Phase transformation of dental zirconia following artificial aging Thomas J. Lucas,1,2 Nathaniel C. Lawson,1 Gregg M. Janowski,2 John O. Burgess1 1 2

Department of Biomaterials, School of Dentistry, University of Alabama at Birmingham, Birmingham, Alabama Department of Materials Science and Engineering, University of Alabama at Birmingham, Birmingham, Alabama

Received 24 July 2014; revised 12 October 2014; accepted 26 October 2014 Published online 1 December 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.33334 Abstract: Aims: Low-temperature degradation (LTD) of yttriastabilized zirconia can produce increased surface roughness with a concomitant decrease in strength. This study determined the effectiveness of artificial aging (prolonged boiling/ autoclaving) to induce LTD of Y-TZP (yttria-tetragonal zirconiapolycrystals) and used artificial aging for transformation depth progression analyses. The null hypothesis is aging techniques tested produce the same amount of transformation, transformation is not time/temperature dependent and LTD causes a constant transformation throughout the Y-TZP samples. Methods: Dental-grade Y-TZP samples were randomly divided into nine subgroups (n 5 5): as received, 3.5 and 7 day boiling, 1 bar autoclave (1, 3, 5 h), and 2 bar autoclave (1, 3, 5 h). A 4-h boil treatment (n 5 2) was performed post-experiment for completion of data. Transformation was measured using traditional X-ray diffraction and low-angle X-ray diffraction.

Results: The fraction of t ! m transformation increased with aging time. The 3.5 day boil and 2 bar 5 h autoclave produced similar transformation results, while the 7 day boiling treatment revealed the greatest transformation. The surface layer of the aged specimen underwent the most transformation while all samples displayed decreasing transformation with depth. Significance: Surface transformation was evident, which can lead to rougher surfaces and increased wear of opposing dentition/materials. Therefore, wear studies addressing LTD C 2014 Wiley of Y-TZP are needed utilizing accelerated aging. V Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater, 103B: 1519–1523, 2015.

Key Words: zirconia, X-ray diffraction, phase transformation, yttria stabilized

How to cite this article: Lucas TJ, Lawson NC, Janowski GM, Burgess JO. 2015. Phase transformation of dental zirconia following artificial aging. J Biomed Mater Res Part B 2015:103B:1519–1523.

INTRODUCTION

Zirconia is on the forefront of prosthetic dental research due to its superior mechanical properties. It is a multiphasic ceramic, which can exist in three temperature-dependent crystallographic structures: monoclinic, tetragonal, and cubic. The addition of a stabilizing oxide such as ytrria (Y2O3) allows zirconia to maintain a tetragonal structure at room temperature, despite the monoclinic phase being thermodynamically favored.1 However, ytrria-stabilized zirconia [yttria-tetragonal zirconia-polycrystals (Y-TZP)] can also undergo the t ! m transformation without stress under certain conditions, which is known as low temperature degradation (LTD). LTD or aging causes yttria-stabilized zirconia in a metastable tetragonal structure to transform into a stable monoclinic structure. A summary of the zirconia LTD literature is2: Degradation proceeds most rapidly in the temperature range of 200–300 C and is time dependent, and degradation causes the tetragonal phase to transform to monoclinic and is accompanied by micro- and macro-cracking. The transformation originates at the surface and progresses to the interior. Water or water vapor contributes to the transformation. A

decrease in grain size and an increase in the stabilizing oxide retard transformation. The t ! m transformation causes a volume increase of approximately 14% and increases surface roughness, which is of particular concern in dentistry.3,4 Conflicting results for mechanical property degradation after LTD have been reported. Some studies report that material properties such as flexural strength are not affected after LTD while another study stated that the Young’s modulus and hardness was reduced by 30% after LTD.5–8 The volume expansion in stabilized zirconia due to the tetragonal to monoclinic phase transformation has advantages and drawbacks.2 The metastable tetragonal structure transforms to the monoclinic structure under stress, which results in a localized volumetric expansion that can halt a propagating crack and ultimately increase toughness by applying pressure at the crack tip.2,9 However, the resulting strain on the surrounding system causes micro-cracks by expanding the surface area, leading to a rougher zirconia surface.10 The micro-cracks may also lead to catastrophic failure if they continue and progress inward.10 Surface roughening, increased wear of opposing surfaces, and decreased surface hardness led to fractures in some TZP hip

Additional Supporting Information may be found in the online version of this article. Correspondence to: N. Lawson; e-mail: [email protected] Contract grant sponsor: National Institute of Dental Research; contract grant number: NIDCR T-90DE022736

C 2014 WILEY PERIODICALS, INC. V

1519

implant recalls.9 Further evidence of LTD, surface degradation, and wear was observed on opposing surfaces of ZrO2 hip implants.10 A magnified effect is expected in the more extreme oral environment due to acid/base fluctuations, water temperature fluctuations and constant cyclic stresses. Several methods of accelerating LTD of zirconia relative to the oral environment have been advanced. The most widely accepted zirconia aging mechanism is cited in ISO standard 13356:2008, which states that autoclave aging under 2 bar of pressure for 5 h should produce less than 25% monoclinic phase for the zirconia to be considered suitable for biomedical purposes.11 While autoclaving is considered in the standard, other methods such as prolonged boiling and storage in solution have also been presented.6,7 One of the ways researchers try to study the longterm effects of LTD is by correlating artificial aging with in vivo aging. Correlations between in vivo LTD and artificial LTD are undergoing scrutiny for the actual likeness of events in which they occur. For example, some studies report a correlation of 3–4 years in vivo for every 1 h of autoclave aging.1,12 On the other hand, zirconia implants extracted after 60–62 months had 1.5–3% monoclinic fraction which correlates to 1 h of autoclave aging.13 Both correlations are drawn from hip implant retrievals which may not be relevant in dentistry where restorations and implants may be subjected to temperature, load, and pH cycling. Furthermore, the monoclinic transformation due to artificial aging does not continue in a linear fashion, but rather follows a sigmoidal curvature, which can be modeled based on the Johnson–Mehl–Avrami (JMA) law and nucleation and growth characteristics.2,11,14 While it has been reported that aging occurs more rapidly in the presence of water and high temperatures, a comparison of methods to determine the role of artificial aging on inducing the transformation has not been presented. In the current study, simulations of the transformation process were investigated by measuring the transformation produced by different methods of accelerated aging and each method analyzed to determine the relative amount of t ! m transformation that occurs. Moreover, the depth progression of the transformation, a function of artificial aging time and treatment was measured using X-ray diffraction. It is expected to find that the amounts of transformation are not constant as a function of depth, and, thus, the aging treatment shows a diminishing affect on the amount of transformation further from the surface. If the transformation was substantial at the surface then the wear affects due to volumetric expansions will need to be addressed. The null hypothesis is the aging techniques tested produce the same amount of transformation, the transformation is not time/temperature dependent and LTD causes a constant transformation throughout the Y-TZP samples, indicating further testing to determine potential surface roughness due to volume expansion.

MATERIALS AND METHODS

Three methods to induce transformation of yttria-stabilized zirconia by artificial accelerated aging were investigated:

1520

LUCAS ET AL.

prolonged boiling, 1 bar autoclaving, and 2 bar autoclaving. Each test method had different exposure times to observe the different behaviors of the TZP under each test. Forty-seven mid-sized grain Y-TZP samples were supplied by 3M ESPE (St Paul, MN) and randomly divided into nine different groups (n 5 5): as-received (control), boiled (3.5 and 7 days), 1 bar autoclave (1, 3, and 5 h), and 2 bar autoclave (1, 3 and 5 h). A 40-h boil group was added for completion of data but was excluded from statistical analysis due to unequal sample population (n 5 2). Prolonged boiling was achieved by utilizing a soxhlet apparatus set up to recycle the condensate.6 The zirconia samples were placed into a glass beaker containing distilled water and heated to boiling. The autoclave-aging test was completed using a dental sterilizer, Midmark M11 Ultraclave Automatic Sterilizer (Midmark Corporation, Versailles, OH). A custom cycle was chosen that consisted of a sterilization temperature of 121 C, and 1 bar pressure, or 14.5 psi, for 60 min. The cycle was repeated to include 3 and 5 h aging treatments. Following these treatments the autoclave cycle was changed to maintain a temperature of 134 C and 2 bar pressure, for 60 min. This cycle was then repeated to achieve 3 and 5 h aging treatments. X-ray diffraction (XRD) was accomplished using a Siemens D 500 X-ray diffraction device (Siemens Corporation, Washington, D.C.) with Cu K alpha radiation. Scans were performed at 40 kV, 30 mA, 0.02 degrees/step from 25 to 33 , and a dwell time of 12 s per step. The monoclinic phase can be calculated with the following equation using the area of intensity for the monoclinic peaks over the total area of intensity for the monoclinic and tetragonal peaks15: Xm 5

 Im ð111Þ1Im ð111Þ  Im ð111Þ1Im ð111Þ1It ð101Þ

(1)

X-ray diffraction was utilized since it is the most common technique for quantitatively determining the fraction monoclinic. Many papers, such as Chevalier’s classic paper, use XRD and optical interferometry.16 Atomic force microscopy (AFM) imaging is very illustrative, but the large number of samples and areas that must be examined to obtain valid quantitative data made it prohibitive. The area that each XRD scan samples and the measurement reproducibility made it the method-of-choice. Statistical analysis was performed using one-way analysis of variance (ANOVA) and Tukey–Kramer (p < 0.05) on each aging treatment to determine significant differences in percent transformation. Further analysis was conducted on two samples (n 5 2) from each boiling and 2 bar autoclave groups to observe the depth progression of monoclinic transformation. XRD scans were performed at 40 kV, 30 mA, 0.02 /step from 27 to 33 , and a dwell time of 12 s per step. Seven different settings for the X-ray incident angle were used to analyze the sample: 1, 2, 3, 5, 8, 10, and varied (h/2h) with the varied setting ran first and last. Previously, it was unknown whether XRD testing methods would cause additional phase

PHASE TRANSFORMATION OF DENTAL ZIRCONIA

ORIGINAL RESEARCH REPORT

TABLE 2. Correlations Between Aging Treatments. Aging Treatment 7 day boil 3.5 day boil 2 bar/5 h 40 h boil 2 bar/3 h 1 bar/5 h 1 bar/3 h 2 bar/1 h 1 bar/1 h As-received

FIGURE 1. Graphical representation of relative X-ray penetration intensity as a function of depth from the surface of partially stabilized zirconia using differing incident angles. As the incident angle increases, the penetration ability of the X-rays increases.

transformations in zirconia samples. Therefore, the varied (h/2h) angle was rescanned to verify that the pattern observed is still indicative of the original aged specimen and to ensure that subsequent X-ray trials do not interfere with observed patterns and cause further transformations due to testing procedures. Figure 1 shows the relative X-ray penetration intensity as a function of depth from the surface. The percent monoclinic transformation was determined for each XRD setting and correlated as a function of depth. RESULTS

Table 1 lists the average transformation and standard deviation produced by each aging treatment as well as the average monoclinic phase present in the as-received samples. The 7 day boiling treatment produced the greatest monoclinic transformation—31.7%, followed by the 3.5-day boil (19.67%) and the 5 h 2 bar autoclave (18.13%). Autoclaving at 1 bar of pressure proved to be the least effective aging test. A maximum of 9.27% monoclinic transformation was observed after aging for 5 h at 121 C under 1 bar pressure. Table 2 shows significant intergroup differences for the aging treatments. The aging treatments that showed similar results were: 3.5 day boil and 5 h autoclave at 2 bar, 40 h boil and 3 h 2 bar autoclave, and the 1 h autoclave (both 1 bar and 2 bar) and the as-received group.

A B B C* C D E F F F

Figure 2 depicts the amount of transformation versus log aging time. The graphical representation of aging time versus monoclinic percentage shows that an initial nucleation period is needed before transformation begins to take place. Each aging group experienced what appears to be an exponential acceleration in transformation after the nucleation period. The current study chose to analyze the samples of the higher transformation groups to determine possible surface alterations. X-ray diffraction methods were employed to determine the amount of transformation as a function of increasing depth. The procedure called for duplication of the typical X-ray diffraction scans used at the end of the analysis to ensure that diffraction testing did not influence the t ! m transformation. While there are small changes in the before and after diffraction data, the after (or rerun) amounts are not consistently higher or less than before. Therefore, it is determined that the X-ray diffraction data collected did not influence the amount of observed transformation. Table 3 lists the average amount of transformation for each aging group at each level of X-ray diffraction. The corresponding depth for each X-ray incident angle is also shown. Autoclave aging for 3–5 h produces approximately 52% monoclinic phase at the surface while prolonged boiling for 3–7 days produces approximately 58% monoclinic phase at the surface. Supporting Information Figure 3(a,b) display the amount of monoclinic phase present for each aging group versus X-ray depth penetration graphically. Higher levels of monoclinic phase are present nearer the

TABLE 1. Mean Fraction of Monoclinic Phase Sample Group As-received 40-h Boil 3.5-day boil 7-day boil 1-h Autoclave 3-h Autoclave 5-h Autoclave 1-h Autoclave 3-h Autoclave 5-h Autoclave

–1 –1 –1 –2 –2 –2

Fraction Monoclinic – (%)

bar bar bar bar bar bar

0.73 13.3 19.7 31.7 2.13 6.30 9.27 2.92 12.6 18.1

(0.1) (1.6) (0.4) (1.8) (1.1) (1.1) (1.1) (0.5) (1.9) (1.6)

FIGURE 2. Percent monoclinic transformation as a function of log aging time for the three different aging treatments is represented graphically.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2015 VOL 103B, ISSUE 7

1521

TABLE 3. Amount of Monoclinic Phase Present for Each Sample Group and the Corresponding X-Ray Incident Angle and Penetration Depth. Incident Angle

X-ray Penetration Depth (lm)

2 Bar 1 Hour

6.05 0.85 1.65 2.35 3.55 4.95 5.55 6.05

3.10 (0.7) 17.1 (0.3) 13.0 (2.3) 13.7 (4.2) 9.40 (5.1) 7.20 (4.1) 5.60 (2.3) 5.1(1.1)

Theta 2 Theta 1 2 3 5 8 10 Theta 2 Theta rerun

2 Bar 3 Hour 15.2 51.6 40.1 35.5 31.1 18.9 17.4 15.0

(1.3) (2.3) (2.7) (5.4) (7.9) (0.1) (0.1) (0.2)

surface and diminishing effects of aging are seen with inward progression for all sample groups. Therefore, when the depth of the analyses is increased the amount of transformation decreased.

DISCUSSION

Similarities of percent transformations across aging groups can be seen. Considering previously stated limitations regarding in vivo correlations, the transformation occurring with each hour of 2 bar autoclaving is roughly equivalent to 3–4 years in vivo, therefore statistically similar treatments of 3.5-days boiling and 5-h of 2 bar autoclave represent 15–20 years in vivo and produce a 20% monoclinic transformation. It is important to emphasize that these transformations occurred under a controlled environment with a constant, high temperature without cyclic stress and acid/base fluctuations present in the oral environment. Correlation data for extracted in vivo zirconia dental restorations is lacking, resulting in inexact correlations, at best. However, the high temperature alone, compared to the oral environment, produced significant transformation. A sigmoidal curve, described by the JMA law, is not immediately evident due to the limited number of data sets used in this study. Furthermore, large numbers of data sets at extended times to achieve the sigmoidal curve have been presented in literature and are not relevant to the current study.16 However, a nucleation and growth mechanism, presumed by the JMA law, can be used to explain the transformation that is occurring. Nucleation and growth curves are temperature dependent: less time is needed for transformation at higher temperatures, while more time is needed for transformation at lower temperatures. Time and temperature dependence can be seen in the graph, strengthening the argument that the transformation is occurring by a nucleation and growth mechanism. The transformation also appears to be preceded by an incubation period, or period of stability. The incubation period is not well defined in the autoclaving treatment under 1 bar pressure due to lower steam temperatures. However, after being aged for 1 h in a 2 bar autoclave treatment the percent monoclinic increases quickly which suggests an incubation period for approximately 1 h. Chevalier et al. reported a 1 h incubation period at 134 C which is supported by this current study.16 Extrapolation of the pro-

1522

LUCAS ET AL.

2 Bar 5 Hour 19.0 53.4 45.6 37.7 29.9 24.3 21.6 19.9

(0.5) (1.8) (1.7) (1.5) (0.2) (1.2) (0.1) (0.3)

40 Hour Boil 13.7 39.8 30.0 26.5 20.2 16.9 14.5 14.2

(0.8) (3.3) (0.6) (2.6) (1.9) (0.4) (0.4) (0.8)

3.5 Day Boil 19.9 58.6 52.2 47.9 33.0 25.7 22.6 20.7

(1.5) (0.7) (5.9) (3.8) (2.4) (2.1) (1.1) (0.7)

7 Day Boil 31.1 (2.2) 57.6 (0.6) 55.1 (4.9) 50.5 (3.8) 42.7 (2.0) 36.7 (0.7) 33.0 (0.0) 31.8(2.5)

longed boiling data indicates an incubation period of 10 h (Figure 2). The current finding is in agreement with reported values suggesting the incubation period at 100 C is approximately 10–20 h.16 A common trend of higher surface transformation was seen in the aging process of all tested conditions. Higher surface transformations strengthen the argument that t ! m transformation follows a nucleation and growth mechanism starting at the surface where a saturation of transformed grains is reached and then transformation progresses inward.2,14 The current study also supports the incubation period theory, which states there is a necessary aging time needed before transformation can begin and progress.12,16 The different aging conditions produced similar results but prolonged boiling method is the harshest treatment, perhaps due to the increased time of aging. The samples that underwent prolonged boiling had slightly higher overall transformations throughout the depth of the sample and higher surface layer transformations. While prolonged boiling is not stated as an artificial accelerated aging treatment in the ISO standard 13356, prolonged boiling has been shown to produce similar transformations and similar aging trends as 2 bar autoclaving. Therefore, if an autoclave device is not readily available then prolonged boiling can be used as an alternative. A few limitations with using the X-ray diffraction data to follow transformation as a function of depth should be noted. X-ray diffraction uses an area approximation method for its measurements. Therefore, when a value is presented for transformation at 6 mm, it is actually an accumulation of transformation within the first 6 mm. Therefore, it must be taken into account that the actual amount of transformation at each depth is less than what is shown due to the aggregate amount of transformation to that depth. Furthermore, the volume fraction measurement depends upon determining the area of XRD peak areas accurately. Below approximately 5% volume transformed, the area of the monoclinic phase peak is too small relative to the background to be reliably measured. While it has been previously reported that the LTD causes the t ! m transformation to originate at the surface and progress inward,3 there is a gap in the understanding of how the transformation affects the surface. Numerous studies have addressed the integrity of the material due to LTD2–7 but there is a paucity of data that addresses the transformation at the surface. The current study shows

PHASE TRANSFORMATION OF DENTAL ZIRCONIA

ORIGINAL RESEARCH REPORT

large transformations are occurring at the surface, combined with the knowledge that the transformation causes a volumetric expansion, there is a need for further research to address the wear of opposing dentition and or materials that contact aged, or aging zirconia. Future studies will likely utilize AFM to characterize surface roughness. CONCLUSIONS

This study presented three in vitro techniques to accelerate the aging process that occurs in vivo for Y-TZP, prolonged boiling at 100 C, steam, sterilization at 1 bar and 121 C, and steam sterilization at 2 bar and 134 C. This study then used two different methods of X-ray diffraction to observe the depth progression of the t ! m transformation in Y-TZP due to prolonged boiling at 100 C and steam sterilization at 2 bar (134 C). Within the limits of this investigation the following conclusions are presented. 1. The amount of monoclinic transformation due to artificial aging increased with an increase in aging treatment time. 2. Prolonged boiling for 3.5 days is statistically similar to autoclaving with 2 bar of pressure for 5 h and produces approximately a 20% monoclinic transformation. 3. The phase transformation is preceded by an incubation period, or period of stability with no significant transformation occurring, which was dependent on temperature and pressure. 4. Higher surface transformation was observed for all aging conditions compared to total transformation up to 6 lm from the surface. 5. A decrease in monoclinic transformation was observed with depth progression from the surface for all tested conditions.

ACKNOWLEDGMENT

Zirconia samples were supplied by 3M ESPE, St. Paul, MN.

REFERENCES 1. Deville S, Gremillard L, Chevalier J, Fantozzi G. A critical comparison of methods for the determination of the aging sensitivy in biomedical grade yttria-stabilized zirconia. J Biomed Mater Res 2005;72:239–245. 2. Lawson S. Environmental degradation of zirconia ceramics. J Eur Ceram Soc 1995;15:485–502. 3. Denry I, Kelly J. State of the art of zirconia for dental applications. Dent Mater 2008;24:299–307. 4. Kelly J, Denry I. Stabilized zirconia as a structural ceramic: An overview. Dent Mater 2008;24:289–298. 5. Swab J. Low temperature degradation of Y-Tzp materials. J Mater Sci 1991;26:6706–6714. 6. Papanagiotou H, Morgano SM, Giordano RA, Pober R. In vitro evaluation of low-temperature aging effects and finishing procedures on the flexural strength and structural stability of Y-TZP dental ceramics. J Prosthet Dent 2006;96:154–164. 7. Piconi C, Maccauro G. Review: Zirconia as a ceramic biomaterial. Biomaterials 1999;20:1–25. 8. Cattani-Lorente M, Scherrer S, Ammann P, Jobin M, Wiskott A. Low temperature degradation of a Y-TZP ceramic. Acta Biomater 2011;7:858–865. 9. Chowdhury S, Vohra Y, Lemons J, Ueno M, Ikeda J. Accelerating aging of zirconia femoral head implants: Change of surface structure and mechanical properties. J Biomed Mater Res B 2007;81: 846–492. 10. Chevalier J. What future for zirconia as a biomaterial? Biomaterials 2006;27:535–543. 11. Lughi V, Sergo V. Low temperature degradation—aging—of zirconia: A critical review of the relevant aspects in dentistry. DentMater 2010;26:807–820. 12. Roy M, Whiteside L, Katerberg B, Steiger J. Phase transformation, roughness, and microhardness of artificially aged yttria- and magnesia-stabilized zirconia femoral heads. J Biomed Mater Res 2007;83:1096–1102. 13. Catledge S, Cook M, Vohra Y, Santos E, McClenny M, Moore K. Surface crystalline phases and nanoindentation hardness of explanted zirconia femoral heads. J Mater Sci Mater Med 2003;14: 863–867. 14. Chevalier J, Gremillard L, Deville S. Low temperature degradation of zirconia and implications for biomedical implants. Rev Mater Res 2007;37:1–32.  miya S. Calibration curve for quantita15. Toraya H, Yoshimura M, So tive analysis of the monoclinic-tetragonal ZrO2 system by X-ray diffraction. J Am Ceram Soc 1984;6:119–121. 16. Chevalier J, Cales B, Drouin J. Low temperature aging of Y-TZP ceramics. J Am Ceram Soc 1999;82:2150–2154.

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2015 VOL 103B, ISSUE 7

1523

Phase transformation of dental zirconia following artificial aging.

Low-temperature degradation (LTD) of yttria-stabilized zirconia can produce increased surface roughness with a concomitant decrease in strength. This ...
150KB Sizes 0 Downloads 4 Views