Introduction: Debonding is the most common complication of fiber-reinforced posts. Airborne-particle abrasion (APA) has been suggested to increase the adhesion of resin cements to the surface of fiber posts. However, it is unclearwhich is the most favorable abrasion protocol and if it affects the failure load of the fiber post. 

Objective: The aim of this study was to compare the surface roughness and characteristic failure load of three fiber post systems following different APA protocols.

Materials and Methods: 150 hundred fiber reinforced posts with continuous unidirectional fibers were tested. The posts belonged to 3 different post systems: GF (GC fiber post; GC America Inc), QF (GT fiber post; Dentsply Sirona USA), ZF (ICELight; Danville Materials Inc) (fig 1.). Post from each system were randomly assigned to 5 different treatment protocols (10 posts in each treatment group): NT (no treatment – as received), E0 (cleaned with 96% ethanol solution), E2 (APA for 5 seconds/mm2), E5 (APA for 5 seconds/mm2), E10 (APA for 10 seconds/mm2). For treatments E2, E5, and E10, APA was performed with 50—m aluminum oxide particles emitted from 10 mm and at a pressure of 0.2 MPa. Post were manually rotated so that the entire post surface was blasted by the aluminum oxide particles. The posts were then ultrasonically cleaned in deionized water for 10 minutes, rinsed with 96% ethanol solution and dried with an oil free stream of air.

  Pre testing laser confocal microscopy was performed on each post under a 3D measuring laser confocal microscope using a 20X magnification dry objective lens. The surface roughness in 3D was recorded in (Sa). The post surface of each post was also evaluated qualitatively for surface alterations as a result of the different treatment protocols.

  Mechanical testing: a universal testing machine (5500 Instron) was used to load the posts until failure. Testing was similar to a three-point bending procedure. A stainless-steel nose was used with crosshead speed of 1mm/min. Failure was defined as the load where the specimen showed a sudden drop in the force-time graph and/or visualization of crack formation. The testing configuration for each of the 3 post systems is shown in figure 2, red lines represent the location of supports and green arrow represent loading point.

Results and Discussion: The (SD) Sa roughness values of each of the 15 groups were measured (Table 2) and ANOVA showed statistically significant difference among the groups. Multiple comparisons with the Games-Howell correction showed that APA for 2 seconds/mm2 resulted in statistically significant increase of surface roughness values in all 3 tested post systems to their respective controls (NT).  Surface roughness after 5 and 10 seconds/mm2 APA was not statistically significant in all post systems compared with 2 seconds/mm2. Cleaning the post surface with 96% ethanol solution resulted in statistically significant increase in surface roughness for GF post system but not for QF or ZF post systems.

  Qualitative evaluation of the obtained confocal laser images lead to the observations: 96% ethanol cleaning did not cause any obvious post surface alterations in all test systems. In GF posts, APA resulted in gradual breaking of the superficial glass fibers and composite resin matrix (Fig 3). QF posts, APA resulted in gradual breaking of the superficial quartz fibers and epoxy resin (Fig 4). ZF posts, APA resulted in almost complete removal of the superficial glass fibers (Fig. 5)

  Table 3 summarizes the results of the two-parameter Weibull analysis. GF: the likelihood ratio plot showed a statistically significant decrease of the failure load after the posts were subjected to APA for 2 seconds/mm2 compared to the control. QF and ZF showed no statistically significant difference of the failure load after APA for 2 seconds/mm2.

  For all systems an increase in time of APA resulted in gradual decrease of failure load but not always significant. QF APA for 5 seconds/mm2 lead to statistically significant decrease in failure load. Failure load was not statistically significant for ZF any treatment groups and between the groups cleaned with 96% ethanol solution and their controls.

  Evaluation of the obtained confocal laser microscopy images of the representative specimens after failure load to the following observations: GF failure load created fracture lines in resin when the posts were subjected to APA for 2 sec/mm2. When APA time increased to 5 or 10 sec/mm2 delamination or debonding of fibers . QF failure load created fracture lines in the resin matrix as well as some bending/rupture of fibers for the control group (similar in QFE0 and QFE2). APA time increased to 5 or 10 sec/mm breaking and debonding fibers. ZF failure load created lines in the fracture matrix that were perpendicular to the fiber orientation as well as bending/rupture of fibers. All tested protocols as well as the control showed similar failure patterns for post system ZF.

The study showed that 3D surface roughness was significantly increased after APA in all post systems. Characteristic failure load was significantly decreased in

posts GF and QF after different APA times. Also, surface topographical features and failure patterns were different among the groups. The results support the rejection of the null hypotheses. Future research could be directed toward comparing the effect of different APA protocols on adhesion to radicular dentin and/or fatigue resistance of endodontically treated teeth.

Conclusions:

3. APA resulted in qualitative surface changes in all posts systems.