Impact of ceria-yttria pigmentation on the mechanical performance and esthetics of zirconia dental restorations

Abstract

Background

An optimum combination of esthetics and mechanical properties is expected of all-ceramic restorations. Consequently, various pigmentation techniques of zirconia have been recommended to enhance the aesthetic results without compromising their long-term survival.

Methodology

Infiltrate solutions of ceria-yttria were prepared by mixing their precursors in various concentrations. Pre-sintered zirconia samples were soaked in the infiltrate solutions and sintered according to the manufacturer’s instructions. Phase analysis, microstructure using scanning electron microscopy, flexural strength, fatigue, CIE Lab, translucency parameter, surface roughness, and aging resistance of infiltrated zirconia were evaluated.

Results

Phase analysis confirmed the presence of the tetragonal phase of zirconia, and the microstructure revealed increased grain size. The flexural strength of infiltrated zirconia ranged from 248 MPa to 512 MPa, and the fatigue limit was lower than control zirconia with reduced surface roughness. The monoclinic content before and after aging was not detectable in the infiltrated samples. The CIE Lab values of the infiltrated samples showed a trend of decreasing lightness, accompanied by higher delta E values, with minimal change in translucency.

Conclusions

Ceria-yttria infiltrated zirconia exhibited reasonable flexural strength and fatigue performance, with improved aging resistance, and color suitable for anterior and low-stress-bearing monolithic restorations.

1. Introduction

Efforts to achieve an optimum combination of esthetics and mechanical properties of all-ceramic restorations have resulted in various generations of dental zirconia [1–3]. However, achieving a close resemblance to natural teeth remains a clinical challenge. Consequently, various pigmentation techniques of zirconia have been recommended to enhance the aesthetic results without affecting their long-term survival. These methods include incorporating pre-shading with metallic pigments into the zirconia powder during block formation, infiltrating coloring liquids into pre-sintered zirconia, and staining and glazing sintered zirconia [4–10].

Infiltration is one of the methods of obtaining pigmented zirconia by infusing green-stage zirconia with coloring liquids. This approach can address the drawbacks associated with pre-shaded zirconia blocks. Pre-shaded zirconia presents challenges in controlling the powder form, size, quantity of pigments, and pigment distribution within the zirconia powders. The infiltration technique using coloring liquids offers a faster and easier approach, and their shades closely resemble natural teeth. Additionally, infiltration can be a surface modification method as it influences the microstructural and crystallographic changes in zirconia [11–18].

Cerium-based salts, such as cerium nitrate, cerium acetate, and cerium chloride, have been used for infiltrating and pigmenting zirconia. The incorporation of cerium affects the lattice parameters of tetragonal zirconia, increasing its grain size [8,19–22]. Additionally, cerium enhances zirconia's resistance to low-temperature degradation or aging [19–22]. However, while higher concentrations of ceria intensify the color, they also decrease the zirconia's flexural strength.This reduction in flexural strength is attributed to the increased grain size and porosity in cerium-infiltrated zirconia [8,19–22]. Given the benefits of cerium-infiltrated zirconia in terms of color and enhanced aging resistance, alternative approaches for strengthening cerium-infiltrated zirconia must be explored [23]. Previous studies reported that an optimal combination of yttria and ceria provides the advantages of high strength and resistance to low-temperature degradation [24–30]. Yttria improved the strength in ceria-doped zirconia due to deformation plasticity with a change in grain size [24–30]. Hence, the present study proposes combining ceria and yttria-based coloring liquids as a pigmentation method to enhance the flexural strength without compromising the aging resistance of zirconia. The objective of the study was to synthesize a ceria-yttria-based coloring liquid for pigmenting zirconia using the infiltration method and evaluate its effect on the phase transformation, microstructure, optical properties, surface roughness, flexural strength, fatigue, and aging resistance of zirconia.

2. Materials and methods

Typically, brush and immersion methods are used for infiltrating pre-sintered zirconia [31]. The study followed the immersion method. Unlike the brush technique, the immersion method is quicker and simpler, producing shades that closely match natural teeth. Moreover, infiltration serves as a surface modification technique by inducing microstructural changes in zirconia, further affecting its properties [8, 19–22].

2.1. Preparation of infiltrate solutions

Infiltrate solutions were prepared by mixing the precursors, cerium nitrate hexahydrate (SpectraChem Pvt.Ltd, Mumbai, India), yttria nitrate hexahydrate powders (SpectraChem Pvt.Ltd, Mumbai, India), at prescribed concentrations (wt%), in de-ionized water and stirred until dissolved using a magnetic stirrer at 500 rpm. The nomenclature of each of the ceria-yttria used, with their weight concentrations is presented in Table 1.

Table 1. Nomenclature of ceria-yttria infiltrated zirconia samples.

Nomenclature	Cerium nitrate hexahydrate (wt%)	Yttrium nitrate hexahydrate (wt%)
1C–9Y	10	90
3C–7Y	30	70
5C–5Y	50	50
7C–3Y	70	30
9C–1Y	90	10

The weight concentration of ceria-yttria in the study was based on previous studies, wherein 1, 5 and 10 wt% of cerium acetate and cerium chloride, and ceria-yttria doping in the ratio of 12–3 mol%, 8.3–3.2 mol%, 8–1 mol%, 7–3 mol,% respectively, were investigated [24–30].

2.2. Infiltration method

Pre-sintered 3 mol% yttria-stabilized zirconia blocks of 12 mm diameter (Zirconia Aidite White) were infiltrated with the prepared wt% concentrations of cerium-yttrium solutions as shown in Table 1. The pre-sintered zirconia blocks were soaked in the above solutions at room temperature for 1 min. The samples were further dried with blotting paper and kept at ambient temperature and humidity inside a sealed dessicant-free container. As previous studies used long immersion times of cerium-based infiltrates ranging from 30 min to 2 h, impacting the mechanical properties of zirconia, a short immersion time of 1 min was chosen in the present study [8,19,20]. The dried samples were further sintered to 1550°C for 2 h, following the conventional sintering required for dental zirconia. The specimen dimensions of sintered zirconia were standardized using a digital spiral micrometer (CD-15AX; Mitutoyo). Considering the objective of evaluating the effect of ceria-yttria-based coloring liquids on the properties of zirconia, phase transformation, microstructure, color and translucency, surface roughness, flexural strength, fatigue resistance, and aging resistance were investigated.

2.3. Phase analysis using X-ray diffraction

X-ray diffraction (XRD) (X′pert Powder model, PANalytical, Westborough, MA, USA) was performed at 10–70°, with a scan step of 10.1600 s, at a 0.0170° step size, with CuK_α radiation. The surface monoclinic content was quantified by measuring the areas of selected peaks and applying the formulas of Garvie and Nicholson and Toraya et al., between 26 and 33°, the volume fraction of the monoclinic phase is given by [32,33] in Eqs. 1 and 2,

$$V_m=\frac{1.311X_m}{1+0.311X_m}$$ (1)

$$X_m=\frac{I{m}_{1+}I{m}_2}{I{m}_{1+}I{m}_2+I_t}$$ (2)

where X_m indicates the weight fraction and V_m, the volume fraction of m-ZrO₂, I_m1, and I_m2 indicate the intensities of monoclinic phase peaks at 2θ = 28.2°and 31.5° and I_t, the intensity of tetragonal phase peak at 2θ= 30.2° [34,35].

2.4. Microstructural characterization using scanning electron microscope

The samples were coated with gold using a sputter coater (Quorum Technologies, UK) for 30 s at 40 mA. Microstructural analysis was conducted using a scanning electron microscope (Schottky Field Emission, Gemini Column) along with energy dispersive spectroscopy (EDS). The SEM images were taken, and the average grain size was measured using the linear intercept method [36–39]. This technique involved counting the number of intersections made by a test line of known length on a digitally calibrated SEM image of the sample surface. ImageJ software was used for the analysis, with six test lines oriented in different directions for each image. The average grain size was calculated using the formula in Eq. 3,

$$D=1.56\frac{C}{MN}$$ (3)

Here, D represents the average grain size, while 1.56 is a proportionality constant accounting for the non-spherical shape of the grains, C denotes the total length of the test line used, N is the number of intercepts, and M refers to the magnification of the photomicrograph, which was set to 1 in this study since the image was already digitally calibrated. Random samples of ceria-yttria infiltrated group and control unpigmented zirconia were sectioned using a slow-speed saw (Metallography Equipment Baincut LSS, Chennai Metco, India), and were subject to SEM imaging with EDS. The infiltration depth was further calculated using ImageJ software.

2.5. Initial three-point flexural strength

The samples (n = 30) were milled as pre-sintered bars, immersed in the infiltrate solutions, and sintered as mentioned in the Methodology 2.2. The bars were milled for flexural strength testing (Single Column LFS System, BISS with 1KN) with dimensions of 40×5×3 mm. A support span of 30 mm was used with a crosshead speed of 1 mm/min. The load at fracture (N) was recorded, and the flexural strength (s) was expressed as MPa by using Eq. 4 [40].

$$Flexuralstrength(s)=\frac{3Pl}{2w{b}^2}$$ (4)

where P is the fracture load (N), l is the span (distance between the center of the supports) (mm), w is the width (mm), and b is the height (mm) of the samples.

2.6. Cyclic fatigue

As mentioned in 2.5, three-point flexural strength was used to determine cyclic fatigue by applying a sinusoidal cyclic compression-tension load-controlled mode on the samples (n = 10 per group) at 3 Hz, at a stress ratio (R) of 0.1, up to 10⁶ cycles. The chosen sample size was based on a similar in-vitro study [40]. About 50, 70, and 90 percent of the mean flexural load was applied in the control and the experimental group [41]. Amongst the experimental infiltrated groups, the group with the highest value of flexural strength was chosen. To simulate the oral environment, the samples were soaked in distilled water at 37°C for 14 days before fatigue testing [42]. The experimental setup for fatigue testing is presented in Fig. 1. The endurance limit was calculated based on Basquin’s power-law fit equation for 10⁶ cycles as presented in Eq. 5 [40]:

$$y=A\cdot x^{-b}$$ (5)

Fig. 1. Experimental set-up for fatigue test.

Wherein y is the percentage of static maximum load, x is the number of cycles to failure, A is the material constant, the theoretical load at 1 cycle, and b is the fatigue strength exponent, slope of the curve in log-log scale.

2.7. Fractographic analysis

The fractured fragments of each sample after flexural strength and fatigue tests were cleaned ultrasonically for 10 min using distilled water and then allowed to air-dry. Subsequently, a general inspection was conducted with an optical microscope to identify the fracture origin and fractographic features [43]. Representative samples were then coated with a gold sputter layer and analyzed using scanning electron microscopy (SEM) at magnifications ranging from 70X to 2000X. SEM images of the fracture surfaces were captured, with the fracture origin regions magnified to examine the fracture patterns.

2.8. CIE Lab coordinates and translucency parameter

Colorimetric analysis of the infiltrated samples was carried out using a UV–VIS–NIR spectrophotometer (PerkinElmer Lambda 950, Massachusetts, USA) equipped with a 150 mm diameter integrating sphere. The spectral data were collected at a wavelength interval of 5 nm. To ensure measurement accuracy and consistency, each sample was analyzed in triplicate (n = 3 readings per group), and the average reflectance values were used for further colorimetric evaluation. To quantify the perceived color difference of the samples, the total color difference, ΔE, was calculated using the Euclidean distance between the two points in the CIELab color space using the formula as follows, wherein ΔL, Δa, and Δb are the color differences of L, a, and b, respectively, in Eq. 6 [8].

$$\text{Δ}{E}^{\ast }=\sqrt{{\left(\text{Δ}{L}^{\ast }\right)}^2+{\left(\text{Δ}{a}^{\ast }\right)}^2+{\left(\text{Δ}{b}^{\ast }\right)}^2}$$ (6)

The translucency parameter (TP) values were obtained using a UV-Vis spectrophotometer (PerkinElmer Lambda 1050 +, Massachusetts, USA). The spectrophotometer was calibrated following the manufacturer's instructions before the measurements. Diffuse reflectance was recorded at 10 nm intervals in the range of 300–800 nm on white and black backgrounds. Three readings were obtained for each sample at a thickness of 2 mm, with five samples per group (n = 15 per group). The TP values were calculated by subtracting the Lab values of the black background from those of the white background using the following equation [44,45].

$$TP=\sqrt{{\left(L_b-L_w\right)}^2+{\left(a_b-a_w\right)}^2+{\left(b_b-b_w\right)}^2}$$ (7)

2.9. Aging resistance

For testing the aging behavior, random samples of each group were exposed to artificial degradation in an autoclave (Woson Tanzo Classic Class B) under water vapor at 134 ± 1 °C and 2 bar pressure for 5 h [46]. This time was chosen to represent more than the expected life of the application, as 1 h at 134 °C has been reported to simulate approximately 1–4 years at 37 °C, corresponding roughly to 20 years in vivo [40,47]. After the treatment, the monoclinic phase volume fraction (V_m%) near the surface was measured using X-ray Diffraction, applying the equation of Toraya et al., and the results were analyzed as mentioned in 2.3, Eq. 2.

2.10. Surface roughness

Surface roughness was evaluated using a microsystem analyser (Polytec MSA-500, Irvine, USA) and a non-contact optical profilometer (Taylor Hobson Precision, TalySurf CCI, Leicester, United Kingdom). The instrument was calibrated before the measurements. Measurements were taken for five samples of each group at the center of each sample (n = 5 for each group). The surface roughness parameters, such as Ra, the arithmetic average of surface roughness; Rq, the standard deviation of the profile heights; and Rz, an estimate of the extreme heights and valleys of the surface, were obtained [34].

2.11. Statistical analysis

Statistical analysis was performed using SPSS-20.0 software (IBM, USA). Data normality was assessed using the Shapiro-Wilk test, and based on the results, appropriate parametric tests or nonparametric statistical tests were selected.

3. Results

The phase analysis of all the ceria-yttria-infiltrated zirconia samples is presented in Fig. 2. Scanning electron microscopic (SEM) images of the samples, including the control zirconia, are presented in Fig. 3(a-f), along with energy-dispersive analysis (EDS) in Fig. 3(g). The SEM-EDS images of the sectioned infiltrated sample and the control are presented in Fig. 4(a) and Fig. 4(b), respectively. The results of flexural strength, surface roughness, CIE Lab values, and translucency parameter are presented as means and standard deviations. As the data followed a normal distribution, a one-way ANOVA was used to compare differences among the groups with a significance level set at p < 0.05. The surface roughness parameters, such as Ra, Rq, and Rz values of the as-received samples, are presented in Table 2 and Fig. 5. Table 3 presents the values of 3-point flexural strength. The fatigue plots of percentage load with the number of cycles to fracture for 5C-5Y and control unpigmented zirconia are presented in Fig. 6(a) and Fig. 6(b). The SEM images of the fractured surfaces of samples after flexural strength and fatigue tests, with a schematic illustration, are presented in Fig. 7(a-c). The CIE lab values of the infiltrated as-received samples are illustrated in Fig. 8 and Table 4 with ΔE values. Photographs of the infiltrated samples with the display of progressively deeper cream tones with a change in ceria and yttria content are presented in Fig. 9.

Fig. 2. XRD plots of ceria-yttria infiltrated zirconia samples depicting 2θ for the tetragonal phase of zirconia.

Fig. 3. SEM images of (a) Control (b) 1C-9Y (c) 3C-7Y (d) 5C-5Y (e) 7C-3Y (f) 9C-1Y and (g) Energy Dispersive Analysis (EDS) of 3C-7Y.

Fig. 4.

SEM images with EDS of sectioned samples, (a) infiltrated 5C-5Y and (b) control zirconia samples. The yellow arrow in (a) shows the ceria-yttria infiltration of 5C-5Y, which is further confirmed in the corresponding EDS.

Table 2. Surface roughness of infiltrated samples.

*Dissimilar superscripts denote statistically significant differences between the groups.

Surface roughness values
Sample	Ra (nm)	Rq (nm)	Rz (nm)
Control	129.8 ± 26.2a	155.5 ± 26.9a	704.1 ± 4.70a
1C–9Y	108.8 ± 29.7a	134.4 ± 29.3a	696.15 ± 102.8a
3C–7Y	97.88 ± 8.04a	121.30 ± 8.90a	611.88 ± 24.2a
5C–5Y	79.01 ± 12.1b	97.54 ± 12.80b	481.90 ± 42.6b
7C–3Y	102.05 ± 30.2a	123.51 ± 35.5a	613.03 ± 168.2a
9C–1Y	108.95 ± 9.2a	132.95 ± 15.0a	661.35 ± 109a

Fig. 5. 3D surface topography of ceria-yttria infiltrated samples with control zirconia.

Table 3. Flexural strength and grain size of the infiltrated samples.

*Dissimilar superscripts denote statistically significant differences between the groups.

Sample	Flexural strength (MPa) Mean ± SD	Grain size (μm) Mean ± SD
Control	704.85 ± 132.68a	0.812 ± 0.2
1C–9Y	284.38 ± 25.59b	5.34 ± 2.7
3C–7Y	453.61 ± 37.66b	1.67 ± 0.2
5C–5Y	521.22 ± 63.5a	0.98 ± 0.21
7C–3Y	478.69 ± 40.72b	1.59 ± 0.23
9C–1Y	420.50 ± 123.49b	1.16 ± 0.14

Fig. 6. Fatigue plots of (a) control and (b) 5C-5Y infiltrated zirconia.

Fig. 7.

Scanning electron images of (a) fractured surfaces after flexural test of 1C-9Y showing compression curls, arrest lines with hackle lines, (b) schematic illustration of fracture propagation across the sample under 3-point flexural strength, and (c) fractured surfaces after fatigue test of control and experimental 5C-5Y depicting crack propagation with compression curls. DCP stands for direction of crack propagation.

Fig. 8. CIE lab diagrams (a) Control (b) 1C-9Y (c) 3C-7Y (d) 5C-5Y (e) 7C-3Y (f) 9C-1Y samples.

Table 4. Mean and standard deviations of CIE Lab values and translucency parameter (TP) of infiltrated samples.

*Dissimilar superscripts denote statistically significant differences between the groups.

Sample	L	a	b	ΔE	Translucency Parameter (TP)
Control	84.64 ± 0.41a	−0.59 ± 0.11a	0.87 ± 0.26a	-	2.70 ± 0.47a
1C–9Y	80.41 ± 0.83b	−1.21 ± 0.18b	−1.43 ± 0.38b	4.85 ± 0.03a	2.14 ± 0.79a
3C–7Y	78.48 ± 0.34c	−1.56 ± 0.06c	3.47 ± 0.5c	6.76 ± 0.05b	2.49 ± 0.53a
5C–5Y	77.89 ± 0.57c	−1.63 ± 0.15c	4.92 ± 0.32d	7.94 ± 0.06c	2.21 ± 0.72a
7C–3Y	76.1 ± 0.36d	−1.74 ± 0.14c	0.49 ± 0.33a	8.63 ± 0.04d	2.38 ± 0.61a
9C–1Y	75.15 ± 0.29e	−1.26 ± 0.21b	3.36 ± 0.38c	9.83 ± 0.07e	2.09 ± 0.60a
Tooth	64–78	−2.6−2.7	10−25	-	-

Fig. 9. Ceria-yttria infiltrated zirconia samples.

4. Discussion

As the objective of the study was to evaluate the effect of ceria-yttria coloring liquid on clinically relevant properties of dental zirconia, the following section discusses the results of phase analysis with microstructural-property correlation for flexural strength, fatigue testing, and fractography, with the results of the CIE lab values, translucency, surface roughness, and aging resistance.

4.1. Phase analysis and microstructural characterization

The phase analysis in Fig. 2 reveals the retention of the tetragonal phase of zirconia in all the ceria-yttria-infiltrated zirconia samples. The findings are similar to the reports in the previous studies of ceria-infiltrated zirconia [8,19–22]. These results indicate ceria and yttria additions were not detectable by X-ray diffraction as they did not form solid solutions with zirconia.

The SEM images showed the largest increase of the grain size in 1C-9Y of 5.34 μm as seen in Table 3 and Fig. 3(b), while the control zirconia showed a grain size of 0.81 μm in Fig. 3(a), and the other infiltrated samples, 3C-7Y, 5C-5Y, 7C-3Y and 9C-1Y showed closely similar values, as shown in Fig. 3(c-f) and Table 3. Energy dispersive analysis (EDS) confirmed the presence of cerium, yttria, and zirconia in the infiltrated sample in Fig. 3(g). It is known that ceria and yttria tend to increase the grain size of zirconia, which makes zirconia restorations susceptible to monoclinic transformation, resulting in premature fractures [8,25,37]. In a similar study by Shah et al., a three-fold increase in the grain size of ceria-infiltrated and a ten-fold increase in bismuth-infiltrated zirconia were observed compared to the control unpigmented zirconia. This was attributed to the vaporization of the solvent and the melting point of the infiltrate solution [8]. In the present study, there was a moderate increase of one to two-fold in all the infiltrated samples (0.98–1.67 μm) except for 1C-9Y, with higher yttria content, which showed a seven-fold increase in the grain size (5.34 μm), compared to the control unpigmented zirconia (0.81 μm). This could be attributed to the influence of the amount of yttria on the grain size of zirconia. Typically, in a 3 mol % yttria-stabilized zirconia, the tetragonal phase is meta-stabilized at room temperature, which limits grain growth during sintering. However, as the amount of yttria increases to 4–5 mol%, more of the zirconia is stabilized in the cubic phase, which does not restrict grain growth during sintering, resulting in increased grain size [22].

Furthermore, the critical grain size of zirconia must be within 1 micrometer (μm) [48]. Above the critical grain size, zirconia tends to be less stable and more inclined to spontaneous tetragonal-to-monoclinic transformation. Smaller grain sizes of less than 1 μm are related to a lower transformation proportion. The infiltration depth measured from the SEM images in Fig. 4(a) and 4 (b) using ImageJ software ranged from 37.6 to 142.4 μm. The observed infiltration range of ceria-yttria may be attributed to variations in porosity defects within the zirconia, which could be a limitation of the immersion technique.

4.2. Surface roughness

There was no statistically significant difference in the surface roughness values between the infiltrated samples, except for 5C-5Y, as can be seen in Table 2 and Fig. 5. The sample 5C-5Y exhibited the lowest surface roughness, which correlates with its relatively higher flexural strength amongst the infiltrated samples. However, despite the high flexural strength of the control unpigmented zirconia, it showed higher surface roughness values compared to the infiltrated samples. The infiltrated solutions of ceria-yttria would have penetrated the porous structure of the sample from the surface to its interior, resulting in lowered surface roughness values. This finding can be attributed to a surface modification achieved through the infiltration process [12].

4.3. Mechanical property characterization

4.3.1. Flexural strength

Studies using ceria-based liquid infiltration of 1, 5, and 10 wt% of cerium acetate and cerium chloride solutions showed reduced flexural strength of zirconia [8,19–22]. Typically, the strength of zirconia decreases with an increase in ceria content (without yttria addition). Interestingly, ceria-yttria infiltrated zirconia reported a varied trend in strength. Previous studies reported that increasing ceria content improved the strength of zirconia until yttria content was less than 2 mol%. However, when the yttria was greater than 2 mol%, its strength decreased with further addition of ceria. A similar trend was observed in the present study, as shown in Table 3. Despite an increase in ceria content up to the threshold of 5 wt% of ceria, the flexural strength of zirconia improved with the addition of yttria until below 5 wt%. The enhanced flexural strength of 5C-5Y (521 MPa), though less than that of control zirconia (704 MPa), was not statistically significantly different from the control unpigmented zirconia. This could be due to the grain size refinement and tetragonal stability by yttria addition till 5 wt%. As can be deciphered from SEM images in Fig. 3(a-f) and grain size values in Table 3, the smaller grain size of control zirconia and the experimental 5C-5Y showed higher strength compared to the larger grain sizes, with reduced strength of 1C-9Y. Furthermore, an increase in the number of defects in zirconia has been associated with increased grain size, with a reduction in its mechanical properties, as can be deciphered from Table 3 [39,48].

The flexural strength values of 3C-7Y, 7C-3Y, and 9C-1Y were comparatively similar, but less than that of control zirconia and 5C-5Y (Table 3), and had a statistically significant difference from the control zirconia. There was a 26 % reduction in the flexural strength of 5C-5Y (521 MPa) and a 60 % reduction for 1C-9Y (420 MPa) compared to the control zirconia (704 MPa). The flexural strength of the infiltrated samples meets the requirements of ISO standard 6872, which recommends a minimum flexural strength of 50–100 MPa for Class I dental ceramics, which are indicated for monolithic single-unit anterior crowns, veneers, inlays, and onlays [49]. Factors such as the nature of cerium solutions, immersion time, and the type of zirconia influence the mechanical properties. The use of cerium nitrate infiltrate, providing better dissolution and a short immersion time of 1 min, could have contributed to the improved strength of 5C-5Y, due to a shorter duration of interaction and the lower concentration of pigments.

4.3.2. Cyclic fatigue

Fatigue testing of dental restorative materials is crucial because the oral environment exposes them to constant moisture and repeated mechanical stress from chewing. Therefore, fatigue behavior serves as a key indicator of the time-dependent performance of dental materials [39,49]. The fatigue plots in Fig. 6(a) and Fig. 6(b) present the run-out and failed samples of each of the groups. The samples of 5C-5Y were chosen due to their high flexural strength among other infiltrated samples. To endure the fatigue effect of masticatory forces, the endurance limit of zirconia as a restorative material should be 40–50 % of its flexural strength, which would be in the range of 300–600 MPa [50]. As can be interpreted in Fig. 6(a) and 6 (b), until 85 % of static flexural strength (600 MPa), the control zirconia samples showed an endurance limit without failure, while the experimental 5C-5Y sustained 71 % of its static flexural strength (370 MPa), the values of which are within the acceptable range. Scatter was observed in control and 5C-5Y zirconia concerning the load failures occurring at high and low numbers of cycles, which could be attributed to microstructural flaws and defects in zirconia [51].

It is well established that the stress required to propagate a crack through a crystalline structure is associated with the grain size [52,53]. Small grains restrict dislocations along the grain boundaries, thus influencing the crack propagation between the grains. Therefore, higher stress levels are required for initiating fracture [52,53]. The lowered fatigue performance of the experimental group, 5C-5Y, can be attributed to localized areas with pigment concentrations associated with larger grain sizes, thereby reducing their fatigue strength [54–56]. Considering the flexural strength and fatigue performance of pigmented 5C-5Y of 521 MPa, these values were close to the reported results of 5 wt% cerium acetate and cerium chloride infiltrated zirconia and commercially available glass-ceramics.

4.4. Fractographic analysis

Fractured samples after static flexural strength of the infiltrated samples revealed the compression curls, hackle lines, arrest lines, and crack origin. These features are illustrated in Fig. 7(a-c). Due to the nature of the flexural strength test, the origin of the crack typically originates from the centre of the sample on the tensile side (bottom) during 3-point flexural testing, as shown in Fig. 7(b). No single point of origin was observed in the fractured samples. Fractured samples after fatigue testing of the control unpigmented zirconia showed surface porosities, which could be related to processing defects. On the other hand, the 5C-5Y showed crack propagation across, ending at the compression curls.

4.5. CIE lab, ΔE values & Translucency Parameter

The CIE Lab color space models human visual perception and describes color using three orthogonal parameters, L, a, and b. The L coordinate represents lightness, ranging from 0 (perfect black) to 100 (perfect white). The a-axis quantifies the color’s position on the red–green spectrum, where positive values indicate a shift toward red and negative values indicate green. Similarly, the b-axis corresponds to the yellow–blue axis, where positive values denote yellow and negative values represent blue. The central point (a = 0, b = 0) represents a neutral gray shade, while the distance from this origin indicates the chroma or color saturation. The a–b plane thus serves to visualize hue and saturation, while L defines how light or dark the color appears [57]. Fig. 8 depicts the CIE Lab color space in two dimensions, plotting a and b along the X- and Y-axes, respectively. Fig. 9 depicts the colored infiltrated samples. The CIE Lab values of the infiltrated samples showed a trend of decrease in the lightness in Table 4. Higher delta E values were observed along with a reduction in the L value (brightness), which could be associated with an increased cerium content.

ΔE values for the 9C-1Y and 5C-5Y samples were similar to earlier studies on cerium acetate at 5 % and 10 %, and cerium chloride at 1 % and 5 % [8]. In contrast to the cerium acetate and cerium chloride solutions used in the previous studies, cerium nitrate was employed in the present study. Cerium nitrate had a coloring impact that was comparable to that of cerium acetate but was less than that of cerium chloride. Typically, natural tooth shades have Lab values within the range of L from 64 to 78, a values from - 2–2.7, and b values from 10 to 25 [58]. In comparison, the cerium-yttria infiltrated zirconia samples in the present study exhibited lightness values between 75 and 85, a values from −0.5 to −1.8, and b values from −1.4–5. Although the lightness closely resembles that of natural teeth, further improvements are needed along the blue-green and red-yellow axes for better color matching for ceriay-ttria infiltrated zirconia. Chromium or bismuth oxide pigments can be further incorporated to enhance the blue-green and red-yellow tints [23].

The translucency parameter (TP) values are presented in Table 4. The TP values of the infiltrated samples showed no statistically significant difference within the groups or with the unpigmented control zirconia. The zirconia used in this study was 3 mol% yttria-stabilized zirconia, which is typically characterized by low translucency with a larger grain size. This remains one of the limitations of 3 mol% yttriastabilized zirconia. To achieve optimal translucency, a material thickness of 2 mm would necessitate a grain size of 70 nm, while a thickness of 1.3 mm would require a grain size of 82 nm [59]. A significant change in translucency demands a grain size modification of less than 70–80 nm. In the present study, though there was an observed increase in the grain size in the infiltrated samples, there was no difference in the TP values. This could be because the grain size of unpigmented control zirconia was 0.81 μm (~810 nm), and that of 5C-5Y was 0.98 μm (~980 nm); the grain size of ceria-yttria infiltrated zirconia was much larger than 80 nm of translucent zirconia. Thus, it can be inferred that pigmentation through ceria-yttria infiltrate solutions does not enhance the translucency of unpigmented zirconia; rather, it retains its translucency.

4.6. Aging resistance

The monoclinic content before and after aging was not detectable in the infiltrated samples except for 1C-9Y, as shown in Fig. 10. This exception could be attributed to the higher content of yttria in 1C-9Y (9 wt% of yttria), contributing to aging degradation with an increase in monoclinic content. Aging resistance is largely influenced by processing parameters and the resulting microstructure. Additionally, the tetragonal-to-monoclinic phase transformation in aqueous environments is primarily associated with the presence of oxygen vacancies. In yttria-stabilized zirconia, the formation of oxygen vacancies plays a critical role in balancing the charge difference between trivalent Y³⁺ and tetravalent Zr⁴⁺. Cerium as a phase stabilizer reduces the formation of oxygen vacancies, thereby improving the aging resistance of zirconia [44,60]. In the present study, all the infiltrated ceria-yttria samples demonstrated negligible monoclinic content post-aging except for 1C-9Y.

Fig. 10. XRD plots of pre-aged and post-aged ceria-yttria infiltrated zirconia samples, 1C-9Y, 3C-7Y, 5C-5Y, 7C-3Y and 9C-1Y and control unpigmented zirconia.

The high yttria coating with lesser ceria in 1C-9Y could have created oxygen vacancies and pathways for the transport of water-related ions from the oral cavity, leading to aging degradation. Furthermore, the high yttria content due to its unstable nature of larger grain size with 5.34 μm (Table 3) could have led to spontaneous tetragonal-monoclinic transformation [28].

The influence of ceria-yttria infiltrate coloring liquid on 3 mol% yttria-stabilized zirconia has been investigated in the present study. Among the ceria-yttria infiltrated groups, 5C-5Y can be recommended for optimal clinical outcomes as it demonstrated a flexural strength close to 521 MPa, with an endurance limit of 71 % of flexural strength, enhanced aging resistance, and CIE Lab values close to natural teeth. Steps were taken to minimize variability within the infiltrated groups by sourcing all zirconia samples from the same manufacturer, maintaining a uniform immersion time to ensure consistent submergence in the infiltrate solution, and calibrating the furnace equipment.

Future research shall explore the impact of ceria-yttria coloring liquid on translucent zirconia, pre-clinical animal studies for toxicity and biocompatibility tests, followed by clinical trials. Considering the clinical relevance of pigmentation techniques, both pre-shaded zirconia and zirconia colored through infiltration have distinct properties and clinical advantages. While pre-shaded zirconia is made by incorporating metal oxides during manufacturing, the infiltration method adds colorants, such as rare earth and transition metals, to the porous zirconia surface and further sintering. The infiltration technique offers superior esthetic customization and is preferred for highly individualized or anterior cases, though it requires greater technical expertise and time. On the other hand, pre-shaded zirconia is well-suited for routine cases due to its ease of use and simplified workflow, making it ideal for posterior restorations [6,23]. The choice between the two techniques depends on the clinical situation, aesthetic demands, and available laboratory resources.

5. Conclusions

Pigmentation was accomplished by infiltrating pre-sintered zirconia with a ceria-yttria-based coloring liquid. Ceria-yttria infiltrated zirconia exhibited reasonable flexural strength and fatigue performance with improved aging resistance, and color suitable for anterior and low-stress-bearing restorations with minimal change in translucency and reduced surface roughness.