9,10-dicarboxylic acid anthracene was synthesized according to the process listed in the references.¹⁴⁾ The ZrCl₄ (0.985 mmol) and acetic acid (1.5 mL) were dissolved in DMF (15 mL), which was recorded as solution A. 9,10-dicarboxylic acid anthracene (0.822 mmol) and triethylamine (0.5 mL) were dissolved in DMF (20 mL); this was recorded as solution B. Then, solution B was slowly added to solution A; this was followed by adding of triethylamine (2 mL) to the reaction solution, which was maintained at 318 K for 15 h. After the reaction, the yellow solid was filtered and repeatedly washed with DMF; it was then immersed in a 323 K water bath in acetone for 3 days. Finally, the yellow solid was filtered, dried under vacuum at 393 K, and kept in a sealed bag; this resulting material was named UIO-66 (H₂DCA)-1. At the same time, the samples that were subjected to solvent thermal reaction were named UIO-66 (H₂DCA)-2; these samples had the same raw material ratio as that of UIO-66 (H₂DCA)-1. Moreover, the chemical stability of UIO-66 (H₂DCA)-1 was tested under different conditions.

Samples of zirconium metal organic framework material UIO-66(H2DCA)-1 were immersed in hydrochloric acid, sulfuric acid, nitric acid, phosphoric acid solution (5 mL,1 mol/L) and sodium hydroxide solution (5 mL,1 mol/L), ammonia solution, triethylamine, and ethylenediamine (5 mL) at 298 K for 24 h. The same amounts of samples were immersed in two high pressure reaction kettles with hydrochloric acid (5 mL, 1 mol/L) and water at 373 K for 5d; then, the product was filtered and dried.

The thermal stability of the materials was measured using a synchronous TG-DSC thermal analyzer (STA-409PC Netzsch, heated to 1173 K at a rate of 5K/min under air gas). To check functional group changes on the samples under different environmental conditions via attenuated total reflectance (ATR) technique, FTIR spectra were obtained using an FTIR instrument (Nicolet 6700, 4000 ~ 500 cm⁻¹ scans range, 2 cm⁻¹ resolution and 16 scans).

SEM analysis (JSM-7500F JEOL) was used to capture and determine the morphologies of the crystalline UIO-66(H₂DCA) samples. X-ray powder diffraction patterns were obtained with an X-ray diffractometer (MiniFlex 600, Cu Ka radiation, λ= 1.5406 Å, 0.02° step size); patterns were used to investigate changes of the crystal structure. N₂ adsorption (JW-BK132F) was used to determine the pore size and surface area of the UIO-66 (H₂DCA) samples. The samples were evacuated at 393 K under high vacuum for 2 h prior to the adsorption measurements.

The way to investigate the ideal crystal structure of UIO-66 (H₂DCA) is as shown in Fig. 1.

Each SBU of UIO-66 (H₂DCA) is coordinated by twelve 9,10-dicarboxylic acid anthracenes and four ligands, as shown in Fig. 1(a), The distances between the zirconium atoms of the SBUs and the parameter of octahedrally shaped pores are 15.837 Å, 15.913 Å, and 16.179 Å, as can be seen in Fig. 1(b); the SBU and the ligands forming the 3D structure of UIO-66 (H₂DCA) are shown in Fig. 1(c). From Fig. 1(d), it can be seen that the XRD patterns of the UIO-66 (H₂DCA) are the same as those of the UIO-66, as has been reported before.⁶⁾ This suggests that UIO-66 (H₂DCA) and UIO-66 have similar crystal structures with good crystallization. The crystalline form of UIO-66(H₂DCA)-2, synthesized with solvent thermal treatment, is better than the crystalline form of UIO-66(H₂DCA)-1. That is say, the crystal grains of UIO-66(H₂DCA)-2 are larger and more complete. The crystal sizes of UIO-66(H₂DCA)-1 and UIO- 66(H₂DCA)-2 were captured and determined by SEM, as shown in Fig. 2. The crystallite size of UIO-66(H₂DCA)-2 was similar to an octahedral shape at about 500 nm, but the particle size of UIO-66 (H₂DCA)-1 was found to have dropped; this was confirmed by XRD.

Figure 3 displays N₂ adsorption isotherms of the UIO-66(H₂DCA) samples. All samples of UIO-66(H₂DCA) have the characteristic of micropore structure. The properties are given in Table 1. UIO-66(H₂DCA)-2 has a BET surface area and pore volume of 586 m²/g and 0.2691 m³/g, respectively.

Thermal stability analysis of UIO-66 (H₂DCA)-1 was conducted using a TG-DSC, with results as shown in Fig. 4. It is clear that four stages of weight loss occurred for UIO-66 (H₂DCA)-1. According to the thermal decomposition process, it can be deduced that the first weight loss at around 323 K should be attributed to the volatilization of acetone; the second weight loss at around 353 K is due to the evaporation of DMF and surface adsorbed water; the third weight loss at about 523 K is due to the beginning of scission of about coordination key chain of Zr-O and the CO₂ decomposition of the 9, 10 - anthracene dicarboxylic acid ligand; and the fourth stage of weight loss at 583 K-773 K is in virtue of the further decomposition of UIO-66 (H₂DCA)-1 to form carbonaceous material and ZrO₂. It is obvious that the thermal stability of UIO-66 (H₂DCA)-1 can reach 583 K. The thermal stability data for UIO-66 (H₂DCA)-1 and other MOF materials is listed in Table 1. From Table 2, we can see that UIO-66 (H₂DCA)-1 has a better thermal stability.

The stability of UIO-66 (H₂DCA)-1 in water was checked by XRD, with results shown in Fig. 5. It can be seen that the UIO-66 (H₂DCA)-1 crystal structure is unchanged when sample is put in boiling water; that is, the UIO-66 (H₂DCA)- 1 sample has a high water resistance.

Figure 6 displays PXRD patterns of UIO-66 (H₂DCA)-1 in different acid environments. The stability of UIO-66 (H₂DCA)-1 in hydrochloric acid was investigated by XRD, with results as shown in Fig. 7. It can be seen that the structure of UIO-66 (H₂DCA)-1 is unable to remain stable in hydrochloric acid and, with increasing time, its structure is gradually destroyed and finally completely decomposes. The structure of UIO-66 (H₂DCA)-1 is known to be incapable of maintaining its stability in hydrochloric acid due to the distribution of Cl⁻¹. Compared with SO₄⁻² and PO₄⁻³ ions, the Cl⁻¹ ion takes a very long time to destroy the structure of UIO-66 (H₂DCA)-1 completely. Fig. 8 provides the FTIR spectra of the UIO-66 (H₂DCA)-1 sample after acid treatment. It can be clearly seen that, after treatment with hydrochloric acid, sulfuric acid, and phosphoric acid, the structural damage was decomposed into the 9,10- two carboxylic acid of H₂DCA and zirconium salt. It should be recognized that a broad IR band in the region of 2500-3000 cm⁻¹ occurred owing to the carboxyl hydroxyl. The peak at 1676 cm⁻¹ belongs to the stretching vibration of carbonyl. It is clear that the peak at 1017 cm⁻¹ is due to PO₄³⁻ vibration, as shown Fig. 6(d). As a result of the strong coordination ability of Cl⁻¹, sulfate ion and phosphate ion were able to destroy the original Zr-O coordination bond; that is, the UIO-66 (H₂DCA)-1 sample was decomposed into 9, 10- dicarboxylic acid anthracene and the corresponding zirconium salt.

As is known to all, UIO-66 (H₂DCA)-1 is a zirconium oxygen coordination compound. It is clear that the structure of UIO-66 (H₂DCA)-1 is stable in nitric acid, slightly damaged in hydrochloric acid, and completely destroyed in sulfuric acid and phosphoric acid; these conditions obtain because nitrate ion cannot provide a lone pair electrons to form a zirconium coordination bond, and UIO-66 (H₂DCA)-1 is naturally stable in nitric acid. Because the electronegativity values of phosphorus and sulfur are less than that of oxygen, and because oxygen has a strong attraction for electrons, it is easy to contribute a lone pair of electrons to oxygen to form the coordination bond, that is, SO₄⁻² and PO₄⁻³, with a zirconium ion form the zirconium phosphate coordination bond and the zirconium sulfate coordination bond; the original Zr⁴⁺ and carboxylic (-OOC) coordination bond is completely destroyed in sulfuric acid and phosphoric acid. The conditions obtain because the levels of electronegativity of chloride ions and oxygen ions are relatively close, and the coordination ability of the chloride ion is not strong enough. In the presence of Cl⁻¹ with zirconium ion, it is more difficult to form zirconium chlorine coordination than it is to form SO₄⁻² or PO₄⁻³ with the zirconium ion. Thus, UIO-66 (H₂DCA)-1 exhibits different levels of stability in different acidic media.

Figure 9 shows the XRD patterns of the UIO-66 (H₂DCA)-1 sample in different alkali environments. It is obvious that the structure of the UIO-66 (H₂DCA)-1 sample was completely destroyed by the sodium hydroxide solution, the ammonia solution, and ethylenediamine. The Zr-O coordination bond is easily disrupted because the volume of hydroxide ion is small and this ion has strong coordination ability. The compound ethylenediamine is an organic alkali and, owing to its strong coordination ability for its two amino groups and its small steric hindrance, also easily causes the structure of UIO-66 (H₂DCA)-1 to become unstable. It is very interesting that triethylamine is also an organic alkali, but the structure of UIO-66 (H₂DCA)-1 still maintains its stability. This is because triethylamine finds it very difficult to enter into the MOF structure due to its large steric effect; thus, UIO-66 (H₂DCA)-1 can be stable in triethylamine.

The FTIR spectra of the samples of UIO-66 (H₂DCA)-1 after alkali treatment are shown in Fig. 10. The peaks at 686 cm⁻¹, 867 cm⁻¹, 1396 cm⁻¹, 1425 cm⁻¹, and 2966 cm⁻¹ are attributed to zirconium hydroxide, as shown in Fig. 8(b). Compared with the data shown in Fig. 8(a), the peaks at 636 cm⁻¹ and 1025 cm⁻¹ disappear in Fig. 8(c) due to the ammonia solution, which can ionize a small amount of OH⁻, which may do great damage to the Zr-O coordination bond. The peaks at 821 cm⁻¹, 1383 cm⁻¹, 1462 cm⁻¹, and 2169 cm⁻¹ emerge owing to the strong basicity and small steric hindrance of ethylenediamine, which make it is easy to attack the Zr-O coordination bond.

Therefore, this research shows that the structure of UIO- 66 (H₂DCA)-1 is easily destroyed by an inorganic alkali environment; however, larger organic alkali molecules may be beneficial to the stability of UIO-66(H₂DCA)-1.