Solid basic K-CN catalyst is formulated by a facile hydrothermal route using dicyandiamide and KI precursors. The obtained solid in pale yellow color K-CN was thoroughly analyzed by physicochemical characterization techniques.
Through the infrared spectrum, the g-CN samples doped with potassium was found to have peaks related to the typical C3N4 heterocyclic skeleton vibration. This shows that doping will not cause changes in its chemical structure. As shown in Fig. 1. The breathing mode of the tris-triazine system exhibits a sharp peak at 812 cm− 1 corresponding to the breathing mode of the tri-s-triazine units (Fang et al. 2015). The stretching vibration mode of C-N/C = N in the tri-s-triazine heterocyclic rings results in an obvious strong absorption peak at 1160–1648 cm− 1 (Gao et al. 2017). The terminal amino group and the absorbed water result in the formation of a wider energy band between 3300–3600 cm− 1, which is caused by N-H and O-H (Guo et al. 2017). It is worth noting that, unlike the usual g-CN, due to the introduction of KI, there is a very obvious peak at 992 cm− 1, which can be attribute to melem (Zhang et al. 2015). In addition, the energy band formed between N and the doped metal K causes the tensile vibration of the azide group, which makes K-CN have a very obvious absorption peak at 2179 cm− 1 (Guo et al. 2016).
The C3N4 and K-CN were characterized by powder ray diffraction (XRD). The result is shown in the Fig. 2. There is a strong diffraction peak at 27.6°, which is caused by the stacking of the (002) plane. The weak diffraction peak caused by the repeating unit of the heptane ring is also found at 13.2° (Han et al. 2015). The above results show that when potassium is doped, the structure of g-CN does not change significantly.
The structure of the prepared K-CN was further observed by scanning electron microscope. As shown in the Fig. 3, the doping of potassium element did not cause the morphology of g-CN to change, and K-CN still had a typical layered structure. Even after several cycles of reused, the morphology has not changed significantly. This shows that the prepared K-CN has good chemical stability.
In addition, elemental analysis of the prepared K-CN showed that the catalyst contained three elements of K, C, and N (Fig. 4). The test results of mapping further confirmed this conclusion (Fig. 5).
The microstructure of K-CN was further observed by transmission electron microscope (TEM). The result is shown in Fig. 6. It can be seen from the TEM image that the potassium-doped carbon nitride has a layered structure similar to that of pure g-CN.
In order to test the catalytic activity of the prepared K-CN, the Knoevenagel reaction between benzaldehyde and malononitrile was tested as a model reaction. As illustrated in Table 1, various solvents including CH2Cl2, CH3CN, ethyl lactate (EL), MeOH, H2O, EtOH and EtOH/H2O were screened (Table 1, entries 1–7). Among them, a mixed solvent of EtOH/H2O gave the best result for achieving a maximum yield of product 3a (Table 1, entry 7). Subsequently, the reaction temperature was screened, and it was found that the increase of the reaction temperature was only helpful for increasing the reaction rate, but had no effect on the yield (Table 1, entry 8). Furthermore, decreasing the amount of the catalyst to 5 mg reduced the yield to 63% (Table 1, entry 9). Increasing the amount of catalyst has no significant effect on the reaction yield (Table 1, entry 10). In addition, shortening the reaction time was not beneficial to the reaction. It is worth noting that expected product 3a was obtained with a low yield of 25% when pure g-CN was used as the catalyst (Table 1, entry 13). It was found that almost no target product was detected when the reaction mixture was stirring for 1 h at room temperature in the absence of catalyst (Table 1, entry 14). To evaluate the practicability of this method, we conducted best synthesis conditions of 3a on a 20-mmol scale and finally obtained 97% yield pure product (Table 1, entry 15).
With the optimal conditions in hand (Table 1, entry 7), the generality of the developed synthetic protocol for various aryl aldehydes was examined. As shown in Table 2, various substituted benzaldehydes bearing electron-donating or electron withdrawing groups reacted with malononitrile, furnishing the desired products in excellent yields. Furthermore, it was observed that the position of the substituent on the benzene ring did not significantly affect the reaction. Additionally, heterocyclic aldehydes such as thiophene-2-carboxaldehyde and 5-methylfurfural were also worked well without any structural damage, affording the corresponding products 3k and 3l in high yield. In addition, the reactions worked also well when more bulky 1-naphthyl, 2-naphthyl-derived aldehydes and 9-anthracenealdehyd was used as a substrate.
The recyclability of the catalyst is one of the important indicators for investigating heterogeneous catalysts. In this reaction system, the catalyst can be separated by simple filtration, wash it with EtOH three times and then dry it, which can be used in the next experiment. The results show that after 6 cycles of testing, the prepared K-CN still maintains high catalytic activity (Fig. 7). After the sixth cycle, the reused catalyst was analyzed by FTIR. The result showed that the catalyst fully maintained its chemical integrity by comparing its FTIR spectra with fresh catalyst (Fig. 1 ). The above results indicate that K-CN has good chemical stability and catalytic activity, and can be used many times in catalytic conversion.
The comparison of this reaction system and other reported Knoevenagel condensation between benzaldehyde and malononitrile is listed in Table 3. As can be seen in Table 3, the present catalytic system is an equally or more efficient to those previously reported for this reaction in respect of the reaction time and yield.