sections transverse to its length resulting in 14 coupons for adhesive testing.
These samples are numbered 1 to 14 starting from one end of the bonded pair of
valve covers for all of the valve cover assemblies. These samples were tested in
tension using the United SFM-20 testing machine manufactured by the United
Calibration Corporation.

This machine is composed of two components — the load frame, where the
actual testing of the specimen takes place and the console, which contains the
computer to process and store data and control the load frame. The rate at which
the load is applied can be controlled. The position of the crosshead is measured by
an optical encoder. The crosshead speed and its position can be monitored. The
load applied to the specimen is measured by a strain-gage type load cell. The
strain imposed on the specimen may also be measured by an extensometer. These
various measurements are collected by the computer and displayed graphically
and tabulated numerically.

For testing the valve covers 1k (1000 lbs) load cell is hooked to the
machine with its position under the crosshead. The crosshead speed is set at l
inch/min. The bonded area is measured using calipers and entered in the

computer. Numerical values of the maximum tensile load and maximum tensile

71

strength appear on the computer for each run. The computer also displays a graph

between force and position for each specimen.

Load

 

Fixture connecting
to the UTS

P Steel Plate

 

 

 

 

__’ Valve Cover

 

 

 

Bonded area

 

 

 

 

 

> Screw that fits on
UTS fixture

 

 

 

 

Load

Figure 5.13: Schematic Representation of the UTS valve cover
Specimen

72

Screws that fit
on UTS

Valve Cover

 

Steel

Figure 5.14: Picture showing the cut section of valve
cover and the testing fixtures

73

5.7) Results and Discussion

The three different adhesives mentioned earlier, Ashland Pliogrip 7779,
Dow AH5643A (LESA) and Essex Betamate 73005 were used to bond the valve
covers and test their adhesive performance. The UTS test results when each of the
above adhesives is used to bond the valve covers are provided in the following
figures.
5.7.1) Ashland Pliogrip Adhesive: This is a two part Polyurethane based
adhesive. The curing time of this adhesive is 72 hours and its work time is 15
minutes. The untreated valve covers bonded using the Ashland Pliogrip adhesive
debonded as soon as the clamping force was removed. Hence, they could not be
tested. The failure was completely adhesive. The test results when the Ashland
Pliogrip adhesive was used to bond the UV treated valve covers valve covers are

shown in Figure 5.15.

500
450

350
300
250
200
1 50
1 00

50

Maximum Tensile Load (lbs)

 

12 3 4 5 678 91011121314
Sample Number

Figure 5.15: UV treated and adhesively bonded Valve cover
test results; Adhesive used: Ashland Pliogrip

74

The average tensile load for the failure is about 335 lbs. All the failures
are completely cohesive in the valve cover material. In order to confirm the
results, the tests were repeated. The results of the repetition experiment are

displayed in Figure 5.16.

500

)
.h
01
O

400
350
300
250
200

_L_L
001
00

Maximum Tensile Load (lbs

(It
0

 

O

12 345 67891011121314
Sample Number

Figure 5.16: UV treated and adhesively bonded Valve cover
regetition test results; Adhesive used: Ashland Pliogrip

The average maximum load for the repetition experiment is 308 lbs. All
the failures again were cohesive and similar to the ones in the previous
experiment. The confirmation results matched well with the previous results.

Upon visual inspection, the UV treated valve covers, bonded using the
Ashland Pliogrip adhesive appeared to have failed cohesively in the valve cover
material. To confirm this, the valve cover lip surfaces were examined using the

ESEM. The lip surfaces of untreated and non-bonded valve cover, untreated and

75

bonded valve cover, UV treated and bonded valve cover and the adhesive were

examined in the ESEM shown below.

    
   

 

Figure 5.17: As Received (untreat lFigure5.18: Untreated and
and non-bonded) valve cover; Bar = bonded valve cover; Bar =

250nm 250nm

    

 

 

{was 35'.- ‘C 3*

Figure 5.19: Cured Ashland Pliogrip Figure 5-22= UV Treated and bonded
Adhesive; Bar = 2001"“ valve cover; Bar = 450nm

76

   

Figure 5.21:UV Treated and bonded Figure 5.22: UV Treated and bonded
valve cover; Bar = 250nm valve cover; Bar = 200nm

It can be observed from the ESEM pictures that the UV treated and
bonded samples have fibers pulled out to the surface from the bulk matrix. The
cured adhesive also does not have any fibers in it. The as received samples do not
have any evidence of fiber pullout on their surface. Since fibers are only present
in the valve cover material, fiber pullout is taken as evidence of cohesive failure
as a result of the improvement in adhesion following UV treatment of the valve
cover samples.

5.7.2) Dow LESA Adhesive: The Dow LESA adhesive is an acrylic—based two
parts adhesive. The curing time of this adhesive is 48 hours and the working time
is 7 minutes. Untreated valve covers bonded with this adhesive failed before they
were tested. The valve covers debonded as soon as the clamping force was

removed. The failure mode was completely adhesive. The UTS test results for UV

77

treated and bonded valve covers using the Dow LESA adhesive are as shown in

Figure 5.23.

 

600 ’

 

500 ’ 4.

 

 

400 .

 

300 ’
200

 

0

 

d»

100

 

Maximum Tensile Load (lbs)
0

 

 

 

To TIIIIITIIIIIT
1234567891011121314

Sample Number

Figure 5.23: UV treated and adhesively bonded Valve cover test
results; Adhesive used: Dow LESA

average tensile failure load was 426 lbs. The failure in this case is not in the
bonded area. The failure is completely within the material of the valve cover
tearing either the top or bottom valve cover completely to failure. In order to

confirm the results the experiment was repeated. The results are as shown in

Figure 5.24.

78

 

 

600

500

 

 

{i

400 e e

 

 

300

200

 

Maximum Tensile Load (lbs)
0

100

 

 

 

 

O I I T T I I I T I I I I I

12 3 4 5 6 7 8 91011121314
Sample Number

Figure 5.24: UV treated and adhesively bonded Valve cover
repetition test results; Adhesive used: Dow LESA

The average tensile failure load for the repetition experiment was 421 lbs.
The failure is exactly the same in both the experimental runs. The bonded area
could not be disturbed and only a failure in the material of the valve cover could
be generated.
5.7.3) Essex Betamate Adhesive: The Essex Betamate adhesive is a two part
adhesive and has a curing time of 36 hours and a working time of 5 minutes.
Untreated valve covers could not be bonded with this adhesive since suffiecient
adhesive was not available but from the PATTI results with this adhesive, it can
be concluded that the untreated samples also have a very poor adhesion and that

the untreated valve covers bonded with this adhesive would fall apart before

79

testing in the same manner as with the other 2 adhesives. The UTS results for UV

treated and bonded valve covers using this adhesive are shown in Figure 5.25.

 

600

 

500

 

400 ‘T . 9 9 ‘ ;

 

 

300

 

200

 

Maximum Tensile Load (lbs)
9
O

100

 

 

 

O IIIIIIIIIIIII
1234567891011121314

Sample Number

Figure 5.25: UV treated and adhesively bonded Valve cover test
results; Adhesive used: ESSEX Betamate

The average load value was about 392 lbs. However the failure mode
appeared to be adhesive to the naked eye. Inspection with the ESEM (as shown in
Figure 5.27) does not suggest complete cohesive failure. For this kind of failure,
the ~390 values of maximum load are very high and comparable to the maximum
loads in the case where the failure was cohesive. This difference might be due to a
difference in toughness of the adhesive itself. Therefore, as a qualitative
comparison, the strain energies for all the three cases were computed for
comparison. Strain energy values for each sample are obtained by calculating the

area under the load-displacement curve. The area was determined from the

80

initiation of the loading cycle until the load reaches a maximum value in all the

cases. The average strain energy values are shown in Figure 5.28.

 

Bar = 350nm Bar = 200nm

Figure 5.26: ESEM pictures of failure surfaces when Essex Betamate
adhe-ve is used for bonding

81

7O

 

 

 

60

 

 

50

 

40

 

30

 

 

 

Strain Energy (lb-in)

20

 

 

 

 

 

 

10

 

 

 

 

 

 

 

Ashland Pliogrip Dow LESA Essex Betamate

Figure 5.27 : Strain Energy Comparisons of Valve cover test results
using different adhesives

These results show that maximum average strain energy was produced for
the case when the Dow LESA adhesive is used. Minimum average strain energy
was obtained in the case when Essex Betamate adhesive was used and Ashland
Pliogrip adhesive produced an intermediate average strain energy. These values
are coincide with the failure modes of the three cases.

It was observed that at a particular position along the length of the valve
cover, the adhesion is low in all the cases regardless of the adhesive used to
fabricate the samples. The position is the location of samples numbered 10 and
11. To investigate this, the lips at the location of sample number 10 were
examined using the optical microscope and compared with the sample located at

position number 6. Sample number 6 is chosen because it required the maximum

82

 

load to debond in some cases. First the as received (referred to as non-polished
surface hereafter) surfaces of the lips were observed under the microscope and
then they were polished and looked under the microscope. The microscopy
pictures are shown in the following page (Figures 5.29, 5.30, 5.31, 5.32).

It can be observed that non-polished sample number 10 has a more matrix
rich surface i.e. it has a smaller amount of fibers on its surface when compared to
the non-polished sample number 6. Whereas the polished samples numbered 10
and 6 appear to be the same and seem have the same amount fibers on their
surface. This difference in the fiber content in this location could be the result of
the how the valve cover was injection molded, i.e. the position of the gates and
the resulting knit line producing a composite with different content and hence
different properties than the rest of the valve cover, and could be the reason for

the lower adhesion of the sample number 10 when compared to the other samples.

83

 

Figure 5.28: Sample Number 10 Non-Polished

 

Figure 5.29: Sample Number 6 Non-Polished

84

 

Figure 5.30: Sample Number 10 Polished

 

Figure 5.31: Sample Number 6 Polished

85

Chapter Six

Conclusions

. Ultraviolet light treatment is an effective surface preparation technique for
adhesively joining polymers and polymer composites.

. UV treatment technique is an environment friendly method, with no
emission or use of volatile organic compounds or other chemicals.

. UV treatment method reduces the complexity of construction and
operation, and is cost effective and less time consuming.

. The PATTI results indicate that the adhesive bond strength between the
polymer substrate and the aluminum stub is dependent on the UV
light/laser treatment conditions.

. The factors influencing the adhesive bond strength of the polymer are UV
light intensity, treatment time, the environment surrounding the substrate

and the nature of the substrate.

. UV/laser treatment enhanced the surface adhesion of PET and PP

composites by a considerable amount.

. Further enhancements in surface adhesion were observed when the silane

treatment was used in a combination with the UV/laser treatment.

. On examining the ESEM and XPS results, it can be concluded that
enhancement in surface adhesion a) in case of UV treatment using the
lamps was due to the introduction of functional groups n the surface of the
polymer and b) in case of laser treatment was due to the roughening of the

surface by the laser.

86

10.

ll.

12.

13.

14.

15.

UV treatment technique was successfully adopted to surface treat
automobile valve cover panels, which have complex geometries in order
to adhesively bond them.

UV treatment of the valve covers gave promising results indicating that
this method is commercially viable. The failure modes changed from
adhesive to cohesive after the UV treatment.

Though UV treatment is a very promising technique, there are certain
challenges that need to be met before it can be used for commercial
applications.

The process needs to be optimized for every material in terms of all the
factors affecting the treatment. This at times could be a huge and time
consuming task.

If a process window which can eliminate some of the treatment parameters
basing on the physical properties of the polymer is developed, it can
simplify the optimization process.

Industrial consumers, researchers and UV lamp manufacturers need to
work together in order to design the lamps that suit a particular set of
applications.

UV treatment has all potential to replace the existing surface treatment

techniques.

87

 

LIST OF REFERENCES

Melvin J. Swanson and Gary W. Opperman, Photochemical modification of
polymers for improved adhesion, Polymer surface modification: Relevance to
adhesion, vsp 1995, pp. 319-325.

Anna Nihlstrand and Thomas Hjertberg, Adhesion properties of oxygen
plasma-treated polypropylene-based copolymers, Polymer, 1997, Vol. 38, No.
7, pp. 1557-1563.

J. Breuer, S. Metev and G. Sepold, Photolytic surface modification of
polymers with UV-laser radiation, J. Adhesion Sci. Technol, 1995, Vol. 9,
No. 3, pp. 351-363.

J. K. Park, J. D. Howard, C. W. Chen and K. Mukherjee, Laser processing of
polymer composites for adhesive bonding, American Society for Composites,
14th Annual technical conference, 1999.

W. S. Gutowski, D. Y. Wu and S. Li, Surface silanization of polyethylene for
enhanced adhesion, J. Adhesion, 1993, Vol. 43, pp.l39-155.

Buchman, H. Dodiuk, M. Rotal and J. Zahavi, Preadhesion treatment of
thermo-plastic adherends using excimer laser, Int. J. Adhesion and Adhesives,

1991, Vol. 11, No. 3, pp. 144-149.

H. Hamada, N. Ikuta. N. Nishida and Z. Maekawa, Effect of interfacial silane
network structure on interfacial strength in glass fiber composites,
Composites, 1995, Vol. 05, No. 7, pp. 512-515.

Hirosuke Watanabe and Masahide Yamamoto, Chemical structure change of a
KrF-laser irradiated PET fiber surface, Journal of Applied Polymer Science,
1999, Vol. 71, pp. 2027-2031.

M. J. Walzak, S. Flynn, R. Foerch, J. M. Hill, E. Karbashewski, A. Lin and M.
Strobel, UV and ozone treatment of polypropylene and poly(ethylene
terepthalate), Polymer Surface Modification: Relevance to adhesion, vsp
1995, pp. 253-272.

Bengt Ranby, Surface photografting onto polymers-a new method for
adhesion control, J. Adhesion Sci. Technol, 1995, Vol. 9, No. 5, pp. 599-613.

W. Chen, K. G. Neoh, E. T. Kang, K. L. Tan, D. J. Liaw, C. C. Huang,
Surface modification and adhesion characteristics of polycarbonate films
after graft copolymerization, J. Polym. Sci: Part A: Polym. Chem., 1998, Vol.
36, pp. 357-366.

88

 

David R. Kyle, Photochemical surface modification of polyolefins using
ultraviolet light, 2000, Radtech 2000 Technical proceedings.

R. Srinivasan and S. Lazare, Modification of polymer surfaces by far-
ultraviolet radiation of low and high (laser) intensities, Polymer, 1985, Vol.
26. pp. 1297-1300.

T. Kavc, W. Kern, M. F. Ebel, R. Svagera and P. Polt, Surface modification of
polyethylene by photochemical introduction of sulfonic acid groups, Chem.
Mater., 2000, Vol. 12, No. 4, pp. 1053-1059.

M. Ouyang, C. Yuan, R. J. Muisener, A. Boulares and J. T. Koberstein,
Conversion of some siloxane polymers to silicon oxide by UV/Ozone
photochemical process, Chem. Mater., 2000, Vol. 12, No. 6, pp. 1591-1596.

L. F. Macmanus, M. J. Walzak, N. S. Mcintyre, Study of ultraviolet light and

ozone modification of polypropylene, J. Polym. Sci.: Part A: Polym. Chem,
1999, Vol. 37, pp. 2489-2501.

D. Knittel and E. Schollmeyer, Surface strucuturing of synthetic fibers by UV
laser irradiation, Polym. Int., 1998, Vol. 45, pp. 103-109.

M. C Matias, M. U. De La Orden, C. G. Sanchez, J. M. Urreaga, Comparative
spectroscopic study of the modification of cellulosic materials with different
coupling agents, J. Applied Polym. Sci., 2000, Vol. 75, pp. 256-266.

J. Jang, J. Y. Lee and J. K. Jeong, Performance improvement of glass fiber
reinforced polystyrene composite using a surface modifier, J. Applied Polym.
Sci., 1996, Vol. 59, pp. 2069-2077.

H. F. Wu, D. W. Dwight and N. T. Huff, Effects of silane coupling agents on
the interphase and performance of glass fiber reinforced polymer composites,
Composites science and technology, 1999, Vol. 57, pp. 975-983.

Y. Suzuki, H. Hamada, Z. Maekawa, A. Yokohama and T. Sugihara,
Influence of silane coupling agents oninterlaminar fracture in glass fiber fabric

reinforced unsaturated polyester laminates, J. Mater. Sci., 1993, Vol. 28, pp.
1725-1732.

M. Charbonnier, M. Romand, H. Esrom and R. Seebock, Functionalization of
polymer surfaces using excimer VUV systems and silent discharges.
Application to electroless metallization, J. Adhesion, 2001, Vol. 75, pp. 381-
404.

89

Q. Song and A. N. Netravali, Excimer laser surface modification of ultra-
high-strength polyethylene fibers for enhanced adhesion with epoxy resins, J.
Adhesion Sci. Technol, 1998, Vol. 12, No. 9, pp. 957-982.

Q. Song and A. N. Netravali, Effect of pulsed XeCl excimer laser on ultra-
high strength polyethylene fiber and its interface with epoxy resin, J.
Adhesion Sci. Technol, 1999, Vol. 13, No. 4, pp. 501-516.

J. M. Hill, e. Karbashewski, A. Lin, M. Strobel and M. J. Walzak, Effects of
aging and washing on UV and ozone-treated poly(ethylene terepthalate and

polypropylene), Polymer surface modification: Relevance to adhesion, vsp
1995. pp. 273-289.

J. Y. Zhang, H. Esrom, U. Kogelschatz and G. Emig, Modification of
polymers and UV excimer radiation from lasers and lamps, Polymer surface
modification: Relevance to adhesion, vsp 1995, pp. 153-184.

R. Srinivasan and B. Braren, Ablative photodecomposition of polymer films
by pulsed far-ultraviolet (193 nm) laser radiation: Dependence of etch depth
on experimental conditions, J. Polym. Sci: Polym. Chem. Edition, 1984, Vol.
22, pp. 2601-2609.

T. Bahners and E. Schollmeyer, Morphological changes of the surface
structure of polymers due to excimer laser radiation: A synergic effect?, J.
Appl. Phys, 1989, Vol. 66, No. 4. pp. 1884-1886.

W. G. Kesting, T. Bahners and E. Schollmeyer, The effect of vacuum-
ultraviolet laser wavelengths on the surface treatment of polyolefmic

polymers, J. Polym. Sci.: Part B: Polym. Physi., 1993, Vol. 31, pp. 887-890.

B. Qu and B. Ranby, Photocrosslinking of low density polyethylene, J. Appl.
Polym. Sci., 1993, Vol. 49, pp. 1799-1807.

90

I

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

II III IIIIIIIIIIIIII IIII IIIIIIIIIIIII

320 4669