Influence of PEEK Surface Modification on Surface Properties and Bond Strength to Veneering Resin Composites Christine Keula / Anja Liebermanna / Patrick R. Schmidlinb / Malgorzata Roosc / Beatrice Senerd / Bogna Stawarczyke Purpose: To test the impact of mechanical and chemical treatments of PEEK on surface roughness (SR), surface free energy (SFE), and tensile bond strength (TBS) to veneering resin composites. Materials and Methods: PEEK specimens (N = 680) were fabricated and divided into treatment groups (n = 170/group): 1. air abrasion (AIA); 2. etching with piranha solution (PIS); 3. air abrasion + piranha acid etching (AIP); and 4. no treatment (NO). Ten specimens of each treatment group were assessed with a contact angle measuring device and profilometer to determine SFE and SR, respectively. The remaining 160 specimens of each group were divided into subgroups according to coupling method (n = 32/subgroup): 1. Monobond Plus/ Heliobond (MH); 2. Visio.link (VL); 3. Clearfil Ceramic Primer (CCP); 4. Signum PEEK Bond (SPB); and 5. control, no coupling (CG). Specimens were veneered using Signum Composite/SiCo or Signum Ceramis/SiCe (both: n = 16), incubated in water (60 days at 37°C) and thermocycled (5000 cycles of 5°C/55°C). TBS was measured and data analyzed by three- and one-way ANOVA, Kruskal-Wallis and Mann-Whitney tests (p < 0.05). Results: A significant effect of surface treatment (p < 0.001) and coupling agent application (p < 0.001) on TBS was observed. AIA specimens with/without PIS showed the highest SFE, SR, and TBS. No differences were measured between PIS and NO, and between AIA and AIP. When no coupling agent was used, no adhesion was obtained. CCP resulted in low adhesion values, whereas MH, SPB, and VL exhibited increased TBS. No significant impact of the veneering resin composite on TBS was found (p = 0.424). Conclusion: AIA and AIP combined with VL, SPB, and MH can be recommended for clinical use. Keywords: tensile bond strength, PEEK, veneering resin composites, coupling agents, surface treatment, acid etching, piranha solution. J Adhes Dent 2014; 16: 383–392. doi: 10.3290/j.jad.a32570
a
Assistant Professor, Department of Prosthodontics, Munich Dental School, Ludwig Maximilians University Munich, Germany. Performed experiments, wrote manuscript.
b
Professor and Head, Division of Periodontology, Clinic of Preventive Dentistry, Periodontology and Cariology, Center of Dental Medicine, University of Zurich, Switzerland. Contributed substantially to idea, discussion, proofreading of manuscript.
c
Senior Statistician, Division of Biostatistics, Institute of Social and Preventive Medicine, University of Zurich, Switzerland. Performed statistical analyses, proofread the manuscript.
d
Senior Laboratory Technican, Clinic of Preventive Dentistry, Periodontology and Cariology, Center of Dental Medicine, University of Zurich, Switzerland. Created SEM images and contributed to their analysis.
e
Senior Materials Scientist, Department of Prosthodontics, Munich Dental School, Ludwig Maximilians University Munich, Germany. Idea, experimental design, hypothesis, co-wrote manuscript. Performed statistical analyses.
Correspondence: Dr. Dipl. Ing. Bogna Stawarczyk, Department of Prosthodontics, Dental School, Ludwig Maximilians University Munich, Goethestrasse 70, 80336 Munich, Germany, Tel. +49-89-5160-9573. e-mail:
[email protected]
Vol 16, No 4, 2014
Submitted for publication: 02.08.13; accepted for publication: 11.04.14
H
igh-performance polyether ether ketone (PEEK) is a linear, aromatic, semicrystalline thermoplastic polymer, which is assigned to the main group of polyaryletherketones (PAEK). Its notable mechanical properties11 and its biocompatibility and stability with nearly all organic and inorganic chemicals11,28 make PEEK attractive for dentistry.23 The fields of application includes implant materials, temporary abutments, and implant-supported bar or clamp materials.3,18,25,26 Even for fixed dental prostheses (FDPs), PEEK might be a suitable dental polymer, and a recent study showed a remarkable fracture load of 1383 N, but with plastic deformation starting at approximately 1200 N.23 The clinical use of PEEK as full-coverage monolithic restorations is, however, limited by its low translucency and a grayish or snow-white color component.23,24 Therefore, additional resin composites for veneering are still necessary. In this context, achieving an adequate bond strength between PEEK surfaces and veneering resin composites still remains a problem due to the low surface energy of 383
Keul et al
PEEK as well as its resistance to surface modification by different mechanical and chemical treatments.19 In general, numerous material characteristics such as adhesive bonding properties, wettability, reflectivity, and coefficient of friction are significantly affected by surface treatment and processing.14,19 One of the most popular procedures for the treatment of dental restorations prior to luting is the use of airborne-particle abrasion, which simultaneously cleans the surface and increases its area.6,13,14,22 Several studies recently investigated bonding properties between PEEK and resin composites, finding that without further treatment of the PEEK surface, no or only insufficient adhesion can be achieved.9,19,21,23,24 One study tested the adhesion between PEEK surfaces and veneering composites after air abrasion and etching without the additional use of adhesives, which led to the recommendation that acid etching should be applied when PEEK is used as a substructure and veneered with composite material.23 The use of acid etching (98% sulfuric acid) resulted in a bond strength of 19 ± 3.4 MPa for RelyX Unicem (3M ESPE) and 18.2 ± 5.4 MPa for Heliobond (Ivoclar Vivadent) in combination with the fine hybrid resin composite Tetric (Ivoclar Vivadent).19 Treatment with piranha solution, consisting of sulfuric acid (H2SO4) and hydrogen peroxide (H2O2), resulted in bond values of 11.8 MPa for RelyX Unicem combined with bis-GMA containing coupling agents, whereas additional airborne-particle abrasion increased the bond strength up to 21.4 MPa.7 Further investigations showed that the application of methacrylate-containing coupling agents on air-abraded PEEK surfaces also represents an auspicious bonding method.21,24 Adhesive systems which contain methyl methacrylate (MMA) groups mediated a higher bond strength between PEEK and all tested veneering resin composites.21,24 Therefore, the present study investigated the influence of mechanical (AIA/air abrasion) and/or chemical treatments (PIS/piranha-solution acid etching) on PEEK surface free energy (SFE) and surface roughness (SR), and the subsequent application of coupling agents on their tensile bond strength (TBS) to two veneering resin composites. The first hypothesis was that mechanical and/or chemical treatment increases the SFE and SR. The second hypothesis tested was that mechanical and/or chemical treatment in combination with coupling agents results in higher TBS when applying veneering resin composites.
(SiC) paper (ScanDia Hans P. Tempelmann) with an automatic polishing device at 25 N pressure (Tegranim-20, Struers; Ballerup, Denmark). The specimens for surface properties and for TBS measurements were divided into four surface treatment groups (n = 10 and n = 160/group, respectively): 1. air abrasion (AIA); 2. acid etching with piranha solution (PIS); 3. air abrasion + piranha-solution acid etching (AIP); and 4. no treatment (NO). AIA was conducted with alumina powder (45-degree angulation, mean powder size 50 μm, Basic Quattro IS) at 0.2 MPa for 10 s. For PIS, a fresh solution was used for half of the polished specimens and half of the AIA specimens. They were cleaned before the application of the piranha solution in an ultrasonic bath with distilled water for 5 min, the piranha solution was applied for 30 s and then rinsed off with distilled water for 10 s. Before the application of coupling agents, all specimens were ultrasonically cleaned again in an ultrasonic bath in distilled water for 5 min. Analysis of Surface Properties For the determination of surface properties, 40 specimens were used, 10 from each treatment group. For the analysis of SFE and SR, the same specimens were used, with SFE being determined first. SFE was analyzed at room temperature (23°C) with a contact angle measurement device (EasyDrop, Krüss; Hamburg, Germany) using the sessile-drop technique. Two different liquids of diverse polarity were used separately: distilled water and diiodomethane 99% (Cat: 15.842-9, Sigma-Aldrich; Steinheim, Germany; batch # S65447-448). For each specimen, three measurements with distilled water and three with diiodomethane were performed. A standardized drop volume of every liquid was applied (distilled water: 10 μl, diiodomethane: 5 μl) and photographed with a digital camera after exactly 5 s. The static contact angle on the basis of the diameter and the height of the drops was measured by special software (Easy Drop DSA4; Krüss). Two different computation methods were used, depending on the angle of the fluid used. For the distilled water, the contact angle was determined with the tangent-1 method, while the circle method was chosen for flat angles. According to Young’s equation, the SFE was calculated on the basis of the contact angles of water and diiodomethane: cos θ =
σS – γLS , σL
MATERIALS AND METHODS Specimen Preparation and Treatment Forty specimens were manually sectioned out of a PEEK blank (Dentokeep, nt-trading; Karlsruhe, Germany; batch #11DK14001) using a diamond cutting disk (918PB.104.220, Komet Dental, Gebr. Brasseler; Lemgo, Germany), with dimensions of 50 mm x 50 mm x 2 mm (n = 40) for analyzing the surface properties and 640 specimens 7 mm x 7 mm x 2 mm (n = 640) for analyzing the TBS. They were embedded in a chemically curing acrylic resin (ScandiQuick, Scan-Dia Hans P. Tempelmann; Hagen, Germany) and polished under water spray for 10 s from P500 up to P2400 silicium carbide 384
where σL represents the SFE of the liquid, σS the SFE of the solid, and γLS the interface SFE.15 SR was investigated with a profilometer using a 90-degree sensing device (MarSurf M400+SD26, Mahr; Göttingen, Germany) and a pressure of 0.7 mN. The diameter of the diamond probe tip was 2 μm. For each specimen, six measurements (3x vertically and 3x horizontally) were performed with a measuring track of 6 mm. The distance between the tracks was 0.25 mm. The mean roughness value SR (μm) of each specimen was calculated. Surface structure topographies of NO, AIA, PIS, and AIP specimens were examined at different magnifications with a scanning electron microscope (SEM) (Carl Zeiss Supra The Journal of Adhesive Dentistry
Keul et al
Table 1 Coupling agents, treatments, and group abbreviations Treatment Coupling agent
Air abrasion (AIA)
Piranha acid solution (PIS)
Air abrasion + piranha acid solution (AIP)
No treatment (NO)
Monobond Plus/Heliobond (MH)
AIA-MH
PIS-MH
AIP-MH
NO-MH
Visio.link (VL)
AIA-VL
PIS-VL
AIP-VL
NO-VL
Clearfil Ceramic Primer (CCP)
AIA-CCP
PIS-CCP
AIP-CCP
NO-CCP
Signum PEEK Bond I+II (SPB)
AIA-SPB
PIS-SPB
AIP-SPB
NO-SPB
No treatment (CG)
AIA-CG
PIS-CG
AIP-CG
NO-CG
Table 2 Chemical composition of coupling agents and veneering resins evaluated Material
Product name (batch number)
Manufacturer
Composition
Treatment
Air abrasion Basic Quattro IS
Renfert; Hilzingen, Germany
50 μm Al2O3
Piranha solution H2SO4 98% (1.12080.1000) H2O2 (K 43190280 206)
Hospital pharmacy, Ludwig Maximilians University, Munich, Germany
H2SO4 (98%):H2O2 (30%) = 1:1
Monobond Plus (R26662)
Ivoclar Vivadent; Schaan, Lichtenstein
Silane methacrylate, phosphoric acid methacrylate, sulfide methacrylate
Coupling agent
Heliobond (R22281)
Veneering resins
Bis-GMA, TEG-DMA
Visio.link (114784)
Bredent; Senden, Germany
MMA, PETIA, dimethacrylates, photoinitiators
Clearfil Ceramic Primer (00022B)
Kuraray Noritake Dental; Tokyo, Japan
3-methacryloxypropyl trimethoxy silane, MDP, ethanol
Signum PEEK Bond I + II (experimental adhesive) (010121/010110)
Heraeus Kulzer; Hanau, Germany
Bond I: bifunctional molecules based on phosphoric acid esters and thiol compounds Bond II: MMA, PMMA, photoinitiators
Signum Composite Dentin A3 (10506)
Heraeus Kulzer
Multifunctional methacrylic acid esters Fillers: 74% w/w, silicon dioxide rheologically effective type, silanized splitter pre-polymerisate; other components: photoinitiators, stabilizing agents, inorganic pigments
Signum Ceramis Dentin A3 (VP090712)
Heraeus Kulzer
Multifunctional methacrylic acid esters, siliciumdioxide, anorganic silanized fillers, photoinitiators, stabilizers, inorganic pigments
MMA: methyl methacrylate; PETIA: pentaerythritol triacrylate; bis-GMA: bisphenol A dimethacrylate; PMMA: poly(methyl methacrylate); TEG-DMA: triethylene glycol dimethacrylate; 10-MDP: 10-methacryloyloxydecyl dihydrogenphosphate; Al2O3: aluminum oxide; H2SO4: sulfuric acid; H2O2: hydrogen peroxide.
50 VP FESEM, Zeiss; Oberkochen, Germany) after sputter coating with gold (Balzers SCD 030, Balzers Union; Balzers, Liechtenstein). Sputter coating took place for 30 s in an argon gas atmosphere at a pressure of 0.05 mbar and a target distance of 50 mm. The acceleration voltage for the SEM analysis was 10 kV with a working distance of 8.5 mm. Analysis of Tensile Bond Strength (TBS) and Failure Types The treated specimens for TBS analyses (N = 640; n = 160/treatment) were further subdivided into five groups with respect to the coupling method, which was performed according to the manufacturer’s recommendations (n = 32/group): 1. MH: application of Monobond Vol 16, No 4, 2014
Plus and vaporization for 60 s, application of Heliobond and light polymerization (Elipar S10; 3M ESPE; Seefeld, Germany; light intensity: 1200 mW/cm2) for 10 s; 2. VL: application and light polymerization at 220 mW/cm2 (Bre. Lux Power Unit, Bredent; Senden, Germany) for 90 s; 3. CCP: application and vaporization; 4. SPB: application of liquid I and vaporization for 10 s, application of liquid II and light polymerization at 225 mW/cm2 (HiLite Power, Heraeus Kulzer; Hanau, Germany) for 90 s; and 5. control group (NO), without coupling. All materials’ abbreviations and combinations are shown in Table 1. The manufacturers, batch numbers, and chemical composition of the used materials are given in Table 2. 385
Keul et al
croscope at 20X magnification (microscope: Stemi 2000-C, light source: CL 6000 LED, Zeiss). The failures were classified by two independent blinded examiners as 1. cohesive failure in PEEK, 2. cohesive failure in veneering resin composite, 3. adhesive failure, and 4. mixed failure. SEM images were made (Carl Zeiss Supra 50 VP FE-SEM) of representative failure types.
F Holding Device
Acrylic Cylinder Veneering Resin
PEEK
Embedding Resin
Fig 1 Schematic overview of tensile bond strength test configuration.
Each coupling group was then divided again into two subgroups (n = 16) with respect to the veneering resin composite materials, which were used according manufacturers’ recommendations: 1. Signum Composite Dentin (SiCo) shade A3, and 2. Signum Ceramis Dentin (SiCe) shade A3. To apply the materials, an acrylic cylinder with an inner diameter of 2.9 mm and a height of 10 mm was manually filled with an increment of 6 mm with one of the resin composite materials and positioned onto the specimen’s surface, fitting seamlessly. Excess material was carefully removed from the bonding margin and the resin composite was polymerized according to the manufacturer’s instructions for 180 s (HiLite Power, Heraeus Kulzer). All bonded specimens were then stored in distilled water at 37°C for 60 days and additionally subjected to 5000 thermal cycles between 5°C and 55°C (dwell time: 20 s) (Thermocycler THE 1100, SD Mechatronik; Feldkirchen-Westerham, Germany). Following artificial aging, all specimens were stored for 24 h at room temperature (23°C) in distilled water before TBS was determined in a universal testing machine (Zwick 1445; Ulm, Germany) at a crosshead speed of 5 mm/min. The acrylic cylinder was held by a collet, allowing the whole system to self-align. Specimens were positioned in the jig with the specimen’s surface perpendicular to the loading direction. The test configuration provided a moment-free axial force application (Fig 1). The jig was attached to the load cell and pulled apart by an upper chain.24 The TBS was calculated according to the following equation: TBS (MPa) = fracture load (N) / bonding area (mm2). For the analysis of fracture type, the fractured surface of each specimen was examined with a reflected-light mi386
Statistical Analyses Using nQuery Advisor (Version 6.04.10, Statistical Solutions; Saugus, MA, USA), a power analysis was calculated on the basis of a pilot study with seven PEEK specimens conditioned with MP and veneered using SiCo (14.47 ± 2.61 MPa). A sample size of 16 in each group was shown to have 95% power to detect a difference of 27% in means (4.5 MPa) caused by aging, assuming that the common standard deviation is 2.6 MPa using a two-group t-test with a Bonferroni corrected twosided significance level of 0.005. The normality of data distribution was tested using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Descriptive statistics (mean, standard deviation, and 95% confidence intervals [CI]) were calculated. For the determination of significant differences between the tested groups, threeand one-way ANOVA with the post-hoc Scheffé test and non-parametric Kruskal-Wallis and Mann-Whitney tests were used. Relative frequencies of failure types, together with the corresponding 95% CI, were computed according to the Wilson procedure.29 The Pearson correlation coefficient evaluated the effect of the association between mean SFE and mean SR for different treatments. P-values less than 5% were considered to be statistically significant in all tests. The data were analyzed using IBM SPSS (Version 20, IBM; New York, NY, USA).
RESULTS Surface Properties One of the four groups (PEEK surface after air abrasion) for the determination of SFE was not normally distributed, whereas all tested SR groups showed normal distribution. Therefore, SFE results were analyzed nonparametrically using the Kruskal-Wallis test, and SR results were parametrically evaluated with one-way ANOVA. Descriptive statistics (mean, SD, 95% CI) for SFE and SR are presented in Table 3. NO-PEEK surfaces presented significantly lower SFE values than AIA or AIP surfaces (p < 0.001). The values of PIS-PEEK surfaces were not significantly different from NO. Both AIA and AIP-PEEK surfaces resulted in significantly higher SR than NO or only PIS specimens (p < 0.001). SR showed a significant positive correlation with SFE (r2 = 0.578, p < 0.001). SEM images after the different PEEK surface treatments are depicted in Fig 2. NO and PIS-PEEK specimens showed a plain surface, whereas AIA resulted in an irregular structured surface with sharp streaks and flaws. The combination of AIA plus PIS displayed irregularities similar to those after AIA, but with rounded edges. The Journal of Adhesive Dentistry
Keul et al
Table 3 Mean values, standard deviations, and 95% confidence intervals (95% CI) of surface roughness and medians and min/max values of surface free energy of different treated PEEK surfaces Treatment
Surface roughness (SR) Mean (SD)
Surface free energy (SFE)
95% CI
Median
Min/max
AIA
0.875
(0.029)b
(0.79; 0.96)
51.2c*
49.7/56.6
PIS
0.041 (0.001)a
(0.03; 0.05)
47.8ab
45.2/55.5
AIP
0.818 (0.020)b
(0.76; 0.87)
50.7bc
49.6/54.1
NO
0.043 (0.002)a
(0.03; 0.06)
48.0a
43.4/51.0
*Not normally distributed. Different superscript letters within the same column indicate significant differences. For abbreviations, see Table 1.
a
b
c
Fig 2 SEM images of PEEK surface after different treatment methods (magnification left: 10,000X, magnification right: 50,000X). a: air abrasion; b: piranha solution; c: air abrasion and piranha solution; d: without treatment.
Vol 16, No 4, 2014
d 387
Keul et al
Table 4 Mean, SD, and 95% confidence intervals of TBS (MPa) of different veneering resin composites on treated PEEK specimens SiCo Coupling agent
SiCe
Mean (SD)
95% CI
Mean (SD)
95% CI
MH
18.4 (4.5)bB
(15.9; 20.9)
20.7 (7.6)cA
(16.5; 24.9)
VL
16.5 (8.0)bA
(12.1; 20.8)
12.8 (12.1)bcA
(6.2; 19.4)
CCP
3.6 (3.9)*aA
(1.4; 5.8)
1.3 (2.3)*aA
(0.7; 2.7)
SPB
14.7 (4.6)bB
(12.1; 17.3)
11.2 (10.9)*bAB
(5.3; 17.2)
CG
0
-
0
-
MH
3.7 (6.5)*aA
(0.1; 7.3)
0
-
VL
23.4 (9.9)bA
(18.0; 28.8)
15.2 (10.6)aA
(9.4; 21.0)
0
-
0
-
AIA
PIS
CCP SPB
5.8
(6.3)*aA
CG
(2.2; 9.3)
11.9
(7.3)aAB
(7.9; 15.9)
0
-
0
-
17.1 (4.5)bcB
(14.5; 19.6)
22.6 (8.3)bA
(18.0; 27.3)
AIP MH VL CCP
19.9 1.8
(8.0)cA
(2.9)*aA
19.7
(10.6)*bA
(13.9; 25.4)
(0.1; 3.5)
1.4
(3.1)*aA
(0.0; 3.2)
(15.5; 24.3)
SPB
11.1 (7.9)bAB
(6.8; 15.4)
18.2 (7.9)bB
(13.8; 22.5)
CG
0
-
0
-
6.5 (10.0)*aA
(1.0; 11.9)
0
-
(8.3; 19.8)
(11.5)aA
NO MH VL
14.1
(10.7)aA
13.9
(7.6; 20.2)
CCP
0
-
0
-
SPB
8.1 (7.1)*aAB
(4.2; 12.1)
7.6 (4.7)aA
(5.0; 10.3)
CG
0
-
0
-
* Not normally distributed. Different superscript lowercase letters within the same column indicate significant differences between coupling agents. Different superscript capital letters within the same column indicate significant differences between surface treatments. Difference between surface SiCo and SiCe not significant.
TBS and Failure Types Results of TBS values for all tested groups are shown in Table 4 and Fig 3. The Kolmogorov-Smirnov and ShapiroWilk tests indicated no violation of the assumption of normality for 75% of the tested groups. Therefore, the data were parametrically analyzed. According to the three-way ANOVA, which was applied for the examination of interaction only, surface treatment (p < 0.001) and coupling (p < 0.001) showed an impact on the TBS results. In contrast, no impact of the veneering resin composite on TBS values was found (p = 0.424). In principle, both AIA and AIP-PEEK surfaces showed significantly higher TBS values compared with NO or PIS 388
specimens (p < 0.001). No differences between NO and PIS, and no significant difference between AIA and AIP in terms of TBS were observed. No application of coupling/ bonding agents resulted in no bond, while the use of CCP resulted in no or low adhesion to PEEK substrates. The use of MH, SPB, and VL significantly increased TBS (p < 0.001). No difference between groups conditioned with MH and SPB was observed (p > 0.05). Significantly different failure types between the tested groups were observed (p < 0.001; Fig 4). All groups showed predominantly adhesive failures. However, cohesive failures in veneering resin composite were rarely observed for groups conditioned with SPB (13%), followed by VL (11%) The Journal of Adhesive Dentistry
Keul et al
35 30
Fig 3a TBS (MPa) of Signum Composite veneering resin composite on treated PEEK specimens. MH: Monobond Plus/Heliobond; VL: Visio.link; CCP: Clearfil Ceramic Primer; SP: Signum PEEK; CG: control group. AIA: air abrasion; PIS: Piranha solution; AIP: air abrasion + Piranha solution; NO: without.
TBS (MPa)
25
AIA
20
PIS AIP
15
No 10 5 0
MH
VL
CCP
SPB
CG
35 30
TBS (MPa)
25
AIA
20
PIS AIP
15
No 10
Fig 3b TBS (MPA) of Signum Ceramis veneering resin composite on treated PEEK speciments. For abbreviations, see Fig 3a.
a
5 0
MH
300μm* 1 mm*
Fig 4
VL
CCP
b Mag = 48 X EHT = 10.00 kV WD = 46 mm Signal A = SE2 Photo No. = 2742 Date: 21 Jun 2013
SPB
CG
300μm* 1 mm*
Mag = 50 X EHT = 10.00 kV WD = 48 mm Signal A = SE2 Photo No. = 2745 Date: 21 Jun 2013
SEM images of failure types after debonding. a: cohesive fracture in veneering resin; b: adhesive failure (magnification: 50X).
Vol 16, No 4, 2014
389
Keul et al
Table 5 Relative frequencies (%) with 95% confidence intervals (CI) of failure types for each tested group.(CP: cohesive in PEEK; CV: cohesive in veneering resin composite; A: adhesive between substrate and resin; M: mixed, combination between A, CV or CP)
Coupling agent CP
SiCo
SiCe
Relative frequencies (95% CI)
Relative frequencies (95% CI)
CV
A
M
CP
CV
A
M
AIA MH
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
13 (1;39)
87 (61;99)
0 (0;21)
VL
0 (0;21)
6 (0;31)
94 (69;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
CCP
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
SPB
0 (0;21)
31 (11;59)
69 (41;89)
0 (0;21)
0 (0;21)
19 (4;46)
81 (54;96)
0 (0;21)
CG
0 (0;21)
0 (0;21.8)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
PIS MH VL
0 (0;21)
31 (11;59)
69 (41;89)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
CCP
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
SPB
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
CG
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
MH
0 (0;21)
19 (4;46)
81 (54;96)
0 (0;21)
0 (0;21)
19 (4;46)
81 (54;96)
0 (0;21)
VL
0 (0;21)
13 (1;39)
87 (61;99)
0 (0;21)
0 (0;21)
13 (1;39)
87 (61;99)
0 (0;21)
AIP
CCP
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
SPB
0 (0;21)
25 (7;53)
100 (79;100)
0 (0;21)
0 (0;21)
31 (11;59)
69 (41;89)
0 (0;21)
CG
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
MH
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
VL
0 (0;21)
13 (1;39)
87 (61;99)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
CCP
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
NO
SPB
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
CG
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
0 (0;21)
0 (0;21)
100 (79;100)
0 (0;21)
For abbreviations, see Table 1.
and MH (6%). No mixed failures were observed. Detailed information about relative frequencies of failure types is given in Table 5.
DISCUSSION The present study investigated the influence of mechanical and chemical surface treatments and coupling/bonding methods on the retentive potential and their ability to promote adhesion between PEEK and veneering resin composites. In terms of the surface treatment, AIA increased the SFE and SR, whereas PIS showed no impact on the 390
surface properties examined. Therefore, the first tested hypothesis was only partially accepted. In contrast to this study, previous studies also observed an increase of the SR after etching.7,23 However, those authors etched the PEEK surface with a different acid – namely, sulfuric acid, and not piranha solution. Another study reported that a high SR influences the contact angle of materials.4 Therefore, it is conceivable that the SFE of a material is correlated to the SR. This study corroborated a positive albeit weak correlation between SR and SFE (r2 = 0.578). The second hypothesis – that mechanical and/or chemical treatment in combination with coupling agents result in higher TBS to veneering resins – was also acThe Journal of Adhesive Dentistry
Keul et al
cepted only partially. The present study included AIA, PIS, AIP, and NO polished PEEK specimens, allowing analysis of potential chemical interactions. In accordance with prior studies that investigated the bond to differently treated PEEK surfaces, the present study observed higher TBS for specimens after air abrasion.7,19 The implemented method of treatment for the specimens represents an important factor in creating a durable bond to PEEK. While air abrasion alone improves the microroughness of the PEEK specimen, the treatment with piranha solution increases the number of functional groups on the surface. Atomic oxygen, released in the reaction of sulfuric acid and hydrogen peroxide, is able to react with the benzene ring of PEEK.7 Piranha solution consists of sulfuric acid (H2SO4) and hydrogen peroxide (H2O2), and is chemically termed peroxymonosulfuric acid (H2SO5). It has strong oxidizing properties and is therefore able to remove most organic remnants. This increases the surface polarity, and due to the breaking of the aromatic ring, the number of functional groups potent for bonding also increases.7 Nevertheless, the present study showed no significantly positive influence of treatment by piranha solution on the long-term bond of veneering resin composites to PEEK. This is in accordance with a prior study that also showed low bond strength for etched-only PEEK surfaces.7 Based on the present results, it could be assumed that only mechanical air abrasion creates sufficient SR for bonding. Due to the mechanically induced surface irregularities, the mechanical anchorage and the penetration of the coupling liquid along the flaws inside the polymer are improved. The chemical composition of the coupling agents plays an important role in creating a bond between participating materials. A previous study investigated the bond strength of veneering resin composites to air-abraded PEEK after the application of coupling agents and thermocycling (vs no thermocycling) and found that Visio.link and Signum Peek Bond I+II resulted in the highest TBS.24 In the current study, the best bonding potential was observed in groups conditioned with VL, which contains pentaerythritol triacrylate (PETIA) in solution, methyl methacrylate (MMA) monomers, as well as additional dimethacrylates. It can be speculated that PETIA dissolves the PEEK surface, whereas MMA monomers swelled the dissolved surface, and the dimethacrylate monomers provided the connection to the veneering resin composites with two methyl groups as binding sites. In contrast, the use of CCP (which contains 10-MDP) as the coupling agent resulted in no or very low bond strength. This can be explained by the fact that one functional group of the bifunctional MDP monomer is occupied by a phosphate group, which cannot further chemically react with the PEEK substrate or the veneering resin composite. The present study investigated the TBS after 60 days of water storage at 37°C plus 5000 thermal cycles (5°C/55°C, dwell time 20 s) according to ISO 10477 with an additional immersion for 24 h at room temperature (23°C) in distilled water. The latter procedure allowed for adequate relaxation and storage before mechanical testing. However, groups without any further artificial aging Vol 16, No 4, 2014
(initial) were not included in the present study. This factor may conceivably be a limitation of the present study and should be addressed in a future study. Depending on the materials examined, long-term water storage as a method of artificial aging leads to degradation within the interface:20 the covalent bonds are chemically split due to the incorporation of water molecules.10 Thermocycling in the lab is a simulation of individual thermal intraoral variations caused by the daily occurrence of breathing,5 eating, and drinking.12,16 Thermocycling may influence the bond strength in two ways. On the one hand, a higher bond strength could be obtained by the postpolymerization of the coupling agent and the veneering resin composite.17 On the other hand, thermal changes may result in mechanical stress caused by volumetric changes.27 Therefore, cracks in the bonding area may initiate, followed by a loss of bond strength. To date, no clinical studies exist which explicitly validate the results of laboratory aging. In the present study, a macrotensile bond strength test was used, although microshear or microtensile tests are also valuable methods in similar contexts.1 These alternatives would probably result in higher bond strength values as a result of the smaller surface area. However, it is noteworthy that they are rather technically demanding and sensitive compared to the more commonly used macro-test methods. Therefore, a macrobond strength test method was selected due to the advantages of quick, direct results and ease of handling, which ensures reliable and reproducible results.1 Although laboratory study designs are unable to reproduce intraoral conditions in detail with all individual factors, they provide some information on the reliability of bonding of materials used in dental restorations.2 In general, bond strength tests under laboratory conditions assess the quality of adhesion.8 Once a veneering and coupling-agent system for dental frameworks passes the laboratory studies, a clinical study with controlled and standardized parameters should evaluate the long-term clinical performance.
CONCLUSION Within the limitations of this laboratory study, the appropriate treatment of PEEK surfaces and the additional use of coupling agents should be recommended for veneering with resin composites. Air abrasion and air abrasion + piranha solution combined with Visio.link, Signum PEEK Bond, and Monobond Plus/Heliobond seemed to generate reliable bond strengths for the veneering of PEEK with resin composites. However, more research will be needed to investigate the long-term durability of veneered PEEK dental restorations.
ACKNOWLEDGMENTS The authors would like to express their gratitude to nt-trading, Heraeus Kulzer, Ivoclar Vivadent, Kuraray, and Bredent for supporting this study with materials.
391
Keul et al
REFERENCES 1.
2.
3.
4.
5. 6.
7.
8.
9. 10. 11. 12. 13.
14.
15. 16. 17.
Andrade AM, Moura SK, Reis A, Loguercio AD, Garcia EJ and Grande RHM. Evaluating resin-enamel bonds by microshear and microtensile bond strength tests: effect of composite resin. J Appl Oral Sci 2010;18:591-598. Bähr N, Keul C, Edelhoff D, Eichberger M, Roos M, Gernet W, Stawarczyk B. Effect of different adhesives combined with two resin composite cements on shear bond strength to polymeric CAD/CAM materials. Dent Mater J 2013;32:492-501. Bayer S, Komor N, Kramer A, Albrecht D, Mericske-Stern R, Enkling, N. Retention force of plastic clips on implant bars: a randomized controlled trial. Clin Oral Implants Res 2012;23:1377-1384. Benetti P, Della Bona A, Kelly JR. Evaluation of thermal compatibility between core and veneer dental ceramics using shear bond strength test and contact angle measurement. Dent Mater 2010;26:743-750. Boehm RF. Thermal environment of teeth during open-mouth respiration. J Dent Res 1972;51:75-78. Ersu, B, Yuzugullu B, Ruya Yazici A, Canay S. Surface roughness and bond strengths of glass-infiltrated alumina-ceramics prepared using various surface treatments. J Dent 2009;37:848-856. Hallmann L, Mehl A, Sereno N, Hämmerle CHF. The improvement of adhesive properties of PEEK through different pre-treatments. Appl Surf Sci 2012;258:7213-7218. Holderegger C, Sailer I, Schuhmacher C, Schläpfer R, Hämmerle CH, Fischer J. Shear bond strength of resin cements to human dentin. Dent Mater 2008;24:944-950. Kern M, Lehmann F. Influence of surface conditioning on bonding to polyetheretherketon (PEEK). Dent Mater 2012;28:1280-1283. Khoroushi M, Rafiei E. Effect of thermocycling and water storage on bond longevity of two self-etch adhesives. Gen Dent 2013;61:39-44. Kurtz SM, Devine JN. PEEK biomaterials in trauma, orthopedic, and spinal implants. Biomaterials 2007;28:4845-4869. Longman CM, Pearson GJ. Variation in tooth surface temperature in the oral cavity during fluid intake. Biomaterials 1987;8:411-414. Marshall SJ, Bayne SC, Baier R, Tomsia AP, Marshall GW. A review of adhesion science. Dent Mater 2010;26:e11-16. Erratum in Dent Mater 2010;26:e195. Ourahmoune R, Salvia M, Mathia TG, Mesrati N. Surface morphology and wettability of sandblasted PEEK and its composites. Scanning 2014;36:64-75. Packham DE. Surface energy, surface topography and adhesion. Int J Adhesion Adhes 2003;23:437-448. Palmer DS, Barco MT, Billy EJ. Temperature extremes produced orally by hot and cold liquids. J Prosthet Dent 1992;67:325-327. Piwowarczyk A, Lauer HC, Sorensen JA. In vitro shear bond strength of cementing agents to fixed prosthodontics restorative materials. J Prosthet Dent 2004;92:265-273.
392
18. Santing HJ, Meijer HJ, Raghoebar GM, Özcan M. Fracture strength and failure mode of maxillary implant-supported provisional single crowns: a comparison of composite resin crowns fabricated directly over PEEK abutments and solid titanium abutments. Clin Implant Dent Relat Res 2010;14:882-889. 19. Schmidlin PR, Stawarczyk B, Wieland M, Attin T, Hämmerle CH, Fischer J. Effect of different surface pre-treatment and luting materials on shear bond strength to PEEK. Dent Mater 2010;26:553-559. 20. Sezinando A, Perdigão J, Regalheiro R. Dentin bond strengths of four adhesion strategies after thermal fatigue and 6-month water storage. J Esthet Restor Dent 2012;24:345-355. 21. Stawarczyk B, Bähr N, Beuer F, Wimmer T, Eichberger M, Gernet W, Jahn D, Schmidlin PR. Influence of plasma pretreatment on shear bond strength of self-adhesive resin cements to polyetheretherketone. Clin Oral Investig 2014;18:163-170. 22. Stawarczyk B, Basler T, Ender A, Roos M, Özcan M, Hämmerle, C. Effect of surface conditioning with airborne-particle abrasion on the tensile strength of polymeric CAD/CAM crowns luted with self-adhesive and conventional resin cements. J Prosthet Dent 2012;107:94-101. 23. Stawarczyk B, Beuer F, Wimmer T, Jahn D, Sener B, Roos M, Schmidlin PR. Polyetheretherketone-A suitable material for fixed dental prostheses? J Biomed Mater Res B Appl Biomater 2013;101:1209-1216. 24. Stawarczyk B, Keul C, Beuer F, Roos M, Schmidlin PR. Tensile bond strength of veneering resin to PEEK: impact of different adhesives. Dent Mater J 2013;32:441-448. 25. Tannous F, Steiner M, Shahin R, Kern M. Retentive forces and fatigue resistance of thermoplastic resin clasps. Dent Mater 2012;28: 273-278. 26. Tetelman ED, Babbush CA. A new transitional abutment for immediate aesthetics and function. Implant Dent 2008;17:51-58. 27. Torstenson B, Brännström M. Contraction gap under composite resin restorations: effect of hygroscopic expansion and thermal stress. Oper Dent 1988;13:24-31. 28. Toth JM, Wang M, Estes BT, Scifert JL, Seim HB 3rd, Turner AS. Polyetheretherketone as a biomaterial for spinal applications. Biomaterials 2006;27:324-334. 29. Wilson EB. Probable inference, the law of succession, and statistical inference. J Am Stat Ass 1927;22:209-212.
Clinical relevance: In order to ensure long-term success, PEEK restorations should be air abraded or air abraded plus etched with piranha solution. In addition, coupling agents such as VL, SPB, and MH should be applied before veneering with resin composites.
The Journal of Adhesive Dentistry
