ao meu marido e aos meus pais - universidade do minho ... · minha “maninha”, e aos meus pais,...
TRANSCRIPT
Ao meu marido e aos meus pais
Master Dissertation
II
Master Dissertation III
Agradecimentos
Gostaria de expressar a minha sincera gratidão a todas as pessoas que directa ou
indirectamente me acompanharam e ajudaram no desenvolvimento e realização desta
dissertação de mestrado. No entanto, devo agradecer mais directamente:
- Ao Professor Doutor Luís Rocha e à Doutora Edith Ariza, meus
orientadores e amigos, por todo o apoio, pelas estimulantes discussões de ideias,
encorajamento e acima de tudo, pela amizade.
- A todos os membros do CIICS – Centro de Investigação em Interfaces e
Comportamento de Superfícies - laboratório onde o trabalho experimental desta
dissertação foi desenvolvido. Trabalhar com um equipa de investigação excelente como
a do CIICS é essencial para o desenvolvimento de um trabalho científico de qualidade.
O companheirismo, a amizade, a entreajuda e o bom ambiente de trabalho é
fundamental para continuar quando a motivação falha.
- À Direcção do CIICS pela amável forma como sempre me trataram e pelo
voto de confiança que em mim depositaram ao me terem aceite como investigadora do
centro de investigação.
- Ao Professor Doutor Jean-Pierre Celis, da Katholieke Universiteit Leuven,
da Bélgica, pelos preciosos ensinamentos e ajuda no desenvolvimento inicial do
trabalho.
- Ao Departamento de Engenharia Mecânica da Universidade do Minho, na
data representado pelo Professor Doutor José Carlos Teixeira, presidente do demais
departamento, assim como às restantes pessoas do departamento, pelo respeito e carinho
com que sempre fui tratada.
- Aos técnicos Sr. Miguel Abreu, Sr.ª Leonor Carneiro, Sr. Vitor Neto, Sr. Leite
e ainda Sr. Sérgio Carvalho, pela ajuda na preparação das amostras assim como na
execução de certos ensaios.
Master Dissertation
IV
Master Dissertation V
- À amiga Ana Ribeiro que trabalhou, directamente comigo no desenvolvimento
desta dissertação e que além da ajuda a nível de trabalho, também a sua amizade foi
muito importante.
- Ao amigo Jorge Pereira que sempre esteve disponível para me ajudar, sendo
também uma pessoa fundamental durante o desenvolvimento da minha dissertação de
mestrado.
- A todos os restantes meus amigos (eles sabem quem são), pela amizade,
encorajamento, compreensão e por estarem comigo, do meu lado quando todas as forças
me pareciam abandonar.
- Ao meu marido André por todo o amor e paciência. Ele é o meu pilar de
orientação.
- E por fim, à minha família, que eu amo muito, por todo o encorajamento,
paciência e pela constante presença. Quero agradecer especialmente à minha irmã, à
minha “Maninha”, e aos meus pais, sem os quais a minha vida não teria sentido.
A todos, muito obrigado!
Master Dissertation
VI
Master Dissertation VII
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial
saliva solutions in the presence of corrosion inhibitors and pH variations
Abstract
Degradation of Ti dental implants is a common process usually caused by
mechanical stress and/or by the physiological environment (human saliva) that surround
the implant. These types of implant are most of the time subjected to micro-movements
at the contact region with bone or at the implant/porcelain interface (due to the
transmitted mastication loads) and chemical solicitations (oral environment). Such
implant becomes part of a tribocorrosion system, which may undergo a complex
degradation process that can lead to implant failure. Additionally, the passive film,
which naturally grows on the metallic implant surface, can be scratched or destroyed
during the insertion and implantation into the hard tissue by abrasion with bone and
other materials.
In this work, two different tribological arrangmets were studied. Fretting-corrosion
and reciprocating pin-on-plate tests were performed in different equipments specially
adapted for tribocorrosion experiments. Artificial saliva was used as electrochemical
solution and an alumina ball (φ = 10 mm) was used as counterbody. Citric acid was
added to artificial saliva in order to investigate the influence of a pH variation on the
tribocorrosion behaviour of the material. Additionally, three different inhibitors were
added to investigate the action of cathodic and anodic reactions on the electrochemical
response. Also, the influence of inhibitors which might be included in the formulation
of tooth cleaning agents or medicines was investigated. During fretting tests, the
degradation mechanisms were investigated by electrochemical noise technique, which
provided information on the evolution of corrosion potential and corrosion current
during fretting tests. In reciprocating tests, two different electrochemical conditions
were imposed: OCP and potentiostatic control in the passive region of the polarization
curve (1V) of Ti samples. Also, to obtain more detailed information on the
characteristics of the original and reformed passive film, EIS measurements were made
before and after the mechanical damage. In both cases, all samples were characterized
using SEM, EDS, and AFM techniques.
Master Dissertation
VIII
Master Dissertation IX
Depassivation and repassivation phenomena occurring during the tests were
detected, and are discussed. The pH decrease and the presence of anodic inhibitors
demonstrate a helpful influence in the improvement of the tribocorrosion properties of
cp Ti. pH decreases The repassivation evolution of commercially pure Ti seems to be
affected by pH decreases. No improvement in the repassivation kinetics was suggested
with the presence of corrosion inhibitors, in artificial saliva solution.
Master Dissertation
X
Master Dissertation XI
Comportamento de Ti comercial em sistemas de fretting - corrosão e evolução da
repassivação, em soluções de saliva artificial, na presença de inibidores de
corrosão e com alterações de pH.
Resumo
A degradação de implantes dentários feitos em titânio é um processo comum que
normalmente acontece devido à acção de solicitações mecânicas e/ou devido à
degradação por parte do ambiente fisiológico em que o implante esta inserido (saliva
humana). Adicionalmente, os implantes dentários estão muitas vezes sujeitos, nas zonas
de contacto com o osso ou nas zonas de contacto implante/porcelana, a micro-
movimentos (devido essencialmente às forças de mastigação) e a solicitações químicas
(ambiente da cavidade oral). Nestas condições, o implante está inserido num sistema de
tribocorrosão, que suscita um processo de degradação complexo podendo levar à falha
do implante. No entanto, o filme passivo que naturalmente cresce na superfície metálica
do implante dentário, pode ser danificado ou mesmo destruído durante a inserção do
implante, por abrasão do metal com o osso ou com outros materiais.
Neste trabalho, foram estudadas duas solicitações tribológicas diferentes. Os testes
de fretting e testes com movimento linear alternativo (pino-placa), ambos combinados
com estudos electroquímicos, foram executados em diferentes equipamentos. Estes
equipamentos foram especialmente adaptados para a execução de testes de
tribocorrosão. A solução electroquímica usada foi saliva artificial e como contra-corpo
foi seleccionada uma bola de alumina (φ = 10 mm). Com o objectivo de estudar a
influência da variação do pH no comportamento do material à tribocorrosão, foi
adicionado ácido cítrico à solução de saliva artificial. Adicionalmente, foram também
adicionados diferentes inibidores de corrosão à solução de saliva artificial, com o
objectivo de investigar a acção destes inibidores nas reacções anódicas e catódicas do
material. É importante referir que estes inibidores podem estar presentes nas
formulações de agentes de limpeza de dentes ou em medicamentos. Durante os testes de
fretting, para avaliar os mecanismos de degradação, foi usada a técnica de ruído
electroquímico. Através desta técnica é possível obter informação sobre a evolução do
Master Dissertation
XII
Master Dissertation XIII
potencial de corrosão e sobre a corrente de corrosão durante todo o teste de fretting. Nos
testes de pino-placa com movimento linear alternativo, foram impostas duas diferentes
condições electroquímicas: OCP e controlo potenciostático na região passiva da curva
de polarização (1000 mV) das amostras de titânio. Adicionalmente, com o objectivo de
obter informação mais detalhada sobre as características do filme passivo original assim
como do filme passivo formado após os testes tribocorrosão, foram efectuados testes de
Espectroscopia de Impedância Electroquímica (EIS), antes e depois do desgaste
mecânico. Em ambos os casos, as amostras foram caracterizadas pelas técnicas de SEM,
EDS e AFM.
Foram detectados fenómenos de despassivação e repassivação das superfícies do
titânio, e os mesmos são discutidos neste trabalho. Ficou demonstrado que o decréscimo
do pH assim como a presença de inibidores de corrosão, melhoram as propriedades do
Ti comercial. A evolução da repassivação do cp. Ti é afectada pelo decréscimo do pH.
No entanto, não há variações da cinética de rapassivação quando se adicionam
inibidores de corrosão à solução de saliva artificial.
Master Dissertation
XIV
Master Dissertation XV
TABLE OF CONTENTS
Abstract
Resumo
Aim and structure of the dissertation
Chapter 1
State of the art
1. Titanium as a biomaterial 2
1.1 What is a biomaterial? 3
1.2 Titanium in dental implants: some important properties 3
a) Biocompatibility and corrosion resistance 5
b) Wear resistance 7
2. Combined fretting and corrosion: Tribocorrosion phenomenon 8
2.1 Tribocorrosion phenomenon definition 8
2.2 Arrangements used in tribocorrosion 8
2.3 Parameters that affect tribocorrosion system 13
a) pH influence 14
b) Corrosion inhibitors’ presence 17
c) Third body particles 18
d) Surface roughness and material transfer 19
2.4 Fretting-corrosion action in Ti dental implants: research works 20
3. Repassivation of titanium passive films 21
3.1 Passive films’ properties 22
3.2 Passive film destruction and regeneration 22
References 33
Master Dissertation
XVI
Master Dissertation XVII
Chapter 2
Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in
artificial saliva
Abstract 39
1. Introduction 40
2. Experimental 42
3. Results and discussion 44
3.1 Tribological measurements 44
3.2 Electrochemical measurements 49
4. Conclusions 55
References 57
Chapter 3
Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion
conditions
Abstract 60
1. Introduction 61
2. Experimental 62
3. Results and discussion 65
3.1 Open-circuit potential (OCP) conditions 65
- Tribocorrosion behaviour 65
- Characterization of the passive film 67
3.1.1 Repassivation evolution with time analyses, in OCP conditions 71
3.2 Potentiostatic control conditions 73
- Tribocorrosion behaviour 73
-Characterization of the passive film 75
3.1.1 Repassivation analyses under potentiostatic control 78
4. Conclusions 80
References 81
Master Dissertation
XVIII
Master Dissertation XIX
Chapter 4
Results discussion
1. Wear and/or electrochemical mechanisms promoted by the combined action of
both
87
1.1 Electrochemical mechanisms 87
1.2 Mechanical mechanisms 88
2. pH decrease influence 89
3. Corrosion inhibitors influence 90
4. Repassivation evolution 91
References 92
Chapter 5
Final conclusions
94
Chapter 6
Suggestions for future work
96
Master Dissertation
XX
Master Dissertation XXI
Aim and structure of the dissertation
In this dissertation, two different main objectives were considered. One of the
objectives was to study the influence of pH and corrosion inhibitors on the
tribocorrosion of Ti in artificial saliva, under fretting conditions. Another aim was to
study the repassivation evolution of Ti in artificial saliva solutions, under tribocorrosion
conditions. Additionally, the effect of changes in pH solution and the presence of
corrosion inhibitors were considered. For the second objective, the tribological
configuration adapted was pin-on-plate.
The first chapter of this dissertation consists of an overview of the application of
titanium as biomaterial, more specially the Ti as dental implant material. Some
distinctively properties important in Ti as dental implant, like biocompatibility,
corrosion resistance and wear resistance are discussed. Combined tribological and
corrosion phenomenon - tribocorrosion – is defined and the arrangements used in
tribocorrosion are presented. A tribocorrosion system can be affected by several
parameters. The parameters that will be studied in this dissertation are discussed in this
chapter, that is, the pH influence and the corrosion inhibitors’ presence. Also, the
influence of the presence of third body particles, the surface roughness and material
transfer are considered and discussed. An over-view about the fretting-corrosion studies
in Ti dental implants is made. In addition, repassivation of titanium passive films, as
well as the passive films’ properties and the passive film destruction and regeneration
topics are considered and discussed.
Chapters 2 and 3 describe the experimental work in the form of papers. Chapter 2
describes the first part of the experimental work in the form of a paper already
published in an international journal: Wear, from Elsevier. This part of the work was
developed with my colleague Ana Ribeiro (second author of the paper). The influence
of the corrosion inhibitors in the fretting-corrosion behaviour of c.p. Ti samples was
Master Dissertation
XXII
Master Dissertation XXIII
studied by me and the influence of pH decrease was studied by Ana Ribeiro.
However, in this master dissertation, the complete paper will be presented and
discussed.
Chapter 3 describes the second aim of the work: repassivation study of c.p. Ti in
different artificial saliva solutions. This chapter is presented in paper format. The work
presented in Chapter 3 will be submitted to Corrosion Science, from Elsevier.
In Chapter 4 a compilation and comparison of the results obtained in the two papers
is made as well as a general discussion.
Finally, in Chapter 5 conclusions are presented.
Master Dissertation
XXIV
Chapter 1 – State of the art
Master Dissertation
2
State of the art
In all types of implants, the typical concern is the implant failure. Dental implants
are not an exception. The complications of dental implants that lead to failure are: bio-
corrosion, electrochemical galvanic coupling, fatigue, fixation failure, fracture, metal
allergy, wear, particulate formation, etc. These complications can be related to the
mechanical-biomechanical aspects or the chemical-biochemical aspects, or both [1]. In
the case of both complications occurring, a tribocorrosion system is created, which
promotes a complex synergism and a significant challenge in the research area.
Additionally, in such conditions, the passivated metallic surface is damaged or
destroyed. The comprehension of the cyclic destruction and repassivation of the passive
film can provide important information about the system.
1. Titanium as biomaterial
Commercially pure titanium (cp titanium) and Ti-6Al-4V are the most used titanium-
based implant biomaterials and the most applied in dental implants, essentially due to its
mechanical properties, good resistance to corrosion in biological fluids and low toxicity.
However, dental implants are subjected to wear degradation, promoted essentially by
the transmitted mastication loads, as well as chemical degradation due to the biological
environment (physiological environment or saliva). For all these reasons, it is important
to appreciate the corrosion and wear properties of Ti as well as the biocompatibility of
this material.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 3
1.1 What is a biomaterial
Biomaterial has been defined by several authors. Biomaterial can be defined as a
“nonviable material used in a medical device, intended to interact with biological
systems”, (Williams, 1987) [1].
The employment of metallic biomaterials has increased in medical applications. It is
possible to use metallic biomaterials as implants to restore lost functions or replace
organs such as bone plates, total joint replacement, dental implants, etc. Main metallic
biomaterials are stainless steels, cobalt based alloys, commercially pure titanium (cp
titanium) and titanium alloys [2].
1.2 Titanium in dental implants: some important properties
In the orthodontic field, dental implant can be defined as artificial tooth surgically
anchored to the jaw bone, i.e., a metal screw that is placed into a jaw bone and acts as
an anchor for a false tooth or a set of false teeth. There are different types of dental
implants but the most important are the endosseous implant and the subperiosteal. The
endosseous implant is implanted into edentulous mandibular or maxillary bone to work
as tooth replacements [2,3]. The subperiosteal implant is conventionally made and
designed to sit on top of the bone, but under the gums, transmitting all the solicitations
to the cortical bone [3]. In Fig. 1.1 a representative scheme of an endosseous dental
implant is presented, the one considered in this dissertation.
Fig. 1.1. Representative scheme of a dental implant [4].
Chapter 1 – State of the art
Master Dissertation
4
In the orthodontic field, cp titanium and titanium alloys (essentially Ti6Al4V) are
the most used metallic biomaterials due to a wide range of properties: low specific
weight (4.5 g/cm3 at 25 ºC), high temperature resistance, high toughness, high corrosion
resistance and excellent biocompatibility [5-10], occupying almost all of the market of
biomaterials.
When the dental implant is replaced into the bone, the periodontal ligament is
removed. So, during the metallic material implant selection, properties such as elasticity
module must be considered in order to allow uniform tensile distributions at
bone/implant interface. Thus, the elasticity module of the metallic material of the dental
implant should be similar to the elasticity modules of the bone. In table 1.1 some
mechanical properties of c.p. Ti, Ti alloys and bone are presented. As it is possible to
see, the elasticity modules of c.p. Ti and Ti alloys are higher than the elasticity modules
of the bone. Although, when compared to the other metallic materials normally used as
dental implant such as stainless steel (elasticity modules approximately 200 GPa) or Co-
Cr-Mo (elasticity modules approximately 240 GPa), cp Ti and Ti alloys present the
lower elasticity module [2 ].
Table 1.1. Mechanical properties of Cp Ti, Ti alloys and cortical bone [2].
Material
Tensile strength
(MPa)
Yield strength
(MPa)
Elongation
(%)
Elasticity
modules
(GPa)
Specific
weight
(g/cm3)
Cp Ti grade 1 240 170 24 102 4.5
Cp Ti grade 2 345 275 20 102 4.5
Cp Ti grade 3 450 380 18 102 4.5
Cp Ti grade 4 550 485 15 104 4.5
Ti-6Al-4V 930 860 10 113 4.4
Cortical Bone 140 -- 1 18 0.7
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 5
a) Biocompatibility and corrosion resistance
The European Society for Biomaterials, define biocompatibility as “the ability of a
material to perform with an appropriate host response in a specific application” [11]. It
encompasses all aspects of the interfacial reaction between a material and the tissues of
the body. The dependence of biocompatibility on several factors is presented in Fig. 1.2.
Fig. 1.2. Biocompatibility depending on a variety of system parameters [12].
In accordance with Fig. 1.2, biocompatibility involves physical, chemical, biological,
medical and design aspects. Biocompatibility is required because the biomaterials will
be in direct contact with the surrounding tissues. If the implant material is not
biocompatible with the body, toxic reactions can occur and consequently, infections or
inflammations will take place in the body [12].
Regarding c.p. Ti and Ti alloys, they present low toxicity, when compared with the
other biomaterials normally used in dental implants. For instance, dental implants
Chapter 1 – State of the art
Master Dissertation
6
produced with stainless steel are not so used as the cp. Ti or Ti alloys essentially due to
the high allergic effect of Ni [2].
Another important issue is the longevity of the biomaterial. In other words, the
biomaterial needs good corrosion resistance, because corrosion of metallic implants can
adversely affect the biocompatibility and mechanical integrity.
The biological environment is surprisingly harsh and can lead to rapid or gradual
breakdown of many materials, promoted by chemical corrosion. Titanium has high
corrosion resistance in a large range of aqueous solutions and over a wide range of
temperature without significant effects. The major exceptions are strong solutions of
some acids, mainly sulphuric, hydrochloric, phosphoric, oxalic, formic and also
solutions that contain fluoride ions [13]. Titanium has good corrosion resistance to
neutral solutions, especially those that contain the chloride ion which attacks a very
large number of metals. Additionally, titanium does not suffer significant degradation
during the sterilization process used in dental implants applications [2].
The titanium corrosion resistance is essentially promoted by the spontaneous
formation of a very protective oxide layer (normally, titanium dioxide film) on its
surface immediately after exposure to oxygen. The breakdown of the titanium oxide
layer is normally followed by a dissolution process, which has an adverse effect on the
corrosion resistance of the material. The passive film dissolution as well as the
corrosion process are two mechanisms for introducing additional ions in the body. Ions’
release of metallic biomaterials is a critical process because it can adversely affect the
biocompatibility and mechanical integrity of implants. Adverse biological reaction can
happen when extensive release of ions from metallic implant occurs which can lead to
mechanical failure of the device. [14].
Also, corrosion in a metallic dental implant can promote roughness on the surface
and weakening of the restoration releasing metallic elements [15]. B. Finet et al [16] in
a study about the titanium release from dental implant, observed some Ti release from
implants: titanium concentrations have been found both in periimplant tissues and in
parenchymal organs, such as lung, liver, spleen, etc. It was also found that Ti
concentration increases proportionally to the time of contact of the metallic implant
with the organs. Additionally, direct contact between connective tissue and metallic
material would appear to be necessary to observe contamination by titanium. M. Aziz-
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 7
Kerrzo et al [17] published some reports, which show the accumulation of titanium in
tissues adjacent to the implant and signifying metallic ions release and corrosion in vivo.
b) Wear resistance
When used as dental implants, titanium is subjected to cyclic micro-movements at
the implant/bone interface or implant/abutment interface, essentially due to the
transmitted mastication loads. This promotes mechanical wear. For this reason it is
important to understand and analyze the tribological properties of titanium, more
specifically the wear resistance [12,18].
The tribological proprieties of titanium have significant differences when this
material is compared to other metals, basically due to the interfacial activity between the
surfaces in contact (sliding surfaces) and the effect of any lubrication on this interaction.
When sliding onto itself, titanium gives a value of the coefficient of friction (≈ 0.47),
lower than other metals tested with similar sliding combinations. Additionally, the oxide
passive layer presented in titanium surface is not sufficiently mechanically stable and,
under load, is disrupted. Thus, the contact formed is metal-to-metal and the reactive
titanium could weld. This causes galling of the surface and the wear rate increases. It is
also important to point out that lubricants are usually ineffective with titanium [13].
J. Qu et al [19] study about the friction and wear properties of titanium alloys sliding
against metal, polymer and ceramic counterbodies (alumina). The authors used different
Ti alloys and the friction and wear tests were performed in a pin-on-disk apparatus. The
tribological conditions adapted were: 0.3 and 1.0 m/s as sliding speed, 10 N as normal
load and 500 m as sliding distance. Ti-6Al-4V alloy sliding against alumina produced
friction coefficient in the range of 0.34–0.50. Large frictional fluctuations occurred,
probably caused the by formation and periodic, localized fracture of a transfer layer.
Also, higher friction coefficient with larger fluctuation and higher wear rate were
observed at the lower sliding speed [19].
Chapter 1 – State of the art
Master Dissertation
8
2. Combined fretting and corrosion: Tribocorrosion phenomenon
As referred to before, dental implants are subjected to mechanical stresses resulting
from small relative displacements between prosthesis and surrounding bone tissues.
Additionally, the dental implants are, at the same time, subject to chemical solicitations
(oral environment) promoted by saliva or by physiological environments. Thus, the
material degradation is a consequence of mechanical stresses resulting in small relative
displacements (fretting ) which, in addition to the corrosive nature of the physiological
environment, leads to fretting–corrosion. This constitutes a tribocorrosion system [12,
18,20].
2.1 Tribocorrosion phenomenon definition
S. Mischler et al [21] defined tribocorrosion as a material degradation process which
results from simultaneous mechanical wear and chemical (or electrochemical) material
removal mechanisms. Is is important to point out that the two mechanisms of
degradation do not proceed separately, but depend on each other in a complex way:
corrosion is accelerated by wear and, similarly wear may be affected by corrosion
phenomena.
Many aspects related to tribocorrosion mechanism are not yet fully understood,
mostly due to the complexity of the chemical, electrochemical, physical and mechanical
processes involved in a tribocorrosion system. Additionally, according to P. Ponthiaux
[22], in practice field the tribocorrrosion occurrence is not yet recognized. Although the
tribocorrosion topic is a very interesting and important subject, it is not widely written
about.
2.2 Arrangements used in tribocorrosion
In accordance with D. Landolt et al [23], different types of contacts can be
considered in the tribocorrosion field: two body or three body contacts (see Fig. 1.3).
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 9
Also, the relative motion of the surfaces can be unidirectional (pin-on-disk test) or it can
be reciprocating (pin-on-plate test). In these cases the reciprocating is relatively big
(some millimetres). When the tribological contact involves a reciprocating motion of
small amplitude motion (few micrometers), a special type of configuration is
considered: fretting. Finally, it is also possible to considered particle impact or erosion
corrosion mode [23].
Fig. 1.3. Different types of tribological contacts involving simultaneous mechanical and
chemical effects (schematic) [23].
In all types of tribological contacts, in tribocorrosion systems, different types of
arrangements involving an antagonist rubbing against a flat plate (Fig. 1.4) can be use.
The counterbody (or antagonist) is normally a pin and it can be: cylindrical (I),
truncated cone (II), or a sphere (III). In the case the pin has a flat surface, the nominal
contact area is well defined. However, the alignment of the contacting surfaces is
extremely critical for the reproducibility of results. Spherical contacts are free of
alignment problems, but the nominal contact area is less well defined (it is calculated
from Hertzian contact mechanics) and it may vary during the experiment due to
formation of a wear scar [23].
Fig. 1.4. Type of counterbody pin: I) cylindrical pin, II) truncated cone, III) sphere [23].
Chapter 1 – State of the art
Master Dissertation
10
To constitute a tribocorrosion system, the sliding contact must contain an electrolyte
solution. The conventional pin-on-disk apparatus includes an oriented rotating pin
rubbing against a stationary disk or plate, immersed in the electrolyte (Fig.1.5 (a)).
Another type of arrangement is pin sliding in reciprocal (linear) motion on a stationary
plate immersed in the electrolyte. This is pin-on-plate arrangement (Fig. 1.5 (b)). The
equipment presented in Fig.1.5 (c) uses a rotating inverted disk rubbing with a
stationary pin that is in contact with the electrolyte. Finally, in the arrangement
presented in Fig. 1.5 (d), a ceramic microtube surrounded by a second tube, containing
the electrolyte, rotates on a stationary metal plate. In this case, the tribocorrosion
behaviour can be measured locally on very small surfaces, at the expense of a less
precise mechanical control. It is suggested by D. Landolt [23] that similar values for the
friction coefficient were obtained using different experimental arrangements. However,
significant quantitative differences were observed in the total wear volume and the
measured current densities. Also, the tribocorrosion phenomena promoted in each
arrangement, are influenced by the position of the plate and the pin [23].
Fig. 1.5. Experimental arrangements used in tribocorrosion studies: a) pin-on-disk, with rotating
pin; b) pin-on-plate, with reciprocating sliding motion of pin; c) pin-on-disk with stationary pin;
rotating ceramic microtube serving as electrolyte conduit [23].
In experimental tests, two different tribocorrosion experimental arrangements were
selected based on the practical application and/or in the objective of the study. To study
the first main goal, tribocorrosion effect in Ti dental implants, fretting-corrosion tests
were promoted. In order to simulate the wear promoted by cyclic micro-movements at
the implant/bone interface or implant/abutment interface, fretting arrangement was
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 11
used. In accordance with the environment where the dental implant suffers the
mechanical damage (oral cavity), artificial saliva was employed as electrochemical
solution. Finally, corundum (alumina) sphere was used as counterbody due to it
chemical inertness, high wear resistance and electrical insulating properties.
In the second experimental work, the electrochemical solution and the counterbody
were the same selected in the first experimental work. The second objective of this
project was to study the repassivation of the dental implant after insertion and
implantation into hard tissue, simulating wear due to the abrasion of the implant with
bone or with other materials. In this case, the wear movements promoted normally
represent some millimetres, the reason why reciprocating sliding arrangement was used.
In relation to the electrochemical experiments used to study the tribocorrosion
mechanisms, P. Ponthiaux et al [22], described some electrochemical techniques
capable of studying tribocorrosion mechanisms: open circuit potential measurements
(OCP), the potentiodynamic polarization measurements, and the electrochemical
impedance measurements (EIS). They present the capabilities and limitations of these
techniques based on a tribocorrosion study of an AISI 316 stainless steel and an iron–
nickel alloy immersed in aerated 0.5M sulphuric acid sliding against a corundum
counterbody. After the results analyses, the authors concluded that the electrochemical
techniques can provide essential information on the tribocorrosion mechanism. They
also concluded that the majority of tribocorrosion mechanisms are related to
electrochemical reactions. Related to the kinetics processes, electrochemical techniques
can provide information about the corrosion rate, rate of depassivation by mechanical
action in the contact area, and rate of passive film restoration. Also, measurements made
with electrochemical techniques can offer information on the tribological conditions of
friction and wear mechanisms. This information can be obtained because under sliding
friction, the wear process, is highly dependent on the electrochemical state of the
surfaces (mild oxidation, or abrasive wear, etc.).
Recently, W. Pei-Qiang et al [24-26] suggested a new electrochemical technique to
study tribocorrosion behaviour of the materials. They applied the electrochemical noise
technique (ENT) to fretting-corrosion studies [26] and to corrosion-wear of stainless
steel in sliding contacts [25]. They defined electrochemical noise technique as the
spontaneous analysis of fluctuations of potential and current at electrodes and suggested
Chapter 1 – State of the art
Master Dissertation
12
three major ways for measuring potential and current noise in a corrosion system: two
identical working electrode (WE), one working electrode coupled to a microelectrode
(e.g. Pt), and two identical working electrodes with a bias potential. Additionally, the
authors noted that there are numerous differences between the ENT in a corrosion
system and in a corrosion–wear system. One of the major differences is that the
mechanical interaction in a corrosion–wear test may induce ENT that is not appearing in
a corrosion test [25].
The same research team recently presented a work where they used the
electrochemical noise technique to monitor the corrosion-wear of TiN coated AISI 316
stainless steel [27]. The experimental set-up used for performing the electrochemical
noise technique during the fretting test, in immersed samples, is presented in Fig. 1.6.
Fig. 1.6. Schematic experimental set-up used for electrochemical noise measurements [27].
A typical graph correlating the corrosion potential and the corrosion current density
with the time, obtained with ENT, is shown in Fig. 1.7. It is possible to see that, when
the fretting test starts, the corrosion potential values decrease and the corrosion current
increases. Under steady-state condition, a very low current is recorded, suggesting that
the material is very corrosion-wear resistant. At the end of the fretting test, the potential
and current of the working electrode are restored to their original values before the
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 13
fretting test, indicating repassivation takes place on the worn surface of the tested
sample [27].
Fig. 1.7. Electrochemical noise measurements on TiN coated AISI 316 stainless steel sliding
against corundum in 0.5 M H2SO4. Test conditions: 5N as normal load, 10 Hz of frequency,
amplituede of 200 µm and 20000 fretting cycles [27].
The authors concluded that the electrochemical noise technique (ENT) has important
advantages in the on-line corrosion-wear and suggest the application of the technique as
a promising on-line monitoring tool in detecting the delamination of a coating [27].
2.3 Parameters that affect tribocorrosion system
As referred to above, a tribocorrosion system can be affected by the electrochemical
and the mechanical parameters. However, there are several other factors that affect a
tribocorrosion system. Fig. 1.8 illustrates the most important factors. Analysing Fig. 1.8,
the performance of electrochemically controlled tribocorrosion systems is conditioned
by the mechanical solicitations which are related to equipment design and operation, the
electrochemical conditions prevailing at the rubbing metal surfaces, the solution
properties in the contact and the materials and surface properties of the sample and the
counterbody. Usually the influence of these parameters on the tribocorrosion behaviour
Chapter 1 – State of the art
Master Dissertation
14
is mutually dependent and they do not act independently. This confirms the importance
of the use of very well defined mechanical and electrochemical variables in a
tribocorrosion experiment [23].
Fig. 1.8. Parameters that affect the tribocorrosion system of a sliding contact, under
electrochemical control [23].
There are some factors that deserve special attention, in accordance with the main
goal of this dissertation. These factors are: pH influence and presence of corrosion
inhibitors in the electrochemical solution, third body particles, surface roughness and
material transfer [23]. The two first factors are directly related with the purpose of this
project: influence of the pH and corrosion inhibitors in the tribocorrosion phenomenon.
a) pH influence
Normally, the pH of normal blood and interstitial fluid is about 7.35–7.45, but can
decrease to 5.2 in the hard tissue due to implantation of the dental implant (or other
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 15
implant). In 2 weeks, the pH value recovers to 7.4. Toxicity and allergy can occur in
vivo, when a decrease is pH is marked, if metallic materials are corroded by body fluid.
In such cases, metallic ions are released into the fluid for a long time, and they will
combine with bio-molecules (proteins and enzymes) promoting toxicity and allergy
[28,29]. The Pourbaix diagram to Ti is presented in Fig. 1.9. The hatch marked regions
indicate conditions of human internal environment. Ti presents a passive region
between the marked lines a and b.
Fig. 1.9. Pourbaix diagram for titanium (hatch marked regions indicate conditions of human
internal environment) [29].
A.M. Al-Mayouf et al [8] studied the effect of pH value on the corrosion behaviour
of Ti–30Cu–10Ag (wt.%) alloy, which is a new titanium alloy used for dental implants.
This alloy, cp Ti and Ti–6Al–4V (for comparison), were examined by electrochemical
methods in artificial saliva solutions. The different pH values studied were 7.2 and 3.
Also, different NaF concentrations were considered. Only the solutions with similar
NaF concentrations and different pH will be analysed in accordance with the aim of the
Chapter 1 – State of the art
Master Dissertation
16
dissertation. The results obtained for this experimental condition are presented in table
2.2.
Table 2.2. OCP values (mV vs. SCE), in artificial saliva, for cp titanium, at different pH values,
in presence of different concentrations of NaF, after 17 h of immersion [8].
pH [NaF] (M) Corrosion potential (V vs. SCE)
0.0 -0.331 7.2
0.01 -0.262
0.0 -0.145 3.0
0.01 -0.952
Some remarks can be made: comparing the potential values obtained with artificial
saliva in absence of NaF, with different pH, it can be suggested that a decrease in pH,
diminishes the tendency of cp Ti to corrosion. Nevertheless, when 0.01M NaF is added
to artificial saliva solution, the increases in the corrosion tendency in the solution with
pH = 3, is very noteworthy. Unfortunately, the reasons associated with these phenomena
are not discussed by these authors [8].
Also, Y. Fovet et al [30] studied the influence of pH and fluoride concentration on
the titanium passivating layer, namely the stability of titanium dioxide. The authors
proceeded to a thermodynamic investigation at 25 °C in the range of concentrations and
pH corresponding to the solutions commonly used. Accordingly, such study gives a
possibility to specify which species could be involved in the corrosion reaction to
calculate the corresponding potentials, and lastly to define the domain of insolubility of
TiO2. Also, the author proposed a model which they believe may be used to anticipate
the behaviour of titanium in vivo. The graph presented in Fig. 1.10 was obtained in an
inorganic composition close to human mixed saliva developed by the authors (SAGF
medium). The model/graph permitted to evaluate, for a given pH in any preparation, the
fluoride content limit, whatever the chemical composition of the medium is. The
domains in the figure show the conditions necessary to prevent an alloy in mouth from
corrosion, considering that TiO2 promotes the corrosion resistance of Ti. For instance, it
can be analysed that solutions in the absence of fluorides, promote stability of TiO2 only
to pH higher then 6.5. Additionally, the authors related that according to the pH,
Ti(OH)22+, Ti(OH)3
+, Ti(OH)4 are the species, which can exist when the solution is free
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 17
of fluoride and Ti(OH)2F+, TiF4, HTiF6 can be formed in a fluoride containing solution.
Finally, it can be suggested in relation to stability of TiO2, that the pH tends to decrease
with the increase of fluoride concentration, achieving a steady-state value, of pH = 1, to
values of fluoride concentration higher than 3.4.
Fig. 1.10. Critical pH curve, for the TiO2 layer in SAGF solution vs. fluoride concentration
[30].
b) Corrosion inhibitors’ presence
All metals, except the noble ones, oxidize spontaneously in an aggressive atmosphere
such as air or water. Also, some acid environments, like H2S or HCl, can be considered
aggressive and can promote oxidation on a metal surface. In most of the cases the
thermodynamic laws would lead to a complete oxidation of the metal. Due to the drastic
effects of corrosion, preventive measures must be taken and the use of corrosion
inhibitors is one of the possibilities [31]. Thus, corrosion inhibitors are chemical
compounds that react, when added in small concentrations, with a metallic surface (or
the environment), in order to stop or reduce the corrosion process. There are anodic,
cathodic and organic inhibitors. An anodic inhibitor retards the anodic reactions and,
similarly, cathodic inhibitors act by retarding the corrosion, inhibiting the reduction of
water to hydrogen gas. As every oxidation requires a reduction to occur at the same time
Chapter 1 – State of the art
Master Dissertation
18
it slows the oxidation of the metal. An inhibitor which acts both as a cathodic and
anodic manner is an organic inhibitor [32].
Although the influence of corrosion inhibitors in dental implants or in Ti used as
biomaterial is not a topic very considered in the literature, it is possible to cite some
works about corrosion inhibition of Ti. It is indicated that strongly oxidizing inorganic
compounds such as HNO3, K2Cr2O7, KMnO4, KIO3, Na2MoO4, NaClO3, Cl2 and H2O2
are able to passivate titanium in 1% H2SO4 and 3% HCl. They are present at a critical
concentration above 10 mM in the corrosive medium [c.f.33]. Other authors stated that
K2Cr2O7, HNO3, TiCl4 are the most powerful corrosion inhibitors at room temperature.
However, in concentrated sulphuric acid solution, corrosion inhibition is observed only
by HNO3 for a very short period [34]. Also, rapid passivation of titanium surface by low
concentrations of MoO2-4 ions in 0.5 M H2SO4 is referred in some works [c.f.33].
F. Mansfeld et al [32], studied the use of organic inhibitors for titanium. The authors
concluded in this work that organic compounds, containing nitro-groups, can reduce
corrosion rates of Ti-6Al-4V in HCl solutions. However, a critical concentration can not
be exceeded. Also, below that concentration level, the additives increase corrosion rates.
c) Third body particles
Relating to the third body particles, this concept was introduced to describe the
velocity accommodation in tribological contacts due to the presence of a fluid or
particles or both. The presence of third body particles in the contact zone are very
critical because they can act as an abrasive which accelerates wear or as a solid
lubricant diminishing friction and wear, depending on their physical properties and on
their number. The quantity of particles present in the tribological contact depends on
their rate of formation due to rubbing and on their rate of ejection from the contact. This
number of third body particles depends on the design of the experimental apparatus
including the geometry of the contact, the presence or not of vibrations (mechanical
stability of the apparatus) and the type of motion (continuous or reciprocating). Also,
the variation of the electrochemical conditions can have a marked influence on the
formation rate and the properties of wear debris.
The mechanical, chemical degradation or both, of metallic materials used in
prosthetic implants, can cause tissues inflammation. Under such conditions, micro and
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 19
nanometer size wear particles are formed. Additionally metallic ions may be released
into the body (see page 6). The wear particles can affect tissue inflammation which will
affect bone regeneration and the stability of the implant. [20].
D. Landolt et al [23], investigated in the third body effects and material fluxes in
tribocorrosion systems involving a sliding contact. All the results of this work were
based on a consideration of material fluxes and take into account the formation and
ejection of third body particles. They considered as first and second bodies, the sample
and counterbody. The third body was defined by wear debris. These authors concluded
that synergistic effects in tribocorrosion are the result of mechanical and
electrochemical mechanisms that control the formation, properties and residence time of
third body particles (or debris).
d) Surface roughness and material transference
In relation to surface roughness, this topic has an important role in tribocorrosion
systems. The surface roughness affects the rate of depassivation and the mechanical
wear. D. Landolt et al [23] observed that when a hard body rubs against a softer body,
the surface roughness of the softer body will adapt rapidly to the rubbing conditions but
not that of the hard body. So, the normal situation is a hard body with a rough surface
promoting abrasive wear in the softer material, where the passive film is locally
removed and the repassivation current is relatively high. When the harder body is well-
polished, abrasion of the soft material is lower and film thinning rather than film
removal occurs. However, plastic deformation can occur which can contribute to
passive film thinning or rupture. Generally, when a smooth antagonist rubs on a ductile
metal some lower re-oxidation current could be expected [23].
Another important factor that can influence the tribocorrosion system is the material
transfer. In tribological contacts one often observes material transfer from the softer to
the harder body, leading to an alteration of the friction and wear conditions in the
contact. The extent of material transfer depends on the prevailing electrochemical
conditions among others. As a consequence, electrochemical conditions can affect
mechanical wear rate [23].
Chapter 1 – State of the art
Master Dissertation
20
2.4. Fretting-corrosion action in Ti dental implants – research works
In literature, fretting-corrosion in dental implants is not a very considered issue.
However, some works related with fretting-corrosion in biomaterials can be mentioned.
S. Barril et al [35] studied the influence of the fretting regimes on the tribocorrosion
behaviour of Ti-6Al-4V in 0.9 wt% sodium chloride solution. In this tribocorrosion
research work, it was observed that the electrochemical response of the alloy to
tribocorrosion is critically affected by the prevailing fretting regime.
L. Duisabeau et al [36], studied the environmental effect on fretting of metallic
materials for orthopaedic implants. Ti–6Al–4V alloy and an austenitic stainless steel
(AISI 316L SS) were the selected and studied biomaterials. Tests were performed both
in air and in an artificial physiologic medium in order to reveal the damage induced by
the physiological medium. The result obtained in this work shows that fretting regime,
accumulated dissipated energy and corrosion activated at the interface are
interdependent. The author shows that the introduction of solution reduces the
interaction between the two surfaces, i.e., reduces the friction. So, it was demonstrated
that the presence of a solution alters the results, as well as the nature of the specimen,
the normal load applied and the electrochemical potential.
The tribological behaviour of some important bio-metallic alloys, namely cp-Ti, Ti–
6Al–4V, Ti–5Al–2.5Fe, Ti–13Nb–13Zr, and Co–28Cr–6Mo under fretting contacts, in
simulated body fluid conditions, was studied by A. Choubey et al [37]. Bearing steel
was used as antagonist part, with 10 N as normal load, 10000 as fretting cycles, a
relative displacement stroke of 80 µm and a frequency of 10 Hz. The electrochemical
solution was a Hank’s balanced salt solution in order to assess the performance of the
materials in simulated body fluid (physiological) solution. The results show that in
tribological conditions immersed in Hank’s solution, a steady state in coefficient of
friction is achieved in Ti alloys (around 0.46–0.50) with the exception of Ti–5Al–2.5Fe
alloy. It was also suggested that the major wear mechanism of the Ti alloys is
tribomechanical abrasion. Transfer of material was observed as well as cracking
phenomenon, which can be related to the extensive plastic deformation under
tribomechanical stress conditions and the formation of persistent slip bands.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 21
3. Repassivation of titanium passive films
In tribological contacts operating in corrosive environments, material removal takes
place simultaneously by mechanical wear and by corrosion [38]. In these systems, the
surface passivation or degradation has an essential influence on the performance and
lifetime of the metal. Interest in the study of the tribocorrosion of passive alloys is
mainly due to the particular chemical and mechanical behavior of these metals once the
oxide protective film is removed [39]. Also, the comprehension of the repassivation
evolution of the surface after the mechanical damage or destruction of the passive film,
can provide important information about the system.
3.1 Passive films’ properties
As already stated, the high corrosion resistance of titanium is due to the spontaneous
formation of a very protective oxide layer (normally titanium dioxide film) on its
surface, immediately after exposure to oxygen or water [8,14,17,40]. The titanium
dioxide passive film normally consists of an amorphous or poorly crystallized and
nonstoichiometric TiO2. This dioxide film voluntarily regenerates even if destroyed.
Normally consists of a very thin layer (about 10 nm) and has a high density of defects.
The nature of this porous layer was found to depend on the nature of the alloy and the
solution anion species [17,28,41,42].
A.K. Shukla et al [42], studied the properties of passive films formed on cp Ti, Ti-
6Al-4V and Ti-13,4Al-29Nb alloys in simulated human body conditions (Hank’s
solution), as a function of immersion time. In this work it was shown that all the alloys
spontaneously passivate on Hank’s solution. Also, when anodic passivation is made to
promote the passive film growth, a titanium passive film with the following
characteristics was obtained: at low potentials, highly disordered titanium dioxide
(TiO2) with some titanium trioxide (Ti2O3) was identified; at voltages up to about 20 V
the surface film has an amorphous structure; from 20 V to about 50 V the film consists
of a mixture of anatase and a quasi-amorphous titanium dioxide (TiO2); from 50 V to 80
V the film consists of small crystals of anatase (titanium dioxide TiO2); above 80 V
Chapter 1 – State of the art
Master Dissertation
22
large crystals were detected. It was also concluded that the passive film breakdown in
Ti–13.4Al–29Nb alloy was related to the greater Al content.
In accordance with the previous work, Y.Z. Huang et al [43] characterized the
titanium oxide film grown in 0.9 % NaCl. The authors indicated that in most aqueous
environments, the oxide typically found in titanium surfaces is TiO2. However it may
consist of mixtures of other titanium oxides including TiO, Ti2O3 and TiO2.
Also, the contact between the metallic implant and the receiving living tissues, in Ti
metallic implant, is made through the oxide layer on the implant surface, which allows
the osseointegration process. The chemical properties of the oxide layer play an
important role in the biocompatibility of titanium implants and the surrounding tissues
and must not break down if the implant is to be successful [6,16,17,40]. However,
C.E.B. Marino et al [14] study the dissolution process of Ti ions of a Ti-dental implant
in artificial saliva and show that oxide layer breakdown after some time of immersion
decreases the corrosion resistance. This process was followed by a dissolution process,
which has a detrimental effect on the corrosion resistance and consequently on the
osseointegration process.
3.2 Passive film destruction and regeneration
The metallic surface with the passive film may be scratched or destroyed during
insertion and implantation into hard tissues, by abrasion with bone or with other tissues
[44]. This mechanical damage, which causes local thinning or removal of the passive
film, promotes a drastic increase of the corrosion rate [28]. Fig. 1.11 presents a
representative scheme of the implant insertion procedure where the abrasion of the
metallic implant with the bone is possible to observe.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 23
Fig. 1.11. Dental implant insertion procedure [4].
In tribocorrosion conditions, after the mechanical damage of the metal surface and
the consequent damage of the passive film, the depassivated surface area re-oxidizes –
repassivation. This process implies a loss of electrons of the reacting metal atoms, that
is, the charge transfer reaction at the interface which yields either dissolved metal ions
or a solid oxide [39,45].
Repassivation can be defined as a nucleation process and growth of a new passive
film on the bare metal surface [39,44,46]. Additionally, the proposed mechanisms for
formation of a passive film can be applied to follow repassivation evolution. Formation
of passive film on metallic surfaces is usually described by the models based on the
pioneering works of Sato and Cohen [c.f 46] and Cabrera and Mott [c.f. 46]. The model
proposed by Cabrera and Mott [c.f.46] consists in High-field ion conduction model, in
which the passive film grows by the transport of ions from the metallic surface across
the film under high electric fields. Depending on which is the slowest step, the rate of
film thickening may be controlled by the movement of: 1) ions across the metallic
surface-film interface; 2) through the film; and 3) across the film-solution interface.
This model can be considered to passive film growth in an aqueous solution.
Nevertheless, according to the Place exchange model proposed Sato and Cohen
[c.f.46], a layer of oxygen is adsorbed onto the surface and then exchanges places
Chapter 1 – State of the art
Master Dissertation
24
(possibly by rotation) with the underlying metallic atoms. A second layer of oxygen is
adsorbed and the two M-O pairs rotate simultaneously. This process is repeated and the
result is the formation of an oxide film [46]. The new surface has different
electrochemical properties from the steady state exterior surface of the metals. Thus,
when mechanical damage is promoted in the metallic surface, the metal will suffer three
processes [39]:
- Reestablishment of the electrical double layer;
- Ion dissolution through the worn area;
- Formation of an oxide film (repassivation).
Additionally, F. Assi et al [39] and P. Jemmely et al [45] considered two distinct
models for describing repassivation of an activated surface. The first model, Surface
coverage model or Lateral Growth (LG), assumes that the passive film on the metallic
surface is removed entirely and the metal oxidation occurs exclusively on the new fresh
metallic surface leading to lateral growth of an oxide. The film is considered completely
removed when the entire oxide film is removed from the surface and the material has lost
its protection due to the aggressive solution which is directly in contact with the fresh
metallic surface (Fig. 1.12). However, the passive film can be considered partially
removed when only a small portion is removed and the metallic surface is still protected
by an oxide layer [39,45]. Both models were based on the assumption that only solid
oxides but no dissolved ions are formed in the anodic reaction. Also, only the non oxide
surface reacts and follows Tafel reaction kinetics; finally, the oxide grows only
laterally.
a) b)
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 25
Fig. 1.12. Abrasion wear mechanism on an oxide film protected alloy: a) with a
complete film removal; 1) Indenter; 2) Reformed passive film; 3) Passive film; 4) Metal. b) with
a partial removal [39].
In relation to the second model presented by F. Assi et al [39] and P. Jemmely et al
[45] - Film growth model or uniform growth (UG) – and in accordance with the authors,
it can be assumed that an anodic oxide grows uniformly on the surface and the growth
rate is determined by high field conduction. They do not assume that this model was
physically realistic in the very first moments of repassivation, but it provides useful
insight on the behaviour at later stages. So, in this model, it is supposed that no
dissolution occurs; the entire charge, Q, is used for the uniform film formation on the
surface; the film formation is limited by ion transport within the oxide, induced by the
high electrical field (some MVcm-1).
Others authors proposed other film growth models considering different points. For
instance, O.A. Olsson et al [47], modelled the current response from a reciprocal motion
pin-on-disc experiment assuming a rate limiting reaction at the film interface. Two
different geometries were considered: inert counterbody on metallic substrate and
metallic counterbody on inert substrate. In this master dissertation, an inert counterbody
on a metallic substrate was the used configuration. When inert counterbody is
considered, it was assumed that the counterbody moves along the wear scar and
instantly removes a certain fraction or all of the film. This gives rise to an instant
current increase followed by a repassivation process at that point. In the metallic
counterbody on inert substrate case, the corrosion current originates from the
counterbody, i.e. a moving current source. Further, the entire current will come from a
crevice-like geometry, which increases the significance of the ohmic drop.
It is assumed that the film is gradually removed during rubbing, until a balance is
reached between film removal, governed by the mechanical parameters, and film re-
growth, governed by the repassivation kinetics. So, the model presented by these
authors is based on the hypothesis of film growth being limited by an interface reaction.
When the pin passes a certain point, it will remove a part or all of the film.
A commonly used and simple qualitative relation describing the repassivation
behaviour is schematic shown in Fig. 1.13.
Chapter 1 – State of the art
Master Dissertation
26
Fig. 1.13. Current measurement during a film rupture event and the associated oxide film
thickness (example curve from experimental data) [48].
Directly related with the figure presented, is the following empirical equation:
)exp(*0ii bb
ktiikti !="=!
(Eq. 1)
where i is the film repassivation current density at the film rupture site, k is a constant
related to the exchange current density of a fresh metal surface in the environment (i0)
and the time required before a fresh metal surface starts to repassivate (t0), t is time of
repassivation (beginning from creation of the fresh surface), and bi (>0) is the
repassivation kinetic exponent [48-53].
If the transient current generated were plotted in a log–log diagram, it was found that
the presented equation was followed. Thus, a plot of the logarithmic of current density
versus the logarithmic of time follows a relation of the type:
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 27
ktbtiktii
bi +=!="
)log()(log (Eq. 2)
where k and bi are constants, but n is one repassivation parameter that indicates how fast
the repassivation occurs: large n values indicate rapid repassivation [48, 52].
Some research works presented analyses based in the repassivation models presented
before. For instance, E. Cho et al [46] analyzed, quantitatively, the repassivation
kinetics of ferritic stainless steels based on the high-field ion conduction model.
Although the materials studied by this author don’t have significance to this
dissertation, the results treatments can provide important information of the scope of the
dissertation. So, the authors studied the effects of the alloying elements (Cr, Mo, W, and
Ni) on the repassivation kinetics of ferritic stainless steels in deaerated MgCl2 solution
at 50 °C, using a rapid scratching electrode technique. Typical current transient curve
when a scratch was made on the surface of the alloy with the passive film (polarized to
a passive potential) was obtained and it is presented in Fig. 1.14.
Fig. 1.14. Current transient curve of Fe–18, 20, 25 and 29 Cr alloys in deaerated MgCl2 solution
(1 M, 50 °C) [46].
Chapter 1 – State of the art
Master Dissertation
28
Once the passive film is broken by the scratch, the anodic current flowing from the
scratch increases abruptly to a peak due to an anodic oxidation reaction, and thereafter
decreases as the repassivation proceeds. Also, the authors plot log i(t) versus log (t) and
demonstrate that the repassivation of the material being studied occurs in consecutive
processes with different kinetics: passive film initially nucleated and grew for about 8–
12 ms according to the place exchange model, and thereafter grew according to the
high-field ion conduction model. The transition of the film growth mechanism from the
place exchange model to the high-field ion conduction model appears to occur because
the activation energy for the place exchange process of M-O pairs increased as the film
thickened, and then reached a value beyond which the place exchange process of M-O
pairs cannot occur.
Also, repassivation rate of an alloy can be analyzed using the repassivation time
needed to achieve a pre-determined degree of repassivation from scratching. The shorter
the repassivation time, the faster will be the repassivation rate of an alloy. The decay of
current density on the scratched surface follows the empirical law, i = kt-bi, presented
before, is shown in Fig. 1.15. The decay gradient determined from the slopes of the log
i(t) versus log (t) was less than 1 in accordance with the high-field ion conduction
model.
Fig. 1.15. log i(t) vs. log (t) plots of Fe–18, 20, 25 and 29 Cr alloys in deaerated MgCl2 solution
(1 M, 50 °C) [46].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 29
M.F. Song et al [49], proposed a model to obtain the repassivation kinetic exponent
of the empirical low (i = kt-bi), that is, the parameter bi. The authors presented some
theoretical predictions for the range of repassivation kinetic exponent bi. They stated
that, for a passive alloy bi will be:
• bi < 0, i increases with time, which indicates active corrosion. This is
contrary to the properties of a passive alloy.
• bi > 1, the ‘‘anodic’’ charge density at a film-ruptured site (q) is
negative. Further, as t increases, the value of q decreases with time,
indicating thinning, with time, of a repassivating film at a film-ruptured
site.
Hence, the theoretical range of bi for a passive alloy is 0 < bi < 1 if the repassivation
kinetics follows the empirical low i = βt-n. Some experimental work was then developed
and the model yields film repassivation kinetic was in reasonable agreement with the
results reported in the literature [49].
In relation to the repassivation evolution based on potential corrosion results, some
works can be cited. J.M.Abd El Kader et al [51], suggested a relation between the
repassivation and the corrosion potential. This author studied the oxide film thickening
on titanium in aqueous solution and presented experimental results obtained with
corrosion potential for study of the repassivation process. He based the study in the
empirical law normally used, i = kt-bi. However, it was suggested that, under open-
circuit conditions, it is reasonable to relate this empirical law with a new one involving
corrosion potential (E) values. E-log(t) curves were suggested and, in general, two
linear segments are obtained: at lower t values (lower times of immersion), the rate of
potential build up is low and is very dependent on the nature and concentration of the
anion. After a certain time, the rate of the potential change increases appreciably.
T. Hanawa et al [44] studied the repassivation of titanium and surface oxide film
regenerated in simulated bioliquid and also suggested the repassivation evolution
analyses based on corrosion potential results. This author referred to the fact that when
abrasion is interrupted, the open-circuit potential values increases, according to the
logarithmic law ∆E = bE logt + k, indicating a thin protective film formation. In the
equation presented, t (s) is the time after the interrupted abrasion and bE and k are
Chapter 1 – State of the art
Master Dissertation
30
constants determined by the kind of solution in which the specimens are abraded. The
repassivation was calculated by fitting the logarithmic plots and further determining the
constant bE and k, 3, 6 and 300 s after the passive film was removed. Large bE values
are related to the large increases of potential at the initial stage of repassivation. In
contrast, the increase of potential is larger at the later stage when bE is large. Logically,
when both bE and k are large, the repassivation will be rapid. The authors concluded that
the repassivation rate of titanium in bio-fluid solutions (Hank’s solution) is slower than
that in saline solutions. Also, the authors suggested that the repassivation of titanium in
biological systems is slower than predict when the surface oxide film is destroyed,
possibly inducing the dissolution of more titanium ions into bio-fluid.
F. Contu et al [53,54] studied the corrosion potentials evolution during mechanical
abrasion and the repassivation rate of the metallic surface. The repassivation rate
analyses of cp Ti and Ti-6Al-4V in inorganic buffer solutions and bovine serum was the
aim of this study [53]. To carry out the repassivation experiments, a tribo-
electrochemical micro-cell apparatus was used. The results obtained with OCP
measuremnents are presented in Fig. 1.16. The variation of the OCP before, during and
after rubbing of cp Ti in inorganic buffer solutions at pH ranging from 2.0 to 12.0 can
be observed. The authors suggested that when the rubbing-tip touches the metallic
surface the potential abruptly decreases due to a creation of a bare metallic electrode
surface and its exposition to the electrolyte. However, when the abrasion action stops,
the potential increases again.
Fig. 1.16. Variation of the OCP before, during and after mechanical disruption of the passive
film of cp Ti in buffer solutions at different pH values [53].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 31
The authors noted that the repassivation rate of the materials was determined
calculating the tangential rate at the initial repassivation curve as well as the percentage
OCP ennoblement 30 s after the rubbing process was stopped. The repassivation rates
determined by the authors are presented in Fig. 1.17. The authors suggested that, when
immersed in the inorganic buffer solutions, the repassivation rate of cp Ti and Ti6Al4V
is affected by the pH 7.0, due to the formation of TiO instead of Ti2+ (aq) as suggested
by the Pourbaix diagram of titanium [53] (see Fig. 1.9, page 15).
Fig. 1.17. Percentage OCP variation of cp Ti immersed in inorganic buffer solutions and bovine
serum at pH 4.0 and 7.0 [53].
Recently G.T. Burstein presented some works related with aluminium [55,56], where
he described the evolution of the corrosion potential. Once again, the material does not
present interest for the dissertation, however, the results obtained by the authors have
significance to the aim of the dissertation. Empirical observation suggested that the free
corrosion potential of freshly generated metallic surfaces increases linearly with log (t)
during repassivation by oxide film growth following E-log(t) [55]. OCP results are
shown in Fig. 1.18 [56]. The initial plunge in corrosion potential in each case is due to
the depassivation of the metallic surface, after which the corrosion potential increases.
The reaction rate must be fastest when the metal surface is freshly bared, and it decays
Chapter 1 – State of the art
Master Dissertation
32
with time as repassivation occurs; the increasing potential with time is representative of
this.
Fig. 1.18. The OCP of aluminium surfaces freshly generated in situ by guillotining in different
electrolytes [56].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 33
References
[1] B.D. Ratner, A.S. Hoffman, F.J. Schoen, J.E. Lemons; Biomaterials Science: An
introduction to materials in medicine; Pressed by Academic Press
[2] K.J. Anusavice; “Philips, Materiais Dentários”; Elsevier, 11ª Edição.
[3] J.B. Brunski, A.F. Moccio, Jr.S.R. Polllack, E. Korostoff and D. Trachtenberg;
Investigation of surfaces of retrieved endesseous implant of commercially pure
titanium; Titanium alloys is surgical implants; ASTM Special Technical Publications
796
[4] www.thedentalimplantcenter.com
[5] E. Eisenbarth, D. Velten, M. Muller, R. Thull, J. Breme; Biocompatibility of β-
stabilizing elements of titanium alloys; Biomaterials 25 (2004) 5705–571
[6] C. Aparicio, F.J. Gil, C. Fonseca, M. Barbosa, J.A. Planell; Corrosion behaviour of
commercially pure titanium shot blasted with different materials and sizes of shot
particles for dental implant applications; Biomaterials 24 (2003) 263–273
[7] Y. Zhou, M. Niinomi, T. Akahori, H. Fukui, H. Toda; Corrosion resistance and
biocompatibility of Ti–Ta alloys for biomedical applications; Materials Science and
Engineering A 398 (2005) 28–36
[8] A.M. Al-Mayouf, A.A. Al-Swayih, N.A. Al-Mobarak, A.S. Al-Jabab; Corrosion
behavior of a new titanium alloy for dental implant applications in fluoride media;
Materials Chemistry and Physics 86 (2004) 320–329
[9] S. L. Assis, S. Wolynec, I. Costa; Corrosion characterization of titanium alloys by
electrochemical techniques; Electrochimica Acta 51 (2006) 1815–1819
[10] M. Niinomi; Review: Recent research and development in titanium alloys for
biomedical applications and healthcare goods; Science and Technology of Advanced
Materials 4 (2003) 445–454
[11] L.A. Rocha; Development of a hermetic metal/ceramic encapsulation system for
injectable telemetric electronic devices; PhD submitted in University of Minho, 1996
[12] F.H. Jones; Teeth and bones: applications of surface science to dental materials
and related biomaterials; Surface Science Report 42 (2001) 75 – 205
Chapter 1 – State of the art
Master Dissertation
34
[13] D. F. Williams; Biocompatibility of Clinical Implant Materials; Vol. I; Senior
Lecture in Dental and Medical Materials, Dep. of Dental Sciences, University of
Liverpool, England
[14] C.E.B. Marino, L. H. Mascaro; EIS characterization of a Ti-dental implant in
artificial saliva media: dissolution process of the oxide barrier; Journal of
Electroanalytical Chemistry 568 (2004) 115–120
[15] Adya N, Alam M, Ravindranath T, Mubeen A, Saluja B; Corrosion in titanium
dental implants: literature review; J. Indian Prosthodont Soc 5 (2005)126-131
[16] B. Finet, G. Weber, R. Cloots; Titanium release from dental implants: an in vivo
study on sheep; Materials Letters 43 (2000) 159–165
[17] M. Aziz-Kerrzo, K.G. Conroy, A.M. Fenelon, S. T. Farrell, C.B. Breslin;
Electrochemical studies on the stability and corrosion resistance of titanium-based
implant materials; Biomaterials 22 (2001) 1531-1539
[18] M. Barry, D. Kennedy, K. Keating, Z. Schauperl; Design of dynamic test
equipment for the testing of dental implants; Materials & Design 26(2005) 209-216
[19] J. Qu, P. J. Blau, T.R. Watkins, O.B. Cavin, N.S. Kulkarni; Friction and wear of
titanium alloys sliding against metal, polymer, and ceramic counterfaces; Wear 258
(2005) 1348–1356
[20] S. Barril, N. Debaud, S. Mischler, D. Landolt; A tribo-electrochemical apparatus
for in vitro investigation of fretting–corrosion of metallic implant materials; Wear 252
(2002) 744–754
[21] S. Mischler, S. Debaud, D. Landolt.; Wear-accelerated corrosion of passive metals
in tribocorrosion systems; Journal of Electrochemical Society, 145 (1998) 750-758
[22] P. Ponthiaux, F. Wenger, D. Drees, J.P. Celis; Electrochemical techniques for
studying tribocorrosion processes; Wear 256 (2004) 459–468
[23] D. Landolt, S. Mischler, M. Stemp; Electrochemical methods in tribocorrosion: a
critical appraisal; Electrochimica Acta 46 (2001) 3913–3929
[24] Z. Quan, W. Pei-Qiang, L. Tang, J.-P. Celis; Corrosion-wear monitoring of TiN
coated ASP23 steel by electrochemical noise measurements; Colloids and Surfaces A:
Physicochem. Eng. Aspects 280 (2006) 241–244
[25] W. Pei-Qiang, J.-P. Celis; Electrochemical noise measurements on stainless steel
during corrosion–wear in sliding contacts; Wear 256 (2004) 480–490
[26] W. Pei-Qiang; A tribocorrosion electrochemical noise techniques applied to
fretting corrosion. MTM Department, Catholic University of Leuven
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 35
[27] Z. Quan, W. Pei-Qiang , L. Tang, J.-P. Celis; Corrosion-wear monitoring of TiN
coated AISI 316 stainless steel by electrochemical noise measurements; Applied
Surface Science 253 (2006) 1194-1197
[28] T. Hanawa; In vivo metallic biomaterials and surface modification; Materials
Science and Engineering A267 (1999) 260–266
[29] D.J. Wood; The characterization of particulate debris obtained from failed
orthopaedic implants; Research report; San Jose State University, College of Materials
Engineering, 1993
[30] Y. Fovet, J. Gal, F. Toumelin-Chemla; Influence of pH and fluoride concentration
on titanium passivating layer: stability of titanium dioxide; Talanta 53 (2001) 1053–
1063
[31] N. Bellakhal, M. Dachraoui; Study of the benzotriazole efficiency as a corrosion
inhibitor for copper in humid air plasma; Materials Chemistry and Physics 85 (2004)
366–369
[32] F. Mansfeld and J. V. Kenkel; Organic corrosion inhibitors for titanium; Corrosion
Science, Volume 15, Issues 6-12 (1975) 767-773
[33] A. S. Mogoda, Y. H. Ahmad and W. A. Badawy; Corrosion inhibition of Ti-6Al-4V
alloy in sulphuric and hydrochloric acid solutions using inorganic passivators;
Materials and Corrosion 55 (2004) 449-456
[34] Z. Abdel Hady and J. Pagetti; Anodic behaviour of titanium in concentrated
sulphuric acid solutions: influence of some oxidizing inhibitors; Journal of Applied
Electrochemistry 6 (1976) 333 - 338
[35] S. Barril, S. Mischler, D. Landolt ; Influence of fretting regimes on the
tribocorrosion behaviour of Ti6Al4V in 0.9 wt.% sodium chloride solution; Wear 256
(2004) 963–972
[36] L. Duisabeau, P. Combrade, B. Forest; Environmental effect on fretting of metallic
materials for orthopaedic implants; Wear 256 (2004) 805–816
[37] A. Choubey, B. Basu, R. Balasubramaniam; Tribological behaviour of Ti-based
alloys in simulated body fluid solution at fretting contacts; Materials Science and
Engineering A 379 (2004) 234–239
[38] S. Mischler, A. Spiegel, D. Landlolt; The role of passive films on the degradation
of steel in tribocorrosion systems; Wear, vol. 225 (1999) 1078-1087
Chapter 1 – State of the art
Master Dissertation
36
[39] F. Assi; Tribo-Electrochemistry at a micrometer scale – measuring techniques,
tribocorrosion and repassivation; Dissertation submitted to the Swiss Federal Institute
of Technology
[40] J. Komotori, B.J. Lee, H. Dong, P.A. Dearnley; Corrosion response of surface
engineered titanium alloys damaged by prior abrasion; Wear 251 (2001) 1239–1249
[41] Y. Okazakia, E. Gotoh; Comparison of metal release from various metallic
biomaterials in vitro; Biomaterials 26 (2005) 11–21
[42] A.K. Shukla, R. Balasubramaniam, S. Bhargava; Properties of passive film formed
on CP titanium, Ti–6Al–4V and Ti–13.4Al–29Nb alloys in simulated human body
conditions; Intermetallics 13 (2005) 631–637
[43] Y.Z. Huang, D.J. Blackwood; Characterisation of titanium oxide film grown in
0.9% NaCl at different sweep rates; Electrochemical Acta 51 (2005) 1099–1107
[44] T. Hanawa, K. Asami and K. Asaoka; Repassivation of titanium and surface oxide
film regenerated in simulated bioliquid; Biomed Mater Res, vol. 40, (1998) 530-538
[45] P. Jemmely, S. Mischler, D. Landolt; Electrochemical modelling of passivation
phenomena in tribocorrosion; Wear 237 (2000) 63–76
[46] E. Cho, C. Kim, J. Kim, H. Kwon; Quantitative analysis of repassivation kinetics
of ferritic stainless steels based on the high field ion conduction model; Electrochemical
Acta 45 (2000) 1933–1942
[47] C.O.A. Olsson, M. Stemp; Modelling the transient current from two rubbing
electrode configurations: insulating pin on metal substrate and metal pin on insulating
substrate; Electrochemical Acta 49 (2004) 2145–2154
[48] R.W. Bosch, B. Schepers, M. Vankeerberghen; Development of a scratch test in an
autoclave for the measurement of repassivation kinetics of stainless steel in high
temperature high pressure water; Electrochemical Acta 49 (2004) 3029–3038
[49] F.M. Song, K.S. Raja, D.A. Jones; A film repassivation kinetic model for potential-
controlled slower electrode straining; Corrosion Science 48 (2006) 285–307
[50] M. A. Barbosa; Passivation kinetics and pitting studies on Cr, Ni and an 18Cr-
10Ni Stainless Steel; PHD Theses, University of Leeds, Dep. of Metallurgy
[51] J.M. Abd El Kader, F.M. Abd El Wahab, H.A. El Shayeb and G.A. Khedr; Oxide
film thickening on titanium in aqueous solutions in relation to ainion type and
concentration; Dr. Corrosion Journal 16 (1981)
[52] S. Mischler, A. Spiegel, M. Stemp, D. Landolt; Influence of passivity on the
tribocorrosion of carbon steel in aqueous solutions; Wear 251 (2001) 1295–1307
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 37
[53] F. Contu, B. Elsener, H. Bohni; A study of the potentials achieved during
mechanical abrasion and the repassivation rate of titanium and Ti6Al4V in inorganic
buffer solutions and bovine serum; Electrochimica Acta 50 (2004) 33–41
[54] F. Contu, B. Elsener, H. Bohni; Corrosion behaviour of CoCrMo implant alloy
during fretting in bovine serum; Corrosion Science 47 (2005) 1863–1875
[55] G. T. Burstein and R. J. Cinderey; Evolution of the corrosion potential of
repassivating aluminium surfaces; Corrosion Science 33 (1992) 475-492
[56] G.T. Burstein, R.M. Organ; Repassivation and pitting of freshly generated
aluminium surfaces in acidic nitrate solution; Corrosion Science 47 (2005) 2932–2955
Chapter 1 – State of the art
Master Dissertation
38
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 39
Influence of pH and corrosion inhibitors on the tribocorrosion
of titanium in artificial saliva
A.C. Vieira a, A.R. Ribeiro a, L.A. Rocha a,b, J.P. Celis c
a Research Centre on Interfaces and Surfaces Performance, Campus Azurém, 4800-058 Guimarães, Portugal
b University of Minho, Department of Mechanical Engineering, Campus Azurém, 4800-058 Guimarães, Portugal
c Katholieke Universiteit Leuven, Department Metallurgy and Materials Engineering,B-3001 Leuven, Belgium
Abstract
Dental implants are used to replace teeth lost due to decay, trauma, or periodontal
diseases. Dental implants are most of the times subjected to micro-movements at the
implant/bone interface or implant/porcelain interface (due to the transmitted mastication
loads) and chemical solicitations (oral environment). Such implant becomes part of a
tribocorrosion system, which may undergo a complex degradation process that can lead
to implant failure. In this work, the fretting–corrosion behaviour of titanium grade 2 in
contact with artificial saliva was investigated under fretting test conditions. Citric acid
was added to artificial saliva to investigate a pH variation on the tribocorrosion
behaviour of the material. Additionally, three different inhibitors were added to
investigate cathodic and anodic reactions on the electrochemical response. Also, the
influence of inhibitors included in the formulation of tooth cleaning agents or medicines
was investigated. Degradation mechanisms were investigated by electrochemical noise
technique that provided information on the evolution of corrosion potential and
corrosion current during fretting tests. Depassivation and repassivation phenomena
occurring during the tests were detected and discussed. Considering the influence of
corrosion inhibitors, it was observed that the degree of protection varies with the nature
of the inhibitors.
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
40
Keywords: Tribocorrosion; Dental implants; Titanium grade 2: Artificial Saliva;
Inhibitors.
1. Introduction
In dentistry, metallic materials are used as implants in reconstructive oral surgery to
replace a single teeth or an array of teeth, or in the fabrication of dental prosthesis such
as metal plates for complete and partial dentures, crowns, and bridges, essentially in
patients requiring hypoallergenic materials [1,2]. Due to its mechanical properties, good
resistance to corrosion in biological fluids and very low toxicity, titanium is the most
commonly material selected for dental implants and prosthesis [1–4]. Corrosion of
metallic implants is of vital importance, because it can adversely affect the
biocompatibility and mechanical integrity of implants [3–5]. The stability of titanium
under corrosion conditions is essentially due to the formation of a stable and tightly
adherent thin protective oxide layer on its surface [5–7]. The passive film stability
depends on its structure and composition, which in turn are dependent on the conditions
in which it was formed [5–7]. For instance pH is known to have a strong influence on
the corrosion resistance of Ti and Ti alloys [8]. Ion release to the surroundings takes
place when the dissolution of the surface passive film is accompanied by corrosion of
the underlying base material. Extensive release of ions from implants can result in
adverse biological reactions, and can lead to mechanical failure of the device [4–6]. It
should be referred that an accumulation of Ti ions in tissues adjacent to implants has
been reported in conditions not totally attributed to wear [6,9].
Despite their attractive corrosion and toxicological properties, titanium and titanium
alloys generally exhibit poor fretting and wear resistance [4,10,11]. In fact, when used
as implants or prosthesis, cyclic micro-movements at the implant/bone interface or
implant/abutment interface may occur, inducing wear [1,4]. The low fretting and wear
resistance of Ti and Ti alloys is attributed to the poor integrity of the TiO2 surface
passive layer, or to the plastic deformation of surface and subsurface layers [11].
Additionally, under sliding wear conditions Ti alloys have a strong tendency for
transferring material to their counter faces, and tribochemical reactions at the contact
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 41
surface are likely to occur [10]. Also, the release of wear debris may lead to cellular
damage, inducing inflammation or encapsulation of the implant by fibrous tissue [4].
These environmental alterations may also alter the corrosion behaviour of the material.
Therefore, Ti dental implants and prostheses exposed to the combined degradation
by corrosion and fretting, constitute a tribocorrosion system. It should be stressed that
the two mechanisms of degradation do not proceed separately, but depend on each other
in a complex way. Normally corrosion is accelerated by wear and, similarly wear may
be affected by corrosion phenomena [12,13]. In fact, wear may lead to local removal of
the passive film resulting on the exposure of the metal surface to the aggressive
environment. Consequently, the corrosion rate will increase (wear accelerated
corrosion) leading to a rapid degradation of a contact. Eventually, corrosion products
will accumulate in the mechanical contact region, influencing the wear regime [12,13].
Several recent studies have focused on the fretting–corrosion behaviour of Ti alloys.
Barril et al. [14] investigated the effect of the displacement amplitude, normal force, and
tribometer stiffness on the tribocorrosion behaviour of Ti6Al4V/Al2O3 pairs, in contact
with a 0.9 wt% NaCl solution. They concluded that wear accelerated corrosion would
only occur, if the amplitude of the displacement was enough for causing slip between
the materials. They proposed a model describing the influence of mechanical parameters
(normal force, elasticity, and speed) on the wear accelerated corrosion of materials.
Duisabeau et al. [15], studied the tribocorrosion behaviour of Ti6Al4V/316L stainless
steel pairs in a Ringer’s solution, under gross slip conditions. They concluded that the
dissipated mechanical energy and fretting regimes are strongly affected by the presence
of a corrosive lubricant, because of the electrochemical phenomena caused by the
electrolyte. The effect of the electrochemical conditions on the tribocorrosion behaviour
of Ti6Al4V /Al2O3 contacts was investigated by Barril et al. [13,16]. They observed that
under gross slip conditions, friction and wear are critically dependent on the applied
potential, affecting the thickness, composition, and stoichiometry of the passive film.
They concluded that the degree of oxidation of the plastically deformed metallic
material would influence the extent of wear.
In this work, the tribocorrosion of titanium grade 2, under fretting regime, in contact
with artificial saliva solutions, is investigated. The influences of pH and corrosion
inhibitors in the artificial solution are considered.
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
42
2. Experimental
Samples made of Ti grade 2 (all from the same sheet) were cut in size as 2.5 cm×2.5
cm, and mechanically polished down to 0.25 µm. The initial sample roughness was Ra
= 0.03 µm ± 0.004 µm. All the samples were polished 1 day before the experiments in
order to allow the formation of an oxide surface layer. After polishing they were
ultrasonically cleaned with ethanol and distilled water and finally dried. The nominal
chemical composition of Ti grade 2 was 0.25 wt% O, 0.03 wt% N, 0.08 wt% C, 0.015
wt% H, 0.3 wt% Fe, and residuals 0.4 wt%.
Fretting–corrosion behaviour was investigated using a triboelectrochemical
approach, in which the electrochemical noise technique was used to monitor the
fluctuations of corrosion potential and corrosion current during the fretting tests.
Corundum balls (Ø 10 mm) were selected as counter body material (Ceratec, The
Netherlands) because of high wear resistance, chemical inertness, and electrical
insulating properties. Titanium grade 2 specimens used as working electrode (WE) were
covered with an adhesive tape to leave an area of 1 cm2 exposed to the test solutions. A
Ag/AgCl reference electrode and a microelectrode consisting in a Pt electrode with a
diameter of 0.25 mm and a tip length of 1.2mm were used. The experimental set-up
used for electrochemical noise measurements during corrosion–wear tests on immersed
samples is schematically shown in Fig. 2.1. As described elsewhere [17], the
configuration of the experimental set-up was optimised to improve accuracy and
minimize external noise.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 43
Fig. 2.1. Schematic representation of the experimental set up used for the tribocorrosion
experiments.
A potentiostat (Solartron electrochemical interface model 1287) was used, which
allows voltage and current measurements at a resolution of 1 µV and 1 pA, respectively.
The microelectrode coupled to the working electrode was used to sense the current
flowing between them. The counter body was lifted away from the WE at the end of
fretting tests. The electrochemical noise data are reported according to ASTM
conventions [22]. The bidirectional sliding (fretting) test equipment was described
elsewhere [19]. The sliding conditions correspond to a fretting test performed at a
normal load of 2 N, an oscillating frequency of 1 Hz, and a linear displacement
amplitude of 200 µm. These fretting tests were performed for 5000 and 10000 cycles at
an ambient temperature of 23 ºC. The number of cycles, the tangential force, the normal
force, the displacement amplitude, and the coefficient of friction were recorded at
equally spaced time increments during the whole test duration.
The solutions used during the experiments were artificial saliva (AS), with different
chemical compositions (Table 2.1). Citric acid was added in order to investigate the
influence of a pH variation on the tribocorrosion behaviour of the contact. Three
different kinds of inhibitors were added in order to investigate the action of the cathodic
and anodic reactions on the electrochemical response. Also, it was found useful to
analyse the influence of corrosion inhibitors, which may be included in the formulation
of tooth cleaning agents or medicines.
Table 2.1: Chemical composition of the artificial saliva solutions used (wt%).
Solution
Compound
Artificial saliva (AS)
AS
+
citric acid
AS
+
anodic inhibitor
AS
+
cathodic inhibitor
AS
+
organic inhibitor
Sodium Chloride, NaCl 0.70 0.70 0.70 0.70 0.70
Potassium Chloride, KCl 1.20 1.20 1.20 1.20 1.20
Citric Acid, C6H8O7.H2O 0.025
Sodium Nitrite, NaNO2 0.16
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
44
Calcium Carbonate, CaCO3
0.5
Benzotriazole, C6H5N3 1.5
pH 5.5 3.8 5.5 5.5 5.5
After the tribocorrosion tests, the samples were ultrasonically cleaned with ethanol
and distilled water during 10 min. The wear scars were investigated by reflected light
microscopy with Nomarski contrast, laser profilometry (Rodenstock RM 600), and
SEM-EDX (Philips XL 30 ESEM FEG). The wear volume was determined by a
profilometric method as described earlier [19].
3. Results and discussion
3.1. Tribological measurements
The evolution of the mechanical contact behaviour was investigated by acquiring
force–displacement hysteresis loops at certain time intervals during the fretting tests.
For tests performed during 5000 fretting cycles, loops at 20, 1000, and 5000 cycles were
obtained. As these tests were performed at a frequency of 1 Hz, the number of cycles
corresponds to the testing time in seconds. Moreover, during tests performed for 10000
fretting cycles tests, the loop at 10000 fretting cycles was also recorded. In Fig. 2.2
representative fretting log diagrams (AS and AS + citric acid) are presented. The shape
of the tangential force–displacement (Ft–D) cycles is, in all cases, a parallelogram,
indicating that the accommodation of displacement occurs under a gross-slip regime
[15,19,20].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 45
20004000
60008000
10000
-2
-1
0
1
2
0
50
100150
200
Ft (
N)
D (µm
)Number of Cycles
Artificial saliva (AS)
(a)
4000
8000
12000
-1
0
1
2
050
100150
200
AS + citric acid
Ft (
N)
D (µm
)Number of Cycles (b)
Fig. 2.2: Fretting logs recorded during tests conducted for 10,000 cycles in (a) AS and (b) AS +
citric acid solution. Fretting test parameters: 2 N, 1 Hz, 200 µm, 10,000 cycles.
In fact, under the imposed fretting conditions, the elastic deformation of the material
and the stiffness of the apparatus do not accommodate the imposed displacement, and
an effective relative motion between the two contacting materials takes place.
Consequently, friction occurs between the two materials, resulting in a measurable
tangential force. Ft increases during the test, and reaches a steady state after 5000
cycles. This behaviour was observed under all test conditions.
In Fig. 2.3 the average coefficient of friction monitored in the AS and artificial saliva
plus organic inhibitor (AS + organic) is presented. In the other solutions, the evolution
of the coefficient of friction with fretting cycles is similar to the one noticed in the AS
solution. In the AS solution three regions can be identified during the fretting tests. A
first region extends up to ca. 2000 cycles where an increase of the coefficient of friction
is observed. This region corresponds to the running-in period in which an adjustment of
the two contacting surfaces occurs by crushing and smearing of the asperities [14]. A
second region expands up to ca. 6000 cycles, during which the coefficient of friction
remains fairly stable. Finally, after ca. 7000 cycles, a monotonic decrease of the
coefficient of friction is observed.
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
46
0 2000 4000 6000 8000 10000
0.5
0.6
0.7
0.8
0.9
1.0
AS + organic inhibitor
Artificial saliva (AS)
Fretting Cycles
Ave
rag
e c
oe
ffic
ien
t o
f fr
ictio
n
Fig. 2.3. Evolution of the average coefficient of friction during fretting tests performed for
10,000 fretting cycles in AS and AS + organic inhibitor solutions. Fretting test parameters: 2 N,
1 Hz, 200 µm, and 10000 cycles.
The coefficient of friction exhibits strong oscillations during the fretting test. After
the running-in period, these oscillations may be attributed to the build-up and
accumulation of third-body particles in the contact region. After the accommodation of
the two surfaces debris are ejected out of the contact as rubbing keeps on [13].
Micrographs of the wear scars of Ti samples are presented in Fig. 2.4. No significant
differences were observed between samples tested in the different solutions. Under the
imposed wear test conditions, two regions can be identified in the wear track. The
central part is characterized by a relative severe material damage and the presence of
wear debris. The surrounding external part of the wear scar is smooth and exhibits some
material smearing [21]. The wear scar is characterized by sliding wear marks aligned in
the fretting direction. The central area of the wear scar reveals scales, most probably
formed by an extensive plastic surface shear. These scales are likely to delaminate and
detach from the surface, inducing the oscillations in Ft. Additionally, it is expected that
during fretting under gross slip regime and oxidizing conditions, cracking, and
delamination of wear particles will be accelerated by their oxidation [13,16].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 47
(a1)
(b1)
(a2)
(b2)
(a3)
(b3)
Fig. 2.4. Micrographs of (a) the wear scar, and (b) of the interior of the scar: (1) AS; (2) AS +
organic inhibitor; (3) AS + citric acid. Fretting test parameters used were: 2 N, 1 Hz, 200 µm,
and 5000 fretting cycles.
As shown in Fig. 2.2, the real displacement is inferior to the imposed displacement
200 µm, because a part of the displacement was accommodated by the elastic
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
48
deformation of the fretting contact and the limited stiffness of the test equipment [20].
The area (A) in the fretting logs shown in Fig. 2.2 can be expressed as [15]:
!= ds)D(FA t (Eq. 1)
with Ft the tangential force and D is the displacement. Thus, area A represents the
dissipated friction energy in the contact during each fretting cycle [16,17]. In Fig. 2.5
(a) the evolution of the dissipated energy as a function of the number of friction cycles
is presented. In all test solutions, the dissipated energy increases with fretting duration.
In the artificial saliva with organic inhibitor (AS + organic) the dissipated energy is
substantially higher than in the other solutions, attending a value of ca. 6.7 J, after
10000 cycles due to the higher coefficient of friction (Fig. 2.2). Also, as observed in
Fig. 2.5 (a), the dissipated energy tends to reach a steady state as the number of fretting
cycles increases. The evolution of the wear volume, calculated from profilometric
measurements after 5000 and 10000 cycles, is plotted in Fig. 2.5 (b) as a function of the
dissipated energy. These wear volumes account for the material removed from the
contact region both by wear and corrosion. When the fretting contact is under gross slip
regime, a linear relation between the wear volume and the cumulated dissipated energy
is commonly observed [15,16]. In Fig. 2.5b the existence of a linear relationship is
assumed. The slope of the wear volume/dissipated energy curves expresses the wear
rate per unit of dissipated energy. Some differences are observed among the different
solutions although these distinctions are dependent on the extent of the fretting tests.
Regarding the test performed during 5000 cycles, it can be observed that the Ti sample
tested in the AS + organic solution suffers a lower weight volume loss that is ca. 2 times
lower than that the one noticed in the AS + cathodic inhibitor solution. However, after
10,000 fretting cycles, Ti has a lower wear volume loss in the AS + citric acid solution,
indicating that some protection is provided by the addition of citric acid to the solution.
Regarding the wear rate per unit of dissipated energy, values between 7.2×104 and
7.8×104 µm3 J-1 have been reported in the literature for the Ti6Al4V alloy in contact with
saline solutions [15,16]. However, such data for pure titanium are not yet available in
literature. In this work, depending on the nature of the solution, values between 8.9×103
and 8.6×104 µm3 J-1 were derived. The nature of the solution appears to influence this
behaviour. The AS + cathodic inhibitor and AS + anodic inhibitor solutions induce a
higher wear rate of Ti per unit of dissipated energy than the other solutions. No
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 49
significant differences in this behaviour were observed between the AS and the AS +
citric acid solutions, while a rather slow evolution of the wear volume loss with
dissipated energy was observed in the AS + anodic solution.
0 2000 4000 6000 8000 10000
0
1
2
3
4
5
6
7
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + cathodic inhibitor AS + organic inhibitor
Dis
sip
ate
d e
ne
rgy (
J)
Fretting Cycles 40000 60000 80000 100000 120000
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
7.5
10000 Cycles
5000 Cycles
Dis
sip
ate
d E
ne
rgy(J
)
Dissipated Energy(J)
Wear Volume (µm3)
Wear Volume (µm3)
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + cathodic inhibitor AS + organic inhibitor
(a) (b)
Fig. 2.5. (a) Evolution of the contact dissipated energy with testing time; (b) evolution of the
wear volume as a function of contact dissipated energy. Fretting test parameters used were: 2 N,
1 Hz, 200 µm, and 10000 cycles.
3.2. Electrochemical measurements
The evolution of the corrosion potential with fretting testing time is shown in Fig.
2.6. Before the start of the fretting tests, the test samples were immersed in the different
electrolytes to reach stabilization. Once stabilization was achieved, fretting tests were
started. A significant drop in potential is observed immediately after the start of the
mechanical action, indicating the destruction of the passive film (depassivation), and the
exposure of fresh active titanium to the test solutions [12–17].
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
50
0 2000 4000 6000 8000 10000 12000-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
End of Fretting
Beginning of Fretting
Artificial saliva (AS) AS + organic inhibitor
Time (s)
Time (s)
Pote
ntial (V
vs.
Ag/A
gC
l)
0 2000 4000 6000 8000 10000 12000
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
End of Fretting
End of Fretting
Beginning of Fretting
Time (s)
Po
ten
tia
l (V
vs.
Ag
/Ag
Cl)
AS + citric acid AS + anodic inhibitor AS + cathodic inhibitor
(a) (b)
Fig. 2.6. Evolution of open-circuit potential: (a) AS and AS + organic inhibitor solutions; (b)
AS + citric acid and AS + anodic and cathodic inhibitor solutions. Fretting test parameters: 2 N,
1 Hz, 200 µm, and 10000 cycles.
The evolution of the corrosion potential at the start of the fretting test is shown in
Fig. 2.7. In the AS (Fig. 2.7 (a)), AS + citric acid and AS + anodic inhibitor (Fig. 2.7
(b)), the potential reaches very low values within a short period of ca. 100 s, before it
evolves to more noble values. Concerning the behaviour of Ti in the AS + organic
inhibitor (Fig. 2.7 (a)) and in the AS + cathodic inhibitor (Fig. 2.7 (b)) solutions, it can
be observed that the drop in potential is significantly lower attending, after some time, a
steady state value that remains almost unchanged during the remaining fretting test
cycles (Fig. 2.6 (a) and (b)).
0 200 400 600 800 1000-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
Beggining of fretting
Artificial saliva (AS) AS + organic inhibitor
Po
ten
tia
l (V
vs.
Ag
/Ag
Cl)
Po
ten
tia
l (V
vs.
Ag
/Ag
Cl)
Time (s)
Time (s)
0 200 400 600 800 1000-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
Beggining of fretting
AS + citric acid AS + anodic inhibitor AS + cathodic inhibitor
Time (s)
Pote
ntial (V
vs.
Ag/A
gC
l)
(a) (b)
Fig. 2.7. Evolution of open-circuit potential values during the running-in of the fretting tests: (a)
AS and AS + organic inhibitor solutions; (b) AS + citric acid and AS + anodic and cathodic
inhibitor solutions. Fretting test parameters: 2 N, 1 Hz, 200 µm, and 10000 cycles.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 51
In all the other solutions, an abrupt increase in potential occurs after the first stage.
Then, the corrosion potential slowly evolves to more noble values, indicating a decrease
in corrosion susceptibility as the fretting tests go on. As shown in Fig. 2.4, this
behaviour may be attributed to the build-up of a tribolayer in the contact region that
creates a barrier between the Ti surface and the test solution. As observed in Fig. 2.6 (a)
and (b), the slow increase in potential is sometimes interrupted by abrupt potential drop
events, probably due to the sudden partial delamination of the tribolayers. The sample
tested in the AS + citric acid solution is the one exhibiting the highest electrochemical
potential before, at the beginning, during and after the fretting test (see Figs. 2.6 (b) and
2.7 (b)).
A representative evolution of the coefficient of friction and of the corrosion potential
as a function of the fretting time is plotted in Fig. 2.8 for Ti in AS + anodic solution.
The potential drop events are accompanied by a sudden decrease of the coefficient of
friction. As explained above, the delamination of the tribolayers formed in the contact
region may explain this behaviour. Remarkable is the evolution of the corrosion
potential towards higher potential values, observed after ca. 7000 cycles in the AS
solution and in the AS + citric acid or AS + anodic inhibitor (Fig. 2.6).
0 2000 4000 6000 8000 10000
-0.6
-0.5
-0.4
-0.3
-0.2
0.24
0.32
0.40
0.48
0.56
0.64
0.72
0.80
Po
ten
tia
l (V
vs.
Ag
/Ag
Cl)
Time (s)
Co
eff
icie
nt
of
fric
tio
n
Potential
Friction coefficient
Fig. 2.8. Evolution of the open-circuit potential and of the coefficient of friction during the
fretting test. AS + anodic inhibitor. Fretting test parameters: 2 N, 1 Hz, 200 µm and 10000
cycles.
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
52
As shown in Fig. 2.8, this variation in potential is accompanied by a decrease of the
coefficient of friction. This new regime might be attributed to the stabilization of the
three body contact area, after 7000 cycles. In other words, as the mechanical action
proceeds in the contact area, wear debris become smeared out and entrapped into the
surface. Consequently, a delamination of the tribolayers at the contact area becomes less
probable, and the formed mechanical mixed layer acts as a protective film both in terms
of wear and corrosion. Finally, it should be noticed that at the end of the fretting test,
the corrosion potential recovers its original value of before the test or becomes even
slightly higher. This behaviour indicates that the newly formed passive film, after a total
removal of the naturally formed passive film by the fretting action, together with the
mechanical mixed layer has quite similar characteristics as the naturally formed film
present on titanium before mechanical loading. The exception to this behaviour are the
Ti samples tested in the AS + cathodic inhibitor and AS + organic inhibitor, in which
such a recovery of the corrosion potential is not observed after the fretting tests. These
solutions by inhibiting the cathodic reaction(s), probably hinder the formation of a new
passive film on the surface of worn Ti.
The evolution of the corrosion current monitored by the electrochemical noise
technique during the fretting tests is presented in Figs. 2.9–2.13. The behaviour of Ti in
the AS, AS + citric acid, and AS + anodic inhibitor solutions (Figs. 2.9–2.11) differs
from that observed in the AS + cathodic inhibitor and AS + organic inhibitor solutions
(Figs. 2.12 and 2.13). Nevertheless, as it can be observed, in all samples the
depassivation of the materials during the initial stage of fretting is accompanied by a
sudden increase in corrosion current density. This increase is much higher in the AS +
cathodic inhibitor and AS + organic inhibitor solutions than in the other solutions (see
current scale in the graphs), indicating that these solutions have a much stronger
corrosive action on fresh (depassivated) titanium surfaces than the other ones. In fact,
the presence of organic and cathodic inhibitors, at concentrations used in this work,
significantly affects the corrosion rate of titanium. Actually, as shown in Figs. 2.12 and
2.13, the corrosion current monitored during the fretting tests on samples immersed in
these solutions, is much higher than the one found in other solutions. Also, after the first
increase in current, arising from the removal of the passive film in the contact region
when sliding starts, no significant variation in the corrosion current is observed during
the fretting tests. Nevertheless, the organic inhibitor seems to be somewhat more
effective based on the corrosion current results. That is in accordance with the slightly
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 53
lower wear volume loss noticed on Ti in this solution (Fig. 2.5 (b)), in comparison with
the AS + cathodic inhibitor solution, notwithstanding the higher coefficient of friction
(Fig. 2.3). In all the other samples (Figs. 2.9–2.11), after the initial increase in corrosion
rate caused by the destruction of the passive film, the corrosion current monotonically
decreases during the fretting test. Again, as the corrosion potential evolution revealed
(Fig. 2.6a and b), some current peaks are observed, which are in good agreement with
the oscillation in the coefficient of friction, as appears from Fig. 2.14. The delamination
of the tribolayers formed at the contact surface, by exposing or facilitating the access of
the solution to the metallic Ti, may explain this behaviour.
A decrease in the corrosion current is observed after ca. 7000 cycles, in the AS, AS +
citric acid, and AS + anodic inhibitor solutions indicating the formation of a third-body
protective layer in the contact region, as already referred. However, the decrease in
corrosion current is more pronounced in the AS + citric acid and AS + anodic inhibitor
solutions, indicating that these additives provide some protection to titanium. The
slightly lower wear volume loss of Ti in these solutions, in comparison to the AS
solution (Fig. 2.5 (b)), may be attributed to the corrosion protection afforded by the
presence of the citric acid or the anodic inhibitor. In other words, the wear and corrosion
behaviour of Ti is influenced by the oxidation and reduction reactions occurring in the
contact area during fretting, depending on the chemical composition of the test
solutions.
0 2000 4000 6000 8000 10000 120000.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
Artifical saliva (AS)
Cu
rre
nt
(µA
)
Time (s)
Time (s)
Fig. 2.9. Evolution of the corrosion current during the fretting tests in AS solution. Fretting test
parameters: 2N, 1 Hz, 200 µm, and 10,000 cycles.
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
54
0 2000 4000 6000 8000 10000 12000 140000.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
AS + citric acid
AS + anodic inhibitor
Time (s)
Curr
ent
(µA
)
Curr
ent
(µA
)
Fig. 2.10. Evolution of the corrosion current during the fretting tests in AS + citric acid. Fretting
test parameters: 2N, 1 Hz, 200 µm, and 10,000 cycles.
0 2000 4000 6000 8000 10000 120000.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
AS + anodic inhibitor
Time (s)
Curr
ent
(µA
)
Fig. 2.11. Evolution of the corrosion current during the fretting tests in AS + anodic inhibitor.
Fretting test parameters: 2N, 1 Hz, 200 µm, and 10,000 cycles.
0 2000 4000 6000 8000 10000 12000 14000-20
0
20
40
60
80
100
120
AS + cathodic inhibitor
Curr
ent
(µA
)
Time (s)
Fig. 2.12. Evolution of the corrosion current during the fretting tests in AS + cathodic inhibitor.
Fretting test parameters: 2N, 1 Hz, 200 µm, and 10,000 cycles.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 55
0 2000 4000 6000 8000 10000 12000 14000-20
0
20
40
60
80
100
120
AS + organic inhibitor
Time (s)
Curr
ent
(µA
)
Fig. 2.13. Evolution of the corrosion current during the fretting tests in AS + organic inhibitor.
Fretting test parameters: 2N, 1 Hz, 200 µm, and 10,000 cycles.
0 2000 4000 6000 8000 100000.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Cu
rre
nt
(µA
)
Time (s)
Co
eff
icie
nt
of
fric
tio
n
Friction coefficient
Current
Fig. 2.14. Evolution of the corrosion current and of the coefficient of friction during the fretting
tests in AS + anodic inhibitor. Fretting test parameters: 2 N, 1 Hz, 200 µm, and 10000 cycles
4. Conclusions
In this work, the influence of pH and corrosion inhibitors in artificial saliva on the
tribocorrosion behaviour of pure Ti under fretting was investigated.
The addition of citric acid or anodic inhibitor to artificial saliva results in a slight
improvement of the tribocorrosion behaviour of Ti. No significant differences were
observed in the wear rate per dissipated energy, but a lower wear volume loss was
obtained that can be attributed to the slightly lower corrosion rate observed in these
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
56
solutions during the fretting tests. The protection noticed by the addition of citric acid or
an anodic inhibitor to artificial saliva is probably due to the nature of the oxidation and
reduction reactions occurring in the contact area during fretting. Tribolayers are formed
in the contact region during the tribocorrosion test. These tribolayers become more
stable after ca. 7000 cycles in solutions containing citric acid or anodic inhibitor, as
revealed by a lower coefficient of friction and a lower corrosion current.
The addition of a cathodic or an organic inhibitor to artificial saliva at concentrations
tested in this work, has a hazardous effect on the fretting–corrosion behaviour of
titanium. Both an increase in the wear volume loss per unit-dissipated energy and a
significant higher corrosion rate during fretting tests, were observed in these solutions.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 57
References
[1] M. Barry, D. Kennedy, K. Keating, Z. Schauperl; Design of dynamic test equipment
for the testing of dental implants; Materials & Design, Vol. 26 (2005) 209–216
[2] Y. Okazaki, Mater. Trans. 43 (2002) 3134–3141
[3] X. Liu, P.K. Chu, C. Ding, Mater. Sci. Eng. R 47 (2004) 49–121
[4] F.H. Jones; Teeth and bones: applications of surface science to dental materials and
related biomaterials; Surface Science Report, Vol. 42 (2001)75 – 205
[5] C.E.B. Marino, L.H. Mascaro; EIS characterization of a Ti-dental implants in
artificial saliva media: dissolution process of the oxide barrier; J. Electroanalytical
Chemistry, Vol. 568 (2004) 115–120
[6] A.W.E. Hodgson, Y. Mueller, D. Forster, S. Virtanen, Electrochem. Acta 47 (2002)
1913–1923
[7] A.K. Shukla, R. Balasubramaniam, S. Bhargava; Properties of passive film formed
on CP titanium, Ti–6Al–4V and Ti–13.4Al–29Nb alloys in simulated human body
conditions; Intermetallics, Vol. 13 (2005) 631–637
[8] M. Nakagawa, Y. Matono, S. Matsuya, K. Udoh, K. Ishikawa; The effect of pt and
pd alloying additions on the corrosion behavior of titanium in fluoride-containing
environments; Biomaterials 26 (2005) 2239–2246.
[9] D.F. Williams, G. Meachim, J. Biomed. Mater. Res. Symp. 5 (Part 1) (1974) 1 – 9
[10] J. Qu, P.J. Blau, T.R. Watkins, O.B. Cavin, N.S. Kulkarni, Friction and wear of
titanium alloys sliding against metal, polymer, and ceramic counterfaces; Wear, Vol.
258 (2005) 1348–1356
[11] M. Long, H.J. Rack; Friction and surface behaviour of selected titanium alloys
during reciprocating sliding motion; Wear, Vol. 249 (2001) 158–168
[12] P. Ponthiaux, F. Wenger, D. Drees, J.-P. Celis; Electrochemical techniques for
studying tribocorrosion processes; Wear, Vol. 256 (2004) 459–468
[13] S. Barril, N. Debaud, S. Mischler, D. Landolt; A tribo-electrochemical apparatus
for in vitro investigation of fretting–corrosion of metallic implant materials; Wear, Vol.
252 (2002) 744–754
[14] S. Barril, S. Mischler, D. Landolt; Influence of fretting regimes on the
tribocorrosion behaviour of Ti6Al4V in 0.9wt% sodium chloride solution; Wear, Vol.
256 (2004) 963–972
Chapter 2 - Influence of pH and corrosion inhibitors on the tribocorrosion of titanium in artificial saliva
Master Dissertation
58
[15] L. Duisabeau, P. Combrade, B. Forest; Environmental effect on fretting of metallic
materials for orthopaedic implants; Wear, Vol. 256 (2004) 805–816
[16] S. Barril, S. Mischler, D. Landolt; Electrochemical effects on the fretting corrosion
behaviour of Ti6Al4V in 0.9% sodium chloride solution; Wear 259 (2005) 282–291
[17] W. Pei-Qiang, J.-P. Celis; Electrochemical noise measurements on stainless steel
during corrosion–wear in sliding contacts; Wear, Vol. 256 (2004) 480–490
[18] ASTM Standard: G3, Annual Book of ASTM Standards, vol. 03.02
[19] H. Mohrbacher, J.-P. Celis, J.R. Roos, Tribol. Int. 28 (1995) 269–278
[20] S. Mischler, in: G. Zambelli, L. Vincent (Eds.), Mat´eriaux et Contacts: Une
Approache Tribologique, Publ. Presses Polytechniques et Universitaires Romandes,
Lausanne (Switzerland), 1998, pp. 107–116
[21] S. Fouvry, P. Kapsa, H. Zahouani, Wear 203–204 (1997) 393–403
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 59
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
60
Repassivation evolution of Ti in artificial saliva solutions under
tribocorrosion conditions
A.C. Vieira a, L.A. Rocha a,b, E. Ariza a,b, J.P. Celis c
a Research Centre on Interfaces and Surfaces Performance, Campus Azurém, 4800-058
Guimar� es, Portugal b University of Minho, Department. of Mechanical Engineering, Campus Azurém, 4800-
058 Guimar� es, Portugal c Katholieke Universiteit Leuven, Department of. Metallurgy and Materials
Engineering, B-3001 Leuven, Belgium Abstract
Degradation of Ti dental implants is a common process usually caused by
mechanical stress or by the physiological environment (human saliva) that surround the
implant. Additionally, the Ti passive film formed on the implant, which naturally grows
on the metallic surface protecting it from corrosion, can be scratched or destroyed
during the insertion and implantation into the hard tissue by abrasion with bone and
other materials [1].
In this work, the repassivation evolution of commercial pure (cp) titanium in
artificial saliva solutions, in a tribocorrosion system was evaluated. Also, the influence
of pH variation and the presence of corrosion inhibitors in the artificial saliva were
investigated. Ti samples were subjected to reciprocating sliding in a pin-on-plate
tribometer against a corundum ball. Two different electrochemical conditions were
imposed: Open-circuit potential (OCP) and potentiostatic control (1 V) in the passive
region of the polarization curve of the Ti. Also, to obtain more detailed information on
the characteristics of the original and reformed passive film, electrochemical impedance
spectroscopy (EIS) measurements were made before and after the mechanical damage.
Citric acid was added to the physiological solutions in order to understand the effect of
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 61
pH on the repassivation. Additionally, different kinds of inhibitors were also added to
the artificial saliva solution in order to investigate the influence of anodic and cathodic
reactions on the repassivation phenomena. Finally, all samples were characterized using
the SEM, EDS, and AFM techniques.
The repassivation evolution of commercially pure Ti seems to be affected by pH
decreases. No improvement in the repassivation kinetics was suggested with the
presence of corrosion inhibitors, in artificial saliva solution.
Keywords: Tribocorrosion, Dental Implants, Repassivation, Titanium
1. Introduction
Nowadays, Ti and its alloys are the most attractive metallic materials for dental
applications because they have superior corrosion resistance and good biocompatibility
[1]. The Ti corrosion resistance is essentially promoted by the presence of a stable oxide
passive film on its surface [2-4]. This protective oxide layer naturally grows on a
metallic material surface, either in air or in wet environments. However, the passive
film can be scratched or destroyed during insertion and implantation into hard tissue by
abrasion with bone or with other materials [1]. Also, during the insertion of the dental
implant, the pH can decrease from 7.35–7.45 to 5.2 in the hard tissue [5,6]. When a
marked decrease in pH happened, metallic materials become corroded and toxicity and
allergy can occur in vivo [5,6].
After insertion, when the dental implant is simultaneously exposed to mechanical
stress (promoted by the mastication loads) and chemical degradation (promoted by the
physiological environment that surround the implant), it becomes part of a
tribocorrosion system [7,8]. The importance of the study of the repassivation of passive
alloys is due to the particular chemical and mechanical behavior of these metals once
the oxide protective film is removed [9]. Also, the comprehension of the repassivation
evolution of the surface after the mechanical damage stops (passive film re-growth) can
provide important information about the system.
The repassivation of Ti used in implants under bio-conditions has been already the
subject of some studies. However, the typical analyses are related to the kinetics of the
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
62
process [10-13]. For instances, F. Contu et al [13] estimate the evolution of the
corrosion potential during mechanical abrasion of Ti and Ti-6Al-4V in inorganic buffer
solutions and the repassivation rate of the metal surface. It is shown that when the
rubbing tip touches the metallic surface, the potential abruptly decreases due to the local
destruction of the passive film and the creation of fresh active metallic surface.
Additionally, the new fresh metallic surface will be exposed to the electrolyte.
However, when the abrasion action stops, the potential increases again. The
repassivation rate of the materials was determined calculating the tangential rate at the
initial repassivation curve as well as the percentage of open circuit potential (OCP)
ennoblement after the rubbing process was stopped. F. Contu et al [13] also suggested
that, when immersed in the inorganic buffer solutions, the repassivation rate of cp
titanium is affected by the pH: repassivation rate of cp titanium significantly decreases
both at acidic and alkaline pH. Although, A.M. Al-Mayouf et al [14] studied the effect
of pH on the corrosion behaviour of cp Ti, Ti–6Al–4V and Ti–30Cu–10Ag (a new
titanium alloy used for dental implants), immersed in artificial saliva solutions. The
different pH values studied were 7.2 and 3. A decrease in the corrosion tendency of cp
Ti, when a decrease in pH occurs, was one of the conclusions of this work.
G.T. Burstein et al [15,16] suggested that the free corrosion potential of freshly
generated metallic surfaces increases linearly with log t during repassivation, due to the
oxide film growth, following ∆E = bE logt + k [15].
The main aim of the present study is to evaluate the repassivation evolution of
commercially pure Ti in different artificial saliva solutions, and to understand how a
decrease in pH and presence of inhibitors could influence the repassivation
phenomenon.
2. Experimental
The material used was commercial pure titanium grade 2 with the following
nominal chemical composition (wt %): 0.25 O, 0.03 N, 0.08 C 0.015 H, 0.3 Fe and 0.4
of residuals. Samples (all from the same ingot) were cut in sections of 3.0 cm × 2.0 cm
× 0.1 cm. After, they were mechanically polished (Ra = 0.03 µm ± 0.01 µm),
ultrasonically cleaned with ethanol during 10 minutes and with distillate water during
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 63
10 minutes. The samples were well dried with hot air and kept in the excicator until the
beginning of the test. It is important to point out that all the samples were polished one
day before being tested in order to try to control the formation of the oxide passive
layer.
A Tribometer TE 67 (Plint, Tribology Products, UK) was used to carry out the
tribological measurements. The tribological arrangement used was pin-on-plate, with
reciprocating sliding motion of the counterbody (total linear stroke of 6 mm) and fixed
flat samples. The tribological conditions used were: 10 N as normal load, 1 Hz as
frequency and 1 m as total displacement (86 s approximately). Corundum balls
(Ceratec, The Netherlands), with 10 mm diameter, were selected as counterbody
especially due to their high wear resistance, chemical inertness and electric insulating
properties.
The electrochemical tests were performed using a Galvanostat-Potentiostat PGZ 100
- Radiometer Analytical - controlled by the Voltamaster-4 software running under a
personal computer. The electrochemical parameters were determined using a standard
calomel electrode – SCE (B20B110- Radiometer Analytical) - as a reference electrode,
a platine electrode as auxiliary electrode (wire B35M110 – Radiometer Analytical), and
Ti as working electrode. The immersion area was kept constant (2.54 cm2).
Two different electrochemical conditions were used to study the Ti repassivation: in
open-circuit potential (OCP) and potentiostatic control, following the sequences
illustrated in Fig. 3.1 and Fig 3.2, respectively. Additionally, electrochemical
impedance spectroscopy tests – EIS (Galvanostat-Potentiostat PGZ 100 – Radiometer
Analytical) – were carried out in both electrochemical conditions, in order to
characterize passive film.
In OCP conditions (Fig. 3.1), the potential was measured before, during and after the
mechanical damage test: first, the samples were immersed in the electrolyte until
potential stabilization after each EIS measurements were performed. After the
stabilization period and EIS experiment, the counterbody was placed in contact with the
titanium surface and then time for stabilisation of the OCP was given again, before the
mechanical damage was started. At the end of tribocorrosion test, and after a certain
time to allow the stabilization of the metallic surface, a new EIS measurement was
performed.
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
64
Co u nterbody
! 7 min ! 3h50min
! 7 min
EIS
8 6 s 5 min
EIS
OCP
Sliding End of
Sliding
OCP
( Stabilization
period )
Fig. 3.1. Schematic representation of the electrochemical experiment done in open circuit
conditions (OCP).
In the potentiostatic control experiments (Fig. 3.2) a cathodic polarization was
performed after the immersion of the samples into the solution in order to promote an
in-situ cleaning of the samples. Then, an anodic potential (1V vs. SCE), in the passive
region of the polarization curve was applied. The sliding began 1 min after the
placement of the counterbody in contact with the Ti surface.
Polarization control
at 1000 mV
Polarization control at 1000 mV Cathodic
polarization
Conterbody
! 7 min ˜ 30min ˜ 7 min
EIS
1min26s 1 min
EIS
Sliding End of
Sliding
3 min ( - 900 )
20 min (1000 )
Polarization control
at 1000 mV
Polarization control at 1000 mV Cathodic
polarization
Counterbody
! 30min ! 7 min
EIS
1min26s 1 min
EIS
Sliding End of
Sliding
3 min 20 min
(1V) (-0.9 V)
Fig. 3.2. Schematic representation of the electrochemical experiments done under potentiostatic
control conditions.
The electrolytes used were artificial saliva (AS) solutions with some additives (see
table 3.1). Citric acid was added to AS to try to understand how the changes in pH can
influence the repassivation phenomenon. In the same way, different kinds of inhibitors
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 65
were added to AS to see how the repassivation phenomenon can be affected by the
anodic and cathodic reactions. The corrosion inhibitors may be included in the
formulations of tooth cleaning agents or medicines. All experiments were performed at
room temperature.
Table 3.1: Chemical composition of the artificial saliva solutions used (wt%).
Solution
Compound
Artificial saliva (AS)
AS
+
citric acid
AS
+
anodic inhibitor
AS
+
organic inhibitor
Sodium Chloride, NaCl 0.70 0.70 0.70 0.70
Potassium Chloride, KCl 1.20 1.20 1.20 1.20
Citric Acid, C6H8O7.H2O 0.025
Sodium Nitrite, NaNO2 0.16
Benzotriazole, C6H5N3 1.5
pH 5.5 3.8 5.5 5.5
After the mechanical damage the samples were inspected using the SEM (JEOL
JSM-610F) and EDS (Noran Voyager) equipments. The AFM (Digital Instruments
Vecco Metrology Group) technique was used to evaluate the surface topography of the
samples after the tribocorrosion test.
3. Results and Discussion
3.1. Open circuit potential (OCP) conditions
- Tribocorrosion behaviour
In Fig. 3.3 (a) the evolution of the corrosion potential (Ecorr) over time, during the
tribocorrosion tests, in the different AS solutions, is presented. As already stated (see
Fig. 3.1), before the beginning of the mechanical damage, a stabilization period was
reached in order to try to obtain a stable passive film. A detail of the evolution of the
potential during the stabilization period can be observed in Fig. 3.3 (b). Some
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
66
differences could be detected between the Ecorr values reached in the different
electrochemical solutions. AS + citric acid solution is the solution who reached the
noblest Ecorr value (approximately 0.05 V vs. SCE), in contrast with the AS + organic
inhibitor, which presents the lower Ecorr values (- 0.24 V vs. SCE). The Ecorr value
reached with AS solution was approximately – 0.15 V vs. SCE and with AS + anodic
inhibitor was – 0.09 V vs. SCE. This could suggest some Ecorr dependency on the
chemical composition of the solutions and/or pH.
After stabilization, at the beginning of the sliding, a sharp drop in Ecorr is observed
(see Fig. 3.3 (a)). In fact, Ecorr values decrease to less-noble values indicating the
depassivation of the surface and the subsequent contact of the fresh active titanium
surface with the electrochemical solutions [8,17].
In Fig. 3.3 (b) it can be also observed in detail the corrosion potential evolution
during sliding. In the AS solution, the potential had reached lower values when
compared to the other solutions. This could indicate a higher corrosion susceptibility of
metallic Ti samples when it is in contact with the AS solution, under sliding
solicitations. Nevertheless, in all solutions, Ecorr values tend to decrease during the
mechanical damage. It is important to point out that the total duration of the mechanical
damage was 86 s. In a tribological test such short time should be in the running-in
period [8]. In a previous work [8], with similar test conditions (Ti sample immersed in
AS solution, but in fretting-corrosion tests and lower normal load) a running-in period
with approximately 2000 s was observed.
At the end of mechanical damage (Fig. 3.3 (a)), an increase of Ecorr, up to a steady-
state condition was observed, indicating the probable restoration of the passive film in
the areas where it was removed, i.e., the surface repassivation [17,19]. This steady- state
is reached approximately after 1200 s. Ecorr values recovering its original value
presented before the test. Again, Ecorr dependency on the chemical composition of the
solutions and/or pH influence can be suggested. Additionally, the increase in Ecorr
values, after the mechanical damage, reached to AS + anodic inhibitor and AS + organic
inhibitor solutions might indicate no efficiency of the corrosion inhibitors, at the
concentrations used in this work. The anodic inhibitor should inhibit the anodic reaction
and the organic inhibitor should inhibit the anodic and cathodic reactions [8].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 67
0 250 500 750 1000 1250 1500 1750 2000-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
Beggining of Reciprocating Sliding
End of Reciprocating Sliding
E (
V v
s.
SC
E)
Time (s)
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
0 100 200 300 400
-1,0
-0,9
-0,8
-0,7
-0,6
-0,5
-0,4
-0,3
-0,2
-0,1
0,0
0,1
0,2Stabilization
period
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
E (
V v
s.
SC
E)
Time (s)
(a) (b)
Fig. 3.3. Evolution of open-circuit potential values during the tribocorrosion tests, for AS
solutions, during: a) all tribocorrosion test; b) the first seconds of the tribocorrosion tests.
Reciprocating sliding parameters: 10 N, 1 Hz, 6 mm of stroke.
- Characterization of the passive film
In Fig. 3.4 (a) and (b), Ti worn surfaces obtained after the mechanical damage, are
presented. Tribolayers can be observed on the worn surface of Ti samples. Tribolayers
are formed with the detached particles (wear debris) which consequently are compacted
in the surface, i.e., these particles are reintegrated on the surface as a smeared layer [7-
9,12]. Additionally, in accordance with the SEM micrograph presented in Fig. 3.4 (b)
and with EDS spectrum presented in Fig. 3.4 (b), oxidized Ti was detected in the worn
metallic surfaces. The oxidized material found seems to be combined with the
tribolayers. This could suggest a mechanical mixed layer constitute by wear debris and
smearing material with oxidized Ti [7-9], contributing also to the ennoblement of the
Ecorr after the mechanical damage (see Fig. 3.3 (a)). Thus, the Ecorr increases after the
mechanical damage additionally indicates that the newly formed passive film, after a
total removal of the naturally formed film by the mechanical action, together with the
mechanical mixed layer has quite similar characteristics as the naturally formed film
presented on Ti before mechanical action [8]. In both SEM micrographs (Fig. 3.4 (a)
and (b)), it is also possible to identify the wear marks aligned in the mechanical damage
direction. It is important to point out that all the samples tested in the different AS
solutions presented similar worn surfaces.
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
68
6 µm
2 µm
EDS
(a) (b)
(c)
Fig. 3.4. SEM micrograph of the wear scar showing scratched tribolayers ((a) and (b)); EDS
spectrum (c). Reciprocating sliding parameters: 10 N, 1 Hz, 6 mm of stroke.
Additionally, wear debris, presented in Fig. 3.5 (a), were detected. Wear debris in the
contact zone forms a third body, and this can influence the mechanical wear mechanism
and the effective contact pressure [7-9,12]. In Fig. 3.5 (b), an AFM image is presented,
where high plastic deformation of Ti worn surfaces can be observed. This phenomenon
is typical when a hard body leads to predominant abrasive wear of the softer material
[20,21].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 69
2 µm
1
2
(a) (b)
Fig. 3.5. a) SEM micrograph of the wear scar showing wear debris; b) AFM micrographs about
the border of the wear scar: 1) the inside zone of the wear scar; 2) the outside zone of the wear
scar. Reciprocating sliding parameters: 10 N, 1 Hz, 6 mm of stroke.
In order to obtain a more complete characterization of the passive film resulting
from the repassivation of the Ti worn surface, EIS experiments were performed before
and after the tribocorrosion tests. In Fig. 3.6 (a) Nyquist plot, obtained with the EIS
experiments, in OCP conditions is presented. It is important to point out that only
results with AS solution and with AS + citric acid are presented, because all the other
solutions presented similar trends to that obtained in AS.
In Nyquist plots (Fig. 3.6 (a)), both AS solutions show only one depressed
semicircle, i.e., a semicircle with an open end at low frequency. This trend indicates a
pure capacitive behaviour of the passive film and high corrosion resistance.
Furthermore, the semicircle does not have significant changes in development, so it can
indicate that the oxide passive film formed on the Ti surface might have high stability
under the experimental conditions [22]. The differences observed in Fig. 3.6 (a),
between the solutions, will be analysed in detail when the resistance of the passive film
is discussed.
Additionally, a simple Randles equivalent circuit was found to describe the EIS
behaviour (chi-square error ≈ 0.3 ± 0.11 %) being presented in Fig. 3.6 (b). The electric
compounds of the circuit are:
- R� represents the electrolyte solution resistance,
- Cp is the constant phase element or capacitance element,
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
70
- Rp is the polarization resistance [23,24].
0 3x104
6x104
9x104
1x105
0
1x105
2x105
3x105
4x105
5x105
6x105
Zreal
(Ohm.cm2)
Zima
g (
Oh
m.c
m2
)
Artificial saliva (AS) AS + citric acid
Before mechanical damage After mechanical damage
(a) (b)
Fig. 3.6. a) Nyquist plot obtained in AS and AS + citric acid solutions tested before and after the
tribocorrosion test, in OCP conditions; b) Equivalent electric circuit used to fit the EIS
experimental data.
The estimated polarization resistance (Rp) and passive film capacitance (Cp) are
presented in Fig. 3.7 (a) and (b), respectively. Regarding the Fig. 3.7 (a), higher Rp
values are achieved after the mechanical damage. This could be due to the tribolayers
presence, mainly constitute by smeared oxidized material, in accordance with results
shown in Fig. 3.4 and Fig. 3.6 (a). Also, comparing AS with AS + citric acid solutions,
it is possible to observe that AS solution presents higher Rp values, before and after the
mechanical damage.
Concerning the passive film capacitance (Cp), presented in Fig. 3.7 (b), no
differences were observed in AS solution, before and after the mechanical damage. The
same tendency is observed with AS + citric acid. However, AS solution presents lower
Cp when compared with AS + citric acid solution, before and after the tribocorrosion
tests. Additionally, and in accordance with several authors [25-27], the relation between
the Cp and the thickness of the passive film, for TiO2 film may be estimated by:
A
C
rd
p
0!!
= (Eq. 1)
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 71
where r is surface roughness factor, � is the dielectric constant and � 0 is the
permittivity of vacuum, A is the exposed area, Cp is the capacitance and d is the
thickness of the passive film. As all the values are constant, with the exception of d and
Cp, the following assumption can be done:
!=
A
C
rd
p
0""
pCd
1! (Eq. 2)
Taking this relation into account and analysing Fig. 3.7 (b), AS solution seems to
provide thicker passive films, before and after the tribocorrosion tests. This could
suggest a detriment effect of pH decrease on the Ti corrosion resistance. The pH
decrease seems to have an inadequately effect in the improvement of the passive film
properties. However, the addition of the corrosion inhibitors to AS, do not promote any
effect on the passive film properties.
Before sliding After sliding0,0
2,0x106
4,0x106
6,0x106
8,0x106
1,0x107
AS
AS
AS +
citric acidRp
(O
hm
.cm2
)
AS +
citric acid
Before sliding After sliding0,0
4,0x10-6
8,0x10-6
1,2x10-5
1,6x10-5
2,0x10-5
2,4x10-5
2,8x10-5
AS AS
AS +
citric acid
AS +
citric acid
CPF
(F
.cm-2
)
(a) (b)
Fig. 3.7. Polarization resistance (Rp) (a) and passive film capacitance (Cp) (b) obtained before
and after the tribocorrosion test, in OCP conditions.
3.1.1. Repassivation evolution with time analyzes, in OCP conditions
Repassivation can be defined as a nucleation and growth process of a new passive
film on the fresh active metal surface. The pioneering works of Sato and Cohen [c.f 28]
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
72
and Cabrera and Mott [c.f. 28], described the formation of passive film on metallic
surfaces. According to the place exchange model proposed by Sato and Cohen, a layer
of oxygen is adsorbed onto the surface and then exchanges places (possibly by rotation)
with the underlying metal atoms. A second layer of oxygen is adsorbed and the two M-
O pairs rotate simultaneously. This process is repeated resulting in the formation of an
oxide film [28].
The repassivation evolution estimated in OCP conditions, has been studied by
several authors [11,13,15,16,29]. It was stated that, when the mechanical damage is
interrupted, the open-circuit potential values increases, according to the logarithmic law:
∆E = bE log(t) + k (Eq. 3)
where t (s) is the time after the interrupted abrasion, bE is the slope and k is a constant.
Thus, repassivation evolution with the time can be calculated by determining bE, i.e., the
slope of the E-log(t) curve. High bE values are related to larger increases of potential at
the initial stage of repassivation.
In Fig. 3.8 (a), the repassivation trends obtained with OCP conditions are presented.
The same graph is presented in Fig. 3.8 (b), although with logarithmic time scale, to
allow the calculation of bE value.
0 4000 8000 12000-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
E (
V v
s S
CE
)
Time (s)
Artificial Saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
100 1000 10000
-1,0
-0,8
-0,6
-0,4
-0,2
0,0
0,2
E (
V v
s S
CE
)
Log time (s)
Artificial Saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
(a) (b)
Fig. 3.8. a) Repassivation trends obtained during the tribocorrosion test in open circuit
conditions; for all the artificial saliva solutions; b) repassivation trends in log time scale.
Reciprocating sliding parameters: 10 N, 1 Hz, 6 mm of stroke.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 73
The bE values obtained from the repassivation trends plotted in Fig. 3.8 (b) are
presented in table 3.2. As it can be observed, the estimated bE values suggest higher bE
value is achieved, in AS + citric acid when compared with the other solutions. This may
be an indication about the effect of pH decrease under OCP conditions, i.e., under these
conditions, the pH decrease seems to have a helpful effect in the repassivation
evolution. The bE values calculated for the AS + anodic inhibitor and AS + organic
inhibitor, are similar to those observed in the AS solution, indicating that the addition of
inhibitors with AS solution does not have any beneficial effect on the repassivation
evolution of the Ti. This is in accordance with the increase in corrosion potential, after
the mechanical damage, presented in Fig. 3.3, to the solutions with corrosion inhibitors.
Table 3.2: Repassivation evolutions obtain in OCP conditions. Reciprocating sliding
parameters: 10 N, 1 Hz, 6 mm of stroke.
Solutions
∆E = bE∗ log(t) + k
(r2 ≈ 0.99 ± 0.004)
Artificial saliva (AS) 0.23
AS + citric acid 0.44
AS + anodic inhibitor 0.20
AS + organic inhibitor 0.18
3.2. Potentiostatic control conditions
- Tribocorrosion behaviour
In Fig. 3.9 the anodic polarisation curves obtained in cp Ti in contact with the AS
solutions are presented. The E(i=0) in these curves varies between - 0.64 V and -0.31 V
vs. SCE. The polarization curves were obtained in order to select a potential value to
promote the controlled formation of the passive film. By applying electrochemical
polarization during wear experiments, a better control of the surface chemistry (passive
oxide film condition) of a metal can be obtained [7]. As referred before an in-situ
cleaning at – 0.9 V vs. SCE was performed prior to the tribocorrosion tests. The applied
potential during the sliding tests was 1 V vs. SCE.
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
74
1E-4 1E-3 0,01 0,1 1 10 100 1000
-1,0
-0,5
0,0
0,5
1,0
1,5
2,0
E
(V
vs S
CE
)
i (µA/cm2)
AS AS + citric acid AS + anodic inhibitor AS + organic inhibitor
Fig. 3.9. Potentiodinamic polarization curves of c.p. Ti obtained with AS solutions.
Fig. 3.10 (a) shows the evolution of corrosion current density (icorr) with the time
during the tribocorrosion tests performed under potentiostatic control. Before the
mechanical damage, a stabilization period was reached, and no significant differences
where observed. When the mechanical damage begins, the corrosion current density
values increase corresponding to the damage or destruction of the oxide passive film,
i.e, the depassivation of Ti samples surface. During the mechanical damage, a slight
increase in the corrosion current density can be observed in all solutions. This indicates
an increase of the corrosion rate during the mechanical action, probably promoted by
the constant exposure of fresh metallic Ti to the electrochemical solution, after each
sliding cycle. In Fig. 3.10 (b) a detail of the corrosion current density during the
mechanical damage is presented. Fluctuations in icorr evolution during the mechanical
damage, in all solutions, can be observed, suggesting passivation and depassivation of
the surface, in the tribo-activated worn surface [30,31]. At the end of the mechanical
damage, also presented in detail in Fig. 3.10 (c), the corrosion current density reached
the original values achieved before the tribocorrosion test indicating the surface
repassivation [18,19]. Some differences between the corrosion current density values
achieved in the different solutions could be observed. These differences results are in
accordance with Fig. 3.9, where differences in the icorr values could be noted at the
applied potential (1 V vs. SCE). AS solution is the solution who reached the lower icorr
value (-1 µA/cm2, i.e. cathodic current) when compared with the other solutions. AS +
anodic inhibitor reached the higher icorr value, approximately 0.07 µA/cm2. This could
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 75
indicate that the corrosion rate of the Ti depends on the electrochemical solution used.
Also, the decrease in icorr in all solutions could indicate corrosion protection.
0 250 500 750 1000 1250 1500 1750 2000
0
20
40
60
80
100
120
140
Beggining of Reciprocating Sliding
End of Reciprocating Sliding
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
i (µ
A/c
m2
)
Time (s)
(a)
60 80 100 120 1400
20
40
60
80
100
120
140
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
i (µ
A/c
m2
)
Time (s)
250 500 750 1000 1250 1500 1750 2000-2
-1
0
1
2
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
i (µ
A/c
m2
)
Time (s)
(b) (c)
Fig. 3.10. a) Evolution of corrosion current density during the tribocorrosion tests, in AS
solutions, under potentiostatic control; b) Detail of the corrosion current density, during the
mechanical damage; c) Detail of the corrosion current density, after the mechanical damage.
Reciprocating sliding parameters: 10 N, 1 Hz, 6 mm of stroke.
- Characterization of the passive film
In relation to EIS measurements promoted under potentiostatic control experiments,
the Nyquist plot obtained is shown in Fig. 3.11. Once again, only results with AS
solution and with AS + citric acid are presented, because all the other solutions
presented a behaviour similar to that observed in AS. According with Fig. 3.11, in both
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
76
solutions, pure capacitive behaviour of the passive film and high corrosion resistance
could be suggested [22].
0 1x105
2x105
3x105
4x105
5x105
6x105
0
2x105
4x105
6x105
8x105
1x106
1x106
Zreal
(Ohm.cm2)
Zima
g (
Oh
m.c
m2
)
Artificial saliva (AS) AS + citric acid
Before the mechanical damage After the mechanical damage
Fig. 3.11. Nyquist plot obtained in AS and AS + citric acid solutions tested before and after the
tribocorrosion test, under potentiostatic control conditions.
Regarding the fit of EIS data using the Randles equivalent circuit [23,24], with chi-
square error ≈ 0.2 ± 0.03 %, the Rp values obtained under potentiostatic control
conditions are presented in Fig. 3.12 (a). Analysing the Rp plot, higher passive film
resistance is achieved after the mechanical damage, in both solutions. Again, this fact
could be explained by the presence of tribolayers in the Ti worn surfaces. However,
some differences between the two solutions, after the mechanical damage could be
observed. AS solution seem to provides higher passive film resistance when compared
to AS + citric acid. Additionally, in Fig. 3.12 (b), a comparison of the passive film
resistance obtained in OCP conditions and in potentiostatic control conditions is
presented. It is possible to see that OCP conditions seem to provide higher passive film
resistance (Rp).
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 77
Before sliding After sliding0,0
5,0x105
1,0x106
1,5x106
2,0x106
AS
AS
AS +
citric acid
AS +
citric acid
Rp
(O
hm
.cm2
)
Before sliding After sliding0,0
2,0x106
4,0x106
6,0x106
8,0x106
1,0x107
Artificial Saliva (AS) AS + citric acid
OCP
1000 mV
1000 mV
OCP
R
p (
Ohm
.cm2
)
(a) (b)
Fig. 3.12. a) Passive film resistance (Rp) before and after the tribocorrosion test, under
potentiostatic control conditions; b) comparison between the Rp values obtained in OCP and in
potentiostatic control conditions.
The passive film capacitance (Cp) values, obtained under potentiostatic control
conditions, are presented in Fig. 3.13 (a). Regarding Fig. 3.13 (a), AS solution presented
a slightly lower Cp when compared with AS + citric acid solution, before and after the
tribocorrosion tests. In accordance with the relation between the Cp and the thickness of
the passive film [25-27] presented before, the AS solution appear to provides a slightly
thicker passive film, both before and after the tribocorrosion tests.
In Fig. 3.13 (b) the Cp values obtained in both electrochemical conditions, i.e., OCP
and potentiostatic control is presented. Observing in Fig. 3.11 (b), it is possible to say
that the passive films obtained with potential application at 1 V vs. SCE are thicker,
when compared to the passive films obtained in OCP conditions. This could suggest that
with the application of a passive potential on Ti samples, thicker films growth on metallic
surface. The OCP conditions provide thin films, although in accordance with Fig. 3.12 (b), it
seems to provide higher passive film resistance.
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
78
Before sliding After sliding0,0
2,0x10-6
4,0x10-6
6,0x10-6
8,0x10-6
1,0x10-5
1,2x10-5
AS AS
AS +
citric acid
CPF
(F
.cm-2
)
AS +
citric acid
Before sliding After sliding0,0
4,0x10-6
8,0x10-6
1,2x10-5
1,6x10-5
2,0x10-5
2,4x10-5
2,8x10-5
3,2x10-5 Artificial Saliva (AS)
AS + citric acid
1000 mV1000 mV
OCP
CPF
(F
.cm-2
)
OCP
(a) (b)
Fig. 3.13. a) Passive film capacitance (Cp) obtained, before and after the tribocorrosion test,
in potentiostatic control conditions; b) Cp obtained, before and after the scratch test, in both
electrochemical conditions used.
3.2.1. Repassivation analyses under potentiostatic control
Traditionally, the repassivation kinetics is studied using the empirical law [10-13]:
i = kt-bi ⇔ ln i = ln (k) + bi*ln (t) (Eq. 4)
where k is a constant and bi is the repassivation rate: large bi values indicate fast
repassivation [10,12]. bi is obtained by the calculation of the repassivation plateaus
slope (log(i) vs. log(t)) after the mechanical damage stops. In Fig. 3.12 repassivation
plateaus both in linear (Fig. 3.12 (a)) and in logarithmic (Fig. 3.12 (b) scale, are
presented.
In Fig. 3.12 (b), two regions could be distinguished: at lower t values, during
approximately 1 second (region A), the curve trend is characterized by a slow rate of
corrosion current density decrease during the time. However, at higher t values, in
region B, the rate of decay is much faster. Region B has approximately 8 seconds. This
change in the behaviour can be associated with the film nucleation (region A) followed
by its growth (region B). The passive film starts to thicken as soon as a monolayer
covers the surface [12]. Thus, the repassivation of Ti passive film, under potentiostatic
control in the passive region, probably occurred in two kinetically different processes:
the passive film is initially nucleated and then the growth of the film will occur [12].
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 79
0 100 200 300 400
0
20
40
60
80
100
120
140
Time (s)
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
i (µ
A/c
m2
)
1x10
-11x10
0
1x100
1x101
1x102
Region B
Artificial saliva (AS) AS + citric acid AS + anodic inhibitor AS + organic inhibitor
Log time (s)
Lo
g i
(µA
/cm2
)
Region A
(a) (b)
Fig. 3.12. a) Repassivation trends obtained during the tribocorrosion test with potentiostatic
control with linear (a) and log (b) scales. Reciprocating sliding parameters: 10 N, 1 Hz, 6 mm of
stroke.
In table 3.3, the repassivation rates - bi values - obtained from the repassivation
trends are presented. It is important to take into account that in bi calculation, the region
A in Fig. 3.12 (b) was not considered. When Ti is immersed in AS the solution, faster
repassivation is observed (higher bi value). The presence of additives (citric acid or
corrosion inhibitors) in AS solution seems to affect negatively the repassivation rate.
However, the difference between the bi achieved in AS and AS + citric acid is
approximately 13 % is not significant when compared with the difference obtained in
bE, between the both solutions (approximately 95 %), in OCP conditions. Additionally,
this could be suggested faster kinetics of the process in OCP conditions.
Table 3.3: Repassivation rate under potentiostatic control conditions.
Solutions
ln (i) = ln (K) + b∗ ln (t)
(r2 ≈ 0.99 ± 0.002)
Artificial saliva (AS) 1.49
AS + citric acid 1.31
AS + anodic inhibitor 1.27
AS + organic inhibitor 1.21
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
80
4. Conclusions
In this work, the repassivation evolution of cp Ti in artificial saliva solutions was
studied. Also, the influence of pH decrease and presence of inhibitors in AS solutions
was considered.
The corrosion potential and the corrosion current density of Ti, seems to be affected
by the presence of additives in AS solutions. Tribolayers namely constitute by smeared
material mixed with oxidized Ti, were founded in the Ti worn surfaces after the
mechanical damage.
In OCP conditions, the pH decrease, results in slight improvement of the
tribocorrosion behaviour of Ti. However, the passive film repassivated seems to be thin
(comparing with AS solution) and have lower resistance. Under potentiostatic control
conditions, the decrease in pH promotes a slightly reduction in the repassivation rate.
The anodic inhibitor and the organic inhibitor, when used in the experimental
conditions considered in this work, seem to be inept in the repassivation kinetics of Ti.
No significant changes where detected in the repassivation kinetics of the passive film,
in OCP conditions and under potentiostatic control conditions when compared to AS
solution.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 81
References
[1] T. Hanawa, K. Asami and K. Asaoka; Repassivation of titanium and surface oxide
film regenerated in simulated bioliquid; Biomed Mater Res, Vol. 40, (1998) 530-538
[2] C.E.B. Marino, L. H. Mascaro; EIS characterization of a Ti-dental implant in
artificial saliva media: dissolution process of the oxide barrier; Journal of
Electroanalytical Chemistry, Vol. 568 (2004) 115–120
[3] A.W.E. Hodgson, Y. Mueller, D. Forster, S. Virtanen; Electrochemical
characterisation of passive films on Ti alloys under simulated biological conditions;
Electrochemica Acta, 47 (2002) 1913-1923
[4] A A.K. Shukla, R. Balasubramaniam, S. Bhargava; Properties of passive film
formed on CP titanium, Ti–6Al–4V and Ti–13.4Al–29Nb alloys in simulated human
body conditions; Intermetallics, Vol. 13 (2005) 631–637.
[5] T. Hanawa; In vivo metallic biomaterials and surface modification; Materials
Science and Engineering, Vol. A267 (1999) 260–266
[6] D.J. Wood; The characterization of particulate debris obtained from failed
orthopaedic implants; Research report; San Jose State University, College of Materials
Engineering, 1993
[7] S. Mischler, A. Spiegel, D. Landlolt; The role of passive films on the degradation of
steel in tribocorrosion systems; Wear, Vol. 225 (1999) 1078-1087
[8] A.C. Vieira, A.R. Ribeiro, L.A. Rocha, J.P. Celis; Influence of pH and corrosion
inhibitors on the tribocorrosion of titanium in artificial saliva; Wear 261 (2006) 994-
1001.
[9] F. Assi; Tribo-Electrochemistry at a micrometer scale – measuring techniques,
tribocorrosion and repassivation; Dissertation submitted to the Swiss Federal Institute
of Technology.
[10] M. A. Barbosa; Passivation kinetics and pitting studies on Cr, Ni and an 18Cr-
10Ni Stainless Steel; PHD Theses, University of Leeds, Dep. of Metallurgy
[11] J.M. Abd El Kader, F.M. Abd El Wahab, H.A. El Shayeb and G.A. Khedr; Oxide
film thickening on titanium in aqueous solutions in relation to ainion type and
concentration; Dr. Corrosion Journal, 1981, Vol. 16, nº2
[12] S. Mischler, A. Spiegel, M. Stemp, D. Landolt; Influence of passivity on the
tribocorrosion of carbon steel in aqueous solutions; Wear 251 (2001) 1295–1307
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
82
[13] F. Contu, B. Elsener, H. Bohni; A study of the potentials achieved during
mechanical abrasion and the repassivation rate of titanium and Ti6Al4V in inorganic
buffer solutions and bovine serum; Electrochimica Acta 50 (2004) 33–418
[14] A.M. Al-Mayouf, A.A. Al-Swayih, N.A. Al-Mobarak, A.S. Al-Jabab; Corrosion
behavior of a new titanium alloy for dental implant applications in fluoride media;
Materials Chemistry and Physics 86 (2004) 320–329
[15] G. T. Burstein and R. J. Cinderey; Evolution of the corrosion potential of
repassivating aluminium surfaces; Corrosion Science, Vol.33, Issue 3 (1992) 475-492
[16] G.T. Burstein, R.M. Organ; Repassivation and pitting of freshly generated
aluminium surfaces in acidic nitrate solution; Corrosion Science 47 (2005) 2932–2955
[17] S. Barril, S. Mischler, D. Landolt ; Influence of fretting regimes on the
tribocorrosion behaviour of Ti6Al4V in 0.9 wt.% sodium chloride solution; Wear, Vol.
256 (2004) 963–972
[18] L. Benea, P. Ponthiaux, F. Wenger, J. Galland, D. Hertz, J.Y. Malo;
Tribocorrosion of stellite 6 in sulphuric acid medium: electrochemical behaviour and
wear; Wear 256 (2004) 948–953
[19] P. Jemmely, S. Mischler, D. Landolt; Electrochemical modelling of passivation
phenomena in tribocorrosion; Wear 237 (2000) 63–76
[20] D. Landolt, S. Mischler, M. Stemp; Electrochemical methods in tribocorrosion: a
critical appraisal; Electrochimica Acta 46 (2001) 3913–3929
[21] Masahiro Seoa and Yusuke Kurata; Nano-mechano-electrochemical properties of
passive titanium surfaces by in-situ nano-identation and nano-scratching;
Electrochimica Acta, Vol. 48, p.3221-3228, 2003
[22] S. Hiromoto, T. Hanawa, Re-passivation current of amorphous Zr65Al7.5Ni10Cu17.5
alloy in a Hanks’ balanced solution; Electrochemical Acta, 47 (2002) 1343-1349
[23] J.E.G. González and J.C. Mirza-Rosca; Study of the corrosion behavior of titanium
and some of its alloys for biomedical and dental implant applications; Journal of
electrochemical chemistry, 471 (1999) 109-115
[24] R. Wen-Wei Hsu, C. Yang, C. Huang and Y. Chen; Electrochemical corrosion
properties of Ti-6Al-4V implant alloy inn the biological environment; Materials Science
and Engineering, vol. A 380, p. 100-109, 2004
[25] C. Liu, Q. Bi, A. Matthews; Corrosion Science, 43 (2001) 1953-1961.
[26] Leonardo M. Da Silva, Karla C. Fernandes, Luiz A. De Faria, Julien F.C. Boodts;
Electrochemical impedance spectroscopy study during accelerated life test of
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 83
conductive oxides: Ti/(Ru + Ti + Ce)O2-system; Electrochimica Acta 49 (2004) 4893–
4906
[27] Maan Aziz-Kerrzo, Kenneth G. Conroy, Anna M. Fenelon, Sinead T. Farrell,
Carmel B. Breslin; Electrochemical studies on the stability and corrosion resistance of
titanium-based implant materials; Biomaterials 22 (2001) 1531-1539.
[28] E. Cho, C. Kim, J. Kim, H. Kwon; Quantitative analysis of repassivation kinetics
of ferritic stainless steels based on the high field ion conduction model; Electrochemical
Acta 45 (2000) 1933–1942
[29] F. Contu, B. Elsener, H. Bohni; Corrosion behaviour of CoCrMo implant alloy
during fretting in bovine serum; Corrosion Science 47 (2005) 1863–1875
[30] P.-Q. Wu, J.P. Celis; “Electrochemical noise measurements on stainless stel during
corrosion-wear in sliding contacts”; Wear 256 (2004) 480-490
[31] J.P. Celis, P.-Q. Wu; Tribo-corrosion of metallic materials: active wear track
concept and electrochemical transients for in-situ analysis of material degradation under
sliding; 2nd World Tribology Congress, 2001.
Chapter 3 – Repassivation evolution of Ti in artificial saliva solutions under tribocorrosion conditions
Master Dissertation
84
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 85
Results Discussion
As point out before, the main scopes of this work were:
- Study the wear and electrochemical mechanisms promoted by combined
action of both. When the dental implant is simultaneously exposed to
mechanical stress (promoted, for instances by the mastication loads or other
mechanical solicitations) and chemical degradation (promoted by the
physiological environment that surrounds the implant), it becomes part of a
tribocorrosion system. In a tribocorrosion system, the two mechanisms do
not proceed separately, and will depend on each other in a complex way.
Normally corrosion is accelerated by wear and, similarly wear may be
affected by corrosion phenomena [1,2].
- Study the tribocorrosion mechanisms in different tribological
arrangements, this is, under fretting conditions and under reciprocating
sliding conditions. Fretting-corrosion was used in order to simulate cyclic
micro-movements at the implant/bone interface or implant/abutment
interface during long time solicitations (mastication loads), after the
insertion of the implant. Reciprocating sliding tests with artificial saliva as
electrochemical media were used to simulate the partial removal or even
total destruction of the passive film, naturally growth on implant metallic
surface, during the inserption of the implant into the bone.
- Study the pH influence. In vivo variations of pH, are normally related with
allergy and toxic reactions. For instances, during the insertion of the dental
implant, the pH can decrease from 7.35–7.45 to 5.2 in the hard tissue. The
Chapter 4 – Results Discussion
Master Dissertation
86
problems associated with pH variations can occur if the metallic
biomaterials become corroded [3,4].
- Understand how the corrosion inhibitors can affect the tribocorrosion
process. It would be interesting if corrosion inhibitors could be included in
the formulations of tooth cleaning agents or medicines intended for patients
holding dental implants or restorations. Thus, the study of corrosion
inhibitors efficiency, in tribocorrosion conditions, promotes an attractive
topic.
- Study the repassivation evolution after the destruction of the passive film.
The corrosion resistance of Ti is essentially promoted by the presence of a
stable oxide passive film, on its surface [5-7]. However, this passive film can
be scratched or destroyed during insertion and implantation into hard tissue
by abrasion with bone or with other materials. Interest in repassivation
study of Ti dental implant material, is due to the particular chemical and
mechanical behavior of these metals once the oxide protective film is
removed.
All experimental results presented in chapter 2 and chapter 3, were performed in
duplicate and were validated. However, to facilitate the interpretation, only one
experimental result, for each solution, was presented. It is also essential to explain
some important experimental decisions. In chapter 2, fretting-corrosion was studied.
However, in chapter 3, the tribological arrangement used was reciprocating sliding. As
stated before, the passive film presented in the dental implant can be scratched or
destroyed during implantation. Also, in accordance with Fig. 1.11 (Chapter 1, page 22),
during implantation the dental implant is subjected to hard wear movements, which
suggest amplitudes of wear higher than some micrometers, that is, reciprocating
sliding.
In fretting-corrosion tests (chapter 2), the normal load used was 2N, and in
reciprocating sliding (chapter 3), the normal load used was 10 N. The chose of different
normal load was based on the main aim on each work: in fretting-corrosion tests the
idea was to simulate solicitations during mastication. In the second experimental work,
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 87
in reciprocating sliding conditions, the idea was to study the material after the
destruction of the passive film, which may happen during the insertion of the implant
into the bone. So, different normal load were used in order to keep the difference
between the two solicitations considered.
1. Wear and/or electrochemical mechanisms promoted by combined
action of both
1.1 Electrochemical mechanisms
The electrochemical phenomena during the tribocorrosion tests, in OCP conditions
identified in both tribological arrangements (fretting and reciprocating sliding) are
illustrated in Fig. 2.6, chapter 2, page 50 and Fig. 3.3, chapter 3, page 67. These
phenomena are:
- Before the start of the tests, the samples were immersed in the different
electrolytes to reach stabilization. The tests only started after the stabilization was
achieved. Ecorr dependency on the chemical composition of the solutions and/or pH
influence is suggested.
- A sharp drop in potential was observed immediately after the start of the
mechanical action. Ecorr values decrease to less-noble values indicating the
depassivation of the surface, i.e., the creation of a new fresh metallic surface and its
exposition to the electrolyte.
- The evolution of Ecorr during the mechanical damage, can not be compared
between the tribocorrosion arrangements, because in reciprocating sliding tests (chapter
3), the duration of tests was to short. In reciprocating sliding tests only a part of the
running-in period was monitored.
- At the end of mechanical damage, the increase of Ecorr to more noble values
indicates the restoration of the passive film on the areas where it was removed by
friction. The Ecorr ennoblement can also be due to the presence of a tribolayer mainly
constitute by smeared wear debris and oxidized material, which was on the worn
surfaces, after the mechanical damage.
Chapter 4 – Results Discussion
Master Dissertation
88
Under potentiostatic control test, in reciprocating sliding, the evolution of corrosion
current density (icorr) with the time during the tribocorrosion tests was: first, a
stabilization period; then, at the beginning pf the mechanical damage, the corrosion
current density values increase corresponding to the depassivation of metallic surface.
Fluctuations in icorr evolution during the mechanical damage were observed, suggesting
passivation and depassivation of the surface, in the tribo-activated worn surface. At the
end of the mechanical damage, the corrosion current density reached the original values
achieved before the tribocorrosion test indicating the surface repassivation.
Some properties of the passive films presented before and after the mechanical
damage, in reciprocating sliding tests, were studied using EIS experiments. In all
solutions, the passive films seem to present pure capacitive behaviour and higher
corrosion resistance after the mechanical damage. Additionally, it could be suggested
high stability of the passive film under the experimental conditions (see Fig. 3.6 and
3.11, Chapter 3, page 70 and 76). Also, in all the cases, thicker passive film were
obtaniend after the mechanical damage suggesting the presence of tribolayers after the
mechanical damage (see Fig. 3.14, Chapter 3, page 68). The existence of tribolayes on
the Ti worn surface results on the protection of the Ti surface form corrosion.
1.2 Mechanical mechanisms
The wear mechanism on the worn surface, under tribocorrosion conditions, is
essentially dependent on the applied normal load as well as the rubbing time. The wear
scar dimension, in fretting-corrosion tests (chapter 2), was about 200 µm. Also, 5000
and 10000 fretting cycles were performed, promoting high wear volumes. In
reciprocating sliding conditions (chapter 3), the dimension of the wear scar was about 6
mm. However, the rubbing time was very short promoting very low wear volumes (not
measured). Thus, no comparison can be made between the wear volume losses between
the different experimental conditions. Also, the coefficient of friction measured during
the reciprocating sliding conditions (chapter 3) was not presented because, as the
rubbing time was lower (86 seg), the coefficient of friction measured belongs to the
running-in period. This running-in period does not represent the real coefficient of
friction trend or the real average value. For this reason, no comparisons are made
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 89
between the coefficient of friction obtained in fretting conditions (chapter 2) and in
reciprocating sliding conditions (chapter 3). Also, no dissipated energy measurements
were made in reciprocating sliding conditions.
In both tribocorrosion systems, the wear scars obtained after the tribocorrosion
present severe material damage inside of the wear scar. In both cases, the presence of
wear debris was detected. Additionally the sliding wear direction can be easily
identified due to the wear marks. Wear debris in the contact zone, which forms a third
body, seems to be presented in both tribocorrosion systems. Build-up of debris
(tribolayers) formed with detached particles and smeared material, delaminations
attributed to the strong plastic deformation, extended plastic flow, plastic shear stress
and cracks were also observed.
Encrustations of alumina (provided from the counterbody) in Ti tribolayers as well
as smeared mixed alumina and Ti surfaces were only found in reciprocating sliding
conditions, suggesting the effect of a high applied normal load in tribological
phenomena.
Nevertheless, predominant abrasive wear of the softer material is suggested in
fretting and in reciprocating sliding conditions, because high plastic deformation on Ti
surfaces is observed in both cases. This type of wear is typical when a hard body rubs
against a softer material. All these phenomena could suggest that the Ti worn surfaces
formed and the tribocorrosion phenomena occurred, are independent of the applied
normal load as well as the duration of the mechanical damage.
2. pH decrease influence
After the mechanical damage, in both tribocorrosion arrangements, the corrosion
potential recovers its original value of before the test. Additionally, the corrosion
potential achieved with AS + citric acid solution was the highest. Thus, it is possible to
suggest that, independent of test duration and tribological arrangement used, citric acid
results in an improvement of the tribocorrosion behaviour of Ti. However, as shown in,
in accordance with EIS results (Fig. 3.7 and Fig. 3.13, chapter 3, page 71 and page 78)
Chapter 4 – Results Discussion
Master Dissertation
90
the passive film repassivated in this solution seems to be thin (comparing with AS
solution) and have lower resistance.
3. Corrosion inhibitors influence
With AS with organic inhibitor, after the mechanical damage, a recovery of the
corrosion potential is not observed after the fretting tests. However, in reciprocating
sliding tests (chapter 3) with lower rubbing times, the corrosion potential recovers of
before the test. Thus, AS with organic inhibitors presented different trends in fretting
and in reciprocating sliding tests. In fretting test, the organic inhibitor was efficient. No
recovering in the corrosion potential was noticed (Fig. 2.6, chapter 2, page 50)
suggesting inhibition of the cathodic and anodic reactions. This probably hinders the
formation of a new passive film. However, in reciprocating sliding conditions, AS +
organic inhibitor was not efficient. The corrosion potential recovering is an indication
about the passive film formation. In sum, the AS + organic inhibitor in the
concentrations used in these experimental works, seems to be affected by the test time,
i.e., the immersion time.
Regarding the AS with anodic inhibitor, in fretting conditions, this inhibitor
promotes a slight improvement of the tribocorrosion behaviour of Ti, probably due to
the nature of the oxidation and reduction reactions occurring in the contact region
during the tribocorrosion test. In reciprocating sliding conditions, the same trend was
observed. This could be suggest that the anodic inhibitors used in the concentrations
promoted in these experimental work, seems to be inefficient.
4. Repassivation evolution analyses
The repassivation phenomenon was studied, in different electrochemical conditions,
in chapter 3. The repassivation of the Ti samples after the mechanical damage in
fretting-corrosion conditions was not evaluated.
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 91
In both electrochemical conditions, the repassivation seems to be affected by the
electrolyte solution. It was suggested that the passive film repassivation process occurs
in two different steps: nucleation, followed by the growth of the passive film.
In the repassivation evolution study, in OCP conditions, the pH decrease was found
helpful. Higher repassivation evolutions were achieved with this solution. The addition
of corrosion inhibitors to AS do not promote any advantage on the repassivation
evolution of Ti, in OCP conditions.
Under potentiostatic control conditions, AS solution is the solution which promotes
with faster repassivation rate. The repassivation rate achieved with AS + citric acid, is
slightly lower than the one promoted by AS solution. The results obtained with AS with
corrosion inhibitors confirm the inefficiently action of these solutions in these
experimental results.
Chapter 4 – Results Discussion
Master Dissertation
92
References [1] P. Ponthiaux, F. Wenger, D. Drees, J.-P. Celis; Electrochemical techniques for
studying tribocorrosion processes; Wear, Vol. 256 (2004) 459–468
[2] S. Barril, N. Debaud, S. Mischler, D. Landolt; A tribo-electrochemical apparatus for
in vitro investigation of fretting–corrosion of metallic implant materials; Wear, Vol.
252 (2002) 744–754
[3] C.E.B. Marino, L. H. Mascaro; EIS characterization of a Ti-dental implant in
artificial saliva media: dissolution process of the oxide barrier; Journal of
Electroanalytical Chemistry, Vol. 568 (2004) 115–120
[4] A.W.E. Hodgson, Y. Mueller, D. Forster, S. Virtanen; Electrochemical
characterisation of passive films on Ti alloys under simulated biological conditions;
Electrochemica Acta, 47 (2002) 1913-1923
[5] A A.K. Shukla, R. Balasubramaniam, S. Bhargava; Properties of passive film
formed on CP titanium, Ti–6Al–4V and Ti–13.4Al–29Nb alloys in simulated human
body conditions; Intermetallics, Vol. 13 (2005) 631–637.
[6] T. Hanawa; In vivo metallic biomaterials and surface modification; Materials
Science and Engineering, Vol. A267 (1999) 260–266
[7] D.J. Wood; The characterization of particulate debris obtained from failed
orthopaedic implants; Research report; San Jose State University, College of Materials
Engineering, 1993
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 93
Chapter 5 – Final conclusions
Master Dissertation
94
Final conclusions
In this work, the influence of pH and corrosion inhibitors in artificial saliva on the
tribocorrosion behaviour of pure Ti was investigated. The tribological arrangements
considered to perform the tribocorrosion tests were: fretting and reciprocating sliding.
Additionally, the repassivation evolution of cp titanium in artificial saliva solutions with
different corrosion inhibitors as pH variations, were studied. The main conclusions of
this master dissertation are:
1) The addition of citric acid to artificial saliva results in a slight improvement of
the tribocorrosion behaviour of Ti. The pH decrease also has a helpful effect in
the repassivation evolution of cp Ti, under OCP conditions.
2) The addition of anodic inhibitor to artificial saliva solution results in
improvement of the tribocorrosion behaviour of Ti. The protection observed by
the addition of anodic inhibitor to artificial saliva is probably due to the nature
of the oxidation and/or reduction reactions occurring in the contact area during
the mechanical damage.
3) The addition of a cathodic inhibitor to artificial saliva at concentrations tested
in this work has a hazardous effect on the fretting–corrosion behaviour of
titanium.
4) In fretting conditions, the addition of an organic inhibitor to artificial saliva at
concentrations tested in this work, seems to inhibit the anodic and the catodic
reactions, promoting hinder of the passive film growth. However, in
reciprocating sliding conditions, the organic inhibitor is not efficient.
5) The repassivation phenomenon of cp Ti in artificial saliva solutions probably
occurs in two kinetically different processes: nucleation followed by the
growth of this film.
6) The repassivation evolution seems to be strongly affected by the electrolyte
solution. In OCP conditions AS + citric acid presented the best repassivation
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 95
evolution. However, under potentiostatic control conditions, the repassivation
rate of the Ti in the artificial saliva solution is faster than that observed in the
solutions with additives.
7) Predominant abrasive wear was detected. High plastic deformation on Ti
surfaces, tribolayers existence in the contact region during the tribocorrosion
test and formation of third bodies, were the principal tribological mechanisms
detected.
8) Tribo-chemical phenomena detected in the Ti worn surface were independent
of the applied normal load as well from the duration of the mechanical damage,
or the tribological configuration.
Chapter 5 – Final conclusions
Master Dissertation
96
Fretting-corrosion behaviour and repassivation evolution of Ti in artificial saliva solutions in the presence of corrosion
inhibitors and pH variations
Master Dissertation 97
Suggestions for future works
- Make a similar study but using other experimental conditions, like human
oral cavity temperature and more complex and real artificial saliva. The real
system is very complicated and the approximation of this real condition can
provide more complete information. For instances, the pH decrease studied
in this master thesis provides the same effect if the temperature in the system
approximately 23 ºC (oral cavity temperature).
- Evaluate how the micro-organisms presented in oral cavity can affect the
dental implants behaviour. In the oral cavity there are a wide range of
organisms, proteins, micro-organisms, etc. So evaluate the individual effect
of each organism in the degradation process of the dental implant, could
provide precious information about the live time of the dental implant.
- Evaluated the stechiometry as well as the thickness of the Ti oxides formed
in which artificial saliva used. The type of the Ti oxide formed in each case
can provide important information in the experimental results. The Ti oxides
stechiometry can be evaluated by X-Ray Photoelectron Spectroscopy (XPS)
and Auger technique.
- Promote in-vivo experiments.