Construction of a supported Ru complex on bifunctional MOF-253 for photocatalytic CO₂ reduction under visible light

Abstract

A MOF-253 supported active Ru carbonyl complex (MOF-253–Ru(CO)₂Cl₂) was constructed for photocatalytic CO₂ reduction under visible light irradiation. Its performance can be further improved by immobilization as a photosensitizer. This study highlights the great potential of using MOFs as a solid ligand and platform for the assembly of a complicated catalytic system.

---

Homogeneous molecular catalysts show high activity in reactions, but suffer from product contamination and limited recyclability. For practical applications, molecular catalysts are usually immobilized on solid supports for easy recovery and recycling. Metal–organic frameworks (MOFs), which have already shown a variety of promising applications, are emerging as a new type of ideal support due to their high surface areas and uniform pores for facile diffusion of reactants.¹ By post-synthetic modifications (PSM), MOFs can be endowed with well defined and isolated sites for the anchoring of a catalytic species, which makes them especially appealing for constructing supported single site catalysts as compared to other supports which have non-homogeneously scattered sites.² Actually, catalytic active species supported MOF materials prepared via PSM methods have been designed to meet a variety of catalytic purposes.^3–5

However, sometimes the anchoring of an active catalytic species on MOFs via PSM functionalizations on the ligand may induce unfavorable interactions between the metal complex and the solid surface, which would lead to a decreased performance of the molecular catalysts. The direct construction of metal complexes using the substrate as a solid ligand would be an ideal strategy to develop the supported molecular catalysts without losing their performance. With opening 2,2′-bipyridine (bpy) moieties in its structure, MOF-253 can be an ideal solid ligand for the construction of a surface supported metal complex featuring bpy ligands.⁶ Given the ubiquitous role of the bpy ligand in coordination chemistry, a series of studies have already been conducted on the functionalization of MOF-253 using its open N,N′-chelating sites.^7–9

Ru complexes containing the bpy ligand have been widely used as homogeneous photocatalysts and photosensitizers.¹⁰ Especially Ru carbonyl complexes containing 2,2′-bpy are photocatalytically or electrochemically active for CO₂ reduction.¹¹ However, the use of a supported Ru complex for photocatalytic CO₂ reduction has not been previously studied. In this communication, we reported for the first time the use of N,N′-chelating centers in MOF-253 to construct the supported active Ru carbonyl complex (MOF-253–Ru(CO)₂Cl₂) for photocatalytic CO₂ reduction under visible light irradiation and its mechanism. In addition, MOF-253 can also be used as a platform for building of a sensitized system with significantly enhanced photocatalytic performance by further incorporation of Ru(bpy)₂Cl₂ as a photo-sensitizer.

Al(OH)(dcbpy) (MOF-253) (dcbpy for 2,2′-bipyridine-5,5′-dicarboxylic acid) was chosen to construct the supported Ru complex since it has open accessible 2,2′-bipyridine units in its framework, which allows for its coordination to metal centers to develop photocatalytically active porous materials. MOF-253 was prepared following the previously reported procedures.⁶ The good agreement between the XRD patterns of the as-prepared product and the calculated MOF-253 suggests the formation of the pure phase of MOF-253 (Fig. 1a). The Langmuir surface area of the as-obtained product is determined to be 1430 m² g⁻¹, larger than that reported previously (1202 m² g⁻¹), indicating that MOF-253 of high quality has been obtained (Fig. S1, ESI†).⁸

Fig. 1 (a) XRD patterns of the as-prepared samples together with that of the calculated MOF-253; (b) FT-IR spectra of prepared samples; (c) Fourier transform magnitude of the EXAFS spectra (FT-EXAFS) for MOF-253–Ru(CO)₂Cl₂; (d) UV/Vis spectra of prepared samples. The inset is the UV/Vis absorption spectrum of homogeneous Ru(5,5′-dcbpy)(CO)₂Cl₂.

To prepare a MOF-253 supported Ru complex (MOF-253–Ru(CO)₂Cl₂), the as-synthesized MOF-253 was de-solvated first under dynamic vacuum and then reflux in anhydrous methanol solution containing [Ru(CO)₃Cl₂]₂. The XRD pattern of the as-obtained product shows characteristic diffraction peaks of the MOF-253 framework, indicating that the introduction of the Ru moiety does not influence the structure of MOF-253 (Fig. 1a). The slight decrease of the diffraction intensity of the resultant product as compared with the parent MOF-253 is probably due to the existence of disorder within the crystal structure after the immobilization process. The presence of Ru(CO)₂Cl₂ in the as-prepared product was confirmed by its FT-IR spectrum (Fig. 1b). As compared with the parent MOF-253, two additional peaks at 2073 cm⁻¹ and 2010 cm⁻¹ was observed, which can be assigned to the asymmetric vibration of CO in Ru(CO)₂Cl₂, indicative of the formation of MOF-253–Ru(CO)₂Cl₂. The coordination of Ru(II) to the free N,N-chelating sites in MOF-253 leads to a slight red shift of the asymmetric vibration of CO as compared with those in the original [Ru(CO)₃Cl₂]₂ (2093 cm⁻¹ and 2071 cm⁻¹). The formation of Ru–N bonds between Ru(CO)₂Cl₂ and N,N-chelating sites in MOF-253 is also evidenced by extended X-ray absorption fine structure (EXAFS) analyses. The experimental Fourier transform spectrum of EXAFS (FT-EXAFS) shows the Ru–N distances to be 2.16 Å and 2.20 Å, while the Ru–Cl distances to be 2.36 Å and 2.40 Å, respectively (Fig. 1c). These distances are in good agreement with the Ru(II)–N and Ru(II)–Cl bonds observed in similar Ru(II) bpy complexes.¹² The amount of Ru incorporated into MOF-253–Ru(CO)₂Cl₂ was determined by inductively coupled optical emission spectrometer (ICP-OES) analysis. The Ru/Al ratio (6.3%) in the as-obtained MOF-253–Ru(CO)₂Cl₂ determined by ICP is a little lower than that added into the reaction system (with Ru/Al at 10.0%). N₂ adsorption–desorption isotherm analysis shows that the as-obtained MOF-253–Ru(CO)₂Cl₂ still shows a relatively high Langmuir specific surface area of 1085 m² g⁻¹, indicating the existence of the permanent porosity (Fig. S1, ESI†). The slight decrease of the surface area is attributed to the partial blocking of the open pores in MOF-253 by the Ru carbonyl complex, rather than the collapse of the framework.

The UV/Vis DRS spectra of the as-obtained MOF-253–Ru(CO)₂Cl₂, compared with that of pure MOF-253 and Ru(5,5′-dcbpy)(CO)₂Cl₂, are shown in Fig. 1d. Ru(5,5′-dcbpy)(CO)₂Cl₂ shows an absorption edge extending to about 470 nm, in accordance with its bright yellow-greenish color (Fig. 1d inset). The absorption in the visible light region observed over Ru(5,5′-dcbpy)(CO)₂Cl₂ can be ascribed to the metal-to-ligand (Ru^II → bipyridine π*) charge transfer (MLCT) transition. Pure MOF-253 alone does not show absorption in the visible light region. However, when coordinated to Ru(II) via N,N-chelating sites, the absorption edge of the as-prepared MOF-253–Ru(CO)₂Cl₂ extends to 470 nm, in accordance with its yellow color.

Since bpy containing Ru carbonyl complexes have been previously reported to be photocatalysts for CO₂ reduction,¹³ we investigated the photocatalytic CO₂ reduction over the as-obtained MOF-253–Ru(CO)₂Cl₂ in a mixture of MeCN/TEOA (10/1) under visible light irradiation. As shown in Fig. 2a, HCOO⁻, CO and H₂ were produced over the as-prepared MOF-253–Ru(CO)₂Cl₂ and the amount of the products increased with the irradiation time. About 0.67 μmol of HCOO⁻, 1.86 μmol of CO as well as 0.09 μmol H₂ were produced after irradiation for 8 h. The calculated TON for the formation of HCOO⁻, CO and H₂ is 2.9, 7.1 and 0.4, respectively (Table 1). No products were detected over pure MOF-253 or MOF-253–Ru(CO)₂Cl₂ without light irradiation, indicating that the formation of the products is truly induced by the photocatalysis over MOF-253–Ru(CO)₂Cl₂. The photocatalytic CO₂ reduction over the homogeneous Ru(5,5′-dcbpy)(CO)₂Cl₂ revealed that only 0.06 μmol of HCOO⁻, 1.27 μmol of CO and 0.12 μmol of H₂ were produced under similar conditions (Fig. 2b). The amount of both CO and HCOO⁻ produced over the homogeneous Ru(5,5′-dcbpy)(CO)₂Cl₂ is lower than that over a MOF supported Ru complex. The even better performance observed over MOF-253–Ru(CO)₂Cl₂ as compared with its homogeneous counterpart is possibly due to the formation of some intermediates inactive for the photocatalytic CO₂ reduction over homogeneous Ru(5,5′-dcbpy)(CO)₂Cl₂.¹⁴ This suggests that the construction of the MOF-253 surface Ru complex via coordination with its N,N′-chelating sites is an efficient strategy to develop the supported molecular catalyst. Although the ICP analysis revealed that about 6.2% of incorporated Ru leached into the reaction filtrate after an 8 h reaction, the filtrate experiment has shown that only about 0.1 μmol of CO and 0.02 μmol of HCOO⁻ were produced over the filtrate irradiated for 4 h (Table S1, ESI†), a confirmation of the heterogeneous nature of the MOF-253–Ru(CO)₂Cl₂ in a photocatalytic CO₂ reduction. Besides this, the XRD of the MOF-253–Ru(CO)₂Cl₂ did not change after the reaction, indicating that the photocatalyst is stable during photocatalytic CO₂ reduction (Fig. S2, ESI†).

Fig. 2 (a) The amount of products produced as a function of irradiation time over MOF-253–Ru(CO)₂Cl₂; (b) the amount of products produced over Ru(5,5′-dcbpy)(CO)₂Cl₂ and MOF-253–Ru(CO)₂Cl₂ after 8 h irradiations. Photocatalysts: 5 mg, MeCN/TEOA (10/1, 6 ml).

Table 1 TON for photocatalytic CO₂ reduction over different samples after irradiated for 8 h^a

Photocatalyst	Reaction time/h	TON
		HCOO^−	CO	H2	Total
a TON for H2 and HCOO^− is defined as mole of the evolved H2 and HCOO^− over per amount of ruthenium, while TON for CO is defined as the number of evolved CO after deduction of those from carbonyl.
Ru(dcbpy)(CO)2Cl2	8	0.3	4.5	0.5	5.3
MOF-253–Ru(CO)2Cl2	8	2.9	7.1	0.4	10.4
Sensitized MOF-253–Ru(CO)2Cl2	8	35.8	7.3	11.9	55.0

---

To study the origin of the reaction products, isotopic ¹³CO₂ was used for the photocatalytic CO₂ reaction (Fig. S3 and S4, ESI†). We observed peaks at 164.4 and 158.7 ppm in the ¹³C NMR spectrum, which can be assigned to HCOO⁻ and HCO₃²⁻ respectively. This suggests that the as-formed HCOO⁻ really comes from CO₂.^1e Additional peaks observed at 161.6, 167.5 and 168.6 ppm in the ¹³C NMR spectrum probably originated from Ru–¹³CO₂ adducts and intermediates (as shown in Scheme 1) leaching into the solution during the reaction.¹⁵ In the meantime, the GC-MS spectra of the gaseous product from the reaction with ¹³CO₂ showed signals at m/z values of 29 and 28, corresponding to ¹³CO and ¹²CO respectively. In contrast, only the signal at the m/z value of 28 was detected in the product from the reaction with ¹²CO₂. This clearly indicates that CO₂ was reduced to CO over MOF-253–Ru(CO)₂Cl₂. The coexistence of ¹²CO in the gaseous product in the ¹³CO₂ reaction may come from those dissociated from the MOF-253–Ru(CO)₂Cl₂ as elucidated in the mechanism.

Scheme 1 Proposed mechanism for photocatalytic CO₂ reduction over the as-prepared MOF-253–Ru(CO)₂Cl₂ under visible light.

Based on the experimental results and the previous studies on the photocatalytic CO₂ reduction over Ru carbonyl complexes, a possible mechanism for the photocatalytic CO₂ reduction over MOF-253–Ru(CO)₂Cl₂ was proposed (Scheme 1). MOF-253–Ru(CO)₂Cl₂ shows light absorption in the visible light region due to the existence of the MLCT. Upon irradiation, the excited MLCT state can be reductively quenched by TEOA, giving the one-electron reduced species (1). Previous reports have shown that irradiation of the Ru-based complex always induces photochemical ligand substitution to give free CO.¹⁶ The observation of ¹²CO in the MS spectrum upon reaction with ¹³CO₂ and the change of the IR spectrum over MOF-253–Ru(CO)₂Cl₂ after the reaction indicated that similar CO dissociation had occurred (Fig. S5, ESI†). The release of CO from the one-electron reduced species can lead to the formation of the pentacoordinated intermediate (2), which would open to the electrophilic attack by CO₂ and result in the formation of a CO₂ adduct (3–4).¹⁵ TEOA can protonate the bound CO₂ in the CO₂ adduct and induce the release of water to complete the catalytic cycle. Such a mechanism leads to the preferential formation of CO as the ultimate product (red cycle in Scheme 1). An alternative mechanism to the direct CO₂ adduction is the formation of the hydride intermediate (5), which followed by CO₂ insertion into the metal–hydride bond (6) can lead to the formation of HCOO⁻ (blue cycle in Scheme 1). The formation of the Ru hydride intermediate was confirmed in our previous study on the photocatalytic hydrogenation over a hybrid of the CdS–Ru carbonyl complex.¹⁷ Actually, a similar dependence of the nature of the ligand coordinated to the Ru(II) sites on the ultimately formed products for photocatalytic CO₂ reduction over other Ru complexes was previously reported.¹³

Although MOF-253–Ru(CO)₂Cl₂ is photocatalytically active for CO₂ reduction, its performance is not satisfactory. Modification of photocatalysts with a photo-sensitizer to enhance their light absorption, especially in the visible light region, is a widely adopted strategy to improve their performance. Ru(bpy)₂Cl₂ was used to prepared photosensitized MOF-253–Ru(CO)₂Cl₂ since Ru(bpy)₂Cl₂ can react with the surface N,N-chelated sites to form MOF-253 supported [Ru(bpy)₂(X₂bpy)²⁺], which shows absorption in visible light region. As shown in Fig. 1d, sensitized MOF-253–Ru(CO)₂Cl₂ shows enhanced absorption in the visible light region with the absorption edge extending to 630 nm. The formation of [Ru(bpy)₂(X₂bpy)²⁺] was also confirmed by the higher catalytic activity over MOF-253 supported Ru(bpy)₂Cl₂ (0.21 μmol of CO, 0.46 μmol of HCOO⁻ and 0.07 μmol of H₂) than that over pure Ru(bpy)₂Cl₂ (0.27 μmol of HCOO⁻ and 0.18 μmol of CO) under similar conditions (Table S2, ESI†). The photocatalytic activity over the sensitized MOF-253–Ru(CO)₂Cl₂ was found to increase significantly as compared with the non-sensitized one. The amount of HCOO⁻, CO and H₂ produced in 8 h over sensitized MOF-253–Ru(CO)₂Cl₂ (with a molar ratio of the Ru(bpy)₂Cl₂/Ru-complex of 1 : 2) was determined to be 4.84 μmol, 1.85 μmol and 0.72 μmol, which is much larger than those produced over non-sensitized MOF-253–Ru(CO)₂Cl₂ under similar conditions (Fig. 3). ICP analyses showed that about 9.3% of Ru leached into the solution after 8 h of irradiation. This value is a little higher than that over pure MOF-253–Ru(CO)₂Cl₂ (6.2%), indicating that part of Ru(bpy)₂Cl₂ leached into the solution. However, the additional amount of Ru leaching into solution showed negligible influence on the photocatalytic activity since homogeneous Ru(bpy)₂Cl₂ exhibited low activity under similar conditions. This indicates that the photocatalytic CO₂ reduction over sensitized MOF-253–Ru(CO)₂Cl₂ is heterogeneous photocatalysis. To provide a better understanding of the sensitized mechanism, the photocatalytic performance over sensitized MOF-253–Ru(CO)₂Cl₂ and MOF-253–Ru(CO)₂Cl₂ were performed under light irradiation with a wavelength larger than 500 nm. As expected, no products were detected over MOF-253–Ru(CO)₂Cl₂ since it shows no absorption in this region. Only 0.83 μmol of CO and 0.02 μmol H₂ were detected over sensitized MOF-253–Ru(CO)₂Cl₂. These values are significantly lower than those obtained over wavelengths larger than 420 nm (1.91 μmol of CO, 8.23 μmol of HCOO⁻ and 2.73 μmol of H₂), indicating that it was induced by the photocatalysis of the sensitizer itself.

Fig. 3 Products distribution over MOF-253–Ru(CO)₂Cl₂ with different amounts of Ru(bpy)₂Cl₂.

The photocatalytic performance of the sensitized MOF-253–Ru(CO)₂Cl₂ was found to be significantly influenced by the amount of Ru(bpy)₂Cl₂ immobilized and an optimum activity was observed over sensitized MOF-253–Ru(CO)₂Cl₂ when the molar ratio of the Ru(bpy)₂Cl₂/Ru-complex was 1 : 1. For this system, the amount of the produced HCOO⁻, CO and H₂ reached 8.23 μmol, 2.73 μmol and 1.91 μmol after irradiation for 8 h. Although the amount of CO produced did not change much as compared to the un-sensitized one, the produced HCOO⁻ over the sensitized system was about 12 times that over the un-sensitized one (0.67 μmol), which corresponded to a TON of 35.8 for HCOO⁻ formation (Table 1). This indicates that preparing sensitized MOF-253–Ru(CO)₂Cl₂ by formation of surface incorporated [Ru(bpy)₂(X₂bpy)²⁺] can promote the photocatalytic CO₂ reduction over MOF-253–Ru(CO)₂Cl₂. However, further increase in the amount of Ru(bpy)₂Cl₂ resulted in a decrease of the reactivity, probably due to blocking of the MOF-253 pore structure by the Ru(bpy)₂Cl₂ moiety. This implies that MOF-253 not only can act as a solid ligand for construction of supported photocatalyst MOF-253–Ru(CO)₂Cl₂, but can also act as a platform for building the composite photocatalytic system, which can promote the charge transfer between the photo-sensitizer and the surface constructed photocatalyst.

In summary, the MOF-253 surface constructed Ru carbonyl complex (MOF-253–Ru(CO)₂Cl₂) shows photocatalytic activity for CO₂ reduction under visible light irradiation. Its performance can be further improved by simultaneous immobilization of the photosensitizer. This work provides an effective method for the direct construction of a surface supported molecular photocatalyst for CO₂ reduction. It also highlights the great potential of using MOFs both as a solid ligand for building a supported molecular catalyst and as a platform for assembly of several active moieties into one composite system to achieve complicated functions.

We are grateful to the Shanghai Synchrotron Radiation Facility (SSRF) of China for the XAFS spectra measurements at the BL14W1 beamline. The work was supported by NSFC (21273035), 973 Programs (2014CB239303) and Specialized Research Fund for the Doctoral Program of Higher Education (20123514110002). Z. Li thanks the Award Program for Minjiang Scholar Professorship for financial support.

Notes and references

1. (a) T. R. Cook, Y. R. Zheng and P. J. Stang, Chem. Rev., 2013, 113, 734; (b) J. R. Li, J. Sculley and H. C. Zhou, Chem. Rev., 2012, 112, 869; (c) A. Corma, H. García and F. X. L. i Xamena, Chem. Rev., 2010, 110, 4606; (d) P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G. Férey, R. E. Morris and C. Serre, Chem. Rev., 2012, 112, 1232; (e) Y. Fu, D. Sun, Y. Chen, R. Huang, Z. Ding, X. Fu and Z. Li, Angew. Chem., Int. Ed., 2012, 51, 3364; (f) H. Liu, B. Yin, Z. Gao, Y. Li and H. Jiang, Chem. Commun., 2012, 48, 2033.
2. (a) S. M. Cohen, Chem. Rev., 2012, 112, 970; (b) Z. Wang and S. M. Cohen, Chem. Soc. Rev., 2009, 38, 1315.
3. M. Pintado-Sierra, A. M. Rasero-Almansa, A. Corma, M. Iglesias and F. Sánchez, J. Catal., 2013, 299, 137.
4. J. Canivet, S. Aguado, Y. Schuurman and D. Farrusseng, J. Am. Chem. Soc., 2013, 135, 4195.
5. C. Wang, Z. Xie, K. E. deKrafft and W. Lin, J. Am. Chem. Soc., 2011, 133, 13445.
6. E. D. Bloch, D. Britt, C. Lee, C. J. Doonan, F. J. Uribe-Romo, H. Furukawa, J. R. Long and O. M. Yaghi, J. Am. Chem. Soc., 2010, 132, 14382.
7. M. Wang, B. Yuan, T. Ma, H. Jiang and Y. Li, RSC Adv., 2012, 2, 5528.
8. F. Carson, S. Agrawal, M. Gustafsson, A. Bartoszewicz, F. Moraga, X. Zou and B. Martín-Matute, Chem. – Eur. J., 2012, 18, 15337.
9. T. Zhou, Y. Du, A. Borgna, J. Hong, Y. Wang, J. Han, W. Zhang and R. Xu, Energy Environ. Sci., 2013, 6, 3229.
10. (a) D. G. Whitten, Acc. Chem. Res., 1980, 13, 83; (b) E. Kimura, X. Bu, M. Shionoya, S. Wada and S. Maruyama, Inorg. Chem., 1992, 31, 4542.
11. (a) T. M. Suzuki, H. Tanaka, T. Morikawa, M. Iwaki, S. Sato, S. Saeki, M. Inoue, T. Kajino and T. Motohiro, Chem. Commun., 2011, 47, 8673; (b) S. Sato, T. Arai, T. Morikawa, K. Uemura, T. M. Suzuki, H. Tanaka and T. Kajino, J. Am. Chem. Soc., 2011, 133, 15240; (c) H. Ishida, K. Tanaka and T. Tanaka, Organometallics, 1987, 6, 181.
12. (a) S. Luukkanen, M. Haukka, E. Eskelinen, T. A. Pakkanen, V. Lehtovuori, J. Kallioinen, P. Myllyperkiö and J. Korppi-Tommola, Phys. Chem. Chem. Phys., 2001, 3, 1992; (b) E. Eskelinen, S. Luukkanen, M. Haukka, M. Ahlgrén and T. A. Pakkanen, J. Chem. Soc., Dalton Trans., 2000, 2745–2752.
13. P. Voyame, K. E. Toghill, M. A. Méndez and H. H. Girault, Inorg. Chem., 2013, 52, 10949.
14. J. R. Pugh, M. R. M. Bruce, B. P. Sullivan and T. J. Meyer, Inorg. Chem., 1991, 30, 86.
15. S. Sato, T. Morikawa, T. Kajino and O. Ishitani, Angew. Chem., Int. Ed., 2013, 52, 988.
16. (a) E. Eskelinen, M. Haukka, T. Venäläinen, T. A. Pakkanen, M. Wasberg, S. Chardon-Noblat and A. Deronzier, Organometallics, 2000, 19, 163; (b) A. Gabrielsson, S. Záliš, P. Matousek, M. Towrie and A. Vlček Jr., Inorg. Chem., 2004, 43, 7380.
17. X. Liu, D. Sun, R. Yuan, X. Fu and Z. Li, J. Catal., 2013, 304, 1.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, additional characterizations. See DOI: 10.1039/c4cc09797a

---