ASM Metals HandBook Vol. 8 - Mechanical Testing and Evaluation
Подождите немного. Документ загружается.
proliferation of different testing machines for adhesion, friction, and wear. The use of established voluntary
standards can help, but only if the standard applies directly to the problem of current concern. In the absence of
widespread, commonly used test methods, it is necessary to analyze the applied conditions associated with each
set of results carefully to determine the extent to which they can be compared to other work.
As with mechanical testing in general, commercial adhesion, friction, and wear testing machines are becoming
increasingly computer automated. While automation has obvious advantages, it also necessitates conscientious
calibration to ensure that the sensors and control mechanisms provide accurate readings to the computer.
In adhesion, friction, and wear testing, as in other forms of mechanical testing, the four most important
requirements are (a) understanding the characteristics of the test method being applied, (b) expecting differing
degrees of repeatability from different material types, (c) selecting the right testing tool for the job, and (d)
coupling measurements with physical observations of contact surfaces to ascertain the causes for the measured
behavior.
Introduction to Adhesion, Friction, and Wear Testing
Peter J. Blau, Oak Ridge National Laboratory
References
1. P.J. Blau, Friction Science and Technology, Marcel Dekker, Inc., 1996
2. H. Czichos, Tribology: A Systems Approach, Elsevier, Amsterdam, Netherlands, 1978, p 322–325
3. P.J. Blau, Wear Testing, Metals Handbook Desk Edition, ASM International, 1998, p 1342–1347
4. P.J. Blau, Glossary, Friction, Lubrication, and Wear Technology, Vol 18, ASM Handbook, ASM
International, 1992, p 1–21
5. D.F. Moore, Principles and Applications of Tribology, Pergamon Press, Oxford, 1975, p 62–85
Introduction to Adhesion, Friction, and Wear Testing
Peter J. Blau, Oak Ridge National Laboratory
Selected References
• B. Bhushan and B.K. Gupta, Handbook of Tribology, McGraw-Hill, 1991
• Friction and Wear Testing Source Book, ASM International, 1997
• Friction, Lubrication, and Wear Technology, Vol 18, ASM Handbook, ASM International, 1992
• Metals Handbook Desk Edition, 2nd ed., ASM International, 1998
• R.G. Bayer, Mechanical Wear Prediction and Prevention, Marcel Dekker, Inc., 1995
• Special Technical Publications, ASTM
• W.A. Glaeser, Characterization of Tribological Materials, Butterworth-Heinemann Ltd., Boston, 1993
• Wear Control Handbook, M.B. Peterson and W.O. Winer, Ed., American Society of Mechanical
Engineers, 1980
Adhesion Testing
Introduction
ADHESION refers to the interfacial bond strength between two materials in close proximity with one another.
Adhesive bond strength can be described in several ways, depending on the nature of the interface. In physical
chemistry, for example, adhesion is a fundamental term that refers to the attractive force between a solid
surface and a second phase in either liquid or solid form. In this context, adhesion is a manifestation of the
innate interatomic and intermolecular bonds that occur between the surfaces of two materials.
Other meanings of adhesion also arise in different disciplines related to mechanical engineering and the
evaluation of coatings and films. In railway engineering, for example, adhesion often means friction (Ref 1) or
the sliding resistance between two materials. In this context, the term adhesion refers to mechanical adhesion,
which is defined as the adhesion produced by the interlocking of protuberances on the surfaces in an interface
(Ref 1).
Adhesion also has important practical meaning in the evaluation of coatings, adhesives, and composite
materials. Thin films (<1 μm, or 0.04 mil), thick films (>1 μm), and bulk coatings (>25 μm, or 98 mils) all
depend on adhesion, which can be evaluated and measured in a variety of ways, depending on the product
configuration and application requirements. Therefore, it is not surprising that many methods are used to
measure adhesion for films, coatings, and adhesive-bonded joints. Indeed, there are so many variations on
adhesion measurements in coatings, surface films, and adhesives (Ref 2, 3, 4, 5, 6) that it would be impossible
to describe them fully here.
Therefore, the purpose of this article is to describe briefly common adhesion measurement techniques for the
three basic types of adhesion outlined by Mittal (Ref 2 and 6):
1. Fundamental (or basic) adhesion
2. Thermodynamic adhesion
3. Practical adhesion
Common measurement methods for each type of adhesion are briefly discussed, with the main focus on
practical adhesion testing of coatings and thin films. However, to illustrate the use of adhesion testing in
materials research, this article also includes a section on the use of adhesion tests in the evaluation of stress-
corrosion cracking (SCC) within bimaterial interfaces.
References cited in this section
1. P.J. Blau, Glossary of Terms, Friction, Lubrication, and Wear Technology, Vol 18, ASM Handbook,
ASM International, 1992
2. K.L. Mittal, Ed., Adhesion Measurements of Films and Coatings, VSP, 1995
3. K.L. Mittal, Ed., Adhesive Joints, Plenum Press, 1984
4. G.L. Schneberger, Ed., Adhesives in Manufacturing, Marcel Dekker, 1983
5. G.P. Anderson, S.J. Bennett, and K.L. DeVries, Analysis and Testing of Adhesive Bonds, Academic
Press, 1977
6. K.L. Mittal, Ed., Adhesion Measurement of Thin Films, Thick Films and Bulk Coatings, STP 640,
ASTM, 1978
Adhesion Testing
Fundamental Adhesion
Fundamental adhesion refers to the basic intermolecular forces that occur whenever two materials are in close
proximity. These intermolecular forces that act between the surfaces of bodies are called surface forces (Ref 7,
8), and adhesion is one manifestation of the existence of surface forces.
Surface Forces
Fundamental adhesion arises from innate surface forces, which have their origins in well-understood
interatomic and intermolecular forces (Ref 7). Such forces are always present, and they can be described by the
summation of individual bond strengths over a unit area or by the energy required to break chemical bonds at
the weakest plane or loci of points within an interface (Ref 2).
It is sometimes convenient to classify surface forces as either short-range or long-range surface forces. Short-
range surface forces are those that act between atoms and molecules that are essentially in contact, say within
0.1 or 0.2 nm of each other. Examples of this are covalent and hydrogen bonding, as well as Born repulsions.
Long-range surface forces act between surfaces that are farther apart, which in this context means on the order
of a few nanometers. Examples of long-range surface forces include van der Waals and electrostatic forces.
However, some short-range surface forces also produce effects over a longer range. An example of this is steric
repulsion. In this case, surfactant molecules adsorbed on a solid surface (via short-range bonding) can prevent a
second surface from approaching the first. Boundary lubricants operate in this way to keep solid surfaces
separated.
In general, short-range forces are stronger than long-range forces and make the most important contributions to
adhesion. Unfortunately, long-range forces are easier to both measure and model. Therefore, knowledge of
surface forces, from both a theoretical and experimental standpoint, is much sounder for long-range effects.
This is why an understanding of fundamental adhesion based on short-range surfaces forces has proven to be
difficult (Ref 9).
Measurement of Surface Forces
“Long-range” surface forces act over surface separations from 1 to 100 nm and cause forces levels in the range
of about 10
-7
to 10
-4
N. Measuring these forces is not a trivial matter. The magnitude of the forces increases
with surface area (or, as shown below, with the radius of curved surfaces). Thus, in order to have measurable
forces, it is desirable to have extended areas of surface. At the same time, to make sensible measurements of
surface separation on a nanometer scale, it is necessary for the surfaces to be extremely smooth. A method of
detecting small forces is required, as are methods of controlling and measuring very small surface separations.
Although surface force measurements have been made for several decades, the inherent difficulties of surface
preparation and cleanliness limited the number of materials studied and the amount of data to a small level,
until about 1970. The most popular technique has been the crossed-cylinder apparatus devised by Tabor (Ref
10) and further developed by Israelachvili (Ref 11). This technique uses a surface force apparatus (SFA), which
is commercially available. A review of the field up until 1982 is also provided in Ref 12.
The SFA consists of a closed stainless steel chamber designed to enclose a variety of liquid or vapor media and
is usually operated at ambient temperature and pressure. Force is measured between two cylindrical surfaces,
with the axes of the cylinders at right angles to each other (Fig. 1). The reason for choosing this geometry is for
purposes of practicality. If two planar surfaces were to be used (as one might have supposed), then there would
be extreme difficulties in forming surfaces of sufficient flatness, mounting them exactly parallel, maintaining
parallelism while moving the surfaces together or apart, and avoiding edge effects, all while working on a
nanometer scale. As shown in the next section, “The Derjaguin Approximation,” it turns out that there is no
difficulty interpreting the forces measured in this odd geometry because there is a simple relationship between
the force measured between crossed cylinders and the energy of interaction between flat surfaces.
Fig. 1 Surface force apparatus, in which two thin solid substrates are mounted as cross cylinders, with
one of them supported by a cantilever spring whose deflection measures the force. An optical
interferometric technique is used to measure the distance between the surfaces.
In order to measure surface separation, the SFA employs an optical interference technique (Ref 11, 13, 14).
Under optimal conditions, this gives a resolution of 0.1 nm or better. Of course, one drawback is that it places a
limitation on the solid materials that can be investigated; namely, that at least one of the pair whose surfaces
approach contact must be transparent and rather thin (ideally, a few mm). Most of the measurements made with
this apparatus have been made with thin foils of mica bent around and glued to cylindrical glass lenses.
To implement the optical method, a 95% reflecting silver layer is coated on the outer (that is, remote) surface of
each solid substrate. Collimated white light is shone through the two substrates and whatever medium separates
them (Fig. 1). Multiple-beam interference between the two silver layers selects only certain wavelengths of
light, which are passed by the interferometer. All other wavelengths interfere destructively and are not
transmitted. The transmitted light is collected and directed to a grating spectrometer, which spreads it according
to wavelength, so that discrete wavelengths appear at the exit port of the spectrometer as spatially separated
fringes of equal chromatic order.
The wavelengths depend on the thicknesses and refractive indices of the materials that are included in the
interferometer: usually, the two transparent substrates and whatever fluid medium is between them.
Measurement and analysis of the wavelengths allow computation of these thicknesses (Ref 13, 14). Because the
two solids are of fixed thickness, those values can be subtracted from the total to give the thickness of the
intervening medium, that is, the separation, D, between the inner (adjacent) surfaces of the solids at their closest
point of approach.
The surface force that one solid substrate exerts on the others is measured by a simple spring-deflection
method. One solid is mounted on a cantilever spring, the remote end of which is moved up or down using a
three-stage drive mechanism. The first stage is a micrometer that allows coarse positioning of the surfaces from
a separation of a few mm to a few μm. The second stage is a micrometer that acts through a differential spring
mechanism, which reduces the motion a thousand-fold, allowing positioning to approximately 1 nm. Finally,
voltage applied to a piezoelectric tube expander gives positioning to a fraction of 1 nm.
After calibrating the drive mechanisms, it is straightforward to monitor any differences between a movement of
the remote end of the spring and the distance moved by the end that bears one of the solids. This difference
corresponds to a deflection of the spring. Multiplying it by the spring stiffness (typically 100 N/m, or 7 lbf/ft)
gives the increment in force resulting from the movement. Because both the calibration and the movement of
the solid are measured with a resolution of ~0.1 nm, it is possible to measure very small force changes (10
-7
to
10
-8
N) using this technique.
The Derjaguin Approximation. The force, F
c
, between two gently curved surfaces is proportional to the
interaction energy per unit area, E
f
, between two flat ones at the same separation. This relationship, known as
the Derjaguin approximation, allows straightforward interpretations to be made of surface force measurements
between crossed cylinders (or between one sphere and another or between a sphere and a flat plate). It is also
helpful in certain adhesion measurements, as described below.
The Derjaguin approximation is derived (see Ref 7, for example) by considering the force between each
element of one curved surface and each element of the other, and then integrating over the two surfaces to
obtain the total force. As long as the radius of curvature is much larger than the range of the surface force, this
is approximately equivalent to integrating the force per unit area, F
f
, between flat surfaces, from the minimum
separation of the curved surfaces, D, to an effectively infinite upper limit, with some geometrical factors to
account for the shape of the surfaces. The integral simply gives the interaction energy between flats, E
f
(D),
which is the work done against the surface forces in moving the flat surfaces from infinity to D. For two
spheres of radius R
1
and R
2
, the geometrical factor is a constant, giving:
F
c
(D) = 2π R E
f
(D)
where 1/R = 1/R
1
+ 1/R
2
. It can be shown that the geometry of crossed cylinders of equal radii, R
c
, is equivalent
to a sphere of radius R
c
approaching a flat plate, or to two spheres of radius 2 R
c
approaching each other.
Substrate Materials. The original and still most common solid material used in surface force measurements is
mica, chosen because it satisfies the requirements (thin and transparent) of the optical interference technique
used in the SFA and because it is easy to prepare large areas of molecularly smooth surface by cleavage.
Experiments have been conducted on mica surfaces immersed in many different liquid and vapor environments
(Ref 8).
Recently, there has been some success in extending these measurements to a wider range of surfaces. One
approach is to coat mica surfaces by various techniques, including Langmuir-Blodgett deposition, surfactant or
polymer adsorption from solution, plasma modification, and evaporative coating of thin metal, carbon, and
metal-oxide films. An alternative approach is to find a means of preparing other transparent materials as
micron-thick foils with very smooth surfaces. This has been done for sapphire, silica, pyrex glass, and certain
polymers. It is reasonable to expect that the range of materials studied will continue to increase in the near
future.
Currently, the best way to prepare metal surfaces for SFA appears to be thin-film evaporation onto mica or
another smooth substrate. Because the optical technique requires some light to pass through the two films, their
thicknesses cannot be more than a few tens of nanometers. It is possible to use the metal films themselves as
one or both optical interferometer mirrors, but the fringes of equal chromatic order would disappear from the
visible spectrum if the two metal surfaces were brought closer together than about 1 μm (40 μin.). In that case,
an alternative method of measuring separation, such as capacitance, would be required.
Environments. Tests with the surface force apparatus can be conducted in many different liquids or vapors, as
long as they are compatible with the materials of the SFA system (namely, stainless steel, silica, Kel-F, and
Teflon). There is a provision to heat the chamber to around 100 °C (212 °F). Use of an appropriate heating
jacket could extend the temperature range from, perhaps, -50 to 150 °C (-60 to 300 °F). At about 150 °C (300
°F), the silver layers used for interferometer mirrors degrade. This limit might be raised by using other optical
coatings. The next limitation of the current design would be the maximum operating temperature of the Teflon
seals, which is 250 °C (480 °F). In principle, the same or comparable techniques could be extended to operate
at several hundred degrees, but in practice this would require a major redesign of the apparatus.
At present, the SFA is intended only to operate at or near ambient pressure. With some modifications to the
seals, it could be made to hold moderate vacuum, say 10
-4
Pa (10
-6
torr). A total redesign would be required to
build a device for making comparable measurements in ultrahigh vacuum (UHV) conditions.
Preparation of Surfaces and Fluids. Any solid to be investigated by the SFA method should be smooth,
compared to the range of forces under examination. Because the adhesion between surfaces is often dominated
by very short-range forces, atomically smooth surfaces would be required to make fundamental and
reproducible measurements of these. However, rough surfaces still adhere, and so measurements can be made
without insisting on atomic smoothness. The drawback, in that case, is that it would be more difficult to obtain
a straightforward interpretation of the results.
Because SFA measurements involve extremely small surface separations, there are stringent requirements for
cleanliness. One speck of dust in the wrong place can spoil the entire test. The relative importance of surface
cleanliness is again related to the range of force under investigation. For very short-range forces, even a
monolayer of adsorbed vapor will dramatically influence results. Considerable care must also be taken in
preparing any liquid or vapor environment. Vapors must be free of dust, and liquids must be free of both
particulate and molecular (for example, polymer or surfactant) contamination that can easily adsorb to solid
surfaces and affect the results.
Other Measurements with SFA. The SFA method can be used to measure other properties of thin liquid or
vapor films and solids at or near contact. These properties include the thickness of an adsorbed layer, the
refractive index of very thin liquid films or adsorbed layers (Ref 13), the viscosity of ultrathin liquid films (Ref
15, 16), and the friction between two molecularly smooth solids (Ref 17).
References cited in this section
2. K.L. Mittal, Ed., Adhesion Measurements of Films and Coatings, VSP, 1995
7. J.N. Israelachvili, Intermolecular and Surface Forces, Academic Press, London, 1985
8. R.G. Horn, Surfaces Forces and Their Action in Ceramic Materials, J. Am. Ceram. Soc., Vol 73, 1990, p
1117–1135
9. R.G. Horn, Measurement of Surface Forces and Adhesion, Friction, Lubrication, and Wear Technology,
Vol 18, ASM Handbook, ASM International, 1992, p 399
10. D. Tabor and R.H.R. Winterton, The Direct Measurement of Normal and Retarded van der Waals
Forces, Proc. R. Soc. (London) A, Vol 312, 1969, p 435–450
11. J.N. Israelachvili and G.E. Adams, Measurement of Forces between Two Mica Surfaces in Aqueous
Electrolyte Solutions in the Range 0 to 100 nm, J. Chem. Soc. Faraday Trans. I, Vol 74, 1978, p 975–
1001
12. K.B. Lodge, Techniques for the Measurement of Forces between Solids, Adv. Colloid Interface Sci., Vol
19, 1983, p 27–73
13. J.N. Israelachvili, Thin Film Studies Using Multiple-Beam Interferometry, J. Colloid Interface Sci., Vol
44, 1973, p 259–272
14. R.G. Horn and D.T. Smith, Analytic Solution for the Three-Layer Multiple Beam Interferometer, Appl.
Opt., Vol 30, 1991, p 59–65
15. D.Y.C. Chan and R.G. Horn, The Drainage of Thin Liquid Films between Solid Surfaces, J. Chem.
Phys., Vol 83, 1985, p 5311–5324
16. J. Peachey, J. Van Alsten, and S. Granick, Design of an Apparatus to Measure the Shear Response of
Ultrathin Liquid Films, Rev. Sci. Instrum., Vol 62, 1991, p 463–473
17. A.M. Homola, J.N. Israelachvili, M.L. Gee, and P.M. McGuiggan, Measurement of and Relation
between the Adhesion and Friction of Two Surfaces Separated by Molecularly Thin Liquid Films, J.
Tribol., Vol 111, 1989, p 675–682
Adhesion Testing
Thermodynamic Adhesion
Thermodynamic adhesion refers to the change in free energy when an interface is formed or separated. This
concept of adhesion is defined in terms of surface energy, interfacial energy, and the work of adhesion.
Surface energy (γ
s
) is the work required to create a unit area of new surface from bulk material. It is commonly
defined as half the reversible work (W
c
) required to first divide a monolithic solid into two new surfaces and
then to pull the two surfaces far apart in a vacuum (where far apart means beyond the range of the applicable
surface forces). The term W
c
is also called the work of cohesion.
In general, the creation of a new surface from a separation of materials involves the breaking of interatomic
bonds, whose number (and possibly strength) depend on crystallographic orientation. Therefore, surface energy
varies from one crystal face to another (Ref 18). The act of creating two surfaces requires work to be done
against the attractive surface forces, and thus the integration of surface force, as a function of separation from 0
to ∞, gives a quantity:
W
SF
= - F
f
(D)dD
(Eq 1)
where F
f
(D) is the surface force between flat solids at a separation, D, using the convention that F < 0 for an
attractive force. The subscript SF indicates that this is the contribution to W
c
from the surface forces.
Sometimes W
SF
is equated to twice the surface energy, but this is not generally correct because other energetic
processes are likely to occur during the creation of new surfaces. In particular, the new surfaces are likely to
reconstruct. Atoms near the new surface rearrange their positions to a configuration that is more favorable near
a surface than was their original configuration in the bulk of the solid. Furthermore, there is the delicate issue of
assigning the energy associated with breaking strong, short-ranged atomic bonds when the original solid is
“magically” cleaved. This is certainly a contribution (probably the major one) to W
c
, but it is unclear whether or
not it should be included in the definition of W
SF
.
Interfacial Energy. When a solid is not in vacuum but is in contact with a liquid or vapor, it is considered in
terms of interfacial energy, γ
SL
or γ
SV
, respectively, rather than surface energy. If the above process is carried
out in either a liquid or vapor environment, the work of cohesion equals twice the interfacial (solid-liquid or
solid-vapor) energy. The same remarks about reversibility and reconstruction still apply. Furthermore, in this
situation there is also the likelihood that molecules of the liquid or vapor will adsorb to the new solid surface
(which could be thought of as “reconstruction” of the fluid at the interface). Adsorption reduces the interfacial
energy.
Note that the reconstruction of both solid and fluid depends on which materials are involved. For example, a
given solid material may reconstruct differently in water than in a hydrocarbon liquid. The details of the
interfacial structure are also likely to depend on the distance between one interface and the other. Careful
consideration also must be given to the question of reversibility. If the adsorption is irreversible, then bringing
the two solids back together after they have been separated will not remove the adsorbate, and the surfaces will
never return to their original intimate contact and will not reform the original atomic bonding that existed in the
monolithic solid.
The interfacial energy, γ
LV
, between a liquid and its own vapor is also called the surface energy, or, more
commonly, the surface tension of that liquid. The term “tension” comes from a real force that resists any
attempt to increase the surface area of a given volume of liquid. Surface tension of a liquid can be measured
directly. Determining either the surface or interfacial energy of a solid is much more problematic. Solid-liquid
and solid-vapor surface energies are related to the liquid surface tension and the contact angle, θ, made by the
liquid on the solid through Young's equation:
γ
SV
= γ
SL
+ γ
LV
cos θ
(Eq 2)
Work of adhesion (W
A
) refers to the energetic cost of creating new interfaces of each solid material in contact
with the environment, γ
1E
and γ
2E
, where the subscript E could be liquid or vapor, as appropriate, less the
energetic gain from removing the original solid-solid interface. The energy associated with the solid-solid
interface is also termed an interfacial energy, γ
12
, which is defined by the Dupré equation:
W
A
= γ
1E
+ γ
2E
- γ
12
(Eq 3)
Adhesion is sometimes used to mean work of adhesion. In a similar way, one could define the work of adhesion
between two pieces of the same solid material. This would differ from the work of cohesion if the two pieces
were brought together in an environment so that adsorbate remained at the interface or if they were formed with
their crystal orientations misaligned, giving a grain boundary at the interface. In that case, the quantity γ
12
would be nonzero and would correspond to a grain-boundary energy, with or without extrinsic molecules
present. Because grain-boundary energy depends on grain-boundary angle, the adhesion between two crystals
of the same material depends on their relative orientation.
Note that there is another effect that can make the above definition of work of adhesion problematic. When two
solids are pulled apart, it is quite possible, and often observed in practice, that separation does not occur
precisely between the two materials. Many solids diffuse into one another when in contact, particularly at high
temperature. This makes it almost impossible to separate them in the ideal way that the simple thought
experiment would imply. Even without diffusion, it is frequently found that joints break parallel to, but not
precisely at, the interface. Many experimentalists have detected at least one of the materials on both sides of the
division following separation (Ref 18).
When the environment includes a condensable vapor, an additional contribution to the adhesion between two
solids occurs from an effect known as capillary condensation. Liquid condenses (as long as the contact angle is
less than 90°) to form a bridge wherever the gap between the solids is small. Curvature of the liquid-vapor
meniscus results in a negative Laplace pressure in the liquid, which acts as a cohesive force that holds the solids
together. The magnitude of this force depends on the relative vapor pressure of the liquid and on the geometry
of the solids (Ref 7 and 19). In some situations, this is the predominant factor in adhesion.
References cited in this section
7. J.N. Israelachvili, Intermolecular and Surface Forces, Academic Press, London, 1985
18. D.H. Buckley, Surface Effects in Adhesion, Friction, Wear and Lubrication, Elsevier, Amsterdam, 1981
19. F.M. Orr, L.E. Scriven, and A.P. Rivas, Pendular Rings between Solids: Meniscus Properties and
Capillary Force, J. Fluid Mech., Vol 67, 1975, p 723–742
Adhesion Testing
Practical Adhesion
There are many types of adhesion tests for specific geometries and modes of separation for coatings, films, and
adhesive joints (Ref 2, 3, 4, 5, 6). A number of them are either empirical or semiempirical. Others give only a
comparative test of adhesion, such as whether the adhesion between a thin film and a substrate is greater or less
than the adhesion between the film and some reference material.
In general, practical adhesion tests involve the application of a known stress to a joint or interface to determine
when it fails. However, the force required to separate two bodies with mating surfaces depends very much on
the manner in which they are separated. For example, two microscope slides held together by a thin film of
moisture are extremely difficult to separate in simple tension or even by wedging them apart, whereas they
separate easily in shear. The reason for this lies partly in the fact that the work of adhesion can be done by
applying a large force over a small distance (uniform tension) or a small force over a large distance (as in
peeling or sliding). In a similar way, the fracture threshold also depends on the mode of separation, such as
tension versus shear. Therefore, an adhesion test carried out in one mode might reveal little about failure in
another.
Adhesive Joints. Various arrangements that can be employed in adhesion tests are illustrated in Fig. 2. For large
pieces (as opposed to films or coatings), the most obvious arrangement is the butt joint (Fig. 2a). Although it
may look simple, the ease of fracture or failure can depend strongly on the presence of flaws in the joint and on
how and whether a crack/separation is initiated at the edges of the sample. A lap joint (Fig. 2b) is appropriate to
test shear strength, for example, in laminates. Care must be taken to avoid excessive bending of the beams
during the test, because that introduces some tensile component. The ring shear, or “napkin ring,” test (Fig. 2c)
applies a more uniform shear stress to the joint, whereas the peg topple test (Fig. 2d) is closer to ideal tension,
but not straightforward to analyze.
Fig. 2 Various arrangements for adhesion tests
The double-cantilever-beam geometry (Fig. 2e) provides an excellent fracture-mechanics type of test if suitable
samples can be prepared. A clever variation on this is to profile the beams in such a way that the fracture
threshold is independent of the length of the crack (Ref 20), making the analysis very simple. These types of
practical adhesion tests are most commonly applied to adhesive-joint testing, which is described in more detail
in the article “Testing of Adhesive Joints” in this Volume.
Films and Coatings. When thin films, thick films, or coatings are involved, there is another set of test
geometries that can be used. The peel test is very common, with the force applied at various angles, not just the
90° angle shown in Fig. 2(g). Blister and delamination tests (Fig. 2f and h) can be very well controlled and
properly analyzed in terms of fracture mechanics (Ref 2 and 21). The scratch test (Fig. 2i) is much more
difficult to analyze, but it at least has the merit of being the only one of these configurations that tests adhesion
under dynamic sliding conditions. More detailed coverage of the various methods is contained in Ref 2 and 6.
Mechanical testing and indentation testing of thin films are also covered in the article “Evaluation of
Mechanical Properties of Thin Films” in the ASM Handbook, Volume 5, Surface Engineering (p 642–646).
References cited in this section
2. K.L. Mittal, Ed., Adhesion Measurements of Films and Coatings, VSP, 1995
3. K.L. Mittal, Ed., Adhesive Joints, Plenum Press, 1984
4. G.L. Schneberger, Ed., Adhesives in Manufacturing, Marcel Dekker, 1983
5. G.P. Anderson, S.J. Bennett, and K.L. DeVries, Analysis and Testing of Adhesive Bonds, Academic
Press, 1977
6. K.L. Mittal, Ed., Adhesion Measurement of Thin Films, Thick Films and Bulk Coatings, STP 640,
ASTM, 1978
20. W.D. Bascom, P.F. Becher, J.L. Bitner, and J.S. Murday, Use of Fracture Mechanics Concepts in
Testing of Film Adhesion, Adhesion Measurement of Thin Films, Thick Films and Bulk Coatings, STP
640, K.L. Mittal, Ed., ASTM, 1978, p 63–81
21. D.B. Marshall and A.G. Evans, Measurement of Adherence of Residually Stressed Thin Films by
Indentation. Mechanics of Interface Delamination, J. Appl. Phys., Vol 56, 1984, p 2632–2638
Adhesion Testing
Adhesion and Interfacial Degradation
Natalia Tymiak and W. Gerberich, University of Minnesota
Gaining an insight into mechanisms of environmentally induced interfacial degradation requires an
understanding of the fundamental aspects of adhesion between two dissimilar materials. For any application
involving multilayers, fiber/matrix, or film/substrate systems, strength of bonding between two materials is one
of the critical reliability issues. Environmental exposure is also a key factor in the interfacial degradation of
various bond interfaces of metal/polymer (Ref 22, 23), ceramic/polymer (Ref 24), metal/ceramic (Ref 25, 26),
and so forth. For interfaces involving metals, the possibility of hydrogen-induced interfacial degradation exists
under exposure to gaseous hydrogen (Ref 27) or during cathodic hydrogen discharge (Ref 28). The latter may