Effect of Argon Plasma and Sandblasting on Bond Strength of PEEK and PEKK to Resin Cement

Introduction

High-performance polymers are thermoplastic materials with various polyarylketone compositions [1,2]. Polyaryletherketone (PAEK) polymers have biocompatibility and superior mechanical properties [3,4]. The 2 best-known PAEK polymers are polyetheretherketone (PEEK) and polyether-ketone-ketone (PEKK) [5]. With the increasing emphasis on aesthetics, marginal integrity problems of metal-supported restorations, metallic taste, and problems such as metallic reflection from the restoration have prompted new studies in the field of materials science [6]. Given the growing interest in metal-free restoration materials and the development of computer-aided design and computer-aided manufacturing (CAD/CAM) technology, CAD/CAM-based high-performance polymers have emerged as alternatives to zirconia, titanium, and other metals [7–9].

PAEK polymers resemble human bones and are considered biocompatible. They also exhibit excellent heat and solvent resistance, near-perfect electrical insulation properties, and high wear and fatigue resistance, while satisfying the esthetic demands of patients who prefer metal-free prostheses [10]. Additionally, high-performance polymers cause fewer artifacts than metal-based restorations on diagnostic imaging using computed tomography (CT), magnetic resonance imaging (MRI), and X-ray because they are radiolucent [10], making them valuable alternatives to ceramic and metal-based restorations [11].

Despite these advantages, the clinical use of high-performance polymers alone is limited. These materials have low translucency, are opaque, and appear grayish or pearly-white in color. Their surfaces must be coated with a veneer to meet esthetic requirements [1]. Regardless of the production technique, the inert structure of PAEK polymers significantly limits their ability to bond with restorative materials to be used as superstructures [12]. The chemical inertness of high-performance polymers, combined with their low surface energy and resistance to surface treatments, makes bonding of veneering materials difficult, partly explaining why these polymers are not yet widely used in dental practice.

Several studies have sought to increase the surface energies of PEEK and PEKK through various treatments, which can be mechanical or chemical [9,10]. Mechanical treatments include the application of silica or aluminum oxide, laser and plasma methods, and burr etching. Chemical treatments include acid etching and the application of adhesive materials [11]. Other methods have also been tested prior to veneer application [13–15]. The simplest procedure is sandblasting, which enhances micromechanical bonding by increasing surface roughness [16]. Aluminum particles were added to the surface through sandblasting surface treatment to increase the bonding strength of PEEK compared with untreated PEEK by enlarging and cleaning the surface area [17,18], thus enhancing mechanical bonding strength [18]. Sandblasting is sometimes combined with acid surface treatment [10,19]. Although PEEK is generally chemically resistant, it can be modified using sulfuric acid or concentrated acids such as piranha solution. However, the clinical application of such acids is inappropriate [9]. In contrast, plasma treatment is environmentally friendly [20] and only changes the surface properties without affecting the bulk material [21,22]. Plasma surface treatment is widely used to modify the surfaces of various polymers. Plasma treatment provides chemical functional groups that increase surface energy and wettability and can improve bonding to coating resins [23,24].

Studies have shown that applying an agent containing functional monomer structures to PAEK polymers before adding the superstructure material increases the bond strength [10,25]. Although many studies have investigated the bond strength of PEEK material, studies on the currently used PEKK material are limited [25,26].

This in vitro study evaluated the bond strength of PEEK and PEKK polymers with resin cement after different surface treatments to compare the bond strength of 2 different polymer materials after surface treatments. Therefore, the objectives of this study were to examine the effect of different surface treatments on bond strength and to compare the bond strength of 2 different polymers after surface treatments. The first null hypothesis was that different surface treatments have no effect on bond strength, and the second null hypothesis was that there is no difference in bond strength between the 2 polymer materials after surface treatments.

Material and Methods

PEEK samples were obtained from BioHPP® (Bredent GmbH & Co. KG, Senden, Germany); PEKK samples were acquired from Pekkton® ivory (Cendres+Métaux, Switzerland). Square prism blocks (40 of each polymer), with dimensions of 7×7×2 mm, were prepared using a CAD/CAM (computer-aided design/computer-aided manufacturing) (Redon GTR®) device. The blocks were produced in a CAD/CAM device with a spindle speed of 60 000 RPM and cooled using a water tank and cooling liquids. Wet milling was used along the x-y-z and a-b axes in the CAD/CAM device.

Length and thickness were measured by a digital caliper (NTS Carbon Fiber Composite, China). All surfaces were pre-treated with 600–2000-grit sandpaper under a stream of water. The 40 samples of each polymer – 80 samples in total – were divided into 8 groups (10 samples per group). Ten samples from each of the PEEK and PEKK polymer samples were selected as the control group to examine both the surface analysis findings and changes in bond strength compared to the surface-treated groups. No surface treatment was applied to the control group samples. The samples were sanded and stored in molds to prevent contact with the bonding surfaces.

Non-thermal atmospheric argon plasma surface treatment was applied to the bonding surface on the samples using a plasma device (Biomedap, Tibbi Chez, Turkey). The device operated at 8 kW and 10 kHz, with a gas flow rate of 5 L/min. This procedure was applied at a 10-mm distance perpendicular to polymer specimens for 20 min. Sandblasting surface roughening was applied to the polymer materials using 125-μm aluminum oxide particles (Renfert 125; Strahlmittel, Germany) under 0.5 Mpa pressure from a 10-mm vertical distance for 10 s. Subsequently, some of these surfaces were also treated with argon plasma.

After the surface treatments were applied, PEEK and PEKK samples were embedded in PVC (polyvinyl chloride) pipe molds manufactured to fit the universal testing device holder with a diameter of 24 mm and a height of 24 mm, with the help of dental tweezers so that the upper surface of the autopolymerized acrylic resin was equal to the surface level of the samples. The upper surfaces of the samples were then coated with a polymeric acrylic resin (Visio.link®, Bredent GmbH & Co. KG, Senden, Germany). The adhesive agent was applied using a disposable brush, then exposed to light with a wavelength of 395–480 nm (Valo Ortho Cordless, Ultradent, USA) for 40 s. Next, a silicone mold – 2 mm thick and 3 mm in diameter – was used to apply resin cement (3M ESPE RelyX U200 Automix) to a thickness of 2 mm and diameter of 3 mm (Figure 1). All resin cement polymerization was completed within 40 s. The specimens were tested after resin cement application using a universal testing machine (Instron 3345 Universal Tester, USA), the samples were conditioned in air under normal conditions (temperature 298.15 K (25°C, 77°F) and absolute pressure 100.0 kPa (14.504 psi, 0.987 atm) for 1 h before bond strength testing. A knife-edge metal loading tip was placed parallel to the bond interface between the samples and the resin cement, and force was applied to the samples at a speed of 1 mm/min. Bond strength, measured in Newtons (N), was recorded at the point of resin cement detachment. To calculate bond strength in MPa, the N value was divided by the bonding area (mm2).

Individual samples taken from all groups, including the control groups, were examined using scanning electron microscopy (SEM) for the examination of the surface topographies of the samples, energy-dispersive spectroscopy (EDS) (FEI QUANTA FEG 250, USA) for the elemental composition, and atomic force microscopy (AFM) (Ez-AFM, Nanomagnetics, USA) for the surface properties, regardless of the bond strength test. SEM analysis was performed in low-vacuum mode at a chamber pressure of 100 pascals. Before analysis, samples were fixed to an aluminum block using adhesive tape, and SEM and EDS analyses were performed on the same device. EDS analysis was applied to the samples within 1 h after argon plasma application. For AFM analysis, samples were scanned in 3×3 μm tapping mode in the XY axis and 1 μm tapping mode in the Z axis at a scanning speed of 1 Nm/s.

The Kolmogorov-Smirnov test was used to verify the assumption that continuous variables were normally distributed. The Levene homogeneity test was used to assess differences among the mean values of independent groups, followed by parametric testing using one-way ANOVA and multiple comparisons with Bonferroni tests. For pairwise comparisons, means were analyzed via independent-samples t tests. A 95% confidence interval was applied in statistical analysis tests; descriptive statistics and analyses were performed using R version 3.2.3 (2015-12-10, copyright© 2015, The R Foundation for Statistical Computing free software package program). Results were considered statistically significant at P<0.05.

Results

COMPARISON OF THE EFFECTS OF SURFACE MODIFICATION ON POLYMER MATERIALS:

The 4 PEEK polymer groups differed significantly (Table 1) (all P<0.05). The lowest bond strength was in the control group (7.1±0.6 MPa), and the highest was in the sandblasting/argon plasma group (12.2±2.4 MPa). The PEKK groups also showed significant differences (all P<0.05). The lowest bond strength was in the control group (7.6±1.2 MPa), and the highest was in the sandblasting/argon plasma group (12.6±2.4 MPa). After the surface treatment application, higher bond strength values were obtained in the surface-treated groups compared to the control groups, and statistically significant results were obtained (all P<0.05). As a result of statistical comparison of the samples with sandblasting/argon plasma application in both polymer materials with samples from other surface treatment applications, a statistically significant difference was found between each group. Based on the results obtained in our study, it was concluded that sandblasting and argon plasma applications increased bond strength, and PEEK and PEKK sandblasting and argon plasma applications had a significant effect on bond strength compared to all samples in the other groups.

PEEK and PEKK polymers were compared statistically within their groups according to the applied surface treatments; the conformity of continuous variables to the normality distribution assumption was investigated with the Kolmogorov-Smirnov test, and homogeneity was investigated with the Levene test.

COMPARING POLYMER MATERIALS WITH EACH OTHER:

Comparisons were made between the control and surface-treated PEEK and PEKK groups. No significant differences were found between the PEEK and PEKK groups treated with argon plasma, sandblasting, or sandblasting/argon plasma (all P>0.05). Based on the values obtained in our study, it was concluded that argon plasma, sandblasting, and sandblasting/argon surface treatments did not produce a significant difference in bond strength when comparing the 2 polymer materials.

Statistical analysis of the means between independent groups between PEEK and PEKK polymers was done using one-way ANOVA, the Bonferroni test was used in multiple comparisons, and the independent-samples t test was used to examine the differences between the means of 2 independent groups.

CONTROL GROUPS: SEM images of the control group samples of PEEK (Figure 2A) and PEKK (Figure 2B) polymers showed small scratches and micro-sized pits on a smooth surface structure.

ARGON PLASMA GROUPS: SEM images of the argon plasma-treated PEEK (Figure 2C) and PEKK (Figure 2D) samples showed more distinct grooves and recessed areas compared to the control group.

SANDBLASTING GROUPS: After sandblasting surface treatment to PEEK (Figure 2E) and PEKK (Figure 2F) polymers, convex protrusions and deeper indentations formed by alumina particles were observed in the SEM images taken from the samples.

SANDBLASTING/ARGON PLASMA GROUPS: SEM analysis of PEEK (Figure 2G) and PEKK (Figure 2H) samples on which sandblasting/argon plasma application was applied showed large convex deposit areas and random alumina particles on the surface.

SURFACE ROUGHNESS VALUES:

AFM revealed that surface roughness was highest in the PEEK sandblasting/argon plasma sample (0.62 μm) and lowest in the control PEEK sample (38.17 nm). The highest Ra value was for the sandblasting/argon plasma PEEK sample, and the lowest was for the control PEEK group. Ra values were 53.92 nm in the PEEK argon plasma group and 0.42 μm in the sandblasting group. For PEKK, Ra values were 88.53 nm in the control group, 105.44 nm in the argon plasma group, 152.57 nm in the sandblasting group, and 199.85 nm in the sandblasting/argon plasma group (Table 2).

EDS:

EDS revealed atomic composition changes on the surfaces of the various samples. The PEEK control surface contained 71.59% carbon (C), 26.56% oxygen (O) and 0.29% aluminum (Al) by weight (Figure 3A). The PEKK control surface contained 76.03% C, 17.68% O, 0.30% Al and 5.47% titanium (Ti) (Figure 3B). After argon plasma treatment, the PEEK surface contained 60.27% C and 25.73% O, along with 10.38% nitrogen (N) and 1.47% fluorine (F) (Figure 4A). In the PEKK argon plasma sample, the surface composition was 60.70% C, 26.67% O, 11.31% N, and 1.31% (Al) (Figure 4B). After sandblasting, a significant decrease in C (to 53.10%) was observed on the PEEK surface, with O increasing to 35.66%. The surface also contained 8.03% N, 1.51% Al, and 1.22% F. For PEKK, sandblasting reduced the C level to 49.65% compared with the control, whereas O increased to 38.55%. Additionally, 7.77% N, 2.19% Al, and 1.60% sodium (Na) were observed.

Parallel with the observation of an increase in aluminum in the sandblasting group’s SEM images, the amount of aluminum was 0.29% for control PEEK and 0.30% for PEKK; and 1.51% for PEEK and 2.19% PEKK for the sandblasting group. In the PEEK sandblasting/argon plasma group, C content was 52.21%, lower than in other groups, accompanied by 33.09% O, 9.74% N, 1.14% Na, and 2.46% F. Compared to the sandblasting group, the Al level in the sandblasting/argon plasma group was lower, at 1.36%. After sandblasting/argon plasma treatment, the PEKK surface contained 51.46% C, 36.46% O, 10.59% N, and 1.35% Al, which were lower than in the sandblasting-alone group.

Discussion

We found higher bond strength values in the surface-treated groups compared to the control groups after the surface treatment. After surface treatments, the argon plasma, sandblasting, and sandblasting/argon plasma groups had greater bond strength for both PEEK and PEKK polymers compared to the control group. When both polymer materials were compared within their respective groups after surface treatments, a statistically ignificant difference was observed. Therefore, our first null hypothesis was rejected. According to ISO 10477.23 requirements, the acceptable shear bond strength for resin-based materials and resin frameworks is 5 MPa. However, it has also been reported that the minimum clinical shear bond strength for resin-based materials for oral environments should be between 10 and 12 MPa [27]. Our bond strength results for all surface treatment groups were higher than the lower limit of the ISO standard (5 MPa), while the sandblasting and sandblasting/argon plasma treatment groups achieved a value within the clinically acceptable range (10 MPa). After surface treatments were applied to PEEK and PEKK samples, no statistically significant difference was found when comparing the 2 polymer materials, including the control groups, in terms of bond strength. Therefore, our second null hypothesis was accepted.

Some high-performance polymers are biocompatible, exhibit good heat and solvent resistance, are near-perfect electrical insulators, and have strong abrasion and fatigue resistance. These properties may satisfy the esthetic demands of patients who prefer metal-free prostheses [19]. However, the use of high-performance polymers alone is not clinically feasible. These materials are poorly translucent, grayish or pearly-white in color, and opaque. Therefore, the polymer surface must be covered with a veneer [1]. Resin cement can be used to bond PEEK and PEKK polymers in prosthesis fabrication [2].

Studies have shown that high surface energy values, and therefore adhesion potential, can be achieved and maintained for months by storing PEEK in a dry atmosphere after plasma treatment. This suggests that atmospheric dryness is the sole parameter responsible for the aging of plasma-activated PEEK surfaces, making a dry atmosphere the most important parameter for storing plasma-activated surfaces [28]. In previous studies, to measure the bond strength with resin adhesive material, the samples were kept in dry air for 1 h and then the bond strength was assessed [29]. In our study, the bond strength test was performed after the application of the sample resin cement and keeping it in dry air for 1 h to ensure activation of the argon plasma application.

Increased surface energies of PEEK and PEKK are essential to improve their bonding with veneers [30,31]. Sandblasting treatments have historically been used to treat PEEK. Studies have reported that increased particle size after sandblasting applied to PEEK material improves bond strength [19].

In medicine and dentistry, plasma therapy is used more frequently, and its areas of use are steadily increasing [32]. Studies suggest that plasma application increases bond strength by making the material surface more hydrophilic. Plasma application has attracted attention due to its use in dentistry [33]. The present study assessed the bond strength achieved using argon plasma, which is increasingly used in medical fields, with polymer materials, singly and in combination with the routine sandblasting surface treatment. The statistical data we obtained also show that atmospheric cold plasma application increases the bond strength of resin-based materials to PAEK polymers [34].

Studies have concluded that surface treatments (sandblasting and plasma treatment) significantly increase the bond strength between the PEEK material and coating composites. The bond strength values obtained in our study, particularly those obtained after sandblasting/argon plasma treatment, are consistent with previous studies. Furthermore, similar results were obtained with PEKK polymers [20,35]. Zhou et al evaluated PEEK samples treated with argon plasma for 5, 15, and 25 min for shear bond strength. They concluded that 5-min argon plasma treatment resulted in lower shear bond strength values than 15-min and 25-min argon plasma treatments. They reported that 15-min and 25-min argon plasma treatments did not produce any difference in shear bond strength values. However, studies examining the effects of non-thermal plasma on PAEK polymers are limited to PEEK materials. The reason for using 20 min of argon plasma treatment in our study was that, based on previous studies, there was no statistically significant difference in bond strength between 15 min and 25 min of argon plasma treatment. Therefore, the results of the 20-min argon plasma treatment were not previously investigated and evaluated in the literature [36]. Another study, in which the bond strength was measured after plasma application for 3 min and 35 min before the sandblasting surface treatment, found that the low-pressure plasma treatment using oxygen plasma for 35 min before sandblasting was the most effective method for PEEK polymer [35].

Labriaga et al studied PEKK surfaces treated with non-thermal plasma, sandblasting, and a combination of both. After application of resin cement, the shear bond strength values of samples that did not undergo thermal aging were 8.2±2.2, 11±2.5, 14.4±2.1, and 17.7±1.1 MPa, respectively. Although the order of bond strengths is consistent with our findings, our values were lower. For the control group, this may reflect our use of finer-grit sandpaper. For the plasma group, we used a different plasma type. For the sandblasting group, our sandblasting time was shorter than that used in the cited study [37]. Zhou et al reported that the bond strengths of PEEK samples subjected to argon plasma and sandblasting, then bonded using resin cement (RelyX™ Unicem®), were 4.0±0.2 MPa in the argon plasma group and 1.4±0.2 MPa in the sandblasting group. Our values were higher, likely because Zhou et al used 50-μm aluminum oxide particles for sandblasting, and the argon plasma conditions differed [38].

In biomedical applications, scanning electron microscopy (SEM) analysis is often used to examine and evaluate the topographic structures of cells or tissues [39]. In our study, SEM was used to characterize the surface and examine the topographic properties of untreated and treated samples. We also compared our SEM images with those from other studies. Zhou et al reported that SEM images of PEEK subjected to various surface treatments generally had regular structures with a few scratches. Argon plasma created grooves, cracks, and deposits, whereas sandblasting induced convex deposits and increased surface roughness. The surface morphological changes observed in our SEM images were consistent with those reported by Zhou et al [38].

Atomic force microscopy (AFM) analysis interacts with the probe tip, which touches the surface of the sample using a cantilever tip. The deflection of the cantilever tip allows the sample’s surface properties and morphology to be visualized [40]. In our study, AFM analysis was used to examine the morphological changes and surface roughness that occur on the surface after the surface treatment application. AFM analysis revealed that the surface roughness values of both PEEK and PEKK followed the order: control < argon plasma < sandblasting < sandblasting/argon plasma. Schwitalla et al subjected different PEEKs (Juvora®, Vestakeep DC4420®, Vestakeep DC4450®) to plasma treatment with argon and oxygen at 70°C, sandblasting with 110-μm aluminum oxide powder, and a combination of both treatments; they investigated the shear bond strength of a composite resin [20]. As in the present study, the highest surface roughness for PEEK was observed after sandblasting/argon plasma treatment.

Energy-dispersive spectroscopy (EDS) analysis is widely used for elemental analysis of solid materials. This method, combined with SEM, provides information about the elemental composition of the samples [41]. In our study, the elemental changes that occurred after surface treatments were compared with the control group using EDS analysis. The effect of surface elemental changes on bond strength was investigated. Non-thermal plasma treatment is a valuable surface roughening option because it enhances surface chemistry without altering key properties of the material. Cold plasma changes the surface chemistry, creating radicals such as O3, OH, H2O2, NO, and OH. We speculate that the higher bond strengths observed after argon plasma treatment (compared with the control) change surface chemistry and the formation of chemical bonds. Bhatnagar et al reported that EDS analysis following low-pressure plasma application may be a suitable method for determining the functional groups added to the PEEK surface. They reported that plasma applications can create various oxygen and nitrogen functionalities on the surface [42]. Wang et al reported that argon treatment is associated with the generation of C radicals, which can form polymeric cross-links due to their unstable nature [43]. In our study, the C content in the PEEK and PEKK control groups was 71.59% and 76.03%, respectively. After argon plasma treatment, the C content decreased to 60.27% in PEEK and 60.70% in PEKK. Additionally, whereas N was absent in both the PEEK control and PEEK argon plasma-treated samples, it was detected at levels of 10.38% and 11.31% in the PEKK sandblasting and plasma-treated groups, respectively. The decrease in C atom ratios on the surfaces of both polymers supports that argon plasma breaks C–C bonds; the resulting free radicals contribute to increased bond strength. Al levels were 1.51% and 2.19% in the sandblasted PEEK and PEKK samples, respectively; they decreased to 1.36% and 1.35% in the PEEK and PEKK samples subjected to sandblasting followed by argon plasma treatment. These decreases, supported by SEM images, may be due to the decrease in the proportion of surface Al atoms due to argon plasma treatment, which reduces the bond strength.

Argon plasma treatment can trigger the formation of free radicals by breaking surface C–C and C–H bonds, allowing the radicals to bind with other polymer molecular structures or adhesive agents (such as Visio.link®), thereby increasing bond strength [6]. We conclude that sandblasting followed by argon plasma treatment significantly increases bond strength, likely due to elemental changes at the surface.

Previous studies found that aging of PEEK polymers following plasma treatment caused surface energies to return to their initial starting values based on the material’s plasma parameters. Regardless of the efficiency of the plasma aging process after plasma application to the polymer and the capabilities offered by plasmas, this process cannot be relied upon unless the aging kinetics are controlled [28]. Therefore, in our study, EDS analysis was performed within 1 h after the argon plasma treatment to determine the surface chemistry of the samples to ensure the plasma aging process was controlled.

One sample from each group was selected for scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), and atomic force microscopy (AFM) analyses. Our results agree with those of a previous study [44]. Schwitalla et al assessed bond strength values in groups where sandblasting and plasma treatments were combined on different PEEK (Juvora®, Vestakeep® DC4420, Vestakeep® DC4450) materials after surface treatments with plasma, sandblasting, and sandblasting+plasma, reporting values of 19.8±2.46, 15.86±4.39, and 9.06±3.1, respectively. Compared with the bond strength values obtained in our study, their results were lower for both the PEEK polymer and the PEKK polymer, but not for the DC4450 PEEK polymer [20]. We believe this is due to the hot plasma application they used and the difference in the PEEK material used. The sandblasting/plasma method applied in these studies is a surface treatment combination that may be superior to many other methods due to its ease of clinical use [6]. Based on our data, we conclude that sandblasting and argon plasma treatment is a suitable method for clinical applications for increasing bond strength.

Conclusions

We concluded that argon plasma treatment was effective in increasing the bond strength of PEEK and PEKK polymers to resin cement. Significant results were obtained in the bond strength of PEEK and PEKK polymers with argon plasma application after sandblasting and resin cement application. Therefore, sandblasting/argon plasma treatment for both polymer materials to increase bond strength may be a promising clinical application.