Vat Photopolymerization of Additively Manufactured Zirconia Ceramic Structures from Slurries of Surface Functionalized Particles: A Critical Review

Abstract

Zirconia (ZrO₂) ceramics and composites have attracted much attention in aerospace, biomedical and energy fields due to their high hardness, high wear resistance, excellent chemical stability and biocompatibility. However, the brittleness of ceramics and the high cost of molds have made it difficult for traditional processing techniques to manufacture complex structural and functional components efficiently. Additive manufacturing technology has successfully overcome these challenges by optimizing the preparation process and improving production efficiency. Among them, vat photopolymeriztion (VPP) has been demonstrated to offer distinct advantages, including high precision, high efficiency and low cost. It provides a novel approach to the preparation of zirconia ceramics. VPP preparation of zirconia ceramics and composites needs to consider various steps such as slurry preparation, structural design and printing, debinding and sintering. This review introduces common VPP technologies related to zirconia ceramics and summarizes the factors affecting the rheological and curing properties of zirconia slurry, in order to provide researchers with a reference for studying VPP preparation of zirconia. The current optimization methods for light-curing zirconia slurry formulations are focused on, and common methods for surface modification and optimization of slurry composition and solid loading are introduced. The influencing factors of the printing process are summarized, and the current research on surface texturing of VPP preparation and the influence of printing parameters on the performance and accuracy of the components are introduced. The effects of debinding/sintering processes on cured zirconia ceramics are also summarized. The applications of VPP zirconia ceramics and composites are proposed, especially for their use in biomedical and energy applications.

1. Introduction

Zirconium dioxide (ZrO₂) ceramics and composites have the advantages of high strength, high hardness and wear resistance, high fracture toughness, excellent chemical stability and biocompatibility [1], which makes them widely used in many fields such as aerospace, biomedical, mechanical production, nuclear power engineering, and chemical energy. In the biomedical field, due to good chemical stability, not easily corroded by acids and alkalis, good biocompatibility and resistance to physiological corrosion, can be used for the human body’s joints, dental implants, ear, nose and throat, and other hard tissues replacement and repair [2,3,4,5]. In the field of mechanical production and chemical energy, zirconia stabilized by the addition of yttrium oxide has sensitive electrical performance parameters and can be used for the preparation of solid oxide fuel cells (SOFCs) [6]. Nano zirconia can emit light when excited by ultraviolet light [7]. In addition, zirconia can be used for the preparation of cutting tools [8,9,10], sensors [11,12,13], catalysts [14,15,16], wear-resistant coating [17], thermal barrier coatings [18,19,20], inert substrates for nuclear power engineering [21], ionizing radiation detectors [7], etc.

ZrO₂ has different crystal structures at different temperatures. When the temperature is below 1170 °C, ZrO₂ presents the state of monoclinic phase. At temperatures ranging from 1200 to 2370 °C, ZrO₂ presents the state of tetragonal phase. At temperatures in excess of 2370 °C, ZrO₂ is in the state of cubic phase. Mechanical properties of ZrO₂ in monoclinic phase at room temperature are extremely unstable; this shortcoming, to a certain extent, limits the industrial applications of ZrO₂ ceramics. In 1972, Garvie and Nicholson [22] found that the doping of alkali-earth or rare earth metallic oxides such as calcium oxide, magnesium oxide and yttrium oxide at the zirconium site can stabilize the tetragonal phase of zirconium oxide to prevent its phase transformation to the monoclinic phase. to prevent it from transforming into monoclinic phase. In 1975, Garvie et al. [23] put forward the theory of phase transition toughening, proposing that room-temperature metastable t-ZrO₂ undergoes a stress-induced transformation to m-ZrO₂. The volume expansion generated in the process can absorb a large amount of energy to enhance its fracture toughness by more than four times. Zirconia remains stable in the cubic phase at room temperature when the stabilizer content is at a high concentration level. While its mechanical properties are reduced, higher doping levels result in greater concentrations of oxygen vacancies in cubic zirconia, thereby enhancing the material’s ionic conductivity. The stabilizers currently used are magnesium-, yttrium- and scandium oxides [24], especially yttrium oxide, which is the most used. For example, the fracture toughness of 3 mol.% Y₂O₃ tetragonal zirconia polycrystal (3Y-TZP) can reach 5–10 MPa·m^1/2, its flexural strength can reach 1200 MPa, and its compressive strength is greater than 2000 MPa. However, in a humid environment, tetragonal-phase zirconia spontaneously transforms into monoclinic phase, a phenomenon known as aging, hydrothermal degradation, or low-temperature degradation (LTD). This phenomenon results in the formation of microcracks, which consequently leads to a decline in mechanical properties. Some problems of LTD can now be mitigated by controlling the content of stabilizer types [25,26], choosing the right alloying or process control [27]. Furthermore, zirconia composite materials are also extensively utilized, particularly zirconia–alumina composite materials. ZrO₂ in zirconia-toughened alumina (ZTA) ceramics provides toughness and reliability. The presence of Al₂O₃ particles within a fine ZrO₂ matrix contributes to the enhancement of toughness, hardness, and strength in alumina-toughened zirconia (ATZ).

The traditional manufacturing methods of molding and shaping zirconia ceramics include slip casting [28,29,30], dry press molding [31,32], isostatic press molding [33,34,35], tape casting [36,37,38] and gel casting [39,40,41]. However, these traditional processes all require the use of complex molds, long development cycles and high processing costs. These processes are compounded by the inherent characteristics of ceramic materials, such as high hardness and brittleness. Moreover, the three-dimensional structure of ZrO₂ ceramic parts has become more and more complex, such as hollow, gradient, and micro/nano structures, and cannot be or is difficult to be prepared by traditional methods, which limits the structural/functional integrated applications of ZrO₂ ceramics in the field of high-end fine ceramics. Currently, a variety of additive manufacturing (AM) technologies such as selective laser sintering/melting (SLS/SLM) [42,43,44], direct ink writing (DIW) [45,46,47], fused deposition molding (FDM) [48,49,50], binder jet (BJ) [51,52,53], laser near-net-shaping (LENS) [54,55], vat photopolymeriztion (VPP) [56] have been widely used. Additive manufacturing can produce more complex structures with high material utilization. It is now widely developed in a variety of fields, including but not limited to aerospace, military defense, energy, biomedical and chemical engineering, and automotive [57]. Furthermore, 4D printing, which incorporates the effect of time on ceramic components, is also developing rapidly. Four-dimensional printing is a technique that uses special materials to deform specimens prepared by 3D printing under other specific conditions [58,59]. Wang et al. [60] realized 4D printing of ceramic structures by using inks with varying solid contents and exploiting the stress mismatch of printed ceramics during sintering. Although there are few studies on zirconia in 4D printing, 4D printing is still one of the possible directions for zirconia.

Compared with other additive manufacturing technologies, VPP uses laser beams or digital micromirror device (DMD) digital micromirrors to control the printing area, with high manufacturing precision, excellent surface quality, and economical raw material application, which has a greater advantage in the preparation of complex shapes and high-precision large parts [61]. Whereas oxide ceramics such as zirconia have an inherent advantage over non-oxide ceramics in VPP, coupled with the great potential of zirconia in fields such as biomedicine and energy, the overall research on zirconia preparation by VPP has increased year by year in recent years, as shown in Figure 1.

The properties of zirconia prepared by VPP are affected by various factors, such as slurry properties, printing parameters and heat-treatment process parameters. To provide researchers studying the preparation of high-performance zirconia by VPP with a reference, this paper reviews the literature on zirconia preparation using VPP technology. The review covers the entire process, from slurry optimization and printing strategies to debinding and sintering. Building on previous research and the advantages of additive manufacturing, the paper also presents examples of integrating slurry powders with component and surface engineering. As nearly 30 related papers were published in the first half of 2025, this review includes outstanding literature from that year, as well as studies from 2024 and earlier. This provides researchers with a comprehensive overview of the latest research developments.

2. Introduction to Vat Photopolymeriztion

VPP is an early-emerging 3D printing technology, and the basic process involves storing a light-curing resin in a container and treating it with UV or visible light. When the photoinitiator is exposed to a specific wavelength of light it breaks down to produce reactive substances that allow the formation of irreversible chains between monomers and oligomers [62]. Figure 2 shows the principle of VPP. Based on this principle, successive cured layers can be processed according to the STL file, layer by layer, and stacked to form the desired embryo. Compared with other additive manufacturing technologies, light-curing molding technology uses laser beams or DMD to control the printing area. This method has significant advantages in manufacturing components with complex structures and high precision requirements [63]. Advancements in technology have given rise to a range of VPP technologies, each with its own distinctive characteristics. These include stereolithography (SLA), digital light processing (DLP), liquid crystal display (LCD), and two-photon polymerization (TPP). The schematic diagram in Figure 3 illustrates the principle of each technology. The light-curing 3D-printing technologies of zirconia ceramics and composites are currently focused on SLA and DLP.

SLA was first invented by Hull and commercialized by 3DSystem in the USA. The laser emits a specific wavelength of light, which is controlled by the system and selectively irradiates the top layer of the photosensitive resin. When the table is close to the top layer of resin, the resin cures onto the table. This process starts with a point, moves to a line, and then forms a surface to create the first layer. The table then moves down and a new layer of photosensitive resin is added to the surface of the first layer. This process can be repeated until the final solid model is obtained. SLA was first used to prepare photosensitive resins for special shape requirements, and then in 1997, Griffith first combined light-curing molding technology with the shaping process of ceramic parts, and proposed the requirements for ceramic pastes based on light-curing technology [64]. Due to the extremely small size of the laser beam used in SLA technology, ceramic blanks prepared using this method are typically highly precise and uniform, with low internal stress. However, the disadvantage is that it is quite time-consuming.

Digital Light Procession (DLP) was developed from mask stereolithography (MSL) proposed by Nakamoto and Yamaguchi in 1996 [65]. The DMD was first designed and produced by Texas Instruments in 2001. Consisting of millions of micromirrors, the device can accurately project images and cure specific cross-sections [66]. The photoinitiators used in DLP technology and SLA technology both absorb single photons to initiate polymerization reactions. Both technologies belong to the category of single-photon polymerization [67]. The biggest difference from SLA is that DLP can project the whole layer cross-section geometry data directly into the whole area and cure the whole layer at the same time. The DLP technology as a fine ceramic forming method has been widely explored, and the performance of zirconia and alumina ceramics with a high density of 97–99% prepared by this method is approaching the level of those fabricated by conventional hot pressing.

LCD is a light-curing molding technology that belongs to the same MSL as DLP and uses a patterned mask to control the exposure area to control the cured shape [68]. Corresponding to the use of DMD for DLP, LCD technology uses liquid crystal display [69]. Despite its relatively low cost, LCD has a short lifespan and needs to be replaced periodically. In addition, the light intensity of the LCD is weak and uneven, affecting the printing accuracy. The bottom photosensitive resin also transitions exposure and requires regular cleaning of the liquid tank. Currently, there are fewer studies on the preparation of zirconia ceramics by LCD, basically in paste development [69,70,71] and print optimization [69].

Figure 3. Schematic illustration showing the principle of various VPP technologies: (a) stereolithography (SLA); (b) digital light processing (DLP); (c) liquid crystal display (LCD); and (d) two-photon polymerization (TPP) [72], copyright 2023, with permission from Royal Society of Chemistry.

Figure 3. Schematic illustration showing the principle of various VPP technologies: (a) stereolithography (SLA); (b) digital light processing (DLP); (c) liquid crystal display (LCD); and (d) two-photon polymerization (TPP) [72], copyright 2023, with permission from Royal Society of Chemistry.

Two-photon polymerization (TPP), an advanced VPP technology, is a micro- and nano-fabrication technique based on nonlinear optical effects that enables fabrication with sub-wavelength resolution [57]. Unlike single-photon absorption in SLA and DLP, molecules in two-photon polymerization absorb two photons from the low-frequency near-infrared to trigger subsequent reactions [73]. TPP was first applied to photosensitive resins, and urethane acrylate resin components with complex 3D microstructures were printed in 1997 [74]. The fact that only the submicron region at the focal point is capable of meeting the curing requirements of photosensitive resins allows for higher resolution than single photon absorption photopolymerization methods and enables curing inside the paste without the need for layer-by-layer fabrication [75]. For ceramic materials, they can be prepared by both TPP of organic-inorganic hybrid materials or ceramic precursors [76], in addition to the use of TPP to prepare organic structures then deposition of inorganic materials and etching of organic materials to prepare hollow ceramic parts [77]. At present, the preparation of ceramic parts by TPP, especially zirconia ceramics, is still at the stage of initial exploration. For example, Chai et al. [78] prepared organic-inorganic hybrid photoresist with silicate and zirconate to prepare microcrystalline glass containing t-ZrO₂ with nanodot structures.

3. Surface Weaving and Topology Design

Unlike the traditional way of subtractive manufacturing, the nature of 3D printing is additive manufacturing that builds layer by layer, requiring modeling and slicing of the modeling file before printing. By using CAD software such as AutoCAD2022 for modeling, it is possible to print components with unconventional structures in addition to preparing components with complex shapes.

Surface weaving is the processing of regularly arranged protrusions or depressions with a specific shape on the surface of a material [79]. Depending on the shape and size parameters, this special microstructure can serve to reduce the friction, increase the wettability, and improve the optical properties of the material surfaces [80]. Commonly used methods for processing surface textures include micro-milling, chemical etching, ion etching, photolithography, as well as direct laser ablation, laser cladding, and laser impact treatment using laser beams. These methods are limited by various factors such as materials design, processing accuracy, time and cost. With the development of VPP technology, research on the preparation of surface textures using the high degree of freedom method is gradually emerging. Several printed surface textures are shown in Figure 4 [81,82,83,84].

The current research on VPP to prepare zirconia with surface weaving is mainly focused on dentistry. The surface weave was utilized to improve the connection between materials. Ye et al. [82] prepared zirconia members with a bionic tree frog toe surface using VPP to enhance its shear bond strength (SBS) with resin cement. Hexagonal weaves with the same dimensions were designed on the surface of zirconia members to compare the effect of weaves with different heights and prism span on SBS. An increase in strut height causes SBS to increase and then decrease, while SBS decreases for strut spans greater than 250 μm. The maximum SBS of the bionic group with a prism span of 250 μm and a strut height of 250 μm is 22.7 ± 1.3 MPa, while the performance of the bionic group with a prism span of 250 μm and a strut height of 150 μm is the least affected by moisture–heat aging. Dai et al. [83], on the other hand, printed zirconia ceramics with hexagonal or square meshes on their surfaces and investigated the effect of grid depth on SBS. In particular, a hexagonal grid with a depth of 0.09 mm resulted in a 21% increase in the SBS value, which showed the best performance. When the depth reaches above 0.2 mm, the interface creates defects at the melt bonding and the role of the weaving is weakened. In addition, due to the larger inner corners and grid area of the hexagonal mesh compared to the square mesh, it is easier for the finish porcelain to penetrate into the hexagonal mesh during melt bonding, resulting in better long-term stability of the hexagonal mesh.

Figure 4. (a) Numerical model and SEM images of 3Y-TZP with different heights and spacing surface weaves [82], copyright 2024, with permission from Elsevier; (b) Design of zirconia ceramics with hexagonal or square lattice and sintered specimens [83], copyright 2023, with permission from MDPI; (c) Convex surface electrolyte structure [84], copyright 2023, with permission from Elsevier; and (d) Model of cellular structured electrolyte with different shapes [81], copyright 2024, with permission from Elsevier.

Figure 4. (a) Numerical model and SEM images of 3Y-TZP with different heights and spacing surface weaves [82], copyright 2024, with permission from Elsevier; (b) Design of zirconia ceramics with hexagonal or square lattice and sintered specimens [83], copyright 2023, with permission from MDPI; (c) Convex surface electrolyte structure [84], copyright 2023, with permission from Elsevier; and (d) Model of cellular structured electrolyte with different shapes [81], copyright 2024, with permission from Elsevier.

In addition, the shape of the surface weave can improve the surface area of the material. In solid oxide fuel cells (SOFC), zirconia electrolyte surfaces can be prepared into concave and convex surface structures and honeycomb surface structures [84] and honeycomb surface structures [81] to improve the cell performance. Gao et al. [85] prepared zirconia ceramic specimens with a honeycomb interlayer structure, and the flexural strength of the ceramic honeycomb interlayer structure could reach up to 170 MPa at a relative density of 41.72%.

Altering the structure of the building blocks, topology optimization can achieve lightweighting of the building blocks while ensuring their performance requirements and reducing the use of materials. Some VPP-prepared zirconia specimens with topological structures are summarized in Figure 5.

Zhao et al. [86] prepared octagonal truss-structured zirconia ceramics with different pillar sizes and cell counts by DLP, with the best performance of compressive strength of 75.3 MPa and energy absorption value of 6.76 × 10⁵ J·m⁻³. Pchelintsev et al. [87], on the other hand, used SLA to prepare Sc₂O₃-stabilized ZrO₂ building blocks with octagonal truss structure, which laid the foundation for the subsequent preparation of SOFC electrolytes. Chen et al. [88] designed 3Y-TZP ceramics with two-dimensional chiral structure, and the effects of structural and geometrical parameters on their mechanical properties were investigated, wherein the ceramics with the inverse tetrachiral structure had the best deformation capacity was the best and the energy absorption capacity was the best with a specific energy absorption (SEA) of 8.17 × 10⁵ J·m⁻³. Jiang et al. [89] prepared zirconia specimens with different triple-period miniaturized surfaces (TPMS) structures using DLP, and compared the effects of different structures on the osteogenic capacity and mechanical properties, and found that the two structures, IWP and Gyroid, had the best overall performance. Schwarzer-Fischer et al. [90] macroscopically combined zirconia and titanium oxide embryos with the designed structures by adjusting the solid content and other factors, and formed macroscopic zirconia/titanium oxide composite members after heat treatment.

Figure 5. (a) Model of octagonal truss structure [86], copyright 2020, with permission from IOP Publishing; (b) Image of zirconia ceramic blank with dot structure [87], copyright 2023, with permission from Elsevier; (c) Chiral structure [88], copyright 2023, with permission from Elsevier; and (d) Model of different types of triple-period miniaturized surfaces structures [91], copyright 2021, with permission from Elsevier.

Figure 5. (a) Model of octagonal truss structure [86], copyright 2020, with permission from IOP Publishing; (b) Image of zirconia ceramic blank with dot structure [87], copyright 2023, with permission from Elsevier; (c) Chiral structure [88], copyright 2023, with permission from Elsevier; and (d) Model of different types of triple-period miniaturized surfaces structures [91], copyright 2021, with permission from Elsevier.

4. Micro/Nano Structures and Micro/Nano Components

With the development of technology and practical needs, in addition to the conventional macroscopic structure design, the research on micro/nano structures and micro/nano components is gradually increasing. Micro/nano structures refer to structures with scales between micrometers and nanometers, whose unique size, surface, and quantum effects give them excellent physical, chemical, and optoelectronic properties that are not available in conventional materials. Liu et al. [92] replicated the structure of tooth enamel and prepared a ceramic coating on the metal surface. The coating is composed of zirconium oxide, characterized by a columnar nanostructure, with amorphous zirconium oxide surrounding it. The nanostructured coating has been demonstrated to enhance the mechanical properties of the components and improve their corrosion resistance. TPP, due to its principal advantages, has a resolution that can break through the diffraction limit and realize micro/nano fabrication. Figure 6 illustrates some of the micro/nanostructures.

Chai et al. [78] used TPP to prepare microcrystalline glass ceramics with nanodot structure and made a study on their properties. t-ZrO₂ and the characteristics of the micro-nano structure make the glass ceramics show excellent mechanical properties. The Young’s modulus of the octahedral truss structure was higher than that of the truss and octahedral structures, reaching 243 MPa. Octa-truss structures showed remarkable ability to withstand stress concentrations, maintaining structural integrity to resist rod bending under a stress of 27.1 MPa. Cao et al. [95] combined zirconia nanoclusters (Zr-NCs) and click-chemistry to prepare an organic-inorganic hybrid photoresist with a maximum lithography speed of 2.0 m·s⁻¹ and a minimum feature size of 59 nm for TPP. The photoresist was used to print recognizable micro 2D codes with high quality. A reference for high-speed preparation of 2D microstructures was provided. Sänger et al. [93] used TPP to print YSZ samples with complex 3D structures at a resolution of 500 nm. The measured compressive strength was 4.5 GPa, with similar properties while the density was significantly lower than that of conventional block YSZ samples. Later, Sänger et al. [94] used TPP to prepare YSZ specimens and investigated the mechanical response behavior of the specimens under compressive loading. Due to the use of zirconia ceramic nanoparticles, the number of grain boundaries in specimens increased. This, together with the effect of the print structure, resulted in an enhanced dissipative mechanical response of the specimens.

In addition, in ceramic micro- and nanostructures, the effect of shrinkage generated during debinding and sintering is more significant than in conventional structures. To address this issue, Desponds et al. [96] used zirconium oxide nanoparticles with a particle size of 5 nm instead of a fraction of zirconium acrylate precursor to reduce the loss and shrinkage during debinding. Cubic zirconia stabilized 3D microstructures with high spatial resolution and a significant reduction in weight and volume loss were prepared using this component. Microlens arrays with excellent focusing efficiency and a pitch of 0.8 μm were also prepared.

5. Optimization of ZrO₂ Ceramic Slurry

Since the ceramic slurry is the basis of VPP, the performance of ceramic slurry determines whether the printing process can be successful or not, and also has a great influence on the performance of the printed parts. A considerable part focuses currently on the modulation of slurry constituents to adjust the performance of the ceramic slurry. This section will summarize the current improvements to zirconia pastes.

5.1. Composition of ZrO₂ Ceramic Slurry

The most basic components of ceramic slurries for vat photopolymerzation are monomers, photoinitiators and ceramic powders [97]. During printing, photoinitiators are activated by light irradiation at specific wavelengths to produce free radicals or cations, which generates a new reaction source after reacting with the monomers in the slurry, and in turn initiates and propagates a polymerization reaction to produce polymers [98]. Monomers with different polarities and different types of photoinitiators are selected according to different ceramic powder materials. In addition to monomers, photoinitiators and ceramic powders, diluents, dispersants, defoamers [99], plasticizers [100,101,102] and light absorbers [81] are added to the pastes as appropriate in order to adjust the performance of the pastes as well as to improve the printing effect. Diluents are basically divided into active diluents and inert diluents. Active diluents are often low-viscosity monofunctional monomers and are involved in the photopolymerization reaction, while inert diluents do not participate in the reaction, but can adjust the viscosity of the slurry, the refractive index, etc., such as glycerol [103,104], 1-octanol [105] and so on.

The solid content of a ceramic slurry is an important factor in determining its suitability for printing. It is generally believed that if the solid content of the slurry is less than 40% by volume, the sample may collapse and deform during subsequent heat treatment [106]. As the organic matter in the slurry will be removed during the debinding and sintering processes, the solid content should be as high as possible to improve densification. However, too high solid content tends to increase the absorption and reflection of incident light by the powder particles, which reduces the curing depth and affects the curing effect [107]. In addition, high solid content will make the viscosity significantly higher. In practice, the solid content is often increased under the premise of ensuring the viscosity and curing effect.

The rheological properties of the ceramic slurry have a great impact on the workability of stereolithography 3D printing, as well as on the densification and mechanical properties of the printed ceramic products. Ceramic pastes for vat photopolymerzation must have a low viscosity, and the shear rate for instrumental operation in VPP is typically 30~100 s⁻¹ [108]. Machine work is generally considered to be slurry viscosity of up to 3 Pa·s to ensure coating and levelling of the ceramic slurry during printing [97,106,109]. In addition, the slurry is usually required to be a non-Newtonian fluid with a shear-thinning characteristic. The Krieger–Dougherty (K-D) model commonly used for viscosity calculation in VPP, which is an improvement of the model proposed by Einstein [110]. It establishes a correlation between the volume fraction of particles and the relative viscosity of UV-cured ceramic slurry [111]:

$${\eta }_r={(1-{\displaystyle \frac{\varphi }{{\varphi }_m}})}^{-\eta {\varphi }_m}$$

(1)

where η_r is the relative viscosity, ϕ is the volume fraction of solids, ϕ_m is the maximum volume fraction of solids, and η is the intrinsic viscosity parameter.

Curing depth (C_d) is one of the most important parameters for enabling to perform VPP, and it has a great impact on the print resolution [112]. Since VPP is built layer by layer, to ensure layer-to-layer bonding the curing depth needs to be about 10–35% larger than the layer thickness of the print setup [113]. Too large a curing depth can lead to overcuring, making the printing accuracy decrease [114]. The curing depth can be calculated according to the formula proposed by Griffith and Halloran [115] based on the Beer equation:

$$C_d=D_{p}ln{\displaystyle \frac{E}{E_c}}$$

(2)

$$D_p={\displaystyle \frac{2d}{3Q}}{\displaystyle \frac{1}{\varnothing }}{\displaystyle \frac{n_0^2}{{∆n}^2}}$$

(3)

$$Q\propto {\displaystyle \frac{s}{\lambda }}$$

(4)

where D_p is the depth of transmission, which is related to the light absorption coefficient of the resin and the scattering effect of the ceramic powder, E is the exposure energy density, and E_c is the critical exposure energy density; d is the average particle size of the ceramic powder, Q is the effective term of the extinction coefficient, φ is the ceramic solids volume fraction, n₀ is the refractive index of the resin, and ∆n is the refractive index difference between ceramic particles and the photosensitive resin; s is the spacing of the powder particles, and λ is the ultraviolet wavelength. The D_p and E_c of the slurry are influenced by various factors, including slurry composition, solid content, and ceramic powder particle size. For example, zirconia slurry with a solid content of 53 vol.% has a D_p of 35 μm and an E_c of 4.1 mJ·cm⁻² [116]. From the formula, the cured thickness is affected by the exposure energy density, the powder particle size, the difference in refractive index between the powder and resin, and the powder spacing. The effect of the difference in refractive index is the most obvious. Compared with the photosensitive resin alone, the ceramic particles in the prepared ceramic vat photopolymerzation slurry affect the refraction of light in the slurry. In other words, the relative refractive index between the ceramic particles and organics is the main factor affecting photopolymerization rate.

5.2. Selection of Ceramic Powders

The particle size distribution and surface morphology of ceramic powders greatly influence the rheological properties of the slurry, the printing process and the material properties. Due to light scattering issues, light absorption and photopolymerization rates decrease as the relative refractive index increases. Additionally, smaller particle sizes enhance the absorption of light by the slurry. The particle size of the zirconia slurry currently used in VPP is primarily at the micron and submicron level.

In general, the slurry’s viscosity increases as ceramic powder’s particle size decreases and the shear stress of the slurry increases. This means the slurry’s viscosity rises with an increase in specific surface area [117], as shown in Figure 7. When ceramic powders are nanoscale, the van der Waals forces between particles are stronger and Brownian motion is more intense. Additionally, particles with larger specific surface areas have more active sites overall. Consequently, when the particles are small, they interact more strongly, making agglomeration more likely to occur [118]. The oversized ceramic particles are prone to sedimentation due to gravity. In addition, ZrO₂ powders with large particle size have low surface activity compared to fine powders, which is not conducive to the sintering densification of ceramics. The use of large particle size powders produces defects and pores, which reduces the mechanical properties of ceramics. Li et al. [119] investigated how the specific surface area of zirconium dioxide powder affects the rheological properties of zirconium dioxide slurry used in SLA. They found that the specific surface area of the powder significantly affected the viscosity of the slurry. As the powder’s specific surface area decreased, the slurry’s viscosity also decreased. Fan et al. [120] compared the rheological properties of slurries prepared from ZrO₂ powders with an average particle size of 2 μm and 4 μm, respectively, applied in photopolymerization 3D printing. The viscosity of the slurry at a shear rate of 160 s⁻¹ was 2218.40 mPa·s for a particle size of 2 μm, while the viscosity was 1053.96 mPa·s for a particle size of 4 μm. The dispersant adsorbed on the surface of the powders separates the powders by steric repulsion and van der Waals forces. The decrease in powder particle size will increase the total surface area, and the adsorption layer formed by dispersant will inevitably become thinner, decreasing the spacing between the powders. In addition, ball milling is more likely to lead to entanglement of the dispersant in the adsorbed layer of the powder, further increasing the viscosity of the slurry.

The shape of the powder particles also has an effect on the properties of the slurry. Wang et al. [121] investigated the effect of different ratios of equiaxed to platelet-shaped Al₂O₃ powders on the properties of ZrO₂-toughened Al₂O₃ ceramic with the overall addition of Al₂O₃ remaining constant using DLP-based stereolithography. The addition of platelet-shaped Al₂O₃ reduces the viscosity of the slurry to 5.10 Pa·s for the slurry containing 8 vol.% flake powder at a shear rate of 30 s⁻¹, whereas the viscosity of the slurry without flake powder is about 17.63 Pa·s. It has been demonstrated that the alignment of platelet-shaped alumina particles along the direction of shear, when subjected to an external shear force, results in a reduction in the drag force between the particles. However, platelet-shaped alumina powders are more prone to agglomeration compared to equiaxed powders. The high aspect ratio of the lamellar powders makes them prone to form an oriented structure in the slurry, which reduces to some extent the stability of the slurry. Therefore, although platelet-shaped alumina powders can improve the filling effect, it is necessary to consider whether or not to add them and at what level. Khakzad et al. [122] used zirconia powder that was irregularly shaped and had a relative roundness of less than 0.37 to prepare a vat photopolymerization slurry. This resulted in zirconia samples that had obvious pores. Increasing the proportion of smaller, irregularly shaped particles decreased the porosity of the samples from 41.8% to 20.3%.

Furthermore, the particle size distribution of ceramic powders is found to have a significant impact on the performance of the slurry. Single-peaked powders often have difficulty in balancing viscosity with the performance of subsequent preparations. The use of ceramic slurries with bimodal or even multimodal particle size distributions reduces the viscosity, increases the solid content and increases the curing depth. In the context of using a bimodal distribution powder, it is widely accepted that in order to reduce the viscosity of the slurry, the particle size ratio of coarse powder to fine powder should be a minimum of 5. The coarse-to-fine powder ratio can be calculated using the following formula [111]:

$${\varnothing }_{01}={\displaystyle \frac{1-{\epsilon }_1}{1-{\epsilon }_1{\epsilon }_2}}$$

(5)

$${\varnothing }_{02}={\displaystyle \frac{(1-{\epsilon }_2){\epsilon }_1}{1-{\epsilon }_1{\epsilon }_2}}$$

(6)

where ${\mathrm{\varnothing }}_{01}$ and ${\mathrm{\varnothing }}_{02}$ are the ideal coarse-to-fine particle ratios for the densest stacking, and ε₁ and ε₂ are the porosities of the two particles.

Fan et al. [120] prepared the VPP slurries using zirconia powders with different particle size gradations such as 2 μm/0.4 μm and 4 μm/0.4 μm, respectively, and compared their rheological properties of the ceramic slurries. The viscosities of the slurries prepared with bimodal distribution of ceramic powders were all lower than those of the large particle powder slurries alone, and the combination of 4 μm + 0.4 μm in particle size was more effective in reducing the viscosity. In combination of theoretical calculations with experiments, the optimal ratio of coarse particles to fine particles of 65%: 35% was obtained. Xing et al. [111] adjusted the contents of 1 μm Al₂O₃ and 200 nm ZrO₂, and when the ratio of large particles to fine particles in the slurry was 70 vol.% to 30 vol.%, the particle spacing in the slurry was the largest, and the viscosity was the lowest at 10.83 Pa·s when the shear rate was 30 s⁻¹. Zheng et al. [123] derived the formula for the optimum ratio of ternary particles and used alumina and zirconia with ternary particle size distribution to prepare ZrO₂-toughened Al₂O₃ (ZTA) samples with a linear shrinkage of less than 20%, and a compressive strength of 781 ± 39 MPa, a fracture toughness of 3.76 ± 0.05 MPa·m^1/2, a Vickers hardness of 19.58 ± 0.33 GPa, and relative density of 99.3%, respectively.

The particle size, size distribution and morphology of zirconia powders have a significant impact on the performance of the ceramic slurry. In order to ensure the optimum outcome in the preparation of slurry, it is imperative that ceramic powders with a moderate particle size are selected. According to the actual conditions to decide whether to use a multi-peak distribution and a suitable morphology of ceramic powders to ensure that the performance of the slurries can meet the printing requirements.

5.3. Selection of Resins and Dispersants

Resin is a pivotal component of the slurry, and the rheology of the slurry can be changed by adjusting the type and content of the resin. The commonly used monomers are acrylamide-based resins, acrylate-based resins and epoxides. Among them, acrylamide-based resins are waterborne resins, while acrylates and epoxides are non-waterborne resins [112]. Although water-based pastes have a low viscosity as compared to non-water-based pastes with the same solid loadings, water-based pastes contain a large amount of water, which makes the prepared green bodies tend to crack. In addition, hydrolysis and oxidation of non-oxide ceramics in water-based pastes may occur. Non-aqueous resins have a high refractive index and tend to have a small refractive index difference in contrast with the added ceramic powder. Non-aqueous resins are more widely used due to excellent rheological properties and curing quality of ceramic slurries. Among the non-aqueous resins, acrylate resins and epoxides belong to free radical polymerization resins and cationic polymerization resins, respectively, and must correspond to the use of different types of photoinitiators [124].

Table 1 summarizes the different ZrO₂ slurry formulations and the corresponding viscosities and curing depths for VPP. In the preparation process of zirconia ceramic slurries, the most used acrylates are 1,6-hexanediol diacrylate (HDDA) and trimethylolpropane triacrylate (TMPTA). HDDA is a bifunctional monomer with low viscosity, which provides good fluidity for the slurry, while TMPTA has three functionalities and provides three reactive cross-linking sites, which can accelerate the formation of cross-linking network. However, a high concentration of TMPTA may make the cured layer strongly adhere to the substrate during printing, leading to the difficulties in mold release [57].

Jia et al. [99] investigated the effect of three combinations of resin monomers on the properties of 8 mol.% yttria-stabilized zirconia ceramic slurries for vat photopolymerzation, namely HDDA/PONPGDA/IBOMA, tetrafunctional ethoxylated (5) pentanediol tetraacrylate (PPTTA)/HDDA and TMPTA/HDDA. In the first group, the combination of bifunctional monomers of HDDA and bifunctional propoxylated neopentyl glycol diacrylate (PONPGDA) with monofunctional monofunctional isoprenyl methacrylate (IBMO) showed good rheology, but the overall crosslink density was insufficient resulting in a low curing depth. The second group, containing tetrafunctional PPTTA, was in the usable range for curing performance and viscosity. However, the cost of using TMPTA was lower and the third group exhibited better overall performance. Selection of resin type and content needs to be considered in terms of curing performance as well as the rheology of the paste.

In addition, since zirconia has a variety of stabilizers, different additives can lead to the differences in performance. In the study of Cailliet et al. [125], two photoinitiators, phenylbis (2,4,6-trimethylbenzoyl) phosphine oxide (BAPO) and 2-methyl-4′-methylthio-2morpholinopropiophenone (MMMP), with higher molar extinction coefficients, were used due to the higher UV absorption of CeO₂ tetragonal zirconia polycrystal (Ce-TZP) than Y-TZP [125]. The absorption property of BAPO at long wavelengths was utilized to avoid the strong absorption region of CeO₂ at short wavelengths in order to improve the utilization of UV light energy.

The purpose of adding dispersants is to improve the rheological properties of ceramic pastes by using the interaction among the dispersants, the ceramic particles and resin. At present, there are two types of commonly used dispersants, one is mainly BYK-103 [126], BYK-180 (an alkylammonium salt of a copolymer containing acidic groups) [127], polyvinyl pyridine [128], KOS110 (Copolymer dispersant) [1] and other polymer dispersants, through the polymer and ceramic powder chemical or physical adsorption to form an adsorbed layer, the formation of the steric repulsion in order to improve the rheological properties of ceramic slurry. The other type is sodium polyacrylate [126,129,130], ammonium polyacrylate and other ionic dispersants, the use of dispersant electrostatic adsorption to produce a bilayer with electrostatic repulsive force, which reduces the viscosity of ceramic paste. With the increase in the amount of dispersant added, the viscosity of the ceramic slurry will often be the first to increase, that is, there is an optimal minimum viscosity point. However, too much dispersants will adversely affect the internal structure of the particle surfaces due to flocculation and other phenomena, which makes the viscosity of the slurry increase as well as poor dispersion [131]. Figure 8 shows the effects of dispersant type and content on slurry viscosity.

Table 1. Viscosity and curing depth of ZrO₂ slurry formulations for VPP.

Method	Type of Powder	D50 (nm)	Monomer	Dispersant	Cd (μm)	Solid Loading (vol.%)	Viscosity (Pa·s)	Reference
SLA	8YSZ	200	TMPTA/HDDA (wt.% 1.5:8.5)	Solsperse 41000 (A 100% active polymeric dispersant)	160	43	3.6 at 30 s−1	[99]
SLA	6Yb4ScSZ	314	TMPTA	Monofax 4.9vol.%	125	47	-	[100]
SLA	3YSZ	40	HDDA	BYK-111 (A copolymer with an acid group.)	-	50	20–30 at 15.8 s−1	[132]
DLP	ZrO2	440	HDDA	Solsperse 41000	-	45	<3 at 10 s−1	[133]
DLP	3YSZ	200	HDDA/polyethylene glycol diacrylate (PEGDA)	DisperBYK (copolymer containing acid groups)	50	83 wt.%	1.23 at 100 s−1	[97]
DLP	3YSZ	200	HDDA/TMPTA	Hypermer KD-1 (polyester/polyamine condensation polymer with a cationic head group)	-	40	-	[134]
DLP	3Y-TZP	100–200	-	BYK-103 (A copolymer solution with affinic filler groups)	-	40.5/43.6	2 at 100 s−1	[135]
DLP	HAP/ZrO2	-	HDDA/acrylamide morpholine (ACMO)/TMPTA/hyperbranched polyester acrylate 45:35:15:5	castor oil phosphate (COPE) (a mixed anionic dispersant)	-	60/3 wt.%	-	[136]

Table 1. Viscosity and curing depth of ZrO₂ slurry formulations for VPP.

Method	Type of Powder	D50 (nm)	Monomer	Dispersant	Cd (μm)	Solid Loading (vol.%)	Viscosity (Pa·s)	Reference
SLA	8YSZ	200	TMPTA/HDDA (wt.% 1.5:8.5)	Solsperse 41000 (A 100% active polymeric dispersant)	160	43	3.6 at 30 s−1	[99]
SLA	6Yb4ScSZ	314	TMPTA	Monofax 4.9vol.%	125	47	-	[100]
SLA	3YSZ	40	HDDA	BYK-111 (A copolymer with an acid group.)	-	50	20–30 at 15.8 s−1	[132]
DLP	ZrO2	440	HDDA	Solsperse 41000	-	45	<3 at 10 s−1	[133]
DLP	3YSZ	200	HDDA/polyethylene glycol diacrylate (PEGDA)	DisperBYK (copolymer containing acid groups)	50	83 wt.%	1.23 at 100 s−1	[97]
DLP	3YSZ	200	HDDA/TMPTA	Hypermer KD-1 (polyester/polyamine condensation polymer with a cationic head group)	-	40	-	[134]
DLP	3Y-TZP	100–200	-	BYK-103 (A copolymer solution with affinic filler groups)	-	40.5/43.6	2 at 100 s−1	[135]
DLP	HAP/ZrO2	-	HDDA/acrylamide morpholine (ACMO)/TMPTA/hyperbranched polyester acrylate 45:35:15:5	castor oil phosphate (COPE) (a mixed anionic dispersant)	-	60/3 wt.%	-	[136]

Many studies have been conducted on the optimization of dispersants. Komissarenko et al. [137] investigated the effect of four dispersants, BYKw969 (Solution of alkyl ammonium salts of acidic copolymers containing hydroxyl functional groups), BYKw996 (Solution of copolymers containing acidic groups), TritonX-45 (Nonionic surfactant) and TritonX-114 (Nonionic surfactants of octylphenyl polyoxyethylene ether type), on the rheological properties of zirconia slurries with 3 mol% yttria stabilized zirconia (3YSZ) powder or 8 mol% yttria stabilized zirconia (8YSZ) powder. At the same addition amount used, BYKw969 had the best effect on viscosity reduction in the slurry, which was less than 2 Pa·s at a shear rate of 10 s⁻¹. Zhang et al. [128] investigated the effects of four dispersants, including polyethylene glycol, polyvinylpyrrolidone, sodium polyacrylate and oleic acid, and their addition amounts on the rheological properties of 8 mol.% yttria-stabilized zirconia ceramic slurries. When 0.1 wt.% of oleic acid was used as the dispersant, the viscosity of zirconia ceramic slurry was the lowest. Zhang et al. [1] investigated the effect of three dispersants, KOS110, KOS163 (A polyurethane-based universal dispersant) and Solsperse17000 (100% active polymer dispersant), on the rheological properties of ZrO₂ slurry. When the solid content of the ZrO₂ slurry was 40 vol.%, the ZrO₂ slurry using KOS110 as the dispersant exhibited the lowest viscosity. At a shear rate of 200 s⁻¹, the viscosity of ZrO₂ slurry was 0.136 Pa·s. Based on this, they optimized the slurry. The viscosity of ZrO₂ slurry was only 1.65 Pa·s with the addition of 2 wt.% KOS110 at a solid content of 55 vol.% in a ZrO₂ slurry.

Borlaf et al. [133] investigated the effect of Solsperse 41,000 dispersant addition on the rheological properties of ZrO₂ and Al₂O₃-ZrO₂ ceramic slurries via UV-LCM-DLP additive manufacturing. The viscosity of the slurry to vary with the dispersant content, which had the lowest viscosity at the addition of 1 wt.% dispersant. In addition, when the dispersant was used at 0.75 wt%, the slurry exhibited shear thickening at high shear rate. Wu et al. [104] investigated the effect of polyvinylpyrrolidone dispersant addition on the viscosity of zirconia-toughened alumina ceramic slurry. It was established that an increase in the content of polyvinylpyrrolidone dispersant resulted in a decrease in the viscosity of the slurry, which then increased. When 1.2 wt.% of dispersant is added, the viscosity of the slurry reaches its lowest value. Sokola et al. [138] investigated the optimum amount of BYK-103 based on conventional measurements of slurry viscosity in conjunction with analytical centrifuges to study the rheological properties of slurries. The addition of 10 wt.% BYK-103 reagent was able to provide a low viscosity for the slurries. BYK-103 reagent also provided good stability and rheological properties of the slurry. The viscosity of the ZrO₂ slurry at a shear rate of 10 s⁻¹ was 0.93 Pa·s at a solid content of 30 vol.%. Kim et al. [131] investigated the effect of the BYK-142 (A solution of high-molecular-weight copolymers containing phosphate ester salt and pigment-affinity groups) dispersant content on the rheological properties of the slurry for VPP. The viscosity of the slurry also showed a first decrease and then increase trend with the increase in BYK-142 content. The lowest viscosity, the best dispersibility, and the slowest settling rate were obtained at a concentration of 2 wt.% BYK-142 dispersant. In addition, when the dispersant content was too high or too low, the curing depth of the slurry was significantly increased due to the poor stability and uneven distribution of ceramic powder in the slurry.

In addition, methods that use AI technology to study formulations have emerged. Machine learning can analyze and predict the performance of different formulations, providing a useful reference for pulp formulation research and development [71].

In summary, according to the type and particle size of ceramic powders, selecting the appropriate type and content of resin is to ensure the curing ability of the slurry. Adding suitable dispersants can reduce the viscosity of ceramic slurries and improve their stability during printing.

5.4. ZrO₂ Composite Ceramics Slurry

In order to improve the properties of VPP-printed zirconia parts to meet various requirements, the option of suitable modifiers for ZrO₂ ceramics is a feasible and commonly used method. Table 2 summarizes the curing depth and viscosity of some ZrO₂ composite slurries.

In the development of zirconia composite ceramics, the most studied composite is ZrO₂-toughened Al₂O₃ ceramics. Alumina provides excellent hardness and wear resistance, while zirconia provides high flexural strength and fracture toughness. Cailliet et al. [125] compared the properties of Ce-TZP and Ce-TZP/Al₂O₃ slurries for digital light processing formulations, and investigated the effect of alumina on zirconia ceramic slurries. Due to the incorporation of alumina using a mixed calcination of aluminum isopropoxide, the lower calcination temperature resulted in the powder producing a mesoporous structure and a large specific surface area (52 m²·g⁻¹). This resulted in poor wettability with the resin and the slurry filler could only reach 25–30 vol.%. Thakur et al. [139] prepared ZrO₂-Al₂O₃ composite slurry for DLP, and discovered that the viscosity of the slurry showed a significant increase with increasing the solid content, while the curing depth increased with increasing the solid content. The curing depth was 69 μm, 92 μm and 95 μm for solid contents of 30 vol.%, 35 vol.% and 42.5 vol.%, respectively. However, 35 vol.% was found to be the optimum solid content for the combined performance of ceramic slurries. Xing et al. [111] prepared ZrO₂-Al₂O₃ pastes with varying amounts of ZrO₂ for SLA. The viscosity of the slurry exhibited an increase from 13,320 to 49,100 mPa·s as the ZrO₂ content was increased from 5 vol.% to 20 vol.%. In particular, the viscosity increased most rapidly when the viscosity of the slurry increased from 15 vol.% to 20 vol.%. The curing depth decreased from 150 μm to 100 μm, and the ZrO₂-Al₂O₃ paste with the content of 15 vol.% ZrO₂ was finally selected.

In addition to the composite of zirconium oxide and aluminum oxide, there have been many studies on the composite of zirconium oxide and other materials. Shi et al. [140] added ZrO₂ to SiC ceramic slurry for digital light processing and explored the effect of its content on the slurry performance. As the ZrO₂ content increased, the powder particles got closer together and the effect of dispersant was weakened. At the same time, the attraction between the particles got stronger because of the high surface energy of the nanoscale ZrO₂ particles. This led to an increase in the slurry’s viscosity. In addition, with increasing the ZrO₂ content, ZrO₂ particles occupied the interstitial space of SiC particles, which hindered the propagation of UV light and inhibited the depth of UV penetration in the slurry, resulting in a decrease in the curing depth. Mao et al. [141] investigated the effect of different ZrO₂ contents on rheological properties of Si₃N₄-ZrO₂ composite slurries for vat photopolymerization. Compared to the irregular shape of Si₃N₄ particles, ZrO₂ powders were more spherical. The viscosity of the slurry gradually decreased with increasing the ZrO₂ content. In addition, the curing depth of the slurry increased with increasing the ZrO₂ content due to relatively low UV absorption of white ZrO₂ powders. Compared with the slurry without ZrO₂, the composite slurry with 20 wt.% ZrO₂ had approximately 5% lower UV absorption than the Si₃N₄ ceramic slurry without ZrO₂, and thus the curing depth increased from 25.5 μm to 64.8 μm. Zhang et al. [108] added graphene oxide (GO) to 3 mol.% yttria stabilized ZrO₂, and investigated the effect of graphene oxide on the properties of the slurry for a comparative study. The presence of graphene oxide led to an increase in the hydrophilicity of zirconia resulting in a decrease in the stability of the slurry, which was more prone to settling. In addition, the absorption of UV light by graphene oxide and the addition of particles to enhance the scattering of light resulted in a decrease in the curing depth of the slurry.

Table 2. Curing depth and viscosity of ZrO₂ composite slurry.

Materials	D50 (nm)	Cd (μm)	Solid Loading (vol.%)	Viscosity (Pa·s)	Reference
Ce-TZP 70 vol.%, Al2O3 30 vol.%	-	75	45	-	[125]
3YSZ 80%, Al2O3 20%	SSA 6.5 m2·g−1	92	35	0.48 at 100 s−1	[139]
Al2O3 85 vol.%, ZrO2 15 vol.%	1000(Al2O3), 200(ZrO2)	108	47	2.82 at 30 s−1	[111]
Al2O3 80 wt.%, 5YSZ 20 wt.%	780(Al2O3), 230(5YSZ)	-	45	3.45 at 30 s−1	[142]
Al2O3 75 wt.%, 3YSZ 25 wt.%	400 (Al2O3), 200 (3YSZ)	45	40	0.382 at 18.6 s−1	[143]
SiC 85 wt.%, ZrO2 10 wt.%, Al2O3 2.5 wt.%, Y2O3 2.5 wt.%	3600 (SiC), 300 (ZrO2), 800 (Al2O3), 3300 (Y2O3)	30	40	-	[140]
Si3N4 81 wt.%, ZrO2 10 wt.%, Al2O3 4.5 wt.%, Y2O3 4.5 wt.%	1660 (Si3N4), 1280 (ZrO2)	48.5	-	<2 at 60 s−1	[141]

Table 2. Curing depth and viscosity of ZrO₂ composite slurry.

Materials	D50 (nm)	Cd (μm)	Solid Loading (vol.%)	Viscosity (Pa·s)	Reference
Ce-TZP 70 vol.%, Al2O3 30 vol.%	-	75	45	-	[125]
3YSZ 80%, Al2O3 20%	SSA 6.5 m2·g−1	92	35	0.48 at 100 s−1	[139]
Al2O3 85 vol.%, ZrO2 15 vol.%	1000(Al2O3), 200(ZrO2)	108	47	2.82 at 30 s−1	[111]
Al2O3 80 wt.%, 5YSZ 20 wt.%	780(Al2O3), 230(5YSZ)	-	45	3.45 at 30 s−1	[142]
Al2O3 75 wt.%, 3YSZ 25 wt.%	400 (Al2O3), 200 (3YSZ)	45	40	0.382 at 18.6 s−1	[143]
SiC 85 wt.%, ZrO2 10 wt.%, Al2O3 2.5 wt.%, Y2O3 2.5 wt.%	3600 (SiC), 300 (ZrO2), 800 (Al2O3), 3300 (Y2O3)	30	40	-	[140]
Si3N4 81 wt.%, ZrO2 10 wt.%, Al2O3 4.5 wt.%, Y2O3 4.5 wt.%	1660 (Si3N4), 1280 (ZrO2)	48.5	-	<2 at 60 s−1	[141]

Furthermore, the doping of zirconia slurry exerts an influence on the rheological properties of the slurry. In particular, the doping of rare earth oxides has been demonstrated to enhance the performance of zirconia slurry. Qi et al. [144] doped a small amount of Pr₆O₁₁, CeO₂, and Er₂O₃, respectively, into yttria stabilized ZrO₂, and investigated the rheological properties of their slurries for digital light processing. Doping with small amounts of rare earth oxides had almost no effect on the viscosity of the slurry, and the slurry properties remained unchanged. However, the curing properties of the pastes were significantly affected. The depth of transmission, D_p, of the pastes decreased significantly after doping with rare earth oxides. The depth of transmission D_p was 31.62 μm when undoped, however, it decreased to 15.2 μm after doping with 0.2 wt.% of Pr₆O₁₁. The depth of transmission D_p was reduced to 17.7 μm and 19.5 μm, respectively, after doping with 4.0 wt.% CeO₂ and 0.8 wt.% Er₂O₃ into yttria stabilized ZrO₂ slurries.

6. Surface Functionalization of ZrO₂ Powders

Hydrophilic hydroxyl groups are usually present on the surface of ceramic powders, whereas the surface of commonly used VPP resins are mostly hydrophobic groups, which usually makes the slurry properties suffer. Although the addition of dispersants is a common method to reduce the viscosity of ceramic slurries, the resins tend to inhibit the adsorption of dispersants on the surface of ceramic powders. In order to improve the interaction between the powders to enhance the rheological properties of ceramic suspensions, surface modification can be considered. Common surface modifications include surface oxidation, use of surface modifiers, and surface coating [57]. Surface oxidation is commonly used for non-oxide ceramic powders such as Si₃N₄ to improve the inherent deficiencies of non-oxide ceramics in VPP. Surface modifiers and ceramic powder particles surface hydroxyl chemical reaction, molecular groups hanging out on the surface of ceramic particles, thereby improving the wettability between ceramic particles and photosensitive resin. The use of external groups of steric repulsion or electrostatic repulsion, improves the dispersion of ceramic powder in the photosensitive resin, so as to improve the solid content of ceramic slurry, stability and reduce the viscosity.

Jang et al. [145] modified 3 mol.% yttria-stabilized zirconia powder using methyltrimethoxysilane coupling agent. The samples prepared from the modified powders were denser and showed better mechanical properties, with strength and hardness increased by 5–6%. Zhang et al. [108] used the silane coupling agent KH560 (γ-Glycidyl ether oxygen propyl trimethoxysilane, an epoxy functional silane) to modify the surface of the slurry powder and analyzed the effect of surface modification on the rheology of the slurry. The modified slurry of 3 mol.% yttria-stabilized zirconia showed significantly less precipitation and increased slurry stability after a settling test of the same duration. Fourier transform infrared spectroscopy (FTIR) results showed the disappearance of hydroxyl (-OH) groups on the surface of the modified pristine 3YSZ powders and the appearance of bonds such as Si-O-Zr bonds, C-O bonds, and C=O bonds, which reflect the effective connection between the surface modifier and the particulate powders. Fan et al. [120] used KH560 for surface modification in order to make KH560 adsorb completely on the surface of ZrO₂ particles. ZrO₂ powder was first added to a mixed solution of anhydrous ethanol and KH560 and ultrasonicated to induce adsorption. Then, solid–liquid separation by centrifugation and washing and drying can complete the surface modification. Through the surface modification, combined with the particle grading, slurries with high solid content and excellent thixotropy were prepared.

Borlaf et al. [133] compared the properties of ZrO₂ and Al₂O₃-ZrO₂ pastes prepared by one-step and two-step methods. The two-step method involves dispersing the Al₂O₃ and ZrO₂ powders in ultrapure water containing MelPers 4350 for functionalization before the conventional step. MelPers 4350 at a concentration of 3 wt.% was found to prepare ZrO₂ ceramic slurries with optimum rheological properties. Sun et al. [118] investigated the effect of five modifiers, stearic acid, oleic acid, BYK, KH560, and variquat CC, on the stability and rheological properties of ZrO₂ slurries, and found that the use of stearic acid and variquat CC-modified ZrO₂ slurry modified with stearic acid and variquat CC, it was found that the powder in ZrO₂ slurry modified with stearic acid and variquat CC all settled after 20 days, while ZrO₂ slurry modified with oleic acid, KH560 and BYK did not settle significantly after 20 days of static storage; and the effect of the concentration of oleic acid, KH560 and BYK modifiers on the viscosity of ZrO₂ slurry was investigated, and the selection of BYK as a surface modifier could reduce the viscosity of ZrO₂ slurry more efficiently than that of the modifiers with oleic acid and KH560. The viscosity of ZrO₂ slurry was the lowest when the addition amount of BYK was 3 wt.%.

In summary, surface modification can tailor the interaction between powders. Treatment of zirconia powders with surface modifiers can reduce the hydrophilicity of the powders and enhance the dispersion and stability of the powders in the slurry. The use of surface modification is an excellent choice for reducing slurry viscosity at high solids content.

7. Effect of Printing Parameters on the Properties of Zirconia Ceramics

The printing process is the most important process for the preparation of zirconia building blocks by VPP. The precision, surface roughness, and densities of the printed explants are affected by the printing parameters. The printing parameters that need to be regulated according to the specific conditions are laser power, printing directions, layer thickness and exposure time. Inappropriate combinations of print parameters can result in poor layer adhesion and incomplete curing during printing, leading to voids between and within layers. Figure 9 shows a typical SEM image of the side of a zirconia component prepared by DLP.

It has been demonstrated that an increase in laser power results in a corresponding enhancement in the strength of the bond between neighboring layers. Nevertheless, an excess of power can generate sufficient heat to cause the resin to burn, resulting in a suspension that cannot be cured. Since SLA prints point-by-point, in addition to the curing depth, the curing width is also a parameter that affects its printability. The cure width is proportional to the square root of the laser power, i.e., it increases with the laser power, but the growth tends to slow down gradually. In addition, scattering of light can also widen the width and this effect becomes stronger with increasing laser power. Fu et al. [147] adjusted the laser intensity for SLA printing of zirconia embers and prepared specimens at 160, 260 and 360 mW laser intensities with constant laser scanning speed and spot diameter and investigated its effect on the material properties. The curing broadband and curing depth increased with increasing laser intensity until a slow growth after 360 mW. The weight of the part increased with increasing laser power, but the size increased more rapidly making the density decrease with increasing laser power. Correspondingly, as the laser power increases, the density and relative density of the sintered sample increase. When the laser power increases, the curing width and depth of the slurry increase, reducing the uncured area between layers. The uncured area provides a channel for the removal of organic matter during the debinding process. The increase in laser power reduces the size of the uncured regions from an unfavorable large oversize. Additionally, as the laser power increases, the hardness slightly increases, while the fracture toughness slightly decreases.

Dai et al. [83] investigated the effect of light intensity on the properties of DLP-printed 3YSZ samples with meshed surfaces. As the light intensity was increased from 8 mW·cm⁻² to 36 mW·cm⁻², the change in curing depth resulted in an increase in wall thickness from 0.28 mm to 0.44 mm. In addition, the flexural strength of the specimens showed an increasing and then decreasing trend as the light intensity was increased from 4 mW·cm⁻² to 10 mW·cm⁻², and reached a maximum of 952 ± 75 MPa at a light intensity of 8 mW·cm⁻². Too low a light intensity makes the embryo curing incomplete, and the softness of embryos printed at a light intensity of 4 mW·cm⁻² is difficult to be post-processed. Radomski et al. [148] varied the total exposure energy dose during the DLP process, and specimens prepared at a medium-energy (150 mJ·cm⁻²) had better lateral resolution as well as maximum flexural strength.

The layer thickness requires a depth of cure that further needs a combination of paste composition, laser intensity and exposure time. Layer thicknesses that are too low not only significantly increase the print time, but also cause unevenness and warping of the printed part due to over-curing. Larger layer thicknesses require higher curing depths that are difficult to achieve and are prone to incomplete curing resulting in poor performance. Since DLP uses the entire layer to be printed simultaneously, the effect of layer thickness and other factors on the printing results is obvious. Additionally, excessive layer thickness can result in inadequate curing, potentially causing voids between layers. Conti et al. [149] studied the effect of exposure energy and layer thickness in DLP printing of zirconia and found that higher exposure energies resulted in oversized parts, while lower energies did not ensure the formation of stable parts. The layer thickness, on the other hand, is very significant on the surface condition, as the formation of laminar structures is hardly observed when the layer thickness is 10 μm, while a layer thickness of 50 μm shows clear steps and laminar structures. The surface quality of the part is therefore significantly affected by this variation. Depending on the requirements of the application, the layer thickness can be increased to increase the printing speed or decreased to improve the surface finish in the z-direction, whereas, increasing the energy dose at high layer thickness printing increases the depth of curing in the z-direction, resulting in more uniform adhesion between layers. Reddy et al. [146] investigated the effect of layer thickness, orientation and light intensity on zirconia embers density and surface unevenness in DLP. An increase in layer thickness from 10 μm to 70 μm, resulted in a decrease in embryo density from 3.23 g·cm⁻³ to 2.92 g·cm⁻³, accompanied by an increase in surface roughness from 5 μm to 10 μm. Conversely, a variation in printing direction from 0° to 90° led to a decline in embryo density from 3.2 g·cm⁻³ to 3.04 g·cm⁻³, and the roughness was maximized to 9.8 μm at 67.5°. When the light intensity was increased from 10 mW·cm⁻² to 50 mW·cm⁻², the embryo density increased from 2.97 g·cm⁻³ to 3.28 g·cm⁻³, and there was no significant change in surface roughness. Mou et al. [150] compared the accuracy of zirconia crowns printed at the layer thicknesses of 30 μm and 50 μm. The edge accuracy of the components was significantly reduced at a layer thickness of 50 μm, with a concave-convex trueness (rms) of 70.7 ± 4.3 μm, compared to a concave-convex trueness (rms) of 43.8 ± 4.3 μm at a layer thickness of 30 μm under the same conditions. In addition, there was an offset in the printed support at a layer thickness of 50 μm. This may be due to the deformation of the paste under gravity and platform movement at large layer thicknesses.

As a consequence of the additive manufacturing process’s inherent layer-by-layer construction, the components prepared by VPP are often oriented. In different printing directions, the printing accuracy and mechanical properties of the components have different performances. In addition, different printing angles will affect the number of printed layers, interlayer interfaces and total interface area. Smaller particle size powders are more likely to adsorb at the interface, which increases the scattering center of the curing light and reduces the interlayer bonding strength. Consequently, samples printed at different angles will exhibit varying degrees of defects. Xiang et al. [151] prepared zirconia samples with two different printing directions, vertical and horizontal, using SLA, and analyzed the accuracy, density, and mechanical properties of the samples; the samples fabricated in the horizontal direction had better accuracy, better fracture toughness of 12.635 ± 1.372 MPa·m^1/2, and better flexural strength of 1151.08 ± 166.41 MPa, while the samples prepared in the vertical direction had better accuracy, higher fracture toughness, and higher flexural strength of 1151.08 ± 166.41 MPa, while the density of the samples prepared in vertical orientation was higher. Shen et al. [152] conducted an experimental study on the effect of exposure time and printing angle on the flexural strength of zirconia ceramics during the DLP process. It was found that the exposure time had a greater effect on the flexural strength. The variations in fracture strength with exposure time showed a wavy undulating trend, with the lowest flexural strength of 371 MPa at an exposure time of 6 s, and the maximum flexural strength of 580 MPa, which appeared at an exposure time of 13 s. It was hypothesized that when the exposure time was longer than 3 s, the light attenuated by the curing layer would still be irradiated resulting in a localized warming or localized polymerization, making localized stresses or defects appear until the secondary conversion was completed. In addition, since 3D-printed samples tend to be anisotropic, the strength of the specimens was lower than 300 MPa at print angles of 15°, 30°, and 45°, and the strength of the specimens prepared at print angles of 60°, 75°, and 90° was fair, with the worst flexural strength of the samples at a print angle of 45°. Marsico et al. [153] conducted experiments on the DLP method for the preparation of different build-up orientations of The mechanical behavior of zirconia was experimentally investigated and it was found that there was no significant difference in Vickers hardness relative to the indentation orientation, with an overall average hardness of 13.1 ± 0.31 GPa, the fracture toughness was the best performance of the specimens constructed at 45°, and the results of the flexural strength were the same as those of the others for the layer perpendicular to the specimen of the maximum positive stress, i.e., the best performance of the specimens constructed at 0°.

In addition to these conventional studies, Song et al. [154] improved the printing accuracy to 2.9~3.8 times by adjusting the feeding speed and feeding height during DLP printing to increase the shear stress during printing to remove the overcured portion of the printing. Cho et al. [155] investigated the angle of the support on the printing effect, and the best realism of the samples was achieved at a maximum support angle of 50°. Zhang et al. [156] improved the DLP equipment by designing a composite oxygen-rich film consisting of microporous PET film and PDMS coating under the slurry, which allows oxygen to penetrate into the ceramic slurry to form a “dead zone” that is difficult to cure, and the existence of the dead zone makes it possible for the slurry to cure without the need for layer-by-layer stopping and mechanical separation, thereby this allows for continuous layerless manufacturing. Although this method results in a reduction in printing accuracy.

8. Influence of the Debinding and Sintering Processes on Properties of Zirconia Ceramics

8.1. Influence of the Debinding Process on Zirconia Ceramics by VPP

Since light-cured 3D printing ceramic slurries contain polymers that often contain 50 vol.% or more, debinding is required to remove the organic polymers after printing. Thermal debinding, solvent debinding, microwave debinding, and catalytic debinding are all common methods of debinding. Due to the nature of the organic material used, most of the debinding of light-cured 3D printed zirconia ceramics is performed using thermal debinding methods. Air, nitrogen or vacuum are commonly used debinding atmospheres. The micropores and channels created by the removal of organics during debinding, as well as the thermal stresses generated by the volatilization of organics, may lead to subsequent defects such as delamination and cracking [157]. Therefore, controlling the parameters such as atmosphere, temperature, and heating rate to allow sufficient and slow volatilization of organic matter is the key to minimize defects.

Thermal debinding decomposes the organics with the help of temperature, so the adjustment of the debinding temperature parameters needs to be based on the knowledge of the decomposition temperature of the organics. Table 3 summarizes some VPP prepared zirconia debinding and sintering parameters and properties. It can be seen that most of the debinding temperatures are in the range of 600 to 1000 °C. This is the range of decomposition temperatures for most of the resins used. The decomposition temperature of the organics is generally determined by the weight loss temperature of the specimen measured by thermogravimetric-differential thermal analysis (TG-DTA) experiments. Ji et al. [132] set up three holding platforms at 300 °C, 380 °C, and 700 °C based on the TG-DTA data to prepare 3YSZ samples with a relative density up to 99.95% and a flexural strength of 1008.5 MPa. Wang et al. [158] used thermal gravimetry-Fourier transform infrared spectroscopy (TG-FTIR) to analyze the changes in binder reaction during debinding and regulate the debinding parameters to prepare defect-free ceramic samples. The thermal debinding process was divided into a low-temperature phase in which organic matter melts and vaporizes and diffuses to the surface of the billet, and a high-temperature phase in which organic matter decomposes and carbonizes. At low temperature, the heating rate will make the gas diffuse faster and cause cracks, and at high temperature, if there is oxygen to produce carbon dioxide and a large amount of heat, which is not discharged in a timely manner, it will also cause cracks to appear.

Referring to the two-part debinding method for light-cured 3D printing of alumina [159,160], He et al. [105] prepared zirconia blades as well as complex components using vacuum debinding followed by air debinding. Wang et al. [161] conducted an in-depth study, comparing the results of the debinding of zirconia ceramic billets prepared in SLA under N₂ and air atmospheres, and investigated the thermal debinding in the effect of different atmospheres on the billets. Similarly, the debinding in air showed white color and deformation, and obvious cracks appeared between layers, while the debinding in nitrogen atmosphere turned the embryo into black color, and there was no obvious structural change in spite of thermal debinding. In the process of debinding, the organic matter decomposed and escaped, and in the nitrogen atmosphere, the carbon was difficult to react with oxygen and remained on the surface showing black color. The debinding rate of zirconia billets in nitrogen is slower than in air, resulting in lower internal stresses and avoiding the defect of interlayer cracking to a certain extent. However, the nitrogen degreased ligands showed defects such as bubbles and delamination during subsequent sintering, which could be caused by residual carbon on the surface, so the debinding process was optimized by debinding with nitrogen before 500 °C and passing air after 500 °C to remove residual carbon on the surface. Zhou et al. [162] compared the effect of degreasing on the embryos in nitrogen, air atmosphere and under vacuum. Among them, the surface of the samples debinding in nitrogen was smooth, while the samples debinding in air atmosphere and vacuum conditions showed obvious deformation and cracks. The intense reaction of organic matter under air atmosphere produced high thermal stress, while the gas pressure difference between inside and outside under vacuum caused cracks to occur. Ultimately, porous 3YSZ with better performance was prepared by nitrogen degreasing for electrolyte preparation. Sun et al. [163] used DLP to prepare zirconia ceramics with 3 mol.% yttria stabilized zirconia and investigated the effect of thermal degreasing atmosphere as well as the rate of temperature increase on the blanks. The results showed that both sets of embers debinding in air had white and cracked surfaces with high oxygen content and low carbon content, while debinding in argon yielded crack-free billets with black surfaces and about 99% densification after subsequent sintering. There was little difference between the two atmospheres with a heating rate of 0.2 and 0.5 °C·min⁻¹, but a slower heating rate gave a better surface quality.

Table 3. Effect of debinding and sintering process on the properties of ZrO₂ prepared by VPP.

Materials	Debinding	Sintering	Relative Density (%)	Grain Size (μm)	Mechanical Properties	Reference
ZrO2 (D50 = 440 nm)	600 °C in air atmosphere, 2 h	5 °C·min−1, 1600 °C, 2 h	99.18	-	Flexural strength 1210 MPa, Fracture toughness 14.10 MPa·m1/2	[101]
ZrO2	-	1450 °C, 1.5 h	99.48	0.5	Vickers hardness 15.11 GPa	[164]
3Y-TZP + ZrO2	1000 °C, in air atmosphere, 2 h	1600 °C, 2 h	96.40	-	Flexural strength 306.53 ± 6.03 MPa	[165]
10Sc1YSZ	1 °C·min−1, in air atmosphere, 800 °C	1600 °C, 2 h, in air	96	9–11	Vickers hardness 10.80 GPa	[24]
3YSZ	0.5 °C·min−1, 600 °C, successively in argon and air atmosphere, 2 h	5 °C·min−1, 1480 °C, 2 h	99.60	0.39	Flexural strength 1566 MPa	[166]
3YSZ	0.83 °C·min−1, 300 °C, 3 h, 380 °C, 3 h, 700 °C, 1 h	1450 °C, 150 min	99.95	2.64	Flexural strength 1008.50 MPa	[132]

Table 3. Effect of debinding and sintering process on the properties of ZrO₂ prepared by VPP.

Materials	Debinding	Sintering	Relative Density (%)	Grain Size (μm)	Mechanical Properties	Reference
ZrO2 (D50 = 440 nm)	600 °C in air atmosphere, 2 h	5 °C·min−1, 1600 °C, 2 h	99.18	-	Flexural strength 1210 MPa, Fracture toughness 14.10 MPa·m1/2	[101]
ZrO2	-	1450 °C, 1.5 h	99.48	0.5	Vickers hardness 15.11 GPa	[164]
3Y-TZP + ZrO2	1000 °C, in air atmosphere, 2 h	1600 °C, 2 h	96.40	-	Flexural strength 306.53 ± 6.03 MPa	[165]
10Sc1YSZ	1 °C·min−1, in air atmosphere, 800 °C	1600 °C, 2 h, in air	96	9–11	Vickers hardness 10.80 GPa	[24]
3YSZ	0.5 °C·min−1, 600 °C, successively in argon and air atmosphere, 2 h	5 °C·min−1, 1480 °C, 2 h	99.60	0.39	Flexural strength 1566 MPa	[166]
3YSZ	0.83 °C·min−1, 300 °C, 3 h, 380 °C, 3 h, 700 °C, 1 h	1450 °C, 150 min	99.95	2.64	Flexural strength 1008.50 MPa	[132]

8.2. Influence of the Sintering Process on Zirconia Ceramics by VPP

Currently, the most used method for sintering of light-cured 3D printed zirconia is conventional sintering, and the atmosphere generally used is air atmosphere, and the sintering temperature is mostly above and below 1450 °C. Wu et al. [104] prepared Al₂O₃-ZrO₂ composite ceramics with different sintering temperatures by using the SLA technique, and investigated the relative densities, microstructures, and mechanical properties of the samples. The results demonstrate that the density of the samples increases initially and subsequently decreases with elevated sintering temperatures, ultimately reaching a maximum density of 4.28 g·cm⁻³ at 1600 °C. It has been demonstrated that an increase in sintering temperature results in a decrease in both the number of grain boundaries and pores at the grain boundaries. Similarly, as the sintering temperature increased, the Vickers hardness of the samples exhibited an initial increase, followed by a subsequent decrease, reaching a maximum value of 17.6 GPa at 1550 °C. Fracture toughness is positively correlated with sintering temperature, with a maximum toughness of 5.2 MPa·m^1/2 measured at 1650 °C.

In addition to conventional sintering, microwave sintering is gradually being used. Khalile et al. [167] prepared 3Y-TZP ceramic dense and dotted parts by DLP and sintered them at 1550 °C using both conventional sintering and microwave sintering methods, respectively. All samples prepared after sintering by both methods showed no detectable interlayer delamination, relative densities greater than 98%, and almost no defects. In addition, the density of the samples prepared by conventional sintering was slightly higher. Compared with conventional sintering, the microwave-sintered ceramics exhibited finer microstructures in the dense and dot matrix samples. The samples with dense and dot matrix structures had close density and microstructure, and microwave sintering had almost no effect on the structure of the samples. They also investigated the results of microwave sintering of dense parts at different heating rates. The specimens had larger grain sizes at a ramp rate of 30 °C·min⁻¹ compared to sintering samples at ramp rates of 10 and 50 °C·min⁻¹. This study suggests that the combination of DLP and microwave sintering is an effective method for the rapid production of technical ceramics. Wang et al. [168] compared the difference in mechanical properties and aging resistance of SLA-prepared 3Y-TZP specimens sintered by conventional sintering and microwave sintering methods at sintering temperatures of 1350, 1425 and 1550 °C. It was found that the samples prepared by microwave sintering had higher Vickers hardness, especially the specimen with a sintering temperature of 1425 °C (14.07 GPa), which was 0.87 GPa higher than that of the sample sintered by the conventional sintering method at the same temperature. In addition, due to its higher surface LTD resistance, the ceramics prepared using microwave sintering exhibited higher low-temperature aging resistance, and the flexural strength showed a slow decreasing trend in the low-temperature aging at 50 h. The flexural strength of the samples prepared by microwave sintering decreased gradually, reaching a final reduction of 33.10 MPa. In contrast, the flexural strength of the samples prepared by conventional sintering exhibited a rapid decline after 50 h of low-temperature aging, from 1029.36 MPa to 825.71 MPa.

Sintering temperature is the parameter that has the greatest influence on the properties of ceramic components during sintering. For zirconia ceramics prepared by VPP, it is usually considered that the sintering temperature should be in the range of 1400–1600 °C. Liu et al. [169] investigated the effect of sintering temperature from 1200 °C to 1600 °C on the properties of YSZ ceramics, and the degree of densification increased with increasing the temperature, the porosity was reduced from 15.00% to 1.59%, and the grain size was increased from 0.13 ± 0.03 μm increased to 0.70 ± 0.20 μm, and thermal conductivity increased from 1.89 W·m⁻¹·K⁻¹ to 2.99 W·m⁻¹·K⁻¹. Zhao et al. [164] prepared zirconia building blocks at temperatures of 1400 °C, 1450 °C and 1500 °C, and investigated the optimum temperature and holding time for sintering. The XRD results showed that the samples prepared at 1450 °C with a holding time of 1.5 h contain the highest percentage of tetragonal zirconia. The samples have a uniform and dense microstructure with a grain size of 500 nm. Too high temperature and too long holding time will cause the samples to suffer from over-sintering phenomenon, grain growth and secondary recrystallization, resulting in a decrease in properties. Liu et al. [170] investigated the microstructure, hardness, and fracture toughness of SLA-printed Al₂O₃-ZrO₂ composite ceramics prepared by different sintering temperatures and holding times. As the sintering temperature increases, t the actual density initially rises and then falls, reaching a maximum value of 3.78 g·cm⁻³ at 1550 °C, and the hardness increases and then decreases, reaching a maximum hardness of 14.1 GPa at 1500 °C. In the case of sintering temperatures lower than 1500 °C, the sintering driving force is too low, resulting in a low densification and relatively poor mechanical properties. Conversely, when sintering temperatures exceed 1500 °C, the migration rate of pores is considerably lower than the migration rate of grain boundaries, so that part of the pores will remain inside the interior of grain to reduce the densification, while the abnormal growth of grains makes the mechanical properties worsen. In addition, prolonging the holding time makes the actual density increase, the holding time of more than 1 h is basically stable, and reaches the maximum density of 3.79 g·cm⁻³ at 1.5 h. The hardness increases with the increase in the holding time, and the maximum hardness is 13.3 GPa at 1 h. Too long a holding time leads to the abnormal growth of some of the grains and the uneven distribution of the size, which results in the reduction in the hardness. Goldberg et al. [171] added different contents of Co to 3Y-TZP powders containing 3 wt.% Al₂O₃, and the solid solution of Co led to an increase in defects, which in turn enhanced the ionic diffusion and lowered the sintering temperature to 1350–1400 °C. The hardness of the sintered powders increased with the holding time.

For the heating rate, Tan et al. [172] used 3Y-TZP as raw material, which was DLP printed, debinded, and then subjected to high-speed sintering, conventional sintering at 1580 °C, and conventional sintering at 1450 °C, respectively. The density, Vickers hardness, and fracture toughness of the sintered samples were also investigated. The relative densities of all the sintered samples were greater than 99.2%, and the densities were 6.02~6.035 g·cm⁻³, which were not much different from the theoretical density of 6.05 g·cm⁻³. Compared with conventional sintering, the grain size of the samples prepared by high-speed sintering method is obviously small, in addition to the balanced mechanical properties, the Vickers hardness of 11.85 GPa is slightly higher, however, the fracture toughness of 6.83 MPa·m^1/2 is obviously low. Compared with conventional sintering method, high-speed sintering cycle is much shorter, and it is one of the development directions of VPP 3D printing.

9. The Properties of Zirconia Ceramics and Composite Prepared by VPP

With the optimization of ceramic slurry formulation as well as heat treatment parameters, VPP has made relatively great progress in the field of ceramic precision manufacturing, and the properties of the prepared components are less different from those of conventional manufacturing methods [173,174]. Zhai et al. [175] compared the mechanical properties of zirconia ceramics prepared by SLA, DLP, and CNC, with hardnesses of 13.80 GPa, 12.13 GPa, and 14.02 GPa, and fracture toughness of 10.28 MPa·m^1/2, 11.54 MPa·m^1/2, and 13.63 MPa·m^1/2, respectively. Kim et al. [176] compared the properties of zirconia ceramics prepared by subtractive milling (SM) and SLA. The Vickers hardness, fracture toughness, and flexural strength of the specimens prepared by SM were 11.6 ± 0.6 GPa, 4.9 ± 0.4 MPa·m^1/2, and 927.4 ± 134.1 MPa, respectively. The Vickers hardness, fracture toughness, and flexural strength of the specimens prepared using SLA were 11.5 ± 0.4 GPa, 4.7 ± 0.3 MPa·m^1/2, and 865.7 ± 148.5 MPa, respectively. The samples prepared by the two methods have similar mechanical properties. Ellakany et al. [177] compared the properties of zirconia ceramics prepared by CAD/CAM milling and VPP prepared interim fixed dental prostheses (IFDPs). The mechanical properties of the IFDPs prepared by SLA did not differ much except that the modulus of elasticity of the IFDPs prepared by SLA was lower than that of the IFDPs prepared by milling. The reduction in the gap between the properties of zirconia prepared by VPP and those of conventional manufacturing methods makes it a solid basis for its wide application.

The overall performance of ceramic composites prepared using paste printing tends to be higher than that of single ceramics. The properties of ceramic composites are shown in Table 4 Coppola et al. [178] used DLP to prepare three types of Al₂O₃-ZrO₂ composite ceramics by mixing off-the-shelf ZrO₂ and Al₂O₃ slurries. Three Al₂O₃-ZrO₂ composites with ZrO₂ volume fractions of 15, 50 and 85%, respectively, were prepared and compared with pure Al₂O₃ and ZrO₂ ceramics. In the composites with low zirconia content, zirconia can pin down the alumina grain boundaries and play the role of grain refinement by inhibiting the abnormal growth of alumina grains, which makes the Vickers hardness of the composite with 15 vol.% ZrO₂ addition even higher than that of Al₂O₃, and however, further increasing the ZrO₂ volume fraction to 85 vol.% will decrease the Vickers hardness. Xing et al. [111] adjusted the content of ZrO₂ in ZTA in the low content range, and prepared Al₂O₃-ZrO₂ composites containing 5, 10, 15, and 20 vol.% ZrO₂. As the ZrO₂ content increases, the Vickers hardness of the samples decreases, but both fracture toughness and bending strength increase. Jiao et al. [129] prepared Al(OH)₃-ZrO₂ green bodies by mixing different contents of Al(OH)₃ and ZrO₂ by digital light processing, and Al₂O₃-ZrO₂ ceramic scaffolds containing 1 μm microscopic pores were prepared with the help of subsequent decomposition of Al(OH)₃. The prepared scaffolds have good biocompatibility and similar strength to human cancellous bone. Additionally, the porosity and pore size of the micropores could be adjusted by changing the amount of Al(OH)₃ added and the sintering temperature. Wang et al. [121] investigated the effect of different ratios of equiaxed to platelet-shaped Al₂O₃ particles on the properties with the overall addition of Al₂O₃ remaining constant. The samples with a powder mixture of 15 vol.% platelet-shaped Al₂O₃ and 5 vol.% equiaxed Al₂O₃ were found to have the most significant fracture toughness enhancement to 16.9 ± 0.8 MPa·m^1/2, which is distinctly higher than that of pure ZrO₂ ceramics, and without significant decrease in strength.

Zhang et al. [179] incorporated α-Al₂O₃ and La₂O₃ into (Ce, Y)-TZP-based ceramics and systematically examined the impact of these two oxides on the microstructure and mechanical properties of the composite ceramics. With the increase in α-Al₂O₃ content, the pinning effect of the second phase particles is enhanced, and the grain size of (Ce, Y)-TZP decreases gradually and reaches a minimum value of 0.75 μm in grain size at the α-Al₂O₃ content of 20 wt.%. However, excessive amount of α-Al₂O₃ would result in the formation of micropores, thereby compromising the mechanical properties of the material. When the addition of α-Al₂O₃ reach 15 wt.%, the material exhibits optimal mechanical properties, with Vickers hardness and flexural strength of 11.53 ± 0.27 GPa and 525.5 ± 21.3 MPa, respectively. The addition of a small amount of La₂O₃ will be the La³⁺ ions undergo grain boundary deviation, hindering the migration of grain boundary and thus refining the grains. The grain size of (Ce, Y)-TZP was reduced to 0.74 μm at the 0.1 wt.% La₂O₃ addition, while the hardness and flexural strength were increased to reach 13.03 ± 0.31 GPa and 550.4 ± 39.8 MPa, respectively. However, excessive amount of La₂O₃ would lead to the coarsening of α-Al₂O₃ grains, which would result in the reduction in mechanical properties. Fournier et al. [180] prepared a composite ceramic by SLA using a commercial powder consisting of Al₂O₃, SrAl₁₂O₁₉, and 11Ce-TZP. Compared to the specimens prepared by cold isostatic pressing followed by sintering, the specimens prepared by SLA showed a larger region of strontium aluminate aggregation and a larger size of the surrounding zirconia grains, which may be due to a number of factors, such as the organic medium and the molding method.

In addition to Al₂O₃-ZrO₂ ceramics, zirconia composites with other materials also tend to have a positive effect on mechanical properties. Zhang et al. [165] improved the strength of ZrO₂ ceramics by adding 3Y-TZP to m-ZrO₂. While ensuring a solid content of 50 vol.%, they added 0, 5, 7.5, and 10 vol.% of 3Y-TZP, respectively. After comparison, the highest relative density of 96.40% was found for the sintered samples with 3Y-TZP concentration of 7.5 vol.%. The flexural strength of the sintered samples exhibited an initial increase, followed by a subsequent decrease, with the maximum flexural strength of 306.53 ± 6.03 MPa observed at an 7.5 vol.% 3Y-TZP addition. The flexural strength demonstrated a strong dependence on the extent of transformation of m-ZrO₂, which reached 11.74 vol.% at this time. Cao et al. [130] combined zirconia with hydroxyapatite, and used the DLP technique to prepare ZrO₂/hydroxyapatite (HA) composite porous scaffolds to evaluate the properties. The compressive strength of the scaffolds first increases and then decreases with increasing the HA content. The highest strength of 52.25 MPa was obtained with the addition of 10 wt.% HA. In comparison, the compressive strength of the ZrO₂ scaffold without HA addition is 39.99 MPa. The improvement in compressive strength due to HA addition can be attributed to the pinning and bridging effects of the small HA crystals. Camargo et al. [181] used alumina and zirconia as raw materials by DLP to print mullite-zirconia composites. Sintering was carried out at 1600 °C and a flexural strength of 84 ± 13 MPa was achieved. The Vickers hardness and fracture toughness increased and then decreased with the increase in IDG content, and the best performance was achieved at an addition of 1 wt.%, which was 13.03 GPa of Vickers hardness and 6.33 MPa·m^1/2 of fracture toughness, respectively. Shi et al. [140] added nanoscale ZrO₂ to SiC for reinforcement, and the mechanical properties increase with the increase in nano ZrO₂ content, with a relative density of 91.1 ± 3.2%, Vickers hardness of 9.76 ± 0.92 GPa, and a flexural strength of 201.5 ± 11.4 MPa at a content of 10 wt.% ZrO₂.

Table 4. Properties of zirconia composite ceramics.

Fabrication Method	Composition of Starting Powders	Relative Density (%)	Grain Size (μm)	Vickers Hardness (GPa)	Fracture Toughness (MPa·m1/2)	Flexural Strength (MPa)	Reference
SLA	Al2O3 80 wt.%, ZrO2 20 wt.%	99.5	1.08 (Al2O3), 0.35 (ZrO2)	17.76 ± 0.21	5.72 ± 0.50	530.25 ± 29.5	[103]
SLA	Al2O3 80 wt.%, ZrO2 20 wt.%	99.7	1.07 (Al2O3), 0.34 (ZrO2)	17.6	5.2	-	[104]
SLA	Al2O3 85 vol.%, ZrO2 15 vol.%	-	-	14.1	4.05	-	[170]
SLA	Large-Al2O3 70 vol.%, Fine-Al2O3 15 vol.%, Fine- ZrO2 15 vol.%	99.4	3 (Al2O3)	19.20 ± 0.89	7.4 ± 1.02	575 ± 87	[111]
SLA	Al2O3 80 wt.%, ZrO2 (Y2O3 1 wt.%) 20 wt.%	96.65	3.05 ± 1.78 (Al2O3), 1.69 ± 0.68 (ZrO2)	16.2	7.4	572.0	[142]
DLP	ZrO2 20 wt.%, MgO 0.5 wt.%, Y2O3 1 wt.%, La2O3 0.25 wt.%	97.93	1.63 ± 0.04 (Al2O3), 1.14 ± 0.02 (ZrO2)	18.8	6.94	556.6	[182]
DLP	Al2O3 75 wt.%, 3YSZ 25 wt.%	99.4	-	17.40	7.76	516.7	[143]
DLP	3Y-TZP 85 vol.%, α-Al2O3 15 vol.%	99.6	0.61 ± 0.19 (ZrO2), 0.65 ± 0.27 (Al2O3)	15.36 ± 0.77	-	764 ± 136	[178]
DLP	3YSZ 80 vol.%, Al2O3 20 vol.% (microplatelet:equiaxed = 3:1)	97.4 ± 0.6	0.94 ± 0.11 (ZrO2)	12.5	16.9 ± 0.8	539	[121]
DLP	3YSZ 80 wt.%, Al2O3 20 wt.%	∼99	∼0.18	-	-	840	[133]
DLP	SiC 85 wt.%, ZrO2 10 wt.%, Al2O3 2.5 wt.%, Y2O3 2.5 wt.%	91.1 ± 3.2%	-	9.76 ± 0.92	-	201.5 ± 11.4	[140]

Table 4. Properties of zirconia composite ceramics.

Fabrication Method	Composition of Starting Powders	Relative Density (%)	Grain Size (μm)	Vickers Hardness (GPa)	Fracture Toughness (MPa·m1/2)	Flexural Strength (MPa)	Reference
SLA	Al2O3 80 wt.%, ZrO2 20 wt.%	99.5	1.08 (Al2O3), 0.35 (ZrO2)	17.76 ± 0.21	5.72 ± 0.50	530.25 ± 29.5	[103]
SLA	Al2O3 80 wt.%, ZrO2 20 wt.%	99.7	1.07 (Al2O3), 0.34 (ZrO2)	17.6	5.2	-	[104]
SLA	Al2O3 85 vol.%, ZrO2 15 vol.%	-	-	14.1	4.05	-	[170]
SLA	Large-Al2O3 70 vol.%, Fine-Al2O3 15 vol.%, Fine- ZrO2 15 vol.%	99.4	3 (Al2O3)	19.20 ± 0.89	7.4 ± 1.02	575 ± 87	[111]
SLA	Al2O3 80 wt.%, ZrO2 (Y2O3 1 wt.%) 20 wt.%	96.65	3.05 ± 1.78 (Al2O3), 1.69 ± 0.68 (ZrO2)	16.2	7.4	572.0	[142]
DLP	ZrO2 20 wt.%, MgO 0.5 wt.%, Y2O3 1 wt.%, La2O3 0.25 wt.%	97.93	1.63 ± 0.04 (Al2O3), 1.14 ± 0.02 (ZrO2)	18.8	6.94	556.6	[182]
DLP	Al2O3 75 wt.%, 3YSZ 25 wt.%	99.4	-	17.40	7.76	516.7	[143]
DLP	3Y-TZP 85 vol.%, α-Al2O3 15 vol.%	99.6	0.61 ± 0.19 (ZrO2), 0.65 ± 0.27 (Al2O3)	15.36 ± 0.77	-	764 ± 136	[178]
DLP	3YSZ 80 vol.%, Al2O3 20 vol.% (microplatelet:equiaxed = 3:1)	97.4 ± 0.6	0.94 ± 0.11 (ZrO2)	12.5	16.9 ± 0.8	539	[121]
DLP	3YSZ 80 wt.%, Al2O3 20 wt.%	∼99	∼0.18	-	-	840	[133]
DLP	SiC 85 wt.%, ZrO2 10 wt.%, Al2O3 2.5 wt.%, Y2O3 2.5 wt.%	91.1 ± 3.2%	-	9.76 ± 0.92	-	201.5 ± 11.4	[140]

In addition, the use of multiple materials in combination has been studied in other additive manufacturing technologies [183,184]. Coupling different material properties can be used to produce components that meet specific requirements. This is also one of the areas in which VPP technology is set to develop in the future.

10. Applications of ZrO₂ Ceramics by VPP

In addition to their excellent mechanical properties, zirconia and composite ceramics exhibit excellent biocompatibility and good electrical properties, which renders them highly versatile in the fields of biomedicine and electrical engineering. The VPP preparation method facilitates the expeditious fabrication of intricate structures or bespoke components, thereby fostering continuous research and development in these domains.

10.1. Biological Materials

Figure 10 demonstrates the current applications of VPP-printed zirconia in biomaterials. The relatively excellent mechanical properties of zirconia ceramics, especially due to their biological inertness, have made them a commonly used material in dentistry for the fabrication of crowns [185,186,187], bridges [188], implants [189] and abutments [22,164,190]. Traditional zirconia products for dental applications includes lost wax casting and CAD/CAM cutting techniques. In contrast, light-cured 3D printing can circumvent the disadvantages of lack of precision, poor adaptability, and material waste of conventional manufacturing methods, and at this stage, the performance of dental components prepared by VPP technology is not significantly different from that of conventional methods [191]. Currently, zirconia ceramics used in dentistry is dominated by 3Y-TZP and 5Y-TZP, of which 3Y-TZP has better mechanical properties, however, 5Y-TZP is more biocompatible and transparent [25].

Zhang et al. [192] added inert dental glass (IDG) to zirconia, with the aim of investigating the effect of IDG addition on microstructure and physical properties of sintered samples. The best performance of the specimens was obtained at 1 wt.% addition of IDG, with Vickers hardness, fracture toughness, and abrasion resistance of 13.03 GPa, 6.33 MPa·m^1/2, and 0.03 mg·mm⁻²·min⁻¹, respectively.

Figure 10. (a) Optical images of DLP-printed 5YSZ embryos and sintered crowns, as well as images of sintered crowns under light [185], copyright 2023, with permission from MDPI; (b) mandible and tooth models prepared using DLP [145], copyright 2022, with permission from MDPI; (c) images of ZrO₂ and ZrO₂-ZnO hips prepared with VPP [193], copyright 2019, with permission from Dove Medical Press; and (d) zirconium oxide denture teeth prepared by adding different oxides [194], copyright 2024, with permission from Elsevier.

Figure 10. (a) Optical images of DLP-printed 5YSZ embryos and sintered crowns, as well as images of sintered crowns under light [185], copyright 2023, with permission from MDPI; (b) mandible and tooth models prepared using DLP [145], copyright 2022, with permission from MDPI; (c) images of ZrO₂ and ZrO₂-ZnO hips prepared with VPP [193], copyright 2019, with permission from Dove Medical Press; and (d) zirconium oxide denture teeth prepared by adding different oxides [194], copyright 2024, with permission from Elsevier.

As a dental material, there is a requirement for wear resistance. Mirt et al. [195] investigated the effect of air particle abrasion (APA) on the subsurface damage and strength distribution of 3Y-TZP and 5 mol% yttria partially stabilized zirconia (5Y-PSZ) zirconia parts prepared by DLP for dental use. After APA treatment, the flexural strength of 3Y-TZP zirconia ceramics increased, which was attributed to the toughening of the t→m phase transition and the generation of residual compressive stresses in the surface layer, which significantly increased the characteristic flexural strength of the 3 mol.% yttria stabilized zirconia, the reason for which may also be the coercivity generated by APA-induced ferroelastic domain transformation. While the flexural strength of 5Y-PSZ zirconia ceramics after APA decreased significantly, the sub-stability of the tetragonal phase is missing, the phase transition toughening effect is partially or completely lost, and the performance is reduced. Due to the nature of zirconia itself, the stability of zirconia under long-term use needs to be ensured. Zhai et al. [196] compared the aging effects of zirconia ceramics prepared by SLA, DLP and computer numerical control (CNC) techniques under low temperature degradation (LTD) conditions, and despite the fact that the samples fabricated by SLA and DLP were more susceptible to the t/m phase transition, there was no significant degradation on mechanical properties, and even the flexural strength of the samples prepared by SLA improved after 5 h of aging. This indicates that the stability of dental zirconia ceramic restorations prepared by SLA and DLP techniques is good. In addition, since the white color of zirconia is different from the color of teeth, the color of denture can be adjusted by adding rare earth elements to zirconia, and so on [144]. Wan et al. [194] used DLP by adjusting the content of Fe₂O₃, Pr₆O₁₀, MnO₂, and Er₂O₃ in 3Y-TZP to prepare a sample that has excellent biocompatibility with a flexural strength of 821.01 ± 51.72 MPa and a compressive strength of 2.67 GPa ± 0.22 MPa with the same color as human natural teeth.

Similarly, zirconium oxide ceramics are often used as orthopedic implants and prosthetic materials in orthopedics due to their better mechanical properties of zirconium oxide ceramics as well as its excellent biocompatibility. Zhu et al. [193] prepared a hip prosthesis using a ZrO₂ material covered with a layer of ZnO nanoparticles on the surface, which did not differ in maintaining the mechanical properties from that of the ZrO₂ material without a layer of ZnO nanoparticles, but it had a higher antimicrobial effect and does not lead to an increase in biotoxicity. In addition, Schiltz et al. [197] investigated the wear behavior of zirconia ceramics prepared by VPP. Differences in wear behavior and phase composition of the laminated structures were investigated by pin-on-disk tests on zirconia specimens prepared with different SLA build-up orientations. All combinations of horizontal and vertical pin-disk assemblies show similar wear coefficients and phase transitions from monoclinic to tetragonal zirconia, with wear dominated by adhesive and abrasive wear. Fatigue wear is possible, but it only occurs after decades of use when zirconia is used as an artificial bone material.

However, as a prosthetic material, pure zirconia material lacks bioactivity, and its applications are often used in conjunction with other biologically active materials. By compositing or coating methods, it is possible to prepare biological components with excellent mechanical properties and biocompatibility. Jiao et al. [198] doped 20 wt.% hydroxyapatite into zirconium oxide to prepare porous bone scaffolds by DLP with a view to combining the mechanical properties of zirconium oxide with the high bioactivity of hydroxyapatite. The porosity was adjusted in the range from 35.53 to 61.75%, and all the scaffolds prepared had mechanical properties comparable to cancellous bone and were able to increase the compressive strength by 30% with appropriate increase in irregularity, which would also increase the bioactivity. Wu et al. [199] prepared slurries containing 1/6, 1/3 and 1/2 of the total volume of synthesized silicocarnotite powder (CPS) and fabricated gradient composite ceramics using stereolithography (SLA). The composite ceramics exhibited excellent mechanical properties, and CPS degradation released Ca²⁺, Si⁴⁺, and P³⁺ ions, which promoted cell growth and differentiation. Coppola et al. [200] prepared porous structured zirconia scaffolds by DLP using the method of coated hydroxyapatite. The coating process was optimized by applying a hydroxyapatite layer to the zirconia scaffolds using a one-step method; the compressive strength was tailored by adjusting different porosities, including the values for both cortical and cancellous bone. In addition to the use of hydroxyapatite to enhance bioactivity, the use of calcium silicate has also been investigated. He et al. [201] prepared zirconia/calcium silicate composite porous scaffolds by DLP. Incorporating with a small amount of calcium silicate destroys the pristine grain boundaries of zirconia ceramics and induces defects, which reduces the compressive strength of the scaffolds, whereas with increasing the calcium silicate content, the calcium silicate aggregates and forms a supportive structure, which in turn improves the compressive strength of the scaffolds. In addition, the increase in calcium silicate increases the cell attachment area and degrades earlier than zirconia, which provides space for new bone growth as a bone scaffold, i.e., this zirconia/calcium silicate composite porous scaffold is more suitable for bone repair than pure zirconia. Li et al. [202] prepared zirconia interference screws for anterior cruciate ligament repair by SLA, and comparatively investigated the mechanical properties of screws with different shapes. The maximum insertion torque for a hexagon drive screw with a top thread width of 0.4 mm and a thread depth of 0.8 mm was 1.064 ± 0.117 N·m, the ultimate load was 446.126 ± 37.632 N, and the stiffness was 66.33 ± 27.48 N·mm⁻¹. The bioactivities of the screws using the different coatings were compared, with the ZrO₂/PDA/RGD/Zn²⁺ bioactive coatings had the best bioactivity for the interfering screws.

Jiang et al. [89] prepared zirconia ceramics with different TPMS with a special porous structure by DLP using 3Y-TZP, and compared the mechanical properties and osteogenic capacity of each group. TPMS is a periodic smooth implicit surface with zero mean curvature, which has a high surface area-to-volume ratio and excellent mechanical energy absorption properties, and has a broader prospect than conventional porous structures for the applications of orthopedic repair materials.

10.2. Solid Oxide Fuel Cell (SOFC) Electrolytes

Ceramic solid oxide fuel cells (SOFCs) have been developed for decades and nowadays have a more mature technology route. However, traditional preparation methods can only produce simple structures, limiting the development of SOFCs. With the development of 3D printing technology, it is possible to prepare parts with complex structures, and zirconia, especially after doping stabilization, has become a suitable material to be used for the preparation of oxide fuel cell electrolytes due to its high ionic conductivity, i.e., mechanical properties. Table 5 summarizes the current information about the performance of zirconia SOFC electrolytes prepared by VPP. In 2014, Hernandez-Rodriguez et al. [203] attempted to prepare SOFC components using VPP. After measuring and comparing the electrical properties, it was concluded that the preparation process does not affect the electrical properties of 8YSZ, i.e., it is feasible to prepare zirconia SOFC electrolytes using VPP. Later, Masciandaro et al. [204] prepared dense self-supporting electrolytes by SLA using 3YSZ and measured the cell performance, but the peak power density at 900 °C was only 100 mW·cm⁻² due to the doping content (3 mol.%) of yttria.

Compared with 3YSZ, 8YSZ has a higher concentration of oxygen vacancies, better electrochemical properties, and a more stable phase at high temperatures, which makes it a more suitable material for the preparation of SOFC electrolytes. Currently, many efforts have been directed to the preparation of solid oxide fuel cell electrolytes using 8YSZ. Wei et al. [205] used 8YSZ to prepare zirconia electrolytes with a maximum power density of 176 mW·cm⁻² at 850 °C by DLP, and the performance has been on the same level as that of the zirconium oxide electrolytes prepared in the conventional way. However, due to the limited conductivity, more conductive scandium oxide stabilized zirconia (ScSZ) has been used, despite its unstable condition under some uses. Komissarenko et al. [24] modulated the slurry to prepare 6ScSZ and 10Sc1YSZ ceramic fittings, both of which had good conductivity. Later, Marquez et al. [100] prepared the ytterbium doped scandium oxide stabilized zirconia (YbScSZ) to break through the relatively low ionic conductivity of yttrium oxide stabilized zirconia. The prepared flat and corrugated electrolytes showed peak power densities of 300 mW·cm⁻² and 470 mW·cm⁻² at 900 °C, respectively, which improved the performance with respect to the conventional 8YSZ electrolyte.

In addition to improving the material to enhance the electrolyte performance, changing the shape of solid electrolytes to increase the effective contact area between the electrode and the electrolyte can also improve the performance of the electrolyte. Xing et al. [206] prepared corrugated electrolytes with 8YSZ by DLP and compared the effect of shape and thickness on the performance. A thickness of 200 μm corrugated at 800 °C increased the power density by nearly 32% compared to the same thickness of flat plate, and the optimal performance in the experiments was achieved by the 170 μm-thick corrugated electrolyte at 800 °C, which showed the maximum power density of 197.618 mW·cm⁻². Zhang et al. [81] prepared an ultrathin honeycomb-structured 8YSZ self-supporting electrolyte by DLP, and the maximum power density of the honeycomb specimen was 215.4 mW·cm⁻² at 800 °C, while that of a flat plate electrolyte prepared by the same method was 114.3 mW·cm⁻² [207]. Zhou et al. [162], on the other hand, prepared a porous three-cell monolithic SOFC using SLA stacks with 3YSZ material as the support and 8YSZ and (Ce_0.90Gd_0.10)O_1.95 as the bilayer electrolyte, and the flexural strength of the material was measured to be 46.6 MPa, while the maximum power density of the prepared samples at 800 °C was 683 mW·cm⁻², which was much better than that of the monolithic cells prepared by other conventional methods.

Table 5. Electrolyte performance of zirconia solid electrolyte for SOFCs prepared by VPP.

Materials	Printing Method	Ionic Conductivity (mS·cm−1)	Geometries (Thickness)	MPDs (mW·cm−2)	Reference
3YSZ	SLA	22 at 900 °C	flat	100 at 900 °C	[204]
			honeycomb-like	110 at 900 °C
8YSZ	SLA	40 at 850 °C	3-tube	230 at 850 °C	[99]
8YSZ	SLA	-	planar (300 μm)	197.6 at 850 °C	[84]
			concavo-convex (300 μm)	288.9 at 850 °C
8YSZ	SLA	30 at 900 °C	planar (250 μm)	260 at 900 °C	[6]
			corrugated (250 μm)	252 at 850 °C 410 at 900 °C
8YSZ	DLP	-	planar	114 at 800 °C	[81]
			honeycomb	215.4 at 800 °C
8YSZ	DLP	-	ripple-shaped (170 μm)	197.6 at 800 °C	[206]
8YSZ	DLP	21.8 at 800 °C	flat	114.3 at 800 °C	[207]
ScSZ	SLA	35 at 850 °C	planar (265 μm)	300 at 900 °C	[100]
			corrugated (265 μm)	470 at 900 °C

Table 5. Electrolyte performance of zirconia solid electrolyte for SOFCs prepared by VPP.

Materials	Printing Method	Ionic Conductivity (mS·cm−1)	Geometries (Thickness)	MPDs (mW·cm−2)	Reference
3YSZ	SLA	22 at 900 °C	flat	100 at 900 °C	[204]
			honeycomb-like	110 at 900 °C
8YSZ	SLA	40 at 850 °C	3-tube	230 at 850 °C	[99]
8YSZ	SLA	-	planar (300 μm)	197.6 at 850 °C	[84]
			concavo-convex (300 μm)	288.9 at 850 °C
8YSZ	SLA	30 at 900 °C	planar (250 μm)	260 at 900 °C	[6]
			corrugated (250 μm)	252 at 850 °C 410 at 900 °C
8YSZ	DLP	-	planar	114 at 800 °C	[81]
			honeycomb	215.4 at 800 °C
8YSZ	DLP	-	ripple-shaped (170 μm)	197.6 at 800 °C	[206]
8YSZ	DLP	21.8 at 800 °C	flat	114.3 at 800 °C	[207]
ScSZ	SLA	35 at 850 °C	planar (265 μm)	300 at 900 °C	[100]
			corrugated (265 μm)	470 at 900 °C

10.3. Other Applications

The high degree of freedom of additive manufacturing and the freeform fabrication with the absence of molds make it possible to use zirconia to conveniently prepare complex components that are more difficult to prepare by traditional methods. Figure 11 shows some other building blocks prepared using VPP. He et al. [105] used DLP to fabricate triangular zirconia tools with retracting grooves and honeycomb ceramic assemblies. The crystalline phase after sintering was t-ZrO₂ phase and the densities of the triangular tool were 97.14%, which was slightly lower than that of the zirconia tools prepared by conventional sintering method. The hardness of the prepared zirconia samples was 13.060 Gpa, and the fracture toughness was 6.038 MPa·m^1/2. Liu et al. [170] prepared ZrO₂-toughened Al₂O₃ gear-like ceramic components using SLA. Tang et al. [208] prepared zirconia orthodontic brackets for orthodontics by using two 3D printing techniques, namely, DLP and Material Jetting (MJ), and the properties of the printed components were similar to those of the conventional approach. Wang et al. [116], on the other hand, used DLP to prepare structurally complex partially stabilized zirconia (PSZ) molds for aluminum casting and subsequently successfully used these molds for aluminum casting. Although VPP-prepared zirconia components are still dominated by dental, orthopedic, and SOFC electrolytes, other application directions such as self-lubricating materials [209], and coupled manufacturing with multi-printing methods are also quite promising.

11. Conclusions and Prospects

This paper summarizes and introduces the mechanism of vat photopolymerization, which is currently common, and summarizes the status of the art on the preparation of ZrO₂ ceramics and composites using VPP, including the improvement of ceramic slurry, the optimization of the printing process, and the effect of debinding and sintering process parameters. Based on the above, methods to improve the rheological properties of ZrO₂ ceramic slurries, mechanical and other properties of ceramic parts are indicated. Finally, vat photopolymerization ZrO₂ ceramics and their composites are highlighted for applications in fields such as biomedical and energy, pointing out future opportunities and challenges. The primary conclusions that can be drawn from this review are as follows:

(1)

This review firstly summarizes the compositional formulations of zirconia and composites slurry for VPP preparation, and introduces the commonly used resins, dispersants and other slurry compositions. The research on the improvement of slurry properties is summarized, including the selection of powders with moderate particle size, suitable particle size distribution, surface modification of powders, and the regulation of the type and content of resins and dispersants. The smoothness of printing as well as the performance of the components was ensured by improving the interaction between powders and lowering the viscosity while ensuring the solid content. In addition, the performance of the composite powders on the slurry and printed products is summarized, and the slurry components need to be adjusted according to the nature of the material.

(2)

In order to further improve the rheological properties of the slurry and make up for the shortcomings in the application of dispersants, the ceramic powders can be surface-functionalized. Surface modification of zirconia powders using surface modifiers such as silane coupling agents reduces the hydrophilicity and improves the dispersion of the powders.

(3)

The design of surface weave and topology is an important ring that can take advantage of VPP. Software modeling such as CAD is used to design structures that are difficult to fabricate by conventional manufacturing methods. Surface texturing has a significant improvement on material properties, especially in terms of abrasion resistance and anti-slip. Bionic weaving will be one of the key directions for surface applications. The design of component topology can play a role in reducing component mass and improving mechanical properties. In addition, the influence of printing parameters such as light intensity, layer thickness, and printing angle on print quality is summarized. The printing parameters need to be dynamically regulated in conjunction with the curing properties of ceramic slurries, and the moderate printing parameters are in turn required for their curing properties.

(4)

The effects of process parameters on the components during debinding/sintering are summarized. Factors such as mode, atmosphere, temperature, and rate of temperature rise all have a significant impact on debinding/sintering results. The debinding method combining an inert atmosphere with an air environment can well reduce the possibility of defects appearing. Regulate the rate of temperature rise to avoid defects from excessive thermal stress. Select the appropriate temperature to remove organic matter during debinding, and promote the densification process without adversely causing excessive grain growth.

(5)

The high degree of liberalization of VPP provides new possibilities for the application of zirconia ceramics. Zirconia itself, with its favorable mechanical properties, high biocompatibility and stabilized electrical properties, has a good potential for applications in biomedicine and electronic energy with the help of VPP. Zirconia dentures, bone repair scaffolds and solid oxide fuel cell electrolytes prepared by VPP have excellent performance.

(6)

In the future, zirconia preparation using VPP will develop in line with additive manufacturing and VPP technology. Firstly, the development of 4D printing technology will extend the use of zirconia components to include the time dimension. Secondly, AI technology will be applied to reduce the time cost of research and development by predicting slurry performance. Additionally, combining multiple materials can produce gradient components tailored to specific requirements. Furthermore, high-throughput in situ monitoring technology can be used to adjust process parameters and improve the performance of zirconia prepared by VPP. Finally, attention should be paid to solving the regulatory and scaling issues that arise when applying VPP technology. The difficulties of establishing evaluation standards and mass production brought about by personalization must be overcome.