d e n t a l m a t e r i a l s 2 2 ( 2 0 0 6 ) 824–831

available at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/dema

Evaluation of resin adhesion to zirconia

ceramic using some organosilanes

Jukka P. Matinlinna a,b,∗ , Timo Heikkinen b , Mutlu Özcan c ,

Lippo V.J. Lassila b , Pekka K. Vallittu b
a NIOM—Nordic Institute of Dental Materials, P.O. Box 70, NO-1305 Haslum, Norway
b Institute of Dentistry, Department of Prosthetic Dentistry and Biomaterials Science,
University of Turku, FI-20520 Turku, Finland
c University of Groningen, Faculty of Medical Sciences, Department of Dentistry and Oral Hygiene,

NL-9713 AV Groningen, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: Objectives. This study evaluated and compared the effect of three trialkoxysilane coupling
Received 3 May 2005 agents on the bond strength of a Bis-GMA-based unfilled resin and a dimethacrylate-based
Received in revised form resin composite luting cement to a zirconia ceramics (Procera® AllZircon, Nobel Biocare,
13 September 2005 Göteborg, Sweden).
Accepted 12 October 2005 Methods. Six square-shaped zirconia specimens were used for each test group, a total of
72 specimens. The specimens in each group were all assigned to air-borne alumina parti-
cle abrasion followed by tribochemical silica-coating and silanization with 1 vol% solutions
Keywords: of 3-methacryloyloxypropyltrimethoxysilane, 3-acryloyloxypropyltrimethoxysilane, or 3-
Silane isocyanatopropyltriethoxysilane in an ethanol–water mixture. The sample stubs were made
Silica-coating of a Bis-GMA/MMA/DMAEMA resin or a commercial resin composite luting cement (RelyXTM
Silanization ARC, 3M ESPE, Seefeld, Germany). They were bonded to the conditioned and silanized silica-
Zirconia coated zirconia specimens using polyethylene molds. All specimens were tested at dry and
Bis-GMA thermo-cycled (6000, 5–55 ◦ C, 30 s) conditions. The shear bond strength of resin stubs to
Luting cement zirconia was measured in a universal testing machine (cross-head speed 1 mm/min).
Results. In dry conditions, the highest shear bond strength was 9.7 MPa (S.D. 3.3 MPa), and
for thermo-cycled samples 7.4 MPa (S.D. 2.4 MPa) was obtained with RelyXTM ARC cement
with 3-methacryloyloxypropyltrimethoxysilane. In general, thermo-cycling decreased the
bond strengths significantly for the Bis-GMA resin (ANOVA, p < 0.005). The resin, cement
and silanes differed significantly silanes (ANOVA, p < 0.005). All samples silanized with
3-isocyanatopropyltriethoxysilane de-bonded during thermo-cycling. De-bonding was dom-
inantly due to adhesive failure.
Significance. Bonding of the experimental resin and commercial cement to silica-
coated zirconia is effective with 3-methacryloyloxypropyltrimethoxysilane or 3-
methacryloyloxypropyltrimethoxysilane, but not with 3-isocyanatopropyltriethoxysilane.
© 2005 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

Corresponding author. Tel.: +358 2 333 8357; fax: +358 2 333 8390.
∗

E-mail address: [email protected] (J.P. Matinlinna).

0109-5641/$ – see front matter © 2005 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.dental.2005.11.035
 d e n t a l m a t e r i a l s 2 2 ( 2 0 0 6 ) 824–831 825

Ceramic repair and its clinical success are strongly dependent

1. Introduction
on the bond between the resin composite and the ceramic
substrate. Various applicable repair techniques have been sug-
New high-performance non-etchable ceramics, such as alu- gested and presented in literature reviews [4–6].
mina and zirconia, for fixed partial dentures and crowns, are There are reports on bond strengths of various types of
becoming commonly used in indirect restorations. Zirconia is ceramics and composites tested with various adhesive sys-
tough, has high strength and is a metal-free material, and its tems [7–10]. However, some information is lacking on shear
color is adequately white; therefore, it is used today in many bond strengths of resin composite luting cements to non-
dental ceramic systems [1], and also in biomedical applica- etchable ceramics, and in particular, no silane comparisons
tions [2]. The traditional approach of cemented restorations have been reported.
is being challenged by the development of new adhesive sys- Silane coupling agents, or more precisely trialkoxysilanes,
tems. In principle, the adhesive bond can eliminate the inter- are hybrid inorganic–organic bifunctional molecules, and their
nal surface flaws and thus reduce the potential for fracture. use as adhesion promoting agents plays a remarkable and rec-
This idea, and the intention of keeping tooth preparations to ognized role [11]. Typical silane coupling agents contain an
the minimum, has led to a growing use of dental ceramics for organofunctional part and three hydrolyzable alkoxy groups
inlays/onlays, crowns, and veneers [3]. (Fig. 1a–c). Before becoming adhesion promoters and to be
The clinical application of resin-bonded fixed partial den- activated, trialkoxysilanes must undergo a hydrolyzation reac-
tures requires a strong and stable resin bond to the ceramic. tion in slightly acidic ethanol–water solvent to form silanols
The luting cement and conditioning methods, as well as the from trialkoxy groups. The organofunctional part, most often
proper treatment of the prosthodontic application under the a methacrylate group, can then be polymerized with the
conditions of the oral cavity, play significant roles in obtain- monomers of a resin composite system [11,12].
ing the successful clinical outcome for non-etchable ceramics, Silane coupling agents lower the surface tension of a sub-
unless a retentive type of tooth preparation is used. In addi- strate, wet it and make its surface energy higher, and hence
tion to cementing, ceramic repair is sometimes a necessary accessible for effective bonding [13]. Thus, a hydrophobic
procedure for slightly fractured partial dentures and crowns. matrix (resin composite) can adhere to hydrophilic surfaces,

Fig. 1 – Silanes used in this study: (a) 3-methacryloyloxypropyltrimethoxysilane, (b) 3-acryloyloxypropyltrimethoxysilane,

(c) 3-isocyanatopropyltriethoxysilane and (d) the MDP molecule: 10-methacryloyloxydecyldihydrogen phosphate.
826 d e n t a l m a t e r i a l s 2 2 ( 2 0 0 6 ) 824–831

such as silica, glass, and glass-ceramics [14,15]. One of the lat- 498), using 110 !m grain sized aluminum trioxide particles
est presented silane coupling agent bonding theories suggests surface-modified with silica, at 280 kPa, from a perpendicular
that the silanes somehow modify the outermost inorganic distance of 10 mm for 2 × 13 s. The samples were cleaned for
interface oxide layer of the substrate [16]. 10 min in an ultrasonic bath in water-free ethanol, and then
Tribochemical silica-coating with Rocatec® (3M ESPE, dried in compressed oil-free air and kept in a desiccator before
Seefeld, Germany) is widely used in dental laboratories. The silanization.
silica-coated alumina particles are blasted onto the surface
(e.g. ceramic) for the bonding of resin or cement [5,17,18]. For 2.1. Activated trialkoxysilane solutions and
intra-oral use, the CoJetTM system has been introduced. It fol- silanization
lows the same physico-chemical principle as RocatecTM [5].
However, both technologies require a silane coupling agent A solution of 95.0 vol% ethanol (Primalco, Helsinki, Finland)
application [17]. and de-ionized water (milli-Q water, 18 M! cm) was prepared
There are some special cements available that con- and allowed to stabilize for 24 h. The pH was adjusted to
tain the acidic phosphate groups, e.g. 10-methacryloylo- 4.5 with 1 M acetic acid (Merck, Darmstadt, Germany).
xydecyldihydrogen phosphate, MDP (Fig. 1d) which is applied Then, 1.0 vol% solutions of 3-methacryloyloxypropyltrimeth-
in PanaviaTM F (Kuraray Medical, Okayama, Japan) or RelyXTM oxysilane (Sigma–Aldrich, Steinheim, Germany, Lot no.
Unicem (with methacrylated phosphoric esters; 3M ESPE, S01603-022), 3-acryloyloxypropyltrimethoxysilane (Dow Corn-
Seefeld) which both provide reliable adhesive results [7]. One ing Toray Silicone, Tokyo, Japan, Lot no. VN02011454), and 3-
cement kit is commercially available in which the silane is isocyanatopropyltriethoxysilane (ABCR, Karlsruhe, Germany,
to be mixed with MDP, and then applied onto the condi- Lot no. 28114-2F24-IS), and were prepared in ethanol in 50 mL
tioned ceramic surface. Finally, the bonding resin is applied polyethylene bottles. The silanes were allowed to hydrolyze
onto the ceramic surface (Kuraray’s Clearfil Repair KitTM ). for 1 h at room temperature before silanization [19]. The silica-
RelyXTM ARC, which has been used in this study, has as coated zirconia samples were assigned randomly to 12 sub-
its indications all ceramic/porcelain crowns, bridges, onlays groups for silanization. One silane coating was applied, each
and inlays. As such, without any silica-coating, silanes can- time with a new, clean brush, onto the ceramic substrate. The
not promote the adhesion of a resin composite onto zirconia silane was allowed to dry and react for 3 min, and then gently
surfaces [9]. air-blasted dry with oil-free air [20].
The aim of this study was to evaluate three trialkoxysilanes
and their adhesive performance on first, air-borne alumina 2.2. Bonding of resin and cement
particle treated, and then silica-coated zirconia, onto which
an experimental Bis-GMA resin and a commercial resin com- The experimental Bis-GMA resin consisted of: 78.4 wt%
posite luting cement RelyXTM ARC were bonded. Bis-GMA (Röhm, Darmstadt, Germany, Lot not available),
19.6 wt% methylmethacrylate (Fluka, Buchs, Switzerland,
Lot no. 436935/1 41702), 1.0 wt% 2-(dimethylamino)-
2. Materials and methods ethylmethacrylate (Sigma–Aldrich, Steinheim, Lot no.
BO05923KU) and 1.0 wt% (±)-camphorquinone (Fluka, Buchs,
All ceramic samples were non-etchable zirconia ceramics Lot no. 395656/1 43901). The commercial cement was RelyXTM
(Procera® AllZircon, Nobel Biocare, Göteborg, Sweden). The ARC (3M ESPE, St. Paul, MN, USA, Lot no. 20041012) and
samples were prepared and applied in a standardized way by according to the manufacturer it contains Bis-GMA, TEGDMA,
the same operator. Twelve experimental groups (n = 6) were silane-treated ceramic and silica fillers, and functionalized
studied for three silanes, and for two resins: one Bis-GMA- DMA.
based resin and a commercial RelyxTM ARC luting cement. The The low-viscous experimental Bis-GMA resin and the com-
samples were then kept in two different storage conditions. mercial cement were bonded to the silanized ceramic spec-
The specimens were planar-shaped plates, 10 mm × 10 mm imens using translucent polyethylene molds with an inner
in surface, and 3 mm in thickness. Before the bonding pro- diameter of 3.6 mm and height of 5 mm. The resin/cement was
cedure was initiated, the specimens were embedded in packed against the substrate with a composite-filling instru-
acrylic resin blocks ensuring that one area of the ceramic ment. The resin/cement stubs were light-polymerized for 40 s.
surface remained uncovered for bonding experiments and Light intensity was 470–520 mW/cm2 , with a wavelength max-
procedures. imum of 470 nm. Polyethylene molds were gently removed
As the first step in the conditioning method protocol, air- from the test specimens. While dry samples were kept in a
borne particle abrasion was performed using 110 !m grain desiccator at room temperature for 24 h prior to testing, the
sized aluminum trioxide, Al2 O3, sand (Korox® , Bego, Bremen, other groups were subjected to thermo-cycling (a custom pro-
Germany, Lot no. 401221) at a pressure of 380 kPa from a dis- cedure made by NIOM—Nordic Institute of Dental Materials,
tance of approximately 10 mm, for 2 × 13 s for the exposed Haslum, Norway) for 6000 cycles between 5 and 55 ◦ C in de-
surface of each specimen. The samples were then cleaned for ionized water. The dwelling time at each temperature was 30 s,
10 min in an ultrasonic bath (Quantrex 90 WT, L&R Manufac- and the transfer time from one bath to the other was 2 s.
turing, Inc., Lahti, Finland) in water-free ethanol. The samples Specimens were mounted in the jig of a universal testing
were gently dried in compressed oil-free air and kept in a machine (Lloyd LRX, Lloyd Instruments Ltd., Fareham, UK) and
desiccator prior to the next stage. Next, the specimens were the shear force was applied to the adhesive interface until
silica-coated with Rocatec® Plus (3M ESPE, Seefeld, Lot no. fracture occurred. The specimens were loaded at a crosshead
 d e n t a l m a t e r i a l s 2 2 ( 2 0 0 6 ) 824–831 827

speed of 1.0 mm/min and the stress–strain curve was analyzed

with Nexygen 2.0 Software (Lloyd LRX, Lloyd Instruments Ltd.).
Statistical analysis was performed using SAS System for
Windows, release 8.02/2001 (Cary, NC, USA). The means of
each group were analyzed by three-way analysis of variance
(ANOVA), with shear bond strength as the dependent variable,
the silane type and the storage type as the independent fac-
tors. Post hoc comparisons were made by Tukey’s test. p-Values
less than 0.05 were considered to be statistically significant in
all tests.

2.3. Scanning electron microscopy (SEM) and energy

dispersive X-ray analysis (EDXA)
Fig. 2 – Shear bond strengths in two storage conditions.
A SEM imaging (JSM 5500, JEOL, Tokyo, Japan) study was under- After thermo-cycling, the groups silanized with
taken to estimate the surfaces and the resin bonding on zir- 3-isocyanatopropyltriethoxysilane revealed spontaneous
conia. A standardless EDXA analysis (PRISM 2000, Princeton de-bonding. Key: ZOAA, Bis-GMA
Gamma-Tech, Princeton, USA) was carried out with acceler- resin + 3-methacryloyloxypropyltrimethoxysilane; ZOAB,
ating voltage of 20 kV in vacuum, with a working distance of Bis-GMA resin + 3-acryloyloxypropyltrimethoxysilane;
20 mm. A liquid nitrogen cooled lithium-drifted silicon detec- ZOAC, Bis-GMA resin + 3-isocyanatopropyltriethoxysilane;
tor with 30 mm2 active area was used to collect the X-ray ZOAARX, RelyXTM ARC
spectra. Analysis (n = 5/group) was made of 1.14 mm × 0.84 mm cement + 3-methacryloyloxypropyltrimethoxysilane;
areas. Position-tagged spectrometry (PTS) mode was applied ZOABRX, RelyXTM ARC
to study the element distribution of the grit-blasted surface. cement + 3-acryloyloxypropyltrimethoxysilane; ZOACRX,
Oxygen and carbon were de-convoluted from the spectra in RelyXTM ARC cement + 3-isocyanatopropyltriethoxysilane.
order to achieve the ratio of Si/Zr/Al.

type and storage on the shear bond strength values

3. Results (p < 0.005), less difference and significance was found for silane
type + resin/cement (p < 0.456), silane + storage (p < 0.140), and
The results of the shear bond strength test for the study groups cement + storage (p < 0.460). Post hoc comparisons were made
are presented in Fig. 2 and Tables 1a and 1b. While ANOVA by Tukey’s HSD test for the type of silane, when strength was
showed a significant influence of the silane type, resin/cement the dependent variable. There was no significant difference

Table 1a – Test groups in dry conditions

Sample code Silane n/group Resin or cement Shear bond Standard deviation
strength (MPa) (MPa)

ZOAA MPS 6 Unfilled 7.6 2.9

ZOAB ACPS 6 Unfilled 7.8 3.8
ZOAC ICS 6 Unfilled 3.4 1.3
ZOAARX MPS 6 PFC 9.7 3.3
ZOABRX ACPS 6 PFC 5.8 1.9
ZOACRX ICS 6 PFC 7.5 4.8

Abbreviations: See also Fig. 2. Key: MPS, 3-methacryloyloxypropyltrimethoxysilane; ACPS, 3-acryloyloxypropyltrimethoxysilane; ICS, 3-
isocyanatopropyltriethoxysilane; PFC, particulate filled composite.

Table 1b – Test groups after thermo-cycling

Sample code Silane used n/group Resin or cement De-bonded during Shear bond Standard deviation
thermo-cycling (%) strength (MPa) (MPa)

ZOAA MPS 6 Unfilled 16 3.9 2.6

ZOAB ACPS 6 Unfilled 32 2.4 0.4
ZOAC ICS 6 Unfilled 100 0 0
ZOAARX MPS 6 PFC 0 7.4 2.4
ZOABRX ACPS 6 PFC 0 6.1 2.4
ZOACRX ICS 6 PFC 100 0 0

Abbreviations as given in Table 1a.

828 d e n t a l m a t e r i a l s 2 2 ( 2 0 0 6 ) 824–831

between 3-methacryloyloxypropyltrimethoxysilane and 3-

Table 2 – Chemical analysis (EDXA) of the silica-coated
acryloyloxypropyltrimethoxysilane, but 3-isocyanatopropyl- zirconia surfaces
triethoxysilane differed significantly from the other two
Element Weight% Atomic concentration (%)
silanes.
The highest shear bond strengths were obtained in dry Si 4.2 11.2
conditions (9.7 ± 3.3 MPa) and they were significantly higher Al 4.8 13.6
than those for thermo-cycled stub samples. Only for RelyXTM Zr 91.0 75.2

ARC and 3-acryloyloxypropyltrimethoxysilane, the shear bond

strengths for dry and thermo-cycled conditions did not dif- that has been conditioned with both abrasion techniques
fer significantly. The highest strength value was for RelyXTM and silanized with 3-acryloyloxypropyltrimethoxysilane and
ARC with 3-methacryloyloxypropyltrimethoxysilane, for both onto which the commercial luting cement has been light-
storage conditions. The lowest result in dry conditions was polymerized, suggests partially cohesive failure (Fig. 3d), for
for Bis-GMA resin and 3-isocyanatopropyltriethoxysilane. All all thermo-cycled samples. According to the EDXA analysis,
silanizations with 3-isocyanatopropyltriethoxysilane had led the silica-coated zirconia surface contained silicon 4.2 wt%, or
to de-bonding during thermo-cycling. Also some de-bonding 11.2% as atomic concentration (Table 2; Fig. 4a–d).
took place when the thermo-cycled resin stubs were of Bis-
GMA resin. 4. Discussion
The SEM image analysis suggests some differences
between alumina sand blasted (Fig. 3b), and alumina Resin bonding to high strength non-etchable ceramics is not
sand + silica-coated zirconia surface (Fig. 3c), in particular always predictable and requires different bonding methods
compared to non-grit blasted zirconia (Fig. 3a). Zirconia from silica based dental ceramics. Thus, further clinical and

Fig. 3 – (a) Zirconia sample before sand blasting (SEM image, magnification 700×). (b) Alumina sand blasted zirconia (SEM
image, magnification 700×). (c) Alumina sand blasted + silica-coated zirconia (SEM image, magnification 700×). (d) The
contact area of the commercial resin composite cement stub and alumina sand blasted + silica-coated zirconia, after shear
bond testing: remains of the luting cement can be seen (SEM image, magnification 100×).
 d e n t a l m a t e r i a l s 2 2 ( 2 0 0 6 ) 824–831 829

Fig. 4 – (a) Silica-coated zirconia surface (Si K!, magnification 250×). The EDXA mapping images (b–d) are based on this SEM
micrograph. Silica-coated zirconia surface image showing elemental distribution of (b) Al, (c) Si and (d) Zr, based on EDXA
analysis (magnification 250×).

in vitro studies are needed [3]. Silane coupling agents, in par- experimental Bis-GMA resin was bonded to silica-coated and
ticular 3-methacryloyloxypropyltrimethoxysilane, are well- silanized titanium [19].
known adhesion promotion reagents between Si-containing According to the EDXA analysis, the Si-content on the
surfaces and organic polymeric material, but there are silica-coated surface was rather low, 11.2% (Table 2). To a large
many studies showing that the silane alone does not pro- extent, silica-coverage originating from the coating particles
vide a reliable bond with zirconia (or alumina) ceramics appears not to have become embedded on the hard zirconia
[3,8–10,20–22]. In some studies, silane coupling agents have surface (cf. Fig. 3c). The presence of silica is a prerequisite for
been found to improve the bond of resin composite to durable siloxane bonding: when one considers the chemical
ceramics by about 25% [23]. In this study, the experimental composition of the ceramic surface, the Si–O–Si bonds that
silanes have been hydrolyzed as reported in literature and might form on it, are known to be hydrolytically more stable
they can be considered activated [19,24]. It has been con- than Si–O–Al bonds on the resin-to-ceramic interface [26].
cluded that for silica-based dental ceramics, air-borne par- The same findings have been confirmed and reported later
ticle conditioning did not provide enough retention with- [27]. On the silica-coated surfaces, there was 75.2% zirconium
out a successive silane treatment [25]. On the other hand, as calculated in atomic concentration: apparently Zr–O–Si
it has been recently observed that some rare silanes, such bonds exist on the interface but they do not withstand the
as 3-isocyanatopropyltriethoxysilane, which have not been hydrothermal effects and changes. The zirconia surfaces were
reported to be used as adhesion promoters in dental materials obviously too silicon-poor for durable bonding.
research, have yielded shear bond strengths that were compa- On the other hand, SEM images with a magnification of
rable to 3-methacryloyloxypropyltrimethoxysilane when an 700× in this study show some differences between alumina
830 dental materials 22 ( 2 0 0 6 ) 824–831

sand blasted and alumina sand blasted + silica-coated zir-

conia surfaces, compared to the original zirconia sur-
5. Conclusions
face. According to the SEM images, the resin stubs have
some remains of the resin or luting cement after thermo- This silane adhesion study, with its limitations, has led the
cycling, which suggests mainly adhesive failure when 3- authors to make the following conclusions:
acryloyloxypropyltrimethoxysilane was used for silanization
(Fig. 3d). In general, the surface texture of the silica-coated 1. Shear bond strengths of the unfilled resin composite and
zirconia has perhaps not been coarse enough to provide resin composite luting cement tested on the zirconia
micromechanical interlocking. A widely accepted suggestion ceramic after three different silanizations varied, depend-
for adequate shear bond strength is 5 MPa according to an ing on the storage and the silane.
ISO standard [28], and all samples of the luting cement 2. 3-Methacryloyloxypropyltrimethoxysilane did not signif-
have exceeded this limit, except those silanized with 3- icantly differ from 3-acryloyloxypropyltrimethoxysilane,
isocyanatopropyltriethoxysilane. when shear bond strength results for dry or thermo-cycled
Currently, methods for hydrolytically stable bonding of samples were compared.
non-etchable ceramics are sought. Grit-blasting with alumina 3. Thermo-cycling decreased the bond strength values signif-
blasting creates high surface energy and in principle pro- icantly for the Bis-GMA resin, but not for the RelyXTM ARC
motes micro-retention. It has been concluded that roughening luting cement. De-bonding was dominantly due to adhe-
of the substrate surface promotes adhesion since it allows sive failure.
the polymer (resin composite) to flow into the surface and 4. 3-Isocyanatopropyltriethoxysilane did not promote adhe-
forms irregularities on the substrate surface [29]. An imme- sion when bonding an experimental Bis-GMA resin or the
diate silica-coating, after ultra-sonic cleaning in ethanol, can RelyXTM ARC luting cement to silica-coated zirconia.
leave a chemically active outer surface for chemical bond-
ing with silanes. However, the pressure and application time
using the RocatecTM system are obviously important fac- Acknowledgements
tors. Some other findings suggest that silica-coating by the
RocatecTM had proved to be insufficient for bonding a Bis-GMA This pilot study was financially supported by grants from
resin to zirconia, in conjunction with a silanization with 3- NIOM (Nordic Institute of Dental Materials, Haslum, Norway)
methacryloyloxypropyltrimethoxysilane [9]. and it was a part of the ‘Bio- and Nanopolymers Research
The three trialkoxysilanes that have now been evaluated Group activity of the Centre of Excellence’ of The Academy
have quite similar molecular structure but the organofunc- of Finland. Nobel Biocare AB (Göteborg, Sweden) is thanked
tional groups vary. The dimethacrylates (cf. methacrylate for the Zirconia samples and Dow Corning Ltd., UK (Mr. Bryan
group in Fig. 1a) existing in RelyXTM ARC have apparently Thomas) is thanked for donating silanes to our study and 3M
polymerized with the methacrylate and acrylate groups of the ESPE Oy, Finland (Mr. Artti Juusela) for donating the commer-
silanes. The chemical reactions with the isocyanato groups cial cement. Mr. John Wright (London, UK) is thanked for the
to form durable bonds have apparently failed (Fig. 1c). It is proofreading.
notable that the 3-acryloyloxypropyltrimethoxysilane group
did not exhibit a decrease in the bond strength after thermo- references
cycling. The reason might be that to some extent the ele-
vated temperature in thermo-cycling process has promoted
the polymerization (post-polymerization) of acrylate groups [1] van Noort R. An introduction to dental materials. Hong
with the remaining monomers of the resin/cement matrix. Kong: Elsevier Science; 2002.
This effect should be studied in detail in the near future. In [2] Piconi C, Maccauro G. Zirconia as a ceramic biomaterial.
Biomaterial 1999;20:1–25.
general, acrylate groups are known to be more reactive than
[3] Blatz MB, Sadan A, Kern M. Resin–ceramic bonding: a
methacrylate groups that most typically are employed as the
review of the literature. J Prosthet Dent 2003;89:268–
organofunctional groups in silanes for numerous applications 74.
[30]. All these silane solutions have been activated according [4] Rosentritt M, Behr M, Kolbeck C, Lang R, Handel G. In vitro
to the literature before their application and they can be sup- repair of all-ceramic and fibre-reinforced composite
posed to be ready for use [19,24]. The shear bond strength crowns. Eur J Prosthodont Restor Dent 2000;8:107–12.
results after thermo-cycling were higher with the filler resin [5] Frankenberger R, Krämer N, Sindel J. Repair strength of
etched vs silica-coated metal-ceramic and all-cearmic
composite on the grit-blasted and silanized zirconia surface,
restorations. Oper Dent 2000;25:209–15.
probably due to the better match of thermal expansion coeffi- [6] Pameijer CH, Louw NP, Fischer D. Repairing fractured
cients. The commercial cement contains fillers in the matrix porcelain: how surface preparation affects shear force
that makes it more suited to the zirconia. On the other hand, resistance. J Am Dent Assoc 1996;127:203–9.
the thermal expansion coefficient of unfilled resin is higher [7] Hummel M, Kern M. Durability of the resin bond strength
than that of a filler composite, and it is probably closer to to the alumina ceramic Procera. Dent Mater
2004;20:498–508.
the thermal expansion coefficient of zirconia. This might also
[8] Wegner SM, Kern M. Long-term resin bond strength to
explain why a commercial cement gave more reliable bonding
zirconia ceramic. J Adhes Dent 2000;2:139–47.
to silica-coated and silanized zirconia [31]. The three activated [9] Kern M, Wegner SM. Bonding to zirconia ceramic: adhesion
silanes gave differences and an optimization in the near future methods and their durability. Dent Mater 1998;14:64–
would be of great interest. 71.
 dental materials 22 ( 2 0 0 6 ) 824–831 831

[10] Kern M, Thompson VP. Bonding to glass infiltrated [21] Kern M, Thompson VP. Sandblasting and silica coating of
alumina ceramic: adhesive methods and their durability. J a glass-infiltrated alumina ceramic: volume loss,
Prosthet Dent 1995;73:240–9. morphology, and changes in the surface composition. J
[11] Matinlinna J, Lassila LVJ, Özcan M, Yli-Urpo A, Vallittu PK. Prosthet Dent 1994;71:453–61.
An introduction to silanes and their clinical applications [22] Özcan M, Vallittu PK. Effect of surface conditioning
in dentistry. Int J Prosthodont 2004;17:155–64. methods on the bond strength of luting cement to
[12] Subramanian PR, van Ooij W. Silane based metal ceramics. Dent Mater 2003;19:725–32.
pretreatments as alternatives to chromating. Surf Eng [23] Diaz-Arnold AM, Aquilino SA. An evaluation of the bond
1999;15:168–72. strengths of four organosilane materials in response to
[13] Marsden JG. Organofunctional silane coupling agents. In: thermal stress. J Prosthet Dent 1989;62:257–60.
Skeist I, editor. Handbook of adhesion. New York: Van [24] Hooshmand T, van Noort R, Keshwad A. Storage effect of
Nostrand Reinhold; 1990. a pre-activated silane on the resin to ceramic bond. Dent
[14] Plueddemann EP. Adhesion through silane coupling Mater 2004;20:635–42.
agents. J Adhes 1970;2:184–201. [25] O’Kray K, Suckak AJ, Stanford JW. Shear strength of
[15] Plueddemann EP. Water is the key to new theory on porcelain repair materials. J Dental Res 1987;66. Abstract
resin-to-fiber bonding. Mod Plast 1970;3:92–8. 805.
[16] Child TF, van Ooij W. Application of silane technology to [26] Plueddemann EP. Silane coupling agents. 2nd ed. New
prevent corrosion of metals and improve paint adhesion. York: Plenum Press; 1991.
Trans Inst Metal Finish 1999;77:64–70. [27] Friederich R, Kern M. Resin bond strength to densely
[17] Guggenberger R. Das Rocatec System – Haftung durch sintered alumina ceramic. Int J Prosthodont 2002;15:
tribochemische Beschichtung. Dtsch Zahnarztl Z 333–8.
1989;44:874–6. [28] ISO, International Organization for Standardization:
[18] Özcan M, Pfeiffer P, Nergiz Y. A brief history and current Dentistry-Polymer-Based Crown and Bridge Materials,
status of metal/ceramic surface-conditioning concepts for Amendment 1996; ISO 10477. 1996.
resin bonding in dentistry. Quintessence Int [29] Jennings CW. Surface roughness and bond strength of
1998;29:713–24. adhesives. J Adhes 1972;4:25–38.
[19] Matinlinna JP, Lassila LVJ, Kangasniemi I, Vallittu PK. [30] Matinlinna J, Lassila LVJ, Vallittu PK. The effect of a novel
Isocyanato- and methacryloxysilanes promote Bis-GMA silane blend system on silica-coated Ti substrate. J Dent
adhesion to titanium. J Dent Res 2005;84:360–4. 2005; in press.
[20] Özcan M. The use of chairside silica coating for different [31] Tezvergil A, Lassila LVJ, Vallittu PK. The effect of fiber
dental applications: a clinical report. J Prosthet Dent orientation on the thermal expansion coefficients of the
2002;87:469–72. fiber reinforced composites. Dent Mater 2003;19:471–7.