these 170x240 cd - uliege.bebictel.ulg.ac.be/etd-db/collection/available/... · je tiens à...

Pr A. Vanheusden, University of Liège, Belgium,

Pr V. Migonney, Paris 13 University, France,

Dr M. Sadoun, Paris Descartes University, France,

President: Pr M. Lamy, University of Liège, Belgium,

Secretary:Pr E. Rompen, University of Liège, Belgium,

Members: Dr JP. Attal, Paris Descartes University, France

Pr J. Dejou, University of Marseille, France

Pr G. Leloup, Louvain Catholic University (UCL), Belgium

Pr E. Marchandise, Louvain Catholic University (UCL), Belgium

Promotors:

Co-promotor :

Jury :

AbstractResidual Stress in Veneering Ceramic I Amélie Mainjot 6

Residual stress measurement in veneering ceramic

The manufacture of dental crowns and bridges generates residual stresses within the

veneering ceramic and framework during the cooling process. Knowing the stress

distribution within the veneering ceramic as a function of depth can help the optimizing

of manufacturing processes and the understanding of failures, particularly chipping, a

frequent complication with zirconia-based fixed partial dentures.

The first objective of this work was to transfer and to adapt an effective industrial method,

the hole-drilling method, for measuring residual stresses to dental use, and to demonstrate

the method for measurement of residual stresses in veneer-metal (VM) and veneer-zirconia

(VZr) disks. The adapted method, presented in the earliest chapters, enables the very low

stresses measurement in comparison with industrial applications, notably due to the

development of a high sensitivity electrical measurement chain.

The second objective was to study the influence of cooling rate, veneer thickness, and

framework thickness on residual stress profile, and to compare measurements in VM and

VZr structures. Results described in the following chapters reveal that VM samples always

exhibited the same type of favorable stress vs. depth profile, starting with compressive

stresses at the ceramic surface, decreasing with depth, and then becoming compressive

again. VZr samples showed varying stress profiles, some describing a worse curve, where

the in-surface compressive stresses were found to turn into tensile stresses in the interior. A

chronological two-step approach is discussed to explain residual stress development in

metal-based samples,and a three-step approach,comprising the hypothesis of the phase

transformation, is proposed for zirconia-based samples. The results highlighted the difficulty

of defining the adequate ratio between veneer and zirconia framework thickness,and the

slow cooling procedures were shown to promote tensile stress development in VZr samples.

Keywords : Residual stress, Hole-drilling, Dental ceramic, Dental crowns, Zirconia,Veneering

ceramic

Abstract I Amélie Mainjot 7

Etude des contraintes résiduelles dans la céramique d’émaillage

L’émaillage des prothèses dentaires génère des contraintes résiduelles dans la céramique

d’émaillage et l’infrastructure durant le procédé de refroidissement. La connaissance de

leur distribution en fonction de la profondeur peut aider à comprendre les échecs,notamment

l’écaillage de la céramique cosmétique, une complication fréquente avec les prothèses

à infrastructure zircone.

Le premier objectif de ce travail a été d’adapter une méthode industrielle, la méthode

du trou incrémental, à l’usage dentaire, et de démontrer sa validité pour mesurer les

contraintes résiduelles dans des disques en métal (VM) ou en zircone (VZr) émaillés. La

méthode développée, présentée dans les premiers chapitres, permet de mesurer des

contraintes de très faible module notamment grâce au développement d’une chaîne

électrique de mesure très sensible.

Le second objectif était d’étudier l’influence de la vitesse de refroidissement, de l’épaisseur

de la céramique d’émaillage et de l’infrastructure, et de comparer les mesures dans les

échantillons VM et VZr. Les résultats décrits dans les chapitres suivants révèlent que les

échantillons VM ont toujours montré un profil qui commence par des contraintes en

compression à la surface, ces contraintes diminuant progressivement avec la profondeur,

puis ré-augmentant à nouveau. Les échantillons VZr ont montré des profils de contraintes

variables,certains décrivant une courbe plus défavorable,dans laquelle les contraintes en

compression à la surface deviennent des contraintes en tension en profondeur. Une

théorie en deux étapes est décrite pour expliquer le développement des contraintes dans

les échantillons métal, et une théorie en trois étapes, comprenant l’hypothèse de la

transformation de phase, est proposée pour les échantillons zircone. Le ratio adéquat

entre l’épaisseur d’émaillage et de zircone est difficile à définir, et les procédures de

refroidissement lent ne semblent pas bénéfiques pour les échantillons VZr.

Mots-clefs : Contraintes résiduelles, Trou incrémental, Céramique dentaire, Couronnes

dentaires, Zircone, Céramique d’émaillage

Résumé

Residual Stress in Veneering Ceramic I Amélie Mainjot 8

© 2011 Amélie MAINJOT

Department of Fixed Prosthodontics

Institute of Dentistry

University Hospital (CHU) of Liège, ULg

Quai G. Kurth 45

4020 Liège

Belgium

tel : + 32 4 270 31 00

fax : + 32 4 270 31 10

email : [email protected]

All right reserved. No part of this book may be reproduced or transmitted in any form or

any mean, electronic or mechanical, including photocopy, recording, or any information

storage and retrieval system, without permission in writing from the copyright owner.

Layout and graphics: www.perspectives.be

Amélie Mainjot 9

"If you don’t know where you are going,any road will get you there"

Lewis Carroll,

Alice's Adventures in Wonderland

À Bab et Sin2

AcknowledgementsResidual Stress in Veneering Ceramic I Amélie Mainjot 10

Je n’aurais pas pu trouver une pensée plus adaptée à ce travail, et aux dernières années

que j’ai vécues,que celle de Lewis Caroll.Quelle que soit la forme de recherche que l’on

entame, scientifique ou personnelle, souvent on ne sait pas où on va arriver, et comment

on va y arriver.Comme le Chat le fait remarquer à Alice : « si tu ne sais pas où tu vas,alors

tous les chemins peuvent t’y mener ». Je pense que c’est cela qui fait la richesse d’un

parcours.Mon chemin est passé pour la première fois par Paris il y a 5 ans déjà, lorsque j’ai

commencé le CES en Biomatériaux. Je ne savais pas à ce moment là que le CES

m’ouvrirait la voie du laboratoire de Biomatériaux de Paris Descartes dans lequel je

travaille depuis trois ans. Ces dernières années ont été extrêmement enrichissantes tant

professionnellement qu’humainement,cette thèse en est un des fruits.Elle n’aurait pas été

possible sans l’aide, le soutien et la contribution d’un grand nombre de personnes.

Tout d’abord, je tiens à rendre hommage au professeur Michel Degrange, directeur du

laboratoire de Biomatériaux, qui nous a quittés il y a à peine plus d’un an. Michel, tu m’as

transmis comme à beaucoup d’autres une partie de ta passion débordante pour les

biomatériaux. Tes qualités de pédagogue sont pour moi une référence inégalée. Tu m’as

accueillie à bras ouverts au CES puis au labo. Tes conseils, tes encouragements ont été

précieux, et je garderai en image la porte de ton bureau, que tu laissais toujours ouverte.

Ton amitié,ta chaleur,ton rire rauque qui résonnait au Montrouge,me manquent cruellement.

Ta disparition brutale nous laisse un goût amer, je n’aurais jamais imaginé que tu ne serais

pas là aujourd’hui.

Je tiens à remercier tout particulièrement le professeur Alain Vanheusden, le promoteur de

cette thèse, ainsi que le docteur Michaël Sadoun, le copromoteur,

Alain, que ce soit en tant que chef de service ou en tant que promoteur, tu m’as toujours

accordé un élément capital tout au long de notre parcours professionnel commun: ta

confiance. Je ne te remercierai jamais assez pour la liberté que tu m’as accordée, pour

ton soutien,pour ton amitié,ainsi que pour tout ce que tu as mis en œuvre pour que cette

thèse et mes déplacements soient réalisables pratiquement.

Michaël, Michel Degrange t’appelait, à juste titre, mon mentor. Tes compétences, ton

exigence, ton investissement et ta disponibilité, font de toi un directeur de thèse hors du

Acknowledgements I Amélie Mainjot 11

commun, très apprécié de ses doctorants. Cette thèse n’aurait jamais existé sans ton

expertise et tes connaissances dans le domaine de la prothèse dentaire mais aussi de

l’ingénierie et de la science des matériaux.Du CES à la thèse,tu m’as énormément appris.

Je tiens également à remercier tous ceux qui ont permis la mise en place de cette

cotutelle : le professeur Véronique Migonney,promoteur au sein de Paris 13,pour ses conseils

et sa disponibilité, le professeur Vincent Lorent, directeur de l’Institut Galilée, le professeur

Gérard Lévy, doyen de la Faculté de Chirurgie Dentaire de Paris Descartes, ainsi que le

docteur Jean-Pierre Attal.Jean-Pierre,tu sais que tu es mon meilleur impresario,merci pour

ton soutien inégalable quelles que soient les circonstances, merci pour ton optimisme

contagieux et ton amitié.

Merci au professeur Marc Lamy, qui a exercé à merveille ses fonctions de président. Marc,

merci pour tes messages d’encouragement, pour tout le temps que tu as consacré à

l’organisation de cette thèse, et pour ton amitié.

Il ne se le rappelle certainement pas car cela ne date pas d’hier, mais je remercie

chaleureusement le professeur Eric Rompen d’avoir été le premier à m’encourager à

entamer un travail de thèse, et surtout de m’avoir convaincue, à une époque où, chez

nous,aucune femme n’avait encore défendu sa thèse en médecine dentaire,que ce travail

ne serait pas incompatible avec mon statut de maman.Merci,d’avoir été et de rester une

locomotive pour notre équipe liégeoise, nous vous devons beaucoup.

Je remercie très sincèrement le professeur Gary Schajer,pour sa grande contribution à ce

travail. Gary, many thanks for your contribution to this work, your clear explanations, your

help regarding the editing of the different papers, your availability and your friendness

whatever the distance. I am looking forward to seeing you again in France or in Belgium.

Merci aussi aux différentes équipes qui collaborent avec nous dans le cadre de ce travail.

A l’Ecole Polytechnique de l’Université Catholique de Louvain : merci aux professeurs

Emilie Marchandise et Laurent Delannay,pour leur accueil, leurs conseils, leurs explications,

leur intérêt et leur motivation, ainsi que Catherine Lambrechts pour son très bon mémoire


de fin d’études. Merci à Leslie Peter de l’université de Paris 13, pour le travail effectué lors

de son stage au laboratoire. Merci au professeur Jérôme Chevalier, aux docteurs Laurent

Gremillard et Thierry Douillard, de l’unité de recherche Mateis UMR CNRS 5510, à l’INSA

Lyon. Jérôme, Laurent et Thierry merci pour votre intérêt et vos magnifiques images, merci

de nous faire bénéficier de votre extraordinaire expertise dans le domaine de la zircone.

Je remercie vivement les professeurs Jacques Dejou et Gaetane Leloup d’avoir accepté

de faire partie du jury de cette thèse.Je profite de l’occasion pour les remercier également

de leur investissement au sein de la société francophone en biomatériaux dentaires et

plus particulièrement dans la création de liens entre la France et la Belgique, afin de

promouvoir la recherche dans ce domaine.

Merci au CHU de Liège, qui a permis, encouragé et soutenu la réalisation de cette thèse.

Je remercie les professeurs Joseph Charpentier, Jean-Pierre Van Nieuwenhuysen

et William D’Hoore, qui m’ont soutenue et aidée dans mon premier projet de thèse.

Jean-Pierre et William, même si nous ne sommes pas arrivés là où nous souhaitions arriver,

merci d’avoir guidé mes premiers pas en recherche.

Du côté liégeois, merci à tous mes collègues et amis de l’Institut de Dentisterie, ainsi qu’à

tout le personnel PATO. Merci de m’avoir encouragée tout au long de cette thèse, de ne

pas m’en avoir trop voulu de vous abandonner pour ma vie parisienne,d’avoir pris de mes

nouvelles régulièrement. Merci au personnel de l’accueil et à Nathalie de m’avoir aidée

à gérer mes agendas abracadabrants, merci à Mme Seghers et ensuite à Patricia qui se

sont occupées de tous mes frais de déplacement. Merci à France pour son coaching

efficace et ses bons plans.Un merci particulier à mes collègues du service de prothèse fixe

pour leur soutien: Thomas, Florence, Charlotte,Vinciane, pour son travail dans le cadre de

l’étude rétrospective sur les prothèses en zircone,ainsi que Sylvie,pour l’organisation.Merci

à Camille et Marine de m’avoir efficacement assistée le peu de temps qu’il me restait

pour soigner mes patients. Merci à ces patients d’avoir été compréhensifs et...patients !

Merci aux céramistes, Jean-Michel et son papa, à qui je rends hommage, ainsi que Mirko,

pour leur collaboration.

Acknowledgements I Amélie Mainjot 13

C’est un plaisir de travailler avec vous tous, je suis extrêmement fière de mon université

mère.

Toujours du côté liégeois, merci à tous mes amis et amies que j’ai délaissés ces derniers

temps. Merci pour leur compréhension et leur soutien, et mille excuses pour les soupers

reportés, les soirées et les matchs de tennis ratés, et l’agenda overbooké. Je sais que je

peux toujours compter sur votre amitié, quoi qu’il arrive. Merci à celles qui ont partagé

avec moi les côtés blancs comme les côtés noirs de ces trois dernières années, je vous en

suis extrêmement reconnaissante.

Du côté parisien, merci à tous les collègues et amis du laboratoire de Biomatériaux, qui

m’ont offert un deuxième chez moi: Hélène et ses bons plans restos, Stéphane et son

dictionnaire franco-belge,Claudine,miss zircone 2,Elizabeth, le meilleur tour operator pour

congrès à petits prix,Jean-François, le seul dans le labo qui apprécie ma musique ringarde

et « l’autre » Hélène, la seule qui ne se moque jamais de mon accent. Bon d’accord, je

vous ai amadoués à coups de chocolat Galler et de bières belges,mais votre accueil m’a

donné le courage d’affronter les aller-retour en thalys et les hôtels miteux de la porte

d’Orléans. Un merci spécial de Superfrite à Mâââthieu, mon Supercamembert, anti-Mac

mais tellement pro, merci pour ton amitié, et ta disponibilité en cas de souci technique.

Merci au professeur Dorin Ruse. Dorin, merci pour ton soutien et tes conseils, merci encore

d’avoir réouvert la porte.

Merci à Frédo,Lulu, la meilleure co-shoppeuse, Boris,Julien,Romain,Stéphane,Marc,Thac,

Nicolas et Cédric.Merci de m’avoir adoptée,merci pour tous ces moments fous que nous

avons passés et que nous passerons encore ensemble. Merci pour votre amitié, merci de

m’avoir accompagnée à Istanbul et maintenant à Liège, vous m’avez énormément

touchée. Un énoooooorme merci à Boris, mon 3D master transportable, réalisateur des

animations pour mes présentations, dépanneur informatique de midi à minuit.

Merci à Meredith, pour son English support de dernière minute, ainsi qu’à Gérard Scrève,

pour son professionnalisme au niveau du graphisme.

Pour terminer, je voudrais remercier le papa de mes enfants, tous mes proches, ma famille

et mon compagnon de route. Sans leur complicité et leur soutien, cette expérience

n’aurait même pas été envisageable,

J’adresse une pensée très émue à mes grands-parents, à L’Amie et à tous ceux que je

chérissais qui se sont arrêtés au bord de la route. J’aurais aimé que vous puissiez être fiers

aujourd’hui, merci de m’avoir emmenée le plus loin possible,

Bab et Sin2, j’ai souvent pensé à la manière dont je vous exprimerais mes remerciements

aujourd’hui mais je n’ai pas réussi à trouver des mots à la hauteur de ce que je ressens.

Merci de contribuer à l’épanouissement de vos enfants quel que soit le chemin qu’ils

choisissent,merci de votre présence auprès de vos petits enfants.De la garde alternée du

chien et du chat, à la surveillance de la maison: sans votre soutien logistique, sans votre

disponibilité, sans l’aide aussi d’Angela que je remercie aussi au passage, mon exil

hebdomadaire et ma carrière professionnelle seraient quasiment impossibles. Je sais que

je peux inconditionnellement m’appuyer sur vous.Et cela n’est rien encore en comparaison

avec votre soutien moral, particulièrement dans les moments difficiles que j’ai traversés

ces trois dernières années. Vous êtes mon meilleur bâton : je vous dédicace la plaque

d’Adhémar,

Mes poussins, Justine et Benjamin, merci de votre patience, jamais vous ne vous êtes

plaints,que ce soit de me voir balader ma valise,d’être vissée à mon ordinateur,ou de ne

pas être disponible cet été.Je vous souhaite des passions,dans la vie,dans votre futur travail,

et je vais vous dire un secret : ma plus grande passion, c’est vous…


Table of contentsTable of contents I Amélie Mainjot 15

1. Introduction1 From Veneered Metal to Zirconia

1.1 Metal-based Restorations: the Porcelain-Fused-to Metal concept1.1.1 Components

- Alloys- Veneering ceramics

1.1.2 Keysteps of the procedure- Framework sandblasting- Framework oxidation firing- Opaque firing

1.1.3 Ceramo-Metal bonding mechanisms- Physico-chemical- Chemical- Mechanical

1.1.4 Clinical background: PFM restorations as a gold standard?1.2 Zirconia-based Restorations

1.2.1 An alternative to metal1.2.2 Zirconia’s unique property: the tetragonal-monoclinic transformation

- Transformation toughening- Low Temperature Degradation

1.2.3 The veneering concept of zirconia frameworks- Framework pretreatment- Ceramo-Zirconia bonding

1.2.4 Clinical background: the chipping problem2 How to study and predict the mechanical behavior of the veneering ceramic?

2.1 Traditional in vitro mechanical tests2.2 Residual stress analysis

2.2.1 Residual stress: Principle and interest2.2.2 Existing data relating to dental restorations2.2.3 Residual stress profile measurement methods

- Diffraction methods- Mechanical methods

A. Hole-drilling methodB. Slitting method

2 Objectives and strategy

18191920202123232323232323242427272828303131313237373838394041414243

47

3 Method selection and development1 Method selection 2 Method development

2.1 Specimen preparation2.2 Strain gage rosette2.3 Electrical measurement chain

2.3.1 Operating principle of strain gages and Wheastone bridges2.3.2 From standardized Wheatstone Bridge to customized Wheastone bridge2.3.3 Nanovoltmeters and filters

2.4 Temperature control2.5 Hole-drilling2.6 Strains measurements, residual stress calculation

4 Residual stress measurement in veneering ceramic by hole-drilling1 Introduction2 Materials and methods

2.1 Specimen preparation2.2 Strain gage rosette installation2.3 Electrical measurement chain2.4 Temperature control2.5 Hole-drilling2.6 Strain measurement and residual stress calculation

3 Results3.1 Measured strains3.2 Calculated residual stresses

4 Discussion5 Conclusions

5 Influence of cooling rate on residual stress profile in veneering ceramic: Measurement by hole-drilling

1 Introduction2 Materials and methods

2.1 Specimen preparation2.2 Determination of temperature profiles2.3 Dilatometric analysis2.4 Hole-drilling method

2.4.1 Strain gage rosette installation2.4.2 Electrical measurement chain2.4.3 Temperature control2.4.4 Hole-drilling2.4.5 Strain measurements and residual stress calculation

3 Results4 Discussion5 Conclusions


505254555656575960606061

6466696970717272737474757678

808284848686868687878788899397

Table of contents I Amélie Mainjot 17

6 Influence of veneer thickness on residual stress profile in veneering ceramic: Measurement by hole-drilling1 Introduction2 Materials and methods

2.1 Specimen preparation2.2 Hole-drilling method



7 Influence of zirconia framework thickness on residual stress profile in veneering ceramic : Measurement by hole-drilling

1 Introduction2 Materials and methods

2.1 Specimen preparation2.2 Hole-drilling method



8 Conclusions and Perspectives1 General conclusion2 Perspectives

2.1 Veneering and zirconia phase transformation: preliminary results 2.2 Further experimentations by hole-drilling2.3 Clinical considerations: What future for zirconia-based restorations ?

9 Associated works

100102104104105105106106106107108111117

120122124124125125125126126127128130133

136137139139143144

147

1 Residual Stress in Veneering Ceramic I Amélie Mainjot 18

1.1 Metal-based Restorations: the Porcelain-Fused-to-Metal concept

The Etruscans already used dental crowns and bridges 2500 years ago,but they fell out of

use during the middle age. Pierre Mouton reintroduced gold crowns in the 18th century,

while accurate casting procedures for dental alloys were developed in the early 20th century.

Metal that exhibits excellent mechanical properties for framework use,even for large dental

prostheses,does not ensure a natural restoration appearance (Figs 1.1 and 1.2).Therefore,

ceramics were rapidly employed as veneering materials, and the first esthetic crown,

composed of porcelain fused on a platinum post, was patented in 1885.

The creation of a bilayered structure by bonding a ceramic material to a metal alloy is a

challenge. This problem was not solved until the early 1960s for the precious alloys, the

1980s for the non-precious alloys, and the 1990s for the titanium alloys. Before the

introduction of the Porcelain-Fused-to-Metal concept (PFM) by Shell and Nielsen in 1962

[1], which provides a micromechanical and chemical bond between the two materials,

From Veneered Metal to Zirconia 1Introduction I Amélie Mainjot 19

1

Fig. 1.1 - Framework for a 5 elements bridge inAu-Pt alloy before the veneering process

Fig. 1.2 - Full metal restorations: current patient demand is directedtowards veneered restorations, which yields a tooth-like appearance tothe restoration


crowns were composed of porcelain veneers retained to the metal frameworks by the use

of pins and macromechanical retentions (Fig.1.3).The PFM concept has revolutionized the

field of prosthetic dentistry and allowed for more esthetic and less invasive restorations by

minimizing the framework dimensions and the need for macromechanical retentions.

Metal recovering has been extended from a buccal veneer to almost the full coverage

of the framework. Currently, PFM restorations are considered the gold standard in

prosthodontics because of the satisfactory esthetic results and high long term survival rate,

which has been demonstrated in clinical studies [2].

1.1.1 Components

- Alloys

Precious, non-precious and titanium alloys employed for PFM restorations have specific

properties. The associated structure, thermal stability, and mechanical properties minimize

deformation during the veneering process. The melting point and coefficient of thermal

expansion (CTE) are adapted to the ceramic: the melting point is approximately 100°C

higher, and the CTE is slightly higher, than the ceramic.

Moreover, these alloys do not contain any chemical elements that can induce coloration

of the ceramic.Finally, the chemical composition is modified to promote the development

of a thin,regular,uniform,and adherent oxide layer on the alloy surface at high temperature.

Fig. 1.3 - Old concepts of porcelain-metal crowns:

Fig. 1.3a and bCeramic buccal veneerretained to the goldframework with pins

Fig. 1.3cGold framework with macroretentions on the surface to promote adhesionof the ceramic veneer

1Introduction I Amélie Mainjot 21

Therefore,elements that promote oxidation layer formation are needed for precious alloys,

and elements that inhibit growth are required for nonprecious alloys. Titanium alloys are

spontaneously oxidized in the presence of air.

- Veneering ceramics

Veneering ceramics employed for PFM restorations are feldspathic ceramics composed

of alkaline aluminosilicates and secondary oxides, which modify thermal and optical

properties. Their microstructure is characterized by the presence of a glass matrix with

some scattered crystals. Their optical properties are adapted to reproduce enamel,

dentin,or specific effects.The opaque ceramic constitutes the first fired layer and contains

a high amount of opacifiers to mask the framework.

The ceramic powders are mixed with a fluid composed of water and organic polymers

and are then applied with a brush to the substructure.The powders are sintered in the liquid

phase, and the tooth shape is generated by the apposition of successive layers that are

fired separately. The manufacturers generally recommend firing classical veneering ceramics

for precious and non-precious alloys at a temperature around 900°C, i.e., approximately

300°C above the Tg. Titanium alloys require ceramics with a lower Tg because of the

crystalline transformation and the high surface contamination by oxygen when titanium

experiences temperatures above 800°C [3].However, the firing results obtained with dental

ceramics depend on the type of furnace,the firing trays and the size of the workpiece.The

ceramist may change the firing schedules in function of the ceramic appearance and,

therefore, the firing process is empirical.

Finally,the term "porcelain" is not appropriate for veneering ceramics because some of the

original components of porcelain, especially clay, are not currently used for this purpose.


Fig. 1.4 - Key steps of the PFM procedure

Fig. 1.4cOpaque layer

Fig. 1.4bFramework after the oxidation firing

Fig. 1.4aFramework after sandblasting

Fig. 1.4dCompleted crown after the successive firings of veneering ceramic


1.1.2 Key steps of the procedure

- Framework sandblasting (Fig. 1.4a)

Metal frameworks are usually manufactured by casting, but they can also be milled,

sintered or fabricated from metal foils.The Al2O3 blasting procedure eliminates investment

material residues, creates surface roughness on the metal surface and hardens it [4].

Depending on the alloy nature (precious, non-precious, or titanium), manufacturers

recommend the use of different alumina granulometries, generally between 100 and 250

µm, at a pressure between 2 and 4 bar. Roughness is an important factor controlling the

ceramo-metal bonding [5] [6] [7].

- Framework oxidation firing (Fig. 1.4b)

After sandblasting, the surface of the framework is oxidized at high temperatures in the air.

In practice, the framework is heated in a furnace for several minutes at temperatures

around 900°C.This treatment modifies the chemical composition of the framework surface

and the associated wetting properties [4].

- Opaque firing (Fig. 1.4c)

The opaque ceramic is sintered to the oxidized surface and has to be deposited as a thin

and flowed intermediate layer, which will mask the framework and ensure the mechanical

bond between metal and ceramic. The different layers of veneering ceramic, dentin and

enamel are then fired successively to the opaque to reproduce the tooth shape and color

(Figs 1.4d ).

1.1.3 Ceramo-Metal bonding mechanisms

- Physico-chemical

The surface energy of the metal is increased by sandblasting and oxidation procedures,

which promote the physico-chemical adhesion between metal and ceramic [8].

Moreover, the wettability of the framework by the opaque is intensified by the high

temperature used during the liquid phase sintering process and the presence of oxides on

both surfaces. These two different surfaces are the oxidized framework and the opaque

ceramic, which is composed of oxides.

- Chemical

The chemical bond between metal and ceramic is ensured by metallic and ceramic

oxide interdiffusion and the creation of bonds between oxygen atoms [8].


Fig. 1.5 - Mechanical ceramo-metal bonding mechanism: the CTE mismatch between the two materials and the presence of roughness on the sandblasted surface of the metal induce a mechanical interlocking

- Mechanical

One of the most interesting mechanisms that contributes to ceramic-metal adherence is

mechanical interlocking. Mechanical bonding is promoted by a stress development at

the interface due to the CTE mismatch between the two materials. The CTE of metal is

slightly higher than the CTE of opaque (around 1 10-6.K-1) so that during cooling process of

the ceramic firing from Tg to room temperature, compressive stresses supposedly develop

within the ceramic, while compensating tensile stresses supposedly develop within the

framework [9]. As a result of the roughness on the metal sandblasted surface, ceramic

penetrates into cracks and cavities during firing,which results in a mechanical interlocking

after solidification (Fig. 1.5). Mechanical bonding plays a major role for non-precious and

titanium alloys [5].

1.1.4 Clinical background: PFM restorations as a gold standard?

Most studies on the longevity of PFM restorations date back to the 1980s and 1990s.Tan et

al. in 2004 [10] published a systematic review of the survival and complication rates of

gold-resin and PFM fixed partial dentures (FPDs). One of the striking points of this review is

the long-term follow-up of the restorations in the selected studies (until 25 years). The 10-yr

survival risk of FPDs was 89.1%, and the success risk was 71.1%. Biological and technical

complications were distinguished among 3548 FPDs analyzed in 19 studies.The 10-yr risk of

caries was 9.5%,the loss of abutment vitality was 10%,and the loss of FPDs due to recurrent

periodontitis was 0.5%.Relating to technical complications,the 10-yr risk for loss of retention was

6.4%, the loss of FPD due to abutment fracture was 2.1%, and the fracture of the framework

and/or veneer was 3.2%, with either ceramic or acrylic resin veneer. In 2007, the same


research team published an update of this review, added two studies and distinguished

metal-ceramic and goldacrylic resin FPDs [11]. The four studies analyzing PFM restorations

allowed for the calculation of a 10-yr survival rate of 89.1% and a 5-yr veneer fracture rate

of 2.9% among 1218 restorations.The veneer fracture was ranked second among technical

complications after the loss of retention, which reached a 5-yr rate of 3.3%. The major

causes of complication were still biological.The fracture rate was similar to a review published

in 2003 on single crowns.These authors reported a mean veneer fracture incidence of 3%

among 199 PFM crowns analyzed in three studies, with two on titanium crowns and one

on a 2-yr follow up [12]. In a recent systematic review, Heintze et al. evaluated and

compared the frequency of veneer and core fracture of zirconia-based and PFM FPDs [2].

This study only accounted for prospective clinical trials with at least 2 years reporting

technical failures and identified 13 studies for zirconia restorations and only two for PFMs

(134 FPDs).These latest studies compared both types of restorations. The chipping rate

registered for PFMs was 34% after 3 years,but the data were highly influenced by the study

effect. However, the study concluded that the veneer fracture frequency for PFMs was

significantly lower than for zirconiabased restorations.As highlighted by the authors,a lack of

valuable data on PFMs restorations was observed because these studies were conducted

at a time when the quality level of clinical dentistry trials was low.

In conclusion,PFMs restorations are often considered the gold standard due to their long-term

survival rate.However,most of these published studies are not recent and analyzed metal-

based restorations without distinction, including all types of alloys and gold-resin FPDs.

Despite promising data, there is no evidence to support the same statement for veneering

ceramic mechanical behavior,which remains a poorly studied cause of failure.Even if the

veneer fractures can be polished or repaired, they decrease the restorations success rate.

At the scale of prosthodontics history, PFM restorations remain a modern treatment that

must still be evaluated and compared to other available systems, such as veneered

ceramic frameworks, to be considered “the gold standard”…


Fig. 1.6 - Posterior zirconia crowns on natural teeth, and anterior screwed-retained zirconia crown on implant

Fig. 1.6aClinical case before prosthodontics

Fig. 1.6bZirconia frameworks before veneering

Fig. 1.6c and dFinal restorations

Fig. 1.7aZirconia frameworks before veneering

Fig. 1.7aFinal restorations

Fig. 1.7 - Screwed retained zirconia bridges on implants


1.2 Zirconia-based Restorations

1.2.1 An alternative to metal

First employed as orthopedic implants in the eighties, zirconia was put on the market for use

as a framework in prosthodontics about 10 years ago, with Computer-Aided-Design (CAD)

and Computer-Aided-Manufacturing (CAM) processes. Because of its biocompatibility,

strength and optical properties, it was considered to be a good alternative to metal-based

restorations (Figs 1.6 and 1.7). Zirconia is a polycrystalline ceramic, i.e. fully composed of

crystals that impart high strength compared to glass ceramics and infiltrated ceramics,

which contain glass in addition to crystals (Fig. 1.8). The mechanical properties of zirconia

are higher than other ceramic materials used as crown and bridge frameworks: its flexural

strength varies between 900 and 1200 MPa,notably according to the grain size.According

to manufacturers, zirconia is the only ceramic material with which frameworks of more

than 3 elements can be made, while reinforced glass ceramics and infiltrated ceramics are

restricted to 3 element anterior bridges.

Fig. 1.8a - Glassceramics

Fig. 1.8b - Infiltrated ceramics

Fig. 1.8c - Poly-crystalline ceramics

Fig. 1.8 - Microstructuralclassificationof ceramics


One of the main advantages of zirconia is its known biocompatibility, in particular regarding

proven cytocompatibility with osteoblasts and fibroblasts [13]. This property is interesting in

regards to abutments and restorations screwed on implants.From an optical point of view,

even if tinted,zirconia has a high refraction index in comparison to other ceramic materials.

It is advantageous to reproduce high value teeth,or to mask colored abutment.However,

for other cases, it requires a lot of room for the veneering ceramic and an experienced

ceramist to obtain satisfactory results.

One disadvantage of zirconia is its poor bonding properties [14] [15],especially due to the

absence of an etchable glass phase.The bonding properties of zirconia have often been

investigated in the literature. In clinical practice this disadvantage is not the most critical:

because of the room required for the framework and the veneering ceramic, zirconia is

essentially indicated for crowns and bridges, which do not necessitate bonding to the

tooth structure to be retentive.However this lack of bonding properties remains a problem

with short abutments or Maryland bridges.

1.2.2 Zirconia’s unique property: the tetragonal-monoclinic transformation [16] [17] [18]

Pure zirconium oxide presents three crystallographic structures at different temperatures:

cubic (c) (from 2680°C,the melting point, to 2370°C); tetragonal (t) (from 2370°C to 1170°C);

and monoclinic (m) (from 1170°C to room temperature). In prosthodontics,zirconium oxide

is not used in the pure form and is always alloyed with a dopant, which stabilizes the t

phase at room temperature. Currently, this dopant is yttrium oxide at 3 mol%. However,

there is ongoing dental research on the use of other zirconia materials,such as polycrystalline

Cerium-Zirconia or alumina-zirconia composites.

Yttrium oxide prevents the phase transformation of zirconia during cooling: crystals are

retained in tetragonal shape at room temperature instead of transforming to monoclinic

crystals. Yttria-tetragonal-zirconia-polycrystal (Y-TZP) is a thermodynamically metastable

state at room temperature and can undergo a crystalline transformation from t to m under

the effect of stress (i.e. by ferro-elastic switching).

- Transformation toughening

The advantage of the t➔m transformation at room temperature is the increasing of the

fracture thoughness of the material, called the “tranformation toughening”, discovered in

the seventies [16]. Should a crack appear in the material, it will concentrate tensile stress


Fig. 1.9 - Schematic illustration of the transformation toughening mechanism,adapted from 'Zirconia ceramics and zirconia dispersed composites',from Chevalier J,Gremillard L., in Bioceramics and Their Clinical Applications,Edited by T. Kokubo C R C, 2008.

on its tip and locally trigger the stress-induced transformation, which is characterized by a

rapid and noticeable volume increase of the crystals (around 4%).

This local volume expansion induces the development of compressive stress which closes

the crack and hinders its propagation (Fig.1.9).That is why zirconia is said to have “healing

properties”, and demonstrates the highest fracture toughness among all other ceramic

materials (6-10 Mpa/m1/2). The very interesting picture from the thesis of El Attaoui (INSA

Lyon,France,2003) (Fig.1.10) shows the expanded material around a crack,closing it,forming

what looks like a scar on the surface. The t➔m transformation is irreversible at room

Fig. 1.10Zirconia stress-induced transformation around a crack, which hinders propagation.From the thesis of El Attaoui, INSA Lyon, 2003.


temperature, and is only reversible by a regeneration treatment, which involves heating

the material to between 900°C and 1000°C.Then the healing property cannot be exploited

twice at the mouth temperature: there is a local aging of the material.

- Low Temperature Degradation

However, the t➔m transformation also occurs in the absence of stress at low temperatures

(from room temperature to around 400°C) and in moist atmospheres (Fig. 1.11). Typically,

the tetragonal crystals slowly transforms to monoclinic by a nucleation and growth

process,which occurs in the material surface,engendering surface alterations,microcracks,

a loss of strength, further moisture penetration and auto-catalytic processes. This aging

phenomenon,called the Low Temperature Degradation (LTD) of zirconia, led to numerous

delayed failures of orthopedic prostheses in 2001.This phenomenon is concerning for dental

indications because of the contact with oral fluids.Currently the kinetics and impact of LTD

on the life span of dental prostheses is unknown [19].

Finally, the properties of zirconia have been voluntary simplified and summarized in this

chapter. However, this topic involves a very complex technology and sensitive material,

particularly in regards to the t➔m transformation and the metastable phase diagram,

Fig. 1.11Schematic illustration of the Low Temperature Degradation mechanism, adapted from'Zirconia ceramics and zirconia dispersed composites', from Chevalier J, Gremillard L., inBioceramics and Their Clinical Applications, Edited by T. Kokubo C R C, 2008.


which are still not completely understood.Besides the tetragonal zirconia polycrystals (TZP)

used in dentistry, different varieties and types of zirconia exist that are applied in different

fields, such as biomedical engineering. These zirconia are constantly evolving in terms of

composition, process, or microstructure with scientific advances, such as the reduction of

the LTD problematic.

1.2.3 The veneering concept of zirconia frameworks

- Framework pretreatment

Zirconia frameworks are CAD-CAM manufactured either by soft machining of presintered

blanks followed by sintering at high temperature or by hard machining of fully sintered

blocks [17]. Soft machining prevents the t➔m stress-induced transformation, while hard

machining may promote microcracking and monoclinic crystals apparition, resulting in a

higher susceptibility to LTD and fatigue damage [17].However,clinical studies have shown

that soft-machined restorations exhibit more framework fractures [20]. This result can be

explained by the influence of other factors,such as the greater density and lower porosity of

hot isostatic pressed (HIP) zirconia blocks used with hard machining processes.

Similarly, grinding adjustments and sandblasting procedures are not recommended by

many manufacturers for avoiding the formation of surface flaws. These procedures were

shown to induce the apparition of a certain amount of monoclinic crystals and to be

detrimental for material performance when submitted to cyclic loading [17] [18].

Furthermore microcracks can develop and promote aging. Conversely, the in-vitro flexural

strength increases because of compressive stress at the surface due to transformation [17].

Grinding procedures are inevitable because the technician must thin the cervical framework

limit, which is always oversized because of manufacturing requirements. This requirement

justifies the recommendation of a regeneration firing of the framework before the veneering

process. In this procedure, the framework is heated to 1000°C for 15 minutes (VM9 firing

procedure, Vita Zahnfabrik, Bad Säckingen, Germany, http://vident.com/files/2009/01/

vm9-firing-chart-with-margin.pdf).

- Ceramo-Zirconia bonding

Contrary to the PFM concept,ceramo-zirconia bonding has not been well described and

is not well understood. Procedures recommended by manufacturers vary. Due to the


aforementioned reasons, the framework is not sandblasted. In comparison with PFM

restorations, the absence of sandblasting is disadvantageous for the physico-chemical and

the mechanical bonding mechanisms.It may decrease the wettability of the framework by

the veneering ceramic and reduce the mechanical interlocking at the interface.

Depending on the system,a “liner”material is thinly fired on the framework as an intermediate

layer with the veneering ceramic, similarly to opaque used for PFM restorations. The liner

provides some chroma and fluorescence and may also assure wetting of the framework.

The chemical bond between the two ceramic materials, which are both composed of

oxides, is supposed to be effective [18],even if the interdiffusion of oxides and dopants has

been suspected to destabilize the zirconia microstructure [17].

The development of veneering ceramics by manufacturers was reduced to CTE

measurements and thermal shock testing [17]. These characteristics were copied and

pasted from the PFM concept. Indeed, the CTE mismatch between the bulk materials was

designed to mimic the ceramo-metal CTE mismatch, with the CTE of veneering ceramics

adapted to be slightly lower than the zirconia (around 10.5 10-6.K-1).Based on the principle

that compressive stress promotes the mechanical behavior of the ceramic, this approach

was intended to develop compressive stress within the ceramic during the veneer cooling

process because this mechanism is supposed to occur in PFM restorations [9] [18].This result

was not verified before the introduction of ceramics on the market because the attention

of manufacturers was focused on the mechanical behavior of the framework. Recently,

slow cooling procedures for the veneering ceramic were introduced by some manufacturers

to improve compressive stress development and to reduce veneer fracture, but these

procedures were contradicted by the in vitro results of interfacial adhesion tests [9].

1.2.4. Clinical background: the chipping problem

Compared to PFM restorations that were introduced 40 years earlier in prosthodontics,clinical

studies of zirconia-based restorations (ZBR) are much more recent.This implies that the clinical

background is shorter in duration, with a maximum follow-up of 5 years; however, studies

exhibit a higher quality level [2]. In the systematic review of Schley et al. [21], the mean

estimated 5-yr survival rate for FDPs was 94.29%. Among biological complications (5-yr

complication free rate of 91.72%), secondary caries were detected most frequently. This

result was partly explained by the lack in precision of fit in the earliest CAD CAM systems.


However, PFM and ZBR differ in technical complications, and ZBR exhibit a 5-yr complication

free rate of only 76.41%. Fracture is the most significant technical complication of ZBR

compared to the loss of PFM retention [11] . Contrary to all other ceramic restorations,

veneering ceramic is the weak link in the restoration, instead of the framework. The fracture

rate of zirconia frameworks is reportedly low, while veneer fracture is an important cause of

failure in all studies,which is reported more frequently than for PFM restorations [20] [2] [21]

[17] [18] (Tab. 1.1).

Brand Study Follow-up Type of restoration Sample Framework Veneer size fracture % fracture %

Cercon zirconia Sailer et al. 2007 5 years 3-5 units FPD 33 8 15(Dentsply) Beuer et al. 2009 3 years 3 units FPD 21 5 0 Cehreli et al. 2009 2 years Single crowns 15 7 0 Schmitter et al. 2009 2 years 4-7 units FPD 30 3 3 Bornemann et al. 2003 1-5 years 3-4 units FPD 59 0 3

Lava (3M ESPE) Raigrodski et al. 2006 2-5 years 3 units FPD 20 0 25 Pospeich et al. 2003 2 years 3 units FPD 38 0 3 Crisp et al. 2008 1 year 3-4 units FPD 38 0 3

Procera zirconia Ortrop et al. 2009 3 years Single crowns 204 0 2(Nobel Biocare)

IPS e.max Zir/CAD Ohlmann et al. 2008 1 year IRFPD 30 10 13(Vivadent-Ivoclar)

Denzir Molin & Karlsson 2008 5 years 3 units FPD 19 0 36(Cadesthetic AB) Larsson et al. 2006 1 year 2-5 units FPD/Ti abut 13 0 54

DC-Zirkon Tinschert et al. 2008 3 years 3-10 units FPD + cantilever 65 0 6(DCS Dental AG) Vult von Steyern et al. 2005 2 years 3-5 units FPD 23 0 15

Digizon Edelhoff et al. 2008 3 years 3-6 units FPD + cantilever 21 0 9.5

Procera zirconia Dpt of Fixed Prosthodontics 6 months single crowns 55 2 11DC-Zirkon University of Liège, 2011 - 5,5 years and 2-6 units FPDLavaCercon Zirconia

Tab. 1.1Percentages of framework fractures in comparison to veneer fractures in ZBR in differentclinical studies, from the systematic review of Al-Amleh et al. [20] . Preliminary results of a retro-spective study conducted at the University of Liège were added.


The cohesive fracture of the veneering ceramic is called “chipping”(Fig.1.12 and Fig.1.13).

Depending on the study, the short-term percentage varies from 0 to 54% (Tab. 1.1). Some

studies distinguish minor chipping, which can be simply polished and does not require

removal from the restoration, from major chipping or delaminations, which implies an

adhesive fracture at the interface between veneer and zirconia.

Delaminations are not easy to diagnose in the mouth (Fig.1.14),since microscopic observation

is required to confirm the interfacial rupture and the framework exposition. Frameworks

fractures often result from improper design, especially connections, from default in the

manufacturing process, or from clinical parameters [20]. Despite many in vitro studies

conducted in that field, the chipping mechanism remains unknown. Several factors are

Fig. 1.12Clinical case: zirconia crowns cemented on zirconia abutments screwed on implants

Fig. 1.12aZirconia abutments

Fig. 1.12bZirconia frameworks

Fig. 1.12cFinal restorations after cementation

Fig. 1.12d and eChipping on the buccal cusp of the premolar,3 months after placement

Fig. 1.12fChipping by scanning-electron microscopy of the crown epoxyresin replica


Fig. 1.13a and bZirconia framework designed to support the veneering ceramic

Fig. 1.13Clinical case: screwed-retained zirconia bridge on implants

Fig. 1.13cFinal restorations after placement

Fig. 1.13d and eChipping on the palatal cusp of the first premolar,5 months after placement

Fig. 1.13fChipping by scanning-electronmicroscopy of the bridge epoxyresin replica

Fig. 1.14Clinical case: chipping on a zirconia molar crow.The framework is visible on the cervical zone,but delamination cannot be confirmed without microscopic evaluation.

Residual Stress in Veneering Ceramic I Amélie Mainjot

taken into account, such as the CTE of the veneering ceramic, the thermal properties of

zirconia,the cooling rate,the thickness of both materials,and the design of the framework,

which was not correctly adapted to support the veneer at the CAD-CAM systems initiation.

The chipping problem has now led to the introduction of new veneering systems, such as

pressed or milled veneering ceramics,but these systems have not yet been proven effective.

Chipping results in a lack of confidence in ZBR by many practioners that may use PFM

restorations or full zirconia crowns, which result in a poor esthetic appearance due to the

absence of veneering ceramic.

The problem of chipping,particularly in ZRB,is very complex.Both in vitro and in vivo data are

important to put the puzzle together,each type of studies contributing to a stepwise approach.

36


2.1 Traditional in vitro mechanical tests

Randomized, controlled clinical trials are often considered the best method to study the

clinical performance of restorations. However, these trials are long, time-consuming, and

costly. Furthermore, the major challenge encountered is the limited analysis of all clinical

parameters.Therefore, the mechanical performance of restorations is often studied in vitro

through fracture mechanical tests and indentation. Different fracture tests are available:

flexure, tensile and shear.The loading conditions can be static or dynamic.These tests can

be performed on standardized samples or on more complex geometrical samples, such

as crowns and bridges.The effect of time on the material resistance can then be investigated,

while under the effect of temperature, thermal cycling,or repetitive stress by cyclic fatigue

loading.These studies can also be coupled with fractographic approaches, which analyze

the flaw type.Fractography can also be performed on replicas derived from in vivo fractured

restorations or from removed fractured restorations. According to Anusavice, this

approach is very helpful and should always be included in clinical trials [22].

Numerous in vitro studies of chipping, crack propagation, and ceramic-zirconia adhesion

have been published.There are both advantages and limitations of the flexure,tensile and

shear tests. These tests are not easy to interpret, such as the influence of sample geometry

[23], influence of testing environment, stressing rate [24],and influence of the crack initiation

site [22].A lack of standardized procedures has been described [22]. Tests using indenters

are influenced by ceramic thickness and can also overestimate adhesion values between

zirconia and veneering ceramic due to the phase transformation process [25]. The

traditional load-to-failure tests of all-ceramic single unit restorations were criticized by Kelly

[26], who affirmed that these tests do not create appropriate stress states, cause failure

from clinically relevant flaws, or create crack systems that model clinical failure. Testing a

sample by loading is not ideal because these results will depend on the load application

and stress repartition within the sample. Therefore, results of the different types of tests are

not comparative.

2How to study and predictthe mechanical behaviorof the veneering ceramic?


In conclusion,even if there is no ideal mechanical test to directly correlate with the clinical

situation,a combination of different in-vitro tests under well-controlled simulated conditions

may be useful in a systematic approach, which tries to solve a problem using a stepwise

strategy. Particularly after residual stress analysis, chewing simulations and fracture load

tests may be used to study the influence of restoration geometry. In vitro tests may also

precede clinical trials to highlight the different study parameters.

2.2 Residual stress analysis

2.2.1 Residual stress: Principle and interest

The origin of cracks in dental ceramic is multifactorial: thermal stress caused by thermal

incompatibility, unbalanced occlusal contact points, unsupportive framework, are a few

examples among all others, still misunderstood. But it is known that in all cases, cracks form

when the tensile stress within the ceramic exceeds the tensile strength [22].

The tensile stress at a specific location of the ceramic is the sum of external and residual

stress. External stresses are stresses generated within the structure by applied loads (i.e.,

during mastication in the case of dental prostheses).Residual stresses are stresses generated

within the structure during the manufacturing processes (i.e.,during the cooling procedure

of the veneering ceramic in the case of dental prostheses). Residual stresses are “locked-

in”stresses present within the veneer and the framework from the time when the veneering

ceramic solidifies. These stresses exist and persist in the materials without the application of

any applied load but add to the stresses temporarily induced by the loading of restoration

[27]. The residual stresses can cause immediate cracking or delayed cracking in the

ceramic [22]. In dental prostheses, residual stresses are mainly engendered by the existence

of a mismatch in thermal properties between the different materials and by the non-uniform

cooling of the restoration during the tempering process [9]. The distribution of stresses

describes a profile through the thickness of both materials (Fig. 1.15). This profile satisfies

equilibrium between tensile and compressive stresses and, therefore, offers no external

evidence of their existence [27].Residual stresses are a universal phenomenon that occurs

with all materials manufacturing processes in any field. They are induced by a lack of

uniformity: a non-uniform material properties, a non-uniform cooling during manufacturing,


local phase transformations,a non-uniform plastic deformation,and non-uniform chemical

reactions or diffusion.This type of stress plays a critical role in failures due to fracture, fatigue

and wear. However, residual stress can also have positive effects and improve the

performance of a material, such as in the case of tempered glass. Indeed the tempering

procedure generates in surface compressive stresses that reinforce the material.

Therefore,knowledge of residual stresses and their formation during manufacturing is crucial

to optimize component design and manufacturing [28]. Particularly, the stress vs. depth

profile in veneering ceramic is critical to understand chipping failures [29] .

2.2.2 Existing data relating to dental restorations

Taskonak et al.[30,31] have analyzed surface residual stresses in bilayered ceramic structures

using a four-step fracture mechanics approach. This method combines indentation and

flexural strength and a fracture mechanics equation is used to quantify in surface stresses.

Their results highlighted the importance of heat treatment parameters and cooling rate

on the biaxial flexural strength of bilayered zirconia based structures.However,this indentation

technique, which was introduced in 1977 by Marshall and Lawn [32] and previously used

by Anusavice on veneering ceramic disks [33],does not provide the stress profile (i.e., stress

distribution in function of depth), which is a key factor for understanding and predicting

failure mechanisms.Moreover, tests applying an external load have limitations,which were

previously discussed.

Fig. 1.15Schematic illustration of a residual stress profile in a bilayer characteristic of dental prostheses.


Until this work, the stress depth profile in bilayers of dental prostheses has been studied

through calculations and finite element analysis simulations but has never been quantified.

One of the difficulties encountered by finite element analysis is to build into calculations

the viscoelastic behaviour of ceramic in the glass transition range, where thermo-physical

properties, such as the coefficient of thermal expansion, glass transition temperature, and

viscosity, have strong temperature dependency and influences on residual stresses [34].

Morevover, the thermal gradients are complicated to simulate due to the amount of

influencing parameters, such as the influence of furnace type and the sample support.

Asaoka [35] and DeHoff [36] have developed analytical models that incorporate variations

in the thermal expansion coefficent for studying the influence of cooling rate and thermal

expansion coefficient mismatch on residual stress profiles in veneered metal structures.

Recently, DeHoff [37] [38] studied the impact of thermal expansion coefficient mismatch

on residual stresses in a 3D model of three-unit FPD with a glass ceramic framework using

viscoelastic finite element stress analysis. Subsequently, Swain highlighted the importance

of the cooling rate, thickness and thermal expansion coefficient on residual stress profiles

in bilayered structures composed of glass ceramics, alumina and zirconia substrates [39].

Although relevant, his approach does not include viscous relaxation effects, variations in

the veneering ceramic thermal expansion coefficient, or variations in the cooling rate

around the glass transition temperature. Arman et al. [40] introduced a more realistic

approach including in their 3D finite element analysis of PFMs crowns, the mechanical and

thermal properties at various temperatures, such as the equivalent heat transfer coefficients

and the instantaneous modulus.

As the development of residual stress involves non-linear material behavior with temperature

and complex thermal gradients,the current predictive capabilities are insufficient to provide

adequate knowledge of residual stresses.Therefore, the ability to measure residual stress is

essential to minimize residual-stress related failures and to aid in developping predictive

capabilities by verifying models [27].

2.2.3 Residual stress profile measurement methods

Various experimental methods for measuring residual stress profiles have been developed

for scientific use and for implementation in industrial process development or quality control

systems[28] [27]. These methods are characterized by the following:


-

-

-

-

-

The physical principle that the method is based on (i.e., mechanical or physical).

Physical methods are mainly diffraction methods.

The scale of measurement from microscopic stresses on the atomic scale to macroscopic

stresses, which cover multiple grain sizes or even the whole component.

The spatial resolution.

The destructiveness, as some methods are completely destructive, semidestructive, or do

not alter the sample.

The sensitivity, accuracy, and measurement time.

As indicated by Prime [27], there is no single best method to measure residual stress, and

the selection of a method should consider the particular application and strengths and

weaknesses of all methods.

Mechanical methods and diffraction are both based on a deformation measurement,

which is utilized to calculate residual stresses. Indeed, the development of residual stress

induces an elastic deformation of the material, and this deformation is proportional to the

stress developed. Mechanical methods measure the component deformation during or

after the manufacturing process, while diffraction measures the distance variation

between two atomic planes.

- Diffraction methods

Diffraction methods measure the crystal lattice deformation by bombarding the sample

with an X-Ray or neutron beam.The angle that the beam is diffracted is related to the crystal

lattice architecture. Lattice deformations can be used to quantitatively calculate stresses

in the material. X-ray diffraction is the most commonly used technique, but it is limited to a

depth range of approximately 20-30 µm. In constrast, neutron diffraction has a higher

penetration depth but a lower spatial resolution [28]. These methods are sensitive to grain

size and texturing effects and are unable to measure non-crystalline materials [27]. These

methods are hardly applicable to dental prostheses because of the composite structure

of veneering ceramic comprised of a glass matrix filled with crystalline particles.

- Mechanical methods

A major class of residual stress measurement methods used in industry is based on the

removal of some stressed material and the measurement of the resulting deformations in


the adjacent material [41], which are linked to the released stresses. The deformations

are measured on the surface,typically using strain gages.The residual stresses can then be

calculated following the Hooke’s law principle :

Û = Â.E

where Û = stress (MPa),Â= ¢L/l= deformation (Microstrains),and E= elasticity modulus (GPa)

In practice,engineers use inverse solutions,which yield a stress distribution that results in the

“best” correlation with the measurements.

Mechanical methods include dissection,layer removal,and slitting for components of simple

geometry in addition to cutting, ringcore drilling and hole-drilling for more complex

geometries [28].All methods are destructive with different degrees of destruction and are

suitable for determination of stress depth profiles in mm range from the surface [28].

A. Hole-drilling method

The hole-drilling method is an incremental step,quasi non-destructive method that is easy-

to-handle and both time and cost efficient. Furthermore, this method can be applied on

almost every material including coatings and layer composites [28]. Holedrilling is widely

used in industry. An advantage is that the hole-drilling method results in residual stress

depth profiles with a high depth resolution and a high accuracy of stress results [28]. The

procedure has been standardized in ASTM Standard Test Method E 837 [42].Fig.1.16 illustrates

the process for the hole-drilling method.First,a strain gage is attached to the specimen surface,

as schematically shown in Fig. 1.16(a). Then a small hole is drilled adjacent to the strain

Fig. 1.16Hole-drilling method: (a) specimen cross-section before hole drilling and (b) specimen cross-sectionafter hole drilling,showing exaggerated surface deformations caused by the local release of residualstresses.

StrainGage

StrainGage


gage.This releases the residual stresses within the hole,and causes the material around the

hole to deform slightly,as shown in exaggerated form in Fig.1.16(b).The strain gage measures

this deformation, from which it is possible to calculate the size of the original residual stress.

In practice, instead of a single strain gage, a specially designed strain gage rosette

consisting of three or six strain gages is used. The hole is drilled incrementally, allowing for

stress calculations at different depths from the surface.

B. Slitting method

The slitting method, or crack compliance method, is based on the same principle as hole-

drilling. Instead of a hole, a slot is machined incrementally into the sample.The slot releases

the residual stress normal to the face of the slot, and the sample deforms. Surface strain

gages measure the resulting strains as a function of slot depth, and these strains are used

to solve for stress [27]. A supplementary strain gage can be bonded on the back of the

sample to increase depth resolution. Following Prime, the depth range of measurement is

greater than the hole-drilling method. However this more recently developed technique

has not been standardized yet, and there is no marketed software that includes inverse

solutions for stress calculation.

Shell JS, Nielsen JP. Study of the bond between gold alloys and porcelain. J Dent

Res. 1962 Nov-Dec;41:1424-37.

Heintze SD, Rousson V. Survival of zirconia- and metal-supported fixed dental

prostheses: a systematic review. Int J Prosthodont. 2010 Nov-Dec;23(6):493-502.

Kononen M, Kivilahti J. Fusing of dental ceramics to titanium. J Dent Res. 2001

Mar;80(3):848-54.

Donald IW. Preparation, Properties and Chemistry of Glass-Ceramic-to-Metal and

Glass-Ceramic-to-Metal Seals and Coatings. Journal of Materials Science. 1993

Jun;28(11):2841-86.

Hautaniemi JA, Juhanoja JT, Hero H. Porcelain Bonding on Ti - Its Dependence on

Surface-Roughness, Firing Time and Vacuum Level. Surf Interface Anal. 1993 May

15;20(5):421-6.

Richmond JC, Moore DG, Kirkpatrick HB, Harrison WN. Relation between roughness

of interace and adherence of porcelain enamel to steel. J Am Ceram Soc. 1953

Dec;36(12):410-6.

Wagner WC, Asgar K, Bigelow WC, Flinn RA. Effect of interfacial variables on metal-

porcelain bonding. J Biomed Mater Res. 1993 Apr;27(4):531-7.

Pask JA, Fulrath RM. Fundamentals of glass-to-metal bonding: nature of wetting

and adherence. J Am Ceram Soc. 1962 Dec;45(12):592-6.

Gostemeyer G, Jendras M, Dittmer MP, Bach FW, Stiesch M, Kohorst P. Influence of

cooling rate on zirconia/veneer interfacial adhesion. Acta Biomater. 2010

Dec;6(12):4532-8.

Tan K, Pjetursson BE, Lang NP, Chan ES. A systematic review of the survival and

complication rates of fixed partial dentures (FPDs) after an observation period of

at least 5 years. Clin Oral Implants Res. 2004 Dec;15(6):654-66.

Pjetursson BE, Bragger U, Lang NP, Zwahlen M. Comparison of survival and

complication rates of tooth-supported fixed dental prostheses (FDPs) and implant-

supported FDPs and single crowns (SCs). Clin Oral Implants Res. 2007 Jun;18 Suppl

3:97-113.

Goodacre CJ,Bernal G,Rungcharassaeng K,Kan JY.Clinical complications in fixed

prosthodontics. J Prosthet Dent. 2003 Jul;90(1):31-41.

Mustafa K, Wennerberg A, Arvidson K, Messelt EB, Haag P, Karlsson S. Influence of

modifying and veneering the surface of ceramic abutments on cellular attachment

and proliferation. Clin Oral Implants Res. 2008 Nov;19(11):1178-87.


References1

2

3

4

5

6

7

8

9

10

11

12

13


Kern M, Swift EJ, Jr. Bonding to zirconia. J Esthet Restor Dent. 2011 Apr;23(2):71-2.

Thompson JY, Stoner BR, Piascik JR, Smith R. Adhesion/cementation to zirconia and

other nonsilicate ceramics: where are we now? Dent Mater. 2011 Jan;27(1):71-82.

Chevalier J, Gremillard L, Virkar AV, Clarke DR. The Tetragonal-Monoclinic

Transformation in Zirconia: Lessons Learned and Future Trends. J Am Ceram Soc.

2009 Sep;92(9):1901-20.

Denry I, Kelly JR. State of the art of zirconia for dental applications. Dent Mater.

2008 Mar;24(3):299-307.

Zarone F, Russo S, Sorrentino R. From porcelain-fused-to-metal to zirconia: Clinical

and experimental considerations. Dent Mater. 2011 Jan;27(1): 83-96.

Lughi V,Sergo V.Low temperature degradation -aging- of zirconia: A critical review

of the relevant aspects in dentistry. Dent Mater. 2010 Aug;26(8):807-20.

Al-Amleh B, Lyons K, Swain M. Clinical trials in zirconia: a systematic review. J Oral

Rehabil. 2010 Aug-Sep-Oct;37(8):641-52.

Schley JS, Heussen N, Reich S, Fischer J, Haselhuhn K, Wolfart S. Survival probability

of zirconia-based fixed dental prostheses up to 5 yr: a systematic review of the lit-

erature. Eur J Oral Sci. 2010 Oct;118(5):443-50.

Anusavice KJ, Kakar K, Ferree N. Which mechanical and physical testing methods

are relevant for predicting the clinical performance of ceramic-based dental

prostheses? Clin Oral Implants Res. 2007 Jun;18 Suppl 3:218-31.

Hammad IA, Talic YF. Designs of bond strength tests for metal-ceramic complexes:

review of the literature. J Prosthet Dent. 1996 Jun;75(6):602-8.

Taskonak B, Griggs JA, Mecholsky JJ, Jr.,Yan JH. Analysis of subcritical crack growth

in dental ceramics using fracture mechanics and fractography. Dent Mater. 2008

May;24(5):700-7.

Taskonak B, Yan J, Mecholsky JJ, Jr., Sertgoz A, Kocak A. Fractographic analyses of

zirconiabased fixed partial dentures. Dent Mater. 2008 Aug;24(8):1077-82.

Kelly JR. Clinically relevant approach to failure testing of all-ceramic restorations. J

Prosthet Dent. 1999 Jun;81(6):652-61.

Prime MB.Residual stress measurement by successive extension of a slot: the crack

compliance method. Applied Mechanics reviews. 1999;52(2):75-96.

Wenzelburger M, Lopez D, Gadow R. Methods and application of residual stress

analysis on thermally sprayed coatings and layer composites. Surface & Coatings

Technology. 2006 Oct;201(5):1995-2001.

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

Swain MV. Unstable cracking (chipping) of veneering porcelain on all-ceramic

dental crowns and fixed partial dentures. Acta Biomater. 2009 Jun;5(5):1668-77.

Taskonak B, Borges GA, Mecholsky JJ, Jr.,Anusavice KJ, Moore BK,Yan J. The effects

of viscoelastic parameters on residual stress development in a zirconia/glass bilayer

dental ceramic. Dent Mater. 2008 Sep;24(9):1149-55.

Taskonak B, Mecholsky JJ, Jr., Anusavice KJ. Residual stresses in bilayer dental

ceramics.Biomaterials. 2005 Jun;26(16):3235-41.

Marshall DB, Lawn BR. An indentation technique for measuring stresses in

tempered glass surfaces. J Am Ceram Soc. 1977;60(1-2):86-7.

Anusavice KJ, DeHoff PH, Hojjatie B, Gray A. Influence of tempering and contraction

mismatch on crack development in ceramic surfaces. J Dent Res. 1989

Jul;68(7):1182-7.

Asaoka K, Kuwayama N, Tesk JA. Influence of tempering method on residual stress

in dental porcelain. J Dent Res. 1992 Sep;71(9):1623-7.

Asaoka K, Tesk JA. Transient and residual stress in a porcelain-metal strip. J Dent

Res. 1990 Feb;69(2):463-9.

DeHoff PH, Anusavice KJ. Viscoelastic stress analysis of thermally compatible and

incompatible metal-ceramic systems. Dent Mater. 1998 Jul;14(4):237-45.

DeHoff PH, Anusavice KJ, Gotzen N. Viscoelastic finite element analysis of an all-

ceramic fixed partial denture. J Biomech. 2006;39(1):40-8.

DeHoff PH, Anusavice KJ. Viscoelastic finite element stress analysis of the thermal

compatibility of dental bilayer ceramic systems. Int J Prosthodont. 2009 Jan-

Feb;22(1):56-61.


dental crowns and fixed partial dentures. Acta Biomater. 2009;5(5(june)):1668-777

Arman Y, Zor M, Gungor MA, Akan E, Aksoy S. Elastic–plastic finite elements analysis

of transient and residual stresses in ceramo-metal restorations. J Biomech

2009;42(13):2104–10.

Schajer GS, Prime MB. Use of inverse solutions for residual stress measurements.

Journal of Engineering Materials and Technology, Transactions of the ASME.

2006;128(3):375-82.

ASTM. Determining residual stresses by the hole-drilling strain-gage method.

Standard Test Method E837-08 West Conshohocken, PA, USA: American Society for

Testing and Materials; 2008.

29

30

31

32

33

34

35

36

37

38

39

40

41

42


2Objectives and strategy I Amélie Mainjot 47

Assumptions...2 Residual Stress in Veneering Ceramic I Amélie Mainjot 48

The main objective of this thesis was to contribute to the understanding of the chipping

mechanism in zirconia-based restorations.

The strategy was based on the following assumptions developed in the introduction:

Clinical studies show that zirconia-based restorations (ZBR) exhibit more chipping than

porcelain-fused-to-metal (PFM) restorations.

Fracture occurs because of the presence of local tensile stresses within the veneering

ceramic, these local tensile stresses being the sum of the residual stresses generated in this

area during the manufacturing process and of applied loads.

The knowledge of residual stress vs. depth profile in the veneering ceramic is a key factor

for preventing chipping and optimizing the manufacturing process. It is the first step of a

holistic approach,which should be followed by well-controlled in vitro traditional mechanical

tests and clinical studies.

The PFM concept is supposed to induce a favorable residual stress profile in the veneering

ceramic, promoting the development of compressive stresses due to the coefficient of

thermal expansion (CTE) mismatch and the tempering effect. In contrast, the concept of

the veneering process of zirconia, which is nevertheless a very sophisticated material, is

empirical.

The residual stress profiles in dental prostheses have been studied by simulations but have

never been measured.

Effective methods allowing residual stress measurement are being developed in industry.

...aims 2Objectives and strategy I Amélie Mainjot 49

Therefore, the aims of this thesis were the following:

To transfer and adapt an effective industrial method for measuring residual stresses to dental

use and to demonstrate the method for measurement of residual stresses in veneer-metal

and veneer-zirconia disks.

To study the influence of the manufacturing process on the residual stress profile, in terms

of the following parameters:

- The cooling rate

- The veneer thickness

- The framework thickness

To compare the residual stress measurements in veneer-metal and veneer-zirconia structures.

To contribute to the explanation of residual stress development origins in veneer-metal

and veneer-zirconia bilayers, comparing measurements to simulations and hypotheses

described previously in the literature.

To contribute to the explanation of failures encountered in clinical practice.

3Method selection and development I Amélie Mainjot 51

The selection and the development of a method for measuring residual stresses in samples

characteristic of dental prostheses took approximately one and a half years, i.e., one

half of the thesis period. The main challenge encountered was achieving a high level

of measurement accuracy (10 nanovolts), according to the low magnitude stresses

developed in dental prostheses, in comparison with industrial applications of the transferred

and adapted method.


Method selection152

Among the industrial methods used for the determination of the residual stress profile

over a depth range from the surface to several millimeters deep, the hole-drilling

method and the slitting method, which are both semidestructive methods based on

deformation and strain gage use (see introduction), were the most appropriate [1]. The

final choice was made by taking into account the following parameters:

The depth range measurement, related to veneer thickness.

The sensitivity of the method, related to the low magnitude of the stresses: the method

has to induce sufficient deformation (strains) because of the sensitivity of the strain gages.

The accuracy of the method, which is also related to the low magnitude of the stresses:

the chain measurement has to be accurate enough to detect low-voltage variations.

The existence of a standardized procedure to transfer and to adapt, and the existence

of a software package containing the mathematical models that facilitate stresses

determination from the measured deformations.

The depth-resolution of the hole-drilling (HD) method is limited to 1.2 mm, while the slitting

method has a higher depth range measurement. Preliminary tests were performed with

both methods (Fig. 3.1). The greater sensitivity of the slitting method was shown to be

very interesting based on the low stresses that needed to be measured, but the lack of

a standardized procedure and, most of all, the absence of software for performing the

-

-

-

-

Fig. 3.1 - Device designed for the preliminary tests with the slitting method


stress calculation from the measured deformations guided the selection toward the HD

method. Indeed, the development of mathematical models adapted to the studied

bilayers seemed more complicated than attempting to improve the sensitivity of the

HD standardized procedure by increasing the accuracy of the measurements. The only

limitation of the HD method remained the depth resolution, which was limited to an

approximately 1.2–1.5 mm depth: solutions were envisaged but not implemented

because the obtained results made it possible to achieve the defined objectives.


Starting from the hole-drilling ASTM norm [2], each step of the procedure was adapted

to reach a measurement accuracy of 10 nanovolts.

Fig. 3.3 - Illustration of the equipment needed for the adapted HD method

Method development154

Fig. 3.2 - Schematicillustration ofthe HD method principle


The principal evolutions and modifications brought to each step of the standardized

procedure are summarized below. The details of the final procedure are presented in

chapter 4.The HD method principle is summarized in Fig.3.2.The equipment needed for

the adapted HD procedure is illustrated in Fig. 3.3.

2.1 Specimen preparation

Bilayers with a CoCr or a zirconia framework were always produced in the same way,

i.e., following the manufacturer’s recommendations, particularly relating to the veneering

process (Fig. 3.4). This procedure is detailed in chapter 4.

The geometry of the sample was a square at the beginning of the study, which was rapidly

changed to a disk because of the circular geometry of the rosettes and the furnace.

For manufacturing purposes, the diameter was reduced to the minimum allowed with

respect to the used rosette.

Fig. 3.4 - Specimen veneering sequence, for CoCr and zirconia frameworks. From left to right: pretreated framework, opaque or liner, veneering ceramic before firing and final specimen.


2.2 Strain gage rosette

Special care was taken when implementing the strain gage bonding procedure,

which is described in chapter 4. It is a delicate procedure that requires some experience:

any defect in the bond, such as the presence of a bubble, can induce errors in the

measurements. In the present case, an etching procedure and cleaning in an ultrasonic

bath were performed before bonding. Moreover, the adhesive was allowed to cure

overnight.All rosettes were checked with an optical microscope after bonding. Initially,

standardized Type A strain gage rosettes from Vishay (Vishay, Malvern, PA, USA) were

used (Fig. 3.5) but were replaced by Type C rosettes (Fig. 3.4), which showed a greater

stability and five times greater output. Type A rosettes contain three strain gages, while

type C rosettes contain six strain gages. They allow deformation measurements along

the three different space axes parallel to the surface sample. A comparison of the

results is discussed in chapter 4.

2.3 Electrical measurement chain

The major change made to the original method was the building of a high-precision

electrical measurement chain. Particularly, the use of a customized Wheatstone bridge

Fig. 3.5 - Standardizedthree-element Type A strain gage rosette on a disk sample


strongly increased the stability and the accuracy of measurements.When the standard

requires a 1-micron strain precision, the final adapted method provides a 0.05 micron

strain precision and stability.

2.3.1 Operating principle of strain gages and Wheastone bridges

Strain gages make use of electrical resistances. When stresses are released by the

hole-drilling process,strain gages deform with the material.A gage deformation generates

a variation in resistance (Fig. 3.6), according to the following formula:

¢R/R= k ¢L/L

where ¢R/R= resistance variation, k= gage factor, and ¢L/L= deformation or strain.

Unfortunately, the resistance can also vary with temperature, which can induce bias in

the stress determination. To measure changes in resistance, each gage of the rosette

needs to be included in a “Wheastone bridge”, which is an electrical circuit comprising

4 resistances and a voltage excitation source (Vex) (Fig. 3.7). When all resistances are

equal, the voltage output (Vo) will be zero and the bridge is said to be balanced. Any

change in resistance in any arm of the bridge will result in a nonzero output voltage,

following the formula:

Vo =Vex (R1R3-R2R4) /(R2 +R1)(R3 +R4)

The output voltage (Vo) is characteristic of the deformation.

Fig. 3.6 - Operating principleof a strain gage

Fig. 3.7 - Quarter Wheastone bridge: the test resistance is orange


For a Type A rosette, each of the three gages is included in a quarter bridge (Fig. 3.7),

i.e.,a Wheastone bridge with a varying resistance (test gage) and three control resistances.

There are three gages and then three bridges connected per rosette, with each gage

facilitating the deformation measurement of the material along three different space axes.

For the Type C rosette, which is a six-element rosette, a paire of gages is connected to

two control resistances in a half-bridge configuration. Then, there are three half bridges

measuring the deformation along three different space axes.By using two varying strain

gages in the bridge, the effect of temperature can be further minimized. Indeed, the

second gage is placed transverse to the applied strain on the rosette, and in this way is

minimally affected by the deformation. However, any change in temperature will affect

both gages in the same way, allowing the elimination of the temperature effect in the

stress calculation. The greater output of the Type C rosette is also promoted by the higher

resistance of the gages.

Fig. 3.8 - Chronological evolution of the electrical measurement chain

Fig. 3.8a - Standardized quarter bridge with the traditional bridge completion accessories from National Instruments (National Instruments, Austin, Texas, USA) connected to a three-element strain gage rosette

Fig. 3.8b and c - Customized quarter bridge: each test gage is connected to three control gages, which are identical gages attached to undisturbed samples.


Usually, Wheastone bridges are built by connecting the strain gages to standard

components called bridge completion accessories, which contain the control resistances.

2.3.2 From standardized Wheatstone Bridge to customized Wheastone bridge

In the beginning,the Type A rosettes were connected to conventional industrial equipment:

the traditional bridge completion accessories from National Instruments (National

Instruments, Austin, Texas, USA) (Fig. 3.8a). Later, accessories from Vishay (Vishay,

Malvern, PA, USA) were purchased and tested. These accessories have to be installed

outside of the sample container.However,the accuracy and the stability of measurements

was not sufficient. Next, customized Wheatstone bridges were built using identical control

gages, i.e., standardized rosettes,attached to undisturbed samples. In practice, rosettes

were bonded on veneered metal control samples: these samples were exposed to the

same conditions as the sample under test, i.e., in the same container and the same

bath (Figs 3.8b and 3.8c for the Type A rosette, Fig. 3.8d for the Type C rosette).

Fig. 3.8d - Final electrical chain: customized half bridge,each pair of test gages is connected to two control gages, which are identical gages attached to an undisturbed sample.


2.3.3 Nanovoltmeters and filters

The very low voltage measurement was also improved by using specific custombuilt

electronic equipment. This equipment comprised a precision Vex source limiting

thermoelectric voltages (DC and AC current source 6221, Keithley Instruments, Inc,

Cleveland, Ohio, USA) and 3 precision nanovoltmeters to measure Vex and Vo

(Nanovolmeters 2182A, Keithley Instruments, Inc, Cleveland, Ohio, USA). Measurements

were filtered, removing external electrical noises, and recorded on a computer using NI

LabView software (National Instruments, Austin, Texas, USA).

2.4 Temperature control

A temperature control feature was eventually added in order to achieve sufficient

stability in the measurements. Indeed, any variation of the ambient temperature was

shown to generate instability and inacurracy in the strain measurements.

The specimens were placed in an aluminum container filled with silicon oil. Water and

glycerol were also tested, but silicon oil was finally found to be the best medium to

enhance the thermal conductivity, electrical insulation, and drilling lubrication. The

bath was thermally controlled and maintained at 36°C± 0.1°C (mouth temperature)

with a Eurotherm 3208 system (Eurotherm Ltd., Worthing, UK). The temperature at the

sample contact was recorded with a thermocouple connected to the NI LabView

data acquisition system. Samples were placed in the thermoregulated bath one night

before starting the experiment in order to allow the temperature of the system to stabilize.

2.5 Hole-drilling

An Isel CAD-CAM machine (CPM 3020, Houdan, France) was transformed for the drilling

procedure—the progression of the drill being controlled manually for greater accuracy—

and checked with a Digimatic indicator (Mitutoyo Corporation, Kawazaki, Japan).

Special attention was given to the following:

the use of a lubricant

the bur rotation speed and the type of drill: different procedures were tested to reduce

-

-


the apparition of cracks. Fine granulometry diamond burs were preferred. However, in

the few cases in which cracks had nevertheless occurred, abnormally large variations

in the microstrains were induced. If these were confirmed by optical microscopy, the

sample was eliminated.

the hole diameter: to increase strain sensitivity, the maximum allowable hole diameter

for the strain gage rosette type was used.

checking the hole diameter and concentricity after the experiment with an optical

microscope and motorized micrometer, Micro Controle CV 78 (Newport, Irvine, CA,

USA). The drill centering procedure was performed with loupes: centering with a binocular

magnifier was also tested but was found to be less efficient.

The time period between each step: from 30 minutes in the early stage of development,

to 10 minutes for the final procedure. This time was shown to be necessary to stabilize

any temperature fluctuations caused by the drilling process.

As for the rosette size used, the hole-drilling method can measure residual stresses to

depths of approximately 1.2 mm, the procedure was stopped at a depth of 1.5 mm.

2.6 Strain measurements, residual stress calculation

Voltage measurements were taken continuously during each step of the drilling

procedure and were recorded in an Excel spreadsheet (Microsoft Corporation,

Redmond,WA, USA). Mean values were evaluated for each strain gage based of the final

200 values (1 Hz acquisition) registered for each step. The mean strain measurements

for each step were calculated from the values registered from the three gages.

For a half bridge, strains are calculated from the registered voltages according to the

following formula:

Vo/Vex= -k ¢L/L

The corresponding profiles of residual stress vs. depth from the specimen surface were

then calculated according to ASTM Standard Test Method E837-08 using H-Drill software

(Vishay, Malvern, PA, USA), elaborated by Schajer. The software develops mathematical

models that translate the strain variations, registered at the surface by the gages, into

-

-

-


a stress profile. The mathematical models contain inverse solutions and calculate the

most likely profile fitting with the measured strains [3]. The H-Drill software offers three

different options for stress profile determination.The uniform stress profile method is used

when stresses do not vary with depth. This option follows the ASTM standard procedure

[2]. The power series method assumes that the residual stresses vary linearly with depth

from the specimen surface. The integral method is the best choice for our samples, as

this method provides a separate evaluation of residual stress for each depth.

However, those models do not take into account any flexure of the sample after the firing

procedure, which can induce additional compressive stresses at the surface. A specific

device, comprising a nanoprofilometer and a two-axis plate, was developed to measure

the sample flexure (Fig. 3.9). The maximum flexure of the samples was approximately 10

microns: this value does not significantly affect the residual stress calculation.

Finally, the HD method was successfully transferred to dental use, allowing accurate

measurement of residual stresses in veneer-metal and veneer-zirconia disks. Experiments

on 36 samples were needed to develop the method.Even if timeconsuming (at least two

days from the moment the samples are manufactured), the procedure enables the

measurement of very low voltages and the calculation of stresses of very low magnitude,

in the range of a few MPa.

Fig. 3.9 - Flexuremeasurement device


Prime MB. Residual stress measurement by successive extension of a slot: the crack

compliance method.Applied Mechanics reviews.1999;52(2):75-96.

ASTM. Determining residual stresses by the hole-drilling strain-gage method. Standard Test

Method E837-08 West Conshohocken,PA,USA: American Society for Testing and Materials;

2008.

Schajer GS, Prime MB. Use of inverse solutions for residual stress measurements. Journal of

Engineering Materials and Technology,Transactions of the ASME.2006;128(3):375-82.

References

1

2

3

4

a Department of Fixed Prosthodontics, Institute of Dentistry, University Hospital of Liège, ULg, 45 Quai G. Kurth, Liège 4020, Belgium

b Department of Mechanical Engineering, University of British Columbia, Vancouver, Canada

c Unité de Recherches Biomatériaux Innovants et Interfaces (URB2I-), Dental Surgery Faculty, University Paris Descartes, Paris, France

Published in: Dental Materials 2011; 27(5): 439-444

Amélie K. Mainjota,c, Gary S. Schajerb, Alain J. Vanheusdena, Michael J. Sadounc


Abstract

Objectives. Mismatch in thermal expansion properties between veneering ceramic

and metallic or high-strength ceramic cores can induce residual stresses and initiate

cracks when combined with functional stresses. Knowledge of the stress distribution

within the veneering ceramic is a key factor for understanding and predicting chipping

failures, which are well-known problems with Yttria-tetragonal-zirconia-polycrystal

based fixed partial dentures. The objectives of this study are to develop a method for

measuring the stress profile in veneering ceramics and to compare ceramic-fused-to-

metal compounds to veneered Yttria-tetragonal-zirconia-polycrystal ceramic.

Methods. The hole-drilling method, often used for engineering measurements, was

adapted for use with veneering ceramic. Because of the high sensitivity needed in

comparison with industrial applications, a high sensitivity electrical measurement chain

was developed.

Results. All samples exhibited the same type of stress vs. depth profile, starting with

compressive at the ceramic surface, decreasing with depth and becoming tensile at

0.5–1.0mm from the surface, and then becoming slightly compressive again. The zirconia

samples exhibited a stress depth profile of larger magnitude.

Significance. The hole drilling method was shown be a practical tool for measuring

residual stresses in veneering ceramics.

Keywords: Residual stress, Hole-drilling, Dental ceramic, Dental crowns, Zirconia

Veneering ceramic

4Residual stress in veneering ceramic by hole-drilling I Amélie Mainjot 65

Residual stress measurement in dental prostheses by hole-drilling. A. Mainjot, G. Schajer, A. Vanheusden, M. Sadoun.Residual Stress Summit, Tahoe City, California, September 2010. Poster session.

A new method to measure residual stresses in veneering ceramic. A.K Mainjot, A.J Vanheusden, M.J Sadoun.International Association for Dental Research general session. Barcelona, Spain, July 2010. Oral Session.

A new method to measure residual stresses in veneering ceramic. Première Journée du Réseau Francilien deRecherche en Sciences et Santé Orales, Paris, France, June 2010. Oral Session.

Development of a residual stresses measurement method in veneering ceramic layered on metal and zirconia frameworks.A.Mainjot, A.Vanheusden, M.Sadoun. Société Française des Biomatériaux Dentaires, annual congress, Reims, France, June2009. Oral Session.

Communicationsabout

this chapter

Yttria-tetragonal-zirconia-polycrystal (Y-TZP) was introduced as a framework material

for dental crowns and fixed partial dentures (FPDs) a decade ago because of its

esthetic and biocompatibility superiority over traditional metal materials.Y-TZP has high

strength and fracture toughness linked to its phase transformation potential, and compares

well with other ceramic framework materials such as glass ceramics containing leucite

or lithium disilicate crystals, or glass-infiltrated ceramics containing spinel, alumina or

crystalline alumina/zirconia.Nevertheless,during the last 5–6 years,clinical reports for Y-TZP

based crowns and FPDs have indicated a high rate of short-term failures linked to

veneering ceramic fracture (chipping), suggesting that the veneering ceramic is a

weak point in the restoration [1]. The mechanism of the chipping is complex and is not

well understood. Veneer chipping is reported more often than with ceramic-fused-to-

metal structures (PFMs) [2], and is impairing confidence in the use of zirconia-based

prostheses.

Mismatches in the thermal expansion properties of the veneer and framework, and

temperature gradients occurring during the cooling and solidification period of the firing

process, induce residual stresses within the structure [3]. The presence of these residual

stresses within the veneer greatly influences the strength and fracture characteristics of

veneered dental restorations, and they are likely a major influence on the observed

chipping. Therefore, knowledge of residual stresses in components and their formation

during manufacture is of great importance when designing and manufacturing

composite components [3].

Until now, stress profiles in dental prostheses have been studied only through mathematical

models. One of the challenges encountered by models is to account for the viscoelastic

behavior of the ceramic in the glass transition range, where thermo-physical properties

such as the coefficient of thermal expansion, glass transition temperature and viscosity

have strong temperature dependence and influence on residual stresses [4]. Asaoka

and Tesk [5] have developed analytical models that incorporate variations of the

coefficient of thermal expansion to study the influence of cooling rate and thermal

expansion coefficient mismatch on residual stress profiles. Recently, DeHoff et al. [6,7]

have investigated the impact of thermal expansion coefficient mismatch on residual

IntroductionResidual Stress in Veneering Ceramic I Amélie Mainjot

166

Fig. 4.1 - Hole-drilling method: (a) specimen cross-sectionbefore hole drilling, (b) specimen cross-section after hole drilling, showing exaggerated surface deformations caused by the local release of residual stresses.

stresses in glass-ceramic-based, three-unit, posterior FPDs by using viscoelastic finite

element stress analysis and a three-dimensional model. Subsequently, Swain highlighted

the critical importance of the cooling rate, thickness and thermal expansion coefficient

on residual stresses profile in bilayered structures composed of glass ceramics, alumina

and zirconia substrates [8]. In other work, Arman et al. [9] introduced a 3D finite element

analysis of PFMs crowns, incorporating the mechanical and thermal properties at various

temperatures including the equivalent heat transfer coefficients and the instantaneous

elastic moduli.

The “locked-in” character of residual stresses makes them challenging to measure

because there are no external loads to manipulate to reveal the internal stresses.

Various measurement techniques have been developed for industrial applications, but

have not been widely used for dental applications. A major class of residual stress

measurement methods used in industry is based on the removal of some stressed material

and the measurement of the resulting deformations in the adjacent material [10]. The

deformations are measured on the surface, typically using strain gages, from which the

residual stresses can be calculated. Fig. 4.1 illustrates this process for the hole-drilling

method, a residual stress measurement method widely used in industry.First,a strain gage

is attached to the specimen surface, as schematically shown in Fig. 4.1(a). Then a small

4Residual stress in veneering ceramic by hole-drilling I Amélie Mainjot 67

StrainGage

StrainGage

hole is drilled adjacent to the strain gage. This releases the residual stresses within the

hole, and causes the material around the hole to deform slightly, as shown in exaggerated

form in Fig. 4.1(b). The strain gage measures this deformation, from which it is possible to

calculate the size of the original residual stress. In practice, instead of a single strain

gage, a specially designed strain gage rosette consisting of three or six strain gages is

used. Fig. 4.2 shows a six-element strain gage rosette that was used in this study. The

drilled hole is at the center.

A major objective here is to transfer and adapt an effective industrial method for measuring

residual stresses to dental use, and to demonstrate the method for measurement of

residual stresses in veneer-metal (VM) and veneer-zirconia (VZr) structures. The hole-

drilling method is chosen because of its flexibility and convenience of use, its demonstrated

reliability in industrial applications, and the existence of a standardized test procedure

[11]. An additional objective is to compare the stress profile in PFM structures to Y-TZP

based structures to gain a better understanding of chipping problems encountered

with Y-TZP based crowns and FPDs in clinical practice.

Fig. 4.2Six-element Type C strain gage rosette installed on a VZr sample, after hole drilling.


4Materials and methods2

Starting Pre-drying t Heating Heating t Firing T Holding t VacuumT (°C) (min) rate (min) (°C) (min) holding t

closing t (°C/min) (min)

Alloy core oxidation 600 3 75 4 900 2 4Vita VM13 Opaque 600 2 75 4 900 1 4Vita VM13 Dentine 600 8 50 6 900 6 6

VM finished 600 8 50 6 900 6 6

Y-TZP coreregeneration firing 500 – 100 5 1000 15 –

Vita VM9 Effect bonder 500 6 75 6 950 1 6Vita VM9 Dentine 500 6 75 6 910 4 6

VZr finished 600 8 50 6 900 6 6

Table. 4.1Firing schedules for VM and VZrsamples.

Residual stress in veneering ceramic by hole-drilling I Amélie Mainjot 69


Bilayered disc samples composed of veneering ceramic sintered either on Y-TZP framework

(VZr, 9 samples), or on dental CoCr alloy framework (VM, 27 samples) were manufactured

following standard dental laboratory procedures and manufacturer’s recommendations.

CoCr core discs (Duceralloy C,DeguDent GmbH,Hanau,Germany),20 mm diameter,were

cast and ground sequentially with 80-grit, 180-grit and 500-grit silicon carbide discs (Struers

LabPol polishing machine, Copenhagen, Denmark) to a thickness of 1.00 ± 0.02 mm. The

surface to be veneered was sandblasted at 4 bars with 125 µm alumina particles.

Y-TZP core discs were made by initially cutting square slices out of a pre-sintered Y-TZP cylinder

(Acerma, Wissembourg, France), with a diamond wheel mounted on an IsoMet saw

(Buehler Ltd., Lake Bluff, IL, USA). These slices were rounded by polishing, and

sintered at 1530°C for 120 min with heating rate 10°C/min, and heating time 149 min

(Zircomat furnace,Vita Zahnfabrik,Bad Säckingen,Germany).The sintered Y-TZP discs were

ground and dimensioned in the same way than CoCr, but not sandblasted.

CoCr and Y-TZP discs were veneered respectively with Vita VM 13 and Vita VM9 feldspar

veneering ceramic (shade 3M2) (Vita Zahnfabrik,Bad Säckingen,Germany).A Vita Vacumat

4000 Premium furnace (Vita Zahnfabrik,Bad Säckingen,Germany) was used for all firing pro-

cedures,as summarized in Tab.4.1.All samples were baked on the same ceramic meshtray.

The sandblasted surfaces of the VM samples were oxidized before ceramic layering

according to the manufacturer’s guidelines. Vita VM 13 Opaque ceramic powder

mixed with Vita VM opaque fluid was applied to the substructure with a brush, and fired

to enhance the bond to the alloy surface. Three layers of dentin ceramic were fired

successively. This layering technique promotes adhesion between opaque and dentin

ceramic and reproduces the dental laboratory procedure. Samples were ground

sequentially with 80-grit, 180-grit and 500-grit silicon carbide discs to the thickness of

3.0 ± 0.02 mm to create a 2 mm thick ceramic layer on a 1mm thick framework.

For the VZr samples preparation, the Y-TZP discs were exposed to a “regeneration firing”,

which is a final thermal treatment of the core to reverse any phase transitions in the

zirconia due to the grinding procedures. A thin coat of Vita VM 9 Effect Bonder was

applied and fired on the surface to be veneered. Then, Vita VM 9 Base Dentine was

progressively layered on the effect bonder and samples were dimensioned in the same

way than VM samples.

After final polishing, all specimens were exposed to a last firing procedure (see Tab. 4.1)

to restore the residual stress profile through the veneering ceramic thickness. The samples

were tempered by opening the furnace door, as performed in dental laboratories, and

removed fromthe mesh-tray at 200°C.

2.2 Strain gage rosette installation

A specialized three-element Type A strain gage rosette (EA-06-062RE-120, Vishay,

Malvern, PA, USA) or a six-element Type C rosette (N2K-06-030RR-350/DP,Vishay, Malvern,

PA, USA) was installed on the center of the veneering ceramic surface (Fig. 4.1). To

promote the strain gage bond, the ceramic surface was prepared by etching with 10%

hydrofluoric acid for 1min, and was then cleaned for 5 min in an ultrasonic bath

containing 90% alcohol. The strain gage rosette was installed with M-Bond 200 Adhesive

(Vishay, Malvern, PA, USA), following the manufacturer’s instructions. The adhesive was

allowed to cure overnight to ensure complete curing. The installation was monitored

using an optical microscope.


2.3 Electrical measurement chain

The strains expected from the strain gages are very small and cannot be measured

with sufficient accuracy using conventional industrial equipment. A specialized data

acquisition system was therefore built where each strain gage was connected in a

Wheatstone bridge circuit with 3 control gages (identical gages attached to an undisturbed

sample). All gages and control rosettes were exposed to identical constant temperature

conditions. Finally, the very low voltage measurements were performed with specific

custom-built electronic equipment (Fig. 4.3) comprising a precision DC and AC current

source 6221 (Keithley Instruments, Inc., Cleveland, OH, USA) and 3 Nanovoltmeters

2182A (Keithley Instruments, Inc., Cleveland, OH, USA). Filtered measurements were

recorded on a computer using NI LabView software (National Instruments,Austin, TX, USA).

4

Fig. 4.3 – Electrical signal processing chain used for the strain gage measurements.


2.4 Temperature control

The specimens were placed in an aluminum container. After sample centering in the

drilling machine, the container was filled with silicon oil to enhance drilling lubrication,

thermal conductivity and electrical insulation. In addition, the silicon oil bath was thermally

controlled and maintained at 36 ± 0.1°C with a Eurotherm 3208 system (Eurotherm Ltd.,

Worthing, UK) to avoid the effects of any ambient temperature variations. Temperature

at the sample contact was recorded with a thermocouple connected to NI LabView

data acquisition system.

2.5 Hole-drilling

An Isel CAD-CAM machine (CPM 3020, Houdan, France) was used for the drilling

procedure. To increase strain sensitivity, the maximum allowable hole diameter for each

strain gage rosette type was used. A cylindrical diamond bur 1.9 mm in diameter

(Dumont Instruments & Co., Brussels, Belgium) was used with the three-element rosettes,

and a 2.5 mm diameter bur (Asahi Diamond Industrial Europe SAS, Chartres, France)

with the six-element rosettes. The bur rotation speed was 19,000 rpm. A hole was cut at

the center of the rosette in steps of 0.1 ± 0.01 mm, as measured by a Digimatic indicator

(Mitutoyo Corporation, Kawazaki, Japan). Hole diameter and concentricity were

checked after the experiment with an optical microscope and motorized micrometer,

Micro Controle CV78 (Newport, Irvine,CA,USA).The protocol of the hole-drilling method

was designed to eliminate/minimize crack initiation through choice of drilling process,

drill type and lubricant used. However, in the few cases where cracks had nevertheless

occurred, abnormal large variations of microstrains were induced. If these were

confirmed by optical microscopy, the sample was eliminated.


42.6 Strain measurements and residual stress calculation

Strain measurements were taken continuously during each step of the drilling procedure

and for 10 min afterwards. This time allowed stabilization of any temperature fluctuations

caused by the drilling process. The strain measurements were recorded in an Excel

spreadsheet (Microsoft Corporation, Redmond,WA, USA). Mean values were evaluated

for each strain gage based of the final 2000 values (1 Hz acquisition) registered for each

step. The corresponding profiles of residual stress vs. depth from the specimen surface

were then calculated according to ASTM Standard Test Method E837-08 using H-Drill

software (Vishay, Malvern, PA, USA).


3.1 Measured strains

Fig. 4.4 shows typical set of strain vs. hole depth measurements. The lower two curves,“VM

(A)” and “VZr (A)” respectively indicate the strains measured using the conventional Type

A rosettes on veneer-metal (VM) and veneer-zirconia (VZr) specimens. The strains are very

small, in the range of just 10 x 10-6, and it was difficult to measure them accurately, even

with the high-precision measurement system with strain resolution 0.1 x 10-6 that was developed

here. To improve measurement resolution, later measurements were done using six-element

Type C rosettes with slightly larger drilled holes, as indicated by the upper two curves “VM

(C)” and “VZr (C)” in Fig. 4.4. The three pairs of strain gages in the Type C rosettes were

connected as half-bridges, and had approximately five times the strain sensitivity of the

quarter-bridge arrangement used for the Type A rosettes.Because of the circular geometry

of the test specimens, the in-plane residual stresses are axisymmetric. Thus, all three strain

gages or strain gage pairs give similar strain vs.depth curves.For pictorial clarity,only a single

example curve for each case is displayed in Fig. 4.4.

Results3Residual Stress in Veneering Ceramic I Amélie Mainjot 74

Fig. 4.4 Typical hole-drilling strains vs. depth:VM = veneer-metal,VZr = veneer-zirconia,(C) = Type C rosette,(A) = Type A rosette.

43.2 Calculated residual stresses

Fig. 4.5 shows the residual stress vs. depth profiles calculated following the procedure

specified in ASTM E837-08, using the software H-Drill. Both VM and VZr samples exhibit

similar stress vs. depth profiles, starting with compressive stresses at the veneer surface,

steadily decreasing and becoming slightly tensile around 0.5 and 1mm in depth, and

then becoming slightly compressive at greater depths. The magnitude of the surface

compressive stresses was larger for the VZr than for the VM samples (-96 MPa vs.-40 MPa),

indicating greater thermal mismatch. The interior tensile stresses were similar in size,

around 4–8 MPa.


Fig. 4.5 - Typical residual stress vs. depth profiles inthe veneering ceramic layer of VM and VZr samples, measuredusing Type C straingage rosettes.

The ability to measure residual stresses is an important need when seeking to predict

the mechanical behavior of bilayered dental prostheses. The mechanical methods,

which are not well-known in the dental community, are based on the removal of some

stressed material and the measurement of the resulting deformations, and are suitable

for the determination of stress depth profiles in the range of some mm from the surface.

The results presented here demonstrate that the hole-drilling method is an effective

procedure for measuring residual stresses in dental prostheses. As these stresses are

small, sensitive and stable measurement equipment is needed. Because the

corresponding strains are also small, it is advantageous to use Type C (six-element)

rosettes instead of (three-element) Type A. Due to the use of half-bridges, increased

hole diameter (2.5 mm vs.1.9 mm),and increased resistance (350 ø vs.120 ø ),the six-element

rosette provided a five times greater output and greater stability than the three-element

rosette.

For the rosette size used, the hole-drilling method can measure residual stresses to

depths to 1.2 mm.The results showed a compressive stress at the surface, decreasing to

a slightly tensile area in the interior. The presence of compressive stress at the surface

confirms hypothesis formulated previously in the literature [8]. Two origins of stresses are

generally identified, thermal expansion mismatch and thermal gradients during the

cooling period [8]. In the case of ceramic fused to metal frameworks for crowns and

fixed partial dentures, the thermal expansion of the ceramic is generally slightly lower

than the framework, so that compressive stresses are developed in the ceramic, with

compensating tensile stresses developed within the framework. The surface compressive

residual stresses enhance the strength of the veneering ceramic by counteracting any

tensile stresses from applied loads, for example, during mastication. The veneering

ceramic is brittle and has a low tensile strength.

Another source of residual stresses in crowns and fixed partial dentures is the occurrence

of thermal gradients during cooling [8]. The size and distribution of these stresses

depends on material conductivity, thickness, and cooling rate. The resulting stresses

expected are compressive at the surface and tensile in the interior,as occurs during the

industrial manufacture of tempered glass.

Discussion4Residual Stress in Veneering Ceramic I Amélie Mainjot 76

4The measured stress profiles in Fig. 4.5 can be compared with those obtained by Swain

for a bilayer plate composed of glass ceramic, alumina or zirconia substrates [8]. His 2D

analytical model,applied to a 3 mm veneering ceramic layered on a zirconia framework,

combining thermal expansion mismatch and tempering effects, showed the same

nature of residual stress profile. His calculations also showed compressive stresses in the

surface and a tensile zone around 0.8 mm depth. However, the modeled stresses were

much larger than those measured here (maximum compressive stress: -120 MPa vs.- 96 MPa,

maximum tensile stress: 55 MPa vs. 4 MPa). The differences are likely due to differences

in the material thermal expansion coefficients, cooling rates, and the omission of viscous

relaxation effects and thermal gradients from the model. Beyond these

differences, the trends of the model results confirm that the distribution of the residual

stresses measured here is realistic. The modeling also illustrates the significance of making

experimental residual stresses measurements such as done here because small changes

in model input data can produce large changes in calculated results.

The presented method allows the study of different parameters influencing residual

stresses independently in PFMs and Y-TZP based structures, such as veneering ceramic

thickness, expansion coefficient mismatch, heat treatment parameters and cooling

rate. It could also be applied to other dental restorations materials such as composite

resins.


Residual stresses can substantially affect veneering ceramic performance. The hole

drilling method is an industrial method to measure these stresses, and it has been

adapted here for dental use. The method was successfully applied on bilayer disc

samples of veneering ceramic sintered either on Co–Cr alloy or on Y-TZP framework

characteristic of dental prostheses. The method allows stress depth profile measurement

in a range of 1.2 mm of depth from surface. Because of the small stresses present in dental

prostheses, sensitive and stable measurement equipment is required.

The residual stress measurements confirm the expectation from the previous literature

of compressive residual stresses in the veneering ceramic surface on VM and VZr

prostheses. The results showed a gradual shift to tensile residual stresses at 0.5–1 mm

depth from the surface. The measured stress profiles in the VZr samples showed residual

stresses of larger magnitude and concentration of tensile stresses in a smaller area,

which could be a potential source of the problems encountered in clinical practice.

Future studies on the mechanical behavior of Y-TZP restorations need to focus on the

influence of each independent parameter, such as veneering ceramic thickness,

expansion coefficient mismatch,heat treatment parameters and cooling rate,on residual

stress profile. The hole-drilling residual stress measurement method described here

provides an effective method for studying these parameters. It would then be possible

to compare the experimental measurements with the results of finite element analyses

of crowns and bridges.

Conclusions5Residual Stress in Veneering Ceramic I Amélie Mainjot 78

4Sailer I, Pjetursson BE, Zwahlen M, Hammerle CH.A systematic review of the survival

and complication rates of all-ceramic and metal-ceramic reconstructions after

an observation period of at least 3 years. Part II. Fixed dental prostheses. Clin Oral

Implants Res 2007;18(Suppl. 3 (June)):86–96.

Pjetursson BE, Sailer I, Zwahlen M, Hammerle CH.A systematic review of the survival

and complication rates of all-ceramic and metal-ceramic reconstructions after

an observation period of at least 3 years. Part I. Single crowns. Clin Oral Implants

Res 2007;18(Suppl. 3 (June)):73–85.

Wenzelburger M, Lopez D, Gadow R. Methods and application of residual stress

analysis on thermally sprayed coatings and layer composites. Surf Coat Technol

2006;201(5 (October)):1995–2001.

Asaoka K, Kuwayama N, Tesk JA. Influence of tempering method on residual stress

in dental porcelain. J Dent Res 1992;71(9 (September)):1623–7.

Asaoka K,Tesk JA.Transient and residual stress in a porcelain-metal strip. J Dent Res

1990;69(2 (February)):463–9.

DeHoff PH, Anusavice KJ, Gotzen N. Viscoelastic finite element analysis of an all-

ceramic fixed partial denture. J Biomech 2006;39(1):40–8.

DeHoff PH, Anusavice KJ. Viscoelastic finite element stress analysis of the thermal

compatibility of dental bilayer ceramic systems. Int J Prosthodont 2009;22

(1(January–February)):56–61.


dental crowns and fixed partial dentures. Acta Biomater 2009;5(5

(June)):1668–777.

Arman Y, Zor M, Gungor MA, Akan E, Aksoy S. Elastic–plastic finite elements analysis

of transient and residual stresses in ceramo-metal restorations. J Biomech

2009;42(13):2104–10.

Rendler NJ,Vigness I. Hole-drilling strain-gage method of measuring residual stresses.

Exp Mech 1966;6(12):577–86.

ASTM. Determining residual stresses by the hole-drilling strain-gage method.

Standard Test Method E837-08.West Conshohocken, PA, USA: American Society for

Testing and Materials; 2008. the experimental measurements with the results of

finite element analyses of crowns and bridges.

References

1

2

3

4

5

6

7

8

9

10

11


5

a Department of Fixed Prosthodontics,Institute of Dentistry, University Hospital of Liège, ULg, 45 Quai G. Kurth, Liège 4020, Belgium


c Unité de Recherches Biomatériaux Innovants et Interfaces (URB2I-), Dental Surgery Faculty, University Paris Descartes, Paris, France

Published in: Dental Materials 2011; 27 (9): 906-914

Amélie K. Mainjota,c,Gary S. Schajerb, Alain J. Vanheusdena,Michael J. Sadounc


Abstract

Objectives. The manufacture of dental crowns and bridges generates residual stresses within

the veneering ceramic and framework during the cooling process. Residual stress is an

important factor that control the mechanical behavior of restorations. Knowing the stress

distribution within the veneering ceramic as a function of depth can help the understanding

of failures, particularly chipping, a well-known problem with Yttria-tetragonal-zirconiapoly-

crystal based fixed partial dentures.The objective of this study is to investigate the cooling rate

dependence of the stress profile in veneering ceramic layered on metal and zirconia

frameworks.

Methods. The hole-drilling method, often used for engineering measurements, was

adapted for use with veneering ceramic. The stress profile was measured in bilayered

disc samples 20 mm in diameter, with a 0.7 mm thick metal or Yttria-tetragonal-zirconia-

polycrystal framework and a 1.5 mm thick veneering ceramic. Three different cooling

procedures were investigated.

Results. The magnitude of the stresses in the surface of the veneering ceramic was

found to increase with cooling rate,while the interior stresses decreased.At the surface,

compressive stresses were observed in all samples. In the interior, compressive stresses

were observed in metal samples and tensile in zirconia samples.

Significance.Cooling rate influences the magnitude of residual stresses.These can significantly

influence the mechanical behavior of metal-and zirconia-based bilayered systems.The

framework material influenced the nature of the interior stresses, with zirconia samples

showing a less favorable stress profile than metal.

Keywords: Residual stress, Hole-drilling, Zirconia, Veneering ceramic, Cooling rate

5Influence of cooling rate on residual stress I Amélie Mainjot 81

Influence of cooling rate on residual stress profile in veneering ceramic: measurement by hole drilling. A. Mainjot, G. Schajer, A. Vanheusden, M. Sadoun. Société Françophone des Biomatériaux Dentaires, annual congress, Toulouse,France, June 2011. Oral Session.

Communicationabout

this chapter

Mismatch in thermal expansion properties between core and veneering ceramic, and

thermal gradients occurring during the cooling/solidification period of the veneer firing

process induce residual stresses in crowns and fixed dental prostheses (FDPs) [1,2].These

residual stresses greatly influence the strength and fracture characteristics of metal-

and zirconia-based restorations. Particularly, the stress vs. depth profile is a key factor for

understanding veneering ceramic fractures [2].Such fractures are a major cause of failure

with Yttria-tetragonal-zirconia-polycrystal (Y-TZP) based FPDs [3]. As the veneering

ceramic cools from the surface to the center, thermal gradients induce non-uniform

solidification, thereby causing contraction mismatch within the ceramic [1]. The resulting

residual stresses are compressive at the surface and tensile in the interior, as also occurs

during the industrial manufacture of tempered glass. The size and distribution of the

stresses generated by thermal gradients depends on material conductivity, thickness,

and cooling rate at temperatures above the glass transition temperature Tg. In the case

of ceramic fused to metal frameworks, the thermal expansion coefficient (CTE) of the

ceramic is generally slightly lower than the framework so that during cooling from Tg to

room temperature, compressive stresses are developed within the ceramic, with

compensating tensile stresses developed within the framework [1]. The thermal gradients

and CTE mismatch effects combine to form the stress profile [2].

Until now, the cooling rate dependence of stress profiles in veneering ceramics layered

on a core material has been studied through mathematical models [2,4,5].Swain studied

the independent influence of the cooling rate,thickness and thermal expansion coefficient

on residual stresses profile within a simple bilayer model composed of glass ceramic,

alumina or zirconia substrates [2].The drawback of such models is the difficulty encountered

when mimicking variations of thermal gradients and materials thermal properties during

the firing process. Asaoka and Tesk [4] developed an analytical model of a porcelain-

fused-to-metal (PFM) beam that incorporates variations of several temperature-

dependent factors, and using a constant cooling rate on both specimen surfaces and

then a constant temperature distribution. They studied the influence of cooling rate

and thermal expansion coefficient mismatch on residual stress profiles. They found

compressive residual stresses at the surface of the ceramic and tensile residual stress at

the opaque interface. Stresses were found to increase with cooling rate.

Introduction1Residual Stress in Veneering Ceramic I Amélie Mainjot 82

5DeHoff et al. [5] developed a metal–ceramic disk model also based on the visco-elastic

theory and reported that tempering by air blasting induces large compressive stresses

in ceramic surface while cooling in air induces small tensile stresses. They also showed

that length of cracks induced in the surface of the ceramic by microhardness

indentation was smaller for tempered samples.

Tempering has often been reported to strengthen the veneering ceramic by developing

compressive stresses in the surface [6].As with tempered glass, these stresses are intended

to counteract external loads. In another hand some ceramic manufacturers have

changed their recommendations relating to the veneering process of zirconia frameworks.

A slow cooling procedure, from temperature of liquid phase sintering to Tg, furnace

closed, is now proposed for the last firing cycle (VM9 firing procedure, Vita Zahnfabrik,

Bad Säckingen, Germany, http://www.vitazahnfabrik. com/resourcesvita/shop/en/en

3055371.pdf). Manufacturers contend that this procedure reduces tensile zones, and

then cracks and chipping within the veneering thickness,but at this time there is no scientific

evidence to support this contention.

Recently a residual stress measurement method was introduced for dental applications [7].

The hole-drilling method, a standardized method developed for industrial applications

[8], was adapted for measuring residual stresses in the veneering ceramic. This method

is based on the removal of some stressed material and the measurement of the resulting

deformations in the adjacent material [9].The deformations are measured on the surface,

typically using strain gages, from which the residual stresses can be calculated. Stresses

are calculated from surface to depth, typically with 0.1mm steps, and giving a stress

profile within a 1.2mm depth.

The first objective of this study is to investigate the cooling rate dependence of stress

profiles in PFM and Y-TZP structures using the hole-drilling method. A second objective is

to investigate how modified firing procedures can influence mechanical behavior of

zirconia-based restorations, comparing stress profile in metal- and zirconia-based structures.

Influence of cooling rate on residual stress I Amélie Mainjot 83



(VZr, 6 samples), or on dental CoCr alloy framework (VM, 6 samples) were manufactured


CoCr core discs (Duceralloy C, DeguDent GmbH, Hanau, Germany), 20 mm diameter,

were cast and sequentially ground with 80-grit,180-grit and 500-grit silicon carbide discs

(Struers LabPol polishing machine,Copenhagen,Denmark) to a thickness of 0.70 ± 0.02 mm.

The surface to be veneered was sandblasted at 4 bars with 125 alumina particles.

Y-TZP core discs were cut out of a pre-sintered Vita In- Ceram YZ blocks (Vita Zahnfabrik,

Bad Säckingen, Germany), were rounded by polishing, and densely sintered at 1530°C

for 120 min with heating rate 10°C/min, and heating time 149 min (Zircomat furnace,

Vita Zahnfabrik, Bad Säckingen, Germany). The sintered Y-TZP discs were ground and

dimensioned in the same way as CoCr, but not sandblasted.

CoCr and Y-TZP discs were veneered respectively with Vita VM 13 and Vita VM9 feldspar

veneering ceramic (shade 3M2) (Vita Zahnfabrik, Bad Säckingen, Germany). A Vita

Vacumat 4000 Premium furnace (Vita Zahnfabrik, Bad Säckingen, Germany) was used

for all firing procedures, as summarized in Tab. 5.1. All samples were baked on the same

ceramic meshtray.

Materials and methods2Residual Stress in Veneering Ceramic I Amélie Mainjot84

Starting Pre-drying t Heating Heating t Firing T Holding t Vacuum Slow cooling Slow coolingT (°C) (min) rate (min) (°C) (min) holding t ending T(°C) rate

closing t (°C/min) (min) (°C/min)



Vita VM9 Effect bonder 500 6 75 6 950 1 6Vita VM9 Dentine 500 6 55 7.27 910 4 7.27

Last firing cycleClassic Cooling (CC) 600 8 50 6 900 6 6

Slow Cooling (SC) ° 10 900 6 Room T° 2Modified Cooling (MC) 600 8 50 6 900 6 6 600 33

Table. 5.1 Firing schedules for VM and VZr samples.

5The sandblasted surfaces of the VM samples were oxidized before ceramic layering



to enhance the bond to the alloy surface.Three layers of dentin ceramic were successively

fired. This layering technique promotes adhesion between opaque and dentin ceramic

and reproduces the dental laboratory procedure. Samples were sequentially ground

with 80-grit, 180-grit and 500-grit silicon carbide discs to the thickness of 2.2 ± 0.02 mm

to create a 1.5 mm thick ceramic layer on a 0.7 mm thick framework.




applied and fired on the surface to be veneered. Then, Vita VM 9 Base Dentin was



After final polishing, all specimens were exposed one by one to a last firing cycle. This

last firing cycle restores the residual stress profile through the veneering ceramic thickness.

All samples were placed in the same position, on the center of the mesh-tray and of the

furnace.Three different cooling schedules were performed. One sample of each group

(VM and VZr) was tempered from 900 °C to room temperature by opening the furnace

door, as classically performed in dental laboratories, and removed from the mesh-tray

at 200 °C (Classic Cooling, CC). This schedule was the one used during the veneering

layering process. In comparison with manufacturer recommendations, the firing

temperature was maintained 6 min in place of 1 min in order to reach 900 °C within the

framework. The second one was submitted to the slow cooling procedure recently

proposed by Vita for VM9 (Vita Zahnfabrik, Bad Säckingen, Germany). This procedure

comprised a slow cooling from 900 °C to 600 °C, furnace closed (Modified Cooling,

MC). The last one was cooled at 2 °C/min in a special furnace (Carbolite LMF 12/2,

Carbolite, Hope Valley, UK), from 900 °C to room temperature (Slow Cooling, SC). All firing

schedules are summarized in Tab. 5.1.


2.2 Determination of temperature profiles

Temperatures on the surface of the veneering ceramic and of the framework during

CC and MC were measured with S-type thermocouples and recorded with NI LabView

software (National Instruments, Austin, Texas, USA). The first thermocouple was placed in

contact with the center of the veneering ceramic.The second one was placed in contact

with the center of the framework surface, through the mesh-tray.

2.3 Dilatometric analysis

The CTE at temperatures above and below Tg for VM9 (Tg ~600°C, Ts ~670°C, following

the manufacturer), VM13 (Tg ~560–565°C, Ts ~635°C, following the manufacturer),

zirconia and CoCr alloy were measured using a single pushrod dilatometer (Netzsch DIL

402C) at a heating rate of 2 K min-1 up to 700°C. Beam samples were manufactured

for each veneering ceramic using a rectangular mold into which powders were

condensed and sintered. For Duceralloy C, a cylindrical sample was cast and fired for

30 min at 950°C to simulate veneering procedure and to eliminate internal stresses due

to casting. For Vita In-Ceram YZ, a rectangular sample was manufactured, sintered and

submitted to a regeneration firing procedure at 1000°C. Each sample was tested two

times.

2.4 Hole-drilling method

2.4.1 Strain gage rosette installation

A specialized six-element Type C rosette (N2K-06-030RR- 350/DP, Vishay, Malvern, PA,

USA) was installed on the center of the veneering ceramic surface. To promote the strain

gage bond, the ceramic surface was prepared by etching with 10% hydrofluoric acid

for 1 min, and was then cleaned for 5 min in an ultrasonic bath containing 90% alcohol.

The strain gage rosette was installed with M-Bond 200 Adhesive (Vishay, Malvern, PA,

USA), following the manufacturer’s instructions. The adhesive was allowed to cure

overnight to ensure complete curing. The installation was monitored using an optical

microscope.


52.4.2 Electrical measurement chain




Wheatstone bridge circuit with 3 control gages (identical gages attached to an undisturbed

sample). All gages and control rosettes were exposed to identical constant temperature

conditions. Finally, the very low voltage measurements were performed with specific

custom-built electronic equipment comprising a precision DC and AC current source

6221 (Keithley Instruments, Inc., Cleveland, OH, USA) and 3 Nanovoltmeters 2182A

(Keithley Instruments, Inc., Cleveland, OH, USA). Filtered measurements were recorded

on a computer using NI LabView software (National Instruments, Austin, TX, USA).

2.4.3 Temperature control




controlled and maintained at 36 ± 0.1°C with a Eurotherm 3208 system (Eurotherm

Ltd.,Worthing, UK) to avoid the effects of any ambient temperature variations.

Temperature at the sample contact was recorded with a thermocouple connected to

NI LabView data acquisition system.

2.4.4 Hole-drilling


procedure. To increase strain sensitivity, the maximum allowable hole diameter for the

strain gage rosette type was performed with a 2.5 mm diameter cylindrical bur (Asahi

Diamond Industrial Europe SAS, Chartres, France). The bur rotation speed was 19,000

rpm. A hole was cut at the center of the rosette in steps of 0.1 ± 0.01 mm, as measured

by a Digimatic indicator (Mitutoyo Corporation, Kawazaki, Japan). Hole diameter and

concentricity were checked after the experiment with an optical microscope and

motorized micrometer, Micro Controle CV 78 (Newport, Irvine, CA, USA). The protocol of

the hole-drilling method was designed to eliminate/minimize crack initiation through

choice of drilling process, drill type and lubricant used. However, in the few cases where



cracks had nevertheless occurred, abnormal large variations of microstrains were

induced. If these were confirmed by optical microscopy, the sample was eliminated.

2.4.5 Strain measurements and residual stress calculation


and for 10 min afterwards. This time allowed stabilization of any temperature fluctuations




step. The corresponding profiles of residual stress vs. depth from the specimen surface

were then calculated according to ASTM Standard Test Method E837-08 using HDrill

software (Vishay, Malvern, PA, USA). For the rosette size used, the hole-drilling method

can measure residual stresses to depths to 1.2 mm.

589

Results3Figs.5.1 and 5.2 illustrate profiles of veneering ceramic and framework surface temperature

for VM and VZr samples vs. time for the CC and MC procedures. When the veneering

ceramic surface was at the softening point temperature (beyond 600°C),CC procedures

induced larger mismatches than MC procedures (around 175°C for CC and around 25°C

for MC) (Figs. 5. 3 and 5.4). Largest mismatches were observed when the veneering

ceramic surface was below Tg. For VM samples, the largest mismatch was higher for MC

procedure (251°C when the veneering surface temperaturewas at 231°C) than for CC

procedure (211°C when the veneering surface temperature was at 435°C). For VZr samples

the largest mismatch was higher for the CC procedure (227 °C when the veneering surface

temperature was at 420 C) than for the MC procedure (200°C when the veneering

surface temperature was at 309°C).

Measured CTE between 50 °C and 500 °C were respectively: 10.12 x 10-6 K-1 for VM9,13.01

x 10-6 K-1 for VM13, 11.39 x 10-6 K-1 for Vita In-Ceram YZ, and 15.32 x 10-6 K-1 for Duceralloy C.

Figs.5.5 and 5.6 show the residual stress vs.depth profiles calculated following the procedure

specified in ASTM E837-08, using the software H-Drill, respectively for VM and VZr samples,

for each cooling rate.

All VM and VZr samples exhibit compressive stresses in the veneer surface.The magnitude

of these stresses increased with cooling rate and was larger for VZr samples.

For VM samples,compressive stresses were found to decrease with depth for the CC and

the MC,becoming more compressive again from 0.5 mm from the surface.For the SC, the

compressive stresses were found to be constant till 0.3–0.4 mm from the surface and then

increased with depth. The magnitude of compressive stresses at 0.3mm from framework

was found to vary inversely to cooling rate.

For VZr samples, stresses were found to decrease with depth, reaching zero at 1.3 mm in

depth for the CC, or becoming tensile around 0.5 and 0.6 mm from the surface for the MC

and the SC. In depth, the magnitude of tensile stresses for MC and SC was found to vary

inversely to cooling rate.

Influence of cooling rate on residual stress I Amélie Mainjot

Residual Stress in Veneering Ceramic I Amélie Mainjot90

Fig. 5.2 - Profiles of veneering ceramic and framework surface temperature for VZr samples vs. time, for the CC and MC procedures.

Fig. 5.1 - Profiles of veneering ceramic and framework surface temperature for VM samples vs. time, for the CC and MCprocedures.

Influence of cooling rate on residual stress I Amélie Mainjot591

Fig. 5.3 - Temperature mismatch (framework surface temperature minusveneering ceramic surface temperature) vs. temperature of the veneering ceramic surface, for VM samples, for the CC and MC procedures.

Fig. 5.4 - Temperature mismatch (framework surface temperature minus veneering ceramic surface temperature) vs.temperature of the veneering ceramic surface, for VZr samples, for the CC and MC procedures.


Fig. 5.5 - Residual stress vs.depth profiles inthe veneeringceramic layer of VM samples.CC= Classic Cooling,MC= Modified Cooling,SC = Slow Cooling.

Fig. 5.6 - Residual stress vs.depth profiles inthe veneering ceramic layer of VZr samples.CC= Classic Cooling,MC=Modified Cooling,SC = Slow Cooling.


Discussion4Due to the viscoelastic properties of the veneering ceramic, some physical and thermal

properties of the material greatly vary around Tg. Therefore the cooling rate around Tg

can substantially influence residual stress during the solidification process of bilayers [6],

as was observed in the measured stress profiles in both VM and VZr samples.

The stress profiles observed in VM samples (Fig. 5.5) confirmed the hypotheses formulated

previously in the literature about the influence of CTE mismatch and thermal gradients

throughout the veneer layer at the temperatures where the veneering ceramic transforms

from a visco-elastic to a solid state [1]. The thermal gradient during the solidification

process increases with cooling rate, as suggested by measured temperatures on veneer

and framework surfaces during NC and MC procedures (Figs. 5.1 and 5.3). For the SC

procedure, the thermal gradient is canceled through the temperature homogenization

induced by the very low cooling rate (2 °C/min). The higher the cooling rate, the higher

was the tempering effect and the magnitude of compressive residual stresses in the surface

[4–6]. For NC and MC, these stresses decreased with depth to reach -3 MPa at 0.5 mm

from surface. In the second part of the profile, from 0.4 to 0.7 mm, compressive stresses

increased,as expected in respect with the CTE mismatch effect [1]. Indeed, following the

manufacturer’s information, the range of CTE is respectively: 13.1–13.6 x 10-6 K-1 for VM13

dentin,and 14–14.3 x 10-6 K-1 for Duceralloy C. In this study measured CTE were 13.01 x 10-6 K-1

for VM13 and 15.3 x 10-6 K-1 for Duceralloy C.The positive mismatch between framework and

veneering ceramic explains the progressive re-emergence of compressive stresses

observed between 0.4–0.7 mm and 1.2 mm from surface.The higher was the cooling rate,

the lower was the magnitude of these stresses which were counterbalanced by the

increasing tempering compressive stresses in the surface, as indicated in the literature

[1]. This was also confirmed by the stress profile induced with the slow cooling procedure:

the tempering effect in the surface was canceled and larger compressive stresses in

depth were revealed. Indeed CTE mismatch is almost effective when the two layers cool

in the same time.When a thermal gradient exists, solidification occurs progressively from

surface to interior, reducing the magnitude of in-depth CTE mismatch induced stresses.

Moreover, due to viscoelastic properties of the veneering ceramic, its modulus of

elasticity decreases and its CTE increases with temperature above Tg [10]. Therefore a

higher cooling rate around Tg increases veneering ceramic contraction during solidification

aResidual Stress in Veneering Ceramic I Amélie Mainjot 94

and then compressive stresses observed in the surface, but decreases the CTE mismatch

between core and veneer and then in-depth compressive stresses. The nature of stresses

observed near the framework contrast with results of Asaoka and Tesk [4] who calculated

tensile stresses in a tempered PFM beam.DeHoff et al. [5] found compressive stresses near

the framework of metal–ceramic disks following a normal cooling procedure, but they

showed tensile stresses in the surface.As reported by the authors, their calculations were

limited since they did not take into account realistic temperature distributions within the

sample, values of CTE up to 600°C, and the effect of cooling rate on ceramic

microstructure. The calculated stresses in both studies did not exceed 40 MPa, as

observed in the present study for CC and MC procedures (max -35 MPa).

Stress profiles for VZr samples (Fig. 5.6) were similar to VM samples within the surface,

despite larger compressive stresses registered for MC and SC procedures (-26 MPa and

-29 MPa, respectively versus -13 MPa and -12 MPa in VM samples). In contrast, profiles

from 0.5 mm depth were opposed to VM samples. Indeed compressive stresses were

found to turn progressively into tensile stresses, the magnitude of tensile stresses increasing

inversely to cooling rate. However measured CTE were 10.12 x 10-6 K-1 for VM9

(vs. 8.8–9.2 x 10-6 K-1 following the manufacturer) and 11.39 x 10-6 K-1 for Vita In-Ceram YZ

(vs. 10.5 x 10-6 K-1 following the manufacturer). The CTE mismatch between veneering

ceramic and zirconia is positive as for VM samples. Due to this positive mismatch and the

elastic moduli of components, which are similar to VM 13 and CoCr alloy, interior com-

pressive stresses were expected for VZr samples, as with VM samples. This profile cannot

be explained by the effect of the viscoelastic processes on residual stress, as shown by

the SC stress profile. Indeed, the slow cooling rate reduced the effect of physical and

thermal properties variations. Additionally this profile cannot be explained by the effect

of the differences between metal and zirconia thermal properties, such as conductivity

or diffusivity, since it is also reduced by the SC procedure. A first hypothesis could be that

during the cooling procedure the framework was colder than the veneering ceramic.This

hypothesis was contradicted by the observed temperature profiles (Fig. 2) and by the

results of the slow cooling procedure,whichwas intended to cancel the thermal gradient.

Another hypothesis could be the possible crystalline transformation of the zirconia, as

described by Tholey et al.,who observed monoclinic crystals at the veneer-zirconia inter-


face with SEM and XRD2 micro-diffraction [11,12]. The martensitic transformation from

tetragonal to monoclinic crystals generates a volume increase of about 4% [11], which

can counteract the CTE mismatch effect and induce tensile stresses during the cooling

process. Tholey et al. linked this transformation to moisture content used for the build-up

of the veneering ceramic, which can degrade Y-TZP during its heating. In any case, if

some monoclinic grains would appear during heating time, they would not induce

residual stress because the expanded grains would be present before the cooling

process.Another explanation could be that this martensitic transformation is enhanced

by the stress developed at the interface during the cooling process due to the CTE

mismatch effect. Indeed, the in-depth compressive stresses in the veneering ceramic are

balanced by tensile stresses in the framework, which can trigger the transformation. The

stress-induced volume expansion is instantaneous and can secondarily generate residual

tensile stresses in place of compressive stresses in the veneering ceramic.The amount of

stress needed to generate the transformation in the absence of water is poorly understood.

Li et al. [13] calculated that 415 MPa can induce the transformation in a bending test

configuration.In the same study they showed that tensile stress accelerates the degradation

of Y-TZP in hot water (100°C). They observed that a tensile stress of 100 MPa, in a bending

configuration in hot water during 5h, generated a monoclinic volume of 5%. It can be

hypothized that the combination of water and temperature during the heating process,

and particularly around 250°C when the transformation is more rapid [14], weakens the

zirconia surface, generating monoclinic grains [12], decreasing the stability of the

tetragonal phase because of the infiltration of water through grain boundaries [13], leading

to microcracks [14]. The microcracks, which can also be originated from the grinding

process of frameworks after sintering, cannot heal with temperature and

may secondarily,during the cooling process,promote the stress-induced transformation.

The determined SC stress profile supports the transformation hypothesis. Indeed, during

SC, solidification and then contraction of the veneering ceramic is presumed to occur

in the same time through the entire thickness of the material, generating more stresses

at the interface due to the CTE mismatch effect.


The results of the present study support the hypothesis that a modified cooling procedure

could promote delaminations of the veneering ceramic, i.e. cohesive fracture of the

ceramic near the interface with the zirconia due to the presence of residual tensile

stresses.This was also suggested in the study of Gostemeyer et al.,which describes more

interfacial adhesion failures with slow cooling firing schedules introduced by manufacturers

[1]. This phenomenon seems to be influenced by the framework thickness; as shown in

a previous study relating to samples with a 1 mm thick framework, the in-depth profile

was more favorable [7].

Finally it must be noticed that the observed stress profiles differ from those calculated

by Swain in bilayered structures comprising a 3 mm thick layer of veneering ceramic and

a 1 mm thick zirconia framework, cooled at 50°C/s [2]. Starting with compressive stresses

in the surface, he described tensile stresses from 0.7 mm to 2.8 mm in depth, these tensile

stresses decreasing with depth and becoming compressive again near the framework.

Differences can be explained by the thickness of framework and veneering ceramic,

by the fact that he did not take into account the possible phase transformation of

zirconia, and finally by the difficulty to model such complex thermal processes as the

veneering ceramic firing procedure.


Conclusions

The residual stress measurements confirm the expectation from the previous literature

of compressive residual stresses in the veneering ceramic surface on VM and VZr

prostheses and illustrate the tempering effect on the magnitude of these stresses. The

results also describe the resulting stress profile generated by the combined effect of

tempering, CTE mismatch and viscoelastic effect in VM samples, showing in depth

compressive stresses. Variations of this stress profile depending on the cooling rate are

highlighted. Surprisingly,VZr samples, with a 0.7 mm framework and a 1.5 mm veneering

ceramic thickness processed using MC and/or SC procedure show a gradual shift to

tensile residual stresses at 0.5 mm depth from the surface. This could be a potential

explanation to delaminations encountered in clinical practice. A hypothesis for the

crystalline transformation of zirconia is proposed to explain these findings, a hypothesis

that needs to be confirmed by further structural analysis, such as X-ray diffraction or

Raman spectroscopy at the interface between zirconia and veneering ceramic. Finally,

this study highlights the complexity and the multifaceted characteristic of residual stress

development in veneering ceramic.

Further studies on the mechanical behavior of Y-TZP restorations are needed to

consider other residual stress influencing parameters such as component thickness and

on the possible transformation of zirconia during the firing procedure.

5


Gostemeyer G,Jendras M,Dittmer MP,Bach FW,Stiesch M,Kohorst P. Influence of cooling rate

on zirconia/veneer interfacial adhesion.Acta Biomater 2010;6 (December (12)):4532–8.

Swain MV. Unstable cracking (chipping) of veneering porcelain on all-ceramic dental

crowns and fixed partial dentures.Acta Biomater 2009;5 (June (5)): 1668–77.

Al-Amleh B,Lyons K,Swain M.Clinical trials in zirconia: a systematic review.J Oral Rehabil

2010;37 (August–October (8)):641–52.

Asaoka K, Tesk JA. Transient and residual stress in a porcelain-metal strip. J Dent Res

1990;69 (February (2)):463–9.

DeHoff PH, Anusavice KJ,Vontivillu SB. Analysis of tempering stresses in metal–ceramic

disks. J Dent Res 1996;75 (February (2)):743–51.

Taskonak B, Borges GA, Mecholsky Jr JJ, Anusavice KJ, Moore BK,Yan J. The effects of

viscoelastic parameters on residual stress development in a zirconia/glass bilayer

dental ceramic. Dent Mater 2008;24 (September (9)):1149–55.

Mainjot AK, Schajer GS, Vanheusden AJ, Sadoun MJ. Residual stress measurement in

veneering ceramic by hole-drilling. Dent Mater 2011; (January 11).

ASTM. Determining residual stresses by the hole-drilling strain-gage method. Standard Test

Method E837-08 West Conshohocken,PA,USA: American Society for Testing and Materials; 2008.

Schajer GS, Prime MB. Use of inverse solutions for residual stress measurements. J Eng

Mater Technol Trans ASME 2006;128 (3):375–82.

Arman Y, Zor M, Gungor MA, Akan E, Aksoy S. Elastic–plastic finite elements analysis of

transient and residual stresses in ceramo-metal restorations. J Biomech 2009;42

(September 18 (13)):2104–10.

Tholey MJ, Swain MV,Thiel N. SEM observations of porcelain Y-TZP interface.Dent Mater

2009;25 (July (7)):857–62.

Tholey MJ,Berthold C,Swain MV,Thiel N.XRD2 micro-diffraction analysis of the interface

between Y-TZP and veneering porcelain: role of application methods. Dent Mater

2010;26 (June (6)):545–52.

References

1

2

3

4

5

6

7

8

9

10

11

12

98


Li JG, Zhang LM, Shen Q, Hashida T. Degradation of yttria stabilized zirconia at 370K

under a low applied stress.Mater Sci Eng A-Struct Mater Prop Microst Process 2001;297

(January (1–2)):26–30.

Chevalier J,Gremillard L,Virkar AV,Clarke DR.The tetragonal-monoclinic transformation in

zirconia: lessons learned and future trends.J Am Ceram Soc 2009;92 (September (9)):1901–20.

13

14

6

a Department of Fixed Prosthodontics, Institute of Dentistry, University of Liège (ULg CHU), 45 Quai G. Kurth, Liège 4020, Belgium


c Unité de Recherches Biomatériaux Innovants et Interfaces (URB2I-EA4462),Université Paris Descartes, Sorbonne Paris Cité, Faculté deChirurgie Dentaire, Paris, France.


Submitted to Dental Materials


Abstract

Objectives: The veneering process of frameworks induces residual stresses and can initiate

cracks when combined with functional stresses. The stress distribution within the veneering

ceramic as a function of depth is a key factor influencing failure by chipping. This is a

well-known problem with Yttria-tetragonal-zirconia-polycrystal based fixed partial dentures.

The objective of this study is to investigate the influence of veneer thickness on the stress

profile in zirconia- and metal-based structures.

Methods: The hole-drilling method, often used for engineering measurements, was

adapted for use with veneering ceramic. The stress profile was measured in bilayered

disc samples of 20 mm diameter, with a 1 mm thick zirconia or metal framework.

Different veneering ceramic thicknesses were performed: 1 mm, 1.5 mm, 2 mm, 2.5 mm

and 3 mm.

Results: All samples exhibited the same type of stress vs. depth profile, starting with

compressive at the ceramic surface, decreasing with depth up to 0.5-1.0 mm from the

surface,and then becoming compressive again near the framework,except for the 1.5 mm-

veneered zirconia samples which exhibited interior tensile stresses. Stresses in the

surface of metal samples were not influenced by veneer thickness. Variation of interior

stresses at 1.2 mm from the surface in function of veneer thickness was inverted for

metal and zirconia samples.

Significance: Veneer thickness influences in an opposite way the residual stress profile

in metal- and in zirconia-based structures. A three-step approach and the hypothesis

of the crystalline transformation are discussed to explain the less favorable residual

stress development in zirconia samples.

Keywords: Residual stress; Hole-Drilling; Zirconia; Veneering ceramic; Thickness

6Influence of veneer thickness on residual stress I Amélie Mainjot 101

Influence of veneer thickness on residual stresses in zirconia prostheses. A. Mainjot, G. Schajer, A. Vanheusden, M.Sadoun. Academy of Dental Materials Meeting, Trieste, Italy, October 2010. Poster Session. Published in : DentalMaterials (2010), vol. 26S, p. e55.

Communicationabout

this chapter

The manufacture of crowns and fixed dental prostheses (FDPs) generates residual stresses

within the veneering ceramic and framework during the cooling process. These “locked-

in”stresses add to functional loads and are an important predictive factor for the mechanical

behavior of restorations,as compressive stresses reinforce ceramic and tensile stresses facilitate

the initiation and the propagation of cracks. Knowledge of the residual stress distribution

within the veneering ceramic as a function of depth is a key factor for understanding and

predicting chipping and delaminations. Such fractures are reported as an important

cause of short-term clinical failure with Yttria-tetragonal-zirconia-polycrystal (Y-TZP) based

FPDs [1]. These complications are more frequent in zirconia-based restorations than in

ceramic-fused-tometal structures (PFMs) [2].

Stress profile in the veneering ceramic is generated by the successive effects of the thermal

gradients occurring during the cooling/solidification period of the veneer liquid phase sintering

process,and the mismatch in thermal expansion properties between core and veneering

ceramic [3] [4]. In the case of ceramic fused to metal frameworks, the thermal expansion

coefficient (CTE) of the ceramic is generally slightly lower than the framework so that during

cooling from Tg to room temperature, interior compressive stresses are developed within

the ceramic near the framework,with compensating tensile stresses developed within the

framework. The second parameter influencing stress profile is thermal gradients, which

induce non-uniform solidification,from the surface to the center,thereby causing contraction

mismatch within the ceramic.

The resulting residual stresses are compressive in the surface of the veneering ceramic, the

magnitude of these stresses decreasing with depth and being influenced by the cooling

rate around the glass transition temperature Tg due to the effect of viscoelastic parameters

[4] [5] [6].Thermal gradients are determined by cooling rate but also by material thickness

and conductivity.

The cooling rate dependence of stress profiles in veneering ceramics layered on different core

materials has been studied through mathematical models [7] [5] [8] and more recently

through residual stress measurements in veneered metal and zirconia disk samples [4].

Introduction1Residual Stress in Veneering Ceramic I Amélie Mainjot 102


To author’s knowledge, the influence of veneering ceramic thickness on residual stress in

metal and zirconia-based structures has been not well described. However core/veneer

thickness ratio and framework design is reported to influence all ceramic restorations reliability,

particularly Y-TZP restorations [9].Ensuring an optimal support to veneering ceramic with an

anatomical framework and avoiding excessive thickness is recommended to prevent

chipping. Swain calculated the independent influence of the cooling rate, thickness and

thermal expansion coefficient on residual stresses profile [7]. His 2D mathematical model

was a bilayer composed of glass ceramic,alumina or zirconia substrates.Among the three

factors studied, thickness predominated as the most influencing parameter.He found that

thick layers of veneering ceramic on framework with low thermal diffusivity, such as Y-TZP,

promotes the development of high tensile interior residual stresses which may result in chipping.

Bonfante et al [10] studied stresses in a 3D zirconia crown model. He observed slight tensile

stress fields at the interface of the core and the veneer layer of both standard and modified

core systems characterized by different veneer thicknesses. But models are limited by the

difficulty encountered when mimicking such a complex thermal process as the veneer firing

procedure.Particularly the cited models did not account for thermal gradients and variations

of thermo-physical properties of ceramic in the glass transition range.

A residual stress measurement method is now available for dental applications [11]. The

hole-drilling method, a standardized method developed for industrial applications [12],

was adapted for measuring residual stresses in the veneering ceramic. This method is

based on the removal of some stressed material and the measurement of the resulting

deformations in the adjacent material [13].The deformations are measured on the surface,

typically using strain gages,from which the residual stresses can be calculated.Stresses are

calculated from surface to depth, typically with 0.1 mm steps, and giving a stress profile

within a 1.2 mm depth. This method was used previously to study the influence of cooling

rate on residual stress profile in veneered metal and zirconia samples [4].

The first objective of this study is to investigate the veneer thickness dependence of stress

profiles in PFM and Y-TZP structures using the hole-drilling method.A second objective is to

understand how the veneer thickness of zirconia-based restorations can influence their

mechanical behavior, comparing stress profile in metaland zirconia-based structures.



(VZr,10 samples),or on dental CoCr alloy framework (VM,10 samples) were manufactured


CoCr core discs (Duceralloy C, DeguDent GmbH, Hanau, Germany), 20 mm diameter,

were cast and sequentially ground with 80-grit,180-grit and 500-grit silicon carbide discs

(Struers LabPol polishing machine, Copenhagen, Denmark) to a thickness of 1.0 ± 0.02

mm.The surface to be veneered was sandblasted at 4 bars with 125 µ alumina particles.

Y-TZP core discs were cut out of a pre-sintered Vita In-Ceram YZ blocks (Vita Zahnfabrik,

Bad Säckingen, Germany), were rounded by polishing, and densely sintered at 1530°C

for 120 min with heating rate 10°C/min, and heating time 149 min (Zircomat furnace,

Vita Zahnfabrik, Bad Säckingen, Germany). The sintered Y-TZP discs were ground and

dimensioned in the same way as CoCr, but not sandblasted.

CoCr and Y-TZP discs were veneered respectively with feldspar veneering ceramic: Vita

VM 13 (Tg ~ 560-565°C,Ts ~ 635°C, following the manufacturer) and Vita VM9 (Tg ~ 600°C ,

Ts ~ 670°C, following the manufacturer) (shade 3M2) (Vita Zahnfabrik, Bad Säckingen,

Germany). A Vita Vacumat 4000 Premium furnace (Vita Zahnfabrik, Bad Säckingen,

Germany) was used for all firing procedures, as summarized in Tab. 6.1.All samples were

baked on the same ceramic mesh-tray.

Materials and methods2Residual Stress in Veneering Ceramic I Amélie Mainjot 104

Starting Pre-drying t Heating Heating t Firing T Holding t VacuumT (°C) (min) rate (min) (°C) (min) holding t

closing t (°C/min) (min)


VM finished 600 8 50 6 900 6 6


Vita VM9 Effect bonder 500 6 75 6 950 1 6Vita VM9 Dentine 500 6 55 7.27 910 4 7.27

VZr finished 600 8 50 6 900 6 6

Tab 6.1Firing schedulesfor VM and VZrsamples.


The sandblasted surfaces of the VM samples were oxidized before ceramic layering



to enhance the bond to the alloy surface.Three layers of dentin ceramic were successively

fired. This layering technique promotes adhesion between opaque and dentin ceramic

and reproduces the dental laboratory procedure. Samples were sequentially ground

with 80-grit, 180-grit and 500-grit silicon carbide discs to obtaineither a 1.0 mm (n=2), a

1.5 mm (n=2), a 2 mm (n=2), a 2.5 mm (n=2) or a 3.0 mm (n=2) thick ceramic layer.




applied and fired on the surface to be veneered. Then, Vita VM 9 Base Dentine was



After final polishing, all specimens were exposed one by one to a last firing cycle (see

Tab. 6.1). This last firing cycle restores the residual stress profile through the veneering

ceramic thickness. All samples were placed in the same position, on the center of the

mesh-tray and of the furnace. They were tempered from 900°C to room temperature

by opening the furnace door, as typically performed in dental laboratories, and

removed from the mesh-tray at 200°C. In compliance with the manufacturer’s

recommendations, the firing temperature was maintained six minutes in place of one

minute in order to reach 900°C within the framework.

2.2 Hole-Drilling method


A specialized six-element Type C rosette (N2K-06-030RR-350/DP,Vishay,Malvern,PA,USA)

was installed on the center of the veneering ceramic surface. To promote the strain

gage bond, the ceramic surface was prepared by etching with 10% hydrofluoric acid

for 1 min, and was then cleaned for 5 min in an ultrasonic bath containing 90% alcohol.


The strain gage rosette was installed with M-Bond 200 Adhesive (Vishay, Malvern, PA,

USA), following the manufacturer’s instructions. The adhesive was allowed to cure

overnight to ensure complete curing. The installation was monitored using an optical

microscope.





Wheatstone bridge circuit with 3 control gages (identical gauges attached to an

undisturbed sample). All gages and control rosettes were exposed to identical constant

temperature conditions. Finally, the very low voltage measurements were performed

with specific custom-built electronic equipment comprising a precision DC and AC current

source 6221 (Keithley Instruments, Inc, Cleveland, Ohio, USA) and 3 Nanovoltmeters

2182A (Keithley Instruments, Inc, Cleveland, Ohio, USA). Filtered measurements were

recorded on a computer using NI LabView software (National Instruments,Austin,Texas,USA).





controlled and maintained at 36°C± 0.1°C with a Eurotherm 3208 system (Eurotherm

Ltd, Worthing, UK) to avoid the effects of any ambient temperature variations.

Temperature at the sample contact was recorded with a thermocouple connected to

NI LabView data acquisition system.

2.2.4 Hole-drilling


procedure. To increase strain sensitivity, the maximum allowable hole diameter for the

strain gage rosette type was performed with a 2.5 mm diameter cylindrical bur (Asahi

Diamond Industrial Europe SAS, Chartres, France). The bur rotation speed was 19000

rpm. A hole was cut at the center of the rosette in steps of 0.1 ± 0.01 mm, as measured


by a Digimatic indicator (Mitutoyo Corporation, Kawazaki, Japan). Hole diameter and

concentricity were checked after the experiment with an optical microscope and

motorized micrometer, Micro Controle CV 78 (Newport, Irvine, CA, USA). The protocol of

the hole-drilling method was designed to eliminate/minimize crack initiation through

choice of drilling process, drill type and lubricant used. However, in the few cases where

cracks had nevertheless occurred, abnormal large variations of microstrains were

induced. If these were confirmed by optical microscopy, the sample was eliminated.



and for 10 minutes afterwards.This time allowed stabilization of any temperature fluctuations




step.Mean strain measurements for each step were calculated from the values registered

from the two samples of each group. The corresponding profiles of residual stress vs.

depth from the specimen surface were then calculated according to ASTM Standard

Test Method E837-08 using H-Drill software (Vishay, Malvern, PA, USA). For the rosette size

used, the hole-drilling method can measure residual stresses to depths to 1.2 mm.

Results3Residual Stress in Veneering Ceramic I Amélie Mainjot 108

All VZr samples with a 3 mm thick veneer layer developed spontaneous cracks in the

veneering ceramic after the grinding process or after the last firing procedure (see Fig.

6.1). Therefore measurements were not performed for this group.

Figs 6.2 and 6.3 show the residual stress vs.depth profiles calculated following the procedure

specified in ASTM E837-08,using the software H-Drill, respectively for VM and VZr samples,

for each veneer thickness.

All VM and VZr samples exhibited compressive stresses in the veneer surface. For VZr

samples, the magnitude of these stresses was larger than VM samples and increased

with veneer thickness (Fig. 6.3).

For VM samples, in surface compressive stresses were found to decrease with depth up

to 0.5-1mm from the surface, becoming slightly tensile for some samples, and then to

become more compressive again near the framework.

Fig. 6.1 - The Methylene blue infiltrated cracks in 3 mm thick veneering ceramic layered on zirconia frameworks .


Fig. 6.3 - Residual stress vs. depth profiles in the veneering ceramic layer of VZr, respectively for 1.0, 1.5, 2.0,and 2.5 mm veneer thickness

Fig. 6.2 - Residual stress vs.depth profiles in the veneeringceramic layer of VM samples,respectively for1.0, 1.5, 2.0, 2.5 and 3.0 mm veneer thickness.


For VZr samples veneered with a 2.5 mm,2 mm or 1 mm thick layer, in surface compressive

stresses were also found to decrease with depth, becoming more compressive again

from 0.5-0.7 mm from the surface. For samples veneered with a 1.5 mm thick layer,

compressive stresses in the veneer surface were found to become tensile

around 0.9 mm from the surface.

Variation of interior stresses at 1.2 mm from the surface in function of veneer thickness

was inverted for metal and zirconia samples, veneer thickness influencing in an opposite

way the residual stress profile in metal- and in zirconia-based bilayers (Fig. 6.4).

Fig. 6.4 - Comparison of measured interior stresses at 1.2 mm from the veneer surface (except for the 1mm-veneered sample: stresses at 1mm from surface) vs. veneer thickness,for VM and VZr samples.

6Influence of veneer thickness on residual stress I Amélie Mainjot

Discussion4As the cooling rate, the veneering ceramic thickness can influence thermal gradients

during the cooling process and then residual stress profile within the material, as was

observed in the measured stress profiles in both VM and VZr samples.

The typical stress profiles observed in VM samples (Fig. 6.2) illustrated the combined

effect of thermal gradients and CTE mismatch on residual stress and were in line with

measurements performed in previous studies, notably showing the influence of cooling

rate [4, 11]. In a first step, as the veneering ceramic cools from the surface to the interior,

the tempering effect induces a contraction mismatch causing the appearance of

compressive stresses in the surface of the veneering ceramic. The magnitude of the

measured stresses in the surface was not influenced by veneer thickness and was similar

(under 30 MPa) to stresses measured in a previous study in samples with a 0.7 mm thick

framework and a 1.5 mm thick veneer, tempered in the same way [4].

These tempering stresses decreased progressively with depth, and should have been

balanced by the appearance of interior tensile stresses (Fig. 6.5a). These tensile stresses

were not observed because when the interior of the veneer solidifies and framework

cools, the positive CTE mismatch between metal and ceramic (13.1-13.6 10-6.K-1 for VM13

dentine, and 14-14.3 10-6.K-1 for Duceralloy C) develops in-depth compressive stresses

within the ceramic. This is illustrated by the high reemerging compressive stresses in the

second part of the registered stress profile in the 1 mm-veneered sample (Figs 6.2 and

6.5b). As shown previously concerning cooling rate [4],veneer thickness influences thermal

gradient. When thickness increases, solidification occurs progressively from surface to

interior and the presence of a fusion zone in the underlayers allows viscorelaxation

effects.These are expected to reduce the magnitude of interior CTE mismatch induced

compressive stresses. In the present study, samples with thin veneering ceramic layer

exhibited favorable stress profile, thereby confirming clinical recommendations [9]. For

thicker samples (beyond 1.5 mm thick), the depth resolution of the method did not allow

the measurement of stresses near the framework, but data suggest advantageous stress

profiles, i.e. mostly in compression.

The nature of stresses observed near the framework contrast with results of Asaoka and

Tesk [5] who calculated tensile stresses in a tempered PFM beam. DeHoff et al [8] found

compressive stresses near the framework of metal-ceramic disks with thinner frameworks

111


(0.3 mm framework thickness, 2 mm veneer thickness), but they showed tensile stresses

in the surface.As reported by the authors, their calculations were limited since they did not

take into account realistic temperature distributions within the sample,values of CTE up to

600°C,and the effect of cooling rate on ceramic microstructure.The magnitude of calculated

stresses in both studies did not exceed 40 MPa,compared with 86 MPa in the present study.

The VZr samples also exhibited compressive stresses in the veneer surface.Their magnitude

was larger than with VM samples,and increased with veneer thickness (Fig.6.3).This could be

explained by the lower heat capacity of zirconia,which induces a higher thermal gradient

in the sample than a metal framework. Indeed, in previous experiments, it was shown that

the temperature mismatch between framework and veneering ceramic surfaces was

lower for VM than for VZr samples [4].As in VM samples, the typical stress profiles observed

in VZr samples veneered with a 2.5 mm,2 mm or 1 mm thick layer illustrated the combined

effect of thermal gradients and CTE mismatch on residual stress: in surface compressive

stresses were also found to decrease with depth,becoming more compressive again from

0.5-0.7 mm from the surface.Variation of interior stresses at 1.2 mm from the surface in function

of veneer thickness was inverted in comparison with metal samples (Fig. 6.4), compressive

stresses in the veneer surface having been found to become tensile around 0.9 mm in

depth for samples veneered with a 1.5 mm thick layer (Fig. 6.3). The inverted variation of

interior stresses with veneer thickness in zirconia samples was expected in the light of the

hypothesis formulated previously to explain the influence of cooling rate on residual stress

profiles in VZr and VM samples [4]. Indeed, due to the positive CTE mismatch between

zirconia and veneering ceramic (CTE = 8.8-9.2 10-6 K-1for VM9 dentine, and 10.5 10-6 K-1 for

Vita In-Ceram YZ,following manufacturers),and the similar elastic moduli of CoCr and zirconia,

and VM13 and VM 9 respectively, interior compressive stresses should have been found to

vary in VZr samples as in VM samples. This tendency cannot be explained by the effect of

the viscoelastic processes or by the effect of the differences between metal and zirconia

properties on residual stress, as it was shown previously studying zirconia and metal samples

slowly cooled in a way to reduce the effect of physical and thermal properties variations

[4]. Moreover it cannot be explained by the fact that the zirconia framework cools faster

than the veneering ceramic, as confirmed by the temperature profiles registered in the

same study.Then an hypothesis could be the possible crystalline transformation of zirconia


during the veneering process [4]. The theory describes residual stress development in zirconia

samples by a three-step approach (Fig. 6.5). The first step is the tempering effect in the

surface of the veneering ceramic and the second is the CTE mismatch effect in the depth,

as in VM samples (Figs 6.5a and 6.5b). The difference with VZr samples could be that and,

at a third time, an extra step occurs: the stress-induced transformation of zirconia, which

generates a volume increase of about 4% [14],and then decreases the interior compressive

stresses or converts them into tensile stresses (Fig. 6.5c). As discussed in the previous study

about the influence of cooling rate [4], this stress-induced or martensitic transformation

could be enhanced by the stress developed at the interface during the second step due to

the CTE mismatch effect. Indeed, the in-depth compressive stresses in the veneering ceramic

are balanced by tensile stresses in the framework, which can trigger the transformation.

The stress-induced volume expansion is instantaneous and can secondarily generate residual

tensile stresses in place of compressive stresses in the veneering ceramic.The amount of stress

needed to generate the transformation in the absence of water is poorly understood.Li et al

[15] calculated that 415 MPa can induce the transformation in a bending test configuration.

In the same study they showed that tensile stress accelerates the degradation of Y-TZP in

hot water (100°C). They observed that a tensile stress of 100 MPa, in a bending configuration

in hot water during 5 hours,generated a monoclinic volume of 5%. It can be hypothized that

the combination of water and temperature during the heating process, and particularly

around 250°C when the transformation is more rapid [16], weakens the zirconia surface,

generating monoclinic grains [17],decreasing the stability of the tetragonal phase because of

the infiltration of water through grain boundaries [15], leading to microcracks [16]. The

microcracks, which can also be originated from the grinding process of frameworks after

sintering, can not heal with temperature and may secondarily, during the cooling process,

promote the stress-induced transformation. The crystalline transformation during the

veneering process was described by Tholey et al,who observed monoclinic crystals at the

veneer-zirconia interface with SEM and XRD2 micro-diffraction [14] [17]. Tholey et al linked

this transformation to moisture content used for the build-up of the veneering ceramic,

which can degrade Y-TZP during its heating. In any case, if some monoclinic grains would

appear during heating time and would not heal during the firing at 900°C, they would not

induce residual stress because the expanded grains would be present before the cooling

process.

113


The influence of veneer thickness on the variation of stress profiles in zirconia samples support

the hypothesis of the transformation.As explained for VM samples, when veneer thickness

decreases, the interior compressive stresses due to the CTE mismatch increase, thereby

increasing the potential of zirconia transformation in the surface of the framework. The tensile

stresses generated in the veneering ceramic due to this transformation lower the magnitude

of the interior compressive stresses and can convert them into tensile stresses when the

veneer thickness reaches 1.5 mm. In VZr samples 1mm-veneered, a reappearance of in-

depth compressive stresses was observed.This can be due to an antagonist phenomenon :

if the veneer thickness decreases, the framework is less stressed by the contraction of the

veneer during the solidification process. In the same way, stress profile can also be

influenced by framework thickness, as shown in the study with 0.7 mm thick framework

samples where exclusively interior tensile stresses were observed [4]. Indeed the ratio

between veneer and framework thickness influences the amount of stress applied to the

framework and the risk of transformation.

Spontaneous cracks observed in 3 mm-veneered VZr samples (see Fig. 6.1) can be

explained by the bending of the framework due to the contraction of a large amount of

veneering ceramic, this bending generating tensile stresses perpendicularly to the frame-

work. Indeed if extrapolating the stress profile curve beyond 1.2 mm depth in 2.5 mm-

veneered samples, very high interior compressive stresses can be expected for 3 mm

veneered-samples, which may subject the bilayer to flexural strengths.

Results of the present study contrast with those obtained with the mathematical model of

Swain [7] who studied the influence of veneer thickness on residual stress in zirconia bilayers.

With his 2D bilayer plate model, he showed in-depth tensile stresses in a 3 mm-veneered

zirconia sample with a 1 mm thick framework, and in-depth compressive stresses in a 0.5

mm-veneered zirconia sample with a 0.5 mm thick framework. The large misfits between

measurements and calculations are linked to the difficulty encountered to mimic the firing

process with models. In this case calculations did not take into account thermal gradients,

variations of thermo-physical properties of ceramic in the glass transition range, and finally

the possible transformation of zirconia. Even if the hole-drilling method is limited to residual


stress profile measurement to depth to 1.2 mm, it gives an accurate and repeatable

tendency based on physical measurements for each veneer thickness and each material

studied, allowing the understanding of these two parameter’s influence.

115

Fig. 6.5 - The chronological three-step approach to explain residual stress developmentin the veneering ceramic:

a - The temperingeffect, whichexplains the development of compressive stresses in the surface of all samples.

b - The CTE mismatch effect, which explains the development of interior compressive stresses near the framework in all samples

c - The stress-induced crystalline transformation effect, which is an hypothesis to explain the inverted variation of measured interior stresses with veneer thickness in zirconia-based samples, in comparison with metal-based samples. The crystalline transformation generates a local volume increase of the surface of the framework counteracting the CTE mismatch effect.



Conclusions5Veneer thickness influences in opposite ways the residual stress profile in metaland in zirconia-

based structures.A chronological two-step approach is discussed to explain residual stress

development in metal-based samples, and a three-step approach, comprising and

additional step, is proposed for zirconia-based samples.

These three steps are respectively :

The tempering effect,which explains the development of compressive stresses in the surface of

both samples,these stresses decreasing with depth.This effect is influenced by thermal gradients.

The CTE mismatch effect, which explains the development of interior compressive stresses

near the framework in both samples. The resulting stress profile measured in metal-based

samples describe a typical curve, starting with compressive at the ceramic surface,

decreasing with depth to 0.5-1.0 mm from the surface, and then becoming compressive

again. Veneer thickness decreases the CTE mismatch effect, increases visco-relaxation,

but does not influence the tempering effect in the surface of VM samples.

The stress-induced crystalline transformation effect, which is an hypothesis to explain the

inverted variation of measured interior stresses with veneer thickness in zirconia-based

samples, in comparison with metal-based samples.

Zirconia-based samples show a gradual shift to tensile residual stresses at 0.9 mm depth

from the surface,when the veneer was 1.5 mm thick.This could be a potential explanation

to delaminations encountered in clinical practice.

The residual stress profiles measured in the present study are in line with the observations

made previously when measuring the influence of cooling rate.Further studies are needed

to confirm the hypothesis of crystalline transformation of zirconia by microscopical observation

methods or by structural analysis, and to consider the influence of other influencing

parameters on the stress profile, such as the framework thickness, or the ratio between the

framework and the veneer, which could influence the risk of transformation.

Finally, this study highlights the difficulty encountered to adapt veneer thickness on zirconia.

If metal-based structures always exhibit a favorable residual stress profile, whatever the

veneer thickness, measured stress profiles in zirconia-based structures suggest to thicken

the veneer,which is in contradiction with the need for support of the ceramic.The adequate

ratio between veneer and framework is difficult to define and point out the lack of tolerance

of zirconia in comparison with metal.

1.

2.

3.


Schley JS, Heussen N, Reich S, Fischer J, Haselhuhn K, Wolfart S. Survival probability of

zirconia-based fixed dental prostheses up to 5 yr: a systematic review of the literature.

Eur J Oral Sci. 2010 Oct;118(5):443-50.

Heintze SD,Rousson V.Survival of zirconia- and metal-supported fixed dental prostheses:

a systematic review. Int J Prosthodont. 2010 Nov-Dec;23(6):493-502.

Gostemeyer G,Jendras M,Dittmer MP,Bach FW,Stiesch M,Kohorst P. Influence of cooling

rate on zirconia/veneer interfacial adhesion. Acta Biomater. 2010 Dec;6(12):4532-8.

Mainjot AK, Schajer GS, Vanheusden AJ, Sadoun MJ. Influence of cooling rate on

residual stress profile in veneering ceramic: measurement by hole-drilling.Dent Mater.

2011 May;27(9):906-14.

Asaoka K, Tesk JA. Transient and residual stress in a porcelain-metal strip. J Dent Res.

1990 Feb;69(2):463-9.

Taskonak B,Borges GA,Mecholsky JJ, Jr.,Anusavice KJ,Moore BK,Yan J.The effects of



Swain MV.Unstable cracking (chipping) of veneering porcelain on all-ceramic dental

crowns and fixed partial dentures. Acta Biomater. 2009 Jun;5(5):1668-77.

DeHoff PH, Anusavice KJ,Vontivillu SB. Analysis of tempering stresses in metal-ceramic

disks. J Dent Res. 1996 Feb;75(2):743-51.

Zarone F, Russo S, Sorrentino R. From porcelain-fused-to-metal to zirconia: Clinical

and experimental considerations. Dent Mater. 2011 Jan;27(1): 83-96.

Bonfante EA, Rafferty B, Zavanelli RA, Silva NR, Rekow ED, Thompson VP, et al.

Thermal/mechanical simulation and laboratory fatigue testing of an alternative yttria

tetragonal zirconia polycrystal core-veneer all-ceramic layered crown design. Eur J

Oral Sci. 2010 Apr;118(2):202-9.

Mainjot AK, Schajer GS,Vanheusden AJ, Sadoun MJ. Residual stress measurement in

veneering ceramic by hole-drilling. Dent Mater. 2011 May;27(5):439-44.

References

1

2

3

4

5

6

7

8

9

10

11

ASTM.Determining residual stresses by the hole-drilling strain-gage method.Standard

Test Method E837-08 West Conshohocken, PA, USA: American Society for Testing and

Materials; 2008.

Schajer GS,Prime MB.Use of inverse solutions for residual stress measurements.Journal

of Engineering Materials and Technology,Transactions of the ASME.2006;128(3):375-82.

Tholey MJ, Swain MV, Thiel N. SEM observations of porcelain Y-TZP interface. Dent

Mater. 2009 Jul;25(7):857-62.

Li JG, Zhang LM, Shen Q, Hashida T. Degradation of yttria stabilized zirconia at 370 K

under a low applied stress. Materials Science and Engineering a-Structural Materials

Properties Microstructure and Processing. 2001 Jan;297(1-2):26-30.

Chevalier J, Gremillard L, Virkar AV, Clarke DR. The Tetragonal-Monoclinic

Transformation in Zirconia: Lessons Learned and Future Trends. J Am Ceram Soc.2009

Sep;92(9):1901-20.

Tholey MJ, Berthold C, Swain MV, Thiel N. XRD2 micro-diffraction analysis of the

interface between Y-TZP and veneering porcelain: role of application methods.Dent

Mater. 2010 Jun;26(6):545-52.

12

13

14

15

16

17


7

a Department of Fixed Prosthodontics, Institute of Dentistry, University ofLiège Hospital (ULg CHU), 45 Quai G. Kurth , Liège, 4020, Belgium.

b Department of Mechanical Engineering, University of British Columbia,Vancouver, Canada.

c Unité de Recherches Biomatériaux Innovants et Interfaces (URB2I-EA4462),Université Paris Descartes, Sorbonne Paris Cité, Faculté deChirurgie Dentaire, Paris, France.

Submitted to Dental Materials



7Influence of framework thickness on residual stress I Amélie Mainjot 121

Abstract

Objectives: Framework design is reported to influence chipping in zirconia-based restorations,

which is an important cause of failure of such restorations. Residual stress profile in the

veneering ceramic after the manufacturing process is an important predictive factor of

the mechanical behavior of the material. The objective of this study is to investigate the

influence of framework thickness on the stress profile measured in zirconia-based structures.

Methods: The stress profile was measured with the hole-drilling method in bilayered disc

samples of 20 mm diameter with a 1.5 mm thick veneering ceramic layer. Six different

framework thicknesses from 0.5 mm to 3 mm were studied.Two different cooling procedures

were also investigated.

Results: Compressive stresses were observed in the surface,and tensile stresses in the depth

of most of the samples.The slow cooling procedure was found to promote the development

of interior tensile stresses, except for the sample with a 3 mm thick framework. With the

tempering procedure,samples with a 1.5 mm thick framework exhibited the most favorable

stress profile,while thicker and thinner frameworks exhibited respectively in surface or interior

tensile stresses.

Significance: The measurements performed highlight the importance of framework thickness,

which determine the nature of stresses and can explain clinical failures encountered,

especially with thin frameworks. The adequate ratio between veneering ceramic and

zirconia is hard to define, restricting the range of indications of zirconia-based restorations

until a better understanding of such a delicate veneering process is achieved.

Keywords: Residual stress; Hole-Drilling; Zirconia; Chipping; Veneering ceramic; Framework;

Thickness

Introduction17 Residual Stress in Veneering Ceramic I Amélie Mainjot 122

Introduced in prosthodontics over ten years ago,Yttria-tetragonal-zirconia polycrystal (Y-TZP)

is reported as a biocompatible and esthetic alternative to metal frameworks. Its good

mechanical properties in comparison with other ceramic materials has led manufacturers

to propose Y-TZP for large bridges on teeth and implants. Unfortunately clinical studies

report short-term clinical failures of zirconia-based restorations mainly due to cohesive

fractures of the veneering ceramic (chipping),which is the weak link of the restoration [1].

This problem is less reported with porcelain-fused to metal (PFM) restorations and has often

been associated to improper framework design [2]. Indeed,the results of some uncontrolled

clinical studies [3] [4] [5] and in vitro studies about crowns fatigue and fracture load resistance

[6] [7] [8] [9] suggest that the use of an alternative core design, which ensures an optimal

support to veneering ceramic, can prevent chipping. Anatomical frameworks, avoiding

excessive veneer thickness and exhibiting an additional lingual shoulder, as proposed for

porcelain-fused-to-metal (PFM) restorations, are then recommended. Moreover the core-

veneer thickness ratio is pointed out as an important influencing factor in terms of failures

[10].But the mechanism of chipping is complex,multifactorial and still not well understood.

A first step to the comprehension of the framework thickness influence in the chipping

problem consists in studying residual stress.Residual stresses are "locked-in" stresses that are

generated within the veneer and the framework during the cooling/solidification period

of the firing process. These stresses are present within the structure without the application

of any external load, but will add to functional loads and then constitute an important

predicting factor for the mechanical behavior of restorations. Indeed,compressive residual

stresses reinforce the ceramic while tensile residual stresses facilitate the initiation and the

propagation of cracks.Therefore the knowledge of the residual stress distribution within the

veneering ceramic as a function of depth, i.e. stress profile, is a key factor for understanding

and predicting chipping and delaminations. The stress profile in the veneering ceramic is

generated by the chronological effects of the thermal gradients occurring during the

cooling/solidification period of the veneer liquid phase sintering process,and the mismatch

in thermal expansion properties between core and veneering ceramic [11] [12]. Thermal

gradients are determined by cooling rate and by material thickness and conductivity.They

induce non-uniform solidification,from the surface to the center,thereby causing contraction

mismatch within the ceramic.Due to the tempering effect, residual stresses are compressive

in the surface of the veneering ceramic, the magnitude of these stresses decreasing with


depth and being influenced by the cooling rate around the glass transition temperature Tg

[12] [13] [14]. Moreover, in the case of ceramic fused to metal frameworks, the thermal

expansion coefficient (CTE) of the ceramic is generally slightly lower than the framework so

that during cooling from Tg to room temperature, interior compressive stresses are developed

within the ceramic near the framework.Swain calculated the independent influence of the

cooling rate, thickness and thermal expansion coefficient on residual stresses profile with a

2D bilayer mathematical model [15].Among the three factors studied,thickness predominated

as the most influential parameter. Some other authors [9] [16] have used 3D-finite element

analysis to study the influence of core design and components thickness on thermal residual

stress and loading stress generated within a crown.Within the limitations of the mathematical

models employed,the results highlighted the importance of cement more than core thickness,

and did not support totally the alternative core design theory. Actually the cited analysis

did not account for thermal gradients and for the viscoelastic behavior of the ceramic in

the glass transition range,where thermo-physical properties are submitted to variations and

influence residual stresses. These parameters are challenging to simulate with models.

Recently a new method was introduced to measure the residual stress profile in the

veneering ceramic [17] [18].This method is based on the removal of some stressed material

and the measurement of the resulting deformations in the adjacent material [19]. The

deformations are measured on the surface, typically using strain gages, from which the

residual stresses can be calculated.Stresses are calculated from surface to depth,typically

with 0.1 mm steps, and giving a stress profile within a 1.2 mm depth.This method was used

to study the influence of cooling rate on residual stress profile in veneered metal and

zirconia disk samples [12]. In this study it was observed that zirconia samples with a 0.7 mm

thick framework and a 1.5 mm thick veneer layer exhibited in-depth tensile stresses in the

veneering ceramic,contrary to metal samples,which exhibited only compressive stresses.The

hypothesis of the crystalline transformation of zirconia was proposed to explain these results.

The objectives of this study were to investigate the framework thickness dependence of

stress profiles in veneered Y-TZP disks using the hole-drilling method and to understand how

the framework thickness of zirconia-based-restorations can influence their mechanical

behavior. Moreover the influence of cooling rate on stress profiles was also studied.

Materials and methods27 Residual Stress in Veneering Ceramic I Amélie Mainjot 124


Bilayered disc samples composed of veneering ceramic sintered on Y-TZP framework (VZr,

12 samples) were manufactured following standard dental laboratory procedures and

manufacturer’s recommendations. Y-TZP core discs were cut out of a pre-sintered Vita

In-Ceram YZ blocks (Vita Zahnfabrik,Bad Säckingen,Germany),were rounded by polishing,

and densely sintered at 1530°C for 120 min with heating rate 10°C/min, and heating time

149 min (Zircomat furnace, Vita Zahnfabrik, Bad Säckingen, Germany). The sintered Y-TZP

discs,20 mm diameter,were cast and sequentially ground with 180-grit and 500-grit silicon

carbide discs (Struers LabPol polishing machine, Copenhagen, Denmark) either to a 0.50

mm (n=2), a 0.70 mm (n=2), a 1.00 mm (n=2), a 1.50 mm (n=2), a 2.00 mm (n=2) or a 3.00

mm (n=2) ± 0.02 mm thickness. The Y-TZP discs were exposed to a “regeneration firing”,

which is a final thermal treatment of the core to reverse any phase transitions in the zirconia

due to the grinding procedures. A thin coat of Vita VM 9 Effect Bonder was applied and

fired on the surface to be veneered. Then, Vita VM9 feldspar veneering ceramic (shade

3M2) (Vita Zahnfabrik, Bad Säckingen, Germany) was progressively layered on the effect

bonder. A Vita Vacumat 4000 Premium furnace (Vita Zahnfabrik, Bad Säckingen,

Germany) was used for all firing procedures, as summarized in Tab. 7.1.

Starting Pre-drying t Heating Heating t Firing T Holding t Vacuum Slow cooling Slow coolingT (°C) (min) rate (min) (°C) (min) holding t ending T(°C) rate

closing t (°C/min) (min) (°C/min)


Vita VM9 500 6 75 6 950 1 6Effect bonder

Vita VM9 Dentine 500 6 55 7.27 910 4 7.27

Last firing cycleClassic Cooling (CC) 600 8 50 6 900 6 6

Slow Cooling (SC) Room T° 10 900 6 Room T° 2

Table. 7.1 - Firing schedules


All samples were baked on the same ceramic mesh-tray. Three layers of dentin ceramic

were successively fired. Samples were sequentially ground with 180-grit and 500-grit silicon

carbide discs to a veneer thickness of 1.50 ± 0.02 mm.

After final polishing, all specimens were exposed one by one to a last firing cycle. This last

firing cycle restored the residual stress profile through the veneering ceramic thickness. All

samples were placed in the same position,on the center of the mesh-tray and of the furnace.

Two different cooling schedules were followed.One sample of each group was tempered

from 900°C to room temperature by opening the furnace door,as classically done in dental

laboratories,and removed from the mesh-tray at 200°C (Classic Cooling,CC).This schedule

was the one used during the veneering layering process. In comparison with manufacturer

recommendations, the firing temperature was maintained six minutes in place of one

minute in order to reach 900°C within the framework. The second sample was cooled at

2°C/min in a special furnace (Carbolite LMF 12/2, Carbolite, Hope Valley, UK), from 900°C

to room temperature (Slow Cooling, SC). All firing schedules are summarized in Tab 7.1.

2.2 Hole-Drilling method


A specialized six-element Type C rosette (N2K-06-030RR-350/DP, Vishay, Malvern, PA, USA)

was installed on the center of the veneering ceramic surface.To promote the strain gage

bond, the ceramic surface was prepared by etching with 10% hydrofluoric acid for 1 min,

and was then cleaned for 5 min in an ultrasonic bath containing 90% alcohol. The strain

gage rosette was installed with M-Bond 200 Adhesive (Vishay, Malvern, PA, USA), following

the manufacturer’s instructions. The adhesive was allowed to cure overnight to ensure

complete curing. The installation was monitored using an optical microscope.


The strains expected from the strain gages are very small and cannot be measured with

sufficient accuracy using conventional industrial equipment. A specialized data acquisition

system was therefore built where each strain gage was connected in a Wheatstone

bridge circuit with 3 control gages (identical gages attached to an undisturbed sample).All


gages and control rosettes were exposed to identical constant temperature conditions.

Finally, the very low voltage measurements were performed with specific custom-built

electronic equipment comprising a precision DC and AC current source 6221 (Keithley

Instruments, Inc, Cleveland, Ohio, USA) and 3 Nanovoltmeters 2182A (Keithley Instruments,

Inc, Cleveland, Ohio, USA). Filtered measurements were recorded on a computer using NI

LabView software (National Instruments, Austin, Texas, USA).



drilling machine,the container was filled with silicon oil to enhance drilling lubrication,thermal

conductivity and electrical insulation.In addition,the silicon oil bath was thermally controlled

and maintained at 36°C ± 0.1°C with a Eurotherm 3208 system (Eurotherm Ltd, Worthing,

UK) to avoid the effects of any ambient temperature variations.Temperature at the sample

contact was recorded with a thermocouple connected to NI LabView data acquisition

system.

2.2.4 Hole-drilling

An Isel CAD-CAM machine (CPM 3020,Houdan,France) was used for the drilling procedure.

To increase strain sensitivity, the maximum allowable hole diameter for the strain gage

rosette type was made using a 2.5 mm diameter cylindrical bur (Asahi Diamond Industrial

Europe SAS,Chartres,France).The bur rotation speed was 19000 rpm.A hole was cut at the

center of the rosette in steps of 0.1 ± 0.01 mm, as measured by a Digimatic indicator

(Mitutoyo Corporation,Kawazaki,Japan).Hole diameter and concentricity were checked

after the experiment with an optical microscope and motorized micrometer, Micro

Control CV 78 (Newport, Irvine, CA, USA). The protocol of the hole-drilling method was

designed to eliminate/minimize crack initiation through choice of drilling process,drill type

and lubricant used. However, in the few cases where cracks had nevertheless occurred,

abnormal large variations of the measured strains were induced. If these were confirmed

by optical microscopy, the sample was eliminated.




and for 10 minutes afterward.This time allowed stabilization of any temperature fluctuations

caused by the drilling process.The strain measurements were recorded in an Excel spread-

sheet (Microsoft Corporation, Redmond,WA, USA). Mean values were evaluated for each

strain gage based of the final 200 values (1 Hz acquisition) registered for each step. The

corresponding profiles of residual stress vs. depth from the specimen surface were then

calculated according to ASTM Standard Test Method E837-08 using H-Drill software

(Vishay, Malvern, PA, USA). For the rosette size used, the hole-drilling method can measure

residual stresses to depths to 1.2 mm.

Results37 Residual Stress in Veneering Ceramic I Amélie Mainjot 128

The VZr sample with a 0.5 mm thick framework performed for the SC procedure developed

spontaneous cracks in the veneering ceramic after the Vita VM9 dentin firing. A second

one was manufactured but cracks appeared after the SC procedure. Therefore

measurements were not performed for this group.

Figs 7.1 and 7.2 show the residual stress vs.depth profiles calculated following the procedure

specified in ASTM E837-08, using the software H-Drill, respectively for samples submitted to

the SC and to the NC procedure, for each framework thickness.

For the NC procedure, the surface stresses in the veneering ceramic were found to be

compressive in all samples except in thicker frameworks samples (2 mm and 3 mm), in

which in surface tensile stresses were observed. Interior stresses were found to vary from

Fig. 7.1 - Residual stress vs. depth profiles in the veneering ceramic layer of VZr samples, for the NCprocedure, respectively for 0.5, 0.7, 1.0, 1.5, 2.0 and 3.0 mm framework thickness.


compressive to tensile with decrease in framework thickness, except for the 3mm thick

framework sample, which exhibited interior tensile stresses. Interior tensile stresses were

observed for samples with frameworks thinner than 1.5 mm.

For the SC procedure, compressive stresses were measured in the surface of all samples.

Interior stresses were also found to vary with framework thickness. Tensile stresses were

observed for all samples,except for the 3 mm thick framework sample,and their magnitude

increased inversely to framework thickness.Compared to the NC procedure, interior stresses

showed a tendency to be more tensile, except for the 3mm thick framework sample, for

which the interior stresses switched from tensile to compressive when the sample was

cooled slowly.

Fig. 7.2 - Residual stress vs. depth profiles in the veneering ceramic layer of VZr samples, for the SCprocedure, respectively for 0.7, 1.0, 1.5, 2.0 and 3.0 mm frameworkthickness.

Discussion47 Residual Stress in Veneering Ceramic I Amélie Mainjot 130

The results of the present study showed that the framework thickness and the cooling rate

influence residual stress profile in the veneering ceramic layered on a zirconia framework.

The stress profiles registered in samples submitted to a NC procedure (Fig. 7. 1), were

expected in regards to the tempering and the CTE mismatch effects, to the viscoelastic

properties of the veneering ceramic [14],and to the hypothesis of the crystalline transformation

of zirconia during the veneering process, formulated in a previous study about the influence

of the cooling rate [12]. Actually, the chronological three-step approach of the residual

stress development, discussed in this study, explains the present measurements.

These three steps are:

The tempering effect, associated to the visco-elastic properties of the veneering ceramic,

which explains the development of compressive stresses in the surface, these stresses

decreasing with depth. Indeed,the vertical thermal gradient existing between the surface

of the veneering ceramic and the framework induce a contraction mismatch during the

solidification process, the solidification occurring progressively from the surface to the interior.

The temperature mismatch between the surface of the veneering ceramic and the surface

of the framework can exceed 200°C in samples with a 0.7 mm framework,as measured in

the study about the influence of cooling rate [12]. In the present study tensile stresses were

measured in the surface of the samples with a 2 mm or a 3 mm thick framework. It can be

explained by the presence of a high horizontal thermal gradient.Actually, as the framework

keeps the center of the veneering ceramic hotter,a supplementary solidification contraction

mismatch is generated from the periphery to the center of the veneering ceramic,putting

the center in tension (Fig. 7.3). This phenomenon is accentuated if the framework is thick.

The temperature distribution in a sample is complex, depending on materials thickness,

cooling rate,materials thermal properties, the presence of a mesh-tray as a support for the

firing procedure, and on temperature distribution within the furnace. The large number of

these factors complicates residual stress calculations through mathematical models.

The CTE mismatch effect induces interior compressive stresses near the framework, as

observed previously in samples with a metal framework [12].

The counteracting of the CTE mismatch effect by stress-induced crystalline transformation

of zirconia. The compressive stresses developed due to the positive CTE mismatch are

compensated by tensile stresses in the framework, which could trigger the transformation

1.

2.

3.


at the interface, inducing a volume expansion of the zirconia.This volume expansion could

secondarily develop tensile stresses in the veneering ceramic, counteracting by this way

the CTE mismatch induced stresses. In the present study this hypothesis can explain the

presence of the interior tensile stresses measured from 0.5 to 1 mm from the surface,

instead of the compressive stresses expected due to the CTE mismatch effect.This hypothesis

also explains their variation, which is inverted to framework thickness, except for the sample

with a 3 mm thick framework. Indeed,the amount of stress applied to the framework varies

with the ratio between veneer and framework thickness,and can potentially influence the

risk of transformation, explaining that the samples with a thicker framework ( 1.5 mm and

2 mm) exhibited interior compressive stresses.The presence of interior tensile stresses in the

sample with a 3 mm thick framework can be explained by the effect of the high horizontal

gradient,which can also counteract the CTE mismatch effect,and not by the the effect of

the transformation.The possible occurrence of phase transformation of zirconia during the

veneering process is corroborated by the observations of Tholey et al.,who showed monoclinic

grains at the zirconia-veneer interface [20, 21].

Fig. 7.3 - Schematicrepresentationof the vertical temperature gradient, in comparison with the horizontaltemperature gradient,which is accentuated in samples with a thicker framework.


The SC procedure was performed to minimize the effects of thermal gradients, and then to

confirm the previous analysis about the residual stress development. Indeed, with the SC

procedure, the samples with a 2 mm and a 3 mm thick framework exhibited compressive

stresses in the surface instead of tensile stresses, and the interior stresses in the sample with a

3 mm thick framework switched from tensile to compressive. In the depth, stresses varied

inversely to framework thickness and showed a tendency to be more tensile than with the

NC procedure. Indeed the CTE mismatch is almost effective when the two layers cool at the

same rate: the SC procedure is supposed to induce more stresses at the interface and then

more risk of transformation,and more tensile stresses in the veneering ceramic.The magnitude

of interior tensile stresses in samples with thinner frameworks explains the appearance of

cracks during the manufacturing of samples with a 0.5 mm thick framework, particularly with

the SC procedure.

The results of the present study highlight the importance of the influence of framework and

veneer thickness,and ratio,on residual stress,as pointed out by the calculations of Swain with his

2D analysis of samples with different thicknesses [15]. However further comparison between

results obtained in the two studies is difficult because of the difficulty encountered by models

when mimicking such a complex process, particularly in regards to thermal gradients and

crystalline transformation potential. Finally, in the present configuration, the slow cooling rate

was shown to aggravate the negative effect of framework thickness.


Conclusions5The results of the present study concur with those obtained in previous studies using the same

method, and support the hypothesis formulated to explain the residual stress development,

especially about the phase transformation of zirconia during the veneering process. This

hypothesis needs to be confirmed by observation and structural analysis at the interface.

The measurements performed highlight the importance of framework thickness and veneer-

framework thickness ratio, both of which determine the nature of interior stresses and can

potentially explain clinical failures encountered with thin frameworks.Actually, in the configuration

of disk samples made using the NC procedure, frameworks thinner than 1.5 mm exhibited

unfavorable stress profiles. The SC procedure was found to have negative effects since it

promoted the development of interior tensile stresses.From a clinical point of view,such a 1.5

mm thick framework cannot be used for a conventional crown in the anterior region,due to

the optical properties of zirconia and the need of space for the veneering ceramic.However

it could be indicated for posterior restorations or screwed restorations on implants.

The framework thickness was also shown to influence surface stresses. Frameworks thicker

than 1.5 mm exhibited tensile stresses, which were attributed to the influence of thermal

gradients occurring during the NC procedure. The stress profile could also be varied by

increasing the veneer thickness, and then by allowing visco-relaxative effects, but it would

be in contradiction with the need of support of the ceramic.Therefore the maximum framework

thickness is difficult to establish, particularly if taking into account the complex geometry of

dental frameworks and its impact on the thermal gradients. Further studies are needed to

analyze the influence of the design of the restorations.

In conclusion,the adequate ratio between veneering ceramic and zirconia,and the adequate

thickness for both materials,are hard to define, restricting the range of indications of zirconia-

based restorations until a better understanding of such a delicate veneering process.

7 Residual Stress in Veneering Ceramic I Amélie Mainjot134

Schley JS, Heussen N, Reich S, Fischer J, Haselhuhn K,Wolfart S. Survival probability of

zirconia-based fixed dental prostheses up to 5 yr: a systematic review of the literature.

Eur J Oral Sci. 2010 Oct;118(5):443-50.

Heintze SD, Rousson V. Survival of zirconia- and metal-supported fixed dental prostheses:

a systematic review. Int J Prosthodont. 2010 Nov-Dec;23(6):493-502.

Molin MK, Karlsson SL. Five-year clinical prospective evaluation of zirconia-based

Denzir 3-unit FPDs. Int J Prosthodont. 2008 May-Jun;21(3):223-7.

Marchack BW, Futatsuki Y, Marchack CB, White SN. Customization of milled zirconia

copings for all-ceramic crowns: a clinical report. J Prosthet Dent.2008 Mar;99(3):169-73.

Tinschert J, Schulze KA, Natt G, Latzke P, Heussen N, Spiekermann H. Clinical behavior

of zirconia-based fixed partial dentures made of DC-Zirkon: 3-year results. Int J

Prosthodont. 2008 May-Jun;21(3):217-22.

Larsson C, Madhoun SE, Wennerberg A,Vult von Steyern P. Fracture strength of yttria-

stabilized tetragonal zirconia polycrystals crowns with different design: an in vitro

study. Clin Oral Implants Res. 2011 Jun 2.

Kokubo Y,Tsumita M, Kano T, Fukushima S.The influence of zirconia coping designs on

the fracture load of all-ceramic molar crowns. Dent Mater J. 2011 Jun 2;30(3):281-5.

Silva NR,Bonfante EA,Rafferty BT,Zavanelli RA,Rekow ED,Thompson VP,et al.Modified

Y-TZP Core Design Improves All-ceramic Crown Reliability. J Dent Res. 2011

Jan;90(1):104-8.

Bonfante EA, Rafferty B, Zavanelli RA, Silva NR, Rekow ED, Thompson VP, et al.

Thermal/mechanical simulation and laboratory fatigue testing of an alternative yttria

tetragonal zirconia polycrystal core-veneer all-ceramic layered crown design. Eur J

Oral Sci. 2010 Apr;118(2):202-9.

Zarone F, Russo S, Sorrentino R. From porcelain-fused-to-metal to zirconia:

Clinical and experimental considerations. Dent Mater. 2011 Jan;27(1): 83-96.

Gostemeyer G,Jendras M,Dittmer MP,Bach FW,Stiesch M,Kohorst P. Influence of cooling

rate on zirconia/veneer interfacial adhesion. Acta Biomater. 2010 Dec;6(12):4532-8.

References

1

2

3

4

5

6

7

8

9

10

11


Mainjot AK, Schajer GS, Vanheusden AJ, Sadoun MJ. Influence of cooling rate on

residual stress profile in veneering ceramic: measurement by hole-drilling.Dent Mater.

2011 May;27(9):906-14.

Asaoka K, Tesk JA. Transient and residual stress in a porcelain-metal strip. J Dent Res.

1990 Feb;69(2):463-9.

Taskonak B,Borges GA,Mecholsky JJ, Jr.,Anusavice KJ,Moore BK,Yan J.The effects of



Swain MV.Unstable cracking (chipping) of veneering porcelain on all-ceramic dental

crowns and fixed partial dentures. Acta Biomater. 2009 Jun;5(5):1668-77.

Rafferty BT,Janal MN,Zavanelli RA,Silva NR,Rekow ED,Thompson VP,et al.Design features

of a three-dimensional molar crown and related maximum principal stress. A finite

element model study. Dent Mater. 2010 Feb;26(2):156-63.

Mainjot AK, Schajer GS,Vanheusden AJ, Sadoun MJ. Residual stress measurement in

veneering ceramic by hole-drilling. Dent Mater. 2011 May;27(5):439-44.

ASTM.Determining residual stresses by the hole-drilling strain-gage method.Standard

Test Method E837-08 West Conshohocken, PA, USA: American Society for Testing and

Materials; 2008.

Schajer GS,Prime MB.Use of inverse solutions for residual stress measurements.Journal

of Engineering Materials and Technology,Transactions of the ASME.2006;128(3):375-82.

Tholey MJ, Berthold C, Swain MV,Thiel N. XRD2 micro-diffraction analysis of the interface

between Y-TZP and veneering porcelain: role of application methods. Dent Mater.

2010 Jun;26(6):545-52.

Tholey MJ, Swain MV, Thiel N. SEM observations of porcelain Y-TZP interface. Dent

Mater. 2009 Jul;25(7):857-62.

12

13

14

15

16

17

18

19

20

21

8Residual Stress in Veneering Ceramic I Amélie Mainjot 136

8Conclusions and perspectives I Amélie Mainjot 137

1

-

-

General conclusion

The hole-drilling method was successfully transferred and adapted to dental use,particularly

to measure residual stresses in veneer-metal (VM) and veneer-zirconia (VZr) disks. Even if

time-consuming, the procedure enables the measurement of very low voltages and then

the calculation of very low stresses, in the range of a few MPa. A total of 82 experiments

were performed. Independently of the studied parameter, VM samples always exhibited

the same type of stress vs. depth profile profile, starting with compressive stresses at the

ceramic surface, decreasing with depth, becoming sometimes slightly tensile at 0.5-1.0

mm from the surface, and then becoming compressive again. The maximum magnitude

of in-surface compressive stresses was -40 MPa, and the maximum magnitude of interior

compressive stresses was -90 MPa. VZr samples showed varying stress profiles depending

on the cooling rate and the veneer-framework ratio. In-surface stresses were always

compressive, except for tempered samples with a framework thickness of 2 mm or 3 mm,

and in the same range of magnitude than VM samples. However, if some types of stress

profiles were similar to VM samples, other described a worse curve, where the in-surface

compressive stresses were found to turn into tensile stresses in the interior, with a maximum

magnitude of 40 MPa. The VZr samples with the following combination of framework

thickness (mm)+veneer thickness (mm):

Normal Cooling 0.5+1.5 0.7+1.5 1+1.5 3+1.5

Modified Cooling 0.7+1.5

Slow Cooling 0.7+1.5 1+1.5 1.5+1.5 2+1.5

exhibited interior tensile stresses.

Normal Cooling 1+1 1+2 1+2.5 1.5+1.5 2+1.5

Slow Cooling 3+1.5

exhibited interior compressive stresses.

One of the striking points is the inverted variation of interior stresses in VM and VZr samples,

with cooling rate and veneer thickness.

Measurements contribute to the explanation of residual stress development origins in VM

and VZr bilayers.A chronological two-step approach is discussed to explain residual stress

development in metal-based samples,and a three-step approach,comprised of the two-step

approach and an additional step, is proposed for zirconia-based samples.


These three steps are respectively :

The tempering effect,which explains the development of in-surface compressive stresses.This

effect is influenced by thermal gradients, which can be schematized as vertical but also

horizontal, as described when studying the framework thickness influence.

The coefficient of thermal expansion (CTE) mismatch effect,which explains the development

of interior compressive stresses when the framework CTE is higher than the veneer CTE.

These stresses increase with cooling rate, but decrease with veneer thickness.

The stress-induced crystalline transformation effect, which is a hypothesis to explain the

presence of interior tensile stresses in zirconiabased samples,and their inverted variation in

comparison with metal-based samples.

If VM samples have always exhibited favorable residual stress profiles,whatever the studied

parameter,VZr samples have often shown unfavorable profiles,particularly in terms of interior

stresses,which have led to the formulation of the hypothesis of the crystalline transformation.

The results highlight the lack of tolerance and the sophisticated behavior of zirconia in

comparison with metal. Notably the measured profiles outline the difficulty of defining the

adequate ratio between veneer thickness and zirconia framework thickness. The reason

underlying this difficulty is that the promotion of residual stress development adversely

affects the ceramic support,which in turn negatively impacts residual stress development.

Thin frameworks,under 1 mm thick,were found to promote tensile stresses.The slow cooling

procedures were shown to promote tensile stress development in VZr samples, this finding

does not support manufacturers recommendations.

Finally, this work is a first step of a global approach, which should be followed by well-

controlled in vitro traditional mechanical tests and clinical studies.

1.

2.

3.


22.1 Veneering and zirconia phase transformation: preliminary results

The hypothesis of the stress-induced crystalline transformation of zirconia came first from

the results about the influence of cooling rate on residual stress profile. This hypothesis was

corroborated by further experimentations about the veneer and the framework thickness

influence.

To proof the presence of a transformed area and monoclinic crystals, different methods

can be useful, such as:

Observation: scanning electron microscopy (SEM) and atomic force microscopy (AFM).

These techniques can demonstrate the presence of transformed grains due to their typical

morphology when transformed into monoclinic: faceting and surface uplifts are

detectable on the surface.

Structural analysis: X Ray or neutron diffraction and Raman spectroscopy.

These methods can demonstrate and quantify the presence of monoclinic phase by

establishing the structural spectra of zirconia, with its different peaks characteristic of the

different types of crystals.The challenge of those techniques is maintaining successful focus

on the interface.

The difficulty encountered of both techniques is the elimination of the veneering ceramic

before the observation of the zirconia surface.This procedure,which involves the application

of fluorhydric acid, is risky, since it can eliminate or modify a thin layer of the transformed

zirconia.

To access the interface, the sample can also be sectioned, but the inconvenience of this

option is that the procedure can potentially generate the transformation. The use of a

focus ionic beam (FIB) allows the accurate elimination of a material layer,without triggering

the transformation. This procedure can be performed before SEM observations.

Preliminary tests were performed on VZr samples with a 0.7 mm thick framework and a

1.5 mm thick veneering ceramic, submitted to the slow cooling procedure. This sample,

which exhibited high interior tensile stress, was sectioned perpendicularly to the veneer

Perspectives

-

-

Residual Stress in Veneering Ceramic I Amélie Mainjot140

zirconia interface, enrobed in metacrylate resin, and polished. In collaboration with the

MATEIS unit research (J.Chevalier, L. Gremillard, T.Douillard, UMR CNRS 5510, INSA Lyon,

France) preliminary observations were carried out on this sample with SEM, and with SEM

after Focus Ionic Beam (FIB) attack.With SEM only, typical faceted crystals were observed

from the surface to 1µm depth of the zirconia framework (Fig. 8.1a and 8.1b).

As the observed faceted crystals only exist near the interface, they cannot have been

generated by the sectioning procedure. A microcrack is also visible on the zirconia surface:

the role of these microcracks, maybe induced by the framework manufacturing process,

needs to be explored in regards to the transformation risk potential. The FIB

Nanotomography (Fig. 8.1c) allowed the 3D-observation of the zirconia framework in the

same area, and confirmed the alteration of the zirconia surface to a depth of 1 µm.

These preliminary results support the hypothesis of the zirconia phase transformation during

the veneering process, and particularly at the interface. However, the origin of the

transformation is still unknown, and the stress-induced process needs to be confirmed.


Fig. 8.1Veneer-zirconia interface of a slow cooled VZr sample. In collaboration with theMATEIS unit research (J.Chevalier, L. Gremillard, T.Douillard, UMR CNRS 5510, INSALyon, France).

Fig. 8.1aSEM photograph at 5000X magnification

Fig. 8.1bSEM photographs at 10000X magnification


Fig. 8.1cFIB Nano-tomography of the same area

Volume: 5.7 X 3.8 X 1.9 µm3

2.2 Further experimentations by hole-drilling

To continue with the chipping problematic understanding, further experimentations are

envisaged, as the study of the influence of:

The coefficient of thermal expansion (CTE) of the veneering ceramic. This parameter is

important, but it will take a long time to investigate it since experimental powders synthesis

is required.

The zirconia surface, relating to the possible presence of microcracks induced by the

manufacturing process. These microcracks could promote the phase transformation and

then potentially modify the stress profile.

The hole-drilling method can also be applied to the study of other veneered framework

materials,metallic or ceramic,such as precious alloys or alumina,and to the study of other

dental materials in general.

-

-



2.3 Clinical considerations: What future for zirconia-based restorations?

Residual measurements are just a piece of the puzzle in the effort to solve the chipping

enigma. They constitute a first step of a global approach comprising in vitro and in vivo

tests,but cannot be directly extrapolated from the clinical situation, since for example, the

geometry of the restoration and the clinical parameters are not taken into account.

However residual stress measurements in veneered zirconia and metal disks highlight the

problematic behavior of zirconia in comparison with metal. If the veneering process of

metal frameworks seems to promote their mechanical behavior, the veneering process of

zirconia frameworks is still not mastered. Moreover zirconia itself is still a misunderstood

material,particularly in regards to the stress-induced phase transformation and the influence

of the manufacturing process, which need to be studied further. In practice, the potential

benefit of slow cooling procedures of the veneering ceramic is questionable, and the

adequate ratio between veneering ceramic and zirconia is hard to define, restricting the

range of indications of zirconia-based restorations until a better understanding of such a

delicate veneering process.Nevertheless, in comparison with metal, zirconia is an excellent

compromise in terms of biocompatibility, strength and optical properties. Zirconia probably

has a future in prosthodontics, if we take the time to master it.We have to look at the past

and remember the porcelain-fused-tometal story, which serves as a reminder that the

development of the veneering concept was a slow process. Today a prudent use of

zirconia-based restorations is recommended, limiting the number of elements,designing and

thickening frameworks (Fig. 8.2), thinning the veneering ceramic and minding occlusal

parameters,which greatly influence chipping,independently of the material choice (Fig.8.3).

Fig. 8.2a - Chipping on the palatal cusp of the first premolar, 5 months after placement, and old chipping on the PFM molar crown.

Fig. 8.2bPFM molar crown:Chipping by scanning-electron microscopy ( epoxy resin replica).

Fig. 8.3Clinical case: chipping on both screwed-retained zirconia bridge on implants and PFM molarcrown, highlighting the importance of occlusal parameters in veneer mechanical behavior.


Fig. 8.2aScrewed-retained zirconia crown on implant: framework andfinal restoration

Fig. 8.2Designed and supportive zirconia frameworks.

Fig. 8.2bZirconia framework for amolar crown

Fig. 8.2cZirconia framework for a screwed-retainedbridge on implants

9Associated works I Amélie Mainjot 147

Associated works

Finite element analysis of residual stress resulting of dental prosthesis manufacture.

Thesis presented by Catherine Lambrechts, to obtain the degree of Mechanical Engineer,

2010, Louvain School of Engineering, Catholic University of Louvain (UCL).

Zirconia framework behavior with the veneering process: comparison of measurements

with finite element analysis.

Internship report presented by Leslie Peters, pregraduated student, 2010, Technology

University Institute, Paris 13 University.

Amélie Mainjot

email : [email protected] of Fixed

ProsthodonticsInstitute of Dentistry

University Hospital (CHU)of Liège, ULg

Quai G. Kurth 454020 Liège

Belgiumtel : + 32 4 270 31 00fax : + 32 4 270 31 10

Keywords

Residual stress, Hole-drilling,

Dental ceramic, Dental crowns,

Zirconia,Veneering ceramic

Residual stress in veneering ceramic

The manufacture of dental crowns and bridges generates residual stresseswithin the veneering ceramic and framework during the cooling process.Knowing the stress distribution within the veneering ceramic as a functionof depth can help the optimizing of manufacturing processes and theunderstanding of failures, particularly chipping, a frequent complicationwith zirconia-based fixed partial dentures.

The first objective of this work was to transfer and to adapt an effectiveindustrial method,the hole-drilling method,for measuring residual stresses todental use, and to demonstrate the method for measurement of residualstresses in veneer-metal (VM) and veneer-zirconia (VZr) disks. The adaptedmethod, presented in the earliest chapters, enables the very low stressesmeasurement in comparison with industrial applications,notably due to the development of a high sensitivity electrical measurement chain.

The second objective was to study the influence of cooling rate, veneerthickness,and framework thickness on residual stress profile,and to comparemeasurements in VM and VZr structures. Results described in the followingchapters reveal that VM samples always exhibited the same type of favorablestress vs. depth profile, starting with compressive stresses at the ceramic surface, decreasing with depth, and then becoming compressive again.VZr samples showed varying stress profiles, some describing a worse curve,where the in-surface compressive stresses were found to turn into tensilestresses in the interior. A chronological two-step approach is discussed toexplain residual stress development in metal-based samples, and a three-step approach, comprising the hypothesis of the phase transformation, isproposed for zirconia-based samples.The results highlighted the difficulty ofdefining the adequate ratio between veneer and zirconia framework thickness, and the slow cooling procedures were shown to promote tensilestress development in VZr samples.

these 170x240 cd - uliege.bebictel.ulg.ac.be/etd-db/collection/available/... · je tiens à...

Documents