The impact of non-thermal plasma on the adhesion of polyetherketoneketone (PEKK) to a veneering composite system

Highlights

• Influence of plasma with various feeding gases on the bond strenght between PEKK and veneering resin wasn’t investigated yet.

• Plasma treatment with acetylen may provid the shear bond strenght results comparable to the adhesive application.

• Acetylene plasma provided the highest bond strength elevation among air, argon, oxygen and nitrogen feeding gases.

Abstract

The PEKK material can be used in prosthodontics for framework manufacturing and is commonly laminated with veneering composites to achieve a better esthetics. Various surface treatment methods including sandblasting, etching, laser and cold plasma treatments were reported to enhance the adhesive properties of dental polymers. Both tensile and shear bond test were employed to quantify the bond strength between PEKK and veneering composites. The present in vitro study aims to evaluate the influence of acetylene, argon, air, nitrogen and oxygen plasma on the shear bond strength between PEKK and one veneering composite.

Firstly, to determine which bond test type should be applied, n = 40 PEKK specimens were treated with argon plasma. Both shear and tensile bond tests were performed and compared to the control group (n = 40). In shear bond testing, values were 8.14 ± 1.70 MPa for Argon plasma while 5.83 ± 1.42 MPa for control group. In tensile bond testing, Argon plasma 1.50 ± 0.51 MPa while control group 0.58 ± 0.50 MPa.

Afterwards n = 160 PEKK specimens were treated with rocatec sandblasting (n = 20), adhesive (n = 20), acetylene (n = 20), argon (n = 20), air (n = 20), nitrogen (n = 20), oxygen (n = 20) plasma types and compared to the untreated control group (n = 20) using shear bond strength test (SBS). Additionally surface roughness and scanning electron microscopy analyses were performed. The following SBS values were revealed: 10.22 ± 1.06 MPa for rocatec; 9.89 ± 3.08 MPa for acetylene, 9.16 ± 1.48 MPa for adhesive, 7.54 ± 1.52 MPa for argon, 7.09 ± 1.99 MPa for air, 7.03 ± 1.48 MPa for nitrogen, 5.69 ± 1.59 MPa for oxygen plasma types and 4.71 ± 1.54 MPa for the control group.

All groups, except control group, showed SBS over 5 MPa, which means that they are suitable for the clinical application, according to ISO 10477. Acetylene showed the highest SBS among all plasma types (p < 0.0001), which was on a level of rocatec sandblasting group. Rocatec and acetylene groups demonstrated Ra values significantly different to the reference group (p < 0.0001).

Plasma treatment especially with acetylene gas can be an effective more convenient surface treatment method for strengthening the bond strength between PEKK and veneering composites than traditional sandblasting/adhesive treatment.

Introduction

Over the past years, high-performance polymers known as polyaryletherketones (PAEK) were introduced in the field of prosthodontics as an alternative to traditional metal or zirconia frameworks (Oh et al., 2017). Because of their chemical nature, these materials became attractive in dentistry as they provide high dimensional stability, chemical resistance, and superior strength-to-weight ratio (Kurtz, 2019). Therefore different fixed and removable prosthetic frameworks could be effortlessly constructed from these materials in thinner sections without affecting their strength (Arvai and Schmid, 2014; Najeeb et al., 2016; Dawson et al., 2017).

Polyetherketoneketone (PEKK) is the most recent material among all PAEK used in prosthodontics and has a big potential in this field (Fuhrmann et al., 2014). Although, because of its opaque grey shade, it must be veneered with composite resins for application in anterior region. Like all polymers, PEKK possesses inert hydrophobic surface, which must be treated before bonding with veneering resins to avoid chipping and fracture of veneering materials (Najeeb et al., 2016).

Various surface treatment methods were described in the literature to improve the bonding between PEKK and veneering composites. Fuhrmann et al. were the first to investigate the influence of different air abrasion techniques on the bonding behavior of two commercially available PEKKs (Fuhrmann et al., 2014). The effect of chemical etching with sulfuric and vinyl sulfonic acids in different concentrations on the bonding to PEKK materials was also assessed in the following researches (Lee et al., 2017; Sakihara et al., 2018). The inconclusive results of these studies prompt to search for alternative surface treatment methods that could provide consistently better bonding.

Generally, the plasma surface treatment was suggested as an effective method for treating different polymer surfaces (Comyn et al., 1996; Grace and Gerenser, 2003). In dentistry the plasma was initially introduced for cleaning and sterilization of dental instruments, root canal disinfection in endodontic treatment, teeth bleaching (Cha and Park, 2014). A significantly positive impact of air, oxygen, and argon plasma treatment on the bonding strength between polyetheretherketone (PEEK) materials and veneering composites was reported in a row of studies (Stawarczyk et al., 2013, Stawarczyk et al., 2014; Schwitalla et al., 2017; Bötel et al., 2018; Younis et al., 2019). Nevertheless, in contrast to PEEK, an extra ketone group in PEKK makes this material more resistant to the surface treatment and little evidence is provided about its surface treatment methods. The cold plasma bombardment of the PEKK may provide a greater adhesion to the resin cements (Labriaga et al., 2018). To authors’ best knowledge the influence of surface treatment with various plasma types on the bonding between PEKK and veneering composites was not compared yet.

In the literature, shear and tensile tests are used mutually to assess the bond strength between PEKK and veneering composite resins or resin cements (Stawarczyk et al., 2013, Stawarczyk et al., 2014; Fuhrmann et al., 2014; Lee et al., 2017; Sakihara et al., 2018; Fokas et al., 2019). Even though, tensile tests show more uniform and homogenous stress distribution, the specimens preparation technique is more sensitive and time-consuming in comparison to shear bond test (Sirisha et al., 2014).

Therefore, the first aim of the present study was to propose, which test among shear and tensile tests suits better for a bond strength analysis of veneering composites to PEKK. The second aim was to quantify the influence of sandblasting, adhesive application and various gaseous plasma treatment on the bonding of veneering composites to PEKK. It was hypothesized that there would be no significant differences between air, argon, oxygen, nitrogen and acetylene plasma types after thermocycling.

Fig. 1 depicts the workflow of this study composed of 2 tests.

PEKK specimens (n = 240) with the following dimensions 20 mm × 10 mm x 2 mm were designed in Solidworks software (Solidworks Corp, Massachusetts, USA) then exported in STL format and milled of PEKK CAD/CAM blocks (Pekkton ivory, Cendres + Métaux, Biel-Bienne, Switzerland). Afterwards, all milled specimens were polished using abrasive papers P180 followed by P320 (Buehler UK LTD, Coventry, England). Subsequently, all specimens were cleaned in an ultrasonic bath containing 70% ethanol for 20 min using an ultrasonic device (Sonorex, Bandelin, Berlin, Germany) then left to dry in the air prior to surface treatment.

For the surface treatment of the milled specimens with different plasmas, special jigs were designed using Siemens NX 10.0 CAD software (Siemens PLM, Texas, USA) following ISO 10477. Two different jigs were constructed. Seating jigs, where specimens were seated on and bonding jigs with dimensions 20 mm × 10 mm x 2.5 mm. With these jigs the bonding steps were made through a hole in its center with a diameter of 5 mm. The seating jigs were printed with Fused Filament Fabrication (FFF) technology (MakerBot Replicator+, MakerBot Industries, NY, USA), and the bonding jigs were printed using stereolithography (SLA) (Form2, Formlabs, Somerville, Massachusetts, USA).

All materials and machines used for the plasma treatment, sandblasting and bonding are listed in Table 1.

80 specimens were divided into four groups. Two control groups were made for both shear and tensile tests (n_{test1_contr_shear} = 20, n_{test1_contr_tensile} = 20), in which the direct bonding with a veneering composite (Sinfony, 3 M ESPE AG, Seefeld, Germany) was performed without any surface treatment. While in the other two groups, the specimens were treated with low-pressure cold argon plasma DENTA PLASPC (Diener electronic GmbH, Ebhausen, Germany) (n_{test1_AR_shear} = 20, n_{test1_Ar_tensile} = 20) according to manufacturer instructions with the following settings of surface treatment: 10 min treatment duration at pressure of 0.3 mbar, temperature of 20 °C, frequency of 40 kHz and power output of 100 W.

160 specimens were sorted randomly, comprising the following 8 groups of 20 specimens each: Control (untreated surface) (n_{test2_contr} = 20), sandblasting (n_{test2_sandbl} = 20), adhesive (application of adhesive without prior plasma treatment) (n_{test2_adh} = 20) and 5 different gaseous cold plasma treatment groups including Argon (n_{test2_argon} = 20), Oxygen (n_{test2_oxygen} = 20), nitrogen (n_{test2_nitrogen} = 20), air (n_{test2_air} = 20) and acetylene (n_{test2_acetyelen} = 20).

In the sandblasting group the specimens were treated with 110 μm Al2O3 (Rocatec Pre, 3 M Espe, Seefeld, Germany) following by tribochemical coating using 110 μm silica-modified Al2O3 (Rocatec Plus, 3 M Espe, Seefeld, Germany). Both were done at a pressure of 2.8 bar, with vertical irradiation at a beam distance of 1 cm and a beam time of 15 s per 1 cm² area. Afterwards the silane (Rocatec Sil, 3 M Espe, Seefeld, Germany) was applied with microbrush for conditioning the surface and left dry for 5 min.

In the adhesive group the adhesive (Visiolink, Bredent, Senden, Germany) was applied with a microbrush on the surface of specimens through the hole in the 3D printed jig. The specimens were immediately cured in a light-curing unit (Speed Labolight, HAGER &WERKEN GmbH &Co.KG, Duisburg, Germany) with wavelengths ranging from 320 to 550 nm for 90 s.

In the plasma group the specimens were treated with cold plasma utilizing corresponding feeding gases for 10 min at the pressure of 0.3 mbar, temperature of 20 °C, frequency of 40 kHz and power output of 100 W.

Following the surface treatment, all specimens were put into the printed jigs. With the use of printed jigs, bonding procedures were performed to produce composite button of 2 mm thickness and 5 mm diameter. First, the opaquer was applied in powder-liquid ratio of 1:1 with clean disposable brushes and was light cured for 10 s using a halogen curing unit with 400–515 nm wavelength. Thereafter the first composite layer of 1 mm was applied through the jig directly from the dispenser and cured for 5 s (ELipar Trilight, 3 M Espe, Seefeld, Germany). The second composite layer was applied to fill the entire jig hole and cure again intermediately for another 5 s. Afterwards then specimens were delicately detached from the jigs and undergone the final curing for 1 min of light exposure without vacuum and 14 min light curing with vacuum (Visia Beta Vario, 3 M Espe, Seefeld, Germany) (Fig. 2).

All veneered specimens (n_test1, n_test2) were subjected to artificial aging by using Thermocycler (SD Mechatronik, Feldkirchen-Westerham, Germany). Following the ISO 10477, the specimens were exposed to thermal stresses by immersing them in distilled water for 5000 cycles at two temperatures 5 °C (±1) and 55 °C (±1). The specimens were kept immersed in water in each bath for 30 s with a dwell time of 20 s in between both temperatures.

The specimens were placed in a holding device, which was installed in an universal testing machine. This holding device is designed to ensure axial force application without any momentum. The specimens were pulled apart axially from bonding jigs through an upper chain, while the system is self-centered the whole procedure to purely tensile stresses. The crosshead had a constant speed of 1 mm/min until debonding or fracture of veneering.

The SBS test was performed using the same universal testing machine (Z010 Zwick Ulm, Germany). For shear testing a customized specimen holder was made, through which specimens were firmly fixed throughout the testing procedure. A chisel-shaped rod applied force constantly parallel to the bond surface at a distance of 0,5 ± 0,02 mm from the surface of the PEKK specimen with a crosshead speed of 1 mm/min and starting from a 0-N load, which increased gradually until fracture of the veneering composite. Then, the maximum force at fracture was detected.

The TBS and SBS were calculated according to the following equation:

$$\mathrm{\sigma }=\text{F}/\text{A}$$

where σ is the TBS or SBS, F is the load applied in Newtons, and A is the bonded area in mm². Specimens that did not survive the aging test and showed premature debonding of veneering composites during thermocycling were assigned 0 MPa and considered as pre-failures. Fig. 3 represents a schematic illustration of shear and bond testing.

Following shear bond testing, the specimens were examined under microscope Wild M400 photomacroscope (Wild Heerbrugg, Gais-Switzerland) to determine the type of failure as follows:

• adhesive failure which means no resin remnants were left on the PEKK surface.

• cohesive failure where failure is in the bulk layer of the resin.

• mixed where resin remnants partially left on PEKK surface exposed.

The surface roughness of a specimen from each group was measured using Perthometer S6P (Mahr, Göttingen, Germany). This device had a needle with a 2 μm diamond tip, which allowed two-dimensional tracing of a given surface. The stylus traversed the center of each specimens surface at a constant speed of 0.5 mm/s in an area of 3 mm length and width. 121 measurement lines with 25 μm distance between the lines were performed. In the MountainsMap Universal 7.3 software (Digital Surf, Besançon, France) the surface roughness (Ra) was calculated.

The surface topography of one specimen per group was further analyzed by using Zeiss LEO 1430 microscope (Zeiss, Oberkochen, Germany) at ×500 and ×5000 magnification.

All gathered data was analyzed with JMP 13.1 software package (SAS Corp. Heidelberg, Germany). The measurements were tested for normality of distribution by goodness of fit with Kolmogorov test using p < 0.05. The bond strength data was distributed normally, therefore ANOVA and Tukey-Kramer post-hoc test were applied on the alpha level of α = 0.05. As the non-normal distribution was revealed for the surface roughness measurements, the Wilcoxon test was applied to determine the statistical significance. The pearson correlation was calculated between the R_a and SBS values (see Table 3).