摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2643–2650 2643 Cobalt-mediated selective C]H bond activation. Direct aromatic hydroxylation in the complexes [CoIII{o-OC6H3(R)N] NC5H4N}2]ClO4? H2O (R 5 H, o-Me/Cl, m-Me/Cl or p-Me/Cl). Synthesis, spectroscopic and redox properties Ananthanarayanan Bharath, Bidyut Kumar Santra, Pradip Munshi and Goutam Kumar Lahiri * Department of Chemistry, Indian Institute of Technology, Bombay, Powai, Mumbai-400 076, India The reactions of low-spin complexes [CoIIL3][ClO4]2?H2O 1 [L = 2-(arylazo)pyridine, (R)C6H4N]] NC5H4N (R = H, o-Me/Cl, m-Me/Cl or p-Me/Cl] with m-chloroperbenzoic acid (m-ClC6H4CO3H) in acetonitrile solvent at room temperature resulted in low-spin [CoIIIL92]ClO4?H2O 2 [L9 = o-OC6H3(R)N]] NC5H4N].In complexes 2 the o-carbon–hydrogen bond of the pendant phenyl ring of both parent ligands L has been selectively and spontaneously hydroxylated. During the transformation of 1 to 2 the metal ion is oxidised from the starting CoII to CoIII. The meridional configuration (cis-trans-cis with respect to the oxygen, azo and pyridine nitrogens respectively) of complexes 2 has been established by 1H and 13C NMR spectroscopy.When one methyl or chloro group was present at the meta position of the pendant phenyl ring of L the reaction resulted in two isomeric complexes due to free rotation of the singly bonded meta-substituted phenyl ring with respect to the azo group. In acetonitrile solvent, complexes 2 systematically display one d–d transition (1A1g æÆ 1T1g) near 850 nm, two metal to ligand charge-transfer transitions in the visible region and intraligand transitions in the UV region.In acetonitrile solution all complexes 2 exhibit irreversible CoIII æÆ CoIV oxidation near 2 V and reversible CoIII CoII reduction near 0.0 V versus Ag–AgCl. The ligand-based expected four azo (N]] N) reductions are observed sequentially for all the complexes at the negative side of the reference Ag–AgCl.Complexes 2 can be quantitatively and stereoretentively reduced to the low-spin cobalt(II) congeners, [CoIIL92] 22 electrochemically as well as chemically by using hydrazine hydrate. These complexes display eight-line EPR spectra in acetonitrile solution at 77 K. Complex 2a2 exhibits a ligand to metal charge-transfer transition at 534 nm and intraligand transition at 345 nm. Two possible d–d transitions, 2E æÆ 2T1 and 2E æÆ 2T2 are observed at 700 and 800 nm respectively.Metal ion-mediated activation of carbon–hydrogen bonds of organic molecules is a fundamentally important chemical process as this may lead to the formation of various interesting products which are otherwise diYcult or even impossible to synthesize by the conventional synthetic routes.1 In this process the presence of metal ion helps to create a suitable chemical platform for the interacting molecules which in turn facilitates the desired reactions.In this article we report one such reaction where the ortho-carbon–hydrogen bond of the pendant phenyl ring of a 2-(arylazo)pyridine ligand (R)C6H4N]] NC5H4N (L) in the cobalt complexes [CoIIL3]21 1 has been selectively and spontaneously hydroxylated in the presence of m-chloroperbenzoic acid at ambient temperature. The hydroxylation of an aromatic ring is known to be an important chemical process in chemistry and biology.2 Herein we describe the cobalt ion-mediated aromatic hydroxylation reaction in a group of seven complexes, spectroscopic characterisation of the final hydroxylated products 2, metal- and ligand-centred electroactivities and solution characterisation of very rare low-spin cobalt(II) octahedral complexes having CoIIN4O2 chromophores.Results and Discussion Synthesis Seven 2-(arylazo)pyridine ligands used for the present study are abbreviated as L1–L7. The reactions of the meridional tris complexes [CoIIL3][ClO4]2?H2O 1 with m-ClC6H4CO3H in a ratio of 1 : 4 in acetonitrile solvent at room temperature result in a bluish green solution immediately. Although the red-brown starting complexes 1 turn to bluish green immediately, stirring is continued to get the maximum yield.Chromatographic puri- fications of the crude green solutions on a silica gel column using chloroform–acetonitrile (2 : 1) as eluent followed by removal of solvents under reduced pressure aVord pure bluish green compounds in the solid state having the composition [CoIII{o-OC6H3(R)N]] NC5H4N}2]ClO4?H2O 2 (Scheme 1).All the starting complexes 1a–1g behave similarly in the presence of perbenzoic acid. In the course of the reaction (Scheme 1) one ligand L is liberated from the co-ordination sphere of the starting tris complexes 1 and the o-carbon atom of the pendant phenyl ring of each of the remaining two ligands L is selectively and directly hydroxylated. The liberated ligand L has been recovered quantitatively by column chromatography.In view of the selective hydroxylation of the phenyl ring, the usual bidentate (N,N) parent ligand L has been transformed into a tridentate (O,N,N) ligand L9 and two such ligands are bound to the metal ion in a meridional fashion. During this carbon–hydrogen bond activation process the cobalt ion is oxidised from its starting bivalent state in 1 to the trivalent state in 2. Since the reactions take place particularly under atmospheric conditions, the oxygen in air may therefore be responsible for the oxidation of the metal ion (see later).In the case of starting complexes 1d and 1e where one methyl group and chloride group respectively are present at the meta position of the active phenyl ring of each ligand, free rotation along the Nazo]Cphenyl bond may lead to the formation of three possible isomers 3–5. However, an intimate mixture of isomers 3 and 4 has been detected in solution for both cases. All attempts to separate the individual isomers 3 and 4 by chromatography have failed. Since in the absence of hydroxylating agent (perbenzoic acid)2644 J. Chem.Soc., Dalton Trans., 1998, Pages 2643–2650 Scheme 1 (i) m-ClC6H4CO3H, MeCN, stirring Np N Na R Na Na Np Na CoII Np Np R R R 2 3 4 6 1 11 10 5 7 8 9 [ClO4]2•H2O Np O Na Na CoIII O Np R R ClO4•H2O L1 L2 L3 L4 L5 L6 L7 R = H R = 7-Me R = 7-Cl R = 8-Me R = 8-Cl R = 9-Me R = 9-Cl 1a 1b 1c 1d 1e 1f 1g 2a 2b 2c 2d 2e 2f 2g (i) Table 1 Microanalytical,a conductivity b and electronic spectral b data Analysis (%) LM/ W21 IR/cm21 Compound 2a C 46.26 (46.12) H 3.30 (3.14) N 14.54 (14.68) cm2 mol21 162 l/nm (e/dm3 mol21 cm21) 858 (170), 667 c (6610), 612 (9330), 367 (24 130), 315 (8710) n(N]] N) 1400 n(ClO4 2) 1140, 624 2b 47.86 (47.96) 3.62 (3.66) 14.10 (13.99) 156 846 (81), 673 c (2600), 620 (3410), 350 (12 530), 305 (6460) 1397 1074, 627 2c 41.02 (41.16) 2.62 (2.49) 12.97 (13.09) 154 875 (30), 673 c (2860), 614 (3860), 350 (14 520), 315 (4860) 1381 1141, 623 2d 48.13 (47.96) 3.59 (3.66) 13.85 (13.99) 160 875 (300), 686 c (4680), 631 (6350), 392 (17 300), 301 (12 940) 1391 1095, 627 2e 41.28 (41.16) 2.59 (2.49) 12.88 (13.09) 166 878 (40), 670 c (3370), 619 (4650), 387 (11 730), 299 (8840) 1401 1091, 630 2f 48.17 (47.96) 3.69 (3.66) 13.83 (13.99) 168 862 (350), 656 c (6350), 603 (8340), 396 (21 840), 314 (10 930) 1380 1086, 620 2g 41.02 (41.16) 2.55 (2.49) 13.26 (13.09) 162 858 (30), 615 c (6350), 593 (8960), 396 (24 600), 314 (7730) 1400 1090, 621 a Calculated values in parentheses.b In acetonitrile solution at 298 K. c Shoulder. the starting complexes 1 remain unchanged in acetonitrile medium, the possibility of the formation of any solventdependent reactive intermediate prior to the activation process may be ruled out. Under identical reaction conditions but in the absence of starting complexes 1 the free L fail to undergo the transformation NC5H4N]] NC6H4(R)H4 æÆ NC5H4N]] NC6H3- (R)OH which indicates the direct involvement of the metal ion in the activation process.The microanalytical data of the products 2 (Table 1) are in Na O Me Na O Me Co Np Np Na O Na O Co Np Np Na O Me Na O Co Np Np Me Me Me 3 4 5 good agreement with the calculated values and thus confirm the composition. Solid state magnetic moment measurements at 298 K establish that the complexes are uniformly diamagnetic (low-spin CoIII, t2g 6, S = 0).In acetonitrile the complexes show 1 : 1 conductivity (Table 1). All the complexes 2 are highly soluble in polar solvents such as acetonitrile, dimethylformamide, dimethyl sulfoxide and moderately soluble in non-polar solvents like chloroform, dichloromethane, tetrahydrofuran and benzene. Infrared spectra The Fourier transform IR spectra of the complexes 2 were recorded as KBr discs in the region 4000–400 cm21. Selected frequencies are listed in Table 1. Two strong bands near 1580 and 1590 cm21 are assigned to n(C]] C) and n(C]] N) stretching frequencies respectively and the n(N]] N) stretching frequency of the ligands is observed in between 1400 and 1380 cm21.The n(N]] N) of free L appears near 1425 cm21; this lowering in azo frequency in complexes 2 is attributed to the presence of dp(CoIII) æÆ p*(L9) back bonding where p*(L9) is primarily dominated by the azo function.3 A strong and broad vibration near 1100 cm21 and a strong and sharp band near 620 cm21 are observed for all the complexes due to the presence of ionic perchlorate.A broad band due to water of crystallisation occurs at 3400 cm21.J. Chem. Soc., Dalton Trans., 1998, Pages 2643–2650 2645 Table 2 Proton and 13C NMR spectral data 1H, d(J/Hz) a Compound 2a 2b 2c 2f 2g H(1) 8.16 (5.9) b 8.14 (6.8) b 8.22 (6.5) b 8.20 (6.3) b 8.18 (9.3) b H(2) 6.90 (8.1, 7.8) c 7.18 (7.4, 7.6) c 7.22 (8.3, 9.6) c 7.35 (6.7, 6.3) c 7.40 (6.2, 6.7) c H(3) 8.10 (6.7, 7.3) c 8.06 (7.4, 7.7) c 8.17 (8.1, 7.7) c 8.06 (6.7, 6.1) c 8.13 (9.3, 9.8) c H(4) 8.07 (8.1) b 7.62 (6.0) b 7.72 (7.2) b 8.02 (6.9) b 8.05 (8.3) b H(8) 6.76 (8.9) b 6.63 (7.4) b 6.70 (8.8) b 6.55 d — 6.80 d — H(9) 7.38 (6.5, 7.2) c 7.38 (6.5, 5.9) c 7.53 (6.1, 5.7) c 2.25 (Me) —— H(10) 7.40 (5.7, 6.4) c 6.60 (10.1) b 6.91 (7.38) b 6.74 (9.2) b 6.94 (7.7) b H(11) 7.61 (5.9) b 2.88 (Me) —— 7.56 (5.8) b 7.54 (9.3) b 13C, d(J/Hz) e 2a C(1) 150.54 (730) b C(2) 120.79 (600) b C(3) 144.38 (690) b C(4) 141.89 (630) b C(5) 142.28 d — C(6) 180.88 d — C(7) 168.36 d — C(8) 118.43 (660) b C(9) 122.99 (660) b C(10) 125.27 (670) b C(11) 129.05 (690) b a Tetramethylsilane as internal standard.b Doublet. c Triplet. d Singlet. e In (CD3)2SO. 1H NMR spectra The complexes 2a–2c, 2f and 2g display well resolved 1H NMR spectra in CDCl3 solvent. Chemical shift and spin–spin splitting (among nearest neighbouring protons) values are depicted in Fig. 1 Proton NMR spectra in CDCl3 of (a) [CoIII{o-OC6H3(Me-o)- N]] NC5H4N}2]ClO4?H2O 2b and (b) methyl peaks for [CoIII{o-OC6- H3(Me-m)N]] NC5H4N}2]ClO4?H2O 2d Table 2.A representative spectrum is shown in Fig. 1(a). The individual proton resonances are assigned (Table 2) on the basis of their relative intensities, spin–spin structure and substituent induced change in splitting pattern.4 The complex 2a displays four doublets and four triplets having equal intensities. A direct comparison of the spectrum of complex 2a with that of free L1 reveals the absence of a H(7) signal for the transformed ligand L19 present in 2a.Similarly the NMR spectra for each of the other substituted complexes (2b,2c and 2f,2g) display seven signals having equal intensities indicating the absence of H(7). This absence therefore unambiguously supports the cobaltmediated activation of the ortho C]H(7) bond of the pendant phenyl ring of ligand L9 in 2. Since the 1H NMR spectra of complex 2a and the other substituted complexes (2b, 2c, 2f, 2g) display eight and seven proton signals respectively corresponding to one hydroxylated ligand L9 present in the complexes 2, it can therefore be inferred that each half of the molecule 2 is equivalent due to localised mirror symmetry around the cobalt centre.Thus 1H NMR spectral data imply a meridional geometry for 2 having a cistrans- cis configuration with respect to oxygen, azo and pyridine nitrogens respectively (Scheme 1). One methyl peak has been observed for each of the complexes 2b and 2f at d 2.88 and 2.25 respectively as imposed symmetry makes the two ligands in 2 equivalent.In the case of complex 2d the aromatic region of the spectrum is complicated due to the presence of two isomers 3 and 4 in solution, however the well resolved upfield methyl signals and direct comparison of the individual methyl intensities with those of the respective aromatic protons enabled us to reach reasonable conclusions.From the symmetry point of view one methyl signal is expected for each of the isomers 3 and 4 and two equally intense peaks for isomer 5. As the spectrum of 2d exhibits two unequally intense methyl peaks at d 2.39 and 2.41 respectively [Fig. 1(b)] having intensity ratio 3 : 4, isomers 3 and 4 are therefore predominant in solution. The downfield portion of the spectrum is overcrowded due to the partial overlapping of the aromatic protons of isomers 3 and 4 (not shown) which has precluded unequivocal assignment of the signals.In the case of complex 2e, the aromatic region of the spectrum is also complicated, like the spectrum of 2d, due to the presence of a mixture of isomers. Since there is no isolated upfield signal (methyl of 2d) in case of 2e, it is diYcult to make any conclusive statement about the nature of the isomeric mixture present. However, the aromatic proton counts match well with those of complex 2d. In view of this we assume that in solution similar isomers 3 and 4 are also present in 2e.2646 J.Chem. Soc., Dalton Trans., 1998, Pages 2643–2650 13C NMR spectra The 13C NMR spectrum of one representative complex (2a) was recorded in (CD3)2SO solvent. Both the decoupled and coupled spectra are shown in Fig. 2. The chemical shift and coupling constant values are listed in Table 2. The decoupled spectrum [Fig. 2(a)] displays eleven distinct peaks corresponding to one ligand (L19) which provides further support in favour of the presence of mirror symmetry in the complex.The corresponding coupled spectrum [Fig. 2(b)] exhibits three singlets and eight doublets. The parent azopyridine ligand L1 should exhibit two singlets [C(5) and C(6)] and nine doublets. Thus on going from ligand L1 in the starting complex 1a to ligand L19 in the final transformed complex 2a one doublet [C(7)] has been changed to a singlet. This change in pattern of the C(7) signal further unequivocally establishes the activation of the C(7)]H bond of the phenyl ring.The individual carbon resonances were assigned on the basis of their electronic environments, like the proton resonances. FAB Mass spectrum The FAB mass spectrum of one representative complex (2a) was recorded. The maximum peak is observed at m/z 455 which corresponds to the molecular ion [CoIII(NC5H4N]] NC6H4O)2]1 (calculated molecular weight 455). Thus the FAB mass, 1H and 13C NMR spectroscopic results along with the microanalytical, conductivity, magnetic moment and IR data collectively establish the composition and stereochemistry of the complexes 2.Electronic spectra Electronic spectra of the complexes 2 were recorded in acetonitrile solvent in the region 1100–200 nm. The spectral data are listed in Table 1 and a representative spectrum is shown in Fig. 3. In the visible region the complexes exhibit one intense band near 600 nm with a shoulder to lower energy near 670 nm Fig. 2 Carbon-13 NMR spectra of complex [CoIII(o-OC6H4N]] NC5- H4N)2]ClO4?H2O 2a: (a) decoupled, (b) coupled Fig. 3 Electronic spectrum of complex 2a in acetonitrile solvent. The inset shows the low-energy d–d transition (Table 1). On the basis of their high intensities the two bands are believed to be charge transfer in nature. Since CoIII in the complexes is in the low-spin t2g 6 configuration, the bands near 600 and 670 nm may be due to dp(CoIII) to ligand LUMO and LUMO 1 1 metal to ligand charge-transfer transitions respectively, where LUMO and LUMO 1 1 (LUMO = lowest unoccupied molecular orbital) are believed to be primarily dominated by the azo function and the pyridine part of the ligand respectively.5 The other two bands in the UV region (Table 1) are presumably due to intraligand p æÆ p* and n æÆ p* transitions respectively. Here the transitions are sensitive to the nature of the substituents present in the ligand framework6 (Table 1).In the lower energy part of the visible region (ª850 nm) all the complexes systematically display one weak transition (Table 1).Based on the low intensity of this lower energy band it is considered to be due to one of the possible d–d transitions.7 Low-spin cobalt(III) complexes are expected to exhibit two spinallowed transitions at relatively lower energies, 1A1g æÆ 1T1g and 1A1g æÆ 1T2g. The observed band near 850 nm is therefore assigned to the 1A1g æÆ 1T1g transition. The other higher energy transition, 1A1g æÆ 1T2g, has not been detected and is possibly masked by the nearby intense charge-transfer transitions.Electron-transfer properties The electron-transfer properties of the complexes 2 have been studied in acetonitrile solution by cyclic voltammetry (CV) using a platinum working electrode. The complexes are electroactive with respect to the metal as well as ligand centres and display five redox processes in the potential range ±2.5 V versus Ag–AgCl electrode (tetraethylammonium perchlorate as electrolyte at 298 K).The reduction potentials are listed in Table 3 and representative voltammograms are shown in Fig. 4. The assignments of the responses to the specific couples I–V in Table 3 are made on the basis of the following considerations. Cobalt(III)–cobalt(II) couple. In acetonitrile solvent the complexes display one reversible reduction couple (Fig. 4, couple II), Eo 298 in the region 20.1 to 0.1 V versus Ag–AgCl reference electrode with characteristic cathodic (Epc) and anodic (Epa) peak potentials.This reversible reduction process is assigned to the cobalt(III)–cobalt(II) couple, equation (1). The one- [CoIIIL92]1 1 e2 [CoIIL92] (1) electron nature of the couple is confirmed by constant-potential coulometry (see later). The presence of the bivalent paramagnetic low-spin cobalt(II) congener (22) in the reduced solution Fig. 4 Cyclic voltammogram (——) and diVerential pulse voltammogram (– – –) (scan rate 50 mV s21) of ª1023 mol dm23 solutions of complex 2a in acetonitrileJ.Chem. Soc., Dalton Trans., 1998, Pages 2643–2650 2647 Table 3 Electrochemical data for complexes 2 and EPR data for 22 Electrochemical data a CoIII]CoIV Couple I CoIII–CoII Couple II Ligand reductions, Eo 298/V(DEp/mV) EPR data b Compound 2a 2b 2c 2d 2e 2f 2g Epa c/V 1.82 1.75 1.97 1.78 1.95 1.80 1.93 Eo 298/V(DEp/mV) 20.005 (80) 20.130 (80) 0.050 (90) 20.090 (90) 0.120 (80) 20.050 (80) 0.080 (90) Couple III 20.62 (80) 20.65 (80) 20.48 (80) 20.66 (90) 20.49 (70) 20.63 (70) 20.52 (80) Couple IV 21.18 (90) 21.31 (90) 21.05 (100) 21.26 (90) 20.96 (90) 21.21 (90) 21.02 (100) Couple V 21.42 (120) 21.49 (110) 21.20 (120) 21.46 (130) 21.34 (120) 21.45 (110) 21.36 (120) g 1.910 2.006 2.005 2.005 2.004 1.996 2.020 A/G 40 30 29 31 26 25 24 a Conditions: solvent, acetonitrile; supporting electrolyte, [NEt4][ClO4]; reference electrode, Ag–AgCl; solute concentration, 1023 mol dm23; working electrode, platinum.Cyclic voltammetric data: scan rate, 50 mV s21; Eo 298 = 0.5 (Epc 1 Epa) where Epc and Epa are the cathodic and anodic peak potentials respectively. b In acetonitrile solution at 77 K (liquid nitrogen). c Epa is considered due to the irreversible nature of the voltammogram. is confirmed by an EPR study (see later). The formal potential of the couple [equation (1)] varies depending on the R group present in the ligand as expected (Table 3).Under identical experimental conditions the cobalt(III)–cobalt(II) potential for the starting complexes 1 appears in the region 0.9–1.2 V versus Ag–AgCl.8 Thus on moving from the complexes 1 to 2 the CoIII]CoII potential decreases by ª1 V. The azopyridine ligand L is known to stabilise low-valent metal complexes (12 in the case of Co) due to its strong p-acidic nature, and this is always reflected in high MII]MIII oxidation potentials.9 Hence the substantial lowering of the cobalt(III)–cobalt(II) reduction potential (ª1 V) in 2 as compared to the starting complexes 1 is primarily due to the combined eVects of (i) insertion of a s-donating phenolato oxygen atom in the selective ortho position of the pendant phenyl ring of L9 and (ii) removal of one strong p-acidic ligand L from the starting complexes 1.Ligand reduction. All the complexes 2 display three successive reversible reductions (couples III–V, Fig. 4) other than the cobalt(III)–cobalt(II) reduction (couple II).The one-electron stoichiometry of couples III and IV and the two-electron stoichiometry of couple V are established by comparison with the CoIII]CoII reduction (couple II) with the help of the cyclic voltammetric current height as well as diVerential pulse voltammetry (Fig. 4). Azopyridine ligand L is known to accept two electrons in the electrochemically accessible LUMO which is predominantly azo in character.3,6,10 Since two such electroactive azo groups are present in complexes 2, four successive one-electron azo reductions equations (2)–(5) are expected for [CoIIL92] 1 e2 [CoIIL9(L9~2)]2 (2) [CoIIL9(L9~2)]2 1 e2 [CoII(L9~2)2]22 (3) [CoII(L9~2)2]22 1 e2 [CoII(L9~2)(L92~2)]32 (4) [CoII(L9~2)(L92~2)]32 1 e2 [CoII(L92~2)2]42 (5) each complex.However, in practice two one-electron reductions corresponding to equations (2) and (3) appeared distinctly (couples III and IV) the other two reductions [equations (4) and (5)] being overlapped (couple V).The electrochemically generated reduced species are too unstable to isolate. Oxidation process. Complexes 2 display an irreversible oxidation process in the region 1.8–2.0 V versus Ag–AgCl (Fig. 4, couple I). Although the current height of the oxidation process (ipa) is ª1.5 times more than the previously observed other reversible processes (couples II–V), the one-electron nature of couple I is determined by direct comparison of its diVerential pulse voltammogram peak height with those of the previous one-electron processes (Fig. 4). This irreversible oxidation process could be either due to cobalt(III)–cobalt(IV) oxidation or oxidation of the co-ordinated ligand. The former assignment seems to be more reasonable as free L does not exhibit any oxidation process within the experimental potential limit (12.5 V). However, the possibility of the oxidation of co-ordinated L9 cannot be ruled out. Electrochemical and chemical reductions of complexes 2 Coulometric reductions of the complexes 2 (couple II) at 20.3 V versus Ag–AgCl in acetonitrile solvent using a platinumgauze working electrode under a dinitrogen atmosphere aVorded a violet solution.The observed Coulomb count corresponds to one-electron transfer for all the complexes (‘n’ = 1.05, 2a; 1.09, 2b; 0.97, 2c; 1.07, 2d; 1.09, 2e; 0.96, 2f; 1.11, 2g; n = Q/Q9, where Q9 is the calculated Coulomb count for a oneelectron transfer and Q is that after exhaustive electrolysis).The resulting violet reduced solutions (22) display voltammograms which are superposable on those of the corresponding parent complexes 2 which suggests the stereoretentive nature of the reduction process. The electrochemical reoxidations of the reduced violet solutions regenerate the corresponding bivalent complexes 2 quantitatively. Although the reduced cobalt(II) complexes can be generated by bulk electrolysis under a dinitrogen atmosphere, the reduced species 22 are unstable at room temperature under atmospheric conditions which has precluded their further isolation in the solid state. In order to confirm that the violet reduced solutions consist of cobalt(II) species as opposed to the reduced ligand, the X-band EPR spectra of the fresh solutions (produced coulometrically in acetonitrile solvent followed by quick freezing in liquid nitrogen at 77 K) of all the complexes were recorded.The spectra exhibit eight lines characteristic of hyperfine splitting by the 59Co nucleus (100% natural abundance, I = 7– 2).8,11 The acetonitrile solutions of complexes 2 can also be reduced chemically to the same violet cobalt(II) complexes by hydrazine hydrate.The reduced complexes 22 are unstable, however we have succeeded in recording their EPR spectra by freezing the reduced solutions immediately in liquid N2. The reduced cobalt(II) complexes 22 obtained either by electrochemical or chemical means exhibit identical EPR spectra.A representative spectrum is shown in Fig. 5. The centre field g values and the average hyperfine splitting (A) are given in Table 3. Although the reduced complexes 22 are unstable, we have managed to record the UV/VIS spectrum of one complex (2a2) generated by electrochemical means. The qualitative spectrum is shown in Fig. 6. It exhibits two intense bands at 534 and 345 nm. The UV band at 345 nm is associated with a shoulder at lower energy, 420 nm.Since CoII in 2a2 is in the low-spin t2g 6eg 1 configuration the band at 534 nm may be due to a ligand to metal charge-transfer transition.8 The higher energy bands at 420 and 345 nm are presumably due to intraligand n–p* and2648 J. Chem. Soc., Dalton Trans., 1998, Pages 2643–2650 p–p* transitions.8 In the lower energy part of the visible region the complex displays two weak transitions at 700 and 800 nm (Fig. 6). These are believed to be possible d–d transitions, 2E æÆ 2T1 and 2E æÆ 2T2 respectively.12 We would like to note here that the cobalt(II) ion in an octahedral environment prefers a high-spin configuration and as a consequence low-spin octahedral cobalt(II) species are rare.8,13 Thus the complexes 22 provide a rare example of the low-spin cobalt(II) octahedral system and to the best of our knowledge this demonstrates the first example of low-spin cobalt(II) complexes of CoN4O2 chromophoric class in the solution state.We were unable to grow suitable single crystals for X-ray characterisation, however the spectral results match well with those of structurally characterised similar ruthenium thiolated complexes.14 This provides additional strong support particularly in favour of the proposed stereochemistry of complexes 2.The mechanism of the conversion of complexes 1 into 2 in the presence of perbenzoic acid is not clearly understood. The conversion involves three primary simultaneously operating steps such as (i) removal of one molecule of ligand L from the starting complex 1, (ii) formation of a new selective C]O centre followed by metallation and (iii) oxidation of the cobalt(II) ion to cobalt(III).Since all the above steps (substitution, insertion and electron transfer) proceed rapidly without any detectable intermediate, it is diYcult to draw any conclusions regarding the mechanism. However, on the basis of the available knowledge Fig. 5 X-Band EPR spectrum of chemically reduced complex 2c in acetonitrile solution at 77 K (G = 1024 T; tcne = tetracyanoethylene) Fig. 6 Electronic spectrum of complex 2a2 in acetonitrile solvent of the chemistry of associated moieties the following tentative mechanism may be proposed. Since orthometallations from the pendant phenyl ring of L 15 and related ligands such as azobenzene (PhN]] NPh),16 azophenol [PhN]] NC6H4(OH)],17 dihydroxyazobenzene [(OH)C6H4N]] NC6H4(OH)],17 azobenzene thioether [(RS)C6H4N]] NPh] 18 and phenolic SchiV bases [(HO)C6H4N]] C(H)Ph,19 PhCH2N]] C(H)C6H4(OH)20] and the labile nature of one of the ligands L of 1 are known, we assume that complex 1 may be transformed initially into a four-membered reactive intermediate (A, Scheme 2) as a first step of the C]H bond activation process.The insertion of oxygen, generated from the perbenzoic acid, into the fourmembered reactive M]C s bond (A) (if it exists) would eventually lead to the formation of B. Insertion of oxygen atoms (generated from perbenzoic acid) into M]C s bonds in palladium and platinum systems has been documented.21 The inaccessibility of the orthometallated intermediate A might be due to its extreme reactivity which has possibly originated from the thermodynamically unfavourable four-membered orthometallated ring.During the transformation of 1 to 2 the metal ion has been oxidised from its bivalent state in the starting complexes 1 to the trivalent state in the hydroxylated complexes 2.The low CoII]CoIII oxidation potential (ª0.0 V) in complexes 2 indicates the possibility of the existence of the cobalt(II) congener B as a reactive intermediate, which may be spontaneously oxidised by the air to the stable cobalt(III) state as in 2. This assumption gets strong support from the experimentally observed fact that the chemically or electrochemically generated B is unstable under atmospheric conditions and spontaneously oxidised to the stable trivalent complexes 2.Further work is in progress in related cobalt systems to understand the mechanism. Conclusion We have observed cobalt ion-prompted direct and selective activation of the C]H bond of the pendant phenyl ring of L. This activation process in turn develops a facile aromatic hydroxylation reaction under ambient conditions. The hydroxy- Scheme 2 Na Np Np Np Na Na CoII Np Na Np Na CoII – L 2 p/3 Np O Na Np Na O CoIII Np O Na Np Na O CoII – e– 1 A O 2 BJ. Chem.Soc., Dalton Trans., 1998, Pages 2643–2650 2649 lated complexes 2 can act as precursors for the chemical as well as electrochemical synthesis of very rare octahedral low-spin cobalt(II) complexes 22. They have shown a complete set of sequential electron-transfer processes which are not often observable.22 Experimental Materials Cobalt carbonate (J. T. Baker, Colorado, USA) was converted into cobalt perchlorate by the standard method. The complexes [CoIIL3][ClO4]2?H2O were prepared according to the reported procedures.8 m-Chloroperbenzoic acid was obtained from Aldrich, USA.Other chemicals and solvents were reagent grade and used as received. Silica gel (60–120 mesh) used for chromatography was of BDH quality. For spectroscopic/ electrochemical studies HPLC-grade solvents were used. Commercial tetraethylammonium bromide was converted into pure tetraethylammonium perchlorate by an available procedure.22 Physical measurements Solution electrical conductivity was checked using a Systronic conductivity bridge-305.Electronic spectra (1100–200 nm) were recorded using a Shimadzu UV-160A spectrophotometer, Fourier-transform IR spectra on a Nicolet spectrophotometer with samples prepared as KBr pellets. Magnetic susceptibility was measured with a PAR vibrating sample magnetometer. The 1H and 13C NMR spectra were obtained with a 300 MHz Varian Fourier-transform spectrometer. Cyclic voltammetric measurements were carried out using a PAR model 273A potentiostat and galvanostat with a platinum working electrode, platinumwire auxiliary electrode and Ag–AgCl reference electrode in a three-electrode configuration. A PAR model 273A coulometer was used for coulometry.The supporting electrolyte was NEt4- ClO4 and the solute concentration ª1023 and ª1022 mol dm23 for the cyclic voltammetric and coulometric experiments respectively. The half-wave potential Eo 298 was set equal to 0.5 (Epc 1 Epa), where Epc and Epa are the cathodic and anodic cyclic voltammetric peak potentials respectively. The scan rate used was 50 mV s21.A platinum-wire gauze electrode was used for coulometry. All the experiments were carried out under a dinitrogen atmosphere. Electrochemical data were collected at 298 K and are uncorrected for the junction potential. The EPR spectra were recorded with a Varian model 109C E-line X-band spectrometer fitted with a quartz Dewar for measurements at 77 K (liquid nitrogen), and calibrated by using tetracyanoethylene (tcne, g = 2.0023). The elemental analyses (C,H,N) were carried out with a Carlo Erba (Italy) elemental analyser.The FAB mass spectrum at 298 K was recorded on a JEOL SX 102/DA-6000 mass spectrometer. Preparation of complexes 2 The hydroxylated complexes 2a–2g were prepared by following a general method. Details are given for complex 2a. Complex 1a (100 mg, 0.12 mmol) and m-chloroperbenzoic acid (82.8 mg, 0.48 mmol) were dissolved separately in acetonitrile (10 cm3).The perbenzoic acid solution was added dropwise to 1a over a period of 15 min with stirring. The progress of the reaction was monitored by TLC. The stirring was continued for 24 h. After completion of reaction the solvent was removed under reduced pressure. The solid product thus obtained was purified by column chromatography on a silica gel column (60–120 mesh) in benzene. The excess of ligand L1 was eluted first with chloroform–acetonitrile (5 : 1) and rejected.The pure bluish green product (2a) was eluted with chloroform– acetonitrile (2 : 1). Evaporation of the solvents under reduced pressure aVorded pure solid 2a. Finally the product was recrystallised from dichloromethane–hexane (1:4). Yields varied in the range 55–60%. Acknowledgements Financial support from the Department of Science and Technology, New Delhi, India, is gratefully acknowledged. Special acknowledgement is made to the Regional Sophisticated Instrumentation Centre (RSIC), Indian Institute of Technology, Bombay, for providing NMR and EPR facilities and RSIC, Central Drug Research Institute, Lucknow, for providing the FAB mass spectrum.The referees’ suggestions at the revision stage were very helpful. References 1 P. S. Braterman, Reactions of Coordinated Ligands, Plenum, New York, 1988; H. L. Chum and M. E. M. Helene, Inorg. Chem., 1980, 19, 876; M. Menon, A. Pramanik, N. Bag and A. Chakravorty, Inorg.Chem., 1994, 33, 403; E. M. Siegbahn, J. Am. Chem. Soc., 1996, 118, 1487; R. H. Schultz, A. A. Bengali, M. J. Tauber, B. H. Weiller, E. P. Wasserman, K. R. Kyle, C. B. Moore and R. G. Bergman, J. Am. Chem. Soc., 1994, 116, 7369; S. I. Murahashi, T. Naota, H. Taki, M. Mizuno, H. Takaya, S. Komiya, Y. Mizuho, N. Oyasato, M. Hiraoko, M. Hirano and A. Fukuoka, J. Am. Chem. Soc., 1995, 117, 12 436; R. G. Bergman, Acc. Chem. Res., 1995, 28, 154; S. Y. Liou, M. Gozin and D. Milstein, J.Am. Chem. Soc., 1995, 117, 9774; P. Ghosh, A. Pramanik, N. Bag, G. K. Lahiri and A. Chakravorty, J. Organomet. Chem., 1993, 453, 237; N. Bag, S. B. Choudhury, G. K. Lahiri and A. Chakravorty, J. Chem. Soc., Chem. Commun., 1990, 1626; S. Murai, F. Kakiuchi, S. Sekine, Y. Tanaka, A. Kamatani, M. Sinoda and N. Chatani, Nature (London), 1993, 366, 529; C. J. Li, D. Wang and L. D. Chen, J. Am. Chem. Soc., 1995, 117, 12 867; G. K. Lahiri and A. M. Stolzenberg, Angew. Chem., Int.Ed. Engl., 1993, 32, 429; M. Menon, A. Pramanik and A. Chakravorty, Inorg. Chem., 1995, 34, 3310. 2 W. A. Lee and T. C. Bruice, Inorg. Chem., 1986, 25, 131; L. Saussine, E. Brazi, A. Robine, H. Mimoun, J. Fischer and R. Weiss, J. Am. Chem. Soc., 1985, 107, 3534; R. A. Sheldon and J. K. Kochi, Metal Catalyzed Oxidation of Organic Compounds, Academic Press, New York, 1981; G. A. Hamilton, Molecular Mechanisms of Oxygen Activation, ed. O. Hayishi, Academic Press, New York, 1974, p. 405. 3 S. Goswami, A. R. Chakravarty and A. Chakravorty, Inorg. Chem., 1981, 20, 2246. 4 B. Pesce, Nuclear Magnetic Resonance in Chemistry, 1965, Academic Press, New York, p. 174; A. K. Mahapatra, B. K. Ghosh, S. Goswami and A. Chakravorty, J. Indian Chem. Soc., 1986, 63, 101. 5 B. K. Santra and G. K. Lahiri, J. Chem. Soc., Dalton Trans., 1997, 129. 6 G. K. Lahiri, S. Goswami, L. R. Falvello and A. Chakravorty, Inorg. Chem., 1987, 26, 3365. 7 A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier, New York, 1984; J.E. Huheey, E. A. Keiter and R. L. Keiter, Inorganic Chemistry, 4th edn., Harper Collins College Publishers, New York, 1993, p. 444. 8 B. K. Santra and G. K. Lahiri, J. Chem. Soc., Dalton Trans., 1998, 139. 9 P. S. Rao, G. A. Thakur and G. K. Lahiri, Indian J. Chem., Sect. A, 1996, 35, 946; N. Bag, A. Pramanik, G. K. Lahiri and A. Chakravorty, Inorg. Chem., 1992, 31, 40. 10 R. A. Krause and K. Krause, Inorg. Chem., 1984, 23, 2195. 11 C. Ohrenberg, P. Ge, P. Schebler, C. G. Riordan, G. P. A. Yap and A. L. Rheingold, Inorg. Chem., 1996, 35, 749; E. J. Hintsa and S. R. Cooper, J. Am. Chem. Soc., 1986, 108, 1208; G. S. Wilson, D. D. Swanson and R. S. Glass, Inorg. Chem., 1986, 25, 3827. 12 J. E. Huheey, E. A. Keiter and R. L. Keiter, Inorganic Chemistry, 4th edn., Harper Collins College Publishers, New York, 1993, p. 441. 13 F. A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, 5th edn., Wiley, New York, 1988, p. 733; R. S. Drago, Physical Methods for Chemists, 2dn edn., Saunders College Publishing, New York, 1992, p. 448; B. B. Wayland and M. E. Abd. Elmageed, J. Am. Chem. Soc., 1974, 96, 4809; F. L. Urbach, R. D. Bereman, J. A. Topich, M. Hariharan and B. J. Kalbacher, J. Am. Chem. Soc., 1974, 96, 5062. 14 B. K. Santra, G. A. Thakur, P. Ghosh, A. Pramanik and G. K. Lahiri, Inorg. Chem., 1996, 35, 3050. 15 P. Bandyopadhyay, D. Bandyopadhyay, A. Chakravorty, F. A. Cotton, L. R. Falvello and S. Han, J. Am. Chem. Soc., 1983, 105, 6327. 16 J. D. Gilbert, D. Rose and G. Wilkinson, J. Chem. Soc. A, 1970, 2765; M. A. Bennett, M. I. Bruce and T. W. Matheson, Com-2650 J. Chem. Soc., Dalton Trans., 1998, Pages 2643–2650 prehensive Organometallic Chemistry, eds. G. Wilkinson, F. G. A. Stone and E. W. Abel, Pergamon, New York, 1982, vol. 4, sect. 32.3, p. 691; G. R. Newkome, W. E. Puckett, V. K. Gupta and G. E. Kiefer, Chem. Rev., 1986, 86, 451. 17 G. K. Lahiri, S. Bhattacharya, M. Mukherjee, A. K. Mukherjee and A. Chakravorty, Inorg. Chem., 1987, 26, 3359. 18 A. K. Mahapatra, S. Datta, S. Goswami, M. Mukherjee, A. K. Mukherjee and A. Chakravorty, Inorg. Chem., 1986, 25, 1715. 19 P. Ghosh, A. Pramanik, N. Bag, G. K. Lahiri and A. Chakravorty, J. Organomet. Chem., 1993, 454, 237. 20 R. Hariram, B. K. Santra and G. K. Lahiri, J. Organomet. Chem., 1997, 540, 155. 21 C. Sinha, D. Bandyopadhyay and A. Chakravorty, Inorg. Chem., 1988, 27, 1173; A. K. Mahapatra, D. Bandhyopadhyay, P. Bandhyopadhyay and A. Chakravorty, Inorg. Chem., 1986, 25, 2214. 22 D. T. Sawyer, A. Sobkowiak and J. L. Roberts, jun., Electrochemistry for Chemists, 2nd edn., Wiley, New York, 1995. Received 21st April 1998; Paper 8/02995D

ISSN:1477-9226    

DOI:10.1039/a802995d

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 2645.2646 2645 Synthesis and structures of neutral and cationic 1-azaallylaluminium methyls ¢Ó Laurence Bourget, Peter B. Hitchcock and Michael F. Lappert * The Chemistry Laboratory, University of Sussex, Brighton UK BN1 9QJ. E-mail: m.f.lappert@sussex.ac.uk Received 24th May 1999, Accepted 30th June 1999 The 1-azaallylaluminium compounds [Al{N(R)C(But)C- (H)R}{N(R)C(But) C(H)R}Me] (X-ray characterised), Al{N(R)C(But)C(H)R}Me2, Al{N(R)C(Ph) C(H)R}Me2 3 and Al{N(R)C(Ph) C(H)R}2Me(thf) were obtained from the appropriate methylaluminium chloride and 1-azaallyllithium; 3 with B(C6F5)3 gave [Al{N(R)C(Ph) C(H)R}- Me(thf)(OEt2)][BMe(C6F5)3] (R SiMe3).There is much current interest in bi- and tri-dentate nitrogencentred spectator ligands, which often are a component of electrophilic neutral or cationic metal alkyls.1 Aluminium complexes have become prominent, following the disclosure of Coles and Jordan that certain cationic amidinatoaluminium methyls, such as [Al(LL)Me][B(C6F5)4] (LL = A), are active catalysts for the polymerisation of ethylene.2 We recently described the synthesis and X-ray molecular structures of a series of neutral and cationic ¥â-diketiminatoaluminium methyls [Al(LL)Me2], [Al(LL)Me(thf)][BMe(C6F5)3]0.5 thf and [Al(LL)Me(OEt2)]- [B(C6F5)4]0.5 Et2O (LL = B, R = SiMe3), the latter two being the .rst crystallographically characterised cationic aluminium methyls.1 The salt [Al{N(Ar)C(Me)C(H)C(Me)N(Ar)}But]- [B(C6F5)4] has been mentioned (Ar = C6H3Pri 2-2,6).3 Molecular orbital calculations on Al.L complexes (L = C6H6, C4H4O, C5H6 or C4H4NH) have been carried out.4 We now report on aluminium methyls containing the 1-azaallyl ligands C5 and D6 (or their tautomers C and D), which are skeletally isoelectronic with amidinates such as A.Treatment of the appropriate methylaluminium chloride with [Li(C)]2 5 or [Li(D)(thf)]2 6 in the correct stoichiometry in hexane at low temperature a.orded in high yield the 1-azaallylaluminium methyls 1 ¢Ô and 2 (Scheme 1) or 3 and 4 ¢Ô (i and ii in Scheme 2). From 3 and tris(penta.uorophenyl)borane, under the same conditions, the salt 5 was obtained (iii in Scheme 2), containing a 1-azaallyl(methyl)aluminium cation. The pale yellow, crystalline complex 1 gave a satisfactory EI (70 eV) mass spectrum; the highest m/z peak corresponded to M (4%), followed by [M Me] (44%).The multinuclear NMR spectra in toluene-d8 showed (Table 1) that (i) the solution structure corresponded to that in the crystal (Fig. 1); (ii) the two 1-azaallyl ligands were tautomers [Al(C)(C)Me]; (iii) ¢Ó No reprints available. the ligands C and C exchanged rapidly on the NMR time scale, eqn. (1); and (iv) the 1H NMR spectral coalescence temperature at Tc = 302 K [based on the C(CH3) signal at ¥ä = 0.99 at 302 K], corresponding to .G¢Ô 302 K = 60.7 kJ mol1. The molecular structure of the crystalline complex 1 is illustrated in Fig. 1.¡× The aluminium atom is almost trigonally disposed with respect to the atoms C(25), C(2) and N(2) (the sum of the angles subtended by these atoms at Al = 349.3 ), the bond to N(1) completing the trigonal monopyramidal geometry about the four-coordinate metal atom. The two 1-azaallyl ligands di.er in that one is of type C and the other C. Thus, (i) the Al.N(1) bond of 1.998(2) A is only slightly longer than the average Al.N bond length of 1.928(3) A in the three-coordinate aluminium ¥â-diketiminate [Al{N(R)C(Ph)C(H)C(Ph)N(R)}- Scheme 1 Synthesis of the 1-azaallylaluminium methyls [Al(C)- (C)Me] 1 and Al(C)Me2 2.Reagents and conditions: i 1/2 (AlMeCl2)2, C6H14, 78 C; ii (AlMe2Cl)2, C6H14, 78 C. Scheme 2 Synthesis of the 1-azaallylaluminium methyls Al(D)- Me2(thf) 3, Al(D)2Me(thf) 4 and [Al(D)Me(thf)(OEt2)][BMe- (C6F5)3] 5. Reagents and conditions: i 1/2 (AlMe2Cl)2, C6H14, 78 C; ii 1/2 (AlMeCl2)2, C6H14, 78 C; iii B(C6F5)3, C6H14, 78 C, then Et2O.2646 J.Chem. Soc., Dalton Trans., 1999, 2645–2646 Me2] 6,1 whereas the Al–N(2) bond of 1.839(2) Å is appreciably shorter; (ii) the Al–C(2) bond of 2.022(2) Å is only marginally shorter than the Al–C(25) bond of 1.990(2) Å, or the average Al–C bond length of 1.964(3) Å in 6,1 and is much shorter than the Al C(13) distance of 2.816(2) Å; and (iii) the C(1)–C(2) bond of 1.468(3) Å is signi.cantly longer than the C(13)–C(14) of 1.347(3) Å.The presence in a single crystalline molecule of two tautomeric 1-azaallyl ligands such as C and C has previously been observed in [Sn{N(R)C(Ph)C(H)CR2}{N(R)C(Ph) CR2}] 7, in which the Sn–N and C(Ph)-CR2 bond lengths were 2.288(4) and 2.153(4) and 1.461(6) and 1.365(3) Å, respectively.7 Moreover, the two ligands of 7 underwent rapid intramolecular exchange in toluene-d8 down to 90 C. The assignment of structures for the complexes 2–5 rests at present on their multinuclear NMR spectra, as indicated for the 1-azaallyl ligands C, C, D and D in Table 1.The presence of an equivalent of thf in each of the complexes 3–5 and also of Et2O in 5 is consistent with their 1H NMR spectra. The ionic character of the salt 5 is borne out by the 11B{1H} NMR spectrum, d = 14.6 (w1/2 = 90 Hz), cf., d = 16.7 for [Al{N(R)C(Ph)C(H)C(Ph)N(R)}Me(thf)][BMe(C6F5)3].1 The Fig. 1 Molecular structure of 1 with selected bond distances (Å) and angles ( ): Al–N(1) 1.998(2), Al C(1) 2.380(2), Al–C(2) 2.022(2), Al–C(25) 1.990(2), Al–N(2) 1.839(2), C(1)–C(2) 1.468(3), N(1)–C(1) 1.320(3), N(2)–C(13) 1.445(3), C(13)–C(14) 1.347(3) Å; N(1)–Al–C(2) 70.1(1), C(2)–Al–C(25) 122.1(1), C(25)–Al–N(2) 111.1(1), N(2)–Al– N(1) 112.5(1), N(1)–Al–C(25) 119.1(1), N(2)–Al–C(2) 116.1(1) .Table 1 Selected NMR spectroscopic chemical shifts (d) and assignments Me3SiNCa(But or Ph)Cb(H)SicMe3 Compound Assignment T (K)/ (Ligand) Ca Cb H 29Sic solvent 1 (Enamide) (.3-1-Azaallyl) 2 (1-Azaallyl) 3 (Enamide) 4 (Enamide) 5 (Enamide) 174.0 221.0 216.1 165.8 165.5 160.3 116.4 45.2 49.6 111.3 114.4 116.2 4.78 2.09 2.47 5.08 5.07 5.04 15.1 1.6 1.0 11.7 11.6 a 10.3 203/C7D8 333/C6D6 298/C6D6 333/C6D6 298/C6D6 a At 298 K. 27Al{1H} NMR spectral signals for each of the complexes 1–5 was broad but distinct: d 125 (1, 1.2), 141 (2, 2.4), 155 (3, 2.4), 128 (4, 3.8) and 167 (5, 4.0) (the second number in each bracket refers to w1/2 in kHz). The di.erence in structures between the monomethylaluminium bis(azaallyls) 1 (C, C) and 4 (D2), is attributed to the ß-carbon substituent (But or Ph) rather than the in.uence for 4 of thf, since structure 1 was also found when 1 was prepared in thf.The bis(enamido)aluminium structure for 4 may have been preferred because of the conjugative e.ect. Compounds related to the 1-azaallyls of aluminium, the 2- pyridylmethyls [Al{C(R)2(C5H4N-2)}2][AlCl4] and [Al{C(R)2- (C5H3Me-6)-2)}nCl3 n] (n = 1 or 2) are known.8 The present results, together with those by Roesky and co-workers on the recently prepared neutral aluminium compounds containing the ligand [C(R)2C(Ph)NR],9 demonstrate the versatility of 1-azaallyl ligands in aluminium chemistry, which is now being explored more extensively.We thank the European Commission for the award of a Marie Curie fellowship for L. B. Notes and references ‡ Synthesis of 1 and 4.Methylaluminium dichloride (1.7 cm3 of a 1 mol dm3 solution in hexanes, 1.7 mmol) was added dropwise to a solution of [Li(C)]2 (0.86 g, 1.72 mmol) in hexane (60 cm3) at –78 C. The mixture was stirred for 12 h at ambient temperature, then .ltered. Upon concentration and cooling at 4 C, pale yellow crystals of 1 (0.87 g, 97%), mp 95–98 C were obtained (Found: C, 56.6; H, 10.9; N, 5.40. C25H59AlN2Si4 requires C, 57.0; H, 11.2; N, 5.31%). In a similar fashion, from AlMeCl2 (2.4 cm3 of a 1 mol dm3 solution in hexanes, 2.4 mmol) and [Li(D)(thf)]2 (1.64 g, 2.4 mmol), there was obtained the pale orange solid 4 (1.4 g, 95%), mp 80–85 C (Found: C, 59.4; H, 9.20; N, 4.46.C33H59AlN2Si4 requires C, 62.0; H, 9.25; N, 4.39%). § Crystallographic data for 1: C25H59AlN2Si4, M = 527.19, monoclinic, space group P21/n, a = 11.360(4), b = 20.099(5), c = 15.366(8) Å, ß = 101.96(4) , U = 3432(2) Å3, Z = 4, .(Mo-Ka) = 0.71073 Å, µ = 0.21 mm1. Data were collected at 173(2) K on an Enraf Nonius CAD4 di.ractometer in the . 2.mode for the range of 2 < . < 25 . The structure was solved by direct methods (SHELXS-97) and re.ned with full-matrix, least-squares on all F2 (SHELXL-97).10 All non-hydrogen atoms were anisotropic, and hydrogen atoms were included in the riding mode with Uiso(H) = 1.2 Ueq(C) or 1.5 Ueq for Me groups. Final residual for 6033 independent re.ections was R1 = 0.066, wR2 = 0.110 and for the 4592 with I > 2s(I), R1 = 0.043, wR2 = 0.098. CCDC reference number 186/1547. See http://www.rsc.org/suppdata/dt/1999/2645/ for crystallographic .les in .cif format. 1 F. Coslédan, P. B. Hitchcock and M. F. Lappert, Chem. Commun., 1999, 705. 2 M. P. Coles and R. F. Jordan, J. Am. Chem. Soc., 1997, 119, 8125. 3 C. E. Radzewich, M. P. Coles and R. F. Jordan, J. Am. Chem. Soc., 1998, 120, 9384. 4 D. Stöckigt, J. Am. Chem. Soc., 1999, 18, 1050. 5 P. B. Hitchcock, M. F. Lappert and D.-S. Liu, J. Chem. Soc., Chem. Commun., 1994, 2637. 6 R. Sablong, unpublished work; for the tmen complex, see P. B. Hitchcock, M. F. Lappert and M. Layh, Chem. Commun., 1998, 201. 7 J. Hu, P. B. Hitchcock, M. F. Lappert, M. Layh and J. R. Severn, Chem. Commun., 1997, 1189. 8 T. R. van den Ancker and C. L. Raston, J. Organomet. Chem., 1995, 500, 289. 9 C. Cui, H. W. Roesky, M. Noltemeyer, M. F. Lappert, H.-G. Schmidt and H. Hao, Organometallics, 1999, 18, 2256. 10 G. M. Sheldrick, SHELXS-97 and SHELXL-97, Programs for crystal structure solution and re.nement, University of Göttingen, 1997. Communication 9/04138I

ISSN:1477-9226    

DOI:10.1039/a904138i

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 2647–2649 2647 An open chain trinuclear titanium(IV) isopropoxide species formed with the tridentate ligand cis,cis–cyclohexane–1,3,5-trialkoxide [C6H9O3]3. Crystal structure of [{Ti(OPri)3}2{-Ti(C6H9O3- O1,O5)2}] Jonathan P. Corden,a William Errington,a Peter Moore,*a Martin G. Partridge b and Malcolm G. H. Wallbridge a a Department of Chemistry, University of Warwick, Coventry, UK CV4 7AL. E-mail: p.moore@warwick.ac.uk b Synetix, Polymers, Inks and Coatings Group, Haverton Hill Site, Haverton Hill Road, Billingham, Cleveland, UK TS23 1PS Received 28th April 1999, Accepted 9th June 1999 Reaction of titanium(IV) tetraisopropoxide with cis,ciscyclohexane- 1,3,5-triol [C6H9(OH)3, H3L] affords the novel trinuclear titanium(IV) alkoxide species [{Ti(OPri)3}2{- Ti(C6H9O3-O1,O5)2}] {L [C6H9O3]3} which possesses a unique open-chain V-shaped arrangement of three titanium atoms, with the central 6-co-ordinate Ti atom bonded to two tridentate L groups, and with each of the two terminal Ti atoms 5-co-ordinate, forming bonds to three nonbridging isopropoxide groups, and to two bridging oxygen atoms, one from each of the two ligands L.Alkoxides of the Group 4 elements, and especially those of titanium, continue to be a subject of interest with respect to their structure and diverse chemistry.1 They also .nd many applications; for example, they act as catalysts in the manufacture of esters and polyole.ns,2 as well as in asymmetric synthesis using the Katsuki–Sharpless epoxidation procedure,3 and they are important precursors in sol–gel technology leading to species such as mesoporous titanium(..) oxides and more recently titania–silica catalysts.4 However, the high susceptibility of [Ti(OR)4] (R = alkyl, aryl) species to hydrolysis remains a problem in relation both to their handling and reactivity, since any hydrolysis can result in some loss of catalytic activity and the formation of Ti–O–Ti bonds, with the eventual precipitation of insoluble oxo species.5 We are currently investigating the potential of compatible supporting ligands which could modify, but not seriously impair, the reactivity of the [Ti(OR)4] alkoxides.The use of multidentate alkoxide ligands in this role is one possibility, since such ligands usually possess a high charge that is compatible with the higher oxidation states of the early transition metals. This together with the presence of a rigid chelating framework could o.er an increase in the stability of the overall alkoxide species.In our initial studies we have used the potentially tridentate cis,cis-cyclohexane-1,3,5-trialkoxide ligand L3, which is derived from the corresponding and easily accessible triol (H3L), and we now report the results of the reaction of H3L with [Ti(OPri)4]. The addition of [Ti(OPri)4] to a stirred suspension of H3L (1 : 1 molar ratio) in toluene at 25 C yields a clear solution from which the colourless solid product [Ti3(OPri)6L2] 1 may be obtained by removal of the solvent [eqn.(1)]. 3[Ti(OPri)4] 2H3L .. [Ti3(OPri)6L2] 6 PriOH (1) It is interesting to note that 1 is apparently the only Ti(..) species formed since the use of other molar ratios (H3L:Ti = 1 : 2, 1 : excess) always yields the same solid.† The product is readily soluble in a range of aromatic, and even to a small extent in aliphatic, solvents.It is slowly hydrolysed by moist air, but it is noticeably less reactive to moisture than the parent [Ti(OPri)4]. The crystal structure of 1 ‡ shows the novel feature of two cyclohexane trialkoxide ligands bonded to the central titanium of an open V-shaped trinuclear metal unit, with a central 6-coordinate titanium .anked by two 5-co-ordinate metal centres (Fig. 1). In the open metal framework the two terminal titanium atoms Ti(1) and Ti(3) make an angle of 120.94(5) at the central Ti(2) atom.The two trialkoxides are both bonded to the central metal atom in a tridentate mode, with the added feature that in each ligand two of the oxygen donor centres are also bridging to each of the terminal titanium atoms. The third oxygen of the two trialkoxide ligands is uniquely bonded to the central Ti(2) atom. The six isopropoxide groups are distributed evenly between the two terminal titanium atoms, and this is a rare example of a complex in which the isopropoxides do not occupy any of the bridging positions.Not unexpectedly the co–ordination around both the central and terminal metal atoms shows a considerable distortion from octahedral and trigonal bipyramidal geometries respectively. For example, the O(6)–Ti(2)–O(3) angle is 72.86(17) and O(3)– Ti(1)–O(7) is 166.0(2). The terminal Ti–O(isopropoxide) and the unique Ti(2)–O(2) distances are all near 1.80 Å, but there is a distinct asymmetry in the Ti2O2 bridge system where the Ti(1)–O(6) and Ti(2)–O(6) distances are 1.978(4) and 2.084(4) Å respectively.These distances are similar to those observed in the dimeric titanium(..) neopentoxide [Ti(µ-OCH2CMe3)- (OCH2CMe3)3]2 where the terminal Ti–O distances are all near 1.80 Å, while the Ti2O2 bridge bond lengths vary from 1.965(4) to 2.104(4) Å.6 Other trinuclear Ti(..) alkoxide species that have been reported are very di.erent to 1. They include the (µ3-oxo)alkoxides, [Ti3(µ3-O)(OPri)10] and [Ti3(µ3-O)(µ3-OMe)- (µ-OPri)3(OPri)6] that contain three 6-co-ordinate Ti(..) centres,7 and the trinuclear Ti(...) species, [Ti3(OPh)9- (TMEDA)2] (TMEDA = Me2NCH2CH2NMe2). The structure of the latter shows that the two terminal Ti atoms are identical and 6-co-ordinate, being bonded to one bidentate TMEDA, and four phenoxide groups.Two of these four phenoxides are also bridged to the central 5-co–ordinate Ti, and the latter is also bonded to a single non–bridging phenoxide.8 The angle between the three Ti atoms is 151.92(6), in contrast to the angle of 120.94(5) found in 1.The structure of 1 in the solid state is in accord with the solution 1H and 13C NMR spectra in toluene-d8. In the 1H NMR the integrals of the resonances from L3 are in a 2 : 1 ratio, consistent with the asymmetric bonding of the ligands. Thus the equatorial and axial CH2 protons are centred at d 2.60 and 2.36 and 1.63 and 1.46 [area ratio 2 : 1; from those attached to {C8(C4) and C10(C6)}, and C12(C2)] respectively, and the equatorial CHO protons are at d 5.08 and 4.41.The2648 J. Chem. Soc., Dalton Trans., 1999, 2647–2649 Fig. 1 Molecular structure of 1. Selected bond lengths (Å) and angles (): Ti(1)–O(3) 2.087(4), Ti(1)–O(6) 1.978(4), Ti(2)–O(6) 2.084(4), Ti(2)–O(3) 1.973(4), Ti(2)–O(2) 1.829(4), Ti(2)–O(1) 2.104(4), Ti(2)–O(4) 1.964(4), Ti(3)–O(1) 1.975(4), Ti(3)–O(4) 2.089(4), Ti(1)–O(7) 1.803(5); O(3)–Ti(1)– O(6) 72.72(17), O(3)–Ti(1)–O(7) 166.0(2), O(3)–Ti(2)–O(6) 72.86(17), O(2)–Ti(2)–O(4) 109.71(19), O(2)–Ti(2)–O(6) 162.55(19), O(1)–Ti(3)–O(4) 72.68(16). six isopropoxide groups are equivalent indicating either .uxionality of the 5-coordinate Ti(..) centres, or rapid exchange of the isopropoxide groups.They show resonances at d 4.83 (septet, 6H; 3J 6 Hz) and 1.29 (doublet, 36H). The proton decoupled 13C spectrum shows the expected six resonances.† Indications are that the same structure is retained in the gas phase, since in the E.I.mass spectrum although the parent ion was not found (RMM = 759, based on 49Ti), ions were observed at m/z values of 700, 641 and 582 corresponding to the loss of one, two and three isopropoxide groups respectively. It is relevant to compare the present results with those from earlier work. While there are previous reports of this trialkoxide ligand bonding in a tridentate fashion to a single metal centre,9 there are no structurally characterised examples of the asymmetric bonding as observed here.However, for the closely related ligand, 1,3,5-(tri.uoromethyl)cyclohexane-1,3,5-trialkoxide, there is one example where the three oxygen atoms are bridge-bonded to Cu(..) and Cd(..) centres, and the more distantly related inositolate(3) ligand is known to bridge between iron atoms.10 In general there have been surprisingly few reports of the use of trialkoxide ligands in combination with the early transition metals, and with the Group 4 elements in particular the only well characterised derivatives with L are a series of substituted cyclopentadienyl compounds of the type [CpTi(L)] [Cp = C5H5, C5Me5, C5H3(SiMe3)2, etc.].11 A report of the reaction of other trialkoxide species, namely 1,1,1-tris(hydroxymethyl)- ethane (H3THME) and 1,1,1-tris(hydroxymethyl)propane (H3THMP) with [M(OPri)4] (M = Ti, Zr) has shown the existence of the tetranuclear cage systems [L2M4(OPri)10] (L = THME, THMP; M = Ti, Zr), which act as precursors to ceramic thin .lms.12 An earlier preliminary report by the same workers claimed that a product with the same composition was formed with L, but this has not been subsequently con.rmed.13 Some related cluster metal oxo trialkoxide complexes are also known for the elements of Groups 5 and 6, for example [NH4]4[V10O16(THMP)4]4H2O and [Bun 4N]2- [Mo3O7 (THME)2].14 Therefore, the present results illustrate that for the .rst time, and in contrast to earlier studies, it is now possible to stabilise a unique open chain titanium alkoxide system by the use of an appropriate alkoxide ligand. It is also noteworthy that the complex contains two terminal titanium centres that are co– ordinatively unsaturated, and hence potentially useful as catalytic sites.We are currently investigating further reactions involving other chelating O-donor ligands in order to determine whether such ligands are able in general to stabilise unusual metal alkoxide species.Acknowledgements We thank Synetix, Polymers, Inks and Coatings Group, for a grant in support of this work, Dr O. W. Howarth and one of the referees for helpful comments on the NMR spectra, the EPSRC Mass Spectrometry Service at the University of Swansea for assistance in recording the mass spectra of the sample, and the EPSRC for provision of NMR and X-ray facilities. Notes and references † Compound 1 was prepared by adding [Ti(OPri)4] (1.37 cm3, 4.53 mmol) to a stirred suspension of dehydrated H3L (0.3 g, 2.26 mmol) in dry toluene (25 cm3) at room temperature. The resulting suspension was stirred further until a clear solution was obtained. The solvent was then removed in vacuo, and the resultant colourless solid was washed thoroughly with hexane, .ltered o.and dried by pumping in vacuo. Yield 40% [Found (required for C30H60O12Ti3): C, 46.8 (47.6); H, 7.9 (8.0%)]. 1H NMR (C6D5CD3, 300 MHz, 295 K): d 5.08 (mult, cyclo- CH, 4Heq), 4.83 (sept, isoprop-CH, 6H), 4.41 (mult, cyclo-CH, 2Heq), 2.62, 2.57 (d/mult, cyclo-CH2, 2Heq) and 2.38, 2.33 (d/mult, cyclo- CH2, 4Heq) [area ratio 1 : 2; doublets of multiplets (2Jav 14 Hz) from equatorial protons attached to C12(C2) and from those attached to {C8(C4) and C10(C6)} respectively], 1.65, 1.61 (d/mult, cyclo-CH2, 4Hax) and 1.48, 1.43 (d/mult, cyclo-CH2, 2Hax) [area ratio 2 : 1; from those attached to {C8(C4) and C10(C6)}, and C12(C2) respectively], 1.30, 1.28 (d, isoprop-CH3, 36H). 13C NMR (75.48 MHz DEPT spectrum): d 77.7 (6 CH, Me2CH-O), 74.4 (2 CH, C5, C9), 73.3, {4 CH, C7(C3) and C11(C1)}, 38.5 {4 CH2, C8(C4) and C10(C6)}, 38.1 (2 CH2, C12,C2), and 26.6 [12 CH3, (CH3)2CHO]. ‡ Data were collected using a Siemens SMART CCD area-detector di.ractometer. An absorption correction was applied using SADABS.15 The structure was solved by direct methods and re.ned by full– matrix least–squares on F2 for all data using SHELXL 97.16 All the isopropoxide groups are disordered.Crystal data: C30H60O12Ti3, M = 756.48, monoclinic, a = 19.1305(6), b = 9.8798(2), c = 20.7186(6) Å,J. Chem. Soc., Dalton Trans., 1999, 2647.2649 2649 ¥â = 91.66(3), U = 3914.29(18) A3, T = 180(2) K, space group P21/n, Z = 4, ¥ì(Mo-K¥á) = 0.649 mm1, 18697 re.ections measured, 6856 unique (Rint = 0.097). The .nal values of R1 [I > 2¥ò(I)] and wR2 are 0.089 and 0.1812 respectively. CCDC reference number 186/1502. See http://www.rsc.org/suppdata/dt/1999/2647/ for crystallographic .les in .cif format. 1 For general reviews see: D. C. Bradley, Chem. Rev., 1989, 89, 1317; R. M. Mehrotra and A. Singh, Prog. Inorg. Chem., 1997, 46, 239; I. P. Rothwell and M. H. Chisholm, in Comprehensive Coordination Chemistry, eds. G. Wilkinson, R. D. Gillard and J. A. McCleverty, Pergamon, Oxford, 1987, vol. 2, p. 335. 2 For summary see: D. E. Putzig and T. W. del Pesco, in Kirk Othmer Encyclopaedia of Chemical Technology, Wiley, New York, 4th edn., 1997, vol. 24, p. 322; M. I. Siling, V. V. Kuznetsov, Y. E. Nosovskii, S. A. Osintseva and A. N. Kharrasova, Kinet. Catal. (Transl. of kinet. katal.), 1986, 27, 88; S. R. Bhutada and V. G. Pangarkar, J. Chem. Technol. Biotechnol., 1986, 36, 61; K. Akagi, K. Mochizuki, A. Yoshifumi and H. Shirakawa, Bull. Chem. Soc. Jpn., 1993, 66, 3444. 3 M. G. Finn and K. B. Sharpless, in Asymmetric Synthesis, Academic Press, New York, 1985, vol. 5, p. 247; E. J. Corey, J. Org.Chem., 1990, 55, 1693; R. O. Duthaler and A. Hafner, Chem. Rev., 1992, 92, 807. 4 H. Dislich and P. Hinz, J. Non-Cryst. Solids, 1982, 48, 11; D. M. Antonelli and J. Y. Ying, Angew. Chem., Int. Ed. Engl., 1995, 34, 2014; R. Hutter, T. Mallat and A. Baiker, J. Catal., 1995, 153, 177. 5 V. W. Day, T. A. Eberspacher, W. G. Klemperer and C. W. Park, J. Am. Chem. Soc., 1993, 115, 8469; E. A. Khrustaleva, Y. G. Yatluk and A. L. Suvorov, Izv. Akad. Nauk SSSR, Ser. Khim., 1989, 6, 1247. 6 T.J. Boyle, T. M. Alam, E. R. Mechenbier, B. L. Scott and J. W. Ziller, Inorg. Chem., 1997, 36, 3293. 7 V. W. Day, T. A. Eberspacher, Y. Chen, J. Hao and W. G. Klemperer, Inorg. Chim. Acta, 1995, 29, 391. 8 R. Minhas, R. Duchateau, S. Gambarotta and C. Bensimon, Inorg. Chem., 1992, 31, 4933. 9 J. Bebendorf, H.-B. Burgi, E. Gamp, M. A. Hitchman, A. Murphy, D. Reinen, M. J. Riley and H. Stratemeir, Inorg. Chem., 1996, 35, 7419. 10 R. Castro, M. L. Duran, J. A. Garcia.Vazquez, J. Romero, A. Sousa, A. Castineiras, W. Hiller and J. Strahle, Polyhedron, 1992, 11, 1195; K. Hegetschweiler, L. Hausherr.Primo, W. H. Koppenol, V. Gramlich, L.Odier, W. Meyer, H. Winkler and A. X. Trautwein, Angew. Chem., Int. Ed. Engl., 1995, 34, 2242. 11 D. M. Choquette, W. E. Buschmann, M. M. Olmstead and R. P. Planalp, Inorg. Chem., 1993, 32, 1062; D. M. Choquette, W. E. Buschmann, R. F. Grace.a and R. P. Planalp, Polyhedron, 1995, 14, 2569. 12 T. J. Boyle, R. W. Schwartz, R. J. Doedens and J. W. Ziller, Inorg. Chem., 1995, 34, 1110. 13 T. J. Boyle and R. W. Schwartz, Comments Inorg. Chem., 1994, 16, 243, 14. 14 M. I. Khan and J. Zubieta, Prog. Inorg. Chem., 1995, 43, 1; M. I. Khan, Q. Chen and J. Zubieta, J. Chem. Soc., Chem. Commun., 1992, 305; E. Gumaer, K. Lettko, L. Ma, D. Macherone and J. Zubieta, Inorg. Chim. Acta, 1991, 179, 47. 15 G. M. Sheldrick, SADABS, Empirical Absorption Correction Program, University of Gottingen, 1996. 16 G. M. Sheldrick, SHELXL 97, Program for Crystal Structure Re.nement, University of Gottingen, 1997. Communication 9/03384J

ISSN:1477-9226    

DOI:10.1039/a903384j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON COMMUNICATION J. Chem. Soc., Dalton Trans., 1999, 2651–2652 2651 An octanuclear pyromellitimidato(2) palladium complex: coordinative assembly of a molecular box Howard M. Colquhoun,*a Richard A. Fairman,b Paula Tootell a and David J. Williams *c a Department of Chemistry, University of Salford, Salford, UK M5 4WT. E-mail: h.m.colquhoun@chemistry.salford.ac.uk b Department of Chemistry, University of the West Indies, St Augustine, Trinidad, WI c Department of Chemistry, Imperial College, South Kensington, London, UK SW7 2AY.E-mail: djw@ic.ac.uk Received 18th June 1999, Accepted 5th July 1999 Reaction of pyromellitimide with the binuclear palladium complex [Pd2(-Cl2)(1,2-C6H4NMe2)2] in the presence of base affords a box-like, pyromellitimidato(2), octapalladium complex exhibiting helical chirality, for which 1H NMR and single crystal X-ray data are reported. In the decade which has elapsed since the .rst metallamacrocyclic complexes were reported,1 interest in the coordinative asssembly of molecular fragments into nanoscale structures with potential for selective complexation and/or catalysis has increased dramatically.2 This type of assembly is generally directed by the relative orientations of donor and acceptor orbitals associated with ligand and transition metal subunits so that, for example, a linearly-coordinating, bidentate donor ligand will generally assemble with a cis-constrained, square planar, acceptor unit to a.ord a square, tetranuclear complex.1 Variations in the angles subtended by donor and acceptor orbitals can, in principle (and increasingly in practice), give rise to a wide variety of high-nuclearity polygonal and polyhedral structures.3 It has long been known that imidato(1) ligands such as those obtained by deprotonation of succinimide or phthalimide are capable of bridging pairs of palladium atoms via N,Ocoordination, giving rise to binuclear complexes such as 1 (where N C = orthometallated dimethylbenzylamine) in which the two imide-derived ligands are directed at right angles to one another.4 The existence of this type of structure suggested to us that the dianion obtained by N,N-deprotonation of pyromellitimide 5 might well act as a linearly-coordinating ligand (2), capable of bridging two pairs of metal atoms and thus perhaps able to generate octanuclear box-like structures.We now report the synthesis and structure of the .rst such complex— a novel octapalladium species exhibiting helical chirality.Reaction of pyromellitimide (0.39 g) with an equimolar quantity of the binuclear, orthopalladated complex [Pd2(µ-Cl2)- (1,2-C6H4NMe2)2] 6 in re.uxing dichloromethane (30 cm3) for four hours, in the presence of a tenfold molar excess of triethylamine, led to formation of an orange solution. After washing with water to remove excess amine and amine hydrochloride this solution was dried over magnesium sulfate and evaporated to dryness.Two recrystallisations of the residue from dichloromethane –diethyl ether a.orded a yellow crystalline compound (3) in ca. 30% yield (on the basis of a 1 : 1 reaction). The 1H NMR spectrum of this material showed a very narrow singlet at 7.88 ppm, assignable to equivalent protons of the pyromellitimidato( 2) ligand, together with singlet resonances at 2.01 and 2.95 ppm assigned to inequivalent N-methyl groups of orthometallated dimethylbenzylamine. Also present was an AB “quartet”, centred at 3.55 ppm and assigned to inequivalent gem-methylene protons.The 1H NMR spectrum strongly suggested that a box-type structure might well have been formed, since inequivalent N-methyl and N-methylene protons in an orthometallated dimethylbenzylamine ligand are only found for binuclear complexes where the bridging ligands [generally imidato(1) or carboxylato(1)] enforce a “U-shaped” or “orthogonal” molecular geometry rather than the “coplanar” structure adopted in halide(1)-bridged structures.4 Single crystals suitable for X-ray analysis were grown by vapour di.usion of diethyl ether into a dichloromethane– methanol solution of 3.† The molecular structure of 3 is shown in Fig. 1. Eight palladium atoms are located at the corners of a slightly skewed square prism, and are linked in pairs by four tetradentate (N,N,O,O) pyromellitimidato(2) ligands. In keeping with the NMR data, these ligands traverse the diagonals of the box, so that the two 1,4-related aromatic protons on each ligand are equivalent.Moreover, one of the two N-methyl groups associated with each orthometallated ligand Fig. 1 X-Ray structure of compound 3 (hydrogen atoms and solvent molecules omitted). Selected bond lengths (Å) and angles (): Pd–C(7) 1.961(7), Pd–N(2) 2.033(5), Pd–N(14) 2.079(5), Pd–O(1) 2.143(4); C(7)–Pd–N(2) 93.6(3), C(7)–Pd–N(14) 82.9(3), N(2)–Pd–N(14) 174.9(2), N(2)–Pd–O(1) 90.5(2), C(7)–Pd–N(14) 82.9(3), C(7)–Pd–O(1) 171.8(3), N(14)–Pd–O(1) 92.5(2).2652 J.Chem. Soc., Dalton Trans., 1999, 2651–2652 is directed towards the face of an aromatic (dimethylbenzylamine) ring bound to the adjacent palladium, thus accounting for the large up.eld shift (.d = 0.94 ppm) relative to that of its adjoining N-methyl substituent. The complex has crystallographic D4 symmetry and, as a consequence, contains only one unique palladium centre in the asymmetric unit. The geometry at palladium is slightly distorted square planar with cis-angles in the range 82.9–93.6(3), the most acute angle being associated with the orthometallated ligand. The transannular separation of parallel pyromellitimidato( 2) planes is 10.98 Å, and adjacent palladium coordination planes are mutually inclined by some 47, with a non-bonded Pd Pd separation of 3.32 Å.The crystal is very heavily solvated, with “chains” of disordered diethyl ether molecules .lling the “internal” channels and molecules of dichloromethane occupying the “external” channels, formed by corner-to-corner stacking of the boxes (Fig. 2). The diagonal orientation of the four pyromellitimidato( 2) ligands gives the molecule as a whole helical chirality (Fig. 3),7 though the crystal is in fact racemic, with alternate layers of molecules having opposite chirality as shown in Fig. 2. Such chirality is also retained in solution, since addition of the aromatic chiral shift reagent (S)-()-tri.uoro-1- (9-anthryl)ethanol (1 molar equivalent) to a solution of 3 in Fig. 2 Crystal packing of compound 3 (orthometallated ligands omitted) showing the corner-to-corner molecular stacking and the alternating chirality of molecules in adjacent layers. Fig. 3 The two enantiomers of compound 3 [bold lines represent the N N axes of pyromellitimidato(2) ligands]. CD2Cl2 results in a splitting of the pyromellitimide singlet resonance into two peaks of equal intensity (.d = 0.03 ppm).Resonances associated with the orthometallated ligand are however una.ected by the chiral shift reagent, suggesting that it must hydrogen-bond to complex 3 in a regiospeci.c as well as a diastereoselective fashion, presumably via one of the noncoordinated imidato carbonyl groups. It should be noted that, prior to recrystallisation, the crude reaction product shows a number of additional resonances in the 1H NMR spectrum, in particular a series of nine weak lines in the range 7.6–8.0 ppm, several of which are split into very narrow doublets (J < 1 Hz). An analysis of possible isomers of compound 3 suggests that these additional resonances may be assigned to other octapalladium species in which either two or four pyromellitimidato(2) ligands traverse the edges, rather than the diagonals of the box.This would give rise to inequivalence of the 1,4-related pyromellitimide protons and hence to .ve-bond 1H–1H coupling between them, so providing a possible explanation for the observed series of doublets.Work aimed at isolating and identifying these additional species is currently in progress. Also, given the known ability of pyromellitimide and its derivatives to form charge-transfer complexes with electron-rich aromatic molecules,8 the ability of 3 to encapsulate such species in solution and in the solid state is under investigation. We are grateful to Dr B. Lygo of the Salford University Chemistry Department for advice on chiral shift reagents, and to the University of the West Indies for granting study leave (to R.A. F.). Notes and references † Crystal data for 3: C112H104N16O16Pd88C4H10O4CH2Cl2, M = 3713.98, tetragonal, space group P4/nnc, a = 22.069(1), c = 17.491(2) Å, V = 8518.3(9) Å3, T = 183 K, Z = 2, Dc = 1.488 g cm3, µ(Cu-Ka) = 83.35 cm1, F(000) = 3776. R1 = 0.0501, wR2 = 0.1080 for 2107 independent observed re.ections [2. = 120, Fo > 4s(Fo)]. Data were measured on a Siemens P4/RA di.ractometer with graphite-monochromated Cu-Ka radiation using .-scans.The data were corrected for Lorentz and polarisation e.ects, and a semiempirical absorption correction (from .-scans) was applied. The structure was solved by the heavy-atom method and non-hydrogen atoms were re.ned anisotropically using F2 data and the SHELXTL program package version 5.03. CCDC reference number 186/1550. See http:// www.rsc.org/suppdata/dt/1999/2651/ for crystallographic .les in .cif format. 1 M. Fujita, J. Yazaki and K. Ogura, J. Am. Chem. Soc., 1990, 112, 5645. 2 B. Olenyuk, A. Fechtenkötter and P. J. Stang, J. Chem. Soc., Dalton Trans., 1998, 1707. 3 For recent examples, see B. Olenyuk, A. Fechtenkötter and P. J. Stang, Nature, 1999, 398, 796; R. D. Schnebeck, E. Freisinger and B. Lippert, Angew. Chem., Int. Ed., 1999, 38, 168; S.-W. Lai, M. C.-W. Chan, S.-M. Peng and C.-M. Che, Angew. Chem., Int. Ed., 1999, 38, 669; T. Kusakawa and M. Fujita, Angew. Chem., Int. Ed., 1998, 37, 3142; F. S. McQuillan, T. E. Berridge, H. L. Chen, T. A. Hamor and C. J. Jones, Inorg. Chem., 1998, 37, 4949; P. Jacopozzi and E. Dalcanale, Angew. Chem., Int. Ed. Engl., 1997, 36, 613; M. A. Houghton, A. Bilyk, M. M. Harding, P. Turner and T. W. Hambley, J. Chem. Soc., Dalton Trans., 1997, 2725. 4 H. Adams, N. A. Bailey, T. N. Briggs, J. A. McCleverty and H. M. Colquhoun, J. Chem. Soc., Dalton Trans., 1986, 813. 5 E. J. Schier, W. Sacher and W. Beck, Z. Naturforsch., Teil B, 1987, 42, 1424. 6 A. C. Cope and D. C. Friedrich, J. Am. Chem. Soc., 1968, 90, 909. 7 For examples of helical chirality in metallamacrocycles see: O. Mamula, A. von Zelewsky and G. Bernardinelli, Angew. Chem., Int. Ed., 1998, 37, 290; Y. S. Zhang, S. N. Wang, G. D. Enright and S. R. Breeze, J. Am. Chem. Soc., 1998, 120, 9398. 8 D. G. Hamilton, D. E. Lynch, K. A. Byriel, C. H. L. Kennard and J. K. M. Sanders, Aust. J. Chem., 1998, 51, 441. Communication 9/04857J

ISSN:1477-9226    

DOI:10.1039/a904857j

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Synthesis and characterization of dinuclear complexes of 3,3,4,4- tetraminobiphenyl with tetramminoruthenium and bis(bipyridine)- ruthenium residues and their two- and four-electron oxidized products including a ZINDO study of orbital mixing as a function of ligand oxidation state† Robert A. Metcalfe,‡a Luiz C. G. Vasconcellos,§b Hameed Mirza,a Douglas W. Franco *b and A. B. P.Lever *a ab Department of Chemistry, York University, 4700 Keele Street, Toronto, Ontario, Canada, M3J 1P3 Department of Chemistry, Universidade de São Paulo, Instituto Física e Química de São Carlos, Av. Dr. Carlos Botelho, 1465, Cx. P. 780- CEP 13560-907, São Carlos, SP, Brasil Received 12th April 1999, Accepted 21st June 1999 This paper reports the synthesis, characterisation and reactivity of new complexes of ruthenium, the symmetric dinuclear complex of ruthenium tetrammine [(Ru(NH3)4)2(catH4catH4)](PF6)4 and the asymmetric dinuclear complex of ruthenium bis(bipyridine) and ruthenium tetrammine [Ru(NH3)4(catH4catH4)Ru(bpy)2](PF6)4 where (catH4catH4) represents the bridging ligand 3,3,4,4-tetraminobiphenyl.These complexes can be oxidized to the two-electron oxidized ligand mixed valence (qH2catH4) species and the four-electron oxidized ligand (qH2qH2) species.Further oxidation yields a range of Ru(...) species. These complexes were studied by proton NMR, UV–VIS spectra, cyclic voltammetry and ZINDO/1 and ZINDO/S calculations. 2]2 Introduction Recent work in this laboratory has focussed on the extensive mixing that exists between the metal d orbitals of [Ru(bpy) and [Ru(NH3)4]2 fragments, (bpy = 2,2-bipyridine) and the molecular orbitals of a series of dioxolene, o-phenylenediamine (opda), aminothiophenolate and o-benzoquinonediimine (bqdi) ligands.1–5 The extent of mixing can be altered by changing the oxidation state of the ligand.Two major objectives are: (i) a greater understanding of the electronic nature of strongly coupled metal–ligand systems and (ii) the construction of polymeric versions of these complexes that may be used as molecular wires, which could be switched by altering the oxidation state of the metal or the bridging ligand.The polymeric versions of these complexes will be capable of forming mixed valence species, not only involving the two metal centres, but also within the ligand itself.The neutral 3,3,4,4-tetraminobiphenyl (Fig. 1), abbreviated as (catH4catH4), contains two o-phenylenediamine groups that are bonded together. Each end can be oxidized, in principle independently, to the semiquinonediimine (sqH2) with loss of two protons from each segment (one-electron–two-proton loss per fragment) and then each fragment can be oxidized by one electron to the quinonediimine oxidation state, the neutral qH2.1–6 The Hn subscript number indicates the number of NH protons in each fragment of the molecule.This ligand has been used previously as a .rst step towards † Supplementary data available: pictures of the molecular orbitals of the various species and xyz .les for all of the species.For direct electronic access see http://www.rsc.org/suppdata/dt/1999/2653/, otherwise available from http://www.chem.yorku.ca/profs/lever/blever.htm, BLDSC (No. SUP 57592, 18 pp.) or the RSC Library. See Instructions for Authors, 1999, Issue 1 (http://www.rsc.org/dalton). ‡ Current address: Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada, N2L 3G1.§ Current address: Depto. de Química Orgânica e Inorgânica, Centro de Ciências, Universidade Federal do Ceará, Campus do Pici, Fortaleza, Ce., Cx. Postal 12200, Brasil. Fig. 1 Sketches of the bridging ligands showing the coordinate frame of reference used. The planar ligand lies in the xz plane.Note the orientations of the axis vectors as a guide to the signs used in the wavefunctions. The centrosymmetric conformer of (qH2qH2) of C2h symmetry is shown. By rotating one ring 180 about the central linking C–C bond, one obtains the planar acentric C2v conformer. the synthesis and understanding of a polymeric version of the dioxolene or o-phenylenediamine type complexes, with the synthesis of the mononuclear complex 3 [Ru(bpy)2(qH2catH4)]- (PF6)2 and the symmetric dinuclear [(Ru(bpy)2)2(qH2qH2)]- (PF6)4 complex.4 These species and several of their reduced forms, including a mixed valence quinone–semiquinonate form, were characterized electrochemically and spectroscopically.These studies and previous work in the literature 7 dealing with dinuclear ruthenium complexes of the ligands 3,3,4,4- tetrahydroxobiphenyl, 3,3,4,4-tetrahydroxo-p-terphenyl, and 1,4,5,8-tetrahydroxonaphthalene, and related ligands, involve J.Chem. Soc., Dalton Trans., 1999, 2653–2667 DALTON FULL PAPER 2653very strongly coupled metal¡Vligand interactions. Such species may form the basis for the synthesis of molecular electronic devices.In this contribution, the (catH4catH4) ligand was used to synthesize the dinuclear ruthenium complexes [(Ru(NH3)4)2- (catH4catH4)](PF6)4 and [Ru(NH3)4(catH4catH4)Ru(bpy)2]- (PF6)4 referred to as the symmetric and asymmetric species respectively. Controlled oxidation of the (catH4catH4) species gives rise to a series of complexes with a range of oxidation states whose nature is the subject of this contribution.In addition to the fully oxidized ligand species, [(Ru(NH3)4)2- (qH2qH2)]4 and [Ru(NH3)4(qH2qH2)Ru(bpy)2]4 we demonstrate the existence of complexes containing the mixed valence (qH2catH4) bridge as well as mixed valence Ru()/Ru() species. The complexes are characterized by a wide range of techniques including NMR, EPR, UV¡VVIS spectroscopy, electrospray ionization mass spectrometry (ESI-MS) and electrochemistry. ZINDO/1 geometry optimizations and ZINDO/S spectroscopic calculations 8 were also performed on this series of complexes.Good agreement was achieved, with closed shell species, between observed and ZINDO/S predicted electronic spectra and the electronic structure of these complexes was deduced therefrom. Extensive mixing between ruthenium d£k orbitals and ligand £k and £k* orbitals is observed when either or both ends of the bridging ligand are in the qH2 oxidation state.The extent of mixing is discussed in terms of the net oxidation state of the species. These studies extend the recent contributions 9,10 on ruthenium pyridine, bipyridine and pyrazine coupling adding to our knowledge of the more complex dinuclear species.11 They further conrm the notion that electron rich Ru() fragments such as the [Ru(NH3)4]2 moiety can couple extremely strongly to organic fragments such as benzoquinonediimine through both the £k and £k* frameworks.1¡V4 Experimental All chemical products used were reagent quality or better and were used without any further purication, unless otherwise stated.The water was doubly distilled, the second time from potassium permanganate, and passed through Barnstead activated charcoal and ion-exchange lters. All other solvents were distilled and dried following literature methods.12 Physical measurements Electronic spectra were recorded on a Cary model 2400 spectrophotometer, or on a Hewlett Packard 8452A diode array instrument.Electrochemical measurements were recorded using Princeton Applied Research Corp. models 173, 174 and 179 instruments, or a Pine Instruments RDE-3 potentiostat. Results were obtained using either CH3CN with 0.15 M tetrabutylammonium hexauorophosphate (TBAPF6) or phosphate buer solutions, solutions typically contained 1 ¡Ñ 103 M of the complex.A platinum disk, or a glassy carbon disk served as the working electrode, a platinum wire as counter electrode, and an SCE, or AgCl/Ag coated wire was used as the reference electrode. In cases where AgCl/Ag was used as the reference electrode, ferrocene was added as an internal standard, and the potentials are reported vs. SCE assuming a ferrocenium¡V ferrocene potential of 0.425 V vs.SCE in CH3CN.13 C, H and N microanalyses were performed by the Canadian Microanalytical Service, Vancouver, British Columbia, Canada, or the Laboratrios de Microanlise da Universidade de So Paulo - Sp - Brasil. Nuclear magnetic resonance (NMR) spectra were recorded with a Bruker AM300 NMR spectrometer. Spectra were recorded with acetone-d6 as the solvent and tetramethylsilane (TMS) as standard.The EPR spectra were recorded in a Bruker spectrometer 2654 J. Chem. Soc., Dalton Trans., 1999, 2653¡V2667 Table 1 Key bond distances (pm) of optimized structures d Ru¡VN(bpy) Ru¡VNH3 Species a C NHb Ru¡VNH(bridge)c 133 133 207(am), 209(bpy) 216(ax), 214(eq) 204 203(q), 209(cat) 216(ax), 214(eq) 204 203(am), 204(bpy) 216(ax), 214(eq) 203 216(ax), 214(eq) 207 203(q), 207(cat) 203 216(ax), 214(eq) 216(ax), 214(eq) 133 133 asymcc asymqc asymqq symcc symqc symqq a Asymmetric and symmetric species: cc = catH4catH4, qc = qH2catH4, qq = qH2qH2.b Bridge imine bond length. c Average Ru¡VN (bridge) bond length, am = tetrammine terminus, bpy = bipyridine terminus, q = qH2 terminus, cat = catH4 terminus. d ax = Axial, eq = equatorial, these distances were restrained in the optimization.See ESI for complete structural information as xyz les which may be directly viewed using CHIME. model ESP300e and a microwave bridge model ER 041 XK (X band), coupled to a standard cavity model 941ST380, in a temperature range of 110 ¡Ó 5 K. INDO/1 and INDO/S calculations used the ZINDO method and the Hyperchem software package (v5.1, Hypercube, Florida, USA).Data were processed on a Silicon Graphics Personal Iris Indigo R4000 or a Pentium 120 MHz Intel computer. Structures of all complexes were obtained by using the modied INDO/1 semiempirical method (ZINDO/1) in the Hyperchem program. The recommended value 8b of the ¡¥resonance¡¦ integral parameter for Ru, £](4d) = 26.5 eV overestimates the Ru¡VN bond strength and produces rather short Ru¡VN distances.Better agreement with X-ray structural data for these complexes was obtained by using a value 14 of £](4d) = 20 eV. All other parameters used in ZINDO/1 were the default parameters in the Hyperchem program. Convergence was assumed when the gradient was lower than 30 cal mol1 1. All bond distances and angles in the optimised structures lay within accepted norms for these types of molecule, see Table 1.Full details of the optimised structures are available as electronic supplementary information (ESI) as xyz les which may be directly viewed using CHIME. The ZINDO/S programme used the ruthenium bases provided by Krogh-Jespersen et al.15 Electronic spectra were calculated with single excitation conguration interaction (CIS).16 The number of congurations used was 20 occupied ¡Ñ 20 unoccupied orbitals.Reasonable convergence of calculated transition energies in the visible region was achieved, and increasing the number of congurations had an insignicant eect on the predicted visible region absorption energies. The overlap weighting factors £m¡V£m and £k¡V£k were set at 1.267 and 0.585, respectively.Small variations in these parameter values had no substantive eect on the calculations. Oscillator strengths were calculated in the dipole length approximation including the one-centre sp and pd atomic terms. Electrospray ionization mass spectrometry (ESI-MS) 17,18 Solution electrospray ionization mass spectrometry (ESI-MS) in analytical grade CH3COCH3 or CH3CN was performed using a Sciex TAGA MS/MS spectrometer, at 2.8 kV applied voltage on the electrospray probe.For the ESI-MS data, mass and intensities were compared to those calculated using the Isopro 3.0 version MS/MS software for PC. Synthesis All syntheses were performed under an inert atmosphere, unless otherwise noted. The starting complexes [Ru(NH3)5Cl]Cl2, Ru(bpy)2Cl22H2O and cis-[Ru(NH3)4(H2O)2](CF3SO3)3 were prepared following literature procedures.19,20 The ligand 3,3,4,4-tetraminobiphenyl tetrahydrochloride dihydrate and the complex RuCl33H2O were purchased from Aldrich.Table 2 NMR results of symmetric and asymmetric complexes (s and m represent singlet or multiplet). All spectra were recorded in acetone-d unless otherwise noted Complex [(Ru(NH3)4)2(catH4catH4)]4 [Ru(NH3)4)2(qH2qH2)]4 [Ru(bpy)2(catH4catH4)Ru(NH3)4]4 [Ru(bpy)2(qH2qH2)Ru(NH3)4]4 [(Ru(bpy)2)2(qH2qH2)]4 a [Ru(bpy)2(dadib)]2 b [Ru(NH3)4(bqdi)]2 c a Recorded in DMSO-d6.b Recorded in CD3CN, dadib = 3,4-diamino-3,4-diimino-3,4-dihydrobiphenyl. c bqdi = o-benzoquinonediimine. (a) [(Ru(NH3)4)2(catH4catH4)](PF6)4. cis-[Ru(NH3)4(H2O)2]- (CF3SO3)3 (0.137 g, 0.21 mmol) was dissolved in an aqueous solution (2 mL), pH = 2.0 (CF3CO2H used to adjust the pH) in the presence of zinc amalgam (3.0 g), and the solution was bubbled with argon.After 30 min the Ru() complex had been reduced to Ru() and was transferred to a separate ask. An aqueous deoxygenated solution (3 mL) containing 3,3,4,4- tetraminobiphenyl tetrahydrochloride dihydrate (0.042 g, 0.105 mmol) was added to the solution containing the ruthenium() tetrammine species.The reaction mixture was stirred for 2 h, and then with the addition of 1 mL of a saturated aqueous solution of NH4PF6, a yellow microcrystalline solid precipitated. The yellow solid was collected by ltration under an inert atmosphere, washed with deoxygenated ethanol and diethyl ether, dried and stored under vacuum in a desiccator.Yield 70% (Calc. for C12H34F24N12P5Ru2: C, 12.77; H, 3.04; N, 14.89. Found: C, 12.47; H, 3.19; N, 13.08%). Diculty was experienced in obtaining a satisfactory nitrogen analysis. (b) [Ru(NH3)4(catH4catH4)Ru(bpy)2](PF6)4. The complex was prepared by adding Ru(bpy)2Cl2H2O (0.1 g, 0.19 mmol) in deoxygenated water, 10 ml, pH = 2.0, containing 3,3,4,4- tetraminobiphenyl tetrahydrochloride dihydrate (0.075 g, 0.19 mmol).The mixture was heated at 40 C for 24 h, during which time the initial purple¡Vblack suspension became an orange solution. At this point, the mononuclear complex [Ru(bpy)2- (catH allowed to cool to room temperature. cis-[Ru(NH 4catH4)]2 had been formed in solution. This was then 3)4(H2O)2]- (CF3SO3)3 (0.12 g, 0.19 mmol) was dissolved in deoxygenated water (2 mL), pH = 2.0, in the presence of zinc amalgam (3.0 g).After 30 min, during which time the Ru() complex was reduced to Ru(), the solution was transferred into the ask which contained the [Ru(bpy)2(catH4catH4)]2 complex. This mixture was stirred for 4 h, at room temperature. Then, with the addition of 1 mL of a saturated aqueous solution of NH4PF6, an orange solid precipitated from the solution.The solid was collected by ltration, under an inert atmosphere, and washed with degassed ethanol and diethyl ether. The product was dried and kept under vacuum in a desiccator. Yield 68% (Calc. for C32H38F24N12P4Ru2: C, 27.99; H, 2.79; N, 12.24. Found: C, £_ (ppm) 2.48 (s) 2.68 (s) 5.81 (s) 7.52 (m), 7.69 (s) 2.48 (s) 4.51 (s), 4.65 (s) 7.38 (d), 7.57 (d), 7.89 (s) 12.22 (s), 12.26 (s) 2.48 (s) 4.63 (s), 4.54 (s) 6.25 (m), 6.75 (m) 7.1¡V9.2 2.51 (s) 4.48 (s), 4.77 (s) 7.25¡V8.85 13.0 (s), 12.9 (s), 12.4 (s), 12.25 (s) 7.27¡V8.50 12.08 (s), 11.98 (s) 3.84 (s), 4.18 (s) 6.70¡V8.48 11.32 (s), 11.78 (s) 2.30 (s) 4.60 (s) 6.85 (d), 7.45 (d) 12.25 (s) 28.15; H, 3.00; N, 11.43%).Diculty was experienced in obtaining a satisfactory nitrogen analysis. The oxidation of the fully reduced (catH4catH4) solids by the addition of 1¡V5 NH3 and O2 yielded solutions of the symmetric and asymmetric species in the Ru()(qH2qH2)Ru() oxidation state as indicated by their electronic spectra.However the solids isolated by this route were EPR active, giving a Ru() signal. In solution these oxidized complexes are EPR silent.Evidently the solid state favours formation of a Ru() species electronic isomer which is not observed in the solution oxidation process under similar conditions (NH3,O2). The reactivity and characterization of these oxidized solids are the subject of a future contribution.21 Results NMR and ESI-MS spectroscopy Details of the NMR data for these complexes and related species are listed in Table 2.[(Ru(NH3)4)2(catH4catH4)]4. In this symmetric, reduced complex, the structure is conrmed by the two sets of resonances for the equivalent pairs of axial and equatorial amine protons, and the signals, at £_ 7.52 and 7.69 that are assigned to the protons of the (catH4catH4) ligand. The NMR spectra of axially and equatorially oriented NH3¡VRu protons have been extensively discussed.11b,22 [(Ru(NH3)4)2(qH2qH2)]4. The NMR spectrum of this symmetric oxidized species in consistent with the proposed structure.The position of the axial NH3 protons is unchanged compared to the fully reduced complex while the equatorial NH3 protons are shifted downeld. These protons are deshielded by the relatively electron decient (qH2) fragment and appear as two signals because the two imine functions are inequivalent. The existence of a singlet and two doublets in the aromatic region accounts for the protons of the (qH2qH2) ligand.The imino protons appear as two singlet resonances located downeld at £_ 12.22 and 12.26 which is consistent with the NMR spectra of related complexes.1¡V5 J. Chem. Soc., Dalton Trans., 1999, 2653¡V2667 6 Assignment Axial NH3 Equatorial NH3 (catH4catH4) amino (catH4catH4) ring protons Axial NH3 Equatorial NH3 (qH2qH2) ring protons Imine protons Axial NH3 Equatorial NH3 (catH4catH4) amino bpy and (catH4catH4) ring protons Axial NH3 Equatorial NH3 (qH2qH2) and bpy protons Imine protons (qH2qH2) and bpy protons Imine protons dadib amino protons dadib and bpy protons Imine protons Axial NH3 Equatorial NH3 bqdi ring protons Imine protons 2655The corrected ESI-MS spectrum of this symmetric (qH2qH2) species in CH3CN shows the presence of several quadruply charged species.The parent ion was observed at m/z 138.25 as an envelope of peaks associated with the ruthenium isotopes. In addition, peaks were observed at m/z 148.5, 158.75, 169 and 179.25 corresponding to the cluster formation of the symmetric species with one, two, three and four CH3CN molecules respectively.The isotopic distribution of these species was in good agreement with the calculated composition. Such cluster assemblies are well known in ESI-MS spectroscopy.17 Other relevant species that could be identi.ed are at m/z 164.75 and 175 corresponding to the loss of one NH3 group from three and four CH3CN clusters respectively.[Ru(NH3)4(catH4catH4)Ru(bpy)2)]4. The NMR spectrum of this asymmetric ion is complicated by its low symmetry. There are a large number of peaks in the aromatic region which correspond to the bpy 23 protons and the protons of the (catH4 catH4) ligand. The amino groups of the (catH4catH4) ligand appear as two multiple signals.All four amino groups are shifted down.eld, compared to the symmetric dinuclear complex [(Ru(NH3)4)2(catH4catH4)]4. The amino groups appear as multiple signals for two reasons. In the asymmetric complex the four amino groups are inequivalent and there is also a chiral [Ru(bpy)2]2 centre in this complex whose presence makes the protons of the amino groups diastereotopic.24 These factors put each proton of each amino group into a magnetically di.erent environment, which produces coupling between the protons on each amino group.Considering the appearance of the NMR spectrum of the [Ru(bpy)2(dadib)]2 complex, listed in Table 2, which shows two di.erent amino signals, it seems likely that the 3- and 3-amino groups will be similar and the 4- and 4-amino groups will be similar.The axial ammines of the [Ru(NH3)4]2 fragment are in almost the same location as they are in the symmetric complex, but the equatorial ammines are shifted down.eld relative to the symmetric complex and appear as two signals. The cause of this shift may be the [Ru(bpy)2]2 fragment which acts as an electron-withdrawing group compared to [Ru(NH3)4]2.The inequivalent amino groups will shield the amine ligands to slightly di.erent degrees producing two signals. 6 [Ru(NH3)4(qH2qH2)Ru(bpy)2]4. The NMR spectrum of this asymmetric oxidized species is consistent with the proposed structure. The axial NH3 groups are essential in the same position as they are in every complex in this series, while the equatorial ammine groups are shifted down.eld, relative to the (catH4catH4) complex, as they are in the symmetric species.The protons of the bpy and phenyl rings of the (qH2qH2) ligand appear as multiplets in the aromatic region. The imino protons on the Ru–bpy fragment and the Ru–NH3 fragment each appear as two singlet resonances with a 1 : 1 intensity ratio. The corrected ESI-MS spectrum of this asymmetric (qH2qH2) species in acetone shows the formation of several doubly charged species.The predominant species which were observed at m/z 396.5, 388, 379.5, 371, 363.5 correspond to the 2 charged species and those with the loss of one, two, three and four NH3 groups respectively.The isotopic distribution observed for these species is in good agreement with the calculated composition. Formation of lower charged complexes in the ESI-MS has previously been observed 18 to occur via proton abstraction by ions such as F, which is a fragment ion of PF . We expect a similar process to occur in our case, as the protons on the ammine groups are relatively acidic and thus could react with F from the PF counter ion forming HF. 6 Redox chemistry (a) Electrochemistry. The typical electrochemical behaviour for complexes of this type of ligand 1–5 in the quinonediimine 2656 J. Chem. Soc., Dalton Trans., 1999, 2653–2667 Fig. 2 Cyclic voltammogram of [(Ru(NH3)4)2(qH2qH2)]4 in acetonitrile solution (0.15 M TBAPF6).Scan rate 100 mV s1. oxidation state, when investigated using cyclic voltammetry shows two reversible one-electron reduction processes at the ligand to form .rst the semiquinonediimine, and then the deprotonated diamide (only one proton per amino group, (catH2) as opposed to two) oxidation states. The metal may also be oxidized from Ru(..) to Ru(...). In the case of symmetric and asymmetric ruthenium complexes of the (catH4catH4) ligand having two o-phenylenediamine ligands bonded together, no well de.ned ligand centred redox processes were observed in the cyclic voltammetry in organic solvents.All processes appeared to be irreversible in the several di.erent solvents explored. However oxidation of the metal centres in these complexes can be seen in the CV experiments.Fig. 2 shows the oxidation of the oxidized symmetric product; [(Ru(NH3)4)2(qH2qH2)]4 in acetonitrile. This complex has one process at 0.86 V (vs. SCE), which corresponds to the oxidation of one ruthenium atom from Ru(..) to Ru(...), to generate the mixed valence [(NH3)4RuII(qH2qH2)RuIII(NH3)4]5 (abbreviated as [II,III- (qH2qH2)] species.A second oxidation process occurs at 1.04 V (vs. SCE), corresponding to the oxidation of the second ruthenium atom to generate the [III,III-(qH2qH2)]. The small signal at 0.65 V likely represents the reduction of a decomposition product, as this peak is not observed until potentials more positive of the .rst oxidation are reached. The .E between the [II,III-(qH2qH2)] and [III-III(qH2 qH2)] redox processes, is 0.18 V which is 60 mV greater than the corresponding processes in the symmetric bipyridine species 4 [(Ru(bpy)2)2(qH2qH2)]4.The increase in .E value re.ects greater electron delocalization between the two Ru(..) centres due to replacing the bpy ligands with the more basic NH3 ligands. This value of .E is comparable to that observed for the 2,2-bipyrimidine bridged bis [Ru(NH3)4]2 dinuclear complex (0.19 V) 25 and is much larger than the .E values observed for many of the complexes of monodentate bridging ligands (.E = 50–80 mV), with the exception of pyrazine (.E = 0.39 V).26 The value of .E allows one to calculate the comproportionation constant, Kc for the following comproportionation reaction [eqn.(1)]. [II,II-(qH2qH2)]4 [III,III-(qH2qH2)]6 2 [II,III-(qH2qH2)]5 (1) Kc has a value of 1.1 × 103 which is approximately 10 times c value of 100 calculated for [(Ru(bpy)2)2(qH2qH2)]4 the K based on the reported .E.5 The asymmetric product [Ru(NH3)4(qH2qH2)Ru(bpy)2]4 has its .rst oxidation at 0.87 V (vs. SCE) which is assigned as oxidation of the ruthenium tetrammine fragment from Ru(..) toRu(), to generate the [RuIII(NH3)4(qH2qH2)RuII(bpy)2]5 mixed valence species.The second oxidation occurs at 1.42 V (vs. SCE) and corresponds to the oxidation of the [Ru(bpy)2] fragment from Ru() to Ru(). There are some small bumps in the region of 1.0 V which are likely due to decomposition products of the mixed valence species.The splitting between the formation of the [RuIII(NH3)4(qH2qH2)RuII(bpy)2]5 species and the [RuIII(NH3)4(qH2qH2)RuIII(bpy)2]6 species is 0.55 V, approximately 130 mV less than in the asymmetric dinuclear complex of bipyrimidine.25 For the asymmetric complex the evaluation of Kc based upon E does not have the same meaning as it does for a symmetric dinuclear complex but also reects the signicant dierence in the local environment of the two metal centres.This is undoubtedly a valence trapped mixed valence species. The complex, in the frozen solution state, exhibits a broad poorly resolved EPR signal close to g = 2. Some additional data for these species are presented below. (b) Chemical oxidation/spectroelectrochemistry. As indicated, these species potentially have ve oxidation states accessible to the bridging ligand (two pairs of catH4 ¡÷ sqH2 ¡÷ qH2), two oxidation states readily accessible to the metal centre (Ru()/Ru()) and in the case of the asymmetric species, additional oxidation states associated with reduction of the bipyridine fragments.We are concerned with characterizing as many of these oxidation states as possible and assessing their electronic structures.This can in principle be achieved by controlled potential oxidations and reductions, and by chemical oxidation or reduction. Both of these techniques have been employed and yield similar data. The chemical oxidations can be controlled readily by adding specic equivalents of oxidizing agent to the RuII(catH4catH4)RuII species, dissolved in 0.1 M phosphoric acid, and these data will be described rst.The individual species can be dened through their electronic spectra. In both the symmetric and asymmetric species the initial starting material can be described by RuII(catH4catH4)RuII and fourelectron oxidation will yield the species RuII(qH2qH2)RuII. In principle one might expect to see the successive oxidation of the bridging ligand as one proceeds from RuII(catH4catH4)RuII with one, two, three and four equivalents of oxidizing agent. The spectra obtained upon successive step-wise oxidation of the symmetric and asymmetric (catH4catH4) species are shown in Fig. 3 and 4. These spectra can be demonstrated to be composites of the spectrum of the starting material [RuII(catH4 catH4)RuII], the spectrum of the nal product, [RuII(qH2 qH2)RuII] and the spectrum of only one intermediate species which we demonstrate below is generically [RuII(qH2catH4)- RuII].No evidence was ever found for a semiquinonate species (as identiable through its electronic spectrum), nor were any bands associated with ligand to Ru() LMCT observed (up to four-electron oxidation per dinuclear molecule).The Ru ¡÷ sq transition usually occurs as an intense band at very low energies, for example in the [(Ru(bpy)2)2(sq,sq)]2 complex the Ru ¡÷ sq transition is at 9260 cm1, while in the [(Ru(bpy)2)2(q,sq)]3 complex it is at 8160 cm1.4 The Ru() spectra of related species,1 have a narrow intense transition at ca. 24000 cm1. No evidence of either of these transitions was observed in the rst four oxidation steps from [RuII(catH4 catH4)RuII].This is true irrespective of whether the system was oxidized with ceric ion [Ce()] or S2O82. The intermediate species can then only be [RuII(qH2catH4)RuII] and we demonstrate below that its formation and electronic spectrum are consistent with this formulation. There are two geometric isomers of the asymmetric intermediate [RuII(NH3)4(qH2catH4)RuII(bpy)2]4 and [RuII(NH3)4- (catH4qH2)RuII(bpy)2]4. Oxidation of the RuII(catH4..) fragment to RuII(qH2..) proceeds via formation of a RuIII(catH4..) intermediate (see detail in next section below).Previous literature experience 1a,4a,27 reveals that the RuII(NH3)4(catH4..) fragment is very much easier to oxidize than the RuII(bpy)2(catH4..) Fig. 3 Ceric titration of a 4.1 ¡Ñ 105 M solution of [(Ru(NH3)4)2- (catH4catH4)](PF6)4 in 0.1 M H3PO4 showing the addition of 0¡V4 equivalents of Ce(). Inset; data for 4¡V6 equivalents of Ce() as designated. Fig. 4 Ceric titration of a 2.2 ¡Ñ 105 M solution of [Ru(bpy)2(catH4 catH4)Ru(NH3)4](PF6)4 in 0.1 M H3PO4 showing the addition of 0¡V4 equivalents of Ce(). fragment and we conclude therefore that the isomer generated is the former of the two cited above.Moreover a ZINDO/1 derivation of the heat of formation of the two isomers shows that the former is some 140 kcal mol1 (in a total of 16840 kcal mol1) more stable than the latter. Thus after one equivalent of ceric ion (in 0.1 M H3PO4) has been added (one electron per dinuclear species) the solution contains 50% unchanged RuII(catH4catH4)RuII and 50% RuII(qH2catH4)RuII, after two equivalents 100% RuII(qH2 catH4)RuII, after three equivalents 50% each of RuII(qH2 catH4)RuII and RuII(qH2qH2)RuII and nally 100% RuII(qH2qH2)RuII after four equivalents have been added. The observed composite spectra of the one- and three-electron oxidized solutions can be simulated precisely by adding 50% of the spectrum of the fully reduced complex to 50% of the spectrum of the two-electron oxidized species and by adding 50% of the spectrum of the two-electron oxidized species to 50% of the spectrum of the four-electron oxidized species, respectively.Further oxidation by ceric ion leads successively to the symmetric [RuIII(NH3)4(qH2qH2)RuII(NH3)4]5 and [RuIII- (NH3)4(qH2qH2)RuIII(NH3)4]6 species.With the asymmetric 2657 J. Chem. Soc., Dalton Trans., 1999, 2653¡V26673)4(qH2 2)RuII(bpy)2]5 is feasible, but the Ce() ion is incapable of species, one-electron oxidation to form [RuIII(NH qH oxidizing the RuII(bpy)2 fragment to the Ru() product. The spectroelectrochemical experiments were carried out beginning with the [RuII(qH2qH2)RuII]4 species and sequentially reducing back to [RuII(catH4catH4)RuII]4. The electronic spectra so obtained by controlled potential electrolysis (reduction) were the same as those obtained by appropriate chemical oxidation (Fig. 3 and 4) but obtained in the inverse sequence.Fig. 3 (insert) shows the Ce() ion oxidation of the symmetric species beginning with the [RuII(qH2qH2)RuII]4 species. The long wavelength tail on the [RuII(qH2qH2)RuIII]5 species is very evident and is likely attributed to the expected intervalence band since this transition is not otherwise evident out to 1600 nm.This shoulder disappears in the spectrum of the [III,III- (qH2qH2)] species which, however, does show a long wavelength tail which must still encompass several transitions. Controlled potential oxidation of the [RuII(qH2qH2)RuIII]5 species in acetonitrile (not shown) gave fairly similar spectra shifted somewhat because of the non-aqueous and aprotic solvent.1d (c) Mechanism of oxidation.The absence of any semiquinonediimine species in the oxidation, chemical or electrochemical, of the (catH4catH4) species, raises the question of the oxidation pathway which may be involved.The oxidation of a (catH4) to a (sqH2) species involves both electron and proton loss and is generally electrochemically irreversible.1 This is an example of oxidative dehydrogenation and the general mechanism via a higher oxidation state of the metal is very well documented.28 If an acidic solution is used for the chemical oxidation, proton loss will be inhibited and the oxidation of a coordinated ¡VNH2 residue will lie at quite a high potential.The diamine fragment can be considered a substituted ammonia which would have a ligand electrochemical parameter value,27 EL(L), near 0 V; thus oxidation of [RuII(NH3)4(catH4..)] to [RuIII(NH3)4(catH4..)]3 is expected to occur before oxidation of (catH4), with formation of [RuIII(catH4catH4)RuII]5. Evidence for this and a subsequent Ru(III) species has been obtained by EPR spectroscopy at low temperature.Thus a sample of the asymmetric [RuII(NH3)4(catH4catH4)RuII- (bpy)2]4 in 0.1 M phosphoric acid was frozen at liquid nitrogen temperature and contacted sequentially with 1¡V4 equivalents of ceric ion. The frozen solution was melted to allow for mixing and then it was refrozen. At liquid nitrogen temperature, EPR spectra consistent with the formation of a Ru() species 29 (and not of a free radical sqH2 species) were observed upon addition of one (g¡æ = 2.60, g|| = 1.59) or three (g¡æ = 2.57, g|| = 1.71) Ce() equivalents, but no signal with two or four equivalents.Upon warming to room temperature these signals disappeared. The proposed sequence of steps involved in the oxidation is as follows.(i) Addition of one equivalent of oxidant produces the following reaction [eqn. (2)]. [RuII(NH3)4(catH4catH4)RuII(bpy)2]4 e [RuIII(NH3)4(catH4catH4)RuII(bpy)2]5 (2) (ii) When warmed towards room temperature the 5 species disproportionates [eqn. (3)]. 2 [RuIII(NH3)4(catH4catH4)RuII(bpy)2]5 2H [RuII(NH3)4(catH4catH4)RuII(bpy)2]4 [RuII(NH3)4(qH2catH4)RuII(bpy)2]4 (3) (iii) After addition of a second equivalent of oxidant the solution is 100% [RuII(NH3)4(qH2catH4)RuII(bpy)2]4. (iv) The addition of a third equivalent of oxidant produces the following reaction [eqn.(4)]. 2658 J. Chem. Soc., Dalton Trans., 1999, 2653¡V2667 [RuII(NH3)4(qH2catH4)RuII(bpy)2]4 e [RuIII(NH3)4(qH2catH4)RuII(bpy)2]5 (4) (v) When warmed towards room temperature this species will disproportionate [eqn.(5)]. 2 [RuIII(NH3)4(qH2catH4)RuII(bpy)2]5 2H [RuII(NH3)4(qH2catH4)RuII(bpy)2]4 [RuII(NH3)4(qH2qH2)RuII(bpy)2]4 (5) (vi) The fourth equivalent of oxidant will convert the solution to 100% [RuII(NH3)4(qH2qH2)RuII(bpy)2]4. This proposed sequence of steps is consistent with the aforementioned observation of a Ru() EPR signal 29 at low temperature for the one- and three-electron oxidized species and no signal for the two- and four-electron oxidized species.Similar experiments were performed on the symmetric species, but no EPR signals were seen for the addition of any number of equivalents of oxidant. This may be due to a more rapid disproportionation of the Ru() species. ZINDO Calculations ZINDO/1 calculations 8 were performed on this series of complexes in their dierent states of protonation and oxidation.The dihedral angle of the bridging ligand was free to rotate. The symmetric and asymmetric complexes optimized with dihedral angles of 69, 44 and 0 and 46, 35 and 4 respectively for the (catH4catH4), (qH2catH4) and (qH2qH2) species. The size of the dihedral angle is governed by competition between the steric repulsion of the 2,6- and 6,2-hydrogen atoms and any electronic demands which might favour co-planarity. The (qH2qH2) oxidation state can be expected to have a dihedral angle close to zero in order to maximise delocalization and ruthenium-bridge mixing.Delocalization and rutheniumbridge mixing across the entire molecule are less likely in the (catH4catH4) and (qH2catH4) oxidation states, so steric repulsion will become the dominant factor and the dihedral angle will increase.ZINDO/1 was used to derive heats of formation as a function of dihedral angle, primarily to assess how sensitive the heat is to the dihedral angle. In both the symmetric and asymmetric series of (qH2catH4) and (catH4catH4) the dependence is very at with only a few kcal mol1 separating the various conformers.We report below the optimized geometries but recognise that they may not be precisely accurate. The symmetric (qH2qH2) shows the deepest well with the at conformer, at 8846 kcal mol1, about 50 kcal mol1 more stable than the 90 twisted conformer. The asymmetric (qH2qH2) species shows similar behavior. Note however that there are two at conformers depending upon the relative orientation of the two pairs of NH groups (see Fig. 1). One conformer has a centrosymmetric bridge and the other does not. For the symmetric (qH2qH2) species, these have C2h and C2v symmetry respectively. ZINDO/ 1 favours the centrosymmetric conformer in both the symmetric and asymmetric (qH2qH2) species but only by a few kcal mol1. We assume henceforth that the centrosymmetric conformer is relevant for all the species under discussion.The ZINDO/S single point calculations were carried out on the optimized structures even though in solution at room temperature, some rotation may occur about the central C¡VC bond. (a) Free ligands. ZINDO/S calculations were performed on the free ligands using the optimized geometries that they have in the complexes.The at (qH2qH2) ligand has C2h symmetry, the twisted (qH2catH4) has C1 symmetry and the twisted (catH4 catH4) has C2 symmetry. The frontier orbitals of the (qH2qH2) ligand include two pairs of lled £k orbitals of bg and au sym-Fig. 5 A selection of ZINDO/S calculated frontier orbitals for the free (qH2qH2) ligand of C2h symmetry. metry (#39,38,36,32) and three empty (£k*) levels, a bg orbital (LUMO 1, #41) and a pair of au orbitals (LUMO #40, 42).Orbitals #40 and #41 are clearly the in and out of phase coupling of the £k* levels of the two qH2 fragments respectively (Fig. 5). Orbitals #36,42 and #43,44 have local density favoring opposite pairs of The (qH NH groups. 2catH4) ligand has frontier £k and £k* orbitals mostly spread over the entire bridge but with local symmetry at the qH2 terminus very similar to that of the (qH2qH2) bridge.The orbitals of the (catH4catH4) ligand do not mix signicantly with the metal d-orbitals. (b) Complexes. It is convenient to dene on each Ru center, a 2v xz framework (Fig. 1) which would be appropriate for the local C environment of the qH2 ligand fragment. The bridge plane is then xz, and orbital dyz is aligned to £k-bond to a local qH2 symmetric fragment, dx2 z2 lies in the qH2 plane and has £m- symmetry, dxy has £_-symmetry with respect to qH2; these then are lled with six electrons in the Ru() species.The empty d and d2y2 x2 z2 point along Ru¡VN bond vectors and are £m*- antibonding. Considering most simply the symmetric (qH2qH2) species of C2h symmetry, symmetry adapted orbitals comprising both Ru centres can be constructed from (dxy ¡Ó dxy), (dyz ¡Ó dyz) and (dx2 z2 ¡Ó dx2 z2) where the primed and unprimed orbitals reside on dierent Ru centres.It is also possible to mix these orbitals on a given Ru atom, namely (d xy ¡Ó xy ¡Ó dyz) so generating two orbitals of £k-symmetry lying along the Ru¡VNH (qH2..) bond vectors.These too can linearly combine across the bridge to form symmetry adapted wavefunctions namely {(d dexclusively as b as a yz) ¡Ó (dxy dyz)}. The £k-orbitals described here transform g if they are even with respect to inversion, and u, if they are odd. They can then mix with the au and bg £k-orbitals of the bridging ligand to provide a mechanism for coupling the remote metals together.The (d combinations which lie in the bridge plane with £m-symmetry, transform as a shown in Fig. 6. Others can be found as ESI. x2 z2 ¡Ó dx2 z2), g bu. Some representative MO pictures are In the low symmetry (C1) asymmetric species coupling across the bridge can be assessed by the relative contributions of each Ru d orbital to the MOs. The resulting MO diagrams are shown in Fig. 6 while the fractional mixing in the frontier orbitals of these complexes can be seen pictorially in Fig. 7 (symmetric) and 8 (asymmetric), with energies displayed in Fig. 9 and 10. The asymmetric (qH2catH4) appears a lot like a mononuclear Ru(NH3)4(qH2) species,1a with a substituent lowering its symmetry. Mixing, symmetry and energy data are collected in Table 3.Orbital mixing One important theme in past studies of these molecules is that of orbital mixing, and the extent of metal¡Vligand coupling. The extent of metal¡Vligand and metal¡Vligand¡Vmetal coupling can be probed as a function of the oxidation state of the bridging ligand. In the symmetric species, where the metal centres are chemically equivalent, i.e. [(NH3)4Ru(catH4catH4)Ru(NH3)4]4 and [(NH3)4Ru(qH2qH2)Ru(NH3)4]4, the dierence in energy between the in- and out-of phase coupled pairs of equivalent d orbitals on each ruthenium is a measure of the extent of communication or coupling between the metal centres.In high symmetry such as D2h, this splitting can be approximately related to the electronic matrix coupling element (2Hab).30 With [(NH3)4Ru(qH2qH2)Ru(NH3)4]4 the C2h symmetry allows for some additional mixing of levels so that this energy dierence can be used as a measure of communication but is not directly equal to 2Hab. 2659 J. Chem. Soc., Dalton Trans., 1999, 2653¡V2667Fig. 6 A selection of ZINDO/S calculated frontier orbitals for (top) the symmetric (qH2qH2) series of complexes and (bottom) the asymmetric (qH2qH2) series of complexes.Additional examples can be found as electronic supplementary information. (a) (catH4catH4) oxidation state. In this oxidation state the extent of metal�Cligand mixing is very small. The symmetric complex has the six t2g orbitals (HOMO�CHOMO 5, #79�C #74) essentially unmixed (¡Ü 3%) with the bridging ligand (we use the octahedral t2g label for convenience). These molecular orbitals are the in- and out-of-phase pairs of the atomic orbitals on each metal centre, combined by symmetry, rather than by any interaction, i.e.there is little if any communication between the orbitals on each ruthenium centre. The HOMO and HOMO 1 (#79,78) are the ¦Ò-symmetry d(t2g) orbitals. The energy separations between the in- and out-of-phase coupled pairs is less than 0.005 eV and this may just reect slight asymmetry in the molecule following ZINDO optimization.3)4] fragment. In the asymmetric species, the t2g set of the Ru(bpy)2 fragment (HOMO�CHOMO 2, #121�C#119) are mixed (ca. 20%) with bpy ¦�-orbitals, and unmixed (< 2%) with the bridging ligand orbitals. There are two bipyridine ¦�-orbitals (#117,118) that lie, in energy, between the d-orbitals of the [Ru(bpy)2] fragment and the [Ru(NH The t2g set of the [Ru(NH3)4] fragment lies at deeper energy and two of these three orbitals are highly localized on ruthenium (ca. 90�C100%).However there is a weak interaction between the dyz combination and a bridge ¦�-orbital (#113, 115) which does show some signicant Ru d-bridge mixing probably owing to a good energy match.The splitting (interaction) energy is very small (ca. 0.09 eV). ¦� (b) (qH2catH4) oxidation state. In this oxidation state, signi- cant metal�Cligand mixing with the qH2 end of the bridge is anticipated. The symmetric species optimizes with a large twist angle of 44 (Table 3). The HOMO (#78) is a mix of Ru d (qH2) mixed with ¦�-bridge, specically an antibonding interaction, (dyz dxy) ¦�.The HOMO 1 (#77) is nearly pure d¦Ò-oriented orbital on the catH4 end. HOMO 2 and HOMO 3 (#76,75) orbitals are in- and out-of-phase mixtures Fig. 7 ZINDO/S calculated fractional mixing for the symmetric series of complexes. 2660 J. Chem. Soc., Dalton Trans., 1999, 2653�C2667Fig. 8 ZINDO/S calculated fractional mixing for the asymmetric series of complexes. Fig. 9 Relative molecular orbital energies of the symmetric series as labelled. The HOMO is indicated by H and the LUMO by L. The catH4 localized Ru d(t2g) orbitals are displayed with hatched lines in the catH4catH4 and qH2catH4 species. of d orbitals on both ends of the molecule with the bridging ligand.However these two orbitals are essentially degenerate and there is no signi.cant coupling across the bridge. The HOMO 4, (#74) orbital is nearly pure dyz orbital on Fig. 10 Relative molecular orbital energies of the asymmetric series as labelled. The HOMO is indicated by H and the LUMO by L. In this case, the bipyridine localized orbitals are dotted, and the Ru d(t2g) bpy localized orbitals by a hatched line.The solid lines indicate orbitals which are fairly extensively mixed as seen in Table 3. the catH4 end of the molecule, while HOMO 5 (#73) is a fairly pure s-oriented orbital on the qH2 end (Table 3). By comparison with the asymmetric (qH2catH4) species discussed below, it is surprising that the HOMO is localized on the qH2 end of the molecule. This is seen to arise from a strong interaction between a bridge p-orbital coupling with (dyz dxy) to provide the stabilized Ru–LL bonding combination HOMO 6 (#72) and the destabilized Ru–LL antibonding combination HOMO (#78).The asymmetric (qH2catH4) species optimizes with an appreciable twist angle of ca. 35. The t2g set of the Ru(bpy)2 fragment (which is bound to the (catH4) end of the bridge) forms the HOMO, HOMO 1 and HOMO 2 (#120–#118) and is mixed with the bpy p* orbitals, but not mixed with the bridging ligand.Two bpy orbitals (#116,117) lie between these Ru(bpy)2 d(t2g) orbitals and the Ru(NH3)4 d(t2g) orbitals. HOMO 5, 6 (#115,114) comprise dxy p and (dyz dxy) p respectively (albeit rather distorted by the low symmetry) while the s–d orbital is HOMO 7 (#113).The (dyz dxy) orbital again mixes strongly with a localized qH2 bridge p-orbital forming HOMO 8 (#112) and HOMO 5 (#115) which are Ru–LL bonding and antibonding respectively. This last orbital is only some 50% metal localized as a consequence of this extensive mixing.The LUMO is an antibonding combination of the dyz orbital from the Ru(NH3)4 fragment and p* orbital (LUMO, #41 of (qH2catH4) free ligand) of the bridging ligand localized on the qH2 end and contains substantial Ru d character. Thus there are components of the d(t2g) orbitals from [Ru(NH3)4]2 in energy both below and above the [Ru(bpy)2]2 d(t2g) orbitals (Table 3, Fig. 8). The observation of Ru d(t2g) NH3 lying appreciably below Ru d(t2g) bpy is surprising since the former Ru is easier to oxidise than the latter.27 This must re.ect the relative stability of the resulting Ru(...) species; a DFT calculation gave a similar result which is being further analysed. (c) (qH2qH2) Oxidation state. In this oxidation state the coupling is very extensive between metal dp and both ligand p and p* orbitals. There is also extensive coupling across the bridge since the dihedral angle at the bridge is essentially zero.However conjugation through the bridge to connect the two ruthenium atoms can only proceed through the ‘meta’ NH groups. In the symmetric species of assumed symmetry C2h, the dp orbitals of the two Ru centres are coupled by mixing with the ligand au and bg orbitals.Indeed, the d-orbital density of the six t2g orbitals of the two metal centres is distributed over at least 10 orbitals from LUMO 1 to HOMO 8 (#79 to #69) (Table 3, Fig. 8). The HOMO 3 and HOMO 4 (#74,73) are the s dx2 z2 orbital combinations which are almost pure d 2661 J.Chem. Soc., Dalton Trans., 1999, 2653–2667Table 3 Percentage mixing in the symmetric and asymmetric dinuclear complexes in their di.erent states of protonation and oxidation %Ru(A2) %Ru(A1) [(Ru(NH3)4)2(catH4catH4)]4 Orbital 0.5 0 95 63 31 40 73 74 75 76 77 78 0.5 950 32 63 55 79 (HOMO) 80 81 82 83 550000 400000 [(Ru(NH3)4)2(qH2catH4)]4 Orbital %Ru(A/catH4) %(Ru(A/qH2) 0095 27 68 93 72 73 74 75 76 77 19 940 47 192 78 (HOMO) 79 80 81 82 20000 61 20000 %Ru(A2) %Ru(A1) [(Ru(NH3)4)2(qH2qH2)]4 Orbital 5813 34 45 49 5813 33 49 45 69 70 71 72 73 74 36 36 28 129.5 0.5 36 36 28 129.5 0.5 75 76 77 (HOMO) 78 79 80 81 0 0 %Ru(bpy) %Ru(A) [Ru(NH3)4(catH4catH4)Ru(bpy)2]4 Orbital 31 92 72 113 114 115 116 117 118 119 120 121 (HOMO) 122 123 124 125 00002284 81 740260 90000000000 [Ru(NH3)4(qH2catH4)Ru(bpy)2]4 %Ru(bpy) %Ru(A) Orbital 27 94 67 112 113 114 115 116 117 00002285 79 73 118 119 120 (HOMO) 5000000 2662 J.Chem. Soc., Dalton Trans., 1999, 2653–2667 %Bridge %NH3 99222333100 100 100 100 03333220000 %Bridge %NH3 803225 103 35 79 100 100 100 13313221000 %Bridge %NH3 87 84 73 314426 26 42 75 80 99 3112222221100100 %bpy %Bridge 000098 98 15 675 26800111 18 250 97 930 10010100 %bpy %Bridge 723 31 000098 98 14 19 26 5000111 Energy/eV 18.183 17.575 17.574 17.511 17.509 17.469 17.465 9.548 9.317 9.139 9.01 Energy/eV Description a (dyz dxy)(Ru-qH2) p ds(Ru-qH2) dyz(Ru-catH4) (dyz dxy)(Ru-qH2) dH2) dxy(Ru-catH4) 18.511 17.853 17.617 17.582 17.551 17.516 ds(Ru-catH4) (dyz dxy)(Ru-qH2) p p* dyz(Ru-qH2) 17.412 12.035 9.738 9.357 8.983 Description symmetry Energy/eV (dyz dyz) p, bg (dxy dxy) p, au (dxy dxy) p, bg {(dyz dxy) (dyz dxy)} p, au (ds ds), bu 20.762 19.191 18.768 17.964 17.949 17.947 (ds ds), ag {(dyz dxy) (dyz dxy)} p, bg (dxy dxy) p, au {(dyz dxy) (dyz dxy)} p, bg 17.647 17.575 17.339 12.367 11.893 9.757 p* (dyz dyz), au p* (dyz dyz), bg au bg 9.074 Energy/eV %NH3 17.297 17.291 17.210 17.172 16.715 16.608 15.496 15.444 15.266 9.133 9.072 8.946 8.785 2322000000010 Energy/eV %NH3 17.883 17.479 17.175 16.961 16.782 16.653 15.594 15.480 15.326 132000010 Description b (dyz dxy) p, Ru(A) ds, RuA (dyz dxy) p, Ru(A) dxy p, Ru(A) dxy, Ru(bpy) dyz, Ru(bpy) ds, Ru(bpy)Table 3 (Contd.) [Ru(NH3)4(qH2catH4)Ru(bpy)2]4 %Ru(A) Orbital 121 122 123 124 19000 [Ru(NH3)4(qH2qH2)Ru(bpy)2]4 %Ru(A) Orbital 15 19 94 54 47 12 110 111 112 113 114 115 116 117 118 119 (HOMO) 120 30031810 121 122 a Approximate description due to low symmetry.Symmetry with respect to the bridge indicated. b Due to the low symmetry, these labels are approximate descriptions; the RuA and Ru(bpy) indicate upon which ruthenium atom the d orbital resides. c Approximate description with respect to the bridge.The orientations of orbitals #117¡V#119 are distorted by interaction with the bpy residues and do not have a simple symmetrical relationship to the bridge. orbitals (Fig. 7). The energy separation between these in- and out-of-phase combinations is essentially zero (Table 3). One can identify the (dxy ¡Ó dxy) combinations at (#70,71,76) (Table 3) and the {(dyz dxy) ¡Ó (dyz dxy)} combinations at (#72,77) separated by 5040 cm1, clearly large due to the conjugated pathway.While the {(dyz dxy) (dyz dxy)} combination can be discerned (#75), the corresponding {(dyz dxy) (dyz dxy)} was not evident. There is clearly very extensive coupling through the various orbital pathways via the bridging ligand but the presence of so many ligand £k and £k* orbitals of the same symmetry makes it very dicult to extract actual electronic coupling matrix elements.This is left for future analysis.31 However comparison of the MOs of the free ligand (Fig. 5) with those of the complex reveals that the low lying orbitals #70,71 are combinations of the d orbitals primarily with free ligand orbitals #38,39 respectively, while the orbitals #72,78 involve d primarily with ligand orbital #40, and the complex pair #75,79 primarily with ligand orbital #41.These orbital combinations are recognizable though clearly there must be additional mixing between orbitals of the same symmetry. The alternative C2v bridge conformation yields data (spectroscopic transition energies, etc.) which do not dier signicantly from the C2h conformation. A calculation for the symmetric (qH2qH2) species but with a dihedral angle of 90 yielded a predicted electronic spectrum in poorer agreement with the experimental spectrum than the zero degree dihedral angle calculation. In the asymmetric species the mixing is extensive and complex, however it only occurs between individual metal centres and the bridging ligand.There is a large energetic dierence between the d(t2g) orbitals of the [Ru(NH3)4]2 fragment and the [Ru(bpy)2]2 fragment,1 which inhibits their mixing. The HOMO¡VHOMO 2 (#119¡V#117) orbitals contain the t2g set of the [Ru(bpy)2]2 fragment mixed extensively with bpy and to a varying degree with the bridging ligand £k and £k* orbitals. Thus HOMO 1 (#118) has £m-symmetry with respect to the bridge and is only minimally mixed therewith (Table 3, Fig. 8). The LUMO is primarily a £k*-bridge orbital coupled to %Ru(bpy) 0016 %Ru(bpy) 040432162 67 513 254 %Bridge 80 72 270 %Bridge 76 693 35 22 123 237 34 77 672 Energy/eV %bpy Description b %NH3 £k* dyz, Ru(A) 11.649 9.150 9.135 9.006 1 28 72 93 0001 Energy/eV %bpy Description c %NH3 (dyz dxy), Ru(A) d£m(RuA) dyz £k, Ru(A) dxy(Ru(A)) 18.318 17.659 17.349 17.088 16.957 16.906 960526 74 92 16.821 16.026 15.989 15.935 11.583 10.727 £k* dyz(Ru(A)) £k* dyz(Ru(bpy)) 15 27 121794 9.277 0122100000100 the dyz of the [Ru(NH3)4]2 fragment and the LUMO 1 is a £k*-bridge orbital coupled to the dyz of the [Ru(bpy)2]2 fragment, both with substantial metal d(t2g) character.The t2g set of the [Ru(NH3)4]2 fragment is distributed (> 12% in each orbital) over HOMO 4 to HOMO 9 (#115¡V#110) and the LUMO.HOMO 3 and HOMO 4 are predominantly bpy orbitals. Table 3 shows the descriptions of each MO although in this low symmetry molecule, it is not always possible to make a simple identication of the d orbital involved. However (#110) and (#113) appear to be the bonding and anti-bonding interactions of a d orbital with a £k-orbital respectively, and (#112) is clearly the £m-symmetry orbital with little coupling to the bridge.Clearly there is extensive coupling of both sets of metal d£k levels to the £k and £k* orbitals of the ligand bridge, as in the symmetric (qH2qH2) species but coupling between the metal ions across the bridge is minimal. (d) Ruthenium(III) species.The ruthenium() species are not very stable and were not isolated. The mixed valence [RuIII(qH2 qH2)RuII] species may, in principle, be localised or delocalised and may be at or twisted at the central C¡VC bridge. Since we were not able to collect much data on these species, we do not attempt to calculate them at present. Relative orbital energies The relative orbital energies are listed in Table 3 and displayed in Fig. 9 and 10. As anticipated the spread in the d(t2g) orbitals is quite small when the metal is attached to the unmixed catH4 end of the (catH4catH4) or (qH2catH4) species. However the splitting of the d(t2g) set attached to the qH2 fragment is signi- cantly larger. In the asymmetric species, the energies of localized bpy orbitals remain essentially constant and the overall splitting of the Ru d(t2g) bpy levels attached to qH2 remains much smaller than for the corresponding Ru d(t2g) NH3 fragment.Competition for a d£k orbital interaction between the bpy and bridge ligands may then cause less eective overlap with the bridge than is present with the tetrammine ruthenium centre. 2663 J. Chem. Soc., Dalton Trans., 1999, 2653¡V2667Electronic spectra Spectroscopic data are listed in Table 4 for both the experimental and the calculated spectra.Assignments of the spectra, based on the ZINDO/S calculations and orbital mixing, follow. (i) Symmetric species. (a) [(Ru(NH3)4)2(catH4catH4)]4. This species has a simple electronic spectrum (Fig. 3, spectrum 0) lacking intense absorption in the visible region.The ZINDO/ S analysis predicts that there are no intense transitions in the visible region for this complex. However a weak feature near 29400 cm1 is reasonably assigned as a d¡Vd transition (d£m(t2g) ¡÷ d£m*) and two transitions in the UV are identied as MLCT and £k¡V£k* transitions with energies well reproduced by the ZINDO calculation (Table 4). Certainly these spectra are completely in accord with the formulation of this species as the fully reduced (catH (b) [(Ru(NH 4catH4) redox isomer. 3)4)2(qH2catH4)]4. The electronic spectrum of this complex is shown in Fig. 3, spectrum 2. Here we anticipate a strong visible region MLCT transition between the d £k* ¡÷ £k* d orbitals and the band at 20160 cm1 is so assigned (#75 ¡÷ #79). At higher energies there are a series of predicted MLCT and £k ¡÷ £k* transitions which reproduce the general experimental features although not all the predicted bands are resolved experimentally.The strong visible region band conrms the presence of a qH (c) [(Ru(NH 2 oxidation fragment. 3)4)2(qH2qH2)]4. The electronic spectrum (Table 4) of this complex is shown in Fig. 3, spectrum 4. This complex is assumed to have C2h symmetry, and the observed electronic transitions are Laporte allowed.An even more intense visible region band is predicted than for the (qH2catH4) species because the number of qH2 chromophores has doubled; otherwise the assignment will be similar. The rst transition is to 1Bu calculated to be at 20080 cm1 (experimentally 19690 cm1) and is a mixture of the #75 and #77 to LUMO (#78) (bg ¡÷ au) transitions.These transitions involve very highly mixed orbitals, with scrambled d components and are MLCT transitions, with £k ¡÷ £k* character. Transition to LUMO 1 (#79) from #76 produces another 1Bu MLCT state seen as a shoulder, while a more intense feature at 33780 cm1 is yet another 1Bu MLCT state terminating on bridge orbital #80. The overall agreement between predicted and experimental band energies is good.Compared with the corresponding band in the symmetric (qH2catH4) species, the oscillator strength of the visible region MLCT band does double in value, and shifts to the red probably because of greater delocalisation. (d) Ruthenium(III) species. Data for the two RuIII species are included in Table 4 (Fig. 3 inset) but for reasons noted above, we have not obtained any calculated data. The broad shoulder on the low energy side of the mixed valence species is reasonably assigned to the inter-valence transition. 1 (ii) Asymmetric species. (a) [Ru(NH3)4(catH4catH4)Ru- (bpy)2]4. The spectra of the asymmetric species are more complex than the symmetric species because they have C symmetry and there is also the complication of having the [Ru(bpy)2]2 chromophore which has electronic transitions of its own (see Fig. 4, spectrum 0). Clearly the visible region absorption can only be Ru(d£k) ¡÷ £k* (1) (bpy) in character (Table 4). A shoulder to higher energy is associated with the expected 1¡V6 Ru(d£k) ¡÷ £k* (2) (bpy) transitions (orbitals #126¡V#129). (b) [Ru(NH3)4(qH2catH4)Ru(bpy)2]4.The oxidation of one end of the bridging ligand creates a new chromophore in the (NH3)4Ru(qH2) end of the molecule (see Fig. 4, spectrum 2) and we now anticipate two visible region transitions corresponding to MLCT to bpy and to the bridge. These are indeed observed, lying fairly close together. The lower energy visible region transition involves orbitals (#114,115 ¡÷ #121) that 2664 J.Chem. Soc., Dalton Trans., 1999, 2653¡V2667 are localized on the (NH3)4Ru(qH2) end of the molecule being d(t2g) (NH3)4Ru ¡÷ £k* (LUMO) qH2 MLCT transition with some £k ¡÷ £k* character. The higher energy component is then the Ru(d£k) ¡÷ £k*(bpy) set of transitions as calculated (Table 4). These MLCT transitions to bipyridine are calculated to be very much weaker than the RuA(d£k) ¡÷ qH2(£k*) MLCT transition. Indeed the actual (Ru(d£k) ¡÷ £k*(bpy)) shoulder on the high energy side of the visible region band is actually quite weak, being built upon the high energy tail of the lower energy transition.Its reported experimental intensity in Table 4 is then a dramatic overestimate of its true (deconvoluted) intensity. At higher energy lie the expected d(t2g) ¡÷ £k*(2) bpy MLCT transitions, some additional MLCT to the bridge and £k¡V£k* transitions (Table 4).(c) [Ru(NH3)4(qH2qH2)Ru(bpy)2]4. In this oxidation state there are many chromophores and many possible transitions (Fig. 4, spectrum 4). Clearly we can expect two dierent Ru(d£k) ¡÷ bridge in addition to the Ru(d£k) ¡÷ £k* (bpy) bands. The increasing oxidation on the bridge causes the ruthenium atoms to be more positive and this shifts the Ru(d£k) ¡÷ £k*(bpy) bands to higher energy where they appear as a shoulder near 430 nm (Fig. 4). The strong visible region band is assigned to the HOMO ¡÷ LUMO transition which is a complex transition containing MMCT, MLCT and £k ¡÷ £k* components as seen from the nature of these MOs in Table 3. Additional MLCT transitions from lower lying d orbitals to the LUMO appear more weakly at higher energy and are responsible for the broad absorption near 24000 cm1.The corresponding MLCT transitions to LUMO 1 are weaker and at higher energy (Table 4). The anticipated Ru(d£k) ¡÷ £k* (bpy) transitions appear at rather higher energies due to the high net oxidation state of the complex. The overall agreement in this series of six complexes between the observed experimental spectra and the ZINDO calculated spectra is remarkably good, with the poorest agreement perhaps with the asymmetric (qH2qH2) species.Conclusions This paper has described the synthesis of two new dinuclear complexes, and the spectroscopic characterization of the two- and four-electron oxidized products. A ZINDO/S analysis of the closed shell Ru() complexes produced a reasonable t to the experimentally determined spectra providing condence in the signicance of the electronic structural analysis.The ZINDO/S analysis reveals in considerable detail the extensive metal¡Vligand orbital mixing that exists in many of these complexes. As would be anticipated the (qH2qH2) species are by far the most mixed and the (catH4catH4) are the least mixed.The extent of the mixing in the (qH2qH2) species is further evidence of the close orbital energy match and the good overlap that occurs in this type of complex. The mixing in these species can be described using the familiar terminology of back-donation. Species such as 2,2- bipyridine are regarded as good £k-back bonding ligands but the extent of £k-back bonding in such species is very much less than observed here with the qH2 fragments.Back-donation in the 2,2-bipyridine ruthenium species involves mixing between metal d£k into £k* (bpy) ligands of the order of 5¡V10% at most, corresponding to a formal transfer of 0.1¡V0.2 electrons. In the qH2 species we commonly observe ca. 20% admixture 1a,5d,9 or a formal transfer of 0.4 electrons.Indeed in the dinuclear qH2qH2 complex, we observe around 0.4 electron back donation from each end of the molecule for a total of approximately 0.8 electrons into the at £k* LUMO and LUMO 1 of qH2qH2. Unfortunately it has so far not proved possible to obtain crystals of these species for X-ray analysis¡Xthis is still being attempted. The extent to which the optimized ZINDO/1 struc-Table 4 Electronic spectroscopic data for this series of complexes and related complexes in 0.1 M aqueous H3PO4 Energy a/cm1 [(Ru(NH3)4)2(catH4catH4)]4 (HOMO = 79) 47900 (0.44) 39680 (0.10) 30710 (0.0002) 47620 39370 (4.25, 8010, 0.62) 29410 (sh) (3.29) [(Ru(NH3)4)2(qH2catH4)]4 (HOMO = 78) 47620 40000 (sh) 34300 (4.12, 7360, 0.42) 28600 (sh) 20160 (4.01, 3740, 0.17) [(Ru(NH3)4)2(qH2qH2)]4 (HOMO 77) 40650 (sh) 33780 (4.24, 9960, 0.75) 28600 (sh) 19690 (4.27, 4950, 0.40) [Ru(NH3)4(catH4catH4)Ru(bpy)2]4 (HOMO = 121) 41100 (4.61) 39100 (sh) 34400 (4.82, 3300, 1) 28300 (sh) 23500 (sh) 20680 (4.14) [Ru(NH3)4(qH2catH4)Ru(bpy)2]4 (HOMO = 120) 41100 (4.55) 39100 (sh) 34400 (4.75, 4000, 1) 28300 (sh) (c) 20300 (4.14) 18420 (4.19, 3440, 0.25) [Ru(NH3)4(qH2qH2)Ru(bpy)2]4 (HOMO = 119) 40000 (sh) 34600 28000 (sh) 23800 (3.89) 18300 (4.33, 4600, 0.45) 14000 (sh) [(Ru(bpy)2)2(qH2qH2)]4 (HOMO = 161) 4,32 21370 (0.35) 17870 (1.29) 23050 17100 Calculated energy/cm1 (f) b Assignment 48450 (0.54) 46700 (0.14) 42300 (0.17) 41450 (0.15) 36950 (0.64) 36900 (0.1) 33170 (0.1) 24230 (0.04) 21540 (0.64) 41780 (0.43) 39500 (0.13) 37800 (0.07) 36220 (0.77) 29470 (0.20) 20080 (1.38) 42000 (0.14) 37400 (0.45) 36500 (0.55) 36200 (0.36) 34720 (0.23) 33040 (0.12) 32760 (0.1) 31990 (0.23) 31300 (0.18) 23370 (0.13) 23120 (0.05) 21760 (0.03) 33000 (0.12) 32520 (0.12) 32500 (0.13) 31880 (0.23) 31400 (0.12) 31340 (0.10) 24200 (0.14) 23400 (0.06) 23000 (0.09) 21250 (0.75) Many transitions 34900 (0.25) 34200 (0.68) 32400 (0.06) 32240 (0.32) 31670 (0.13) 26520 (0.12) 25600 (0.05) 23670 (0.10) 22470 (0.27) 17500 (1.37) 11000 (0.01) ¦� ¡ú¦�*; 72 ¡ú 81 ¦� ¡ú ¦�*; MLCT, 77 ¡ú 80 and 76 ¡ú 81 d ¡ú d transitions, 79 ¡ú 86 and 87 ¦� ¡ú ¦�*, 69 ¡ú 79 ¦� ¡ú ¦�*; MLCT, 78 ¡ú 81 ¦� ¡ú ¦�*; MLCT, 78 ¡ú 82 ¦� ¡ú ¦�*; MLCT, 75 ¡ú 82 ¦� ¡ú ¦�*, MLCT, 78 ¡ú 81 d ¡ú d; LMCT, 78 ¡ú 86 ¦� ¡ú ¦�*, d ¡ú d, 73 ¡ú 84 and 70 ¡ú 79 ¦� ¡ú ¦�*; MLCT, 70 and 72 ¡ú 79 MLCT; ¦� ¡ú ¦�*, 75 ¡ú 79 MLCT; ¦� ¡ú ¦�*, 76 ¡ú 81, 75 ¡ú 82 MLCT; ¦� ¡ú ¦�*, 75 ¡ú 80 ¦� ¡ú ¦�*; MLCT, 71 ¡ú 78, 70 ¡ú 79 ¦� ¡ú ¦�*; MLCT, 77 ¡ú 80 ¦� ¡ú ¦�*; MLCT, 76 ¡ú 79 ¦� ¡ú ¦�*; MLCT, 77 and 75 ¡ú 78 Ru ¡ú bpy, MLCT, 120 ¡ú 134, 121 ¡ú 135 Ru ¡ú bpy, MLCT, 120 ¡ú 132 119 ¡ú 131 120 ¡ú 131, 121 ¡ú 132 Ru ¡ú bpy, MLCT, 121 ¡ú 131 Ru(bpy) ¡ú bridge, MLCT, 121 ¡ú 126 Ru(bpy) ¡ú bridge, MLCT, 121 ¡ú 126 Ru ¡ú bpy, MLCT 121 ¡ú 127 121 ¡ú 128 and 129 Ru ¡ú bpy, MLCT 119 ¡ú 123 120 ¡ú 124 121 ¡ú 123 Ru ¡ú bpy, MLCT, 118 ¡ú 128, 120 ¡ú 131 Ru ¡ú bpy, MLCT, 119 ¡ú 127, 118 ¡ú 128 ¦� ¡ú ¦�* bridge, 107 and 111 ¡ú 121 Ru ¡ú bpy, MLCT, 120 ¡ú 127 ¦� ¡ú ¦�* bridge, 111 ¡ú 121 Ru ¡ú bpy, MLCT, 120 ¡ú 128 Ru(NH3)4 ¡ú bridge; MLCT, 112 ¡ú 121 Ru ¡ú bpy, MLCT 118 ¡ú 123 119 ¡ú 124 Ru(NH3)4 ¡ú bridge; ¦� ¡ú ¦�*; MLCT, 115 and 114 ¡ú 121 117 ¡ú 127 119 ¡ú 124 ¦�(bpy) ¡ú ¦�*(bridge) Ru(bpy) Ru(NH3), LLCT, LMCT, 116 ¡ú 120 and 121 ¦� ¡ú ¦�* bpy, 116 ¡ú 123 and 115 ¡ú 122 ¦� ¡ú ¦�* bpy, 116 ¡ú 120 and 122 Ru(bpy)2 ¡ú bpy ¦�* MLCT, 118 ¡ú 122 115 ¡ú 121, bpy ¡ú bridge, LLCT Ru(NH3)4 ¡ú bridge, ¦� ¡ú ¦�*; MLCT, 113 ¡ú 120 Ru(bpy)2 ¡ú bridge; ¦� ¡ú ¦�* bridge, 117 ¡ú 120 Ru(bpy)2(qH2) ¡ú Ru(NH3)4(qH2), 119 ¡ú 120 Ru(NH3)4, ¦�-bpy, ¦�-bridge ¡ú Ru(NH3)4(qH2), 114 ¡ú 120 (bpy)2Ru ¡ú bridge, 161 ¡ú 163, 156 ¡ú 162 (bpy)2Ru ¡ú bridge, 161 and 159 ¡ú 162 2665 J.Chem. Soc., Dalton Trans., 1999, 2653�C2667Table 4 (Contd.) Assignment Calculated energy/cm1 (f) b Energy a/cm1 [(Ru(NH3)4)2(qH2qH2)]5 MLCT and LMCT Intervalence MMCT? 34700 (4.19) 21700 (4.18) 18100 (sh) (4.09) [(Ru(NH3)4)2(qH2qH2)]6 34300 (sh) (4.13) 22400 (4.13) LMCT a Data in parentheses are (log (¥å/dm3 mol1 cm1), half band width (cm1), oscillator strength).b Principal calculated bands in the visible and near- UV region with calculated oscillator strengths (in parentheses) >0.05. c A broad background absorption is evident in this region, see Fig. 3, insert. tures are close to the real structures still remains to be assessed. However the overall features of the structures, including bond distance, are most likely correct and the uncertainty remains primarily with the twist angle at the biphenyl link. Small changes in the twist angle have no substantive eect on the descriptions of the molecules discussed here.In particular the extent of interaction between the metal centre and its local diimino fragment is not greatly aected by the twist angle. Acknowledgements We would like to thank Drs Elaine Dodsworth, Yu-Hong Tse, Hitoshi Masui, Carlos da Cunha, and Mr Sergei I. Gorelsky for valuable discussion. Thanks are also due to Professor Michael Siu and his students for assistance with the electrospray mass spectroscopy.We would also like to thank CNPq (Conselho de Desenvolvimento Cientico e Tecnologico da Secretaria de Ciencia e Tecnologia do Brasil), and the Natural Sciences and Engineering Research Council of Canada (Ottawa) for nancial support. Some of the RuCl3 used in this project was a generous loan from Johnson Matthey Co.Inc. References 1 (a) R. A. Metcalfe and A. B. P. Lever, Inorg. Chem., 1997, 36, 4762; (b) R. A. Metcalfe, E. S. Dodsworth, S. S. Fielder, D. J. Stufkens, A. B. P. Lever and W. J. Pietro, Inorg. Chem., 1996, 35, 7741; (c) R. A. Metcalfe, Ph.D. Thesis, York University, Canada, 1988. 2 C. J. da Cunha, S. S. Fielder, D. V. Stynes, H. Masui, P.R. Auburn and A. B. P. Lever, Inorg. Chim. Acta., 1996, 242, 293. 3 R. A. Metcalfe, E. S. Dodsworth, W. J. Pietro, A. B. P. Lever and D. J. Stufkens, Inorg. Chem., 1993, 32, 3581. 4 P. R. Auburn and A. B. P. Lever, Inorg. Chem., 1990, 29, 2551. 5 (a) H. Masui, A. B. P. Lever and P. R. Auburn, Inorg. Chem., 1991, 30, 2402; (b) H. Masui, A. B. P. Lever and E. S. Dodsworth, Inorg.Chem., 1993, 32, 258; (c) H. Masui, Ph.D. Thesis, York University, Canada, 1994; (d ) H. Masui, A. L. Freda and A. B. P. Lever, Inorg. Chem., submitted; (e) C. J. da Cunha, E. S. Dodsworth, M. A. Monteiro and A. B. P. Lever, Inorg. Chem., 1999, 38, in the press. 6 P. Belser, A. von Zelewsky and M. Zehnder, Inorg. Chem., 1981, 20, 3098; H. Y. Cheng and S.M. Peng, Inorg. Chim. Acta, 1990, 169, 23; E. H. Cutin, N. E. Katz, P. A. M. Williams and P. J. Aymonino, Transition Met. Chem., 1991, 16, 155; S. Nemeth and L. I. Simandi, Inorg. Chem., 1994, 33, 5964; G. Ricciardi, A. Rosa, G. Morelli and F. Lelj, Polyhedron, 1991, 10, 955; O. Carugo, K. Djinovic and C. B. Castellani, J. Chem. Soc., Dalton Trans., 1991, 1551; J. Rall, A.F. Stange, K. Hubler and W. Kaim, Angew. Chem., Int. Ed., 1998, 37, 2681; T. Justel, J. Bendix, N. Metzler, T. Weyhermuller, B. Nuber and K. Wieghardt, Inorg. Chem., 1998, 37, 35; S.-M. Peng, K. Peters, E. M. Peters and A. Simon, Inorg. Chim. Acta, 1985, 101, L35; S. Peng, C. Chen, D. Liaw, C. Chen and Y. Wang, Inorg. Chim. Acta, 1985, 101, L31; L. I. Simandi and S.Nemeth, Inorg. Chim. Acta, 1998, 270, 326. 7 J. F. Joulie, E. Schatz, M. D. Ward, F. Weber and L. J. Yellowlees, J. Chem. Soc., Dalton Trans., 1994, 799; S. Bruni, F. Cariati, A. Dei and D. Gatteschi, Inorg. Chim. Acta, 1991, 186, 157; A. Dei, D. Gatteschi and L. Pardi, Inorg. Chem., 1990, 29, 1442; M. D. Ward, Inorg. Chem., 1996, 35, 1712; F. Heinze, S. Mann, G. Huttner and L.Zsolnai, Chem. Ber., 1996, 129, 1115; A. Dei, D. Gatteschi, 2666 J. Chem. Soc., Dalton Trans., 1999, 2653¡©2667 L. Pardi and U. Russo, Inorg. Chem., 1991, 30, 2589; A. Dei, D. Gatteschi and L. Pardi, Inorg. Chim. Acta, 1991, 189, 125; M. A. Calvo, A. M. M. Lanfredi, L. A. Oro, M. T. Pinillos, C. Tejel, A. Tiripicchio and F. Ugozzoli, Inorg. Chem., 1993, 32, 1147; S.Cueto, H. P. Straumann, P. Rys, W. Petter, V. Gramlich and F. S. Rys, Acta Crystallogr., Sect. C, 1992, 48, 458; A. Elduque, Y. Garces, L. A. Oro, M. T. Pinillos, A. Tiripicchio and F. Ugozzoli, J. Chem. Soc., Dalton Trans., 1996, 2155; R. F. Johnston, P. K. Sengupta, M. B. Ossain, R. Vanderhelm, W. Y. Jeong and R. A. Holwerda, Acta Crystallogr., Sect. C, 1990, 46, 1796; Ac., Chem. Commun., 1991, 1054; J. Zubieta, J. Chem. Soc., Chem. Commun., 1988, 1017; T. E. Keyes, R. J. Forster, P. M. Jayaweera, C. G. Coates, J. J. McGarvey and J. G. Vos, Inorg. Chem., 1998, 37, 5925. 8 (a) W. P. Anderson, T. Cundari, R. S. Drago and M.C. Zerner, Inorg. Chem., 1989, 29, 1; (b) W. P. Anderson, T. Cundari and M. C. Zerner, Int. J. Quantum Chem., 1991, 39, 31; (c) W. P. Anderson, W. D. Edwards and M. C. Zerner, Inorg. Chem., 1986, 25, 2728; (d ) A. D. Bacon and M. C. Zerner, Theor. Chim. Acta, 1979, 53, 21; (e) M. C. Zerner, Metal¡©Ligand Interactions, Kluwer Academic Publishers, Dordrecht, 1996, p. 493; ( f ) M.C. Zerner, G. H. Loew, R. F. Kirchner and U. T. Mueller-Westerho, J. Am. Chem. Soc., 1980, 102, 589; ( g) J. C. Culberson, P. Knappe, N. Rosch and M. C. Zerner, Theor. Chim. Acta, 1987, 71, 21; (h) J. Ridley and M. C. Zerner, Theor. Chim. Acta, 1973, 32, 111; (i) J. Ridley and M. C. Zerner, Theor. Chim. Acta, 1976, 42, 223. 9 S. I. Gorelsky, E. S. Dodsworth, A.B. P. Lever and A. Vlcek, Coord. Chem. Rev., 1998, 174, 469. 10 Y. K. Shin, B. S. Brunschwig, C. Creutz, M. D. Newton and N. Sutin, J. Phys. Chem., 1996, 100, 1104 and references therein; K. J. Lachance-Galang, P. E. Doan, M. J. Clarke, U. Rao, A. Yamano and B. M. Homan, J. Am. Chem. Soc., 1995, 117, 3529; L. T. Zhang and M. Ondrechen, Inorg. Chim. Acta, 1994, 226, 43; O.V. Sizova, V. I. Baranosvskii, N. V. Ivanova and A. B. Nikol¡�skii, Russ. J. Gen. Chem., 1997, 67, 1667; O. V. Sizova, V. I. Baranovski, N. V. Ivanova and A. I. Panin, Int. J. Quantum Chem., 1997, 65, 183; O. V. Sizova, V. I. Baranovskii, N. V. Ivanova and A. I. Panin, Russ. J. Coord. Chem., 1997, 23, 197; O. V. Sizova, N. V. Ivanova and A. B. Nikolskii, Zh. Obshch.Khim., 1997, 67, 26; O. V. Sizova, V. I. Baranovski, N. V. Ivanova and A. I. Panin, Int. J. Quantum Chem., 1997, 65, 183; O. V. Sizova, V. I. Baranovskii, N. V. Ivanova and A. I. Panin, Russ. J. Coord. Chem., 1997, 23, 197; O. V. Sizova, N. V. Ivanova, A. Y. Ershov, A. B. Nikolskii and I. V. Rogachevskii, Zh. Obshch. Khim., 1997, 67, 1409; O. V. Sizova, N. V. Ivanova and A.B. Nikolskii, Zh. Obshch. Khim., 1997, 67, 26; O. V. Sizova, A. I. Panin, N. V. Ivanova and V. I. Baranovskii, J. Struct. Chem., 1997, 38, 366; Y. K. Shin, D. J. Szalda, B. S. Brunschwig, C. Creutz and N. Sutin, Inorg. Chem., 1997, 36, 3190; J. Yang, D. Seneviratne, G. Arbatin, A. M. Andersson and J. C. Curtis, J. Am. Chem. Soc., 1997, 119, 5329; A. Broo and P.Lincoln, Inorg. Chem., 1997, 36, 2544; H. Rensmo, S. Lunell and H. Siegbahn, J. Photochem. Photobiol. A: Chem., 1998, 114, 117; O. V. Sizova, V. I. Baranovskii, A. I. Panin, N. V. Ivanova and A. B. Nikolskii, Zh. Obshch. Khim., 1997, 67, 1611; O. V. Sizova, I. Panin, V. I. Baranovskii, A. B. Nikolskii and N. V. Ivanova, Zh. Obshch. Khim., 1997, 67, 1761. 11 See, for example: (a) S.Ernst, V. Kasack and W. Kaim, Inorg. Chem., 1988, 27, 1146; (b) S. Joss, H. Reust and A. Ludi, J. Am. Chem. Soc., 1981, 103, 981; (c) S. Joss, H. B. Burgi and A. Ludi, Inorg. Chem., 1985, 24, 949; (d ) S. Joss, K. M. Hasselbach, H. B. Burgi, R. Wordel, F. E. Wagner and A. Ludi, Inorg. Chem., 1989, 28, 1815; (e) J. E. B. Johnson, C. Degro and R. R. Ruminski, Inorg.Chim. Acta, 1991, 187, 7380; ( f ) M. A. S. Aquino, F. L. Lee, E. J. Gabe, C. Bensimon,J. E. Greedan and R. J. Crutchley, J. Am. Chem. Soc., 1992, 114, 5130; ( g) W. Bruns, W. Kaim, E. Waldhor and M. Krejcik, J. Chem. Soc., Chem. Commun., 1993, 1868; (h) C. Creutz, M. D. Newton and N. Sutin, J. Photochem. Photobiol. A, 1994, 82, 47; (i) H. Y. Huang, W. J. Chen, C.C. Yang and A. Yeh, Inorg. Chem., 1991, 30, 1862; ( j) V. Kasack, W. Kaim, H. Binder, J. Jordanov and E. Roth, Inorg. Chem., 1995, 34, 1924; (k) J. A. Roberts and J. T. Hupp, Inorg. Chem., 1992, 31, 157; (l) M. D. Ward, Chem. Soc. Rev., 1995, 24, 121; (m) A. E. Almaraz, L. A. Gentil, L. M. Baraldo and J. A. Olabe, Inorg. Chem., 1997, 36, 1517; (n) J. F. Endicott, M. A.Watzky, X. Q. Song and T. Buranda, Coord. Chem. Rev., 1997, 159, 295; (o) F. Salaymeh, S. Berhane, R. Yusof, R. de la Rosa, E. Y. Fung, R. Matamoros, K. W. Lau, Q. Zheng, E. M. Kober and J. C. Curtis, Inorg. Chem., 1993, 32, 3895. 12 D. D. Perrin, W. L. F. Armarego and D. R. Perrin, Puri.cation of Laboratory Chemicals, Pergamon Press, Elmsford, New York, 2nd edn., 1980. 13 A.B. P. Lever, E. R. Milaeva and G. Speier, Phthalocyanines: Properties and Applications, eds. C. C. Lezno. and A. B. P. Lever, VCH, New York, 1995, vol. 3, p. 3. 14 S. I. Gorelsky and A. B. P. Lever, unpublished work. 15 K. Krogh-Jespersen, J. D. Westbrook, J. A. Potenza and H. J. Schugar, J. Am. Chem. Soc., 1987, 109, 7025. 16 A. Szabo and N. S. Ostlund, Modern Quantum Chemistry, Macmillan Publ.Co., New York, 1982. 17 X. Cheng, Q. Gao, R. D. Smith, E. E. Simanek, M. Mammen and G. M. Whitesides, J. Org. Chem., 1996, 61, 2204; J. Manna, J. A. Whiteford, P. J. Stang, D. C. Muddiman and R. D. Smith, J. Am. Chem. Soc., 1996, 118, 8731; G. J. Langley, E. Hecquet, I. P. Morris and D. G. Hamilton, Rapid Commun. Mass Spectrom., 1997, 11, 165; A.Marquis-Rigault, A. Dupont-Gervais, A. Van Dorsselaer and J.-M. Lehn, Chem. Eur. J., 1996, 2, 1395; D. P. Funeriu, J.-M. Lehn, G. Baum and D. Fenske, Chem. Eur. J., 1997, 3, 99; C. Moucheron, A. Kirsch-De Mesmaeker, A. Dupont-Gervais, E. Leize and A. Van Dorsselaer, J. Am. Chem. Soc., 1996, 118, 12834; B. C. Gilbert, N. W. J. Kamp, J. R. Lindsay Smith and J.Oakes, J. Chem. Soc., Perkin Trans. 2, 1998, 1841; W. Henderson, J. S. McIndoe, B. K. Nicholson and P. J. Dyson, J. Chem. Soc., Dalton Trans., 1998, 519; C. Jiang, W. Henderson, T. S. A Hor, L. J. McCa.rey and Y. K. Yan, Chem. Commun., 1998, 2029; A. S. Batsanov, A. V. Churakov, M. A. M. Easson, L. J. Govenlock, J. A. K. Howard, J. M. Moloney and D. Parker, J.Chem. Soc., Dalton Trans., 1999, 323; M. J. Deery, O. W. Howarth and K. R. Jennings, J. Chem. Soc., Dalton Trans., 1997, 4783; G. Hopfgartner, C. Piguet and J. D. Henion, J. Am. Soc. Mass Spectrom., 1994, 5, 748; C. Piguet, G. Bernardelli and H. Hopfgartner, Chem. Rev., 1997, 97, 2005; K. A. Hirsch, S. R. Wilson and J. S. Moore, J. Am. Chem. Soc., 1997, 199, 10401. 18 I. K. Chu, T.-C. Lau and K. W. M. Siu, J. Mass Spectrom., 1998, 33, 811. 19 C. G. Kuehn and H. Taube, J. Am. Chem. Soc., 1976, 98, 689; B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17, 3334. 20 S. Diamond, Ph.D Thesis, Stanford University, 1975. 21 L. Vasconcellos, R. A. Metcalfe, H. Mirza and A. B. P. Lever, to be published. 22 J. K. M. Sanders and B. K. Hunter, Modern NMR Spectroscopy, A Guide for Chemists, Oxford University Press, New York, 1978; J. B. Lambert, H. F. Shurvell, D. Lightner and R. G. Crooks, Introduction to Organic Spectroscopy, Macmillan Publishing Company, New York, 1987; R. D. Foust and P. C. Ford, J. Am. Chem. Soc., 1972, 94, 5686; C. A. Stein and H. Taube, J. Am. Chem. Soc., 1978, 100, 336; H. Lehmann, K. J. Schenk, G. Chapuis and A. Ludi, J. Am. Chem. Soc., 1979, 101, 6197; I. P. Evans, G. W. Everett and A. M. Sargeson, J. Am. Chem. Soc., 1976, 98, 8041. 23 M. Maruyama, H. Matsuzawa and Y. Kaizu, Inorg. Chim. Acta, 1995, 237, 159; M. Bourgault, T. Renouard, B. Lognone, C. Mountassir and H. LeBozec, Can. J. Chem., 1997, 75, 318; L. S. Kelso, D. A. Reitsma and F. R. Keene, Inorg. Chem., 1996, 35, 5144. 24 J. K. M. Sanders and B. K. Hunter, Modern NMR Spectroscopy, A Guide for Chemists, Oxford University Press, New York, 1978; J. B. Lambert, H. F. Shurvell, D. Lightner and R. G. Crooks, Introduction to Organic Spectroscopy, Macmillan Publishing Company, New York, 1987. 25 R. R. Ruminski and J. D. Petersen, Inorg. Chem., 1982, 21, 3706. 26 C. Creutz, Prog. Inorg. Chem., 1983, 30, 1 and references therein. 27 A. B. P. Lever, Inorg. Chem., 1990, 29, 1271. 28 P. Bernhard, D. J. Bull, H. B Burgi. P. Osvath, A. Raselli and A. M. Sargeson, Inorg. Chem., 1997, 36, 2804; M. J. Ridd and F. R. Keene, J. Am. Chem. Soc., 1981, 103, 5733; V. L. Goedkin and D. H. Busch, J. Am. Chem. Soc., 1972, 94, 7355; P. Bernhard, D. J. Bull, H. B. Burgi. P. Osvath, A. Raselli and A. M. Sargeson, Inorg. Chem., 1997, 36, 2804; E. Jandrasics, Kolp, J. A. Wolny and A. von Zelewsky, Inorg. Chim. Acta, 1998, 272, 153; J. C. Dabrowiak, F. V. Lovecchio and D. H. Busch, J. Am. Chem. Soc., 1972, 94, 5502; R. Arakawa, S. Mimura, G. Matsubayashi and T. Matsuo, Inorg. Chem., 1996, 35, 5725; M. A. Fox and M. T. Dulay, Chem. Rev., 1993, 93, 341; B. C. Lane, J. E. Lester and F. Basolo, Chem. Commun., 1971, 1618; D. F. Mahoney and J. K. Beattie, Inorg. Chem., 1973, 12, 2561; F. R. Keene, P. A. Lay, G. E. Sneddon and G. W. Whebell, Aust. J. Chem., 1993, 46, 1763; P. Berhard and F. C. Anson, Inorg. Chem., 1989, 28, 3272; P. A. Lay, G. M. McLaughlin and A. M. Sargeson, Aust. J. Chem., 1987, 40, 1267; F. R. Keene, Coord. Chem. Rev., 1999, 187, 121. 29 P. H. Rieger, Coord. Chem. Rev., 1994, 135/136, 203 and references therein; A. R. Chakravarty and A. Chakravorty, Inorg. Chem., 1981, 20, 3138; N. Bag, G. K. Lahiri and A. Chakravorty, J. Chem. Soc., Dalton Trans., 1990, 1557; G. K. Lahiri, S. Bhattacharya, M. Mukherjee, A. K. Mukherjee and A. Chakravorty, Inorg. Chem., 1987, 26, 3359; G. K. Lahiri, S. Bhattacharya, B. K. Ghosh and A. Chakravorty, Inorg. Chem., 1987, 26, 4324; H. K. Gupta and S. Dikshit, Polyhedron, 1987, 6, 1009; N. S. Hush and A. Edgar, Chem. Phys. Lett., 1980, 69, 128; K. Matsumoto, T. Matsumoto, M. Kawano, H. Ohnuki, Y. Shichi, T. Nishide and T. Sato, J. Am. Chem. Soc., 1996, 118, 3597. 30 C. Joachim, S. P. Launay and S. Woitellier, Chem. Phys., 1990, 147, 131; J. P. Collin and J. P. Sauvage, J. Am. Chem. Soc., 1998, 120, 3717. 31 A. B. P. Lever, work in progress. 32 A. Freda, undergraduate thesis project. Paper 9/02850A 2667 J. Chem. Soc., Dalton Trans., 1999, 2653–2667

ISSN:1477-9226    

DOI:10.1039/a902850a

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 2657 Heterobimetallic nickel–sodium and cobalt–sodium complexes of pyridonate ligands Euan K. Brechin, Liam M. Gilby, Robert O. Gould, Steven G. Harris, Simon Parsons and Richard E. P. Winpenny * Department of Chemistry, The University of Edinburgh, West Mains Road, Edinburgh, UK EH9 3JJ Seven new mixed-metal complexes have been prepared and their structures determined. Reaction of anhydrous nickel(II) chloride with Na(xhp) (xhp = 6-methyl- or 6-chloro-2-pyridonate, mhp or chp respectively) produced the discrete polynuclear complexes [Ni4Na4(mhp)12(Hmhp)2] 1 and [Ni2Na2(chp)2(H2O)(MeCN)4] 2, depending on the pyridonate derivative used.A closely related complex [Ni2Na2(chp)6(H2O)2(MeCN)4] 3 cocrystallised with 2 from a reaction scheme involving phenylacetate. A polymeric complex [{Ni2Na2(chp)6(H2O)}n] 4 can be isolated from a similar reaction to that which gave 2 but by recrystallisation from ethyl acetate rather than the co-ordinating solvent MeCN.The cobalt analogue of 4, [{Co2Na2(chp)6(H2O)}n] 5 was also prepared and a closely related cobalt–sodium polymer, [{Co2Na2(chp)6}n] 6, crystallised from ethyl acetate. A further polymer featuring the mhp ligand, [{Co2Na2(mhp)6(Hmhp)(H2O)}n] 7, was produced from reaction of anhydrous cobalt(II) acetate with Na(mhp) followed by recrystallisation from toluene. The structures of the molecular species 1, 2 and 3 can be rationalised as co-ordination of Na by [Ni(xhp)3]2 ‘complex ligands’ in which the nickel centres are surrounded by three chelating pyridonates.In the polymeric species 4, 5, 6 and 7 the structures can be viewed as involving co-ordination of Na by [M2(xhp)6]22 complex ligands. In 6 the absence of any additional solvent molecules leads to interaction of the sodium site with Cl-donors. We have been exploring routes to high nuclearity co-ordination complexes using mixtures of 1,3-bridging ligands to stabilise the polynuclear arrays.During this work we have occasionally noticed involvement of sodium as a structural element, for example as a nucleation centre in the ‘metallocrown’ [Cu6- Na(mhp)12][NO3] 1 (mhp = 6-methyl-2-pyridonate) or as an anchoring point in the ‘supra-cage assembly’ [Ni16Na6(chp)4- (phth)10(Hphth)2(MeO)10(OH)2(MeOH)20] (chp = 6-chloro-2- pyridonate, phth = phthalate).2 Here we report seven further examples of the co-ordination of sodium by a 3d-metal complex.Experimental Preparations All reagents, metal salts and ligands were used as obtained from Aldrich. Sodium salts of pyridone ligands were obtained by deprotonation of the ligand in MeOH using Na(OMe) followed by evaporation to dryness. Analytical data were obtained on a Perkin-Elmer 2400 Elemental Analyser by the University of Edinburgh Microanalytical Service and are given in Table 1. Mass spectra were obtained by fast atom bombardment (FAB) of samples in a 3-nitrobenzyl alcohol matrix on a Kratos MS50 spectrometer. [Ni4Na4(mhp)12(Hmhp)2] 1.The salt Na(mhp) (0.506 g, 3.86 mmol) was added to a stirred solution of anhydrous nickel(II) chloride (0.250 g, 1.93 mmol) in MeOH (30 cm3). After 24 h the green solution was filtered and the filtrate evaporated to dryness under reduced pressure, leaving a green paste. Recrystallisation of this paste from acetonitrile or ethyl acetate produced green crystals of complex 1. FAB mass spectra (significant peaks, possible assignments): m/z 1843, [Ni4Na4(mhp)12(Hmhp)2]1; 1734, [Ni4Na4(mhp)12(Hmhp)]1; 1625, [Ni4Na4(mhp)12]1; 1409, [Ni4Na4(mhp)10]1; 1326, [Ni3Na3(mhp)10]1; 704, [Ni2Na2- (mhp)6]1; 596, [Ni2Na2(mhp)4]1; 537, [NiNa2(mhp)4]1; 429, [NiNa4(mhp)3]1; 406, [NiNa(mhp)3]1; and 298, [NiNa(mhp)2].[Ni2Na2(chp)6(H2O)(MeCN)4] 2. This was prepared in a similar fashion to that of complex 1 using Na(chp) in place of Na(mhp). Recrystallisation from MeCN gave green crystals of 2.FAB mass spectrum (significant peaks, possible assignments): m/z 1117, [Ni2Na2(chp)6(MeCN)4(H2O)]1; 806, [Ni2- Na2(chp)6]1; 655, [Ni2Na(chp)4]1; and 339, [NiNa(chp)2]1. [Ni2Na2(chp)6(H2O)(MeCN)4]?[Ni2Na2(chp)6(H2O)2(MeCN)4] 2.3. The salt Na(chp) (0.318 g, 2.10 mmol) was added to a stirred solution of hydrated nickel(II) chloride (0.250 g, 1.05 mmol) and Na(O2CCH2Ph) (0.166 g, 1.05 mmol) in MeOH (30 ml). The solution was stirred for 24 h, filtered and evaporated to dryness, leaving a green paste.Crystallisation from MeCN produced crystals of complex 2.3. FAB mass spectrum (significant peaks, possible assignments): m/z 1117, [Ni2Na2(chp)6(MeCN)4- (H2O)]1; 806, [Ni2Na2(chp)6]1; 655, [Ni2Na(chp)4]1; and 339, [NiNa(chp)2]1. [{Ni2Na2(chp)6(H2O)}n] 4. This was prepared in a similar fashion to that of complex 1 using Na(chp) in place of Na(mhp). Recrystallisation from ethyl acetate gave green crystals of 4. FAB mass spectrum (significant peaks, possible assignments): m/z 766, [NiNa2(chp)4(H2O)]1; 614, [NiNa(chp)4- (H2O)]1; 485, [NiNa(chp)3(H2O)]1; 444, [Ni(chp)3]1; 303, [Na2(chp)2]1; 280, [Na(chp)2]1; 210, [NiNa(chp)]1; 187, [Ni- (chp)]1; and 152, [Na(chp)]1.[{Co2Na2(chp)6(H2O)}n] 5. This was prepared in a similar fashion to that of complex 1 using Na(chp) in place of Na(mhp) and anhydrous cobalt(II) chloride in place of anhydrous nickel(II) chloride. Recrystallisation from MeCN gave pink crystals of 5.No significant FAB mass spectral peaks. [{Co2Na2(chp)6}n] 6. Anhydrous cobalt(II) chloride (0.25 g, 1.9 mmol) and Na(chp) (0.59 g, 3.9 mmol) were added to a solution of sodium formate (0.14 g, 1.9 mmol) dissolved in MeOH (40 ml), and the reaction mixture stirred for 24 h. The solution was then filtered and evaporated to dryness producing a purple2658 J. Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 Table 1 Analytical data * for compounds 1 to 7 Analysis (%) Complex 122 .3 4567 [Ni4Na4(mhp)12(Hmhp)2] [Ni2Na2(chp)6(H2O)(MeCN)4] [Ni2Na2(chp)6(H2O)(MeCN)4]? [Ni2Na2(chp)6(H2O)2(MeCN)4] [{Ni2Na2(chp)6(H2O)}n] [{Co2Na2(chp)6(H2O)}n] [{Co2Na2(chp)6}n] [{Co2Na2(mhp)6(Hmhp)(H2O)}n] C 54.6 (54.7) 40.6 (40.8) 40.6 (40.5) 37.8 (37.8) 37.8 (37.8) 38.4 (38.5) 54.4 (53.7) H 4.6 (4.7) 2.9 (2.9) 2.7 (2.9) 2.1 (2.1) 2.0 (2.1) 1.9 (1.9) 5.4 (4.8) N 10.5 (10.6) 12.1 (12.5) 11.9 (12.4) 8.8 (8.8) 8.6 (8.8) 8.8 (9.0) 9.7 (10.4) Yield (%) 70 75 6 38 10 8 12 * Calculated values are given in parentheses.paste which was dried under vacuum overnight. The paste was extracted with ethyl acetate (15 ml) and purple crystals of 6 formed after 1 d. FAB mass spectrum (significant peaks, possible assignments): m/z 806, [Co2Na2(chp)5]1; 655, [Co2Na- (chp)4]1; 504, [Co2(chp)3]1; 339, [CoNa(chp)2]1; and 303, [Na2(chp)2]1. [{Co2Na2(mhp)6(Hmhp)(H2O)}n] 7. Anhydrous cobalt(II) acetate (0.34 g, 1.9 mmol) and Na(mhp) (0.55 g, 4.2 mmol) were stirred together in dry CH2Cl2 under N2 for 24 h during which time the solution became very dark purple.It was filtered and the filtrate evaporated to dryness under reduced pressure, leaving a purple paste. The paste was dissolved in toluene and placed at 25 8C. A small number of pink crystals grew over a period of 4 weeks. No significant FAB mass spectral peaks. Crystallography Crystal data and data collection and refinement parameters for compounds 1–7 are given in Table 2, selected bond lengths and angles in Tables 3–7. Data collection and processing.Data were collected on a Stoë Stadi-4 four-circle diVractometer equipped with an Oxford Cryosystems low-temperature device,3 using graphite-monochromated Cu-Ka radiation for complexes 1, 2, 2.3, 4 and 5, and Mo-Ka radiation for 6 and 7; w–q scans for 1, 2 and 6, w scans for 2.3 and 5, and w–2q scans for 4 and 7. Data were corrected for Lorentz-polarisation factors. Semiempirical absorption corrections based on azimuthal measurements4 were applied to data for 2, 2.3 and 7; lower crystal quality led to an absorption correction using DIFABS5 being applied to data for 1 and 4.Complex 5 crystallised as very thin plates and the crystal was mounted on a short length (0.2 mm) of glass wool in order to minimise background scatter. Although it is not normally advisable for cobalt compounds, Cu-Ka radiation was used for data collection for 5 because of its higher intensity relative to Mo-Ka; peak profiles were broad and w scans of width (2.2 1 0.15 tan q) were used.The data for 5 were corrected for absorption by numerical integration based on face-indexing. Structure analysis and refinement. All structures were solved by direct methods using SHELXS 866 or SIR 92 7 and completed by iterative cycles of DF syntheses and full-matrix leastsquares refinement. In 7 one of two dicobalt moieties in the polymer is disordered about a crystallographic inversion centre. All non-H atoms were refined anisotropically with global rigid body and rigid bond restraints applied to the ligands in 5 and 7.Pyridone ligands were also restrained to have similar geometries in 5 and 7. Fourier-diVerence syntheses were employed in positioning idealised methyl-hydrogen atoms which were assigned isotropic thermal parameters [U(H) = 1.5 Ueq(C)]. Ring H atoms were included in idealised positions, allowed to ride on their parent C atoms (C]H 0.93 Å), and assigned isotropic thermal parameters [U(H) = 1.2 Ueq(C)].All refinements were against F2 and used SHELXL 93.8 CCDC reference number 186/1042. See http://www.rsc.org/suppdata/dt/1998/2657/ for crystallographic files in .cif format. Results Reaction of nickel(II) chloride with 2 equivalents of Na(mhp) leads, after recrystallisation from either MeCN or ethyl acetate, to [Ni4Na4(mhp)12(Hmhp)2] 1 in moderate yield. DiVraction studies reveal a centrosymmetric structure containing four chemically identical [Ni(mhp)3]2 units surrounding a central sodium ‘chair’ (Fig. 1). Each nickel is co-ordinated to three chelating mhp ligands, two of which also bind to one sodium through their exocyclic oxygen atoms, which are thus m-bridging. The oxygen atom of the third chelating mhp binds to two further sodiums, and hence is m3-bridging. Two Hmhp ligands are also found in the structure and show a novel binding mode, each bridging three Na atoms through the exocyclic oxygen atom alone.This mode has not to our knowledge previously been reported either for Hxhp (6-substituted 2-pyridone) or xhp ligands and represents bridging by the keto-tautomer of the pyridonate ligand as the ring N atom is protonated. The nickel centres in complex 1 have distorted octahedral coordination geometries as a result of the small bite angles of the chelating mhp ligands (Table 3). The Ni]N bond lengths are slightly shorter than the Ni]O bonds, although the two bond length ranges overlap allowing for statistical significance.The two crystallographic unique sodium sites show distinct geometries: Na(1) has five contacts to O-donors between 2.230 and Fig. 1 Structure of complex 1 in the crystal, showing the atomnumbering scheme. The molecule lies disposed about an inversion centreJ. Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 2659 Table 2 Experimental data for the X-ray diVraction studies of compounds 1, 2, 2.3, 3, 4, 5, 6 and 7 Formula M Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 T/K Z Dc/g cm23 Crystal shape and colour Crystal size/mm m/mm21 Unique data Observed data (Fo > 4sFo) Parameters Maximum D/s ratio R1, wR2c Weighting scheme,d w21 Goodness of fit Largest residuals/e Å23 1 C84H86N14Na4Ni4O14 1842.5 Triclinic P1� 10.818(4) 14.336(5) 16.281(6) 109.90(2) 100.46(2) 107.50(2) 2147.8 150.0(2) 1 a 1.424 Green plate 0.39 × 0.35 × 0.12 1.764 6251 4727 548 0.006 0.0512, 0.1450 s2(Fo 2) 1 (0.097P)2 1.013 10.404, 20.439 2 C38H32Cl6N10Na2Ni2O7 1116.8 Monoclinic I2 15.667(2) 9.1628(5) 16.4418(11) 90.525(10) 2360.1 220.0(2) 2 b 1.572 Green block 0.27 × 0.23 × 0.23 4.791 2929 2751 294 0.001 0.0338, 0.0882 s2(Fo 2) 1(0.0633P)2 1.038 10.238, 20.483 2.3 C38H32Cl6N10Na2Ni2O7? C38H34Cl6N10Na2Ni2O8 2251.7 Monoclinic P2/c 16.4177(15) 9.1699(5) 31.597(2) 91.096(11) 4756.0 220.0(2) 2 1.572 Green block 0.27 × 0.16 × 0.10 4.768 6939 4810 591 0.012 0.0534, 0.1263 s2(Fo 2) 1 (0.0316P)2 1 8.23P 1.051 10.284, 20.363 4 C30H20Cl6N6Na2Ni2O7 952.6 Monoclinic C2/c 21.407(6) 11.860(3) 13.912(2) 91.04(2) 3531.5 220.0(2) 4 1.792 Green lath 0.35 × 0.16 × 0.08 6.243 2584 1414 240 0.001 0.0515, 0.1091 s2(Fo 2) 1 (0.0358P)2 0.966 0.305, 20.289 5 C30H20Cl6Co2N6Na2O7 953.1 Monoclinic C2/c 21.582(5) 11.898(3) 13.873(3) 91.45(2) 3561.2 220.0(2) 4 1.778 Purple rod 0.66 × 0.27 × 0.21 1.463 3134 2259 240 0.001 0.0438, 0.1084 s2(Fo 2) 1 (0.0612P)2 1.013 0.620, 20.595 6 C15H9CoN3NaO3 467.5 Triclinic P1� 7.261(2) 11.316(2) 12.105(3) 112.204(10) 93.119(14) 105.192(13) 875.5 220.0(2) 2 1.773 Purple plate 0.35 × 0.18 × 0.08 1.484 3100 1959 235 0.001 0.0538, 0.1196 s2(Fo 2) 1 (0.0502P)2 0.981 0.529, 20.613 7 C42H45Co2N7Na2O8 939.7 Triclinic P1� 11.618(5) 12.931(5) 17.540(6) 87.86(10) 71.45(10) 64.36(10) 2236.6 293 2 1.395 Pink block 0.819 5783 3455 838 20.001 0.0478, 0.1453 s2(Fo 2) 1 (0.0504P)2 1 4.93P 1.012 0.308, 20.537 a The molecule lies on an inversion centre.b The molecule lies on a two-fold axis.c R1 based on observed data, wR2 on all unique data. d P = 1– 3[max(Fo 2, 0) 1 2Fc].2660 J. Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 2.487 Å with one longer contact of 3.008 Å to O(64); Na(2) has five contacts to O-donors which fall in the range 2.269 to 2.541 Å but no further long contacts. There is no diVerence between Na]O bonds to mhp or Hmhp ligands which is perhaps surprising. The Na]O (Hmhp) bonds average 2.35 Å which is extremely short for a Na]O (ketone) distance.This is still more surprising considering that this ketone oxygen is bridging three sodium sites, and is therefore formally pentavalent. Only four previous examples 9–12 have been reported of m3-oxygen bridges where the O-donor is derived from a ketone. These examples involve lithium,9,10 vanadium11 and cadmium.12 The metal polyhedron can be described as four linked cubes with each cube missing a vertex; for example, the cube comprising Na(1), O(67), Na(2a), O(63), Ni(1), O(61) and O(62).Each cube contains one Ni, two Na and four O atoms and is linked to two others, sharing an edge [Na(2), O(67) or symmetry equivalents] with one, and a vertex [Na(1) or Na(1a)] with a second. The average Na ? ? ? Na contact is 3.499 Å, with the closest Ni ? ? ? Ni contact being over 6 Å. Reaction of nickel(II) chloride with Na(chp) followed by crystallisation from MeCN gives a tetrametallic species [Ni2- Na2(chp)6(H2O)(MeCN)4] 2 which crystallises about a two-fold rotation axis.As in 1 the molecule contains [Ni(xhp)3]2 ‘complex ligands’ linked to a central sodium core (Fig. 2). Oxygen atoms from two of the chelating chp ligands each bridge to one sodium centre as in 1, however in 2 the third O atom [O(1R)] is bound only to the Ni, whereas in 1 it is m3 bridging. Again the nickel centres have distorted octahedral geometries due to the small bite angle of the chp ligands (Table 4).The sodium is fiveco- ordinate, bound to two chp oxygen donors, a m-bridging water and two molecules of MeCN. The bridging H2O is involved in two strong hydrogen bonds to the non-bridging O Table 3 Selected bond lengths (Å) and angles (8) for compound 1 Ni(1)]N(11) Ni(1)]N(12) Ni(1)]N(13) Ni(1)]O(61) Ni(1)]O(62) Ni(1)]O(63) Na(1)]O(66) Na(1)]O(62) Na(1)]O(67) Na(1)]O(61) Na(1)]O(67A) N(11)]Ni(1)]N(13) N(11)]Ni(1)]N(12) N(13)]Ni(1)]N(12) N(11)]Ni(1)]O(62) N(13)]Ni(1)]O(62) N(12)]Ni(1)]O(62) N(11)]Ni(1)]O(63) N(13)]Ni(1)]O(63) N(12)]Ni(1)]O(63) O(62)]Ni(1)]O(63) N(11)]Ni(1)]O(61) N(13)]Ni(1)]O(61) N(12)]Ni(1)]O(61) O(62)]Ni(1)]O(61) O(63)]Ni(1)]O(61) O(66)]Na(1)]O(62) O(66)]Na(1)]O(67) O(62)]Na(1)]O(67) O(66)]Na(1)]O(67A) O(62)]Na(1)]O(67A) O(67)]Na(1)]O(67A) O(66)]Na(1)]O(61) O(62)]Na(1)]O(61) O(67)]Na(1)]O(61) O(67A)]Na(1)]O(61) 2.044(3) 2.095(3) 2.061(3) 2.144(3) 2.107(3) 2.144(3) 2.230(3) 2.325(3) 2.352(3) 2.487(3) 2.367(3) 101.34(13) 105.24(14) 109.91(13) 102.73(12) 155.93(12) 64.00(12) 161.27(12) 64.01(12) 91.31(12) 92.29(11) 63.73(13) 96.77(12) 152.94(12) 93.26(11) 104.76(11) 109.58(12) 162.88(12) 87.14(11) 99.39(11) 97.08(11) 81.59(10) 93.19(11) 79.84(11) 85.98(10) 167.35(11) N(14) Ni(2)]N(15) Ni(2)]N(16) Ni(2)]O(64) Ni(2)]O(65) Ni(2)]O(66) Na(2)]O(64) Na(2)]O(65) Na(2)]O(63A) Na(2)]O(62A) Na(2)]O(67A) N(14)]Ni(2)]N(15) N(14)]Ni(2)]N(16) N(15)]Ni(2)]N(16) N(14)]Ni(2)]O(66) N(15)]Ni(2)]O(66) N(16)]Ni(2)]O(66) N(14)]Ni(2)]O(65) N(15)]Ni(2)]O(65) N(16)]Ni(2)]O(65) O(66)]Ni(2)]O(65) N(14)]Ni(2)]O(64) N(15)]Ni(2)]O(64) N(16)]Ni(2)]O(64) O(66)]Ni(2)]O(64) O(65)]Ni(2)]O(64) O(63A)]Na(2)]O(64) O(63A)]Na(2)]O(67A) O(64)]Na(2)]O(67A) O(63A)]Na(2)]O(65) O(64)]Na(2)]O(65) O(67A)]Na(2)]O(65) O(63A)]Na(2)]O(62A) O(64)]Na(2)]O(62A) O(67A)]Na(2)]O(62A) O(65)]Na(2)]O(62A) 2.067(3) 2.074(3) 2.088(3) 2.149(3) 2.117(3) 2.108(3) 2.302(3) 2.357(3) 2.269(3) 2.541(3) 2.327(3) 101.93(14) 112.80(14) 103.60(13) 100.83(13) 157.00(12) 64.02(12) 150.29(12) 64.05(12) 96.36(12) 96.85(11) 63.88(12) 102.72(12) 153.53(13) 90.23(12) 92.56(11) 151.38(13) 105.41(11) 102.48(12) 98.73(11) 82.84(11) 99.39(10) 78.29(10) 99.12(11) 82.79(10) 176.72(11) Symmetry transformation used to generate equivalent atoms: A 2x 1 1, 2y, 2z 1 1.atoms [O(1R) and its symmetry equivalent], [O(W) ? ? ? O(1R) 2.705 Å]. A quite diVerent reaction involving sodium phenylacetate, nickel(II) chloride and Na(chp) also led to a small yield of crystals containing complex 2, however here 2 was found cocrystallised with a further tetranuclear complex [Ni2Na2(chp)6(H2O)2- (MeCN)4] 3 (Fig. 3). There is a close similarity between 2 and 3, the significant diVerence being that two m-water molecules are found in 3 bridging between the central sodium centres. Each bridging H2O forms only one strong hydrogen bond in 3, whereas the bridging H2O in 2 formed two. The nickel coordination spheres are identical in the two complexes (Table 4) but the sodium environment in 3 is six-co-ordinate, involving two bridging water molecules, two MeCN ligands and two moxygen donors from chp ligands.Complex 2 is formed from reaction of nickel(II) chloride with Na(chp) followed by recrystallisation of the product from MeCN. If the same product is recrystallised from ethyl acetate a polymeric complex [{Ni2Na2(chp)6(H2O)}n] 4 is obtained (Fig. 4). The change from 2 may be caused by the absence of co-ordinating solvate molecules which lead to quite diVerent metal co-ordination environments.In 4 the nickel site is sixco- ordinate, but there are only two chelating chp ligands bound to the metal with the remaining two sites occupied by two m-oxygen donors [O(23) and its symmetry equivalent] from trinucleating chp ligands (Table 5). The nickel site therefore has Fig. 2 Structure of complex 2 in the crystal, showing the atomnumbering scheme. The molecule lies disposed about a two-fold axis Fig. 3 Structure of complex 3 in the crystal of 2.3, showing the atomnumbering scheme. The molecule lies disposed about a two-fold axisJ. Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 2661 2 N- and 4 O-donors co-ordinated, rather than 3 N- and 3 O-donors as in complexes 1–3. The crystallographically unique sodium site is quite diVerent to those found in complexes 1–3. It is five-co-ordinate, involving one nitrogen and four oxygen atoms. The N-donor [N(13)] is derived from the chp ligand which provides the oxygen donor [O(23)] which bridges Ni(1) and Ni(1A), while the four oxygen donors are derived from a variety of ligands.Atom O(1) is from a m-bridging water molecule, similar to those found in 1 to 3, and bonds to two sodium centres. A further oxygen [O(21)] is m bridging between a Ni and Na, and is derived from a chelating chp. The final two oxygen donors are derived from chelating chp ligands, but also bridge between two sodium sites and are hence m3 bridging.The result is that two neighbouring sodium sites are bridged by three O atoms, and within the polymer there are alternating dinuclear nickel and dinuclear sodium fragments (Fig. 5). Extending the concept of nickel complexes acting as ligands for sodium centres, the polymer 4 could be regarded as consisting of [Ni2(chp)6]22 units ligating [Na2- (H2O)]21 units, where in 2 there are two [Ni(xhp)3]2 complex ligands binding to the [Na2(H2O)(MeCN)4]21 core.Table 4 Selected bond lengths (Å) and angles (8) for structures 2 and 2.3 2.3 Ni]N(1R) Ni]N(2R) Ni]N(3R) Ni]O(1R) Ni]O(2R) Ni]O(3R) Na]N(1E) Na]N(1F) Na]O(2R) Na]O(3RA) Na]O(W) Na]O(WA) N(3R)]Ni]N(2R) N(3R)]Ni]N(1R) N(2R)]Ni]N(1R) N(3R)]Ni]O(2R) N(2R)]Ni]O(2R) N(1R)]Ni]O(2R) N(3R)]Ni]O(1R) N(2R)]Ni]O(1R) N(1R)]Ni]O(1R) O(2R)]Ni]O(1R) N(3R)]Ni]O(3R) N(2R)]Ni]O(3R) N(1R)]Ni]O(3R) O(2R)]Ni]O(3R) O(1R)]Ni]O(3R) O(2R)]Na]O(W) O(2R)]Na]O(3RA) O(W)]Na]O(3RA) O(2R)]Na]N(1E) O(W)]Na]N(1E) O(3RA)]Na]N(1E) O(2R)]Na]N(1F) O(W)]Na]N(1F) O(3RA)]Na]N(1F) N(1E)]Na]N(1F) O(2R)]Na]O(WA) O(W)]Na]O(WA) O(3RA)]Na]O(WA) N(1E)]Na]O(WA) N(1F)]Na]O(WA) 2 2.086(3) 2.062(3) 2.038(4) 2.115(3) 2.107(3) 2.144(3) 2.434(5) 2.473(6) 2.334(3) 2.425(3) 2.342(4) 102.20(13) 105.42(14) 106.17(13) 92.28(14) 63.95(12) 161.52(14) 157.89(12) 99.46(13) 63.80(13) 101.20(12) 63.99(12) 151.75(13) 101.42(13) 90.98(13) 98.02(11) 81.78(10) 161.77(13) 80.52(10) 92.08(14) 109.9(2) 98.02(15) 106.03(14) 149.2(2) 87.27(14) 99.6(2) Molecule 1 (º3)* 2.078(4) 2.069(4) 2.057(4) 2.100(4) 2.121(3) 2.144(4) 2.631(7) 2.428(6) 2.294(4) 2.414(4) 2.425(4) 2.404(5) 101.3(2) 108.8(2) 103.2(2) 93.18(15) 63.44(15) 156.6(2) 158.79(15) 99.8(2) 64.27(15) 97.77(14) 63.60(15) 155.4(2) 100.3(2) 96.56(2) 97.00(14) 94.74(14) 169.21(15) 74.75(14) 91.1(2) 84.5(2) 90.3(2) 99.0(2) 166.3(2) 91.5(2) 95.4(2) 81.9(2) 84.8(2) 74.75(14) 166.7(2) 96.8(2) Molecule 2 (º2) 2.046(4) 2.059(4) 2.066(4) 2.139(4) 2.123(4) 2.108(3) 2.438(6) 2.468(6) 2.479(4) 2.347(4) 2.328(4) 105.3(2) 102.6(2) 106.5(2) 98.8(2) 64.17(15) 158.4(2) 153.7(2) 100.4(2) 63.8(2) 97.65(14) 64.23(15) 160.8(2) 91.8(2) 100.33(14) 92.65(15) 79.66(12) 161.9(2) 82.68(12) 97.3(2) 112.8(2) 93.2(2) 88.1(2) 148.6(2) 105.2(2) 97.4(2) Symmetry transformation used to generate equivalent atoms: A for 2 2x, y, 2z; for 2.3 (molecule 1) 2x 2 1, 2y 2 2, 2z 2 1; for 2.3 (molecule 2) 2x, y, 2z 2 ��� .* To match atom labels to those in Fig. 3 add 3 to the label for N or O in column 1, i.e. N(1R) º N(4R). A similar reaction, but using cobalt in place of nickel, leads to the complex [{Co2Na2(chp)6(H2O)}n] 5, which is isostructural with 4. The metal–ligand bonds in 5 are slightly longer than in 4, e.g. Co]O bonds average 2.123 Å in 5, while Ni]O bonds Fig. 4 The tetranuclear Co2Na2 building-block from which the polymer 5 is constructed. Complex 4 is isostructural and the numbering scheme shown is common to both Table 5 Selected bond lengths (Å) and angles (8) for compounds 4 and 5 4 5 M(1)]O(21) M(1)]O(22) M(1)]O(23) M(1)]O(23A) M(1)]N(11) M(1)]N(12) Na(1)]O(1) Na(1)]O(22) Na(1)]O(21B) Na(1)]O(22C) Na(1)]N(13C) Na(1A)]O(1) O(23)]M(1)]O(23A) O(23)]M(1)]N(12) O(23A)]M(1)]N(12) O(21)]M(1)]O(23) O(21)]M(1)]O(23A) O(21)]M(1)]N(12) N(11)]M(1)]O(23) N(11)]M(1)]O(23A) N(11)]M(1)]N(12) N(11)]M(1)]O(21) O(22)]M(1)]O(23) O(22)]M(1)]O(23A) O(22)]M(1)]N(12) O(22)]M(1)]O(21) O(22)]M(1)]N(11) O(22)]Na(1)]O(1) O(22)]Na(1)]O(21B) O(1)]Na(1)]O(21B) O(22)]Na(1)]O(22C) O(1)]Na(1)]O(22C) O(21B)]Na(1)]O(22C) O(22)]Na(1)]N(13C) O(1)]Na(1)]N(13C) O(21B)]Na(1)]N(13C) O(22C)]Na(1)]N(13C) M = Ni 2.117(4) 2.182(4) 2.052(4) 2.035(5) 2.118(5) 2.052(5) 2.365(6) 2.380(5) 2.366(5) 2.407(5) 2.491(6) 2.365(6) 78.7(2) 96.8(2) 165.5(2) 98.4(2) 93.6(2) 100.6(2) 161.3(2) 97.7(2) 90.9(2) 63.4(2) 101.6(2) 103.9(2) 63.4(2) 155.7(2) 97.1(2) 79.4(2) 137.6(3) 72.4(2) 86.0(3) 78.0(2) 117.2(3) 136.7(3) 139.8(2) 82.7(3) 86.4(3) Co 2.152(7) 2.197(6) 2.057(7) 2.084(7) 2.180(8) 2.119(8) 2.372(8) 2.371(7) 2.377(8) 2.439(8) 2.513(8) 2.372(8) 79.9(3) 96.4(3) 165.1(3) 98.9(3) 93.1(3) 101.7(3) 160.9(3) 98.3(3) 89.8(3) 62.1(2) 101.7(3) 104.2(3) 62.2(2) 155.1(2) 97.1(3) 79.33(15) 137.5(2) 72.1(2) 86.5(2) 78.80(15) 116.9(2) 135.7(2) 141.1(2) 83.5(2) 86.2(2) Symmetry transformations used to generate equivalent atoms: A 2x 1 1, 2y, 2z 1 2; B x, 2y, z 2 ��� ; C 2x 1 1, y, 2z 1 ��� .2662 J.Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 average 2.096 Å in 4 (Table 5). There are no further significant diVerences between the structures. Iture contains sodium formate, crystallisation of the paste from ethyl acetate leads to a further polymer, [{Co2Na2(chp)6}n] 6 in low yield, which does not contain coordinated solvent or water (Figs. 6 and 7). The [M2(chp)6]22 complex ligand is similar to this fragment in 3 and 4, with each Co bound to two chelating chp ligands and two m-oxygen donors from further pyridonates. The sodium site is quite diVerent from that in 3 and 4, although it is probably best described as five-co-ordinate, involving one Cl-, one N- and three O-donors (Table 6).The N atom [N(13)] is again derived from the pyridonate which also provides the m-oxygen atom which bridges Co(1) and Co(1A), and the oxygen atoms are again two m3 donors [O(21) and symmetry equivalents] which bridge two Na and one Co, and one oxygen [O(22)] which bridges to a Co.The fifth site in 3 and 4 is provided by a bridging water molecule, however as this group is absent in 6 a chlorine atom from chp [Cl(2) or symmetry equivalents] occupies the fifth site with a contact of 3.07 Å. This appears to be a new binding mode for the chp ligand: chelating to one metal through the O- and Fig. 5 A fragment of the one-dimensional polymer 5; 4 is isostructural Fig. 6 The tetranuclear Co2Na2 building-block from which the polymer 6 is constructed, also showing the atom-numbering scheme Fig. 7 A fragment of the one-dimensional polymer 6 N-donors, bridging to a second metal through the oxygen atom and contacting a third metal through the chlorine. We have thus far been unable to isolate the cobalt equivalents for the molecular species 1–3, however we have isolated a further polymeric complex [{Co2Na2(mhp)6(Hmhp)(H2O)}n] 7 which was prepared in very low yield from a reaction involving cobalt acetate and Na(mhp) (Fig. 8).The structure is extremely disordered, with two positions observed for one cobalt and four mhp ligands in the asymmetric unit. This disorder renders bond lengths and angles of dubious significance (Table 7), however the connectivity of the structure is clearly established. It is related to that of 4 and 5. The cobalt sites in 7 are coordinated to two chelating mhp ligands, and two m-oxygen atoms which bridged to a symmetry equivalent cobalt site.Thus Fig. 8 One of the two crystallographically independent Co2Na2 building-blocks from which the polymer 7 is constructed, showing the atom-numbering scheme Fig. 9 A fragment of the one-dimensional polymer 7 Table 6 Selected bond lengths (Å) and angles (8) for compound 6 Co(1)]O(23) Co(1)]O(23A) Co(1)]O(22) Na(1)]O(22A) Na(1)]O(22B) Na(1)]Cl(2B) O(23)]Co(1)]O(23A) O(23A)]Co(1)]N(11) N(11)]Co(1)]N(12) O(23)]Co(1)]O(22) N(11)]Co(1)]O(22) O(23)]Co(1)]O(21) N(11)]Co(1)]O(21) O(22)]Co(1)]O(21) O(22A)]Na(1)]O(21B) O(22A)]Na(1)]N(13C) O(21B)]Na(1)]N(13C) O(21)]Na(1)]Cl(2B) N(13C)]Na(1)]Cl(2B) O(21)]Na(1)]Cl(3C) N(13C)]Na(1)]Cl(3C) 2.045(4) 2.083(4) 2.168(4) 2.289(4) 2.409(4) 3.073(3) 83.4(2) 150.1(2) 88.0(2) 90.6(2) 119.7(2) 107.4(2) 61.5(2) 161.36(14) 119.2(2) 140.6(2) 94.8(2) 173.07(14) 89.34(13) 103.04(13) 50.96(12) Co(1)]O(21) Co(1)]N(11) Co(1)]N(12) Na(1)]O(21) Na(1)]N(13C) Na(1)]Cl(3C) O(23)]Co(1)]N(11) O(23)]Co(1)]N(12) O(23A)]Co(1)]N(12) O(23A)]Co(1)]O(22) N(12)]Co(1)]O(22) O(23A)]Co(1)]O(21) N(12)]Co(1)]O(21) O(22A)]Na(1)]O(21) O(21)]Na(1)]O(21B) O(21)]Na(1)]N(13C) O(22a)]Na(1)]Cl(2B) O(21B)]Na(1)]Cl(2B) O(22A)]Na(1)]Cl(3C) O(21B)]Na(1)]Cl(3C) Cl(2B)]Na(1)]Cl(3C) 2.195(4) 2.131(4) 2.152(5) 2.327(4) 2.539(5) 3.298(3) 104.1(2) 152.1(2) 98.7(2) 88.7(2) 61.7(2) 88.52(14) 100.5(2) 99.1(2) 95.37(14) 96.6(2) 78.30(12) 80.54(11) 90.21(12) 142.29(12) 83.47(8) Symmetry transformations used to generate equivalent atoms: A 2x 1 1, 2y 1 1, 2z 1 1; B 2x, 2y 1 1, 2z 1 1; C x 2 1, y, z; D x 1 1, y, z.J.Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 2663 [Co2(mhp)6]22 can be recognised as a structural feature similar to [M2(chp)6]22 in 4 and 5. The sodium unit in 7 is diVerent, in that each five-co-ordinate Na is bound to two m-O-donors from mhp ligands chelating to cobalt, the N-donor of a mhp ligand which provides the O atom which bridges between cobalt centres, a m-oxygen from water and finally a m-oxygen from an Hmhp ligand.This final ligand represents the main diVerence between 7 and 4 and 5 with [Na2(Hmhp)(H2O)] units each bound by two chelating ‘complex ligands’ of formula [Co2- (mhp)6]22. As each complex ligand binds to two Na-containing fragments, a polymer results (Fig. 9). Discussion The structures described illustrate how Na atoms can become involved in linking complexes of other metals into more complicated frameworks. The ability of the sodium to adopt many diVerent co-ordination geometries is important in this role.Within the seven structures discussed here several diVerent coordination geometries are adopted which are all more or less distorted. In compound 2 there is one crystallographically unique sodium site, and it has a five-co-ordinate square-based pyramidal environment. A similar five-co-ordinate site [Na(2)] is found in 1. The five bond lengths in these sites fall in the range from 2.269 to 2.542 Å, and bond angles involving the atom at the vertex of the pyramidal are between 82.1 and 109.98 while those within the square base are between 78.3 and 106.08 when cis and between 149.2 and 176.78 for trans.In compound 1 the Na(1) site can be regarded as somewhat between five- and six-co-ordinate. Five bonds between 2.230 and 2.487 Å are found, plus one further contact of 3.008 Å. The angles are consistent with a distorted octahedron, with trans angles between 162.9 and 178.08 and cis angles between 68.5 and 109.68.In compound 3 the sodium is six-co-ordinate, with no markedly longer bonds and angles close to that required for an octahedron. In complexes 4 and 5 the question of co-ordination number becomes more debatable. The sodium site in both compounds has five contacts to O- or N-donors between 2.365 and 2.491 Å, however the angles at Na bear no relation to those of either a square-based pyramid or a trigonal bipyramid. In particular there is no angle near 1808, with the largest angle found 141.18.Table 7 Selected bond lengths (Å) for compound 7 Ordered cobalt site Co(1)]O(24) Co(1)]O(24A) Co(1)]O(25) 2.053(4) 2.061(4) 2.124(4) Co(1)]O(26) Co(1)]N(15) Co(1)]N(16) 2.137(5) 2.145(4) 2.147(5) Disordered cobalt site Co(2)]O(21) Co(2)]O(21B) Co(2)]O(22) Co(2)]O(23) Co(2)]N(12) Co(2)]N(13) 2.042(7) 2.049(10) 2.128(7) 2.125(10) 2.177(7) 2.148(7) Co(29)]O(219) Co(29)]N(129) Co(29)]O(239) Co(29)]O(219A) Co(29)]N(139) Co(29)]O(229) 2.056(7) 2.101(7) 2.140(10) 1.821(7) 2.132(7) 2.159(7) Sodium Na(1)]O(1) Na(1)]O(21) Na(1)]O(22) Na(1)]O(27) Na(1)]N(11) Na(1)]O(23A) Na(1)]O(229) Na(1)]O(279) Na(1)]N(119) 2.325(6) 2.628(7) 2.182(7) 2.461(10) 2.640(8) 2.646(8) 2.470(8) 2.266(10) 2.290(7) Na(2)]O(1) Na(2)]O(26) Na(2)]O(27) Na(2)]O(24B) Na(2)]O(25B) Na(2)]N(14B) Na(2)]O(279) Na(1)]O(239A) 2.424(6) 2.309(5) 2.401(11) 2.817(5) 2.424(6) 2.421(5) 2.299(11) 2.360(10) Symmetry transformations used to generate equivalent atoms: A 2x 1 1, 2y 1 1, 2z 1 1; B 2x, 2y 1 2, 2z 1 2.There is a much longer contact to a Cl atom [Na(1) ? ? ? Cl(1) 3.360 Å] and inclusion of this atom within the sodium coordination sphere reveals a geometry close to that of a trigonal prism. The three O-donors which bridge between sodium sites [O(1), O(22) and O(22A)] form one trigonal plane which eclipses a plane described by O(21C), N(13A) and Cl(1). Whether the Na(1) ? ? ? Cl(1) contact is suYciently short to be regarded as a bond is debatable, however in any case the coordination geometry appears to be based on a trigonal prism. In complex 6 there is again a question of co-ordination number.There are four bonds to O- or N-donors between 2.289 and 2.539 Å. The shortest contact to a Cl atom is 3.073 Å, and inclusion of this atom within the sodium co-ordination sphere leads to a geometry closer to trigonal bipyramidal than square pyramidal, with the trans angle between the axial atoms [Cl(2B) and O(21)] 173.18. The angles between the axial and equatorial atoms fall between 78.3 and 99.18 while those within the equatorial plane are between 94.8 and 140.68.The next nearest contact to a Cl is 3.298 Å but the geometry which results if this atom is regarded as bound to the Na is extremely irregular, and therefore it is probably simplest to regard this Na as five-coordinate. In complex 7 the sodium geometries are still more distorted. The Na(1) site is disordered, however both models have very similar geometries as has the Na(2) site.The best description of the sites, which are six-co-ordinate, is as a plane of five oxygen donors capped with an additional N-donor at one apical site of a pentagonal bipyramid. The degree of distortion is considerable, mainly caused by the N- and one O-donor being derived from a chelating mhp ligand in every case. This has a concomitant eVect of increasing all other angles involving these atoms.The angles found in the structures fall within the following ranges: between the two donors of the chelating mhp 50.3– 50.88, other angles to the apical N-donor 91.1–115.68 and the cis angles within the pentagonal plane 64.2–91.48. The degree of distortion in each of these sites makes the description chosen debatable but the important point is that Na adopts a variety of five- and six-co-ordinate geometries readily and therefore imposes no constraints on the overall structure of the co-ordination framework.This geometric flexibility, which reflects the non-directional electrostatic nature of the interaction between Na and donor atoms, seems complementary to those of transition metals. For these reasons sodium can play a primary role in mixed-metal complex formation, producing an array of structural motifs. Those relevant to the new compounds described here are discussed below. In many of these structures, and certainly in compounds 1–7, the motif shown in Scheme 1 is important, where a chelating ligand is attached to the 3d metal, and the oxygen donor of the chelator is shared with the sodium site.This motif is common for SchiV-base ligands, and examples are found where Na is surrounded by one,13 two14 or three 15 SchiV-base complexes of 3d metals. Several other ‘sodium-centred’ mixed-metal complexes have the core shown in Scheme 2. This common core can be described as a raft of six face-sharing ‘broken cubes’, and has been found with nuclearities {NaCu6}8,1 {NaCo6}9,16 {NaFe6}1017 and {Na3Co4}11.18 Sodium occupies the M1 site in all these structures, as well as M2 and M3 in 11.Only six molecular structures are known which contain this core,19 but which do not contain sodium. Other structures contain sub- or super-sets of Scheme 2. Examples are a {Na2V2} 20 12 structure with M1 and M2 as Na and M4 and M6 as V. Changing the ligand from 2-sulfenyl- Scheme 1 The bridging mode found in many heterometallic compounds featuring Na.D = N, O or S, and O and D belong to a chelating ligand; TM = a transition metal Na O TM D2664 J. Chem. Soc., Dalton Trans., 1998, Pages 2657–2664 phenol to catechol leads to an ‘isomeric’ {Na2V2} structure 13 21 with the metal positions exchanged. An expanded version of the core in 12 exists for {Na2Fe4},22 with the additional two metals bound at M7 and at a vertex related to M7 through an inversion centre. The {Na4Ni4} core in 1 can be understood in a similar way with Na at M1, M2, M4 and M6 while Ni is found at M5 and M7 and their inverse relations.The Na2M2 cores of 12 and 13 can be combined in a third variant, a chain-polymer based around a {Na2M2} repeat.23 Most of the structures reported here belong to this category, with some variation in the exact propagation mode involved in the formation of each polymer. The range of structural types reported here is therefore quite limited, and all can be related to Scheme 2.Previous work indicates an apparently limitless range of structural arrangements are possible from such mixed-metal inorganic syntheses. Examples include: a triangular structure with a {Na3Co3} nuclearity 24 where at the centre of the molecule a pair of m3-oxides bridge three sodiums, which are surrounded by three peripheral cobalt atoms; a tetrahedral pattern of MnII surrounding a m4-oxo at the centre of a {Na2Mn4} structure 25 which has sodium sites capping two opposite edges of the tetrahedron; an octahedral {Na4Cr2} O-bridged core,26 where the sodium sites are mutually trans; and perhaps the most beautiful architecture which is an {Na4Cu4} complex27 with four parallel O]CuI]O linear units which are capped by threeco- ordinate sodiums, resulting in a pair of edge-sharing (Cu3Na2) trigonal bipyramids.Acknowledgements We are grateful to the EPSRC for funding for a diVractometer and for studentships (to E. K.B., L. M. G. and S. G. H.). References 1 A. J. Blake, R. O. Gould, P. E. Y. Milne and R. E. P. Winpenny, J. Chem. Soc., Chem. Commun., 1991, 1453. Scheme 2 A heptanuclear metal–oxygen core (M = Na or a transition metal) O M4 O M5 O O O M2 O M1 O M3 O M6 O M7 O 2 E. K. Brechin, R. O. Gould, S. G. Harris, S. Parsons and R. E. P. Winpenny, J. Am. Chem. Soc., 1996, 118, 11 293. 3 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105. 4 A. C. T. North, D. C. Phillips and F.S. Matthews, Acta. Crystallogr., Sect. A, 1968, 24, 351. 5 N. Walker and D. Stuart, DIFABS, Acta. Crystallogr., Sect. A, 1983, 39, 158. 6 G. M. Sheldrick, SHELXS 86, University of Göttingen, 1986. 7 A. Altomare, G. Cascarano, C. Giacovazzo and A. Guagliardi, SIR 92, J. Appl. Crystallogr., 1993, 26, 343. 8 G. M. Sheldrick, SHELXL 93, University of Göttingen, 1993. 9 B. Bogdanovic, C. Krüger and B. Wermeckes, Angew. Chem., Int. Ed. Engl., 1980, 19, 817. 10 P. G. Williard and J. M. Salvino, Tetrahedron Lett., 1985, 26, 3931. 11 D. C. Bradley, M. B. Hursthouse, A. N. de M. Jelfs and R. L. Short, Polyhedron, 1983, 2, 849. 12 I. G. Dance, R. G. Garbutt and M. L. Scudder, Inorg. Chem., 1990, 29, 1571. 13 S. Gambarotta, F. Arena, C. Floriani and P. F. Zanazzi, J. Am. Chem. Soc., 1982, 104, 5082. 14 F. Arena, C. Floriani and P. F. Zanazzi, J. Chem. Soc., Chem. Commun., 1987, 183. 15 F. Corazza, C. Floriani and M. Zehnder, J. Chem. Soc., Chem. Commun., 1986, 1270. 16 S. McConnell, M. Motevalli and P. Thornton, Polyhedron, 1995, 14, 459. 17 A. Caneschi, A. Cornia, A. C. Fabretti, S. Foner, D. Gatteschi, R. Grandi and L. Schenetti, Chem. Eur. J., 1996, 2, 1379. 18 F. Jiang, M. Hong, X. Xie, R. Cao, B. Kang, D. Wu and H. Liu, Inorg. Chim. Acta, 1995, 231, 153. 19 S. L. Heath and A. K. Powell, Angew. Chem., Int. Ed. Engl., 1992, 31, 191; Z. Sun, P. K. Gantzel and D. N. Hendrickson, Inorg. Chem., 1996, 35, 6640; M. Tesmer, B. Muller and H. Vahrenkamp, Chem. Commun., 1997, 721; E. K. Brechin, S. G. Harris, A. Harrison, S. Parsons, A. G. Whitaker and R. E. P. Winpenny, Chem. Commun., 1997, 653. 20 L. Weng, B. Kang, X. Chen, M. Hong, X. Lei, Y. Hu and H. Liu, Chin. J. Chem. (Huaxue Xuebao) (Engl. Transl.), 1993, 11, 30. 21 C. Floriani, M. Mazzanti, A. Chiesi-Villa and C. Guastini, Angew. Chem., Int. Ed. Engl., 1988, 27, 576. 22 F. Arena, C. Floriani, A. Chiesi-Villa and C. Guastini, J. Chem. Soc., Chem. Commun., 1986, 1369. 23 V. G. Kessler, N. Ya Turova, A. V. Korolev, A. I. Yanovskii and Yu. T. Struchkov, Mendeleev Commun., 1991, 89. 24 W. Klaui, A. Muller, W. Eberspach, R. Boese and I. Goldberg, J. Am. Chem. Soc., 1987, 109, 164. 25 E. Gallo, E. Solari, S. De Angelis, C. Floriani, N. Re, A. Chiesi-Villa and C. Rizzoli, J. Am. Chem. Soc., 1993, 115, 9850. 26 J. J. H. Edema, S. Gambarotta, F. van Bolhuis and A. L. Spek, J. Am. Chem. Soc., 1989, 111, 2142. 27 A. P. Purdy and C. F. George, Polyhedron, 1995, 14, 761. Received 7th May 1998; Paper 8/03442G

ISSN:1477-9226    

DOI:10.1039/a803442g

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2665–2669 2665 Carbonyl sulfide insertion into Group 2 metal–isopropoxide bonds; synthesis and crystal structures of [Mg(OCSOPri)2(PriOH)4]? 2PriOH and [{Sr(OCSOPri)2(PriOH)2}n] Izoldi K. Bezougli,a Alan Bashall,b Mary McPartlin b and D. Michael P. Mingos a a Department of Chemistry, Imperial College of Science Technology and Medicine, South Kensington, London, UK SW7 2AY b School of Applied Chemistry, University of North London, Holloway Road, London, UK N7 8DB The insertion reactions of COS into the metal–alkoxide bonds of some alkaline-earth metals have been investigated.The resulting compounds have been studied spectroscopically (IR, 1H and 13C NMR), and by analytical techniques (TGA). The crystal structures of the strontium complex [{Sr(OCSOPri)2(PriOH)2}n] and the magnesium complex [Mg(OCSOPri)2(PriOH)4]?2PriOH have shown that whereas the former has a three dimensional structure the latter is monomeric.The renewed interest in the chemistry of alkaline-earth-metal complexes can be attributed primarily to their potential use as molecular precursors in the preparation of thin films via metal organic chemical vapour deposition (MOCVD), for electroceramics such as superconductors and other speciality electronic materials.1 Also, in the course of the understanding of their co-ordination chemistries several researchers have recently reported some significant developments in the chemistry of these materials, for example the structural characterisation of a wide range of organomagnesium complexes,2 the facile cleavage of a wide range of organic ligands by the Group 2 metals and their complexation and characterisation of Group 2 metal alkoxides and b-diketonates.3 Although several papers have examined the eVects of diVerent solvents on the volatility and reactivity of alkaline-earth-metal alkoxides,4 the insertion reactions of small unsaturated molecules such as SO2, CO2, COS and CS2 with alkaline-earth-metal alkoxides have been relatively neglected compared with the corresponding reactions with transition metals.5–8 This may be attributed to the diYculty of synthesizing and handling Group 2 metal alkoxides which often led to their poor characterisation.9 The first example of such an insertion compound to be structurally characterised was [{Ca(O2SOMe)2(MeOH)2}n],10 resulting from the insertion of SO2 into the calcium–methoxide bonds and later we also described the thiocarbonato-bridged dimer [{Ca(OCSOMe)2- (MeOH)3}2] resulting from the insertion reaction of COS into the calcium–methoxide bonds.11 More recently we reported the crystal structure of the magnesium and strontium ethoxide analogues [Mg(OCSOEt)2(EtOH)4] 1 and [Sr3(OCSOEt)6- (EtOH)8] 2,12 and in this paper we report the extension of this research to the two Group 2 metal thiocarbonato isopropoxides [Mg(OCSOPri)2(PriOH)4] 3 and [{Sr(OCSOPri)2(PriOH)2}n] 4.The structures of 3 and 4 have been confirmed by single crystal X-ray crystallography. Results and Discussion The two thiocarbonato complexes 3 and 4 were synthesized in a similar way to that described previously for 1 and 2.1 The crystalline PriOH-solvated metal isopropoxides [{Mg(OPri)2- (PriOH)4}n] and [{Sr(OPri)2(PriOH)4}n] were suspended in PriOH and COS bubbled through the suspension at room temperature to give an exothermic reaction which reached completion within 10 min.The products were isolated as colourless crystals after reducing the volume of the solvent. On the basis of single-crystal X-ray studies, analytical data and spectroscopic measurements, the products have been formulated as the monomeric and polymeric thiocarbonato complexes [Mg(OCSOPri) 2(PriOH)4] 3 and [{Sr(OCSOPri)2(PriOH)2}n] 4, respectively. Complexes 3 and 4 are moisture sensitive, but may be O Sr Sr O O O Et Et Et Et O O S C O O C S Et C S O Et O Sr O O O Et Et Et Et O O S C O O C S Et C S O Et O Et O Et H H H H H H H H Sr S S O O C C O Sr S O C O Sr S O C O O O O Pri Pri Pri O Pri H O Pri H Pri Pri Pri Mg O O C S O R C S O R O O O O R R R H H H H R 1, R = Et 3, R = Pri 2 42666 J.Chem. Soc., Dalton Trans., 1998, Pages 2665–2669 stored indefinitely under an inert atmosphere at room temperature without losing COS, although some reversible desolvation occurs. They are soluble in alcohols, co-ordinating and polar organic solvents, but have poor solubilities in hydrocarbons.Crystal structures [Mg(OCSOPri)2(PriOH)4]?2PriOH 3. The X-ray structural analysis established a discrete monomeric molecule as the insertion product of reaction of COS with [Mg(OPri)2- (PriOH)4]. The centrosymmetric structure is shown in Fig. 1 and selected bond lengths and angles are listed in Table 1. Insertion of the COS into two Mg]OPri bonds has resulted in two trans-monodentate isopropyl thiocarbonato ligands bonded through the oxygen atom of the COS group, and an octahedral co-ordination is completed by four oxygen atoms of equatorial PriOH molecules.The overall structure is very similar to that of complex 1 formed by insertion of COS into Mg]OEt, but unlike 1 the solid state structure of 3 incorporates two add- Fig. 1 Molecular structure of [Mg(OCSOPri)2(PriOH)4]?2PriOH 3, formed by the insertion of COS into the M]O bonds of magnesium isopropoxide Table 1 Selected bond lengths (Å) and angles (8) for complex 3 Mg]O(1bI) Mg]O(3a) Mg]O(1c) S(1a)]C(2a) C(2a)]O(4a) C(5a)]C(6a) O(1b)]C(2b) C(2b)]C(4b) C(2c)]C(4c) O(1d)]C(2d) C(2d)]C(3d) O(1bI)]Mg]O(1b) O(1b)]Mg]O(3a) O(1b)]Mg]O(3aI) O(1bI)]Mg]O(1c) O(3a)]Mg]O(1c) O(1bI)]Mg]O(1cI) O(3a)]Mg]O(1cI) O(1c)]Mg]O(1cI) C(2b)]O(1b)]Mg 2.060(2) 2.060(2) 2.067(3) 1.712(4) 1.333(4) 1.492(6) 1.442(4) 1.507(6) 1.500(6) 1.429(6) 1.472(7) 180.0 92.68(9) 87.32(9) 90.24(11) 89.46(10) 89.76(11) 90.54(10) 180.0 137.8(2) Mg]O(1b) Mg]O(3aI) Mg]O(1cI) C(2a)]O(3a) O(4a)]C(5a) C(5a)]C(7a) C(2b)]C(3b) O(1c)]C(2c) C(2c)]C(3c) C(2d)]C(4d) O(1bI)]Mg]O(3a) O(1bI)]Mg]O(3aI) O(3a)]Mg]O(3aI) O(1b)]Mg]O(1c) O(3aI)]Mg]O(1c) O(1b)]Mg]O(1cI) O(3aI)]Mg]O(1cI) C(2a)]O(3a)]Mg C(2c)]O(1c)]Mg 2.060(2) 2.060(2) 2.067(3) 1.238(4) 1.459(4) 1.501(5) 1.494(6) 1.438(5) 1.505(6) 1.455(8) 87.32(9) 92.68(9) 180.0 89.76(11) 90.54(10) 90.24(11) 89.46(10) 144.5(2) 128.9(2) Symmetry transformation used to generate equivalent atoms: I 2x, 2y, 2z.itional alcohol molecules per asymmetric unit. Each of these symmetry related PriOH molecules forms hydrogen bonds with the sulfur atom of an axial OCS(OPri)2 ligand [O(H1d) ? ? ? S(1a) 2.188, O(H1bI) ? ? ? S(1a) 2.150 Å] and the hydrogen of an equatorial PriOH ligand [H(1c) ? ? ? O(1d) 2.650 Å] thus forming a bridge between these two ligands (Fig. 1). In the structure of complex 1 the symmetry related Mg]OCSOEt bond lengths (Mg]O 2.036 Å) were found to be signifi- cantly shorter than the two independent Mg]O(H)Et lengths [mean 2.083(2) Å].In the present study all the Mg]O bond lengths in 3 are very similar lying in the range 2.060(2)–2.067(3) Å; the apparent lengthening of the Mg]O bond to the alkyl thiocarbonato ligand in 3 may be attributed to steric factors associated with the greater bulk of the isopropyl group compared to the ethyl group in 1. [{Sr(OCSOPri)2(PriOH)2}n] 4. X-Ray structural analysis of complex 4, [{Sr(OCSOPri)2(PriOH)2}n], established the presence of a polymeric material.Each strontium atom lies on a site of C2 symmetry and is co-ordinated by oxygen atoms of two symmetry related PriOH ligands [O(1b) and O(1bIII)] and is chelated by the sulfur and oxygen atoms of two symmetry related isopropyl thiocarbonato ligands as illustrated in Fig. 2. Selected bond lengths and angles are listed in Table 2. Eightco- ordination of the strontium centre is completed by bonds from two oxygen atoms [O(3aI) and O(13aII)] of two OCS- (OPri)2 ligands on adjacent strontium atoms, generated by the crystal symmetry, resulting in a three-dimensional polymeric structure. This contrasts with the discrete trimeric molecules [Sr3(OCSOEt)6(EtOH)8] 2 formed from the insertion reaction of COS with strontium ethoxide.Polymer formation in 4 may be facilitated by the co-ordination of only two relatively bulky PriOH molecules to each strontium centre compared to the four smaller ethanol molecules which co-ordinate each of the terminal strontium atoms in 2.Bidentate bonding of an alkyl thiocarbonato ligand via sulfur and oxygen atoms has been observed previously for the COS insertion products [{Ca(OCSOMe)2(MeOH)3}2] 11 and [Sr3(OCSOEt)6(EtOH)8] 2, and in each of these it is the oxygen of the COS group which is involved. In contrast, the alkyl thiocarbonato ligand in 4 forms a chelate ring involving the sulfur and isopropyl oxygen atoms of OCS(OPri)2. The Sr]O distance in the sulfur–oxygen chelate ring of 4 is considerably longer [Sr]O(4a) 2.796(3) Å] than that in 2 [Sr]O 2.785(5) Å], consistent with the weaker donor character of the ‘ether’ oxygen in OCS(OPri)2 in 4; the Sr]S(1a) length of 3.061(1) Å in the chelate ring of 4 is slightly longer than that of 3.025(3) Å in 2.Infrared spectroscopy The main stretching vibrations of complexes 1–4 are summarised in Table 3 and have been assigned on the basis of previously published infrared data for similar compounds.13 In view of their similar crystal structures it is not surprising that 1 and 3 exhibit similar stretching vibrations in their infrared spectra.Both complexes are co-ordinated in a monodentate fashion to the terminal oxygen of the (OCS)OR2 ligand, exhibiting only one set of vibrations for the ligand. The band at 1551 cm21 of 3 has been assigned to the n(C]] O) stretching mode of the COS moiety and the band at 1162 cm21 to the n(C]] S) stretching mode.These bands are similar to those observed for 1 at 1554 and 1174 cm21, respectively. The IR spectra of complexes 2 and 4 are clearly diVerent and imply dissimilar alkyl thiocarbonato co-ordination modes. Complex 4 shows only one n(CO) at 1579 cm21 and a n(C]S) stretch at 950 cm21. In contrast, 2 exhibits a more complicated infrared spectrum, with three CO stretching frequencies at 1620, 1598 and 1568 cm21, suggesting multiple co-ordination modes for the OCS(OEt)2 ligand.Nevertheless, the infrared spectra were measured as solids and therefore theJ. Chem. Soc., Dalton Trans., 1998, Pages 2665–2669 2667 above analysis has to be treated with some caution because of possible solid state eVects. Nuclear magnetic resonance The 1H and 13C-{1H} NMR spectroscopic studies for complexes 1–4 were made in (CD3)2SO solution and are summarised in Tables 4 and 5, respectively. Fig. 2 Part of the three dimensional structure of the COS insertion product [{Sr(OCSOPri)2(PriOH)2}n] 4, showing the unique m3-bridging mode adopted by the alkylthiocarbonato ligand Table 2 Selected bond lengths (Å) and angles (8) for complex 4 Sr]O(3aI) Sr]O(1bIII) Sr]O(4a) Sr]S(1a) S(1a)]C(2a) C(2a)]O(4a) O(4a)]C(5a) C(5a)]C(7a) C(2b)]C(3b) O(3aI)]Sr]O(3aII) O(3aII)]Sr]O(1bIII) O(3aII)]Sr]O(1b) O(3aI)]Sr]O(4a) O(1bIII)]Sr]O(4a) O(3aI)]Sr]O(4aIII) O(1bIII)]Sr]O(4aIII) O(4a)]Sr]O(4aIII) O(3aII)]Sr]S(1a) O(1b)]Sr]S(1a) O(4aIII)]Sr]S(1a) O(3aII)]Sr]S(1aIII) O(1b)]Sr]S(1aIII) O(4aIII)]Sr]S(1aIII) C(2a)]S(1a)]Sr C(2a)]O(4a)]Sr C(2b)]O(1b)]Sr 2.505(3) 2.540(3) 2.796(3) 3.0607(13) 1.703(4) 1.370(5) 1.478(5) 1.492(7) 1.473(8) 140.3(2) 81.14(11) 70.92(11) 125.10(9) 163.43(9) 86.56(10) 96.57(11) 81.45(14) 125.66(8) 80.21(9) 85.81(7) 74.18(7) 144.78(8) 51.72(6) 88.1(2) 106.9(2) 135.0(3) Sr]O(3aII) Sr]O(1b) Sr]O(4aIII) Sr]S(1aIII) C(2a)]O(3a) O(3a)]SrIV C(5a)]C(6a) O(1b)]C(2b) C(2b)]C(4b) O(3aI)]Sr]O(1bIII) O(3aI)]Sr]O(1b) O(1bIII)]Sr]O(1b) O(3aII)]Sr]O(4a) O(1b)]Sr]O(4a) O(3aII)]Sr]O(4aIII) O(1b)]Sr]O(4aIII) O(3aI)]Sr]S(1a) O(1bIII)]Sr]S(1a) O(4a)]Sr]S(1a) O(3aI)]Sr]S(1aIII) O(1bIII)]Sr]S(1aIII) O(4a)]Sr]S(1aIII) S(1a)]Sr]S(1aIII) C(2a)]O(3a)]SrIV C(5a)]O(4a)]Sr 2.505(3) 2.540(3) 2.796(3) 3.0607(13) 1.213(5) 2.505(3) 1.481(7) 1.422(5) 1.482(8) 70.92(11) 81.14(11) 89.9(2) 86.56(10) 96.57(11) 125.10(9) 163.43(9) 74.18(7) 144.78(8) 51.72(6) 125.66(8) 80.21(9) 85.81(7) 125.62(5) 154.5(3) 135.9(2) Symmetry transformations used to generate equivalent atoms: I x 1 1– 4, 2y 1 1– 4, z 1 1– 4; II 2x 2 1– 4, y 2 1– 4, z 1 1– 4; III 2x, 2y, z; IV x21– 4, 2y1 1– 4, z2 1– 4.Table 3 Characteristic infrared data (cm21) for complexes 1–4. The spectra were obtained in the solid state as Nujol mulls Complex 1234 n(CO) 1554 1620, 1598, 1568 1551 1579 n(C]] S) 1174 1176, 1152 1162 — n(C]S) — 972, 946 — 950 n(C]OR) 1091, 1050 1051 1095 1103, 1066, 1041 The proton NMR studies clearly diVerentiate the alcohol and alkyl thiocarbonato ligands.A doublet of quartets for the CH of the co-ordinated PriOH molecules was observed at d 3.77 for 3 and 3.78 for 4, while a multiplet at d 4.65 for 3 and 4.75 for 4 was assigned to the CH of the isopropyl thiocarbonato moieties. The two diVerent CH3 environments of the PriOH and isopropyl thiocarbonato moieties appeared in the same region as two superimposed doublets for 3 (d 1.05) and as two distinct doublets for 4 (d 1.05 and 1.09).Similar proton NMR assignments were obtained for 1 and 2.12 The 13C-{1H} NMR spectra of complexes 1–4 were assigned in a similar manner. A characteristic signal for the thiocarbonato carbon, OCS(OR)2, was observed at higher chemical shifts at around d 184 for all complexes. The data clearly reveal the presence of two distinct types of alkyl environments, one corresponding to the bound alcohol molecules and one corresponding to the alkyl thiocarbonato ligands.Darensbourg et al.14,15 described similar chemical shift positions for COS insertion products with tungsten alkoxides. Thermogravimetric analysis (TGA) and diVerential scanning calorimetry (DSC) The diVerential scanning calorimetry (DSC) trace for the magnesium complex 3 shows an endothermic peak between 19 and 130 8C. This is mirrored in the TGA plot with a 75% weight loss, corresponding to loss of the co-ordinated PriOH and the loss of the COS molecules. The subsequent weight loss percentage ca. 15% represents the decomposition of Mg(OPri)2 to MgO between 130 and 800 8C leaving a residue of 9.30% (calc. 9.05%). A similar DSC trace to that of 3 was obtained for compound 4. An endotherm between 100 and 160 8C corresponds to a 40% weight loss on the TGA plot and is assigned to the loss of the COS molecules. The subsequent weight loss of ca. 30% represents the decomposition of Sr(OPri)2 to SrO between 160 and 800 8C leaving a residue of 30.1% (calc. 31.82%).The TGA plot of 4 between 20 and 100 8C suggests that all the PriOH molecules attached to the metal were lost before the measurement was made. Similarly, from the residue of the TGA of 3 it is deduced that only three molecules of PriOH remained in the solid sample when the TGA analysis was made. This may be due to the fact that the compounds were exposed to air for some seconds before they were transferred to the TGA apparatus for the measurements. The observation of an endotherm for 3 and an exotherm for 4 in closely related complexes suggests that the thermodynamics of the decomposition process is sensitive to the nuclearity of the complex and the mode of co-ordination of the isopropyl thiocarbonato ligand.Similar simultaneous thermal analysis (STA) was obtained for 1 and 2.12 In all cases the alcohol molecules were released first below 100 8C, followed by the COS molecules at ca. 140 8C, leaving a residue of the metal oxide above 700 8C.Table 4 Proton NMR data (d) for complexes 1–4 in (CD3)2SO Complex 1234 OH of ROH 4.41 (t) 4.39 (t) 4.40 (d) 4.44 (d) CH2 or CH of OCSOR 3.74 (q) 3.80 (q) 4.65 (q) 4.75 (m) CH2 or CH of ROH 3.41 (dq) 3.42 (dq) 3.77 (dq) 3.78 (dq) CH3 of OCSOR/ ROH 1.04 (t) 1.04 (t) 1.05 (d) 1.09/1.05 (d) Table 5 13C-{1H} NMR data (d) for complexes 1–4 in (CD3)2SO Complex 1234 OCSOR 183.42 184.76 183.53 185.59 CH2 or CH of OCSOR 59.53 60.13 65.99 67.73 CH2 or CH of ROH 56.66 56.67 63.12 63.33 CH3 of ROH 19.19 19.19 26.59 26.58 CH3 of OCSOR 15.62 15.15 23.60 23.462668 J.Chem. Soc., Dalton Trans., 1998, Pages 2665–2669 Mass spectrometric studies using fast atom bombardment (FAB) (positive and negative ion) techniques for all complexes 1–4 yielded complicated fragmentation patterns. Molecular ions were not observed probably because of the poor volatilities and thermal stabilities of these complexes. Nevertheless, the complexes exhibited fragments which have been assigned in the Experimental section.Conclusion The research described in this paper has demonstrated that the insertion of COS into Group 2 metal alkoxide bonds leads to a range of crystalline derivatives which are reasonably air stable and soluble in organic solvents. Both IR and NMR techniques were very useful in characterising these complexes. The IR spectra revealed diVerent co-ordination modes of the COS(OR)2 ligand, while in the 1H and 13C-{1H} NMR spectra the diVerent OR2 and COS(OR)2 ligands were clearly diVerentiated.The NMR spectra confirmed the stoichiometries of the compounds. The thermogravimetric results indicate that the insertion of the COS molecule may be reversed at high temperatures. The crystallographic structural determinations of complexes 3 and 4 together with those previously studied, 1 and 2, have revealed a variety of bonding motifs. The OCS(OR)2 ligand appears to be a flexible ligand which can use its alternative donor sites and bridging modes to co-ordinate to a wide range of metals and thereby form a series of crystalline derivatives with low molecular weights.Moving down the group from Mg21 in 3 to Sr21 in 4, the increase in the ionic radii of the metal ion [from Mg21 (0.72) to Sr21 (1.18 Å)] 16 leads to a preference for higher co-ordination numbers which are achieved by oligomerisation or polymerisation processes. That results in the introduction of the S and OR groups into the co-ordination sphere in order to make up the co-ordination number. This was also observed in the previously described complexes 1 and 2.Therefore, the (OCS)OR2 ligand adopts a monodentate co-ordination mode in 1 and 3 and bidentate bridging co-ordination modes in 2 and 4. The diVerent co-ordination modes adopted by the OCS(OEt)2 ligand in 1 and 2 have been previously discussed.12 In the polymeric structure associated with 4 a new bonding mode is observed which involves a novel ligand bridge A and, for the first time, all three potential donor atoms of the ligand are simultaneously bonded to metal ions.Experimental General procedures All manipulations were carried out under an atmosphere of dry nitrogen using standard glove-box (Miller-Howe FF160) and Schlenk techniques. All solvents were rigorously dried and deoxygenated by standard procedures. The samples for NMR, infrared studies and thermogravimetric analyses were handled in a glove-box, but those for microanalyses were not.The complexes desolvate when exposed to air and that has led to some diVerences concerning the extent to which PriOH molecules of crystallisation were detected by the spectroscopic and analytical techniques. Instrumentation Infrared spectra were recorded on a Perkin-Elmer FTIR 1720 spectrometer using Nujol mulls between 25 × 4 mm CsI plates. The Nujol was dried with 4 Å molecular sieves prior to use (and stored in a glove-box); the samples were protected from M S OR C O M (A) the atmosphere by an O-ring-sealed Presslok holder (Aldrich Chemicals).The NMR spectra were recorded on a JEOL GS 270 MHz spectrometer; 1H NMR was referenced internally to the residual 1H impurity present in the deuteriated solvent. Chemical shifts are recorded in parts per million (d) relative to SiMe4 (d 0) using (CD3)2SO (d 2.52), 13C NMR spectra to (CD3)2SO (d 40.6). Controlled thermal analyses of the complexes were investigated using a Polymer Laboratories 1500H simultaneous thermal analyser, controlled by an OmniPro 486DX-33 personal computer.The mass of the samples was between 10 and 20 mg. The measurements were carried out in alumina crucibles under an atmosphere of flowing (25 cm3 min21) nitrogen gas, using heating rates of 5 8C min21. Starting materials Strontium granules and dibutylmagnesium from Aldrich Chemicals Co. were used as received. Preparations Tetra(isopropylalcohol)bis(O-isopropyl thiocarbonato)magnesium, [Mg(OCSOPri)2(PriOH)4] 3.Dibutylmagnesium in heptane (10 cm3, 1 M, 10 mmol) was added to PriOH (30 cm3) at 240 8C resulting in an exothermic reaction. The reaction mixture was slowly warmed to room temperature, the volume of the solvent was reduced until all the heptane was removed and precipitation of a white solid [{Mg(OPri)2(PriOH)4}n] was observed. Addition of PriOH (30 cm3) resulted in a suspension of the magnesium isopropoxide. Carbonyl sulfide gas was bubbled through the suspension at room temperature for 30 min.The solution was stirred for 12 h until all the alkoxide had dissolved to aVord a yellow solution. A crystalline solid was isolated after cooling the solution to 220 8C (yield: 3.80 g, 75.7%) (Found: C, 46.4; H, 9.1. Calc. for C20H48MgO9S2: C, 46.9; H, 9.3%) [analysis based on Mg(OCSOPri)2(PriOH)4? H2O]. IR (cm21) (Nujol): 3369s (br), 3197s (br), 1551s, 1342w, 1301w, 1261w, 1162s, 1095s, 950m, 875w, 817m, 694m, 524w and 437w.NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.65 [q, OCSOCH(CH3)2, 2 H], 4.40 [d, (CH3)CHOH, 4 H], 3.77 [dq, (CH3)2CHOH, 4 H] and 1.05 (d, CH3, OCSOPri/PriOH, 36 H); 13C-{1H} (67.94 MHz), d 183.53 (OCSOPri), 65.99 [OCSOCH( CH3)2], 63.12 [(CH3)2CHOH], 26.59 [CH(CH3)2OH] and 23.60 [OCSOCH(CH3)2]. Mass spectrum: (positive-ion FAB), m/z 120, (OCS)2; 152, (OCSO)2; 290, [Mg(OCSOPri)2(OC)] and 443, [Mg(OCSOPri)2(PriOH)3]1; (negative-ion FAB), m/z 175, [Mg(OCSO)2]2; 192, [Mg(OCSO)2O] and 223, [Mg- (OCSO)2O3]2.Di(isopropyl alcohol)bis(O-isopropyl thiocarbonato)strontium, [{Sr(COSOPri)2(PriOH)2}n] 4. Strontium metal (0.97 g, 11.07 mmol) was suspended in PriOH (40 cm3) and the mixture was stirred for 12 h at room temperature resulting in dissolution of the metal and evolution of hydrogen gas, yielding a transparent slightly purple solution. The solution was filtered and COS gas bubbled through the clear solution for 15 min at room temperature, resulting in an exothermic reaction to give a clear yellow solution.The solution was stirred for 1 h and a crystalline solid isolated by cooling the solution to 220 8C (yield: 3.81 g, 77.3%) (Found: C, 36.6; H, 5.9. Calc. for C14H30O6S2Sr: C, 37.7; H, 6.8%) [analysis is based on Sr(OCSOPri)2(PriOH)2]. IR (cm21) (Nujol): 3295m, 1579m, 1303w, 1159m, 1147m, 1103m, 1066m, 1041m, 973w, 950s, 917m, 852m, 815m, 723m and 682w. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.75 [m, OCSOCH(CH3)2, 2 H], 4.44 [d, (CH3)2CHOH, 2 H], 3.78 [dq, (CH3)2CHOH, 2 H], 1.09 [d, OCSOCH(CH3)2, 6 H] and 1.05 [d, (CH3)2CHOH, 6 H]; 13C-{1H} (67.94 MHz), d 185.59 (OCSOR), 67.73 [OCSOCH( CH3)2], 63.33 [(CH3)2CHOH], 26.58 [(CH3)2CHOH] and 23.46 [OCSOCH(CH3)2].Mass spectrum: (positive-ion FAB), m/z 88, [Sr]1; 105, [SrO]1; 120, [SrO2]; 165, [Sr(OCSO)]1 and 267, [Sr(OCSOPri)(PriOH)]1.J. Chem. Soc., Dalton Trans., 1998, Pages 2665–2669 2669 Table 6 Crystal data and structure refinement for complexes 3 and 4 Formula Mr Crystal system Space group a/Å b/Å c/Å b/8 U/Å3 Z Dc/Mg m23 F(000) Crystal size/mm l/Å m(Mo-Ka)/mm21 q Range/8 hkl Ranges Reflections collected Unique reflections Data, restraints, parameters Goodness of fit on F2, S Transmission factors (minimum/maximum) Final R1, wR2 I > 2s(I) all data Weighting, w21 Largest peak and hole/e Å23 3 C26H62MgO10S2 623.19 Monoclinic P21/n 11.332(2) 14.535(2) 11.538(2) 93.800(9) 1896.2(5) 2 1.091 684 0.24 × 0.40 × 0.50 0.710 73 0.200 2.26 to 25.00 213 to 13, 217 to 17, 213 to 13 7028 3342 3339, 0, 186 1.017 0.187/0.982 0.0631, 0.1538 0.1094, 0.1836 s2(Fo)2 1 (0.082 60P)2 1 1.0358P 0.609, 20.353 4 C14H30O6S2Sr 446.12 Orthorhombic Fdd 2 17.283(3) 15.544(5) 15.769(6) 4236(2) 8 1.399 1856 0.56 × 0.38 × 0.36 0.710 73 2.764 2.18 to 25.09 220 to 20, 218 to 18, 218 to 18 1964 1890 1889, 1, 110 1.039 0.254/0.356 0.0338, 0.0735 0.0411, 0.0765 s2(Fo)2 1 (0.049 95P)2 1 0.00P 0.739, 20.347 S = [Sw(Fo 2 2 Fc 2)2/(n 2 p)]� �� , R1 = S |Fo| 2 |Fc| /S|Fo|, wR2 = Sw(Fo 2 2 Fc 2)2/S[w(Fo 2)2]� �� , P = [max(Fo 2, 0) 1 2(Fc 2)]/3 where n = number of reflections and p = total number of parameters.X-Rgraphy Data were collected using a Siemens P4 diVractometer equipped with a Siemens LT2 low temperature device with graphite monochromated radiation Mo-Ka using w–2q scans at 198 K. No significant decay in the intensity of three standard reflections measured after every 100 was observed.The data were corrected for Lorentz-polarisation factors and for absorption (y scans). The crystal data, data collection and refinement details are summarised in Table 6. Both structures were solved by direct methods and in each case all non-hydrogen atoms were located from subsequent Fourier-diVerence syntheses. All non-hydrogen atoms were assigned anisotropic thermal parameters and refined using fullmatrix least squares on Fo 2.17 The hydrogen atoms for each of the compounds were included at calculated positions with C]H bond distances of 1.00 and 0.98 Å for the methine and methyl groups, respectively.The hydroxyl hydrogens of both compounds were located from Fourier-diVerence syntheses. During refinement all the hydrogens were allowed to ride on their parent atom (the positional parameters of the directly located hydrogens were not refined) and were assigned isotropic thermal parameters equal to 1.2 Ueq of the parent atom for the methine groups and 1.5 Ueq for the methyl and hydroxyl groups.CCDC reference number 186/1051. See http://www.rsc.org/suppdata/dt/1998/2665/ for crystallographic files in .cif format. Acknowledgements The EPSRC is thanked for financial support and BP plc for endowing D. M. P. M.’s Chair. References 1 D. C. Bradley, Polyhedron, 1994, 13, 1111. 2 P. R. Markies, O. S. Akkerman, F. Bickelhaupt, W. J. J. Smeets and A. L. Spek, Adv. Organomet. Chem., 1991, 32, 147. 3 S. R. Drake, M. H.Chisholm, K. G. Caulton and J. C. HuVman, Inorg. Chem., 1991, 31, 1500; S. R. Drake, M. H. Chisholm, K. G. Caulton and K. Folting, J. Chem. Soc., Chem. Commun., 1990, 1345, 1498. 4 P. H. Dickenson, T. H. Geballe, A. Sanjurjo, D. Hildenbrand, G. Craig, M. Zisk, J. Collman, S. A. Banning and R. A. Sievers, J. Appl. Phys., 1989, 66, 444; C. I. M. A. Spee and A. Mackor, Science of Technology of Thin Film Superconductors, eds. R. D. McConnell and S. A. Wolf, New York, 1989, p. 281; S. Matsuno, F. Uchikawa and K. Yoshizaki, Jpn. J. Appl. Phys., 1990, 29, L947; J. M. Buriak, L. K. Cheatham, J. J. Graham, R. G. Gordon and A. R. Barron, Mater. Res. Soc. Symp. Proc., 1991, 204, 545. 5 G. J. Kubas, Acc. Chem. Res., 1994, 27, 183; W. A. Schenk, Angew. Chem., Int. Ed. Engl., 1987, 26, 98; R. Ros, G. Carturan and M. Graziani, Transition Met. Chem., 1975, 1, 13; G. R. Hughes, P. C. Minshall and D. M. P. Mingos, Transition Met. Chem., 1979, 4, 147; S. L. Randall, C.A. Miller, T. S. Janik, M. R. Churchill and J. D. Atwood, Organometallics, 1994, 13, 141. 6 K. K. Panday, Coord. Chem. Rev., 1995, 140, 37. 7 D. J. Darensbourg and R. A. Kudarovski, Adv. Organomet. Chem., 1983, 22, 129; T. Tsuda, S. Sanada, K. Ueda and T. Saegusa, Inorg. Chem., 1976, 15, 2329; T. Yamamoto, M. Kuboto and A. Yamamoto, Bull. Chem. Soc. Jpn., 1980, 53, 680; M. H. Chisholm, W. W. Reichert, F. A. Cotton and C. A. Murillo, J. Am. Chem. Soc., 1977, 99, 1652. 8 J.A. Ibers, Chem. Soc. Rev., 1982, 11, 57; K. K. Pandey and H. L. Nigam, Rev. Inorg. Chem., 1984, 6, 69. 9 W. E. Lindsell, Comprehensive Organometallic Chemistry, eds. G. Wilkinson, F. G. A. Stone and E. Abel, Pergamon, Oxford, 1982, vol. 1, ch. 3. 10 V. C. Arunasalam, I. Baxter, M. B. Hursthouse, K. M. A. Malik, D. M. P. Mingos and J. C. Plakatouras, J. Chem. Soc., Chem. Commun., 1994, 2695. 11 V. C. Arunasalam, D. M. P. Mingos, J. C. Plakatouras, I. Baxter, M. B. Hursthouse and K.M. A. Malik, Polyhedron, 1995, 14, 1105. 12 I. K. Bezougli, A. Bashall, M. McPartlin and D. M. P. Mingos, J. Chem. Soc., Dalton Trans., 1997, 287. 13 A. J. Goodsel and G. Blyholder, J. Am. Chem. Soc., 1972, 94, 6725. 14 D. J. Darensbourg, B. L. Mueller, C. J. BischoV, S. S. Chojnacki and J. H. Reibenspies, Inorg. Chem., 1991, 30, 2418. 15 D. J. Darensbourg, K. M. Sanchez, J. H. Reibenspies and A. L. Rheingold, J. Am. Chem. Soc., 1989, 111, 7094. 16 T. P. Hanusa, Chem. Rev., 1993, 93, 1023. 17 SHELXTL, PC version 5.03, Siemens Analytical Instruments Inc., Madison, WI, 1994. Received 6th April 1998; Paper 8/02595IJ. Chem. Soc., Dalton Trans., 1998, Pages 2665–2669 2669 Table 6 Crystal data and structure refinement for complexes 3 and 4 Formula Mr Crystal system Space group a/Å b/Å c/Å b/8 U/Å3 Z Dc/Mg m23 F(000) Crystal size/mm l/Å m(Mo-Ka)/mm21 q Range/8 hkl Ranges Reflections collected Unique reflections Data, restraints, parameters Goodness of fit on F2, S Transmission factors (minimum/maximum) Final R1, wR2 I > 2s(I) all data Weighting, w21 Largest peak and hole/e Å23 3 C26H62MgO10S2 623.19 Monoclinic P21/n 11.332(2) 14.535(2) 11.538(2) 93.800(9) 1896.2(5) 2 1.091 684 0.24 × 0.40 × 0.50 0.710 73 0.200 2.26 to 25.00 213 to 13, 217 to 17, 213 to 13 7028 3342 3339, 0, 186 1.017 0.187/0.982 0.0631, 0.1538 0.1094, 0.1836 s2(Fo)2 1 (0.082 60P)2 1 1.0358P 0.609, 20.353 4 C14H30O6S2Sr 446.12 Orthorhombic Fdd 2 17.283(3) 15.544(5) 15.769(6) 4236(2) 8 1.399 1856 0.56 × 0.38 × 0.36 0.710 73 2.764 2.18 to 25.09 220 to 20, 218 to 18, 218 to 18 1964 1890 1889, 1, 110 1.039 0.254/0.356 0.0338, 0.0735 0.0411, 0.0765 s2(Fo)2 1 (0.049 95P)2 1 0.00P 0.739, 20.347 S = [Sw(Fo 2 2 Fc 2)2/(n 2 p)]� �� , R1 = S |Fo| 2 |Fc| /S|Fo|, wR2 = Sw(Fo 2 2 Fc 2)2/S[w(Fo 2)2]� �� , P = [max(Fo 2, 0) 1 2(Fc 2)]/3 where n = number of reflections and p = total number of parameters.X-Ray crystallography Data were collected using a Siemens P4 diVractometer equipped with a Siemens LT2 low temperature device with graphite monochromated radiation Mo-Ka using w–2q scans at 198 K.No significant decay in the intensity of three standard reflections measured after every 100 was observed. The data were corrected for Lorentz-polarisation factors and for absorption (y scans). The crystal data, data collection and refinement details are summarised in Table 6. Both structures were solved by direct methods and in each case all non-hydrogen atoms were located from subsequent Fourier-diVerence syntheses.All non-hydrogen atoms were assigned anisotropic thermal parameters and refined using fullmatrix least squares on Fo 2.17 The hydrogen atoms for each of the compounds were included at calculated positions with C]H bond distances of 1.00 and 0.98 Å for the methine and methyl groups, respectively. The hydroxyl hydrogens of both compounds were located from Fourier-diVerence syntheses.During refinement all the hydrogens were allowed to ride on their parent atom (the positional parameters of the directly located hydrogens were not refined) and were assigned isotropic thermal parameters equal to 1.2 Ueq of the parent atom for the methine groups and 1.5 Ueq for the methyl and hydroxyl groups. CCDC reference number 186/1051. See http://www.rsc.org/suppdata/dt/1998/2665/ for crystallographic files in .cif format. Acknowledgements The EPSRC is thanked for financial support and BP plc for endowing D.M. P. M.’s Chair. References 1 D. C. Bradley, Polyhedron, 1994, 13, 1111. 2 P. R. Markies, O. S. Akkerman, F. Bickelhaupt, W. J. J. Smeets and A. L. Spek, Adv. Organomet. Chem., 1991, 32, 147. 3 S. R. Drake, M. H. Chisholm, K. G. Caulton and J. C. HuVman, Inorg. Chem., 1991, 31, 1500; S. R. Drake, M. H. Chisholm, K. G. Caulton and K. Folting, J. Chem. Soc., Chem. Commun., 1990, 1345, 1498. 4 P. H. Dickenson, T. H. Geballe, A. Sanjurjo, D.Hildenbrand, G. Craig, M. Zisk, J. Collman, S. A. Banning and R. A. Sievers, J. Appl. Phys., 1989, 66, 444; C. I. M. A. Spee and A. Mackor, Science of Technology of Thin Film Superconductors, eds. R. D. McConnell and S. A. Wolf, New York, 1989, p. 281; S. Matsuno, F. Uchikawa and K. Yoshizaki, Jpn. J. Appl. Phys., 1990, 29, L947; J. M. Buriak, L. K. Cheatham, J. J. Graham, R. G. Gordon and A. R. Barron, Mater. Res. Soc. Symp. Proc., 1991, 204, 545. 5 G. J. Kubas, Acc. Chem. Res., 1994, 27, 183; W. A. Schenk, Angew. Chem., Int. Ed. Engl., 1987, 2s, G. Carturan and M. Graziani, Transition Met. Chem., 1975, 1, 13; G. R. Hughes, P. C. Minshall and D. M. P. Mingos, Transition Met. Chem., 1979, 4, 147; S. L. Randall, C. A. Miller, T. S. Janik, M. R. Churchill and J. D. Atwood, Organometallics, 1994, 13, 141. 6 K. K. Panday, Coord. Chem. Rev., 1995, 140, 37. 7 D. J. Darensbourg and R. A. Kudarovski, Adv. Organomet. Chem., 1983, 22, 129; T. Tsuda, S. Sanada, K. Ueda and T. Saegusa, Inorg. Chem., 1976, 15, 2329; T. Yamamoto, M. Kuboto and A. Yamamoto, Bull. Chem. Soc. Jpn., 1980, 53, 680; M. H. Chisholm, W. W. Reichert, F. A. Cotton and C. A. Murillo, J. Am. Chem. Soc., 1977, 99, 1652. 8 J. A. Ibers, Chem. Soc. Rev., 1982, 11, 57; K. K. Pandey and H. L. Nigam, Rev. Inorg. Chem., 1984, 6, 69. 9 W. E. Lindsell, Comprehensive Organometallic Chemistry, eds. G. Wilkinson, F. G. A. Stone and E. Abel, Pergamon, Oxford, 1982, vol. 1, ch. 3. 10 V. C. Arunasalam, I. Baxter, M. B. Hursthouse, K. M. A. Malik, D. M. P. Mingos and J. C. Plakatouras, J. Chem. Soc., Chem. Commun., 1994, 2695. 11 V. C. Arunasalam, D. M. P. Mingos, J. C. Plakatouras, I. Baxter, M. B. Hursthouse and K. M. A. Malik, Polyhedron, 1995, 14, 1105. 12 I. K. Bezougli, A. Bashall, M. McPartlin and D. M. P. Mingos, J. Chem. Soc., Dalton Trans., 1997, 287. 13 A. J. Goodsel and G. Blyholder, J. Am. Chem. Soc., 1972, 94, 6725. 14 D. J. Darensbourg, B. L. Mueller, C. J. BischoV, S. S. Chojnacki and J. H. Reibenspies, Inorg. Chem., 1991, 30, 2418. 15 D. J. Darensbourg, K. M. Sanchez, J. H. Reibenspies and A. L. Rheingold, J. Am. Chem. Soc., 1989, 111, 7094. 16 T. P. Hanusa, Chem. Rev., 1993, 93, 1023. 17 SHELXTL, PC version 5.03, Siemens Analytical Instruments Inc., Madison, WI, 1994. Received 6th April 1998; Paper 8/02595I

ISSN:1477-9226    

DOI:10.1039/a802595i

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 2669–2673 2669 Structure, spectroscopic and electrochemical properties of novel binuclear ruthenium(II) copper(I) complexes with polypyridyl bridging ligands † Sonya M. Scott,a Keith C. Gordon *a and Anthony K. Burrell *b a Department of Chemistry, University of Otago, PO Box 56, Dunedin, New Zealand. E-mail: kgordon@alkali.otago.ac.nz b Department of Chemistry, IFS, Massey University, Private Bag 11222, Palmerston North, New Zealand Received 13th April 1999, Accepted 25th June 1999 Binuclear complexes of the type [(bpy)2Ru(BL)Cu(PPh3)2]3, where bpy = 2,2-bipyridine, BL = 2,3-di-2-pyridylpyrazine (dpp), 2,3-di-2-pyridylquinoxaline (dpq), or 6,7-dimethyl-2,3-di-2-pyridylquinoxaline (dpqMe2), were readily formed by the reaction of [Cu(PPh3)4] with mononuclear complexes [Ru(bpy)2(BL)]2.The binuclear complexes are stable in CH2Cl2 solution at concentrations above 103 mol dm3 having equilibrium constants for formation in the range 1000–2500 dm3 mol1.Single crystal structures for [(bpy)2Ru(dpp)Cu(PPh3)2]3 and [(bpy)2Ru(dpqMe2)Cu(PPh3)2]3 show distortions of the bridging ligand in the form of twisting and splaying of the ring systems. Electrochemical and UV/Visible data suggest the {Cu(PPh3)2} moiety has little a.ect in stabilising the BL p* orbital. Resonance Raman spectra show the bichromophoric nature of the visible absorptions of the heteroleptic complexes; both bpy and BL ligand vibrations are enhanced depending on the excitation wavelength.Observation of a Ru–N vibration suggests that the dominant transition in the visible region is Ru(dp).BL(p*) CT. Introduction There has been increased interest in recent years in the photochemical and electrochemical properties of supramolecular assemblies composed of mononuclear metal polypyridyl complexes.1 Such systems have applications in solar energy harvesting 2 and in molecular device technology.3 Of interest in the use of such assemblies is the possibility of programming them, through molecular design, so they may transduce energy in one particular direction.This is possible by using binuclear complexes with di.erent metals, by having di.erent terminal ligand substituents on each metal or by using asymmetric bridging ligands.4 A number of heterobimetallic systems have been studied using metals such as ruthenium(..)/osmium(..),5 rhenium(.)/ ruthenium(..) 6 or ruthenium(..) with chromium(...) and rhodium(...).7 All of these heterobimetallics have metal sites with six-co-ordinate octahedral systems.d6d8 Heterobimetallic systems based on ruthenium(..) and platinum(..) have also been reported.8 The ground and excited states of mononuclear complexes with these metals tend to have modest geometry changes upon photoexcitation into their metal-to-ligand charge-transfer (MLCT) excited states. Copper(.) polypyridyl complexes have signi.cant co-ordination di.erences between the ground and 3MLCT excited state,9 as the formation of an excited state results in a copper(..) centre which prefers a .ve- or six-co-ordinate geometry.The synthesis of a binuclear system containing a ruthenium and copper site would produce a complex in which the excited state properties of the copper centre may be programmed by the steric substituents on the ligand.10 The ruthenium site may be programmed using ligands which possess di.erent electronic e.ects.11 † Supplementary data available: UV/Visible spectra for 1 and 2.Available from BLDSC (No. SUP 57598, 1 pp.). See Instructions for Authors, 1999, Issue 1 (http://www.rsc.org/dalton). Experimental Complex synthesis Mononuclear complexes, [Ru(bpy)2(dpp)][BF4]2 1(BF4)2 (dpp = 2,3-di-2-pyridylpyrazine), [Ru(bpy)2(dpq)][BF4]2 2(BF4)2 (dpq = 2,3-di-2-pyridylquinoxaline) and [Ru(bpy)2- (dpqMe2)][BF4]2 3(BF4)2 (dpqMe2 = 6,7-dimethyl-2,3-di-2- pyridylquinoxaline), were prepared by literature procedures.12 [(bpy)2Ru(dpp)Cu(PPh3)2][BF4]3 4(BF4)3.This was prepared by the addition of four equivalents (0.58 g (0.48 mmol)) of [Cu(PPh3)4]BF4 to 0.1 g (0.12 mmol) of 1(BF4)2 in CH2Cl2 (10 mL). No change was observed from the original red-brown colour. The complex was recrystallised by diethyl ether di.usion and crystals collected. Yield 47% based upon ruthenium complex. Calc. for [(bpy)2Ru(dpp)Cu(PPh3)2][BF4]32CH2Cl2: C, 51.91; H, 3.63; N, 6.72.Found: C, 52.43; H, 3.42; N, 6.40%. 1H NMR (CDCl3): d 8.90 (m, 4 H); 8.12–8.38 (broad m, 8 H); 7.89–8.02 (broad m, 6 H); 7.45–7.72 (m, 8 H); 7.3–6.8 (m, 15 H, PPh3) and 5.30 (s, CH2Cl2). [(bpy)2Ru(dpq)Cu(PPh3)2][BF4]3 5(BF4)3. This was prepared in an analogous fashion to complex 4 and recrystallised by ether di.usion. Yield 32% based upon ruthenium complex. Calc. for [(bpy)2Ru(dpq)Cu(PPh3)2][BF4]3: C, 59.03; H, 3.90; N, 5.99. Found: C, 58.67; H, 3.71; N, 5.44%. 1H NMR (CDCl3): d 8.70 (broad m, 4 H); 8.48 (broad m, 4 H); 8.15–8.05 (broad m, 8 H); 7.67 (td, 2 H); 7.64 (td, 2 H); 7.58 (d, 2 H); 7.48–7.35 (m, 6 H) and 7.3–6.8 (m, 15 H, PPh3). [(bpy)2Ru(dpqMe2)Cu(PPh3)2][BF4]3 6(BF4)3.This was prepared in an analogous fashion to complex 4 and recrystallised by ether di.usion. Yield 54% based upon ruthenium complex. Calc. for [(bpy)2Ru(dpqMe2)Cu(PPh3)2][BF4]3CH2Cl2: C, 55.73; H, 3.89; N, 6.75. Found: C, 55.84; H, 4.04; N, 7.08%. 1H NMR (CDCl3): d 8.7 (broad m, 4 H); 8.5 (broad m, 4 H); 8.222670 J.Chem. Soc., Dalton Trans., 1999, 2669–2673 (td, 2 H); 8.16 (td, 2 H); 8.06 (d, 2 H); 8.02 (s, 2 H); 7.99 (d, 2 H); 7.7 (broad m, 4 H); 7.5 (m, 4 H); 7.2–6.8 (m, 15 H, PPh3) and 2.14 (s, 6 H, 2 CH3). Physical measurements The instrumentation used in the measurement of UV/Visible, electrochemical and resonance Raman properties and the protocols used have been described elsewhere.13 Crystallography Single crystals of complexes 4(BF4)32H2O2CH2Cl2 and 6(BF4)3H2OCH2Cl2 were grown by slow di.usion of diethyl ether into a dichloromethane solution of 4(BF4)3 or 6(BF4)3, respectively.Red plate shaped crystals with approximate dimensions 0.26 × 0.18 × 0.15 and 0.26 × 0.25 × 0.24 mm were secured to the ends of glass .bres with cyanoacrylate glue and cooled to 100 C in a nitrogen stream. Data collection, reduction, solution and re.nement were performed as previously described.13b,14,15 Crystal data for 4(BF4)32H2O2CH2Cl2.C72H56B3Cl4CuF12- N8O2Ru, M = 1694.03, monoclinic, space group P21/n, a = 12.905(3), b = 13.264(3), c = 45.357(4) Å, ß = 94.51(3), U = 7740(2) Å3, T = 173 K, Z = 4, µ(Mo-Ka) = 0.727 mm1, 14852 re.ections measured, 7242 unique (Rint = 0.1652) used in all calculations. The .nal Rw(Fo 2) = 0.1826 (R(Fo) = 0.0745). Crystal data for 6(BF4)3H2OCH2Cl2. C77H64B3Cl2CuF12- N8OP2Ru, M = 1675.24, triclinic, space group P1� , a = 11.676(6), b = 12.804(5), c = 27.340(12) Å, a = 103.48(4), ß = 90.68(3), .= 112.34(3), U = 3671(3) Å3, T = 173 K, Z = 2, µ(Mo-Ka) = 0.695 mm1, 7224 re.ections measured, 6767 unique (Rint = 0.0222) used in all calculations. The .nal Rw(Fo 2) = 0.1265 (R (Fo) = 0.0482). CCDC reference number 186/1541. See http://www.rsc.org/suppdata/dt/1999/2669/ for crystallographic .les in .cif format. Results and discussion It was found that the addition of [Cu(PPh3)4] to solutions of complexes 1, 2 and 3 resulted in a slight deepening in colour of the solutions.Fig. 1 shows the changes in the UV/Visible spectrum upon addition of [Cu(PPh3)4] to 3. Solutions of 1, 2 and 3 containing an excess of [Cu(PPh3)4] yielded crystalline samples of the binuclear complexes 4, 5 and 6 respectively. Attempts to precipitate the binuclear complexes resulted in powdered samples containing mono- and bi-nuclear materials Fig. 1 Changes in the UV/Visible spectrum, in CH2Cl2, upon addition of [Cu(PPh3)4] to 3.Initial concentration of 3 = 1 × 104 mol dm3; concentration of [Cu(PPh3)4] is (a) 0, (b) 1 × 104, (c) 2.5 × 104, (d) 5 × 104, (e) 1 × 102 and (f) 1 × 101 mol dm3. as a mixture. The paucity of the copper(.) to polypyridyl linkages also precluded the use of chromatography to purify the binuclear complexes. However the samples crystallised in pure form and were used for microanalysis and physical measurements. The complexes were stable in the solid state, and in solution they remained stable, at su.cient concentration (see below), for a period of days.The changes in the UV/Visible spectra, in CH2Cl2 solution, as a function of concentration of [Cu(PPh3)4] present provide a method of determining the equilibrium constant (K) for the formation of the binuclear complex: K = [(bpy)2Ru(BL)Cu(PPh3)2 3]/[Ru(bpy)2(BL)2] [Cu(PPh3)4 ]. The series of spectra in Fig. 1 shows how the binuclear species, 6, forms with increased concentrations of [Cu(PPh3)4].If one assumes the binuclear complex is dominant at high concentrations of [Cu(PPh3)4] then the intermediate spectrum, at which the band associated with the binuclear complex is half as intense as in the .nal spectrum, is measuring a system in which the concentration of [Ru(bpy)2(BL)]2 (BL = bridging ligand) is equivalent to that of [Ru(bpy)2(BL) Cu(PPh3)2]3. Assuming the concentration of [Cu(PPh3)4] is much greater than those of the other species present then K = 1/[Cu(PPh3)4].The values of K for the binuclear complexes based on the aforementioned assumptions are: K(4) = 1300, K(5) = 2000 and K(6) = 2500 dm3 mol1. The values are approximate, incorporating an error of 20%. The dissociation of the binuclear complexes at low concentration meant that all physical measurements were made on solutions of 1 × 103 mol dm3. UV/Visible absorption data for 1 × 103 mol dm3 solutions for samples 4–6 are shown in Table 1.The structures of the complexes 4 and 6 are shown in Figs. 2 and 3, respectively. They are generally similar in that both contain the {Ru(bpy)2}2 and {Cu(PPh3)2} fragments coordinated at each of the binding sites on the ligands. While these binding sites are e.ectively equivalent, the extra aromatic ring and methyl groups on dpqMe2 result in signi.cant di.erences in the structures of complexes 4 and 6. For example, the copper and ruthenium atoms are held at similar distances in each compound (6.82 Å distant in 4 and 6.88 Å distant in 6).However little else is similar between the two complexes. This is most simply shown in Fig. 4 where the metals and bpy and PPh3 ligands have been removed for clarity. The two related ligands appear to react to co-ordination in quite di.erent fashions. The dpp ligand splays the two pyridyl groups apart, while dpqMe2 both splays and twists the pyridyl groups. Surprisingly, the rest of the dpqMe2 ligand appears una.ected by the distortions of the pyridyl groups.The consequence of these distortions is a major di.erence in the relative orientations of the {Cu(PPh3)2} and {Ru(bpy)2}2 fragments in 4 and 6. In Figs. 2 and 3 the {Cu(PPh3)2} portions are drawn from the same relative orientations to highlight the di.erences in orientation of the {Ru(bpy)2}2 fragment. The reason for the di.erences in the distortions between 4 and 6 become apparent when CPK (Corey–Pauling–Koltun) models are examined. The space .lling drawings shown in Fig. 5 indicate that the methyl groups on the dpqMe2 are not causing the distortion because they are too distant from the co-ordination sites. However, the aromatic protons on the extra aromatic ring do cause signi.cant steric Table 1 Electronic absorption data for complexes in acetonitrile at 298 K Complex ./nm (e × 103/dm3 mol1 cm1) 456 387 (19.9) 394 (25.9) 436 (12.7) 421sh (12) 428sh (14) 481 (12.8) 509sh (11) 576 (14) 563 (15.3) sh = Shoulder.J.Chem. Soc., Dalton Trans., 1999, 2669–2673 2671 problems for the complexation on 6. No such steric problems are apparent for 4. The aromatic protons on 6 (dark) protrude directly in the positions that the {Ru(bpy)2}2 occupies in 4. To minimise unfavourable steric interactions the pyridyl groups on the dpqMe2 ligand and both metal fragments on 6 distort from what could be considered the optimum geometry displayed by 4. However, it is apparent that as the distortions in the dpqMe2 ligand are restricted to the pyridyl groups changes in functionality on the body of the ligand should not a.ect the structure Fig. 2 An ORTEP16 drawing of the cation of complex 4. All but the ipso-carbons on the phenyl rings of the PPh3 groups have been removed for clarity. Thermal ellipsoids are drawn at the 40% level. Fig. 3 An ORTEP drawing of the cation of complex 6. Details as in Fig. 2. signi.cantly, i.e. replacing methyl groups by hydrogens for example.This is relevant as there are no reported structures of bimetallic complexes involving dpqMe2 although at least one structurally characterised complex is known.17 In contrast a greater number of structurally characterised complexes of the dpq ligand 13,17–20 (where methyl groups are replaced by hydrogens) are known as well as the structure of the “free” ligand.19 The only bimetallic complex involving dpq as a ligand is [(Cu(PPh3)2)2(dpq)]2,18 which shows none of the signi.cant distortions observed in 6.However, the structure of the [Ru- (dpq)(bpy)2]2 complex has been described 20 and although the distortions we note in 6 are not reported by the authors they are apparent in the structure. Clearly, the {Ru(bpy)2}2 fragment is sterically demanding and its unique steric requirements are the dominant feature in the di.erences between the structures of 4 and 6. Fig. 4 Side on views of the dpp (top) and dpqMe2 (bottom) ligands in complexes 4 and 6 respectively. The metal complexes have been removed for clarity.Fig. 5 Space .lling drawings of complexes 4 (top) and 6 (bottom), with the aromatic protons of interest highlighted.2672 J. Chem. Soc., Dalton Trans., 1999, 2669¡V2673The electrochemical data for the ruthenium¡Vcopper complexesand their mononuclear counterparts are given in Table 2.The ruthenium() copper() binuclear complexes show anirreversible oxidation at 0.8 to 0.9 V vs. SCE. These are assignedas the oxidation CuI/II couple.They lie at similar Eo values forother polypyridyl complexes with {Cu(PPh3)2} moieties.21 Inthe binuclear complexes, 4¡V6, the rst reduction is at lessnegative potentials than for the corresponding mononuclearruthenium complexes. This reduction is assigned asBL ¡÷ BL; consistent with this the ease of reduction followsthe order dpq > dpqMe2 > dpp. The reduction is madeeasier by ca. 0.1 V on binding of the {Cu(PPh3)2} unit. Thisis considerably less than the stabilisation aorded on goingfrom mono- to bi-nuclear complexes with {Re(CO)3Cl} or{Ru(bpy)2}2 units.These stabilise by ca. 0.6 and 0.4 V respectivelyfor the ligands used herein.19,22 The {Cu(PPh3)2} is lesseective at stabilising the BL reduction because of its lowcharge and poor £k-acid character.23 The second reductionsfor 4¡V6 lie at ca. 1.4 V vs. SCE. These are at the Eo valuesassociated with bpy reduction.22,24The UV/Visible spectra of complexes 4¡V6 are consistent withthe electrochemistry ndings (Table 1).The strong visibleabsorptions observed for each of the complexes are assigned asMLCT Ru (d£k)¡÷BL (£k*). The energies of these transitionscorrelate with the ease of reduction of the bridging ligand.The shifts on going from mono- to bi-nuclear complexesfor ruthenium() to ruthenium copper are less (ca. 2000 cm1)than those observed for ruthenium to a diruthenium complex(typically 3000 cm1).24Resonance Raman spectra of complex 6 at a series of excitationwavelengths reveal something of the nature of thetransitions in the region 450 to 632 nm (Fig. 6, Table 3).Theresonance Raman spectrum at 457.9 nm is dominated by modesof 2,2-bipyridine (bpy). These lie at 1599, 1564, 1490 and 1316cm1.25 They indicate that the dominant transition at this wavelengthis Ru(d£k)¡÷bpy(£k*) charge transfer (CT) in nature. At488 nm the intensity of the bpy modes is reduced with dpqMe2modes at 1468 and 1558 cm1 increased in intensity.This trendcontinues at 514.5 and 632.8 nm. At 632.8 nm no bpy modes areobserved; the spectrum shows features that are dpqMe2 based.The Ru(d£k)¡÷dpqMe2(£k*) and Cu(d£k)¡÷dpqMe2(£k*) CT transitionscan give rise to the enhancement of dpqMe2 modes. Thelow wavenumber region of the resonance Raman spectrumshows very weak features. The most prominent of these is at338 cm1 and is assigned as a Ru¡VN stretch.26 No bands areobserved that may be assigned to the Cu¡VP stretch.Metal¡Vphosphorus stretches for rst row transition metals bonded toPPh3 typically lie at less than 200 cm1.27 The fact that the Ru¡VN band is enhanced suggests that the dominant transition in thevisible region is Ru(d£k)¡÷dpqMe2(£k*) CT. The complex appearsstable in solution as there is no observable change in its electronicspectra over a period of hours. The resonance Ramanspectrum of 6 in solution shows no features that could beassigned to the decomposition product 3. The band positionsfor 4 and 5 are also presented in Table 3.These complexesTable 2 Electrochemical data for complexes in acetonitrile at 298 K aE/VComplex Oxidation Reduction1234561.0 (i)0.9 (i)0.8 (i)1.381.401.371.471.491.451.060.800.850.910.600.761.48 (i)1.48 (i)1.49 (i)1.461.381.40a Potentials versus SCE ¡Ó 0.02 V. Supporting electrolyte 0.1 mol dm3NBu4ClO4. (i) = Irreversible process.behave in a very similar manner to that of 6, with bpy modesenhanced at shorter wavelengths while the BL modes becomeenhanced to lower energy.No features of the monomers can beobserved suggesting no detectable decomposition has occurredat the 1 mmol dm3 concentrations used in these studies.Electronic spectroelectrochemistry reveals that upon oxidationand reduction the binuclear complexes decompose to themononuclear ruthenium() complexes within the timescale ofthe electrochemical experiment. This is unsurprising given theelectrochemical data with the rst oxidation being irreversibleand the inherent instability of complexes with {Cu(PPh3)2} toelectrochemical oxidation and reduction processes.21,23 Weare currently investigating methods of increasing the electrochemicalstability of such complexes.The excited states of the complexes are weakly luminescent inCH2Cl2. Attempts to generate time-resolved resonance Ramanspectra of the MLCT excited state of 4¡V6 by variable powersingle-colour experiments reveal that at low pulse energy aground state spectrum is generated and as the power isincreased several ground state features are reduced in intensityrelative to solvent. However, no features are observed to growin with increased photon ux at the excitation wavelength of532 or 630.7 nm.The fact that only ground state features areobserved, which appear barely aected by increased laserpower, suggests that the excited state lifetime of 6 in CH2Cl2 isless than 5 ns.28ConclusionCrystal structures are presented for complexes 4 and 6.Dierencesin orientations of the {Ru(bpy)2}2 fragment areobserved, caused by the presence of the aromatic protons onthe extra aromatic ring as these protons on 6 protrude directlyin the positions that the {Ru(bpy)2}2 occupy in 4.The reduction of the BL occurs at a slightly more positivepotential than is found in the monometallic ruthenium systemswhile a small red shift in the Ru¡÷BL MLCT transition is foundupon complexation of the {Cu(PPh3)2} moiety. These smallshifts are caused by the stabilisation of the ligand £k* orbital bythe substitution of the second metal centre.The changesobserved are not as large as has been found for other bimetallicsystems as the d10 copper system is not as ecient at stabilisingthe ligand £k* orbital. Dilution of these complexes in CH2Cl2below 5 ¡Ñ 104 mol dm3 resulted in signicant decomposition.Fig. 6 Resonance Raman spectra of complex 6 in dichloromethane(1 ¡Ñ 103 mol dm3): (a) excitation wavelength = 457.9 nm, 30 mW; (b)excitation wavelength = 488 nm, 30 mW; (c) excitation wavelength= 514.5 nm, 30 mW; (d) excitation wavelength = 632.8 nm, 10mW.* denotes solvent bands.J. Chem. Soc., Dalton Trans., 1999, 2669.2673 2673 Table 3 Observed Raman bands (cm1) for complexes 4 5 6 ¥ëexc/nm 457.9 514.5 457.9 514.5 457.9 514.5 632.8 1554s c 1508s c 1489s a 1470wc 1452wc 1402wc 1315wa 1267wa 1246wc 1597wa 1565wa 1554s c 1508s c 1489wa 1470s c 1452wc 1402s c 1315wa 1301w 1267s a 1246s c 1587wb 1563wa 1491s a 1469s b 1315wa 1598wa 1587wb 1563wa 1491s a 1469s b 1395wb 1362s b 1314s a 1281s b 1599wa 1564wa 1558wd 1515wd 1490s a 1468s d 1409wd 1363wd 1328wd 1316wa 1599wa 1564wa 1558wd 1490wa 1468s d 1409wd 1363wd 1324s d 1288wd 1558s d 1468wd 1363wd 1319s d 1288s d s = Strong, w = weak. a bpy mode.b dpq mode. c dpp mode. d dpqMe2 mode. Acknowledgements Support from the New Zealand Lottery Commission and the University of Otago Research Committee for the purchase of the Raman spectrometer, and from the University of Otago Division of Sciences for chemicals, is gratefully acknowledged.We also thank the University of Otago for the award of a Ph.D. scholarship to S. M. S., and the Massey University Research Fund for funding toward low-temperature crystallographic data collection. References 1 F. Scandola, M. T. Indelli, C. Chiorboli and C. A. Bignozzi, Top. Curr. Chem., 1990, 158, 73; V. Balzani, S. Campagna, G. Denti, A. Juris, S.Serroni and M. Venturi, Acc. Chem. Res., 1998, 31, 26. 2 M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphrey-Baker, E. Mueller, P. Liska, N. Vlachopoulos and M. Gratzel, J. Am. Chem. Soc., 1993, 115, 6382; K. Kalyanasundaram and M. Gratzel, Coord. Chem. Rev., 1998, 177, 347; J. E. Moser, P. Bonnote and M. Gratzel, Coord. Chem. Rev., 1998, 177, 245; C. A. Bignozzi, J. R. Schoonover and F. Scandola, Prog. Inorg. Chem., 1997, 44, 1. 3 V. Balzani, M. Gomez-Lopez and J. F. Stoddart, Acc.Chem. Res., 1998, 31, 405; M. Venturi, S. Serroni, A. Juris, S. Campagna and V. Balzani, Top. Curr. Chem., 1998, 197, 193; R. Ziessel, M. Hissler, A. El-Ghayoury and A. Harriman, Coord. Chem. Rev., 1998, 180, 1251; R. Ziessel and A. Harriman, Coord. Chem. Rev., 1998, 171, 331; F. Scandola, R. Argazzi, C. A. Bignozzi, C. Chiorboli, M. T. Indelli and M. A. Rampi, in Supramolecular Chemistry, eds. V. Balzani and L. DeCola, Kluwer, Dordrecht, 1992, p. 235. 4 K. Kalyanasundaram and M.K. Nazeeruddin, Inorg. Chim. Acta, 1994, 226, 213; V. Balzani, A. Credi and M. Venturi, Coord. Chem. Rev., 1998, 171, 3; P. Belser, S. Bernhard, E. Jandrasics, A. von Zelewsky, L. DeCola and V. Balzani, Coord. Chem. Rev., 1997, 157, 1. 5 L. De Cola and P. Belser, Coord. Chem. Rev., 1998, 177, 301; M. D. Ward, C. M. White, F. Barigelletti, N. Armaroli, G. Calogero and L. Flamigni, Coord. Chem. Rev., 1998, 171, 481; M. M. Richter and K. J. Brewer, Inorg. Chem., 1993, 32, 5762; F.Vogtle, M. Frank, M. Nieger, P. Belser, A. von Zelewsky, V. Balzani, F. Barigelletti, L. DeCola and L. Flamigni, Angew. Chem., Int. Ed. Engl., 1993, 32, 1643; V. Grosshenny, A. Harriman and R. Ziessel, Angew. Chem., Int. Ed. Engl., 1995, 34, 1100. 6 J. R. Schoonover, K. C. Gordon, R. Argazzi, W. H. Woodru., K. A. Peterson, C. A. Bignozzi, R. B. Dyer and T. J. Meyer, J. Am. Chem. Soc., 1993, 115, 10996. 7 Y. Lei, T. Buranda and J. F. Endicott, J. Am. Chem. Soc., 1990, 112, 8820; M.T. Indelli, F. Scandola, J.-P. Collin, J.-P. Sauvage and A. Sour, Inorg. Chem., 1996, 35, 303. 8 M. Milkevitch, E. Brauns and K. J. Brewer, Inorg. Chem., 1996, 35, 1737 and refs. therein. 9 D. R. Crane, J. DiBenedetto, C. E. A. Palmer, D. R. McMillin and P. C. Ford, Inorg. Chem., 1988, 27, 3698. 10 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and A. von Zelewsky, Coord. Chem. Rev., 1988, 84, 85; K. Kalyanasundaram, Photochemistry of Polypyridine and Porphyrin Complexes, Academic Press, London, 1992. 11 S. M. Scott, A. K. Burrell, P. A. Cocks and K. C. Gordon, J. Chem. Soc., Dalton Trans., 1998, 3679. 12 H. A. Goodwin and F. Lions, J. Am. Chem. Soc., 1959, 81, 6415; C. H. Braunstein, A. D. Baker, T. C. Strekas and H. D. Gafney, Inorg. Chem., 1984, 23, 858. 13 (a) T. J. Simpson and K. C. Gordon, Inorg. Chem., 1995, 34, 6323; (b) M. R. Waterland, T. J. Simpson, K. C. Gordon and A. K. Burrell, J. Chem. Soc., Dalton Trans., 1998, 185. 14 SDP, Structure Determination Package, Enraf-Nonius, Delft, 1985. 15 G. M. Sheldrick, SHELXL 95, Institut fur Anorganische Chemie der Universitat Gottingen, 1993. 16 C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. 17 A. Escuer, R. Vicente, T. Comas, J. Ribas, M. Gomez, X. Solans, D. Gatteschi and C. Zanchini, Inorg. Chim. Acta, 1991, 181, 51. 18 K. C. Gordon, A. H. R. Al-Obaidi, P. M. Jayaweera, J. J. McGarvey, J. F. Malone and S. E. J. Bell, J. Chem. Soc., Dalton Trans., 1996, 1519. 19 S. C. Rasmussen, M. M. Richter, E. Yi, H. Place and K. J. Brewer, Inorg. Chem., 1990, 29, 3926. 20 D. P. Rillema, D. G. Taghdiri, D. S. Jones, C. D. Keller, L. A. Worl, T. J. Meyer and H. A. Levy, Inorg. Chem., 1987, 26, 578. 21 S. M. Scott, K. C. Gordon and A. K. Burrell, Inorg. Chem., 1996, 35, 2452. 22 K. J. Brewer, W. R. Murphy, S. R. Spurlin and J. D. Petersen, Inorg. Chem., 1986, 25, 882; M. M. Richter and K. J. Brewer, Inorg. Chim. Acta, 1991, 180, 125; R. Lin and T. F. Guarr, Inorg. Chim. Acta, 1994, 226, 79; J. Sherborne, S. M. Scott and K. C. Gordon, Inorg. Chim. Acta, 1997, 260, 199. 23 M. R. Waterland, K. C. Gordon, J. J. McGarvey and P. M. Jayaweera, J. Chem. Soc., Dalton Trans., 1998, 609. 24 S. M. Molnar, K. R. Neville, G. E. Jensen and K. J. Brewer, Inorg. Chim. Acta, 1993, 206, 69. 25 P. G. Bradley, N. Kress, B. A. Hornberger, R. F. Dallinger and W. H. Woodru., J. Am. Chem. Soc., 1981, 103, 7441. 26 D. P. Strommen, P. K. Mallick, G. D. Danzer, R. S. Lumpkin and J. R. Kincaid, J. Phys. Chem., 1990, 94, 1357; O. Poizat and C. Sourisseau, J. Phys. Chem., 1984, 88, 3007. 27 J. Bradbury, K. P. Forest, R. H. Nuttal and D. W. A. Sharp, Spectrochim. Acta, Part A, 1967, 23, 2704. 28 D. S. Caswell and T. G. Spiro, Inorg. Chem., 1987, 26, 18; J.-H. Perng and J. I. Zink, Inorg. Chem., 1988, 27, 1403; K. C. Gordon and J. J. McGarvey, Chem. Phys. Lett., 1989, 162, 117. Paper 9/02934F

ISSN:1477-9226    

DOI:10.1039/a902934f

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 2671 Insertion of CS2 into the Group 2 metal–alkoxide bonds of [{M(OR)2}n] (M 5 Mg, Ca, Sr or Ba; R 5 Et or Pri); crystal structures of [Ca(S2COPri)2(PriOH)3]?2PriOH and [{Ba(S2COEt)2}•] Izoldi K. Bezougli,a Alan Bashall,b Mary McPartlin b and D. Michael P. Mingos a a Department of Chemistry, Imperial College of Science Technology and Medicine, South Kensington, London, UK SW7 2AY b School of Applied Chemistry, University of North London, Holloway Road, London, UK N7 8DB The insertion of carbon disulfide into the M]O bonds of the alkaline-earth-metal ethoxides and isopropoxides has been studied.The resulting products have been characterised by IR, 1H and 13C NMR spectroscopy, mass spectrometry and thermogravimetric analysis. Two of these complexes have been structurally characterised by single crystal X-ray crystallography and the data have confirmed the insertion of CS2 into both Ba]O bonds of [{Ba(OEt)2(EtOH)4}n] giving the polymer [{Ba(S2COEt)2}•] and into both Ca]O bonds of [{Ca(OEt)2(EtOH)4}n] giving the monomer [Ca(S2COPri)2(PriOH)3]?2PriOH.Carbon disulfide has proved to be a very versatile ligand which forms complexes with almost every transition metal.1 A large number of carbon disulfide metal complexes have been prepared and several review articles have been devoted to their co-ordination chemistry.1–6 Carbon disulfide has the potential to insert into a variety of M]X bonds (X = H, C, N, P, S, Cl or O),2,7 more specifically its insertion into M]O bonds of alkoxides leads to the formation of either metal O-alkyl thiocarbonates or dithiocarbonates. The chemical literature has very many structurally characterised alkyl dithiocarbonato complexes of the transition metals, but very little is known about the analogous complexes of the alkaline-earth metals.In this paper we report the insertion reactions of CS2 into the M]O bonds of the Group 2 metal ethoxides and isopropoxides.Experimental General procedures All manipulations were carried out under an atmosphere of dry nitrogen using standard glove-box (Miller-Howe FF160) and Schlenk techniques. All solvents were rigorously dried and deoxygenated by standard procedures. The samples for NMR, infrared studies and thermogravimetric analysis were handled in a glove-box, but those for microanalysis were not. This has led to some diVerences concerning the extent to which alcohol molecules of crystallization were detected by the spectroscopic and analytical techniques.Instrumentation Infrared spectra were recorded on a Perkin-Elmer FTIR 1720 spectrometer using Nujol mulls between 25 × 4 mm CsI plates. The Nujol was dried with 4 Å molecular sieves prior to use (and stored in a glove-box); the samples were protected from the atmosphere by an O-ring-sealed Presslok holder (Aldrich Chemicals). The NMR spectra were recorded on a JEOL GS 270 MHz spectrometer; 1H NMR was referenced internally to the residual 1H impurity present in the deuteriated solvent.Chemical shifts are recorded in parts per million (d) relative to SiMe4 (d 0) using (CD3)2SO (d 2.52), 13C NMR to (CD3)2SO (d 40.6). Controlled thermal analyses of the complexes were investigated using a Polymer Laboratories 1500H simultaneous thermal analyser, controlled by an OmniPro 486DX-33 personal computer. The mass of the samples was between 10 and 25 mg.The measurements were carried out in alumina crucibles under an atmosphere of flowing (25 cm3 min21) nitrogen gas, using heating rates of 5 8C min21. Starting materials Dibutylmagnesium, calcium, strontium and barium granules from Aldrich Chemicals Co. were used as received. Preparations [{Mg(S2COEt)2(EtOH)x}n] 1. Dibutylmagnesium in heptane (20 cm3, 1 M, 20 mmol) was added to ethanol (30 cm3) at 240 8C resulting in an exothermic reaction. The reaction mixture was slowly warmed to room temperature where the volume of the solvent was reduced until all the heptane had been removed and precipitation of a white solid [{Mg(OEt)2(EtOH)x}n] was observed.Addition of ethanol (30 cm3) resulted in a suspension of the magnesium ethoxide. Carbon disulfide (3.10 cm3, 50 mmol) was added and was stirred for 12 h at room temperature. This resulted in dissolution of the ethoxide to yield a yellow solution, which was filtered to remove any insoluble species.A crystalline solid was isolated after cooling the solution to 220 8C (yield: 3.87 g, 66.7%) (Found: C, 31.3; H, 6.7. Calc. for C10H24MgO5S4: C, 31.9; H, 6.4%) [analysis based on Mg(S2COEt)2(EtOH)2?H2O]. IR (cm21) (Nujol): 3362s, 1261w, 1184s, 1120m, 1042s, 1009m, 883s, 802m, 722w, 561s and 491w. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.38 (t, CH3CH2OH, 3 H), 4.24 (q, S2COCH2CH3, 4 H), 3.46 (m, CH3CH2OH, 6 H), 1.19 (t, S2COCH2CH3, 6 H) and 1.07 (t, CH3CH2OH, 9 H); 13C-{1H} (67.94 MHz), d 230.49 (S2COR), 66.63 (S2COCH2CH3), 56.68 (CH3CH2OH), 19.21 (CH3CH2OH) and 15.17 (S2COCH2CH3).Mass spectrum: (positive-ion FAB), m/z 176, [Mg(S2C)2]; (negative-ion FAB), m/z 232, [Mg2(S2CO)2] and 289, [Mg2(S2COEt)2]2. [{Ca(S2COEt)2(EtOH)x}n] 2. Calcium metal (1.32 g, 32.95 mmol) was suspended in ethanol (50 cm3) and the mixture heated under reflux for 4 h resulting in dissolution of the metal and evolution of hydrogen gas. Carbon disulfide (5.20 cm3, 82.402672 J.Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 mmol) was added to the resulting suspension of [{Ca(OEt)2- (EtOH)x}n] in ethanol and the solution stirred for 8 h at room temperature. This resulted in dissolution of the ethoxide to yield a yellow solution which was filtered. A crystalline solid was isolated after cooling the solution to 220 8C (yield: 4.71 g, 50.7%) (Found: C, 31.2; H, 5.8. Calc. for C18H38Ca2O7S8: C, 30.8; H, 5.5%) [analysis based on Ca2(S2COEt)4(EtOH)3]. IR (cm21) (Nujol): 3229m, 1268w, 1177s, 1119s, 1079m, 1039s, 1011s, 875s, 803m, 771w, 723m, 674w, 655m, 446m, 372w and 320m.NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.38 (s, CH3CH2OH, 3 H), 4.23 (q, S2COCH2CH3, 4 H), 3.45 (q, CH3CH2OH, 6 H), 1.18 (t, S2COCH2CH3, 6 H) and 1.06 (t, CH3CH2OH, 9 H); 13C-{1H} (67.94 MHz), d 230.48 (S2COEt), 66.67 (S2COCH2CH3), 56.70 (CH3CH2OH), 19.23 (CH3CH2- OH) and 15.19 (S2COCH2CH3). Mass spectrum: (positive-ion FAB), m/z 161, [Ca(S2COEt)]; 206, [Ca(CS)(S2COEt)]1; 238, [Ca(CS2)(S2COEt)]1; 329, [Ca(S2COEt)(EtOH)]1; 467, [Ca(S2- COEt)2(EtOH)4]1; 499, [CaS(S2COEt)2(EtOH)4]; 551, [Ca2(CS)- (S2COEt)2(EtOH)4]1; 583, [Ca2(CS2)(S2COEt)2(EtOH)4]1 and 565, [Ca2(S2COEt)4]1.[{Sr(S2COEt)2(EtOH)x}n] 3. A similar experimental procedure was adopted to that described for complex 2, starting from strontium metal (1.61 g, 18.37 mmol) in ethanol (60 cm3) and adding carbon disulfide (2.86 cm3, 46.0 mmol). In this case the reaction was exothermic, yielding a clear yellow solution within 5 min of addition of CS2 to the strontium ethoxide solution (yield: 4.77 g, 78.8%) (Found: C, 24.4; H, 4.2.Calc. for C8H16O3S4Sr: C, 25.6; H, 4.3%) [analysis based on Sr(S2COEt)2- (EtOH)]. IR (cm21) (Nujol): 3355w (br), 1290w, 1176s, 1066s, 1037s, 869m, 802m, 738w and 723w. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.41 (s, CH3CH2OH, 1 H), 4.26 (q, S2COCH2CH3, 4 H), 3.48 (q, CH3CH2OH, 2 H), 1.22 (t, S2COCH2CH3, 6 H) and 1.10 (t, CH3CH2OH, 3 H); 13C-{1H} (67.94 MHz), d 231.28 (S2COR), 67.38 (S2COCH2CH3), 57.45 (CH3CH2OH), 19.98 (CH3CH2OH) and 15.94 (S2COCH2CH3).Mass spectrum: (positive-ion FAB), m/z 165, [Sr(CS2)]1; 240, [Sr(CS2)2]; 394, [Sr(S2COEt)2S2]; 615, [Sr2(S2COEt)3(CS2)] and 631, [Sr2(S2COEt)3(S2CO)]; (negative-ion FAB), m/z 375, [Sr- (S2COEt)2(EtOH)]2; 407, [Sr(S2COEt)2S(EtOH)]2; 438, [Sr- (S2COEt)2S2(EtOH)]2; 451, [Sr(S2COEt)2(CS2)(EtOH)]2; 467, [Sr(S2COEt)2(S2CO)(EtOH)]2 and 483, [Sr(S2COR)2(S2CO)- O(EtOH)].[{Ba(S2COEt)2}•] 4. A similar synthetic method to that described for complex 2 was used here, starting from barium metal (1.47 g, 10.70 mmol) in ethanol (60 cm3) and adding carbon disulfide (1.66 cm3, 26.76 mmol) (yield: 2.81 g, 69.2%) (Found: C, 18.9; H, 2.6. Calc. for C6H10BaO2S4: C, 20.0; H, 2.7%) [analysis based on Ba(S2COEt)2]. IR (cm21) (Nujol): 3328w, 3185w, 1297w, 1261w, 1168s, 1143m, 1099s, 1083s, 1060s, 1043s, 1016m, 879w, 804w, 723s, 588w, 487w, 449w and 318w.NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.35 (t, ROH, 1.5 H), 4.21 (q, S2COCH2CH3, 4 H), 3.42 (dq, CH3CH2OH, 3 H), 1.16 (t, S2COCH2CH3, 6 H) and 1.03 (t, CH3CH2OH, 4.5 H); 13C-{1H} (67.94 MHz), d 230.39 (S2COR), 66.59 (S2COCH2CH3), 56.58 (CH3CH2OH), 19.12 (CH3CH2OH) and 15.07 (S2COCH2CH3). Mass spectrum: (positive-ion FAB), m/z 259, [Ba(S2COEt)]1; 518, [Ba2(S2COEt)2]1; 638, [Ba2(S2COEt)3]1 and 760, [Ba2(S2COEt)4]1; (negative-ion FAB), m/z 121, (S2COEt, 87%) and 322, [Ba(S2COEt)S2].[{Mg(S2COPri)2(PriOH)x}n] 5. Dibutylmagnesium in heptane (10 cm3, 1 M, 10 mmol) was added to PriOH (30 cm3) at 240 8C resulting in an exothermic reaction. The reaction mixture was slowly warmed to room temperature and the volume of the solvent reduced until all the heptane was removed and precipitation of white [{Mg(OPri)2(PriOH)4}n] was observed. Addition of PriOH (30 cm3) resulted in a suspension of the magnesium isopropoxide. Carbon disulfide (1.50 cm3, 25 mmol) was added and the solution stirred at 80 8C for 12 h.This aVorded a yellow precipitate which was isolated by filtration (yield: 1.81 g, 61.6%) (Found: C, 33.0; H, 6.1. Calc. for C8H14MgO2S4: C, 32.6; H, 4.8%) [analysis based on Mg(S2COPri)2]. IR (cm21) (Nujol): 1365s, 1348m, 1259w, 1162s, 1149s, 1042s, 987s, 831m, 794w, 580s, 520m and 433m. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), 4.29 [s, (CH3)2CHOH], 3.75 [m, S2COCH(CH3)2/(CH3)2- CHOH], 1.04 [d, S2COCH(CH3)2] and 1.02 [d, (CH3)2CHOH]; 13C-{1H} (67.94 MHz), d 230.11 [S2COCH(CH3)2], 62.83 [S2COCH(CH3)2/(CH3)2CHOH] and 26.47 [(CH3)2CHOH/ S2COCH(CH3)2].Mass spectrum: (positive-ion FAB), m/z 355, [Mg(S2COPri)2(PriOH)]1 and 414, [Mg(S2COPri)2(Pri- OH)2]; (negative-ion FAB), m/z 100, [Mg(S2C)]; 175, [Mg- (S2C)2, 91%]2; 192, [Mg(S2C)2(O)]; 296, [Mg(S2C)2(PriOH)2]; 414, [Mg(S2COPri)2(PriOH)2] and 474, [Mg(S2COPri)2(Pri- OH)3]. [Ca(S2COPri)2(PriOH)3]?2PriOH 6. Calcium metal (0.64 g, 15.98 mmol) was suspended in PriOH (50 cm3) and heated under reflux for 36 h resulting in dissolution of the metal and evolution of hydrogen gas.Carbon disulfide (2.40 cm3, 39.94 mmol) was added to the resulting suspension of [{Ca(OPri)2- (PriOH)x}n] in PriOH. The solution was stirred for 12 h at room temperature after which time dissolution of the alkoxide had occurred producing a clear yellow solution. The solution was filtered to remove any insoluble species, its volume was reduced and the product isolated as big cubic crystals at 220 8C (yield: 3.34 g, 67.5%) (Found: C, 41.5; H, 7.7.Calc. for C17H38CaO5S4: C, 41.6; H, 7.8%) [analysis based on Ca(S2COPri)2(PriOH)3]. IR (cm21) (Nujol): 3343s, 1301m, 1263w, 1189s, 1159s, 1137s, 1091s, 1037s, 948s, 906w, 815m, 723m and 659w. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 5.47 [m, S2COCH(CH3)2, 2 H], 4.35 [d, (CH3)2CHOH, 5 H], 3.77 [m, (CH3)2CHOH, 5 H], 1.17 [d, S2COCH(CH3)2, 12 H] and 1.05 [d, (CH3)2CHOH, 30 H]; 13C-{1H} (67.94 MHz), d 229.94 [S2COCH(CH3)2], 72.71 [S2COCH(CH3)2], 62.73 [(CH3)2CHOH], 26.00 [(CH3)2- CHOH] and 22.40 [S2COCH(CH3)2].Solid state 13C NMR: d 231.30, 230.13 [S2COCH(CH3)2], 77.86, 77.50 [S2COCH- (CH3)2], 67.70 [(CH3)2CHOH], 26.69, 25.84 [(CH3)2CHOH], 23.11, 22.82, 22.27 [S2COCH(CH3)2]. Mass spectrum: (positiveion FAB), m/z 57, [CaO]; 192, [Ca(S2CPri)O]1; 252, [Ca(S2- COPri)O(PriOH)]1; 328, [CaO(S2COPri)2(PriOH)]1 and 345, [CaO2(S2COPri)2(PriOH)]1; (negative-ion FAB), m/z 135, [S2COPri].[{Sr(S2COPri)2(PriOH)2}n] 7. Strontium metal (0.99 g, 11.30 mmol) was suspended in PriOH (40 cm3) and the mixture stirred at room temperature for 12 h resulting in dissolution of the metal and evolution of hydrogen gas, yielding a transparent slightly purple solution. The solution was filtered and into the clear solution carbon disulfide (1.7 cm3, 28.2 mmol) added. The solution turned immediately yellow and was stirred for 1 h at room temperature.The product was isolated after reducing the volume of the solution and cooling to 220 8C (yield: 3.29 g, 81.4%) (Found: C, 35.9; H, 5.0. Calc. for C14H30O4S4Sr: C, 35.2; H, 6.3%) [analysis based on Sr(S2COPri)2(PriOH)2]. IR (cm21) (Nujol): 3360m, 1615m, 1261m, 1190s, 1183s, 1173s, 1092s, 1069s, 1037s, 902m, 807w, 722m and 662w. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 5.46 [m, S2COCH(CH3)2, 2 H], 4.36 [d, (CH3)2CHOH, 3 H], 3.78 [m, (CH3)2CHOH, 3 H], 1.16 [d, S2COCH(CH3)2, 12 H] and 1.04 [d, (CH3)2CHOH, 18 H]; 13C-{1H} (67.94 MHz), d 229.94 [S2COCH(CH3)2], 72.73 [S2COCH(CH3)2], 62.70 [(CH3)2CHOH], 26.01 [(CH3)2CHOH] and 22.53 [S2COCH(CH3)2].Mass spectrum: (positive-ion FAB), m/z 240, [Sr(S2CPri)O]1; 300, [Sr(S2COPri)O(PriOH)]1; 447, [Sr2(S2COPri)2]1 and 581, [Sr2(S2COPri)3]1; (negativeion FAB), m/z 135, [S2COPri]; 234, [Sr(S2COPri)C]2; 370, [Sr(S2COPri)2C]2; 592, [Sr2(S2COPri)3C]2 and 728, [Sr2(S2COPri) 4C]2.J. Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 2673 Table 1 The IR data (cm21) for the dithiocarbonate group and 13C NMR data in (CD3)2SO for compounds 1–8 d (13C) Complex 123456 78 [{Mg(S2COEt)2(EtOH)x}n] [{Ca(S2COEt)2(EtOH)x}n] [{Sr(S2COEt)2(EtOH)x}n] [{Ba(S2COEt)2}•] [{Mg(S2COPri)2(PriOH)x}n] [{Ca(S2COPri)2(PriOH)3]? 2PriOH [{Sr(S2COPri)2(PriOH)2}n] [{Ba(S2COPri)2(PriOH)2}n] n(Ca]O) 1184, 1120 1177, 1119 1176 1168, 1143 1162, 1149 1189, 1159, 1137 1190, 1183, 1173 1187, 1170, 1159 n(CS2) 1042, 883 1039, 875 1037, 869 1043 1042 1037, 815 1037 1049 n(Cb]O) 1009 1011 1000 1016 987 948 1069 1039 CS2 230.49 230.48 231.28 230.39 230.11 229.94 229.94 230.07 CH2/CH of S2COR 66.63 66.67 67.38 66.59 62.83 72.71 72.73 72.87 CH2/CH of ROH 56.68 56.70 57.45 56.58 62.83 62.73 62.70 62.75 CH3 of ROH 19.21 19.23 19.98 19.12 26.47 26.00 26.01 26.32 CH3 of CS2OR 15.17 15.19 15.94 15.07 26.47 22.40 22.53 22.72 [{Ba(S2COPri)2(PriOH)2}n] 8.Barium metal (1.34 g, 9.76 mmol) was suspended in PriOH (60 cm3) and the mixture stirred at room temperature for 12 h which resulted in dissolution of the metal and evolution of hydrogen gas, yielding a transparent slightly purple solution.The solution was filtered and into the clear solution carbon disulfide (1.47 cm3, 24.39 mmol) added. A yellow precipitate started forming in the solution soon after the addition of carbon disulfide. The solution was stirred for 1 h at room temperature and the product isolated as a microcrystalline yellow precipitate by filtration (yield: 2.65 g, 66.6%) (Found: C, 30.9; H, 3.1.Calc. for C14H30BaO4S4: C, 30.8; H, 5.9%) [analysis based on Ba(S2COPri)2(PriOH)2?H2O]. IR (cm21) (Nujol): 1307w, 1261w, 1187s, 1170m, 1159m, 1095s, 1049s, 1039s, 904w, 800w and 723m. NMR [(CD3)2SO, 20 8C]: 1H (270 MHz), d 4.46 [d, (CH3)2CHOH, 2 H], 3.86 [m, (CH3)2CHOH/ S2COCH(CH3)2, 4 H], 1.25 [d, S2COCH(CH3)2, 12 H] and 1.12 [d, (CH3)2CHOH, 12 H]; 13C-{1H} (67.94 MHz), d 230.07 [S2COCH(CH3)2], 72.87 [S2COCH(CH3)2], 62.75 [(CH3)2- CHOH], 26.32 [(CH3)2CHOH] and 22.72 [S2COCH(CH3)2].Solid state 13C NMR: d 230.59 [S2COCH(CH3)2], 81.38 [S2COCH( CH3)2], 76.29 [(CH3)2CHOH], 24.12 [(CH3)2CHOH], 22.75 and 21.57 [S2COCH(CH3)2]. Mass spectrum: (positiveion FAB), m/z 138, [Ba]; 155, [BaO]1; 273, [Ba(S2COPri)]1; 290, [Ba(S2COPri)O]1 and 681, [Ba2(S2COPri)3]1; (negative-ion FAB), m/z 135, [S2COPri]; 284, [Ba(S2COPri)C] and 420, [Ba(S2COPri)2C]. X-Ray crystallography Data were collected using a Siemens P4 diVractometer, equipped with a Siemens LT2 low temperature device with graphite monochromated radiation using w–2q scans at 173 K.No significant decay in the intensity of three standard reflections measured after every 100 was observed. The data were corrected for Lorentz-polarisation factors and for absorption (y scans). The crystal data, data collection and refinement details are summarized in Table 4. Both structures were solved by direct methods and in each case all non-hydrogen atoms were located from subsequent Fourier-diVerence syntheses.All non-hydrogen atoms were assigned anisotropic displacement parameters and refined using full-matrix least squares on Fo 2.8 The hydrogen atoms for each of the compounds were included at calculated positions with C]H bond distances of 1.00 and 0.98 Å for the tertiary CH and methyl groups, respectively. The hydroxyl hydrogens in complex 6 were located in a Fourier-diVerence synthesis, and included in structure factor calculation with thermal parameters at 0.09 Å2 but were not refined.During refinement all the hydrogens were allowed to ride on their parent atom and assigned isotropic thermal parameters equal to 1.2Ueq of the parent atom for the tertiary CH groups and 1.5Ueq for the methyl groups. CCDC reference number 186/1052. See http://www.rsc.org/suppdata/dt/1998/2671/ for crystallographic files in .cif format. Results and Discussion Synthesis The products resulting from the reactions of the Group 2 metal ethoxides and isopropoxides [{M(OR)2(ROH)x}n] (M = Mg, Ca, Sr or Ba; R = Et or Pri) with CS2 are listed in Table 1 together with selected IR and 13C NMR data.The metal alkoxides were synthesized by the methods established by previous research workers.9 They are not suYciently crystalline to have been structurally characterised, nonetheless NMR studies indicated that they are solvated and x in the general formula is either 3 or 4 depending on the nature of R.9 Liquid carbon disulfide (2 equivalents) was added to a suspension of the metal alkoxide in alcohol and the solution stirred until dissolution of the metal alkoxide was complete and a clear yellow solution had formed [equation (1)].The reactions were done under an [{M(OR)2(ROH)x}n] 1 2CS2 æÆ [{M(S2COR)2(ROH)x}n] (1) inert atmosphere and oxygen and moisture were excluded during and after the reactions. The rates of reaction of CS2 with the Group 2 metal alkoxides showed great variations depending on the specific alcohols and metals involved.The addition of CS2 to ethanol solutions of Mg(OEt)2 or Ca(OEt)2 resulted in the dissolution of the ethoxide after stirring for 12 or 8 h, respectively. In contrast, on addition of CS2 to ethanol solutions of Sr(OEt)2 or Ba(OEt)2 a very rapid exothermic reaction occurred, accompanied by dissolution of the alkoxides. In all cases clear yellow solutions were obtained which were filtered in order to remove any insoluble residues and compounds 1–4 were isolated as crystalline materials after cooling the solution to 220 8C.In the case of Mg(OPri)2 the alkoxide was suspended in PriOH and after addition of CS2 the reaction mixture was refluxed for 12 h. A yellow product was formed at the end of the reaction. During the preparation of the CS2 calcium isopropoxide adduct the solution was stirred for 12 h after which time almost all the alkoxide had dissolved.The solution was filtered and complex 6 was isolated as a crystalline yellowish material at 220 8C. In the case of Sr(OPri)2 and Ba(OPri)2 the alkoxides were dissolved in PriOH at room temperature and CS2 was added to the clear colourless solutions. These solutions turned yellow and after stirring for 1 h at room temperature 7 and 8 were isolated as yellow microcrystalline solids after cooling to 220 8C. Spectroscopic characterisation The IR spectra of complexes 1–8 were studied as Nujol mulls between CsI plates and the stretching vibrations of the O-alkyl dithiocarbonate groups are summarised in Table 1.The infrared data have been assigned on the basis of previously reported2674 J. Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 IR spectra of such complexes.10 The C]O]C linkage frequencies of the ligand were separated into two ranges according to the two diVerent carbon environments: [S2]Ca]O]Cb]R]2. The most important features of the 1H NMR spectra for complexes 1–8 are recorded in the Experimental section. The protons of the two diVerent CH2 environments of the ethyl dithiocarbonato and the ethanol ligands in 1–4 appeared as two distinct peaks at d ca. 4.25 and at ca. 3.46, respectively. Similarly, the protons of the two diVerent CH3 groups appeared as two distinct triplets at d ca. 1.19 for the CH3 of the dithiocarbonato moiety, and at d ca. 1.08 for the co-ordinated ethanol molecules. The OH proton was observed at higher chemical shifts, at d ca. 4.38, as a triplet for 1 and 4, and as a singlet for 2 and 3. Similar assignments were made for the isopropyl dithiocarbonato complexes 5–8. The diVerent CH environments of the isopropyl dithiocarbonato and co-ordinated PriOH molecules appeared at d ca. 5.47 and ca. 3.78, respectively, for 6 and 7 while the two sets of CH3 groups were observed at d ca. 1.17 and ca. 1.05 as distinct triplets. In contrast, in the 1H NMR spectra of 5 and 8 the peaks associated with the two diVerent CH environments of the isopropyl dithiocarbonato and the bound PriOH molecules appeared in the same region as a broad multiplet at d 3.75 and 3.86, respectively. This is believed to be due to an exchange process occurring between the PriOH and isopropyl dithiocarbonato ligands on the NMR timescale.The 13C NMR data for complexes 1–8 in (CD3)2SO are summarised in Table 1. They support the presence of [S2C(OR)]2 and ROH ligands.The carbon atom of the CS2 group was observed at low field, d ca. 230, for all the dithiocarbonato complexes. Free CS2 has a 13C resonance at d 192. Since the alkyl dithiocarbonato complexes 1–8 had low solubilities in common organic solvents it was necessary to use dimethyl sulfoxide as the solvent for the measurement of the NMR spectra. Since it is a co-ordinating solvent the NMR assignments which were obtained must be treated with some caution, nonetheless the 1H and 13C NMR data are useful for discriminating the diVerent OR environments and therefore provide a basis for confirming the stoichiometries of the compounds.The degree of aggregation of the complexes, however, cannot be established by NMR. Mass spectrometry Mass spectroscopic studies by positive and negative ion fast atom bombardment (FAB1 and FAB2, respectively) for complexes 1–8 yielded complicated fragmentation patterns. Molecular ions were not observed, something that is believed to be due to the poor volatilities and thermal stabilities of these complexes.Nevertheless, certain complexes exhibited fragments in both FAB1 and FAB2 spectra consistent with the presence of M(S2COR), M(S2COR)2, M2(S2COR)2, M2(SCOR)3 and solvated fragments of the type Mn(S2COR)n(ROH)n (see Experimental section). Surprisingly, in the FAB-spectra of the isopropyl dithiocarbonato complexes 6–8 the S2C(OPri)2 ion m/z 135 was observed as the most intense peak. Thermogravimetric analysis (TGA) and diVerential scanning calorimetry (DSC) Thermogravimetric analyses and diVerential scanning calorimetric studies were performed for all alkyl thiocarbonato complexes 1–8.The DSC plot of complex 1 showed one exotherm with a shoulder between 25 and 140 8C. This was associated in the TGA plot with a single rapid weight loss of 70%, corresponding to the loss of three molecules of ethanol, one molecule of [S2C(OR)]2 and a CH2CH3 group. The resulting intermediate was stable up to 200 8C whereupon the OCS group was eliminated to yield MgS at 440 8C.The residue of 13.74% (calc. for MgS 13.86%) was stable to 800 8C. The decomposition of 5 showed a more complicated pattern. The residue (31.53%) remained as a very high percentage even at 800 8C, and possibly corresponds to MgCS2 or MgS2. The DSC plot of 2 shows one endotherm between 25 and 200 8C which is reflected in the TGA trace with a single weight loss of 70%. This corresponds to the loss of one and a half molecules of ethanol and two CSOR groups.The subsequent weight loss of 10% corresponds to the loss of one S atom to yield CaS at 400 8C (residue ca. 20%, calc. for CaS 20.5%). A similar decomposition pattern was obtained for 3 leaving a residue of SrS at 500 8C (observed residue 38.03%, calc. for SrS 36.29%). The DSC curve of complex 4 showed several interesting features. A broad endotherm was observed over the range 25–210 8C within which three distinct peaks were apparent; the first is centred at 55 8C, the second at 100 8C and the third at 180 8C.This was mirrored in the TGA plot with a weight loss of 38.5% corresponding to a third of an ethanol molecule, a COEt group and a CS2 fragment to yield [BaS2(OEt)]. The subsequent weight loss of 18% corresponds to the loss of the SOEt fragment to yield BaS at 500 8C (residue 42.8%, calc. for BaS 42.9%). The DSC of 6 showed two endotherms, the first centred at 70 8C and the second at 120 8C (Fig. 1). This was mirrored in the TGA plot with a single rapid weight loss of 70% corresponding to the loss of one molecule of PriOH and two CSOR groups. The subsequent weight loss of 10% represents the decomposition of CaS2 to CaS (residue 20.07%, calc. for CaS 19.5%). A similar decomposition pattern was observed for 7. The TGA curve showed one major weight loss between 25 and 200 8C, representing the loss of two CSOR groups. The subsequent weight loss of ca. 10% represented the decomposition of SrS2 to SrS at 800 8C (residue 32.84%, calc. for SrS 33.52%). The TGA of 8 showed a continuous weight loss between 25 and 240 8C, corresponding to the loss of one OR and one CSOR group. The second weight loss represented the loss of two S atoms to yield BaS and presumably some carbonaceous deposits at 800 8C. In general the observation that these dithiocarbonato complexes decompose to the corresponding metal sulfides at 500 8C is not too surprising. For example.a similar decomposition pattern to that described for 6 has been previously reported for the calcium methyl dithiocarbonato complex [{Ca(S2COMe)2- (MeOH)4}n].11 Also, Greenwood and Earnshaw12 have described the alkaline-earth-metal sulfides MS as high melting point solids. Crystal structures [{Ba(S2COEt)2}•] 4. X-Ray single crystal analysis confirmed that the insertion of CS2 into both Ba]O bonds of [{Ba- (OEt)2(EtOH)4}n] had occurred to give the polymeric structure [Ba(S2COEt)2}•] illustrated in Fig. 2 in which each barium ion is Fig. 1 The TGA/DSC curves of complex [Ca(S2COPri)2(PriOH)3]? 2PriOH 6; cal = 4.184 JJ. Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 2675 Fig. 2 Part of the polymeric structure of [{Ba(S2COEt)2}•] 4 formed by CS2 insertion into the Ba]O bonds of barium ethoxide showing the nine-coordination at barium. The Roman superscripts refer to atoms at equivalent positions listed in Table 2 nine-co-ordinate. Selected bond lengths and angles are listed in Table 2.The asymmetric unit of the crystal consists of two ethyl dithiocarbonato ligands each linked in a bidentate mode to a barium atom via a sulfur atom [Ba]S(11) 3.209(1), Ba(1)]S(21) 3.237(1) Å] and the oxygen donor [Ba]O(14) 2.967(2), Ba]O(24) 3.042(2) Å]. Each ligand also bonds in chelating mode via both sulfur atoms to adjacent symmetry related barium ions [Ba(1III)]S(11) 3.237(1), Ba(1III)]S(12) 3.200(1); Ba(1IV)]S(21) 3.234(1), Ba(1IV)]S(22) 3.280(1) Å], and in addition one sulfur atom forms a slightly longer monodentate link to a third symmetry related barium ion [Ba(1I)]S(22) 3.335(1) Å].The first ligand therefore bridges two symmetry related barium atoms and the second bridges three. [Ca(S2COPri)2(PriOH)3]?2PriOH 6. The X-ray study of complex 6 confirmed the formation of the monomer [Ca(S2COPri) 2(PriOH)3]?2PriOH resulting from the insertion of CS2 into both M]O metal alkoxide bonds. The molecular structure is shown in Fig. 3 and selected bond lengths and angles are summarised in Table 3. The calcium is seven-co-ordinate with bonds to the two sulfur atoms of both S2C(OPri)2 molecules and to the oxygen atoms of three PriOH ligands; it may be envisaged as having a distorted pentagonal bipyramidal geometry, with two PriOH molecules in the axial sites. Two chelate ligands [mean S]Ca]S 60.85(3)8] are in the equatorial plane together with the third PriOH ligand [deviations from mean plane Ca 0.053, S(1A) 0.069, S(1B) 0.311, S(2A) 0.141, S(2B) 0.369, O(1D) 0.321 Å].The equatorial Ca]O distance [Ca]O(1D) 2.374 Å] is slightly longer than the distance to the two axial PriOH [mean Ca]O 2.351(2) Å]. The two dithiocarbonato ligands are co-ordinated in a slightly asymmetric bidentate fashion each having a relatively short Ca]S bond [Ca]S(2A) 2.897(1) and Ca]S(1B) 2.886(1) Å] and a long Ca]S bond [Ca]S(1A) 2.922(1) and Ca]S(2B) 2.969(1) Å]. The longest of these bonds [Ca]S(2B)] is similar in length to the Ca]S bond [2.961(1) Å] found in the COS Table 2 Selected bond lengths (Å) and angles (8) for complex 4 Ba(1)]O(14) Ba(1)]S(11) Ba(1)]S(21) S(21)]Ba(1IV) S(22)]Ba(1I) S(12)]C(13) S(22)]C(23) O(14)]C(15) O(24)]C(25) C(25)]C(26) S(12I)]Ba(1)]S(11) S(11)]Ba(1)]S(2111) S(11)]Ba(1)]S(11I) S(121)]Ba(1)]S(21) S(21II)]Ba(1)]S(21) S(12)]Ba(1)]S(22II) S(21)]Ba(1)]S(22II) S(21)]Ba(1)]S(22II) S(11)]Ba(1)]S(22III) S(11I)]Ba(1)]S(22III) S(22II)]Ba(1)]S(22III) O(14)]Ba(1)]S(12I) O(14)]Ba(1)]S(11) O(14)]Ba(1)]S(21II) O(14)]Ba(1)]S(11I) O(14)]Ba(1)]S(21) O(14)]Ba(1)]S(22II) O(14)]Ba(1)]S(22II) 2.967(2) 3.2088(9) 3.2374(9) 3.2338(9) 3.3351(9) 1.664(3) 1.690(3) 1.468(4) 1.463(4) 1.482(6) 82.03(2) 155.97(2) 126.80(2) 86.45(3) 122.658(10) 109.20(2) 54.64(2) 163.91(3) 80.79(2) 137.80(2) 76.39(2) 65.59(5) 48.63(5) 121.33(5) 119.20(5) 116.00(5) 69.45(5) 103.00(5) Ba(1)]O(24) S(12)]Ba(1III) S(11)]Ba(1III) S(22)]Ba(1IV) S(11)]C(13) S(21)]C(23) O(14)]C(13) O(24)]C(23) C(15)]C(16) S(12I)]Ba(1)]S(21II) S(12II)]Ba(1)]S(11I) S(21II)]Ba(1)]S(11I) S(11)]Ba(1)]S(21) S(11I)]Ba(1)]S(21) S(11)]Ba(1)]S(22II) S(11)]Ba(1)]S(22II) S(12I)]Ba(1)]S(22III) S(21II)]Ba(1)]S(22III) S(21)]Ba(1)]S(22III) O(14)]Ba(1)]O(24) O(24)]Ba(1)]S(12I) O(24)]Ba(1)]S(11) O(24)]Ba(1)]S(21II) O(24)]Ba(1)]S(11I) O(24)]Ba(1)]S(21) O(24)]Ba(1)]S(22II) O(24)]Ba(1)]S(22III) 3.042(2) 3.1997(10) 3.2368(9) 3.2799(9) 1.707(3) 1.692(3) 1.350(4) 1.333(4) 1.506(5) 115.37(3) 55.19(2) 77.08(2) 72.40(2) 74.71(2) 105.29(2) 116.93(2) 162.80(2) 81.34(2) 87.55(3) 162.88(6) 111.92(5) 114.79(5) 75.50(5) 64.78(5) 47.39(5) 125.19(5) 74.94(5) Symmetry transformations used to generate equivalent atoms: I x, 1– 2 2 y, 1– 2 1 z; II 1 2 x, 1y, 1– 2 2 z; III x, 1– 2 2 y, 1– 2 1 z; IV 1 2 x, 21– 2 1 y, 1– 2 2 z.2676 J.Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 insertion product [{Ca(OCSOMe)2(MeOH)3}2], the only other Ca]S linkage from a bidentate ligand co-ordinated to calcium that has been structurally characterised.13 In the solid state molecules of complex 6 form dimers, linked by hydrogen bonding between the proton of an axial PriOH ligand of one molecule and a sulfur donor of a centrosymmetrically related molecule [H(1C) ? ? ? S(2AI) 2.223 Å] as illustrated in Fig. 3(b). The two remaining co-ordinated PriOH ligands form a hydrogen bonded ‘chelate ring’ via hydrogen bonds from their protons to an oxygen atom of one of the solvent PriOH [H(1D) ? ? ? O(1F) 1.846 and H(1E) ? ? ? O(1F) 1.906 Å].One sulfur donor is linked via a relatively weak hydrogen bond to the second solvent PriOH [S(1B) ? ? ? H(1G) 2.416 Å]; hydrogen bonding also links the two solvent molecules at diVerent equivalent positions [H(1F) ? ? ? O(1GIII) 1.822 Å] so that the overall dimeric units are linked into polymeric chains parallel to the a axis. Conclusion This research has demonstrated the insertion of CS2 into both the M]O bonds of alkaline-earth-metal alkoxides, resulting in the formation of the alkyl dithiocarbonato complexes with the general formula [{M(S2COR)2(ROH)x}n].We have recently reported a series of Group 2 metal alkyl thiocarbonato complexes resulting from the insertion of COS into metal–alkoxide bonds.14 The monomeric complexes [Mg(OCSOEt)2(EtOH)4] and [Mg(OCSOPri)2(PriOH)4], the dimer [{Ca(OCSOMe)2- (MeOH)3}2], the ‘trimer’ [Sr3(OCSOEt)6(EtOH)8] and finally the polymer [{Sr(OCSOPri)2(PriOH)2}•] have demonstrated how the nuclearity increases with an increase in the ionic radius of the metal ion.Moving from Mg21, to Ca21 to Sr21, there is a preference for higher co-ordination numbers which are achieved by oligomerisation or polymerisation processes. A similar trend has resulted from the study of the two structurally Table 3 Selected bond lengths (Å) and angles (8) for complex 6 Ca]O(1C) Ca]O(1D) Ca]S(2A) Ca]S(2B) S(2A)]C(3A) O(4A)]C(5A) C(5A)]C(6A) S(2B)]C(3B) O(4B)]C(5B) C(5B)]C(7B) C(2C)]C(3C) O(1D)]C(2D) C(2D)]C(3D) C(2E)]C(4E) O(1F)]C(2F) C(2F)]C(4F) C(2G)]C(3G) O(1C)]Ca]O(1E) O(1E)]Ca]O(1D) O(1E)]Ca]S(1B) O(1C)]Ca]S(2A) O(1D)]Ca]S(2A) O(1C)]Ca]S(1A) O(1D)]Ca]S(1A) S(2A)]Ca]S(1A) O(1E)]Ca]S(2B) S(1B)]Ca]S(2B) S(1A)]Ca]S(2B) C(3A)]S(2A)]Ca C(3B)]S(2B)]Ca Ca]O(1C)]H(1C) Ca]O(1D)]H(1D) Ca]O(1E)]H(1E) 2.346(2) 2.374(2) 2.8974(11) 2.9694(11) 1.690(3) 1.462(3) 1.507(4) 1.671(3) 1.466(3) 1.507(4) 1.495(4) 1.431(3) 1.505(4) 1.495(4) 1.443(3) 1.499(4) 1.501(4) 167.77(6) 76.91(6) 87.30(5) 86.40(5) 81.02(5) 95.50(6) 140.80(5) 61.01(3) 97.74(5) 60.68(3) 135.65(3) 88.10(9) 86.78(9) 122.65(12) 116.44(12) 122.91(13) Ca]O(1E) Ca]S(1B) Ca]S(1A) S(1A)]C(3A) C(3A)]O(4A) C(5A)]C(7A) S(1B)]C(3B) C(3B)]O(4B) C(5B)]C(6B) O(1C)]C(2C) C(2C)]C(4C) C(2D)]C(4D) O(1E)]C(2E) C(2E)]C(3E) C(2F)]C(3F) O(1G)]C(2G) C(2G)]C(4G) O(1C)]Ca]O(1D) O(1C)]Ca]S(1B) O(1D)]Ca]S(1B) O(1E)]Ca]S(2A) S(1B)]Ca]S(2A) O(1E)]Ca]S(1A) S(1B)]Ca]S(1A) O(1C)]Ca]S(2B) O(1D)]Ca]S(2B) S(2A)]Ca]S(2B) C(3A)]S(1A)]Ca C(3B)]S(1B)]Ca C(2C)]O(1C)]Ca C(2D)]O(1D)]Ca C(2E)]O(1E)]Ca 2.357(2) 2.8862(9) 2.9218(11) 1.675(2) 1.335(3) 1.498(4) 1.691(3) 1.337(3) 1.496(4) 1.442(3) 1.498(4) 1.506(4) 1.443(3) 1.510(4) 1.491(4) 1.438(3) 1.501(4) 91.18(7) 103.75(5) 138.48(5) 88.92(5) 137.51(3) 92.10(5) 76.85(3) 83.48(5) 83.45(5) 161.26(3) 87.56(9) 89.20(9) 129.7(2) 133.7(2) 130.8(2) characterised dithiocarbonato complexes [Ca(S2COPri)2- (PriOH)3] 6 and [{Ba(S2COEt)2}•] 4.The increasing size of the metal ion [from Ca (1.06) to Ba (1.43) Å] results in polymerisation. The seven-co-ordinate calcium ion in 6 is able to achieve co-ordinative saturation by co-ordination through the two sulfur atoms of the two dithiocarbonato ligands and three PriOH molecules. The barium ion in 4 is in a nine-co-ordinate environment with S2C(OEt)2 ligands bridging adjacent barium ions forming polymeric sheets.In the previously reported alkyl thiocarbonato Group 2 complexes as well as in 6 all complexes incorporated alcohol molecules in their co-ordination sphere, in order to achieve co-ordinative saturation. Interestingly, the crystal structure of 4 shows that co-ordinative saturation is achieved by utilising the bridging capability of the ethyl dithiocarbonato ligand. The OCS(OR)2 ligand in the previously studied alkyl thiocarbonato complexes, demonstrated a variety of bonding motifs, monodentate, bidentate or bridging two adjacent metal ions.14 In this research the S2C(OPri)2 ligand has proved to be flexible, adopting diVerent co-ordination modes but also bridging three metal ions.Acknowledgements The EPSRC is thanked for financial support and BP plc for endowing D. M. P. M.’s Chair. Fig. 3 (a) The distorted pentagonal geometry of the CS2 insertion product [Ca(S2COPri)2(PriOH)3]?2PriOH 6. (b) The dimer formed in the crystal of complex 6 by hydrogen bonding between an isopropyl dithiocarbonato ligand of one molecule and the proton of a co-ordinated propanol on a centrosymmetrically related molecule.The symmetry operations relating equivalent atoms are: I 2x, 2y, 2z; II 21 2 x, 2y, 2z; III 1 1 x, y, zJ. Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 2677 Table 4 Crystal data and structure refinement details for compounds 4 and 6 Formula Mr Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 Z Dc/Mg m23 F(000) Crystal size/mm l/Å m(Mo-Ka)/mm21 q Range/8 hkl Ranges Reflections collected Unique reflections Minimum and maximum transmission Data, restraints, parameters Goodness of fit on F2, S Final R1, wR2 I > 2s(I) All data Weighting, w21 Largest peak and hole/e Å23 4 C6H10BaO2S4 379.72 Monoclinic P21/c (no. 14) 10.2864(11) 11.3404(14) 10.8724(14) 108.885(8) 1200.0(2) 4 2.102 728 0.28 × 0.28 × 0.12 0.710 73 3.973 2.09 to 24.99 21 to 12, 213 to 1, 212 to 12 2766 2115 0.724/0.811 2115, 0, 118 1.058 0.0217, 0.0512 0.0257, 0.0534 s2(Fo)2 1 (0.0311P)2 1 0.3023P 0.500, 20.535 6 C23H54CaO7S4 610.98 Triclinic P1� (no. 2) 9.611(2) 12.741(3) 16.012(3) 92.33(2) 103.74(2) 112.00(2) 1747.5(6) 2 1.161 664 0.56 × 0.50 × 0.48 0.710 73 0.451 1.32 to 25.00 21 to 11, 214 to 14, 219 to 18 7325 6123 0.500/0.861 6119, 0, 330 1.026 0.0402, 0.0731 0.0703, 0.0925 s2(Fo)2 1 (0.0283P)2 1 0.6910P 0.248, 20.226 S = [Sw(Fo 2 2 Fc 2)2/(n 2 p)]� �� , R1 = S||Fo| 2 |Fc||/S|Fo|, wR2 = Sw(Fo 2 2 Fc 2)2/S[w(Fo 2)2]� �� , P = [max(Fo 2, 0) 1 2(Fc 2)]/3 where n = number of reflections and p = total number of parameters.References 1 J. A. Ibers, Chem. Soc. Rev., 1982, 11, 57. 2 K. K. Pandey, Coord. Chem. Rev., 1995, 140, 37. 3 I. S. Butler and A. E. Fenster, J. Organomet. Chem., 1974, 66, 161. 4 I. S. Butler, Acc. Chem. Res., 1977, 10, 359. 5 P. V. IaneV, Coord. Chem. Rev., 1977, 23, 183. 6 H. Werner, Coord. Chem. Rev., 1982, 43, 165. 7 M. F. Lappert, Adv. Organomet. Chem., 1967, 5, 257. 8 SHELXTL, PC version 5.03, Siemens Analytical Instruments Inc., Madison, WI, 1994. 9 S. A. S. Miller, Ph. D. Thesis, Imperial College, London, 1995; D. J. Otway, Ph. D. Thesis, Imperial College, London, 1994. 10 A. J. Goodsel and G. Blyholder, J. Am. Chem. Soc., 1972, 94, 6725; L. J. Bellamy, The infrared spectra of complex molecules, 2nd edn., Wiley, New York, 1958, pp. 353–356; C. N. R. Rao and R. Venkatarachavan, Spectrochim. Acta, 1962, 18, 541; G. W. Watt and B.J. McCormick, Spectrochim. Acta, 1965, 21, 753; L. H. Little, G. W. Poling and J. Leja, Can. J. Chem., 1961, 39, 745; N. B. Colthur and L. Porter-Powell, Spectrochim. Acta, Part A, 1987, 43, 317. 11 V. C. Arunasalam, Ph. D. Thesis, Imperial College, London, 1995. 12 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements, Pergamon, Oxford, 1984, pp. 803–806. 13 V. C. Arunasalam, D. M. P. Mingos, J. C. Plakatouras, I. Baxter, M. B. Hursthouse and K. M. A. Malik, Polyhedron, 1995, 14, 1105. 14 I. K. Bezougli, A. Bashall, M. McPartlin and D. M. P. Mingos, J. Chem. Soc., Dalton Trans., 1997, 2, 287; preceding paper. Received 8th April 1998; Paper 8/02681EJ. Chem. Soc., Dalton Trans., 1998, Pages 2671–2677 2677 Table 4 Crystal data and structure refinement details for compounds 4 and 6 Formula Mr Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 Z Dc/Mg m23 F(000) Crystal size/mm l/Å m(Mo-Ka)/mm21 q Range/8 hkl Ranges Reflections collected Unique reflections Minimum and maximum transmission Data, restraints, parameters Goodness of fit on F2, S Final R1, wR2 I > 2s(I) All data Weighting, w21 Largest peak and hole/e Å23 4 C6H10BaO2S4 379.72 Monoclinic P21/c (no. 14) 10.2864(11) 11.3404(14) 10.8724(14) 108.885(8) 1200.0(2) 4 2.102 728 0.28 × 0.28 × 0.12 0.710 73 3.973 2.09 to 24.99 21 to 12, 213 to 1, 212 to 12 2766 2115 0.724/0.811 2115, 0, 118 1.058 0.0217, 0.0512 0.0257, 0.0534 s2(Fo)2 1 (0.0311P)2 1 0.3023P 0.500, 20.535 6 C23H54CaO7S4 610.98 Triclinic P1� (no. 2) 9.611(2) 12.741(3) 16.012(3) 92.33(2) 103.74(2) 112.00(2) 1747.5(6) 2 1.161 664 0.56 × 0.50 × 0.48 0.710 73 0.451 1.32 to 25.00 21 to 11, 214 to 14, 219 to 18 7325 6123 0.500/0.861 6119, 0, 330 1.026 0.0402, 0.0731 0.0703, 0.0925 s2(Fo)2 1 (0.0283P)2 1 0.6910P 0.248, 20.226 S = [Sw(Fo 2 2 Fc 2)2/(n 2 p)]� �� , R1 = S||Fo| 2 |Fc||/S|Fo|, wR2 = Sw(Fo 2 2 Fc 2)2/S[w(Fo 2)2]� �� , P = [max(Fo 2, 0) 1 2(Fc 2)]/3 where n = number of reflections and p = total number of parameters. References 1 J. A. Ibers, Chem. Soc. Rev., 1982, 11, 57. 2 K. K. Pandey, Coord. Chem. Rev., 1995, 140, 37. 3 I. S. Butler and A. E. Fenster, J. Organomet. Chem., 1974, 66, 161. 4 I. S. Butler, Acc. Chem. Res., 1977, 10, 359. 5 P. V. IaneV, Coord. Chem. Rev., 1977, 23, 183. 6 H. Werner, Coord. Chem. Rev., 1982, 43, 165. 7 M. F. Lappert, Adv. Organomet. Chem., 1967, 5, 257. 8 SHELXTL, PC version 5.03, Siemens Analytical Instruments Inc., Madison, WI, 1994. 9 S. A. S. Miller, Ph. D. Thesis, Imperial College, London, 1995; D. J. Otway, Ph. D. Thesis, Imperial College, London, 1994. 10 A. J. Goodsel and G. Blyholder, J. Am. Chem. Soc., 1972, 94, 6725; L. J. Bellamy, The infrared spectra of complex molecules, 2nd edn., Wiley, New York, 1958, pp. 353–356; C. N. R. Rao and R. Venkatarachavan, Spectrochim. Acta, 1962, 18, 541; G., Spectrochim. Acta, 1965, 21, 753; L. H. Little, G. W. Poling and J. Leja, Can. J. Chem., 1961, 39, 745; N. B. Colthur and L. Porter-Powell, Spectrochim. Acta, Part A, 1987, 43, 317. 11 V. C. Arunasalam, Ph. D. Thesis, Imperial College, London, 1995. 12 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements, Pergamon, Oxford, 1984, pp. 803–806. 13 V. C. Arunasalam, D. M. P. Mingos, J. C. Plakatouras, I. Baxter, M. B. Hursthouse and K. M. A. Malik, Polyhedron, 1995, 14, 1105. 14 I. K. Bezougli, A. Bashall, M. McPartlin and D. M. P. Mingos, J. Chem. Soc., Dalton Trans., 1997, 2, 287; preceding paper. Received 8th April 1998; Paper 8/02681E

ISSN:1477-9226    

DOI:10.1039/a802681e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 2675–2681 2675 Synthesis and characterization of dinuclear complexes containing the FeIII–F (H2O)MII motif Morten Ghiladi,a Kenneth B. Jensen,a Jianzhong Jiang,b Christine J. McKenzie,*a Steen Mørup,b Inger Søtofte c and Jens Ulstrup c a Department of Chemistry, University of Southern Denmark, Main campus: Odense University, DK-5230 Odense M, Denmark. E-mail: chk@chem.sdu.dk b Department of Physics, Technical University of Denmark, Building 307, DK-2800 Lyngby, Denmark c Department of Chemistry, Technical University of Denmark, Building 207, DK-2800 Lyngby, Denmark Received 6th April 1999, Accepted 23rd June 1999 The dinucleating phenolate-hinged ligand 4-tert-butyl-2,6-bis[bis(2-pyridylmethyl)aminomethyl]phenolate (bpbp) has been used to prepare a series of FeIIIMII complexes containing independent species at the exogenous binding sites.These sites are occupied by .uoride and water ligands and show the general formulation [(bpbp)Fe(F)2M(H2O)n]- [BF4]2, M = Zn or Cu, n = 1; M = Co or Fe, n = 2.Two terminal .uoride ions are bound to the iron(...) ion and one or two water ligands to the adjacent divalent metal ion. The .uoride ligands are derived from the hydrolysis of tetra.uoroborate. In the crystal structure of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O. The copper(..) and iron(...) atoms are linked asymmetrically by the phenolic oxygen atom hinge of bpbp with Cu–Ophenolato 2.270(2) and Fe–Ophenolato 2.041(2) Å with a Cu Fe distance of 3.828(1) Å.The two terminal .uoride ions are bound to the Fe atom (Fe–F 1.818(2), 1.902(2) Å) and one of them is strongly hydrogen bonded to the water molecule on the adjacent Cu atom (F–H O 2.653(4) Å). The metal ions in the aqua.uoride complexes [(bpbp)Fe(F)2M(H2O)2][BF4]2, M = Fe or Co, are weakly antiferromagnetically coupled (J = 8 and 10 cm1 respectively) and in [(bpbp)Fe(F)2Cu(H2O)][BF4]2 are weakly ferromagnetically coupled (J = 2 cm1).The spectroscopic, electrochemical and magnetic properties of these complexes are compared to those of an analogous series of complexes containing two acetate bridging groups in the exogenous site. Electrochemical results indicate that the iron(...) ions in the bis-.uoride complexes are stabilized by about 300 mV towards reduction compared to the bis-µ-acetate complexes. The crystal structure of one bis-µ- acetate complex, [Fe2(bpbp)(CH3CO2)2][BF4]2, shows the expected arrangement; the iron-(..) and -(...) atoms are triply bridged by the phenolic oxygen atom of bpbp and two µ-acetate groups with FeII–Ophenolato 2.088(4) and FeIII– Ophenolato 1.951(5) Å and an Fe Fe distance of 3.380(2) Å.The crystal structure at 120 K indicates that the iron atoms are valence trapped and in accordance with this Mössbauer measurements between 80 and 200 K show clearly distinguishable iron-(..) and -(...) components.The Mössbauer spectra of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O are in.uenced by paramagnetic relaxation e.ects with relaxation times of the order of 1 ns. The relaxation time increases when a magnetic .eld is applied. This e.ect can be explained by a model for cross-relaxation in conjunction with the crystal symmetry of the compound. Introduction Mixed-valence diiron and dimanganese and heterometallic complexes using acyclic phenolate-hinged ligands with two chemically identical metal binding compartments are well known, and with few exceptions 1,2 they contain bridging exogenous ligands, usually oxo-acids or tetrahedral oxoanions.3 The present article describes a continuation of our work on the isolation and characterization of mixed-valence and heterodimetallic complexes of 4-tert-butyl-2,6-bis[bis(2-pyridylmethyl) aminomethyl]phenolate (bpbp) in which the remaining metal co-ordination sites are occupied by terminal ligands.The compounds here contain the [(bpbp)FeIIIF2MII]2 core, where M = Fe, Cu, Co or Zn.The .uoride ions are bound as terminal ligands to the iron(...) atoms and water(s) are coordinated as terminal ligands to the divalent metal ions. These systems are rare examples of dimetallic complexes in which the endogenous ligand furnishes the only group linking the two metal ions; the species bound at the exogenous sites are independent (Fig. 1). One of these compounds, [(bpbp)Fe- (F)2Fe(H2O)2][BF4]24H2O, was reported earlier,2 and a poor quality crystal structure was obtained; unfortunately doubt remained as to the location of hydrogen atoms, and indeed as to the nature of the mononegative terminal ligands (.uoride or hydroxide ions).The characterization of analogous mixed metal complexes here now supports this original formulation. Apart from the binary metal .uorides, only two structurally characterized iron co-ordination compounds containing the .uoride ion as a terminal ligand are known.4 In contrast to the systems described here these are iron(...) porphyrin complexes. The preparation of transition metal complexes containing terminal .uoride ligands is not trivial since addition of F ions to a reaction mixture seldom leads to the pure .uoro complex.This obstacle has been overcome by using anhydrous metal .uorides as starting materials in suitable solvents.5 Fluoridecontaining transition metal complexes have been obtained also through partial decomposition of poly.uorinated counter ions, Fig. 1 Core structures based on [(bpbp)FeIIIMII]2: (a) triply bridged bis-µ-acetato and (b) singly bridged di.uoroaqua structures (the phenolate-hinged dinucleating ligand, bpbp, is represented by dashed lines). Fe O O C H3C M O O C CH3 Fe F F M OH2 Fe F F M H2O OH2 (a) (b)2676 J. Chem. Soc., Dalton Trans., 1999, 2675–2681 a synthetic strategy which is based on the work of Musgrave and Linn.6 Using this method Reedijk and co-workers 7 have prepared copper and cobalt complexes of azole ligands containing .uoride as bridging moieties.We have found in the present work that the preparation of complexes containing the Fe– F (H2O)M moiety can also be accomplished using this method. Heterodimetallic Fe/M or mixed-valence diiron complexes containing the .uoride ligands are relevant structural models for .uoride-inhibited purple acid phosphatases (PAPs). To date no structural data using single crystal methods or EXAFS to analyse .uoride inhibited PAPs have been reported.Crowder and co-workers 8 have carried out kinetic and EPR studies of the reaction of bovine spleen PAP with an excess of .uoride, however the number of .uoride ions co-ordinated to the active site was not ascertained. The crystal structure of red kidney bean PAP without substrate or inhibitor at the dimetallic active site has been solved and the distance between the iron(...) and zinc(..) ions re.ned to 3.26 Å.9 Consistent with this distance, and apart from the endogenous protein-derived bridging aspartate oxygen atom, a bridging exogenous hydroxide ion was assigned on the basis of the spectroscopic and kinetic studies.In the .uoride inhibited PAP it has been anticipated that a .uoride ion can replace co-ordinated hydroxide ion. Thus the characterization of iron(...)/metal(..) complexes with bridging and/or terminal .uoride ions is pertinent in establishing possible F ion inhibition modes in the PAPs. Experimental Dry acetonitrile and dry MeOH were obtained from Sigma- Aldrich Chemical Company, Inc.(>99.8%) and used as received. The UV/VIS spectra were recorded on a Shimadzu UV-3100 spectrophotometer, IR spectra of the complexes in KBr discs using an Hitachi 270-30 IR spectrometer, fast atom bombardment mass spectra on a Kratos MS50RF spectrometer and 19F NMR spectra were recorded on a Varian Gemini 2000 using a,a,a-tri.uorotoluene as an internal reference. Elemental analyses were performed at the Chemistry Department II at Copenhagen University and Atlantic Microlab, Inc., Norcross, Georgia, USA.Fluoride analyses were carried out by Dr J. Theiner, Institut für Physikalische Chemie der Universität Wien, Mikroanalytisches Laboratorium, A-1090 Vienna, Austria. Cyclic voltammograms were recorded using an Eco Chemie Autolab potentiostat equipped with an ECD lowcurrent auxiliary module and controlled by the General Purpose Electrochemical Systems v.3.2 software (Eco Chemie software).The all-glass cell consisted of a working and a reference compartment connected via a Luggin capillary. The working compartment contained a platinum disc (5 mm in diameter) working electrode and a semi-cylindrical platinum gauze auxiliary electrode. The reference compartment contained a silver wire reference electrode immersed in a 0.01 M AgNO3 solution in dry solvents separated from the bulk solution by a porous Vycor plug.Dry acetonitrile, methanol or 20% v/v acetone in CH2Cl2 solutions, with 0.1 M tetrabutylammonium perchlorate as the supporting electrolyte, were used. For measurements in acetone–CH2Cl2 solutions the reference electrode contained dry acetonitrile. Ambient temperature and an N2 atmosphere were used throughout. The ferrocenium–ferrocene couple was used to check the reference electrode potential in acetonitrile, methanol and 20% v/v solution of acetone in CH2Cl2 and was found to occur at 89, 175 and 200 mV, respectively. Magnetic susceptibility measurements were performed by the Faraday method in the temperature range 78–300 K at a .eld strength of 1.3 T using instrumentation described elsewhere.10 The variation of susceptibility with temperature can be described reasonably well by the equations derived from the Heisenberg– Dirac–Van Vleck model for isotropic binuclear magnetic exchange interactions (H = 2JS1S2).11 The J values are estimated to be accurate to within 20%. The molar susceptibility was corrected for underlying diamagnetism by the use of Pascal’s constants.Mössbauer spectra were obtained with a conventional Mössbauer spectrometer in the constant acceleration mode using a 50 mCi source of 57Co in Rh. The calibration was performed using a 12.5 µm thick foil of a-Fe at room temperature, relative to which all isomer shifts are given. The spectra were computer .tted using a least squares procedure.The preparations of 4-tert-butyl-2,6-bis[bis(2-pyridylmethyl)- aminomethyl]phenol (Hbpbp), [(bpbp)Fe(F)2Fe(H2O)2][BF4]2 4H2O and [(bpbp)FeM(CH3CO2)2][ClO4]2 (M = Co, Ni, Cu, Zn or Fe) are described elsewhere.2 Preparations [(bpbp)Fe(F)2M(H2O)n][BF4]2, for M Zn or Cu, n 1; for M Co, n 2. A solution of Hbpbp (0.3000 g, 0.52 mmol) in 2 mL of acetone was added to a solution of Fe(BF4)26H2O (0.1755 g, 0.52 mmol) in 10 mL water. One equivalent of the appropriate hydrated M(BF4)2 salt was then added.The products crystallized in ca. 95% yield within two days. Calc. for [(bpbp)Fe(F)2Co(H2O)2][BF4]22.5H2O, C36H48B2CoF10Fe- N6O5.5: C, 44.16; H, 4.94; F, 19.40; N, 8.58. Found: C, 44.13; H, 4.66; F, 19.91; N, 8.40%. FAB mass spectrum: m/z 724 (35, [(bpbp)FeCo(F)2]) and 743 (28%, [(bpbp)FeCo(F)3]). 19F NMR (300 MHz): d 8.9 (s) and 78.0 (s). UV-Vis (methanol): .max/nm (e/dm3 mol1 cm1) 255 (14500), 295 (6800, sh) and 476 (650). Calc. for [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O, C36H49B2- CuF10FeN6O6: C, 43.55; H, 4.97; F, 19.14; N, 8.46.Found: C, 43.86; H, 4.60; F, 20.15; N, 8.37%. FAB mass spectrum: m/z 260 (100, [C12H11N3Cu]), 709 (25, [(bpbp)FeCu(F)]), 728 (40, [(bpbp)FeCu(F)2]) and 747 (33%, [(bpbp)FeCu(F)3]). 19F NMR (300 MHz): d 4.0 (s) and 80.6 (s). UV-Vis (methanol): .max/nm (e/dm3 mol1 cm1) 253 (15800, sh), 300 (7180, sh) and 494 (860). Calc. for [(bpbp)Fe(F)2Zn(H2O)][BF4]23.5H2O, C36H48N6O5.5B2F10FeZn: C, 43.47; H, 4.97; F, 19.10; N, 8.45.Found: C, 44.01; H, 4.72; F, 18.75; N, 8.41%. FAB mass spectrum: m/z 729 (70, [(bpbp)FeZn(F)2]) and 748 (75%, [(bpbp)FeZn(F)3]). 19F NMR (300 MHz): d 5.8 (s) and 78.2 (s). UV-Vis (methanol): .max/nm (e/dm3 mol1 cm1) 258 (13100), 295 (6020) and 472 (600). [Fe2(bpbp)(CH3CO2)2][BF4]2. A solution of Hbpbp (0.0711 g, 0.124 mmol) in 1 mL of acetone was added to a solution Fe(BF4)26H2O (0.0838 g, 0.248 mmol) in 1 mL water. Ethyl acetate was di.used into the mixture and dark blue crystals of the product deposited over several days in 70% yield.Calc. for [Fe2(bpbp)(CH3CO2)2][BF4]2, C40H45B2F8Fe2N6O5: C, 49.27; H, 4.65; N, 7.18. Found: C, 49.36; H, 4.50; N, 7.27%. UV-Vis (CH3CN): .max/nm (e/dm3 mol1 cm1) 256 (21900), 296 (sh, 8440), 329 (sh, 5140), 381 (sh, 3000) and 555 (1040). X-Ray crystallography Crystals of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O and [Fe2- (bpbp)(CH3CO2)2][BF4]2 were obtained directly from reaction mixtures.Table 1 contains the crystal data and details of the structural determinations. The crystals were cooled to 120 K using a Cryostream nitrogen gas cooler system.12 The data were collected on a Siemens SMART Platform di.ractometer with a CCD area sensitive detector. Three of the water molecules of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O are disordered. The nonhydrogen atoms were re.ned anisotropically. The hydrogen atoms of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O except for the disordered water molecules were located from electron-density di.erence maps and re.ned isotropically.The hydrogen atoms of the disordered water molecules were not included in the re.nement. The hydrogen atoms of [Fe2(bpbp)(CH3CO2)2]- [BF4]2 were placed at calculated positions using a riding model with .xed thermal parameters [U(H) = 1.2U for attached atom].J. Chem. Soc., Dalton Trans., 1999, 2675–2681 2677 Programs used for data collection, data reduction and absorption were SMART, SAINT and SADABS.13 The program SHELXTL 9514 was used to solve the structures and for molecular graphics; PLATON15 was used for molecular geometry calculations. CCDC reference number 186/1536.See http://www.rsc.org/suppdata/dt/1999/2675/ for crystallographic .les in .cif format. Results and discussion Syntheses The dinuclear complexes [(bpbp)FeIII(F)2M(H2O)n][BF4]2, M = Zn or Cu, n = 1; M = Co or Fe,2 n = 2, are prepared from a mixture of Hbpbp and one equivalent each of iron(..) tetra- .uoroborate and the appropriate divalent metal tetra.uoroborate. The compounds crystallize in almost quantitative yields within two days at ambient temperature.The two .uoride ions bound to the iron(...) are derived from hydrolysis of the counter anion. Attempts to prepare the same series of complexes using .uoride salts or metal .uorides as the .uoride ion source did not result in tractable products. Interestingly di.usion of ethyl acetate into a 1 : 2 mixture of Hbpbp and iron tetra.uoroborate in acetone led to the precipitation of the bis-acetate-bridged complex, [Fe2(bpbp)(CH3CO2)2]- [BF4]2 rather than the bis .uoro complex [(bpbp)Fe(F)2- Fe(H2O)2][BF4]24H2O that is obtained in its absence.The acetate bridging groups are derived from the hydrolysis of ethyl acetate, thus hydrolysis rather than abstraction of .uoride from tetra.uoroborate is apparently favoured under these reaction conditions. The crystal structure of [Fe2(bpbp)(CH3- CO2)2][BF4]2 was determined in order to verify the presence of two acetate bridges.The novel alternative of a combination of one acetate bridge and terminal .uoride, water or hydroxide ligands was considered a possibility given the unusual reaction conditions. The preparations of this bis-acetato-bridged complex and those we have previously reported with perchlorate counter anions 2 contrast markedly to the preparations of other similar acetate-bridged complexes for which the bridging groups are provided by acetate ion sources.3 Elemental analyses of the .uoride complexes have been .tted with extra water corroborating the proposal of one or two waters as ligands on the divalent metal ions.Water as lattice solvent is present also in all the complexes and con.rmed in the crystal structures of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O (Fig. 2) and [(bpbp)Fe(F)2Fe(H2O)2][BF4]24H2O.2 It is not possible unequivocally to assign the number of water ligands in the complexes without crystal structures however a formulation has been made on the basis of expected geometrical similarities of the Fe/Fe (crystal structure) and Fe/Co and the Fe/Cu (crystal structure) and Fe/Zn complexes.Thus there is one terminal aqua ligand in the complexes containing CuII and ZnII, and two aqua ligands in the complexes containing FeII and CoII. Five- or six-co-ordination is feasible for the Zn atom in the noncrystallographically characterized Fe/Zn complexes, thus it is not possible to distinguish between the structural formulations of [(bpbp)Fe(F)2Zn(H2O)]2 or [(bpbp)Fe(F)2Zn(H2O)2]2 for the cations in the Fe/Zn complexes in the absence of a crystal structure.A geometry close to that found for the copper containing analogue in Fig. 2 is considered as the most likely, hence the formulation given for the zinc complex. Crystals of the Fe/ Co and the Fe/Zn complexes di.racted poorly and in attempts to obtain better crystals these complexes were isolated as perchlorate salts by using zinc or cobalt perchlorate in place of the tetra.uoroborate starting materials.The full characterization of these perchorate salts was not made since even though suitably sized crystals of [(bpbp)Fe(F)2Co(H2O)2][ClO4]2 and [(bpbp)Fe(F)2Zn(H2O)][ClO4]2 were formed these also diffracted too weakly for crystal structure analysis. Establishing the presence of .uoride ligands (e.g. rather than OH) by elemental analysis was di.cult in light of the presence of .uoride in the counter anions.However .uorine analyses combined with 19F NMR spectroscopy con.rm the presence of both co-ordinated .uoride and the .uoride of the counter anion. The signal for the co-ordinated .uoride ion at ca. d 78 for all the complexes is broad and paramagnetically shifted. The FAB mass spectra of all the tetra.uoroborate complexes show intense peaks corresponding to the di.uorinated cations [(bpbp)FeM(F)2].Ions containing three .uoride atoms were also observed; these may be assigned to the tri.uorinated cations [(bpbp)FeM(F)3] or ion pairs {[(bpbp)FeM- (F)2]2[F]} which may result from the gas phase decomposition of the counter anion. In no case is the water ligand(s) retained. The mass spectrometry results are consistent with the 100% formation of the mixed metal complexes; no signals that can be assigned to homodinuclear analogues are detected. Crystal structure of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O Selected bond distances and angles are listed in Table 2 (see Table 1 for other crystallographic parameters).The copper(..) and iron(...) atoms in [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O (Fig. 2) are bridged asymmetrically by the phenolic oxygen atom of bpbp with Cu–O 2.270(2) and Fe–O 2.041(2) Å with a Cu Fe distance of 3.828(1) Å. As support for the formulation, the re.nement of this structure was clearly improved by using .uorine atoms rather than hydroxide oxygen atoms as the exogenous terminal ligands attached to FeIII in the model.The two terminal .uoride ions are bound to the Fe atom and one of them, F1, is strongly hydrogen bonded to the water molecule on the adjacent Cu atom (F1 H–O 2.653(4) Å). The other .uoride ion is hydrogen bonded to the ordered water molecule (F2 H–O 2.688(4) Å). The hydrogen atoms of the water molecules were located on electron density di.erence maps. Hydrogen bonding probably stabilizes the terminal .uoride ligands.The geometries around the iron(...) and copper(..) are octahedral and square pyramidal respectively. The apical bond on the copper ion is that to the phenolate oxygen atom. As a result the complex is highly unsymmetrical with a di.erence of 0.14 Å between the Fe–O1 and the Cu–O1 bonds. Notably, and similarly to terminal .uoride ligands, terminal hydroxide ligands are rarely characterized in the solid state in iron(...) complexes.4,16 The crystal structure of [(bpbp)Fe- (F)2Fe(H2O)2][BF4]24H2O reported earlier was of extremely poor quality and the result ambiguous with regard to Fig. 2 The molecular arrangement of the cation in [(bpbp)Fe(F)2- Cu(H2O)][BF4]24H2O, showing 50% probability ellipsoids. The hydrogen atoms are omitted for clarity.2678 J. Chem. Soc., Dalton Trans., 1999, 2675¡V2681Table 1 Crystallographic data[(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O [Fe2(bpbp)(CH3CO2)2][BF4]2FormulaFormula weightCrystal symmetrySpace groupa/b/c/£]/V/3Z(MoK£\)/mm1T/KTotal no.unique reectionsNo. observed reections [I > 2£m(I)]RwR2C36H49B2CuF10FeN6O6976.82MonoclinicP21/c17.9414(8)13.6413(6)18.8617(9)112.260(1)4272.3(3)40.931201084475900.0538 (obs.)0.1481 (all)C40H45B2F8Fe2N6O5975.14MonoclinicP21/n12.191(2)21.792(4)16.435(3)93.92(3)4356.0(15)40.75120854139410.0850 (obs.)0.2144 (all)distinguishing between the presence of terminal uoride vs.terminal hydroxide ligands.2 The structure of [(bpbp)Fe-(F)2Cu(H2O)][BF4]24H2O is of signicantly higher quality.Thus assignment of uoride rather than hydroxide as the terminalligands on the FeIII is beyond doubt, thereby lendingcredence to our previous assignment of terminal uorideligands in [(bpbp)Fe(F)2Fe(H2O)2][BF4]24H2O.Fig. 3 The molecular arrangement of the cation in [Fe2(bpbp)(CH3-CO2)2][BF4]2. Details as in Fig. 2.Table 2 Selected bond lengths () and angles () for the cation of[(bpbp)Fe(F)2Cu(H2O)][BF4]24H2OFe¡VF2Fe¡VF1Fe¡VO1Fe¡VN1Fe¡VN2Fe¡VN3F2¡VFe¡VF1F2¡VFe¡VO1F1¡VFe¡VO1F2¡VFe¡VN1F1¡VFe¡VN1O1¡VFe¡VN1F2¡VFe¡VN2F1¡VFe¡VN2O1¡VFe¡VN2N1¡VFe¡VN2F2¡VFe¡VN3F1¡VFe¡VN3O1¡VFe¡VN31.818(2)1.902(2)2.041(2)2.112(3)2.149(3)2.234(3)100.20(9)106.01(9)90.15(9)162.81(10)89.99(9)87.62(9)91.24(10)168.43(9)84.94(10)79.36(10)86.86(10)87.21(9)167.12(9)Cu¡VN6Cu¡VO2Cu¡VN5Cu¡VN4Cu¡VO1N1¡VFe¡VN3N2¡VFe¡VN3N6¡VCu¡VO2N6¡VCu¡VN5O2¡VCu¡VN5N6¡VCu¡VN4O2¡VCu¡VN4N5¡VCu¡VN4N6¡VCu¡VO1O2¡VCu¡VO1N5¡VCu¡VO1N4¡VCu¡VO1Fe¡VO1¡VCu1.967(3)2.024(2)2.025(3)2.070(3)2.270(2)79.78(10)95.26(10)95.35(11)164.55(11)97.77(11)86.16(11)177.74(11)80.50(11)98.90(9)87.55(9)89.80(9)93.88(9)126.72(10)Crystal structure of [Fe2(bpbp)(CH3CO2)2][BF4]2Selected bond distances and angles are listed in Table 3 (seeTable 1 for other crystallographic parameters).The structure of[Fe2(bpbp)(CH3CO2)2][BF4]2 (Fig. 3) shows an arrangementof the cation similar to that found for the closely related complex[Fe2(bpmp)(C2H5CO2)2][BPh4]2CH3COCH30.5CH3CN,bpmp = 4-methyl-2,6-bis[bis(2-pyridylmethyl)aminomethyl]-phenolate(1).3f The signicant dierences in the Fe¡VO bondlengths about the two iron centres indicate that Fe1 is theiron() and Fe2 the iron() atom.The Fe Fe distance of3.380(2) is comparable to that of 3.365(1) found for [Fe2-(bpmp)(C2H5CO2)2]2,3f and signicantly shorter than theFe M distances in [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O and[(bpbp)Fe(F)2Fe(H2O)2][BF4]24H2O3 (3.828(1) and 3.819(4) respectively).ElectrochemistryCyclic voltammograms of the bis-uoro FeIII/MII complexes arenot as straightforward as for their bis-acetato-bridged counterparts.The latter show clearly reversible behaviour, with noapparent complicating features.However they also pointstrongly towards stabilization of the mixed-valence state. Voltammetricanalysis of the bis-uoro FeIII/FeII complex suggeststhat the exogenous water and/or uoride ligand can easily besubstituted by the external solvent. This is illustrated by the CVTable 3 Selected bond lengths () and angles () for the cation of[Fe2(bpbp)(CH3CO2)2][BF4]2Fe1¡VO5Fe1¡VO1Fe1¡VO4Fe1¡VN6Fe1¡VN5Fe1¡VN4O5¡VFe1¡VO1O5¡VFe1¡VO4O1¡VFe1¡VO4O5¡VFe1¡VN6O1¡VFe1¡VN6O4¡VFe1¡VN6O5¡VFe1¡VN5O1¡VFe1¡VN5O4¡VFe1¡VN5N6¡VFe1¡VN5O5¡VFe1¡VN4O1¡VFe1¡VN4O4¡VFe1¡VN4N6¡VFe1¡VN4N5¡VFe1¡VN4O3¡VFe2¡VO12.019(5)2.088(4)2.108(5)2.148(6)2.184(6)2.206(6)97.8(2)94.4(2)89.1(2)99.1(2)162.8(2)86.7(2)91.8(2)84.7(2)171.8(2)97.6(2)168.6(2)87.4(2)95.9(2)76.5(2)78.5(2)101.0(2)Fe2¡VO3Fe2¡VO1Fe2¡VO2Fe2¡VN3Fe2¡VN2Fe2¡VN1O3¡VFe2¡VO2O1¡VFe2¡VO2O3¡VFe2¡VN3O1¡VFe2¡VN3O2¡VFe2¡VN3O3¡VFe2¡VN2O1¡VFe2¡VN2O2¡VFe2¡VN2N3¡VFe2¡VN2O3¡VFe2¡VN1O1¡VFe2¡VN1O2¡VFe2¡VN1N3¡VFe2¡VN1N2¡VFe2¡VN1Fe2¡VO1¡VFe11.947(5)1.951(5)1.974(5)2.152(6)2.165(6)2.200(5)96.0(2)95.0(2)94.2(2)164.6(2)85.5(2)91.0(2)85.2(2)172.8(2)92.5(2)164.4(2)89.6(2)94.4(2)75.0(2)78.4(2)113.6(2)J. Chem.Soc., Dalton Trans., 1999, 2675–2681 2679 of [(bpbp)Fe(F)2Fe(H2O)2][BF4]2 in methanol shown in Fig. 4. It is dominated by two peaks with the midpoint potentials at ca. 50 and ca. 550 mV. These match approximately the two peaks of the bis-acetato complex, and as for the latter the 600 mV peak separation re.ects strong interaction between the two iron centres. The cathodic and anodic peak separations of the individual signals, however, exceed signi.cantly 59 mV, and additional peaks are conspicuously apparent. This is indicative of the presence of more than a single species and irreversible voltammetric patterns.By contrast to the diiron .uoride complex the heteronuclear complexes show a single reversible FeIII–FeII peak. The signals are weak but di.erential pulse voltammetry clearly substantiates the reversible one-electron nature of the voltammetric signal, by the 95–110 mV peak half-width, and the (1 s)/(1 s) dependence of the peak height, with s = exp(nF.E/2RT), n = number of electrons transferred, F = Faraday’s number, .E = potential relative to the equilibrium potential, R = gas constant and T = temperature.Redox potentials obtained from di.erential pulse voltammetric experiments are listed in Table 4 Fig. 4 The cyclic voltammogram of a 3.1 × 103 M solution [(bpbp)- Fe(F)2Fe(H2O)2][BF4]24H2O in dry methanol. Sweep rate, 250 mV s1; reference, Ag–AgNO3. Table 4 Redox potentials for the FeIIIMII–FeIIMII couples in the di.uoroaqua complexes and their bis-µ-acetato counterparts Redox couple Compound E/mV FeIIIZnII–FeIIZnII FeIIICuII–FeIICuII FeIIICoII–FeIICoII FeIIIFeII–FeIIFeII FeIIIFeIII–FeIIIFeII [(bpbp)Fe(F)2Zn(H2O)][BF4]2 [(bpbp)FeZn(CH3CO2)2][ClO4]2 [(bpbp)Fe(F)2Cu(H2O)][BF4]2 [(bpbp)FeCu(CH3CO2)2][ClO4]2 [(bpbp)Fe(F)2Co(H2O)2][BF4]2 [(bpbp)FeCo(CH3CO2)2][ClO4]2 [(bpbp)Fe(F)2Fe(H2O)2][BF4]2 [Fe2(bpbp)(CH3CO2)2][ClO4]2 [(bpbp)Fe(F)2Fe(H2O)2][BF4]2 [Fe2(bpbp)(CH3CO2)2][ClO4]2 639 a 334 b 623 a 308 b 569 a 352 b 610 a 334 b 359 a 386 b a In 20% v/v acetone in dichloromethane vs.the ferrocene–ferrocenium couple. b From ref. 2; in acetonitrile vs. the ferrocene–ferrocenium couple. which also includes the corresponding values for the FeIIIMII bis-µ-acetato-bridged analogues for comparison. The FeIII–FeII couples for the iron(...) ions in the bis-.uoro complexes are lower by about 300 mV compared to those of the bis-µ-acetatobridged complexes. This result is not unexpected given the higher concentration of negative charge around the FeIII when two .uoride ions are co-ordinated.The similarity of the reduction potential for the FeIII/CoII complex with the other FeIII/MII complexes supports the proposed oxidation state distribution for the metal atoms. Magnetochemistry The magnetic susceptibilities of the aqua.uoride complexes [(bpbp)Fe(F)2Fe(H2O)2][BF4]24H2O, [(bpbp)Fe(F)2Co(H2O)2]- [BF4]22.5H2O, [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O and [(bpbp)Fe(F)2Zn(H2O)][BF4]23.5H2O were recorded between 78 and 300 K. All the complexes, apart from [(bpbp)Fe(F)2Zn- (H2O)][BF4]23.5H2O, show weak exchange coupling, antiferromagnetic in the case of the FeIII/FeII and FeIII/CoII and ferromagnetic in the case of FeIII/CuII (Table 5).For comparison Table 5 contains the magnetic coupling constants obtained for the FeIII/MII bis-µ-acetato-bridged analogues which were not previously reported. The FeIII/ZnII complexes behave as simple paramagnets with magnetic moments of 5.95–6.05 µB over the temperature range measured.The magnetic susceptibility measurements further support the assignment of an FeIII/CoII complex rather than the alternative CoIII/FeII assignment, by the fact that antiferromagnetic coupling is observed and can be best .tted to a high-spin iron(...) and a low-spin cobalt(..) ion. The alternative of a CoIII/FeII complex is excluded since a low spin diamagnetic CoIII is not expected to contribute to the magnetism, such that a simple high-spin iron(..) paramagnet should be observed.To our knowledge no comparable CoIII/FeII or FeIII/CoII complexes exist to which we can compare our results. There is a notable di.erence in going from the singly µ-phenolato-bridged .uoride complexes to the triply µ- phenolato-bis-µ-acetato-bridged complexes in the case of Fe/Cu. Weak ferromagnetic and antiferromagnetic coupling respectively is observed. The result is consistent with the signi.cant geometrical di.erences of the copper ions in [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O and [(bpbp)FeCu(CH3- CO2)2][ClO4]20.5CH3OH.The copper ion in [(bpbp)- FeCu(CH3CO2)2][ClO4]20.5CH3OH shows an octahedral geometry (con.rmed by X-ray crystallography 3), while that in [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O is square pyramidal. A consequence of these geometrical di.erences is the orientation of the magnetic orbitals of the high spin iron(...) and the copper(..) atoms. The Cu–O1 bond is axial in the case of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O suggesting that the magnetic orbital for CuII (dx2 y2) is not coplanar with any of the magnetic orbitals of FeIII.Thus overlap of the magnetic orbitals of the two metal ions, via the hinging phenolate oxygen atom, is vanishing in [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O. This contrasts with the situation for the octahedral copper ion’s magnetic orbital in [(bpbp)FeCu(CH3CO2)2][ClO4]20.5CH3OH which can be orientated such that overlap with the FeIII-based Table 5 Magnetic coupling constants and where possible M–Ophenolato–M angles for the series of di.uoroaqua complexes and their bis-µ-acetatobridged FeIIIMII counterparts Compound J/cm1 M–Ophenolato–M/ FeIIIFeII FeIIICuII FeIIICoII FeIIINiII [(bpbp)Fe(F)2Fe(H2O)2][BF4]24H2O [Fe2(bpbp)(CH3CO2)2][ClO4]2 [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O [(bpbp)FeCu(CH3CO2)2][ClO4]2 [(bpbp)Fe(F)2Co(H2O)2][BF4]22.5H2O [(bpbp)FeCo(CH3CO2)2][ClO4]2 [(bpbp)FeNi(CH3CO2)2][ClO4]2 8 4 2 20 10 6 11 124.6(3) a 113.6(2) 126.7(1) 118.0(2) a a From ref. 2.2680 J. Chem. Soc., Dalton Trans., 1999, 2675.2681 Table 6 Isomer shifts, ¥ä, and quadrupole splittings, .EQ, obtained from .ts of the spectra shown in Figs. 5 and 6 Compound T/K ¥ä/mm s1 .EQ/ mm s1 ¥ä/mm s1 .EQ/ mm s1 Relative area of iron(...) component (%) [(bpbp)Fe(F)2Fe(H2O)n][BF4]2 [(bpbp)Fe(F)2Co(H2O)n][BF4]2 [Fe2(bpbp)(CH3CO2)2][BF4]2 80 80 280 150 80 0.47(1) 0.47(1) 0.44(4) 0.46(4) 0.48(2) 0.20(1) 0.21(1) 0.53(4) 0.47(4) 0.46(2) 1.17(1) . 0.98(4) 1.10(4) 1.13(2) 3.26(1) . 1.63(4) 2.03(4) 2.59(2) 52(4) 100 52(5) 49(5) 49(4) magnetic orbitals via the hinging phenolate oxygen atom is possible. With regard to the other matched pairs of singly and triply bridged complexes there is very little di.erence in the strength of the antiferromagnetic coupling. Within experimental error a trend is clear: acetate groups are sometimes assumed to furnish a minor contribution to magnetic exchange pathways, however the singly bridged complexes (apart from the Fe/Cu discussed above) show the largest values for J.However to counter this e.ect, the F HOH moiety gives rise to a larger ¡°bite¡± distance compared to an acetate bridge with the consequence of a larger M.Ophenolato.M angle (listed in Table 5). Since it is expected that the phenolate oxygen atom provides the major magnetic exchange pathway then a more obtuse M. Ophenolato.M angle will give rise to better ¥� orbital overlap of the metal ion¡�s magnetic orbitals via this oxygen atom.Mossbauer spectroscopy Mossbauer parameters are listed in Table 6. In accordance with the results of magnetic susceptibility and electrochemistry measurements the Mossbauer parameters of [(bpbp)Fe(F)2- Co(H2O)2][BF4]22.5H2O show that all of the iron atoms are in thon(...) high spin state. The spectrum of [(bpbp)Fe(F)2- Fe(H2O)2][BF4]24H2O contains both iron-(..) and -(...) high spin components. The relative area of the two components is within experimental uncertainties 1 : 1 in accordance with an FeII:FeIII ratio of 1 :1.Mossbauer spectra of [Fe2(bpbp)- (CH3CO2)2][BF4]2 were obtained in the temperature range Fig. 5 The Mossbauer spectra of [Fe2(bpbp)(CH3CO2)2][BF4]2 obtained in the temperature range 80.280 K. 80.280 K (Fig. 5) in order to determine the extent of any intramolecular electronic delocalization between the two inequivalent iron atoms. In the whole temperature range the spectra show the presence of clearly distinguishable components due to FeII and FeIII in the high spin state.Thus there is no evidence for electronic delocalization like the e.ect seen in [Fe2(bpmp)(ena)2][BF4]2 (ena = deprotonated heptanoic acid).17 The area ratio of the iron-(..) and -(...) components is very close to 1 : 1 and independent of temperature within experimental uncertainty. This indicates that the thermal vibrations of FeII and FeIII can be described by the same e.ective Debye temperature.From a .t of the temperature dependence of the total spectral area an e.ective Debye temperature of 140 ¡¾ 10 K was obtained. Fig. 6 shows Mossbauer spectra of [(bpbp)Fe(F)2Cu(H2O)]- [BF4]24H2O obtained at 15 K in zero applied magnetic .eld and at 80 K with and without an applied magnetic .eld of 0.5 T. The zero .eld spectra essentially consist of a single line with a linewidth of about 1.8 mm s1, i.e. almost ten times the natural linewidth indicating that the spectra are in.uenced by paramagnetic relaxation e.ects with relaxation times of the order of 1 ns.18 An increase in the linewidth when a magnetic .eld is applied indicates an increase in the relaxation time.A similar e.ect has been observed for Fe(NO3)39H2O and explained on the basis of the crystal symmetry.19 The space group of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O is the same as that of Fe(NO3)39H2O (P21/c) and the relaxation e.ects can therefore be explained by the same type of mechanism.Model complexes for .uoride-inhibited purple acid phosphatases? It is interesting that the .uoride ligands are terminally bound at the exogenous bridging site of [(bpbp)Fe(F)2M(H2O)n]2, since .uoride ligands bound in the terminal mode are rarer than bridging .uorides. Since a phenolate-hinged diiron complex Fig. 6 The Mossbauer spectra of [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O obtained at 80 K without and with an applied magnetic .eld of 0.5 T and at 15 K without an applied .eld.J.Chem. Soc., Dalton Trans., 1999, 2675–2681 2681 with Fe Fe distance as small as 3.193(2) Å has been reported 20 (with an exogenous methoxide as bridging group) there do not appear to be geometrical contraints preventing the formation of a more usual µ-.uoride bridge rather than the aqua-.uoride-bridged compounds we have structurally characterized. This result has implications for the mode of inhibition by .uoride ions in PAPs. The M M distances in the structures of [(bpbp)Fe(F)2Fe(H2O)2][BF4]24H2O (3.726(2) Å) and [(bpbp)Fe(F)2Cu(H2O)][BF4]24H2O (3.828(2) Å) are signi.- cantly larger than that found in the crystal structure of red kidney bean PAP (3.26 Å) 11 without substrate or inhibitor.The two metal ions in these complexes and in red kidney bean PAP are linked by one endogenous oxygen atom, and we suggest that if F can substitute the isoelectronic hydroxide ion(s) proposed to be bound at the iron(...) atom in PAPs then it might bind in a terminal, rather than bridging mode, and be stabilized by hydrogen bonding to water molecules/ligands or amino acid side chains.Thus a structural motif like the FeIII– F (H2O)MII moieties that we have structurally characterized is pertinent. Conclusion The preparation of iron(...) complexes containing terminal .uoride ligands is unusual and in our hands their synthesis is possible only by exploiting the controlled decomposition of the counter anion, tetra.uoroborate. The complexes are novel mixed-metal and mixed-valence systems of an acyclic phenolate-hinged dinucleating ligand in which the exogenous site is not occupied by a bidentate bridging group.Stabilization of the terminal .uoride ligands by hydrogen bonding to adjacent water ligands and lattice water is likely to be important for their existence. The iron(...) ions of the bis-.uoride complexes are stabilized by ca. 300 mV towards reduction compared to the corresponding FeIIIMII bis-µ-acetato-bridged complexes.Thus the mixed-valence state for the singly bridged diiron complex is more stable that its triply bridged counterpart. X-Ray analysis shows that the “bite” distance of the (HO–H F) group is larger that that for a bidentate carboxylate bridge with the structural consequence of an approximately 0.4 Å greater metal–metal separation for the .uoride complexes compared to their di-µ-carboxylate-bridged counterparts. With respect to the possible structural relevance of [(bpbp)Fe(F)2M(H2O)n][BF4]2, M = Zn or Cu, n = 1; M = Co or Fe, n = 2, to .uoride-inhibited PAPs, we believe that if .uoride is bound at the active site then it will be bound to the iron(...) centre, and that on the basis of the present work that it is necessary to consider the hydrogen bonded structural moiety Fe–F H–O(H)–M, M = FeII or ZnII, in the case of the mammalian and plant PAPs respectively.Acknowledgements This work was supported by a Ph.D.grant (M. G.) and grant no. 28808 (C. J. M.) from the Danish Natural Science research council. We are grateful to Dr J. Theiner, Institut für Physikalische Chemie der Universität, Vienna, Austria for the .uoride analyses. References 1 A. Hazell, C. J. McKenzie, B. Moubaraki and K. S. Murray, Acta Chem. Scand., 1997, 51, 470. 2 M. Ghiladi, C. J. McKenzie, A. Meier, A. K. Powell, J. Ulstrup and S. Wocadlo, J. Chem Soc., Dalton Trans., 1997, 4011. 3 (a) M.Suzuki, M. Mikuriya, S. Murata, A. Uehara and H. Oshio, Bull. Soc. Chem. Jpn., 1987, 60, 4305; (b) A. S. Borovik and L. Que, Jr., J. Am. Chem. Soc., 1988, 110, 2345; (c) H. Diril, H.-R. Chang, M. J. Nilges, X. Zhang, J. A. Potenza, H. J. Schugar, D. N. Hendrickson and S. S. Isied, J. Am. Chem. Soc., 1988, 110, 625; (d ) R. M. Buchanan, K. J. Oberhausen and J. F. Richardson, Inorg. Chem., 1988, 27, 971; (e) H. Diril, H.-R. Chang, M. J. Nilges, X. Zhang, J. A. Potenza, H. J. Schugar, S.S. Isied and D. N. Hendrickson, J. Am. Chem. Soc., 1989, 111, 5102; ( f ) A. S. Borovik, V. Papaefthymiou, L. F. Taylor, O. P. Anderson and L. Que, Jr. J. Am. Chem. Soc., 1989, 111, 6183; ( g) T. R. Holman, C. Juarez- Garcia, M. P. Hendrich, L. Que, Jr. and E. Münck, J. Am. Chem. Soc., 1990, 112, 7611; (h) M. S. Mashuta, R. J. Webb, J. K. McCusker, E. A. Schmitt, K. J. Oberhausen, J. F. Richardson, R. M. Buchanan and D. N. Hendrickson, J. Am. Chem. Soc., 1992, 114, 3815; (i) W.Kanda, W. Moneta, M. Bardet, E. Bernard, N. Debaecker, J. Laugier, A. Boussekou, S. Chardon-Noblat and J.-M. Latour, Angew. Chem., Int. Ed. Engl., 1995, 34, 588; ( j) A. Neves, M. A. de Brito, I. Vencato, V. Drago, K. Griesar and W. Haase, Inorg. Chem., 1996, 35, 2360. 4 K. Anzai, K. Hatano, Y. J. Lee and W. R. Scheidt, Inorg. Chem., 1981, 20, 2337; W. R. Scheidt, Y. J. Lee, S. Tamai and K. Hatano, J. Am. Chem. Soc., 1983, 105, 778; S. C. Lee and R. H. Holm, Inorg.Chem., 1993, 32, 4745; S. Christie, S. Subramanian, L. Wang and M. J. Zaworotko, Inorg. Chem., 1993, 32, 5415. 5 P. Glavic, J. Slivnik and A. Bole, J. Inorg. Nucl. Chem., 1979, 41, 248; K. C. Patil and E. A. Secco, Can. J. Chem., 1972, 50, 567. 6 T. R. Musgrave and T. S. Linn, J. Coord. Chem., 1973, 2, 323. 7 J. Reedijk, J. C. Jansen, H. van Koningsveld and C. G. van Kralingen, Inorg. Chem., 1978, 17, 1990; R. W. M. ten Hoedt, J. Reedijk and G. C. Verschoor, Recl. Trav. Chim. Pays-Bas, 1981, 100, 400; W.C. Velthuizen, J. G. Haasnoot, A. J. Kinneging, F. J. Rietmeijer and J. Reedijk, J. Chem. Soc., Chem. Commun., 1983, 1366; F. J. Rietmeijer, R. A. G. de Graa. and J. Reedijk, Inorg. Chem., 1984, 23, 151; J. Verbiest, J. A. C. van Ooijen and J. Reedijk, J. Inorg. Nucl. Chem., 1980, 42, 971; J. Reedijk, Comments Inorg. Chem., 1982, 1, 379; W. Vreugdenhil, P. J. M. W. L. Birker, R. W. M. ten Hoedt, G. C. Verschoor and J. Reedijk, J. Chem. Soc., Dalton Trans., 1984, 429; J. Reedijk and R. W. M. ten Hoedt, Recl. Trav. Chim. Pays-Bas, 1982, 101, 49. 8 J. B. Vincent, M. W. Crowder and B. A. Averill, Biochemistry, 1991, 30, 3025; M. W. Crowder, J. B. Vincent and B. A. Averill, Biochemistry, 1992, 31, 9603. 9 T. Klabunde, N. Sträter, R. Frölich, H. Witzel and B. Krebs, J. Mol. Biol., 1996, 259, 737. 10 E. Pedersen, Acta Chem. Scand., 1972, 26, 333; J. Josephsen and E. Pedersen, Inorg. Chem., 1977, 16, 2534. 11 C. J. O’Connor, Prog. Inorg. Chem., 1982, 29, 203. 12 J. Cosier and A. M. Glazer, J. Appl. Crystallogr., 1986, 19, 105. 13 Siemans SMART and SAINT, Area-Detector Control and Integration Software, Siemens Analytical X-Ray Instruments Inc., Madison, WI, 1995; G. M. Sheldrick, SADABS, Program for Absorption Correction, Siemens Analytical X-Ray Instruments Inc., Madison, WI, 1996. 14 G. M. Sheldrick, SHELXTL 95, Siemens Analytical X-Ray Instruments Inc., Madison, WI, 1995. 15 A. L. Spek, Acta Crystallogr., Sect. A, 1990, 46, C-34. 16 A. Hazell, K. B. Jensen, C. J. McKenzie and H. Toftlund, Inorg. Chem., 1994, 33, 3127. 17 Y. Maeda, Y. Tanigawa, N. Matsumoto, H. Oshio, M. Suzuki and Y. Takashima, Bull. Soc. Chem. Soc. Jpn., 1994, 67, 125; T. Manago, S. Hayami, H. Oshio, S. Osaki, H. Hasuyama, R. H. Herber and Y. Maeda, J. Chem. Soc., Dalton Trans., 1999, 1001. 18 S. Mørup, Mössbauer E.ect Methodology, eds. I. J. Gruverman C. W. Seidel and D. K. Dieterly, Plenum, New York, 1974, vol. 9, p. 127; Paramagnetic and Superparamagnetic Relaxation Phenomena Studied by Mössbauer Spectroscopy, Polyteknisk Forlag, Lyngby, 1981. 19 S. Mørup and N. Thrane, Chem. Phys. Lett., 1973, 21, 363; S. Mørup, J. Phys. Chem. Solids, 1974, 35, 1159; S. Mørup, F. Sontheimer, G. Ritter and R. Zimmermann, J. Phys. Chem. Solids, 1978, 39, 123. 20 C. Belle, I. Gautier-Luneau, J.-L. Pierre, C. Scheer and E. Saint- Aman, Inorg. Chem., 1996, 35, 3706. Paper 9/02678I

ISSN:1477-9226    

DOI:10.1039/a902678i

出版商:RSC    

年代:1999

数据来源: RSC