DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1999, 2281–2292 2281 In situ syntheses of trans-spanned octahedral ruthenium complexes. Crystal structures of trans-[Ru(Cl)(trpy){Ph2PC6H4CH2O(CO)- (CH2)4(CO)OCH2C6H4PPh2}][PF6]?0.25C6H5Me?0.5CH2Cl2 and trans-[Ru(Cl)(trpy)(PPh3)2][BF4]?CH2Cl2 † Willie J. Perez,a Charles H. Lake,a Ronald F. See,a Laurence M. Toomey,a Melvyn Rowen Churchill,a Kenneth J. Takeuchi,*a Christopher P. Radano,b Walter J. Boyko b and Carol A. Bessel *b a Department of Chemistry, Natural Sciences Complex, State University of New York at BuValo, BuValo, NY 14260, USA b Department of Chemistry, Villanova University, 800 Lancaster Ave., Villanova, PA 19085, USA Received 29th September 1998, Accepted 21st May 1999 The formation of stable, undistorted octahedral transition metal complexes which contain a trans-spanning bidentate ligand remains a synthetic challenge.The reported complexes are of the type trans-[Ru(Cl)(trpy){Ph2PC6H4CH2O- (CO)Y(CO)OCH2C6H4PPh2}][PF6] [where trpy = 2,29:69,20-terpyridine and Y = (CH2)3 = C3SPAN, 6; (CH2)4 = C4SPAN, 7; or isophthalate = ISPAN, 8] and represent the first examples of trans-spanned transition metal complexes which display little bond angle distortion from octahedral geometry and also contain a bridging linkage which is stable towards oxidation, reduction and hydrolysis.These complexes were characterized by elemental analyses, cyclic voltammetry, conductivity and UV-VIS spectroscopy.COSY, HETCOR and variable temperature (1H and 13C) NMR spectra of the complexes are consistent with a flexible spanning linkage that does not demonstrate restricted rotation about either the P–Cipso or the Ru–P bonds while the X-ray crystal structure analysis of 7 showed that the spanning linkage is positioned to one side of the meridional chloride. The formation of stable, undistorted octahedral transition metal complexes which contain a trans-spanning bidentate ligand presents a synthetic challenge in several respects.1,2 To date, trans-spanning bidentate ligands have been prepared by two distinct methods: the synthesis of a bidentate ligand and its subsequent coordination to the transition metal center (preformed ligand strategy), and the bonding of two monodentate ligands to the metal center followed by the joining of the ligands with a trans-spanning linkage (in situ ligand strategy).The preformed ligand strategy has been used eVectively by Shaw and coworkers as they prepared several series of large ring complexes of the general formula trans-[M(Cl)2{Bu2P- (CH2)nPBu2}] (M = Pd or Pt; n = 5–10),3–5 or trans-[{M(Cl)2- [Bu2P(CH2)nPBu2]}x] (M = Pd or Pt, n = 8, 9, 10 or 12; x = 1–3) 6–9 where each complex contained a flexible diphosphine ligand. Initially, Shaw and coworkers proposed that the presence of the bulky tert-butyl groups attached to the phosphorus donor atoms resulted in repulsive interactions between the substituents favoring a trans-geometry.Alcock10–12 and McAuliVe 13–17 later prepared a number of trans-spanning Rh, Ni, Pt and Pd complexes with bis(diphenylphosphino)ethers, bis(diphenylphosphino)alkanes and bis(dimethylarsino)alkanes suggesting that the length of the spanning ligand was a large contributor to obtaining the trans-geometry. This argument was in agreement with those of McAuliVe 17 and others 18 who prepared ligands with both methyl, ethyl or phenyl substituents on the pnictogen or chalcogen donor atoms. In addition to preformed bidentate ligands with flexible chains, Venanzi and coworkers 19–26 and others 27 demonstrated the use of bis(dialkyl † Supplementary data available: NMR spectra.Available from BLDSC (No. SUP 57568, 19 pp.). See Instructions for Authors, 1999, Issue 1 (http://www.rsc.org/dalton). or diaryl phosphinomethyl)benzophenanthrene ligands as rigid trans-spanning “spacers.” These rigid ligands have been used to span distorted trigonal, pseudo-tetrahedral, square planar, square pyramidal and octahedral metal centers.While the preformed ligand strategy has been successful, it is not without limitations. First, the successful trans-positioning of preformed spanning ligands with flexible backbones can be crucially dependent on the choice of starting material. For example, in the preparation of PtCl2[Ph2As(CH2)nAsPh2] (n = 6–12, 16), the use of the starting material K2PtCl4 leads to the formation of the cis isomer, the use of K[PtCl3(H2C]] CH2)] leads to the formation of the trans isomer, and the use of Pt(C6H5- CN)2Cl2 leads to the formation of a cis–trans mixture.17c This marked dependency on the starting material makes the rational and systematic design of trans-spanned complexes diYcult.Second, the use of preformed ligands can result in dimerization when the preformed ligands can bridge between two metal centers.16,17d,18,23,28 This behavior was observed for [Pt{Ph2As- (CH2)nAsPh2}Cl2] (n = 6–12, 16) where both trans monomers and cis dimers were formed,17d as well as for [PtCl2{Ph2P(CH2)n- PPh2}] complexes where both cis and trans monomers and cis and trans dimers were formed.16 A third limitation is that preformed ligands with flexible backbones can form cyclometallated complexes 28–37 such as [IrH(Cl)(Bu2PCH2CH2CHCHRCH2PBu2)], where the C forms an iridium–carbon bond.29 Such cyclometallation products are often observed with square planar or trigonal bipyramidal geometries.A fourth limitation is that the preformed rigid spanning ligands can cause strain which produces transition metal complex geometries distorted from octahedral geometry. The strain caused by the bis- (diphenylphosphinomethyl)benzo[c]phenanthrene ligand (SL1) is evident in complexes of the type [M(SL1)]X or [M(SL1)X] where Venanzi reported that the P–M–P angle ranges from 1328 for Cu to 1418 for Ag, to 1768 for Au.202282 J.Chem. Soc., Dalton Trans., 1999, 2281–2292 The second method for formation of complexes with a transspanning bidentate ligand is the in situ ligand strategy. Using this strategy, Takeuchi and coworkers synthesized transspanning octahedral complexes of the form [Ru(Cl)(trpy)- (SL2)]1 [trpy = 2,29:69,20-terpyridine and SL2 = Ph2PC6H4- CH2N(Me)(CH2)nN(Me)CH2C6H4PPh2 or Ph2PC6H4CH2N- (Me)2(CH2)nN(Me)2CH2C6H4PPh2; n = 5 or 6] by linking two coordinated trans-positioned tertiary phosphine ligands with a diamine.38 These complexes were the first reported cases of in situ generated, trans-spanning ligands on an octahedral metal center where the spanning ligand bridged over the cis coordinated meridional chloride ligand.Like the preformed ligand strategy, the in situ ligand strategy oVered flexibility in terms of both backbone length and composition. In addition, with the in situ ligand strategy the major reaction product was the transspanned complex and no cyclometallated complex was found.Disadvantages in the in situ strategy resulted from the use of the amine linkages. In the trans-spanned complex, the tertiary amine linkages were readily oxidized and evidence for dealkylation of the quaternary amine linkages was observed. To overcome these disadvantages, we developed a new ligand system that retained the benefits of the in situ ligand strategy while producing a ligand that is stable to oxidation and reduction.In this work we report the formation of [Ru(Cl)(trpy)(L)]1 [where L = Ph2PC6H4CH2O(CO)Y(CO)OC6H4PPh2 and Y = (CH2)3 = C3SPAN, 6; (CH2)4 = C4SPAN, 7; and isophthalate = ISPAN, 8]. The use of the in situ strategy with an ester linkage retains the span versatility in both length and structure, while making the span stable against degradation by oxidation and reduction. Additionally, decomposition due to hydrolysis is not observed on exposure of the complexes to either mildly acidic or basic solutions (pH = 2 to 10).Finally, the new transspanning complexes display good solubility in organic solvents, which facilitated the collection of variable temperature NMR spectra and the formation of single crystals suitable for X-ray diVraction studies. Experimental Materials RuCl3?nH2O was obtained on loan or purchased from Johnson Matthey/Alfa/Aesar. 2,29:69,20-Terpyridine was purchased from G. F. Smith Chemical Company or was synthesized by literature methods.39 Triphenylphosphine was purchased from Aldrich Chemical Company or Strem Chemical. Methylene chloride (J.T. Baker) was dried over activated 5 Å molecular sieves and distilled under N2.40 All other solvents and materials were of reagent quality and were used as received. Reactions were conducted under a nitrogen atmosphere unless otherwise noted. Measurements Elemental analyses were performed by Atlantic Microlabs, Norcross, GA. UV-VIS spectra were recorded using a Bausch and Lomb Spectronic 2000, a Milton Roy Spectronic 3000 Diode Array Spectrophotometer, or a Cary 1G UV-VIS Spectrophotometer.Conductivity measurements were performed in acetonitrile using a YSI Model 31 conductivity bridge. Electrochemical measurements were made versus a saturated sodium chloride calomel reference electrode (SSCE) using either an IBM EC/225 voltammetric analyzer, a PAR Model 173 Potentiostat/Galvanostat equipped with a PAR Model 175 Universal Programmer or a BAS CV-50W Voltammetric Analyzer.A platinum disc working electrode was used along with a platinum wire common electrode. Electrochemical measurements used 0.1 M tetrabutylammonium tetrafluoroborate (TBAB) as the electrolyte and were conducted with ferrocene (E1/2 = 10.40 V vs. SSCE in CH3CN, E1/2 = 10.50 V vs. SSCE in CH2Cl2) as the internal standard. All NMR spectra were obtained on a Varian XL-300 spectrometer in CD2Cl2. 1H spectra were obtained at 299.9 MHz and referenced to tetramethylsilane. 13C spectra were obtained at 75.4 MHz and referenced to CD2Cl2 (d 53.8). Proton–proton COSY and carbon–hydrogen HETCOR were run with standard Varian-supplied pulse sequences. The one-bond HETCOR direct detection sequence utilized BIRD pulses 41 to suppress proton–proton couplings in the f1 domain of the 2D maps. Quaternary carbon resonances in trpy were assigned using the Varian-supplied direct detection HETCOR sequence (no BIRD pulse) optimizing defocussing/refocussing delays for the appropriate nJC–H constants.Several sets of delays were used near each desired J-value in order to avoid loss of the appropriate correlation signals due to one-bond modulation of the longrange response intensity.42 Crystallography Data collection. Crystals were aligned on a Siemensupgraded Syntex P21/R3 diVractometer equipped with a highlyoriented graphite crystal monochromator. The determination of the Laue symmetry, crystal class, unit-cell parameters and the crystal orientation matrix were carried out by previously described techniques.43 Room-temperature data were collected with Mo-Ka radiation (l = 0.71073 Å), using the q–2q scan technique for 7, and the w scan technique for 5 where peak overlap was a possible problem.Details of the data collection are in Table 1. All reflections in each data set were corrected for Lorentz and polarization eVects and for absorption (semiempirical). Solution and refinement of the structures. All crystallographic calculations were carried out on a VAX3100 workstation with the use of the Siemens SHELXTL PLUS44 program set.The analytical scattering factors for neutral atoms were corrected for both the Df 9 and the iDf 0 components of anomalous dispersion. The structures were solved by a combination of direct methods and Fourier-diVerence techniques. All non-hydrogen atoms were refined anisotropically, and hydrogen atoms were included in calculated positions with d(C–H) = 0.96 Å.45 Details of each structure solution and its refinement may be found in Table 1.A diagram of one structure was generated using ORTEP II.46 CCDC reference number 186/1479. Preparations The complexes RuCl3(trpy),47 trans-[Ru(Cl)2(trpy){Ph2PC6H4- (CH2OC5H9O)-p}], 1,38 cis-[Ru(Cl)2(trpy){Ph2PC6H4(CH2OC5H9O)- p}], 2,38 trans-[Ru(Cl)(trpy){Ph2PC6H4(CH2OC5H9O)- p}2][PF6], 3,38 trans-[Ru(Cl)(trpy){Ph2PC6H4(CH2OH)-p}2]- [PF6], 4,38 and trans-[Ru(Cl)(trpy)(PPh3)2]1, 5,47 were synthesized following published procedures. trans-[Ru(Cl)(trpy)(C3SPAN)][PF6], 6.A 0.174 g (0.158 mmol) sample of 4 was dissolved in 17 mL of CH2Cl2 and the solution was outgassed with N2 for 5 min. Glutaryl dichloride (0.027 mL, 0.21 mmol) was added and the reaction mixture was heated to reflux for 24 h. After cooling, the volume of the solution was reduced to dryness on a rotary evaporator. The residue was redissolved in CH2Cl2 and purified by passing through an alumina column using 100 : 1 (v/v) CH2Cl2–MeOH as the eluent.The first tan-orange band was collected and the solution was reduced to dryness with a rotary evaporator. The product was redissolved in a minimal amount of CH2Cl2 and precipitated by dropwise addition to Et2O (ca. 100 mL). The solid was collected by vacuum filtration, washed with a minimum amount of Et2O and air dried. A 0.111 g (0.093 mmol, 59% yield) sample of yellow-brown product was obtained (Calc.for C58H49ClF6N3O4P3Ru?2H2O: C, 56.57; H, 4.02. Found: C, 56.58; H, 4.13%).J. Chem. Soc., Dalton Trans., 1999, 2281–2292 2283 Table 1 Details of X-ray diVraction studies of trans-[Ru(Cl)(trpy)(PPh3)2][BF4]?CH2Cl2 5 and trans-[Ru(Cl)(trpy)(C4SPAN)][PF6]?0.25C6H5Me? CH2Cl2 7 5 7 Formula M Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 Z Dc/Mg m23 m(Mo-Ka)/mm21 Independent reflections Reflections > 6s(F ) Final R indices (all data): R, wR Final R indices (6s data): R, wR C51H41ClN3P2BF4Ru?CH2Cl2 1066.1 Monoclinic P21/c 17.8306(38) 12.7954(26) 21.7707(40) 90.00 104.195(15) 90.00 4815.4(1.7) 4 1.470 0.606 6329 3118 0.063, 0.038 0.049, 0.035 C59H51ClO4N3P3F6Ru?0.25C6H5Me?0.5CH2Cl2 1275 Triclinic P1� 10.9656(18) 13.9782(22) 19.8449(26) 89.212(12) 85.938(12) 68.071(12) 2814.4(7) 2 1.505 0.529 7391 4070 0.092, 0.076 0.044, 0.046 trans-[Ru(Cl)(trpy)(C4SPAN)][PF6], 7.A 0.101 g (0.0916 mmol) sample of 4 was dissolved in 7.2 mL of CH2Cl2. Adipoyl dichloride (0.020 mL; 0.14 mmol) was added and the reaction mixture was heated to reflux for 4 h.The reaction mixture was reduced to dryness with a rotary evaporator. The residue was passed through an alumina column using a 150: 1 (v/v) CH2Cl2–MeOH solution as the eluent. The first tan-orange band was collected and reduced to dryness with a rotary evaporator. This residue was redissolved in a minimum amount of CH2Cl2 and precipitated by dropwise addition to toluene. The yellow-brown product was collected by vacuum filtration, washed with Et2O and air dried; yield 0.067 g (0.055 mmol); 59% (Calc.for C59H51ClF6N3O4P3Ru?C6H5Me: C, 60.90; H, 4.56. Found: C, 60.57; H, 4.62%). trans-[Ru(Cl)(trpy)(ISPAN)][BF4], 8. A 0.100 g (0.0907 mmol) sample of 4 was dissolved in 7.0 mL of CH2Cl2. Isophthaloyl dichloride (0.022 g; 0.11 mmol) was added and the reaction mixture was heated to reflux for 4 h and then reduced to dryness with a rotary evaporator. The residue was passed through an alumina column using a 100: 1 (v/v) CH2Cl2– MeOH solution as the eluent.The first red-brown band was collected and reduced to dryness with a rotary evaporator, redissolved in a minimum amount of CH2Cl2, and precipitated by dropwise additon to Et2O. The red-brown product was collected by vacuum filtration, washed with Et2O and air dried; yield 0.051 g (0.039 mmol); 43% (Calc. for C61H47ClF4N3O4P2- BRu?0.5H2O: C, 59.16; H, 3.91. Found: C, 59.13; H, 4.05%).Results and discussion Fig. 1 shows the general scheme for the synthesis of the transspanning ruthenium complex, trans-[Ru(Cl)(trpy)(C3SPAN)]1, 6. The synthesis is initiated by combining RuCl3(trpy) with one protected phosphine ligand and a reducing agent to form the trans-[Ru(Cl)2(trpy){Ph2PC6H4(CH2OC5H9O)-p}], 1. Irradiation of the reaction mixture with light 47 converts 1 to cis- [Ru(Cl)2(trpy){Ph2PC6H4(CH2OC5H9O)-p}], 2. The combination of excess phosphine with 2 produces trans-[Ru(Cl)(trpy)- {Ph2PC6H4(CH2OC5H9O)-p}2][PF6], 3.Prior to the transspanning reaction, the phosphine groups are deprotected in an acid catalyzed step resulting in the formation of trans- [Ru(Cl)(trpy){Ph2PC6H4(CH2OH)-p}2][PF6], 4. trans-Spanning is achieved by reacting 4 with an organic dioxychlide to give 6, 7 or 8 in a double esterification reaction. The product is purified by column chromatography to isolate the monomeric species with yields of 40–60%. The use of a large volume of solvent during the spanning procedure prevents dimerization and/or oligimerization and increases the yields of the spanned complexes to values similar to those of the model reactions (i.e.the formation of dibenzyl adipate and dibenzyl glutarate).48 Through the use of three diVerent spans, –(CH2)3–, –(CH2)4– and –isophthalate–, we demonstrate that the trans-spanning linkage can be modified in terms of chain length and structure. Electronic spectroscopy and cyclic voltammetric data The UV-VIS spectroscopic and cyclic voltammetry data for the complexes are summarized in Table 2.Transitions at ca. 700, 550 and 400 nm in the trans-[Ru(Cl)2(trpy)(PR3)] complexes are assigned to Ru(dp)Æp*(trpy) metal-to-ligand charge transfer bands (MLCT) as observed for other trans-ruthenium complexes.47,49–52 The transitions observed at 380, 330, 320, 286 and 275 nm are assigned to pÆp* ligand (trpy, triphenylphosphine)- localized transitions.47,53,54 These trans-(dichloro)- ruthenium complexes display one reversible couple assigned to the ruthenium(III/II) potential at approximately 0.50 V vs.SSCE in CH2Cl2. The cis-[Ru(Cl)2(trpy)(PR3)] complexes are characterized by two MLCT transitions at ca. 530 and 490 nm. These wavelengths have shifted to shorter wavelengths relative to the trans- [Ru(Cl)2(trpy)(PR3)] complexes, however the absorbances maintain similar molar absorptivity values. The four ligand localized pÆp* transitions occur at ca. 360, 320, 285 and 275 nm. These cis-(dichloro)ruthenium complexes display one reversible couple assigned to the ruthenium(III/II) potential at approximately 0.60 V vs. SSCE in CH2Cl2. The shifts to shorter wavelengths which accompany the increases in E1/2 values are observed for other polypyridyl ruthenium complexes.47,49–52,55,56 Both 1 and 2 have E1/2 values which are 50 mV higher than the corresponding PPh3 complexes. This increase in E1/2 is consistent with the electron withdrawing nature of the protected group on the phosphine.The addition of a second phosphine ligand to 2 and the consequent change from a neutral to a positively charged molecule result in a shift of the absorption maxima to higher energies. These absorbances are again assigned to MLCT bands from Ru (dp)Æligand (p*) transitions. The absorption maxima of the complexes at 330, 310, 270, 230, and 210 nm are assigned to ligand localized pÆp* transitions. Interestingly, the reversible ruthenium(III/II) redox couples for complexes 3–8 demonstrate a small range in E1/2 values from 10.88 to 10.93 V vs.SSCE in CH3CN and a linear relationship between the peak current (ip,c) and the square root of the scan rate (n1/2). This linear relationship indicates that the electron transfer is diVusion controlled2284 J. Chem. Soc., Dalton Trans., 1999, 2281–2292 Table 2 UV-VIS spectroscopic and cyclic voltammetry data for selected ruthenium complexes Complex a E2� 1 /V vs.SSCE lmax/nm (1023e/cm21 M21) trans-[Ru(Cl)2(trpy)(PPh3)] 10.46 e 705 (sh), 549(4.66), 403(4.71), 375 (sh), 331(17.9), 320 (sh), 286 (sh), 275(16.5) trans-[Ru(Cl)2(trpy)(Ph2PR)], 1 b 10.51 701 (sh), 552(4.84), 403(4.86), 381 (sh), 331(20.5), 320 (sh), 286 (sh), 275(21.0) cis-[Ru(Cl)2(trpy)(PPh3)] 10.58 e 531(4.79), 488 (sh), 363 (sh), 319(20.8), 286 (sh), 275(16.3) cis-[Ru(Cl)2(trpy)(Ph2PR)], 2 b 10.63 533(5.35), 488 (sh), 362 (sh), 319(22.9), 285 (sh), 276(17.9) trans-[Ru(Cl)(trpy)(Ph2PR)2][PF6], 3 c 10.88 474(3.83), 431 (sh), 334 (sh), 310(29.8), 270(78.6), 236(62.0), 209(96.6) trans-[Ru(Cl)(trpy)(Ph2PR9)2][PF6], 4 c,d 10.88 473(3.35), 431 (sh), 332 (sh), 311(22.2), 270(45.8), 231 (sh), 208(98.6) trans-[Ru(Cl)(trpy)(PPh3)2][PF6], 5 c 10.90 473(3.62), 431 (sh), 330 (sh), 312(23.2), 268(43.2) trans-[Ru(Cl)(trpy)(C3SPAN)][PF6], 6 c 10.92 473(3.50), 432 (sh), 334 (sh), 311(21.7), 271(38.5), 232(52.1) trans-[Ru(Cl)(trpy)(C4SPAN)][PF6], 7 c 10.91 473(3.84), 432 (sh), 334 (sh), 311(21.4), 271(39.9), 231(51.0) trans-[Ru(Cl)(trpy)(ISPAN)][PF6], 8 c 10.93 471(4.10), 432 (sh), 333 (sh), 311(25.0), 271(42.0), 233(75.6) a Measured in methylene chloride unless otherwise stated.b R = C6H4(CH2OC5H9O)-p. c Measured in acetonitrile. d R9 = C6H4(CH2OH)-p. e From ref. 47. Fig. 1 The reaction scheme for the preparation of trans-[Ru(Cl)(trpy)(C3SPAN)]1, 6. and is not influenced by the steric bulk of the spanning linkage (for complexes 6–8).Finally, peak current ratios [cathodic peak current (ip,c)/anodic peak current (ip,a)], determined using the Nicholson Method,57 ranged from 0.90–1.0 : 1 for all of the complexes used in this study. These data imply that electron transfer at the ruthenium metal center is reversible and that the ester linkages found in complexes 6–8 are stable to oxidation and reduction under standard electrochemical conditions. Thus, the ester linkage overcomes the disadvantages observed with the Takeuchi SL2 ligand (see above).While changes in the electron donating or electron withdrawing nature of ligand substituents can result in variation of electronic properties at the metal center previous observations have also shown that the redox potential of a metal complex can change with steric ligand eVects.58–60 An advantage of the in situ ligand synthesis procedure is the ability to compare the span precursors to the spanned complexes.This comparison has enabled us to separate the electronic contributions from the steric contributions of the spanning linkage. This is most notable in light of studies involving the Venanzi preformed ligand SL1 complexes (see above),19–26 where isolating the electronic eVects of the spanning ligand from the steric eVects caused by the strained geometries proved diYcult.21,24 The consistency of the spectroscopic and electrochemical data for the protected (3), deprotected (4), unsubstituted (5) and spanned (6–8) complexes indicates that changes in the periphery of theJ.Chem. Soc., Dalton Trans., 1999, 2281–2292 2285 ligand structure leave the electronic environment about the metal center relatively unchanged and implies that the geometry about the metal center is relatively undistorted from an ideal octahedral arrangement (this was later proven by X-ray crystallography, see below). Conductivity analysis Conductivity measurements were performed on 6 and 7 using the method of Feltham and Hayter.61 The charge and nuclearity of the complexes were determined by measuring the equivalent conductivity (Le) over a range of concentrations (expressed in terms of equivalent concentration, c).The data are obtained as Lo, the conductance at infinite dilution and B, the slope of the plot of (Lo 2 Le) versus the square root of c. The use of equivalent concentrations in this technique eliminates the uncertainty in concentrations of complexes with unknown nuclearity and it allows for the diVerentiation of complexes having the same empirical formula but diVerent molecular weight.The B values were 297 and 373 for 6 and 7 respectively, and are consistent with the values proposed by Davies for monomeric complexes with a 1 : 1 electrolyte in acetonitrile,62 and by Leising for monomeric trans-diphosphine ruthenium(II) complexes and for ruthenium(II) complexes containing the Takeuchi SL2 spanning linkages.38,50 Table 3 Selected interatomic distances (Å) and angles (8) for trans- [Ru(Cl)(trpy)(PPh3)2][BF4]?CH2Cl2, 5 (A) Ruthenium–ligand distances Ru(1)–P(1) Ru(1)–P(2) Ru(1)–Cl(1) 2.398(2) 2.415(2) 2.457(2) Ru(1)–N(1) Ru(1)–N(2) Ru(1)–N(3) 2.098(6) 1.964(7) 2.072(6) (B) Phosphorus–carbon distances P(1)–C(11) P(1)–C(21) P(1)–C(31) 1.841(8) 1.831(8) 1.847(8) P(2)–C(41) P(2)–C(51) P(2)–C(61) 1.831(7) 1.813(9) 1.824(9) (C) Distances within terpyridyl systems N(1)–C(71) C(71)–C(72) C(72)–C(73) C(73)–C(74) C(74)–C(75) C(75)–N(1) C(75)–C(81) N(2)–C(81)(81)–C(82) C(82)–C(83) 1.363(12) 1.376(13) 1.351(18) 1.380(16) 1.377(13) 1.380(12) 1.473(13) 1.351(12) 1.386(14) 1.365(16) N(3)–C(91) C(91)–C(92) C(92)–C(93) C(93)–C(94) C(94)–C(95) C(95)–N(3) C(85)–C(95) C(83)–C(84) C(84)–C(85) C(85)–N(2) 1.350(10) 1.377(12) 1.397(15) 1.356(14) 1.383(13) 1.357(11) 1.476(12) 1.390(16) 1.376(12) 1.357(11) (D) Angles around the ruthenium atom P(1)–Ru(1)–P(2) P(2)–Ru(1)–Cl(1) P(2)–Ru(1)–N(1) P(1)–Ru(1)–N(2) Cl(1)–Ru(1)–N(2) P(1)–Ru(1)–N(3) Cl(1)–Ru(1)–N(3) N(2)–Ru(1)–N(3) 178.1(1) 90.7(1) 90.4(2) 90.3(2) 174.9(2) 90.1(2) 95.9(2) 79.5(3) P(1)–Ru(1)–Cl(1) P(1)–Ru(1)–N(1) Cl(1)–Ru(1)–N(1) P(2)–Ru(1)–N(2) N(1)–Ru(1)–N(2) P(2)–Ru(1)–N(3) N(1)–Ru(1)–N(3) 87.5(1) 90.8(2) 106.2(2) 91.4(2) 78.5(3) 89.4(2) 157.9(3) (E) Angles around phosphorus atoms Ru(1)–P(1)–C(11) Ru(1)–P(1)–C(21) Ru(1)–P(1)–C(31) Ru(1)–P(2)–C(41) Ru(1)–P(2)–C(51) Ru(1)–P(2)–C(61) 114.0(2) 121.1(3) 111.8(2) 118.2(2) 108.9(2) 119.6(3) C(11)–P(1)–C(21) C(11)–P(1)–C(31) C(21)–P(1)–C(31) C(41)–P(2)–C(51) C(41)–P(2)–C(61) C(51)–P(2)–C(61) 101.0(4) 108.4(4) 98.9(4) 105.9(4) 99.0(4) 103.6(4) (F) Angles at Cipso in PPh3 ligands C(16)–C(11)–C(12) C(26)–C(21)–C(22) C(36)–C(31)–C(32) 117.4(8) 117.8(7) 117.2(8) C(46)–C(41)–C(42) C(56)–C(51)–C(52) C(66)–C(61)–C(62) 118.5(7) 118.3(8) 117.7(8) Crystal structure analysis Crystals of 5 were grown using the double vial diVusion technique in a CH2Cl2–toluene solution and crystallize with CH2Cl2 molecules in the lattice.Crystallographic-grade crystals of 7 were grown, after purification by column chromatography, from a solution of 12 mg complex–0.5 ml chloroform at approximately 18 8C. Selected bond distances and bond angles for 5 and 7 are listed in Tables 3 and 4 respectively. Crystals of 7 contain a disordered array of C6H5Me and CH2Cl2 molecules of crystallization about the inversion center at 1/2 1/2 0.Fig. 2 and 3 give perspective views and numbering schemes for 5 and 7 respectively. The crystal structures of 5 and 7 are compared to determine whether distortions are brought about by the presence of the trans-spanning linkage. The crystal structures of 5 and 7 consist of arrays of ordered ruthenium cations and anions (BF4 2 for 5 and PF6 2 for 7) in a 1 : 1 stoichiometry along with solvent of crystallization. Results of the structural analyses indicate that 5 adopts a C2v geometry, while 7 adopts a C1 geometry due to the position of the spanning ligand.One of the most significant aspects of the crystal structure of 7 is that the complex is monomeric, confirming the conductivity studies. Both complexes show small distortions from an ideal octahedral geometry. First, the N(trpy terminal pyridine)–Ru– N(trpy terminal pyridine) bond angle for 5 is 157.9(3)8 and for 7 is 157.7(2)8. These deviations are attributed to the geometrical constraints of the trpy backbone.63 These angles diVer greatly from those of the free trpy ligand at 1288,64 however, they are within the range [156.9(5)–158.3(3)8] observed for other ruthenium( II) complexes.50–52,65,66 The Ru–N(central trpy) bond length is 1.964(7) and 1.955(5) Å for 5 and 7 respectively; both structures show some dissymmetry in the Ru–N(trpy terminal pyridines) bond distances.Comparisons between complexes containing two triphenylphosphine ligands and other trans-spanning ligands have been made in the literature.For example, the crystal structures of Fig. 2 ORTEP diagram of the trans-[Ru(Cl)(trpy)(PPh3)2]1 cation, 5.2286 J. Chem. Soc., Dalton Trans., 1999, 2281–2292 Au(SL1)Cl 20,67 and Au(PPh3)2Cl 68 have significantly diVerent P–Au–P bond angles [140.7(1)8 and 132.1(1)8 respectively]; as do CuCl(SL1)67 and CuBr(PPh3)2 69 [at 131.98 and 126.0(1)8, respectively]. These data imply that the rigid spacer ligand, SL1, may be better at enforcing linear trans-geometries than two monodentate PPh3 ligands.However, the latter diVerences must be interpreted with caution, especially in light of the significant changes in P–Ag–P bond angle that occur with change in anion.20,24 The P–Ru–P angles in both 5 and 7 deviate slightly from linearity (1808). Complex 5 has a P–Ru–P bond angle of 178.1(1)8 while 7 has a P–Ru–P bond angle of 175.0(1)8. Although the P–Ru–P bond angle in 7 is smaller than in 5, and could indicate strain on the octahedral geometry due to the spanning ligand, an investigation of trans-diphosphine- (terpyridyl)ruthenium(II) complexes which contain PMe3,50 PEt3,51 PPr3 66 or PPh3 65 ligands shows that P–Ru–P bond angles normally range between 175.1(1) and 178.2(2)8.Thus, the P–Ru–P bond angle observed in 7 is probably not caused by the spanning linkage. The P–Ru–P bond axis for both 5 and 7 appears to bend in the direction of the chloride ligand, implying that this region has less steric crowding than the region around the trpy ligand. The diagrams of 5 and 7 show that the phosphine ligands align in a similar fashion over and under the trpy ligand.That is, for both structures one phenyl ring of the phosphine is parallel to the central pyridine ring of trpy; a second phenyl ring is orthogonal to the plane of the trpy ligand; and a third phenyl group of the phosphine is tilted from the terminal pyridine of the trpy ligand. The Ru–P bond distances in both complexes fall within the expected range found for other phosphine ruthenium complexes, which is 2.26–2.41 Å.50–52,64–66 Such bond lengths are also similar to those observed with other metals and Table 4 Interatomic distances (Å) and angles (8) for trans-[Ru(Cl)(trpy)(C4SPAN)][PF6]?0.25C6H5Me?0.5CH2Cl2 7 (A) Ruthenium–ligand distances Ru(1)–P(1) Ru(1)–P(2) Ru(1)–Cl(1) 2.397(2) 2.397(2) 2.481(2) Ru(1)–N(71) Ru(1)–N(81) Ru(1)–N(91) 2.073(5) 1.955(5) 2.126(6) (B) Phosphorus–carbon distances P(1)–C(11) P(1)–C(21) P(1)–C(31) 1.836(8) 1.831(7) 1.825(6) P(2)–C(41) P(2)–C(51) P(2)–C(61) 1.852(6) 1.839(6) 1.817(9) (C) Distances within PPh2(C6H4) moieties C(11)–C(12) C(12)–C(13) C(14)–C(15) C(21)–C(22) C(22)–C(23) C(24)–C(25) C(31)–C(32) C(32)–C(33) C(34)–C(35) C(41)–C(42) C(42)–C(43) C(44)–C(45) C(51)–C(52) C(52)–C(53) C(54)–C(55) C(61)–C(62) C(62)–C(63) C(64)–C(65) 1.380(11) 1.385(12) 1.371(12) 1.392(8) 1.379(12) 1.381(11) 1.379(9) 1.386(10) 1.390(11) 1.371(9) 1.387(10) 1.373(10) 1.364(9) 1.364(10) 1.368(11) 1.387(9) 1.398(13) 1.371(12) C(11)–C(16) C(13)–C(14) C(15)–C(16) C(21)–C(26) C(23)–C(24) C(25)–C(26) C(31)–C(36) C(33)–C(34) C(35)–C(36) C(41)–C(46) C(43)–C(44) C(45)–C(46) C(51)–C(56) C(53)–C(54) C(55)–C(56) C(61)–C(66) C(63)–C(64) C(65)–C(66) 1.395(9) 1.375(9) 1.384(14) 1.387(10) 1.356(12) 1.388(13) 1.392(11) 1.361(13) 1.376(10) 1.397(12) 1.375(12) 1.384(9) 1.373(10) 1.362(13) 1.385(10) 1.420(11) 1.372(14) 1.379(16) (D) Distances within the trans-spanning bridge C(1)–O(2) O(2)–C(3) C(4)–C(5) C(6)–C(7) C(8)–O(9) C(10)–C(44) 1.451(9) 1.324(10) 1.524(12) 1.511(12) 1.325(10) 1.501(10) C(1)–C(14) C(3)–C(4) C(5)–C(6) C(7)–C(8) O(9)–C(10) 1.507(12) 1.512(10) 1.481(14) 1.512(14) 1.436(11) spanning linkages, for example [Ir(Cl)(CO)(SL1)] 22 has an Ir–P bond length of 2.310(4) Å and [Ir(Cl)3(CO)(SL1)] 22 has an average Ir–P bond length of 2.411 Å.The Cl–Ru–N(central pyridine) bond angles are also less than the ideal 1808 for both complexes: 174.9(2)8 for 5 and 174.7(2)8 for 7.The phenyl groups of the phosphines that are oriented orthogonal to the trpy plane appear to push the chloride ligand in the opposite direction and expand the Cl–Ru– N(trpy terminal pyridine) bond angle to 106.2(2)8 in 5 and 106.8(2)8 for 7. These angles are 10.38 and 11.38 greater than the second Cl–Ru–N(trpy terminal pyridine), i.e. where the phenyl group of the phosphine is approximately parallel to the terminal pyridine ring.This variation in bond angle is significant in light of other ligand (nitro or aqua)–Ru–N(trpy terminal pyridine) angles which vary between 99.6 and 102.58.50,51,65,66 The Ru–N(trpy terminal pyridine) bonds are diVerent by 0.026 Å in 5 and by 0.053 Å for 7. Also, the Ru–Cl bond distance in 7 is 0.024 Å longer than the Ru–Cl bond distance in 5. NMR spectroscopy 1H and 13C NMR spectroscopies were used to confirm the structures of the low spin, d6, ruthenium(II) complexes and to investigate the eVects of the trans-spanning linkages.The chemical shifts and coupling constants for all of the compounds are given in Table 5 (1H NMR) and Table 6 (13C NMR). Analysis of the free trpy spectra. In 1971, Carlson et al. assigned the 1H NMR spectrum of free trpy by drawing an analogy between trpy and 2,29-bipyridine (bpy).70 Assignments for bpy had previously been made by Castellano et al.71 Our analysis of the 1H NMR spectrum of the free trpy ligand diVers from that reported by Carlson et al.70 in our assignment of the (E) Distances within carbonyl groups C(3)–O(3) 1.204(9) C(8)–O(8) 1.188(9) (F) Angles around the ruthenium atom P(1)–Ru(1)–P(2) P(2)–Ru(1)–Cl(1) P(2)–Ru(1)–N(71) P(1)–Ru(1)–N(81) Cl(1)–Ru(1)–N(81) P(1)–Ru(1)–N(91) Cl(1)–Ru(1)–N(91) N(81)–Ru(1)–N(91) 175.0(1) 87.9(1) 91.2(2) 93.6(2) 174.7(2) 92.7(2) 106.8(2) 78.4(2) P(1)–Ru(1)–Cl(1) P(1)–Ru(1)–N(71) Cl(1)–Ru(1)–N(71) P(2)–Ru(1)–N(81) N(71)–Ru(1)–N(81) P(2)–Ru(1)–N(91) N(71)–Ru(1)–N(91) 87.1(1) 88.5(2) 95.5(2) 91.3(2) 79.3(2) 89.5(2) 157.7(2) (G) Angles around phosphorus atoms Ru(1)–P(1)–C(11) Ru(1)–P(1)–C(21) Ru(1)–P(1)–C(31) Ru(1)–P(2)–C(41) Ru(1)–P(2)–C(51) Ru(1)–P(2)–C(61) 114.4(2) 110.6(3) 119.7(2) 112.9(2) 120.7(2) 111.5(2) C(11)–P(1)–C(21) C(11)–P(1)–C(31) C(21)–P(1)–C(31) C(41)–P(2)–C(51) C(41)–P(2)–C(61) C(51)–P(2)–C(61) 108.0(3) 101.4(3) 101.5(3) 99.7(3) 107.9(3) 102.7(3) (H) Angles at Cipso in PPh3 ligands C(16)–C(11)–C(12) C(26)–C(21)–C(22) C(36)–C(31)–C(32) 116.6(7) 117.9(7) 117.5(6) C(46)–C(41)–C(42) C(56)–C(51)–C(52) C(66)–C(61)–C(62) 118.7(6) 118.2(6) 117.8(8) (I) Angles within trans-spanning bridge O(2)–C(1)–C(14) C(1)–O(2)–C(3) O(2)–C(3)–C(4) C(3)–C(4)–C(5) C(5)–C(6)–C(7) 110.7(8) 116.8(6) 110.0(6) 115.6(6) 114.3(8) O(9)–C(10)–C(44) C(7)–C(8)–O(9) C(8)–O(9)–C(10) C(4)–C(5)–C(6) C(6)–C(7)–C(8) 106.6(6) 111.6(7) 117.0(6) 115.3(8) 116.5(8) (J) Angles about carbonyl groups O(2)–C(3)–O(3) C(7)–C(8)–O(8) 125.8(7) 125.3(8) O(3)–C(3)–C(4) O(8)–C(8)–C(9) 124.2(8) 123.1(8)J.Chem. Soc., Dalton Trans., 1999, 2281–2292 2287 Table 5 1H NMR spectroscopy for ruthenium complexes Complex a d (ppm) (integration, multiplicity,b coupling/Hz, assignment c) trpy 7.35 (2 H, ddd, Jba = 4.8, Jbc = 7.5, Jbd = 1.2, b), 7.88 (2 H, ddd, Jcd = 8.0, Jcb = 7.5, Jca = 1.8, c), 7.96 (1 H, t, Jhg = 8.0, h), 8.47 (2 H, d, Jgh = 8.0, g), 8.63 (2 H, dt, Jdc = 8.0, d ), 8.69 (2 H, ddd, Jab = 4.8, Jdb = 1.2, Jda = 1.0, a) trans-[Ru(Cl)(trpy)(PPh3)2][PF6], 5 7.08 (12 H, cm, k), 7.11 d (2 H, —e, Jba = 5.4, Jbc = 7.4, Jbd e, b), 7.14 d (12 H, cm, j), 7.24 (6 H, cm, l ), 7.47 (3 H, s, Jgh e, g and h), 7.69 (2 H, td, Jca = 1.3, Jcb = 7.4, Jcd = 8.2, c), 7.75 (2 H, dd, Jdc = 8.2, Jdb = 1.6, Jda e, d ), 9.01 (2 H, d, Jab = 5.4, Jac = 1.3, Jad e, a) trans-[Ru(Cl)(trpy)(C3SPAN)][PF6], 6 2.09 (2 H, p, t), 2.71 (4 H, t, s), 4.90 (4 H, s, q), 6.17 (4 H, bs, n), 6.77 (4 H, bd, Jon = 7.7, o), 7.10 (2 H, bt, Jba = 5.8, Jbc = 7.6, Jbd e, b), 7.22 (8 H, cm, k), 7.29 f (2 H,e, Jgh = 8.0, g), 7.30 (4 H, cm, l ), 7.49 (1 H, t, Jhg = 8.0, h), 7.58 (2 H, d, Jda e, Jdb e, Jdc = 8.1, d), 7.67 (10 H, td, Jca = 1.2, Jcb = 7.6, Jcd = 8.1, c and j), 8.99 (2 H, bs, Jab = 5.8, Jac = 1.2, Jad e, a) trans-[Ru(Cl)(trpy)(C4SPAN)][PF6], 7 1.81 (4 H, cm, t), 2.56 (4 H, cm, s), 4.88 (4 H, s, q), 6.24 (4 H, cm, n), 6.80 (4 H, bd, o), 7.11 (2 H, ddd, Jba = 5.6, Jbc = 7.3, Jbd = 1.7, b), 7.21 (8 H, cm, k), 7.30 (4 H, cm, l), 7.35 (2 H, d, Jgh = 7.3, g), 7.44 (1 H, cm, Jhg = 7.3, h), 7.60 (2 H, d, Jda e, Jdb = 1.7, Jdc = 8.2, d), 7.60 (8 H, cm, j), 7.67 (2 H, td, Jca = 1.3, Jcb = 7.3, Jcd = 8.2, c), 9.00 (2 H, bd, Jab = 5.6, Jac = 1.3, Jad e, a) trans-[Ru(Cl)(trpy)(ISPAN)][PF6], 8 4.99 (4 H, s, q), 5.83 (4 H, p, n), 6.66 (4 H, bd, o), 7.10 (2 H, ddd, Jba = 5.3, Jbc = 6.7, Jbd = 2.2, b), 7.22 d (8 H, cm, k), 7.23 d (4 H, cm, l), 7.49 (2 H, dd, Jda = 1.7, Jdb = 2.2, Jdc = 8.5, d), 7.44 (3 H, bs, g and h), 7.54 (2 H, cm, Jca = 1.2, Jcb = 6.7, Jcd = 8.5, c), 7.66 (1 H, t, Jut = 7.9, u), 7.92 (8 H, cm, j), 8.37 (2 H, dd, Jtu = 7.9, Jtv = 1.6, t), 8.50 (1 H, bt, Jvt = 1.6, v), 9.20 (2 H, bd, Jab = 5.3, Jac = 1.2, Jad = 1.7, a).a NMR spectra were measured in CD2Cl2 and referenced against CDHCl2. b Abbreviations: s = singlet, d = doublet, t = triplet, p = pentuplet, bs = broad singlet, bd = broad doublet, bt = broad triplet, dd = doublet of doublets, dt = doublet of triplets, td = triplet of doublets, ddd = doublet of doublet of doublets, cm = complex multiplet. c Letters used in assignments correlate to trpy Fig. 4; 5 Fig. 5; 6 Fig. 6; 7 Fig. 7; 8 Fig. 8. d Shift determined from corrected HETCOR data. e Could not determine due to overlap or broad linewidth. f Shift determined by COSY. chemical shifts for the Hd and Hg protons (see Fig. 4). The terminal pyridine rings in trpy are magnetically equivalent and have the first order sub-spectrum of a 2-substituted pyridine while the middle ring has the first order sub-spectrum of a 2,6- disubstituted pyridine.Free trpy demonstrates a trans,transconfiguration both in solution 63 and in the solid state 64 due to the repulsion of the non-bonding electrons on the nitrogen atoms. This has consequences for the proton chemical shifts. Protons Ha, Hb, Hc and Hh in free trpy have shifts similar to the equivalent positions in pyridine.However, Hd and Hg are shifted over 1 ppm downfield from their position in pyridine due to their interaction with the non-bonding electron pairs on the adjacent pyridine rings. The two-dimensional COSY spectrum (see SUP 57568, Fig. 9) of free trpy is consistent with our current assignments. Notably, Jab is about half the size of Jcd for both free trpy and for trpy contained in complexes 5–8. When this information is used in conjunction with the carbon shifts determined by HETCOR spectroscopy, confirmation of the proton assignments in the ruthenium complexes is possible. The proton decoupled 13C NMR resonances for free trpy are assigned (Fig. 4) based on one-bond HETCOR experiments (SUP 57568, Fig. 10) except in the case of Ce and Cf which are Fig. 3 Molecular geometry of the trans-[Ru(Cl)(C4SPAN)(trpy)]1 cation, 7. Hydrogen atoms are omitted for clarity. assigned based on N-bond HETCOR experiments. The 13C NMR resonances for free trpy are within 0.3 to 2.5 ppm of the literature values for the resonances of free pyridine (Cortho to N d 149.8, Cmeta to N d 123.6 and Cpara to N d 135.7) 72 except for Ce and Cf which are shifted by 6.7 and 5.9 ppm downfield, respectively, due to substituent eVects.Analysis of triphenylphosphine spectra. The 1H NMR spectum of free triphenylphosphine (PPh3) shows one broad singlet in the aromatic region, d 7.28 in CDCl3.73 The 13C NMR spectrum of free PPh3 shows seven peaks at d 137.27, 137.12, 133.79, 133.53, 128.60, 128.44 and 128.35 in CDCl3.73 NMR studies have been successful in characterizing the dynamic processes of PPh3 ligands in many transition metal complexes.74–77 Coordinated PPh3 is capable of rotating about the three P–Cipso bonds as well as as the metal–P bond.For steric reasons, the three phenyl groups generally adopt a chiral propeller-like conformation with either a clockwise or anticlockwise screw configuration.77 Interconversion of the two enantiomeric configurations or full rotation about any P–Cipso bond requires cooperative motion within PPh3.78 In a conformational study of free triphenylphosphine, Brock and Ibers 78 estimated both of these barriers to rotation to be less than 2 kcal mol21.For [Fe(h5-C5H5)(CO)(PPh3)(COMe)], Davies determined that the P–Fe rotational activation energy barrier (DG‡) was 10.3 kcal mol21 and that the phenyl ring rotation about the P–Cipso was rapid on the NMR time scale down to 290 8C.77 In order to determine if the steric eVects of the trans-spanning linkage resulted in restricted rotation of the phosphine ligand we employed (see below) variable temperature NMR studies on ruthenium complexes which contain phosphine ligands.Analysis of the ruthenium complex spectra. The literature contains little information on the NMR behavior of trans-spanning complexes due at times to the diYculties associated with the separation of isomers (cis and trans, monomers and dimers) and also the insolubility of many of the complexes in common NMR solvents.21 In contrast, the trans-spanning complexes 6–8 are monomeric, trans-isomers and are soluble in a variety of common NMR solvents.The 1H and 13C spectra of 6, 7 and 8 are shown in Fig. 6–8 [HETCOR and COSY data are given in SUP 57568, Fig. 15, 16 (complex 6), 19, 20 (7), and 23, 24 (8); variable temperature data are also given in SUP 57568, Figs.2288 J.Chem. Soc., Dalton Trans., 1999, 2281–2292 Table 6 13C NMR spectroscopy for ruthenium complexes Complex a d (ppm) b (coupling/Hz, assignment c) trpy 121.10 c (b), 121.20 c ( g), 121.24 c (d ), 137.1 c (c), 138.2 c (h), 149.5 c (a), 155.7 d ( f ), 156.5 d (e) trans-[Ru(Cl)(trpy)(PPh3)2][PF6], 5 122.66 ( g), 122.83 (d ), 126.84 (b), 128.57 (|3JPCk 1 5JP9Ck| = 9.1, k), 130.10 (|4JPCl 1 6JPCl| = 1.6, l), 130.13 (|1JPCi 1 3JP9Ci | = 39.2, i), 132.52 (s, h), 133.23 (|2JPCj 1 4JP9Cj | = 10.3, j), 136.87 (c), 155.79 (a), 157.66 ( f ), 158.18 (e) trans-[Ru(Cl)(trpy)(C3SPAN)][PF6], 6 20.72 (t), 33.07 (s), 66.40 (q), 122.74 ( g), 122.76 (d ), 127.17 (b), 128.75 (|3JPCk 1 5JP9Ck| = 9.6, k), 129.74 (|1JPCm 1 3JP9Cm| = 36.9, m), 129.85 (|3JPCo 1 5JP9Co| = 7.7, o), 130.64 (|4JPCl 1 6JP9Cl | = 1.9, l), 130.96 (2 C, |2JPCn 1 4JP9Cn| = 8.6, n; |1 PCi 1 3JP9Cl | = 41.0, i), 132.45 (h), 134.15 (|2JPCj 1 4JP9Cj | = 10.8, j), 136.75 (c), 137.21 (|4JPCp 1 6JP9Cp| = 1.9, p), 155.45 (a), 157.37 ( f ), 157.97 (e), 172.91 (r) trans-[Ru(Cl)(trpy)(C4SPAN)][PF6], 7 24.67 (t), 34.17 (s), 66.00 (q), 122.55 (d ), 122.64 ( g), 127.08 (b), 128.67 (|3JPCk 1 5JP9Ck| = 9.8, k), 128.73 (|3JPCo 1 5J)P9Co| = 7.2, o), 128.94 (|1JPCm 1 3JP9Cm| = 36.4, m), 130.51 (|4JPCl 1 6JP9Cl | = 0.0, l), 130.82 (|1JPCi 1 3JP9Ci | = 41.4, i), 131.13 (|2JPCn 1 4JP9Cn| = 10.3, n), 132.32 (h), 134.07 (|2JPCj 1 4JP9Cj | = 11.0, j), 136.73 (c), 137.94 (|4JPCp 1 6JP9Cp| = 0.0, p), 155.54 (a), 157.43 ( f ), 157.87 (e), 173.26 (r) trans-[Ru(Cl)(trpy)(ISPAN)][PF6], 8 68.38 (q), 122.91 (d), 123.31 ( g), 126.96 (b), 127.28 (v), 128.80 (|3JPCk 1 5JP9Ck| = 9.7, k), 129.40 (|1JPCm 1 3JP9Cm| = 36.0, m), 130.08 (u), 130.78 (|4JPCl 1 6JP9Cl | = 0.0, l), 130.84 (|1JPCi 1 3JP9Ci | = 40.9, i), 130.89 (|2JPCn 1 4JP9Cn| = 8.7, n), 130.90 (s), 131.03 (|3JPCo 1 5JP9Co| = 7.4, o), 132.40 (h), 134.24 (|2JPCj 1 4JP9Cj | = 11.0, j), 135.59 (t), 135.70 (|4JPCp 1 6JP9Cp| = 0.0, p), 136.82 (c), 155.78 (a), 156.82 ( f ), 158.22 (e), 165.65 (r) a NMR spectra were measured in CD2Cl2 and referenced against CDHCl2.b Letters used in assignments correlate to trpy Fig. 4; 5 Fig. 5; 6 Fig. 6; 7 Fig. 7; 8 Fig. 8. c Determined by one-bond HETCOR. d Determined by N-bond HETCOR. 17, 18 (complex 6), and 21, 22 (complex 7)]. Room temperature spectra will be the subject of discussion below unless otherwise noted. I. Analysis of coordinated trpy spectra. The coordination of the trpy ligand to a ruthenium(II) cation can result in a down- field shift of the bonded trpy proton resonances relative to free trpy;50,70 however, while Ha moves downfield 0.30–0.51 ppm in 5–8 as expected, the chemical shifts of the other trpy protons in 5–8 move 0.19–1.18 ppm upfield relative to the free trpy ligand.A possible reason for the downfield shift of Ha is given below. The 13C NMR chemical shifts of bonded trpy in 5 (Fig. 5) vary little when compared with the free trpy resonances with the following exceptions: Ca is 6.3 ppm and Cb is 5.7 ppm downfield of the free trpy positions, while the chemical shift of Ch in 5 is 5.7 ppm upfield of that in free trpy. Similar changes in chemical shifts are observed for 6–8.II. Analysis of the coordinated phosphine ligand spectrum for 5. The proton chemical shifts of the triphenylphosphine ligands of 5 are assigned as complex multiplets: meta (Hk, d 7.08), ortho (Hj, d 7.14) and para (Hl, d 7.24).When compared to the free triphenylphosphine ligand, the chemical shifts of 5 are Fig. 4 1H and 13C NMR spectra (300 MHz) of 2,29:69,60-terpyridine (trpy) in methylene chloride-d2. 0.04–0.21 ppm upfield of their expected position. Thus, there seems to be a mutual anisotropic deshielding between the phenyl rings and the trpy ligand. These results can be explained for 5 if the possible motions of the triphenylphosphine groups are considered. As only three proton resonances are observed for the PPh3 moieties of 5, free rotation about all three P–Cipso bonds as well as the Ru–P bond is indicated.The ORTEP diagram of 5 (Fig. 2) shows that (a) each phenyl group of triphenylphosphine has a diVerent orientation depending on its position relative to the trpy ligand and (b) both PPh3 ligands are similarly arranged. Two phenyl rings are located over and under the central pyridine ring of trpy in an essentially parallel arrangement. The other two pairs of phenyl rings are positioned on each side of the trpy ligand and are nearly perpendicular to the plane of the terminal trpy pyridines.Thus, from the crystal structure it may be postulated that as each triphenylphosphine rotates along the Ru–P bond each phenyl ring will adjust its orientation along the P–Cipso axis. That is, as each phenyl ring moves around the Ru–P bond, it may travel a monotonic or slightly oscillatory path over the trpy ring, but, as it clears the plane above (or Fig. 5 1H and 13C NMR spectra (300 MHz) of trans-[Ru(Cl)(trpy)- (PPh3)2]1, 5 in methylene chloride-d2.J. Chem. Soc., Dalton Trans., 1999, 2281–2292 2289 Fig. 6 1H and 13C NMR spectra (300 MHz) of trans-[Ru(Cl)(trpy)(C3SPAN)]1, 6 in methylene chloride-d2. below) the trpy ligand, it may rotate about the P–Cipso bond to become perpendicular to the trpy ring thereby time averaging the ortho and meta proton resonances and resulting in only three unique proton resonances.If this model is correct, the chemical shifts of the PPh3 ligands are expected to be upfield relative to the free PPh3, since the PPh3 ligands of 5 spend more time in the shielding areas over the trpy compared to the two pockets between the chlorine and each of the terminal pyridines of trpy. Likewise the trpy protons of 5 (except for Ha) should experience shielding by the phenyl rings of the PPh3 ligands and should also be found upfield when compared to the free trpy ligand.As these general upfield shifts are indeed observed in both the PPh3 and trpy ligands, anisotropic deshielding may be the cause. Notably, the Ha protons are expected to be unique since the phenyl rings of the PPh3 may freely rotate once they clear the plane of the trpy ligand. The 0.3 ppm downfield shift for Ha of 5 may be due to anisotropic deshielding from the phenyls in the pockets. The triphenylphosphine ligands in the 13C NMR spectrum of 5 show a set of four triplets, although that from the para position, Cl, looks like a singlet at low resolution.The signals are triplets due to virtual coupling to the second phosphorus. Since the two-bond P–Ru–P9 coupling is much larger than the P–C couplings, the high order pattern is a triplet rather than a doublet-of-doublets or a pentuplet. One can then measure only the algebraic sum of the two P–C couplings across the outer line of the triplet. For 5 the values are ipso (|1JP–Ci 1 3JP–Ci | = 39.2 Hz), ortho (|2JP–Cj 1 4JP–Cj | = 10.3 Hz), meta (|3JP–Ck 1 5JP–Ck| = 9.1 Hz) and para (|4JP–Cl 1 6JP–Cl | = 1.6 Hz).The peak assignments for 5 have been corroborated through HETCOR and COSY experiments. III. Analysis of the coordinated phosphine ligand spectra for 6–8. Proton shifts for the phenyl groups of PPh3 show some changes on going from 5 to the span complexes, 6–8. First, complexes 6–8 show a sub-spectrum for a para-disubstituted phenyl ring in the aromatic region as part of the spanning linkage and 8 shows an additional 1,3-disubstituted phenyl pattern for the isophthalate linkage.For any given phenyl ring on the phosphine ligand, the two ortho positions are equivalent to each other as are the two meta positions. These results indicate that the phenyl rings are undergoing rapid rotation about the P–Cipso bond at room temperature. Second, meta and para protons maintain similar chemical shift values for 5–8 but there are significant diVerences between the complexes in the positions of the ortho protons of the phenyl rings, Hj.The Hj chemical shifts of 6 and 7 are approx. 0.5 ppm downfield relative to 5; while the chemical shift of Hj in 8 is 0.78 ppm downfield relative to 5. These downfield shifts will be discussed below in terms of motional averaging in the span complexes. Finally, the 0.3 ppm downfield shift for Ha of 6–8 may also be attributed to the anisotropic deshielding from the phenyls in the pockets.There is little diVerence in the 13C shifts between the spanned complexes, 6–8, and that containing the triphenylphosphine ligand, 5. IV. Analysis of the variable temperature spectra for 5. Restricted rotation about the P–Cipso bonds or Ru–P bond should be observed if the phosphine ligands are sterically con- fined; variable temperature studies were conducted to detect such restricted rotations. If the rotation about each P–Cipso becomes slow on cooling, it is expected that the ortho and meta peaks of the triphenylphosphine moiety will split into two distinct resonances while the ipso (13C only) and para resonances should remain singular.If rotation about the Ru–P bond becomes slow on the NMR time scale, the peaks observed from the ipso, ortho, meta and para carbons should split into three, one for each of the rotating phenyl groups. Variable temperature 1H and 13C NMR spectra for 5 (see SUP 57568, Fig. 13 and 14) show no line broadenings until 290 8C where Hj, Ci and Cj show significant broadening and Cl shows slight broadening.Since the ipso carbon, Ci and the para carbon, Cj, should not show broadening for slowed rotation about the P–Cipso bond, it is possible but not conclusive that rotation about the P–Ru bond is becoming slow at low temperatures. Interestingly, [Fe(h5-C5H5)(CO)(PPh3)(COMe)] shows a similar pattern of low temperature behavior in the 13C spectrum.77 For the iron complex, the ipso, ortho and para carbons show broadening at2290 J.Chem. Soc., Dalton Trans., 1999, 2281–2292 Fig. 7 1H and 13C NMR spectra (300 MHz) of trans-[Ru(Cl)(trpy)(C4SPAN)]1, 7 in methylene chloride-d2. Fig. 8 1H and 13C NMR spectra (300 MHz) of trans-[Ru(Cl)(trpy)(ISPAN)]1, 8 in methylene chloride-d2. 260 8C with complete freezing out of the spectrum at 290 8C. The meta carbons are only starting to broaden at 290 8C and there is no sign of broadening for the acetyl methyl group or the cyclopentadienyl ring.For [Fe(h5-C5H5)(CO)(PPh3)(COMe)], Davies determined an experimental DG = 10 kcal mol21 for the Fe–P rotation at 62.90 MHz. As 5 requires a lower temperature for the start of broadening, even at a higher frequency (75 MHz), the Ru–P rotation in 5 is faster. V. Analysis of the variable temperature spectra for 6–8. The variable temperature NMR spectra of 6 and 7 are qualitativelyJ. Chem. Soc., Dalton Trans., 1999, 2281–2292 2291 similar to that of 5 except that even though there is no sign of broadening for the trpy resonances of 5 at 290 8C, the Ha, Hj and Hn resonances of 6 and 7 are already broad at room temperature.Maximum line broadenings for Ha, Hj, Hn and Ho are found between 5 8C and 240 8C. Below 240 8C these resonances sharpen though they do not split into separate resonances as described above for slowed rotations about the Ru–P or P–Cipso bonds. At 260 8C, these resonances are sharp indicating that the motions about the Ru–P and P–Cipso bonds are still fast on the NMR time scale.At 290 8C, Ha, Hj and Hn start to broaden but again, there is no indication of which motion is slowing. The methylene proton signals for the span also broaden but it is not possible to tell if this is an independent conformational process in the spanning linkage or if this is related to rotation about the P–Ru bond. Complex 7 follows the same behavior as 6 but maximum coalescence starts about 40 8C lower.Complex 7 is expected to be more accommodating than 6 for rotation about either the trpy or chloride moieties. The diVerence in the coalescence temperatures may be related to subtle changes in span rotation about the P–Ru bond. Finally, two types of motion appear to be available in the spanned complexes. The first involves a complete 3608 rotation of the span about the P–Ru–P axis, where the span passes over both the chloride and the meridional trpy ligand. The second involves a restricted “fan-like” motion of the spanning linkage limited by the two terminal pyridines of the trpy ligands, where the span passes over only the chloride ligand and not the trpy ligand.Both mechanisms would time average the trpy resonances. At this time, it is not possible to distinguish which of these motional mechanisms is operating. Conclusions The in situ strategy for the preparation of trans-spanning ligands with ester linkages resulted in spanning linkages which are stable to oxidation and reduction, hydrolysis and cyclometallation while maintaining the benefits of span variability and the formation of ruthenium complexes which display bond angles close to ideal octahedral geometry.The NMR spectral analyses yielded several observations: (1) the original assignment of the 1H NMR spectrum of trpy by Carlson is inconsistent with our COSY and HETCOR analyses, (2) the NMR spectra of 6–8 are consistent with a flexible spanning linkage that does not demonstrate restricted rotation about either the P–Cipso or the Ru–P bonds even at low temperatures, and (3) maximal coalescence increased with temperature with shorter alkyl chain lengths in the trans-spanning linkage.Finally, the X-ray crystal structure analysis of 7 showed that the spanning linkage is positioned in one of the two pockets defined by the chloride ligand and the terminal pyridine groups of trpy. Acknowledgements This work was supported in part by the National Science Foundation (CHE 9120602) and the ARCO Chemical Company.Purchase of the Siemens R3m/v diVractometer was made possible by Grant 89-13733 from the Chemical Instrumentation Program of the National Science Foundation. The authors also gratefully acknowledge Johnson Matthey/Alfa/ Aesar for the loan of the RuCl3?nH2O used in these experiments. The authors acknowledge the assistance of Mr Simit D. Shah, Mr Sejal C. Patel and Mr Brian M. Scull of Villanova University in preparation, purification and characterization of the complexes. References 1 J.C. Bailar, Jr., Coord. Chem. Rev., 1990, 100, 1. 2 D. M. A. Minahan, W. E. Hill and C. A. McAuliVe, Coord. Chem. Rev., 1984, 55, 31. 3 A. Pryde, B. L. Shaw and B. Weeks, J. Chem. Soc., Dalton Trans., 1976, 322. 4 N. A. Al-Salem, H. D. Empsall, R. Markham, B. L. Shaw and B. Weeks, J. Chem. Soc., Dalton Trans., 1979, 1972. 5 C. Crocker, R. J. Errington, R. Markham, C. J. Moulton, K. J.Odell and B. L. Shaw, J. Am. Chem. Soc., 1980, 102, 4373. 6 B. L. Shaw, J. Am. Chem. Soc., 1975, 97, 3856. 7 A. Pryde, B. L. Shaw and B. Weeks, J. Chem. Soc., Dalton Trans., 1976, 322. 8 H. D. Empsall, E. Mentzer, D. Pawson, B. L. Shaw, R. Mason and G. A. Williams, J. Chem. Soc., Chem. Commun., 1977, 311. 9 F. C. Marsh, R. Mason, K. M. Thomas and B. L. Shaw, J. Chem. Soc., Chem. Commun., 1975, 584. 10 N. W. Alcock, J. M. Brown and J. C. JeVrey, J. Chem. Soc., Chem. Commun., 1974, 829. 11 N. W. Alcock, J. M. Brown and J. C. JeVrey, J. Chem. Soc., Dalton Trans., 1976, 583. 12 N. W. Alcock, J. M. Brown and J. C. JeVrey, J. Chem. Soc., Dalton Trans., 1977, 888. 13 W. E. Hill, J. G. Taylor, C. A. McAuliVe, K. W. Muir and L. Manojlovic-Muir, J. Chem. Soc., Dalton Trans., 1982, 833. 14 W. E. Hill, C. A. McAuliVe, I. E. Niven and R. V. Parish, Inorg. Chim. Acta, 1979, 38, 273. 15 W. Levason, C. A. McAuliVe and S. G. Murray, J. Organomet. Chem., 1976, 110, C25. 16 W.E. Hill, D. M. A. Minahan, J. G. Taylor and C. A. McAuliVe, J. Am. Chem. Soc., 1982, 104, 6001. 17 (a) W. E. Hill, D. M. A. Minahan, J. G. Taylor and C. A. McAuliVe, J. Chem. Soc., Perkin Trans. 2, 1982, 327; (b) J. C. Briggs, C. A. McAuliVe, W. E. Hill, D. M. A. Minahan and G. Dyer, J. Chem. Soc., Dalton Trans., 1982, 321; (c) J. C. Briggs, C. A. McAuliVe, W. E. Hill, D. M. A. Minahan, J. G. Taylor and G. Dyer, Inorg. Chem., 1982, 21, 4204; (d) W. E. Hill, D. M. A. Minahan, C.A. McAuliVe and K. L. Minten, Inorg. Chim. Acta, 1983, 74, 9; (e) W. E. Hill, D. M. A. Minahan and C. A. McAuliVe, Inorg. Chem., 1983, 22, 3382; ( f ) J. C. Briggs, C. A. McAuliVe and G. Dyer, J. Chem. Soc., Dalton Trans., 1984, 423; (g) W. E. Hill, M. Q. Islam, T. R. Webb and C. A. McAuliVe, Inorg. Chim. Acta, 1988, 146, 111. 18 H. M. K. K. Pathirana, A. W. Downs, W. R. McWhinnie and P. Granger, Inorg. Chim. Acta, 1988, 143, 161. 19 P. N. Kapoor, P. S. Pregosin and L.M. Venanzi, Helv. Chim. Acta, 1982, 65, 654. 20 M. Barrow, H.-B. Bürgi, M. Camalli, F. Caruso, E. Fischer, L. M. Venanzi and L. Zambonelli, Inorg. Chem., 1983, 22, 2356. 21 L. I. Elding, B. Kellenberger and L. M. Venanzi, Helv. Chim. Acta, 1983, 66, 1676. 22 E. Baumgartner, F. J. S. Reed, L. M. Venanzi, F. Bachechi, P. Mura and L. Zambonelli, Helv. Chim. Acta, 1983, 66, 2572. 23 P. Boron-Rettore, D. M. Grove and L. M. Venanzi, Helv. Chim. Acta, 1984, 67, 65. 24 M. Camalli, F.Caruso, S. Chaloupka and L. M. Venanzi, Helv. Chim. Acta, 1988, 71, 703. 25 M. Camalli, F. Caruso, S. Chaloupka, P. N. Kapoor, P. S. Pregosin and L. M. Venanzi, Helv. Chim. Acta, 1984, 67, 1603. 26 H.-B. Bürgi, J. Murray-Rust, M. Camalli, F. Caruso and L. M. Venanzi, Helv. Chim. Acta, 1989, 72, 1293. 27 D. M. P. Mingos, D. J. Sherman and I. D. Williams, Transition Met. Chem., 1987, 12, 493. 28 C. Crocker, R. J. Errington, R. Markham, C. J. Moulton and B. L. Shaw, J. Chem. Soc., Dalton Trans., 1982, 387. 29 R. J. Errington and B. L. Shaw, J. Organomet. Chem., 1982, 238, 319. 30 S. D. Perera, M. Shamsuddin and B. L. Shaw, Can. J. Chem., 1995, 73, 1010. 31 J. R. Briggs, A. G. Constable, W. S. McDonald and B. L. Shaw, J. Chem. Soc., Dalton Trans., 1982, 1225. 32 H. Rimml and L. M. Venanzi, J. Orgamomet. Chem., 1983, 259, C6. 33 S. D. Perera, B. L. Shaw and M. Thornton-Pett, J. Chem. Soc., Dalton Trans., 1992, 1469. 34 S. D. Perera, B. L. Shaw and M. Thornton-Pett, J.Chem. Soc., Dalton Trans., 1994, 3311. 35 K. K. Hii, S. D. Perera and B. L. Shaw, J. Chem. Soc., Dalton Trans., 1995, 625. 36 C. Crocker, H. D. Empsall, R. J. Errington, E. M. Hyde, W. S. McDonald, R. Markham, M. C. Norton, B. L. Shaw and B. Weeks, J. Chem. Soc., Dalton Trans., 1982, 1217. 37 R. J. Errington, W. S. McDonald and B. L. Shaw, J. Chem. Soc., Dalton Trans., 1982, 1829. 38 R. A. Leising, J. J. Grzybowski and K. J. Takeuchi, Inorg. Chem., 1988, 27, 1020. 39 D.L. Jameson and L. E. Guise, Tetrahedron Lett., 1991, 32, 1989.2292 J. Chem. Soc., Dalton Trans., 1999, 2281–2292 40 D. D. Perrin and W. L. F. Armarego, Purification of Laboratory Chemicals, Pergamon Press, Oxford, 3rd edn., 1988. 41 J. R. Garbow, D. P. Weitekamp and A. Pines, Chem. Phys. Lett., 1982, 93, 504. 42 G. E. Martin and A. S. Zektzer, Magn. Reson. Chem., 1988, 26, 631. 43 M. R. Churchill, R. A. Lashewycz and F. J. Rotella, Inorg. Chem., 1977, 16, 265. 44 G.M. Sheldrick, SHELXTL PLUS, Siemens Analytical Instruments, Madison, WI, 1983. 45 M. R. Churchill, Inorg. Chem., 1973, 12, 1213. 46 C. K. Johnson, ORTEP II, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1974. 47 B. P. Sullivan, J. M. Calvert and T. J. Meyer, Inorg. Chem., 1980, 19, 1404. 48 B. S. Furness, A. J. Hannaford, P. W. G. Smith and A. R. Tatchell, in Vogel’s Textbook of Practical Organic Chemistry, J. Wiley and Sons, New York, 5th edn., 1989. 49 R. A. Leising, S. A. Kubow and K. J. Takeuchi, Inorg. Chem., 1990, 29, 4569. 50 R. A. Leising, S. A. Kubow, M. R. Churchill, L. A. Buttrey, J. W. Ziller and K. J. Takeuchi, Inorg. Chem., 1990, 29, 1306. 51 H. J. Lawson, T. S. Janik, M. R. Churchill and K. J. Takeuchi, Inorg. Chim. Acta, 1990, 174, 197. 52 L. F. Szczepura, S. A. Kubow, R. A. Leising, W. J. Perez, M. H. V. Huynh, C. H. Lake, D. G. Churchill, M. R. Churchill and K. J. Takeuchi, J. Chem. Soc., Dalton Trans., 1996, 1463. 53 R. P. Thummel and Y. Jahng, J. Org. Chem., 1985, 50, 2407. 54 K. Nakamoto, J. Phys. Chem., 1960, 64, 1420. 55 H. S. Lim, D. J. Barclay and F. C. Anson, Inorg. Chem., 1972, 11, 1460. 56 T. Matsubara and P. C. Ford, Inorg. Chem., 1976, 15, 1107. 57 R. S. Nicholson, Anal. Chem., 1966, 38, 1406. 58 C. A. Bessel, J. A. Margarucci, J. H. Acquaye, R. S. Rubino, J. Crandall, A. J. Jircitano and K. J. Takeuchi, Inorg. Chem., 1993, 32, 5779. 59 D. T. Walker and H. Taube, Inorg. Chem., 1981, 20, 2828. 60 D. A. Bohling, J. F. Evans and K. R. Mann, Inorg. Chem., 1982, 21, 3546. 61 R. D. Feltham and R. G. Hayter, J. Chem. Soc., 1964, 4587. 62 J. A. Davies, F. R. Hartley and S. G. Murray, Inorg. Chim. Acta, 1980, 43, 69. 63 E. Constable, J. Chem. Soc., Dalton Trans., 1985, 2687. 64 C. A. Bessel, R. F. See, D. J. Jameson, M. R. Churchill and K. J. Takeuchi, J. Chem. Soc., Dalton Trans., 1992, 3223. 65 M. R. Churchill, R.F. See, C. A. Bessel and K. J. Takeuchi, J. Chem. Crystallogr., 1996, 26, 853. 66 M. R. Churchill, L. Krajkowski, L. F. Szczepura and K. J. Takeuchi, J. Chem. Crystallogr., 1996, 26, 543. 67 M. Barrow, H.-B. Bürgi, D. K. Johnson and L. M. Venanzi, J. Am. Chem. Soc., 1976, 98, 2356. 68 N. C. Baezinger, K. M. Dittemore and J. R. Doyle, Inorg. Chem., 1974, 13, 805. 69 P. H. Davis, R. L. Belford and I. C. Paul, Inorg. Chem., 1973, 12, 213. 70 F. E. Lytle, L. M. Petrosky and L. R. Carlson, Anal. Chim. Acta, 1971, 57, 239. 71 S. Castellano, H. Gunther and S. Ebersole, J. Phys. Chem., 1965, 69, 4166. 72 E. Pretsch, J. Seibl, W. Simon and T. Clerc, Tables of Spectral Data for Structure Determination of Organic Compounds, Springer- Verlag, New York, 1983. 73 C. J. Pouchert and J. Behnke, Aldrich Library of 13C and 1H FT NMR Spectra, Aldrich Chemical Company, Inc., Milwaukee, WI, 1st edn., 1993, vol. 2, p. 1653. 74 J. W. Faller and B. V. Johnson, J. Organomet. Chem., 1975, 96, 99. 75 J. Vicente, M.-T. Chicote, M.-C. Lagunas, P. G. Jones and B. Ahrens, Inorg. Chem., 1997, 36, 4938. 76 E. E. Wille, D. S. Stephenson, P. Capriel and G. Binsch, J. Am. Chem. Soc., 1982, 104, 405. 77 S. G. Davies, A. E. Derome and J. P. McNally, J. Am. Chem. Soc., 1991, 113, 2854 and refs. therein. 78 C. P. Brock and J. A. Ibers, Acta Crystallogr., Sect. B, 1973, 29, 2426. Paper 8/07574C