Abstract
Objectives
The metastability of the tetragonal crystal structure of yttria partial stabilized zirconia polycrystalline (Y-TZP) ceramics is a basis of concern for dental restorations. Reactions between the porcelain and the Y-TZP framework may result in a reduction of the stability of the zirconia and interface bonding caused by a transformation from tetragonal to monoclinic crystalline structure during veneering.
Methods
XRD 2 micro-diffraction measurements were carried out on tapered veneered cross-sections of the interface area to generate locally resolved information of the phase content in this region. To get a high intensity X-ray beam for short measurement times a focussing polycapillary with a spot size of app. 50 μm was used to evaluate the porcelain zirconia interface.
Results
Under almost all conditions the phase composition of zirconia grains at the interface revealed both the monoclinic and tetragonal structure. These observations indicate that destabilization of the tetragonal phase of zirconia occurs at the interface during veneering with porcelain.
Significance
These results and their relevance to the long-term stability of the interface adhesion between zirconia and veneering porcelain as well as the tetragonal to monoclinic crystal transformations at the interface are discussed.
1
Introduction
Zirconia holds an exclusive place amongst dental restorative materials compared with other oxide ceramics, such as alumina, because of its excellent mechanical properties as a consequence of transformation toughening that was identified in the mid-1970s . Pure zirconia can exist in three different crystal structures depending on temperature. At room temperature up to 1170 °C, the symmetry is monoclinic, the structure is tetragonal between 1170 and 2370 °C and cubic above 2370 °C up to the melting point . The transformation from tetragonal (t) to monoclinic (m) during a cooling process is accompanied by a volume increase (approximately 4%) and shear distortion, sufficient to cause catastrophic failure. Alloying pure zirconia with stabilizing oxides such as Y 2 O 3 allows the preservation of the meta-stable tetragonal structure at room temperature and therefore the potential to enable stress-induced t → m transformation, which can enhance resistance to crack extension leading to higher toughness compared to alumina .
As a consequence of the metastability of tetragonal zirconia, stress-generating surface treatments, such as grinding or sandblasting, are able to trigger the t → m transformation with the associated volume increase leading to the formation of surface compressive stresses, thereby increasing the flexural strength. However such metastability of the material also increases the susceptibility to aging . The low temperature degradation (LTD) of zirconia is a well-documented phenomenon, dependent upon the presence of moisture and modest heat . The consequences of this aging process are multiple and include surface degradation with grain pullout and micro-cracking as well as strength degradation. Although LTD has been shown to be associated with a series of orthopedic hip prostheses failures in 2001 and despite a well established definition of the conditions under which LTD occurs, there is currently no clear relationship between LTD and failure predictability when zirconia is used as a dental bio-ceramic .
3Y-TZP is now widely used in dentistry for the fabrication of dental restorations, mostly processed by machining of partially sintered blanks followed by sintering at high temperature. The mechanical properties of 3Y-TZP strongly depend on its grain size . Above a critical size, Y-TZP is less stable and more vulnerable to spontaneous t → m transformation than smaller grain sizes (<1 μm) . Moreover, below a certain grain size (∼0.2 μm), the stress-induced transformation is not possible, leading to reduced fracture toughness .
Porcelain fused to metal (PFM) restorations developed over decades have placed great attention on the preparation of the interface prior to building up of the porcelain as well as the reactions that promote adhesion. The nature of the interface between Y-TZP and its veneering porcelains however has not been carefully studied. Although diffusion processes are time-dependent, chemical reactions may occur between the two ceramic materials . In a previous paper the authors used high resolution SEM to investigate the surface features of Y-TZP bonded to porcelain and found that wet thick layers of porcelain when sintered generated highly textured Y-TZP grains . The aim of this paper is to use X-ray micro-diffraction to evaluate the interface region between veneering porcelain and Y-TZP prepared using standard dental fabrication techniques.
2
Materials and methods
In this study yttria partially stabilized tetragonal zirconia polycrystalline ceramic (Y-TZP, VITA In-Ceram ® YZ, VITA Zahnfabrik, Germany) is used as a framework material. This material was prepared by sintering blocks, suitable for CAD/CAM machines Cerec ® and InLab ® (Sirona, USA), at a temperature of 1530 °C for 2 h. For the observation of the interface system rectangular plates with dimensions of 10 mm × 10 mm × 1 mm were used. As veneering porcelain, VITA VM ® 9 (colour shade 2M2), prepared in three variations were built up on the zirconia surface using the methods listed in Table 1 .
| No. | Liquid medium | Porcelain | Firing process |
|---|---|---|---|
| 1. | VITA VM liquid | VITA VM9 | Wash-firing (950 °C) |
| 2. | VITA VM liquid | VITA VM9 | Wash-firing thicker layer (950 °C) |
| 3. | No | VITA VM9 | Wash-firing (950 °C) |
Prior to veneering the samples the flat zirconia surface was evaluated by X-ray diffractometry to determine the crystal phases of the sintered zirconia grains. This step ensured that changes at the zirconia interface following porcelain build up were not associated with transformations due to exposure to room temperature humidity prior to porcelain veneering.
To assess the initial wetting of the Y-TZP a so-called Wash-Dentin-firing procedure was used. This involves the firing of a very thin layer of veneering porcelain (approximately 0.05 mm thick) to 950 °C at a heating rate of 45 °C/min with a holding time of 2 min, following a preheating temperature at 500 °C for 6 min in a 70% closed furnace chamber. The same Wash-Dentin firing temperature cycle was used for each preparation method. For all preparation methods, after preheating at the holding period for 2 min, vacuum was applied upon closing of the furnace chamber before heating to the highest temperature. For sample method number 2, a thicker moist layer (approximately 2 mm thick) of veneering porcelain was applied prior to initial heating. To observe only the initial bonding reaction between the two different materials no further firing process was conducted and no other porcelain layer was applied. The liquid application medium for the porcelain was systematically varied from complete absence in sample method 3 of the standard proprietary wetting liquid (VITA VM Modelling Liquid, VITA Zahnfabrik, Germany) while for samples prepared according to method number 1 and 2 this was included. The choice of the standard liquid was based upon observations in the previous study that faceting occurred in the presence of both distilled water and the VITA VM liquid . The furnace used during this study was a VITA Vacumat 4000 (VITA Zahnfabrik, Germany).
To enable a magnified view of the interface, taper sections at approximately 5° to the interface of all samples were cut with a water-cooled diamond saw (600 microns Struers, Germany) and polished with diamond paste (3 μm VITA Zahnfabrik, Germany). The cross-sections of the samples were examined by scanning electron microscopy (LEO 438 VPi Carl Zeiss SMT and LEO 1530 Gemini, field emission microscope) and an XRD 2 micro-diffractometer (BRUKER-D8 DISCOVER with General Area Diffraction Detection System (GADDS) equipped with a specially developed focusing micro-lens to generate a analysing spot of approximately 50 μm diameter (FWHM) ).
Traditionally X-ray diffractometers used for the purpose of local area analysis are typically equipped with pinhole collimators to generate a small analysis spot on the sample surfaces. The use of such pinhole collimators however decreases the intensity of the X-ray beam on the sample dramatically, which results in long measurement times. In the last decade pinhole collimators have been replaced by monocapillary optics and in the last few years by focussing polycapillary micro-lenses in order to focus the generated X-ray beam on the sample. This has increased the localized intensity on the sample by orders of magnitude depending on the X-ray source and type of optics in comparison to a standard pinhole collimator .
The X-ray micro-diffractometer (μ-XRD 2 ) used in this study was a modification of typical powder diffractometer with a focussing micro-lens to achieve a micrometer-sized X-ray beam and a 2-dimensional detector system (BRUKER-HiStar), which covers app. 30° in 2 and app. 30° at the same time . The advantage of such a focussing polycapillary micro-lens instead of the common pinhole collimator or a monocapillary is the short measurement time required, down to a few seconds, due to the high brilliance of the X-ray beam on the sample combined with a spot size currently down to approximately 50 μm diameter FWHM. A general disadvantage of small spot sizes in powder diffraction setups is the potentially poor crystallite statistics in the analyzed volume depending on the crystallite size. In addition and to avoid this disadvantage the additional 2-dimensional HiStar-detector provides direct assessment of texture effects and crystallite size in the analyzed sample.
Before commencing the μ-XRD 2 -analysis of the interface region various sample preparation methods of the taper sections were examined by locally scanning the surface with the micro-diffractometer. This indicated that the cutting and polishing preparation did not generate significant t → m transformation of the zirconia grains on the surface. Monoclinic peaks and broadening of the tetragonal/cubic peak were only observed on the ground surface of the prepared area and not on the polished area in the vicinity of the veneered interface with the relatively low spatial resolution of the X-ray diffraction system used in the former study .
All three preparation procedures were scanned with the micro-diffractometer in the manner shown in Fig. 1 . The veneered part and the polished zirconia surfaces were measured at 8 different positions in steps of 50 μm across the interface with a measurement time of 120 s per pattern using the 50 μm micro-lens and Co Kα (radiation with 30 kV/30 mA setting) with a fixed incident angle of 10° to the flat interface. Due to this incident angle the measurement spot has an elliptical geometry with a length of app. 400 μm and a width of app. 50 μm. Three of the scanning positions are shown as an example for the working area in Fig. 1 where the middle position focuses on the location directly at the interface between Y-TZP and porcelain and the others on the adjacent positions. The schematic on the left hand side vertically through the tapered interface sample shows positions of the spots indicated by the elliptical areas on the right hand side. No etching process such as HF content gel was used to ensure that other potential influences on the framework material were minimized.
2
Materials and methods
In this study yttria partially stabilized tetragonal zirconia polycrystalline ceramic (Y-TZP, VITA In-Ceram ® YZ, VITA Zahnfabrik, Germany) is used as a framework material. This material was prepared by sintering blocks, suitable for CAD/CAM machines Cerec ® and InLab ® (Sirona, USA), at a temperature of 1530 °C for 2 h. For the observation of the interface system rectangular plates with dimensions of 10 mm × 10 mm × 1 mm were used. As veneering porcelain, VITA VM ® 9 (colour shade 2M2), prepared in three variations were built up on the zirconia surface using the methods listed in Table 1 .
| No. | Liquid medium | Porcelain | Firing process |
|---|---|---|---|
| 1. | VITA VM liquid | VITA VM9 | Wash-firing (950 °C) |
| 2. | VITA VM liquid | VITA VM9 | Wash-firing thicker layer (950 °C) |
| 3. | No | VITA VM9 | Wash-firing (950 °C) |
Prior to veneering the samples the flat zirconia surface was evaluated by X-ray diffractometry to determine the crystal phases of the sintered zirconia grains. This step ensured that changes at the zirconia interface following porcelain build up were not associated with transformations due to exposure to room temperature humidity prior to porcelain veneering.
To assess the initial wetting of the Y-TZP a so-called Wash-Dentin-firing procedure was used. This involves the firing of a very thin layer of veneering porcelain (approximately 0.05 mm thick) to 950 °C at a heating rate of 45 °C/min with a holding time of 2 min, following a preheating temperature at 500 °C for 6 min in a 70% closed furnace chamber. The same Wash-Dentin firing temperature cycle was used for each preparation method. For all preparation methods, after preheating at the holding period for 2 min, vacuum was applied upon closing of the furnace chamber before heating to the highest temperature. For sample method number 2, a thicker moist layer (approximately 2 mm thick) of veneering porcelain was applied prior to initial heating. To observe only the initial bonding reaction between the two different materials no further firing process was conducted and no other porcelain layer was applied. The liquid application medium for the porcelain was systematically varied from complete absence in sample method 3 of the standard proprietary wetting liquid (VITA VM Modelling Liquid, VITA Zahnfabrik, Germany) while for samples prepared according to method number 1 and 2 this was included. The choice of the standard liquid was based upon observations in the previous study that faceting occurred in the presence of both distilled water and the VITA VM liquid . The furnace used during this study was a VITA Vacumat 4000 (VITA Zahnfabrik, Germany).
To enable a magnified view of the interface, taper sections at approximately 5° to the interface of all samples were cut with a water-cooled diamond saw (600 microns Struers, Germany) and polished with diamond paste (3 μm VITA Zahnfabrik, Germany). The cross-sections of the samples were examined by scanning electron microscopy (LEO 438 VPi Carl Zeiss SMT and LEO 1530 Gemini, field emission microscope) and an XRD 2 micro-diffractometer (BRUKER-D8 DISCOVER with General Area Diffraction Detection System (GADDS) equipped with a specially developed focusing micro-lens to generate a analysing spot of approximately 50 μm diameter (FWHM) ).
Traditionally X-ray diffractometers used for the purpose of local area analysis are typically equipped with pinhole collimators to generate a small analysis spot on the sample surfaces. The use of such pinhole collimators however decreases the intensity of the X-ray beam on the sample dramatically, which results in long measurement times. In the last decade pinhole collimators have been replaced by monocapillary optics and in the last few years by focussing polycapillary micro-lenses in order to focus the generated X-ray beam on the sample. This has increased the localized intensity on the sample by orders of magnitude depending on the X-ray source and type of optics in comparison to a standard pinhole collimator .
The X-ray micro-diffractometer (μ-XRD 2 ) used in this study was a modification of typical powder diffractometer with a focussing micro-lens to achieve a micrometer-sized X-ray beam and a 2-dimensional detector system (BRUKER-HiStar), which covers app. 30° in 2 and app. 30° at the same time . The advantage of such a focussing polycapillary micro-lens instead of the common pinhole collimator or a monocapillary is the short measurement time required, down to a few seconds, due to the high brilliance of the X-ray beam on the sample combined with a spot size currently down to approximately 50 μm diameter FWHM. A general disadvantage of small spot sizes in powder diffraction setups is the potentially poor crystallite statistics in the analyzed volume depending on the crystallite size. In addition and to avoid this disadvantage the additional 2-dimensional HiStar-detector provides direct assessment of texture effects and crystallite size in the analyzed sample.
Before commencing the μ-XRD 2 -analysis of the interface region various sample preparation methods of the taper sections were examined by locally scanning the surface with the micro-diffractometer. This indicated that the cutting and polishing preparation did not generate significant t → m transformation of the zirconia grains on the surface. Monoclinic peaks and broadening of the tetragonal/cubic peak were only observed on the ground surface of the prepared area and not on the polished area in the vicinity of the veneered interface with the relatively low spatial resolution of the X-ray diffraction system used in the former study .
All three preparation procedures were scanned with the micro-diffractometer in the manner shown in Fig. 1 . The veneered part and the polished zirconia surfaces were measured at 8 different positions in steps of 50 μm across the interface with a measurement time of 120 s per pattern using the 50 μm micro-lens and Co Kα (radiation with 30 kV/30 mA setting) with a fixed incident angle of 10° to the flat interface. Due to this incident angle the measurement spot has an elliptical geometry with a length of app. 400 μm and a width of app. 50 μm. Three of the scanning positions are shown as an example for the working area in Fig. 1 where the middle position focuses on the location directly at the interface between Y-TZP and porcelain and the others on the adjacent positions. The schematic on the left hand side vertically through the tapered interface sample shows positions of the spots indicated by the elliptical areas on the right hand side. No etching process such as HF content gel was used to ensure that other potential influences on the framework material were minimized.
