Impact Resistance of Yttria- and Ceria-Doped Zirconia Ceramics in Relation to Their Tetragonal-to-Monoclinic Transformation Ability

Abstract

In this work, the impact resistance of three zirconia ceramics was investigated: two yttria-stabilized zirconia (3Y-TZP and 1.5Y-TZP) and a ceria-stabilized-zirconia (Ce-TZP) composite. The impact resistance was evaluated through drop-ball impact tests on disk-shaped samples. The results are discussed in terms of the materials’ transformability, which was correlated to the size of tetragonal-to-monoclinic (t-m) transformation zones observed after the impact tests and to the volume fraction of the monoclinic content on fractured surfaces. The findings show that impact resistance increases with the ability of the material to undergo t-m transformation. The Ce-TZP composite exhibited the highest transformability and consequently the highest impact resistance, followed by 1.5Y-TZP, and then 3Y-TZP.

1. Introduction

Zirconia ceramics are widely used in various industrial and biomedical applications due to their remarkable mechanical properties, biocompatibility, and wear resistance. The high toughness of zirconia, when compared to other oxide ceramics, is largely attributed to its stress-induced phase transformation from the tetragonal (t) to monoclinic (m) phase. This t-m phase transformation is accompanied by a volume expansion (~5%), resulting in compressive stresses, which lead to crack shielding from applied tensile stresses. Consequently, this contributes to the toughening of zirconia ceramics by increasing their toughness or crack propagation resistance [1,2,3].

Ce-TZP and Y-TZP are zirconia ceramics stabilized in the tetragonal phase by the addition of CeO₂ and Y₂O₃, respectively. In the biomedical field, Y-TZP is favored for its aesthetic qualities and high mechanical strength, which may exceed 1 GPa, typically for ceramics containing 3 mol.% of yttria. However, this material exhibits moderate fracture toughness, ranging from 4 to 6 MPa$\sqrt{m}$ [4,5,6]. In contrast, Ce-TZP ceramics offer higher toughness but lower strength (~500 MPa for zirconia stabilized with 12 mol.% ceria) [7,8,9,10] due to grain coarsening during sintering.

Numerous studies have explored strategies to achieve a better balance between strength and toughness in TZP ceramics [3,6,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25]. Certain approaches involve the development of finer-grained Ce-TZP composites to enhance strength [3,6,9,15,19,21,26]. Another approach focuses on reducing Y₂O₃ content in Y-TZP ceramics to increase the material’s ability to undergo t-m transformation, thereby improving toughness without significantly compromising strength [27,28,29,30,31,32].

While the strength-toughness relationship in zirconia ceramics is well documented, less emphasis has been placed on impact resistance and energy absorption capacity. These properties, which are important for ballistic or shock-resistance applications, have been understudied and most previous investigations were focused on resistance to impact at very high speeds, known as ballistic impact loading [33,34,35,36,37,38,39,40]. In this study, we present a comparative analysis of the impact resistance of three TZP-based zirconia ceramics with increasing degrees of transformability: a standard 3Y-TZP zirconia, which exhibits transformation toughening but a quite modest crack-tip transformation zone size; a newly developed yttria-doped zirconia with less yttria content (1.5Y-TZP, doped with 1.5 mol.% yttria), which offers an attractive combination of toughness and strength [41]; and a Ce-TZP composite, showing a high transformation ability, a large transformation zone size, and, to a certain extent, some transformation-induced plasticity and ductility [9]. The aim of this work was also to clarify the question: ‘Is strength or toughness the most relevant figure of merit when designing a shock-resistant ceramic?’.

2. Materials and Methods

Three zirconia-based ceramic materials were used in this study: 1.5 mol.% and 3 mol.% yttria-stabilized zirconia, denoted by 1.5Y-TZP and 3Y-TZP, respectively, and a Ce-TZP composite consisting of 84 vol.% 11Ce-TZP, 8 vol.% Al₂O₃, and 8 vol.% SrAl₁₂O₁₉ aluminate platelets. The 1.5Y-TZP material was supplied by TOSOH (Zgaia 1.5Y-HT, Tokyo, Japan) and contained 0.25 wt.% Al₂O₃; the two other materials were provided by DOCERAM (Dortmund, Germany). All the samples were received in the shape of sintered disks with a diameter of 20 mm and a thickness of 1 mm. They were mirror-polished using diamond pastes (in the following order of grades: 6, 3, and 1 μm) and a colloidal silica suspension (0.03 μm). Density was measured using the Archimedes method in water [42] and the percentage of theoretical density (TD, from rule of mixture) was reported. The microstructure of the studied ceramics was investigated using Scanning Electron Microscopy (SEM; Zeiss SUPRA VP55, Oberkochen, Germany) at 1 kV on fractured surfaces. The zirconia grain size was estimated from the SEM micrographs from a minimum of 150 grains using ImageJ software (Version 1.54). Vickers hardness was measured by Vickers indentation (FV-700 tester, Future-Tech, Kanagawa, Japan) on mirror-polished samples with a load of 1 kgf applied for 10s (Hv₁).

To evaluate the impact resistance of the studied materials, drop-ball impact tests were performed using a device developed in a laboratory, a schematic representation of which is shown in Figure 1. The mirror-polished surface of the disk-shape samples was placed on three balls equidistant from its center, arranged in a circle of 16 mm in diameter (as would be the case in biaxial bending tests). A tungsten carbide ball of 26.5 g weight and 15 mm in diameter was then dropped onto the center of the disk. The drop heights were chosen between 10 cm and 40 cm (with increments of 5 cm or 2.5 cm for more precision) to include, for all three ceramics, heights where no failure was observed towards heights for which all samples broke. For each material, six samples were tested at each drop height. The results were then analyzed in terms of failure probability, calculated as the ratio of the number of fractured samples to the number of tested samples under the same conditions (i.e., six samples), versus height. For each material, a critical height (H_C)—corresponding to the drop height at which at least one sample among the six fractured (non-null failure probability)—was determined.

The impact energy, E_imp, was calculated considering only the potential energy, i.e., ignoring the loss due to air resistance, as follows:

E_imp = mgH

(1)

where m is the mass of the tungsten carbide drop ball (26.5 g), g is the gravitational acceleration (9.8 m/s²), and H is the drop height.

The absorbed energy, E_a, during impact tests was calculated, as follows:

E_a = mg(H − H_B)

(2)

where H_B is the bounce height, which was measured using a high-speed camera.

To determine the phase composition and assess the transformability of the studied materials, X-ray Diffraction (XRD) analysis was conducted on mirror-polished and fractured surfaces using CuKα radiation (1.5418Å) in the angular domain 2θ ranging from 27 to 33° (D8 Advance Bruker AXS diffractometer, Billerica, MA, USA). The mass fraction of the monoclinic phase (X_m) was calculated according to the following expression [43]:

$${\mathrm{X}}_{\mathrm{m}}={\displaystyle \frac{{\mathrm{l}}_{\mathrm{m}}(\stackrel{\mathrm{-}}{1}11)+{\mathrm{l}}_{\mathrm{m}}\left(111\right)}{{\mathrm{l}}_{\mathrm{m}}(\stackrel{\mathrm{-}}{1}11)+{\mathrm{l}}_{\mathrm{m}}\left(111\right)+{\mathrm{l}}_{\mathrm{t}}\left(101\right)}}$$

(3)

where I_m (hkl) denotes the area of the peak generated by the hkl plane in the monoclinic (m) or the tetragonal (t) phase.

The volume fraction of the monoclinic phase (V_m) was then calculated according to [44]:

$${\mathrm{V}}_{\mathrm{m}}={\displaystyle \frac{1.311{\mathrm{X}}_{\mathrm{m}}}{{1+0.311\mathrm{X}}_{\mathrm{m}}}}$$

(4)

The bottom (tensile) side of the samples was observed for signs of t-m phase transformation zones induced by impact tests using optical microscopy (ZEISS Axiophot, Oberkochen, Germany) as well as 3D confocal microscopy, employing a S-Neox (Sensofar, Barcelona, Spain) 3D non-contact optical profiler machine. which combines confocal and interferometry techniques with a lateral resolution of 0.26 μm. The acquired data were processed using Mountains Map Universal software^® (Digital Surf, Besançon, France).

3. Results

3.1. Microstructure and Properties

The density, Vickers hardness, and the zirconia grain size are presented in

Table 1. Summary of density, hardness, and zirconia grain size of sintered disks.

Sample	1.5Y-TZP	3Y-TZP	Ce-TZP Composite
Density (% TD)	99.7	99.7	99.7
Hardness (Hv1)	1180	1305	1000
Zirconia grain size (µm ± Sd)	0.4 ± 0.1	0.7 ± 0.1	0.9 ± 0.4

. SEM micrographs of fractured surfaces are show in Figure 2. All materials exceeded 99% of the theoretical density, indicating almost full densification. The Vickers hardness falls within the typical range for Y-TZP and Ce-TZP composites [45,46,47], with a lower value for the latter. The initial monoclinic phase content on polished surfaces was less than 5 vol.% for all studied materials.

3.2. Critical Drop Height

Figure 3 summarizes the results of the impact tests in terms of failure probability as a function of drop height. It can be observed that the range of drop heights over which the risk of fracture under impact is 100% is higher for 1.5Y-TZP and Ce-TZP composite (35 cm) compared to 3Y-TZP (25 cm). Both 1.5Y-TZP and Ce-TZP composite show similar drop heights for failure probabilities of 0.5 (30 cm), whereas this height is significantly lower for 3Y-TZP (~17.5 cm). Figure 4 shows the critical drop heights, H_C, determined for the studied materials. The Ce-TZP composite exhibited the highest H_C value of 30 cm. An intermediate value of 22.5 cm was observed for the 1.5Y-TZP zirconia, whereas the lowest value of 17.5 cm was observed for the 3Y-TZP zirconia.

3.3. Absorbed Energy

Figure 5 shows the absorbed energy, E_a, during impact testing at drop heights of 10, 15, 20, and 25 cm. For the 1.5Y-TZP and 3Y-TZP materials, only rebound measurements performed below their respective critical drop heights were considered for E_a calculations. For all the materials, the absorbed energy increased as drop height increased. At the smaller drop heights of 10 cm and 15 cm, 1.5Y-TZP and 3Y-TZP absorbed roughly the same amount of energy, whereas the Ce-TZP composite absorbed significantly more energy at all drop heights.

3.4. Impact-Induced Phase Transformation

Optical observations of the tensile side of impacted and unbroken samples revealed the presence of t-m transformation zones in 1.5Y-TZP and the Ce-TZP composite, while no phase transformation was detected in the impacted 3Y-TZP zirconia. Figure 6 shows typical optical micrographs of the impact-induced t-m phase transformation zones for different drop heights. For heights above H_C, only unbroken tested samples were observed. The transformation zones exhibited a star-shaped geometry, with a circular central transformed region from which radial bands extended towards the edge of the disk. The area of the transformation zone was evaluated by considering the transformed area within a circle corresponding to the effective surface in biaxial tests, as in [48]. Figure 7 shows the evolution of the impact-induced transformation zone area with drop height for the 1.5Y-TZP and the Ce-TZP composite. For both materials, the size of the transformation zone increases linearly with drop height (Figure 7), with consistently higher transformed areas for the Ce-TZP composite, which reached almost five times that of 1.5Y-TZP at a drop height of 30 cm (i.e., around 2.5 and 0.5 mm², respectively).

3D confocal microscopy observations confirmed the t-m phase transformation during impact testing and allowed quantitative analysis of the topography of impacted areas. Volume increase related to the t-m transformation (i.e., around 5 vol.%) was evidenced by circular dome-shaped zones, typical images of which are shown in Figure 8, where radial roughness profiles are also displayed. The effect of the impact on the t-m transformation extended over the sample (at least 5 mm in diameter around the impact point), as evidenced by the absence of a flat area in the 3D confocal images and analyses. For 1.5Y-TZP, roughness profiles revealed an increase of 2.4 µm in the maximum height of the transformation zone from 15 to 25 cm drop heights (1 and 3.4 µm, respectively). Similarly, for the Ce-TZP composite, the maximum height of the transformation zone increased by 9 µm over the same range of drop heights (i.e., from 9 µm at 15 cm to 18 µm at 25 cm).

Evidence of t-m phase transformation in the studied materials during impact tests was also provided by XRD analysis of polished and fractured surfaces, as shown in Figure 9. The monoclinic phase content, (V_m), estimated on fractured surfaces after impact testing, was 80% for 1.5Y-TZP zirconia and the Ce-TZP composite, and 30% for 3Y-TZP zirconia.

4. Discussion

The phase transformation observed after impact testing of 1.5Y-TZP and Ce-TZP composite disks was induced by the stresses generated by the highly localized contact force between the ceramic surface and the WC drop ball, which is harder (Hv~2400) and stiffer (E = 500–700 MPa). The star-shaped transformation zones induced during impact tests (Figure 6) are similar to those previously observed for these materials on the tensile side of biaxial piston-on-three-ball test samples (i.e., in the center of the disk, over the piston, where the stresses are the highest) or three-ball-on-three-ball tests (i.e., under the three closest balls in contact with the surface under compression, but on the side under tension) [9,41,48,49].

The results revealed a strong correlation between impact resistance, expressed either by the critical drop height (H_C) or the absorbed energy (E_a), and the stress-induced t-m phase transformation of the materials. The Ce-TZP composite, with the largest impact-induced transformation zones (Figure 6 and Figure 8), exhibited the highest critical drop height compared to the Y-TZP materials (Figure 4). Among the latter, the under-stabilized 1.5Y-TZP, for which impact-induced t-m phase transformation occurred, exhibited a higher critical drop height than 3Y-TZP, for which no transformation was observed on impacted surfaces.

The significantly high energy absorption of the Ce-TZP composite, even at low drop heights (Figure 5), can also be attributed to the high transformability of this material, which is known to provide a ductile-like behavior to these composite ceramics under quasi-static bending and tensile testing [48]. In previous works [10,48], it has been shown that this material and other similar composites [6,9,10,50] exhibited low critical t-m transformation stresses, σ_c^(t-m), compared to their flexural strengths, allowing t-m phase transformation to occur at relatively low stresses (e.g., σ_c^(t-m)~300–400 MPa) before any crack propagation, inducing nonlinear behavior (~0.4% plastic strain at failure) and high energy dissipation before fracture. The plasticity in these Ce-TZP composites is exclusively linked to the stress-induced t-m zirconia phase transformation, as no microcracking and/or damage occurred during plastic deformation under flexural loading [10]. At low drop heights (10–20 cm), no significant difference was observed between the energy absorption in 3Y-TZP and 1.5Y-TZP, as the t-m phase transformation zone was not sufficiently developed in the latter under such conditions. This agrees with the high critical t-m transformation stress of 1.5Y-TZP, which is of the same order as its flexural strength, σ_R (i.e., σ_c^(t-m) = σ_R~1000 MPa), placing it at the transition between brittle and ductile behavior [41]. This characteristic also explains the wider range of drop heights over which the risk of fracture under impact increases from 0 to 100% for 1.5Y-TZP (Figure 3). At low drop heights, the transformation zones in this material were not sufficiently developed, and its behavior was close to that of 3Y-TZP (brittle-like). However, it then becomes similar to that of the Ce-TZP composite at high drop heights (ductile-like), at which its t-m transformation becomes important.

After sintering and polishing, mainly tetragonal zirconia was detected by XRD (Figure 9a). The volume fraction of the monoclinic phase was estimated to be 3% for 3Y-TZP and 5% for both 1.5Y-TZP and the Ce-TZP composite, related to transformation at the surface of the ceramics due to applied polishing stresses. The ability to undergo t-m phase transformation under stress, estimated from XRD performed on as-fractured surfaces (Figure 9b), was much higher for 1.5Y-TZP and the Ce-TZP composite (80 vol.% monoclinic phase) compared to the less impact-resistant material, 3Y-TZP (only 30 vol.%). It is worth noting, however, that the amount of monoclinic phase on fractured surfaces reached the same value for 1.5Y-TZP and Ce-TZP composite materials (80%), despite the higher impact resistance of the Ce-TZP composite. This is due to the limitations of XRD in assessing the extent of monoclinic phase transformation in zirconia, as X-rays only penetrate to a depth of ~5 µm [51]. Microstructural analysis of cross-sections should be carried out to better characterize the volume affected by t-m phase transformation under impact, but the transformability is already well described and quantified from topography analyses of the transformation areas.

As outlined by several authors [52,53,54,55,56,57,58], the fracture toughness (K_IC) is a key material parameter affecting the ballistic resistance of ceramic materials, as it measures the resistance to crack initiation. For the studied materials (

Table 2. Impact resistance parameters (H_C and absorbed energy at H_C, E_ac), strength (σ_R), and fracture toughness (K_IC) of the studied materials.

Material	HC (cm)	Eac (J)	σR * (MPa)	KIC * ($\mathbf{MPa}\sqrt{\mathbf{m}}$)
3Y-TZP	17.5	1.64	830	6.1
1.5Y-TZP	22.5	2.21	1000	8.5
Ce-TZP composite	30	5.02	600	10.2

* Determined using the Single Edge V-Notched Beam (SEVNB) method in previous works [41,48,59].

), the higher the fracture toughness, the higher the impact resistance, following a linear trend, as shown in Figure 10a. In contrast, no direct relationship was observed between H_C and strength, (σ_R), previously determined from four-point bending tests (Figure 10b) [41,48,59]. In particular, the material most resistant to impact (i.e., the Ce-TZP composite) displayed the lowest value of strength, indicating that strength has a weaker correlation with impact resistance than fracture toughness under the conditions of this study.

The correlation between K_IC and impact resistance can be understood by considering the positive influence of stress-induced t-m phase transformation on both properties. On one hand, crack shielding due to this transformation decreases the effective applied stress at the crack front, likely to propagate, which leads to improved fracture toughness. The size of the transformation zone around the crack (h) and the volume fraction of the monoclinic transformed phase are key parameters for transformation toughening, as previously verified for the studied materials during crack propagation under flexural loading [41,48]. It has been shown that h, typically about 5 µm for 3Y-TZP (known for its brittle behavior), reaches 20 µm for 1.5Y-TZP and 100 µm for the ductile Ce-TZP composite, which exhibited a significant crack growth resistance (rising R-curve behavior). On the other hand, the effect of t-m phase transformation on impact resistance can be attributed to the increased energy dissipation associated with the material’s ability to exhibit t-m phase transformation and ductile behavior. This finding is consistent with the impact-induced t-m phase transformation results, which highlight the intermediate behavior of 1.5Y-TZP, allowing it to perform relatively well in terms of impact resistance, though not as effectively as the ductile Ce-TZP composite. It would be interesting to better characterize the t-m transformed zones induced by impact to verify whether, as observed during biaxial flexural tests performed on Ce-TZP composites, the t-m transformation takes place without microcracking. If this is the case, strategies for “regenerating” impacted ceramics could be explored by inverse m-t phase transformation, which would be achieved by increasing the temperature of the material.

5. Conclusions

Our work clearly highlights the role of stress-induced t-m zirconia phase transformation in the impact resistance of zirconia ceramics. The higher the transformability, the higher the ability to absorb energy upon impact and the higher the critical height for failure. The highest transformability was observed in the Ce-TZP composite, which had the highest critical height (30 cm), followed by 1.5Y-TZP (22.5 cm) and 3Y-TZP (17.5 cm). A linear relationship between the fracture toughness and critical drop height was observed in these zirconia-transformable materials. However, further studies on other zirconia-based compositions and/or composites are needed to validate this dependence. Conversely, strength measured in bending is not a relevant figure of merit for shock resistance, at least under the conditions of this study. These findings show the positive influence of stress-induced t-m zirconia phase transformation on the impact resistance of zirconia materials, suggesting the potential use of highly transformable materials in applications where shock resistance is of prime importance.