IOP PUBLISHING JOURNAL OF PHYSICS D: APPLIED PHYSICS
J. Phys. D: Appl. Phys. 40 (2007) 5502–5521 doi:10.1088/0022-3727/40/18/S10
Tribo-corrosion of coatings: a review
RobertJKWood
Surface Engineering and Tribology Group, School of Engineering Sciences, Highfield,
University of Southampton, SO17 1BJ, UK
Received 16 May 2007
Published 30 August 2007
Online at
stacks.iop.org/JPhysD/40/5502
Abstract
This paper reviews the available literature relating to the emerging research
into the performance of coatings under combined wear and corrosion
conditions. Understanding how coatings perform under these
tribo-corrosion conditions is essential if the service life of equipment is to be
predicted and to allow service life to be extended. Therefore, the
tribo-corrosion performance of coatings deposited by a variety of techniques
is discussed and the main mechanisms associated with their degradation
under combined wear and corrosion highlighted. Coating composition,
microstructure, defect level, adhesion, cohesion and substrate properties are
seen as some of the critical elements in coating performance when subjected
to tribo-corrosion contacts. The importance of post-coating deposition
treatments such as laser resurfacing and sealing are also discussed.
Interactions between wear and corrosion mechanisms are identified along
with some models and mapping techniques that aim to inform coating
selection and predict performance. Recent investigations into mono-layer as
well as multilayered and functionally graded coatings are reviewed as
candidates for wear–corrosion resistant surfaces. The review reveals the
need for a more considered approach to tribo-corrosion testing and the way
in which the results are analysed and presented. For example, the test
conditions should be appropriate to the coating system under test; the level
of in situ instrumentation deployed and the post-test analysis of in situ
electrochemical data should be carefully selected as well as details given of
the composition of any surface tribofilms formed and the identification of
the degradation mechanisms.
(Some figures in this article are in colour only in the electronic version)
1. Introduction and definitions
1.1. Background and definition of tribo-corrosion
The need to select or design new surfaces for future equipment
as well as minimize the operating costs and extend the life
of existing machinery has led to demands for a much better
understanding of surface degradation processess particularly
when tribological components are operating in corrosive
environments. This has given rise to the active research area
of tribo-corrosion which seeks to address the concerns above
and understand the surface degradation mechanisms when
mechanical wear and chemical/electrochemical processes
interact with each other.
Tribo-corrosion involves the interaction between
mechanical wear processes and electrochemical and/or
chemical corrosion processes and leads to a material loss
rate that is a summation of these effects, as shown below in
equation (1):
Wear–corrosion = mechanical wear processes
+electrochemical (and/or chemical) response. (1)
The subject started to be researched in the late 1980s
and has now emerged as an active research area as
advanced experimental techniques have been developed to
yield substantial insight into the complex processes present
in tribo-corrosion contacts. The development of in situ
electrochemical techniques and post-test analysis techniques
for surface film examination for example are powerful
tools that can be deployed in tribo-corrosion experimental
programmes. The subject, therefore, includes the interaction
of corrosion and erosion (solids, droplets or cavitation
bubbles), abrasion, adhesion, fretting and fatigue wear
0022-3727/07/185502+20$30.00 © 2007 IOP Publishing Ltd Printed in the UK 5502
Tribo-corrosion of coatings: a review
processes. For example, during friction the adhesive
dissipation of energy is often influenced by chemical effects,
Fischer and Mischler [1].
1.2. Industrial and medical relevance
Tribo-corrosion is encountered in many technological areas
where it causes damage to human joints, slurry handling
equipment and machines and can compromise safe operation
and in the case of biotribocorrosion of prosthesis and
restorative dentistry tribo-corrosion has implications for
human health and quality of life. However, the chemo-
mechanical mechanisms of tribo-corrosion are not well
understood and are extremely complex as they involve the
properties of contacting material surfaces or coatings, the
mechanics of the contact and the corrosion conditions.
Understanding is further hampered by the experimental
difficulty in characterizing surface phenomena occurring
within the tribo-contacts. A multitude of reactions probably
occur simultaneously within the contact and the quantity of
reaction products is likely to be very small, making their
detection and analysis difficult. In addition, metastable
phases could be generated within the contact but these are
subsequently transformed into stable reaction products outside
of the contact or have time dependent transformations.
Tribo-corrosion is often linked to the synergy resulting
from the coupling of mechanical and environmental effects.
This synergism or antagonism results in material degradation
that is often much larger or smaller than would be expected
from a simple summing of the mechanical and environmental
effects.
On the positive side, tribo-corrosion phenomena can
be used as a manufacturing process such as in chemical–
mechanical polishing of silicon wafers. The coupling of
mechanical and environmental effects can also create surfaces
of specific reaction layers on materials which could inhibit
corrosion and/or wear. Examples of this are self-lubricating
and/or self-healing surface layers as mentioned by Celis and
Ponthiaux [2].
Tribo-corrosion degradation affects components in
numerous industries such as mining, automotive, food, nuclear,
offshore, marine and biomedical just to name a few. These
industries expend the equivalent of millions of pounds every
year to repair damage, typically in ignorance that it has been
caused by tribo-corrosion processes. Typical examples of
this kind of material removal are erosion–corrosion damage
to pumps, impellers, propellers, valves, heat exchanger tubes
and other fluid handling equipment. The performance data
and models published in the open literature aspire to allow
informed surface selection for combined wear and corrosion
resistance, but this information is vulnerable as few systematic
test programmes are reported and many reports are incomplete.
Therefore, there is little basic understanding of how even
the most popular surfaces used in machinery perform when
exposed to corrosive conditions which may often include
temperature fluctuations. The transitions from relatively
benign surface loss rates to unacceptable loss rates are not
defined. Hence, identical machines often suffer from vastly
different surface damage rates due to only subtle differences
in the conditions of operation or for example in countries with
different climates.
Substrate
Passive film
Coating
Counter face
Figure 1. Schematic showing depassivation of a passivating coated
surface by three body rolling abrasion induced by relative motion
between the coated surface and a counter face.
1.3. Interfacial aspects and the role of oxides
Stainless steels rely on a 1–10 nm thick surface oxide
passive film for their protection from aggressive and corrosive
environments. This oxide film forms instantaneously when
oxygen is available in the environment but abrasion can lead
to the local rupture or complete removal of these films.
This can lead to areas of the substrate being exposed to the
aggressive environment and unless repassivation (repair or
self-healing) mechanisms reform the passive film, accelerated
anodic dissolution will occur within these sites. The response
of some of these passivating surfaces relates to the rate of
depassivation due to stripping (or damage) to the protective
passive (e.g. oxide) film and the rate of repassivation (healing
of the passive film). Clearly if the former is higher than the
latter, then nascent bulk material is exposed to the corrosive
environment and high dissolution rates may result. Therefore,
the wear–corrosion performance of metallic coatings relates
not only to the integrity of the coating and whether the substrate
is exposed to the environment but also to the passive film state
on the coating surface. Figure 1 illustrates passive film and
coating removal that can accelerate wear–corrosion under three
body rolling abrasion while figure 2 shows how an impinging
solid particle can damage passive films of ductile metallic
surfaces and how this is dependent on impact angle. However,
there is no direct relationship between surfaces that readily
passivate and are thought to be corrosion resistant and lower
loss rates under wear–corrosion. This lack of understanding
has badly hampered improving the performance of machines
in aggressive environments.
The presence of oxides could also affect the level of
plastic deformation and the depth of surface penetration made
by the harder asperities into the softer counter body surface.
The presence of well-adhered oxide surface films or loosely
adhered layers of corrosion products and the ability to reform
these films and layers is also critical in many engineering
applications where control of friction and wear is important.
Friction forces for certain materials can be modelled by
considering summing the forces associated with adhesive and
deformation processes. Adhesive forces are often related to the
shear strength of junctions made between contacting asperities
of surfaces in contact with each other but in relative motion.
The presence of oxide films on these asperities could clearly
influence the resulting forces opposing motion and surface
temperatures generated during contact.
5503
RJKWood
1.4. Additional factors associated with coatings
Understanding the tribo-corrosion of coatings is far harder
compared with many monolithic surfaces as the coating
processes used can unintentionally degrade the coating
microstructure by adding significant porosity and pockets
of oxides into the coating and the coating process can also
affect the substrate properties. Also remnants from the pre-
coating substrate treatments (such as shot or grit blasting)
can be embedded into the surface and then subsequently
coated, thereby degrading the adhesive properties of the
coating/substrate interface or the coating cohesion by
weakening the bonding between any inclusions and the matrix.
These add to the already complex interactions that can occur
in wearing contacts as shown in figure 3, which include
electrochemical activity of the wear debris with the contacting
Figure 2. Removal or fracture of passive films dependent on
erodent impingement angle. (1) Initial undamaged passivated
surface, (2) low angle impingement results in cutting and removal of
passive film and substrate material, (3) intermediate angle of
impingement results in some cutting and removal of passive film
plus some plastic cratering and fracture of the passive film and
(4) impingement normal to the surface leads to plastic
cratering/lipping and fracture of the passive film.
Figure 3. Schematic illustration of the tribo-corrosion contact involving two metallic samples first body and second body and abrasive
particles (third body). Wear particles originating from the first and second bodies (4 and 5) form additional contact components. The
corresponding material fluxes are also indicated.
surfaces and material transfer between contacting surfaces
which can radically change the composition of the surface and
the wear and corrosion resistance of that surface.
Complex microstructures with multiple phases of coatings
can lead to micro-galvanic activity and selective phase
corrosion. Examples of such surfaces include composites
or surfaces that undergo tribologically induced compositional
changes. For instance the presence of carbides in a metallic
surface, typically formed for improved wear resistance,
establishes a micro-corrosion cell as the carbide is likely to
be cathodic with respect to the surrounding metallic matrix.
This can result in preferential anodic dissolution of the metallic
matrix close to or at the matrix/carbide interface and thereby
accelerate carbide removal from surfaces and reduce the anti-
wear properties of the surface, [3].
1.5. Basic phenomena to consider
Tribo-corrosion includes the interaction of corrosion with
• solid particle erosion,
• abrasion,
• cavitation erosion,
• fretting,
• biological solutions,
• sliding wear and tribo-oxidation.
These interactions are represented schematically in
figure 4.
Cost reduction considerations favour replacing expensive
solid alloy components with coatings on inexpensive carbon
steel substrates. The relatively cheap option of organic
coatings tends to perform poorly under high energy solid
particle impingement but have a use as corrosion barrier
coatings within low energy flow components, see Percy [4],
whereas ceramic materials are usually too expensive to use,
except in particularly critical applications. Sprayed metallic
coatings are a relatively unexplored possibility for this duty.
The most likely metallic coating materials for thermal spray
systems, on cost grounds, are aluminium, copper, nickel and
zinc, their alloys and possibly composite materials based
on them. For more aggressive environments Co-based
(e.g. sprayed and fused stellite) or Ni based (Inconel) coatings
are preferred as they provide enhanced performance against
wear–corrosion. For highly aggressive (very energetic flows)
WC–CoCr or WC–NiCr cermets are used either in sintered
form or thermally sprayed by detonation type or high velocity
oxy-fuel (HVOF) thermal spray guns [5–7].
5504
Tribo-corrosion of coatings: a review
Cavitation
Liquid
droplet
Solid
particle
Erosion
Rolling
(3-body)
Grooving
(2-body)
Abrasion Fatigue Fretting
Corrosion
Wear
Biological
systems
Figure 4. Possible interactions between corrosion and the various wear mechanisms.
To achieve long term corrosion resistance, it is important
to understand the galvanic interaction between the metallic
coating and the substrates. For example, anodic coatings to
carbon steel will protect the steel galvanically in electrolytes.
For cathodic coatings where defects in the coating occur
(i.e. porosity), exposed steel is likely to form the anode of
an electrochemical cell and rapidly corrode, adding to wear–
corrosion attack and spallation of the coating. It would be a
condition for using cathodic coatings that either the coating is
defect free or that a sealant is used if substrate protection is to
be achieved. Coating interfaces which can be coating/coating,
coating/substrate or coating/interlayer/substrate are vulnerable
parts of any coating system. They can distort subsurface
stress fields induced by tribo-contacts and cause spallation
of the coating by lowering the interfacial adhesion of the
coating and promote interfacial crack propagation. As a way
of summarizing the above introductory comments, and to
introduce the breadth of this research field, the possible surface
responses to a variety of contact conditions are given in table 1.
The deposition of coatings typically involves multiple
steps such as in PVD deposition of thin Cr
3
C
2
films on stainless
steels. The process starts with chromium ion etching of the
substrate to enhance epitaxial film growth and coating adhesion
to the substrate. This is followed by the deposition of a
CrN base layer to further enhance adhesion and finally the
deposition of the top Cr
3
C
2
coating [8]. PVD multilayered
coatings are even more complex and vary the composition
of layers to maximize E/H (elastic modulus/hardness) ratios.
Thus, this can give non-linear relationships between tribo-
corrosion rates and coating depths.
1.5.1. Fretting. Fretting damage, such as suffered by gas
turbine blade roots and by replacement joints in humans, is
associated with wear–corrosion processes. Barril et al [9]
found only fretting regimes involving slip led to measurable
wear and to an enhancement of the anodic current. Previously
developed tribo-corrosion models for passivating metals were
found to describe the effect of normal force and displacement
amplitude on the anodic current. For a smooth hard indenter
sliding against a soft passive metal the current can be
given by I
r
:
I
r
= k
1
λ
v
s
F
n
H
Q
p
, (2)
where k is the probability factor, λ is the length of the sliding
direction, v
s
is the sliding velocity, F
n
is the normal force, H
is the metal hardness and Q
p
the passivation charge density.
The fretting corrosion behaviour of a Ti6Al4V alloy in
contact with an alumina ball has been investigated by Barril
et al [10] in a 0.9% sodium chloride solution. The tests
were carried out using a fretting test rig equipped with an
Table 1. Summary of tribological effects on corrosion/surface
phenomena.
Contact conditions Response
Contact load 1. Contact stress (surface and
and geometry subsurface) and contact area
2. Elastic or elastic-plastic or
fully plastic surface and subsurface
response
3. Surface deflection (elastic or
permanent)
Relative motion 1. Induces depassivation
under load of passive surface
(causing wear) and primary ion release
in a corrosive 2. Effects E
corr
(mixed potential
environment of worn and unworn surfaces).
3. Induces surface and near
surface phase transformations
4. Effects repassivation kinetics
5. Increases local mass transport
6. Increases contact temperature
7. Increases roughness and
thus area of surfaces in contact
8. Changes surface layer compositions
9. Induces a mechanically mixed
layer on the surfaces of the
tribocouples including
embedded wear debris.
10. Establishes a galvanic
process between worn area
and unworn area as a result of (2).
11. Induces surface strain hardening
12. Affects shear strength of
asperity/asperity junctions
between tribocouples and thus
adhesive contributions
to the friction force.
Varying the applied 1. Free corrosion potential [8]
load of dynamic 2. Local current density [8]
contacts 3. Number of active anodic
areas and anodic site locations
(removable anodes) [8]
4. Surface recovery time [8]
5. Change wear mechanism
or force transition to
more/less severe loss rates
Tribo-corrosion 1. Effects friction
2. Local pH of environment
(where appropriate)
3. Total mass loss from
surface (mechanical
and electrochemical)
Wear debris 1. Generation of metallic
debris causes secondary
ion release as
debris react with environment
and corrode themselves.
5505
RJKWood
electrochemical cell. The effects of applied potential and
oscillation frequency were evaluated for micro-motions of
50 and 100 pm amplitude. The extent of wear was strongly
influenced by the applied potential, especially for potentials
above approximately −0.2 V Ag/AgCl. This effect was related
to the increasing oxidation of the third body formed by plastic
deformation of the metal found at higher anodic potentials.
On the other hand, the mechanical energy necessary for wear
was found to be dramatically decreased when anodic potentials
were applied.
1.5.2. Medical significance of tribo-corrosion. Researchers
are becoming increasingly interested in metal-on-metal
(MoM) joint implants with studies being focused on their
wear and corrosion (ion release) behaviour and the subsequent
influence on human biology. The corrosion, wear and wear–
corrosion behaviour for three materials (high carbon CoCrMo,
low carbon CoCrMo and UNS S31603) have been discussed by
Yan et al [11]. Corrosion effects on the overall performance
for the three materials are analysed. Two distinct regimes
have been found for the three materials: (a) the running-in
regime and (b) the steady state regime. Even in the steady state
regime, 20–30% of the material degradation can be attributed
to corrosion-related damage. High carbon CoCrMo showed
excellent corrosion, wear and corrosion–wear resistance and
therefore it delivered the best overall performance in terms of
a lower wear rate, a lower friction coefficient and a higher
resistance to corrosion. The role of proteins in the joint fluids
and attachment onto surfaces are thought to play a key role but
this mechanism is not well understood.
1.6. Aims of this review
The aim of this review is to overview research into the
wear–corrosion of engineering coatings and dental overlays.
It will identify the coating systems that have been, or are
being, investigated and, wherever possible, identify their
application and suitability for industrial environments. It
will highlight the additional issues associated with coated
surfaces compared with monolithic surfaces as well as the
main degradation mechanisms. It also aims to discuss the
current state of the art in the understanding of the interactions
between mechanical wear processes and electrochemical
and/or chemical processes. Appropriate models will also be
reviewed. However, the subject area is in its infancy and the
literature, to date, is dispersed making generic conclusions
difficult and the conclusions that can be made are very
coating/substrate/environment system specific.
2. Wear–corrosion interactions
Wear–corrosion interactions can be defined as follows. The
total damage under erosion–corrosion, T , can be represented
as:-
T = E + C + S, (3a)
while under wear–corrosion, the total damage, T , is given by
(i.e. sliding or abrasive wear):
T = W + C + S, (3b)
where E is the pure erosion material loss, W is the mass
loss due to wear, C is the solids free flow corrosion rate
and S is the synergistic or interactions term. There are
numerous interactions between wear and corrosion and almost
as many ways experimentally to quantify them. Mostly these
interactions are referred to as synergistic and quantify the level
of synergy. Synergy is defined as ‘the difference between
wear–corrosion and the summation of its two parts’ and can
be expressed by equation (4).
S = T − (E + C) = (C
e
+ E
c
), (4)
where T , C and E are typically gravimetric terms relating
to wear–corrosion, electrochemical corrosion without tribo-
influence and mechanical wear mechanisms, respectively. The
interactive processes can be broken down into two components,
E
c
and C
e
, where E
c
is the corrosion-enhanced wear and
C
e
is the wear-enhanced corrosion. Recent literature has
defined E
c
as the synergy term and C
e
as the additive term.
For the erosion case, the synergistic effect (interactive
term), S, is referred to as E
c
or (C
e
+ E
c
) depending
on the literature source and under what conditions C has been
obtained. Thus care must be taken when extracting synergistic
levels for different materials when using multiple sources of
literature. The S terms and how they should be measured are
given by the ASTM G119-93 standard which is a useful guide
to evaluate synergy [12].
Wear can mechanically strip the protective corrosion
film creating fresh reactive corrosion sites and producing
C
e
[13], dependent on the rate of repassivation and
the integrity of the film formed. Other possible wear-
enhanced corrosion mechanisms include (i) local acidification
at wear sites, accelerating corrosion rates and prohibiting film
formation, (ii) increased mass transport by high turbulence
levels, (iii) lowering the fatigue strength of a metal by
corrosion, (iv) anodic wear scars can cathodically polarize
the surrounding unworn surfaces and destabilize passive films
in these regions enhancing corrosion [14] and (v) surface
roughening of the specimen during wear-enhanced mass
transfer effects increasing the corrosion rate [15]. Corrosion-
enhanced wear mechanisms are also possible (E
c
). The E
c
wear rate could be due to (vi) the removal of work hardened
surfaces by corrosion processes which expose the underlying
base metal to erosion mechanisms [16], (vii) preferential
corrosive attack at grain boundaries resulting in grain loosening
and eventual removal [17], (viii) the increase in the number of
stress concentration defects resulting from micropitting and
(ix) detachment of plastically deformed flakes on the metal
surface due to stress corrosion cracking. Most of the above
mechanisms, if dominant, would be expected to lead to positive
synergy.
Malka et al [18] give an example of where positive
synergies are dominant and have looked at whether erosion
enhances corrosion and/or corrosion enhances erosion in pipe
loop experiments on uncoated AISI 1018 carbon steel in 1 wt%
NaCl solutions purged with CO
2
(partial pressure of 1.2 bar)
and containing 2 wt% silica sand (275 µm). They found that
erosion enhances corrosion and corrosion enhances erosion but
the dominant synergistic effect was that of corrosion on erosion
i.e. E
c
. For roughly equal corrosion (C) and erosion rates
5506
![Figure 8. Erosion–corrosion regime map with electrochemical material loss versus mechanical material loss for erosion–corrosion of carbon steel AISI 1020, HVOF nickel–aluminium bronze coating and cast nickel–aluminium bronze under jet impingement conditions with 2% w/w sand in seawater. Reprinted with permission from [24]. Copyright 2007 Elsevier.](/figures/figure-8-erosion-corrosion-regime-map-with-electrochemical-2dhw2w36.webp)