A new Ru,Ru,Pt supramolecular architecture for photocatalytic H₂ production

Abstract

A new polyazine-bridged RuRuPt trimetallic supramolecular architecture emulates the photophysical properties of the previously reported Ru₂RuPt tetrametallic architecture that exhibits photoinduced charge separation. The RuRuPt complexes are more robust H₂O reduction photocatalysts with enhanced stability compared to the Ru₂RuPt tetrametallic analogues.

---

Solar-to-chemical energy conversion is an important topic in the quest for clean and renewable energy.^1–5 Converting H₂O to H₂ fuel by harvesting solar energy is a complicated multi-electron process that involves bond breaking and formation.^1,6–8 Ru(II)–polyazine complexes, such as the prototypical [Ru(bpy)₃]²⁺ (bpy = 2,2′-bipyridine), are attractive light absorbers (LA) for harnessing solar energy due to their broad UV and visible light absorption and long-lived, strongly reducing/oxidizing excited states.^9–12 Molecular photocatalysts provide a means for analysing and understanding the complicated processes involved in H₂O reduction.¹³

Supramolecular complexes,¹⁴ individual components each possessing their own function assembled into systems that perform a complex task, are an important class of molecular photocatalysts for H₂ production from H₂O. Several supramolecular photocatalysts featuring a Ru–polyazine LA coupled to a reactive metal (RM) such as Co,¹⁵ Rh,^16–21 Pd,^22–24 and Pt,^25–29 are reported. The photocatalytic activity of a series of Ru,Pt bimetallic complexes, [(bpy)₂Ru{phenNHCO(Rbpy)}PtCl₂]²⁺ (phen = 1,10-phenanthroline, R = –COOH, –COOEt, or –CH₃) depends on the nature of R. The charge separated (CS) excited state is most stabilized in the –COOH complex, providing the greatest activity in aqueous solution with 5 TON in 10 h.^25,27 Dimerization of this architecture through the R-bpy unit doubles the efficiency by enhancing Pt–Pt dimerization for proton coupled electron transfer. A Ru,Pt bimetallic system [(^tBu₂bpy)₂Ru(tpphz)PtX₂]²⁺ (^tBu₂bpy = 4,4′-di-tert-butyl-2,2′-bipyridine; tpphz = tetrapyrido[3,2-a:2′,3′-c:3′′,2′′-h:2′′′,3′′′-j]phenazine; X = Cl⁻ or I⁻) produces H₂ from H₂O with 7 TON when X = Cl⁻, while a 40 fold increase in TON and enhanced stability is achieved when X = I⁻.³⁰ It was recently discovered that the active catalyst of the Pd^IICl₂ analogue, [(^tBu₂bpy)₂Ru(tpphz)PdCl₂]²⁺, is colloidal Pd,³¹ corroborating results of a previously reported Ru,Pd photocatalyst.³²

The supramolecular architecture [{(TL)₂Ru(dpp)}₂Ru(BL)PtCl₂]⁶⁺ (Ru₂RuPt; TL = phen or Ph₂phen = 4,7-diphenyl-1,10-phenanthroline; BL = dpp = 2,3-bis(2-pyridyl)pyrazine or dpq = 2,3-bis(2-pyridyl)quinoxaline) is active in photocatalytic H₂ production from H₂O by virtue of photoinduced charge separation in which the HOMO is localized on the terminal Ru and the LUMO is localized on BL coordinated to the Pt RM.^33,34 The catalytic efficiency of this architecture is strongly influenced by the nature of BL; the complexes with BL = dpq display greatly enhanced catalysis compared to their BL = dpp counterparts owing to the stabilized LUMO and enhanced driving force for intramolecular electron transfer toward the reactive Pt centre. Spectroscopic analysis of these complexes is difficult due to the presence of multiple overlapping, strongly absorbing intraligand (IL) π → π* and metal-to-ligand charge transfer (MLCT) transitions in the UV and visible regions, respectively.

Reported herein is the new trimetallic supramolecular architecture [(Ph₂phen)₂Ru(dpp)Ru(bpy)(BL)PtCl₂]⁴⁺ (labelled RuRuPt; BL = dpp or dpq) designed to provide analogous redox, spectroscopic, photophysical, and photocatalytic properties compared to the Ru₂RuPt tetrametallic complexes. Reducing the number of (Ph₂phen)₂Ru^II(dpp) LA units from two to one simplifies the molecular architecture and provides more active photocatalysts for H₂O reduction compared to the Ru₂RuPt analogues. Fig. 1 depicts the structures of the new RuRuPt trimetallic complexes, [(Ph₂phen)₂Ru(dpp)Ru(bpy)(dpp)PtCl₂]⁴⁺ (RuRudppPt) and [(Ph₂phen)₂Ru(dpp)Ru(bpy)(dpq)PtCl₂]⁴⁺ (RuRudpqPt), and the previously reported Ru₂RuPt tetrametallic complexes (Ru₂RudppPt and Ru₂RudpqPt). The synthesis, redox and spectroscopic properties, and photocatalytic activity towards H₂O reduction of this new architecture are discussed below.

Fig. 1 Structural representations of the RuRuPt and Ru₂RuPt complexes.

The RuRuPt complexes were assembled using a six step building block process. Synthetic details are provided in ESI.† The synthetic approach to these structurally complicated supramolecules required initial assembly of the bimetallic complex [(Ph₂phen)₂Ru(dpp)RuCl₂(bpy)](PF₆)₂ by coupling [(Ph₂phen)₂Ru(dpp)](PF₆)₂³⁵ and [(bpy)RuCl₂(DMSO)₂]³⁶ monometallic precursors. This complex was reacted with either dpp or dpq to provide an open coordination site to extend the molecular architecture, thereby generating bimetallics of the design [(Ph₂phen)₂Ru(dpp)Ru(bpy)(BL)](PF₆)₄. The RuRuPt trimetallics, [(Ph₂phen)₂Ru(dpp)Ru(bpy)(BL)PtCl₂](PF₆)₄ (RuRudppPt and RuRudpqPt), were prepared upon reacting the BL-containing bimetallic with cis-[PtCl₂(DMSO)₂]. ESI-MS data is consistent with the desired product for each complex.

Electrochemical analysis of the RuRuPt complexes and their bimetallic precursors indicates orbital energetics that are similar to the analogous Ru₂RuPt tetrametallic complexes and their trimetallic precursors. Square wave voltammograms are provided in Fig. S1 and the data and assignments are given in Table S1 (ESI†). Fig. 2a highlights the similarities between the cyclic voltammograms of RuRudpqPt and Ru₂RudpqPt. The new RuRuPt complexes possess the same spatially separated HOMO and LUMO that was reported in the Ru₂RuPt complexes and is necessary for photoinduced charge separation.^33,34,37 The first oxidation process at 1.54–1.57 V vs. Ag/AgCl is assigned the terminal Ru^II/III oxidation in all four Pt-containing complexes. This couple for Ru₂RudppPt and Ru₂RudpqPt has approximately twice the peak current of RuRudppPt and RuRudpqPt, consistent with the number of terminal Ru centres (two and one, respectively). The first reduction is assigned as BL^0/−; this reduction occurs at −0.39 V and −0.08 V for RuRudppPt and RuRudpqPt, respectively, and this trend is consistent with dpq's stabilized π* orbitals compared to those of dpp. The BL^0/− potentials are quite similar to those of the corresponding tetrametallic complexes (−0.33 V and −0.02 V for Ru₂RudppPt and Ru₂RudpqPt, respectively). This supports the validity of the RuRuPt complexes as analogues for the Ru₂RuPt complexes.

Fig. 2 Cyclic voltammograms measured in CH₃CN with 0.1 M Bu₄NPF₆ (a) and electronic absorption spectra in CH₃CN (b) of Ru₂RudpqPt (black) and RuRudpqPt (blue).

The electronic absorption spectrum for each of the RuRuPt complexes is similar to the Ru₂RuPt analogue while exhibiting a decrease in the molar absorptivity throughout the UV and visible regions due to the absence of a strongly absorbing (Ph₂phen)₂Ru^II(dpp) unit. This is shown in Fig. 2b and Fig. S2 and Table S2 (ESI†). Similar to related Ru(II)–polyazine complexes, the UV and visible regions are dominated by IL and MLCT transitions, respectively, with transitions involving dpp or dpq red-shifted compared to Ph₂phen transitions. The broad, lowest energy absorption band for RuRudpqPt is centred at 543 nm (ε = 22 000 M⁻¹ cm⁻¹) with contributions from Ru → dpp and Ru → dpq ¹MLCT transitions. The lowest energy band for Ru₂RudpqPt is centred at 550 nm with ε = 31 000 M⁻¹ cm⁻¹ due to additional central Ru → dpp and terminal Ru → dpp ¹MLCT transitions. Similar trends are observed in comparing RuRudppPt with Ru₂RudppPt (Table S2, ESI†). The RuRuPt architecture maintains the broad light absorption that covers the entire visible region which is desirable in solar energy conversion, while decreasing the number of possible transitions to allow for further studies such as transient absorption spectroscopy.

The emissive nature of the RuRuPt complexes provides a convenient probe into the excited state dynamics that are similar to those observed for the Ru₂RuPt complexes. A simplified state diagram for RuRudpqPt is pictured in Fig. S3 and emission spectroscopy data is presented in Table S3 and Fig. S4 and S5 (ESI†). Emission is observed from the terminal Ru → dpp ³MLCT excited state in each case. The two [(Ph₂phen)₂Ru(dpp)Ru(bpy)(BL)](PF₆)₄ bimetallic complexes emit at 762 nm with τ = 120 ns and Φ^em = 1.5 × 10⁻³. These homobimetallics serve as photophysical models to study charge separation in the Pt-containing complexes. Coordination of cis-PtCl₂ results in quenched emission (Φ^em = 1.1 × 10⁻³ and 5.2 × 10⁻⁴ for BL = dpp and dpq, respectively) and shortened excited state lifetimes (τ = 90 and 100 ns when BL = dpp and dpq, respectively) as the Pt unit stabilizes the BL(π*) orbitals and enables intramolecular electron transfer to populate a low-lying, non-emissive charge separated (³CS) state. As observed in the Ru₂RuPt complexes, the degree of emission quenching and charge separation is strongly dependent on the nature of BL.^34,37 The emissive ³MLCT state is populated with 98% and 43% efficiency for RuRudppPt and RuRudpqPt with λ^exc = 540 nm (equations in ESI†). These efficiencies are close to the analogous Ru₂RuPt complexes (99% and 51% for Ru₂RudppPt and Ru₂RudpqPt, respectively).³⁴ Less efficient emissive state population results from more efficient population of the ³CS state. The degree by which Φ^em and τ of the ³MLCT state are quenched varies substantially due to population of the ³CS from both the emissive ³MLCT state and a higher-energy ³MLCT state, as discussed previously.^34,37 The agreement in photophysical properties of the Ru₂RuPt and RuRuPt complexes further demonstrates the suitability of the new architecture as analogues to the tetrametallic complexes.

The RuRuPt complexes exhibit remarkably enhanced photocatalytic activity and stability towards H₂ production from H₂O compared to their Ru₂RuPt analogues. Table 1 features the amount of H₂ produced and turnover number (TON = moles of H₂ produced/moles of catalyst) for RuRudppPt and RuRudpqPt as well as for the previously reported Ru₂RudppPt and Ru₂RudpqPt following 470 nm irradiation (flux = 2.3 × 10¹⁹ photons per min) for 20 hours in RT CH₃CN with 50 μM metal complex, 0.62 M H₂O, 1.5 M DMA sacrificial electron donor, and 110 μM [DMAH⁺][SO₃CF₃⁻]. The volume of the solution and the headspace are 4.5 mL and 15.3 mL, respectively. The reported values are the average of three experiments. To highlight the greater activity of the new architecture, a plot of H₂ production vs. time for RuRudpqPt and Ru₂RudpqPt is provided in Fig. 3, and the plot for the dpp analogues are provided in Fig. S6 (ESI†). RuRudpqPt is the most active catalyst in this series, producing 52 ± 4 μmol of H₂ in 20 hours and undergoing 230 ± 20 TON (quantum yield of H₂ production, Φ = 1.1 × 10⁻³). The enhancement in photocatalytic activity within the trimetallic series of RuRudpqPt and RuRudppPt (29 ± 2 μmol of H₂, 130 ± 9 TON, Φ = 6.2 × 10⁻⁴) is related to the enhanced ³CS state population in the BL = dpq complex and agrees with the trend observed in the analogous Ru₂RuPt complexes.³⁴ For each BL, the H₂ production of the RuRuPt complex is greater than the Ru₂RuPt analogue (shown in Fig. 3 for the BL = dpq analogues) despite the smaller molecule's less efficient absorption at 470 nm. The RuRuPt architecture is expected to provide less demanding steric bulk near the cis-PtCl₂ site, whereas the presence of two bulky (Ph₂phen)₂Ru^II(dpp) units in the Ru₂RuPt architecture may hinder interactions between the substrate and the catalytic site. Three dimensional models highlighting the difference in sterics for different geometric isomers of the Ru₂RudpqPt and RuRudpqPt architectures are provided in Fig. S7 (ESI†). The RuRuPt and Ru₂RuPt complexes are expected to exist in up to 16 and 32 isomers, respectively, as a result of Λ and Δ optical isomers as well as the AB chelating nature of the bridging ligands. The distribution of isomers in each sample is unknown; however, steric effects on reactivity, charge separation distance, and orbital overlap are expected to result from structural variations among isomers.

Table 1 Photocatalytic H₂ production data for RuRuPt and Ru₂RuPt supramolecular complexes^a

Complex	μmol H2	TON^b	Φ × 10^4
a 50 μM catalyst in spectral grade CH3CN, 0.62 M H2O, 1.5 M DMA, 110 μM [DMAH^+][SO3CF3^−], λirr = 470 ± 10 nm (flux = 2.3 × 10^19 photons per min). Values represent H2 production after 20 h photolysis. b TON = moles of H2 produced/moles of catalyst. c Quantum yield of H2 production.
Ru2RudppPt	7.1 ± 3.2	32 ± 14	1.6 ± 0.7
Ru2RudpqPt	25 ± 1	110 ± 6	5.5 ± 0.3
RuRudppPt	29 ± 2	130 ± 9	6.2 ± 0.4
RuRudpqPt	52 ± 4	230 ± 20	11 ± 1

---

Fig. 3 Photocatalytic H₂ production with RuRudpqPt (black circles) and Ru₂RudpqPt (white circles) with 50 μM catalyst in spectral grade CH₃CN, 0.62 M H₂O, 1.5 M DMA, and 110 μM [DMAH⁺][SO₃CF₃⁻]. Solutions were irradiated with λ = 470 ± 10 nm.

The new RuRuPt trimetallic supramolecular architecture features a Ru(II)–polyazine LA and a cis-PtCl₂ RM site with a spatially separated HOMO and LUMO that imparts unusual photophysical properties. This architecture provides redox and photophysical properties analogous to the previously reported Ru₂RuPt architecture, demonstrating the ability to design supramolecular complexes with desired properties through knowledge of previously studied systems. The architecture features a terminal Ru-based HOMO and a remote BL-based LUMO which enables photoinduced charge separation. The nature of BL dictates the energy of the LUMO and largely impacts the population of the ³CS state in competition with population of the emissive ³MLCT state, with BL = dpq affording a complex with a more efficiently populated ³CS state compared to the BL = dpp analogue. The enhanced charge separation in RuRudpqPt is important in providing an active and stable catalyst for H₂ production from H₂O (52 μmol of H₂ and 230 TON in 20 h) while RuRudppPt is less active (29 μmol of H₂ and 130 TON in 20 h). Both of these complexes are superior to their Ru₂RuPt analogues, possibly due to steric factors provided by the bulky (Ph₂phen)₂Ru^II(dpp) units or a varied distribution of geometric isomers. Future studies include further probing the unusual excited state dynamics using transient absorption spectroscopy enabled by the less complicated spectroscopy by virtue of the absence of a strongly absorbing LA unit.

Acknowledgement is made to the Chemical Sciences, Geosciences and Biosciences Division, Office of Basic Energy Sciences, Office of Sciences, U.S. Department of Energy (DE FG02-05ER15751) for their generous support of the development of new LAs used in this research.

References

1. A. J. Bard and M. A. Fox, Acc. Chem. Res., 1995, 28, 141–145.
2. D. Gust and T. A. Moore, Science, 1989, 244, 35–41.
3. D. Gust, T. A. Moore and A. L. Moore, Acc. Chem. Res., 2009, 42, 1890–1898.
4. N. S. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 15729–15735.
5. N. D. McDaniel and S. Bernhard, Dalton Trans., 2010, 39, 10021–10030.
6. N. Armaroli and V. Balzani, ChemSusChem, 2011, 4, 21–36.
7. M. Grätzel, Acc. Chem. Res., 1981, 14, 376–384.
8. W. Lubitz and W. Tumas, Chem. Rev., 2007, 107, 3900–3903.
9. B. Durham, J. V. Caspar, J. K. Nagle and T. J. Meyer, J. Am. Chem. Soc., 1982, 104, 4803–4810.
10. K. Kalyanasundaram, Coord. Chem. Rev., 1982, 46, 159–244.
11. T. J. Meyer, Pure Appl. Chem., 1986, 58, 1193–1206.
12. R. J. Watts, J. Chem. Educ., 1983, 60, 834–842.
13. A. J. Esswein and D. G. Nocera, Chem. Rev., 2007, 107, 4022–4047.
14. V. Balzani, L. Moggi and F. Scandola, in Supramolecular Photochemistry, ed. V. Balzani, NATO ASI Series 214, Reidel, Dordrecht, The Netherlands, 1987, pp. 1–28.
15. A. Fihri, V. Artero, M. Razavet, C. Baffert, W. Leibl and M. Fontecave, Angew. Chem., Int. Ed., 2008, 47, 564–567.
16. M. Elvington, J. Brown, S. M. Arachchige and K. J. Brewer, J. Am. Chem. Soc., 2007, 129, 10644–10645.
17. G. F. Manbeck and K. J. Brewer, Coord. Chem. Rev., 2013, 257, 1660–1675.
18. T. A. White, S. L. H. Higgins, S. M. Arachchige and K. J. Brewer, Angew. Chem., 2011, 123, 12417–12421.
19. T. A. White, B. N. Whitaker and K. J. Brewer, J. Am. Chem. Soc., 2011, 133, 15332–15334.
20. R. Zhou, G. F. Manbeck, D. G. Wimer and K. J. Brewer, Chem. Commun., 2015, 51, 12966–12969.
21. T. Stoll, M. Gennari, J. Fortage, C. E. Castillo, M. Rebarz, M. Sliwa, O. Poizat, F. Odobel, A. Deronzier and M.-N. Collomb, Angew. Chem., Int. Ed., 2014, 53, 1654–1658.
22. M. Karnahl, C. Kuhnt, F. W. Heinemann, M. Schmitt, S. Rau, J. Popp and B. Dietzek, Chem. Phys., 2012, 393, 65–73.
23. M. Karnahl, C. Kuhnt, F. Ma, A. Yartsev, M. Schmitt, B. Dietzek, S. Rau and J. Popp, ChemPhysChem, 2011, 12, 2101–2109.
24. S. Rau, B. Schäfer, D. Gleich, E. Anders, M. Rudolph, M. Friedrich, H. Görls, W. Henry and J. G. Vos, Angew. Chem., Int. Ed., 2006, 45, 6215–6218.
25. H. Ozawa, M.-A. Haga and K. Sakai, J. Am. Chem. Soc., 2006, 128, 4926–4927.
26. H. Ozawa and K. Sakai, Chem. Lett., 2007, 36, 920–921.
27. H. Ozawa, Y. Yokoyama, M.-A. Haga and K. Sakai, Dalton Trans., 2007, 1197–1206.
28. K. Sakai and H. Ozawa, Coord. Chem. Rev., 2007, 251, 2753–2766.
29. K. Sakai, H. Ozawa, H. Yamada, T. Tsubomura, M. Hara, A. Higuchi and M.-A. Haga, Dalton Trans., 2006, 3300–3305.
30. M. G. Pfeffer, T. Kowacs, M. Wächtler, J. Guthmuller, B. Dietzek, J. G. Vos and S. Rau, Angew. Chem., Int. Ed., 2015, 54, 6627–6631.
31. M. G. Pfeffer, B. Schäfer, G. Smolentsev, J. Uhlig, E. Nazarenko, J. Guthmuller, C. Kuhnt, M. Wächtler, B. Dietzek, V. Sundström and S. Rau, Angew. Chem., Int. Ed., 2015, 54, 5044–5048.
32. P. Lei, M. Hedlund, R. Lomoth, H. Rensmo, O. Johansson and L. Hammarström, J. Am. Chem. Soc., 2007, 130, 26–27.
33. J. D. Knoll, S. M. Arachchige and K. J. Brewer, ChemSusChem, 2011, 4, 252–261.
34. J. D. Knoll, S. L. H. Higgins, T. A. White and K. J. Brewer, Inorg. Chem., 2013, 52, 9749–9760.
35. M. T. Mongelli and K. J. Brewer, Inorg. Chem. Commun., 2006, 9, 877–881.
36. M. Toyama, K.-I. Inoue, S. Iwamatsu and N. Nagao, Bull. Chem. Soc. Jpn., 2006, 79, 1525–1534.
37. J. D. Knoll, S. M. Arachchige, G. Wang, K. Rangan, R. Miao, S. L. H. Higgins, B. Okyere, M. Zhao, P. Croasdale, K. Magruder, B. Sinclair, C. Wall and K. J. Brewer, Inorg. Chem., 2011, 50, 8850–8860.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental details, electrochemical, spectroscopic, and photocatalytic data, three dimensional models. See DOI: 10.1039/c5cc06463e

‡ Current address: Department of Chemistry and Biochemistry, The Ohio State University, Columbus, OH, 43210, USA.

§ Karen J. Brewer is deceased (October 24, 2014).

---