|
Sign In to gain access to subscriptions and/or personal tools.
|
Biomaterials & Bioengineering |
Isocyanato- and Methacryloxysilanes Promote Bis-GMA Adhesion to Titanium
J.P. Matinlinna1,*,
L.V.J. Lassila1,
I. Kangasniemi2 and
P.K. Vallittu1
1 Institute of Dentistry, Department of Prosthetic Dentistry and Biomaterials Research, University of Turku, Lemminkäisenkatu 2, FI-20520 Turku, Finland; and
2 Vivoxid Ltd., Tykistökatu 4 A, FI-20520 Turku, Finland;
Correspondence: * corresponding author, jukka.matinlinna{at}utu.fi
 |
ABSTRACT
|
|---|
In dentistry, adhesion promotion with 3-methacryloyloxypropyltrimethoxysilane is usually sufficient, but its hydrolytic stability is a continuous concern. The hydrolytic stability of an alternative, 3-isocyanatopropyltriethoxysilane, was compared with that of conventional 3-methacryloyloxypropyltrimethoxysilane. Two silanes, both in 0.1 and 1.0 vol-% in ethanol-water, were evaluated in the attachment of an experimental bis-phenol-A-diglycidyldimethacrylate (Bis-GMA) resin to grit-blasted (with two different systems) titanium. Silane hydrolysis was monitored by FTIR spectrometry. Bis-GMA resin was applied and photo-polymerized on titanium. The specimens were thermocycled (6000 cycles, 5–55°C). Surface analysis was carried out with scanning electron microscopy. Statistical analysis (ANOVA) showed that the highest shear bond was achieved with 0.1% 3-isocyanatopropyltriethoxysilane (12.5 MPa) with silica-coating, and the lowest with 1.0% 3-methacryloyloxypropyltrimethoxysilane (3.4 MPa) with alumina-coating. The silane, its concentration, and the grit-blasting method significantly affected the shear bond strength (p < 0.05). SEM images indicated cohesive failure of bonding, and, in conclusion, 3-isocyanatopropyltriethoxysilane is a potential coupling agent.
Key Words: silanization • FTIR • silica-coating • trialkoxysilanes • titanium
|
INTRODUCTION
|
|---|
The resin system most associated with restorative dentistry is bis-phenol-A-diglycidyldimethacrylate, Bis-GMA (Bowen, 1956). It can be polymerized quickly via a free-radical mechanism. Bis-GMA yields a cross-linked polymer that is also used in veneering composites. The relatively large Bis-GMA monomer molecule has the effect of reducing polymerization shrinkage on setting. The effective shrinkage is further reduced through the incorporation of an inert silanized filler (Bowen and Rodriguez, 1962; Bowen and Marjenhoff, 1992; Venhoven et al., 1994). Bis-GMA is an extremely viscous liquid, and it needs to be diluted with monomers such as methylmethacrylate to be applicable. In prosthodontics, the bonding of veneering composite to metal alloys is required. Titanium is popular in biomedicine and dentistry, due to its excellent biocompatibility. Ti is a commonly used metal, e.g., in a CAD/CAM milled implant fixed partial denture substructure that is veneered with a particulate filled composite. The extremely thin and compact oxide layer on Ti is mainly thermodynamically stable rutile TiO2, but non-stoichiometric oxides also occur (Murray and Wriedt, 1990).
Bifunctional trialkoxysilane esters may join essentially different materials together. Two frequently used are vinyltriethoxysilane (Bowen, 1963; Clark and Plueddemann, 1963) and 3-methacryloyloxypropyltrimethoxysilane (Paffenbarger et al., 1967). Silanes alter the surface energy and wettability of an inorganic substrate, and they have an organofunctional group that can co-polymerize with the composite resin. The hydrolyzable alkoxy groups react in aqueous alcohol solution, forming silanols,  Si-OH.
Two key condensation reactions are supposed to take place during activation. The intermediate reactive silanol groups condense to form dimeric  oligomeric molecules. With metals containing hydroxyl groups, Si-O-M-, i.e., metal-siloxane bonds, will be formed. The rate of condensation between silanol groups is minimized at ca. pH 4-5 (Plueddemann, 1970; Arkles, 1997; Child and van Ooij, 1999):
 | (1) |
With silica surfaces, Si-O-Si(interface) siloxane bonds form. The second condensation reaction takes place among the adsorbed excess Si-OH groups on the substrate. This yields a highly cross-linked siloxane film:
 | (2) |
The silanization of the E-glass fibers for prosthetic dental appliances has been investigated (Vallittu, 1997a; Vallittu and Ekstrand, 1999). Silanes are also applied for surface treatment of filler materials, such as silica (Mohsen and Craig, 1995), titania (Yoshida et al., 2001), and glass particles (Condon and Ferracane, 1997). Trialkoxysilane-treated polished Ti, without grit-blasting, has shown some adhesion to veneering composite resin (Matinlinna et al., 2004b).
Tribochemical silica-coating with Rocatec® (3M ESPE, Seefeld, Germany) is used in dental laboratories. The silica-coated alumina particles are blasted onto the surface for the bonding of resin to a metal, ceramic, or acrylic surface. The particles hit the surface, creating locally high temperatures (up to 1200°C) due to kinetic energy. The impregnation and embedding of a fresh SiO2 layer into the surface takes place, and a high-energy surface is formed. An immediate 3-methacryloyloxypropyltrimethoxysilane application provides chemical covalent bonds between the substrate layer and the resin (Guggenberger, 1989). In BIFA® (BIFA, Ramat Gan, Israel), the grit-blasted surface is alumina-coated. Both systems also provide micromechanical retention for the resin.
The silane performance, the effects of two grit-blasting systems, and the adhesion of a Bis-GMA resin to titanium were evaluated. To the best of our knowledge, 3-isocyanatopropyltriethoxysilane has not been widely investigated. There are similarities between 3-isocyanatopropyltriethoxysilane and 3-methacryloyloxypropyltrimethoxysilane (Figs. 1a , 1b). Both are (a) trialkoxysilanes, (b) bifunctional, and (c) organofunctional with a propylene link to the Si atom. The hypothesis was that 3-isocyanatopropyltriethoxysilane performs better than 3-methacryloyloxypropyltrimethoxysilane in bonding a Bis-GMA resin to silica-coated titanium, due to its different chemical properties (Fig. 1c).
View larger version (13K):
[in this window]
[in a new window]
|
Figure 1. Molecular structures of silanes, an isocyanato group bonding to a hydroxyl-rich surface, and the shear bond strength results. (a) 3-Isocyanatoproplytriethoxysilane, (b) 3-methacryloyloxypropyltrimethox-silane, (c) a suggested mechanism for the isocyanato group reaction with an inorganic surface, and (d) shear bond strengths of the Bis-GMA resin stubs (N = 10) to titanium with various silanes, silane concentration, and grit-blasting systems (Abbreviations: MPS = 3-methacryloyloxypropyltrimethoxysilane; ICS = 3-isocyanatopropyltriethoxysilane).
|
|
|
MATERIALS & METHODS
|
|---|
Titanium Preparation
Commercially pure grade 2 Ti (Permascand, Ljungaverk, Sweden) was cut into 20 mm x 40 mm x 1 mm slides (N = 16). The upper halves of the surfaces (ca. 10 mm x 40 mm) were grit-blasted with Rocatec® Junior, with the use of Rocatec® Plus abrasive sand, which was 110-µm-grain-sized alumina surface-modified with silica (3M ESPE, Seefeld, Germany), in a jet at 280 KPa from a perpendicular distance of 10 mm for 10 sec. The BIFA® system was utilized in an identical manner, with Korox® 50 sand (BEGO, Bremen, Germany), 99.6% alumina particles with 50-µm grain size.
Trialkoxysilanes
A 95vol% ethanol solution (Primalco, Helsinki, Finland) in de-ionized water (milli-Q water, 18 M cm) was prepared and stabilized for 24 hrs. The pH was adjusted to 4.5 with 1 M acetic acid (Merck, Darmstadt, Germany). Solutions (0.1 vol% and 1.0 vol%) of both 3-isocyanatopropyltriethoxysilane (ABCR, Karlsruhe, Germany) and 3-methacryloyloxypropyltrimethoxysilane (Sigma-Aldrich, Steinheim, Germany) were prepared in the ethanol in 50-mL polyethylene bottles. The silanes were allowed to hydrolyze for 1 hr at room temperature.
Bonding
One silane coating was applied to the Ti substrate, each time with a new, clean brush (Özcan, 2002) (N = 16). The silane was allowed to react for 5 min, and was then gently air-blasted dry. The Bis-GMA resin consisted of: 78.4% Bis-GMA (Röhm, Darmstadt, Germany), 19.6% methylmethacrylate (Fluka, Buchs, Switzerland), 1.0% 2-(dimethylamino)ethylmethacrylate (Sigma-Aldrich, Steinheim, Germany), and 1.0% (±)-camphorquinone (Fluka, Buchs, Switzerland). It was applied in polyethylene molds as 2-mm-diameter and 4-mm-high stubs, with 5 evenly placed on the upper horizontal borders of the Ti slide. The stubs were photo-polymerized with a hand-unit (Optilux 501, SDS Kerr, Danbury, CT, USA) for 40 sec. Final polymerization was carried out in an ESPE Visio® BETA Vario light-curing vacuum unit (ESPE, Seefeld, Germany) for 15 min. We removed the molds carefully by pressing the cured resin stub to the substrate.
Thermocycling and Shear Bond
The Ti slides with resin stubs were subjected to thermocycling for 6000 cycles at temperatures alternating between 5 and 55 °C, with an immersion time of 30 sec (Heto CBN 18-30 Baths, Alleroed, Denmark). Shear bond tests for the stubs were carried out with a universal materials testing machine (LRX®, Lloyd Instrument, Fareham, UK) with Nexygen® software, at a cross-head speed of 1.0 mm min–1.
Statistical Analysis
The data were statistically analyzed by two-way analysis of variance, ANOVA (SPS, Chicago, IL, USA). The dependent variable (shear bond strength) was explained by the independent variables: type of silane, silane concentration, and grit-blasting method.
Scanning Electron Microscopy (SEM) and Energy-dispersive X-ray Analysis (EDS)
We undertook a SEM (JSM 5500, Jeol, Tokyo, Japan) study to analyze the surfaces and the resin bonding on Ti. A standardless EDS analysis was carried out with accelerating voltage of 20 kV in vacuum, with a working distance of 20 mm. A liquid-nitrogen-cooled lithium-drifted silicon detector with 30 mm2 active area (PRISM 2000, Princeton Gamma-Tech, Princeton, NJ, USA) was used to collect the x-ray spectra. The analysis (n = 5/group) was comprised of 1.14 mm x 0.84 mm areas. We applied the position-tagged spectrometry (PTS) mode to look at the element distribution of the grit-blasted surface. Oxygen and carbon were de-convoluted from the spectra to achieve the ratio of Al/Ti/Si.
Infrared Spectroscopy
We used a Fourier transform infrared (FTIR) spectrophotometer (Perkin Elmer Spectrum One, Perkin-Elmer, Beaconsfield, UK) at a resolution of 2 cm–1 to take 32 scans. Silane hydrolysis was followed with an inert, attenuated total reflectance (ATR) device (Perkin-Elmer, Beaconsfield, UK), with 2–3 drops of fresh silane applied to the ATR device. The observation points for hydrolysis were t = 0 min, 30 min, and 60 min.
|
RESULTS
|
|---|
Mechanical Properties
The highest shear bond strengths for thermocycled, grit-blasted silanized samples (12.5 MPa, standard deviation 5.8 MPa) were obtained by the use of 0.1% 3-isocyanatopropyltriethoxysilane for silica-coating. For 1.0% 3-isocyanatopropyltriethoxysilane, the mean was 5.6 (± 3.1) MPa, and for 1.0% 3-methacryloyloxypropyltrimethoxysilane, the mean was 7.2 (± 4.3) MPa. The lowest shear bond strengths were obtained with 1.0% 3-methacryloyloxypropyltrimethoxysilane for alumina-coating, 3.4 (± 1.4) MPa. For all the samples, when only the grit-blasting method was considered, alumina-coating yielded 5.6 (± 3.1) MPa, while silica-coating gave 9.0 (± 4.7) MPa. For the silane type, the mean shear bond strengths were, for 3-isocyanatopropyltriethoxysilane, 7.8 (± 4.8) MPa, and, for 3-methacryloyloxypropyltrimethoxysilane, 6.7 (± 3.7) MPa (Fig. 1d ). ANOVA showed that the means differed significantly in the silane concentration (p < 0.05, F = 5.387), in the grit-blasting (p < 0.001, F = 19.377), and in the silane and its concentration (p < 0.001, F = 12.941).
SEM/EDS Analysis
A grit-blasted non-silanized Ti substrate contained the embedded silica particles (Figs. 2a, 2c). Results of EDS analysis (Table) revealed that alumina-coating resulted in an average 38.1% (atomic concentration, ac) of Al2O3 on the titanium surface. After silica-coating, the Al2O3 content was 21.5 ac-%, and SiO2 19.4 ac-%. An EDS morphological analysis is presented (Figs. 2b, 2d). An uneven distribution of Al and Si on the titanium substrate could be observed. A quantitative analysis taken from a mapped PTS file suggested that both techniques led to Al2O3-, Ti-rich areas, and, in particular, silica-coating left a SiO2-rich area. Si is found in cases where Al content is high as well as low. Both grit-blasting methods seemed to leave a Ti-rich area with a relatively low Al and Si content. The silica-coated Ti sample silanized with 0.1% 3-isocyanatopropyltriethoxysilane, after shear bond testing, is shown in Fig. 2e. The Bis-GMA resin left in the surface texture suggested that the de-bonding had been due to a cohesive failure. Images of an alumina-coated sample, silanized with 1.0% 3-methacryloyloxypropyltrimethoxysilane (the lowest shear bond value), suggested a mostly adhesive failure.
FTIR Analysis
The 1.0% 3-isocyanatopropyltriethoxysilane gave stronger –N=C=O and –N=C signals (ca. 2268 cm–1) than the 0.1%. Si-O-CH2CH3 peaks (1480–1300 cm–1) appeared smaller compared with those associated with the silanols (ca. 3385 cm–1, overlapped). The signal at 2886 cm–1 indicated –C=O from acetic acid. At ca. 1650 cm–1, signals for water and double-bond C=C overlapped (Fig. 3a). In the 3-methacryloyloxypropyltrimethoxysilane spectra, a symmetric C-H stretch of the Si-O-CH3 group became smaller (ca. 2974 cm–1) and the Si-OH stretching mode increased (ca. 879 cm–1). Strong bands at ca. 1055 cm–1 were due to the asymmetric siloxane network stretching. An increase of free water and hydroxyl groups (and silanols) could be observed at ca. 3340 cm–1. Water was also identified at ca. 1650 cm–1 (hiding the C=C stretching, and CH3, CH2 bending modes). The hydrogen-bonded OH-group band (ca. 3400–3200 cm–1) was overlapped by a broad band. –C=O peaks at 1718 cm–1 seemed to be stronger for 1.0% than 0.1% 3-methacryloyloxypropyltrimethoxysilane, but at ca. 2887–2893 cm–1 were similar (Fig. 3b).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 3. Activation (hydrolysis) of the experimental silanes. (a) Hydrolysis of 0.1% 3-isocyanatopropyltriethoxysilane. A: t = 0 min, B = 30 min, C = 60 min. X-axis: wave number (cm–1) and Y-axis absorbance A (in arbitrary units). (b) Hydrolysis of 1.0% 3-methacryloyloxypropyltrimethoxysilane. A: t = 0 min, B = 30 min, C = 60 min. X-axis: wave number (cm–1) and Y-axis absorbance A (in arbitrary units).
|
|
|
DISCUSSION
|
|---|
Grit-blasting is particularly relevant to dental casting alloys, titanium, ceramics, composite, and also to dental hard tissues. It creates high surface energy, promotes microretention, and can leave an active outer surface for chemical bonding with silanes. Rocatec® sand has a bigger particle size (110 µm) than Korox® sand (50 µm). Chemical bonding to silica-coated surfaces could be stronger, since there was a fresh silica layer on the titanium surface, whereas BIFA®-treatment created a fresh alumina-coating. It is known that Si-O-Si bonds are hydrolytically more stable than Si-O-Al= bonds at the interface (Plueddemann, 1991). The same findings were reported later (Friederich and Kern, 2002). The EDS analysis revealed quite high alumina content on the titanium surface, which could be due to soft titanium. However, the distribution of elements on the surface (for both grit-blasting techniques) was very uneven, when the surface was scanned at higher magnification (x1000), contrary to some findings reported (Vallittu, 1997b). One study found a 9 wt-% SiO2 content on a ceramic surface after silica-coating (Kern and Thompson, 1994). The alumina-coated Ti surface revealed 2.1 ac-% of Si that might be due to impurities and silanization.
EDS is not a totally surface-sensitive method, but the recorded x-ray signals came from a volume near the surface (ca. 1 µm depth), depending on the material and the acceleration voltage. Oxygen was de-convoluted from the analysis, because it was uncertain whether the signal was coming from Ti-metal or TiO2 (Friel, 1998). It should also be noted that an optimal EDS analysis requires a flat surface, because the x-rays cannot escape from deep cavities. This effect can be recognized as black areas (Figs. 2b, 2d).
Commercial 3-methacryloyloxypropyltrimethoxysilane products are pre-hydrolyzed and widely applied clinically (Matinlinna et al., 2004a). A complete (100%) hydrolysis is not required for the trialkoxysilanes to perform effectively (Arkles, 1997). Hydrolysis of the fresh 3-methacryloyloxypropyltrimethoxysilane and 3-isocyanatopropyltriethoxysilane took place within 60 min, and in particular for 3-isocyanatopropyltriethoxysilane as an aminosilane, almost immediately (Figs. 3a, 3b). All shear bond strength values for 0.1% silane solutions exceeded the limiting bond value (Fig. 1d), which is 5 MPa according to ISO Standard 10477 (ISO, 1996). Resin samples exceeded the same limit, when both silanes were used (on silica-coating) at a 1.0% concentration. The results for 0.1% 3-isocyanatopropyltriethoxysilane were comparable with those of 1.0% 3-methacryloyloxypropyltrimethoxysilane. The reason for the good, promising bonding properties of 3-isocyanatopropyltriethoxysilane might be explained by the ability of the isocyanato group to bond to both matrices in two ways, viz., with a nitrogen bond to the Bis-GMA (Antonucci et al., 2002) and with a =N-C(O2)Si bond to the silica-rich surface. A reaction mechanism (Fig. 1c) for the isocyanato group bonding covalently with the surface hydroxyl groups has been presented (van Noort, 2002). The methacrylate group in the 3-methacryloyloxypropyltrimethoxysilane does not have this dual bonding ability. 3-Isocyanatopropyltriethoxysilane also forms typical siloxane bonds with silica. The 1.0% 3-methacryloyloxypropyltrimethoxysilane provided relatively high bond strength for the Bis-GMA resin, in accordance with earlier observations (Ekstrand et al., 1988). In general, the temperature change during the thermocycling (+5°C-+55°C) did not de-bond the resin samples from substrates.
The experimental Bis-GMA resin did not contain fillers, in contrast to veneering composites. The thermal expansion coefficient of unfilled resin is higher than that of a filler composite (Tezvergil et al., 2003). The shear bond strength results might have been higher with the addition of filler composite to the grit-blasted and silanized Ti surface, due to the better match of thermal expansion coefficients. This also merits further study. The hypothesis of this study was confirmed: 3-isocyanatopropyltriethoxysilane appeared to have a role in the adhesion promotion bonding of Bis-GMA resin to silica-coated titanium.
|
ACKNOWLEDGMENTS
|
|---|
This investigation was financially supported by TEKES (National Technology Agency of Finland), and it was a part of the Bio- and Nanopolymers Research Group activity of the Centre of Excellence of Academy of Finland. Turku-Dental Laboratory Ltd. and Mr. Mikko Jokinen (BSc) are both acknowledged.
Received for publication September 29, 2003.
Revision received January 20, 2005.
Accepted for publication January 28, 2005.
|
REFERENCES
|
|---|
- Antonucci JM, Fowler BO, Dickens SH (2002). Interaction of a silane coupling agent with dental monomers (abstract). J Dent Res 81(Spec Iss A):A-140.
- Arkles B (1997). Silicon esters. In: Kirk-Othmer encyclopedia of chemical technology. Othmer K, editor. London: John Wiley & Sons, pp. 69–81.
- Bowen RL (1956). Use of epoxy resins in restorative materials. J Dent Res 35:361–369.
- Bowen RL (1963). Properties of a silica-reinforced polymer for dental restorations. J Am Dent Assoc 66:57–64.[Medline]
[Order article via Infotrieve]
- Bowen RL, Marjenhoff WA (1992). Dental composites/glass ionomers: the materials. Adv Dent Res 6:44–49.[Abstract/Free Full Text]
- Bowen RL, Rodriguez MS (1962). Tensile strength and modulus of elasticity of tooth structure and several restorative materials. J Am Dent Assoc 64:378–387.[Medline]
[Order article via Infotrieve]
- Child TF, van Ooij W (1999). Application of silane technology to prevent corrosion of metals and improve paint adhesion. Trans Inst Metal Finish 77(2):64–70.
- Clark HA, Plueddemann EP (1963). Bonding of silane coupling agents in glass-reinforced plastics. Modern Plastics 40(10):133–196.
- Condon JR, Ferracane JL (1997). In vitro wear of composite with varied cure, filler level, and filler treatment. J Dent Res 76:1405–1411.
- Ekstrand K, Ruyter IE, Oysaed H (1988). Adhesion to titanium of methacrylate-based polymer materials. Dent Mater 4(3):111–115.[Medline]
[Order article via Infotrieve]
- Friederich R, Kern M (2002). Resin bond strength to densely sintered alumina ceramic. Int J Prosthodont 15(4):333–338.[Medline]
[Order article via Infotrieve]
- Friel JJ (1998). X-ray and image analysis in electron microscopy. Princeton: Princeton Gamma-Tech, pp. 22–27.
- Guggenberger R (1989). Das Rocatec system—Haftung durch tribochemische Beschichtung. Dtsch Zahnärztl Z 44:874–876.[Medline]
[Order article via Infotrieve]
- ISO (1996). International Organization for Standardization: dentistry-polymer-based crown and bridge materials, Amendment 1996. ISO 10477.
- Kern M, Thompson VP (1994). Sandblasting and silica coating of a glass-infiltrated alumina ceramic: volume loss, morphology, and changes in the surface composition. J Prosthet Dent 71:453–461.[CrossRef][Medline]
[Order article via Infotrieve]
- Matinlinna J, Lassila LVJ, Özcan M, Yli-Urpo A, Vallittu PK (2004a). An introduction to silanes and their clinical applications in dentistry. Int J Prosthodont 17:155–164.[Medline]
[Order article via Infotrieve]
- Matinlinna J, Özcan M, Lassila LVJ, Vallittu PK (2004b). The effect of a 3-methacryloxypropyltrimethoxysilane and vinyltriisopropoxysilane blend and tris(3-trimethoxysilylpropyl)isocyanurate on the shear bond strength of composite resin to titanium metal. Dent Mater 20:804–813.[Medline]
[Order article via Infotrieve]
- Mohsen NM, Craig RG (1995). Effect of silanation of fillers on their dispersability by monomer systems. J Oral Rehabil 22:183–189.[Medline]
[Order article via Infotrieve]
- Murray JL, Wriedt HA (1990). O-Ti (oxygen-titanium). In: Binary alloy phase diagrams. Kacprzak L, editor. Materials Park, OH: ASM International, pp. 2924–2927.
- Özcan M (2002). The use of chairside silica coating for different dental applications: a clinical report. J Prosthet Dent 87:469–472.[Medline]
[Order article via Infotrieve]
- Paffenbarger GC, Sweeney WT, Bowen RL (1967). Bonding porcelain to acrylic resin denture bases. J Am Dent Assoc 74:1018–1023.[Medline]
[Order article via Infotrieve]
- Plueddemann EP (1970). Adhesion through silane coupling agents. J Adhes 2(7):184–201.
- Plueddemann EP (1991). Silane coupling agents. 2nd ed. New York: Plenum Press.
- Tezvergil A, Lassila LVJ, Vallittu PK (2003). The effect of fiber orientation on the thermal expansion coefficients of the fiber reinforced composites. Dent Mater 19:471–477.[CrossRef][Medline]
[Order article via Infotrieve]
- Vallittu PK (1997a). Curing of a silane coupling agent and its effect on the transverse strength of autopolymerizing polymethylmethacrylate-glass fibre composite. J Oral Rehabil 24:124–130.[Medline]
[Order article via Infotrieve]
- Vallittu PK (1997b). Bonding of hybrid composite resin to the surface of gold-alloy used in porcelain-fused-to-metal restorations. J Oral Rehabil 24:560–567.[Medline]
[Order article via Infotrieve]
- Vallittu PK, Ekstrand K (1999). In vitro cytotoxicity of fibre-polymethyl methacrylate composite used in dentures. J Oral Rehabil 26:666–671.[Medline]
[Order article via Infotrieve]
- van Noort R (2002). Principles of adhesion. In: An introduction to dental materials. van Noort R, editor. Hong Kong: Elsevier, pp. 68–78.
- Venhoven BA, de Gee AJ, Werner A, Davidson CL (1994). Silane treatment of filler and composite blending in a one-step procedure for dental restoratives. Biomaterials 15:1152–1156.
- Yoshida K, Taira Y, Atsuta M (2001). Properties of opaque resin composite containing coated and silanized titanium dioxide. J Dent Res 80:864–868.
Journal of Dental Research, Vol. 84, No. 4,
360-364 (2005)
DOI: 10.1177/154405910508400413
 CiteULike  Complore  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter What's this?
|
|
TOP
ABSTRACT
INTRODUCTION