MOF-Derived Cu@N-C Catalyst for 1,3-Dipolar Cycloaddition Reaction

Cu(im)2-derived Cu@N-C composites were used for the first time as efficient heterogeneous catalysts for one-pot 1,3-dipolar cycloaddition of terminal alkynes, aryl halides, and sodium azide to preparation of 1,4-disubstituted 1,2,3-triazoles with broad substrate scope and high yields. The catalyst can be easily reused without the changes of structure and morphology, and the heterogeneity nature was confirmed from the catalyst recyclability and metal leaching test.


Synthesis of Cu@N-C(x) (X Represents Different Pyrolysis Temperature)
The powder Cu(im) 2 (1.5 g) was placed in a tube furnace and calcined up to 400 • C/600 • C/800 • C with a heating rate of 5 • C·min −1 under argon flow. Maintaining the targeted temperature for 5 h, the resulting solid was cooled to room temperature to afford the copper supported on nitrogen-doped carbon, denoted as Cu@N-C(400), Cu@N-C(600), and Cu@N-C(800), respectively.

General Procedure for the Cycloaddition Reaction
A 25 mL Schlenk tube was charged with Cu@N-C(x) (10 mg), benzyl halide (1, 0.5 mmol), NaN 3 (2, 0.6 mmol), alkyne (3, 0.6 mmol) and 2 mL mixture of t-BuOH/water (v/v = 3:1). The mixture was stirred at 50 • C and monitored by TLC until the benzyl halide was consumed. The reaction mixture was then extracted with ethyl acetate (3 × 10 mL). The combined organic phases were washed with water and brine, dried over anhydrous Na 2 SO 4 , and concentrated in vacuo. The residue was purified by flash column chromatograph on silica gel (ethyl acetate/petroleum ether as the eluent) to provide the target products 4 and 5.

Recycling of Cu@N-C(600) Catalyst
After complete separation of the organic phase of the reaction by centrifugation, the catalyst was washed with ethyl acetate and then dried in an oven at 100 • C for 12 h. The dried solid catalyst was reused in a new cycle and repeated the process 4 times under the standard conditions.

Metal Leaching Test of Cu@N-C(600) Catalyst
After completion of the reaction, the reaction mixture was filtered hot under vacuum. The solid was washed with t-BuOH, and the liquid phase was analyzed by ICP-MS.

Results and Discussion
The powder XRD pattern ( Figure S1) and FE-SEM image (Figure 2a) of the Cu(im) 2 demonstrate that the crystal structure and morphology are all in suitable agreement with the literature [45][46][47], confirming a successful synthesis of Cu(im) 2 . Considering that Cu(im) 2 begins to decompose when the temperature increases to ca. 300 • C, the applied calcination temperature are varied from 400 to 800 • C under argon flow [47].
respectively, which indicate that the bivalent Cu 2+ cations in Cu(im)2 are in situ reduced to zero-valent Cu at high temperature. Interestingly, similar XRD patterns were observed for the three Cu@N-C composites due to the same crystal structures. These results suggested that the thermolysis temperature and N-containing imidazole ligands played important roles in particle size and species of copper.  The surface appearances of the as-synthesized Cu@N-C materials were firstly analyzed by FE-SEM analysis, and the results are shown in Figure 2. It is obviously indicated that thermolysis of Cu(im) 2 destroyed the rod structure of the MOF template and formed a flat-like structure. Moreover, it can be seen that the copper-containing particles tended to aggregate gradually with the increase in pyrolysis temperature. Elemental mapping and EDS spectra show the distribution of C, N, and Cu present in the materials ( Figures S2-S4), which further supported the FE-SEM results. In addition, we can see that N atoms homogeneously distribute in all three catalysts, suggesting the N atoms are doped successfully into the catalysts. Figure 3 shows the powder XRD pattern of the three catalysts. The XRD diffraction peaks at 43.4 • , 50.5 • , and 74.2 • 2θ values correspond to the (111), (200) and (220) lattice planes of metallic copper (JCPDS 85-1326), respectively, which indicate that the bivalent Cu 2+ cations in Cu(im) 2 are in situ reduced to zero-valent Cu at high temperature. Interestingly, similar XRD patterns were observed for the three Cu@N-C composites due to the same crystal structures. These results suggested that the thermolysis temperature and N-containing imidazole ligands played important roles in particle size and species of copper. The content of Cu was detected by ICP-OES, while the proportions of C, H, and N were determined by element analysis. As illustrated in Table 1, we can see that higher pyrolysis temperature could lead to a significant increase in the Cu's content, and about 73.73% of Cu content is observed when the temperature rises to 800 • C. On the contrary, the contents of C, N, and H decreased. In addition, the actual measured value of Cu(im) 2 is slightly lower than the theoretical calculated value, which may be related to the existence of a few impurities. With the Cu@N-C composites prepared, the cycloaddition reaction between benzyl bromide (1a), sodium azide (2), and phenylacetylene (3a) was selected as the model reaction to investigate the catalytic activity. The results are summarized in Table 2. The parent Cu(im) 2 only gave low activity, suggesting that cupric ions coordinated with imidazole were not suitable for this reaction (entry 1). The reaction could not proceed without copper catalyst (entry 2). To our delight, Cu(im) 2 pyrolyzed composites showed high activities in this transformation, and Cu@N-C(600) exhibited the highest efficiency, affording the corresponding 1,2,3-triazole 4a in 94% isolated yield (entries 3-5). Lower or higher thermolysis temperature all resulted in lower reactivity of the resulting materials. Investigation of a variety of solvents (H 2 O, alcohols, or H 2 O/alcohols mixture) showed that the mixed solvents of t-BuOH/H 2 O (3/1, v/v) gave a better yield (98%) than others (entries 6-17). Inferior results were found when reducing the catalyst loading or lowering the reaction temperature (entries [18][19][20]. As the classical catalytic system, the mixture of CuSO 4 and sodium ascorbate (NaAsc), which was used as the catalyst under the same reaction conditions, also resulted in obviously decreased yield (entry 21). Having established the optimized conditions ( Table 2, entry 15), we then evaluated the substrates scope of alkynes and aryl halides, and the results were listed in Table 3. For alkynes, we found that all arylacetylenes bearing electron-donating, electron-withdrawing, or electron-neutral groups at the para-, meta-, or ortho-positions of the aromatic ring could smoothly be converted the desired triazoles in 90-98% yield (4a-4k). Sterically 2-substituted substrate also reacted without any problem to give 4f in 90% yield. Aliphatic alkynes seemed to be less reactive than aryl alkynes, and longer reaction times or higher temperatures were needed (4l and 4m). For benzyl bromides, most substrates worked well to give the corresponding cycloaddition products in 70-95% yields under the standard conditions (4n-4ae). However, the electron effect of substituents was more obvious than the aryl acetylenes. For example, the substrates containing electron-withdrawing groups at the para-condition resulted in lower yields than other ones (4s-4v). Longer reaction time was required for benzyl chloride due to the lower activity for the nucleophilic substitution reaction between sodium azide and benzyl chloride, which is the first step for the one-pot 1,3-dipolar cycloaddition (4a and 4s). Based on the heterogeneous catalyst and mild reaction conditions, as well as excellent functional-group compatibility, the catalytic system was also used to derivatize two selected complex drug-like substrates bearing alkyne or azide moieties ( Figure 4). Ethisterone showed suitable efficiency, and the corresponding cycloaddition product was obtained in 82% yield with only increasing the temperature to 70 • C (5a). Surprisingly, a quantitative yield was gained for zidovudine within 1 h (5b). Furthermore, in order to establish the industrial viability of our method, a gram-scale reaction was carried out under the optimal reaction conditions, and the expected product 4a was formed in 90% yield (1.18 g), indicating the synthetic utility of this method from a practical point of view.
Recoverability and reusability of Cu@N-C(600) were investigated in the cycloaddition between benzyl bromide, NaN 3, and phenylacetylene. According to Figure 5, the recycled catalyst Cu@N-C(600) could be recovered and reused without considerable deterioration in catalytic activities, and the yield of 4a always remains above 90% after four runs. The XRD patterns of the recycled catalyst showed no obvious differences of the fresh Cu@N-C(600), which indicates that the crystallinity and structure of the catalyst can be maintained well during the process of the reaction ( Figure S5). However, different results were observed from the FE-SEM image of the reused catalysts ( Figure 6). The metal particles on the surface of fresh Cu@N-C(600) disappeared after the first run, but no changes in the second and the fourth run. The loss of surface copper species may result from the weak binding affinity and mechanical abrasion-induced exfoliation during the reaction process. The results were also supported by the EDS and element mapping analysis. As illustrated in Figures S6-S8, the copper element mapping of the first reused catalyst showed a weaker signal than the fresh one, but the second and fourth cycles were basically unchanged from the first one.   Leaching experiments for model reactions between benzyl bromide, sodium azide, and phenylacetylene were conducted to check the stability of the catalyst. We can see that 1.7% of the initial copper content was detected in the reaction solution by ICP-MS analysis, which was collected by hot filtration after the first cycle. Lower Cu leaching was observed in the next three cycles (Table 4), which was in suitable agreement with the FE-SEM results. Moreover, the reaction with the solution after removal of catalyst via hot filtration at approximately 20% yield stopped, and the yield of cycloaddition product did not increase further even after 10 h under the same conditions (Figure 7). These aforementioned results suggested that Cu(im) 2 -derived Cu@N-C composites was an excellent stable and reusable heterogeneous catalyst for this type of reaction.

Conclusions
In summary, by employing a MOF-templated method, we have reported a Cu(im)2derived Cu@N-C composite, which was applied for the first time as an efficient, recyclable heterogeneous catalyst for the synthesis of 1,4-disubstituted 1,2,3-triazoles with high yields. The catalyst features easily prepared, broad substrate scope with excellent functional tolerance and regioselectivity, low metal leaching, and ambient reaction conditions. The catalyst can be separated by simple filtration and recovered at least four times without declining activity, and the structure maintained well during the reaction process. Further investigations into other types of Cu@N-C-catalyzed reactions are Finally, we compared the activity of the present MOF-derived catalyst with other reported heterogeneous copper catalysts in the one-pot 1,3-dipolar cycloaddition reaction (Table S1). The results demonstrated that the present catalyst exhibited a higher efficiency with higher yields, mild reaction conditions, and broad substrate scopes than other reported methods.

Conclusions
In summary, by employing a MOF-templated method, we have reported a Cu(im) 2derived Cu@N-C composite, which was applied for the first time as an efficient, recyclable heterogeneous catalyst for the synthesis of 1,4-disubstituted 1,2,3-triazoles with high yields. The catalyst features easily prepared, broad substrate scope with excellent functional tolerance and regioselectivity, low metal leaching, and ambient reaction conditions. The catalyst can be separated by simple filtration and recovered at least four times without declining activity, and the structure maintained well during the reaction process. Further investigations into other types of Cu@N-C-catalyzed reactions are currently ongoing in our laboratory and will be reported in due course.
Author Contributions: J.X. conceived and designed the experiments; Z.W. and X.Z. performed the experiments and analyzed the data; S.G. analyzed the data; Z.W., J.X. and S.G. wrote the paper; J.X. supervised the research. All authors have read and agreed to the published version of the manuscript.