1

Introduction

All-ceramic restorations have well been adopted in daily clinical practice as fixed dental prostheses or FDPs. Thanks to its biocompatibility and promising mechanical properties, yttria-stabilized tetragonal zirconia polycrystalline (Y-TZP) ceramics can be used as an alternative for conventional metal or metal-ceramic restorations. More recently, dental zirconia has also been introduced for full-contour ‘all-zirconia’ crowns and bridges, as well as for implant and implant abutments.

In most cases, all-ceramic restorations are CAD/CAM machined from pre-sintered ceramic blocks, and sintered to full density afterwards. To improve the shape of the zirconia core, the dental technician often needs to additionally abrade the zirconia core . Furthermore, due to the high chemical inertness, different mechanical surface pre-treatments have been recommended to improve the bonding of composite cement to zirconia, this in light of an adhesive luting procedure. For instance, aluminum-oxide (Al ₂ O ₃ ) sandblasting or tribochemical silica sandblasting with 30- and 110-μm silica-coated Al ₂ O ₃ particles have been shown not only to roughen, but also to chemically activate zirconia, the latter thus making it more receptive for chemical bonding via silane coupling agents . Because zirconia ceramics exhibit a stress-induced transformation, the surface of the abraded or sandblasted zirconia will be transformed, i.e., constrained as well as damaged; this may influence its long-term performance under clinical conditions .

It is well known that hundreds of zirconia total hip prosthesis heads failed catastrophically between 1999 and 2001, which led to its withdrawal from the market soon after . Later in 2007, the problem of catastrophic failures was attributed to low-temperature degradation (LTD), i.e. transformation of the metastable tetragonal to monoclinic phase (at 20–250 °C), initiated and accelerated by water penetration . The cause of the failures was related to an accelerated tetragonal to monoclinic phase transformation of zirconia in a limited number of batches . Although the manufacturing process of those orthopedic zirconia femoral heads is significantly different from that of dental zirconia, more research attention is recently also devoted to LTD of dental zirconia . At the moment, however, only few papers reported on the influence of surface treatment on LTD of dental zirconia , whereas any direct comparison among different dental zirconia is currently missing. The objective of this study was therefore to evaluate the influence of different surface treatments on the LTD behavior of dental zirconia. The null hypothesis tested was that different surface pre-treatments do not affect the LTD behavior of dental zirconia.

2

Materials and methods

The study design is schematically explained in Fig. 1 . Fully-sintered zirconia specimens, five so-called ‘Y-TZP’ zirconia (Aadva, GC, Tokyo, Japan; In-Ceram YZ, VITA, Bad Säckingen, Germany; IPS e.max ZirCAD, Ivoclar Vivadent, Schaan, Lichtenstein; LAVA Frame and LAVA Plus, 3M ESPE, Seefeld, Germany), one ceria-stabilized tetragonal zirconia polycrystalline (Ce-TZP)/Alumina (Al ₂ O ₃ ) zirconia (NANOZR, Panasonic, Osaka, Japan), and one Y-TZP/Al ₂ O ₃ zirconia (ZirTough, Kuraray-Noritake Dental, Tokyo, Japan) were provided by the manufacturers. All specimens were obtained in the form of sintered rectangular bars (10.0 mm × 5.0 mm × 3.0 mm) from the different suppliers and all surface treatments were directly applied to these as received samples without additional machining operations. Those specimens were ultrasonically cleaned in acetone for 10 min and thoroughly dried with compressed air. All specimens of each grade were assigned into four groups of four specimens each, and either were kept as sintered (untreated) or were rough polished using a polishing disk (MD Allegro 250, Struers, Ballerup, Denmark) with a diamond suspension (15 μm water-based diamond suspension, Kemet Europe, Kapellen, Belgium), sandblasted with 50 μm Al ₂ O ₃ particles (Danville Engineering, San Ramon, CA, USA), or tribochemical silica sandblasted with 30 μm Al ₂ O ₃ particles using CoJet (3M ESPE). Details of the surface treatments are summarized in Table 1 . Both the top and bottom surfaces received the same surface treatment. One specimen from each zirconia grade was used for microstructural investigation using scanning electron microscopy (SEM, XL30-FEG, FEI, Eindhoven, Netherlands) employing the following conditions (gold coated, 10 ⁻⁵ mbar pressure, 10 kV energy range, 144 μA beam current, secondary electron image).

Flow chart of the experimental study set-up.

Surface roughness was measured on the three remaining specimens of each grade using an optical interferometer (Wyko NT3300, Veeco, Tuscon, AZ, USA) with 5× magnification. Quantification of the three-dimensional (3D) surface roughness parameter Sa (arithmetic mean deviation) was performed using Vision32 software (Veeco). For each specimen, five regions (effective field of view was 1.210 mm × 0.921 mm) on the surface-treated sides were selected. For surface roughness, a linear mixed-effects model (nlme package, R3.01, R foundation for Statistical Computing, Vienna, Austria) was constructed to assess the influence of the different surface treatments. For this model, the ‘as sintered’ (untreated) condition was excluded, since data largely varied, depending on the sample preparation process conducted by the manufacturers. In this model, two fixed effects, ‘ZIRCONIA GRADE’ and ‘SURFACE TREATMENT’ were considered, as well as their mutual interaction. The specimen measured was considered as a random effect.

Micro-Raman spectroscopy was next performed in order to detect potential residual stress on the surface . Raman spectra (SENTERRA, BrukerOptik, Ettlingen, Germany) were collected using the following conditions: Ar-ion laser with a wavelength of 532 nm, 20 mW power at sample and 100× objective. The spectrum integration time was 20 s with the recorded spectra averaged over three successive measurements. For each specimen, at least 12 measurements were performed using a pinhole aperture of 50 μm. The degree of correlation between the Raman wavenumber of the tetragonal ( t -ZrO ₂ ) band around 147 cm ⁻¹ and the monoclinic ( m -ZrO ₂ ) volume fraction was calculated (R3.01, R Foundation for Statistical Computing) for each grade.

Cu K _α (40 kV, 40 mA) X-ray diffraction (XRD, Seifert 3003 T/T, Seifert, Ahrensburg, Germany) analysis was used for phase identification and calculation of the relative phase content of m -ZrO ₂ and t -ZrO ₂ . Both the top and bottom surfaces of each specimen were analyzed from 20 to 90° 2 θ with a step size of 0.01° for 3 s. The transformed zirconia fraction was defined as the difference in m -ZrO ₂ content between the (partially) degraded and the initially untreated (control) or surface-treated specimens. The volume fraction of m -ZrO ₂ was calculated according to the method of Toraya et al. . Grazing incidence XRD (GIXRD) was conducted to obtain information of the upper 2 μm of the surface with a grazing incident angle ( ω ) set to 1.6709°. The other GIXRD settings were the same as for the θ –2 θ XRD analysis. For both θ –2 θ XRD and GIXRD, the I _t ^{(0 0 2)} / I _t ^{(2 0 0)} intensity ratio was calculated.

LTD tests were performed following the ISO 13356 standard: at 134 °C under a standard H ₂ O pressure of 2 bar up to 40 h. The autoclave was placed in an oil bath to establish an internal temperature of 134 °C, as monitored by a thermocouple inside the autoclave. The amount of m -ZrO ₂ on the exposed sample surfaces was measured by θ –2 θ XRD with the same equipment and under the same experimental conditions as mentioned above. Comparing the m -ZrO ₂ volume fraction on the initially ‘as sintered’ (untreated; control) or surface-treated surfaces and on the hydrothermally tested surfaces allowed to calculate the increased m -ZrO ₂ volume fraction. The degree of correlation between the surface roughness (Sa) and the increase in m -ZrO ₂ volume fraction from XRD was calculated for the five Y-TZP zirconia (Aadva, GC; In-Ceram YZ, VITA; IPS e.max ZirCAD, Ivoclar Vivadent; LAVA Frame and LAVA Plus, 3M ESPE). Changes in m -ZrO ₂ fraction were assessed by a linear mixed-effects model. In this model, two fixed effects, ‘ZIRCONIA GRADE’ and ‘SURFACE TREATMENT’, as well as their interaction were included. The specimen measured was considered as a random effect. All tests were performed at a significance level of α = 0.05 using the statistical software (nlme package, R3.01, R Foundation for Statistical Computing).

Finally, one specimen from each grade was mirror polished and thermally etched at 1350 °C for 20 min in air to reveal the grain boundary network. The average grain size of the zirconia grades was measured using SEM (XL-30 FEG, FEI) employing the following conditions (gold coated, 10 ⁻⁵ mbar pressure, 10 kV energy range, 144 μA beam current, secondary electron image and 50,000× magnification). The grain size of at least 300 zirconia grains were measured according to the linear intercept method .

2

Materials and methods

The study design is schematically explained in Fig. 1 . Fully-sintered zirconia specimens, five so-called ‘Y-TZP’ zirconia (Aadva, GC, Tokyo, Japan; In-Ceram YZ, VITA, Bad Säckingen, Germany; IPS e.max ZirCAD, Ivoclar Vivadent, Schaan, Lichtenstein; LAVA Frame and LAVA Plus, 3M ESPE, Seefeld, Germany), one ceria-stabilized tetragonal zirconia polycrystalline (Ce-TZP)/Alumina (Al ₂ O ₃ ) zirconia (NANOZR, Panasonic, Osaka, Japan), and one Y-TZP/Al ₂ O ₃ zirconia (ZirTough, Kuraray-Noritake Dental, Tokyo, Japan) were provided by the manufacturers. All specimens were obtained in the form of sintered rectangular bars (10.0 mm × 5.0 mm × 3.0 mm) from the different suppliers and all surface treatments were directly applied to these as received samples without additional machining operations. Those specimens were ultrasonically cleaned in acetone for 10 min and thoroughly dried with compressed air. All specimens of each grade were assigned into four groups of four specimens each, and either were kept as sintered (untreated) or were rough polished using a polishing disk (MD Allegro 250, Struers, Ballerup, Denmark) with a diamond suspension (15 μm water-based diamond suspension, Kemet Europe, Kapellen, Belgium), sandblasted with 50 μm Al ₂ O ₃ particles (Danville Engineering, San Ramon, CA, USA), or tribochemical silica sandblasted with 30 μm Al ₂ O ₃ particles using CoJet (3M ESPE). Details of the surface treatments are summarized in Table 1 . Both the top and bottom surfaces received the same surface treatment. One specimen from each zirconia grade was used for microstructural investigation using scanning electron microscopy (SEM, XL30-FEG, FEI, Eindhoven, Netherlands) employing the following conditions (gold coated, 10 ⁻⁵ mbar pressure, 10 kV energy range, 144 μA beam current, secondary electron image).

Flow chart of the experimental study set-up.

Surface roughness was measured on the three remaining specimens of each grade using an optical interferometer (Wyko NT3300, Veeco, Tuscon, AZ, USA) with 5× magnification. Quantification of the three-dimensional (3D) surface roughness parameter Sa (arithmetic mean deviation) was performed using Vision32 software (Veeco). For each specimen, five regions (effective field of view was 1.210 mm × 0.921 mm) on the surface-treated sides were selected. For surface roughness, a linear mixed-effects model (nlme package, R3.01, R foundation for Statistical Computing, Vienna, Austria) was constructed to assess the influence of the different surface treatments. For this model, the ‘as sintered’ (untreated) condition was excluded, since data largely varied, depending on the sample preparation process conducted by the manufacturers. In this model, two fixed effects, ‘ZIRCONIA GRADE’ and ‘SURFACE TREATMENT’ were considered, as well as their mutual interaction. The specimen measured was considered as a random effect.

Micro-Raman spectroscopy was next performed in order to detect potential residual stress on the surface . Raman spectra (SENTERRA, BrukerOptik, Ettlingen, Germany) were collected using the following conditions: Ar-ion laser with a wavelength of 532 nm, 20 mW power at sample and 100× objective. The spectrum integration time was 20 s with the recorded spectra averaged over three successive measurements. For each specimen, at least 12 measurements were performed using a pinhole aperture of 50 μm. The degree of correlation between the Raman wavenumber of the tetragonal ( t -ZrO ₂ ) band around 147 cm ⁻¹ and the monoclinic ( m -ZrO ₂ ) volume fraction was calculated (R3.01, R Foundation for Statistical Computing) for each grade.

Cu K _α (40 kV, 40 mA) X-ray diffraction (XRD, Seifert 3003 T/T, Seifert, Ahrensburg, Germany) analysis was used for phase identification and calculation of the relative phase content of m -ZrO ₂ and t -ZrO ₂ . Both the top and bottom surfaces of each specimen were analyzed from 20 to 90° 2 θ with a step size of 0.01° for 3 s. The transformed zirconia fraction was defined as the difference in m -ZrO ₂ content between the (partially) degraded and the initially untreated (control) or surface-treated specimens. The volume fraction of m -ZrO ₂ was calculated according to the method of Toraya et al. . Grazing incidence XRD (GIXRD) was conducted to obtain information of the upper 2 μm of the surface with a grazing incident angle ( ω ) set to 1.6709°. The other GIXRD settings were the same as for the θ –2 θ XRD analysis. For both θ –2 θ XRD and GIXRD, the I _t ^{(0 0 2)} / I _t ^{(2 0 0)} intensity ratio was calculated.

LTD tests were performed following the ISO 13356 standard: at 134 °C under a standard H ₂ O pressure of 2 bar up to 40 h. The autoclave was placed in an oil bath to establish an internal temperature of 134 °C, as monitored by a thermocouple inside the autoclave. The amount of m -ZrO ₂ on the exposed sample surfaces was measured by θ –2 θ XRD with the same equipment and under the same experimental conditions as mentioned above. Comparing the m -ZrO ₂ volume fraction on the initially ‘as sintered’ (untreated; control) or surface-treated surfaces and on the hydrothermally tested surfaces allowed to calculate the increased m -ZrO ₂ volume fraction. The degree of correlation between the surface roughness (Sa) and the increase in m -ZrO ₂ volume fraction from XRD was calculated for the five Y-TZP zirconia (Aadva, GC; In-Ceram YZ, VITA; IPS e.max ZirCAD, Ivoclar Vivadent; LAVA Frame and LAVA Plus, 3M ESPE). Changes in m -ZrO ₂ fraction were assessed by a linear mixed-effects model. In this model, two fixed effects, ‘ZIRCONIA GRADE’ and ‘SURFACE TREATMENT’, as well as their interaction were included. The specimen measured was considered as a random effect. All tests were performed at a significance level of α = 0.05 using the statistical software (nlme package, R3.01, R Foundation for Statistical Computing).

Finally, one specimen from each grade was mirror polished and thermally etched at 1350 °C for 20 min in air to reveal the grain boundary network. The average grain size of the zirconia grades was measured using SEM (XL-30 FEG, FEI) employing the following conditions (gold coated, 10 ⁻⁵ mbar pressure, 10 kV energy range, 144 μA beam current, secondary electron image and 50,000× magnification). The grain size of at least 300 zirconia grains were measured according to the linear intercept method .

3

Results

SEM photomicrographs of the polished and thermally etched zirconia are shown in Fig. 2 . For NANOZR (Panasonic) and ZirTough (Kuraray Noritake), alumina particles with a darker contrast were distinguished. The results of the ZrO ₂ grain-size measurements are summarized in Fig. 2 . In-Ceram YZ (VITA) revealed a larger grain size than the others.

Representative SEM photomicrographs of polished and thermally etched specimens, and summary of the grain-size distribution. (a) Aadva (GC); (b) In-Ceram YZ (VITA); (c) IPS e.max ZirCAD (Ivoclar Vivadent); (d) LAVA Frame (3M ESPE); (e) LAVA Plus (3M ESPE); (f) NANOZR (Panasonic): dark contrast alumina particles can be observed; (g) ZirTough (Kuraray Noritake): dark contrast alumina particles can be observed.

The results of the Sa surface-roughness parameter are summarized in Table 2 . The linear mixed-effects model revealed increased Sa values for Al ₂ O ₃ sandblasting and tribochemical silica (CoJet) sandblasting by 0.37 μm and 0.28 μm, respectively. LAVA Plus (3M ESPE) had a significantly higher surface roughness in the as sintered (untreated) state and after Al ₂ O ₃ sandblasting than the other grades. Correlation analysis between Sa (μm) and increase in monoclinic volume fraction after 40 h LTD testing revealed only a weak correlation with an r value of 0.32 ( p = 0.013). Representative SEM photomicrographs of as sintered (untreated; control) and surface-treated IPS e.max ZirCAD (Ivoclar Vivadent) are shown in Fig. 3 .

Table 2

Summary of surface-roughness measurements.

Zirconia	As sintered (untreated)		Rough polished		Al 2 O 3 sandblasted		Tribochemical silica sandblasted
	Sa (μm)	SD	Sa (μm)	SD	Sa (μm)	SD	Sa (μm)	SD
Aadva	0.28	0.04	0.10	0.00	0.47	0.03	0.39	0.04
In-Ceram YZ	0.29	0.03	0.10	0.01	0.47	0.03	0.37	0.03
IPS e.max ZirCAD	0.39	0.08	0.10	0.00	0.47	0.02	0.34	0.03
LAVA Frame	0.63	0.29	0.08	0.01	0.60	0.04	0.50	0.19
LAVA Plus	0.56	0.07	0.08	0.00	1.15	0.35	0.56	0.12
NANOZR	0.70	0.19	0.10	0.00	0.50	0.01	0.36	0.04
ZirTough	0.25	0.02	0.09	0.00	0.52	0.05	0.35	0.04

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Slow cooling protocol improves fatigue life of zirconia crowns
2. Cytotoxicity and terminal differentiation of human oral keratinocyte by indium ions from a silver–palladium–gold–indium dental alloy
3. Retrospective evaluation of posterior direct composite restorations: 10-Year findings
4. Influence of ultraviolet photofunctionalization on the surface characteristics of zirconia-based dental implant materials
5. Can quaternary ammonium methacrylates inhibit matrix MMPs and cathepsins?
6. Two regression methods for estimation of a two-parameter Weibull distribution for reliability of dental materials

Stay updated, free dental videos. Join our Telegram channel

Join