TMCnet News
Synthesis and Electron-Transfer Processes in a New Family of Ligands for Coupled Ru-Mn [ChemPlusChem](ChemPlusChem Via Acquire Media NewsEdge) A series of [Ru(bpy)3]2+ -type (bpy = 2,2'-bipyridine) photosensitisers have been coupled to a ligand for Mn, which is expected to give a dinuclear complex that is active as a water oxidation catalyst. Unexpectedly, photophysical studies showed that the assemblies had very short lived excited states and that the decay patterns were complex and strongly dependent on pH. One dyad was prepared that was capable of catalysing chemical water oxidation by using [Ru(bpy)3]3+ as an oxidant. However, photochemical water oxidation in the presence of an external electron acceptor failed, presumably because the short excited-state lifetime precluded initial electron transfer to the added acceptor. The photophysical behaviour could be explained by the presence of an intricate excited-state manifold, as also suggested by time-dependent DFT calculations. Keywords : electron transfer · manganese · photophysics · quantum chemistry · water splitting Introduction With growing awareness of the limitations and disadvantages of the energy sources that form the basis for today's society, interest in exploring alternative energy systems is increasing. One approach to develop systems for fuel production would be to mimic the charge-separation processes in photosystem II (PS II).[1-5] This multi-component protein complex uses light for the extraction of electrons from water to generate biomass from carbon dioxide. The oxygen evolving complex (OEC) is responsible for carrying out the oxidation of H2O and consists of a tetranuclear manganese cluster.[6] Inspired by natural systems, several artificial water oxidation catalysts (WOCs) have been developed, based on a range of different metals.[7-29] In addition, a few model systems for photoinduced electron transfer from manganese complexes to a chromophore have been prepared and characterised.[30-45] Typically, metal porphyrins or [Ru(bpy)3]2+-type (bpy= 2,2'-bipyridine) photosensitisers have been used as chromophores (photosensitisers), which upon excitation can transfer an electron to an external acceptor. The photosensitiser is subsequently regenerated by electron transfer from a nearby electron donor, such as an attached manganese complex. We have previously prepared a series of ruthenium-manganese dyads, but none of them proved capable of catalysing the oxidation of water, which requires the consecutive transfer of four electrons.[46] This is most likely to be related to degradation of the ligands. However, we have recently managed to prepare a dinuclear manganese complex (1), which does catalyse water oxidation (Figure 1).[47] We therefore decided to try to prepare ruthenium-manganese dyads ; this involved coupling of this dinuclear Mn unit to a [Ru(bpy)3]2 + -type photosensitiser. Herein, we present the synthesis, characterisation, and electrochemical and photophysical properties of the ligand complexes 7a-c. The complexes 7a and 7b were first prepared, but turned out to have unexpected and unusual photochemistry, with very short excited-state lifetimes. It seemed likely that this was due to the large conjugated systems and complex 7c was therefore prepared with the hope that the phenyl substituent on the central imidazole would decrease the planarity owing to steric strain. However, this complex also had a very short excited-state lifetime. These results were rather unexpected and prompted us to study the photophysical properties of the complexes in more detail, and also investigate the structural and electronic properties by means of time-dependent density functional theory (TDDFT) calculations. We also prepared dyad 8, although it seemed unlikely that this would have a longer lifetime. It seemed interesting, nonetheless, to see if this complex could catalyse water oxidation, either chemically or by light induction. Results and Discussion Synthesis The preparation of ligand complexes 7 a-c and dyad 8 is depicted in Scheme 1. The synthesis is based on the reaction of 1,10-phenanthroline-5,6-dione (2)withp-hydroxybenzaldehyde and ammonium acetate to yield compound 3a. In the presence of aniline, this reaction gave the N-arylated product 3b. Reaction of 3a with the ruthenium precursor in ethylene glycol at 120 8C gave complex 5a as the triflate salt. To avoid transesterification, the reactions of precursor 4b with compounds 3a and 3b were performed in ethanol at 120 8Cto yield complexes 5b and 5c, respectively. Treatment of complexes 5a-c with 20 molar equivalents of hexamethylenetetramine in TFA at 120 8C for 3 days, followed by hydrolysis of the intermediate imines by aqueous HCl, gave the diformylated ruthenium complexes 6a-c as salts containing a mixture of triflate and trifluoroacetate counterions. Finally, a reductive cyclisation with 2-amino-3-nitrobenzoic acid in a solution of water/ ethanol, with sodium dithionite as the reducing agent, gave the desired ruthenium complexes 7a-c. The coupled Ru^Mn2 dyad 8, which was essentially insoluble in organic solvents, was then obtained by using a Soxhlet extractor to slowly extract the ligand complex 7c into a solution of methanol containing Mn(OAc)2·4H2O and NaOAc. UV/Vis spectroscopy The electronic absorption spectra of 7a-c in a 1:1 mixture of acetonitrile/water (v/v) are shown in Figure 2. They all display an intense band at around l ^ 300 nm, corresponding to the p-p* transitions in the bpy ligands, as also demonstrated by ligands 3 and complexes 5 (Figure S1 in the Supporting Information). The introduction of the ruthenium moiety adds a broad MLCT band centred at l ^ 450 nm, which partly overlaps with the imidazole absorption at l ^350 nm (Figure 2 and Figure S1 in the Supporting Information). The results are consistent with published data[43, 48] and our TDDFT calculations (see below). The MLCT band is sensitive to pH, as shown by the inset in Figure 2. The UV/Vis spectrum of dyad 8 is very similar to that of ligand 7c (Figure 3), although 8 has a higher absorption at l ^ 400 nm, for which the manganese imidazole complex has a distinct absorption (Figure S2 in the Supporting Information) and confirms the coordination of manganese in dyad 8. Adjusting the pH to 1 with HCl had a profound effect on the spectra of 7a-c and 8. An example is shown for 7c in the inset in Figure 2. At neutral pH, the absorption at l ^ 450 nm is broad, with a hump at l ^ 480 nm, suggesting a mixed excited state that contains the MLCT state. At high pH, the spectrum is very similar, but the hump becomes slightly more visible. By contrast, at low pH, a single peak at l ^ 460 nm appears, which is probably due to the unperturbed MLCT state, as also suggested by the long lifetime of this excited state (see below). Owing to the limited solubility of the complexes, determination of the exact molar extinction coefficients was difficult. Electrochemical measurements The limited solubility of the compounds precluded electrochemical characterisation by cyclic voltammetry and 7a- c and 8 were therefore studied by means of differential pulse voltammetry (DPV) in phosphate buffer (0.1 m, pH 7.2), under similar conditions to those of the catalytic experiments (see below). The respective oxidative DPV results are shown in Figure 4 and summarised in Tables 1 and 2. A first redox process at 0.86, 0.87 and 0.97 V versus a normal hydrogen electrode (NHE)[49] was observed for 7a-c, respectively ; this was assigned to ligand-based oxidations, as previously reported for similar complexes.[48] A second oxidation assigned to a metal-based oxidation of the ruthenium centre was observed at 1.38, 1.60 and 1.56 V versus NHE, respectively. The potential of the metal-centred oxidation is higher for 7a-c than that of pristine [Ru(bpy)3]2 + and related complexes (Table 1), which can be explained by the presence of the electron-with-drawing imidazole-phenanthro-line-type ligand. As expected, the oxidation of the ruthenium centre is shifted ( ^0.20 V) to higher potentials for 7b and 7c, which contain electron-with-drawing ester groups on the bpy ligands. Such a potential difference has previously been observed by our group with [Ru(bpy)3]2 + -type complexes 9a and 9b.[46, 50] The metal-based oxidation potential of 7a is rather low in comparison with related compound 10 (Table 1). However, the choice of solvents may have influenced the diverging electrochemical results.[48] The DPV results for 8 also displayed two different oxidation processes, at approximately 0.80 and 1.60 V versus NHE ; the latter is attributed to a ruthenium-centred oxidation. Between the two peaks the current increased, relative to values in 7a-c, possibly because of overlapping redox processes, owing to the introduction of manganese. Interestingly, complex 8 gave rise to a catalytic current with an onset potential at about 1.20 V versus NHE. This suggests that it is an active catalyst for water oxidation. Reductive DPV results were subsequently collected for complexes 7a- c and dyad 8 , and are shown in Table 2. Water oxidation experiments To study whether dyad 8 was capable of catalysing light-driven water oxidation, it was dissolved in phosphate buffer at pH 7.2 and added to sodium persulfate, which acted as the external sacrificial electron acceptor, and the solution was irradiated with visible light. However, at most, trace amounts of oxygen were detected. To ensure that the photosensitiser motif in 8 did not affect the efficiency of the manganese centre as a WOC, experiments with a chemical oxidant were performed. In a typical run, complex 8 was dissolved in phosphate buffer (pH 7.2) and [Ru(bpy)3]3 + was added as the chemical oxidant. Immediate oxygen evolution was observed, showing similar activity to that previously reported for WOC 1, although the measured turnover number was only approximately one (under non-optimised conditions). Control experiments for which the [Ru(bpy)3]3 + oxidant was added to the aqueous phosphate buffer solution, lacking dyad 8, resulted in negligi- ble amounts of produced O2. This validates that the manganese moiety in 8 is also capable of promoting oxidation of water when it is bound to the photosensitiser unit, and that the observed catalytic activity is mediated by dyad 8. Emission properties Steady-state emission spectra for 5a, 7a-c and 8 in argonpurged mixtures of acetonitrile/water at various pH values were recorded after photoexcitation with l = 425 nm light. To determine the effect of pH on the photophysics, the complexes were studied at different pH values and quantum yields were calculated by using [Ru(bpy)3]2 + as a reference. Emission maxima and quantum yields are reported in Table 3, and some examples are shown in Figure 5. All compounds exhibited a weak steady-state emission that could be attributed to 3MLCT emission. Notably, compound 7a displayed its 3MLCT emission maximum at l = 613 nm in neutral pH, whereas the maxima observed for 7b, 7cand 8 were l^ 660 nm; this clearly indicates that, if (CO2Et)2bpy ligands are present, this is where the excited state is delocalised. This is further supported by the fact that the emission maxima for 7b and 7c were not affected to a large extent by changes in pH, whereas for 7a the emission maximum shifted to l = 613 nm at neutral pH and to l = 630 nm at low pH. Thus, in this regard the design strategy, driving the excited state further away from the manganese moiety, was successful. The quantum-yield measurements indicate unexpectedly strong quenching for the 7a- c series, and they also revealed a strong pH dependence, such that the quantum yields were around 2-4 % at low pH and decreased to < 0.2 % at pH 7, with a further decrease at higher pH. To confirm the presence of manganese in complex 8, photophysical control experiments were performed, in which Mn(OAc)2 was added to 7c. The 3MLCT peak intensity immediately decreased. The same experiment with 8 did not result in any appreciable changes in 3MLCT emission intensity ; this confirmed the coordination of manganese in complex 8. We have also recorded HRMS (ESI) spectra on solutions of complex 8,at the concentrations used for the photophysical measurements, immediately and after standing overnight, and found no peak corresponding to free ligand 7c. Notably, at low pH, complex 8 exhibits the highest quantum yield of all of the compounds studied. This is another indication that much of the quenching observed is due to properties of the ligand structure rather than the incorporation of manganese. For some of the complexes (see Figure 5), the emission spectra also revealed a weak emission peak centred at l ^ 480- 500 nm. This peak had a much higher intensity at neutral and high pH relative to that at low pH, which is particularly visible for 7c. Upon addition of Mn(OAc)2 to 7c, this peak was completely removed over the whole pH range investigated. Its nature remains elusive and is discussed in more detail (see below). The excited-state lifetimes (see Table 3 and Figure 6) were measured in argon-purged solutions at low, neutral and high pH. From the emission decays of 5a and 7a, it is evident that even the excited states of precursors 5 are strongly quenched at neutral and high pH. Quenching becomes stronger when the photosensitiser is linked to the conjugated benzimidazole and phenol moieties (as in 7a). Monitoring the emission at l = 610 nm revealed that, at low pH, complex 5a had one excitedstate lifetime (ca. 860 ns), which corresponded to the unperturbed 3MLCT state of the [Ru(bpy)3]2 + photosensitiser (Table 3). At neutral and high pH, complex 5a displayed one additional shorter excited-state lifetime with significant amplitude relative to unperturbed 3MLCT. At low pH, complexes 7 a-c and 8 showed similar pattern to that of 5a at neutral and high pH, with one component with a lifetime corresponding to the unperturbed 3MLCT state and one with an intermediate lifetime (300-540 ns). At high pH, complexes 7 a-c behaved very differently from that of 5a, for which a very short lifetime, 1.4- 3.6 ns, dominated. This was also true for 8, although it also displayed a second short lifetime of 8.3 ns. Finally, at neutral pH, the excited state of complex 7a showed triphasic decay, with one short, one intermediate and one long lifetime. In contrast, complex 7bonly displayed one major fast decay, with a lifetime of 3 ns, and a minor component with a lifetime of 58 ns. Finally, complex 7cdisplayed a single fast component and complex 8 displayed a biphasic decay, similar to the pattern at high pH, with one short (ca. 8 ns) and one intermediate lifetime (ca. 60 ns). The excited-state lifetimes corresponding to species with an emission maximum at l= 480 nm were also determined by using the same setup. In this case, the data clearly showed that all emission components (one or two) disappeared within 30 ns, which again highlighted that this emission did not originate from the 3MLCT state. A quantum chemical description To shed more light on the behaviour of the complexes, TDDFT calculations were performed on 5a, 7a and 8. Optimised structures for 7a and 8 are shown in Figures 7 and 8 and Figure S3 in the Supporting Information. Complex 7aturned out to have four different possible structures with low to moderate energies (Figure 7 and Figure S3 in the Supporting Information), whereas one structure was found for 8 (Figure 8). The optimised structures of 7a were found to be essentially planar (Figure 7 and Figure S3 in the Supporting Information). This is also true for 8, for which the phenyl group at the imidazole is essentially perpendicular to the ring system and causes only a minor distortion of planarity ; the dihedral angle of C1-C2-C3-N1 is 12.78 (Figure 8). It should be noted that TDDFT calculations predict the experimentally observed electronic spectra quite well ; thus there is good reason to believe that the calculated geometries are relevant for our discussion. The system will thus be perfectly conjugated and manganese and ruthenium will strongly interact in 8. This is expected to lead to the observed quenching of the excited state of the photosensitiser and prevent electron transfer from the ruthenium photosensitiser to an external acceptor. Perhaps more surprising is that the imidazole ligand itself is also a strong quencher, which will greatly interfere with this electron transfer even in the absence of manganese. The reason may be that the phenol is deprotonated or at least strongly hydrogen bonded in the relevant lowenergy ground states (see below). The lowest observed excitation for 5a is the HOMO-LUMO transition, which is of chargetransfer character, from the phenol to p* orbitals on bpy, and was calculated to occur at l = 451 nm in water (Table S2 and Figure S8 in the Supporting Information), with significant oscillator strength. Geometry optimisation of the triplet state of 5agives the 3MLCT state 5 a-T1, in which the spin density is located on the bpy ligand and on ruthenium (Figure S7 in the Supporting Information). The transition from 5 a-T1 to S0 corresponds to a wavelength of l = 641 nm from our calculations and is in close agreement with the observed value of l=627 nm. If 5ais mono-deprotonated (total charge of + 1), two isomers can be envisioned : deprotonation of the phenol (5adeprotonated-A) and of the imidazole (5adeprotonated-B). In the ground state, complex 5adeprotonated-B is more stable than that of 5adeprotonated-A by 7.2 kcal mol^1. TDDFT calculations on 5adeprotonated-B show that the lowest transition is the HOMO-LUMO excitation, which corresponds to a wavelength of l = 573 nm. The significant redshift compared with 5a is due to the formation of an anionic ligand. Calculations show that, in the trip- let state, complex 5adeprotonated-A becomes 3.2 kcal mol^1 lower in energy than that of 5adeprotonated-B. The transition from the 5adeprotonated-B triplet to the ground-state singlet corresponds to a wavelength of l = 804 nm, although it was predicted to occur at l = 1045 nm for 5adeprotonated-A triplet. In contrast to the long-lived 3MLCT state of protonated 5a, both of these are p* states that are expected to be short lived, in accordance with experimental observations. In the ground state, the proton affinity of 5adeprotonated-B is calculated to be 298.6 kcal mol^1, which becomes 287.6 kcal mol^1 in the triplet state. This suggests that the pKa of 5a is decreased by approximately eight units upon excitation from the ground state to the triplet state, indicating that proton-transfer reactions should be facile. Relative to 5a, a redshift is observed for 7a (to l = 494 nm (Table S1 in the Supporting Information) from l = 451 nm; observed l ^ 470 nm). Because of the different hydrogen-bonding options introduced by the imidazole groups in the ligand, four possible isomers were found for 7a in the singlet ground state. The two structures with lowest energy (7 a-S0 and 7a-S0-B) are shown in Figure 7, and the other two, 7 a-S0-C and 7a-S0-D, are shown in Figure S3 in the Supporting Information. In the last two isomers, the phenol proton is transferred to the connecting imidazole, which is doubly protonated. In 7 a-S0, the phenol proton is intramolecularly transferred to the adjacent imidazole group to form an ion pair, whereas it is only hydrogen bonded to one imidazole unit in 7 a-S0-B, which was found to be 1.0 kcal mol^1 higher in energy. Isodensity surface plots of selected molecular orbitals (MOs) for 7 a-S0 are reported in Figure S4 in the Supporting Information. The HOMO is the highly conjugated p orbital of the phenanthroline-benzimidazole ligand framework, and the LUMO is a combination of the p* orbitals on the three ligands, with an equal distribution over the two bpy moieties and slightly higher contribution of the phenantroline motif. Because the LUMO is essentially a p* orbital, the excited state is again expected to be short lived, as observed. In addition, HOMO-1 and HOMO-3 consist of mainly ruthenium-centred d orbitals and the HOMO-LUMO gap is calculated to be 2.86 eV (Table S1 in the Supporting Information). Isomer 7 a-S0 has the lowest energy and 7a-S0-B is the second lowest. To further investigate the excited-state manifold of 7a, several low-energy excited singlet states were found by using the TDDFT method and are reported in Table S1 in the Supporting Information. The lowest energy transition was calculated to be at l = 494 nm in water, which corresponded to mainly HOMO- LUMO excitation and is an intraligand charge transfer (ILCT) state, as evidenced by the dipole moment change of about 10 D during excitation. Excited-state 3 in Table S1 in the Supporting Information is of MLCT character, with an absorption wavelength of l =477 nm. A number of transitions with great oscillator strength can be observed in the range of l = 400- 500 nm, which is in good agreement with experimental absorption spectra. From the ground-state structure, a TDDFT geometry optimisation was performed for the first singlet excited state (7 a-S1), which is assigned to HOMO-LUMO excitation and is 53.2 kcal mol^1 above the ground state. In 7 a-S1 (Figure S5 in the Supporting Information), the transition from S0 to S1 was calculated to occur at l= 588 nm with an oscillator strength of 0.1486. To obtain the theoretical emission spectra, we fully optimised the lower triplet excited states of 7a in water. Three close-lying triplet states were obtained, which are labelled as 7 a-T1, 7 a-T2 and 7 a-T3 (Figure S5 in the Supporting Information), according to their relative energies. State 7 a-T1 (46.5 kcal mol^1 higher in energy than that of 7 a-S0)isan3ILCT transition and does not involve ruthenium, as confirmed by no spin density on the ruthenium metal centre (Figure S6 in the Supporting Information) and the emission from 7 a-T1 occurs with a wavelength of l = 679 nm (experimentally observed at l = 620 nm). Excited-state 7a-T2 (48.7 kcal mol^1 higher in energy than that of 7 a-S0) is a metal centred d-d transition (3MC), for which the spin density at ruthenium is 1.87. In 7 a-T2, the two axial Ru^N bonds are elongated to about 2.5 ^, owing to excitation of an electron to this anti-bonding orbital, which has also been observed from previous calculations on related ruthenium-based complexes.[52] The 7 a-T3 state (49.8 kcal mol^1 higher in energy than that of 7 a-S0) is the result of a 3MLCT, as evidenced by the spin density at ruthenium (0.97) and also at one of the two bpy ligands. Similar results have been obtained in previous calculations on a number of ruthenium-based complexes.[52-58] The 3MC states are typically extremely short lived, owing to a large non-radiative decay rate constant, and because it is close to the 3MLCT state in energy, it can be expected to cause activated decay of the 3MCLT state, which contributes to the short lifetime observed ; this clearly also applies to 7b and 7c.[51] For dyad 8, which is formally in the Mn2III,III oxidation state, the high-spin ferromagnetic coupled nonet was first considered and the optimised structures are shown in Figure 8. Spin density analysis, which suggests that the electronic structure of 8 is a nonet state, can be interpreted as featuring a high-spin MnII (SMn1= 5/ 2) ferromagnetically coupled to a high-spin MnIII ( S Mn1 = 2) and antiferromagnetically coupled to a ligand radical (S = 1/2). The ligand radical is mainly centred at the phenolate-imidazole part and at the two bpy moieties, although a small spin density can also be seen at the ruthenium centre. These results suggest that in dyad 8 the ligand backbone is oxidised rather than the second manganese. An antiferromagnetically coupled singlet state was also considered for 8 and calculations show that this state is 0.7 kcal mol^1 higher in energy. The singlet-state electronic structure of 8 can be described as a high-spin MnII (SMn1= 5/2) antiferromagnetically coupled to a high-spin MnIII (SMn1= 2) and a ligand radical (S= 1/2). The undectet with all three sites ferromagnetically coupled lies + 0.9 kcal mol^1 higher in energy than that of the nonet. Discussion of the quenching mechanism We have previously shown that the quenching of the ruthenium excited state by manganese is very efficient if the manganese and ruthenium centres are in close proximity in space. As the distance between the ruthenium and manganese centres becomes shorter than about 10 ^, this results in insufficient electron transfer to the external acceptor to generate RuIII owing to a diffusion-limited electron-transfer process.[39, 59] Although this could be an explanation for the quenching observed for 8, which is highly quenched compared with the parent [Ru(bpy)3]2 + complex, it is clear that here the quenching is also very efficient in the absence of manganese. The finding that the excited states of the ligand complexes 7 a-c were highly quenched was initially surprising. Under neutral and basic conditions, even precursor 5a proved to be highly quenched compared with pristine [Ru(bpy)3]2 + . Furthermore, the fact that 5a exhibits a bi-exponential decay under neutral conditions clearly demonstrates complex photophysical behaviour in the absence of manganese, even for this simple system. The general trend for 5a is seen for all of the complexes studied : at low pH the emission quantum yield increases and the longer-lived components that could be attributed to more or less unperturbed 3MLCT emissions are favoured. By contrast, the short-lived components became more prominent as the pH increased, and for 7 the decay became very short and essentially mono-exponential at high pH. For dyad 8, the decay is still complex at high pH, which indeed confirms previous findings that the introduction of manganese opens up additional quenching processes that are not present in precursors 7 a-c . When the photosensitisers were designed, it was anticipated that the phenyl ring in 7c and 8 would have a great impact on the photophysical properties of the compounds because it could potentially destroy the planarity of the Mn2^Ru ligand, although this does not seem to be the case. The differences between 7b and 7c are so small that no significant effect can be attributed to the introduced phenyl substituent. Judging from the steady-state emission data, it is clear that the 3MLCT emission originates from the (CO2Et)2bpy ligands in the cases involving 7b, 7c and 8, whereas it most likely resides on the phenantroline-benzimidazole part in complex 7a. This is seen from the emission maxima, which are essentially the same for 7b, 7c and 8, and is further supported by the fact that emission maxima for 7b, 7c and 8 are less affected by pH than 7a; this indicates that the excited electron is delocalised over different parts of the molecule in 7a compared with the b and c complexes. Thus, in this regard, the design strategy was successful ; however, it was evidently not enough to reduce the unwanted quenching of the 3MLCT state. So, what is the origin of the quenching of 5a and 7a-c ? NMR spectroscopy and DFT calculations suggest the presence of an intramolecular hydrogen bond between the nitrogen of the benzimidazole and the hydrogen of the phenol, resulting in a formal phenolate, which could play an important role in explaining the observed time-resolved emission of the studied dyads. As the pH is increased, more negative charge would be allocated around the phenol group and cause more efficient electron-transfer quenching from the phenolate to the excited state of ruthenium, which is essentially what is observed. The phenol and imidazole motifs are what distinguish 5a and 7a-c from [Ru(bpy)]/[Ru(phen)]-type complexes, which typically give rise to mono-exponential decays with a lifetime close to 1000 ns. It is also evident that the quenching becomes stronger when the photosensitiser is linked to the combined benzimidazole and phenol moieties (as in 7). A complex structurally related to 5a, which has been reported by Aukauloo and co-workers, contains the phenanotroline-imidazole motif, but lacks the phenol, and yet displays a biphasic decay for which one component can be attributed to unperturbed 3MLCT emission.[60] According to the calculations, the lowest triplet state for 5a is the expected 3MLCT state, whereas for the deprotonated complex 5adeprotonated the state mainly involves the imidazole- phenantroline ligand and is 3ILCT in character, which is essentially a p-p* excited state, and is expected to be very short lived. It is possible that this state could exist in equilibrium with the 3MLCT state, especially because the solvent mixture used contains both protic (water) and non-protic (acetonitrile) solvents. This type of interaction has previously been observed for assemblies with an organometallic photosensitiser attached to an aromatic backbone.[61, 62] The next question is whether this can also be used to explain the dual emission observed in the steady-state emission. Many claims of observed dual emission have turned out to be due to impurities in the sample. In this case, however, it seems more likely that both emission peaks are inherent to the samples, especially because 8 was prepared directly from 7c, but these two complexes do not exhibit the same behaviour upon the addition of manganese to the respective samples. Thus, can the proposed equilibrium, depending on the protonation states, also explain this observed behaviour ? There is indeed literature precedence for emission from derivatives of 2-(2'-hydroxyphenyl)benzimidazole that show pH-dependent emission centred at l ^ 500 nm, just as that observed for some of the complexes in this study. The excited-state lifetimes recorded for these compounds are very similar to what we observe at l= 480 nm: emission that decays back to the ground state in less than 30 ns.[63, 64] It was suggested that the dual emission was caused by proton transfer between excited states. Because calculations show that the excited state of, for example, precursor 5a is a strong acid, such proton transfer could also be responsible for the complex behaviour of complexes 5 and 7. However, to ascertain that this is the reason for the observed behaviour, we would have to perform ultrafast transient absorption studies, which is beyond the scope of the current study. Irrespective of the exact mechanism, it can be concluded that the current ligand design results in quenching of the ruthenium 3MLCT state that is too efficient and precludes the desired consecutive four-electron transfer from manganese to ruthenium, and thus, also photosensitised water oxidation. Conclusion Three new ruthenium-containing ligands, 7 a-c, aimed at the preparation of coupled Ru^Mn2 complexes, such as 8, have been synthesised. Unexpectedly, these ligands turned out to have short-lived excited states, particularly at neutral and high pH. With time-resolved emission studies, it was shown that the decays of their excited states were intricate and fast, and could not be modelled with a mono-exponential decay function. The pH also proved to have a significant influence on the photophysics of the different complexes. Complementary TDDFT calculations revealed the presence of a complex excited-state manifold, as well as possible large effects of different protonation states, which may have contributed to the observed complex decay patterns. One dyad 8 was also prepared and, although it was capable of catalysing the oxidation of water, electrochemically or using [Ru(bpy)3]3 + as an oxidant, it failed to perform water oxidation upon illumination in the presence of an external acceptor, which was anticipated because of its short excited-state lifetime. Although dyad 8 failed to photochemically oxidise water, the results of our study are informative and show that a highly conjugated system containing a combination of imidazole and phenol groups may affect efficient quenching of excited [Ru(bpy)3]2+ -type photosensitisers, presumably by electron transfer from the hydrogen-bonded phenol. Experimental Section Materials and general methods All reagents, including solvents, were obtained from commercial suppliers and used directly without further purification. 1H NMR spectra were recorded at 400 MHz and 13C NMR spectra were recorded at 100 MHz. Chemical shifts (d) are reported in ppm relative to the residual solvent signals ([D6]DMSO: d(H) =2.50 ppm and d(C) =39.52 ppm, D2O: d(H) =4.79 ppm, CD2Cl2: d(H) = 5.32 ppm and d(C) =53.84 ppm, CDCl3: d(H)= 7.26 ppm and d(C)= 77.16 ppm). Splitting patterns are denoted as s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet) and br (broad). High-resolution mass spectra measurements were recorded on a Bruker Daltonics microTOF spectrometer with an electrospray ioniser. The UV/Vis absorption spectra were measured on a CARY 300 Bio UV/ Vis spectrophotometer. DPV measurements were performed in phosphate buffer (0.1 m, pH 7.2) with an Autolab potentiostat with a GPES electrochemical interface (Ec°Chemie) equipped with a glassy carbon working electrode (diameter 3 mm), a saturated calomel reference electrode (SCE) and a Pt wire as the auxiliary electrode. The redox couple [Ru(bpy)3]3+/2+ (E = 1.26 V vs. NHE) was used as an internal standard. Photophysical measurements The emission quantum yields, F,of7a-c and 8 were calculated according to Equation (1): F ¼ðh=href ÞðAref =AÞðI=Iref ÞFref ð1Þ in which [Ru(bpy)3]Cl2 in a 1:1 mixture of acetonitrile/water is used as a reference (which in turn is relative to the quantum yield of [Ru(bpy)3]Cl2 in acetonitrile with a value of 0.059).[51] A and Aref are the absorbance values of the samples at the excitation wavelength (l = 425 nm); I and Iref are the emission intensities of the samples, given by the signed area of the emission spectra ; Fref is the quantum yield of the reference ; and h is the refractive index. The excited-state lifetimes were retrieved by measuring the samples in a deaerated 1:1 mixture of acetonitrile/water (v/v) by using a l = 405 nm pulsed picosecond diode laser as the excitation source. The detection wavelength was set at l = (610^10) nm. The measurement was stopped when 10 000 counts had been collected in the peak channel. The emitted light was detected at magic angle and perpendicular to the excitation light through a monochromator tuned to the maximum emission wavelength of the sample. The photons were collected by a microchannel plate photomultiplier tube from Hammamatsu and fed into a multichannel analyser with 4096 channels. The excited-state decays were evaluated by using the Fluorofit software. The emission spectra of the received samples were measured on a Spex Fluorolog 3 equipped with a xenon lamp. The spectral bandwidth for the emission and excitation monochromators were 3 nm for a good signal-to-noise ratio. The spectra are a result of three averaged measurements. The excitation wavelength was l= 425 nm for all samples. The absorbance of the samples was fixed to 0.1 ^ 0.02 at the excitation wavelength. All samples were measured in a 1:1 mixture of acetonitrile/water (v/v) and the emission spectra were recorded after argon purging. To adjust the pH value, HCl (12 m) and NaOH (5 m) were used. Water oxidation experiments In the water oxidation experiment attempts, the gas phase of the reaction was measured by mass spectrometry. We have previously used the MS system to study oxidation of water, for further experimental details of the MS system, see refs. [18] and [65] . The light source in these experiments was a halogen lamp. Stock solutions of the dyad 8 (1 mm) were made in acetonitrile/water (1:1). The catalyst solutions used in the experiments were made by diluting the stock solutions 10 times with phosphate buffer (0.1 m, pH 7.2). The catalyst solutions (100 mm) were then deoxygenated by bubbling with N2 for at least 5 min before use in the experiments. Na2S2O8 (3.1 mg, 13 mmol) was placed in the reaction chamber. The system was then evacuated with a rough pump and, after the background pressure was reached, 18 mbar of He was introduced into the system. After 10 min, the catalyst (1.0 mL, 100 mm) was injected and after approximately an additional 10 min the light was switched on. The reaction chamber in these experiments was placed in a 100 mL glass with a small flow of cooling water (approximately 0.3 L min^1). The function of this glass vessel was to avoid heating of the system and to act as a UV filter. Quantum chemical calculations DFT calculations were performed with the hybrid B3LYP[66] functional in the Gaussian 09 program.[67] Previous theoretical studies on a number of ruthenium-based systems have shown that B3LYP gives quite good results for absorption and emission spectra.[52-58] The 6-31G(d,p) basis set was used for the C, N, O and H elements, and the SDD[68, 69] pseudopotential was used for Mn and Ru. All calculations were performed in water by employing the CPCM continuum model.[70] Geometry optimisations were first performed for 5a, 7a and 8 in their ground state. For 8, the ferromagnetically coupled high-spin state was considered. For the optical absorption spectrum of 5a and 7a, the ground state of which is closed shell, the 20 lowest singlet-singlet transitions were calculated by using the TDDFT method. The first excited singlet state was optimised by using TDDFT, followed by fluorescence emission spectrum calculations. Three different triplet states were optimised by using different initial guesses, and the phosphorescence emission spectrum was calculated by using TDDFT. Isodensity surface plots were constructed by using the GaussView 5 program.[71] Synthesis 1,10-phenanthroline-5,6-dione (2): This compound was synthesised according to a previously published procedure with minor modifications.[72] A mixture of concentrated H2SO4 (200 mL) and concentrated HNO3 (200 mL) were added to a mixture of 1,10-phenanthroline (20.0 g, 0.111 mol) and KBr (20.0 g, 0.168 mol) under stirring at 0 8C. The red reaction mixture was kept at reflux for 6 h and excess of Br2(g) was quenched by conducting the gas through a saturated aqueous solution of Na2S2O3 by using vacuum suction at the top of the reflux condenser. After being heated to reflux, the pale orange solution was poured into 2 L of ice, resulting in a yellow/green solution. A yellow precipitate was formed by adjustment of the pH to approximately 6 with 10 m NaOH (910 mL). The suspension was divided into two parts ; each part was extracted with CH2Cl2 (5^ 100 mL). The organic phases were combined, dried with MgSO4 and evaporated. Recrystallisation of the orange crude product from EtOH yielded yellow needle-shaped crystals (15.6g, 67%). 1HNMR (400 MHz, CD2Cl2,258C): d=9.04 (dd, J= 4.68, 1.88 Hz, 2H), 8.43 (dd, J=7.84, 1.88 Hz, 2H), 7.56 ppm (dd, J=7.84, 4.68Hz, 2H); 13CNMR (100MHz, CD2Cl2,258C): d=177.77, 154.35, 152.29, 135.67, 129.08, 125.22 ppm ; HRMS (ESI): m/z calcd for C13H11N2O3 [2 +MeOH + H]+: 243.0764; found: 243.0767. 2-(4'-hydroxyphenyl)imidazo[4,5-f]-1,10-phenanthroline (3 a): Compound 3awas synthesised by using a modified published procedure.[73] A solution of 2 (10.0 g, 47.6 mmol) and NH4OAc (73.3 g, 952 mmol) in glacial acetic acid (100 mL) was heated at 90 8C for 3.5 h. Upon removal from the heat, a solution of 4-hydroxybenzaldehyde (7.26 g, 59.8 mmol) in glacial acetic acid (100 mL) was added dropwise. After heating the mixture at 90 8C for 3 h, the mixture was poured into H2O (2 L) and neutralised to pH ^ 7 with an aqueous solution of NH3 (25 wt %, 200 mL). Filtration, washing with H2O (2 L) and acetone (2 L), and drying under vacuum resulted in 3aas an orange compound (8.75 g, 59 %). 1H NMR (400 MHz, [D6]DMSO, 258C): d= 13.50 (s, 1H), 9.98 (s, 1 H), 9.01 (dd, J =4.28, 1.76 Hz, 2H), 8.90 (dd, J=8.12, 1.76 Hz, 2 H), 8.15-8.09 (m, 2 H), 7.87-7.77 (m, 2H), 7.03-6.96 ppm (m, 2H); 13C NMR (100 MHz, [D6]DMSO, 258C): d = 179.32, 156.55, 153.35, 137.25, 128.50, 125.84 ppm; HRMS (ESI): m/z calcd for C19H13N4O[3a+ H]+: 313.1084 ; found : 313.1082. 2-(4'-hydroxyphenyl)-1-phenylimidazo[4,5-f]-1,10-phenanthroline (3 b): The condensation product 3b was prepared according to a published procedure with minor modifications.[74] 4-Hydroxybenzaldehyde (1.162 g, 9.51 mmol) was dissolved in glacial acetic acid (22 mL) followed by addition of aniline (1.04 mL, 11.4 mmol). Compound 2 (2.00 g, 9.51 mmol) and NH4OAc (7.34 g, 95.2 mmol) were added to this orange solution, resulting in a suspension. Another portion of glacial acetic acid (18 mL) was added and the mixture was heated at reflux for 24 h under an argon atmosphere, resulting in a colour change from dark-red to brown/yellow. After being cooled to RT, the reaction mixture was poured into water (400 mL), forming a yellow precipitate. By increasing the pH of this suspension to approximately 5 with the addition of an aqueous solution of NH3 (25 wt %, 14 mL), more precipitate formed. The solid was collected by filtration, washed with water (300 mL) and acetone (350 mL), and dried under vacuum to yield a grey powder (2.677g, 74%). 1H NMR (400 MHz, [D6]DMSO, 258C): d=9.08 (dd, J=4.34, 1.82Hz, 1H), 9.00 (dd, J=8.10, 1.82Hz, 1H), 8.93 (dd, J= 4.26 and 1.62Hz, 1H), 7.86 (dd, J=8.10, 4.34Hz, 1H), 7.77-7.68 (m, 5H), 7.47(dd, J=8.48, 4.26Hz,1H), 7.43-7.38 (m, 2H), 7.33(dd, J= 8.48, 1.62 Hz, 1H), 6.74-6.70 ppm (m, 2H); 13C NMR (100 MHz, [D6]DMSO, 25 8C): d = 158.52, 152.32, 148.38, 147.29, 143.82, 143.61, 137.80, 135.07, 130.63, 130.56, 130.43, 129.72, 128.99, 127.14, 126.31, 123.73, 123.39, 122.41, 120.41, 119.44, 115.14 ppm; HRMS (ESI): m/z calcd for C25H16N4ONa [3b+ Na]+: 411.1206; found: 411.1206. 4,4'-dicarboxy-2,2'-bipyridine :[75] K2Cr2O7 (24.0 g, 81.6 mmol) was slowly added to a solution of 4,4'-dimethyl-2,2'-bipyridine (5.0 g, 27.1 mmol) in sulfuric acid (95 %, 125 mL). After addition, the reaction solution was stirred for 20 h at RT. The deep-green mixture was then diluted with cold water (800 mL), filtered and washed with water until the colour became light yellow. Subsequently, the solid was heated to reflux in nitric acid (50 %, 170 mL) for 4 h and then poured over ice, diluted with water (1 L) and cooled to 5 8C. The precipitate was filtered, washed with water (5^ 50 mL) and acetone (2^ 20 mL), and finally dried under reduced pressure to yield the product as a white solid (5.0 g, 91 %). 1H NMR (400 MHz, 0.9m NaOD): d=8.78 (dd, J=5.2, 0.67 Hz, 2H), 8.40 (dd, J=1.5, 0.67Hz, 2H), 7.87ppm (dd, J=5.2, 1.5Hz, 2H); 13CNMR (100MHz, 0.9m NaOD): d=173.7, 156.4, 150.3, 147.0, 123.9, 121.9 ppm; HRMS (ESI): m/z calcd for C12H7N2O4 [M^H]^ : 243.0411; found: 243.0408. cis-[Ru(bpy)2Cl2]H2O(4a):[76] RuCl3·xH2O (assumption: x=4, 1.6 g, 5.7 mmol), bipyridine (1.9 g, 12 mmol) and LiCl (1.7 g, 40 mmol) were heated at reflux in DMF (10 mL) for 8 h. After the reaction mixture was cooled to RT, acetone (50 mL) was added and the resultant solution was cooled at 0 8C overnight. The solid was filtered, washed with H2O (3 ^ 15 mL) and Et2O (3 ^ 15 mL), and dried to yield a dark-green microcrystalline product (2.0 g, 72 %). 1H NMR (400 MHz, [D6]DMSO): d=9.97 (dd, J=5.56, 1.14 Hz, 2H), 8.64 (d, J=8.15 Hz, 2H), 8.48 (d, J=8.15 Hz, 2H), 8.07 (dt, J=7.76, 1.54 Hz, 2H), 7.77 (dt, J=6.63, 1.32Hz, 2H), 7.68 (dt, J=7.76, 1.54Hz, 2H), 7.51 (d, J=5.56 Hz, 2H), 7.10 ppm (dt, J=6.63, 1.32 Hz, 2H); 13C NMR (100 MHz, [D6]DMSO, 258C): d = 160.18, 158.18, 153.17, 151.94, 134.54, 133.27, 125.31, 125.23, 122.82, 122.46 ppm ; HRMS (ESI): m/z calcd for C20H16Cl2N4NaRu [M+ Na]+: 506.9688; found: 506.9693. [Ru(deeb)2Cl2] (deeb =2,2'-bipyridinyl-4,4'-dicarboxylic acid diethyl ester) (4 b): This ruthenium complex was synthesised according to a previously reported protocol with minor changes.[77] EtOH (99%, 7.5 mL) was added to RuCl3·x H2O (0.321 g, 1.23 mmol) and 2,2'-bipyridine-4,4'-dicarboxylic acid (0.60 g, 2.46 mmol) under an argon atmosphere. The reaction mixture was heated for 17 h in a sealed tube at 110 8 C covered with aluminium foil. After being cooled to RT, the mixture was kept in a refrigerator for 3 h. The product was filtered and washed with EtOH, a saturated solution of Na2B4O7·10 H2O in MeOH/H2O (1:1), MeOH/H2O (1:1) and finally EtOH to yield a black/purple powder (0.71 g, 75%). 1H NMR (400MHz, CDCl3): d=10.45 (d, J=5.90Hz, 2H), 8.83 (d, J=1.74Hz, 2H), 8.66 (d, J=1.74 Hz, 2H), 8.16 (dd, J=5.90, 1.74 Hz, 2H), 7.70 (d, J=6.00 Hz, 2H), 7.49 (dd, J=6.00, 1.74 Hz, 2H), 4.57 (q, J= 7.14Hz, 4H), 4.42 (q, J=7.14Hz, 4H), 1.52 (t, J=7.14 Hz, 6H), 1.39ppm (t, J=7.14Hz, 6H); 13CNMR (400MHz, CDCl3): d=164.3, 163.8, 160.5, 158.2, 155.3, 152.6, 136.5, 135.2, 125.1, 124.6, 122.0, 121.7, 62.7, 62.6, 14.5, 14.3 ppm ; HRMS (ESI): m/z calcd for C32H32Cl2N4O8NaRu [M +Na]+: 795.0533; found: 795.0531. 5a: Compound 3a (1.50 g, 4.80 mmol) and [Ru(bpy)2Cl2] (2.07 g, 4.00 mmol) in ethylene glycol (40 mL) were heated in a round-bottomed flask at 120 8C for 20 h. The reaction mixture was cooled to RT before the addition of H2O (720 mL), which resulted in a clear red solution. A solution of NaOTf (13.8 g, 10 mmol) in H2O (40 mL) was added dropwise with stirring. The precipitate was collected by filtration and washed with a small amount of cold H2O and large amounts of diethyl ether and finally dried under vacuum to give an orange powder (3.78 g, 92 %). 1H NMR (400 MHz, [D6]DMSO, 258C): d=14.14 (brs, 1H), 10.11 (s, 1H), 9.09 (d, J=8.22 Hz, 2H), 8.88 (d, J=8.25 Hz, 2H), 8.84 (d, J=7.95 Hz, 2H), 8.22 (ddd, J= 8.02, 7.94, 1.50 Hz, 2H), 8.16 (AA' of AA'BB', 2H), 8.11 (ddd, J =7.91, 7.88, 1.48 Hz, 2H), 8.04 (dm, J=5.05 Hz, 2H), 7.91 (m, 2 H), 7.84 (dm, J= 5.71 Hz, 2 H), 7.67-7.51 (m, 4H), 7.34 (ddd, J = 7.20, 5.68, 1.26 Hz, 2H), 7.02 ppm (AA' of AA'BB', 2H); 13C NMR (100 MHz, [D6]DMSO, 25 8C): d = 159.64, 156.78, 156.56, 153.32, 151.44, 151.37, 149.76, 149.45, 144.99, 144.48, 137.93, 137.77, 136.87, 130.49, 130.30, 128.43, 127.87, 127.72, 127.15, 126.31, 126.05, 125.48, 124.44, 124.35, 122.27, 121.15, 120.38, 119.07, 115.96, 62.78 ppm ; 19F NMR (376 MHz, [D6]DMSO, 258C): d=^77.75 ppm; HRMS (ESI): m/z calcd for C39H27N8ORu [5a^2 OTf^H]+: 725.1346; found: 725.1362. 5b: Compounds 3a (0.262 g, 0.84 mmol) and 4b (0.541 g, 0.70 mmol) in EtOH (99 %, 7 mL) were heated in a sealed tube at 120 8C for 19 h. The reaction mixture was cooled to RT before dilution with H2O (126 mL) and the dropwise addition of an aqueous solution (7 mL) containing NaOTf (2.41 g, 14 mmol). The formed gel was heated at 100 8C for 1 h with slow stirring and cooled to RT with continued slow stirring, and then stored in the refrigerator overnight. The product was collected by filtration, washed with small amounts of cold water followed by a large amount of Et2O, and finally dried in vacuum to yield an orange powder (0.88 g, 96%). 1H NMR (400 MHz, [D6]DMSO, 258C): d=14.13 (brs, 1H), 10.15(s,1H), 9.35 (d, J=1.83Hz, 2H), 9.33 (d, J=1.77Hz, 2H),9.11 (m, 2H), 8.15 (AA' of AA'BB', 2H),8.10 (d,J=5.96Hz, 2H), 8.08 (dm, J=4.77 Hz, 2H), 7.91 (brm, 2H), 7.88 (dd, J=5.83, 1.78 Hz, 2H), 7.81 (m, 2H), 7.73 (m, 2 H), 7.03 (AA' of AA'BB', 2 H), 4.46 (q, J= 7.05 Hz, 4H), 4.40 (q, J=7.11 Hz, 4H), 1.37 (t, J=7.12 Hz, 6H), 1.31ppm(t,J=7.03Hz,6H); 13CNMR(100MHz,[D6]DMSO,258C): d= 163.47, 163.37, 159.73, 157.32, 157.06, 153.48, 153.00, 152.54, 138.40, 138.27, 138.17, 138.04, 131.04, 128.44, 126.69, 123.91, 123.84, 122.27, 120.31, 119.06, 116.00, 62.34, 62.26, 14.05, 13.98 ppm; 19F NMR (376 MHz, [D6]DMSO, 258C): d=^77.76 ppm; HRMS (ESI): m/z calcd for C51H43N8O9Ru [5b^2 OTf^H]+: 1013.2191; found : 1013.2189. 5c: Compounds 3b (0.70 g, 1.80 mmol) and 4b(1.16 g, 1.50 mmol) in EtOH (99 %, 15 mL) were heated in a sealed tube at 120 8C for 19 h. After being cooled to RT, H2O (210 mL) was added. A clear orange solution was obtained by filtration and being subsequently washed with H2O (60 mL). The addition of NaOTf (5.16 g, 30 mmol) in H2O (15 mL) under stirring resulted in the formation of a fine precipitate. Larger particles formed when this suspension was heated at 100 8C for 1 h. The precipitate was filtered, washed with small amounts of H2O and large amounts of Et2O, and dried in vacuum to give 5c as a powder (1.94 g, 93 %). 1H NMR (400 MHz, [D6]DMSO,258C):d=10.04(s, 1H),9.36(d,J=1.77Hz,1H),9.34(d, J=1.80Hz,1H),9.34(d,J=1.78Hz,1H), 9.30(d,J=1.86Hz,1H), 9.24 (dd, J = 8.30, 1.31 Hz, 1 H), 8.16 (dd, J = 5.29, 1.39 Hz, 1 H), 8.12 (d, J=5.83 Hz, 1H), 8.06 (d, J=5.82Hz, 1H), 8.01 (dd, J=5.27, 1.23 Hz, 1H), 7.95 (dd, J=8.19, 5.31 Hz, 1H), 7.89 (dd, J=5.81, 1.78 Hz, 1H), 7.86 (dd, J=5.86, 1.76 Hz, 1H), 7.83-7.66 (m, 9 H), 7.61 (dd, J=8.66, 5.25 Hz, 1H), 7.47-7.41 (m, 3 H), 6.77 (AA' of AA'BB', 2H), 4.52-4.35 (m, 8H), 1.42-1.28 ppm (m, 12H); 13C NMR (100 MHz, [D6]DMSO, 25 8C): d= 163.46, 163.44, 163.37, 163.36, 159.13, 159.11, 157.29, 157.19, 157.06, 157.04, 154.28, 153.03, 152.92, 152.54, 152.48, 150.85, 149.99, 144.57, 144.54, 138.45, 138.41, 138.33, 137.78, 136.80, 136.17, 131.42, 131.08, 130.84, 130.81, 130.61, 130.56, 128.96, 128.74, 128.57, 127.52, 126.71, 126.65, 125.87, 125.48, 123.94, 123.90, 123.82, 122.26, 121.45, 119.53, 119.06, 115.36, 115.12, 62.34, 62.30, 62.27, 14.05, 14.03 ppm ; 19FNMR (376MHz,[D6]DMSO, 258C):d=^77.75ppm; HRMS (ESI): m/z calcd for C57H48N8O9Ru [5c^2 OTf]2+: 545.1288; found: 545.1307. 6a: Compound 5a (1.0 g, 0.98 mmol) and hexamethylenetetramine (2.74 g, 19.5 mmol) were added to a microwave tube. The vessel was flushed with argon followed by the addition of TFA (6 mL) and finally the tube was sealed. After heating for three days at 110 8C, 4 m HCl (30 mL) was added and the mixture was transferred to a 500 mL E-flask and stirred at RT for 3 h. The pH was adjusted to approximately three with an aqueous saturated solution of NaHCO3 ( ^ 110 mL), resulting in a red and sticky fraction, which was dissolved by successively adding small portions of H2O, vigorous stirring and decantation. The resulting slightly cloudy red solution was filtered and washed with H2O. In total, 200 mL of H2O was added. An orange precipitate formed upon the addition of NaOTf (3.36 g, 19.5 mmol) dissolved in water (6 mL). This suspension was left to stand overnight before final filtration and drying under vacuum. The orange product, 6a(0.802 g, 78%; xTFA = 0.904 and xOTf = 1.096), appeared as a double salt with varying distribution of the triflate and trifluoroacetate anions, depending on the batch (determined by 19F NMR spectroscopy). 1H NMR (400 MHz, [D6]DMSO, 25 8C): d=14.64 (br s, 1 H), 10.42 (s, 2 H), 9.15 (m, 2 H), 8.94 (s, 2H), 8.89 (d, J=8.22Hz, 2H), 8.85 (d, J=7.98Hz,2H), 8.22 (t,J=7.82Hz,2H),8.11(t,J=8.03Hz,2H),8.06 (d,J=5.20Hz,2H), 7.94 (m,2H), 7.85 (d,J=5.27Hz,2H),7.67-7.53 (m,4H),7.35ppm (t, J=6.76 Hz, 2H); 13C NMR (100 MHz, [D6]DMSO, 258C): d= 191.31, 164.13, 156.79, 156.55, 151.52, 151.38, 150.96, 149.87, 145.04, 137.96, 137.80, 133.86, 130.56, 127.88, 127.75, 124.95, 124.47, 124.39, 122.27, 120.76, 119.07 ppm; 19F NMR (376 MHz, [D6]DMSO, 258C): d=^73.47, ^77.75 ppm; HRMS (ESI): m/z calcd for C41H27N8O3Ru [6a^OTf^TFA^H] +: 781.1244 ; found : 781.1246. 6b: TFA (3 mL) was added to an argon-flushed tube containing 5b (0.656 g, 0.5 mmol) and hexamethylenetetramine (1.402 g, 10.0 mmol). The tube was sealed and heated at 110 8C for 3 days. 4 m HCl (15 mL) was added to the reaction mixture and the red solution was quickly transferred to a 250 mL Erlenmeyer flask, and within a few minutes a dark-red sticky fraction was formed. The reaction mixture was stirred for another 3 h at RT before the pH was adjusted to approximately three with a saturated aqueous solution of NaHCO3 ( ^ 44 mL). With successive additions of small portions of water, vigorous stirring and decantation, the sticky fraction of water was dissolved. In total, 500 mL of water was used. This slightly cloudy solution was filtered before the addition of a solution of NaOTf (1.720 g, 10.0 mmol) in water (5 mL), which resulted in the formation of an orange precipitate. After keeping the suspension in the refrigerator overnight, the powder was collected by filtration, washed with a small amount of cold water and a large amount of diethyl ether, and dried under vacuum to yield 6b (0.479 g, 71%) as a double salt, in which xTFA= 0.45 and xOTf=1.55. 1H NMR (400 MHz, [D6]DMSO, 258C): d= 14.28 (br s, 1 H), 10.37 (s, 2H), 9.35 (m, 2H), 9.32 (m, 2 H), 9.15 (m, 2 H), 8.74 (s, 2 H), 8.10 (d, J=5.80 Hz, 2H), 8.07 (dm, J=4.81 Hz, 2H), 7.91 (brm, 2H), 7.88 (dm, J=5.40 Hz, 2H), 7.82 (m, 2 H), 7.73 (m, 2 H), 4.46 (q, J= 7.07 Hz, 4H), 4.40 (q, J=7.01 Hz, 4H), 1.37 (t, J=7.10 Hz, 6H), 1.31ppm (t, J=7.05Hz, 6H); 13CNMR (100MHz, [D6]DMSO, 258C): d= 191.24, 164.91, 163.47, 163.38, 157.33, 157.05, 153.01, 152.61, 152.02, 150.09, 138.41, 138.28, 133.48, 131.21, 126.69, 126.39, 126.13, 125.47, 123.90, 123.84, 122.26, 119.06, 62.34, 62.26, 14.05, 13.98 ppm; 19F NMR (376 MHz, [D6]DMSO, 25 8C): d = ^73.44, ^77.76 ppm ; HRMS (ESI): m/z calcd for C53H43N8O11Ru [6b^OTf^TFA^H] +: 1069.2089 ; found : 1069.2106. 6c: Compound 5c (0.694 g, 0.500 mmol) and hexamethylenetetramine (1.402 g, 10.00 mmol) were added to a microwave tube. After purging with argon, TFA (3.0 mL) was added and the tube was sealed. The reaction mixture was heated at 110 8C for 3 days. HCl (15 mL, 4 m) was added in portions to the reaction and the mixture was quickly transferred to a 100 mL round-bottomed flask. Within a few minutes a two-phase system was formed : a red viscous mass and an almost colourless aqueous phase. After stirring at RT for 3 h, the pH was adjusted to approximately three through the addition of a saturated aqueous solution of NaHCO3 (50 mL). The viscous mass was gradually dissolved by the addition of water in small portions and removal of the liquid part after each addition. The total amount of water added was 400 mL. The solution was filtered to remove undissolved particles. A precipitate was formed by the addition of an aqueous solution of NaOTf (1.721 g, 10.00 mmol) dissolved in water (3 mL) to the filtrate. The suspension was refrigerated overnight before it was filtered, washed with cold water (6 mL) and diethyl ether (200 mL), and finally dried in vacuum to yield 6c (0.514 g, 71%; xTFA = 0.16 and xOTf =1.84). 1H NMR (400 MHz, [D6]DMSO, 25 8C): d=10.22 (s, 2 H), 9.37 (m, 1 H), 9.34 (m, 1H), 9.33 (m, 1H), 9.31 (m, 1H), 9.29 (dm, J=8.56Hz, 1H), 8.17 (dm, J=5.83 Hz, 1H), 8.13 (s, 2H), 8.12 (d, J=5.75 Hz, 1H), 8.06 (d, J=5.86 Hz, 1H), 8.03 (dm, J=5.08 Hz, 1H), 7.96 (dd, J= 8.30, 5.25 Hz, 1H), 7.89 (dd, J=5.80, 1.86 Hz, 1 H), 7.86 (dd, J =5.81, 1.84Hz, 1H),7.85-7.70(m,9H)7.64(dd, J=8.63, 5.31Hz,1H), 7.49 (d, J=8.64Hz, 1H), 4.52-4.35 (m, 8H), 1.43-1.27 ppm (m, 12H); 13C NMR (100 MHz, [D6]DMSO, 258C): d= 190.74, 163.46, 163.43, 163.36, 157.30, 157.19, 157.06, 157.05, 152.91, 152.55, 152.25, 151.03, 150.26, 144.74, 144.69, 138.47, 138.42, 138.36, 136.32, 136.15, 136.02, 131.47, 131.37, 131.05, 128.68, 128.54, 127.72, 127.04, 126.73, 126.67, 126.55 ppm ; 19F NMR (376 MHz, [D6]DMSO, 25 8C): d=^73.46, ^77.74 ppm; HRMS (ESI): m/z calcd for C59H48N8O11Ru [6c^OTf^TFA]2+: 573.1238 ; found: 573.1252. 7a: Compound 6a (0.682 g, 0.65 mmol) and 2-amino-3-nitrobenzoic acid (0.237 g, 1.3 mmol) were dissolved in EtOH (5.2 mL) at 70 8C. This mixture was removed from the heat and a solution of Na2S2O4 (85 wt%, 0.799 g, 3.90 mmol) in H2O (2.6 mL) was added. After heating at 70 8C for 5 h, the reaction mixture was cooled to RT before it was kept at 4 8C overnight. Filtration and washing with large amounts of EtOH, H2O and acetone yielded an orange powder (0.643 g, 95 %). 1H NMR (400 MHz, [D6]DMSO, 25 8C): d = 14.5-13.9 (m, 3H), 9.32 (s, 2H), 9.27 (m, 2H), 8.89 (d, J=7.96 Hz, 2H), 8.85 (d, J=8.35 Hz, 2H), 8.22 (t, J=8.05 Hz, 2H), 8.11 (t, J= 7.76 Hz, 2H), 8.02 (m, 2H), 7.97-7.81 (m, 4H), 7.86 (d, J=5.80 Hz, 2H), 7.76 (d, J=7.53Hz, 2H), 7.65 (d, J=4.48Hz, 2H), 7.59 (t, J= 6.86Hz, 2H), 7.37 (t, J=6.72Hz, 2H), 7.29ppm (t, J=7.78Hz, 2H); HRMS (ESI): m/z calcd for C55H35N12O5Ru [7a+ H]+: 1045.1891; found : 1045.1905. 7b: Compound 6b (338 mg, 0.25 mmol) and 2-amino-3-nitrobenzoic acid (91 mg, 0.5 mmol) were dissolved in absolute EtOH (2 mL) at 70 8C in a microwave tube. A solution of Na2S2O4 (85 wt %, 0.307 g, 1.50 mmol) in water (1 mL) was added and the tube was capped and heated at 70 8C for 5 h. After cooling to RT, the reaction mixture was stored in the refrigerator overnight. The precipitation was filtered and washed with EtOH, water and EtOH to yield a red/orange powder (320 mg, 96 %) 1H NMR (400 MHz, [D6]DMSO, 25 8C): d =14.4-13.9 (m, 3H), 9.40-8.95 (m, 8H), 8.18- 8.01 (m, 4H), 7.98-7.80 (m, 8 H), 7.80-7.60 (m, 4H), 7.30 (t, J = 7.57Hz, 2H), 4.46 (q, J=7.14Hz,4H), 4.39 (q, J=7.07Hz, 4H), 1.38 (t, J=7.08Hz, 6H), 1.31ppm (t, J=7.10Hz, 6H); HRMS (ESI): m/z calcd for C67H51N12O13Ru [7b+ H+]+: 1333.2737; found: 1333.2730. 7c: A solution of NaS2O4 (85 wt %, 31 mg, 0.15 mmol) in H2O (0.1 mL) was added to a solution of 6c (36 mg, 0.025 mmol) and 2amino-3-nitrobenzoic acid (9.1 mg, 0.05 mmol) in EtOH (0.2 mL) dissolved at 70 8C. The reaction mixture was heated in a sealed tube at 70 8C for 5 h. A precipitate was formed upon cooling to RT. After refrigeration overnight, the suspension was filtered and washed with EtOH (8 mL), H2O (2 mL) and EtOH (5 mL), and finally dried under vacuum to yield 7c (28 mg, 80 %). 1H NMR (400 MHz, [D6]DMSO, 25 8C): d = 13.96 (brs, 2H), 9.46-9.18 (m, 5 H), 8.71 (br s, 2H), 8.17 (d, J=5.43Hz, 1H), 8.13 (d, J=5.49Hz, 1H), 8.06 (d, J= 5.83Hz, 1H), 8.00 (d, J=5.10Hz, 1H), 7.98-7.68 (m, 16H), 7.62 (m, 1H), 7.49 (d, J=8.69Hz,1H), 7.28(t,J=7.80Hz, 2H),4.52-4.34(m, 8 H), 1.44-1.25 ppm (m, 12 H); HRMS (ESI): m/z calcd for C73H56N12O13Ru [7c+ 2H]2+: 705.1561; found: 705.1588. Dyad 8: Complex 7c(25.0mg, 0.0178mmol)was placed ina Soxhlet sock. Mn(OAc)2·4H2O (10.9 mg, 0.0444 mmol) and NaOAc (14.6 mg, 0.178 mmol) were added to a 25 mL flask followed by MeOH (10 mL). After Soxhlet extraction for 17 h, the resulting orange suspension was centrifuged and washed with MeOH (2 ^ 4 mL) and dried in vacuum to yield a dark-orange powder (14 mg, 38 %). HRMS (ESI): m/z calcd for C75H56N12O17Mn2Ru [8 + 2OH^MeO^2H]^ : 1608.1696 ; found 1608.1629 ; elemental analysis calcd (%) for C78H75F6Mn2N12O30RuS2 (8+ 2OTf^+8H2O): C 45.71, H 3.69, N 8.20, Mn 5.36, Ru 4.96 ; found : C 45.69, H 3.47, N 9.42, Mn 5.93, Ru 5.17. Acknowledgements We thank the Swedish Energy Agency, the Swedish Research Council (VR), Gçran Gustafsson Foundation for Natural Science and Medical Research and the Knut and Alice Wallenberg Foundation for financial support, and Professor Bo Albinsson at Chalmers University of Technology for fruitful discussions. [1] J. J. Concepcion, R. L. House, J. M. Papanikolas, T. J. Meyer, Proc. Natl. Acad. Sci. USA 2012, 109, 15560 - 15564. [2] L. Sun, L. Hammarstrçm, B. ^kermark, S. Styring, Chem. Soc. Rev. 2001, 30, 36 - 49. [3] K. S. Joya, J. L. Vall^s-Pardo, Y. F. Joya, T. Eisenmayer, B. Thomas, F. Buda, H. J. M. de Groot, ChemPlusChem 2013, 78, 35 - 47. [4] H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, P. Strasser, ChemCatChem 2010, 2, 724 - 761. [5] M. D. K^rk^s, E. V. Johnston, O. Verho, B. ^kermark, Acc. Chem. Res. 2014, 47, 100 - 111. [6] Y. Umena, K. Kawakami, J. Shen, N. Kamiya, Nature 2011, 473, 55 - 60. [7] S. W. Gersten, G. J. Samuels, T. J. Meyer, J. Am. Chem. Soc. 1982, 104, 4029-4030. [8] M. D. K^rk^s, T. ^kermark, E. V. Johnston, S. R. Karim, T. M. Laine, B.-L. Lee, T. ^kermark, T. Privalov, B. ^kermark, Angew. Chem. Int. Ed. 2012, 51, 11589 - 11593; Angew. Chem. 2012, 124, 11757 - 11761. [9] C. Sens, I. Romero, M. Rodr^guez, A. Llobet, T. Parella, J. Benet-Buchholz, J. Am. Chem. Soc. 2004, 126, 7798-7799. [10] R. Zong, R. P. Thummel, J. Am. Chem. Soc. 2005, 127, 12802 - 12803. [11] Y. Xu, A. Fischer, L. Duan, L. Tong, E. Gabrielsson, B. ^kermark, L. Sun, Angew. Chem. Int. Ed. 2010, 49, 8934- 8937; Angew. Chem. 2010, 122, 9118-9121. [12] M. D. K^rk^s, T. ^kermark, H. Chen, J. Sun, B. ^kermark, Angew. Chem. Int. Ed. 2013, 52, 4189-4193; Angew. Chem. 2013, 125, 4283-4287. [13] A. Sartorel, M. Carraro, G. Scorrano, R. D. Zorzi, S. Geremia, N. D. McDaniel, S. Bernhard, M. Bonchio, J. Am. Chem. Soc. 2008, 130, 5006 - 5007. [14] S. Maji, I. L^pez, F. Bozoglian, J. Benet-Buchholz, A. Llobet, Inorg. Chem. 2013, 52, 3591 - 3593. [15] M. D. K^rk^s, E. V. Johnston, E. A. Karlsson, B.-L. Lee, T. ^kermark, M. Shariatgorji, L. Ilag, ^. Hansson, J.-E. B^ckvall, B. ^kermark, Chem. Eur. J. 2011, 17, 7953-7959. [16] H.-W. Tseng, R. Zong, J. T. Muckerman, R. Thummel, Inorg. Chem. 2008 , 47, 11763-11773. [17] J.D.Blakemore,N.D.Schley,D.Balcells,J.F.Hull,G.W.Olack, C.D.Incarvito, O. Eisenstein, G. W. Brudvig, R. H. Crabtree, J. Am. Chem. Soc. 2010, 132, 16017-16029. [18] Y. Gao, T. ^kermark, J. Liu, L. Sun, B. ^kermark, J. Am. Chem. Soc. 2009, 131, 8726-8727. [19] J. Limburg, J. S. Vrettos, L. M. Liable-Sands, A. L. Rheingold, R. H. Crabtree, G.W. Brudvig, Science 1999, 283, 1524-1527. [20] A. K. Poulsen, A. Rompel, C. J. McKenzie, Angew. Chem. Int. Ed. 2005, 44, 6916-6920; Angew. Chem. 2005, 117,7076-7080. [21] P. Kurz, Dalton Trans. 2009, 6103 - 6108. [22] W. A. A. Arafa, M. D. K^rk^s, B.-L. Lee, T. ^kermark, R.-Z. Liao, H.-M. Berends, J. Messinger, P. E. M. Siegbahn, B. ^kermark, Phys. Chem. Chem. Phys. 2014, DOI: 10.1039/C3CP54800G. [23] W. C. Ellis, N. D. McDaniel, S. Bernhard, T. J. Collins, J. Am. Chem. Soc. 2010, 132, 10990-10991. [24] J. L. Fillol, Z. Codol^, I. Garcia-Bosch, L. G^mez, J. J. Pla, M. Costas, Nat. Chem. 2011, 3, 807 - 813. [25] S. M. Barnett, K. I. Goldberg, J. M. Mayer, Nat. Chem. 2012, 4, 498 - 502. [26] Z. Chen, T. J. Meyer, Angew. Chem. Int. Ed. 2013, 52, 700-703; Angew. Chem. 2013, 125, 728-731. [27] M.-T. Zhang, Z. Chen, P. Kang, T. J. Meyer, J. Am. Chem. Soc. 2013 , 135 , 2048-2051. [28] T. Nakazono, A. R. Parent, K. Sakai, Chem. Commun. 2013, 49, 6325 - 6327. [29] M. W. Kanan, D. G. Nocera, Science 2008, 321, 1072 - 1075. [30] A. Magnuson, Y. Frapart, M. Abrahamsson, O. Horner, B. ^kermark, L. Sun, J. Girerd, L. Hammarstrçm, S. Styring, J. Am. Chem. Soc. 1999, 121, 89-96. [31] L. Sun, L. Hammarstrçm, T. Norrby, H. Berglund, R. Davydov, M. Andersson, A. Bçrje, P. Korall, C. Philouze, M. Almgren, S. Styring, B. ^kermark, Chem. Commun. 1997, 607- 608. [32] L. Sun, M. K. Raymond, A. Magnuson, D. LeGourri^rec, M. Tamm, M. Abrahamsson, P. H. Ken^z, J. M^rtensson, G. Stenhagen, L. Hammarstrçm, S. Styring, B. ^kermark, J. Inorg. Biochem. 2000, 78, 15 - 22. [33] P. Huang, A. Magnuson, R. Lomoth, M. Abrahamsson, M. Tamm, L. Sun, B. van Rotterdam, J. Park, L. Hammarstrçm, B. ^kermark, J. Inorg. Biochem. 2002, 91, 159-172. [34] L. Sun, H. Berglund, R. Davydov, T. Norrby, L. Hammarstrçm, P. Korall, A. Bçrje, C. Philouze, K. Berg, A. Tran, M. Andersson, G. Stenhagen, J. M^rtensson, M. Almgren, S. Styring, B. ^kermark, J. Am. Chem. Soc. 1997, 119, 6996-7004. [35] D. Burdinski, K. Wieghardt, S. Steenken, J. Am. Chem. Soc. 1999 , 121, 10781 -10787. [36] D. Burdinski, E. Bothe, K. Wieghardt, Inorg. Chem. 2000, 39, 105 - 116. [37] L. Sun, M. Burkitt, M. Tamm, M. K. Raymond, M. Abrahamsson, D. LeGourri^rec, Y. Frapart, A. Magnuson, P. H. Ken^z, P. Brandt, A. Tran, L. Hammarstrçm, S. Styring, B. ^kermark, J. Am. Chem. Soc. 1999, 121, 6834-6842. [38] M. Sjçdin, S. Styring, B. ^kermark, L. Sun, L. Hammarstrçm, J. Am. Chem. Soc. 2000, 122, 3932-3936. [39] K. E. Berg, A. Tran, M. K. Raymond, M. Abrahamsson, J. Wolny, S. Redon, M. Andersson, L. Sun, S. Styring, L. Hammarstrçm, H. Toftlund, B. ^kermark, Eur. J. Inorg. Chem. 2001, 1019 - 1029. [40] S. Romain, C. Baffert, S. Dumas, J. Chauvin, J.-C. LeprÞtre, D. Daveloose, A. Deronzier, M.-N. Collomb, Dalton Trans. 2006, 5691 - 5702. [41] S. Romain, J.-C. LeprÞtre, J. Chauvin, A. Deronzier, M.-N. Collomb, Inorg. Chem. 2007, 46, 2735-2743. [42] M.-N. Collomb, A. Deronzier, Eur. J. Inorg. Chem. 2009, 2025 - 2046. [43] C. Herrero, J. L. Hughes, A. Quaranta, N. Cox, A. W. Rutherford, W. Leibl, A. Aukauloo, Chem. Commun. 2010, 46, 7605 - 7607. [44] C. Herrero, A. Quaranta, S. Protti, W. Leibl, A. W. Rutherford, D. R. Fallahpour, M.-F. Charlot, A. Aukauloo, Chem. Asian J. 2011, 6, 1335 - 1339. [45] C. E. Castillo, S. Romain, M. Retegan, J.-C. LeprÞtre, J. Chauvin, C. Duboc, J. Fortage, A. Deronzier, M.-N. Collomb, Eur. J. Inorg. Chem. 2012, 5485 - 5499. [46] M. L. A. Abrahamsson, H. B. Baudin, A. Tran, C. Philouze, K. E. Berg, M. K. Raymond-Johansson, L. Sun, B. ^kermark, S. Styring, L. Hammarstrçm, Inorg. Chem. 2002, 41, 1534-1544. [47] E. A. Karlsson, B.-L. Lee, T. ^kermark, E. V. Johnston, M. D. K^rk^s, J. Sun, ^. Hansson, J.-E. B^ckvall, B. ^kermark, Angew. Chem. 2011, 123, 11919 - 11922;Angew.Chem. Int.Ed.2011, 50,11715-11718. [48] F. Lachaud, A. Quaranta, Y. Pellegrin, P. Dorlet, M.-F. Charlot, S. Un, W. Leibl, A. Aukauloo, Angew. Chem. 2005, 117, 1560 -1564; Angew. Chem. Int. Ed. 2005, 44, 1536-1540. [49] Throughout this article, redox potentials are presented against the normal hydrogen electrode (NHE). Occasionally the referenced potentials have been determined versus different electrodes and have been converted to NHE for the purpose of comparison. For conversion constants, see : V. V. Pavlishchuk, A. W. Addison, Inorg. Chim. Acta 2000, 298, 97 - 102. [50] E. Galoppini, W. Guo, W. Zhang, P. G. Hoertz, P. Qu, G. J. Meyer, J. Am. Chem. Soc. 2002, 124, 7801 -7811. [51] A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser, A. von Zelewsky, Coord. Chem. Rev. 1988, 84, 85 - 277. [52] E. Jakubikova, W. Chen, D. M. Dattelbaum, F. N. Rein, R. C. Rocha, R. L. Martin, E. R. Batista, Inorg. Chem. 2009, 48, 10720 - 10725. [53] M.-F. Charlot, Y. Pellegrin, A. Quaranta, W. Leibl, A. Aukauloo, Chem. Eur. J. 2006, 12, 796-812. [54] M.-F. Charlot, A. Aukauloo, J. Phys. Chem. A 2007, 111, 11661 - 11672. [55] H.A. Meylemans, N.H. Damrauer, Inorg. Chem. 2009, 48, 11161-11175. [56] E. Badaeva, V. V. Albert, S. Kilina, A. Koposov, M. Sykora, S. Tretiak, Phys. Chem. Chem. Phys. 2010, 12, 8902 - 8913. [57] S. Verma, P. Kar, A. Das, H. N. Ghosh, Dalton Trans. 2011, 40, 9765 - 9773. [58] C. I. Oprea, B. Frecus, B. F. Minaev, M. A. G^rtu, Mol. Phys. 2011, 109, 2511 -2523. [59] H. Berglund-Baudin, L. Sun, R. Davidov, M. Sundahl, S. Styring, B. ^kermark, M. Almgren, L. Hammarstrçm, J. Phys. Chem. A 1998, 102, 2512 - 2518. [60] A. Quaranta, F. Lachaud, C. Herrero, R. Guillot, M.-F. Charlot, W. Leibl, A. Aukauloo, Chem. Eur. J. 2007, 13, 8201 - 8211. [61] T. A. Grusenmeyer, J. Chen, Y. Jin, J. Nguyen, J. J. Rack, R. H. Schmehl, J. Am. Chem. Soc. 2012, 134, 7497-7506. [62] J. E. Yarnell, J. C. Deaton, C. E. McCusker, F. N. Castellano, Inorg. Chem. 2011, 50, 7820-7830. [63] V. Khorwal, B. Sadhu, A. Dey, M. Sundararajan, A. Datta, J. Phys. Chem. B 2013, 117, 8603 - 8610. [64] M. Novo, M. Mosquera, F. Rodriguez Prieto, Can. J. Chem. 1992, 70, 823 - 827. [65] Y. Xu, T. ^kermark, V. Gyollai, D. Zou, L. Eriksson, L. Duan, R. Zhang, B. ^kermark, L. Sun, Inorg. Chem. 2009, 48, 2717 - 2719. [66] A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5652. [67] M. J. Frisch, et al. , Gaussian 09, Revision A.1, Gaussian Inc. , Wallingford, CT, 2009 . See the Supporting Information for full citation. [68] M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys. 1987, 86, 866-872. [69] D. Andrae, U. H^ußermann, M. Dolg, H. Stoll, H. Preuß, Theor. Chim. Acta 1990, 77, 123-141. [70] M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24, 669-681. [71] R. Dennington, T. Keith, J. Millam, GaussView, Version 5, Semichem Inc. , Shawnee Mission, KS, 2009 . [72] A. He, C. Zhong, H. Huang, Y. Zhou, Y. He, H. Zhang, J. Lumin. 2008, 128, 1291 -1296. [73] P. Lenaerts, A. Storms, J. Mullens, J. D'Haen, C. Gçrller-Walrand, K. Binnemans, K. Driesen, Chem. Mater. 2005, 17, 5194-5201. [74] C. Cheng, C. Liu, F. Wu, Bis-Phenanthroimidazolyl Compound and Electroluminescent Device Using the Same, Patent US 2010/0253208 A1, 2010. [75] A. R. Oki, R. J. Morgan, Synth. Commun. 1995, 25, 4093-4097. [76] B. P. Sullivan, D. J. Salmon, T. J. Meyer, Inorg. Chem. 1978, 17, 3334-3341. [77] I. Katsuhiro, Method for Producing Ruthenium Complex, Patent JP2007031390, 2007. Received: February 5, 2014 Published online on March 12, 2014 Erik A. Karlsson,[a] Bao-Lin Lee,[a] Rong-Zhen Liao,[a] Torbjçrn ^kermark,[a] Markus D. K^rk^s,[a] Valeria Saavedra Becerril,[b] Per E. M. Siegbahn,[a] Xiaodong Zou,[c] Maria Abrahamsson,*[b] and Bjçrn ^kermark*[a] [a] Dr. E. A. Karlsson, Dr. B.-L. Lee, Dr. R.-Z. Liao, Dr. T. ^kermark, Dr. M. D. K^rk^s, Prof. P. E. M. Siegbahn, Prof. B. ^kermark Department of Organic Chemistry Arrhenius Laboratory, Stockholm University 106 91 Stockholm (Sweden) E-mail : [email protected] [b] V. S. Becerril, Prof. M. Abrahamsson Department of Chemical and Biological Engineering Chalmers University of Technology 412 96 Gothenburg (Sweden) E-mail : [email protected] [c] Prof. X. Zou Department of Materials and Environmental Chemistry Arrhenius Laboratory, Stockholm University 106 91 Stockholm (Sweden) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cplu.201402006. (c) 2014 Blackwell Publishing Ltd. |