DOI: 10.1002/chem.201404549

Full Paper

& Mixed-Valent Compounds

Strongly Coupled Cyclometalated Ruthenium–Triarylamine Hybrids: Tuning Electrochemical Properties, Intervalence Charge Transfer, and Spin Distribution by Substituent Effects Chang-Jiang Yao , Hai-Jing Nie , Wen-Wen Yang, Jiang-Yang Shao, Jiannian Yao, and YuWu Zhong*[a]

Abstract: Nine cyclometalated ruthenium complexes with a redox-active diphenylamine unit in the para position to the RuC bond were prepared. MeO, Me, and Cl substituents on the diphenylamine unit and three types of auxiliary ligands—bis(N-methylbenzimidazolyl)pyridine (Mebip), 2,2’:6’,2’’-terpyridine (tpy), and trimethyl-4,4’,4’’-tricarboxylate-2,2’:6’,2’’-terpyridine (Me3tctpy)—were used to vary the electronic properties of these complexes. The derivative with an MeO-substituted amine unit and Me3tctpy ligand was studied by single-crystal X-ray analysis. All complexes display two well-separated redox waves in the potential region of + 0.1 to + 1.0 V versus Ag/AgCl, and the potential splitting ranges from 360 to 510 mV. Spectroelectrochemical meas-

Introduction Mixed-valent (MV) compounds[1] have been the subject of intense research activities since the pioneering work of Creutz and Taube on charge delocalization in diruthenium complexes.[2] At the same time, the seminal work by Hush and others laid the foundation of the electron-transfer theory of MV compounds.[3] To date, MV chemistry has become a powerful means to examine the fundamental electron-transfer process among redox-active components. By analyzing the intervalence charge-transfer (IVCT) transitions of MV compounds, important electron-transfer parameters, such as reorganization energy and electronic coupling parameter Vab, can be estimated. According to the classification of Robin and Day,[4] three categories of MV compounds can be distinguished. In addition, intermediate class II/III borderline systems have received much attention.[5] [a] C.-J. Yao ,+ H.-J. Nie ,+ W.-W. Yang, J.-Y. Shao, Prof. Dr. J. Yao, Prof. Dr. Y.-W. Zhong Beijing National Laboratory for Molecular Sciences CAS Key Laboratory of Photochemistry Institute of Chemistry, Chinese Academy of Sciences Beijing 100190 (P. R. China) E-mail: [email protected] [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201404549. Chem. Eur. J. 2014, 20, 17466 – 17477

urements show that these complexes display electrochromism at low potentials and intense near-infrared (NIR) absorptions. In the one-electron oxidized form, the complex with the Cl-substituted amine unit and Mebip ligand shows a moderate ligand-to-metal charge transfer at 800 nm. The other eight complexes show asymmetric, narrow, and intense intervalence charge-transfer transitions in the NIR region, which are independent of the polarity of the solvent. The Mebip-containing complexes display rhombic or broad isotropic EPR signals, whereas the other seven complexes show relatively narrow isotropic EPR signals. In addition, DFT and time-dependent DFT studies were performed to gain insights into the spin distributions and NIR absorptions.

Beside fundamental electron-transfer studies, the recent resurgence in MV compounds is largely driven by the current need for functional molecular materials for miniaturized electronic devices. MV compounds are well-known for their appealing optical, electronic, and magnetic properties. More importantly, these properties are significantly dependent on the redox states of the materials and the externally applied potentials. This feature makes MV compounds particularly useful in a wide range of optoelectronic applications, such as electrochromic devices,[6] molecular switches,[7] molecular conductance,[8] information storage,[9] two-photon absorption,[10] and high-spin materials.[11] To become useful in the above-mentioned applications, MV compounds should have strong charge delocalization, low redox potentials, and wide structural diversity. We recently reported that the combination of cyclometalated ruthenium and triarylamine (complex 5(PF6)2, Scheme 1) gives rise to a redoxasymmetric MV system with these desirable electronic properties.[12] The strong electronic coupling between ruthenium and amine results in low RuIII/II and NC + /0 potentials with wide potential separation and rich near-infrared (NIR) absorptions in the MV state. Electropolymerized films of a related complex show interesting NIR electrochromism with long retention time.[13] We consider that, with respect to structurally symmetric MV systems with either ruthenium complexes[14] or triarylamines[15] as the redox sites, an asymmetric ruthenium–amine hybrid would allow a wider structural diversity and property

17466

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Scheme 1. Synthesis of 4(PF6)–12(PF6) and chemical structures of model complexes 13(PF6)–15(PF6).

tuning by varying the substituents on the two components.[16] In this context, a series of related complexes 4(PF6)–12(PF6) with three different amine groups and three N,N,N-auxiliary ligands was prepared (Scheme 1). Methoxyl, methyl, and chloro groups were used to tune the properties of the triarylamine subunit. The tridentate ligands bis(N-methylbenzimidazolyl)pyridine (Mebip),[17] 2,2’:6’,2’’-terpyridine (tpy), and trimethyl4,4’,4’’-tricarboxylate-2,2’:6’,2’’-terpyridine (Me3tctpy)[18] were Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

used to vary the properties of the ruthenium subcomponent. This article presents experimental and computational studies on these complexes by single-crystal X-ray crystallography, electrochemistry, IVCT analysis, EPR spectroscopy, and DFT and time-dependent (TDDFT) calculations. In addition, model complexes 13(PF6)–15(PF6) without the amine substituent were prepared for a comparison study.

17467

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper Results and Discussion Synthesis and characterization As outlined in Scheme 1, N,C,N-type amine ligands 1–3 were synthesized in moderate yield by palladium-catalyzed CN coupling of 3,5-bis(pyrid-2-yl)bromobenzene[12] with 4,4’-dimethoxydiphenylamine, 4,4’-dimethyldiphenylamine, and 4,4’-dichlorodiphenylamine, respectively. Following the known procedure for the synthesis of cyclometalated ruthenium complexes,[19] the treatment of [(Mebip)RuCl3] (Mebip = bis(N-methylbenzimidazolyl)pyridine), [(tpy)RuCl3] (tpy = 2,2’:6’,2’’-terpyridine), and [(Me3tctpy)RuCl3] (Me3tctpy = trimethyl-4,4’,4’’tricarboxylate-2,2’:6’,2’’-terpyridine) with AgOTf, followed by reaction with ligand 1, 2, or 3 under microwave heating and anion exchange with KPF6, gave the desired complexes 4(PF6)– 12(PF6) in 34–65 % yield. Complex 5(PF6)2 was isolated with two anions, as was previously reported.[12] Other new complexes were isolated with one anion. A short reaction time (microwave heating for 30 min) is preferred for the isolation of the complexes with one anion. Complexes 13(PF6)[17] and 14(PF6)[20] were synthesized according to the known procedures. Complex 15(PF6) was obtained from the reaction of 1,3bis(pyrid-2-yl)benzene with [(Me3tctpy)RuCl3]. The new compounds were fully characterized by NMR spectroscopy, mass spectrometry, and microanalysis. Signals consistent with the chemical structures of the above complexes (after loss of the PF6 anion) were observed on their MALDITOF mass spectra. Complexes with the Mebip or tpy ligand are easily oxidized in solution. The presence of the resulting paramagnetic impurities makes it difficult to obtain well-defined 1 H NMR signals. However, we found that this issue could be remedied by adding a small amount of aqueous hydrazine to the deuterated solvent (see the NMR spectra in the Supporting Information). The hydrazine is believed to reduce the small amount of the oxidized paramagnetic species and ensure the complex has a charge of 1 + . We previously found that complex 52 + can be fully converted to 5 + by aqueous hydrazine.[12] A single crystal of 6(PF6) was obtained by slow diffusion of hexane into a solution in CHCl3. The single-crystal X-ray structure is shown in Figure 1. The ruthenium ion has a distorted octahedral configuration. The RuC bond is 1.981  in length, which is slightly shorter than the RuN bonds. The distance between the redox-active Ru center and N6 is 6.194 .

Figure 1. Single-crystal X-ray structure of 6(PF6)·2 CHCl3 with 30 % probability thermal ellipsoids. Anion, solvent, and H atoms are omitted for clarity. Selected bond lengths [] and angles [8]: RuC32 1.981(4), RuN1 2.059(3), RuN2 2.007(3), RuN3 2.064(3), RuN4 2.098(3), RuN5 2.098(3), RuN6 6.194, C32-Ru-N2 176.83(14), C32-Ru-N5 77.64(14), C32-Ru-N1 100.81(13), N1Ru-N5 91.73(12), N4-Ru-N5 155.04(12), N1-Ru-N3 156.63(12).

Figure 2. Anodic CVs of a)–c) 4(PF6)–12(PF6) and d) 1–3 and 13(PF6)–15(PF6) in 0.1 m Bu4NClO4/CH3CN.

Electrochemical studies The electronic properties of these complexes were first studied by electrochemical analysis (Figures 2 and 3, Table 1, and Figure S1 in the Supporting Information). In the anodic scan, complexes 4(PF6)–12(PF6) all show two well-separated redox waves in the range of + 0.1 to + 1.0 V versus Ag/AgCl, as shown by the cyclic voltammograms (CVs) and differential pulse voltammograms (DPVs) in Figures 2 and 3. By comparing complexes 4(PF6)–6(PF6) with the same amine substituent but different N,N,N ligands, it is clear that both redox waves of each complex shift to more positive potential as the N,N,N ligand beChem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

Figure 3. Anodic DPVs of 4(PF6)–12(PF6) in 0.1 m Bu4NClO4/CH3CN.

17468

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper significantly increases from 4(PF6) to 7(PF6) and 10(PF6). This is caused by the increasing redox asymmetry DG0 from 4(PF6) to 10(PF6), instead of increasing electronic coupling. The following spectroscopic studies indeed showed that complex 102 + has the lowest degree of charge delocalization among the nine complexes studied. For the compounds of the tpy and Me3tctpy series, the variation of DE versus s is much smaller. In these complexes, the effect of electronic coupling probably plays a more important role in the DE values. Figure 2 d shows the NC + /0 process of ligands 1–3 and the RuIII/II process of model complexes 13(PF6)–15(PF6). The potentials of these processes can be varied effectively by changing the substituent on the amine ligand (MeO, Me, or Cl) or the N,N,N ligand (Mebip, tpy, or Me3tctpy). In the region of more positive potential, some irreversible oxidations are present for all complexes (Supporting Information, Figure S2), which are possibly due to the RuIV/III processes or further oxidation of the aminium component. In the cathodic scan, complexes with Mebip or tpy show one redox wave, whereas complexes with Me3tctpy show two waves. These waves are attributed to the reduction events of these auxiliary ligands.

Table 1. Electrochemical data in CH3CN.[a]

1 2 3 4(PF6) 5(PF6) 6(PF6) 7(PF6) 8(PF6) 9(PF6) 10(PF6) 11(PF6) 12(PF6) 13(PF6) 14(PF6) 15(PF6)

E1/2,anodic [V]

DE[b] [mV]

KC[c]

E1/2,cathodic [V]

+ 0.77 + 0.95 + 1.15 + 0.20, + 0.27, + 0.40, + 0.27, + 0.33, + 0.50, + 0.33, + 0.42, + 0.54, + 0.43 + 0.56 + 0.74

– – – 450 410 360 470 440 370 510 450 380 – – –

– – – 4.2  107 8.9  106 1.3  106 9.2  107 2.9  107 1.9  106 4.4  108 4.2  107 2.8  106 – – –

– – – 1.60 1.50 1.07, 1.42 1.57 1.51 1.07, 1.42 1.58 1.50 1.09, 1.44 1.56 1.51 1.11, 1.46

+ 0.65 + 0.68 + 0.76 + 0.74 + 0.77 + 0.87 + 0.84 + 0.87 + 0.92

[a] The electrochemical potential is reported as the E1/2 value versus Ag/ AgCl. Potential versus ferrocene + /0 can be estimated by subtracting 0.45 V. [b] Potential splitting of the two anodic redox waves of 4(PF6)– 12(PF6). [c] Comproportionation constant, determined as 10(DE/59).

comes increasingly electron-deficient on going from Mebip to tpy and Me3tctpy. Similar trends hold for the 7(PF6)–9(PF6) and 10(PF6)–12(PF6) series. On the other hand, in complexes 4(PF6), 7(PF6), and 10(PF6), with the same N,N,N ligand but different amine substituents, changing the amine substituent from MeO to Me and Cl also leads to a positive shift of both redox waves of each complex. Again, similar trends are shown by the 5(PF6), 8(PF6), 11(PF6) and 6(PF6), 9(PF6), 12(PF6) series. Thus, changing either the amine substituent or the N,N,N ligand causes both redox potentials of any complex to shift significantly in the same direction, which is suggestive of strong coupling between the amine segment and the ruthenium component. In stark contrast, we previously reported that in a related weakly coupled hybrid system,[21] in which the amine substituent and the RuC bond are located in the meta position of the cyclometalated phenyl ring, only very small changes in the RuIII/II potential was observed when the electronic nature of the amine substituent was varied. The potential splitting DE between the two anodic redox waves of 4(PF6)–12(PF6) is in the range of 360–510 mV. Accordingly, the comproportionation constant KC for the equilibrium [N-Ru] + + [N-Ru]3 + !2[N-Ru]2 + is estimated to be in the range of 1.3  106 to 4.4  108. This means that the MV states of these complexes have high thermodynamic stability against disproportionation. Note that the DE values of asymmetric MV compounds are more complicated than those of symmetric analogues. They contain contributions from the electronic coupling, the energy difference DG0 between the reactant and product states of the electron-transfer process, the electrostatic effect, and the inductive effect, among others.[22] Figure S1 in the Supporting Information shows the plots of the potential splitting DE of 4(PF6)–12(PF6) versus the Hammett constant s of the substituents on the diphenylamine unit (MeO, Me, Cl). For the Mebip series compounds, the DE value Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

Spectroscopic studies The electronic absorption spectra of complexes 4(PF6)–15(PF6) are shown in Figure S3 of the Supporting Information. Complexes 4(PF6), 7(PF6), and 10(PF6) with Mebip show similar absorption bands in the visible region. However, the absorptions between 500 and 600 nm are distinctly expanded with respect to 13(PF6) without the amine substituent (Supporting Information, Figure S3a). The main absorption bands in the visible region are largely assigned to the metal-to-ligand charge-transfer (MLCT) transitions. The additional expanded shoulder bands between 500 and 600 nm of 4(PF6), 7(PF6), and 10(PF6) are a result of the intraligand charge-transfer (ILCT) transitions from the amine group. This is discussed below with the aid of TDDFT results. The situations for the series 5(PF6), 8(PF6), 11(PF6), 14(PF6) and 6(PF6), 9(PF6), 12(PF6), 15(PF6) are very similar (Supporting Information, Figure S3b and S3c), in that the shape of the absorption band is largely dependent on the auxiliary ligand. The presence of the amine unit distinctly expands the visible absorption. However, the substituent (MeO, Me, or Cl) of the amine unit makes little difference to the shape and intensity of the absorption. On stepwise oxidative electrolysis at a transparent indium tin oxide (ITO)/glass electrode in CH3CN, complexes 4(PF6)– 12(PF6) all display reversible two-step electrochromism (Figure 4). In the one-electron oxidized states (the MV state), strong NIR absorptions were observed for complexes 42 + –92 + , 112 + , and 122 + . These absorptions are attributed to the IVCT transitions, and they decreased significantly on double oxidation. In principle, these IVCT transition should completely disappear if the complex is fully oxidized to have a charge of 3 + . However, the highly oxidized complexes (particularly 113 + and 123 + ) had poor solubility in the presence of the electrolyte, and precipitates appeared on exhaustive electrolysis at high

17469

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Figure 4. Absorption spectral changes of 4 + –12 + on electrolysis at ITO in 0.1 m Bu4NClO4/CH3CN. The applied potentials are referenced to Ag/AgCl.

potentials. For complex 10(PF6), a moderate absorption band at 800 nm appeared on single oxidation, and this new band decreased a little on further oxidation. Theoretical calculations suggest that the free spin of 102 + is largely localized on the ruthenium component, and the absorption band at 800 is attributed to the ligand-to-metal charge transfer (LMCT) transition. The absorption maxima labs,max and molar extinction coefficients e of these complexes with different total charges are summarized in Table 2, together with those of the three model Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

complexes. When the complex has a charge of 3 + , the absorption pattern is largely dependent on the amine substituent. For instance, complexes 43 + –63 + with the methoxy-substituted amine group all show sharp and intense absorptions around 700 nm, while multiple absorption bands (labs,max around 640 nm) are observed for 73 + –93 + with the methylsubstituted amine group. For the Cl-containing complexes 103 + –123 + , the absorption maxima shift to a region of even higher energy. The e values of the absorptions of 43 + –123 + are more or less underestimated, because some amounts of the

17470

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper Table 2. Absorption data in CH3CN.[a] labs,max [nm] (e [105 M1 cm1]) 4+ 42 + 43 + 5+ 52 + 53 + 6+ 62 + 63 + 7+ 72 + 73 + 8+ 82 + 83 + 9+ 92 + 93 + 10 + 102 + 103 + 11 + 112 + 113 + 12 + 122 + 123 + 13 + 14 + 15 +

332 466 528 424 488 558 323 516 419 332 458 526 424 504 401 323 504 416 332 452 450 424 506 577 323 425 420 334 315 325

(0.36), 348 (0.40), 460 (0.10), 524 (0.17) (0.14), 511 (0.14), 968 (0.108) (0.13), 698 (0.37) (0.090), 502 (0.14), 546 (0.13) (0.12), 1048 (0.124) (0.13), 689 (0.43) (0.29), 339 (0.26), 420 (0.16), 507 (0.13), 583 (0.10) (0.13), 697 (0.077), 1107 (0.126) (0.11), 581 (0.12), 696 (0.31) (0.39), 348 (0.43), 460 (0.11), 520 (0.19) (0.11), 520 (0.12), 920 (0.087) (0.15), 593 (0.17), 640 (0.23) (0.083), 505 (0.14), 542 (0.13) (0.11), 978 (0.055) (0.10), 507 (0.11), 542 (0.11), 585 (0.12), 637 (0.15) (0.28), 340 (0.24), 419 (0.16), 506 (0.11), 586 (0.091) (0.13), 1024 (0.095) (0.13), 507 (0.13), 590 (0.18), 640 (0.26) (0.40), 348 (0.42), 461 (0.11), 520 (0.17) (0.13), 800 (0.048) (0.15), 540 (0.12) (0.083), 504 (0.15), 538 (0.14) (0.11), 953 (0.091) (0.17) (0.37), 340 (0.31), 418 (0.19), 503 (0.14), 582 (0.11) (0.14), 501 (0.15), 982 (0.127) (0.16), 502 (0.15), 583 (0.14) (0.29), 349 (0.39), 457 (0.115), 516 (0.17) (0.37), 372 (0.086), 423 (0.095), 499 (0.14) (0.20), 341 (0.22), 419 (0.19), 587 (0.092)

Figure 5. NIR transitions of 42 + –122 + in CH3CN.

[a] Complexes with a charge of 1 + were obtained according to Scheme 1 and the counteranions are PF6 . Complexes with a charges 2 + or 3 + were obtained by electrolysis, whereby ClO4 counteranions were introduced from the electrolyte.

complexes were not fully oxidized and still had a charge of 2 + , as discussed above. Spectroelectrochemical measurements on these complexes were also carried out in CH2Cl2, acetone, CH3NO2, and DMF. Figure S4 in the Supporting Information shows a comparison of the NIR transitions of the one-electron oxidized forms 42 + –122 + in different solvents. The spectral shapes and energies and the IVCT transitions of these complexes are virtually inde-

pendent of solvent polarity. The largest energy difference is 400 cm1 for complex 52 + between the IVCT bands in CH3NO2 with respect to that in DMF. This solvent effect suggests that these complexes belong to the fully delocalized class III or the borderline II/III systems. Note that the solvatochromic effects of these complexes are much smaller than those of the previously reported bis-triarylamine MV systems. For instance, a class III bis-triarylamine MV system bridged by 1,4-phenylene has a solvatochromic effect of 860 cm1 in CH3CN versus CH2Cl2.[23] In addition, a class II MV system bridged by paraC6H4CCC6H4 has a solvatochromic effect of 2170 cm1 in CH3CN versus CH2Cl2.[24] Figure 5 shows the NIR bands of 42 + –122 + in CH3CN as a function of the wavenumbers (u in cm-1). The variation of the amine substituent and the auxiliary ligand makes a big difference to the shape, energy, and intensity of the IVCT transitions. The IVCT parameters are summarized in Table 3 (complex 102 +

Table 3. Parameters for IVCT transitions.[a]

42 + 52 + 62 + 72 + 82 + 92 + 112 + 122 +

nmax [cm1]

emax [m1 cm1]

Dn1/2(exptl) [cm1][b]

Dn1/2(calcd) [cm1][c]

Dn1/2(high) [cm1][d]

Dn1/2(low) [cm1][e]

Dn1/2(high)/ Dn1/2(low)

nmax/2 [cm1]

mge [D]

Vab,II[f] [cm1]

10 330 9540 9030 10 870 10 220 9770 10 490 10 180

10 800 12 400 12 650 8700 5500 9500 9100 13 000

3880 3540 3320 3320 3400 3430 2950 3200

3935 3796 3763 4045 3927 3960 3981 4054

4400 4420 3890 3680 3960 4230 3500 3720

3350 2670 2740 2960 2840 2630 2400 2690

1.32 1.66 1.42 1.24 1.39 1.61 1.46 1.38

5165 4770 4515 5365 5110 4885 5245 5090

5.4 5.9 5.0 4.7 3.6 4.8 4.6 5.6

1900 1920 1540 1740 1260 1600 1650 1950

[a] Based on the spectroelectrochemical results. [b] The experimentally observed width at half-height. [c] The theoretically predicted width at half-height, as estimated by Dn1/2(calcd) = [2310(nmaxDE)]1/2, where DE is the electrochemical potential splitting (in cm1). [d] Twice the band width at half-height of the high-energy side. [e] Twice the band width at half-height of the low-energy side. [f] Vab values for class II compounds, calculated by (mgenmax)/eR. Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

17471

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper is not included, because only LMCT transitions were observed). The energy nmax of the IVCT band varies in the range of 9030– 10 870 cm1. The emax value varies in the range of 5500– 13 000 cm1. The full width at half-height Dn1/2(exptl) is around 3000–4000 cm1. According to the Hush theory,[3] the predicted band half-width Dn1/2(calcd) for the IVCT of asymmetric MV systems is [2310(nmaxDG0)]1/2. The energy difference DG0 can be estimated to an upper limit by the potential splitting DE (in cm1).[16, 25] Thus, the theoretical lower limits of Dn1/2(calcd) can be estimated for these complexes, which are all slightly wider than the Dn1/2(exptl) values. However, due to the strong ruthenium–amine coupling, a large portion of the DE value should arise from the charge delocalization effect, and the exact DG0 value must be much smaller relative to DE. In this sense, the observed Dn1/2 values should be much narrower with respect to the theoretical values. The IVCT bands of the above complexes are all asymmetric. The ratios between the bandwidth of the high-energy side and the low-energy side Dn1/2(high)/Dn1/2(low) lie in the range of 1.24– 1.66. The asymmetric shape, high intensity, narrow bandwidth, and solvent effect of the IVCT bands suggest that these complexes are very likely fully delocalized class III systems. The Vab values for class III compounds Vab,III can be directly estimated as nmax/2,[3] and the Vab,III values of the above complexes are in the range of 4515–5365 cm1. Note that the use of “IVCT band” is inappropriate for class III compounds, because the valence is averaged in such compounds. A better term to describe the nature of these NIR absorptions is charge-resonance transition. For comparison, the Vab values for class II compounds Vab,II, estimated by mgenmax/eR, are provided in Table 3, where mge is the transition dipole moment of the IVCT band and can be calculated from the integrated absorbance of the IVCT band.[15a,m] The diabatic electron transfer distance R is taken from the Ru– N geometrical distance (6.1  for all), and e is the elementary charge. The Vab,II values of the above complexes are in the range of 1260–1950 cm1. These values are much smaller than the corresponding nmax/2 values. The Vab,II values may be underestimated, since the true electron-transfer distance is very likely shorter than the Ru–N geometrical distance due to the strong electron delocalization. For redox-asymmetric MV systems, nmax is the sum of DG0 and the reorganization energy.[3, 16] If the reorganization energy is assumed to be constant for a series of structurally related compounds and DG0 varies linearly as a function of DE, a good linear relationship between nmax and DE would be expected. Figure S5 in the Supporting Information shows plots of the IVCT energies nmax versus the potential splitting DE of the above complexes. In general, the IVCT energies increase with increasing potential splitting, and the plot of nmax versus DE could be fitted to a linear equation. However, the R2 value for the linear equation is poor (0.7187). This may suggest that the substituent effect has a significant impact on the electronic properties of these MV complexes, which leads to a big variation of the reorganization and/or the DG0 values.

Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

EPR analysis The EPR signals of 42 + –122 + , obtained by chemical oxidation with cerium ammonium nitrate (0.5 equiv oxidant was used in order to prevent over-oxidation), are shown in Figure 6. All measurements were performed at 77 K in frozen CH3CN. Complexes 42 + –62 + and 82 + –122 + show isotropic signals with g factors of 2.108, 2.086, 2.062, 2.095, 2.077, 1.131, 2.112, and

Figure 6. EPR signals of 42 + –122 + .

2.097, respectively. The EPR signal of 102 + is much broader than those of the other complexes. Complex 72 + shows a rhombic signal with g1, g2, and g3 values of 2.288, 2.116, and 1.943, respectively. anisotropy Dg (=g1g3) and the aver The 
 1= age g factor hgi ¼ g21 þ g22 þ g23 3 2 are 0.345 and 2.120, respectively. The rhombic EPR signal of 72 + and broad isotropic signal of 102 + suggest that the ruthenium components of these two complexes make significant contributions to the free spins,[14, 26] whereas the spins of the other seven complexes are more biased toward the triarylamine segment.

DFT and TDDFT calculations DFT and TDDFT calculations were performed to complement the above experimental findings. The TDDFT results of 4 + , 5 + , and 6 + are shown in Table S1 and Figures S6 and S7 in the Supporting Information. Taking complex 4 + as an example, the visible absorptions are mainly associated with the predicted S4, S6, S8, S10, and S15 excitations. Among them, the S4, S8, and S15 excitations have dominant contributions from the HOMO! LUMO + 2, HOMO!LUMO + 3, and HOMO!LUMO + 4 transitions. These transitions are mainly of ILCT character from the amine unit to the cyclometalated ligand. The S6 and S10 excitations are dominated by the MLCT transitions from HOMO1 and HOMO2. Similar ILCT and MLCT transitions were predicted for 5 + and 6 + (Supporting Information, Table S1), and account for the slight expansion of the visible absorption of these complexes with respect to those without the amine substituent (Supporting Information, Figure S3). Complexes 7 + –12 + are believed to have similar transitions, as they have very similar absorption shapes with respect to the methoxy-containing series (4 + –6 + ).

17472

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper bution from the N1 atom increases from 0.238 to 0.257 and 0.278. The methyl-containing series (72 + –92 + ) and chloro-containing series (102 + –122 + ) show similar trends. A similar analysis can be applied to complexes with the same auxiliary ligand but different amine substituent. When the substituent varies from MeO to Me and Cl, the ruthenium contribution of the Mebip-containing series increases from 0.268 (42 + ) to 0.385 (72 + ) and 0.441 (102 + ), and the contribution from the N1 atom decreases from 0.238 to 0.197 and 0.159, respectively. Again, a similar trend is present for the tpy-containing series (52 + , 82 + , and 112 + ) and chlorocontaining series (62 + , 92 + , and 122 + ). The spin-population difference Figure 7. DFT-calculated Mulliken spin-density distributions of 42 + –122 + . See details in Table 4. can be reflected by the spin ratio between the ruthenium ion and the N1 atom, which varies in the range of 0.58 (62 + ) to The geometries of the MV compounds 42 + –122 + were optimized by DFT calculations. It has been reported that class III 2.77 (102 + ). Complex 102 + has the largest Ru/N1 spin ratio, compounds show fully delocalized spin populations in polar which is significantly larger than those of the other complexes. solvents.[27] Figure 7 and Table 4 show the spin-density distribuThis supports the previous spectroscopic results that complex 102 + shows a weak LMCT band rather than a strong IVCT trantions calculated by taking into account the solvent effect in CH2Cl2. The electronic nature of the amine substituent and the sition. The Mebip-containing complexes 72 + and 102 + , which auxiliary ligand has a significant impact on the spin distribushow rhombic or broad isotropic EPR signals, have relatively tions. Taking the methoxy-containing series as an example, as high Ru/N1 spin ratio. However, this rule is not strictly correct. the auxiliary ligand becomes increasingly electron-deficient on Complex 112 + also has a large Ru/N1 ratio, but has a narrow 2+ 2+ 2+ going from 4 to 5 and 6 , the ruthenium contribution isotropic EPR signal. Note that the spin is also rather delocaldecreases from 0.268 to 0.220 and 0.160. Meanwhile the contriized across the Ph1, Ph2, and Ph3 rings. In addition, DFT calculations are known to overestimate charge delocalization, and these results should be taken with care. To rationalize the absorption spectra of the one-electron oxiTable 4. DFT-calculated Mulliken spin density (ab) distribution.[a] dized compounds 42 + –122 + , TDDFT calculations were performed on the above optimized structures. The predicted lowenergy doublet excitations are summarized in Table 5. Chargeresonance transitions with strong oscillator strength were reproduced for 42 + –92 + , 112 + , and 122 + . These transitions are all associated with the spin transitions from the b highest occupied spin orbital (HOSO) to the b LUSO (Figure 8 and Figure S8 42 + 52 + 62 + 72 + 82 + 92 + 102 + 112 + 122 + in the Supporting Information). Other predicted transitions in Ru 0.268 0.220 0.160 0.385 0.356 0.291 0.441 0.414 0.354 the NIR region have zero or negligible oscillator strength (Sup0.230 0.216 0.209 0.263 0.268 0.284 0.265 0.265 0.281 Ph3 porting Information, Table S2). 1 0.238 0.257 0.278 0.197 0.213 0.242 0.159 0.174 0.200 N The TDDFT results of 102 + are different from the others. The 1 0.103 0.124 0.146 0.063 0.071 0.086 0.045 0.055 0.067 Ph 1 predicted D3 excitation with oscillator strength of 0.4457 is re0.020 0.025 0.029 0.003 0.003 0.004 0.004 0.005 0.007 X 0.104 0.125 0.146 0.062 0.070 0.085 0.046 0.055 0.067 Ph2 sponsible for the observed absorption band at 800 nm. This 0.020 0.025 0.029 0.003 0.003 0.004 0.004 0.005 0.007 X2 excitation is dominated by the spin transition from b-HOSO1 1.95 1.67 1.20 2.77 2.38 1.77 Ru/N1 1.13 0.86 0.58 to b-LUSO (right panel in Figure 8), which is mainly involved in [a] Method of calculation: UB3LYP/LANL2DZ/6-31G*/CPCM(solvent = the LMCT transitions from the Mebip ligand to the ruthenium CH2Cl2). ion. Figure S9 in the Supporting Information shows the plot of Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

17473

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper purities. However, the presence of small amount of aqueous hydrazine successfully resolves this issue, and satisfactory NMR spectra have been recorded for all new cyclometalated ruthenium complexes studied herein. We believe that these two experimental techniques would be highly useful for the future study and characterization of cyclometalated complexes. Thirdly, this work demonstrates that, by combining organic and inorganic redox sites, a library of structurally asymmetric MV compounds can be easily expanded by changing the substituent or auxiliary ligand. With m types of amine substituents and n types of auxiliary ligand at hand, a library of m  n twocenter compounds can be readily obtained. Such a method would be very efficient for expanding structural diversity of related materials. Fourthly, the substituent effect has a significant impact on the electrochemistry, IVCT transitions, and spin distributions of the resulting MV compounds. All nine compounds, except 102 + , show interesting electrochromism at low potentials and with intense NIR absorptions. They belong to the fully delocalized class III systems, according to the asymmetric shape, high intensity, narrow bandwidth, and solvent effect of the IVCT bands. Such compounds are highly useful as information-storage media.[13] The variation of the redox potentials and IVCT energies would allow us to make films with multistate electrochromism by copolymerization of two complexes (e.g., complexes 5 and 11 with large potential difference),[29] which will be the subject of future work. Finally, DFT and TDDFT calculations were found to provide useful and complementary information on the spin populations and IVCT transitions, and they are expected to attract increasing attention in the study of MV compounds and transition-metal complexes.[27, 30]

Table 5. TDDFT-predicted doublet (D) excitations.[a]

42 + 52 + 62 + 72 + 82 + 92 + 102 + 112 + 122 +

Excitation

E [cm1]

l [nm]

f

Major contribution

D3 D3 D1 D3 D3 D3 D3 D3 D3

8979.4 8838.3 9030.2 9794.1 9457.7 9382.7 10264.3 9827.9 9565.0

1113.7 1131.4 1107.4 1021.0 1057.3 1065.8 974.3 1017.5 1045.5

0.4577 0.4680 0.4297 0.4408 0.4654 0.5116 0.4165 0.4457 0.5068

H(b)!L(b) (93 %) H(b)!L(b) (93 %) H(b)!L(b) (86 %) H(b)!L(b) (94 %) H(b)!L(b) (94 %) H(b)!L(b) (92 %) H-1(b)!L(b) (92 %) H(b)!L(b) (94 %) H(b)!L(b) (93 %)

[a] The computational method was UB3LYP/LANL2DZ/6-31G*/CPCM/ CH2Cl2. f is the predicted oscillator strength. H = HOSO; L = LUSO.

Figure 8. Spin orbitals of a) 42 + and b) 102 + involved in the TDDFT results shown in Table 5. Other spin orbitals are provided in Figure S6 in the Supporting Information.

Experimental Section

the calculated NIR absorption energies of 42 + –122 + versus the observed values. The trend in transition energies is basically reproduced by the calculation.

Conclusion We have presented an experimental and theoretical study on a series of strongly coupled cyclometalated ruthenium–triarylamine hybrid compounds. We consider that the following information is of high importance for the design and study of cyclometalated complexes, new MV systems, and electrochromic molecular materials. Firstly, Me3tctpy seems to be a good auxiliary ligand for obtaining single crystals of cyclometalated complexes. In addition to the ruthenium complex 6(PF6), we have recently reported a Me3tctpy-containing cyclometalated osmium complex,[28] the structure of which has also been confirmed by single-crystal X-ray analysis. Cyclometalated complexes are generally easily oxidized by oxygen, and the stability can be enhanced by the electron-withdrawing Me3tctpy ligand. Secondly, due to the stability issue of cyclometalated ruthenium complexes, it is sometimes difficult to obtain well-defined NMR signals as a result of the presence of paramagnetic imChem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

Spectroscopic measurements Absorption spectra were recorded on a PE Lambda 750 UV/Vis/NIR spectrophotometer at room temperature. Spectroelectrochemical measurements were performed in a thin-layer cell (optical length = 0.2 cm), in which an ITO/glass working electrode (< 10 W/square) was set in the indicated solvent containing the compound to be studied (c  5  105 m) and 0.1 m Bu4NClO4 as the supporting electrolyte. A platinum wire and Ag/AgCl in saturated aqueous NaCl were used as the counter electrode and reference electrode, respectively. The cell was put into the spectrometer to monitor the spectral change during electrolysis.

Electrochemical measurements All electrochemical measurements were performed by using a CHI 660D potentiostat with one-compartment electrochemical cell under an atmosphere of nitrogen. All measurements were carried out in 0.1 m Bu4NClO4 in indicated solvents at a scan rate of 100 mV s1. The glassy carbon working electrode had a diameter of 3 mm. The electrode was polished prior to use with 0.05 mm alumina and rinsed thoroughly with water and acetone. A large-area platinum wire coil was used as the counter electrode. All potentials are referenced to an Ag/AgCl electrode in saturated aqueous NaCl without regard for the liquid-junction potential. The potential versus ferrocene + /0 can be estimated by subtracting 0.45 V.

17474

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper X-ray crystallography X-ray diffraction data were collected on a Rigaku Saturn 724 diffractometer with a rotating anode (MoKa radiation, 0.71073 ) at 173 K. The structure was solved by the direct method by using SHELXS-97[31] and refined with Olex.[32] Crystallographic data for 6(PF6)·2 CHCl3 : C53H43Cl6F6N6O8PRu, M = 1350.67, triclinic, space group P1¯, a = 10.585(2), b = 13.993(1), c = 20.209(4) ; a = 81.71(3), b = 77.29(3), g = 75.14(3)8; V = 2810.1(10) 3 ; T = 173 K; Z = 2; final R indices: R1 = 0.0664, wR2 = 0.1505, R indices (all data): R1 = 0.0748, wR2 = 0.1573. CCDC 996696 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/ data_request/cif.

Computational methods DFT calculations were carried out with the B3LYP exchange correlation functional,[33] as implemented in the Gaussian 09 package.[34] The electronic structures were optimized by using a general basis set with the Los Alamos effective core potential LanL2DZ basis set for Ru[35] and 6-31G* for other atoms.[36] The computations were carried out with solvation effects taken into account by means of the conductor polarizable continuum model (CPCM).[37] No symmetry constraints were used in the optimization (nosymm keyword was used). Frequency calculations were performed at the same level of theory to ensure that the optimized geometries were local minima. All orbitals were computed at an isovalue of 0.02 e bohr3.

EPR measurements EPR measurements were performed on a Bruker ELEXSYS E500-10/ 12 spectrometer at 77 K in frozen CH3CN. The spectrometer frequency was 9.7  109 Hz. The MV complex was obtained by chemical oxidation with 0.5 equiv cerium ammonium nitrate and used directly for the EPR measurement.

Synthesis NMR spectra were recorded on a Bruker Avance 400 MHz spectrometer. Spectra are reported in parts per million (ppm) from residual proton signals of deuterated solvent. Mass data were obtained with a Bruker Daltonics Inc. Apex II FT-ICR or Autoflex III MALDI-TOF mass spectrometer. The matrix for MALDI-TOF measurement was a-cyano-4-hydroxycinnamic acid. Microanalysis was carried out on a Flash EA 1112 or Carlo Erba 1106 analyzer at the Institute of Chemistry, Chinese Academy of Sciences. 3,5-Bis(pyrid2-yl)-bromobenzene,[12] ligand 1,[12] and complexes 5(PF6)2,[12] 13(PF6),[17] and 14(PF6)[20] were prepared according to the published procedures. Synthesis of N,N-Bis(p-tolyl)-3,5-bis(pyrid-2-yl)aniline (2): [Pd2(dba)3] (dba = trans,trans-dibenzylideneacetone; 0.02 mmol, 18.3 mg), 1,1’-bis(diphenylphosphino)ferrocene (dppf; 0.02 mmol, 11.1 mg), and NaOtBu (1.2 mmol, 115.3 mg) were added to a solution of 3,5-bis(pyrid-2-yl)bromobenzene (1.0 mmol, 323 mg) and dip-tolylamine (1.5 mmol, 296 mg) in 20 mL of dry toluene under N2 atmosphere. The mixture was bubbled with nitrogen for 10 min, followed by heating to reflux for 24 h. After cooling to room temperature, the solvent was removed under reduced pressure. The residue was subjected to flash column chromatography on silica gel (eluent: CH2Cl2/ethyl acetate/NH4OH 100/5/1) to afford 192 mg of ligand 2 as a white solid in 45 % yield. 1H NMR (400 MHz, CDCl3): d = 2.30 (s, 6 H), 7.05 (s, 8 H), 7.18–7.23 (m, 2 H), 7.70–7.72 Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

(m, 6 H), 8.20 (s, 1 H), 8.63 (d, J = 6.3 Hz, 2 H); 13C NMR (100 MHz, CDCl3): d = 20.8, 119.6, 121.0, 122.2, 122.4, 124.5, 130.0, 132.4, 136.8, 141.0, 145.4, 149.4, 149.5, 157.2; EI-MS (m/z): 427 for [M] + ; EI-HRMS: m/z calcd for C30H25N3 : 427.2048; found: 427.2055. Synthesis of N,N-bis(p-chlorophenyl)-3,5-bis(pyrid-2-yl)aniline (3): By using the same procedure for the synthesis of 2, ligand 3 was prepared from 3,5-bis(pyrid-2-yl)bromobenzene and bis(pchlorophenyl)amine in 50 % yield as a white solid. 1H NMR (400 MHz, CD3COCD3): d = 7.14 (d, J = 8.8 Hz 4 H), 7.30 (t, J = 5.8 Hz, 2 H), 7.32 (d, J = 8.8 Hz, 4 H), 7.83 (t, J = 7.7 Hz, 2 H), 7.93 (s, 2 H), 7.94 (d, J = 9.2 Hz, 2 H), 8.53 (s, 1 H), 8.61 (d, J = 4.4 Hz, 2 H); 13 C NMR (100 MHz, CD3COCD3): d = 119.7, 119.8, 122.3, 122.8, 124.9, 127.2, 129.1, 136.5, 140.8, 145.9, 147.7, 149.1, 155.4; EI-MS: m/z = 467 [M] + ; EI-HRMS: m/z calcd for C28H19Cl2N3 : 467.0956; found: 467.0962. General procedure for the synthesis of ruthenium complexes 4(PF6), 6(PF6)–12(PF6), and 15(PF6): [(Mebip)RuCl3], [(tpy)RuCl3], or [(Me3tctpy)RuCl3] (0.10 mmol) and AgOTf (0.33 mmol) were added to 20 mL of dry acetone. The mixture was heated to reflux for 3 h. One hour later, the mixture was filtered to remove the AgCl precipitate. The filtrate was concentrated to dryness. Ligand 1, 2, 3, or 1,3-di(pyrid-2-yl)benzene (0.11 mmol), DMF (10 mL), and tBuOH (10 mL) were added to the residue. The resulting mixture was heated to reflux under microwave heating (power = 375 W) for 30 min. After cooling to room temperature, the solvent was removed under reduced pressure. The residue was dissolved in 2 mL of methanol, and an excess of aqueous KPF6 was added. The resulting precipitate was collected by filtration and washed with water and Et2O. The crude solid was purified by chromatography on silica gel (eluent: CH3CN/H2O/aq. KNO3 300/30/0.1), followed by anion exchange with KPF6 to give the desired complexes as black solids in 34–65 % yield. Some of these complexes are easily oxidized under ambient conditions. To obtain satisfactory 1H NMR signals, a small amount of aqueous hydrazine was added to the solution in [D6]acetone to remove the oxidized impurities (see the NMR spectra in the Supporting Information). Complex 4(PF6): Prepared from [(Mebip)RuCl3] and ligand 1 in 60 % yield. 1H NMR (400 MHz, CD3COCD3): d = 3.76 (s, 6 H), 4.61 (s, 6 H), 6.38 (d, J = 8.4 Hz, 2 H), 6.58 (t, J = 6.5 Hz, 2 H), 7.01 (m, 6 H), 7.16 (d, J = 4.8 Hz, 2 H), 7.30 (t, J = 7.6 Hz, 2 H), 7.34 (d, J = 9.2 Hz, 4 H), 7.46 (t, J = 7.8 Hz, 2 H), 7.60 (d, J = 8.4 Hz, 2 H), 8.04 (d, J = 8.0 Hz, 2 H), 8.27 (s, 2 H), 8.38 (t, J = 8.2 Hz, 1 H), 9.06 (d, J = 8.0 Hz, 2 H); MALDI-MS: m/z = 899.0 [MPF6] + ; elemental analysis (%) calcd for C51H41F6N8O2PRu·2 H2O: C 56.72, H 4.20, N 10.38; found: C 56.63, H 4.03, N 10.37. Complex 6(PF6): Prepared from [(Me3tctpy)RuCl3] and ligand 1 in 65 % yield. 1H NMR (400 MHz, CD3COCD3): d = 3.84 (s, 6 H), 3.94 (s, 6 H), 4.19 (s, 3 H), 6.68 (m, 2 H), 6.95–6.97 (m, 4 H), 7.07 (d, J = 5.2 Hz, 2 H), 7.24–7.25 (m, 4 H), 7.64–7.72 (m, 6 H), 8.16 (m, 4 H), 9.25 (s, 2 H), 9.65 (s, 2 H); MALDI-MS: m/z = 966.9 [MPF6] + ; elemental analysis (%) calcd for C51H41F6N6O8PRu·H2O: C 54.21, H 3.84, N 7.44; found: C 54.50, H 3.96, N 7.84. Complex 7(PF6): Prepared from [(Mebip)RuCl3] and ligand 2 in 34 % yield. 1H NMR (400 MHz, CD3COCD3): d = 2.35 (s, 6 H), 4.61 (s, 6 H), 6.39 (d, J = 8.0 Hz, 2 H), 6.59 (t, J = 6.4 Hz, 2 H), 7.00 (t, J = 7.8 Hz, 2 H), 7.17 (d, J = 5.2 Hz, 2 H), 7.23 (d, J = 8.4 Hz, 4 H), 7.31 (m, 6 H), 7.47 (t, J = 7.7 Hz, 2 H), 7.61 (d, J = 8.4 Hz, 2 H), 8.07 (d, J = 8.0 Hz, 2 H), 8.29 (s, 2 H), 8.39 (t, J = 8.2 Hz, 1 H), 9.07 (d, J = 8.0 Hz, 2 H); MALDI-MS: 867.2 [MPF6] + ; elemental analysis (%) calcd for C51H41F6N8PRu: C 60.53, H 4.08, N 11.07; found: C 60.54, H 4.24, N 11.02.

17475

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper Complex 8(PF6): Prepared from [(tpy)RuCl3] and ligand 2 in 55 % yield. 1H NMR (400 MHz, CD3COCD3): d = 2.32 (s, 6 H), 6.74 (t, J = 6.4 Hz, 2 H), 7.15 (m, 12 H), 7.34 (d, J = 5.2 Hz, 2 H), 7.64 (t, J = 7.8 Hz, 2 H), 7.84 (t, J = 7.8 Hz, 2 H), 8.17 (d, J = 6.4 Hz, 2 H), 8.18 (s, 2 H), 8.39 (t, J = 8.0 Hz, 1 H), 8.69 (d, J = 8.0 Hz, 2 H), 9.02 (d, J = 8.0 Hz, 2 H); MALDI-MS: m/z = 761.2 [MPF6] + ; elemental analysis (%) calcd for C45H35F6N6PRu·H2O: C 58.50, H 4.04, N 9.10; found: C 58.75, H 3.92, N 8.90. Complex 9(PF6): Prepared from [(Me3tctpy)RuCl3] and ligand 2 in 52 % yield. 1H NMR (400 MHz, CD3COCD3): d = 2.34 (s, 6 H), 3.94 (s, 6 H), 4.19 (s, 3 H), 6.69 (t, J = 6.6 Hz, 2 H), 7.09 (d, J = 5.2 Hz, 2 H), 7.16–7.22 (m, 8 H), 7.63–7.72 (m, 6 H), 8.19 (d, J = 8.0 Hz, 2 H), 8.21 (s, 2 H), 9.25 (s, 2 H), 9.66 (s, 2 H); MALDI-MS: m/z = 935.2 [MPF6] + ; elemental analysis (%) calcd for C51H41F6N6O6PRu: C 56.72, H 3.83, N 7.78; found: C 56.48, H 4.01, N 7.82. Complex 10(PF6): Prepared from [(Mebip)RuCl3] and ligand 3 in 45 % yield. 1H NMR (400 MHz, CD3COCD3): d = 4.61 (s, 6 H), 6.34 (d, J = 8.4 Hz, 2 H), 6.61 (t, J = 6.5 Hz, 2 H), 7.02 (t, J = 7.4 Hz, 2 H), 7.20 (d, J = 5.2 Hz, 2 H), 7.30 (t, J = 7.4 Hz, 2 H), 7.46 (s, 7 H), 7.48 (t, J = 8.0 Hz, 2 H), 7.61 (d, J = 8.4 Hz, 2 H), 8.13 (d, J = 8.4 Hz, 2 H), 8.36 (s, 2 H), 8.41 (t, J = 8.2 Hz, 1 H), 9.07 (d, J = 8.4 Hz, 2 H); MALDI-MS: m/ z = 907.0 [MPF6] + ; elemental analysis (%) calcd for C49H35Cl2F6N8PRu: C 55.90, H 3.35, N 10.64; found: C 55.70, H 3.52, N 10.23. Complex 11(PF6): Prepared from [(tpy)RuCl3] and ligand 3 in 63 % yield. 1H NMR (400 MHz, CD3COCD3): d = 6.77 (t, J = 6.2 Hz, 2 H), 7.11–7.17 (m, 4 H), 7.22 (d, J = 5.4 Hz, 2 H), 7.30–7.37 (m, 8 H), 7.66 (t, J = 7.7 Hz, 2 H), 7.84 (t, J = 7.8 Hz, 2 H), 8.18 (s, 1 H), 8.25 (d, J = 9.6 Hz, 2 H), 8.26 (s, 1 H), 8.40 (t, J = 8.0 Hz, 1 H), 8.70 (d, J = 8.0 Hz, 2 H), 9.03 (d, J = 8.4 Hz, 2 H); MALDI-MS: m/z = 801.1 [MPF6] + ; elemental analysis (%) calcd for C43H29Cl2F6N6PRu: C 54.56, H 3.09, N 8.88; found: C 54.60, H 3.33, N 8.91. Complex 12(PF6): Prepared from [(Me3tctpy)RuCl3] and ligand 3 in 42 % yield. 1H NMR (400 MHz, CD3COCD3): d = 3.94 (s, 6 H), 4.20 (s, 3 H), 6.72 (t, J = 6.6 Hz, 2 H), 7.13 (d, J = 4.8 Hz, 2 H), 7.33–7.39 (m, 8 H), 7.65–7.71 (m, 6 H), 8.28 (d, J = 8.0 Hz, 2 H), 8.31 (s, 2 H), 9.26 (s, 2 H), 9.66 (s, 2 H); MALDI-MS: m/z = 975.2 [MPF6] + ; elemental analysis (%) calcd for C49H35Cl2F6N6O6PRu·2H2O: C 50.87, H 3.40, N 7.26; found: C 50.52, H 3.52, N 7.16. Complex 15(PF6): Prepared from [(Me3tctpy)RuCl3] and 1,3-bis(pyrid-2-yl)benzene in 40 % yield. 1H NMR (400 MHz, CD3CN): d = 3.89 (s, 6 H), 4.18 (s, 3 H), 6.59 (t, J = 6.6 Hz, 2 H), 6.82 (d, J = 5.6 Hz, 2 H), 7.42 (d, J = 6.0 Hz, 2 H), 7.48 (d, J = 6.0 Hz, 2 H), 7.53 (t, J = 7.6 Hz, 1 H), 7.61 (t, J = 7.7 Hz, 2 H), 8.14 (d, J = 8.0 Hz, 2 H), 8.28 (d, J = 7.6 Hz, 2 H), 9.01 (s, 2 H), 9.41 (s, 2 H)); MALDI-MS: m/z = 740.0 [MPF6] + ; elemental analysis (%) calcd for C37H28F6N5O6PRu: C 50.23, H 3.19, N 7.92; found: C 49.86, H 3.53, N 7.43.

Acknowledgements We thank the National Natural Science Foundation of China (grants 21271176, 91227104, 21472196, and 21221002), the National Basic Research 973 Program of China (grant 2011CB932301), and the Strategic Priority Research Program of the Chinese Academy of Sciences (grant XDB 12010400) for funding support. Keywords: electrochemistry · charge transfer · mixed-valent compounds · ruthenium · tridentate ligands

Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

[1] a) D. M. D’Alessandro, F. R. Keene, Chem. Rev. 2006, 106, 2270; b) W. Kaim, G. K. Lahiri, Angew. Chem. Int. Ed. 2007, 46, 1778; Angew. Chem. 2007, 119, 1808; c) P. Aguirre-Etcheverry, D. O’Hare, Chem. Rev. 2010, 110, 4839; d) P. J. Low, N. J. Brown, J. Cluster Sci. 2010, 21, 235; e) M. H. Chisholm, B. J. Lear, Chem. Soc. Rev. 2011, 40, 5254; f) C. P. Kubiak, Inorg. Chem. 2013, 52, 5663; g) S. E. Nelsen, Chem. Eur. J. 2000, 6, 581; h) S. V. Rosokha, J. K. Kochi, Acc. Chem. Res. 2008, 41, 641; i) J. Hankache, O. S. Wenger, Chem. Rev. 2011, 111, 5138; j) S. Heckmann, C. Lambert, Angew. Chem. Int. Ed. 2012, 51, 326; Angew. Chem. 2012, 124, 334; k) A. Hildebrandt, H. Lang, Organometallics 2013, 32, 5640; l) G. E. Pieslinger, P. Albores, L. D. Slep, L. M. Baraldo, Angew. Chem. Int. Ed. 2014, 53, 1293; Angew. Chem. 2014, 126, 1317. [2] C. Creutz, H. Taube, J. Am. Chem. Soc. 1969, 91, 3988. [3] a) N. S. Hush, Prog. Inorg. Chem. 1967, 8, 391; b) N. S. Hush, Coord. Chem. Rev. 1985, 64, 135; c) N. S. Hush, Electrochim. Acta 1968, 13, 1005. [4] M. B. Robin, P. Day, Adv. Inorg. Chem. Radiochem. 1967, 9, 247. [5] B. S. Brunschwig, C. Creutz, N. Sutin, Chem. Soc. Rev. 2002, 31, 168. [6] a) M. D. Ward, J. Solid State Electrochem. 2005, 9, 778; b) W. Kaim, Coord. Chem. Rev. 2011, 255, 2503; c) C.-J. Yao, Y.-W. Zhong, H.-J. Nie, H. D. AbruÇa, J. Yao, J. Am. Chem. Soc. 2011, 133, 20720; d) S.-H. Hsiao, H.-M. Wang, S.-H. Liao, Polym. Chem. 2014, 5, 2473; e) L. Ma, H. Nie, J. Cai, P, Zhao, C. Wang, Y. Lian, X. Bai, W. Wang, J. Mater. Chem. C 2014, 2, 2272; f) C.-J. Yao, Y.-W. Zhong, J. Yao, Inorg. Chem. 2013, 52, 10000; g) N. Deibel, S. Hohloch, M. G. Sommer, D. Schweinfurth, F. Ehret, P. Braunstein, B. Sarkar, Organometallics 2013, 32, 7366; h) N. Deibel, M. G. Sommer, S. Hohloch, J. Schwann, D. Schweinfurth, F. Ehret, B. Sarkar, Organometallics 2014, 33, 4756; i) H. S. Das, D. Schweinfurth, J. Fiedler, M. M. Khusniyarov, S. M. Mobin, B. Sakar, Chem. Eur. J. 2014, 20, 4334. [7] a) C. Sporer, I. Ratera, D. Ruiz-Molina, Y. Zhao, J. Vidal-Gancedo, K. Wurst, P. Jaitner, K. Clays, A. Persoons, C. Rovira, J. Veciana, Angew. Chem. Int. Ed. 2004, 43, 5266; Angew. Chem. 2004, 116, 5378; b) J. C. Barnes, A. C. Fahrenbach, S. M. Dyar, M. Frasconi, M. A. Giesener, Z. Zhu, Z. Liu, K. J. Hartlieb, R. Carmieli, M. R. Wasielewski, J. F. Stoddart, Proc. Natl. Acad. Sci. USA 2012, 109, 11546. [8] a) I. P.-C. Liu, W.-Z. Wang, S.-M. Peng, Chem. Commun. 2009, 4323; b) R. Sakamoto, S. Katagiri, H. Maeda, H. Nishihara, Coord. Chem. Rev. 2013, 257, 1493; c) J.-W. Ying, I. P.-C. Liu, B. Xi, Y. Song, C. Campana, J.-L. Zuo, T. Ren, Angew. Chem. Int. Ed. 2010, 49, 954; Angew. Chem. 2010, 122, 966; d) J.-P. Launay, Coord. Chem. Rev. 2013, 257, 1544. [9] C. Simao, M. Mas-Torrent, J. Casado-Montenegro, F. Oton, J. Veciana, C. Rovira, J. Am. Chem. Soc. 2011, 133, 13256. [10] K. Kamada, S.-i. Fuku-en, S. Minamide, K. Ohta, R. Kishi, M. Nakano, H. Matsuzaki, H. Okamoto, H. Higashikawa, K. Inoue, S. Kojima, Y. Yamamoto, J. Am. Chem. Soc. 2013, 135, 232. [11] a) Y. Yokoyama, D. Sakanaki, A. Ito, K. Tanaka, M. Shiro, Angew. Chem. Int. Ed. 2012, 51, 9403; Angew. Chem. 2012, 124, 9537; b) D. Sakamaki, A. Ito, K. Furukawa, T. Kato, M. Shiro, K. Tanaka, Angew. Chem. Int. Ed. 2012, 51, 12776; Angew. Chem. 2012, 124, 12948; c) P. Bujak, I. KulszewiczBajer, M. Zagorska, V. Maurel, I. Wielgus, A. Pron, Chem. Soc. Rev. 2013, 42, 8895. [12] C.-J. Yao, R.-H. Zheng, Q. Shi, Y.-W. Zhong, J. Yao, Chem. Commun. 2012, 48, 5680. [13] B.-B. Cui, C.-J. Yao, J. Yao, Y.-W. Zhong, Chem. Sci. 2014, 5, 932. [14] a) C. Olivier, K. Costuas, S. Choua, V. Maurel, P. Turek, J.-Y. Saillard, D. Touchard, S. Rigaut, J. Am. Chem. Soc. 2010, 132, 5638; b) S. H. Wadman, R. W. A. Havenith, M. Lutz, A. L. Spek, G. P. M. van Klink, G. van Koten, J. Am. Chem. Soc. 2010, 132, 1914; c) C.-J. Yao, Y.-W. Zhong, J. Yao, J. Am. Chem. Soc. 2011, 133, 15697; d) B. Xi, I. P.-C. Liu, G.-L. Xu, M. M. R. Choudhuri, M. C. DeRosa, R. J. Crutchley, T. Ren, J. Am. Chem. Soc. 2011, 133, 15094; e) S. D. Glover, C. P. Kubiak, J. Am. Chem. Soc. 2011, 133, 8721; f) M. Linseis, S. Zalis, M. Zabel, R. F. Winter, J. Am. Chem. Soc. 2012, 134, 16671; g) Y.-P. Qu, J. Xia, J. Zhang, M. Xu, J. Yin, G.-A. Yu, S. H. Liu, Chem. Asian J. 2013, 8, 3085; h) K. Kobayashi, M. Ishikubo, K. Kanaizuka, K. Kosuge, S. Masaoka, K. Sakai, K. Nozaki, M.-a. Haga, Chem. Eur. J. 2011, 17, 6954; i) F. Weisser, R. Huebner, D. Schweinfurht, B. Sarkar, Chem. Eur. J. 2011, 17, 57277. [15] a) C. Lambert, G. Nçll, J. Am. Chem. Soc. 1999, 121, 8434; b) C. Lambert, G. Nçll, J. Schelter, Nat. Mater. 2002, 1, 69; c) S. Barlow, C. Risko, S.-J. Chung, N. M. Tucker, V. Coropceanu, S. C. Jones, Z. Levi, J.-L. Brdas, S. R. Marder, J. Am. Chem. Soc. 2005, 127, 16900; d) A. Heckmann, S. Amothor, C. Lambert, Chem. Commun. 2006, 2959; e) J. C. Lacroix, K. I.

17476

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

[16]

[17] [18] [19]

[20] [21] [22]

[23]

Chane-Ching, F. Maquere, F. Maurel, J. Am. Chem. Soc. 2006, 128, 7264; f) R. Sakamoto, T. Sasaki, N. Honda, T. Yamamura, Chem. Commun. 2009, 5156; g) S. F. Vçlker, M. Renz, M. Kaupp, C. Lambert, Chem. Eur. J. 2011, 17, 14147; h) B. He, O. S. Wenger, J. Am. Chem. Soc. 2011, 133, 17027; i) S. Barlow, C. Risko, S. A. Odom, S. Zheng, V. Coropceanu, L. Beverina, J.-L. Bredas, S. R. Marder, J. Am. Chem. Soc. 2012, 134, 10146; j) S. C. Jones, V. Coropceanu, S. Barlow, T. Kinnibrugh, T. Timofeeva, J.-L. Brdas, S. R. Marder, J. Am. Chem. Soc. 2004, 126, 11782; k) C.-J. Yao, R.-H. Zheng, H.-J. Nie, B.-B. Cui, Q. Shi, J. Yao, Y.-W. Zhong, Chem. Eur. J. 2013, 19, 12376; l) M. Parthey, K. B. Vincent, M. Renz, P. A. Schauer, D. S. Yuft, J. A. K. Howard, M. Kaupp, P. J. Low, Inorg. Chem. 2014, 53, 1544; m) H.J. Nie, X. Chen, C.-J. Yao, Y.-W. Zhong, G. R. Hutchison, J. Yao, Chem. Eur. J. 2012, 18, 14497. a) A. Ceccon, S. Santi, L. Orian, A. Bisello, Coord. Chem. Rev. 2004, 248, 683; b) A. Heckmann, C. Lambert, J. Am. Chem. Soc. 2007, 129, 5515; c) I. Ratera, C. Sporer, D. Ruiz-Molina, N. Ventosa, J. Baggerman, A. M. Brouwer, C. Rovira, J. Veciana, J. Am. Chem. Soc. 2007, 129, 6117; d) K. Kowalski, M. Linseis, R. F. Winter, M. Zabel, S. ZliSˇ, H. Kelm, H.-J. Krger, B. Sarkar, W. Kaim, Organometallics 2009, 28, 4196; e) A. Das, T. M. Scherer, A. D. Chowdhury, S. M. Mobin, W. Kaim, G. K. Lahiri, Inorg. Chem. 2012, 51, 1675; f) J.-Y. Shao, Y.-W. Zhong, Chem. Eur. J. 2014, 20, 8702; g) S.-H. Wu, J.-Y. Shao, H.-W. Kang, J. Yao, Y.-W. Zhong, Chem. Asian J. 2013, 8, 2843; h) J. C. Chen, C.-H. Kao, S. J. Lin, C.-C. Tai, K. S. Kwan, Inorg. Chem. 2000, 39, 189; i) H. Vahrenkamp, A. Gei, G. N. Richardson, J. Chem. Soc. Dalton Trans. 1997, 3643; j) G. Barrado, G. A. Carriedo, C. Diaz-Valenzuela, V. Riera, Inorg. Chem. 1991, 30, 4416; k) S. Santi, L. Orian, C. Durante, A. Bisello, F. Benetollo, L. Crociani, P. Ganis, A. Ceccon, Chem. Eur. J. 2007, 13, 1955; l) F. Pevny, E. D. Piazza, L. Norel, M. Drescher, R. F. Winter, S. Rigaut, Organometallics 2010, 29, 5912; m) S. Bellinger-Buckley, T.-C. Chang, S. Bag, D. Schweinfurth, W. Zhou, B. Torok, B. Sarkar, M.-K. Tsai, J. Rochford, Inorg. Chem. 2014, 53, 5556. W.-W. Yang, Y.-W. Zhong, S. Yoshikawa, J.-Y. Shao, S. Masaoka, K. Sakai, J. Yao, M.-a. Haga, Inorg. Chem. 2012, 51, 890. K. C. D. Robson, B. Sporinova, B. D. Koivisto, E. Schott, D. G. Brown, C. P. Berlinguette, Inorg. Chem. 2011, 50, 6019. a) V. Duprez, J.-P. Launary, A. Gourdon, Inorg. Chim. Acta 2003, 343, 395; b) J.-P. Djukic, J.-B. Sortais, L. Barloy, M. Pfeffer, Eur. J. Inorg. Chem. 2009, 817. P. G. Bomben, K. C. D. Robson, P. A. Sedach, C. P. Berlinguette, Inorg. Chem. 2009, 48, 9631. J.-H. Tang, S.-H. Wu, J.-Y. Shao, H.-J. Nie, Y.-W. Zhong, Organometallics 2013, 32, 4564. a) D. E. Richardson, H. Taube, J. Am. Chem. Soc. 1983, 105, 40; b) D. E. Richardson, H. Taube, Coord. Chem. Rev. 1984, 60, 107; c) C. E. B. Evans, M. L. Naklicki, A. R. Rezvani, C. A. White, V. V. Kondratiev, R. J. Crutchley, J. Am. Chem. Soc. 1998, 120, 13096; d) R. J. Crutchley, Adv. Inorg. Chem. 1994, 41, 273. A. V. Szeghalmi, M. Erdmann, V. Engel, M. Schmitt, S. Amthor, V. Kriegisch, G. Nçll, R. Stahl, C. Lambert, D. Leusser, D. Stalke, M. Zabel, J. Popp, J. Am. Chem. Soc. 2004, 126, 7834.

Chem. Eur. J. 2014, 20, 17466 – 17477

www.chemeurj.org

[24] S. Barlow, C. Risko, V. Coropceanu, N. M. Tucker, S. C. Jones, Z. Levi, V. N. Khrustalev, M. Y. Antipin, T. L. Kinniburgh, T. Timofeeva, S. R. Marder, J.-L. Brdas, Chem. Commun. 2005, 764. [25] a) N. Dowling, P. M. Henry, N. A. Lewis, H. Taube, Inorg. Chem. 1981, 20, 2345; b) N. Dowling, P. M. Henry, Inorg. Chem. 1982, 21, 4088. [26] S. Patra, B. Sarkar, S. M. Mobin, W. Kaim, G. K. Lahiri, Inorg. Chem. 2003, 42, 6469. [27] a) M. Renz, K. Theilacker, C. Lambert, M. Kaupp, J. Am. Chem. Soc. 2009, 131, 16292; b) M. Kaupp, M. Renz, M. Parthey, M. Stolte, F. Wrthner, C. Lambert, Phys. Chem. Chem. Phys. 2011, 13, 16973. [28] H.-J. Nie, J.-Y. Shao, C.-J. Yao, Y.-W. Zhong, Chem. Commun. 2014, 50, 10082. [29] B.-B. Cui, H.-J. Nie, C.-J. Yao, J.-Y. Shao, S.-H. Wu, Y.-W. Zhong, Dalton Trans. 2013, 42, 14125. [30] a) D. Yokogawa, H. Sato, Y. Nakao, S. Sakaki, Inorg. Chem. 2007, 46, 1966; b) L.-J. Li, F.-Q. Bai, H.-X. Zhang, H.-Y. Li, Sci. China Chem. 2013, 56, 641; c) Z.-G. Shuai, W. Xu, Q. Peng, H. Geng, Sci. China Chem. 2013, 56, 1277; d) S.-J. Liu, C.-J. Yang, W.-J. Xu, Y.-Q. Huang, S.-H. Ye, Q. Zhao, X.M. Liu, W. Huang, Sci. China Chem. 2013, 56, 137. [31] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112. [32] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann, J. Appl. Crystallogr. 2009, 42, 339. [33] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785. [34] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, T. Vreven Jr., K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 09, revision A.2; Gaussian, Inc.: Wallingford CT, 2009. [35] a) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270; b) W. R. Wadt, P. J. Hay, J. Chem. Phys. 1985, 82, 284; c) P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299. [36] T. H. Dunning, P. J. Hay, in Modern Theoretical Chemistry; Schaefer, H. F., Ed.; Plenum: New York 1976; Vol. 3, p. 1. [37] a) V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995; b) M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem. 2003, 24, 669. Received: July 23, 2014 Published online on November 4, 2014

17477

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim