Hydrofluoric acid concentration and etching time on bond strength to lithium disilicate glass-ceramic

The objective of this study to evaluate the effect of different HF concentrations and etching times on the microshear bond strength (μSBS) of LD to resin cement. Forty LD sections (8x8 mm) of 3-mm thickness were randomly distributed (n=10) in accordance with the HF concentration (5 or 10%) and surface etching time (20 or 60 sec). The specimens were silanized and received an air-thinned layer of a light-curable adhesive. Six translucent tubes (0.8-mm diameter and 1-mm height) were positioned over each LD section, filled with resin cement and light-cured. After 24 h of storage, the tubes were carefully removed and the specimens were submitted to the μSBS test. The results submitted to a two-way analysis of variance and Sidak post hoc test (α=.05). Representative HF-etched specimens and one non-etched LD specimen were observed under a field-emission scanning electron microscope. The interaction between the HF concentrations and etching times was not significant (p=0.075). No significant differences were observed regarding HF concentrations and etching times (p=0.06 and p=0.059, respectively). Surfaces of specimens etched with 10% HF for 60 sec were found with grooves and microcracks. The μSBS of LD to resin cement was not significantly influenced by different HF concentrations and etching times; however, the LD surface morphology was found considerably modified . = significant differences were observed regarding HF concentrations and etching times (p=0.06 and p=0.059, respectively). Regarding each etching time, no significant difference was found between different HF concentrations (p=0.13 and p=0.31).

An adequate surface treatment has a great influence on the tensile bond strength between lithium disilicate glass-ceramic (LD) and resin cement (Aboushelib & Sleem, 2014;Vila-Nova et al., 2019).

Methodology
This is an experimental, in vitro, quantitative and qualitative study. LD blocks were cut in 8x8 mm sections of 3-mm thickness by using a water-cooled and low-speed diamond saw (Isomet 1000, Buehler, Lake Bluff, IL, USA). LD sections were ultrasonically cleaned with distilled water for 10 min and fired following the crystallization program recommended by the manufacturer. After cooling, the specimens were positioned in polyvinyl chloride (PVC) plastic rings and embedded in epoxy resin (Epo-Thin Resin, Buehler, Lake Buff, IL, USA). Each specimen was mechanically wet polished (Struers DP-10, Panambra, São Paulo, SP, Brazil) with 600-, 800-and 1200-grit silicon carbide paper (Acqua Flex-Norton, São Paulo, SP, Brazil) to establish a uniform surface and ultrasonically cleaned.
The LD specimens were randomly distributed in four groups (n=10) in accordance with the HF concentration (5% or 10%) and surface etching time (20 sec or 60 sec). After rinsing with air/water spray for 30 sec and air-drying, the LD specimens were silanized and left undisturbed for 60 sec. A single layer of a light-curable adhesive was applied, air-thinned and light-cured for 20 sec using a light-emitting diode (LED) unit (Bluephase N, Ivoclar Vivadent, Schaan, Liechtenstein) with an output of 1,000 mW/cm 2 . Each LD specimen was covered with an acid-resistant and double-sided adhesive tape perforated with six equally distant 1-mm-diameter holes to delimit the bonding áreas. Translucent polyethylene Tygon tubes (Tygon Medical Tubing, Saint-Gobain, Akron, OH, USA) with an internal diameter of 0.8 mm and 1-mm height were positioned over the lumen of the tape perforations. After careful insertion of light-curable resin cement into each tube, a Mylar strip was gently pressed over the filled tube to avoid displacement and light-cured for 40 sec using the LED unit. After 24 h of storage in distilled water at 37 o C, the tubes were carefully removed using a sharp blade to expose the resin cement cylinders. Specimens were examined under magnification to exclude possible irregularities at the bonding area.
The PVC rings were attached to a universal testing machine (4444, Instron, Canton, MA, USA). A sharpened stylus was used to apply shear forces on to the side of each resin cement cylinder at a crosshead speed of 0.5 mm/min until failure. The μSBS results were calculated in MPa by dividing the load at the failure by the bonding surface area (mm 2 ). Since the Kolmogorov-Smirnov showed normal data distribution (p=0.082) and the Levene test demonstrated the equality of variances (p=0.062), the results were submitted to two-way analysis of variance (ANOVA) and Sidak multi-comparison post hoc test

Results
The mean μSBS values are shown 9195734173 in Table 1. The interaction between the HF concentrations and etching times was found not significant (p=0.075). No significant differences were observed regarding HF concentrations and etching times (p=0.06 and p=0.059, respectively). Regarding each etching time, no significant difference was found between different HF concentrations (p=0.13 and p=0.31).
The SEM evaluation revealed different surface morphology of LD in function of HF concentration and etching time.

Discussion
HF etching is recommended for LD surface treatment due to its selective dissolution of the glassy matrix, which exposes 0.5 to 4 μm-length crystals and creates microporosities that increases surface area and resin cement interlocking.
Although the manufacturer of IPS e.max CAD (Ivoclar Vivadent, Schaan, Liechtenstein) specifically recommends the use of HF 5% for 20 sec, many clinicians extend the etching time of LD restorations. This misunderstanding may be associated with the former surface treatment protocol of feldspathic ceramics and the reduced commercial availability of HF 5%.
Since the results of this study showed that different HF concentrations and etching times have no significant effect on bond strength to resin cement, the null hypotheses were not accepted. However, SEM evaluation showed that etching with HF 5% and 10% during 60 sec created grooves and microcracks on the LD surface possibly due to glassy matrix dissolution and removal of unsupported crystals. Although Posritong, et al., (2013), observed that the structural changes caused by HF after different etching times did not significantly affect the flexural strength of a glass-based veneering ceramic, the results of this study showed that the extension of HF etching time can deeply damage the surface and may weak LD structure (Brentel, et al., 2007;Zogheib, et al., 2011, Cardoso, et al., 2021. Other studies have also reported that the increase in surface roughness caused by HF etching is associated with a reduction in LD flexural strength (Zogheib, et al., 2011;Hooshmand, et al., 2008). Ramakrishnaiah, et al., (2016), reported that longer HF etching times of silica-based glass-ceramics results in larger, deeper, and irregular grooves due to a significant change in pore pattern, crystal structure, and surface roughness. Therefore, it seems reasonable to follow the HF etching manufacturer's recommendation of 20 sec (Guarda, et al., 2013).
Since different HF concentrations used for 20-sec etching did not significantly affect the immediate bond strength to resin cement, it is recommended to use 5% HF to minimize LD damage. The SEM evaluation of this study showed that 5% HF etching for 20 sec dissolved an adequate amount of glassy matrix, which corroborates with previous research (Sundfeld, et al., 2018;Prochnow, et al., 2017). Although no severe or even lethal accidents involving the use of HF in dentistry have been found in the literature, this acid must be used at the current lowest effective concentration of 5% due to toxicological and biological aspects (Sundfeld, et al., 2018;Trakyali, et al., 2009;Lise, et al., 2015). Since HF is highly volatile and toxic to organic tissues, it can cause injuries to the respiratory or digestive system if when inhaled or ingested (Kalavacharla, et al., 2015). Moreover, its corrosiveness can harm both patients and clinicians soft tissues and eyes, and the severity of injuries is directly associated with the exposure time and HF concentration (Ozcan, et al., 2012).
It is believed that shear stresses are involved in the majority of ceramic restorations failures (Pattanaik & Wadkar, 2011;Shimada, et al., 2002), thus, the μSBS was used in this in vitro study to evaluate the bond strength between resin cement and LD. The choice for micromechanical tests is based on the reduction of specimen size that minimizes the occurrence of cohesive failures and the stress around the adhesive interface is more evenly distributed (Lise, et al., 2015). In addition, the generation of pre-testing stresses at the adhesive interface of μSBS specimens is substantially reduced when compared to the specimen preparation for microtensile bond strength test (Panah, et al., 2008). Some authors claim that μSBS allows multiple specimens testing on the same ceramic surface (Shimada, et al., 2002).
The results of this research must be carefully interpreted since in vitro studies are carried out in dry and static conditions. The exposition to oral fluid could cause hydrolysis and change the mechanical properties of LD, while changes in temperature and pH could have an influence on bond strength (Venturini, et al., 2015;Pattanaik, et al., 2011).

Conclusion
The use of 5% HF for 20 sec is the most appropriate LD etching protocol due to no reduction of bond strength to resin cement, shorter application time, lower risk of toxic/corrosive, and damage to the ceramic structure.