Satas D., Tracton A.A. (ed.). Coatings Technology Handbook
Подождите немного. Документ загружается.
62
ZORLL
3.0
DELAMINATION PROCEDURES
There is a certain contrast between the tensile methods, in which stress conditions at the
interface are
of
primary concern. and another group
of
test methods, in which the delamina-
tion effects that sometimes occur in the practical use of the coating are duplicated. In
these methods, the test piece is, under idealized conditions, subjected to peeling forces.
These forces attack the bond between paint film and substrate either at well-defined lines,
so
that the coating is detached in strips,
or
at one point only. thus causing delamination
that progresses radially
in
the form of
a
blister. Both situations are encountered in practice.
Thus. an appropriate method can always be selected according to the actual needs.
Such methods are equivalent
to
a
model of the typical treatment that the coating
may undergo. This is the most essential advantage of the methods in practical terms. But
they do not as a rule lead to results
so
clearly defined as the maximum stresses at the
interface. Only under certain conditions can the values of the stresses be calculated from
results
of
these tests, and this is possible only if all geometrical and dynamical factors,
on
which the test procedure depends are available.
As
a test result, the work of adhesion is regularly obtained. This quantity is in any case
useful for comparative purposes (i.e., without reference to a particular absolute adhesion
standard). But the work of adhesion also provides some insight into the mechanism of
bonding in energetic terms.
3.1
Knife-Cutting Method
There are two special methods for testing the degree of adhesion, and the procedures for
loosening the bond between paint film and substrate are rather similar (Fig.
6).
In
one
method. film separation is obtained by means of a sharp knife, pushed along the interface
with an exactly measured force.' Although this seems to be
a
simple test method, the
process of detachment
is,
in fact, complicated, comprising both shear and tensile stresses,
which finally cause disbonding of the film. Moreover, a leverage effect plays a part in
the separation process of the coating, and the intensity
of
that effect can be characterized
by the angle under which the force is acting upon the film.
knife-cuttmg method peel adhesion test
adhesion value
V
:
(force per unit length
l
1
V=L7
1
-cos
0
Figure
6
Devices for measurmg adhesion
on
thc
basis
of
delaminatmn procedures.
ADHESION
TESTING
63
hard and brittle
film
n\
A
low
adhesion strenoth
m
hard and brittle film
igh adhesion strength
elastic and flexible
film
low
adhesion strength
"""
\I
l?/
high adhesion strength
elastic and flexible
film
Figure
7
Influence of mechanical paint film properties
on
the
results of
the
knife-cutting test.
In the knife-cutting method, the principle of removing the paint film is analogous
to the machining of metal in
a
lathe. A particular system of forces becomes effective here.
It is determined by rake angle of the knife, coating thickness, friction between cutting
tool and coating
as
well
as
substrate, amount of energy stored elastically in the film and
energy losses caused by plastic deformation, fracture energy occurring during decomposi-
tion within the film, and other effects
of
minor importance.
To
obtain meaningful results, all those parameters should be controlled strictly, or
their influence at least estimated
as
exactly
as
possible. In many cases, however, the details
of the separation process reveal,
as
indicated in Figure
7,
how much relevance can be
attributed to the various factors.
3.2
Peel
Test
In practice,
a
coating often fails not by becoming simultaneously detached over an extended
area, but by gradual peeling, for example, starting from
a
scarcely covered edge, or from
a
line-shaped damage zone. Thus, it appears sensible to duplicate these conditions in an
appropriate test procedure.
The peel test was originally developed for adhesive tape, where its use appears
natural and simple. For measuring the adhesion
of
a
coating, however,
a
strip
of
adequate
width must first be marked on the sample by two parallel cuts of sufficient length.
An analysis
of
the test conditions,
as
indicated in Figure
6,
reveals the influence
of
the angle under which the load is applied.' Thus, in contrast to the first impression, that
the test procedure ought to be simple, the details of the separation process turns out
to
be
rather complicated.x Detachment of the film occurs under the combined effects
of
both
positive and negative tensile stresses and shear stress.
64
ZORLL
Besides, the viscoelastic character of the coating material must be taken into account,
and this feature is also affected by the amount
of
pigmentation in the film.' As a conse-
quence of the viscoelasticity, the test results depend to a remarkable degree, on the velocity
of the detachment process.
Thus, the general experience is that the peel test, when used for measuring adhesion
of paint films, can certainly be considered to be a practical method, but it is also understand-
able that the results obtained cannot be interpreted in terms of the bonding mechanism.
3.3
Blister Method
Formation
of
blisters is generally a first sign
of
deterioration
of
any coating designed for
protection against corrosion. It is therefore sensible to investigate the behavior
of
such a
coating material under conditions that finally may lead to film detachment in the form of
blisters,
This test
(Fig.
8) is normally carried out in the following way. First, before the
liquid paint is spread over the surface of the sample, a hole is bored into the substrate at
the site at which the blister is to occur. The hole is plugged with a material such as Teflon,
on which there will be no adhesion during painting at all, to permit easy removal of the
plug after film formation.'" It is also possible to generate the hole after painting, using
spark erosion techniques.
The detachment
of
the film is initiated, and supported, by providing hydrostatic
pressure in the hole, either with a fluid (oil, mercury etc.)
or
with pressurized air.'' In
any case, the pressure is a primary measure of the progress of the debonding process.
To
obtain adhesion values, either as maximum stress
or
as bonding energy (work of adhesion),
both height and diameter of the blister must be monitored. An optical system with relatively
low magnification can prove sufficient for that purpose." From these geometrical data,
together with the tensile modulus of the film and its thickness
a
critical pressure value
can be calculated. It is pressure that causes the blister to grow, and thus pressure can serve
as a basis for determining the adhesion strength.I3
work
of
detachment
W
=
-65
Pd P
4.75
E
-
d.y3
ab
E:
tensile modulus
,-
blister diameter 2a
coating
substrate
4
inflow
of
pressurized medium
Figure
8
Blow-off
test
for
measuring
adhesion
based
on
determination
of
blister
dimensions
and
pressure.
ADHESION
TESTING
65
Preparation
of
test pieces as well as procedures of measurement are comparatively
complicated, but this disadvantage is virtually compensated
for
by the basic information
the test results yield in terms of the mechanism of blister formation.
4.0
LOCAL
DEBONDING
SYSTEMS
In practice, mechanical damage on the coating often occurs locally, in form of scratches
or
impact deformation.
As
a consequence, not only are the appearance properties of the
film deteriorated, but its adhesion may be lost
as
well, if the intensity of the external
forces, transferred through the film, is high enough. Idealized conditions for such treatment
have become the basis of special adhesion test methods. Prototypes
of
pertinent devices
exist, and their scope of application has been studied in detail. Moreover, the deformation
caused by such external loading processes has been analyzed, and theoretical relations are
now available for assessing the extent of debonding at the interface.
4.1
Scratch
Technique
Figure
9
illustrates the principle
of
a test method in which scratch resistance
as
well as
adhesion of the coating can be measured.
A
loaded stylus is drawn
across
the film surface.
There is obviously a simple relationship between the applied load,
of
which the intensity
probe under defined load
balancmg load
/
/
=L
,
,
A
bar bearmg pwot
"-
specmen motion
hardness operation
force
(adheston value)
I(
Bm
Figure
9
Scratch technique for determining the bonding
of
polymer films subjected to local
surface forces.
66
ZORLL
can be controlled easily in the balance system carrying the stylus, and the shearing force
eventually causing detachment of the coating.
A
critical quantity is the radius at the tip
of the tool, and it must be measured with the same accuracy as the width
of
the contact
area generated during the scratching procedure.
It is re~ommended'~ that the information obtained directly from the test be supple-
mented by additional data, such
as
results of surface profilemeters or scanning electron
microscopes, which provide insight into the details of the scratch topography. Under such
provision, even the adhesion values in multilayer systems can be estimated. It can be seen
what type of film failure occurs in those systems when they are subjected to scratch
loading.
4.2
Indentation Debonding
If
a
needlelike indenter is perpendicularly pressed into the surface
of
a
coating, which is
bonded to
a
virtually undeformable substrate, most of the deformation will occur within
the film, but there also will be
a
certain debonding effect at the interface.
According to that situation, indicated in Figure
10,
a
peeling moment can be calcu-
lated, which may serve as
a
measure of the film's capacity
to
withstand delamination in
\
1
i/
indenter
coating
\
/
start phase
/
substrate
reversible
deformation
\'l
indentation
tension
rn
bond
peeling moment
Figure
10
Principle
of
the indentation process
for
measuring adhesion at the polymer-substrate
interface.
ADHESION
TESTING
67
the vicinity of the indentation site.
Is
It is convenient, especially for thin coatings, to monitor
the gradually increasing area of debonding with optical devices capable of leading to an
evaluation on the basis of Newton's rings, for instance, with transparent coatings.
It is by no means necessary
to
restrict the scope of the indentation test to using
a
needle for penetration
of
the coating. Indenters of other typical shapes also are used
successfully. It has been shown that the best approach for taking into account the boundary
conditions at the interface, as
a
basis for the calculation
of
adhesion values, is to use
a
60"
angle cone
as
an indenter.16 An essential advantage
of
the identification test can be
seen in the fact that it yields values for the bond strength in absolute terms, as Well
as
information about the durability of the connection between coating and substrate under
those specific loading conditions.
4.3
Impact
Tests
Especially for determining the stone-chip resistance of coatings that are supposed to
provide efficient protection against corrosion
of
a
metal substrate, the value
of
adhesion
at the interface is of primary interest. The situation encountered in practice can be duplica-
ted, under only slightly idealized conditions, by means
of
a steel ball impinging
on
the
test piece (Fig.
11).
The transfer of forces through the film is, in
a
first approximation,
equal to the case
of
static loading; that is, it can be calculated in virtually the same way
as
for an indentation test.
Thus, in debonding area, two types
of
stress are active: negative tensile (i.e., compres-
sive) stress
in
the center
of
the circular detachment site, and, still more essential, shear
stress in the annular region. The dimensions of the debonding area, mainly the maximum
diameter, can serve
as
a
measure of adhesion at the interface, it is convenient to use either
that diameter
or,
better still, the area of the debonding zone, both
as
(reciprocal) measures
of adhesion. An extended site of separation is thus an indication
of
a
low adhesion level."
compresslon
shear
tC".L"LI
c
).
).
-
-
shear
-""
"-"
V"""
influence
of
film thickness
Figure
11
Information
about
adhesion on the basis
of
the circular debonding figure produced
by
impact loading of
the
paint film.
68
ZORLL
It is possible, in principle at least,
to
calculate the value of adhesion in absolute
terms, but then some additional parameters must be taken into account. There is first the
remarkable influence of film thickness.
For
a thin film, the detachment area is more
concentrated in the region in the vicinity
of
the axis of impact, areas with thicker coatings,
under the same test conditions, a larger area is affected by the stresses leading to delamina-
tion. The details
of
this relationship can easily be understood on the basis of the particular
scheme that is valid
for
the stress distribution within the film as a whole.
The other parameters for the calculation
of
adhesion strength are related to the
impinging steel ball. Its mass, diameter, and speed-before and after impact-must be known,
as well as the (generally very short) time of impact, in which the energy is being transferred
into the film.
5.0
FLAW
DETECTION METHODS
In practice, it is often necessary
to
know as fast as possible about any deterioration
of
the
bonding strength between film and substrate; quantitative details of the adhesion values
however, are not required. The occurrence at the interface,
of
any flaws, even those of tiny
dimensions, is an event belonging
to
this category. Thus tests allowing for identification
of
the first signs of adhesion failure are
of
great interest.
Some methods, based on acoustical or thermo-optical principles, have been proposed
and. to a considerable degree, tested already for that purpose. The unique potential
of
these test methods, most of them introduced only recently, promises very interesting appli-
cations in the future.
5.1
Ultrasonic Pulse-Echo System
Monitoring of ultrasonic echoes, according to the time domain principle,lx is widely used
to identify any irregularities in materials that are supposed to be flawless. This method
thus can be applied also to evaluate the quality
of
bonding between
a
polymer coating
and the substrate, whatever its nature.
According
to
Figure
12,
an incident
of
ultrasound will be partly reflected and trans-
mitted at each interface at the test piece, including the backing. It is the pulse partly
coating
substrate
t
intensity Intensity
time time
ADHESION
TESTING
69
transmitted at the interface that then undergoes more
or
less total reflection at this free
surface. The bonding quality is expressed in the relative heights of the pulses.
With an intact bonding at the interface, the amplitude of the pulse reflected there
will be fairly low. in contrast to the amplitude
of
the transmitted pulse that travels through
the substrate and is reflected on its free boundary, appearing later at the sensor accordingly.
If
a
defect is assumed at the interface, containing air or indicating in another way that
the joint has been disbonded, the amplitude
of
the related ultrasonic pulse will increase
considerably, owing to the very low acoustic impedance now prevailing at that site. There
is thus less intensity available for the transmitted pulse, and its amplitude would then
decrease considerably.
The region subjected to that ultrasonic treatment can be restricted to a relatively
small size. Consequently, there is a possibility
of
virtually scanning the test piece, line
after line, to identify any sites at which adhesion strength might have been lost. The extent,
even the particular shape, of that defective area can be determined in the test procedure."
5.2
Acoustic Emission Analysis
Since it has been observed, by means of very sensitive acoustical sensors, that any debond-
ing effects within an originally uniform and coherent material are accompanied by a
specific burst of (mostly ultrasonic) pulses, this principle has also been proposed for
monitoring the behavior of adhesive joints under load."
Thus, the method can also be applied with slight modifications (Fig.
13),
to examine
the extent
of
debonding phenomena occurring in a coating system on a given substrate,
if
the whole test piece is subjected to gradual stretching, normally in one dimension."
The acoustical signals obtained are related to individual fracture events in the test system.
As
a rule. it will be the detachment process of the film that is indicated in this way.
Also,
orlglnal coating
I
pigmented
I
substrate
sources
of
acoustlc emisslon
"
"
tension
t
"c
detachment
of
coating
t
acoustic emission data
pulse rate
pulse sum
l
time
-
I1
typlcal acoustic emission process
time or Intensity
of
tenslon
-
Figure
13
Application
of
acoustic emission analysis
for
monitoring the onset
of
coating detach-
ment.
70
ZORLL
however, any separation within the film (e.g., between binder matrix and pigment), could
cause acoustic signals, although most probably, of course, on a lower level
of
intensity.
This possibility, however, together with the knowledge that there may be a wide
variety in form and distribution
of
the acoustic pulse spectrum, elucidates the present
situation with this test. It provides for a promising potential in terms of flaw detection in
coating systems; nevertheless, a great deal of information is required if the primary results
are to be interpreted correctly.
5.3
Thermographic Detection of Defects
The main advantage of thermographic methods can be seen in the fact that they yield
remote mapping of the distribution of temperature on a surface, whatever the cause may
have been. This principle has been applied successfully for the detection of subsurface
flaws, which may occur as a consequence of any deterioration in the bonding quality of
coating and substrate.”
As
indicated schematically in Figure
14,
it is the heat flux in the test piece, proceeding
toward the surface, that, if there is an obstacle in its way in the form of a flaw or a
debonded area, produces more or less pronounced differences in the otherwise uniform
temperature on the surface. The shape of the area in which the temperature decreases is
made visible at the surface by means of an infrared sensor. It is a rather exact reproduction
of the real defect, which would have remained invisible otherwise. The agreement between
real shape of debonded area and its “image” on the surface will be the better, the lower
the distance is between the position of the flaw and the surface. The situation is similar,
on optical grounds, to the formation of a shadow from
a
distant object.
effect
of
thickness
d
t
L“
Lateral dimension
x-
coating
effect of convectlon
E
intense convection
t“‘__
faible convection
Figure
14
Thermographic detection
of
debonding areas by their impedance
to
heat
flow
though
the substrate-film system.
ADHESION TESTING
71
As a rule of thumb, the linear dimension of the debonded area should be at least
twice the distance from the interface to the surface. Under these conditions, a distinct
reproduction of the defect on the surface can be expected.
6.0
OUTLOOK
Although adhesion is in principle a well-defined quantity for a system consisting of paint
film,
or
adhesives, and its pertinent substrate, there are
so
many possibilities for practical
combinations that it would be unrealistic to look for universal adhesion test systems. On
the contrary, the relatively great number of test methods proposed
so
far is the best proof
of the diversity
of
the means available for tackling the problem of adhesion measurement
in practical terms.
In many cases, the selected method will lead to satisfactory results, but it is neverthe-
less advisable to make it a practice to follow up the development activities in the field
of
adhesion testing. Many of the existing methods were proposed when there was a need for
a particular reproduction of typical bonding situations. It is thus not unrealistic to expect
that the successful methods of today will be supplemented, in the not
so
distant future,
by even more sophisticated methods with accordingly high performance qualities.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
R.
J.
Good, “Theory of cohesive vs adhesive separation in an adhering system,”
J.
Adltes.,
4, 133-154, (1972).
International Standard
IS0
2409, Paint and Varnishes-Cross-cut Test,”
International Standard
IS0
4624, Paint and Vanishes-Pull-off Test for Adhesion.”
J.
Sickfeld, “Experimental aspects of adhesion testing-An IUPAC study,”
J.
Oil
Color.
Chem.
P. C. Hopman, “The direct pull-off test,”
J.
Oil
Color.
Cl7em.
Assoc.,
67, 179-184 (1984).
W.
K.
Asbeck, “The measurement of adhesion in absolute units by knife-cutting methods:
The Hesiometer,” presented at the Eleventh FATIPEC Congress, Brussels, 1972.
Congress
A.
N.
Gent and G.
R.
Hamed, “Peel mechanics,”
J.
Adhes.,
7, 97-95 (1975).
A. D. Crocombe and R.
D.
Adams, “Peel analysis using the finite element method,”
J.
Adhes.,
P. M. Heertjes and
J.
de Jong, “The peeling-off of paint film,”
J.
Oi/
Color. Chetn. Assoc.,
A. Elbasir,
J.
D. Scantlebury, and
L.
M. Callow, “The adhesion
of
chlorinated rubber to mild
steel using
a
blister technique,”
J.
Oil
Color.
Chem.
Assoc..
68, 282-284 (1985).
H. Dannenberg, “Measurement of adhesion by
a
blister method,”
J.
Appl.
Pol.ym.
Sei.,
5,
L.
A. van der Meer-Lerk and P. M. Heertjes, “The influence of pressure on blister growth,”
J.
Oil
Color.
Chem.
Assoc.,
64, 30-38 (1981).
M. G. Allen and
S.
D. Senturia, “analysis of critical debonding pressures
of
stressed thin films
in the blister test,”
J.
Adhes.,
25, 303-315 (1988).
J.
Ahn,
K.
L.
Mittal, and
R.
H. McQueen, “Hardness and adhesion of filmed structures
as
determined by the scratch technique,” ASTM STP 640. Philadelphia: American Society for
Testing and Materials, 1978
pp.
134-157.
P. A. Engel and D. D. Roshon, “Indentation-debonding
of
an adhered surface layer,”
J.
Adhes.,
IO,
237-253 (1979).
ASSOC.,
61 292-298 (1978).
Book,
pp.
P78-P87.
12, 127-139 (1981).
55,
1096-1106 (1972).
125-134 (1961).