27 December

1

Chapter one Introduction 1. Restorative dental material Restorative dental materials consist of all synthetic components that can be used to repair or replace tooth structure (Fig 1.1). These materials may be used for temporary purposes (e.g. temporary cement, temporary crowns), or for longer applications (e.g. inlays, onlays, crowns and bridges) 1.

1.1 Classification: Restorative dental materials can be classified as direct or indirect restorative materials 2 (Table 1.1). Table 1.1 Categories of restorative dental materials 3. Direct materials

Indirect materials

Amalgam

All-ceramic

Resin based composite

Metal-ceramic

Glass ionomer

Cast gold (noble metal) alloys

Resin modified glass ionomer

Base metal (non noble metal) alloys

Fig 1-1 Schematic illustration of a cross sectional view of . a natural anterior tooth and supporting tissues 1

2 1.1.1 Direct restorative materials: Are used intra-coronally to fabricate restorations directly on the teeth or tissues (Table 1.2). Examples include: •

Amalgam: A mixture of mercury and silver alloy powder that forms a hard solid metal filing which is self hardening at mouth temperature.

• Direct resin based composite: A mixture of submicron glass filler and acrylic that form a solid tooth coloured restoration which is self or light hardening at mouth temperature. •

Glass ionomer: Self-hardening mixture of fluoride containing glass powder and organic acid that forms a solid tooth colour restoration able to release fluoride.

•

Resin modified glass ionomer: Self or light hardening mixture of sub micron glass filler with fluoride–containing glass powder and acrylic resin that forms a solid tooth coloured restoration able to release fluoride.

Table 1-2 Comparison of direct restorative dental materials 3.

Resin based composite (direct and indirect)

Glass ionomer

Factor

Amalgam

Principle use

Dental filling and heavily loaded posterior tooth restoration

Durability

Good to excellent in large load-bearing restorations

Good in small to moderate size restoration

Resistance to wear

High resistance to wear

Moderately resistant

Biocompatibility

Well tolerated with rare occurrence of an allergic response

Aesthetics

Silver or grey metallic colour does not mimic tooth colour

Aesthetic dental filling and veneers

Mimics natural tooth colour and translucency, but can be subject to staining and discolouration

Resin modified glass ionomer

Small in non-loadbearing fillings, cavity liners and cements for crowns and bridges Moderate to good in non–load-bearing restorations, poor in load bearing High wear when placed on chewing surface

Mimics natural tooth colour, but lacks natural translucency of enamel

3 1.1.2 Indirect restorative materials: Are used extraorally in which the materials are formed indirectly on casts or other replicas of the teeth and other tissues (Table 1.3). Examples include: •

Ceramic: Porcelain, ceramic or glass-like fillings and crowns.

•

Metal-ceramic: Ceramic is fused to underlying metal structure to

provide strength to inlays, crowns and bridges (bridges are also known as fixed partial dentures). •

Cast gold (high noble) alloy: Alloy of gold copper and other metals

resulting in a strong, effective inlays, crowns and bridges. •

Base-metal alloy: Alloy of non noble metals with silver appearance

resulting in a high-strength crowns and bridges . Table 1-3 Comparison of indirect restorative dental materials 3.

Metal ceramic

Cast-gold (high noble) alloys

Base metal alloys (base metal)

Crowns and fixed bridges

Inlays, onlays, crowns and fixed bridges

Crowns, fixed bridges and partial dentures

Factor

Ceramic

Principle use

Inlays, onlays, crowns, bridges and veneers

Durability

Brittle material, may fracture under heavy biting load

Very strong and durable

High corrosion resistance prevent tarnishing high strength and toughness resist fracture and wear

Resistance to wear

Highly resistant to wear but ceramic can rapidly wear the opposing teeth

Highly resistant to wear but ceramic can rapidly wear the opposing teeth

Resistant to wear and gentle to opposing teeth

Aesthetic

Colour and translucency mimics natural tooth appearance

Ceramic can mimic natural tooth appearance, but metal limits translucency

Metal colours don’t mimic natural teeth

4

1.2 Ceramics The word ceramics can be traced back to the ancient Greek terminology. The term keramos meaning "a potty" or "potty", keramos in turn is related to an older Sanskrit origin meaning (to burn) 4. In dental science, ceramics are referred to as a non-metallic, inorganic structure, primarily composed of oxygen with one or more metallic or semi-metallic elements like aluminium, zirconium, calcium, magnesium, phosphate, potassium, silicon, and titanium 5. 1.2.1 Composition: In its mineral state, feldspar the main raw material of dental porcelain is crystalline and opaque with an indefinite colour between grey and pink 2. Chemically it is designated as potassium aluminium silicate with a composition of KO.Al2O3.6SiO. The fusion temperature of feldspar varies between 1125 °C and 1170 °C depending upon its purity. Pure quartz SIO2 is used in dental porcelain. Silica is added and contributes stability to the mass during heating by providing a glassy framework for the other ingredients 6. This feldspar goes through a manufacturing process, after which feldspathic dental ceramic consists of two phases. The first is the glass (or vitreous) phase which has properties typical of glass such as brittleness, translucency and lack of definite melting point temperature (amorphous structure). The second is the crystalline (or mineral) phase, the amount present controls the thermal expansion coefficient of the porcelain, leucite also contributes strength to porcelain, high leucite porcelains are approximately twice as strong as those containing low concentrations 2 (Figs 1.2 and 1.3).

5

Fig 1-2 Dimensional structure of .sodium silicate glass 2

Fig 1-3 Three dimensional structure of .leucite (KAlSi2O6) 2

‫حمحخ‬-= 1.2.2 Applications: Ceramics have three major applications in dentistry 7 (Table 1.4): 1- Ceramic denture teeth (this application will not be discussed further). 2- Ceramics for metal crowns, and bridges. 3- All-ceramic crowns, inlays, onlays, and veneers. Table 1-4 Classification of dental ceramic materials 2. Applications in dentistry

Processing methods

Denture teeth

Manufactured

Ceramic-metal

Sintered

All-ceramic

Machined, slip cast, heat pressed, sintered

1.2.3 Ceramic metal restorations: Ceramic metal restorations consist of a cast metallic framework on which layers of ceramic are baked. The first layer applied is the opaque layer consisting of a ceramic rich in pacifying oxides (tin oxide zirconium oxide or titanium oxide), its role is to mask the darkness of the oxidized metal framework to achieve adequate aesthetics, this first layer also provides the ceramic-metal bond which involves mechanical retention by the

6 rough surface of the metal oxide, and chemical retention by the diffusion of atoms from the metal and porcelain into the oxide. The next step is the build up of dentine and enamel (more translucent) ceramics to obtain an aesthetic appearance similar to that of a natural tooth. After building up the ceramic metal crown it is sintered in a porcelain furnace, as the porcelain is heated adjacent particles fuse and bond together. After the incisal layer is added, the porcelain is brought to the final stage called the glaze bake, a thin layer of a low fusing glass or glaze is added to the surface and fired to the flow temperature of the glaze 8 (Fig 1.4). Advantages of metal-ceramic restorations 7: •

High strength.

•

Excellent fit because of casting.

Disadvantages of metal- ceramic restorations 9: •

Appearance of metal margins

•

Lack of translucency.

•

Discoloration by metal.

•

Bond failure with metal.

.Fig

1-4 Cross section of metal-ceramic crown 1

7 1.2.4 Classification of all-ceramic restorations according to processing methods: 1.2.4.1 Ceramic jacket crowns: It is a glass ceramic with 40-50 % Alumina by weight. The alumina-reinforced porcelain jacket crowns (PJC) are providing better aesthetics for anterior teeth than metal-ceramic crowns. However the strength of the core porcelain used for PJCs is not enough for posterior teeth 1. 1.2.4.2 Bonded platinum-foil coping: The objective of this technique is to improve the aesthetics of anterior teeth by replacing the thicker metal coping with a thin platinum foil thus allowing more room for porcelain 2

. Attachment of the porcelain is secured by electroplating the platinum foil with a thin

layer of tin and then oxidizing it in a furnace to provide tin oxide layer. However, these restorations are unacceptable for posterior teeth because of the high rate of fracture 1. 1.2.4.3 Castable and pressed-glass ceramic: Glass ceramic consists of a glass matrix phase, and at least one crystal phase that is produced by the controlled crystallization of the glass. Controlled crystallization generally involves the melting, formatting and cooling of glasses to form appropriate shapes or powders that are subsequently re-heated to promote crystallization via a nucleation and growth process 10. Castable ceramic systems are used to cast crowns by the lost wax process, and are indicated for use in an anterior and posterior crown. The restoration is waxed on the die and the wax pattern of the crown is invested in a phosphate bonded investment. An ingot of the ceramic material is placed in a special crucible and melted and cast with a motor-driven centrifugal casting machine at 1380 °C. The cast crown is a clear glass that must be heat treated to form a crystalline ceramic; the crystallization procedure takes several hours in a heat treatment furnace with a final temperature of 1075 °C. The

8 fired ceramic crown has a white shade with a translucency of around 50 %. Final shading is achieved using a series of coloured surface porcelain 5 (Fig 1.5).

Fig 1-5 Cross-section of cemented Dicore cast glass-ceramic 1. .crown 1 Advantages of glass -ceramic restorations: •

Ability to be etched and bonded to dentine, and good marginal

adaptation 8. •

High flexural strength which allow it to be used in the fabrication of

anterior dental bridges 11. •

Good optical properties, and good casting ability, easily prepared by the

lost wax technique 12. Disadvantages of glass-ceramic restorations: •

Learning period required for technicians 8.

1.2.4.4 Magnesia-based core material: A high expansion magnesia core material has been developed that is compatible with the same porcelain used for ceramic- metal restoration. The flexural strength of magnesia core ceramic is twice as high (131 MPa) as that of conventional feldspathic porcelain (70 MPa). The main advantage of this core

9 material is that it allows a laboratory to veneer it with the more widely available porcelain for ceramic-metal restorations 3. 1.2.4.5 Vita In-Ceram: In Ceram is supplied as one of three core ceramics 13: •

In Ceram Alumina (Al2O3) with high strength and moderate

translucency, for anterior and posterior crown, and anterior bridges. •

In Ceram Spinal (MgO-Al2O3), the most translucent with moderately

high strength, for anterior crowns. •

In Ceram Zirconia (Al2O3-ZrO2) with high strength and lower

translucency for three-unit anterior and posterior bridges, and anterior and posterior crowns. These ceramic materials are composed of an infiltrated core veneered with feldspathic porcelain, the core is made from fine grained particles that are mixed with water to form a suspension referred to as a ̋ slip ̏ . The slip is then placed on a gypsum die and fired at 1120 °C for 10 hours 5, usually the gypsum die shrinks more than the condensed slip and the piece can be separated easily after firing. The fired porous core is infilitrated with glass, a process in which molten glass is drawn into the pores by capillary action at high temperature 14. Aluminous veneering porcelain is then applied using conventional powder-slurry techniques to create the proper shade and contour 5 ( Fig 1.6).

10

Advantages of Vita In-Ceram ceramic restorations 13, 5: •

High strength.

•

Good fit.

Disadvantages of Vita In–Ceram ceramic restorations 5: •

High initial cost.

•

Long processing time.

•

The core is highly dense and so cannot be etched or bonded to dentine.

1.2.4.6 Milled ceramic restorations:

.Fig 1-6 Cross-section of an In-Ceram (glass infiltrated core) crown1 .

11 Computer technologies introduced in 1987, have enriched dentistry with new ways of producing ceramic cores for full crowns. The goal of porcelain veneering these cores is to improve anatomy and aesthetics 15. Computer technology consists of digital image generation and data acquisition, computer-assisted milling systems and tooling systems 16

(Figs 1.7-1.10). The indication for complete ceramic restoration can be limited by

numerous problems such as low breaking strength of conventional dental ceramics and complex manufacturing techniques, with CAD/CAM system complete ceramic individual crowns are available 17. In addition high strength milled restorations allow for the use of all-ceramic restorations for multiple-unit posterior and anterior bridges.

Fig 1-7 Computer–aided optical triangle system .for data acquisition 16

Fig 1-8 Dental crown generation using computer graphics software, and computer aided design of a crown profile with five feature curves 16 .

12

16 .Fig 1-9 Design of tool path for milling of a crown 16 .Fig 1-10 The computer-aided milling system

Currently, there are two major CAD/CAM systems; in the first system, the computer assisted milling process can be used to machine the machinable ceramics directly from their blanks. In the second system, the milling process is firstly conducted from the presintered blanks of the difficult-to-machine ceramics, and then the sintering is followed to harden the ceramic prostheses 16. Advantages of CAD/CAM system restorations 18: •

Work can be done chair side.

13 •

High strength milled restorations allow for multiple-unit posterior and

anterior bridges. Disadvantages of CAD/CAM system restorations: •

High initial equipment cost 8.

•

Lack of marginal accuracy 16.

•

Require polishing to reduce surface roughness 16, 19.

1.2.4.6.1 Common CAD/CAM systems CEREC (Chairside Economical Restoration of Esthetic Ceramics) With CEREC 1 and CEREC 2 an optical scan of the prepared tooth is made with a couple charged device (CCD) camera, and a three dimension digital image is generated on the monitor, the restoration is then designed and milled. With the newer CEREC 3D, the operator recodes multiple images within seconds, enabling clinicians to prepare multiple teeth in the same quadrant and create a virtual cast for the entire quadrant. The restoration is then designed and transmitted to a remote milling unit for fabrication. While the system is milling the first restoration, the software can virtually seat the restoration back into the virtual cast to provide the adjacent contact while designing the next restoration 20.

CEREC inLab CEREC inLab is a laboratory system in which working dies are laser-scanned and a digital image of the virtual model is displayed on a lap top screen. After designing the coping or framework, the laboratory technician inserts the appropriate Vita In-Ceram block into the CEREC inLab machine for milling. The technician then verifies the fit of the milled coping or framework. The coping or framework is glass infiltrated and veneering porcelain is added 20.

14 DCS Precident (Distributed Control System) Preciscan laser scanner and Precimill CAM multi-tool milling center. The DCS Dentform soft-ware automatically suggests connector sizes and pontic forms for bridges. It can scan 14 dies simultaneously and mill up to 30 framework units in 1 fully automated operation. Materials used with DCS include porcelain, glass ceramic, InCeram, dense zirconia, metals, and fiber re-in-forced composites. This system is one of the few CAD/CAM systems that can mill titanium and fully dense sintered zirconia 20. Procera Procera/AllCeram was introduced in 1994. Procera uses an innovative concept of generating its alumina and zirconia copings. A scanning stylus acquires 3D images of the master dies that are sent to the processing centre via modem. The processing centre then generates enlarged dies designed to compensate for the shrinkage of the ceramic material. Coping are manufactured by dry pressing high-purity alumina powder against the enlarged dies. These densely packed copings are then milled to the desired thickness. Subsequent sintering at 2000 ˚C imparts maximum density and strength to the milled coping. The complete procedure for Procera coping fabrication is very technique sensitive because the degree of die enlargement must precisely match the shrinkage produced by sintering the alumina or zirconia 20. Lava Introduced in 2002, Lava uses a laser optical system to digitise information from multiple abutment margins and the edentulous ridge. The lava CAD software automatically finds the margin and suggests a pontic. The framework is designed to be 20 % larger to compensate for sintering shrinkage. After the design is complete, a properly sized semi-sintered zirconia block is selected for milling. The block is bar coded to register the special design of the block. The computer controlled precision

15 milling unit can mill out 21 copings or bridge frameworks without supervision or manual intervention. Milled frameworks then undergo sintering to achieve their final dimensions, density, and strength. The system also has 8 different shades to colour the framework for maximum aesthetics 20. Everest Marketed in 2002, the Everest system consists of scan, engine, and therm components. In the scanning unit, a reflection-free gypsum cast is fixed to the turntable and scanned by a CCD camera in a 1:1 ratio with an accuracy of measurement of 20 µm. A digital 3D model is generated by computing 15 point photographs. The restoration is then designed on the virtual 3D model with windows-based software. Its machining unit has 5-axis movement that is capable of producing detailed morphology and precise margins from a variety of materials including leucite-reinforced glass ceramics, partially and fully sintered zirconia, and titanium. Partially sintered zirconia frameworks require additional heat processing in its furnace 20. Cercon The Cercon System is commonly referred to as a CAM system because it does not have a CAD component. In this system, a wax pattern (coping and pontic) with a minimum thickness of 0.4 mm is made. The system scans the wax pattern and mills a zirconia bridge coping from pre-sintered zirconia blanks. The coping is then sintered in the Cercon heat furnace (1350 oC) for 6 to 8 hours. Veneering porcelain is then used to provide the aesthetic contour 20. Renishaw's Incise TM process 21 In this system, to be used in this study, the Incise TM dental contact scanning system digitises a stone model, which has been formed from an impression of the patient's bite taken by the dentist. Data from the scan is then sent electronically to Renishaw's milling

16 centre where the zirconia framework is manufactured. It is then returned to the laboratory for the application of veneering porcelain and colour-matching, before final fitting by the dentist. A summary of CAD/CAM systems along with the restorative materials is presented in table 1.5 Table1-5 Common restorative materials for dental CAD/CAM systems 20 Restorative Material

CAD/CAM

Dicor (fluormica)

CEREC

Vita Mark (feldspathic)

CEREC

ProCAD (leucite-reinforced)

CEREC

In-Ceram Spinell (magnesium oxide)

CEREC 3D, CEREC inLab

In-Ceram Alumina (aluminium oxide)

CEREC 3D, CEREC inLab, DCS Precident

Alumina (aluminium oxide)

Procera

In-Ceram Zirconia (zirconium oxide)

CEREC 3D, CEREC inLab, DCS Precident

Partially sintered zirconia (zirconium oxide)

DCS Precident, Lava, Procera, Everest, Cercon

Fully sintered zirconia (zirconium oxide)

DCS Precident, Everest

1.2.5 Properties of dental ceramics The properties of most ceramics are: hard, brittle, refractory, thermal insulators, oxidation resistant, and chemically stable 4. 1.2.5.1 Brittle material: The most important feature of ceramic is that it is a brittle material; the case of strength reduction in ceramic is the presence of small micro structural defects on the surface or within the internal structure 1. These flaws are especially critical in brittle materials in

17 areas of tensile stress because the stress at the tips of these flaws is greatly increased and may lead to crack initiation (Fig 1.10). Although the tensile stress is increased at the flaw tip it increases by a small amount in the ductile material which means that metals have the ability to deform to an applied stress without fracturing 22. The stress at areas far from these flaws will be much lower if flaws are absent in these areas. The flaws do not play a significant role when the material is subjected to compressive force, due to the fact that compressive stress tends to inhibit the propagation of cracks from the surface by keeping them closed 23, 24. Flaws on the surface, such as porosity, grinding roughness, and machining damage are associated with higher stresses than are flaws of the same size in interior regions, such as voiding and inclusions (Fig 1.11), therefore surface finishing of ceramics is extremely important to reduce the depth of the flaw 1. To overcome brittle fracture, strong ceramic core materials have been developed to support the weaker veneering ceramic materials 25.

Fig1-10 Influence of tensile and compressive stresses on flaws in brittle and . ductile material 1

18

Fig 1-11 Catastrophic fracture due to internal surface . cracks 18

1.3 Composites Composite resin has been used for nearly 50 years as a restorative material in dentistry; the use of this material has recently increased as a result of consumer demands for aesthetic restorations, coupled with problems associated with mercury containing dental amalgam 26. 1.3.1 Composition

19 Resin composites are composed of four major components. Organic polymer matrix, inorganic filler particles, coupling agent, and the initiator accelerator system 27. The organic polymer matrix: In most composites is either an aromatic or urethane dracylate oligmer, the two most common oligmers that are used in dental composites are bisphenol A-glycidyl methacrylate (bis-GMA), and urethane dimethacrylate (UDMA). Oligmers are viscous liquids the viscosity of which is reduced to a useful clinical level by the additional of a diluent monomer 2. The simplified formula for (UDMA) is 8:

Where R may be any of a number of organic groups, such as methyl-, hydroxyl-, and carboxyl. The oligmers have in common reactive double bonds at each end of the molecule, which are able to undergo addition polymerization in the presence of free radicals 8. The fillers: The dispersed inorganic particles such as glass or quartz (fine particles) or colloidal silica (micro fine particles). The benefits of fillers include: • Reduction

in

thermal

expansion

and

contraction,

and

reduction

in

polymerization shrinkage 1. • Reinforcement of the matrix resin, resulting in increased hardness, strength, and decreased wear 28. • Reduction of water resorption, softening, and staining 28. Coupling agent: A bond between the filler and matrix in the set composite is achieved by the use of a silicone silane coupling agent, silane enhances filler-resin bond strength

20 by promoting the wetting of the etched filler surface and facilating the diffusion of the fluid composite resin into the retentive space among the exposed fibres 29. Bonding is accomplished by the manufacturer, which treats the surface of the filler with a coupling agent before mixing it with the oligmer. During polymerization double bonds on the silane molecule also react with the polymer matrix. A bond between filler and matrix allows the distribution of stresses generated under function, bonding also enhances the retention of the filler particle during abrasive action at the composite surface. This bond can be degraded by water absorbed by the composite during clinical use 2. Initiator and accelerator: Polymerization of composites may be achieved by chemical means (self-curing), visible light activation, or via dual cure which is a combination of light and chemical cure 7. Other ingredients: Inorganic oxide pigments are added to composites in small amounts to provide a range of standard shades. Polymerization inhibitors and stabilizers are added to the composite to lengthen shelf life 8.

1.3.2 Classification of dental composites Resin composites are often classified according to the size of the ceramic filler particles (Fig 1.8). 1.3.2.1 Macrofilled resin composites: This is a composite containing fine-sized filler particles 0.5-300 µm in diameter, which occupies 60-65 % by volume, 70-80 % by weight of the composite 30. Comparing this composite with those of unfilled acrylic materials, the compressive strength, tensile

21 strength, hardness and elastic modulus were increased, while polymerization shrinkage and thermal expansion coefficient were decreased 31. However these composites have a rough surface, which is very difficult to polish, also they have a tendency to discolour due to the susceptibility of the rough textured surface to retain stain 30, and in addition, high wear rates due to the relatively large interparticle distances which results in breaking or plucking out of filler particles 1. 1.3.2.2 Microfilled resin composites: Contain spherical colloidal silica particles 0.01-0.12 µm in diameter, filler content may be increased and properties improved by grinding a polymerized microfilled composite into particles 10-20 µm in diameter and subsequently using these reinforced particles as filler alongside the colloidal silica to improve the handling characteristics 8. This composite exhibits a smooth surface and so it has become popular for aesthetic restorations of anterior teeth particularly in non-stress situations 30. However because of 50-70 % by volume of these materials is made up of resin, this results in a higher coefficient of thermal expansion, greater water absorption, and a decrease in the elastic modulus 31. The resin matrix lowers the mechanical properties. Thus, indications for microfilled composites usually are limited to low stress-bearing anterior restorations 31.

1.3.2.3 Blended composites: Three or more types of filler: macrofiller, microfiller, and pre-polymerised resin particles are found in blended composites. These types of composites can be polished more than would be expected for their particle size if they are refined with a dry discs. This is due to the heat generated from finishing which smears the resin in the prepolymerised particles on the surface. This smeared resin surface wears away quickly,

22 exposing the macrofiller particles just below the surface, intermittent re-polishing is necessary for these materials in order to maintain a smooth surface 30. 1.3.2.4 Hybrid resin composites: Combinations of colloidal and fine particles fill the matrix between fine particles resulting in a filler content of around 60-65 % by volume 8. The hybrid composites differ from blended composites because they do not contain pre-polymerized resin particles. The increased filler loading of hybrid improves the stress transfer between particles in the resin composite; resulting in a resin that acts more like an adhesive (nonstress-bearing) and less like a matrix (stress-bearing) 30. The hybrid composites are widely used for stress-bearing restorations 1 (Fig 1.12). 1.3.2.5 Microhybrid resin composites: The generic term for advanced composites, are combinations of micro-filled and fine particle composites and are so called because of their small diameter (0.4-1.0 µm) filler particles. They were introduced as all-purpose universal composites offering both aesthetic and superior wear resistance for use in anterior and posterior teeth 8. 1.3.2.6 Nanofiller composites: These are small particle hybrid composites which have filler particles with an average particle size of less than 1 µm, and a typical range of particle size of 0.1-6.0 µm. The particle sizes are below the wavelength of visible light, so they do not scatter or absorb visible light

32

. By using nanotechnology it is possible to improve the important

properties of composites and provide a dental composite system that offers high translucency, high polish and polish retention 33, 34. In addition nanofiller particles that have not been treated with a silane coupling agent have been found to lead to lower polymerization stress levels 32, 33.

23

.Fig 1-12 The micro structure of resin composite materials 8

24 1.3.3 Average properties of two types of resin composites (Table 1.6) 8: Table 1-6 Average properties of microfilled and microhybrid composites Property Inorganic filler content (vol %)

Microfilled

Microhybrid

20-25

60-70

Insulator

Insulator

Coefficient of thermal expansion (/°C×10-6)

50-68

20-40

Hardness (Knoop)

22-36

50-60

Water sorption (mg/cm²)

1.2-2.2

0.5-0.7

Compressive strength (Mpa)

225-300

300-350

Tensile strength (Mpa)

25-35

35-60

Young's modulus (Gpa)

3-5

7-14

Polymerization shrinkage (%)

2-4

1.5-1.7

Thermal conductivity

25

1.4 Finishing and polishing of restorative dental materials 1.4.1 Definitions Abrasion: Is the process of wear on the surface of one material by another material by scratching, tumbling or other mechanical means, the object causing the wear is called the abrasive (e.g. sandpaper); the material being abraded is called the substrate 8

(Fig

1.13). Grinding: Is the gross reduction of the surface of a substrate by the process of abrasion; it is usually performed with a large particle size abrasive. The surface texture of the substrate after grinding is usually rough to the touch and gives a different reflection to incident light 8. Erosion: Erosion wear is caused by hard particles impacting on a substrate surface carried by either a stream of liquid or a stream of air such as occurs when sandblasting a surface 1. Finishing: Refers to the process by which a restoration is contoured to remove excess material and produce a reasonably smooth surface 7. Polishing: Is a process of providing lustre or glass on a material surface 1. 1.4.2 Types of abrasives: There are two types of abrasives: •

Natural abrasives: Such as quartz, sand, pumice, diamond, and chalk.

•

Manufactured abrasives: Which are synthesized materials that are

generally preferred because of their more predictable physical properties. Silicone carbide, synthetic diamond, and aluminum oxide are examples of manufactured abrasives. 1.

26 Fig 1-13 Illustration of two-body abrasion, and hard particle erosion (A) two body abrasion occurs when abrasive particles are tightly bonded to the abrasive instrument that is removing materials from the substrate surface (B) three-body abrasion occurs when abrasive particles are free to translate and rotate between two surfaces (C) hard particle erosion is produced when abrasive particles are propelled against a substrate by air pressure 1 .

1.4.3 Factors affect the rate of abrasion An important consideration in cutting is the difference between the hardness of the abrasive and the hardness of the substrate, large difference allows more efficient grinding to take place. Another factor is the shape of the particles as irregular shaped particles abrade a surface more rapidly, and it will also produce deeper scratches than rounded particles 7. Particles usually are classified by their size: fine (0-10 µm), medium (10-100 µm), and coarse (100-500 µm). In general under the same conditions, large particles abrade a surface more rapidly than smaller ones. The speed of the abrasive also influences the cutting efficiency, faster speeds result in faster cutting rates. However, this tends to create higher temperature at the surface of the substrate. Pressure should be considered as well, the greater the pressure applied the more rapid will be the abrasion for a given abrasive. Greater pressure produces deeper and wider scratches and creates higher temperature. Heat build-up can be reduced by the use of a lubricant (e.g. silicone, water spray) which washes away debris to prevent clogging of the abrasive

27 instrument; care must be taken when using a lubricant as too much can reduce the abrasive rate 7, 8. 1.4.4 Benefits of finishing and polishing restorative materials: Good polishing is valued as one of the most important criteria for the success of the dental restorations as poor polishing lead to clinical problems such as; gingival sensitivity, secondary caries, pulp infection and periodontal disease surface roughnesses influence the marginal integrity posterior restorations

36, 37

35

.

Moreover

, and wear behaviour of

36

. Furthermore a smooth surface texture is important for the

colour of the restoration since a smooth surface will reflect a greater amount of light than rough surface 38. In addition polishing improves mechanical properties of dental ceramics 39.

1.5 Roughness Roughness consists of surface irregularities which result from the various machining processes. These irregularities combine to form surface texture. 1.5.1 Roughness Height: It is the height of the irregularities with respect to a reference line. It is measured in millimeters or micrometers. It is also known as the height of unevenness 40 (Fig 1.14). 1.5.2 Roughness Width: The roughness width is the distance parallel to the nominal surface between successive peaks or ridges which constitute the predominate pattern of the roughness. It is measured in millimeters 40 (Fig 1.14).

28

.Fig 1-14 Surface roughness 40 1.5.3 Types of roughness The resultant roughness produced by a machining process can be thought of as the combination of two independent quantities: •

Ideal roughness

Ideal surface roughness represents the best possible finish which can be obtained for a given tool shape and feed. It can be achieved only if the built-up-edge and inaccuracies in the machine tool movements are eliminated completely 40. •Natural roughness In practice, it is not usually possible to achieve conditions such as those described above, and normally the natural surface roughness forms a large proportion of the actual roughness. One of the main factors contributing to natural roughness is the occurrence of a built-up edge. Thus, the larger the built up edge, the rougher would be the surface produced, and factors tending to eliminate or reduce the built-up edge would give improved surface finish 40.

29

1.5.4 Factors contributing to overall surface roughness The defects or features which contribute to surface roughness may be random or regular (Fig 1.15). For surfaces produced by a grinding or polishing process, the most obvious roughness is the unevenness of the surface itself, i.e. scratches, pits and ridges. In the case of non random wear processes such as diamond turning, these surface imperfections may have an inherent symmetrical pattern, rather like a smooth version of a phonograph record 41.

Random unevenness

Dirt particles

Film

Regular grooves or features

Cracks

.Figure 1-15 Examples of the common factors contributing to overall surface roughness 41 1.5.5 Measurement of surface roughness 1.5.5.1 Optical microscopes: The optical microscope, often referred to as a "light microscope", uses visible light and a system of lenses to magnify images of small samples. Optical microscopes are the oldest and simplest of the microscopes 42. 1.5.5.2 Examples of optical microscopes that are used to measure surface roughness: Profilometer: There are two types: contact and non-contact, according to whether or not there is direct contact of the profilometer probe with the specimen 42.

30 •

Contact profilometers:

A related instrument is the stylus profilometer. As the name suggests, in this type of instrument a pre-loaded stylus or needle (usually diamond) is dragged across the surface. The resultant vertical motion of the stylus compresses a piezoelectric element which generates a fairly linear voltage response. This is a good method for looking at small areas or single transacts of very hard surfaces. It is not suited for soft surfaces such as semiconductor materials or coated and precision optics, since a heavy preload on the stylus will scratch the surface and a light preload may not be sufficient to register all surface features. Another related problem is choice of resolution. Clearly the maximum resolution (i.e. smallest surface spatial frequency) depends on the sharpness of the stylus tip. Although capable of high resolution, the sharper tips are more likely to scratch the test surface 41. •

Non-contact profilometers:

Any form of profilometry where there is no direct contact with the specimen. Some optical profilometers use a monochromatic laser light source whereas others use white light, but the basic principle of operation is the same in both cases 42, a collimated beam of light, coupled to an optical microscope system, is split and focused to a small spot on both the test surface and a reference surface. Height differences on the test surface and a reference surface result in optical path differences which are seen as light and dark fringes on a video camera or diode array detection system. By precise movement of the reference surface, phase information can be computed for the interference pattern. Analysis of the intensity and phase of the complex interference pattern yields information about the shape and roughness of the test surface 41.

31 Interferometer: Is the science and technique of superposing (interfering) two or more waves which creates an output wave different from the input waves; this in turn can be used to explore the difference between the inputs and derive quantitative data on the properties of the matter 42. Light microscopes can yield different surface roughness parameters including Ra, Sa, Rv and Rp. 1.5.5.3 Line Roughness: The average surface roughness (Ra) can be defined as the arithmetic mean value of all absolute distance of the roughness profiles from the center line within the measuring length

43

.

1.5.5.4 Area roughness (Sa): The optimal characterization of surface texture is often expressed with area roughness calculations that are made on the entire surface. Surface roughness calculations are similar to line roughness calculations but they include data in the x and y plane of the surface 44. Ra has been used as a roughness parameter in the majority of the previous studies 30, 35, 38, although other studies have criticized the use of Ra measurement as it cannot distinguish peaks from valleys, and also it does not distinguish roughness from an undulating surface 45. The roughness root mean squared (Rq) parameter has been used instead which is sensitive to peaks and valleys, additionally Rv and Rp both provide direct information about the amplitude of peaks or valleys in a surface 45. 1.5.5.5 Scanning electron microscopy (SEM): SEM uses electrons instead of light to form an image. A beam of electrons are produced at the top of the microscope by heating a metallic filament. The electron beam follows a vertical path through the column of the microscope. It makes its way through

32 electromagnetic lenses which focus and direct the beam down towards the sample. Once it hits the sample, other electrons (secondary) are ejected from the sample. Detectors collect the secondary or backscattered electrons, and convert them to a signal that is sent to a monitor, producing an image 46 (Figs 1.16-1.17).

. Fig 1-16 An example of a SEM 46

Fig 1-17 The path of the beam through the . Column 46

1.5.5.6 Why SEM is often used to determine topography 47: SEM is chosen because of its depth of focus and resolving capability. This can be evidenced by figure 1.18 which shows striking contrast between (a) optical and (b) .SEM micrograph of a radiolarian at the same magnification

A

B

.Fig 1-18 (A) Scanning Electron Microscopy (B) X-Ray Microanalysis 47

33 In the optical micrograph taken at high resolution only a section of the radiolarian is in sharp focus, whereas in the SEM image the entire specimen is in focus. For the optical microscope, the depth of focus in optical microscopy is the distance above and below the image plane over which the image appears in focus. As the magnification is increased the depth of the focus decreases. The three-dimensional appearance of the specimen image is as a direct result of the large depth of focus of SEM. The other useful feature of SEM is its resolving power which is orders of magnitude better than that of an optical microscope because the wavelength of the probing beam is orders of magnitude smaller. SEM examination can yield important information such as crystallographic information about the atoms arrangement in the object, as well as the surface features of an object or "how it looks" (topography), in addition to the shape and size of the particles making up the object (morphology), composition can be known as well which is the elements and compounds that the object is composed of and the relative amounts of them 48.

34

1.6 Literature review 1.6.1 Surface roughness of restorative dental materials. A considerable amount of literature has been published on the surface roughness of restorative dental materials; however the aims and the results of the studies have not always been in agreement. 1.6.1.1 Different purpose of studies comparing polishing of restorative materials: A number of studies have evaluated the effect of various finishing and polishing systems on the surface roughness of a selected type of ceramic or composite 38, 49-51. Other studies used one or more polishing system with a range of different types of ceramics or composites 52, 53. A small number of studies tended to compare the polished surface of ceramic with the glazed surface 54, 55. A limited number of studies have investigated the use of the same polishing system with different restorative materials. For example, Baseren 56 evaluated the effect of several finishing and polishing procedures on the surface roughness of nanofill, nanohybrid composites and ormocer-based dental restorative materials, they found that the best results were achieved by the same polishing kit (Super-snap abrasive discs). Another study assessed the finishing and polishing of a hybrid composite and a glass-ceramic, the investigators reported that the ceramic specimens were able to be polished to lower roughness values than the composite 57. 1.6.1.2 Methods of finishing and polishing: The polishability of dental materials in relation to polishing systems is normally tested in vitro using flat specimens prepared with dental handpieces, a defined rotation speed and a pre-defined polishing time. Most of the studies have used currently available commercial manual finishing and polishing systems. One study investigated the efficacy of using laser radiation for the

35 polishing and coating of dental ceramics and it was reported that the topography of dental ceramic surfaces could be significantly reduced by laser

58

. Nevertheless, the

laser-treated surface was in no case completely fused and the formation of microcracks was observed. Therefore, after using the laser for finishing dental ceramic alloys, further polishing of the laser-treated ceramic surfaces would be required 58. 1.6.1.3 Surface roughness parameters: Most of the studies have used surface profilometry for quantitative evaluation and SEM for qualitative evaluation

35, 50, 53

35, 51-52

,

. A recent study by Al-shammery and Bubb 45

has used confocal laser scanning microscopy to assess surface roughness of a commercial and experimental dental ceramic. 1.6.1.4 Sample size, polishing time, and effect of pressure: Most of studies used a ready made mold in different diameters (5x2 mm) 53, 56 (10x2 mm) 50, (6x6 mm) 35, (10x4 mm) 59 to fabricate the specimens. Bizar and Anglada used shade guide discs 51. Prefabricated specimens were also used by some researchers 60. Regarding polishing time some researches did not standardize the polishing time, they polished their specimens until it was shiny to the naked eye 50. Others used a limited measured time with the applied polishing systems 56, 51 e.g. 30 s was applied for each step of finishing and polishing whilst 20 s was used by Ozgunaltay 53. Sasahara

35

used

different times with different polishing procedures (20 s and 30 s). The finishing direction in all of the reviewed studies was either light in a single direction 52 or in a multiple directions

53-54

. Polishing load with the applied polishing systems was in most

of the studies described as constant light manual pressure

59, 61

, while Kou and Molin

60

controlled the load on the specimens during polishing and finishing and calibrated it using a balance. With most of the finishing/polishing systems no specific polishing

36 times are indicated, manufacturers leave it down to the dental technician skill and experience. 1.6.2 Marginal fit of all-ceramic crowns generated by CAD/CAM systems: All-ceramic core crowns offer excellent aesthetics and have been used successfully for restoring anterior as well as posterior teeth. Apart from fracture resistance and aesthetics, marginal and internal accuracy of fit is valued as one of the most important criteria for the success of all-ceramic crowns. Increased marginal discrepancy of a crown favours the increased rate of cement dissolution and of microleakage causing inflammation of the vital pulp

62

. In addition, a poor marginal adaptation of crowns

increases plaque retention and changes the composition of the subgingival microflora indicating the onset of periodontal disease 63. Moreover, misfit in the axial wall area and occlusal plateau can reduce the resistance to fracture of all-ceramic restorations and thus reduce longevity 17. In order to prevent crown fracture it is not only necessary to use as strong a material as possible (which is possible with CAD/CAM systems), but also fabricate the crowns with the best fit possible 64. There have been a number of studies investigating the marginal fit of all-ceramic crowns. Some studies indicated that the fitting of all-ceramic crown copings gives an indication of the fitting of the whole crown 62, while other studies did not rely on the core fitting and investigated the fitting of the whole crown after each of the different fabrication steps

64

. In their study, Groten and Girthofer

65

demonstrated that

manufacturing steps after copy milling have no obvious influence on the external gap width.

37 1.6.2.1 Die fabrication: Different methods have been used to fabricate the dies; these include duplication of human teeth 66, acrylic model teeth 62, and the base metal alloy die

65

in order to prepare

full crowns. 1.6.2.2 Finish line design: Regarding the finish line design, many investigations have reported that the finish line design i.e. chamfer, shoulder, beveled shoulder, has no effect on the marginal discrepancy of a conventional full crown 67. While another study by Bottino 68 found that the best cervical adaptation was achieved with the chamfer type of finish line. For allceramic crowns fabricated by CAD/CAM systems all the reviewed studies found that the finish line preparation design had no effect on marginal adaptation 36, 66. 1.6.2.3 Methods for measurement of the marginal fit: Methods for measurement of the marginal fit include: Cross sectional view which has been used to measure the marginal fit of cemented -1 and sectioned restorations, 2- direct view of the crown on a die which is a non destructive technique and has been used to measure the distortion during manufacturing process of the restoration, 3- impression replica technique used to simulate crown gap space after cementation. The copings are filled with light body silicone impression material and inserted in place on their die. A constant load is then uniformly applied to maintain the seated position of the copings. After full setting of the material, the copings are removed from the primary casts with the light body silicone film firmly attached to them. The assembly (coping and silicone film) is then reinforced and stabilized by filling its inner side with heavy body silicone impression material, resulting in a combination of a light and heavy body silicone core ready to section for measurement of the thickness of the light body silicone impression material

38 69

. The final method is the clinical examination, where a clinically acceptable gap is not .visible to naked eye and is not detectable with a sharp explorer 70

1.6.2.4 Measurement methods used to recorded marginal fit: Regarding the measurement methods used to determine the marginal and internal gap, numerous studies of CAD/CAM restorations have used light microscopy 17, 53, travelling microscopy 36, SEM

63, 65, 66

or microscopic digital imaging systems

15

to measure the

marginal gap between the restoration and the tooth either directly or from an epoxy replica. 1.6.2.5 CAD/CAM systems and innovation and conventional systems: A few studies which used different CAD/CAM systems reported that the marginal fit of all-ceramic crowns generated by this technique is in the range of clinical acceptance and similar to other innovative and conventional systems 15, 17, 63, 66.

39

1.7 Aims and objectives The surface finish and marginal fit of restorative dental material have an extensive effect on the overall success of the restoration. Other studies have emphasized the polishing quality of chosen polishing systems; however the aim of this study was to understand the relationship between the polishing system and the polished restorative material. Ceramic and composite discs will be fabricated, and two polishing kits will be used (EnamelPlus HFO, CeraMaster). Specimens will be viewed using scanning electron microscopy and white light interferometery. ANOVA and Tukeys HSD test will be used to analyse the data. For the second part of the study stone dies will be fabricated. The Renishaw Incise machine will be used to scan the preparations from which all-ceramic cores will be fabricated. Veneering ceramic will be used to build up the crowns and the marginal fit assessed by the same microscopes used in the first part of the study. Data will be analysed using the same statistical tests.

40

Chapter two Methodology 2.1 Methodology of surface roughness measurements 2.1.1 Specimen's fabrication 2.1.1.1 Ceramic specimens' fabrication: Thirty metal bonded porcelain specimens were fabricated, 15 specimens each of Duceram (Duceram, DuguDent, Germany), and DuceraGold (DuceraGold, DuguDent, Germany), using a plastic mold (5 mm in diameter and 2 mm thick). The porcelain was mixed with sculpting liquid, and placed into a ready made mold. The excess moisture was absorbed by using tissue (Georgia-Pacific, Sheffield, UK). After removal from the mold, the specimens were placed in a porcelain-firing furnace (Ivoclar, Programmate P100, Liechtenstein, UK), and fired according to the manufacturer's instructions (Tables 2.1-2.2). Table 2-1: Firing recommendations for DuceraGold ceramic: Stand by temp (°C)

Closing time (min)

Heating rate (°C/min)

Final temp (ºC)

Holding time (min)

Dentine

403

6:00

55

780

1:00

Auto glaze

403

4:00

55

770

1:00

Table 2-2: Firing recommendation for Duceram ceramic: Stand by temp (°C)

Closing time (min)

Heating rate (°C/min)

Final temp (ºC)

Holding time (min)

Dentine

550

6.00

80

980

1.00

Autoglaze

650

5.00

60

920

3.00

41 Specimens were allowed to cool and then were finished with a medium-grit diamond (Jacobs Ltd, Surry, UK), with a slow speed handpiece (Kavo, Germany), rotating at approximately 10,000 rpm to remove surface irregularities for a maximum of 20 seconds, (stop clock was used). The specimens were soaked in distilled water for five minutes, and dried with a blast of air for 20 seconds, and then placed into the porcelain-firing furnace to obtain autoglaze, in which autoglazing, the surface of the ceramic itself was allowed to melt at a high temperature to provide the glazed layer (Tables 2.1-2.2). The fifteen specimens of each ceramic material were divided into three groups of 5 specimens. One group served as a control and had no surface treatment, in the experimental two groups; the glaze layer of one side of each specimen was removed using a medium-grit diamond. To differentiate the two surfaces, each specimen was marked on the non working surface with a blue permanent marker. Specimens were again soaked in distilled water for five minutes and allowed to dry; an initial surface roughness was evaluated by using white light interferometer (Section 2.1.4). The specimens were then finished and polished using one of the two polishing kits (Section 2.1.2). 2.1.1.2 Zirconia ceramic specimen's fabrication: Ten machined (auto milling procedure) disc-shaped specimens were used (5 mm in diameter and 2 mm thick), (Renishaw, Gloucestershire, UK). 2.1.1.3 Composite specimen's fabrication: A light cured microhybrid resin composite was used in this study (Table 2.3). Fifteen disc-shaped specimens (5 mm in diameter and 2 mm thick) were prepared in the ready made mold.

42 :Table 2-3 Composite resin material details

Composite resin

Mean particle size (µm)

Type

Tender fiber

Microhybrid

Batch number

0.012

0297

Manufacturer GDF Gmbh, Rosbach, Germany

Composite resin was inserted into the mold in two increments, covered with Mylar strip (Directa AB. Upplands Vasby. Sweden), pressed flat using a plastic slide, and light cured for half an hour (TRIAD, Dentsply), the polymerized specimens were then stored in distilled water at 37 °C for 24 hours. The baseline average roughness values (Ra, Sa) were measured on each specimen immediately after light curing under Mylar strip by means of an interferometer (Section 2.1.4). The 15 specimens were divided into three groups of 5 specimens; one group had no surface treatment and served as a control as curing composite under Mylar strip gives the smoothest surface possible. The specimens were then finished and polished using one of the two polishing kits (Section 2.1.2). 2.1.2 Finishing and polishing kits: The two polishing systems used were: CeraMaster polishing kit, and EnamelPlus Shiny (Table 2.4). Table 2-4 The two polishing kits: Product name

Batch no

Manufacturer

CeraMaster

0044

Shofu Dental products Ltd, UK

EnamelPlus Shiny

0123

Micerium S.p.A, Italy

43 2.1.2.1 Shofu’s CeraMaster: Together Shofu’s CeraMaster Coarse and CeraMaster finishing and polishing points provide a simple two step system for finishing, polishing and super polishing porcelain and enamel restorations either in the lab or in the clinic (Fig 2.1). CeraMaster (white and blue bands) silicone rubber polishers are filled with a high density of pure diamond particles 71, grain size of these particles was not specified by the manufacturer.

A

B

Fig 2-1 (A) CeraMaster coarse for finishing and polishing 71 .

(B) CeraMaster for super polishing available in different shapes 71 . in bullet, knife, cup, and mini

point.

2.1.2.2 Enamel plus Shiny 72: The system includes the use of Rotative instruments: The rotative instruments consist of diamond burs and abrasive • soft silicone rubbers. Quality and grain measure of the abrasive material were not .specified by the manufacturer Diamond pastes: The shiny pastes are available in two grain measures: 3µm (paste A) • .(and 1 µm (paste B .(Goat hair brushes and felts (Fig 2.2 •

.Fig 2-2 EnamelPlus kit 72

44 2.1.3 Finishing and polishing procedures In this study 5x2 mm ceramic and composite disks were fabricated. Because different restorative materials were used, and the manufacturers did not recommend an exact polishing time with correspondence materials, and based on researcher experience, specimens size, and the review of the literature of section 1.6.1.4, 20 s was chosen to be applied for each finishing and polishing step except for the last polishing step where the time was adjusted but still measured according to the material used and system applied until all the specimens looked shiny to the naked eye. The application of the two techniques allowed standardization of steps and a better comparison of results. Manual constant pressure was applied in one direction. :(Finishing and polishing procedure 1 (CeraMaster kit 2.1.3.1 Specimens were surfaced with a medium grit finishing diamond. They were then treated for 20 s with blue diamond-based polisher points followed by a 20 s treatment using white diamond-based polisher points. With the polishing white diamond-based polisher points time was amended. (Stop clock was used) 2.1.3.2 Finishing and polishing procedure 2 (EnamelPlus kit): Specimens were surfaced with medium grit diamond. They were then treated for 20 s with an ultra fine diamond, followed by a 20 s with soft silicone rubber polisher, followed by a 20 s treatment using a goat hair brush together with diamond paste with different grains (3 µm and 1 µm). With the silicone rubber polisher time was amended. 2.1.3.3 The application of the two polishing systems The two polishing systems were applied to the specimens using a handpiece rotating at approximately 10,000 rpm (Kavo, Germany), with a light and intermittent pressure controlled by the operator as much as possible. A new bur was used after application on five surfaces. Specimen preparation, finishing and polishing procedures were performed

45 by the same investigator. After each step of finishing and polishing the specimens were ultrasonically cleaned (Ultrawave Ltd. Cardiff, UK) with distilled water for five minutes, and dried with a blast of air for 20 seconds before the surface roughness measurements were recorded. The two polishing systems were applied to the specimens of the different restorative materials following the same procedures (procedure 1 and procedure 2). The medium grit diamond was used with great care on the composite as recommended by the manufacturer. A rubber bur was dipped in pumice and applied to the specimens in a bath of water to avoid burning the resin matrix of the composite material by the generated heat. With the CeraMaster polishing technique in the last step of polishing, 20 s was not considered sufficient to produce a shiny surface as observed with the naked eye. Additional polishing was carried out to produce a shiny surface with the ceramic (5-7 s extra), and with composite specimens (7-10 s extra). Surface roughness measurements were then repeated on these materials. However with the EnamelPlus polishing kit, 1015 s of polishing with rubber polisher was sufficient to produce a shiny surface and this agreed with the manufacturer’s recommendation that the surface must be treated at a very slow speed, with a short working time. However for the ceramic specimens the investigator had to increase the polishing time by 10 s to 30 s in order to achieve a shiny surface comparable with the other specimens. For zirconia ceramic no extra time was required to achieve a shiny surface to the naked eye with either of the polishing systems. New surface roughness readings of the specimens were carried out after each additional finishing/ polishing treatment.

46 2.1.4 Surface roughness measurement: A white light interferometer (Micro XAM, ADE phase shift, Tucson, Arizona) was used to measure the mean roughness values Ra and Sa (µm). The scan height was set to 20 mm, the horizontal width was 86.14 µm, and the vertical height was 64.07 µm. Specimens were visualized at × 100 magnification. Four measurements in different directions and perpendicular to the finishing and polishing scratch directions were recorded (8 measurements per specimen) for the control specimens (no surface treatment) and for all the other specimens after each finishing and polishing treatment. The mean Ra and Sa values were determined for each specimen, and an overall mean and standard deviation values determined for the specimens in each group. Ra was measured to allow direct comparison with previous studies as this parameter has been frequently used 35, 50-51. However the option to record Sa was also available with this software, enabling the topography of a feature surface area to be measured. 2.1.5 SEM evaluation SEM (JEOL, JSM-5600 LV, Japan) was used for qualitative evaluation.

47

2.2 Methodology of marginal fit assessments of zirconia based allceramic crowns 2.2.1 Master die fabrication A prepared right central pourable polyurethane based model resin tooth (Nissan, Japan) with a shoulder line design was selected in order to prepare two master casts. The acrylic tooth was positioned on a wax base, and enclosed in a plastic ring before taking an impression using a silicone based impression material (Shera Duosil H, Germany), portioning and mixing were done in accordance to manufacturer instructions (mixing ratio P:W 100 g/25 ml). After setting of the impression material, the ring and the plastic tooth were removed. To permit a more accurate flow of the die stone, a surface tension reduction agent (Yeti, Germany) was applied to the surface of the silicone impression. The mixed die stone (Zhermack S.p.A of Italy) was carefully vibrated (Bracon, East Sussex, UK) into the impression and left to set for 30 minutes. The two produced casts were suitable for a full-coverage crown with a shoulder line design. The marginal line of one cast was changed manually to a chamfer line design using a sharp carver, resulting in two master casts with two different marginal designs: (1) a chamfer, (2) a 90˚ shoulder (Fig 2.3).

Fig 2-3 Two different marginal designs 36 .

48 An impression was taken of each cast to fabricate a series of 5 replica dies for each marginal design using the materials and procedure previously described in this section. The stone dies were sealed with die hardener (Clear Coat Model hardener, Yeti, Germany) 2.2.2 Dies scanning: A scanning design device (Incise machine, Renishaw, UK) was used by the investigator. All scanning and designing procedures were repeated for each replica separately; the scanner was calibrated at the beginning of the study and recalibrated with each die with the use of calibration specimens provided with the system. The luting space was set to a 0.07 mm, and the core thickness was set to thin (0.25 mm). The acquired data was transmitted to the machining centre (laboratories of Renishaw, Gloucestershire, UK). The cores were machined and sent back to the investigator. Each core was placed on the respective stone die without applying any additional pressure, the fit of the crown on its respective tooth was assessed visually. No adjustments were needed to ensure complete seating. 2.2.3 Crown fabrication (building up): Veneering ceramic (Lava and 3M ESPE, Leicestershire, UK) was used to build up the crowns. The framework modifier was mixed with modelling liquid, a thin coat (estimated to be 0.1-0.2 mm) was applied to the entire surface of the core as the manufacturer recommended, and gentle vibration was performed to avoid bubble entrapment. Dentine and incisal materials were mixed with modelling liquid and applied to the framework modifier while the core was seated on the die. The first firing was carried out in accordance with the manufacturer’s recommendation (Table 2.5).

49 2.2.4 Crowns finishing and glazing: Medium-grain diamond burs were used at low speed (10,000 rpm) to finish the crowns. The final glazing firing was completed in accordance to manufacturer’s recommendations (Table 2.5). Table 2-5 Firing recommendation for Lava ceramic: Stand by temp (°C)

Closing time (min)

Heating rate (°C/min)

Final temp (°C)

Holding time (min)

Dentine and incisal materials

450

6:00

45:00

810

1:00

Auto glaze

480

2:00

45:00

820

1:00

2.2.5 Evaluation of the marginal adaptation using SEM: SEM (JEOL, JSM-5600 LV, Japan) was used to visualize and record micrographs of the marginal gaps (the distance from the internal surface of the crown margin to the preparation finish line) on all four axial walls of the cores with 15 measurements on each crown, which equates to 75 measurements per test group and 150 measurements for all the ten specimens for each stage of fabrication. A representative SEM micrograph is presented in figure 2.4. The working distance used was 20 mm, with a spot size set at around 30. Magnification varied from × 200 to × 1000 depending on image quality. Measurements of the marginal gap were recorded when the margins of both the core and the stone were clearly visible (Fig 2.4). Where the shadow of one parts (core or die) was observed on the corresponding part, measurements were avoided (Fig 2.5). Measurements were taken after the following stages of fabrication: 1- The fabrication of the core. 2- Dentine and enamel application. 3- Glazing.

50 The image processing and analysis program ImageJ (Java, U.S. National Institutes of Health) was used to carry out the measurements. All measurements were performed by the same operator to avoid undue errors as much as possible. The investigator was blind to the description of the micrographs and the test groups at the measurement stage .

Ceramic core Stone die

Fig 2-4 A representative SEM micrograph of the marginal gap (330 × magnification). The lines represent the distances measured using .ImageJ software

Ceramic crown

Stone die

Fig 2-5 Ceramic core shading on the stone die making this image unsuitable to record a measurement.

51

Chapter three Results 3.1 Surface roughness measurements of restorative materials The mean of the Ra and Sa values (µm) and standard deviations (SD) are presented in tables 3.1-3.6, and graphically depicted in figures 3.1-3.4 for the two polishing systems. Raw data is presented in Appendix 1. 3.1.1 Finishing and polishing procedure 1 (CeraMaster kit): Table 3-1 Overall Ra means (µm) ± the standard deviations recorded after finishing/polishing with CeraMaster polishing kit (n=5 per material). Initial reading

Blue polisher

White polisher

Extra polishing

Duceram ceramic

1.23 ±0.20

0.73 ±0.12

0.25 ±0.09

0.18 ±0.06

DuceraGold ceramic

1.53 ±0.27

0.89 ±0.18

0.35 ±0.09

0.21 ±0.08

Composite

1.80 ±0.58

0.91 ±0.21

0.53 ±.0.09

0.39 ±0.10

Zirconia ceramic

0.54 ±0.11

0.30 ±0.05

0.16 ±0.05

Table 3-2 Overall means Sa (µm) ± the standard deviations recorded after finishing/polishing with CeraMaster polishing kit (n=5 per material). Initial reading

Blue polisher

White polisher

Extra polishing

Duceram ceramic

1.42 ±0.19

0.58 ±0.09

0.29 ±0.09

0.21 ±0.06

DuceraGold ceramic

1.73 ±0.26

1.00 ±0.07

0.40 ±0.08

0.24 ±0.07

Composite

1.69 ±0.44

1.01 ±0.35

0.55 ±0.09

0.42 ±0.11

Zirconia ceramic

0.70 ±0.12

0.37 ±0.04

0.20 ±0.04

Figures 3.1-3.2 present the effect of consequential finishing and polishing steps of the CeraMaster polishing system on each of the restorative materials used.

52 The plot of overall mean of Ra (Fig 3.1) shows that the composite had the highest initial surface roughness base line value between the four restorative materials used (1.80 µm), while zirconia ceramic had the lowest value (0.54 µm); the composite specimens continued having the highest surface roughness values with the last observed value of 0.53 µm. The effect of this polishing technique on the two veneering ceramics was similar. The lowest final result was observed with zirconia ceramic (0.16 µm) which did not require further polishing to give a shiny appearance comparing to the other three restorative materials tested (Fig 3.1). The plot of overall mean of Sa (Fig 3.2) shows that DuceraGold ceramic had marginally the highest initial surface roughness value (1.73 µm), but showed similar trends to figure 3.1. Although Sa values were generally higher than Ra values, probably due to the larger surface area being measured, as opposed to linear Ra readings. For some restorative material types (e.g. composite) there was considerable variability between specimens as indicated by large error bars particularly foe initial readings (Figs 3.1 and 3.2).

53

Fig 3-1 The mean Ra (µm) with error bars denoting standard deviation of restorative materials after finishing/polishing with CeraMaster polishing kit (n=5).

Fig 3-2 The mean Sa (µm) with error bars denoting standard deviation of restorative materials after finishing/polishing with CeraMaster polishing kit (n=5).

54 3.1.2 Finishing and polishing procedure 2 (EnamelPlus kit): Table 3-3 Ra means (µm) ± the standard deviations recorded after finishing/polishing with EnamelPlus polishing kit (n=5 per material). Initial reading

Ultra fine diamond

Rubber polisher

Goat hair brush

Duceram ceramic

1.34 ±0.27

1.19 ±0.22

0.42 ±0.15

0.20 ±0.07

DuceraGold ceramic

1.68 ±0.34

1.27 ±0.24

0.51 ±0.15

0.20 ±0.06

Composite

1.33 ±0.28

0.68 ±0.16

0.29 ±0.12

0.12 ±0.04

Zirconia ceramic

0.53 ±0.12

0.38 ±0.07

0.36 ±0.08

0.11 ±0.04

Table 3-4 Sa means (µm) ± the standard deviations recorded after finishing/polishing with EnamelPlus polishing kit (n=5 per material). Initial reading

Ultra fine diamond

Rubber polisher

Goat hair brush

Duceram ceramic

1.56 ±0.16

1.29 ±0.16

0.40 ±0.13

0.20 ±0.07

DuceraGold ceramic

1.79 ±0.26

1.39 ±0.16

0.51 ±0.14

0.22 ±0.07

Composite

1.56 ±0.16

1.23 ±0.16

0.32 ±0.10

0.13 ±0.05

Zirconia ceramic

0.68 ±0.13

0.42 ±0.06

0.40 ±0.07

0.10 ±0.04

55 Table 3-5 Ra means (µm) ± the standard deviations recorded for the control groups and after finishing/polishing with the two applied polishing kits (n=5 per material). Duceram

DuceraGold

Composite

Control group

0.26 ±0.10

0.25 ±0.12

0.04 ±0.02

Procedure 1

0.18 ±0.06

0.21 ±0.08

0.39 ±0.10

Procedure 2

0.20 ±0.07

0.20 ±0.06

0.12 ±0.04

Procedure 1: CeraMaster technique Procedure 2: EnamelPlus technique

Table 3-6 Sa means (µm) ± the standard deviations recorded for the control groups and after finishing/polishing with the two applied polishing kits (n=5 per material). Duceram

DuceraGold

Composite

Control group

0.29 ±0.10

0.28 ±0.10

0.06 ±0.03

Procedure 1

0.21 ±0.06

0.24 ±0.07

0.42 ±0.11

Procedure 2

0.20 ±0.07

0.22 ±0.07

0.13 ±0.05

56 Figures 3.3-3.4 present the effect of consequential polishing steps of EnamelPlus polishing system on each restorative material used. The plot of overall mean of Ra (Fig 3.3) shows that DuceraGold ceramic had the highest initial surface roughness value (1.68 µm), followed by the composite and DuceraGold ceramic (1.34 µm and 1.33 µm respectively), while zirconia ceramic had the lowest value (0.53 µm). Similar to the previous polishing system zirconia had the lowest surface roughness values at each polishing stage. But unlike the CeraMaster, surface roughness values for the composite after each polishing step were lower than the two veneering ceramics (DuceraGold and Duceram). Zirconia ceramic and composite had the lowest final values. While Duceram and DuceraGold had higher values which were comparable. The plot of overall mean of Sa (Fig 3.4) gave similar observations to figure 3.3. For both finishing/polishing techniques surface roughness parameters (Ra and Sa) gave similar indications, with Sa values (Tables 3.2-3.4) again being generally slightly higher than Ra (Tables 3.1-3.3) particularly for the initial values, with these differences decreasing as the surfaces became smoother. Error bars in these plots varied between specimens of different restorative materials under the application of the same polishing kit and between different finishing/polishing steps for the same material. The level of variability reduced as the surfaces became smoother.

57

Fig 3-3 The mean Ra (µm) with error bars denoting standard deviation of restorative materials after finishing/polishing with EnamelPlus polishing kit (n=5).

Fig 3-4 The mean Sa (µm) with error bars denoting standard deviation of restorative materials after finishing/polishing with EnamelPlus polishing kit (n=5).

58 3.1.3 Statistical analysis Data from surface roughness measurements were analyzed using the SPSS (SPSS 15.0 for windows). One way analysis of variance (ANOVA), and Tukey HSD multiple comparison test were used to determine the difference between the different restorative materials within each of the surface treatments. The T test was used to find out the effect of different finishing/polishing techniques on the same restorative material, and to determine the effect of consequential polishing steps within each polishing system on each restorative material used. A p-value of <0.05 was considered statistically significant. The statistical analysis showed significant differences (p<0.05) between the restorative materials used for each finishing/polishing technique (Tables 3.7-3.8), and between the finishing/polishing systems within each restorative material (Tables 3.9-3.12). The polishing techniques had a significant effect on the surface roughness of materials tested as determined by interferometry. The use of CeraMaster resulted in higher surface roughness values on the composite material when compared to the other material tested (p≤0.001), while no significant difference was observed between the three ceramic materials used (Duceram, DuceraGold and zirconia ceramic) (Table 3.8). Data analysis revealed that the composite exhibited the highest Sa and Ra values of all the materials used (Tables 3.1-3.2) with the CeraMaster kit. The three different surface treatments resulted in a significant difference in surface roughness values (Table 3.11); Mylar strip had the lowest Ra and Sa values (Tables 3.5-3.6). The application of the EnamelPlus system resulted in the higher final roughness values for both Duceram and DuceraGold, when compared to the composite and zirconia ceramic (Table 3.4). Composite and zirconia ceramic had lower final surface roughness

59 values after polishing specimens with this kit with no statically significant difference between the two materials (p≤0.001) (Table 3.8). EnamelPlus resulted in low final Ra and Sa values when applied to zirconia ceramic (Tables 3.3-3.4) and so a significant difference was found between the two different surface treatments (p= 0.017) (Table 3.12). Regarding the material used, for Duceram and DuceraGold ceramics there was no significant difference in surface roughness values between either the two polishing systems or between the polished and the autoglazed surfaces (Tables 3.9 and 3.10). The last statistical analysis (T test to compare between the effects of each fabrication step of the polishing system) was designed for both Ra and Sa to determine any difference between these two parameters. No significant differences were found between the majority of Ra and Sa data (Tables 3.13-3.16). Only one difference was observed with the Ra showing that extra polishing of Duceram ceramic was statistically different (p=0.044) (Table 3.13), while the Sa result was not statically significant (p=0.223) (Table 3.14). For both polishing kits, the effect of each finishing and polishing step was found to be statistically significant, except for the extra polishing of Duceram ceramic (Table 3.14), and the polishing of zirconia with the silicone rubber polisher (Tables 3.15-3.16). For the two metal bonded ceramics no statistically significant difference was observed using either technique (Tables 3.9-3.10), while significant difference was found for composite and zirconia using EnamelPlus when compared to CeraMaster (Tables 3.113.12). Composite surface roughness values were approximately four times lower using EnamelPlus (Ra=12 µm, Sa=13 µm) when compared to CeraMaster (Ra=39 µm, Sa=42 µm). Zirconia had approximately half surface roughness values using EnamelPlus (Ra=11 µm, Sa=10 µm) compared to CeraMaster (Ra=16 µm, Sa=20 µm).

60 (Table 3-7) Summary of statistical analysis of surface roughness of the restorative materials after polishing with two polishing kits.

Polishing 1

Polishing 2

Sum of Squares

df

Mean Square

F

Sig.

Between groups

0.155

3

0.052

17.567

0.000

Within groups

0.047

16

0.003

Total

0.202

19

Between groups

0.151

3

0.050

22.957

0.000

Within groups

0.035

16

0.002

Total

0.186

19

Polishing 1: CeraMaster polishing kit Polishing 2: EnamelPlus polishing kit. Values recorded in grey represent the statistical significance

61 Table 3-8 Comparison of the four restorative materials after each finishing technique (one way ANOVA).

Dependent Variable

(I) material

1.00

2.00

Polishing 1

Tukey HSD 3.00

4.00

1.00

2.00

Polishing 2

Tukey HSD 3.00

4.00

(J) material

Mean Difference (I-J)

Std. Error

Sig.

2.00

-0.04000

0.03434

3.00

-.20600(*)

4.00

95% Confidence Interval Upper Bound

Lower Bound

0.656

-0.1382

0.0582

0.03434

0.000

-0.3042

-0.1078

0.01600

0.03434

0.965

-0.0822

0.1142

1.00

0.04000

0.03434

0.656

-0.0582

0.1382

3.00

-.16600(*)

0.03434

0.001

-0.2642

-0.0678

4.00

0.05600

0.03434

0.390

-0.0422

0.1542

1.00

.20600(*)

0.03434

0.000

0.1078

0.3042

2.00

.16600(*)

0.03434

0.001

0.0678

0.2642

4.00

.22200(*)

0.03434

0.000

0.1238

0.3202

1.00

-0.01600

0.03434

0.965

-0.1142

0.0822

2.00

-0.05600

0.03434

0.390

-0.1542

0.0422

3.00

-.22200(*)

0.03434

0.000

-0.3202

-0.1238

2.00

0.02200

0.02958

0.878

-0.0626

0.1066

3.00

.16400(*)

0.02958

0.000

0.0794

0.2486

4.00

.20000(*)

0.02958

0.000

0.1154

0.2846

1.00

-0.02200

0.02958

0.878

-0.1066

0.0626

3.00

.14200(*)

0.02958

0.001

0.0574

0.2266

4.00

.17800(*)

0.02958

0.000

0.0934

0.2626

1.00

-.16400(*)

0.02958

0.000

-0.2486

-0.0794

2.00

-.14200(*)

0.02958

0.001

-0.2266

-0.0574

4.00

0.03600

0.02958

0.625

-0.0486

0.1206

1.00

-.20000(*)

0.02958

0.000

-0.2846

-0.1154

2.00

-.17800(*)

0.02958

0.000

-0.2626

-0.0934

3.00

-0.03600

0.02958

0.625

-0.1206

0.0486

Results of Post-hoc comparison carried out using Turkey’s (HSD) test. N=5 for each tested material Material 1: Duceram ceramic Material 2: DuceraGold ceramic Material 3: Composite Material 4: Zirconia ceramic

62 Table 3-9 Comparison between different surface treatments on Duceram ceramic. Paired Differences

Pair 1 Pair 2 Pair 3

Poli 1poli 2 Poli lglazing Poli 2glazing

Mean

Std. Deviation

Std. Error Mean

-0.0120

0.0630

-0.0860 -0.0740

95% Confidence Interval of the Difference Upper

Lower

t

df

Sig. (2tailed)

0.028

-0.0902

0.0662

-0.426

4

0.692

0.0695

0.031

-0.1722

0.0002

-2.767

4

0.050

0.0918

0.041

-0.1880

0.0400 0

-1.802

4

0.146

Table 3-10 Comparison between different surface treatments on DuceraGold ceramic. Paired Differences

Pair 1 Pair 2 Pair 3

Poli 1poli 2 Poli lglazing Poli 2glazing

Mean

Std. Deviation

Std. Error Mean

-0.02800

0.07120

-0.02400 -0.05200

95% Confidence Interval of the Difference Upper

Lower

t

df

Sig. (2tailed)

0.0318

-0.06041

0.11641

0.879

4

0.429

0.07829

0.0350

-0.12122

0.07322

-0.685

4

0.531

0.05975

0.0267

-0.12619

0.02219

-1.946

4

0.124

Table 3-11 Comparison between different surface treatments on microhybrid composite. Paired Differences

Pair 1 Pair 2 Pair 3

Poli 1poli 2 Poli lmylar Poli 2mylar

Mean

Std. Deviation

Std. Error Mean

0.28800

0.09706

0.38600 0.09800

95% Confidence Interval of the Difference Upper

Lower

t

df

Sig. (2tailed)

0.0434

0.16749

0.40851

6.635

4

0.003

0.06950

0.0310

0.29971

0.47229

12.419

4

0.000

0.03271

0.0146

0.05738

0.13862

6.699

4

0.003

63 Table 3-12 Comparison between different surface treatments on zirconia ceramic. Paired Differences

Pair 1

Poli 1poli 2

Mean

Std. Deviation

Std. Error Mean

0.19800

0.11234

0.050

95% Confidence Interval of the Difference Upper

Lower

t

df

Sig. (2tailed)

0.05851

0.33749

3.941

4

0.017

Table 3-13 Comparison between mean Ra values of different fabrication steps using CeraMaster kit.

Mean Step 1-step 2

Step 2-step 3

Step 3-step 4

Std. Deviation

df

Sig. (2-tailed)

Duceram

0.554

0.07765

4

0.00

DuceraGold

0.61

0.05148

4

0.001

Composite

0.5316

0.3698

4

0.032

Zirconia

0.21

0.04359

4

0.00

Duceram

0.482

0.09365

4

0.00

DuceraGold

0.538

0.0545

4

0.001

Composite

0.4424

0.23394

4

0.013

Zirconia

0.128

0.0295

4

0.001

Duceram

0.072

0.05541

4

0.044

DuceraGold

0.126

0.0532

4

0.006

Composite

0.132

0.09935

4

0.041

Step 1= Finishing with medium grit diamond bur Step 2= Polishing with blue polisher Step 3= Polishing with white polisher Step 4= Extra polish for 5-7 s

64 Table 3-14 Comparison between mean Sa values of different fabrication steps using CeraMaster kit.

Step 1-step 2

Step 2-step 3

Step 3-step 4

Mean

Std. Deviation

df

Sig. (2-tailed)

Duceram

0.588

0.12478

4

0.00

DuceraGold

0.688

0.15611

4

0.001

Composite

0.896

0.18663

4

0.00

Zirconia

0.342

0.0687

4

0.00

Duceram

0.562

0.12091

4

0.00

DuceraGold

0.602

0.0507

4

0.00

Composite

0.378

0.07887

4

0.00

Zirconia

0.176

0.03362

4

0.00

Duceram

0.078

0.12112

4

0.223

DuceraGold

0.148

0.0705

4

0.009

Composite

0.138

0.04604

4

0.003

Table 3-15 Comparison between mean Ra values of different fabrication steps using EnamelPlus kit.

Step 1-step 2

Step 2-step 3

Step 3-step 4

Mean

Std. Deviation

df

Sig. (2-tailed)

0.158

0.1006

4

0.025

DuceraGold 0.43

0.17117

4

0.005

Composite

0.698

0.1757

4

0.001

Zirconia

0.122

0.01643

Duceram

0.764

0.09965

4

0.00

DuceraGold 0.754

0.08295

4

0.00

Composite

0.386

0.11992

4

0.002

Zirconia

0.02

0.03937

4

0.319

Duceram

0.224

0.05595

4

0.001

DuceraGold 0.286

0.09965

4

0.003

Composite

0.178

0.06979

4

0.005

Zirconia

0.256

0.05727

4

Duceram

Step 1= Finishing with medium grit diamond bur Step 2= Polishing with ultrafine diamond bur Step 3= Polishing with silicone rubber Step 4= Polishing with diamond past

0.00

65 Table 3-16 Comparison between mean Sa values of different fabrication steps using EnamelPlus kit.

Step 1-step 2

Step 2-step 3

Step 3-step 4

Mean

Std. Deviation

df

Sig. (2-tailed)

Duceram

0.276

0.0445

4

0.00

DuceraGold

0.4

0.12309

4

0.002

Composite

0.28

0.13583

4

0.01

Zirconia

0.32

0.10173

4

0.002

Duceram

0.836

0.10922

4

0.00

DuceraGold

0.896

0.13297

4

0.00

Composite

0.972

0.10964

4

0.00

Zirconia

0.022

0.03633

4

0.247

Duceram

0.228

0.05541

4

0.001

DuceraGold

0.268

0.0753

4

0.001

Composite

0.18

0.06819

4

0.004

Zirconia

0.302

0.0676

4

0.001

3.1.4 SEM evaluation: On completion of interferometry evaluation, representative specimens from each group were prepared for SEM (JEOL, JSM-5600 LV, Japan) to observe the effect of polishing systems on the topography of the tested materials, the working surface of each representative specimens were visualized at × 500 magnification. Figs 3.5-3.13 show micrographs of the surface of the materials tested after finishing and polishing with the CeraMaster and EnamelPlus polishing kits. The results of the interferometery measurements were consistent with the SEM micrographs. Examination of the micrographs indicated that the final polished surface for all materials tested appeared to have a smoother surface. The polished sides of Duceram and DuceraGold ceramics looked similar but contained many pitted areas and voids (Figs 3.5, 3.6, 3.11 and 3.12). Polished composite was different and contained numerous scratches and irregularities following the application of CeraMaster polishing kit (Fig

66 3.7); however it had a more regular finish with shallow scratches following the application of EnamelPlus kit (Fig 3.10). Mylar strip produced the smoothest surface on the composite (Fig 3.13). Zirconia ceramic polished surfaces had scratches but were free of voids following the application of the two polishing systems (Figs 3.8, 3.9). Rubber polisher did not seem to have any effect on the surface topography of this ceramic (Fig 3.9). Glazed surfaces of Duceram and DuceraGold looked similar and had a few scratches and voids (Fig 3.13). Rough surfaces of zirconia ceramic specimens (Figs 3.8, 3.9) and composite specimens (Figs 3.7, 3.10) had scratches in a clear linear direction, while rough surfaces of metal bonded ceramics specimens (Figs 3.5-3.6, 3.11-3.12) looked similar with irregular roughness features.

A

B

C

Fig 3-5 SEM micrographs of Duceram specimens after (A) grinding with diamond bur, (B) polishing .with blue diamond polisher, (C) final polishing with white diamond polisher

67

B

A

C

Fig 3-6 SEM micrographs of DuceraGold specimens after (A) grinding with diamond bur, (B) polishing with blue diamond polisher, (C) final polishing with white diamond polisher.

A

B

C

Fig3-7 SEM micrographs of Composite specimens after (A) grinding with diamond bure, (B) polishing with blue diamond polisher, (C) final polishing with white diamond polisher.

68

B

A

C

Fig 3-8 SEM micrographs of zirconia ceramic specimens after (A) grinding with diamond bur, (B) polishing with blue diamond polisher, (C) final polishing with white diamond polisher.

A

B

C

D

Fig 3-9 SEM micrographs of zirconia ceramic specimens after (A) grinding with medium-grit diamond bur, (B) grinding with ultra fine diamond bur, (C) polishing with rubber, (D) polishing with diamond paste.

69

A

B

C

D

Fig 3-10 SEM micrographs of the composite specimens after (A) grinding with mediumgrit diamond bur, (B) grinding with ultrafine diamond bur, (C) polishing with rubber, (D) polishing with diamond paste.

A

B

C

D

Fig 3-11 SEM micrographs of Duceram specimens after (A) grinding with medium-grit diamond bur (B) grinding with ultrafine diamond bur, (C) polishing with rubber, (D) polishing with diamond paste.

70

A

B

C

D

Fig 3-12 SEM micrographs of DuceraGold specimens after (A) grinding with medium-grit diamond bur, (B) grinding with ultrafine diamond bur, (C) polishing with rubber, (D) polishing with diamond paste.

B

A

C

.Fig 3-13 (A) glazed Duceram, (B) glazed DuceraGold, (C) composite cured under Mylar strip

71

3.2 Results of marginal fit assessments of zirconia based all ceramic crowns 3.2.1 SEM evaluation: SEM imaging (JEOL, JSM-5600 LV, Japan) was used as a technique to visualize the marginal gap for the milled cores and the completed crowns before and after glazing. (Figs 3.14-3.16)

Ceramic core Glazed crown Stone die Stone die

Stone die Fig 3.14 Representative SEM micrographs show the marginal gap of the zirconia ceramic core.

All-ceramic crown

Stone die

Fig 3.15 Representative SEM micrographs show marginal gap of the all ceramic crown after dentine and enamel firing cycle.

Fig 3.16 Representative SEM micrographs show marginal gap of the all ceramic crown after glazing Fig 3.16 Representative SEM micrograph show marginal gap of the all ceramic crown after glaze firing cycle.

72 The mean marginal gap of 15 measurements per crown were determined to give an overall mean for each specimen in two preparation groups (shoulder and chamfer) at the following fabrication steps 1- fabrication of the core, 2- dentine and enamel porcelain applications, 3- glazing. The overall mean for each test group and for all the fabricated crowns and the standard deviation were calculated and listed in tables 3.17 and 3.18, and shown graphically in figs 3.17 and 3.18. Raw data is presented in appendix 2. For both test groups the measurements obtained after the porcelain firing cycle were greater than those for the fabrication of the copings; whilst a minimal difference was recorded between the porcelain and the glaze firing cycle. Table 3-17 Means marginal gaps (µm) ± the standard deviation of each marginal design recorded after different fabrication steps (n=5 per marginal design)

Group

Fabrication steps (µm) Step 1

Step 2

Step 3

Shoulder

20.86 ±4.81

52.61 ±8.32

50.06 ±6.28

Chamfer

21.55 ±2.48

45.20 ± 14.89

47.78 ±6.45

Step 1= Zirconium core fitting. Step 2= Dentine and enamel building up. Step 3= Glazing.

Table 3-18 The overall mean marginal gaps (µm) ± the standard deviation of all the specimens recorded after three fabrication steps (n=10 per fabrication step) Fabrication steps (µm) Step1

Step 2

Step 3

21.20 ± 3.62

48.91 ± 5.23

48.92 ± 4.34

The marginal gap varied at different locations of each crown. The range of the marginal gap of all the fabricated crowns regardless of the construction stage was between 10 to 95 µm, with the standard deviation ranging from 6 to 23 µm.

73

Fig 3-17 The mean marginal gap (µm) with error bars denoting the standard deviation of all ceramic crowns with two different marginal lines at each construction stage (n=5). Step 1= Zirconium core fitting. Step 2= Dentine and enamel build up. Step 3= Glazing.

Fig 3-18 The mean marginal gap (µm) with error bars denoting the standard deviation of all ceramic crowns with two different marginal lines at each construction stage (n=10).

74 3.2.2 Statistical analysis Data from marginal gap measurements were analyzed using SPSS (SPSS 15.0 for windows). A 2- way analysis of variance (ANOVA) was performed to determine any significant differences between different finish line designs and different production steps and their interactive effect on the marginal gap. A p-value of <0.05 was considered statistically significant. Statistical analysis revealed that the p value for the construction step factor was less than 0.05 (fabrication step p= 0.00) (Table 3.19), therefore the results were significantly different. A T test was subsequently performed to compare the three fabrication steps, showing a significant difference between the core marginal fit and the all ceramic crown before glaze firing cycle (Table 3.20). No significant difference was found between the mean marginal fit before and after the glazing (p=0.96) (Table 3.20). The interactive effect of marginal finish and fabrication step was not significant (p=0.327), indicating that there was no significant difference in the marginal gap between shoulder and chamfer finish line and fabrication step (Table 3.19). Table 3-19 Summary of the results of two-way ANOVA analysis.

Fabrication step Line design group Step * group

Type III Sum of Squares

df

Mean square

F

Sig

518.223 82.105 105.562

2 1 1

25559.11 82.105 52.781

33.113 0.683 1.08

0.000 0.515 0.327

Table 3-20 Results of the T test. Paired Differences Mean Std. Deviation

Pair 1 Pair 2

step1 step2 step2 step3

Std. Error Mean

95% Confidence Interval of the Difference Upper Lower

t

df

Sig. (2tailed)

14.17652 27.85380

4.48301

-37.99

-17.71253

-6.213

9

0.000

0.13900

3.20566

-7.11271

7.39071

0.043

9

0.966

10.13719

75

Chapter four 4.1 Discussion of surface roughness measurements The first part of the present study was designed to determine the effect of two polishing systems on the topography of two different types of restorative materials (composite and ceramic). The study has investigated the use of a polishing system that has been formulated for use with a particular tooth coloured restorative material to determine whether clinically acceptable surface roughness values can be achieved when used with a different type of tooth coloured restorative material with the intention of simplifying product range for dental technicians to use one system for a range of restorative materials. Although numerous studies have evaluated the effect of polishing systems on different types of restorative materials 38, 51-53, very little was found in the literature that addresses this topic 56, 57. A further aim of this study was to evaluate the effect of grinding and polishing using different techniques on the surface roughness of zirconia ceramic core material. Many studies have investigated the mechanical properties and the biocompatibility of this restorative material 73, 74, and also the effect of finishing procedures on the strength and structural stability of zirconia ceramic 23. However in a survey of the literature, only one article was found that evaluated the surface roughness of different dental ceramic core materials after grinding and polishing 60, whilst another study mentioned surface roughness of zirconium core ceramic, but with the aim of determining the effect of the firing procedure on the mechanical properties and surface roughness of the core ceramic 75

.

Zirconium dioxide ceramic (Yttrium-stabilized tetragonal zirconia polycrystals) is a high strength oxide ceramic, developed by adding small amount of Y2O3 to ZrO2, in order to stabilize the ceramic in a tetragonal phase, which is normally unstable at room

76 temperature 76. Zirconia may exist in several crystal types, depending on the addition of minor components such as CaO, MgO, or Y2O3. Specific phases are said to be stabilized by minor components, if about 8 to 12 percent of a component is added, a fully stabilized cubic phase can be produced. If a smaller amount (3-5 weight percentages) is added, then partially stabilized zirconia is produced 77. This ceramic showed flexural strength far exceeding that of other ceramic systems, (up to 1200 MPa); it is densely sintered with minimal voids, flows and cracks 78. Although the cores must be veneered with suitable ceramics, adjustment to improve occlusion by grinding of allceramic restorations means that part of the core can be exposed to the oral environment 78

. In addition when the marginal line of a prepared tooth is very thin (narrow) and in

order to avoid overextending the crown, the gingival part of the crown may be left uncovered with the veneering ceramic. Moreover due to the superior mechanical properties of this ceramic there is the potential to construct a complete crown. For all these reasons it was essential to investigate the effect of polishing this restorative material with the opportunity to use the same polishing systems to finish and polish zirconia ceramic. Finishing refers to gross contouring or reducing of the restoration to obtain the desired anatomy. Diamond and carbide burs are usually used to finish a tooth coloured restoration 59. Polishing reduces the roughness and scratches created by finishing instruments 79. Proper finishing and polishing are important to enhance both aesthetics and longevity of the restoration. The results of this study show that the polished surface of Duceram was smoother than the glazed surface when the CeraMaster technique was applied, and this finding supports previous studies 55, 56. For DuceraGold ceramic no difference was found between the glazed and the polished surfaces 38.

77 CeraMaster polishing system has been designed to be used with dental ceramics, and it has produced clinically acceptable topography values in the ceramics tested in this study. The relation between the Ra of intraoral hard surfaces and bacterial adhesion is one way in which clinical acceptability can be determined. Ra values below a threshold level of 0.2 µm have been shown to have no effect on supra- and sub-gingival microbial adhesion or colonization 80. Therefore, the roughness of all intraoral hard surfaces should approximate Ra value of 0.2 µm or lower. The initial Ra and Sa values of the DuceraGold ceramic were higher than the Duceram ceramic, the reason for this is not clear but it might be related to the leucite content of the DuceraGold. Metal-bonded ceramics are a leucite-containing glass ceramics; the amount of precipitated leucite can be carefully controlled to correct the thermal coefficient for a particular alloy, and can be 30-40 % of the volume of the material 6. A possible explanation for this finding might be the particle size of the ceramic. Otherwise the overall Ra and Sa values were similar for the two metal bonded ceramics and this might be related to the effectiveness of the CeraMaster polishing technique. It could therefore be suggested that although the surface roughness of restorative materials differ due to their type, this difference can be overcome with the appropriate polishing technique. In this study as well as others 53, 59, Mylar strip formed the smoothest surface for the composite, and that is because of the resin rich layer at the surface 57. The composite used in this study was a microhybrid composite with a small particle size (0.012 µm), and the polishing system included two diamond polishers with different diamond grain grit sizes with a high cutting efficiency, and a good ability to abrade the filler phase. However the overall Ra and Sa values for the composite were higher than all the other materials tested. This result might be related to the heterogeneous nature of

78 the composite (organic matrix and inorganic fillers), so the diamond burs were not able to produce a homogenous abrasion of the fillers and the resin matrix. In addition the polishers were not able to remove the scratches left by the diamond burs. The present findings seem to be consistent with other studies which found that using diamond finishing points produced rougher surfaces on all composites, and scratches could be seen on the surface of the composites 49, 53. The observed high surface roughness value could be attributed to the unsuitability of the finishing/polishing system used with this type of composite, and not to the other factors that related to the composite material itself like the large particle size of the composite, which was the conclusion of a previous study 81. These findings are also in agreement with Paravina’s findings 82, which showed that flexibility and suitability of the finishing/polishing material has the most important role in determining the surface roughness of the composite. Lu and Roeder 83 who used different types of microhybrid composites found significant differences between materials and polishing systems, and concluded that polishing systems had the most important role in determining the surface roughness of resin composites. Another study evaluated the effect of a range of polishing systems on different types of composites and concluded that the surface roughness of the composite is determined by the size, hardness, and amount of filler particles, which enhance the mechanical properties of the composites, and by the hardness of the abrasive and its grit size 59. The EnamelPlus polishing kit has been designed to be used with composites, contrary to expectations, the two veneering ceramics exhibited similar values following the two different surface treatments, and also no significant difference between the surface roughness values of the polished surfaces and the control group was found. A previous study by Grieve and Jeffrey 84 concluded similar results with no difference between

79 polished and glazed surfaces of the tested ceramic. SEM micrographs of these ceramics showed the polished sides of Duceram and DuceraGold ceramics contained many pitted areas and voids. This result might be explained by the fact that although vacuum firing results in a reduction in number and size of voids within the porcelain, it does not completely eliminate these bubbles. When the surface layer of the porcelain is removed by grinding, the voids following incomplete condensation cannot be significantly reduced in size 85. In addition grinding the surface layer will expose these pores resulting in a rough surface and the polishing procedure was not able to reduce these voids completely. For zirconia ceramic CAD-CAM restorations are not produced with a surface glaze generated in a porcelain furnace and therefore comparison with a control group was not possible. What was surprising that although EnamelPlus has been designed to be used with composites, it produced low Ra and Sa values with zirconia ceramic, and these values were lower and significantly different from the other two ceramics. A possible explanation for this result is that zirconia ceramic is stronger and has higher fracture toughness and therefore more force would need to be applied to remove material from the surface in order to produce or reduce roughness. Medium grit diamond was not able to produce deep scratches, and so for both polishing kits it was easy to remove any slight irregularity caused by the diamond bur. Another possible explanation for the low initial surface roughness value of zirconia ceramic is the high quality of sintering, in which the fusion of individual particles to form a coherent solid 6, compared to the layering ceramics. This allowed coherent cutting by the medium grit diamond. SEM micrographs showed the surface of zirconia ceramic had a more regular appearance following the application of medium grit diamond compared to the other two ceramics

80 tested. Another unexpected finding was noticed when applying the silicone rubber polisher, no statically significant different was found in the surface roughness values before and after the use of silicone rubber polisher with zirconia ceramics, whereas the difference in the surface roughness values for the other ceramics was significant. Again it could be due to the strength of the zirconia ceramic, as silicone polisher was applied with insufficient force to reduce the scratches produced by the diamond bur or to cause any protruding crystals to either pull out or fracture from the surface or to be smoothed. Further force is not recommended as it causes heat to generate, which can cause microfissures on the ceramic surface 86. Fissures can decrease the ceramic structural strength and contribute to its failure 61. It is encouraging to compare this finding with that of by Al-shammery and Bubb 45, who found that rubber wheel polishing of Vita Mk II samples (a homogenous fine grained feldspathic porcelain), resulted in their smoothest surfaces whilst having no significant effect on an experimental stronger ceramic with higher fracture toughness (a glass ceramic consisting of fine crystals of mixed barium/potassium fluormica in an aluminosilicate glassy matrix). Sasahara 35 concluded in his study that porcelains with lower leucite content tended to present lower roughness compared to those with higher leucite content after being polished with rubbers or discs followed by diamond pastes. The rubber that was used in this study was described by the manufacturer as the abrasive soft silicone rubber; this means that this rubber includes grains in a particular size. The composition, quality and particle size of the polishing media was not able to be known from these manufacturers The use of the goat hair brush with polishing paste exhibited smoother surfaces for the three ceramics used and the composite; this might be explained by the good retention of the paste on the surface of the brush, and the good abrasive potential of the paste 61. This

81 can only be confirmed by the structural analysis of the composition of the paste and the materials tested. Camacho and Vinha 61 found that the use of a brush was efficient for mechanical polishing of tested ceramics used with polishing paste. Al-Wahadni 87 recommended that regardless of the type of ceramic or pre-treatment any adjusted ceramic restoration should be subjected to a finishing sequence that is followed through to a final stage of polishing with diamond paste. It is interesting to find that although the initial surface roughness values were significantly different between zirconia ceramic and the composite, the overall Ra and Sa values were not found to be statistically significant. This might be related to the ability of the polishing procedure to abrade the filler phase of the composite and minimize the scratches left by the finishing bur. In previous studies microhybrid composite has shown high polishability 52, 31. Compared to the other composite types microfilled and microhybrid composites can be finished to a very smooth surface due to their smaller particle size and arrangement 31. Although both polishing kits exhibited comparable Ra and Sa values with the metal bonded ceramics, it should be noticed that the EnamelPlus kit included three polishing steps, whilst only two were required for the CeraMaster kit. If other essential factors are considered, rather than the final surface roughness values that can be produced by a polishing system, such as clinical and laboratory time, labour, cost, and durability (longevity) of the polishing components, then the polishing technique must be chosen in accordance with its application. The finding of this study has proved the high polishability of zirconia ceramic as it exhibited low surface roughness values following the application of two different existing commercial polishing kits. This is an important issue for future research.

82 There were some limitations in this study. The sample size was small, 5 specimens for each test group. However eight Ra and four Sa surface roughness measurements were performed on each specimen, which means 60 surface roughness measurements for each test group. Although a variety of restorative dental materials were used, another limitation was the diversity of surface treatments applied. In addition, the pressure applied to the handpiece was not able to be precisely controlled; however Siegel and Von Fraqunhofer 88 evaluated the cutting efficiency of diamond dental burs using different handpiece loads against simulated enamel and a machinable glass ceramic (Macor) and demonstrated that when the pressure applied to the handpiece was markedly increased, there was no increase in the cutting efficiency for medium-grit burs. Further investigation is necessary to fully evaluate the effect of different polishing systems on the topography of different types of restorative materials.

83

4.2 Discussion of the marginal fit assessment results of zirconia based all ceramic crowns. Marginal fit is a very important aspect for fixed restorations as a large marginal opening allows more plaque to accumulation, gingival secular fluid flow, and bone loss, resulting in microleakage, recurrent caries, and periodontal disease 89. All ceramic core crowns offer excellent aesthetics and biocompatibility 90, and can be made from different high strength materials using different technologies. The slip casting technique is used to fabricate a slip cast crown coping, a refractory die is needed to compensate for the shrinkage of the slip during the sintering process. The slip is loaded onto plaster models using a ceramic brush to form the crown coping. The material is then sintered to an open porous state and infiltrated with glass 64. The heat pressing technique is based on the lost wax principle. Prefabricated ceramic ingots of lithium-disilicate ceramic are heated and then pressed into the lost wax form of the crown coping 5. Dental computer aided design/computer aided manufacturing systems utilize a scanning design and machining process to custom-shape copings from industrially prefabricated ceramic blocks 18. However there is a lack of information as to how the fit is affected by the firing procedures that are used for all ceramic crown systems in general and zirconium based crowns in particular. The aim of this part of the study was to evaluate the influence of veneering procedure on the marginal fit of milled zirconia ceramic core, and whether two different types of finish line preparation design would affect the marginal adaptation. Although there have been a few studies that have evaluated the marginal fit of all ceramic restorations, comparison of the results can be confusing. Variations in the type of tooth used, differences in restoration designs, preparation procedures, testing

84 methods, the way of measuring the marginal gap, and whether the fit was determined before or after luting, are parameters that will influence the result obtained. Light microscopy and SEM imaging have been compared as marginal gap measurement techniques for milled and CAD/CAM restorations 66, 91. One investigation 91

concluded that SEM imaging was superior to light microscopy to evaluate marginal

gaps of class two CAD/CAM inlays, especially when a small gap was present. Another investigation 66 reported that there was no significant difference between the accuracy of the two techniques. However those authors 66 also concluded that SEM imaging provides more appropriate and realistic observations than a light microscopy analyzing system. Using SEM analysis, restorations seated on die samples were thought an accepted method for analyzing accuracy of fit in vitro 63, 92. The marginal fit of the nonsectioned specimens is generally examined with direct microscopic view of the interface. Measurements using an optical microscope may be faulty due to the limited depth of the field 93. In the current study as in a previous study 91 optical microscope could not be used because unless the two points to be measured were on the same plane, it was not possible to focus on both points at once. The number of measurements per crown used in previous studies has varied considerably. Groten and Axmann 94 suggested that ideally at least 20-25 measurements per crown be recorded according to the aimed precision level. Fewer points (4-12) may not be considered adequate for comparison of different crown systems or the successive manufacturing process 94. In many studies, sample size ranged from 5 to 10 specimens for each test group 65, 95, 96, 97. Therefore in the present study 10 specimens for the CAD/CAM system was used, 5 specimen for each marginal design, and 15 measurements for each crown at each testing step.

85 The perpendicular measurement from the internal surface of the casting to the axial wall of the preparation is called the internal gap, and the same measurement at the margin is called marginal gap 98. Holmes 98 who established measurements of internal and external marginal fit concluded that the best alternative was absolute marginal discrepancy, this was defined as the angular combination of marginal gap and any extension error (overextension or underextension), and described more simply as the distance from the margin of a casting to the cavosurface of a tooth preparation 36 (Fig 4-1). Overcontour and undercontour were not evaluated in this study.

Fig 4-1 Points of measurements of the marginal gap of different marginal situations 36 . Although there are no accepted standards 93, Lofstrom and Barakat 99 who used SEM to measure the supragingival margins of crowns that were considered clinically well fitting by several dentists and reported marginal discrepancy values in a range of 7 to 65 µm. Acceptable marginal opening is described as that which not visible to the naked eye, and is undetectable with a sharp explorer 93. While Mclean and Fraunhofer 100 reported that marginal opening of 120 µm should be the maximum limit of clinical acceptability. In this study the 50 µm reference was used as the criterion, as it is a size clinicians are able to control and realize in the sense of good clinic practice 66. The fit of the cores and the all ceramic crowns were comparable to other conventional all ceramic crown systems, and came within the low documented values. In vitro studies on conventional In-Ceram crown revealed a mean marginal gap width of 57 µm 93. According to Klaus 17 in a previous study about marginal fit of aesthetic crowns means of marginal opening

86 were 158 µm (Cerestone) and 177 µm (collarless porcelain-fused to metal crowns). In vitro studies of the fit of Procera all-Ceram crowns revealed mean marginal opening below 63 µm 101. The Procera System embraces the concept of computer-assisted design and computer-assisted machining to fabricate an all-ceramic crown composed of a densely sintered, high-purity aluminum oxide coping combined with compatible veneering porcelain 20. Following a survey of the literature, few studies have investigated the clinical fit of zirconia coping. In this study the mean marginal gap between the coping and the stone dies was 21 µm, and between the glazed all ceramic crowns and the stone die was 48.92 µm. These results are supported by a previous study which addressed the fit of a zirconia coping and found that the mean marginal discrepancy between the copings and master models was below 50 μm, with a range of 0 to 115 μm 102. Another study by Luthardt and Sandkuhl 103 who used zirconia-TZP-frameworks to construct the crowns concluded that the precision of fit of the veneered crowns were comparable with established systems. Another study For other all-ceramic crown systems, a previous study by Akbar and Petrie 66 suggested that the finish line preparation design had no effect on the marginal adaptation for a CEREC 3 composite crown. Quintas and Oliveira 104 evaluated the effect of different finish lines, ceramic manufacturing techniques, and luting agents on the vertical discrepancy of ceramic copings, and documented that the ceramic manufacturing technique was the only significant factor tested that influenced vertical marginal discrepancy when this factor was considered separately. The same conclusion was reported by Suliman and Chai 64 when they compared the marginal fit of three allceramic crown systems (In-Ceram, IPS, and Procera crowns) at different fabrication

87 stages. In a survey of the literature no study was found about the effect of marginal line design on zirconia ceramic crowns and so comparison with other studies was not possible. Shoulder design is recommended for the preparation of all ceramic crowns rather than a chamfer preparation 105. Shillingburg 106 found that the shoulder finish line resisted distortion. However some of CAD/CAM manufacturers who utilize scanning machines recommend the chamfer finish line to gain improved scanning at the edge shaped areas. In the current study marginal design had no effect on the marginal fit of the zirconia core, and on the all-ceramic crowns. Contrary to the current study, results from Puntsag and Bae 107 who evaluated the marginal fit of CAD-CAM zirconia cores with different proximal height and after the veneering procedure suggested that no significant differences were observed in the firing stages of all the test groups and the mean marginal gap of zirconia cores with different proximal heights fell well within the acceptable clinical ranges. While in their study Nonogaki and Uno 108 who aimed to examine the effect of a post-milling heat treatment on the marginal fit of the all-ceramic crowns fabricated by using CAD/CAM system found that the fitting accuracy differed significantly after the post-milling heat treatment. In reviewing the literature a study was found that used Renishaw Incise process, the aim of the study was to measure and compare the marginal and internal fit of zirconia ceramic copings produced by the Procera, Lava and Renishaw Incise systems, the means and standard deviations of the marginal fit of the cores produced by Renishaw Incise process in the mentioned study was 47.3 ±12.4 µm 109. Conflicting results of studies which addressed the same issues using the same procedure were identified. Shearer 110 found that the addition of porcelain to the conventional InCeram coping did not affect the fit, while Balkaya 93 demonstrated that porcelain firing

88 significantly altered the marginal fit of the crowns, except for the fit in the horizontal plane of the conventional In-Ceram coping. In the present study the results showed high variability as reflected by standard deviations in the range of (6-23 µm). The mean of the marginal discrepancies at all measurement locations reflects the level of marginal discrepancy of the entire crown. However, the marginal discrepancy of each crown may vary greatly at different locations 94. Because of high variation of the fit within a crown; the mean values of all measurement locations can show a large local discrepancy and result in an increase in the standard deviation 93. This can be due to the fact that contemporary chairside or laboratory based CAD/CAM systems have additional factors that may affect the accuracy of the fit, including software limitations in designing restorations, and hardware limitations of the camera, scanning equipment, and milling machines 111. Each single step could contribute to the overall fit of the crown. The standard deviation in previous studies has been reported to be approximately 20 µm 96, 106, 112. In another study the standard deviation ranged from 4 µm to 32 µm 93. Whereas Holmes 97 found that the fit at various locations around the margin was not significantly different. The heat treatment influenced the marginal fit during the initial firing only, this result is supported by other studies who investigated the influence of multiple firing on ceramic cores 75, 113 this finding indicates that multiple firings that might be performed at the dental technology laboratory do not further affect the integrity of marginal fit. Zirconia is a polymorphic material that occurs in three forms. At its melting point of 2680 °C, the cubic structure exists and transforms into the tetragonal phase below 2370 °C. The tetragonal-to-monoclinic phase transformation occurs below 1170 °C and is accompanied by a 3-5 % volume expansion which causes high internal stresses 114. Yttrium-oxide (Y2O3 3 % mol) is added to pure zirconia to control the volume

89 expansion and to stabilize it in the tetragonal phase at room temperature. Known as Partially Stabilized Zirconia (PSZ) whose microstructure at room temperature generally consists of cubic zirconia as the major phase with monoclinic and tetragonal zirconia precipitates as the minor phase 115. This partially stabilized zirconia has high initial flexural strength and fracture toughness 78. Tensile stresses at a crack tip will cause the tetragonal phase to transform into the monoclinic phase with an associated 3-5 % localized expansion 116. The volume increase creates compressive stresses at the crack tip that counteract the external tensile stresses. This phenomenon is known as transformation toughening and retards crack propagation. In the presence of higher stress, a crack can still propagate. The toughening mechanism does not prevent the progression of a crack; it just makes it harder for the crack to propagate 116. Yttrium-oxide partially stabilized zirconia (Y-TZP) has mechanical properties that are attractive for restorative dentistry; such as chemical stability, high mechanical strength, and fracture-toughness 117. The cores have a radiopacity comparable to metal which enhances radiographic evaluation of marginal integrity, excess cement removal, and recurrent decay 118. Y-TZP can be manufactured in two methods through computeraided design/computer-aided manufacturing (CAD/CAM) technology. First, an enlarged coping/framework can be designed and milled from a homogenous ceramic soft green body blank of zirconia 119. The framework structure has a linear shrinkage of 20-25 % during sintering until it reaches the desired final dimensions. Processing with this softer pre-sintered material not only shortens the milling time, but also reduces the wear on the milling tools. Although zirconia frameworks can be milled directly from a fully sintered pre-fabricated blank into the final dimensions 120, milling fully sintered zirconia may compromise the microstructure and strength of the material 121 and it takes a longer time

90 77

. However milling of fully sintered blanks has no shrinkage so the marginal fit is

superior 122. The veneering process involves a firing procedure at high temperature (750-900 ºC) and subsequent cooling of the restoration. This process is performed at least once, but usually two to five times 75. During the porcelain firing cycle, veneering porcelain particles melt and fuse by filling up voids, and the contracting mass of fused porcelain exerts a compressive force on the coping during cooling 123. If the coping margin begins to deform under the stress of the contracting porcelain, the stress is spread further around the circumference of the margin. Consequently, because the porcelain shrinks toward its greatest mass, the labiopalatal distance decreases, while the mesiodistal distance increases in the coping margin 124. The distortion that occurred during the porcelain firing cycle might be due to non-uniform porcelain mass 93. As an extra bulk of the porcelain used during the manufacturing process may be the primary reason for variations in the marginal distortion after the first heat treatment 124, 75. The thermal behaviour of ceramic core materials and veneering porcelains of allceramic systems is more complex than that of porcelain fused to metal (PFM) system. Studies have demonstrated that porcelains designed to be bonded to a metal, that are composed of leucite crystals embedded in a glass matrix, showed a change in thermal dimensional behaviour with each heat treatment, whilst that of metal remained constant 125. The change in thermal dimensional behaviour observed in porcelain is explained by a change in leucite content as well as a decoupling of leucite from the glass matrix during the cooling process and their re-coupling to the glass matrix with repeated firings 126. Therefore, dental ceramics are not constant because of phase changes as a result of thermal history (i.e. number of firings). While this change in thermal dimensional behaviour does not lead to compatibility problems for PFM

91 systems, it may affect the thermal compatibility of the ceramic materials (i.e. ceramic core, veneering porcelain) used in all-ceramic systems 113. For PFM restoration systems, it is preferable to have a positive expansion/contraction mismatch

between

the

metal

and

veneering

porcelain.

The

thermal

expansion/contraction of the metal should be slightly higher than that of porcelain. It is thought that the greater thermal contraction of the metal during cooling subjects the porcelain layer to a compressive stress. By contrast, a negative mismatch as a consequence of a greater expansion/contraction of the porcelain would produce tensile stresses in the porcelain layer 9. Compressive stress increases the strength of the whole restoration because ceramics are able to withstand higher compressive stress than tensile stress 9 due to the fact that compressive stress tends to inhibit the propagation of cracks from the surface by keeping them closed 8. Nielsen and Tuccillo 127 concluded that metal and porcelain are thermally

compatible when their difference in thermal

expansion/contraction is less than 0.5 ppm/8 ºC in order to generate safe levels of residual stress. It is important to realize that metals show a ductile behaviour under stress, while ceramics are brittle, which means that metals have the ability to deform to an applied stress without fracturing

8

. For an all-ceramic system, the thermal

compatibility between ceramic substrate and veneering porcelain is based on the same principle of thermal expansion/contraction mismatch. Now, in place of the metal at room temperature, the high-strength ceramic core is subjected to a tensile stress while the weak veneering porcelain is still subjected to compressive stress. In contrast to metal, this condition might have a negative effect on the ceramic core, as the tensile strength of the brittle ceramic is much lower than its compressive strength. More likely, in an ideal all-ceramic system both veneering porcelain and ceramic core should not have significant mismatch in their thermal contraction values

113

. This might explain

92 why core material manufacturers recommend the use of their manufactured layering ceramics. Other manufacturers like Renishaw guarantee excellent core fitting and give wider choices of the veneering ceramics to be used. Zirconia has proved its chemical and dimensional stability, high mechanical strength and fracture toughness; otherwise it has been proved that heat treatment influences the core materials in several ways such as microhardness and flexure strength 75, 113, 128. Therefore it would not be unexpected that the dimensional stability is affected, even though the temperature was below the core materials own temperature. The temperature of the veneering firing are within the stability range temperature of Y-TZP, but the stability of the forced tetragonal state may, nevertheless, be altered 75. The yttrium-stabilized zirconia is in a forced tetragonal shape that is stable between 1150° C to below room temperature. The phase stability of the individual materials may depend on grain size and the initial sintering process. One more explanation is that the structural change throughout the specimens might have given some dimensional distortion; this suggestion is supported by other studies indicating that heating in veneering temperature range reduces the number of monocline shaped grains on the surface 113,128-129. Stabilizing content and/or zirconia grain size has an important effect on the transformation process and proceeds more rapidly between 56 °C and 500 °C 130. Further work on the structure of the core and the veneering ceramics would be needed in order to address this suggestion. In the present study the investigator performed the technical procedures which included using a rotating diamond, and the application of veneering ceramic with extreme care to maintain marginal integrity, As such the effect of technical work on marginal fit is minimized. The method used in the current study was related as strongly as possible to the real laboratory procedure situation. A stone die was used to simulate the dental

93 laboratory procedure, worn and chipped finish lines of the die were avoided when selecting points of measurements. Coli and Karlsson 131 who evaluated coping-stone die and coping-master model internal and marginal misfits documented that the misfit detected between the coping and the stone die did not differ from the misfit detected between the coping and the acrylic resin master model. Finally, based on the observations it can be concluded that the mean marginal gap of zirconia all ceramic crowns fabricated with the use of CAD/CAM Renishaw Incise system were excellent and within the clinical range of acceptance. The heat treatment had a significant influence on the fit of zirconia core after the first firing, but the mean marginal gap was still within the clinical range of acceptance and the marginal fit accuracy exceeded documented values for ceramic core fitting in other studies. It can be suggested that the ceramic coping fabrication and layering procedure both accounted for the marginal fit of the all-ceramic restoration. The outcome of this study introduces zirconia ceramic and the production system (Renishaw Incise process) as one of the best choices for all ceramic restorations. However more studies would be needed to support these results as very few studies have addressed the effect of layering procedure on the marginal fit of zirconia ceramic cores and as such it is still not well evaluated.

94

4.3 Conclusion CeraMaster polishing kit exhibited clinically acceptable surface roughness values with the tested ceramics but not with the composite. While EnamelPlus kit resulted in acceptable values for all the tested materials. It can be concluded that although some polishing systems can be used with different restorative materials, others would only produce clinically acceptable surface roughness values with exacting restorative materials. Two issues have been investigated with zirconia ceramics and ended with satisfactory results. Zirconia ceramic was easily polished within a reasonable time; this can be added to the advantages of zirconia ceramic. This study has also introduced this restorative material as a good choice for milled frameworks, although the veneering procedure had an effect on the marginal fit, it was still within clinical acceptance. More study is needed to support these findings and this will lead investigators to other issues such as aesthetic and wear reduction of opposing dentition of zirconia ceramic.