Dental Materials (2004) 20, 498–508
http://www.intl.elsevierhealth.com/journals/dema
Durability of the resin bond strength
to the alumina ceramic Procera
Michael Hummel, Matthias Kern*
Department of Prosthodontics, Propaedeutics and Dental Materials, School of Dentistry,
Christian-Albrechts University, Kiel, Germany
Received 26 February 2003; received in revised form 28 August 2003; accepted 8 October 2003
KEYWORDS
Alumina ceramic;
Composite resin; Silane;
Silica coating; Tensile
bond strength; Thermal
cycling
Summary Objectives. The purpose of this in vitro study was to evaluate the tensile
bond strength of adhesive bonding systems to the densely sintered alumina ceramic
Procera, and its durability.
Methods. Plexiglas tubes filled with composite resin were bonded to Procera
ceramic discs (99% Al2O3), which were either in their original state as supplied by the
manufacturer or which were sandblasted for surface conditioning. Groups of 20
specimens were bonded in an alignment apparatus using 10 bonding methods.
Subgroups of 10 bonded specimens were tested for tensile strength following storage
in distilled water at 37 8C either for 3 days or for 150 days. In addition, the 150 days
specimens were thermal cycled 37,500 times. The statistical analyses were conducted
with the Kruskal –Wallis test followed by multiple pair-wise comparison of groups using
the Wilcoxon rank sum test.
Results. Not sandblasted groups showed relatively poor initial bond strengths
independent from bonding resins. During 150 days storage time all specimens in the not
sandblasted groups debonded spontaneously.
Moderate to relatively high initial bond strengths between 18 and 39 MPa were
achieved to sandblasted specimens by using the PMMA luting resin Superbond C & B
or the composite resin Variolink II or by silica coating and silanation in combination
with Variolink II. However, in these groups after 150 days storage time the bond
strength decreased significantly. The phosphate monomer containing composite resin
Panavia 21 showed the highest bond strength to sandblasted Procera ceramic which did
not decrease significantly over storage time. In addition, the bond strengths of
sandblasted groups bonded with Variolink II after priming the ceramic with Alloy
Primer or the silane Monobond S were relatively high and did also not decrease
significantly after 150 days storage time.
Significance. Using ten bonding systems, a stable resin bond to Procera ceramic
could be achieved after sandblasting and by using Panavia 21 or by using Variolink II
after priming the ceramic with Alloy Primer or Monobond S.
Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
Introduction
*Corresponding author. Tel.: þ 49-4315972874; fax: þ 494315972860.
E-mail address: mkern@proth.uni-kiel.de
Recently, the densely sintered alumina ceramic
Procera has become more and more popular
0109-5641/$ - see front matter Q 2004 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dental.2003.10.014
Durability of the resin bond strength to the alumina ceramic Procera
as a restorative dental material. It is already used as
a material for crowns, fixed partial dentures and
abutments for dental implants.1 A 5-year follow-up
study for anterior and posterior crowns revealed
successful clinical results.2 Although bonding with
resin to Procera ceramic would be advantageous for
many clinical applications, only a few laboratory
studies evaluated the short-term bond strength to
Procera ceramic.3,4 However, there is only limited
data available about reliable bonding methods to
this ceramic where thermal cycling and long-term
water storage have been included.5 In a recent
studies with storage times up to 180 days, thermal
cycling bond strengths were either low initially or
decreased significantly over storage time.6,7 Similar
results were reported for another densely sintered
pure alumina ceramic.8
Therefore, the purpose of this study was to
evaluate using thermal cycling, the bond strength
to the densely sintered alumina ceramic Procera
and its durability after a storage time of 150 days
and to investigate bond failure modes by SEM
characterization. The hypothesis tested was that
bond strength is influenced by the bonding methods
and storage conditions.
Materials and methods
Industrially manufactured densely sintered alumina
ceramic discs (99% Al2O3, Nobel Biocare, Gothenborg, Sweden) with a diameter of 6.8 – 7.5 mm and a
thickness of 3.4 mm were used for this study. The
bonding surfaces were either left in their original
state as supplied by the manufacturer or they were
sandblasted with 50 mm Al2O3 at 2.5 bars pressure
(13 s) at a distance of 10 mm. All specimens were
then ultrasonically cleaned in 96% isopropanol for
3 min.
Ten different bonding methods were used. The
materials utilized and their characteristics are
listed in Table 1.
The following ten different test groups with
twenty specimens each were arranged:
S-V
The conventional BisGMA composite resin
Variolink II together with its adhesive resin
Heliobond (both Ivoclar Vivadent, Schaan,
Liechtenstein) were used on sandblasted
specimens. Heliobond was applied in a thin
layer onto the bonding surface and lightcured for 20 s with a dental curing light
(Optilux 500, Demetron Company, Danbury,
USA). The base and catalyst paste of
Variolink II were then mixed for 20 s
S-RV
O-AV
S-AV
O-SV
S-SV
O-SB
S-SB
O-P
S-P
499
and applied to bond the composite resin
filled Plexiglas tubes to the ceramic surface.
After tribochemically silica coating and
silanization of the ceramic with the Rocatec
system (3M Espe, Seefeld, Germany) the
resins Heliobond and Variolink II were used
for bonding as described for group S-V.
Alloy Primer (Kuraray, Osaka, Japan) was
used for priming the original, i.e. untreated,
ceramic surface, then Heliobond and Variolink II were used for bonding as described
for group S-V.
Alloy Primer, Heliobond and Variolink II
were used on sandblasted specimens as
described for group O-AV.
After silanization of the ceramic with the
silane Monobond S (Ivoclar Vivadent,
Schaan, Liechtenstein), Heliobond and Variolink II were used on the original ceramic
surface as described for group S-V.
The silane Monobond S, Heliobond and
Variolink II were used on sandblasted specimens as described for group O-SV.
Superbond C & B (Sun Medical, Kyoto,
Japan), which contains 4-META (4-methacryloxyethyl-trimellitat-anhydrid), was
applied to the original ceramic surface.
Superbond C & B was applied to sandblasted
specimens.
Panavia 21 (Kuraray), which contains a
phosphate monomer (10-methacryloyloxydecyl dihydrogenphosphate MDP), was
applied to the original ceramic surface.
Panavia 21 was used on sandblasted
specimens.
Plexiglas tube (Fig. 1) specimens were filled with
self-curing composite resin Clearfil F2 (Kuraray).
This composite resin has been chosen as it was
successfully used in previous bonding studies without showing any incompatibility to various bonding
materials.9 – 11 The tube rim aligned on the ceramic
sample was beveled to a final thickness of 0.1 mm
to ensure minimal contact of the tube itself and for
thinning the luting agent out to a minimal thickness
at the rim. Two retentive grooves prepared on the
opposite inner side of the tube ensured that the
polymerizing resin shrank from the bonding surface
providing a space for the luting composite resin.
The space provided as a result of polymerization
shrinkage of the auto-curing composite resin was
estimated at approximately 60 mm in height,
because the linear shrinkage of Clearfil F2 is
approximately 0.40% after 8 min.12 After 8 min
from mixing Clearfil F2, these tubes were bonded,
according to the above listed procedures, to
500
M. Hummel, M. Kern
Table 1 List of materials used.
System
Component
Batch No.
Main compositiona
Manufacturer
Clearfil F2
Base paste
1660
BisGMA/TEGDMA/DMA/
barium sulfate/
silica cont. com-posite resin
Kuraray, Osaka,
Japan
Rocatec
Catalyst paste
Rocatec Pre
Rocatec plus
1561
082
462
110 mm Al2O3
Silica cont.110 mm Al2O3
Espe-Sil
Alloy primer
123
091BA
MPS silane
MDP/VBATDT cont. primer
Monobond S
532888
MPS silane
Panavia 21 TC
Variolink II
Superbond
C and B
Base paste
Catalyst paste
42252
41152
Oxyguard II
42251
Heliobond
Base paste
B11128
46759
Catalyst paste
Liquid-Strip
Polymer L-Type
Catalyst
548787
B15614
VX1
EM12
Monomer
EM12
BPEDMA/MDP/DMA/
Ba–B–Si-glass/
silica cont. com-posite resin
Polyethyleneglycol/glycerine/
sodium benzensulfinate cont. gel
BisGMA/TEGDMA cont. resin
BisGMA/UDMA/TEGDMA/
DMA/barium sulfate/
Ba–Al-Fluoro-Si-glass/
silica cont. com-posite resin
3M Espe, Seefeld,
Germany
Kuraray, Osaka,
Japan
Ivoclar-Vivadent,
Schaan,
Liechtenstein
Kuraray, Osaka,
Japan
Ivoclar-Vivadent,
Schaan,
Liechtenstein
Glycerine cont. gel
4-META/TBB/
PMMA containing resin
Sun Medical,
Kyoto, Japan
4-META ¼ 4-methacryloxyethyltrimellitic acid anhydride, BisGMA ¼ Bisphenol-A-diglycidylmethacrylate, BPEDMA ¼ Bisphenol-Apolyethoxy dimethacrylate, DMA ¼ aliphatic dimeth-acrylate, MDP ¼ 10-methacryloyloxy-decyl-dihydrogenphosphate, MPS ¼ 3methacryloyloxypropyl trimethoxysilane, PMMA ¼ polymethylmethacrylate TEGDMA ¼ triethyleneglycol dimethacrylate, TBB ¼
tri-n-butylborane, UDMA ¼ urethane dimethacrylate, VBATDT ¼ 6-(4-Vinylbenzyl-n-propyl)amino-1,3,5-triazine-2,4-dithione, Al ¼
aluminum, B ¼ boron, Ba ¼ barium, F ¼ fluorine, Si ¼ silicium, cont. ¼ containing.
a
According to the information provided by the manufacturers.
the Procera ceramic specimens using an alignment
apparatus. The alignment apparatus consisted of
parallel guides, a tube holder, a silicon pad and an
added weight of 750 g. The apparatus ensured that
the tubes axis was perpendicular to the surface
(Fig. 2).
Excess resin was removed from the bonding
margin using cotton pellets, then an oxygenblocking gel was applied with a syringe to the
bonding margins for the composite resin luting
agents (Liquid-Strip, Ivoclar Vivadent, for Variolink
II and Oxyguard II, Kuraray, for Panavia 21 TC).
After 7 min, all specimens with Variolink II were
light-cured from two sides, 180 degrees apart, for
40 s with a dental curing light (Optilux 500,
Demetron Company, Danbury, USA), then further
cured in a xenon strobe light-curing unit (Dentacolor XS, Heraeus-Kulzer, Wehrheim, Germany) for an
additional 90 s. After 10 min in an incubator at
a temperature of 37 8C all specimens were stored in
distilled water at 37 8C.
Each bonding group with a total of 20 specimens
was divided into two subgroups (each with 10
specimens) and stored in distilled water at 37 8C
either for 3 days without thermal cycling (TC) or for
150 days with additional 37,500 TC between 5 and
55 8C at regular intervals (Fig. 2).
After these times, the tensile bond strength
test was performed at a crosshead speed of 2 mm
per min in a universal testing machine (Zwick Z
010/TN2A, Ulm, Germany) using a special test
configuration, which provided a moment-free
axial force application. A collet held the tube
while an alignment jig allowed self-centering of
the specimens. The jig was attached to the load
cell and crosshead by upper and lower chains,
allowing the whole system to be self-aligning
(Fig. 2) .
Durability of the resin bond strength to the alumina ceramic Procera
501
Statistical analyses of the test results were
conducted using the Kruskal –Wallis test followed
by multiple pair-wise comparisons of groups using
the Wilcoxon rank sum test for independent specimens.13 Significance levels were adjusted with the
Bonferroni –Holm correction for multiple testing.14
The fracture interfaces on the debonded specimens were examined under a light microscope
(Wild Makroskop M 420, Heerbrugg, Germany) at
30 £ magnification to assess the mode of failure
classified as adhesive, cohesive or mixed. The total
bonding area was divided into 75 squares using a
special eyepiece and areas with remaining adhesive
on the debonded surface were visually inspected,
counted, and the percentage of adhesive failure
mode calculated. This method assured a standardized examination process with a variability of
about 5%. After sputtering using a gold-alloy
conductive layer of approximately 30 nm, representative specimens were examined using a scanning electron microscope (XL30CP, Philips,
Eindhoven, Netherlands) with an acceleration voltage of 15 keV and a working distance of 10 mm.
Results
Figure 1 Schematic drawing of a Plexiglas tube. All
specifications are in mm unless otherwise noted.
Mean tensile bond strengths are summarized in
Table 2 for the ten bonding groups and the two
storage conditions. Statistically significant differences between the bonding groups and the storage
conditions are indicated in the same table. The
hypothesis that bond strength is influenced by
Figure 2 Test configuration. Alignment apparatus (left), thermal cycling apparatus (middle), and self-aligning
debonding jig with upper and lower rings for chain suspension (right).
502
M. Hummel, M. Kern
Table 2 Tensile bond strength to Procera ceramic after
different storage times. Medians, means and standard
deviations in MPa ðN ¼ 8Þ:
Groups
Storage time
3 d/0 TC
S-V
S-RV
O-AV
S-AV
O-SV
S-SV
O-SB
S-SB
O-P
S-P
150 d/37,500 TC
Medians
Means (SD)
Medians
Means (SD)
17.0Da
31.9Ca
9.2Ea
33.1Ca
7.5Fa
38.8Ba
0.0E,F
a
31.3Ca
0.0E,F
a
45.0Aa
18.2
33.3
9.5
32.5
5.3
39.1
5.6
32.1
3.9
45.1
0.0Eb
24.2Cb
0.0Eb
31.7B,C
a
0.0Eb
32.9Ba
0.0Eb
17.4Db
0.0Eb
35.8Aa
0.0 (0.0)
25.4 (5.5)
0.0 (0.0)
30.3 (4.5)
0.0 (0.0)
32.8 (3.1)
0.0 (0.0)
16.4 (4.5)
0.0 (0.0)
37.1 (6.2)
(3.0)
(2.6)
(3.4)
(3.4)
(4.5)
(1.7)
(7.6)
(5.3)
(5.4)
(1.0)
TC ¼ thermal cycles. Within the same column, means with
the same superscript letter are not statistically different ðp .
0:05Þ: Within the same row, means with the same Greek
subscript letter are not statistically different ðp . 0:05Þ:
Global Kruskal–Wallis test followed by pair-wise comparison
using the Wilcoxon test modified by Bonferroni–Holm.
the bonding methods was confirmed. However, only
for some test groups was the bond strength
influenced by the storage conditions.
All groups with surfaces in their original, i.e. not
sandblasted state, showed relatively low initial
bond strengths. During 150 days storage time all
specimens in these groups debonded spontaneously, independent of the bonding resins
used. Specimens bonded with Variolink II to
sandblasted Procera ceramic initially showed
a bond strength of about 18 MPa and also debonded
spontaneously after 150 days.
Silica coating and silanation of Procera ceramic
using the Rocatec system increased the bond
strength of Variolink II statistically significantly to
about 33 MPa, but did not remain stable over
storage time.
The bond strength of the specimens bonded with
Alloy Primer and Variolink II was initially relatively
high at about 32 MPa, and after 150 days the bond
strength remained stable at 30 MPa. Similary,
sandblasted specimens bonded with the silane
Monobond S and Variolink II showed a high initial
bond strength at about 39 MPa and the bond strength
remained stable at about 33 MPa over storage time.
Using Superbond C & B on sandblasted Procera
ceramic, a high initial bond strength of about
32 MPa was found. However, over 150 days storage
time with thermal cycling the bond strength
decreased significantly to about 16 MPa. In contrast, Panavia 21 showed the highest bond strength
to sandblasted ceramic of all test groups at 45 MPa,
which decreased not significantly to about 37 MPa
over 150 days storage time.
The distribution of the failure modes for all
groups is shown in Fig. 3. After 150 days of storage
with 37,500 thermal cycles, the failure mode was
purely adhesive for bonding groups S-V, S-RV, O-AV,
O-SV, O-SB and O-P. The failure mode was either
mostly or exclusively cohesive in groups S-AV, S-SV
and S-P. Figs. 4 – 6 show representative SEM photographs of the fracture interfaces after the tensile
test following 3 days or 150 days storage time with
37,500 thermal cycles.
Figure 3 Mean percentages of areas assigned to the failure modes observed in the bonding groups after different
storage times.
Durability of the resin bond strength to the alumina ceramic Procera
503
Figure 4 (a) Sample of group SP after 150 days/37,500 thermal cycles showing cohesive failure of the resin composite
(SEM photograph, magnification 100 £ , white bar 500 mm). (b) Sample of group SP after 150 days/37,500 thermal cycles
showing blank squares indicating torn filler particles of the composite resin (SEM photograph, magnification 750 £ ,
white bar 50 mm).
Discussion
As the results of this study show, it was not possible
to achieve a stable long-term bond to the original
Procera ceramic surface, i.e. without sandblasting.
All specimens with the original surface had a low
initial bond strength and debonded spontaneously
after 150 days. The spontaneous debonds in the not
sandblasted groups show that neither stable micromechanical retention nor stable chemical bonds
could be achieved without sandblasting.
Fig. 7 showing the surface topography of the
original ceramic surface as provided by the
manufacturer, reveals that a certain surface
504
M. Hummel, M. Kern
Figure 5 (a) Sample of group SV after 3 days water storage showing adhesive failure mode (SEM photograph,
magnification 100 £ , white bar ¼ 500 mm). (b) Sample of group SV after 150 days/37,500 thermal cycles showing total
adhesive failure mode (SEM photograph, magnification 750 £ , white bar ¼ 50 mm).
roughness but no undercuts or porosities are
present on the original surface. However after
sandblasting many porosities are seen on the
ceramic surface which obviously improved microretention of the luting agents by interlocking (Fig.
8). So the sandblasting step seems necessary to
open near surface porosities which are covered by
a dense superficial layer after the manufacturing
process. Such porosities are not only near
the ceramic surface but can be also found in
deeper ceramic layers (Fig. 9).
In addition, without sandblasting a stable chemical bond to Procera was also not achieved by any of
the bonding systems. So neither with priming,
silanization nor by using resins with adhesive
monomers could a stable bond strength be achieved
to the not sandblasted alumina ceramic. This can be
explained by the fact that sandblasting hot only
Durability of the resin bond strength to the alumina ceramic Procera
505
Figure 6 (a) Sample of group RSV after 3 days water storage showing a mixed failure mode (SEM photograph,
magnification 100 £ , white bar ¼ 500 mm). (b) Sample of group RSV after 3 days showing a mainly adhesive debonded
area of the specimen with remains of the composite resin (SEM photograph, magnification 750 £ , white bar ¼ 50 mm).
creates a microretentive surface but also cleans the
surface from any contaminants which might prevent chemical bonding.15,16 Therefore it seems that
sandblasting is an indisposable step to clean and
activate the alumina ceramic prior to using chemical bonding methods.
Low initial bond strength values to the not
sandblasted Procera intaglio surface were also
reported in a recent laboratory study, in which
a shear bond strength of only 7.6 MPa after 3 days
storage was achieved,7 which decreased significantly to 4.3 MPa after 180 days storage time in
water with TC. Although test specimens did not
debond spontaneously in the cited study, these low
bond strength values seem to be insufficient for
clinical applications.
Although silica coating with silanization
increased the initial bond strength of Variolink II
506
M. Hummel, M. Kern
Figure 7 SEM photograph of alumina ceramic sample with its untreated surface as supplied by the manufacturer
(magnification 1,500 £ , white bar 20 mm).
to sandblasted Procera ceramic significantly, water
storage with TC resulted in a decrease in bond
strength for the silica coated and silanated bonding
group. This may be caused by a missing glass-phase
in the Procera ceramic and it corresponds with the
results of previous studies in which silica coating
was also not successful for glass-phase-free and
densely sintered alumina and zirconia ceramics.8,11
In contrast to this study Kern and Thompson
showed, that silica coating and silanation of a
glass-infiltrated alumina ceramic resulted in stable
and high bond strengths.10 A significant decrease
was also found in a recent study with water storage
up to 180 days.7
The bond strength of Variolink II to sandblasted
Procera ceramic was improved significantly through
Figure 8 SEM photograph of a sandblasted ceramic sample (magnification 1,500 £ , white bar 20 mm).
Durability of the resin bond strength to the alumina ceramic Procera
507
Figure 9 SEM photograph of alumina ceramic sample in a sectional view (magnification 1,000 £ , white bar 50 mm).
the use of Alloy Primer, which was also hydrolytic
stable after 150 days water storage with TC. It can
be assumed that the phosphate monomer in Alloy
Primer is the reason for the stability of the bond.
The phosphate ester group of the MDP is supposed to
directly bond to metal oxides.17 Obviously, the MDP
within the Alloy Primer is able to create a stable
chemical bond to alumina ceramic Procera.
The bond strength of group S-SV was initially
relatively high with about 39 MPa. Also after 150
days storage with TC no significant decrease in bond
strength was found. The values were still above
30 MPa. These results are unlike preceding studies
that did not show a significantly positive effect of
silanization on the bond strength to oxide ceramics.8,18,19 The reason for these results might be
related to the surface structure of the Procera
ceramic. As the SEM of a sandblasted ceramic
sample shows (Fig. 8), there are deep porosities
unlike the original surface (Fig. 7). The function of
the silane might be to increase the wettability,
which allows flow of the bonding resins into the
undercuts and porosities. These undercuts can be
seen in the entirely sectional view (Fig. 9). So the
results of a former study in which a silane
significantly increased the bond strength to the
sandblasted alumina ceramic In-Ceram can be
explained.20
The bond strength between the PMMA resin
Superbond C & B and the sandblasted ceramic
specimens was initially high. The anhydride group
of 4-META is supposed to create chemical bonds to
metal oxides.21 Considering the results of the
current study it can be assumed that Superbond C
& B also bonds chemically to the alumina ceramic
Procera. However after 150 days storage time with
TC, the bond strength decreased statistically
significantly. This can be explained by the water
absorption of PMMA over long-term storage, which
seems to diminish the chemical bond.22 Also in
other studies a stable bond strength between
Superbond C & B and alloys was not achieved.21,23
The bond strength of the MDP-containing composite resin Panavia 21 (group S-P) showed
the highest bond strength. Although a slight
decrease in bond strength over 150 days storage
time was observed, it was not statistically significant. Since a completely cohesive failure mode was
found for both the 3-days and 150-days specimens,
this decrease in bond strength might be related to a
certain degeneration of the resin itself, which has
previously been shown in other resins.24 As already
shown for glass-infiltrated alumina ceramic10,25 and
zirconia ceramic19 the results of the current study
demonstrate that MDP promotes water resistant
chemical bonds to Procera ceramic. The phosphate
ester group of the monomer is reported to bond
directly to metal oxides,26 and the data of the
current and previous studies suggest that MDP
bonds chemically to both alumina and zirconia.
However given the present results, the need for
long-term investigations regarding the bond
strength of resins to restorative materials is
evident.
508
Conclusions
In summary, after 150 days storage with thermocycling, only the MDP-containing composite resin
and the combination of the MDP-containing Alloy
Primer and Variolink II or the combination of
silanation and Variolink II did not exhibit a
statistically significant decrease in bond strength
to sandblasted Procera ceramic as compared with
the initial values after 3 days. Therefore, only these
methods can currently be recommended for clinical
use when a stable clinical bond to Procera ceramic
is required.
Acknowledgements
This study was supported by Nobel Biocare (Gothenburg, Sweden).
References
1. Hegenbarth EA. Procera aluminum oxide ceramics: a new
way to achieve stability, precision and esthetics in allceramic restorations. Quintessence Dent Technol 1996;19:
21—34.
2. Odén A, Andersson M, Krystek Ondracek I, Magnusson D.
Five-year clinical evaluation of Procera AllCeram crowns.
J Prosthet Dent 1998;80:450—6.
3. Awliya W, Odén A, Yaman P, Dennison JB, Razzoog ME. Shear
bond strength of a resin cement to densely sintered highpurity alumina with various surface conditions. Acta Odontol
Scand 1998;56:9—13.
4. Blixt M, Adamczak E, Lindén L-Å, Odén A, Arvidson K.
Bonding to densely sintered alumina surfaces: effect of
sandblasting and silica coating on shear bond strength
of luting cements. Int J Prosthodontics 2000;13:
221—6.
5. Blatz MB, Sadan A, Kern M. Resin-ceramic bonding: a review
of the literature. J Prosthet Dent 2003;89:268—74.
6. Blatz MB, Sadan A, Arch GH, Lang BR. In vitro evaluation of
long-term bonding of Procera AllCeram alumina restorations
with a modified resin luting agent. J Prosthet Dent 2003;89:
381—7.
7. Blatz MB, Sadan A, Blatz U. The effect of silica coating on the
resin bond to the intaglio surface of Procera AllCeram
restorations. Quintessence Int 2003;34:542—7.
8. Friederich R, Kern M. Resin bond strength to densely sintered
alumina ceramic. Int J Prosthodontics 2002;15:333—8.
M. Hummel, M. Kern
9. Kern M, Thompson VP. Influence of prolonged thermal
cycling and water storage on the tensile bond strength of
composite to NiCr alloy. Dent Mater 1994;10:19—25.
10. Kern M, Thompson VP. Bonding to a glass infiltrated alumina
ceramic: adhesion methods and their durability. J Prosthet
Dent 1995;73:240—9.
11. Kern M, Wegner SM. Bonding to zirconia ceramic: adhesion
methods and their durability. Dent Mater 1998;14:64—71.
12. Komatsu K. Shrinkage stress developed by polymerization of
composite resin. J Jpn Soc Dent Mater Devices 1990;9:
520—40.
13. Siegel J, editor. Statistix. Version 4.0. User’s Manual. St.
Paul: Analytical Software; 1992. p. 127—9.
14. Holm S. A simple sequentially rejective multiple test
procedure. Scand J Stat 1979;6:65—70.
15. Kern M, Thompson VP. Sandblasting and silica-coating of
dental alloys: volume loss, morphology and changes in the
surface composition. Dent Mater 1993;9:155—61.
16. Kern M, Thompson VP. Sandblasting and silica coating of a
glass-infiltrated alumina ceramic: volume loss, morphology,
and changes in the surface composition. J Prosthet Dent
1994;71:453—61.
17. Omura I, Yamauchi J. Correlation between molecular
structure of adhesive monomer and adhesive property.
International Congress on Dental Materials. University of
Hawaii, Honolulu, Hawaii, November 4, 1989. Kurashiki,
Japan: Kuraray; 1989.
18. Lüthy H, Loeffel O, Schärer P. Zirconia posts and cores:
factors influencing retention. J Dent Res 1995;74(939):228.
19. Wegner S, Kern M. Long-term resin bond strength to zirconia
ceramic. J Adhes Dent 2000;2:139—45.
20. Edelhoff D, Abuzayeda M, Yildirim M, Spiekermann H, Marx
R. Adhäsion von Kompositen an hochfesten Strukturkeramiken nach unterschiedlicher Oberflächenbehandlung.
Deutsche Zahnärztliche Zeitschrift 2000;55:617—23.
21. Masuhara E, Yamashita A. Zahnärztliche Adhäsivkunststoffe
und ihre klinische Anwendung. Berlin: Quintessenz; 1990.
22. Marx R, Eggert G. Dentalkleber bei Wasserdiffusion und
thermischer Belastung. Deutsche Zahnärztliche Zeitschrift
1986;41:285—8.
23. Atta MO, Smith BGN, Brown D. Bond strengths of three
chemical adhesive cements to a nickel—chromium alloy for
direct bonded retainers. J Prosthet Dent 1990;63:137—43.
24. Söderholm K-JM, Roberts MJ. Influence of water exposure on
the tensile strength of composites. J Dent Res 1990;69:
1812—6.
25. Isidor F, Stokholm R, Ravnholt G. Tensile bond strength of
resin luting cement to glass infiltrated porous aluminium
oxide cores (In-Ceram). Eur J Prosthodontics Restorative
Dent 1995;3:199—202.
26. Wada T. Development of a new adhesive material and its
properties. Proceedings of the international symposium
on adhesive prosthodontics, 1986 June 24. Amsterdam,
Netherlands, Chicago: Academy of Dental Materials;
1986.