The Effect of Aging on Composition and Surface of Translucent Zirconia Ceramic.

OBJECTIVES: To examine the effect of two aging protocols on the chemical and phase composition as well as the surface state of monolithic translucent zirconia ceramics.
MATERIAL AND METHODS: Translucent zirconia ceramics KATANA-Zirconia STML with different surface treatments (no treatment, K1, K2; glazed, G1-G8; polished, P1-P8) underwent testing in order to examine how the two aging protocols (three-hour hydrothermal degradation in an autoclave at 134 °C and 2 bars: G1-G4, P1-P4, and sixteen-hour chemical degradation in four-percent acetic acid at 80 °C (ISO 6872): G5-G8, P5-P8) affect chemical composition, particularly the share of stabilizing yttrium oxide (Energy Dispersive X-Ray Fluorescence - EDXRF), phase composition (X-ray diffraction - XRD) and surface state in terms of roughness and gloss.
RESULTS: Aging protocols did not affect the tested chemical composition stability of specimens and a high share of stabilizing yttrium-oxide (≥10% of total content), which correlates with the absence of monoclinic phase. A decrease in gloss on all specimens is statistically significant. Chemical degradation substantially increased the surface roughness of tested specimens.
CONCLUSIONS: Translucent monolithic zirconia demonstrated a stable chemical composition and resistance to tetragonal-to-monoclinic transformation. Surface gloss was significantly reduced, especially in polished specimens. Contrary to glazed specimens, the tested polished specimens manifested an increase in surface roughness. Glazing the surface of translucent monolithic zirconia produces better esthetic, tribological and hygienic effects than polishing.

Introduction

For several years, zirconia based ceramics has been the material of choice in the fabrication of fixed partial dentures (FPD), especially for the more strain-prone posterior region of dental arch, because of its good mechanical properties (flexural strength 900-1,200 MPa, hardness HV0.1 1200) (). Due to insufficient translucency, it has been unable to meet high esthetic criteria for usage in the restoration of the anterior part of dental arch (). For that reason, it has been veneered with ceramic materials that have better optical properties (glass and feldspathic ceramics) (-).

Two problems have arisen in the clinical application of this material. One problem is common to all bilayer systems and is caused by a discord in the thermal expansion coefficient between the veneering material and the core material. Because of this discord, strain is exerted on the “weaker” material, which may result in chipping of the veneering material (-). Another big problem is aging (low temperature degradation – LTD - spontaneous tetragonal-to-monoclinic transformation of crystal structure in a moist medium at room temperature) (-). The transformation of crystal structure generates the expansion of grain volume (4-5%) (-). Grain growth exerts strain on grain boundaries (-), which may lead to microcracks in the material and, once critical fracture toughness has been exceeded, prosthetic restoration fracture (-). Spontaneous phase transformation of zirconia ceramics is inhibited by adding various shares of stabilizers (yttrium, calcium, magnesium or cerium oxide) (, , ). Simultaneously, the expansion of the volume of grains volume and transformation toughening (closing cracks) provide these materials with a high value of flexural strength. The scientific and expert communities have been making efforts to eliminate the aforementioned problems and improve zirconia materials while maintaining their good properties. Through multi-parameter experimental protocols, they have endeavored to examine the long-term behavior of materials in the oral cavity as realistically as possible by analyzing both mechanical and esthetic properties. Improving the esthetic quality of these materials extends their application to the anterior region of dental arch. The most recent research studies indicate that optical properties of zirconia ceramics could be improved by increasing the share of transformation-resistant cubic crystalline structure (-).

Aside from microstructure, surface state also has a great impact on optical properties (-). Due to specular reflection, shiny and glossy surfaces seem lighter and fresher, whereas unpolished and rough surfaces appear darker and more saturated (-). Tribological relations of opposite substrates also give importance to surface state (-). Greater surface roughness of previous generations of zirconia ceramics and the consequential greater consumption of antagonists in the chewing process were caused by phase transformation and the ensuing presence of the monoclinic phase; as crystals in that phase had a larger volume than crystals in the tetragonal phase, they rose above the surface of the material (-). In their extensive review of relevant literature, Passos et al. highlighted the importance of polishing zirconia fixed partial dentures, noting that enamel abrasion of antagonist teeth in such tribo-pairs is reduced to a minimum (). Sripetchdanond et al. formed a tribo-pair made up of tooth enamels with monolithic zirconia ceramics and glass ceramics, respectively, and found that the former material wore a natural tooth out much less than the latter (). Another important factor in plaque control should not be ignored – the manner and quality of the final surface treatment on a restoration ().

The aim of this research was to investigate how aging protocols affect chemical composition (in particular, the share of stabilizing yttrium oxide), phase composition and the manifestation of tetragonal-to-monoclinic transformation as well as surface state (conveyed as surface gloss and surface roughness).

The following hypotheses were tested: the new generation of zirconia material contains a share of yttrium oxide that correlates with structural stability, i.e. the absence of tetragonal-to-monoclinic phase transformation. Experimental aging protocols will cause neither phase transformation nor a decrease in the share of stabilizing yttrium oxide in the material. Experimental aging protocols will have an effect on the surface gloss of the specimens. Experimental aging protocols will have an effect on the surface roughness of the specimens.

Material and methods

Specimen Preparation

The material employed in this research was a monolithic zirconia based ceramics KATANA-Zirconia Super Translucent Multi Layered - STML (Kuraray Noritake Dental Inc., Tokyo, Japan), shade A2. Using the CAD/CAM technology (Zenotec Easy Wieland Dental, Pforzheim, Germany), 18 disc-shaped specimens (11 x 11 x 1.5 mm with ±5% tolerance) were fabricated. The specimens were sintered in a furnace (Wieland, Pforzheim, Germany) at 1,550 °C for two hours according to manufacturer’s instructions. The temperature was increased until the sintering temperature was reached and then decreased in the cooling process after the final sintering at a rate of 10 °C/min. The specimens were divided into two groups, according to the final surface treatment applied: eight specimens were glazed following the manufacturer’s instructions (G1-G8), whereas eight specimens were polished with rubber (Ceragloss, EDENTA AG, Switzerland) accompanied by water cooling (P1-P8). The remaining two specimens served as control specimens in the research and received no surface treatment (K1, K2) (Figure 1).

Figure 1

Graphic division of groups of specimens.

Measurements and Analyses

All measurements and analyses of the specimens were conducted in two phases: before and after experimental aging protocols.

Two specimens (K1, P2) were analyzed using Energy Dispersive X-ray Fluorescence (EDXRF) (Ruđer Bošković Institute, Zagreb, Croatia). A Philips W X-ray tube (Philips Co., Amsterdam, Netherlands) was utilized as a source. The specimens were irradiated by the secondary Mo target in rectangular geometry. The signal was detected using a semiconductor SiLi detector (Canberra Packard, Vienna, Austria) with 30 mm2 of active surface area and a thickness of 3 mm. The thickness of beryllium window of the detector was 0.025 mm and the resolution at 5.9 keV was 170 eV (FWHM). The working parameters were 35 kV and 5 mA in a 100-bar vacuum. The measurement period was 100 s. The samples were analyzed for Y2O3, ZrO2, HfO2 as well as Sr and Zn. Relative errors of measurement, defined as slope coefficient errors for correlation lines were as follows: Y2O3 – 2.11%, ZrO2 – 1.84%, HfO2 – 2.50%, Zn and Sr – 0.1%. Minimum detection limits (MDL) were: Y2O3 – 0.11%, ZrO2 – 1.35%, HfO2 – 0.10%, Zn and Sr – 0.01%.

X-ray diffraction (XRD) was performed on two polished specimens (P1, P5) with a diffractometer (Philips PW 1820, Philips Co., Amsterdam, Netherlands) in order to attain the initial crystal structure. These two specimens (P1, P5) were subjected to CuK-alpha radiation in the space between 10° and 70° of the 2theta angle, with a step size of 0.02° and step time of 1 s/step, because it was assumed that the initial phase composition of all specimens was identical.

A glossmeter (Elcometer 407, Elcometer Inc., Michigan, USA) was utilized in null measurements of the surface gloss on the specimens. Elcometer is a multiangle glossmeter, meaning that light is reflected (detected) off a surface at 20°, 60° and 85° angles (60° being the reference angle). Ten measurements were conducted on each specimen (Faculty of Graphic Arts, University of Zagreb, Croatia).

Ten roughness profiles were recorded on specimens G3, G6, P3 and P6 using stylus instrument Perthometer S8P (Perthen Mahr, Göttingen, Germany) under the following conditions:

Gauss filter: λc = 0.8 mm

Tip radius: r = 5 µm

Evaluation length: ln = 4 mm.

Roughness parameters R and R were calculated on the recorded profiles. The traceability of measurement results was ensured by Croatian national roughness standards.

Two experimental protocols were conducted in this research. A three-hour hydrothermal degradation was carried out in an autoclave (SKO 7, Faro, Italy) at a temperature of 134 °C and under a pressure of 2 bars (six 30-minute cycles), with distilled water. On the other hand, chemical degradation was carried out in a corrosive medium (four-percent acetic acid (CH3COOH), pH 2.49) at 80 °C over a sixteen-hour period (ISO 6872). The specimens were divided into four subgroups according to the experimental protocol they were subjected to. Four glazed specimens (G1-G4) (first subgroup) and four polished specimens (P1-P4) (second subgroup) were sterilized in an autoclave, whereas the remaining four glazed specimens (G5-G8) (third subgroup) and the remaining four polished specimens (P5-P8) (fourth subgroup) were immersed into a corrosive medium in a 1,000-mL glass measuring flask. Instead of being subjected to experimental protocols, control specimens (K1, K2) were kept at a standard temperature of 20 °C and under a standard pressure of 1 atm (101,325 Pa) throughout the study.

The same measurements and analyses from the beginning of the research were made after the aging protocols. One specimen from each subgroup (G2, G6, P2, P6) and control specimen K1 were analyzed using the EDXRF methodology so as to identify the chemical composition of the specimens after aging protocols. The phase composition was determined with a post-aging XRD analysis of one specimen from each subgroup (G1, G5, P1, P5). Surface gloss was measured on all specimens. Post-aging surface roughness was analyzed on the same specimens as at the start of the research.

Analysis of Results

The results of the research are displayed in tables (chemical composition) and figures (diffraction analyses). In order to statistically analyze the results of surface gloss measurements, a Bonferroni test and an ANOVA (Analysis of Variance) test with a 95% confidence interval were employed. All p values under 0.05 were considered to be statistically significant. The values of roughness parameters R and R are expressed as arithmetical means of 10 recorded profiles, with the addition of a one-way ANOVA.

Results

The pre-aging examination of chemical composition on specimens K1 and P2 (Table 1) yielded the following shares (arithmetic mean ± standard deviation) of chemical compounds: yttrium oxide 11.97 ± 0.13%; zirconium oxide 85.87 ± 0.19%; hafnium oxide 2.16 ± 0.15%. Multiple specimens were not analyzed because the chemical composition of all specimens was found to be identical at the start of the research.

Table 1

Pre-aging chemical composition of zirconia ceramic KATANA-Zirconia STML (arithmetic mean ± standard deviation (AM ± SD)).

SPECIMEN	DESCRIPTION	Y2 O3 (%)	Zr O2 (%)	HfO2 (%)
K1a	Control specimen – upper surface	11.79	85.97	2.24
K1b	Control specimen – lower surface	12.10	85.63	2.26
P2a	Polished specimen – polished surface	12.01	86.06	1.94
P2b	Polished specimen – unpolished surface	11.98	85.82	2.20
	AM ± SD	11.97 ± 0.13	85.87 ± 0.19	2.16 ± 0.15

The results of chemical composition after hydrothermal degradation in an autoclave (specimens P2, G2) and chemical degradation in a corrosive medium (specimens P6, G6) are displayed in Table 2. The difference in the share of chemical elements before and after aging protocols is not a significant one. On the glazed side of glazed specimens, the difference in the share of yttrium oxide (10.90 ± 0.30%) was somewhat larger than on the non-glazed side of glazed specimens (11.85 ± 0.00%). Non-glazed sides of glazed specimens had almost an identical share of chemical compounds as all other specimens.

Table 2

Post-aging chemical composition of zirconia ceramic KATANA-Zirconia STML.

SPECIMEN	DESCRIPTION	Y2O3	ZrO2	HfO2	Zn	Sr
		(%)	(%)	(%)	(%)	(%)
P2a	Polished specimen – polished surface	11.91	86.03	2.06	0	0
P2b	Polished specimen – unpolished surface	11.93	85.85	2.23	0	0
P6a	Polished specimen – polished surface	11.77	86.31	1.92	0	0
P6b	Polished specimen – unpolished surface	11.91	86.14	1.95	0	0
	AM ± SD	11.88 ± 0.07	86.08 ± 0.19	2.04 ± 0.14
K1a	Control specimen – upper surface	11.75	86.22	2.04	0	0
K1b	Control specimen – lower surface	11.74	86.29	1.97	0	0
	AM ± SD	11.75 ± 0.01	86.26 ± 0.05	2.01 ± 0.05
G2a	Glazed specimen – glazed surface	10.68	83.23	1.21	3.38	1.5
G6a	Glazed specimen – glazed surface	11.11	84.36	1.47	2.14	0.92
	AM ± SD	10.90 ± 0.30	83.80 ± 0.80	1.34 ± 0.18	2.76 ± 0.88	1.21 ±0.41
G2b	Glazed specimen – non-glazed surface	11.85	86.35	1.8	0	0
G6b	Glazed specimen – non-glazed surface	11.85	86.31	1.85	0	0
	AM ± SD	11.85 ± 0.00	86.33 ± 0.03	1.83 ± 0.04

Pre-aging recordings of specimens subjected to X-ray diffractometry (P1 and P2) show the diffraction maximum of tetragonal zirconia at a position of 30°2Theta and cubic phase peaks (Figure 2). Phase shares are expressed as percentages: tetragonal phase 62.7%, and cubic phase 37.3%. The phase composition analysis of specimens subjected to hydrothermal degradation in an autoclave (glazed G1, polished P1) (Fig. 3) revealed a full cubic-to-tetragonal phase transition. The phase composition of specimens subjected to chemical degradation in a corrosive medium (G5, P5) was virtually without change (Figure 4).

Figure 2

X-ray diffraction analysis of polished specimens P1 and P5 before aging protocols.

Figure 3

X-ray diffraction analysis after aging in an autoclave for glazed specimen G1 (2A) and polished specimen P1 (2B).

Figure 4

X-ray diffraction analysis after chemical degradation in a corrosive medium for glazed specimen G5 (3A) and polished specimen P5 (3B).

Pre-aging surface gloss measurements have manifested significantly higher gloss values (arithmetic mean (AM) = 22.26 GU1) in glazed specimens (G1-G8) than in polished specimens (P1-P8) (AM = 19.77 GU; p<0.05). After aging protocols, a statistically significant reduction in gloss value occurred in both specimen groups (glazed AM = 18.54 GU; polished AM = 14.89 GU). With regard to the aging protocols applied, gloss change (ΔG) was larger in the subgroup of glazed specimens subjected to chemical degradation (G5-G8) (ΔGavg = -4.38 GU) than in the subgroup subjected to hydrothermal degradation (G1-G4) (ΔGavg = -3.08 GU) (Table 3). In polished specimens, surface gloss reduction after both aging protocols was nearly identical (average gloss change after hydrothermal degradation (P1-P5): ΔGavg = -4.65 GU; after chemical degradation (P5-P8): ΔGavg = -4.97 GU) (Table 4). Control specimens displayed low gloss values from the beginning of the research as they received no surface treatment and their average gloss change ΔGavg = -1.37 GU was not statistically significant.

Table 3

Inter-group statistical analysis of surface gloss measurements before and after aging protocols (Bonferroni test with 95% confidence interval, p<0.05).

DEPENDENT VARIABLE			MEAN DIFFERENCE (GU)	STD. ERROR	p	95% CONFIDENCE INTERVAL
						LOWER BOUND	UPPER BOUND
GLOSS BEFORE PROTOCOLS	G1-G8	P1-P8	2.49	0.70	0.001	0.80	4.18
		K1, K2	16.31	1.11	<0.001	13.64	18.98
	P1-P8	G1-G8	-2.49	0.70	0.001	-4.18	-0.80
		K1,K2	13.82	1.11	<0.001	11.15	16.49
	K1,K2	G1-G8	-16.31	1.11	0.000	-18.98	-13.64
		P1-P8	-13.82	1.11	<0.001	-16.49	-11.15
GLOSS AFTER PROTOCOLS	G1-G8	P1-P8	3.64	0.31	<0.001	2.90	4.39
		K1,K2	13.95	0.49	<0.001	12.78	15.13
	P1-P8	G1-G8	-3.64	0.31	<0.001	-4.39	-2.90
		K1,K2	10.31	0.49	<0.001	9.13	11.48
	K1,K2	G1-G8	-13.95	0.49	<0.001	-15.13	-12.78
		P1-P8	-10.31	0.49	<0.001	-11.48	-9.13

Table 4

Statistical analysis of results of gloss measurements (GU) in subgroups, with regard to the aging protocol applied (One-way ANOVA, p<0.05).

SPECIMENS	BEFOREAM ± SD(GU)	PROTOCOL	AFTERAM ± SD(GU)	∆Gavg(GU)	p
G1-G4	22.55 ± 2.13	Hydrothermal degradation	19.49 ± 1.36	-3.06	<0.001
G5-G8	21.96 ± 2.73	Chemical degradation	17.58 ± 3.27	-4.38	<0.001
P1-P4	19.43 ± 6.49	Hydrothermal degradation	14.78 ± 1.87	-4.65	<0.001
P5-P8	19.96 ± 5.96	Chemical degradation	14.99 ± 1.66	-4.97	<0.001
K1, K2	5.95 ± 0.41	Without degradation	4.58 ± 0.36	-1.37	0.263

The arithmetic means of roughness parameters R and R on polished specimens P3 and P6 obtained after experimental aging protocols were higher than the values obtained on the same specimens before aging protocols were performed (Table 5).

Table 5

Values of parameters Ra and Rz on polished specimens (P3 and P6) before and after aging protocols.

Measurement No.	POLISHED SPECIMENS BEFORE AGING				POLISHED SPECIMENS AFTER AGING
	SPECIMEN P3		SPECIMEN P6		SPECIMEN P3		SPECIMEN P6
	Ra,µm	Rz,µm	Ra,µm	Rz,µm	Ra, µm	Rz,µm	Ra,µm	Rz,µm
1	0.73	4.5	0.52	3.26	0.71	4.39	0.7	4.48
2	0.56	3.77	0.55	3.64	0.65	4.45	0.66	4.16
3	0.71	4.19	0.53	3.67	0.58	3.9	0.68	4.32
4	0.7	4.11	0.56	3.68	0.62	4.32	0.63	3.96
5	0.62	3.83	0.54	3.5	0.69	4.47	0.64	3.65
6	0.61	3.86	0.58	3.66	0.63	3.89	0.58	3.91
7	0.67	3.84	0.6	3.78	0.63	4.2	0.64	3.75
8	0.57	3.22	0.57	3.41	0.72	4.4	0.63	3.93
9	0.69	4.45	0.65	4.16	0.66	4.28	0.55	3.84
10	0.67	4.52	0.54	3.7	0.69	4.75	0.59	3.41
	0.65	4.03	0.56	3.65	0.66	4.31	0.63	3.94
s, nm	59	408	39	239	44	260	46	316

The results of one-way ANOVA point to a statistically significant increase in the values of estimated standard deviations for roughness parameters R and R on polished specimen P6 (Table 7).

Table 7

Statistical analysis of roughness parameters Ra and Rzz before and after aging protocols on G3, G6, P3 and P6 specimens.2(One-way-ANOVA, p<0,05).

SPECIMEN	PROTOCOL	RaAM ± SD before	RaAM ± SD after	p	RzAM ± SD before	RzAM ± SD after	p
G3	HD	0.86 ± 0.035	0.68 ± 0.02	0.03	3.13 ± 0.77	2.76 ± 0.58	0.33
G6	CD	0.93 ± 0.02	1.26 ± 0.05	0.001	3.16 ± 0.11	4.03 ± 0.06	0.005
P3	HD	0.65 ± 0.004	0.66 ± 0.002	0.834	4.03 ± 0.166	4.31 ± 0.07	0.09
P6	CD	0.56 ± 0.001	0.63 ± 0.002	0.003	3.65 ± 0.06	3.94 ± 0.1	0.03

The arithmetic means of roughness parameters R and R on glazed specimen G3 obtained after the aging protocol were lower than before the aging protocol was performed. An increase in the values of roughness parameters R and R after the aging treatment was present on glazed specimen G6 (Table 6).

Table 6

Values of parameters Ra and Rz on glazed specimens (G3 and G6) before and after aging protocols.

MeasurementNo.	GLAZED SPECIMENS BEFORE AGING				GLAZED SPECIMENS AFTER AGING
	SPECIMEN G3		SPECIMEN G6		SPECIMEN G3		SPECIMEN G6
	Ra,µm	Rz,µm	Ra,µm	Rz,µm	Ra,µm	Rz,µm	Ra,µm	Rz,µm
1	1.05	3.72	1.01	3.35	0.52	1.83	1.71	5.61
2	0.64	2.43	1.16	3.18	0.47	1.79	1.24	3.44
3	0.92	3.8	1.16	3.93	0.53	2.07	1.35	3.9
4	0.78	2.67	0.76	2.87	0.71	4.27	1.13	3.5
5	0.82	2.94	0.79	2.72	0.7	2.54	1.06	3.41
6	0.95	3.76	0.83	3.33	0.64	2.98	0.86	3.11
7	1.22	4.89	0.89	3.03	0.88	3.28	1.22	3.64
8	0.68	2.31	0.89	2.92	0.66	2.66	1.38	4.74
9	0.86	2.49	0.97	3.12	0.76	3.23	1.42	4.44
10	0.64	2.24	0.82	3.19	0.89	2.93	1.26	4.46
	0.86	3.13	0.93	3.16	0.68	2.76	1.26	4.03
s, nm	187	876	144	335	144	759	228	773

The results of one-way ANOVA point to a statistically significant change in the values of estimated standard deviations for roughness parameter R on glazed specimens G3 (decrease) and G6 (increase) as well as for roughness parameter R on glazed specimen G6 (increase) (Table 7).

Discussion

Various aging protocols provide an insight into the possible behavior of the constituent material in the mouth over a lengthy period of time. This research comprised several contemporary and costly protocols. A three-hour period was chosen for exploring hydrothermal degradation in an autoclave because it corresponded to a restoration usage period of 10-15 years (-), which should be the expected restoration usage period both from a therapist’s and a patient’s point of view. Submersing specimens into a medium of low pH value (pH 2.49) at a temperature of 80 °C for 16 hours (ISO 6872) is a method of examining the stability of materials under the conditions similar to those in the oral cavity, with constantly changing temperature and pH values (-). Plaque that clings to the surface of dental hard tissue and dentures has a pH value in its depth similar to the pH value of the acid utilized in this research; hence the effect of plaque was simulated in the present study, as well (). The quantity of specimens in this research was defined after a pilot study, in agreement with associates/experts under whose supervision the tests were administered. The tested number of specimens sufficed for a part of the testing (EDXRF, XRD, surface roughness) because the materials had been manufactured in a controlled environment. Furthermore, these testing methods were employed using very expensive, sophisticated apparatuses and methods which would have obtained the results relevant even on a small number of specimens. It is to be expected that a small quantity of specimens would be representative of a certain test type and that the obtained results could be repeated and applied to a greater number of specimens. It would have certainly been desirable to have more specimens for other tests (XRD, surface roughness), but one ought to take into account the extremely high price of the tested material, great costs for specimen manufacture as well as the cost and delicacy of measuring instruments.

At room temperature, pure zirconia exists in the monoclinic phase and great tension inside the material renders it unusable in the field of dental medicine, i.e. fixed prosthodontics (). Yttrium oxide is the most frequently used stabilizer in zirconia ceramics (-, ). When its share is 3-8%, it is possible to stabilize the tetragonal phase of zirconia ceramics, the result being so-called Y-TZP (yttrium-stabilized tetragonal zirconia polycrystal material). In the present study, the post-aging share of yttrium oxide (polished specimens 11.88 ± 0.07%, control specimens 11.75 ± 0.01%, glazed specimens 10.90 ± 0.30% - 11.85 ± 0.00%) was not significantly lower than the pre-aging share (11.97 ± 0.13%), i.e. the chemical composition of the specimens was not significantly altered (Table 2). A somewhat lower share of yttrium oxide on the glazed side of glazed specimens (10.90 ± 0.30%) can be explained by the presence of strontium and zinc in the glaze. Tested specimens manifested the stability of their chemical composition after being subjected to both hydrothermal degradation in an autoclave and chemical degradation in acetic acid. So far, there has been no mechanism for which one could say with certainty that it ages zirconia, although there have been some theories that could account for that phenomenon. A theory by Lange et al., based on an analysis of SEM images, suggests that water reacts with Y2O3, creating yttrium hydroxide (Y (OH) 3), a chemical compound that prompts the loss of stabilizers in the surrounding grains and a tetragonal-to-monoclinic transformation (). According to Yoshimura, water evaporation causes the bond between zirconium and oxygen in zirconia ceramics to break; the newly-free -OH ions exert stress and strain on the material and the outcome is a tetragonal-to-monoclinic transformation (). Chevalier et al. believe that free oxygen radicals O2-, formed as a consequence of water dissociation, destabilize the structure of Y-TZP and trigger LTD (). It has been proven that a tetragonal-to-monoclinic transformation starts on the surface of the material and progresses into its interior (-). This transformation leads to an increase in the volume of surface grains, which rise above the rest of the surface, create micro-cracks and open up the way for water and thereby the transformation to penetrate deeper into the material. Under the strain of e.g. masticatory force in the mouth, a restoration will ultimately crack (-). A greater post-aging share of stabilizing yttrium oxide calculated in this study (up to 12%) suffices for maintaining the stability of the crystal structure of specimens (-), i.e. for preventing a phase transformation. The first hypothesis is thus confirmed.

At the start of the present study, two phases were found to co-exist in specimens: tetragonal (approx. 63%) and cubic (approx. 37%). After being exposed to hydrothermal degradation in an autoclave, the crystal structure of the specimens fully transitioned to a tetragonal structure (100%). Unlike the first aging protocol, degradation in a corrosive medium did not bring about a change in the crystal structure of the specimens, i.e. tetragonal and cubic structures continued to co-exist in nearly the same ratio as at the start of the research. These analyses did not establish a phase (tetragonal-to-monoclinic) transformation of monolithic zirconia specimens; therefore the second hypothesis is accepted. A monoclinic phase in monolithic zirconia with a mixed tetragonal-cubic microstructure was not manifested in a study by Muñoz et al., although the share of the cubic phase increased after hydrothermal degradation (). Kolakarnprasert et al. did not prove the manifestation of a monoclinic phase in the material after exposing specimens submerged in water to a temperature of 120 °C for a twelve-hour period; their findings correspond to the results of this study ().

Long-term stability, smoothness and gloss of a restoration are important from a hygienic (-, ), tribological (, , ) and esthetic point of view (, , ). Surface roughness and final surface treatment on FPD affect the color stability of dental materials (). Discoloration of dentures, caused by beverages such as coffee and tea, has been described in literature (). In the present study, zirconia ceramic was subjected to two types of final surface treatment – polishing and glazing. Glazing is a standard surface finishing treatment which closes any pores left after sintering, allows for a better aesthetic impression of the restoration and reduces the accumulation of biofilm (). In this research, surface state was expressed in terms of two parameters – surface gloss and surface roughness. In the imitation of a natural tooth in fixed prosthodontic therapy, a lot of emphasis is placed on the successful reproduction of surface texture (). An enamel-like texture is attained by manually spreading a thin layer of glaze over a finished restoration (). Because of divergences in the refraction of light, differences in texture generate differences in color perception. In this study, both experimental protocols brought about a statistically significant reduction of surface gloss in both specimen groups (G1-G8, P1-P8), which prompts the acceptance of the third hypothesis (p<0.05). From the beginning, glazed specimens manifested significantly higher surface gloss values than polished specimens. This study established that the surface gloss of specimens was reduced regardless of the final surface treatment applied, with both subgroups of polished specimens (P1-P4, P5-P8) having a higher mean gloss change ΔG than glazed specimens (G1-G4, G5-G8). It needs to be noted that surface gloss reduction in an oral cavity would unfold at a slower pace than during experimental protocols. Control specimens (K1, K2) did not manifest significant surface gloss reduction in this research, although their gloss values were low from the onset because they were not subjected to any surface treatment. Gloss reduction in glazed specimens is most likely a result of congruent dissolution of glaze during experimental protocols, especially during chemical degradation in a corrosive medium, which is confirmed by the results relating to surface roughness. A glazed specimen subjected to aging in an autoclave (G3) recorded a decrease of parameters R (p=0.03) and R (p=0.33), while a specimen submerged in a corrosive medium (G6) recorded a significant increase in roughness parameters R and R (p<0.05). As an amorphous material, glaze is not resistant to aggressive external influences nor is it chemically stable/inert. In an aggressive medium, surface ions perish from glaze, some irregularities appear on the surface and roughness is increased, which is in agreement with findings of Manicone et al. () and Milleding et al. (, ). The reduction of roughness parameters R and R in a glazed specimen aged in an autoclave (Table 6) supports the finding that aging a specimen in an autoclave “smoothed” its surface, therefore mean gloss change ΔG in those specimens (G1-G4) was lower than in polished specimens (P1-P4). The effect of hydrothermal degradation in an autoclave on glazed specimens (G1-G4) was reduced by a protective coating on the surface (glaze) so that the penetration of water and degradation of base material were inhibited, unlike in specimens without a protective coating. This was confirmed by Palla et al., who proved that water infiltrates through the surface of a non-glazed glass ceramic restoration, which results in the disintegration of the material’s structure (). Camposilvan et al. have suggested spreading glaze on all zirconia restorations so as to inhibit the impact of a hydrothermal aging protocol on the surface of the material (). In polished specimens (P1-P8), surface gloss decreased and gloss change ΔG was higher than in glazed specimens (G1-G8), which could be attributed to an increase in the values of measured roughness parameters during both experimental protocols (Tables 5 and 7). Štefančić confirmed these findings in her doctoral dissertation (). An increase in roughness was not substantial in a specimen subjected to hydrothermal degradation (P3), when compared to a specimen subjected to chemical degradation (P6), but it undoubtedly had an impact on surface gloss reduction in all polished specimens (). Judging by the results relating to roughness parameters, the fourth hypothesis can be partially accepted because a significant change in roughness occurred in specimens G6 and P6, which belong to subgroups of glazed and polished specimens respectively, which were subjected to chemical degradation in a corrosive medium (p<0.05). Surface roughness of control specimens was not measured using profilometry since these specimens had not been subjected to any type of final surface treatment and their surface roughness was high; hence it serves no purpose to compare it to the roughness of either polished or glazed specimens.

Relevant literature points to the conclusion that all types of dental ceramics (silica-based and oxide-based) exhibit reactivity in an aqueous medium, thus a completely inert ceramic material does not exist (, , , ). Polycrystalline, oxide-based ceramics also release ions, but to a much lesser extent than silica-based ceramics (, , ).

Conclusions

Within the limitations of this study, it can be concluded that experimental aging protocols do not reduce the share of stabilizing yttrium oxide and consequently do not generate a tetragonal-to-monoclinic phase transformation, i.e. aging of translucent multilayer zirconia ceramic.

Regardless of the surface treatment applied, surface gloss on all specimens was significantly reduced by experimental aging protocols, with surface gloss reduction being somewhat greater in polished specimens. The reduction of the gloss of specimens is an indicator of possible repercussions on the esthetic properties of dentures, particularly in polished specimens.

In contrast to the changes in the value of roughness parameters in glazed specimens subjected to aging in an autoclave (decrease) and glazed specimens submerged in a corrosive medium (increase), a significant increase in the value of roughness parameters in polished specimens following both aging protocols is a signal of substantial unwanted changes in surface state caused by polishing.

Glazing will have better esthetic, hygienic and tribological effects on the surface of a FPD than polishing.