Phosphorus pentoxide, alumina, and chromium trioxide were purchased from Zhongfan Dongsheng New Material Technology Co., Ltd., and zirconium dioxide was purchased from Aladdin (98.5%≤ purity ≤100%) with a particle size of 80-110 nm. Deionized water was made in the laboratory using LD DI Micro.

In a given amount of deionized water, certain amounts of phosphorus pentoxide, aluminum oxide, and chromium trioxide were added sequentially, where the molar ratio of phosphorus pentoxide/aluminum oxide/chromium trioxide/water was 2:1:1:5. Mixing was performed by magnetic stirring for 20 min to ensure that the reaction between the ingredients was sufficient, and finally a dark green aluminum-chromium phosphate viscous liquid was formed. Subsequently, weighed zirconium dioxide was inverted in the mortar. Dark green alumi-num-chromium phosphate solution was poured into the mortar containing zirconium dioxide, where the zirconium dioxide and alumi-num-chromium phosphate had a mass ratio of 2:1. Next, grinding was performed at room temperature for 10 min. No granularity was obvious until the mixing of the homogeneous. The resulting solution was poured into the oscillation mold and then placed into an 80^∘C curing oven for 1.5 h to obtain the corrosion-resistant zirconium-based phosphate ceramics.

The microstructures of the samples were analyzed using a field emission scanning electron microscope (SEM; Quanta, FEG250, USA), and the elemental composition of the samples were determined using an SEM equipped with an energy-dispersive spectrometer (EDS). Nanostructure analyses of the samples was performed using a field emission transmission electron microscope (FE-TEM; JEOL, JEM-2800) equipped with EDS. The TEM test samples of the cured zirconium-based phosphate ceramics were prepared using a focused ion beam scanning electron microscope (FIB-SEM; FEI Scios). The chemical bonding state and physical phases of the samples were obtained by X-ray diffraction (XRD; Panalytical Empyrean X-ray Diffractometer). The thermal stability of the samples was measured by performing thermogravimetric differential scanning calorimetry (TG-DSC; Netzsch, Germany) at 25-1400^∘C with an air heating rate of 10^∘C/min. Compressive testing was performed using a mechanical testing machine (Instron 3369) at a displacement loading rate of 2.0 mm/min. The thermal conductivity of the samples was measured at room temperature (LFA457, NETZSCH, Germany).

The ablation resistance and ultra-high-temperature thermal insulation properties of the zirconium-based phosphate ceramics were tested using an oxyacetylene flame according to the GJB323A-96 standard. An infrared thermometer (ENDURANCE 1R) was used to measure the temperatures of the ablated surfaces, and a thermocouple was employed to test the back temperatures of the samples. The mass ablation rate and line ablation rate were calculated using the following equations [35]:

where Rm is the mass ablation rate, Δm is the mass change between before and after ablation, Rl is the line ablation rate, Δl is the change in center thickness between before and after ablation, and t is the ablation time.

Based on the exothermic self-curing characteristics of phosphate material sol-gel polymerization, combined with the high-melting-point oxidation resistance of ZrO₂, a zirconium-based phosphate ceramic composite material with integrated heat protection and insulation was constructed. We firstly designed an aluminum-chromium phosphate matrix solution, which was prepared from phosphoric acid, aluminum hydroxide, chromium oxide, and hydrogen peroxide through a complex chemical reaction, with excellent thermal stability and high temperature resistance [36], as shown in Fig. 1. Then, the solution of alumi-num-chromium phosphate and ZrO₂ underwent an acid-base reaction and specific crosslinking condensation to achieve room temperature curing. Specifically, the aluminum-chromium phosphate solution and ZrO₂ were mixed uniformly through rapid grinding and stirring. Then, the weak alkaline ZrO₂ partially reacted with the acidic solution to generate Zr²⁺, which partially replaced the Al³⁺ and Cr³⁺ in the aluminum-chromium phosphate as the reaction continued. The generated molecular structure continued to form crosslinks to the surrounding area and gradually created a three-dimensional network of macromolecules. These processes released a certain amount of heat to promote gradual evaporation of the free water and other substances, causing the above mixed solution to lose its macro-adhesion gradually, until complete curing was achieved. The prepared zirconium-based phosphate ceramic composites exhibited many fascinating properties, such as a high density of 2.42 g/cm³ and a low porosity of 21.7% after curing at room temperature, excellent thermal stability, small macroscopic shrinkage during heat treatment from room temperature to 1500^∘C, and a mass loss of less than 2.2% in thermogravimetric tests from 25^∘C to 1400^∘C. The zirconium-based phosphate ceramics exhibited outstanding high temperature resistance and thermal insulation properties, as evidenced by the protection of flowers from scorching during 100 s of flame heating with an alcohol lamp (Fig. 1(c)), the maintenance of the structural integrity of samples during 30 s of ablation at an ablation temperature of 2527^∘C (Fig. 1(d)), and the maintenance of a back temperature of the samples (with a thickness of approximately 10 mm) of less than 250^∘C at all times.

The macroscopic morphology of the zirconium-based phosphate ceramic material, which exhibits a light green color, is clearly visible in Fig. 2(a), and the XRD data indicate that the main physical phases are ZrO₂ as well as zirconium-containing phosphate materials (Fig. 2(b)). The microstructure of the zirconium-based phosphate ceramic material was characterized using SEM. The SEM images in Fig. 2(c) clearly shows that the zirconium-based phosphate ceramic material mainly consists of white and gray-white phases that were identified by EDS analysis to be ZrO₂ and zirconium-containing aluminum-chromium phosphate. The ZrO₂ particles are densely surrounded by fine zirco-nium-containing phosphate particles (Fig. 2(d)). These particles are formed by crosslinking polycondensation to form a three-dimensional mesh structure distributed around the ZrO₂, thus effectively bonding the larger particles together.

To further analyze the microstructure of the zirconium-based phosphate material, this study utilized transmission electron microscopy to analyze the phase distribution. Fig. 3(a) shows a high-angle annular dark-field (HAADF) image of the zirconium-based phosphate material, which shows that the phases are closely combined, and no defects such as cracks and pores appear. Subsequently, EDS mapping was used to analyze the material, as shown in Fig. 3(b). Clearly, the material is divided into two phases. Phase I is primarily composed of zirconia phase and phosphorus phase enriched on its surface, while Phase II is dominated by aluminum phosphate phase. Comparative analysis revealed a distinct selective enrichment of phosphorus in Phase I. The formation mechanism of this phenomenon can be attributed to the following synergistic processes[1-3]: first, Al³⁺ and Cr³⁺ preferentially coordinate with phosphate ions in the solution to form an amorphous AlPO₄-CrPO₄ network structure; second, due to the significant interfacial energy difference between this amorphous network and the zirconia surface, free phosphate ions (PO₄ ^3-) that did not participate in coordination are expelled during the condensationcrystallisation process; Subsequently, these negatively charged phosphate ions are captured by the abundant Zr-OH₂ ⁺/Zr-OH groups on the zirconia surface through electrostatic interactions; Finally, during the dehydration-condensation process, the phosphate ions form stable compounds with the zirconia surface through the formation of Zr-O-P covalent bonds or the generation of zirconium pyrophosphate (ZrP₂O₇, confirmed by XRD in Fig. 2b), thereby creating a characteristic phosphorus enrichment layer on the surface of the zirconia particles. Moreover, it is found that the atomic percentages of oxygen and zirconium were found to be relatively high, reaching 59.22% and 33.74% respectively. The selected-area electron diffraction (SAED) of the zirconium-based phosphate material provides a deeper understanding of the crystal structure of the composite phase. Fig. 3(c) clearly confirms that the two phases are closely combined and that no obvious intermediate phase has formed. The electron diffraction pattern of ZrO₂ was analyzed, and the interplanar spacing was measured to be 3Å (Fig. 3(d)), corresponding to the (111) plane. The interplanar spacing of the (-211) plane measured in another direction was 2.78Å. Therefore, this ZrO₂ mainly had a cubic phase structure. The corresponding aluminum-containing phosphate phase was mainly amorphous. No lattice fringes appear in this image, and only a small number of diffraction spots exist in the electron diffraction pattern (Fig. 3(f)).

The dynamic oxidation of zirconium-based phosphate ceramic composites was characterized by conducting TG-DSC tests in an air atmosphere from room temperature to 1400^∘C. The TG analysis results in Fig. 4(a) indicate that the sample mass exhibits a significant nonlinear change with increasing temperature, which can be divided into three characteristic transition stages. In the low-temperature range of 0-67^∘C, an abnormal increase in mass is observed, primarily due to the inherent hygroscopic properties of phosphate materials. Previous studies have confirmed [1] that such materials can adsorb 2-4wt% moisture at room temperature and form a characteristic "weight gain peak" when heated to ∼70^∘C. As the temperature rises to the 67-800^∘C range, the sample mass decreases continuously from 102.1% to 97.9%. This mass loss process involves multiple mechanisms [2,3], including the gradual removal of physically adsorbed water and crystalline water, condensation dehydration reactions of interlayer and framework hydroxyl groups, and the release of volatile phosphorus-oxygen species during the condensation or phase transformation of a small amount of phosphate. Notably, in the high-temperature range of 800-1400^∘C, the mass change exhibits a fluctuating pattern of first increasing then decreasing: the slight mass increase observed between 800-1200^∘C (97.9%→98.4%) can be attributed to the reoxidation process of lowvalent ZrO_2-x and P₂O_5-x components in the material [4], while the subsequent decrease in mass (98.4%→97.8%) observed in the 1200-1400^∘C range is directly related to the high-temperature decomposition reaction of pyrophosphate (ZrP₂O₇→ZrO₂+P₂O₅↑). This conclusion is highly consistent with the phase evolution results shown in Fig. 4(d) for the 1200-1500^∘C temperature range. The aforementioned mass change characteristics systematically reveal the complex thermal transformation behavior of phosphate materials at different temperature intervals.

In order to analyze in more detail, the heat resistance and the changes in the morphology of the phases during the warming process, a long-time sintering test was performed in a muffle furnace. Fig. 4(b) presents the changes in density and compressive strength after treatments at different temperatures, wherein the density first decreases and then increases. By contrast, the compressive strength initially increases gradually with the temperature rise, then gradually decreases, and finally increases again. This indicates that the temperature treatment affects the fundamental properties of the material. Therefore, the microstructural changes of the phases under different temperature treatments must be examined.

To that end, Fig. 4(c) shows macroscopic photographs of the samples after curing at room temperature and after treatment at 1500^∘C for 5 h. The samples before and after treatment are intact and crack-free, and the shrinkage rate is almost zero, verifying the excellent heat resistance. The differences in the changes after different temperature treatments can also be seen in the XRD patterns of the prepared zir-conium-based phosphate ceramics (Fig. 4(d)), which contain more amorphous diffraction peaks after curing at room temperature. Further, crystallization occurs with the gradual loss of physical phases such as bound water as the temperature rises. The complete disappearance of the diffraction peaks of AlPO₄H₂O after treatment at 800^∘C is attributed to the dehydration of AlPO₄H₂O into AlPO₄. The disappearance of the diffraction peaks of ZrP₂O₇ from 1200^∘C to 1500^∘C is mainly due to the transformation of ZrP₂O₇ into ZrO₂, and after a long time of treatment at 1500^∘C, the samples gradually form phosphate materials with ZrO₂ as the main physical phase.

Fig. 5 shows microstructural maps of the surface and cross-sections of the zirconium-based phosphate ceramic material after treatment at room temperature, 500^∘C,800^∘C,1200^∘C, and 1500^∘C. The zirconiumbased phosphate ceramic material exhibits a denser microstructure both on the surface and in the cross-section after curing at room temperature. With increasing temperature, the internal microstructure of the sample after treatment at 500^∘C starts to decrease in densification, which is manifested by the appearance of loose pores in the internal microstructure (Fig. 5(b)). This characteristic is due to the heat absorption and volatilization of the residual free water sequestered in the sample by the ambient temperature curing. This micromorphological transformation phenomenon is consistent with the density change before and after the treatment temperature (Fig. 4(b)), where the density decreases from 2.469 g/cm³ at room temperature to 2.216 g/cm³ at 500^∘C. As the treatment temperature continues to increase, the particles on the sample surface and in the cross-sectionm, as well as the particle boundaries, all become progressively more pronounced, and a similar phenomenon of sinter crystallization occurs. When the temperature increases to 1500^∘C, white and grayish-white particulate matter on the nanoscale appears on the surface and in the cross-section of the sample, and the particles are distributed in the sample with a close inter-particle arrangement (Fig. 5(d)). Combining these results with the XRD data (Fig. 4(d)) indicates that the white particulate matter is mainly ZrO₂ and that the gray phase is primarily AlPO₄.AlPO₄ mainly plays the role of bonding ZrO₂ particles, which also ensures that the zirconium-based phosphate ceramic composites have good high-temperature stability and temperature resistance.

Next, based on the systematic microstructural and phase analysis above, the trends in the density and compressive strength data shown in Fig. 4(b) were analyzed, as discussed below.

With the increasing processing temperature, the density is low at first and then becomes higher. The physical phase and morphology analysis above revealed that with low-temperature curing, zirconium-based phosphate ceramic solidifies more free water and part of the bonded water and other phases, increasing the sample mass. This phenomenon is due to the high density and ambient curing. The same low-temperature-cured sample has many defects, which leads to the compressive strength of this sample being the lowest. As the treatment temperature increases, the free and bound water volatilize, and the sample mass and density decrease. When the temperature is 800^∘C, the density appears to increase, but the quality decreases. The data in Fig. 4(b) demonstrate that the density augmentation originates mainly from the dominant sample contraction mechanism. This increase in density with increasing temperature is mainly due to the gradual sintering and ceramization of the sample, whereas an increase in the sample mass results in a slow increase in density.

In addition, the zirconium-based phosphate ceramic composites exhibit excellent thermal insulation properties. The back temperature of the composite with a thickness of 10 mm was always lower than 193^∘C when heated on a high-temperature platform at 400^∘C for a continuous period of 30 min. As can be seen from the infrared thermograms in Fig. 6, the temperature difference between the heated side and the back side of the composite sample is stably controlled above 207^∘C. Similarly, the thermal conductivity of the material was evaluated using three-dimensional temperature distribution maps during the heating process, as shown in Fig. 6(f-k). Throughout the continuous heating process, a distinct temperature gradient persisted within the sample. The purple box in the figure designates the sample area. The red region at the top represents the temperature at the bottom of the sample, while the opposite side of the box corresponds to the top. By drawing a line from the top of the sample to the temperature axis, the temperature at the top can be observed to increase with the heating time. However, the rate of temperature rise is slow, and the variation in the width of the isotherms is minimal. This indicates that the sample has low thermal conductivity at this temperature, demonstrating excellent insulating properties. The thermal conductivity in the Z-direction after different temperature treatments is 1.027,0.827,0.934,1.221, and 1.794 W/m⋅K (The data are presented as shown in S1). The thermal conductivity decreases and then increases as the temperature increases, which illustrates that the zirconium-based phosphate ceramic materials should be treated at certain temperatures. The changes in the combination of the mixture phases, microscopic morphology, and density indicate that the thermal conductivity of the zirconium-based phosphate ceramic materials decreases due to the formation of a certain pore structure in the bulk of the heat-treated sample. However, the phosphate phase gradually transforms and volatilizes at an excessively high temperature, and ceramization occurs, which increases the thermal conductivity.

A higher temperature gun was used to characterize the thermal insulation properties of the samples. Fig. 7 shows the infrared thermograms of a 10-mm-thick zirconium-based phosphate ceramic material under a high-temperature gun. After achieving a surface temperature of 727^∘C with 240 s of continuous heating, the back temperature reaches a maximum of 156^∘C. Thus, cooling of 571^∘C in the thickness direction is achieved. Moreover, an obvious temperature gradient curve also appears at the heating surface in Fig. 7(b)(c), indicating that the zirco-nium-based phosphate ceramic material has excellent thermal insulation properties in both the Y- and Z-directions. Similarly, Fig. 7(e) depicts the three-dimensional temperature distribution of the sample during ablation by a butane torch. Analysis of the overall isothermal curves reveals both longitudinal and transverse temperature gradients across the sample. Notably, the width of the isothermal lines gradually increases from the high-temperature regions to the low-temperature ones, which suggests that high-temperature zones are localized. As a result, heat accumulates at the ablation site and cannot rapidly disperse throughout the sample. These findings further confirm the outstanding insulating properties of the sample.

The insulation mechanism was also analyzed and is schematically illustrated in Fig. 8. The thermal conductivity of the zirconium-based phosphate ceramics is mainly composed of the gas thermal conductivity (λg), solid thermal conductivity (λs), radiation (λr), and convection (λ con). In this case, λ con is negligible because the pore size of the homogeneous composite is much less than 1 mm. The mean free path of gas molecules is given by Eq. (3). In this equation, λg represents the mean free path of air molecules, k is the Boltzmann constant, T is the thermodynamic temperature, d is the effective diameter of the molecules, and p is the gas pressure. As can be seen from the formula, the mean free path of gas molecules is positively correlated with temperature. When the temperature rises, the mean free path increases significantly to approximately 500 nm, which is larger than the majority of internal pores of the zirconium-based phosphate material (the pore size distributions after various heat treatments are shown in Fig. S2). Thus, the mean free path can be neglected [36,37]. During heating, thermal energy is transferred mainly through the phosphate phase and its interfaces. The phonon scattering effect dissipates a significant amount of energy as it passes through the abundant interfaces of the matrix.

Consequently, the composite exhibits excellent thermal insulation properties. The effective synergy of low thermal conductivity and high mechanical properties makes these zirconium-based phosphate ceramics among the most competitive candidates for thermal protection in extreme aerodynamic thermal environments.

Next, the zirconium-based phosphate ceramic was tested at 2527^∘C under ablation with an oxyacetylene flame for 30 s. During ablation, the temperature of the ablated surface of each sample was tested using an infrared thermometer, and the back temperature of each sample was measured using a thermocouple. The temperature of the ablation surface rapidly reached 2527^∘C, whereas the back temperature of the sample peaked at 247^∘C (Fig. 9(a)). Thus, the sample cooled down by at least 2280^∘C over a distance of 10 mm in the longitudinal direction, confirming its excellent thermal insulation properties. Fig. 9(b) shows the microscopic topography of the ablated surface as a whole, demonstrating that many micrometer-sized holes had formed on the surface; the surface profile after ablation is shown in Fig. S3. The intricate mechanism governing the generation of this extensive array of micropores on the ablated surface is elucidated in Fig. 9(c). Under elevated temperatures, phosphates melt, during which the molten phosphates encapsulate a profusion of zirconia particles. As ablation proceeds, the molten phosphates and internal components partially volatilize and decompose, creating voids. Simultaneously, zirconia particles initiate a process of sintering-induced growth. Ultimately, a distinct hierarchical structure is formed, characterized by a zirconia layer interspersed with a pore-structured layer derived from the residues of volatile decomposition. Thermodynamically, the volatilization and decomposition processes play a crucial role in dissipating substantial amounts of heat. By efficiently removing thermal energy, these processes impede the inward conduction of heat, thus endowing the material with outstanding high-temperature insulation capabilities.

Under oxyacetylene flame ablation at 2500^∘C, the zirconium-based phosphate material exhibited mass ablation and linear ablation rates of 0.0173 g/s and 0.0114 mm/s, respectively. In this case, mass ablation primarily resulted from three concurrent mechanisms: (i) thermal volatilization, (ii) pyrolytic decomposition, and (iii) mechanical erosion. During high-temperature ablation, the phosphate phase underwent progressive decomposition and molten-phase migration, while the exposed ZrO₂ ceramic phase on the ablated surface experienced grain coarsening through sintering. The ZrO2-rich layer that formed in situ contributed significantly to the ablation resistance of the material through two key effects: (1) the formation of a protective ceramic barrier, and (2) enhanced thermal stability via sinter-induced densification. Fig. 10 comparatively evaluates the linear ablation rates of zirconium-phosphate-based materials and other refractory materials under extreme thermal conditions. The experimental data reveal that the zirconium phosphate system demonstrates a 15-30% lower linear ablation rate than conventional ultra-high-temperature ceramics while maintaining comparable structural stability above 2500^∘C. Although its linear ablation rate is slightly higher than that of some polymer-derived ceramics and aerogels, the zirconium phosphate exhibits a substantially higher operational temperature threshold, exceeding these organic-inorganic hybrids by 700-900^∘C. This unique combination of exceptional ablation resistance (0.0114 mm/s at 2500^∘C) and exceptional thermal stability (up to 2500^∘C) positions this zirconium phosphate material as a promising candidate for thermal protection systems requiring balanced performance under extreme thermomechanical loads.

Fig. 11 shows the microscopic morphology and composition of the sample ablation surface after the sample was ablated in the oxyacetylene flame for 30 s. The ablated surface has obviously formed a dense white layer of material, and combining this finding with the EDS results (Fig. 11(d)(e)) indicates that the white material phase is ZrO₂. The ZrO₂ oxide layer formed in the central zone of ablation is much denser, with no obvious defects. The ablation transition zone forms a massive ZrO₂ of approximately 20μ m, whereas the ablation transition zone forms smaller ZrO₂ clasts. The reason for the formation of this unique shape is that when the oxyacetylene flame reaches the surface of the sample, the ablation center area preferentially undergoes high-temperature ablation. During which the phosphate phase decomposes and melts at high temperature, and the ZrO₂ phase leaks out of the surface. This ZrO₂ then grows with the help of the molten aluminum phosphate and heat, flat sintering to a certain degree, which is the most important factor in the formation of this unique shape. Thermal ablation occurs later in the ablation transition and edge zones than it does in the ablation center. Further, the ablation flame has no mechanical impact, the phosphate phase melts and sinks, and ZrO₂ in the edge region grows slowly, resulting in the appearance of a certain gap. Ablation occurs on the surface as the decomposing phosphate phase both volatilizes and melts, and the molten phosphate phase deposits on the lower layer. Hence, the phosphate phase is not detected on the surface; instead, a large amount of bare ZrO₂ leakage occurs during ablation, and the molten phosphate phase acts as a binder to firmly bind the surface layer of ZrO₂ during ablation, thus improving its resistance to ablative corrosion.

The microstructure of the sample cross-section after ablation is shown in Fig. 12. The ablation layer has a dense ZrO₂ layer with a thickness of approximately 2.12 mm (Fig. 12(a)), and the oxide layer is structurally intact without cracks or other defects, which can effectively prevent the ablation flame from directly destroying the interior of the sample. The structural changes can be clearly seen in the ablation transition layer in Fig. 12(b), where the top-down microphase gradually becomes less dense, and ZrO₂ tends to increase in granularity, with no obvious structural deformation or visible collapse. This is attributed to the ablation that occurs in this region, where heat transfer from the surface layer to the transition ablation layer results in the decomposition of the phosphate in this layer. This decrease is also caused by the release of pyrolysis gases, raising the pressure inside the sample, which attenuates the continued transfer of heat to the lower layers to some degree (Fig. 12(d)). On the other hand, the sintering growth of ZrO₂ particles is hindered by the pressure, which gradually forms 2-6μ m particles with smooth surfaces and a certain pore structure between the particles. The heat from the ablation surface is transferred to the original material layer shown in Fig. 12(c), but by then, its heat has already been mostly dissipated. Hence, the bottom of the layer is at a maximum temperature of 247^∘C, which is generally low, and the bonding morphology features within the zirconium-based phosphate ceramic are fully preserved.

This comprehensive analysis of the ablation process reveals the underlying mechanism, as illustrated in Fig. 12(d). The post-ablated sample exhibits a distinct gradient structure comprising three well-defined layers from top to bottom: (i) the ablation layer, (ii) the ablation transition layer, and (iii) the original material layer. During the initial ablation stage, the surface of the ablation layer experiences severe degradation, including the thermal decomposition and melting of phosphate phases, accompanied by the sintering-induced grain growth of zirconia particles. As ablation progresses to the intermediate stage, the accelerated sintering of zirconia particles forms a continuous zirconiarich protective layer (ablation layer), which effectively impedes flame penetration. Concurrently, the molten phosphate phase partially volatilizes and partially migrates downward, with the descending molten phosphate providing thermal energy to promote gas evolution in the transition layer. In the final ablation stage, the rapid release of pyrolysis gases from the transition layer generates significant internal pressure, which establishes a thermal barrier that effectively limits heat transfer to deeper regions. This mechanism maintains the temperature of the bottommost layer at 247^∘C, thereby preserving the structural integrity of the original material. The resulting multi-layered architecture demonstrates synergistic protection: the ZrO₂-rich surface layer provides ablation resistance, and the porous transition layer offers thermal insulation to protect the original material layer below.

The above results show that zirconium-based phosphate ceramics undergo complex oxidation, decomposition, volatilization, and other processes at high temperatures. The relevant literature states that the possible reactions of zirconium-based phosphate ceramics at high temperatures are as follows [47-49]:

In order to understand the ablation behavior of the zirconium-based phosphate ceramics further, the equilibrium thermodynamics of these materials was calculated, and the gas volatility was plotted as a function of P_O2 and temperature. The graph was obtained using FactSage 7.3 software and data from the NIST-JANAF database, and the fluctuation graph corresponding to 2500^∘C in Fig. 13 shows that the wave diagrams of the zirconium-based phosphate ceramic materials as a function of the P_O2 partial pressure can be divided into four states[50]:

(I) When P_O2<10-11.5, the oxygen potential is low, the oxides are unstable, and only gas-phase monomers exist, including aluminum, zirconium, dichromated phosphorus, and phosphorus, with no condensed phase.

(II) When 10-11.5<P_O2<10-7, elemental gas phases such as phosphorus, aluminum, and di-zirconiated phosphorus are still present; the oxide phase is unstable; and the condensed phases are Al₂O₃,CrO, ZrO₂,P₂O₅, and Cr₂O₃.

(III) When 10-7<P_O2<10-5, P_O2 gas appears, the gas phase of phosphorus decreases, and the condensed phase consists of Al₂O₃,CrO, ZrO₂,P₂O₅, and Cr₂O₃ slag.

(IV) When P_O2>10-5, the phosphorus gas phase is completely transformed into PO and PO 2 gas phases, and the condensed phases are Al₂O₃,CrO,ZrO₂,P₂O₅, and Cr₂O₃.

In conclusion, the volatilization of low-melting-point phases occurs first in the zirconium-based phosphate ceramics under ablation at 2500^∘C. As the partial pressure of oxygen decreases, Al-O and Cr-O gas phases begin to appear. Finally, the Zr-O gas phase appears, suggesting that ZrO₂ is the most difficult phase to volatilize because it has the lowest vapor and decomposition pressures. The formation of a dense zirconia layer on the surface after ablation was also verified.