Improvement in Dental Porcelain Bonding to Milled, Noncast Titanium Surfaces by Gold Sputter Coating Mau-Chin Lin, BSc, MSc, PhD1,2 & Her-Hsiung Huang, BSc, MSc, PhD1,3,4,5,6,7 1

Department of Dentistry, National Yang-Ming University, Taipei, Taiwan Department of Dental Technology and Materials Science, Central Taiwan University of Science and Technology, Taichung, Taiwan 3 Graduate Institute of Basic Medical Science, China Medical University, Taichung, Taiwan 4 Department of Biomedical Informatics, Asia University, Taichung, Taiwan 5 Department of Stomatology, Taipei Veterans General Hospital, Taipei, Taiwan 6 Institute of Oral Biology, National Yang-Ming University, Taipei, Taiwan 7 Department of Medical Research, China Medical University Hospital, Taichung, Taiwan 2

Keywords Gold sputter coating; milled noncast titanium; dental porcelain; bond strength. Correspondence Her-Hsiung Huang, Department of Dentistry, National Yang-Ming University, No. 155, Sec. 2, Li-Nong Street, Taipei 112, Taiwan. E-mail: [email protected] This study was financially supported by the National Science Council (NSC 97–2314-B-166–002-MY3 and NSC 100-2314-B-166-002), Taiwan. The authors deny any conflicts of interest. Accepted October 3, 2013 doi: 10.1111/jopr.12145

Abstract Purpose: This study evaluated the adherence of dental porcelain to a milled, noncast titanium (Ti) surface with a gold sputter coating to evaluate a possible new practical surface treatment for enhancing the bond strength between Ti and porcelain. Materials and Methods: Milled, noncast Ti strips were created by computer-aided design and manufacturing processes. The milled, noncast Ti strips were sandblasted with alumina particles and were then sequentially subjected to gold sputter coating treatments of 150- and 300-second duration. Low-fusion dental porcelain was then sintered onto the surface-treated Ti strips. The bond strengths of the Ti/porcelain specimens were evaluated using a three-point bending test (ISO 9693). Surface characterizations of the specimens were carried out with X-ray photoelectron spectrometry, scanning electron microscopy, and energy dispersive X-ray spectroscopy. Results: The results indicated that the bond strengths of all the Ti/porcelain groups were greater than the minimum requirement (25 MPa) as prescribed by ISO 9693. The gold sputter coating increased the oxidation resistance (or decreased the oxide content) of the Ti surface during porcelain sintering, which positively affected the bond strength of Ti/porcelain (approximately 36 MPa) compared to the untreated Ti/porcelain specimen (approximately 29 MPa). The fracture morphologies of all the Ti/porcelain groups revealed an adhesive bond failure as the interfacial fracture mode between the Ti and the porcelain. Conclusions: A practical and simple sandblasting/gold sputter coating treatment of Ti surfaces prior to porcelain sintering significantly strengthens the bond between the milled, noncast Ti and the dental porcelain.

Titanium (Ti) tends to react with oxygen at high temperatures, resulting in a thick oxide layer, which can adversely affect Ti/porcelain bonding in prosthetic appliances.1-3 Porcelainfused-to-metal (PFM) dental crowns are typically created with a thin interfacial oxide layer to enhance bonding between the porcelain and metal.4,5 Many researchers have created low-fusing dental porcelains suitable for sintering with Ti surfaces with thin oxide layers.6-8 Nevertheless, the bond strength between low-temperature porcelain and Ti is merely slightly greater than the lowest standard of metal/porcelain bond strength given by ISO 9693.9 Related studies have also indicated that Ti/porcelain bonding can be inferior to the bonding between porcelain and other metals.7,10 Therefore, developing methods to improve Ti/porcelain bonding could benefit clinical applications that require durable PFM dental crowns. 540

The bond strength between metal and porcelain is affected by a number of factors, including the chemical composition of materials, treatment of the metal surface, porcelain sintering procedures, and the methods employed for strength analysis.11-13 Bond strength is primarily related to the interfacial mechanical and chemical properties between the porcelain and the metal.14,15 Despite extensive research into the effects of Ti surface treatment on bonding characteristics prior to porcelain sintering procedure,14,16-22 the fundamental mechanism underlying the bonding of porcelain to Ti has yet to be fully elucidated. Recent reports have used gold sputter coating on cast pure Ti to increase the adhesion of porcelain to Ti surfaces.23-26 In these studies, a sputter coater was used to produce a gold coating layer of 0.3 to 2 μm on the cast Ti surface following

C 2014 by the American College of Prosthodontists Journal of Prosthodontics 23 (2014) 540–548 

Lin and Huang

Bonding between Milled Noncast Titanium and Porcelain

Table 1 Surface treatment stages used to process the milled, noncast Ti strips Group SB Au150

Au300

Description of surface treatment procedures Step 1: sandblasting with 125 μm Al2 O3 particles Step 2: ultrasonic cleaning in distilled water for 5 minutes Step 1: sandblasting with 125 μm Al2 O3 particles Step 2: ultrasonic cleaning with distilled water for 5 minutes Step 3: gold sputter coating at 15 mA for 150 seconds Step 1: sandblasting with 125 μm Al2 O3 particles Step 2: ultrasonic cleaning with distilled water for 5 minutes Step 3: gold sputter coating at 15 mA for 300 seconds

previous studies have proposed simple, practical techniques to improve the bond strength between noncast, milled Ti and dental porcelain. In a previous study,40 the authors of this report used similar testing methods to investigate the porcelain bonding to Ti surfaces treated by combining the mechanical sandblasting and different chemical etching processes. In this study, our hypothesis is that gold sputter coating can improve the bond strength between Ti and porcelain. This study aimed to identify a practical and simple combination of mechanical sandblasting and gold sputter coating treatments for increasing the bond strength between porcelain and the CAD/CAM-produced milled, noncast Ti surface.

Materials and methods approximately 15 minutes of coating treatment, and then the adhesion morphology of the cast Ti/porcelain was evaluated qualitatively. These studies suggest that gold sputter coating on cast Ti surfaces is an effective technique for improving adhesion in the Ti/porcelain system; however, no quantitative bond strength was reported. For potential dental applications, a reduction of the gold sputter coating treatment time and quantitative bond strength analysis of the Ti/porcelain material are necessary. One problem associated with the traditional lost-wax method of dental cast Ti is the formation of a thick, hardened alphacase layer on the Ti, which can vary from ten to a few hundred microns.27,28 The alpha-case layer is considerably harder than the inner bulk structure. This tends to impede the grinding and polishing procedures required to process cast Ti prostheses. In addition, this layer is susceptible to the initiation and propagation of cracks.29 Computer-aided design (CAD) and computer-aided manufacturing (CAM) techniques have recently been incorporated into the manufacture of high-quality Ti prosthetic appliances.30-32 When CAD/CAM milling procedures are used to fabricate dental copings and crowns, no thick oxide layer forms on the Ti surface. This results in a more homogeneous surface structure.33 Furthermore, Ti surfaces milled using these techniques exhibit stronger Ti/porcelain bonding than the surfaces of conventionally cast Ti.34-36 These results have fueled the adoption of this approach in modern dentistry. Most previous research has focused on the bonding characteristics between porcelain and milled, noncast Ti surfaces without surface treatment.37-39 Few reports have discussed how surface treatments affect the bonding characteristics between porcelain and CAD/CAM-milled Ti surfaces. In addition, no

Specimen preparation

Specimen preparation before gold sputter coating treatments was conducted in the same way as our previous report,40 as described below. Commercially available noncast pure Ti (grade 2) strips were used as test specimens. Strips were mechanically milled following a commercial CAD/CAM procedure (DATRON D5; Datron Dynamics, Inc., Milford, NH). The final Ti test strips had dimensions of 25.0 ± 1 × 3.0 ± 0.1 × 0.5 ± 0.05 mm3 , as measured using digital calipers. The milled, noncast Ti surfaces exhibited no evidence of the thick, hardened alpha-case layer that normally forms on cast Ti surfaces. The surface roughness (Ra ) of the milled, noncast Ti strip was approximately 0.35 μm. Thirty Ti strips were divided into three groups of ten specimens. The control group, denoted SB, was sandblasted with 125-μm Al2 O3 particles at a pressure of four bars for 10 seconds in a sandblaster (BL-22; Fu Ming Co., Taichung, Taiwan). The remaining two groups were both experimental: ten SB strips were further gold sputtered at 15 mA for 150 seconds (Au150 group), and the final ten SB strips were further gold sputtered at 15 mA for 300 seconds (Au300 group). An ion sputter coater (E-1010; Hitachi, Tokyo, Japan) with a pure gold target was used to perform the coating process. The detailed procedures used for the Ti strip surface treatments are listed in Table 1. In accordance with ISO 9693, a low-fusing porcelain (Titankeramik; Vita Zahnfabrik, Bad S¨ackingen, Germany) plate (8.0 ± 0.1 × 3.0 ± 0.1 × 1.0 ± 0.1 mm3 ) was sintered to the center of each Ti strip. In accordance with the manufacturer-recommended sintering conditions (Table 2), we employed the same porcelain sintering procedure as used in our previous report.40 An opaque porcelain with a thickness of

Table 2 Porcelain sintering conditions recommended by the manufacturer Procedure (1) Bonder (2) Opaque (3) First dentin (4) Second dentin (5) Glazing

Initial temp. (°C)

Temp. rising rate (°C /min)

Final temp. (°C)

Holding time (min)

Vacuum values (bar)

400 400 400 400 400

60 110 50 50 100

800 790 770 770 770

1 1 1 1 1

6 4 8 8 4

C 2014 by the American College of Prosthodontists Journal of Prosthodontics 23 (2014) 540–548 

541

Bonding between Milled Noncast Titanium and Porcelain

Lin and Huang

Figure 2 X-ray photoelectron spectroscopy (XPS) depth profile analyses, in terms of the atomic concentration of gold by depth, of the milled, noncast Ti strips after the gold sputter coating treatments (Au150 and Au300 groups; n = 10).

Bond strength of the Ti/porcelain specimens

Bond strength of Ti/porcelain specimens was measured using the same methods outlined in our previous report.40 In brief, a universal test machine (EZ-L; Shimadzu, Tokyo, Japan) was used for three-point bending tests of Ti/porcelain strip specimens in accordance with ISO 9693.9 Throughout the bending test, the porcelain side of the specimens was facing downwards, and a downward load was applied to the center of the Ti side at a rate of 0.5 mm/min. In situ load versus displacement curves were recorded throughout the bending tests. We also recorded the fracture load (or debonding load) of the specimens as the point at which a debonding crack first appeared at one end of the porcelain layer. Bond strength of the Ti/porcelain specimens was then calculated using the following equation:9 Bond strength = F × k

Figure 1 (A) Scanning electron microscopy (SEM) micrographs and (B) surface roughnesses, Ra , of the milled, noncast Ti strips with and without gold sputter coating treatments (SB, Au150, and Au300) described in Table 1 (error bar represents standard deviation; n = 10).

approximately 0.2 mm was sintered onto the Ti strips. Then, a dentin porcelain was sintered onto the surfaces. The total thickness of all porcelain sintered onto the Ti strip was approximately 1.1 mm. The opaque and dentin porcelain layers were each applied with a different custom-made jig, which adjusted the dimension of each. Subsequently, a glazing treatment was applied to complete the final procedure. The sintering processes were carried out under vacuum. 542

where F is the fracture load (N), and coefficient k is a function of the thickness and elastic modulus of the Ti strip, in accordance with ISO 9693. The elastic modulus of the Ti strip, as determined from the bending test data, was calculated as follows:41 E = L3 P/4bh3 d where E is the elastic modulus of the Ti strip; L is the distance between the two supporting positions (20.0 mm); b is the width of the Ti strip (3.0 mm); h is the thickness of the Ti strip (0.5 mm); and P and d are the load (N) and displacement (mm) increments, respectively. Surface characterization

Surface characterization of the test specimens was performed using methods similar to those employed in our previous report.40 In brief, after subjecting the Ti strips to various surface

C 2014 by the American College of Prosthodontists Journal of Prosthodontics 23 (2014) 540–548 

Lin and Huang

Bonding between Milled Noncast Titanium and Porcelain

Oxidation resistance of the gold-coated Ti surfaces during the porcelain sintering simulation

To analyze the effect of the gold sputter coating on the oxidation of the Ti surface during porcelain sintering, the test Ti specimens (SB, Au150, Au300) underwent heat treatment at 780°C for 1 minute at a heating rate of 50°C/min under vacuum, followed by furnace cooling to room temperature. Afterwards, XPS was used to analyze the surface oxygen content of the sintering-simulated Ti strips. Statistical analysis

We used one-way ANOVA to determine how a gold sputter coating on the Ti surface affected the surface roughness and Ti/porcelain bond strength. A p value of < 0.05 was considered statistically significant for all tests. Tukey’s test was used as a post hoc test. Ten specimens were included in each test group.

Results

Figure 3 X-ray photoelectron spectroscopy (XPS) spectra of O1s on the milled, noncast Ti strips, both with and without the gold sputter coating treatments, (A) before and (B) after porcelain sintering simulation heat treatment.

treatments, X-ray photoelectron spectroscopy (XPS) was used to analyze the surface chemical composition prior to porcelain sintering. Scanning electron microscopy (SEM) was used to observe the surface morphology of the Ti strips before porcelain sintering. We also used SEM to observe a cross-section of the interface between the porcelain and the Ti following the bending test. The surface arithmetic mean roughness Ra of the Ti strips was then measured using a surface profilometer following surface treatment. Following completion of the three-point bending tests, we analyzed the fractured side surfaces of the Ti and the porcelain using SEM and energy-dispersive X-ray spectroscopy (EDS). This allowed us to further investigate the surface morphology and chemical composition of the specimens, respectively.

Table 1 shows the surface treatment stages used to process the milled, noncast Ti strips. The corresponding surface morphology and surface roughness data are presented in Figures 1A and 1B, respectively. Figure 1A shows a rough, irregular surface morphology on the sandblasted Ti surface (SB group), with a surface roughness (Ra ) of 1.60 ± 0.02 μm. After gold sputter coating, the surface roughness of the sandblasted Ti specimens was reduced slightly: Ra = 1.55 ± 0.03 μm for the Au150 specimens and Ra = 1.44 ± 0.02 μm for the Au300 specimens. There was a slight difference in Ra between the two gold-coated test groups (p < 0.05). Figure 2 shows the XPS depth profile analyses, in terms of the atomic concentration of gold by depth, of the milled, noncast Ti strips that had received the gold sputter coating treatments (Au150 and Au300 groups). The XPS results indicate that the thicknesses of the gold coatings of the Au150 and Au300 groups were approximately 20 and 40 nm, respectively. Figure 3 shows the XPS spectra of O1s on the milled, noncast Ti strips, both with and without the gold sputter coating treatments, (A) before and (B) after porcelain sintering simulation heat treatment. Prior to heat treatment, the gold-coated Ti strips show negligible O1s intensities on the outermost surface, while a significant O1s intensity can be observed for the untreated Ti strip (SB group). After the porcelain sintering heat treatment, the O1s intensity of the gold-coated Ti strips increased, but was still significantly less than that of the SB group. Figure 4A shows SEM micrographs of an interfacial crosssection of the milled, noncast Ti-porcelain specimens, both with and without the gold sputter coating treatments, following the three-point bending test; higher magnifications of Figure 4A are shown in Figure 4B. In each case, fracture occurred near the Ti/porcelain interface. The fracture type of the Ti/porcelain specimens through the bending test was a failure of the adhesive bond.

C 2014 by the American College of Prosthodontists Journal of Prosthodontics 23 (2014) 540–548 

543

Bonding between Milled Noncast Titanium and Porcelain

Lin and Huang

Table 3 Means and standard deviations of bond strengths of the Ti/porcelain strips obtained from the three-point bending tests Group

Mean ± SD (MPa)

SB Au150 Au300

28.76 ± 1.76a 35.68 ± 1.52b 36.55 ± 1.43b

Mean values with different superscript letters are significantly different (p < 0.001).

Table 3 shows the Ti/porcelain bond strengths of the three test Ti/porcelain groups obtained from the three-point bending tests according to ISO 9693 specification. The result of oneway ANOVA suggests that the use of gold sputter on the sandblasted Ti surface prior to porcelain sintering had a significant influence on the Ti/porcelain bond strength (p < 0.001). The Au150 and Au300 groups exhibited bond strengths of 35.68 and 36.55 MPa, respectively; the SB group bond strength was 28.76 MPa. Tukey’s post hoc testing showed no significant differences between the Au150 and Au300 groups. Figure 5 shows an SEM micrograph and the EDS spectra of the Ti side of the fracture surface of a specimen from the Au300 group following the three-point bending test, and Figure 6 shows the porcelain side of the same specimen. The EDS spectra demonstrated the presence of Si, Al, Au, and Ti on both the Ti and porcelain sides of the fracture surface. We found the same result for the other two test groups. Thus, for all test groups, the Ti side of the fracture surface had some remaining porcelain, and the porcelain side had some remaining metals after the bending tests.

Discussion Our experimental results demonstrate that the use of a gold sputter coating treatment on the sandblasted milled, noncast Ti surfaces can increase the bond strength between the Ti and porcelain phases prior to porcelain sintering (p < 0.001). Regardless of the gold sputter coating time, the bond strengths of all test Ti/porcelain groups were approximately 1.5 times the minimum requirement (25 MPa) given by ISO 9693.9 Although the bond strength of the noncoated specimen (SB group) was acceptable according to ISO 9693, the significant enhancement in the bond strength of the gold-coated specimens (Au150 and Au300 groups) is expected to be of use in dental PFM crowns developed for durable clinical applications. Previous studies have reported that the rough cast Ti surface is a significant feature for increasing the bonding of Ti to porcelain.7,18,42 Different Ti surface treatments create different surface roughness values, and these have an effect on the mechanical bonding characteristics between Ti and porcelain. In this study, the SEM observations showed that the use of mechanical sandblasting creates irregular and rough Ti surfaces (Fig 1A). The use of a gold sputter coating treatment only slightly smoothed the irregular and sharp angles of the sandblasted Ti surfaces (Fig 1B); however, the bond strength of the Au150 and Au300 groups was significantly increased compared to that of the SB control group. This implies that the surface chemical properties of the milled, noncast Ti play a role in determining the bond strength between the Ti and the porcelain. The use of gold sputter coating did not significantly alter the Ti surface roughness (Fig 1B); however, it did increase the oxidation resistance (or decreased the oxide content) of the milled, noncast Ti surface during porcelain sintering (Fig 3), which positively affected the Ti/porcelain bond strength (Table 3).

Figure 4 (A) Scanning electron microscopy (SEM) micrographs of an interfacial cross-section of the milled, noncast Ti/porcelain specimens (TM: Ti metal, OP: opaque porcelain, DP: dentine porcelain), both with and without the gold sputter coating treatments, following the three-point bending test; higher magnifications of (A) are shown in (B).

544

C 2014 by the American College of Prosthodontists Journal of Prosthodontics 23 (2014) 540–548 

Lin and Huang

Bonding between Milled Noncast Titanium and Porcelain

Figure 5 (A) Scanning electron microscopy (SEM) micrograph of the Ti side of the fracture surface of a specimen from the Au300 group following the three-point bending test; (B) and (C) represent the energydispersive X-ray spectroscopy (EDS) spectra of the regions indicated by the dotted and solid boxes in (A), respectively.

Figure 6 (A) Scanning electron microscopy (SEM) micrograph of the porcelain side of the fracture surface of a specimen from the Au300 group following the three-point bending test; (B) and (C) represent the energy-dispersive X-ray spectroscopy (EDS) spectra of the regions indicated by the dotted and solid boxes in (A), respectively.

This finding indicates that the thin gold coating layer, which is only a few tens of nanometers thick (Fig 2), is nonetheless sufficient to significantly hinder oxidation and suppress the formation of a thick Ti oxide layer, resulting in an increased Ti/porcelain bond strength after the porcelain sintering treatment. A thinner oxide layer on the Ti surface could increase Ti/porcelain bonding,2,23 which was believed to be attributable to an increase in the mechanical interlocking and chemical reaction at the Ti/porcelain interface. This was shown by the ad-

hesive fracture morphology between Ti and porcelain, namely residual porcelain on the Ti surface and residual metal on the porcelain surface, after the bending test (Figs 5 and 6). The adhesive fracture morphology, ascribed to the thin gold coating on the Ti surface, between the Ti and porcelain, represented a good bonding index. On the Ti side of the fractured specimens in the Au300 group, the EDS spectra revealed residual Si and Al derived from the sintered porcelain side (Figs 5 and 6). Similarly, the porcelain side of the fractured specimens presented Ti derived

C 2014 by the American College of Prosthodontists Journal of Prosthodontics 23 (2014) 540–548 

545

Bonding between Milled Noncast Titanium and Porcelain

Lin and Huang

Table 4 Comparison of the results of Ti/porcelain bond strength obtained in this research with other studies using different surface treatments

Reference 7 14 16 18 21 43 44 45 45 46 47 This research

Surface treatment Sandblasting Hydrochloric acid by etching Silicon coated by sol-gel NaOH + HNO3 solution by immersion SnOx coating by sol-gel SiO2 coating by sol-gel Micro-arc oxidation Niobium nitride (NbN) by sputtering Zirconium nitride (ZrN) by sputtering ZrN coating by sputtering Sandblasting + acid etching Gold sputter coating

Bond strength (MPa) according to ISO 9693 26.7 24.65 37.77 34.8

± ± ± ±

4.1 4.61 0.78 2.7

42 38 46.46 43.1

± ± ± ±

1.48 1.27 4.35 0.59

52.4 ± 0.80 45.99 ± 0.65 34.60 ± 1.79 36.55 ± 1.43

from the Ti side. Residual Au from the gold sputtering was also identified on both the Ti and porcelain sides of specimens in the Au300 group. Similar results were found for the SB and Au150 test specimens, suggesting that the fractures occurring in Ti/porcelain specimens during the bending test resulted from adhesive bond failure. As mentioned earlier, the bond strengths of all the Ti/porcelain groups were greater than the minimum requirement (25 MPa) prescribed by ISO 9693. This finding indicates that the interface between the milled, noncast Ti and porcelain has sufficient bonding character. Wang and Fung2 noted that a metallic oxide layer measuring approximately 5 to 7 μm thick was produced on the Ti and could not well adhere to the Ti surface; therefore, the bonding failure of Ti/porcelain happens within the oxide layer on the Ti surface. This thick oxide layer produced on the Ti surface at porcelain sintering temperature is porous and not well adherent, and therefore makes Ti unsuitable for porcelain bonding. In this study, the sintering temperature of the low-fusing porcelain was less than 800°C, which reduced the thickness of the oxide layer formed on the Ti surface through porcelain sintering.8,43 Furthermore, the milled, noncast Ti surface was roughened by the sandblasting treatment, and the gold sputter coating treatment increased the Ti surface resistance to oxidation during the porcelain sintering treatment. These factors significantly improved the bonding properties between the milled, noncast Ti and the porcelain for the Au150 and Au300 groups. The fracture mode that occurred between the low-fusing porcelain and all the test Ti strips showed an adhesive type of fracture. The bond strengths of the gold-coated Ti strips were approximately 1.5 times greater than the minimum value (25 MPa) specified by ISO 9693. Metallographic grinding and polishing, rather than sandblasting with Al2 O3 particles, can generally be used for the preparation of Ti surfaces prior to porcelain sintering, thereby 546

avoiding surface contamination of the Ti surface by the Al2 O3 particles.8 However, sandblasting with Al2 O3 is usually employed as the first step in the treatment of Ti surfaces before porcelain sintering. Our EDS analysis results indicated that no contamination by the Al2 O3 particles was present on the sandblasted Ti surface (SB group), which also demonstrated sufficient bonding with the porcelain (bond strength 28.76 MPa). This bond strength value was similar to previously reported values.43,44 To increase the bond strength of Ti-porcelain prosthetic appliances, dental technicians should first sandblast the Ti surface. Next, a gold layer should be applied by sputter coating, all prior to the porcelain sintering process. These processes are simple and quick. Although earlier reports have suggested that the ceramic coatings (e.g., oxides.44,45 and nitrides20,46 ) on Ti surfaces considerably increase the cast Ti/porcelain bond strength (>38 MPa), some investigations have revealed that surface modifications (e.g., machining,19 and oxide coating17,21,47 ) of cast Ti surfaces only slightly improve the Ti/porcelain bond strength (

Improvement in dental porcelain bonding to milled, noncast titanium surfaces by gold sputter coating.

This study evaluated the adherence of dental porcelain to a milled, noncast titanium (Ti) surface with a gold sputter coating to evaluate a possible n...
899KB Sizes 0 Downloads 3 Views