...

Document 1009586

by user

on
Category: Documents
214

views

Report

Comments

Transcript

Document 1009586
ADVERTIMENT. La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents
condicions d'ús: La difusió d’aquesta tesi per mitjà del servei TDX (www.tesisenxarxa.net) ha
estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats
emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats
de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei TDX. No s’autoritza la
presentació del seu contingut en una finestra o marc aliè a TDX (framing). Aquesta reserva de
drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita
de parts de la tesi és obligat indicar el nom de la persona autora.
ADVERTENCIA. La consulta de esta tesis queda condicionada a la aceptación de las siguientes
condiciones de uso: La difusión de esta tesis por medio del servicio TDR (www.tesisenred.net) ha
sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos
privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción
con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio TDR.
No se autoriza la presentación de su contenido en una ventana o marco ajeno a TDR (framing).
Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus
contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la
persona autora.
WARNING. On having consulted this thesis you’re accepting the following use conditions:
Spreading this thesis by the TDX (www.tesisenxarxa.net) service has been authorized by the
titular of the intellectual property rights only for private uses placed in investigation and teaching
activities. Reproduction with lucrative aims is not authorized neither its spreading and availability
from a site foreign to the TDX service. Introducing its content in a window or frame foreign to the
TDX service is not authorized (framing). This rights affect to the presentation summary of the
thesis as well as to its contents. In the using or citation of parts of the thesis it’s obliged to indicate
the name of the author
Influence off sand
dblasting on
n zirconia in
restorrative dentis
stry
Ravi Kira
an Chinttapalli
T
Thesis
prresented to obtain
n PhD deg
gree in Materials
M
Science and
a
Engineering from th
he Univers
sitat Politècnica de
e Cataluny
ya
T
Thesis
d
director
Prof. Ma
arc Angla
ada
Departme
ent of Mate
erials Scie
ence and Metallurgy
M
y Engineering
Barcelona
a, Spain
J
July
2012
1
2
To
My parents Rajeswari and Nageswara Rao
And
My wife Jyothi
3
4
ACKNOWLEDGEMENTS
I thank the CIEFMA group (ETSEIB, UPC) for giving me an opportunity to carry out my
doctoral thesis work. This thesis work was made possible by many people, whose direct or
indirect help is greatly appreciated. I gratefully acknowledge the financial support given for the
doctoral program.
I owe my sincere gratitude to my supervisor and also the director of CIEFMA Prof. Marc
Anglada for his outstanding guidance and help. I am indebted to him particularly for his ever
availability to discuss the work. I also thank him for valuable suggestions he made during
discussions. His motivation and encouragement to get best out of the work is greatly appreciated.
I would also like to extend my sincere thanks to Prof. Emilio Jimenez, Prof. Luis llanes, Prof.
Antonio Mateo and Prof. Gemma Fargas for their wonderful suggestions and co-operation.
I would like to thank the MAEC-AECID and expresses my honest appreciation to Ministerio de
Asuntos Exteriores y de Co-operación of the Government of Spain for providing the funding for
carrying out this thesis.
I want to convey my sincere thanks Dr. Fernando Garcia Marro and Dr. Alvaro Mestra for their
help in experimental work and fruitful discussions. I would also like to thank Mr. Francesc
Heredero for his all-round help in the lab throughout this thesis work. Overall, during the last
four years I have immensely enjoyed working in CIEFMA group and I want to thank all the
people in the group for their wonderful cooperation. My special thanks goes to Dr. Trifon
Trifonov of Nanocenter for his help during FIB sessions and peppy discussions.
I am grateful to Prof. Mike Reece for allowing me to carry out some experimental work in his
lab at Nanoforce, QMUL, London. I thank Mr. Ben Milsom for the help at Nanoforce lab.
Finally, I thank my parents, my wife, and other family members for their unconditional love,
support, encouragement and best wishes.
5
6
Contents LISTOFFIGURES................................................................................................................................................................11
LISTOFTABLES..................................................................................................................................................................19
ABSTRACT............................................................................................................................................................................21
RESUMEN..............................................................................................................................................................................24
CHAPTER 1
1.ZIRCONIA CERAMICS AND SANDBLASTING............................................................................................27
1.1 BACKGROUNDOFBIOCERAMICS.......................................................................................................................28
1.2 ZIRCONIACERAMICS..............................................................................................................................................30
1.2.1ZIRCONIAPOLYMORPHS...................................................................................................................................................................30
1.2.2STABILIZATIONOFZIRCONIA...........................................................................................................................................................31
1.2.3TETRAGONAL(T)TOMONOCLINIC(M)TRANSFORMATION......................................................................................................32
1.2.3.1Martensitictransformations...............................................................................................................................................32
1.2.3.2Stressinducedtransformations.........................................................................................................................................36
1.2.3.3Martensiteformationandpropagation.........................................................................................................................38
1.2.4TOUGHENINGMECHANISMSINZIRCONIA......................................................................................................................................41
1.2.4.1Transformationtoughening................................................................................................................................................41
1.2.4.2Ferroelastictoughening........................................................................................................................................................43
1.2.5R‐CURVEBEHAVIOUR.....................................................................................................................................................................44
1.2.6LOWTEMPERATUREDEGRADATION(AGING)OFZIRCONIA.......................................................................................................45
1.2.6.1Agingmechanism.....................................................................................................................................................................46
1.2.6.2Nucleation(N)andGrowth(G)..........................................................................................................................................48
1.2.6.3Agingkinetics.............................................................................................................................................................................49
1.2.6.4Factorsinfluencingaging.....................................................................................................................................................50
1.2.6.5Consequencesofaging............................................................................................................................................................51
1.3ZIRCONIACERAMICSINRESTORATIVEDENTISTRY.......................................................................................51
1.3.1NANOCRYSTALLINE/NANOCOMPOSITEMATERIALS..................................................................................................................53
1.3.2TYPESOFDENTALCERAMICSANDCLINICALAPPLICATIONS......................................................................................................53
1.3.3PROCESSINGOFZIRCONIAFRAMEWORKS.....................................................................................................................................54
1.3.4IMPLICATIONS....................................................................................................................................................................................55
1.4SANDBLASTING...........................................................................................................................................................56
1.4.1PRINCIPLEOFSANDBLASTING.........................................................................................................................................................57
1.4.2EROSIONINCERAMICS......................................................................................................................................................................58
1.4.3EFFECTOFPROCESSPARAMETERS.................................................................................................................................................58
1.4.3.1ImpactAngle...............................................................................................................................................................................59
1.4.3.2Particlevelocity.........................................................................................................................................................................59
1.4.3.3Particlesize.................................................................................................................................................................................60
1.5SANDBLASTINGINZIRCONIA..................................................................................................................................60
1.5.1INTRODUCTION..................................................................................................................................................................................60
1.5.2STATEOFTHEART............................................................................................................................................................................61
1.5.3SURFACEROUGHNESS.......................................................................................................................................................................61
1.5.4PHASETRANSFORMATION...............................................................................................................................................................63
1.5.5STRENGTH..........................................................................................................................................................................................67
1.5.6FATIGUEBEHAVIOUR........................................................................................................................................................................70
1.5.7SANDBLASTINGDAMAGE..................................................................................................................................................................72
1.5.8HYDROTHERMALDEGRADATIONAFTERSANDBLASTING............................................................................................................73
1.5.9BONDSTRENGTH...............................................................................................................................................................................75
1.6SUMMARY......................................................................................................................................................................76
7
1.7OBJECTIVES..................................................................................................................................................................78
1.7.1SPECIFICOBJECTIVES........................................................................................................................................................................78
1.8REFERENCES.................................................................................................................................................................79
CHAPTER 2
2.EXPERIMENTAL METHODS..................................................................................................................................89
2.1SPECIMENPREPARATION........................................................................................................................................90
2.1.1BASEMATERIAL.................................................................................................................................................................................90
2.1.2MATERIALAS‐300..........................................................................................................................................................................90
2.1.3NANOCRYSTALLINEZIRCONIA.........................................................................................................................................................91
2.1.4NANOCOMPOSITES............................................................................................................................................................................93
2.1.5METALLOGRAPHICPREPARATION..................................................................................................................................................94
2.2MICROSTRUCTURALCHARACTERIZATIONS......................................................................................................94
2.2.1CONFOCALMICROSCOPE..................................................................................................................................................................94
2.2.2SCANNINGELECTRONMICROSCOPE(SEM)..................................................................................................................................95
2.2.3FIB‐SEM............................................................................................................................................................................................95
2.2.4AFM....................................................................................................................................................................................................96
2.3MECHANICALCHARACTERIZATIONS...................................................................................................................96
2.3.1DENSITY..............................................................................................................................................................................................96
2.3.2SURFACEPROFILOMETER.................................................................................................................................................................96
2.3.3HARDNESS&FRACTURETOUGHNESS............................................................................................................................................96
2.3.4BIAXIALSTRENGTH...........................................................................................................................................................................98
2.3.5NANOINDENTATION..........................................................................................................................................................................99
2.3.6RESIDUALSTRESSES.......................................................................................................................................................................101
2.4PHASECHARACTERIZATIONS...............................................................................................................................102
2.4.1X‐RAYDIFFRACTION......................................................................................................................................................................102
2.4.2RAMANSPECTROSCOPY.................................................................................................................................................................102
2.5SANDBLASTING.........................................................................................................................................................103
2.6LOWTEMPERATUREDEGRADATION.................................................................................................................104
2.7GRAINSIZE..................................................................................................................................................................104
2.8HEATTREATMENTS.................................................................................................................................................104
2.9REFERENCES...............................................................................................................................................................105
CHAPTER 3
3.MATERIAL PROPERTIES AND HYDROTHERMAL DEGRADATION...............................................107
3.1INTRODUCTION.........................................................................................................................................................108
3.2MONOLITHICMATERIALS......................................................................................................................................108
3.2.1MICROSTRUCTURE.........................................................................................................................................................................108
3.2.2PROPERTIESOFMONOLITHICMATERIALS..................................................................................................................................109
3.2.3HYDROTHERMALDEGRADATION(HDORLTD)&EFFECTONPROPERTIES........................................................................110
3.2.4DISCUSSION.....................................................................................................................................................................................112
3.3SPHERICALINDENTATIONBEHAVIOROFPOROUSZIRCONIA...................................................................116
3.3.1INTRODUCTION...............................................................................................................................................................................116
3.3.2THEORETICALMODELFORSPHERICALINDENTATION..............................................................................................................116
3.3.3TIPCALIBRATION...........................................................................................................................................................................118
3.3.4DETERMINATIONOFCONTACTPOINT.........................................................................................................................................118
3.3.5LOAD–DISPLACEMENTCURVES....................................................................................................................................................119
8
3.3.6INDENTATIONSTRESS–STRAINCURVES......................................................................................................................................120
3.3.7DAMAGECHARACTERIZATION......................................................................................................................................................122
3.3.8DISCUSSION.....................................................................................................................................................................................124
3.4NANOCOMPOSITES...................................................................................................................................................128
3.4.1INTRODUCTION...............................................................................................................................................................................128
3.4.2MICROSTRUCTURE.........................................................................................................................................................................130
3.4.3HYDROTHERMALDEGRADATION.................................................................................................................................................132
3.4.4PROPERTIESOFTHECOMPOSITES...............................................................................................................................................133
3.4.5PHASETRANSFORMATIONINTHECRACKTIP............................................................................................................................135
3.4.6DISCUSSION.....................................................................................................................................................................................135
3.5SUMMARY....................................................................................................................................................................137
3.6REFERENCES...............................................................................................................................................................138
CHAPTER 4
4.ROUGHNESS, PHASE TRANSFORMATIONS AND DAMAGE INDUCED BY SANDBLASTING
...............................................................................................................................................................................................143
4.1INTRODUCTION.........................................................................................................................................................144
4.2SURFACEROUGHNESS.............................................................................................................................................144
4.2.1ZIRCONIAAS‐300.........................................................................................................................................................................144
4.2.1.1Surfacemorphology.............................................................................................................................................................144
4.2.1.2Surfaceroughness.................................................................................................................................................................146
4.2.2NANOCRYSTALLINEZIRCONIAS120...........................................................................................................................................147
4.2.2.1Morphology&Surfaceroughness..................................................................................................................................147
4.2.3NANOCOMPOSITE3YTZP‐2CNT...............................................................................................................................................148
4.2.3.1Morphology&Surfaceroughness..................................................................................................................................148
4.3STATISTICALANALYSIS..........................................................................................................................................148
4.4PHASETRANSFORMATION....................................................................................................................................149
4.4.1ZIRCONIAAS300...........................................................................................................................................................................149
4.4.1.1Depthoftransformation.....................................................................................................................................................151
4.4.2NANOCRYSTALLINEZIRCONIAS120...........................................................................................................................................154
4.4.3NANOCOMPOSITE3Y‐TZP‐2CNT..............................................................................................................................................155
4.5EFFECTOFHEATTREATMENTAFTERSANDBLASTING...............................................................................156
4.6DAMAGEINDUCEDBYSANDBLASTING.............................................................................................................157
4.6.1ZIRCONIAAS300...........................................................................................................................................................................157
4.6.2ZIRCONIAS120..............................................................................................................................................................................162
4.6.3NANOCOMPOSITE3YTZP‐2CNT...............................................................................................................................................162
4.7DISCUSSION................................................................................................................................................................163
4.7.1SURFACEROUGHNESS....................................................................................................................................................................163
4.7.2PHASETRANSFORMATION............................................................................................................................................................164
4.7.3DAMAGE...........................................................................................................................................................................................167
4.8SUMMARY....................................................................................................................................................................167
4.9REFERENCES...............................................................................................................................................................168
CHAPTER 5
5.MECHANICAL PROPERTIES AFTER SANDBLASTING...........................................................................171
5.1INTRODUCTION.........................................................................................................................................................172
5.2BI‐AXIALSTRENGTH................................................................................................................................................172
9
5.1.2EFFECTOFANNEALING..................................................................................................................................................................175
5.1.3FRACTOGRAPHY..............................................................................................................................................................................175
5.3ELASTICMODULUSANDCONTACTHARDNESS...............................................................................................177
5.4RESIDUALSTRESSES................................................................................................................................................179
5.5DISCUSSION................................................................................................................................................................181
5.5.1BI‐AXIALSTRENGTH.......................................................................................................................................................................181
5.5.1.1Effectofparticlesizeandpressure................................................................................................................................182
5.5.1.2Effectofimpactangle..........................................................................................................................................................184
5.5.1.3Effectofphasetransformation........................................................................................................................................185
5.5.1.4Influenceoftheresidualstress........................................................................................................................................186
5.5.1.5Effectofannealing................................................................................................................................................................195
5.5.2ELASTICMODULUSANDHARDNESS.............................................................................................................................................197
5.5.3RESIDUALSTRESSES.......................................................................................................................................................................197
5.6SUMMARY....................................................................................................................................................................198
5.7REFERENCES...............................................................................................................................................................199
CHAPTER 6
6.HYDROTHERMAL DEGRADATION AFTER SANDBLASTING............................................................201
6.1INTRODUCTION.........................................................................................................................................................202
6.2HYDROTHERMALDEGRADATIONOFAFTERSANDBLASTING..................................................................202
6.2.1PHASETRANSFORMATION............................................................................................................................................................202
6.3CHANGEINMECHANICALPROPERTIES.............................................................................................................205
6.3.1BI‐AXIALSTRENGTH.......................................................................................................................................................................205
6.3.2ELASTICMODULUSANDHARDNESS............................................................................................................................................206
6.4DISCUSSION................................................................................................................................................................208
6.4.1ZIRCONIAAS300...........................................................................................................................................................................208
6.4.2NANOCRYSTALLINEZIRCONIA.....................................................................................................................................................211
6.5SUMMARY....................................................................................................................................................................212
6.6REFERENCES...............................................................................................................................................................213
CHAPTER 7
7.CONCLUSIONS AND FUTURE WORK............................................................................................................215
7.1CONCLUSIONS............................................................................................................................................................216
7.2FUTUREWORK..........................................................................................................................................................216
APPENDIX...........................................................................................................................................................................219
REFERENCES..............................................................................................................................................................................................221
PUBLICATIONS AND CONFERENCE CONTRIBUTIONS............................................................................223
10
LIST OF FIGURES
Figure 1.1
Schematic representations of the three crystal structures of ZrO2 (a) cubic,
(b) tetragonal, and (c) monoclinic
30
Figure 1.2
Phase diagram of ZrO2-rich sections of the phase diagrams and of ZrO2–
Y2O3 system
31
Figure 1.3
Invariant plane strain (IPS), IPS (shaded area) is composed of shear
parallel to habit plane and expansion/contraction normal to habit plane
32
Figure 1.4
Stepwise illustration of phenomenological theory of martensitic
transformations
33
Figure 1.5
Schematic illustration of general correspondences for A, B and C for t-m
transformation with two expected orientations for each correspondence
34
Figure 1.6
Schematic illustration of a) internally twinned and b) internally slipped
martensitic plates
35
Figure 1.7
Examples of transformation twinning in zirconia ceramics. A) TEM
micrograph of partially transformed t-grain in Ce-TZP, b) TEM image of
hydrothermally degraded 3Y-TZP, c) Transformed pentagonal grain with
grain boundary cracks and d) SEM image of a section perpendicular to the
degraded surface of D60H. Grains with monoclinic transformation twins
as well as microcracks can be observed
35
Figure 1.8
Relationship between free energy changes for t-m transformation in
zirconia
37
Figure 1.9
Schematic illustration of stages in transformation of spherical tetragonal
zirconia particle to self accommodating monoclinic variants
39
Figure 1.10
Scheme of autocatalytic transformation a) single martensite plate b) two
additional martensite plates in the neighbor grains triggered by the initial
plate c) a fourth plate in the next grain
40
Figure 1.11
Illustration of transformation toughening mechanism in front of a
propagating crack
41
Figure 1.12
a) Cardioid-shaped transformation zone associated with a purely dilatant
transformation at a crack tip. Two dotted lines define a sector at the front
of the zone with an included angle of 120° (2/3), which actually leads to
a decrease in toughness. (b) Transformation wake of half height hy
associated with the movement of the crack tip by a distance a
43
Figure 1.13
R-curve behavior of different ceramics
44
Figure 1.14
OHˉ ion trigger mechanism for degradation
47
Figure 1.15
Nucleation and growth aging mechanism
48
11
Figure 1.16
Monoclinic content versus time at different temperatures
50
Figure 1.17
a) dental crowns b) three crown bridge c) all ceramic implants d) fitted
abutment e) cemented zirconia restorations, f) radiographic evaluation of
the two zirconia restorations
52
Figure 1.18
LTD of LAVA dental ceramic: a) XRD patterns obtained with a fixed
incidence angle of 2 before and after 168 h low temperature degradation
at 140 0C, b)Monoclinic fraction Xm after hydrothermal aging
55
Figure 1.19
Schematic representation of sandblasting
56
Figure 1.20
Particle impact as indentation process
57
Figure 1.21
Single particle impact process
58
Figure 1.22
Effect of impact angle on erosion rate in ductile and brittle materials
59
Figure 1.23
Effect of sandblasting on the surface roughness of different dental
materials for different sandblasting conditions. 25-4B indicates as the
particle size 25 m and air pressure 4 bars
62
Figure 1.24
XRD patterns of DC-Zirkon dental ceramic ground and sandblasted. ttetragonal phase, m-monoclinic phase
64
Figure 1.25
Changes in monoclinic ZrO2 with heat treatment temperature
66
Figure 1.26
X-ray diffraction patterns of DC-Zirkon after heat treatment
66
Figure 1.27
Changes in strength of different dental materials before and after
sandblasting. (Results from various authors)
67
Figure 1.28
SEM micrograph of a polished interface perpendicular to the sandblasted
surface of a fine-grained Y-TZP
69
Figure 1.29
S–N fatigue data for Lava (LV) (3M Espe), Lava colored (LVB) (3M
Espe), Everest ZS (KV) (KaVo), Zeno (ZW) (Wieland), Fatigue limits at
106 cycles at 10Hz in water
71
Figure 1.30
Maximum tensile stress in ceramic layer versus effective time to radial
fracture tR for a) as-polished b) sandblasted surfaces of Y-TZP and
alumina. Solid lines are data fits in accordance with slow crack growth
relations
71
Figure 1.31
SEM Micrograph showing partial top and cross-section view of sandblast
damage (50 m Al2O3 particles) in Y-TZP
73
Figure 2.1
Cold isostatic pressing of 3Y-TZP powder
90
Figure 2.2
Sintering curve for material AS300
91
Figure 2.3
Spark plasma sintering (SPS FCT HP D25I, FCT system GmBh)
Nanoforce Ltd UK, a) SPS furnace, b) die punch setup c) die during
sintering temperature
92
12
Figure 2.4
SPS curve for nanocrystalline zirconia
92
Figure 2.5
SPS curve for Zirconia-MWCNT nanocomposites
93
Figure 2.6
Principle of focus ion beam technique
95
Figure 2.7
Scheme of SURFTEST profilometer
96
Figure 2.8
Vickers indentation scheme with Palmqvist crack system
97
Figure 2.9
a) ball on three ball test configuration b) Finite element model of the three
balls test assembly
98
Figure 2.10
Schematic illustration of a) the unloading process showing parameters
characterizing the contact geometry b) indentation load–displacement data
showing important measured parameters
99
Figure 2.11
a) BEGO, Easy Blast sandblaster, b) & c) sandblasting schemes in 90º and
30º
103
Figure 3.1
SEM microstructures of thermally etched samples a) S65, b) S95, c) S120,
d) AS300, e) S800.
108
Figure 3.2
X-ray diffraction patterns of samples: a) after sintering and fine polishing;
b) after 60 hours autoclave ageing. (t-tetragonal, m-monoclinic)
110
Figure 3.3
Elastic modulus and hardness as a function of penetration depth before
and after LTD: (a) & (c) for fine grain materials, (b) & (d) for coarse grain
materials respectively.
111
Figure 3.4
Schematic representation of spherical indentation.
117
Figure 3.5
i) P–h curve for fused silica, ii) hc vs a, for spherical indentation on fused
silica.
118
Figure 3.6
Stiffness against contact radius and raw data before correction (inset).
119
Figure 3.7
P–h curves with Hertz fit of i) S65, (ii) S90, (iii) S120, and (iv)
comparison curve.
120
Figure 3.8
Indentation stress–strain curves of i) S65, ii) S90, and iii) S120
121
Figure 3.9
AFM tapping mode height images showing residual spherical indentation
imprint on (a) S65 at 1 N, (b) S90 at 2.4 N, (c) S120 at 2.9 N load and (d)
residual depth profiles.
122
Figure 3.10
Micro Raman spectra of the residual indentation imprints of (i) S65, (ii)
S90 and (iii) S120
123
Figure 3.11
SEM micrographs of FIB cross-sections of spherical indentation (i) S65,
(ii) S90 and (iii) S120. Notations a and b are the locations of indentation
and surface, respectively.
123
Figure 3.12
High magnification views of marked zones in Figure 3.11(i)
124
13
Figure 3.13
Effect of pore shape and porosity content on elastic modulus
125
Figure 3.14
SEM image of dried composite powder with 2 vol% MWCNTs
130
Figure 3.15
SEM images of thermally etched polished surfaces i) 0% ii) 0.5 % iii) 1%
iv) 2%
131
Figure 3.16
SEM images of fracture surfaces i) 0% ii) 0.5 % iii) 1% iv) 2% MWCNTs
132
Figure 3.17
X-ray diffraction patterns of a) polished samples b) hydrothermally aged
for 200 hours.
132
Figure 3.18
Fracture toughness with respect to volume % of MWCNTs.
134
Figure 3.19
Phase transformation around the crack tip
135
Figure 3.20
SEM image of indentation crack in 3YTZP-2CNT, where some nanotubes
are bridging the crack faces
136
Figure 4.1
a1) & b1) Surface morphologies, a2) & b2) surface profiles obtained from
the center of the image, for the surfaces sandblasted with 110 m particle
size at an angle 90º
145
Figure 4.2
a1) & b1) Surface morphologies, a2) & b2) surface profiles obtained from
the center of the image for the surfaces sandblasted with 250 m particle
size at an angle 90º
145
Figure 4.3
a1) & b1) Surface morphologies, a2) & b2) surface profiles obtained from
the center of the image for the surfaces sandblasted with 110&250 m
particle size at an angle 30º
146
Figure 4.4
Surface roughness Ra for different sandblasting conditions a) at 2 bars, b)
at 4 bars
147
Figure 4.5
a1) Surface morphology of S120, a2) surface profile obtained from the
center of the image for the surface sandblasted with 110 m particle size
at an angle 90º
147
Figure 4.6
a1) Surface morphologies of 3YTZP-2CNT, a2) surface profile obtained
from the center of the image for the surface sandblasted with 110 m
particle size at an angle 90º
148
Figure 4.7
Tukey´s mean statistical difference in roughness of AS300
149
Figure 4.8
X-ray diffraction patterns of samples sandblasted with different conditions
150
Figure 4.9
Monoclinic volume fraction of samples sandblasted with different
conditions measured by XRD
151
Figure 4.10
Illustration of Raman map acquisition on the cross-sections of sandblasted
samples. Maps were acquired at a lateral resolution of 2 microns.
151
Figure 4.11
micro Raman spectra obtained from the cross-sections at different depths
a) 110-2B-90º b) 250-2B-90º
152
14
Figure 4.12
Cross-sectional phase transformation maps of AS300 sandblasted with
110-2B-90º,a) & b) maps taken from different zones in the center of the
specimen
153
Figure 4.13
Cross-sectional phase transformation maps of AS300 sandblasted with
250-2B-90º,a) & b)maps taken from different zones in the center of the
specimen
153
Figure 4.14
X-ray diffraction patterns of samples polished and sandblasted for S120
154
Figure 4.15
Cross-sectional phase transformation maps of S120 sandblasted with 1102B-90º,a) & b)maps taken from different zones in the center of the
specimen
155
Figure 4.16
X-ray diffraction patterns of samples polished and sandblasted for
3YTZP-2CNT
155
Figure 4.17
Cross-sectional phase transformation maps of 3YTZP-2CNT sandblasted
with 110-2B-90º, a) & b) maps taken from different zones in the center of
the specimen.
156
Figure 4.18
X-ray diffraction patterns after different treatments for materials i) AS300, ii) S120 and iii) 3YTZP-2CNT
156
Figure 4.19
Change in monoclinic volume fraction with temperature of AS300 after
sandblasting
157
Figure 4.20
SEM images of polished cross-section of sandblasted AS300 with 1102B-90º a) and b) images taken in different zones
158
Figure 4.21
SEM image of first FIB trench of material AS300 sandblasted 110-2B-90º
159
Figure 4.22
SEM image of first FIB trench after ion beam etching of material AS300
sandblasted 110-2B-90º
159
Figure 4.23
SEM image of a second FIB trench after ion beam etching of material
AS300 sandblasted 110-2B-90º
160
Figure 4.24
Close-up view of the top region of figure 4.20 and twin orientation
160
Figure 4.25
SEM image of FIB trench of material AS300 sandblasted 110-2B-30º
161
Figure 4.26
SEM image of FIB trench of material AS300 sandblasted 250-2B-90º
161
Figure 4.27
SEM image of FIB trenches of material S120 sandblasted 110-2B-90ºa) &
b) FIB trenches in different sites
162
Figure 4.28
SEM image of FIB trenches of material 3Y-TZP-2CNT sandblasted 1102B-90º
163
Figure 4.29
Effect of monoclinic volume fraction on the ratio of I002/I200
165
Figure 5.1
Mean bi-axial strength of AS300 after sandblasting at a) 2 and b) 4 bars
pressure
173
Figure 5.2
Weibull plot of AS300 after sandblasting at a) 2 and b) 4 bars pressure.
174
15
Figure 5.3
Tukey´s mean statistical difference in strength among groups
174
Figure 5.4
Effect of annealing on after sandblasting on bi-axial strength a) mean
biaxial strength and b) Weibull plot
175
Figure 5.5
SEM micrographs of defects found in different groups
176
Figure 5.6
Sample P-h curves of 110-2B-900 on the surface at different depths
177
Figure 5.7
Elastic modulus and hardness on the cross-section of AS300 sandblasted
with110-2B-900
178
Figure 5.8
Indentation cracks of 110-2B-900 during polishing steps
179
Figure 5.9
a) Measured crack lengths at different depths and b) estimated residual
stresses (c > > d)
179
Figure 5.10
Theoretical compressive residual stresses vs. monoclinic volume fraction
180
Figure 5.11
Theoretical compressive residual stresses for different treatments, a) at
pressure 2 bars, b) at pressure 4 bars,
181
Figure 5.12
Fracture surfaces of 110-2B-900 a) mimic of sharp indentation and b)
mimic of blunt indentation
183
Figure 5.13
Particle-sample interactions with different orientations
184
Figure 5.14
Relationship between strength and amount of monoclinic phase
185
Figure 5.15
a) Schematic representation of the residual stresses along the crack depth
b) Surface elliptical crack showing the definition of the different
geometrical terms
186
Figure 5.16
Change in fracture toughness with increasing crack size caused by
residual stress gradient
188
Figure 5.17
Effect of stress intensity factor on the indentation crack
190
Figure 5.18
Change in crack length with respect to the ratio of thickness removed to
the depth
192
Figure 5.19
Effect of residual stress and defect size on strength
193
Figure 5.20
Residual stresses with respect depth and linearity
195
Figure 5.21
Influence of stress state on a crack under different treatments (crack
perpendicular to the tensile stress direction)
196
Figure 6.1
X-Ray diffraction patterns after various treatments i) AS300, ii) S120 and
iii) 3YTZP-2CNT
203
Figure 6.2
Monoclinic volume fractions with respect to different treatments
204
Figure 6.3
Effect of different treatments on the bi-axial strength of AS300 a) bi-axial
strength, b) Weibull plot
205
16
Figure 6.4
Fracture surfaces of AS300 treated with sandblasting (110-2B-90º) and
hydrothermal degradation (100 hours). a) Micrograph showing degraded
layer with high intergranular fracture appearance with a crack parallel to
the surface, b) crack at high magnification
206
Figure 6.5
Changes in elastic modulus for different materials
207
Figure 6.6
Changes in contact hardness for different materials
207
Figure 6.7
Cross-sectional property changes in AS 300
208
Figure 6.8
SEM micrographs of fracture surfaces of AS300 after hydrothermal
degradation of 100 hours
211
Figure A1
Relationship between residual stress intensity factor and the ratio of crack
to depth
221
17
18
LIST OF TABLES
Table 1.1
Cell attachment mechanisms with bio-ceramics
28
Table 1.2
Characteristics of some bio-ceramics
29
Table 1.3
Commercial zirconia frame works for clinical applications
53
Table 1.4
Dental ceramic materials and forming methods
54
Table 2.1
Chemical composition of 3Y-TZP powder
90
Table 2.2
Sintering temperatures and material labels
93
Table 2.3
Concentration of MWCNT and material labels
94
Table 2.4
Calculated values for the constants
99
Table 2.5
Sandblasting nomenclature
104
Table 3.1
Properties of monolithic materials
109
Table 3.2
Properties obtained by spherical indentation
121
Table 3.3
Properties of nanocomposites materials
133
Table 4.1
Peak intensity ratios of (002)t and (200)t for different surface
treatments
150
Table 5.1
Estimated surface defect size
177
Table 5.2
Properties obtained from surface of AS 300 sandblasted with 110-2B90º
178
19
20
Abstract
ABSTRACT
The use of tetragonal zirconia polycrystals (3Y-TZP) in dental restorations such as crowns and
implants has recently increased attention due to their very good aesthetic appearance and
mechanical properties in addition to biocompatibility. The restorations undergo several surface
treatments such as sandblasting for better adhesion to luting cements and veneering porcelain.
However, there is some controversy about using sandblasted crowns, as sandblasting introduces
surface flaws and defects that can compromise the strength of the crown as well as
crystallographic changes at the surface.
Though the effect of sandblasting in zirconia has been previously studied to some extent, many
issues like severity of the conditions, effect on surface mechanical properties, subsurface damage
and phase transformation zone size have not been still fully addressed. Comprehensive
understanding of these aspects will help in choosing better sandblasting conditions and also to
improve the microstructural design of the materials for long term performance of the restorations
so that clinical failures can be avoided or delayed.
In this thesis, the effect of sandblasting on 3 mol% yttria stabilized zirconia (3Y-TZP) with
different grain sizes has been studied under different sandblasting conditions. Additionally,
nanocomposites formed by adding multiwall carbon nanotubes (0.5-2 vol. %) to 3Y-TZP matrix
have been also studied.
Initially, the study has been focused in the mechanical properties and hydrothermal degradation
resistance of nanometric grain size 3Y-TZP and zirconia multiwall carbon nanotubes
nanocomposites (3YTZP-MWCNT). Nanometric grain size 3Y-TZP (90-150 nm) produced by
spark plasma sintering have slightly lower toughness compared to standard zirconia with grain
size 300 nm. Adding multiwall carbon nanotubes improve the indentation fracture toughness
nearly 15% compared to monolithic materials. Elastic modulus hardly changes while hardness
decreases slightly for 2 vol.% nanotubes.
3Y-TZP presents severe hydrothermal degradation in tests carried out at temperature of 131 °C
in water vapour at a pressure of 2 bars. By contrast, resistance to hydrothermal degradation of
nanometric grain size 3Y-TZP is very high. In the 3Y-TZP-MWCNT composites the resistance
Ravi K Chintapalli
21
Abstract
to hydrothermal degradation is found similar to nano grain size 3Y-TZP since the composites
have similar grain sizes.
The materials were subjected to sandblasting using two particle sizes (110 and 250 m), two
pressures (2 and 4 bars) and two impact angles (300 and 900). After sandblasting the materials
were analyzed looking for roughness, phase transformation and damage. In addition the change
in mechanical properties and in hydrothermal degradation resistance induced by sandblasting
was evaluated.
It has been found that increasing particle size and pressure increases surface roughness, while
highest surface roughness can be achieved using low impact angle (30º) and large particle size
(250 mm) without affecting the strength.
The bi-axial strength of zirconia has been studied only in standard 300 nm 3Y-TZP. The main
result has been to show that at impact angle of 90º the biaxial strength increases when
sandblasted with 110 m particles while it decreases with 250 m particles. On the other hand,
the strength slightly increases when sandblasted under an impact angle of 30° irrespective of the
particle size. By using nanoindentation it is shown that mild sandblasting conditions (110 m
particle size, 2 bars pressure) have no effect on the surface mechanical properties such as, elastic
modulus and contact hardness. A model based on the formation of residual compressive stresses
is presented in order to explain the indentation the shorter length of the indentation cracks in
sandblasted material as well as to rationalize the increase in strength of sandblasted material
under mild sandblasted conditions. The decrease in strength in severe sandblasting conditions is
associated to the nucleation of cracks during sandblasting which are longer than the critical crack
length in the non-sandblasted material.
The microstructural change induced by sandblasting near the surface consists of: i) a thin layer
of plastically deformed grains; ii) phase transformation; and iii) occasional microcracking. The
fraction of monoclinic volume fraction induced after sandblasting under the studied conditions
is of about 10-15%, and with transformation up to a depth of about 10-13 m. In the nanometric
and nanocomposites materials there is also transformation by sandblasting but the volume
fraction of monoclinic phase induced is lower than in the standard 300 nm 3Y-TZP.
Ravi K Chintapalli
22
Abstract
Finally it is shown that in sandblasted conventional 3Y-TZP the kinetics of hydrothermal
degradation is slower than in the starting material and that hydrothermal degradation of the
sandblasted material diminishes the strength to the values close to the initial control condition.
Ravi K Chintapalli
23
Abstract
RESUMEN
La utilización de circona tetragonal policristalina estabilizada con 3 % mol de itria (3Y-TZP)
para fabricar coronas e implantes ha sufrido una fuerte expansión recientemente debido a las
buenas propiedades mecánicas, estéticas y de biocompatibilidad que posee este material. La
microestructura y composición exacta de 3Y-TZP
son específicamente diseñadas por los
fabricantes para adecuarse a las normativas existentes. Durante su procesamiento, las prótesis
cerámicas son tratadas superficialmente por métodos como el arenado para mejorar su adhesión
al cemento protésico y a la porcelana que cubre la pieza dental. Ahora bien, no todos los
fabricantes dentales recomiendan el arenado de las coronas previamente a su implantación, ya
que el arenado puede introducir defectos superficiales que pueden afectar la integridad
estructural de la prótesis así como producir cambios cristalográficos en la superficie.
A pesar de que el efecto de arenado en circona ya ha sido parcialmente estudiado, no se han
considerado exhaustivamente muchos aspectos como la severidad de las condiciones de arenado,
el efecto en las propiedades superficiales, el daño subsuperficial y los cambios de fase. Una
comprensión detallada de estos aspectos es necesaria para escoger correctamente las condiciones
de trabajo del arenado y para mejorar el diseño microestructural de estos materiales. Esto
mejoraría la vida a largo plazo de los implantes cerámicos evitando o retrasando posibles fallos
de la pieza.
En este trabajo, se ha estudiado el efecto del arenado en 3Y-TZP con diferentes tamaños de
grano bajo diferentes condiciones de arenado. También se ha estudiado el efecto de la adición de
de nanotubos de carbono multicapa (MWCNT, 0.5-2 vol. %) a una matriz de 3Y-TZP.
Una parte del trabajo ha consistido en el estudio de las propiedades mecánicas y de resistencia a
la degradación
hidrotérmica de 3Y-TZP nanométrica y de los nanocompuestos 3YTZP-
MWCNT con tamaño de grano nanométrico (90-150 nm) producidos por “spark plasma
sintering”, los cuales se ha encontrado que poseen una menor tenacidad de fractura por
indentación que 3Y-TZP con tamaño de grano de 300 nm. La adición de un 2% en volumen de
MWCNT aumenta la tenacidad de fractura por indentación en alrededor de un 15% con respecto
a la matriz del mismo tamaño de grano. El módulo de elasticidad apenas cambia mientras que la
dureza disminuye ligeramente.
Ravi K Chintapalli
24
Abstract
La circona convencional con tamaño de grano de 300 nm presenta severa degradación
hidrotérmica en ensayos realizados a 131 °C en vapor de agua y a una presión de 2 bares. Por el
contrario, la resistencia a la degradación hidrotérmica de 3Y-TZP de tamaño nanométrico es
muy alta. En los nanocompuestos 3Y-TZP-MWCNT la resistencia a la degradación hidrotérmica
es similar a la de 3Y-TZP con el mismo tamaño de grano.
Los materiales fueron sometidos a arenado usando dos tamaños de partícula (110 and 250 m),
dos presiones (2 y 4 bares) y dos ángulos de impacto (300 and 900). Después del arenado la
superficie de los materiales fue analizada con respecto a su rugosidad, transformación de fase, y
daño. También se midió el cambio en propiedades mecánicas y en la resistencia a la degradación
inducido por el arenado.
Se ha determinado que aumentando el tamaño de partícula y la presión de arenado aumenta la
rugosidad superficial, mientras que la mayor rugosidad superficial se obtiene utilizando ángulos
de impacto bajos (30º) y tamaños de partícula grandes (250 mm) sin que ello afecte a la
resistencia a flexión biaxial.
La resistencia a la flexión biaxial se ha estudiado solamente en 3Y-TZP convencional. El
principal resultado encontrado ha sido poner de manifiesto que
impactos a 90º aumentan la
resistencia a la flexión biaxial de las probetas arenadas con tamaños de partícula de 110 m
mientras que la resistencia disminuye si se utilizan tamaños de 250 m partículas. Por otro lado,
la resistencia aumenta ligeramente para los dos ángulos de impacto independientemente del
tamaño de la partícula. Mediante nanoindentación se ha puesto de relieve que condiciones
suaves de arenado (110 m, 2 bares) no afectan a las propiedades mecánicas superficiales, tales
como módulo de elasticidad y dureza. Se presenta un modelo basado en la influencia de las
tensiones residuales de compresión inducidas durante el arenado para explicar la reducción de
las grietas de indentación en el material arenado y para explicar los cambios en la resistencia
biaxial. En tanto el arenado no produce grietas superiores al tamaño crítico, se produce un
aumento en la resistencia a flexión. (arenado suave). En cuanto el arenado induce defectos
superiores al defecto crítico inicial, se puede producir una disminución de la resistencia
dependiendo de la severidad del arenado.
Los cambios microestructurales inducidos por el arenado cercanos a la superficie consisten en: i)
una película delgada de granos deformados; ii) una zona con transformación de fase; and iii)
Ravi K Chintapalli
25
Abstract
microgrietas ocasionales. La fracción de volumen de fase monoclínica después del arenado bajo
las condiciones estudiadas está en el rango entre 10 y 15%, y la profundidad de la zona
transformada está comprendida entre 10 y 13 m. En los materiales nanométricos y en los
nanocomposites también se observa transformación de fase pero a una escala menor y se induce
una cantidad menor de fase monoclínica.
Finalmente se ha puesto también de relieve que en 3Y-TZP convencional arenada la cinética de
la degradación hidrotérmica es más lenta que en la 3Y-TZP inicial y que la resistencia a la
flexión biaxial después la degradación hidrotérmica de las probetas arenadas alcanza valores
similares a los correspondientes al material de partida.
Ravi K Chintapalli
26
Introduction to zirconia ceramics and sandblasting
Chapter
1
Zirconia ceramics and
sandblasting
Ravi K Chintapalli
27
Introduction to zirconia ceramics and sandblasting
1.1 Background of bio ceramics
Ceramic materials have been of great interest for mankind since prehistoric times. Now many
products in our daily life are made of ceramics, for example dishes, knives, resistors, bearings,
turbine rotors, and many others. For long period of time, ceramics are employed in chemical,
steel and glass industries for special conditions such as, high temperature, corrosive, and
reducing condition environments. But recently ceramics are used as surgical implants for
humans due to their good mechanical properties and the ability to form a stable interface with the
connecting tissue. Although the implantation of materials as restoratives in humans dates back to
pre-Christian era; progress has been made since mid 19th century to employ foreign materials as
restoratives.
Modern material science has developed wide range of alloys and materials to use as orthopedic
devices such as plates and screws for fractured bones and artificial hip joints. But, the use of
metallic materials as implants in the human body is limited by their corrosive nature. Unlike
metals, some oxide ceramics perform well under oxidative environments. Furthermore, their
chemical inertness and biocompatibility made ceramics very attractive candidates for biomedical
applications such as total hip replacements. In addition, in dentistry, ceramics are used as
restorative materials, such as gold porcelain crowns, glass-filled ionomer cements, dentures etc.
The requirement to use bio inert ceramics as replacements for damaged parts of the musculoskeletal system led to the use of advanced ceramics such as high-density alumina (Al2O3) and
zirconia (ZrO2). But the longevity and the acceptation of bio-ceramics by the living host tissue
depends on the interface. Generally no material placed in living tissues is totally inert with the
physiological environment except the autogenous tissue, as all other materials instigate a
response from living tissue. For example, a fibrous tissue of variable thickness forms at the
interface of nontoxic and biologically inactive ceramics. A summary of the attachment
mechanisms with different class of ceramics is given in table 1.1.
Table 1.1 Cell attachment mechanisms with bio-ceramics1
Ravi K Chintapalli
28
Introduction to zirconia ceramics and sandblasting
High-density polycrystalline alumina (Al2O3) has been the first clinically used bio-ceramic for its
combination of excellent corrosion resistance, good biocompatibility, low friction, high wear
resistance and high strength. However, tetragonal zirconia (ZrO2) ceramics stabilized with yttria
(Y2O3) and magnesia (MgO2) were later introduced as medical grade bio-ceramics because of
their superior fracture toughness and tensile strength over alumina. Further research on zirconia
lead to wide spread use of this material as ball heads for Total Hip Replacements (THR). In early
stages, various solid solutions (ZrO-MgO, ZrO2-CaO, ZrO2-Y2O3) were examined for such
applications, but more recently the focus has been on yttria-zirconia ceramics2. An ISO 13356
standard was developed to describe the minimum requirements of TZP (tetragonal zirconia
polycrystal) ceramics for use as surgical implants. Table 1.2 shows the properties of different
bioinert bio-ceramics.
Table 1.2 Characteristics of some bio-ceramics2
Stabilized tetragonal zirconia has excellent properties like high toughness, low thermal
expansion, high corrosion resistance, low thermal conductivity, and large electrical conductivity
at high temperature. The phase instabilities in pure zirconia made it useless without being
stabilized with dopants like Y2O3, MgO, and CeO. As mentioned earlier, zirconia ceramics
doped with yttria (Y2O3) in different concentrations are of great interest in biomedical as well as
in several engineering applications.
The enhanced fracture toughness of zirconia is attributed to the stress induced phase
transformation from tetragonal (t) to monoclinic (m), which is also known as transformation
toughening3. Moreover, the detection of ferroelastic switching which also leads to toughening in
zirconia boosted the use of this material in critical applications4. In spite of its superior
properties5 over other ceramics, it was found that some zirconia compositions could suffer a
slow t-m transformation at the sample surface in a humid atmosphere for applications near 250
0
C, which resulted in microcracking and a loss in strength as discovered by Kobayashi et al6. The
Ravi K Chintapalli
29
Introduction to zirconia ceramics and sandblasting
phenomenon is widely called as hydrothermal degradation (HD) or low temperature degradation
(LTD).
Biomedical grade 3Y-TZP (3mol % yttria doped tetragonal zirconia) was widely popular as
restorative material for THR. More than 300 000 TZP ball heads had been implanted in THR
surgeries with only a couple of failures reported until 1997. The promising success of this
material was eventually followed by a hundreds of failures of implanted femoral heads in 2001.
These failures were attributed to accelerated aging of femoral heads produced in particular
batches7. Since then, this material has been intensively studied for the causes of failure and for
the mechanisms of degradation. The excellent reviews by Piconi2, Chevalier7 and Lawson8 on
low temperature degradation also deal with the use of zirconia as biomaterial. Recently, 3Y-TZP
ceramics are increasingly used in dental restorations; a more detailed description will be given
later in this chapter.
1.2 Zirconia ceramics
1.2.1 Zirconia polymorphs
Pure zirconia ceramics exists in three different crystallographic forms depending on the
temperature. At ambient pressure and high temperatures (>2370 0C) it exhibits cubic (c) (Fm3m)
structure. At intermediate temperatures (1200–2370 0C) it changes the crystallographic form
from cubic to tetragonal (t) (P42/nmc) structure. It shows monoclinic (m) (P21/c) crystal structure
at low temperatures (<950 0C). At high pressures, it exhibits two distinct orthorhombic structures
(space groups Pbca and Pnam), and a third orthorhombic form (space group Pbc21), closely
related to the first of these, has been observed in some partially stabilized zirconias (PSZs)9. The
three crystallographic forms of pure zirconia are shown in figure 1.1.
Figure 1.1: Schematic representations of the three crystal structures of ZrO2 (a) cubic, (b)
tetragonal, and (c) monoclinic9.
Ravi K Chintapalli
30
Introductio
on to zirconia ceramics and
d sandblasting
g
The t-m
m phase traansformationns are assoociated with
h volume changes,
c
whhich can cause microo
crackingg during cooling. Thee tetragonaal to monoclinic transsformation begins at 950 0C onn
cooling and is reveersible at 11150 0C up on heating10. The t unit cell containns two ZrO2 units, andd
me is half in
i comparisson with thee c and m un
nit cells whhich containn four ZrO2 units. Bothh
its volum
the t and m structuures are desccribed as diistortions off c structuree with majoor distortion
n associatedd
m transform
mation correesponding too a shear an
ngle of 90 parallel
p
to thhe basal plaane of the t
with t-m
unit celll, thus form
ming a monooclinic  anggle of 810.9
1.2.2 Sttabilization
n of zirconiia
Pure zirrconia is staabilized by some oxidee dopants su
uch as MgO
O, CaO andd Y2O3. Theese dopantss
suppress the phase transformaations and alllow the cerramic to stabbilize eitherr in cubic orr tetragonall
forms at
a any tempperature. Diffferent typees of stabiliized zirconiia can be prepared usiing suitablee
cooling rates durinng sintering. Fully stabiilized zircon
nia (FSZ) iss formed wiith full cubiic structure.
Partiallyy stabilizedd zirconia (PSZ) is formed wiith duplex structure ccontaining cubic andd
tetragonnal phases as
a shown inn figure 1.2. Polycrystaalline tetraggonal zirconnia (TZP) iss formed inn
metastaable state wiith full tetraagonal phasee when coolled down too room tempperature.
Figurre 1.2: Phasee diagram off ZrO2-rich sections
s
of the phase diaggrams and off ZrO2–Y2O3 system11.
Commeercial grade zirconia ceeramics aree generally produced with
w 1.75–33.5 mol% (3
3.5– 8.7 wtt
%) yttriia (Y2O3). As
A mentioneed above, too produce ZrrO2-based materials,
m
oxxide additiv
ves are usedd
to generrate metastable t-phase. In TZP, oxides thatt have highher solubilitty at low teemperaturess
chemicaally stabilize the t phasse12. The PS
SZ and TZP ceramics derived
d
from
m these systems are thee
Ravi K Chintapalli
31
Introduction to zirconia ceramics and sandblasting
two most commonly used ZrO2-based engineering materials9. The content of the tetragonal and
cubic phases varies with a minimum 60% of t-phase depending on the composition, firing time,
and temperature. The microstructure of as sintered 3Y-TZP consists of 0.5–2 m diameter
uniform equiaxed grains. Moreover, a glassy grain-boundary phase, rich in SiO2 and Y2O3, is
almost always present in varying thickness, depending on the method of consolidation9.
1.2.3 Tetragonal (t) to Monoclinic (m) transformation
Tetragonal to monoclinic phase transformation has a significant role in Y-TZP ceramics in either
improving their toughness under externally applied stress or decreasing the properties under
humid environment. The nature of t-m transformation is explained by phenomenological
martensitic crystallography. The fundamentals of martensitic transformation in zirconia are
given below.
1.2.3.1 Martensitic transformations
Martensitic transformation by definition is an athermal diffusionless change in crystal structure
which involves simultaneous cooperative movement of atoms over distance less than an atomic
diameter, resulting in macroscopic shape change in the transformed region. According to the
phenomenological theory explained by Kelly and Rose13 martensitic transformations are
described in purely mathematical terms rather than describing the actual physical mechanisms.
An invariant plane strain (IPS) resulting from a transformed region is the basis of this theory. An
IPS is a strain which leaves invariant plane (the habit plane) in two phases. Figure 1.3 illustrates
the invariant plane strain (shape strain S), which consists of an expansion or contraction ()
normal to the invariant plane together with a shear (parallel to the invariant plane.
Figure 1.3: Invariant plane strain (IPS), IPS (shaded area) is composed of shear parallel to habit plane
and expansion/contraction normal to habit plane 13.
From figure 1.3, the strain perpendicular to both the shear direction and to the normal of the
habit plane is zero. The transformation induced volume change (V) is accommodated by the
expansion or contraction normal to the invariant plane (V). The overall strain S can be
Ravi K Chintapalli
32
Introductio
on to zirconia ceramics and
d sandblasting
g
mathem
matically desscribed as the
t product of three diifferent trannsformations: i) a latticce invariantt
shear (L
LIS) that dooes not channge the cryystal structu
ure as, for example,
e
in slip or twin
nning; ii) a
Bain strrain (B) whhich is the strain
s
needeed to changee the parent crystal strructure into that of thee
productt; iii) a rigidd body rotation R whichh is required to keep thhe habit planne unrotated. Then thee
total strain S is given by equattion 1.1
S = RBL
L
(1.11)
The stepps involvedd in the abovve equation are schemaatically illusstrated in figgure 1.4. The
T changess
in the crystal structure are shoown on the left and maacroscopic transforminng volume on
o the rightt
t figure 1.4.
side of the
Figgure 1.4: Steepwise illustrration of pheenomenologiccal theory off martensitic transformattions13.
Lattice corresponddence betweeen the pareent and pro
oduct phase in martenssitic transformations iss
importaant to deteermine the Bain straain. Kirven
n14 initiallyy proposed three sim
mple latticee
correspoondences foor t-m transfformation inn zirconia and
a later theey were refiined by Hay
yakawa andd
coworkeers15–17. Foor t-m transsformation in zirconia, a tetragonnal zirconiaa crystal iss arbitrarilyy
denotedd as the axiss at, bt and ct. The resuulting product monoclinnic phase iss denoted ass am, bm andd
cm. The corresponddence A occcurs when at becomes am, similarlly corresponndence B occurs whenn
bt becom
mes bm. Figuure 1.5 show
ws the scheematic illusstration of thhe three geeneral corresspondencess
for A, B and C forr t-m transfformation with
w two exp
pected orienntations forr each correespondence.
In general, choice of the orienntation relaationship wiill indicate the respecttive corresp
pondence orr
on.
selectinng a correspoondence will suggest thhe orientatio
Ravi K Chintapalli
33
Introductio
on to zirconia ceramics and
d sandblasting
g
Figure 1.5:
1 Schemattic illustratioon of generall correspond
dences for A, B and C for t-m transforrmation with
two exp
xpected orienntations for each
e
correspoondence13.
For exaample from
m figure 1.55, in case of
o t-m transsformation in zirconiaa with correespondencee
CAB; thhe monoclinnic axis bm is perpendiicular to both cm and am in order tto be parallel with onee
of the teetragonal axxis. Similarrly either am or cm willl be parallel to a tetraggonal axis. This yieldss
two posssible orienntation relaationships i)) bm will be
b approxim
mately paraallel to corrrespondingg
tetragonnal axis andd (001)m pllane is paraallel to (10
00)t or (0011)t plane, iii) cm is app
proximatelyy
parallel to a tetragoonal axis, annd [100]m parallel
p
to [001]
[
00]t plane. T
These two orientations
o
s
t or [10
dences resuulting in sixx possible orientationn
combinee with eacch of the thhree basic correspond
relationnships as shoown in figuure 1.5. Forr an in-deptth descriptioon on marteensitic transsformationss
readers are referredd to the exceellent review
w of Kelly and
a Rose13.
Ravi K Chintapalli
34
Introductio
on to zirconia ceramics and
d sandblasting
g
As menntioned earlier, that thee habit planne is unrotaated and disstorted durinng transform
mation; thee
lack of this rotationn is accomppanied by a microscop
pic inhomoggeneous sheear of marteensite plate.
This inhhomogeneouus shear is accommodaated in the martensite
m
p
plate
by eithher internall twining orr
internal shear18. Fiigure 1.6 shhows the schematic
s
illlustration of
o twin relaated and sh
hear relatedd
martenssite plates.
Figure 1.6: Schemaatic illustratioon of a) interrnally twinneed and b) intternally slippped martensitic plates19.
The stacck of monooclinic variaants formedd due to pro
ogression off transformation is rep
ported to bee
only twin related20. The progression of traansformatio
on will be described
d
latter in this seection. T-M
M
transforrmation by hydrotherm
mal degradaation and/o
or by externnally applieed stress leeads to thee
formatioon of thin plates of monoclinicc phase9,13,21,22. Figuree 1.7 show
ws some ex
xamples off
monocliinic plates in
i zirconia ceramics.
c
Figure 1.7: Examples of transfo
formation twiinning in zircconia ceramiics. A) TEM micrograph of partially
formed t-graiin ince-TZP9, b)TEM image of hydro
othermally deegraded 3Y-T
TZP21, c)Tra
ansformed
transfo
penttagonal grainn with grain boundary crracks and d) SEM image of a section pperpendicula
ar to the
degraaded surface of D60H. Grains
G
with monoclinic
m
tw
wins as well as
a microcraccks can be ob
bserved23.
Ravi K Chintapalli
35
Introduction to zirconia ceramics and sandblasting
1.2.3.2 Stress induced transformations
As mentioned before, the shape strain induced by the transformation t-m can be represented by a
shear parallel to the habit plane and dilatational strain normal to the habit plane. The dilatational
strain corresponds to a volume change (∆V) in the form of expansion. The strain energy
associated with the martensitic transformation involves large shear strains which are, larger (3 to
4 time larger) than the dilatational strain13.
In order to initiate the transformation the net change in free energy must be sufficient to
overcome a free energy barrier. The net change in free energy accompanied with the nucleation
of a martensite plate is the sum of change in chemical free energy per volume associated with tm transformation and the energy required to form the nucleus. Lange24 described conditions for
the transformation in terms of energy contributions considering the energy of a tetragonal
particle embedded in an infinite matrix.
The change in total free energy (∆Gt-m) for the t-m transformation is given by
∆Gt-m = ∆Gc + ∆USE + ∆US
(1.2)
where ∆Gc is the difference in chemical free energy between the tetragonal and monoclinic
phases (<0 at temperatures below T0, see figure 1.8). This term depends on temperature and
composition, and also on the oxygen vacancy content. It can be written as,
∆Gc = ∆St-m (T-T0)
(1.3)
where ∆St-m (>0) is the transformational entropy change, and T0 is the transformation temperature
at which unconstrained tetragonal phase begins to transform to the monoclinic phase. ∆USE is
positive and refers to the change in elastic strain energy associated with the transformation of
particles. This term depends on the modulus of the surrounding matrix, the size, and shape of the
particle, and the presence of internal or external stresses. The term, ∆US is the change in energy
associated with the formation of new interfaces associated with transformation, which is also
positive. Transformation will take place when ∆Gt-m 0, that is,
-∆Gc  ∆USE + ∆US
Ravi K Chintapalli
(1.4)
36
Introductio
on to zirconia ceramics and
d sandblasting
g
At the temperature Ms, -∆Gc = ∆USE + ∆US. Deccreasing thee term ∆U
USSE or the surrounding
s
g
constraiint raises Ms Temperatuure and vicee versa.
n driving force for the
t transforrmation and
d results inn
Externaally appliedd stress moddifies the net
change of effectivee transform
mation tempperature. In such case, for each ppotential varriant of thee
habit pllane, the coomponents of the appplied stress parallel annd normal tto the habit plane aree
superim
mposed on the
t shear and
a dilatational compo
onents of the
t shape sstrain respeectively25,26.
Eshelbyy 27 describeed the dilataational compponent of ∆U
∆ SE as:
∆
1
2
1.5
Where, σ is the unniform tensiile constrainnt of the traansformed particle witthout any in
nitial stresss
and e iss the deform
mation by traansformatioon without any
a constraiint. The sheear componeent of ∆USEE
is givenn by
∆
1
2
1.6
Where, τ and γ are the shear sttress and shhear deformaation of thee shape defoormation com
mponent off
the transformation..
F
Figure
1.8: Relationship
R
between freee energy changes for t-m transformattion in zircon
nia13.
Ravi K Chintapalli
37
Introduction to zirconia ceramics and sandblasting
The large shear component of the shape strain of the t–m transformation will have a greater
effect than the dilatational component. The main difference between the dilatational and shear
strain is that shear strain can change the sign but not the dilatational strain; which implies that for
a positive volume change, the uniaxial tension will provide positive stress-assistance, whereas
uniaxial compression will oppose the transformation. However, when a suitable orientation of
the variant is available, both the uniaxial tension and compression leads to positive stress
assistance13.
The last term of the equation 1.2 is the change in interfacial surface energy, which can be
explained by the size effect by including various surface phenomena, associated with the
transformation constraint. The size effect due to change in interfacial energy is given by Garvie
and Swain 28 as
∆
1.7
Where Am and At are the interfacial surface areas, γm and γm are interfacial energies per unit area
for tetragonal and monoclinic respectively. V is the volume of the transformed inclusion. By
assuming the inclusion as spherical with diameter D (V= π/6D3), Lange 29 described the equation
1.6 as
∆
6
1.8
Where gs =At / Am.
1.2.3.3 Martensite formation and propagation
Stress induced transformations are not only limited to the situation of externally applied stresses.
Internal stresses generated during sintering and subsequent cooling leads to anisotropic
expansion or contraction of individual grains, which can also stress-induce the transformation. In
zirconia transformation begins with the formation of the self-accommodating martensite plates.
Figure 1.9 shows various stages of the transformation. When the first martensite plate forms, the
shape strain automatically generates opposing stresses in the surrounding matrix. These stresses
increases with the growth of the martensite plate and they eventually stop the progress of the
Ravi K Chintapalli
38
Introductio
on to zirconia ceramics and
d sandblasting
g
transforrmation. Wiith this situuation, the surrounding
s
g matrix is subjected tto local inteernal stresss
that leadds to stress--induced maartensite plaate formatio
on when thee following cconditions are
a met 13:
-
T stress induced
The
i
subbsequent maartensite vaariant must have
h
a habiit plane thatt is close too
t originall martensite plate generrating the in
the
nternal stresss.
-
T shear component
The
c
o the shapee strain of th
of
he second variant
v
mustt be equal an
nd oppositee
t that of thhe initial varriant.
to
-
T dilatatiional compoonents of thhe shape straain have alw
The
ways the sam
me sign.
mmodatingg
Wheen all the above connditions aree satisfied the formattion of thee self accom
marrtensite variiants can reesult in trannsformed vo
olume, wheere the overrall shear sttrain of thee
pairr is effectiively zero as shownn in figuree 1.9 (c). Nucleationn of the second
s
selff
accoommodatingg variants is most likelly to occur in the regioons of stresss concentraation of thee
initiial variant marked
m
withh X in figuree 1.9 (b). Th
he dilatationnal componnent of the shape
s
strainn
assoociated withh the plates is the sum of the two individual plates in thhe pair. As the processs
conttinues form
mation of self accommoodating pairrs will occuupy the whoole transforrmed grain.
The transformeed region inn figure 1.99 (f) appearss as a stackk of parallell variants fo
ormed by a
sequuential proccess for whicch an initiall single variiant is necesssary.
Figure 1.9:
1 Schemattic illustratioon of stages in
i transformation of spheerical tetragoonal zirconia
a particle to
13
seelf accommoodating mono
oclinic variants .
Ravi K Chintapalli
39
Introductio
on to zirconia ceramics and
d sandblasting
g
The nuccleation straains to form
m the first variant
v
are dominated
d
b shear coomponent of
by
o the shapee
strain raather than dilatational strain. Nevvertheless, the effect of shear strrain is acco
ommodatedd
with traansformatioon progresssion and thhe strains associated
a
with the nnet transforrmation aree
dilatatioonal. Moreoover, internnal energy is required
d by the eaach pair off self accommodatingg
variantss to generatte an additiional interfaace with th
he transform
med volumee. This interrnal energyy
must bee balanced between
b
thee strain enerrgy decreasee and the ennergy increaase associatted with thee
formatioon of additiional surfaces 30,31.
Figuure 1.10: Schheme of autoocatalytic traansformation
n a) single maartensite plaate b) two add
ditional
martenssite plates in the neighbor grains trigg
ggered by thee initial platee c) a fourth pplate in the next
n grain13.
The tetrragonal mattrix surrounnding the trransforming
g region is also
a able too transform;; if the firstt
martenssite plate in one grain will
w stress-iinduce transsformation in
i the neighhboring grain resultingg
in formation of a second
s
marttensite platee provided with a suitaably orienteed variant is available.
The habbit plane off the secondd variant is approximat
a
ely parallell to the firstt variant. Ad
dditionally,,
the secoond variantt must posssess a shaape strain in
n the direcction close to that off the initiall
martenssite plate. This
T
triggeriing of marttensite platees in adjaceent grains liike a chain reaction iss
termed as autocataalytic transfo
formation32 which is sch
hematicallyy shown in ffigure 1.10.
ormation off
Howeveer, self acccommodatinng martensiite variantss are limiteed only to the transfo
isolatedd particles and
a results in
i minimal or zero sheear of the trransformed volume. On
O the otherr
Ravi K Chintapalli
40
Introductio
on to zirconia ceramics and
d sandblasting
g
hand, auutocatalyticc transformaation occurrs in homog
genous singgle phase m
materials ressulting in a
significaant value of the neet shear inn the transsformed voolume withh the prog
gression off
transforrmation awaay from thee initial zonne. The diff
fference in net
n shear iss because of
o the shearr
componnents of shaape strain inn two variannts are in op
pposite direcction for self accommo
odation andd
in samee direction inn autocatalyytic transforrmation.
oughening
g mechanis
sms in zirco
onia
1.2.4 To
1.2.4.1 Transformat
T
tion toughen
ning
Garvie et al33 firrst reportedd transform
mation toug
ghening inn zirconia ceramics. The namee
a toughnesss of the m
material incrreases as a
“transfoormation tooughening” is itself deescriptive as
result of
o transform
mation. In case of zirrconia ceraamics, t-m transformaation is thee source off
enhanceed toughnesss. Ceramics usually coontain flawss such as cracks due too processing
g, and whenn
the material is subjjected to exxternally appplied loads the pre-existing crackss tend to prropagate. Inn
such cirrcumstancess, in zirconiia, the energgy required
d to propagaate a crack tthrough the metastablee
12
phase iss increased and
a the streess on the t-pphase is relieved to alloow it to trannsform to m-phase
m
.
Figurre 1.11: Illusstration of traansformationn toughening
g mechanism in front of a propagating
g crack34.
Figure 1.11
1
shows the mechannism of touughening du
ue to the traansformationn. The tensile stress att
the cracck tip inducces phase transformati
t
ion, which results in increase
i
in volume. Th
his volumee
change will generrate compreessive stressses around
d the crackk, which w
will stop furrther crackk
propagaation35. McM
Meeking annd Evans
35
have develloped a moddel for transsformation tougheningg
based onn crack shieelding.
Ravi K Chintapalli
1.9
41
Introduction to zirconia ceramics and sandblasting
where KItip is the stress intensity factor at the crack tip, KI is the applied stress intensity factor
and KIsh is the shielding factor, which includes the mechanisms of the toughening. In zirconia
ceramics, three principal shielding mechanisms have been identified as contributing to KIsh 36–38;
transformation toughening, transformation induced microcracks toughening, and crack
deflection. The extent of the contribution from each mechanism depends on the scale,
morphology, distribution and volume fraction of transforming t-phase 9.
These mechanisms decrease the applied stress intensity factor at the crack tip so that the real
stress intensity factor at the tip is lower than the applied one. Previous works
35,39
on this effect
have shown that the higher the applied stress intensity factor, the larger the transformation zone,
1.10
where,
0.214
1
1
√3
1.11 12
where E is the elastic modulus Vf is the volume fraction of the transformable particles eT is the
dilatational strain associated with the transformation,  is the Poisson ratio and mc is the critical
stress leading to transformation.
McMeeking35 proposed a uniform dilatational component of the transformation strain to
compute the crack shielding effect based on the shape of the transformation zone wake. In the
model adopted by McMeeking35 with dilatant strain assumption, a cardioid-shaped frontal zone
of transformation is formed around a stationary crack tip (see 1.12 (a)). This leads to no increase
in toughness. The transformation strains actually lead to a reduction in toughness in the 120°
sector of the transformed frontal zone immediately ahead of the crack ( 60° in Figure.1.12
(a)), but this is counteracted by a toughness increase in the remaining sector (=60° to 300°),
leading to an overall null effect.
Ravi K Chintapalli
42
Introduction to zirconia ceramics and sandblasting
Figure 1.12: a) Cardioid-shaped transformation zone associated with a purely dilatant transformation at
a crack tip. Two dotted lines define a sector at the front of the zone with an included angle of 120° (2/3),
which actually leads to a decrease in toughness. (b) Transformation wake of half height hy associated
with the movement of the crack tip by a distance a 9.
Toughness increases in the crack front, when the shape of the zone is altered to reduce the
transformation in the 120º sector ahead of the crack (where the contribution is negative); which
is possible by the inclusion of a shear component. Additionally, toughness increases if the crack
tip moves forward leaving a wake of transformed zone behind it (Figure 1.12 (b)). For more indepth knowledge on transformation toughening readers are referred to the works of Hannik,
Kelly and Rose 9,20.
1.2.4.2 Ferroelastic toughening
An additional mechanism of toughening in zirconia ceramics is ferroelastic toughening. Virkar
and Matsuimoto4 first reported higher toughness in some zirconia based ceramics, which they
attributed to ferroelastic domain switching. The domain switching in tetragonal zirconia occurs
through a ferroelastic transition. In this mechanism, the stress-induced alignment of the c-axis of
the t-phase along the maximum principal stress axis results in a shape change of a pure shear
type. The extent of the shear is of the order of 0.04 or less; it arises from the different cell
spacing along c and a directions. This mechanism operates in the stable and metastable t-phase
containing materials as an active toughening mechanism.
The domain switching was originally observed by XRD in ground zirconia4,40. The main
observation was the absence of monoclinic phase with c axis aligned preferentially normal to the
ground surface. The stress state of the ground surface is the biaxial compression with zero stress
normal to the ground surface41. Ferro elastic switching occurs in different tetragonal grains, since
the domains in the grain depend on whether the grain is newly formed or is a transformed one. In
case of formation of an individual tetragonal grain, it results in six different crystallographically
Ravi K Chintapalli
43
Introduction to zirconia ceramics and sandblasting
equivalent orientations of the c-axis. Each of these variants has the same energy but can re-orient
when a stress is applied. On the other hand, for tetragonal grains formed directly, each grain can
be a single domain. Then, individual grains or portions within each grain can be switched to a
different orientation by an applied stress12.
The toughness contribution by domain switching was quantified in high solute containing t′
(tetragonal prime) phase in 3Y-TZP between temperatures 200-1000 0C by Foitzik et al42, they
calculated the toughness contribution in the order of 3-4 MPa√m. Also in single phase cubic
zirconia this contribution is around 2.4 MPa√m. Later, Virkar43 reported that domain switching
in t′ alloys is possibly a high temperature toughening mechanism, because this mechanism is less
sensitive to temperature than t-m transformation. In spite of some experimental knowledge on
toughening contribution, according to Hannik9 and Chevalier12 the reported values are only an
approximation and an additional work is needed to validate these values.
1.2.5 R - Curve behaviour
Crack resistance or resistance curve is generally referred as R-curve. The crack resistance
behavior is well reported for metals44 as a characteristic of the material, generally associated
with plastic deformation resulting from increasing plastic flow at a crack tip. In ZrO2 ceramics
R-curve behavior is described as a phenomenon wherein crack resistance increases with
increasing crack propagation45. McMeeking and Evans35 have attributed increase in crack
resistance to the transformation wake zone behind a propagating crack.
Figure 1.13: R-curve behavior of different ceramics46.
Ravi K Chintapalli
44
Introduction to zirconia ceramics and sandblasting
R-curve behavior in different ceramics subjected to compact tension is shown in figure 1.13. CeTZP and Mg-PSZ exhibited a strong crack resistance where as Y-TZP has no signs of R-curve
behavior. The strong crack resistance in Mg-PSZ is attributed to the transformation toughening
including other extrinsic mechanisms such as crack deflection and crack shielding47.
In other works48,49, it has been shown that the crack resistance in Y-TZP depends on grain size.
Eichler et al48 found that plateau values exhibiting a linear dependence with grain size. For a 2YTZP with grain size of 300 nm the plateau values of toughness are 5.4 and 6.7 MPa√m in air and
in vacuum respectively. However, the raising crack resistance with increasing crack length in air
was not observed due to the smaller transformation zone. Whereas, Fargas et al50 observed a
slight increase in plateau toughness in air from 3.9 to 4.3 MPa√m in a 300 nm grain size Y-TZP.
This is because of 90% of the plateau of fracture energy is reached at five times the height of the
transformation zone, and zone widths in most transformation-toughened ceramics are in the
range 0.1 to 15 m. Additionally, plateau toughness is found 30-40 % higher in vacuum than in
air.
R-curve is generally obtained from macrocracks and microcracks introduced by various
indentation techniques and crack extension studies from natural flaws. Munz51 in a recent review
on R-curve measurements reported that, the R-curve behavior of a material can be measured with
macrocracks; a quantitative prediction of the R-curve for natural flaws is not possible from
measurements in specimens with macrocracks or indentation cracks. Moreover, independent of
how R-curve is obtained for a material with natural flaws, materials with a rising R-curve lead to
an increase in the strength compared with a material with the same crack tip toughness and a flat
R-curve. Additionally, It was shown that coarse-grained materials lead to a rising R-curve and,
therefore, to high strength values.
1.2.6 Low temperature degradation (aging) of zirconia
A critical and widely discussed characteristic of 3Y-TZP is the phenomenon of low temperature
degradation (LTD) or hydrothermal degradation or aging in air and aqueous environments at
relatively low temperatures. This effect was first observed at temperatures close to 250 °C in air
by Kobayashi et al6 and it is widely documented in the literature7,8,52–54. The interaction of water
activates surface t-m transformation and deteriorates surface mechanical properties.
Ravi K Chintapalli
45
Introduction to zirconia ceramics and sandblasting
For several years, it has been thought that low temperature degradation was negligible at the
human body temperature, but recently there have been many reports, which document the
existence of monoclinic phase at the surface of explanted femoral heads several years after
implantation55–57. Thus, LTD at the human body temperature is one of the main reasons for the
failures of femoral heads.
Since the discovery of Kobayashi et al6 many authors7,8,52–54 have investigated the mechanism of
LTD, but precise mechanism is still at debate. The study of LTD in medical grade zirconia at
human body temperature is usually carried out in vitro at accelerated conditions in an autoclave7.
Yoshimura58 summarized the experimental observations of low temperature degradation on
zirconia as follows.
- The degradation proceeds most rapidly at temperatures of 200-300°C and is time dependent.
- The degradation is caused by the tetragonal-monoclinic transformation accompanied by
micro- and macro-cracking.
- The transformation progresses from the surface to the interior of the specimen.
- Water or water vapor enhances the transformation.
- A decrease in grain size and an increase in stabilizer content delay the transformation.
1.2.6.1 Aging mechanism
In spite of abundant research work on degradation of zirconia, the researchers have not yet
reached to a consensus for the precise mechanism of degradation. However, several theories are
accepted. Widely accepted aging mechanism is presented here.
Sato and Shimada59 reported a reaction between water and Zr-O-Zr bonds. The reaction between
water and zirconia is described on the basis of activation energies for transformation in Y- and
Ce-ZrO2. As the results in both stabilizers were found to be similar, it was concluded that the
water reacted primarily with the Zr-O-Zr bonds on the surface and not with the stabilizing oxide.
Similarly, Yoshimura58 and Guo in his series of works53,60–62 proposed a mechanism based on
formation and penetration of OHˉ ion. They observed the inclusion of OHˉ ion upon aging and
exclusion of OHˉ ion upon annealing by infrared spectroscopy. Based on their findings the
sequential mechanism is summarised as follows and is shown in figure 1.14.
- Chemical-adsorption of H2O at the surface.
Ravi K Chintapalli
46
Introduction to zirconia ceramics and sandblasting
- Reaction of H2O with O2 on the ZrO2 surface to form OHˉ ions resulting in stressed sites at
the surface.
- Penetration of OHˉ ions into the inner part by grain-boundary diffusion resulting in
accumulation of strain in the lattice.
- Annihilation of oxygen vacancies by OHˉ ions leads to destabilization of tetragonal phase
- The nucleation of monoclinic phase in the tetragonal grains: then the tetragonal-monoclinic
transformation yields micro and macrocracking.
1st step
Adsorption
2nd step
Surface
Stress
Vacancy
3rd step
4th step
Migration
Migration
Stress accumulated area
Nucleation
Figure 1.14: OHˉ ion trigger mechanism for degradation58.
The following equations describe the degradation mechanism.
→ 2
1.12
Where O′′surf is the oxygen at the surface and OH′surf is the OHˉ ion formed on the surface of
zirconia. Stabilizing agents such as yttria or magnesia usually introduces oxygen vacancies in
zirconia and diffusion of these vacancies due to the reaction in equation (1.12) will trigger
another reaction as
Ravi K Chintapalli
••
→
•
,
1.13
47
Introduction to zirconia ceramics and sandblasting
Where, V ••O is the oxygen vacancy and (OH) •O is an OHˉ on the vacant oxygen site in zirconia
lattice, and SxO,surf is an oxygen vacancy at the surface. Additionally, it is also possible that the
diffusion of OHˉ ions takes place across the grain boundary due to their high concentrations at
grain boundaries and because of diffusion is generally easier along these paths. This activates the
reaction of OHˉ ions with the oxygen vacancies at the grain boundary. In this case, the equation
1.13 can be written as
••
→
•
,
1.14
Where OH′gb is the OHˉ ion at the grain boundary and SxO,gb is the oxygen at the grain boundary.
By reaction (equation 1.13), the oxygen vacancies in the surface layers are annihilated. When the
oxygen vacancy concentrations are reduced, the tetragonal phase is no longer stable and t-m
transformation occurs in the surface. When adequate amount of the phase transformation take
place, both micro and macrocracks are produced in the transformed surface due to the volume
expansion associated with the phase transformation. These cracks open up new surfaces to react
with water species, leading to a further spontaneous transformation. Due to this phenomenon
degradation proceeds further along grain boundaries and cracks.
1.2.6.2 Nucleation (N) and Growth (G)
In support to the above described atomic scale mechanism that is hydrothermal ageing is related
to the diffusion of water, as mentioned earlier, the martensitic transformation is triggered by the
presence of water and its progress inside the material; this progress is governed by a mechanism
of diffusion which is a thermally activated process. This makes that t-m transformation to
proceed from one grain to the neighbouring grains by a nucleation and growth mechanism as
shown in figure 1.15. This phenomenon has been experimentally validated by Chevalier52. A
monoclinic plate can pass through two or more grains, but in most cases martensitic laths can be
restricted by grain boundaries to pass from one grain to another. Transformation occurs
randomly and also preferentially21 on the neighbouring grains.
Figure1.15: Nucleation and growth aging mechanism63.
Ravi K Chintapalli
48
Introduction to zirconia ceramics and sandblasting
Nucleation occurs on the unstable grains depending on the grain size, yttria content and/or level
of tensile stresses (either internal or applied). The number of nuclei increases steadily with the
stresses, due to the penetration of water. Simultaneously, growth occurs because the
transformation of one grain puts its neighbours under tensile stresses, favouring their
transformation under the effect of water7. In a recent study Muñoz et al21 also report a nucleation
and growth mechanism from surface to bulk. Additionally, they found that thickness of
monoclinic phase on the surface is more than that of the microcracked layer. N-G mechanism is
emphasized by the fact of diffusion of water species which trigger the non-diffusional
martensitic transformation
1.2.6.3 Aging kinetics
Time dependent transformations are often described by Mehl–Avrami–Johnson (MAJ) equations
that are very much used in metals64. In zirconia ceramics, during aging, the transformation
occurs at the surface and proceeds slowly into the bulk as a nucleation and growth mechanism.
Nucleation and growth of aging is directly referred to formation of m-phase and its increase with
time. Several authors52,64–66 have studied t-m transformations with respect to time and found a
sigmoidal behavior. The relationship between the time and the amount of m-phase fraction is
described by MAJ equation as
1
1.15
Where, f is the transformation fraction, t is the time, and b and n are constants. The exponent n is
related to the nucleation and growth conditions. The parameter b is related to the activation
energy Q and is given by
1.16
Where, b0 is a constant, R is the gas constant and T is the absolute temperature.
Chevalier et al52 studied the relationship between the amount of monoclinic phase and aging
time at various temperatures and found that m-phase increases with aging according to the
sigmoidal behavior. Figure 1.16 shows the m-phase fraction measured by X-ray diffraction at
various temperatures with respect to time. MAJ equations are fitted well with an exponent n of
3.6 to the experimental observations. Chevalier et al52 also emphasized the fact that the model
Ravi K Chintapalli
49
Introduction to zirconia ceramics and sandblasting
gives good prediction of monoclinic phase for a given time and temperature for a particular
zirconia ceramic. For zirconia with different microstructure and stabilizer content parameters
models must be experimentally determined
Figure1.16: Monoclinic content versus time at different temperatures52.
1.2.6.4 Factors influencing aging
Several factors influence aging and the extent of damage caused by aging. The main factors
influencing aging are grain size, stabilizer content, porosity, internal stresses and composition.
For instance decreasing grain size influences the surface term in equation 1.2, so that the
transformation becomes difficult. The surface term, leads to a dependence of the activation
barrier for t-m transformation on the particle size. Therefore, the activation barrier to form a
critical nucleus is increased by a decrease in particle size. This effect qualitatively explains the
dependence of amount of transformation on grain size67. Subsequently several works68–70 found
that 3Y-TZP with grain sizes below 200 nm is resistant to low temperature degradation.
The other influencing factor is the stabilizer content due to the number of oxygen vacancies in
the crystal lattice induced by doping. As mentioned earlier, the oxygen vacancies are replaced by
the newly formed OHˉ ions to destabilize the t-grains. So, it is expected that increasing the
stabilizer content will delay the degradation. For instance, 4% molar yttria stabilised zirconia
shows better resistance to degradation as compared to the 3% molar yttria stabilised zirconia71.
The stress distribution is a critical factor for transformation72, stresses with the same sign as the
component of transformation tensor are known to assist transformation, so that shear and tensile
are destabilizing, whereas compression is stabilizing. Schmauder and Shubert73 found that
unconstrained grains were stable under humid conditions, whereas constrained grains
transformed. They attributed this behavior to the principal stresses from the thermal expansion
Ravi K Chintapalli
50
Introduction to zirconia ceramics and sandblasting
anisotropy, which enhances the free energy of stressed grains and decreases the nucleation
barrier to destabilize the phase.
1.2.6.5 Consequences of aging
Hydrothermal degradation of zirconia ceramics has detrimental effect on their performance in
biomedical applications. The degradation over the surface leads to the following consequences,
which depend on many critical parameters like grain size, size distribution and content of yttria,
density, and sintering process:
•
Volume increase (4%)
•
Micro cracking
•
Loss of strength in the material
Due to these consequences the reliability and performance of bio medical grade zirconia in
biological environment is a concern. For instance, in vivo case studies have shown that, volume
increase at the surface leads to increased wear rate of the implant when in contact with other
surface. Microcracking and surface roughening leads to osteolysis and eventually cause aseptic
loosening. Additionally, mechanical stability of the implant is decreased due to loss of strength
in the ceramic.
1.3 Zirconia ceramics in restorative dentistry
Due to the high aesthetic demand for all types of dental restorations all-ceramic materials are
replacing the metal systems that were previously used. Initially in 1960´s feldspathic porcelain
was used to make crowns and eventually other ceramics such as alumina and zirconia made
inroads into field of dental restorations. However, zirconia due to its superior strength and
toughness (table 2) has overshadowed alumina. Use of zirconia based ceramics in restorative
dentistry has increased since 1990 for applications like, implants and implant abutments,
orthodontic brackets, cores for crowns, included endodontic posts and fixed partial denture
prosthesis (FPDP) frameworks74–76. Figure 1.17 shows the several types of frame works made by
zirconia based ceramics.
Dental community is particularly interested in zirconia based ceramics due its highest
mechanical properties among other dental ceramics77. On the other side, the metastable
tetragonal zirconia is prone to t-m transformation due to stress-generating surface treatments
such as grinding or sandblasting. As a consequence, surface compressive stresses are induced,
Ravi K Chintapalli
51
Introduction to zirconia ceramics and sandblasting
which are believed to increase the flexural strengtht78. Three different types of zirconia materials
are mainly employed in dental restorations. Bio medical grade yttria stabilized zirconia (3YTZP), magnesia partially stabilized zirconia (Mg-PSZ) and glass infiltrated zirconia-toughened
alumina (ZTA)77.
Figure 1.17: a) dental crowns b) three crown bridge34 c) all ceramic implants d) fitted abutment e)
cemented zirconia restorations, f)radiographic evaluation of the two zirconia restorations34.
The mechanical properties of 3Y-TZP strongly depend on its grain size3. Below a critical grain
size, 3Y-TZP is stable and spontaneous t-m transformation is not favored. The grain size of 3YTZP ceramics for dental applications consists of small equiaxed grains in the range of 0.2–0.5
m depending on the sintering temperature79. Sintering temperature influences the grain size,
which further affects the stability and mechanical properties. Higher sintering temperatures and
longer sintering times lead to larger grain sizes11,80. The flexural strength and fracture toughness
of zirconia ceramics used in dentistry are approximately ~1000 MPa and ~7 MPa√m
respectively77.
Magnesia partially stabilized zirconia (Mg-PSZ) is less successful because of very large grain
size (30–60 m) and porosity compared to 3Y-TZP. The commercial Mg-PSZ (Denzir-M,
Dentronic AB) is usually stabilized with 8 and 10 mol% of magnesia3. Mg-PSZ has low
mechanical properties and is less stable than 3Y-TZP due to the difficulty of obtaining Mg-PSZ
precursors free of SiO2. Due to this difficulty magnesium silicates can form that lower the Mg
content in the grains and promote t-m transformation77. On the other hand, glass infiltrated
zirconia-toughened alumina also has some degree of porosity due to poor dispersion of different
phases. As a result the mechanical properties are reported to be low compared to 3Y-TZP79.
Ravi K Chintapalli
52
Introduction to zirconia ceramics and sandblasting
Recently developed nanocomposites NANOZR (Panasonic, Japan) based on ceria stabilized
zirconia and alumina (10 mol % CeO2 and 30 vol. % Al2O3) is claimed to have higher strength
1500 MPa, fracture toughness 18 MPa√m (Indentation fracture method) and are highly resistant
to low temperature degradation compared to conventional 3Y-TZP34,81. However, the fracture
toughness is found to be nearly 11 MPa√m using the standard toughness measurements82. The
reason for very high fracture toughness of this material is its heavy transformation (t-m) under
stress.32
1.3.1 Nanocrystalline / Nanocomposite materials
Due to several drawbacks in the above mentioned materials, new materials with microstructure
refinement and nanocomposites by adding carbon nanotubes to the zirconia matrix are being
developed69,70. The aim of developing these materials is to achieve superior resistance to
hydrothermal degradation with improved fracture toughness for use in bio medical
applications83,84.
As mentioned before in 3Y-TZP, decreasing the grain size will reduce its capability to transform
thus limiting the material toughness by transformation. Recent studies found that toughness is
improved by alternative methods such as adding carbon nanotubes84–86. Nevertheless, this is not
very clear as these materials are still in development phase; however, biocompatibility and
toxicity aspects must be resolved before using them in biomedical applications. Giving due
consideration to the emerging materials, this thesis also focuses on the behavior of these
materials under several conditions.
1.3.2 Types of dental ceramics and clinical applications
Table 1.3 and 1.4 lists several bio medical grade dental ceramics systems used by commercial
manufacturers and their clinical applications.
Table 1.3 Commercial zirconia frame works for clinical applications74
Core material
Ravi K Chintapalli
System
Manufacturing techniques
Clinical indications
53
Introductio
on to zirconia ceramics and
d sandblasting
g
Table 1.44 Dental ceraamic materials and forminng methods34
a frame worrks
1.3.3 Prrocessing of zirconia
In earlyy days, the restorative
r
f
frame
workks were prod
duced by hoot pressing, dry pressin
ng, and slipp
casting or machiniing, but moore recentlyy CAD/CAM
M (computter aided deesign/ comp
puter aidedd
manufaccturing) systems are increasingly
i
y used87 beecause of thheir high aaccuracy in machiningg
complexx features. Due to the high flexxural streng
gth and fraacture toughhness of ziirconia, thee
feasibiliity of makiing custom
m fit restorations with the use of CAD/CAM
M has beco
ome reality.
Howeveer, zirconiaa materials from differrent manuffacturers, may
m be proccessed diffeerently andd
have vaarying levelss of stabilityy. Final restoration quaality is direectly dependdent on the careful andd
88
accuratee control off the manuffacturing prrocess and thorough
t
teesting of thee material reliability
r
.
Due to several proocessing methods
m
usedd by different manufaacturers andd a full desscription off
them is beyond thee context off this thesis (see table 1.4 for the forming
f
meethods used
d by variouss
manufaccturers).
Ravi K Chintapalli
54
Introduction to zirconia ceramics and sandblasting
A general processing method for LAVA dental ceramics is briefly described. For fabrication of
LAVA zirconia frame works, the spray dried zirconia powder (0.07-0.3 m) is isostatically
pressed. The pressed green bodies are then pre-sintered. The partially sintered blocks are
mounted in a holder to be placed in a CAD/CAM system for milling. The pre sintered and milled
blocks are finally sintered in a furnace at temperatures between 1350°C and 1500°C. The porous
pre-sintered zirconia shape shrinks by approximately 20%, thus achieving its strength and optical
properties88.
1.3.4 Implications
Before the clinical use, the crowns and implants undergo several finishing procedures like
grinding, polishing and sandblasting74,77,78,89,90. These surface treatments will trigger t-m phase
transformation in the material and also will induce surface defects91. As mentioned earlier, the
treatments like grinding and sandblasting induce surface compressive stresses which counter act
the defects, and increase the strength92. Although the effect of these treatments has been studied
by several authors34,81,91,92 the severity and the extent of damage is not well established.
Moreover, once in the oral environment the crowns and implants are in contact with aggressive
oral fluids, in which, zirconia may be subjected to poor performance at humid environments due
to low temperature degradation. Several authors93,94 have studied the low temperature
degradation behavior of commercial 3Y-TZP frameworks, found that the degradation occurs in
oral simulative in-vitro conditions with a 30% reduction in material properties such as hardness
and elastics modulus. Figure 1.18 shows phase transformation after degradation of LAVA 3MESPE dental ceramic.
Figure 1.18: LTD of LAVA dental ceramic: a)XRD patterns obtained with a fixed incidence angle of 2
before and after 168 h low temperature degradation at 140 0C, b)Monoclinic fraction Xm after
hydrothermal aging94.
Ravi K Chintapalli
55
Introduction to zirconia ceramics and sandblasting
1.4 Sandblasting
Sandblasting, also known as powder blasting or abrasive blasting, is a process of impinging a
stream of particles on to a target surface with pressure. Abrasive blasting is an age-old technique,
which was patented by Tilghman95 on October 18, 1870 in USA, and which was initially used
for priming a surface for the application of paint or a sealant. Different types of blasting media
(particles) used in the process, such as are silica sand, aluminium oxide, silicon carbide, steel
balls, glass beads etc. Sandblasting has recently gained significance due to its use in several
industrial applications at low cost. Some important applications of sandblasting are listed below
- Cleaning scale or dirt from surfaces
- Protective coating from corrosion
- Micromachining
- Carving surfaces
- Roughening surfaces for adhesion
- Introducing compressive stresses
A typical sandblasting scheme is shown in figure 1.19. The scheme consists of particles flowing
with the air speed through a nozzle and hitting the target surface from some standoff distance
and with particle flow oriented at 90º to the surface.
Compressed air flow
Nozzle
90º
Particles
Stand off distance
Target surface
Figure 1.19: Schematic representation of sandblasting.
The immediate consequence of sandblasting is erosion of material from the target surface. This
process of erosion occurs by removing chunks of material due to repeated impacts of small-sized
Ravi K Chintapalli
56
Introduction to zirconia ceramics and sandblasting
particles on the surface, which further leads to the nucleation of successive imprints and cracks
close to the contact zone. There are many parameters that affect the erosion phenomenon96–99, in
particular the erosion of brittle materials is governed by the following parameters100:
- The material properties of eroded surface (especially surface hardness and toughness).
- The impact properties of erodent particles (shape, size and velocity or kinetic energy).
- The erosion conditions (contact angle, temperature during the process and of the medium
which carries the particles).
1.4.1 Principle of sandblasting
According to Slikkerveer et al101 solid particle impact can be considered as quasi-static
indentation, since the impact speeds are much smaller than the velocity of elastic waves in
materials. Figure 1.20 shows the particle impact as a quasi-static indentation process.
Considering a single particle impact, when the particle hits the target surface, high compressive
stresses are generated leading to a plastically deformed zone. At high impact velocities, the
tensile stresses around the plastic zone begin to initiate microcracks. Two different types of
cracks are generated, lateral cracks, which are parallel to the target surface, and radial/median
cracks which run into the material.
Figure 1.20: Particle impact as indentation process102.
The mechanism of material removal depends on the properties of the target material. The process
of erosion involves an energy transfer between the particles and target: the kinetic energy stored
by the particles is partially or totally converted into strain energy in the target material. The
erosion process begins by repeated impacts leading to growth of lateral cracks, and eventually
Ravi K Chintapalli
57
Introductio
on to zirconia ceramics and
d sandblasting
g
coalesceence of these cracks give
g
rise to chipping off the material. On the other hand
d, the radiall
cracks will
w propagaate into the material.
1.4.2 Errosion in ceramics
c
As menntioned prevviously, erosion beginss with the teensile stressses around tthe plastic zone
z
of thee
impact generating the lateral cracks. Figgure 1.21 shows
s
the sequential
s
pprocess of erosion.
e
Too
nsidered. Inn
simulatee the erosioon process, often a sinngle particle impact onn a flat surrface is con
literaturre, this impaact process is comparedd to the sharp (Vickerss) indentatioon.
Figgure 1.21: Siingle particlee impact proccess103.
i
are often used
u
to dettermine eroosion. Marsshall et al977
The moodels basedd on sharp indentation
establishhed empiriccal equationns based onn the inden
ntation presssure and thhe lateral crrack length.
Additionally, several other annalytical models
m
are available
a
too predict errosion
104
. The laterall
crack mechanism
m
i mainly reesponsible for
is
f erosion in brittle materials
m
succh as glass.. In case off
homogeenous polyccrystalline ceramics,
c
thhe erosion rate
r
dependds on the grrain size an
nd porosity.
When the grain sizze is smalleer than the particles, erosion
e
occcurs in the fform of ejeecting grainn
clusters.
1.4.3 Efffect of pro
ocess param
meters
The maain sandblassting processs parameteers that inflluence the erosion
e
andd damage in
n the targett
materiall are very important
i
f comprehhensive und
for
derstanding of the sanndblasting process
p
andd
designinng the sandbblasting connditions. Thhe three maiin parameteers are brieflly explained
d below.
Ravi K Chintapalli
58
Introduction to zirconia ceramics and sandblasting
1.4.3.1 Impact Angle
In sandblasting, the angle of impact and the resultant erosion depends on target material. Two
main characteristic erosions are described in literature: brittle and ductile erosion. The
mechanism of material removal in brittle erosion is due to crack formation while in ductile
erosion it is due to cutting and ploughing. Figure 1.22 shows the erosion as a function of the
impact angle in brittle and ductile materials.
Figure 1.22: Effect of impact angle on erosion rate in ductile and brittle materials 98,105.
From the figure, it is clear that, the erosion in ductile materials is high at low impact angles.
Erosion rate peaks between 15 and 200 and eventually decreases with the increase in the angle.
In contrast, erosion in brittle materials increases with the impact angle and it shows a maximum
at 900. At low impact angles, the particles will plastically deform the material instead of
generating the cracks in brittle erosion. Additionally, generation of cracks can also be avoided by
decreasing particle size and velocity.
1.4.3.2 Particle velocity
The particle velocity also influences erosion; at high velocities the kinetic energy of the particle
is higher which results in high erosion rate and vice versa. Models have been established for
erosion loss with respect to the velocity. For instance, Tilly et al106 presented a simple power law
aVwhere, is erosion loss, a is material constant and exponent  is a constant equals to 2.
For brittle materials, Sheldon et al105 found that erosion rate is proportional to Vwhere  is
different for oblique and normal angles, and the value of for brittle materials lies between 3.5
and 6.
Ravi K Chintapalli
59
Introduction to zirconia ceramics and sandblasting
1.4.3.3 Particle size
The erosion is governed by the kinetic energy of the particles in brittle materials, as the particle
size is directly proportional to the erosion rate. Large sized particles remove large chunks of
material resulting in high erosion rate. Tilly et al106 also proposed a simple power law for glass
and similar brittle materials, =ad2 where a is constant and d is the diameter of the particle.
Recently, Ismail et al100 found a similar behavior in erosion rate of glass for large particles (150
and 220 m).
1.5 Sandblasting in zirconia
1.5.1 Introduction
The demand for ceramic-based restorations in dental practice has increased unabatedly because
of two factors: strong aesthetic demand and biocompatibility. Tetragonal Zirconia polycrystals
(TZP), especially 3 mol % Y2O3 - stabilized Zirconia (3Y-TZP), is widely employed as a
restorative material for dental crowns and pillars. Surface treatments are typically performed on
all ceramic dental restorations for better mechanical retention and proper fitting. Sandblasting is
one of the most commonly used surface treatments to improve the surface area by increasing the
roughness. Sandblasting is mainly used to improve the bonding between dental crown and luting
cement and also in some cases for better adhesion between crown and veneering porcelain107,108.
Air particle abrasion systems (sandblasting) are supposed to roughen zirconia increasing the
bonding area and modifying the ceramic superficial energy and wettability, thus facilitating the
formation of resin–ceramic micromechanical interlocks109.
The important aspect here is the material perspective, which implies that how 3Y-TZP ceramics
responds to sandblasting treatment. In literature, it is reported that, due to sandblasting, the
surface of the metastable tetragonal zirconia might be transformed, and also damaged, which
influences the mechanical properties and reliability of the material. While most current dental
ceramics exhibit excellent in situ chemical durability, the transformability of Y-TZP ceramics
under isothermal conditions warrants attention. When exposed to an aqueous environment
around 250°C over long periods of time, Y-TZP ceramics start transforming spontaneously into
the monoclinic structure. As mentioned earlier, this t-m transformation is athermal but its
progression into material is diffusion-controlled and is accompanied by extensive microcracking,
which ultimately leads to strength degradation.
Ravi K Chintapalli
60
Introduction to zirconia ceramics and sandblasting
But dentists do not seem to be concerned by ageing problems, presumably anticipating that
veneering and luting materials, separating the core from the oral environment and hard dental
tissues, provide for a durable protection of dental zirconia against hydrothermal degradation. In
addition, the temperature of exposure in the oral cavity is much lower than the temperature at
which LTD is usually reported. However, now it is known that LTD may also act at human body
temperature55–57 and it has been shown recently110 that commonly used luting cements absorb
water via dentine tubules, thereby exposing the zirconia core to moisture, which, in turn, may
lead to ageing problems over a shorter period of time than anticipated.
From a process perspective, a mechanical surface treatment such as sandblasting will induce
microstructural flaws apart from desired increase of wettability. The range of these flaws will
depend on sandblasting conditions. Severe process conditions will introduce larger defects while
mild conditions will induce smaller defects. As mentioned in the earlier section, the main
sandblasting conditions such as impact angle, particle size and particle velocity will influence the
extent of the damage induced. In literature, some amount of information on sandblasting in
zirconia is available, though the information regarding the effect of different sandblasting
conditions is limited. As such, this is the main focus of this thesis; the following section presents
a brief literature review on effects of sandblasting in zirconia dental ceramics.
1.5.2 State of the art
This section will be further divided into the following topics on Y-TZP ceramics with respect to
sandblasting.
-
Surface roughness
-
Phase transformation
-
Strength
-
Fatigue behavior
-
Damage induced
-
Hydrothermal degradation after sandblasting
-
Bond strength
1.5.3 Surface roughness
The assessment of surface roughness is important for any surface treated material, which will be
subjected to contact loads in service. In case of dental crowns, higher surface roughness is
Ravi K Chintapalli
61
Introduction to zirconia ceramics and sandblasting
achieved by sandblasting for the reasons mentioned above. Surface roughness is most commonly
represented as Ra in literature, which is universally, recognized parameter. Ra is the roughness
derived from the arithmetic mean of the absolute deviations of the roughness profile from the
mean line. According to the theory of sandblasting mentioned previously, the roughness is
generated by removal of chunks of material as an erosive wear process. Therefore particle size
and velocity have direct influence over the roughness; larger particles produce higher surface
roughness by removing larger chunks of material. However, the roughness produced by
sandblasting also depends on the previous state of dental crowns, for example, the crowns are
processed from solid blocks which are subjected to milling, grinding and polishing before
sandblasting.
The surface roughness data of dental ceramics from different authors is compiled and shown in
figure 1.23. In addition, the figure also shows the effect of sandblasting conditions such as
particle size and pressure on surface roughness. The data in the figure 1.23 indicates that
roughness increases with the particle size as can be seen in case of 50, 70 and 110m particles.
2
1.8
Control
Sandblasted
1.4
1.2
1
110-2B
110-4B
70-4B
70-4B
50-4B
0.4
50-3B
0.6
50-3B
0.8
25-4B
Surface roughness, Ra (m)
1.6
0.2
0
LAVA YIPS
IPS
LAVA Y- Y-TZP
TZP
Empress Empress
TZP
2
Materials
Nanozr
LAVA Y- VITA-YZ
TZP
Figure 1.23: Effect of sandblasting on the surface roughness of different dental materials for different
sandblasting conditions108,111,112. 25-4B indicates as the particle size 25 m and air pressure 4 bars
(results taken from various authors).
Authors often report weather the roughness differences before and after sandblasting have
statistical significance. Usually the condition before sandblasting is referred as control, which is
usually different in each study. In IPS Empress dental material, Albakry et al 111 did not find any
Ravi K Chintapalli
62
Introduction to zirconia ceramics and sandblasting
statistical difference in the roughness of control and sandblasted groups. The control group in
this case was not a polished condition but as received from the manufacturer; however, they
found significant differences in roughness between polished and sandblasted group. For LAVAY-TZP dental ceramics, Curtis et.al112 reported that the roughness is not statistically significant
after sandblasting; additionally, the effect of particle size on roughness was not observed. On the
other hand, Fischer et al108, Scherrer et al113 and Ban et al114 have found statistically significant
differences in the roughness of the control and sandblasted groups for particle sizes between 70
and 110 m, in different zirconia based dental ceramics.
However, it should be noted that, in each study the surface condition of the control group was
probably different as many authors report only the roughness for the control condition, but not
the method used to prepare the surface. The probable difference can arise from processing and
finishing procedures of manufacturers before supplying the samples for testing, since in some
cases the roughness of control samples was high. Therefore, it is very difficult to make
meaningful correlations between increase in surface roughness and the previous surface state,
and the data presented here is only based on the available information in the literature. From the
data from figure 1.23, it is clear that the effect of particle size and velocity on roughness is not
well established. As sandblasting is mostly used to increase the surface roughness, the knowledge about the
adequate roughness for good bond strength provides a basis for selecting the process parameters.
The effectiveness of increase in roughness for bond strength will be presented based on the data
available in the literature later in this section. This will probably give insights into the degree of
roughness that can provide good bond strength.
1.5.4 Phase transformation
3Y-TZP ceramics are a class of materials, which undergo t-m phase transformation when
subjected to externally applied stresses by grinding, impact, and fracture33. Sandblasting in
zirconia will induce a spontaneous t-m phase transformation by the impact of particles34,115,116.
As explained previously, the t-m transformation may generate compressive stresses. A thin
compressive stress layer at the surface may increase the resistance to fracture from flaws induced
by sandblasting.
Figure 1.24 shows the X-ray diffraction patterns of polished, grounded and sandblasted DCZircon dental ceramic. The phase transformation can be seen as appearance of monoclinic peak
Ravi K Chintapalli
63
Introduction to zirconia ceramics and sandblasting
(m) at 2equal to28.20. Additionally, broadening of (t) peak and reversal tetragonal doublet for
2between 34 and 36º was also observed113,116. The interchange of t doublet is referred as
ferroelastic domain switching. This phenomenon is widely discussed in zirconia subjected to
surface grinding and is already explained in section 1.5.2. Swain and Hannik117 attribute the
reversal of t doublet peaks to the reversible m-t transformation during grinding, activated by high
local temperatures or by stresses. However, Kosmac et al116 reported that sandblasting is less
effective to raise local temperatures for the reversible m-t transformation to occur and the peak
reversal in the sandblasted zirconia is probably due to the distortion of crystal structure.
Figure 1.24: XRD patterns of DC-Zirkon dental ceramic ground and sandblasted115. T- tetragonal phase,
M-monoclinic phase.
Some dental materials such as EVEREST KV and LAVA zirconia have cubic phase up to 28wt% which is referred as cubic 1 phase
112,113
. The presence of cubic phase is attributed to the
“grain-boundary segregation-induced phase transformation mechanism” in which the cubicphase regions start to form in grain boundaries and/or triple junctions in which Y3+ ions
segregate in grain interiors adjacent to the grain boundaries118. Scherrer et al113 suggested that, in
EVEREST KV the cubic 1 phase broadening is an indication of emergence of new cubic 2
phase, which was also observed by Curtis et al112 in LAVA Y-TZP.
In relation to the phase transformation induced by sandblasting, there are two other aspects in
which the studies have been focused: i) amount of phase transformation with respect to the
treatment applied, ii) depth of the transformation zone into the surface. Curtis et al112 mentioned
that the occurrence of the phase transformation was influenced by the severity of the
Ravi K Chintapalli
64
Introduction to zirconia ceramics and sandblasting
sandblasting treatment employed. By increasing the sandblasting mean particle size they found
an increased amount of transformed phase in LAVA Y-TZP dental ceramics. The amount of
monoclinic phase determined by XRD is usually calculated in the literature by using two
different equations: i) equation given by Garvie and Nicholson119 for the monoclinic phase
fraction Xm, ii) equation given by Toraya et al120 for the monoclinic volume fraction Vm,. On the
other hand, monoclinic volume fraction can be also determined from the measurements by
Raman spectroscopy by using the equation of Katagiri et al121.
There are many indications about the increase in monoclinic content with the sandblasting
particle size, which is in agreement with the observation of Curtis et al112. For instance,
sandblasting conditions using a particle size in between 30 and 110 m have resulted in
monoclinic content in between 5 to 15%112–114,116,122,123. Similarly, Sato et al124 pointed out that
sandblasting with 110m-sized SiC particles produced larger monoclinic content than with 70
m-sized alumina particles.
For surface treated Y-TZP materials, Kosmac et al92 proposed a model to estimate the depth of
the transformed layer based on the measurements of XRD. In this method the thickness of the
transformed zone is calculated using the amount of monoclinic phase. Additionally, it is assumed
that all the grains in this zone were transformed to monoclinic phase. The equation for
transformed zone depth (TZD) is given as
2
1
1
1.17
Where  = 150 is the angle of reflection and  is the absorption coefficient and Xm the
monoclinic fraction obtained from the XRD analysis on the basis of Equation of Garvie and
Nicholson119. Using this model in their work, sandblasting with 110 m particle size will induce
transformed zone depths for 3Y-TZP between 0.30 (fine grain size) and 0.33 m (coarse grain
size). However, they emphasized that the values obtained from the above model are
conservative, and be used for comparing results from different surface treatments.
Sato et al81 investigated the depth of the transformed zone by Raman spectroscopy and found the
presence of the monoclinic phase towards the interior and decreasing with depth. They
emphasized that the thickness of the transformed layer mainly depends on the size of
sandblasting particles; however, the homogeneity of the transformed layer is not known.
Ravi K Chintapalli
65
Introduction to zirconia ceramics and sandblasting
Another important aspect to note here is that usually after machining zirconia crowns and
bridges are heat treated between 800 and 950 oC for veneering porcelain to get the final aesthetic
appearance34,116. This treatment is generally preformed after sandblasting the frameworks. Ban et
al34,81 and Kosmac et al116 has shown that heat treatment after sandblasting will restore the
tetragonal phase in the material. Figure 1.25 shows monoclinic content with heat treatment
temperature for different materials (Y-TZP and CZA or NANOZR) that are sandblasted either
with alumina or SiC. It is clear that increasing the heat treatment temperature from 500 to 1200
o
C strongly decreased the monoclinic phase to almost zero.
81
Figure 1.25:Changes in monoclinic ZrO2 with heat treatment temperature .
Another effect of the heat treatment on the phase transformation is that the reversal of the
tetragonal doublet (002) and (200) is preserved and the broadened (111) peak is narrowed back
in the XRD patterns as seen in figure 1.26 115,116 for 2between 34º and 36º.
Figure 1.26: X-ray diffraction patterns of DC-Zirkon after heat treatment115.
Ravi K Chintapalli
66
Introduction to zirconia ceramics and sandblasting
1.5.5 Strength
In zirconia based dental ceramics, mechanical treatments such as sandblasting, grinding, or
machining may have strong influence over their mechanical properties such as strength and
fatigue resistance etc. Zirconia dental restorations are subjected to cyclic contact loads during
mastication. Therefore the strength and fatigue resistance of the material are critical parameters
that influence the long-term performance of the restorations. The first question that arises when
dealing with sandblasting and the nature of the induced damage is, whether sandblasting affects
the strength. Additionally, the severity of the sandblasting should be considered for proper
designing of the process parameters to achieve the desired result (high surface area) with
minimal adverse effects. However, the effect of the sandblasting conditions on the strength of
dental restorations is still not well known. Therefore an attempt is made here to put the results of
several authors together for different sandblasting conditions used with respect to the particle
size to understand the effect of this treatment on flexural strength.
Figure 1.27:Changes in strength of different dental materials before and after sandblasting. (Results
taken from different authors)111,112,116,122,125.
With respect to figure 1.27, for particle sizes up to 50 m the strength of the materials did not
practically change after sandblasting, But for particle size between 70 and 125 m the strength
increased considerably with a few exceptions. Wang et al87 found in CAD/CAM zirconia
frameworks, that the strength increased with 50 m particle and decreased with 120 m particle
size. Ban34 and Sato et al81 reported that strength increases after sandblasting, and attributed this
increase to the transformation from tetragonal to monoclinic which in turn generate compressive
Ravi K Chintapalli
67
Introduction to zirconia ceramics and sandblasting
stresses on the surface. Additionally, the biaxial flexure strengths increased with increase in
monoclinic ZrO2 content. In this sense, particle sizes up to 50 m are not effective to induce a
compressive stress zone compared to the larger particles (70-125 m). Similarly Kosmac et
al92,126,127 also found that strength increased after sandblasting but with lower values of Weibull
modulus as compared to control group.
However, Curtis et al112 and Karakoca et al123 did not found significant change in the mean biaxial flexure strength of zirconia based Lava and Cercon dental ceramics after alumina abrasion
with 25, 50 and 110 m particle sizes compared to untreated group. With reference to other
dental materials, Albakry et al111 found no significant differences in mean strength of
sandblasted (particle size 50 m) and control groups for pressable reinforced glass ceramics (IPS
Empress). By contrast, Guazzato et al125 reported that, in glass-infiltrated alumina-reinforced
dental ceramics, the strength decreased significantly after sandblasting (particle size 110 m)
compared to polished group. The differences in other sandblasting parameters such as working
distance and sandblasting time in addition to microstructural differences in the materials studied
is another possible reason for the contrasting results found by different authors.
With respect to the above mentioned findings, it can be concluded that the effect of sandblasting
depends on the microstructural properties of the material receiving the treatment and the
sandblasting particle size. In this sense, it is well known that the strength of all-ceramic materials
is directly related to the size, population, and distribution of structural defects and flaws. Surface
flaws are also more detrimental to the strength because these flaws directly act as stress
concentration sites, which magnify the applied stresses according to the severity of the surface
modification technique128.
Sandblasting results in damage and distortion of the surface and introduces defects in addition to
the phase transformation in zirconia ceramics. Kosmac et al92 have explained the increase in
strength in zirconia ceramics by the hypothesis that, the size of sandblasted surface defects are
smaller than the thickness of the surface compressive layer. Moreover, radial cracks which are
usually the critical defects were found to be absent after sandblasting, and only lateral cracks
were found in the microscopic examination (figure 1.28)129. Similar conclusions were achieved
by Andriotelli130 and emphasized that, surface flaws are contained within the compression zone
and are not detrimental for the strength.
Ravi K Chintapalli
68
Introduction to zirconia ceramics and sandblasting
Figure 1.28:SEM micrograph of a polished interface perpendicular to the sandblasted surface of a finegrained Y-TZP129.
Kosmac et al92 has made an approximate estimation of the critical defect size as 15.2 and 10.9
m for as sintered and sandblasted fine grain zirconia, and 20.8 and 11.9 m for as sintered and
sandblasted coarse grain zirconia respectively. For the estimation of the defect size they assumed
the existence of penny shaped cracks. It is clear that, the defect size is larger in coarse grain
materials, while the defect size is smaller in the sandblasted material compared to as sintered.
Sato et al81 estimated the compressive residual stresses based on the equation 1.18 proposed by
Swain131. Where ΔV is volume expansion due to phase transformation, Vm is the monoclinic
content, E is the elastic modulus, and  is the Poisson’s ratio. The estimated theoretical residual
stresses are found to be in the range of 160-400 MPa for 3Y-TZP, whereas 900-1935 MPa for
NANOZR, both sandblasted with 70 and 110 m particles.
∆ .
3 1
.
1.18
From the results of Kosmac et al92and Sato et al81, the critical defect size and the transformation
zone depth of sandblasted 3Y-TZP were found to be similar (approximately 10 m as in
NANOZR as found by Sato et al81). Therefore, it can be suggested that the critical defect is
contained in the region of residual stress. But the depth of the transformation zone may change
with along the surface.
Many authors also studied the effect of heat treatment on strength after sandblasting. The mean
strength of the zirconia increased after sandblasted (different types) and decreased after heat
Ravi K Chintapalli
69
Introduction to zirconia ceramics and sandblasting
treatment around 1000 ºC
34,81,115,116
, nevertheless the strength is similar to the non sandblasted
specimens. This decrease in strength in sandblasted specimens is explained by Ban34 as the
reverse transformation from monoclinic to tetragonal by heat treatment, which reduces the
compressive stress on the surface, resulting in decrease of strength. But, Wang et al87reported
that the strength of sandblasted zirconia did not change significantly after heat treatment (910 ºC
for 1 min in vacuum), as they found that there is no contribution of the residual stress to the
strength after sandblasting.
1.5.6 Fatigue behaviour
Fatigue is a very important mechanism of fracture in materials, which needs to be addressed with
respect to the long-term performance. Fatigue in ceramics has been less studied than in metals
due to absence of crack tip plasticity, since for many years it was thought that ceramics are
immune to cyclic fatigue. But in new tougher class of ceramics such as zirconia, which exhibit
R-curve behavior, the presence of fatigue behavior has been well documented. For dental
materials, resistance to fracture under cyclic stresses is an important criterion for selection of
materials. Surface treatments, such as sandblasting and grinding can influence the long-term
strength and reliability of dental ceramics under sustained and cyclic loading.
Kosmac et al126 reported the mechanical fatigue of sintered and sandblasted (110 m particle
size) Y-TZP under cyclic stresses induced by loads between 50 and 850 N at a frequency of 15
Hz for 106 cycles in air and artificial saliva. After fatigue in air, the survival strength of the as
sintered group (1070 MPa) was less than the strength of the sandblasted group (1250 MPa). The
survival strengths of both groups decreased after fatigue in artificial saliva, although the
sandblasted group had slightly higher strength. The survival rate under cyclic stresses of the as
sintered group was 64 and 50 % in air and artificial saliva respectively, whereas the sandblasted
groups reported 100% survival rate in both media.
Scherrer et al113 reported a 15 to 30% increase in the fatigue limits of different zirconia based
dental frameworks in water after sandblasting (30 m particle size). Figure 1.29 shows the
fatigue data for different dental Y-TZP ceramics. The fatigue strength of the control group was
primarily controlled by the catastrophic propagation of machining flaws such as critical blank
pressing flaws. The improved fatigue strength of the sandblasted group was attributed to the
presence of compressive stresses on the tensile surface which resulted from changes in the
crystalline structure (t-m) of the Y-TZP ceramics.
Ravi K Chintapalli
70
Introduction to zirconia ceramics and sandblasting
Figure 1.29:S–N fatigue data for Lava (LV) (3M Espe), Lava colored (LVB) (3M Espe), Everest ZS (KV)
(KaVo), Zeno (ZW) (Wieland), Fatigue limits at 106 cycles at 10Hz in water113.
In contrast to the above findings, Zhang et al91,132 reported that fatigue resistance and lifetime of
Prozyr dental ceramics decreased after sandblasting (figure 1.30) under dynamic and cyclic
fatigue conditions. Fatigue resistance after sandblasting decreased about 10% under dynamic
loading and about 30% under cyclic loading. The reduction in the strength by cyclic loading
compared to dynamic loading is attributed to enhanced and mechanically driven flaw extension
under cyclic loading.
Figure 1.30: Maximum tensile stress in ceramic layer versus effective time to radial fracture tR for a) aspolished b) sandblasted surfaces of Y-TZP and alumina. Solid lines are data fits in accordance with slow
crack growth relations91.
The results presented above are contradictory, as in some cases fatigue strength increased while
in other cases it decreased after sandblasting. Sandblasting induced flaws may have an important
role in controlling the fatigue resistance if they are larger than natural defects. According to
Zhang et al91, the concerning factor is the underlying nature of the sandblasting induced flaws;
Ravi K Chintapalli
71
Introduction to zirconia ceramics and sandblasting
and concluded that sandblasting flaws will have the nature of true microcracks, since they found
decrease in elastic modulus after sandblasting.
However, no correlation between sandblasting conditions and fatigue resistance can be
established; as the positive effect of sandblasting is often attributed to the surface compressive
stresses while the negative effect is attributed to the larger defects. Therefore, the comprehensive
understanding of the nature of the damage and defects will provide a basis for establishing a
relationship between sandblasting conditions and resultant properties.
1.5.7 Sandblasting damage
It is obvious that the damage induced by sandblasting could have adverse consequences on the
performance and reliability of the Y-TZP dental ceramics; as mentioned earlier, that properties
decreased in some cases after sandblasting. The previous studies indicate that sandblasting
induces surface defects, but the nature and character of these defects are not fully understood.
Fundamentally sandblasting generates two types of defects i) lateral cracks and ii) radial or
median cracks. It is suggested, in Y-TZP based dental ceramics these defects are counteracted by
the compressive stresses associated with the t-m phase transformation.
The influence of the compressive stresses will depend on their magnitude and on their
distribution in the depth direction as well as on the size of defects either induced by sandblasting
or already present in the starting material. The effect of compressive stress zone will be smaller
if a crack is longer than if it is completely embedded in this zone. Attempts to relate the size of
the defects and the stressed zone to the severity of the sandblasting treatment have not been
fruitful. The reason in part may be due to a lack of precise information on defect population and
precise knowledge of the compressive stress field after sandblasting. For instance, Kosmac et
al126 reported a defect size about 10 m while Zhang et al91found damage zone size as  4 m
after sandblasting as shown in figure 1.31, which cloud be due to difference in sandblasting
conditions and the material microstructure.
Scanning electron microscopy examinations by Kosmac et al127 on the specimens sandblasted
reveals lateral crack chipping (figure 1.28), which is the most prevalent mechanism of damage.
Similarly Guazatto et al115 also reported that the impact of sand particles on the surface of YTZP caused significant damage due to extensive erosive wear and lateral cracks.
Ravi K Chintapalli
72
Introduction to zirconia ceramics and sandblasting
Figure 1.31: SEM Micrograph showing partial top and cross-section view of sandblast damage (50 m
Al2O3 particles) in Y-TZP91.
In another work on glass infiltrated alumina, Guazatto et al125 found that sandblasted group is the
weakest among others due to the severe damage by the impact. The possible reason for this
adverse effect of sandblasting is the absence of strong residual stresses associated with phase
transformation unlike in zirconia ceramics.
Zhang et al91 emphasized that examinations by SEM may not resolve any individual
microcracks, especially in microstructures with submicrometer grain sizes. Nevertheless,
modulus reductions deduced from nanoindentation measurements within the damage zones
provide supportive evidence for their existence, as reduction in elastic modulus is related to the
presence of micro cracks23,133.
1.5.8 Hydrothermal degradation after sandblasting
Although 3Y-TZP ceramics have excellent mechanical properties there is a serious limitation for
their use in humid environment, where they are prone to relatively low-temperature aging. This
effect was first observed by Kobayashi et al6 at temperatures close to 250 0C in air. Aging
triggers surface phase transformation from tetragonal to monoclinic phase and rapidly
deteriorates mechanical properties. At lower temperatures, that is, between 250 0C and room
temperature, ageing also occurs but at much slower rates of transformation as already explained
in section 1.7.
Also many authors93,94,134,135 reported that hydrothermal degradation occurs in different dental YTZP ceramics, and highlighted the fact that it is a concerning factor for long-term sustainability
of restorations in the oral cavity. But it should be noted that Y-TZP ceramics used as dental
inlays, on lays, crowns and bridges go through different surface treatments before final usage.
These treatments will change the kinetics of aging, for example Deville et al136 found that rough
Ravi K Chintapalli
73
Introduction to zirconia ceramics and sandblasting
polishing, which generated compressive stresses in Y-TZP ceramics, increased the resistance to
degradation. Therefore surface state plays major role in controlling degradation. In this sense, as
sandblasting generates a rough surface with a compressive stress layer, it is expected that
sandblasted Y-TZP should be in principle more resistant to hydrothermal degradation than in
polished condition. Limited work has been done on this front, and the effect of sandblasting on
hydrothermal aging is not fully explored.
Kosmac et al116,126 studied hydrothermal degradation of 3Y-TZP in artificial saliva at 1340 C for
24 hours after sandblasting. They found the monoclinic volume fraction of sandblasted samples
was less compared to as sintered samples after hydrothermal degradation. Considering
monoclinic content as the degradation parameter, sandblasted group was slightly more resistant
to degradation compared to as sintered group. They concluded that, the partitioned tetragonal
zirconia grains and pre-existing monoclinic zirconia in the sandblasted surface impede the
propagation of the diffusion-controlled transformation during subsequent hydrothermal
degradation.
Moreover, the bi-axial strength of the sandblasted zirconia after hydrothermal degradation did
not change significantly compared to pristine samples with coarse grains; whereas, strength
decreased in fine grain materials116,126. It is then suggested that coarse grain sandblasted
materials are more resistant than fine grain materials to degradation in comparison with pristine
materials. No further suggestion was given by the author to explain this phenomenon, as this is
contrary to the normal belief that Y-TZP with grain sizes smaller than 0.3 m are more resistant
to hydrothermal degradation.
Recently, Munoz et al137 have also shown that 3Y-TZP subjected to grinding does not suffer
hydrothermal degradation at 1310 C in water vapor for 96 hours. This resistance to degradation
has been attributed to the formation of a very thin surface layer of tetragonal nanometric grains
in the highly deformed surface due to recrystallization. The size of these grains is smaller than
the critical size of the nucleus for phase transformation in humid environment. In addition to
this, 3Y-TZP ground and annealed at 12000C for 1 hour also shows excellent resistance to
hydrothermal degradation. This is attributed to the texture of the tetragonal phase although the
typical microstructure (phase and grain size) of the as sintered zirconia is recovered after
annealing treatment.
Ravi K Chintapalli
74
Introduction to zirconia ceramics and sandblasting
It should also be noted here that, during grinding the deformation and the temperature in the
surface are very high117 and they both could promote recrystallization. Whereas, it is reported
that the local temperatures during sandblasting are somewhat lower compared116 to grinding, and
presence of recrystallization has not been detected.
1.5.9 Bond strength
Zirconia crowns are bonded with veneering porcelain on the outer side and with the luting
cement on the inner side. The strength of the bond in both cases plays crucial role in fixing the
fabricated restorations and its reliability in long-term performance108. Usually, the bond strength
depends on several factors like strength of the chemical bonds, mechanical interlocking, type and
concentration of defects at the interface, wetting properties, and the degree of compressive stress
in the veneering layer due to the difference in the coefficients of thermal expansion between
zirconia and veneering ceramics.
On the other hand, adhesive cementation is preferable for cementation of zirconium oxide
restorations with traditional luting agents to ensure better retention and marginal adaptation138.
Surface pre-treatments such as sandblasting have been used for the reliable adhesion of ceramics.
Several authors have reported the durability of the bond strengths with different surface
treatments and different selection of cements34,107,108,138,139. As bonding methods with different
types of cements is beyond the context of this thesis, a brief overview of the effectiveness of
sandblasting treatment on the bond strength is presented here.
Kern et al107 reported sandblasting of zirconia resulted in significant initial bond strength to a
conventional dual-curing BisGMA resin composite. But the strength later decreased to almost
zero after long-term storage in water and thermal cycling. This is attributed to the fact that,
minimal undercuts produced by sandblasting of zirconia; due to this, the physical and chemical
bonds between conventional BisGMA resin composite and zirconia were not water resistant.
Similarly Kokubo et al139 found that, sandblasting was an effective method to improve the
durability of the adhesion, but this was linked to specific hard cement (Hy bond temporary
cement). Similarly, Castillo de Oyague et al138 found that micro tensile bond strengths were
significantly influenced by the luting cement but not by the surface treatment.
Ban34 reported that, the shear bond strength of different zirconia based dental ceramics did not
change significantly after sandblasting with 70 and 125 m sized particles in spite of the fact that
Ravi K Chintapalli
75
Introduction to zirconia ceramics and sandblasting
surface roughness in case of 125 m particles was twice higher than in 75 m particles.
Similarly, Fischer et al108 found that increasing surface roughness by sandblasting of zirconia did
not enhance shear bond strength, while the polished surface of zirconia provided strong bonding
to the veneering ceramics due to the formation of chemical bonds between both materials during
firing.
From the above findings, it can be summarized that sandblasting proves to be an effective
treatment in combination with specific cements; whereas degree of surface roughness as well as
the severity of the sandblasting treatment is less relevant in increasing the bond strength, while
selection of luting cement seems to be more relevant factor when bonding to zirconium oxide
ceramics than the surface treatment. However, sandblasting can enhance the bond strength with
proper selection of the luting cement.
1.6 Summary
The ever-increasing demand for high quality, highly aesthetic, high performance, and sustainable
materials in restorative dentistry is replacing metal-metal and metal-ceramic systems with all
ceramics materials. Dentists are overwhelmed with the introduction of zirconia ceramics with
their unmatched superiority over other ceramics or previously used porcelain based ceramics.
However, some failures in zirconia may be promoted by low temperature degradation. In the
past, manufacturers of dental implants have overlooked this fact. In recent years, new evidence
of hydrothermal degradation at body temperatures has opened new research interest in how this
effect may impact the performance of zirconia dental ceramics
Additionally, as mentioned earlier the restorative frame works undergo several surface finishing
procedures before clinical application. As a consequence of these treatments, material at the
surface is affected up to a few microns of depth. The influence of these treatments on the
material performance must be well studied to avoid clinical failures. As this study is aimed at
investigating the changes on the surface microstructure of zirconia, zirconia ceramics are
thoroughly introduced and several aspects such as stabilization, t-m transformation, and low
temperature degradation are discussed in detail within the limitations of this thesis.
The effects of sandblasting on dental zirconia ceramics were also thoroughly introduced and
explained in terms of changes in microstructure and severity of the treatment whenever possible.
Although increasing the wettability of the surface is the prime application of this treatment,
Ravi K Chintapalli
76
Introduction to zirconia ceramics and sandblasting
choice of the luting cement plays significant role in enhancing the bond strength. Moreover
roughness levels after sandblasting depend on the pre-surface state. On the other hand, the effect
of sandblasting on the bi-axial strength and fatigue resistance is significant. While some studies
have shown increase in strength and fatigue resistance after sandblasting, others have shown
decrease in both properties of dental zirconia ceramics. The increase in properties is attributed to
the t-m phase transformation associated compressive stresses, whereas the decrease in properties
is due to the large defects. Moreover, the improved properties after sandblasting were again
decreased with annealing, which is due to the relaxation of the stresses and reverse m-t
transformation.
In addition to the contradicting results in literature, the information about depth of transformed
zone induced by sandblasting and its homogeneity is not known. Moreover, no clear correlation
has been established between the severity of sandblasting and the properties. With respect to
sandblasting damage, only lateral crack mechanism was evident by SEM, which does not have a
strong effect on degradation of strength. In other cases strength degradation was attributed to the
nature of the damage, and presence of microcracks.
As the presence of microcracks can affect the hardness and elastic modulus, it is also important
to assess these properties after sandblasting. With the objective of establishing a correlation
between sandblasting conditions and material properties, this thesis will be focused to study the
severity of the sandblasting conditions on strength degradation, and the nature of the damage
along with other consequences on 3Y-TZP ceramics.
Ravi K Chintapalli
77
Introduction to zirconia ceramics and sandblasting
1.7 Objectives
The complex nature of 3Y-TZP ceramics under stress and humid environment have arisen the
need to understand the effect of sandblasting conditions on phase transformation, strength and
aging behavior. This study enables to increase the life of the zirconia based restorations by
optimizing the sandblasting process while minimizing the unfavorable effects such as damage
and transformation.
1.7.1 Specific objectives

Develop new nanocrystalline zirconia materials and study the resistance of hydrothermal
degradation.

Processing nanocomposites with improved fracture toughness and resistance to
degradation.

Study the effect of sandblasting particle size and pressure on the surface roughness, phase
transformation and bi-axial strength.

Characterize the sandblasting damage in the near surface zone.

Investigate the phase transformation depth and residual stresses induced by sandblasting.

Study the changes in elastic modulus and hardness after sandblasting and after
hydrothermal degradation in the near surface zone.

Study the influence of heat treatments after sandblasting on phase transformation and biaxial strength.
Ravi K Chintapalli
78
Introduction to zirconia ceramics and sandblasting
1.8 References
1. Hench LL. Bioceramics. Journal of the American Ceramic Society. 1998;81(7):1705-1728.
2. Piconi C, Maccauro G. Zirconia as A Ceramic Biomaterial. Biomaterials. 1999;20(1):1-25.
3. Green J, Hannink RHJ, Swain MV. Transformation Toughening Of Ceramics. Crc Press Inc. Florida.
1989:1989.
4. Virkar A, Matsumoto R. Ferroelastic Domain Switching as A Toughening Mechanism In Tetragonal
Zirconia. Journal of the American Ceramic Society. 1986;69:224-226.
5. Gupta TK, Bechtold JH, Kuznicki R, Cadoff LH, Rossing BR. Stabilization of Tetragonal Phase in
Polycrystalline Zirconia. Journal of Materials Science. 1977;12:2421-2426.
6. Kobayashi K, Kuwajima H, Masaki T. Phase Change and Mechanical Properties of Zro2-Y2O3 Solid
Electrolyte After Ageing. Solid State Ionics. 1980;3(4):489-493.
7. Chevalier J, Gremillard L, Deville S. Low-Temperature Degradation of Zirconia and Implications for
Biomedical Implants. Annual Review of Materials Research. 2007;37:1-32.
8. Lawson S. Environmental Degradation of Zirconia Ceramics. Journal of the European Ceramic
Society. 1995;15:485-502.
9. Hannink RHJ, Kelly PM, Muddle BC. Transformation Toughening in ZrO2-Containing Ceramics.
Journal of the American Ceramic Society. 2000;83(3):461-487.
10. Garvie RC. Zirconium Dioxide and Some of Its Binary Systems. Academia Press, New York.
1970:117-166.
11. Scott HG. Phase Relationships in the Zirconia–Yttria System. Journal of Materials Research.
1975;10:1527-1535.
12. Chevalier J, Gremillard L, Virkar AV, Clarke DR. The Tetragonal-Monoclinic Transformation in
Zirconia: Lessons Learned and Future Trends. Journal of the American Ceramic Society.
2009;92(9):1901-1920.
13. Kelly PM, Rose LRF. The Martensitic Transformation In Ceramics — Its Role In Transformation
Toughening. Progress in Materials Science. 2002;47(5):463-557.
14. Kriven WM, Fraser WL, Kennedy SW. The Martensite Crystallography of Tetragonal Zirconia. in H.
Heuer and L. W. Hobbs (Eds.), Advances in Ceramics, Science and Technology of Zirconia. American
Ceramic Society,Columbus, OH. 1981;3:82-97.
15. Hayakawa M, Adachi K, Oka M. Crystallographic Analysis of the Monoclinic Herringbone Structure
in an Arc-Melted ZrO2-2 mol% Y2O3 Alloy. Acta Metallurgica et Materialia. 1990;38(9):1753-1759.
16. Hayakawa M, Kuntani N, Oka M. Structural Study on The Tetragonal To Monoclinic Transformation
In Arc-Melted Zro2-2mol.%Y2O3-I Experimental Observations. Acta Metallurgica. 1989;37:2223-2228.
17. Hayakawa M, Kuntani N, Oka M. Structural Study on the Tetragonal To Monoclinic Transformation
In Arc-Melted Zro2-2mol%Y2O3-II Quantitative Analysis. Acta Metallurgica. 1989;37:2229-2235.
18. Green DJ, Hannik RHJ, Swain MV. Transformation Toughening of Ceramics. Boca Raton, Florida:
CRC press; 1989.
Ravi K Chintapalli
79
Introduction to zirconia ceramics and sandblasting
19. Wayman CM. Introduction to Crystallography of Martensitic Transformations. New York: Mac
Millan; 1965.
20. Kelly PM, Ball CJ. Crystallography of Stress-Induced Martensitic Transformations in Partially
Stabilized Zirconia. Journal of American Ceramic Society. 1986;69:259-264.
21. Muñoz-Tabares JA, Jiménez-Piqué E, Anglada M. Subsurface Evaluation of Hydrothermal
Degradation of Zirconia. Acta Materialia. 2011;59(2):473-484.
22. Muñoz-Tabares J., Jiménez-Piqué E, Reyes-Gasga J, Anglada M. Microstructural Changes In Ground
3Y-TZP And Their Effect On Mechanical Properties. Acta Materialia. 2011;59(17):6670-6683.
23. Marro FG, Chintapalli R, Hvizdos P, et al. Study of Near Surface Changes In Yttria-Doped
Tetragonal Zirconia After Low Temperature Degradation. International Journal of Materials Research.
2009;100(1):92-96.
24. Lange FF. Transformation Toughening 1 Size Effects Associated with the Thermodynamics of
Constrained Transformations. Journal of Materials Research. 1982;17:225-234.
25. Olson G. Deformation, Processing and Structure. Proceedings of 1982 Conference. American Society
for Metals, Metals Park, OH. 1984:391-424.
26. Patel JR, Cohen M. Criterion for the Action of Applied Stress in the Martensitic Transformation. Acta
Metallurgica. 1953;1(9):531-538.
27. Eshelby J. The Determination of the Elastic Field of an Ellipsoidal Inclusion, and Related Problems.
Proceedings of the Royal Society of London A. 1957;241:376-396.
28. Garvie RC, Swain MV. Thermodynamics of The Tetragonal To Monoclinic Phase Transformation In
Constrained Zirconia Microscrystals, Part I In Absence of An Applied Stress Field. Journal of Materials
Science. 1985;20:1193-1200.
29. Lange FF. Transformation Toughening Part 2 Contribution to Fracture Toughness. Journal of
Materials Science. 1982;17:235-239.
30. Evans AG, Burlingame N, Drory M, Kriven WM. Martensitic Transformations in Zirconia-Particle
Size Effects and Toughening. Acta Metallurgica. 1981;29:447.
31. Kato M, Monzen R, Mori T. A Stress-Induced Martensitic Transformation of Spherical Iron Particles
in A Cu—Fe Alloy. Acta Metallurgica. 1978;26:605-613.
32. Bokros JC, Parker ER. The Mechanism of The Martensite Burst Transformation In Fe-Ni Single
Crystals. Acta Metallurgica. 1963;11:1291-1301.
33. Garvie RC, Hannink RH, Pascoe RT. Ceramic Steel? Nature. 1975;258:703-704.
34. Ban S. Reliability and Properties of Core Materials for All-Ceramic Dental Restorations. Japanese
Dental Science Review. 2008;44:3-21.
35. McMeeking RM, Evans A. Mechanics of Transformation Toughening In Brittle Materials. Journal of
American Ceramic Society. 1982;65:242-246.
36. Evans AG, Heuer AH. Review—Transformation Toughening in Ceramics Martensitic
Transformations in Crack-Tip Stress Fields. Journal of American Ceramic Society. 1980;63:241-248.
Ravi K Chintapalli
80
Introduction to zirconia ceramics and sandblasting
37. Ritchie RO. Mechanisms of Fatigue Crack Propagation in Metals, Ceramics, and Composites Role of
Crack Tip Shielding. Materials Science and Engineering: A. 1988;103:15-28.
38. Evans A. Perspective on the Development of High-Toughness Ceramics. Journal of American
Ceramic Society. 1990;73(2):187-206.
39. Chevalier J, Olagnon C, Fantozzi G. Subcritical Crack Propagation in 3Y-TZP Ceramics: Static and
Cyclic Fatigue. Journal of the American Ceramic Society. 1999;82(11):3129-3138.
40. Virkar AV, Matsumoto R. Toughening Mechanism in Tetragonal Zirconia Polycrystalline (TZP)
Ceramics. published in Advances in Ceramics, Science and Technology of Zirconia 111. Edited by S .
Somiya and M. Yoshimura. American Ceramic Society, Westerville, OH.
41. Li B-S, Cherng J-S, Bowman KJ, Chen I-W. Domain Switching as a Toughening Mechanism in
Tetragonal Zirconia. Journal of the American Ceramic Society. 1988;71(7):C362-C364.
42. Foitzik, Statwald-Klenke M, Rühle M. Ferroelasticity of t -ZrO2. Z. Metallkd. 1993;84(6):397-404.
43. Virkar AV. Role of Ferroelasticity in Toughening of Zirconia Ceramics. Key Engineering Materials,
Zirconia Engineering Ceramics: Old Challenges New Ideas, Edited by E. H. Kisi. TransTech, UteikonZurich. 1998;153-154:1-36.
44. Knott F. Fundamentals of Fracture Mechanics. Butterworths, London; 1976:126.
45. Heuer AH. Transformation Toughening in ZrO2-Containing Ceramics. Journal of the American
Ceramic Society. 1987;70(10):689-698.
46. Steinbrech RW. Toughening Mechanisms in Ceramic Materials. Journal of European Ceramic
Scociety. 1992;10:131-142.
47. Hoffman MJ, Dauskardt RH, Mai Y-W, Ritchie RO. A Review of the Mechanics and Mechanisms of
Cyclic Fatigue Crack Propagation in Transformation Toughened Zirconia Ceramics. Science and
Technology of Zirconia. Technomic, Lancaster, PA. 1993:321-328.
48. Eichler J, Hoffman M, Eisele U, Rödel J. R-Curve Behaviour of 2Y-TZP With Submicron Grain Size.
Journal of the European Ceramic Society. 2006;26(16):3575-3582.
49. Casellas D, Alcala J, Llanes L, Anglada M. Fracture Variability and R-Curve Behavior in YttriaStabilized Zirconia Ceramics. Journal of Materials Science. 2001;36:3011 - 3025.
50. Fargas G, Casellas D, Llanes L, Anglada M. Thermal Shock Resistance of Yttria Stabilized Zirconia
With Palmqvist Indentation Cracks. Journal of European Ceramic Scociety. 2003;23:107-114.
51. Munz D. What Can We Learn from R-Curve Measurements? Journal of the American Ceramic
Society. 2007;90(1):1-15.
52. Chevalier J, Cales B, Drouin JM. Low-Temperature Aging of Y-TZP Ceramics. Journal of the
American Ceramic Society. 1999;82(8):2150-2154.
53. Guo X. Hydrothermal Degradation Mechanism of Tetragonal Zirconia. Journal of Materials Science.
2001;36(1):3737 - 3744.
54. Lange FF, Dunlop GL, Davis BI. Degradation During Aging of Transformation-Toughened Zr02Y203 Materials at 250°C. Journal of the American Ceramic Society. 1986;69(3):237-240.
Ravi K Chintapalli
81
Introduction to zirconia ceramics and sandblasting
55. Catledge SA, Cook M, Vohra YK, et al. Surface Crystalline Phases and Nanoindentation Hardness of
Explanted Zirconia Femoral Heads. Journal of Materials Science. Materials in Medicine. 2003;14:863867.
56. Santos EM, Vohra S, Catledge SA, et al. Examination of Surface and Material Properties of
Explanted Zirconia Femoral Heads. Journal of Arthroplasty. 2004;19:30-34.
57. Haraguchi K, Sugano N, Nishii T, et al. Phase Transformation of a Zirconia Ceramic Head after Total
Hip Arthroplasty. Journal of Bone and Joint surgery. 2001;83:996-1000.
58. Yoshimura M. Phase Stability of Zirconia. American Ceramic Society Bulletin. 1988;67(12):1988.
59. Sato T, Shimada M. Transformation of Yttria-Doped Tetragonal Zr02 Polycrystals by Annealing in
Water. Journal of the American Ceramic Society. 1985;68(6):356-359.
60. Guo X. On The Degradation Of Zirconia Ceramics During Low-Temperature Annealing In Water Or
Water Vapor. Journal of Physics and Chemistry of Solids. 1999;60(4):539-546.
61. Guo X. Low Temperature Stability of Cubic Zirconia. Physica Status Solidi (a). 2000;177(1):191201.
62. Guo X, Schober T. Water Incorporation in Tetragonal Zirconia. Journal of the American Ceramic
Society. 2004;87(4):746-748.
63. Chevalier J. What Future for Zirconia as a Biomaterial? Biomaterials. 2006;27:535-543.
64. Tsubakino H, Hamamoto M, Nozato R. Tetragonal to Monoclinic Phase Transformation during
Thermal Cycling and Isothermal Ageing in Partially Stabilized Zirconia. Journal of Materials Science.
1991;26:5521-5226.
65. Tsubakino H, Sonoda K, Nozata R. Martensite Transformation Behavior during Isothermal Ageing in
Partially Stabilized Zirconia with or without Alumina Addition. Journal of Materials Science.
1993;12:196-198.
66. Zhu W, Lei T, Zhou Y. Time Dependent Tetragonal to Monoclinic Transition in Hot Pressed Zirconia
Stabilized with 2 mol% Yttria. Materials Chemistry and Physics. 1993;34:317-320.
67. Suresh A, Mayo MJ, Porter WD, Rawn CJ. Crystallite and Grain-Size-Dependent Phase
Transformations in Yttria-Doped Zirconia. Journal of American Ceramic Society. 2003;86(2):360-362.
68. Djurado E, Dessemond L, Roux C. Phase Stability of Nanostructured Tetragonal Zirconia
Polycrystals Versus Temperature And Water Vapor. Solid State Ionics. 2000;136-137:1249-1254.
69. Chintapalli RK, Mestra A, García Marro F, et al. Stability of Nanocrystalline Spark Plasma Sintered
3Y-TZP. Materials. 2010;3(2):800-814.
70. Paul A, Vaidhyanathan B, Binner JGP. Hydrothermal Aging Behavior of Nanocrystalline Y-TZP
Ceramics. Journal of American Ceramic Society. 2011;94(2):2146-2152.
71. Jue JF, Chen J, Virkar AV. Low-Temperature Aging of t’ Zirconia The Role of Microstructure on
Phase Stability. Journal of American Ceramic Society. 1991;74(8):1811-1820.
72. Chen IW. Model of Transformation Toughening in Brittle Materials. Journal of American Ceramic
Society. 1991;74:2564-2572.
Ravi K Chintapalli
82
Introduction to zirconia ceramics and sandblasting
73. Schmauder S, Schubert H. Significance of Internal Stresses for the Martensitic Transformation in
Yttria-Stabilized Tetragonal Zirconia Polycrystals During Degradation. Journal of the American Ceramic
Society. 1986;69(7):534-540.
74. Conrad HJ, Seong W-jin, Pesun IJ. Current Ceramic Materials and Systems With Clinical
Recommendations- a Systematic Review. The Journal of Prosthetic Dentistry. 2007;98(5):389-404.
75. Devigus A, Lombardi G. Shading Vita YZ Substructures Influence on Value and Chroma, Part I.
International Journal of Computerized Dentistry. 2004;7:293-301.
76. Raigrodski AJ. Contemporary Materials and Technologies For All-Ceramic Fixed Partial Dentures A
Review of The Literature. Journal of Prosthetic Dentistry. 2004;92:557-562.
77. Denry I, Kelly JR. State of The Art of Zirconia for Dental Applications. Dental Materials.
2008;24(3):299-307.
78. Denry I, Holloway JA. Ceramics for Dental Applications A Review. Materials. 2010;3:351-368.
79. Guazzato M, Albakry M, Ringer SP, Swain MV. Strength, Fracture Toughness and Microstructure Of
A Selection of All-Ceramic Materials Part II Zirconia-Based Dental Ceramics. Dental Materials.
2004;20(5):449-56.
80. Chevalier J, Deville S, Münch E, Jullian R, Lair F. Critical Effect of Cubic Phase on Aging In 3mol%
Yttria-Stabilized Zirconia Ceramics for Hip Replacement Prosthesis. Biomaterials. 2004;25(24):5539-45.
81. Sato H, Yamada K, Pezzotti G, Nawa M, Ban S. Mechanical Properties of Dental Zirconia Ceramics
Changed with Sandblasting and Heat Treatment. Dental Materials Journal. 2008;27(3):408-414.
82. Marro FG, Mestra A, Anglada M. Contact Damage in a Ce-TZP/Al2O3 Nanocomposite. Journal of
the European Ceramic Society. 2011;31:2189-2197.
83. Garmendia N, Santacruz I, Moreno R, Obieta I. Zirconia-MWCNT Nanocomposites for Biomedical
Applications Obtained by Colloidal Processing. Journal of Materials Science. Materials in Medicine.
2010;21(5):1445-51.
84. Garmendia N, Grandjean S, Chevalier J, et al. Zirconia–Multiwall Carbon Nanotubes Dense NanoComposites with an Unusual Balance Between Crack and Ageing Resistance. Journal of the European
Ceramic Society. 2011;31(6):1009-1014.
85. Mazaheri M, Mari D, Schaller R, Bonnefont G, Fantozzi G. Processing of Yttria Stabilized Zirconia
Reinforced With Multi-Walled Carbon Nanotubes with Attractive Mechanical Properties. Journal of the
European Ceramic Society. 2011;31(14):2691-2698.
86. Cho J, Inam F, Reece MJ, et al. Carbon Nanotubes: Do They Toughen Brittle Matrices? Journal of
Materials Science. 2011;46(14):4770-4779.
87. Wang H, Aboushelib MN, Feilzer AJ. Strength Influencing Variables on CAD/CAM Zirconia
Frameworks. Dental Materials. 2008;24(5):633-638.
88. 3M. Scientific Facts-Lava Zirconia. Dental Products,3M ESPE.
89. Manicone PF, Iommetti PR, Raffaelli L. An Overview of Zirconia Ceramics Basic Properties and
Clinical Applications. Journal of Dentistry. 2007;35:819-826.
Ravi K Chintapalli
83
Introduction to zirconia ceramics and sandblasting
90. Rieger W, Stefan K, Weber W. Zirconia Ceramics for Dental Applications. Spectrum Dialogue.
2008;(March):2-11.
91. Zhang Y, Lawn BR, Rekow ED, Thompson VP. Effect of Sandblasting on the Long-Term
Performance of Dental Ceramics. Journal of Biomedical Materials Research. Part B, Applied
biomaterials. 2004;71B:381-386.
92. Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. The Effect of Surface Grinding and
Sandblasting On Flexural Strength and Reliability of Y-TZP Zirconia Ceramic. Dental Materials.
1999;15:426-433.
93. Kohorst P, Borchers L, Strempel J, et al. Low-Temperature Degradation of Different Zirconia
Ceramics for Dental Applications. Acta Biomaterialia. 2011;In press.
94. Cattani-Lorente M, Scherrer SS, Ammann P, Jobin M, Wiskott HWA. Low Temperature Degradation
of a Y-TZP Dental Ceramic. Acta Biomaterialia. 2011;7(2):858-65.
95. Tilghman BC. Sandblasting. 1870.
96. Lawn BR, Evans AG, Marshall DB. Elastic/Plastic Indentation Damage in Ceramics: The
Median/Radial Crack System. Journal of the American Ceramic Society. 1980;63:574-581.
97. Marshall DB, Lawn BR, Evans AG. Elastic/Plastic Damage in Ceramics: The Lateral Crack System.
Journal of the American Ceramic Society. 1982;65:561-566.
98. Finnie I. Some Reflections on the Past and Future of Erosion. Wear. 1995;186-187:1-10.
99. Slikkerveer PJ, Bouten PCP, in t Veld FH, Scholten H. Erosion and Damage by Sharp Particles.
Wear. 1998;217:237-250.
100. Ismail J, Zaïri F, Naït-Abdelaziz M, Bouzid S, Azari Z. Experimental and Numerical Investigations
on Erosion Damage in Glass by Impact of Small-Sized Particles. Wear. 2011;271(5-6):817-826.
101. Slikkerveer PJ, Bouten PCP, De Haas FCM. High Quality Mechanical Etching of Brittle Materials
by Powder Blasting. Sensors And Actuators A:Physical. 2000;85:296-303.
102. Pourreuix C. Strength of Glass Eroded by Round Particles. Nat.Lab. Unclassified Report, Philips
Electronics. 1997.
103. Lawn BR. Fracture of Brittle Solids. Second Edition, Cambridge University press, Cambridge,
United Kingdom. 1993.
104. Wensink H. Fabrication of Microstructures by Powder Blasting. PhD Thesis, University of Twente,
Enschede The Netherlands. 2002.
105. Sheldon GL, Finnie I. On The Ductile Behaviour of Nominally Brittle Materials during Erosive
Cutting. Trans. ASME. 1966;88B:387-392.
106. Tilly GP, Sage W. The Interaction of Particle and Material Behavior in erosion Processes. Wear.
1970;16:447-465.
107. Kern M, Wegner SM. Bonding to Zirconia Ceramic: Adhesion Methods and Their Durability.
Dental Materials. 1998;14(1):64-71.
Ravi K Chintapalli
84
Introduction to zirconia ceramics and sandblasting
108. Fischer J, Grohmann P, Stawarczyk B. Effect of Zirconia Surface Treatments on the Shear Strength
of Zirconia / Veneering Ceramic Composites. Dental Materials Journal. 2008;27(3):448-454.
109. Wolfart M, Lehmann F, Wolfart S, Kern M. Durability of The Resin Bond Strength to Zirconia
Ceramic After Using Different Surface Conditioning Methods. Dental Materials. 2007;23(1):45-50.
110. Jevnikar P, Serša I, Sepe A, Jarh O, Funduk N. Effect of Surface Coating On Water Migration into
Resin-Modified Glass Ionomer Cements- A Magnetic Resonance Micro-Imaging Study. Magnetic
Resonance in Medicine. 2000;44:686-691.
111. Albakry M, Guazzato M, Swain MV. Effect of Sandblasting, Grinding, Polishing And Glazing on
The Flexural Strength of Two Pressable All-Ceramic Dental Materials. Journal of Dentistry.
2004;32(2):91-99.
112. Curtis AR, Wright AJ, Fleming GJP. The Influence of Surface Modification Techniques on The
Performance of a Y-TZP Dental Ceramic. Journal of Dentistry. 2006;34(3):195-206.
113. Scherrer SS, Cattani-Lorente M, Vittecoq E, et al. Fatigue Behaviour In Water of Y-TZP Zirconia
Ceramics After Abrasion With 30 m Silica-Coated Alumina Particles. Dental Materials.
2011;27(2):e28-e42.
114. Ban S, Sato H, Suehiro Y, Nakanishi H, Nawa M. Effect of Sintering Condition, Sandblasting and
Heat Treatment on Biaxial Flexure Strength of Zirconia. Key Engineering Materials. 2008;361-363:779782.
115. Guazzato M, Quach L, Albakry M, Swain MV. Influence of Surface and Heat Treatments on the
Flexural Strength of Y-TZP Dental Ceramic. Journal of Dentistry. 2005;33(1):9-18.
116. Kosmac T, Oblak C, Marion L. The Effects of Dental Grinding and Sandblasting on Ageing and
Fatigue Behavior of Dental Zirconia (Y-TZP) Ceramics. Journal of the European Ceramic Society.
2008;28:1085-1090.
117. Swain MV, Hannink RHJ. Metastability of the Mertensitic Transformation in a 12mol% Ceria–
Zirconia Alloy. II. Grinding Studies. Journal of the American Ceramic Society. 1989;72:1358-1364.
118. Matsui K, Horikoshi H, Ohmichi N, et al. Cubic-Formation and Grain-Growth Mechanisms in
Tetragonal Zirconia Polycrystal. Journal of the American Ceramic Society. 2003;86(8):1401-1408.
119. Garvie RC, Nicholson PS. Phase Analysis in Zirconia Systems. Journal of the American Ceramic
Society. 1972;55:303-305.
120. Toraya H, Yoshimura M, Somiya S. Calibration Curve for Quantitative Analysis of the Monoclinic
Tetragonal ZrO2 System by X-Rays Diffraction. Journal of the American Ceramic Society. 1984;67:119121.
121. Katagiri G, Ishida H, Ishitani A, Masaki T. Direct Determination By A Raman Microprobe of The
Transformation Zone Size In Y2O3 Containing Tetragonal ZrO2 Polycrystals. Advances in Ceramics.
1988;24:537-544.
122. Sato H, Yamashita D, Ban S. Relation between Biaxial Flexure Strength and Phase Transformation
of Zirconia with Surface Treatments. Journal of the Ceramic Society of Japan. 2008;116(1):28-30.
Ravi K Chintapalli
85
Introduction to zirconia ceramics and sandblasting
123. Karakoca S, Yilmaz H. Influence of Surface Treatments on Surface Roughness, Phase
Transformation, and Biaxial Flexural Strength of Y-TZP Ceramics. Journal of Biomedical Materials
Research. Part B, Applied Biomaterials. 2009;91(2):930-7.
124. Sato H, Ban S, Nawa M, Suehiro Y, Nakanishi H. Effect of Grinding, Sandblasting and Heat
Treatment on the Phase Transformation of Zirconia Surface. Key Engineering Materials. 2007;330332:1263-1266.
125. Guazzato M, Albakry M, Quach L, Swain MV. Influence of Grinding, Sandblasting, Polishing and
Heat Treatment on the Flexural Strength of A Glass-Infiltrated Alumina-Reinforced Dental Ceramic.
Biomaterials. 2004;25(11):2153-2160.
126. Kosmac T, Dakskobler A, Oblak C, Jevnikar P. The Strength and Hydrothermal Stability of Y-TZP
Ceramics for Dental Applications. International Journal of Applied Ceramic Technology. 2007;4(2):164174.
127. Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. Strength and Reliability of Surface Treated
Y-TZP Dental Ceramics. Journal of Biomedical Materials Research. 2000;53:304-313.
128. Curtis AR, Wright AJ, Fleming GJP. The Influence of Simulated Masticatory Loading Regimes on
the Bi-Axial Flexure Strength and Reliability of a Y-TZP Dental Ceramic. Journal of Dentistry.
2006;34:317-325.
129. Kosmac T, Oblak C, Jevnikar P. The Fracture and Fatigue of Surface-Treated Tetragonal Zirconia
(Y-Tzp) Dental Ceramics. Materilas and Technology. 2007;41(5):237-241.
130. Andreiotelli M. Survival Rate And Fracture Resistance of Zirconium Dioxide Implants After
Exposure to the Artificial Mouth an In-Vitro Study. PhD Thesis,Universität Freiburg. 2006.
131. Swain MV. Grinding-Induced Tempering of Ceramics Containing Metastable Tetragonal Zirconia.
Journal of Materials Science. 1980;15:1577-1579.
132. Zhang Y, Lawn BR, Malament K a, Van Thompson P, Rekow ED. Damage Accumulation and
Fatigue Life of Particle-Abraded Ceramics. The International Journal of Prosthodontics. 2005;19(5):4428.
133. Gaillard Y, Jimenez-Pique E, Soldera F, Mucklich F, Anglada M. Quantification of Hydrothermal
Degradation in Zirconia by Nanoindentation. Acta Materialia. 2008;56:4206-4216.
134. Ardlin BI. Transformation-Toughened Zirconia for Dental Inlays, Crowns and Bridges Chemical
Stability and Effect of Low-Temperature Aging On Flexural Strength and Surface Structure. Dental
Materials. 2002;18:590-595.
135. Lughi V, Sergo V. Low Temperature Degradation –Aging of Zirconia: A Critical Review of the
Relevant Aspects in Dentistry. Dental Materials. 2010;26(8):807-20.
136. Deville S, Chevalier J, Gremillard L. Influence of Surface Finish and Residual Stresses on the
Ageing Sensitivity of Biomedical Grade Zirconia. Biomaterials. 2006;27(10):2186-92.
137. Muñoz-Tabares JA, Anglada M. Hydrothermal Degradation of Ground 3Y-TZP. Journal of the
European Ceramic Society. 2012;32(2):325-333.
Ravi K Chintapalli
86
Introduction to zirconia ceramics and sandblasting
138. Castillo de oyague R, Monticelli F, Toledano M, et al. Influence of Surface Treatments And Resin
Cement Selection on Bonding to Densely-Sintered Zirconium-Oxide Ceramic. Dental Materials.
2009;5:172-179.
139. Kokubo Y, Kano T, Tsumita M, et al. Retention of Zirconia Copings on Zirconia Implant Abutments
Cemented With Provisional Luting Agents. Journal of Oral Rehabilitation. 2010;37(2):48-53.
Ravi K Chintapalli
87
Introduction to zirconia ceramics and sandblasting
Ravi K Chintapalli
88
Experimental methods
Chapter
2
Experimental methods
Ravi K Chintapalli
89
Experime
ental methodss
2.1 Spe
ecimen prreparation
n
2.1.1 Ba
ase materiial
The com
mmercially available yttria
y
stabilized Zircon
nia powder (TZ-3YSB
B-E, Tosoh Co, Japan))
with 3%
% molar of yttria
y
contennt was chossen for this study.
s
The crystallite
c
size and partticle size off
the pow
wder are 30 and
a 600 nm
m respectively. The pow
wder granulee size is 60 m.
Table 2.1 Chemicaal compositio
on of 3Y-TZ
ZP powder
Elements
Composition in wt %
Y2O3
4.9
95-5.35
Al2O3
0.1
15-.035
SiO2
Maax 0.02
Fe2O3
Maax 0.01
NaO2
Maax 0.04
Monolitthic materiaals were prrocessed byy using con
nventional cold
c
isostattic pressing
g (CIP) andd
furnace sintering and also by
b spark pllasma sinteering (SPS)). In additiion to the monolithicc
materialls, nanocom
mposites weere processeed by adding
g multiwall carbon nannotubes to the 3Y-TZP
P
powder and were sintered
s
by SPS. The materials
m
weere divided in three grooups accord
ding to theirr
processiing methodd and composition. The detailed processing
p
methods arre described
d below forr
each maaterial.
2.1.2 Material
M
AS--300
The 3Y-TZP powdder is isostattically comppacted in a cylindrical polymer moould in a co
old isostaticc
press too form a greeen compacct as shown schematicaally in figurre 2.1.The ccompaction pressure iss
200 MP
Pa and the dwell
d
time iss 5 minutes..
Figure 2.1: Cold isostatic presssing of 3Y-TZ
ZP powder.
Ravi K Chintapalli
90
Experimental methods
The carefully removed green bodies are sintered at 1450 0C for 2 hours at heating rate of 3
0
C/min. The sintering curve is shown in figure 2.2. During the heating ramp, the temperature was
maintained at 700 0C for one hour to burn out the binder compounds. Finally, after holding the
temperature at 1450 0C for 2 hours, the bars were cooled down at a rate of 30C/min until room
temperature. The sintered cylinders are then cut into discs of 2 mm thick with a diamond wheel.
The typical measurements of the sample are 10 mm in diameter and 2 mm in thickness. This
material is labeled as “AS-300”throughout the thesis.
Figure 2.2: Sintering curve for material AS300.
2.1.3 Nanocrystalline zirconia
Nanocrystalline zirconia materials were processed by spark plasma sintering. The powder was
sintered using spark plasma sintering (SPS FCT HP D25I, FCT system GmBh) in Nanoforce Ltd
laboratory, London, UK. The basic SPS principle is that pulsed DC current directly passes
through a graphite die and into the ceramic powder generating heat internally. The heating is
fairly rapid due to the internal heat generation and can reach upto 700 0C/min.
Zirconia powder was pre-heated at 700 0C for 1 hour before sintering to burn of the binder
compounds. A graphite die with internal diameter of 20 mm is employed for preparing the
samples. The lower punch is introduced into the die and thin circular graphite sheets were then
placed inside the die. The appropriate amount of preheated zirconia powder was then introduced
in the die and closed with upper punch. The die punch set up was pressed in hand press to ensure
that both the punches are properly fitted in the die. The setup was then placed into the sintering
chamber of SPS furnace.
Ravi K Chintapalli
91
Experimental methods
Figure 2.3: Spark plasma sintering (SPS FCT HP D25I, FCT system GmBh)Nanoforce Ltd UK, a) SPS
furnace, b)die punch setup c)die during sintering temperature.
The powder was sintered at various temperatures between 1100 °C and 1600 °C for 5 minutes at
a pressure of 100 MPa. The general sintering cycle (for a temperature of 1175 0C) is shown in
figure 2.4. A pressure of 100MPa was applied when the temperature reached 850 ºC during the
ramp up; this pressure was maintained until the end of the sintering dwell time and then released.
The vacuum chamber was maintained at 1 Torr; heating current and voltage were 1300 A and
5.8 V, respectively. The final samples had a disc shape with a 4 mm thickness and 20 mm
diameter. As mentioned above, the materials were sintered with different temperatures keeping
other parameters constant. Sintering temperatures with respective material labels are given in
table 2.1.
Figure 2.4: SPS curve for nanocrystalline zirconia.
Ravi K Chintapalli
92
Experimental methods
Table 2.2 Sintering temperatures and material labels
Sintering
temperature (0C)
1100
Material label
S65
1150
S90
1175
S120
1600
S800
2.1.4 Nanocomposites
Zirconia multiwall carbon nanotube composites are processed by mixing 3Y-TZP powder with
0.5, 1 and 2 vol. % of multiwall carbon nanotubes (MWCNT) (Graphistrength C100, Arkema,
France, outer diameter 10-15 nm, inner diameter 2-6 nm and length 0.1-10 m) to prepare the
composite materials.
Multiwall carbon nanotubes were initially dispersed in N, N dimethylforamamide (DMF) using
ultrasonication for 2 hours. Zirconia powder was preheated at 750 ºC for 1 hr and then added to
the ultrasonicated MWCNT-DMF solution. The solution was milled in a planetary ball mill for 4
hrs using different sized zirconia balls (10-5-3 mm) at a velocity of 300 rpm. The milled slurry
was collected into a stainless steel tray and dried on a hot plate at 70 ºC for 24 hr. Finally the
powder was sieved using a mesh of 250 m.
Figure 2.5: SPS curve for Zirconia-MWCNT nanocomposites.
Ravi K Chintapalli
93
Experimental methods
The prepared powder was introduced into a graphite die and sintered by spark plasma sinteringat
maximum temperature of 1350 0C with a dwell time of 5 min, with heating and cooling rates of
1000C/min. A pressure of 50 MPa was applied when the temperature reached 650 ºC during the
ramp up; this pressure was maintained during the sintering cycle until cooling down below 800
ºC as shown in the figure 2.5. The vacuum chamber was maintained at 1 Torr; heating current
and voltage were 1300 A and 5.8 V. The final samples had a disc shape with a 2 mm thickness
and 20 mm diameter. For comparison monolithic zirconia samples were also sintered at the same
conditions. Table 2.3 shows the material labels with respective quantities of MWCNTs.
Table 2.3 Concentration of MWCNT and material labels
Concentration
of MWCNT
(% volume)
0
0.5
Material label
3YTZP-0CNT
3YTZP-0.5CNT
1
3YTZP-1CNT
2
3YTZP-2CNT
2.1.5 Metallographic preparation
All the disc shaped sample surfaces were prepared in steps. Samples were subjected to mild
grinding against diamond grinding disc (Struers MD Piano 220) until a uniform plane was
achieved. In the second step samples were polished using a MD/DP-Plan grade white polishing
cloth with 30m diamond paste for removing the larger scratches produced during grinding. In
the subsequent steps samples were polished using MD/DP-Dac blue cloth with 30 and 3 m
diamond paste to remove the scratches completely. Finally samples were polished with MD/DP
Nap brown cloth with 0.03 m colloidal silica to achieve mirror like surface.
2.2 Microstructural characterizations
2.2.1 Confocal Microscope
Laser confocal microscope (Olympus, LEXT OLS 3100) was used for obtaining surface
topographic images for sandblasted samples and also for characterizing the Vickers indentations
in all materials. The microscope was used in both optical and confocal mode wherever
necessary.
Ravi K Chintapalli
94
Experimental methods
2.2.2 Scanning electron microscope (SEM)
SEM (JEOL JSM 6400) was used for initial microstructural characterization such as grain size in
all materials and for fractographic examinations of the samples broken under bi-axial strength
testing. Sample surfaces were prepared with a thin layer of gold coating before introducing the
samples in the SEM chamber to improve the electron conductivity for better imaging.
2.2.3 FIB-SEM
The principle of focus ion beam (FIB) system is shown in figure 2.6. Gallium ions (Ga+) are
primarily used for sputtering the surfaces. The Ga+ ion beam is focused onto the area of interest
on the surface, once the beam hits the surface a small amount of material is sputtered. The
returning ions leave the surface as secondary ions (i+ or i-) or neutral atoms (n0) and also
emitting as secondary electrons (e–). The primary beam is used to raster the surface while the
signals from the emitting ions are used to form the image. Usually high beam currents are used
for milling large areas quite rapidly, and low beam currents are used for polishing the milled
surface for imaging. FIB-SEM is dual system that comprises focus ion beam coupled with
scanning electron microscope for in-situ characterizations of sputtered or milled area.
Figure 2.6: Principle of focus ion beam technique.
In this thesis FIB trenches were prepared for characterizing the damage below spherical
nanoindentations of nanocrystalline zirconia and for investigation of the subsurface damage in
sandblasted samples. Silver is painted at the border of the sample to glue it to the stub. All
samples surfaces were prepared by sputtering a thin layer of gold-palladium. Focused Ion Beam
(FIB) cross sections were carried out in a Zeiss Neon 40 microscope. A platinum layer was
deposited before milling to protect the material. The surface was milled with Ga+ ions and
Ravi K Chintapalli
95
Experime
ental methodss
decreasiing currentss at 30 kV down
d
to 1000 pA. Trencches were milled
m
with currents off 10 nA andd
polishedd in steps off 500, 150 and
a 100 pA. Images weere taken wiith both elecctrons and ions.
i
FM
2.2.4 AF
Atomic force miccroscope (A
AFM Dimeension Veecco Inc) waas used to observe th
he residuall
sphericaal nanoindentation impprints in tappping mode.
2.3 Mechanical characteri
c
izations
2.3.1 De
ensity
Bulk deensity and apparent porosity
p
weere measureed by Archhimedes meethod (Metttler Toledoo
XS205)) and also according
a
too the ASTM
M standard C 20-001 for
f materialls with morre than 5%
%
porosityy.
2.3.2 Su
urface proffilometer
Surface roughnesss of polished and sandblasted
s
d samples was measured with a surfacee
profilom
meter (SUR
RFTEST) cooupled withh software SURFPAK.
S
The schem
matic illustraation of thee
SURFT
TEST profiloometer is shhown in figuure 2.7.The sensitivity parameter () is the cu
ut-off valuee
which depends
d
on the sampling length (length
(
of measuremen
m
nt) and corrresponds to a standardd
range of
o roughnesss. Thus foor measurinng Ra for polished
p
sam
mples c iss 0.08 mm
m (Ra rangee
betweenn 0.02-0.066 m, samppling lengthh 0.4 mm) and
a for sanndblasted saamples is 0.8
0 mm (Ra
range beetween 0.1--2 m, samppling lengthh 4 mm).
Figgure2.7: Scheeme of SURF
FTEST profillometer.
2.3.3 Ha
ardness & Fracture to
oughness
Hardnesss and factuure toughneess values were
w
measu
ured with 1 and 20 kgg Vickers in
ndentationss
2
respectiively. For estimating the fracturre toughnesss, the equuation propposed by Niihara
N
forr
Palmqvist cracks was
w used. Additionally
A
y, for nanoccomposites the
t equationn proposed
d by Anstis3
Ravi K Chintapalli
96
Experimental methods
for median cracks was used to compare the results with those in literature. Figure 2.8 shows the
scheme of Vickers indentation and its parameters with Palmqvist cracks.
Figure2.8: Vickers indentation scheme with Palmqvist crack system.
Vickers Hardness of the material is given by the following equation
1.854
2
2.1
Fracture toughness, KIC, was obtained by using the equation given by Niihara2et al
.
0.018 √
1
.
2.2
as well as by using the equation given by Anstis3et al
/
0.016
/
2.3
where c is the crack length and E is the elastic modulus, H is the hardness calculated from the
indentation load P and half diagonal a of the Vickers imprint as
2
Ravi K Chintapalli
2.4
97
Experime
ental methodss
2.3.4 Biiaxial stren
ngth
In this thesis biaxiial strength was measurred for only
y AS300 maaterial. The availability
y of numberr
of speciimens in caase of nanocrystalline zirconia an
nd nanocom
mposites matterials was limited forr
an adeqquate Weibuull analysis, therefore strength waas not meassured. The bbiaxial strength of thee
“AS3000” samples after polishhing and sandblasting
g was meassured usingg “Ball on three
t
Ball””
(B3B) configuratio
c
on. The scheematic illusttration and the test parrameters4 off the ball on
n three ballss
test is shhown in figgure 2.9.
Figgure 2.9: a) ball
b on three ball test conf
nfiguration b)) Finite elem
ment model off the three ba
alls test
assembly4.
Borger et al4 propoosed an equuation for caalculation of
o the strenggth based onn their resu
ults on theirr
investiggations of thhe stress distribution in the disc by
y finite elem
ment modelinng.
2.5
Where max is the maximum
m
t
tensile
stresss in the dissc, F is the applied loaad, t is the th
hickness off
the discc, and f is a dimensionle
d
ess parametter which is given by
,
,
1
1
2.6
R is raddius of the disc,
d
Rb is thhe radius of the ball and
d the supporrt radius Ra is given by
y (see figuree
2.9 b)
2√3
3
Ravi K Chintapalli
2.7
98
Experimental methods
C0 to C6 are the constants which depend on the passion ratio. The constant values are taken from
the work of Borger et al4 (=0.3 for zirconia) are shown in table 2.4.
Table 2.4 Calculated values for the constants4
C0
C1
C2
C3
C4
C5
C6
-17.346
20.774
622.62
-76.879
50.383
33.736
0.0613
The samples were tested in INSTRON 8511 universal testing machine with a load cell of 10 KN.
The tensile surface (polished or sandblasted) of the samples is positioned downward and the
sample is preloaded to -50N, and then the load is applied with a velocity of 20 N/s until the
fracture occurs.
2.3.5 Nanoindentation
Nanoindentation or instrumented indentation is a widely accepted method for determining the
elastic modulus and hardness of materials at small scales which was introduced by Oliver and
Pharr5. This method determines some mechanical properties directly from the indentation load
and displacement measurements. Berkovich indenter is the most commonly used tip for
determining the mechanical properties. A schematic representation of unloading processes and a
typical data set obtained with a Berkovich indenter is shown in figure 2.10. The measured
parameters are the applied load (P), the total displacement relative to the initial undeformed
surface (h), the contact depth (hc), the final depth (hf), and the contact stiffness, S dP/dh, which
is defined as the slope of the upper portion of the unloading curve during the initial stages of
unloading.
Figure 2.10: Schematic illustration of a) the unloading process showing parameters characterizing the
contact geometry b) indentation load–displacement data showing important measured parameters5.
Ravi K Chintapalli
99
Experimental methods
The values of hardness and Young’s modulus were calculated as a function of penetration depth
using the method of Oliver and Pharr5. In this method, the contact depth is estimated from the
load-displacement data using the following equation.
2.8
where Pmax is the peak indentation load and ε is a constant which depends on the indenter
geometry and is 0.75 for a Berkovich indenter5. From the measurements of load and
displacement data, the projected contact area, A, of the indentation impression is estimated by
evaluating the indenter shape function at the contact depth, hc, that is,A=f(hc).Once the contact
area is determined from the load displacement data, the hardness and effective elastic modulus,
Eeff, were calculated as follows
2.9
2
√
√ 2.10
where β depends on the indenter tip geometry and is equal to 1.034 for a Berkovich tip5 and Eeff
is the effective elastic modulus defined by
1
1
1
2.11
The effective elastic modulus takes into account the fact that elastic displacements occur in both
the specimen, with Young’s modulus E and Poisson’s ratio , and the indenter, with elastic
constants Ei and i.
In this thesis nanoindentations were made with an MTS Nanoindenter XP equipped with a CSM
module. Two different type of indenter were used: Berkovich and spherical indenters. The tests
were performed with a Berkovich diamond tip under a load of 630 mN with a constant
deformation rate of 0.05 s-1. On the other hand, a diamond spherical indenter tip with a nominal
radius of 25 m was used. Indentations were made to a maximum depth of 2000 nm and under a
Ravi K Chintapalli
100
Experimental methods
constant deformation rate of 0.05 s−1. Both indenter tips were initially calibrated with standard
fused silica. Arrays of 4 by 4 indentations were carried out in each material; about 1200 data
points (load, displacement, stiffness and time) were collected for each indentation.
2.3.6 Residual stresses
Sandblasting induces compressive residual stresses, which lead to an apparent fracture
toughness. Some information on compressive stress field can be obtained by measuring the
apparent fracture toughness. In this thesis residual stresses induced by sandblasting are estimated
in terms of change in apparent indentation fracture toughness by using a linear stress gradient
model proposed by Portu and Conoci6.
/
/
2.12
Where KIC is the fracture toughness of stress free material and KICA is the apparent fracture
toughness of material with residual stresses. cis the residual stress and d is the depth of the
residual stresses, C is the crack length (refer to figure 2.8). Residual stresses are estimated in
material AS300 for one sandblasting condition. Residual stresses are estimated as follows.

Niihara2 model was used to calculate the apparent and stress free fracture toughness.

Fracture toughness of the stress free material and apparent fracture toughness of a
sandblasted sample is measured by Vickers indentation of 150N

Apparent fracture toughness was measured in small intervals (few microns) of depths
from the sandblasted surface by polishing and removing material.

The process was repeated until the apparent fracture toughness and stress free fracture
toughness are equal.

The thickness of the material removed is calculated using an area function of the Vickers
indenter.
24.504

2.13
Where A is the area of the residual indentation imprint and hp is the depth of the
indentation. A Vickers indentation of 295 N is carried out and the area is measured after
each step to calculate the thickness of the material removed.

All the indentations are performed in the center of the sandblasted sample.
Ravi K Chintapalli
101
Experimental methods
2.4 Phase characterizations
2.4.1 X-Ray diffraction
The polished and sandblasted samples were analysed for phase composition by X-ray diffraction
with Cu-Kα radiation (Bruker AXS D8). A Bragg-Brentano configuration was used, in
mode. Spectra were obtained for 2between 24 and 780, with an increment of 0.020 and a
velocity of one second per step. Cu K with wave length =1.5418 Å was used as the radiation
source. Voltage and current was set to 40 kV and 40 mA respectively.
The phase analysis of the samples was done using the equation of Toraya et al7. By using this
equation the monoclinic and tetragonal phases are quantified. For the quantification of these
phases, 2between 24 and 370 is considered as it is mostly done in the literature.
1.311
1.311
〈
〈
〉
〈
〉
〈
〉
〉
〈
〉
2.14
Where, Im (111) and Im (111) are the monoclinic peak intensities at 2equal to 28.20 and 31.400,
respectively. Whereas It (111) is the tetragonal peak intensity at 31.10. Integrated peak areas
were obtained to calculate the monoclinic volume fraction.
2.4.2 Raman Spectroscopy
Micro Raman spectroscopy was used to quantify the local phase transformation around spherical
nanoindentations in nanocrystalline zirconia samples (SPS sintered zirconia), around Vickers
indentations in nanocomposites (3YTZP-MWCNT) and in the cross-sections of the sandblasted
samples.
A triple monochromator spectrometer (Horiba JobinYvon- LABRAM HR) with a coupled CCD
detector (liquid nitrogen cooled) was used to obtain Raman spectra on the residual indentation
imprints. An Ar-ion laser Innova 300 (Coherent Laser Group) was used as a light source and
excited at a wavelength of 514 nm. Raman spectra were collected at 100x magnification and the
spectrum integration time was 60 seconds, with the recorded spectra averaged over four
successive measurements.
In nanocrystalline zirconia, Raman spectra were collected from the centre and the border of the
spherical nanoindentations. Innanocomposites (3YTZP-MWCNT), Raman spectra were
Ravi K Chintapalli
102
Experimental methods
collected by mapping an area of 100 m2 around the crack tip with a lateral resolution of 2.5 m.
In case of sandblasted samples, Raman Spectra were collected by mapping an area of the
polished cross-sections with a lateral resolution of 2 m. The phase quantification was done by
using the equation proposed by Katagiri et al8.
0.5
2.2
0.5
2.15
Where, Im (180) and Im (190) are the characteristic monoclinic peaks and It (150) is the tetragonal
peak at respective Raman displacements. To determining the monoclinic fraction, the Raman
spectra were analyzed by using the integrated areas under tetragonal peak at 150 cm-1 and under
the monoclinic doublet peaks at 180 and 190 cm-1. These areas were measured by using the
linear background method and integrating the peak areas of all spectra9.
2.5 Sandblasting
Samples were sandblasted using BEGO, Easy Blast sandblaster (figure 2.11). A total of seven
sandblasting conditions were chosen with the combination of two particle sizes (110, 250 m)
and two air pressures (2 and 4 bars) and two impact angles (30 and 900). Sandblasting was done
for all the conditions with a constant standoff distance of 25 mm (the distance between sample
surface and nozzle tip) and time of 10 seconds. The nozzle tip diameter is 2.5 mm. The selected
conditions are shown in the table 2.5. Samples were cleaned in ethanol under ultrasonication for
15 minutes after sandblasting to remove any free particles over the surface.
Figure 2.11: a) BEGO, Easy Blast sandblaster, b)& c) sandblasting schemes in 90º and 30º.
Ravi K Chintapalli
103
Experimental methods
Table 2.5 Sandblasting nomenclature
Sandblasting (SB)
surface treatment
Description
110-2B-900
Sandblasted with particle size 110 m under pressure 2 bars at an angle of 900
110-4B-900
Sandblasted with particle size 110 m under pressure 4 bars at an angle of 900
250-2B-900
Sandblasted with particle size 250m under pressure 2 bars at an angle of 900
250-4B-900
Sandblasted with particle size 250m under pressure 4 bars at an angle of 900
110-2B-300
Sandblasted with particle size 110 m under pressure 2 bars at an angle of 300
250-2B-300
Sandblasted with particle size 250m under pressure 2 bars at an angle of 300
2.6 Low temperature degradation
Low temperature degradation (LTD) tests are carried out in a JP Selecta autoclave. The
equipment has a controlled environment chamber for 100% steam. The conditions for
accelerated aging are 131 ºC temperature and 2 bars of pressure. The duration of aging is varied
between60 and 200hours.
2.7 Grain size
Grain size is measured for all materials on thermally etched samples (thermal etching at 200 0C
below the sintering temperature for 1 hour in air). It is estimated in SEM micrographs using
Buehler OMNIMET 5.4software which uses grain size module.
2.8 Heat treatments
Heat treatments were performed on sandblasted samples in a tubular furnace (HOBERSAL ST18). The samples of AS300 were heat treated at temperatures ranging between 500-10000C for 1
hour in air after sandblasting. The nanocrystalline and nanocomposites samples were heat treated
at 1000ºC for 1 hour in air after sandblasting.
Ravi K Chintapalli
104
Experimental methods
2.9 References
1. ASTM. C 20-00 Standard Test Methods for Apparent Porosity, Water Absorption, Apparent Specific
Gravity, and Bulk Density of Burned Refractory Brick and Shapes by Boiling Water.
2. K. Niihara. A Fracture Mechanics Analysis of Indentation-Induced Palmqvist Crack in Ceramics.
Journal of Materials Science Letters. 1983;2(5):221-223.
3. Anstis GR, Chantikul P, Lawn BR, Marshall DBA. A Critical Evaluation of Indentation Techniques for
Measuring, Fracture Toughness I Direct Crack Measurements. Journal of American Ceramic Society.
1981;64:533-538.
4. Borger A, Supancic P, Danzer R. The Ball on Three Balls Test for Strength Testing of Brittle Discs:
Stress Distribution in the Disc. Journal of the European Ceramic Society. 2002;22:1425-1436.
5. Oliver WC, Pharr GM. Measurement of Hardness and Elastic Modulus by Instrumented Indentation:
Advances in Understanding and Refinements to Methodology. Journal of Materials Research.
2004;19(1):3-20.
6. Portu GD, Conoci S. Simplified Equation for Evaluating the Influence of Surface Residual Stresses on
the Toughness of Zirconia Ceramics. Journal of American Ceramic Society. 1997;80(12):3242-3244.
7. Toraya H, Yoshimura M, Somiya S. Calibration Curve for Quantitative Analysis of the Monoclinic
Tetragonal ZrO2 System by X-Rays Diffraction. Journal of the American Ceramic Society. 1984;67:119121.
8. Katagiri G, Ishida H, Ishitani A, Masaki T. Direct Determination by a Raman Microprobe of The
Transformation Zone Size in Y2O3 Containing Tetragonal ZrO2 Polycrystals. Advances in Ceramics.
1988;24:537-544.
9. Muñoz Tabares JA, Anglada M. Quantitative Analysis of Monoclinic Phase in 3Y-TZP by Raman
Spectroscopy. Journal of the American Ceramic Society. 2010;93(6):1790-1795.
Ravi K Chintapalli
105
Experimental methods
Ravi K Chintapalli
106
Properties and hydrothermal degradation
Chapter
3
Material properties and
hydrothermal degradation
Ravi K Chintapalli
107
Properties and hydrothermal degradation
3.1 Introduction
The final objective of this thesis partially deals with preparing and characterizing nanocrystalline
monolithic and composite materials which could be used as restorative materials for dentistry.
This requires materials with improved resistance to hydrothermal degradation and fracture
toughness. With this objective, monolithic and nanocomposites materials were prepared with
different microstructures and the processing details are already given chapter 2. Therefore this
chapter deals with the characterization of different materials prepared as explained in the
previous chapter. Microstructure, mechanical properties and hydrothermal degradation behavior
are characterized. Additionally, the knowledge of the material properties and hydrothermal
degradation behavior of pristine materials is required to observe the changes induced by
sandblasting treatments. Characterization is presented separately for monolithic and composite
materials.
3.2 Monolithic materials
3.2.1 Microstructure
Figure 3.1 shows the microstructures of all materials. S65 and S90 have some degree of porosity
(Refer to table 3.1) whereas AS300 and S120, S800 are fully dense. From figure 3.1, in S65
pores are interconnected and elongated, while in S90 they are closed and dispersed. It is clear
that pores become increasingly irregular in shape with the increase of porosity which
corresponds to the decrease in sintering temperature. This non-uniformity arises from the
sintering process which depends on dwell time, temperature, and pressure applied and release
conditions during spark plasma sintering. This leads to incomplete densification leaving the
inhomogeneous submicron undensified zones in the material.
Figure 3.1: SEM microstructures of thermally etched samples a) S65, b) S95, c) S120, d) AS300, e) S800.
Ravi K Chintapalli
108
Properties and hydrothermal degradation
Additionally, it can be observed that in S65, S90 and S120 grains appear to have round borders
whereas in AS300 and S800 the grains appear with sharp edges. The change in grain sizes is
associated with sintering temperature.
3.2.2 Properties of monolithic materials
Table 3.1 shows different properties of the specimens with respect to their sintering temperature.
Because of the small starting crystallite size and good sintering activity, moderate to fully dense
specimens are obtained using the SPS technique whereas a fully dense specimen is obtained with
conventional sintering (1450 C for 2 hours). Fully dense materials were also achieved by
increasing the sintering temperature in SPS. Grain growth is also evident as the sintering
temperature increases. Very fine grain specimens were obtained at low temperatures compared
to conventional sintering.
Table 3.1 Properties of monolithic materials
Material
S65
S90
S120
AS300
S800
Sintering temperature (0C)
1100
1150
1175
1450
1600
Bulk density (g/cm3)
4.79
5.49
5.98
6.08
6.05
Relative density (%)
78.52
90.01
98.03
99.60
99.14
Apparent porosity (%)
21.26
9.36
1.91
-
-
Mean grain size (nm)
65±20
90±30
120±20
300±10
800±90
Elastic modulus (GPa)a
121±5
200±7
230±3
231±3
223±2
Contact hardness (GPa)a
6.5±1.9
12.5±1.8
17.2±1.4
17.4±1.1
16.1±2.1
5.1±0.7
3.7±0.9
5.0±0.5
5.0±0.5
11.4±0.3
14.5±0.2
12.8±0.2
12.1±0.4
Fracture toughness (MPa√m)b 5.0±0.8
Hardness (HV1) (GPa)
6.2±0.5
a
Berkovichnanoindentation
Niihara equation
b
Indentation toughness and hardness are average values calculated from four indentations on each
sample. The material with high density and fine grain size (S120) has higher hardness but
relatively lower toughness as compared with larger grain size specimens and similar density
(S800 or AS300). On the other hand, low density specimens (S65 and S90) have low hardness,
but, surprisingly, higher indentation fracture toughness than dense S120. This aspect will be
discussed later in this section. Elastic modulus and contact hardness are similar in dense
specimens irrespective of the grain size, whereas these properties are found decreasing with
increasing porosity in porous materials.
Ravi K Chintapalli
109
Properties and hydrothermal degradation
3.2.3 Hydrothermal degradation (HD or LTD)& effect on properties
Figure 3.2 (a) shows the X-ray diffraction patterns of all samples after sintering and fine
polishing. The material is purely tetragonal and no monoclinic phase is found. Figure 3.2 (b)
displays the XRD patterns of samples after 60 hours of ageing in autoclave at 131 °C. It is
observed that the stability of the tetragonal phase is retained after hydrothermal ageing in
samples with fine grain size (S65, S90 and S120), as no monoclinic peaks were found. The
remarkable fact is that, in spite of having porosity, specimens with the smaller grain sizes (S65
and S90), which were sintered at the low temperatures, are resistant to LTD.
By contrast, the tetragonal phase is destabilized and transformed to monoclinic in samples with
larger grain size (AS300and S800). From figure 3.2 (b), monoclinic peaks can be observed at 2θ
values of 28.1°, 31.4° and 55.8°. The amount of m-phase was calculated using equation 2.13 (see
chapter 2). Large amount of monoclinic volume fraction, 65.4 and 71.4%, were detected in
specimens AS300 and S800, respectively.
t
a)
t
tt
t
Intensity
tt
t
AS300 t t
m
b)
t
t
m
t
m
tt
AS300
S800
S800
S120
S120
S90
S90
S65
S65
t
24 28 32 36 40 44 48 52 56 60 64 68 72 76 24 28 32 36 40 44 48 52 56 60 64 68 72 76
2 Theta
2Theta
Figure 3.2: X-ray diffraction patterns of samples: a) after sintering and fine polishing; b) after 60 hours
autoclave ageing. (t-tetragonal, m-monoclinic)
Elastic modulus and hardness were determined by nanoindentation using a Berkovich indenter
and a continuous stiffness measurement unit and the results are shown with respect to indenter
displacement in Figure 3.3. Indentations were made before and after hydrothermal ageing. Both
modulus and hardness were found to change widely in the specimens before ageing because of
their different porosity. Nevertheless, there is a clear correlation between density and contact
Ravi K Chintapalli
110
Properties and hydrothermal degradation
hardness and elastic modulus in specimens S65, S90 and S120: elastic modulus and contact
hardness decrease by increasing porosity.
Elastic modulus (GPa)
240
200
160
120
80
S120
S90
S65
40
a)
0
S120-60H
S90-60H
S65-60H
S800
S800-60H
AS300
AS300-60H
b)
20
Hardness (GPa)
16
12
8
4
0
S120
S90
S65
c)
0
400
800
Displacement (nm)
1200
S120-60H
S90-60H
S65-60H
1600
S800
S800-60H
AS300
AS300-60H
d)
0
400
800
1200
1600
Displacement (nm)
Figure 3.3: Elastic modulus and hardness as a function of penetration depth before and after LTD: (a) &
(c) for fine grain materials, (b) & (d) for coarse grain materials respectively.
After hydrothermal degradation, elastic modulus and contact hardness decrease only in
specimens with larger grain size, but they remain unchanged in specimens with grain size equal
or smaller than 120 nm. Figure 3.3 shows elastic modulus and contact hardness for fine grain
(≤120 nm) and coarse grain (≥300 nm) materials. It can be appreciated that, in coarse grain
materials, ageing induces a large decrease in elastic modulus and contact hardness. A closer
inspection of the curves of aged specimens that suffer LTD reveals that, after a certain
penetration depth, the elastic modulus and hardness tend to recover their values for non-aged
specimens because of the effect of the healthy non degraded material beneath the affected
surface layer.
Ravi K Chintapalli
111
Properties and hydrothermal degradation
3.2.4 Discussion
It is well known that long dwell time has a great influence on the density in SPS1–3. The dwell
time of 5 minutes used here is too short for obtaining a high density material at the lower
temperatures used, but for temperatures equal or higher than 1,175 °C relatively high density and
small grain size are achieved. Materials will be porous (relative density between 61–90% of
theoretical density) when the sintering temperature is below 1050 °C and the soak time is limited
to 15 minutes, but dense materials are achieved when the soak time is above 60 minutes4. Under
the present conditions, porous to dense materials were obtained between temperatures 1100 and
1600 °C due to limited dwell time (5 minutes).In contrast long dwell time (2 hours) is required to
obtain full density at 14500C in conventional sintering (AS300).
Porosity is obviously the main reason for the low Vickers hardness and contact Berkovich
hardness (in S65, Vickers hardness drops to one third of its value compared to dense AS300
specimens) while the elastic modulus is only reduced to about half. These changes have an
influence on the measured indentation fracture toughness from the lengths of cracks emanating
from Vickers indents5,6.
Indentation of low porosity material produces not only plastic deformation but also it may induce
compaction of the volume underneath the contact area so that there is a reduction in the driving
force for cracks to grow from the residual elastic strain field after unloading. This may be the
main reason for the larger indentation fracture toughness encountered in the most porous
specimen as compared to dense materials. That is, the porous material is not tougher; instead it is
the residual indentation stress field that is weaker.
The reason for finding higher values of indentation fracture toughness by using equation (2.2) of
chapter 2 originates by the increase in (E/H) in the most porous specimens with respect to the
dense material (about 1.5). Indentation fracture toughness decreases in dense materials when the
grain size is reduced as can be appreciated by comparing AS300 and S120. Recently the fracture
strength of 3Y-TZP has been studied for a range of grain sizes between 110 and 480 nm by
Eichler et al7, who have shown that the biaxial fracture strength is reduced by the decrease of
grain size, which was related to the concurrent change in fracture toughness. Therefore the
increase in indentation fracture toughness for the smaller grain size is believed to be caused by
the use of Equation (2.2) (refer to chapter 2).
Ravi K Chintapalli
112
Properties and hydrothermal degradation
Under the conditions of hydrothermal degradation studied here, LTD does not occur for
specimens with grain sizes equal or smaller than 120 nm. In materials with conventional grain
size, in which degradation takes place, it is well known that porosity increases degradation as has
been reported by Masaki8. Here, in nanometric grain size specimens, since there is no
degradation by water species, the presence of porosity has no effect on degradation of the
surface.
By preparing sections perpendicular to the surface by focused ion beam machining and
observation by SEM of degraded 3Y-TZP,Gaillard et al9 have shown the existence of
microcracks mostly parallel to the surface which have been associated to the drop in elastic
modulus and hardness in degraded specimens, because of a reduction in contact stiffness. Also
Muñoz et al10 have recently shown the effect of surface degradation on these properties with
respect to degradation time, showing that properties decrease with increasing degradation time.
By contrast, in our nanometric grain size specimens, since there is no degradation of the surface,
no reduction of elastic modulus and hardness is detected by nanoindentation [see figures 3.3 a)
and c)].
Eichler et al7 have studied hydrothermal degradation of 3Y-TZP with grain sizes in the range of
110–480 nm and also fracture toughness in 2Y-TZP with average grain sizes between 150 and
900 nm11. Although their degradation tests used shorter times and higher temperature and vapour
pressure, our results are similar to theirs in the sense that significant degradation occurred only in
specimens with grain sizes larger than 210 nm. Only very small amount of monoclinic phase was
found by these authors in sintered bodies with 110 nm grain size. With respect to fracture
toughness, they found that in 2Y-TZP (which is more transformable than 3Y-TZP) both, the
extent of stress induced tetragonal to monoclinic phase transformation at the crack flanks and the
fracture toughness, diminished by decreasing the grain size with the exception of the largest
grain-size sample studied (900 nm).
Though the mechanism of low temperature degradation is still at debate, it is agreed that water
species from moisture environment penetrate into the tetragonal lattice and occupy oxygen
vacancies during hydrothermal ageing. These oxygen vacancies are annihilated near the surface
and their concentration could be low enough to destabilise the tetragonal phase, resulting in t–m
transformation at surface12. Therefore, it seems likely that in porous specimens with nanometric
grain size (S65, S90 and S120); the diffusion of water species should be more effective because
Ravi K Chintapalli
113
Properties and hydrothermal degradation
of the presence of larger area of grain boundaries and of larger free surface in contact with water.
In spite of that, no degradation was detected in these specimens after ageing, in contrast with the
large monoclinic content measured in dense specimens with large grain size (AS300 and S800).
The conditions for the phase transformation in terms of energy given by Lange13 are already
presented in chapter 1. With reference to equations (1.2) and (1.4) in chapter 1, by decreasing the
number of vacancies near the surface, the driving force for t-m transformation increases, that is,
(-ΔGc increases); ΔUse will increase less in the surface than in the bulk, since at the surface,
deformation perpendicular to the surface is not restricted.
According to Schubert and Frey14, the strain energy contribution has a dominating influence over
the surface term. It is composed of a term related to the volume increase that accompanies
transformation and makes it more difficult, and another term associated to the residual stresses
present in the manufactured body and to those created by the diffusion of water species inside
the material.
Apparently, the water radicals lead to a change in lattice parameters which shows basically a
lattice contraction of a tetragonal unit cell. This contraction due to water penetration increases
the free energy difference between t-and m-phase with respect to free powder, which has a
destabilising effect because it leads to a smaller activation barrier for transformation. The strain
energy term is clearly revealed by the fact that free tetragonal powder has higher resistance to
LTD than sintered bodies with tetragonal microstructure and with grain size near to the
crystallite size of the powder14.
It is expected that decreasing the grain size has no important effect on the strain energy, but it
increases the surface term of equation (3.1) so that transformation is more difficult. The surface
term, Us, leads to a dependence of the activation barrier for t-m transformation on the particle
size. Therefore, the activation barrier to form a critical nucleus is decreased by an increase in
particle size. This effect qualitatively explains the dependence of amount of transformation on
grain size15. If degradation was induced only by a change in the free energy because of the
removal of vacancies near the surface, its effect would be stronger in the smaller grain sizes
which have larger area of grain boundaries and of contact with water.
Ravi K Chintapalli
114
Properties and hydrothermal degradation
The stability of a grain depends also on internal stresses induced by the level of anisotropy of the
thermal expansion. For 2Y-TZP the stress levels are higher than in 3Y-TZP because of higher
thermal expansion anisotropy16. It is also found that larger grains contain high internal stresses in
the area close to the grain border compared to smaller grains17. Starting from the same level, the
stresses decrease more rapidly for smaller grain sizes. Additionally, from figure 3.1 it can be
observed that the grains appear with round edges in samples S65, S90 and S120. The stress level
in grains with round edges is lower than in perfectly sharp-edged grains17.
Ravi K Chintapalli
115
Material properties and hydrothermal degradation
3.3 Spherical indentation behavior of porous zirconia
3.3.1 Introduction
Porous zirconia ceramics are attractive candidates for a wide array of applications such as
prosthetics, dentistry, environmental filters, and sensors18–20 among other applications.
Additionally, porous zirconia ceramics are highly regarded materials as bone scaffolds coated
with hydroxyapatite (HA)21,22 where zirconia is used as a load bearing component and the HA
coating is for promoting bioactivity. Furthermore, researchers have proposed different surface
modifications23,24 to enhance bone in-growth in dental implants by increasing the surface
roughness and/or by surface micro porosity. Increasing the surface roughness by sandblasting is
a common method, but, due to the adverse effects of sandblasting on the strength of the implants,
micro porous surfaces25.
For contact loading applications such as in dental implants, the capacity of the material to absorb
energy and sustain the deformation becomes critical, so that mechanical properties such as
stress–strain behaviour and yield strength are of great interest in order to assess the material
deformation behaviour. Given the quasi-plastic nature of zirconia ceramics, it is important to
quantify the deformation and yield behaviour as these properties are critical in applications
where the material is under contact loading.
Indentation studies are commonly used for characterising fracture and deformation behaviour of
brittle ceramics; more specifically, contact damage is analyzed by Hertzian testing using
spherical indenters26.On the other hand, the influence of porosity on indentation stress–strain
behaviour and contact damage response has severe implications concerning the capacity of
ceramics to sustain mechanical damage and for improving reliability. Therefore this section
presents the indentation stress–strain curves and yield points of porous materials S65, S90 and
compared with dense S120at sub micron scale, and possible damage mechanisms are discussed.
3.3.2 Theoretical model for spherical indentation
Figure 3.4 schematically represents the interaction between a flat surface and a spherical
indenter. The response of the sample plus indenter under spherical indentation is divided into
elastic and elasto-plastic regimes. The elastic regime can be described by Hertz equations for
small penetration depths27–29:
Ravi K Chintapalli
116
Material properties and hydrothermal degradation
Figure 3.4: Schematic representation of spherical indentation.
P
3
E eff
4
3
R ht 2
1
1  2 1   2


Eeff
E
E
(3.1)
(3.2)
Where P is the applied load, R is the indenter radius, ht is the total displacement into the surface,
Eeffis the effective modulus of indenter-sample system, E and E′ are the elastic modulus of
sample and indenter respectively. For a spherical indenter, Sneddon27 demonstrated that the
elastic displacement is related to the contact radius a as:
he  htotal 
a2
R
(3.3)
Substituting equation (3.3) in equation (3.1) yields:
4
P
a

E eff  
2
3
a
R
(3.4)
The left-hand side of equation (3.4) is the Meyer hardness, represented as indentation stress and
the term in parenthesis, on the right-hand side, stands for indentation strain27. During a
nanoindentation test, load-displacement data and contact stiffness (from the CSM module) are
recorded. In particular, the contact depth is calculated from the direct stiffness measurement, S,
for the elasto-plastic regime as29:
hc  ht 
Ravi K Chintapalli
3P
4S
(3.5)
117
Material properties and hydrothermal degradation
Then the contact radius can be calculated as:
a
2 Rh c  hc
2
(3.6)
Therefore, the nanoindentation test allows calculating the contact radius using equations (3.5)
and (3.6). Then, it is possible to obtain the indentation average stress P/πa2 and the indentation
average strain, a/R. Additionally, it has been shown30 that for an isotropic material indented with
a sphere, the contact stiffness is proportional to the contact radius. This linear dependency was
used in the present work for tip calibration and to correct the zero contact point:
S  2 Eeff a
(3.7)
3.3.3 Tip calibration
In order to check the correct calibration of the tip, tests were firstly done on fused silica as a
reference material. Figure 3.5 (i) shows the obtained P-h curve, where it can be seen that the
unloading curve overlaps the loading curve, indicating that the deformation is elastic. Figure 3.5
(ii), shows the hc vs a curve which reflects the tip shape. From the known elastic modulus of
fused silica and using equations (3.4), (3.8) and (3.9), the tip radius is calculated as 23.9±0.4 m
for depths equal or less than 1200 nm, which is close to the nominal tip radius of 25 m.
Figure 3.5:i) P–h curve for fused silica, ii) hc vs a, for spherical indentation on fused silica.
3.3.4 Determination of contact point
At reduced length scales, determining the initial contact point where load and displacement are
zero is the key to plot the correct material response. The contact point in nanoindentation,
especially with tips with a large radius of curvature, may be incorrectly determined due to
Ravi K Chintapalli
118
Material properties and hydrothermal degradation
surface-tip interactions, such as capillary forces or electrostatic interactions. Several different
methods to determine the contact point have been discussed in literature31–34.
Moseson et al33used the S vs. a relationship (eq. 3.7) to fit the curve through the origin and
correct the contact point. Figure 3.6 (inset) shows the raw stiffness vs. contact radius; here the
apparent lack of linearity is clearly visible. A linear regression analysis was performed in the
apparent linear part of the curves and the contact point was found by extrapolation to zero, as
shown in figure 3.5. The plot of S vs. a in all the range studied shows a linear tendency with a
slope corresponding to 2Eeff, which is also an indication that contact stiffness S was not affected
by the plastic deformation35.
6
2.0x10
5
6.0x10
5
4.0x10
Harmonic stiffness (N/m)
6
1.6x10
5
2.0x10
6
1.2x10
0.0
0
400
800
1200
1600
2000
5
8.0x10
5
4.0x10
0.0
S65
S90
S120
Fused silica
0
1000
2000
3000
Contact radius a(nm)
4000
5000
Figure 3.6: Stiffness against contact radius and raw data before correction (inset).
3.3.5 Load–displacement curves
Figure 3.7 shows the sample load-displacement curves for the three materials. The indentations
were displacement -controlled until a maximum penetration depth of 2 µm was reached. The
maximum load, Pmax, in each material was 1, 2.4, and 3 N for S65, S90 and S120, respectively.
The solid line represents the Hertzian fit according to equation (3.1) by considering the
determined spherical nanoindentation modulus. Details of how the elastic modulus is obtained
are given in next section.
Figure 3.7 displays the initial loading part of the curves, which can be satisfactorily fitted with
Hertz equations. Deviations from Hertz curve is attributed to yielding, that is, the onset of elasto-
Ravi K Chintapalli
119
Material properties and hydrothermal degradation
plastic deformation in all samples. At this stage, deformation is controlled by both elastic and
plastic processes. In each sample, yielding occurs at different load and displacement due to the
microstructural dissimilarities and/or porosity in the samples. For instance, in the porous S65,
elasto-plastic response starts around ~ 100 mN load and a depth of ~ 350 nm, whereas in the
dense S120 it occurs around ~ 400mN load and a depth of ~ 450 nm.
1200
2500
i)
1000
Load (mN)
ii)
Hertz fit
Hertz fit
2000
800
1500
600
1000
400
500
200
0
0
0
400
800
1200
1600
400
800
1200
1600
2000
3000
3000
iii)
Hertz fit
2500
Load (mN)
0
2000
iv)
S120
2500
2000
2000
1500
1500
1000
1000
500
500
S90
S65
0
0
0
400
800
1200
Displacement (nm)
1600
2000
0
400
800
1200
1600
2000
Displacement (nm)
Figure 3.7: P–h curves with Hertz fit of i) S65, (ii) S90, (iii) S120, and (iv) comparison curve.
Figure 3.7 (iv) shows a comparison of the load - unloading curve for the three materials with
different porosity. The load required for penetration depths up to ~ 2 µm is decreasing with
increasing porosity. The area between the load and unloading curves describes the amount of
irreversible deformation. The large area in S65 is attributed to the large plastic response and low
elastic recovery, and in S120 the small area is due to small plastic response and high elastic
recovery.
3.3.6 Indentation stress–strain curves
Figure 3.8 illustrates the indentation stress-strain curves for the different materials. These curves
are the average for all indentations for each material and are presented with standard deviation in
Ravi K Chintapalli
120
Material properties and hydrothermal degradation
both stress and strain. In spite of porosity, the curves for S65and S90 exhibit low standard
deviation in the elastic region, but slight increase was observed after yielding. The increase in
the scatter in the elasto-plastic regime is believed to be caused by the local inhomogeneties in
porosity beneath each indentation and/or associated deformation events. The dashed line is the fit
obtained by using linear regression analysis of the elastic part of the curve. The elastic moduli
were determined from this slope and are presented in table 3.2. Pop-in events (sudden
discontinuities in displacement due to cracking and other deformation mechanisms) were not
observed in the P-h curves. Therefore, yield strength is obtained from the corresponding
deviation point of the Hertz fit of the load-displacement curve. The average yield strength for all
indentations together with the standard deviation is given in table 3.2.
Figure 3.8: Indentation stress–strain curves of i) S65, ii) S90, and iii) S120.
Repeatability of the material behaviour under indentation was not affected by the porosity
because of very small pore size ~ 0.24 m compared to the contact radius. This is because the
used contact stiffness for determining the point of contact by extrapolation from the linear
behaviour was selected from measurements with a relative large contact radius of at least ~2µm.
Therefore many pores were pressed in the volume affected by the contact during each
indentation.
Table 3.2 Properties obtained by spherical indentation
Material
S65
S90
S120
Yield stress, σy (GPa)
2.1± 0.3
6.7± 0.2
9.4± 0.4
Elastic modulusSph (GPa)
89±8
181±6
223±4
It can be observed that the indentation stress required for yielding decreases with increasing
porosity. After yielding, the apparent stress required to cause further deformation increases in all
Ravi K Chintapalli
121
Material properties and hydrothermal degradation
materials. The elasto-plastic transition is clearly observed in S65 and S90, where in the case of
S120, the deviation of stress-strain curve from elastic line is less abrupt, and is closer to the
elastic trend. Spherical indentation modulus (ESph) is similar to the values obtained by Berkovich
nanoindentation (EBer) (refer to tables 3.1 and 3.2), except in the case of S65. The cause of this
discrepancy is discussed later.
3.3.7 Damage characterization
Residual indentation imprints and depth profiles of the spherical indentations are displayed in
figure 3.9. Permanent deformations on the surface without any cracks were observed in all
samples. Residual indentation depths ranged from ~0.85 to ~0.25 µm for S65 to S120
respectively. The indentation profiles (see fig 3.9 (iv)) show no significant pileup around the
indentations. Additionally, the imprints were larger on the more porous materials.
Figure 3.9: AFM tapping mode height images showing residual spherical indentation imprint on (a) S65
at 1 N, (b) S90 at 2.4 N, (c) S120 at 2.9 N load and (d) residual depth profiles.
Phase transformation was then locally analyzed by micro-Raman spectroscopy on the
nanoindentation imprints. Figure 3.10 shows the spectra obtained in the centre and border of the
imprints, with the peaks labeled with the respective phase (t-tetragonal, m-monoclinic).
Additionally a spectrum obtained from 2 mm away from the indentations indicates that the
material has tetragonal phase.
In S65, a clear phase transformation from tetragonal to monoclinic was observed at the centre
and border of the indentation. In S90 and S120, very low and no phase transformation,
respectively, was detected. Raman spectra are analyzed, as explained in chapter 2, for
quantification of monoclinic phase. Monoclinic volume fraction was quantified using the model
proposed by Katagiri et al36 and it is found to be 17±2 % in S65. Quantifying monoclinic volume
fraction in S90 is complex due to the low amount of transformation (less than 7 %).
Ravi K Chintapalli
122
Material properties
p
and
d hydrotherma
al degradation
n
Figuree 3.10: Micro Raman spectra
s
of thhe residual indentationn imprints off (i) S65, (ii)
i) S90 and
0.
(iii) S120
Focusedd ion-beam (FIB) crosss-sections were
w
prepared on indenntion impressions in alll materials.
The FIB
B cut was made from
m the centtre of indentations. Sample surffaces were previouslyy
protecteed with thinn platinum coating foor ion millin
ng. Figure 3.11 shows the SEM images off
cross-seections of indentations
i
s. In porouss materials,, curtain efffect can bee observed due to thee
porosityy and the larrge cut areaa. The indenntation imprrint profile is
i highlighteed with a daashed line.
Figuree 3.11: SEM micrographss of FIB crosss-sections off spherical inndentation (i)
i) S65, (ii) S9
90 and (iii)
S120. Nootations a annd b are the locations
l
of indentation
i
a surface, respectively..
and
n interconnnected and the region
n under thee
In S65,, the porossity is highhly irregulaar and often
indentattion shows some indiication of compaction
c
n. High maagnification views of the dashedd
Ravi K Chintapalli
123
Material properties
p
and
d hydrotherma
al degradation
n
rectanglles from figgure 3.11(i) are shown in
i figure 3.1
12 with the respective zzones mark
ked as 1 andd
2. The main
m
differrence in zonne 1 and zoone 2 is thaat consideraable shrinkaage of porees is clearlyy
seen in zone1, whiich is not thhe case in zone
z
2. Allso, it appeaars that in zzone 1 som
me pores aree
f
a crack like appearancee, which arre indicatedd by arrowss. A closer inspectionn
closed forming
shows that
t
indeed these crackk like lines are in fact formed byy pore facess which cam
me togetherr
partiallyy under the load. In S90, the porossity is dispeersed and thhe interactioon between the
t pores iss
minimaal. In all materials,
m
irrespectivee of porossity, no crracks were observed under thee
indentattions.
Figuure 3.12: Higgh magnificaation views off marked zonnes in Figuree 3.11(i).
3.3.8 Diiscussion
The elasstic propertties of porouus ceramicss depend of many factoors like totall porosity an
nd shape off
the porees and of thhe solid phaase, but no universal
u
model exists to describe the elastic modulus37.
One sim
mple empiriical equatioon for the elastic
e
mod
dulus with a fitting parameter wh
hich can bee
38
related to
t the shapee of pores was
w used byy Luo and Stevens
S
inn conventionnally sintered 3Y-TZP
P
and it caan be expreessed as
E  E0 1  P  / 1  P  (3.8)
E0 correesponds to the elastic modulus of a pore-freee material (230 GPa) and P is th
he porosityy
volume fraction. The
T parametter α has been related to the shappe of the ppores in 3Y-TZP38; forr
Ravi K Chintapalli
124
Material properties and hydrothermal degradation
spherical pores it has a value close to 1 and increases as the spherical pores change to oblate
spheroids. In the present case, the values of α for fitting our measurements of the elastic modulus
by spherical indentation are approximately 1 for relatively low porosity (P = 0.02 and 0.09), and
5 for large porosity (P = 0.21) as shown in figure 3.13.
In S65, the intersection of pores is significant (see figure 3.1a), so that actually the effective
shape of the pores change; they appear to be elongated and often interconnected. If we assume
they are spheroids, the average ratio between their radii (β) is about 0.02. On the other hand, in
S90 most of the pores have near spherical shape and are dispersed with β close to the unity
(0.86). Due to this sharp contrast between the pore shape in S65 and S90, the constant α in
equation 8 for fitting should be different. It can be seen that α=1 for S90 is in full agreement with
the measured pore shape (β0.86), but in S65 we find α=5 and this value corresponds to
spheroids that are more oblate (β0.1) than those measured (β0.2). However, the result is
consistent with oblate spheroids and one has to take into account that porosity in S65 is
interconnected, which is not taken into account in the model.
Figure 3.13: Effect of pore shape and porosity content on elastic modulus.
The values of hardness and yield strength decrease, as expected, with porosity. Their reduction
follows a similar trend as the elastic modulus in the sense that they show a sharp reduction of
properties in S65, which is attributed not only to the increase in porosity but also to the pore
shape and to its interconnected character.
Ravi K Chintapalli
125
Material properties and hydrothermal degradation
The discrepancy between the elastic modulus measured by Berkovich and spherical indentation
is higher with increasing porosity. In the dense sample S120, the elastic moduli measured by
both techniques are similar. The values of the elastic modulus determined by spherical
indentation are more representative of the actual porosity than those measured by using a
Berkovich indenter. This is related to the indenter tip shape and the fact that the modulus in
spherical indentation is obtained from fitting the elastic part of the stress-strain curve. In sharp
indentation, plastic deformation is produced from the initial contact which results in partial
closure of pores and densification; by contrast, beneath the spherical indenter, where the elasticplastic transition is produced at relatively high loads, the pore structure will be more intact
during the initial contact loads, and, therefore, the spherical elastic modulus will be lower than in
sharp indentation.
Gaillard et al39 studied nanoindentation behavior of fully dense 3Y-TZP of larger grain size (300
nm) under sharp indentation. They observed pile up around the indentation, which was attributed
to the increase in volume by phase transformation. In the same work39, but under spherical
indentation, phase transformation was clearly observed at 2.7 N with a similar indenter to that
used here. In our results for S120 of lower grain size (120 nm), no phase transformation was
detected at 2.9 N with the same indenter radius. This can be attributed to the smaller grain size
that precludes tetragonal - monoclinic transformation under contact loading. The tetragonal
phase can remain stable when the grain size is below a critical value, since the activation barrier
for t-m transformation increases when the size of the critical nucleus decreases15. Along the
same lines, it is interesting to notice that the three materials studied do not suffer low
temperature degradation40, presumably because the stresses produced during hydrothermal
ageing are not large enough to activate t-m transformation in the present nanometric grain sizes.
The fact that in the more porous material transformation is observed under a load below 2.7 N,
suggests that porosity has a clear effect on phase transformation. The latter is associated with a
volume expansion in 3Y-TZP, which is seen as a pileup around the indentation in dense zirconia
of micrometric grain size. However, since in S65 pileup was not observed in spite of the
presence of phase transformation as detected by Raman analysis, it is suggested that the increase
in volume due to phase transformation has been accommodated by neighboring porosity.
It is well known that t-m transformation with its associated volume expansion is easier when
tensile and shear stresses are high and the volume expansion is not constrained. The presence of
Ravi K Chintapalli
126
Material properties and hydrothermal degradation
large amount of porosity decreases the average components of the stress but not necessarily the
local stresses and reduces the constraint for volume expansion. The stress distribution is a critical
factor for transformation41, stresses with the same sign as the component of transformation
tensor are known to assist transformation, so that shear and tensile are destabilizing, whereas
compression stresses are stabilizing42,43.
In case of dense material (S120), the volume in contact with the indenter is experiencing
homogeneous higher compression stresses, but the material below the indentation is highly
constrained. In the more porous material (S65) the average components of the stress are smaller,
but they change locally so that in some regions may reach high values together with much less
constraint of the surrounding material due to its porosity.
Ravi K Chintapalli
127
Material properties and hydrothermal degradation
3.4 Nanocomposites
3.4.1 Introduction
Tetragonal zirconia (3Y-TZP) with high toughness, strength and resistance to hydrothermal
degradation is very attractive material for bio medical applications. The tetragonal-to-monoclinic
(t-m) phase transformation plays a fundamental role by shielding the crack from external stresses
as the crack extends thereby improving the toughness. The effect depends on the transformed
wake formed at both sides of the crack. The tendency for transformation in front of the crack tip
diminishes as the grain size decreases and the Yttria content increases. For a review of
transformation toughening see, for example, the work of Hannink, et al44.
Increasing resistance to hydrothermal degradation can be achieved by reducing the grain size,
since then the spontaneous t-m transformation activated by water species is reduced. Therefore,
the strategy of reducing the grain size for avoiding hydrothermal degradation is limited by the
concurrent parallel reduction in toughening as observed in the previous section (see table 3.1).
One possible way to increase the toughness without affecting the resistance to hydrothermal
ageing is designing 3Y-TZP matrix composites with carbon nanotubes as reinforcements45. In
fact, in recent years, there has been a considerable interest in adding carbon nanotubes especially
multiwall carbon nanotubes (MWCNT) to ceramics46–49 because of improvements in fracture
toughness in some ceramic-MWCNT composites50,51.
However, up to date, the effect of MWCNT addition to zirconia is not fully understood. Some
researchers46,52,53 report increases in fracture toughness by the incorporation of MWCNTs in 3YTZP, while others have found no improvements at all54,55. For instance, Garmendia et al45 and
Mazaheri et al56 reported improved fracture toughness, which was attributed to crack bridging
and MWCNT pull-out mechanisms. On the contrary, Sun et al55 reported that the formation of
MWCNT agglomerates at grain boundaries and a weak bonding between MWCNT and zirconia
are responsible for decreasing hardness and no improvement in other properties.
In fact, there are many reasons for this disparity of mechanical properties reported in the
literature since the measured properties are sensitive to different processing methods, difficulty
in maintaining the microstructure of the matrix when the reinforcement is added, and type of test
used to measure the fracture toughness. With reference to the first point, a key step for
processing such composites is the overcoming of the strong Van der Waals forces between
MWCNTs and dispersing the isolated MWCNTs in the ceramic matrix. Several methods such as
Ravi K Chintapalli
128
Material properties and hydrothermal degradation
colloidal processing, pre-treating MWCNTs for wet milling and/or dry milling, and pre-coated
MWCNTs for preparing composite powders are discussed in the literature52,54,57,58. For example,
high power ultrasonication of MWCNTs immersed in dispersant is known to produce
agglomerate-free carbon nanotubes59; however, the effectiveness of this pre-treating method
depends on dispersant properties as shown by Inam et al60who reported that a solvent-based
dispersant, dimethylformamide (DMF), is effective in producing agglomerate-free MWCNTs.
Colloidal processing based on heterocoagulation mechanism was also shown to be effective
when the MWCNTs were pre-treated57. In this sense, Garmendia et al52 found that partially
coated MWCNTs with zirconia showed better dispersion than uncoated MWCNTs in zirconia
matrix, which was attributed to the increased surface area due to coating. Datye et al46 observed
that in-situ growth of MWCNTs directly onto zirconia particles by chemical vapour deposition
resulted in good dispersion after final sintering.
In addition, most of the reported works involve different methods for sintering. In general, it is
usually done in Ar or N atmosphere and/or hot isostatic pressing52,54,57, except a few using spark
plasma sintering (SPS)46,53. In most cases, the final density achieved by SPS is still low as
compared to monolithic ceramics, which is probably one of the main reasons for the adverse
effect on some mechanical properties.
Isolating the contribution of reinforcement to the fracture toughness is difficult to assess if the
matrix microstructure is changing in the composite. This is more likely to happen in a
transformable zirconia matrix if a large volume fraction of carbon nanotubes are added since
these may induce a strong decrease of the grain size of the matrix46,48. As a consequence, the
matrix in the composite may have a different contribution to toughening with respect to the
monolithic form. This may be the case of transformation toughening in 3Y-TZP, since this
mechanism is very sensitive to grain size.
With reference to the measurement of fracture toughness, standard 3Y-TZP biomedical grade
(grain size about 350 nm) has increasing crack resistance to crack extension (R-curve) but its
magnitude is relatively small. In 3Y-TZP matrix MWCNT composites, R-curve related to
transformation shielding is expected to be even weaker in the composite because the reduced
grain size of the matrix may prevent transformation toughening and hydrothermal ageing to take
place7,40.
Ravi K Chintapalli
129
Material properties
p
and
d hydrotherma
al degradation
n
An addditional efffect that can
c
compliccate the measuremen
m
nt of the ffracture toughness iss
environnmental subccritical cracck growth during
d
crack
k extension when the ddeterminatio
on of the R-curve iss carried ouut in air. Buut, even moore importan
nt is the facct that veryy often the indentationn
micro fracture
fr
test (IF) is usedd to obtain KIC, and th
his method gives
g
frequeently only trends
t
validd
for com
mparison purrposes but not
n for obtaaining the actual
a
fractuure toughneess. Howeveer, it is stilll
used because of itss simplicity,, and very often
o
is the only methood availablee for compaaring resultss
of fractuure toughneess betweenn different innvestigation
ns or in smaall volumes of material..
o the addittion of MW
WCNTs onn
This seection preseents the characterizatioon of the influence of
microstrructure andd mechaniccal propertties of 3Y-TZP. Thee maximum
m volume fraction off
MWCN
NTs has beeen limited to
t 2% in order
o
to ach
hieve similaar density, matrix graiin size andd
hardnesss in the monolithic cerramic and inn the compo
osite.
3.4.2 Microstructu
M
ure
Figure 3.14
3
shows SEM microograph of prepared
p
zirrconia/MWC
CNT powdeers. It can be
b observedd
that thee nanotubes seem to bee well dispeersed betweeen zirconiaa particles w
with minim
mal signs off
clusters. The MWC
CNTs appeaar intact, fleexible and ap
ppended beetween zircoonia particlees.
Figurre 3.14: SEM
M image of drried composiite powder with
w 2 vol% M
MWCNTs.
Figure 3.15
3
shows the final microstructu
m
ure of the co
omposite saamples. In ffigure 3.15 ii), iii) andd
iv), tinyy holes cann be seen at the graiin boundariies which are the loccations of burned
b
outt
MWCN
NTs during thermal etcching in air in order to
o reveal thee microstruccture. Thesee are largerr
Ravi K Chintapalli
130
Material properties
p
and
d hydrotherma
al degradation
n
and morre irregularr for 2% MW
WCNTs. Thhe differencces in meann grain size among materials withh
differennt contents of MWCN
NTs are nott significan
nt (see Tabble 3.3), annd this is because
b
thee
conditioons for SPS
S of the com
mposites weere selected
d in order too produce tthe same grrain size ass
monolitthic 3Y-TZP
P producedd by SPS. The
T grain siize (in the range
r
153-1182 m) is practicallyy
indepenndent of thhe volume fraction off nanotubess and is about
a
less tthan half the
t size off
conventtionally sinttered biomeedical gradee 3Y-TZP (aabout 350 nm
m average ggrain size).
Figgure 3.15: SE
EM images of
o thermally etched polish
hed surfacess i) 0% ii) 0.55 % iii) 1% iv) 2%.
Figure 3.16 showss the fractuure surfaces of the sam
mples with and withouut MWCNT
Ts. A closee
observaation indicattes that MW
WCNTs aree mostly lo
ocated at thhe grain bouundaries. The
T fracturee
surface appearancee points out that both innter and intrra granular fracture
f
occcurs in all materials.
m
Ravi K Chintapalli
131
Material properties
p
and
d hydrotherma
al degradation
n
Figure 3.166: SEM imagges of fracturre surfaces i)) 0% ii) 0.5 % iii) 1% iv) 2% MWCN
NTs.
3.4.3 Hy
ydrothermal degrada
ation
X-ray diffraction
d
p
patterns
werre obtained from the su
urface of thhe samples iin order to analyse thee
phase compositionn (figure 3.17 (a)). It caan be observed that onnly tetragonaal phase is present; noo
monocliinic contentt was deteccted in any specimen. Therefore,
T
t additionn of MWCN
the
NTs had noo
influencce on the phhase structuure of the coomposite materials
m
as detected byy XRD. Fig
gure 3.17(b))
shows the
t XRD paatterns of alll materials after hydro
othermal deggradation foor 200 hourrs, in whichh
monocliinic phase is not deteccted. The reasons for resistance to
t degradattion of nano
ocrystallinee
materialls are alreaddy discussedd in the prevvious sectio
ons.
t
Polished
a)
Intensity
t t
m
3YTZP-1CNT
T
3YTZP-1CNT
3YTZP-0.5 CNT
C
3YTZP-0.5 CNT
3YTZP-0CNT
T
3YTZP-0CNT
24
26
28
8
30
32
2Theta
34
b)
3YTZP-2CNT
t t
3YTZP-2CNT
T
Degraded 200H
H t
36
24
4
26
28
3
30
32
34
36
2T
Theta
Figure 3.17:
3
X-ray diffraction
d
paatterns of a) polished
p
sam
mples b) hydrrothermally aaged for 200
0 hours.
Ravi K Chintapalli
132
Material properties and hydrothermal degradation
3.4.4 Properties of the composites
The physical and mechanical properties of the composites are listed in table 3.3. Increasing the
fraction of MWCNTs reduces very slightly the final density in the range studied. The relative
density of 3YTZP-0CNT is 99.3% of the theoretical density for pure zirconia. In the case of
composites, the theoretical density is calculated according to the rule of mixtures taking the
MWCNT density as 2.1 g/cm3 from manufacturer specifications. The relative density of
composites is between 99.1 and 98.7 %; implying that the compaction of ceramic powders is
good with low levels of porosity. Only a low additional porosity is observed in figure 3.15, and
there is a homogeneous dispersion of MWCNTs in the zirconia matrix.
Table 3.3 Properties of nanocomposites materials
Material
Density
Grain size HV1
KIC
E
(g/cm3)
(nm)
(MPam1/2)a
(GPa) b
(GPa)
HBerk(GPa)
b
3YTZP-0CNT
6.06
157±84
14.9 ± 0.2
3.9 ± 0.1
221 ± 3
16.1 ± 0.2
3YTZP-0.5CNT
6.03
182±82
14.5 ± 0.2
4.0 ± 0.1
231 ± 2
15.9 ± 1.2
3YTZP-1CNT
6.01
153±81
14.2 ± 0.1
4.2 ± 0.1
231 ± 5
15.2 ± 0.8
3YTZP-2CNT
5.95
161±81
13.3 ± 0.2
4.5 ± 0.1
234 ± 4
15.3 ± 0.9
a
b
Indentation Fracture IF method Antsis equation
Berkovich nanoindentation
Vickers hardness for 3YTZP-0CNT is high compared to conventionally sintered AS300 and is
similar to SPS sintered S120, that are reported in previous section; this is probably mainly due to
its smaller grain size. On the other hand, a decrease of about 10 % in Vickers hardness was
observed in 3YTZP-2CNT, but note that hardness is still slightly higher than conventionally
sintered AS300. By contrast, a small increase in elastic modulus (less than 5%) can be observed
with the addition of MWCNTs (table 3.3). The parameter (E/HV)1/2 changes from 3.85 to 4.19,
or, (E/H)1/2 between 3.70 and 4.04, by the addition of 2 vol. % of nanotubes. Contact hardness
for all materials is found to be similar with a slight decreasing trend by the addition of
MWCNTs. However, the scatter in the experimental values was higher in the composites.
Indentation fracture toughness in terms of volume fraction of MWCNTs is given in figure 3.18
by using equations of Niihara and Anstis (equations 2.2. and 2.3 from chapter 2), since
indentation fracture toughness of 3Y-TZP is sometimes given in the literature by using either of
these two equations. Because of the limitations of indentation fracture toughness testing61,62, the
values obtained for KIC should be taken with caution. From figure 3.18 it can be observed that
indentation fracture toughness increases with volume fraction of nanotubes, independently of the
Ravi K Chintapalli
133
Material properties and hydrothermal degradation
equation used to calculate K1C. In fact for the small range of experimental values of E, H and c/a
studied here, the equation of Niihara always gives values of fracture toughness near 30 % higher
than Anstis equation.
Figure 3.18: Fracture toughness with respect to volume % of MWCNTs.
Anstis indentation fracture toughness of 3YTZP-0CNT with grain size of about 160 nm is 3.9
MPam. This is slightly lower than KIC for conventionally sintered biomedical grade full density
zirconia of grain size around 350 nm (4.1 MPam), and this is in line with the smaller crack tip
shielding expected with decreasing grain size. In 3YTZP-2CNT, indentation fracture toughness
increases with respect to 3YTZP-0CNTbecause of the addition of MWCNTs without changing
the grain size. This increase is small, but significant (from 3.9 to 4.5 MPam as measured by the
IF method).
It should be recalled that when calculating fracture toughness by the IF method, equation (2.3)
(chapter 2) has two terms, one from the length of the cracks and the other one from the (E/H)
ratio. In the present case, the length of the cracks generated during indentation in 3YTZP-2CNT
is about 15% shorter than in 3YTZP-0CNT. A slightly higher value of the E/H ratio also
contributes to the calculated fracture toughness, but the main contribution to the increase in the
calculated KIC comes from the fact that cracks in 3YTZP-2CNT are shorter than in 3YTZP0CNT for the same indentation load.
Ravi K Chintapalli
134
Material properties and hydrothermal degradation
3.4.5 Phase transformation in the crack tip
Contact loading by Vickers indenters can activate t-m transformation around the imprint, which
can be quantified by micro-Raman spectroscopy63. In the present case, the small grain size of the
sintered materials restricts the extent of t-m transformation around imprints as well as on the
wake of the indentation cracks. Maps of phase transformation obtained by micro-Raman
spectroscopy around the crack are shown in figure 3.19.
Figure 3.19: Phase transformation around the crack tip.
The area scanned was a square of 10x10 m2 with the crack tip located at the centre of the
square. It can be seen that the amount of monoclinic phase is close to zero everywhere except in
a very narrow layer on the crack path. In general, no monoclinic bands appear at distances longer
than  2.0 m from the crack faces, and the highest value of monoclinic volume fraction that
was detected ( 6 % in the composites with 0 and 2 vol. % examined) was at the crack edge. It
can be noticed that transformation in 3YTZP-2CNTwas slightly larger than in the monolithic
material, but the difference is very small. The transformation map of AS300 with larger grain
size (300 nm) is also shown and it can be appreciated the higher monoclinic volume fraction and
the extension of the transformation as compared to the SPS specimens with lower average grain
size
3.4.6 Discussion
The prerequisite for achieving improved mechanical properties is to consolidate a well dispersed
and defect free material along with high density. In a ceramic-CNT system the main hurdle to
obtaining a fully dense composite is the agglomeration of CNTs due to its poor dispersion. Some
works reported on CNT dispersion in ceramic matrix yielded poor mechanical properties47,55 due
to high volume of CNT agglomerates, while others achieved better properties45,53 with minimal
amount of CNT clusters, which is mainly related to the dispersing methods used. For instance,
our observations on CNT dispersion are consistent with those reported for pre-treated MWCNTs
Ravi K Chintapalli
135
Material properties and hydrothermal degradation
49,51
. On the other hand, Woodman et al65dry mixed non-pretreated MWCNTs with boron
carbide and they found MWCNT agglomerates in the composite powders due to poor dispersion.
Therefore, a prior treatment of carbon nanotubes is essential to obtain agglomerate free
dispersions.
The extent of toughening for grain size of range 153-180 nm is expected to be very small since it
is already weak for larger grain sizes of about 350 nm. This is in line with the small thickness
and volume fraction of the transformed wake along the crack faces, as has been shown in figure
3.19. The transformation width has also been measured by Eichler et al11 in 2Y-TZP, which is a
zirconia ceramic more transformable than 3Y-TZP because of its lower yttria content. It was
tested in air for grain sizes in the same range as in the present work (150-300 nm) and no Rcurve could be detected in air. The thickness of the transformed wake in 2Y-TZP measured by
local Raman-spectroscopy was always less than 3 m for all the above grain sizes. Since in
3YTZP-0CNT (153 nm grain size) weaker shielding is expected from transformation
toughening, its contribution to the crack tip fracture toughness will be very small.
Figure 3.20: SEM image of indentation crack in 3YTZP-2CNT, where some nanotubes are bridging the
crack faces.
It can be assumed that the magnitude of transformation toughening in the composite is the same
as in 3YTZP-0CNTsince the grain size is very similar. Any difference in their behaviour should
then be associated to the presence of MWCNTs. Figure 3.20 shows the presence of nanotubes
bridging the crack faces. This observation is also in line with previous findings by Garmendia et
al45and Mazaheri et al56 in 3Y-TZP as well by other authors in MWCNT ceramic and glass
composites66,67. The increase in toughness is attributed to crack bridging by nanotubes by these
authors. However, contribution of bridging stresses to the toughness should be further
investigated with higher concentration of MWCNT and longer cracks may be necessary in order
that their contribution on the measured properties could be more clearly established.
Ravi K Chintapalli
136
Material properties and hydrothermal degradation
3.5 Summary
Properties and the hydrothermal degradation behaviour are characterised for both monolithic and
composite materials. Additionally, the effect of porosity on mechanical properties was
investigated. The main points from this chapter are summarised as follows:
i) Materials with grain sizes less than 300 nm were highly resistant to hydrothermal
degradation for the conditions studied here.
ii) Severe degradation was observed in materials with grain sizes equal or larger than 300
nm.
iii) Mechanical properties such as elastic modulus, hardness and indentation yield strength
were affected by the porosity. Once the porosity becomes interconnected the decrease in
mechanical properties is more significant.
iv) Porosity has no effect on hydrothermal degradation in nanometric grain size materials
(<300nm). However, the presence of significant porosity promotes t–m phase
transformation under the indentation because of high local effective shear stresses and
reduction of constraint for volume expansion close to the pores
v) Indentation fracture toughness increased nearly 15% in with the addition of 2 vol.%
multiwall carbon nanotubes.
The influence of sandblasting in zirconia in various aspects is discussed in the following
chapters. AS300 will be more focused and thoroughly studied under several aspects of
sandblasting. However due to the limited availability of nanocrystalline materials, S120 and
3YTZP-2CNT are studied only for some aspects of sandblasting.
Ravi K Chintapalli
137
Material properties and hydrothermal degradation
3.6 References
1. Anselmi-Tamburini U, Garay JE, Munir ZA, et al. Spark Plasma Sintering and Characterization of
Bulk Nanostructured Fully Stabilized Zirconia: Part I. Densification Studies. Journal of Materials
Research. 2004;19(11):3255-3262.
2. Bernard-Granger G, Guizard C. Spark Plasma Sintering of a Commercially Available Granulated
Zirconia Powder: I. Sintering Path and Hypotheses about the Mechanism(S) Controlling Densification.
Acta Materialia. 2007;55:3493-3504.
3. Bernard-Granger G, Guizard C, Surblé S, Baldinozzi G, Addad A. Spark Plasma Sintering of a
Commercially Available Granulated Zirconia Powder—II. Microstructure after Sintering and Ionic
Conductivity. Acta Materialia. 2008;56(17):4658-4672.
4. Bernard-Granger G, Addad A, Fantozzi G, et al. Spark Plasma Sintering of a Commercially Available
Granulated Zirconia Powder: Comparison with Hot-Pressing. Acta Materialia. 2010;58(9):3390-3399.
5. Anstis GR, Chantikul P, Lawn BR, Marshall DBA. A Critical Evaluation of Indentation Techniques
For Measuring, Fracture Toughness I Direct Crack Measurements. Journal of American Ceramic Society.
1981;64:533-538.
6. Lawn BR, Evans AG, Marshall DB. Elastic/Plastic Indentation Damage in Ceramics: The
Median/Radial Crack System. Journal of the American Ceramic Society. 1980;63:574-581.
7. Eichler J, Rödel J, Eisele U, Hoffman M. Effect of Grain Size on Mechanical Properties of
Submicrometer 3Y-TZP: Fracture Strength and Hydrothermal Degradation. Journal of the American
Ceramic Society. 2007;90(9):2830-2836.
8. Masaki T. Mechanical Properties of Y2O3-Stabilized Tetragonal Zro2 Polycrystals after Aging at High
Temperature. Journal of American Ceramic Society. 1986;69:519-522.
9. Gaillard Y, Jimenez-Pique E, Soldera F, Mucklich F, Anglada M. Quantification of Hydrothermal
Degradation in Zirconia by Nanoindentation. Acta Materialia. 2008;56:4206-4216.
10. Muñoz-Tabares JA, Jiménez-Piqué E, Anglada M. Subsurface Evaluation of Hydrothermal
Degradation of Zirconia. Acta Materialia. 2011;59(2):473-484.
11. Eichler J, Hoffman M, Eisele U, Rödel J. R-Curve Behaviour of 2Y-TZP with Submicron Grain Size.
Journal of the European Ceramic Society. 2006;26(16):3575-3582.
12. Guo X. Hydrothermal Degradation Mechanism of Tetragonal Zirconia. Journal of Materials Science.
2001;36(1):3737 - 3744.
13. Lange FF. Transformation Toughening 1 Size Effects Associated with the Thermodynamics of
Constrained Transformations. Journal of Materials Research. 1982;17:225-234.
14. Schubert H, Frey F. Stability of Y-TZP during Hydrothermal Treatment: Neutron Experiments and
Stability Considerations. Journal of the European Ceramic Society. 2005;25(9):1597-1602.
15. Suresh A, Mayo MJ, Porter WD, Rawn CJ. Crystallite and Grain-Size-Dependent Phase
Transformations in Yttria-Doped Zirconia. Journal of American Ceramic Society. 2003;86(2):360-362.
16. Schmauder S, Schubert H. Significance of Internal Stresses for the Martensitic Transformation in
Yttria-Stabilized Tetragonal Zirconia Polycrystals During Degradation. Journal of the American Ceramic
Society. 1986;69(7):534-540.
Ravi K Chintapalli
138
Material properties and hydrothermal degradation
17. Schubert H, Petzow G. Microstructural Investigations on the Stability of Ytrria Stabilised Tetragonal
Zirconia. In Advances in Ceramics, Science and Technology of Zirconia III; Somiya, S., Yamamoto, N.,
Yanagida, H., Eds.; American ceramic society Inc.: Columbus, OH, USA. 1988;24:21-28.
18. Maca K, Dobsak P, Boccaccini AR. Fabrication of Graded Porous Ceramics Using Alumina-Carbon
Powder Mixtures. Ceramics International. 2001;27:577-584.
19. Biasini V, Parasporo M, Bellosi A. Fabrication and Characterisation of Al2O3 Porous Bodies by Hot
Isostatic Pressing. Thin Solid Films. 1997;297:207-211.
20. Gain AK, Song H-Y, Lee B-T. Microstructure and Mechanical Properties of Porous Yttria Stabilized
Zirconia Ceramic Using Poly Methyl Methacrylate Powder. Scripta Materialia. 2006;54(12):2081-2085.
21. Kim H-W, Lee S-Y, Bae C-J, et al. Porous ZrO2 Bone Scaffold Coated With Hydroxyapatite With
Fluorapatite Intermediate Layer. Biomaterials. 2003;24(19):3277-3284.
22. Miao X, Hu Y, Liu J, Huang X. Hydroxyapatite Coating on Porous Zirconia. Materials Science and
Engineering:C. 2007;27(2):257-261.
23. Kohal R, Weng D, Bächle M, Strub J. Load Custom Made Zirconia and Titanium Implants Show
Similar Osseointegration-An Animal Experiment 2004;75:1262–8. Journal of Periodontology.
2004;75:1262-1268.
24. Sennerby L, Dasmah A, Larsson B, Iverhed M. Bone Tissue Responses to Surface- Modified Zirconia
Implants-A Histomorphometric and Removal Torque Study in the Rabbit. Clinical Implant Dentistry and
Related Research. 2005;(7):13-20.
25. Chevalier J, Loh J, Gremillard L, Meille S, Adolfson E. Low-Temperature Degradation In Zirconia
with a Porous Surface. Acta Biomaterialia. 2011;7(7):2986-93.
26. Lawn BR. Fracture of Brittle Solids. Second Edition, Cambridge University press, Cambridge, United
Kingdom. 1993.
27. Tabor D. Hardness of Metals. Clarendon Press; 1951.
28. Field JS, Swain MV. A Simple Predictive Model for Spherical Indentation. Journal of Materials
Research. 1993;8(2):297-305.
29. Oliver WC, Pharr GM. Measurement of Hardness and Elastic Modulus by Instrumented Indentation:
Advances in Understanding and Refinements to Methodology. Journal of Materials Research.
2004;19(1):3-20.
30. Johanson KL. Contact Mechanics. Cambridge: Cambridge University Press; 1985.
31. Linss V, Schwarzer N, Chudoba T, Karniychuk M, Richter F. Mechanical Properties of a Graded B–
C–N Sputtered Coating with Varying Young’s Modulus: Deposition, Theoretical Modelling and
Nanoindentation. Surface and Coatings Technology. 2005;195(2-3):287-297.
32. Liang Y-H, Arai Y, Ozasa K, Ohashi M, Tsuchida E. Simultaneous Measurement of Nanoprobe
Indentation Force and Photoluminescence of InGaAs/GaAs Quantum Dots And Its Simulation. Physica
E. 2007;36(1):1-11.
33. Moseson AJ, Basu S, Barsoum MW. Determination of the Effective Zero Point of Contact for
Spherical Nanoindentation. Journal of Materials Research. 2008;23(1):204-209.
Ravi K Chintapalli
139
Material properties and hydrothermal degradation
34. Kalidindi SR, Pathak S. Determination of the Effective Zero-Point and the Extraction of Spherical
Nanoindentation Stress–Strain Curves. Acta Materialia. 2008;56(14):3523-3532.
35. Basu S, Moseson A, Barsoum MW. On The Determination of Spherical Nanoindentation Stress–
Strain Curves. Journal of Materials Research. 2006;21(10):2628-2637.
36. Katagiri G, Ishida H, Ishitani A, Masaki T. Direct Determination by a Raman Microprobe of the
Transformation Zone Size in Y2O3 Containing Tetragonal ZrO2 Polycrystals. Advances in Ceramics.
1988;24:537-544.
37. Rice RW. Porosity of Ceramics. New York: Marcel Dekker; 1998.
38. Luo J, Stevens R. Porosity-Dependence of Elastic Moduli and Hardness of 3Y-TZP Ceramics.
Ceramics International. 1999;25(3):281-286.
39. Gaillard Y, Anglada M, Jimenez-Pique E. Nanoindentation of Yttria-Doped Zirconia, Effect of
Crystallographic Structure on Deformation Mechanisms. Journal of Materials Research. 2009;24(3):719727.
40. Chintapalli R, Mestra A, García Marro F, et al. Stability of Nanocrystalline Spark Plasma Sintered
3Y-TZP. Materials. 2010;3(2):800-814.
41. Chen IW. Model of Transformation Toughening in Brittle Material. Journal of the American Ceramic
Society. 1991;74:2564-2572.
42. Heuer A, Claussen N, Keriven W, Ruhle M. Stability of Tetragonal ZrO2 Particles in Ceramic
Matrices. Journal of the American Ceramic Society. 1982;65(12):642-650.
43. Schmauder S, Mader W, Ruhle M. Calculations of Strain Distributions in and Around ZrO2
Inclusions. In Claussen N, Ruhle M, Heuer AH, edi-tors. Advances in ceramics. Science and technology
of zirconia II, vol. 12. Columbus: The American Ceramic Society. 1984;12:251-256.
44. Hannink RHJ, Kelly PM, Muddle BC. Transformation Toughening in ZrO2-Containing Ceramics.
Journal of the American Ceramic Society. 2000;83(3):461-487.
45. Garmendia N, Grandjean S, Chevalier J, et al. Zirconia–Multiwall Carbon Nanotubes Dense NanoComposites with an Unusual Balance Between Crack and Ageing Resistance. Journal of the European
Ceramic Society. 2011;31(6):1009-1014.
46. Datye A, Wu K-H, Gomes G, et al. Synthesis, Microstructure and Mechanical Properties of Yttria
Stabilized Zirconia (3YTZP) – Multi-Walled Nanotube (Mwnts) Nanocomposite by Direct In-Situ
Growth of MWNTS on Zirconia Particles. Composites Science and Technology. 2010;70(14):2086-2092.
47. Dusza J, Blugan G, Morgiel J, et al. Hot Pressed and Spark Plasma Sintered Zirconia/Carbon
Nanofiber Composites. Journal of the European Ceramic Society. 2009;29(15):3177-3184.
48. Inam F, Yan H, Peijs T, Reece MJ. The Sintering and Grain Growth Behaviour of Ceramic–Carbon
Nanotube Nanocomposites. Composites Science and Technology. 2010;70(6):947-952.
49. Shi S-L, Liang J. Effect of Multiwall Carbon Nanotubes on Electrical and Dielectric Properties of
Yttria-Stabilized Zirconia Ceramic. Journal of the American Ceramic Society. 2006;89(11):3533-3535.
50. Peigney A. Tougher Ceramics with Nanotubes. Nature Materials. 2002;2:15-16.
Ravi K Chintapalli
140
Material properties and hydrothermal degradation
51. Guo-Dong Zhan and Amiya K. Mukherjee. Carbon Nanotube Reinforced Alumina-B ased Ceramics
with Novel Mechanical, Electrical, and Thermal Properties. International Journal of Applied Ceramic
Technology.. 2004;1(2):161-171.
52. Garmendia N, Santacruz I, Moreno R, Obieta I. Zirconia-MWCNT Nanocomposites for Biomedical
Applications Obtained by Colloidal Processing. Journal of Materials Science. Materials in Medicine.
2010;21(5):1445-51.
53. Mazaheri M, Mari D, Schaller R, Bonnefont G, Fantozzi G. Processing of Yttria Stabilized Zirconia
Reinforced with Multi-Walled Carbon Nanotubes with Attractive Mechanical Properties. Journal of the
European Ceramic Society. 2011;31(14):2691-2698.
54. Duszov A, Dusza J, Tomasek K, Blugan G, Kuebler J. Microstructure and Properties of Carbon
Nanotube/Zirconia Composite. Journal of the European Ceramic Society. 2008;28(5):1023-1027.
55. Sun J, Gao L, Iwasa M, Nakayama T, Niihara K. Failure Investigation of Carbon Nanotube/3Y-TZP
Nanocomposites. Ceramics International. 2005;31(8):1131-1134.
56. Mazaheri M, Mari D, Hesabi ZR, Schaller R, Fantozzi G. Multi-Walled Carbon
Nanotube/Nanostructured Zirconia Composites: Outstanding Mechanical Properties in a Wide Range of
Temperature. Composites Science and Technology. 2011;71(7):939-945.
57. Duszova A, Dusza J, Tomasek K, et al. Zirconia / Carbon Nanofiber Composite. Scripta Materialia.
2008;58:520-523.
58. Zhou JP, Gong QM, Yuan KY, et al. The Effects of Multiwalled Carbon Nanotubes on The HotPressed 3 mol% Yttria Stabilized Zirconia Ceramics. Materials Science and Engineering: A. 2009;520(12):153-157.
59. Hilding J, Grulke EA, Zhang ZG, Lockwood F. Dispersion of Carbon Nanotubes in Liquids. Journal
of Dispersion Science and Technology. 2003;24(1):1-41.
60. Inam F, Yan H, Reece MJ, Peijs T. Dimethylformamide: an Effective Dispersant for Making
Ceramic–Carbon Nanotube Composites. Nanotechnology. 2008;19(19):195710 (5).
61. Quinn GD, Bradt RC. On the Vickers Indentation Fracture Toughness Test. Journal of the American
Ceramic Society. 2007;90(3):673-680.
62. Kruzic JJ, Ritchie RO. Comments on “Measurement of the Microstructural Fracture Toughness of
Cortical Bone Using Indentation Fracture”. Journal of Biomechanics. 2008;41(6):1379-80.
63. Reece MJ, Tetlow PL, Galiotis C. Phase-Transformation around Indentations in Zirconia. Journal of
Materials Science Letters. 1992;11(9):557-575.
64. Echeberria J, Ollo J, Bocanegra-Bernal MH, et al. Sinter and Hot Isostatic Pressing (HIP) of MultiWall Carbon Nanotubes (Mwcnts) Reinforced ZTA Nanocomposite: Microstructure and Fracture
Toughness. International Journal of Refractory Metals and Hard Materials. 2010;28(3):399-406.
65. Woodman R, Klotz B, Dowding R. Evaluation of a Dry Ball-Milling Technique as a Method for
Mixing Boron Carbide and Carbon Nanotube Powders. Ceramics International. 2005;31(5):765-768.
66. Cho J, Inam F, Reece MJ, et al. Carbon Nanotubes: Do They Toughen Brittle Matrices? Journal of
Materials Science. 2011;46(14):4770-4779.
Ravi K Chintapalli
141
Material properties and hydrothermal degradation
67. Mukhopadhyay A, Chu BTT, Green MLH, Todd RI. Understanding the Mechanical Reinforcement of
Uniformly Dispersed Multiwalled Carbon Nanotubes in Alumino-Borosilicate Glass Ceramic. Acta
Materialia. 2010;58(7):2685-2697.
Ravi K Chintapalli
142
Roughness, phase transformation and damage induced by sandblasting
Chapter
4
Roughness, phase
transformations and damage
induced by sandblasting
Ravi K Chintapalli
143
Roughness, phase transformation and damage induced by sandblasting
4.1 Introduction
The state and properties of the surface and the near surface are crucial for dental crowns and
implants for good bonding with the other surface in contact and for long-term performance of the
union. Sandblasting, as already mentioned, changes the surface roughness, activates the
tetragonal–monoclinic transformation, and may also induce damage. The first change is desired
while the phase transformation and damage are only consequential and could be adverse.
Therefore, it is important to study these consequential changes comprehensively to understand
their effect on the zirconia used to fabricate crowns and implants. Any adverse effect should be
minimized to increase the life time of the dental restorations. In this sense one strategy could be
using moderate to mild sandblasting conditions and/or changing the microstructure of the
starting zirconia.
With the objective of studying the effect of sandblasting conditions, this chapter presents the
effect
on
roughness
and
phase
transformation
in
conventional
zirconia
and
in
nanocrystalline/nanocomposite materials. Additionally, the transformation zone depth and the
subsurface damage are also characterized.
4.2 Surface roughness
4.2.1 Zirconia AS-300
4.2.1.1 Surface morphology
Figure 4.1 shows three-dimensional surface morphologies of sandblasted specimens with 110
m particles at different pressures. The morphologies were obtained from the center of the
sandblasted specimens. Also the roughness profiles traced at the center of the image are shown
below. The impact of the particles is clearly visible in the form of peaks and valleys due to
erosion and displacement of material around the impact.
The erosion has taken place in the form of ejection of material and partition of grains thereby
generating undercuts (valleys) and peaks. The undercuts are approximately 2 m deep. A subtle
difference in the morphology under different impact pressure conditions is the slight increase of
surface roughness with impact pressure from 2 to 4 bars (refer to figure 4.4).
Ravi K Chintapalli
144
Roughness, phase transformation and damage induced by sandblasting
Figure 4.1: a1) & b1) Surface morphologies, a2) & b2) surface profiles obtained from the center of the
image, for the surfaces sandblasted with 110 m particle size at an angle 90º.
Figure 4.2 shows the morphologies of specimens sandblasted with 250 m particles. In figure
4.2 (a1), a large particle impacted area can be seen, which is in slightly of spherical shape with
approximately 20 m in diameter and 3 m in depth. The undercuts are larger in size with the
250 m particles as compared to the 110 m particles, which originates a higher roughness. On
the other hand, when samples were sandblasted with an angle of 300,(figure 4.3) the undercuts
were observed as long scratches. Additionally, the length of the scratch is larger with large
particles. The depth of the impact zones is small with the lower sandblasting angle 300 as
compared to 900.
Figure 4.2: a1) & b1) Surface morphologies, a2) & b2) surface profiles obtained from the center of the
image for the surfaces sandblasted with 250 m particle size at an angle 90º.
Ravi K Chintapalli
145
Roughness, phase transformation and damage induced by sandblasting
Figure 4.3: a1) & b1) Surface morphologies, a2) & b2) surface profiles obtained from the center of the
image for the surfaces sandblasted with 110&250 m particle size at an angle 30º.
It is clear, from the residual size of the impact zone measured from roughness profiles, that the
erosion in low angle (30º) sandblasting is less compared to high angle (90º) sandblasting. This
observation is in agreement with the findings of Finne1 and Sheldon2 figure 1.22 of chapter 1 for
brittle materials.
4.2.1.2 Surface roughness
The earlier section presented only the surface morphology, but the surface roughness was
measured for a length of 4 mm at four different places in the surface of the sample. The
roughness parameter Ra is considered to represent the surface roughness and it was measured on
seven different surface conditions achieved by different treatments. For each treatment 15
samples were tested and for each sample four measurements of roughness were taken.The mean
value of Ra for the 15 samples is the value reported here in figure 4.4 for each condition.
The initial surface roughness (Ra) for the control sample (for polished condition) is 0.03±0.01
m. Sandblasting increased the surface roughness significantly under all conditions. At high
angle (90º) sandblasting, large particles (250 m) produced significantly higher roughness
compared to small particles (110 m). While at low angle (30º) sandblasting, surface roughness
increased in comparison with control conditions but decreased significantly with large particles
compared to high angle sandblasting.
Ravi K Chintapalli
146
Roughness, phase transformation and damage induced by sandblasting
2.0
1.8
1.4
a)
1.8
1.6
Pressure 2 bars
Roughness, Ra (m)
Roughness, Ra (m)
1.6
2.0
Polished
110m
250 m
1.2
1.0
0.8
0.6
0.4
0.2
0.0
1.4
Polished
110 m
250 m
b)
Pressure 4 bars
1.2
1.0
0.8
0.6
0.4
0.2
Control
SB-90º
SB-30º
Surface treatments
0.0
Control
SB-90º
Surface treatments
Figure 4.4: Surface roughness Ra for different sandblasting conditions a) at 2 bars, b) at 4 bars.
Therefore, the effect of the impact angle on surface roughness with small particles is less, but
with 250 m particles it is significant, as Ra at 90º is twice in comparison with 30º sandblasting.
The surface roughness further increased by increasing the impact pressure (from 2 to 4 bars) at
90º sandblasting for both particle sizes. Statistical significance of the mean roughness values for
all conditions will be presented later in this section.
4.2.2 Nanocrystalline zirconiaS120
4.2.2.1 Morphology & Surface roughness
Figure 4.5 shows the morphology and roughness profile of small grain size zirconia S120. Due
to the limited availability of the material produced by SPS, only one sample was sandblasted
under one condition, 110-2B-90º.
Figure 4.5: a1) Surface morphology of S120, a2) surface profile obtained from the center of the image for
the surface sandblasted with 110 m particle size at an angle 90º.
Ravi K Chintapalli
147
Roughness, phase transformation and damage induced by sandblasting
Shallow undercuts are observed compared to AS300. Here the Ra values are presented as mean
values of the five different measurements in the same specimen. The roughness Ra of the
polished sample is 0.04±0.02 m. Sandblasting at 90º under the condition 110-2B increased the
roughness to 0.74±0.09 m. In spite of having small grain size, the Ra values of S120 are quite
similar to that of AS300 under the same sandblasting conditions.
4.2.3 Nanocomposite 3YTZP-2CNT
4.2.3.1 Morphology & Surface roughness
Figure 4.6 shows the morphology and roughness profile of composite 3YTZP-2CNT sandblasted
under the condition 110-2B-90º. The undercuts are like small indents and are shallow. Similar to
S120, Ra values are presented as mean values of the four different measurements in the same
specimen. The roughness Ra of the polished sample was 0.05±0.03 m. Sandblasting at 90º
under the condition 110-2B increased the roughness to 0.63±0.1 m. In spite of the
microstructural differences and small grain size, the Ra values of 3YTZP-2CNT are comparable
to AS300 and S120 under the same sandblasting conditions.
Figure 4.6: a1) Surface morphologies of 3YTZP-2CNT, a2) surface profile obtained from the center of
the image for the surface sandblasted with 110 m particle size at an angle 90º.
4.3 Statistical analysis
A statistical one-way ANOVA with Tukey´s test was performed on the surface roughness data
obtained for all conditions at P=0.05, that is at 95 % confidence level for AS300. The difference
in mean values can be seen in figure 4.7. The figure shows that all population means of all
treatments are significantly different except for 110-2B-90º and 250-2B-30º.
Ravi K Chintapalli
148
Treatment combinations
Roughness, phase transformation and damage induced by sandblasting
significant difference
nonsignificant difference
250-2B-30 250-4B-90
250-2B-30 250-2B-90
250-2B-30 110-2B-30
250-2B-30 110-4B-90
250-2B-30 110-2B-90
250-2B-30 Control
250-4B-90 250-2B-90
250-4B-90 110-2B-30
250-4B-90 110-4B-90
250-4B-90 110-2B-90
250-4B-90 Control
250-2B-90 110-2B-30
250-2B-90 110-4B-90
250-2B-90 110-2B-90
250-2B-90 Control
110-2B-30 110-4B-90
110-2B-30 110-2B-90
110-2B-30 Control
110-4B-90 110-2B-90
110-4B-90 Control
110-2B-90 Control
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
Roughness, Ra, Mean difference
Figure 4.7: Tukey´s mean statistical difference in roughness of AS300.
4.4 Phase transformation
4.4.1 Zirconia AS300
Figure 4.8 shows X-ray diffraction patterns for all treatment conditions. The spectrum of the
polished sample shows only t-phase characteristic peaks (111)t, (002)t and (200)t implying that
material is fully tetragonal. After sandblasting, irrespective of the specific conditions, three main
aspects can be noticed from the spectra. The first one is the appearance of m-phase peak (‐111)tat
2 28.2º,the second one is the broadening of the t-phase peak(111)t at 2 31.1º and the third one
is the peak reversal of the intensity of t-phase peaks (002)t and (200)t which is also referred as
ferroelastic domain switching (refer to section 1.2.4.2 in chapter 1).This domain switching
originates the change in the peak intensity ratio (I002/I200).
Table 4.1 shows the intensity ratios of peaks (002)t and (200)t for all treatment conditions.
(I002/I200) is increasing with the severity of the surface treatment. From the table, it can be
observed that increasing the particle size and pressure significantly changes the peak intensity
ratio at high angle sandblasting, whereas only a slight change in this parameter can be seen at
low angle sandblasting.
Ravi K Chintapalli
149
Roughness, phase transformation and damage induced by sandblasting
111t
o
250-2B-30
002t
-111m
200t
o
110-2B-30
Intensity
o
250-4B-90
o
250-2B-90
o
110-4B-90
o
110-2B-90
Control-Polished
24
26
28
30
32
34
36
2Theta
Figure 4.8: X-ray diffraction patterns of samples sandblasted with different conditions.
Table 4.1 Peak intensity ratios of (002)t and (200)t for different surface treatments
Surface treatment
I002/I200
Control
0.71
110-2B-900
1.39
110-4B-900
1.95
250-2B-90
0
2.27
250-4B-90
0
2.77
110-2B-300
1.01
250-2B-300
1.33
Figure 4.9 shows the amount of monoclinic phase with respect to the surface treatments.
Monoclinic phase was measured in five samples for each treatment condition and the mean
values are given along with standard deviation. After polishing the amount of monoclinic phase
is considered to be negligible as can be concluded from the observations made from figure 4.8.
After sandblasting at 90º, mean monoclinic volume fraction induced is between 12-15%
Ravi K Chintapalli
150
Roughness, phase transformation and damage induced by sandblasting
irrespective of the particle size and pressure. At low angle (30º) sandblasting the volume fraction
of monoclinic phase induced is between 6-8% for both particle sizes. It is clear that sandblasting
at low angle (30º) induces significantly less (about half) transformation compared to high angle
(90º).
Monoclinc volume fraction Vm%
18
16
Pressure 2 bars
Polished
110 m
250 m
a)
18
16
14
14
12
12
10
10
8
8
6
6
4
4
2
2
0
Control
SB-90º
SB-30º
Surface treatments
0
Polished
110 m
250 m
Pressure 4 bars
Control
b)
SB-90º
Surface treatments
Figure 4.9: Monoclinic volume fraction of samples sandblasted with different conditions measured by
XRD.
4.4.1.1 Depth of transformation
The local phase transformation and transformation depth are characterized by micro Raman
spectroscopy. The methodology is shown in figure 4.10. The sandblasted sample is cut in to two
symmetrical pieces and the cross-section of one piece is finely polished. Micro Raman spectra
are collected by mapping an area of 140 m2. Two maps were acquired from 1 mm distance
from the center of the sample on either side as shown in figure 4.10.
Figure 4.10: Illustration of Raman map acquisition on the cross-sections of sandblasted samples. Maps
were acquired at a lateral resolution of 2 m.
Ravi K Chintapalli
151
Roughness, phase transformation and damage induced by sandblasting
t
b)
t
t
m 14 m
Intensity
mm
100
t
a)
t
200
300
t
t
mm
t
m 14 m
12 m
12 m
10 m
10 m
8 m
8 m
6 m
6 m
4 m
4 m
2 m
2 m
0 m
0 m
400
500 100
-1
Raman displacement cm
200
300
400
500
-1
Raman displacement cm
Figure 4.11: micro Raman spectra obtained from the cross-sections at different depths a) 110-2B-90º b)
250-2B-90º.
Figure 4.11 shows the Raman spectra obtained from the cross-sections of the materials
sandblasted with different conditions. The spectra shown are from single location at different
depths. The main aspects observed from the spectra are as follows: a) strong monoclinic bands
appear at 180 and 190 cm-1 of Raman displacements in the surface (0 m) b) the intensity of the
monoclinic bands decreases with increasing depth. However, the intensity of the monoclinic
bands is different for different conditions as can be seen in figure 4.11.
Figure 4.12 shows the transformation maps obtained from two zones for the sandblasting
condition 110-2B-90º. The maps can be read as follows
 Phase transformation in the depth is not homogeneous.
 A transformation gradient is present in the depth direction; however the gradient is also
not homogeneous in the direction parallel to the surface.
 The highest amount of monoclinic volume fraction estimated is between 20-25% is
located within the depth of first micron.
 The transformed region extends up to a depth of approximately 12 m.
 No phase transformation is observed after 12 m in depth in both maps, therefore for the
said sandblasting condition the depth of the transformed zone is ~12±1m.
Ravi K Chintapalli
152
Distance from the surface (m)
Roughness, phase transformation and damage induced by sandblasting
0
0
%vm
2
2
32
4
4
6
6
8
8
10
10
12
12
28
24
20
16
14
0
2
4
x (m)
6
8
10
14
12
8
4
0
0
2
4
x (m)
6
8
10
Figure 4.12: Cross-sectional phase transformation maps of AS300 sandblasted with 110-2B-90º,a) & b)
maps taken from different zones in the center of the specimen.
Figure 4.13 shows the transformation maps obtained from two zones for the sandblasting
condition 250-2B-90º. The observations from the maps are very similar to that of the figure 4.12,
except slightly higher amount of monoclinic phase (25-30%) is found, which is mainly located
within the depth of 4 microns. The transformation depth slightly changes locally as can be seen
from figure 4.13 (a) and (b). Transformation depth for the sandblasting condition 250-2B-90º is
~13±1m. Two main differences can be observed for the sandblasting conditions110-2B-90º and
250-2B-90º: i) slightly higher amount of monoclinic phase ~5% located close to the surface, ii)
slight increase in the transformation depth, which is in the order of a micron in case of
Distance from the surface (m)
sandblasting with large particles.
0
0
%vm
2
2
32
4
4
6
6
8
8
10
10
12
12
28
24
20
16
14
0
2
4
6
x (m)
8
10
14
12
8
4
0
0
2
4
x (m)
6
8
10
Figure 4.13: Cross-sectional phase transformation maps of AS300 sandblasted with 250-2B-90º,a) &
b)maps taken from different zones in the center of the specimen.
Ravi K Chintapalli
153
Roughness, phase transformation and damage induced by sandblasting
4.4.2 Nanocrystalline zirconiaS120
Figure 4.14 shows X-ray diffraction patterns for samples polished and sandblasted. The control
sample is fully tetragonal as only t-phase characteristic peaks (111)t, (002tand (200)t can be seen
from the spectrum.
After sandblasting, the characteristics observed from the spectrum are similar to AS300 which
are, appearance of m-phase peak (‐111)t at 2 28.2º, t-phase peak (111)t broadening at 2angle
31.1º and change in intensity of t-phase peaks (002)t and (200)t. The amount of monoclinic phase
after sandblasting is estimated as ~10%, which is slightly less compared to AS300 due to the
smaller grain size. The peak intensity ratios I002/I200 are 0.31 and 1.10 for polished and
sandblasted conditions respectively.
111t
Intensity
110-2B-90º
002m
-111m
200m
Control-Polished
24
26
28
30
32
34
36
2Theta
Figure 4.14: X-ray diffraction patterns of samples polished and sandblasted for S120.
Figure 4.15 shows the transformation maps obtained from two zones for the sandblasting
condition 110-2B-90º. The scale of the maps is kept constant for all materials for the sake of
simplicity in comparisons. The observations from the maps are very similar to that of the AS300.
Phase transformation is not homogeneous and a gradient is present along the depth direction.
The highest amount of monoclinic phase is estimated between 10-14% and is located within the
depth of 2-4 m. Transformation depth in this case is ~10±1m, which is slightly less than
AS300; however the depth of the transformation slightly changes locally.
Ravi K Chintapalli
154
Distance from the surface (m)
Roughness, phase transformation and damage induced by sandblasting
0
a)
0
b)
%v
m
2
2
4
4
28
6
6
20
8
8
16
10
10
12
12
14
14
0
2
4
X (m)
6
8
10
32
24
12
8
4
0
0
2
4
X (m)
6
8
10
Figure 4.15: Cross-sectional phase transformation maps of S120 sandblasted with 110-2B-90º,a) &
b)maps taken from different zones in the center of the specimen.
4.4.3 Nanocomposite 3Y-TZP-2CNT
Figure 4.16 shows X-ray diffraction patterns for samples polished and sandblasted. The control
sample is fully tetragonal; after sandblasting, the characteristics observed from the spectrum are
similar to those of AS300 and S120. That is, appearance of m-phase peak (‐111t) at 2 28.2º, tphase peak (111t) broadening at 2angle 31.1º and change in intensity of t-phase peaks (002t) and
(200t). The amount of monoclinic phase after sandblasting is estimated as ~8%.The peak
intensity ratios of peaks (002t) and (200t) are 0.29 and 1.02 for polished and sandblasted
conditions respectively.
111t
110-2B-90º
002t
Intensity
-111m
200t
Control-Polished
24
26
28
30
32
34
36
2 Theta
Figure 4.16: X-ray diffraction patterns of samples polished and sandblasted for 3YTZP-2CNT.
Ravi K Chintapalli
155
Roughness, phase transformation and damage induced by sandblasting
Figure 4.17 shows the transformation maps obtained from two zones for the sandblasting
condition 110-2B-90º. The main observations are very similar to those observed in AS300 and
S120. The highest amount of monoclinic phase estimated is between 8-10% and is located within
the depth of 3 m. Transformation depth in this case is ~8±2 m, which is less than AS300 and
Distance from the surface (m)
S120.
0
a)
0
b)
%v
m
2
2
4
4
6
6
8
8
10
10
8
12
12
4
14
14
32
28
24
20
16
12
0
2
4
6
X (m)
8
10
0
0
2
4
6
8
X (m)
10
Figure 4.17: Cross-sectional phase transformation maps of 3YTZP-2CNT sandblasted with 110-2B-90º,a)
& b)maps taken from different zones in the center of the specimen.
4.5 Effect of heat treatment after sandblasting
Figure 4.18 shows X-ray diffraction patterns for all materials after polishing, sandblasting and
sandblasting plus heat treatment (10000C, for 1 hour). The characteristics of the XRD spectra for
polished and sandblasted specimens were already explained in the previous sub sections for the
different grain sizes of 3Y-TZP and for the zirconia-MWCNT composites.
i)
111
t
AS300
ii)
110-2B-90º-HT
Intensity
-111
m
002
m 200m
S120
111
t
-111
m
002
m 200
m
iii)
111
t
3Y-TZP-2CNT
-111
m
002
m 200
m
110-2B-90º
Polished
24 26 28 30 32 34 36
2 Theta
24 26 28 30 32 34 36
2 theta
24 26 28 30 32 34 36
2 Theta
Figure 4.18: X-ray diffraction patterns after different treatments for materials i) AS-300, ii) S120 and iii)
3YTZP-2CNT .
Ravi K Chintapalli
156
Roughness, phase transformation and damage induced by sandblasting
After sandblasting plus heat treatment all materials behave in similar manner: i) disappearance of
monoclinic peak (‐111t) at 2 28.2º due m-t transformation, ii) the broadened tetragonal peak
(111t) at 2 31.1º is narrowed back, but not to the full extent iii) peak reversal of (002t) and (200t)
t-phase peaks and their intensity ratios remain unchanged.
In case of AS300, samples were heat treated at different temperatures for one hour between 500
and 1000 0C after sandblasting. The change in the amount of monoclinic volume with respect to
the heat treatment temperature is shown in figure 4.19. The reverse monoclinic-tetragonal
transformation initiated with heat treatment at 500 ºC and monoclinic phase gradually decreased
with increasing heat treatment temperature. The m-t transformation is completed and no
monoclinic phase is found after heat treatment at 1000 ºC.
Monoclinic volume fraction %
16
14
110-2B-90º-HT
250-2B-90º-HT
Sandblasted only
12
10
8
6
4
2
0
20 24 28
500
600
700
800
900
1000 1100
0
Temperature C
Figure 4.19: Change in monoclinic volume fraction with temperature of AS300 after sandblasting.
4.6 Damage induced by sandblasting
The sub surface damage induced by sandblasting is investigated by making FIB cross-sections of
the sandblasted samples. FIB trenches are made in the center of specimen. The dimensions of the
trench are 8-10 m in length and 5-8 m in depth. The analyzed conditions and the respective
materials are presented below.
4.6.1 Zirconia AS300
Figure 4.20 shows SEM micrographs of polished cross-section of a sandblasted specimen taken
from two different regions in the same cross-section.
Ravi K Chintapalli
157
Roughness, phase transformation and damage induced by sandblasting
Figure 4.20: SEM images of polished cross-section of sandblasted AS300 with 110-2B-90º a) and b)
images taken in different zones.
The particle impact zones are shown with arrowsand appear like deep cuts into the material from
the surface due to erosion. Lateral cracks are formed below the impact zones. The size of the
observed cuts and cracks is dissimilar in different zones due to the differences in the local impact
and the kinetic energy of that particular particle hitting the surface. Since these micrographs have
not revealed any microcracks at this resolution, it is necessary to observe the sub surface damage
(microcracks and any other associated damage mechanisms) at higher resolutions in small
regions.
Four FIB trenches are made AS300 for three different sandblasting conditions. Figure 4.21
shows a FIB trench of AS300 sandblasted by condition 110-2B-90º. The primary observations
from the cross-section are i) a plastically deformed zone of approximately 2 to 3 microns (above
the dotted line in figure 4.21), ii) grains are highly deformed in this layer and below this layer the
grain structure is intact, iii) a grain boundary microcrack (shown with a black arrow), iv)
martensite plate formations (monoclinic variants) in the grains (shown with white
arrows).Martensite plates are commonly found in zirconia as a result of phase transformation (tm), which is shown in figure 1.6 of chapter 1.
However, the thickness of the deformed layer is not uniform due to the difference in impact
received by the material locally. Figure 4.22 shows the top right region of figure 4.21 after
polishing with Gallium ions. More grain boundary microcracks can be seen (shown with white
arrows) near to the surface. Two of these grain boundary microcracks are oriented parallel to the
surface while others are oriented at an angle of 120º to the surface. Black arrows show the
monoclinic twins from the grain boundary to the interior of the grain just below the surface.
Ravi K Chintapalli
158
Roug
ghness, phase
e transformation and damag
ge induced by
y sandblasting
g
Figure 4.23
4
shows another FIB
B trench maade on AS30
00 sandblassted by conddition 110-2
2B-90º. Forr
this conndition no microcracks
m
were obserrved and on
nly the very first layer oof the grains is slightlyy
distortedd. Also maartensite plaates are cleaarly visiblee. Figure 4.224 shows thhe high maagnificationn
view off near surfaace region. The orientaation of thee martensitee plates is aanalysed in this regionn
and no preferential
p
l orientationn is found.
Figure 4.211: SEM imagge of first FIB
B trench of material
m
AS3000 sandblastted 110-2B-9
90º.
Figure 4.22: SEM image
i
of first FIB trench after ion beam etching of
o material A
AS300 sandbllasted 1102B-90º.
Ravi K Chintapalli
159
Roughness, phase transformation and damage induced by sandblasting
Figure 4.23: SEM image of a second FIB trench after ion beam etching of material AS300 sandblasted
110-2B-90º.
Figure 4.24: Close-up view of the top region of figure 4.20 and orientation of monoclinic variants.
Figure 4.25 shows a FIB trench of AS300 sandblasted by condition 110-2B-30º. The dotted line
in the figure separates the two distinguished regions. The grain structure is plastically deformed
in the region above the dotted line and the thickness of this layer is ~3.5 m. Below this layer
the grain structure remains unchanged; however, martensite plates are present to a lesser extent.
Ravi K Chintapalli
160
Roughness, phase transformation and damage induced by sandblasting
Figure 4.25: SEM image of FIB trench of material AS300 sandblasted 110-2B-30º.
Figure 4.26 shows a FIB trench of material AS300 sandblasted by condition 250-2B-90º. In this
case a crack parallel to the surface with the length of 2.5 m is observed. The crack is located
near an impact zone and is within a micron distance from the surface. Moreover some micro
cracking is also observed on one side of the crack. The grain boundary microcracks are oriented
parallel to the surface unlike the microcracks observed in 110-2B-90º (figure 4.21). However, a
clearly distinguished damaged layer is not visible in this case.
200 nm
1 m
250-2B-90º
Figure 4.26: SEM image of FIB trench of material AS300 sandblasted 250-2B-90º.
The damage induced by sandblasting clearly depends on the impact received in a region.
Observing large area in the polished cross-section reveals some zones in which no damage can
be detected and zones where damage can be found. On the other hand, observing a few square
microns also shows that microcracks are present in some regions only. This suggests that
Ravi K Chintapalli
161
Roug
ghness, phase
e transformation and damag
ge induced by
y sandblasting
g
sandblaasting damaage is inhom
mogeneous and concen
ntrated in high
h
impacct zones. A systematicc
study iss therefore needed
n
to quuantify the representati
r
ive damage..
4.6.2 Ziirconia S12
20
In S1200two FIB trenches
t
aree made forr one sandb
blasting condition 1100-2B-90º. Figure
F
4.277
shows SEM
S
microggraphs of tw
wo FIB trennches made at the centeer of the sanndblasted sample. Thee
main obbservations from figuure 4 a) annd b) are: thin
t
deform
med grain llayer within
n a micronn
distancee from the surface,
s
no evidence
e
off microcrack
ks and sparsse porosity. The monocclinic twinss
are not clearly
c
visibble due to thhe very smaall grain size of S120.
Figure 4.27: SEM image
i
of FIB
B trenches off material S12
20 sandblastted 110-2B-990ºa) & b) FIIB trenches
i different sites.
in
s
anocompo
osite 3YTZP
P-2CNT
4.6.3 Na
Figure 4.28presentts a SEM micrographh of one FIB
F trench made in 3YTZP-2CN
NT for onee
sandblaasting condiition 110-2B
B-90º. Whiite arrows point to CN
NT agglom
merations, black
b
arrow
w
shows a crack paraallel to the surface.
s
Thee region abo
ove the dotteed line is a thin layer of
o deformedd
Ravi K Chintapalli
162
Roughness, phase transformation and damage induced by sandblasting
grains and the thickness of this region is ~ 0.8 m from the surface. The tiny holes are locations
of multiwall carbon nanotubes oriented perpendicular to the face of the FIB trench. Some carbon
nanotubes are also oriented parallel to the face of the trench are shown with thick white arrows.
1 m
Figure 4.28: SEM image of FIB trenches of material 3Y-TZP-2CNT sandblasted 110-2B-90º.
4.7 Discussion
4.7.1 Surface roughness
The surface roughness after polishing and sandblasting has been presented for different
conditions and materials. It is clear, as expected, that sandblasting increases the surface
roughness significantly in all conditions and materials studied. Increasing the sandblasting
particle size and pressure increase the surface roughness.
Impact angle has no influence on roughness when sandblasted with small particles (110 m),
whereas significant reduction in roughness is observed when sandblasted with large particles
(250 m) at 30º impact angle in comparison with 90º.
It is also important to emphasize the effect of sandblasting parameters such as particle size,
pressure and impact angle, as the final roughness will depend on these conditions. Moreover
these parameters can be used as design criteria to achieve a pre-determined roughness range.
From the morphology observation and measured Ra, it is clear that increasing the particle sizeand
Ravi K Chintapalli
163
Roughness, phase transformation and damage induced by sandblasting
velocity results in severe erosion due to the higher kinetic energy of the particles3–5.Increase in
surface roughness is the result of high erosion.
The main purpose of sandblasting is to improve the wetting ability of the intaglio surface of the
dental crowns and pillars6–9 for better adhesion and retention. The wetting ability of the surface
is improved by increasing the surface area by means of increasing the surface roughness.
However, the choice of design criteria (sandblasting parameters) is only possible when forehand
information is available on the degree of surface roughness required for better adhesion.
As several studies10–12on influence of sandblasting on bond strength conclude that though
sandblasting increases the bond strength, the increase is highly associated with cement selection
rather than surface roughness. On the hand, in a study13 of cell attachment to the smooth
(grounded) and rough surfaces (sandblasted) of zirconia based materials it was concluded that
cell attachment was greater in sandblasted surfaces but no information given on degree of
roughness required.
The above conclusions leads to less relevance to the degree of the surface roughness required for
better adhesion. Therefore it would be ideal to choose a sandblasting condition with less damage
irrespective of the surface roughness.
4.7.2 Phase transformation
The phase changes produced by sandblasting are presented for all materials and conditions
studied. Sandblasting induces t-m phase transformation in all materials irrespective of the
conditions used. The results observed in this study are in partial agreement with those in
literature8,14–18in the sense that the amount of phase transformation is influenced by the severity
of the sandblasting treatment.
In this study it is found that the amount of monoclinic phase induced depends on the impact
angle rather than particle size or pressure. Changes in particle size and pressure have slight effect
on the phase transformation; however, increasing the impact angle increases induced amount of
monoclinic phase. Among the different materials used for 110-2B-90º sandblasting, it is shown
that S120 and 3YTZP-2CNT have slightly less amount of monoclinic phase compared to AS
300.For a commercial grade zirconia, the amount of monoclinic phase reported14,15,19 is
between 10-15% which is very similar for AS300 for similar sandblasting conditions.
Ravi K Chintapalli
164
in
Roughness, phase transformation and damage induced by sandblasting
The asymmetrical peak (111t) broadening after sandblasting was attributed to different
explanations by several authors8,16,18,20. This broadening was initially associated in zirconia
subjected to grinding and was attributed to the presence of rhombohedral phase20,21; however, in
more recent works Curtis et al8 and Scherrer et al16 have attributed this broadening to the
emergence of a new cubic phase called cubic 2 phase, as they have cubic 1 phase in their starting
material. It is also reported18 that the lattice parameters of rhombohedral phase are similar to
those of the cubic phase. In a recent study of zirconia subjected to grinding Muñoz et al22
suggested the peak broadening is due to an overlap of (111) tetragonal peak with another
symmetrical peak corresponding to the reflection of the (111) plane of the cubic phase.
On the other hand, Kondoh23 attributed this broadening to the lattice distortion in zirconia
ceramics. No cubic phase is detected by XRD in the starting materials studied here, so we
suggest the peak broadening is probably related to lattice distortion. In spite of observing peak
broadening in zirconia for many years, this phenomenon is not fully studied and several
hypothetical explanations exist in literature.
3.0
2.5
I002/I200
2.0
1.5
1.0
0.5
0
4
8
12
16
Monoclinc volume fraction (Vm)
Figure 4.29: Effect of monoclinic volume fraction on the ratio of I002/I200.
The severity of sandblasting treatment influences the amount of monoclinic phase and in turn the
ratio of peaks I002/I200 (table 4.1). The peak intensity ratio increases with amount of monoclinic
phase (see figure 4.29); however, any trend could not be detected because of limited
experimental values. Zhu et al24reported that the change in intensity of peaks (200) and (002) is
Ravi K Chintapalli
165
Roughness, phase transformation and damage induced by sandblasting
associated with phase transformation in zirconia subjected to grinding. They found that peak
intensity ratio changes only when phase transformation (t-m) occurs and it increases with the
transformation. It is also suggested that, the tetragonal variant reorientation under stress resulted
in peak reversal. Similarly, Virkar and Matsumoto25also reported that grinding of zirconia lead to
the reorientation of domains in the tetragonal grains and the interchange of t doublet.
Sandblasting and grinding are different, in the sense that, local temperatures during grinding are
high enough to activate reverse t-m transformation while under sandblasting temperatures are not
high. However, both the processes have a similarity that they induce t-m transformation and
compressive stresses at the surface. As mentioned above that the change in peak intensity ratio is
associated with transformability, therefore peak reversal under sandblasting is believed to be
caused by tetragonal variant reorientation via t-m phase transformation.
On the other hand, upon annealing (1000 0C) the monoclinic phase has transformed back to
tetragonal, but the peak intensity ratio of I002/I200 is not recovered. This phenomenon was also
observed previously14,19,24 in literature but no explanation was given. A detailed study of
crystallography of t and m phases is beyond the scope of this thesis; readers are referred to the
works of Kelly and Rose26,27 and Hannik28.
The phase transformation depth after sandblasting is measured on the cross-sections by micro
Raman spectroscopy. The phase transformation maps for the studied conditions are presented.
The main observations are: presence of transformation gradient in the depth direction and
inhomogeneous transformation parallel to the surface irrespective of the sandblasting conditions
and materials studied. However, slight differences are observed among materials in the amount
of monoclinic phase detected and its depth. In case AS300 the depth of the transformation is ~13
m while S120 and 3YTZP-2CNT the depth is ~10 and ~ 8m respectively. Also higher
monoclinic amount is detected in AS300 compared to S120 and 3YTZP-2CNT, which is
attributed to the former´s larger grain size compared to the later.
In the literature few studies6,15 have reported the depth of transformation after sandblasting.
However large differences exist in these results due to the measuring or estimating method.
Kosmac et al6 reported a transformation depth of 0.3 m whereas Sato et al15 reported ~10 m.
Sato et al15also found the presence of the transformed monoclinic phase toward the interior in a
decreasing trend. The results presented here are similar with results of Sato et al15.
Ravi K Chintapalli
166
Roughness, phase transformation and damage induced by sandblasting
4.7.3 Damage
The information regarding the nature of damage produced by sandblasting is very limited in the
literature. Most of the studies have given either a hypothetical descriptions of the damage or no
information, while a few studies29,30 gave microscopical evidence and observed lateral cracks in
the polished cross-sections of the sandblasted materials. In this work FIB cross-sections were
made after sandblasting to investigate the sub-surface damage.
The main damage mechanisms observed after sandblasting are: i) an inhomogeneous layer of
highly deformed grains, ii) phase transformation (martensite plates) iii) local microcracks. In
case of 110-2B-90º few microcracks were observed and are oriented at an angle of 120º to the
surface in one of the FIB trench in material AS300, whereas no microcracks were observed in
S120 and 3YTZP-2CNT. Additionally, no clear preferential orientation of the monoclinic twins
is found in AS300. On the other hand, for 250-2B-90º a macrocrack (~2.5 m) oriented parallel
to the surface is observed in AS300 with adjacent microcracks. This indicates that increasing the
sandblasting particle size will induce larger cracks.
4.8 Summary
The effect sandblasting conditions on roughness, phase transformation and damage is studied.
The main observations are summarized below.
‐
Increasing sandblasting particle size and pressure increases the surface roughness and
large particles produce deep under cuts.
‐
Impact angle has no influence on roughness when sandblasted with small particles (110
m), whereas significant reduction in roughness is observed when sandblasted with large
particles (250 m) at 30º.
‐
Changes in particle size and pressure have slight effect on the phase transformation,
while decreasing the impact angle decreases the amount of monoclinic phase.
‐
For similar sandblasting conditions, S120 and 3YTZP-2CNT have slightly less amount of
monoclinic phase compared to AS 300 due to their smaller grain size.
‐
The maximum average depth of the transformation in AS300 is ~13 m whereas in case
of S120 and 3YTZP-2CNT the depth is ~10 and ~ 8 m respectively; additionally, the
transformation is not homogenous in this zone and its depth is changing from one
position to the other.
‐
Sandblasting damage micro mechanisms are plastic deformation and localised
microcracks.
Ravi K Chintapalli
167
Roughness, phase transformation and damage induced by sandblasting
4.9 References
1. Finnie I. Some Reflections on the Past and Future of Erosion. Wear. 1995;186-187:1-10.
2. Sheldon GL, Finnie I. On The Ductile Behaviour of Nominally Brittle Materials during Erosive
Cutting. Transactions. ASME. 1966;88B:387-392.
3. Tilly GP, Sage W. The Interaction of Particle and Material Behaviour in Erosion Processes. Wear.
1970;16:447-465.
4. Ismail J, Zaïri F, Naït-Abdelaziz M, Bouzid S, Azari Z. Experimental and Numerical Investigations on
Erosion Damage in Glass by Impact of Small-Sized Particles. Wear. 2011;271(5-6):817-826.
5. Ramesh P. Failure Mechanisms in Impact Erosion of Ductile Materials. Master Thesis. West Virginia
University.2004.
6. Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. The Effect of Surface Grinding and
Sandblasting on Flexural Strength and Reliability of Y-TZP Zirconia Ceramic. Dental Materials.
1999;15:426-433.
7. Wolfart M, Lehmann F, Wolfart S, Kern M. Durability of the Resin Bond Strength to Zirconia Ceramic
after Using Different Surface Conditioning Methods. Dental Materials. 2007;23(1):45-50.
8. Curtis AR, Wright AJ, Fleming GJP. The Influence of Surface Modification Techniques on the
Performance of a Y-TZP Dental Ceramic. Journal of Dentistry. 2006;34(3):195-206.
9. Gahlert M, Gudehus T, Eichhorn S, et al. Biomechanical and Histomorphometric Comparison Between
Zirconia Implants with Varying Surface Textures and a Titanium Implant in the Maxilla of Miniature
Pigs. Clinical Oral Implants Research. 2007;18:662-668.
10. Ban S. Reliability and Properties of Core Materials for All-Ceramic Dental Restorations. Japanese
Dental Science Review. 2008;44:3-21.
11. Castillo de oyague R, Monticelli F, Toledano M, et al. Influence of Surface Treatments and Resin
Cement Selection on Bonding to Densely-Sintered Zirconium-Oxide Ceramic. Dental Materials.
2009;5:172-179.
12. Kern M, Wegner SM. Bonding to Zirconia Ceramic: Adhesion Methods and Their Durability. Dental
Materials. 1998;14(1):64-71.
13. Yamashita D, Machigashira M, Miyamoto M, et al. Effect of Surface Roughness on Initial Responses
of Osteoblast-Like Cells on Two Types of Zirconia. Dental Materials Journal. 2009;28(4):461-470.
14. Kosmac T, Oblak C, Marion L. The Effects of Dental Grinding and Sandblasting On Ageing and
Fatigue Behavior of Dental Zirconia (Y-TZP) Ceramics. Journal of the European Ceramic Society.
2008;28:1085-1090.
15. Sato H, Yamada K, Pezzotti G, Nawa M, Ban S. Mechanical Properties of Dental Zirconia Ceramics
Changed with Sandblasting and Heat Treatment. Dental Materials Journal. 2008;27(3):408-414.
16. Scherrer SS, Cattani-Lorente M, Vittecoq E, et al. Fatigue Behaviour in Water of Y-TZP Zirconia
Ceramics after Abrasion with 30 m Silica-Coated Alumina Particles. Dental Materials. 2011;27(2):e28e42.
Ravi K Chintapalli
168
Roughness, phase transformation and damage induced by sandblasting
17. Fischer J, Grohmann P, Stawarczyk B. Effect of Zirconia Surface Treatments on the Shear Strength of
Zirconia / Veneering Ceramic Composites. Dental Materials Journal. 2008;27(3):448-454.
18. Cattani Lorente M, Scherrer SS, Richard J, et al. Surface Roughness and EDS Characterization of a
Y-TZP Dental Ceramic Treated with the CoJetTM Sand. Dental Materials. 2010;26(11):1035-1042.
19. Guazzato M, Quach L, Albakry M, Swain MV. Influence of Surface and Heat Treatments on the
Flexural Strength of Y-TZP Dental Ceramic. Journal of Dentistry. 2005;33(1):9-18.
20. Denry IL, Holloway JA. Microstructural and Crystallographic Surface Changes after Grinding
Zirconia-Based Dental Ceramics. Journal of Biomedical Materials Research. Part B, Applied
Biomaterials. 2006;76(2):440-448.
21. Hasegawa H. Rhombohedral Phase Produced in Abraded Surfaces of Partially Stabilized Zirconia
(PSZ). Journal of Materials Science Letters. 1983;2:91-93.
22. Muñoz-Tabares J, Jiménez-Piqué E, Reyes-Gasga J, Anglada M. Microstructural Changes in Ground
3Y-TZP and their Effect on Mechanical Properties. Acta Materialia. 2011;59(17):6670-6683.
23. Kondoh J. Origin of the Hump on the Left Shoulder of the X-Ray Diffraction Peaks Observed in
Y2O3-Fully and Partially Stabilized ZrO2. Journal of Alloys and Compounds. 2004;375(1-2):270-282.
24. Zhu H, Xu Z. Tetragonal Domain Switching via Reversible t-m Phase Transformation. Journal of
Materials Science Technology. 1996;12:225-260.
25. Virkar A, Matsumoto R. Ferroelastic Domain Switching as a Toughening Mechanism in Tetragonal
Zirconia. Journal of the American Ceramic Society. 1986;69:224-226.
26. Kelly PM, Rose LRF. The Martensitic Transformation in Ceramics - Its Role in Transformation
Toughening. Progress in Materials Science. 2002;47(5):463-557.
27. Kelly PM, Ball CJ. Crystallography of Stress-Induced Martensitic Transformations in Partially
Stabilized Zirconia. Journal of American Ceramic Society. 1986;69:259-264.
28. Hannink RHJ, Kelly PM, Muddle BC. Transformation Toughening in ZrO2-Containing Ceramics.
Journal of the American Ceramic Society. 2000;83(3):461-487.
29. Kosmac T, Oblak C, Jevnikar P. The Fracture and Fatigue of Surface-Treated Tetragonal Zirconia (YTZP) Dental Ceramics. Materilas and Technology. 2007;41(5):237-241.
30. Zhang Y, Lawn BR, Rekow ED, Thompson VP. Effect of Sandblasting on the Long-Term
Performance of Dental Ceramics. Journal of Biomedical Materials Research. Part B, Applied
Biomaterials. 2004;71B:381-386.
Ravi K Chintapalli
169
Roughness, phase transformation and damage induced by sandblasting
Ravi K Chintapalli
170
Effect of sandblasting on mechanical properties
Chapter
5
Mechanical properties after
Sandblasting
Ravi K Chintapalli
171
Effect of sandblasting on mechanical properties
5.1 Introduction
Strength of dental ceramics plays a significant role in determining the performance and life time
of a restoration. The strength is controlled by the size and nature of the surface and subsurface
defects, as well as by surface residual stresses induced during fabrication. For dental zirconia, as
mentioned earlier, the surfaces are often sandblasted. This treatment will induce surface defects
that can compromise the strength of the ceramic1. On the other hand, the t-m transformation
induces compressive stresses which may counter act these flaws and increase the strength2.
Therefore, the net effect of sandblasting is not clear, since it will depend of the influence of each
phenomenon. The net result of the two contrasting effects of sandblasting can be examined by
considering the effect of the treatment.
In principle, severe sandblasting conditions induce large defects, while mild conditions induce
smaller defects. As mentioned in chapter 1, many authors1,3,4 have studied only one sandblasting
condition except a few5,6. For instance, Curtis et al5 studied the effect of sandblasting on the
strength of LAVA Y-TZP using different particle sizes (25-110 m) and reported that strength
did not change significantly. As mentioned in chapter 1, the effect of sandblasting conditions on
strength is not clear. In this work an attempt is made to study the effect of particle size, impact
pressure and angle on the strength.
In the literature1,4, it is reported that sandblasting induced defects have the nature of true
microcracks. But these observations are only hypothetical, as no microscopic evidence was given
by those authors. In this work, in chapter 4, the sub surface damage is observed using FIB and
few localised microcracks were found. Additionally, the occurrence of microcracks depends on
the severity of the conditions and whether the region is one of high impact.
In this chapter, the bi-axial strength, elastic modulus hardness and presence of residual stresses
after sandblasting are presented. Due to limited availability and geometry limitations of the
nanocrystalline and nanocomposite materials these properties are investigated only in AS300.
5.2 Bi-axial strength
The bi-axial strength was measured in AS 300 samples by using a ball on three ball test
configuration. Figure 5.1 shows the mean strength after sandblasting with seven different
conditions. The mean bi-axial strength increased after sandblasting with 110-2B-90º, while with
250-2B-90º the strength decreased in comparison with the control sample. For an impact angle of
Ravi K Chintapalli
172
Effect of sandblasting on mechanical properties
30º, the strength slightly increased for both particle sizes compared to control sample; however,
the standard deviation was higher. While, on the other hand, when the pressure is increased to 4
bars, the strength increased slightly with 110-4B-90ºand decreased significantly with 250-4B-90º
with higher standard deviation in both cases.
1800
Mean Bi-axial strength (MPa)
1600
Polished
110 m
250 m
Pressure:2 Bars
a)
Polished
110 m
250 m
Pressure:4 Bars
Control
SB-90º
b)
1400
1200
1000
800
600
400
200
0
Control
SB-90º
SB-30º
Surface treatments
Surface treatments
Figure 5.1: Mean bi-axial strength of AS300 after sandblasting at a) 2 and b) 4 bars pressure.
The Weibull analysis of the strength for all sandblasted conditions is shown in figure 5.2 and the
Weibull modulus (m) is listed for each condition. The highest m value is observed for 110-2B90º and is significantly higher compared to control sample. The lowest m is observed for 2504B-90ºand is slightly less compared to the control sample. In case of impact angle 30º the
Weibull modulus is similar in both particle sizes and with the control sample.
A statistical one-way ANOVA with Tukey´s test was performed on all conditions. At p=0.05, the
mean strengths are significantly different. The significant statistical difference is found between
groups 110-2B-90º, 250-4B-90º, 110-2B-30º and control. On the other hand, the statistical
difference is not significant between groups 110-4B-90º, 250-2B-90º, 250-2B-30º and control.
The highest difference is found between groups 110-2B-90º and 250-4B-90º having highest and
lowest mean strength values respectively.
Ravi K Chintapalli
173
Effect of sandblasting on mechanical properties
1.5
Control
o
110-2B-90
o
110-2B-30
o
250-2B-90
o
250-2B-30
1.0
0.5
m=12
Control
m=12
o
110-4B-90 m=14
o
250-4B-90 m=9
m=21
m=14
m=9
m=14
ln(-ln(1-F))
0.0
-0.5
-1.0
-1.5
-2.0
-2.5
-3.0
Pressure 2 Bars
-3.5
600
800
a)
b)
Pressure 4 Bars
1000 1200 1400 1600 1800 400
Mean bi-axial strength (MPa)
600
800 1000 1200 1400 1600 1800
Mean bi-axial strength (MPa)
Groups
Figure 5.2: Weibull plot of AS300 after sandblasting at a) 2 and b) 4 bars pressure.
Level7
Level7
Level7
Level7
Level7
Level7
Level6
Level6
Level6
Level6
Level6
Level5
Level5
Level5
Level5
Level4
Level4
Level4
Level3
Level3
Level2
Level6
Level5
Level4
Level3
Level2
Level1
Level5
Level4
Level3
Level2
Level1
Level4
Level3
Level2
Level1
Level3
Level2
Level1
Level2
Level1
Level1
Significant difference
Nonsignificant difference
-750
-500
-250
0
250
500
Level 1
Control
Level 2
110-2B-90º
Level 3
250-2B-90º
Level 4
110-4B-90º
Level 5
250-4B-90º
Level 6
110-2B-30º
Level 7
110-2B-30º
750
Mean difference in Bi-axial strength (MPa)
Figure 5.3: Tukey´s mean statistical difference in strength among groups.
Figure 5.3 shows the Tukey’s mean difference in strength for all treatment combinations. The
sandblasting conditions such as particle size, pressure and impact angle have significant
influence on the bi-axial strength of AS300in certain combinations. Increasing particle size
reduces the strength, while increasing pressure reduces the Weibull modulus in both particle
sizes. Reducing the impact angle to 30º has a positive effect on the strength with both particles.
Ravi K Chintapalli
174
Effect of sandblasting on mechanical properties
5.1.2 Effect of annealing
The two sandblasting conditions (110-2B-90º and 110-2B-30º) with significantly higher strength
than the control group were annealed at 1000 ºC for 1 hour and then tested for bi-axial strength.
The results for the strength are shown in figure 5.4 a) in comparison with the strength without
heat treatment. The mean biaxial strength of specimens subjected to the annealing treatment was
similar and decreased below the strength of the control group. The Weibull plot (figure 5.4 b)
shows that the two annealed groups have higher Weibull modulus than the control group. A
higher Weibull modulus indicates less statistical deviation due to a similar characteristic defect
in all the samples of these two groups.
1600
Mean Bi-axial strength (MPa)
1400
1.0
SB-90º Annealed at 1000 ºC, 1hr
SB-30º Annealed at 1000 ºC, 1hr
1200
b)
0.5
0.0
1000
ln(-ln(1-F))
-0.5
800
-1.0
-1.5
600
-2.0
400
-2.5
200
0
1.5
Polished
110 m
110 m
a)
m= 12
m= 14
m= 14
-3.0
Control
SB-90º
SB-30º
Surface treatments
-3.5
600
800
Control
110-2B-90º
110-2B-30º
1000 1200 1400 1600
Mean bi-axial strength (MPa)
Figure 5.4: Effect of annealing on after sandblasting on bi-axial strength a) mean biaxial strength and b)
Weibull plot.
5.1.3 Fractography
The critical defects for fracture of the material under bi-axial loading were observed by SEM.
Figure 5.5 shows the SEM micrographs of defects found in different treatment conditions. In
control sample (polished), a surface defect seems to be the cause for fracture. This defect could
be either from processing of material or introduced during grinding and polishing steps.
In sandblasted samples, two types of defects were found. Defects induced by the sandblasting
and microstructural defects. In case of 110-2B-90º, a deep cut was found at the surface which
was introduced by sandblasting. In 110-4B-90º and 250-4B-90º, large cracks which are induced
by sandblasting can be seen at the surface. Microstructural flaws such as porosity and large grain
clusters are found in other conditions 250-2B-90º, 110-2B-30º and 250-2B-30º. These defects are
Ravi K Chintapalli
175
Effect of sandblasting on mechanical properties
internal and found at a distance between 20-135 m from the surface in the previously
mentioned conditions.
Table 5.1 shows the estimated critical defect size for all conditions using the following equation
√ 5.1
Where KIC is the fracture toughness, σ is the fracture strength and c is the defect size and Y0 is
geometric constant equal to 1.29.
Figure 5.5: SEM micrographs of defects found in different sandblasting treatments.
Ravi K Chintapalli
176
Effect of sandblasting on mechanical properties
The defect size estimated is given in table 5.1. It is observed that the defect size increases with
the increase in particle size and pressure under impact angle 90º. The defect size is smaller than
the control sample when the impact angle is reduced to 30º irrespective of the particle size.
Table 5.1 Estimated surface defect size
Surface condition
Control
Defect size (m)
9.3
110-2B-90º
6.3
250-2B-90º
11.5
110-4B-90º
7.4
250-4B-90º
16.2
110-2B-30º
8.3
250-2B-30º
8.8
5.3 Elastic modulus and contact hardness
Nanoindentation was performed on sample sandblasted with condition 110-2B-90ºto obtain the
elastic modulus and contact hardness. The nanoindentations were done in the surface and in the
cross-section. First, sandblasted samples were carefully polished with colloidal silica (0.3m) to
remove the peaks that are formed by sandblasting until some smooth areas appeared on the
sample. These smooth areas were then indented at two different depths. Secondly, cross-sections
of the sandblasted samples were prepared by polishing to the required roughness of the
nanoindentation and then indented from the surface to interior up to 12m.
700
600
Load (mN)
500
400
300
200
100
0
0
300 600 900 1200 1500 1800
Displacement (nm)
Figure 5.6: Sample P-h curves of 110-2B-900 on the surface at different depths.
Ravi K Chintapalli
177
Effect of sandblasting on mechanical properties
Figure 5.6 shows the P-h curves for different depths investigated. The P-h curves demonstrate
usual behaviour during loading or unloading, which suggests that the curve was not affected by
the presence of cracks induced during sandblasting. Table 5.2 shows the results obtained from
the surface indentation data. At both depths the properties are found to be similar and also to that
of the control sample (refer to table 3.1 in chapter 3). However, at 2 m depth the standard
deviation is higher in both elastic modulus and hardness. This high scatter is due to the
surrounding uneven areas of the indent that were not completely removed during the careful
polishing step to ensure that no additional changes in the material are induced by polishing. This
local surface unevenness is believed to be the cause of raise in standard deviation.
Table 5.2 Properties obtained from surface of AS300 sandblasted with 110-2B-90º
0.05
Elastic
modulus
(GPa)
227±23
2
228±40
Depth
(m)
Hardness
(GPa)
17±3
16±5
The properties obtained from the measurements carried out on the cross-section are shown in
figure 5.7. The elastic modulus and hardness are practically constant in the cross-section and are
similar to that obtained from the surface. Both the properties obtained after sandblasting are
similar to the control specimens (refer to table 3.1). Therefore, it is evident that elastic modulus
and contact hardness did not change in spite of sandblasting.
300
20
Elastic Modulus (GPa)
19
250
18
225
17
200
16
175
15
150
0
2
4
6
8
10
Distance from surface (m)
Hardness (GPa)
Elastic modulus
Hardness
275
14
12
Figure 5.7: Elastic modulus and hardness on the cross-section of AS300 sandblasted with110-2B-900.
Ravi K Chintapalli
178
Effect of
o sandblastin
ng on mechaniical propertiess
5.4 Res
sidual stre
esses
It is weell known that
t
during sandblastinng compresssive residuual stresses are induceed in a thinn
surface layer1,6–8. In
I the preseence of a crrack induced
d by indenttation, thesee sandblastiing residuall
stresses act as to innduce an appparent fractture toughneess higher than
t
the actuual fracture toughness.
By takinng this in to
t account, residual strresses inducced by sanddblasting w
were estimatted on 110-2B-900usingthe
u
inndirect methhod of inddentation crrack; in othher words, residual stresses
s
aree
estimateed from thee changes inn the crack size in a strressed and stress-free state. The method
m
andd
the moddel used are already expplained in chapter
c
2.
Fiigure 5.8: Indentation crracks of 110-2B-90º durinng polishing steps.
Figure 5.8 shows the indentaations durinng differen
nt material removal
r
steeps. No craacks at thee
imprint edges weree found im
mmediately after
a
sandbllasting. Thiis is an inddication of presence
p
off
residuall stresses which
w
are noot allowing the crack to
t extend. Subsequent
S
ly, after each materiall
removall step, indeentation prooduced visibble cracks and
a their siize was incrreasing. Att around 122
m of depth,
d
crackk size is almost similaar to that in
n the polishhed sample (stress freee material).
Figure 5.9
5 a) show
ws the crackk size with respect to the
t materiall removed. The figure shows twoo
regions in the curvve. The firsst region coomprises fro
om 2-12m
m where craack size incrreases withh
materiall removal, but
b in the seecond regionn from 12-1
18m crackk size is stabble.
60
50
40
30
20
1
1.2
a)
Compressive stress (GPa)
Mean crack length (m)
70
0
4
8
12
16
6
Thickness removed (m)
m
20
b)
1
1.0
0
0.8
0
0.6
0
0.4
0
0.2
0
0.0
0
4
8
12
16
Thick
kness remove
ed (m)
20
0
Figurre 5.9: a) Meeasured crackk lengths at different
d
dep
pths and b) esstimated resiidual stressess (c>>d).
Ravi K Chintapalli
179
Effect of sandblasting on mechanical properties
Figure 5.9 b) shows the estimated residual stresses from a fracture mechanics analysis (equation
2.12 from chapter 2). The range of the stress is in between ~ 0-1.2 GPa, which is compressive in
nature. Two significantly different regions can be observed from the figure; in the first region the
compressive stresses ranging between 0.5 to 1.2GPa are located in the first 4m from the
surface. In the second region stresses between 0-0.5 GPa are found from 4 to 12microns. This
indicates that sandblasting under 110-2B-90ºconditions will induce compressive residual stresses
up to ~12 microns from the surface; however, the magnitude of stresses is much higher in the
first 4 microns.
The above results give a clear idea about the depth of the residual stresses in a specific zone in
terms of the indentation crack size. The properties of 3Y-TZP are influenced by the amount of
monoclinic phase formed during sandblasting since this is responsible for the residual
compressive stress. This can be also estimated based on the amount of phase transformation
using the following equation proposed by Green et al9 as mentioned in chapter 1. Nevertheless,
the equation is given here again for the convenience.
1 ∆
3
1
5.2
Where σc is the theoretical compressive residual stress, ΔV/V is the change in volume due to t-m
phase transformation (approximately 4%), E is the elastic modulus and ν=0.3 is the Poisson ratio
of zirconia. Vi is the monoclinic volume fraction measured by XRD.
Compressive stress,c (MPa)
900
750
600
450
300
150
0
0.00
0.04
0.08
0.12
0.16
Monoclinic volume fraction
Figure 5.10: Theoretical compressive residual stresses vs monoclinic volume fraction.
The compressive stresses calculated from equation (5.2) are shown in figure 5.10, where it can
be seen that residual stresses increase linearly to high values with a relatively reduced
Ravi K Chintapalli
180
Effect of sandblasting on mechanical properties
monoclinic volume fraction. Figure 5.11 shows the theoretical compressive residual stresses
calculated from equation (5.2) for all sandblasting conditions. Increasing particle size and
pressure increases the amount of monoclinic volume fraction, therefore, from equation 5.2 it can
be seen that they also increase the residual stress. Sandblasting at 30 º induces less residuals
stresses due to the low monoclinic fraction induced in comparison with sandblasting at 90º.
Residual compressive stress (MPa)
800
700
110 m
250 m
Pressure 2 bars
a)
110 m
250 m
b)
Pressure 4 bars
600
500
400
300
200
100
0
Control
SB-90
SB-30
Surface treatments
Control
SB-90
Surface treatments
Figure 5.11: Theoretical compressive residual stresses for different treatments, a) at pressure 2 bars, b)
at pressure 4bars.
5.5 Discussion
5.5.1 Bi-axial strength
The purpose of studying varied conditions of sandblasting on strength is to understand the effect
of test parameters such as particle size and pressure on the biaxial flexural strength as in
literature most of the investigations were made under a single sandblasting condition3,4,10–12 and
found different results. Only a few authors6,13 have used different conditions in the same study.
In this work different sandblasting conditions are used by varying particle size, pressure and
impact angle. While small particles (110 m) and low pressure (2 bars) condition increase the
mean strength large, particles (250 m) and high pressure (4 bars) reduce the strength in
comparison with control sample. In addition low impact angle (30º) slightly increases the
strength for sandblasting with both small and large particles as compared to control samples.
The failure of ceramics under bi-axial stress occurs from a particular defect that has the most
favourable orientation to the stress field under an externally applied load14. This particular defect
is assumed to be perpendicular to stress direction and is located in the maximum tensile stress
Ravi K Chintapalli
181
Effect of sandblasting on mechanical properties
region. The defects that are parallel to the stress direction within the maximum tensile stress
region and the defects that are outside this region irrespective of the orientation, have the least
influence on strength. The failure probability of ceramics is generally characterised by the
Weibull statistics. As mentioned earlier, a higher Weibull modulus indicates less scatter in the
strength data suggesting a single parameter defect population, whereas a lower Weibull modulus
indicates high scatter in the strength data suggesting multi parameter defect population.
From figure 5.2, Weibull modulus increases under sandblasting with small particle and low
pressure, while it decreases with large particles and high pressure. To understand this
observation it is important to emphasize here what change does sandblasting induces at the
surface. Sandblasting at 90º induces a localised stress region in the centre of the specimen, as the
nozzle and specimen are aligned to the centre, at 30º no localised stress region is generated due
to the sample tilting. Additionally, in both cases erosion of material occurs.
Weibull modulus increases significantly under 110-2B-90º, this increase can be attributed to the
high compressive stresses generated in the maximum tensile stress region. Moreover, due to the
erosion of material from this zone reduces the size of large cracks already present in the material,
which decreases the scatter in the strength. Some previous studies15–17 also reported that Weibull
modulus increases significantly in zirconia subjected to grinding and sandblasting, which they
have attributed to either elimination of large cracks or reduction in their size due to material
removal.
For sandblasting at 30º irrespective of the particle size, Weibull modulus is slightly higher
compared to control group. On the other hand, Weibull modulus decreases by sandblasting under
250-2B-90º and 250-4B-90º. In addition to generating residual stress and erosion of material,
sandblasting with large particles might induce cracks in different sizes, which increases the
scatter in the strength. In an earlier study, Wang et al18 found strength and Weibull modulus
decreases after grinding and sandblasting in CAD/CAM zirconia. It is suggested that grinding
and sandblasting further enhanced the milling induced surface flaws instead of generating any
counter acting effect, which decreased the strength and increased the scatter.
5.5.1.1 Effect of particle size and pressure
The results presented here are in line with those of the literature particularly in case of the
strength increase4,6,10–12,19–21,7. For instance, in several works of Kosmac et al4,12,19–21, the mean
Ravi K Chintapalli
182
Effect of sandblasting on mechanical properties
strength increased but with low Weibull modulus when sandblasted with 110 m particles at 4
bars on Y-TZP; similarly, in this work, it is found that the mean strength is increased with lower
Weibull modulus under similar sandblasting conditions.
Most of the previous works involved in sandblasting used 110 m particles or less in size.
However, to generate a high surface roughness larger particles are used. Although, large particles
(250 m) generate high surface roughness, the strength decreased significantly irrespective of
the other sandblasting conditions. With higher pressure the decrease in the strength is even more
severe. From above, it is clear that large particles decrease the strength due to the large defects
induced (see figure 5.5 250-4B-900).
The effect of sandblasting particle size is also explained by considering sandblasting as
simultaneously repeated multiple indentations mimicking the particles hitting the sample surface.
However, the type of the indentation depends on the particle shape. Figure5.12shows the
particles impact mimics both sharp and blunt indentations. Thus sandblasting produces both
sharp and blunt type impacts and generating defects at numerous sites on the sample surface and
larger particles generate large and deep defects. In case of larger particles local defects are either
large and/or connected to the surface which is detrimental to the strength. Whereas, in case of
small particles the local defects are smaller and might be contained in the region of high residual
stresses, which opposes crack propagation.
Figure 5.12: Fracture surfaces of 110-2B-900 a) mimic of sharp indentation and b) mimic of blunt
indentation.
Another important test parameter is sandblasting pressure, as mentioned earlier in the literature
pressures from 4 bars and above was used. However, most of the works1,6,7,21,22 have done
sandblasting using only pressure condition. In this study, two pressure conditions were studied (2
Ravi K Chintapalli
183
Effect of sandblasting on mechanical properties
and 4 bars).Under both cases strength and Weibull modulus increases compared to control, but
the effect is more at 2 bars than 4 bars.
It can be suggested that, particles projected at low pressure have low kinetic energy, which
induces damage mostly limited to plastic deformation around the impact zone. As mentioned in
chapter 1, that particle velocity is directly proportional to erosion of material and increasing the
impact pressure will increase the particle velocity. At high velocities, the kinetic energy of the
particles is high; due to this the tensile stresses around the impact zone initiate lateral and radial
cracks23. In an ideal situation, all the particles projected at same pressure have the same kinetic
energy if the particles are of exactly same size and shape. But in practice, particles do not
possess the same size and shape; due to this their kinetic energy is believed to be different.
Additionally, at high impact velocities particle rebounding increases and the rebounded particles
hit the surface again with much lower velocity. Therefore, the type of the crack induced at a
specific site depends on kinetic energy and shape of particle hitting that site.
From the above observations, it is believed that the magnitude of increase in strength and
Weibull modulus of zirconia at low pressures is higher because of not inducing any critical
defects in the maximum tensile stress region.
5.5.1.2 Effect of impact angle
The results show that, with respect to the control samples, the strength increases significantly for
110-2B-90º, but only slightly for 110-2B-30º. However, for larger particle size the strength
decreased for 250-2B-90º, while increased slightly for 250-2B-30º. It is clear that the sample
orientation has a clear effect i) for small particles both impact angles have a positive effect on
strength; although 90º has more positive effect than 30º, ii) for large particles, the effect is
negative for impact angle of 90º but positive at 30º.
90º
Sharp
Blunt
Particle
Sample
30º
Figure 5.13: Particle-sample interactions with different orientations.
Ravi K Chintapalli
184
Effect of sandblasting on mechanical properties
The above behaviour can be explained with a simple illustration of sample particle interaction
shown in figure 5.13. As mentioned previously particle impact can be considered as numerous
indentations, in this sense at 90º a single particle impact can be considered as indentation while
at 30º a combination of indentation and scratch. In each case the strength governing
microstructural changes depends on the corresponding stress field. Muñoz et al24 reported the
monoclinic phase distribution under Vickers indentation and scratch. They concluded that under
indentation, highest monoclinic volume is found at a depth equal to half of the diagonal of the
imprint. Whereas, under scratch highest monoclinic volume is found on the surface. This
explains that, for small particles, the high residual stresses are located slightly deeper at 90º
compared to 30º which gives more positive effect on strength in the former than the later. For
the large particles, it is contrary, because the strength depends on the defect size and the depth of
residual stress. This will be explained in the next section in detail.
5.5.1.3 Effect of phase transformation
Figure 5.14 shows the effect of the amount of monoclinic phase on biaxial strength. Strength
increased with monoclinic phase up to 9%. While strength shows an irregular behaviour between
9 and 14% of monoclinic phase, however, it decreased depending on the severity of the
sandblasting condition. This is in agreement with the result of Ban et.al11, who reported that after
sandblasting, the strength increased with monoclinic volume up to 10% and above decreased.
1800
Control
110-2B-90º
110-2B-30º
110-4B-90º
250-2B-90º
250-2B-30º
250-4B-90º
Biaxial strength (MPa)
1600
1400
1200
1000
800
600
0
2
4
6
8
10
12
14
16
Monoclinic volume fraction % Vm
Figure 5.14: Relationship between strength and amount of monoclinic phase.
According to Ban et al7 the increase in strength is limited to certain extent of m-phase, and a
further increase in m phase will induce excessive micro cracking due to increase of volume11.
Ravi K Chintapalli
185
Effect of sandblasting on mechanical properties
However, in this work the surface treatments induced an amount of monoclinic phase in between
9 and 13%. As this difference is rather small, and large defects induced by large sandblasting
particles are believed to be more critical for the decrease of strength.
5.5.1.4 Influence of the residual stress
Sandblasting induces compressive stresses in the near surface region of the specimen. These
residual stresses influence the fracture toughness and in turn the strength of the material. The
material toughness then becomes the apparent fracture toughness plus the residual stress
intensity factor.
5.3
If a specimen is tested using Vickers indentation in a sandblasted surface, the indentation crack
length that results is shorter than in the same material before sandblasting (see figure 5.9 a), so
there is an apparent rise in fracture toughness. Therefore, the residual stress intensity factor, KRes
(negative) should be added to the Vickers residual stress intensity factor induced by indentation,
KIndA, so that the crack will grow until the equilibrium condition is reached, that is,
5.4
In addition, the exact value of the apparent fracture toughness will change along the crack front
following the change in the residual stress depth profile, which depends itself of the depth of the
crack tip. Therefore, for a surface semicircular flaws (see Fig.5.15), the total stress intensity
factor may have different values at the surface and at the deepest point, even if the surface effect
is neglected25.
a r ind (a)
r b)
Figure 5.15: a) Schematic representation of the residual stresses along the crack depth b) Surface
elliptical crack showing the definition of the different geometrical terms25.
Ravi K Chintapalli
186
Effect of sandblasting on mechanical properties
As a consequence, during indentation the flaw will tend to grow more at the deepest point (A)
than at the surface (point B of figure 5.15(b)), since at the surface the compressive residual stress
is maximum. If the indentation crack length is only measured at the surface and it is assumed to
be semicircular, the apparent indentation fracture toughness may be overestimated if the stresses
are compressive. As mentioned earlier, that point B is constrained by the residual compressive
stress at the surface and the maximum tensile stress is located at point A, so the specimen might
fail from this point. In this sense assuming a straight crack if fairly sufficient.
i) Uniform residual stress
However, in a first step and in order to simplify the treatment, it will be assumed that the residual
stress distribution is uniform over the depth of the surface cracks, so that a rough estimation of
the stress intensity factor for a semicircular crack can be given as26:
√ 5.5 Where c is the constant residual stress. This approximation describes well the situation for
which the crack length is much shorter than the thickness of the surface transformation layer. For
a constant stress field,YRes= Y0= 1.128, so that
√ 5.6
ii) Gradient residual stress
The residual stress, c, in the depth direction (perpendicular to the surface) usually decreases
until reaching a constant value, equal to 0 at a depth d. For a stress field with a gradient the stress
intensity factor for a surface crack depends on the crack depth (c) and d (depth of the stress
field). The residual stress field is given by the expression (A4) in the appendix. In this case the is
an apparent fracture toughness given by
| |√ 5.7
For the gradient residual stress, YRes is function that depends on the ratio of crack lent and the
depth of the stress field c/d. The expressions of YRes for c<d and c>d are given in the
appendix.Dividing the equation 5.7 with c√d gives
Ravi K Chintapalli
187
Effect of sandblasting on mechanical properties
5.8
√
1.0
0.8
0.6
K
A
IC
 K IC
c d
0.4
0.2
c<d
0.0
0.0
0.5
c>d
1.0
1.5
2.0
2.5
3.0
c
d
Figure 5.16: Change in fracture toughness with increasing crack size caused by residual stress gradient.
The equation 5.8 is plotted in figure 5.16. Figure 5.16 show that the factor KICA-KIC increases and
reaches to maximum when c=0.52d and then the apparent fracture toughness decreases with
increasing crack size.
iii) Indentation crack length under compressive stress
Indentation crack length under residual stress can now be estimated. For Palmqvist indentation
cracks, the fracture toughness is given by the equation of Niihara27, which can be written as
.
5.9
8
Where
.
0.025
Ravi K Chintapalli
5.10
188
Effect of sandblasting on mechanical properties
Where P is the indentation load, H is the hardness, E is the elastic modulus and c0 is the
indentation crack length without residual stress. By substituting equations (5.5) and (5.9) in
equation (5.3)
.
√
.
8
8
.
√
5.11
5.12
Simplifying equation 5.12
1 5.13
a) Constant stress field
In this case YRes= Y0, therefore
1
.
5.14
Where
≡
0
Since
5.15
Substituting equation 5.15 in 5.14 gives
.
1
0 5.16
Further solving the equation 5.16 gives
Ravi K Chintapalli
189
Effect of sandblasting on mechanical properties
1
.
4
1
2
5.17
The above equation can be useful for estimation of how the indentation crack length is affected
by the level of compressive residual stresses. This expression is plotted in figure 5.17. It can be
observed that crack length decreases with increasing residual stress intensity factor, and when
c=c0the residual stress intensity factor is zero.
1.0
0.8
c/c0
0.6
0.4
0.2
0.0
0.0
0.4
0.8
1.2
o
K
1.6
2.0
/KIC
Res
Figure 5.17: Effect of stress intensity factor on the indentation crack.
b) Gradient stress field
If it is assumed that c >> d, and using equation (A8) given in the appendix, then the indentation
crack length expected in a material with compressive residual stress can be calculated in the
same manner shown before for a constant stress field. However, now for a gradient stress field
when c >> d, from appendix it can be seen that:
√
√
√
/
√
1
/
5.18
Then equation 5.13 becomes
Ravi K Chintapalli
190
Effect of sandblasting on mechanical properties
1
.
/
√
1
/
0 5.19
Simplifying the above equation
1
1
1
5.20
√
2
|
| 1
√
| |
2
1
5.21
Considering that it has been assumed that c/d = s= constant we have thatc d=s*d2 so that the
ratio c/c0 should be proportional to d2.
1
2
1
2
1
5.22
Where d0 is the depth of the stress filed and x is the thickness of the material removed. Figure
5.18 shows the plot of equation 5.22 with the experimental values. The plot shows that, with
increasing x the crack length c increases and c becomes c0 when x is equal to d0.
The
experimental data is fitted with equation 5.22with the constant value ~ 0.91. This value is
reasonable since
2
1
2
1
2
By taking representative values of the above parameters:
Ravi K Chintapalli
1
0.46 5.23
and
6
.
191
Effect of sandblasting on mechanical properties
1.0
Experimental
Fit Eq (5.22)
0.8
c/c0
0.6
0.4
0.2
0.0
0.0
0.2
0.4
0.6
0.8
1.0
x/d0
Figure 5.18: Change in crack length with respect to the ratio of thickness removed to the depth .
Let us now consider the contribution of sandblasting induced residual stresses to the strength of
zirconia. Before sandblasting, the specimen has a natural crack size c0.But sandblasting induces
defects that can be larger or smaller than the natural critical defect size and which are at least
partly inside the compressive residual stress field. Therefore, the increase or decrease of strength
depends on the size of the crack c and the stress level induced by sandblasting. For a crack
subjected to residual stress, by using the equation proposed by Green and Maloney28 for a linear
stress gradient (developed in the appendix), see for example Lawn and Marshall29, equation 5.3
leads to
| |
√ √ 5.24 YRes is given as a function of (c/d) (see appendix, equation A6 and A7). Dividing the above
equation with 0 Y0 and further solving gives
| |
5.25
Where and c are the strength and crack size of the material after sandblasting, 0andc0 are the
strength and crack size of polished specimens.c is the maximum compressive residual stress.
Ravi K Chintapalli
192
Effect of sandblasting on mechanical properties
Figure 5.19 shows the effect of residual stress and crack size on strength based on equation
(5.25) for different maximum compressive stresses. Figure 5.19 (i) show the effect on strength
considering that crack the critical natural crack does not change by sandblasting. In this case the
strength of the sandblasted specimens is always higher than the polished specimens. The
magnitude of the strength increase is high when c0/d< 1 depending on the compressive residual
stress level.
Figure 5.19 (ii), (iii) and (iv) shows the effect on strength in case sandblasting induces defect
sizes larger than those present in polished specimens. This is the case sometimes since annealing
after sandblasting (removing residual stresses) diminishes the strength below that of polished
specimens. Three cases are presented here. For c = 1.25c0, it can be appreciated that, the strength
of sandblasted specimens for c0 = d starts to be equal to the strength of polished specimens (for
c=-0.50 and declining when c0> d in comparison with the polished samples. It is clear that
strength decreases with increasing crack size c and c0/d.
4
I
c
I/
0
0
I
c
I
=1
I/
c
0
I/
0
1
i)
c=1.25c0
.5
=1
I/  0
I c
.5
=1
I/  0
I c
3
2
c=c0
=0
.5
I I
c
IcI/0 =0
ii)
=1
/ =
0
0.5
IcI/0 =0
0
4
I
0
c
I/
0
I I
c
1
0
iii)
I
=1
/ =
0 0.
c
I/
0
I I
5
c
IcI/0 =0
0
iv)
c=2c0
.5
=1
I/  0
I c
.5
=1
I/  0
I c
3
2
c=1.75c0
=1
/ =
0
0.5
IcI/0 =0
1
2
c0/d
3
4
0
1
2
3
4
c0/d
Figure 5.19: Effect of residual stress and defect size on strength.
Ravi K Chintapalli
193
Effect of sandblasting on mechanical properties
From table 5.1, for four sandblasting conditions (110-2B-90º, 110-4B-90º110-2B-30º and 250-2B30º) the estimated defect size is smaller than the control (polished) specimens. This can be
explained from figure 5.19(i) where c=c0, that is, sandblasting did not change the critical crack
size. In this sense, the strength increases after sandblasting with respect to the polished-control
specimens. The increase in strength depends on the compressive stress level irrespective of
c0/din comparison with polished specimens for which c is zero.
For sandblasting conditions 250-2B-90º and 250-4B-90º, the decrease in strength observed may
be estimated by considering that now the defect size is larger than in the control specimen. From
figure 4.13 in chapter 4, the transformation depth for 250-2B-90º is found to be ~13 m.
Therefore, for 250-2B-90º, if the crack increases to c=1.25c0 after sandblasting from figure 5.19
(ii) with c0/d ~ 1the strength may already be similar to the strength of control specimens.
Since the amount of phase transformation is similar in 250-2B-90º and 250-4B-90º (refer to
figure 4.9 in chapter 4), the depth of the residual stress might also be similar. In this sense, for
250-2B-90º the parameters are c=1.23c0 and c0/d =0.88 without considering the effect of residual
stress (see table 5.1). By considering the effect of residual stress the above parameters may be
about 30% higher, since the residual stress affects the defect size nearly 30% compared to
control specimen. Then the parameters become c=1.7c0 and c0/d =1.15. For 250-4B-90º the
parameters are c=1.75c0 and c0/d =1.25 without the effect of residual stress and c=2.25c0 and
c0/d =1.62 with the effect of residuals tress. For the said c and c0/d values, the figure 5.19 (ii)
and (iii) show that the strength is decreased compared to control specimen.
However, it should be note that the model presented here has a certain limitation in the form of
linear stress gradient, as in most cases this is considered for the sake of simplicity. But figure 5.9
(b) shows that the stress gradient is not linear for the one sandblasting condition that has been
studied here. Moreover, from the chapter 4, the phase transformation maps show that the
transformation is not liner along the depth and also not homogenous along the surface. Figure
5.20 explains the deviation of liner stress gradient model from the experimental data.
Ravi K Chintapalli
194
Effect of sandblasting on mechanical properties
Compressive stress (GPa)
1.2
Experimental
1.0
0.8
L1
0.6
0.4
0.2
0.0
L2
0
2
4
6
8
10
12
14
Thickness removed (m)
Figure 5.20: Residual stresses with respect depth and linearity.
Let us assume a similar stress profile for all sandblasting conditions, however, with slight
differences in the stress levels. Figure 5.20 shows experimental stress values with two linear fits
L1 and L2. If we consider d ~ 12 m, the linear stress gradient assumes the stress levels at L1. But
in fact the experimental stress levels are less compared to L1. The strength decreases at lower
residual stress levels. Whereas, the depth is much less than the experimental data for stress
gradient L2. If d decreases c0/d increases and the strength decreases. From the above mentioned
discrepancies in the stress gradient, the above presented model is in partial agreement with
experimental results.
5.5.1.5 Effect of annealing
For specimens treated with 110-2B-90º and 110-2B-30º, the mean strength decreased
significantly after heat treatment compared to sandblasted group but only slightly in comparison
with the control group. Upon heating, monoclinic zirconia will be transformed back to tetragonal
phase (refer chapter 4) and the stresses in the material tend to relax, which decreases or
eliminates the contribution of residual stresses to the strength.
The above observations are consistent with those reported in literature6,11,13,21. It is reported that
the strength decreased in sandblasted and grounded samples after heat treatment, although it did
not practically change compared to the control group. They suggested that the sandblasting flaws
are not detrimental to the strength of zirconia even after residual stresses are released. This can
Ravi K Chintapalli
195
Effect of sandblasting on mechanical properties
be explained by; sandblasting did not induce any additional defects (for the said conditions) and
reduces the critical flaw size due to erosion of material. Therefore, heat treatment only removed
the contribution of residual stresses, and the strength came back to original state (strength of
control group) with slight increase in Weibull modulus.
Similarly, for Y-TZP ceramics subjected to abrasive grinding, Ho et.al15 reported that strength
increases after grinding due to residual stresses up to 1 GPa, and decreased after annealing
(11000 C, 2 hr), but then it is similar to the control group. This indicates that grinding did not
induce any additional defects. They showed that higher residual stress implies higher strength,
with a possible reason that most of the critical flaws are located inside the stressed region. If the
tip of the flaw is free from any stress, the effect on the strength will be in contrary.
Figure 5.21 shows the crack under different stress situations for various treatments. From the
figure 5.21, sandblasting under these conditions (110-2B-90º and 110-2B-30º) induced only
residual compressive stresses but not any additional cracks. These residual stresses oppose the
crack extension under loading, which increases the strength. Annealing of sandblasted specimens
only relaxed the compressive stresses bringing back the crack under same situation like in
polished specimens. Strength of polished and sandblasted plus annealed specimens is then
practically same. However, this is the case observed here for the above said sandblasting
conditions, but in case of severe sandblasting conditions the strength changes must be
experimentally determined before any conclusions can be drawn.
Compressive stress
zone
Crack subjected to tensile stress in a
polished specimen
Crack subjected to compressive
stress in a sandblasted specimen
Crack subjected to tensile stress in a
sandblasted + annealed specimen
Figure 5.21: Influence of stress state on a crack under different treatments (crack perpendicular to the
tensile stress direction). Another important aspect in annealing is crack healing, if annealing temperatures are close to the
sintering temperatures. In this work the annealing temperature (1000º C) is well below the
sintering temperature (1350º C) and so crack healing does not happen. Therefore, it is believed
that in case of 110-2B-90º and 110-2B-30º, after annealing the strength is nearly similar to
Ravi K Chintapalli
196
Effect of sandblasting on mechanical properties
control group as no crack healing occurs; which, otherwise the strength would have been higher
than control group.
5.5.2 Elastic modulus and hardness
Elastic modulus and hardness changes are very sensitive to the presence of microcracks and
porosity (refer chapter 3). Gaillard et.al30have reported that microcracks are the reason for
reduction elastic modulus in hydrothermally degraded zirconia, as they observed microcracks
below a nanoindentation by making FIB trenches. With respect to sandblasting, no previous
studies have shown the evidence of microcracks. However, it is often hypothetically reported in
literature that sandblasting defects have the nature of microcracks12,21. Zhang et.al1 found
reductions in elastic modulus by nanoindentation in zirconia. They believed that the reduction in
modulus is linked to the microcracks but with no microscopical observations.
In this work, nanoindentation results show that there is no change in elastic modulus and
hardness after sandblasting with 110-2B-900 either at the surface or in interior up to 10 microns.
FIB-SEM observations for the specimen sandblasted with same condition have shown few
scattered microcracks (refer to figures 4.20 & 4.21 in chapter 4). Additionally these microcracks
are located very close to the surface. Nanoindentations done at depths of 0.5 and 2 m are
probably not affected by these scattered microcracks. Therefore these are not detrimental to the
elastic modulus and hardness as observed in nanoindentation results and it can be appreciated
from the above results that sandblasting with 110-2B-90º has no significant effect in the
material´s elastic modulus and contact hardness.
5.5.3 Residual stresses
Surface residual stresses play a significant role in determining the contact damage in Y-TZP
ceramics. As there are several ways to introduce residual stresses, the predominant method in
zirconia is to induce phase transformation. In zirconia ceramics, a common phenomenon is,
phase transformation will induce compressive residual stresses due to volume increase of
transformed grains31. It is already explained that sandblasting induces phase transformation in
the surface and up to around 10-13 microns into depth.
Figure 5.9 b) shows the residual stress profile, which indicates that stresses are more intense in
near surface but decreasing in depth. The phase transformation shown in chapter 4 is not
homogenous in depth, also the FIB characterizations show partially transformed grains is an
Ravi K Chintapalli
197
Effect of sandblasting on mechanical properties
indication that those grains received less sandblasting impact were transformed less. As a result
of this, a non linear behaviour in residual stress profile is seen.
Sato et al6 has estimated the theoretical residual stresses based on the phase transformation in YTZP after sandblasting, are in order of 381 to 417 GPa, which are slightly lower than observed
here for similar sandblasting conditions (110-2B-90º). This is because of slightly lower phase
transformation reported by them compared to those observed here.
5.6 Summary
Effect of sandblasting on mechanical properties of 3Y-TZP is studied and the main results are
summarized as follows.
 The best sandblasting conditions for strength are 110-2B at 90º and 30º impact angle.
 Air pressure more than 2 bars decreases the reliability of strength at high angle 90º
sandblasting.
 Large particles (250 m) decreases the strength significantly at high angle (90º)
sandblasting while slightly increases the strength at low angle (30º).
 Annealing at 1000 0C after sandblasting eliminates the contribution of residual stress for
the studied sandblasting conditions.
 Elastic modulus and contact hardness are not affected under sandblasting condition (1102B-900).
 Non linear compressive residual stresses are present up to first 12 µm from the surface.
Ravi K Chintapalli
198
Effect of sandblasting on mechanical properties
5.7 References
1. Zhang Y, Lawn BR, Rekow ED, Thompson VP. Effect of Sandblasting on the Long-Term
Performance of Dental Ceramics. Journal of Biomedical Materials Research. Part B, Applied
Biomaterials. 2004;71B:381-386.
2. Guazzato M, Quach L, Albakry M, Swain MV. Influence of Surface and Heat Treatments on the
Flexural Strength of Y-TZP Dental Ceramic. Journal of Dentistry. 2005;33(1):9-18.
3. Albakry M, Guazzato M, Swain MV. Effect of Sandblasting, Grinding, Polishing and Glazing on the
Flexural Strength of two Pressable All-Ceramic Dental Materials. Journal of Dentistry. 2004;32(2):91-99.
4. Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. The Effect of Surface Grinding and
Sandblasting On Flexural Strength and Reliability of Y-TZP Zirconia Ceramic. Dental Materials.
1999;15:426-433.
5. Curtis AR, Wright AJ, Fleming GJP. The Influence of Simulated Masticatory Loading Regimes on the
Bi-Axial Flexure Strength and Reliability of a Y-TZP Dental Ceramic. Journal of Dentistry.
2006;34:317-325.
6. Sato H, Yamada K, Pezzotti G, Nawa M, Ban S. Mechanical Properties of Dental Zirconia Ceramics
Changed with Sandblasting and Heat Treatment. Dental Materials Journal. 2008;27(3):408-414.
7. Ban S. Reliability and Properties of Core Materials for All-Ceramic Dental Restorations. Japanese
Dental Science Review. 2008;44:3-21.
8. Denry I, Kelly JR. State of the Art of Zirconia for Dental Applications. Dental Materials.
2008;24(3):299-307.
9. Green DJ, Lange FF, James MR. Factor Influencing Residual Surface Stresses due to a Stress-Induced
Phase Transformation. Journal of American Ceramic Society. 1983;66(9):623-629.
10. Guazzato M, Albakry M, Quach L, Swain MV. Influence of Grinding, Sandblasting, Polishing and
Heat Treatment on the Flexural Strength of A Glass-Infiltrated Alumina-Reinforced Dental Ceramic.
Biomaterials. 2004;25(11):2153-2160.
11. Ban S, Sato H, Suehiro Y, Nakanishi H, Nawa M. Effect of Sintering Condition, Sandblasting and
Heat Treatment on Biaxial Flexure Strength of Zirconia. Key Engineering Materials. 2008;361-363:779782.
12. Kosmac T, Oblak C, Jevnikar P, Funduk N, Marion L. Strength and Reliability of Surface Treated YTZP Dental Ceramics. Journal of Biomedical Materials Research. 2000;53:304-313.
13. Sato H, Ban S, Nawa M, Suehiro Y, Nakanishi H. Effect of Grinding, Sandblasting and Heat
Treatment on the Phase Transformation of Zirconia Surface. Key Engineering Materials. 2007;330332:1263-1266.
14. Munz D, Fett T. Ceramics- Mechanical Properties Failure Behaviour, Material selection. New York:
Springer; 1998:170-175.
15. Ho C-J, Liu H-C, Tuan W-H. Effect of Abrasive Grinding on the Strength of Y-TZP. Journal of the
European Ceramic Society. 2009;29(12):2665-2669.
Ravi K Chintapalli
199
Effect of sandblasting on mechanical properties
16. Papanagiotou HP, Morgano SM, Giordano RA, Pober R. In Vitro Evaluation of Low-Temperature
Aging Effects and Finishing Procedures on the Flexural Strength and Structural Stability of Y-TZP
Dental Ceramics. The Journal of Prosthetic Dentistry. 2006;96(3):154-164.
17. Tuan WH, Kuo JC. Contribution of Residual Stress to the Strength of Abrasive Ground Alumina.
Journal of European Ceramic Scociety. 1999;19:1593-1597.
18. Wang H, Aboushelib MN, Feilzer AJ. Strength Influencing Variables on CAD/CAM Zirconia
frameworks. Dental Materials. 2008;24(5):633-638.
19. Kosmac T, Dakskobler A, Oblak C, Jevnikar P. The Strength and Hydrothermal Stability of Y-TZP
Ceramics for Dental Applications. International Journal of Applied Ceramic Technology. 2007;4(2):164174.
20. Kosmac T, Oblak C, Jevnikar P. The Fracture and Fatigue of Surface-Treated Tetragonal Zirconia (YTzp) Dental Ceramics. Materilas and Technology. 2007;41(5):237-241.
21. Kosmac T, Oblak C, Marion L. The Effects of Dental Grinding and Sandblasting on Ageing and
Fatigue Behavior of Dental Zirconia (Y-TZP) Ceramics. Journal of the European Ceramic Society.
2008;28:1085-1090.
22. Curtis AR, Wright AJ, Fleming GJP. The Influence of Surface Modification Techniques on the
Performance of a Y-TZP Dental Ceramic. Journal of Dentistry. 2006;34(3):195-206.
23. Slikkerveer PJ, Bouten PCP, De Haas FCM. High Quality Mechanical Etching of Brittle Materials by
Powder Blasting. Sensors And Actuators A:Physical. 2000;85:296-303.
24. Muñoz-Tabares JA, Jiménez-Piqué E, Reyes-Gasga J, Anglada M. Microstructural Changes in 3YTZP Induced by Scratching and Indentation. Journal of the European Ceramic Society. 2012. in press
25. Fett T. Estimation of Stress Intensity Factors for Semi-Elliptical Surface Cracks. Engineering
Fracture Mechanics. 2000;66(4):349-356.
26. Newman JC, Raju IS. An Empirical Stress Intensity Factor Equation for the Surface Crack.
Engineering Fracture Mechanics. 1981;15:185-192.
27. K. Niihara. A Fracture Mechanics Analysis of Indentation-Induced Palmqvist Crack in Ceramics.
Journal of Materials Science Letters. 1983;2(5):221-223.
28. Green DJ, Maloney BR. Influence of Surface Stress on Indentation Cracking. Journal of American
Ceramic Society. 1986;69(3):223-225.
29. Lawn BR, Marshall DB. Contact Fracture Resistance of Physically and Chemically Tempered Glass
Plates. Physical Chemistry of Glasses. 1977;18(1):7-18.
30. Gaillard Y, Jimenez-Pique E, Soldera F, Mucklich F, Anglada M. Quantification of Hydrothermal
Degradation in Zirconia by Nanoindentation. Acta Materialia. 2008;56:4206-4216.
31. Hannink RHJ, Kelly PM, Muddle BC. Transformation Toughening in ZrO2-Containing Ceramics.
Journal of the American Ceramic Society. 2000;83(3):461-487.
Ravi K Chintapalli
200
Hydrothermal degradation after sandblasting
Chapter
6
Hydrothermal degradation after
sandblasting
Ravi K Chintapalli
201
Hydrothermal degradation after sandblasting
6.1 Introduction
It is well known that 3Y-TZP ceramics in the annealed or as sintered condition suffer time
dependent hydrothermal degradation in aqueous environments1. In prosthetic dentistry
applications, the surface of 3Y-TZP of dental implants and crowns is sandblasted before
implantation for better adhesion2,3.
In case of implants, degradation could begin as soon as they come in contact with the oral fluids,
as the implants are directly fitted to the bone. On the other hand, crowns are fitted to the
abutments using a luting agent, which was believed to work as a protective layer to the ceramic
core from the surrounding oral fluids. However, recent studies have reported that, commonly
employed luting agents or cements absorb water via dentine tubules4, thus offering no protection
to the ceramic core.
Therefore, it is necessary to study and understand the hydrothermal degradation behavior after
sandblasting and its influence over mechanical properties of 3Y-TZP ceramics. In addition, in
the case of crowns, they are veneered to porcelain with a treatment at high temperature after
sandblasting. Therefore, it is of interest to study the effect of heat treatment on sandblasted 3YTZP since this is a standard practice.
Moreover, it has been shown in previous chapters that zirconia with grain size smaller than
300nm suffered severe hydrothermal degradation with a decrease inelastic modulus and contact
hardness. While Zirconia with grain size less than 300 nm is resistant to degradation. Having
discussed various aspects of hydrothermal degradation and its preclusion in nanocrystalline
materials (Chapter 3), this chapter will focus on the behavior of sandblasted materials after
degradation.
6.2 Hydro thermal degradation of after sandblasting
6.2.1 Phase transformation
Figure 6.1 shows the X-ray diffraction patterns of different materials treated under several
conditions. The phase changes in the material after polishing and sandblasting were already
discussed in chapter 4; nevertheless they are shown here for comparison with further treatments.
In case of AS300 after SB (sandblasting)+HD (hydrothermal degradation) a large monoclinic
peak appears at 2theta angle of 28.2 and 31.5º suggesting there is a significant increase in phase
Ravi K Chintapalli
202
Hydrothermal degradation after sandblasting
transformation compared to SB only. Therefore sandblasted specimens are not immune to
hydrothermal degradation.
Now, it is interesting to compare the hydrothermal degradation behavior of zirconia after
sandblasting and annealing treatment (SB+HT) in order to recover the tetragonal structure. In
this sense, it should be recalled from chapter 4 that SB+HT triggered reverse m-t transformation
turning the zirconia into full tetragonal state. For SB+HT+HD, transformation induced by water
vapor takes place as expected and the intensity of the monoclinic peaks is larger than after
SB+HD, so that sandblasted specimens are more resistant to degradation than specimens
annealed after sandblasting.
i)
Intensity
SB+HT+HD
AS300
m
t
m
m
t t
ii)
t
m
S120
t t
iii)
t
3YTZP-2CNT
tt
m
SB+HD
SB
Polished
24 26 28 30 32 34 36
2theta
24 26 28 30 32 34 36
2theta
24 26 28 30 32 34 36
2theta
Figure 6.1: X-Ray diffraction patterns after various treatments i) AS300, ii) S120 and iii) 3YTZP2CNT.[SB- Sandblasted (110-2B-900); HT- Heat treated (1000 0C); HD- Hydrothermally degraded (100 hours)]
In nanocrystalline materials S120 and 3YTZP-2CNT after a SB+HD treatment, a slight increase
in monoclinic peaks is clearly visible from figure 6.1 ii) & iii) compared to the specimen with
only SB treatment. This is an indication that transformation has progressed during hydrothermal
degradation in the sandblasted nanocrystalline materials. On the contrary, no monoclinic peaks
are found after hydrothermal degradation of SB+HT specimens, suggesting no phase
transformation takes place.
Figure 6.2 shows the quantity of monoclinic volume fraction with respect to the different
treatments for all the materials. The amount of monoclinic phase for each separated treatment,
that is, polishing, SB and HD are already given in chapters 3 and 4. It should be recalled that
Ravi K Chintapalli
203
Hydrothermal degradation after sandblasting
after polishing all zirconias studied are fully tetragonal with negligible amount of monocline
phase. SB induced monocline phase between 12-15 % irrespective of the material. Whereas, HD
(100 hours) induced about ~80% of monoclinic phase in AS300 and no phase transformation in
S120 and 3YTZP-2CNT.
90
Monoclinic volume fraction,% vm
75
AS300
S120
3YTZP-2CNT
0
SB- Sandblasted (110-2B-90 )
HD- Hydrotermal degradation (100 hours)
0
HT- Heat treated (1000 C,1 hour)
60
45
30
15
0
Polished
SB
HD
SB+HD
SB+HT+HD
Treatments
Figure 6.2: Monoclinic volume fractions with respect to different treatments.
For specimens hydrothermally degraded after sandblasting, that is, in the SB+HD condition, the
pre-existing monoclinic phase increased irrespective of the grain size. However, the increase is
higher in AS300 (larger grain size) than in the other zirconias with smaller grain size. The
difference is much larger if we compare the materials in the SB+HT+HD condition, since the
amount of monoclinic phase induced in AS300 is about ~70% while no phase transformation is
produced in S120 and 3YTZP-2CNT.
In all treatments involving hydrothermal degradation, high amounts of monoclinic phase are
found only in AS300. Comparing the treatments, hydrothermal degradation alone has more
significant effect in phase transformation than other any other treatment such as SB+HD and
SB+HT+HD. AS mentioned earlier, due to the nanometric grain size, pristine S120 and 3YTZP2CNT are resistant to hydrothermal degradation. However, these materials have transformed
slightly from tetragonal to monoclinic under externally applied stress (sandblasting). This preRavi K Chintapalli
204
Hydrothermal degradation after sandblasting
existing monoclinic phase increased during hydrothermal degradation, however, significantly
much less compared to AS300. The reasons for this will be discussed later in the chapter.
6.3 Change in mechanical properties
6.3.1 Bi-axial strength
The bi-axial strength after hydrothermal degradation was studied in sandblasted AS300 (see
figure 6.3) with different treatments. It should be recalled from the chapter 5 that strength and
Weibull modulus of all the conditions were compared to the control samples.
For two sandblasting conditions presented in figure 6.3, the strength increased significantly after
SB. But after SB+HD treatment, the strength decreased significantly in both conditions as
compared to only sandblasted specimens, being even slightly lower than that of the control
samples. Figure 6.3 b) shows the Weibull plots for the respective treatments. For SB+HD
specimens, Weibull modulus remains unchanged in 110-2B-90º and increased for 110-2B-30º. It
is interesting to notice that after hydrothermal degradation the strength decreases while Weibull
modulus either unchanged or increases, which implies that the scatter in the strength after
hydrothermal degradation is less.
1800
Polished
110-2B-90º
110-2B-30º
HD : 100 hours
a)
1400
1
b)
0
1200
ln(-ln(1-F))
Mean bi-axial strength (MPa)
1600
Control
m=12
110-2B-90º m=21
110-2B-30º m=14
110-2B-90º+HD m=19
110-2B-30º+HD m=20
1000
800
600
-1
-2
400
200
0
-3
Control
SB
Treatments
SB+HD
600
800
1000 1200 1400 1600 1800
Mean bi-axial strength (MPa)
Figure 6.3: Effect of different treatments on the bi-axial strength of AS300 a) bi-axial strength, b) Weibull
plot.
Figure 6.4 shows the fracture surface of a sample treated with SB+HD. Two distinguishable
layers are observed: i) fully surface degraded layer of 10±2 m thickness with intergranular
Ravi K Chintapalli
205
Hydrothermal degradation after sandblasting
fracture appearance and ii) mixed mode fracture appearance below the degraded layer. In
addition, a crack parallel to the surface can be seen in the degraded layer (in figure 6.4 b) with
few microcracks on either side of the crack.
Figure 6.4: Fracture surfaces of AS300 treated with sandblasting (110-2B-90º) and hydrothermal
degradation (100 hours). a) Micrograph showing degraded layer with high intergranular fracture
appearance with a crack parallel to the surface, b) crack at high magnification.
6.3.2 Elastic modulus and Hardness
Sandblasted plus hydrothermally degraded samples were carefully polished by removing the
rough peaks to produce smooth areas in order to be able to measure elastic modulus and contact
hardness by nanoindentation. Figure 6.5 shows the elastic modulus with respect to the indenter
displacement for all materials. For AS300, after hydrothermal degradation of 200 hours, the
elastic modulus is about 170±15 and after SB+HD, it is 165±10GPa. The elastic modulus
decreases approximately 25% for both conditions as compared to the control condition (refer
table 3.1, chapter 3).But the elastic modulus is practically same for SB and SB+HD. In contrast,
for S120 and 3YTZP-2CNT,the elastic modulus did not change after hydrothermal degradation
(HD) compared to control condition, and only a slight decrease of about 5% was observed after
SB+HD (see figure 6.5 ii and iii).
Figure 6.6 shows the contact hardness for all materials. For AS300, after hydrothermal
degradation of 200 hours, the contact hardness is 11±1GPa and after SB+HD it is 10±1 GPa,
which is practically similar for both conditions, and this represents a hardness decrease of about
35% in both HD and SB+HD compared to the control condition (refer table 3.1, chapter 3).
Again, in S120 and 3YTZP-2CNT the contact hardness did not practically change after
Ravi K Chintapalli
206
Hydrothermal degradation after sandblasting
hydrothermal degradation as compared to control condition, only a slight decrease of about 5%
was observed after SB+HD as seen from figure 6.6 ii) and iii).
320
AS300
Elastic modulus (GPa)
280
i)
ii)
S120
iii)
3YTZP-2CNT
HD: Hydrothermal degradation 200 hours
SB: Sandblasting:110-2B-90º
240
200
160
120
80
HD
SB+HD
40
0
0
400
800
1200
HD
SB+HD
HD
SB+HD
1600 0
Displacement (nm)
400
800
1200
1600 0
Displacement (nm)
400
800
1200
1600
Displacement (nm)
Figure 6.5: Changes in elastic modulus for different materials.
From the above results, it is clear that after HD and SB+HD the elastic modulus and the contact
hardness decrease only in AS300, which has a higher grain size compared to the other zirconia
materials. Although a slight decrease in both properties is observed in nanocrystalline materials
after SB+HD, by considering the standard deviation in the measurements, the properties are
practically similar to that of the control samples. Therefore nanocrystalline materials are highly
resistant to both the treatments shown in figure 6.5 and 6.6.
Contact hardness (GPa)
22
20
AS300
18
HD: Hydrothermal degradation 200 hours
SB: Sandblasting:110-2B-90º
16
i)
ii)
S120
3YTZP-2CNT
iii)
14
12
10
8
6
4
HD
SB+HD
2
0
0
400
800
1200
Displacement (nm)
1600 0
HD
SB+HD
HD
SB+HD
400
800
1200
Displacement (nm)
1600 0
400
800
1200
1600
Displacement (nm)
Figure 6.6: Changes in contact hardness for different materials
The elastic modulus and the contact hardness were also measured in polished cross-sections of
AS300. The nanoindentation tests were performed at different distances from the surface to a
maximum depth of about 40 m. Figure 6.7 shows the changes in both properties along the
cross-section. From figure 6.7 i) it can be seen that the lowest elastic modulus is found in the
Ravi K Chintapalli
207
Hydrothermal degradation after sandblasting
surface and thereafter a linear increase is observed up to 30 m in depth, where the elastic
modulus of SB+HD reaches its original value (modulus of control sample), whereas in HD, it is
still below the original modulus. This indicates that the thickness of the layer with decreased
properties in SB+HD is less (~22m) than in HD (~30m).
275
Polished+HD
SB+HD
225
200
175
150
Sandblasted (SB):110-2B-90º
Hydrothermal degradation (HD): 200 hours
125
100
0
5
10
15
20
25
30
35
40
Distance from the surface (m)
ii)
Polished+HD
SB+HD
18
Contact Hardness (GPa)
Elastic modulus (GPa)
250
20
i)
16
14
12
10
8
45
6
0
5
10
15
20
25
30
35
40
45
Distance from surface (m)
Figure 6.7: Cross-sectional property changes in AS 300.
Figure 6.7 ii) shows the increase in contact hardness with respect to the depth. Unlike elastic
modulus, hardness is found to be approximately constant up to a depth of ~22m in SB+HD and
~30m in HD. The hardness increases steeply between 30 and 37m of depth and reaches the
value of the control sample.
From the results observed in figure 6.2 and 6.7, it can be said that sandblasting before
hydrothermal degradation provides increasing resistance to degradation as compared with the
resistance of as sintered polished specimens (control specimens). This is manifested in the
appearance of smaller amount of transformation and also in a shorter degraded layer with
degraded elastic modulus and contact hardness as compared to AS300.
6.4 Discussion
6.4.1 Zirconia AS300
The influence of sandblasting and heat treatment over hydrothermal degradation is analyzedin
terms of t-m phase transformation and change in mechanical properties. It is very important to
clearly distinguish the changes observed in sub-micron and nanometric grain size materials.
Ravi K Chintapalli
208
Hydrothermal degradation after sandblasting
In AS300, hydrothermal degradation induces less amount of m-phase in sandblasted specimens
than in the pristine materials. Not many previous studies have focused into the study the
hydrothermal degradation behavior of sandblasted zirconia, nevertheless, Kosmac et al5,6 and
Kim et al7reported less m-phase in sandblasted zirconia samples compared to the pristine and
fully sintered CAD/CAM samples respectively after hydrothermal degradation. Kosmacet al5,6
concluded that the growth of diffusion-controlled transformation is hindered by the partitioned
tetragonal grains and pre-existing monoclinic phase in the surface of the sandblasted samples.
While, Kim et al7 concluded that sandblasting or grinding induced compressive stresses are able
to suppress the degradation, as it is well known that surface compressive stresses are beneficial
for aging resistance as reported by Devillie et al8.
In a recent study, Muñoz et al9 reported the hydrothermal degradation behavior of ground 3YTZP. They found that ground zirconia is fully resistant to degradation and it was attributed both
to the grinding induced recrystallised layer with nano grains at the surface and to the
compressive residual stress field. In addition, grounded and annealed specimens were also found
to be resistant to degradation. An annealing treatment (1200 ºC for 1 hour) after grinding
recovers its original grain structure, though grinding induced texture in t-phase remains
unchanged. The degradation resistance for ground and annealed specimens was attributed to the
presence of texture in tetragonal phase of recrystallised layer. Unlike in grinding, it is not known
for sandblasting if it induces a recrystallised nano grain layer at the surface; although it induces
tetragonal peak reversal, which is termed as texture in t-phase.
Based on the above observations from the literature, it can be suggested that the partial
hydrothermal degradation resistance of sandblasted specimens can be attributed to the surface
compressive stresses. The SB+HT+HD specimens are found to have higher m-phase compared
to the SB+HD because the zirconia heat treatment (around 1000 ºC for 1 hour) after sandblasting
relaxes the residual stresses and promote reverse m-t transformation. It can be suggested that
eliminating the contribution of the compressive stresses will normalize the degradation, which
implies that degradation proceeds as in control specimens.
The bi-axial strength of AS300 after SB+HD is decreased significantly as compared to SB for
the two sandblasting conditions and slightly in comparison with control conditions. From figure
6.4, it is clear that a transformed and microcracked layer with a thickness of about ~10 m is
formed in the surface after hydrothermal degradation.
Ravi K Chintapalli
209
Hydrothermal degradation after sandblasting
Kosmac et al5,6found that strength of sandblasted zirconia slightly decreased after hydrothermal
degradation (140º C 24 hrs in water), while Papanagiotou et al10found a significant decrease of
strength after hydrothermal degradation (7 days in boiling water)compared to only sandblasted
group. This indicates that the significant decrease in strength is linked to longer degradation time
in case of the later. In the present case the strength decrease is also linked to long degradation
time (100 hours) and is consistent with the observation of Papanagiotou et al10. The decrease in
strength is attributed to the weak monoclinic layer with severe micocracking at the surface.
Moreover, this microcracked layer has the thickness of size similar to the natural defect in the
specimen without any treatment.
The Weibull modulus for 110-2B-90º remains unchanged after hydrothermal degradation, which
implies that the scatter in the strength is not changed. However, for 110-2B-30º the Weibull
modulus increased after degradation. At this point it is not very clear why the Weibull modulus
is increased; nevertheless it could be linked to the flaw population in starting specimens.
The change in elastic modulus and contact hardness in AS300 is significant after both treatments
(HD and SB+HD) compared to pristine condition. Both properties decreased about 25-35% after
the treatments. As mentioned in chapter 3, many authors11–13reported similar results in pristine
zirconia and have attributed the decrease to the existence of microcracks in degraded specimens.
Moreover for the sandblasted sample, microcracked layer is clearly evident (see figure 6.4) and
both the properties were decreased due to loss of contact stiffness between the individual grains.
Additionally, it has been also reported that elastic modulus and contact hardness decrease with
degradation time11–13. Although properties depend on degradation time, it is important to notice
the fact that for a given time of degradation, the degraded layer has certain thickness and this
will increase with time. The properties measured by nanoindentation are in the depth of 2 m.
As the thickness of degraded layer is larger than the thickness in which properties are measured,
therefore the properties did not change within this layer for range studied here (60-200 hours).
This is the reason for the similar properties found in the surface degraded layer after 60 and 200
hours.
The other important observation is that sandblasted material is slightly more resistant to
hydrothermal degradation compared to control sample. The decrease in properties is found
deeper in control sample than in sandblasted (see figure 6.7), is an indication that the degraded
Ravi K Chintapalli
210
Hydrothermal degradation after sandblasting
layer is thicker in control compared to sandblasted sample; this is clearly evident in figure 6.8,
which shows the fracture surfaces of both samples after hydrothermal degradation of 100 hours.
Figure 6.8: SEM micrographs of fracture surfaces of AS300 after hydrothermal degradation of 100
hours.
6.4.2 Nanocrystalline Zirconia
In case of nanocrystalline materials (S120 and 3YTZP-2CNT), recalling the results from chapter
3, control samples are resistant to hydrothermal degradation (no phase transformation) and no
decrease in elastic modulus and hardness is observed. Whereas, the sandblasted samples
exhibited a nearly 10% increase of m-phase after hydrothermal degradation with no significant
decrease in properties. Therefore it is clear that the presence of m-phase prior to hydrothermal
degradation has certain effect in increasing transformation hydrothermal treatment. To
understand this observation, it is important to recall the phase transformation mechanism under
hydrothermal degradation.
Chevalier proposed14that t-m transformation occurs by nucleation and growth process.
Nucleation of the transformation in one grain leads to a volume increase, which induces stress on
the neighboring grains and generates microcraking. These microcracks offer a path for the water
to penetrate into to the material. This phenomenon occurs in materials with grain sizes larger
than critical grain size. But in nanometric grain size materials (smaller than critical grain size),
nucleation of transformation is more difficult due to high activation barrier15,16.
In nanocrystalline materials, it is observed that a small amount of monoclinic phase is formed
after sandblasting which very slightly increases after hydrothermal degradation. In this sense it
can be suggested that in nanocrystalline materials preexisting monoclinic sites induced by
sandblasting act as nucleus for the transformation to grow during hydrothermal degradation.
Ravi K Chintapalli
211
Hydrothermal degradation after sandblasting
Therefore sandblasting is completely necessary in order these specimens suffer hydrothermal
degradation. When the m-phase induced by sandblasting is transformed back to tetragonal during
a heat treatment, it becomes fully resistant to degradation. So, hydrothermal degradation did not
induce t-m transformation in the absence of preexisting monoclinic phase, which can act as a
nucleus. Therefore it is clear that in nanometric grain size materials, transformation grows
slightly only in the presence of a pre-existing nucleus. Nevertheless, in the presence of
preexisting monoclinic phase the growth of t-m transformation during hydrothermal degradation
is very less compared to AS300 which has higher grain size.
6.5 Summary
The influence of sandblasting on hydrothermal degradation and associated changes in
mechanical properties has been studied in materials with different grain sizes. The results are
summarized as follows:
 Sandblasting delays the t-m transformation growth during hydrothermal degradation and
a heat treatment before degradation recovers the usual degradation behavior observed in
polished specimens in AS300.
 Both contact hardness and elastic modulus decrease about 25-35% in AS300 after
hydrothermal degradation, but the thickness of this layer is less if the samples have been
previously sandblasted.
 Bi-axial strength decreases significantly in sandblasted AS300 after hydrothermal
degradation.
 Sandblasting in nanocrystalline materials (S120 and 3YTZP-2CNT) slightly favored
transformation during hydrothermal degradation, but the growth kinetics is far less
compared to AS300. Heat treatment after sandblasting offers full protection to
degradation.
Ravi K Chintapalli
212
Hydrothermal degradation after sandblasting
6.6 References
1. Kobayashi K, Kuwajima H, Masaki T. Phase Change and Mechanical Properties of ZrO2-Y2O3 Solid
Electrolyte after Ageing. Solid State Ionics. 1980;3(4):489-493.
2. Albakry M, Guazzato M, Swain MV. Effect of Sandblasting, Grinding, Polishing and Glazing on the
Flexural Strength of two Pressable All-Ceramic Dental Materials. Journal of Dentistry. 2004;32(2):91-99.
3. Curtis AR, Wright AJ, Fleming GJP. The Influence of Surface Modification Techniques on the
Performance of a Y-TZP Dental Ceramic. Journal of Dentistry. 2006;34(3):195-206.
4. Jevnikar P, Sersa I, Sepe A, Jarh O, Funduk N. Effect of Surface Coating on Water Migration Into
Resin-Modified Glass Ionomer Cements: A Magnetic Resonance Micro-Imaging Study. Magnetic
Resonance in Medicine. 2000;44(5):686-91.
5. Kosmac T, Oblak C, Marion L. The Effects of Dental Grinding and Sandblasting On Ageing and
Fatigue Behavior of Dental Zirconia (Y-TZP) Ceramics. Journal of the European Ceramic Society.
2008;28:1085-1090.
6. Kosmac T, Dakskobler A, Oblak C, Jevnikar P. The Strength and Hydrothermal Stability of Y-TZP
Ceramics for Dental Applications. International Journal of Applied Ceramic Technology. 2007;4(2):164174.
7. Kim J-W, Covel NS, Guess PC, Rekow ED, Zhang Y. Concerns of Hydrothermal Degradation In
CAD/CAM Zirconia. Journal of Dental Research. 2010;89(1):91-5.
8. Deville S, Chevalier J, Gremillard L. Influence of Surface Finish and Residual Stresses On the Ageing
Sensitivity of Biomedical Grade Zirconia. Biomaterials. 2006;27(10):2186-92.
9. Muñoz-Tabares JA, Anglada M. Hydrothermal Degradation of Ground 3Y-TZP. Journal of the
European Ceramic Society. 2012;32(2):325-333.
10. Papanagiotou HP, Morgano SM, Giordano RA, Pober R. In Vitro Evaluation of Low-Temperature
Aging Effects and Finishing Procedures on the Flexural Strength and Structural Stability of Y-TZP
Dental Ceramics. The Journal of Prosthetic Dentistry. 2006;96(3):154-164.
11. Cattani-Lorente M, Scherrer SS, Ammann P, Jobin M, Wiskott HWA. Low Temperature Degradation
of a Y-TZP Dental Ceramic. Acta Biomaterialia. 2011;7(2):858-65.
12. Gaillard Y, Jimenez-Pique E, Soldera F, Mucklich F, Anglada M. Quantification of Hydrothermal
Degradation in Zirconia by Nanoindentation. Acta Materialia. 2008;56:4206-4216.
13. Muñoz-Tabares JA, Jiménez-Piqué E, Anglada M. Subsurface Evaluation of Hydrothermal
Degradation of Zirconia. Acta Materialia. 2011;59(2):473-484.
14. Chevalier J. What Future for Zirconia as a Biomaterial? Biomaterials. 2006;27:535-543.
15. Chintapalli RK, Mestra A, García Marro F, et al. Stability of Nanocrystalline Spark Plasma Sintered
3Y-TZP. Materials. 2010;3(2):800-814.
16. Suresh A, Mayo MJ, Porter WD, Rawn CJ. Crystallite and Grain-Size-Dependent Phase
Transformations in Yttria-Doped Zirconia. Journal of American Ceramic Society. 2003;86(2):360-362.
Ravi K Chintapalli
213
Hydrothermal degradation after sandblasting
Ravi K Chintapalli
214
Conclusions
Chapter
7
Conclusions and future work
Ravi K Chintapalli
215
Conclusions
7.1 Conclusions
In this thesis, the influence of sandblasting in 3Y-TZP ceramics is studied for various aspects
such as different sandblasting conditions and microstructures of zirconia. A summary at the end
of each chapter is already given to highlight the results found. This chapter will be restricted to
general conclusions from this thesis.
i. It has been shown that fine grain size zirconia (< 200 nm) is resistant to hydrothermal
degradation and this has been explained in terms of the small grain size, as the the
activation barrier to form a critical nucleus for t-m transformation increases by decreasing
the grain size.
ii. The addition of 2 vol% multiwall carbon nanotubes increases the indentation fracture
toughness up to 15%, as observed by the shorter cracks observed in the composite in
comparison with monolithic zirconia, and Vickers hardness slightly decreases.
iii. Sandblasting, if done with proper care and control is beneficial for dental restorations.
Within the limitations of the study with respect to the sandblasting conditions it can be
concluded in the following way
a)
Sandblasting is recommended with particles size equal or less than 110 m and
pressures less than 4 bars for better reliability of the dental crowns and implants.
b)
Low impact angle (30º) should be employed when sandblasting with particles
larger than 110 m to achieve high surface roughness.
c)
Mild sandblasting increases the bi-axial strength by inducing residual
compressive stresses, and also do not change the surface mechanical properties
such as elastic modulus and contact hardness.
d)
Sandblasting delays hydrothermal degradation.
7.2 Future work
This work has been performed with specific objectives and a time frame. Within the limitations
of this thesis, effect of sandblasting is studied for materials with different microstructures.
However, this work can be carried out further by studying the following aspects.
i.
The fracture toughness and strength are very important factors for materials subjected to
contact loads, especially in case of dental crowns and implants. In this thesis, the strength
is only studied for zirconia with grain size 0.3 mm due to lack of number of specimens of
nanocrystalline zirconia. It would be interesting to study the strength of the
Ravi K Chintapalli
216
Conclusions
nanocrystalline zirconia before and after sandblasting. Since this material has slightly less
fracture toughness and undergoes less t-m transformation after sandblasting, the effect of
this on the strength should be studied.
ii.
With regard to nanocomposites, it is found here that an increase about 15 % in the
indentation fracture toughness by adding 2 vol % of MWCNT to 3Y-TZP matrix.
Nevertheless, measuring fracture toughness by indentation has limitations and fracture
toughness shall also be measured by traditional methods such as notched specimens with
sharp cracks under bending.
iii.
Understanding the role of the interface between MWCNT and zirconia grains will be
helpful to comprehend the weakness of the toughening mechanisms induced by adding
multi wall carbon nanotubes to zirconia matrix.
iv.
As the grain size becomes smaller of about 100 nm, the strengthening as well as damage
mechanisms under contact and impact loads in the material tends to change. So
transmission electron microscopy studies to understand these mechanisms will help to
better design and selection of materials particularly in case of nanocrystalline zirconia.
v.
The bond strength of between the sandblasted zirconia and cement is crucial for longterm performance of the crowns, so the studies of bond strength using the proposed best
sandblasting conditions here will compliment the results presented here and in general
provides suitable information for the manufacturers of dental restorations.
Ravi K Chintapalli
217
Conclusions
Ravi K Chintapalli
218
Appendix
Appendix
For gradient residual stresses, the stress decreases with depth. In case of semicircular crack under
the gradient residual stresses, the apparent fracture toughness at the surface (point B figure 5.15
in chapter 5) is much higher than at the deepest point (point A figure 5.15 in chapter 5).
However, for simplicity assuming a straight crack will be sufficient as the specimen might fail
from the point A. In order to obtain analytical solutions we shall consider the material is
isotropic and that all natural or indentation cracks can be assimilated to edge cracks. For
simplicity it will be assumed that the residual stress can be represented by a linear relationship,
that is,
0
1
1
2
Where c is the residual stress, x is the thickness removed and d is depth remaining after
removing x. Let us now assume that the stress intensity factor of the residual stress in the
material increases with depth according to equation 5.21 in chapter 5. The stress intensity factor
can be calculated by the weight function method and assuming that the crack is in an infinite
body,
√
3
Where g(x)= 2c/√c2-x2 is a so called Green´s function1 and Y0=1.12, by substituting this in the
above equation gives
2
4
By integrating and solving the above equation by using equations A1, A2 and A4, we obtain
√ Ravi K Chintapalli
5
219
Appendix
For a crack c smaller than or equal to the depth of the stress field d, c ≤ d
√ 1
2
6
And for a crack c larger than or equal to the depth of the stress field d, c ≥ d
√ 2
2
/
1
1
7
If c >> d the last expression can be approximated by:
√
8
We can compare the residual stress intensity factor acting on the critical defect size, c0,
corresponding to the non-sandblasted material. For c0 ≤ d
1
2
9
For c0 > d
2
1
10
With the true fracture toughness
where
is the of the strength of the non-sandblasted material.
The result is plotted in figure A1.The above model describes relationship between how the
residual stress intensity factor changes with depth of the stress field in an isotropic material with
respect to the indentation crack. The equations A9 and A10 are plotted for c0 > d and c0 < d in
Ravi K Chintapalli
220
Appendix
figure A1. Now it is clear from the figure A1 that with decreasing the depth of the stress field the
0.5
KRes/(Y0c0) )
residual stress intensity factor decreases and vice versa.
1.0
1.0
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
d<c0
c0<d
d>c0
c0>d
0.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
c0/d
d/c0
Figure A1: Relationship between residual stress intensity factor and the ratio of crack to depth
References
1. Lawn BR. Fracture of Brittle Solids. Second Edition, Cambridge University press, Cambridge, United
Kingdom. 1993.
Ravi K Chintapalli
221
Appendix
Ravi K Chintapalli
222
Publications and conference contributions
Publications and conference
contributions
Ravi K Chintapalli
223
Publications and conference contributions
Peer reviewed publications
1. Chintapalli RK, Marro FG, Milsom B, Reece MJ, Anglada M. Processing and characterization of
high-density zirconia-carbon nanotube composites, Materials Science & Engineering A, 2012 549:559.
2. Marro FG, Mestra A, Jiménez-Piqué E, Chintapalli RK, Anglada M. Surface mechanical properties
of advanced zirconia after hydrothermal exposure. IOP Conference Series: Materials Science and
Engineering. 2012; 31:012015.
3. Chintapalli RK, Jimenez-Pique E, Marro FG, Yan H, Reece MJ, Anglada M. Spherical
instrumented indentation of porous nanocrystalline zirconia. Journal of the European Ceramic
Society. 2011; 32(1):123-132.
4. Chintapalli RK, Mestra A, Marro FG,, Yan H, Reece MJ, Anglada M. Stability of Nanocrystalline
Spark Plasma Sintered 3Y-TZP. Materials. 2010; 3(2):800-814.
5. Chintapalli RK, Marro FG, Valle JA, Anglada M. Degradation resistance of 3Y-TZP ceramics
sintered using spark plasma sintering. IOP Conf. Series: Materials Science and Engineering. 2009;
5(012014):1-8.
Conference contributions
1. Chintapalli RK, Marro FG, Mestra A, Anglada M. Influence of sandblasting on the strength of 3Y-
TZP. 13th European Inter-regional Conference on Ceramics, 12-14 September, 2012, Barcelona,
Spain, Invited, Oral
2. Chintapalli RK, Marro FG, Jiménez-Pique E, Anglada M. Effect of sandblasting on hydrothermal
degradation of zirconia ceramics. 13th European Inter-regional Conference on Ceramics, 12-14
September, 2012, Barcelona, Spain, Invited, Poster.
3. Chintapalli RK, Marro FG, Mestra A, Anglada M. Influence of sandblasting on bi-axial strength of
3Y-TZP. XII Congreso Nacional de Materiales (12th National congress of Materials), 30 May-1
June, 2012 Alicante, Spain, Poster.
4. Mestra A, Marro FG, Chintapalli RK, Zamir de Armas, Anglada M, Influencia de la degradación
hidrotérmica en la resistencia mecánica y en el comportamiento al desgaste de 3y-tzp (Influence of
hydrothermal degradation on bi-axial strength and wear behaviour of 3Y-TZP). XXIX Encuentro Del
Grupo Español De Fractura, (29th Meeting of Spanish fracture group), 21-23 March 2012, Bilbao,
Spain, Oral.
5. Marro FG, Chintapalli RK, Mestra A, and Anglada M. Determination of the intrinsic fracture
toughness from the cod analysis of indentation cracks in spark plasma sintered 3y-tzp reinforced
with MWCNT. XXIX Encuentro Del Grupo Español De Fractura, (29th Meeting of Spanish fracture
group), 21-23 March 2012. Bilbao, Spain, Oral.
6. Chintapalli RK, Marro FG, Milsom B, Reece MJ and Anglada M. Processing and characterization
of Zirconia-MWCNT nanocomposites, 6th EEIGM International Conference on Advanced Materials
Research. 7-8, November, 2011, EEIGM, Nancy, France, Poster.
7. Marro FG, Mestra A, Jiménez-Piqué E, Chintapalli RK, Anglada M. Surface mechanical properties
of advanced zirconia after hydrothermal exposure. 6th EEIGM International Conference on
Advanced Materials Research. 7-8, November, 2011, EEIGM, Nancy, France, Poster.
Ravi K Chintapalli
224
Publications and conference contributions
8. Chintapalli RK, Marro FG, Mestra A., Reece MJ, Anglada M, Relationship Between Density And
Mechanical Properties Of Spark Plasma Sintered Zirconia Ceramics, 12th European Inter-regional
Conference on Ceramics, 7- 9, September 2010, Mons, Belgium, Oral.
9. Chintapalli RK, Marro FG, Valle JA, Yan H, Reece MJ and Anglada M. Degradation resistance of
3Y-TZP
ceramics
sintered
using
spark
plasma
sintering,
5th
International
EEIGM/AMASE/FORGEMAT Conference on Advanced Materials Research, 4-5, November,
2009, EEIGM, Nancy, France, Oral.
Ravi K Chintapalli
225
Fly UP