ao meu marido e aos meus pais - universidade do minho ... · minha “maninha”, e aos meus pais,...

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].


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




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


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-




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].


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).




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].


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




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


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 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


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




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


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




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


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




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].


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.




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


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.




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


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)




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.


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:




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].


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].




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


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].




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


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].




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


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




[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


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




[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


Master Dissertation

38




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




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.


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.




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


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].




20004000

60008000

10000

-2

-1

0

1

2

0

50

100150

200

Ft (

N)

D (µm

)Number of Cycles


(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.


Master Dissertation

46

0 2000 4000 6000 8000 10000

0.5

0.6

0.7

0.8

0.9

1.0

AS + organic inhibitor


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].




(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


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




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,


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].


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,


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.




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.


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




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.


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.





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.


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


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.




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


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

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




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


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




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.


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




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


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


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].




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


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.


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].




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,


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)




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


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]


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.





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.


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




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


i (µ

A/c

m2

)

Time (s)

(a)

60 80 100 120 1400

20

40

60

80

100

120

140


i (µ

A/c

m2

)

Time (s)

250 500 750 1000 1250 1500 1750 2000-2

-1

0

1

2


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.


- 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


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).





5,0x105

1,0x106

1,5x106

2,0x106

AS

AS

AS +

citric acid

AS +

citric acid

Rp

(O

hm

.cm2

)


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.


Master Dissertation

78


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


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].




0 100 200 300 400

0

20

40

60

80

100

120

140

Time (s)


i (µ

A/c

m2

)

1x10

-11x10

0

1x100

1x101

1x102

Region B


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


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.




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



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.


Science and Engineering, Vol. A267 (1999) 260–266



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


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




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.


Master Dissertation

84




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,




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.


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




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)


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.




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.


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



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.


Science and Engineering, Vol. A267 (1999) 260–266



Engineering, 1993

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




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




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.

ao meu marido e aos meus pais - universidade do minho ... · minha “maninha”, e aos meus pais,...

Documents