Adhesion of resin composites to biomaterials in dentistry Özcan, Mutlu

University of Groningen

Adhesion of resin composites to biomaterials in dentistryÖzcan, Mutlu

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2003

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Özcan, M. (2003). Adhesion of resin composites to biomaterials in dentistry: an evaluation of surfaceconditioning methods Groningen: s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 13-04-2018

https://www.rug.nl/research/portal/en/publications/adhesion-of-resin-composites-to-biomaterials-in-dentistry(d97f3bdc-de61-495e-bbdc-e506899c2733).html

Adhesion of Resin Composites

to Biomaterials in Dentistry:

An Evaluation of Surface

Conditioning Methods

MUTLU ÖZCAN

Adhesion of R

esin Com

posites to Biom

aterials in Dentistry:A

n Evaluation of S

urface Conditioning M

ethodsM

UT

LU

ÖZ

CA

N





MUTLU ÖZCAN

1

binnenwerk 17-11-2003 11:12 Pagina 1

MUTLU ÖZCAN

Adhesion of Resin Composites to Biomaterials in Dentistry: An Evaluation ofSurface Conditioning Methods.Thesis at the University of Groningen. With references and Dutch summary.ISBN 90-367-1942-9

Subject headings: adhesion / air-borne particle abrasion / amalgam / ceramics /dental materials / fiber-reinforced-composites / fracture / hydrofluoric acidetching / particulate filler composites / post-cores / repair / silica coating /silanization / surface conditioning methods

Cover design by Mutlu ÖzcanThe front cover shows the Scanning Electron Microscope photo of hydrofluoricacid conditioned lithium disilicate ceramic (x1000) and the back cover showsthe silica coated surface of the particulate filler composite (x5000).

Copyright © M. ÖZCAN, Groningen, 2003All rights reserved. No part of this publication may be reproduced ortransmitted in any form or by any means-graphics, electronic, or mechanical,including photocopying, recording, taping or information storage and retrievalsystems without permission of the copyright owner.Printed by Facilitairbedrijf, Groningen

2


RIJKSUNIVERSITEIT GRONINGEN

Adhesion of Resin Composites to

Biomaterials in Dentistry:



PROEFSCHRIFTter verkrijging van het doctoraat in de

Medische Wetenschappenaan de Rijksuniversiteit Groningen

op gezag van deRector Magnificus, dr. F. Zwarts, in het openbaar te verdedigen op

woensdag 10 december 2003om 14:15 uur

door

Mutlu Özcan

geboren op 17 juni 1969te Erzincan, Turkije

3


Promotores:Prof. dr. M-Ch.D.N.J.M. HuysmansProf. dr. W. Kalk

Beoordelingscommissie:Prof. dr. ir. H.J. BusscherProf. dr. A.J. FeilzerProf. dr. P.K. Vallittu

ISBN-nummer: 90-367-1942-9

4


To my parents

5


Paranimfen: Michiel H.J. DoffJohannes Brendeke

6


Contents

Page

Chapter I General Introduction 9

Chapter II Effect of Surface Conditioning Methods on the Bond Strength of Luting Cement to Ceramics 57

Chapter III Resistance of Core Materials Against TorsionalForces on Differently Conditioned Titanium Posts 73

Chapter IV Bonding Polycarbonate Brackets to Ceramic:Effects of Substrate Treatment on Bond Strength 91

Chapter V Effect of Three Surface Conditioning Methods to ImproveBond Strength of Particulate Filler Resin Composites 109

Chapter VI Bond Strength of Resin Composite to DifferentlyConditioned Amalgam 125

Chapter VII General Discussion and Future Research 141

Chapter VIII Summary 153

Dutch Summary 161

Acknowledgements 167

Curriculum Vitae 168

7


8


Chapter General Introduction

9

1


10


The phenomenon of adhesion, here meaning the physico-chemical adhesionof two dental materials rather than the adhesion of biological substancesexisting in the oral cavity, is involved in almost all disciplines of dentistry.Basically, adhesion is the force that causes two substances to attach whenthey are brought into intimate contact with each other. The molecules of onesubstance adhere or are attracted to molecules of another. This force is calledadhesion when unlike molecules are attracted and cohesion when moleculesof the same kind are attracted. The material or film added to produce adhesionis known as the adhesive and the material to which it is applied is called theadherend or substrate.1 Adhesion in dental applications can be considered toinclude two categories: one to the dental tissues like enamel, dentin or cementand the other to artificial materials. Artificial materials used in dentistry are, ingeneral terms, biomaterials. By definition, any material or substance (otherthan a drug) or combination of materials, synthetic or natural in origin, whichcan be used as a whole or as a part of a system which treats, augments, orreplaces any tissue, organ, or function of the body are called biomaterials.2

A plethora of dental biomaterials is being introduced in dentistry forvarious indications that require attachment to the tooth substance or tosubstrates using adhesive means. Despite the increased effort to improve theadhesion between various restorative and prosthetic materials in dentalapplications, adhesive and/or cohesive failures are still being experiencedeither in the form of debonding, delamination or fractures.3,4

Many factors play a role in the failure of dental restorations such as theinherent physical features of the materials, external factors or insufficientadhesion. In fact, even if the restorations were made at ideal conditions,experiencing failures is not surprising. The differences between the elasticproperties and thermal expansion coefficients of the dental materials that theclinicians try to adhere to are sometimes large. They function in a moist andaggressive oral environment and they are exposed to thermal and mechanicalfatigue, static and impact forces caused by the occluding teeth. The result iscatastrophic failure of the restoration either at the interface or within thematerial itself.

The presence of minute flaws contributes to failure in many ways,primarily where the restoration or restoration-tooth system flexes underloading, causing deflection and resulting in loss of adhesion or debondingbetween materials. The failures may also be simply experienced due totechnical mistakes during the preparation of the restorations and occasionalpresence of flaws. Mechanical fatigue of restorative materials is mainlygoverned by mechanisms that are related to material properties includingmicrostructure, crack length and fracture toughness, as well as to applied

11


stress intensity. Other factors like properties of resin adhesives, lutingcements, design of the restoration, voids in cement layers, surfacemicrodefects within the material, all may influence the fracture resistance ofthe restoration.

Failures related to debonding and difficulties in achieving durableadhesion required, have initiated this research in order to find the bettermethods for conditioning a number of dental biomaterials prior to cementation,recementation, lamination or repair actions. Although the body of research ondental adhesion is large, some basic principles and knowledge still seem to beneglected leading to shorter survival rate than would be expected from oftenexpensive dental appliances. Some of the reported failure rates are listedbelow:

1-Clinical failure rates of adhesively luted ceramic restorations rangedbetween 0.6 to 5 % per year. By location of the pontic, failure rates were notedas 0 to 11 % for premolars and 24 % for molars. Mechanical problemsaccounted for 13 to 33 % of the failed units up to 3 years.5-7

2-Posts for endodontically treated teeth have received considerable attentionin the dental scientific literature, but there is sparse in vivo research andcertain information from in vitro investigations is contradictory about theadhesion of core materials to the metal posts. The retrospective studiesindicated failure rate ranging between 1 to 7.5 % related to loss of adhesion ofcores to metal posts.8

3-With the increased demand for adult orthodontics, the clinicians often lutebrackets and retainer wires to the ceramic parts of fixed-partial-dentures.However, the failure rates are reported to be 9.8 % in 2 years. Even taking intoaccount the short duration of bracket applications in orthodontics compared toother adhesive procedures in restorative dentistry, this failure rate is high.9

4-Clinical studies related to the survival rate of direct or indirect resincomposite restorations revealed 30-60 % of all restorations had been replacedafter 3-8 years mainly due to secondary caries followed by discoloration,degradation, microleakage, wear or ditching at the margins. It was reportedthat the clinicians spend the majority of their chair side time replacingrestorations and the amount of time was found the highest for resin compositerestorations.10-14

5-Amalgam has served dentistry for more than a century. Longevity is reportedup to 6 years with annual failure rates of 0.5-6.6 %.5 The results of recentsurveys from cross-sectional studies indicate that complete cusp fracture ofposterior teeth associated with amalgam restorations is a problem in dentalpractice.The failure rate range between 4.4 and 14 occasions per 100 subjects

12


or 20.5 teeth per 1000 persons a year.15-17 Although only a few surveys existin the literature, strictly speaking, the reported failure rates are not high.Nevertheless any kind of failure is an unpleasant experience that posesaesthetic and functional dilemma and is costly both for the patients and thedentist.

Replacement of a failed restoration is not necessarily the most practicalsolution. Before making an attempt to restore failed restoration, cliniciansshould know the underlying reasons for the failure.Therefore knowledge on thefailure phenomena and the adhesion characteristics related to the underlyingmaterial should be well-recognized to receive the best possible adhesion.

A summary of the materials of particular concern in this study is listed inTables 1a-c.

Background information

A good adhesion to the dental tissues has been achieved successfully over thefew decades. Current research efforts are also aimed at the optimization of theadhesion between composites to metals (alloys), composites to ceramics andcomposites to other composites. A number of surface conditioning methodshave been developed over the last few decades to produce adequate adhesionto the adherend for restorations. Advances in adhesive dentistry have resultedin the introduction of surface conditioning methods using chemical agents orair-borne particle abrasion of the adherend prior to bonding the adhesive inorder to achieve optimum adhesion.18 A summary of conditioning techniquesfor restorative materials is presented in Table 2. These methods were initiallybased on macro or micromechanical retention either lying on the adherend likeretention beads or located within the adherend such as grooves or undercuts.However, the methods based on mechanical retention mechanisms sufferedfrom unreliable bonding strengths or gap formation between the adhesive andthe adherend leading to microlekage. Also, while external retentions promotedovercontouring of the restoration, internal ones produced by drilling couldintroduce additional cracks or deformations and distortions in the adherend.Other surface conditioning methods used for alloys were based on creatingmicromechanical retention through electrochemical etching or electrolytic tinplating but unfortunately they were not applicable to all alloys.

Chemical conditioning methods rely on less invasive techniques viaparticle deposition on the adherend surface followed by coating the surfacewith a silane coupling agent and/or an intermediate monomer resin. One suchsystem is tribochemical silica (SiO2) coating. In this technique, surfaces are

13


air-abraded with aluminium trioxide (Al2O3) particles coated with silica. Theblasting pressure results in the embedding of these silica coated aluminaparticles in the surface, rendering the silica-modified surface chemically morereactive to the adhesive through silane coupling agents.19-23

The most popular chemical agent for conditioning dental ceramic orparticulate filler composite surfaces prior to cementation is hydrofluoric (HF)acid gel. The greatest advantage of the use of HF acid gel is that it is verysimple to apply in chair side procedures. Furthermore, restorations can be re-etched in case of failure without the need of laboratory procedures. Itselectively dissolves the glassy matrix and causes physical alteration topromote adhesion of resin composite to the porous surface of the ceramic.Unfortunately, HF acid gel is a strong caustic agent that may create irritationsof soft tissues.24 Although intraoral use of this acid gel should be seriouslyquestioned, most of the manufacturers still do recommend it particularly priorto cementation of adhesively luted restorations.

When two dissimilar materials are to be attached, whethermicromechanical or chemical conditioning is chosen, the use of silanecoupling agents could provide increased attachment.

Silane coupling agents

Silanes are hybrid organic-inorganic compounds that can function asmediators, coupling agents to promote adhesion between dissimilar, inorganicand organic materials.25-27 There are several explanations that describe whatcould happen at the interface during silane reactions. Briefly listed: a)Chemical bonding theory is the best known and its main idea is that silanesimprove adhesion by formation of stable, covalent siloxane (Si-O-Si), bondsand metallo-siloxane bonds (Si-O-M); b) Deformable layer theory points to theplasticity of the interface region; c) Surface wettability theory suggests theimprovement of adhesive strength by physical adsorption; d) The restrainedlayer theory states that some mechanical stress transfer between the phasestake place; e) The reversible hydrolytic bond theory combines aspects ofchemical bonding with the rigid interface of the restrained layer theory; f) Anadhesion between silica and silane, made by two types of bonds, viz. siloxanebridges and hydrogen bonds; g) Consisting of ionomer bonding,interpenetration, both soft and rigid layer theories; h) The silane modifies theoxide layer on the substrate and forms a conversion layer. This layer is differentwith its electrochemical properties than those on the silanes and the metaloxide.

14


Silane coupling agents used in dentistry were at first vinyltrimethoxysilaneand subsequently mainly 3-methacryloxypropyltrimethoxysilane, �-MPS (also�-MPTS) (3-methoxysilylpropylmethacrylate). These trialkoxyorganosilaneshave an organic functional radical (R’, e.g. vinyl- or methacryloxypropyl) thatcan co-polymerize with a resin composite. They have also three alkoxy groups(e.g. methoxy -O-CH3) ready to hydrolyze in a water-alcohol solution, and thento react with surface hydroxyl group on an inorganic substrate (e.g. silicasurface, oxide layer on a metal surface, ceramic, E-glass). In other words,surfaces to be silanated must contain OH-groups sufficiently. The functionalalkoxyl groups react as acid catalysed (usually in pH 4-5) in aqueous alcoholsolution, to form at the first stage (i.e. hydrolysis) labile intermediate acidic silanolgroups (-Si-OH):

R’-Si(OR)3 + 3 H2O ➔ R’-Si(OH)3 + 3 R-OH (1)

The silanol groups then condense to form dimeric (and then oligomeric) molecules:

R-Si(OH)3 + R-Si(OH)3 ➔ R-Si(OH)2-O-Si-(R)(OH)2 + H2O (2)

In the next fast step, they form a three dimensional highly cross-linkedpolysiloxane (-Si-O-Si-) layer with covalent bonds, and also -Si-O-M (M =metal) bonds with hydroxyl groups on the substrate surface. Water iseliminated in this reaction and methanol is released but in minute amounts.

R R| |

…R-Si-(OH)2-O-Si-(R)(OH)-… + 2OH-M ➔ -R-Si-O-Si-O-… + ➔ etc. (3)I IO OI IM M

Commercial dental silanes are typically pre-hydrolyzed or in other words,ready to use. The advantage of silane coupling agents appear to enhancebond strength by promoting a chemical bond between the adhesive and theadherend. Although early commercial silane solutions suffered from instability,they have been steadily improved demonstrating higher bond strengths.28,29

15


Relevance and objectives of this thesis

The work described in this thesis was initiated as a result of the recognition offailures experienced in clinical dentistry. It aims to elucidate the relationsbetween the material properties and the adhesion principles. Moreover thelarge variations of new dental biomaterials needed to be studied from theperspective of their adhesive behaviour in order to find the optimalcombination. Therefore, the overall aim of this thesis was to increase ourknowledge on the interactions between surface conditioning methods used fordifferent dental restorative, prosthetic and orthodontic materials, and theadhesives, with a particular emphasis on ceramics, amalgam, titanium metaland particulate filler resin composites and to chose the most suitable methodfor the specific substrate. Such knowledge would help the clinicians to prolongthe service life of the restorations. This allows minimally invasive and costeffective manner of treatment.

References:

1. Phillips RW. Skinner`s Science of Dental Materials, WB Saunders Co, p.5-10, 1991.

2. Bronzino JD. The Biomedical Engineering Handbook, 2nd Ed. Volume II,Boca Raton: CRC Press LLC, p.XII-10, 2000.

3. Creugers NH, Kayser AF, Van't Hof MA. A seven-and-a-half-year survivalstudy of resin-bonded bridges. Journal of Dental Research, 71:1822-1855,1992.

4. Özcan M, Niedermeier W. Clinical study on the reasons and location of thefailures of metal-ceramic restorations and survival of repairs. InternationalJournal of Prosthodontics, 15:299-302, 2002.

5. Roulet J-F. Benefits and disadvantages of tooth-coloured alternatives toamalgam. Journal of Dentistry, 25:459-473, 1997.

6. Sorensen JA, Kang SK, Torres TJ, Knode H. In-Ceram fixed partialdentures: three-year clinical trial results. Journal of Californian DentalAssociation, 26:207-214, 1998.

7. Gemalmaz D, Özcan M, Alkumru HN. A clinical evaluation of ceramicinlays bonded with different luting agents. Journal of Adhesive Dentistry,3:273-283, 2001.

8. Käyser AF, Mentink AG, Meeuwissen R, Leempoel PJ. The clinicalbehavior of metal-composite post and core build-up. Results of a pilotstudy. Nederlands Tijdschrift voor Tandheelkunde, 99:401-403, 1992.

16


9. Zachrisson BU. Orthodontic bonding to artificial tooth surfaces: Clinicalversus laboratory findings. American Journal of Orthodontics andDentofacial Orthopedics, 117:592-594, 2000.

10. Mjør IA, Toffenetti F. Placement and replacement of resin-based compositerestorations in Italy. Operative Dentistry, 17:82-85, 1992.

11. Mjør IA, Moorhead JE. Selection of restorative materials, reasons forreplacement, and longevity of restorations in Florida. Journal of AmericanCollege of Dentistry, 65:27-33, 1998.

12. Burke FJ, Cheung SW, Mjør IA, Wilson NH. Restoration longevity andanalysis of reasons for the placement and replacement of restorationsprovided by vocational dental practitioners and their trainers in the UnitedKingdom. Quintessence International, 30:234-242, 1999.

13. Al-Negrish AR. Composite resin restorations: a cross-sectional survey ofplacement and replacement in Jordan. International Dental Journal,52:461-468, 2002.

14. Mjør IA, Shen C, Eliasson ST, Richter S. Placement and replacement ofrestorations in general dental practice in Iceland. Operative Dentistry,27:117-123, 2002.

15. Bader JD, Martin JA, Shugars DA. Preliminary estimates of the incidenceand consequences of tooth fracture. Journal of American DentalAssociation, 126:1650-1654, 1995.

16. Heft MW, Gilbert GH, Dolan TA, Foerster U. Restoration fractures, cuspfractures and root fragments in a diverse sample of adults: 24-monthincidence. Journal of American Dental Association, 131:1459-1464, 2000.

17. Fennis WM, Kuijs RH, Kreulen CM, Roeters FJ, Creugers NH, BurgersdijkRC. A survey of cusp fractures in a population of general dental practices.International Journal of Prosthodontics, 15:559-563, 2002.

18. Özcan M, Pfeiffer P, Nergiz I. A brief history and current status ofmetal/ceramic surface conditioning concepts for resin bonding in dentistry.Quintessence International, 29:713-724, 1998.

19. Guggenberger R. Rocatec system-adhesion by tribochemical coating.Deutsche Zahnärztliche Zeitschrift, 44:874-876, 1989.

20. Özcan M, Alkumru H, Gemalmaz D. The effect of surface treatment on theshear bond strength of luting cement to a glass infiltrated alumina ceramic.International Journal of Prosthodontics, 14:335-339, 2001.

21. Özcan M. The use of chairside silica coating for different dentalapplications. Journal of Prosthetic Dentistry, 87:469-472, 2002.

22. Özcan M, Akkaya A. New approach to bonding all-ceramic adhesive fixedpartial dentures: A clinical report. Journal of Prosthetic Dentistry, 88:252-254, 2002.

17


23. Pfeiffer P, Nergiz I, Özcan M. Metal surface conditioning concepts for resinbonding in dentistry. In: Adhesion Aspects of Polymeric Coatings. K.L.Mittal (Ed). Vol. 2, p:137-149, VSP, Zeist NL, 2003.

24. MacKinnon MA. Hydrofluoric acid burns. Dermatologic Clinics. 6:67-74,1988.

25. Clark HA, Plueddemann EP. Bonding of silane coupling agents in glass-reinforced plastics. Modern Plastics, 40:133-96, 1963.

26. Plueddemann EP. Adhesion through silane coupling agents. Journal ofAdhesion, 2:184-201, 1970.

27. Matinlinna JP, Lassila LVJ, Özcan M, Yli-Urpo A, Vallittu PK. Anintroduction to silanes and their clinical applications in dentistry.International Journal of Prosthodontics, (in press, 2003).

28. Bowen RL. Properties of a silica-reinforced polymer for dental restorations.Journal of American Dental Association, 66:57-64, 1963.

29. Matinlinna JP, Özcan M, Lassila LVJ, Vallittu PK. Comparison of effect of a3-Methacryloxypropyltrimethoxysilane and vinyltriisoproxysilane blend andtris(3-trimethoxysilylpropyl)isocyanurate on the shear bond strength ofcomposite resin to titanium metal. Dental Materials, (in press, 2003).

18


Table 1a. Trade names, types of ceramics, titanium, amalgam used as substrates and the manufacturing company names.

Trade name Type Manufacturers

Finesse leucite reinforced Ceramco, Burlington, NJ, USA

In-Ceram glass-infiltrated alumina (70%) Vita Zahnfabrik, Bad Säckingen, Germany

Zirkonia Blank for Celay glass-infiltrated zirconia Vita Zahnfabrik, Bad Säckingen, Germany

IPS Empress 2 lithium disilicate Ivoclar, Vivadent AG, Schaan, Liechtenstein

Procera AllCeram high alumina (99.9%) Nobel Biocare AB, Göteborg, Sweden

Experimental alumina high alumina (99.7%) Tampere University of Technology, Tampere, Finland

VMK68 Feldspathic ceramic Vita Zahnfabrik, Bad Säckingen, Germany

Titanium DIN 17850-T4/3.70651

Amalgam non-gamma 2, lathe-cut, ANA 2000 Duet, Nordiska Dental AB, Ängelholm, Sweden

high-copper alloy

Table1b. Monomer matrix types of particulate filler composites used as substrates and the manufacturing company names.

Trade name Matrix type Manufacturer

Gradia UEDMA/ ethylene dimethacrylate1 GC, Alsip, IL, USA

Sculpture dimethacrylate2 Jeneric Pentron, Wallingford, CT, USA

Sinfony HEMA/diacrylate3 3M ESPE, Seefeld, Germany

Targis Bis-GMA, DDDMA, UEDMA, TEGDMA4 Ivoclar Vivadent AG, Schaan, Liechtenstein

Tetric Ceram Bis-GMA, UEDMA, TEGDMA5 Ivoclar Vivadent AG, Schaan, Liechtenstein

Bis-GMA= Bis-phenol-A-glycidylmethacrylate

UEDMA= Urethane dimethacrylate

TEGDMA= Triethyleneglycoldimethacrylate

DDDMA= Decandioldimethacrylate

HEMA= 2-hydroxyethylmethacrylate

1 UDMA (10-25 %) and ethylene dimethacrylate (5-10 %)

2 dimethacrylate

3 10-30- % (octahydro-4,7-methano-1H-indenediyl) bis(methylenediacrylate)

4 Bis-GMA (9 %), DDMA (4,8 %), UEDMA (9,3 %)

5 Bis-GMA (< 9 %), TEGDMA (< 5 %), UEDMA (< 8 %)

19


Table 1c. Filler types and percentages of particulate filler composites.

Trade name Filler type and content

Gradia Alumina silicate glass (40-50 w-%), amorphous precipitated silica (5-10 w-%)

Sculpture Glass-infiltrated alumina (70 w-%)

Sinfony Strontium-aluminium borosilicate glass, silicon dioxide (50 w-%)

Targis Silanized Ba-glass fillers (46.2 w-%), highly dispersed silica (11.8 w-%), mixed oxides (18.2 w-%),

catalyst and stabilizers (0.6 w-%), pigments (≤ 0.1 w-%)

Tetric Ceram Silanized Ba-glass, ytterbium trifluoride, silanized unknown metal oxide, silanated

barium-aluminium-fluoro-silicate glass, silanated silica glass (79 w-%)

Table 2. Characteristics of conditioning principles with some details, brand names and the manufacturers of surface

conditioning methods assessed.

Conditioning principle Details Brand names and manufacturers

Orthophosphoric acid (37 %, 60 s) Ultradent® Ultraetch, South Jordan, UT, USA

Hydrofluoric acid (9.5 %, 90 s) Ultradent Porcelain Etch®, South Jordan, UT, USA

(5 %, 20 s) IPS Empress Ceramic Etch, Vivadent AG, Schaan,

Liechtenstein

Air-particle abrasion (110 µm aluminium trioxide, Korox®, Bego, Bremen, Germany

380 kPa, 10 mm, 13 s)

Chair side air-particle abrasion (30 µm aluminium trioxide, Korox®, Bego, Bremen, Germany

250 kPa, 10 mm, 4 s)

Tribochemical silica coating Rocatec® Pre, Rocatec® Plus Rocatec™, 3M ESPE AG, Seefeld, Germany

(280 kPa, 10 mm, 13 s)

Chair side silica coating (CoJet®-Sand, 30 µm CoJet®, 3M ESPE AG, Seefeld, Germany

SiO2 coated Al2O3,

so called SiOx,

250 kPa, 10 mm, 4 s)

Silica coating-silanization Silicoater® Classical, Silicoater® MD, Siloc®,

Heraeus-Kulzer GmbH, Wehrheim, Germany

Acrylization Kevloc®, Heraeus-Kulzer GmbH, Wehrheim, Germany

Silane coupling agent 5 min ESPE®-Sil, 3M ESPE AG, Seefeld, Germany

Alloy primer Alloy Primer™, Kuraray Medical Co, Ltd, Tokyo, Japan

Pre-impregnated bidirectional everNET™, StickTech, Turku, Finland

E-glass fiber sheets

20


Outline of this thesis

In Chapter II the effect of three surface conditioning methods on the bondstrength of a Bis-GMA based luting cement to six dental ceramics wereassessed by shear bond test.

In Chapter III the resistance of six core materials on titanium posts after usingfive surface conditioning methods and two types of opaquers were evaluatedby means of electronic rotational torque device.

In Chapter IV the effect of five surface conditioning methods on the bondstrength of polycarbonate brackets to feldspathic ceramic surfaces wereevaluated by shear bond test. Furthermore after debonding, the adhesiveremnant index was used for classifying failure modes and the fracture surfaceswere analyzed using scanning electron microscope.

In Chapter V the effect of three surface conditioning methods on the bondstrength of a diacrylate resin to five particulate filler resin composites wereassessed by shear bond test and the surfaces were analyzed using scanningelectron microscope after the conditioning methods.

In Chapter VI the effect of seven conditions on the bond strength of a resincomposite to amalgam surfaces were evaluated using shear bond test and thecorrelation between surface roughness and bond strength was assessed.

In Chapter VII the results are discussed and ongoing or further plannedstudies are mentioned and in Chapter VIII these studies together with thediscussion and future research ideas are summarized.

21


This thesis is based on the following papers:

1- Özcan, M.: Evaluation of Alternative Intraoral Repair Techniques forFractured Ceramic-fused-to-metal Restorations. Journal of OralRehabilitation, 30(2): 194-203, 2003.

2- Özcan, M.: Fracture Reasons in Ceramic-fused-to-metal Restorations.Journal of Oral Rehabilitation, 30(3): 265-269, 2003.

3- Özcan, M., Vallittu, P.K.: Effect of Surface Conditioning Methods on theBond Strength of Luting Cement to Ceramics. Dental Materials, 19(8):725-731, 2003.

4- Akıslı, I., Özcan, M., Nergiz, I.: Resistance of Core Materials AgainstTorsional Forces on Differently Conditioned Titanium Posts. Journal ofProsthetic Dentistry, 88;367-374, 2002.

5- Özcan, M., Vallittu, P.K., Peltomäki, T., Huysmans, M-Ch., Kalk, W.:Bonding Polycarbonate Brackets to Ceramic: Effects of SubstrateTreatment on Bond Strength. American Journal of Orthodontics andDentofacial Orthopedics, (in press, 2003).

6- Özcan, M., Alander, P., Vallittu, P.K., Huysmans, M-Ch., Kalk, W.: Effect ofThree Surface Conditioning Methods to Improve Bond Strength ofParticulate Filler Resin Composites. Journal of Materials Science:Materials in Medicine, (submitted, 2003).

7- Özcan, M., Vallittu, P.K., Huysmans, M-Ch, Kalk, W, Vahlberg, T.: BondStrength of Resin Composite to Differently Conditioned Amalgam.Operative Dentistry, (submitted, 2003).

22


Fracture Reasons in Ceramic-fused-to-metal

Restorations

Özcan, M.: Fracture Reasons in Ceramic-fused-to-metal Restorations. Journalof Oral Rehabilitation 30(3): 265-269, 2003. (reproduced with permission ofBlackwell Publishing)

23


24


Fracture Reasons in Ceramic-fused-to-metal

Restorations

MUTLU ÖZCAN

*Marmara University, Dentistry Faculty, Department of Prosthodontics, Istanbul, Turkey

SUMMARY Ceramic-fused-to-metal restorations are widely used in dentistry with a highdegree of general success. Fracture of the ceramic veneers as a result of oralfunction or trauma is not an uncommon problem in clinical practice. Althoughfractures of such restorations do not necessarily mean the failure of therestoration, the renewal process is both costly and time consuming andtherefore remains a clinical problem. Fractures in the anterior region pose anaesthetic problem but when they are in the posterior, chewing function couldalso be affected. The published literature reveals that reasons for failures covera wide spectrum from iatrogenic factors to laboratory mistakes or due tofactors related to the inherent structure of the ceramics or simply to trauma.

Introduction

Because of their excellent biocompatibility and superior aesthetic qualities,ceramic-fused-to-metal crowns and bridges are commonly applied in fixedprosthodontics. Despite the increased effort to improve the bond strengthbetween the ceramic and the metal substrate, on occasion, fractures ofceramic veneers still occur under clinical conditions. The reasons for suchfailures are frequently repeated stresses and strains during chewing functionor trauma. Clinical studies indicated that the prevalence of ceramic fracturesranged between 5-10% over 10 years of use (Coornaert, Adrians & de Boever,1984).

Ceramic fractures are serious and costly problems in dentistry. Moreover,they pose an aesthetic and functional dilemma both for the patient and thedentist. Therefore the intent of this paper is to review the published literatureon the reasons for fractures, concentrating on the data obtained both from invitro and in vivo studies.

25


Failure rates

Many patients are still in need of fixed restoration replacements due to somefailures in those restorations. Only a few studies in the literature have dealt withthe survival rates of metal-ceramic restorations.

In a clinical follow up study by Coornaert et al. (1984), the prevalence offractures in metal-ceramic crowns was found to be approximately 5% over 10years of function. Strub, Stiffler & Schärer (1988) observed a failure rate ofmetal-ceramic restorations of only between 1% to 3% over 5 years. Studies byKarlsson (1986) revealed a 93% success rate for fixed bridge restorationsduring a 10-year period, while Palmqvist & Schwartz (1993) reported a 79%success rate over a 18 to 23 year period. The survival rates obtained by Glantzet al. (1993) as a function of time between 1979 and 1994 indicated that mostof the debondings occured over 15 years and almost all recordeddislodgements were observed within 5 years of placement. Subsequent clinicalresults from Hankinson & Cappetta (1994) and Kelsey et al. (1995) exhibited 2to 4% failure rates after two years of function, rising from 20 to 25% after 4 to5 years due to consistent repeating occlusal contacts.

In another clinical retrospective analysis, 1219 three-unit fixed bridgesand 1618 single crowns in the anterior region were evaluated between 1969and 1989 (Kerschbaum, Seth & Teeuwen, 1997). The results of the studysupported the superiority of metal-ceramic systems over acrylic-veneeredcrowns with 2 to 4% failure rates after 2 years of function. Statistical analysishowever, showed that after 10 years, 88.7% of the metal-ceramic crowns and80.2% of the metal-ceramic bridges were still in function.

Overall survival rate of metal-ceramic restorations demonstrate a paradoxin the different survival rate values in the literature. It is well recognized thatmany factors are involved in the success rate assessments of fixed partialdentures limiting the longevity of the restorations.

Factors affecting failure

Failure of the restorations is in fact a multifactorial problem which could berelated to a combination of different reasons. Optimization of the metal-ceramic restorations requires knowledge of the failure phenomena. Numerousstudies over the years have focused on reasons for failure.

Mechanical failures of metal-ceramic systems are not surprisingconsidering the vast differences in modulus between the metal and ceramicmaterials. When feldspathic dental porcelain is cooled, the leucite crystals

26


contract more than the surrounding glass matrix leading to the development oftangential compressive stresses around the leucite particles as well as tomicrocracks within and around the crystals (Hasselman & Fulathy, 1966;Mackert, 1988; Anusavice & Zhang, 1998; Denry, Hollowey & Rosentiel, 1998).

Twenty to 30% reduction in metal-ceramic strength was found in a moistenvironment (Sherill & O`Brein, 1974). Michalske & Freiman (1982) indicatedthat silicate bonds in the glassy ceramic matrix are susceptible to hydrolysis byenvironmental moisture in the presence of mechanical stress. The porcelainrestoration functions in a moist environment, which may allow static fatigue tocause the propagation of fractures along the microcracks resulting in failure ofthe restoration. The environment of the oral cavity was found to aggravate thestrength of dental ceramics. The silicon-oxgen bond was found to becomeweaker between the metal and ceramic in the presence of moisture which abetfailure in many ways primarily because of the water propagation at the cracktip (Dauskardt, Marshall & Ritchie, 1990).

The minute scratches present on the surfaces of nearly all materialssometimes behave as sharp notches whose tips are as narrow as the spacingbetween atoms in the materials. Thus, the stress concentration at the tips ofthese minute scratches causes the stress to reach the theoretical strength ofthe material at relatively low average stress. When the theoretical strength ofthe material is exceeded at the tip of the notch, the bond at the notch tipbreaks. As the crack propagates through the material, the stress concentrationis maintained at the crack tip until the crack moves completely through thematerial (Lamon & Evans, 1983). Long anterio-posterior metal substructurealso flexes under heavy or complex loading causing porcelain fracture (Reuter& Brose, 1984).

It was also noted that other reasons for the ceramic fractures are technicalmistakes during the preparation of the restorations and claimed thatoccasional presence of pores inside the ceramic could account for theirweakness and eventual fracture at that site (Oram & Cruicshank-Boyd, 1984).The same results were also found by Øilo (1988) who agreed that suchmistakes markedly increase the failures.

Microcracks in ceramic could also be caused by the condensation,melting, and sintering process of the ceramics on metal due to thermalcoefficient differences (Yamamato, 1985). Faulty design of the metalsubstructure, incompatible thermal coefficients of expansion between themetal substructure and ceramic, excessive porcelain thickness withinadequate metal support, technical flaws in the porcelain application, occlusalforces or trauma were also included as the failure reasons (Diaz-Arnold,Schneider & Aquilino, 1989). Due to the heterogeneous nature of many dental

27


materials, they are likely to contain defects or flaws in various amounts andsizes. Such flaws remain at fixed length unless under load but then theybecome unstable and propagate, catastrophically culminating in fracture.Small changes in microstructure or surface treatment can lead to drasticalterations in service life of fixed restorations and repeated stresses andstrains can cause slow crack growth and mechanical fatigue. Even a singleload cycle can produce measurable cracking at the contact area and damageaccumulation during load cycling (Chadwick, Mason & Sharp, 1993; White etal., 1995).

Wiederhorn (1968, 1974) stressed that during actual masticatoryconditions, restorations are subject to repeated loading over long periods, withsuperposed tangential motion and further claimed that, especially inchemically active aqueous environments, this could greatly exacerbatedamage build up. He stated that the ceramic fracture process might beaccelerated by the environment. It was reported that facings may crack, befractured or damaged as a result of trauma, parafunctional occlusion orinadequate retention between the veneer and the metal (Farah & Craig, 1975).Mechanical fatigue of ceramics on the other hand, is probably governed byseveral mechanisms which are related to material properties includingmicrostructure, crack length and fracture thoughness, as well as to appliedstress intensity (Ban & Anusavice, 1990). Evans et al. (1990) indicated thatevery effort should be made to minimize air entrapment between ceramicparticles since porosity does occur during ceramic application and can impairaesthetics as well as promote fracture. Properties of resin adhesives,cementation agents, preparation designs, voids in cement layers, andthickness of the ceramic restorations were reported to affect the fractureresistance as well (Tsai et al., 1998). In a finite element analysis however, itwas found that the presence of a void in the ceramic structure did have asignificant effect on the fracture (Abu-hassan, Abu-hammad & Harrison, 1998).Another reason for porcelain fracture was attributed to inadequate toothpreparation, which results in too little interocclusal space for the metalsubstructure and porcelain. It was concluded that the improper design of therestoration for the occlusion is the major cause of failure (Creugers, Snoek &Käyser, 1992).

Llobell et al. (1992) described the reasons for intraoral ceramic fracture asimpact load, fatigue load, improper design, microdefects within the material,and added that clinically, mastication, parafunction and intraoral occlusalforces create repetetive dynamic loading. The fatigue failure is preceded by acombination of crack initiation and crack propagation. Finally catastrophicfailure occurs in the form of fracture. It was emphasized that fatigue is of

28


considerable importance for metal-ceramic restorations which are subjected tosmall alternating forces during mastication.

In order to minimize the formation of microcracks a fairly uniform thicknesswas recommended, which may occur during the firing of the ceramic.Avoidance of acute line angled preparations was advised since they enhancethe formation of microcracks within the porcelain during the firing procedures(Burke, 1996). Bertolotti (1997) described the reasons in detail why ceramicmaterials do not yield in the same manner as metals. It is noted thatamorphous materials like glasses or glassy materials do not possess anordered crystalline structure as do metals. Dislocations of a crystalline latticedo not exist in glassy materials and they have no mechanism for yieldingwithout fracture. Dislocations exist in crystalline ceramic materials, but theirmobility is severely limited. The energy required to do this is so large thatdislocations are essentially immobile in crystalline ceramic materials.

Stress direction is another contributory factor for failure as sometimesfailure occurs at sites of relatively low local stress merely because there is aparticularly large flaw oriented in a stress field which is ideal for causingfractures. The possible sites from which failure may start were found to behighly unpredictable, since this depends on flaw size and is related to thestress distribution (White et al., 1997). High biting forces, destructivepremature contacts and common beverages with low pH ranges were reportedto cause glass-containing dental restorations to break down (Anusavice &Zhang, 1998).

The possible failure of ceramics were sometimes attributed toinadequately registered occlusion, material type, spanning of the restoration orinadequate marginal adaptation (Niedermeier et al., 1998).

Özcan (1999) observed that the majority of the ceramic fractures occurduring normal chewing function followed by either trauma or some kinds ofaccidents. Since complications involving fixed partial dentures can also occurduring the preprosthetic preparation phase, Raustia et al. (1998) noted that theclinical skill of the dentist or dental student is extremely important.

Conclusions

Fracture of porcelain is often considered an emergency treatment and therestoration process can present a difficult challenge to the dentist. Clinicalstudies indicated that the prevalence of ceramic fractures ranged between 5-10% over 10 years of use (Coornaert et al., 1984).

Because of the nature of the porcelain processing, new porcelain cannot

29


be added to an existing restoration intraorally. The manual fabrication of metalframeworks and the porcelain veneers is time consuming and requires a highlevel of skill (Freilich et al., 1998). It is an unpleasant experience for the patientand arduous for the dentist to remove these restorations from the mouth.Replacement of a failed restoration is not necessarily the most practicalsolution because of the obviously substantial costs and the complex nature ofthe restoration (Fan, 1991).

The complexities of the oral environment and varied surface topographyof dental restorations make it difficult to precisely define the magnitude andmode of stresses precipitating clinical fracture. The laboratory cannotreproduce intraoral variables and the complexities of the oral environment.When the crowns are cemented intraorally, factors other than inherentmechanical strength of the materials come into play. Under continuousapplication of the mechanical environmental loads, progressive degradationmay lead to crack initiation and growth and ultimately to a catastrophic failureof the restoration. Although failures of ceramic-fused-metal restorations can beovercome by either some repair techniques or renewal of the restoration, it isbeneficial to know the reasons for the failures, especially those due toiatrogenic or technical mistakes, which would help to increase the service timeof such restorations.

References

ABU-HASSAN, M.I., ABU-HAMMAD, O.A. & HARRISON, A. (1998) Strainsand tensile stress distribution in loaded disc-shaped ceramicspecimens:An FEA study. Journal of Oral Rehabilitation, 25, 490-495.

ANUSAVICE, K.J. & ZHANG, N.Z. (1998) Chemical durability of dicor andfluorocanasite-based glass-ceramics. Journal of Dental Research, 77,1553-1559.

BAN, S. & ANUSAVICE, K.J. (1990) Influence of test method on failure stressof brittle dental materials. Journal of Dental Research, 69, 1791-1799.

BERTOLOTTI, R.L. (1997) Alloys for porcelain-fused-to-metal restorations. In:Dental materials and their selection (ed W. O`Brien), pp. 225.Quintessence Publishing Co., USA.

BURKE, F.J.T. (1996) Fracture resistance of teeth restored with dentin-bondedcrowns: The effect of increased tooth preparation. QuintessenceInternational, 27, 115-121.

30


CHADWICK, R.G., MASON, A.G. & SHARP, W. (1998) Attempted evaluationof three porcelain repair systems-What are we really testing? Journal ofOral Rehabilitation, 25, 610-615.

COORNAERT, J., ADRIANS, P. & de BOEVER, J. (1984) Long-term clinicalstudy of porcelain-fused-to-gold restorations. Journal of ProstheticDentistry, 51, 338-342.

CREUGERS, N., SNOEK, P. & KÄYSER, A.F. (1992) An experimentalporcelain repair system evaluated under controlled clinical conditions.Journal of Prosthetic Dentistry, 68, 724-727.

DAUSKARDT, R.H., MARSHALL, D.B. & RITCHIE, R.O. (1990) Cyclic fatigue-crack propagation in magnesia-partially-stabilized zirconia ceramics.Journal of American Ceramic Society, 73, 893-890.

DIAZ-ARNOLD, A.M., SCHNEIDER, R. L. & AQUILINO, S.A. (1989) Bondstrengths of intraoral porcelain repair materials. Journal of ProstheticDentistry, 61, 305-309.

DENRY, I.L., HOLLOWEY, J.A. & ROSENTIEL, S.F. (1998) Effect of ionexchange on the microstructure, strength and thermal expansion behaviorof a leucite-reinforced porcelain. Journal of Dental Research, 77, 583-588.

EVANS, D., BARGHI, N., MALLOY, C. & WINDELER, S. (1990) The influenceof condensation method on porosity and shade of body porcelain. Journalof Prosthetic Dentistry, 63, 380-389.

FAN, P.L. (1991) Porcelain repair materials. Council on dental materials,instrument and equipment prepared at the request of the council. Journalof American Dental Association, 122, 124-130.

FARAH, J.W. & CRAIG, R.G. (1975) Distribution of stresses in porcelain fusedto metal and porcelain jacket crowns. Journal of Dental Research, 54, 255-261.

FREILICH, M.A., KARMAKER, A.C., BURSTONE, C.J. & GOLDBERG, J.(1998) Development and clinical applications of a light-polymerized fibrereinforced composite. Journal of Prosthetic Dentistry, 80, 311-318.

GLANTZ, P.O., NILNER, K., JENDERSEN, M.D. & SUNBERG, H. (1993)Quality of fixed prosthodontics after 15 years. Acta OdontologicaScandinavia, 51, 247-252.

HANKINSON, J.A. & CAPPETTA, E.G. (1994) Five years clinical experiencewith a leucite-reinforced porcelain crown system. International Journal ofPeriodontics and Restorative Dentistry, 14, 138-153.

HASSELMAN, D.P.H. & FULATHY, R.M. (1966) Proposed fracture theory of adispersion-strengthened glass matrix. Journal of American CeramicSociety, 49, 68-76.

31


KARLSSON, S. (1986) Clinical evaluation of fixed bridges 10 years followinginsertion. Journal of Oral Rehabilitation, 13, 423-432.

KELSEY, W.P., CAVEL, T., BLANKENAU, R.J., BARKMEIER, W.W.,WILWERDING, T.M. & LATTA, M.A. (1995) 4 year clinical study of castableceramic crowns. American Journal of Dentistry, 8, 259-262.

KERSCHBAUM, Th., SETH, M. & TEEUWEN, U. (1997) Verweildauer vonKunststoff- und Metal-Keramisch verblendeten Kronen und Brücken.Deutsche Zahnärztliche Zeitschriften, 52, 404-406.

LAMON, J. & EVANS, A.G. (1983) Statistical analysis of bending strength forbrittle solids: A multiaxial fracture a problem. Journal of American CeramicSociety, 66, 177-183.

LLOBELL, A., NICHOLLS, J.I., KOIS, J.C. & DALY, C.H. (1992) Fatigue life ofporcelain repair systems. International Journal of Prosthodontics, 5, 205-213.

MACKERT, J.R. (1988) Effects of thermally induced changes on porcelain-metal compatibility. In: Perspectives in dental ceramics. Proceedings of theforth international symposium on ceramics (ed Preston J.D.), pp. 53.Quintessence Publishing Co., Chicago.

MICHALSKE, T.A. & FREIMAN, S.W. (1982) A molecular interpretation ofstress corrosion in silica. Nature, 295, 511.

NIEDERMEIER, W., PROANO, F.P., ÖZCAN, M., MAYER, B., NERGIZ, I. &PFEIFFER, P. (1998) Enorale Reparaturen mit tribochemischem Verbund.Zahnärztliche Mitteilungen, 16, 54-58.

ØILO, G. (1988) Flexural strength and internal defect of some dental porcelain.Acta Odontologica Scandinavia, 46, 313-322.

ORAM, D.A. & CRUICKSHANK-BOYD, E.H. (1984) Fracture of ceramic andmetalloceramic cylinders. Journal of Prosthetic Dentistry, 52, 221-230.

ÖZCAN M. (1999) Fracture strength of ceramic-fused-to-metal crownsrepaired with two intraoral air-abrasion-techniques and some aspects ofsilane treatment-A laboratory and clinical study (In English). Med Diss,Cologne, Germany.

PALMQVIST, S. & SWARTZ, B. (1993) Artificial crowns and fixed partialdentures 18 years after placement. International Journal ofProsthodontics, 6, 279-285.

RAUSTIA, A.M., NÄPÄNKANGAS, R. & SALONEN, M.A.M. (1998)Complications and primary failures related to fixed metal ceramic bridgeprosthesis made by dental students. Journal of Oral Rehabilitation, 25,677-680.

32


REUTER, J.E. & BROSE, M.O. (1984) Failures in full crown retained dentalbridges. British Dental Journal, 157, 61-63.

SHERILL, C.A. & O`BREIN, W.J. (1974) Transverse strength of aluminous andfeldspathic porcelain. Journal of Dental Research, 53, 683-690.

STRUB, J.R., STIFFLER, S. & SCHÄRER, P. (1988) Causes of failure followingoral rehabilitation: Biological versus technical factors. QuintessenceInternational, 19, 215-222.

TSAI, Y.L., PETSCHE, P.E., ANUSAVICE, K.J. & YANG, M.C. (1998) Influenceof glass-ceramic thickness on hertzian and bulk mechanisms. InternationalJournal of Prosthodontics, 18, 27-32.

WHITE, S., ZHEN, C.L., ZHAOKUN, Y. & KIPNIS, V. (1995) Cyclic mechanicalfatigue of a feldspathic dental porcelain. International Journal ofProsthodontics, 8, 213-220.

WHITE, S.N., LI, Z.C., YU, Z. & KIPNIS, V. (1997) Relationship between static,chemical and cyclic mechanical fatigue in a feldspathic porcelain. DentalMaterials, 13, 103-110.

WIDERHORN, S.M. (1968) Moisture assisted crack growth in ceramics.International Journal of Fracture Mechanism, 4, 171-178.

WIDERHORN, S.M. (1974) Subcritical crack growth in ceramics. In: Fracturemechanics of ceramics. (eds Bradt, D.C., Hasselman, D.P.H., Lange, F.F.),Vol 2., pp. 613. Plenum Press, New York, USA.

YAMAMOTO, M. (1989) In: Metal ceramics, principles and methods of MakotoYamamoto. pp. 447. Quintessence Publishing. Co., Chicago, USA.

33


34


Evaluation of Alternative Intraoral Repair

Techniques for Fractured Ceramic-fused-to-metal

Restorations

Özcan, M.: Evaluation of Alternative Intraoral Repair Techniques for FracturedCeramic-fused-to-metal Restorations. Journal of Oral Rehabilitation,30(2):194-203, 2003. (reproduced with permission of Blackwell Publishing)

35


36


Evaluation of Alternative Intraoral Repair

Techniques for Fractured Ceramic-fused-to-metal

Restorations

MUTLU ÖZCAN

*Marmara University, Dentistry Faculty, Department of Prosthodontics, Istanbul, Turkey

SUMMARY Ceramic fractures are serious and costly problems in dentistry.Moreover, they pose an aesthetic and functional dilemma both for the patientand the dentist. This problem has created demand for the development ofpractical repair options which do not necessitate the removal and remake ofthe entire restoration. Published literature on repair techniques for fracturedfixed partial dentures, concentrating on the data obtained both from in vitroand in vivo studies, reveals that the repair techniques based on sandblastingand silanization are the most durable in terms of adhesive and cohesivefailures compared to those using different etching agents.

Introduction

Despite the increased effort to improve the bond strength between the ceramicand the metal substrate, on occasion, fractures of ceramic veneers still occurunder clinical conditions. Clinical studies indicated that the prevalence ofceramic fractures ranged from between 5-10% over 10 years of use(Coornaert, Adrians & de Boever, 1984).

Although fractures of such restorations do not necessarily mean thefailure of the restoration, the renewal process is both costly and timeconsuming and therefore remains a clinical problem. Fractures in the anteriorregion pose an aesthetic problem but when they are in the posterior region,chewing function could also be affected. The published literature reveals thatthe reasons for failures cover a wide spectrum from iatrogenic causes tolaboratory mistakes, or related to the inherent structure of the ceramics orsimply due to trauma.

It is well recognized that many factors are involved in the success rateassessments of fixed partial dentures limiting the longevity of the restorations.

37


Need for an intra-oral repair technique

Fracture of porcelain is often considered an emergency treatment and therestoration process can present difficult challenges to the dentist. Because ofthe nature of the porcelain processing, new porcelain cannot be added to anexisting restoration intraorally. The manual fabrication of metal frameworks andporcelain veneers is time consuming and requires a high level of skill (Freilichet al., 1998). It is an unpleasant experience for the patient and arduous for thedentist to remove these restorations from the mouth. Replacement of a failedrestoration is not necessarily the most practical solution because of theobviously substantial costs and the complex nature of the restoration (Fan,1991).

Besides some economic and technical reasons, it was reported that thecracks or crazing in the fractured area might become a haven formicroorganisms and plaque accompanied by staining (Walton, Gardner &Agar, 1986). On the basis of previous studies, a consensus was reached thatthe repeated firing cycles cause distortion of the ceramic restorations.Deformation or most of the distortion was found to occur especially during theinitial oxidation of the alloys but small changes from 30 to 99.6 µm were alsoexamined at the margins of the restoration during the subsequent heating andceramic applications (Van Rensburg & Strating, 1984; Richter-Snapp et al.,1988).

Intra-oral repair options provide the possibility of repairing the veneer inthe patient`s mouth preventing replacement of the complete restoration.Aesthetic and functional repair, wherever possible, has many advantages overtime-consuming and expensive remakes of crowns or bridges. Given theseproblems and concerns, it is desirable to repair the fixed restorations in themouth so that the service time can be increased in a more conservativeapproach. Various intraoral repair alternatives for metal-ceramic restorationshave been the subject of numerous studies.

Previous intraoral repair trials

The clinical success of the ceramic repair system is almost entirely dependenton the integrity of the bond between the ceramic and the composite resin. Thisintegrity is achieved either by chemical or mechanical bonds. Many of thepreviously advocated techniques were dependent on mechanical retention butthe results of these earlier repairs were unsatisfactory because of aestheticand mechanical limitations. Various repair techniques have been suggested in

38


the literature, many of which are considered interim but are still preferable asit is important to salvage an extensive restoration for even a few years. 3conditions for the repair of ceramic fractures were suggested (Chung &Hwang, 1997):1. Fracture in ceramic only2. Fracture with both ceramic and metal exposed3. Fracture with substantial metal exposure.

Hydrofluoric acid

Intraoral repair systems based on topical acid application have become verypopular in bonding resin to ceramic. The greatest advantage of these systemsis that chair-side application is very simple. Furthermore the restoration can bere-etched in the case of failure without the need for sophisticated laboratoryprocedures. The most often cited etching agent for the ceramic surface hasbeen hydrofluoric acid.

It has been postulated that acid concentrations and etching times shouldbe adjusted with specific ceramics to optimize bond strength (Calamia &Simonsen, 1984). Furthermore, the bond strength of composite resin toaluminous porcelain was found to be inferior to that of feldspathic porcelain. Inprinciple, chemical etching agents dissolve the glass matrix selectively andcause physical alteration to promote adhesion of composite-resin to theporous surface of fractured ceramic (Calamia et al., 1985; Sheth, Jensen &Tolliver, 1988; Thurmond, Barkmeier & Wilwerding, 1994).

Ceramics etched with hydrofluoric acid demonstrate a microstructure thatappeared most conducive to the development of high strength as a function ofthe number of large porosities within its amorphous surface. Resin penetrationof these spaces enhance micro-mechanical retention (Stangel, Nathanson &Hsu, 1987) and produces greater roughness on the ceramic surface than otheracid agents (Aida, Hayakawa & Mizukawa, 1995).

Alumina content of the ceramic materials plays a significant role on theeffect of hydrofluoric acid. It was stated that reducing the etching time to lessthan three minutes dissolved less of the glass matrix (Tjan & Nemetz, 1988).Sorenson et al. (1991) observed that etching feldspathic porcelain with 20%hydrofluoric acid for 3 min significantly increased its bond strength tocomposite resin. Although many commercially available, porcelains are similarin chemical formula, there are distinct differences in constituents, crystallinestructure, particle size, sintering behaviour and microtopography which effectthe etched surface. Alumina increases the strength of the ceramic but it is

39


highly resistant to chemical attack and therefore does not etch well. Higherbond strength after etching and a high percentage of cohesive failures in Vitaceramics containing 10% alumina has been observed.

Lacy et al. (1988) observed that etching the ceramic surface without usinga silane coupling agent did not provide greater bond strength to the compositeresin than mechanical roughening with a fine diamond bur. Llobell et al. (1992)found significantly higher bond strengths with hydrofluoric acid compared withphosphoric acid and advised use of hydrofluoric acid for mechanical retentionand silane coupling agents for chemical retention. While some studies showedenhanced bond strength with the application of silane to the etched ceramicsurface (Lacy et al. 1988), others exhibited significant variation in bondstrengths between proprietary brands of silane. On the other hand, especiallyafter hydrofluoric acid treatment, the use of silane coupling solutions promotedgood results (O`Kray, Suchak & Stanford, 1987; Nicholls, 1988; Bailey, 1989).

From a clinical point of view, hydrofluoric acid application alone wasconsidered inadequate when preparing a ceramic surface for composite resinbonding (Pameijer, Louw & Fischer 1996). Matsumara et al. (1989) concludedthat acid treatment might only be useful, in practice, to remove the smearsfrom the ceramic. In another study, increased incidence of cohesive failureswere observed in samples pretreated with 9.5% hydrofluoric acid due to deepacid penetration but 5 minutes of hydrofluoric acid application to be too long(Wolf, Powers & O`Keefe, 1992). Durability of bonding between compositeresin and ceramic formed with chemical agents was markedly inferior toalteration of the ceramic surface with either aluminum oxide air abrasion,hydrofluoric acid or a combination of both (Thurmond, Barkmeier &Wilwerding, 1994).

Although new chemical etching systems claimed to provide adequateretention, the study by Tylka & Stewart (1994) indicated that these chemicaletchants unfortunately produce a shallower etch pattern on metal. They alsoreported that even though an optimal bond could be achieved with eitheretchant or in conjuction with an organosilane, the intraoral use of dangeroushydrofluoric acid should be seriously questioned.

The hazards of hydrofluoric acid are well recognized. Despite itseffectiveness, hydrofluoric acid presents severe hazards to human tissue andadvised more reasonable repair alternatives (Chung & Hwang, 1997).Practitioners were warned, indicating that the problem is particularly acutewhen adequate rubber dam isolation is not possible, such as repair cases offixed partial dentures where a tight cervical seal cannot be attained.

There has been only one clinical study conducted using etching gel for therepair process (Creugers, Snoek & Käyser, 1992). In this study, in order to

40


study the effect of 37% phosphoric acid application, the surfaces of 20 ceramiccrowns were fractured on purpose. 12 of them included metal exposure, and 8of them had fractures with no metal exposure. Crowns were cemented and thepatients were recalled at 2 weeks, 3, 6 and 12 months after the repair. Thefailure rate was found to be 50% after 12 months. Failures were mostlyobserved at the bonding interface between the crown and the repair resin withno cohesive failures. The survival rate was noted to be 59% at the end of 12months of the evaluation period. Because of the low survival rate, this methodwas not recommended for use, especially in occlusal repair of metal-ceramiccrowns.

All though hydrofluoric acid is considered to be a dangerous, harmful, anirritating compound and categorized as a poisonous reagent (Llobell et al.1992), both laboratory evaluations and clinical procedures concerning its usefor intraoral porcelain repair have been reported. Etching with hydrofluoric acidmay not be practicable due to the biological risks in vivo. It still seems intraoralrepair options with acid agents are effective on an interim basis. Moreover, acidetching is a method which could be used in ceramic fractures with no metalexposure.

The studies on the use of hydrofluoric acid have significant findings.Concentration of the acid and the application period are apparently importantfactors to note. Considering the vast range of ceramics in today`s dentalpractice, the choice of suitable acid etching process clearly needs furtherresearch in order to avoid misleading information for the practitioners.

Acidulated Phosphate Fluoride

The hazards, extreme caustic effects to soft tissues and the danger for clinicaluse of hydrofluoric acids are well known. For this reason some studiesquestioned whether 1.23% acidulated phosphate fluoride gels might serve asa safe and effective substitute for etching ceramic surfaces to bond compositeresin because of the reduced risk it presents. Some studies demonstrated thatthe bond strength of composite resin to silanized ceramic after being etchedby acidulated phosphate fluoride was comparable to that of hydrofluoric acidetching (Sposetti, Shen & Levin, 1986; Wunderich & Yaman, 1986; Abbasi etal., 1988).

Lacy et al. (1988) reported that ceramic surfaces could be etched with1.23% acidulated phosphate fluoride gels in relatively short periods of time. Itwas concluded that 1.23% acidulated phosphate fluoride gels can besubstituted for 9.5% hydrofluoric gels as prolonged etching times were

41


required with the lower concentrations of hydrofluoric acid.Remarkable differences in the etched ceramic surface morphology were

observed in visual comparisons. Application1.23% acidulated phosphatefluoride gel was found to create smooth, homogenous surfaces on theexposed ceramic, whereas hydrofluoric acid produced a porous, amorphoussurface. The widely accepted theory that hydrofluoric acid enhances thecomposite resin bond to ceramic more than an acidulated phosphate fluoridewas not substantiated (Senda, Suzuki & Jordan, 1989; Tylka & Stewart, 1994).The SEM findings showed that etching by acidulated phosphate fluoride gelmight not be adequate (Nelson & Barghi, 1989).

No significant difference was found between the tensile bond strengths forspecimens etched with 9.6% hydrofluoric acid and those of specimens etchedwith 4% acidulated phosphate fluoride gel in the data obtained by Della Bona& van Noort (1995). However, the group etched with 4% acidulated phosphatefluoride gel, showed a wider statistical spread than the one etched with 9.6%hydrofluoric acid. This suggested that hydrofluoric acid etching might wellproduce a more reliable and consistent result but this has not been confirmedsince the sample size was too small.

This literature review led to the conclusion that intraoral use of acid agentsappears to be unwarranted.

Micromechanical roughening

Some practitioners have relied on mechanical retention such as grooves orundercuts to retain the composite resin to ceramic or metal. Owing tomicroleakage and humid intraoral conditions, this type of repair wasconsidered as an interim procedure. It was reported that the use of fine andcoarse diamond burs increases crack initiation and propagation through theceramic which could result in failure (Wood et al., 1992). These trials did notgive long lasting, predictable results in ceramic repair.

Air abrasion with Al2O3

One easy method for intraoral repair is roughening the surface by air abrasionwith Al2O3, thereby increasing the surface area for bonding and decreasingthe surface tension. This technique was based on direct sandblasting of thesurfaces by an intra-oral device. Air abrasion (or sandblasting) promotesmicromechanical retention. Physical alteration of the ceramic surface with

42


Al2O3 was mostly achieved using a particle size of 50 µm. Air abrasionimproves the retention between the metal and resin by cleaning oxides or anygreasy materials from metal surfaces, creating very fine roughness enhancingmechanical and chemical bonding between some resins and metals. WhenAl2O3 treatment was performed on the alloy, microscopically cleaned androughened surfaces were observed which allowed efficient wetting by resinsand stronger composite-alloy bonds (Schneider, Powers & Pierpoint, 1992).

Higher bond values with Al2O3 were obtained than those with typicalsilane application on etched ceramic surface and advised its use in lieu offluoride etching (Lacy et al., 1988).

A variety of treatment regimens including medium diamond bur, airabrasion with 50 µm Al2O3, hydrofluoric acid, phosphoric acid, silane andbonding agent were compared. The shear test results revealed that the mostdurable bond values were obtained with physical alteration of the ceramicusing Al2O3 air abrasion followed by hydrofluoric acid (Thurmond, Barkmeier& Wilwerding, 1994).

Sandblasting was described as the most effective surface treatment forthe fractured metal-ceramic restorations no matter whether the surface wassimplified with metal, porcelain, or a combination of the two. Sufficient bondstrength was obtained with Al2O3, eliminating the use of caustic andpotentially harmful acid agents (Chung & Hwang, 1997). However thecompulsory use of silane together with Al2O3 was advised in order to avoidchanges in retention (Shahverdi et al., 1998).

Combined data from the literature revealed that sandblasting with Al2O3,is an effective surface treatment regardless of whether the fracture was metal,porcelain, or a combined exposure. It was also stressed that air abrasion doesnot expose patients to the risk of severe acid burns. Controversial reports onthe effect of whether Al2O3 should be used alone, followed by silaneapplication or together with hydrofluoric acid, needs to be identified.Furthermore, concerns on the mechanism of each treatment regimen shouldalso be clarified.

Combined surface treatments

Some trials combined the above-mentioned methods in order to obtain betterbond strengths.

Combined use of silane with hydrofluoric acid or air abrasiondemonstrated better results with Al2O3 air abrasion than those with etchedceramic surfaces (Bertolotti, Lacy & Watanabe, 1989). Llobell et al. (l992)

43


observed that silane and hydrofluoric acid combinations did not affect the bondstrengths positively.

Various surface treatments including air abrasion with Al2O3 of 50 µm,roughening with a diamond, etching with 9.6% hydrofluoric acid and acombination of the latter two methods were evaluated (Suliman, Swift &Perdigao, 1993). Shear tests revealed that the most effective surface treatmentcombinations were: mechanical roughening with diamond burs and thenchemical etching with hydrofluoric. In another study, it was advised to acidifythe surface with 32% phosphoric acid in combination with Al2O3 air abrasionor roughen with a diamond instrument to alter the ceramic surface. It was alsofound that the durability of bonds between composite and ceramic formed withchemical agents was markedly inferior to alteration of the ceramic surface witheither Al2O3 air abrasion and hydrofluoric acid or a combination of both(Thurmond et al., 1994).

Castellani et al. (1994) roughened the exposed metal and ceramicsurfaces with a diamond bur and created mechanically retentive areas on themetal surface. The best results were observed with the use of 50 µm Al2O3sandblasting on the etched surface of the metal. Pameijer, Louw & Fischer(1996) obtained the best results in their study with the combined use ofsandblasting and hydrofluoric acid application. Shahverdi et al. (1998) foundthat the combination of chemical and mechanical retention techniques seempromising for improved bond strength. In their study, the samples treated firstwith air abrasion, then with hydrofluoric acid and silane exhibited the highestshear bond values compared to those of the air abraded and silanized orhydrofluoric acid etched and silanized groups.

Although the data appear to document the efficacy of air abrasion, itappears that optimum protocol for the treatment of either ceramic or metalusing these methods is yet to be defined.

Air abrasion with SiOx

Although satisfactory bonding between ceramic and metal is achieved incurrent dental practice, many attempts have been made to develop bettertechniques for bonding composite resin materials to dental alloys. The natureof the metal-resin junction is critical; therefore, the strength of the bondingsystem, its resistance to microleakage, and the minimum space required forthe system are very important. As an alternative to the conventionalmechanical retention systems, chemical retention systems aim to develop abond between metal and resin. This has led to the development of various

44


surface conditioning techniques.Guggenberger (1989) introduced the Rocatec® System*, which

presented a new kind of acrylic-metal bonding system. The principle istribochemical application of a silica layer by means of sandblasting. Accordingto the extraoral use of the Rocatec® System, samples are blasted with 110 µmgrain size aluminum oxide particles modified with silicic acid, so called,Rocatec® Plus*. The blasting pressure results in the embedding of silicaparticles on the metal surface rendering the surface chemically more reactiveto resin via silane. The Rocatec® System was proclaimed to be a novelacrylic/metal bonding system. Shear, compression and tensile tests revealedincreased bond strength values with this system compared to those obtainedfrom mechanical bead retention, even after thermocycling and storage in waterfor one year.

Edelhoff & Marx (1995) conducted a study in which different surfaceconditioning methods were used for ceramic surfaces including diamondroughening, sandblasting, silica coating, and acid etching. The resultsobtained by silica coating showed significantly higher bond strengths of resinon ceramic surfaces compared with other systems. Best results were obtainedwhen the nozzle of the intraoral sandblaster was held perpendicular to thesurface at a distance of approximately 10 mm. Depending on the size of thefracture, it was advised that the surface be sandblasted for approximately 13 s(Proano et al., 1998).

In another study which was performed on disc samples, removing thedebris layer with SiOx of 30 µm particle size resulted in higher bond strengthsof resins to ceramic surfaces with no metal exposures. Mostly cohesive failureswere observed and use of particles of 110 µm grain size was found todecrease the bond strengths compared to the etching technique after 24 h ofwater storage at 37�C (Sindel, Gehrlicher & Petschelt, 1996). The sameresearch group compared 5% hydrofluoric acid etching with use of SiOx of 30and 110 µm particle size. In that study, 30µm silica coating showedsignificantly higher bond strength values with cohesive failure modes thanthose obtained with acid etching after 24 h of storage in distilled water withoutthermocycling (Sindel, Gehrlicher & Petschelt, 1997). This study hassignificant findings but it could be criticized on the grounds that storage periodwas too short.

In a subsequent study, bond strengths using two different coatingmethods were evaluated. After storage in distilled water at 37�C for 30, 90,150, 360 days without thermocycling, the test samples were subjected totensile loads until they fractured. Significant differences in bond strength wereobtained especially after an interval of 360-day-period. The tensile bond

45* ESPE, Seefeld, Germany


strength for the intraoral silica coating technique using SiOx of 110 µm grainsize showed better results than that of 30 µm SiOx and Al2O3 after 60 sec ofapplication on NiCr alloys (Edelhoff, Marx & Spiekermann, 1998).The outcomeof this study is in contrast with the findings of Sindel et al. (1997).

Some aspects of silane pre-treatment

The system of bonding composite resin to dental porcelain using silanesolutions produced reliable bonds. It was thought to be an effective method forintraoral repair of fractured or chipped ceramic restorations. However, thismethod, reported in the 1970s by Newburg & Pameijer, suffered fromdifficulties at first because of the instability of the silane solutions used toprepare the ceramic surface. Silane coupling agents have been steadilyimproved, producing higher bond strengths. For an effective bond of resin tofeldspathic porcelain and metal, the use of silane in combination with a surfacetreatment is compulsory. Silane promotes adhesion between the fracturedceramic and the repair resin. Recent advances in silane coupling agentsappear to enhance bond strength by promoting a chemical bond between thecomposite resin and the porcelain (Calamia et al., 1985; Tjan & Nemetz, 1988;Hayakawa et al., 1992; Mueller, Olsson & Söderholm, 1997).

Eames et al. (1977) evaluated various organosilanes to establish theirbonding to ceramic or metal and observed that they did not bond to the metalsurface as they had with the ceramic. In other studies, silane coupling agentswere found to improve the bonding of composite resin to ceramic byapproximately 25%. These studies demonstrated the use of silane or itsdegraded solutions to be completely ineffective when used on a glazedceramic surface (Newburg & Pameijer, 1978; Diaz-Arnold, Schneider &Aquilino, 1989; Lacy et al., 1988).

Rapid increase in the amount of water absorbed by the compositematerial causes hydrolysis and degradation of the silane. Water storage andthermocycling were described as detrimental for the silane-ceramic bond(Roulet, 1987). Reuter & Brose (1987) reported that silanized interfacesappear to be unstable in humid conditions and the silane bond was found todeteriorate under atmospheric moisture. Since the resins are permeable towater, the bond between silane and composite resin was expected todeteriorate by hydrolysis over time. It was concluded that in humid conditionsthis may lead to stress corrosion and subcritical crack growth.

In other studies, it was indicated that the use of silane is a must butdifferent composite systems yield different values. It was noted that there is

46


little information on the bond strength between organosilane and ceramicrepair materials (O’Kray, Suchak & Stanford, 1987; Bailey, 1989).

The use of the Rocatec® System (SiOx) increased the bond strengtheffectively because of the increase in silica content, which provided a basis forthe silanes to enhance the bonding with the resin. For a better clinical success,Guggenberger (1989) advised the use of silane coupling agents as crucialingredients in creating long-term bonds of resin to ceramic or metal.

A study conducted by Shahverdi et al. (1998) concluded that althoughsilane coupling agents are capable of forming bonds with both inorganic andorganic surfaces, silane itself was not found to help in bonding. Therefore, itsuse in combination with silica coating was recommended. In this study, in thecases where silane was not used, the bond strengths were less after waterstorage for 30 days. The data showed that when only silane was applied on theceramic surfaces, the bond strength did not improve because of insufficientmechanical retention.

Studies indicated that silane coupling agents are important in theadhesion of composite resin to ceramic. The main contribution to the obtainedvalues was made, not by the mechanical interlocking of the composite resin,but by the formation of siloxane bonds via silane (Söderholm et al., 1984).

The implication found in these studies was that silane coupling agentsimproves wettability and contributes to covalent bond formation between theceramic and the resin composite. Literature supports silanization of ceramics,which provides a more reliable bond than etching with hydrofluoric acid onlybut little is known about the hydrolytic stability of the silanes especially inhumid conditions.

Repair composite-resins

Composite resins are commonly used for the repair of ceramic fractures. Ifthere is a small part missing, composite resins of appropriate shade have beenthe material of choice for aesthetic appearance and ease of manipulation.

In order to withstand the functional loads, the bond between the repairmaterial and the restoration must be sufficiently strong. The repair materialwhich ensures this bond should have a minimal coefficient of thermalexpansion and minimal polymerization shrinkage. The type of composite resinalso affects its bond strength to ceramic. Larger particle size composite resinsor hybrid type resins at the ceramic interface result in higher bond strengththan those of microfilled composite resins (Gregory & Moss, 1990). For repairpurposes, use of the hybrid composite resins was advised as the most suitable

47


ones (Lutz & Phillips, 1983). Bond strengths are also dependent on the type ofthe composite resin used. Hybrid composite resin was found to increasestrength and decrease stress compared with a microfilled one (Simonsen &Calamia, 1983; Stangel, Nathanson & Hsu, 1987). The problems of wear andsurface changes are not related to the repair system but to the use of themicrofilled composite resin which could be minimized if a hybrid compositeresin is used. It is also recommended to be used where fatigue loading is ofconsideration (Creugers, Snoek & Käyser, 1992; Llobell et al., 1992).

A large number of studies investigated the effect of surface treatmentregimens on the bond strength of composite resins to ceramic surfaces. Thedata from these studies should be interpreted cautiously as the type of therepair resins used in these studies exhibit different structures.

Effect of thermocycling

The durability of the bond values under the stresses of the oral environment isimportant for clinical predictability of dental materials. Usually, dental materialsare subject to mechanical, thermal, and chemical stresses in the mouth duringoral functions. Thermocycling and water storage in vitro is a common way oftesting dental materials to establish their suitability for in vivo use. Exposing thespecimens to thermocycling speeds up the diffusion of water in between thecomposite resin and the metal or ceramic. Changing the temperature createsstress at the interface of the two materials because of different coefficients ofthermal expansion. Most of the studies with repair process involved differentthermocycling times but the common consensus was that the thermocyclingdecreased the bond strength as it weakens the resin structure (Cochran et al.1988).

Water storage and thermocycling are detrimental to the silane-ceramicbond as well. However, it was not clarified whether the silane was broken downby the water storage or thermocycling (Cochran et al., 1988; Pratt et al., 1989).

With the use of silica coating, Peutzfeld & Asmussen (1988) found nostatistical decline in the adhesive strength from the initial bonding resultsobtained after 20 h of water storage at 36�C plus 6 h thermocycling repeated180 times between 15-70�C and those after one year water storage andrepeated 900 times thermocycling between 15�C to 70�C. However in anotherstudy, thermocycling caused decreased bond strength values for samplessandblasted with 50 µm Al2O3 (Wolf, Powers & O`Keefe, 1992).

A comparative study was performed by Kern & Thompson (1993) between5 different resin-bonding systems to cobalt-chromium alloys. The samples

48


were stored in artificial saliva for 150 days at 37�C and every second day theywere subjected to 1000 thermocycles in a temperature range of 5 to 55�C for75000 cycles. Samples were tested after 24 h, 10, 30, 90 days and after 150days of water storage. The results indicated that, in contrast to themicromechanical bonding systems, silica coating showed no significantchange in the tensile bond strength during this observation period. The systemwas recommended as suitable for cobalt-chromium alloys used in resin-bonded restorations.

The relevance of the studies in which thermocycling was applied for ashorter period of time should be questioned. There seems to be a lack ofagreement that water storage and thermocycling have decreasing effects onthe resin-ceramic bond. The main reason for this could be attributed to variousthermocycling times in the experiments.

Conclusion

Successful intraoral repair of fixed partial dentures has been a great problemespecially when the metal substructure is exposed (Chung & Hwang, 1997).

From the previously introduced intraoral repair techniques, organosilanecoupling agents are not able to bond to metal surfaces as they do to dentalceramics (Bailey 1989). Hydrofluoric acid and acidulated phosphate fluoridefacilitate micromechanical retention but these chemical agents are notapplicable to the fractures where metal is exposed and they are alsohazardous to soft tissues. Mechanical roughening of the metal or ceramic withfine and coarse diamond burs however, are reported to provoke crack initiationand propagation through the ceramic. Both experimental and clinical reportsprovided evidence of significant differences between the repair techniques butthe results were not uniform and therefore they were considered to be interimprocedures.

For the repaired restoration to withstand functional loads, the bondbetween the repair resin and the remaining restoration must be strong anddurable. Recently, the advantages of extraoral silica coating (tribochemicalcoating) using the Rocatec® System were combined with the practical use ofan intraoral sandblaster in order to get a better bond strength in repairingfractured veneers in vivo. Although the intraoral sandblasters had already beendesigned to be used with Al2O3, because of their superior advantages, SiOx(aluminium oxide coated SiOx particles) was used instead of Al2O3, togetherwith silane application (Proano et al., 1998, Özcan 1999).

At present, the minimum bond strength for retention of an adhesive to a

49


metal-ceramic restoration in the oral environment is not known. Maximum biteforce capability of each patient, the estimated biting force on specific teeth, thepresence or absence of surface damage may reduce the success rate. Thereis however, insufficient clinical data available at this time to predict the clinicalperformance from in vitro studies and the performance of ceramic repairs invivo (Özcan, Schulz & Niedermeier, 1999).

From the available literature, it could be interpreted that the innovative airabrasion technique with SiOx, recently called the CoJet®-System*, does notexpose the patients to the risk of severe acid burns with the advantage ofrepairing fractures with both ceramic and metal exposure. Owing to theincreasing number of composite resin materials on the market, it is still noteasy to choose the best one. When the composites are used in the anteriorregion, more aesthetic expectations should be fulfilled and the clinician mustmeet both aesthetic and functional challenges. They should behave similarly todentin and enamel with respect to the properties of reflection, refraction,scattering and transmission of light to give the illusion of natural teeth.

Before any attempt at a repair, the underlying metal substructure shouldfirst be found to be sound and that it is not the real cause of the failure. If thisis the reason, instead of attempting the repair process, the restoration shouldbe renewed.

When evaluating the current literature on ceramic repair techniques, thevariables of composite resin, storage conditions and silane application shoudbe taken into consideration.

References

ABBASI, J., BERTOLOTTI, R.L., LACY, A.M. & WATANABE, L.G. (1988) Bondstrengths of porcelain repair monomers. Journal of Dental Research, 67,223.

AIDA, M., HAYAKAWA, T. & MIZUKAWA, K. (1995) Adhesion of composite toporcelain with various surface conditions. Journal of Prosthetic Dentistry,73, 464-470.

BAILEY, J.H. (1989) Porcelain-to-composite bond strengths using fourorganosilane materials. Journal of Prosthetic Dentistry, 61, 174-177.

BERTOLOTTI, R.L., LACY, A. & WATANABE, L.G. (1989) Adhesive monomersfor porcelain repair. International Journal of Prosthodontics, 2, 483-489.

CALAMIA, J.R. & SIMONSEN, R.J. (1984) Effect of coupling agents on bondstrength of etched ceramics. Journal of Dental Research, 63, 179.

CALAMIA, J.R., VAIDYANATHAN, J., VAIDYANATHAN, T.K. & HIRSCH, S.M.

50


(1985) Effects of coupling agents on bond strength of etched porcelain.Journal of Dental Research, 64, 296.

CASTELLANI, D., BACCETTI, T., CLAUSER, C. & BERNARDINI, U.D. (1994)Thermal distortion of different materials in crown construction. Journal ofProsthetic Dentistry, 72(4), 360-366.

COCHRAN, M., CARLSON, T., MOORE, B., RICHMOND, N. & BRACKETT, W.(1988) Tensile bond strengths of five porcelain repair systems. OperativeDentistry, 13, 112-117.

CHUNG, K.H. & HWANG, Y.C. (1997) Bonding strengths of porcelain repairsystems with various surface treatments. Journal of Prosthetic Dentistry,78, 267-274.

COORNAERT, J., ADRIANS, P. & de BOEVER, J. (1984) Long-term clinicalstudy of porcelain-fused-to-gold restorations. Journal of ProstheticDentistry, 51, 338-342.

CREUGERS, N., SNOEK, P. & KÄYSER, A.F. (1992) An experimentalporcelain repair system evaluated under controlled clinical conditions.Journal of Prosthetic Dentistry, 68, 724-727.

DELLA BONA, A. & VAN NOORT, R. (1995) Shear vs. tensile bond strength ofresin composite bonded to ceramic. Journal of Dental Research, 74, 1591-1596.

DIAZ-ARNOLD, A.M., SCHNEIDER, R. L. & AQUILINO, S.A. (1989) Bondstrengths of intraoral porcelain repair materials. Journal of ProstheticDentistry, 61, 305-309.

EAMES, W.B., ROGERS, L.B., FELLER, P.R. & PRINCE, W.R. (1977) Bondingagents for repairing porcelain and gold: An evaluation. Operative Dentistry,2, 118-124.

EDELHOFF, D. & MARX, R. (1995) Adhäsion zwischen Vollkeramik undBefestigungskomposit nach unterschiedlicher Oberflächenvorbehandlung.Deutsche Zahnärztliche Zeitschrift, 50, 112-117.

EDELHOFF, D., MARX, R. & SPIEKERMANN, H. (1998) ReparaturAbgeplatzter Verblendungen durch Intraorale Silicatisierung- Eine In vitro-Untersuchung. Deutsche Zahnärztliche Zeitschrift, 53, 115-119.

FAN, P.L. (1991) Porcelain repair materials. Council on dental materials,instrument and equipment prepared at the request of the council. Journalof American Dental Association, 122, 124-130.

FREILICH, M.A., KARMAKER, A.C., BURSTONE, C.J. & GOLDBERG, J.(1998) Development and clinical applications of a light-polymerized fibrereinforced composite. Journal of Prosthetic Dentistry, 80, 311-318.

51


GREGORY, W.A. & MOSS, S.M. (1990) Effects of heterogeneous layers ofcomposite and time on composite repair of porcelain. Operative Dentistry,15, 18-22.

GUGGENBERGER, R. (1989) Das Rocatec-System-Haftung durchtribochemische Beschichtung. Deutsche Zahnärztliche Zeitschrift, 44, 874-876.

HAYAKAWA, T., HORIE, K., AIDA, M., KANAVA, H., KOBAYASHI, T. &MURATA, Y. (1992) The influence of surface conditions and silane agentson the bond of resin to dental porcelain. Dental Materials, 8, 238-240.

KERN, M. & THOMPSON, V.P. (1993) Sandblasting and silica coating of dentalalloys: Volume loss, morphology and changes in surface composition.Dental Materials, 9, 155-161.

LACY, A.M., LALUZ, J., WATANABE, L.G. & DELLINGES, M. (1988) Effect ofporcelain treatments on the bond strength of composite. Journal ofProsthetic Dentistry, 60, 288-291.

LLOBELL, A., NICHOLLS, J.I., KOIS, J.C. & DALY, C.H. (1992) Fatigue life ofporcelain repair systems. International Journal of Prosthodontics, 5, 205-213.

LUTZ, F. & PHILLIPS, R.W. (1983) A classification and evaluation of compositeresin systems. Journal of Prosthetic Dentistry, 50, 480-488.

MATSUMARA, H., KAWAHARA, M., TANAKA, T. & ATSUTA, M. (1989) A newporcelain repair system with silane coupler, ferric chloride, and adhesiveopaque resin. Journal of Dental Research, 68, 813-818.

MUELLER, H., OLSSON, S. & SÖDERHOLM, K.J. (1997) The effect ofcomonomer, composition, silane heating and filler type on aqueousTEGDMA leachability in model resin composites. European Journal of OralSciences, 105, 363-368.

NELSON, E. & BARGHI, N. (1989) Effect of APF etching time on resin bondedporcelain. Journal of Dental Research, 68, 271.

NEWBURG, R. & PAMEIJER, C.H. (1978) Composite resin bonded toporcelain with silane solution. Journal of American Dental Association, 96,288-293.

NICHOLLS, J.I. (1988) Tensile bond to resin cements to porcelain veneers.Journal of Prosthetic Dentistry, 60, 443-447.

O`KRAY, K., SUCHAK, A.J. & STANFORD, J.W. (1987) Shear strength ofporcelain repair materials. Journal of Dental Research, 66, 207.

ÖZCAN M. (1999) Fracture strength of ceramic-fused-to-metal crownsrepaired with two intraoral air-abrasion-techniques and some aspects ofsilane treatment-A laboratory and clinical study (In English). Med Diss,Cologne, Germany.

52


ÖZCAN, M., SCHULZ, A. & NIEDERMEIER, W. (1999) Clinical outcomes onfractured ceramic-fused-to-metal restorations repaired with an intraoralsilicacoating method. Journal of Dental Research, 78, 302.

PAMEIJER, C.H., LOUW, N.P & FISCHER, D. (1996) Repairing fracturedporcelain: How surface preparation affects shear force resistance? Journalof American Dental Association, 127, 203-207.

PEUTZFELDT, A. & ASMUSSEN, E. (1988) Silicoating. Evaluation of a newmethod of bonding composite resin to metal. Scandinavian Journal ofDental Research, 96, 171-177.

PLEUDDEMANN, E.P. (1982) Nature of adhesion through silane couplingagents. pp. 111 Plenum Press, New York, USA.

PRATT, R.C., BURGESS, J.O., SCHWARTZ, R.S. & SMITH, J.H. (1989)Evaluation of bond strength of six porcelain repair systems. Journal ofProsthetic Dentistry, 62, 11-13.

PROANO, P., PFEIFFER, P., NERGIZ, I. & NIEDERMEIER, W. (1998) Shearbond strength of repair resin using an intraoral tribochemical coating onceramometal, ceramic and resin surfaces. Journal of Adhesion Scienceand Technologies, 12, 1121-1130.

REUTER, J.E. & BROSE, M.O. (1984) Failures in full crown retained dentalbridges. British Dental Journal, 157, 61-63.

RICHTER-SNAPP, K., AQUILINO, S.A., SVARE, C.W. & TURNER, K.A. (1988)Change in marginal fit as related to margin design, alloy type, andporcelain proximity in porcelain-fused-to-metal restorations. Journal ofProsthetic Dentistry, 60, 435-439.

ROULET, J.F. (1987) In: Degradation of Dental Polymers. pp. 60 Karger, Basel,Germany.

SCHNEIDER, W., POWERS, J.M. & PIERPOINT, H.P. (1992) Bond strength ofcomposites to etched and silica coated porcelain fusing alloys. DentalMaterials, 8, 211-215.

SENDA, A., SUZUKI, M. & JORDAN, R.E. (1989) The effect of fluorides andHF acids on porcelain surfaces. Journal of Dental Research, 68, 236.

SHAHVERDI, S., CANAY, S., SAHIN, E. & BILGE, A. (1998) Effects of differentsurface treatment methods on the bond strength of composite resin toporcelain. Journal of Oral Rehabilitation, 25, 699-705.

SHETH, J., JENSEN, M. & TOLLIVER, D. (1988) Effect of surface treatment onetched porcelain bond. Dental Materials, 4, 328-337.

SIMONSEN, R.J. & CALAMIA, J.R. (1983) Tensile bond strength of etchedporcelain. Journal of Dental Research, 62, 297.

53


SINDEL, J., GEHRLICHER, S. & PETSCHELT, A. (1996) Untersuchungen zurHaftung von Komposit an VMK-Keramik. Deutsche ZahnärztlicheZeitschrift, 51, 712-718.

SINDEL, J., GEHRLICHER, S. & PETSCHELT, A. (1997) Haftung vonKomposit an VMK-keramik bei freiliegendem Metallgerüst. DeutscheZahnärztliche Zeitschrift, 52, 193-198.

SORENSON, J.A., ENGELMAN, M.J., TORRES, T.J. & AVERA, S.P. (1991)Shear bond strength of composite resin to porcelain. International Journalof Prosthodontics, 4, 17-23.

SÖDERHOLM, K.J., ZIGAN, M., RAGAN, M., FISCHLSCHWEIGER, W. &BERGMAN, M. (1984) Hydrofluoric degradation of dental composites.Journal of Dental Research, 63, 1248-1254.

SPOSETTI, V., SHEN, C. & LEVIN, A. (1986) The effect of topical fluorideapplication on porcelain restorations. Journal of Prosthetic Dentistry, 55,677-682.

STANGEL, I., NATHANSON, D. & HSU, C.S. (1987) Shear strength of thecomposite bond to etched porcelain. Journal of Dental Research, 66,1460-1465.

SULIMAN, A.A., SWIFT, E. & PERDIGAO, J. (1993) Effects of surfacetreatments and bonding agents on bond strength of composite resin toporcelain. Journal of Prosthetic Dentistry, 70, 118-120.

TJAN, A.H. & NEMETZ, H. (1988) A comparison of the shear bond strengthbetween two composite resins and two etched ceramic materials.International Journal of Prosthodontics, 1, 73-79.

TYLKA, D.F. & STEWART, G.P. (1994) Comparison of an APF and HF forporcelain-composite repair. Journal of Prosthetic Dentistry, 72, 121-127.

THURMOND, J.W., BARKMEIER, W. & WILWERDING, T.M. (1994) Effect ofporcelain surface treatments on bond strengths of composite resin bondedto porcelain. Journal of Prosthetic Dentistry, 72, 355-359.

VAN RENSBURG, F. & STRATING, H. (1984) Evaluation of the marginalintegrity of ceramometal restorations: Part II. Journal of ProstheticDentistry, 52(2), 210-214.

WALTON, J.N., GARDNER, F.M. & AGAR, J.R. (1986) A survey of fixed partialdenture failures: Length of service and reasons for replacement. Journal ofProsthetic Dentistry, 56, 416-421.

WOLF, D.M., POWERS, J.M. & O`KEEFE, K.L. (1992) Bond strength ofcomposite to porcelain treated with new porcelain repair agents. DentalMaterials, 8, 158-161.

54


WOOD, M., LITKOWSKI, L.J., THOMPSON, V.P. & CHURCH, T. (1992) Repairof porcelain/metal restoration with resin bonded overcasting. Journal ofEsthetic Dentistry, 4, 110-117.

WUNDERICH, R.C. & YAMAN, P. (1986) In vitro effect of topical fluoride ondental porcelain. Journal of Prosthetic Dentistry, 55, 385-388.

55


56


Chapter Effect of Surface Conditioning Methods on the

Bond Strength of Luting Cement to Ceramics

This chapter has been published as:Özcan, M., Vallittu, P.K.: Effect of Surface Conditioning Methods on the BondStrength of Luting Cement to Ceramics. Dental Materials, 19(8):725-732,2003. (reproduced with permission of Elsevier Science Ltd.)

57

2


58


Effect of Surface Conditioning Methods on the

Bond Strength of Luting Cement to Ceramics

Mutlu Özcana, Pekka K. Vallittub

aUniversity of Groningen, Faculty of Medical Sciences,

Department of Dentistry and Oral Hygiene, Groningen, The NetherlandsbUniversity of Turku, Institute of Dentistry,

Department of Prosthodontics and Biomaterials Research, Turku, Finland

Abstract

Objectives: This study evaluated the effect of three different surfaceconditioning methods on the bond strength of a Bis-GMA based luting cementto six commercial dental ceramics.Methods: Six disc shaped ceramic specimens (glass ceramics, glassinfiltrated alumina, glass infiltrated zirconium dioxide reinforced alumina) wereused for each test group yielding a total number of 216 specimens. Thespecimens in each group were randomly assigned to one of the each followingtreatment conditions: (1) Hydrofluoric acid etching, (2) Airborne particleabrasion, (3) Tribochemical silica coating. The resin composite luting cementwas bonded to the conditioned and silanized ceramics using polyethylenemolds. All specimens were tested at dry and thermocycled (6.000, 5ºC-55ºC,30 s) conditions. The shear bond strength of luting cement to ceramics wasmeasured in a universal testing machine (1 mm/min).Results: In dry conditions, acid etched glass ceramics exhibited significantlyhigher results (26.4-29.4 MPa) than those of glass infiltrated alumina ceramics(5.3-18.1 MPa) or zirconium dioxide (8.1 MPa) (ANOVA, P < 0.001). Silicacoating with silanization increased the bond strength significantly for high-alumina ceramics (8.5-21.8 MPa) and glass infiltrated zirconium dioxideceramic (17.4 MPa) compared to that of airborne particle abrasion (ANOVA, P< 0.001). Thermocycling decreased the bond strengths significantly after all ofthe conditioning methods tested.Significance: Bond strengths of the luting cement tested on the dentalceramics following surface conditioning methods varied in accordance with theceramic types. Hydrofluoric acid gel was effective mostly on the ceramicshaving glassy matrix in their structures. Roughening the ceramic surfaces withair particle abrasion provided higher bond strengths for high-alumina ceramicsand the values increased more significantly after silica coating/silanization.

59


Keywords: Surface conditioning; Ceramics; Acid etching; Silica coating;Zirconia ceramics; Alumina ceramics

1. Introduction

Numerous attempts have been made to develop ceramic systems thateliminate metal infrastructures and provide optimal distribution of reflectedlight. Currently clinicians have an increasing range of ceramics capable ofdelivering high quality aesthetic restorations to choose from for many clinicalindications. New ceramic systems involve reinforced ceramic cores throughdispersion with leucite [1-6], glass infiltration into sintered alumina (Al2O3)[7,8], the use of high-purity alumina [9] or zirconium dioxide (zirconia, ZrO2)[10].

To enhance the bond strength of luting cement to the ceramic surface, anumber of techniques have been reported which mechanically facilitate resin-ceramic bonding. Etching the inner surface of a restoration with hydrofluoricacid followed by the application of a silane coupling agent is a well-known andrecommended method to increase bond strength. Although hydrofluoric acid isefficient in roughening feldspathic ceramic for bonding composite resin [11-16], neither etching with these solutions nor adding silane resulted in anadequate resin bond to some new ceramics [17-19]. Particularly high-alumina[20,21] or zirconia ceramics [22,23] cannot be roughened by hydrofluoric acidetching since such ceramics do not contain a silicon dioxide (silica) phase. Forthis reason, special conditioning systems are indicated for these types ofceramics.

Advances in adhesive dentistry have resulted in the recent introduction ofmodern surface conditioning methods that require airborne particle abrasion ofthe surface before bonding in order to achieve high bond strength. One suchsystem is silica coating. In this technique, the surfaces are air abraded withaluminium trioxide particles modified with silica [24-27]. The blasting pressureresults in the embedding of these silica coated alumina particles on theceramic surface, rendering the silica-modified surface chemically morereactive to the resin through silane coupling agents. Silane molecules reactwith water to form three silanol groups (-Si-OH) from the correspondingmethoxy groups (-Si-O-CH3). The silanol groups then react further to form asiloxane (-Si-O-Si-O-) network with the silica surface. Monomeric ends of thesilane molecules react with the methacrylate groups of the adhesive resins byfree radical polymerization process.

When a ceramic exhibits very similar surface compositions and chemical

60


states of silicon and oxygen, then it is reasonable to hypothesize that thesiloxane bond will be achieved as these represent the binding sites for thecoupling agent to the ceramic surface. Since silane coupling agents do not bondwell to alumina, the bond strengths of resin composite to the ceramic will beaffected [19]. However, when alumina or zirconia ceramics are glass infiltrated,they are melted together at high temperatures to form a ceramic matrix. Thechemical components of the ceramics (traces such as Li2O, Na2O, K2O, CaO,MgO) are then bounded to each other by strong covalent bonds with hydroxylgroups at the surface of the ceramic material [28]. When the surface is acidetched and rinsed, this would generate more hydroxyl groups on the surface andalso enhance the micro-mechanical retention. Furthermore, the methoxy groupsof silane would react with water to form silanol groups that in turn will react withthe surface hydroxyl groups to form siloxane network. It was hypothesized in thisstudy that amphoteric alumina in the ceramic matrix could form strong enoughchemical adhesion bonds, covalent bridges, through its surface hydroxyl groupswith hydrolysed silanol groups of the silane: -Al-O-Si-.

The microstructure, morphology and mechanical properties of theintermediate region adjacent to the silane-modified surface of the substrateand to the matrix are also important considerations. If contact is suppliedbetween a polymer and the uncross-linked siloxane/nonreacted silanolbridges, the bonding can take several forms including copolymer formationand interpenetrating polymer networks via methacrylate groups [29, 30].Increased cross-linking of the siloxane structure in the interphase region byadhesive monomers can give higher bond strength and superior resistance tomoisture. One other function of adhesive silane monomer is to achieve betterwetting of the substrate surface. Although intermediate resin is not necessarilyneeded with flow viscosity, some products clearly benefit using them [30].

Although comparative studies exist, showing the advantages of varioustypes of surface conditioning methods on various ceramics [31-39], there hasbeen no consensus in the literature regarding the best surface conditioningmethod for optimum bond strength depending on the luting cements orceramics used. Therefore, the objectives of this study were to evaluate theeffect of current surface conditioning methods on the bond strength of a resincomposite luting cement bonded to ceramic surfaces and to identify theoptimum method to be used for conditioning the ceramics prior to cementation.

2. Materials and methods

Thirty-six experimental groups (n=6) of six types of ceramic materials, namely

61


Finesse (FIN), In-Ceram (INC-AL), Celay (INC-ZR), IPS Empress 2 (EMPII),Procera® AllCeram (PRO) and Experimental alumina (EAL) were obtainedfrom the manufacturers. The specimens were in disc forms with 10 mm indiameter and 2 mm in thickness. Three surface conditioning techniques wereassessed for the ceramic materials at both dry and thermocycled storingconditions. Tables 1 and 2 summarize the characteristics of surfaceconditioning methods and ceramic types with codes and manufacturingcompany names. Before initiating the bonding procedure, the specimens wereembedded in acrylic resin blocks ensuring that one surface of the discremained uncovered for bonding procedures. The exposed surface of eachspecimen was ground finished to 1200 grit silicone carbide abrasive (StruersRotoPol 11, Struers A/S, Rodovre, Denmark) and cleaned for 10 min in anultrasonic bath (Quantrex 90 WT, L&R Manufacturing, Inc., Kearny, NJ, USA)containing ethylacetate and air-dried. Subsequently, the specimens wererandomly assigned to one of the following three conditioning methods:

Table 1. Characteristics of surface conditioning methods assessed.

Conditioning principle Manufacturer

Hydrofluoric acid (9.5%, 90 s) Ultradent Porcelain Etch, South Jordan, USA

(5%, 20 s) IPS Empress Ceramic Etch, Vivadent, Schaan,

Liechtenstein

Air particle abrasion (110 µm alumina, Korox, Bego, Bremen, Germany

380 kPa, 10 mm, 13 s)

Tribochemical silica coating Rocatec Pre, Rocatec Plus 3M ESPE AG, Seefeld, Germany

(280 kPa, 10 mm, 13 s),

Silane (5 min)

2.1. Surface conditioning methods

In hydrofluoric acid-etched groups, the ceramic substrates were etched with9.5% hydrofluoric acid gel for 90 s except EMPII for which etching wasperformed for 20 s with 5% hydrofluoric acid gel according to themanufacturer`s strict regulations. The ceramic surfaces were etched in thelaboratory under ventilation, wearing acid-resistant gloves and protectiveglasses. The etching gel was rinsed in a polyethylene cup and the dilutedsolution was neutralized using the neutralizing powder (calcium carbonate,CaCO3 and sodium carbonate, Na2CO3) for 5 min and washed thoroughly for

62


20 seconds as recommended by the manufacturers of FIN and EMPII. Theetched substrates were washed and rinsed thoroughly to remove the residualacid after etching, air-dried and coated with a 3-methacryloxypropyltrimethoxysilane coupling agent (Monobond S, Vivadent, Schaan, Liechtenstein). Silanewas allowed to remain in contact for 60 s. The surface was then dried with air.

As an alternative conditioning method to the etching, airborne particleabrasion was performed using 110 µm grain sized aluminium trioxide powderat a pressure of 380 kPa from a distance of approx. 10 mm, for 13 s. Followingair particle abrasion, silane coupling agent (ESPE-Sil, 3M ESPE AG, Seefeld,Germany) was applied and waited for its evaporation for 5 min.

The third conditioning method was tribochemical silica coating in whichthe specimens were first conditioned by air-abrasion with 110 µm grain sizedaluminium dioxide particles at a pressure of 280 kPa with Rocatec Preabrasive in a Rocatector Delta device (3M ESPE). Then the specimens wereair-abraded with Rocatec Plus abrasive, which was 110 µm grain sizedaluminium dioxide modified with silisic acid, at 280 kPa from a distance of 10mm for 13 s. The surfaces were coated with silane coupling agent (ESPE-Sil)and allowed to dry for 5 min.

Table 2. Types of ceramics with codes, and manufacturing company names.

Trade name Abbreviation Ceramic Type Manufacturer

Finesse FIN leucite reinforced Ceramco, Burlington, NJ, USA

In-Ceram INC-AL glass-infiltrated alumina (70%) Vita Zahnfabrik, Bad Saeckingen,

Germany

Zirkonia Blank for Celay INC-ZR glass-infiltrated zirconia Vita Zahnfabrik, Bad Saeckingen,

Germany

IPS Empress 2 EMPII lithium disilicate Ivoclar, Schaan, Liechtenstein

Procera AllCeram PRO high alumina (99.9%) Nobel Biocare AB, Göteborg, Sweden

Experimental alumina EAL high alumina (99.7%) Technical University, Tampere, Finland

2.2 Bonding procedure

Throughout the experiments, the bonding procedures were carried out inaccordance with the manufacturers` instructions. All materials were mixed andapplied in a standardized way by the same operator. In the acid etched groups,adhesive resin (Heliobond, Vivadent) was applied a thin layer, excess resinwas removed with air and it was light polymerized (Elipar, 3M ESPE) for 20 s.

63


Light-intensity was 800 mW/cm2. The low-viscous resin cement (Variolink® II,Vivadent) was then bonded to the conditioned ceramic specimens usingtranslucent polyethylene molds with inner diameter of 3.6 mm and height of 5mm.The low-viscous resin was packed against the substrate with a composite-filling instrument. The resins were light polymerized for 40 s. Polyethylenemolds were gently removed from the test specimens. While dry samples werekept in a dessicator at room temperature for 24 h prior to testing, the othergroups were subjected to thermocycling (custom procedure made by NIOM-Scandinavian Institute for Dental Materials, Haslum, Norway) for 6.000 cyclesbetween 5ºC and 55ºC in deionised grade 3 water. The dwelling time at eachtemperature was 30 s. The transfer time from one bath to the other was 2 s.

Specimens were mounted in a jig (Bencor Multi-T shear assembly,Danville Engineering Inc., San Ramon, CA, USA) of the universal testingmachine (Llyod LRX, Lloyd Instruments Ltd, Fareham, UK) and the shear forcewas applied to the adhesive interface until fracture occurred. The specimenswere loaded at a crosshead speed of 1.0 mm/min and the stress-strain curvewas analysed with Nexygen 2.0 software (Llyod LRX, Lloyd Instruments Ltd,Fareham, UK).

Statistical analysis was performed using SAS System for Windows,release 8.02/2001 (Cary, NC, USA). The means of each group were analysedby two-way analysis of variance (ANOVA), with shear bond strength as thedependent variable, the surface conditioning methods and the ceramic typesas the independent factors. P values less than 0.05 are considered to bestatistically significant in all tests. Multiple comparisons were made by Tukey`sadjustment test. Furthermore, one-way ANOVA was used to determine thesignificant differences between dry and thermocycled conditions.

3. Results

The results of the shear bond strength test for hydrofluoric acid etching,airborne particle abrasion and tribochemical silica coating are presented inFigs. 1a-c. While ANOVA showed significant influence of the ceramic type onthe shear bond strength values (P < 0.0001), less difference was found forsurface conditioning methods (Tables 3 and 4).

The highest shear bond strengths in dry conditions were obtained withglass ceramics (FIN and EMPII) in all surface conditioning groups varyingbetween 20.1 and 38.8 MPa. The lowest bond strengths were found with PROin all conditioning methods ranging from 5.3 to 8.5 MPa.

One-way ANOVA showed that shear bond strength was significantly

64


affected by thermocycling (P < 0.001). The least reduction in shear bondstrength values after thermocycling was with EMPII ceramic following acidetching conditioning. With other ceramic substrates reduction was higher.

Table 3. Results of 2-way analysis of variance for dry conditions.

Source of variation df Sum of squares Mean square F-value P value

Ceramic type (A) 5 7469.397 1493.879 26.078 <.0001

Surface Conditioning (B) 2 544.882 272.441 4.756 0.0110

A*B 10 985.058 98.506 1.720 0.0891

Error 86 4926.516 57.285

Total 103 14262.468

65

Acid etching

0

10

20

30

40

50

FIN INC CEL EMPII PRO EAL

Bo

nd

Str

eng

th (

MP

a)DryTC

Air particle abrasion

0

10

20

30

40

50


Bo

nd

Str

eng

th (

MP

a)

DryTC

Silica coating

0

10

20

30

40

50


Bo

nd

str

eng

th (

MP

a)

DryTC

Fig.1 a-c. Shear bond strengths after a) Hydrofluoric acid etching, b) Airborne particle abrasion

and c) Tribochemical silica coating at dry and thermocycled conditions. Vertical lines represent

the standard deviations. For abbreviations, see Table 2.


Table 4. Results of 2-way analysis of variance for thermocycled conditions.


Ceramic type (A) 5 1421.607 284.321 10.017 <.0001

Surface Conditioning (B) 2 216.582 108.291 3.815 0.0256

A*B 10 2127.236 212.724 7.495 <.0001

Error 92 2611.261 28.383

Total 109 6515.943

4. Discussion

Requirement for successful function of ceramic restorations over the years isadequate adhesion between ceramic and tooth substance. Bond strengths areinfluenced by several factors one of which is the luting cement type [40,41].Bonding of ceramic to tooth substance is based on the adhesion of lutingcement and its bonding resin to the ceramic substrate together with theadhesion of luting cement to enamel and dentine.

Hydrofluoric acid selectively dissolves glassy or crystalline components ofthe ceramic and produces a porous irregular surface that increases the surfacearea and facilitates the penetration of the resin into the microretentions of theetched ceramic surfaces. In this study, while acid etching demonstrated higherresults for glass ceramics (FIN and EMPII), it did not improve the bond strengthof the luting cement to high-alumina ceramics or zirconium oxide ceramic. Thedifferences obtained in bond strength can be explained on the basis ofvarieties in surface morphology. FIN and EMPII are glass ceramics as the firstone is a leucite reinforced and the latter a lithium disilicate ceramic. Theprimary function of leucite is to raise the coefficient of thermal expansion,consequently increasing the hardness and fusion. The FIN ceramic includes 8-10% leucite crystals which are very receptive to hydrofluoric acid etchingbefore bonding with the resin cement.

The great influence of the type of substrates on the bond strength of Bis-GMA resin to ceramics can be clearly seen in the case of high-aluminaceramics. INC-AL, PRO, EAL are loosely sintered high-alumina ceramics.Principally, acid etching will only affect the grain boundaries visible on thesurface. Hydrofluoric acid etching did not create sufficient bond strength on theINC-AL due to its high alumina content and it was almost ineffective fordissolving the glassy phase for micromechanical bonding. Our results are incompliance with the earlier report by Sorenson et al. [42] who showed thathydrofluoric acid etching significantly increased the bond strength of most of

66


feldspathic ceramics but did not improve the bond strength to the core part ofthe INC-AL ceramic. Furthermore, statistically significant difference wasobserved between PRO and EAL which were also high-alumina ceramics. Thereason for this finding could be attributed to the sintering temperatures andprocessing such ceramics. EAL contained refractory “ivory“ alumina (99.7%)and has been sintered in higher temperatures which could have effected thegrain size. Usually grain size in alumina ceramics are 1-3 µm but since EAL isat an experimental stage at the moment, the manufacturer claimed that thesintering process was not precise yet and therefore it contained grain size of20-30 µm including some pores. In the case of PRO, resin composite lutingcement exhibited poor adhesion to the ceramic substrates.

While some studies found no obvious correlation between different acids[43], the optimal concentration and duration of their application are not well-established, which is reflected in the variety of concentrations of commerciallyavailable hydrofluoric acids. Although less concentration and less duration wasused, high bond results were obtained for EMPII after acid etching at both dryand thermocycled conditions. In a study by Madani et al. [33], 5% and 9.5%acid gel was compared and bond strength values with 5% hydrofluoric acidwas found to be lower but not significantly different. It should also be noted thatin this study, all ceramics tested after acid etching showed higher standarddeviations compared with other surface conditioning methods. Oneconceivable explanation for high standard deviations could be that the poorlyadherent precipitates that are deposited at the bottom surface of the groovesand channels, created by acid treatment and rinsing, may weaken resin-ceramic bonds and lead to failure [44,45]. Ultrasonic cleaning could be oneoption but in this experiment, washing and rinsing were performed using air-water syringe. In clinical applications however, when etching will becontemplated by chairside, this finding might have a big impact on the marginalareas of the restorations.

Air-particle abrasion is a prerequisite for achieving sufficient bond strengthbetween the resins and ceramics. Significant improvement was observed in allceramic groups after air-particle abrasion followed by silanization except forFIN and EMPII. Although satisfactory bond results were obtained using air-particle abrasion, the material loss from these procedures after employing ondifferent substrates is important [33]. The data showed however that aluminumoxide particles were essential for creating micro-mechanical retention on high-alumina ceramics compared to hydrofluoric acid etching.

The tribochemical silica coating followed by silanization, which increasedthe silica content on the ceramic surface, evidently enhanced the bondbetween the ceramic surfaces and the luting cement. Since the silica layer is

67


attached well to the ceramic surface, this provides a basis for silanes toenhance the resin bond. Particular increase was observed for INC-AL, EALand INC-ZR. Similar findings were obtained in previous studies [21,31,33,34].

In this study a Bis-GMA based resin was used as the luting cement. A highand reliable resin bond to alumina and zirconia ceramics was also achievedwith airborne particle abrasion and by using a phosphate monomer (MDP)containing resin composite luting cement. Although Wegner et al. [46] reportedbetter long-term results with MDP containing cements than using thetribochemical silica coating procedure, Özcan et al. [21] did not observesignificant differences between MDP or Bis-GMa containing resin cementswhen tribochemical silica coating was employed. The question still needs to beaddressed in further studies whether the luting cement alone and/or thecombination with the conditioning method play the curicial role in long-termadhesion to the ceramic.

The possible influence of water storage in experimental studies must alsobe addressed. Different findings compared to others [23,38,39], especially forPRO after air particle abrasion or silica coating may be due to the storageconditions of the specimens. Usually bond strength values decreased afterthermocycling [19, 47-49], while some others reported no decrease [20]. Suchdifferences might be explained by the differences in experimental set-up, whichis important to keep in mind when in vitro studies are extrapolated to a clinicalsituation. In this study, the specimens were subjected to shear test after 6.000thermocycles. Although it was well above the recommended cycle numberaccording to ISO [50], one limitation of this study could still be the short-termwater storage and lower thermal cycling in comparison to other studies thatmight make it difficult to predict the long-term durability of the tested bondingmethods.

The present study did not find an ideal surface conditioning technique thatcould be applied to all types of ceramics. Because many factors affect the bondstrengths of resin luting cements to ceramics, it is necessary for dentists tounderstand the characteristics of the ceramics and the surface conditioningmethods in accordance with the cements to be chosen.

5. Conclusions

1. Bond strengths of the resin composite luting cement tested on the dentalceramics after surface conditioning techniques varied in accordance withthe ceramic types.

2. The findings confirmed that the use of hydrofluoric acid appeared to be the

68


method of choice for bonding the Bis-GMA resin composite luting cementto the ceramics having glassy matrix in their structures.

3. Roughening the ceramic surfaces with air particle abrasion prior tocementation provided higher bond strengths for high-alumina ceramicsand the values increased more significantly after silica coating/silanization.

4. Thermocycling decreased the bond strength values significantly after allsurface conditioning methods tested.

Acknowledgements

Special thanks are due to the manufacturing companies for the donation and provision of the

ceramic substrates. The authors are also grateful to Tero Vahlberg, M. A., Department of

Biostatistics, University of Turku, for his assistance with statistical analysis and Jukka Matinlinna,

M.Sc., Research associate, University of Turku, Department of Prosthodontics and Biomaterials

Research, for the helpful discussions.

References

[1] Kon M, Kawano F, Asaoka K, Matsumoto N. Effect of leucite crystals onthe strength of glassy porcelain. Dent Mater 1994;13:138-47.

[2] Seghi RR, Sorensen JA. Relative flexural strength of six new ceramicmaterials. Int J Prosthodont 1995;8:239-46.

[3] Seghi RR, Denry IL, Rosenstiel SF. Relative fracture toughness andhardness of new dental ceramics. J Prosthet Dent 1995;74:45-50.

[4] Denry IL, Mackert JR Jr, Holloway JA, Rosenstiel SF.Effect of cubic leucite stabilization on the flexural strength of feldspathicdental porcelain. J Dent Res 1996;75:1928-35.

[5] Mackert JR, Russell CM. Leucite crystallization during processing of aheat-pressed dental ceramic. Int J Prosthodont 1996;9:261-5.

[6] Mackert JR Jr, Williams AL, Ergle JW, Russell CM. Water-enhancedcrystallization of leucite in dental porcelain. Dent Mater 2000;16:426-31.

[7] Probster L, Diehl J. Slip-casting alumina ceramics for crown and bridgerestorations. Quintessence Int 1992;23:25-31.

[8] Sadoun M, Asmussen E. Bonding of resin cements to an aluminousceramic: a new surface treatment. Dent Mater 1994;10:185-9.

[9] Andersson M, Oden A. A new all-ceramic crown – A dense-sintered, highpurity alumina coping with porcelain. Acta Odontol Scand 1993;51:59-64.

69


[10] Piconi C, Maccauro G. Zirconia as a ceramic biomaterial. Biomaterials1999;20:1-25.

[11] Calamia JR. Etched porcelain veneers: The current state of the art.Quintessence Int 1985;1:5-12.

[12] Stangel I, Nathanson D, Hsu CS. Shear strength of the composite bondto etched porcelain. J Dent Res 1987;66:1460-5.

[13] Thurmond JW, Barkmeier W, Wilwerding TM. Effect of porcelain surfacetreatments on bond strengths of composite resin bonded to porcelain. JProsthet Dent 1994;72:355-9.

[14] Tylka DF, Stewart G. Comparison of acidulated phosphate fluoride geland hydrofluoric acid etchants for porcelain-composite repair. J ProsthetDent 1994;72:121-7.

[15] Aida M, Hayakawa T, Mizukawa K. Adhesion of composite to porcelainwith various surface conditions. J Prosthet Dent 1995;73:464-70.

[16] Nicholls JI.Tensile bond to resin cements to porcelain veneers. J ProsthetDent 1998;60:443-7.

[17] Rosenstiel SF, Gupta PK, Van der Sluys RA, Zimmermann MH. Strengthof a dental glass-ceramic after surface coating. Dent Mater 1993;9:274-9.

[18] Wolf DM, Powers JM, O'Keefe KL. Bond strength of composite to etchedand sandblasted porcelain. Am J Dent 1993;6:155-8.

[19] Kern M, Thompson VP. Bonding to glass infiltrated alumina ceramic:Adhesive methods and their durability. J Prosthet Dent 1995;73:240-9.

[20] Lu YC, Tseng H, Shih YH, Lee SY. Effects of surface treatments on bondstrength of glass-infiltrated ceramic. J Oral Rehabil 2001;28:805-13.

[21] Özcan M, Alkumru HN, Gemalmaz D. The effect of surface treatment onthe shear bond strength of luting cement to a glass-infiltrated aluminaceramic. Int J Prosthodont 2001;14:335-9.

[22] Kern M, Wegner SM. Bonding to zirconia ceramic: Adhesion methods andtheir durability. Dent Mater 1998;14:64-71.

[23] Derand P, Derand T. Bond strength of luting cements to zirconium oxideceramics. Int J Prosthodont. 2000;13:131-5.

[24] Peutzfeldt A, Asmussen E. Silicoating. Evaluation of a new method ofbonding composite resin to metal. Scand J Dent Res 1988;96:171-6.

[25] Özcan M, Pfeiffer P, Nergiz I. A brief history and current status ofmetal/ceramic surface conditioning concepts for resin bonding indentistry. Quintessence Int 1998;29:713-24.

[26] Özcan M. The use of chairside silica coating for different dentalapplications: A clinical report. J Prosthet Dent 2002;87:469-72.

70


[27] Özcan M, Niedermeier W. Clinical study on the reasons for and locationof failures of metal-ceramic restorations and survival of repairs. Int JProsthodont 2002;15:299-302.

[28] Shimada Y, Yamaguchi S, Tagami J. Micro-shear bond strength of dual-cured resin cement to glass ceramics. Dent Mater 2002;18:380-8.

[29] Hooshmand T, van Noort R, Keshvad A. Bond durability of the resin-bonded and silane treated ceramic surface. Dent Mater 2002;18:179-88.

[30] Kallio TT, Lastumaki TM, Vallittu PK. Bonding of restorative and veneeringcomposite resin to some polymeric composites. Dent Mater 2001;17:80-6.

[31] Bailey LF, Bennett RJ. DICOR surface treatments for enhanced bonding.J Dent Res 1988;67:925-31.

[32] Sorensen JA, Kang SK, Avera SP. Porcelain-composite interfacemicroleakage with various porcelain surface treatments. Dent Mater1991;7:118-23.

[33] Kern M, Thompson VP. Sandblasting and silica coating of a glass-infiltrated alumina ceramic: Volume loss, morphology and changes in thesurface composition. J Prosthet Dent 1994;71:453-61.

[34] Isidor F, Stokholm R, Ravnholt G. Tensile bond strength of resin lutingcement to glass infiltrated porous aluminium oxide cores (In-Ceram). EurJ Prosthodont Restor Dent 1995;3:199-202.

[35] Wood DJ, Bubb NL, Millar BJ, Dunne SM. Preliminary investigation of anovel retentive system for hydrofluoric acid etch-resistant dentalceramics. J Prosthet Dent 1997;78:275-80.

[36] Awliya W, Oden A, Yaman P, Dennison JB, Razzoog ME. Shear bondstrength of a resin cement to densely sintered high-purity alumina withvarious surface conditions. Acta Odontol Scand 1998;56:9-13.

[37] Thompson JY, Rapp MM, Parker AJ. Microscopic and energy dispersivex-ray analysis of surface adaptation of dental cements to dental ceramicsurfaces. J Prosthet Dent 1998;79:378-83.

[38] Blixt M, Adamczak E, Linden LA, Oden A, Arvidson K. Bonding to denselysintered alumina surfaces: effect of sandblasting and silica coating onshear bond strength of luting cements. Int J Prosthodont 2000;13:221-6.

[39] Madani M, Chu FC, McDonald AV, Smales RJ. Effects of surfacetreatments on shear bond strengths between a resin cement and analumina core. J Prosthet Dent 2000;83:644-7.

[40] Kelly JR, Campbell SD, Bowen HK. Fracture-surface analysis of dentalceramics. J Prosthet Dent 1989;62:536-41.

[41] Rosenstiel SF, Land MF, Crispin BJ. Dental luting agents: A review of thecurrent literature. J Prosthet Dent 1998;80:280-301.

71


[42] Sorenson JA, Engelman, MJ, Torres, TJ, Avera SP. Shear bond strengthof composite resin to porcelain. Int J Prosthodont 1991;4:17-23.

[43] al Edris A, al Jabr A, Coley RL, Barghi N. SEM evaluation of etch patternsby three etchants on three porcelains. J Prosthet Dent 1990;64:734-9.

[44] Phoenix RD, Shen C. Characterization of treated porcelain surfaces viadynamic contact angle analysis. Int J Prosthodont 1995;8:187-94.

[45] Canay S, Hersek N, Ertan A. Effect of different acid treatments on aporcelain surface. J Oral Rehabil 2001;28:95-101.

[46] Wegner SM, Gerdes W, Kern M.Effect of different artificial agingconditions on ceramic-composite bond strength. Int J Prosthodont2002;15:267-72.

[47] Roulet JF, Soderholm KJ, Longmate J. Effects of treatment and storageconditions on ceramic/composite bond strength. J Dent Res1995;74:381-7.

[48] Yoshida K, Kamada K, Atsuta M. Effects of two silane coupling agents, abonding agent, and thermal cycling on the bond strength of a CAD/CAMcomposite material cemented with two resin luting agents. J ProsthetDent 2001;85:184-9.

[49] Wegner SM, Kern M. Long-term resin bond strength to zirconia ceramic.J Adhes Dent. 2000;2:139-47.

[50] International Organization for Standardization, Polymer-based, Crownand Bridge Materials, ISO 10477, Amendment 1, 1998.

72


ChapterResistance of Core Materials Against Torsional

Forces on Differently Conditioned Titanium Posts

This chapter has been published as:Akıslı, I., Özcan, M., Nergiz, I.: Resistance of Core Materials Against TorsionalForces on Differently Conditioned Titanium Posts. Journal of ProstheticDentistry, 88;367-374, 2002. (reproduced with the permission from TheEditorial Council of The Journal of Prosthetic Dentistry.)

73

3


74


Resistance of Core Materials Against Torsional

Forces on Differently Conditioned Titanium Posts

I. Akislia, M. Özcanb, I. Nergiza

aUniversity of Hamburg, Dental School, Hamburg, Germany,bUniversity of Groningen, Faculty of Medical Sciences,

Department of Dentistry and Dental Hygiene, Groningen, The Netherlands

ABSTRACT

Statement of problem. The separation of core materials from titanium posts,which have low modulus of elasticity, has been identified as a problem inrestorative dentistry.Purpose. This study evaluated the resistance to torsional forces of variouscore materials adapted to differently conditioned titanium posts.Material and methods. Sixty-six groups (10 specimens per group) of custommade pure titanium posts (DIN 17850-Ti4/3.7065) were conditioned with thefollowing products: Silicoater Classical, Silicoater MD, Rocatec, Kevloc andSiloc surface conditioning systems. Subsequently, 6 core materials withdifferent compositions (Durafill, Adaptic, Coradent, Ti-Core, Hytac Aplitip andPhotac-Fil Aplitip) were applied to titanium posts that were previously coatedwith two types of light-polymerized opaquers, either Artglass or Dentacolor.Sixty air-abraded titanium posts (250 µm, 30 seconds) were used as controlsfor each core material. Following thermocycling (5-55�C, 30 seconds, 5000cycles), maximum torsional forces were determined with an electronic torquemovement key. Data were statistically analyzed by using 1-way ANOVAfollowed by 2-way ANOVA (P<.05).Results. Significantly higher (P<.001) mean torsional forces were observedwith respect to Siloc (20.4 dNm), Silicoater Classical (18.6 dNm), SilicoaterMD (18.2 dNm) and Rocatec (17.0 dNm) systems compared with the mean forthe untreated control group (14.6 dNm). Kevloc (10.4 dNm) systemdemonstrated no significant difference (P>0.001) compared to the controlgroup. Kevloc system in combination with Artglass opaquer and Photac-FilAplitip (0.00 dNm) core material showed no resistance against torsional forces.Significant differences (P<.001) were observed between hybrid core materialsand microfilled composite, compomer or resin-modified glass ionomer corematerials.Conclusion. Within the limitations of this study, the resistance to torsional

75


forces for the core materials on titanium posts increased with the use ofchemical surface conditioning techniques and varied in accordance with theopaquer type. Type of core material also significantly influenced the resistanceafter thermocycling.

Clinical implications

In this study, microfilled composite and resin-modified glass ionomer corematerials offered poor resistance to torsional forces. Surface conditioningsystems based on silica coating/silanization or silica coating/acrylization canbe recommended for composite cores used with titanium posts.

INTRODUCTION

The long-term clinical performance of prefabricated posts and cores dependson a number of factors,1-2 and retention of the core material is one of the mostsignificant of these.3-7 A great number of screw or threaded prefabricated postsystems are commercially available. These systems, which vary according totheir geometrical design and structural properties, are made from puretitanium or titanium alloys in addition to the traditional stainless steel posts.Titanium is biocompatible, although its relatively low modulus of elasticitycompared with most stainless steels is viewed as a major disadvantage.8,9

Experimental and clinical reports provide evidence of significantdifferences in the survival of posts.10-13 Reduced chair time and ease ofmanipulation of prefabricated posts, as compared with cast posts and cores,make the procedure appealing to practitioners. Silver amalgam, composite andresin-reinforced glass ionomer restorative materials have been suggested ascore materials for use with prefabricated posts.14-22 Core reconstructionsmade from silver amalgam offer high compressive strength and ease ofmanipulation17 but composites can be prepared immediately after placementand impressed for restoration with a complete crown.16,22 With dentin bondingagents, composites also offer the advantage of bonding to tooth structure.Another material used for core reconstruction is resin-reinforced glassionomer. The advantages of this material are fluoride release to the adjacenttooth structure and rapid setting. However, glass ionomers including resin-reinforced materials are inherently brittle, vulnerable to moisture23 and lackingin strength to withstand occlusal loading under simulated occlusal forces.24

Most in vitro studies have been performed to evaluate the posts and cores

76


subjected to tensile,25,26 compressive,26,27 shear forces,28-30 trauma31 andfatigue tests.32 However, clinically, posts are also subjected to torsional orrotational forces produced by functional tooth contacts.33-37

Advances in adhesive dentistry have resulted in the recent introduction ofmodern surface treatment methods. These new systems involve theconditioning of the substrate to produce bifunctional molecules that adhere tothe metal surface through silane by means of a polymerizable double bond.Such molecules react with the methacylate groups contained in the monomersof the applied opaque/composite in a radical polymerization process. Themanufacturers of most of the new surface conditioning systems requireairborne particle abrasion of the metal prior to bonding to achieve high bondstrengths. From those systems, Silicoater Classical system was introduced in1984 as a system of molecular bond between resins and the surface of dentalalloys. The procedure produces an intermediate layer containing SiOx to allowbonding of the resin through silane bonding. The technique consists of airabrasion followed by heating SiOx with a flame in a specially designedapparatus that burns chrome, producing a silica layer on the surface.38-41

Silicoater MD system is the new version of the Silicoater Classicaltechnique, and uses the same principle. The difference between the twotechniques is the method of coating the metal surface. The surface is coatedwith SiOx in a liquid form in Silicoater MD apparatus in which firing takes placeat highly controlled temperatures. This system has the advantage of avoidingflame adjustment problems, eliminating of the human factor. 39,40,42

The Rocatec conditioning procedure presented a kind of resin-metalbonding procedure based on silica coating and silanization. The principle istribochemical application of a silica layer by means of air-abrasion which takesplace in the Rocatector Delta unit with Rocatec Pre and Rocatec Plus abrasivesfollowed by silane application. The Rocatec Plus particles hit the alloy surfacewith a theoretically calculated speed of 200 m/s, producing spot heating up to1000�C. This spot heating together with the air-abrasion pressure results inembedding of silica particles on the metal surface, rendering the metal surfacechemically more reactive to resin through silane. The particular advantages ofthe process are the speed and accuracy of coating and the ability to visuallycheck the adhesive layer so that thermal stressing of the framework is avoided.In these aforementioned procedures, by increasing the roughness of the metalsurface, the air-abrasion contributes to the formation of a mechanical bond. Asa result of silanization, the metal surface is provided with a layer of doublebonds.40,42,43

To improve bond strengths of the Silicoater technique, new bondingsystems, namely Kevloc and Siloc were introduced. Kevloc system works with

77


acrylization and Siloc with both silica coating and acrylization principles.Kevloc is a relatively new system introduced in 1995 which offers a

combination of mechanical and chemical bonding, presenting promising resultsin resin veneer-alloy bonding, inlays, onlays, implant-supported restorationsand removable prostheses. The temperature needed for the activation of thebonding layers is generated by contact heat transfer and heat radiation in theactivation chamber of the Kevloc apparatus. The bonding layers in this systemconsist of a fused layer of the rigid acrylonitrile and a layer of the water-resistant, highly cross-linked urethane resin. Both layers are reinforced by thedirect acrylizing network and have the capability of sustaining the loadingforces.40,42-45

Siloc system involves air particle abrasion followed by the application ofSiloc-Pre silane which is dried at room temperature first and then placed in theSiloc apparatus (Heraeus-Kulzer). Because of the application of both Siloc-Presilane and Siloc-Bond (Heraeus-Kulzer) bonding agent, the working principle ofthe system is called silica coating and acrylization.40

Thermocycling relaxes stresses within the composites produced bypolymerization shrinkage. In general, the surface treatment studies conductedhave used different thermocycling times and the common consensus was thatthe thermocycling decreased the bond strength as it weakened the resinstructure.46-48

Although comparative studies exist showing the advantages of varioustypes of surface conditioning methods on both base and noble alloys,49-53

limited information is available concerning the use of these techniques ontitanium substrate.54,56 Therefore, the aim of this study was to evaluate theeffect of current surface conditioning methods and opaquers on the resistanceof various core materials against torsional forces on titanium posts afterthermocycling.

MATERIAL AND METHODS

Seven hundred twenty custom-made pure titanium posts, each 3 mm indiameter and 15 mm in length, were prepared according to DIN 17850-Ti4/3.70651.57 Five surface conditioning techniques, two opaquers and sixpost/core materials were assessed (6x5x10x2=600). Tables I and IIsummarize the characteristics of surface conditioning methods, types andpolymerization procedures for the core materials and opaquers. Specimenswere divided into groups according to core material, opaquer, and surfaceconditioning method. Sixty titanium posts (10 groups of core materials and 2

78


types of opaquers) air-abraded with 250 µm aluminum oxide for 30 secondseach, were used as controls. The surface roughness of the control specimens,which were initially approximately equal to 2 µm, was measured to beapproximately equal to Rz≈13 µm after air-abrasion (Pruefgerät Type 402,Mitutoyo, Kawasaki, Japan).

Table I. Characteristics of surface conditioning methods assessed.

Trade name Abbreviation Conditioning principle Manufacturer

Silicoater Classical SC Silica coating-silanization Heraeus-Kulzer GmbH, Wehrheim, Germany

Silicoater MD SMD Silica coating-silanization Heraeus-Kulzer GmbH, Wehrheim, Germany

Rocatec RC Silica coating-silanization ESPE AG, Seefeld, Germany

Kevloc KV Acrylization Heraeus-Kulzer GmbH, Wehrheim, Germany

Siloc SL Silica coating-silanization Heraeus-Kulzer GmbH, Wehrheim, Germany

Table II. Type of the core materials and opaquers.

Trade name Abbreviation Core/opaquer type Polymerization Manufacturer

Durafill DL Microfilled Light polymerized Heraeus-Kulzer GmbH, Wehrheim, Germany

Adaptic AD Hybrid Chemically polymerized Johnson & Johnson, Skillman, USA

Coradent CD Hybrid Chemically polymerized Vivadent, Schaan, Liechtenstein

Ti-Core TC Hybrid Chemically polymerized EDS, Hackensack, USA

Hytac Aplitip HA Compomer Light polymerized ESPE AG, Seefeld, Germany

Photac-Fil Aplitip PF Glass ionomer Light polymerized ESPE AG, Seefeld, Germany

Dentacolor Dent Methacrylate Light polymerized Heraeus-Kulzer GmbH, Wehrheim, Germany

Artglass Art Bismethacrylate Light polymerized Heraeus-Kulzer GmbH, Wehrheim, Germany

Surface conditioning methods

Titanium posts were conditioned and underwent silanization chemicallyaccording to the manufacturers` recommendations.

With the Silicoater Classical (SC) method, the posts were air-abraded with250 µm aluminum oxide for 30 seconds at 4 bar. The surfaces were thencoated with Siliclean solution (Heraeus-Kulzer). After being washed, thesurfaces were air-dried for 2 minutes at room temperature. This techniqueconsisted of heating SiOx with a flame (Siliflam, Heraeus-Kulzer) in a speciallydesigned Silicoater apparatus at temperatures higher than 250�C. The flame

79


was adjusted to provide enough SiOx molecular deposition on the metalsurface. Siliflam was applied for 5 minutes (air/propan 20:1, air appr. 130 l/h,propan 6.5 l/h). After the sample was dried for 4 minutes in air, a silane layerof a mixture of Silicoup A and B (Heraeus-Kulzer) was applied.

Silicoater MD system worked in manner similar to the Silicoater Classical.The difference between the two techniques was the coating process for themetal surface. In Silicoater MD, SiOx was applied in a liquid form in a specialapparatus in which firing took place at highly controlled temperatures. With thisapparatus, flame adjustment problems and human factor were eliminated.

Rocatec was based on tribochemical application of a silica layer by meansof airborne particle abrasion. First, the specimens were conditioned by air-abrasion with 110 µm aluminum oxide grains at a pressure of 0.25 MPa withRocatec Pre abrasive by using a Rocatector Delta device. The specimenswere then air-abraded with Rocatec Plus abrasive, which was 110 µmaluminum oxide modified with silisic acid, at 0.25 MPa at a distance of 1 cmfrom the metal surface for 13 sec/cm2. The surfaces were then coated withsilane (ESPE-Sil, ESPE) to render them chemically more reactive to the resin.

The Kevloc system offers a combination of mechanical and chemicalbonding. In this system, the specimens were air-abraded with fresh 110 µmgrains of aluminum oxide at 0.2 MPa. The air-abraded surfaces were cleanedwith a brush and the loose particles were removed. Kevloc-Primer (Heraeus-Kulzer) was applied with a clean brush and dried at room temperature for 2minutes. Then, Kevloc Bond (Heraeus-Kulzer) was applied with a brush. Afterit was dry, the specimens were placed in the chamber of the Kevloc ACapparatus where the activation of the bonding layers was generated by contactheat transfer and heat radiation for 8 minutes. The specimens were cooled atroom temperature for 5 minutes.

With the Siloc technique, the surfaces were air-abraded with 250 µmgrains of aluminum oxide and dried with water-free and oil-free air at 0.3 MPa.The activated surfaces were then coated with Siloc-Pre silane with a brush anddried at room temperature for 2 minutes. Later, the specimens were placed inthe Siloc apparatus. Finally, they were cooled at room temperature for 4minutes and the activated surfaces were coated with Siloc-Bond and dried inthe air for 5 minutes.

Opaquer application

After conditioning the titanium posts, opaquers were painted on the titaniumposts and light-polymerized accordingly.

80


The powder and liquid components of Dentacolor Opaquer were mixed(1:1) to produce a thin consistency, stirring for at least 30 seconds to achievea homogeneous distribution of the pigments. A thin layer (Shade No. 123) wasapplied with a brush and light-polymerized for 90 seconds in the DentacolorXS unit (Heraeus-Kulzer GmbH). Second and third thin layers were separatelyapplied and polymerized until complete masking of the metal was achieved.

The components of Artglass Opaquer were mixed on a mixing pad. A thinlayer (Shade No. OA3) of the opaquer was painted on the alloy surface with ashort bristle brush. As the first layer did not mask the metal sufficiently, furtherlayering was performed. Each layer was light-polymerized for 90 seconds inthe Dentacolor XS unit.

To produce the specimens in a standardized form, a hexagonal metalmatrix with 6 mm diameter holes was used (Fig. 1). Subsequently, 6 corematerials were homogeneously placed within the matrix and cured accordingto the manufacturer`s instructions. While chemically polymerized corematerials were polymerized (Optilux 401; Demetron Research, Danbury,Conn., USA) in one increment, the light-polymerized cores were placed in themetal matrix in two increments to ensure that the lower portion was alsopolymerized. The light output of the curing unit was initially tested to be 630mW/cm2 with the Cure Rite instrument (Model 8000, EFOS Inc. Williansville,NY, USA). The irradiation distance between the exit window and the resinsurface was maintained at 10 mm to obtain adequate polymerization.

The specimens were then subjected to 5000 thermocycles at atemperature range between 5-55�C and a dwelling time of 30 sec.58 Anelectronic rotational torque device (Wear Werk; Hermann Werner, Germany)was connected to a compression test machine (Zwick 1445; Zwick GmbH,Germany) for the measurement of maximal torsional forces. The direction ofthe torsional forces was clockwise (Fig. 2). Failure was established at the pointat which the specimen could not withstand an increase in load.

The means of each group were analyzed by analysis of variance (ANOVA)(StatView 5.0; SAS Institute Inc., Cary, NC). If a significant difference wasestablished, 2-way ANOVA was used to determine the probability values witha significance level P<0.05.

81


RESULTS

Tables III and IV display the mean values obtained, together with the standarddeviations associated with surface conditioning techniques and core materialsfor Dentacolor and Artglass opaquer. The 2-way ANOVA revealed significantdifferences (P<.0001) between groups dependent on the combinations ofsurface conditioning techniques and opaquers and the interaction with corematerials (Tables V and VI).

Significant differences (P<.0001) were observed between hybrid andmicrofilled composite, compomer or resin-modified glass ionomer corematerials. The highest torque values were obtained with Siloc-Dentacolor-Adaptic (23.9 dNm). The most profound and significant improvements wereobserved with Siloc-Dentacolor-Hytac Aplitip (19.0 dNm). Kevloc system in

82

Fig. 1. Hexagonal metal matrix with holes 6 mm in diameter

and 15 mm in length.

Fig. 2. Schematic diagram of the electronic rotational torque

device connected to a compression test machine.


combination with Artglass opaquer and Photac-Fil Aplitip core material (0.00dNm) however showed no resistance against torsional forces.

Significantly higher (P<.0001) torsional forces with respect to Siloc (20.4dNm), Silicoater Classical (18.6 dNm), Silicoater MD (18.2 dNm) and Rocatec(17.0 dNm) systems were observed compared with the untreated controlgroup (14.6 dNm).

The mean torque force achieved for the group with the application of theKevloc (10.4 dNm) system was significantly lower than that attained for groupsprovided with other surface conditioning methods (P<.001). Improvement intorque forces were not significantly different for Dentacolor-Durafill, Adaptic, Ti-Core, Hytac Aplitip core materials compared with their control group (P>.001).

In the Artglass-Photac Fil group, none of the surface conditioningtechniques improved the torque forces.

Table III. Statistical significance of differences between the torque forces (dNm) within combinations of core materials and

surface conditioning techniques using Dentacolor at each combination.

Core Material

Surface Treatment DL AD CD TC HA PF

Control 12.5 (1.8)* 16.1 (1.1)* 15.1 (1.0) 12.2 (0.8)* 4.8 (1.9)* 7.0 (5.0)

SC 13.9 (3.0)* 20.8 (4.3) 21.1 (3.6) 19.6 (4.5) 16.5 (1.8) 19.5 (3.9)

SMD 14.4 (2.6)* 22.2 (2.9) 19.1 (3.4) 19.7 (3.7) 16.2 (2.1) 17.5 (3.9)

RC 15.6 (2.8) 21.5 (3.2) 18.5 (4.6) 19.5 (2.9) 6.5 (4.3)* 20.0 (1.4)

KV 11.1 (2.8)* 16.2 (1.5)* 12.3 (2.5) 12.9 (1.4)* 10.2 (3.2)* 0 (0)

SL 14.4 (3.4)* 23.9 (3.4) 22.2 (3.0) 22.6 (2.9) 21.0 (1.7) 18.7 (3.1)

Abbreviations shown in Tables I and II.

* Indicates no significant differences in the same column between surface treatment/core material in comparison to control group

as determined by 2-way ANOVA (P >.05).

83


Table IV. Statistical significance of differences between the torque forces (dNm) within combinations of core materials and sur-

face conditioning techniques using Artglass at each combination.

Core Material

Surface Treatment DL AD CD TC HA PF

Control 12.5 (1.8)* 16.1 (1.1)* 15.1 (1.0)* 12.2 (0.8)* 4.8 (1.9) 7.0 (5.0)*

SC 9.0 (3.1) 12.5 (3.2) 13.2 (3.6)* 11.0 (3.4)* 17.0 (1.1) 8.9 (3.2)*

SMD 7.1 (1.3) 6.7 (1.3) 11.1 (2.2) 6.2 (2.0) 12.9 (2.3) 0.5 (1.5)

RC 7.7 (1.7) 13.8 (3.5)* 12.6 (3.7) 7.6 (2.0) 16.2 (1.5) 5.6 (3.9)*

KV 8.9 (2.4) 8.2 (1.7) 10.4 (1.3) 8.5 (2.4) 9.7 (2.4) 0(0)

SL 13.4 (2.6)* 9.0 (1.5) 13.5 (1.6)* 4.5 (0.7) 19.3 (1.6) 4.4 (3.5)*

Abbreviations shown in Tables I and II.

* Indicates no significant differences in the same column between surface treatment/core material in comparison to control group

as determined by 2-way ANOVA (P >.05).

Table V. Results of 2-way analysis of variance for Dentacolor.


Surface conditioning (A) 5 4641.079689 928.215918 112.54 <.0001

Core material (B) 5 1776.213409 355.242682 43.07 <.0001

A*B 25 4715.052881 188.602115 22.87 <.0001

Error 324 2672.27652 8.24777

Total 359 13804.62240

Table VI. Results of 2-way analysis of variance for Artglass.


Surface conditioning (A) 5 6784.054378 1356.810876 207.75 <.0001

Core material (B) 5 2397.833771 479.566754 73.43 <.0001

A*B 25 2206.338247 88.253530 13.51 <.0001

Error 324 2116.00767 6.53089

Total 359 13504.23407

84


DISCUSSION

To withstand functional loads, the bond between the core material and the postshould be strong and durable. Recently, the technology for bonding of resin todental alloys has significantly improved. Various commercial surfaceconditioning techniques are available that claim to provide a durable bondbetween the resin and a metal substrate.

In this study, the ranking of surface conditioning methods with regard toresistance of core materials after thermocycling was more favorable with Siloc,Silicoater Classical, Silicoater MD, Rocatec and least favorable for Kevlocsystem. The advantages of extraoral surface conditioning have beenreported38, 43-45 and are in accordance with the results of the present study.The surface conditioning techniques tested in this study substantiallyincreased the attachment of the core materials tested in comparison to theunconditioned control groups.

In the Kevloc-treated groups with either Dentacolor or Artglass opaquerhowever, no resistance to torsion was obtained with Photac-Fil, which is aresin-modified glass-ionomer. The working principle of Kevloc is onlyacrylization whereas the other systems are based on silica coating/silanizationor silica coating/acrylization. The results obtained in Kevloc-treated groupsmay be attributed to either the nonuniform distribution of heat transfer or thelack of silane application. In a study by Vojvodic et al,44 this difference wasattributed to differences in bonding layer thickness, opaque viscosity oropaque liquid proportion. The common finding of several studies was that thebonding results decreased after artificial aging of specimens treated withKevloc.47,48,50-55 The individual mechanism producing this result needs to befurther clarified.

The type of composite and surface conditioning technique are especiallyimportant because the type/composition of current composites differ widely. Ofthe core materials used, the type of core material also significantly influencedthe resistance after thermocycling. In this study, three hybrid composites, onemicrofilled resin, one compomer and a resin-modified glass-ionomer wereinvestigated, with the results revealing that resin-modified glass ionomer corematerials were especially affected by the type of conditioning method. Oneexplanation for the non-resistance of Photac-Fil to Kevloc-treated titaniumposts might be the low resin content of this core material. The disadvantagesof composite cores include microleakage and problems with dimensionalstability that may affect marginal adaptation of castings.4 The results of thisstudy indicated that hybrid composites resisted torsional forces moreeffectively even after thermocycling. This result may be due to the resin matrix

85


or to particle size and filler composition.The results varied in accordance with surface conditioning methods and

the type of the opaquer for the core materials. When Dentacolor was used asan opaquer, significantly higher bond strengths were obtained compared withArtglass opaquer. Dentacolor and Artglass opaquers vary in composition: theformer is a methacrylate provided in a powder-liquid system, while the latter isa bismethacrylate provided in paste form. It is likely that bond strengthsproduced by opaquers are dependent on their chemical composition andconsistency. Handling characteristics of opaquers therefore, may be animportant variable. During experimental process, it was noted that when theywere used in the proportions specified by the manufacturers, the consistenciesof the opaquers were slightly different, which may have contributed todifferences in thickness between the resin and the conditioned metal surface,thereby affecting their resistance.

In clinical practice, both the configuration of the post and the type of corematerial are thought to affect the core’s retention on the posts. To eliminate theeffect of post design, titanium posts with smooth surfaces were used in theexperiments.

The current surface conditioning methods required additional equipmentfor conditioning the titanium posts before cementation, which is costly.Recently, attempts have been made to perform silica coating chairside bymeans of an air-abrasion device. Such technique might be an alternativeapproach to condition the posts chairside.

Because many factors affect the resistance of core materials on titaniumposts, it is necessary for dentists to understand the characteristics of surfaceconditioning methods in accordance with the opaquers and core materials tobe chosen. More information about the clinical performance of theseincreasingly popular techniques is needed. Further evaluation clinically ismerited to validate the findings of this in vitro study.

CONCLUSIONS

Within the limits of this study, the resistance of core materials to dislodgementfrom titanium posts after the application of torsional forces and thermocyclingwas affected by the conditioning systems used. Resistance was greatest withSiloc and then, in descending order, with Silicoater Classical, Silicoater MD,Rocatec, and Kevloc surface conditioning systems. In addition the resistanceof core materials based on silica coating/silanization or silicacoating/acrylization varied in accordance with the opaquer used. Dentacolor

86


opaquer demonstrated higher resistance values than Artglass opaquer. Hybridcomposites and compomers used as core materials demonstrated highertorque resistance compared with microfilled composites or resin-modifiedglass ionomer.

AcknowledgementsSpecial thanks are due to ESPE AG and Hereaus-Kulzer GmbH and Vivadent for their generous

provision and donation of the testing materials. We are also grateful to Tero Vahlberg for his

assistance with statistical analysis and Wilhelm Niedermeier, Prof Dr Med Dent, for his valuable

comments.

REFERENCES

1. Standlee JP, Caputo AA, Hanson EC. Retention of endodontic dowels:effect of cement, dowel length, diameter and design. J Prosthet Dent1978;39:401-5.

2. Standlee JP, Caputo AA, Holcomb J, Trabert KC. The retentive and stress-distributing properties of a threaded endodontic dowel. J Prosthet1980;44:398-404.

3. Olivia RA, Lowe JA. Dimensional stability of composite used as a corematerial. J Prosthet Dent 1986;56:554-61.

4. Kovarik RE, Breeding LC, Caughman WF. Fatigue life of three corematerials under simulated chewing functions. J Prosthet Dent1992;68:584-90.

5. Chang WC, Millstein PL. Effect of design of prefabricated post heads andcore materials. J Prosthet Dent 1993;69:475-82.

6. Cohen BI, Condos S, Deutsch AS, Musikant BL. Fracture strength of threedifferent core materials in combination with three different endodonticposts. Int J Prosthodont 1994;7:178-82.

7. Huysmans MCDNJM, Van der Varst PGT. Mechanical longevity estimationmodel for post-and-core restorations. Dent Mater 1995;11:252-7.

8. Zardiackas LD, Disegi J, Givan D. Torsional properties of implant gradetitanium. J Biomed Mater Res 1991;25:281-3.

9. Christensen GJ. Posts: necessary or unnecessary? J Am Dent Assoc1996;127:1522-6.

10. Sorenson JA, Martinoff JT. Intracoronal reinforcement and coronalcoverage: A study of endodontically treated teeth. J Prosthet Dent1984;51:780-4.

87


11. Sorenson JA, Martinoff JT. Clinically significant factors in dowel design. JProsthet Dent 1984;52:28-35.

12. Heydecke G, Butz F, Strub JR. Fracture strength and survival rate ofendodontically treated maxillary incisors with approximal cavities afterrestoration with different post and core systems: an in-vitro study. J Dent2001;29:427-33.

13. Strub JR, Pontius O, Koutayas S. Survival rate and fracture strength ofincisors restored with different post and core systems after exposure in theartificial mouth. J Oral Rehabil 2001;28:120-4.

14. Chan RW, Bryant RW. Post-core foundations for endodontically treatedposterior teeth. J Prosthet Dent 1982;48:401-6.

15. Hoag EP, Dwyer TG. A comparative evaluation for three post and coretechniques. J Prosthet Dent 1982;47:177-81.

16. Linde LA. The use of composites as core material in root filled teeth. SwedDent J 1983;7:92-9.

17. Nayar A. Coronal-radicular buildup for endodontically treated teeth.In:Clark JW, ed. Clinical dentistry. Vol 4. Philadelphia:Harper & Row,1983;1-28.

18. Chapman KW, Worley JL, Von Fraunhofer JA. Retention of prefabricatedposts by cements and resins. J Prosthet Dent 1985;54:649-52.

19. Plasmans PJJM, Visseren LGH, Vrijhoef MMA, Kayser AF. In vitrocomparison of dowel and core techniques for endodontically treatedmolars. J Endodontol 1986;12:382-7.

20. Brandal JL, Nicholls JI, Harrington GW. A comparison of three restorativetechniques for endodontically treated anterior teeth. J Prosthet Dent1987;58:161-5.

21. Marshak BL, Shemen BB, Cardash HS. Use of a special matrix system forconstructing amalgam and composite cores. J Prosthet Dent 1987;57:21-2.

22. Taleghani M, Morgan RW. Reconstructive materials for endodonticallytreated teeth. J Prosthet Dent 1987;57:446-9.

23. Attin T, Buchalla W, Kielbassa AM, Hellwig E. Curing shrinkage andvolumetric changes of resin-modified glass ionomer restorative materials.Dent Mater 1995;11:359-62.

24. Engelman MJ. Core materials. J Calif Dent Assoc 1988;16:41-5.25. Lepe X, Bales DJ, Johnson GH. Tensile dislodgment evaluation of two

experimental prefabricated post systems. Oper Dent 1996;21:209-12.26. Burgess JO, Summit JB, Robbins JW. The resistance to tensile,

compression and torsional forces provided by four post systems. JProsthet Dent 1992;68:899-903.

88


27. Cohen BI, Condos S, Musikant BL, Deutsch AS. Retentive properties ofthreaded split-shaft posts with titanium-reinforced composite cement. JProsthet Dent 1992;68:910-2.

28. Levartovsky S, Goldstein GR, Georgescu M. Shear bond strength ofseveral new core materials. J Prosthet Dent 1996;75:154-8.

29. Stockton LW, Williams PT. Retention and shear bond strength of two postsystems. Oper Dent 1999;24:210-6.

30. Stockton LW, Williams PT, Clarke CT. Post retention and post/core shearbond strength of four post systems. Oper Dent 2000;25:441-7.

31. Wadhwani CP. A single visit, multidisciplinary approach to themanagement of traumatic tooth crown fracture. Br Dent J 2000;188:593-8.

32. Huysmans MCDNJM, Van der Varst PGT, Schäfer R, Peters MCRB,Plasschaert AMJ, Soltesz U. Fatigue behaviour of direct post-and-corerestored premolars. J Dent Res 1992;71:1145-50.

33. Hemmings KW, King PA, Setchell DJ. Resistance to torsional forces ofvarious post and core designs. J Prosthet Dent 1991;66:325-9.

34. Cohen BI, Pagnillo M, Condos S, Deutsch AS. Comparison of the torsionalforces at failure for seven endodontic post systems. J Prosthet Dent1995;74:350-7.

35. Cohen BI, Pagnillo MK, Newman I, Musikant BL, Deutsch AS. Effects ofthree bonding systems on the torsional resistance of titanium-reinforcedcomposite cores supported by two post designs. J Prosthet Dent1999;81:678-83.

36. Cohen BI, Pagnillo MK, Musikant BL, Deutsch AS, Cofrancesco G. Pilotstudy of the effects of three bonding systems on the torsional resistance ofa titanium-reinforced composite core. J Prosthet Dent 1999;82:277-80.

37. Cohen BI, Penugonda B, Pagnillo MK, Schulman A, Hittelman E. Torsionalresistance of crowns cemented to composite cores involving threestainless steel endodontic post designs. J Prosthet Dent 2000;84:38-42.

38. Peutzfeldt A, Asmussen E. Silicoating. Evaluation of a new method ofbonding composite resin to metal. Scand J Dent Res 1988;96:171-6.

39. Vojvodic D, Predanic-Gasparac H, Brkic H, Celebic A. The bond strengthof polymers and metal surfaces using the 'silicoater' technique. J OralRehabil 1995;22:493-9.

40. Özcan M, Pfeiffer P, Nergiz I. A brief history and current status ofmetal/ceramic surface conditioning concepts for resin bonding in dentistry.Quintessence Int 1998;29:713-24.

41. Moulin P, Degrange M, Picard B. Influence of surface treatment onadherence energy of alloys used in bonded prosthetics. J Oral Rehabil1999;26:413-21.

89


42. Mazurat RD, Pesun S. Resin-metal bonding systems: a review of theSilicoating and Kevloc systems. J Can Dent Assoc 1998;64:503-7.

43. Pesun S, Mazurat RD. Bond strength of acrylic resin to cobalt-chromiumalloy treated with the Silicoater MD and Kevloc systems. J Can Dent Assoc1998;64:798-802.

44. Vojvodic D, Jerolimov V, Celebic A, Catovic A. Bond strengths of silicoatedand acrylic resin bonding systems to metal. J Prosthet Dent 1999;81:1-6.

45. Watanabe I, Kurtz KS, Kabcenell JL, Okabe T. Effect of sandblasting andsilicoating on bond strength of polymer-glass composite to cast titanium. JProsthet Dent 1999; 82:462-7.

46. Söderholm KJ, Zigan M, Ragan M, Fischlschweiger W, Bergman M.Hydrofluoric degradation of dental composites. J Dent Res 1984;63:1248-54.

47. Moulin P, Picard B, Degrange M. Water resistance of resin-bonded jointswith time related to alloy surface treatment. J Dent 1999;27:79-87.

48. Kawano F, Ohguri T, Ichikawa T, Matsumoto N. Influence of thermal cyclesin water on flexural strength of laboratory-processed composite resin. JOral Rehabil 2001;28:703-7.

49. Magneville B, Dejou J. A comparison of two methods of adheringcomposite to metal. J Prosthet Dent 1996;76:97-101.

50. Kourtis SG. Bond strengths of resin-to-metal bonding systems. J ProsthetDent 1997;78:136-45.

51. Mukai M, Fukui H, Hasegawa J. Relationship between sandblasting andcomposite resin-alloy bond strength by a silica coating. J Prosthet Dent1995;74:151-5.

52. Stoknorm R, Isidor F, Ravnholt G. Tensile bond strength of resin lutingcement to a porcelain-fusing noble alloy. Int J Prosthodont 1996;9:323-30.

53. Na Badalung DP, Powers JM, Connelly ME. Comparison of bond strengthsof denture base resins to nickel-chromium-beryllium removable partialdenture alloy. J Prosthet Dent 1997;78:566-73.

54. Hansson O. Strength of bond with Cospan Opaque to three silicoatedalloys and titanium. Scand J Dent Res 1990;98:248-56.

55. May KB, Fox J, Razzoog ME, lang BR. Silane to enhance the bondbetween polymethyl methacrylate and titanium. J Prosthet Dent1995;73:428-31.

56. DIN 17850. Titan Chemische Zusammensetzung (Titanium ChemicalComposition), Berlin, 1990-11.

57. International Organization for Standardization. Dentistry-polymer-basedcrown and bridge materials, Amendment 1996;ISO 10477.

58. Kern M, Thompson VP. Durability of resin bonds to pure titanium. JProsthodont 1995;4:16-22.

90


ChapterBonding Polycarbonate Brackets to Ceramic:

Effects of Substrate Treatment on Bond Strength

This chapter is in press for publication as:Özcan, M., Vallittu, P.K., Peltomäki, T., Huysmans, M-Ch., Kalk, W.: BondingPolycarbonate Brackets to Ceramic: Effects of Substrate Treatment on BondStrength. American Journal of Orthodontics and Dentofacial Orthopedics, (inpress, 2003). (reproduced with permission of Mosby Inc.)

91

4


92


Bonding polycarbonate brackets to ceramic: Effects

of substrate treatment on bond strength

M. Özcana, P.K. Vallittub, T. Peltomäkic, M-Ch. Huysmansa, W. Kalka

aUniversity of Groningen, Faculty of Medical Sciences, Department of Dentistry and Dental

Hygiene, Groningen, The Netherlands.bUniversity of Turku, Institute of Dentistry, Department of Prosthodontics and Biomaterials

Research, Turku, Finland.cUniversity of Turku, Institute of Dentistry, Department of Oral Development and Orhodontics,

Turku, Finland.

ABSTRACTThis study evaluated the effect of five different surface conditioning methodson the bond strength of polycarbonate brackets bonded to ceramic surfaceswith resin based cement. Six disc-shaped ceramic specimens (feldspathicporcelain) with glazed surfaces were used for each group. The specimens ineach group were randomly assigned to one of the following treatmentconditions of the ceramic surface: (1) Orthophosphoric acid+primer+bondingagent, (2) Hydrofluoric (HF) acid gel+primer+bonding agent, (3) Tribochemicalsilica coating (SiOx, 30µm)+silane, (4) Airborne particle abrasion (Al2O3,30µm)+silane, (5) Airborne particle abrasion (Al2O3, 30µm)+silane+bondingagent. Brackets were bonded to the conditioned ceramic specimens using alight-polymerized resin composite. All specimens were stored in water for 1week at 37ºC and then thermocycled (1000 cycles, 5ºC-55ºC, 30 s). The shearbond strength values were measured at a universal testing machine at acrosshead speed of 1 mm/min. Silica coating together with silanization (13.6MPa) exhibited significantly higher (P=0.01) results than those oforthophosphoric acid (8.5 MPa). There was no significant difference (P=0.97)between the bond strengths obtained after airborne particle abrasion withaluminium trioxide particles followed by silanization (12 MPa) and HF acidapplication (11.2 MPa) (ANOVA and Tukey`s test). While orthophosphoric acidconditioning exhibited only adhesive failures of the luting cement from theceramic surface, other conditioning methods showed mixed type of failures.Airborne particle abrasion with aluminium trioxide or silica coating followed bysilanization demonstrated the most favourable bond strengths. The type offailures observed after debonding indicated that the critical parameter was thestrength of the adhesive joint of the luting cement to both the bracket and theceramic.

93


INTRODUCTION

With the increased demand for adult orthodontics, the clinician is often facedwith the problem of luting brackets and retainer wires to metal-ceramic fixedpartial dentures, crowns, veneers or even full ceramic restorations.1-4

Recently, more esthetic and relatively invisible brackets have gained popularityin orthodontics. These brackets are generally either polycarbonate or ceramicand the latter are either mono or polycrystalline. Ceramic brackets can have anepoxy or a polycarbonate base. The bases of polycarbonate brackets arereported to be sufficiently flexible to allow plastic deformation at levels similarto that of metal brackets.5-7 However, lack of durable bonding between thebrackets and ceramic restorations is still a major problem in adultorthodontics.8,9 Because bonding in orthodontics is semi-permanent in nature,bond strength should be high enough to resist accidental debonding during thewhole course of treatment but also low enough so that excessive force neednot be applied during debonding at the end of the treatment.10

In order to achieve sufficient bond strength of brackets to ceramics,pretreatment of the ceramic surfaces is a prerequisite. Numerous options havebeen suggested that were generally combinations of various mechanical andchemical conditioning methods.11-15 Roughening of the surface is generallyregarded as compulsory for reliable bond strength.8,9,15 Etching the ceramicsurfaces with either orthophosphoric or HF acid followed by the application ofprimer and bonding agent are recommended methods.16,17 The use of HFacid in the laboratory may be recommended but it is hazardous in clinical use.However, the failure rate on ceramic surfaces is still reported to be as high as9.8% even when HF acid is used followed by silane application.4

Advances in silane coupling agents during the last two decades appear toenhance the bond strength by promoting a chemical bond between the resincomposite and the ceramic. Silane molecules after being hydrolized to silanol,can form polysiloxane network or hydroxyl groups to cover the silica surface.Monomeric ends of silane molecules react with the methacrylate groups of theadhesive resins by free radical polymerization. One system, in which silanesare also used, is tribochemical silica coating that provides ultrafine mechanicalretention by airborne particle abrasion. The surfaces are abraded with 30 µmgrain size aluminium trioxide modified with silisic acid. The blasting pressureresults in embedding of silica particles on the surface rendering the surfacechemically more reactive to resin via silane. Silica coating has been tried indifferent applications in dentistry.18-20 Although the literature contains manyreports on bonding of brackets to ceramic surfaces, this method has not beenwidely investigated for bracket bonding purposes.21

94


Review of relevant literature also indicates that little research has been carriedout to quantify the bond strength of polycarbonate brackets.22-25 Due to theinert polycarbonate matrix of plastic brackets, adequate adhesion is difficult toachieve.26,27 The manufacturing method of such brackets is based on injectionmolding, where the dispersion of fillers in a polymer is forced into the mold atcontrolled temperature. The fillers added to each plastic bracket are fairly welldistributed with a limited part of the filler exposed on the bracket surface. Inview of this component information, and since silanes are frequently used ascoupling agents to bond glass fillers to polymers, it was hypothesized that ifsilica coating was as efficient as it was reported in other dental applications,its use especially on the ceramic parts of restorative materials could beconsidered as a conditioning method prior to bracket bonding.

The objectives of this study were twofold, namely to identify the effect offive different surface conditioning methods on the shear bond strength ofpolycarbonate brackets bonded with a resin composite luting cement toceramic surfaces that simulate ceramic restorations and to evaluate the failuretypes after debonding.

MATERIALS AND METHODS

Thirty feldspathic ceramic specimens with a diameter of 6 mm and a thicknessof 2 mm (Shade A3, VMK68, Vita Zahnfabrik, Bad Säckingen, Germany) werefabricated with glazed surfaces in disc forms according to the manufacturer`srecommendations. Six specimens were used for each experimental group. Fivesurface conditioning methods were assessed for the ceramic materials afterwater storage and thermocycling. Table 1 summarizes the characteristics ofsurface conditioning methods and manufacturing company names. Beforeinitiating the bonding procedure, the specimens were embedded in acrylicresin blocks ensuring that one surface of the disc remained uncovered forbonding procedures. The exposed surface of each specimen was cleaned for10 min in an ultrasonic bath (Quantrex 90 WT, L&R Manufacturing, Inc.,Kearny, NJ, USA) containing ethylacetate and then air-dried. Subsequently, thespecimens were randomly assigned to one of the following five conditioningmethods:

95


Table 1. Characteristics of five surface conditioning methods to ceramic.

Conditioning principle Manufacturer

Method 1

Orthophosphoric acid (37%, 60 s) Ultradent® Ultraetch, South Jordan, USA

Primer Ormco, Glendora, USA

Bonding agent Transbond XT, 3M, USA

Method 2

Hydrofluoric acid (9.5%, 90 s) Ultradent Porcelain Etch, South Jordan, USA

Primer Ormco, Glendora, USA

Bonding agent Transbond XT, 3M, USA

Method 3

Air particle abrasion (30 µm aluminium trioxide, Korox, Bego, Bremen, Germany

250 kPa, 10 mm, 4 s)

Silane coupling agent 5 min ESPE-Sil, 3M ESPE AG, Seefeld, Germany

Method 4

Tribochemical silicacoating (CoJet®-Sand, 30 µm SiOx 3M ESPE AG, Seefeld, Germany

250 kPa, 10 mm, 4 s),


Method 5

Air particle abrasion (30 µm aluminium trioxide, Korox, Bego, Bremen, Germany

250 kPa, 10 mm, 4 s)


Bonding agent Transbond XT, 3M, US


In the first group, the ceramic substrates were etched with 37%orthophosphoric acid for 60 s. In the second group, 9.5% HF acid gel wasapplied for 90 s and rinsed. After both etching procedures, the substrates werewashed and rinsed thoroughly to remove the residual acid and then air-dried.

In the third group, silica coating process was achieved using an intraoralair abrasion device (Dento-PrepTM, RØNVIG A/S, Daugaard, Denmark) filledwith 30 µm SiOx (CoJet®-Sand), from a distance of approximately 10 mm ata pressure of 250 kPa bar for 4 s.

In the fourth and fifth groups, airborne particle abrasion was performed

96


using aluminium trioxide particles in the same device under the sameconditions with silica coating.

Bonding procedure

Throughout the experiments, the bonding procedures were carried out by thesame operator (M.Ö.) in accordance with the manufacturer`s instructions. Inthe acid etched groups, following the application of primer, bonding agent wasapplied a thin layer, excess resin was removed with air and it was lightpolymerized (Elipar, 3M ESPE) for 20 s according to the manufacturer`sinstructions. In the third and fourth conditioning groups, only silane couplingagent was applied and allowed to evaporate for 5 min. In the fifth group,bonding agent was applied in addition to aluminium trioxide and silaneapplication.

A total of 30 polycarbonate brackets designed for mandibular incisors(Ormco Spirit MB, Glendora, USA) were bonded to each conditioned ceramicsurface with a light polymerized resin composite luting cement (3M Unitek,Monrovia, CA, USA). The surface area for the bracket base was 10 mm2

according to the manufacturer. The bracket was placed onto the ceramicsurface using a bracket plier under manual control. Before light polymerization,excess resin was removed from the bracket periphery and polymerization ofthe luting resin was performed for 40 s from two directions. Light-intensity was800 mW/cm2. The irradiation distance between the exit window and the resinsurface was maintained at 10 mm to obtain adequate polymerization.

Storage and testing procedure

The specimens were first stored in water for 1 week at 37ºC and thensubjected to thermocycling (custom made by NIOM-Scandinavian Institute forDental Materials, Haslum, Norway) for 1,000 cycles between 5ºC and 55ºC indeionised grade 3 water. The dwelling time at each temperature was 30 s.Transfer time from one bath to the other was 2 s.

Specimens were mounted in a jig (Bencor Multi-T shear assembly,Danville Engineering Inc., San Ramon, CA) of the universal testing machine(Lloyd LRX, Lloyd Instruments Ltd, Fareham, UK) and a shear force wasapplied to the adhesive interface until fracture occurred (Fig.1).The specimenswere loaded at a crosshead speed of 1.0 mm/min.

After debonding, the fracture sites were examined visually to determine

97


the location and the manner of the failure and classified according to themodified Adhesive Remnant Index (ARI) system of Årtun and Bergland.28,29

Ceramic specimens and the bracket bases were further examined underscanning electron microscopy (SEM; Cambridge Instruments Streoscan,Electron Microscopy Ltd) after debonding.

Statistical analysis was performed using SAS System for Windows,release 8.02/2001 (Cary, NC, USA). Shapiro Wilk`s test was used to test thenormality of the data. The means of each group were analysed by one-wayanalysis of variance (ANOVA) determine the significant differences betweenthe conditioning methods and Tukey`s test was used for multiple comparisons.P values less than 0.05 are considered to be statistically significant.

RESULTS

The results of the shear bond strength test for five conditioning methods arepresented in Fig 2. Mean bond strengths for each group along with theirminimum, maximum values, standard deviations and standard errors areshown in Table 2. The data were normally distributed and variances wereequal in each five groups. ANOVA showed significant influence of theconditioning methods on the bond strength values (P=0.03).

The highest bond strength of 13.6 MPa was obtained with silica coatingand silanization that was significantly higher than with phosphoric acidtreatment (8.5 MPa) that exhibited the lowest bond strengths (P=0.01).

98

Fig. 1. Specimen mounted in the jig of the

universal testing machine and the bracket

being submitted to a shear force.


Bonding agent application did not increase the bond strength (10.9 MPa) whencompared with the group where only silane was applied after airborne particleabrasion with aluminium trioxide (12 MPa) (P=0.93). There were no significantdifferences (P=0.40) between HF acid (11.2 MPa) and phosphoric acid gel (8.5MPa) treated group.

Table 3 displays a tabulation of the modes of failure for the brackets afterdifferent surface conditioning methods. In the phosphoric acid etched group,the brackets failed mainly at the ceramic/resin interface with all of the lutingcement remaining on the bracket base. In the HF acid treated group,predominantly less than half of the composite was left on the ceramic surfaceafter debonding. In both airborne particle abraded groups, more than half ofthe luting cement was left adhered to the ceramic surface and the bracketbase. On the contrary, in the silica coated group, luting cement was mainlydebonded from the bracket base being left adhered to the ceramic surfaceswith distinct impression of bracket mesh (Fig 3a-d).

No fracture in the body of the bracket was observed in any of the groupstested, but two incidents of ceramic fractures after silica coating, one after HFacid and one after air particle abrasion+silane treated group wereexperienced.

99

0

2

4

6

8

10

12

14

16

18

Bo

nd

Str

eng

th (

MP

a)

Phosphoric acid + primer +b.agent

HF (9.5%) + primer +b.agent

SiOx + silane

Al2O3 + silane

Al2O3 + silane + b.agent

Fig. 2. Shear bond strengths after surface conditioning methods in thermocycled conditions.

Vertical lines represent the standard deviations.


Table 2. Mean shear bond strength (X), minimum (Min), maximum (Max) values, standard deviations (SD), standard errors (SE)

and number of specimens (N) for each group.

Groups N X Min Max SD SE

Group 1 6 8.5 3.5 11.2 2.8 1.15364

Group 2 6 11.2 7.8 12.8 2.3 0.91860

Group 3 6 13.6 10.2 17.7 2.2 0.89329

Group 4 6 12 8.7 15.6 2.9 1.22155

Group 5 6 10.9 6.6 15.4 2.8 1.16226

Table 3. Modes of failure of polycarbonate brackets bonded with light-polymerized resin composite luting cement to feldspathic

porcelain after various surface treatments. All samples were stored in water for 1 week at 37ºC and then subjected to

thermocycling for 1.000 cycles between 5ºC and 55ºC.

Adhesive Remnant Index Score

Dislodged† 0 1 2 3 Ceramic fracture

Orthophosphoric acid+primer+bonding agent - 6 - - - -

Hydrofluoric acid gel+primer+bonding agent - 1 4 - - 1

Tribochemical silicacoating+silane - - - - 4 2

Airborne particle abrasion+silane - - - 5 - 1

Airborne particle abrasion+silane+bonding agent - 1 - 5 - -

*Adhesive Remnant Index (ARI)

score 0= no composite left on ceramic surface.

score 1= less than half of composite left.

score 2= more than half of composite left.

score 3= all composite left on ceramic surface, with distinct impression of bracket mesh.

†During thermocycling or testing

100


DISCUSSION

Bonding of brackets to the ceramic substrates has two adhesion aspects:adhesion of the luting cement to the conditioned ceramic substrate andadhesion of the luting cement to the bracket base. In fact, there are a numberof other factors that might also influence bond strength such as thearchitecture of the bracket base, the composition and elastic modulus of theadhesive. In this study, brackets for the mandibular incisors were used due totheir flat bases that ensure optimal adaptation to the ceramic surface. Thebracket base used relied on entirely mechanical interlocking principle. No

101

Fig. 3a Fig. 3b

Fig. 3a-d. SEM micrograph of a) the base of a debonded bracket after orthophosphoric acid

treatment. Note that the ceramic surface is devoid of the composite luting cement. b) the base of

a debonded bracket after airborne particle abrasion+silane. Note that more than half of the resin

is left adhering to the bracket base. c) the base of a debonded bracket after airborne particle

abrasion+silane+bonding agent. d) the ceramic surface following debonding after silica coating

and silanization. Note that all resin luting cement is left adhering to the ceramic with distinct

impression of bracket mesh.

Fig. 3c Fig. 3d


conditioning or adhesive was used on the bracket base.It has been frequently suggested that clinically adequate bond strength for

a metal orthodontic bracket to enamel should be between 6 to 8 MPa.2,3,9,10

The mean shear bond strengths of polycarbonate brackets to ceramic surfacesachieved in this study do fall within this range or exceed these limits andtherefore could be considered sufficient for clinical applications.

Among the conditioning methods tested, conventional acid-etching with37% orthophosphoric acid revealed the lowest values. It has been previouslyproved that phosphoric acid is relatively ineffective for providing mechanicalretention on ceramics when they are used in combination with the resincomposite.9 Although in the phosphoric acid etched group, adhesive failures ofthe luting agent from the ceramic surface was experienced after debonding,the bond strength was higher than it was reported. The possible contribution ofthe primer and the bonding agent may have had an impact on the results. Thismay also be due to the differences in luting agent used. The least favourablebond results were obtained with 8.5 MPa in this group but the type of failureswas desirable.

HF acid is well recognized to have hazardous effects in vivo since it wasfound to be a harmful and irritating compound for soft tissues. Nevertheless, itsefficiency in improving the bracket bonding on ceramics has been widelyaccepted.8,17,27 When the two acid agents were compared, higher bondvalues were obtained in the HF acid treated group. This indicated that HF acidwas more effective than phosphoric acid in dissolving the crystalline andglassy phase of the ceramic and therefore facilitates better micromechanicalretention however the results were not statistically significant.

In this investigation, no statistically significant differences were found inthe bond strengths between the two aluminium trioxide particle abraded groupfollowed by either only silane or bonding agent plus silane application. In arecent study on the survival rate of the brackets on ceramic surfaces,Zachrisson postulated that air-particle abrasion did not provide clinicallyacceptable bond strengths when it was used in combination with a chemicallycured resin composite luting cement.4 In the case of aluminium trioxidetreatment, the silica layer is missing and therefore although this bond may beconsidered clinically sufficient, it should require constant monitoring.

The recommended duration for air abrasion application was 2 to 4 s.5

Since it has been previously shown that variation within this time interval haslittle effect on the bond strength,8,15 in this study, air abrasion was applied for4 s. As the restorations generally remain in the mouth after debonding thebrackets, damage to the ceramic due to extreme roughening of the surfacesduring the pretreatment should be avoided. Longer application time might

102


result in more material loss from the ceramic surface creating more roughnessthan it is desired. Therefore, the duration of air abrasion still remains to beinvestigated.

Tribochemical silica coating followed by silanization evidently enhancedthe bond between the ceramic surfaces and the luting cement. The silica layerleft by silica coating on the ceramic surface provides a basis for silane. In theceramic composite bond, silane functions as a coupling agent, which adsorbsonto and alters the surface of the ceramic, thereby facilitating interaction.31,32

One important requirement in bracket bond is that there should be no orminimal risk of iatrogenic damage to the ceramic surface during debonding.Failures at the bracket/composite interface could be expected to be morefavourable. No matter how one applies the load on a configuration that consistsof two different materials, the stresses at the interface will not be uniform. Inthe orthophosphoric acid etched group, only adhesive failures of the lutingcement from the ceramic occurred. This demonstrates a weak connectionbetween the resin composite and the ceramic. In order to avoid ceramicfractures at the time of debonding, Cochran9 suggested solely etching theceramic with HF acid, however there was also one incident of ceramic fracturein this group.

In general, increased bond strength resulted in failures within the resinsuch that some resin was left on both the bracket and the ceramic surfaces.This was the typical case in air particle abraded groups. When bonding agentwas used after air particle abrasion, notably there were no cohesive failureswithin the ceramic. Therefore, the cohesive failure occurred within the resinwith all or most of the resin remaining on both the ceramic and the bracketbase indicated the influence of the cohesive strength of the resin lutingcement.

Only in the silica coated and silanized group after debonding the resin waspredominantly left adhered on the ceramic surface. This type of failure in theadhesive-bracket interface reveals the fact that the chemical bonding wasequal to or exceeded the mechanical retention provided by the bracket baseand the bond strength to the ceramic surface was greater than the cohesivestrength of the luting resin. When some remnants of the luting agent is left onthe ceramic surface, clinicians need to polish the ceramic surface.

Some studies reported that the use of silanes without removing the glazefrom the ceramic surface resulted in the least damage to surface withacceptable bond strength values.1,25,33 In this study, no additional attempt wasmade to remove the glaze. This study also did not deal with conditioning thebracket base but this may be important for clinical performance.

Although in many in vitro studies, high bond strengths were obtained, in

103


vivo failures of debonding are still being reported at a considerable rate.4 Onereason for this could be the differences between the experimental set-up invarious in vitro studies. Durability evaluation in wet environment is necessaryto judge the performance of adhesive interface. Most of the earlier studieshave included neither long-term water storage, nor thermocycling regimensbefore testing.2,7,9,30,34 This makes it difficult to make a direct comparisonbetween previous studies and the present one in which 1000 thermocyclingwas applied. When no or limited thermocycling was performed, high bondstrengths may be found that may not correspond to the chair side experiences.Such studies might also offer figures that misinterpret the long-term in vivobond strength and therefore should be evaluated with caution. If the specimenswere only stored in water without thermocycling, the bond strengths toceramics, as well as the incidence of cohesive ceramic fractures were found tobe excessively high.2,11,27,35 Therefore the clinical relevance of some of theprevious studies appears limited. Since controversial findings exist in theliterature, whether longer periods of thermocycling regimens are neededremains a matter of discussion.

Bond strengths are influenced by several other factors one of which is theluting cement type. In many of the studies, a chemically polymerized, largefiller particle adhesive, namely Concise was used as a luting resin incombination with metal brackets.4,8,12,16 In contrast to metal brackets, it ispossible to bond the polycarbonate brackets with light polymerized resincomposites that exhibit markedly less porosity than chemically polymerizedresins. In this study, heavily filled luting cement with inorganic filler contentvarying from 67-80% in weight was used. Further investigations could also beaddressed with other resins.

The present study did not find an ideal conditioning method withoutlimitations. Because many factors affect the bond strengths of brackets,caution must be exercised to understand the characteristics of the surfaceconditioning methods in accordance with the cements and brackets to bechosen when comparing in vitro studies and furthermore when extrapolatingthe in vitro measurement to the clinics.

Recently, alternative methods for conditioning ceramics have beendeveloped, that may become important replacements of the conventionalmethods. There are however insufficient clinical data available at this time topredict the clinical performance, which also cannot be interpreted from in vitrostudies.

104


CONCLUSIONS

Within the limitations of the present study, the following conclusions weremade:1. Bond strengths of the polycarbonate brackets luted with resin composite

cement tested on the dental ceramics after surface conditioningtechniques varied in accordance with the conditioning methods.

2. By using air particle abrasion with aluminium trioxide for the ceramicsurfaces, the critical parameter was found to be the strength of adhesivejoint of the luting cement to the bracket rather than to the ceramic.

3. The air-particle abrasion either with SiOx or aluminium trioxide togetherwith silanization eliminated the need for acid etching, primer and/orbonding agent applications with sufficient bond strengths.

4. The use of HF acid would be still appropriate for orthodontic reasons, if thecritical aspect was accepted to use this chemical agent intraorally.

Acknowledgements

We express our appreciation to the Ortomat Herpola Ltd in Turku for supplying the brackets and

the adhesives for this study. The authors are also grateful to Tero Vahlberg, M.Sc., Department of

Biostatistics, University of Turku, for his assistance with statistical analysis.

REFERENCES

1. Kao EC, Johnston WM. Fracture incidence on debonding of orthodonticbrackets from porcelain veneer laminates. J Prosthet Dent 1991;66:631-7.

2. Gillis I, Redlich M. The effect of different porcelain conditioning techniqueson shear bond strength of stainless steel brackets. Am J Orthod DentofacOrthop 1998;114:387-92.

3. Bourke BM, Rock WP. Factors affecting the shear bond strength oforthodontic brackets to porcelain. Brit J Orthodont 1999;26:285-90.

4. Zachrisson BU. Orthodontic bonding to artificial tooth surfaces: Clinicalversus laboratory findings. Am J Orthod Dentofac Orthop 2000;117:592-4.

5. Miura F, Nakagawa K, Masuhara E. New direct bonding system for plasticbrackets. Am J Orthod 1971;59(4):350-61.

6. Zachrisson BU. Bonding in Orthodontics. In: Graber TM, Swain BF (eds)Orthodontics. Current principles and techniques, 3rd Edn. CV Mosby, StLouis. 1985, pp. 5-59 and p.485.

105


7. Huang TH, Kao CT. The shear bond strength of composite brackets onporcelain teeth. Eur J Orthod. 2001;23(4):433-9.

8. Zachrisson BU, Büyükyilmaz T. Recent advances in bonding to gold,amalgam and porcelain. J Clin Orthodont 1993;27:661-75.

9. Cochran D, O`Keefe KL, Turner DT, Powers JM. Bond strength oforthodontic composite cement to treated porcelain. Am J Orthod DentofacOrthop 1997;111:297-300.

10. Reynolds IR, von Fraunhofer JA. Direct bonding in orthodonticattachments to teeth: The relation of adhesive bond strength to gauzemesh size. Brit J Orthodont 1975;3:91-5.

11. Newman SM, Dresseler KB, Grenadier MR. Direct bonding of orthodonticbrackets to esthetic restorative materials using a silane. Am J OrthodDentofac Orthop 1984;86:503-6.

12. Eustaquio R, Garner LD, Moore BK. Comparative tensile strengths ofbrackets bonded to porcelain with orthodontic adhesive and porcelainrepair systems. Am J Orthod Dentofac Orthop 1988;94:421-5.

13. Kupiec KA, Wuertz KM, Barkmeier WW, Wilwerding TM. Evaluation ofporcelain surface treatments and agents for composite-to-porcelain repair.J Prosthet Dent 1996;76:119-24.

14. Özcan M, Pfeiffer P, Nergiz I. A brief history and current status of metal-ceramic surface conditioning concepts for resin bonding in dentistry.Quintessence Int 1998;29:713-24.

15. Jost-Brinkmann PG, Böhme A. Shear bond strengths attained in vitro withlight-cured glass ionomers vs composite adhesives in bonding ceramicbrackets to metal or porcelain. J Adhes Dent 1999;1:243-53.

16. Gwinnett AJ.A comparison of shear bond strengths of metal and ceramicbrackets. Am J Orthod Dentofacial Orthop 1988;93:346-8.

17. Barbosa VLT, Almeida MA, Chevitarese O, Keith O. Direct bonding toporcelain. Am J Orthod Dentofac Orthop 1995;107:159-64.

18. Edelhoff D, Marx R, Spiekermann H, Yildirim M. Clinical use of an intraoralsilicoating technique. J Esthet Restor Dent 2001;13:350-6.

19. Özcan M, Niedermeier W. Clinical study on the reasons and location of thefailures of metal-ceramic restorations and survival of repairs. Int JProsthodont 2002;3:299-302.

20. Özcan M. The use of chairside silica coating for different dentalapplications. J Prosthet Dent. 2002;87:469-72.

21. Schmage P, Nergiz I, Herrmann W, Özcan M. Influence of various surfaceconditioning methods on the bond strength of metal brackets to ceramic.Am J Orthod Dentofacial Orthop 2003;123:540-546.

106


22. Buzzitta VA, Hallgren SE, Powers JM. Bond strength of orthodontic direct-bonding cement-bracket systems as studied in vitro. Am J Orthod1982;81:87-92.

23. Blalock KA, Powers JM. Retention capacity of the bracket bases of newesthetic orthodontic brackets. Am J Orthod Dentofacial Orthop1995;107:596-603.

24. Fernandez L, Canut JA. In vitro comparison of the retention capacity ofnew aesthetic brackets. Eur J Orthod 1999;21:71-7.

25. Lai PY, Woods MG, Tyas MJ. Bond strengths of orthodontic brackets torestorative resin composite surfaces. Aust Orthod J 1999;15:235-45.

26. Chaconas SJ, Caputo AA, Niu GS. Bond strength of ceramic brackets withvarious bonding systems. Angle Orthod 1991;61:35-42.

27. Fox NA, McCabe JF. An easily removable ceramic bracket? Br J Orthod1992;19:305-9.

28. Årtun J, Bergland S. Clinical trials with crystal growth conditioning as analternative to acid-etch enamel pretreatment. Am J Orthod 1984;85:333-40.

29. Zachrisson YØ, Zachrisson BU, Büyükyilmaz T. Surface preparation fororthodontic bonding to porcelain. Am J Orthod Dentofac Orthop1996;109:420-30.

30. Kocadereli I, Canay S, Akca K. Tensile bond strength of ceramicorthodontic brackets bonded to porcelain surfaces. Am J Orthod DentofacOrthop 2001;119:617-20.

31. Özcan M, Alkumru HN, Gemalmaz D. The effect of surface treatment onthe shear bond strength of luting cement to a glass-infiltrated aluminaceramic. Int J Prosthodont 2001;14:335-9.

32. Peutzfeldt A, Asmussen E. Silicoating. Evaluation of a new method ofbonding composite resin to metal. Scand J Dent Res 1988;96:171-6.

33. Nebbe B, Stein E. Orthodontic brackets bonded to glazed and deglazedporcelain surfaces. Am J Orthod Dentofac Orthop 1996;109:431-6.

34. Chung CH, Brendlinger EJ, Brendlinger DL, Bernal V, Mante FK. Shearbond strengths of two resin-modified glass ionomer cements to porcelain.Am J Orthod Dentofacial Orthop 1999;115:533-5.

35. Smith GA, Mclnnes-Ledoux P, Ledoux WR, Weinberg R. Orhodonticbonding to porcelain - Bond strength and refinishing. Am J OrthodDentofac Orthop 1988;94:245-52.

107


108


ChapterEffect of Three Surface Conditioning Methods to

Improve Bond Strength of Particulate Filler Resin

Composites

This chapter is submitted for publication in Journal of Materials Science:Materials in Medicine as:Özcan, M., Alander, P., Vallittu, P.K., Huysmans, M-Ch., Kalk, W.: Effect ofThree Surface Conditioning Methods to Improve Bond Strength of ParticulateFiller Resin Composites. Journal of Materials Science: Materials in Medicine,(submitted, 2003).

109

5


110


Effect of Three Surface Conditioning Methods to

Improve Bond Strength of Particulate Filler Resin

Composites

M. Özcana, P.K. Vallittub, P. Alanderb, M-Ch. Huysmansa, W. Kalka



Research, Turku, Finland.

ABSTRACT

The objective of this study was to evaluate the effect of three surfaceconditioning methods on the shear bond strength of a particulate filler resin-composite (PFC) to 5 PFC substrates. The specimens were randomlyassigned to one of the following surface conditioning methods: (1) Hydrofluoric(HF) acid gel (9.5%) etching, (2) Air-borne particle abrasion (50 µm Al2O3), (3)Silica coating (30 µm SiOx, CoJet®-Sand). After each conditioning method,silane coupling agent was applied. Adhesive resin was then applied a thin layerand light polymerized. The low-viscous diacrylate resin composite was bondedto the conditioned substrates using polyethylene molds. All specimens weretested at dry and thermocycled (6.000, 5 ºC-55 ºC, 30 s) conditions. One-wayANOVA showed significant influence of the surface conditioning methods (p <0.001), and the PFC types (p < 0.0001) on the shear bond strength values.Significant differences were observed in bond strength values between theacid etched specimens (5.7-14.3 MPa) and those treated with either air-borneparticle abrasion (13.0-22.5 MPa) or silica coating (25.5-41.8 MPa) in dryconditions (ANOVA, p < 0.001). After thermocycling, the silica coating processresulted in the highest bond values in all material groups (17.2-30.3 MPa).Keywords: Air-borne particle abrasion, fracture, hydrofluoric acid, repair,silanization, silica coating

1. Introduction

In today`s dentistry PFC materials, applied directly or indirectly, occupy aparamount position and they achieve acceptable longevity with much lowercost than their ceramic counterparts. PFC restorations, especially when used

111


as laminates, allow for minimally invasive preparations, or no preparations atall, for the replacement of missing dental tissues by means of basic layeringtechnique.

Novel dental polymers today generally consist of a monomeric matrixresin, silanated inorganic fillers, a polymerization initiator system, inhibitors forstorage stability, and pigmentation for shading. Although markedimprovements have been noted in terms of physical and mechanical propertiesduring the last 10-20 years, enzymes and alcohol present in the oral cavity, forinstance, can degrade the composite matrix [1-3]. Moreover composites areless stable in fluids and their degradation rate is higher in saliva simulatingconditions, depending on the chemical nature of the monomers, amount ofdimers and oligomers, the degree of cross-linking in the polymerized matrix,and other intraoral impact [4-6]. In addition, fatigue can accelerate the wearprocess in composite materials. All these factors provoke discoloration,degradation, microleakage, wear, ditching at the margins, delamination orsimply fracture being often experienced in clinical conditions, which in turn,may require repair or replacement of the restoration [7-11].

Repair as an alternative to complete removal, would preserve the tooth asit is often difficult to remove an adhesive restoration without removing anintegral part of the tooth or the restoration itself and thereby prolong theservice life of such restorations [12, 13].

A number of techniques have been proposed to improve composite repairstrengths through roughening, etching the substrate surface with acidulatedphosphate fluoride, HF acid gel, air-borne particle abrasion or using adhesiveresins [14-21]. While several researchers found that the surface roughness ofthe composite was an important factor in developing high repair strength [21,22], others reported that grinding or roughening of the bonding surfacedecreased the bond strength [15, 20, 23]. Despite the hazardous effects of HFacid gel, etching the surface of a composite restoration with this acid followedby the application of a silane coupling agent is a well-known andrecommended method to increase bond strength. Although HF acid was foundeffective in roughening the composite surface for bonding resin composite [24,25], neither etching with these solutions nor adding silane resulted in anadequate resin bond to some resin composites [26-29].

Recent developments in surface conditioning methods have resulted inimproved resin-to-resin bond strengths. One alternative has been introducedwith the use of silica coating and silanization [30]. Although comparativestudies showing the advantages of various types of surface conditioningmethods on different composites exist, there seems to be no consensus in theliterature regarding the best conditioning method for individual PFCs.

112


Therefore, the objectives of this study were to evaluate the effect of threesurface conditioning methods on the shear bond strength of a PFC to five PFCmaterials and to identify whether there exists an optimum method.

2. Materials and methods

Thirty-six specimens were made from five brands of PFC materials, namelyGradia (GRA), Sculpture (SCU), Sinfony (SIN), Targis (TAR), Tetric Ceram(TET) (total number of specimens=180). The composites were dispensed witha hand instrument into cylindrical (diameter: 6 mm, thickness: 2 mm) undercutcavities prepared in auto-polymerized poly(methylmethacrylate) (Palapress,Vario, Heraeus Kulzer) and polymerized incrementally according to eachmanufacturers` recomendations. Table I and II summarize the characteristicsof PFC types with codes and manufacturing company names.

TABLE I. Monomer matrix types and percentages of particulate filler composites with codes, and manufacturing company names.

Trade name Abbreviation Matrix type Manufacturer

Gradia GRA UEDMA/ ethylene dimethacrylate1 GC, USA

Sculpture SCU dimethacrylate2 Jeneric Pentron, USA

Sinfony SIN HEMA/diacrylate3 3M ESPE, Germany

Targis TAR Bis-GMA, DDDMA, UEDMA, TEGDMA4 Ivoclar Vivadent AG, Liechtenstein

Tetric Ceram TET Bis-GMA, UEDMA, TEGDMA5 Ivoclar Vivadent AG, Liechtenstein

Bis-GMA= Bis-phenol-A-glycidylmethacrylate

UEDMA= Urethane dimethacrylate

TEGDMA= Triethyleneglycol dimethacrylate

DDDMA= Decandiol dimethacrylate

HEMA=2-hydroxyethyl methacrylate

1 UDMA (10-25%) and ethylene dimethacrylate (5-10%)

2 dimethacrylate

3 10-30-% (octahydro-4,7-methano-1H-indenediyl) bis(methylene)diacrylate)

4 Bis-GMA (9%), DDMA (4,8%), UEDMA (9,3%)

5 Bis-GMA (<9%), TEGDMA (<5%), UEDMA (<8%)

113


TABLE II. Filler types and percentages of particulate filler composites with codes.

Trade name Abbreviation Filler type and content

Gradia GRA alumina silicate glass (40-50 w-%), amorphous precipitated silica (5-10 w-%)

Sculpture SCU glass-infiltrated alumina (70 w-%)

Sinfony SIN strontium-aluminium borosilicate glass, silicon oxide (50 w-%)

Targis TAR silanized barium glass fillers (46.2 w-%), highly dispersed silica (11.8 w-%), mixed

oxides (18.2 w-%), catalyst and stabilizers (0.6 w-%), pigments (≤ 0.1 w-%)

Tetric Ceram TET Silanated Ba-glass, ytterbium trifluoride, silanated metal oxide, silanated

barium-aluminum-fluoro-silicate glass, silanated silica glass (79 w-%)

The exposed surface of each specimen was ground finished to 1200 gritsilicon carbide abrasive (Struers RotoPol 11, Struers A/S) and cleaned for 10min in an ultrasonic bath (Quantrex 90 WT, L&R Manufacturing Inc.) containingdistilled water and air-dried. Subsequently, the substrates in each PFC group(n=6) were randomly assigned to each of the following three conditioningmethods:

2.1 Surface conditioning methods

Method 1: The substrates were etched with 9.5% HF acid gel (Ultradent®Porcelain Etch) for 90 s in accordance with the manufacturer`srecommendations.Method 2: Air-borne particle abrasion with 50 µm Al2O3 (Korox®, Bego) wasapplied using an intraoral air abrasion device (Dento-PrepTM, RØNVIG A/S)from a distance of approximately 10 mm at a pressure of 2.5 bars for 4 s.Method 3 was based on silica coating process that was achieved using thesame device under the same conditions but this time it was filled with 30 µmSiOx (CoJet®-Sand, 3M ESPE AG).The conditioned substrates were then coated with a 3-methacryloxypropyltrimethoxysilane coupling agent, �-MPS (ESPE®-Sil, 3MESPE AG) and waited for its reaction for 5 min. Intermediate monomer resin(IMR) (Schotchbond Multipurpose Adhesive, 3M Dental Products) was applieda thin layer and it was light-polymerized (Optilux 501, Kerr) for 10 s.

In an additional experiment, (n=15, 1/group) the conditioned surfaces ofthe five substrates were first gold sputtered and then examined using ascanning electron microscopy, SEM, (JSM-5500, Jeol Instruments).

114


2.2 Bonding procedures

The bonding procedures were carried out in accordance with themanufacturers` instructions by the same operator (M.Ö.) throughout theexperiments. The low-viscous diacrylate resin composite (Sinfony Dentin,Shade A2, 3M ESPE) was bonded onto the conditioned PFC substrates usingtranslucent polyethylene molds with inner diameter of 3.6 mm and height of 5mm. The resin composite was packed against the substrate with a composite-filling instrument and polymerized in a light-curing oven (Visio Beta Vario, 3MESPE) for 40 s. Polyethylene molds were gently removed from the testspecimens.

All groups of PFC/conditioning method combinations were randomlydivided into 2 groups (n=6) for dry and thermocycled storing conditions. Whiledry samples were kept in a dessicator at room temperature for 24 h prior totesting, the other groups were subjected to thermocycling (Thermocycler2000, Heto-Holten A/S) for 6.000 cycles between 5 ºC and 55 ºC in deionisedgrade 3 water. The dwelling time at each temperature was 30 s. The transfertime from one bath to the other was 2 s.

Specimens were mounted in a jig (Bencor Multi-T shear assembly,Danville Engineering Inc.) of the universal testing machine (Llyod LRX, LloydInstruments Ltd) and the shear force was applied to the adhesive interface untilfailure occurred. The specimens were loaded at a crosshead speed of 1.0mm/min and the stress-strain curve was analyzed with Nexygen 2.0 software(Llyod LRX, Lloyd Instruments Ltd).

Statistical analysis was performed using SAS System for Windows,release 8.02/2001 (SAS Institute Inc). P values less than 0.05 were consideredto be statistically significant in all tests.The differences in means of each groupwere analysed by analysis of variance (ANOVA) with shear bond strength asthe dependent variable, the surface conditioning methods and the PFC typesas the independent factors. Since the interaction between surface conditioningmethods and PFC types were statistically significant (two-way ANOVA, p <0.0001) in dry and thermocycled conditions, one-way ANOVA with multiplecomparisons using Tukey-Kramer adjustment test was used for furtheranalyses. Furthermore, two-sample t-test was used to determine thesignificant differences between dry and thermocycled conditions.

3. Results

The results of the shear bond strength test for HF acid etching, airborne

115


particle abrasion and silica coating are presented in Figs. 1a-c. One-wayanalysis of variance (ANOVA) showed significant influence of the surfaceconditioning methods (p < 0.0001), and PFC type on the bond strength values(p < 0.001). The differences in bond strength between storage conditions weresignificant except for TAR after air-borne particle abrasion (two-sample t-test,p < 0.05).

116

Acid etching

0

5

10

15

20

GRA SCU SIN TAR TET

Bo

nd

Str

eng

th (

MP

a)

Dry

TC

Air particle abrasion

05

1015202530

GRA SCU SIN TAR TETBo

nd

Str

eng

th (

MP

a)

Dry

TC

Silica coating

0

10

20

30

40

50

GRA SCU SIN TAR TET

Bo

nd

str

eng

th (

MP

a)

Dry

TC

Fig. 1 a-c. Shear bond strengths after a) Hydrofluoric acid etching, b) Air-borne particle

abrasion and c) Silica coating at dry and thermocycled conditions. Vertical lines represent the

standard deviations. For abbreviations, see Table I.

Fig. 1c

Fig. 1b

Fig. 1a


Conditioning the PFC substrates with the HF acid etching resulted in theleast favourable bond strength values amongst all conditioning methodsranging between 5.7 MPa and 14.3 MPa in dry conditions and between 3.3MPa and 9.5 MPa after thermocycling. In dry conditions after HF acid etching,there were significant differences between SIN vs TET (p = 0.047), TAR vs TET(p = 0.002). After thermocycling, there were significant differences betweenSIN vs TAR (p = 0.0007), SIN vs TET (p = 0.0002) in the HF acid etchedgroups.

In the air particle treated group, bond strengths increased significantlycompared to HF acid etching for GRA (13.0 MPa, p = 0.0007), SCU (15.7 MPa,p = 0.0001), SIN (22.5 MPa, p = 0.0001), TAR (20.0 MPa, p = 0.02) and TET(14.8 MPa, p = 0.009) in dry conditions and for SIN (18.0 MPa, p = 0.01), TAR(21.2 MPa, p < 0.0001) and TET (12.5 MPa, p < 0.0001 MPa) afterthermocycling.

Significantly higher (p < 0.0001), bond strengths were achieved aftersilica coating and silanization amongst all the conditioning methods for alltypes of PFCs ranging between 25.5 MPa and 41.8 MPa except for TAR afterair-borne particle abrasion vs silica coating (p = 0.17) in dry conditions.

SEM analysis, complementary to the shear bond strength tests, revealedthat HF acid gel dissolves the filler components of the PFCs and producesporous irregular surfaces (Fig. 2a-e). On the other hand, airborne particletreated groups either with Al2O3 or SiOx exhibited similar rough surfacescovered with abundant sand particles on the substrate surfaces (Fig. 3a-b).

117

Fig. 2a Fig. 2b


118

Fig. 2c Fig. 2d

Fig. 2e

Fig. 2 a-e. Typical SEM view of a) GRA, b) SCU, c) SIN, d) TAR, e) TET PFC substrates

exposed to 9.5% HF acid gel application for 90 s and rinsing. Note that the acid treatment

dissolved the filler components of the substrates (original magnification x 5000).

Fig. 3a Fig. 3b

Fig. 3 a-b. Typical SEM view of a) GRA and b) TET after airborne particle abrasion with alumina

and SiOx, respectively. Note that the surfaces were covered with abundant sand particles

(original magnification x 5000).


4. Discussion

The results of this study indicated that conditioning the substrates with HF acidgel adversely affected the morphological features of PFC substrates therebyresulting in poor repair strength when compared with other methods tested.

Usually inorganic fillers are integrated into the polymer matrix by silanecoupling agents, that form an interface between hydrophobic resin matrix andhydrophilic filler particles. In most situations, hydrolyzed �-MPS is used as acoupling agent for the fillers. Fluoride ion is implicated in depolymerizationreactions of matrix-filler interface. When PFC substrates are exposed to HFacid gel, water monolayer may penetrate via voids to fillers, that in turn, maydisorganize the siloxane network formed from the condensation ofintermolecular silanol groups, which is responsible for stabilizing the filler-resininterface [31]. All these mechanisms may weaken the particle-matrix interfacethat leads to filler dissolution. This phenomenon was evidently observed in theSEM analysis where a great portion of the fillers were detached from thematrix after they were exposed to HF acid etching.

One important aspect of filler erosion after HF acid gel treatment isdependent on the filler type. It has been reported that barium,boroaluminosilicate, silicate, strontium glass, and zinc glasses exhibitedextended degradation on acid attack, whereas quartz, silica, lithiumaluminosilicates and their mixtures showed less involvement [32, 33]. TheSEM findings revealed that more fillers were dissolved in highly filled TET (79w-%) that is composed of silica and barium fillers. A high filler contentadversely affects processing and on the other hand, too much cross-linking ofthe resins could embrittle the material that was observed with TET. Themorphologic and compositional changes in patterns obtained for the materialsafter etching are also dependent on the type of acid used as well as thecomposition of the restorative material. Although flourides with lowerconcentrations like 1.23% APF were used, similar findings were reported insome previous studies where APF was found to dissolve the fillers and causeddegradation [34]. Nevertheless, etching the PFCs with a 9.5% HF gel for 90 sresulted in variations in repair strength dependent on the composite material.This is in line with the study of Swift et al [24] where they found that etchingwith 9.5% HF acid gel for 30 s either increased or decreased the repairstrengths of composites. This finding together with ours could be explained onthe grounds of variations in matrix composition. All the monomers used incurrent composite techniques are organic esters of methyl methacrylatederivatives. Generally, organic esters in low pH undergo hydrolytic cleavage ofthe ester group.

119


Increased filler dissolution after HF acid conditioning might result inincreased surface area exposure of the resin matrix and consequently anaccelerated hydrolytic effect. This phenomenon was very evident in high filledPFCs with similar filler types (TAR, TET) when compared to a relatively lowfilled one (SIN).

Bond strength is dependent on unconverted C=C double bonds on theresin, which may be attributed to either to a low conversion rate or a highmatrix portion. The bond between the substrate and new resin is also based onunreacted C=C double bonds of the functional groups on the surface of polymermatrix. A high degree of conversion that resulted from the use of heat and lightused for polymerization causes improvement in mechanical strength andhardness and therefore makes the attachment of the new composite moredifficult. Controversy still exists regarding whether the degree of conversion iscompromised when PFCs are photo-activated [33]. However, it was alsoassumed that a certain percentage of unconverted C=C double bonds areavailable even after laboratory processing [35]. In this study, in dry conditionssatisfactory bond strengths were obtained according to ISO standards [36].Although the substrates were only ground finished, there might be someunreacted carbon bonds available on the surface. However to what extent thosetypes of unreacted bonding sites existed in the PFC materials, needs furtherclarification, namely using e.g. infrared spectroscopy for the surfacecharacterization.

In FTIR (Fourier transformation infrared spectroscopic) evaluations thathas been already published, it was found that the UEDMA/TEGDMA phase hada conversion rate of 70% and exhibited superior wear resistance, while the Bis-GMA/TEGDMA had a conversion rate of 55% [37]. The least favourable resultsobtained with GRA could be attributed to its high UEDMA matrix content. InPFCs with such matrix content, the possibility to obtain free radicalpolymerization bonding is low because of relatively small number of unreactedC=C double bonds on the polymer surface [38]. Monomer mixtures of Bis-GMAand TEGDMA give rise to polymers in which the quantity of remaining doublebonds increases with the content of Bis-GMA, without the mechanicalproperties being significantly effected [39].

An interesting result was achieved with the SIN composite, with octahydro-4,7-mathano-1H-indenediyl)bis(methylene)-diacrylate, in its monomer matrix. Itshowed high bond strengths, similar to those of Bis-GMA/TEGDMA PFCs, withalso less decrease after thermocycling. One reason for this may be related tothe function of IMR that may bond covalently to the pendent, unreactedmethacrylate groups. It has been reported earlier that swelling of the compositesubstrate surface with different solvents and the use of low-viscosity IMR

120


influence the bond between two composites [20, 38, 40]. The functions of IMRare to achieve better wetting of the substrate surface and to some degreedissolve and swell the polymer surface of the substrate. The hydrophilicadhesive joint in GRA, SCU, TAR, TET may be less rigid than the adhesive jointmade by hydrophobic monomer resin such as the main monomer of SIN.

After air-particle abrasion, either with Al2O3 or SiOx followed bysilanization, significant increase in bond strengths was noted. When comparingthe results of the acid etched groups with those of air-abraded ones, thefunction of silane coupling agent should not be disregarded. In this study, �-MPS was chosen because of the compatibility of the methacrylate moeity forcopolymerization with the PFC. Silane treatment also improves the wettability ofthe filler, affects its surface energy, hence its dispersion in the matrix. Howeverwhen little or no filler remains after HF conditioning, this effect of silane couldobviously not profited from. One can anticipate that alumina or silica on thesurface of the substrate could form strong enough chemical bonds, covalentbridges, through its surface hydroxyl groups with hydrolyzed silanol groups ofthe silane:-Al-O-Si- or –Si-O-Si. The methacrylate groups of the organosilane �-MPS compound form covalent bonds with the resin when polymerized.

It is difficult to compare our results with previous studies as storageconditions are not the same but our findings after thermocycling with the use ofair-particle abrasion with either alumina or silica followed by silane couplingagent and IMR application were higher than those reports where specimenswere tested either after short term water storage [34] or lower number ofthermocycles [28]. After thermocycling, the bond strengths provided were wellabove the recommended ISO standards [36]. While some studies reported thatthe type and chemical structure of repair resin have no influence on the strengthof the repair [35], some others proved that the use of silica coating provided asignificant improvement in the repair strength [28, 30]. Boyer et al [40] found thathighly filled resin composites provided higher bond strengths which contradictswith our findings and warrants further research.

Water uptake has an important role in the chemical degradation ofcomposite materials that mainly takes place in the resin matrix that is a diffusioncontrolled process with the diffusion coefficient decreasing with theconcentration of water in the matrix. Many resin composite bonding studieshave addressed the effect of storage time in water on bond strength of repairs.Söderholm and Roberts [41] found that the repair resin had a tendency toweaken when they were stored in water for 3 and 12 months. The variation incoefficient of thermal expansion of materials and especially intermediate resincould be factors responsible for the reduction of bond strength afterthermocycling.

121


In clinical situations, when PFC substrates need to be conditioned prior tobonding new resin, factors such as age of the substrate, surface chemistry andmorphology play role on the bond strength due to degradation in the oralenvironment. Of course they affect the adhesion and require furtherinvestigations.

5. Conclusions

Within the limitations of this study, the following conclusions were made:1. Composite-to-composite adhesion strengths varied in accordance with the

PFC types and surface conditioning methods tested.2. HF acid gel dissolved the filler particles but resulted in lower bond strengths

than alumina particle abrasion and silica coating.3. Air-particle abrasion with silica particles followed by silanization increased

the bond strengths regardless of the PFC type.4. When compared to dry testing conditions, bond strengths decreased after

thermocycling in all HF acid gel treated substrates but no significant changewas noted after alumina particle abrasion or silica coating followed bysilanization.

References

1. I. B. LARSEN, M. FREUNF and E. C. MUNKSGAARD, J. Dent. Res. (1992)1851.

2. K-J. SÖDERHOLM, R. MUKHERJEE and J. LONGMATE, J. Dent. Res.(1996) 1692.

3. D. C. SARRETT, D. P. COLETTI and A. R. PELUSO, Dent. Mater. (2000) 62.4. J-F. ROULET, in “Degradation of dental polymers” (Basel, Karger, 1987) p.

60.5. J. L. FERRACANE and J. R. CONDON, Dent. Mater. (1990) 282.6. U. ØSTENGREN, H. WELLENDORF, S. KARLSSON and I. E. RUYTER, J.

Oral Rehabil. (2001) 1106.7. I. A. MJØR and F. TOFFENETTI, Oper. Dent. (1992) 82.8. I. A. MJØR and C. M. UM, Int. Dent. J. (1993) 311.9. W. D. BROWNING and J. B. DENNISON, Oper. Dent. (1996) 160.10. F. J. BURKE, S.W. CHEUNG, I. A. MJØR and N. H. WILSON, Prim. Dent.

Care. (1999) 17.11. A. R. Al-NEGRISH, Int. Dent. J. (2002) 461.

122


12. I. A. MJØR and V. V. GORDAN, Oper. Dent. (2002) 528.13. M. ÖZCAN and W. NIEDERMEIER, Int. J. Prosthodont. (2002) 299.14. C. H. LLOYD, D. A. BAIGRIE and I. W. JEFFREY, J. Dent. (1980) 171.15. K. C. CHAN and D. B. BOYER, J. Prosthet. Dent. (1983) 345.16. D. B. BOYER, K. C. CHAN and J. W. REINHARDT, J. Dent. Res. (1984)

1241.17. P. AZARBAL, D. B. BOYER and K. C. CHAN, Dent. Mater. (1986) 153.18. E. C. KAO, H. G. PRYOR and W. M. JOHNSTON, J. Prosthet. Dent. (1988)

328.19. A. D. PUCKETT, R. HOLDER and J. W. O'HARA, Oper. Dent. (1991) 136.20. T. T. KALLIO, T. M. LASTUMAKI and P. K. VALLITTU, Dent. Mater. (2001)

80.21. K. A. KUPIEC and W. W. BARKMEIER, Oper. Dent. (1996) 59.22. S. A. SHAHDAD and J .G. KENNEDY, J. Dent. (1998) 685.23. H. MITSAKI-MATSOU, A. KARANIKA-KOUMA, Y. PAOADOYIANNIS and

S. THEODRIDOU-PAHINE, Quintessence Int. (1991) 475.24. E. J. Jr. SWIFT, B. D. LeVALLEY and D. B. BOYER, Dent. Mater. (1992)

362.25. C. W. TURNER and J. C. MEIERS, Oper. Dent. (1993) 187.26. G. M. IMAMURA, J. W. REINHARDT, D. B. BOYER and E. J. Jr. SWIFT,

Oper. Dent. (1996) 249.27. E. J. Jr. SWIFT, B. C. CLOE and D. B. BOYER, Am. J. Dent. (1994) 200.28. M. R. BOUSCHLICHER, J. W. REINHARDT and M. A. VARGAS, Am. J.

Dent. (1997) 279.29. P. F. CESAR, P. M. MEYER FAARA, R. MIWA CALDART, R. GASTALDONI

JAEGER et al., Am. J. Dent. (2001) 373.30. M. ÖZCAN, J. Prosthet. Dent. (2002) 469.31. E. P. PLUDDEMANN, J. Adhesion. (1970) 184.32. K. KULA, S. NELSON and V. J. THOMPSON, J. Dent. Res. (1983) 846.33. K. KULA, S. NELSON, T. KULA and V. J. THOMPSON, J. Prosthet. Dent.

(1986) 161.34. L. PAPAGIANNOULIS, J. TZOUTZAS and G. ELIADES, J. Prosthet. Dent.

(1997) 405.35. W. A. GREGORY, B. POUNDER and E. J. BAKUS, J. Prosthet. Dent. (1990)

664.36. INTERNATIONAL ORGANIZATION for STANDARDIZATION, Amendment

ISO 10477 (1996).37. J. L. FERRACANE and E. H. GREENER, J. Biomed. Mater. Res. (1986)

121.38. I. E. RUYTER and S. A. SVENDSEN, Acta Odontol. Scand. (1978) 75.

123


39. E. ASMUSSEN and A. PEUTZFELDT, J. Dent. Res. (2001) 1570.40. D. B. BOYER, K. C. CHAN and J. W. REINHARDT, J. Dent. Res. (1984)

1241.41. K-J. SÖDERHOLM, M. J. ROBERTS, Scand. J. Dent. Res. (1991) 173.

124


ChapterBond Strength of Resin Composite to Differently

Conditioned Amalgam

This chapter is submitted for publication in Operative Dentistry as:Özcan, M., Vallittu, P.K., Huysmans, M-Ch, Kalk, W, Vahlberg, T.: BondStrength of Resin Composite to Differently Conditioned Amalgam. OperativeDentistry, (submitted, 2003).

125

6


126


Bond Strength of Resin Composite to Differently

Conditioned Amalgam

M. Özcana, P.K. Vallittub, M-Ch. Huysmansa, T. Vahlbergc, W. Kalka



Research, Turku, Finland.cUniversity of Turku, Institute of Dentistry, Department of Biostatistics, Turku, Finland.

SUMMARY

Bulk fracture of teeth, where a part of the amalgam restoration and/or the cuspis fractured, is a common clinical problem. The aim of this study was toevaluate the effect of different surface conditioning methods on the shear bondstrength of a hybrid resin composite to fresh amalgam. Amalgams (N=84) werecondensed into acrylic and randomly assigned to one of the followingtreatments (N=6): (1) Alloy primer + opaquer, (2) Air-particle abrasion (50 µmAl2O3) + alloy primer + opaquer, (3) Silica coating (30 µm SiOx) + silanization+ opaquer, (4) Opaquer + pre-impregnated continuous bidirectional E-glassfibre sheets, (5) Silica coating + silanization + fibre sheets, (6) Silica coating +silanization + opaquer + fibre sheet application. Non-conditioned amalgamsurfaces were considered as control group (7). The mean surface roughnessdepth (Rz) was measured from the control group and air-abraded amalgamsurfaces. The resin composite was bonded to the conditioned amalgamspecimens using polyethylene molds. All specimens were tested under dry andthermocycled (6.000, 5ºC-55ºC, 30 seconds) conditions. The shear bondstrength of resin composite to amalgam substrates was measured in auniversal testing machine (1 mm/min). Surface roughness values for the non-conditioned control group (Rz~ 0.14 µm) and for air-particle abraded surfaceswith either Al2O3 or SiOx (Rz~ 0.19 µm and Rz~ 0.16 µm, respectively) did notshow significant differences (p=0.23) (One-way ANOVA). In dry conditions,silica coating and silanization followed by fibre sheet application exhibitedsignificantly higher results (14.8±5.6 MPa) than those of the groupsconditioned with alloy primer (2.2±0.7 MPa) (p<0.001), air-particleabrasion+alloy primer (4.4±2.0 MPa, p<0.001), silica coating + silanizationalone (6.2±0.8 MPa, p=0.009) or non-conditioned group (1.4±0.6, p<0.001).Silica coating and silanization followed by additional fibre sheets with opaquer

127


application (23.6±6.9 MPa) increased the bond strength significantlycompared to those of other groups (group 5 vs group 6, p=0.007; other groupsvs group 6, p<0.001). Thermocycling decreased the bond strengthssignificantly for all of the conditioning methods tested (for group 1, p<0.001; forgroup 2, p=0.013; for group 3, p=0.002; for group 4, p=0.026; for group 5,p=0.002; for group 6, p<0.001 and for group 7, p<0.001).

Clinical Relevance

Combination of silica coating and silanization with addition of E-glass fibersheets at the adhesive interface can be considered as an alternative methodto improve adhesion of resin composite to amalgam.

INTRODUCTION

Amalgam has served dentistry for more than a century. Although amalgamfillings undergo constant corrosion and they might not fulfil all cosmetic-estheticdemands, they are still commonly used. The results of recent surveys fromcross-sectional studies indicate that complete cusp fracture of posterior teethassociated with amalgam restorations is a common problem in dental practice.The failure rate range between 4.4 (Bader, Marin & Shugars, 1995) and 14occasions (Heft & others, 2000) per 100 subjects or 20.5 teeth per 1000 personsa year (Fennis & others, 2002).

A number of factors seem to contribute to the fracture of teeth with orwithout loss of tooth substance and amalgam material, such as occlusalinstability, impact load, fatigue load during mastication, secondary caries,microdefects, technical errors, insufficient sound tooth material availablesurrounding the restoration (undermining cusps) or occlusal prematurity. Inaddition, the more surfaces restored and/or the wider the isthmus, the greaterthe chance of cusp fracture (Cavel, Kelsey & Blankenau, 1985; Morin & others,1988; Lagouvardos, Sourai & Douvitsas, 1989; Rees, 1998). Thus it is likely thatthe restorative status of the tooth has an influence on the incidence of fracture.The majority of the fractures were observed in the supragingival location, whichsuggested that the fractured tooth could be restored easily (Fennis & others,2002). Although there is little published literature on the subject, repair of arestoration is more cost-effective than total replacement where ever appropriate(Randall, Vrijhoef & Wilson, 2001; Mjør & Gordon, 2002). It can be considered asfitting in a trend towards a less interventionist procedure (Elderton, 1990).

128


Recently, veneering the restorations with resin materials has beenattempted to solve the problem of repairing fractured teeth with amalgamrestorations. Various repair techniques have been suggested in the literature,many of which are based on either mechanical and/or chemical adhesiontechniques (Liatukas, 1970; Lubow & Cooley, 1986; Plasmans & Reukers,1993; Bichacho & others, 1995; Ruse, Sekimoto & Feduik, 1995; Salama & el-Mallakh, 1997; Al-Jazairy, 2001). Mechanical means include roughening theamalgam, preparing undercuts, creating grooves or placing self-threading pins(Gordon, Laufer & Metzger, 1985). Chemical means on the other hand, usemultipurpose adhesive agents (Watts, Devlin & Fletcher, 1992; Giannini,Paulillo & Ambrosano, 2002). While primers and adhesives form ionic bondswith metal oxides or with the active metal compounds of the amalgam(McConnell, 1993), several studies have shown that air-borne particleabrasion modifies the metal surface and provides micro-roughness that isessential for mechanical bonding (Lubow & Cooley, 1986; Ruse, Sekimoto &Feduik, 1995; Giannini, Paulillo & Ambrosano, 2002). However, the existenceof a true chemical bond between amalgam and resin composites iscontroversial (Miller & others, 1992; Bichacho & others, 1995).

The techniques that facilitate alloy-resin bonding have significantlyimproved over the last decade (Özcan, 1998) and rely on both(micro)mechanical and chemical adhesion. Numerous intraoral repair systemsare available and a growing number of systems are being introduced. Modernsurface treatment methods mostly require air-borne particle abrasion of themetal prior to bonding. These new systems also involve the conditioning of thesubstrate to produce bifunctional silane molecules that adhere to the metalsurface after being hydrolized to silanol and forming polysiloxane network onthe subtstrate and finally reacting with the monomers of theopaquer/composite (Özcan, 2002). One system, in which silanes are alsoused, is tribochemical silica coating. The surfaces are air-abraded with 30 µmgrain size aluminium trioxide modified with silisic acid. The blasting pressureresults in embedding of silica particles on the surface rendering the surfacechemically more reactive to resin via silane. One other repair alternative hasbeen proposed with the use of reinforcing fibers for the composite whereimproved fatigue resistance of composites were noted (Vallittu, 2002).

The literature contains many reports on bonding of resin composites toalloy surfaces but these methods have not been investigated for the purposesof bonding composites to amalgam. Although the concept of veneeringamalgam restorations with composite is not new to restorative dentistry, thereseems to be no consensus in the literature regarding the best method forrepairing such restorations. Therefore the aim of this study was to assess the

129


bond strength between amalgam and resin composite mediated by recentsurface conditioning methods.

METHODS AND MATERIALS

The amalgam (non-gamma 2, lathe-cut, high-copper alloy with 43 % Ag, 25.4% Cu) (N=84) (ANA 2000 Duet, Nordiska Dental AB, Angelholm, Sweden) wastriturated according to the manufacturer`s recomendations regarding speedand time and then condensed with a hand instrument into a cylindrical(diameter: 6 mm, depth: 2 mm) undercut cavity prepared in auto-polymerizedPMMA (Palapress, Vario, Heraeus Kulzer, Wehrheim, Germany) until theywere slightly overfilled. The exposed surface of each specimen was groundfinished to 1200 grit silicone carbide abrasive (Struers RotoPol 11, StruersA/S, Rodovre, Denmark) and cleaned for 10 minutes in an ultrasonic bath(Quantrex 90 WT, L&R Manufacturing Inc., Kearny, NJ 07032-0607) containingdistilled water and air-dried. Subsequently, the amalgam specimens(n=6/group) were randomly distributed in seven testing groups according totheir surface treatment (Table1, Fig 1):

Table 1. Surface treatment and testing groups (n=6/per group).

Conditioning principle Group 1 Group 2 Group 3 Group 4 Group 5 Group 6 Group 7

Air particle abrasion √

(50 µm aluminium trioxide)

Silica coating √ √ √

(CoJet®-Sand, 30 µm SiOx)

Alloy primer √ √

Silane coupling agent √ √ √

Pre-impregnated bidirectional

E-glass fibre sheets √ √ √

Opaquer √ √ √ √ √

Bonding agent √ √

Composite resin √ √ √ √ √ √ √

130



Group1: In the alloy primer treated groups, the primer [6-(4-vinylbenzyl-n-propyl) amino-1, 3,5 triazune-2, 4-dithiol (VBATDT)] (Alloy Primer™, KurarayMedical Co, Ltd, Tokyo, Japan) was first applied on amalgam substrates and itwas allowed to dry prior to bonding procedures.

Group 2: Alloy primer was applied following air-borne particle abrasionwith 50 µm Al2O3 (Korox®, Bego, Bremen, Germany) using an intraoral airabrasion device (Dento-PrepTM, RØNVIG A/S, Daugaard, Denmark) from adistance of approximately 10 mm at a pressure of 2.5 bars for 4 seconds.

Group 3: Silica coating process was achieved using the same deviceunder the same conditions but this time it was filled with 30 µm SiOx (CoJet®-Sand, 3M ESPE AG, Seefeld, Germany) (Fig 2). Following air-particleabrasion, a 3-methacryloxypropyltrimethoxy silane coupling agent (MPS)(ESPE®-Sil, 3M ESPE AG) was applied and waited for its evaporation for 5minutes.In the glass fiber treated groups, two pieces of polymer-monomer gel pre-impregnated photopolymerizable bidirectional E-glass-fiber sheets (thickness:0.06 mm, StickTech, Turku, Finland) were cut (diameter of the circular sheet:3.6 mm) and placed in one group on the non-conditioned amalgam surfaces

131

Amalgam

50 µm Al2O3/30 µm SiOx

Alloy primer/Silane coupling agent

Opaquer

Pre-impregnated bidirectionalE-glass fibre sheets+bonding agent

Composite resin

Figure 1. Schematic representation of the amalgam-composite specimens with corresponding

surface conditioning methods.


(Group 4) and in two groups on silica coated and silanized substrates (Group5 and 6) (Fig 3). Adhesive resin (Schotchbond Multipurpose Adhesive, 3MDental Products, St Paul, MN 55144) was applied after the placement of eachfiber sheet and it was light-polymerized (Optilux 501, Kerr, West CollinsOrange, CA 92867) for 10 seconds. The polished amalgam surfaces wereconsidered as control group (7).

In order to mimic the clinical situations where the amalgam is exposedand interfere with cosmetic-esthetic perspective, except group 5, opaquer

132

Figure 3. Application of polymer-monomer gel

pre-impregnated photopolymerizable

bidirectional E-glass-fiber sheets on the

amalgam surfaces. The thickness of each

fiber layer is 0.06 mm.

Figure 2. Application of chair side airborne

particle abrasion on the amalgam surfaces

using the intraoral sandblaster.


(Visiogem, 3M ESPE AG) was applied a thin layer in all groups and light-polymerized for 20 seconds.

Bonding procedures

The bonding procedures were carried out in accordance with themanufacturers` instructions by the same operator throughout the experiments.The highly filled (79 w-% filler) resin composite (Tetric Ceram, Shade A2,Ivoclar, Schaan, Liechteinstein) was bonded to the conditioned amalgamspecimens using translucent polyethylene molds with inner diameter of 3.6mm and height of 5 mm. The resin composite was packed against thesubstrate with a composite-filling instrument.The resins were light polymerizedfor 40 seconds. Light-intensity was 770 mW/cm2. Polyethylene molds weregently removed from the test specimens.

In an additional study (n=18, 6/group) the mean surface roughness depth (RZ)from the polished and air-abraded amalgam surfaces either with Al2O3 orSiOx was measured (Perthometer S8P 4.51, Feinprüf GmbH, Göttingen,Germany). The mean roughness value was calculated from 3 singlemeasurements. Each value represented the distance between the lowest andthe highest point of the surface profile. These specimens were not used for thebond test in case the measurements may damage the surfaces.

All experimental groups were assessed at both dry and thermocycledstoring conditions. While dry samples were kept in a dessicator at roomtemperature for 24 hours prior to testing, the other groups were subjected tothermocycling (Thermocycler 2000, Heto-Holten A/S, Allerod, Denmark) for6.000 cycles between 5ºC and 55ºC in deionised grade 3 water. The dwellingtime at each temperature was 30 seconds. The transfer time from one bath tothe other was 2 seconds.

Specimens were mounted in a jig (Bencor Multi-T shear assembly,Danville Engineering Inc., San Ramon, CA 94583) of the universal testingmachine (Llyod LRX, Lloyd Instruments Ltd, Lloyd, Canada) and a shear forcewas applied to the adhesive interface until fracture occurred. The specimenswere loaded at a crosshead speed of 1.0 mm/min and the stress-strain curvewas analysed with Nexygen 2.0 software.

Statistical analysis was performed using SAS System for Windows,release 8.02/2001 (Cary, NC). The comparisons between surfaceconditionings at dry conditions were made by one-way analysis of variance(ANOVA) with multiple comparisons using Tukey's honestly significant

133


difference test. Student’s t-test was used to determine the differences betweendry and thermocycled conditions and Pearson`s correlation coefficient wasused to evaluate the association between surface roughness and bondstrength. P values less than 0.05 were considered to be statistically significantin all tests.

RESULTS

Surface roughness values for the non-conditioned control group (Rz~ 0.14 µm)and for air-particle abraded surfaces with either Al2O3 (Rz~ 0.19 µm) or SiOx(Rz~ 0.16 µm) did not show significant differences (p=0.23) (Fig 4).Furthermore the surface roughness was not significantly correlated with thebond strength values (Pearson’s correlation coefficient r=0.42, p=0.09).

One-way ANOVA showed that shear bond strength was significantlyaffected by conditioning methods (p<0.001). Figure 5 displays box plots of thebond strength values associated with surface conditioning techniques at dryand thermocycled conditions. In dry conditions, silica coating and silanizationfollowed by fibre sheet application exhibited significantly higher results with(14.8±5.6 MPa) and without opaquer (23.6.8±6.9 MPa) than those of thegroups conditioned with alloy primer (2.2±0.7 MPa, p<0.001), air-particleabrasion + alloy primer (4.4±2.0 MPa, p<0.001), silica coating + silanizationalone (6.2±0.8 MPa, p=0.009) or non-conditioned group (1.4±0.6, p<0.001).Silica coating and silanization followed by additional fibre sheets with opaquerapplication (23.6±6.9 MPa) increased the bond strength significantlycompared to those of other groups (for group 5, p=0.007 and for other groups,p<0.001).

Thermocycling decreased the bond strengths significantly for all of theconditioning methods tested (group 1, p<0.001; group 2, p=0.013; group 3,p=0.002; group 4, p=0.026; group 5, p=0.002; group 6, p<0.001; group 7,p<0.001) and the least favourable results were obtained with the alloy primertreated groups with (4.4±2.0 MPa, 1.0±1.9 MPa) or without air-particleabrasion (2.2±0.7 MPa, 0 MPa) at both dry and thermocycled conditions,respectively.

134


DISCUSSION

Replacement of amalgam restorations is associated with loss of tooth tissueby progressive cavity enlargement and repeated insults to the pulp. Sincerepair of amalgam restorations with amalgam is not reliable, experienceindicates that an adhesive approach should be considered. Moreover, for therepaired restoration to withstand functional loads, strong and durable bond isneeded.

Roughening the amalgam surface can increase the surface area andfacilitate mechanical interlocking of the adhesive. The results of this studysuggest that roughening the amalgam surface by air-particle abrasion providesome micro-roughness that was also in accordance with several previousstudies (Okabe & others, 1978; Lubow & Cooley, 1986; Quiroz & Swift, 1986;Ruse, Sekimoto & Feduik, 1995; Salama & el-Mallakh, 1997) but in factsurface roughening itself did not dictated the bond strength. The present studywas performed on fresh amalgam surfaces where a high surface energy canbe expected but a change in surface roughness of high-copper amalgams overtime, due to the formation of Cu6Sn5 crystals could also provide someroughness for micro-mechanical bonding.

Alloy primers are designed for conditioning both noble and base alloysand they are claimed to promote the bond strength. The postulated interaction

135

Figure 5. Shear bond strengths (MPa) of the

resin composite bonded to conditioned

amalgam substrates at dry and thermocycled

conditions. Box plots represent the minimum

value, first quartile, median, third quartile and

maximum value.

Figure 4. The surface roughness (µm) (Rz) of

the tested amalgam (n=6 per group) for air-

particle abraded groups with either Al2O3(Group 2) or SiOx (Group 3) and non-

conditioned group (Group 7). Box plots

represent the minimum value, first quartile,

median, third quartile and maximum value.


mechanism of alloy primers involves adhesion to the alloy via hydrophiliccarboxylate groups and the exposure of a hydrophobic moiety that is able tointeract with the resin composite. The alloy primer used in this study containsVBATDT as the functional monomer for noble alloys and hydrophobic 10-methacryloyloxydecyl dihydrogen phosphate monomer (MDP) as thefunctional monomer for base alloys (Yanagida, Matsumura & Atsuta, 2001). Atpresent, the majority of the commercially available amalgam products are thehigh-copper amalgams. Use of these products avoids the formation of theeasily corroding �2(Sn8Hg) phase. Amalgam is an alloy where mercury playsan essential role but the exact mechanism of adhesion to amalgam is unclear.Principally, silane (oligomers) monomers/molecules react with each otherforming branched siloxane bonds, -Si-O-Si-, and with an inorganic substrate(matrix) (i.e. silica, metal oxides that contain basic hydroxyl –OH groups) withwhich they can form –Si-O-M- bonds (M=metal). It is likely that the oxide layeron the surface of the amalgam surface used in this study was not sufficientlyformed as it is formed on other alloys. Therefore MPS silane coupling agentand MDP-VBATDT primer did not form durable covalent bonds with amalgamssince there might not have been excess of hydroxyl groups on the surface. Thetype of amalgam may also influence the results. Higher bond strengths havebeen reported for bonding to spherical dental amalgam compared to lathe-cutor admixed amalgams (Watts, Devlin & Fletcher, 1992). However, in clinicalpractice it is not always possible for a clinician to determine the specific typeof dental amalgam.

Tribochemical silica coating followed by silanization enhanced the bondbetween the resin composite and the amalgam compared with the non-conditioned or alloy primer treated groups.The quality and concentration of thesurface oxides after silica coating and silanization affect the extent ofmolecular orientation providing a configuration that sterically favours cross-linking of the monomers of the resinous phase composites and thus increasesthe polymerization at the interface (Schneider, Powers & Pierpoint, 1992;Anagnostopoulos, Eliades & Palaghias, 1993). However, no statisticallysignificant differences were found in the bond strengths between the two air-particle abraded groups followed by either only alloy primer or MPS silaneapplication.

One interesting finding of this study was the significant influence of theapplication of glass fibers on bond strength especially on the silica coated andsilanized amalgam surfaces. The glass fibers used in this study were pre-impregnated with polymer-monomer gel. The results obtained in this groupexhibited mean bond values ranging between 9.2 MPa and 23.6 MPa thatexceeds the recommended ISO standard (1996) and therefore could be

136


considered strong enough for clinical applications. In general, stressconcentrations within the resin and the interface can be relieved by initiation ofa crack and its propagation through the resin until it meets the fibers, resultingin debonding of the resin composite. This phenomenon together with theinteraction between the silica coated and silanized oxides needs furtherinvestigation on a large number of specimens. It is also possible that thepolymer matrix between the glass fibers (semi-interpenetrating polymernetwork of polymethylmethacrylate and cross-linked dimethacrylates) couldhave behaved as low modulus stress breaker between the amalgam and therepair composite resin.

The observation of a significant decline in bond strength after long-termthermocycling is probably due to the hydrolytic degradation of the chemicalbond between the active monomers in the coupling agents studied and theamalgam substrates. Either water sorption or thermally-initiated reorientationof the coating might cause stress relaxation (Schneider, Powers & Pierpoint,1992). Direct comparison with previous studies is difficult to make since theydiffered in storage conditions but our findings after thermocycling with the useof alloy primer alone or air-particle abrasion with Al2O3 followed by alloyprimer application were lower than those reports where specimens weretested either after short term water storage (Hadavi and others, 1991; Cooley,Tseng & Barkmeier, 1991; Chang & others, 1992; Watts, Devlin & Fletcher,1992; Ruse, Sekimoto & Feduik, 1995) or lower number of thermocycles(Bichacho & others, 1995). After thermocycling, except for the glass fiber sheettreated groups, the bond strengths provided in other groups were lower thanthe recommended ISO standards.

In this study, opaquer has been advocated to mask the amalgam prior tocomposite bonding in order to simulate the clinical situations where fracture ofamalgam was encontoured with esthetics such as in some visible areas of themouth. The opaquer used in this study was dimethacrylate based that isprovided in a powder-liquid system. One can anticipate that bond strengthsmay vary with chemical composition and consistency of the opaquer. Lack ofinformation also exists on the influence of the thickness of opaquer layer onthe bond strength. It could be expected that the cohesional strength of theopaquer is lower than that of resin composite. Therefore a thick layer ofopaquer might decrease the bond strength. However, this needs furtherinvestigation.

In clinical situations where an amalgam fracture is experienced, factorssuch as existence of intact enamel/dentin, repair resins with different elasticmodulus, surface chemical composition, morphology and age of the amalgamcould also affect the adhesion of resin composites to amalgam surfaces.

137


CONCLUSIONS

From this in-vitro study, the following conclusions were drawn:1. Bond strengths of resin composite to amalgam substrates varied in

accordance with the surface conditioning techniques.2. Combination of silica coating and silanization with addition of optional pre-

impregnated bidirectional e-glass fiber sheets at the adhesive interfaceincreased the bond strengths significantly and therefore can be consideredas an alternative method to improve attachment of resin composite toamalgam.

3. Thermocycling decreased the bond strength values substantially after allsurface conditioning methods tested.

References

Al-Jazairy YH (2001) Shear peel bond strength of compomers veneered toamalgam Journal of Prosthetic Dentistry 85 396-400.

Anagnostopoulos T, Eliades G & Palaghias G (1993) Composition, reactivityand surface interactions of three dental silane primers Dental Materials 9182-90.

Bader JD, Martin JA & Shugars DA (1995) Preliminary estimates of theincidence and consequences of tooth fracture Journal of American DentalAssociation 126 1650-1654.

Bichacho N, Pilo R, Brosh T, Berkovich M & Helft M (1995) Shear bondstrength of composite resin to fresh amalgam Operative Dentistry 20 68-73.

Cavel WT, Kelsey WP & Blankenau RJ (1985) An in vivo study of cuspalfracture Journal of Prosthetic Dentistry 53 38-42.

Chang J, Scherer W, Tauk A & Martini R (1992) Shear bond strength of a 4-META adhesive system Journal of Prosthetic Dentistry 67 42-45.

Cooley RL, Tseng EY & Barkmeier WW (1991) Dentinal bond strengths andmicroleakage of a 4-META adhesive to amalgam and composite resinQuintessence International 22:979-983.

Elderton RJ (1990) Clinical studies concerning re-restoration of teethAdvances in Dental Research 4 4-9.

Fennis WM, Kuijs RH, Kreulen CM, Roeters FJ, Creugers NH & BurgersdijkRC (2002) A survey of cusp fractures in a population of general dentalpractices International Journal of Prosthodontics 15 559-563.

138


Giannini M, Paulillo LA & Ambrosano GM (2002) Effect of surface roughnesson amalgam repair using adhesive systems Brazilian Dental Journal 13179-183.

Gordon M, Laufer BZ & Metzger Z. (1985) Composite-veneered amalgamrestorations Journal of Prosthetic Dentistry 54 759-762.

Hadavi F, Hey JH, Ambrose ER & elBadrawy HE (1991) The influence of anadhesive system on shear bond strength of repaired high-copperamalgams Operative Dentistry 16 175-80.

Heft MW, Gilbert GH, Dolan TA & Foerster U (2000) Restoration fractures,cusp fractures and root fragments in a diverse sample of adults: 24-monthincidence Journal of American Dental Association 131 1459-1464.

International Organization for Standardization (1996) Dentistry-polymer-basedcrown and bridge materials Amendment ISO 10477.

Lagouvardos P, Sourai P & Douvitsas G (1989) Coronal fractures in posteriorteeth Operative Dentistry 14 28-32.

Liatukas EL (1970) Amalgam restorations with silicate cement facings foranterior teeth Journal of Prosthetic Dentistry 23 560-561.

Lubow RM & Cooley RL (1986) Effect of air-powder abrasive instrument onrestorative materials Journal of Prosthetic Dentistry 55 462-465.

McConnell RJ (1993) Metal-resin bonding. Journal of Californian DentalAssociation 21 38-42.

Miller BH, Arita K, Tamura N, Nishino M, Guo I & Okabe T (1992) Bondstrengths of various materials to dentin using Amalgambond AmericanJournal of Dentistry 5 272-276.

Mjør IA & Gordan VV. (2002) Failure, repair, refurbishing and longevity ofrestorations Operative Dentistry 27 528-534.

Morin DL, Douglas WH, Cross M & DeLong R (1988) Biophysical stressanalysis of restored teeth: experimental strain measurement DentalMaterials 4 41-48.

Okabe T, Mitchell R, Butts MB & Fairhurst CW (1978) A study of high copperamalgams III. SEM observations of amalgamation of high copper powdersJournal of Dental Research 57 975-982.

Özcan M, Pfeiffer P & Nergiz I (1998) A brief history and current status ofmetal/ceramic surface conditioning concepts for resin bonding in dentistry.Quintessence International 29 713-724.

Özcan M (2002) The use of chairside silica coating for different dentalapplications. Journal of Prosthetic Dentistry 87 469-472.

Plasmans PJ & Reukers EA (1993) Esthetic veneering of amalgamrestorations with composite resin-combining the best of both worlds?Operative Dentistry 18 66-71.

139


Quiroz L & Swift EJ Jr (1986) A technique for esthetic veneering of amalgam.Compendium of Continuous Education in Dentistry 7 352-354.

Randall RC, Vrijhoef MM & Wilson NH (2002) Current trends in restorativedentistry in the UK: a Delphi approach Journal of Dentistry 30 177-187.

Rees JS (1998) The role of cuspal flexure in the development of abfractionlesions: a finite element study European Journal of Oral Sciences 1061028-1032.

Ruse ND, Sekimoto RT & Feduik D (1995) The effect of amalgam surfacepreparation on the shear bond strength between composite and amalgam.Operative Dentistry 20 180-185.

Salama FS & el-Mallakh BF (1997) An in vitro comparison of four surfacepreparation techniques for veneering a compomer to stainless steelPediatric Dentistry 19 267-272.

Schneider W, Powers JM & Pierpont HP (1992) Bond strength of compositesto etched and silica-coated porcelain fusing alloys Dental Materials 8 211-215.

Vallittu PK (2002) Use of woven glass fibres to reinforce a composite veneer.A fracture resistance and acoustic emission study Journal of OralRehabilitation 29 423-429.

Watts DC, Devlin H & Fletcher JE (1992) Bonding characteristics of aphosphonated anaerobic adhesive to amalgam Journal of Dentistry 20245-249.

Yanagida H, Matsumura H & Atsuta M (2001) Bonding of prosthetic compositematerial to Ti-6Al-7Nb alloy with eight metal conditioners and a surfacemodification technique American Journal of Dentistry 14 291-294.

140


ChapterGeneral Discussion and Future Research

141

7


142


The use of conditioning methods was found to be a prerequisite to promote theadhesion between some of the substrates and the resin composites tested inthis thesis.

Methodology

The composite to ceramic/alloy bond is susceptible to chemical, thermal andmechanical influences under intraoral conditions. The simulation of suchinfluences in the laboratory is compulsory to draw tentative conclusions on thelong-term durability of specific bonding procedure and to identify superiormaterials and techniques. Long term water storage and thermocycling ofbonded specimens are unanimously accepted methods to simulate aging andto stress the bonding interface. The influence of prolonged thermal cycling andwater storage seems to affect the durability of adhesion of polymericadhesives to various alloys, ceramics and polymers used in dentistry. In thisthesis, long term thermocycling was applied prior to testing according to ISO(International Organization for Standardization) except where semi-permanentbonding was envisaged, i.e. bracket adhesion in orthodontics. The validity ofthe studies existing in the literature that have not fulfilled this requirementshould be considered with caution. Moreover, the available ISO standard itselfmay also not be sufficient to predict long-term clinical durability of therestorations bonded using these adhesion systems. Although bond valuesabove the recommended ISO were obtained after silica coating followed bysilanization in almost all experiments of this thesis, the relationship betweenthe nature of the substrate, adhesive and technique often leads to uncertainitywhen carrying out clinical work. The question was posed, have we yetdeveloped the optimal bonding that can serve for long-term clinicalapplications?

Material selection and clinical recommendations on resin bonding arebased on mechanical laboratory tests that show great variability in materialsand methods. One of the most common testing method is the shear bond test;however some researchers prefer modified tensile tests to eliminate theoccurrence of non-uniform interfacial stresses. The specific fracture pattern inshear testing may cause cohesive failure in the substrate which may lead toerraneous interpretation of the actual data and taint an abolute ranking of thetested materials when shear test is employed. Nevertheless since the resultsfrom tensile tests are reported to be greatly influenced by specimen geometryand the occurrence of non-uniform stress distributions during load application,for the experiments in this thesis mainly a shear test was employed. In further

143


studies, the findings of this thesis can be confirmed with the use of tensiletests.

A review of literature on particularly ceramic bonding revealed a samplesize of 6 to 10 specimens per group is commonly used. A sample size of 4specimens per group was reported to be sufficient to achieve 90 % power indetecting differences between group means.1 A sample size of 10 specimensper group provides more than 99 % power. Except for the study on theadhesion of core materials on to titanium posts where 10 specimens wereused, throughout this thesis, 6 specimens were used for the experiments fromwhich above 90 % power could be expected.

Surface conditioning with HF acid gel

Adhesive bonding techniques and modern ceramic and polymer systems offera wide range of highly aesthetic and conservative treatment options. Currentlyvarious ceramic and particulate filler resin (PFC) substrates are being used inrestorative dentistry as alternatives to metal infrastructures.The applications ofnew ceramics range from implant abutments, fixed-partial-dentures, laminates,posts, to inlay, onlay restorations either processed in the laboratory in specialfurnaces and equipment or in combination with CAD-CAM (Computer AidedDesign/ Computer Aided Manufacturing) technology. The most frequently usedones are zirconium dioxide (ZrO2), lithium disilicate, leucite-reinforcedceramics or high alumina ceramics (with 70-90 % alumina). Bonding totraditional silica-based ceramics is a predictable procedure yielding durableresults when certain guidelines are followed. In this study, HF acid etchingdemonstrated satisfactory results for ceramics with glassy matrix. However, itshould also be noted that previous reports including our findings revealed highstandard deviations in this application compared with other surfaceconditioning methods. One conceivable explanation for high standarddeviations could be that the poorly adhering precipitates that are deposited atthe bottom surface of the grooves and channels, created by acid treatment andrinsing which may weaken resin-ceramic bonds and lead to failure. Ultrasoniccleaning could be one option but in this experiment, washing and rinsing wereperformed using air-water syringe. Investigation of the possible influence ofcleaning regimens on the removal of acidic gel products is warranted. Inclinical applications however, when etching will be contemplated by chair side,this finding might have a big impact on the marginal areas of the restorations.The ceramic surfaces that are not etched properly might correspond to themargins, which in turn may lead to ditching between the tooth and the

144


restorations, thereby create locations for plaque accumulation. They may alsolead to wear of the luting resin at the interface, or loss of adhesion andeventually uneven stress distribution and fracture of the restoration. In furtherclinical studies, this issue will be given a close look.

Although HF acid gel worked fairly well in terms of receiving high bondstrength on glass matrix ceramics, the results were poor when it was used forconditioning the reinforced ceramics. The composition and physical propertiesof high-strength ceramic materials, such as aluminum trioxide-based andZrO2-based ceramics differ substantially from silica-based ceramics andobviously require alternative bonding techniques to achieve a strong, long-term, durable resin bond. The results of this study indicated that HF acid geldissolved the glassy matrix or crystalline components of the glass ceramicsbut did not dissolve either ZrO2 or the grain boundaries in high aluminaceramics. With the increasing number of ceramics with different components,there is still no optimized method available. As long as the availableconditioning methods will not be optimized, the development in the ceramicfield is expected to continue experiencing failures.

The results of the study on the use of various conditioning methods forPFC substrates prior to layering indicated that conditioning the substrates withHF acid gel adversely affected the morphological features of PFC substratesthereby resulting in poor repair strength when compared with other methodstested. Usually, inorganic fillers are integrated into the polymer matrix by silanecoupling agents, that form an interface between hydrophobic resin matrix andhydrophilic filler particles. When PFC substrates are exposed to HF acid gel,water monolayer may penetrate via voids to fillers, that in turn, maydisorganize the siloxane network formed from the condensation ofintermolecular silanol groups, which is responsible for stabilizing the filler-resininterface. All these mechanisms may weaken the particle-matrix interface thatleads to filler dissolution. This phenomenon was evidently observed in theSEM analysis where a great portion of the fillers were detached from thematrix after they were exposed to HF acid etching. The results of this studyshowed that the use of HF acid gel could not be a predictable option forconditioning PFC materials for repair or relamination purposes. Themorphologic and compositional changes in patterns obtained for the materialsafter etching were also found to be dependent on the type of acid used as wellas the composition of the restorative material. Increased filler dissolution afterHF acid gel conditioning might result in increased surface area exposure of theresin matrix and consequently an accelerated hydrolytic effect. Thisphenomenon was very evident in high filled PFCs with similar filler types whencompared to a relatively low filled one.

145


Many of the manufacturers of non-silica-based ceramics still dorecommend the conditioning methods based on HF acid gel application tocondition the ceramic surfaces in clinical practice, notwithstanding itsinsufficient effectiveness and well-recognized hazardous effects. Etchingceramics or composites with HF acid is a practical option since it can be easilyapplied at the chair side without any requirement of additional devices.However, they may lead to serious clinical manifestations. HF acid gel belongstogether with HCl (hydrochloric acid, a very strong acid) to the group ofhydrogen halogens. These are corrosive, hazardous and cytotoxic acids. HFacid gel (a medium strong acid) distinguishes itself from the other hydrogenhalogens by only partly ionizing in water. HF acid, when penetrating the skin,causes rapid destruction of underlying tissue. The dissociated hydrogen ionleads to dehydration and corrosion of tissues similar to other acid burns. Thecaustic fluoride ion can induce liquefaction necrosis of soft tissue anddestruction of the supporting bone. If HF-vapors are inhaled, they can causelung-edema and after weeks liver and kidney insufficiency can appear. It is thepain associated with hydrofluoric acid burns that requires medical assistance.Clinicians should exercise extreme caution when handling this material eitherin vitro or in vivo at chair side application. The use of HF acid gel is the mostfrequently advised method for conditioning the inner surface of suchrestorations prior to cementation. HF acid gel selectively dissolves glassy orcrystalline components of the ceramic and produces a porous irregular surfacethat increases the surface area and facilitates the penetration of the resin intothe microretentions of the etched ceramic surfaces.

However, HF acid gel can achieve proper surface texture and roughnesson the ceramics having glassy matrix in their structures. The concentration (5 - 9.6 %) and the durations (20 s-2 min) of HF gel application vary betweendifferent manufacturers and needs to be standardized with regard to theceramics used. Further surface and chemical analysis, such as EDS (EnergyDispersive Spectroscopy) and XRF (X-Ray Fluoresence) analysis, for a betterunderstanding of the effects of this gel on various ceramics are needed toperform in order to find out the corrosion rate and degree of the glassy matrixdissolution and the remnants of acid left on the ceramics. The relation afteracid etching and the corrosion rate or by-products and whether they caninitiate crack propagation are of future research interest.

Bond strengths are influenced by several factors one of which is the lutingcement type. The restorations made out of ceramics were bonded to toothpreparations using generally resin-based luting cements. The results of thisstudy revealed satisfactory bond strengths when the Bis-GMA based resincement was used for glass matrix ceramics. The question still needs to be

146


addressed in further studies whether the luting cement alone and/or thecombination with the conditioning method play the curicial role in long-termadhesion particularly for the reinforced ceramics. Future studies will also focuson the use of MDP (10-metharcyloyloxydecyldihydrogenphosphate) monomerbased luting cements in combination with the conditioning methods tested inthis thesis.

Surface conditioning with airborne particle abrasion

Although clinical evidence is lacking for an optimal method for conditioning allceramics, so far in vitro data using alumina air-particle abrasion andtribochemical silica coating together with silanization revealed satisfactory bondresults between the luting cement and the densely sintered high alumina andZrO2 based ceramics.

The air abrasion systems rely on air-particle abrasion with different particlesizes (110 and 30 µm). The abrasive process removes loose contaminatedlayers and the roughened surface provides some degree of mechanicalinterlocking or “keying” with the adhesive. It can be argued that the increasedroughness also forms a larger effective surface area for the bond. While thesemechanisms explain some of the general characteristics of adhesion toroughened surfaces, it may also introduce physico-chemical changes that affectsurface energy and wettability. The determination of surface energy and theassociated phenomenon of wettability by contact angle measurements are keysubjects for future research.

The use of silica coating followed by silanization or acrylization (applicationof silanization in a special furnace) was not only very effective on reinforcedceramics but also gave very promising results when they were employed ontitanium posts prior to adhering resin based core materials. Originally theequipment designed for silica coating process used in this study was brought todental market to be used in laboratory conditions. In order to avoid possiblecontamination during delivery of the restoration from the laboratory to the clinic,recently these techniques have been brought to the chair side and can beapplied using an intraoral air-abrasion device. Our recent data showedcomparable results when silica coating was applied at “chair side” for differentalloys, ceramics or composites.2,3 Due to the numerous occurrences ofresinous and/or metal interfaces in dentistry, the application range of newconditioning methods is wide. The promising results obtained for titanium alloysin this thesis could also be tried for other alloys used in dentistry andrestorative/prosthetic dentistry can benefit from these conditioning systems.

147


Aluminium trioxide treatment and tribochemical silica coating followed bysilanization evidently enhanced the bond between the ceramic surfaces andthe luting cement. Silica coating followed by silanization has also been tried forthe first time in this study in conjunction with transparent polycarbonatebrackets. One important aspect of bond tests in orthodontic applications is theevaluation of the failure types. Failures at the bracket/composite interface couldbe expected to be more favourable. Only in the silica coated and silanizedgroup after debonding, the resin was predominantly left adhered on theceramic surface. This type of failure in the adhesive-bracket interface revealsthe fact that the chemical bonding was equal to or exceeded the mechanicalretention provided by the bracket base and the bond strength to the ceramicsurface was greater than the cohesive strength of the luting resin. When someremnants of the luting agent is left on the ceramic surface, this type of failureis favourable for the clinicians since they only need then to polish the ceramicsurface.

Bonding of brackets in orthodontics has two adhesion aspects: adhesionof the luting cement to the conditioned substrate and adhesion of the lutingcement to the bracket base. Among the conditioning methods tested onfeldspathic ceramic prior to bonding polycarbonate brackets in this study,conventional acid-etching with 37% orthophosphoric acid revealed the leastfavourable bond results. HF acid gel on the other hand was more effective thanphosphoric acid (H3PO4) in dissolving the crystalline and glassy phase of theceramic and therefore facilitated better micromechanical retention however theresults were not statistically significant after thermocycling. With the bonding ofbrackets in orthodontics, the clinician has to address the critical question ofwhether a bond was strong enough to withstand the forces. On the other hand,this bond should allow for safe debonding, leaving the tooth substance or theunderlying restoration intact at the end of the orthodontic treatment.

It has been frequently suggested that clinically adequate bond strength fora metal orthodontic bracket to enamel should be between 6 to 8 MPa. Howeverwhen the literature suggesting these values were traced, it was noted that thewater storage or thermal conditioning were either missing or too short. Thisstudy did not deal with conditioning the bracket base but this may be importantfor clinical performance of brackets. Furthermore the new conditioningsystems could also be tried on metal or ceramic brackets.

Silanization

One of the initial steps after all the surface treatment tested in this thesis,

148


involves silane coupling agent application. Application of silane coupling agentto the preconditioned surfaces provides a covalent and hydrogen bond. Silanolgroups bond chemically to the silicon dioxide formed on the surface of thesubstrate. Silanized interfaces appear to be unstable in humid conditions andthe siloxane bond was found to deteriorate under moisture.4 Since the resinsare permeable to water, the stability of siloxane layer on the resin compositewas expected to change with hydrolysis over time. The decrease afterthermocycling in all surface conditioning methods followed by silanizationcould be explained on these grounds. However, the hydrolysis time varies,depending on the silane concentration, solution or temperature. Silaneoligomers react with each other forming branched hydrophobic siloxanebonds, –Si-O-Si-, and with an inorganic matrix (e.g. silica, metal oxides thatcontain hydroxyl –OH groups) they can form –Si-O-M– bonds (M = metal).Fresh metal surfaces have very high surface energies. In air the outermostsurfaces are oxidized and become covered by hydroxyl groups. The acidicsilanol groups then can react with the basic OH- groups on the metal. Thebasicity of the metal surfaces depends upon e.g. the chemical nature of themetal itself and the pretreatment before the silane treatment. On the inorganicsubstrate (metal) thus will be formed siloxane bonds of both types, -Si-O-M-and above and between them, –Si-O-Si-. According to the latest ideas, therewill be a film, a hydrophobic and branched polysiloxane layer that may containalso free hydrogen bonded oligomers. If the substrate is silica (quartz, SiO2)or silicate, then only a siloxane layer, -Si-O-Si- will be formed. The branchedsilane layer (film) thickness is dependent only on the concentration of thesilane solutions. Theoretically, it should be a monolayer but in practice it maybe 50-150 nm thick, approximately. There are numerous factors that affect theadhesion of silanes to metals, e.g. isoelectronic point of the metal oxide, thechemical character of oxide bond, the metal oxide solubility in water. None ofthese issues have been investigated in this thesis but since silane couplingagents have versatile effects, based on the various working mechanismsdescribed in the introduction, dominant working principle of these agents areunder investigation in our ongoing studies.

Alternative methods tested

In the final part of this thesis, where conditioning methods were tested onamalgam substrates, the use of E-glass fibers was tried for the first time incombination with silica coating and silanization. Interestingly, the addition ofoptional resin-impregnated bidirectional E-glass fiber sheets at the adhesive

149


interface increased the bond strengths significantly and therefore can beconsidered as an alternative method to improve attachment of resin compositeto amalgam. The glass fibers used in this study were pre-impregnated withpolymer-monomer gel. In general, stress concentrations within the resin andthe interface can be relieved by initiation of a crack and its propagation throughthe resin until it meets the fibers. In other words, the use of fibers mostprobably absorbed the forces and diverted the crack propagation and as aresult of this, higher bond strengths were achieved. This phenomenon togetherwith the interaction between the silica coated and silanized oxides needsfurther investigation on a large number of specimens. It is also possible thatthe polymer matrix between the glass fibers (semi-interpenetrating polymernetwork of poly(methylmethacrylate), PMMA, and cross-linkeddimethacrylates) could have behaved as low modulus stress breaker betweenthe amalgam and repair composite resin.The fiber layer therefore can even thedebonding stress concentrations. It would be interesting to investigate its effecton other alloys.

When these conditioning methods are used often opaquers have beenadvocated to mask the underlying alloy when the resulting restoration has tomeet aesthetic criteria such as in some visible areas of the mouth. In thisstudy, methacrylate opaquer demonstrated higher resistance values thanbismethacrylate opaquer. Monomethacrylates are linear polymers with highflexibility but providing better adhesion and on the other hand, polymerizationis more difficult to be obtained by conventional light-curing devices. However,dimethacrylates are highly cross-linked with increased brittleness.5 Thevarying results due to the chemical composition and consistency of theopaquer may require future studies. Lack of information exists on the influenceof the thickness of the opaquer layer on the bond strength. It could be expectedthat the cohesional strength of the opaquer is lower than that of resincomposite. Therefore thick layer of opaquer might decrease the bond strength.However, this needs further investigation.

Clinical implications

Clinicians often do expect the same result using one conditioning method forall materials used in dentistry. It should be the ultimate goal of a clinician toknow and understand the basic structure and properties of adherend andadhesives in their clinical practices and choose the right surface conditioningmethod that chemically suits the materials and of course depending on theapplication. The general outcome of this thesis suggests that in contrast to

150


what is being regularly applied for adhesion principles creatingmacromechanical principles, relatively recent surface conditioning techniquesbased on a combination of micromechanical retention and chemical coatingshould be considered for durable restorations applied in adhesive dentistry.More importantly these methods seem to offset the importance of the varietiesof the substrates and therefore could be applicable to a wide range of dentalmaterials. The equipment to apply these techniques was sophisticated andexpensive during the last two decades but they are recently simplified andbrought to the chair side.

Further clinical studies should seriously consider the use of rightconditioning methods for the specific substrate. Future in vitro experimentsshould focus on the worst-case scenario to assess the performance of thecurrent conditioning methods in dental materials science so that the resultscould be extrapolated to clinical situations.

References

1. Blatz MB, Sadan A, Arch GH Jr, Lang BR. In vitro evaluation of long-termbonding of Procera AllCeram alumina restorations with a modified resinluting agent. Journal of Prosthetic Dentistry, 89:381-387, 2003.

2. Schmage P, Nergiz I, Herrmann W, Özcan, M. Influence of various surface-conditioning methods on the bond strength of metal brackets to ceramicsurfaces. American Journal of Orthodontics and Dentofacial Orthopedics,123:540-546, 2003.

3. Nergiz I, Schmage P, Herrmann W, Özcan M. Effect of Metal Conditioningon the Bond Strength of Brackets. American Journal of Orthodontics andDentofacial Orthopedics, (in press, 2003).

4. Özcan M, Matinlinna J, Vallittu PK, Huysmans M-Ch. Effect of drying timeof 3-methacryloxypropyltrimethoxysilane on the shear bond strength ofcomposite resin to silica-coated base/noble alloys. Dental Materials, (inpress, 2003).

5. Akıslı I, Özcan M, Nergiz I. Effect of surface conditioning techniques on theresistance of composite resin core materials to titanium posts.Quintessence International, 34:725-731, 2003.

151


152


ChapterSummary

153

8


154


Since previous investigations revealed that most clinical failures in adhesivelyluted ceramic restorations initiate from the cementation or internal surfaces,the study presented in Chapter II evaluated the effect of three different surfaceconditioning methods on the bond strength of a Bis-GMA based luting cementto glass ceramics, glass infiltrated alumina, glass infiltrated ZrO2 reinforcedalumina. The three conditioning methods assesed were: (1) HF acid etching,(2) Air-borne particle abrasion, (3) Tribochemical silica coating. All specimenswere tested for shear bond strength at dry and long-term thermocycled (6000times) conditions. It was hypothesized in this study that amphoteric alumina inthe ceramic matrix could form strong enough chemical adhesion bonds,covalent bridges, through its surface hydroxyl groups with hydrolysed silanolgroups of the silane: -Al-O-Si-. The results indicated that bond strengths of theBis-GMA based composite luting cement tested on the dental ceramics aftersurface conditioning techniques varied in accordance with the ceramic types.HF acid gel was effective mostly on the ceramics having glassy matrix in theirstructures. The findings confirmed that the use of HF acid appeared to be themethod of choice for bonding the Bis-GMA resin composite luting cement tothose ceramics having glassy matrix in their structures. Conditioning theceramic surfaces with air-particle abrasion followed by silanization providedhigher bond strengths for high-alumina ceramics and the values increasedmore significantly after silica coating followed by silanization. Thermocyclingdecreased the bond strength values significantly after all surface conditioningmethods tested and the least favourable results were obtained with ZrO2 andglass infiltrated ZrO2 reinforced alumina.

Experimental and clinical reports provide evidence of significantdifferences in the survival of metal posts.The separation of core materials fromtitanium posts has been identified as one of the clinical problems related topost-core restorations. To withstand functional loads, the bond between thecore material and the post should be strong and durable. In order to study theinteraction between conditioned titanium posts and cores, in Chapter III, 6brands of core materials with different compositions (microfilled, hybrid,compomer, resin-modified glass ionomer) were applied to titanium posts thatwere previously conditioned and coated with either of the two types of light-polymerized opaquers (methacrylate, bismethacrylate). Five conditioningmethods based on silica coating followed by silanization and acrylizationnamely, Silicoater Classical, Silicoater MD, Rocatec, Kevloc and Siloc wereused. The resistance of the various core materials adapted to differentlyconditioned titanium posts were evaluated using a torque test that is quite anaggressive method compared to previous test methods existing in the

155


literature. When compared with the non-conditioned control group, the resultsindicated that the resistance to torsional forces for the core materials ontitanium posts increased with the use of chemical surface conditioningtechniques and varied in accordance with the opaquer type. Type of corematerial also significantly influenced the resistance after long termthermocycling. Resistance against torque forces was the greatest with Silocand then, in descending order, with Silicoater Classical, Silicoater MD,Rocatec, and Kevloc surface conditioning systems. The resistance of corematerials based on silica coating and silanization or silica coating andacrylization varied in accordance with the opaquer used. Methacrylateopaquer demonstrated higher resistance values than bismethacrylateopaquer. Monomethacrylates are linear polymers with high flexibility butproviding better adhesion and on the other hand, polymerization is moredifficult to be obtained by light curing. However, dimethacrylates are highlycross-linked with increased brittleness. Hybrid composites and compomersused as core materials demonstrated higher torque resistance compared withmicrofilled composites or resin-modified glass ionomer.

With the increased demand for adult orthodontics, the orthodontists areoften faced with the problem of luting brackets to metal-ceramic fixed-partial-dentures. Recently, more aesthetic and relatively invisible brackets, satisfyingpatient desires gained popularity in orthodontics. Unfortunately lack of durablebonding between the brackets and ceramic restorations is still a major problemin adult orthodontics. Bonding concepts in orthodontics are somewhat differentthan in the other restorative applications in operative dentistry. Since bondingin orthodontics is semi-permanent in nature, bond strength should be highenough to resist debonding during the whole course of treatment but also lowenough so that damage to the existing tooth or restoration would not occurduring debonding. In the study presented in Chapter IV, the effect of fivedifferent surface conditioning methods, namely (1) Phosphoric acid (H3PO4) +primer+ bonding agent, (2) HF acid gel + primer +bonding agent, (3) chair sidetribochemical silica coating (4) Air-borne particle abrasion with alumina +silane, (5) Air-borne particle abrasion with alumina + silane + bonding agentwere tested for the shear bond strength of polycarbonate brackets to glazedfeldspathic ceramic surfaces using light-polymerized resin-based cement.Since the orthodontic treatment duration is shorter than conventionalrestorative procedures, bond tests were performed after only 1000 cycles. Theresults indicated that bond strengths of the polycarbonate brackets luted withresin composite cement tested on the dental ceramics after surfaceconditioning techniques varied in accordance with the conditioning methods.

156


Air-borne particle abrasion with aluminium trioxide or silica coating followed bysilanization demonstrated the most favourable bond strengths.The satisfactorybond strengths obtained after air-particle abrasion either with silica or aluminatogether with silanization could eliminate the need for acid etching, primerand/or bonding agent applications. After debonding, the fracture sites of theceramic specimens and the bracket bases were further examined underscanning electron microscopy to evaluate the changes on the surface. Thefailure modes were classified according to modified Adhesive Remnant Index(ARI) system. While in the phosphoric acid etched group, the brackets failedmainly at the ceramic/resin interface with all of the luting cement remaining onthe bracket base, in the HF acid treated group, predominantly less than half ofthe composite was left on the ceramic surface after debonding. In both air-borne particle abraded groups, more than half of the luting cement was leftadhered to the ceramic surface and the bracket base. On the contrary, in thesilica coated group, luting cement was mainly debonded from the bracket basebeing left adhered to the ceramic surfaces with distinct impression of bracketmesh. The type of failures observed after debonding indicated that the criticalparameter was the strength of the adhesive joints of the luting cement to boththe bracket and the ceramic. Bond strengths of the polycarbonate bracketsluted with resin composite cement tested on the dental ceramics after surfaceconditioning techniques varied in accordance with the conditioning methods.The use of HF acid would still be appropriate for orthodontic reasons withsufficient bond strength and favourable failure modes after debonding if thecritical aspect was accepted to use this chemical agent intraorally.

Adhesion of resins to processed composites has been difficult to achieve.Aggressive oral environment and enzymatic changes all provoke discoloration,degradation, microleakage, wear, ditching at the margins, delamination orsimply fracture being often experienced in clinical conditions, that may requirereplacement of the restoration. We hypothesized that if the right conditioningmethod could be found then a new layer of composite could be applied to thealready polymerized one in an attempt to prolong the service life ofrestorations suffering from small deficiencies. Therefore the objective of thestudy presented in Chapter V was to evaluate the effect of three surfaceconditioning methods 1) HF acid gel (9.5 %) etching, (2) Air-borne particleabrasion (50 µm Al2O3), (3) Silica coating (30 µm SiOx, CoJet®-Sand) on theshear bond strength of a low-viscous diacrylate veneering particulate fillerresin-composite (PFC) to 5 PFC substrates. The bond strengths wereevaluated at both dry and thermocycled (6.000 cycles) conditions. Bondstrengths of low-viscous diacrylate veneering resin to PFC substrates tested,

157


increased with the use of silica coating and silanization and varied inaccordance with the PFC types. HF acid gel appeared to dissolve the fillerparticles but resulted in the least favourable bond strengths when comparedwith other methods tested. Air-borne particle abrasion with silica particlesincreased the bond strengths regardless of the PFC type. When compared todry testing conditions, bond strengths decreased after thermocycling in all HFacid gel treated substrates but no significant change was noted after air-particle abrasion or silica coating followed by silanization.

Complete or partial cusp fracture of posterior teeth associated withamalgam restorations is a common problem in dental practice. In an attemptto find a reliable method to restore the fractured surfaces without drilling andremoving the sound amalgam restorations, in the study presented in ChapterVI, the effect of different surface conditioning methods on the shear bondstrength of a hybrid resin composite to fresh amalgam were evaluated.Amalgams were conditioned using one of the following conditioning methods:(1) Alloy primer + opaquer, (2) Air-borne particle abrasion (50 µm Al2O3) +alloy primer + opaquer, (3) Silica coating (30 µm SiOx) + silanization +opaquer, (4) Opaquer + pre-impregnated continuous bidirectional E-glass fibresheets, (5) Silica coating + silanization + fibre sheets, (6) Silica coating +silanization + opaquer + fibre sheet application. Non-conditioned amalgamsurfaces were considered as control group (7). The mean surface roughnessdepth (Rz) was measured from the control group and air-abraded amalgamsurfaces. All specimens were tested at dry and thermocycled (6.000 cycles)conditions. The results revealed that combination of silica coating andsilanization with addition of glass fiber sheets at the adhesive interface couldbe considered as an alternative method to improve adhesion of resincomposite to amalgam. Bond strengths of the resin composite to amalgamsubstrates varied in accordance with the surface conditioning techniques.Conditioning the amalgam surface with air-borne particle abrasion prior tobonding resin composite provided higher bond strengths compared to the non-conditioned control group or alloy primer treated groups in dry conditions. Theuse of optional E-glass fibers was tried for the first time in this study incombination with silica coating and silanization. The addition of optional resin-impregnated bidirectional E-glass fiber sheets at the adhesive interfaceincreased the bond strengths significantly and therefore can be considered asan alternative method to improve attachment of resin composite to amalgam.Thermocycling decreased the bond strength values significantly after allsurface conditioning methods tested.

158


In Chapter VII the methodological aspects of the experiments conductedin Chapters II-IV are evaluated and the results obtained after thermocycling inrelation to the recommended ISO standard are discussed. The efficacy ofvarious adhesion methods and some hazardous aspects related to the use ofhydrofluoric acid are critically evaluated and the future research ideas that areplanned on surface analysis are mentioned. Furthermore in this chapter,recent data obtained by the author after the completion of this thesis usingairborne particle abrasion, silica coating together with silanization arementioned and the ongoing research, further investigation on the workingmechanisms and durability of these methods are discussed. Finally the resultsof this thesis were extrapolated to clinical situations. The results of this thesisseem to be more in favour of chemical conditioning methods for varioussubstrates. Using these methods in clinical practice may prolong the servicelife of dental restorations in a cheaper way, avoid the total replacement of theexisting restorations and preserve the tooth structure in a non-traumatic way.

159


160


DUTCH SUMMARY-SAMENVATTING

161


SAMENVATTING

De meeste klinische mislukkingen van adhesief bevestigde keramischerestauraties vinden een oorsprong aan het grensvlak tussenbevestigingscement en het interne oppervlak van de restauratie. Daaromevalueerde de studie in hoofdstuk II het effect van drie methoden vanoppervlakteconditionering op de hechtsterkte van een composietcement (opbasis van Bis-GMA) aan de volgende keramieken: glaskeramiek, met glasgeïnfiltreerd alumina, met glas geinfiltreerd en met zirkoniumoxide versterktalumina.

De drie conditioneringmethoden waren: (1) etsen met hydrofluoride (HF),(2) zandstralen met aluminiumoxide en (3) tribochemische silicacoating, allegevolgd door aanbrengen van silaan. Van deze preparaten werd deafschuifsterkte bepaald na droog bewaren en na thermocycling in water (6000cycli). De hypothese in deze studie was dat alumina in de keramische matrixvoldoende sterke hechting aan composietcement zou kunnen bereiken doorcovalente bruggen tussen oppervlakkige hydroxylgroepen en gehydrolyseerdesilanolgroepen van het silaan: -Al-O-Si-. De hechtsterkte van het Bis-GMAcement na de verschillende conditioneringmethoden was afhankelijk van hettype keramiek. Etsen met HF gaf de beste resultaten bij keramieken met eenglasmatrix. Zandstralen gevolgd door silanisering resulteerde in een hogerehechtsterkte voor de hoog-alumina keramieken. Silicacoating gevolgd doorsilanisering gaf nog hogere hechtsterktes. De wijze van bewaren van depreparaten had significante invloed op de hechtsterkte, waarbij thermocyclingleidde tot lagere hechtsterktes, vooral in de met glas geinfiltreerde met ZrO2versterkt alumina.

Van plastische stiftopbouwen is bekend dat opbouwmaterialen los kunnenraken van titanium stiften. Om functionele belasting te kunnen weerstaan moetde hechting tussen stift en opbouwmaterialen sterk en duurzaam zijn. In destudie in hoofdstuk III wordt de interactie onderzocht tussen 6 merken vanverschillende typen opbouwmaterialen en titanium stiften. De stiften warenvoorbehandeld met een van de volgende methoden: Silicoater Classical,Silicoater MD, Rocatec, Kevloc en Siloc. Deze conditioneringmethoden zijnalle gebaseerd op silicacoating en silanisering of acrylisering. In decontrolegroepen werd geen voorbehandeling toegepast. Na conditioneringwerden 2 verschillende lichthardende opaquers op de stiften aangebracht, éénop basis van methacrylaat en één op bismethacrylaatbasis.

De hechtsterkte van de opbouwmaterialen aan de stiften werd onderzochtmet een torque test. Deze methode is vrij aggressief in vergelijking met andere

162


bestaande hechtsterkte onderzoeken. Het type opbouwmateriaal beinvloeddede hechtsterkte na thermocycling significant. Hybride composieten encompomeren vertoonden een hogere torsieweerstand dan de microfillcomposieten en harsgemodificeerd glasionomeercement.

De weerstand tegen torsiekrachten was het grootst bij Siloc, inafnemende volgorde gevolgd door Silicoater Classical, Silicoater MD, Rocatecen Kevloc.

Hechtsterktes werden ook beinvloed door het type opaquer: demethacrylaat opaker gaf hogere waarden dan het bismethacrylaat type.Monomethacrylaten zijn lineaire polymeren met hoge flexibiliteit en goedeadhesie, maar polymerisatie met licht is moeilijker. Daarentegen vertonen debismethacrylaten in hoge mate crosslinking, maar zijn ze brosser.

Doordat in toenemende mate volwassen patiënten vragen om orthodontischebehandeling komt het vaker voor dat brackets op porseleinen restauratiesmoeten worden geplaatst en dat patiënten vragen om bijna onzichtbarebrackets.

Helaas is de hechting van brackets aan keramische restauraties nogsteeds niet betrouwbaar. De eisen die worden gesteld aan hechting zijn in deorthodontie anders dan in de restauratieve tandheelkunde. In de orthodontiehebben hechtprocedures een semi-permanent karakter: de hechting moet zosterk zijn dat brackets niet tijdens behandeling los raken, maar ook zo zwakdat ze gemakkelijk zijn te verwijderen na behandeling, zonder beschadigingvan tandmateriaal of ander substraat.

Hoofdstuk IV beschrijft een onderzoek naar het effect van vijf verschillendeoppervlaktebehandelingen op de hechting van polycarbonaat brackets aanveldspaat porselein.

De brackets werden gecementeerd met lichthardend composiet, nadat hetporselein was behandeld met een van de volgende conditioneringmethoden:

163


(1) etsen met fosforzuur + primer + bonding agent, (2) etsen methydrofluoridezuur + primer + bonding agent, (3) tribochemische silicacoating(chair-side), (4) zandstralen met aluminapartikels + silaan + bondingagent.De hechtsterktes werden bepaald met afschuiftest, na thermocycling. Omdatorthodontische behandelingen relatief kort duren zijn de hechtsterktesbepaald na “slechts” 1000 cycli.

De conditioneringtechnieken hadden verschillende hechtsterktes totgevolg. De beste resultaten werden bereikt na zandstralen of silicacoatinggevolgd door silanisering. Dit zou kunnen betekenen dat etsen en primer en/ofbondingapplicatie niet langer nodig is. Na los raken van de brackets werden debreukvlakken onder scanning elektronenmicroscoop onderzocht op de wijzevan falen. Deze werd geclassificeerd volgens de gemodificeerde AdhesiveRemnant Index (ARI).

In de fosforzuur-groep trad breuk voornamelijk op ter plaatse van hetgrensvlak van porselein met composiet, waarbij alle cement achter bleef op debracket. In de met hydrofluoridezuur behandelde groep bleef meestal minderdan de helft van de composiet achter op het porseleinoppervlak.

In de beide gezandstraalde groepen bleef meer dan de helft van hetbevestigingscomposiet zitten aan het porselein en aan de brackets. In desilicacoating groep echter, was het composiet grotendeels los geraakt van debracket maar achter op het porselein.

Uit observatie van de breukvlakken blijkt dat de kritieke parameter desterkte van de adhesieve verbinding van het cement aan zowel bracket alsporselein is.

Etsen met hydrofluoride zou geschikt kunnen zijn voor orthodontischegebruik, maar de gezondheidsrisico’s van intraorale toepassing mogen nietonderschat worden.

In composietrestauraties treden door blootstelling aan het aggressievemondmilieu veranderingen op, die verantwoordelijk zijn voor processen alsverkleuring, microlekkage, slijtage, randbreuk, delaminatie. Ook kanbulkfractuur optreden.

164


Bij geringe schade aan een overigens goed functionerende restauratie,zou reparatie van de restauratie te verkiezen zijn. Maar omdat hechting vankunststof aan reeds uitgehard composiet moeilijk is te realiseren, wordt vaakgekozen voor vervanging van de restauratie.

Een adequate conditioneringmethode zou het mogelijk moeten makenhechting tussen nieuw en oud composiet te verkrijgen, waarmee de ouderestauratie kan worden bedekt met een nieuwe laag composiet. In hoofdstuk Vwordt een onderzoek beschreven waarin de effecten van drieconditioneringmethoden op afschuifsterkte van oud aan nieuw composietwerden geëvalueerd. Vijf verschillende composieten werden voorbehandeldmet drie methoden: (1) etsen met 9,5% hydrofluoride (HF), (2) zandstralen met50 �m aluminiumoxide en (3) silicacoating met 30 �m siliciumoxide (CoJet®-zand). Na voorbehandeling werd een laag viskeus diacrylaat veneeringparticulate filler composiet (PFC) aangebracht. De hechtsterkte werd bepaaldin afschuiftest na droog bewaren en na thermocycling (6000) cycli.Hechtsterktes waren groter na silicacoating + silanisering en warenverschillend voor de diverse geteste composieten.

HF blijkt de vulstofpartikels op te lossen, maar resulteerde in de minstgunstige hechtwaardes vergeleken met de andere methodes. Zandstralenverhoogt de hechtsterktes onafhankelijk van het type composiet. Bij alle metHF behandelde preparaten werden lagere hechtwaarden gevonden dan nabewaren in droge conditie, maar na zandstralen en silicacoating plussilanisering waren geen significante verschillen tussen de samples na droogbewaren en na thermocycling.

Complete of gedeeltelijke knobbelfractuur in met amalgaam gerestaureerdeelementen is een bekend klinisch probleem in de tandartspraktijk. In de studiegepresenteerd in hoofdstuk VI wordt geprobeerd een betrouwbare methodete vinden voor het restaureren van fracturen zonder boren en zonderverwijderen van nog goed functionerende amalgaamrestauraties. Het effectvan verschillende oppervlaktebehandelingen op de afschuifsterkte van eenhybride composiet op vers amalgaam werd geëvalueerd. Amalgaam werd op3 manieren voorbehandeld: (1) Alloy primer + opaquer, (2) zandstralen met 50 �m aluminiumoxide + alloy primer + opaquer, (3) silicacoating met 30 �msiliciumoxide + silanisatie + opaquer, (4) opaquer + gepreïmpregneerdebidirectionele E-glas vezel sheets, (5) silicacoating + silanisatie +glasvezelsheets, (6) silicacoating + silanisatie + opaquer + glasvezelsheets.Ongeconditioneerd amalgaam vormde de controlegroep.

Oppervlakteruwheid (Rz) werd gemeten bij de controlegroep en de

165


gezandstraalde groep. Afschuifsterktes werden bepaald na droog bewaren enna thermocycling (6000 cycli).

De resultaten laten zien dat silicacoating + silaan gecombineerd metglasvezel sheets een veelbelovende techniek kan zijn om composiet aanamalgaam te hechten.

De conditioneringmethoden hadden invloed op de hechtsterktes.Zandstralen van het amalgaam gaf hogere hechtsterktes dan decontrolegroep en de alloy primer groep in droge omstandigheden. Nathermocycling werden significant lagere hechtsterktes gemeten voor alleconditioneringmethoden.

In hoofdstuk VII worden de methodologische aspecten van de onderzoekenin hoofdstukken II tot en met VI en de resultaten verkregen na thermocyclingin relatie tot de aanbevolen ISO standaard besproken. De effectiviteit vanverschillende hechtprocedures en enkele schadelijke aspecten vanhydrofluoridezuur worden kritisch tegen het licht gehouden en toekomstigonderzoek omtrent oppervlakteanalyse wordt beschreven.Verder worden in dithoofdstuk recente data genoemd, die na het voltooien van het proefschriftdoor de auteur zijn verkregen uit onderzoek naar zandstralen en silicacoating+ silanisatie. Nader onderzoek naar de werkingsmechanismen enduurzaamheid van deze methoden worden besproken. Tot slot worden deresultaten van dit proefschrift geëxtrapoleerd naar de klinische situatie. Deresultaten van dit proefschrift laten een gunstig effect zien van combinatiesvan micromechanische en chemische hechttechnieken op verschillendesubstraten. Klinische toepassing kan de levensduur van tandheelkundigerestauraties verlengen en kan voorkomen dat volledige vervanging vanrestauraties noodzakelijk is bij geringe onvolkomenheden. Hiermee kanonnodig trauma aan tandweefsel worden voorkomen en kunnen kostenbespaard worden.

166


ACKNOWLEDGEMENTS

Yes, I do enjoy writing in general but I must admit I enjoy writing theacknowledgements the most. It is not possible to avoid being emotional writingthis part of a thesis that is opposite to keeping rational while doing research.Prof. Dr. MARIE-CHARLOTTE HUYSMANS: Marie-Charlotte, you could notconvince me to play rugby but to write a second Ph.D. Like I always say, Ibelong to the softy group of people and find it safer to continue my gymlessons and dancing classes. I do hope that I will find some time for those afterthis thesis. I also thank you for being there when Murphy was ruling my worldparticularly before the congresses.Prof. Dr. WARNER KALK: Warner, during this 1.5 years our conversationsoften involved time and people-management. I think I was and still am good intime-management but I am sure I will learn more from your experiences thatmeet the latter. Thank you for the support.Prof. Dr. PEKKA VALLITTU: Pekka, many thanks for giving me one of the mostimportant things in my life: My Freedom in Research.Assoc. Prof. JOOST ROETERS: Joost, sometimes receptions at thecongresses may change your life, don`t they? Thanks for introducing me toMarie-Charlotte at the IADR Annual Meeting in Washington.Drs. HANS SCHOLTANUS: Hans, many thanks for your contribution to thetranslation of the Dutch summary of this thesis.All my friends all over the world who crossed my path over the years, I thankyou all for being so patient, keeping loyal and understanding especially attimes I could not reply your E-mails, often declined joining you at the partieswhen I switched the world off and stayed in my cocoon. And my dear colleguesat RuG THK/MH, thanks for making me feel at home and putting up with mybroken Dutch.Mum, you missed sharing this part and also the rest of my life but I will alwaysremember your saying “Never look back but ahead”. And dad, I know you arestrong!

167


CURRICULUM VITAE

Mutlu Özcan was born in Erzincan, Turkey on June, 17th, 1969.

• 1987: Graduated from High School (Bornova Anadolu Lisesi) in Izmir.• 1993-1997: Qualified DDS from the University of Marmara, Dental School

in Istanbul and started working as a research assistant at the Departmentof Prosthetic Dentistry at the same university.

• 1997-1999: Invited to do her Ph.D thesis (Fracture Strength of Ceramic-fused-to-metal Crowns Repaired with Two Intraoral Air-abrasionTechniques and Some Aspects of Silane Pretreatment- A Laboratory andClinical study) at the University of Cologne, Medical and Dental School,Department of Prosthodontics in Cologne, Germany.

• 1999-2000: Worked as an assistant professor at the University ofMarmara, Dental School, at the Department of Prosthetic Dentistry inIstanbul.

• 2001- 2002: Invited as a visiting research scientist to the University ofTurku, Institute of Dentistry, Department of Prosthetic Dentistry andBiomaterials Research in Turku, Finland.

• April 2nd 2002: Achieved the oral exam to be entitled as an associateprofessor in Turkey.

• 2002- Working as an assistant professor and research scientist at theState University of Groningen, Department of Dentistry and DentalHygiene in Groningen, The Netherlands.

168






MUTLU ÖZCAN

Adhesion of R

esin Com

posites to Biom

aterials in Dentistry:A

n Evaluation of S

urface Conditioning M

ethodsM

UT

LU

ÖZ

CA

N

Documents

Adhesion of resin composites to biomaterials in dentistry Özcan, Mutlu