Preparation of MgAl₂O₄ Porous Ceramics for High-Temperature Flue Gas Filtration Application by In-Situ Decomposition Method

Although energy sources such as wind and solar power have developed rapidly, fossil fuels still remain the main source of energy generation. However, their combustion emits large amounts of high-temperature flue gases [1,2] containing acid corrosive gases (CO, CO₂, SO₂) [3] and dust particles (PM), which can destroy the ecological environment [4,5,6]. To reduce such environmental pollution, porous ceramics are used as filters to filter the flue gas before emission. Generally, SiC porous ceramics with their exceptional thermal stability [7] and substantial specific surface area are suitable for filtering flue. However, high-temperature flue gas can erode SiC ceramics, thereby reducing their service life. Furthermore, the susceptibility of SiC ceramics to oxidation limits their operating temperature and makes them prone to failure. Hence, it is worth investigating alternative materials to produce advanced porous ceramics with excellent inoxidizability, greater chemical stability, and improved filtration performance. Compared with SiC, MgAl₂O₄ offers high melting point, high temperature stability, and good corrosion resistance, making it a promising candidate for high-temperature filtration applications [2,8,9,10].

Gas permeability is one of the most important properties of porous ceramic filters and is affected by porosity, pore size, and pore connectivity [11]. To introduce pores into the ceramic, the pore-forming agent method, gel-casting method, and freeze-drying [12] have been applied. Kumar et al. [11] prepared Al₂O₃ based ceramic filters by adding a pore-forming agent method with characteristics of an apparent porosity of 39–70%, a compressive strength of 6–82 MPa, and a relatively high gas permeability coefficient of 0.38 × 10⁻¹⁰–9.15 × 10⁻¹⁰ m². Liu et al. [13] use gel-casting method and freeze-drying method to prepare the mullite porous ceramics with a high apparent porosity of 56.04–75.34%, a compressive strength of 0.61–3.03 MPa, and a gas permeability coefficient of 1.11 × 10⁻¹⁰–4.73 × 10⁻¹¹ m². Based on the above analysis, various porous ceramics with a high apparent porosity have been fabricated by these methods. However, the high porosity results in a low compressive strength. Compared with the above method, the in-situ decomposition method, which relies on the decomposition of raw materials, offers a more uniform pore size distribution, thereby helping to preserve the compressive strength of porous ceramic filters. For instance, Zhou et al. [14] prepared MgAl₂O₄ porous ceramics via the decomposition of bauxite and magnesite, with characteristics of a density of 3.43 g/cm³ and a flexural strength of 93.5 MPa. Wang et al. [15] prepared MgAl₂O₄ porous ceramics by the in-situ decomposition method, achieving an optimal sample with an apparent porosity of 50%, a flexural strength of 35.06 MPa, and a nitrogen gas flux of 3057 m³·m⁻²·h⁻¹ under the gas pressure of 0.1 MPa. It can therefore be concluded that the in-situ decomposition method is an effective method for improving the strengths of MgAl₂O₄ porous ceramic.

EDTA-MgNa₂ is an organic magnesium compound that decomposes to generate more gas during sintering. When used for the in-situ decomposition, it provides a new opportunity to enhance the pore interconnectivity between pores. During the sintering process, EDTA-MgNa₂ also releases Na⁺ ions that react with impurity ions in the raw materials to form a transitional liquid phase, thereby promoting the sintering reaction. Combined with the in-situ decomposition method, this approach enables the formation of uniformly sized pores, ultimately optimizing the balance between compressive strength and gas permeability. Based on the above analysis, preparing porous ceramics via the in-situ decomposition method using EDTA-MgNa₂ as a raw material is expected to yield high-temperature porous MgAl₂O₄ ceramic with high compressive strength and high gas permeability.

In this study, a MgAl₂O₄ porous ceramic filter was prepared via the in-situ decomposition method, using α-Al₂O₃, MgO, and EDTA-MgNa₂ as raw materials. The influence of EDTA-MgNa₂ content on the phase composition, microstructure, apparent porosity, pore size distribution, thermal expansion coefficient, mechanical properties, and gas permeability of MgAl₂O₄ porous ceramics was investigated to verify the feasibility of MgAl₂O₄ as a high-temperature flue gas filter.

α-Al₂O₃ (purity ≥ 99.99%, D₅₀ = 4.00 μm, Shandong Higiant High-Purity Alumina Technology Co., Ltd., Binzhou, China), MgO (purity ≥ 99.90%, D₅₀ = 5.00 μm, Hebei Zhuyan Alloy Materials Co., Ltd., Xingtai, China), and EDTA-MgNa₂ (Shandong Smatts Agricultural Science and Technology Co., Ltd., Weifang, China) were used as the main material. C₂H₅OH (Tianjin Fuyu Chemical Co., Ltd., Tianjin, China) was chosen as a binder.

In order to form stoichiometric MgAl₂O₄ spinel, the weight ratio of Al₂O₃ and MgO was chosen as 2.53:1. Firstly, the raw materials were mixed with ethyl alcohol using a mortar and pestle for 5 min according to the designed compositions described in Table 1. EDTA-MgNa₂ with the contents of 0.0 wt%, 2.5 wt%, 5.0 wt%, 7.5 wt%, or 10.0 wt% was introduced, and the samples were named as M1, M2, M3, M4, M5. Dry mixtures were pressed uniaxially in a hydraulic press at 20 MPa to φ30 × H10 mm and φ20 × H20 mm. The green samples were then dried in an oven at 110 °C for 24 h. Finally, the samples were placed in a muffle furnace with an air atmosphere and sintered at a heating rate of 5 °C/min. During the sintering process, they were pre-fired at 700 °C for 3 h, then sintered at 1500 °C for 3 h, and allowed to cool down to room temperature at a rate of 5 °C/min inside the furnace.

Thermal analysis of the EDTA-MgNa₂ was detected using a simultaneous TG-DSC apparatus (STA449, NETZSCH, Selb, Germany). The phase compositions of samples were determined via X-ray diffraction (Rigaku Smart Lab 9kW, Rigaku, Tokyo, Japan) with Cu Kα radiation in the 2θ range of 5–90° and a scanning speed of 2°/min. The microstructure was studied by field emission scanning electron microscopy (QUANTA 250 FEG, FEI Company, Hillsboro, OR, USA). The pore size distribution was examined via mercury porosimetry (AutoPore Ⅴ 9600, Micromeritics, Norcross, GA, USA). The bulk density and apparent porosity were determined by the Archimedes method using water as the medium.

The linear change of the sample before and after calcination was determined by the following equation:

where l₁ and l₂ are the diameters of the samples before and after sintering.

The weight change of the sample before and after calcination was determined by the following equation:

where m₁ and m₂ are the weights of the samples before and after sintering.

The compressive strength was assessed using a universal testing machine (CMT5305, MTS Systems (China) Co., Ltd., Shanghai, China). The sample was cylindrical with 20 mm in diameter and 20 mm in height. Thermal expansion analysis was carried out in a thermomechanical analyzer (TMAPT1000, LINSEIS, Selb, Germany) from room temperature to 600 °C. The theoretical liquid phases were calculated by the Equilib mode of the thermo-chemical software Factsage 6.2.

As shown in Figure 1, a self-made pressure drop tester was used to measure the pressure drop of a gas across the ceramics. The device pipe’s inner diameter and the test samples’ diameter were 30 mm each. The tests were carried out at the gas velocity of 10–19 m/s. According to previous research [16,17,18,19], the gas permeability coefficient and the ceramic permeance were calculated by using Equation (3) and Equation (4):

where ψ is the permeability coefficient (m²), η is the dynamic viscosity of air, Q is the gas flow rate (m³/s), t is the thickness of the sample (m), A is the filter area (m²), and ΔP is the pressure drop (Pa).

The term “permeance” refers to gas permeance (mol·m⁻²·s⁻¹·Pa⁻¹), while n is the flow rate of the gas (mol·s⁻¹).

In order to determine the heat treatment process, the dehydration and thermal stability of EDTA-MgNa₂ were investigated through the TG-DSC analysis, as depicted in Figure 2. The DSC curve can be divided into three stages. In the first stage, with the temperatures ranging from ambient to 186.27 °C, the mass of specimens was reduced to 79%, due to the vaporization of the physisorbed water. In the second stage, when the temperature increased to 624.17 °C, EDTA-MgNa₂ decomposed, releasing gases such as CO and CO₂ [20], and formed MgCO₃ and Na₂CO₃, as shown in Equation (5).

Further increasing the temperature to 667.67 °C, MgCO₃ decomposed into MgO and CO₂, as depicted in Equation (6).

Upon further heating to 700 °C, the mass of MgO (the decomposition products of EDTA-MgNa₂) remained stable, with a mass residual rate of 15%. Therefore, since the decomposition of EDTA-MgNa₂ is completed at 700 °C, a holding step was chosen to be carried out at this temperature to ensure complete decomposition of EDTA-MgNa₂.

Figure 3 shows the XRD pattern of MgAl₂O₄ samples with different EDTA-MgNa₂ contents, indicating that the main phase of these ceramics was MgAl₂O₄. It is well established that Al₂O₃ and MgO can react in situ to form MgAl₂O₄ (Equation (7)). In this study, MgO originates from two sources: the MgO present in the raw materials and the MgO derived from the decomposition of EDTA-MgNa₂ (Figure 2). During the sintering process, both types of MgO can react with the Al₂O₃ in the raw materials to form MgAl₂O₄.

Figure 4 shows the micromorphology of fired samples M1–M5 with different EDTA-MgNa₂ contents. Combined with EDS, the irregular particles are confirmed as MgAl₂O₄ grains, consistent with the XRD results shown in Figure 3. As shown in Figure 4a1–e1, the grain growth mechanism of MgAl₂O₄ can be described as follows: Initially, Al₂O₃ and MgO react to form MgAl₂O₄. As the reaction proceeds, more MgAl₂O₄ grains are formed and accumulated, resulting in larger grains until they reach the critical size. Since the MgO derived from the decomposition of EDTA-MgNa₂ has higher reactivity, increasing the EDTA-MgNa₂ content leads to a greater number of fine grains.

The SEM images of polished sample M5 sintered at 1500 °C are given in Figure 5. As shown in Figure 5a, the matrix is mainly MgAl₂O₄, with a large number of uniformly distributed pores. Additionally, a few regions exhibiting a multi-particle packing structure surrounded by MgAl₂O₄ and annular pores (Figure 5b) are observed. By combining Figure 5d–g, it reveals that this region with multi-particle packing structure corresponds to partially reacted MgO, and there are pores formed by the decomposition of EDTA-MgNa₂ between the MgO particles.

The pore structure parameters of MgAl₂O₄ porous ceramics are depicted in Figure 6. Analysis of Figure 6a reveals a unimodal distribution of the pore size in the porous ceramics with different EDTA-MgNa₂ contents, indicating that the pore size of the samples is uniformly distributed. As shown in Figure 6a, with the EDTA-MgNa₂ content increasing from 0 to 10 wt%, the pore size of MgAl₂O₄ porous ceramics shows an overall upward trend. The prominent peak value of the M5 sample reaches 4.09 μm, and a large pore size is conducive to enhancing the gas permeability of MgAl₂O₄ porous ceramics. As illustrated in Figure 6b, higher EDTA-MgNa₂ content resulted in an upward shift in the cumulative pore volume curve, indicating that samples containing higher EDTA-MgNa₂ content have better pore structure. The steep curve in the range of 0.70–7.20 μm reflects a rapid increase in cumulative pore volume, indicating that most pores fall within this size range, and a large and uniform pore size distribution is conducive to gas passing and flue gas filtration.

Figure 7 shows the weight change and linear change of the MgAl₂O₄ porous ceramics with different EDTA-MgNa₂ contents. As shown in Figure 7, with the increase in EDTA-MgNa₂ content from 0.0 to 10.0 wt%, the weight change gradually increased from 1.32% to 10.57%, due to the decomposition of EDTA-MgNa₂ and release of gas [20]. The linear change of MgAl₂O₄ porous ceramics indicates that all samples exhibit thermal expansion characteristics, mainly due to the formation of spinel, which is accompanied by volume expansion [15,21].

Figure 8 illustrates the apparent porosity and bulk density of MgAl₂O₄ porous ceramics with different EDTA-MgNa₂ contents. The results showed that the apparent porosity of MgAl₂O₄ porous ceramics showed an upward trend from 47.27% to 56.28%, and the bulk density showed a down trend from 1.72 g·cm⁻³ to 1.45 g·cm⁻³ with increasing EDTA-MgNa₂ content. Then, by combining the apparent porosity with the linear change in Figure 6, it can be found that the MgAl₂O₄ spinel formation is beneficial for the apparent porosity due to the volume expansion. And as mentioned above, the EDTA-MgNa₂ is evenly distributed between the particles. After sintering, a large number of pores formed due to the decomposition of EDTA-MgNa₂. Thus, the MgAl₂O₄ porous ceramics prepared by this in-situ decomposition method exhibited high porosity, an important condition for achieving high gas permeability.

Figure 9 displays the compressive strength of MgAl₂O₄ porous ceramics with different EDTA-MgNa₂ contents. As observed, the compressive strength of MgAl₂O₄ porous ceramic decreased gradually from 35.56 to 10.93 MPa with increasing EDTA-MgNa₂ content. Compressive strength is a crucial index of material mechanical properties, while porosity is a key factor that is usually negatively correlated with material strength. The relationship between the two parameters can be described by the Gibson-Ashby model, as expressed by Equation (8) [19,22,23]:

where σ and σ_s represent the compressive strength of porous and dense material, respectively, P represents the porosity, and m is a constant.

As shown in Figure 8 and Equation (8), the compressive strength is negatively correlated with porosity (Figure 9b), indicating that higher apparent porosity reduces the load-bearing area of the material, thereby decreasing the compressive strength (Figure 9a). Thus, the porosity of the MgAl₂O₄ porous ceramics increased, and the strength decreased. However, the MgAl₂O₄ porous ceramics produced by the in-situ decomposition method still possess sufficient compressive strength to enable their industrial applications for high-temperature gas filtration.

Figure 10 presents the coefficients of thermal expansion (CTE) of MgAl₂O₄ porous ceramics prepared with varying EDTA-MgNa₂ contents. It is evident from the figure that the CTE of the MgAl₂O₄ porous ceramic increased gradually with the increasing test temperature, due to the average distance between atoms in the lattice increasing with the increase of temperature [24,25]. At 600 °C, as the content of EDTA-MgNa₂ increases from 0.0 to 2.5 wt% and from 5.0 to 7.5 wt%, the CTE values of the MgAl₂O₄ porous ceramic increase. This is because MgAl₂O₄ itself expands during heating. Although pores in the sample can absorb the displacement of some solid particles caused by heating, the poor connectivity of the pores limits their ability to improve the CTE. As the content of EDTA-MgNa₂ increases from 2.5 to 5.0 wt% and from 7.5 to 10.0 wt%, the CTE values of the MgAl₂O₄ porous ceramics decrease. This is attributed to the greater number of pores in the sample, which absorb the displacements of some solid particles caused by heating, thereby reducing the CTE. As a result, the MgAl₂O₄ porous ceramic with 10.0 wt% EDTA-MgNa₂ shows the lowest CTE of 7.93 × 10⁻⁶/K.

Pressure drop is an important parameter to characterize the gas permeance of ceramic filters, which is caused by the energy loss of gas flow in the pipeline. Figure 11a shows the relationship between EDTA-MgNa₂ content and the pressure drop at flow velocities of 10–19 m/s. As shown in Figure 11a, the results indicate that the pressure drop gradually increased with increasing gas velocity, because resistance is generated as gas molecules collide with each other due to internal friction as they flow through a pipeline [26]. This resistance reduces the gas velocity, thereby increasing resistance to gas flow and leading to a rise in the pressure drop. However, as the gas velocity increases, the pressure drop of the M5 sample generally remains at a lower value, which is conducive to enhancing the gas permeability of MgAl₂O₄ porous ceramic in practical applications. As gas velocity increased to 19 m/s, the pressure drop of MgAl₂O₄ porous ceramic first increased and then decreased. The reason why the M2 sample has a higher pressure drop than the M1 sample is attributed to the presence of isolated small pores in the M2 sample, which will lead to longer and more tortuous paths in the sample when gas flows through. Subsequently, the number of interconnected pores in MgAl₂O₄ porous ceramic gradually improves, leading to a decreasing trend in pressure drop. Ultimately, the M5 sample achieved the minimum pressure drop of 3.39 hPa.

Figure 11b shows the gas permeability coefficient and gas permeance values of the MgAl₂O₄ porous ceramics with different EDTA-MgNa₂ contents at a gas flow velocity of 19 m/s. It can be observed that, as fundamental performance parameters of filter materials, the gas permeability coefficient and gas permeance indicate that all MgAl₂O₄ porous ceramics exhibit good gas permeability, consistent with the gas permeability coefficient reported in the literature [27]. Among them, the M5 sample had the maximum gas permeability coefficient and gas permeance of 8.84 × 10⁻⁹ m² and 1.52 mol·m⁻²·s⁻¹·Pa⁻¹, respectively, consistent with the phenomenon that the pressure drop decreases to 3.39 hPa. In summary, the content of EDTA-MgNa₂ can improve the gas permeability while maintaining the high filtration precision of MgAl₂O₄ porous ceramics, and the M5 sample is the optimal one.

The above results indicate that the MgAl₂O₄ porous ceramics with high gas permeability are successfully prepared using MgO, Al₂O_3, and EDTA-MgNa₂ as raw materials through the in-situ decomposition method. The process will be discussed in three stages, as illustrated in Figure 12, focusing on EDTA-MgNa₂ decomposition, spinel formation, and pore structure.

(a) Initial Stage: As shown in Figure 12a, the green body primarily consists of large MgO and Al₂O₃ particles, and smaller EDTA-MgNa₂ particles. Pores (Pore 1) are observed in the green body, arising from particle packing during the dry pressing process.

(b) Decomposition Stage: During the heating process, the organic backbone of EDTA-MgNa₂ decomposes, and the C, H, and N elements react with oxygen to generate gases such as CO and CO₂, which are subsequently released. When the temperature reaches about 700 °C, EDTA-MgNa₂ particles completely decompose and form MgO grains with numerous pores (Pore 2), which may be the main reason for the significant increase in apparent porosity as the content of EDTA-MgNa₂ increases (Figure 8). As shown in Figure 12b, the presence of pores isolates most particles from one another (Figure 5b), thereby providing more interfacial area that serves as nucleation sites.

(c,d) Sintering Stage: As the sintering temperature increases, the liquid phase forms on MgO particles due to the presence of impurities in the raw materials and the Na⁺ released from the EDTA-MgNa₂ decomposition. The effects of temperature and EDTA-MgNa₂ content on the liquid phase content of samples are shown in Figure 13. As can be seen, the liquid phase content increases gradually with higher EDTA-MgNa₂ content, reaching 2.85 wt% at 1500 °C when the EDTA-MgNa₂ content is 10 wt%. This increase in liquid phase accelerates the mutual diffusion of Mg²⁺ and Al³⁺, thereby promoting the formation and deposition of MgAl₂O₄ spinel.

To achieve the electric neutrality during the formation of spinel, 2-mol Al³⁺ diffuses into the MgO side to produce 1 mol MgAl₂O_4, whereas 3-mol Mg²⁺ diffuses into the Al₂O₃ side [28,29]. As shown in Figure 5a,b, the presence of MgAl₂O₄ rims around those pores indicates that they were previously occupied by MgO particles, which led to a quicker formation of MgAl₂O₄ on the Al₂O₃ side, thereby triggering the Kirkendall effect and creating Kirkendall voids (Pore 3 and Figure 5c) [29,30]. Due to the different rate of ion diffusion, neck bonds will appear between particles, leading to agglomeration (Figure 4), which helps to reduce the negative impact of increased apparent porosity on the compressive strength of MgAl₂O₄ porous ceramic. In turn, during the sintering stage, MgAl₂O₄ grains will rearrange, and pores between particles will merge and grow. This phenomenon is conducive to improving the pore size (Figure 6) and ultimately enhancing the gas permeability of MgAl₂O₄ porous ceramic (Figure 10).

When compared to various porous ceramic filters (Figure 14 and Table 2), the MgAl₂O₄ porous ceramic filters developed in this work exhibit an outstanding overall balance of gas permeability and compressive strength. Notably, the compressive strength reaches 10.93 MPa, the highest among all compared samples, indicating superior structural integrity and resistance to mechanical stress, which are critical in harsh, high-temperature filtration environments. In addition, the material shows a gas permeability coefficient of 8.84 × 10⁻⁹ m², surpassing many porous ceramic filters. Overall, this finding facilitates further optimization of the gas permeance in high-temperature flue gas filters.

To enhance the gas permeability and achieve high compressive strength of MgAl₂O₄ porous ceramics, EDTA-MgNa₂ was introduced as a pore former and sintering aid by the in-situ decomposition method. The effect of the content of EDTA-MgNa₂ on the gas permeability and performance of ceramics was investigated. The conclusions reached are as follows.

The authors gratefully acknowledge the financial support provided by the Taishan Scholar Program of Shandong Province, Shandong Postdoctoral Science Foundation, Leader of Scientific Research Studio Program of Jinan, University of Jinan Disciplinary Cross-Convergence Construction Projects 2023, the Natural Science Foundation of Shandong Province, the Jinan City-School Integration Development Strategy Project, and the Shandong Provincial Key Research and Development Program.

Methodology, J.L.; Investigation, J.L.; Writing—Original Draft Preparation, J.L.; Data Curation, J.L., C.Z.; Writing—Review & Editing, W.Z., Z.Z.; Supervision, L.Z., J.K., X.J.; Project Administration, W.Z.; Funding Acquisition, D.Z.

Not applicable.

Not applicable.

The data are available from the corresponding author on reasonable request.

This research was funded by the Taishan Scholar Program of Shandong Province (No. tsqn202306225), Shandong Postdoctoral Science Foundation (No. SDBX2023025), Leader of Scientific Research Studio Program of Jinan (No. 2021GXRC082), University of Jinan Disciplinary Cross-Convergence Construction Projects 2023 (Nos. XKJC-202301 and XKJC-202311), the Natural Science Foundation of Shandong Province (Nos. ZR2024QE044 and ZR2025QC605), the Jinan City-School Integration Development Strategy Project (Nos. JNSX2023015 and JNSX2023018), and the Shandong Provincial Key Research and Development Program (2025CXGC020107).

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.