SEM image in Fig. 1(a) demonstrates that ZIF-8 particles have a uniform size. This observation is consistent with previous report which have found ZIF-8 to be an isometric crystal with a diameter of around 100 nm. Moreover, EDS mapping result as shown in Fig. 1 (b) reveals the even Zn element distribution on ZIF-8, suggesting the uniform ZIF-8 distribution. Figure 2 presents the STA of ZIF-8 in air and nitrogen atmosphere to assess thermal stability of ZIF-8 nanoparticles. When STA was done under air flow, a slight weight loss is detected below 300 ℃, which is attributed to the evaporation of guest molecules, including methanol, 2-methylimidazole, H2O. However, the weight loss of ZIF-8 becomes significant after 300 ℃, with a sharp weight loss at 400 ℃, due to the thermal decomposition of ZIF-8. This results indicate that the thermal stability of ZIF-8 in air is up to 300 ℃, with a noticeable weight loss of around 64% on the STA curve as ZIF-8 (relative molecular mass: 299) is converted to ZnO (relative molecular mass: 81)[42]. For the STA done under nitrogen, the thermal stability of ZIF-8 is found to be much higher as expected, reaching reach 550 ℃ compared to the 300 ℃ thermal stability under air. However, above 550 ℃, ZIF-8 decomposes drastically in nitrogen, resulting in significant weight loss. Therefore, to preserve the integrity of the crystal, ZIF-8 is calcinated under 300 ℃ in an air atmosphere, and under nitrogen for comparison purposes.
The XRD patterns of pristine and calcined ZIF-8 are presented in Fig. 3. The peaks observed between 2° and 40° are attributed to the ordered porous structure of ZIF-8 as shown in Fig. 3(a). These results, along with FE-SEM images, indicate the successful synthesis of ZIF-8 with a sodalite structure in a methanol solution at room temperature. The XRD analysis also reveals a significant improvement in crystallinity among the ZIF-8 samples after calcination, as observed by notably sharp peaks. The strongest characteristic peaks of the C200-T6 sample (Fig. 3(a)) and C300-T3 sample (Fig. 3(b)) demonstrate that a longer time of thermal treatment is beneficial in improving crystallinity at a lower calcination temperature. However, it should be noted that the crystallinity of ZIF-8 decreased when samples were calcinated for 9 hours, possibly due to damage to the crystal of ZIF-8.
Table 1 shows the surface area and pore size of ZIF-8 and calcined ZIF-8 samples, which were determined by N2 absorption-desorption measurements. The results show that thermal treatment has a minimal impact on the pore volume and pore diameter of ZIF-8, confirming its thermal stability. In addition, the surface area of ZIF-8 calcinated after 3 hours slightly increased. There was negligible change in the surface area of ZIF-8 calcined below 300℃ in the air atmosphere, and a similar trend was observed for ZIF-8 samples calcined in nitrogen. However, when the thermal treatment time increased to 9 hours at 300℃ in the air atmosphere, the surface area of ZIF-8 decreased from 1264.9 m²/g to 917.3 m²/g. This decrease suggests that the structure of ZIF-8 may have been lightly damaged by the longer calcination time, which is consistent with the XRD patterns.
Table 1
Physical properties of the ZIF-8 and ZIF-8 after thermal treatment
Sample | Surface Areaa (m²/g) | Pore Volumeb (cm3/g) | Pore Diameterb (nm) |
ZIF-8 | 1162.0 | 1.1 | 1.4 |
C200-T3 | 1268.0 | 1.1 | 1.4 |
C200-T6 | 1259.6 | 1.1 | 1.4 |
C200-T9 | 1233.2 | 1.0 | 1.4 |
C300-T3 | 1264.9 | 1.1 | 1.4 |
C300-T6 | 1045.8 | 1.1 | 1.4 |
C300-T9 | 917.3 | 1.2 | 1.4 |
N2-C300-T3 | 1248.7 | 1.2 | 1.4 |
N2-C300-T6 | 1157.1 | 1.0 | 1.4 |
N2-C300-T9 | 1172.0 | 0.9 | 1.4 |
a: Data of surface area is calculated by BET method
b: Data of pore volume and pore diameter is based on desorption
Valence states of C 1s and Zn 2p in both ZIF-8 and calcined ZIF-8 were elucidated by XPS. As shown in Fig. 4(a), peaks from C 1s of ZIF-8 can be observed at 284.3 eV, 284.9 eV and 285.9 eV which are attributed to C = C, C-sp3 and C-N respectively[43]. Interestingly, after calcined in air, ZIF-8 shows an additional peak at 288.3 eV, indicating the presence of N = C = O group. This peak is not observed in fresh ZIF-8 or nitrogen-treated ZIF-8, suggesting that -N = C- of ZIF-8 is oxidized to N = C = O when treated by air[44]. Moreover, higher calcination temperature can enhance the oxidation of ZIF-8 in air, as evidenced by the increased intensity of N = C = O peak in C300-T3 compared to C200-T3. In addition, the intensity of C-N peak and C-sp3 peak decreases after calcination for 9 hours, suggesting the loss of C-N and -CH3 (Fig. 4(b))[43, 45].
For the XPS of Zn, as displayed in Fig. 4 (c), peaks of the Zn 2p 1/2 and Zn 2p 3/2 are observed at 1044.7 eV and 1021.6 eV, respectively. The Zn 2p 3/2 may be a combination of signal from Zn-N at 1021.1 eV and ZnO at 1021.8 eV[46, 47]. As the thermal treatment time increases, the formation of Zn-O from Zn-N is enhanced as evidenced by the decreasing intensity of Zn-N peak shown in Fig. 4 (d). Overall, XPS analysis confirms that isocyanate groups are formed when ZIF-8 is only calcinated in air, and C-N, Zn-N and -CH3 break down with increasing thermal treatment time.
The prepared samples were then tested as a catalyst for transesterification of triglyceride in soybean oil by using methanol. Figure 5 (a) summarizes the conversion of triglyceride at various reaction time with the use of prepared catalyst. The triglyceride conversion reaches 98.1% by using C300-T3 while the conversions are 70.3%, 81.5% and 81.0% when using ZIF-8, C200-T3 and N2-C300-T3, respectively. The C300-T3 presents the highest triglyceride conversion possibly because C300-T3 contains more isocyanate groups than others, as observed in XPS spectra in Fig. 4 (a). Isocyanate is known to convert triglycerides with alcohol to alkyl esters[48]. In addition, the catalyst calcined at 200℃ for 9 hours (C200-T9) provides higher conversion for triglyceride than C200-T3 and C200-T6, as shown in Fig. 5 (b) This suggests that longer calcination time can result in higher activity of ZIF-8 at 200℃. However, as illustrated in Fig. 5 (c), among all catalysts, C300-T3 showed the highest triglyceride conversion. Comparing to C300-T6 and C300-T9, activity of catalysts decreased when the calcination was longer, possibly due to the damage of both ZIF-8 and isocyanate groups. A similar trend is also observed for the catalysts calcined at 200 C, Fig. 5 (b). XPS spectra as shown in Fig. 4 (b) supports this observation. Moreover, Fig. 5 (d) reveals 6 hours calcination is suitable time to treat ZIF-8 in nitrogen. Overall, catalyst calcined at 300 ℃, 3 hours and in air (C300-T3) has best activity for transesterification, and activity of catalyst decreases if calcination time is extended.
The leaching of Zn2+ from ZIF-8 and calcined ZIF-8 was subsequently investigated as a measure of their catalytic activity. Leaching of Zn2+ ions from ZIF-8 is considered as a primary component of its catalytic activity during the reaction, with higher leaching generally indicating better activity[49]. The Zn2+ concentration ratio (Ct/C0, C0: initial concentration of Zn2+ in reactants, Ct: concentration of Zn2+ in product after reaction ) is used to describe changes of the concentration of Zn2+ in solution after the reaction. The higher value of, the more Zn2+ leach into product from catalysts. As shown in Table 2, Ct/C0 of product of ZIF-8 was highest at 10.4, while the ratio of C300-T3 was lowest at 2.5. Nevertheless, C300-T3 converted the most triglyceride (98.1%) to fatty acid methyl ester (FAME) and ZIF-8 only converted 70.3%. The Ct/C0 for N2-C300-T3 and C200-T3 fell between those of ZIF-8 and C300-T3 in terms of both Zn2+ concentration and conversion. These results can be explained by the fact that isocyanate groups from oxidized ZIF-8 are essential components for catalyzing triglyceride, rather than Zn2+. Therefore, C300-T3, which contains the most isocyanate groups, presents the best conversion for triglyceride, but with the lowest Zn2+ leaching ratio, while ZIF-8 converts less triglyceride but with the highest Zn2+ leaching ratio.
Table 2
Conversion of triglyceride and Zn leaching analysis among different catalysts
Catalyst | Conversion | Zinc Leaching (ct /c0) |
Blank Experiment | 22.1% | 1.0 |
ZIF-8 | 70.3% | 10.4 |
N2-C300-T3 | 81.0% | 6.7 |
C200-T3 | 81.5% | 6.3 |
C300-T3 | 98.1% | 2.5 |
The stability of the catalysts was also investigated in this study and the results are summarized in Table 3. The activity of all catalysts decreased after being used repeatedly for 4 times. The activity of C200-T3 and C300-T3 declined moderately, while ZIF-8 and N2-C300-T3 showed a significant decrease in activity. The reason for this can be attributed to the fact that leaching of Zn2+ from ZIF-8 and N2-C300-T3 accelerates destruction of the crystal, while less leaching of Zn2+ in C300-T3 relatively preserves the integrity of the crystal. Furthermore, the isocyanate groups in C200-T3 and C300-T3 can still provide active sites for transesterification even after repeated use. Moreover, XRD patterns of catalysts after reaction, shown in supplementary materials (S1), reveal that C300-T3 still maintains the integrity of the crystal structure after use, while the crystal structures of other samples show varying degrees of corrosion. Therefore, the C300-T3 catalyst presents great potential for catalytic applications.
Table 3
Stability of catalysts for 5 cycles of transesterification
Run Catalyst | 1st | 2nd | 3rd | 4th | 5th |
ZIF-8 | 70.3% | 71.2% | 65.2% | 58.1% | 47.2% |
C200-T3 | 81.5% | 80.3% | 79.4% | 76.8% | 65.5% |
N2-C300-T3 | 81.0% | 79.3% | 72.1% | 65.8% | 58.4% |
C300-T3 | 98.1% | 99.5% | 99.7% | 94.5% | 86.8% |