Hydroquinone-treated Cu₃(BTC)₂: a mixed-valence Cu(I/II) MOF catalyst for efficient cycloadditions

Abstract

We present mixed-valence Cu(I)₁Cu(II)₂(BTC)₂ [henceforth Cu(I/II)-HKUST-1], post-synthetically prepared via the hydroquinone (H₂Q) treatment of Cu(II)₃(BTC)₂ (also referred to as HKUST-1) and its subsequent catalytic activity. This Cu(I/II)-HKUST-1 exhibits exceptional structural integrity and superior catalytic performance in the copper-catalyzed azide–alkyne cycloaddition (CuAAC) reaction between phenylacetylene and benzyl azide. These findings highlight the potential of mixed-valence Cu-based MOFs as effective and sustainable heterogeneous catalysts for organic transformations, paving the way for future advancements in MOF-based catalysis.

---

Metal–organic frameworks (MOFs) are a highly versatile class of crystalline porous materials, composed of metal ions (or clusters) bridged by organic linkers, which exhibit extensive structural diversity. The pore size and chemical environment within MOFs can be precisely tuned by selecting appropriate metal ions and functionalized linkers, depending on the intended application.¹ Notably, a wide variety of MOFs possess open metal sites (OMSs), which are vacant Lewis-acidic sites created by the removal of labile molecules, typically Lewis-basic solvents, from the primary coordination sphere of the metal center.² Due to these distinctive characteristics, MOFs have garnered significant interest for applications, for example, in gas adsorption and separation,^3,4 energy storage,^5,6 and heterogeneous catalysis.^7–9

Among the various types of MOFs, those with metal centers in multiple oxidation states (mixed-valence MOFs) have exhibited exceptional properties, such as charge transfer and charge delocalization, which contribute to high conductivity a desirable feature in redox chemistry and semiconductor development,^10,11 and catalytic reactions.^12,13 In these mixed-valence MOFs, the incorporation or removal of counter-ions, essential for maintaining the charge balance, not only alters the chemical environment of the pores but also induces structural changes, affecting pore size, window aperture, and available pore volume.¹⁴ However, the synthesis of intrinsically mixed-valence MOFs has been challenging because it requires metals and ligands to cooperate together precisely to enable redox isomerism.¹⁵ Specifically, direct synthetic methods often require modulators to control the formation of metal clusters in mixed oxidation states, but achieving precise control over the metal nodes in these mixed states remains difficult.¹⁶ Therefore, post-synthetic strategies have been developed to efficiently change the oxidation state of the metal centers while preserving the original MOF structure.^17–20

Cu-based mixed-valence MOFs, in particular, have attracted considerable attention due to the enhancements in properties such as conductivity,²¹ hydrolytic stability,^20,22 adsorption,^3,4 and catalytic activity^7,8 that arise from changes in their electronic structure. Among these, Cu₃(BTC)₂ (BTC = benzene-1,3,5-tricarboxylate), also known as HKUST-1, serves as a benchmark material for the reduction of Cu(II) ions. For example, heating HKUST-1 under vacuum conditions partially reduces Cu(II) to Cu(I).³ Duke et al. demonstrated that a nearly complete reduction of Cu(II) in HKUST-1 to Cu(I) could be achieved at 225 °C under an argon atmosphere.²¹ Additionally, Szanyi and coworkers reported the effective reduction of Cu(II) centers in HKUST-1 using CO or H₂ atmospheres, though the extent of reduction was influenced by the temperature rather than the nature of the reductive atmosphere, with the highest reduction occurring at 200 °C.²³ However, a comprehensive structural characterization, including whether the reduction of Cu(II) occurs completely or partially to produce a mixed-valence MOF, is still lacking. Consequently, the precise mechanism underlying Cu(II) reduction and defect formation remains partially resolved. Recently, by contrast, our group introduced a novel approach, the coordinative reduction of Cu(II) in HKUST-1 using post-synthetic hydroquinone (H₂Q) treatment, yielding a Cu(I)₁/Cu(II)₂ mixed-valence MOF [hereafter referred to as Cu(I/II)-HKUST-1] while maintaining structural integrity. In Cu(I/II)-HKUST-1, approximately 33% of Cu(II) ions were reduced to Cu(I), with half of these reduced ions remaining within the paddlewheel structure, thereby serving as potential catalytic active sites.²⁰ By contrast, the remaining 67% of Cu(II) ions contribute to preserving the structural stability of Cu(I/II)-HKUST-1.

Meanwhile, it has been considered challenging to exploit a Cu(I)-MOF as catalyst in organic reactions, in particular, for the coupling of terminal alkynes and organic azides to form 1,4-disubstituted 1,2,3-triazoles (Fig. 1a),²⁴ known as the copper-catalyzed azide–alkyne cycloaddition (CuAAC) reaction (also called as a type of click reaction). Although the reaction has been extensively studied using various molecular Cu(I) complexes generated by the reduction of Cu(II), oxidation of Cu(0), or through ligand-coupled reduction of Cu(II),^25,26 the use of heterogeneous MOF catalysts without structural damage and significant loss of activity has rarely been successful, despite their preference for easier recovery, recyclability, and molecular diffusion all crucial for more efficient and sustainable processes. Therefore, mixed-valence Cu-based MOFs have recently been recognized as promising candidates for such catalytic applications.^7,8,27,28

Fig. 1 (a) Schematic illustration of the Cu(I)-catalyzed azide–alkyne cycloaddition reaction, (b) schematic illustration for the H₂Q-induced coordinative reduction of Cu(II) in the paddlewheel node of HKUST-1. Hydrogen atoms bound to carbon atoms in the benzene moieties are omitted for clarity, and (c) the three types of nodes that possibly exist in H₂Q-treated Cu(I/II) HKUST-1: Node I, Node II, and Node III, each constituting 33.3% of the structure.

Herein, we demonstrate that post-synthetically H₂Q-treated mixed-valence Cu(I/II)-HKUST-1 can be utilized as an efficient catalyst. This Cu(I/II)-HKUST-1 exhibits excellent catalytic activity in the CuAAC reaction between phenylacetylene and benzyl azide. Furthermore, Cu(I/II)-HKUST-1 demonstrates remarkable structural stability and recyclability, paving the way for further advancements in the design and synthesis of mixed-valence Cu-based MOFs.

Cu(I/II)-HKUST-1, containing 33% of Cu(I), was prepared according to our previously reported method.²⁰ In brief, thermally activated HKUST-1 was treated with H₂Q in anhydrous acetonitrile (MeCN) at 80 °C for 12 hours. During the H₂Q treatment, 33% of Cu centers were reduced via a coordinative reduction mechanism, where H₂Q, upon coordination with Cu(II), donates an electron to the Cu(II) center through the coordination bond, forming HQ• and Cu(I) (Fig. 1b). The subsequent change in the coordination number and geometry of Cu(I), which prefers two- or four-coordination, promotes the dissociation of HQ˙, enabling further reduction of remaining Cu(II) ions and the formation of p-benzoquinone (Q) (Fig. S1, ESI†). Consequently, three distinct types of nodes are formed in Cu(I/II)-HKUST-1: node I, formed by Cu(II)–Cu(I) paddlewheels; node II, consisting of Cu(II) Vacancy units; and node III, composed of Cu(II)–Cu(II) paddlewheels (Fig. 1c).

To validate whether the reduction arose, we examined the oxidation state of H₂Q by analyzing the supernatant H₂Q/MeCN solution (1 : 2 ratio of H₂Q : Cu) after the H₂Q treatment. The ¹H NMR spectrum of the supernatant solution shows proton peaks corresponding to H₂Q (84%) and quinone (16%), indicating that approximately 32% of Cu(II) centers were reduced to Cu(I) (Fig. S2 and S3, ESI†). XPS analysis of Cu(I/II)-HKUST-1 further supports this finding, showing a band with a shoulder at approximately 933–935 and 952–954 eV, indicative of a mixture of Cu(II) and Cu(I), whereas MeCN coordinated HKUST-1 (MeCN-HKUST-1), which is not H₂Q-treated, displays bands solely corresponding to Cu(II) (Fig. S4, ESI†). Integration of the fitted bands in the XPS spectrum of Cu(I/II)-HKUST-1 revealed that Cu(II) and Cu(I) are present at approximately 66% and 34%, respectively (Table S1, ESI†), corroborating the ¹H NMR results.

Half of the reduced Cu(I) centers likely dissociated from the node due to the protonation of BTC's carboxylates by a proton released from H₂Q during the reaction, subsequently yielding a [Cu(MeCN)₄]⁺ complex. The formation of the [Cu(MeCN)₄]⁺ complex was confirmed by FT-IR spectroscopy, which showed bands at 2275 and 2303 cm⁻¹ in the Cu(I/II)-HKUST-1 spectrum, corresponding to C≡N vibration modes that agreed well with those of the commercial [Cu(MeCN)₄]BF₄ complex (Fig. S5, ESI†).²⁹ The dissociation of Cu(I) ions from the node subsequently to form [Cu(MeCN)₄]⁺ ions did not alter the HKUST-1 crystal structure (Fig. S6, ESI†), however, it did created defects in the framework, as confirmed by the presence of mesopores in the N₂ adsorption/desorption isotherm of Cu(I/II)-HKUST-1 performed at 77 K (Fig. S7, ESI†). Nevertheless, the surface area of Cu(I/II)-HKUST-1 (1196 m² g⁻¹), although reduced compared to MeCN-HKUST-1 (1965 m² g⁻¹) due to the presence of the [Cu(MeCN)₄]⁺ complex and defect formation, was still comparable to typical HKUST-1 samples reported in other literature.²⁰ The formation of mesopores with an average diameter of 38 Å, as indicated by the DFT pore size distribution analysis, also confirmed the presence of defects induced by the H₂Q treatment (Fig. S8, ESI†). In addition, we observed that Cu(I/II)-HKUST-1 exhibits high thermal stability comparable to the pristine HKUST-1 (Fig. S9, ESI†).

Given the outstanding structural stability and large surface area of Cu(I/II)-HKUST-1, we explored its efficacy as a Cu(I)-based heterogeneous MOF catalyst for the CuAAC reaction between phenylacetylene and benzyl azide in ethanol (EtOH) media over 12 hours (Table 1). The conversion efficiency was monitored by ¹H NMR spectroscopy, focusing on the characteristic signals of methylene groups in benzyl azide and 1-benzyl-4-phenyl-1,2,3-triazole (Fig. S10, ESI†). Initially, we optimized the reaction by controlling temperatures from 25 to 80 °C, using 0.007 mmol of Cu(I/II)-HKUST-1. While the conversion efficiency was modest at 25 °C (15%, entry 1 in Table 1), the efficiency was improved upon the temperature increase (66% at 50 °C, entry 2, Table 1). Eventually, the 100% conversion was achieved at 80 °C (entry 3, Table 1). We then optimized the reaction by verifying the catalyst loading amount. Subsequently, we found that 0.007 mmol yielded the highest conversion, while lower amounts resulted in significantly reduced conversions (66% and 64%, entries 4 and 5, Table 1). Additionally, we evaluated the reaction in different solvents, with MeCN resulting in the lowest conversion (4%, entry 6, Table 1), likely due to its strong coordination to the Cu(II) sites and its hindrance of the neighboring Cu(I) sites, thus, rendering them inactive.³⁰ In contrast, toluene, a non-polar solvent, yielded a high conversion (95%, entry 7, Table 1), suggesting that non-polar and less-polar (EtOH) solvents are preferable for this reaction.

Table 1 Reaction conditions of Cu(I/II)-HKUST-1-catalyzed azide–alkyne cycloaddition reaction, showing catalyst loading, temperature, solvent, and corresponding conversion efficiencies

Entry	Catalyst	Loading (mmol)	Temp. (°C)	Solvent	Conversion (%)
1	H2Q-HK	0.007	25	EtOH	15
2	H2Q-HK	0.007	50	EtOH	66
3	H2Q-HK	0.007	80	EtOH	100
4	H2Q-HK	0.003	80	EtOH	64
5	H2Q-HK	0.005	80	EtOH	66
6	H2Q-HK	0.007	80	MeCN	4
7	H2Q-HK	0.007	80	Toluene	95
8	Act-HK_150 °C	0.007	80	EtOH	31
9	Act-HK_200 °C	0.007	80	EtOH	65
10	None	0	80	EtOH	0

---

To assess the absence of Cu(I/II)-HKUST-1 and the impact of Cu's oxidation state on catalytic activity, we examined the reactions without any catalyst and with a thermally activated HKUST-1, which predominantly contains Cu(II). As expected, the reaction without any catalyst yielded no conversion (entry 10, Table 1), and the HKUST-1 samples activated at 150 °C and 200 °C (Act-HKUST-1_150 °C and Act-HKUST-1_200 °C, respectively) exhibited limited activities (31% and 65%, respectively, entries 8 and 9 in Table 1). The conversion profiles of Cu(I/II)-HKUST-1 and Act-HKUST-1_150 °C along the reaction time were further tested (Fig. 2a and Fig. S11, S12, ESI†). As a result, Cu(I/II)-HKUST-1 exhibited a steady increase in benzyl azide conversion, reaching 100% after 12 hours with no signs of catalyst deactivation, while Act-HKUST-1_150 °C showed only minimal catalytic activity, with the benzyl azide conversion reaching just 31% after 12 hours, likely due to the small presence of Cu(I) sites inevitably generated during thermal activation.³ To determine Cu ions leaching during the catalytic reaction, a hot filtration experiment and inductively coupled plasma (ICP) analysis were performed. The Cu(I/II)-HKUST-1 catalyst was removed after 4 hours of reaction, and the conversion remained unchanged over the subsequent 8 hours (Fig. 2a and Fig. S13, ESI†). ICP analysis of the supernatant revealed that 0.007% of the total Cu ions were detected, indicating negligible leaching occur (Table S2, ESI†). These results show that the reaction is heterogeneous, and the catalyst exhibits high stability during the reaction. Moreover, PXRD analysis confirmed that the structural integrity of Cu(I/II)-HKUST-1 remained intact after the reaction (Fig. S14, ESI†), underscoring its potential for efficient recyclability. Finally, we observed that the Cu(I/II)-HKUST-1 catalyst was easily recollected and recovered by washing with fresh EtOH and reactivated, with conversion above 96% over five cycles, confirming the excellent catalytic activity and recyclability of Cu(I/II)-HKUST-1 (Fig. 2b, Fig S15, and Table S3, ESI†). The framework of the Cu(I/II)-HKUST-1 remained intact during five cycles (Fig. S16, ESI†).

Fig. 2 (a) Time-dependent conversion profiles of the azide–alkyne cycloaddition reaction catalyzed by Cu(I/II)-HKUST-1 (blue) and Act-HKUST-1_150 °C (black). A profile where the Cu(I/II)-HKUST-1 catalyst was removed 4 hours after the reaction started (pink). (b) Catalytic activity of Cu(I/II)-HKUST-1 over five cycles in the azide–alkyne cycloaddition reaction, showing high and consistent conversion rates across all cycles.

To demonstrate the general applicability of the Cu(I/II)-HKUST-1 catalyst, we examined the reaction using various substrates with different functional groups. As a result, the catalyst showed high conversion for all substrates with efficiencies of 95% except for 2-fluorobenzyl azide (1i), which can be attributed to steric hindrance (Fig. 3 and Fig. S17, ESI†). Thus, these results demonstrate that Cu(I/II)-HKUST-1 is an effective heterogeneous catalyst for a wide range of substrates in the CuAAC reaction.

Fig. 3 Various catalytic substrates and their conversion efficiencies for Cu(I/II)-HKUST-1 catalyzed azide–alkyne cycloaddition reaction.

In this study, we successfully demonstrate the post-synthetic modification of HKUST-1 to produce mixed-valence Cu(I/II)-HKUST-1 via hydroquinone treatment, highlighting its strong potential for catalytic azide–alkyne cycloaddition applications. This mixed-valence Cu(I/II)-HKUST-1 not only maintained its structural integrity but also exhibited remarkable catalytic activity in the CuAAC reaction between phenylacetylene and benzyl azide. The Cu(I/II)-HKUST-1 catalyst showed superior performance in terms of conversion efficiency, structural stability, and recyclability, significantly outperforming the thermally activated HKUST-1, which primarily contains Cu(II) sites. These findings highlight the potential of mixed-valence Cu-based MOFs as effective and sustainable heterogeneous catalysts for the transformations of organic compounds. The ability of Cu(I/II)-HKUST-1 to retain high catalytic activity across multiple cycles further underscores its promise for future applications in catalysis and beyond.

This work was supported by the Ministry of Science and ICT (MSIT) of Korea under the auspices of the Basic Science Research Program sponsored by the National Research Foundation (NRF-2023R1A2C2004838) and by the DGIST R&D Program (24-HRHR+-01).

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. Y. Belmabkhout and K. E. CordovaReticular Chemistry and Applications, De Gruyter, 2023.
2. U. Kokcam-Demir, A. Goldman, L. Esrafili, M. Gharib, A. Morsali, O. Weingart and C. Janiak, Chem. Soc. Rev., 2020, 49, 2751–2798.
3. P. S. Petkov, G. N. Vayssilov, J. Liu, O. Shekhah, Y. Wang, C. Woll and T. Heine, Chem. Phys. Chem., 2012, 13, 2025–2029.
4. H. Sun, X. Han, K. Liu, B. Shen, J. Liu, D. Wu and X. Shi, Ind. Eng. Chem. Res., 2017, 56, 9541–9550.
5. X. Tang, C. Liu, H. Wang, L.-P. Lv, W. Sun and Y. Wang, Coord. Chem. Rev., 2023, 494, 215361.
6. Z. Zhu, Y. Zeng, Z. Pei, D. Luan, X. Wang and X. W. D. Lou, Angew. Chem., Int. Ed., 2023, 62, e202305828.
7. Q. Fu, K. Xie, S. Tan, J. M. Ren, Q. Zhao, P. A. Webley and G. G. Qiao, Chem. Commun., 2016, 52, 12226–12229.
8. L. Xue, Y. Yang, S. Wu, Y. Huang, J. Li, Y. Xiang and G. Li, Anal. Chem., 2020, 92, 2972–2978.
9. R. Yepez, S. García, P. Schachat, M. Sánchez-Sánchez, J. H. González-Estefan, E. González-Zamora, I. A. Ibarra and J. Aguilar-Pliego, New J. Chem., 2015, 39, 5112–5115.
10. J. G. Park, M. L. Aubrey, J. Oktawiec, K. Chakarawet, L. E. Darago, F. Grandjean, G. J. Long and J. R. Long, J. Am. Chem. Soc., 2018, 140, 8526–8534.
11. S. Y. Fu, C. H. Chang, A. S. Ivanov, I. Popovs, J. L. Chen, Y. F. Liao, H. K. Liu, S. Chirra, Y. W. Chiang, J. C. Lee, W. L. Liu, W. Kaveevivitchai and T. H. Chen, Angew. Chem., Int. Ed., 2023, 62, e202312494.
12. H. Chen, Y. Liu, T. Cai, W. Dong, L. Tang, X. Xia, L. Wang and T. Li, ACS Appl. Mater. Interfaces, 2019, 11, 28791–28800.
13. T. Yang, D. Yu, D. Wang, T. Yang, Z. Li, M. Wu, M. Petru and J. Crittenden, Appl. Catal., B, 2021, 286, 119859.
14. D. M. D'Alessandro, Chem. Commun., 2016, 52, 8957–8971.
15. Y.-X. Li, J.-X. Shen, Z.-J. Diao, S.-C. Qi, X.-Q. Liu and L.-B. Sun, Fund. Res., 2022 DOI:10.1016/j.fmre.2022.08.012.
16. M. Ronda-Lloret, I. Pellicer-Carreño, A. Grau-Atienza, R. Boada, S. Diaz-Moreno, J. Narciso-Romero, J. C. Serrano-Ruiz, A. Sepúlveda-Escribano and E. V. Ramos-Fernandez, Adv. Funct. Mater., 2021, 31, 2102582.
17. J. W. Yoon, Y.-K. Seo, Y. K. Hwang, J.-S. Chang, H. Leclerc, S. Wuttke, P. Bazin, A. Vimont, M. Daturi, E. Bloch, P. L. Llewellyn, C. Serre, P. Horcajada, J.-M. Grenèche, A. E. Rodrigues and G. Férey, Angew. Chem., Int. Ed., 2010, 49, 5949–5952.
18. T. F. Liu, L. F. Zou, D. W. Feng, Y. P. Chen, S. Fordham, X. Wang, Y. Y. Liu and H. C. Zhou, J. Am. Chem. Soc., 2014, 136, 7813–7816.
19. B. Tu, Q. Pang, H. Xu, X. Li, Y. Wang, Z. Ma, L. Weng and Q. Li, J. Am. Chem. Soc., 2017, 139, 7998–8007.
20. D. Song, J. Bae, H. Ji, M.-B. Kim, Y.-S. Bae, K. S. Park, D. Moon and N. C. Jeong, J. Am. Chem. Soc., 2019, 141, 7853–7864.
21. A. S. Duke, E. A. Dolgopolova, R. P. Galhenage, S. C. Ammal, A. Heyden, M. D. Smith, D. A. Chen and N. B. Shustova, J. Phys. Chem. C, 2015, 119, 27457–27466.
22. M. Kanno, T. Kitao, T. Ito and K. Terashima, RSC Adv., 2021, 11, 22756–22760.
23. J. Szanyi, M. Daturi, G. Clet, D. R. Baer and C. H. F. Peden, Phys. Chem. Chem. Phys., 2012, 14, 4383–4390.
24. M. Meldal and C. W. Tornøe, Chem. Rev., 2008, 108, 2952–3015.
25. V. D. Bock, H. Hiemstra and J. H. van Maarseveen, Eur. J. Org. Chem., 2005, 51–68.
26. M. S. Ziegler, K. V. Lakshmi and T. D. Tilley, J. Am. Chem. Soc., 2017, 139, 5378–5386.
27. H. Noei, S. Amirjalayer, M. Müller, X. Zhang, R. Schmid, M. Muhler, R. A. Fischer and Y. Wang, ChemCatChem, 2012, 4, 755–759.
28. W. Xue, J. Wang, H. Huang, Y. Cui and D. Mei, J. Phys. Chem. C, 2022, 126, 9652–9664.
29. I. S. Kritchenkov, J. R. Shakirova and S. P. Tunik, RSC Adv., 2019, 9, 15531–15535.
30. S. H. Park, R. A. Peralta, D. Moon and N. C. Jeong, J. Mater. Chem. A, 2022, 10, 23499–23508.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cc04294h

‡ These authors contributed equally to this work.

---