摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3745–3750 3745 Synthesis and characterization of complexes of PdII and PtII containing the iminophosphorane ligand Ph3P] NC]] N Larry R. Falvello, Susana Fernández, María M. García, Rafael Navarro * and Esteban P. Urriolabeitia Departamento de Química Inorgánica, Instituto de Ciencia de Materiales de Aragón, Universidad de Zaragoza, Consejo Superior de Investigaciones Científicas, 50009 Zaragoza, Spain. E-mail: rafanava@posta.unizar.es; esteban@posta.unizar.es Received 3rd July 1998, Accepted 30th September 1998 The chemistry of the iminophosphorane Ph3P]] NC]] ] N towards diVerent palladium(II) and platinum(II) subtrates has been investigated. This ligand shows a high selectivity for co-ordination through the terminal N atom (that of the cyanide group), while the phosphoiminic N atom shows very poor nucleophilic ability.The reaction of [PdCl2(NCPh)2] with Ph3P]] NCN (1: 2 molar ratio) gives trans-[PdCl2(NCNPPh3)2] 1.The crystal structure of 1 has been determined and shows the iminophosphorane co-ordinated selectively through the nitrilic N atom. The reaction of the cationic derivatives [M(C–P)(NCMe)2][ClO4] [M = Pd or Pt; C–P = o-CH2C6H4P(C6H4Me-o)2] with Ph3P]] NCN (1: 2 molar ratio) resulted in the formation of [M(C–P)(NCN]] PPh3)2][ClO4] (M = Pd 2 or Pt 3), for which the same terminal N-co-ordination has been observed. The reaction of [Pd(dmba)(NCMe)2][ClO4] (dmba = C6H4CH2NMe2-2) with Ph3P]] NCN (1: 1 molar ratio) gave the dinuclear derivative [{Pd(m-NCN]] PPh3)- (dmba)}2][ClO4]2 4 in which the iminophosphorane acts as an N,N-bridging ligand, while reaction in 1 : 2 molar ratio gave [Pd(dmba)(NCN]] PPh3)2][ClO4] 5 which possess N-terminal co-ordination.The reaction of this iminophosphorane with monosolvate complexes [Pd(dmba)(PPh3)(THF)][ClO4] or [Pt(C–P)(PPh3)(NCMe)][ClO4] (1 : 1 molar ratio) resulted in the formation of [Pd(dmba)(PPh3)(NCN]] PPh3)][ClO4] 6 or [Pt(C–P)(PPh3)- (NCN]] PPh3)][ClO4] 7, and the reaction with the halide bridging derivatives [{Pd(m-Cl)(dmba)}2] or [{Pt(m-Cl)- (C–P)}2] (2 : 1 molar ratio) produced cleavage of the bridging system and formation of [PdCl(dmba)(NCN]] PPh3)] 8 and [PtCl(C–P)(NCN]] PPh3)] 9.The same N-terminal co-ordination mode is observed in complexes 6–9. Introduction We have been studying the co-ordination chemistry of astabilized phosphorus ylides Ph3P]] C(H)R (R = COMe, COPh, COOMe, CONMe2 or CN) and their behaviour as ambidentate ligands.1–6 The reactivity of these ylides towards several orthometallated complexes of PdII and PtII has been examined and their ambidentate character observed.Furthermore, we have found that, in many cases, it is possible to predict both the coordination site of the ylide and the donor atom linked to the metal (O- versus C-bonding, N- versus C-bonding) using as simple a concept as the antisymbiotic eVect.7,8 The iminophosphoranes, compounds of general structure R3P]] NR9, are an interesting class of compounds closely related to the phosphorus ylides and showing numerous applications to organic synthesis (for instance, the formation of C]] N bonds through the Aza–Wittig reaction).9a Moreover, they can behave as ligands towards transition metals through, at least, the lone pair on the N atom.In spite of the similarities, their use as ligands is less developed than that of the ylides,9 although in recent years several groups have focused their interest on the co-ordinative properties and reactivity of iminophosphoranes. 9,10 Prompted by the results obtained with the a-stabilized ylides, we have decided to explore the co-ordination chemistry of the a-stabilized iminophosphorane Ph3P]] NC]] ] N,11 a compound for which there is no precedent in co-ordination chemistry.Its reactivity has been explored towards several palladium(II) and platinum(II) substrates (see Scheme 1).We have chosen complexes with two labile ligands in diVerent arrangements and with diVerent trans atoms, such as [MCl2(NCPh)2] (M = Pd or Pt),12 [M(C–P)(NCMe)2][ClO4] (M = Pd† or Pt 13) and [Pd(dmba)(NCMe)2][ClO4] 16 [C–P = o-CH2C6H4P(C6H4- Me-o)2; dmba = C6H4CH2NMe2-2], complexes with one labile ligand such as [Pt(C–P)(PPh3)(NCMe)][ClO4] 5 and [Pd(dmba)- (PPh3)(THF)][ClO4] 17 and dinuclear complexes such as [{M(m- Cl)(C–P)}2] (M = Pd14 or Pt 13) and [{Pd(m-Cl)(dmba)}2] 18 in order to explore the co-ordination properties (nucleophilic ability of the iminic N atom and steric requirements) of this ligand in diVerent environments.In this paper we report the results obtained from this chemistry. Results and discussion The reaction of [PdCl2(NCPh)2] with Ph3P]] NCN (1: 2 molar ratio) in acetone at room temperature results in the precipitation of an orange solid of stoichiometry [PdCl2(NCNPPh3)2] 1 as deduced from its elemental analysis.The reaction of [PtCl2(NCPh)2] with Ph3P]] NCN (1: 2 molar ratio) performed under the same conditions (acetone; room temperature, r.t.) gives a mixture of the starting products. If the reaction is performed in refluxing toluene the spectroscopic data of the resulting solid show the presence of both starting materials together with at least three products which appear to contain co-ordinated iminophosphorane. However, the recrystallization of this mixture failed in all attempted cases and we were not able to isolate a Pt–NCN]] PPh3 derivative free of the starting † The complex [Pd(C–P)(NCMe)2][ClO4] was prepared in a way similar to that described in ref. 13 for [Pt(C–P)(NCMe)2][ClO4], with the exception that the reaction was performed starting from [{Pd(m-Cl)(C–P)}2] (see ref. 14) under a nitrogen atmosphere in dry NCMe (ref. 15).3746 J. Chem. Soc., Dalton Trans., 1998, 3745–3750 Scheme 1 C N Ph3P N Cl Pd NCN=PPh3 Cl Ph3P=NCN M C P NCN=PPh3 NCN=PPh3 Pd C N NCN=PPh3 NCN=PPh3 Pd C N N N PPh3 C C N N PPh3 Pd C N Pd C N PPh3 NCN=PPh3 Pt C P PPh3 NCN=PPh3 M C X NCN=PPh3 Cl (A) 1 M = Pd, X = NMe2 8 M = Pt, X = P(C6H4Me- o)2 9 M = Pd 2, Pt 3 2 (C) (C) (C) Ph3P=NCN 4 5 6 7 (D) (E) (F), (G) (B) M H2 C P (C6H4Me- o)2 NCMe NCMe Pd N Me2 NCMe NCMe M H2 C P (C6H4Me- o)2 Cl Pd N Me2 Cl Pt H2 C P (C6H4Me- o)2 PPh3 NCMe Pd N Me2 PPh3 THF Cl M NCPh Cl PhCN (B); M = Pd, Pt 2 (G); M = Pd, Pt (E) (A); M = Pd, Pt 2 (C) (D) (F) materials, probably due to the similar solubility of the species involved.The molecular structure of complex 1 has been determined by X-ray diVraction methods. A drawing of the molecule is shown in Fig. 1 and selected bond distances and angles are presented in Table 1. The palladium atom lies on a crystallographic inversion center and has a very slightly distorted square-planar environment. The Pd–Cl bond distance [2.2969(9) Å] falls within the usual range of distances found for this kind of bond,19 while the Pd–N(1) bond distance [1.971(3) Å] is shorter than other Pd–N (nitrile) bond lengths found in the literature (2.065(14) Å for [(MeNC)PdCl2(m-dpmp)- PdCl2]?NCMe20 and 2.076(3) Å for [Pd(dmba){P(OMe)3}- {NCC(H)]] PPh3}][ClO4]2).With respect to the internal bond distances and angles of the iminophosphorane ligand, the N(1)–C(1) bond distance [1.151(4) Å] is fairly typical for nitriles.2,21 The C(1)–N(2) bond Fig. 1 Thermal ellipsoid plot of trans-[PdCl2(NCN]] PPh3)2] 1.Hydrogen atoms have been omitted for clarity. Atoms are drawn at the 50% probability level; Pd(1) lies on a crystallographic inversion center. distance [1.292(4) Å] is shorter than other distances reported for this kind of bond in iminophosphoranes [ranging from 1.351(6) to 1.461(6) Å],9,10,21,22 regardless of whether the ligand is co-ordinated or not, and is similar within experimental error to that reported for [Ph3P]] NC]] ] NC10H15-1][SbCl6] 23 [1.25(2) Å] (C10H15 = adamantyl). The N(2)–P(1) bond [1.609(3) Å] is shorter than that reported for the latter salt [1.64(1) Å], is similar to (within experimental error) or shorter than those found in N-co-ordinated iminophosphoranes [range 1.604(7)–1.627(4) Å], but is longer than those found in free iminophosphoranes [range 1.569(4)–1.585(4) Å].10 This phosphorus–nitrogen bond is seen to have considerable single bond character if compared to the “true” P]] N double bond in [Ph3P]] N]] PPh3]1 [1.547(2) Å].24 Another interesting feature of the structure is the bond angle Pd(1)–N(1)–C(1) [161.9(3)8], the value of which is less than the 1808 expected for sp hybridization of the nitrogen atom.The bond angle N(1)–C(1)–N(2) is 175.2(3)8 and indicates an almost Table 1 Selected bond lengths (Å) and angles (8) for complex 1 Pd(1)–N(1) Pd(1)–Cl(1) P(1)–N(2) P(1)–C(8) N(1A)–Pd(1)–N(1) N(1)–Pd(1)–Cl(1) N(2)–P(1)–C(2) N(2)–P(1)–C(14) C(2)–P(1)–C(14) N(1)–C(1)–N(2) 1.971(3) 2.2969(9) 1.609(3) 1.789(3) 180.0 91.16(9) 111.91(14) 112.23(14) 110.22(14) 175.2(3) P(1)–C(2) P(1)–C(14) N(1)–C(1) C(1)–N(2) N(1A)–Pd(1)–Cl(1) N(2)–P(1)–C(8) C(8)–P(1)–C(2) C(8)–P(1)–C(14) C(1)–N(1)–Pd(1) C(1)–N(2)–P(1) 1.791(3) 1.796(3) 1.151(4) 1.292(4) 88.84(9) 105.74(14) 107.65(14) 108.84(14) 161.9(3) 122.3(2) Symmetry transformation used to generate equivalent atoms: A 2x, 2y, 2z.J.Chem. Soc., Dalton Trans., 1998, 3745–3750 3747 linear disposition of these bonds, in accord with the presence of a cyanide unit. The environment around the N(2) atom is trigonal planar [C(1)–N(2)–P(1) 122.3(2)8] and that around the phosphorus atom P(1) is tetrahedral.All of these facts imply that there is an extensive delocalization of the charge density in the P(1)–N(2)–C(1)–N(1) system. The distances and angles taken together suggest participation of the resonance forms B and D in the bonding description with only small contributions of A and C.However, the question of why the iminophosphorane co-ordinates selectively through the nitrilic N atom instead of through the iminic N atom remains unanswered. In our experience, the co-ordinative ability of the iminic N atom is worse than those of other donor atoms present in the ligand. For instance, in Ph3P]] NCOC6H4N-2 both the carbonylic oxygen atom and the pyridinic N atom have the same or even a better co-ordination ability than the iminic N atom.22 This fact would probably be due to the delocalization of the electronic density, which results in a lower charge density at the iminic N atom.It is also worth noting the diVerent behaviour of the phosphoylide Ph3P]] C(H)CN and the iminophosphorane Ph3P]] NCN towards the same precursor, in this case [PdCl2(NCPh)2]. While the former ligand co-ordinates selectively through the ylidic carbon atom, giving a mixture of diastereoisomers (RR/SS and RS/SR) trans-[PdCl2{C(H)- (CN)PPh3}2], the latter gives 1 which shows selective “end-on” N(nitrile) co-ordination.Complex 1 was also characterized by spectroscopic methods. The IR spectrum shows a trans arrangement of the chloride ligands, and hence of the iminophosphorane ligands, as deduced from the observation of a sharp absorption at 348 cm21. Moreover, the IR spectrum shows a strong band at 2256 cm21 attributed to the C]] ] N stretch. This absorption is shifted to higher energy with respect to that of free Ph3P]] NCN (2176 cm21); this shift can be interpreted as a consequence of co-ordination of the iminophosphorane through the iminic N atom, by analogy with the results obtained with the C-coordinated phosphoylide Ph3P]] C(H)CN2 and N(imine)-bonded keto-stabilized iminophosphoranes Ph3P]] NC(O)R.22 The 31P- {1H} NMR spectrum of 1 shows a singlet resonance at d 27.44 which is shifted to low field by only about 2 ppm with respect to free Ph3P]] NCN (d 25.25).The deshielding of the phosphorus nucleus that would be expected to result from iminic N-bonding is not observed 22 thus suggesting that an “end-on” nitrile coordination has taken place, as is depicted in Scheme 1.Further study of the reactivity of this iminophosphorane ligand gives additional evidence of the poor co-ordinating ability of the iminic N atom. The reaction of [M(C–P)(NCMe)2][ClO4] (M = Pd or Pt) with 2 equivalents of Ph3P]] NCN results in the formation of white solids of stoichiometry [M(C–P)- (NCN]] PPh3)2][ClO4] (M = Pd 2 or Pt 3) as indicated by their elemental analyses and mass spectra (see Experimental section).If the reaction is carried out in 1 : 1 molar ratio complex mixtures of 2 or 3 and other unidentified products are obtained. The IR spectra of 2 and 3 show absorptions corresponding to the C]] ] N group at 2215 and 2228 cm21, respectively, and the 31P-{1H} NMR spectra show resonances attributed to the phosphorus atom of the C–P ligands [d 37.54 for 2 and 16.10 for 3 (1JPtP = 4551 Hz)] together with resonances attributed to the iminophosphorane which appear as two close singlets at Pd N C N PPh3 Pd N C N PPh3 Pd N C N PPh3 Pd N C N PPh3 A B C D d 27.76 and 29.18 for 2 and at 28.06 and 29.52 for 3.The close similarity of these resonances to each other and to that of free Ph3P]] NCN clearly suggests an N-terminal (nitrile) co-ordination of the ligand, as also deduced for complex 1. Additional evidence is the absence of coupling between the P atom of the C–P ligand and the P atom of the trans Ph3P]] NCN ligand.Such coupling is observable when a C-coordinated ylide is trans to the P atom of the C–P group5 and also when an iminophosphorane is co-ordinated through the iminic N atom22 trans to the P atom of the C–P ligand. The structure represented in Scheme 1 accounts for all of these facts. The observed reactivity contrasts with that reported for the phosphoylide Ph3P]] C(H)CN5 and C–P complexes. There, [Pt(C–P)(NCMe)2][ClO4] reacts with 1 equivalent of Ph3- P]] C(H)CN to give [Pt(C–P){C(H)(CN)PPh3}(NCMe)][ClO4] which reacts with a second equivalent of the ylide to give [Pt- (C–P){C(H)(CN)PPh3}{NCC(H)]] PPh3}][ClO4].It is clear that for the phosphoylide there is a high tendency to co-ordinate through the ylidic carbon atom and that the N-co-ordination of the second ylide is driven by the electronic and steric requirements of the first C-co-ordinated ylide. This tendency does not carry over to the iminic N atom of the iminophosphorane and, hence, selective N(nitrile) co-ordination is observed.The reactivity of the iminophosphorane towards [Pd(dmba)(NCMe)2][ClO4] more nearly resembles that reported for the analogous phosphoylide. Thus, [Pd(dmba)- (NCMe)2][ClO4] reacts with Ph3P]] NCN (1: 1 molar ratio) giving a pale yellow solid of stoichiometry [Pd(dmba)- (NCNPPh3)][ClO4], as deduced from its elemental analysis. The mass spectrum (positive ion FAB) of this solid shows a peak at m/z 542 which corresponds to the cationic fragment [Pd(dmba)(NCNPPh3)]1.The IR spectrum shows a sharp absorption at 2249 cm21, attributed to the cyano group of the iminophosphorane ligand, and does not show the presence of co-ordinated NCMe which usually appears at higher frequencies. The 1H NMR spectrum shows the presence of the expected resonances for the Ph, C6H4, CH2N and NMe2 groups, which integrated as 15:4:2:6, thus confirming the presence of only one iminophosphorane group per dmba ligand. The observation of only one set of signals indicates the presence of only one isomer.Similarly, the 31P-{1H} NMR spectrum shows a singlet resonance at d 29.16. Excluding the possibility of chelation of the ligand because of the constraint imposed by the nitrile group, we can propose two possible structures for this complex. One is mononuclear with a covalent perchlorate group. This proposition is immediately discarded because the absorptions corresponding to the ClO4 2 group are not split.2 The other possibility is a dinuclear structure with the iminophosphorane acting as a bridging ligand.This second proposition is supported by two facts: the position of the n(CN) absorption at 2249 cm21 shifted to high wavenumber with respect to that of [Pd(dmba)(NCNPPh3)2][ClO4] 5 (2211 cm21, see below) in which both iminophosphorane ligands are N(nitrile)-co-ordinated, and the position of the phosphorus resonance at d 29.16, shifted downfield with respect to those of 5 (d 26.60 and 26.41, see below).Both the increase in the n(CN) wavenumber and the deshielding of the phosphorus resonance strongly suggest that both N atoms of Ph3P]] NCN are bonded to the palladium center. A more accurate definition of the nature of the product is given by the measurement of the molar conductance in acetone solution (c = 5 × 1024 M), which is LM = 268 W21 cm2 mol21 corresponding to a 2 : 1 electrolyte.25 Thus, the product can be formulated as the dinuclear complex [{Pd(m-NCN]] PPh3)(dmba)}2][ClO4]2 4.The observation of only one set of resonances in the NMR spectra shows that 4 must be a product of high symmetry and that of the two highest symmetry isomers only one is present. The 1H–1H NOESY spectrum of complex 4 shows a strong NOE interaction between the ortho protons of phenyl groups and the methyl protons of the NMe2 group, showing their mutually cis3748 J.Chem. Soc., Dalton Trans., 1998, 3745–3750 disposition. The dimeric structure depicted in Scheme 1 accounts for these observations. When the reaction between [Pd(dmba)(NCMe)2][ClO4] and Ph3P]] NCN is performed in 1: 2 molar ratio a solid of stoichiometry [Pd(dmba)(NCNPPh3)2][ClO4] 5 is obtained as deduced from its elemental analysis and mass spectrum. The characterization of complex 5 follows the same key features as those reported for 1–4. The n(CN) of 5 appears at 2211 cm21, shifted to high frequencies with respect to free Ph3P]] NCN, and the 31P-{1H} NMR spectrum shows two close singlet resonances (d 26.60 and 26.41), very similar to the resonance of free Ph3P]] NCN.All of these facts mean that both iminophosphorane ligands are N(nitrile)-co-ordinated, as depicted in Scheme 1. Complexes 4 and 5 can also be obtained by mutual interconversion (see Scheme 1). Thus, 4 reacts with an additional equivalent of Ph3P]] NCN to give 5 while 5 reacts with 1 equivalent of the solvate [Pd(dmba)(NCMe)2][ClO4] to give 4.The reactivity observed for the iminophosphorane is in this case somewhat similar to that reported for the phosphoylide towards C,N-orthometallated derivatives of PdII.2 Thus, Ph3- P]] C(H)CN reacted with [Pd(dmba)(NCMe)2][ClO4] (1 : 1 molar ratio) to give the dimer [{Pd[m-NCC(H)PPh3](dmba)}2][ClO4]2. To this point the behaviour of the two ligands is analogous. However, the dinuclear complex further reacts with the ylide giving [Pd(dmba){C(H)(CN)PPh3}{NCC(H)]] PPh3}][ClO4] in which the two ylides are co-ordinated in diVerent ways, one through the C atom and the other through the N atom.Its analogous derivative 5 possess the two iminophosphorane ligands co-ordinated in the same way as a result, once again, of the poor nucleophilicity of the iminic N atom. The reaction of the monosolvate complexes [Pd(dmba)- (PPh3)(THF)][ClO4] or [Pt(C–P)(PPh3)(NCMe)][ClO4] with Ph3P]] NCN (1: 1 molar ratio) gives the corresponding cationic derivatives [Pd(dmba)(PPh3)(NCN]] PPh3)][ClO4] 6 and [Pt- (C–P)(PPh3)(NCN]] PPh3)][ClO4] 7, as deduced from their elemental analyses and mass spectra. The IR spectra of 6 and 7 show a strong absorption at 2239 cm21 in both cases, and their 31P-{1H} NMR spectra show the expected resonances for the presence of the PPh3 (6) and C–P and PPh3 in trans positions (7) together with the signals attributed to the iminophosphorane (d 27.69 for 6 and 27.27 for 7). As discussed in the preceding paragraphs, these data imply an N(nitrile) coordination of the Ph3P]] NCN ligand.Similar results (N-coordination) have been obtained with the ylide Ph3P]] C(H)CN and the same precursors.2,5 Finally, the iminophosphorane Ph3P]] NCN is able to promote the cleavage of the halide bridging systems in the dinuclear complexes [{Pd(m-Cl)(dmba)}2] and [{Pt(m-Cl)(C–P)}2]. Thus, the reaction of these compounds with Ph3P]] NCN (1: 2 molar ratio) results in the formation of solids of stoichiometry [PdCl(dmba)(NCN]] PPh3)] 8 and [PtCl(C–P)(NCN]] PPh3)] 9 in accord with their elemental analyses.However, the reaction of [{Pd(m-Cl)(C–P)}2] with Ph3P]] NCN (1: 2 molar ratio) under the same conditions does not promote cleavage and the starting products are recovered. The IR spectrum of 8 shows the n(C]] ] N) absorption at 2243 cm21 and the Pd–Cl stretch at 305 cm21, while that of 9 shows the n(C]] ] N) absorption at 2233 cm21 and the Pt–Cl stretch at 292 cm21.Both facts, the increase in energies of the n(C]] ] N) absorption and the shift to higher frequency of the M]Cl stretch, point to the presence of a terminal chloride ligand and an N-bonded iminophosphorane ligand as a result of the cleavage of the halide bridging system. That is, these products are not simply mixtures of the starting materials. The NMR spectra of 8 and 9 (see Experimental section) show the expected resonances for all groups present in these molecules and also show only one set of resonances; that is, only one isomer (of the two possible) is present in solution.The sterochemistry of this isomer has been assigned as Cl-trans-to- C, as depicted in Scheme 1, on the basis of the spectroscopic data and by analogy to the splitting of the M(m-Cl)2M (M = Pd or Pt) system by N-donor ligands.5,17 Conclusion The iminophosphorane Ph3P]] NCN behaves as an N(nitrile)- donor ligand towards complexes of PdII and PtII, regardless of the electronic and steric requirements of the other ligands around the metal centre.Only in one case we have been able to co-ordinate this ligand through both the iminic N atom and the nitrilic N atom, in a bridging mode. No examples have been found with selective iminic N-co-ordination. This behaviour could be due to an extensive delocalization of the electron density along the P–NC–N system, which reduces the charge density, and thus the coordinating ability, of the iminic N atom.Experimental CAUTION: perchlorate salts of metal complexes with organic ligands are potentially explosive. Only small amounts of these materials should be prepared and handled with great caution. See ref. 26. General procedures Solvents were dried and distilled by standard methods prior to use. Elemental analyses of C, H, N were carried out on a Perkin-Elmer 2400 microanalyser. Infrared spectra (4000–200 cm21) were recorded on a Perkin-Elmer 883 spectrophotometer in Nujol mulls between polyethylene sheets, 1H (300.13 MHz) and 31P-{1H} (121.49 MHz) NMR spectra from CDCl3 or CD2Cl2 solutions at room temperature (unless otherwise stated) on a Bruker ARX-300 spectrometer; 1H spectra were referenced using the solvent signal as an internal standard and 31P-{1H} spectra were externally referenced to H3PO4 (85%).The two dimensional 1H–1H NOESY experiment for complex 4 was performed at a measuring frequency of 300.13 MHz.The data were acquired into a 512 × 1024 matrix, and then transformed into 1024 × 1024 points using a sine window in each dimension. The mixing time was 400 ms. Mass spectra (positive ion FAB) were recorded on a V.G. Autospec. spectrometer from CH2Cl2 solutions. Electrical conductivity measurements were performed in acetone solutions with concentrations of about 5 × 1024 M using a Philips PW 9509 conductivity cell. The ligand Ph3P]] NC]] ] N11 and the complexes [MCl2(NCPh)2] (M = Pd or Pt),12 [M(C–P)(NCMe)2][A] (M = Pd15 or Pt 13), [Pd(dmba)(NCMe)2][A],16 [Pd(dmba)(PPh3)(THF)][A],17 [Pt(C–P)(PPh3)(NCMe)][A] 5 (A = ClO4 2), [{M(m-Cl)(C–P)}2] (M = Pd,14 or Pt 13) and [{Pd(m-Cl)(dmba)}2] 18 were prepared according to published methods.Preparations trans-[PdCl2(N]] ] CN]] PPh3)2] 1. To a solution of [PdCl2- (NCPh)2] (0.100 g, 0.260 mmol) in 20 cm3 of acetone the iminophosphorane Ph3P]] NC]] ] N (0.157 g, 0.521 mmol) was added. After a few minutes an orange solid precipitated which was collected, washed with additional acetone (10 cm3), Et2O (25 cm3), air-dried and identified as complex 1, 0.117 g (58% yield).The acetone solution was evaporated to dryness and the residue treated with Et2O (25 cm3), giving a second fraction of 1, 0.035 g (17% yield, net yield 75%) (Found: C, 57.98; H, 3.86; N, 7.13. Calc. for C38H30Cl2N4P2Pd: C, 58.37; H, 3.87; N, 7.16%). IR (n& /cm21): 2256 (nNC), 348 (nPd–Cl). 1H NMR (CDCl3): d 7.63–7.48 (m, Ph). 31P-{1H} NMR (CDCl3): d 27.44 (s, NPPh3). [Pd(C–P)(N]] ] CN]] PPh3)2][ClO4] 2. To a solution of [Pd(C–P)- (NCMe)2][ClO4] (0.100 g, 0.169 mmol) in 20 cm3 of CH2Cl2 the iminophosphorane Ph3P]] NC]] ] N (0.102 g, 0.338 mmol) wasJ. Chem. Soc., Dalton Trans., 1998, 3745–3750 3749 added, and the resulting solution stirred at room temperature for 30 min. The solvent was then evaporated to dryness and the residue washed with Et2O (25 cm3), giving complex 2 as a white solid which was collected and air-dried, 0.104 g (55% yield) (Found: C, 63.79; H, 4.18; N, 5.00.Calc. for C59H50- ClN4O4P3Pd: C, 63.62; H, 4.52; N, 5.03%). IR (n& /cm21): 2215 (nNC). Mass spectrum (FAB1) [m/z, (%)]: 1013 (15, M1) and 711 (55% [M 2 NCNPPh3]1). 1H NMR (CDCl3, 213 K): d 7.69–6.64 (m, 42H, Ph 1 o-MeC6H4), 2.95 (s, broad, 2H, CH2Pd), 2.44 and 2.16 (2s, 6H, 2o-MeC6H4). 31P-{1H} NMR (CDCl3, 213 K): d 37.54 (s, C–P), 29.18 (s, NPPh3) and 27.76 (s, NPPh3).[Pt(C–P)(N]] ] CN]] PPh3)2][ClO4] 3. Complex 3 was synthesized in the same way as 2: [Pt(C–P)(NCMe)2][ClO4] (0.100 g, 0.147 mmol) was treated with Ph3P]] NC]] ] N (0.088 g, 0.29 mmol) in CH2Cl2 to give 3 as a white solid, 0.117 g (66% yield) (Found: C, 58.63; H, 3.85; N, 4.63. Calc. for C59H50ClN4O4P3Pt: C, 58.93; H, 4.19; N, 4.66%). IR (n& /cm21): 2228 (nNC). Mass spectrum (FAB1) [m/z, (%)]: 1102 (35%, M1) and 800 (100, [M 2 NCNPPh3]1). 1H NMR (CDCl3, 213 K): d 7.72–6.61 (m, 42H, Ph 1 o-MeC6H4), 2.99, 2.77 (AB spin system, 2H, CH2Pt, 2JHH = 16.4 Hz), 2.45 and 2.10 (2s, 6H, 2o-MeC6H4). 31P-{1H} NMR (CDCl3, 213 K): d 29.52 (s, NPPh3), 28.06 (s, NPPh3) and 16.10 (s, C–P, 1JPt–P = 4551 Hz). [{Pd(dmba)(Ï-N]] ] CN]] PPh3)}2][ClO4]2 4. Complex 4 was synthesized in the same way as 2: [Pd(dmba)(NCMe)2][ClO4] (0.084 g, 0.20 mmol) was treated with Ph3P]] NC]] ] N (0.060 g, 0.20 mmol) in CH2Cl2 to give 4 as a pale yellow solid, 0.104 g (82% yield) (Found: C, 51.85; H, 4.17; N, 6.43.Calc. for C28- H27ClN3O4PPd: C, 52.35; H, 4.23; N, 6.54%). IR (n& /cm21): 2249 (nNC). Mass spectrum (FAB1) [m/z, (%)]: 542 [100, (M/2)1]. LM/W21 cm2 mol21 = 268 (5 × 1024 M in acetone solution). 1H NMR (CDCl3, 223 K): d 7.71–7.61 (m, 15 H, Ph), 6.94 (m, 1H, C6H4), 6.85 (m, 1H, C6H4), 6.72 (m, 2H, C6H4), 3.81 (s, 2H, CH2) and 2.72 (s, 6H, NMe2). 31P-{1H} NMR (CDCl3): d 29.16 (s, NPPh3). [Pd(dmba)(N]] ] CN]] PPh3)2][ClO4] 5. Complex 5 was synthesized in the same way as 2: [Pd(dmba)(NCMe)2][ClO4] (0.200 g, 0.474 mmol) was treated with Ph3P]] NC]] ] N (0.286 g, 0.948 mmol) in CH2Cl2 to give 5 as a white solid, 0.307 g (69% yield) (Found: C, 58.90; H, 4.18; N, 7.08.Calc. for C47- H42ClN5O4P2Pd: C, 59.76; H, 4.48; N, 7.41%). IR (n& /cm21): 2211 (nNC). Mass spectrum (FAB1) [m/z, (%)]: 844 (60, M1). 1H NMR (CDCl3): d 7.69–7.55 (m, 30H, Ph), 6.88–6.80 (m, 3H, C6H4), 6.56 (m, 1H, C6H4), 3.79 (s, 2H, CH2) and 2.52 (s, 6H, NMe2). 31P-{1H} NMR (CDCl3): d 26.60 and 26.41 (2s, NPPh3). [Pd(dmba)(PPh3)(N]] ] CN]] PPh3)][ClO4] 6. To a solution of [Pd(dmba)Cl(PPh3)] (0.100 g, 0.186 mmol) in THF (20 cm3), AgClO4 (0.038 g, 0.19 mmol) was added. The resulting suspension was stirred in the dark at room temperature for 20 min and then filtered. To the freshly obtained solution of [Pd(dmba)(THF)(PPh3)][ClO4] (0.186 mmol) the iminophosphorane Ph3P]] NC]] ] N (0.056 g, 0.19 mmol) was added and the resulting solution stirred at room temperature for 15 min.The solvent was evaporated to dryness and the oily residue treated with Et2O (25 cm3), giving complex 6 as a white solid, 0.143 g (85% yield) (Found: C, 60.77; H, 4.29; N, 4.65. Calc. for C46H42ClN3O4P2Pd: C, 61.07; H, 4.68; N, 4.64%). IR (n& /cm21): 2239 (nNC). Mass spectrum (FAB1) [m/z (%)]: 804 (40, M1), 542 (45, [M 2 PPh3]1) and 502 (100, [M 2 NCNPPh3]1). 1H NMR (CDCl3): d 7.74–7.29 (m, 30 H, Ph), 6.96 (d, 1H, C6H4, 3JHH = 6.5), 6.80 (t, 1H, C6H4, 3JHH = 7.2), 6.34 (t, 1H, C6H4, 3JHH = 7.2), 6.19 (t, 1H, C6H4, 3JHH ª 4JPH = 6.8), 4.00 (d, 2H, CH2, 4JPH = 1.9) and 2.50 (d, 6H, NMe2, 4JPH = 2.6 Hz). 31P-{1H} NMR (CDCl3): d 42.11 (s, Pd–PPh3) and 27.69 (s, NPPh3).[Pt(C–P)(PPh3)(N]] ] CN]] PPh3)][ClO4] 7. Complex 7 was synthesized in the same way as 2: [Pt(C–P)(PPh3)(NCMe)][ClO4] (0.100 g, 0.111 mmol) was treated with Ph3P]] NC]] ] N (0.034 g, 0.11 mmol) in CH2Cl2 to give 7 as a white solid, 0.089 g (69% yield) (Found: C, 59.56; H, 4.23; N, 2.34.Calc. for C58H50ClN2O4P3Pt: C, 59.92; H, 4.33; N, 2.41%). IR (n& /cm21): 2353, 2340 (nNCMe), 2239 (nNC). Mass spectrum (FAB1) [m/z (%)]: 1062 (100, M1), 800 (25, [M 2 PPh3]1), 760 (65, [M 2 NCNPPh3] 1). 1H NMR (CDCl3, 213 K): d 7.71–6.76 (m, 42 H, Ph 1 o-MeC6H4), 3.14, 2.21 (AX spin system, 2H, CH2Pt, 2JHH = 15.9 Hz), 2.38, 2.26 (2s, 6H, 2o-MeC6H4). 31P-{1H} NMR (CDCl3, 213 K): d 31.55, 27.22 (AB spin system, C– P 1 PPh3, 1JPtP = 3005, 1JPtP9 = 2968, 2JPP9 = 408 Hz) and 27.27 (s, NPPh3).[PdCl(dmba)(N]] ] CN]] PPh3)] 8. To a solution of [{Pd- (m-Cl)(dmba)}2] (0.100 g, 0.181 mmol) in CH2Cl2 (20 cm3) the iminophosphorane Ph3P]] NC]] ] N (0.109 g, 0.362 mmol) was added, and the resulting solution stirred at room temperature for 15 min. The solvent was evaporated to dryness and the oily residue treated with Et2O (25 cm3), giving complex 8 as a yellow solid, 0.170 g (81% yield) (Found: C, 57.88; H, 4.49; N, 7.19.Calc. for C28H27ClN3PPd: C, 58.15; H, 4.70; N, 7.26%). IR (n& /cm21): 2243 (nNC) and 305 (nPdCl). 1H NMR (CDCl3): d 7.70–7.52 (m, 15 H, Ph), 6.94–6.84 (m, 3H, C6H4), 6.69 (m, 1H, C6H4), 3.84 (d, 2H, CH2) and 2.84 (s, 6H, NMe2). 31P-{1H} NMR (CDCl3): d 26.78 (s, NPPh3). [PtCl(C–P)(N]] ] CN]] PPh3)] 9. To a suspension of [{Pt(m-Cl)- (C–P)}2] (0.100 g, 0.094 mmol) in 20 cm3 of CH2Cl2 the iminophosphorane Ph3P]] NC]] ] N (0.057 g, 0.19 mmol) was added. The original suspension gradually dissolved and after 30 min stirring at room temperature a colorless solution was obtained.The solvent was then evaporated to dryness and the residue washed with Et2O (25 cm3), giving complex 9 as a white solid which was collected and air dried, 0.124 g (79% yield) (Found: C, 57.39; H, 4.02; N, 3.28. Calc. for C40- H35ClN2P2Pt: C, 57.45; H, 4.22; N, 3.35%). IR (n& /cm21): 2233 (nNC) and 292 (nPtCl). 1H NMR (CDCl3, 213 K): d 7.65–6.66 (m, 27H, Ph 1 o-MeC6H4), 3.42, 3.33 (AB spin system, 2H, CH2Pt, 2JHH = 16.7 Hz), 2.59, 2.31 (2s, 6H, 2o-MeC6H4). 31P-{1H} NMR (CDCl3, 213 K): d 27.07 (s, NPPh3) and 19.00 (s, C–P, 1JPtP = 4743 Hz). Crystallography Data collection. Crystals of complex 1 of adequate quality for X-ray purposes were grown from slow diVusion of n-hexane into a CH2Cl2 solution of 1 kept at 230 8C. An orange crystal was mounted on a quartz fiber and covered with epoxy. Normal procedures were used on a Enraf-Nonius CAD4 diVractometer for the determination of the unit cell constants and for the measurement of intensity data.After preliminary indexing and transformation of the cell to a conventional setting, axial photographs were taken of the a, b and c axes to verify the Laue symmetry and lattice dimensions. Accurate unit cell dimensions were determined from 25 centered reflections in the range 22.4 £ 2q £ 31.28. For intensity data collection, w–2q scans were used with Dw = 1.12 1 0.35tan q. Three monitor reflections were measured after 3 h of beam time, and the orientation of the crystal was checked after every 400 intensity measurements.Absorption corrections 27 were based on azimuthal scans of 11 reflections, 4 of which had the Eulerian angle c near 908. The other reflections used for this purpose had their bisectingposition c values distributed in the range 12–708. Structure solution and refinement. The structure was solved and developed by Patterson and Fourier methods.28 All nonhydrogen atoms were assigned anisotropic displacement parameters.The hydrogen atoms of the aromatic carbon atoms3750 J. Chem. Soc., Dalton Trans., 1998, 3745–3750 were constrained to idealized geometries and the isotropic displacement parameter of each of these hydrogen atoms was set to a value of 1.2 times the equivalent isotropic displacement parameter of its parent carbon atom. The data-toparameter ratio in the final refinement was 14.1. The structure was refined on Fo 2, and all reflections were used in the leastsquares calculations.29 The residuals and other pertinent parameters are summarized in Table 2.Crystallographic calculations were done on a Local Area VAXCluster computer (VAX/VMS V5.5-2). Data reduction was done by the program XCAD4B.30 CCDC reference number 186/1184. See http://www.rsc.org/suppdata/dt/1998/3745/ for crystallographic files in .cif format. Acknowledgements We thank the Dirección General de Enseñanza Superior (Spain) for financial support (Projects PB95-0792 and PB95- 0003-C02-01) and Professor J.Forniés for his invaluable logistical support. References 1 L. R. Falvello, S. Fernández, R. Navarro and E. P. Urriolabeitia, Inorg. Chem., 1996, 35, 3064. 2 L. R. Falvello, S. Fernández, R. Navarro and E. P. Urriolabeitia, Inorg. Chem., 1997, 36, 1136 and refs. therein. 3 L. R. Falvello, S. Fernández, R. Navarro, I. Pascual and E. P. Urriolabeitia, J. Chem. Soc., Dalton Trans., 1997, 763 and refs. therein. 4 L. R. Falvello, S. Fernández, R. Navarro and E. P. Urriolabeitia, New J. Chem., 1997, 21, 909. 5 S. Fernández, M. M. García, R. Navarro and E. P. Urriolabeitia, J. Organomet. Chem., 1998, 561, 67. 6 I. C. Barco, L. R. Falvello, S. Fernández, R. Navarro and E. P. Urriolabeitia, J. Chem. Soc., Dalton Trans., 1998, 1699. 7 R. G. Pearson, Inorg. Chem., 1973, 12, 712. 8 J. A. Davies and F. R. Hartley, Chem. Rev., 1981, 81, 79. 9 (a) A. W. Johnson, W. C. Kaska, K.A. O. Starzewski and D. A. Dixon, Ylides and Imines of Phosphorus, Wiley, New York, 1993, ch. 13 and refs. therein; (b) J. Vicente, M. T. Chicote, J. Fernández- Baeza, F. J. Lahoz and J. A. López, Inorg. Chem., 1991, 30, 3617 and refs. therein. 10 J. S. Miller, M. O. Visscher and K. G. Caulton, Inorg. Chem., 1974, 13, 1632; K. A. O. Starzewski and H. T. Dieck, Inorg. Chem., 1979, Table 2 Crystal data and structure refinement for complex 1 Formula Formula weight Crystal system Space group a/Å b/Å c/Å b/8 V/Å3 Z Dc/Mg m23 m/mm21 Reflections collected Unique reflections Data/restraints/parameters Goodness of fit Final R1, wR2 indices [I > 2s(I)] T/K l(Mo-Ka)/Å C38H30Cl2N4P2Pd 781.90 Monoclinic P21/n 9.5452(7) 12.9923(5) 13.8799(7) 93.862(6) 1717.4(2) 2 1.512 0.823 3226 3022, Rint = 0.0399 3022/0/214 1.045 0.0349, 0.0937 150(2) 0.71073 18, 3307 and refs.therein; P. Dapporto, G. Denti, G. Dolcetti and G. Ghedini, J. Chem. Soc., Dalton Trans., 1983, 779; M.J. Fernández, J. J. del Val, L. A. Oro, F. Palacios and J. Barluenga, Polyhedron, 1987, 6, 1999; K. V. Katti and R. G. Cavell, Inorg. Chem., 1989, 28, 413; Organometallics, 1989, 8, 2147; K. V. Katti, R. J. Batchelor, F. W. B. Einstein and R. G. Cavell, Inorg. Chem., 1990, 29, 808; P. ImhoV, C. J. Elsevier and C. H. Stam, Inorg. Chim. Acta, 1990, 175, 209; P. ImhoV, R. van Asselt, C. J. Elsevier, M. C. Zoutberg and C. H. Stam, Inorg. Chim. Acta, 1991, 184, 73; K.V. Katti and R. G. Cavell, Organometallics, 1991, 10, 539; K. V. Katti, B. D. Santarsiero, A. A. Pinkerton and R. G. Cavell, Inorg. Chem., 1993, 32, 5919; M. S. Balakrishna, B. D. Santarsiero and R. C. Cavell, Inorg. Chem., 1994, 33, 3079; M. W. Avis, K Vrieze, H. Kooijman, N. Veldman, A. Spek and C. J. Elsevier, Inorg. Chem., 1995, 34, 4092; A. Mahieu, A. Igau, J. Jaud and J. P. Majoral, Organometallics, 1995, 14, 944; W. K. Holley, G. E. Ryschkewitsch, A. E. Koziol and G.J. Palenik, Inorg. Chim. Acta, 1995, 239, 171; J. Li, R. McDonald and R. G. Cavell, Organometallics, 1996, 15, 1033; M. W. Avis, K. Vrieze, J. M. Ernsting, C. J. Elsevier, N. Veldman, A. Spek, K. V. Katti and C. L. Barnes, Organometallics, 1996, 15, 2376; D. M. Hankin, A. A. Danopoulos, G. Wilkinson, T. K. N. Sweet and M. B. Hursthouse, J. Chem. Soc., Dalton Trans., 1996, 4063; M. W. Avis, C. J. Elsevier, N. Veldman, H. Kooijman and A. L. Spek, Inorg. Chem., 1996, 35, 1518; R.W. Reed, B. Santarsiero and R. G. Cavell, Inorg. Chem., 1996, 35, 4292; J. Li, A. A. Pinkerton, D. C. Finnen, M. Kummer, A. Martin, F. Wiesemann and R. G. Cavell, Inorg. Chem., 1996, 35, 5684; J. Vicente, M. T. Chicote, M. A. Beswick and M. C. Ramírez de Arellano, Inorg. Chem., 1996, 35, 6592; M. W. Avis, M. E. van der Boom, C. J. Elsevier, W. J. J. Smeets and A. L. Spek, J. Organomet Chem., 1997, 527, 263; R.-Z. Ku, D.-Y. Chen, G.-H. Lee, S.-M. Peng and S.-T. Liu, Angew. Chem., Int. Ed. Engl., 1997, 36, 2631. 11 S. Bittner, M. Pomerantz, Y. Assaf, P. Krief, S. Xi and M. K. Witczak, J. Org. Chem., 1988, 53, 1. 12 G. K. Anderson and M. Lin, Inorg. Synth., 1990, 28, 60. 13 J. Forniés, A. Martín, R. Navarro, V. Sicilia and P. Villarroya, Organometallics, 1996, 15, 1826. 14 W. A. Herrmann, C. Brossmer, K. Öfele, C.-P. Reisinger, T. Priermeier, M. Beller and H. Fischer, Angew. Chem., Int. Ed. Engl., 1995, 34, 1844; L. R. Falvello, J. Forniés, A. Martín, R. Navarro, V. Sicilia and P. Villarroya, Inorg. Chem., 1997, 36, 6166. 15 P. Villarroya, Ph. D. Thesis, University of Zaragoza, 1998. 16 J. Forniés, R. Navarro and V. Sicilia, Polyhedron, 1988, 7, 2659. 17 A. J. Deeming, I. P. Rothwell, M. B. Hursthouse and L. New, J. Chem. Soc., Dalton Trans., 1978, 1490. 18 A. C. Cope and E. C. Friedrich, J. Am. Chem. Soc., 1968, 90, 909. 19 A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson and R. Taylor, J. Chem. Soc., Dalton Trans., 1989, S1. 20 M. M. Olmstead, R. R. Guimerans, J. P. Farr and A. L. Balch, Inorg. Chim. Acta, 1983, 75, 199. 21 F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen and R. Taylor, J. Chem. Soc., Perkin Trans. 2, 1987, 685. 22 L. R. Falvello, S. Fernández, M. M. García, I. Lázaro, R. Navarro and E. P. Urriolabeitia, New J. Chem., submitted. 23 J. C. Jochims, M. A. Rahman, L. Zsolnai, S. Herzberger and G. Huttner, Chem. Ber., 1983, 116, 3692. 24 R. Usón, A. Laguna, M. Laguna, P. G. Jones and G. M. Sheldrick, J. Chem. Soc., Dalton Trans., 1981, 366. 25 W. Geary, Coord. Chem. Rev., 1971, 7, 81. 26 J. Chem. Educ., 1973, 50, A335. 27 SHELXTL PLUS, Release 4.21/V, Siemens Analytical X-ray Instruments, Madison, WI, 1990. 28 G. M. Sheldrick, SHELXS 86, Acta Crystallogr., Sect. A, 1990, 46, 467. 29 G. M. Sheldrick, SHELXL 93, FORTRAN program for the refinement of crystal structures from diVraction data, Göttingen University, 1993. 30 K. Harms, personal communication, 1995. Paper 8/05147J

ISSN:1477-9226    

DOI:10.1039/a805147j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3751–3763 3751 Platinum group metal complexes of arylphosphine ligands containing perfluoroalkyl ponytails; crystal structures of [RhCl2(Á5-C5Me5){P(C6H4C6F13-4)3}] and cis- and trans- [PtCl2{P(C6H4C6F13-4)3}2] John Fawcett, Eric G. Hope,* Raymond D. W. Kemmitt, Danny R. Paige, David R. Russell and Alison M. Stuart Department of Chemistry, University of Leicester, Leicester, UK LE1 7RH Received 3rd July 1998, Accepted 23rd September 1998 The triarylphosphine ligands PPh32x(C6H4C6F13-4)x, x = 1, 2 or 3, reacted with [{RhCl2(h5-C5Me5)}2], [{RhCl- (CO)2}2], [{IrCl(COD)}2], [PdCl2(MeCN)2] or [PtCl2(MeCN)2] to yield the complexes [RhCl2(h5-C5Me5)L] 1–3, trans-[RhCl(CO)L2] 4–6, trans-[IrCl(CO)L2] 7–9, trans-[PdCl2L2] 11–13 or cis-/trans-[PtCl2L2] 14–16 respectively.Spectroscopic studies and structural studies (EXAFS for 4–9, 11–15 and X-ray single crystal for 3 and 16) indicated that the aryl groups are fairly good insulators of the electronic influence of the perfluoroalkyl substituents whilst solubility studies indicated that at least six C6F13 units are necessary for preferential perfluorocarbon solvent solubility and that the type of metal complex is important, i.e.the Vaska’s analogues 6 and 9 are perfluorocarbon solvent soluble whereas the dichloride complexes 13 and 16 are not. Studies on the addition of dioxygen to 7–10 identified a stepwise reduction in rate following the introduction of the perfluoroalkyl ponytails.One of the important areas of development in homogeneous catalysis is improvements in catalyst/product separation and catalyst recycling. Recently, the fluorous (perfluoroalkyl) biphase system (FBS) was proposed as a new approach.1 This entails anchoring a catalyst in a perfluorinated solvent by the attachment of long, perfluorinated, aliphatic side-chains (called fluorous “ponytails”) and product/catalyst separation arises from the immiscibility of many conventional organic and per- fluorinated solvents.This approach has now been adopted in a series of homogeneous catalysis processes.1–6 Recently, we reported the preparations of a range of related perfluoroalkylderivatised phosphorus(III) ligands 7 and here describe the coordination chemistry of analogues of triphenylphosphine with 1, 2 or 3 ponytails with a range of platinum group metals, including the structure of a highly unusual mixture of cis- and trans-[PtCl2L2] complexes, to assess the influence of these fluorous ponytails on the co-ordination, reactivity and solubility properties of metal complexes. Experimental Proton, 19F and 31P NMR spectroscopies were carried out on a Bruker ARX250 spectrometer at 250.13, 235.34 and 101.26 MHz or a Bruker DRX 400 spectrometer at 400.13, 376.50 and 161.98 MHz and were referenced to external SiMe4 (1H), to external CFCl3 (19F) and to external 85% H3PO4 (31P) using the high-frequency positive convention.The IR spectra were recorded on a Digilab FTS40 Fourier-transform spectrometer at 4 cm21 resolution for the complexes as Nujol mulls held between KBr discs or for solutions in CHCl3 held between KBr plates. Elemental analyses were performed by Butterworth Laboratories Ltd.. Mass spectra were recorded on a Kratos Concept 1H mass spectrometer. Rhodium and palladium K-edge and iridium and platinum LIII-edge EXAFS data were collected at the Daresbury Synchrotron Radiation Source at 2 GeV (ca. 3.2 × 10210 J) with an average current of 190 mA in transmission mode, at room temperature, on stations 7 : 1 and 9 : 2 using an order sorting Si(111) or a double-crystal Si(220) monochromator oVset to 50% of the rocking curve for harmonic rejection. Samples were diluted with dry boron nitride and mounted between Sellotape strips in 1 mm aluminium spacers. The EXAFS data treatment utilised the programs EX8 and EXCURV 92.9 Several data sets were collected for each compound in k space, and averaged to improve the signal-tonoise ratio.The pre-edge background was removed by fitting the spectrum to a quadratic polynomial, and subtracting this from the whole spectrum. The atomic contribution to the oscillatory part of the absorption spectrum was approximated using a polynomial, and the optimum function judged by minimising the intensity of chemically insignificant shells at low r in the Fourier transform. Curve fitting used single- or multiplescattering curved-wave theory with phase shifts and backscattering factors calculated using normal ab initio methods.10 Ground state potentials of the atoms were calculated using Von Barth theory and phase shifts using Hedin–Lundqvist potentials.The fits discussed below are for model data compared to raw (background-subtracted) EXAFS, and no Fourier filtering or smoothing has been applied. The distances and Debye– Waller factors were refined for all the shells, as well as the Fermi energy diVerence.The ligands,7 cis-[PtCl2(MeCN)2],11 cis-[PdCl2(MeCN)2],11 [{IrCl(COD)}2] 12 and [{PtCl2(PEt3)}2] 13 were prepared as previously described. The complexes, [(RhCl2Cp*)2] and [{RhCl- (CO)2}2] were commercial samples (Aldrich) and used as supplied. Dichloromethane, chloroform and perfluoro-1,3- dimethylcyclohexane (PP3) were each dried by refluxing over calcium hydride under dinitrogen, distilled under nitrogen and stored in closed ampoules over molecular sieves; PP3 was also frozen/pumped/thawed three times to remove all dissolved gases.Hexane was dried by refluxing over potassium metal under nitrogen, distilled and stored similarly. Toluene and diethyl ether were dried by refluxing over sodium metal under nitrogen, distilled and stored similarly. Preparations trans-[PdCl2{PPh2(C6H4C6F13-4)}2] 11. The ligand (0.348 g, 0.60 mmol) and trans-[PdCl2(MeCN)2] (0.078 g, 0.30 mmol)3752 J. Chem. Soc., Dalton Trans., 1998, 3751–3763 were stirred in refluxing dichloromethane (60 cm3) under dinitrogen for 2 h.The solvent was removed in vacuo, the product washed with light petroleum (bp 40–60 8C), recrystallised from boiling light petroleum and dried in vacuo (yellow powder; yield 0.034 g, 65%). The corresponding trans-[PdCl2L2] [L = PPh(C6H4C6H13-4)2 or P(C6H4C6H13-4)3] were made similarly. Yields 60–80%. The corresponding cis-[PtCl2{PPh2(C6H4C6F13-4)]2] 14, cis- [PtCl2{PPh(C6H4C6F13-4)2}2] 15 and a mixture of cis- and trans-[PtCl2{P(C6H4C6F13-4)3}2] 16 were made, as white powders, in the same way as their palladium analogues from [PtCl2- (MeCN)2].Yields 60–78%. [RhCl2(Á5-C5Me5){PPh2(C6H4C6F13-4)}] 1. The ligand (0.144 g, 0.25 mmol) and [(RhCl2Cp*)2] (0.77 g, 0.12 mmol) were refluxed in ethanol (60 cm3) under dinitrogen for 2 h. The solvent was removed in vacuo to yield a red solid which was recrystallised from CH2Cl2–hexane to give a fine red powder.Yield 0.15 g, 70%. The corresponding [RhCl2(h5-C5Me5)L] were made similarly. Yields 75–81%. trans-[RhCl(CO){PPh2(C6H4C6F13-4)}2] 4. The ligand (0.270 g, 0.47 mmol) and [{RhCl(CO)2}2] (0.044 g, 0.11 mmol) were stirred for 1.5 h under dinitrogen in CH2Cl2 (50 cm3). The solvent was removed in vacuo to give a yellow powder which was washed with light petroleum (bp 40–60 8C) (2 × 10 cm3). Yield 0.13 g, 66%. The corresponding trans-[RhCl(CO)L2] were made similarly. Yields 73–75%.trans-[IrCl(CO){PPh2(C6H4C6F13-4)}2] 7. The ligand (0.246 g, 0.42 mmol) and [{IrCl(COD)}2] (0.071 g, 0.11 mmol) were stirred in dry, degassed CH2Cl2 (15 cm3) under a CO atmosphere (1 atm) for 30 min at room temperature. The solvent was removed in vacuo. Dry, degassed, hexane (5 cm3) was added and the resulting yellow slurry stirred for 5 min to give a yellow solid which was quickly filtered in air and dried in vacuo. Yield 0.152 g, 70%. The corresponding trans-[IrCl(CO)L2] were made similarly. Yields 70–76%.Mixture of cis- and trans-[PtCl2(PEt3){P(C6H4C6F13-4)3}] 17. The ligand (0.68 g, 0.56 mmol) and [{PtCl2(PEt3)}2] (0.183 g, 0.28 mmol) were refluxed in dichloromethane (50 cm3) for 3 h under nitrogen. After cooling to room temperature, the solvent was removed in vacuo to give a white solid which was washed with light petroleum (bp 40–60 8C), filtered and dried in vacuo. Yield 0.56 g, 63%. Kinetic studies Method 1. The trans-[IrCl(CO)L2] [L = PPh32x(C6H4C6- F13-4)x; x = 0, 1, 2, 3 or L = P(C6H4CF3-4)3] (12.8 mmol) was dissolved in chloroform (5 cm3) which had been saturated with O2 (8.52 mmol cm23 at 20 8C)14 in a closed system.The solution was stirred and O2 was continually bubbled through the solution at room temperature. Samples (0.2 cm3) were removed at regular intervals and the relative ratio of product:reactant measured by IR spectroscopy before the sample was returned to the reaction vessel. Method 2.Three oxidation experiments were performed at 296 K with 20×, 15× and 10× molar excesses of oxygen respectively. The initial concentrations of O2 were established by mixing appropriate aliquots of saturated (O2 and N2) solutions of chloroform. The same metal complexes (6.1 mmol) were dissolved in 14 cm3 of each chloroform solution in a 10 cm3 closed round-bottomed flask submersed in a thermostatically controlled water-bath. Product : reactant ratios were measured using the procedure outlined under method 1.Solubility studies The metal complexes (0.05 g), PP3 (1.5 cm3) and [2H8]toluene (1.5 cm3) were shaken in a test-tube and allowed to settle. The phases were separated by syringe, loaded into 4 or 5 mm outside diameter NMR tubes and 31P-{1H} NMR spectra recorded for each phase. A quantitative evaluation of the partition coef- ficients was not undertaken in view of the accuracy of integration associated with routine 31P-{1H} NMR experiments.15 Crystal structure determinations [RhCl2(Á5-C5Me5){P(C6H4C6F13-4)3}] 3.Crystal data. C46- H27Cl2F39PRh, M = 1525.46, monoclinic, a = 7.911(3), b = 45.183(13), c = 15.330(2) Å, b = 93.43(1)8, U = 5470(3) Å3, T = 150 K, space group P21/c, Z = 4, Dc = 1.852 g cm23, F(000) = 2992, dimensions 0.18 × 0.14 × 0.10 mm, m(Mo-Ka) = 0.608 mm21, 21472 reflections measured (Enraf Nonius FAST area detector diVractometer), 8258 unique (Rint = 0.1835) which were used in all calculations. The final R1 = 0.079, wR(F2) (all data) was 0.182.cis- and trans-[PtCl2{P(C6H4C6F13-4)3}2] 16. Crystal data. C216H72Cl6F234P6Pt3, M = 4048.26, triclinic, a = 18.254(2), b = 19.180(2), c = 20.290(2) Å, a = 101.17(1), b = 100.47(1), g = 92.76(1)8, U = 6827.5(12) Å3, T = 120 K, space group P1� (no. 2), Z = 1, Dc = 1.969 g cm23, F(000) = 3900, dimensions 0.66 × 0.33 × 0.31 mm, m(Mo-Ka) = 1.838 mm21, 30707 reflections measured (Siemens P4 diVractometer), 27923 unique (Rint = 0.0433) which were used in all calculations.The final R1 = 0.0816, wR(F2) (all data) was 0.1818. The final Fourierdi Verence map had 11.7 and 21.2 e Å23 peaks < 1 Å from the fluorine chain atoms [i.e. highest peak at 0.89 Å from F(112)]. CCDC reference number 186/1173. See http://www.rsc.org/suppdata/dt/1998/3751/ for crystallographic files in .cif format. Results and discussion The reactions between the fluorous-derivatised triarylphosphine ligands and conventional platinum-group metal starting materials yield analogues of well established co-ordination and organometallic complexes either by the cleavage of chloridebridged dimers or by the displacement of weakly co-ordinating ligands.The reactions are relatively straightforward aVording, predominantly, air-stable metal products in reasonable (60– 80%) yields which analyse well in view of the significant numbers of perfluoroalkyl ponytails. Spectroscopic studies on the platinum complexes of the tris-derivatised ligands (see below) indicated a mixture of cis and trans isomers which we were unable to separate even by recrystallisation.For the [PtCl2- {P(C6H4C6F13-4)3}2] complex 16 we obtained single crystals suitable for X-ray crystallography and, remarkably, the cis and trans isomers co-crystallised (see below). For these complexes, the spectroscopic data reported below are, therefore, for these mixtures of isomers. The complexes were characterised by FAB mass spectrometry, IR (Table 1), 1H, 19F and 31P NMR spectroscopies (Table 2), EXAFS (Table 8) and single crystal X-ray diVraction.The mass spectra for most of the complexes showed either [M 2 Cl]1 or [M 2 CO]1 as the most intense fragments, in line with mass spectral data for many metal–phosphine complexes. Assignment of the co-ordination geometry for the [MCl2L2] (M = Pd or Pt) comes from a combination of 31P NMR data (see below) and n(M–Cl). For M = Pd, a single IR active Pd–Cl stretch suggests a trans arrangement whilst for M = Pt [except when L = P(C6H4C6F13-4)3] the observation of two Pt–Cl stretches implies a cis geometry.These results are in line with the thermodynamically favoured products in these systems established with conventional arylphosphine ligands.17 For the platinum complexes with the tris-derivatised ligand, 16 and 17, the observation of three Pt–Cl stretches oVers the first evidence of the mixture of isomers for these complexes. The rhodium(I)J.Chem. Soc., Dalton Trans., 1998, 3751–3763 3753 Table 1 Analytical, mass and IR spectral data Analysis (%) a n& (M–Cl) or Complex 1 [RhCl2(h5-C5Me5){PPh2(C6H4C6F13-4)}] 2 [RhCl2(h5-C5Me5){PPh(C6H4C6F13-4)2}] 3 [RhCl2(h5-C5Me5){P(C6H4C6F13-4)3}] 4 trans-[RhCl(CO){PPh2(C6H4C6F13-4)}2] 5 trans-[RhCl(CO){PPh(C6H4C6F13-4)2}2] 6 trans-[RhCl(CO){P(C6H4C6F13-4)3}2] 7 trans-[IrCl(CO){PPh2(C6H4C6F13-4)}2] 8 trans-[IrCl(CO){PPh(C6H4C6F13-4)2}2] 9 trans-[IrCl(CO){P(C6H4C6F13-4)3}2] 10 trans-[IrCl(CO){P(C6H4CF3-4)3}2] 11 trans-[PdCl2{PPh2(C6H4C6F13-4)}2] 12 trans-[PdCl2{PPh(C6H4C6F13-4)2}2] 13 trans-[PdCl2{P(C6H4C6F13-4)3}2] 14 cis-[PtCl2{PPh2(C6H4C6F13-4)}2] 15 cis-[PtCl2{PPh(C6H4C6F13-4)2}2] 16 cis-/trans-[PtCl2{P(C6H4C6F13-4)3}2] 17 cis-/trans-[PtCl2(PEt3){P(C6H4C6F13-4)3}] C 45.94(45.89) 40.24(39.77) 37.11(36.21) 44.32(44.33) 37.70(37.30) 33.52(33.71) 40.58(40.62) 36.26(35.68) 32.39(32.60) 43.37(43.45) 43.33(43.07) 36.45(36.49) 33.33(33.11) 40.59(40.39) 34.74(34.92) 32.54(32.04) 31.71(31.51) H 3.30(3.26) 2.19(2.32) 1.94(1.79) 1.96(2.11) 1.16(1.32) 0.92(0.92) 1.92(1.93) 0.91(1.27) 0.85(0.89) 2.00(2.02) 1.97(2.09) 1.30(1.32) 0.83(0.92) 1.71(1.96) 1.06(1.26) 0.93(0.90) 1.80(1.70) Xb 4.29(3.49) 2.21(2.57) 45.48(48.50)d 21.47(37.24)d 3.36(3.16) 2.52(2.39) 4.82(4.28) 3.04(3.02) 3.81(2.31) 4.99(5.22) 6.05(4.64) 3.92(3.14) 2.03(2.38) 33.80(34.64)d 2.40(3.01) 50.46(54.91) 45.25(46.29)d m/z c 853 [(M 2 Cl)1], 818 [(M 2 2Cl)1] 1206 [M1], 1171 [(M 2 Cl)1], 1136 [(M 2 2Cl)1] 1524 [M1], 1489 [(M 2 Cl)1], 1454 [(M 2 2Cl)1] 1298 [(M 2 CO)1], 1263 [(M 2 CO 2 Cl)1] 1934 [(M 2 CO)1], 1899 [(M 2 CO 2 Cl)1] 2535 [(M 2 CO 2 Cl)1] 1416 [(M 1 H)1], 1381 [(M 1 H 2 Cl)1] 1988 [(M 2 CO 2 Cl)1] 2688 [M1] 1188 [M1], 1160 [(M 2 CO)1] 1302[(M 1 H 2 Cl)1], 1266[(M 2 2Cl)1] 1902[(M 2 2Cl)1] 2538 [(M 2 2Cl)1] 1391 [(M 1 H 2 Cl)1] 2027 [(M 1 H 2 Cl)1], 1991 [(M 2 2Cl)1] 2699 [M1], 2663[(M 2 Cl)1], 2627[(M 2 2Cl)1] 1600 [M1], 1565 [(M 2 Cl)1] n& (CO)/cm21 2 2 2 1982 1983 1993 1959 1972 1979 1975 364 364 364 323, 303 323, 303 340, 312, 292 350, 316, 293 a Calculated values in parentheses.Microanalysis for metal complexes with perfluoroalkyl sidechains may be inaccurate due to poor combustion as described earlier.16b X = P, unless otherwise stated. c Fast-atom bombardment with m-nitrobenzyl alcohol matrix. d X = F. and iridium(I) complexes show a single band assignable as n(CO). The variation in n(CO) with electron density at the metal centre is well established and here, for each series, there is a general increase in n(CO) (Table 3) with the addition of two fluorous ponytails (one per ligand). Interestingly, the eVects of introducing a CF3 group and a C6F13 ponytail are similar.These results indicate that the phosphorus atoms and arene rings do not completely insulate the metal atoms from the highly electron withdrawing fluorous ponytails but, by comparison with the data for the closely related trans-[IrCl(CO){P(C2H4- C6F13)3}2] [n(CO) 1977 cm21] 16,20 and trans-[IrCl(CO)(PEt3)2] [n(CO) 1929 cm21],21 the arene rings are better insulators than the linear ethyl spacers.NMR Spectroscopictudies For the Cp* rhodium complexes, the 1H NMR spectra show, in addition to the aryl resonances associated with the ligands, doublets at ca. d 1.2 readily assigned to the Cp* methyl protons. The hydrogen-phosphorus coupling constant (ca. 4 Hz) is typical for this class of complex.For the mixture of cis- and trans-[PtCl2(PEt3){P(C6H4C6F13-4)3}] 17, the observation of two well resolved sets of resonances for the ethyl groups con- firms the presence of both isomers in solution. For the metal complexes containing ligands with one or two ponytails the aryl regions in the 1H NMR spectra are complicated, overlapping, multiplets due to the C6H4 and C6H5 protons but, for the metal complexes with the P(C6H4C6F13-4)3 ligand, the ortho- and meta-protons are resolved.In addition to the 3JHH interaction, the protons ortho to phosphorus show identifiable 3JHP couplings which can result in their 1H NMR resonances appearing as triplets, as seen for the ‘free’ ligand.7 The 19F NMR spectra for all the complexes show five or six, highly consistent, multiplet resonances which are similar to those for the ‘free’ ligands. The highest frequency resonances are assigned to the terminal CF3 groups. The remaining, CF2, resonances are assigned according to Scheme 1 from 19F–19F COSY experiments. The 31P NMR spectral data oVer further insights into the electronic influence of the perfluoroalkyl substituents.Except for complex 16 and 17, the 31P NMR spectra of these complexes exhibit a single resonance (M = Ir or Pd), a single resonance with satellites (M = Pt) or a doublet (M = Rh). For M = Pt (14, 15), the 1JPtP coupling constants (3653 and 3635 Hz) con- firm the cis configuration at the metal centre suggested from the IR data.For 16, the observation of two singlets with signifi- cantly diVerent coupling constants confirms the presence of both the cis and trans isomers in solution. Similarly for 17, in addition to the resonances for the triethylphosphine ligand, the observation of two multiplets assignable to co-ordinated arylphosphines, a lower frequency multiplet with a large 1JPtP and a small 2JPP coupling constant consistent with a cis- PtCl2LL9 arrangement and a higher frequency multiplet with a small 1JPtP and a large 2JPP coupling constant consistent with a trans-PtCl2LL9 arrangement, confirm the presence of both isomers in solution.We have previously shown that d(31P) is insensitive to the electronic nature of the phosphorus atom whilst a variation in 1JMP is an indicator of electronic eVects.22 A comparison of the 31P NMR spectral data with those for the analogous triphenylphosphine complexes (Table 4), particularly D31P(dcomplex 2 dfree ligand) and 1JMP, illustrates that there are general increases/decreases with the number of ponytails and that the introduction of perfluoroalkyl groups has a small influence on the electronic properties of the phosphine ligands.It has recently been argued26 that for platinum(II) complexes a decrease in 1JPtP can be correlated with the Hammett constant (sP) 27 which can be rationalised in terms of decreased PÆPt s donation which occurs with electron-withdrawing substituents.Using their correlation, sP for P(C6H4C6F13-4)3 can be estimated as 0.36, cf. 0.00 for PPh3, which should result in a weaker Pt–P bond.28 Structural studies During the course of this work we obtained crystals suitable for structural characterisation for two of the metal complexes. Scheme 1 CaF2 CbF2 CgF2 CdF2 CeF2 CF33754 J. Chem. Soc., Dalton Trans., 1998, 3751–3763 Table 2 NMR Spectral dataa Complex 1 1H 1.36 (15 H, d, JHP 4, Cp*), 7.3–8.0 (14 H, m, C6H4/C6H5) 19F-{1H} 281.2 (3F, t, 3JFF 10, CF3), 2111.5 (2F, t, 3JFF 14, CaF2), 2121.9 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.5 (2F, m, CgF2) 31P-{1H} 29.9 (d, 1JRhP 146) 2 1.37 (15 H, d, JHP 4.5, Cp*), 7.3–8.0 (13 H, m, C6H4/C6H5) 281.2 (3F, t, 3JFF 10, CF3), 2111.6 (2F, t, 3JFF 14, CaF2), 2121.8 (2F, m, CbF2), 2122.0 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 30.2 (d, 1JRhP 144) 3 b 1.19 (15 H, d, JHP 4, Cp*), 7.64 (6 H, m, m-H of C6H4P), 7.97 (6 H, t, 3JHP = 3JHH = 9, o-H of C6H4P) 281.6 (3F, t, 3JFF 10, CF3), 2111.8 (2F, t, 3JFF 14, CaF2), 2122.0 (2F, m, CbF2), 2122.3 (2F, m, CdF2), 2123.4 (2F, m, CeF2), 2126.8 (2F, m, CgF2) 29.6 (d, 1JRhP 147) 4 7.2–7.8 (14 H, m, C6H4/C6H5) 281.3 (3F, t, 3JFF 10, CF3), 2111.4 (2F, t, 3JFF 14, CaF2), 2121.8 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.5 (2F, m, CgF2) 28.8 (br)(r.t.) 30.0 (d, 1JRhP 127) (224 K) 5 7.3–7.9 (13 H, m, C6H4/C6H5) 281.3 (3F, t, 3JFF 11, CF3), 2111.6 (2F, t, 3JFF 14, CaF2), 2121.9 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 29.9 (d, 1JRhP 129) 6 7.67 (6 H, d, 3JHH 9, m-H of C6H4P), 7.81 (6 H, t, 3JHP = 3JHH 9, o-H of C6H4P) 281.4 (3F, t, 3JFF 10, CF3), 2111.7 (2F, t, 3JFF 14, CaF2), 2121.9 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.5 (2F, m, CgF2) 30.0 (d, 1JRhP 131) 7 7.2–7.8 (14 H, m, C6H4/C6H5) 281.2 (3F, t, 3JFF 10, CF3), 2111.5 (2F, t, 3JFF 14, CaF2), 2121.9 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.5 (2F, m, CgF2) 24.5 (s) 8 7.3–7.9 (13 H, m, C6H4/C6H5) 281.2 (3F, t, 3JFF 10, CF3), 2111.6 (2F, t, 3JFF 14, CaF2), 2121.8 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.2 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 24.7 (s) 9 7.63 (6 H, d, 3JHH 8, m-H of C6H4P), 7.76 (6 H, dd, 3JHP 9.5, 3JHH 8, o-H of C6H4P) 281.3 (3F, t, 3JFF 10, CF3), 2111.7 (2F, t, 3JFF 14, CaF2), 2121.9 (2F, m, CbF2), 2122.1 (2F, m, CdF2), 2123.3 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 24.7 (s) 10 7.70 (6 H, d, 3JHH 9, m-H of C6H4P), 7.84 (6 H, d, 3JHH 9, o-H of C6H4P) 263.6 (s) 24.7 (s) 11 7.3–7.8 (14 H, m, C6H4/C6H5) 281.3 (3F, t, 3JFF 10, CF3), 2111.5 (2F, t, 3JFF 12, CaF2), 2122.0 (4F, m, CbF2/CdF2), 2123.2 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 23.6 (s) 12 7.4–7.9 (13 H, m, C6H4/C6H5) 281.3 (3F, t, 3JFF 10, CF3), 2111.7 (2F, t, 3JFF 14, CaF2), 2122.0 (4F, m, CbF2/CdF2), 2123.2 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 23.5 (s) 13 7.67 (6 H, d, 3JHH 8, m-H of C6H4P), 7.80 (6 H, dd, 3JHP 10, 3JHH 8, o-H of C6H4P) 281.3 (3F, t, 3JFF 9, CF3), 2111.8 (2F, td, 3JFF 14, 4JFF 2, CaF2), 2122.0 (4F, t, 3JFF 9, CbF2/CdF2), 2123.3 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 23.3 (s) 14 7.1–7.6 (14 H, m, C6H4/C6H5) 281.2 (3F, t, 3JFF 12, CF3), 2111.6 (2F, t, 3JFF 14, CaF2), 2121.9 (2F, m, CbF2), 2122.1 (2F, t, 3JFF 14, CdF2), 2123.2 (2F, m, CeF2), 2126.5 (2F, m, CgF2) 14.6 (s, 1JPtP 3635) 15 7.4–7.9 (13 H, m, C6H4/C6H5) 281.3 (3F, t, 3JFF 9, CF3), 2111.8 (2F, t, 3JFF 14, CaF2), 2122.0 (4F, m, CbF2/CdF2), 2123.3 (2F, m, CeF2), 2126.6 (2F, m, CgF2) 14.6 (s, 1JPtP 3653) 16 (cis) c 7.82 (6 H, dd, 3JHH 8, 3JHP 11, o-H of C6H4P), 7.57 (6 H, d, 3JHH 8, m-H of C6H4P) 282.2 (3F, t, 3JFF 10, CF3), 2112.1 (2F, t, 3JFF 13, CaF2), 2122.5 (4F, m, CbF2/CdF2), 2123.8 (2F, m, CeF2), 2127.2 (2F, m, CgF2) 15.5 (s, 1JPtP 3631) 16 (trans) c 7.95 (6 H, m, o-H of C6H4P), 7.79 (6 H, d, 3JHH 8, m-H of C6H4P) 282.1 (3F, t, 3JFF 10, CF3), 2111.8 (2F, t, 3JFF 14, CaF2), 2122.5 (4F, m, CbF2/CdF2), 2123.8 (2F, m, CeF2), 2127.2 (2F, m, CgF2) 22.8 (s, 1JPtP 2719) 17 (cis) c 0.88 (9 H, dt, 3JHP 18, 3JHH 8, PCH2CH3), 1.56 (6 H, dq, 2JHP 10, 3JHH 8, PCH2CH3), 7.79 (6 H, d, 3JHH 8, m-H of C6H4P), 8.08 (6 H, dd, 3JHP 11, 3JHH 8, o-H of C6H4P) 282.1 (3F, t, 3JFF 10, CF3), 2111.8 (2F, t, 3JFF 14, CaF2), 2122.3 (2F, m, CbF2), 2122.6 (2F, m, CdF2), 2123.8 (2F, m, CeF2), 2127.2 (2F, m, CgF2) 10.1 (1P, d, 1JPtP 3285, 2JPP 15, PEt3), 15.9 (1P, d, 1JPtP 3830, 2JPP 15, aryl P) 17 (trans) c 1.11 (9 H, m, PCH2CH3), 1.91 (6 H, m, PCH2CH3), 7.74 (6 H, d, 3JHH 8, m-H of C6H4P), 7.9 (6 H, vt, 3JHH 8, 3JHP 10, o-H of C6H4P) 282.1 (3F, t, 3JFF 10, CF3), 2111.7 (2F, t, 3JFF 15, CaF2), 2122.3 (2F, m, CbF2), 2122.6 (2F, m, CdF2), 2123.8 (2F, m, CeF2), 2127.2 (2F, m, CgF2) 18.1 (1P, d, 1JPtP 2640, 2JPP 472, PEt3), 24.2 (1P, d, 1JPtP 2401, 2JPP 427, aryl P) a Spectra recorded at room temperature in CDCl3 unless otherwise stated. Data given as d (intensity, multiplicity, J/Hz, assignment).b In CD2Cl2. c In (CD3)2CO. Table 3 Variation of n& (CO)a in trans-[MCl(CO)L2] (M = Rh or Ir) with the number of perfluoroalkyl groups Ligand PPh3 PPh2(C6H4C6F13-4) PPh(C6H4C6F13-4)2 P(C6H4C6F13-4)3 P(C6H4CF3-4)3 Rh 1965 b 1982 1983 1993 1990 b Ir 1953 c 1959 1972 1979 1975 a n& (CO)/cm21. Recorded as Nujol mulls, unless otherwise stated. b In CH2Cl2 solution.Data taken from ref. 18. c Data taken from ref. 19. These represent only the third and fourth structural characterisations of metal complexes with fluorous ponytails 16,20,29 and, in spite of the interest in the structural characteristics of perfluoro-aliphatic derivatives,30,31 relatively few compounds with five or six CF2 units have been crystallographically characterised. 32,33 Here, metal complexes with three and nine independent fluorous ponytails have been determined.In [RhCl2(h5-C5Me5){P(C6H4C6F13-4)3}] 3 (Figs. 1 and 2; Table 5) the asymmetric unit cell contains one discrete molecule which adopts an archetypal piano-stool geometry. Considering the C5Me5 ring as a single co-ordination centre represented by its centroid, the rhodium co-ordination might be described as very distorted tetrahedral in which the bulky Cp* ligand forcesJ. Chem. Soc., Dalton Trans., 1998, 3751–3763 3755 Table 4 31P-{1H} NMR data for related arylphosphine metal complexes a [RhCl2(h5-C5Me5)L] trans-[RhCl(CO)L2] trans-[IrCl(CO)L2] trans-[PdCl2L2] cis-[PtCl2L2] Ligand PPh3 PPh2(C6H4C6F13-4) PPh(C6H4C6F13-4)2 P(C6H4C6F13-4)3 P(C6H4CF3-4)3 D(31P) 35.2 34.9 35.6 35.6 — 1JMP 144 146 144 147 — D(31P) a 29.5 b 35.0 b 35.4 36.0 36.8 1JMP 125 127 129 131 121 D(31P) 28.5 c 29.5 30.2 30.7 30.1 D(31P) 33.5 d 28.0 28.9 29.3 — D(31P) 18.9 e 19.6 20.0 21.5 19.9 f 1JMP 3676 e 3653 3635 3631 3648 f a D(31P) = dmetal complex 2 dfree ligand/ppm, 1JMP/Hz.b Spectrum recorded at 224 K. c Data taken from ref. 23. d Data taken from ref. 24. e Data taken from ref. 25. f Data taken from ref. 26. the interligand angles between the other ligands close to 908. Surprisingly, there have been relatively few structural characterisations on [RhCl2Cp*L] [where L = phosphorus(III) donor ligand] complexes.22,34–36 For 3, the Rh–CCp* and Rh–Cl distances are within the range defined by this group of complexes, and the Rh–P distance at 2.332(3) Å is close to the longest [2.327(5) Å, L = PPh2(C2H4SiMe2OH)], but this is not particularly surprising since the other crystallographically characterised complexes include trialkyl phosphite and triarylphosphonite ligands.Similarly, the slight asymmetry in the P–Rh–Cl and Cl–Rh–Cl bond angles is mirrored by this group of complexes and it can be concluded that, for this complex, the perfluoroalkyl groups have a negligible influence on the metal co-ordination environment in line with the NMR data.The perfluoroalkyl groups radiate away from the metal centre and, although the terminal CF2CF3 units have large thermal ellipsoids as a result of motion even at 150 K, in marked contrast to the previously characterised iridium and rhodium complexes with the P(C2H4C6F13)3 ligand,16,20,29 this complex does not suVer from any disorder. The perfluoroalkyl groups experience the usual steric congestion resulting in twisting of the CF2 units with respect to each other.There are two types of C6F13 ponytails. The first, extending from C38, adopts a consistently trans and staggered conformation with torsion angles between 162 and 1668 resulting in a linear ponytail. Here, the C–C bond lengths are similar and unremarkable. The others include a pseudo boat, cis conformation in the middle of the perfluoroalkyl chain where the tor- Fig. 1 Molecular structure of [RhCl2(h5-C5Me5){P(C6H4C6F13-4)3}]. Displacement ellipsoids are shown at the 30% probability level. The H atoms are omitted for clarity.sion angle is much smaller (38 or 548) causing the ponytail to kink. In these chains there is an alternation of long and short C–C distances which appear to be associated with the strained conformation. In an extended view of the lattice (Fig. 2), the preference for the perfluoroalkyl chains to align is illustrated and it is this interaction which appears to be important in holding the structure together.This strong preference has been highlighted previously 16,20,29 and appears to arise from electrostatic interactions between fluorous fragments which underpin the fluorous biphase concept. Here, there are a number of short F ? ? ? F contacts between adjacent molecules. Most of the interactions involve the chains radiating from C29 holding pairs of molecules together and these chains point towards chains radiating from C17 and both sets of chains appear to have to kink to accommodate this arrangement.As indicated by the spectroscopic data, [PtCl2{P(C6H4C6F13- 4)3}2] 16 exists in the solid state and in solution as a mixture of cis and trans isomers. Single crystals of this complex were grown by slow evaporation of an acetone solution over 4 weeks. The very large unit cell contains 465 non-hydrogen atoms and the thermal motion of the perfluoroalkyl groups necessitated collecting the data at 120 K. The structure is highly unusual since it possesses both the cis and the trans isomers of the complex in the same unit cell in a 2 : 1 ratio.The asymmetric unit has one cis-molecule in a general position of the space group P1� and half of a trans-molecule with the platinum atom located at Fig. 2 Extended structure of [RhCl2(h5-C5Me5){P(C6H4C6F13-4)3}] showing short intramolecular interactions.3756 J. Chem. Soc., Dalton Trans., 1998, 3751–3763 Table 5 Selected bond lengths (Å), angles (8), torsion angles (8) and non-bonded short interchain distances (Å) with estimated standard deviations (e.s.d.s) in parentheses for [RhCl2(h5-C5Me5){P(C6H4C6F13-4)3}] a 3 Rh(1)–C(1) Rh(1)–C(3) Rh(1)–C(5) Rh(1)–Cl(2) Cp† ? ? ? Rh(1) av.C(cp*)–C(cp*) av. C(ring)–C(ring) C(17)–C(18) C(19)–C(20) C(21)–C(22) C(30)–C(31) C(32)–C(33) C(41)–C(42) C(43)–C(44) C(45)–C(46) P(1)–Rh(1)–Cl(1) Cl(1)–Rh(1)–Cl(2) C(23)–P(1)–Rh(1) C(11)–P(1)–C(23) C(23)–P(1)–C(35) Cl(1)–Rh(1) ? ? ? Cp† C(14)–C(17)–C(18)–C(19) 2 C(18)–C(19)–C(20)–C(21) 2 C(26)–C(29)–C(30)–C(31) 2 C(30)–C(31)–C(32)–C(33) C(38)–C(41)–C(42)–C(43) C(42)–C(43)–C(44)–C(45) F(17A) ? ? ? F(34B) F(20A) ? ? ? F(43B) F(30A) ? ? ? F(31B) F(31A) ? ? ? F(33A) 2.215(9) 2.164(9) 2.179(9) 2.398(3) 1.824 1.431 1.38 1.47(2) 1.46(2) 1.45(3) 1.56(2) 1.75(3) 1.544(14) 1.52(2) 1.51(2) 88.28(9) 93.95(9) 113.0(2) 102.3(3) 106.4(3) 120.5 167.8 171.4 176.1 176.1 165.9 163.3 2.791 2.697 2.789 2.880 Rh(1)–C(2) Rh(1)–C(4) Rh(1)–Cl(1) Rh(1)–P(1) av.P–C av. C(cp*)–C(Me) av.C(ring)–C(tail) C(18)–C(19) C(20)–C(21) C(29)–C(30) C(31)–C(32) C(33)–C(34) C(42)–C(43) C(44)–C(45) av. C–F P(1)–Rh(1)–Cl(2) C(11)–P(1)–Rh(1) C(35)–P(1)–Rh(1) C(11)–P(1)–C(35) P(1)–Rh(1) ? ? ? Cp† Cl(2)–Rh(1) ? ? ? Cp† C(17)–C(18)–C(19)–C(20) C(19)–C(20)–C(21)–C(22) C(29)–C(30)–C(31)–C(32) C(31)–C(32)–C(33)–C(34) C(41)–C(42)–C(43)–C(44) C(43)–C(44)–C(45)–C(46) F(19B) ? ? ? F(33B) F(29B) ? ? ? F(34A) F(31A) ? ? ? F(31A) 2.204(9) 2.201(8) 2.393(2) 2.332(3) 1.831 1.498 1.517 1.63(2) 1.65(2) 1.499(14) 1.30(2) 1.16(3) 1.547(14) 1.57(2) 1.375 86.00(9) 118.2(2) 113.0(2) 102.5(3) 134.7 122.0 253.8 178.9 237.8 151.8 162.6 165.5 2.662 2.815 2.701 a Cp† denotes the cyclopentadienyl centroid.Fig. 3 Molecular structure of trans-[PtCl2{P(C6H4C6F13-4)3}2]. Details as in Fig. 1. Primed atoms generated by symmetry (2x, 2y, 1 2 z). the special position (0, 0, ��� ). Selected bond lengths and bond angles are given in Table 6 and the molecular structures of the two isomshown in Figs. 3 and 4. For the trans isomer the trans geometry and planarity of the “PtCl2P2” unit is imposed by crystallographic constraints. For the cis isomer the geometry around the metal atom is very similar to that for cis- [PtCl2(PPh3)2],37 i.e. there is slight asymmetry in the Pt–P and Pt–Cl bond lengths and the bond distances and angles atJ. Chem. Soc., Dalton Trans., 1998, 3751–3763 3757 Table 6 Selected bond lengths (Å), angles (8) and torsion angles (8) with estimated standard deviations (e.s.d.s) in parentheses for the mixture of cis- and trans-[PtCl2{P(C6H4C6F13-4)3}2] 16 cis Pt(1)–Cl(1) Pt(1)–P(1) C(7)–C(8) C(9)–C(10) C(11)–C(12) C(20)–C(21) C(22)–C(23) C(31)–C(32) C(33)–C(34) C(35)–C(36) C(44)–C(45) C(46)–C(47) C(55)–C(56) C(57)–C(58) C(59)–C(60) C(68)–C(69) C(70)–C(71) C(71)–C(729) P(1)–Pt(1)–P(2) P(1)–Pt(1)–Cl(1) P(2)–Pt(1)–Cl(2) C(1)–P(1)–Pt(1) C(25)–P(1)–Pt(1) C(49)–P(2)–Pt(1) C(1)–P(1)–C(13) C(13)–P(1)–C(25) C(37)–P(2)–C(61) C(4)–C(7)–C(8)–C(9) 2 C(8)–C(9)–C(10)–C(11) C(16)–C(19)–C(20)–C(21) 2 C(20)–C(21)–C(22)–C(23) 2 C(28)–C(31)–C(32)–C(33) C(32)–C(33)–C(34)–C(35) C(40)–C(43)–C(44)–C(45) C(44)–C(45)–C(46)–C(47) 2 C(52)–C(55)–C(56)–C(57) C(56)–C(57)–C(58)–C(59) C(64)–C(67)–C(68)–C(69) 2 C(68)–C(69)–C(70)–C(71) 2 2.349(3) 2.254(3) 1.55(2) 1.56(2) 1.61(2) 1.71(2) 1.64(2) 1.57(2) 1.57(2) 1.57(2) 1.54(2) 1.57(2) 1.57(2) 1.54(2) 1.55(2) 1.55(2) 1.53(2) 1.43(2) 97.73(9) 89.97(9) 86.36(9) 115.7(3) 112.7(3) 114.8(3) 100.1(5) 109.7(5) 103.7(4) 171.68 251.99 169.84 168.81 174.25 167.21 175.81 179.77 165.27 162.30 171.72 167.30 Pt(1)–Cl(2) Pt(1)–P(2) C(8)–C(9) C(10)–C(11) C(19)–C(20) C(21)–C(22) C(23)–C(24) C(32)–C(33) C(34)–C(35) C(43)–C(44) C(45)–C(46) C(47)–C(48) C(56)–C(57) C(58)–C(59) C(67)–C(68) C(69)–C(70) C(71)–C(72) C(79)–C(80) P(1)–Pt(1)–Cl(2) P(2)–Pt(1)–Cl(1) Cl(2)–Pt(1)–Cl(1) C(13)–P(1)–Pt(1) C(37)–P(2)–Pt(1) C(61)–P(2)–Pt(1) C(1)–P(1)–C(25) C(37)–P(2)–C(49) C(49)–P(2)–C(61) C(7)–C(8)–C(9)–C(10) 2 C(9)–C(10)–C(11)–C(12) 2 C(19)–C(20)–C(21)–C(22) C(21)–C(22)–C(23)–C(24) 2 C(31)–C(32)–C(33)–C(34) C(33)–C(34)–C(35)–C(36) C(43)–C(44)–C(45)–C(46) 2 C(45)–C(46)–C(47)–C(48) C(55)–C(56)–C(57)–C(58) C(57)–C(58)–C(59)–C(60) C(67)–C(68)–C(69)–C(70) 2 C(69)–C(70)–C(71)–C(72) 2.328(2) 2.271(3) 1.58(2) 1.46(2) 1.47(2) 1.40(2) 1.45(2) 1.56(2) 1.55(2) 1.58(2) 1.56(2) 1.56(2) 1.56(2) 1.58(2) 1.54(2) 1.52(2) 1.43(2) 1.59(2) 174.28(9) 171.01(9) 86.30(9) 113.7(3) 110.1(3) 119.5(3) 103.8(4) 104.5(4) 102.6(4) 168.01 162.73 250.06 165.00 169.55 164.06 179.81 175.28 159.43 163.85 168.02 248.57 trans Pt(2)–Cl(3) C(80)–C(81) C(82)–C(83) C(91)–C(92) C(93)–C(94) C(95)–C(96) C(104)–C(105) C(106)–C(107) P(3)–Pt(2)–Cl(3) P(3)–Pt(2)–Cl(39) C(73)–P(3)–Pt(2) C(97)–P(3)–Pt(2) C(85)–P(3)–C(97) C(76)–C(79)–C(80)–C(81) C(80)–C(81)–C(82)–C(83) C(88)–C(91)–C(92)–C(93) 2 C(92)–C(93)–C(94)–C(95) C(100)–C(103)–C(104)–C(105) 2 C(104)–C(105)–C(106)–C(107) 2 2.331(2) 1.54(2) 1.62(2) 1.57(2) 1.45(2) 1.67(2) 1.55(2) 1.53(2) 86.66(9) 93.34 108.9(3) 115.3(3) 101.5(5) 166.96 160.98 172.04 257.18 171.68 163.73 Pt(2)–P(3) C(81)–C(82) C(83)–C(84) C(92)–C(93) C(94)–C(95) C(103)–C(104) C(105)–C(106) C(107)–C(108) P(3)–Pt(2)–P(39) Cl(3)–Pt(2)–Cl(39) C(85)–P(3)–Pt(2) C(73)–P(3)–C(85) C(73)–P(3)–C(97) C(79)–C(80)–C(81)–C(82) C(81)–C(82)–C(83)–C(84) C(91)–C(92)–C(93)–C(94) C(93)–C(94)–C(95)–C(96) 2 C(103)–C(104)–C(105)–C(106) 2 C(105)–C(106)–C(107)–C(108) 2 2.330(3) 1.58(2) 1.40(2) 1.63(2) 1.43(2) 1.58(2) 1.58(2) 1.54(2) 180.0 180.0 116.2(3) 105.8(5) 108.7(5) 168.37 63.97 260.57 160.52 171.68 163.66 Non-bonded short interchain distances (Å) for the mixture F(7) ? ? ? F(96) F(13) ? ? ? F(69) F(27) ? ? ? F(33) F(41) ? ? ? F(78) F(43) ? ? ? F(65) F(63) ? ? ? F(95) F(769) ? ? ? F(108) F(99) ? ? ? F(104) F(101) ? ? ? F(102) F(103) ? ? ? F(104) 2.833 2.802 2.745 2.819 2.845 2.782 2.597 2.603 2.836 2.098 F(21) ? ? ? F(21) F(16) ? ? ? F(91) F(32) ? ? ? F(769) F(41) ? ? ? F(789) F(53) ? ? ? F(82) F(71) ? ? ? F(99) F(89) ? ? ? F(101) F(101) ? ? ? F(104) F(102) ? ? ? F(104) F(110) ? ? ? F(115) 2.875 2.831 2.720 2.828 2.696 2.783 2.852 2.736 2.147 2.7543758 J.Chem. Soc., Dalton Trans., 1998, 3751–3763 Fig. 4 Molecular structure of cis-[PtCl2{P(C6H4C6F13-4)3}2]. Details as in Fig. 1. Disorder indicated by dashed bonds. platinum for both cis complexes in the essentially planar “PtCl2P2” cores are virtually identical. In particular, the insignificant diVerences between the platinum–phosphorus bond lengths for 16 [2.254(3) and 2.271(3) Å] and those for cis- [PtCl2(PPh3)2] [2.251(2) and 2.265(2) Å] suggest the introduction of the electron withdrawing perfluoroalkyl chains appears to have little influence on the length of the Pt–P bond.Phosphorus( 1) is slightly closer to platinum than phosphorus(2) and, consequently, has a larger influence on the angles at the metal centre. The Pt–P bond lengths in the cis isomer are shorter than that for the trans isomer, in line with the diVerence in the trans influence between chloride and phosphorus.The P–C, C–C and C–F bond lengths in the nine unique substituents on the phosphorus atoms are very similar and are entirely consistent with co-ordinated phosphines 38 and highly fluorinated aliphatic compounds.16,20,29–31 It has been suggested that the variation in Pt–P–C angles reflects the degree of overcrowding at the metal centre for these square planar [PtCl2- (phosphine)2] complexes.Here, only phosphorus(2) shows any significant asymmetry in the Pt–P–C angles (which is translated into a longer Pt–P distance), and the largest Pt–P–C angle is 38 smaller than that for cis-[PtCl2(PPh3)2] 37 implying that the fluorous-ponytail-derivatised ligand has less steric bulk than its protio-congener. The perfluoroalkyl groups radiate away from the metal centres and one chain, that radiating from C64, experiences disorder of the CF3 group and a fluorine atom on C71, the two arrangements occurring in a 50 : 50 ratio. This disorder is similar to that observed for trans-[MCl(CO){P(C2H4- C6F13)3}2] (M = Rh or Ir).16,20,29 The nine unique fluorous ponytails adopt three diVerent conformations (Fig. 5). Four (radiating from C28, C40, C52 and C100) adopt the trans and staggered conformation (torsion angles vary from 159 to 1808) giving virtually linear perfluoroalkyl chains.Four (radiating from C4, C16, C64 and C76) contain one cis conformation (torsion angles of 52, 50, 49 and 648 respectively), although the location of this varies along the ponytail, aVording kinked ponytails. These structural motifs are the same as those seen for [RhCl2(h5-C5Me5){P(C6H4C6F13-4)3}] and a similar, although not as varied, distribution of C–C bond lengths is also present. The ponytail radiating from C88 is unique. Here, two adjacent cis conformations (torsion angles of 61 and 578) result in a perfluoroalkyl unit which bends at virtually 908 in the middle.The extended view of this structure is not as straightforward as that for 3. The perfluoroalkyl chains still lie in layers within the structure but are criss-crossed. Short interchain interactions, similar to those seen for 3, hold the extended structure together. These interactions are predominantly associated with the bent perfluoroalkyl chains, however there are also interactions with the near-linear chains.The unique double-bent chain (radiating from C88) is interesting since it points directly at an identical chain eminating from an adjacent molecule (Fig. 6). The double bending is, therefore, necessary to accommodate these two chains. Even with this arrangement this area is very crowded with five short intermolecular interactions between the terminal C3F7 units including two extremely short distances (2.098 and 2.147 Å) between fluorines on the terminal CF3 groups.However, these short interactions do not appear to influence the C–C bond lengths which are as regular as those for the straight ponytails. One of the important aims of this work was to evaluate the eVect of the perfluoroalkyl substituents on the donor properties of these ligands and, consequently, on the properties of their co-ordination compounds. As indicated above, the spectroscopic probes do not oVer a clear-cut view of these eVects. We have been unable to obtain single crystals suitable for structural determination for all of the co-ordination compounds described in this work, however an assessment of the structural impact of the ponytails can be made from variations in the metal’s first co-ordination sphere.We have shown that the EXAFS technique can be a valuable probe of the metal coordination sphere which does not require single crystals.39 Hence, we have collected and modelled the metal-edge EXAFS data for 4–9, 11–15, together with structurally characterised analogues as models to test the reliability of our data collection and treatment (Tables 7 and 8).For each complex, transmission EXAFS data were collected to k = 15 Å21 (k = photoelectronJ. Chem. Soc., Dalton Trans., 1998, 3751–3763 3759 Fig. 5 Nine unique perfluoroalkyl groups present in [PtCl2{P(C6H4C6F13-4)3}2]. wave vector) beyond the edge but, due to poor signal-to-noise ratio at high k, the data were, typically, truncated at k = 12.5 Å21.At least five data sets were collected for each complex, averaged and the data multiplied by k3 to compensate for dropo V in intensity at higher k. No smoothing or Fourier filtering was applied and the fits discussed below are compared with the averaged raw (background-subtracted) EXAFS data. Throughout the analyses we have iterated the distances and Debye– Waller factors for each shell and the Fermi energy; each shell was added stepwise and the best fits tested for statistical signifi- cance in the usual way.42,43 For the [MCl2L2] (M = Pd or Pt) complexes the data were modelled to a two-shell model (2Cl, 2P).Although the Fourier transforms revealed features associated with longer nonbonding interactions and even though modelling these distances to carbon atoms in the backbone of the ligands resulted in significant reductions in the Fit Index and R factors, since we are interested in only the metal co-ordination environment we follow convention 44 in not modelling these longer interactions.We note that fitting two shells at similar distance may be dangerous. However, the validity of the fits we obtained can be demonstrated by the level of correlation between the two shells (<0.7) and our attempts to fit the data to chemically unreasonable single- or two-shell (3Cl, 1P and 1Cl, 3P) models which gave uncharacteristic Debye–Waller factors, unreasonable bond lengths and markedly poorer fits.The very good agreement between the EXAFS and crystallographic data for the triphenylphosphine complexes (Table 7) illustrates the reliability and value of the application of EXAFS to this system. Furthermore, these data for the model complexes compare very well with those for the perfluoroalkyl-derivatised metal complexes (Tables 6 and 8) indicating that, within the accuracy of the3760 J. Chem. Soc., Dalton Trans., 1998, 3751–3763 EXAFS analysis, the co-ordination environment at the metal centre for 11–16 is not significantly aVected by the introduction of the perfluoroalkyl substituents.A similar analysis procedure for the trans-[MCl(CO)L2] (M = Rh or Ir) used a three-shell model (1C, 3P, 1O) with multiple scattering for the non-bonded oxygen shell with a fixed M–C–O angle of 1808 (Fig. 7). For this set of complexes we found that in a four-shell model (1C, 2P, 1Cl, 1O) the phosphorus and chlorine shells were very strongly correlated which significantly reduced the confidence in the modelling.Hence, we adopted the three-shell model in which the heavier backscatterers were modelled using a single shell of three phosphorus Fig. 6 Unusual interactions between the unique perfluoroalkyl chains in trans-[PtCl2{P(C6H4C6F13-4)3}2]. atoms. For the model compounds (Table 7) this gave entirely reasonable, averaged, distances and, again, the close agreement between the data for these model compounds and the perfluoroalkyl-derivatised species indicates that the perfluoroalkyl substituents have, within the accuracy of the EXAFS analysis, little influence on the metal co-ordination environment.These results indicate that the aryl spacer group is a good insulator of the electronic influence of the perfluoroalkyl ponytail and that any eVects which are still present can be balanced by other factors at the metal centres. Kinetic studies An alternative way to assess the influence of ligands is to study the reactions of the metal complexes. Here, we have, initially, chosen to use the well established rate of O2 addition to the [IrCl(CO)L2] 7–9 complexes as a probe.45 For comparison purposes we have also investigated the rates for the analogous PPh3 and P(C6H4CF3-4)3 10 complexes under the same conditions.In these studies we have examined the variation in substrate/ product concentrations using IR spectroscopy in the carbonyl stretching region and have chosen chloroform as the solvent.This is not ideal since the solubility of oxygen in CHCl3 is relatively low, but it is the only solvent in which all the complexes would dissolve at a high enough concentration to oVer accurate analysis of the data. Two experimental methods have been adopted (Experimental section). In the first the well established pseudo-first-order kinetics for this oxidation has been created by bubbling oxygen through an oxygen-saturated solution of the metal complex throughout the experiment.The results (Fig. 8) indicate a sequential drop in the rate of oxidation with the stepwise introduction of pairs of ponytails. The half-lives are calculated to be PPh3 62 min (which is comparable to that observed previously45), PPh2(C6H4C6F13-4) 132 min, PPh(C6H4C6F13-4)2 380 min, P(C6H4C6F13-4)3 1395 min and P(C6H4CF3-4)3 980 min. A comparable drop in reaction rate has been reported between trans-[IrCl(CO)(PPh3)2] and trans- [IrCl(CO){P(C2H4C6F13)3}2] from NMR studies in THF solution. 20 At first glance it is tempting to attribute this variation Fig. 7 Background–subtracted EXAFS ( ——, experimental × k3; – – – –, curved–wave theory × k3) and the Fourier transform (——, experimental; – – – –, theoretical) for (a) trans-[IrCl(CO)(PPh3)2] and (b) trans-[IrCl(CO){P(C6H4C6F13-4)3}2]; k is the photoelectron wave vector and r is the radial distance from the absorbing atom.J. Chem. Soc., Dalton Trans., 1998, 3751–3763 3761 Table 7 Rhodium and palladium K-edge and iridium and platinum LIII-edge EXAFS data for model compoundsa Complex trans-[RhCl(CO)L2] g trans-[IrCl(CO)L2] i trans-[PdCl2L2] j cis-[PtCl2L2] k d(M–C)b/Å 1.807(5) 1.791(13) —— d(M–C)/Å 1.799(3) 1.791(5) —— 2s2 c/Å2 0.001(1) 0.004(1) —— d(M–P)b/Å 2.302(1) 2.330(1) 2.337 2.258 d(M–P)/Å 2.326(2) h 2.337(2) h 2.373(4) 2.269(3) 2s2 c/Å2 0.008(1) 0.008(1) 0.003(1) 0.004(1) d(M–Cl)b/Å 2.356(2) 2.382(3) 2.290 2.345 d(M–Cl)/Å —— 2.272(3) 2.362(2) 2s2 c/Å2 —— 0.003(1) 0.003(1) d(M ? ? ? O)b/Å 2.959 2.952 —— d(M ? ? ? O)/Å 2.928(4) 2.937(5) —— 2s2 c/Å2 0.004(1) 0.004(1) —— EFd 21.75(32) 29.29(40) 22.72(30) 211.51(42) F.I.e 3.17 5.22 1.04 3.57 Rf 22.11 21.02 13.37 23.82 a Standard deviations in parentheses.The systematic errors in bond distances arising from the data collection and analysis procedures are ca. ±0.02 Å for the first co-ordination shells and ca. ±0.04 Å for subsequent shells. b Distance taken from crystallographic studies.c Debye–Waller factor. d Fermi energy. e Fit Index = Si[(cT 2 cE)ki 3]2. f R = [Ú(cT 2 cE)k3dk/ÚcEk3dk] × 100%. g L = P(C2H4C6F13)3; Crystallographic data taken from ref. 29. h M–P and M–Cl modelled to a single shell of 3 × P atoms (see text). i L = PPh3; crystallographic data taken from ref. 40. j L = PPh3; crystallographic data taaken from ref. 41. k L = PPh3; crystallographic data taken from ref. 37.3762 J. Chem. Soc., Dalton Trans., 1998, 3751–3763 Table 8 Rhodium and palladium K-edge and iridium and platinum LIII-edge EXAFS data for complexes 4–14a Complex 456789 10 11 12 13 14 d(M–C)/Å 1.806(3) 1.825(4) 1.813(3) 1.802(6) 1.820(5) 1.816(5) ————— 2s2/Å2 0.002(1) 0.002(1) 0.001(1) 0.013(1) 0.004(1) 0.002(1) ————— d(M–P)/Å 2.346(2) 2.348(2) 2.343(2) 2.343(1) 2.338(1) 2.337(2) 2.370(3) 2.362(5) 2.359(4) 2.268(5) 2.281(6) 2s2/Å2 0.009(1) 0.005(1) 0.010(1) 0.009(1) 0.006(1) 0.006(1) 0.001(1) 0.003(1) 0.002(1) 0.006(1) 0.006(1) d(M–Cl)/Å —————— 2.259(3) 2.270(3) 2.261(3) 2.350(5) 2.342(4) 2s2/Å2 —————— 0.002(1) 0.003(1) 0.002(1) 0.006(1) 0.008(2) d(M ? ? ? O)/Å 2.932(4) 2.952(4) 2.944(14) 2.934(2) 2.944(5) 2.944(5) ————— 2s2/Å2 0.006(1) 0.006(1) 0.009(1) 0.001(1) 0.004(1) 0.004(1) ————— EF 20.42(29) 21.82(43) 21.32(37) 27.55(24) 28.76(26) 29.26(34) 22.05(34) 23.63(32) 24.21(33) 213.66(44) 213.95(62) F.I. 2.60 2.91 2.62 2.80 3.45 3.31 1.12 0.84 0.93 3.18 2.50 R 18.42 19.90 17.98 20.45 22.16 22.19 13.76 10.90 11.47 22.87 19.48 a Details as in Table 7.solely to the electronic influence of the perfluoroalkyl substituents. However, the spectroscopic and structural data (see above) suggest that the introduction of the ponytails in these aryl ligands has a limited eVect; the IR data for [IrCl(CO)(O2){P(C6H4- C6F13-4)3}2] [(n(CO) 2025 and n(O–O) 861 cm21)] are very similar to those for the PPh3 complex (2015 and 858 cm21 respectively).46 In an attempt to confirm that these reactions are genuinely pseudo-first-order a second set of experiments were undertaken (method 2).Here, the rates of reaction were investigated at fixed (closed-system) excess oxygen concentrations (20 : 1, 15 : 1 and 10 : 1). For the triphenylphosphine complex the well established pseudo-first-order kinetics was confirmed under our conditions and the calculated second order rate constant (0.02 M21 s21 in chloroform at 23 8C) is similar to that calculated previously.47 However, the introduction of only two ponytails under this experimental set-up had an enormous eVect on the rate of oxidation.In contrast to the results Fig. 8 Rates of reaction of trans-[IrCl(CO)L2] (2.566 mM) and O2 (8.52 mM) in CHCl3 at room temperature. Fig. 9 Plots of ln [a/(a 2 x)] vs. time for the reactions of trans- [IrCl(CO)L2] [L = PPh3 or PPh2(C6H4C6F13-4); 0.436 mM] and O2 (8.710 mM) in CHCl3 at room temperature. obtained under method 1, 50% conversion was only reached after ca. 24 h (cf. t2� 1 = 132 min) and analysis did not reveal pseudo-first-order kinetics (Fig. 9). For the tris-derivatised ligand metal complex this decelaration was multiplied several fold. These results indicate that in these closed systems, even at a 20:1 O2 :complex ratio, there is insuYcient oxygen to maintain pseudo-first-order kinetics. It is also unlikely that pseudo-firstorder kinetics was established in the earlier study 20 where the O2 :complex ratio was only 3 : 1. These results suggest that for the fluorous-derivatised metal complexes there is an additional “reaction” and we believe that this may oVer an additional explanation for the relative reaction rates illustrated in Fig. 8. It is well known that high concentrations of oxygen dissolve in perfluorocarbon solvents and that this occurs by trapping the oxygen molecules in voids in the solvent structure.48 Guillevic et al. 20 have shown that the rate of O2 addition to trans- [IrCl(CO){P(C2H4C6F13)3}2] is very much slower in perfluoromethylcyclohexane than in THF, which they have ascribed to the high aYnity of the solvent for the oxygen molecules. We have noted that it is very diYcult to remove dissolved oxygen from our ligands in solution, and we believe that the ponytails on our ligands and metal complexes have an aYnity for small molecules which may be similar to that for the perfluorocarbon solvents.Consequently, during these kinetic studies there is an alternative interaction whereby the oxygen molecules interact strongly with the perfluoroalkyl chains decreasing the apparent oxygen concentration in solution.Further kinetic and physical studies are necessary in this area to establish the exact nature of any interaction and the role of the fluorous ponytails in reactions at metal centres. Solubility studies The value of the FBS approach to catalysis depends upon the partition of the metal complex/catalyst in a two-phase organic/ fluorous solvent system.In this study at least six ponytails per metal centre are required to make a metal complex preferentially soluble in a perfluorocarbon solvent as shown by the complete absence of signals in the 31P-{1H} NMR spectra of the toluene phases after shaking these complexes in a toluene– PP3 two-phase system. At this six ponytail limit the type of metal complex appears to be important i.e. trans-[MCl(CO)L2] (M = Rh or Ir) are preferentially PP3 soluble whilst cis- and trans-[MCl2L2] (M = Pd or Pt) are not.Increasing the number of ponytails to nine, for example in [RhClL3], gives metal complexes which are very perfluorocarbon solvent soluble and dif- ficult to isolate except as very viscous oils. The NMR spectra † † 31P-{1H} NMR data for [RhCl{PPh32x(C6H4C6F13-4)x}]. x = 0 (CH2Cl2): d 31.5 (2P, dd 1JRhP 142, 2JPP 38, Ptrans-P) and 48.0 (1P, dt, 1JRhP 193, 2JPP 38 Hz, Pcis-P).49 x = 1 (CDCl3): d 31.0 (2P, dd, 1JRhP 143, 2JPP 38, Ptrans-P) and 47.8 (1P, dt, 1JRhP 193, 2JPP 38 Hz, Pcis-P).x = 3 (PP3): d 31.5 (2P, dd, 1JRhP 144, 2JPP 37, Ptrans-P) and 48.0 (1P, dt, 1JRhP 190, 2JPP 37 Hz, Pcis-P).J. Chem. Soc., Dalton Trans., 1998, 3751–3763 3763 for these complexes confirm the presence of three ligands at the metal centre, but we have been unable to obtain further characterisation data for these derivatives. Conclusion The reactions of ligands with perfluoroalkyl ponytails with platinum group metal starting materials readily aVords metal complexes of these ligands either by cleavage of chloride bridges or by displacement of weakly co-ordinating ligands.Spectroscopic and structural studies indicate that the introduction of the ponytails reduces the s-donor strength of the ligands but this is compensated by p eVects resulting in little overall aVect on the metal co-ordination environment. Solubility studies indicate that at least six ponytails are required for preferential solubility in perfluorinated solvents.However, increasing the number of ponytails can aVord complexes which are diYcult to isolate in a pure form as a consequence of the number of perfluoroalkyl substituents. Preliminary kinetic studies on the reaction of oxygen with analues of Vaska’s complex indicate that the oxidation is not as straightforward as that described for complexes without perfluoroalkyl substituents and further work on this aspect of this work is underway. Acknowledgements We would like to thank the Royal Society (E.G. H.), the EPSRC (D. R. P., A. M. S.) and BP Chemicals Ltd. (D. R. P.) for financial support, Dr G. Eaton for recording the mass spectral data, the EPSRC X-ray crystallography service for data collection for complex 3, the Director of the Daresbury laboratory for the provision of facilities, Dr M. Jones (BP Chemicals Ltd.) for helpful discussions and Drs J. Burgess and E. Raven (née Lloyd) for assistance with the kinetic studies.References 1 I. T. Horváth and J. Rábai, Science, 1994, 266, 72; US Pat., 5 463 082, 1995. 2 J. J. J. Juliette, I. T. Horváth and J. A. Gladysz, Angew. Chem., Int. Ed. Engl., 1997, 36, 1610. 3 I. Klement, H. Lütjens and P. Knochel, Angew. Chem., Int. Ed. Engl., 1997, 36, 1454; J. M. Vincent, A. Rabion, V. K. Yachandra and R. H. Fish, Angew. Chem., Int. Ed. Engl., 1997, 36, 2346. 4 D. P. Curran and S. Hadida, J. Am. Chem. Soc., 1996, 118, 2531; D. P. Curran and M.Hoshino, J. Org. Chem., 1996, 61, 6480; A. Studer and D. P. Curran, Tetrahedron, 1997, 53, 6681; A. Studer, S. Hadida, R. Ferritto, S. Y. Kim, P. Jeger, P. Wipf and D. P. Curran, Science, 1997, 275, 823. 5 B. Betzemeier and P. Knochel, Angew. Chem., Int. Ed. Engl., 1997, 36, 2623. 6 G. Pozzi, S. Banfi, A. Manfredi, F. Montanari and S. Quici, Tetrahedron, 1996, 52, 11879; G. Pozzi, F. Montanari and S. Quici, Chem. Commun., 1997, 69; G. Pozzi, I. Colombani, M. Miglioli, F. Montanari and S.Quici, Tetrahedron, 1997, 53, 6145; G. Pozzi, F. Cinato, F. Montanari and S. Quici, Chem. Commun., 1998, 877. 7 P. Bhattacharyya, D. Gudmunsen, E. G. Hope, R. D. W. Kemmitt, D. R. Paige and A. M. Stuart, J. Chem. Soc., Perkin Trans. 1, 1997, 3609. 8 A. K. Brisdon, EX, A Program for EXAFS Data Reduction, University of Leicester, 1992. 9 N. Binsted, S. J. Gurman and J. W. Campbell, EXCURV 92, SERC Daresbury Laboratory Program, 1992. 10 S. J. Gurman, N. Binsted and I.Ross, EXCURVE, J. Phys. C, 1984, 17, 143; 1986, 19, 1845. 11 F. R. Hartley, S. G. Murray and C. A. McAuliVe, Inorg. Chem., 1979, 18, 1394. 12 J. Herde, J. C. Lambert and C. V. SenoV, Inorg. Synth., 1974, 15, 18. 13 R. J. Goodfellow and L. M. Venanzi, J. Chem. Soc., 1965, 7533. 14 Solubilities of Inorganic and Metal Organic Compounds, 3rd edn., ed. A. Seidell, D. Van Nostrand, New York, 1940, vol. 1, p. 1359. 15 S. R. Davies, M. C. Mitchell, C. P. Cain, P. G. Devitt, R. J. Taylor and T.P. Kee, J. Organomet. Chem., 1998, 550, 29. 16 M. A. Guillevic, A. M. Arif, I. Horváth and J. A. Gladysz, Angew. Chem., Int. Ed. Engl., 1997, 36, 1612. 17 F. R. Hartley, The Chemistry of Platinum and Palladium, Applied Science, London, 1973, p. 456. 18 K. G. Moloy and J. L. Peterson, J. Am. Chem. Soc., 1995, 117, 7696. 19 K. Vrieze, J. P. Collman, C. T. Sears and M. Kubota, Inorg. Synth., 1968, 11, 101. 20 M. A. Guillevic, C. Rocaboy, A. M. Arif, I. T. Horváth and J. A. Gladysz, Organometallics, 1998, 17, 707. 21 A. J. Deeming and B. L. Shaw, J. Chem. Soc. A, 1968, 1887. 22 M. J. Atherton, J. Fawcett, A. P. Hill, J. H. Holloway, E. G. Hope, D. R. Russell, G. C. Saunders and R. M. J. Stead, J. Chem. Soc., Dalton Trans., 1997, 1137. 23 A. F. Williams, S. Bhaduri and A. G. Maddock, J. Chem. Soc., Dalton Trans., 1975, 1958. 24 J. J. MacDougall and J. H. Nelson, Inorg. Nucl. Chem. Lett., 1979, 15, 315. 25 R. G. Goel, Inorg. Nucl. Chem. Lett., 1979, 15, 437. 26 C. J. Cobley and P. G. Pringle, Inorg. Chim. Acta., 1997, 265, 107. 27 J. March, Advanced Organic Chemistry, Wiley, New York, 4th edn., 1992, p. 280. 28 A. Crispini, K. N. Harrison, A. G. Orpen, P. G. Pringle and J. R. Wheatcroft, J. Chem. Soc., Dalton Trans., 1996, 1069 and refs. therein. 29 J. Fawcett, E. G. Hope, R. D. W. Kemmitt, D. R. Paige, D. R. Russell, A. M. Stuart, D. J. Cole-Hamilton and M. J. Payne, Chem. Commun., 1997, 1127. 30 B. Albinsson and J. Michl, J. Phys. Chem., 1996, 100, 3418. 31 L. Xue, D. D. DesMarteau and W. T. Pennington, Angew. Chem., Int. Ed. Engl., 1997, 36, 1331. 32 C. R. Jablonski and Z. Zhou, Can. J. Chem., 1992, 70, 2544. 33 P. Kromm, J. -P. Bideau, M. Cotrait, C. Destrade and H. Nguyen, Acta Crystallogr., Sect. C, 1994, 50, 112. 34 R. D. Brost, G. C. Bruce, S. L. Grundy and S. R. Stobart, Inorg. Chem., 1993, 32, 5195. 35 W. Keim, P. Kraneburg, G. Dahmen, G. Deckers, U. Englert, K. Linn, T. P. Spaniol, G. Raabe and C. Kruger, Organometallics, 1994, 13, 3085. 36 S. Ogo, T. Suzuki, Y. Ozawa and K. Isobe, Inorg. Chem., 1996, 35, 6093. 37 G. K. Anderson, H. C. Clark, J. A. Davies, G. Ferguson and M. Parvez, J. Crystallogr. Spectrosc. Res., 1982, 12, 449. 38 A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson and R. Taylor, J. Chem. Soc., Dalton Trans., 1989, S1. 39 M. J. Atherton, K. S. Coleman, J. Fawcett, J. H. Holloway, E. G. Hope, A. Karaçar, L. A. Peck and G. C. Saunders, J. Chem. Soc., Dalton Trans., 1995, 4029; H. C. S. Clark, J. Fawcett, J. H. Holloway, E. G. Hope, L. A. Peck and D. R. Russell, J. Chem. Soc., Dalton Trans., 1998, 1249. 40 M. R. Churchill, J. C. Fettinger, L. A. Buttrey, M. D. Barkan and J. S. Thompson, J. Organomet. Chem., 1988, 340, 257. 41 G. Ferguson, R. McCrindle, A. J. McAlees and M. Parvez, Acta Crystallogr., Sect. B, 1982, 38, 2679. 42 R. W. Joyner, K. J. Martin and P. Meehan, J. Phys. C, 1987, 20, 4005. 43 N. Binsted, S. L. Cook, J. Evans, G. N. Greaves and R. J. Price, J. Am. Chem. Soc., 1987, 109, 3669. 44 See, for example, N. R. Champness, W. Levason, D. Pletcher, M. D. Spicer and M. Webster, J. Chem. Soc., Dalton Trans., 1992, 2201. 45 C. A. McAuliVe and R. Pollock, J. Organomet. Chem., 1974, 77, 265. 46 P. B. Chock and J. Halpern, J. Am. Chem. Soc., 1966, 88, 3511. 47 H. Stieger and H. Kelm, J. Phys. Chem., 1973, 77, 290. 48 J. G. Riess and M. LeBlanc, Pure Appl. Chem., 1982, 54, 2383. 49 T. H. Brown and P. J. Green, J. Am. Chem. Soc., 1970, 92, 2359. Paper 8/05141K

ISSN:1477-9226    

DOI:10.1039/a805141k

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3765–3770 3765 Platinum group metal complexes of a bis(diphenylphosphino)ethane ligand containing perfluoroalkyl ponytails Eric G. Hope,* Raymond D. W. Kemmitt and Alison M. Stuart Department of Chemistry, University of Leicester, Leicester, UK LE1 7RH Received 17th July 1998, Accepted 23rd September 1998 The perfluoroalkyl-derivatised bidentate phosphine (4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2 reacted with [PtCl2(MeCN)2], [PdCl2(MeCN)2] or [{RhCl(COD)}2] to yield the complexes [PtCl2(L–L)] 1, [PdCl2(L–L)] 2 and [Pd(L–L)2]212Cl2 3 or [{RhCl(L–L)}2] 5 respectively.Complex 3 was reduced with NaBH4 to [Pd(L–L)2] 4 and the chloride bridges in 5 were cleaved with triphenylphosphine to yield [RhCl(L–L)(PPh3)] 6. The reaction of the ligand with [{RhCl2(h5-C5Me5)}2] in a 1 : 1 or a 2 : 1 ratio yielded either [{RhCl2(h5-C5Me5)}2(L–L)] 7 or, after metathesis with NH4BF4 in acetone, [RhCl(h5-C5Me5)(L–L)]1BF4 2 8 respectively.Throughout, spectroscopic studies indicated that the ligand co-ordinates in a similar fashion to bis(diphenylphosphino)ethane (dppe). Complex 7 and the analogous dppe complex are fluxional at room temperature. At the low temperature limit, all the nuclei within these molecules are chemically inequivalent for which simulation suggests a gauche-eclipsed conformation around the PCH2–CH2P bond. Solubility studies indicated that only 4 and 5 are preferentially soluble in perfluorocarbon solvents.Following the proposal of the fluorous biphasic concept as a solution to the problem of catalyst/product separation in homogeneous catalysis,1 a number of ligands derivatised with long perfluoroalkyl sidechains have been prepared and their application in this field tested.1–6 Many of these systems involve monodentate phosphine ligands but, in view of the additional catalyst stability oVered by chelating ligands and their widespread application in homogeneous catalysis, we have recently prepared a perfluoroalkyl-derivatised analogue of bis(diphenylphosphino)ethane (dppe).7 Here, we illustrate the influence of the fluorous ponytails on the donor properties of this ligand by the synthesis and characterisation of a series of platinum metal group complexes.Experimental Proton, 19F and 31P NMR spectroscopies were carried out on a Bruker ARX250 spectrometer at 250.13, 235.34 and 101.26 MHz or a Bruker DRX 400 spectrometer at 400.13, 376.50 and 161.98 MHz and were referenced to external SiMe4 (1H), to external CFCl3 (19F) and to external 85% H3PO4 (31P) using the high-frequency positive convention.Elemental analyses were performed by Butterworth Laboratories Ltd. Mass spectra were recorded on a Kratos Concept 1H mass spectrometer. The ligand,7 [{RhCl(COD)}2],8 [PdCl2(MeCN)2] and [PtCl2(MeCN)2] 9 were prepared by the literature routes. Toluene and diethyl ether were dried by refluxing over sodium, dichloromethane by refluxing over calcium hydride and perfluoro-1,3-dimethylcyclohexane (PP3) by refluxing over calcium hydride and then freeze/thaw/degassed. Each solvent was stored in a closed glass ampoule over molecular sieves. Preparations [PtCl2{(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2}] 1.A mixture of [PtCl2(MeCN)2] (0.139 g, 0.4 mmol), the ligand (0.67 g, 0.4 mmol) and dichloromethane (50 cm3) was refluxed under nitrogen for 5 h. After cooling to room temperature the solvent was removed on a rotary evaporator and the white solid (0.68 g, 88%) washed with hexane (Found: C, 32.0; H, 1.0.C50H20Cl2F52P2Pt requires C, 31.0; H, 1.0%). MS (FAB): m/z 1901 (M 2 Cl)1. 19F NMR [(CD3)2CO]: d 280.8 (12F, t, 3JFF 10, CF3), 2110.5 [8F, unresolved(u)m, CaF2], 2121.1 (16F, um, CbF2 and CdF2), 2122.4 (8F, um, CeF2) and 2125.8 (8F, um, CgF2). [PdCl2{(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2}] 2. A mixture of [PdCl2(MeCN)2] (0.078 g, 0.3 mmol), the ligand (0.52 g, 0.3 mmol) and dichloromethane (50 cm3) was refluxed under nitrogen for 4 h.After cooling to room temperature the solvent was removed on a rotary evaporator and the products were washed with light petroleum (bp 40–60 8C) to give a pale yellow solid (0.50 g, 90%) (Found: C, 32.8; H, 0.8. C50H20Cl2F52P2Pd requires C, 32.5; H, 1.1%). MS (EI): m/z 1811 (M 2 Cl)1 and 1776 (M 2 HCl2)1. 19F NMR (CDCl3): d 281.2 (12F, t, 3JFF 10, CF3), 2111.9 (8F, um, CaF2), 2121.8 (16F, um, CbF2 and CdF2), 2123.2 (8F, um, CeF2) and 2126.6 (8F, um, CgF2).[Pd{(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2}2]212Cl2 3. A mixture of [PdCl2(MeCN)2] (0.039 g, 0.15 mmol), the ligand (0.530 g, 0.32 mmol) and dichloromethane (100 cm3) was refluxed under nitrogen for 19 h. After cooling to room temperature the volume of solvent was reduced to 20 cm3 on a rotary evaporator and the product filtered oV. It was washed with light petroleum and dichloromethane to give an insoluble white solid (0.435 g, 82%) (Found: C, 34.0; H, 1.1; F, 50.1; P, 4.0.C100H40Cl2F104P4Pd requires C, 34.1; H, 1.15; F, 56.2; P, 3.5%). MS (FAB): m/z 3483 (M 2 Cl)1 and 3447 (M 2 Cl2)1. [Pd{(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2}2] 4. A solution of NaBH4 (0.11 g, 2.9 mmol) in water (10 cm3) was added dropwise over 15 min to a suspension of complex 3 (0.34 g, 0.1 mmol) in water (15 cm3) and acetone (15 cm3) under nitrogen. An exotherm (50 8C) was observed during the addition. The reaction mixture was stirred at room temperature under nitrogen for 2 h.The product was extracted into ether and washed with water. After drying the ether layer over MgSO4 the ether was removed. The product was washed twice with dichloromethane and then dried under vacuum to give an orange solid (0.19 g, 57%) which decomposed slowly in solution and was light sensitive (Found: C, 34.3; H, 1.2. C100H40F104P4Pd requires C, 34.8; H, 1.2%). MS (FAB): m/z 3448 (M)1. 19F3766 J. Chem. Soc., Dalton Trans., 1998, 3765–3770 NMR (CD2Cl2): d 281.8 (24F, t, 3JFF 10, CF3), 2111.7 (16F, um, CaF2), 2122.3 (32F, um, CbF2 and CdF2), 2123.6 (16F, um, CeF2) and 2126.9 (16F, um, CgF2).[{RhCl[(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2]}2] 5. A solution of the ligand (1.0 g, 0.6 mmol) in perfluoro-1,3- dimethylcyclohexane (PP3) (20 cm3) was added to a solution of [{RhCl(COD)}2] (0.15 g, 0.3 mmol) in toluene (15 cm3) with stirring at 60 8C under nitrogen. The reaction mixture was heated to 90 8C for 4 h and the bottom layer changed from cloudy white to yellow to an orange-red clear solution.After cooling the bottom layer was transferred under nitrogen to another Schlenk flask and the solvent removed in vacuo to leave an orange-red solid (0.89 g, 82%) (Found: C, 33.6; H, 0.9; F, 51.2. C100H40Cl2F104P4Rh requires C, 33.2; H, 1.1; F, 54.6%). 19F NMR ([2H8]toluene): d 281.6 (24F, t, 3JFF 9, CF3), 2111.4 (16F, t, 3JFF 14, CaF2), 2121.8 (16F, um, CbF2), 2121.9 (16F, um, CdF2), 2123.3 (16F, um, CeF2) and 2126.7 (16F, um, CgF2).[RhCl{(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2}(PPh3)] 6. A solution of the ligand (0.70 g, 0.4 mmol) in PP3 (30 cm3) was added to [{RhCl(COD)}2] (0.098 g, 0.2 mmol) with stirring in toluene (20 cm3) at 50 8C under nitrogen. The reaction mixture was then heated at 80 8C for 3 h. The bottom, fluorous, layer was then transferred, under nitrogen, on to triphenylphosphine (0.11 g, 0.42 mmol) with stirring in toluene (15 cm3) and stirred at 50 8C for 2 h under nitrogen.After cooling the solvents were removed in vacuo to leave a yellow solid. The product was finally washed with toluene to remove traces of any excess of triphenylphosphine (0.56 g, 68%) (Found: C, 39.8; H, 1.4; F, 42.4; P, 4.0. C68H35ClF52P3Rh requires C, 39.4; H, 1.7; F, 47.7; P, 4.5%). MS (FAB): m/z 1773 (M 2 Cl 2 PPh3)1. 19F NMR (diethyl ether): d 281.9 (12F, um, CF3), 2111.4 (8F, um, CaF2), 2122.0 (16F, um, CbF2 and CdF2), 2123.4 (8F, um, CeF2) and 2126.9 (8F, um, CgF2).[{RhCl2(h5-C5Me5)}2{(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13- 4)2}] 7. A solution of [{RhCl2(h5-C5Me5)}2] (Aldrich) (0.105 g, 0.17 mmol) and the ligand (0.29 g, 0.17 mmol) in dichloromethane (50 cm3) was stirred at room temperature under nitrogen for 2.5 h. After removing the solvent on the rotary evaporator, the product was washed with hexane to give a red solid (0.23 g, 59%) (Found: C, 37.1; H, 2.1; P, 3.6. C70H50Cl4F52P2Rh2 requires C, 36.7; H, 2.2; P, 2.7%).MS (FAB): m/z 2253 (M 2 Cl)1, 2118 (M 2 Cp*Cl)1, 2081 (M 2 Cl2Cp*)1, 1978 (M 2 RhCp*Cl2)1, 1943 (M 2 RhCp*Cl3)1, 1908 (M 2 RhCp*Cl4)1 and 1808 (M 2 RhCp*2Cl3)1. 19F NMR ([2H8]toluene): 363 K, d 281.5 (12F, t, 3JFF 10, CF3), 2110.8 (8F, t, 3JFF 14, CaF2), 2121.2 (16F, um, CbF2 and CdF2), 2122.6 (8F, um, CeF2) and 2126.0 (8F, um, CgF2); 233 K, d 281.2 (3F, t, 3JFF 9, CF3), 281.3 (3F, t, 3JFF 9, CF3), 281.4 (3F, t, 3JFF 9, CF3), 281.5 (3F, t, 3JFF 9, CF3), 2111.5 (2F, um, CaF2), 2111.7 (2F, t, 3JFF 14, CaF2), 2111.9 (2F, t, 3JFF 14, CaF2), 2112.2 (2F, t, 3JFF 14, CaF2), 2122.2 (16F, br m, CbF2 and CdF2), 2123.6 (8F, br m, CeF2), 2126.8 (4F, br m, CgF2) and 2127.1 (4F, br m, CgF2). [RhCl(h5-C5Me5){(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13- 4)2}]1BF4 2 8.A solution of [{RhCl2(h5-C5Me5)}2] (Aldrich) (0.049 g, 0.079 mmol) and the ligand (0.271 g, 0.16 mmol) in toluene (50 cm3) was stirred at room temperature under nitrogen for 6 h.After filtering the solution the toluene was removed on the rotary evaporator to give a red solid which was a mixture of products, [{RhCl2(h5-C5Me5)}2{(4-F13C6C6H4)2PCH2CH2P- (C6H4C6F13-4)2}] and [RhCl2(h5-C5Me5){(4-F13C6C6H4)2- PCH2CH2P(C6H4C6F13-4)2}] and unchanged ligand. This mixture (0.241 g) was dissolved in acetone (50 cm3) and an excess of NH4BF4 (1 g) added. The suspension was stirred under nitrogen at room temperature for 20 h and changed from orange-red to lemon yellow.The acetone was removed on the rotary evaporator and the product washed with water and light petroleum to give the yellow product (0.161 g, 50%). The [RhCl(h5-C5Me5){(4-F13C6C6H4)2PCH2CH2P(C6H4C6F13- 4)2}]1BF4 2 was finally recrystallised from dichloromethane (Found: C, 35.8; H, 1.5; P, 3.0. C60H35BClF56P2Rh requires C, 35.5; H, 1.7; P, 3.1%). MS (FAB): m/z 1943/5 (M 2 BF4)1 and 1909 (M 2 Cl 2 BF4)1. 19F NMR [(CD3)2CO]: d 282.1 (12F, m, CF3), 2111.8 (4F, t, 3JFF 14, CaF2), 2112.0 (4F, br m, CaF2), 2122.4 (6F, um, CbF2 and CdF2), 2122.5 (2F, um, CbF2), 2123.8 (8F, um, CeF2) and 2127.1 (8F, um, CgF2).Results and discussion The reactions between the new fluorous-ponytail-derivatised analogue of bis(diphenylphosphino)ethane and some conventional platinum-group metal starting materials yields analogues of well established co-ordination and organometallic complexes either by cleavage of chloride-bridged dimers or by the displacement of weakly co-ordinating ligands.The products were all obtained as solids in 50–90% yield and were characterised by mass spectrometry, 1H, 19F and 31P NMR spectroscopies. In the mass spectra most of the complexes showed either the parent ion or [M 2 Cl]1, in line with mass spectral data for the analogous DPPE complexes. NMR spectroscopic studies Fluorine-19 NMR data (Experimental section) for all the complexes show, principally, five or six highly consistent multiplet resonances which are similar to those for the ‘free’ ligand 7 and for metal complexes of related perfluoroalkylderivatised triarylphosphines.10 The highest frequency resonances are assigned to the terminal CF3 groups and the remaining, CF2, resonances are assigned according to Scheme 1 from 19F–19F COSY experiments.For the platinum and palladium complexes 1, 2 and 4, the 1H and 1H-{31P} NMR data (Table 1) are similar to those for the ‘free’ ligand 7 and to those for the analogous dppe complexes prepared by the literature routes.12,13 Complex 3 is insoluble in all solvents and, hence, NMR data are unavailable.For 1, 2 and 4 a doublet/multiplet resonance is observed for the C2H4 protons for which 3JHP is comparable to that for the analogous DPPE complexes, whilst the aryl protons are shifted to higher frequency than those for the dppe complexes, due to the electron withdrawing perfluoroalkyl chain, and show well resolved 3JHH and 3JHP couplings. The 31P-{1H} NMR spectra (Table 1) exhibit a single resonance (Pd) or a single resonance with satellites (Pt) where dP is virtually identical to that for the analogous dppe complexes.For 1, 1JPtP, which is normally a fair indicator of electronic eVects, is smaller than the value for [PtCl2(dppe)] (3568 cf. 3594 Hz). From data for a series of related [PtCl2(PR3)2] (R = aryl) complexes it has been concluded that a reduction in 1JPtP can be correlated with the Hammett function for the ligand which can be accounted for by a reduction in the PÆPt s donation of the ligand and a weaker Pt–P bond.14 Therefore, for 1, these NMR data suggest that the aryl spacer groups do not completely insulate the metal from the electronic eVects of the perfluoroalkyl substituents.However, we have noted 10 a comparable reduction in 1JPtP (3676 to 3631 Hz) for cis-[PtCl2L2] [L = PPh3 2 x(C6H4C6F13-4)x; x = 0, 1, 2 or 3] but structural data for the platinum complexes when x = 0 and 3 indicate that the metal–phosphorus bond lengths are unaVected by the introduction of the perfluoroalkyl units.For the chloride-bridged dimer 5 the NMR data are also very similar to those for the analogous dppe complex.11 In Scheme 1 CaF2 CbF2 CgF2 CdF2 CeF2 CF3J. Chem. Soc., Dalton Trans., 1998, 3765–3770 3767 Table 1 Proton, 1H-{31P} and 31P-{1H} NMR data (d, J/Hz) for some DPPE complexes and their analogues with perfluoroalkyl substituents a Compound 1 [PtCl2(L–L)] b,c 1H NMR Data 2.88 (4 H, br d, 3JHP 19.0, PCH2), 7.77 (8 H, d, 3JHH 8.3, m-H of C6H4P), 8.14 (8 H, dd, 3JHH 8.4, 3JHP 11.9, o-H of C6H4P) 1H-{31P} NMR Data 2.88 (4 H, s, PCH2), 7.77 (8 H, d, 3JHH 8.3, m-H of C6H4P), 8.14 (8 H, d, 3JHH 8.3, o-H of C6H4P) 31P-{1H} NMR Data 43.4 (s, 1JPtP 3568) [PtCl2(dppe)] c 2.49 (4 H, d, 3JHP 18.9, PCH2), 7.40 (8 H, tm, 3JHH 7.5 and 7.0, m-H of C6H5P), 7.46 (4 H, m, p-H of C6H5P), 7.82 (8 H, m, o-H of C6H5P) 2.49 (4 H, s, PCH2), 7.40 (8 H, t, 3JHH 7.5 and 7.0, m-H of C6H5P), 7.46 (4 H, m, p-H of C6H5P), 7.82 (8 H, d, 3JHH 6.7, o-H of C6H5P) 42.8 (s, 1JPtP 3594) 2 [PdCl2(L–L)] b,d 2.60 (4 H, d, 3JHP 23.2, PCH2), 7.68 (8 H, d, 3JHH 8.3, m-H of C6H4P), 7.98 (8 H, dd, 3JHH 8.2, 3JHP 11.7, o-H of C6H4P) 2.59 (4 H, s, PCH2), 7.68 (8 H, d, 3JHH 8.3, m-H of C6H4P), 7.98 (8 H, d, 3JHH 8.2, o-H of C6H4P) 62.7 (s) [PdCl2(dppe)] d 2.38 (4 H, d, 3JHP 22.9, PCH2), 7.42 (8 H, td, 3JHH 7.3, 4JHP 2.5, m-H of C6H5P), 7.48 (4 H, m, p-H of C6H5P), 7.81 (8 H, m, o-H of C6H5P) 2.38 (4 H, s, PCH2), 7.42 (8 H, t, 3JHH 7.3, m-H of C6H5P), 7.49 (4 H, m, p-H of C6H5P), 7.81 (8 H, dm, 3JHH 7.0, o-H of C6H5P) 64.2 (s) 4 [Pd(L–L)2] b,e 2.18 (4 H, m, PCH2), 7.24 (8 H, d, 3JHH 7.9, m-H of C6H4P), 7.34 (8 H, m, o-H of C6H4P) 2.18 (4 H, s, PCH2), 7.24 (8 H, d, 3JHH 8.3, m-H of C6H4P), 7.34 (8 H, d, 3JHH 8.2, o-H of C6H4P) 29.6 (s) [Pd(dppe)2] e 2.00 (8 H, m, PCH2), 6.95 (16 H, t, 3JHH 7.0, m-H of C6H5P), 7.06 (8 H, t, 3JHH 7.0, p-H of C6H5P), 7.28 (16 H, m, o-H of C6H5P) Not recorded 31.4 (s) 5 [{RhCl(L–L)}2] b 1.56 (8 H, br d, 3JHP 18.5, PCH2), 7.31 (16 H, d, 3JHH 8.2, m-H of C6H4P), 7.77 (16 H, t, 3JHH 8.2, 3JHP 8.8, o-H of C6H4P) Not recorded 74.9 (d, 1JRhP 198) [{RhCl(dppe)}2] e,f 2.07 (8 H, br d, 3JHP 20, PCH2), 7.20–7.85 (40 H, br m, aryl protons) Not reported 74.2 (d, 1JRhP 198) 6 [RhCl(L–L)(PPh3)] b,e 2.03 (2 H, dm, 3JHP 32, Ha), 2.11 (2 H, dm, 3JHP 31, Hb), 7.15 (6 H, td, 3JHH 7.7, 4JHP 1.6, Hh), 7.31 (3 H, tm, 3JHH 7.5, Hi), 7.40 (6 H, t, 3JHP 9.8, 3JHH 7.6, Hg), 7.46 (4 H, d, 3JHH 8.2, Hf), 7.70 (4 H, t, 3JHP 10.7, 3JHH 7.9, He), 7.73 (4 H, d, 3JHH 7.3, Hd), 8.14 (4 H, t, 3JHP 10.7, 3JHH 7.3, Hc) 7.13 (6 H, t, 3JHH 7.6, 7.1, Hh), 7.27 (3 H, t, 3JHH 7.1, Hi), 7.47 (6 H, d, 3JHH 7.6, Hg), 7.53 (4 H, d, 3JHH 7.9, Hf), 7.73 (4 H, d, 3JHH 7.9, Hd), 7.80 (4 H, d, 3JHH 7.6, He), 8.29 (4 H, d, 3JHH 7.9, Hc) g 30.6 (1 P, ddd, 2JPcPa 358, 1JRhP 131, 2JPcPb 36, Pc), 61.0 (1 P, ddd, 2JPaPc 358, 1JRhP 142, 2JPaPb 33, Pa), 75.5 (1 P, dt, 1JRhP 187, 2JPcPb 36, 2JPaPb 33, Pb) 7 [{RhCl2(h5-C5Me5)}2- (L–L)] b (373 K) 1.11 (30 H, d, JHP 3.4, CH3), 3.10 (4 H, d, 3JHP 2.4, PCH2), 7.45 (8 H, d, 3JHH 8.1, m-H of C6H4P), 8.02 (8 H, t, 3JHP 9.3, 3JHH 8.1, o-H of C6H4P) 1.11 (30 H, s, CH3), 3.10 (4 H, s, PCH2), 7.45 (8 H, d, 3JHH 8.1, m-H of C6H4P), 8.02 (8 H, d, 3JHH 8.1, o-H of C6H4P) 29.3 (AA9XX9 spectrum, 1JRhP 145, 3JPP 36, 4JRhP 0) (233 K) 0.94 (15 H, d, JHP 3.1, CH3), 1.11 (15 H, d, JHP 3.1, CH3), 2.52 (1 H, m, PCH2), 2.68 (1 H, m, PCH2), 3.45 (1 H, m, PCH2), 3.75 (1 H, m, PCH2), 6.77 (2 H, d, 3JHH 7.7, m-H of C6H4P), 7.39 (2 H, d, 3JHH 8.6, m-H of C6H4P), 7.52 (2 H, d, 3JHH 7.8, m-H of C6H4P), 7.66 (2 H, t, 3JHP 8.1, 3JHH 7.3, o-H of C6H4P), 8.28 (3 H, br s, C6H4), 8.93 (1 H, br s, C6H4) h 0.95 (15 H, s, CH3), 1.11 (15 H, s, CH3), 2.52 (1 H, t, 2JHaHc 4.6, 3JHaHb 214.6, 3JHaHd 11.9, Ha), 2.68 (1 H, t, 2JHaHc 4.6, 3JHbHc 11.9, 3JHcHd 214.6, Hc), 3.44 (1 H, t, 2JHbHd 4.6, 3JHaHb 214.6, 3JHbHc 11.9, Hb), 3.75 (1 H, t, 2JHbHd 4.6, 3JHcHd 214.6, 3JHaHd 11.9, Hd), 6.79 (2 H, d, 3JHH 7.7, m-H of C6H4P), 7.40 (2 H, d, 3JHH 7.9, m-H of C6H4P), 7.52 (2 H, d, 3JHH 8.0, m-H of C6H4P), 7.67 (2 H, d, 3JHH 7.3, o-H of C6H4P), 8.28 (3 H, br s, C6H4), 8.91 (1 H, br s, C6H4) h 28.8 (1 P, 1JRhP 144, 3JPP 35, 4JRhP 0, Pa), 29.4 (1 P, 1JRhP 144, 3JPP 35, 4JRhP 0, Pb) (ABXY spectrum) 9 [{RhCl2(h5-C5Me5)}2- (dppe)] (343 K) 1.27 (30 H, m, JHP 3.2, CH3), 3.27 (4 H, d, 3JHP 2.3, PCH2), 7.17 (12 H, m, m- and p-H of C6H5), 8.09 (8 H, br t, o-H of C6H5) 1.27 (30 H, s, CH3), 3.27 (4 H, s, PCH2), 7.17 (12 H, m, m- and p-H of C6H5), 8.09 (8 H, d, 3JHH 7.0, o-H of C6H5) 30.3 (AA9XX9 spectrum, 1JRhP 143, 3JPP 33, 4JRhP 0) (243 K) 1.16 (15 H, d, JHP 3.1, CH3), 1.45 (15 H, d, JHP 3.2, CH3), 2.96 (1 H, br m, PCH2), 3.21 (1 H, br m, PCH2), 3.46 (1 H, br m, PCH2), 3.81 (1 H, br m, PCH2), 6.94 (4 H, br s, aryl), 7.07 (4 H, br s, aryl), 7.44 (4 H, br s, aryl), 7.73 (4 H, t, 3JHH 7.0, 3JHP 9.0, o-H of C6H5), 7.81 (2 H, br s, aryl), 8.44 (2 H, br s, aryl) 1.18 (15 H, s, CH3), 1.47 (15 H, s, CH3), 2.97 (1 H, br t, PCH2), 3.23 (1 H, br t, PCH2), 3.48 (1 H, br t, PCH2), 3.83 (1 H, br t, PCH2), 6.96 (4 H, br s, aryl), 7.09 (4 H, br m, aryl), 7.46 (4 H, br s, aryl), 7.75 (4 H, d, 3JHH 7.0, o-H of C6H5), 7.84 (2 H, br s, aryl), 8.26 (2 H, br s, aryl) 28.1 (1 P, dd, 1JRhP 143, 3JPP 36, 4JRhP 1, Pa), 30.5 (1 P, dd, 1JRhP 143, 3JPP 36, 4JRhP 1, Pb) (ABXY spectrum) 8 [RhCl(h5-C5Me5)- (L–L)][BF4] b,c 1.54 (15 H, t, JHP 3.5, CH3), 3.01 (2 H, m, PCH2), 3.32 (2 H, m, PCH2), 7.55 (4 H, t, 3JHP 11.1, 3JHH 7.9, o-H of C6H4P), 7.86 (8 H, d, 3JHH 7.9, m-H of C6H4P), 7.96 (4 H, t, 3JHP 11.5, 3JHH 7.9, o-H of C6H4P) 1.54 (15 H, s, CH3), 3.01 (2 H, m, PCH2), 3.32 (2 H, m, PCH2), 7.55 (4 H, d, 3JHH 8.1, o-H of C6H4P), 7.85 (8 H, d, 3JHH 7.9, m-H of C6H4P), 7.96 (4 H, d, 3JHH 8.3, o-H of C6H4P) 66.7 (d, 1JRhP 133) 10 [RhCl(h5-C5Me5)- (dppe)][BF4] c 1.45 (15 H, t, JHP 3.4, CH3), 2.62 (2 H, m, PCH2), 3.16 (2 H, m, PCH2), 7.19 (4 H, m, o-H of C6H5P), 7.54 (16 H, m, aryl) 1.45 (15 H, s, CH3), 2.61 (2 H, m, PCH2), 3.10 (2 H, m, PCH2), 7.19 (4 H, d, 3JHH 7.3, o-H of C6H5P), 7.54 (16 H, m, aryl) 66.2 (d, 1JRhP 132) a Spectra recorded in [2H8]toluene unless otherwise stated.b L–L = (4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2. c Spectra recorded in (CD3)2CO.d Spectra recorded in CDCl3. e Spectra recorded in CD2Cl2. f Data taken from ref. 11. g Data recorded in Et2O using D2O as lock substance. Backbone protons hidden under solvent signals. h Other aryl peaks hidden under solvent peaks.3768 J. Chem. Soc., Dalton Trans., 1998, 3765–3770 particular, dP, 1JRhP and 3JHP (PCH2) are virtually identical implying that the perfluoroalkyl groups do not have a significant eVect on the co-ordination properties of the ligand in this system.Cleavage of the dimer with triphenylphosphine gives an analogue of Wilkinson’s complex, 6. Here, the NMR spectral data are complicated and the data in Table 1 have been assigned according to the labels in Scheme 2. The 31P-{1H} NMR data confirm that the two ends of the bidentate ligand are inequivalent. Consequently, the spectra show two well resolved mutually coupled doublets of doublets of doublets and a doublet of apparent triplets in contrast to the highly second-order spectrum for the structurally characterised [RhCl(PPh3)(dfppe)] [dfppe = (F5C6)2PCH2CH2P(C6F5)2].15 There are few compounds with which to compare these spectroscopic data.The 1JRhP couplings for Ptrans-P are smaller than that for Ptrans-Cl as expected from the relative trans influence of phosphorus and chlorine and as seen for Wilkinson’s complex; 16 the cis-2JPP couplings are also similar to those for Wilkinson’s complex. The trans-2JPP coupling constant is an order of magnitude larger than these cis-2JPP values as seen for the comparable couplings in cis- and trans-[PtCl2(PEt3)L] (L = monodentate phosphine).17 The inequivalent phosphorus atoms create asymmetry in the bidentate ligand.The backbone protons are inequivalent and give two unresolved multiplets, in the 1H NMR spectrum, both showing resolvable coupling to phosphorus. Similarly, the protons on the aryl rings are now inequivalent and the assignments, including those for the triphenylphosphine protons, have been made using 1H–31P Scheme 2 C6F13 Hd Hc Hc Hd Pa Ha C6F13 Hf He He Hf Pb Ha Hb Hb Cl Hi Hh Hg Hg Hh Pc Rh 2 2 3 COSY and HMQC (heteronuclear multiple quantum correlation) (selected for JHP = 10 and 30 Hz) and then 1H homonuclear decoupling experiments.We note that the resonances in the 19F NMR spectrum for this complex are unusually broad and no resolvable couplings could be identified. It is likely that this occurs due to the inequivalence of the aryl rings on Pa and Pb which would render the perfluoroalkyl chains inequivalent and so each peak in the 19F NMR spectrum arises from the overlap of at least two sets of fluorine resonances.In the reactions of [{RhCl2(h5-C5Me5)}2] with bidentate phosphine ligands (L–L) three types of product can be obtained, a ligand-bridged dinuclear complex, [(RhCl2Cp*)2- (L–L)], a neutral mononuclear complex [RhCl2Cp*(L–L)] or a cationic [RhClCp*(L–L)]1 species.18–21 We have observed, by NMR spectroscopy, all three types of complex in the reaction of (4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2 with [{RhCl2(h5- C5Me5)}2], but have only obtained the first (7) and the last (8) complexes analytically pure.As a part of this study we have also prepared the analogous complexes with dppe (9 and 10), to allow a direct comparison between the NMR spectral data for complexes with our ligand and with dppe. The 31P NMR spectrum of the cationic complex 8 is a simple doublet in which 1JRhP is comparable to that for related complexes.In the 1H NMR spectrum the protons on the pentamethylcyclopentadienyl ring show equivalent couplings to both phosphorus atoms, confirming that this ligand is undergoing unrestricted rotation. Although there is a single doublet for the aryl protons meta to P, the resonances for the ortho-aryl and backbone protons are split into two multiplets. This arises from chelation of the ligand and asymmetry at the metal centre, i.e.two of the backbone protons are on the same side of the chelate ring as the Cp* ligand, whilst the other two protons are on the opposite side. Similarly, although there is unhindered rotation about the P–C (aryl) bonds, one aryl ring is cis-Cp* whilst the other is trans-Cp*. This eVect is mirrored in the 1H NMR spectrum of 10 and in the 19F NMR spectra of the fluorinated dppe analogues [RhCl(h5-C5Me5)(L–L)]1 (L–L = dfppe 21 or ddfppe {1,2-bis[bis(2,6-difluorophenyl)phosphino]ethane} 22).Interestingly, this asymmetry is also seen in the 19F NMR spectrum of 8 (Experimental section) where two sets of multiplets can be assigned to the CaF2 and CbF2 fluorine atoms. The fluorine atoms further along the perfluoroalkyl chains cannot be distinguished but the resonances for these fluorine atoms are significantly broader than those for the ‘free’ ligand 7 or for 1–4. Fig. 1 Experimental [at 363 (a) and 233 K (c)] and simulated (b and d) 31P-{1H} NMR spectra of [{RhCl(h5-C5Me5)}2{(4-F13C6C6H4)2- PCH2CH2P(C6H4C6F13-4)2}] 7.J.Chem. Soc., Dalton Trans., 1998, 3765–3770 3769 Fig. 2 Experimental [at 373 (a) and 243 K (c)] and simulated (b and d) 31P-{1H} NMR spectra of [{RhCl(h5-C5Me5)}2(dppe)] 9. The NMR spectra for complexes 7 and 9 are not simple and variable temperature experiments and simulation were required for assignment. As identified in the earliest preparation of 9,18 the Cp* protons show an unusual pattern in the 1H NMR spectrum which was thought to be a result of the magnetic inequivalence of the phosphorus atoms.Subsequently, the related complex [{RhCl2(h5-C5Me5)}2{Ph2PCH2CH(CH3)- PPh2}],19 in which the phosphorus atoms are chemically inequivalent, was shown to be fluxional at room temperature, although no mention of fluxionality or magnetic inequivalence was made in a paper on a comprehensive series of diphosphinebridged dirhodium complexes.20 For 7 ( at 363 K) and 9 (at 373 K), in [2H8]toluene, the high temperature limit reveals classical AA9XX9 second order 31P-{1H} NMR spectra [Figs. 1(a,b) and 2(a,b)] which can be simulated 23 using the values listed in Table 1. The similarity between the values for 7 and 9 illustrates, again, that the aryl rings are very good insulators of the electronic influence of the perfluoroalkyl groups. Similarly, at this high temperature limit, the 19F NMR spectrum for 7 (Experimental section) reveals the five resonances typical for equivalent perfluoroalkyl groups.Apart from the unusual threeline pattern assigned to the Cp* protons for 9, the magnetic inequivalence of the phosphorus atoms is not obvious from the 1H NMR spectra. In particular, the simpler doublet pattern for the Cp* protons for 7 suggests that the longer range phosphorus–proton coupling in this molecule is too small to aVect the appearance of this resonance. On cooling the NMR spectra look very diVerent to those at the high temperature limit and the spectra are well resolved at 233 [7, Figs. 1(c,d) and 3] and at 243 K [9, Fig. 2(c,d)]. The low temperature spectra for complexes 7 and 9 are similar and can be simulated (Table 1) and interpreted in the same way. For 7, two resonances in the 31P NMR spectrum (Fig. 1) indicate that the phosphorus atoms are now inequivalent. This is confirmed from the 1H NMR spectrum in which two low-frequency doublets, assigned to the Cp* protons, indicate that the two ends of the molecule are inequivalent and, from the 1H-{31P} NMR spectrum (Fig. 3), four mutually coupled resonances can be assigned to the backbone protons indicating that there is restricted rotation about the C–C (backbone) bond. Furthermore, in the 19F NMR spectrum in the high frequency region associated with the terminal CF3 fluorine atoms, four well resolved triplet resonances (Experimental section) indicate that all four perfluoroalkyl chains are inequivalent. The 31P NMR spectrum can be simulated in terms of a second order ABXY spin system in which d(PA) and d(PB) have only slightly diVerent values (Table 1; Fig. 1). The backbone region of the 1H-{31P} NMR spectrum can be simulated as an ABCD spin system (Table 1; Fig. 3) for which the 3JHH and 2JHH coupling constants suggest that the Karplus angle 24 for the backbone at the low temperature limit is 68 (Fig. 4), a gauche-eclipsed conformation. Hence, the molecule has no symmetry and all the nuclei are chemically inequivalent.The similarity in the spectral parameters for 7 and 9 indicates, again, that the aryl groups are good insulators of the electronic influence of the perfluoroalkyl ponytails. Solubility studies Part of the rationale behind this study was to prepare potential Fig. 3 Experimental at 233 K (a) and simulated (b) 1H-{31P} NMR spectra of complex 7 in the region d 2.3 to 4.0.3770 J. Chem. Soc., Dalton Trans., 1998, 3765–3770 catalysts with a bidentate ligand for catalysis in a fluorous biphase.However, although the ‘free’ ligand is preferentially soluble in perfluorocarbon solvents, as shown by qualitative 31P-{1H} NMR studies (see preceding paper), only complexes 4 and 5, which contain two substituted ligands, are similarly soluble. Consequently, to make appropriate catalysts for catalysis in a fluorous biphase, longer and/or more perfluoroalkyl substituents need to be introduced in this system. Conclusion Co-ordination complexes of the perfluoroalkyl-derivatised analogue of dppe, (4-F13C6C6H4)2PCH2CH2P(C6H4C6F13-4)2, have very similar spectroscopic properties to those containing dppe indicating that the aryl rings are good insulators of the electron withdrawing eVects of the perfluoroalkyl groups.However, only those complexes with two of these ligands are preferentially soluble in perfluorocarbon solvents, restricting their applicability for fluorous biphase catalysis. Acknowledgements We would like to thank the Royal Society (E.G. H.) and the EPSRC (A. M. S.) for financial support. We would also like to thank Dr M. Jones (BP Chemicals Ltd.) for helpful discussions and Dr H. C. S. Clark for assistance with the NMR simulations. References 1 I. T. Horváth and J. Rábai, Science, 1994, 266, 72; US Pat., 5 463 082, 1995; I. T. Horváth, G. Kiss, R. A. Cook, J. E. Bond, P. A. Stevens, J. Rábai and E. J. Mozeleski, J. Am. Chem. Soc., 1998, 120, 3133. 2 J. J. J. Juliette, I. T. Horváth and J.A. Gladysz, Angew. Chem., Int. Ed. Engl., 1997, 36, 1610. 3 I. Klement, H. Lütjens and P. Knochel, Angew. Chem., Int. Ed. Engl., 1997, 36, 1454; J. M. Vincent, A. Rabion, V. K. Yachandra and R. H. Fish, Angew. Chem., Int. Ed. Engl., 1997, 36, 2346. Fig. 4 Newman projection of complex 7 along the CH2–CH2 axis at the low-temperature limit. 4 D. P. Curran and S. Hadida, J. Am. Chem. Soc., 1996, 118, 2531; D. P. Curran and M. Hoshino, J. Org. Chem., 1996, 61, 6480; A. Studer and D.P. Curran, Tetrahedron, 1997, 53, 6681; A. Studer, S. Hadida, R. Ferritto, S. Y. Kim, P. Jeger, P. Wipf and D. P. Curran, Science, 1997, 275, 823. 5 B. Betzemeier and P. Knochel, Angew. Chem., Int. Ed. Engl., 1997, 36, 2623. 6 G. Pozzi, S. Banfi, A. Manfredi, F. Montanari and S. Quici, Tetrahedron, 1996, 52, 11879; G. Pozzi, F. Montanari and S. Quici, Chem. Commun., 1997, 69; G. Pozzi, I. Colombani, M. Miglioli, F. Montanari and S. Quici, Tetrahedron, 1997, 53, 6145; G.Pozzi, F. Cinato, F. Montanari and S. Quici, Chem. Commun., 1998, 877. 7 P. Bhattacharyya, D. Gudmunsen, E. G. Hope, R. D. W. Kemmitt, D. R. Paige and A. M. Stuart, J. Chem. Soc., Perkin Trans. 1, 1997, 3609. 8 G. Giordano and R. H. Crabtree, Inorg. Synth., 1979, 19, 218. 9 F. R. Hartley and C. A. McAuliVe, Inorg. Chem., 1979, 18, 1394. 10 J. Fawcett, E. G. Hope, R. D. W. Kemmitt, D. R. Paige, D. R. Russell and A. M. Stuart, preceding paper. 11 D. P. Fairlie and B. Bosnich, Organometallics, 1988, 7, 936. 12 C. H. Lindsay, L. S. Benner and A. L. Balch, Inorg. Chem., 1980, 19, 3503. 13 M. J. Hudson, R. S. Nyholm and M. H. B. Stiddard, J. Chem. Soc. A, 1968, 40; E. G. Hope, W. Levason and N. A. Powell, Inorg. Chim. Acta, 1986, 115, 187. 14 C. J. Cobley and P. G. Pringle, Inorg. Chim. Acta, 1997, 265, 107. 15 M. J. Atherton, K. S. Coleman, J. Fawcett, J. H. Holloway, E. G. Hope, A. Karaçar, L. A. Peck and G. C. Saunders, J. Chem. Soc., Dalton Trans., 1995, 4029. 16 A. J. Naaktyeboren, R. J. M. Nolte and W. Drenth, J. Am. Chem. Soc., 1980, 102, 3350; T. H. Brown and P. J. Green, J. Am. Chem. Soc., 1970, 92, 12359. 17 M. J. Atherton, J. Fawcett, A. P. Hill, J. H. Holloway, E. G. Hope, D. R. Russell, G. C. Saunders and R. M. J. Stead, J. Chem. Soc., Dalton Trans., 1997, 1137. 18 J. W. Kang, K. Moseley and P. M. Maitlis, J. Am. Chem. Soc., 1969, 91, 5970. 19 D. Carmona, F. J. Lahoz, L. A. Oro, M. P. Lamata, F. Viguri and E. S. José, Organometallics, 1996, 15, 2961. 20 W. Keim, P. Kraneburg, G. Dahmen, G. Deckers, U. Englert, K. Linn, T. P. Spaniol, G. Raabe and C. Krüger, Organometallics, 1994, 13, 3085. 21 M. J. Atherton, J. Fawcett, J. H. Holloway, E. G. Hope, A. Karaçar, D. R. Russell and G. C. Saunders, J. Chem. Soc., Dalton Trans., 1996, 3215. 22 J. Fawcett, S. Freidrichs, J. H. Holloway, E. G. Hope, V. McKee, M. Nieuweenhuyzen and G. C. Saunders, J. Chem. Soc., Dalton Trans., 1998, 1477. 23 gNMR, version 3.6, Cherwell Scientific Publishing Ltd., Oxford, 1995. 24 M. Karplus, J. Chem. Phys., 1959, 30, 11; J. Am. Chem. Soc., 1963, 85, 2870. Paper 8/05560B

ISSN:1477-9226    

DOI:10.1039/a805560b

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3771–3776 3771 Palladium diphenyl-2-pyridylphosphine complexes Athanasia Dervisi,a Peter G. Edwards,*a Paul D. Newman,a Robert P. Tooze,b Simon J. Coles a and Michael B. Hursthouse a a Department of Chemistry, University of Wales Cardiff, Cardiff, UK CF1 3TB b ICI Acrylics, Wilton, Middlesbrough, UK TS80 8JE Received 2nd July 1998, Accepted 30th September 1998 The stoichiometric reaction of diphenyl-2-pyridylphosphine (Ph2PC5H4N-2) and palladium(II) acetate aVorded the palladium(I) dimer [Pd2(m-Ph2PC5H4N)2(OAc)2] 1.Attempts to isolate the monomer, [Pd(m-Ph2PC5H4N)2(OAc)2], from 2: 1 stoichiometry reactions (Ph2PC5H4N: Pd), resulted only in the isolation of 1. When >3 mole equivalents of Ph2PC5H4N were used the zerovalent palladium complex [Pd(Ph2PC5H4N)3] was isolated. Other palladium(I) dimers with carboxylic or sulfonic anionic ligands have been prepared by a metathesis reaction of the co-ordinated acetates.The complexes 1 and [Pd2(m-Ph2PC5H4N)2(O2CCF3)2] 2 have been structurally characterised as head-to-tail dimers. The mononuclear complex [Pd(Ph2PC5H4N)2(O2CCF3)2] has been obtained by reaction of Ph2C5H4N with [Pd(CH3CN)2(O2CCF3)2]. Complex 1 reacts with dimethyl acetylenedicarboxylate to give [Pd2(m-Ph2PC5H4N)2- (OAc)2(m-MeO2CC]] CCO2Me)], while attempts to isolate an A-frame carbonyl complex were unsuccessful. Introduction Both in terms of their high activity and superior selectivity attained under mild conditions, complexes generated in situ from mixtures of palladium acetate, diphenyl-2-pyridylphosphine (Ph2PC5H4N-2) and sulfonic acids are highly eY- cient catalysts for the methoxycarbonylation of propyne to methyl methacrylate (MMA).1 With this in mind we have undertaken a study on the co-ordination and reaction chemistry of this catalyst system.Broadly speaking two catalytic mechanisms may be in operation, namely the hydride and the methoxycarbonyl cycles.Drent et al.2 have presented a compelling argument for the operation of the latter based on selectivity improvements obtained on increasing the steric bulk of the phosphine ligand. More recently others have provided evidence for the existence of the hydride cycle.3 Palladium complexes of arylphosphines show enhanced catalytic performance in carbonylation reactions when the well established triphenylphosphine ligand is replaced by Ph2PC5- H4N-2.1 Optimum rates and selectivities are realised when a protic acid co-catalyst whose conjugate base is weakly coordinating, e.g.trifluoromethanesulfonic acid, is present.4 However, much of the co-ordination chemistry of Ph2PC5H4N dimers has been established with halides as terminal donors and the possible non-innocent involvement of the secondary ligands has been largely ignored. The disposition of nitrogen and phosphorus donors in (Ph2PC5H4N) render it an excellent bridging ligand for late-transition metals, such that a host of homo- and hetero-binuclear dimers have been reported.5 The usual synthetic approach to the dimers is through the coupling of two preformed mononuclear complexes (Scheme 1).A change in the formal oxidation state of the metals is implicit in such a reaction. The resulting dimers exist in two geometric forms: the more stable head-to-tail (H-T) isomer has the two like donors bound to diVerent metals, whereas the head to head (H-H) isomer has the nitrogen and phosphorus donors at the same metal (Scheme 1).Several examples of the less stable H-H isomers have been isolated.5a,f,6 Prompted by the above observations, we report here aspects of the co-ordination chemistry of Ph2PC5H4N with palladium in the presence of oxygen-bearing secondary ligands, including acetate. Unlike the known systems with halides, the oxyligands promote the formation of palladium(I) dimers from solely palladium(II) precursors.Results and discussion Palladium(I) dimers Phosphorus NMR data for the compounds discussed below are collected in Table 1. The dimer [Pd2(m-Ph2PC5H4N)2(OAc)2] 1 is obtained in high yield from the reaction of palladium(II) acetate and Ph2PC5H4N in a 1: 1 molar ratio. This contrasts with analogous compounds containing halides as terminal donors; dimers are only obtained when two preformed mononuclear species are combined (Scheme 1). In the present case, the redox chemistry is not solely metal-based, rather two acetate groups undergo a reductive elimination to generate a palladium(0) species which reacts with a palladium(II) complex to give the palladium(I) dimer and the elimination product, diacetyl peroxide (Scheme 2).The inability to synthesize Pd(Ph2PC5H4N)2X2 (X = CH3- CO2 2, CF3CO2 2, CH3SO3 2, p-H3CC6H4SO3 2 or CF3SO3 2) compounds by simple mixing (see below) is indicative of their inherent instability toward reductive elimination. It is well established that triphenylphosphine can reduce palladium(II) to palladium(0) when oxygen-bearing anions are present,7 e.g.[Pd(PPh3)2(OAc)2] is reduced spontaneously (but slowly) to a palladium(0) species with concomitant production of phosphine oxide.8 When a 1 : 1 mixture of Pd(OAc)2 and Ph2PC5H4N is monitored by 31P NMR spectroscopy the initial spectrum reveals a number of signals between d 10 and 20. Complex 1 (d 4.5) is present as a minor product 20 min after mixing, with a species giving a signal at d 9.4 predominating.The peak at d 9.4 is Scheme 13772 J. Chem. Soc., Dalton Trans., 1998, 3771–3776 assigned to [Pd(Ph2PC5H4N)2(OAc)2] which could not be isolated; the remaining peaks are presumably due to other intermediates. As the reaction proceeds the signals due to [Pd(Ph2PC5H4N)2(OAc)2] and other intermediates decay while that for 1 grows until it is the major component after 24 h following which 1 can be isolated in good yield (ca. 80%).Our failure to isolate the bis(acetate) complex is due to the spontaneous elimination of both acetates to form a zerovalent complex of lower co-ordination number (akin to the PPh3 analogue); evidence for the formation of the unstable diacyl peroxide species (Scheme 2) appears to account for the fate of the acetate groups (see below). The unassigned signals are likely to arise from diVerent isomers of this palladium(0) intermediate where diphenyl-2-pyridylphosphine may act as a chelating, bridging or mono co-ordinated ligand and which then reacts with Pd(OAc)2 or [Pd(Ph2PC5H4N)2(OAc)2] to form the dimer 1.The unstable diacyl peroxide species is likely to decompose to acetic anhydride and oxygen as indicated in the 13C-{1H} NMR spectrum of the mixture after the reaction is complete; signals at d 21.13 and 165.40 being assigned to acetic anhydride. Acetic anhydride is also observed in the 1H NMR spectrum of this mixture along with a trace of acetic acid, presumably arising from trace hydrolysis due to adventitious water; these resonances account for approximately half of the total acetate CH3 resonance intensities.Their identity is confirmed by further addition of an acetic anhydride–acetic acid mixture to the NMR sample. Loss of oxygen from the elimination by-product was further established by the addition of 2,3-dimethylbut-2-ene to the reaction mixture. Formation of the corresponding epoxide was observed by GCMS analysis.When iodomethane was added to a 1 : 2 mixture of Pd(OAc)2 and Ph2PC5H4N, monomeric [Pd(Ph2PC5H4N)2I2 and dimeric Scheme 2 Table 1 The 31P NMR data for the Ph2PC5H4N complexes Compound 1 [Pd2(m-Ph2PC5H4N)2(OAc)2] 2 [Pd2(m-Ph2PC5H4N)2(O2CCF3)2] 3 [Pd2(m-Ph2PC5H4N)2(O3SCH3)2] 4 [Pd2(m-Ph2PC5H4N)2(O3SCF3)2] 5 [Pd2(m-Ph2PC5H4N)2(O3SC6H4CH3-p)2] 6 [Pd(Ph2PC5H4N)2(O2CCF3)2] 7 [Pd(Ph2PC5H4N)2(C]] ] CPh)2] 8 [Pd(Ph2PC5H4N)2(C]] ] CMe)2] 9 [Pd2(m-Ph2PC5H4N)2(OAc)2(m-DMAD)] 10 [Pd(Ph2PC5H4N)3] b Ph2PC5H4N 31P-{1H} (d) a 14.5 10.1 23.3 22.4 23.1 114.8 123.7 123.4 131.6 121.7 26.7 a In CDCl3, unless otherwise mentioned.b In d6-benzene. [Pd2(Ph2PC5H4N)2I2] were obtained. The methyl iodo complex, [Pd(Ph2PC5H4N)2(CH3)I], which might be expected from oxidative addition to a palladium(0) intermediate, was not observed. The mononuclear diodo complex could arise from successive additions of MeI to a palladium intermediate with the liberation of ethane. It is noteworthy that Amatore et al.8 reported the formation of the oxidative addition product [Pd- (PPh3)2(Ph)I] from the reaction of [Pd(PPh3)2] with iodobenzene.The 31P NMR spectrum of a 1 : 2 Pd(OAc)2–Ph2PC5H4N mixture is substantially diVerent from that of the 1 : 1 mixture. A very broad signal appears at d 26.2 (in CH2Cl2) with a second at d 16.2 but neither could be assigned. At the end of the reaction several phosphorus containing species were observed, but only the dimer 1 was isolated upon crystallisation.Ratios of pyridyl phosphine to Pd(OAc)2 greater than 3 : 1 aVord the zerovalent complex [Pd(Ph2PC5H4N)3] which has been isolated and characterised as detailed in the Experimental section. In contrast to previous observations with triphenylphosphine, 9 formation of Ph2PC5H4N oxide is not taking place in the stoichiometric mixture of Ph2PC5H4N–Pd(OAc)2, as evidenced by the absence of a resonance at d 20 in the final spectrum of the reaction mixture.The possible formation of the N-oxide is also excluded as all the phosphine ligand in the 1 : 1 mixture is utilised to form the palladium complex 1. The operation of an alternative reaction pathway for Ph2PC5H4N with respect to PPh3 is likely to be a result of the second available (pyridine) donor in Ph2PC5H4N. Thus, stable dimeric complexes such as 1 form readily when the palladium(0) intermediate species is generated in situ in the presence of unchanged Pd(OAc)2; the bridging configuration of the phosphine in the palladium(I) dimeric product would no doubt render the ligand less labile and hence less available for subsequent oxidation than in related PPh3 systems.In the case of PPh3 such an intermediate is likely to decompose in the absence of any other substrate along with (in the presence of an appropriate oxygen donor) formation of the phosphine oxide. The detection of acetic anhydride in the final solution suggests an electron transfer directly from the acetate to the metal centre without Ph2PC5H4N involvement and is probably best described as a reductive elimination in the classical sense.The dimers 2–5 are conveniently prepared by the addition of one mol equivalent of the protic acid of the appropriate oxyanion to the 1 : 1 mixture of Pd(OAc)2 and Ph2PC5H4N. Anion exchange proceeds readily as the sulfonic and trifluoroacetic acids employed herein are appreciably stronger protic acids than is acetic.X-Ray quality crystals of [Pd2(Ph2PC5H4N)2(OAc)2]?CH2Cl2 1 and [Pd2(Ph2PC5H4N)2(O2CCF3)2]?0.5CH2Cl2 2 were obtained by slow diVusion of light petroleum into solutions of the appropriate complex in dichloromethane. There are two independent molecules of 2 in the asymmetric unit which, although crystallographically distinct, are structurally similar. The molecular structures of 1 and one of the independent molecules of 2 with the adopted numbering scheme are shown in Figs. 1 and 2.Selected bond lengths and angles are summarised in Table 2. Although several structures of complexes with bridging Ph2C5H4N ligands have been reported,5a,f,6b,10 the structures of 1 and 2 are the first to contain an unsupported palladium–oxygen bond. The Pd–Pd bond lengths of 2.579(1) and 2.561(1) Å, respectively, are the shortest reported for the palladium(I)– Ph2PC5H4N dimeric series {cf. 2.597 Å for [Pd2(Ph2PC5H4N)2I2] and 2.594 Å for [Pd2(Ph2PC5H4N)2Cl2] 5a,11} reflecting the weaker interaction of oxygen donors with the metal.In both 1 and 2 the two Pd–O bond lengths are quite distinct being 2.189(3), 2.139(4) and 2.183(6), 2.307(8) Å, respectively. There appears to be a correlation between the Pd–O bond length and the accompanying P–Pd–O bond angle: the longer terminal Pd–O bonds in both complexes are associated with the tighter P–Pd–O angle (Table 2) and it would appear that the steric bulkJ. Chem. Soc., Dalton Trans., 1998, 3771–3776 3773 of the two phenyl substituents on the phosphorus atoms causes a lengthening of the Pd–O bond as the P–Pd–O angle contracts.The Pd–O distances of 2.18 ± 1 Å in 1 and 2 compare well with those observed in [Pd2(dppm)2(O2CCF3)2].12 The mean of the Pd–O lengths is considerably shorter for the acetate ligand reflecting its better donor ability with respect to the more weakly co-ordinating trifluoro analogue. However, due to the trans influence of the metal–metal bond 13 and the larger radius of PdI, both are longer than equivalent Pd–O bonds in mononuclear palladium(II) compounds.The Pd–P (1, average 2.199; 2, 2.207 and 2.209 Å for the second independent molecule) and Pd–N (1, average 2.116; 2, 2.112 and 2.128 Å for the second independent molecule) distances are similar to those for the iodo- (Pd–P, average 2.212 and Pd–N, 2.112 Å) and chloro- (Pd–P, average 2.205 and Pd– N, 2.114 Å) analogues. The Pd–P distances are shorter than those found in complexes of bridging diphosphines and this is attributed to the relative trans influences of the P and N atoms. The Pd–Pd–O angles deviate from linearity by 9.66/7.18 for 1 and 11.6/8.18 for 2 and are transoid with respect to the M–M axis in both complexes as opposed to the dimers [Pd2(dppm)2- (O2CCF3)2] and [Pd2(Ph2PC5H4N)2I2] where the terminal Fig. 1 Molecular structure and atom labelling scheme of [Pd2(Ph2- PC5H4N)2(OAc)2]. Hydrogen atoms are omitted for clarity. Table 2 Selected bond lengths (Å) and angles (8) for [Pd2(Ph2PC5- H4N)2(OAc)2] 1 and [Pd2(Ph2PC5C4N)2(O2CCF3)2] 2 1 Pd(1)–N(2) Pd(1)–P(1) Pd(1)–Pd(2) Pd(1)–O(1) N(2)–Pd(1)–P(1) P(2)–Pd(2)–Pd(1) N(2)–Pd(1)–Pd(2) O(3)–Pd(2)–Pd(1) N(1)–Pd(2)–P(2) N(2)–Pd(1)–Pd(2) 2.117(5) 2.197(2) 2.5795(7) 2.189(3) 174.2(1) 80.7(2) 93.7(1) 170.3(1) 173.9(1) 93.7(1) Pd(2)–O(3) Pd(2)–N(1) Pd(2)–P(2) N(2)–Pd(1)–O(1) P(2)–Pd(2)–O(3) P(1)–Pd(1)–Pd(2) N(1)–Pd(2)–O(3) O(1)–Pd(1)–P(1) O(1)–Pd(1)–Pd(2) 2.139(4) 2.114(4) 2.201(2) 92.1(2) 98.0(1) 80.61(4) 86.9(2) 93.5(1) 173.0(1) 2 Pd(1)–N(2) Pd(1)–P(1) Pd(1)–Pd(2) Pd(2)–O(1) N(2)–Pd(1)–P(1) N(2)–Pd(1)–O(3) N(2)–Pd(1)–Pd(2) O(3)–Pd(1)–Pd(2) N(1)–Pd(2)–P(2) N(1)–Pd(2)–Pd(1) 2.110(7) 2.197(3) 2.561(1) 2.183(6) 174.7(2) 92.2(3) 93.6(2) 168.4(2) 173.6(2) 94.4(2) P(1)–C(5) Pd(1)–O(3) Pd(2)–N(1) Pd(2)–P(2) P(2)–Pd(2)–Pd(1) P(1)–Pd(1)–O(3) P(1)–Pd(1)–Pd(2) N(1)–Pd(2)–O(1) O(1)–Pd(2)–P(2) O(1)–Pd(2)–Pd(1) 1.817(8) 2.307(8) 2.114(7) 2.216(2) 80.62(8) 92.9(2) 81.11(9) 87.9(2) 97.6(2) 171.9(2) ligands are cisoid.5a,12 The palladium centres have planar co-ordination geometries with the angles between the palladium–ligand bonds varying from 80.6(4) to 98.0(1)8 in both complexes. Unlike [Pd2(Ph2PC5H4N)2I2], the two halves of the dimer are quite distinct with respect to their L–Pd–L9 bond angles.The Pd–N–C–P–Pd five-membered rings are nonplanar, and the eight–membered framework adopts a twisted boat conformation. The phosphorus atoms in the binuclear complexes 1–5 are equivalent giving a sharp singlet in their 31P-{1H} NMR spectra.It is reasonable to assign the solution species as H-T isomers in accord with the solid-state structures presented above and the established thermodynamic stability of such a geometry. Palladium dimers of Ph2PC5H4N preferentially adopt a H-T arrangement as the labile nature of the Pd–P bond allows this thermodynamically more stable isomer to predominate. Variable temperature 31P-{1H} NMR studies show no evidence of fluxionality of the Ph2PC5H4N ligand, precluding the possibility of H-T�H-H isomerisation.The 31P-{1H} NMR data show that there is a tend in d(31P) as a function of the anionic ligand, with the dimers 2–5 containing the less basic trifluoroacetate and especiay sulfonate groups showing 31P shifts to high field of those of the parental acetate dimer. The 1H NMR spectra of the complexes closely resemble that of uncomplexed Ph2PC5H4N.At least three of the pyridyl protons give discrete signals, whereas two unresolved multiplets appear for the phenyl groups. The only signifi- cant co-ordination shift is for the proton ortho to the phosphorus centre, H(6), which resonates between 0.4 and 0.7 ppm upfield of its position in the spectrum of free Ph2PC5H4N. Changes in the electronic framework of the aromatic rings on co-ordination are inferred from the 13C NMR spectra, where low-field shifts of up to 5 ppm are observed for all carbons in the pyridine ring with respect to those of Ph2PC5H4N.The phenyl carbons are largely unaltered except for the ipsocarbons, C(1), which shift 8 ppm upfield upon co-ordination. In the solid state IR spectrum of complex 1 the acetates have n(C]] O) and n(C–O) at 1581 and 1366 cm21, respectively; these frequencies suggest that inferred structural information should be treated with caution in these systems since the former is lower than that reported for monodentate palladium acetates and compares to anticipated values for bridging or anionic acetates, whereas the latter is within the range expected for monodentate co-ordination.14 The symmetric carbonyl stretch is assigned at 1681 cm21 for 2, again within the expected range (1680–1720 cm21) for monodentate binding.12 For the complexes 3–5 broad, strong n(SO3) stretches are seen between 1260 Fig. 2 Molecular structure and atom labelling scheme of [Pd2(Ph2- PC5H4N)2(O2CCF3)2]. Hydrogen atoms are omitted for clarity.3774 J.Chem. Soc., Dalton Trans., 1998, 3771–3776 and 1006 cm21 which are split implying bound (monodentate) sulfonates.15 Palladium(II) complexes The only complex of the type [PdX2(Ph2PC5H4N)2], where X is an oxyanion, isolated during the present study was cis- [Pd(Ph2PC5H4N)2(O2CCF3)2] 6. Although as already detailed for the parent acetate system, such species are believed to be present during the early stages of reaction, they are unstable with respect to reductive elimination and give palladium(I) dimers as the only isolable compounds.The bis(ligand)bis(tri- fluoroacetate) complex was synthesized by the reaction of [Pd(CH3CN)2(O2CCF3)2] with 2 moles of Ph2PC5H4N. Both cis and trans isomers of [Pd(Ph2PC5H4N)2(O2CCF3)2] are possible, although the formation of only one is indicated by 31P NMR spectroscopy where a single phosphorus resonance is observed (d 14.8). Inspection of the 13C NMR spectrum confirms the geometry as cis, where pertinent ligand resonances occur as doublets showing coupling to only one phosphorus atom.It is known that 3JCP is usually of equivalent magnitude to 1JCP in trans isomers giving virtual triplets for carbons directly bound to phosphorus: 16 this is clearly not the case here, and the isolated complex is assigned a cis-[Pd(Ph2PC5H4N)2(O2CCF3)2] stereochemistry. Only palladium(I) dimers were isolated when the same procedure was applied for the attempted preparation of [Pd(Ph2- PC5H4N)2X2] (where X = CH3SO3 2, CF3SO3 2 or p-H3CC6H4- SO3 2).EVorts to procure palladium(II) complexes from other starting materials were also unsuccessful: the reaction of [Pd(Ph2PC5H4N)2Cl2] with two equivalents of CF3SO3Ag aVorded only the dimer 4, re-emphasising the inherent stability of these dimers. The alkynyl complexes [Pd(Ph2PC5H4N)2(C]] ] CPh)2] 7 and [Pd(Ph2PC5H4N)2(C]] ] CMe)2] 8 were obtained from the reaction of palladium acetate with diphenyl-2-pyridylphosphine and a ten-fold excess of the appropriate acetylene.Both white solids reveal an infrared band of medium intensity characteristic of n(C]] ] C) stretches (2105 cm21 for 7 and 2110 cm21 for 8). Alkynylpalladium(II) complexes can be obtained by oxidative addition of alkynyl halides to palladium(0) complexes or via metathesis reactions of halogenopalladium( II) complexes with Li(C]] ] CR).17 It has been reported recently that alkynylpalladium(II) complexes can be prepared by the direct reaction of a palladium(II) complex with a terminal alkyne.18 When complex 10 is dissolved in acetonitrile Ph2PC5H4N remains chelated to the palladium centre without a solvent molecule replacing the co-ordinated pyridyl nitrogen.Additionally, variable temperature 31P-{1H} NMR spectroscopy (CDCl3, £60 8C) showed a lack of exchange between the mono co-ordinated and chelated ligand. These results suggest formation of a kinetically rigid four-membered chelate.Balch and co-workers 19 have shown that [Pd2(Ph2PC5H4N)2- Cl2] reacts with CO to give an unbridged Pd–Pd complex with terminal carbonyl ligands, as indicated by the presence of new bands at 2019 and 1994 cm21 in the solution infrared spectrum. Attempts to isolate the carbonyl complex resulted only in the recovery of [Pd2(Ph2PC5H4N)2Cl2]. Exposure of a CDCl3 solution of complex 1 to 1 atm of carbon monoxide causes no change in the 1H or 31P-{1H} NMR spectra, although new infrared bands appear in solution.A very strong sharp absorbance arises at 2253 cm21 and although not assigned is very close to the value of 2248 cm21 reported for [Pd(CO)4]21,20 while others at 1820.9, 1793.4 and 1712.9 cm21 may indicate the presence of bridging carbonyls. The solid that precipitates from solution was isolated and its IR spectrum recorded as a KBr disc. A medium intensity band at 1873 cm21 was observed while the rest of the spectrum remains unchanged.The available data show an interaction of 1 with carbon monoxide in solution although the nature of the species is not clear since the NMR and IR spectra of the adduct are otherwise very similar to those of 1. Despite the reluctance of the dimers to insert carbon monoxide into the Pd–Pd bond, [Pd2(Ph2PC5H4N)2(OAc)2] 1 reacts readily with DMAD (MeO2C]] ] CO2Me) to give the expected dimetallated alkene product 9, in accord with the previously reported chloride analogue.Other unactivated alkynes such as butyne and terminal alkynes did not aVord the corresponding insertion products. The two phenyl rings attached to each phosphorus atom in the co-ordinated Ph2PC5H4N ligands are no longer equivalent in complex 9. Consequently, two sets of resonances appear in the 1H and 13C-{1H} NMR spectra, one for the phenyl ring facing the acetate group and one for the ring facing the alkene group. In the IR spectrum a strong doublet appears due to the carbonyl groups of DMAD at 1726.4 and 1697.5 cm21 [n(C]] O)].An absorption due to the C]] C bond could not be assigned. Conclusion Equimolar solutions of palladium(II) acetate and Ph2PC5H4N are unstable with respect to reductive elimination to give initially palladium(0) species which appear to conproportionate with PdII to give ultimately palladium(I) dimers. No external reductant is required for this reaction which occurs more readily than with Ph3P in place of Ph2PC5H4N or than for palladium halides.These observations may be relevant to the alkoxycarbonylation of alkynes where palladium diphenyl-2- pyridylphosphine complexes with oxyanions are known to be far more eVective catalysts than triphenylphosphine analoges or than with palladium halides. Experimental Reactions were performed under an atmosphere of nitrogen using standard Schlenk techniques, solid complexes being stored in a desiccator under air. All solvents were refluxed under nitrogen over sodium–benzophenone and were distilled immediately prior to use with the exception of dichloromethane which was dried over CaH2 and toluene which was refluxed over sodium.Light petroleum had boiling point 40–608. The 31P- {1H} NMR spectra (referenced to 85% H3PO4) were collected on a JEOL FX90Q spectrometer operating at 36.2 MHz, 1H (400.13) and 13C-{1H} (100 MHz) spectra on a Bruker DPX400 unless stated otherwise and referenced to SiMe4. The NMR assignments are with respect to the labelling scheme shown below.Infrared spectra were recorded as KBr discs on a Nicolet 510 FT-IR spectrophotometer. The compounds [PdCl2(PhCN)2], [Pd2(dba)3] (dba = dibenzylideneacetone) 21 and [PdCl2(Ph2PC5H4N)2] 5a,c were prepared by literature procedures. The [Pd(CH3CN)nX2] (X = O2CCF3, O3SCH3, O3SCF3, or O3SC6H4CH3-p) complexes were prepared by a procedure similar to that reported for [Pd(CH3CN)4(O3SCF3)2].22J. Chem. Soc., Dalton Trans., 1998, 3771–3776 3775 Preparation Ph2PC5H4N.A modified literature procedure 23 was followed for the synthesis of Ph2PC5H4N. Sodium metal (1.06 g, 46 mmol) was dissolved in liquid ammonia (100 ml) at 280 8C and triphenylphosphine (6.00 g, 23 mmol) added as a solid. The resultant mixture was stirred for 30 min. A diethyl ether solution of tert-butyl chloride (10 ml, 23 mmol) was then added to destroy the phenylsodium formed, followed by a diethyl ether solution (10 ml) of 2-bromopyridine (2.2 ml, 23 mmol).The temperature was allowed to rise slowly (8 h) to room temperature during which time the ammonia evaporated to leave a yellow solid. Saturated aqueous ammonium chloride was added to the solid, and the aqueous phase extracted with CH2Cl2 (3 × 50 ml). The organic extracts were combined and the solvent removed under reduced pressure. The product was obtained as white crystals after recrystallisation from hot light petroleum. Yield 5.0 g, 83%. (Found: C, 77.3; H, 5.4; N, 5.3.Calc. for C17H14NP: C, 77.56; H, 5.57; N, 5.32%). 1H NMR (CDCl3, d): 7.30 (H6, d), 7.76 (H7, t), 7.36 (H8, t) and 8.86 (H9, d). 13C NMR/DEPT (CDCl3, d): 136.2 (C1, d, J 11.6), 134.2 (C2, d, J 20.0), 128.7 (C3, d, J 7.4), 129.1 (C4, s), 164.0 (C5, d, J 3), 135.8 (C6, d, J 2.1), 127.9 (C7, d, J 15.2), 122.2 (C8, s), and 150.4 (C9, d, J 13 Hz). [Pd2(Ph2PC5H4N)2X2] 1–5. A solution of Ph2PC5H4N (0.24 g, 0.9 mmol) in dichloromethane (10 ml) was added slowly to a suspension of palladium(II) acetate (0.2 g, 0.9 mmol) in dichloromethane (20 ml).For complexes 2–5, one equivalent of the corresponding acid, HX, was subsequently added and the mixture stirred (3 h). The red solution was filtered through Celite and evaporated to dryness. The resultant red solid was washed with Et2O (2 × 30 ml) and dried in vacuo. Red crystals were obtained by slow diVusion of light petroleum into a dichloromethane solution. Typical yields >80%. Infrared as KBr discs, and NMR data in CDCl3.[Pd2(m-Ph2PC5H4N)2(OAc)2] 1 (Found: C, 53.7; H, 4.4; N, 3.7. Calc. for C38H34N2O4P2Pd2: C, 53.23; H, 4.00; N, 3.27%): IR n(C]] O) 1581.0s, n(C–O) 1365.6vs, 1318.7m cm21; 1H NMR d 6.90 (H6, br s), 7.70 (H7, t, 7.8 J Hz), 8.94 (H9, br s) and 1.44 (OAc); 13C NMR/DEPT d 133.6 (C2, t, J 5.6), 128.4 (C3, t, J 4.9), 130.1 (C4, s), 169.4 (C5, dd, J 70.1/5.3), 137.6 (C6, s), 129.4 (C7, d, J 24.1), 125.6 (C8, s), 152.4 (C9, t, J 5.6 Hz), 23.4 (CH3) and 176.1 (CO2 2).[Pd2(m-Ph2PC5H4N)2(O2CCF3)2] 2 (Found: C, 47.2; H, 2.9; N, 3.0. Calc. for C38H28F6N2O4P2Pd2: C, 47.28; H, 2.92; N, 3.08%): n(C]] O) 1681.2vs, n(C–O) 1409.4m, n(CF3) 1193.7/1131.2 cm21; 1H NMR d 6.88 (H6, br s), 7.76 (H7, t, J 7.8 Hz) and 8.77 (H9, br s); 13C NMR/DEPT d 128.5 (C1, t, J 27.9), 133.5 (C2, t, J 6.0), 129.0 (C3, t, J 5.3), 130.9 (C4, s), 167.7 (C5, dd, J 70.9/6.6), 138.5 (C6, s), 129.8 (C7, d, J 4.9), 126.2 (C8, s), 151.6 (C9, t, J 6.0), 116.0 (CF3, q, 292.5 Hz) and 160.3 (CO2 2, q, J 35 Hz).[Pd2(Ph2PC5H4N)2(O3SCH3)2] 3 (Found: C, 46.4; H, 3.7; N, 3.0. Calc. for C36H34N2O6P2Pd2S2: C, 46.52; H, 3.69; N, 3.01%): n(SO3) 1249.7vs, 1142.4s, 1097.6s, 1006.3vs (Ph2PC5H4N), 535.55s cm21; 1 H NMR d 6.70 (H6, br s), 7.70 (H7, t, J 7.8), 7.48 (H8, t, J 7.8 Hz), 8.97 (H9, br s), 2.06 (CH3SO3), 7.2–7.4 (Ph, m); 13C NMR/DEPT d 128.6 (C1, dd, J 58.3/4.9), 134.3 (C2, t), 129.6 (C3, t), 131.4 (C4, t), 167.9 (C5, dd, J 79/6.3), 139.2 (C6, s), 126.7 (C8, s), 153.4 (C9, t, J 6.3 Hz) and 39.1 (CH3, s).[Pd2(m-Ph2PC5H4N)2(O3SCF3)2] 4 (Found: C, 41.5; H, 3.5; N, 3.3. Calc. for C36H28F6N2O6P2Pd2S2: C, 41.68; H, 2.72; N, 2.70%): n(SO3) 1257.8vs, 1180s, 1032.0/1004.7s cm21; 1H NMR d 6.80 (H6, br s), 7.80 (H7, t, 7.8 Hz), 8.76 (H9, br s), 7.30–7.75 (Ph, m). [Pd2(m-Ph2PC5H4N)2(O3SC6H4CH3-p)2] 5 (Found: C, 52.4; H, 3.9; N, 2.5. Calc. for C48H42N2O6P2Pd2S2: C, 53.29; H, 3.91; N, 2.59%): n(SO3) 1193.7, 1031.2/1009.0vs cm21; 1H NMR d 6.64 (H6, br s), 7.69 (H7, t, J 7.6 Hz), 8.92 (H9, br s), 7.34 (Ph, m), 6.93 (tosyl, d, 8 Hz), 6.75 (tosyl, d, 8 Hz) and 2.22 (tosyl, s); 13C NMR/DEPT d 134.0 (C2, t, J 5.8), 129.0 (C3, t, J 5.8), 130.9 (C4, s), 167.9 (C5, d, J 71.9), 138.9 (C6, s), 125.9 (C8, s), 153.3 (C9, t, J 5.8 Hz), 141.7/139.3/128.2/26.1 (tosylate). cis-[Pd2(Ph2PC5H4N)2(O2CCF3)2] 6.A solution of Ph2PC5- H4N (0.25 g, 0.96 mmol) in acetonitrile (10 ml) was added slowly to a solution of [Pd(CH3CN)2(O2CCF3)2] (0.2 g, 0.48 mmol) in acetonitrile (20 ml) to give a pale yellow solution. After stirring for 2 h the solvent was evaporated and the resultant yellow oil stirred overnight with light petroleum to aVord a yellow powder (Found: C, 53.1; H, 3.2; N, 3.1.Calc. for C38H28F6- N2O4P2Pd: C, 53.13; H, 3.29; N, 3.26%). 1H NMR (CDCl3, d): 7.29 (H6, br), 7.78 (H7, br), 7.69 (H8, br) and 8.27 (H9, br). 13C NMR/DEPT (CDCl3, d): 125.2 (C1, d, J 17.4 Hz), 134.2 (C2, d, J 10.9), 128.8 (C3, d, J 11.8), 132.1 (C4, s), 156 (C5, d), 149.5 (C6, d, J 16.2), 130.8 (C7, d, 20.7 J Hz), 124.8 (C8, s), 137.8 (C9), 116.1 (CF3, q, J 292) and 160.3 (CO2, q, J 36 Hz).cis-[Pd2(Ph2PC5H4N)2(C]] ] CR)2] (R 5 Ph 7 or Me 8). To the red solution formed on addition of Ph2PC5H4N (0.48 g, 1.9 mmol) to a suspension of Pd(OAc)2 (0.2 g, 0.9 mmol) in CH2Cl2 (40 ml) was added phenylacetylene (1 ml) or propyne (1 ml) as appropriate. The complexes [Pd(Ph2PC5H4N)2(C]] ] CR)2] precipitated as white solids.[Pd(Ph2PC5H4N)2(C]] ] CPh)2] 7 (Found: C, 71.9; H, 4.4; N, 3.2. Calc. for C50H38N2P2Pd: C, 71.90; H, 4.59; N, 3.35%): 1H NMR (CDCl3, d 8.7 (d, 2H), 8.23 (br, 2H), 8.03 (br, 6H), 7.5 (t, 2H), 7.3 (m, 12H), 7.19 (m, 4H), 6.9 (m, 6H) and 6.34 (d, 4H). [Pd(Ph2PC5H4N)2(C]] ] CMe)2] 8 (Found: C, 68.2; H, 4.9; N, 4.0. Calc. for C40H34N2P2Pd: C, 67.56; H, 4.82; N, 3.94%). [Pd2(Ph2PC5H4N)2(OAc)2(Ï-DMAD)] 9. Dimethyl acetylenedicarboxylate (0.3 ml) was added to a solution of [Pd2- (Ph2PC5H4N)2(OAc)2] (0.09 g, 0.4 mmol) in toluene (30 ml) and the red solution stirred for 3 h.Addition of light petroleum (100 ml) precipitated a yellow solid (Found: C, 44.7; H, 3.2; N, 3.2. Calc. for C40H42N2O8P2Pd2: C, 46.44; H, 3.03; N 3.01%). IR (KBr disk, cm21): n(DMAD) 1726.4s, 1697.5s, 1263.5s, 1217.5s, (OAc) 1583.7s, 1386.9m, 1327.1m, (Ph2PC5H4N) 1435.1s, 1096.6s, 696.35m and 534.32s. 1H NMR (CDCl3, d): 7.58 (H6, m), 9.40 (H9, br), 7.98 (m, 4H), 7.76 (m, 4H), 7.45 (m, 4H), 7.25 (m, 4H), 2.86 (s, 3H, DMAD) and 1.35 (s, OAc). 13C-{1H} NMR (CDCl3, d): 133.1 (o-C of Ph, d, J 50), 132.0 (o-C of Ph, d, J 46), 129.6 (p-C of Ph, s), 129.2 (p-C of Ph, s), 127.5 (m-C of Ph, d, J 43), (m-CPh, d, J 43 Hz), 161.2 (s, O2CCH3), 14.27 (s, O2CCH3), 176.46 (s, CO2CH3) and 49.88 (s, CO2CH3). [Pd2(Ph2PC5H4N)3] 10. A solution of Ph2PC5H4N (1.66 g, 6.3 mmol) in methanol (5 ml) was added to a suspension of Pd(OAc)2 (0.2 g, 0.9 mmol) in methanol (20 ml).The mixture was stirred for 5 h during which time a bright yellow solid precipitated. Diethyl ether (50 ml) was added to complete precipitation and the resultant yellow solid filtered oV, washed with Et2O (2 × 30 ml) and dried in vacuo. Yellow crystals of complex 10 were obtained by slow diVusion of light petroleum into a toluene solution of the complex (Found: C, 71.0; H, 4.6; N, 4.7. Calc. for C51H42N2P2Pd: C, 68.35; H, 4.72; N, 4.69%). 1H NMR (C6D6, d): 8.46 (d, 2H), 7.78 (m, 12H), 7.42 (m, 3H), 7.12 (m, 18H), 6.86 (t, 3H) and 6.52 (dd, 3H).Crystallography Data for compounds 1 and 2 were recorded on a FAST TV Area detector diVractometer, with a molybdenum target [l(Mo-Ka) = 0.71069 Å], equipped with an Oxford Cryosystems cryostat and driven by MADNES software operating on a MicroVax 3200 computer, following previously described procedures.24 Crystals of 1 and 2 were mounted on glass fibres using the oil drop technique and collected at 150 and 120 K respectively.The structures were solved via heavy atom methods3776 J. Chem. Soc., Dalton Trans., 1998, 3771–3776 Table 3 Crystal data for [Pd2(m-Ph2PC5H4N)2(OAc)2] 1 and [Pd2(m-Ph2PC5H4N)2(O2CCF3)2] 2 Empirical formula Formula weight Crystal system Space group a/Å b/Å c/Å b/ 8 U/Å3 Z Dc/Mg m23 q/mm21 F(000) Crystal size/mm q range/8 Index ranges Reflections collected Independent reflections Rint Data/restraints/parameters Goodness of fit on F2 Final R, wR2 [I > 2s(I)] (all data) Largest diVerence peak and hole/e Å23 Absolute structure parameter 1 C38H34N2O4P2Pd2?CH2Cl2 940.06 Monoclinic P21/a 12.532(2) 20.159(4) 14.7610(6) 104.24(3) 3614.7(10) 4 1.663 1.201 1812 0.18 × 0.07 × 0.145 1.75 to 25.01 210 < h < 14, 223 < k < 23, 217 < l < 16 14885 5391 0.0959 5383/0/455 1.027 0.0478, 0.1067 0.0630, 0.1143 1.956 and 20.992 2 C38H28F6N2O4P2Pd2 965.08 Monoclinic P21 12.413(9) 20.703(10) 15.580(9) 108.53(10) 3796(4) 4 1.762 1.176 1992 0.2 × 0.11 × 0.2 1.84 to 25.03 213 < h < 13, 22 < k < 21, 18 < l < 18 16331 9636 0.0689 9628/1/1000 1.059 0.0438. 0.1078 0.0472, 0.1112 1.436 and 20.874 0.02(3) (SHELXS),25 and then subjected to full matrix least-squares refinement on Fo 2 (SHELXL 93).26 Non-hydrogen atoms were made anisotropic, with hydrogens in calculated positions (C]H 0.96 Å, with Uiso tied to Ueq of the parent atoms). The weighting scheme used was w = 1/[(2Fo 2)]. The disordered solvent molecule present in 1 was freely refined as two components, with the major part having 75% occupancy.Absorption corrections were applied using DIFABS27 and diagrams drawn with SNOOPI.28 Further details are given in Table 3. CCDC reference number 186/1182. See http://www.rsc.org/suppdata/dt/1998/3771/ for crystallographic files in .cif format. Acknowledgements We are grateful to ICI Acrylics for financial support for this study (A. D. and P. D. N.) and to EPSRC for support for the crystallography unit at CardiV.References 1. E. Drent, D. Arnoldy and P. H. M. Budzelaar, J. Organomet. Chem., 1993, 455, 247. 2 E. Drent, W. W. Jager, J. J. Keijsper and F. G. M. Niele, in Applied homogeneous catalysis with organometallic compounds, eds. B. Cornils and W. A. Hermann, VCH, Weinheim, 1996, vol. 2, p. 1119. 3 A. Scrivanti, V. Beghetto, E. Campagna, M. Zanato and U. Matteoli, Organometallics, 1998, 17, 630. 4 E. Drent, D. Arnoldy and P. H. M. Budzelaar, J.Organomet. Chem., 1994, 475, 57. 5 (a) Y. Xie, C. Lee, Y. Yang, S. Rettig and B. R. James, Can. J. Chem., 1991, 70, 751; (b) J. P. Farr, M. M. Olmstead, C. H. Hunt and A. L. Balch, Inorg. Chem., 1981, 20, 1182; (c) G. De Munno, J. Organomet. Chem., 1993, 450, 263; (d ) J. P. Farr, M. M. Olmstead and A. L. Balch, Inorg. Chem., 1983, 22, 1229; (e) M. L. Kullberg and C. P. Kubiak, Organometallics, 1984, 3, 632; ( f ) J. P. Farr, M. M. Olmstead, N. M. Rutherford, F. E. Wood and A.L. Balch, Organometallics, 1983, 2, 1758; ( g) C. G. Arena, F. Faraone, M. Lanfranchi, E. Rotondo and A. Tiripicchio, Inorg. Chem., 1992, 31, 4797. 6 (a) F. E. Wood, J. Hvoslef, H. Hope and A. L. Balch, Inorg. Chem., 1984, 23, 4309; (b) J. P. Farr, F. E. Wood and A. L. Balch, Inorg. Chem., 1983, 22, 3387. 7 L. Malatesta and M. Angoletta, J. Chem. Soc., 1957, 1186. 8 C. Amatore, E. Carre, A. Jutand and M. A. M’Barki, Organometallics, 1995, 14, 1818. 9 C. Amatore, E. Carre, A. Jutand and M. A. M’Barki, Organometallics, 1992, 11, 3009. 10 Z.-Z. Zhang, H.-K. Wang, H.-G. Wang and R.-J. Wang, J. Organomet. Chem., 1986, 314, 357. 11 T. Suzuki, N. Iitaka, S. Kuradi, M. Kita, K. Kashiwava, S. Ohba and J. Fujita, Bull. Chem. Soc. Jpn., 1992, 65, 1817. 12 T. E. Kraft, C. I. Hejna and J. S. Smith, Inorg. Chem., 1990, 29, 2682. 13 D. J. Wink, Acta Crystallogr., Sect. C, 1990, 46, 56. 14 K. Nakamoto, Infrared and raman spectroscopy of inorganic and coordination compounds, Wiley, New York, 4th edn., 1986, p. 232. 15 S. P. Gejji, K. Hermansson and J. Lindgren, J. Phys. Chem., 1993, 97, 3712. 16 G. Balimann, H. Motschi and P. S. Pregosin, Inorg. Chim. Acta, 1977, 23, 191. 17 M. Akita, M. Terada, S. Oyama and Y. Moro-oka, Organometallics, 1990, 9, 816; J. H. Nelson, A. W. Verstuyft, J. D. Kelly and H. B. Jonassen, Inorg. Chem., 1974, 13, 27; H. Hoberg and H. J. Riegel, J. Organomet. Chem., 1983, 241, 245; H. F. Klein, B. Zettel, U. Florke and H. J. Haupt, Chem. Ber., 1992, 125, 9. 18 A. Bacchi, M. Carcelli, M. Costa, P. Pelagatti and C. Pelizzi, Gazz. Chim. Ital., 1994, 124, 429. 19 A. Maisonnat, J. P. Farr and A. L. Balch, Inorg. Chim. Acta, 1981, 53, L217. 20 M. Bochmann, Organometallics 1: complexes with transition metal carbon s-bonds, Oxford chemistry primers, 1996, p. 13. 21 M. F. Rettig and P. M. Maitlis, Inorg. Synth, 1977, 17, 134. 22 E. Drent, J. Organomet. Chem., 1991, 417, 235. 23 P. A. A. Klusener, J. C. L. Sukerbuyk and P. A. Verbrugge, Eur. Pat. Appl., EP-A-499328I, 1992. 24 S. R. Drake, M. B. Hursthouse, K. M. A. Malik and S. A. S. Miller, Inorg. Chem., 1993, 32, 4653. 25 G. M. Sheldrick, Acta Crystallogr., Sect. A, 1990, 46, 467. 26 G. M. Sheldrick, SHELXL 93, University of Göttingen, 1993. 27 N. P. C. Walker and D. Stuart, Acta Crystallogr., Sect. A, 1983, 39, 158. Adapted for FAST geometry by A. Karaulov, University of Wales, CardiV, 1991. 28 K. Davis and K. Prout, University of Oxford, 1993. Paper 8/05102J

ISSN:1477-9226    

DOI:10.1039/a805102j

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3777–3784 3777 Synthesis and electrospray mass spectrometry of palladium(II) diphosphine complexes from oxidative addition of 2-bromopyridine to Pd0 Clavius C. H. Chin,a Jeremy S. L. Yeo,a Z. H. Loh,a J. J. Vittal,a W. Henderson b and T. S. Andy Hor *a a Department of Chemistry, Faculty of Science, National University of Singapore, Kent Ridge, 119260 Singapore b Department of Chemistry, University of Waikato, Private Bag 3105, Hamilton, New Zealand Received 19th May 1998, Accepted 22nd September 1998 Oxidative addition reactions of palladium(0) phosphine complexes with 2-bromopyridine gave rise to a series of structurally distinctive complexes, namely trans-(N,P)-[Pd2Br2(PPh3)2(m-C5H4N-C2,N)2] 1, [Pd2(m-C5H4N-C2,N)2- (m-dppm)2]Br2 2, [Pd2(h1-dppp)2(m-C5H4N-C2,N)2(m-dppp)]Br2 3, trans-[{PdBr(h1-C5H4N-C2)(m-dppb)}n] 4, trans- (N,P)-[Pd2Br2(m-C5H4N-C2,N)2(m-dppb)] 5 and cis-[PdBr(h1-C5H4NH-C2)(h2-dppf)]Br 6 [Ph2P(CH2)nPPh2, n = 1(dppm), 3(dppp) or 4(dppb); dppf = Fe(Ph2PC5H4)2].Similarly, trans-(N,P)-[Pd2Cl2(m-C9H6N-C2,N)2(m-dppb)] 7 has been obtained from 2-chloroquinoline and [Pd(dppb)2]. An array of structural possibilities is envisaged based on the diVerent co-ordination modes of the pyridine (C or/and N bonded; terminal or bridging; pyridyl, pyridine or pyridinium), phosphine (terminal, bridging or chelating) bromide (terminal or ionised) ligands. Complexes 2 and 3, but not the others, can be obtained from phosphine exchange reactions of 1.Complexes 5 and 6 were analysed by X-ray single-crystal crystallographic methods. The former reveals a dinuclear structure with a dppb ligand bridging diagonally two metals that are juxtaposed by two syn-bridging pyridyl groups. It represents an unusual dinuclear core stabilised by two types of bridging ligands of contrasting steric and geometric demands. The latter shows a cationic and mononuclear square planar palladium(II) complex containing a chelating dppf, terminal bromide and an unusual C-bonded pyridyl group with the N-site protonated.The fragmentation of these complexes was investigated by electrospray mass spectrometry under diVerent cone voltages. Breakdown of the dinuclear framework is facilitated by addition of H–Br to the N–Pd bonds of the bridging pyridyl group. Introduction Since the turn of the century the chemistry of polymers has expanded enormously.1 Organic bifunctional polymers containing heterocyclic rings have wide applications as they exhibit properties such as optical non-linearity 2 and anisotropic electrical conductivity.3 Hitherto, these polymers have been used as electrochemical energy storage 4 and electrochromic devices,5 as molecular catalysts 6 and in molecular electronics.7 A convenient synthetic pathway for these polymers is the polymerisation of their respective monomers.The future applications of these polymers hence hinge on our ability chemically to manipulate their monomeric or oligomeric building blocks.Among the most successful methodologies for the current synthesis of these precursors are metal-catalysed hetero-coupling reactions, e.g. Grignard,8 Stille 9 and Suzuki 10 couplings, in which organometallic catalyst precursors, especially palladium(0) and palladium( II) phosphine complexes, are inevitably involved. The common use of PPh3 as supporting ligands leads to the synthesis of a range of monomeric complexes that are catalytically related.For example, oxidative addition reaction of 3- or 4- bromopyridine to [Pd(PPh3)4] results in trans-[PdBr(h1-C5H4NCn)( PPh3)2] (n = 3 or 4),11,12 which readily undergoes metathesis with anions such as N3 2, NCO2, NCS2, NCSe2, ClO4 2, BF4 2 and PF6 2.13 These mononuclear complexes are generally considered as catalytically active. By a careful choice of incoming nucleophile (Nuc), one would obtain [Pd(Nuc)(h1-C5H4NCn)( PPh3)2] (n = 3 or 4) which could reductively eliminate NH4C5-Nuc as a hetero-coupled product and regenerate [Pd(PPh3)2] as the active catalyst.Even in the cases when dinuclear complexes are formed, e.g. 2-bromopyridine giving trans-(N,P)-[Pd2Br2(PPh3)2(m-C5H4N-C2,N)2] 1,13 they could break down to mononuclear complexes upon ligand exchange. 14a The involvement of dinuclear palladium(II) complexes in the catalytic cycle is generally neglected despite their first synthesis back in 1980 12 and subsequent works.14 Accordingly, the current understanding of the mononuclear catalytic intermediates is inadequate for PdCl2(P–P) 15 which contain bidentate phosphines (P–P) such as dppe, dppp [Ph2P(CH2)3- PPh2] and dppf as supporting ligands. The literature is poorly developed on the structures of these catalytic intermediates and the co-ordination and structural roles of diphosphine on a dinuclear framework.In this paper, using 2-bromopyridine as a model, we examine the stabilising eVects and structural influence of diVerent diphosphines on a dinuclear skeleton.The complexes formed are characterised by X-ray single-crystal diVractometry and electrospray mass spectrometry (ESMS). Results and discussion Synthesis Oxidative addition of 2-bromopyridine to [Pd(PPh3)4] occurs readily at room temperature (r.t.) to give a near-quantitative yield of trans-(N,P)-[Pd2Br2(PPh3)2(m-C5H4N-C2,N)2] 1. The same complex was reported to be obtainable only under reflux conditions.13 Its identity is substantiated by NMR spectroscopy and ESMS (see below).Similar reaction occurs between [Pd2(dppm)3] and 2-bromopyridine in refluxing toluene to give [Pd2(m-C5H4N-C2,N)2(m-dppm)2]Br2 2. The 31P NMR analysis shows two sets of triplets indicating two pairs of phosphines which are chemically distinct. The parent molecular ion was detected by ESMS analysis (m/z ca. 569). Complex 2 is one of the few palladium dinuclear complexes with no terminal but3778 J.Chem. Soc., Dalton Trans., 1998, 3777–3784 four bridging ligands, namely two C- and N-bonded pyridyl and two dppm groups. It can also be obtained from the stoichiometric phosphine-exchange reaction of 1 with dppm. Similar reaction of [Pd(dppp)2] with 2-bromopyridine yields a third structural type of dinuclear complex [Pd2(h1-dppp)2- (m-C5H4N-C2,N)2(m-dppp)]Br2 3. It contains three dppp ligands, two of which are pendant and one is in a diagonalbridging mode; this is consistent with the three discrete resonances in the 31P NMR spectrum.The ESMS spectrum gives a molecular peak at m/z = 803.4, which agrees with the dicationic configuration. Similar to 2, phosphine exchange reaction of 1 with an excess of dppp yields 3 as the major product. Reaction between 2-bromopyridine and [Pd(dppb)2] [dppb = Ph2- P(CH2)4PPh2] gives diVerent products depending on the reaction temperatures. At r.t. a complex analysed to be trans- [{PdBr(h1-C5H4N-C2)(m-dppb)}n] 4 resulted. Under toluene reflux the major product is trans-(N,P)-[Pd2Br2(C5H4NC2, N)2(m-dppb)] 5.Although the nuclearity of complex 4 is presently unclear, its role as a precursor of 5 based on a trinuclear structural model can be proposed.† Conversion is achieved by basic attack of the free pyridyl nitrogen on the neighbouring metal followed by elimination of the phosphine (Scheme 1). The by-product in this conversion, cis-[PdBr(h1- C5H4N-C2)(h2-dppb)], can add to the yield of 5 by a simple dimerisation process with the elimination of phosphine. Similar reaction between 2-bromopyridine and [Pd(dppf)2] in refluxing toluene gives a dark red-brown complex characterised as cis- [PdBr(h1-C5H4NH-C2)(h2-dppf)]Br 6.The 31P NMR spectrum shows two sets of doublets expected for a mononuclear structure with inequivalent phosphine sites. There is no evidence for the expected product cis-[PdBr(h1-C5H4N-C2)(h2-dppf)].The free pyridyl nitrogen is probably very prone to protonation by a trace of HBr in the 2-bromopyridine.‡ The fact that 6 cannot be obtained from the exchange reaction of 1 with dppf also sup- Scheme 1 † The complex cannot be monomeric since the phosphine sites are chemically equivalent (31P NMR). A dimeric model is possible but not likely as it would entail an unusual [Pd2(m-dppb)2] moiety that is geometrically constrained. A trimeric model is possible as it has a reasonable mechanistic path for the generation of a monomeric and dimeric (5) pyridyl complexes under ESMS conditions (described later) although there is no definitive experimental support.The ESMS spectrum at 5 V only reveals dppbO2, M 1 H1 and 2M 1 H1. Addition of the pyridine only breaks up the “oligomer” and gives monomeric species. ‡ One alternative is that oxidation gives rise to cis-[Pd(h2-C5H4- N-C2,N)(dppf)]Br which rapidly undergoes H–Br addition across the N–Pd bond (in a manner similar to the formation of 2a from 2 described later) to relieve the stress on the strained chelate.ports the formulation of 6. Complex trans-(N,P)-[Pd2Cl2(m- C9H6N-C2,N)2(m-dppb)] 7, the chloroquinolyl analogue of 5, is isolated from a r.t. reaction between [Pd(dppb)2] and 2- chloroquinoline. Structures The isolated complexes illustrate a wide range of structural variations. All six complexes 1–6 are structurally distinct whereas 5 and 7 are isostructural. Complexes 1–3 and 5 are dinuclear with two bridging pyridyl ligands and none to two bridging phosphine ligands.Single-crystal X-ray diVraction analysis of 6 reveals a mononuclear d8 cation with a chelating dppf and an unusual C-co-ordinated pyridyl with the nitrogen site protonated giving a ligand best described as a neutral pyridinium (Fig. 1, Table 1). The close contact of N(1)H ? ? ? Br(2) [N(1)–H(11) 0.860, Br(2)–H(11) 2.367(4) Å; N(1)–H(11)–Br(2) 177.61(9)8] strongly supports the protonated model.The established higher trans influence of the Pd–C bond weakens significantly the opposite Pd–P bond [2.3863(8) Å] which faces Br. A similar phenomenon has been observed in a cis-[PdBr(aryl)(dppe)] complex.16 The P–Pd–P chelate angle [102.08(3)8] is comparable to those found in other square-planar complexes [97.5(1)–104.4(2)8].17 The staggered conformation of the ferrocenyl rings [t 40.8(6)8] is consistent with those in other square-planar chelates.18 The pyridyl ligand is found primarily in the forms of C-bonded pyridyl and pyridinium and C,N-bonded pyridyl.It is a terminal ligand C-bonded in both 4 and 6, with 6 having the nitrogen end protonated. The phosphine is bridging in most cases (2, 3, 4, 5 and 7) but chelating in 6 and unidentate in 3. In 5, the bridging phosphine is in the diagonal-bridging mode trans to the nitrogen end of pyridyl, while for 2 the diphosphine displays the syn-bridging mode and is necessarily trans to both C and N ends.In 3 the bridging dppp is trans to the N-donors whilst the terminal ones are opposite the carbon ends. These structural variations are attributed to both electronic and steric factors. The high trans-directing phosphine causes the N of the bridging pyridyl to adopt the trans-P position. This governs the diagonal-bridging mode for dppb in 5 and accounts for the chemically equivalent phosphine sites on a hetero-bridged system. In 3 the distinction is less obvious as there are phosphines Fig. 1 An ORTEP drawing (50% probability level) of the molecular structure of cis-[PdBr(h1-C5H4NH-C2)(h2-dppf)]Br 6.J. Chem. Soc., Dalton Trans., 1998, 3777–3784 3779 trans to both carbon and nitrogen ends. The two bridging pyridyl rings juxtapose the metal atoms and determine the Pd–Pd separation, thereby restricting the type of diphosphine ligands that are suitable for the diagonal-bridging mode. The long and flexible carbon skeletal chains of dppp and dppb do not favour a syn-bridge conformation but permit the diagonalbridge formation.These ideas are illustrated in the single-crystal X-ray molecular structure of 5 (Fig. 2). It is an unusual dinuclear structure showing how two sterically contrasting bridging ligands can be accommodated within a dinuclear framework. With the two square-planar palladium atoms locked by two pyridyl bridges in a syn conformation [dihedral angle between the Pd(1)–C(6)– N(2)–Pd(2) and Pd(1)–N(1)–C(1)–Pd(2) planes is 77.06(7)8], the dppb ligand takes advantage of its long alkyl chain by traversing in a diagonal-bridging mode.This forces the Br(1)– Pd(1) ? ? ? Pd(2)–Br(2) torsional angle to twist from an ideal 0 to 88.11(4)8. As a result the phosphines fit into the Pd2 pocket and are chemically equivalent, which is consistent with the solution NMR data. The similar Pd–P bond lengths [2.2609(10) and 2.2678 (11) Å] (and similar Pd–Br and Pd–N bond lengths on Fig. 2 An ORTEP drawing (50% probability level) of the molecular structure of trans-(N,P)-[Pd2Br2(m-C5H4N-C2,N)2(m-dppb)] 5. Table 1 Selected bond lengths (Å) and angles (8) for (a) trans-(N,P)- [Pd2Br2(m-C5H4N-C2,N)2(m-dppb)]?MeCN 5 and (b) cis-[PdBr(h1- C5H4NH-C2)(h2-dppf)]Br?0.5MeOH 6 (a) Pd(1)–C(6) Pd(1)–P(2) Pd(2)–N(2) Pd(2)–Br(2) N(1)–C(5) N(2)–C(10) C(6)–Pd(1)–N(1) N(1)–Pd(1)–P(1) N(1)–Pd(1)–Br(1) C(6)–Pd(1)–Pd(2) P(1)–Pd(1)–Pd(2) C(1)–Pd(2)–N(2) N(2)–Pd(2)–P(2) N(2)–Pd(2)–Br(2) C(1)–Pd(2)–Pd(1) P(2)–Pd(2)–Pd(1) 1.993(4) 2.2609(10) 2.096(3) 2.5196(5) 1.352(5) 1.348(5) 87.4(2) 169.31(9) 91.12(9) 64.29(12) 108.83(3) 87.10(14) 170.06(10) 90.62(9) 64.30(12) 109.68(3) Pd(1)–N(1) Pd(1)–Br(1) Pd(2)–C(1) Pd(2)–P(1) N(1)–C(1) N(2)–C(6) C(6)–Pd(1)–P(2) C(6)–Pd(1)–Br(1) P(1)–Pd(1)–Br(1) N(1)–Pd(1)–Pd(2) Br(1)–Pd(1)–Pd(2) C(1)–Pd(2)–P(2) C(1)–Pd(2)–Br(2) P(2)–Pd(2)–Br(2) N(2)–Pd(2)–Pd(1) Br(2)–Pd(2)–Pd(1) 2.093(3) 2.5197(5) 2.003(4) 2.2678(11) 1.352(5) 1.356(5) 92.52(12) 173.78(11) 90.95(3) 61.63(9) 119.28(2) 93.29(11) 172.35(11) 90.22(3) 61.64(10) 120.72(2) (b) Pd(1)–C(11) Pd(1)–P(1) P(1)–C(1) N(1)–C(11) N(1)–H(11) C(11)–Pd(1)–P(2) P(1)–Pd(1)–P(2) P(2)–Pd(1)–Br(1) N(1)–H(11)–Br(2) 2.037(3) 2.3863(8) 1.812(3) 1.341(5) 0.860 87.50(9) 102.08(3) 172.00(2) 177.61(9) Pd(1)–P(2) Pd(1)–Br(1) P(2)–C(6) N(1)–C(15) Br(2)–H(11) C(11)–Pd(1)–P(1) C(11)–Pd(1)–Br(1) P(1)–Pd(1)–Br(1) 2.2837(8) 2.4695(4) 1.804(3) 1.353(5) 2.367(4) 169.77(9) 84.50(9) 85.91(2) both palladium planes) also reflect such parity.It is notable that the Pd–Br [2.5197(5) Å] bond is significantly longer than that in 6 [2.4695(4) Å]. This is indicative of the higher trans influence of the pyridyl carbon in 5 (compared to the phosphino group in 6) and suggests that the terminal bromide is susceptible to ligand exchanges. This is verified in the formation of 2 and 3 from 1 and is an important consideration in the design of these Pd2 species as catalysts for hetero-couplings.Comparison of the crystal structures of 1 and 5 shows that the diphosphine ligand increases the torsional twist of Br–Pd ? ? ? Pd–Br. The compounds dppm and dppf, on the other hand, are too short and crowded respectively to fit into the diagonal-bridge ‘cavity’ and hence their equivalents of 5 cannot be obtained. For entropic reasons it is not favourable for 5 to take up two additional dangling dppb ligands like that in 3. The pin-wheellike structure of 2 is stabilised by four bridging ligands of comparable steric requirements.There is little driving force for such a structure to disrupt and give way to any structure with dangling phosphines. The steric demand of dppf favours a chelating mode that provides an impetus for 6 to be mononuclear. Although we have successfully assembled a quinolyl derivative of a tribridged system, viz. 7, it is worth noting that refluxing 2-chloroquinoline with [Pd2(dppm)3] in toluene, contrary to what is found with 2 or 7, does not result in the isolation of any quinolyl complex but an unexpected product, [Pd2Cl2- (m-dppm)2].One possible explanation is that, although similar oxidative addition can occur, a quinolyl-bridged structure analogous to that of 2 or 7 could break down under forcing conditions (Scheme 2). Chloride attack would open the bridge by converting a C,N-bonded to a C-bonded quinolyl group. Intramolecular reductive elimination would follow and yield 2,29-biquinoline and the observed product, [Pd2Cl2(m-dppm)2].Similar PdI–PdI complexes of the dppp, dppb and dppf analogues are rare and their synthesis from this path is a subject of our current investigation. These results suggest that a slight change in the skeletal property of the diphosphine would have a significant influence on the dinuclear framework. This intricate influence is manifested in diVerent bonding modes of the phosphines, which result in diVerent structures.For a diphosphine that favours a chelating mode the dinuclear unit can break down to a mononuclear complex, as witnessed in 6. Attempts to synthesize the dppe analogue of 5 by oxidative addition have so far been unsuccessful. This is reminiscent of the sensitivity of the dinuclear structure to its diphosphine skeletal residue. In the phosphine exchange reactions, dppm and dppp react smoothly with 1 to give 2 and 3 respectively. However, dppb, dppf and dppe, whose co-ordinating ability is not known to be weaker, do not show Scheme 23780 J.Chem. Soc., Dalton Trans., 1998, 3777–3784 any activity. These diVerences emphasise the stability bridging of dppm (e.g. in A-frame type complexes) and possible isolation of the unidentate mode of dppp due to its optimum chain length. The similar ease of replacement of PPh3 or Br2 in 1 makes structural prediction of any ligand-exchanged products diYcult. Among all the complexes isolated, only 6 is mononuclear.Preliminary results indicate that metathesis (and deprotonation) of 1 with Li1(C4H3S)2 (C4H3S = 2-thienyl) gives trans- (N,P)-[Pd2(h1-C4H3S)2(PPh3)2(m-C5H4N-C2,N)2] which undergoes thermolysis in toluene to give 2-thienylpyridine.19 This suggests that at least some of the many known Pd-catalysed hetero-coupling reactions do not proceed through mononuclear intermediates, especially when the phosphines used are bidentate, and that the dinuclear complexes isolated in this work are catalytically significant.Electrospray mass spectrometry The emerging importance of ESMS as a soft ionisation mass spectrometry for organometallic complexes is evident.20 In this work it is used to investigate the dinuclear-to-mononuclear breakdown of the Pd2 species and unveil the various fragmentation possibilities of a dinuclear core. The information thus obtained would help to understand the Pd-catalysed coupling mechanisms and design of new catalysts using diphosphines as ligand support.Similar use of ESMS to observe catalytic intermediates in the Pd0-catalysed Suzuki coupling has been reported.21 At a very low cone voltage of 5 V, both monocationic trans-(N,P)-[Pd2Br(MeCN)(PPh3)2(m-C5H4N-C2,N)2]1 1a (m/z=1013.9) and dicationic species trans-(N,P)-[Pd2- (MeCN)2(PPh3)2(m-C5H4N-C2,N)2]21 1b, (m/z = 487.9), from MeCN displacement of Br in 1, are observed with the latter as the dominant species (Fig. 3). The displacement of halide and subsequent co-ordination of MeCN to the parent fragment is a common phenomenon in ESMS.22 The BF4 2 salt of 1b was independently synthesized and its ESMS spectra analysed.They generally agree with those obtained from 1 and support that 1b is an essential fragmentation product of 1. At 20 V ligand fragmentation begins to occur, giving rise to [Pd2- (PPh3)2(m-C5H4N-C2,N)2]21 1c (m/z = 446.3) and [Pd2(MeCN)- (PPh3)2(m-C5H4N-C2,N)2]21 1d (m/z = 467.4). The preferred dissociation of Br2 and MeCN suggests that the bond strength of Pd–Y decreases in the order Y = PPh3>MeCN > Br2.At 30 V the amount of 1c and 1d increases and 1c becomes the dominant species. A species 1e (m/z = 340.4) is formed whose isotope pattern suggests a phosphonium cation [Ph3PC5H4N]1, formed by dissociative coupling between the phosphine and pyridyl ligands. At 40 V fragmentation to monopalladium species occurs. The peak of 1b is overshadowed by the monomeric ion [Pd(MeCN)(C5H4N)(PPh3)]1 1f, having the same m/z value but peaks separated by 1 mass unit, and 1c by peaks due to [Pd(C5H4N)(PPh3)]1 1g.The pyridyl ligand remains intact even at 40 V, which supports C-bonded 13,14b,23 rather than the generally weaker N-bonded pyridine. Another new species, [Pd2Br(C5H4N)2(PPh3)2]1 1h, at m/z = 972.4 also surfaces at this high voltage. These results are consistent with the general observation that increasing the cone voltage results in ions of lower charge.These experiments provide unequivocal support that the dinuclear structure is maintained in solution. Dissociation of Br2 and MeCN occurs prior to the monomerisation through cleavage of the bridging pyridyl groups. Bromide dissociation is most facile and is probably assisted by the MeCN present. Similar results have been observed for other metal phosphine–halide complexes.24 To substantiate the latter point, the ESMS spectra of 1 were recorded with the sample solution doped with a few drops of pyridine.As expected, species arising from bromide/acetonitrile replacement by pyridine, viz. trans-(N,P)-[Pd2Br(h1-C5H5N-N)(PPh3)2(m-C5H4N-C2, N)2]1 1i (m/z = 1051.5) {and its MeCN substitution complex, trans-(N,P)-[Pd2(MeCN)(h1-C5H5N-N)(PPh3)2(m-C5H4N-C2, N)2]21 1k (m/z=507.0)} and trans-(N,P)-[Pd2(h1-C5H5NN) 2(PPh3)2(m-C5H4N-C2,N)2]21 1j (m/z = 525.5) were detected (Scheme 3). These results suggest a rich ligand replacement chemistry in 1 which can be developed in line with the catalytic initiatives. The parent ion of complex 2 (m/z = 569) was detected as a Fig. 3 Positive-ion electrospray mass spectra of complex 1 in MeCN– water solvent (1 : 1 v/v) at cone voltages of (a) 5, (b) 20, (c) 30 and (d) 40 V. The spectra show the formation of the bis(acetonitrile) dication 1b at low cone voltages, which loses MeCN ligands (giving 1d and 1c) and fragments to the monomeric species 1f and 1g as the cone voltage is raised. Scheme 3J.Chem. Soc., Dalton Trans., 1998, 3777–3784 3781 minor species at low cone voltage (5 V). The occurrence of two major peaks (m/z = 609.1 and 1217.7) merits special comment. The former peak is attributed to [Pd2H(Br)(C5H4N)2(dppm)2]21 2a, which is formed from protonation of 2, while the latter is most likely an ion pair of 2 1 Br2. The parent ion of 3 (m/z = 803.4) was detected at 5 V. The presence of three diVerent types of ligands, viz. bridging pyridyl, bridging phosphine and unidentate phosphine, creates a number of possibilities for ligand transformation and molecular fragmentation.A large number of molecular peaks observed at such a low cone voltage supports this projection. Similar to 2, complex 3 readily adds H1Br2 to give [Pd2H(Br)(C5H4N)2(dppp)3]21 3a (m/z = 843.3), [Pd2H2Br2(C5H4N)2(dppp)3]21 3b (m/z=884.5), [PdH(Br)- (C5H4N)(dppp)]1 3c (m/z=677.5) and [PdH(Br)(C5H4N)- (dppp)2]21 3d (m/z = 1089.9). Chelation of one of the dangling phosphines in 3 could lead to an immediate rupture of the phosphine bridge which triggers the splitting of 3 to two mononuclear fragments: [Pd(C5H4N)(dppp)2]21 3f (m/z = 1007.8) and [Pd(C5H4N)(dppp)]21 3g (m/z = 595.6).In the presence of trace H1Br2, and especially at higher voltages (e.g. 40 V), the amount of 3d, 3f and 3g would decrease while that of 3c is observed to increase relatively. Species 3c is thus a possible sink in a series of acid-promoted fragmentations. As it is structurally identical to 6, detection of 3c together with its precursors under ESMS conditions provides a mechanistic mapping for the formation of 6 under synthetic conditions.Ligand displacement of dppp in 3 by Br2 giving [Pd2Br(C5H4N)2(dppp)2]1 3e (m/z = 1433.8) is also evident. At 5 V, H1Br2 elimination from complex 6 generates [Pd- (C5H4N)(dppf)]1 6a (m/z = 737.4); dimeric species corresponding to formulae [Pd2HBr2(C5H4N)2(dppf)2]1 6b (m/z = 1637.7) and [Pd2H2Br3(C5H4N)2(dppf)2]1 6c (m/z = 1718.0) are also observed.The protonation site and hence the structures of these species are uncertain. At 20 V a fresh species [Pd(C5H4- NH)(dppf)]21 6d (m/z = 369.5) is detected, which can either be formed from 6 through a bromide dissociation or 6a through protonation. As the cone voltage is increased to 80 V, 6d disappears and 6a emerges as the dominant species. Conclusion Our work demonstrated that the structures of the palladium(II) pyridyl complexes are highly sensitive to the supporting phosphines.Although their structures are not easily predictable, they usually represent a compromise of conflicting demands of the diphosphine and pyridyl ligands. In some cases the resultant complexes arise from a series of molecular rearrangements whose mechanisms are complex and unknown. Based on the products formed under ESMS conditions, one can gain some valuable insight on the ligand transformation and coordination mode changes during the molecular rearrangement process.Although the ESMS data alone cannot provide definitive structural proof of various transient species detected, a meaningful discussion on the structural changes can be presented when supported by synthetic and crystallographic data. Collectively, this information provides a platform for us to understand the formation mechanism of the complexes concerned. Experimental General procedures All reactions involving light-sensitive compounds were performed shielded from light.Unless stated otherwise, all reactions were performed under a positive pressure of purified argon. All 1H NMR spectra were recorded at 300 MHz at 25 8C on a Bruker ACF 300 spectrometer. Chemical shifts are reported in ppm to high frequency with Me4Si as internal standard for 1H NMR spectra. The 31P NMR spectra were recorded at 121.39 MHz with 85% H3PO4 as external reference. Elemental analyses were performed by the Microanalytical Laboratory staV of the Department of Chemistry in the National University of Singapore. The facile solvation of many dppf (other diphosphines) complexes precludes an accurate microanalysis of some complexes.25 All solvents were of analytical grade and freshly distilled before use.All chemical reagents were purchased commercially and used as received. The complexes [Pd(PPh3)4], [Pd2(dppm)3], [Pd(dppp)2], [Pd- (dppb)2] and [Pd(dppf)2] were prepared by the literature methods with slight modification.26 Electrospray mass spectrometry Electrospray mass spectra were recorded using a VG Platform II instrument, using MeCN–water (1 :1) or MeOH as the mobile phase solvent.Cone voltages were typically varied from 5 to 80 V in order to investigate the eVect of higher voltages on fragmentation pathways of the various compounds observed. The peaks in the ESMS are identified by the most intense m/z value within the isotopic mass distribution. Further details of the experimental set-up used are available elsewhere.27 Isotope patterns were recorded under high-resolution conditions for all major ions and compared with theoretical patterns obtained using the Isotope program.28 In all cases there was good agreement between the experimental and calculated isotopic mass distributions.Syntheses trans-(N,P)-[Pd2Br2(PPh3)2(Ï-C5H4N-C2,N)2] 1. Freshly prepared [Pd(PPh3)4] 26 [obtained in situ from PdCl2 (1.000 g, 5.64 mmol)] was dissolved in toluene and 2-bromopyridine (1.620 g, 10.25 mmol) added slowly.The resulting mixture was stirred for 2 h, resulting in the formation of a yellow-green suspension. The product was filtered oV and washed thoroughly with an excess of diethyl ether before recrystallisation from chloroform –hexane (2.91 g, 98%) (Found: C, 52.6; H, 3.8; Br, 13.5; N, 2.8; P, 5.8; Pd, 19.8. C23H19BrNPPd requires C, 52.5; H, 3.7; Br, 15.2; N, 2.7; P, 5.9; Pd, 20.2%). dP(CDCl3) 29.8 (s). [Pd2(Ï-C5H4N-C2,N)2(Ï-dppm)2]Br2 2. 2-Bromopyridine (0.724 g, 4.58 mmol) was added to a toluene solution of [Pd2(dppm)3] 26 [obtained in situ from PdCl2 (0.310 g, 1.75 mmol)]. The resulting mixture was refluxed for ca. 1 d after which light yellow crystals were formed. The mixture was filtered and the crystals washed with an excess of toluene and diethyl ether and recrystallised from methanol (0.96 g, 85%) (Found: C, 54.1; H, 4.1; Br, 15.5; N, 2.1; P, 9.3; Pd, 13.0. C30H26BrNP2Pd requires C, 55.5; H, 4.0; Br, 12.3; N, 2.2; P, 9.6; Pd, 16.4%).dC(CD3OD) 152.6 (C2). dP(CD3OD) 11.8 (t) and 3.5 (t) [J(P–P) = 30 Hz]. [Pd2(Á1-dppp)2(Ï-C5H4N-C2,N)2(Ï-dppp)]Br2 3. Freshly prepared [Pd(dppp)2] 26 [obtained in situ from PdCl2 (0.182 g, 1.02 mmol)] was dissolved in toluene and 2-bromopyridine (0.178 g, 1.13 mmol) added slowly. The resulting mixture was refluxed for 2 h, resulting in the formation of oV-white crystals. The crystals were filtered oV, washed with an excess of toluene and diethyl ether and recrystallised from methanol (0.90 g, 99%) (Found: C, 62.2; H, 5.2; Br, 9.5; N, 1.7; P, 11.3; Pd, 10.2.C91H86Br2N2P6Pd2 requires C, 61.9; H, 4.9; Br, 9.1; N, 1.6; P, 10.5; Pd, 12.1%). dC(CD3OD) 183.5 (C2). dP(CD3OD) 12.9 (dd) [2J(P–P) 330, 3J(P–P) 35], 3.4 (dd) [2J(P–P) 330, 3J(P–P) 54 Hz] and 24.5 (unresolved multiplets). trans-[{PdBr(Á1-C5H4N-C2)(Ï-dppb)}n]?C6H5Me 4. 2-Bromopyridine (0.604 g, 3.82 mmol) was added to a toluene solution of [Pd(dppb)2] 26 [obtained in situ from PdCl2 (0.184 g, 1.04 mmol)].The resulting mixture was stirred at r.t. until bright yellow crystals were precipitated. After filtration the crystals3782 J. Chem. Soc., Dalton Trans., 1998, 3777–3784 were washed with an excess of toluene and diethyl ether and recrystallised from methanol–diethyl ether (0.37 g, 75%) (Found: C, 56.5; H, 5.0; Br, 11.5; N, 1.9; P, 7.9; Pd, 14.8. C33H32BrNP2Pd requires C, 57.4; H, 4.7; Br, 11.6; N, 2.0; P, 9.0; Pd, 15.4%).dC(CDCl3) 131.7 (C2). dP(CDCl3) 32.8 (s). trans-(N,P)-[Pd2Br2(Ï-C5H4N-C2,N)2(Ï-dppb)] 5. 2-Bromopyridine (0.604 g, 3.82 mmol) was added to a toluene solution of [Pd(dppb)2] 26 [obtained in situ from PdCl2 (0.184 g, 1.04 mmol)]. The resulting mixture was refluxed until light yellow crystals were precipitated. After filtration the crystals were washed with an excess of toluene, methanol and diethyl ether. The product was recrystallised from dichloromethane–pentane (0.37 g, 75%) (Found: C, 47.1; H, 3.9; Br, 23.0; N, 3.3; P, 6.4; Pd, 18.7.C19H18BrNPPd requires C, 47.9; H, 3.8; Br, 16.7; N, 3.0; P, 6.5; Pd, 22.3%). dP(CDCl3) 25.2 (s). cis-[PdBr(Á1-C5H4NH-C2)(Á2-dppf)]Br?0.5H2O 6. 2-Bromopyridine (0.433 g, 2.74 mmol) was added to a toluene solution of [Pd(dppf)2] 26 [obtained in situ from PdCl2 (0.405 g, 2.28 mmol)]. The resulting mixture was refluxed until deep red crystals were precipitated from an indigo solution (about 2–3 d). After filtration, the crystals were washed with an excess of toluene and diethyl ether and recrystallised from methanol–diethyl ether (0.14 g, 14%) (Found: C, 51.5; H, 3.7; Br, 14.7; N, 1.6; P, 7.5; Fe, 6.6; Pd, 12.3.C39H33Br2FeNP2Pd?0.5H2O requires C, 51.5; H, 3.8; Br, 17.6; Fe, 6.2; N, 1.5; P, 6.8; Pd, 11.7%). dC(CDCl3) 140.1 (C2). dP(CDCl3) 29.4 (d) and 14.2 (d) [2J(P–P) 19 Hz]. trans-(N,P)-[Pd2Cl2(Ï-C9H6N-C2,N)2(Ï-dppb)] 7. 2-Chloroquinoline (0.210 g, 1.28 mmol) was added to a toluene solution of [Pd(dppb)2] 26 [obtained in situ from PdCl2 (0.227 g, 1.04 mmol)].The resulting mixture was stirred at r.t. for ca. 2 d until oV-white crystals were precipitated. After filtration the crystals were washed with an excess of toluene and diethyl ether (0.05 g, 17%) (Found: C, 57.6; H, 4.3; Cl, 6.4; N, 2.5; P, 4.9; Pd, 23.3. C23H20ClNPPd requires C, 57.2; H, 4.2; Cl, 7.3; N, 2.9; P, 6.4; Pd, 22.0%). dP(CDCl3) 25.7 (s). Complex 2 from phosphine exchange reaction between 1 and dppm in 1 : 1 ratio.Complex 1 (0.020 g, 1.73 × 1022 mmol) and dppm (0.007 g, 1.73 × 1022 mmol) were suspended in methanol and stirred at r.t. for 3 d. The resultant yellow solution was filtered, concentrated and diethyl ether was added for precipitation. The light yellow solid (complex 2) was collected by filtration, washed with an excess of diethyl ether and dried in vacuo (0.02 g, 62%) (Found: C, 53.8; H, 4.6; Br, 12.3; N, 2.1; P, 8.4; Pd, 15.4. C60H52Br2N2P4Pd2 requires C, 55.5; H, 4.0; Br, 12.3; N, 2.2; P, 9.6; Pd, 16.4%).dP(CD3OD) 11.8 (t) and 3.5 (t) [2J(P–P) 30 Hz]. Complex 3 from phosphine exchange reaction between 1 and an excess of dppp. Complex 1 (0.050 g, 4.33 × 1025 mmol) and an excess of dppp (0.098 g, 1.73 × 1024 mmol) were dissolved in toluene and stirred at r.t. for 2 d. The resultant oV-white suspension was filtered, and the residue (complex 3) was washed with an excess of toluene and diethyl ether (0.01 g, 85%) (Found: C, 61.8; H, 4.5; Br, 7.0; N, 1.3; P, 10.6; Pd, 12.6.C91H86Br2N2P6Pd2 requires C, 61.9; H, 4.9; Br, 9.1; N, 1.6; P, 10.5; Pd, 12.1%). dP(CD3OD) 12.9 (dd) [2J(P–P) 330, 3J(P–P) 35], 3.4 (dd) [2J(P–P) 330, 3J(P–P) 54 Hz] and 24.5 (unresolved multiplets). Reaction of [Pd2(dppm)3] with 2-chloroquinoline. 2-Chloroquinoline (0.198 g, 1.21 mmol) was added to a fresh toluene solution of [Pd2(dppm)3] 26 [obtained in situ from PdCl2 (0.212 g, 1.20 mmol)]. The resulting mixture was refluxed for ca. 2 d to give bright orange crystals of [Pd2Cl2(m-dppm)2]. The crystals were collected by filtration, washed with an excess of toluene and diethyl ether and recrystallised from methanol (0.30 g, 47 %) (Found: C, 57.9; H, 4.2; Cl, 6.7; P, 8.9; Pd, 15.1. C25H22ClP2Pd requires C, 57.1; H, 4.2; Cl, 6.7; P, 11.7; Pd, 20.2%). dP(CDCl3) 23.1 (s). trans-(N,P)-[Pd2(MeCN)2(PPh3)2(Ï-C5H4N-C2,N)2][BF4]2 1b. The salt AgBF4 (0.342 g, 1.76 mmol) was added to a fresh acetonitrile solution of complex 1 (0.853 g, 0.81 mmol).The resulting mixture was stirred at r.t. for ca. 2 h. The mixture was filtered and diethyl ether was added to the filtrate for precipitation (yield 0.76 g, 85%) (Found: C, 53.9; H, 4.1; B, 2.0; F, 13.8; N, 4.6; P, 5.0; Pd, 18.1. C25H22BF4N2PPd requires C, 54.2; H, 4.0; B, 1.9; F, 13.3; N, 5.1; P, 5.6; Pd, 19.2%). dP(CD3CN) 29.5 (s). X-Ray crystallography Single-crystal X-ray diVraction experiments were carried out on a Siemens SMART CCD diVractometer with a sealed tube at 23 8C and Mo-Ka radiation (l 0.71073 Å).Preliminary cell constants were obtained from 45 frames (width of 0.38 in W) data. Final cell parameters were obtained by global refinements of reflections obtained from integration of all the frame data. The data were collected with a frame width of 0.38 in W and a counting time of 20 s per frame at a crystal-to-detector distance of 5.027 cm. The software SMART29 was used for collecting frames of data, indexing reflections and determination of lattice parameters, SAINT17 for integration of intensity of reflections and scaling, SADABS30 for absorption correction and SHELXTL31 for space group and structure determination, refinements, graphics and structure reporting.Light yellow crystals of complex 5 were grown from a sample solution in acetonitrile and dichloromethane by slow evaporation at r.t. A suitable crystal of dimensions 0.35 × 0.25 × 0.2 mm was selected, wedged inside a glass capillary tube and flame-sealed.A total of 12 548 reflections were collected in the q range 1.74–29.048 (212 < h < 13, 216 < k < 17, 222 < l < 22). For Z = 2, the space group P1� was chosen in the triclinic system. All the non-hydrogen atoms in the neutral molecule were refined anisotropically. One disordered molecule of acetonitrile was located in the Fourier-diVerence synthesis (occupancy 0.5/ 0.5). Ideal bond distances and angles were imposed for these acetonitrile solvates and common isotropic thermal parameters were refined for each model.A riding model was used to place the hydrogen atoms in their idealised positions. In the final least-squares refinement cycles on F2 the model converged to R = 0.0394, R9 = 0.0956, and S = 1.042 for 7322 reflections with Fo > 4s(Fo) and 437 parameters, and R = 0.0529 and R9 = 0.1033 for all 8929 data. In the final Fourier-diVerence synthesis the electron density fluctuated in the range 0.77 to 20.84 e Å23.The top peak was associated with N(1s) at a distance of 0.48 Å. There was no shift in the final cycles. An extinction correction was refined to 0.0009(2). There is a non-crystallographic twofold symmetry in the molecule. However, MISSYM32 did not reveal any additional symmetry. Reddish brown rectangular crystals of complex 6 were grown by a layering method using methanol–diethyl ether. A suitable crystal of dimensions 0.50 × 0.23 × 0.13 mm was mounted at the end of a glass fibre.A total of 12 093 reflections were collected in the q range 2.07–29.348 (212 < h < 13, 210 < k < 14, 225 < l < 25). All the non-hydrogen atoms were refined anisotropically. One disordered molecule of water was located in the Fourier-diVerence synthesis (occupancy 0.25/0.25). A riding model was used to place the hydrogen atoms in their idealised positions except for the water molecule. In the final leastsquares refinement cycles on F2 the model converged to R = 0.0395, R9 = 0.1060, and S = 1.036 for 7000 reflections with Fo > 4s(Fo) and 434 parameters, and R = 0.0510 and R9 = 0.1126 for all 8582 data.In the final Fourier-diVerence synthesisJ. Chem. Soc., Dalton Trans., 1998, 3777–3784 3783 Table 2 Crystallographic data for trans-(N,P)-[Pd2Br2(m-C5H4N-C2,N)2(m-dppb)]?MeCN 5 and cis-[PdBr(h1-C5H4NH-C2)(h2-dppf)]Br? 0.5MeOH 6 Chemical formula M Colour and habit Crystal system Space group a/Å b/Å c/Å a/8 b/8 g/8 U/Å3 ZF (000) Dc/g cm23 Rint m/mm21 Reflections collected Independent reflections Observed reflections, n Residual electron densities/e Å23 RR 9 Goodness of fit, S 5?MeCN C38H36Br2N2P2Pd2?MeCN 996.30 Light yellow block Triclinic P1� 10.4392(2) 12.6975(2) 16.8905(3) 70.198(1) 75.766(1) 73.886(1) 1994.96(6) 2 984 1.659 0.0201 3.016 12548 8929 7322 10.770 to 20.844 0.0394 0.0956 1.042 6?0.5MeOH C39H33Br2FeNP2Pd?0.5MeOH 908.68 Brownish red block Triclinic P1� 9.7095(1) 10.6396(1) 18.9842(1) 89.378(1) 85.003(1) 67.925(1) 1809.98(3) 2 902 1.667 0.0256 3.224 12093 8582 7000 11.092 to 21.151 0.0395 0.1060 1.036 the electron density fluctuated in the range 1.092 to 21.151 e Å23. The top four peaks were associated with Pd(1).There was no shift in the final cycles. Although we were not able to locate the pyridinium proton directly, all the data are more consistent with a C- rather than an N-bonded model. The former gives a lower R and R9 factors.The diVerence in the isotopic thermal parameters between the atoms in question is less in the C-bonded model. The residual electron density in the N-bonded case also vanishes in the C-bonded model. Further crystallographic details are given in Table 2. CCDC reference number 186/1171. Acknowledgements The authors acknowledge the National University of Singapore (NUS) (Grant RP950695) for financial support and the technical staV for supporting services. We thank Y.Xie for some preliminary work and Y. P. Leong for assistance in the preparation of this manuscript. C. C. H. C. acknowledges National University of Singapore for a scholarship award. W. H. thanks the University of Waikato and the New Zealand Lottery Grants Board for financial assistance, and W. Jackson for technical support. References 1 J. H. Gladstone and W. Hibbert, Philos. Mag., 1989, 28, 38; M. Polanyi, Naturwissenschaften, 1922, 10, 411; R. Staudinger, H. Johner, M. Luthy, W.Kern, D. Russidis and O. Schweitzer, Liebigs Ann. Chem., 1929, 474, 145; H. W. Carothers, Chem. Rev., 1931, 8, 353; J. R. Bates and H. S. Taylor, J. Am. Chem. Soc., 1927, 49, 2438; K. Ziegler, L. Jakob, H. Wollthan and A. Wenz, Liebigs Ann. Chem., 1934, 511, 13; A. G. Evans and M. Polanyi, J. Chem. Soc., 1947, 252; P. H. Plesch, Nature (London), 1952, 169, 828; P. H. Plesch, J. Chem. Soc., 1953, 1659; M. Doi and S. F. Edwards, in The Theory of Polymer Dynamics, Clarendon Press, Oxford, 1986, ch. 1. 2 J.-M. Lehn, in Frontiers in Supramolecular Organic Chemistry and Photochemistry, eds. H.-J. Schneider and H. Dürr, VCH, Weinheim, 1991, ch. 1, pp. 4–8; see also: J.-M. Lehn, in Nobel Lectures in Chemistry (1981–1990), eds. T. Frängsmyr and B. G. Malmström, World Scientific, Singapore, 1992, pp. 444–491. 3 F. Vögtle, Supramolecular Chemistry, Wiley, Chichester, 1991, ch. 10, pp. 291–310. 4 J. Roncali, Chem. Rev., 1992, 92 5 F. Garnier, G. Tourillon, M.Garzrd and J. C. Dubois, J. Electroanal. Chem. Interfacial Electrochem., 1983, 148, 301. 6 F. Garnier, Angew. Chem., Int. Ed. Engl., 1989, 28, 513. 7 T. KauVman, E. Wienhöfer and A. Woltermann, Angew. Chem., Int. Ed. Engl., 1971, 10, 741. 8 M. Ishikura, M. Kamada and M. Terashima, Synthesis, 1984, 11, 96; K. Kahmann, H. Sigel and H. Erlenmeyer, Helv. Chim. Acta, 1964, 47, 1754; H. Wynberg, T. J. van Bergen and R. M. Kellogg, J. Org. Chem., 1969, 34, 3175. 9 S. Gronowitz, J.Malm and A.-B. Hörnfeldt, J. Organomet. Chem., 1993, 460, 127. 10 A. Suzuki, Acc. Chem. Res., 1982, 15, 178; N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457. 11 K. Isobe, E. Kai, Y. Nakamura, K. Nichisuke, T. Miwa, S. Kawaguchi, K. Kinoshita and K. Nakatsu, J. Am. Chem. Soc., 1980, 102, 2475. 12 K. Nakatsu, K. Kinoshita, H. Kanda, K. Isobe, Y. Nakamura, S. Kawaguchi and U. Nishinomiya, Chem. Lett., 1980, 913. 13 K. Isobe and S. Kawaguchi, Heterocycles, 1981, 16, 1603. 14 (a) K. Nakatsu, K. Kafuku and H. Yamaoka, Inorg. Chim. Acta, 1981, 54, L69; H. Tanaka, K. Isobe, S. Kawaguchi and S. Okeya, Bull. Chem. Soc. Jpn., 1984, 57, 456; K. Isobe, K. Nanjo, Y. Nakamura and S. Kawaguchi, Bull. Chem. Soc. Jpn., 1986, 59, 2141; K. Isobe, Y. Nakamura, T. Miwa and S. Kawaguchi, Bull. Chem. Soc. Jpn., 1987, 60, 149; A. Mantovani, J. Organomet. Chem., 1983, 255, 385; (b) K. Isobe, Y. Nakamura and S. Kawaguchi, Bull. Chem. Soc. Jpn., 1989, 62, 1802; (c) L. Canovese, F.Visentin, P. Uguagliati, F. Dibianca, A. Fontana and B. Crociani, J. Organomet. Chem., 1996, 525, 43. 15 K.-S. Gan and T. S. A. Hor, in Ferrocenes – Homogeneous Catalysis, Organic Synthesis, Materials Science, eds. A. Togni and T. Hayashi, VCH, Weinheim, 1995, ch. 1, pp. 3–104, and refs. therein. 16 H. Trauner, P. Lefloch, J. M. Lefour, L. Ricard and F. Mathey, Synthesis-Stuttgart, 1995, 717. 17 W. R. Cullen, T.-J. Kim, F. W. B. Einstein and T. Jones, Organometallics, 1985, 4, 346; U.Casellato, B. Corain, R. Graziani, B. Longato and G. Pilloni, Inorg. Chem., 1990, 29, 1193; I. R. Butler, W. R. Cullen, S. J. Rettig and J. Trotter, Organometallics, 1985, 4, 972; T. Hayashi, M. Konishi, Y. Kobori, M. Kumada, T. Higuchi and K. Hirotsu, J. Am. Chem. Soc., 1984, 106, 158; L.-T. Phang, T. S. A. Hor and T. C. W. Mak, unpublished work; C. E. Housecroft, S. M. Owen, P. R. Raithby and B. A. M. Shaykh, Organometallics, 1990, 9, 1617; D. A. Clemente, G.Pilloni, B. Corain, B. Longato and M. Tiripicchio-Camellini, Inorg. Chim. Acta, 1986, 115, L9; P. F. Kelly, A. M. Z. Slawin, D. J. Williams and J. D. Woollins, Polyhedron, 1988, 7, 1925; A. L. Bandini, G. Banditelli, M. A. Cinellu, G. Sanna, G. Minghetti, F. Demartin and M. Manassero, Inorg. Chem., 1989, 28, 404; B. Longato, G. Pilloni, G. Valle and B. Corain, Inorg. Chem., 1988, 27, 956; M. Zhou, Y. Xu, A. M. Tan, P.-H. Leung, K. F. Mok, L.-L. Koh and T. S. A. Hor, Inorg.Chem., 1995, 34, 6425; B. S. Haggerty, C. E. Housecroft, A. L. Rheingold and B. A. M. Shaykh, J. Chem. Soc., Dalton Trans., 1991, 2175; G. Bandoli, G. Trovò, A. Dolmella and B. Longato, Inorg. Chem., 1992, 31, 45.3784 J. Chem. Soc., Dalton Trans., 1998, 3777–3784 18 P. Kalck, C. Randrianalimanana, M. Ridmy, A. Thorez, H. T. Dieck and J. Ehlers, New. J. Chem., 1988, 12, 679; L.-T. Phang, S. C. F. Au-Yeung, T. S. A. Hor, S. B. Khoo, Z.-Y. Zhou and T. C. W. Mak, J. Chem.Soc., Dalton Trans., 1993, 165. 19 Z. H. Loh, C. C. H. Chin, Y. Xie and T. S. A. Hor, unpublished work. 20 R. Colton, A. D. Agustine and J. C. Traeger, Mass Spectrum Rev., 1995, 14, 79; M. T. Caudle, R. D. Stevens and A. L. Crumbliss, Inorg. Chem., 1994, 33, 6111; P. Falaras, C.-A. Mitsopoulou, D. Argyropoulos, E. Lyris, N. Psaroudakis, E. Vrachnou and D. Katakis, Inorg. Chem., 1995, 34, 4536; S. Blanc, P. Yakirevitch, E. Leize, M. Meyer, J. Libman, A. Van Dorsselaer, A.-M. Albrecht- Gary and A.Shanzer, J. Am. Chem. Soc., 1997, 119, 4934; P. B. O’Connor and F. W. McLaVerty, J. Am. Chem. Soc., 1995, 117, 12826; A. J. Canty, R. Colton, A. D’Agostino and J. C. Traeger, Inorg. Chim. Acta, 1994, 223, 103; M. Ferrer, R. Reina, O. Rossell, M. Seco and G. Segales, J. Organomet. Chem., 1996, 515, 205; H. Hori, J. Ishihara, K. Koike, K. Takeuchi, T. Ibusuki and O. Ishitani, Chem. Lett., 1997, 273; R. Colton, V. Tedesco and J. C. Traeger, Inorg. Chem., 1992, 31, 1865. 21 A. O. Aliprantis and J. W. Canary, J. Am. Chem. Soc., 1994, 116, 6985. 22 K. J. Fisher, W. Henderson, I. G. Dance and G. D. Willett, J. Chem. Soc., Dalton Trans., 1996, 4109; W. Henderson and B. K. Nicholson, J. Chem. Soc., Chem. Commun., 1995, 2531; W. Henderson and G. M. Olsen, Polyhedron, 1996, 15, 2105; W. Henderson, J. S. McIndoe, B. K. Nicholson and P. J. Dyson, J. Chem. Soc., Dalton Trans., 1998, 4109; C. E. C. A. Hop, J. T. Brady and R. Bakhtiar, J. Am. Soc. Mass Spectrom., 1997, 8, 191; M. I. Bruce, M. Ke and P. J. Low, Chem. Commun., 1996, 2405; M. T. Caudle, R. D. Stevens and A. L. Crumbliss, Inorg. Chem., 1994, 33, 843. 23 A. Mantovani and B. Crociani, J. Organomet. Chem., 1982, 236, C37. 24 J. Fawcett, W. Henderson, R. D. W. Kemmitt, D. R. Russell and A. Upreti, J. Chem. Soc., Dalton Trans., 1996, 1897; H. C. Ehrsson, I. B. Wallin, A. S. Andersson and P. O. Edlund, Anal. Chem., 1995, 67, 3608; R. Colton and D. Dakternieks, Inorg. Chim. Acta, 1993, 208, 173. 25 T. S. A. Hor and L.-T. Phang, J. Organomet. Chem., 1990, 381, 121; T. S. A. Hor and L.-T. Phang, J. Organomet. Chem., 1990, 380, 345; T. S. A. Hor, L.-T. Phang, L.-K. Liu and Y.-S. Wen, J. Organomet. Chem., 1990, 397, 29; T. S. A. Hor, H. S. O. Chan, K.-L. Tan, L.-T. Phang, Y. K. Yan, L.-K. Liu and Y.-S. Wen, Polyhedron, 1991, 10, 2437; L.-T. Phang, K.-S. Gan, H. K. Lee and T. S. A. Hor, J. Chem. Soc., Dalton Trans., 1993, 2697; S. P. Neo, Z.-Y. Zhou, T. C. W. Mak and T. S. A. Hor, J. Chem. Soc., Dalton Trans., 1994, 3451; Z.-G. Fang, Y.-S. Wen, R. K. L. Wong, S.-C. Ng, L.-K. Liu and T. S. A. Hor, J. Cluster Sci., 1994, 5, 327; G. Li, S. Li., A. L. Tan, W.-H. Yip, T. C. W. Mak and T. S. A. Hor, J. Chem. Soc., Dalton Trans., 1996, 4315. 26 D. R. Coulson, Inorg. Synth., 1990, 28, 108. 27 W. Henderson, J. Fawcett, R. D. W. Kemmitt, P. McKenna and D. R. Russell, Polyhedron, 1997, 16, 2455. 28 L. J. Arnold, J. Chem. Educ., 1992, 69, 811. 29 SMART & SAINT Software Reference Manuals, version 4.0, Siemens Energy & Automation, Inc., Analytical Instrumentation, Madison, WI, 1996. 30 G. M. Sheldrick, SADABS, a software for empirical absorption correction, University of Göttingen, 1993. 31 SHELXTL Reference Manual, version 5.03, Siemens Energy Automation, Inc., Analytical Instrumentation, Madison, WI, 1996. 32 Y. LePage, J. Appl. Crystallogr., 1987, 20, 264. Paper 8/03779E

ISSN:1477-9226    

DOI:10.1039/a803779e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3785–3789 3785 Enantiopure planar chiral monomers and di-Ï-bromo-bridged dimers of ferrocenylhydrazones from asymmetric cyclopalladation Gang Zhao, Qi-Guang Wang and Thomas C. W. Mak* Department of Chemistry, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong. E-mail: tcwmak@cuhk.edu.hk Received 3rd August 1998, Accepted 26th August 1998 Planar chiral cyclopalladated ferrocenylhydrozones [PdBr{(h5-C5H3CH]] NN(CH2)3CHCH2OCH3)Fe(h5-C5H5)}] (2)-(Sp,R) and (1)-(Rp,S) and di-m-bromo-bridged dimers [{Pd[(h5-C5H3C(CH3)]] NN(CH2)3CHCH2OCH3)- Fe(h5-C5H5)](m-Br)}2] syn-(2)-(Sp,R,R,Sp), syn-(1)-(Rp,R,R,Rp), syn-(1)-(Rp,S,S,Rp) and syn-(2)-(Sp,S,S,Sp) have been conveniently synthesized in enantiomerically pure form with a high level of diastereoselectivity and their absolute configurations elucidated by single-crystal X-ray analysis.Owing to the well known reactivity of the s(Pd–C) bond,1 planar chiral cyclopalladated derivatives of ferrocene have found useful applications in organic synthesis, resolution of enantiomers, catalysis, photochemistry, bioinorganic chemistry, and the study of ordered mesophases.Although many cyclopalladated derivatives of ferrocene are known,2 most of them have been obtained as racemic mixtures due to diYculties in isolating these compounds in enantiomerically pure form,3 and accordingly their stereochemistries remain unsubstantiated.Several papers 2d,4,5 have reported the synthesis and crystal structures of planar chiral cyclopalladated monomers of ferrocene. However, anion-bridged dimers of such systems exhibit more complex structures and interesting aspects of stereoisomerism. 5 There are six possible diastereomeric di-m-chlorobridged dimers of a planar chiral cyclopalladated ferrocenylimine, namely syn-(Rp,Rp), syn-(Sp,Sp), anti-(Rp,Sp), syn-(Rp, Sp), anti-(Sp,Sp) and anti-(Rp,Rp) that diVer in regard to planar chiralities, cis/trans relationship of the pair of co-ordinating nitrogen atoms, or syn/anti arrangement of the ferrocenyl moieties.6 Three of these, syn-(Rp,Rp), syn-(Sp,Sp) and anti- (Rp,Sp), were reported in our study,6 whereas an analog of the anti-(Rp,Rp) isomer is also known.7 Development of a convenient and eYcient synthetic procedure for enantiopure ferrocene compounds of this type is important in the advancement of cyclopalladation chemistry.Previous attempts to carry out cyclopalladation of ferrocenylhydrazones have not been successful.8 Here we present a simple and highly stereoselective synthesis of enantiopure bromide-ligated monomeric and di-m-bromo-bridged dimeric planar chiral cyclopalladated derivatives of ferrocenylhydrazones.Their absolute configurations and structural relationship were also elucidated. In comparison with cyclopalladation employing Na2PdCl4 and NaO2CMe?3H2O in our previous studies on the analogous di-m-chloro-bridged dimers,5,6 the present synthetic procedure using Pd(O2CMe)2 and NaO2CMe? 3H2O in MeOH (non-acidic medium) at room temperature results in a superior yield and significantly higher diastereoselectivity.To investigate whether the presence of a chiral substituent in the cyclopentadienyl ring would preferentially induce the activation of one of the two ortho s(C–H) bonds of the ferrocene moiety, we employed the pure chiral ferrocenylhydrazones (2)-(R)-2, (1)-(S)-2, (2)-(R)-3 and (1)-(S)-3 from the condensation reaction of ferrocenecarbaldehyde and acetylferrocene, respectively, with (1)-(R)- or (2)-(S)-1-amino-2- (methoxymethyl)pyrrolidine [(1)-(R)-1 or (2)-(S)-1], which serves as a chirality marker and facilitates product isolation (Scheme 1).Results and discussion Synthesis of planar chiral cyclopalladated derivatives of ferrocene Asymmetric cyclopalladation of the ferrocenylimines (2)-(R)-2 and (1)-(S)-2 with Pd(O2CMe)2 and NaO2CMe?3H2O in MeOH, followed by treatment with LiBr, gave the corresponding planar chiral cyclopalladated monomers [PdBr{(h5-C5H3CH]] NN(CH2)3CHCH2OCH3)Fe(h5-C5H5)}] (2)-(Sp,R)-4, [a]D 20 2680.7, and (1)-(Rp,S)-4, [a]D 20 1681.9 deg cm3 g21 dm21, respectively, in ca. 63% yield with a high level of diastereoselectivity. No evidence of the formation of any other isomer was detected by 1H NMR (300 MHz). The enantiopure compounds were isolated from the two reaction mixtures through column-layer chromatography.Asymmetric cyclopalladation of the ferrocenylimines (2)-(R)-3 and (1)-(S)-3 in the same manner gave di-mbromo- bridged planar chiral cyclopalladated dimers [{Pd[(h5-C5H3C(CH3)]] NN(CH2)3CHCH2OCH3)(Fe(h5-C5H5)]- (m-Br)}2] syn-(2)-(Sp,R,R,Sp)-5, [a]D 20 23506.8, syn-(1)- (Rp,R,R,Rp)-5, [a]D 20 12352.9, and syn-(1)-(Rp,S,S,Rp)-5, [a]D 20 13510.7, syn-(2)-(Sp,S,S,Sp)-5, [a]D 20 22359.3 deg cm3 g21 dm21, respectively, in ca. 70% yield with a high level of diastereoselectivity [product ratio syn-(2)-(Sp,R,R,Sp)-5 : syn-(1)- (Rp,R,R,Rp)-5 ª syn-(1)-(Rp,S,S,Rp)-5 : syn-(2)-(Sp,S,S,Sp)-5 ª 9:1].The results show that ligands (2)-(R)-2 and (1)-(S)-2 lead to higher stereoselectivity than do (2)-(R)-3 and (1)-(S)-3. Notably, cyclopalladation of (2)-(R)-2 and (1)-(S)-2 gives monomers in which the ether oxygen atom co-ordinates to palladium to form a six membered palladacycle, whereas cyclopalladation of (2)-(R)-3 and (1)-(S)-3 gives bromobridged dimers.This diVerent behavior can be explained by the model shown in Fig. 1, assuming prior co-ordination by the methoxymethyl group, a process often encountered in asymmetric synthesis.9 There is a repulsive interaction between hydrogen atoms at methyl C(12) and ethylene C(16) in the transition state of (2)-(R)-3 and (1)-(S)-3, thus favouring the formation of dimeric products in their cyclopalladation reaction. Cyclopalladation is usually carried out in acetic acid, one of the standard solvents.However, a complex reaction mixture is obtained when cyclopalladation of complexes (1)-(S)-2 and3786 J. Chem. Soc., Dalton Trans., 1998, 3785–3789 Scheme 1 Synthesis of planar chiral cyclopalladated ferrocenes; (i) Pd(O2CMe)2, NaO2CMe, MeOH, room temperature (r.t.), 24 h; (ii) LiBr, MeOH, r.t., 15 min. Pd N N O CpFe Br Br Pd N N O CpFe Pd N N O CpFe Br Br Pd N N O CpFe syn -(–)-( Sp, R, R, Sp)-5 syn -(+)-( Rp, R, R, Rp)-5 9 : 1 (–)-( R)-3 i, ii + Pd N N O CpFe Br Br Pd N N O CpFe syn -(+)-( Rp, R, R, Rp)-5 Pd N N O CpFe Br Br Pd N N O CpFe syn -(–)-( Sp, S, S, Sp)-5 (–)-( S)-3 i, ii (+)-( S)-2 i, ii Pd N FeCp N O H Br Pd N FeCp N O H (+)-( Rp, S)-4 (–)-( Sp, R)-4 (–)-( R)-2 N N C R Fe O R = H, (+)-( S)-2 R = CH3, (+)-( S)-3 N N C R Fe O R = H, (–)-( R)-2 R = CH3, (–)-( R)-3 N O NH2 O C R Fe (+)-( S)-1 N O NH2 (+)-( R)-1 + 9 : 1 i, ii Fig. 1 Idealized model of asymmetric cyclopalladation promoted by the methoxymethyl group and destabilization of monomeric cyclopalladated products derived from complexes (2)-(R)-3 and (1)-(S)-3.(1)-(S)-3 is conducted in this solvent due to the acid sensitivity of imine, which causes diYculty in purification. We found that cyclopalladation of this kind of imine with Pd(O2CMe)2 and NaO2CMe?3H2O in MeOH (non-acidic medium) at room temperature gives planar chiral compounds of ferrocene in good yield with a high level of stereoselectivity. Work-up of the reac-J.Chem. Soc., Dalton Trans., 1998, 3785–3789 3787 tion mixture is generally more eVective and convenient. In comparison with cyclopalladation using Na2PdCl4 and NaO2CMe?3H2O,5,6 the present reaction condition leads to significantly higher diastereoselectivity. Moreover, the new procedure can readily be applied to the preparation of related planar chiral anion-bridged dimers such as the chloro-, iodo-, acetato- and nitrito-bridged derivatives.10 Characterization Chemical shifts (1H NMR spectra in CDCl3) for substituted cyclopentadienyl protons of enantiomeric complexes (2)- (Sp,R)-4 and (1)-(Rp,S)-4 are d 5.07, 4.32 and 4.22.Moreover, the signals due to unsubstituted cyclopentadienyl protons of (2)-(Sp,R)-4 and (1)-(Rp,S)-4 shifted downfield by 0.34 ppm at d 4.43. Chemical shifts for substituted cyclopentadienyl protons of enantiomeric syn-(2)-(Sp,R,R,Sp)-5 and syn-(1)-(Rp,S, S,Rp)-5 are d 4.83, 4.34 and 4.24, respectively, but those of syn- (1)-(Rp,R,R,Rp)-5 and syn-(2)-(Sp,S,S,Sp)-5 are d 4.87, 4.34 and 4.25, respectively. Moreover, the signals due to unsubstituted cyclopentadienyl protons of syn-(2)-(Sp,R,R,Sp)-5 and syn-(1)-(Rp,S,S,Rp)-5 shifted downfield by 0.26 ppm at d 4.34 and those in syn-(1)-(Rp,R,R,Rp)-5 and syn-(2)- (Sp,S,S,Sp)-5 shifted downfield by 0.15 ppm at d 4.13.In all spectra of cyclopalladated compounds, H3 and H5 shifted upfield and H4 downfield for syn-(1)-(Rp,S,S,Rp)-5 and syn- (2)-(Sp,S,S,Sp)-5.The 13C-{1H} NMR spectra of ‘free’ ligands and their cyclopalladated complexes diVer in regard to the splitting of the resonance due to the C3, C4 pair of carbon atoms since the formation of the metallacycle causes a decrease in the symmetry of the substituted cyclopentadienyl ring. Crystal structure of compounds (2)-(Sp,R)-4 and (1)-(Rp,S)-4 X-Ray analysis confirmed that the desired enantiomeric compounds had indeed been obtained and established the absolute configuration of the planar chiralities of the ferrocenyl moieties (Fig. 2). The six-membered ring in each enantiomer is in a distorted chair form, and all of the substituents are equatorial. The palladium atom in the metallacycle is bound to a bromide and an oxygen atom, with the latter cis to the imino nitrogen atom which is unusual for palladocyclic compounds, thus leading to a slightly distorted square-planar co-ordination environment around it. The deviations (in Å) from the least-squares plane defined by the atoms Pd(1), Br(1), O(1), N(1) and C(1) are 0.038, 0.002, 20.022, 0.006 and 20.025 for (2)-(Sp,R)-4 and 20.038, 20.004, 0.023, 20.009 and 0.028 for (1)-(Rp,S)-4.The cyclopentadienyl rings are each planar and nearly parallel to each other [tilt angle: 3.88 for (2)-(Sp,R)-4 and 4.18 for (1)-(Rp,S)-4], and the two rings involved in the bicyclic system Fig. 2 Molecular structure (30% thermal ellipsoids) and absolute configuration of complex (1)-(Rp,S)-4 with atom-numbering scheme.formed by fusion of the palladocycle with the ferrocenyl C5H3 moiety are approximately coplanar, the relevant dihedral angle being 3.08 for (2)-(Sp,R)-4 and 3.38 for (1)-(Rp,S)-4. Crystal structures of compounds syn-(1)-(Rp,R,R,Rp)-5?2C6H6 and syn-(2)-(Sp,S,S,Sp)-5?2C6H6 X-Ray analysis has established the absolute planar chiralities of the ferrocenyl moieties based on the (1)-(R)-1 or (2)-(S)-1 marker in the enantiomers syn-(1)-(Rp,R,R,Rp)-5?2C6H6 (Fig. 3) and syn-(2)-(Sp,S,S,Sp)-5?2C6H6. The pair of co-ordinating N atoms bear a trans relationship, and each palladium atom in the metallacycle is in a slightly distorted square-planar coordination environment. With reference to the mean plane of the Pd and Br atoms, the pair of ferrocenyl groups take a syn arrangement. Accordingly the Pd2Br2 ring is slightly folded [the angle between the two planes defined by the atoms Pd(1), Br(1), Br(2) and Pd(2), Br(1), Br(2) is 22.08 for syn-(1)-(Rp,R,R,Rp)-5 and 21.98 for syn-(2)-(Sp,S,S,Sp)-5].The two five-membered palladocycles are nearly coplanar [tilt angle: 20.08 for syn-(1)- (Rp,R,R,Rp)-5 and 20.18 for syn-(2)-(Sp,S,S,Sp)-5]. The cyclopentadienyl rings are each planar and nearly parallel to each other [tilt angle: 1.48 for syn-(1)-(Rp,R,R,Rp)-5 and 1.18 for syn- (2)-(Sp,S,S,Sp)-5], and the two rings involved in the bicyclic system formed by fusion of the palladocycle with the ferrocenyl C5H3 moiety are approximately coplanar, the relevant dihedral angle being 4.78 for syn-(1)-(Rp,R,R,Rp)-5 and 4.98 for syn-(2)- (Sp,S,S,Sp)-5.Crystal structure of compound syn-(2)-(Sp,R,R,Sp)-5 As shown in Fig. 4, two ferrocenyl units of Sp configuration are bridged by two bromo ligands in syn-(2)-(Sp,R,R,Sp)-5. Unlike its diastereomers syn-(1)-(Rp,R,R,Rp)-5 and syn-(2)-(Sp,S,S, Sp)-5, the Pd2Br2 ring is markedly folded [the angle between the two planes defined by the atoms Pd(1), Br(1), Br(2) and Pd(2), Br(1), Br(2) is 52.18].Each palladium atom in the metallacycle is in a slightly distorted square-planar co-ordination environment. The two five-membered palladocycles exhibit a nonplanar, open-book shape (fold angle: 123.38). The pair of cyclopentadienyl rings are each planar and nearly parallel to each other (tilt angle: 1.78), and the two rings involved in the bicyclic system formed by fusion of the palladocycle with the ferrocenyl C5H3 moiety are virtually coplanar, the relevant dihedral angle being 0.28.Experimental General methods Proton and 13C-{1H} NMR were recorded on a Bruker DPX Fig. 3 Molecular structure (30% thermal ellipsoids) and absolute configuration of complex syn-(1)-(Rp,R,R,Rp)-5 with atom-numbering scheme.3788 J. Chem. Soc., Dalton Trans., 1998, 3785–3789 300 instrument using CDCl3 (99.8%) as solvent. Optical rotations were measured in CHCl3 in a 1 dm cell at 20 8C with a Perkin-Elmer model 341 polarimeter. Elemental analyses were performed by MEDAC Ltd.of the Department of Chemistry at Brunel University. Ferrocenecarbaldehyde, acetylferrocene, (1)-(R)- or (2)-(S)-1-amino-2-(methoxymethyl)pyrrolidine, palladium(II) acetate, and 5 Å molecular sieves were products of Aldrich and used as received. Compounds (1)-(S)-2, (2)- (R)-2, (1)-(S)-3 and (2)-(R)-3 were prepared by published methods.5,6 Preparations (2)-(Sp,R)- and (1)-(Rp,S)- [PdBr{(Á5-C5H3CH]] NN(CH2)3CHCH2OCH3)Fe(Á5-C5H5)}] 4. The hydrazone (2)-(R)-2 or (1)-(S)-2 (0.33 g, 1.0 mmol) was added to a methanolic (30 mL) solution containing Pd(O2CMe) 2 (0.22 g, 1.0 mmol) and NaO2CMe?3H2O (0.14 g, 1.0 mmol), and stirred at room temperature for 24 h.A solution of LiBr in 10 cm3 of methanol was added to the reaction mixture, and the red suspension stirred at room temperature for 15 min. The resulting reaction mixture was dried under high vacuum. The product was extracted into chloroform and isolated as a red solid via column chromatography (silica 60 and chloroform as eluent). The solid was subsequently recrystallized from dichloromethane by addition of n-hexane as red plates, yield 0.34 g (66%) for (2)-(Sp,R)-4 and 0.31 g (61%) for (1)-(Rp,S)-4.Characterization data for (2)-(Sp,R)-4: [a]D 20 2680.7 deg cm3 g21 dm21 (c 1.0 in CHCl3); 1H NMR (selected data) d 4.22 (s, C5H3), 4.32 (s, C5H3), 4.42 (s, C5H5), 5.07 (s, C5H3), 7.00 (s, CH]] N); 13C-{1H} NMR (selected data) d 68.3 [C4 (C5H3)], 70.9 (C5H5), 72.6 [C3 (C5H3)], 79.3 [C5 (C5H3)], 95.9 [C1 (C5H3)] and 148.8 (C]] N) (Found: C, 39.86; H, 4.08, N, 5.53. Calc.for C17H21BrFeN2OPd: C, 39.91; H, 4.14; N, 5.48%). For (1)- (Rp,S)-4: [a]D 20 1681.9 deg cm3 g21 dm21 (c 1.0 in CHCl3); 1H NMR (selected data) d 4.24 (s, C5H3), 4.33 (s, C5H3), 4.43 (s, C5H5), 5.08 (s, C5H3) and 6.99 (s, CH]] N); 13C-{1H} NMR (selected data) d 68.3 [C4 (C5H3)], 70.9 (C5H5), 72.4 [C3 (C5H3)], 79.5 [C5 (C5H3)], 95.9 [C1 (C5H3)] and 148.7 (C]] N) (Found: C, 39.99; H, 4.16; N, 5.39%).syn-(2)-(Sp,R,R,Sp)- and syn-(1)-(Rp,R,R,Rp)- [{Pd[(Á5-C5H3C(CH3)]] NN(CH2)3CHCH2OCH3)Fe(Á5-C5H5)]- (Ï-Br)}2] 5. The above procedure was repeated using hydrazone (2)-(R)-3 (0.34 g, 1.0 mmol). The product was eluted through a column of SiO2 with chloroform. Concentration of the eluted Fig. 4 Molecular structure (30% thermal ellipsoids) and absolute configuration of complex syn-(2)-(Sp,R,R,Sp)-5 with atom-numbering scheme.solution of two successive red bands produced complexes syn- (2)-(Sp,R,R,Sp)-5 and syn-(1)-(Rp,R,R,Rp)-5 in that order, which were recrystallized from dichloromethane–n-hexane (1 : 3) as red plates [product ratio 9 : 1, total yield 0.38 g (72%)]. Characterization data for syn-(2)-(Sp,R,R,Sp)-5: [a]D 20 23506.8 deg cm3 g21 dm21 (c 1.0 in CHCl3); 1H NMR (selected data) d 3.34 (s, 6 H, H3CC]] N), 4.24 [d, J = 6.0, 2 H, H3 (C5H3)], 4.34 [s, 12 H, C5H5 1 H5 (C5H3)] and 4.83 [t, J = 1.8 Hz, 2 H, H4 (C5H3)]; 13C-{1H} NMR (selected data) d 86.4 [C5 (C5H3)], 76.1 [C3 (C5H3)], 68.6 [C4 (C5H3)], 71.6 (C5H5), 102.6 [C1 (C5H3)] and 187.9 (C]] N) (Found: C, 41.50; H, 4.45; N, 5.55.Calc. for C18H23BrFeN2OPd: C, 41.13; H, 4.41; N, 5.33%). For syn-(1)- (Rp,R,R,Rp)-5: [a]D 20 12352.9 deg cm3 g21 dm21 (c 1.0 in CHCl3); 1H NMR (selected data) d 3.20 (s, 6 H, H3CC]] N), 4.25 [t, J = 2.7, 2 H, H3 (C5H3)], 4.19 (s, 10 H, C5H5), 4.87 [s, 2 H, H5 (C5H3)] and 4.34 [s, J = 2.7 Hz, 2 H, H4 (C5H3)]; 13C-{1H} (selected data) d 71.9 [C5 (C5H3)], 69.0 [C3 (C5H3)], 66.6 [C4 (C5H3)], 74.7 (C5H5), 100.7 [C1 (C5H3)] and 187.1 (C]] N) (Found: C, 41.48; H, 4.20; N, 5.07%).syn-(1)-(Rp,S,S,Rp)- and syn-(2)-(Sp,S,S,Sp)- [{Pd[(Á5-C5H3C(CH3)]] NN(CH2)3CHCH2OCH3)Fe(Á5-C5H5)]- (Ï-Br)}2] 5. Red plates of syn-(1)-(Rp,S,S,Rp)-5 and syn- (Sp,S,S,Sp)-5 were prepared according to the procedure described above using hydrazone (1)-(S)-3 as starting material. Product ratio 9 : 1 [syn-(1)-(Rp,S,S,Rp)-5 : syn-(2)-(Sp,S,S,Sp)- 5], total yield 0.36 g (69%)].Characterization data for syn-(1)- (Rp,S,S,Rp)-5: [a]D 20 13510.7 deg cm3 g21 dm21 (c 1.0 in CHCl3); 1H NMR (selected data) d 3.34 (s, 6 H, H3CC]] N), 4.24 [d, J = 6.0, 2 H, H3 (C5H3)], 4.34 [s, 12 H, C5H5 1 H5 (C5H3)] and 4.83 [t, J = 1.8 Hz, 2 H, H4 (C5H3)]; 13C-{1H} NMR (selected data) d 86.6 [C5 (C5H3)], 76.2 [C3 (C5H3)], 68.8 [C4 (C5H3)], 71.6 (C5H5), 102.9 [C1 (C5H3)] and 187.5 (C]] N) (Found: C, 41.32; H, 4.17; N, 5.60%).For syn-(2)-(Sp,S,S,Sp)-5: [a]D 20 22359.3 deg cm3 g21 dm21 (c 1.0 in CHCl3); 1H NMR (selected data) d 3.26 (s, 6 H, H3CC]] N), 4.25 [s, 2 H, H3 (C5H3)], 4.13 (s, 10 H, C5H5), 4.34 [s, 2 H, H5 (C5H3)] and 4.87 [s, J = 3.0 Hz, 2 H, H4 (C5H3)]; 13C-{1H} NMR (selected data) d 71.0 [C5 (C5H3)], 68.8 [C3 (C5H3)], 66.4 [C4 (C5H3)], 73.0 (C5H5), 97.5 [C1 (C5H3)] and 185.8 (C]] N) (Found: C, 41.24; H, 4.15; N, 5.18%). Crystallographic studies Crystallographic data of complexes (2)-(Sp,R)-4, (1)-(Rp,S)-4, syn-(2)-(Sp,R,R,Sp)-5, syn-(2)-(Sp,S,S,Sp)-5?2C6H6 and syn- (1)-(Rp,R,R,Rp)-5?2C6H6 measured on a MSC/Rigaku RAXIS IIC imaging-plate diVractometer are summarized in Table 1.Intensities were collected at 294 K using graphitemonochromatized Mo-Ka radiation (l = 0.7103 Å) from a rotating-anode generator operating at 50 kV and 90 mA (2qmin = 38, 2qmax = 558, 2–58 oscillation frames in the range of 0–1808, exposure 8 min per frame).11 A self-consistent semiempirical absorption correction based on Fourier coeYcient fitting of symmetry-equivalent reflections was applied using the ABSCOR program.12 The crystal structures of all four compounds were solved with the Patterson superposition method, and Fourierdi Verence syntheses.All the non-hydrogen atoms were refined anisotropically. The benzene solvate molecules in syn-(1)- (Rp,R,R,Rp)-5?2C6H6 and syn-(2)-(Sp,S,S,Sp)-5?2C6H6 are located in a general position with full site occupancy for their component atoms.Hydrogen atoms were all generated geometrically (C–H bond lengths fixed at 0.96 Å), assigned appropriate isotropic thermal parameters and allowed to ride on their parent carbon atoms. Full-matrix least-squares refinement on F2 was performed on an IBM-compatible 486 PC with the SHELXTL PC program package.13 Although the Flack x parameter 14 failed to give a clear indication of the absolute structure in the X-ray analysis, the known chiralities of (1)- (R)-1 and (2)-(S)-1 as starting synthetic materials ensured the correct assignment of the absolute configurations of all threeJ.Chem. Soc., Dalton Trans., 1998, 3785–3789 3789 Table 1 Crystal data Formula M Shape (color) Size/mm Crystal system Space group a/Å b/Å c/Å b/8 V/Å3 ZF (000) Dc/g cm23 m(Mo-Ka)/mm21 Reflections collected Independent reflections (Rint) R1 wR2 (1)-(Rp,S)-4 C17H21BrFeN2OPd 511.5 Prism (red) 0.35 × 0.25 × 0.15 Orthorhombic P212121 8.296(2) 13.865(3) 15.585(3) 1793(1) 4 1008 1.895 4.045 6054 3291 (0.0465) 0.0377 0.0899 syn-(2)-(Sp,R,R,Sp)-5 C36H46Br2Fe2N4O2Pd2 1051.1 Prism (red) 0.30 × 0.35 × 0.50 Monoclinic P21 12.872(1) 11.296(1) 13.337(1) 96.37(1) 1927(1) 2 1040 1.811 3.766 6579 6458 (0.0609) 0.0553 0.1571 syn-(1)-(Rp,R,R,Rp)-5?2C6H6 C48H58Br2Fe2N4O2Pd2 1207.3 Prism (red) 0.20 × 0.01 × 0.30 Orthorhombic P212121 10.654(2) 13.156(3) 34.577(7) 4847(2) 4 2416 1.655 3.007 14539 8380 (0.0870) 0.0615 0.1387 Table 2 Selected bond lengths (Å) and angles (8) of complexes (1)-(Rp,S)-4, syn-(2)-(Sp,R,R,Sp)-5 and syn-(1)-(Rp,R,R,Rp)-5?2C6H6 (1)-(Rp,S)-4 syn-(2)-(Sp,R,R,Sp)-5 syn-(1)-(RpR,R,Rp)-5?2C6H6 Pd(1)–C(1) Pd(1)–N(1) Pd(1)–O(1) Pd(1)–Br(1) N(1)–C(11) C(2)–C(11) C(1)–Pd(1)–N(1) C(1)–Pd(1)–O(1) N(1)–Pd(1)–O(1) C(1)–Pd(1)–Br(1) N(1)–Pd(1)–Br(1) O(1)–Pd(1)–Br(1) C(11)–C(2)–C(1) N(1)–C(11)–C(2) 1.937(2) 2.055(2) 2.228(2) 2.417(1) 1.288(3) 1.429(3) 81.3(1) 171.2(1) 90.6(1) 92.2(1) 173.4(1) 95.8(1) 115.9(2) 114.4(2) Pd(1)–C(1) Pd(1)–N(1) Pd(1)–Br(2) Pd(1)–Br(1) N(1)–C(11) C(2)–C(11) C(20)–C(29) C(1)–Pd(1)–N(1) C(1)–Pd(1)–Br(2) Br(2)–Pd(1)–Br(1) Pd(2)–Br(1)–Pd(1) Pd(1)–Br(2)–Pd(2) C(1)–C(2)–C(11) N(1)–C(11)–C(2) 1.955(3) 2.104(3) 2.465(1) 2.582(1) 1.292(4) 1.451(5) 1.437(5) 79.6(1) 93.0(1) 84.7(1) 83.6(1) 82.7(1) 116.6(3) 113.3(3) Pd(1)–C(1) Pd(1)–N(1) Pd(1)–Br(1) Pd(1)–Br(2) N(1)–C(11) C(2)–C(11) C(20)–C(29) C(1)–Pd(1)–N(1) C(1)–Pd(1)–Br(1) Br(1)–Pd(1)–Br(2) Pd(1)–Br(1)–Pd(2) Pd(2)–Br(2)–Pd(1) C(1)–C(2)–C(11) N(1)–C(11)–C(2) 1.951(4) 2.097(3) 2.455(1) 2.579(1) 1.322(5) 1.443(6) 1.424(6) 80.0(2) 91.8(1) 86.0(1) 91.4(1) 92.5(1) 118.2(4) 111.6(4) diastereomers.The final R1 and wR2 indices and other refinement parameters are presented in Table 1, and Table 2 gives selected bond distances and angles. CCDC reference number 186/1134. See http://www.rsc.org/suppdata/dt/1998/3785/ for crystallographic files in .cif format.Acknowledgements This work is supported by Hong Kong Research Grants Council Earmarked Grant CUHK 4022/98P. References 1 G. R. Newkome, W. E. Puckett, V. K. Gupta and G. E. Kiefer, Chem. Rev., 1986, 86, 451; J. Dupont, M. PfeVer, J.-C. Daran and J. Gouteron, J. Chem. Soc., Dalton Trans., 1988, 2421; J.-P. Sutter, M. PfeVer, A. de Cian and J. Fischer, Organometallics, 1992, 11, 386; M. PfeVer, J.-P. Sutter, A. de Cian and J. Fischer, Organometallics, 1993, 12, 1167; C.López, R. Bosque, X. Solans, M. Font-Bardía, J. Silver and G. Fern, J. Chem. Soc., Dalton Trans., 1995, 2445. 2 (a) R. Bosque, C. López, J. Sales, J. Solans and M. Font-Bardía, J. Chem. Soc., Dalton Trans., 1994, 735; (b) S. Q. Huo, Y. J. Wu, C. X. Du, Y. Zhu, H. Z. Yuan and X. A. Mao, J. Organomet. Chem., 1994, 483, 139; (c) R. Bosque, C. López, J. Sales, D. Tramuns and X. Solans, J. Chem. Soc., Dalton Trans., 1995, 2445; (d ) C. López, R. Bosque, D. Sainz, X. Solans and M.Font-Bardía, Organometallics, 1997, 16, 3261. 3 F. Maassarani, M. PfeVer, G. L. Borgne, J. T. B. H. Jastrzebski and G. van Koten, Organometallics, 1987, 6, 1111. 4 L. G. Kuzmina, Yu. T. Struchkov, L. L. Troitskayoa, V. I. Sokolov and O. A. Reutov, Izv. Akad. Nauk SSSR, Ser. Khim., 1979, 95, 1528; G. Zhao, F. Xue, Z.-Y. Zhang and T. C. W. Mak, Organometallics, 1997, 16, 4023. 5 G. Zhao, Q.-G. Wang and T. C. W. Mak, Organometallics, 1998, 16, 3405. 6 G. Zhao, Q.-G. Wang and T. C. W. Mak, Tetrahedron: Asymmetry, 1998, 9, 1557. 7 C. López, R. Bosque, X. Solans and M. Font-Bardía, Tetrahedron: Asymmetry, 1996, 7, 2527. 8 A. I. Meyers and K. A. Lutomski, in Asymmetric Synthesis, ed. J. D. Morrison, Academic Press, 1984, vol. 3, p. 213; D. Enders, ibid., p. 275; O. Riant, O. Samuel and H. B. Kagan, J. Am. Chem. Soc., 1993, 115, 5835. 9 J. Granell, R. Moragas, J. Sales, X. Solans and M. Font-Bardía, J. Chem. Soc., Dalton Trans., 1992, 822; J. Granell, R. Moragas, J. Sales, M. Font-Bardía and X. Solans, J. Organomet. Chem., 1992, 431, 359; C. López, R. Bosque, X. Solans and M. Font-Bardía, J. Organomet. Chem., 1997, 535, 99. 10 G. Zhao and T. C. W. Mak, unpublished work. 11 J. Tanner and K. Krause, The Rigaku Journal, 1994, 11, 4; 1990, 7, 28; K. Krause and G. N. Phillips, jun., J. Appl. Crystallogr., 1992, 25, 146; M. Sato, M. Yamamoto, Y. Katsube, N. Tanaka and T. Higashi, J. Appl. Crystallogr., 1992, 25, 348. 12 T. Higashi, ABSCOR, An Empirical Absorption Correction Based on Fourier CoeYcient Fitting, Rigaku Corporation, Tokyo, 1995. 13 SHELXL/PC Version 5.0 Reference Manual, Siemens Energy & Automation Inc., Madison, WI, 1996; G. M. Sheldrick, in Computational Crystallography 6, eds. H. D. Flack, L. Párkányi and K. Simon, International Union of Crystallography and Oxford University Press, 1993, p. 111. 14 J. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876; H. D. Flack and D. Schwarzenbach, Acta Crystallogr., Sect. A, 1988, 44, 499. Paper 8/06052E

ISSN:1477-9226    

DOI:10.1039/a806052e

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3791–3799 3791 Co-ordination chemistry of bis(ferrocenylcarbaldimine) SchiV bases Peiyi Li,a Ian J. Scowen,b John E. Davies a and Malcolm A. Halcrow*†a a Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge, UK CB2 1EW b Department of Chemistry and Chemical Technology, University of Bradford, Bradford, UK BD7 1DP Received 25th August 1998, Accepted 2nd October 1998 The complex chemistries of 1,2-bis(ferrocen-1-ylmethyleneamino)ethane (L1), 1,2-bis(ferrocen-1-ylmethyleneamino) benzene (L2) and its 4-methyl (L3), 4-chloro (L4) and 4-nitro (L5) derivatives have been reexamined.Complexation of L1 by ZnCl2 aVorded [ZnCl2(L1)], whose single crystal structure reveals a distorted tetrahedral zinc(II) centre with Zn ? ? ?Fe 4.730(2) and 4.803(2) Å. The complexes [ZnCl2(L)] (L = L2–L5), [ZnBr2(L)] (L = L2–L5) and [CoBr2(L3)] are accessible by MX2-templated condensation of 2 equivalents of ferrocenecarbaldehyde (fcCHO) with the appropriate 1,2-diaminobenzene.Treatment of [Cu(NCMe)4]X (X2 = BF4 2 or PF6 2) with L1, or fcCHO and the corresponding 1,2-diaminobenzene, yielded [Cu(L)2]X (L = L1–L3). The single crystal structure of [Cu(L3)2]- PF6?1.7CH2Cl2 shows a tetrahedral copper(I) centre, the chelate ligands being substantially distorted from planarity. Compounds [ZnCl2(L)], [ZnBr2(L)] (L = L2–L5) and [CoBr2(L3)] exhibit weak electronic communication between the two ferrocenyl centres, showing by cyclic voltammetry two chemically reversible FeII–FeIII oxidations separated by 50–60 mV in CH2Cl2–0.5 M NBun 4PF6 at 293 K; [ZnCl2(L1)] and [Cu(L)2]X (L = L1–L3) exhibit a single FeII–FeIII couple under these conditions.Attempted template syntheses of L2–L5 employing other MX2 (M = Mn, Co, Ni or Cu; X2 = Cl2, Br2, NO3 2, BF4 2 or ClO4 2) salts yielded primarily 2-ferrocenylbenzimidazole intramolecular cyclisation derivatives; the crystal structure of one such product was determined.Introduction Ferrocene-containing molecules are continuing to attract great interest as components in homogeneous catalysts,1 molecular sensors,2 and molecular magnetic 3 and non-linear optical 4 materials. Several groups have exploited SchiV base condensation of ferrocenecarbaldehyde [Fe(h-C5H5)(h-C5H4CHO)] as a facile method of substituting a ferrocene group onto an organic residue, and have studied the resultant SchiV bases [Fe(h-C5H5)(h-C5H4CR]] NR9)] (R = H, Me or Ph; R9 = alkyl or aryl) for their non-linear optical 5 and ligating 6 properties.Some ferrocenyl di-SchiV base macrocycles derived from ferrocene-1,19-dicarbaldehyde have also been structurally characterised.7 While monoferrocenyl SchiV bases appear to be relatively robust, previous attempts to prepare complexes of the bidentate diferrocenyl SchiV bases fcCH]] NYN]] CHfc (fc = ferrocenyl, Y = C2H4-1,2 L1 or C6H4-1,2 L2) have been unsuccessful.Two groups have previously reported the synthesis of L1; 8,9 however, it was found that complexation of L1 resulted in its spontaneous hydrolytic degradation, and it was necessary to reduce the L1 aldimine moieties with AlH4 2 before a useful ligand was obtained. Similarly, attempts to prepare the related ligand L2 by condensation of ferrocenecarbaldehyde with 1,2-diaminobenzene resulted in spontaneous intramolecular cyclisation of the inital di-SchiV base, forming instead the 2-ferrocenylbenzimidazole derivatives 1H and 2H.10 Similar observations regarding the chemical sensitivity of the ferrocenecarbaldimine moiety have also been made by others.11 Given our interest in the preparation of ferrocene-substituted complexes,12 we decided to re-examine this chemistry. We noted † Current address: School of Chemistry, University of Leeds, Woodhouse Lane, Leeds, UK LS2 9JT.E-mail: M.A.Halcrow@ chemistry.leeds.ac.uk that (fcCH]] N)2C6H4-1,4, a geometric isomer of L2, is stable.13 In addition, other groups have previously reported the synthesis and complex chemistry of monoferrocenyl ligands related to L2, of type [fcCH]] NC6H4X-2]2 (X2 = O214 or S2 15).In particular, it was reported that treatment of 3 16 with salts of NiII, ZnII or PdII resulted in opening of the thiazoline ring, aVording the complexes trans-[M(N,S-fcCH]] NC6H4S-2)2] (M = Ni, Zn or Pd).15 We therefore suspected that complexes of L1 and L2 might be accessible by metal-templated condensations of fcCHO with the appropriate diamine, thus removing the N Fe N Fe L1 R = H; L2 R = Me; L3 R = Cl; L4 R = NO2; L5 N Fe N Fe R3792 J.Chem. Soc., Dalton Trans., 1998, 3791–3799 Table 1 Analytical and selected FAB mass spectrometric data for the new compounds Analysis (%) Compound 4 [ZnCl2(L1)] 5 [ZnCl2(L2)] 6 [ZnCl2(L3)] 7 [ZnCl2(L4)] 8 [ZnCl2(L5)] 9 [ZnBr2(L2)] 10 [ZnBr2(L3)] 11 [ZnBr2(L4)] 12 [ZnBr2(L5)] 13 [CoBr2(L3)] 14?PF6?2H2O [Cu(L1)2]PF6?2H2O 15?BF4?CH3NO2 [Cu(L2)2]BF4?CH3NO2 16?BF4 [Cu(L3)2]BF4 16?PF6 [Cu(L3)2]PF6 [1HH]BF4 C 48.5 (49.0) 52.9 (52.8) 53.7 (53.5) 50.4 (50.1) 49.3 (48.6) 47.0 (46.4) 46.6 (47.1) 44.3 (44.3) 43.6 (43.7) 47.1 (47.5) 50.2 (50.2) 56.2 (56.5) 59.1 (59.1) 55.9 (56.3) 57.0 (57.1) H 4.1 (4.1) 3.8 (3.8) 4.1 (4.0) 3.5 (3.5) 3.4 (3.4) 3.4 (3.3) 3.5 (3.5) 3.0 (3.1) 3.0 (3.0) 3.5 (3.6) 4.4 (4.6) 4.3 (4.2) 4.8 (4.4) 4.4 (4.2) 4.3 (4.3) N 4.7 (4.8) 4.4 (4.4) 4.5 (4.3) 4.2 (4.2) 6.1 (6.2) 3.8 (3.9) 3.7 (3.8) 3.5 (3.7) 5.3 (5.5) 3.6 (3.8) 4.7 (4.9) 5.5 (5.8) 4.6 (4.8) 4.4 (4.5) 4.6 (4.8) m/z b 586, 551, 452 634, 599, 500 648, 613, 514 668, 633, 534 644, 545 643, 500 736, 657, 517 756, 677, 534 688, 545 731, 652, 514 967, 515 1063, 563 1091, 577 —— a Calculated values in parentheses.b Peaks for compounds 4–13 are assigned to the ions [MX2(L)]1 (not always observed), [MX(L)]1 and [L]1 (M = 59Co or 64Zn; X2 = 35Cl2 or 79Br2; L = L1–L5) and exhibit the correct isotopic distributions.Peaks for 14?BF4–16?BF4 are assigned to the ions [63Cu(L)2]1 and [63Cu(L)]1 (L = L1–L3), and exhibit the correct isotopic distributions. requirement for the free SchiV bases. We report here the results of this study. Results and discussion Complex syntheses Attempted condensations of ferrocenecarbaldehyde with 1,2- diaminoethane in the presence of zinc(II) or copper(II) salts as templates aVorded only en-containing complex products. We therefore decided to reinvestigate the coordination chemistry of preformed L1.8,9 As in a previous report,9 treatment of L1 with anhydrous MX2 (M = Ni or Cu; X2 = Cl2 or Br2) or hydrated M(BF4)2 (M = Ni, Cu or Zn) in CH2Cl2 or MeCN aVorded no isolable ferrocene-containing products.However, reaction of L1 with 1 molar equivalent of ZnCl2 in CH2Cl2 yields large brown air-stable crystals after layering with hexanes, which analyse as [ZnCl2(L1)] 4 (Table 1). This product is always contaminated with a small number of paler crystals of fcCHO, from which it must be separated manually.The 1H and 13C NMR spectra of 4 confirm that L1 is intact in the complex, but also consistently show the presence of 10–25 mol% of fcCHO. That this contaminant originates from 4 was confirmed by 1H NMR analysis of a single crystal of 4, which clearly showed the presence of the carbaldehyde. Fe Fe R N N Fe N X 1R (R = H, Me, Cl, NO2) X = NH; 2R (R = H, Me, Cl, NO2) X = S, R = H; 3 R A large number of MX2-templated (M = Mn, Co, Ni, Cu or Zn; X2 = Cl2, Br2, MeCO2 2 or NO3 2) condensations of fcCHO with 1,2-diaminobenzene, 1,2-diamino-4-methylbenzene, 1,2-diamino-4-chlorobenzene or 1,2-diamino-4-nitrobenzene were attempted, with the aim of producing complexes of stoichiometry [MX2(L)] (L = L2–L5).When ZnCl2 or ZnBr2 was used, syntheses in refluxing CHCl3 gave the complexes [ZnCl2(L)] (L = L2 5, L3 6, L4 7 or L5 8) and [ZnBr2(L)] (L = L2 9, L3 10, L4 11 or L5 12; Table 1) in moderate yields.Similar reactions involving all these diaminobenzene derivatives with other zinc(II) salts or salts of MnII, CoII or NiII yielded crude solids containing predominantly 1R, [1RH]X and/or 2R (R = H, Me, Cl or NO2).10 In only one case it was possible to purify a [MX2(L)] complex from these reactions containing a metal other than zinc, namely [CoBr2(L3)] 13. Compounds 5–13 are all moderately soluble deep red microcrystalline solids. The 1H and 13C NMR spectra of 5–12 in CDCl3 are free from fcCHO and are consistent with the presence of intact L2–L5 (13C spectra of 8 and 12 could not be obtained because of their reduced solubility), showing that intramolecular cyclisation to 1R has not occurred.The paramagnetic 1H NMR spectrum of 13 in CDCl3 is also consistent with the presence of L3, and was assigned from the peak integrals and by comparison with literature spectra of related complexes (Experimental section).17 Interestingly, the ferrocenyl proton resonances exhibit upfield contact shifts, consistent with a p-spin-delocalisation pathway.This implies that substantial conjugation exists between the ferrocene and cobalt(II) centres of the complex. In an attempt to produce complexes of formula [M(L)2]21 (L = L2–L5), hydrated M(BF4)2 (M = Co, Ni, Cu or Zn) and Mn(ClO4)2 salts were treated with fcCHO and the appropriate diaminobenzene in a 1:4:2 mole ratio in refluxing CH3OH. Generally, these reactions again aVorded only 1R or [1RH]X.However, when M = Cu and L = L2 or L3, deep red products were obtained of stoichiometry [Cu(L)2]BF4 (L = L2, 15?BF4 or L3, 16?BF4; Table 1). Both compounds aVorded diamagnetic 1H NMR spectra, confirming that reduction of CuII to CuI had taken place. The reductant is probably fcCHO, since increased yields were obtained when excess of it was employed in these syntheses. Further improved yields of 15?X and 16?X (X2 = BF4 2 or PF6 2) could be obtained by employing [Cu(NCMe)4]X rather than a copper(II) salt as template under the conditions described above.Complexation of [Cu(NCMe)4]PF6 by preformed L1 aVorded an orange product [Cu(L1)2]PF6 14?PF6. Although air-stable as solids, 14?PF6–16?BF4 decompose in chlorinated solvents, MeNO2 or MeCN over a period of days.J. Chem. Soc., Dalton Trans., 1998, 3791–3799 3793 Single crystal structures A single crystal X-ray analysis of complex 4 shows the expected pseudo-tetrahedral zinc(II) centre (Fig. 1, Table 2).The Zn–Cl bond lengths are typical of those observed for such compounds; however, the Zn–N distances are somewhat short compared to other zinc(II) SchiV base complexes.18 Aside from the acute N(1)–Zn(1)–N(2) angle of 84.3(1)8 enforced by the bite of L1, the donor atoms adopt a nearly regular tetrahedral geometry. Hence, the average bond angle at Zn(1) is 108.98, compared to 109.58 for an ‘ideal’ tetrahedron, while the dihedral angle ‘q’ between the planes of [Zn(1), Cl(1), Cl(2)] and [Zn(1), N(1), N(2)] is 87.3(2)8, compared to an ideal value of 908.The ferrocenyl substituents are in a transoid arrangement with respect to each other and are almost coplanar with the L1 aldimine groups, the dihedral angle between the planes [C(11)–C(15)] and [C(11)–C(10)–N(1)] being 8.7(2)8 and that formed by [C(21)–C(25)] and [C(21)–C(20)–N(2)] being 8.5(2)8. This means that one a-C–H bond of each substituted C5H4 ring is oriented close to the zinc(II) ion, with Zn(1) ? ? ? H(12) 2.90 and Zn(1) ? ? ? H(22) 2.86 Å.Other bond lengths and angles within the molecule are unexceptional. The intermetallic distances are Zn(1) ? ? ?Fe(1) 4.803(2), Zn(1) ? ? ?Fe(2) 4.730(2) and Fe(1) ? ? ?Fe(2) 9.522(2) Å. While salts of 141 and 151 do not crystallise well, platelets of 16?BF4 and 16?PF6 could be obtained by layering concentrated CH2Cl2 solutions of the complexes with hexanes. Structural analyses of both salts were undertaken; unfortunately, crystals of 16?BF4?CH2Cl2 [orthorhombic, space group Pbca, a = 22.387(5), b = 21.830(4), c = 22.877(5) Å] included a second (minor) crystal domain while 16?PF6?1.7CH2Cl2 suVered substantial disorder of the lattice solvent.Since the complex cations in both structures were crystallographically indistinguishable, only the structure of 16?PF6 will be discussed in detail. The structure contains one complex cation in the asymmetric unit, which exhibits a distorted tetrahedral structure with Cu–- Fig. 1 Structure of the [ZnCl2(L1)] molecule in the crystal of complex 4, showing the atom numbering scheme employed. For clarity, all hydrogen atoms have been omitted. Table 2 Selected bond lengths (Å) and angles (8) for [ZnCl2(L1)] 4 Zn(1)–Cl(1) Zn(1)–Cl(2) Zn(1)–N(1) Zn(1)–N(2) N(1)–C(10) C(1)–N(1) Cl(1)–Zn(1)–Cl(2) Cl(1)–Zn(1)–N(1) Cl(1)–Zn(1)–N(2) Cl(2)–Zn(1)–N(1) Cl(2)–Zn(1)–N(2) N(1)–Zn(1)–N(2) C(1)–N(1)–Zn(1) C(10)–N(1)–Zn(1) C(1)–N(1)–C(10) N(1)–C(1)–C(2) 2.223(1) 2.210(1) 2.073(4) 2.063(3) 1.281(6) 1.466(6) 114.06(6) 116.3(1) 112.8(1) 111.6(1) 114.6(1) 84.3(1) 106.6(3) 135.9(3) 117.6(4) 109.1(4) C(1)–C(2) N(2)–C(2) N(2)–C(20) C(10)–C(11) C(20)–C(21) C(1)–C(2)–N(2) C(2)–N(2)–Zn(1) C(20)–N(2)–Zn(1) C(2)–N(2)–C(20) N(1)–C(10)–C(11) C(10)–C(11)–C(12) C(10)–C(11)–C(15) N(2)–C(20)–C(21) C(20)–C(21)–C(22) C(20)–C(21)–C(25) 1.506(7) 1.487(6) 1.270(6) 1.440(6) 1.441(6) 107.7(4) 107.2(3) 135.5(3) 117.3(4) 127.1(4) 129.8(4) 121.9(4) 126.6(4) 130.2(4) 122.3(4) N distances of 2.030(9)–2.071(8) Å (Table 3, Fig. 2). The average N–Cu–N angle in the structure is 110.48, while the dihedral angle ‘q’ between the planes of the L3 ligands [N(1A), Cu(1), N(2A)] and [N(1B), Cu(1), N(2B)] is 87.8(3)8. Therefore, while the geometry at copper is distorted by the bite of the L3 chelate, there is minimal flattening of the CuN4 tetrahedron. Other reported copper(I) complexes of diimine chelates often exhibit substantially flattened structures, with q as low as 498.19,20 In contrast to 4, the ferrocenyl substituents on each ligand are cisoid to each other. The Cu ? ? ?Fe distances are Cu(1) ? ? ?Fe(1) 4.783(2), Cu(1) ? ? ?Fe(2) 5.097(3), Cu(1) ? ? ?Fe(3) 4.651(2), Cu(1) ? ? ?Fe(1) 5.010(2) Å, while the Fe ? ? ?Fe distances range from 6.498(2) to 9.876(2) Å.Despite the regular geometry at copper, the L3 ligands are substantially distorted from planarity (Fig. 3).This distortion is characterised by the dihedral angles between the phenylene and ferrocenecarbaldimine moieties, and between the phenylene ring and the plane formed the two N donors and copper ion (Table 4), both of which should be zero for a planar, fully con- Fig. 2 Structure of the [Cu(L3)2]1 cation in the crystal of 16?PF6? 1.7CH2Cl2, showing the atom numbering scheme employed. For clarity, all hydrogen atoms have been omitted. The methyl substituent on one ligand is disordered over two sites C(7A) and C(8A).Fig. 3 Alternative view of the [Cu(L3)2]1 cation in the crystal of 16?PF6?1.7CH2Cl2, emphasising the bending of the L3 ligands. Details as for Fig. 2.3794 J. Chem. Soc., Dalton Trans., 1998, 3791–3799 jugated ligand. These distortions are a consequence of a close contact formed by one a-C–H bond of each ferrocenyl group with Cu(1) (cf. 4, see above). These distances are Cu(1) ? ? ? H(12) 2.74, Cu(1) ? ? ? H(22) 2.89, Cu(1) ? ? ? H(32) 2.70, Cu(1) ? ? ? H(42) 2.83 Å.Thus, the ferrocenyl units are forced to twist away from the copper ion and, in order to retain planarity at the sp2 N atoms, the L3 phenylene groups become displaced out of the CuN2 plane. Crystals of compound [1HH]BF4 were isolated from a crude reaction mixture of Zn(BF4)2.6H2O, fcCHO and 1,2-diaminobenzene in MeCN. By comparison with the previously reported structure of unprotonated 1H,10 the angle C(1)–N(2)–C(2) in [1HH]BF4 is significantly greater [110.7(4) vs. 104.3(4)8], while N(1)–C(1)–N(2) is correspondingly smaller [106.7(4) vs. 112.4(4)8] (Fig. 4, Table 5). All other bond lengths and angles in the complex molecules are crystallographically indistinguishable in the two structures. There is a hydrogen bond between the benzimidazolium proton and BF4 2 anion (Fig. 4), with N(2) ? ? ? F(1) 2.772(6), H(2) ? ? ? F(1) 2.03(3) Å and N–H ? ? ?F 161(5)8. UV/visible spectroscopy Ferrocene exhibits two absorptions in the visible and near UV at nmax = 22.7 (emax = 110) and 30.8 × 103 cm21 (49 M21 cm21), assigned to spin-allowed d–d transitions.21 Compounds L1 and 4–13 also exhibit two absorptions in CH2Cl2 solution (Table 6).These transitions lie at lower energy for 4 compared to uncom- Table 3 Selected bond lengths (Å) and angles (8) for [Cu(L3)2]PF6? 1.7CH2Cl2 16?PF6?1.7CH2Cl2 Cu(1)–N(1A) Cu(1)–N(1B) Cu(1)–N(2A) Cu(1)–N(2B) N(1A)–C(10) C(1A)–N(1A) C(1A)–C(2A) N(2A)–C(2A) N(2A)–C(20) N(1A)–Cu(1)–N(1B) N(1A)–Cu(1)–N(2A) N(1A)–Cu(1)–N(2B) N(1B)–Cu(1)–N(2A) N(1B)–Cu(1)–N(2B) N(2A)–Cu(1)–N(2B) C(1A)–N(1A)–Cu(1) C(10)–N(1A)–Cu(1) C(1A)–N(1A)–C(10) N(1A)–C(1A)–C(2A) N(1A)–C(1A)–C(6A) N(2A)–C(2A)–C(1A) N(2A)–C(2A)–C(3A) C(2A)–N(2A)–Cu(1) C(20)–N(2A)–Cu(1) C(2A)–N(2A)–C(20) N(1A)–C(10)–C(11) C(10)–C(11)–C(12) C(10)–C(11)–C(15) 2.071(8) 2.030(9) 2.050(8) 2.041(9) 1.32(1) 1.43(1) 1.39(1) 1.42(1) 1.32(1) 128.7(3) 81.9(3) 121.5(3) 125.5(3) 81.9(4) 123.1(3) 107.2(6) 132.4(7) 120.1(8) 118.7(9) 123.1(9) 116.1(8) 123.2(9) 109.1(6) 131.5(7) 118.1(9) 125.7(9) 129.1(9) 122(1) C(10)–C(11) C(20)–C(21) N(1B)–C(30) C(1B)–N(1B) C(1B)–C(2B) N(2B)–C(2B) N(2B)–C(40) C(30)–C(31) C(40)–C(41) N(2A)–C(20)–C(21) C(20)–C(21)–C(22) C(20)–C(21)–C(25) C(1B)–N(1B)–Cu(1) C(30)–N(1B)–Cu(1) C(1B)–N(1B)–C(30) N(1B)–C(1B)–C(2B) N(1B)–C(1B)–C(6B) N(2B)–C(2B)–C(1B) N(2B)–C(2B)–C(3B) C(2B)–N(2B)–Cu(1) C(40)–N(2B)–Cu(1) C(2B)–N(2B)–C(40) N(1B)–C(30)–C(31) C(30)–C(31)–C(32) C(30)–C(31)–C(35) N(2B)–C(40)–C(41) C(40)–C(41)–C(42) C(40)–C(41)–C(45) 1.40(1) 1.45(1) 1.30(1) 1.41(1) 1.38(2) 1.40(1) 1.27(1) 1.45(1) 1.43(2) 125(1) 127(1) 126(1) 108.9(7) 132.1(8) 119(1) 119(1) 123(1) 116(1) 123(1) 110.4(7) 130.7(8) 117(1) 127(1) 128(1) 124(1) 126(1) 127(1) 129(1) Table 4 Selected torsion angles (8) for [Cu(L3)2]PF6?1.7CH2Cl2 16?PF6?1.7CH2Cl2, describing the structural distortions of the ligands Ligand A [N(1A), N(2A), C(1A)–C(6A)]–[N(1A), C(10)–C(15)] [N(1A), N(2A), C(1A)–C(6A)]–[N(2A), C(20)–C(25)] [N(1A), N(2A), C(1A)–C(6A)]–[Cu(1), N(1A), N(2A)] 36.6(3) 53.7(3) 27.5(2) Ligand B [N(1B), N(2B), C(1B)–C(6B)]–[N(1B), C(30)–C(35)] [N(1B), N(2B), C(1B)–C(6B)]–[N(2B), C(40)–C(45)] [N(1B), N(2B), C(1B)–C(6B)]–[Cu(1), N(1B), N(2B)] 24.3(4) 48.1(4) 21.4(4) plexed L1, suggesting that the iron centres become less electronrich upon co-ordination.Compared to 4, the lower energy visible transition exhibited by 5–12 is red-shifted, as expected upon replacement of the N-alkyl substituent in 4 by an Fig. 4 Structure of the molecule in the crystal of [1HH]BF4, showing the atom numbering scheme employed. For clarity, all C-bound hydrogen atoms have been omitted. Table 5 Selected bond lengths (Å) and angles (8) for [1HH]BF4 N(1)–C(1) N(1)–C(7) N(1)–C(10) C(1)–N(2) C(1)–C(31) N(2)–C(2) N(2)–H(2) C(1)–N(1)–C(7) C(1)–N(1)–C(10) C(7)–N(1)–C(10) N(1)–C(1)–N(2) N(1)–C(1)–C(31) N(2)–C(1)–C(31) C(1)–N(2)–C(2) C(1)–N(2)–H(2) C(2)–N(2)–H(2) N(2)–C(2)–C(3) N(2)–C(2)–C(7) C(3)–C(2)–C(7) 1.355(6) 1.392(7) 1.485(5) 1.362(6) 1.432(7) 1.367(7) 0.77(3) 109.3(4) 125.8(5) 124.9(4) 106.7(4) 130.0(4) 123.3(4) 110.7(4) 121(5) 128(5) 132.3(4) 106.3(4) 121.4(5) C(2)–C(7) C(2)–C(3) C(3)–C(4) C(4)–C(5) C(5)–C(6) C(6)–C(7) C(10)–C(11) C(2)–C(3)–C(4) C(3)–C(4)–C(5) C(4)–C(5)–C(6) C(5)–C(6)–C(7) N(1)–C(7)–C(2) N(1)–C(7)–C(6) C(2)–C(7)–C(6) N(1)–C(10)–C(11) C(10)–C(11)–C(12) C(10)–C(11)–C(15) C(1)–C(31)–C(32) C(1)–C(31)–C(35) 1.389(6) 1.396(8) 1.380(8) 1.378(8) 1.372(8) 1.394(7) 1.507(6) 116.9(5) 121.1(5) 122.9(5) 116.6(5) 107.0(4) 132.0(4) 121.0(5) 110.9(4) 127.3(4) 124.4(5) 123.4(5) 129.6(4) Table 6 The UV/visible spectroscopic data for the compounds (CH2Cl2, 293 K) Compound L1 456789 10 11 12 13 14?PF6 15?BF4 16?BF4 1023 n& max/cm21 (emax/M21 cm21) 22.2 (760), 31.0 (2 500) 21.1 (2 200), 28.2 (3 800) 19.5 (7 100), 29.8 (21 600) 19.5 (6 600), 29.4 (21 000) 19.1 (7 900), 29.3 (24 200) 18.4 (7 900), 28.6 (23 000) 19.5 (6 400), 29.7 (19 400) 19.5 (7 600), 29.3 (22 800) 19.1 (8 400), 29.2 (24 700) 18.3 (9 300), 28.3 (23 900) 19.0 (7 700), 28.3 (22 200), 32.3 (sh) 22.0 (3 600), 29.0 (14 600), 36.6 (27 900) 20.1 (12 400), 27.8 (sh), 32.2 (36 300) 20.1 (12 300), 27.8 (sh), 31.8 (36 000)J. Chem.Soc., Dalton Trans., 1998, 3791–3799 3795 electron-withdrawing phenylene group; anomalously, however, the higher energy near-UV absorption lies at higher energy for 5–12 than for 4.For a given ligand L, the peak energies shown by [ZnX2(L)] are essentially identical for X2 = Cl2 and Br2, showing the trends in nmax of L5 < L4 ª L3 < L2 for the low energy band, and L5 < L4 < L3 ª L2 for the higher energy peak. The intensities of these bands show small variations between compounds, however, suggesting that some solvolysis may be taking place in this solvent. The dependence of nmax on the identity of the ligand phenylene substituent suggests that the two absorption maxima shown by compounds 5–12 may not have pure d–d character.22 In order to confirm this suggestion, UV/VIS spectra of 9–12 were run in two other weakly co-ordinating solvents, in which solvolysis of the zinc(II) centres is likely to be small. There is a small but reproducible solvatochromism associated with both UV/VIS absorptions (Table 7), which follows the polarity of the solvents as expressed by the ET scale of Reichardt 23 reasonably well.This is behaviour expected of charge transfer rather than d–d transitions,24 and contrasts with ferrocene whose two UV/VIS absorptions exhibit no solvatochromism.25 Hence, it appears that both absorptions for 5–12 have a charge-transfer as well as (presumably) a d–d component.The UV/VIS spectrum of compound 13 exhibits the same two absorptions as 5–12, although these are red-shifted compared to 6 and 10 (Table 6). While tetrahedral cobalt(II) complexes generally show a d–d absorption in the range 15 000–- 20 000 cm21,24 this was not detected for 13.Presumably, this is obscured by the stronger L3-based absorption at 19 000 cm21. For the copper(I) complexes 14?PF6–16?BF4, the lowest energy absorption is blue-shifted compared to those of the zinc(II) complexes of the same ligands (Table 6), consistent with the less Lewis acidic nature of the copper(I) ion. The peak near 29 000 cm21 for 14?PF6 is also blue-shifted compared to that of 4, and is more intense than might be expected, suggesting that this band may also contain a CuÆp* MLCT component.19,26 For 15?BF4 and 16?BF4 this peak forms a shoulder, so that its energy and intensity cannot be accurately measured.Compounds 13–- 16?BF4 also exhibit an additional very intense peak above 30000 cm21, which we assign to a Fe-based MLCT transition.12,21 Electrochemistry Cyclic voltammograms of compounds 4–16?BF4 were measured in CH2Cl2–0.5 M NBun 4PF6 at 293 K. All potentials are quoted vs.the ferrocene–ferrocenium couple, and were measured at a scan rate of 100 mV s21 unless otherwise stated. The voltammetric data obtained are summarised in Table 8. Complex 4 exhibits a single chemically reversible FeII–FeIII oxidation at E2� 1 = 10.29 V, with a peak-to-peak separation similar to that shown by ferrocene under our conditions. This half-potential for 4 is more positive than that shown by L1 in the same solvent (10.14 V 9), reflecting an inductive interaction between the Lewis acidic zinc(II) ion and the L1 iron centres.No splitting of this wave into separate one-electron components was observed [Fig. 5(a)], so that there is negligible communication between the two ferrocenyl moieties in co-ordinated L1. Table 7 The UV/visible spectroscsopic data for [ZnBr2(L)] (L = L2–L5) in diVerent solvents (293 K) ET a/kcal mol21 Toluene 33.9 Ethyl acetate 38.1 CH2Cl2 41.1 Compound 1023 n& max/cm21 9 10 11 12 19.3, 29.3 19.3, 29.0 19.0, 28.8 18.3, 28.2 19.6, 29.8 19.3, 29.3 19.4, 29.1 18.6, 28.7 19.5, 29.7 19.5, 29.3 19.2, 29.2 18.3, 28.3 a Ref. 23. Importantly, this shows that electronic interactions between the iron ions in L1 and, by extension in L2–L5, are not mediated by the co-ordinated zinc(II) ion. Weak communication between two ferrocenylcarbaldimine centres co-ordinated in a trans disposition about a nickel(II) ion has been described by others.15,27 In contrast to 4, the complexes of L2–L4 display two distinct chemically reversible FeII–FeIII oxidations at E2� 1 = 10.32 ± 0.01 and 10.38 ± 0.01 V [Table 8; Fig. 5(b)]. The L5 complexes 8 and 12 gave broadened cyclic voltammograms, so that no splitting of this couple could be discerned. The FeII–FeIII halfpotentials for the ligands in 5–12 follow the sequence L3 < L2 ª L4 < L5, which is consistent with the inductive properties of the phenylene substituents on these ligands. For 5 and 9, whose ferrocene substituents are in identical chemical environments, the splitting DE2� 1 = 60 mV can be taken as a measure of communication between the ferrocene centres across the phenylenediimine moiety.28 Given the lack of a similar splitting for 4, this electronic communication is probably mediated by the ligand p system.Complexes 6, 7, 10, 11 and 13 Fig. 5 Semiderivative cyclic voltammograms in CH2Cl2–0.5 M NBun 4- PF6 at 293 K and 100 mV s21 of (a) [ZnCl2(L1)] 4 and (b) [ZnCl2(L2)] 5. The irreversible ligand reduction near 21.9 V for 5 is not shown.Table 8 Voltammetric data for the compounds (CH2Cl2–0.5 M NBun 4PF6 or MeCN–0.1 M NBun 4PF6, 293 K, 100 mV s21). All potentials quoted vs. an internal ferrocene–ferrocenium standard Compound 456789 10 11 12 13 14?PF6 15?BF4 16?BF4 Solvent CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 CH2Cl2 MeCN CH2Cl2 MeCN CH2Cl2 MeCN E2� 1 (FeII–FeIII)/V 10.29 10.32, 10.38 10.31, 10.36 10.32, 10.38 10.37 10.32, 10.38 10.31, 10.36 10.33, 10.38 10.36 10.31, 10.37 10.31 a 10.31 10.29 10.30 10.27 10.23 Other peaks Epa/V ————————— 10.69 10.10 b — 10.9 c — 10.93 — a Irreversible process, Epa quoted.b Chemically reversible process, E2� 1 value quoted. c Broad peak.3796 J. Chem. Soc., Dalton Trans., 1998, 3791–3799 show an identical DE2� 1 to those of the L1 complexes, showing that substitution at the 4 position of the phenylene ring in L2 aVects both iron centres in the molecule to an approximately equal extent.Complex 13 exhibits an additional irreversible oxidation with Epa = 10.69 V, Ipa ª 2� 1 Ipa[(FeII–FeIII)] and no detectable daughter peaks, which is not exhibited by 5–12. This is assigned to a CoII–CoIII process. The irreversibility of this oxidation may arise from the reduced co-ordinating ability of the ‘soft’ diimine ligand L3 for a high oxidation state cobalt(III) ion, the desire of a d6 cobalt(III) ion to attain an octahedral geometry, and/or the vulnerability of the ferrocenecarbaldimine C]] N bond towards a highly Lewis acidic cobalt(III) centre.The cyclic voltammograms of 14?PF6–16?BF4 in CH2Cl2 show a single FeII–FeIII oxidation, whose return wave is partially obscured by a desorption spike of variable intensity, which diminishes at increased scan rates; for 14?PF6, this spike is particularly intense, so that the oxidation is eVectively irreversible. We ascribe this behaviour to deposition of the very highly charged [Cu(L)2]n1 [L = L1; n = 6 (see below); L = L2 or L3; n = 5] oxidation products onto the electrode, reflecting their insolubility in the non-polar CH2Cl2 medium.Consistent with this, cyclic voltammograms of 14?PF6–16?BF4 in the more polar solvent MeCN–0.1 M NBun 4PF6 now show a single chemically reversible FeII–FeIII oxidation (Table 8). The trend in E2� 1 (FeII– FeIII) for these complexes is 16?BF4 < 15?BF4 < 14?PF6, which contrasts with the behaviour observed for the [ZnCl2(L)] complexes of the same ligands, namely 4 < 6 < 5 (Table 7).The lack of communication between the ferrocenyl groups in 15?BF4 and 16?BF4 may reflect reduced conjugation within the ligands caused by twisting of the carbaldimine moieties upon coordination (Fig. 3). The copper(I) complexes also exhibit a second oxidation, with Ipa ca. 4� 1 Ipa(FeII–FeIII). For 14?PF6 this is a chemically reversible process in CH2Cl2 with E2� 1 = 10.10 V, which we assign to a CuI–CuII couple.This is typical behaviour for copper(I) bis(diimine) complexes, which generally exhibit chemically reversible CuI–CuII couples in the range E2� 1 = –0.3 to 0.7 V.19,29 In MeCN this peak is not observed, and is probably obscured by the FeII–FeIII oxidation. Since E2� 1 (CuI–CuII) < E2� 1 (FeII–FeIII) in CH2Cl2, the FeII–FeIII half-potential measured for 14?PF6 in this solvent is in fact derived from the species [Cu(L1)2]21. This probably explains the increased Epa(FeII–FeIII) value measured for 14?PF6 compared to that for 4 (see above).For 15?BF4 and 16?BF4 a broad irreversible oxidation occurs in CH2Cl2 at Epa ª 10.9 V. This is a very anodic potential for a CuI–CuII process, however, and given its irreversibility the assignment of this peak is therefore uncertain. This peak was not detected in MeCN. Concluding remarks Our results have shown that, while in contrast to previous reports 9,10 complexes of L1–L5 can be prepared, these ligands only form isolable complexes with certain metal centres.We ascribe these observations mainly to steric factors. It is unusual for hydrolysis of an imine like L1 to be promoted by coordination to a metal ion, since p-back donation into the C]] N p* orbital generally reduces the electrophilicity of the imine C atom.30 However, as this work has shown, coordinated ferrocenecarbaldimines exhibit short M ? ? ? H–C contacts which can lead to substantial structural distortions within the co-ordinated ligands (cf. 16?PF6, Fig. 3). We therefore suggest that for L1 it is the introduction of steric strain within the imine linkers upon co-ordination which leads to its rapid hydrolysis in the presence of transition ions. Ligands L2–L5 are less sensitive to nucleophilic attack, because of conjugation between their imine moieties and the phenylene backbone. However, given their increased conformational rigidity, steric interactions may well prevent a strong complex being formed to most metal ions, so that the intramolecular cyclisation reaction undertaken by free L2–L5 10 will proceed readily even in the presence of a metal ion template.It is also noteworthy that all the complexes isolated contain metal ions that favour tetrahedral co-ordination, namely ZnII, CoII and CuI. It is therefore likely that the steric properties of L2–L5 prevent the formation of tetragonal complexes, or of coordination numbers >4. The preponderance of zinc(II) products in this work may reflect the relatively long metal–ligand bonds formed by ZnII,18 which minimise the unfavourable Zn ? ? ?H–C steric contacts. For CuI, the excellent p-donor capability of this electron-rich low-valent metal ion is rcome the extremely short Cu ? ? ? H–C contacts and concomitant ligand distortions observed in 141–161.Experimental Unless stated otherwise, all manipulations were performed in air using commercial grade solvents.Ferrocenecarbaldehyde, 1,2-diaminoethane, 1,2-diaminobenzene, 1,2-diamino-4-methylbenzene and all metal salts were used as supplied. 1,2- Bis(ferrocen-1-ylmethyleneamino)ethane (L1) was prepared by the literature method.9 Microanalytical, FAB mass spectrometric and UV/VIS data for the complexes in this study are listed in Tables 1 and 5. Syntheses [ZnCl2(L1)] 4. A mixture of L1 (0.45 g, 1.00 × 1023 mol) and ZnCl2 (0.14 g, 1.00 × 1023 mol) in CH3OH (30 cm3) was stirred at room temperature for 3 h.The resultant orange precipitate was filtered oV, air-dried and recrystallised from CH2Cl2–Et2O. Yield 0.39 g, 68%. NMR spectra (CDCl3, 293 K): 1H, d 8.37 (s, 2 H, N]] CH), 5.13 (t, J 1.8, 4 H, fc H2 and H5), 4.67 (t, J 1.8, 4 H, fc H3 and H4), 4.35 (s, 10 H, fc C5H5) and 3.78 (s, 4 H, CH2); 13C; d 170.3 (N]] CH), 74.3 (fc C1), 73.7 (fc C2 and C5), 71.6 (fc C3 and C4), 70.1 (fc C5H5) and 59.2 (CH2). [ZnCl2(L2)] 5. Ferrocenecarbaldehyde (0.43 g, 2.00 × 1023 mol), 1,2-diaminobenzene (0.11 g, 1.00 × 1023 mol) and ZnCl2 (0.14 g, 1.00 × 1023 mol) were refluxed in CH3OH (30 cm3) for 3 h, yielding a deep red solution.Concentration of the solution and addition of an excess of Et2O yielded a dark red solid which was recrystallised from CH2Cl2–Et2O. Yield 0.46 g, 72%. NMR spectra (CDCl3, 293 K): 1H, d 8.84 (s, 2 H, N]] CH), 7.53 (m, 2 H, Ph H3 and H6), 7.41 (m, 2 H, Ph H4 and H5), 5.31 (t, J 1.8, 4 H, fc H2 and H5), 4.86 (t, J 1.8, 4 H, fc H3 and H4) and 4.38 (s, 10 H, fc C5H5); 13C; d 165.5 (N]] CH), 140.4 (Ph C1 and C2), 128.5 (Ph C3 and C6), 117.5 (Ph C4 and C5), 76.0 (fc C1), 75.5 (fc C2 and C5), 72.6 (fc C3 and C4) and 70.3 (fc C5H5).[ZnCl2(L3)] 6. Method as for compound 5, using 1,2- diamino-4-methylbenzene (0.12 g, 1.00 × 1023 mol). The product formed dark red microcrystals from CH2Cl2–Et2O. Yield 0.25 g, 38%. NMR spectra (CDCl3, 293 K): 1H; d 8.81 (s, 1 H), 8.80 (s, 1 H, N]] CH), 7.32 (m, 3 H, Ph H3 1 H5 1 H6), 5.29 (br s, 4 H, fc H2 and H5), 4.82 (br s, 4 H, fc H3 and H4), 4.38 (s, 5 H), 4.37 (s, 5 H, fc C5H5) and 2.37 (s, 3 H, CH3); 13C, d 165.3, 164.5 (N]] CH), 140.0, 138.8, 138.1 (Ph C1, C2 and C4), 129.3 (Ph C5), 117.9, 117.2 (Ph C3 and C6), 76.0 (fc C1), 75.3, 75.2 (fc C2 and C5), 72.6, 72.4 (fc C3 and C4), 70.3, 70.2 (fc C5H5) and 21.5 (CH3).[ZnCl2(L4)] 7. Method as for compound 5, using 1,2- diamino-4-chlorobenzene (0.14 g, 1.00 × 1023 mol). The product formed a rose-red solid from CH2Cl2–Et2O.Yield 0.30 g, 45%. NMR spectra (CDCl3, 293 K): 1H, d 8.81 (s, 1 H), 8.77 (s, 1 H, N]] CH), 7.41 (m, 3 H, Ph H3 1 H5 1 H6), 5.33 (br s, 2 H), 5.29 (br s, 2 H, fc H2 and H5), 4.92 (t, J 1.8, 2 H), 4.88 (t, J 1.8, 2 H, fc H3 and H4), 4.41 (s, 5 H) and 4.38 (s, 5 H, fc C5H5); 13C,J. Chem. Soc., Dalton Trans., 1998, 3791–3799 3797 d 166.2, 165.7 (N]] CH), 141.1, 139.0 (Ph C1 and C2), 134.0 (Ph C4), 128.2 (Ph C5), 118.6, 117.5 (Ph C3 and C6), 76.1, 75.8 (fc C2 and C5), 72.8, 72.7 (fc C3 and C4), 70.5, 70.4 (fc C5H5).The peak from fc C1 was obscured. [ZnCl2(L5)] 8. Method as for compound 5, using 1,2- diamino-4-nitrobenzene (0.15 g, 1.00 × 1023 mol). The product formed a violet-red solid from CH2Cl2–Et2O. Yield 0.14 g, 21%. NMR spectrum (CDCl3, 293 K): 1H, d 8.94 (s, 1 H), 8.92 (s, 1 H, N]] CH), 8.41 (d, J 2.3, 1 H, Ph H3), 8.27 (dd, J 2.3 and 9.0, 1 H, Ph H5), 7.65 (d, J 9.0, 1 H, Ph H6), 5.37 (br s, 4 H, fc H2 and H5), 5.03 (br s, 4 H, fc H3 and H4), 4.45 (s, 5 H) and 4.44 (s, 5 H, fc C5H5).[ZnBr2(L2)] 9. Method as for compound 5, using ZnBr2 (0.23 g, 1.00 × 1023 mol). The product formed a deep red solid from CH2Cl2–Et2O. Yield 0.47 g, 65%. NMR spectra (CDCl3, 293 K): 1H, d 8.82 (s, 2 H, N]] CH), 7.53 (m, 2 H, Ph H3 and H6), 7.43 (m, 2 H, Ph H4 and H5), 5.33 (t, J 1.9, 4 H, fc H2 and H5), 4.87 (t, J 1.9, 4 H, fc H3 and H4), 4.40 (s, 10 H, fc C5H5); 13C, d 165.5 (N]] CH), 140.5 (Ph C1 and C2), 128.6 (Ph C3 and C6), 117.7 (Ph C4 and C5), 76.4 (fc C1), 75.4 (fc C2 and C5), 73.0 (fc C3 and C4) and 70.0 (fc C5H5).[ZnBr2(L3)] 10. Method as for compound 9, using 1,2- diamino-4-methylbenzene (0.12 g, 1.00 × 1023 mol). The product formed dark red microcrystals from CH2Cl2–Et2O. Yield 0.30 g, 40%. NMR spectra (CDCl3, 293 K): 1H, d 8.79 (s, 1 H), 8.77 (s, 1 H, N]] CH), 7.42 (m, 3 H, Ph H3 1 H5 1 H6), 5.31 (br s, 4 H, fc H2 and H5), 4.81 (br s, 4 H, fc H3 and H4), 4.38 (s, 5 H), 4.36 (s, 5 H, fc C5H5) and 2.43 (s, 3 H, CH3); 13C, d 165.4, 164.5 (N]] CH), 140.1, 138.8, 138.1 (Ph C1, C2 and C4), 129.4 (Ph C5), 118.1, 117.4 (Ph C3 and C6), 75.9 (fc C1), 75.3, 75.2 (fc C2 and C5), 73.1, 73.0 (fc C3 and C4), 70.3, 70.2 (fc C5H5) and 21.5 (CH3).[ZnBr2(L4)] 11. Method as for compound 9, using 1,2- diamino-4-chlorobenzene (0.14 g, 1.00 × 1023 mol). The product formed a dark red solid from CH2Cl2–Et2O. Yield 0.40 g, 53%. NMR spectra (CDCl3, 293 K): 1H, d 8.82 (s, 1 H), 8.71 (s, 1 H, N]] CH), 7.37 (m, 3 H, Ph H3 1 H5 1 H6), 5.35 (br s, 2 H), 5.29 (br s, 2 H, fc H2 and H5), 4.90 (t, J 1.8, 2 H), 4.85 (t, J 1.8, 2 H, fc H3 and H4), 4.41 (s, 5 H) and 4.38 (s, 5 H, fc C5H5); 13C, d 166.1, 165.9 (N]] CH), 141.0, 139.0 (Ph C1 and C2), 133.8 (Ph C4), 128.2 (Ph C5), 118.9, 117.6 (Ph C3 and C6), 76.4 (fc C1), 75.4 (fc C2 and C5), 73.0 (fc C3 and C4) and 70.0 (fc C5H5).[ZnBr2(L5)] 12. Method as for compound 9, using 1,2- diamino-4-nitrobenzene (0.15 g, 1.00 × 1023 mol).The product formed a violet-red solid from CH2Cl2–Et2O. Yield 0.30 g, 39%. NMR spectrum (CDCl3, 293 K): 1H, d 8.91 (s, 1 H), 8.89 (s, 1 H, N]] CH), 8.40 (d, J 2.3, 1 H, Ph H3), 8.27 (dd, J 2.3 and 9.0, 1 H, Ph H5), 7.65 (d, J 9.0, 1 H, Ph H6), 5.39 (br s, 4 H, fc H2 and H5), 5.03 (br s, 4 H, fc H3 and H4), 4.46 (s, 5 H) and 4.45 (s, 5 H, fc C5H5). [CoBr2(L3)] 13. Method as for compound 5, using 1,2- diamino-4-methylbenzene (0.12 g, 1.00 × 1023 mol) and CoBr2 (0.22 g, 1.00 × 1023 mol).The product formed a violet-red solid from CH2Cl2–Et2O. Yield 0.25 g, 34%. NMR spectrum (CDCl3, 293 K): 1H, d 19.7 (1 H, Ph H5), 6.8 (1 H), 6.5 (1 H, Ph H3 1 H6), 5.3 (3 H, CH3), 21.2 (5 H), 21.3 (5 H, fc C5H5), 22.7 (4 H, fc H3 and H4) and 221.9 (v br, ca. 4H, fc H2 and H5). [Cu(L1)2]PF6 14?PF6. A solution of L1 (0.45 g, 1.00 × 1023 mol) and [Cu(NCCH3)4]PF6 (0.19 g, 5.00 × 1024 mol) in CH3OH (30 cm3) was stirred at room temperature under N2 for 3 h.The orange solution was filtered, then concentrated to 3 cm3. Layering with Et2O aVorded orange microcrystals, which were recrystallised from CH3NO2–Et2O. Yield 0.48 g, 86%. NMR spectra (CD3NO2, 293 K): 1H, d 8.37 (s, 2 H, N]] CH), 5.13 (t, J 1.8, 4 H, fc H2 and H5), 4.67 (t, J 1.8, 4 H, fc H3 and H4), 4.35 (s, 10 H, fc C5H5), 3.78 (s, 4 H, CH2); 13C, d 165.6 (N]] CH), 79.9 (fc C1), 74.2 (fc C2 and C5), 70.4 (fc C3, C4 and C5H5) and 54.9 (CH2).[Cu(L2)2]BF4 15?BF4. Method A. Ferrocenecarbaldehyde (0.43 g, 2.00 × 1023 mol), 1,2-diaminobenzene (0.11 g, 1.00 × 1023 mol) and Cu(BF4)2?xH2O (0.18 g, 5.00 × 1024 mol) were refluxed in CH3OH (30 cm3) for 3 h, yielding a deep red solution. Concentration of the solution and addition of an excess of Et2O yielded a dark red solid which was recrystallised from CH3NO2–Et2O. Yield 0.29 g, 50%. Method B. As for method A, using [Cu(NCCH3)4]BF4 (0.16 g, 5.00 × 1024 mol). Yield 0.39 g, 68%.NMR spectra (CDCl3, 293 K): 1H, d 8.47 (s, 4 H, N]] CH), 7.46 (m, 8 H, Ph H3–H6), 4.53 (t, J 1.8, 8 H, fc H2 and H5), 4.27 (t, 1.8 Hz, 8 H, fc H3 and H4) and 4.07 (s, 20 H, fc C5H5); 13C, d 161.7 (N]] CH), 143.8 (Ph C1 and C2), 128.3 (Ph C3 and C6), 118.6 (Ph C4 and C5), 78.4 (fc C1), 72.7 (fc C2 and C5), 69.8 (fc C3 and C4) and 69.6 (fc C5H5). [Cu(L3)2]BF4 16?BF4. Method as for compound 15?BF4, using 1,2-diamino-4-methylbenzene (0.12 g, 1.00 × 1023 mol). The product formed dark red microcrystals from CH3NO2–Et2O.Yields: method A, 0.30 g, 51%; B, 0.40 g, 68%. NMR spectra (CDCl3, 293 K): 1H, d 8.43 (s, 2 H), 8.41 (s, 2 H, N]] CH), 7.27 (m, 6 H, Ph H3, H5 and H6), 4.52 (t, J 2.0, 8 H, fc H2 and H5), 4.27 (t, J 2.0, 8 H, fc H3 and H4), 4.08 (s, 10 H), 4.06 (s, 10 H, fc C5H5) and 2.54 (s, 6H, CH3); 13C; d 161.3, 160.6 (N]] CH), 143.5, 141.4, 138.5 (Ph C1, C2 and C4), 118.8, 118.2 (Ph C5 and C6), 78.4 (fc C1), 75.5 (fc C2 and C5), 72.6 (fc C3 and C4), 70.3 (fc C5H5) and 21.4 (CH3).[Cu(L3)2]PF6 16?PF6. Method as for compound 16?BF4, using [Cu(NCCH3)4]BF4 (0.19 g, 5.00 × 1024 mol). Yield: 0.44 g, 74%. Single crystal structure determinations Single crystals of [ZnCl2(L1)] 4 and [Cu(L3)2]PF6 16?PF6 were grown from CH2Cl2–hexanes. Crystals of [1HH]BF4 were obtained by vapour diVusion of Et2O into a MeCN solution of the crude solid obtained from the reaction of Zn(BF4)2.6H2O, fcCHO and 1,2-diaminobenzene in MeOH; the microanalysis of this product is given in Table 1.Experimental details from the structure determinations are given in Table 9. Intensity data for compounds 4 and [1HH]BF4 were collected on a Stoe IPDS diVractometer (Mo-Ka, l = 0.71073 Å), for 16?PF6?1.7CH2Cl2 on a Rigaku RAXIS IIc diVractometer (Mo-Ka, l = 0.71069 Å). All structures were solved by direct methods31 and refined by full matrix least squares on F2.32 [ZnCl2(L1)] 4. During refinement, high thermal parameters at carbon for both [C5H5]2 units in the molecule were suggestive of librational disorder in these groups.One of these was successfully modelled over two orientations C41–C45 and C412–- C452 with a 65 : 35 occupancy ratio. After isotropic refinement of all non-H atoms, an empirical absorption correction 33 was applied to the data. In the final cycles of refinement all full occupancy non-hydrogen atoms were assigned anisotropic displacement parameters. [Cu(L3)2]PF6?1.7CH2Cl2 16?PF6?1.7CH2Cl2.During refinement, the methyl substituent of ligand ‘A’ was found to be disordered equally over two sites, C(7A) and C(8A). No disorder in the [C5H5]2 rings or PF6 2 anion was detected. However, the CH2Cl2 solvent was disordered over a substantial fraction3798 J. Chem. Soc., Dalton Trans., 1998, 3791–3799 Table 9 Experimental details for the single crystal structure determinations Formula Mr Crystal class Space group a/Å b/Å c/Å b/8 U/Å3 Z m(Mo-Ka)/mm21 T/K Measured reflections Independent reflections Rint R(F) wR(F2) Goodness of fit Flack parameter 4 C24H24Cl2Fe2N2Zn 588.42 Orthorhombic P212121 (no. 19) 11.297(2) 14.614(3) 14.625(3) — 2414.5(8) 4 2.406 293(2) 4174 4174 — 0.035 0.087 1.046 0.00(2) 16?PF6?1.7CH2Cl2 C59.7H55.4Cl3.4CuF6Fe4N4P 1381.32 Monoclinic P21/c (no. 14) 17.091(5) 20.019(5) 17.797(5) 95.67(4) 6059(3) 4 1.518 180(2) 12736 7409 0.112 0.075 0.167 0.947 — [1HH]BF4 C28H25BF4Fe2N2 588.01 Monoclinic P21/c (no. 14) 10.966(2) 15.935(3) 14.829(3) 108.45(3) 2458.1(9) 4 1.232 293(2) 8511 3770 0.074 0.061 0.175 1.032 — of the asymmetric unit.This was modelled using 6 diVerent molecules with occupancies of 0.1–0.5, 3 of which showed disorder in their Cl atoms; all C–Cl distances were fixed to a common value, which refined to 1.69(2) Å. All wholly occupied non-H atoms were refined anisotropically. [1HH]BF4. During refinement high thermal parameters at carbon for one of the [C5H5]2 moieties in the molecule were suggestive of librational disorder in this group.This was modelled over two orientations C41–C45 and C412–C452 with a 80 : 20 occupancy ratio. After isotropic refinement of all non-H atoms, an empirical absorption correction 34 was applied to the data. In the final cycles of refinement all full occupancy non-H atoms were assigned anisotropic displacement parameters. The benzimidazolium atom H(2) was located during refinement, and allowed to refine freely. All other H atoms were placed in calculated positions.CCDC reference number 186/1185. See http://www.rsc.org/suppdata/dt/1998/3791/ for crystallographic files in .cif format. Other measurements Infrared spectra were obtained as Nujol mulls between 4000 and 400 cm21 using a Perkin-Elmer Paragon 1000 spectrophotometer, UV/VIS spectra with a Perkin-Elmer Lambda 12 spectrophotometer operating between 1100 and 200 nm in 1 cm quartz cells, NMR spectra on a Bruker DPX250 spectrometer, operating at 250.1 (1H) and 62.9 MHz (13C) and fast atom bombardment mass spectra on a Kratos MS890 spectrometer, employing a 3-nitrobenzyl alcohol matrix.The CHN microanalyses were performed by the University of Cambridge Department of Chemistry microanalytical service. Electrochemical measurements were carried out using an Autolab PGSTAT20 voltammetric analyser, in CH2Cl2 or MeCN containing 0.1 or 0.5 M NBun 4PF6, respectively, as supporting electrolyte. Voltammetric experiments involved the use of a double platinum working/counter electrode and a silver wire reference electrode; all potentials are referenced to an internal ferrocene–- ferrocenium standard and were obtained at a scan rate of 100 mV s21.Acknowledgements The authors thank the Royal Society (M. A. H.), the Committee of Vice-Chancellors and Principles and the Cambridge Commonwealth Trust (P. L.), the University of Cambridge and St. Catharine9s College for financial support. References 1 A.Togni and T. Hayashi (Editors), Ferrocenes. Homogeneous Catalysis. Organic Synthesis. Materials Science, VCH, Weinheim, 1995. 2 P. D. Beer, Adv. Inorg. Chem., 1992, 39, 79; P. D. Beer and D. K. Smith, Prog. Inorg. Chem., 1997, 46, 1. 3 J. S. Miller and A. J. Epstein, Angew. Chem., Int. Ed. Engl., 1994, 33, 385. 4 N. J. Long, Angew. Chem., Int. Ed. Engl., 1995, 34, 21; S. R. Marder, in Inorganic Materials, eds. D. W. Bruce and D. O’Hare, Wiley, Chichester, 1996, ch. 3, pp. 121–169. 5 A. Houlton, N. Jasim, R. M. G. Roberts, J. Silver, D. Cunningham, P. McArdle and T. Higgins, J. Chem. Soc., Dalton Trans., 1992, 2235; A. Houlton, J. R. Miller, J. Silver, N. Jasim, M. J. Ahmet, T. L. Axon, D. Bloor and G. H. Gross, Inorg. Chim. Acta, 1993, 205, 67; J. Silver, J. R. Miller, A. Houlton and M. J. Ahmet, J. Chem. Soc., Dalton Trans., 1994, 3355; R. Bosque, C. Lopez, J. Sales, X. Solans and M. Font-Bardia, J. Chem. Soc., Dalton Trans., 1994, 735; Y. J. Wu, S.Q. Huo and Y. Zhu, J. Organomet. Chem., 1995, 485, 161; S. Q. Huo and Y. J. Wu, J. Organomet. Chem., 1995, 490, 243; C. Lopez, R. Bosque, X. Solans and M. Font-Bardia, New J. Chem., 1996, 20, 1285. 6 C. Lopez, J. Sales, X. Solans and J. Reedijk, J. Chem. Soc., Dalton Trans., 1992, 2321; J. Blanco, E. Gayoso, J. M. Vila, M. Gayoso, C. Maichle-Mossmer and J. Strahle, Z. Naturforsch., Teil B, 1993, 48, 906; S. Q. Huo, Y. J. Wu, Y. Zhu and L. Yang, J. Organomet. Chem., 1994, 470, 17; Y.J. Wu, S. Q. Huo, Y. Zhu and L. Yang, J. Organomet. Chem., 1994, 481, 235; R. Bosque, C. Lopez, J. Sales and X. Solans, J. Organomet. Chem., 1994, 483, 61; C. Lopez, R. Bosque, X. Solans, M. Font-Bardia, D. Tramuns, G. Fern and J. Silver, J. Chem. Soc., Dalton Trans., 1994, 3039; Y. Yoshida, K. Onitsuka and K. Sonogashira, J. Organomet. Chem., 1996, 511, 47. 7 M. J. L. Tendero, A. Benito, J. M. Lloris, R. Martínez-Mañez, J. Soto, J. Payá, A. J. Edwards and P. R. Raithby, Inorg.Chim. Acta, 1996, 247, 139; M. J. L. Tendero, A. Benito, R. Martínez-Mañez, J. Soto, J. Payá, A. J. Edwards and P. R. Raithby, J. Chem. Soc., Dalton Trans., 1996, 343. 8 E. W. Nause, M. G. Meirim and N. F. Blom, Organometallics, 1988, 7, 2562. 9 A. Benito, J. Cano, R. Martínez-Máñez, J. Soto, J. Payá, F. Lloret, M. Julve, J. Faus and M. D. Marcos, Inorg. Chem., 1993, 32, 1197. 10 A. Benito, R. Martínez-Máñez, J. Payá, J. Soto, M. J. L. Tendero and E. Sinn, J. Organomet.Chem., 1995, 503, 259. 11 J. Cano, A. Benito, R. Martínez-Máñez, J. Soto, J. Payá, F. Lloret, M. Julve, M. D. Marcos and E. Sinn, Inorg. Chim. Acta, 1995, 231, 45. 12 R. J. Less, J. L. M. Wicks, N. P. Chatterton, M. J. Dewey, N. L. Cromhout, M. A. Halcrow and J. E. Davies, J. Chem. Soc., Dalton Trans., 1996, 4055. 13 C. Li, X. Peng and X.-Z. You, Synth. React. Org. Metal-Org. Chem., 1990, 20, 1231; C. Li, X.-Z. You, X. Peng, C.-Y. Zhao and H.-Y. Wu, Transition Met. Chem., 1995, 20, 300. 14 D. Freiesleben, K. Polborn, C. Robl, K. Sünkel and W. Beck, Can. J. Chem., 1995, 73, 1164.J. Chem. Soc., Dalton Trans., 1998, 3791–3799 3799 15 T. Kawamoto and Y. Kushi, J. Chem. Soc., Dalton Trans., 1992, 3137; I. Nagasawa, T. Kawamoto, H. Kuma and Y. Kushi, Bull. Chem. Soc. Jpn., 1998, 71, 1337. 16 P. J. Palmer, R. B. Trigg and J. V. Warrington, J. Med. Chem., 1971, 14, 248. 17 R. H. Holm, A. Chakravorty and G. O. Dudek, J. Am. Chem. Soc., 1964, 86, 379. 18 A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson and R. Taylor, J. Chem. Soc., Dalton Trans., 1989, S1. 19 E. Müller, C. Piguet, G. Bernardinelli and A. F. Williams, Inorg. Chem., 1988, 27, 849; D. A. Bardwell, A. M. W. Cargill Thompson, J. C. JeVrey, E. E. M. Tilley and M. D. Ward, J. Chem. Soc., Dalton Trans., 1995, 835; S. M. Scott, K. C. Gordon and A. K. Burrell, Inorg. Chem., 1996, 35, 2452. 20 M. A. Halcrow, N. L. Cromhout and P. R. Raithby, Polyhedron, 1997, 16, 4257 and refs. therein. 21 Y. S. Sohn, D. N. Hendrickson and H. B. Gray, J. Am. Chem. Soc., 1971, 93, 3603. 22 S. Toma, A. Gáplovsky, M. Hudecek and Z. Langfelderová, Monatsh. Chem., 1985, 116, 357; M. M. Bhadbhade, A. Das, J. C. JeVrey, J. A. McCleverty, J. A. Navas Badiola and M. D. Ward, J. Chem. Soc., Dalton Trans., 1995, 2769. 23 C. Reichardt, Angew. Chem., Int. Ed. Engl., 1965, 4, 29. 24 A. B. P. Lever, Inorganic Electronic Spectroscopy, 2nd edn., Elsevier, Amsterdam, 1984. 25 D. R. Scott and R. S. Becker, J. Chem. Phys., 1961, 35, 516. 26 S. Kitagawa and M. Munakata, Inorg. Chem., 1981, 20, 2261; M. A. Masood and P. S. Zacharias, J. Chem. Soc., Dalton Trans., 1991, 111. 27 Z. Yu, Y. Zhou, W. Yang, Y. Tian, C.-Y. Duan, R. Liu and X. You, Chem. Lett., 1996, 957. 28 M. D. Ward, Chem. Soc. Rev., 1995, 24, 121. 29 P. Federlin, J.-M. Kern, A. Rastegar, C. Dietrich-Buchecker, P. A. Marnot and J.-P. Sauvage, New J. Chem., 1980, 14, 9. 30 E. C. Constable, Metals and Ligand Reactivity, 2nd edn., VCH, Weinheim, 1996. 31 G. M. Sheldrick, SHELXTL PLUS, PC version, Siemens Analytical X-Ray Instruments Inc., Madison, WI, 1990. 32 G. M. Sheldrick, SHELXL 97, University of Göttingen, 1997. 33 S. Parkin, B. Moezzi and H. Hope, J. Appl. Crystallogr., 1995, 28, 53. 34 N. Walker and D. Stuart, Acta Crystallogr., Sect. A, 1983, 39, 158. Paper 8/06657D

ISSN:1477-9226    

DOI:10.1039/a806657d

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3801–3804 3801 A Ï-hydroxy-Ï-nitritodicopper(II) core embedded in a AgNO2 matrix: synthesis, structure and magnetism Ulrich Flörke,a Hans-Jürgen Haupt a and Phalguni Chaudhuri *b a Anorganische und Analytische Chemie, Universität-Gesamthochschule Paderborn, D-33098 Paderborn, Germany b Max-Planck-Institut für Strahlenchemie, Stiftstraße 34-36, D-45470 Mülheim an der Ruhr, Germany. E-mail: Chaudh@mpi-muelheim.mpg.de Received 2nd July 1998, Accepted 29th September 1998 The copper complex [L2Cu2(m-NO2)(m-OH)Ag2(NO2)4] (L = 1,4,7-trimethyl-1,4,7-triazacyclononane) has been synthesized and characterized by X-ray crystallography and from temperature-dependent susceptibility measurements which established a fairly strong, intramolecular, antiferromagnetic exchange coupling (J = 2193 cm21, H = 22JS1?S2, S1 = S2 = 1/2) and the 1,2-co-ordination mode of the nitrito bridge to be an extremely good mediator of spin coupling.Introduction The co-ordination mode of the nitrite ion is unusual, because of the number of diVerent ways in which it can bind to a metal.1–15 Towards CuII it behaves as a monodentate ligand via either nitrogen or oxygen; it can also act either in a symmetrical end-to-end m-1,3 ONO-bridging mode, the unsymmetrical twoatom O,N mode or via a single oxygen atom. A few examples of tridentate behaviour are also known. Although there is a long history and a prolific literature on the subject of exchange coupling in binuclear copper(II) complexes,16 unsymmetrical dibridged systems, particularly those containing nitrite ions, are, however, not well documented.2,7 So it is of interest to learn how nitrite reacts with binuclear copper species particularly from the viewpoints of structure and exchange coupling.Moreover, Cu–NOx complexes are relevant in the study of copper-containing enzymes involved in the denitrification process,17 i.e. the bacterial dissimilatory reduction of nitrate and nitrite to NO, N2O and N2.In particular, copper–nitrite complexes are potentially relevant to the nitrite reductases which convert NO2 2 into NO and/or N2O, reactions presumed to proceed by nitrite binding to copper(II). With the above points in mind we prepared the copper complex [L2Cu2(m-NO2)(m-OH)Ag2(NO2)4] 1 (L = 1,4,7-trimethyl- 1,4,7-triazacyclononane), which contains one two-atom O,N-bridging nitrite and one hydroxo group and can be considered as a binuclear copper(II) complex, magnetically diluted with Ag1 and NO2 2 ions.Here we report the synthesis, crystal structure and the exchange coupling constant determined from the temperature-dependent magnetic susceptibility data of 1. In this connection 1 will also be compared with another copper(II) complex,7 in which the metal ions are also unsymmetrically dibridged but via an alkoxide and a nitrito ligand. Experimental The ligand 1,4,7-trimethyl-1,4,7-triazacyclononane (L) was prepared according to a published procedure.18 All other reagents were used as received.Microanalyses were performed by Mikroanalytisches Laboratorium Dornis and Kolbe, Mülheim an der Ruhr. Copper was determined gravimetrically by using N-benzoyl-N-phenylhydroxylamine. Fourier transform IR spectroscopy on KBr pellets was performed on a Perkin-Elmer 2000 FT-IR instrument. Magnetic susceptibilities of powdered samples were recorded on a SQUID magnetometer (MPMS, Quantum Design) in the temperature range 2–290 K with an applied field of 1 T.Experimental susceptibility data were corrected for the underlying diamagnetism using Pascal’s constants. Mass spectra were recorded in the electrospray-ionization (ESI) mode in acetonitrile using a V 8200 spectrometer. Preparation of [L2CuII 2(Ï-NO2)(Ï-OH)Ag2(NO2)4] 1 1,4,7-Trimethyl-1,4,7-triazacyclononane (0.51 g, 3 mmol) was added to a solution of CuCl2?2H2O (0.40 g, 2.4 mmol) in distilled methanol (80 ml).The solution was stirred at room temperature for 15 min. To the resulting deep green solution AgNO2 (1.85 g, 12 mmol) was added and the suspension stirred for 2 h. The precipitated AgCl and the residual AgNO2 were filtered oV and the deep green filtrate was concentrated to about 50 ml, when the crystallization started. The suspension was cooled to 4 8C for 2 h and the green crystals were collected by filtration and air-dried. Yield: 0.27 g (24%) (Found: C, 23.2; H, 4.7; Ag, 23.1; Cu, 23.4; N, 16.6.Calc. for C18H43Ag2- Cu2N11O11: C, 23.19; H, 4.65; Ag, 23.14; Cu, 23.19; N, 16.52%). IR(KBr, cm21): 3500–3400(br)m, 1465s,m, 1270s and 1009s,m. UV-VIS in CH3CN: l/nm (e/M21 cm21) 635(154) and 1000(58.6). ESI-MS(CH3CN): m/z (relative intensity, %) 934 (1), 914 (1), 606 (100), 560 (40), 514 (38) and 280 (65). Crystal structure determination The crystallographic data for complex 1 are summarized in Table 1. Graphite monochromated Mo-Ka X-radiation was used.Intensity data collected at 293(2) K were corrected for Lorentz-polarization and absorption eVects (y scans) in the usual manner. The structures were solved by direct methods by using the Siemens SHELXTL-V5 package.19 The function minimized during full-matrix least-squares refinement was Sw(|Fo| 2 |Fc|)2. The hydrogen atoms were placed at calculated positions with isotropic thermal parameters and refined with a riding model. All non-hydrogen atoms were refined with anisotropic thermal parameters.CCDC reference number 186/1181. See http://www.rsc.org/suppdata/dt/1998/3801/ for crystallographic files in .cif format.3802 J. Chem. Soc., Dalton Trans., 1998, 3801–3804 Results and discussion That the reaction of CuCl2?2H2O with the macrocyclic amine ligand L in methanol aVords the dichloride salt of the cation [Cu2(m-Cl)2L2]21 is known from our earlier work.20 Treatment of the mentioned dication with AgNO2 aVords complex 1, which is presumably formed by substitution of a bridging hydroxo group21 in [Cu2(m-OH)2L2]21 by a nitrite ion.Complex 1 can be considered as a cluster formed through the interaction of the ions [Cu2(NO2)(OH)L2]21 and [Ag2(NO2)4]22. It is interesting in this connection that the azide analogue of the cation in 1 contains a symmetrical 1,1-azide bridging mode.20 Besides the bands belonging to the macrocyclic amine L in the IR spectrum complex 1 exhibits a broad, medium-intense band in the region 3500 cm21, which can be assigned to n(OH).The sharp bands at 1465 and 1009 cm21 are due to n(N]] O) and n(N–O), respectively, indicating the m(NO) bridging mode of the nitrito ligand, whereas the strong band at 1270 cm21 is assignable to the nsym(NO2) mode showing a three-co-ordinate nitrite group. A sharp but weak band is also observed at 893 cm21, which may be due to the wagging mode d(ONO) vibration. Thus the IR spectrum clearly excludes the monodentate bonding via a single oxygen atom of the nitrite ion 1 in 1.The ESI mass spectrum of complex 1 was recorded in acetonitrile solution. The indication that 1 contains silver atoms was first obtained from the ESI-MS data. The isotope pattern of the signals characterizes the number of silver atoms in the masses. Selected signals (m/z) in the ESI-MS of 1, including the assignments, are 934[L2Cu2(OH)(NO2)5Ag2], 914[L2Cu2(NO2)5Ag2], 606[L2Cu2(NO2)3, 100], 560[L2Cu2- (NO2)2, 40], 514[L2Cu2NO2, 38] and 280[LCuNO, 65%].These data are in accord with the structure determination described below. The electronic spectrum of complex 1 in acetonitrile is consistent with the square-pyramidal CuN3O2 and CuN4O chromophores since it displays two weak d–d transitions in the visible region at 635 (e = 154) and at 1000 nm (59 M21 cm21) in agreement with a (dx2 2 y2) 1 ground state of the copper(II) ions. Molecular structure of complex 1 A view of the asymmetric unit together with the atomnumbering scheme is given in Fig. 1(a). The structure deter- Table 1 Crystallographic data for [L2Cu2(NO2)(OH)Ag2(NO2)4] 1 Formula Formula weight Crystal size/mm Crystal system Space group a/Å b/Å c/Å V/Å3 Z Dc/g cm23 DiVractometer l(Mo-Ka)/Å m/mm21 F(000) q Range/8 Index ranges Reflections collected Maximum, minimum transmission Data/restraints/parameters R1,a wR [I > 2s(I)] (all data) C18H43Ag2Cu2N11O11 932.44 0.44 × 0.15 × 0.13 Orthorhombic P212121 8.726(2) 17.216(3) 21.631(4) 3249.6(11) 4 1.904 Siemens R3m/V 0.71073 2.546 1868 2.22–27.56 211 £ h £ 0, 0 £ k £ 22, 0 £ l £ 28 4215 0.903, 0.344 4213/0/406 0.0716, 0.1586 0.1382, 0.2012 a R1 = S(|Fo| 2 |Fc|)/S|Fo|.mination and hence the derived geometrical parameters suVer from rather bad crystal quality and disordered nitrito groups N(7)O2 and N(10)O2. The disorder is indicated by large and elongated anisotropic displacement parameters of the oxygen atoms O(72) and O(101).Refinement with a split model did not lead to any improvement. Selected bond lengths and angles for 1 are listed in Table 2. The packing consists of dinuclear copper moieties Cu2L2- (m-NO2)(m-OH) and infinite zigzag chains of [Ag(NO2)]x along [100]. The Cu atoms Cu(1) and Cu(2) are linked through a symmetric m-OH bridging ligand with Cu–O(1) bond lengths of 1.928(10) Å and an unsymmetric bridging O,N nitrito ligand with Cu(1)–O(71) 2.062(10), O(71)–N(7) 1.25(2) and Cu(2)–N(7) 2.07(2) Å.Similar geometries of the Cu2 core with bridging m-OR and m-NO2 ligands are known for complexes [Cu2(bipy)2(m-OCH3)(m-NO2)][NO2]2 2 and [Cu2(L-Et)- (m-NO2)][ClO4]2 3, where “bipy” stands for 2,29-bipyridine and “HL-Et” for a binucleating ligand N,N,N9,N9-tetrakis(1-ethylbenzimidazol- 2-yl)-2-hydroxy-1,3-diaminopropane. It is of interest that the Cu ? ? ? Cu separation of 3.299(2) Å in 1 is shorter than that in 2, 3.403(1) Å, and 3, 3.325(2) Å. Two tridentate amine ligands L complete the distorted square-pyramidal co-ordination sphere of the two copper(II) centres.The basal planes, defined by Cu(1)O(1)O(71)N(1)N(2) and Cu(2)O(1)N(7)N(5)N(6), set up a dihedral angle of 408. The average basal Cu–Namine bonds which are trans to the Fig. 1 Views of (a) the aymmetric unit in complex 1, (b) the Ag(NO2)x chains along [100].J. Chem. Soc., Dalton Trans., 1998, 3801–3804 3803 hydroxo and nitrito ligands are shorter (by 0.16 to 0.20 Å) than the apical Cu–Namine bonds as is always observed in squarepyramidal copper(II) complexes containing the macrocyclic ligand L.22 Interestingly, the vacant sixth co-ordination site trans to the apical Cu–Namine bond is occupied either by one oxygen atom, Cu(1) ? ? ? O(81) 2.45 Å, or one nitrogen atom, Cu(2) ? ? ? N(8) 2.71 Å, of a AgNO2 group. These intermolecular contacts are rather long, indicating that they are very weak. Thus a square-pyramidal geometry rather than a distorted octahedral co-ordination for both copper centres is a rather appropriate description for 1.Complex 1 can thus be envisaged as a dinuclear copper(II) complex embedded in a matrix of diamagnetic AgNO2, as is described below and confirmed by the magnetic susceptibility study. Silver atom Ag(1) is fivefold co-ordinated in an irregular arrangement by one nitrogen and four oxygen atoms. The nitrito anions act as bidentate ligands through their two oxygen atoms. The resulting Ag–O contact lengths for Ag(1) range from 2.392(13) to 2.520(12) Å for the oxygen atoms numbered 8 and 9 and those for Ag(2) from 2.43(2) to 2.73(2) for O atoms numbered 10 and 11.A nitrogen atom from a third NO2 group then occupies the remaining co-ordination site with bond lengths of Ag(1)–N(10) 2.32(2) and Ag(2)–N(11) 2.49(3) Å. The infinite AgNO2 zigzag chains [Fig. 1(b)] are solely formed by Ag(2) and the nitrito group 11, and the symmetry related equivalents to give the repetitive arrangement –Ag(O2N)– Ag(O2N)Ag– with a Ag–Ag–Ag angle of 115.6(5)8. Additionally, each Ag(2) atom is linked via the O2N(10) anion to a terminal Ag(1)(O2N)2 unit including nitrito groups 8 and 9.Of these, group 8 takes part in the above described co-ordination pattern of the copper centres. There are few other reported structures 23 containing nitrito-co-ordinated silver atoms. Most of them comprise isolated Ag(NO2) moieties. A more complex structure with NO2-co-ordinated Ag and Co atoms has been described recently.24 A somewhat related arrangement of infinite Pb–NO2 chains, but with ninefold co-ordination of Table 2 Selected bond lengths (Å) and angles (8) for complex 1 Cu(1)–O(1) Cu(1)–O(71) Cu(1)–N(1) Cu(1)–N(2) Cu(1)–N(3) Ag(1)–N(10) Ag(1)–O(82) Ag(1)–O(91) Ag(1)–O(92) Ag(1)–O(81) Ag(2)–O(101) Ag(2)–N(11) Ag(2)–O(102) O(1)–Cu(1)–O(71) O(1)–Cu(1)–N(1) O(71)–Cu(1)–N(1) O(1)–Cu(1)–N(2) O(71)–Cu(1)–N(2) N(1)–Cu(1)–N(2) O(1)–Cu(1)–N(3) O(71)–Cu(1)–N(3) N(1)–Cu(1)–N(3) N(2)–Cu(1)–N(3) N(10)–Ag(1)–O(82) N(10)–Ag(1)–O(91) O(82)–Ag(1)–O(91) N(10)–Ag(1)–O(92) O(82)–Ag(1)–O(92) O(91)–Ag(1)–O(92) N(10)–Ag(1)–O(81) 1.928(10) 2.062(10) 2.078(13) 2.108(13) 2.257(12) 2.32(2) 2.392(13) 2.40(2) 2.43(2) 2.520(12) 2.43(2) 2.49(3) 2.56(2) 86.2(4) 173.3(4) 90.9(5) 99.1(5) 173.9(5) 84.2(5) 103.5(4) 92.6(4) 82.6(4) 83.2(5) 109.6(6) 109.8(8) 121.1(7) 100.4(9) 149.6(8) 48.5(9) 138.8(7) Cu(2)–O(1) Cu(2)–N(5) Cu(2)–N(7) Cu(2)–N(6) Cu(2)–N(4) N(8)–O(81) N(8)–O(82) N(9)–O(92) N(9)–O(91) N(10)–O(101) N(10)–O(102) N(11)–O(112) N(11)–O(111) N(7)–O(71) N(7)–O(72) O(1)–Cu(2)–N(5) O(1)–Cu(2)–N(7) N(5)–Cu(2)–N(7) O(1)–Cu(2)–N(6) N(5)–Cu(2)–N(6) N(7)–Cu(2)–N(6) O(1)–Cu(2)–N(4) N(5)–Cu(2)–N(4) N(7)–Cu(2)–N(4) N(6)–Cu(2)–N(4) O(82)–Ag(1)–O(81) O(91)–Ag(1)–O(81) O(92)–Ag(1)–O(81) O(101)–Ag(2)–N(11) O(101)–Ag(2)–O(102) N(11)–Ag(2)–O(102) 1.927(9) 2.064(14) 2.070(15) 2.089(11) 2.275(13) 1.22(2) 1.22(2) 1.18(2) 1.18(2) 1.16(3) 1.19(2) 1.03(3) 1.07(2) 1.25(2) 1.26(2) 96.5(5) 83.9(5) 175.2(7) 173.5(4) 83.6(5) 95.4(5) 104.3(4) 80.8(7) 103.7(7) 82.1(5) 49.1(4) 111.3(7) 104.0(8) 151.9(7) 47.7(7) 104.2(7) the metal nitrito and nitrate groups, was reported for K2Pb(NO2)3(NO3)?H2O.4 Magnetic susceptibility study Magnetic susceptibility data for a polycrystalline sample of complex 1 were collected in the temperature range 2–290 K in order to characterize the nature and magnitude of the exchange interaction propagated by the bridging 1,2-m-nitrito ligand.We use the Heisenberg spin Hamiltonian in the form H = 22JS1?S2 for an isotropic exchange coupling with S1 = S2 = 1/2. The cryomagnetic property of 1 is shown in Fig. 2 in the form of a meff vs. T plot. The magnetic moment of 1.79 mB at 290 K decreases very rapidly with decreasing temperature, reaching a value of 0.35 mB at 70 K which remains nearly constant until 15 K with a value of 0.30 mB and then it starts to decrease reaching a value of 0.23 mB at 2 K.This magnetic behaviour is quite characteristic of an antiferromagnetic coupling between the paramagnetic copper(II) centres. The experimental magnetic data were simulated using a least-squares fitting computer program with a full-matrix diagonalization approach including exchange coupling, Zeeman splitting and the temperature independent paramagnetism (TIP) for the copper(II) ion. The least-squares fitting, shown as the solid line in Fig. 2, of the experimental data leads to J = 2193 cm21 and g = 2.28. A paramagnetic impurity (3.3%) with an S = 1/2 had also to be considered to account for the residual moment of ª0.30 mB at low temperatures. Thus the copper centres in 1 are antiferromagnetically exchange coupled with a singlet ground state, and the triplet state lying 386 cm21 above the ground state. Previously we reported in a short communication the structural and preliminary susceptibility (93–293 K) results of the dinuclear copper(II) complex [Cu2(m-OH)2L2][ClO4]2.21 For the purpose of comparison low-temperature (2–290 K) susceptibility data for the above bis(m-hydroxo)dicopper(II) complex have been obtained. The exchange coupling constant J = 250.2 cm21 and g = 2.286 have been evaluated from the new measurement, indicating that the antiferromagnetic coupling through the nitrito group is strong in 1.Since the observed J value is composed of ferromagnetic and antiferromagnetic contributions and these are in turn influenced by a variety of structural and electronic factors,25 it is diYcult to give a definite explanation for the relatively strong coupling in complex 1, although some pertinent observations and partial rationalization will be attempted. Complex 1 can be viewed as a dinuclear square-pyramidal d9 copper(II) species where the magnetic dx2 2 y2 orbitals are intermolecularly bridged by one hydroxo and one 1,2-m-nitrito group.The electronic spectrum of 1 in CH3CN is also in complete agreement with the (dx2 2 y2) 1 ground state orbitals.It has been pointed out by several authors 7,26,27 that the extent of Fig. 2 Plot of meff vs. T for complex 1. The solid line represents the simulation with the spin Hamiltonian H = 22JS1?S2.3804 J. Chem. Soc., Dalton Trans., 1998, 3801–3804 superexchange interaction through two diVerent bridging ligands in dinuclear copper(II) complexes can be rationalized on the basis of the concept of ligand orbital complementarity.Inspection of the ligand HOMOs reveals complementarity; as a consequence hydroxide and nitrite ligands provide orbital interaction pathways that act in concert to make the antisymmetric (fA) significantly diVerent in energy from symmetric (fS) combinations of metal and ligand orbitals, thus resulting in strong antiferromagnetism. It is interesting that for a similar m-alkoxo-m-1,2-nitrito copper(II) complex [Cu2(LEt)( NO2)]21 3,7 having comparable large coupling (J = 2139 cm21) but via dz2 magnetic orbitals, the Cu–O–Cu angle is larger (127.1o) than that in 1 (117.68).Hence, we are prone to recognize that the present co-ordination mode of the nitrito bridge is an extremely good mediator of antiferromagnetic coupling. Similar strong antiferromagnetic coupling between copper(II) sites via N,O bridging has been documented before in oximato species.22b Acknowledgements We acknowledge the support of this work by the Fonds der Chemischen Industrie.Our thanks are also due to the skilful technical assistance of Frau D. Kreft and Herr U. Pieper. References 1 M. A. Hitchman and G. L. Rowbottom, Coord. Chem. Rev., 1982, 42, 55. 2 A. Camus, N. Marsich and G. Nardin, Acta Crystallogr., Sect. B, 1977, 33, 1669. 3 F. S. Stephens, J. Chem. Soc. A, 1969, 2081. 4 M. Nardelli and G. Pelizzi, Inorg. Chim. Acta, 1980, 38, 15. 5 A. Gleizes, A. Meyer, M. A. Hitchman and O. Kahn, Inorg. Chem., 1982, 21, 2257. 6 B. J. Hathaway, Struct. Bonding, (Berlin), 1984, 57, 56. 7 V. McKee, M. Zvagulis and C. A. Reed, Inorg. Chem., 1985, 24, 2914. 8 W. B. Tolman, Inorg. Chem., 1991, 30, 4877. 9 F. Jiang, R. R. Conry, L. Bubacco, Z. Tyeklár, R. R. Jacobsen, K. D. Karlin and J. Peisach, J. Am. Chem. Soc., 1993, 115, 2093. 10 N. Komeda, H. Nagao, G. Y. Adachi, M. Suzuki, A. Uehara and K. Tanaka, Chem. Lett., 1993, 1521. 11 J. A. Halfen, S. Mahapatra, M. M. Olmstead and W. B. Tolman, J. Am. Chem.Soc., 1994, 116, 2173. 12 L. Casella, O. Carugo, M. Gullotti, S. Doldi and M. Frassoni, Inorg. Chem., 1996, 35, 1101. 13 A. Escuer, R. Vicente and X. Solans, J. Chem. Soc., Dalton Trans., 1997, 531. 14 N. Arulsamy, D. Scott Bohle, B. Hansert, A. K. Powell, A. J. Thomson and S. Wocaldo, Inorg. Chem., 1998, 37, 746. 15 J. P. Costes, F. Dahan, J. Ruiz and J. P. Laurent, Inorg. Chim. Acta, 1995, 239, 53. 16 See, for example, K. Geetha, M. Netaji, A. R. Chakravarty and N. Y. Vasanthacharya, Inorg. Chem., 1996, 35, 7666. 17 P. M. H. Kroneck, J. Beurle and W. Schumacher, Metal Ions Biol. Syst., 1992, 28, 455. 18 K. Wieghardt, P. Chaudhuri, B. Nuber and J. Weiss, Inorg. Chem., 1982, 21, 3086. 19 SHELXTL-V5, Siemens Analytical X-ray Instruments Inc., Madison, WI, 1995. 20 P. Chaudhuri and K. Oder, J. Chem. Soc., Dalton Trans., 1990, 1597. 21 P. Chaudhuri, D. Ventur, K. Wieghardt, E. M. Peters, K. Peters and A. Simon, Angew. Chem., 1985, 97, 55; Angew. Chem., Int. Ed. Engl., 1985, 24, 57. 22 (a) U. Flörke, H.-J. Haupt, I. Karpenstein and P. Chaudhuri, Acta Crystallogr., Sect. C, 1993, 49, 1625; (b) P. Chaudhuri, M. Winter, U. Flörke and H.-J. Haupt, Inorg. Chim. Acta, 1995, 232, 125. 23 R. H. Benno and C. Fritchie, Jr., Acta Crystallogr., Sect. B, 1973, 29, 2493; H. Lang, M. Herres and L. Zsolnai, Organometallics, 1993, 12, 5008. 24 B. Prelesnik, K. Andjelkovic, M. Malinar and N. Juranic, Acta Crystallogr., Sect. C, 1995, 51, 1767. 25 O. Kahn, Molecular Magnetism, VCH, Weinheim, 1993. 26 P. J. Hay, J. C. Thibeault and R. J. HoVmann, J. Am. Chem. Soc., 1975, 97, 4884. 27 Y. Nishida, M. Takeuchi, K. Takahashi and S. Kida, Chem. Lett., 1985, 631. Paper 8/05104F

ISSN:1477-9226    

DOI:10.1039/a805104f

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3805–3813 3805 Nitrogen versus oxygen co-ordination of carboxamidefunctionalized triazacyclononane ligands in transition metal ion complexes Thomas Weyhermüller, Karl Weighardt and Phalguni Chaudhuri* Max-Planck-Institut für Strahlenchemie, PO Box 101365, D-45413 Mülheim an der Ruhr, Germany. E-mail: chaudh@mpi-muelheim.mpg.de Received 17th August 1998, Accepted 22nd September 1998 Mononuclear complexes of zinc(II), copper(II), nickel(II), cobalt(II), iron(II), manganese(II), chromium(III), vanadium(III) and vanadyl(IV) with the ligands 1,4,7-tris(carbamoylmethyl)-1,4,7-triazacyclonanone (L) and its N-methyl derivative C6H12N3(CH2CONHMe)3 (MeL) have been prepared and structurally, magnetochemically and spectroscopically characterized.The ligands on complexation provide, in general, the MN3O3 moiety, i.e. they co-ordinate through the carboxamide oxygens with the exception of chromium(III). Chromium(III) forms complexes not only with the carboxamide oxygens, but also with the deprotonated amide nitrogen, thus yielding both CrN3O3 and CrN4O2 chromophores, respectively, which have been confirmed by crystal structures.The crystal structure of the VO21 complex reveals a typical six-co-ordinate geometry with one amide dangling free, while that of the vanadium(III) compound shows seven-co-ordinate pentagonal bipyramidal geometry with the amide co-ordinated to the metal via its carbonyl oxygen atoms together with a chloride ion.The pendant carboxamide groups hydrolyse slowly to the corresponding carboxylic acid groups. The ligands L and MeL provide weak ligand fields and, relative to the trimethyl substituted cyclic amine [9]aneN3, i.e. 1,4,7-trimethyl-1,4,7-triazacyclononane, are more eVective in stabilizing the metal centers in the oxidation state II. Introduction This work stems from our interest in utilizing 1,4,7-triazacyclononane (TACN) based ligands for the synthesis of highand low-valent organometallics as well as for the preparation of models of metalloprotein active sites.1 Owing to the kinetic and thermodynamic stability which these macrocycles impart upon their complexes, they have become highly valued as supporting ligands not only for the synthesis of monomeric and homopolynuclear but also for heteropolynuclear complexes.2 This macrocyclic amine is readily derivatized at the nitrogen atoms,3–5 resulting in C3-symmetric N,N9,N0-trisubstituted ligands.A variety of pendant-arm ligands such as CH2CO2H,6,7 CH2CH2OH,8 CH2C6H4OH,9 CH2C6H4SH,10 CH2C6H4NH2,11 CH2CH2NH2,12 CH2C5H4N,13,14 bipyridylmethyl,15 imidazolylmethyl, 16 pyrazolylmethyl,17 CH2CONHCH(CH2Ph)CO2Me,18 CH2CH2C(NH2)]] NOH19 and CH2C(Me)]] N]OH19 have been prepared to generate a battery of ligands possessing divergent steric and electronic properties and co-ordination environments. Our interest in this type of functionalization has prompted us to prepare pendant-amide ligands based on TACN.Ligands with pendant-amide groups, CH2CONH2, are particularly interesting because they can form complexes in which the NH2 group can remain protonated and the metal coordination of the CH2CONH2 group occurs through the carbonyl oxygen, or the NH2 group can deprotonate and so act as a deprotonated amide nitrogen thus providing a strong metal binding site. This co-ordination behavior is very similar to that of the CONH group in small peptides.20 We believe that this study provides new information on the metal–peptide bonding mode,21 which at present is still scarcely explored.The ligand 1,4,7-tris(carbamoylmethyl)-1,4,7-triazacyclononane (L) has the advantage of yielding complexes containing both carbonyl oxygens and amide nitrogens as donor atoms. Additionally, it might stabilize unusual higher oxidation states of metal centers through its deprotonated N-amidate group. It is also interesting to compare the ligating property of the CH2CONH2 with that of the CH2CO2 2 group.7 We found that the divalent metal ions are ineVective in promoting amide nitrogen deprotonation, but a trivalent metal ion like CrIII can easily deprotonate the NH2 group in the ligand. Our goal is to characterize the interaction of the chromium(III) ion with the amide bond and to compare it with that of the other metals with the peptide ligands.In this paper we describe the mononuclear complexes of ZnII, CuII, NiII, CoII, FeII, MnII, CrIII, VIII and VO21 with the new ligands L and its N-methyl derivative MeL and their characterization by spectroscopic methods together with structural characterization of five complexes.While our research was in progress, the synthesis of L and its complexes with YIII and LuIII were reported.22a Very recently Berreau et al.22b have described copper complexes of 1,4-diisopropyl-1,4,7-triazacyclononane linked to secondary and tertiary amide groups and where their results and present work overlap good agreement is observed. Experimental The macrocyclic amine 1,4,7-triazacyclononane was prepared according to a modified method23 described by Atkins et al.24 All other chemicals were obtained from commercial sources and used as received.Microanalyses were performed by Mikroanalytisches Laboratorium Dornis and Kolbe, Mülheim an der Ruhr. The perchlorate anion was determined gravi-3806 J. Chem.Soc., Dalton Trans., 1998, 3805–3813 metrically as tetraphenylarsonium perchlorate. Solution electronic spectra were measured on a Perkin-Elmer Lambda 19 spectrophotometer, Fourier transform infrared spectroscopy on KBr pellets on a Perkin-Elmer 2000 FT-IR instrument. Magnetic susceptibilities of the polycrystalline samples were recorded on a SQUID magnetometer (MPMS, Quantum Design) in the temperature range 2–290 K with an applied field of 1 T. Experimental susceptibility data were corrected for the underlying diamagnetism using Pascal’s constants. Mass spectra were recorded either in the ESI or FAB mode using a VG 8200 spectrometer.Ligand syntheses 1,4,7-Tris(carbamolymethyl)-1,4,7-triazacyclononane L was synthesized by a modification of the reported method.22 To a solution of 1,4,7-triazacyclononane (5.56 g, 43 mmol) in distilled ethanol (240 ml) was added 2-bromoacetamide (20.0 g, 145 mmol) with stirring at room temperature. The clear solution was charged with either solid sodium methoxide (7.0 g, 130 mmol) or sodium hydroxide (5.20 g, 130 mmol) and the mixture refluxed for 8 h.The precipitated NaBr was filtered oV and discarded. The solvent ethanol was evaporated from the filtrate under reduced pressure to yield a white solid (11.6 g), which was recrystallized from ethanol–water (400 : 100 ml). The microcrystalline product, collected in three crops during 48 h, was filtered oV, washed with diethyl ether and air-dried.Yield: 9.15 g (ª70%). Mp 233–235 8C (Found: C, 48.2; H, 8.10; N, 28.0. Calc. for C4H8N2O: C, 47.99; H, 8.05; N, 27.98%). 1H NMR (D2O): d 2.80 (s, 12 H, TACN ring CH2) and 3.35 (s, 6 H, carboxamide CH2). IR (KBr, cm21): 3402s, 3303s (br), 3169s (br), 1711ms and 1655s (br). MS: m/z 301[M] and 187 (100% abundance). The N-methyl derivative of the ligand, MeL, C6H12N3- (CH2CONHMe)3 was prepared by carbamoylalkylation of 1,4,7-triazacyclononane with 2-bromo-N-methylacetamide.25,26 1,4,7-Triazacyclononane (6.25 g, 48 mmol) was stirred in dry acetonitrile (500 ml) under an atmosphere of argon.Potassium carbonate (20.8 g, 150 mmol) and 2-bromo-N-methylacetamide (22.8 g, 150 mmol) were added, and the mixture was heated to reflux for 36 h. The mixture was then cooled to room temperature, the inorganic salts were filtered oV and the filtrate was evaporated under reduced pressure. The waxy orange-yellow solid was purified by repeated crystallization from acetonitrile– ether to yield a white solid.Yield: 6.5 g (ª40%) (Found: C, 51.7; H, 8.7; N, 23.9. Calc. for C5H10N2O: C, 52.61; H, 8.83; N, 24.54%). 1H NMR (CD3OD): d 2.77/2.76 (two s, 21 H tacn ring CH2 1 carboxamide CH3) and 3.25 (s, 6 H, carboxamide CH2). MS: m/z 342[M] and 215 (100% abundance). IR (KBr, cm21): 3323s, 3291s, 1655s and 1565s. Preparation of complexes [V(Cl)L][ClO4]2?2MeOH 1. All operations were done under an argon atmosphere. A suspension of the amide ligand L (0.3 g, 1 mmol) and [V(MeCN)3Cl3] (0.28 g, 1 mmol) in methanol (50 ml) was stirred under argon for 1 h, during which time it changed from green to rose.The suspension was filtered to remove the colored solid and the orange-red filtrate treated with a solid sample of NaClO4?H2O (0.5 g). The solution yielded after 24 h orange crystals, which became turbid on exposure to air. The crystals were preserved under argon to avoid the loss of solvent molecules. Yield: 0.18 g (ª28%) (Found: C, 25.8; H, 4.4; N, 12.9; V, 8.0; ClO4, 31.9.Calc. for C12H24Cl3N6O11V?2CH3OH: C, 25.88; H, 4.96; N, 12.93; V, 7.84; ClO4, 30.61%). IR (KBr, cm21) 3398s, 3280s, 1661s, 1584s, 1088s and 626m. UV/VIS in water: lmax/nm (e/M21 cm21) 363(14.9), 479(24.8), 807(29.8), ª970 (sh) and ª1031 (sh) (24.8). [V(O)L]Br[ClO4]?MeOH 2. A suspension of the ligand L (0.3 g, 1 mmol) and [Et4N]2[VOBr4] (0.65 g, 1 mmol) in methanol (50 ml) was stirred in a closed vessel for 20 h to obtain a blue-green solution.The solution was filtered to remove any solid particles and to the filtrate was added NaClO4?H2O (0.42 g). The solution yielded after 3 d blue crystals of complex 2, which were filtered oV and air-dried. Yield: 0.38 g (66%) (Found: C, 27.8; H, 4.7; Br, 13.6; N, 14.5; V, 8.6; ClO4, 18.1. Calc. for C12H24BrClN6O8V?CH3OH: C, 26.98; H, 4.88; Br, 13.81; N, 14.52; V, 8.80; ClO4, 17.19%). IR(KBr, cm21) 3425s (sharp), 3332s (br), 3220m, 3171m, 1680s, 1663s, 1114s, 978m and 623m.UV/VIS in water: l/nm (e/M21 cm21) 583(40) and 765(25). [Cr(L 2 H)]Cl[PF6]?Me3OH 3. To an argon-scrubbed suspension of L (0.60 g, 2 mmol) in methanol (100 ml) was added CrCl2 (0.32 g, 2 mmol) under argon with constant stirring. The stirring was continued for 2.5 h at ambient temperature to obtain an aubergine colored solution with negligible amount of green solid, presumably unchanged CrCl2. The solution was filtered in the air to remove the insoluble particles.To the resulting violet solution was added NaPF6 (0.74 g, 4.4 mmol) and the clear solution, kept in a closed vessel for 24 h, yielded red-violet crystals of complex 3, which were filtered oV and air-dried. Yield: 0.26 g (23%) (Found: C, 27.3; H, 4.7; Cl, 6.1; Cr, 9.0, N, 14.9. Calc. for C12H23ClCrF6N6O3P?CH3OH: C, 27.69; H, 4.83; Cl, 6.29; Cr, 9.22; N, 14.91%). IR (KBr, cm21) 3413s, 3351m, 3256m, 3012m, 1674m, 1609s, 1566s, 841s and 558m. UV/VIS(KBr): 273 (br) and 507 nm.UV/VIS in water: lmax/nm (e/M21 cm21) 383(218) and 509(261). meff /mB (T/K): 3.70(290), 3.69(180), 3.64(80), 3.49(10) and 3.28(2). [Cr(L 2 H)][ClO4]2 3*. This microcrystalline perchlorate salt was obtained similarly by using, in place of NaPF6, NaClO4, giving a better yield (66%). [CrL][ClO4]3 4. Complex 4 was obtained by adding concentrated HClO4 (1.5 ml) to the violet solution (in the preparation of 3), instead of NaPF6. Microcrystalline pink-violet 4 was filtered oV and air-dried. Yield: 0.45 g (69%) (Found: C, 22.2; H, 3.7; Cr, 7.9; N, 12.8; ClO4, 43.9.Calc. for C12H24Cl3CrN6O15: C, 22.15; H, 3.72; Cr, 8.00; N, 12.92; ClO4, 45.88%). IR (KBr, cm21) 3368s, 3269m, 3208m, 1677s (sharp), 1567m, 1089s and 625m. UV/VIS in water: lmax/nm (e/M21 cm21) 387(150) and 512(256). meff /mB (T/K): 3.70(290), 3.69(210), 3.68(150) and 3.65(100). [Cr(MeL)][ClO4]3?3.5H2O 5. An argon-scrubbed solution of MeL (0.34 g, 1 mmol) in methanol (50 ml) was treated with 0.16 g (1.3 mmol) of CrCl2 and the suspension stirred under argon for 6 h. The resulting suspension was filtered to remove a green solid, presumably an excess of CrCl2, in the air and to the violet filtrate perchloric acid (1 ml) was added.The precipitated red solid was collected after 1 h by filtration and air-dried. The solid was recrystallized from water acidified with HClO4 (pH ª 1) at 60 8C to yield X-ray quality red crystals. Yield: 0.5 g (66%) (Found: C, 23.9; H, 4.9; Cr, 7.0; N, 11.3; ClO4, 39.1.Calc. for C15H30Cl3CrN6O15?3.5H2O: C, 23.83; H, 4.93; Cr, 6.88; N, 11.12; ClO4, 39.47%). IR(KBr, cm21) 3432s (br), 1659s, 1542m, 1109, 1088s and 626m. UV/VIS in water (pH ª 3): l/nm (e/M21 cm21) 389(162) and 513(278). meff /mB (T/K): 3.83(290), 3.82(160), 3.80(15 K), 3.75(5) and 3.59(2). Compound 5 could also be obtained by recrystallizing 6 from water, acidified with HClO4 (pH ª 1). [Cr(MeL 2 H)][ClO4]2?MeOH 6. A solution of MeL (0.34 g, 1 mmol) in dry methanol (50 ml) was treated with solid CrCl2 (0.12 g, 1 mmol) under argon.The suspension was stirred under argon for 1 h at ambient temperature and left overnight. The resulting violet solution was filtered to remove any solid particles and the filtrate after addition of NaClO4 (0.36 g) concentrated by passing argon through it. The solution yielded afterJ. Chem. Soc., Dalton Trans., 1998, 3805–3813 3807 cooling at 4 8C deep violet crystals of 6. Yield: 0.17 g (27%) (Found: C, 30.7; H, 5.3; Cr, 8.8; N, 13.4; ClO4, 32.7.Calc. for C15H29Cl2CrN6O11?CH3OH: C, 30.78; H, 5.33; Cr, 8.33; N, 13.46; ClO4, 31.86%). IR(KBr, cm21) 3535m, 3415m (br), 3324m, 1646s, 1596s, 1088s and 624m. meff /mB (T/K): 3.80(290), 3.79(150), 3.78(100), 3.74 (15 K), 3.72 (10 K) and 3.43(2). [MIIL][ClO4]2 (MII 5 Zn 7, Cu 8, Ni 9, Co 10, Fe 11 or Mn 12). As these complexes were prepared in a very similar way a representative method only is described. An argon-scrubbed solution of M(O2CMe)2?xH2O (1 mmol) in methanol (30 ml) was treated with a sample of the amide ligand L (0.30 g, 1 mmol) under vigorous stirring.The mixture was stirred at room temperature until a clear solution was obtained (ca. 2 h). Upon addition of NaClO4?H2O (0.6 g) crystalline solids were obtained. The microcrystalline product was collected by filtration and air-dried. Yield: 70–80%. Complex 7 [Found (Calc. for C12H24Cl2N6O11Zn): C, 25.4(25.53); H, 4.3(4.28); N, 14.8(14.88); Zn, 11.7 (11.61); ClO4, 35.0(35.23)%], colorless; IR(KBr, cm21) 3430m, 3380m, 3246m (br), 3107m, 1665s, 1611m, 1596m, 1103s, 1087s and 626m.Complex 8 [Found (Calc. for C12H24- Cl2CuN6O11): C, 25.7(25.61); H, 4.3(4.30); Cu, 10.9(11.29); N, 14.9(14.93); ClO4, 34.9(35.34)%], blue; IR(KBr, cm21) ª3400s (br), 3150m, 1663s, 1591m, 1145s, 1110s, 1088s and 627m; UV/ VIS in water lmax/nm (e/M21 cm21) 764(96.5); meff /mB (T/K) 1.97(290), 1.93(200), 1.90(130), 1.85(30), 1.83(10), 1.82(7) and 1.60(2).Complex 9 [Found (Calc. for C12H24Cl2N6NiO11: C, 25.9(25.97); H, 4.4(4.36); N, 15.2(15.14); Ni, 9.7(10.04); ClO4, 36.0(35.84)%], violet; IR(KBr, cm21) 3459m, 3386s, 3336m, 3294m (br), 3133m, 3096s, 1676s, 1622m, 1594m, 1146s, 1115s, 1093s and 627m; UV/VIS in water l/nm (e/M21 cm21) 354(19.6), 561(14.7), 802(19.6) and 920(39.3); UV/VIS in MeCN 358(20.7), 561(15.5), 798(20.7) and 920(41.5); meff /mB (T/K) 3.13(290), 3.10(200), 3.05(30), 3.04(10), 3.00(5) and 2.85(2).Complex 10 [Found (Calc. for C12H24Cl2CoN6O11: C, 25.7(25.82); H, 4.3(4.33); Co, 10.1(10.56); N, 15.0(15.06); ClO4, 34.9(35.63)%], rose; IR(KBr, cm21) 3431s, 3355m, 3300m, 3241m, 1665s, 1595m, 1090s (br) and 625m; UV/VIS in water lmax/nm (e/M21 cm21) 500(20), 668(5) and 1146(15); meff /mB (T/K) 4.91(290), 4.97(200), 5.00(100), 4.91(50), 4.59(7) and 3.67(2). Complex 11 [Found (Calc. for C12H24Cl2FeN6O11): C, 25.8(25.96); H, 4.2(4.36); Fe, 9.1(10.06); N, 15.0(15.14); ClO4, 35.6(35.83)%], pale yellow; IR(KBr, cm21) 3374m, 3323w, 3252m, 3094m (br), 1681m, 1659s, 1608m, 1589m, 1144s, 1116s, 1089s and 628m; meff /mB (T/K) 4.90(290), 4.80(50), 4.60(15), 4.10(5) and 3.21(2). Complex 12 [Found (Calc.for C12H24Cl2MnN6O11): C, 25.9(26.04); H, 4.4(4.37); Mn, 9.9(9.91); N, 15.0(15.16); ClO4, 36.0(35.91)%], colorless; IR(KBr, cm21) 3462m, 3379m (br), 3361s (br), 1695w, 1655s, 1611m, 1145s, 1120s, 1088s and 627m; meff /mB (T/K) 5.88(290), 5.87(50), 5.85(30), 5.84(20), 5.80(10 K), 5.76(7) and 4.74(2).The metal complexes of MeL, [MII(MeL)][ClO4]2 (MII = Zn 13, Cu 14, Ni 15, Co 16, Fe 17 or Mn 18, have been prepared in a very similar way. Complex 13 (Found: C, 29.5; H, 4.9; N, 13.8; ClO4, 33.1. Calc. for C15H30Cl2N6O11Zn: C, 29.72; H, 4.99; N, 13.86; ClO4, 32.81%): IR(KBr, cm21) 3106m, 1634s, 1145s, 1086s and 625m; 1H NMR (D2O) d 2.74–2.67 (m), 2.81 (s), 2.88–2.95 (m) and 3.66 (s); 13C NMR (D2O): d 27.04, 33.40 (CH3N), 50.56 (CH2N), 57.90 (CH2CO) and 173.71 (C]] O); ESI-MS (MeCN) m/z 505.1 (40, M 2 ClO4) and 405.1 (100%, M 2 2ClO4).Complex 14 (Found: C, 29.16; H, 5.0; Cu, 9.5; N, 13.6; ClO4, 33.0. Calc. for C15H30Cl2CuN6O11: C, 29.79; H, 4.99; Cu, 10.51; N, 13.89; ClO4, 32.88%): IR(KBr, cm21) 3364s, 3233m, 3091m, 1638s, 1619s, 1145s, 1115s, 1086s and 626m; UV/VIS in water lmax/nm (e/M21 cm21) 767(96.6); meff /mB (T/K) 1.82(5.0) and 1.94(150.0); ESI-MS (MeCN) m/z 504 (M 2 ClO4) and 404.2 (100%, M 2 2ClO4). Complex 15 (Found: C, 29.2; H, 5.1; N, 13.7; Ni, 9.6; ClO4, 32.9.Calc.: C, 30.03; H, 5.04; N, 14.01; Ni, 9.78; ClO4, 33.15%). IR(KBr, cm21) 3426s, 3233m, 3073m, 1630s, 1145s, 1117s, 1088s and 626m; UV/VIS in water lmax/nm (e/M21 cm21) 357(19.3), 561(14.5) and 918(43.5); meff /mB (T/K) 2.95(7.0) and 3.00(150); ESI-MS (MeCN) m/z 499.2 (50, M 2 ClO4) and 399.1 (100%, M 2 2ClO4). Complex 16 (Found: C, 29.4; H, 4.9; Co, 9.6; N, 13.9; ClO4, 33.1. Calc.: C, 30.01; H, 5.04; Co, 9.82; N, 14.00; ClO4, 33.13%): IR(KBr, cm21) 3430s, 3246s, 3106s, 1630s, 1145s, 1081s and 625m; UV/VIS in water lmax/nm (e/M21 cm21) 507(20.4), 659(6.8) and 1160(15.3); meff /mB (T/K) 2.86(2.0), 3.24(5.0), 3.36(10) and 3.73(150); ESI-MS(MeCN) m/z 500.2 (90, M 2 ClO4) and 400.2 (100%, M 2 2ClO4).Complex 17 (Found: C, 28.9; H, 5.2; Fe, 9.0; N, 13.8; ClO4, 33.0. Calc.: C, 30.17; H, 5.06; Fe, 9.33; N, 14.07; ClO4, 33.31%): IR(KBr, cm21) 3378s, 3277s, 3109m, 1631s, 1148mn, 1090s and 624m; meff /mB (T/K) 4.01(2.0), 4.79(5.0), 5.09(10) and 5.07(150); ESIMS( MeCN) m/z 497 (100%, M 2 ClO4) and 397.0 (91%, M 2 2ClO4). Complex 18 (Found: C, 29.4; H, 5.0; Mn, 8.3; N, 13.8; ClO4, 33.9.Calc.: C, 30.21; H, 5.07; Mn, 9.21; N, 14.09; ClO4, 33.36%): IR(KBr, cm21) 3402s, 3245m, 3091m, 1632s, 1144s, 1106s, 1088s and 626m; meff /mB (T/K) 5.72(5.7), 5.77(10.0), 5.80(20) and 5.81(290); ESI-MS(MeCN) m/z 496.2 (100, M 2 ClO4) and 396.2 (43%, M 2 2ClO4).[V(MeL)Cl][ClO4]2 19. This orange crystalline compound was obtained in the same way as 1 (Found: C, 29.3; H, 4.7; Cl, 5.5; N, 13.5; ClO4, 32.0. Calc. for C15H30Cl3N6O11V: C, 28.70; H, 4.82; Cl, 5.64; N, 13.39; ClO4, 31.69%): IR(KBr, cm21) 3395s, 3105m, 1639s, 1165m, 1091s and 624m; meff /mB (T/K) 2.23(2), 2.70(7.1), 2.71(10) and 2.75(150). Crystallography The crystallographic data of [VIII(Cl)L][ClO4]2?2MeOH 1, [VIV(O)L]Br[ClO4]?MeOH 2, [CrIII(L 2 H)]Cl[PF6]?MeOH 3, [CrIII(MeL)][ClO4]3?3.5H2O 5 and [NiIIL][ClO4]2 9 are summarized in Table 1.Graphite monochromated Mo-Ka X-radiation (l 0.71073 Å) was used throughout. Intensity data collected at 293 K for complex 9 and 100 K for 1, 2, 3 and 5 were corrected for Lorentz-polarization and absorption eVects for 1, 2, 3 and 5 using the program SADABS.27 No absorption correction was done for 9. The structures were solved by direct methods by using SHELXTL-PLUS.28 The function minimized during fullmatrix least-squares refinement was Sw(|Fo| 2 |Fc |)2.Neutral atom scattering factors and anomalous dispersion corrections for non-hydrogen atoms were taken from ref. 29(a). Hydrogen atoms attached to carbon were placed at calculated positions with isotropic thermal parameters. Those bound to nitrogen were located from the diVerence map. All non-hydrogen atoms were refined with anisotropic thermal parameters. Absolute configurations for 2 and 3 were checked using the Flack X parameter.29b CCDC reference number 186/1172.See http://www.rsc.org/suppdata/dt/1998/3805/ for crystallographic files in .cif format. Results and discussion The ligands L and MeL were prepared using adaptations to the methods described in the literature for their syntheses.22,26 This involved reaction of 3 equivalents of 2-bromoacetamide or 2-bromo-N-methylacetamide with the free amine 1,4,7- triazacyclononane in the presence of a base. The ligands were conveniently isolated in reasonable yields and their purities established by elemental analyses, NMR, IR and mass spectrometry. The transition metal complexes, prepared in methanol from the respective metal salts and the free base form of the ligands L or MeL, were isolated as perchlorate salts.Although the ligands L and MeL both have potentially dissociable protons present as NH2 or NHMe groups, respectively, only one of the three groups is deprotonated upon reaction with3808 J. Chem. Soc., Dalton Trans., 1998, 3805–3813 chromium(II) chloride.The reaction of [VOBr4]22 with the ligand L yields blue crystals of [(V]] O)L]Br[ClO4]?MeOH 2, in which the ligand is only pentadentate, the sixth co-ordination site being occupied by a terminal oxo ligand. One carboxamide group is not co-ordinated to the vanadium(IV) center, consistent with the presence of two strong bands in the carbonyl region and a V]] O structural unit (nV]] O ª 978 cm21). In all other cases the physical data suggest that the ligand is hexadentate, at least, in the solid state.As the properties of the complexes 13–19 containing the methylated ligand MeL are very similar to those of the parent ligand L, and they were characterized similarly, we will refrain mostly from the discussion of the complexes with the ligand MeL. Characterization The IR spectra of all the complexes derived from L exhibit characteristic bands due to the ligand [n(NH), and n(C]] O) at about ª3200 and ª1670 cm21], and non-co-ordinated perchlorate anions (ª1090 and ª625 cm21).In the case of 3, [Cr(L 2 H)]21, a sharp strong band observed at 1609 cm21 (together with a medium strong band at 1674 cm21) can be attributed to the presence of the deprotonated amide moiety, while for 4, [CrL]31, this band is absent and a strong band is observed at 1677 cm21, a region where the n(C]] O) vibrations for all other compounds appear (see Experimental section). This is a strong indication that in 4, together with the other compounds, all three carboxamide oxygen atoms are co-ordinated to the metal centers, but not in 3.This structural proposal has been confirmed by the X-ray structural determinations. The electronic spectra of divalent and trivalent first-row transition-metal complexes (given in the Experimental section) are consistent with these metal ions being in a pseudooctahedral environment (facial N3O3 donor set). Judged on the basis of absorption coeYcients, the bands and the shoulders for 1, a seven-co-ordinated d2 species, are ascribed to the d–d transitions of the vanadium(III) centre.In contrast complex 2, a d1 VO21 species, displays only two low-intensity d–d transitions at 583 (40) and 765 nm (25 M21 cm21). These bands are assigned 30 as 2B2(dxy) æÆ 2B1(dx2 2 y2) and 2B2(dxy) æÆ 2E(dxz, dyz) transitions, respectively, assuming a C4v symmetry for 2. This leads to the 10Dq value of 17 150 cm21, which is slightly smaller than that of the corresponding tris(acetate) analog, [VO(TCTA)]7 (TCTA = 1,4,7-triazacyclononane-1,4,7-triacetate) (17 450 cm21) and in good agreement with the other reported spectra for VO21 species.31 In the visible region a single absorption band at 764 nm with e = 96.5 M21 cm21 is observed for the copper(II) complex, 8; this is similar to the spectrum for the corresponding complex with the ligand containing three acetate groups7 as pendant arms to the parent macrocycle [9]aneN3, and taken as further evidence that, with the exception of [VO(L)]21 2, the ligand L is invariably hexadentate and the copper(II) ion is six-co-ordinate. The visible spectrum of nickel(II) complex 9 has features typical of octahedral nickel(II) complexes with 3A2g æÆ 3T2g and 3A2g æÆ 3T1g(F) transitions occurring at 920 and 561 nm, respectively.A third band at 354 nm is most likely due to a 3A2g æÆ 3T1g(P) transition. The spike of 802 nm is considered to arise from the “spin-flip” transition 3A2g(3F) æÆ 1Eg(1D).Three spin-allowed transitions, as is expected for the octahedral d7 complexes, are observed for 10 with a very weak (e = 5 M21 cm21) band at 668 nm, which can be attributed to the so-called “two-electron jump” 4T1g æÆ 4A2g transition.31 Owing to the low symmetry of the CrIIIN3O3 and CrIIIN4O2 moieties the molar absorption coeYcients of the d–d transitions in the visible region are large for complexes 3 and 4. The bands at 509 (261), 383 nm (218 M21 cm21) for 3 and 512 (256), 387 nm (150 M21 cm21) for 4 are assigned to the first and second spin-allowed transitions 4A2g æÆ 4T2g and 4A2g æÆ 4T1g(F), respectively, of the d3 chromium(III) center, noting that the band assignments have been performed by assuming an octahedral ligand field.It allows us to estimate the splitting parameter 10Dq for 3 and 4 to be 19 646 and 19 531 cm21, respectively, indicating that the ligand-field strengths of L and is deprotonated form L 2 H (i.e.the amide form) are not very diVerent. That 3 with the CrN4O2 chromophore is stable in neutral aqueous solution and does not become protonated to form 4 with the CrN3O3 chromophore has been shown by the fact that the solid state and solution spectra of 3 are identical. It is also interesting in this connection that the N-methyl derivative, MeL, is also of similar ligand-field strength. Stepwise hydrolysis of the pendant carboxamide groups to the carboxylic groups That the co-ordinated carboxamide ligands are amenable to hydrolysis, resulting in carboxylic acids, is evidenced from the isolation of three chromium(III) complexes containing one, two or three co-ordinated pendant carboxylate groups, respectively; the reaction conditions are shown in Scheme 1.An aqueous solution of 4, [CrL][ClO4]3, on refluxing yielded red needles of the reported [Cr(TCTA)].7 On the other hand, an aqueous solution of 4, adjusted to pH ª1 by addition of perchloric acid and stirred at room temperature for 4 h, aVorded upon evaporation of water red crystals of the chromium compound containing two pendant CONH2 ligands and one CO2 2 group.When the above solution is stirred for 3 d the chromium compound containing only one pendant carboxamide and two acetate groups is obtained. Interestingly, the absorption maxima for the d–d bands of the above mentioned four compounds do not vary at all, viz. 512 and 387 nm. We will describe in detail the reactions of the above compounds in a future publication.Similarly compound 11, the iron(II) complex, in water yields on refluxing in the presence of air the expected hydrolysed product [FeIII(TCTA)].7 Crystal structures The seven-co-ordinate structure for the cation [VIII(Cl)L]21 is shown in Fig. 1. Among the co-ordination polyhedra33 considered as being ideal for this co-ordination number are the 1:5:1 pentagonal bipyramid (D5h), the 1:4:2 monocapped trigonal prism (C2v), the 1:3:3 monocapped octahedron (C3v) and the 4 : 3 tetragonal base-trigonal cap piano stool (Cs) structures.An analysis of the metal geometry showed that the structure of the cation in 1 could not clearly be assigned to any of these polyhedral forms. However, the geometry approximates with moderate distortions that of the 1:5:1 pentagonal bipyramid. The axial angle Cl(3)–V(1)–N(2) of 169.68 diVers from the ideal 1808, while the angle sum in the approximate plane comprising N(1)N(3)O(3)O(2)O(1) atoms about V(1) is 368.48.The sum of in-plane angles exceeds 3608 by ca. 88, which causes slight ruZing in the positions of the “in-plane” O and N donors, O(2) exhibiting the largest deviation from the mean plane. The deviation from ideal angles of 72.08 presumably occurs to accom- Scheme 1J. Chem. Soc., Dalton Trans., 1998, 3805–3813 3809 modate the formation of the five-membered chelate rings of the [9]aneN3 fragment. The largest “in-plane” angle of 79.38, O(1)– V(1)–O(2), occurs between the co-ordinated oxygens provided by the two neighboring amide pendant arms.Deviation from linearity of the axial ligands is observed even for the six-coordinate complexes; for example the axial O–M–N angles in comparable MIII(TCTA) complexes 7 are ca. 1708. It has been pointed out34 that the d2 electron configuration is well suited to the pentagonal bipyramidal geometry as two electrons are placed in the doubly degenerate dxz,dyz orbitals, and thus the dxy and dx2 2 y2 are eVectively antibonding.The other geometries, having lower symmetry, will not have degenerate orbitals and one electron will have to be placed above the other, thus lowering the maximum possible ligand field stabilization energy and resulting in an orbital singlet ground state. Selected bond lengths and angles for complex 1 are listed in Tables 2 and 3. The axial V(1)–N(2) distance, 2.088(2) Å, is appreciably shorter than the “in-plane” V(1)–N(3) and V(1)–N(1) distances, 2.213(2) and 2.313(2) Å, respectively. The latter V–N bond distances are comparable to those of related vanadium(III) complexes with aminopolycarboxylate ligands, e.g.EDTA.35 The V–O bond lengths of the donor oxygens O(1), O(2) and O(3) range from 2.095(2) to 2.157(2) Å, which are slightly longer than the comparable V–O distances. The amide nitrogens N(4), N(5) and N(6) exhibit hydrogen bonding with the oxygens O(13) and O(40) of the solvent of crystallization, methanol: N(4) ? ? ? O(13) 3.180, N(6) ? ? ? O(13) 2.889 and N(5) ? ? ? O(40) 2.785 Å.Blue crystals of complex 2 consist of the dication [V(O)L]21, one non-co-ordinated bromide and perchlorate anion and a methanol molecule of crystallization. Fig. 2 shows the cation together with the atom-labeling scheme; Tables 2 and 3 summarize important bond distances and angles. The vanadium(IV) ion is in a distorted octahedral environment with three tertiary amine nitrogen atoms, two carbamoyl oxygen atoms, and one terminal oxo ligand ( fac-N3O3 donor set).The ligand is only pentadentate; one amide pendant group is not co-ordinated to the vanadium center. The V–O(4) distance is short, 1.606(2) Å, indicating the considerable double-bond character typical of vanadyl(IV) complexes.36 The V–N(1) bond trans to the V]] O(4) group is long 2.283(3) Å, which is characteristic for the strong trans influence of the V]] O group. The other two V–N bonds are shorter, 2.179(2) and 2.114(2) Å.The conformation of the three five-membered V–N–C–C–N chelate rings is (lll) or (ddd). Interestingly, the two five-membered chelate rings built by the Fig. 1 An ORTEP32 diagram of the dication [VIII(Cl)L]21 in complex 1. co-ordinated carbamoylmethyl groups V–O–C–C–N adopt the (ll) conformation if the parent cyclononane rings are (ddd)- configured and vice versa. The molecular structure of the cation in complex 3 is shown in Fig. 3. The geometry about the CrIII is pseudo-octahedral, with three facial sites occupied by the nitrogen donors of the parent cyclononane base fragment, two carbamoyl oxygen atoms, and one deprotonated amide nitrogen of the ligand thus aVording a CrN4O2 chromophore and completing the coordination sphere. Facial co-ordination of the [9]aneN3 moiety results in similar bond angles and distances (Tables 2 and 3), within this portion of the ligand, to those found in other chromium( III) complexes of triazacyclononane derivatives with different pendant arms.Thus, the Cr–N (TACN) distances, 2.054(4), 2.065(4) and 2.068(4) Å, and N–Cr–N bite angles (average 85.98) to the TACN nitrogens match with those in the [Cr(TCTA)] complex.7 In contrast to [Cr(TCTA)], a twist angle q of 47.28 (608 for octahedral co-ordination) indicates a considerable trigonal twisting of the co-ordination environment generated by the N4O2 ligand donors. The Cr–O (carbamoyl) bond lengths, 1.987(3) and 1.970(3) Å, are comparable to those reported 37,38 and slightly longer than those in 5.The conformation of the five-membered chelate rings is very similar to that found in 2. The bond length of the deprotonated amide nitrogen to the chromium, Cr–N(4) 1.963(4) Å, is significantly shorter than the Cr–N (amine) bonds, average 2.062(8) Å, reflecting the diVer- Fig. 2 Molecular structure of [V(O)L]Br[ClO4]?MeOH 2. Fig. 3 Structure of the dication [Cr(L 2 H)]21 in crystals of complex 3.3810 J.Chem. Soc., Dalton Trans., 1998, 3805–3813 Table 1 Crystallographic data for [V(Cl)L][ClO4]2?2MeOH 1, [V(O)L]Br[ClO4]?MeOH 2, [Cr(L 2 H)Cl[PF6]?MeOH 3, [Cr(MeL)[ClO4]3?3.5H2O 5 and [NiL][ClO4]2 9 Formula M Crystal size/mm Crystal system Space group a/Å b/Å c/Å b/8 V/Å3 Z Dc/g cm23 DiVractometer m/mm21 F(000) qmax/8 No. independent reflections [F > 4.0s(F)] Data/parameters RF [I > 2s(I)] T/K 1 C14H32Cl3N6O13V 649.75 0.35 × 0.42 × 0.32 Orthorhombic Pbcn 13.284(3) 11.925(3) 32.508(6) 5150(2) 8 1.676 Siemens SMART 0.770 2688 28 5726 5261/356 0.053 100(2) 2 C13H28BrClN6O9V 578.71 0.53 × 0.39 × 0.07 Orthorhombic P212121 9.877(2) 12.315(3) 17.779(3) 2162.6(8) 4 1.777 Siemens SMART 2.489 1180 33 8060 6653/300 0.036 100(2) 3 C13H27ClCrF6N6O4P 563.83 0.18 × 0.25 × 0.21 Orthorhombic P212121 11.886(1) 16.362(2) 11.335(1) 2204.4(4) 4 1.699 Siemens SMART 0.797 1156 25 3684 2975/306 0.044 100(2) 5 C15H37Cl3CrN6O18.5 755.86 0.54 × 0.54 × 0.25 Monoclinic C2/c 29.058(5) 13.680(3) 15.286(3) 94.39(3) 6059(2) 8 1.657 Siemens SMART 0.730 3136 28 6168 4615/439 0.074 100(2) 9 C12H24Cl2N6NiO11 557.98 1.12 × 0.42 × 0.35 Monoclinic P21/c 18.034(1) 11.874(1) 10.573(2) 106.43(1) 2171.6(5) 4 1.707 Enraf Nonius CAD4 1.207 1152 26 4306 2799/301 0.066 293(2) ence between sp2 and sp3 hybridization. A very similar relative order of metal–ligand bond lengths 39 M–N (amine) > M–O (carboxy) > M–N (peptide) is found in the peptide complexes of CoIII, NiII, CuII and CrIII.It is noteworthy that the Cr–N(3) (amine) bond length trans to Cr–N(4) (amide) is not signifi- cantly diVerent from those of the other Cr–N (amine) lengths, indicating the absence of the trans influence. The hydrogen atom bonded to N(4) was directly located in the Fourierdi Verence map. The diVerence between the carbonyl C–O bond lengths for both co-ordinated [O(3)–C(12) 1.271(5) and O(2)– C(10) 1.277(6) Å] and unco-ordinated [O(1)–C(8) 1.254(6) Å] amide carbonyls is small, but not negligible.This is in complete accord with the description that, after deprotonation of the NH2 group in the amide ligand, the bonding situation between C(8) and O(1) atoms is comparable with that in a ketone >C]] O. In other words the delocalization of the p electrons over the whole amide >C(O)NH2 group decreases on deprotonation, rendering a more single bond character to the C–N bond in the pendant arm. The corresponding C–N bond lengths belonging to the pendant arms are as follows: C(12)–N(6) 1.298(7), C(10)– N(5) 1.301(6) and C(8)–N(4) 1.320(6) Å.The molecular structure of the cation [NiL]21 in complex 9, as illustrated in Fig. 4, shows that the geometry about the NiII is pseudo-octahedral, with three facial sites occupied by the nitro- Fig. 4 An ORTEP representation of the [NiL]21 cation of complex 9. gen donors of the macrocycle and three carbonyl oxygens of the pendant amide groups completing the co-ordination sphere.Facial co-ordination of the 1,4,7-triazacyclononane moiety results in similar bond angles and distances, within this portion of the ligand, to those found in other nickel(II) complexes of TACN derivatives with diVerent pendant arms.40 Thus, the Ni–N distances, 2.070(5), 2.049(5) and 2.051(6) Å, and the N–Ni–N bite angles (average 86.28) to the TACN nitrogens match those in [Ni(TCTA)]2.41 The base fragment [9]aneN3 is chiral once co-ordinated to a metal ion, having (lll) or (ddd) conformation of the five-membered chelate rings M–N–C– C–N.For a given arrangement, the pendant amide groups may attach themselves in a clockwise (Type I) or anticlockwise (Type II) fashion. Type II has been found in 9 similar to [NiII- (TCTA)]2 containing the acetate pendant groups. Considerable trigonal twisting of the amide oxygens relative to the nitrogens of [9]aneN3 is found in 9 indicating significant distortion from octahedral geometry.The oxygens are rotated by 15.98 (q = 44.18) from the expected 608 for octahedral co-ordination. Fig. 5 shows a perspective view of complex 5. The structure consists of monomeric, tricationic ions [Cr(MeL)]31. The chromium is co-ordinated to six donor atoms, three nitrogens of the base fragment [9]aneN3 and three oxygens of Fig. 5 Perspective view of the trication [Cr(MeL)]31 in complex 5.J. Chem. Soc., Dalton Trans., 1998, 3805–3813 3811 the amide groups occupying facial positions, respectively of a distorted octahedron.The degree of distortion from the regular octahedral arrangement of the N3O3 donor set, expressed by the twist angle q, is 498, indicating the propensity of CrIII with d3 electron configuration for an octahedral environment. The average Cr–N and Cr–O distances, 2.04 and 1.95 Å, respectively are nearly identical to those of [Cr(TCTA)].7 Bond distances and angles within the [9]aneN3 fragment in the methyl derivative of the amide ligand are very similar to those observed in 9, indicating its rigid nature, and selective metrical parameters are summarized in Tables 2 and 3.Magnetic susceptibility measurements, EPR and Mössbauer spectra Variable-temperature (2–290 K) magnetic susceptibility data were collected for the vanadium(III) complex 1. The magnetic moment of 2.32mB at 2 K increases monotonically with increasing temperature reaching a value of 2.92mB at 290 K; selected values of meff are 2.83 mB at 7 K, 2.88 mB at 30 K, 2.90 mB at 100 K and 2.91 mB at 200 K.The experimental magnetic data were simulated by a least-squares fitting computer program with a full-matrix diagonalization approach. The Hamiltonian used to describe the paramagnetism is given by eqn. (1) where spin S is H = gbHS 1 D[Sz 2 2 1 3 – S(S 1 1)] (1) 1 and D is the axial zero-field splitting parameter. The measured magnetic moments meff are reported in Fig. 6, along with the best fit using the above equation.The best fit parameters are · gÒ = 1.97 and D = 3.63 cm21. This best fit also required the addition of a temperature-independent paramagnetism term (TIP) of 1.45 × 1024 cm3 mol21. Plots of meff vs. T with each sign of D showed that very much larger values of D (<220 cm21) are required if D is negative. Moreover, all previous values of D for the vanadium(III) ion have been shown to be positive.42 The g and D values for 1 are in accord with the literature values for vanadium(III) complexes.43 The deviation from Fig. 6 Plot of meff vs. T for complex 1. The solid line is the simulation using the Hamiltonian described in the text. Table 2 Selected bond distances (Å) for complexes 1, 2, 3, 5 and 9 M–N(1) M–N(2) M–N(3) M–O(1) M–O(2) M–O(3) V–Cl(3) V]] O(4) Cr–N(4) 1 2.313(2) 2.088(2) 2.213(2) 2.157(2) 2.104(2) 2.095(2) 2.372(1) —— 2 2.283(2) 2.179(2) 2.114(2) 2.011(2) — 2.019(2) — 1.606(2) — 3 2.054(4) 2.065(4) 2.068(4) — 1.970(3) 1.987(3) —— 1.963(4) 5 2.041(4) 2.039(4) 2.046(4) 1.959(4) 1.946(3) 1.959(3) ——— 9 2.070(5) 2.051(6) 2.049(5) 2.046(4) 2.026(5) 2.073(4) ——— the Curie law for 1 is less pronounced than anticipated, due to partial quenching of the orbital momentum resulting from the covalence (k < 1) and the lowering of symmetry, resulting in an orbital singlet ground state.As a result D is positive. Calculations on trivalent vanadium have appeared 44 which lend credence to the presence of a TIP term.The magnetic moment of 1.69 mB at 2 K for complex 2 increases slowly, but steadily, with increasing temperature reaching a value of 1.75 mB at 290 K; a few selected meff values are 1.70 (5), 1.71 (30), 1.72 (110), 1.73 (170) and 1.74 mB (230 K). Several room temperature magnetic susceptibility measurements have been made for both powdered samples and aqueous solutions of vanadyl complexes.45 The moments all are approximately equal to the spin only value of 1.73 mB for one unpaired spin, as is expected when the orbital contribution is completely quenched in the low symmetry field.The experimental magnetic moments were fitted by the theoretical equation containing only the Zeeman term and the temperature- independent contributions to the susceptibility (TIP, so-called high-frequency terms), and the best fit parameters Table 3 Selected bond angles (8) of complexes 1, 2, 3, 5 and 9 1 N(2)–V(1)–O(3) O(3)–V(1)–O(2) O(3)–V(1)–O(1) N(2)–V(1)–N(3) O(2)–V(1)–N(3) N(2)–V(1)–N(1) O(2)–V(1)–N(1) N(3)–V(1)–N(1) O(3)–V(1)–Cl(3) O(1)–V(1)–Cl(3) N(1)–V(1)–Cl(3) 94.78(9) 67.76(7) 138.29(7) 74.58(8) 128.43(8) 90.38(8) 143.78(8) 78.57(8) 93.54(6) 75.15(6) 79.28(6) N(2)–V(1)–O(2) N(2)–V(1)–O(1) O(2)–V(1)–O(1) O(3)–V(1)–N(3) O(1)–V(1)–N(3) O(3)–V(1)–N(1) O(1)–V(1)–N(1) N(2)–V(1)–Cl(3) O(2)–V(1)–Cl(3) N(3)–V(1)–Cl(3) 77.28(8) 102.39(8) 79.30(8) 72.55(8) 148.65(8) 148.06(8) 70.21(8) 169.61(7) 111.74(6) 102.11(6) 2 O(4)–V–O(1) O(1)–V–O(3) O(1)–V–N(3) O(4)–V–N(2) O(3)–V–N(2) O(4)–V–N(1) O(3)–V–N(1) N(2)–V–N(1) 101.88(7) 94.94(6) 153.77(7) 93.85(8) 158.08(7) 172.67(8) 85.25(6) 79.39(7) O(4)–V–O(3) O(4)–V–N(3) O(3)–V–N(3) O(1)–V–N(2) N(3)–V–N(2) O(1)–V–N(1) N(3)–V–N(1) 102.02(7) 104.33(8) 78.98(7) 96.49(7) 82.53(7) 76.29(6) 77.78(7) 3 N(4)–Cr(1)–O(2) O(2)–Cr(1)–O(3) O(2)–Cr(1)–N(1) N(4)–Cl(1)–N(2) O(3)–Cr(1)–N(2) N(4)–Cr(1)–N(3) O(3)–Cr(1)–N(3) N(2)–Cr(1)–N(3) 98.4(2) 94.86(13) 167.54(14) 97.1(2) 165.82(14) 165.9(2) 82.00(14) 84.4(2) N(4)–Cr(1)–O(3) N(4)–Cr(1)–N(1) O(3)–Cr(1)–N(1) O(2)–Cr(1)–N(2) N(1)–Cr(1)–N(2) O(2)–Cr(1)–N(3) N(1)–Cr(1)–N(3) 79.1(2) 81.2(2) 97.55(14) 82.56(14) 85.1(2) 95.73(14) 85.0(2) 5 O(2)–Cr–O(1) O(1)–Cr–O(3) O(1)–Cr–N(2) O(2)–Cr–N(1) O(3)–Cr–N(1) O(2)–Cr–N(3) O(3)–Cr–N(3) N(1)–Cr–N(3) 96.7(2) 96.41(14) 168.74(14) 96.8(2) 167.90(14) 168.3(2) 82.49(14) 85.5(2) O(2)–Cr–O(3) O(2)–Cr–N(2) O(3)–Cr–N(2) O(1)–Cr–N(1) N(2)–Cr–N(1) O(1)–Cr–N(3) N(2)–Cr–N(3) 95.27(14) 82.4(2) 94.85(14) 83.13(14) 85.8(2) 94.9(2) 86.4(2) 9 O(2)–Ni–O(1) O(1)–Ni–N(3) O(1)–Ni–N(2) O(2)–Ni–N(1) N(3)–Ni–N(1) O(2)–Ni–O(3) N(3)–Ni–O(3) N(1)–Ni–O(3) 96.1(2) 96.9(2) 167.0(2) 102.1(2) 86.6(2) 91.0(2) 81.7(2) 165.8(2) O(2)–Ni–N(3) O(2)–Ni–N(2) N(3)–Ni–N(2) O(1)–Ni–N(1) N(2)–Ni–N(1) O(1)–Ni–O(3) N(2)–Ni–O(3) 165.3(2) 82.9(3) 86.0(3) 81.5(2) 86.0(2) 91.9(2) 101.1(2)3812 J.Chem. Soc., Dalton Trans., 1998, 3805–3813 evaluated are · gÒ = 1.97 and TIP = 5.04 × 1024 cm3 mol21.This is in good agreement with the g values reported for VO21 ions by EPR measurements.46 A plot of cM vs. 1/T gives a straight line with slope 0.381 cm3 mol21 and intercept 238 × 1026 cm3 mol21 K21, which is in good agreement with the theoretical equation for VO21 complexes. Thus magnetic data for 2 indicate an orbitally non-degenerate ground state, and hence the E orbitals are less stable than B2 in 2, as is assumed for the band assignment in electronic spectroscopy. The X-band EPR spectrum of complex 2 in MeCN at 25 K, shown in Fig. 7, demonstrates the low-symmetry structure in the molecule and is very similar to those of other vanadyl complexes reported.46 The Hamiltonian parameters obtained from the simulation of the spectrum are gx = 1.997, gy = 1.995, Ax = 65 × 1024 Ay = 58 × 1024, and Az = 165 × 1024 cm21. The magnetic moments for complexes 3 (CrIII), 5 (CrIII), 8 (CuII), 9 (NiII), 10 (CoII), 11 (FeII) and 12 (MnII) have been measured (Experimental section) and are in accord with their high-spin nature.The measured magnetic moments were fitted by the Hamiltonian described above and the best fit parameters are as follows: · gÒ = 1.89, |D| = 1.89 cm21 for 3; ·gÒ = 1.97, |D| = 0.28 cm21 for 5; · gÒ = 2.10 for 8; · gÒ = 2.145, TIP = 269 × 1026 cm3 mol21 for 9; · gÒ = 1.95, |D| = 5.8 cm21, TIP = 596 × 1026 cm3 mol21 for 11; · gÒ = 1.98 for 12. The zero-field Mössbauer spectrum of solid complex 11 at 80 K is a symmetrical doublet and yields isomer shift d = 1.09 mm s21 and quadrupole splitting DEQ = 3.98 mm s21 with linewidths G = 0.25 mm s21.The Mössbauer isomer shift together with the quadrupole splitting support the six-co-ordinated iron(II) highspin state for the iron site in 11. Conclusion The hexadentate ligand 1,4,7-tris(carbamoylmethyl)-1,4,7-triazacyclononane and its N-methyl derivative form stable complexes with many first-row transition metals in the oxidation states II and III.The co-ordination number is dependent on the metal center, varying between 6 and 7. Octahedral geometry is favored over the prismatic geometry for the ligand of chromium(III) and nickel(II) complexes with twist angles q of 49 and 448, consistent with the simple ligand field energy arguments. It is interesting that chromium(III) forms complexes not only with carboxamide oxygens but also with the deprotonated amide nitrogen. Another important point that we note is that vanadium in its oxidation states III and IV forms very stable complexes through the oxygen atoms of the ligand.The d–d transitions of the chromium complex with this ligand are found to be at the same energies as those of the corresponding TCTA complex. The ligand provides a weak ligand field and stabilizes, in general, the 12 oxidation state. Fig. 7 The X-band EPR spectrum of complex 2 in MeCN at 25 K (microwave frequency 9.450 GHz; power 2.0 mW, modulation amplitude 9.9 G, modulation frequency 100 kHz).Acknowledgements We acknowledge the support of this work by the Fonds der Chemischen Industrie. Our thanks are also due to the skillful technical assistance of Frau D. Kreft, Frau H. Schucht and Herr A. Göbels. We also thank Dr. E. Bill for measuring the EPR spectrum. References 1 P. Chaudhuri and K. Wieghardt, Prog. Inorg.Chem., 1987, 35, 329. 2 See, for example, D. Burdinski, F. Birkelbach, T. Weyhermüller, U. Flörke, H.-J.Haupt, M. Lengen, A. X. Trautwein, E. Bill, K. Wieghardt and P. Chaudhuri, Inorg. Chem., 1998, 37, 1009; F. Birkelbach, U. Flörke, H.-J. Haupt, C. ButzlaV, A. X. Trautwein, K. Wieghardt and P. Chaudhuri, Inorg. Chem., 1998, 37, 2000. 3 K. P. Wainwright, Coord. Chem. Rev., 1997, 166, 35. 4 T. A. Kaden, Top. Curr. Chem., 1984, 121, 154. 5 P. V. Bernhardt and G. A. Lawrence, Coord. Chem. Rev., 1990, 104, 297. 6 M. Takahasi and S. Takamoto, Bull. Chem. Soc. Jpn., 1977, 50 3413. 7 K. Wieghardt, U.Bossek, P. Chaudhuri, W. Herrmann, B. C. Menke and J. Weiss, Inorg. Chem., 1982, 21, 4308. 8 B. A. Sayer, J. P. Michael and R. D. Hancock, Inorg. Chim. Acta, 1983, 77, L63. 9 D. A. Moore, P. E. Fanwick and M. J. Welch, Inorg. Chem., 1989, 28, 1504. 10 T. Beissel, K. S. Bürger, G. Voigt, K. Wieghardt, C. ButzlaV and A. X. Trautwein, Inorg. Chem., 1993, 32, 124. 11 O. Schlager, K. Wieghardt, H. Grondey, A. Rufinska and B. Nuber, Inorg. Chem., 1995, 34, 6440. 12 L.R. Gahan, G. A. Lawrence and A. M. Sargeson, Aust. J. Chem., 1982, 35, 1119. 13 L. Christiansen, D. N. Hendrickson, H. Toflund, S. R. Wilson and C. L. Xie, Inorg. Chem., 1986, 25, 2813. 14 K. Wieghardt, E. SchöVmann, B. Nuber and J. Weiss, Inorg. Chem., 1986, 25, 4877. 15 R. Ziessel and J.-M. Lehn, Helv. Chim. Acta, 1990, 73, 1149. 16 M. Di Vaira, F. Manni and P. Stoppioni, J. Chem. Soc., Chem. Commun., 1989, 126. 17 M. Di Vaira, B. Cosimelli, F. Manni and P. Stoppioni, J.Chem. Soc., Dalton Trans., 1991, 331. 18 A. A. Watson, A. C. Willis and D. P. Fairlie, Inorg. Chem., 1997, 36, 752. 19 P. Chaudhuri and M. Winter, unpublished work; F. Birkelbach, Dissertation, Bochum, 1995. 20 H. Sigel and R. B. Martin, Chem. Rev., 1982, 82, 385. 21 J. K. Moran and C. F. Meares, in Encyclopedia of Inorganic Chemistry, ed. R. B. King, Wiley, New York, 1994, vol. 6, p. 3067. 22 (a) S. Amin, C. Marks, L. M. Toomey, M. R. Churchill and J. R. Morrow, Inorg. Chim.Acta, 1996, 246, 99; (b) L. M. Berreau, J. A. Halfen, V. G. Young and W. B. Tolman, Inorg. Chem., 1998, 37, 1091. 23 K. Wieghardt, W. Schmidt, B. Nuber and J. Weiss, Chem. Ber., 1979, 112, 2220. 24 T. J. Atkins, J. E. Richman and W. F. Oettle, Org. Synth., 1978, 58, 86. 25 W. E. Weaver and W. M. Whaley, J. Am. Chem. Soc., 1947, 69, 515. 26 R. Kataky, K. E. Matthes, P. E. Nicholson, D. Parker and H.-J. Buschmann, J. Chem. Soc., Perkin Trans. 2, 1990, 1425. 27 G. M. Sheldrick, SADBAS, University of Göttingen, 1994. 28 G. M. Sheldrick, SHELXTL PLUS, Siemens Analytical Instruments, Madison, WI, 1990. 29 (a) International Tables for X-Ray Crystallography, Kynoch Press, Birmingham, 1974, vol. 4; (b) H. D. Flack, Acta Crystallogr., Sect. A, 1983, 39, 876. 30 C. J. Ballhausen and H. B. Gray, Inorg. Chem., 1962, 1, 111. 31 A. B. P. Lever, Inorganic Electronic Spectroscopy, Elsevier, Amsterdam, 1984. 32 C. K. Johnson, ORTEP, Report ORNL-5138, Oak Ridge National Laboratory, Oak Ridge, TN, 1976. 33 M. G. B. Drew, Prog. Inorg. Chem., 1977, 23, 67 and refs. therein. 34 R. A. Levenson and R. L. R. Townsend, Inorg. Chem., 1974, 13, 105. 35 M. Shimoi, Y. Saito and H. Ogino, Bull. Chem. Soc. Jpn., 1991, 64, 2629. 36 A. Neves, W. Walz, K. Wieghardt, B. Nuber and J. Weiss, Inorg. Chem., 1988, 27, 2484 and refs. therein; J. C. Dutton, G. D. Fallon and K. S. Murray, Inorg. Chem., 1988, 27, 34; A. D. Keramidas, A. B., Papaiaanou, A. Vlahos, T. A. Kabanos, G. Bonas,J. Chem. Soc., Dalton Trans., 1998, 3805–3813 3813 A. Makriyannis, C. P. Rapropoulou and A. Terzis, Inorg. Chem., 1996, 35, 357 and refs. therein. 37 T. J. Collins, B. D. Santarsiero and G. H. Spies, J. Chem. Soc., Chem. Commun., 1983, 681. 38 C. M. Murdoch, M. K. Cooper, T. W. Hambley, W. N. Hunter and H. C. Freeman, J. Chem. Soc., Chem. Commun., 1986, 1329. 39 H. C. Freeman, Adv. Protein Chem., 1967, 22, 257. 40 O. Schlager, K. Wieghardt and B. Nuber, Inorg. Chem., 1995, 34, 6449. 41 M. J. van der Merwe, J. C. A. Boeyens and R. D. Hancock, Inorg. Chem., 1985, 24, 1208. 42 J. N. McElearney, R. W. Schwartz, A. E. Sieigel and R. L. Carlin, J. Am. Chem. Soc., 1971, 93, 4337; R. L. Carlin, C. J. O’Connor and S. N. Bhatia, Inorg. Chem., 1976, 15, 985; A. K. Gregson, D. M. Doddrell and P. C. Healy, Inorg. Chem., 1978, 17, 1216. 43 J. N. McElearney, R. W. Schwartz, S. Merchant and R. L. Carlin, J. Chem. Phys., 1971, 55, 466. 44 H. U. Rahman, J. Phys. C., 1971, 4, 3301. 45 Theory and Applications of Molecular Paramagnetism, eds. E. A. Boudreaux and L. N. Mulay, Wiley, New York, 1976. 46 Electron Paramagnetic Resonance of d Transition Metal Compounds, eds. F. E. Mabbs and D. Collison, Elsevier, Amsterdam, 1992. Paper 8/06466K

ISSN:1477-9226    

DOI:10.1039/a806466k

出版商:RSC    

年代:1998

数据来源: RSC

摘要:

DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, 3815–3818 3815 The structure of R3PBr2 compounds in the solid state and in solution; geometrical dependence on R, the crystal structures of tetrahedral ionic Et3PBr2 and molecular trigonal bipyramidal (C6F5)3PBr2 Stephen M. Godfrey, Charles A. McAuliVe, Imran Mushtaq, Robin G. Pritchard and Joanne M. SheYeld Department of Chemistry, University of Manchester Institute of Science & Technology, Manchester, UK M60 1QD Received 16th September 1998, Accepted 28th September 1998 Twenty-one triorganophosphorus dibromide compounds, R3PBr2 (R3 = substituted aryl, mixed aryl–alkyl, triaryl or trialkyl) have been synthesized from diethyl ether solution and characterised by analytical and 31P-{H} NMR data in CDCl3 solution, the vast majority being reported for the first time.All but one of the compounds are ionic, [R3PBr]Br in CDCl3 solution, in keeping with analogous species containing iodine or chlorine, [R3PX]X (X = I or Cl) according to 31P-{H} NMR studies.In contrast, (C6F5)3PBr2 has a molecular five-co-ordinate trigonal bipyramidal structure both in CDCl3 solution and in the solid state. The single crystal structure of this compound has been determined and it represents the only reported R3PBr2 species which contains a five-co-ordinate phosphorus atom. It has D3 symmetry and perfect trigonal bipyramidal geometry. Why (C6F5)3PBr2 is the only R3PBr2 compound which adopts trigonal bipyramidal geometry is reasoned to be due to the very low basicity of the parent tertiary phosphine, which favours this geometry for the dihalogen adducts, a phenomenon previously observed for dichlorine adducts of tertiary phosphines and dihalogen adducts of tertiary arsines.The crystal structure of the first non-solvated trialkylphosphine dibromide compound, Et3PBr2, has also been determined and contains a tertrahedral phosphorus atom, exhibiting a long Br ? ? ? Br contact, 3.303(2) Å, and is probably better described as ionic, [Et3PBr]Br, with significant cation–anion interaction.Introduction Although compounds of formula R3PX2 (X = Cl, Br or I) have been recognised for over a century,1 it is only recently that their precise structural nature has been elucidated. The interesting variety, and in some cases unexpected structures of such compounds in the solid state, has sparked considerable renewed interest in this area.2–7 Perhaps surprisingly, no comprehensive study of compounds of formula R3PBr2 has ever been reported despite the fact that Ph3PBr2 is available commercially and has found significant use in synthetic organic chemistry, being widely used in many bromination reactions.8 Recent studies by us 2 and other workers 9 have established the diiodide compounds, R3PI2, as charge-transfer species, (R3 = Ph3, 2PhMe2 2 or But 3 9).However, a sample of Ph3PI2 prepared from nitrobenzene, a more polar solvent, was claimed 10 to be ionic, [Ph3PI]I, from solid state 31P-{H} NMR spectroscopy, thus illustrating the structural dependence of R3PI2 compounds on the nature of the solvent in which they are prepared.The analogous dibromine compounds, R3PBr2, have received very little attention, especially in the solid state. Solution 31P-{H} NMR studies regarding such compounds are limited to Ph3PBr2,10 Bun 3PBr2, Pri 3PBr2,11 as well as the nitrogen containing compounds (Et2N)3PBr2, (Me2N)3PBr2 and Me(Me2N)2PBr2.12 All the compounds were assigned an ionic structure [R3PBr]Br, in acetonitrile solution.Such reports are in agreement with conductimetric studies by Harris and coworkers 13 who also assigned R3PBr2 compounds an ionic structure, [ R3PBr]Br, again in acetonitrile solution. Solid state investigations of R3PBr2 compounds are limited to an infrared study of Me3PBr2 14 and a solid state 31P-{H} NMR study of Ph3PBr2 prepared from nitrobenzene.10 Both compounds were assigned an ionic structure, [R3PBr]Br (R = Me or Ph) in agreement with solution studies.More recently, a sample of Ph3PBr2 prepared from diethyl ether solution was examined crystallographically 3 and shown to adopt the four-co-ordinate charge-transfer structure Ph3PBr–Br, previously established for the diiodo compound, Ph3PI2. However, this has been challenged by Gates and co-workers 15 who spectroscopically investigated Ph3PBr2, prepared from both dichloromethane and toluene solution; these workers compared the Raman spectrum of this material with those of the salts [Ph3PBr][BBr4] and [Ph3PBr][AlBr4].The similarity between the Raman spectra led the authors to conclude that Ph3PBr2 is better described as ionic, [Ph3PBr]Br, in the solid state despite the fact that the Br–Br distance, 3.12(1) Å, lies well within bonding distance if the van der Waals radius for two bromine atoms, 3.9 Å, is considered. Clearly triorgano-phosphorus and -arsenic dihalide compounds could conceivably be regarded as either molecular charge-transfer ‘spoke’ structures R3EX–X (E = P or As; X = Br or I) which exhibit a weak donor–acceptor X–X bond or as ionic species, [R3EX]X (E = P or As; X = Br or I), which exhibit significant cation–anion interactions; which of these structures is correct is probably dependent on the R, E and X variables for any given R3EX2 compound. It has been established by Raman spectroscopy that Ph3AsI2 does exist as a donor–acceptor charge-transfer compound, Ph3AsI–I, similarly Ph3PI2 has been shown to be a donor-acceptor compound of [Ph3PI]1 and I2.The perceived ambiguity regarding the solidstate structure of R3EX2 compounds indicates that a theoretical study of these compounds would certainly be worthwhile. No trigonal bipyramidal compound of formula R3PBr2 has been crystallographically characterised and the ionic structure of R3PBr2 compounds, [R3PBr]Br, which has been suggested from several NMR, infrared and conductimetric studies, had never been confirmed crystallographically.However, very recently3816 J. Chem. Soc., Dalton Trans., 1998, 3815–3818 du Mont and co-workers 7 reported the crystal structure of the solvated compound Pri 3PBr2?0.5CH2Cl2. This molecule exhibits a long Br ? ? ? Br contact [3.369(2) Å] and is essentially ionic; the bromide ions of two such ion pairs are ‘bridged’ by a CH2Cl2 molecule. We are currently engaged on a comprehensive study of R3EX2 compounds (E = P or As;16 X2 = Cl2,5 Br2,3 I2 2 or IBr 4).Our studies regarding the related chloro-compounds, R3PCl2,5 have shown that the structure of such compounds is dependent on the nature of R and, in some cases, the nature of the solvent used for their preparation. In most cases the compounds are ionic both in the solid state and in solution, this being illustrated crystallographically for Prn 3PCl2. In contrast, the compounds R3PCl2 [R3 = (C6F5)3 or Ph2(C6F5)] are fiveco- ordinate molecular trigonal bipyramidal species.Interestingly, Ph3PCl2 appears to represent a borderline case and is solvent dependent, Ph3PCl2 prepared from diethyl ether again being molecular trigonal bipyramidal whereas the same material prepared in the same way but using CH2Cl2 as a solvent produces the ionic solvated dinuclear species [Ph3PCl ? ? ? Cl ? ? ? ClPPh3]Cl?2CH2Cl2. The aims of the present study are therefore as follows: first to report a comprehensive study of R3PBr2 compounds which contain a wide variety of diVerent R groups on the phosphorus atoms, secondly to characterise crystallographically the first five-co-ordinate trigonal bipyramidal R3PBr2 compound and thirdly to characterise crystallographically the first ionic solvent-free [R3PBr]Br compound to compare with du Mont’s Pri 3PBr2?0.5CH2Cl2.7 Considering the familiarity of R3PBr2 compounds to both inorganic and organic chemists alike, establishing their structures is of fundamental importance and is of considerable current interest.Results and discussion All of the triorganophosphorus dibromides synthesized for this study were prepared by the reaction of equimolar quantities of tertiary phosphine and dibromine in diethyl ether, eqn. (1) R3P 1 Br2 Et2O, N2 r.t. R3PBr2 (1) (r.t. = room temperature). Reaction times were dependent on the tertiary phosphine; triaryl tertiary phosphines took ca. 3 d to reach complete reaction with dibromine whereas the corresponding trialkyl tertiary phosphines reacted with dibromine in a matter of hours.In all cases a white flocculent solid was produced which was isolated by standard Schlenk techniques. All the products described are very moisture sensitive, especially those containing parent trialkyl tertiary phosphines, which smoke profusely when exposed to the atmosphere, therefore strictly anhydrous conditions must be adhered to during their synthesis and subsequent manipulation.Elemental analyses of the compounds, together with their 31P-{H} NMR chemical shifts recorded in CDCl3 solution, are presented in Table 1. With one notable exception, (C6F5)3PBr2, all of the R3PBr2 compounds exhibit high positive 31P-{H} NMR resonances. Such resonances are indicative of an ionic structure in CDCl3 solution, irrespective of their solid state structure. Furthermore, the values recorded herein are in good agreement with the limited studies regarding such compounds performed by previous workers.10–12 Of the compounds described here only Ph3PBr2,10 Bun 3PBr2,11 and (Me2N)3PBr2 12 have previously been the subject of a solution 31P-{H} NMR spectroscopic study; the present values for these compounds d 49.2, 102.5 and 48.2, are in good agreement with the previously reported values of d 48.3, 105.0 and 47.0 respectively.Previous workers 10–12 also assigned an ionic structure, [R3PBr]Br, in solution to these compounds.The chemical shifts recorded in Table 1 are also similar to those previously observed for analogous R3PI2 1 and R3PCl2 5 compounds which have also been shown to adopt the ionic [R3PX]X (X = Cl or I) structure in CDCl3 solution. We have previously observed 2,5 that the CDCl3 solution 31P-{H} NMR shifts for [R3PCl]Cl are more positive than those of [R3PI]I. In keeping with this phenomenon, the values recorded here for [R3PBr]Br are intermediate between those recorded for [R3PCl]Cl and [R3PI]I, for a given parent tertiary phosphine.The 31P-{H} NMR resonance for (C6F5)3PBr2, d 259.1, is clearly anomalous and is particularly interesting since it arises from the R3PBr2 compound which contains the more acidic (or least basic) parent tertiary phosphine. In addition, this value is similar to the chemical shift in the 31P-{H} NMR of (CF3)3PBr2, d 264.5, recorded by Cavell et al.,17 who assigned a trigonal bipyramidal structure to this R3PBr2 compound.However, no five-co-ordinate R3PBr2 species has been crystallographically characterised. Consequently we decided to investigate the structure of (C6F5)3PBr2 by single crystal X-ray diVraction. Recrystallisation of (C6F5)3PBr2 from dichloromethane solution at room temperature produced a large quantity of huge colourless crystals on standing for ca. 7 d. Of these, one was selected for analysis by X-ray diVraction. In contrast to all previous reports regarding compounds of formula R3PBr2, which relate to phosphorus in tetrahedral geometry, (C6F5)3- PBr2 is trigonal bipyramidal, with a five-co-ordinate phosphorus atom, Fig. 1. In addition to being the first reported trigonal bipyramidal R3PBr2 compound, (C6F5)3PBr2 is also only the second non-solvated R3PBr2 compound to be studied crystallographically. The reason why (C6F5)3PBr2 adopts this geometry in contrast to all the other reported R3PBr2 compounds reported herein must be due to the very low basicity of the parent tertiary phosphine.It exhibits crystallographically imposed (space group R3� c) trigonal bipyramidal geometry (D3 symmetry) with d (P–Br) of 2.4105(9) Å, significantly longer than that exhibited by Ph3PBr2, 2.181(2) Å3, as expected with the higher co-ordination number at the phosphorus atom. In addition to crystallographically characterising the first trigonal bipyramidal R3PBr2 compound, we were also interested in crystallographically characterising an R3PBr2 compound which adopts an ionic structure but doesn’t contain a dichloro- Table 1 Analytical and spectroscopic data for the compounds R3PBr2 Analysis (%) a 31P-{H}, Compound (C6F5)3PBr2 (p-FC6H4)3PBr2 (p-FC6H4)Ph2PBr2 (p-ClC6H4)3PBr2 (m-MeC6H4)3PBr2 (o-MeC6H4)Ph2PBr2 (p-MeC6H4)Ph2PBr2 Ph3PBr2 Ph2(C5H4N)PBr2 Ph2(C6H11)PBr2 Ph2PrnPBr2 Ph2MePBr2 PhMe2PBr2 PhBun 2PBr2 PhPrn 2PBr2 Bun 3PBr2 Prn 3PBr2 Et3PBr2 (PhCH2)3PBr2 (C6H11)3PBr2 (Me2N)3PBr2 C 31.0 (31.2) 45.4 (45.4) 47.3 (47.2) 40.8 (41.1) 55.4 (54.3) 51.6 (52.3) 48.7 (48.4) 51.5 (51.2) 49.4 (49.1) 51.5 (50.5) 46.1 (46.4) 43.3 (43.3) 31.9 (32.2) 43.8 (43.8) 40.3 (40.7) 39.5 (39.8) 33.6 (33.7) 26.1 (25.9) 54.3 (54.3) 48.5 (49.1) 22.0 (22.3) H 0.0 (0.0) 2.8 (2.5) 3.1 (2.8) 2.4 (2.3) 4.9 (4.5) 3.9 (4.2) 3.9 (4.2) 3.8 (3.6) 7.7 (7.5) 5.1 (4.9) 4.2 (4.4) 3.5 (3.6) 3.8 (3.7) 6.3 (6.5) 5.4 (5.4) 7.2 (7.5) 6.9 (6.6) 5.6 (5.4) 4.6 (4.5) 8.3 (7.5) 5.9 (5.6) Br 23.2 (23.1) 33.2 (33.6) 34.9 (34.9) 30.3 (30.4) 33.3 (34.5) 33.4 (33.9) 33.4 (33.9) 37.7 (37.9) 36.1 (36.4) 36.0 (37.4) 39.9 (41.2) 44.5 (44.4) 53.7 (53.7) 41.8 (41.7) 45.2 (45.1) 43.8 (44.2) 49.6 (50.0) 57.1 (57.5) 34.1 (34.5) 35.8 (36.4) 49.0 (49.5) d b 259.1 47.0 44.3 48.1 45.9 54.4 52.3 49.2 64.4 72.2 55.0 63.0 67.7 83.7 81.2 102.5 101.6 100.0 89.8 105.5 48.2 a Calculated values in parentheses.b Shifts recorded in CDCl3 relative to concentrated phosphoric acid standard.J. Chem. Soc., Dalton Trans., 1998, 3815–3818 3817 methane solvent molecule since, as observed in [Ph3PCl ? ? ? Cl ? ? ? ClPPh3]Cl?2CH2Cl2 and Pri 3PBr2?0.5CH2Cl2, the longrange electrostatic interactions between the solvent and the ionic molecule influence the product formed.Consequently, we recrystallised a sample of Et3PBr2 previously prepared in Et2O from Et2O (and not CH2Cl2 to avoid its possible inclusion in the structure) at room temperature. On standing for ca. 6 d a number of colourless crystals formed which were removed from the reaction vessel in an inert atmosphere and plunged into an inert oil.From these, a suitable crystal was chosen for examination by single crystal X-ray diVraction. The crystal structure of Et3PBr2 is illustrated in Fig. 2. The structure represents the first crystallographically characterised non-solvated trialkylphosphine dibromide compound and contains the phosphorus atom in tetrahedral geometry, as expected. A long contact [3.303(2) Å] exists between the two bromine atoms and Fig. 1 Molecular structure of (C6F5)3PBr2 (the molecule has crystallographically imposed D3 symmetry). Selected bond lengths (Å) and angles (8): Br(1)–P(1) 2.4105(9), P(1)–C(1) 1.819(7); C(1)–P(1)–C(1) 120, C(1)–P(1)–Br(1) 90, Br(1)–P(1)–Br(1) 180, C(2)–C(1)–P(1) 121.6(3). Fig. 2 Molecular structure of [Et3PBr]Br (only one of each of the disordered methylene groups is illustrated for clarity). Selected bond lengths (Å) and angles (8): Br(2)–Br(1) 3.303(2), Br(2)–P(1) 2.173(3); Br(1)–Br(2)–P(1) 177.5(1).the structure is probably best described as ionic, [Et3PBr]Br with cation–anion interaction. This interaction is similar to that exhibited by Pri 3PBr2?0.5CH2Cl2, 3.369(2) Å, reported by du Mont and co-workers.7 Conclusion The results reported here clearly show that all of the R3PBr2 compounds except one ionise in CDCl3 solution to produce [R3PBr]Br, from 31P-{H} NMR studies. However when a very weakly basic parent tertiary phosphine is employed, (C6F5)3P, a trigonal bipyramidal R3PBr2 compound is revealed in the solid state which also persists in CDCl3 solution.The geometrical dependence of R3PBr2 compounds on R is therefore clearly illustrated. This phenomenon has previously been observed for R3AsBr2 15 and R3PCl2 5 compounds. The solid state structure of Et3PBr2, prepared and recrystallised from Et2O, is interesting to compare to the solvated Pri 3PBr2?0.5CH2Cl2.7 Both structures exhibit long Br–Br contacts [3.303(2) and 3.369(2) Å, respectively] and are essentially ionic.Experimental All of the compounds reported here are moisture sensitive, some intensely so, decomposing in a few seconds if exposed to the atmosphere. Consequtly, strictly anaerobic and anhydrous conditions were employed for their synthesis. Any subsequent manipulations were carried out inside a Vacuum Atmospheres HE-493 glove-box. Diethyl ether (BDH) was dried by standing over sodium wire for ca. 1 d and subsequently refluxed over CaH2 in an inert atmosphere and distilled directly into the reaction vessel. Anhydrous CH2Cl2 was obtained commercially (Aldrich) and used as received. Tertiary phosphines were either synthesized by standard Grignard techniques, R3P [R3 = (p-FC6H4)3, (p-FC6H4)Ph2, (p-ClC6H4)3, (m-MeC6H4)3, (o-MeC6H4)Ph2, (p-MeC6H4)Ph2, PhBun 2, PhPrn 2 or Ph2- (C6H11)] or obtained commercially, R3P [R3 = (C6F5)3, Ph3, Ph2(C5H4N), Ph2Prn, Ph2Me, PhMe2, Bun 3 or Prn 3 (Aldrich); R = Et3, (CNCH2CH2)3, (PhCH2)3 or (C6H11)3, (Strem); R = Me2N (Lancaster)].The purity of all the tertiary phosphines used was confirmed by elemental analysis and 31P-{H} NMR spectroscopy prior to use. Dibromine was obtained commercially (Aldrich) and used as received. All the R3PBr2 compounds were synthesized in a similar way, that of Ph3PBr2 being typical. Triphenylphosphine (2.00 g, 7.63 mmol) was suspended in Et2O (ca. 75 cm3) and subsequently dibromine (1.22 g, 0.39 cm3, 7.63 mmol) was added.After reaction completion, the resultant white solid was isolated using standard Schlenk techniques. The solids were then transferred to pre-dried argon-filled ampoules which were flame sealed. Elemental analyses were performed by the analytical laboratory of this department. The 31P-{H} NMR spectra were recorded as CDCl3 solutions on a Bruker AC200 highresolution multiprobe spectrometer relative to concentrated phosphoric acid as standard.Crystallography Crystals of (C6F5)3PBr2 were mounted in Lindemann tubes under an atmosphere of dry argon. Crystals of Et3PBr2 were submerged in an inert oil under anaerobic conditions and a suitable crystal was chosen by examination under the microscope. The crystal, with its protective coating of oil, was then mounted on a glass fibre and transferred to the diVractometer and cooled to ca. 183(2) K in the cold gas stream derived from liquid nitrogen. Measurements were performed on a MAC 3 CAD 4 diVractometer employing graphite-monochromated Mo-Ka radiation (l = 0.71069 Å) and w–2q scans. Both3818 J.Chem. Soc., Dalton Trans., 1998, 3815–3818 structures were solved by direct methods. Unit-cell dimensions were derived from the setting angles of 25 accurately centred reflections. Lorentz-polarisation corrections were applied. Details of the X-ray measurements and subsequent structure determinations are presented in Table 2. Hydrogen atoms were confined to chemically reasonable positions.In the structure of [Et3PBr]Br each of the methylene groups is disordered over two semi-populated sites. Neutral atom scattering factors were taken from ref. 18. Anomalous dispersion eVects were taken from ref. 19. All calculations were performed using SHELXS 86 and SHELXL 93 crystallographic software packages.20,21 CCDC reference number 186/1174. Acknowledgements We are grateful to the EPSRC for a research studentship (to J. M. S.).References 1 A. M. Liebigs, Ann. Chem., 1876, 181, 256. 2 S. M. Godfrey, D. G. Kelly, A. G. Mackie, C. A. McAuliVe, R. G. Pritchard and S. M. Watson, J. Chem. Soc., Chem. Commun., 1991, 1163; N. Bricklebank, S. M. Godfrey, A. G. Mackie, C. A. Table 2 Crystal data and details of refinement for R3PBr2 MT /K Crystal system Space group a/Å b/Å c/Å U/Å3 Z Dc/g cm23 F(000) m/cm21 Total data measured Maximum 2q/8 No. unique reflections No. observed reflections [I > 2.00s(I)] No.parameters Final R Final R9 (C6F5)3PBr2 691.97 293(2) Rhombohedral R3� c (no. 167) 11.539(3) — 26.389(5) 3043(1) 6 2.266 1968 42.17 596 49.8 596 596 58 0.0317 0.0529 [Et3PBr]Br 277.97 183(2) Orthorhombic Pbca (no. 61) 16.387(6) 14.585(2) 9.369(2) 2239(1) 8 1.649 1088 73.20 1746 50.0 1746 1746 95 0.063 0.012 McAuliVe, R. G. Pritchard and P. J. Kobryn, J. Chem. Soc., Dalton Trans., 1993, 101; N. Bricklebank, S. M. Godfrey, H. P. Lane, C. A. McAuliVe, R. G. Pritchard and J.M. Moreno, J. Chem. Soc., Dalton Trans., 1995, 2421. 3 N. Bricklebank, S. M. Godfrey, A. G. Mackie, C. A. McAuliVe and R. G. Pritchard, J. Chem. Soc., Chem. Commun., 1992, 355. 4 N. Bricklebank, S. M. Godfrey, C. A. McAuliVe and R. G. Pritchard, J. Chem. Soc., Dalton Trans., 1993, 2261. 5 S. M. Godfrey, C. A. McAuliVe, R. G. Pritchard and J. M. SheYeld, Chem. Commun., 1996, 2521; S. M. Godfrey, C. A. McAuliVe, R. G. Pritchard, J. M. SheYeld and G. M. Thompson, J. Chem. Soc., Dalton Trans., 1997, 4823. 6 M. A. H. A. Al-Juboori, P. N. Gates and A. S. Muir, J. Chem. Soc., Chem. Commun., 1991, 1270. 7 F. Ruthe, W. W. du Mont and P. G. Jones, Chem. Commun., 1997, 1947. 8 G. A. Wiley, B. M. Rein and R. L. Hershkowitz, Tetrahedron Lett., 1964, 2509; A. G. Anderson and F. J. Freenor, J. Am. Chem. Soc., 1964, 86, 5037; J. Org. Chem., 1972, 27, 626. 9 W. W. du Mont, M. Bätcher, S. Pohl and W. Saak, Angew. Chem., Int. Ed. Engl., 1987, 26, 912. 10 K. B. Dillon and T. C. Waddington, Nature (London) Phys. Sci., 1971, 230, 158. 11 G. A. Wiley and W. R. Stine, Tetrahedron Lett., 1967, 24, 2321. 12 R. Bartsch, O. Stelzer and R. Schmutzler, Z. Naturforsch, Teil B, 1981, 36, 1349; J. Fluorine Chem., 1982, 20, 85. 13 A. D. Beveridge and G. S. Harris., J. Chem. Soc., 1964, 6077; A. D. Beveridge, G. S. Harris and F. Inglis, J. Chem. Soc. A, 1966, 520; A. D. Beveridge, G. S. Harris and D. S. Payne, J. Chem. Soc. A, 1966, 726; G. S. Harris and M. F. Ali, Tetrahedron Lett., 1968, 37; Inorg. Nucl. Chem. Lett., 1968, 4, 5; M. F. Ali and G. S. Harris, J. Chem. Soc., Dalton Trans., 1980, 1545; G. S. Harris and J. S. McKechnie, Polyhedron, 1985, 4, 115. 14 J. Goubeau and R. Baumgartner, Z. Electrochem., 1960, 64, 598. 15 M. A. H. A. Al-Juboori, P. N. Gates and A. S. Muir, J. Chem. Soc., Dalton Trans., 1994, 1441. 16 N. Bricklebank, S. M. Godfrey, H. P. Lane, C. A. McAuliVe, R. G. Pritchard and J. M. Moreno, J. Chem. Soc., Dalton Trans., 1995, 3873. 17 R. G. Cavell, J. A. Gibson and K. I. The, J. Am. Chem. Soc., 1977, 99, 784. 18 International Tables for X-Ray Crystallography, Kynoch Press, Birmingham, 1974, vol. 4, Table 2.2A. 19 International Tables for X-Ray Crystallography, Kynoch Press, Birmingham, 1974, vol. 4, Table 2.3.1. 20 G. M. Sheldrick, SHELXS 86, in Crystallographic Computing 3, ed. G. M. Sheldrick, Oxford University Press, 1985, p. 175. 21 G. M. Sheldrick, SHELXL 93, University of Göttingen, 1993. Paper 8/0724

ISSN:1477-9226    

DOI:10.1039/a807241h

出版商:RSC    

年代:1998

数据来源: RSC